Photoelectron generation apparatus

Inspection device

INSPECTION APPARATUS

The present invention relates to an inspection apparatus for inspecting a pattern defect or the like formed on a surface of an inspection object. In detail, it relates to secondary charged particles for capturing the properties of the surface of the object to be inspected to form image data, and according to the image data thereof, the surface formed on the inspection object is inspected by a high throughput (Through put). Inspection device such as pattern and inspection method.

Conventional semiconductor inspection devices are devices and techniques that correspond to 100 nm design rules. However, the sample to be inspected includes a wafer, an exposure mask, an EUV (very ultraviolet light) mask, a NIL (Nano-imprinted lithography) mask, and a substrate. It is now required to correspond to the design rules and equipment of the sample of 5 to 30 nm. That is, the L/S (line/space) or hp (half pitch) node corresponding to the pattern is required to be a generation of 5 to 30 nm. When such a sample is inspected by an inspection device, high resolution must be obtained.

Here, the sample refers to an exposure mask, an EUV mask, a mask for imprinting (and a template), a semiconductor wafer, a substrate for an optical element, and a substrate for an optical circuit. Some of these samples have patterns and some have no patterns. Among the patterns, there are bumps and bumps. Patterns without bumps are formed from different materials. Among the unpatterned ones, there are coated oxide films, and Oxide film coated.

The subject matter of the conventional inspection apparatus is summarized as follows.

First, there is a problem of insufficient resolution and throughput. In the conventional technique of the imaging optical system, the pixel size is 50 nm and the aberration is about 200 nm. In order to further increase the high resolution and throughput, it is necessary to reduce aberrations, reduce the energy width of the irradiation current, small pixel specifications, and increase the amount of current.

Second, in the case of the SEM (Scanning Electron Microscope) type inspection, the more the inspection of the fine structure, the greater the problem of the throughput. Because, if you do not need a smaller pixel size, the resolution of the image will be insufficient. These problems are caused by the fact that SEM mainly uses edge contrast to perform image formation and defect inspection. For example, if it is 5nmPx specification, 200MPPS, it is about 6hr/cm.². This time is 20 to 50 times that of the image projection, which is not practical in inspection.

[Patent Document]

(previous technical literature)

(Patent Document 1) International Publication WO2002/001596

(Patent Document 2) Japanese Patent Laid-Open Publication No. 2007-48686

(Patent Document 3) Japanese Patent Laid-Open No. Hei 11-132975

<p>Therefore, the present invention has been made in view of the above-described conventional inspection apparatus, and provides an inspection method and an inspection apparatus which can improve the inspection precision and can be applied to a design rule of 5 to 30 nm.</p>

<p>Furthermore, according to an embodiment of the present invention, an inspection device is provided The present invention is characterized in that: a beam generating means is provided to generate either charged particles or electromagnetic waves in the form of a beam; and the primary optical system guides and irradiates the beam to an inspection object held in the work chamber; The optical system includes: a movable numerical aperture; and a first detector for detecting secondary charged particles generated from the inspection target and passing through the numerical aperture; and an image processing system according to the first detector Detecting secondary charged particles to form an image; and a second detector disposed between the movable numerical aperture and the first detector to detect secondary charged particles generated from the inspection object at a polycondensation position Location and shape.</p><p>The first detector may detect the secondary charged particles in a state where the position of the movable numerical aperture has been adjusted based on the detection result of the second detector.</p><p>The beam may be a charged particle beam, and the beam generating means may include a photoelectric element formed by coating an optoelectronic material on the planar portion of a substrate made of a light-transmitting element having a flat portion to receive irradiation. The photoelectron material generates photoelectrons by light; one or more lenses are arranged at a predetermined interval in a subsequent stage of the photovoltaic element to accelerate photoelectrons generated from the photovoltaic element; and a numerical aperture is disposed in a rear portion of the lens; And a cathode lens disposed in a subsequent stage of the numerical aperture.</p><p>The beam is an electromagnetic wave beam, and the beam generating means may generate a plurality of the beams having different wavelengths.</p><p>The first detector may detect secondary charged particles generated from the beam irradiation surface of the inspection target.</p><p>The first detector can also detect the aforementioned beam from the inspection object Secondary charged particles produced by the illuminated surface.</p><p>The first detector may detect secondary charged particles generated from a surface of the inspection object opposite to the beam irradiation surface.</p><p>The first detector can also use TDI.</p><p>The second detector may also use an EB-CCD.</p><p>The movable numerical aperture may be formed by a cross hole or a slit in the opening.</p><p>The present invention includes an optical microscope and an SEM (scanning electron microscope) for observing the object to be inspected, and the beam generating means, the primary optical system, the secondary optical system, the image processing system, and the optical microscope The SEM may be disposed in the working chamber.</p><p>The secondary charged particles are a part or a mixture of electron emission, mirror reflection electrons, and photoelectrons. When an electromagnetic wave is irradiated, photoelectrons are generated from the surface of the sample. When charged particles such as electron beams are applied to the surface of the sample, electrons are emitted twice from the surface of the sample. In addition, mirror-reflected electrons are also formed. The electron beam collides with the surface of the sample to produce electrons twice. In other words, the secondary electron emission is a part or a mixture of secondary electrons, reflected electrons, and backscattered electrons. Furthermore, the irradiated electron beam does not collide with the surface of the sample, and the reflection near the surface is called mirror reflection electron.</p>

<p>According to the present invention, it is possible to provide an inspection method and an inspection apparatus which are improved in inspection precision.</p>

<p>1Semiconductor inspection device</p><p>10Carton holding box</p><p>11 Lifting platform</p><p>12 Lifting mechanism</p><p>20Small environmental installations</p><p>21Small environmental space</p><p>22Shell</p><p>23 gas circulation device</p><p>24Draining device</p><p>25 Pre-aligner</p><p>27Close device</p><p>30Host shell</p><p>31Workroom</p><p>32Chassis body</p><p>32aShell body</p><p>32bChassis body</p><p>33Chassis support device</p><p>33aShell support device</p><p>33bChassis support device</p><p>36Set up the gantry</p><p>36b gantry</p><p>37Anti-vibration device</p><p>37bAnti-vibration device</p><p>40Loader casing</p><p>40bLoader casing</p><p>411st load room</p><p>422nd loading room</p><p>43Chassis body</p><p>45Close device</p><p>46Close device</p><p>47 Wafer rack</p><p>49b hanging parts</p><p>50 platform device</p><p>51Fixed platform</p><p>52Y platform</p><p>53X platform</p><p>54Rotating platform</p><p>55Holding box</p><p>56 platform drive mechanism</p><p>60Loader</p><p>61Robot-type first handling unit</p><p>63Robot type second handling unit</p><p>70electron optical device</p><p>71Mirror tube</p><p>71aLight source control unit</p><p>71bray tube control unit</p><p>721 times optical system</p><p>76Detection system</p><p>76-22nd detector</p><p>76-33rd detector</p><p>81Precharge unit</p><p>83potentiometer application mechanism</p><p>85Electron beam calibration mechanism</p><p>87Alignment control device</p><p>91-1vacuum exhaust passage</p><p>91-2vacuum exhaust passage</p><p>93 platform</p><p>94Sample table</p><p>95Y direction activity department</p><p>95Y platform</p><p>96X direction activity department</p><p>96X platform</p><p>96a Guide surface</p><p>97 platform platform</p><p>97a guiding surface</p><p>98Chassis</p><p>221 top wall</p><p>222 bottom wall</p><p>223Wall</p><p>225 entrance</p><p>226 entrances and exits</p><p>231 gas supply unit</p><p>232Recycling catheter</p><p>233 catheter</p><p>241Inhalation catheter</p><p>242Air blower</p><p>243 catheter</p><p>244 catheter</p><p>271Closed parts</p><p>272 threshold</p><p>273 drive</p><p>321 bottom wall</p><p>321a bottom wall</p><p>321b bottom wall</p><p>322 top wall</p><p>323Wall</p><p>323 side wall</p><p>325 entrance</p><p>331Framework</p><p>331aRectangular steel plate</p><p>336Framework</p><p>336bFrame Structure</p><p>337b vertical frame</p><p>431 bottom wall</p><p>432 top wall</p><p>433Wall</p><p>433 side wall</p><p>434 partition wall</p><p>435 entrance</p><p>436 Entrance</p><p>437 entrance</p><p>451Closed parts</p><p>452 threshold</p><p>453 drive</p><p>461 threshold</p><p>471Rectangular substrate</p><p>472 pillar</p><p>473Support Department</p><p>474Support Department</p><p>501 box body</p><p>502Introduction to substrate loading and unloading</p><p>503 cover</p><p>505ULPA (Ultra Low Penetration Air) Filter</p><p>506Chemical filter</p><p>507Fan motor</p><p>521Servo motor</p><p>522Encoder</p><p>531Servo motor</p><p>532Encoder</p><p>551 wafer mounting surface</p><p>611 Drive Department</p><p>612Multi-section arm</p><p>6121st handling unit</p><p>615 Lifting mechanism</p><p>616 pick up device</p><p>632arms</p><p>724 Objective Optical System</p><p>724cathode lens</p><p>724Electrostatic lens</p><p>724-1 lens</p><p>724-2 lens</p><p>741 lens</p><p>761 detector</p><p>763Image Processing Department</p><p>780Control unit</p><p>781CPU (Central Processing Unit)</p><p>821hot silk</p><p>822hot wire power supply</p><p>823Electronic extraction power supply</p><p>824Extracted electrode</p><p>825Electroray</p><p>826Shielding box</p><p>827Discharge power supply</p><p>828Anode</p><p>830Ar gas</p><p>831Pipe</p><p>840Charge control</p><p>841electrode</p><p>842Toggle switch</p><p>843terminal</p><p>844Voltage generator</p><p>845 terminals</p><p>846Charge detector</p><p>849Timed Signal Oscillator</p><p>871Light microscope</p><p>873 monitor</p><p>906 pedestal</p><p>907aY direction guide</p><p>907bY direction guide</p><p>908aX direction guide</p><p>908bX direction guide</p><p>909aStatic bearing</p><p>909bStatic bearing</p><p>911aStatic bearing</p><p>911bStatic bearing</p><p>912 partition board</p><p>913 space</p><p>914 partition board</p><p>915 space</p><p>917Differential exhaust groove</p><p>918Differential exhaust groove</p><p>924 space</p><p>931Flexible piping</p><p>932Linear motor</p><p>933Linear motor</p><p>934Flexible piping</p><p>950narrowing</p><p>951narrowing</p><p>952 Turbo Molecular Pump</p><p>972Vacuum piping</p><p>973Vacuum piping</p><p>20001 times optical system</p><p>2020Photoelectric generating device</p><p>2021Photoelectric surface</p><p>2022 lens</p><p>2022electrode</p><p>2023 lens</p><p>2023electrode</p><p>2024 lens</p><p>2024electrode</p><p>2025 numerical light</p><p>2030 aligner</p><p>20311st aligner</p><p>20322nd aligner</p><p>2040Light</p><p>2060 aligner</p><p>20611st aligner</p><p>20622nd aligner</p><p>2091 lens</p><p>2091 lens group</p><p>2092 lens</p><p>2092 lens group</p><p>2093 lens</p><p>2093Lens Group</p><p>3000Light microscope</p><p>3002Scanning Electron Microscopy (SEM)</p><p>7000Light source tube</p><p>7241electrode</p><p>7242electrode</p><p>7245Parts</p><p>10000Light source</p><p>10000ALight</p><p>10001Mirror</p><p>10002Mirror</p><p>10004Mirror</p><p>10006 lens (TL1)</p><p>10008Numerical Optical Fiber (NA)</p><p>10009 lens (TL2)</p><p>12000Correct lens</p><p>12002 Field lens</p><p>12004Magnifying lens</p><p>12006Magnifying lens</p><p>20211Material</p><p>20212Photoelectric materials</p><p>20213Electrical materials</p><p>20214 Keeping parts</p><p>20215Lighting for laser irradiation</p><p>20216Inner diameter area</p><p>cCarmen</p><p>WRetaining the substrate</p><p>W wafer</p><p>W objects</p>

Fig. 1 is a view showing the main body of an inspection apparatus according to an embodiment of the present invention. An elevational view of the constituent elements, which is a view as viewed along the line AA of Fig. 2.

Fig. 2A is a plan view showing main components of the inspection apparatus shown in Fig. 1, which is a view taken along line BB of Fig. 1.

Fig. 2B is a schematic cross-sectional view showing another embodiment of the substrate carrying device of the inspection apparatus according to the embodiment of the present invention.

Fig. 3 is a cross-sectional view showing the small environment device of Fig. 1, which is a view taken along line CC.

Fig. 4 is a view showing the loader casing of Fig. 1, which is a view taken along line DD of Fig. 2.

Fig. 5 (A) and (B) are enlarged views of the wafer holder, in which (A) is a side view and (B) is a cross-sectional view taken along line EE of (A).

Fig. 6 is a view showing a variation of the support method of the main casing.

Fig. 7 is a view showing a variation of the support method of the main casing.

Fig. 8 is a schematic view showing the schematic configuration of an electro-optical device of the inspection apparatus shown in Fig. 1.

Figure 9 is a related diagram of an embodiment of the present invention.

Figure 10 is a related diagram of an embodiment of the present invention.

Figure 11 is a related diagram of an embodiment of the present invention.

Figure 12 is a related diagram of an embodiment of the present invention.

Figure 13 is a related diagram of an embodiment of the present invention.

Figure 14 is a related diagram of an embodiment of the present invention.

Fig. 15 (A) and (B) are diagrams related to an embodiment of the present invention.

Figure 16 is a related diagram of an embodiment of the present invention.

Fig. 17 (A) and (B) are diagrams related to an embodiment of the present invention.

Figure 18 is a related diagram of an embodiment of the present invention.

Figure 19 is a related diagram of an embodiment of the present invention.

Figure 20 is a related diagram of an embodiment of the present invention.

Figure 21 is a related diagram of an embodiment of the present invention.

Figure 22 is a related diagram of an embodiment of the present invention.

Figure 23 is a related diagram of an embodiment of the present invention.

Figure 24 is a related diagram of an embodiment of the present invention.

Figure 25 is a related diagram of an embodiment of the present invention.

Figure 26A is a related diagram of an embodiment of the present invention.

Figure 26B is a related diagram of an embodiment of the present invention.

Figure 26C is a related diagram of an embodiment of the present invention.

Figure 27A is a related diagram of an embodiment of the present invention.

Figure 27B is a related diagram of an embodiment of the present invention.

Figure 28 is a related diagram of an embodiment of the present invention.

Figure 29A is a related diagram of an embodiment of the present invention.

Figure 29B is a related diagram of an embodiment of the present invention.

Figure 30 is a related diagram of an embodiment of the present invention.

Figure 31 is a related diagram of an embodiment of the present invention.

Fig. 32A (a) and (b) are diagrams related to an embodiment of the present invention.

Figure 32B is a related diagram of an embodiment of the present invention.

Figure 32C is a related diagram of an embodiment of the present invention.

Figure 32D is a related diagram of an embodiment of the present invention.

Figure 32E is a related diagram of an embodiment of the present invention.

Fig. 33 (a) and (b) are diagrams related to an embodiment of the present invention.

Fig. 34 (a) and (b) are diagrams related to an embodiment of the present invention.

Figure 35 is a related diagram of an embodiment of the present invention.

Figure 36 is a related diagram of an embodiment of the present invention.

Figure 37 is a related diagram of an embodiment of the present invention.

Figure 38 is a related diagram of an embodiment of the present invention.

Figure 39 is a related diagram of an embodiment of the present invention.

Figure 40 is a related diagram of an embodiment of the present invention.

Figure 41 is a related diagram of an embodiment of the present invention.

Figure 42 is a related diagram of an embodiment of the present invention.

Figure 43 is a related diagram of an embodiment of the present invention.

Figure 44 is a related diagram of an embodiment of the present invention.

Figure 45 is a related diagram of an embodiment of the present invention.

Figure 46 is a related diagram of an embodiment of the present invention.

Fig. 47 (a) and (b) are diagrams related to an embodiment of the present invention.

Figure 48 is a related diagram of an embodiment of the present invention.

Figure 49 is a related diagram of an embodiment of the present invention.

Figure 50 is a related diagram of an embodiment of the present invention.

Figure 51 is a related diagram of an embodiment of the present invention.

Figure 52 is a related diagram of an embodiment of the present invention.

Fig. 53 (a) and (b) are diagrams related to an embodiment of the present invention.

Figure 54 is a related diagram of an embodiment of the present invention.

Figure 55 is a related diagram of an embodiment of the present invention.

Figure 56 is a related diagram of an embodiment of the present invention.

Figure 57 is a related diagram of an embodiment of the present invention.

Fig. 58 (a) to (c) are related diagrams of an embodiment of the present invention.

Figure 59 is a related diagram of an embodiment of the present invention.

Figure 60 is a related diagram of an embodiment of the present invention.

Figure 61 is a related diagram of an embodiment of the present invention.

Figure 62 is a related diagram of an embodiment of the present invention.

Figure 63 is a correlation diagram of an embodiment of the present invention.

Fig. 64 (a) and (b) are diagrams related to an embodiment of the present invention.

Fig. 65 (a) and (b) are diagrams related to an embodiment of the present invention.

Figure 66 is a related diagram of an embodiment of the present invention.

Figure 67 is a related diagram of an embodiment of the present invention.

Figure 68 is a correlation diagram of an embodiment of the present invention.

Figure 69 is a related diagram of an embodiment of the present invention.

Figure 70 is a related diagram of an embodiment of the present invention.

Figure 71 is a correlation diagram of an embodiment of the present invention.

Figure 72 is a related diagram of an embodiment of the present invention.

Figure 73 is a related diagram of an embodiment of the present invention.

Figure 74 is a correlation diagram of an embodiment of the present invention.

Figure 75 is a related diagram of an embodiment of the present invention.

Figure 76 is a correlation diagram of an embodiment of the present invention.

Figure 77 is a related diagram of an embodiment of the present invention.

Figure 78 is a correlation diagram of an embodiment of the present invention.

Figure 79 is a related diagram of an embodiment of the present invention.

Figure 80 is a related diagram of an embodiment of the present invention.

Fig. 81 (a) and (b) are diagrams related to an embodiment of the present invention.

Figure 82 is a related diagram of an embodiment of the present invention.

Figure 83 is a related diagram of an embodiment of the present invention.

Figure 84 is a related diagram of an embodiment of the present invention.

Figure 85 is a related diagram of an embodiment of the present invention.

Figure 86 is a correlation diagram of an embodiment of the present invention.

Figure 87 is a related diagram of an embodiment of the present invention.

Figure 88 is a related diagram of an embodiment of the present invention.

Figure 89 is a related diagram of an embodiment of the present invention.

Figure 90 is a related diagram of an embodiment of the present invention.

Figure 91 is a related diagram of an embodiment of the present invention.

Figure 92 is a related diagram of an embodiment of the present invention.

Fig. 93 (a) and (b) are diagrams related to an embodiment of the present invention.

Fig. 94 (a) and (b) are diagrams related to an embodiment of the present invention.

Fig. 95 (a) and (b) are diagrams related to an embodiment of the present invention.

Figure 96A is a related diagram of an embodiment of the present invention.

Fig. 96B (a) to (c) are diagrams related to an embodiment of the present invention.

Fig. 97 (A) and (B) are related diagrams of an embodiment of the present invention.

Figure 98 is a related diagram of an embodiment of the present invention.

Fig. 99 (A) and (B) are diagrams related to an embodiment of the present invention.

Figure 100 is a related diagram of an embodiment of the present invention.

Figure 101 is a related diagram of an embodiment of the present invention.

Figure 102 is a related diagram of an embodiment of the present invention.

Figure 103 is a related diagram of an embodiment of the present invention.

Figure 104 is a related diagram of an embodiment of the present invention.

Figure 105 is a related diagram of an embodiment of the present invention.

Figure 106 is a related diagram of an embodiment of the present invention.

Figure 107 is a related diagram of an embodiment of the present invention.

Figure 108 is a related diagram of an embodiment of the present invention.

Figure 109 is a related diagram of an embodiment of the present invention.

Figure 110 is a related diagram of an embodiment of the present invention.

Figure 111 is a related diagram of an embodiment of the present invention.

Figure 112 is a related diagram of an embodiment of the present invention.

Figure 113 is a related diagram of an embodiment of the present invention.

Figure 114 is a related diagram of an embodiment of the present invention.

Figure 115 is a related diagram of an embodiment of the present invention.

Figure 116 is a related diagram of an embodiment of the present invention.

Figure 117 is a related diagram of an embodiment of the present invention.

Figure 118 is a related diagram of an embodiment of the present invention.

Figure 119 is a related diagram of an embodiment of the present invention.

Figure 120 is a related diagram of an embodiment of the present invention.

Figure 121 is a related diagram of an embodiment of the present invention.

Figure 122 is a related diagram of an embodiment of the present invention.

Figure 123 is a related diagram of an embodiment of the present invention.

Figure 124 is a related diagram of an embodiment of the present invention.

Figure 125 is a related diagram of an embodiment of the present invention.

Figure 126 is a related diagram of an embodiment of the present invention.

Figure 127 is a related diagram of an embodiment of the present invention.

Figure 128 is a related diagram of an embodiment of the present invention.

Figure 129 is a related diagram of an embodiment of the present invention.

Figure 130 is a related diagram of an embodiment of the present invention.

Figure 131 is a correlation diagram of an embodiment of the present invention.

Figure 132 is a related diagram of an embodiment of the present invention.

Figure 133 is a related diagram of an embodiment of the present invention.

Figure 134 is a related diagram of an embodiment of the present invention.

Figure 135 is a related diagram of an embodiment of the present invention.

Figure 136 is a related diagram of an embodiment of the present invention.

Figure 137 is a related diagram of an embodiment of the present invention.

Figure 138 is a related diagram of an embodiment of the present invention.

Figure 139 is a related diagram of an embodiment of the present invention.

Figure 140 is a related diagram of an embodiment of the present invention.

Figure 141 is a related diagram of an embodiment of the present invention.

Figure 142 is a related diagram of an embodiment of the present invention.

Figure 143 is a related diagram of an embodiment of the present invention.

Figure 144 is a related diagram of an embodiment of the present invention.

Figure 145A is a related diagram of an embodiment of the present invention.

Figure 145B is a related diagram of an embodiment of the present invention.

Figure 145C is a related diagram of an embodiment of the present invention.

Figure 146A is a related diagram of an embodiment of the present invention.

Figure 146B is a related diagram of an embodiment of the present invention.

Figure 147A is a related diagram of an embodiment of the present invention.

Figure 147B is a related diagram of an embodiment of the present invention.

Figure 148A is a related diagram of an embodiment of the present invention.

Figure 148B is a related diagram of an embodiment of the present invention.

Figure 148C is a related diagram of an embodiment of the present invention.

Figure 149A is a related diagram of an embodiment of the present invention.

Figure 149B is a related diagram of an embodiment of the present invention.

Figure 150 is a related diagram of an embodiment of the present invention.

Figure 151A is a related diagram of an embodiment of the present invention.

Figure 151B is a related diagram of an embodiment of the present invention.

Figure 152 is a related diagram of an embodiment of the present invention.

Figure 153A is a related diagram of an embodiment of the present invention.

Figure 153B is a related diagram of an embodiment of the present invention.

Figure 154 is a related diagram of an embodiment of the present invention.

Figure 155 is a related diagram of an embodiment of the present invention.

Figure 156 is a related diagram of an embodiment of the present invention.

Figure 157A is a related diagram of an embodiment of the present invention.

Figure 157B is a related diagram of an embodiment of the present invention.

Figure 157C is a related diagram of an embodiment of the present invention.

Figure 158 is a related diagram of an embodiment of the present invention.

Figure 159 is a related diagram of an embodiment of the present invention.

Figure 160A is a related diagram of an embodiment of the present invention.

Figure 160B is a related diagram of an embodiment of the present invention.

Figure 161 is a related diagram of an embodiment of the present invention.

Figure 162A is a related diagram of an embodiment of the present invention.

Figure 162B is a related diagram of an embodiment of the present invention.

Figure 163 is a related diagram of an embodiment of the present invention.

Figure 164 is a related diagram of an embodiment of the present invention.

Figure 165 is a related diagram of an embodiment of the present invention.

Figure 166 is a related diagram of an embodiment of the present invention.

Figure 167 is a related diagram of an embodiment of the present invention.

Figure 168 is a related diagram of an embodiment of the present invention.

Figure 169 is a related diagram of an embodiment of the present invention.

Figure 170 is a related diagram of an embodiment of the present invention.

Figure 171 is a related diagram of an embodiment of the present invention.

Figure 172 is a related diagram of an embodiment of the present invention.

Figure 173 is a related diagram of an embodiment of the present invention.

Figure 174 is a related diagram of an embodiment of the present invention.

Figure 175 is a related diagram of an embodiment of the present invention.

Figure 176 is a related diagram of an embodiment of the present invention.

Figure 177 is a related diagram of an embodiment of the present invention.

Figure 178 is a related diagram of an embodiment of the present invention.

Figure 179 is a related diagram of an embodiment of the present invention.

Figure 180 is a related diagram of an embodiment of the present invention.

Figure 181 is a related diagram of an embodiment of the present invention.

Figure 182 is a related diagram of an embodiment of the present invention.

Figure 183 is a related diagram of an embodiment of the present invention.

Figure 184 is a related diagram of an embodiment of the present invention.

Figure 185 is a related diagram of an embodiment of the present invention.

Figure 186 is a related diagram of an embodiment of the present invention.

Figure 187 is a related diagram of an embodiment of the present invention.

Figure 188 is a related diagram of an embodiment of the present invention.

Figure 189 is a related diagram of an embodiment of the present invention.

Figure 190 is a related diagram of an embodiment of the present invention.

Figure 191 is a related diagram of an embodiment of the present invention.

Figure 192 is a related diagram of an embodiment of the present invention.

Figure 193 is a related diagram of an embodiment of the present invention.

Figure 194A is a related diagram of an embodiment of the present invention.

Figure 194B is a related diagram of an embodiment of the present invention.

Figure 195A is a related diagram of an embodiment of the present invention.

Figure 195B is a related diagram of an embodiment of the present invention.

Figure 196 is a related diagram of an embodiment of the present invention.

Figure 197 is a related diagram of an embodiment of the present invention.

Figure 198 is a related diagram of an embodiment of the present invention.

Figure 199 is a related diagram of an embodiment of the present invention.

Figure 200 is a related diagram of an embodiment of the present invention.

Figure 201 is a related diagram of an embodiment of the present invention.

Fig. 202 (a) and (b) are related diagrams of an embodiment of the present invention.

Figure 203 is a related diagram of an embodiment of the present invention.

Figure 204 is a related diagram of an embodiment of the present invention.

Figure 205 is a related diagram of an embodiment of the present invention.

Figure 206 is a related diagram of an embodiment of the present invention.

Figure 207 is a related diagram of an embodiment of the present invention.

Figure 208 is a related diagram of an embodiment of the present invention.

Fig. 209 (a) to (c) are related diagrams of an embodiment of the present invention.

Fig. 210 (a) and (b) are related diagrams of an embodiment of the present invention.

Fig. 211 (a) and (b) are related diagrams of an embodiment of the present invention.

Fig. 212 (a) and (b) are related diagrams of an embodiment of the present invention.

Fig. 213 (a) and (b) are related diagrams of an embodiment of the present invention.

Figure 214 is a related diagram of an embodiment of the present invention.

Figure 215 is a related diagram of an embodiment of the present invention.

Figure 216 is a related diagram of an embodiment of the present invention.

Figure 217 is a related diagram of an embodiment of the present invention.

Fig. 218 (a) to (c) are related diagrams of an embodiment of the present invention.

Fig. 219 (a) to (c) are related diagrams of an embodiment of the present invention.

Fig. 220 (a) to (c) are related diagrams of an embodiment of the present invention.

Fig. 221 (a) and (b) are related diagrams of an embodiment of the present invention.

Figure 222 (a) and (b) are related diagrams of an embodiment of the present invention.

Figure 223 is a related diagram of an embodiment of the present invention.

Figure 224 is a related diagram of an embodiment of the present invention.

(Embodiment of the invention)

Hereinafter, a semiconductor inspection apparatus for inspecting a substrate (i.e., a wafer) having a pattern formed on a surface as an inspection object will be described with reference to the drawings in the embodiment of the present invention. Further, the following embodiments are examples of the inspection apparatus and the inspection method of the present invention, but the present invention is not limited thereto.

In the first and second drawings, the main components of the semiconductor inspection device 1 of the present embodiment are shown in an elevation view and a plan view.

The semiconductor inspection apparatus 1 of the present embodiment includes a cassette holding case 10 for holding a cassette containing a plurality of wafers, and a small environment. The device 20, the main casing 30 partitioned into the working chamber, the loader casing 40 disposed between the small environment device 20 and the main casing 30 and partitioned into two loading chambers, and loading the wafer from the cassette holding box 10 The loader 60 disposed on the platform device 50 in the main casing 30, the electro-optical device 70 mounted on the vacuum casing, the optical microscope 3000, and the scanning electron microscope (SEM) 3002. These devices are arranged in the positional relationship shown in Fig. 1 and Fig. 2A. The semiconductor inspection apparatus 1 further includes a pre-charging unit 81 disposed in the vacuum main casing 30, a potential applying mechanism 83 (shown in FIG. 14) for applying a potential to the wafer, and an electron beam calibration mechanism 85 (FIG. Shown in Fig. 15) and an optical microscope 871 which constitutes an alignment control device 87 for wafer positioning on a platform device. The electro-optical device 70 has a lens barrel 71 and a light source tube 7000. The internal structure of the electro-optical device 70 will be described later.

- Card Holder -

A plurality of (two in this embodiment) cassettes c are held in the cassette holding case 10 (for example, SMIF (standard mechanical interface) manufactured by Ashisuto Co., FOUP (front opening type crystal) A closed cassette such as a round carrying case) in which a plurality of sheets (for example, 25 sheets) are accommodated in parallel in a vertical direction. In the form of the cassette holding case, there is a structure suitable for carrying a cassette by a robot or the like and automatically loading the cassette holding case 10, and an open cassette structure suitable for manual loading, and any of the cassettes can be used. Select settings. In this embodiment, the cassette holding case 10 is in the form of automatically loading the cassette c, and includes, for example, an elevating table 11 and an elevating mechanism 12 for moving the elevating table 11 up and down. The cassette c can be automatically placed on the lifting table in the state shown by the chain line of Fig. 2A. After placement, automatically rotate to The state shown by the solid line in Fig. 2A is directed to the rotation axis of the first transport unit in the small environment device. Further, the lifting platform 11 is lowered to the state shown by the chain line in the first drawing. As described above, the cassette holding case used for the automatic loading or the cassette holding case used for the manual loading is a well-known structure, and a detailed description of the structure and function thereof is omitted as long as it is appropriately selected.

In another embodiment, as shown in FIG. 2B, a plurality of 300 mm substrates are housed in a state of being stored in a pocket (not shown) fixed inside the case body 501 for transporting. Keep it, etc. The substrate carrying case 24 is a box body 501 having a rectangular tubular shape, and is connected to a substrate loading/unloading automatic opening and closing device, and can be used to mechanically open and close the side opening portion of the case body 501. On the opposite side, a cover 503 for covering the opening for loading and unloading the filter and the fan motor; and a grooved bag (not shown) for holding the substrate W, and a ULPA (ultra low penetration air) filter 505, a chemical filter 506, and a fan motor 507 are formed. In this embodiment, the robot-type first transport unit 612 of the loader 60 carries the substrate in and out.

Further, the substrate, that is, the wafer, which is housed in the cassette c, is a wafer to be inspected. This inspection is performed in the semiconductor manufacturing process after the process of processing the wafer or in the middle of the program. Specifically, a substrate (that is, a wafer) that has been subjected to a film formation process, chemical mechanical polishing (CMP), ion implantation, a wafer on which a wiring pattern is formed, or a wafer that has not yet formed a wiring pattern is stored in a card. Inside. The wafers accommodated in the cassette c are arranged such that a plurality of sheets are spaced apart in the vertical direction and arranged in parallel. In order to hold the wafer at an arbitrary position by the first transfer unit described later, the first transport sheet can be used. The arm of the element is set to be movable up and down.

- Small environmental installations -

In the first to third figures, the small-sized environment device 20 is provided with a casing 22 for constituting a small environmental space 21 for performing atmosphere control, and a gas such as clean air is circulated in the small-sized environmental space 21. a gas circulation device 23 for controlling the atmosphere; a discharge device 24 for collecting and discharging a part of the air supplied to the small environmental space 21; and a discharge device 24 disposed in the small environment space 21 for the substrate to be inspected (also That is, the wafer) is a pre-aligner 25 that performs a rough positioning.

The casing 22 has a top wall 221, a bottom wall 222, and a peripheral wall 223 surrounding the periphery, and is configured to separate the small environmental space 21 from the outside. In order to control the atmosphere of the small-sized environmental space, the gas circulation device 23 is provided in the small-sized environmental space 21, and is attached to the top wall 221 to purify the gas (in the present embodiment, air) and pass through. One or more gas outlets (not shown) allow clean air to flow downwardly into a laminar gas supply unit 231; in a small environmental space, disposed on the bottom wall 222 for flowing downward toward the bottom A recovery duct 232 that blows air for recovery; and a duct 233 for connecting the recovery duct 232 and the gas supply unit 231 to return the recovered air to the gas supply unit 231. In the present embodiment, the gas supply unit 231 takes in approximately 20% of the supplied air from the outside of the casing 22 to be purified. The ratio of the gas taken in from the outside can be arbitrarily selected. The gas supply unit 231 has a well-known configuration type HEPA (High Efficiency Particle Air) or ULPA filter to produce clean air. a laminar downflow of clean air (ie, a blown airflow), mainly by configuration The conveyance surface generated by the first conveyance unit, which will be described later, in the small-sized environmental space 21 is supplied to prevent the dust particles generated by the conveyance unit from adhering to the wafer. Therefore, the air outlet of the lower blowing airflow does not have to be located close to the top wall as shown in the drawing, and may be provided on the upper side of the conveying surface generated by the conveying unit. Moreover, it does not have to flow over the entire surface of a small environmental space. In addition, the ionic wind can be used as clean air to ensure cleanliness. Moreover, a sensor for observing cleanliness can also be provided in a small environmental space, and the device is turned off when the cleanliness is deteriorated. In the peripheral wall 223 of the casing 22, an inlet and outlet 225 is formed in a portion adjacent to the cassette holding case 10. A known structural closing device may be provided in the vicinity of the entrance and exit 225, and the inlet and outlet 225 may be closed from the side of the small environment device. The laminar flow of the blown air generated near the wafer may be carried out by, for example, a flow rate of 0.3 to 0.4 m/sec. The gas supply unit is not only disposed in a small environmental space but also on the outside.

The discharge device 24 includes a suction duct 241 disposed at a lower portion of the transport unit than a wafer transport surface of the transport unit, a blower 242 disposed outside the casing 22, and a suction duct 241 and a blower 242. Catheter 243. The discharge device 24 sucks the gas which is moved to the periphery of the transport unit and contains dust particles which may be generated by the transport unit by the suction duct 241, and is discharged to the casing 22 via the ducts 243 and 244 and the blower 242. The outside. In this case, it may also be discharged into an exhaust pipe (not shown) that is pulled to the vicinity of the casing 22.

The pre-aligner 25 disposed in the small environment space 21 is used to form an oriented flat portion of the wafer (meaning flatness formed on the outer circumference of the circular wafer) a portion, hereinafter referred to as a flattening portion) or one or more V-notches (ie, grooves) formed on the outer periphery of the wafer, are optically or mechanically detected to thereby position the rotational direction about the wafer axis OO Pre-positioned with a precision of about ±1 degree. The pre-aligner is a part of the coordinate determining mechanism constituting the inspection object of the invention described in the patent application, and serves as a rough positioning of the inspection object. The pre-aligner itself may be of a well-known structure, and the description of its structure and operation is omitted.

Further, although not shown, a recovery duct for the discharge device may be provided in the lower portion of the pre-aligner, whereby the dust-containing air discharged from the pre-aligner is discharged to the outside.

-Host shell -

In the first and second figures, the main casing 30 constituting the work chamber 31 is provided with an organic casing main body 32. The casing body 32 is supported by a casing supporting device 33 provided on a vibration blocking device (i.e., the vibration isolating device 37) disposed on the gantry 36. The cabinet supporting device 33 has a frame-shaped structural body 331 assembled in a rectangular shape. The casing body 32 is disposed and fixed to the frame structure 331 and includes a bottom wall 321 that is placed on the frame structure, a top wall 322, and a peripheral wall that is connected to the bottom wall 321 and the top wall 322 and surrounds the periphery. 323, while the workroom 31 is isolated from the outside. In the embodiment, the bottom wall 321 is formed of a thick steel plate in order to prevent deformation due to an increase in weight due to a device such as a platform device placed above, but another structure may be employed. In this embodiment, the casing body 32 and the casing support device 33 are assembled in a rigid structure, and the vibration from the floor on which the gantry 36 is placed is prevented from being transmitted to the rigid structure by the vibration isolating device 37. The peripheral wall 323 of the casing body 32 In the peripheral wall adjacent to the loader casing to be described later, a wafer inlet/outlet 325 is formed.

Further, the anti-vibration device may be an active device having an air spring, a magnetic bearing, or the like, or a passive device having such components. Since any of the known structures can be used, its own construction and function description are omitted. The workroom 31 is a vacuum device (not shown) having a known structure to maintain a vacuum atmosphere. Below the gantry 36, a control device 2 for controlling the overall operation of the apparatus is disposed.

- loader casing -

In the first, second, and fourth figures, the loader casing 40 includes a casing body 43 for constituting the first loading chamber 41 and the second loading chamber 42. The casing body 43 has a bottom wall 431, a top wall 432, a peripheral wall 433 surrounding the periphery, and a partition wall 434 separating the first loading chamber 41 and the second loading chamber 42 to isolate the two loading chambers from the outside. The dividing wall 434 forms an opening for the wafer to be taught between the two loading chambers, that is, the inlet and outlet 435. Further, the peripheral wall 433 is formed with inlets and outlets 436 and 437 at portions adjacent to the small environment device and the main casing. The casing body 43 of the loader casing 40 is mounted on the frame structure 331 of the casing support device 33 for support. Therefore, the vibration of the floor is not transmitted to the loader casing 40. The inlet and outlet 436 of the loader casing 40 and the inlet and outlet 226 of the casing 22 of the small-sized environmental unit are integrated, and a closing device 27 for selectively preventing communication between the small environmental space 21 and the first loading chamber 41 is provided therein. The closing device 27 has a sealing member 271 that surrounds the periphery of the inlet and outlet ports 226 and 436 and is closely fixed to the side wall 433; and a sealing member 271 that cooperates to block the flow of air through the inlet and outlet. 272; and a driving device 273 that drives the threshold. Further, the inlet and outlet 437 of the loader casing 40 and the inlet and outlet 325 of the casing body 32 are integrated, and a closing device 45 for selectively sealing and preventing the second loading chamber 42 from communicating with the working chamber 31 is provided. The closing device 45 has a sealing member 451 which surrounds the inlet and outlet ports 437 and 325 and is closely adhered to and fixed to the side walls 433 and 323, and a sealing member 451 which cooperates with the sealing member 451 to block air from flowing through the inlet and outlet; and a driving for driving the threshold Device 453. Further, the opening formed in the partition wall 434 is provided with a closing device 46 that closes the door 461 to selectively seal the communication between the first and second loading chambers. When the closing devices 27, 45 and 46 are in the closed state, the chambers can be hermetically sealed. These closing devices may be of a known device, and detailed descriptions of their construction and operation are omitted. In addition, the support method of the casing 22 of the small-sized environment device 20 is different from the support method of the loader casing 40, and in order to prevent vibration from the floor from being transmitted to the loader casing 40 and the main casing 30 via the small-sized environmental device, the machine is in the machine Between the casing 22 and the loader casing 40, the vibration-damping cushioning material may be disposed in an airtight manner around the inlet and outlet.

In the first loading chamber 41, a wafer holder 47 in which a plurality of wafers (two sheets in the present embodiment) are supported in a horizontal state is provided in a horizontal state. As shown in FIG. 5, the wafer holder 47 is provided with four pillars 471 and 474 formed in two corners of the rectangular substrate 471 in an upright state with the pillars 472 spaced apart from each other. The periphery of the wafer W is placed on top of the support portion. Then, the tips of the arm portions of the first and second transport units, which will be described later, can approach the wafer from between adjacent pillars, and the wafer can be taken up by the arm portion.

The load chambers 41 and 42 are controlled by a known structural vacuum exhauster (not shown) including a vacuum pump (not shown) in a high vacuum state (in terms of vacuum, the system 10)^-5To 10^-6Pa). In this case, the first load chamber 41 can be maintained in a low vacuum atmosphere as a low vacuum chamber, and the second load chamber 42 can be maintained in a high vacuum atmosphere as a high vacuum chamber to effectively prevent contamination of the wafer. By adopting such a configuration, the next wafer to be subjected to the defect inspection in the loading chamber can be transported into the work chamber without delay. By adopting such a loading chamber, it is possible to increase the throughput of the defect inspection, and to maintain a high vacuum state as much as possible in the vacuum state around the electron source in which the storage state is required to be maintained in a high vacuum state.

Each of the first and second loading chambers 41 and 42 is connected to a vacuum exhaust pipe and a ventilation pipe for inert gas (for example, dry pure nitrogen gas) (all not shown). Thereby, the atmospheric pressure state in each of the loading chambers can be achieved by an inert gas venting pipe (injecting an inert gas to prevent oxygen or the like other than the inert gas from adhering to the surface). The apparatus for performing such inert gas aeration can be a known structure itself, and a detailed description thereof will be omitted.

- Platform device -

The platform device 50 includes a fixed platform 51 disposed on the bottom wall 321 of the main casing 30, and a Y platform 52 moving in the Y direction (the direction perpendicular to the paper surface in FIG. 1) on the fixed platform; on the Y platform The X platform 53 moving in the X direction (in the first figure, in the left-right direction); the rotating platform 54 rotatable on the X platform; and the holding box 55 disposed on the rotating platform 54. The wafer is releasably held on the wafer mounting surface 551 of the holding case 55. Keep the box to release the wafer mechanically or by electrostatic chucking A well-known structure can be taken. The platform device 50 is operated by a servo motor, an encoder, and various sensors (not shown) to hold the above-described plurality of stages, and holds the mounting surface 551 on the wafer in the holding case, and the electrons irradiated from the electron optical device. The beam is positioned with high precision in the X direction, the Y direction, and the Z direction (in the first figure, in the up and down direction), and is highly precise about an axis direction (θ direction) perpendicular to the wafer support surface. Positioning. Further, the positioning in the Z direction may be performed by, for example, slightly adjusting the position of the mounting surface on the holding case in the Z direction. In this case, the reference position of the mounting surface is detected by a position measuring device for a small-diameter laser (a laser interference distance measuring device using the principle of the jammer), and the position is controlled by a feedback circuit (not shown); At the same time (or instead) to determine the position of the groove or flattening of the wafer to detect the planar position and rotational position of the wafer relative to the electron beam, and to rotate using a stepper motor or the like that can be controlled by micro angle. The platform is subjected to rotation control. In order to prevent dust from occurring in the work chamber as much as possible, the servo motors 521 and 531 and the encoders 522 and 532 for the platform device are disposed outside the main casing 30. In addition, the platform device 50 may be a known structure such as a stepper, and a detailed description of its structure and operation will be omitted. Further, the above-described laser interference distance measuring device may be a well-known structure, and a detailed description of its structure and operation will be omitted.

The reference signal of the obtained signal can be obtained by inputting the rotational position or the X and Y positions of the wafer with respect to the electron beam in advance to a signal detecting system or an image processing system to be described later. Further, the wafer chuck mechanism provided in the holding case is configured to input a voltage for holding the wafer to the electrode of the electrostatic chuck to hold the three points on the outer peripheral portion of the wafer (preferably in the circumferential direction, etc.) Positioning is performed at intervals). Wafer chuck mechanism with two fixed positioning pins and one push type Crank pin. The crank pin is configured to realize automatic clamping and automatic release, and is configured to have a conductive portion to which a voltage is applied.

Further, in the second embodiment, the platform that moves in the left-right direction is the X platform, and the platform that moves in the vertical direction is the Y platform. However, the platform that moves in the left-right direction may be used as the Y platform in the figure. The platform that moves up and down is used as the X platform.

- loader -

The loader 60 includes a robot-type first transport unit 61 disposed in the casing 22 of the small-sized environment device 20 and a robot-type second transport unit 63 disposed in the second load chamber 42.

The first transport unit 61 has an axis O with respect to the drive unit 611₁-O₁A rotating multi-section arm 612. In the multi-section arm, any structure can be used. However, in this embodiment, there are three portions that are rotatably mounted. One portion of the arm portion 612 of the first transport unit 61, that is, the first portion closest to the drive portion 611 side is attached to a known structural drive mechanism (not shown) that can be provided in the drive portion 611. Rotating shaft 613. The arm portion 612 is transmitted through the shaft 613 about the axis O₁-O₁Rotate, and can use the relative rotation between the parts to the axis O as a whole₁-O₁Scales in the radial direction. A distal end of the third portion farthest from the shaft 613 of the arm portion 612 is provided with a pick-up device 616 for picking up the wafer, such as a mechanical chuck or an electrostatic chuck known in the art. The drive unit 611 is configured to be movable in the vertical direction by the known structural type elevating mechanism 615.

In the first transport unit 61, the arm portion 612 protrudes in a direction M1 or M2 held in either of the two cassettes c of the cassette holding case, and is housed in the direction M1 or M2. One wafer in the cassette c is placed on the arm or taken up and taken out by a chuck (not shown) attached to the front end of the arm. Then, the arm portion is retracted (as shown in Fig. 2), and is rotated until it can be extended to the direction M3 of the pre-aligner 25, and is stopped at this position. As such, the arms will again extend and the wafer held on the arms will be placed in the pre-aligner 25. After receiving the wafer from the pre-aligner opposite to the foregoing, the arm is rotated again and stopped at a position (direction M4) at which the second loading chamber 41 can be extended, and the wafer is delivered into the second loading chamber 41. Wafer holder 47. Further, when the wafer is mechanically taken, the peripheral portion of the wafer (a range of about 5 mm from the peripheral edge) is taken. This is because, in addition to the peripheral portion of the wafer, a device (circuit wiring) is formed on the entire surface, and if the portion is taken, the device is broken and defects are generated.

The second transport unit 63 has basically the same structure as the first transport unit, and differs in the transport of the wafer between the wafer holder 47 and the mounting surface of the platform device. .

In the loader 60, in the first and second transport units 61 and 63, the wafer is transported from the cassette held by the cassette holding case to the table device 50 disposed in the work chamber 31, and the wafer is reversed. The conveyance is carried out while being kept at a substantially level. The up-and-down operation of the arm of the transport unit is performed only when the wafer is taken out from the cassette and inserted into the wafer, the wafer is placed on the wafer holder, taken out from the wafer, and the wafer is placed on and removed from the platform apparatus. Therefore, the movement of large wafers, such as wafers having a diameter of 30 cm, can also proceed smoothly.

- Wafer handling -

Next, the operation of transporting the wafer from the cassette c supported by the cassette holding case to the stage unit 50 disposed in the work chamber 31 will be described in order.

When the cassette holding case 10 utilizes the manual placement of the cassette in the above manner, a structural design suitable for this manner is used, and when the cassette is placed in an automatic manner, a configuration design suitable for the manner is used. In the present embodiment, the cassette c is placed on the lifting platform 11 of the cassette holding box 10, and the lifting platform 11 is lowered by the lifting mechanism 12 to integrate the cassette c into the inlet and outlet 225.

The cassette is integrated into the inlet and outlet 225, and the lid (not shown) provided on the cassette is opened, and a cylindrical cover is disposed between the cassette c and the inlet and outlet 225 of the small environment device, and the cassette interior and the small environment are arranged. The interior and exterior of the space are partitioned. These configurations are well-known techniques, and detailed descriptions of their constructions and operations are omitted. Further, when the small environment device 20 side is provided with a closing device for opening and closing the door 225, the closing device operates to open the door 225.

On the other hand, the arm portion 612 of the first transport unit 61 is stopped in the M1 or M2 direction (in the M1 direction in the present description), and the arm opening 225 is opened, and the arm portion is extended, and the front end will be One piece is picked up in the wafer contained in the cassette. Further, the arm portion and the position of the wafer to be taken out from the cassette are adjusted in the vertical direction. In the present embodiment, the driving portion 611 of the first transport unit 61 and the arm 612 are moved up and down, and the holding box may be stuck. The up and down movement of the lifting platform or both.

When the operation of picking up the wafer by the arm portion 612 is completed, the arm portion is retracted, the closing device is operated to close the inlet and outlet (when the closing device is provided), and then the arm portion 612 is wound around the axis O.₁-O₁Rotate to become a state that can be elongated in the M3 direction. Then, the arm is extended to mount the wafer placed on the front end or sandwiched by the chuck on the pre-aligner 25, and the wafer is rotated in the direction of the wafer by the pre-aligner The orientation (with an orientation perpendicular to a central axis of the wafer plane) is positioned within a predetermined range. When the positioning is completed, the first transport unit 61 receives the wafer from the pre-aligner 25 to the tip end of the arm, and then retracts the arm portion, thereby causing the arm portion to extend in the direction M4. Then, the threshold 227 of the closing device 27 operates to open the inlets and outlets 226 and 436, and then the arm portion 612 is extended to place the wafer on the upper side or the lower side of the wafer holder 47 in the first loading chamber 41. Further, as described above, before the closing device 27 is opened to allow the wafer to be placed on the wafer holder 47, the opening 435 formed in the partition wall 434 is closed in an airtight state by the threshold 461 of the closing device 46.

In the process of transporting the wafer by the first transport unit, the clean air flows in a laminar flow (in the form of a downward blowing airflow) from the gas supply unit 231 provided in the casing of the small-sized environment device, and is prevented during transportation. Dust particles adhere to the wafer. A part of the air around the transport unit (in the present embodiment, about 20% of the air supplied from the supply unit, mainly contaminated air) is sucked by the suction duct 241 of the discharge device 24 and discharged outside the casing. The remaining air is recovered through a recovery conduit 232 provided at the bottom of the casing and returned to the gas supply unit 231.

When the wafer is placed in the wafer holder 47 in the first loading chamber 41 of the loader casing 40 through the first conveyance unit 61, the closing device 27 is closed, and the inside of the load chamber 41 is sealed. Then, the first load chamber 41 is filled with an inert gas to discharge the air, and then the inert gas is also discharged, so that the inside of the load chamber 41 becomes a vacuum atmosphere. The vacuum atmosphere of the first loading chamber may be a low degree of vacuum. When the degree of vacuum in the load chamber 41 reaches a certain level, the closing device 46 operates to open the entrance and exit that has been sealed by the threshold 461. 434 is opened, and the arm portion 632 of the second transport unit 63 is extended and a wafer is picked up from the wafer holder 47 by the holding device at the front end (placed on the front end or held by the chuck attached to the front end). At the end of the wafer picking up, the arm is retracted and the closing device 46 is again actuated to close the door 435 with the threshold 461. Further, before the closing device 46 is opened, the arm portion 632 can be previously extended in the direction N1 of the wafer holder 47. Further, as described above, before the closing device 46 is opened, the inlets 437 and 325 are closed by the threshold 452 of the closing device 45, and the communication in the second loading chamber 42 and the working chamber 31 is prevented in an airtight state. 2 Vacuum evacuation is performed in the load chamber 42.

When the closing device 46 closes the inlet and outlet 435, the second loading chamber is again evacuated by vacuum and is set to have a higher degree of vacuum than the first loading chamber. During this period, the arm portion of the second transport unit 61 is rotated to a position that can be extended in the direction of the stage device 50 in the work chamber 31. On the other hand, on the platform device in the workroom 31, the Y platform 52 moves to the top of the second figure until the center line X of the X platform 53₀-X₀And the axis of rotation O passing through the second transport unit 63₂-O₂X axis X₁-X₁The position is approximately the same, and the X platform 53 moves to a position closest to the leftmost side in Fig. 2, and stands by in this state. When the second loading chamber arrives at substantially the same vacuum state as the working chamber, the threshold 452 of the closing device 45 operates to open the inlets and outlets 437, 325, and the front end of the arm portion that has been extended and holds the wafer is then close to the working chamber 31. Platform device. The wafer is then placed on the mounting surface 551 of the platform unit 50. At the end of the wafer placement, the arms are retracted and the closing device 45 closes the ports 437, 325.

The above is the movement of the wafer in the cassette c to the platform device. To be explained, when the wafer placed on the platform device and the end of the processing is returned from the platform device to the cassette c, the reverse operation is performed. Further, since the wafer holder 47 is placed with a plurality of wafers in advance, the first transport unit can be used for the cassette and the wafer holder during the wafer transfer between the wafer holder and the platform device by the second transfer unit. The wafer is transported between them, so that the inspection process can be performed with good efficiency.

Specifically, when the wafer carrier 47 of the second transport unit has the processed wafer A and the unprocessed wafer B,

(1) First, the unprocessed wafer B is moved to the stage device 50, and processing is started.

(2) In this process, the processed wafer A is moved from the stage device 50 to the wafer holder 47 by the arm portion, and the unprocessed wafer C is also extracted from the wafer holder by the arm portion and pre-paired. After the positioning is performed, the wafer holder 47 is moved to the loading chamber 41.

By such an operation, in the wafer holder 47, the processed wafer A can be replaced with the unprocessed wafer C during the wafer B processing.

In addition, according to the manner in which such a device is inspected or evaluated, a plurality of wafers can be processed by moving the wafers from one wafer holder 47 to each device by placing the platform devices 50 in parallel. .

Fig. 6 is a view showing a variation of the host casing supporting method. In the modification shown in Fig. 6, the casing supporting device 33a is formed of a thick rectangular steel plate 331a, and the casing main body 32a is placed on the steel plate. Therefore, the bottom wall 321a of the casing body 32a is formed into a thinner structure than the bottom wall of the foregoing embodiment. In the variation shown in Fig. 7, the casing support device 33b is utilized. The frame structure 336b supports the casing body 32b and the loader casing 40b in a hanging state. The lower ends of the plurality of vertical frames 337b fixed to the frame structure 336b are fixed to the four corners of the bottom wall 321b of the casing body 32b, and are configured to support the peripheral wall and the top wall by the bottom wall thereof. Next, the vibration isolating device 37b is disposed between the frame structure 336b and the gantry 36b. Further, the loader casing 40 is also suspended by a hanging member 49b fixed to the frame structure 336. In the variation shown in Fig. 7, since the casing body 32b is supported by the hanging type, the main casing and various devices provided therein can reduce the center of gravity as a whole. In the support method including the main casing and the loader casing of the above-described variation, the vibration from the floor is not transmitted to the main casing and the loader casing.

In other variations not shown, only the casing body of the main casing is supported from below by the casing supporting device, and the casing of the loader is disposed on the floor in the same manner as the adjacent small environmental device. Further, in another modification not shown, only the casing body of the main casing is supported by the frame structure in a hanging manner, and the loader casing is disposed on the floor in the same manner as the adjacent small environment device.

According to the above embodiment, the following effects can be achieved.

(A) The overall configuration of the image projection type inspection apparatus using electron beams can be obtained, and the inspection object can be processed with high throughput.

(B) The object to be inspected can be cleaned in a small environmental space to prevent dust from adhering, and the degree of cleanness can be observed by providing a sensor, and the object to be inspected can be inspected while monitoring the dust in the space.

(C) because the load chamber is supported in an integrated manner by the vibration preventing device Since the working room is not affected by the external environment, the platform device can be supplied and inspected.

-Electronic optical device -

The electro-optical device 70 includes a lens barrel 71 fixed to the casing main body 32, and is provided with a primary light source optical system (hereinafter simply referred to as "primary optical system") 72 as shown schematically in Fig. 8, and a secondary electron optical system. The optical system (hereinafter referred to as "secondary optical system") 74 and the detection system 76. The primary optical system 72 is an optical system that irradiates light onto the surface of the wafer W to be inspected, and includes light that emits light. A light source 10000 and a mirror 10001 that changes the angle of the light. In this embodiment, the optical axis of the light ray 10000A emitted from the light source is inclined with respect to the optical axis of the photoelectrons emitted from the wafer W to be inspected (perpendicular to the surface of the wafer W).

The detection system 76 includes a detector 761 and an image processing unit 763 that are disposed on the imaging surface of the lens system 741.

-Light source (light source) -

In the present embodiment, the light source 10000 is a DUV (deep ultraviolet light) laser light source. DUV laser light is emitted from the DUV laser source 10000. In addition, UV, DUV, EUV (extreme ultraviolet light) light and laser, and X-ray and X-ray laser, etc., as long as the substrate is illuminated by the light from the light source 10000, the other light source is also be usable.

-1 optical system -

The primary light is formed by the light emitted from the light source 10000, and a rectangular, or circular (also round) beam is irradiated on the surface of the wafer W. It is called a primary optical system. The light emitted from the light source 10000 is irradiated onto the wafer WF on the stage device 50 through the objective optical system 724 as a primary ray.

-2 times optical system -

A two-dimensional image formed by photoelectrons generated by light irradiated on the wafer W is passed through a hole formed in the mirror 10001, and passed through a numerical aperture 10008 by electrostatic lenses (transmission lenses) 10006 and 10009. The field of view is imaged at a contracted position, and the lens 741 of the subsequent stage is enlarged and projected, and detected by the detection system 76. This imaging projection optical system is referred to as a secondary optical system 74.

At this time, a negative bias is applied to the wafer. The potential difference between the electrostatic lens 724 (the lenses 724-1 and 724-2) and the wafer accelerates the photoelectrons generated from the sample surface, thereby reducing the chromatic aberration. The objective optical system 724 has an extracted electric field of 3 kV/mm to 10 kV/mm, which is a very high electric field. When the electric field is increased, the effect of reducing the aberration is related to the improvement of the resolution. On the other hand, when the electric field is increased, the voltage gradient becomes large, and discharge is likely to occur. Therefore, it is important to extract the electric field with an appropriate value. The electrons amplified by the lens 724 (CL) to a predetermined magnification are converged by the lens (TL1) 10006, and a polycondensation (Crossover, CO) is formed on a numerical aperture (NA) 10008. Further, magnification can be amplified by a combination of the lens (TL1) 10006 and the lens (TL2) 10009. Thereafter, the projection is enlarged with a lens (PL) 741, and imaged on the MCP (Micro Channel Plate) of the detector 761. In this optical system, an NA is disposed between TL1 and TL2, and by optimizing it, light that can be reduced by off-axis aberration is formed. Learning system.

-Detector-

The photoelectron image sent by the wafer imaged by the secondary optical system is first amplified by a microchannel plate (MCP) and then converted into a light image by colliding with the phosphor screen. In the principle of MCP, millions of very thin conductive glass capillaries having a diameter of 6 to 25 μm and a length of 0.24 to 1.0 mm are bundled and shaped into a thin plate shape, and a capillary tube is applied by applying a predetermined voltage. Act as a separate electronic amplifier to form an electronic amplifier as a whole.

The image converted into light by the detector is projected to the TDI (Time Delay Integration) by a FOP (Fiber Optical Plate) system placed in the atmosphere via a vacuum penetration window. - CCD (Charge Coupled Device). Further, in other methods, the FOP coated with the fluorescent material is sometimes connected to the TDI sensor surface, and the signal subjected to the electronic/optical conversion in the vacuum is introduced to the TDI sensor. In this way, the transmittance or MTF (Modulation Transfer Function) efficiency is better than in the case of being placed in the atmosphere. For example, in the transmittance and the MTF, a height value of ×5 to ×10 can be obtained. At this time, as the detector, as described above, although MCP+TDI is used, it may be replaced by EB (Electron Bombardment)-TDI or EB-CCD. When EB-TDI is used, photoelectrons generated from the surface of the sample and forming a two-dimensional image are directly incident on the EB-TDI sensor surface, so that an image signal can be formed without deterioration of resolution. For example, in the case of MCP+TDI, after electronic amplification with MCP, it will be applied by fluorescent material, scintillator, etc. Electronic/optical conversion, the information of the light image reaches the TDI sensor. On the other hand, when EB-TDI or EB-CCD is used, there is no transmission/loss of electron/optical conversion or optical enhancement information, so there is no deterioration of the image, and the signal reaches the sensor. For example, when MCP+TDI is used, the MTF or contrast becomes 1/2 to 1/3 as compared with EB-TDI or EB-CCD.

Further, in this embodiment, it is assumed that a high voltage of 10 to 50 kV is applied to the objective lens system 724, and the wafer W is provided.

- The relationship between the main functions of the image projection method and the description of the overall image -

Fig. 9 is a view showing the overall configuration of the embodiment. However, a part of the illustration has been omitted.

In Fig. 9, the inspection apparatus has a lens barrel 71, a light source tube 7000, and a chamber 32. A light source 10000 is provided inside the light source tube 7000, and an optical system 72 is disposed on the optical axis of the light (primary light) emitted from the light source 10000. Further, a platform device 50 is disposed inside the chamber 32, and a wafer W is placed on the platform device 50.

On the other hand, inside the lens barrel 71, cathode lenses 724 (724-1 and 724-2) and transfer lenses (Transfer Lens) 10006 and 10009 are disposed on the optical axis of the secondary beam emitted from the wafer W. , numerical aperture (NA) 10008, lens 741 and detector 761. Further, the numerical aperture (NA) 10008 corresponds to an aperture stop, which is a metal (Mo or the like) thin plate having a circular hole.

On the other hand, the output of the detector 761 is input to the control unit 780, and the output of the control unit 780 is input to the CPU (Central Processing Unit) 781. The control signal of the CPU 781 is input to the light source control unit 71a, the lens barrel control unit 71b, and the stage drive mechanism 56. The light source control unit 71a performs light The source control of the source 10000, the lens barrel control unit 71b performs a cathode lens (Cathod Lens) 724, lenses 10006 and 10009, lens voltage control of the lens 741, and voltage control of the aligner (not shown) (bias vector control). .

Further, the platform drive mechanism 56 transmits the position information of the platform to the CPU 781. Further, the light source cylinder 7000, the lens barrel 71, and the chamber 32 are connected to a vacuum exhaust system (not shown), and are exhausted through the turbine pump of the vacuum exhaust system to maintain the internal vacuum state. Further, on the downstream side of the turbo pump, a vacuum exhaust system that performs rough suction using a general dry pump or a rotary pump is provided.

When primary light is applied to the sample, photoelectrons are generated from the light-irradiated surface of the wafer W as a secondary beam.

The secondary beam is guided to the detector by the cathode lens 724, the TL lens groups 10006 and 10009, and the lens (PL) 741.

However, the cathode lens 724 is composed of three electrodes. The lowermost electrode is designed to form a positive electric field between it and the potential on the side of the sample W to pull in electrons (especially the second electrons with less directivity) to be introduced into the lens with excellent efficiency. Therefore, if the cathode lens forms two telecentrics, it is very effective. The secondary beam imaged through the cathode lens passes through the aperture of mirror 10001.

When the cathode lens 724 images the secondary beam in only one stage, the lens action becomes strong, and aberration is likely to occur. Therefore, a two-stage double lens system is provided to perform imaging once. At this time, the intermediate imaging position is between the lens (TL1) 10006 and the cathode lens 724. At this time, as described above, if two telecentricities are provided, the reduction of the aberration is very effective. Secondary beam The cathode lens 724 and the lens (TL1) 10006 are constricted on the numerical aperture (NA) 10008 to form a polycondensation (crossover), and an imaging is performed between the lens 724 and the lens (TL1) 10006, and then the lens is used ( TL1) 10006 and lens (TL2) 10009 determine the intermediate magnification, which is amplified by lens (PL) 741 and imaged on detector 761. That is, in this example, the total imaging is performed 3 times.

Furthermore, the lenses 10006 and 10009 and the lens 741 are all lenses of a rotational axisymmetric type called a Unipotential lens or a single lens (Einzel Lens). Each lens has a configuration of three electrodes. Usually, the two outer electrodes are set to a zero potential, and a lens is applied by a voltage applied to the center electrode to control the lens. Further, the present invention is not limited to such a lens structure, and has a structure for adjusting the focus adjustment electrode in the first stage, the second stage, or both of the lenses 724, or a focus adjustment electrode that operates dynamically, and belongs to the 4-pole. The situation is either a 5-pole situation. Further, the PL lens 741 is also effective in that the field lens function is added to reduce the off-axis aberration and the magnification is increased to provide four or five poles.

The secondary beam is magnified and projected using a secondary optical system and imaged on the detection surface of the detector 761. The detector 761 is composed of an MCP that magnifies electrons, a fluorescent plate that converts electrons into light, a lens or other optical element for relaying and transmitting optical images of a vacuum system and the outside, and an imaging element (CCD, etc.). Composition. The secondary beam is imaged and amplified on the MCP detection surface, and the electrons are converted into optical signals by the fluorescent plate, and converted into photoelectric signals by the imaging elements.

The control unit 780 reads the image signal of the wafer W from the detector 761 and transmits it to the CPU 781. CPU781 uses sample ratio from image signal Perform defect inspection of a pattern such as (Template matching). Further, the platform device 50 can be moved in the XY direction by the platform drive mechanism 56. The CPU 781 reads the position of the platform device 50, and outputs the drive control signal to the platform drive mechanism 56 to drive the platform device 50 to sequentially perform image detection and inspection.

Further, even if the magnification of the lens is changed, even if the setting magnification of the lens conditions of the lenses 10006 and 10009 is changed, a uniform image can be obtained on the entire surface of the detection side. Further, in the present embodiment, a uniform image having no unevenness can be obtained. However, when the magnification is set to a high magnification, the brightness of the image is lowered. Therefore, in order to improve this problem, when the lens condition of the secondary optical system is changed to change the magnification, the lens condition of the primary optical system is set such that the amount of electrons emitted per unit pixel is fixed.

- Pre-charging unit -

As shown in FIG. 1, the pre-charging unit 81 is disposed adjacent to the lens barrel 71 of the electro-optical device 70 in the working chamber 31. In the inspection apparatus, the electron beam is irradiated to the substrate (that is, the wafer) to be inspected, and the device pattern formed on the surface of the wafer is applied to the inspection form, so that the light is irradiated. The information of the light electronics is used as information on the surface of the wafer. However, there are cases where the surface of the wafer is charged-up due to the conditions of the wafer material, the illuminating light, or the wavelength and energy of the laser. Moreover, there is also the possibility of generating a portion with a strong charge and a weak portion on the surface of the wafer. When the amount of charge on the surface of the wafer is staggered, the optoelectronic information will also be staggered, and the correct information cannot be obtained. Therefore, in this embodiment, in order to prevent A precharge unit 81 having a charged particle irradiation unit 811 is provided to prevent such a variation in the amount of charge. Before the light or laser is irradiated to a predetermined portion of the wafer to be inspected, the charged particles are irradiated from the charged particle irradiation portion 811 of the precharge unit to eliminate the parallax of the charged portion. When the surface of the wafer is charged, an image of the wafer surface to be detected is formed in advance, and the image is evaluated by the image evaluation, and the precharge unit 81 is operated based on the detection.

Fig. 10 is a view showing the main part of an embodiment of the precharge unit of the present invention.

The charged particles 818 are emitted from the charged particle irradiation line source 819 in the direction of the wafer W, and are accelerated by the voltage set by the bias power source 820, and then irradiated. Both the inspected area 815 and the area 816 show the portion irradiated with the charged particles before the implementation, and the area 817 shows the portion where the charged particles are being irradiated. In the figure, the sample substrate W is scanned in the direction of the arrow in the drawing, but when reciprocating scanning is performed, as shown by the dotted line in the figure, a charged particle beam is further provided on the opposite side of the primary electron beam source. The source 819 may alternately turn ON and OFF the charged particle beam sources 819 and 819 in synchronization with the scanning direction of the sample substrate W. At this time, if the energy of the charged particles is too high, the secondary electron yield from the insulating portion of the sample substrate W exceeds 1, and the surface is positively charged. Furthermore, if the energy is generated twice, the electrons are generated twice, the phenomenon becomes complicated, and the irradiation effect is reduced. Therefore, the input voltage (Landing voltage) is set to be less than 100 eV in which the generation of the second electrons is reduced (in terms of It is more effective than 0eV and below 30eV.

Figure 11 is a view showing a second embodiment of the precharge unit of the embodiment. state. This figure discloses an illumination line source in the form of an illuminated electron beam 825 as a charged particle beam. The illumination line source is composed of a hot wire 821, an extraction electrode 824, a shielding box 826, a filament power source 827, and an electron extraction power source 823. The extraction electrode 824 is provided with a slit having a thickness of 0.1 mm, a width of 0.2 mm, and a length of 1.0 mm, and its positional relationship with the hot wire 821 having a diameter of 0.1 mm forms a three-electrode electron gun. The shield case 826 is provided with a slit having a width of 1 mm and a length of 2 mm, which is spaced apart from the extraction electrode 824 by a distance of 1 mm, and is formed in such a manner that the centers of the two slits are formed in a uniform manner. The hot wire material is tungsten (W), and the current of 2A, the extraction voltage of 20V, and the bias voltage of -30V can obtain an electron current of several μA.

The example shown here is only an example. For example, the hot wire material may also use a high melting point metal such as Ta, Ir, or Re, or a tungsten oxide (Thoria-coated), an oxide cathode, etc., depending on the material and the line. The diameter and length of the hot wire current will change when there is no rumor. As for other types of electron guns, any electron beam irradiation area, electron current, and energy can be set to an appropriate value.

Fig. 12 is a view showing a third embodiment in which an irradiation ion 829 is shown as an irradiation line source of the present charged particle beam type. The illumination source is composed of a hot wire 821, a hot wire power supply 822, a discharge power source 827, and an anode shielding box 826. The anode 828 and the shielding box 826 are provided with slits of the same size of 1 mm × 2 mm, and the centers of the two slits are formed in a uniform manner at intervals of 1 mm. In the shield case 826, an Ar gas 830 of about 1 Pa is introduced through the pipe 831, and the arc wire 821 is operated by the hot wire 821. The bias voltage is set at a positive value.

Fig. 13 is a view showing a state of the plasma irradiation method according to the fourth embodiment. The structure is the same as that of Fig. 20, and the operation is also performed in the same manner as described above. The hot wire 821 is operated by arc discharge, and the bias potential is set at 0 V, and the plasma 832 is ejected from the slit by the air pressure, and is irradiated to the sample. Substrate. In the case of plasma irradiation, the surface potential of either the positive or negative surface of the sample substrate can be close to zero due to the group of particles having positive and negative electric charges compared to other methods.

The charged particle irradiation portion 819 disposed close to the wafer W has a structure as shown in FIGS. 10 to 13 , which is different by the difference in surface structure of the oxide film or the nitride film of the wafer W. After the step, the wafer W is irradiated with appropriate conditions to irradiate the charged particles 818. After irradiating the wafer W under optimal irradiation conditions, that is, smoothing the surface potential of the wafer W, or saturating it with charged particles, using irradiation light or laser, or by electron beam 711 and The sub-charged particles 712 form an image to detect defects.

As described above, in the present embodiment, since the measurement is performed by the irradiation of the charged particles, the measurement image is not distorted by charging, or the degree of distortion is small, so that the presence or absence of the defect can be accurately measured.

In addition, the amount of light, laser, or primary electron beam that causes problems in the past can be used for a large number of illumination and scanning platforms. The secondary charged particles such as photoelectrons, secondary electron emission, and mirror reflection electrons can also be large. Detected and able to obtain a good detection signal with a good S/N ratio, the reliability of defect detection is improved.

Moreover, since the S/N ratio is large, good image information can be produced even if the platform is scanned more quickly, and the throughput of the inspection is expanded.

- potential application mechanism -

In Fig. 14, the potential application unit 83 transmits the setting of the stage on which the wafer is placed, based on the fact that the generation rate of the secondary charged particles discharged from the wafer or the transmittance of the secondary system depends on the wafer potential. A potential of ± V is applied to the stage to control the generator of the secondary charged particles. When the electron beam is irradiated to the primary system, the potential application mechanism decelerates the energy originally generated by the irradiation electrons, and the application of the irradiation electron energy of about 0 to 500 eV is set for the wafer. Further, it is possible to form a reference voltage as a sample wafer and determine the energy of electrons moving in the secondary system.

As shown in FIG. 14, the potential application unit 83 includes a voltage application device 831 that is electrically connected to the mounting surface 541 of the stage device 50, and a charging investigation and voltage determination system (hereinafter referred to as a survey and determination system) 832. The investigation and determination system 832 includes a monitor 833 electrically connected to the image forming unit 763 of the detection system 76 of the electro-optical device 70, a manipulator 834 connected to the monitor 833, and a CPU 835 connected to the manipulator 834. The CPU 835 supplies a signal to the aforementioned voltage applying means 831.

The potential application mechanism is designed to find a potential at which the wafer to be inspected is difficult to charge, and apply the potential to the wafer.

In the method of inspecting an electrical defect as a sample, in the case where the originally electrically insulating portion and the portion are in an energized state, a phenomenon in which the voltage of the portion is different may be used.

This method is, first, to apply a charge to the sample beforehand, so that this A voltage difference is generated between the voltage of the portion that has been electrically insulated and the portion of the portion that is electrically insulated, but is partially energized for some reason, and then the voltage is obtained by irradiating the beam of the present invention. Poor data and parse the obtained data to detect its power-on status.

- Illumination beam calibration mechanism -

In Fig. 15, the irradiation beam calibration unit 85 includes a plurality of Faraday cups 851 and 852 and a standard sample 853 for measuring beam current. These elements are a plurality of portions provided on the side of the wafer mounting surface 541 on the rotating platform 54. The standard sample 853 is a person who uses a lattice pattern including a plain plane portion having no pattern and a reference pitch forming a dot pattern. The plain plane is made of a conductive material. When the light or the laser is irradiated to the same portion, in order to measure the irradiation area of the beam, the photoelectron image generated by the portion irradiated at a low magnification is photographed, and the contour is changed from the degree of darkness (Gradation) Profile) to find out. In the case of electrical conductivity, the surface potential is stabilized, so that a stable measurement can be obtained. Also. If there is a pattern of dots, the pitch can be known, for example, a pitch of 2 μm. Therefore, the change profile of the specification and the intensity of the irradiated portion can be measured from the relationship between the change in the photoelectron image of the irradiated portion and the size of the irradiated portion. Moreover, it can also be used in the Faraday cup. When a Faraday cup is used, light or laser is irradiated to a portion having a hole to obtain a photoelectron image of the obtained irradiation region. Moreover, since the specifications of the holes of the Faraday cup are known, the specifications and coordinates of the irradiation area can be measured by comparing the specifications with the specifications of the irradiation area. Further, when the beam is irradiated with an electron beam, the original Faraday cup can be used. When an electron beam is injected into the Faraday cup, the current of the electron beam can be measured. Faraday Cup 851 is a thin beam (about φ 2 μm), and a Faraday cup 852 is a large beam (about φ 30 μm). There are two types. The beamlet Faraday cup 851 is stepped to push the rotating platform 54 to determine the beam change profile, and the large beam Faraday cup 852 is the total amount of current measured by the beam. The Faraday cups 851 and 852 are disposed such that the upper surface thereof and the upper surface of the wafer W placed on the mounting surface 541 are at the same level. In this way, it is possible to constantly monitor the primary electron beam emitted from the electron gun. This is because the electron gun does not always emit a fixed electron beam, but changes its injection amount during use.

- Alignment control device -

The alignment control device 87 is a device that uses the platform device 50 to position the wafer W relative to the electro-optical device 70. Including the optical microscope 871, the wafer is subjected to a broad field of view (the magnification used is lower than that of the electro-optical system), the electron optical system using the electro-optical device 70 is aligned at a high magnification, the focus is adjusted, and the inspection area is set. , pattern alignment and other controls. The reason why the wafer is inspected at a low magnification by the optical system is to perform the pattern inspection of the wafer in an automatic manner, and to perform wafer alignment by observing the wafer pattern with a narrow field of view by using the photoelectron image generated by light or laser irradiation. It is necessary to easily detect the alignment mark through the photoelectron image. At this time, an electron beam can also be used as an irradiation beam instead of photoelectrons.

The optical microscope 871 is provided in the casing (which is also arranged to be movable in the casing), and the light source for operating the optical microscope is not shown, but is also provided in the casing. Further, the electron optical system that performs high-magnification observation shares the electron optical system (primary optical system 72 and secondary optical system 74) of the electro-optical device 70. If it is schematically shown, it will be the type shown in Figure 16. state. To observe the observed point on the wafer at a low magnification, the X-platform 53 of the platform device 50 is moved in the X direction to move the observed point of the wafer into the field of view of the optical microscope. The wafer is first recognized by the optical microscope 871 in a wide field of view, and then the observed position on the wafer is displayed on the monitor 873 via the CCD 872, and the observation position is roughly determined. At this time, the magnification of the optical microscope can also be changed from a low magnification to a high magnification.

Next, the platform device 50 is moved only by the distance δ x corresponding to the optical axis of the electron optical device 70 and the optical axis of the optical microscope 871, and the observed point on the wafer determined in advance by the optical microscope is moved to the electro-optical device. The position of the field of view. At this time, the axis O of the electro-optical device₃-O₃And the optical axis of the optical microscope 871₄-O₄The distance between them (in the present embodiment, the two are only displaced in the direction along the X axis, but may be displaced in the Y-axis direction and the Y-axis direction). The δ x system is known in advance, so that the value δ is moved. x, you can move the observed point to the recognized position. After the movement of the observed position to the identification position of the electro-optical device is completed, the observed point is photographed at a high magnification through the electro-optical system, and the image is memorized or displayed on the monitor 765 through the detector 761. At this time, as a primary system, light or a laser beam can be irradiated and an optoelectronic image can be utilized. Further, when an electron beam is used as a primary system, electrons or mirror-reflected electron images can be obtained twice and used for alignment.

In this manner, after the observation point of the wafer is displayed on the monitor at a high magnification by the electro-optical system, the rotation center of the wafer relative to the rotation center of the rotating platform 54 of the platform device 50 is offset by a known method. That is, the optical axis of the wafer relative to the electro-optical system₃-O₃The deviation δ θ in the direction of rotation is detected, and the predetermined pattern is relative to the electro-optical device on the X-axis and The deviation in the Y-axis direction is detected. Then, based on the detected value and the data of the inspection mark provided on the wafer or the shape of the wafer pattern, the operation of the platform device 50 is controlled to perform alignment of the wafer.

- Vacuum exhaust system -

The vacuum exhaust system is composed of a vacuum pump, a vacuum valve, a vacuum gauge, a vacuum pipe, and the like, and the electron optical system, the detector portion, the sample chamber, and the load lock chamber are vacuum-exhausted in a predetermined order. In each section, the vacuum valve is controlled to obtain the necessary degree of vacuum. The degree of vacuum is often monitored. When there is an abnormality, the interlocking function is used to perform emergency control such as an isolation valve to ensure the degree of vacuum. As a vacuum pump, the main exhaust can use a turbo molecular pump.

The rough pump can use the Roots-type dry pump. The pressure at the inspection site (electron irradiation unit) is 10^-3To 10^-5Pa, preferably 10 with a low 1-digit number^-4To 10^-6Pa is more practical.

-Control System-

The control system is mainly composed of a main controller, a control controller, and a platform controller.

The main controller has a man-machine interface, and the operator's operation is performed here (input of various instructions/commands, program steps, etc., instructions for starting the inspection, switching between automatic and manual inspection modes, and all necessary for manual inspection mode) Input of instructions, etc.). Other aspects, communication with the factory main computer, control of the vacuum exhaust system, handling of samples such as wafers, control of position alignment, command transmission to other control controllers or platform controllers, or information reception, etc. The main controller executes. In addition, it has: The optical microscope image signal, and the platform change signal is fed back to the electron optical system to correct the deteriorated platform vibration correction function of the image; the displacement of the sample observation position in the Z direction (the axis direction of the secondary optical system) is Detect and feed back to the electro-optical system to automatically correct the focus of the auto focus correction function. The transmission and reception of signals such as feedback signals to the electronic optical system and the signals from the platform are respectively performed via the control controller and the platform controller.

The control controller is mainly used as a high-precision power supply for the control of the primary optical system and the secondary electron optical system (light source or laser light source, mirror, optical system lens, electron optical system lens, aligner, Wien filter, etc.). Control, etc.). Specifically, in the irradiation region, when the magnification is changed, in order to maintain a constant irradiation beam density at all times, automatic voltage setting or the like is performed for each lens system or aligner corresponding to each magnification, and corresponding to each operation mode. The automatic voltage setting of each lens system or aligner is controlled (linked control). Further, when the magnification is changed, if the control of changing the irradiation beam density is performed, the number of electrons (electron number/Px) of each Px obtained by the detector is constant, and it is effective, and electrons having different magnifications can be obtained with a certain brightness. image.

The platform controller mainly performs control related to the movement of the platform, and can obtain a precise movement of μm level in the X direction and the Y direction (error of about ±0.05 μm). Furthermore, the platform also controls the rotation direction (θ control) within ±0.1 seconds of the error precision.

- Electrode cleaning -

When the electron beam device of the present invention operates, it is close to each other (Table When the particles in the vicinity of the surface are charged, the target substance will float and be attracted to the high-pressure region, so that organic substances are deposited on various electrodes formed or deflected by the electron beam. Insulators that slowly accumulate due to surface charging can adversely affect the formation or deflection mechanism of the electron beam. Therefore, the deposited insulator must be periodically removed. The periodic removal of the insulator is to use hydrogen or oxygen or fluorine and a compound containing the elements HF, O in a vacuum by using an electrode near the region where the insulator is deposited.₂, H₂The plasma of O, CMFN, etc. maintains the potential of the plasma in the space on the electrode surface to generate a sputtering potential (several kV, for example, 20V to 5kV), and removes only organic substances by oxidation, hydrogenation, and fluorination.

-Changes in platform devices -

Figure 17 shows a variation of the platform apparatus of the detecting apparatus of the present invention. A partition plate 914 that substantially protrudes horizontally in the +Y direction and the -Y direction (the left-right direction of FIG. 17B) is attached to the upper surface of the Y-direction movable portion 95 of the platform 93, and the X-direction movable portion 96 The upper portion always forms a narrowing portion 950 having little conductivity. Further, the upper surface of the X-direction movable portion 96 is formed such that the same partition plate 912 protrudes in the ±X direction (the left-right direction of Fig. 17A), and a narrow portion is always formed between the upper surface of the platform platform 97 and the upper surface of the platform platform 97. 951. The platform platform 97 is secured within the housing 98 to the bottom wall by known methods.

Therefore, even if the sample stage 94 is moved to any position, the narrowed portions 950 and 951 can always be formed, so that when the movable portions 95 and 96 move, even if the gas is discharged from the guiding faces 96a and 97a, the narrowed portion 950 is also formed. With 951, the movement of the released gas is hindered, so it is irradiated by a charged beam. The pressure rise in the space 924 near the sample can be suppressed to be very small.

On the side and the lower surface of the movable portion 95 of the platform and the lower surface of the movable portion 96, a differential exhaust groove as shown in Fig. 18 is formed around the hydrostatic bearing 90, and vacuum evacuation is performed through the groove. Therefore, when the narrowed portions 950 and 951 are formed, the released gas from the guide surface is mainly exhausted through the differential exhaust portions. Therefore, the pressures of the spaces 913 and 915 inside the platform form a state higher than the pressure in the chamber C. Therefore, the spaces 913 and 915 are not only exhausted by the differential exhaust grooves 917 and 918, but if the portion to be evacuated is separately provided, the pressure of the spaces 913 and 915 can be lowered, and the pressure rise of the vicinity 924 of the sample can be further reduced. . Therefore, vacuum exhaust passages 91-1 and 91-2 are provided for this purpose. The exhaust passage extends through the platform cap 97 and the casing 98 and opens to the outside of the casing 98. Further, the exhaust passage 91-2 is formed in the X-direction movable portion 96 and opens below the X-direction movable portion 96.

Further, when the partition plates 912 and 914 are provided, it is necessary to increase the chamber in order to prevent the chamber C and the partition plate from interfering with each other, but it is also possible to improve the partition plate by using a stretchable material or a structure. this point. As such an embodiment, it is conceivable that the partitioning plate is made of rubber or is formed in a telescopic tubular shape so that the end portion in the moving direction is fixed to the X-direction movable portion 96 in the case of the partitioning plate 914, and is disposed on the partitioning plate 912. In the case of the case, it is fixed to the inner wall of the casing 98.

A second modification of the platform device is shown in Fig. 19.

In this embodiment, the front end portion of the lens barrel, that is, the periphery of the charged beam irradiation portion 72 is formed with a cylindrical spacer 916, and is on the top of the sample W. A narrow portion can be formed between the two. With this configuration, even if the gas is released from the XY stage and the pressure in the chamber C rises, the inner portion 924 of the partition member is partitioned by the partition member 916, and is evacuated by the vacuum pipe 710, so that the chamber C and the partition member are separated. A pressure difference is generated between the interiors 924, so that the pressure rise of the space 924 inside the partition is suppressed to a low level. The gap between the separator 916 and the sample surface varies depending on how much the pressure in the chamber C and the periphery of the irradiation portion 72 is maintained, but it is appropriate to be about several tens of μm to several mm. Further, the inside of the partition 916 and the vacuum piping are connected by a known method.

Further, in the charged beam irradiation apparatus, a high voltage of about several kV is applied to the sample W, and if a conductive material is provided in the vicinity of the sample, discharge may occur. In this case, if the material of the separator 916 is made of ceramic or an insulating material covering the surface of a conductive material or the like, the sample W and the separator 916 are not formed. There is a situation in which a discharge occurs.

Further, the annular member 94-1 disposed around the sample W (wafer) is fixed to the plate-shaped regulating member of the sample stage 94. Even in the case where a charged beam is irradiated to the end of the sample such as a wafer, in order to form a minute gap 952 around the entire front end portion of the spacer 916, the member is set at the same height as the wafer. In this manner, even if the charged beam is irradiated to any position of the sample W, the tip end portion of the spacer 916 is always formed with a certain minute gap 952, so that the pressure of the space 924 around the front end portion of the lens barrel is kept stable.

Another variation is shown in Fig. 20.

A separator 919 having a built-in differential exhaust structure is provided around the charged beam irradiation portion 72 of the lens barrel 71. The shape of the partition 919 is cylindrical, and A circumferential groove 920 is formed inside, and the exhaust passage 921 extends upward from the circumferential groove, and the exhaust passage is connected to the vacuum pipe 923 via the internal space 922. A minute gap of several tens of μm to several mm is formed between the lower end of the separator 919 and the upper surface of the sample W.

In this configuration, as the platform moves, gas is released from the platform, causing the pressure in the chamber C to rise, so that even if the gas is intended to flow into the front end portion (ie, the charged beam irradiation portion 72), since the partition member 919 has The gap between the sample and the sample W is narrowed, the conductivity is reduced very small, the inflow of gas is hindered, and the inflow is reduced. Further, the inflowing gas is discharged from the circumferential groove 920 to the vacuum pipe 923, so that almost no gas flows into the space 924 around the charged beam irradiation portion 72, and the pressure of the charged beam irradiation portion 72 can be maintained at a desired high vacuum. status.

Fig. 21 shows still another variation.

A partition 926 is provided around the chamber C and the charged beam irradiation portion 72 to separate the charged beam irradiation portion 72 from the chamber C. The separator 926 is connected to the refrigerator 930 via a support member 929 made of a material having good thermal conductivity such as copper or aluminum, and is cooled to about -100 ° C to -200 ° C. The member 927 is for obstructing heat conduction between the partition member 926 and the lens barrel during cooling, and is made of a material having poor thermal conductivity such as ceramic or resin material. Further, the member 928 is made of a non-insulating body such as ceramics, and is formed at the lower end of the partition member 926 to prevent the sample W and the partition member 926 from being discharged.

With this configuration, the gas molecules which are to flow from the chamber C into the charged beam irradiation portion are hindered from flowing due to the partition member 926, and even if they flow in, they are trapped and trapped on the surface of the partition member 926, so that they are charged. Beam photo The pressure of the firing portion 924 can be kept low. In this way, if vacuum evacuation is performed using a frozen gas repairer or a Cryo panel, it is very effective for local exhaust.

Further, in the case of the refrigerator, it is possible to use a cooling of liquid nitrogen or a refrigerator of various colors such as a He refrigerator or a pulse tube type refrigerator.

Figure 22 shows yet another variation.

The two movable parts of the platform 93 are provided with partition plates 912 and 914 in the same manner as those shown in Fig. 17. Even if the sample stage 94 is moved to an arbitrary position, the partitions can be used to separate the space 913 in the platform and the chamber C through the narrowed portions 950, 951. Further, a spacer 916 similar to that shown in Fig. 19 is formed around the charged beam irradiation portion 72, and the space 924 in the chamber C and the charged beam irradiation portion 72 is separated by the narrow portion 952. . Therefore, when the platform moves, even if the gas adsorbed to the stage is released to the space 913, the pressure of the portion rises, the pressure rise of the chamber C is suppressed to be low, and the pressure rise of the space 924 is suppressed to be lower. Thereby, the pressure of the charged beam irradiation space 924 is maintained at a very low state. Further, the space 924 can be made lower by using the partition member 919 (see Fig. 20) having the built-in differential exhaust mechanism as shown by the partition member 916 or the partition member 926 which is cooled by the refrigerator as shown in Fig. 21. The pressure is maintained steadily.

According to these embodiments, the following effects can be achieved.

(A) The platform device can exhibit high-precision positioning performance in a vacuum, and the pressure at the charged beam irradiation position is hard to rise. That is, the charged beam treatment of the sample can be performed with high precision.

(B) The gas released from the hydrostatic bearing support is almost impossible to pass. The spacer reaches the side of the charged beam irradiation area. Thereby, the degree of vacuum of the charged beam irradiation position is made more stable.

(C) It is difficult for the released gas to reach the side of the charged beam irradiation region, and the degree of vacuum of the charged beam irradiation region is easily kept stable.

(D) The vacuum chamber is divided into three chambers, such as a charged beam irradiation chamber, a static pressure bearing chamber, and an intermediate chamber thereof, with a small conductivity (Conductance). Then, the vacuum exhaust system is configured such that the pressure of each chamber is a charged beam irradiation chamber, an intermediate chamber, and a static pressure bearing chamber in a lower order. The pressure fluctuation in the intermediate chamber is suppressed lower by the separator, and the pressure fluctuation in the charged beam irradiation chamber is further reduced by a plurality of separators, and the pressure fluctuation can be reduced to substantially no problem.

(E) The pressure rise when the platform moves can be suppressed to a low level.

(F) The pressure rise when the platform moves can be suppressed lower.

(G) It is possible to realize an inspection device with high-precision positioning performance and a vacuum degree in a charged beam irradiation area, so that an inspection device with high inspection performance and no sample contamination can be provided.

(H) It is possible to realize an exposure apparatus with high-precision positioning performance and a stable vacuum degree in a charged beam irradiation area, so that an exposure apparatus with high exposure precision and no sample contamination can be provided.

(I) It is possible to manufacture a semiconductor by using a device having a high-precision positioning performance and a vacuum degree in a charged beam irradiation region, so that a fine semiconductor circuit can be formed.

Further, the platform device of Figs. 17 to 22 can be applied to the platform device 50 of Fig. 1 without any doubt.

Further embodiments of the XY stage of the present invention will be described with reference to Figs. 23 to 25.

In the present specification, the term "vacuum" means a vacuum referred to in the technical field, and does not necessarily mean an absolute vacuum.

Figure 23 shows other aspects of the XY stage.

The front end portion of the lens barrel 71 (that is, the charged beam irradiation portion 72) that irradiates the charged beam toward the sample is attached to the casing 98 that constitutes the vacuum chamber C. The sample W placed on the movable stage in the X direction of the XY stage 93 (the left-right direction in FIG. 23) is disposed directly under the lens barrel 71. This sample W is transmitted through the high-precision XY stage 93, and the charged beam is accurately irradiated with respect to an arbitrary position on the sample surface.

The pedestal 906 of the XY stage 93 is fixed to the bottom wall of the casing 98, and the Y platform 95 moving in the Y direction (the direction perpendicular to the paper surface in Fig. 23) is mounted on the pedestal 906. The two side faces of the Y stage 95 (the left and right side faces in Fig. 23) are formed with projections that protrude in the grooves, and the grooves are formed in the pair of Y-direction guides 907a and 907b placed on the pedestal 906. On the side of the platform, the groove extends in the Y direction over substantially the entire length of the Y-direction guide. Static pressure bearings 911a, 909a, 911b, and 909b of a known structure are respectively provided on the upper, lower, and side surfaces of the protrusion protruding in the groove. By blowing high-pressure gas through these hydrostatic bearings, the Y stage 95 is supported in a non-contact manner with respect to the Y-direction guides 907a, 907b, and can smoothly reciprocate in the Y direction. Further, a linear motor 932 of a known configuration is disposed between the pedestal 906 and the Y platform 95, and driving in the Y direction is performed by the linear motor. The Y platform supplies high-pressure gas through the high-pressure gas supply flexible pipe 934. The body is supplied with high-pressure gas to the hydrostatic bearings 909a to 911a and 909b to 911b through a gas passage (not shown) formed in the Y stage. The high-pressure gas system supplied to the hydrostatic bearing is ejected to a gap of several micrometers to several tens of micrometers formed between the guide faces opposed to the Y-direction guides, and the Y-platform is oriented in the X direction and the Z direction with respect to the guide faces ( In Fig. 23, the effect of correct positioning in the up and down direction).

On the Y platform, the X stage 96 is placed so as to be movable in the X direction (the left and right direction in Fig. 23). On the Y stage 95, a pair of X-direction guides 908a and 908b (illustration 908a only) having the same structure as the Y-direction guides 907a and 907b for the Y stage are provided so as to sandwich the X stage 96. A groove is also formed in the side of the X-direction guide facing the X platform. The side portion of the X platform (the side facing the X-direction guide) is formed with a projection that protrudes into the groove. The groove extends over substantially the entire length of the X-direction guide. The upper surface, the lower surface, and the side surface of the protrusion of the X-direction stage 96 projecting in the groove are provided with the same static pressure bearing as the above-described hydrostatic bearings 911a, 909a, 910a, 911b, 909b, and 910b (not shown). Show). A linear motor 933 of a known configuration is disposed between the Y platform 95 and the X platform 96, and the X-direction drive of the X platform is performed by the linear motor. Then, the X stage 96 supplies high-pressure gas through the flexible pipe 931, and supplies high-pressure gas to the hydrostatic bearing. The high-pressure gas is ejected from the hydrostatic bearing to the guiding surface of the X-direction guide, and the X-platform 96 is supported in a non-contact manner with high precision with respect to the Y-direction guide. The vacuum chamber C is exhausted through vacuum pipes 919, 920a, and 920b connected to a vacuum pump or the like of a known structure. The inlet side (vacuum chamber side) of the pipes 920a, 920b passes through the pedestal 906, and on the upper side, from the XY stage An opening is provided near the position where the high-pressure gas is discharged, and the pressure in the vacuum chamber is prevented from rising due to the high-pressure gas ejected from the hydrostatic bearing.

A front end portion of the lens barrel 71 (that is, a periphery of the charged beam irradiation portion 72) is provided with a differential exhaust mechanism 925. Even if the pressure in the vacuum chamber C is high, the pressure of the charged beam irradiation space 930 can be sufficiently lowered. That is, the annular member 926 of the differential exhaust mechanism 925 mounted around the charged beam irradiation portion 72 has a small gap (a few micrometers) formed between the lower surface (the surface on the sample W side) and the sample. The way of tens of micrometers 940 is positioned relative to the casing 98, and an annular groove 927 is formed below the casing 98. The annular groove 927 is connected to a vacuum pump or the like (not shown) through the exhaust pipe 928. Therefore, the minute gap 940 is exhausted through the annular groove 927 and the exhaust port 928, and even if the gas molecules are to enter the space 930 surrounded by the annular member 926 from the vacuum chamber C, they are discharged. With this configuration, the pressure in the charged beam irradiation space 930 can be kept low, and the charged beam can be irradiated without any problem.

The annular groove can form a double structure or a triple structure depending on the pressure in the room and the pressure in the charged beam irradiation space 930.

Dry nitrogen is generally used for the high pressure gas supplied to the hydrostatic bearing. However, if possible, a higher purity inert gas is preferred. This is because when the gas contains impurities such as moisture or oil, these impurity molecules adhere to the surface of the inner surface of the casing or the platform constituent member constituting the vacuum chamber, so that the degree of vacuum is deteriorated and adheres to the surface of the sample to cause a charged beam. The degree of vacuum of the irradiation space is deteriorated. Moreover, in terms of the use of the factory, the cost is also an important factor in the operation, so it is useful to clean dry air. At this time, in order to remove impurities, However, various chemical filters are used, and in order to reduce particles, there is a case where an ultra-high-precision filter is used. For example, a 1 μm filter and a 3 nm filter are used in series to introduce clean dry air or the like.

Further, in the above description, the sample W is usually not placed directly on the X stage, but the sample is placed on a detachable manner, or may be subjected to some micro position change with respect to the XY stage 93. On the sample stage. The presence or absence of the sample stage and its structure are not related to the gist of the present embodiment, and therefore will be omitted for simplification of the description.

In the charged beam device described above, the hydrostatic bearing type platform mechanism used in the atmosphere can be used almost directly, so that the high-precision XY stage equivalent to the high-precision platform used for the exposure apparatus can be substantially equivalent in cost. And size, realized on the XY platform for charged beam devices.

Furthermore, the configuration or arrangement of the static pressure guide described above and the actuator (linear motor) are at best only one embodiment, as long as it is a static pressure guide or actuator that can be used in the atmosphere. Be applicable. For example, the y direction is a linear motor and the x direction is a combination of ultrasonic motors, the y direction is a linear motor and the x direction is a combination of air driven positioning platforms, the y direction is a linear motor and the x direction is a combination of a ball screw type pulse motor drive. Can be applied.

Next, the numerical value of the annular element 926 of the differential exhaust mechanism and the annular groove formed in the element is shown in Fig. 24. Further, in this example, the annular groove has a dual structure of 927a and 927b. These components are spaced apart in a radial direction.

The flow rate of the high-pressure gas supplied to the hydrostatic bearing is usually about 20 L/min (converted to atmospheric pressure). Assume that the vacuum chamber C is transmitted through the inner diameter When a vacuum pipe of 50 mm and a length of 2 m is exhausted by a dry pump having an exhaust speed of 20,000 L/min, the pressure in the vacuum chamber is about 160 Pa (about 1.2 Torr). At this time, if the size of the annular element 926 and the annular groove of the differential exhaust mechanism is set to the mode shown in Fig. 24, the pressure in the charged beam irradiation space 930 can be set at 10.^-4Pa(10^-6Torr).

Figure 25 shows another embodiment of the XY stage. The vacuum chamber C composed of the casing 98 is connected to the dry vacuum pump 953 through vacuum pipes 974 and 975. Further, the annular groove 927 of the differential exhaust mechanism 925 is connected to a turbo molecular pump 951 belonging to the ultra-high vacuum pump through a vacuum pipe 970 connected to the exhaust port 928. Further, the inside of the lens barrel 71 is connected to the turbo molecular pump 952 through a vacuum pipe 971 connected to the exhaust port 710. These turbomolecular pumps 951 and 952 are connected to the dry vacuum pump 953 through vacuum piping 972, 973 (in the figure, the pumping of the turbo molecular pump and the vacuum chamber of the vacuum chamber are 1 The dry vacuum pump can be used in combination. According to the flow rate of the high pressure gas supplied to the hydrostatic bearing of the XY platform, the volume or internal surface area of the vacuum chamber, the inner diameter or the length of the vacuum piping, these exhaust gases can also be considered as dry type of other systems. The vacuum pump is used.) The hydrostatic bearing of the XY stage 93 supplies high-purity inert gas through the flexible pipes 921 and 922 (N₂Gas, Ar gas, etc.). These gas molecules ejected by the hydrostatic bearing are diffused into the vacuum chamber, and are discharged through the exhaust ports 919, 920a, and 920b by the dry vacuum pump 953. The gas molecules intruding into the differential exhaust mechanism or the charged beam irradiation space are sucked from the front end of the annular groove 927 or the lens barrel 71, and are pumped through the exhaust ports 928 and 710 by the turbo molecules 951 and 952. Discharge it. From the turbo molecular help After the discharge, the dry vacuum pump 953 is used for discharge.

Thus, the high purity inert gas system supplied to the hydrostatic bearing is concentrated on the dry vacuum pump and discharged.

On the other hand, the exhaust port of the dry vacuum pump 953 is connected to the compressor 954 through the pipe 976, and the exhaust port of the compressor 954 is connected to the flexible pipe through the pipes 977, 978, and 979 and the regulators 961 and 962. Pipes 931, 932. Therefore, the high-purity inert gas discharged from the dry vacuum pump 953 is again pressurized by the compressor 954, and is adjusted to an appropriate pressure by the regulators 961, 962, and then supplied to the hydrostatic bearing of the XY stage.

In addition, the gas supplied to the hydrostatic bearing must be as high as possible as described above, and should not contain moisture or oil as much as possible. Therefore, turbomolecular pumps, dry pumps, and compressors are required to be structures that do not incorporate moisture or oil into the gas flow path. Further, a cold trap or a filter 960 is provided in the middle of the discharge side pipe 977 of the compressor, and impurities such as moisture or oil mixed in the circulating gas are collected so as not to be supplied to the hydrostatic bearing. The practice is also valid.

Through this design, high-purity inert gas can be recycled, so that high-purity inert gas can be saved. Moreover, the room in which the device is installed does not overflow the inert gas, and the occurrence of an accident such as suffocation due to the inert gas can be eliminated.

Further, the circulation piping system is connected to a high-purity inert gas supply system 963. When the gas starts to circulate, the entire circulation system including the vacuum chamber C, the vacuum piping 970 to 975, and the pressurizing side piping 976 to 980 is provided with the effect of filling a high-purity inert gas, and for some reason, the circulation gas flow. Provides the utility of the insufficient portion when the amount is reduced.

Further, by providing the dry vacuum pump 953 with a function of compressing to a pressure higher than atmospheric pressure, the dry vacuum pump 953 and the compressor 954 can also be used in combination with one pump.

Moreover, the ultra-high vacuum pump for the exhaust of the lens barrel can also replace the turbo molecular pump with a pump such as an ion pump or an air pump. However, when using these accumulation pumps, this part cannot be used to construct a circulating piping system. In addition, it is of course possible to replace the dry vacuum pump with a dry pump such as a diaphram type dry pump or the like.

(Embodiment 2)

Reference is made to Figure 26A. Fig. 26A is a schematic configuration diagram of an optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention. The electro-optical device 70 of the present embodiment includes a light source 10000, mirrors 10002 and 10004, lens optical systems 724 (724-1 and 724-2), electrostatic lenses 10006 and 10009, a numerical aperture 10008, an electrostatic lens optical system 741, and Detection system 76. In the present embodiment, the light source 10000 is a UV laser light source. However, as in the first embodiment, UV light, DUV light, EUV light, X-ray light, and laser light of the light may be used as long as the light source 10000 is used. The irradiated substrate generates a source of photoelectrons, and other light sources can be used. In addition, the same configurations as those in the first embodiment are denoted by the same reference numerals.

Two systems of laser light are generated from the light source 10000 and incident on the mirrors 10002 and 1004, respectively, and reflected from the reflecting surfaces of the mirrors, and the traveling direction of the light is curved toward the wafer WF on the stage device 50. In reflection The laser light reflected by the reflecting surfaces of the mirrors 10002 and 10004 is irradiated onto the wafer WF on the stage device 50 by the objective optical system 724 as a primary beam. The two-dimensional image formed by the two times of emitting photoelectrons generated by the primary beam irradiated on the wafer passes through the mirrors 10002 and 10004, and passes through the numerical aperture 10008 through the electrostatic lenses 10006 and 10009, and has a narrow field of view. The reduced position is imaged, and the rear lens 741 is used to magnify the projection, and the detection system 76 performs the detection.

Fig. 27A is a schematic configuration diagram of an electro-optical device 70 of the inspection apparatus 1 of the present invention of the present embodiment. In the present embodiment, the configuration of the primary lens system is the same as that of the first embodiment, and the configuration of the first embodiment is the same as that of the first embodiment. Further, in the figures 26 and 27, the above-described example is a type in which the laser is introduced into the two systems, but one system may be introduced. For example, in FIGS. 26B and 27B, light or laser light is generated from the light source 10000, and the light 10000A of one system is reflected by the mirror 10002 of the secondary optical system, and reaches the sample surface (there is no 1000B). At this time, for example, the mirror provided in the secondary optical system is located at a position deviating from the central axis of the secondary optical system, 1 to 10 mm, etc., and the reflection position formed by the mirror is also offset from the center of the secondary optical system axis. s position. Thus, the angle of illumination of the sample is not vertical but angled. When the angle of the secondary optical system axis (z axis) is θ p , the irradiation light is irradiated onto the sample surface at the angle θ p . According to this method of use, although the one side edge of the unevenness is irradiated, the other side edge portion becomes a female back surface and is hard to be irradiated. Thereby, the contrast of the one-sided edge is improved, and the concave-convex pattern can be photographed. Furthermore, it is not limited to 2 systems, and most systems are also possible. at this time, If using most systems, the angle θ Different lasers or light can be introduced, and uniform beam irradiation can be performed. In addition, it is also possible to perform beam irradiation with a strong bias. Laser or light beam systems, such as the 3 to 20 system, can be used, but more. When the central axisymmetric illumination of the secondary optical system is performed, uniformity is better, and a highly uniform photoelectron image can be obtained. Furthermore, when a plurality of beam systems that irradiate lasers or light in an asymmetric manner are formed, a stronger beam irradiation is formed. At this time, the formed image can obtain a high contrast locally. Further, as for the mirror, as shown in Fig. 26C, a structure that can be used as a mirror in one system or a plurality of systems can be used. For example, a triangular structure has a hole formed in a central portion thereof, and a sloped surface thereof has a structure having a mirror function. At this time, it is important that the surface is coated with a conductive material or made of a conductive material. The hole in the center is for the passage of photoelectrons from the surface of the sample. Therefore, if there is an insulating material, the potential change due to the charging action may cause adverse effects such as changes in the track or deterioration of the aberration. The cross-sectional shape of the hole is more round, but square. The shape of the axisymmetric is preferred. In terms of shape, it may be a pentahedron whose surface is all square, or a bevel and a flat surface, and the side may be a curved surface. It is important that the surface roughness of the bevel having the mirror function (for example, the average surface roughness Ra) is smaller than the wavelength of the irradiated laser or light. It is preferably about 1/2 to 1/16. If the wavelength is larger than the surface roughness, the increase in scattered light and the decrease in reflectance will result in lower efficiency. Further, in the structure of the mirror, even in the two-system beam irradiation or the majority system beam irradiation of Figs. 26, 27, and 29, one such mirror can be used as the member. The mirror portion 10002 and the mirror portion 10004 may be integrated Shot mirror. Specifically, the substrate material can be used as a glass, and the mirror surface and other portions are aluminum-plated articles. At this time, the surface roughness of the mirror surface, as described above, is smaller than the beam wavelength. Further, any of Au, Ru, Os, carbon, Pt, Ti, or Cr or a plurality of materials may be coated on the mirror-plated aluminum layer in an ultra-thin film state.

(Embodiment 3)

In the optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention in the second embodiment, the optical system used for the illumination of the substrate from the light source 10000 is different from that of the optical system used for the substrate. The other components are the same as those in the second embodiment, and the description thereof will be omitted.

Refer to Figure 28. Fig. 28 is a schematic configuration diagram of an optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention. The electro-optical device 70 of the present embodiment includes a light source 10000, fiber plate 11000A and 11000B, a hole portion 11002, an objective optical system 724 (724-1 and 724-2), electrostatic lenses 10006 and 10009, a numerical aperture 10008, and an electrostatic lens. Optical system 741 and detection system 76. In the present embodiment, the light source 10000 is a UV laser light source. However, in the same manner as in the first embodiment, UV light, DUV light, EUV light, X-ray or the like may be used, and any substrate that is irradiated with light from the light source 10000 may be generated. For photonics, other light sources can also be used. In addition, the same structures as those of the first embodiment are denoted by the same reference numerals.

Laser light is generated from the light source 10000 and injected into the fiber sheets 11000A and 11000B, respectively. The light rays incident on the fiber sheets 11000A and 11000B are bent toward the traveling direction of the wafer WF on the stage device 50, and passed through the objective optical system. 724 is used as a primary beam to illuminate wafer WF on platform device 50. A two-dimensional image of photoelectrons generated by a primary beam irradiated on the wafer is passed through the hole portion 11002, and is passed through the numerical aperture 10008 by the electrostatic lenses 10006 and 10009 to image at a narrowing position of the field of view, and later. The lens 741 is enlarged and projected and detected by the detection system 76.

(Embodiment 4)

In the optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention according to the second embodiment, the two-dimensional secondary electron image generated by the primary beam irradiated on the wafer is guided to the detector. The optical system is different. The other components are the same as those in the second embodiment, and thus the description thereof will be omitted.

Refer to Figure 29A. Fig. 29A is a schematic configuration diagram of an optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention. The electro-optical device 70 of the present embodiment includes a light source 10000, mirrors 10002 and 10004, an objective optical system 724, a correction lens 12000, a numerical aperture 10008, a field lens 1202, a magnifying lens 12004, a magnifying lens 12006, and an electrostatic lens optical system 741. And detection system 76. In the present embodiment, the light source 10000 is a DUV laser light source. However, in the same manner as in the first embodiment, UV light, EUV light, X-ray or the like may be used, and if a light source is generated from a substrate irradiated with light from the light source 10000, Other light sources can also be used. In addition, the same configurations as those in the first embodiment are denoted by the same reference numerals.

Two systems of laser light (1000A and 10000B) are generated from the light source 10000 and injected into the mirrors 10002 and 1004, respectively, on the reflecting surfaces of the mirrors Reflection is made and the direction of travel is bent toward the wafer WF on the platform device 50. The laser light reflected by the reflecting surfaces of the mirrors 10002 and 10004 is irradiated onto the wafer WF on the stage device 50 as a primary beam by the lens optical system 724. The two-dimensional photoelectron image of the emitted photoelectrons generated by the primary beam irradiated on the wafer passes between the mirrors 10002 and 10004, and is contracted by the correction lens 12000 at the position of the numerical aperture (NA) 10008. , forming a polycondensation. The field lens 1202, the magnifying lens 12004, the magnifying lens 12006, and the electrostatic lens optical system 741 in the subsequent stage are enlarged and projected, and are detected by the detecting system 76. The correction of the off-axis aberration is performed by the field lens 12002. The continuous magnification setting can be performed through the magnifying lens, and the electrostatic lens 741 is used to image the image at the detector and enlarge the projection. Further, in the present embodiment, as shown in Fig. 29B, the laser can be introduced in one system. The efficacy in this case is the same as that described in Fig. 26B and Fig. 27B.

(Embodiment 5)

The optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention according to the fourth embodiment differs from the optical system for irradiating light from the light source 10000 to the substrate. The other components are the same as those of the fourth embodiment, and thus the description thereof is omitted here.

Refer to Figure 30. Fig. 30 is a schematic configuration diagram of an optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention. The electro-optical device 70 of the present embodiment includes a light source 10000, fiber plates 11000A and 11000B, a hole portion 11002, a lens optical system 724 (cathode lenses 724-1 and 724-2 (not shown)), a correction lens 12000, and numerical light. (NA) 10008, field lens 1202, magnifying lens 12004, magnifying lens 12006, electrostatic lens optical system 741, and detection system 76. In the present embodiment, the light source 10000 uses a DUV laser light source, but a light source, a laser light source, or the like may be used in the same manner as in the first embodiment, as long as it is from the light source 10000. If the substrate to which the light is irradiated generates photoelectrons, other light sources may be used. In addition, the same configurations as those in the first embodiment are denoted by the same reference numerals.

Two systems of laser light are generated from the light source 10000 and injected into the fiber sheets 11000A and 11000B, respectively. The light incident on the fiber sheets 11000A and 11000B is irradiated onto the wafer WF on the stage device 50 as a primary beam by the lens optical system 724. The two-dimensional image of the emitted photoelectrons generated by the primary beam irradiated on the wafer is imaged by the correction lens 12000 at the narrowing position of the field of view, and the field lens 1202, the magnifying lens 12004, and the magnifying lens in the subsequent stage. In 12006, the electrostatic lens optical system 741 enlarges the projection and performs detection at the detection system 76.

(Embodiment 6)

The optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention according to the second embodiment differs from the optical system in which a large amount of light of different wavelengths is applied to the substrate. The other components are the same as those in the second embodiment, and thus the description thereof will be omitted.

Refer to Figure 31. Fig. 31 is a schematic configuration diagram of an optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention. The electro-optical device 70 of the present embodiment includes a light source 10000A (wavelength λ 1) and 10000B (wavelength λ 2 ), mirrors 10002 and 10004, and a lens optical system. System 724 (cathode lenses 724-1 and 724-2), electrostatic lenses 10006 and 10009, numerical aperture 10008, electrostatic lens optical system 741, and detection system 76. In the present embodiment, the light sources 10000A and 10000B are DUV laser light sources. However, as in the first embodiment, UV light, DUV light, EUV light, X-ray, or the like, a light source or a laser light source may be used. If the substrate from which the light from the light sources 10000A and 10000B is irradiated generates photoelectrons, other light sources may be used. Incidentally, the same configurations as those in the first embodiment are denoted by the same reference numerals.

The light source 10000A and the light source 10000B generate laser light A (wavelength λ 1) and B (wavelength λ 2) of different wavelengths, and respectively enter the mirrors 10002 and 10004, and reflect on the reflecting surfaces of the mirrors, and The wafer WF on the platform device 50 bends the direction of travel. The laser light reflected by the reflecting surfaces of the mirrors 10002 and 10004 is irradiated onto the wafer WF on the stage device 50 by the objective optical system 724 as a primary beam. The laser light A (wavelength λ 1) and B (wavelength λ 2 ) of the different wavelengths may be simultaneously irradiated on the wafer W, or may be alternately irradiated. The two-dimensional secondary electron image generated by the primary beam irradiated on the wafer is passed between the mirrors 10002 and 10004, and is imaged by the numerical aperture 10008 through the numerical aperture 10008 by the electrostatic lenses 10006 and 10009. The rear lens 741 is then used to magnify the projection and the detection system 76 performs the detection.

Further, in the present embodiment, two light sources of different wavelengths are used, but two or more light sources of different wavelengths may be used.

(Embodiment 7)

The electronic device of the inspection apparatus 1 of the present invention according to the embodiment and the first embodiment The optical system used in the optical device 70 differs in the incident direction of the light from the light source with respect to the wafer substrate, and the other components are the same as those in the second embodiment, and thus the description thereof will be omitted.

Refer to Figure 32A. Fig. 32A is a schematic configuration diagram of an optical system used in the electro-optical device 70 of the inspection apparatus 1 of the present invention. The electro-optical device 70 of the present embodiment includes a light source 10000, a lens optical system 724 (cathode lenses 724-1 and 724-2), electrostatic lenses 10006 and 10009, a numerical aperture 10008, an electrostatic lens optical system 741, and a detection system 76. In the present embodiment, the light source 10000 is a DUV laser light source. However, as in the first embodiment, a light source, a laser light source, or the like may be used, such as a light source or a laser light source, as long as it is from the light source 10000. If the substrate to which the light is irradiated generates photoelectrons, other light sources may be used. In addition, the same configurations as those in the first embodiment are denoted by the same reference numerals.

In Fig. 32A(a), the laser light 10000 is generated from the light source 10000, and the back surface of the sample W is irradiated onto the stage device 50 as a primary beam. When the laser 10000A irradiated as a single beam is irradiated on the sample W, a two-dimensional secondary electron image according to the pattern on the wafer is generated, and the two-dimensional image of the photoelectron is passed through the cathode lens 724-1 and the cathode. The lens 724-2 is contracted by the electrostatic lens 10006, and polycondensation is formed in the vicinity of the position of the numerical aperture (NA) 10008. Further, the electrostatic lens 10006 and the electrostatic lens 10009 have an amplification function, and magnification control can be performed. Then, the lens 741 in the latter stage is enlarged and projected, and imaged on the detection system 76.

Figure 32A (b) shows the emission of light from the surfaces P1 and P2 of the wafer W. The state of electronics. At this time, if P1 is a material that easily generates photoelectrons, and P2 is difficult to generate an optoelectronic material, P1 generates a large amount of photoelectrons by laser irradiation from the back surface, but P2 generates only a small amount of photoelectrons. Therefore, the pattern shape composed of P1 and P2 can greatly obtain the contrast of the pattern generated by the photoelectrons.

The example shown in the above-mentioned 32A is an example in which a laser or a light is irradiated from the back surface of a sample, and an embodiment in which a laser or a light is irradiated toward the surface is disclosed in FIG. 32B, FIG. 32C, and FIG. 32D. Figure 32E. The laser or light used is a DUV laser light source. However, as in the first embodiment, UV light, DUV light, EUV light, X-ray, or the like, a light source or a laser light source may be used, as long as it is light from a light source. If the substrate to be irradiated generates photoelectrons, other light sources may be used.

Figs. 32B and 32C have the same apparatus, illumination system, and secondary optical system as those of Figs. 8, 9, 26A, 27A, and 29A. In addition, in the case of the 32B and 32C, when the sample has irregularities, the amount of photoelectrons generated from the concave portion and the convex portion is different, and thus contrast is formed, and high-resolution imaging can be performed. The difference in the material of the uneven portion of the sample and the wavelength of the primary beam irradiation may cause the amount of photoelectrons from the concave portion to be larger or opposite to the convex portion. This example of high resolution is achieved due to the difference in the amount of photoelectron emission, and a detailed example thereof will be described later.

In addition, in the 32D and 32E drawings, the image formation by the difference in the amount of photoelectrons from the surface of the sample having the unevenness is the same as that of the 32B and 32C, and the primary beam irradiation is revealed. An example of when the system is illuminated at different angles. Figure 32B and Figure 32C are approximate For the case where the sample surface is nearly perpendicular, the 32D and 32E are examples of oblique angles. In this case, although a laser, a mirror, or a lens is used for introduction, the surface of the sample may be irradiated with an optical fiber or the like. When a short-wavelength beam such as UV, DUV, EUV, or X-ray is used, if a normal quartz fiber is used, the transmission efficiency of laser or light may be poor. If a hollow fiber or hollow tube is used, the transmission efficiency will be good.

Further, as described above, the irradiation angle of the primary beam is particularly affected by the angle on the patterned sample. Therefore, in this case, it is important to optimize the angle of the light or laser to be irradiated, that is, to select the S/N of the pattern or defect and the condition of the highest contrast, and to select the irradiation angle.

(Embodiment 8)

This embodiment is an electro-optical device including a primary optical system using electron irradiation instead of a primary system using light irradiation.

In the above description, it is a form in which light or a laser is irradiated on the surface of the sample to generate photoelectrons from the surface of the sample. Here, the form of the primary system in which the electron beam is irradiated instead of the irradiated light is stated. First, an example of an inspection apparatus including a general-purpose electron gun is shown in Fig. 33. Fig. 33(a) shows the overall structure, and Fig. 33(b) is a partially enlarged schematic view of the electron gun. However, the illustration of a part of the structure is omitted.

In Fig. 33(a), the inspection apparatus has a primary column 71-1, a secondary cylinder 71-2, and a chamber 32. An electron gun 721 is provided inside the primary cylinder 71-1, and an optical system 72 is disposed on the optical axis of the electron beam (primary beam) irradiated from the electron gun 721. Moreover, a platform device is disposed inside the chamber 32. 50, sample W is placed on platform device 50. On the other hand, inside the secondary cylinder 71-2, on the optical axis of the secondary beam generated from the sample W, a cathode lens 724, a numerical aperture NA-2, a Wien filter 723, and the like are disposed. 2 lens 741-1, numerical aperture NA-3, third lens 741-2, fourth lens 741-3, and detector 761. Further, the numerical aperture NA-3 corresponds to an opening , and is a thin plate made of a metal (Mo or the like) having a circular hole. Further, the numerical aperture NA-2 is arranged such that its opening portion is at the focus position of the primary beam and the focus position of the cathode lens 724. Therefore, the cathode lens 724 and the numerical aperture NA-2 constitute a telecentric electron optical system. In particular, the cathode lens 724 is a two-stage double lens and has two telecentric electron optical systems in which the first intermediate imaging point is formed near the center of E×B. With this configuration, compared with a single telecentric or non-telecentric situation, the aberration can be reduced, and a high-resolution imaging of a two-dimensional electronic image with a wide field of view can be achieved. In short, 1/2 to 1/3 aberration can be achieved.

Furthermore, the output of the detector 761 is input to the control unit 780, and the output of the control unit 780 is input to the CPU 781. The control signal of the CPU 781 is input to the primary cylinder control unit 71a, the secondary cylinder control unit 71b, and the platform drive mechanism 56. The primary cylinder control unit 71a performs lens voltage control of the primary optical system 72, and the secondary cylinder control unit 71b performs lens voltage control applied to the cathode lens 724, the second lens 741-1 to the fourth lens 741-3. And electromagnetic field control of the Wien filter 723. Further, the platform drive mechanism 56 transmits the position information of the platform to the CPU 781. Further, the primary cylinder 71-1, the secondary cylinder 71-2, and the chamber 32 are connected to a vacuum exhaust system (not shown), and are exhausted by a turbo molecular pump of the vacuum exhaust system. Keep the interior in a vacuum.

- one shot -

The primary beam system from the electron gun 721 is incident on the Wien filter 723 while receiving the lens through the primary optical system 72. Here, in the case of the tip of the electron gun, it is possible to use a rectangular shape, a round flat shape, a curved shape (for example, r=50 μm or so), and a LaB6 which can take out a large current. Further, the primary optical system 72 is a four-pole or eight-pole electrostatic (or electromagnetic) lens having a rotational axis asymmetric type. In this way, in the same manner as the so-called cylindrical lens, convergence and divergence occur in the X-axis and the Y-axis, respectively. The lens is formed by two or three segments, and by optimizing the conditions of the respective lenses, the beam irradiation region on the sample surface can be shaped into an arbitrary rectangular shape or a circular shape without irradiation. Electronic loss. Specifically, when an electrostatic lens is used, four cylindrical rods (Rod) are used to form opposing electrodes with equal potentials and to give opposite voltage characteristics to each other. Further, as the quadrupole lens, the circular plate which is generally used may be a four-divided lens by an electrostatic deflector instead of a cylindrical shape. At this time, it is possible to reduce the size of the lens.

The primary beam passing through the primary optical system 72 is bent by the biasing action of the Wien filter 723. When the Wien filter 723 is orthogonal to the magnetic field and the electric field, and the electric field is set to E, the magnetic boundary is B, and the speed of the charged particles is v, only the charged particles satisfying the Wien condition of E=vB advance linearly. The track of charged particles is bent. For a single beam, the force FB generated by the magnetic field and the force FE generated by the electric field are generated, and the beam trajectory is curved. On the other hand, for the secondary beam, because of the force FB and The forces FE act in opposite directions, so they cancel each other out, so the secondary beam is kept straight. The lens voltage of the primary optical system 72 is set in advance so that the primary beam is imaged at the opening of the numerical aperture NA-2. The numerical aperture NA-2 is used to prevent unwanted electron beams scattered in the device from reaching the sample surface and to prevent the charged or contaminated sample W from being charged. Further, the Field Aperture NA-2 and the cathode lens 724 (two-stage double lens not shown) constitute two telecentric electron optical systems, so that the primary beam passing through the cathode lens 724 is parallel shot. The bundle can be uniformly and uniformly irradiated on the sample W. That is, Koehler illumination called optical microscopy can be realized.

-Secondary beam -

When the primary beam is irradiated onto the sample, secondary electrons, reflected electrons, or backscattered electrons are generated from the beam irradiation surface of the sample as a secondary beam. According to the irradiation energy, the mirror reflection electrons are formed. The secondary beam passes through the lens while receiving the lens of the cathode lens 724. However, the cathode lens 724 is constructed of 3 or 4 electrodes. The lowermost electrode is designed to form a positive electric field between the potential on the side of the sample W, and the electrons (especially the two-shot electrons and the mirror-reflecting electrons with less directivity) are pulled in, and the lens is introduced into the lens with good efficiency. Inside. Next, the lens action is performed by applying a voltage to the first and second electrodes of the cathode lens 724 and setting the third electrode to a zero potential. Further, the voltage is applied to the first, second, and third electrodes, and the fourth electrode is set to a zero potential. The third electrode in the case of four electrodes is used for focal length adjustment. Further, the numerical aperture NA-2 is disposed at the focus position of the cathode lens 724, that is, from the back focus position of the sample W. Therefore, beyond view The beam of the electron beam outside the field center (off-axis) also forms a parallel beam, and the center position of the numerical aperture NA-2 does not cause the corner concealment. Further, the numerical aperture NA-3 functions to suppress the lens aberration of the cathode lens 724 and the second lens 741-1 to the fourth lens 743-1 with respect to the secondary beam. The secondary beam passing through the numerical aperture NA-2, i.e., not biased by the Wien filter 723, can maintain a straight forward passage. Further, by changing the electromagnetic field applied to the Wien filter 723, only electrons having a specific energy (for example, secondary electrons, or reflected electrons, or backscattered electrons) can be guided from the secondary beam to the detector 761. The cathode lens 724 is an important lens that determines the aberration of the emitted electrons twice from the surface of the sample. Therefore, you cannot expect too large a magnification. Thus, in order to reduce the aberration, the two-stage dual lens structure type cathode lens is formed into two telecentric structures. Further, in order to reduce the aberration (non-point aberration, etc.) generated by the Wien filter formed by E × B, the intermediate imaging is set near the intermediate position of the E × B. Through this method, the effect of suppressing the increase in aberration is very large. Further, the beam is converged by the second lens 741-1, and polycondensation is formed in the vicinity of the numerical aperture NA-3. Further, the second lens 741-1 and the third lens 741-2 have a magnifying lens function, and magnification control can be performed. In the latter stage, there is a fourth lens 741-3 which is magnified and imaged on the detector surface. The fourth lens is a 5-segment lens structure, and the 1, 3, and 5 segments are GND, and a positive high voltage is applied to the 2 and 4 segments to form a lens. At this time, the second stage has a field lens function, and the second intermediate imaging is performed in the vicinity. At this time, the correction of the off-axis aberration can be performed by the field lens function. Then, the magnified imaging is performed through the lens function of the fourth stage. In this way, the imaging is collectively performed 3 times here. In addition, the cathode lens and the second through The mirror 741-1 is combined and imaged on the detection surface (total 2 times). Further, the second lens 744-1 to the fourth lens 743-1 may be a rotational axisymmetric lens called a single potential lens or a single lens. Each lens may be a configuration of three electrodes. In general, the outer 2 electrodes are set to a zero potential, and the voltage applied to the center electrode is performed to control the lens action. Further, a field stop FA-2 (not shown) may be disposed at an intermediate imaging point. The field stop FA-2 is provided in the vicinity of the second stage when the fourth lens 741-3 is a 5-segment lens, and is disposed in the vicinity of the first stage when it is a 3-stage lens. This field stop FA-2 is limited to the required range in the same manner as the field aperture of the optical microscope. In the case of an electron beam, the excess beam is shielded to prevent charging or contamination of the detector 761. The secondary beam is magnified by a secondary optical system and imaged on the detection surface of the detector 761. The detector 761 is composed of: an MCP that magnifies electrons; a fluorescent plate that converts electrons into light; a lens and other optical elements for relaying and optical image transmission of a vacuum system and the outside; and an imaging element (CCD, etc.) Composition. The secondary beam is imaged and amplified on the MCP detection surface, and the electrons are converted into optical signals by the fluorescent plate, and converted into photoelectric signals by the imaging element. The control unit 780 reads the image signal of the sample from the detector 761 and transmits it to the CPU 781. The CPU 781 performs defect inspection of the pattern from the image signal through sample comparison or the like. Further, the platform device 50 can be moved in the XY direction by the platform drive mechanism 56. The CPU 781 reads the position of the platform device 50, and outputs the drive control signal to the platform drive mechanism 56 to drive the platform device 50 to sequentially perform image detection and inspection.

Thus, in the inspection apparatus of this embodiment, the numerical aperture NA-2 Since the cathode lens 724 constitutes a telecentric electro-optical system, the beam can be uniformly irradiated onto the sample for the primary beam. That is, the Geller illumination can be easily realized. Moreover, for the secondary beam, all of the principal rays from the sample W are perpendicular (parallel to the optical axis of the lens) to the cathode lens 724, and pass through the numerical aperture NA-2, and the peripheral light is not blocked. The image brightness of the peripheral portion is not lowered. In addition, due to the energy variation of the electrons, the imaging position will be different, although the so-called chromatic aberration of magnification occurs (especially, the energy of the secondary electrons is very large, so the chromatic aberration of magnification is large), but By arranging the numerical aperture NA-2 at the focus position of the cathode lens 724, the chromatic aberration of magnification is suppressed.

Further, since the change in the magnification is performed after passing through the numerical aperture NA-2, even if the lens condition setting magnification of the third lens 741-2 and the fourth lens 741-3 is changed, the overall field of view on the detection side can be obtained. Even image. Further, in the present embodiment, a uniform image having no unevenness can be obtained. However, when the magnification is set to a high magnification, the problem of lowering the brightness of the image occurs. Therefore, in order to improve this, when the lens condition of the secondary optical system is changed to change the magnification, the lens condition of the primary optical system is set, and the effective field of view on the sample surface determined by the change in magnification is adjusted, and the irradiation is performed. The electron beams on the sample surface reach the same size.

In other words, if the magnification is increased all the time, the field of view becomes narrower, but by simultaneously increasing the electron beam irradiation energy density, the signal density of the electrons is detected even if the secondary optical system is subjected to the enlarged projection. It will always be fixed, and the brightness of the image will not decrease. Further, in the inspection apparatus of the present embodiment, it is used to bend the orbit of the primary beam to make a secondary shot. The Wien filter 723 that travels straight in a straight line is not limited to this embodiment, and an inspection apparatus may be configured by using a Wien filter that linearly travels the track of the primary beam and bends the track of the secondary beam. Further, in the present embodiment, a rectangular beam is formed by a rectangular cathode and a quadrupole lens, but it is not limited thereto, and a rectangular beam or a circular beam may be made from, for example, a circular beam, or a circular shape may be used. The beam passes through the slit to take a rectangular beam.

In this example, two numerical apertures such as a numerical aperture NA-2 and a numerical aperture NA-3 are provided. This is because it can be used separately depending on the amount of electrons to be irradiated. In the case where the amount of electrons to be irradiated to the sample is small, for example, 0.1 to 10 nA, in order to reduce the aberration of the primary beam and the secondary beam by the numerical aperture NA-2, an appropriate beam diameter is selected, for example, Φ 30 to φ 300 μm. However, when the amount of irradiated electrons increases, the numerical aperture NA-2 is charged by the adhesion of dirt, and the image quality is deteriorated instead. At this time, a larger aperture, for example, a numerical aperture NA-2 of φ 500 to φ 3000 μm is used, and the surrounding stray electrons are eliminated. In addition, it is also used to determine the aberration of the secondary beam and the transmittance according to the numerical aperture NA-3. Since the numerical aperture NA-3 does not illuminate the primary beam, the adhesion of the dirt is small, and the image is not deteriorated due to charging. Therefore, it is very efficient to select a numerical aperture according to the magnitude of the amount of illumination current.

When electron irradiation is performed on the primary beam of this type, in the semiconductor inspection apparatus 1 using the electron gun as the primary optical system 72 of the electro-optical device 70, when a large irradiation current is to be obtained, there is a problem that the electron energy width is enlarged. . The following is a detailed description using the illustration. Figure 33 (b) shows the primary optical system 72 of the electro-optical device 70 of the general-purpose electron gun 2300. schematic diagram.

In the electron gun 2300, a heating current is supplied from the heating power source 2313 for generating hot electrons to the cathode 2310, and an acceleration voltage Vacc is set to the cathode 2310 through the acceleration power source 2314. On the other hand, with respect to the cathode 2310, a voltage is applied to the anode 2311 so as to have a positive voltage therebetween and have a voltage difference of, for example, 3,000 to 5,000 volts. When the cathode 2310 is -5000 V, the anode 2311 can be 0 V. The amount of electron emission at this time is controlled by the voltage applied to the Weinel 2312. The Venus pole 2312 is superimposed on the accelerating voltage Vacc. For example, the overlap voltage: 0 to -1000V. When the voltage difference between the accelerating voltage Vacc is large, the amount of emission becomes small; when it is small, the amount of emission becomes large. Furthermore, due to the Venezue pole voltage, the initial polycondensation (first polycondensation: 1)_ST The CO) position is offset in the direction of the axis. Moreover, when the center of the cathode and the center of the anode and the center of the anode are shifted, the x and y directions perpendicular to the z-axis are also deviated, and the emitted electrons are diffused. Among them, the one used to select an effective beam and determine the shape of the beam is the field stop FA2320. The transmittance with respect to the amount of emission at this time is usually 0.1 to 0.5%. For example, the emission is 5 μA and the irradiation current is 5 to 25 nA. Therefore, when an irradiation current of, for example, 1 μA is to be obtained, an emission of 200 μA to 1 mA is required. At this time, the transmission emission becomes large, and from the cathode to the first polycondensation, from the first polycondensation to the field stop FA, the energy width of the electron is enlarged due to the Boersch effect. For example, at the FA position, it will expand from 1.2eV to 10 to 50eV.

The energy width causes problems especially at low LEs because the expansion of the electron orbital near the surface of the sample in the z direction becomes large. It is explained according to the schema. Figure 34 is a view showing the irradiation of an electron beam irradiated on the surface of the sample. A graph of current intensity (amount) and energy state, and the state of the beam that is incident on the surface of the sample. Fig. 34(a) shows the intensity of the irradiation current and the state of the energy of the beam irradiated on the surface of the sample, and Fig. 34(b) shows the state of the beam irradiated on the surface of the sample. When the irradiation current energy of the beam irradiated to the sample is optimal, the beam is set to the beam c, and when the beam irradiation current energy is low, the beam is set to the beam a, and the irradiation current of the beam is maximized. The beam is set to beam b. Further, the beam when the irradiation current energy of the beam is high is set as the beam d. The relationship between the energy of the electron beam and the intensity (amount) of the irradiation current is in the form of a hot electron formation such as LaB6, which follows the Maxwell distribution, as shown in Fig. 34(a). At this time, as described above, the electron beam having the characteristic of the corresponding energy level is set to the beam a to the beam d.

The case where the beam d of higher energy just hits the surface of the sample is taken as an example and is shown in Fig. 34(b). At this time, the beam d collides with the surface but does not reflect (the mirror does not form electrons). On the other hand, the beam c, the beam b, and the beam a are reflected at the reflection potential point, that is, the mirror-reflected electrons are formed. Further, the position of the axial direction reflected by the beam c, the beam b, and the beam a having different energies, that is, the Z position is also different. The difference in Z position produces ΔZ. The larger the ΔZ, the larger the image bluriness imaged in the secondary optical system. This is because the mirror-reflected electrons formed at the same surface position are displaced in position on the image plane. In particular, in mirror-reflected electrons, the energy shift causes a shift in the reflection point and a shift in the middle track, so the influence is large. These phenomena can be said to be the same in the image formed by the electrons reflected by the mirror or the image formed by the electrons emitted by the mirror + two times. In addition, when the electron beam energy width of the irradiation is large, the adverse effect is also large (ΔZ change) Big). Therefore, it is very effective if there is a primary beam that can be irradiated on the surface of the sample in a state where the energy width is narrowed. The technique invented for this purpose is an electron generating source and a primary optical system shown in Figs. 35 to 41 described below. Compared with the conventional technology, these technologies can not only narrow the energy width of the electron beam, but also improve the beam transmittance of the primary system. Therefore, a large current can be irradiated to the sample with a narrow energy width. surface. That is, since the above ΔZ can be small, the offset at the imaging plane of the secondary optical system becomes small, and low aberration, high resolution, large current, and high throughput can be realized. Generally, in a thermoelectric electron source (Gun) such as LaB6, the electron generating portion has an energy width of about 2 eV. Therefore, as the amount of generated current is increased, the energy width is increased due to the Wells effect due to the repulsive force of the reservoir wheel or the like. For example, when the emission current of the electron source increases from 5 μA to 50 μA, the energy width is expanded from, for example, 0.6 eV to 8.7 eV. That is, when the current value is set to 10 times, the energy width is expanded to about 15 times. Further, in the primary optical system that passes through the middle, the energy width of the space charge effect or the like is increased. In view of this feature, it is important to reduce the energy width at the electron generating source and to increase the transmittance of the primary optical system to reduce the emission current of the electron generating source in order to make the electron beam of the narrow energy width reach the sample. Although there is no means to achieve this in the above, the present invention is to achieve these characteristics. The functions and descriptions of these techniques will be described with reference to the embodiments shown in Figs. 35 to 41 to be described later.

Moreover, the intensity of the electron beam (in the case of a high amount, beam b) is not necessarily optimal for shooting. For example, in the case of an energy distribution with a quasi-Maxwell distribution, where the lower energy portion has the maximum beam intensity (amount) A lot (beam b). At this time, since there are many beams having a higher energy than the beam b, there is a case where the image formed by the beams becomes a different image quality. That is, the beam d collides with the sample, and thus the electron image is emitted twice, and relatively, the energy of the beam b is low, so that the surface unevenness of the sample is small, and the mirror-reflected electron is easily formed. That is, the influence of the unevenness or potential difference on the surface is small, and the mirror-reflected electrons are formed, and as the quality is concerned, it is easy to become an image having a low overall contrast or a shining image. Empirically, it is difficult to obtain images with higher resolution. In particular, when there is an oxide film on the uppermost surface of the surface, the influence of the amount of electrons colliding with the surface becomes large. For example, when the emission current becomes large (for example, 10 times) as compared with the case where the emission current is small, the energy width is therefore Expand more than 10 times. At this time, if the electron beam is irradiated to the surface of the sample with the same incident energy LE, the absolute amount of the beam b colliding on the surface of the sample increases at a portion where the energy is higher than the beam b, and thus the oxide film is charged. Become bigger. Due to the influence of the charging, the orbit or imaging condition of the mirror-reflecting electrons may be disordered, and there may be cases where the shooting cannot be performed normally. This phenomenon is one of the reasons why the irradiation current cannot be increased. In this case, the amount by which the beam d collides with the surface of the sample can be reduced, and the energy c can be used to suppress the change in the potential of the oxide film to a smaller beam c (the beam of the optimum energy). In this way, the amount of the beam colliding with the sample can be suppressed to obtain a stable image. However, as can be seen from Fig. 34(a), the intensity (amount) of the beam c is lower than the beam b. If the beam c of the optimum energy can be brought close to the beam b of the maximum intensity, the amount of electrons contributing to the formation of the image will increase due to the amount of proximity, thereby increasing the throughput of detection. Therefore, it is important to narrow the energy width to reduce the electrons that hit the surface of the sample. The present invention is to achieve this object. The following will be Figures 35 through 41 illustrate embodiments thereof.

In addition, in Fig. 34(b), if the LE is continuously increased, the beam d will collide with the surface of the sample, and the beam c will collide, and the colliding electron beam will increase, so that two electrons are emitted. Also increased. The region in which the mirror-reflected electrons and the two-shot electrons are mixed is referred to as a transition region. When all the first shots collide with the surface of the sample, the mirror-reflected electrons disappear, and only two electrons are emitted. Moreover, when there is no collision electron, all of them become mirror-reflected electrons.

Moreover, when the Vernett voltage is changed to change the amount of emission, the first polycondensation position also changes, so the aligner or lens located downstream must be adjusted each time.

In addition, in the inspection of semiconductors, it is necessary to perform EUV mask inspection (inspection of mask for ultraviolet lithography method) or NIL inspection (mask inspection for nano-imprint lithography) in accordance with new technology. Defect inspection. Therefore, semiconductor inspection devices have the need to reduce aberrations to improve resolution.

To reduce the aberration to improve the resolution, the aberration of the optical system must be reduced in particular. The main reason for the deterioration of the imaging system is the so-called energy aberration (also known as chromatic aberration) and<img file="TW201634917A_D0001.tif" wi="299" he="62" img-format="tif" img-content="character" orientation="portrait" inline="no" />. Therefore, in order to improve the aberration of the secondary optical system, it is required to increase the acceleration energy in a short time.

Therefore, in order to solve such a problem, the inventors of the present invention have invented a primary optical system including a novel photoelectron generating device and an electro-optical device having the primary optical system. This primary optical system uses DUV light or DUV laser as a light source. However, the light source is not limited to this, and UV, EUV or X-rays can also be used. The contents thereof will be described below based on Fig. 35.

As shown in Fig. 35, the present optical system 2000 is roughly provided with light: Source (not shown), field stop (FA) 2010, optoelectronic generator 2020, aligner 2030, E×B deflector (Wien filter) (not shown), aperture 2040, cathode lens (CL) 2050.

The field stop 2010 is provided with an optoelectronic surface 2021 of a photoelectron generating device 2020 to be described later, and a hole having a predetermined shape disposed between the photoelectron surface and the light source. Light or laser light that is radiated from the light source to the field stop 2010 passes through the aperture of the field stop 2010, and the optoelectronic surface 2021 is illuminated with a hole-like light or laser. That is, light or a laser beam irradiated from the light source thereby generates photoelectrons having the same shape and hole shape from the photoelectron surface 2021. Further, as the light source, light or laser such as DUV (deep ultraviolet ray), UV (ultraviolet light), EUV (extreme ultraviolet ray), or X-ray which generates photoelectron wavelengths is used.

The photoelectron generation device 2020 is configured by an optoelectronic surface 2021, a first segment lens 2022 belonging to a three-stage extraction lens, a second segment lens 2023, and a third segment lens 2024 to constitute one extraction lens. Moreover, a numerical aperture 2025 is provided. The extraction lens is a magnetic field lens or an electrostatic lens. However, when a magnetic field lens is used, the magnetic field corrector is provided in the vicinity of the numerical aperture 2025 described later. Further, it is also effective in the vicinity of the downstream of the field lens (not shown) of the secondary optical system or in the vicinity of the objective lens (not shown). This is because the image is sometimes bent by the influence of the magnetic field, so that it is corrected. Further, the number of segments in which the lens is taken out is not limited to this.

The photoelectron surface 2021 is a planar portion in which a photoelectronic material is coated on a substrate made of a light-transmitting element such as quartz, quartz glass, quartz glass, or magnesium fluoride glass. As the optoelectronic material, a material having a low work function such as ruthenium or gold (a material having excellent photoelectron generation efficiency) is used. this In the embodiment, an optoelectronic material such as ruthenium or gold is coated on the substrate material to a thickness of 1 to 10 nm. The shape of the photoelectron surface 2021 is, for example, a circle or a rectangle of 10 μm to 50 mm, but is not limited thereto. Light or laser light is introduced through the window of the substrate to reach the photocathode and produce photoelectrons on the photocathode.

The extraction lens (extraction electrode) composed of the first segment lens 2022, the second segment lens 2023, and the third segment lens 2024 is configured to extract photoelectrons generated from the photoelectron surface 2021 from the light source in opposite directions, and accelerate the extracted photoelectrons. The role. These extraction lenses are electrostatic lenses. Moreover, the extraction of the lenses 2022, 2023, and 2024 does not use a Vernite pole, and the extraction electric field will remain constant. Further, it is preferable that the first extraction electrode 2022, the second extraction electrode 2023, and the third extraction electrode 2024 use a one-sided telecentric or two telecentric structures. Since a very uniform extracted electric field region can be formed, the generated photoelectrons can be transported with low loss.

When the voltage applied to the extraction lens is V1, the voltages of the first extraction electrode 2022, the second extraction electrode 2023, and the third extraction electrode 2024 are V2, V3, and V4, respectively, for example, V2 and V4 are set to V1+3000 to 30000V, and V3 is set to V4+10000 to 30000V. However, it should not be limited to this.

A numerical aperture 2025 is disposed between the third extraction electrode 2024 of the photoelectron generation device 2020 and an aligner 2030 to be described later. The numerical aperture 2025 is used to perform selection of an effective beam such as a polycondensation formation position, a beam amount, and an aberration.

The aligner 2030 includes a first aligner 2031, a second aligner 2032, and a third aligner 2033, which are used for adjustment of optical axis conditions and the like. First pair The aligner 2031 and the second aligner 2032 are aligners that perform static operation, and perform tasks of tilting (tilt) and shifting (Shift) when adjusting optical axis conditions. Further, the third aligner 2033 is an aligner used when the dynamic deflector is operated at a high speed, and is used, for example, for a dynamic blanking operation.

The aperture 2040 is disposed downstream of the aligner 2030 (sample side. Hereinafter, in the positional relationship with each member, the light source side is referred to as the upstream side, and the sample side is referred to as the downstream side). The aperture 2040 is used to receive a beam at the time of blanking, to eliminate stray electrons, and to center the beam. Further, the measurement of the amount of electron beam is performed by measuring the absorption current of the aperture 2040.

Downstream of the aperture 2040 is an E x B area, which is the area that intersects the secondary optical system, where an E x B deflector (Wien filter) (not shown) is provided. The E×B deflector deflects the primary electron beam so that its optical axis is perpendicular to the sample surface.

A cathode lens 2050 is provided downstream of the E x B region. The cathode lens 2050 is a lens in which a primary optical system and a secondary optical system coexist. The cathode lens 2050 may be composed of two stages of the first cathode lens 2051 and the second cathode lens 2052, or may be composed of one sheet. When the cathode lens 2050 is composed of two stages, polycondensation is formed between the first cathode lens 2051 and the second cathode lens 2052. When the number of cathode lenses is one, polycondensation is formed between the cathode lens 2050 and the sample.

In addition, the amount of photoelectrons is determined by the intensity of light or laser light that is incident on the optoelectronic surface. Therefore, the primary optical system 2000 can also apply a method of adjusting the output of the light source or the laser light source. Moreover, although not shown in the figure It is also shown that an output adjustment mechanism such as an Attenuator or a beam splitter may be provided between the light source or the laser source and the substrate.

The formation of polycondensation of the primary optical system 2000 of the present invention will be described using the drawings. Fig. 36 is a schematic view showing the formation of polycondensation of the primary optical system 2000 of the present invention. In Fig. 36, the photoelectrons generated by the optoelectronic surface are schematically expressed as a method of vertically illuminating the sample, but actually they are deflected by the E×B deflector.

As shown in Fig. 36, the light or laser light is incident on the optoelectronic surface 2021 from the light source or laser source through the field stop 2010. Thereby, the photoelectrons generated on the photoelectron surface 2021 form a first polycondensation at the position of the numerical aperture 2025, and are vertically deflected toward the sample by the E x B deflector via the aperture 2040, and are in the first cathode lens 2051 and the second Polycondensation is formed between the cathode lenses 2052. Then, the photoelectrons forming the polycondensation are irradiated onto the sample surface in the form of a surface beam. Therefore, the electron emission shape of the photoelectron surface 2021 and the shape of the electron beam irradiated on the sample surface become conjugate. On the other hand, in the primary optical system including the general-purpose electron gun, as shown in Fig. 33(b), the photoelectrons generated from the cathode 2310 form a first polycondensation between the cathode 2310 and the anode 2311, and then pass through the anode 2311. The field stop 2320 is irradiated onto the sample surface. Therefore, the shape of the field stop 2320 and the shape of the electron beam irradiated on the surface of the sample become conjugate.

The application voltage setting of the primary optical system 2000 of the present invention is explained below. The invention of the present invention differs from the general type electron gun in the structural design by extracting and accelerating the photoelectrons generated by the light or laser irradiation on the optoelectronic surface 2021 by the extraction lens in the rear stage. No Venus pole or suppressor (Suppressor) is accelerated by a uniform electric field, so the voltage applied to each component is different from that of a general-purpose electron gun.

The following is explained based on Fig. 35. The voltages applied to the respective constituent elements are set as follows. The voltage of the photoelectron surface 2021 is V1. The voltage of the first extraction electrode 2022 is V2, the voltage of the second extraction electrode 2023 is V3, and the voltage 2024 of the third extraction electrode is V4. The voltage of the aperture 2025 is V5, and the voltage of the aperture 2040 is V6. Furthermore, the wafer surface voltage (also known as the retarding voltage) is set to RTD. In the primary optical system 2000 of the present invention, when the voltage is expressed by the voltage V1 of the optoelectronic surface 2021, a voltage is applied to each component in the following manner. That is, at low LE, V1 = RTD - 10V to RTD + 5V, V2, V4 = V1 + 3000 to 30000V, V3 = V4 + 10000 to 30000V, and V5, V6 = reference potential. In one embodiment of the primary optical system of the present invention, it is set to RTD=-5000V, V1=-5005V, V2, V4=GND, and V3=+20000V. With the voltage application of the above method, at a low LE, high throughput can be achieved with high resolution. However, this is only an example, and the applied voltage to each component is not limited to this.

Further, when the reference potential is represented by V0 and the voltage of the electron entrance surface of the detector is represented by DV, the relationship between the applied voltage and the RTD in the primary optical system 2000 of the present invention is as shown in Table 1 below. .

<tables><img id="" he="654" wi="1369" file="tw201634917a_d0002.tif" alt="" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></tables>

The primary optical system 2000 of the present invention having the above-described configuration and the electro-optical device having the primary optical system 2000 of the present invention can obtain the following effects.

First, the primary optical system 2000 of the present invention can achieve very high transmittance. The transmittance is 5 to 50%, and the transmittance of 10 to 100 times is ensured with respect to the transmittance of 0.1 to 0.5% of the optical system of the general type electron gun. This is because, first, the configuration of the planar cathode surface and the novel extraction lens can form a very uniform extracted electric field region, so that the formed photoelectrons can be transported with low loss. By increasing or decreasing the amount of photoelectrons, the distribution of the extracted electric field is kept constant. Thereby, a stable operation can be achieved with high transmittance. In a primary optical system with a general-purpose electron gun, since it is necessary to have a Venus pole or a suppression mechanism, the electric field distribution is changed by the amount of electrons generated (that is, the amount of emission), and the electric field portion is uniformly extracted, and the effective shot is made. The beam area becomes narrow, so it is difficult to increase the transmittance. In the primary optical system 2000 of the present invention, high transmittance is obtained because a Veneta pole or a suppression mechanism is not required. Secondly, in the primary optical system 2000 of the present invention, the first polycondensation position is located downstream of the lens, and the setting of the numerical aperture or the like is easy, so that it is easy to perform the optical system in which the aberration of the beam is reduced and the Wells effect is reduced. Achieved. 1st optics with a general electron gun In the system, the first polycondensation position is located near the Wei Naite pole, so it is difficult to set a numerical aperture at this position. Moreover, the position is shifted by the amount of emission, and even if a numerical aperture or the like can be set at the position, it is difficult to use it effectively. In the primary optical system 2000 of the present invention, since the position of the first polycondensation is placed downstream of the lens, this problem can be solved.

Second, the primary optical system 2000 of the present invention can achieve high throughput with high resolution. Since high transmittance can be achieved as described above, in order to obtain a high throughput, for example, an electron irradiation amount of 1 μA, it is sufficient that a very small amount of cathode discharge current is 2 to 10 μA. Thus, the Wells effect is very small and can be solved. For example, at a numerical aperture position, the energy width is 0.5 to 1.2 eV. Therefore, the electron irradiation amount can be increased with a small energy width, and the beam shift of the imaging in the secondary optical system is small, and high resolution can be maintained. According to the above results, high throughput can be achieved with high resolution.

Thirdly, the primary optical system 2000 of the present invention can maintain an optical system that is always in a stable state. The reason for this is that the positional deviation of the first polycondensation does not occur in the primary optical system 2000 of the present invention.

Next, the effects of the electro-optical device having the primary optical system 2000 of the present invention will be described in detail below.

First, since the primary optical system 2000 having the above configuration is used, the shape of the electron beam irradiated on the surface of the sample can be set to "magnification × 10 times to × 0.1 times" with respect to the electron emission shape of the surface of the surface. In particular, it is possible to use a reduction ratio of magnification × 1 or less, and the specification of the photoelectron surface does not need to be reduced, and the photoelectron density generated can be suppressed to be low. Thereby, the electro-optical device having the primary optical system 2000 of the present invention can reduce the effect of Wells Should the expansion of the energy width be suppressed.

Second, in the axial center of the electron generating portion of the optoelectronic surface, the photoelectron generating portion can be easily formed at the center position formed by the extraction lens. This effect can be achieved by illuminating the laser or laser at the center of the axis. In the 35th and 36th drawings, although the position of the light source is not shown, it is easy to achieve with a lens, a mirror, or the like regardless of the position of the light source. The primary optical system 2000 of the present invention is disposed in a lens barrel fixed to the main casing, but the light source is generated by using light or laser, so that the light source does not have to be disposed in the lens barrel, and the light source is disposed in, for example, a mirror. When the outside of the cylinder is used, it may be guided to the shaft center of the electron generating portion of the optoelectronic surface by a mirror, a lens or the like. Therefore, since the electro-optical device of the primary optical system 2000 of the present invention can be disposed on the atmospheric side, the adjustment of the center position is easy. In the inspection apparatus using the general type electron gun shown in Fig. 33(b), the center positions of the cathode 2310, the Veneta pole 2312, the anode 2311, and the field stop 2320 are shifted by assembly. Moreover, the baking performed after the atmosphere is opened also causes a bias, that is, a positional change occurs after assembly due to temperature changes during the thermal expansion and cooling processes. In order to correct these offsets, an aligner is typically placed upstream of the field stop 2320, and the aligner is used for correction. When the deviation is severe, it must be repeatedly decomposed, assembled, adjusted, and baked. On the other hand, in the electro-optical device using the primary optical system 2000 of the present invention, since only light or a laser is irradiated to the axial center position, the photoelectron generating portion can be easily formed at the center position formed by the electrostatic lens. Even if a misalignment due to assembly occurs, it can be easily adjusted. In addition, since the light source can be disposed on the atmosphere side, it is less susceptible to positional changes after assembly, and Even if the positional change after assembly occurs, it is easy to adjust. Therefore, the work process can be greatly shortened and the cost can be reduced. Further, since the field stop 2010 for determining the shape of the electrons generated on the optoelectronic surface can be disposed on the atmosphere side, it is easy to perform the replacement operation of the field stop 2010. Due to this feature, the operation process can be greatly shortened and reduced. When the field diaphragm is installed on the vacuum side, it is necessary to perform operations such as breaking the vacuum, disassembling the column, assembling, adjusting, vacuuming, baking, and adjusting the optical axis in order to replace the work. However, the present invention does not require such an operation.

Thirdly, the electro-optical device having the primary optical system 2000 of the present invention has an improved degree of freedom in beam specifications. The shape of the electron generating surface of the photoelectron surface is determined by the field stop 2010, so it is not limited to a circular or rectangular shape, and a rectangular or non-axisymmetric shape can also be used. The inspection apparatus including the primary optical system 2000 of the present invention may, for example, have a circular shape of φ 100 μm on the surface of the photoelectron, a circular shape of φ 50 μm to 100 μm on the surface of the sample, and a rectangular shape of 100 × 100 μm on the surface of the optomechanical surface. The sample surface may have a rectangular shape of 50 × 50 μm to 100 × 100 μm.

Fourth, in the electro-optical device having the optical system 2000 of the present invention, the number of components in the vacuum can be greatly reduced. In an electro-optical device having a general-purpose electron gun, an aligner is required in front of the field stop 2320 shown in Fig. 33(b) because of the offset correction of the cathode center, the Venus pole, the anode, and the field stop center. Furthermore, in order to image the beam shape formed by the diaphragm 2320 on the sample surface, a lens of 1 to 3 segments is required. Since the electro-optical device having the primary optical system 2000 of the present invention does not require these components, the number of components in the vacuum can be greatly reduced.

When the electro-optical device having the optical system of the present invention described above is applied to a semiconductor inspection device, high throughput can be achieved with high resolution, which is suitable for EUV mask inspection or NIL mask inspection. Moreover, even in the case of low LE (incident energy), high resolution can be achieved.

(Embodiment 9)

- Second embodiment of the optical system -

A second embodiment of the primary optical system of the present invention will be described. Figure 37 is a view showing a second embodiment of the primary optical system of the present invention. The primary optical system 2100 is generally provided with a light source (not shown), a field stop (FA) 2110, an optoelectronic generator 2120, an aligner 2130, and an E×B deflector (Wien filter) (not In the drawings, the aperture 2140, the cathode lens (CL) 2150, the first tube 10071, and a second tube (not shown) for accommodating the primary optical system. The second embodiment of the primary optical system of the present invention is characterized in that the reference potential is set to a high voltage. Hereinafter, the differences between the present embodiment and the above-described first optical system of the present invention will be mainly described.

In the present embodiment, a two-layer structure including a first tube 10071 and a second tube is provided, and the photoelectron generating device 2120 includes an optoelectronic surface 2121, a 1-piece extraction lens 2122, and a numerical aperture 2125.

The first tube 10071 is a tube for forming a reference voltage when the reference voltage is a high voltage, and a high voltage is applied to the first tube. The first tube 10071 is disposed inside the hole that is provided in the extraction lens 2122, the numerical aperture 2125, and the aligner 2130 to pass the primary beam, and forms an inscribed with the hole, and then is in the aperture 2140. The later section forms a large diameter and forms a large straight A cathode lens 2150 is disposed inside the portion of the diameter.

The material of the first tube 10071 is not particularly limited as long as it is not a magnetic material, but it is preferably a thin copper tube or a thin titanium tube, a plastic copper plating or a titanium plating product. Thereby, when a high voltage is applied to the first tube 10071, a magnetic field is formed inside the first tube 10071, and the primary electron beam generated by the photoelectron surface 2121 irradiated with light or laser light is highly accelerated.

On the other hand, although not shown in Fig. 37, the second tube system covers the field diaphragm (FA) 2110, the photoelectron generating device 2120, the aligner 2130, and the E x B deflector (Wien filter). (not shown), the aperture 2140, the cathode lens (CL) 2150, and the first tube 10071 are set to GND (ground). This configuration is because it belongs to the outermost structure of the cylinder device, so the portion is held at GND to prevent conductor connection with other devices, and electric shock when the human body comes into contact.

In the second embodiment of the optical system of the present invention, the lens-type electromagnetic lens is taken out. The other configurations are the same as those of the above-described embodiment, and the description thereof will be omitted.

By configuring the double-layered structural tube, the primary optical system 2100 of the present invention can apply a high voltage to the first tube 10071 of the inner tube belonging to the double-tube structure by setting the surface voltage of the sample to GND. The electron beam generated by the optoelectronic surface 2121 is highly accelerated. Thus, the primary optical system of the present invention can be referred to as a high acceleration cylinder.

In the primary optical system 2100 of the present invention (see Fig. 37), the voltages applied to the respective constituent elements are set as follows. The voltage of the photoelectron surface 2121 is set to V1, and the voltage of the first tube 10071 is V2, and the numerical aperture is The voltage of NA2025 is V5, and the voltage of aperture 2140 is V6. In addition, the wafer surface voltage (also known as the deceleration voltage) is set to RTD. Under low LE conditions, V1 = RTD-10V to RTD + 5V. V2, V5, and V6 are reference potentials. In an embodiment of the invention, it is set to RTD=0, V1=-5V, and the reference potential=40000V. Through the above voltage application, high throughput can be achieved with high resolution at low LE.

At this time, if a magnetic field lens is used, the rotation of the beam occurs due to the generated longitudinal magnetic field (the residual magnetic field in the optical axis direction). Therefore, the two-dimensional photoelectron formed by the photoelectron surface generates a shape that is rotated after passing through the generating portion and the magnetic field lens. In order to correct this situation, the rotation correcting lens is placed near the NA or downstream of the magnetic field lens to correct this effect. The correction lens at the position downstream of the magnetic field lens may be disposed at a position as close as possible to the magnetic field lens (just behind), and may be rotated and corrected.

Further, in the primary optical system 2000 (see FIG. 35) of the present invention, the electrostatic lens is represented by the voltage V1 of the optoelectronic surface 2021 as an example of the double-tube structure. The voltage is applied to the element. That is, at low LE, V1 = RTD-10V to RTD + 5V. V2, V5, and V6 are reference potentials, and V3 = reference voltage +10 to 100 kV. In one embodiment of the present invention, the system is set to: RTD=0, V1=-5V, V2=reference potential +40000V, and V3=65000V. Further, in order to make the reference voltage a reference space voltage, the tube 1 having the lens built therein is provided, and the lens, the aperture, and the aligner of Fig. 35 are built in the tube 1 to which the reference voltage is applied. Further, a tube 2 having a GND potential is provided on the outer side thereof. The tube 1 and the tube 2 are fixed by an insulating member (tube 1 and tube 2 are not shown). Through the above By applying voltage in a manner, high throughput can be achieved with high resolution at low LE.

The primary optical system 2100 of the present invention can obtain the effect of performing inspection in a state where the sample surface voltage RTD is set to 0V. Further, the primary optical system 2100 of the present invention can obtain the same effects as the primary optical system 2000 of the above-described invention. Further, the electro-optical device having the primary optical system of the present invention has the same effects, and the description thereof will be omitted.

-1 photoelectron generation device variation of optical system -

Other examples of the photoelectron generating device of the optical system of the invention of the present invention are disclosed. Figures 38 and 39 show an example in which light or laser is transmitted from a position in the middle of the system to the optoelectronic surface using a mirror provided in the cylinder.

Fig. 38 is an example of a case where the reference voltage is a high voltage, for example, 40 kV. That is, the primary optical system 2000 of the present invention is applied to the example of the second embodiment. At this time, in order to form the reference voltage, a voltage of V2 = 40 kV is applied to the tube 10071 to which a high voltage is to be applied. The inside of the tube 10071 is the same voltage space. Therefore, in this example, a mirror having a photoelectron passage hole in the center portion, for example, a triangular mirror 2170, is used, and DUV light or UV laser is introduced through a hole (not shown) provided in the tube 10071, and the triangle is transmitted through the triangle. The mirror 2170 reflects it and illuminates the optoelectronic face 2121. Next, photoelectrons are generated from the illuminated surface, and then the photoelectrons are irradiated onto the sample surface through the EX lens 2120, NA2125, and the downstream aligner. At this time, in order to form the generated photoelectrons into the track of the primary system, a voltage of a predetermined value is applied to the photoelectron surface 2121, and is determined so that LE=RTD voltage -V1.

In addition, in the same manner as the example shown in FIG. 38, in the same manner as the example shown in FIG. 38, the light or the laser beam generated by the triangular mirror 2070 is irradiated to the surface of the photoelectron, which is an example of the reference voltage GND. That is, the primary optical system 2000 of the present invention is applied to an example of an embodiment. At this time, for example, V2, V4, and V5 are GND, and the vicinity thereof serves as a reference voltage space. Then, by setting the same mirror as in Figure 38, the light-laser can be introduced. At this time, the amount of photoelectrons generated is determined by the irradiation intensity of light or laser, and therefore the control of the irradiation intensity is performed. Here, the strength control method described above is used. At this time, the mirror is such that the mirror surface and the structure are integrally formed as a conductor or a conductor is coated with a conductor. Then, the potential is made the same as the reference potential. This is to form the same potential without causing the spatial potential to be disordered. Further, in order to allow the primary beam to pass without being affected by the mirror, a hole is formed in the central portion of the optical axis of the mirror, and the primary beam passes through the hole. In order to make the inside of the hole form the same potential as the reference voltage, a conductor material or a conductor is used and connected to the reference voltage portion.

Furthermore, regarding the shape of photoelectrons, two methods are disclosed. It is illustrated by Figure 39. One is to use a FA diaphragm 2010 for defining a beam shape before entering a mirror positioned in the cylinder to form a beam having a shape of a field stop (FA) 2010, and to shoot the beam. The beam is irradiated onto the optoelectronic surface to produce photoelectrons of the shape. At this time, the projection specification of the field stop (FA) 2010 is controlled by the lens position located upstream of the field stop (FA) 2010.

Yet another method is to cover the mask material on the photoelectron surface. Figure 40 shows an example of using a photoelectron-coated pattern mask material for this purpose. An example of the primary optical system 2100 of the second embodiment of the optical system of the present invention is described. As shown in Fig. 40, the optoelectronic surface 2121 is covered with a mask material 2122. The masking material 2122 has a pattern-shaped aperture. The hole portion is not covered with the mask material. With this coating, photoelectrons are not generated from the portion, and photoelectrons are generated from the portion where the material is not provided. That is, when the DUV light is irradiated, a pattern-shaped photoelectron is generated from the unmasked patterned photoelectron surface portion. In this case, as the mask material, a material that does not generate photoelectrons may be coated in advance. Use a material with a large work function or a material with low efficiency. For example, carbon, Pt, Cr, and the like. However, if the mask material is charged, a potential unevenness is formed, and the track for emitting electrons is bent to cause an adverse effect, so that a conductive material is used.

Fig. 41 is a view showing a method of irradiating the transmitted light or the laser reflection to the optoelectronic surface again in order to further improve the efficiency. Light and laser light incident from the side of the photoelectron surface 2121 are reflected by the light having the reflecting surface structure (reflecting surface 2123) and the laser penetrating element, and are returned to the optoelectronic surface 2121 for re-irradiation. According to this method, the light or the laser light is irradiated to the photoelectron surface 2121 many times, so that the efficiency is improved. For example, the light/laser transmittance of the optoelectronic surface 2121 is set to 60%, and by irradiating 60% of the transmitted light/laser again, an increase in the amount of photoelectron generated can be obtained in accordance with the number of times of irradiation. Moreover, it is not limited to this example as long as the method of irradiating at most several times is effective. In particular, 2 to 5 shots are effective. Above this number, the light/laser intensity will decrease and the effectiveness will be greatly reduced. As described above, if the irradiation is performed for a plurality of times, the intensity of the incident light and the laser can be expected to be effective as long as it is 1/2 to 1/5 of the single irradiation. For example, the illumination light/laser intensity must be 1W When it is 0.2 to 0.5W. Especially when a large output light source is required, when the light source itself is lacking, or when the operation management cost is large. In this case, if a low-output light source can be used, the effects such as cost, efficiency, heat, and deterioration of components of the light introduction system can be reduced, which is very effective.

In addition, the example described in FIG. 40 and FIG. 41 shows an example of the primary optical system 2100 used in the second embodiment of the primary optical system of the present invention. However, the present invention is not limited to this form, and can be applied to the primary optical system 2000 of another embodiment.

(Embodiment 10)

- Semiconductor inspection device with double tube construction tube -

As described above, the electro-optical device 70 including the primary optical system 2100 of the second embodiment of the optical system of the present invention has a voltage setting applied to each component different from that of the general-purpose electron gun. That is, the reference potential V2 is set to a high voltage (for example, +40000V). Therefore, in the semiconductor inspection apparatus 1 including the electro-optical device 70 of the present invention, the first point is a double tube structure.

Use Figure 42 to illustrate. Fig. 42 is a view schematically showing the structure of a double tube of the semiconductor inspection apparatus according to the embodiment of the present invention. In Fig. 42, the first tube and the second tube are emphasized. Actually, the cross sections of the first tube and the second tube are not the same. As shown in Fig. 42, the electro-optical device 70 including the primary optical system 2000 of the present invention comprises two tubes, a first tube 10071 and a second tube 10072 provided outside the first tube 10071. In other words, it is a double tube structure. Moreover, the double-tube structure houses a light source, a primary optical system, a secondary optical system, and Detector. Further, a high voltage (for example, +40000 V) is applied to the first tube 10071, and the second tube 10072 is set to GND. The space reference potential V0 of the high voltage is secured by the first tube 10071, and the second tube is used to form the GND. Thereby, the GND connection of the device setting mode and the prevention of electric shock are realized. The tube 10071 is secured to the tube 10072 by an insulating member. The tube 10072 is GND and is mounted to the main housing 30. A primary optical system 2000, a secondary optical system, a detection system 76, and the like are disposed inside the first tube 10071.

The internal partitions of the first tube 10071 and the second tube 10072, such as screws, are made of a non-magnetic material so as not to affect the magnetic field, so that the magnetic field does not act on the electron beams. Further, although not shown in Fig. 42, a space is provided on the side surface of the second tube 10072, and a part of the primary optical system 2000 such as a light source and a photoelectron generating portion is connected to the inside. Similarly, the first tube 10071 is also provided with the same space as the space provided in the second tube 10072. The photoelectrons generated by the photoelectron generating portion are irradiated to the sample through the spaces. Further, the light source is not necessarily provided inside the second tube 10072, and may be disposed on the atmosphere side, and the irradiation light is introduced into the photoelectron generating portion housed in the second tube 10072 on the vacuum side. However, the primary optical system and the secondary optical system must be housed inside the double tube structure. The detector is provided in the first tube 10071 and is provided in an independent potential independent of the first tube and the second tube. This is to set the detection surface potential of the detector arbitrarily, and to control the electron energy incident on the detector to an appropriate value. When the tube 1 and the tube 2 are subjected to potential separation by an insulating member, an arbitrary voltage is applied to the surface of the detector detecting sensor surface to operate. At this time, if the surface potential of the sensor is set to VD, The energy incident on the surface of the sensor is determined by the VD-RTD. When the detector is used with EB-CCD or EB-TDI, in order to reduce the damage of the sensor for long-term use, the incident energy is more effective to use 1 to 7 keV.

Further, another configuration of the semiconductor inspection apparatus 1 including the electro-optical device 70 of the present invention will be described. Figure 43 is a view showing the overall configuration of a semiconductor inspection apparatus 1 according to an embodiment of the present invention. As shown in Fig. 43, a semiconductor inspection apparatus 1 according to an embodiment of the present invention has a second vacuum chamber 900. In other words, the semiconductor inspection apparatus 1 is provided with a second vacuum chamber 900, and a power supply 910 for generating a high voltage is disposed in the second vacuum chamber 900, and the lens barrel 71 and the second vacuum chamber for accommodating the first tube and the second tube are disposed. 900 is connected by a connecting pipe 920, and the wiring is disposed in the connecting pipe 920. The reason is that, as described above, the electro-optical device 70 of the present invention differs from the conventional one in that the reference potential V0 is set to a high voltage. Since the reference potential V0 is set to a high voltage, the semiconductor inspection apparatus 1 having the electro-optical device 70 of the present invention has a tube structure of a double tube. Further, a high voltage is applied to the inner first tube 10071. In the case where such a high voltage is applied, the feed-through between the atmosphere and the vacuum requires a large feedthrough in order to ensure the creep resistance at the atmosphere side due to the low withstand voltage on the atmosphere side. For example, a withstand voltage of 1 kV/mm is required, and at 40 kV, an insulating member having an insulating creeping distance of 40 mm or more and a large connector corresponding thereto are required. When such a large-sized connector has a large number, the space in which the setting portion is disposed in the lens barrel occupies a large proportion, and the specification of the lens barrel and the cost thereof become large. Therefore, in the present invention, a vacuum chamber dedicated to the power source is provided. Therefore, the feedthrough from the output is not required as long as the wiring is connected to the electrode. At this time, from The gas generated at the power source becomes a major contributor to pollution. Therefore, in order to cut off the vacuum conduction during wiring, it is effective to apply a vacuum insulation of the power supply vacuum chamber and the lens barrel to the insulating member. In addition, when a high voltage is applied, the wiring becomes thick. In the semiconductor inspection apparatus 1, when the voltage applied to the sample is increased, it is necessary to provide a large number of thick wirings around the stage. When such a large-diameter wiring is placed inside the work chamber, the wiring moves with the platform operation, so a large torque is required. For example, the friction between the wiring and the wall surface becomes large, so that particles are generated and a large problem is formed. Therefore, it is very effective to set the sample potential to GND and the reference voltage to a high voltage. At this time, it is more effective to apply the voltage on the surface of the detector to reduce the damage of the sensor. The sample potential, the reference space potential, and the sensor surface potential are set to different values. At this time, for example, it is effective to set the sample potential to GND, the reference voltage to be 10 to 50 kV, and the surface potential of the sensor to be 3 to 7 kV. In the above-described manner, the second vacuum chamber 900 is disposed to accommodate the power source 910, and is connected to the lens barrel or the like by the connection tube 920, and the wiring is disposed in the connection tube 920 to realize vacuum wiring. As for the power supply, the power supply (AC100V or DC24V, etc.) is externally imported, and the communication system uses optical communication. If this is the level of power supply, it can be solved with a small feedthrough, and it is easy to connect from the atmosphere side.

Further, as described above, since the inner tube (tube 1) has a high vacuum because of the double tube structure, the outer tube (tube 2) and the inner tube (tube 1) may be in an atmospheric pressure state. In this case, the electrostatic electrode is provided in the tube 1, because the number of wires connected to the wall of the tube 1 is large, and the vacuum/atmosphere feedthrough becomes large, which is sometimes not practical. At this point, the lens, aligner, corrector A lens with a magnetic field, an aligner, and a corrector are used. Thereby, it is not necessary to provide the feedthrough in the pipe 1, which is effective in forming a reference space of high voltage. The use of such a configuration can also be applied to the above-described first to ninth embodiments.

By providing the second lens chamber, the second vacuum chamber for power supply, and the connection tube for vacuum wiring for connecting the lens barrel and the second vacuum chamber to each other, the optical system having the above-described first optical system of the present invention can be provided. Semiconductor inspection device 1 of 2000. However, it is merely an example, and the semiconductor inspection apparatus 1 including the primary optical system 2000 of the present invention is not limited thereto. Further, in the above-described embodiment, for example, in the embodiment of the primary system and the secondary system shown in the first to ninth embodiments, the double pipe structure of the present embodiment can be used.

(Embodiment 11)

- a beam measuring method at a polycondensation position, a method of adjusting an electron beam irradiation and a NA position using the method, and a semiconductor inspection device using the same -

A semiconductor inspection method using an electro-optical device having the above-described first optical system of the present invention will be described. Further, the following method can also be applied to a semiconductor inspection apparatus using an electro-optical apparatus having a general-purpose electron gun.

In the present embodiment, a sample is observed using a projection projection type observation device (an electron beam observation device having a projection projection optical system). Such an electron beam observation apparatus includes a primary optical system and a secondary optical system. In the primary optical system 2000, an electron beam emitted from a photoelectron generating unit is irradiated onto a sample to generate an electron having information such as a sample structure. 2nd optical system A detector is provided to generate an electronic image produced by irradiation with an electron beam. In the image projection type observation apparatus, a large-diameter electron beam is used, and a wide range of images can be obtained. That is, instead of shrinking a spot beam like a normal SEM, it is irradiated with a surface beam.

When an electron beam is irradiated onto a sample, a plurality of electron systems can be used to detect a plurality of electrons. The various electrons are referred to as mirror reflection electrons, secondary electrons, reflected electrons, and backscattered electrons. In the present embodiment, the secondary electrons, the reflected electrons, and the backscattered electrons are referred to as secondary electrons. Then, the sample was observed by the characteristics of the mirror reflection electrons and the electron emission twice. Mirror-reflected electrons are electrons that do not collide with a sample and bounce back before the sample is reached. The phenomenon of specular reflection electrons is caused by the electric field on the surface of the sample.

As described above, the secondary electrons, the reflected electrons, and the backscattered electrons are referred to as secondary electron emission. In the case of these three kinds of electronic mixing, the term "release electrons twice" is also used. In the second release of electrons, "2nd electron" is representative. Therefore, the secondary electrons are sometimes explained as a representative of the secondary electron emission. Both the specular reflected electrons and the secondary electron emission can be expressed by "sending from the sample", "reflecting from the sample", and "generating by electron beam irradiation".

Fig. 44 is a graph showing the relationship between the incident energy LE and the gradation DN when the sample is irradiated with an electron beam. The incident energy LE is the energy imparted to the electron beam to be irradiated to the sample. Further, an acceleration voltage Vacc is applied to the electron gun, and a deceleration voltage Vrtd is applied to the sample. In this case, the incident energy LE is the acceleration voltage and the deceleration voltage. The difference is expressed.

Further, in Fig. 44, the degree of darkness DN of the vertical axis indicates the brightness of an image generated from electrons detected by the detector of the secondary optical system. That is, the degree of DN indicates the number of electrons detected. The more electrons are detected, the greater the DN is.

Figure 44 shows the degree of darkness of the small energy region near 0 [eV]. As shown in the figure, in the area where LE is larger than LEB (LEB&lt;LE), the degree of darkness DN shows a small certain value. In the LE below LEB and above LEA (LEALELEB), the smaller the LE, the more the darkness DN increases. In the area where LE is smaller than LEA (LE&lt;LEA), the degree of darkness DN shows a large value.

The above-mentioned faintness characteristics are related to the type of electrons detected. In the region of LEB&lt;LE, almost all electrons detected are emitted electrons twice. This area can be referred to as a secondary discharge electron zone. On the other hand, in the region of LE &lt; LEA, almost all of the electrons detected are mirror-reflected electrons. This area can be referred to as a mirror reflection area. As shown in the figure, the degree of darkness of the mirror-reflected electron region is greater than that of the secondary electron-emitting region. The reason is that the distribution range of the mirror-reflected electrons is smaller than that of electron emission twice. Since the distribution range is small, more electrons can reach the detector, and the degree of darkening becomes larger.

In addition, the LEALELEB region is a transition region from the transition of the secondary electron emission region to the mirror reflection region (or vice versa). This region is a region where the mirror-reflected electrons and the two electrons are mixed, which may also be referred to as a mixed region. In the transition region (mixed region), the smaller the LE, the generation of mirror-reflected electrons The more the amount increases, the more the degree of darkness increases.

LEA and LEB mean the lowest incident energy and the highest incident energy of the transition region. The specific values of LEA and LEB are described as follows. As a result of the inventors' research, LEA is -5 [eV] or more, and LEB is 5 [eV] or less (that is, -5 [eV] LEA LEB 5 [eV]).

The advantages of the transition zone are as follows. In the mirror-reflected electron region (LE&lt;LEA), all electrons generated by beam irradiation are mirror-reflected electrons. Therefore, it has nothing to do with the shape of the sample. The detected electrons are all mirror-reflected electrons. The difference in the degree of darkness is small, whether it is the concave or convex portion of the sample, and the S/N and contrast of the pattern and the defect are both Become smaller. Thus, it is sometimes difficult to use the mirror reflection electron region for inspection. On the other hand, in the transition region, mirror-reflected electrons are characteristically and specifically generated at the edge portion of the shape, and electrons are emitted twice in other portions. Therefore, the edge S/N and contrast are improved. Therefore, the transition area is very effective when it is inspected. The following is a detailed explanation of this point.

Figure 45 shows the phenomenon of the above transition zone. Figure 45 is a diagram showing the phenomenon of the transition region. In Fig. 45, in the mirror-reflected electron region (LE &lt; LEA), all electrons do not collide with the sample, but become mirror-reflected electrons. On the other hand, in the transition region, a part of the electrons collide with the sample, and the sample emits electrons twice. The larger the LE, the more the ratio of electrons released twice. In addition, although not shown, when LE exceeds LEB, only two discharge electrons are detected.

The invention of the present invention includes a secondary discharge electron region, a transition region, and a mirror-reflecting electron region, and includes a pattern having a concave-convex structure and a non-convex pattern. The method of making and adjusting the electron beam conditions of the secondary optical system that illuminates the electron beam and forms an image. According to the present invention, the efficiency is drastically improved, and high-precision adjustment and condition production can be achieved. It is stated below.

In the present invention, the position and shape of a beam which is located at a polycondensation position (hereinafter referred to as a CO position) in the middle of a secondary optical system is measured as a major feature. In the prior art, the beam coming to the CO position is not measured, but the NA is moved to obtain an image, and the contrast of the image is evaluated, which takes a lot of time. The order of operation of the conventional practice is as follows.

a. Forming the imaging conditions with a lens located between the CO position and the detector.

b. If there is NA, set it to a large diameter or remove it.

It is better to be able to observe the CO as a whole. For example, φ 1000 to φ 5000 μm.

c. Shoot the beam at the CO position.

In the present invention, the composition of the apparatus will be described later in order to perform photographing and adjustment with good efficiency, or to improve deterioration or replacement and maintainability due to contamination. In terms of features, it has a movable numerical aperture (NA) of 10008. Thereby, a measurement example of the beam shape with respect to the CO position of LE is shown in Fig. 46. Fig. 46 is a view showing an example of measurement of the beam shape at the CO position with respect to LE. In Fig. 46, the shape of the beam coming to the CO position is shown in the upper stage, and the phenomenon of the mirror reflection area, the transition area, and the secondary electron emission area of the beam irradiated on the surface of the sample is shown in the lower stage. Further, in the upper stage, the mirror-reflected electrons are represented by black dots, and the secondary electron-emitting electrons are represented by circles. Compared with LE, only mirror reflection can be observed in the mirror reflection electron region. child. In the transition region, it is observed that the mirror reflects electrons and emits electrons twice. Only two electrons were emitted during the discharge of the electron region twice, but no electrons were observed by the mirror. The position, size, strength, and specifications and strength of the electrons emitted twice are measured using the image data obtained by the shooting.

Further, by this observation, when the irradiated electron beam hits the target sample, it is possible to immediately determine which of the three states is the three states. In the conventional technique, it is only possible to make a fuzzy prediction from the irradiation conditions and the obtained image, and it is impossible to judge such a correct situation. In addition, the error due to the precision of the power supply setting and the influence due to the optical axis condition cannot be correctly judged. This is because the formation of the mirror-reflecting electron region and the transition region is sensitive to LE and optical axis conditions, and errors in equipment or conditions for controlling it are also affected. For example, the setting precision of the power supply is generally about 0.1%, and the setting error of the 5000V setting power supply is 5V. When a 5V change occurs, it is sufficient to release the electron region from the transition regionmirror reflection region or from the transition region2 times. Since it is impossible to confirm it, it is only possible to make a fuzzy prediction by predicting the mirror reflection area or the transition area by the set value.

Further, in the present invention, a method of performing such measurement is used, and a method of setting the adjustment of the electron beam and the setting of the NA position of the image formation are described. The orientation of the sample such as the mask and the wafer is assumed to be the coordinate and position adjustment of the secondary optical system (column).

Fig. 47 is a view showing the incident angle of the primary beam with respect to the sample in the inspection method according to the embodiment of the present invention. As shown in Fig. 47, the irradiation angle of the incident electron beam is set to θ, and the irradiation of the sample (or the column coordinates) is performed. The direction is set to α. That is, the angle with respect to the surface of the sample and the vertical direction (the Z direction, which is the same as the optical axis direction of the secondary optical system) is θ. For example, when θ = 0, it is perpendicular to the surface of the sample. When θ = 90 degrees, the sample is horizontally incident. When θ=45 degrees in the oblique direction, the sample surface was incident at 45 degrees. Moreover, θ can represent the absolute value from the Z axis. With respect to the Z axis, θ is the same value if both the right side and the left side are the same angle. Generally, θ can be used in the range of 0 to 45 degrees. In the example of α, in the sample (or the column coordinates), the E direction of E×B is taken as the Y direction, and the B direction is taken as the X direction. For example, the E+ side of E×B (the direction in which the primary optical system has) is Y+, and the E-side is Y-. At this time, when the sample is observed from the detector side, the right 90 degree direction is X+ and X- is the left 90 degree direction with respect to Y+. Further, for example, when the sample has a pattern area of a vertical line/space (L/S) and a horizontal line/space (L/S), the X and Y directions are set such that the vertical line is the Y direction and the horizontal line is the X direction. Just understand. At this time, for example, as shown in Fig. 47 (a), the incident angle α of the sample in which the X + direction is 0 degrees can be determined. When α=0, the incident direction of the primary electron beam becomes the X+ direction. When α=45 degrees in the oblique direction example, it is 45 degrees obliquely incident in the middle direction of X+ and Y+. This makes it possible to form the same electron beam irradiation direction for the vertical L/S and the horizontal L/S. Moreover, electronic signals from the same line and space can be formed to achieve the same contrast and S/N. When the above-described values of θ and α are adjusted, when the beam at the CO position of the NA diaphragm of the secondary optical system is observed, the case shown in Fig. 48 is formed. Fig. 48 is a view showing an example of beam observation of the CO position, and this example is an example of adjustment in the transition region.

The beam of electrons emitted twice is rounded at the CO position. This is electronics The beam collides with the surface formed by the sample to emit electrons, and since the direction emitted from the surface is isotropic, the CO position becomes circular. On the other hand, since the mirror reflects in the vicinity of the surface in the direction affected by the aforementioned θ and α, the mirror-reflected electrons are formed at the positions reflecting θ and α at the CO position.

For example, when the incident angle of the sample is α, the CO position is such that the position is formed in the angular direction of α with respect to the circular shape of the electron emission twice. Further, when the vertical direction of the sample surface is Z and the detector direction is Z+, the incident angle with respect to Z is θ. Depending on the magnitude of θ, the mirror reflection position of the CO position is affected. That is, as shown in Fig. 48, when θ (absolute value) is large, the distance Lm from the CO center at which electrons are emitted twice becomes large. That is, when the incident angle θ is large at oblique incidence, the mirror reflection position is formed at a position deviating from the CO center of the electron emission twice. Further, when the electron beam is incident vertically, the mirror reflection position is formed at the CO center position at which electrons are emitted twice.

This example is shown in Figure 49. Fig. 49 is a view showing a mirror reflection position formed in accordance with the incident angle of the primary electron beam. When the electron beam in the X direction is irradiated, the mirror reflection electron position is formed on the X axis with respect to the CO that emits electrons twice. When the electron beam in the Y direction is irradiated, the mirror reflection electron position is formed on the Y axis with respect to CO that emits electrons twice. When irradiated from the oblique direction α, the mirror-reflected electron position is formed in the α direction with respect to the CO that emits electrons twice. Commonly used angles α are 0 degrees, 30 degrees, 45 degrees, 60 degrees, 90 degrees, 120 degrees, 150 degrees, 180 degrees, 210 degrees, 240 degrees, 270 degrees. Further, the θ angle is used in a range of 0 to 45 degrees. Furthermore, for those having a high-contrast, S/N uneven surface, for example, EUV masks or nano-imprinted masks or semiconductor wafers are used in a range of 0 to 20 degrees.

The incident angle control of the primary system can be performed with a beam aligner of the primary system. In addition, the beam aligner of the system can be used once in the X direction, and the E direction can be adjusted in the Y direction. Moreover, in the Y direction, a beam aligner can also be used instead of E×B.

In the present invention, the NA position is adjusted in order to form an electronic image condition with high contrast and S/N. This is because depending on the relationship between the electron position of the mirror reflection and the position of the NA, the resulting image information will be different and the image quality will vary greatly. E.g,

a. Image with multiple mirror reflection electrons: set NA near the mirror reflection electron position

b. is a concave-convex pattern, and has a concave white/convex black image with more mirror reflection electrons in the concave portion

c. is a concave-convex pattern, and has a concave black/convex white image with less mirror reflection electrons in the concave portion

d. The contrast is an asymmetric image, vertical/horizontal pattern, etc.

e. An image of a mirror-reflected electron or the like is formed at an edge portion of the uneven portion.

Therefore, in order to obtain the desired image, the relationship between the mirror reflection position and the NA position must be required and set. In the conventional technique, since the understanding of the phenomenon that occurs is insufficient and the adjustment method is not known, the NA is moved to acquire an image and determine conditions. According to the present invention, work efficiency is improved, and time-cost can be greatly reduced. At this time, in order to adjust and set the NA position, there must be an NA activity mechanism. Moreover, it is more suitable if it is a two-dimensional moving mechanism. The reason is that, in the case of one-dimensional movement, the MC (mirror reflected electron position) is in an oblique direction or an axis direction in which the CO center of the electrons is discharged twice (for example, it can only move in the x direction or in the y direction). When it is not possible to set the NA between the MC and CO center positions.

Fig. 50 and Fig. 51 show examples of mirror reflection electron positions and NA positions. Fig. 50 and Fig. 51 are diagrams showing an example of the mirror reflection electron position and the NA position. Further, the same condition adjustment method can be applied to a sample having a flat surface and a flat sample. Even if it is a flat sample, if it is desired to form an image capable of grasping a change in potential or a change in material, conditions for easily changing can be obtained and produced according to the present invention. For example, it can be applied to detection of minute foreign matter, washing residue, contamination, etc. located on a flat sample surface, or mixed pattern detection of a conductive material and an insulating material. At this time, in the same manner as described above, in order to obtain the contrast of the defect or the pattern and the condition that the S/N is high, the above-described condition production method can be used. Moreover, high sensitivity detection that cannot be achieved by conventional techniques can be achieved. By making this adjustment, the fact that the contrast × 1.2 to × 2, S/N × 1.5 to × 5 can be obtained is confirmed by the method of detecting the image while viewing the image, and the adjustment time Tc and Reproducibility is very effective, for example, Tc = 1/2 to 1/10 can be obtained as compared with the prior art.

The classification of the NA setting position can be roughly divided into a case where it is disposed around the mirror reflection electron position, and a case where it is set at the separation position. The farther away the position, the smaller the effect of the mirror reflection electrons.

The formation example of the image is disclosed below.

1) is a concave-convex pattern, which is a white image of a concave white/convex black signal

Partially generated mirror reflection is formed, and the amount of electrons passing through the portion is increased to obtain an example of high contrast and high S/N.

Fig. 52 is a graph showing the relationship between the incident energy LE and the degree of gradation of the edge portion of the uneven structure on the surface of the sample. The edge portion is located at both ends of the concave portion and belongs to a portion where the height of the sample varies. In Fig. 52, the broken line indicates the degree of darkness of the edge portion, and the solid line indicates the degree of darkness of other portions. The characteristics of the other parts correspond to the characteristics of Fig. 44.

As shown in Fig. 52, the characteristic lines are different at the edge portion and other portions. The characteristic line of the edge portion is biased toward the direction in which the incident energy becomes larger. That is, at the edge portion, the lower limit of the transition region is larger, and the upper limit of the transition region is LEB+5 [eV]. Here, LEB is the upper limit of the transition region of the portion other than the edge. The reason for this shift in the characteristic line is due to the difference in shape, structure, and material between the edge portion and other portions. Moreover, due to the shift of the characteristic line, a difference ΔDN in the edge portion and other portions is generated.

Secondly, according to Fig. 52, the reason why the characteristics of the edge portion differs from other parts, and even the difference in the degree of faintness DN is reviewed.

Fig. 53 is a cross-sectional view showing the structure of the unevenness of the sample, showing a fine line/space shape. For example, the convex portion is a line and the concave portion is a space. The line width and space width are 100 μ or less. In the shape of Fig. 53(a), the conductor (Si) has an uneven shape, and an oxide film (SiO) is formed on the uppermost portion of the convex portion.₂Wait). Similarly, in the shape of Fig. 53 (b), TaBO is formed on the uppermost portion of the convex portion.

Fig. 54 is a view showing a phenomenon in which mirror-reflected electrons are generated at the edge portion of the concavo-convex structure when the electron beam is irradiated to the structure of Fig. 53(a). Figure 54 In the middle, a pattern of vertical stripes is formed. At the time of electron beam irradiation, the irradiated electrons change the orbit around one edge of the concave portion (groove), are curved in the lateral direction, and travel toward the opposite side edge of the groove. Then, the illuminating electrons change the orbit again near the edge of the opposite side, and return to the top. In this way, the irradiated electrons do not collide with the sample, but become mirror-reflected electrons. Such mirror-reflecting electrons generated at the edges can be referred to as edge mirror reflection electrons. The edge mirror reflects electrons that are symmetrical from the edges of the ends. Fig. 55 is also a view showing the reflected electrons of the edge mirror generated in the structure of Fig. 53(a), similarly to Fig. 54. In Fig. 55, a horizontal stripe pattern is formed. At this time, since the electron beam is reflected by the edge mirror, the electrons other than the electron generating portion collide with the surface to generate electrons twice. Therefore, the contrast and S/N of the pattern can be determined according to, for example, the mirror-reflecting electrons at the edge portion and the second-time electron emission from other portions. Since the transmittance of the mirror-reflected electrons is high, high contrast and S/N can be obtained. Further, depending on the shape of the pattern or the performance of the secondary optical system, there is a case where the edge mirror reflection is completely decomposed and an image cannot be formed. For example, since the degree of aberration reduction of the secondary optical system is insufficient, the edge mirror reflection electrons are integrated and observed. Therefore, there is a case where the space signal is reflected by the edge mirror and the black signal formed by the electron emission from the line portion is observed twice. Further, depending on the incident direction of the primary electron beam, there is a case where the edge mirror reflection is generated only in one direction of the edge portion. At this time, there is a case where the space signal is reflected by the edge mirror and the black signal formed by the electron emission from the line portion is observed twice.

Next, Fig. 56 shows an example of another electron orbit of the electrons being changed to reflect electrons of the edge mirror. In this example, the illuminating electrons are directed to one of the recesses. The edge is incident and enters the recess along an arcuate track passing near one edge, but does not collide with the bottom of the recess to change the direction of travel, and passes through the vicinity of the other edge of the recess to become a mirror-reflecting electron. This mirror-reflecting electron is also the edge mirror reflecting electron. In terms of the edge structure, it can be considered that each of the irradiated electrons passes through the track of Fig. 54 or Fig. 56, or passes through the intermediate track of Fig. 54 and Fig. 56, and becomes an edge mirror reflecting electron.

Next, the reason why the electron orbit is easily bent near the edge will be described. Fig. 53 is a view showing an example in which an oxide film is formed on the surface of the convex portion of the conductor. In the configuration of Fig. 53, an oxide film is formed on the surface of the convex portion of the conductor. In this configuration, the oxide film on the surface of the sample is negatively charged. Further, the conductor potential in the concave portion is relatively higher than the potential of the oxide film. Since the potential near the edge changes, the electron orbital is easily bent as described above, resulting in edge mirror reflection electrons.

In the present embodiment, precharging is also suitable. Pre-charging is electron beam irradiation performed before the sample is observed. By pre-charging, the insulating region of the sample is negatively charged (in the case of Figure 54 and the like, the oxide film on the surface of the sample is negatively charged). By precharging, the potential of the insulating region is stabilized. Thereby, the reflected electrons of the edge mirror are stably generated, and the characteristics of Fig. 52 can be stably obtained. Therefore, the sample observation can be performed in detail, and the inspection precision using the sample observation result can be improved.

For pre-charged electron beam irradiation, an electron optical system for sample observation can be used. Alternatively, an electron gun may be additionally provided for pre-charging.

Fig. 57 shows another example of the uneven structure of the sample. Figure 57 is also a cross-section of the line/space shape. In Fig. 57, the Si surface is formed Oxide film (SiO₂The convex part of etc.) In this configuration, the equipotential surfaces are meandered at the both side edges of the concave portion. Due to the influence of the equipotential surface tortuosity, the orbit of the irradiated electrons will bend. As a result, in the configuration of Fig. 57, the irradiated electrons pass through the orbits shown in Figs. 54 to 56, and become edge mirror reflection electrons. The configuration of Fig. 57 is also very suitable for precharging, whereby the oxide film potential of the convex portion can be stabilized.

Further, there is a case where the uneven structure is formed only by the conductive material. In this case, the equipotential surface is also formed along the unevenness. Moreover, at the both side edges of the concave portion, the equipotential surface is tortuous. Due to the influence of the tortuosity of the equipotential surface, the orbit of the irradiated electrons is curved. As a result, the illuminating electrons become the edge mirror reflecting electrons by the above-described orbital. Further, in the configuration of Fig. 53 (b), the construction mask without the TaBO can be considered the same. This is the case for EUV masks.

Further, even in the case where the uneven surface is formed only of the conductive material, a natural oxide film exists on the surface of the conductive film. Therefore, it is suitable for precharging, whereby the potential can be stabilized.

As explained in detail above, in the concave portion of the sample, electrons are U-turned near the edges of the both ends, and the edge mirror reflects electrons. Therefore, the edge mirror reflects electrons more easily than the mirror reflection electrons in the general portion. As a result, as shown in Fig. 52, at the edge portion, the transition region expands toward the high energy side as compared with the portion other than the edge.

Further, in the above region, the mirror-reflected electrons and the secondary discharge electrons are mixed. As described above, the electrons emitted twice are secondary electrons, reflected electrons, or backscattered electrons (or a mixture of these electrons). 2 times release of electronic system, etc. Expanded sexually. Therefore, at most only a few percent of the electrons will reach the detector. In contrast, the edge mirror reflected electrons are generated by direct reflection of the irradiated electrons. Therefore, as far as the edge mirror reflects electrons, the transmittance (the arrival rate to the detector) is almost 100%. Therefore, high brightness (darkness) can be obtained, and the difference ΔN from the surrounding degree is increased.

As described above, in the edge portion, it is easy to generate mirror-reflected electrons, and the transmittance of the mirror-reflected electrons is large. As a result, as shown in Fig. 52, the characteristic line of the darkness of the edge portion is shifted in the direction in which the incident energy LE is large, and the difference in depth between the edge portion and other portions is ΔDN.

In the present embodiment, the above-described phenomenon is utilized to produce a pattern image having a high resolution and a large contrast. The concave structure described above corresponds to the concave pattern of the present invention. In the present embodiment, the incident energy LE is set such that the edge mirror reflects electrons with excellent efficiency using a concave pattern. As shown, the incident energy LE can be set at a very low value compared to conventional viewing techniques. With this energy setting, the difference between the pattern and the surrounding depth ΔDN becomes large, and an image with high resolution and high contrast can be obtained.

Specifically, the incident energy LE is set in such a manner as to form LEALELEB or LEALELEB+5[eV]. Thereby, the incident energy LE can be set for the region where the mirror reflected electrons and the secondary electrons are mixed.

As described above, according to the study by the inventors, the result is -5 [eV] LEA LEB 5 [eV], and is, for example, LEA = -5 [eV], LEB = 5 [eV]. At this time, the incident energy LE is set at -5 [eV] LE 5 + 5 [eV] = 10 [eV]. In more detail, depending on the incident energy LE, the mixing state of the mirror-reflected electrons and the two-time discharge electrons will change, and the difference in depth will also change. Chemical. Therefore, it can be considered that a large effect can be obtained by setting the incident energy LE to a region where the number of mirror-reflected electrons is small.

2) is a concave-convex pattern, but forms a black image of the concave black/convex portion

This is a case where the specular reflection is locally generated, and since the mirror reflection fails to reach the detector and becomes a black signal, an example of high contrast and high S/N is obtained.

The mirror-reflected electrons formed in the concave portion collide with the side wall or the like, or the track is offset and fail to reach the upper CO position or the detector position, so that the signal of the concave portion is reduced and detected as a black signal.

At this time, in Fig. 52, the signal characteristic of the concave portion is (b), and the other portion is (c). The signal of the convex part is: the case where the electron is reflected by the mirror, the case where the mirror reflects the electron + the electron is released twice, and the electron is released twice. At this time, in the case where the mirror-reflecting electrons of the concave portion collide with the side wall surface, electrons from the side material are emitted twice. In contrast, the mirror-reflected electrons of the convex portion are regarded as white signals to form an image due to the large amount of electrons reaching. At this time, the black signal is formed by releasing the electron twice. Through the concave black signal and the convex white signal, the contrast of the concave-convex pattern (that is, the line/space structure) and the condition that the S/N ratio is high can be obtained, and the defect inspection with high sensitivity can be performed. Further, there are cases in which the electrons are emitted twice from the convex portion due to the collision of one of the mirror-reflecting electrons and the primary electron beam to the surface of the sample. At this time, since the electrons from the convex portion reach a large amount of electrons to the detector, the contrast and S/N of the concave-convex pattern (that is, the line/space structure) can be obtained by the black signal of the concave portion and the white signal of the convex portion. All of the higher conditions allow for more sensitive defect inspections. At this time, it is also affected by the material of the outermost layer. Compared to the side material, there will be SiO₂Or two discharges of oxide film such as TaBO A situation in which the amount of electrons is large. At this time, higher contrast and S/N can be obtained. Further, at this time, the mirror-reflected electrons formed by the concave portion may have a large deviation of the orbit, or the track may be shifted to a position where the NA cannot pass, or a large number of electrons collide with the side wall. At this time, in the same manner as described above, the electrons from the concave portion become the secondary electrons emitted when one of the primary beams incident on the concave portion partially collides with the side wall, and the amount is a small amount, so that a black signal is formed. On the other hand, although the convex portion has a mixture of mirror-reflecting electrons, mirror-reflected electrons, and electrons discharged twice, or electrons are emitted twice, in either case, the total amount of incident primary beams has an influence, so the convex portion The amount of electrons can be relatively larger than that of the concave portion, and a higher contrast ratio and S/N can be obtained.

In this case, it is also greatly affected by the relative relationship between the incident angles θ and α of the above-described one-shot electron beam and the position of the mirror-reflected electron position NA of the secondary optical system. The conditions of the NA position can also be utilized to eliminate the mirror-reflected electrons of the recess. That is, this adjustment is performed, and the conditions of obtaining high contrast and S/N of the concave-convex pattern are obtained and set.

- Beam measuring mechanism at the CO position 2nd detector -

By measuring the position and shape of the beam at the CO position, it is possible to adjust the conditions of the electron beam and the high precision for various patterns. In order to make this adjustment, it is effective to have an active numerical aperture. In particular, it must be a diaphragm that can move in the direction of 2 axes (x, y directions). Further, since the CO position changes in the z direction due to the lens condition, it is more suitable when moving in the three-axis directions of x, y, and z.

However, even if you have a movable numerical stop and the detector is one, the detector is used for each adjustment. Mask in 2 optical system imaging or The electronic image from the wafer is amplified by the microchannel plate (MCP) of the detector and then converted into a light image by hitting the fluorescent screen. The image converted into light by the detector is projected on the TDI-CCD in a one-to-one manner via a vacuum penetrating window through a FOP system placed in the atmosphere. If the detector is used frequently during each adjustment, the microchannel plate (MCP) and the like may be damaged, and the detector must be frequently replaced. When the optical axis adjustment or the electronic image adjustment of the optical system is performed twice, since the static image is to be used, the electron intensity distribution in the static image needs to be maintained for a long time. In other words, imaging is performed in a state where a portion having a large amount of electrons and a small portion are kept for a predetermined period of time or longer. At this time, since the components having a large amount of electrons and the components of the smaller portions are different in degradation, a difference in gain (Gain) is locally generated, and the gain variation of the detector itself continues to occur. When this is checked in the next stage, it becomes a major factor in the deterioration of the detection performance such as an increase in the amount of virtual defects. Therefore, when shooting a still image, it is more appropriate to have another detector. At the time of inspection, the detector continuously changes the position at which the electron distribution is changed in a short time while moving the stage, so that the deterioration due to the difference in gain is small.

Therefore, the detector does not need to be frequently replaced, and as a means for measuring the beam position and shape of the CO position and an apparatus for adjusting the optical axis, the second detector can be disposed as the CO position directly in front of the inspection detector. A detector for beam measurement. Fig. 58 is a schematic view showing the second detector of the invention of the present invention. Fig. 58(a) is a view showing the secondary optical system of the invention of the present invention, and Fig. 58(b) is a view showing the electron beam which emits electrons and mirror reflected electrons twice at the position of the numerical aperture (NA) 10008 through the lens. The image is imaged in the second detector 76-2. The numerical aperture 10008 and detection shown in Figure 58 (b) The second detector 76-2 according to the embodiment of the present invention is provided between the systems 76, and the movable numerical aperture (NA) 10008 is moved, and the beam position and shape of the CO position are captured by the second detector. Here, the beam shape and position of the CO position (or NA position) can be photographed by static drawing. Based on the information captured by the second detector 76-2, the adjustment is repeated and the inspection is performed after the adjustment.

The electron emission or mirror reflection electrons through the numerical aperture (NA) 10008 are imaged on the sensor surface of the detector. The two-dimensional electronic image obtained by the imaging is obtained by the second detector 76-2, and converted into an electrical signal and transmitted to the image processing unit. In order to capture the electron beam image of the CO position by the second detector 76-2, a transfer lens or an electrostatic lens for projection projection may be used between the numerical aperture 10008 and the second detector 76-2.

An EB-CCD or a C-MOS type EB-CCD can be used as the second detector 76-2. The component specification may be 1/2 to 1/3 of the component size of the EB-TDI which is the first detector (detector 761). Thereby, Px size shooting smaller than the first detector can be performed. The Px specification is the value of the component specification divided by the optical magnification, that is, the image segmentation specification on the surface of the sample. For example, when the component size is 10 μm and the magnification is 1000 times, the Px specification = 10 μm / 1000 times = 10 nm. In the case of the second detector having a component size smaller than that of the first detector, the surface can be observed from the Px size which is smaller than that of the first detector. The EB-TDI of the first detector, the EB-CCD of the second detector, or the EB-CCD of the C-MOS type do not require an optoelectronic conversion mechanism and an optical transmission mechanism. The electrons are injected directly into the EB-TDI sensor face or the EB-CCD sensor face. Therefore, without the deterioration of the resolution, a high MTF (modulation transfer function, Modulation Transfer Function) and contrast. Compared with the conventional EB-CCD, the C-MOS type EB-CCD can significantly reduce the background noise, so it is very effective for reducing the noise generated by the detector. When shooting under the same conditions, the contrast can be improved and the S/N can be improved compared to the conventional technique. In particular, it is effective when the number of electrons is small. In terms of noise reduction, it has the effect of 1/3 to 1/20 of the conventional EB-CCD.

The beam imaged by the numerical aperture (NA) 10008 on the detector surface is detected using the second detector 76-2, and the condition and shape of the electron beam are adjusted according to the position and shape of the detected beam. The position of the light (NA) 10008. After various adjustments are made based on the detection results obtained by the second detector 76-2, the sample inspection using the detection system 76 is performed. Therefore, since the detection system 76 is only used during inspection, the frequency of replacement of the detection system 76 can be suppressed. Further, since the second detector 76-2 only captures a still picture, even if deterioration occurs, it does not affect the inspection. In order to achieve such imaging conditions, for example, the condition in which the first detector images the electronic image, the condition in which the second detector is imaged, and the beam shape that comes to the CO position in order to observe the beam at the CO position are detected in the second detection. For the conditions of imaging of the device, etc., referring to the example of Fig. 33(a), the lens intensity of the transmission lens 10009 is adjusted to obtain the optimum conditions for the first detector and the second detector. A situation in which some imaging condition is used. In addition, a lens 741 may be used instead of the transmission lens 10009. Since the distance between the center of the lens and the detector changes, the magnification is also changed when the transmission lens 10009 and the lens 741 are used, so that an appropriate lens and magnification can be selected.

The second detector 76-2 described above can be used in combination with the above-described adjustment method of the present invention for measuring the beam position and shape of the CO position to perform electron beam conditions and high-precision adjustment. Further, the second detector 76-2 is not only applicable to an electro-optical device including the novel photoelectron generating portion of the present invention, but also applicable to an electro-optical device including a general-purpose electron gun. This embodiment can also be applied to the devices described in the first to eleventh embodiments. In the above-described example of the method of adjusting the beam and the NA position, an example in which the primary beam is an electron beam is described, but the irradiation system may be applied to light or a laser. The laser or the light is irradiated to generate photoelectrons from the surface of the sample, and the photopolymerization polycondensation specification or the relationship between the center position and the NA setting position can be used. Thereby, a photoelectron image having excellent resolution can be formed.

Further, as another example, as shown in Fig. 58(c), the third detector 76-3 that moves in the x and y directions in conjunction with the NA may be provided at the NA position. For example, the third detector 76-3 may be integrally provided on the setting plate of the NA. At this time, in order to observe the shape-position of the beam coming to the NA position, the setting plate can be moved to move the third detector 76-3 to a coordinate whose center coincides with the center of the optical axis, and the third detector 76 is directly used. -3 Observe the beam coming, whereby the rear lens adjustment is not required.

(Embodiment 12)

-Inspection device with optical microscope and SEM in the same room -

Further, when the sample is inspected by the above detector, it is necessary to observe it by SEM. Therefore, it is very effective if the imaging optical inspection apparatus and the SEM are installed in the same room (refer to Fig. 59). For example, in an EUV mask, In the inspection of fine patterns such as NIL (Nano-imprinted lithography) masks, it is required to perform inspection of ultra-fine patterns with high sensitivity using imaging conditions of pattern optical patterns and pattern defects. When the optical inspection apparatus and the SEM are installed in the same room, the sample is mounted on the same platform, and the sample can be observed and inspected by both the image method and the SEM method. The methods and advantages at this time are as follows.

First, since the coordinate relationship is obtained in an unambiguous manner when the sample is moved between the image and the image on the same platform, the same portion can be easily defined with high precision. When the sample is moved by the separate devices, in order to set the other platform, the alignment of the sample must be performed separately, or even if the alignment is performed, the specific error of the same portion may be 5 to 10 μm or more. In the current device, due to such a positional error, a portion where the deviation from the defect portion is detected occurs, resulting in a false detection of a defect-free portion and a erroneous determination as a defect-free condition. In particular, in the case of a non-patterned sample, the error is greater because the positional reference cannot be defined. The error is about 2 to 10 times compared to the case of a pattern.

Secondly, since it is installed in the same room and platform, even if the sample is moved by the imaging method and the SEM, the same portion can be defined with high precision, so that the definition of the portion can be performed with high precision. The precision can be, for example, 1 μm or less, and the position of the foreign matter or defect to be observed can be moved in the range of 0.05 to 1 μm. Thereby, in the case of performing the inspection of the pattern and the pattern defect by the image method, it is very effective in performing the definition and detailed observation of the detected defect by the SEM. Since the definition of the part can be made, it is not only possible to judge the presence or absence of the defect (if it is not, it is a virtual test), The correct size-shape of the defect can be judged at high speed. For other devices, pattern defects and their definition take a lot of time.

As described above, the imaging method of the pattern optical pattern and the image defect of the pattern defect are used to inspect the ultra-small pattern with high sensitivity, and the device system in which the imaging optical method and the SEM are mounted in the same chamber, in particular, ultra-light of 100 nm or less. The inspection of small patterns and their determination-classification can be carried out very efficiently and at high speed. The embodiments thereof are described in detail below.

(Example 1)

When the optical microscope, the imaging optical system, and the SEM described above are installed in one room, they are characterized by the following functions and mechanisms.

It is important to obtain the optical system center of each of the above-mentioned optical microscope, imaging optical system, and SEM in advance; secondly, it is important to store the coordinate relationship of each center in the memory; and to move to the center coordinates of each optical system to be memorized. Therefore, it is better to have an organization that can meet these needs. In one room, when the same portion is observed for the sample set on the same platform, it is possible to simply move from the control screen of the PC between the optical centers of the memory by using a button or a click operation. As for the precision, since it is disposed on the same platform, it can be moved with high precision and the position is stationary, which can reach a precision of, for example, 0.05 to 1 μm. Moreover, if the control at the stop is carefully designed, the precision of 0.05 to 0.1 μm can be achieved. The control example is a method in which a plurality of stop allowable values are used to make them stand still. For example, two kinds of allowable values of the allowable value A 1 μm and the allowable value B 0.1 μm are used. The first stop control system uses the allowable value A, and then uses the allowable value B to stop it. Use the majority of tolerances in this way and use the tolerances step by step. At a small value, the stop operation can be smoothly performed with high efficiency and high precision.

(alignment procedure)

In order to perform these actions, it is necessary to align the sample, align the detector sensor in the imaging optical system, and in each optical system (ie, optical microscope, imaging optical system, SEM). The relationship between the alignments formed by the SEM image. Depending on how these factors are predetermined, the operational process, its time or positional precision, and the virtual judgment-defective classification performance will be affected. In order to efficiently determine this with high precision, the following procedures are performed.

a) Determine the alignment of the sample using an optical microscope (see Figure 60).

This procedure determines the direction of the sample by performing an operation that matches the direction of movement of the platform with the direction of the sample. For example, the sample is rotated by a rotating platform or the like, and the rotation angle θ s of the sample is determined, so that the moving direction of the stage (y direction) coincides with the y direction of the sample. For example, when the mark of the representative pattern has two or more points in the y direction of the sample, for example, a pattern or mark having a distance of about 10 to 300 mm in the y direction in the two portions is used. At this time, the rotation angle θ s of the data is determined and determined so that the two patterns or marks that conform to each other come to the optical center of the optical microscope. At this time, by performing image processing such as pattern matching, the alignment precision can be obtained from 1/10 to 1/100 Px.

b) Determine the rotation angle θ t of the detector for detecting the electronic image of the image optical system (refer to Fig. 60 and Fig. 61).

This angle is for the TDI sensor (time integral CCD-TDI sense) The y direction of the detector or the CCD sensor (the direction in which the pixels are arranged in the y direction) is adjusted in accordance with the moving direction of the platform, and the rotation angle θ t is adjusted. In the specific example, the following operations are performed. Patterns or marks are used with a distance of 10 to 300 mm in the y direction. Through the above a), the moving direction of the platform and the rotation angle θ s of the sample are adjusted, and the adjustment of the platform moving direction y and the y direction of the sample is performed with high precision, for example, 1/1000 to 1/100000 rad, or Adjusted to 1/10000 to 1/100000 rad with high precision. In this state, the platform is moved in the y direction, and the TDI image is acquired in synchronization with this, and the resolution of the pattern or the marked TDI image is obtained as an optimum state, for example, the rotation angle θ t when the contrast is maximum. . As for the pattern used, a one-dimensional L/S pattern or the like can be used in the Y direction. If the rotation angle θ t is shifted, the L/S pattern is blurred and the contrast is lowered. When the rotation angle θ t reaches an appropriate value, the contrast rises and an optimum value can be obtained. By performing this operation, the platform moving direction y and the y-direction offset of the TDI or CCD sensor can be set to 1/1000 to 1/100000 rad. Moreover, it can reach 1/10000 to 1/100,000 rad through high-precision adjustment.

c) Next, find the center coordinates of the TDI frame (refer to Figure 62 and Figure 63).

Since the TDI image is a two-dimensional continuous image, the continuous image is divided into frames by image processing. For example, a frame of 1000×1000 Px, 2000×2000 Px, 4000×4000 Px, or the like is used. The adjustment is made in such a way that the center of each of the frames becomes the target position. An example of the method of determining the target coordinates is listed.

A) For example, a pattern portion or a marking portion having a feature is used. Place these in the optical center of the optical microscope and remember these coordinate values. Of course Then, the TDI image acquisition start position or end position is determined such that the pattern portion or the mark portion becomes the center of the frame of the TDI image. The adjustment of the position in the y direction enables parameter adjustment on the image processing (that is, the frame division start position). Since the frame division start position can be adjusted in units of 1Px, the adjustment can be easily performed. For example, if it is 1 Px 50 nm, it can be adjusted with a precision of 50 nm, and usually 10 to 500 nm/Px is used.

B) The adjustment of the position in the x direction is performed by slightly adjusting the coordinate position of the optical center in such a manner that the target pattern or mark comes to the center position of the sensor in the x direction. Alternatively, fine adjustment can be performed by the last stage deflector. At this time, it can be adjusted with a precision of 1/10Px to 10Px, or an adjustment of 1/10 to 1Px is often used.

C) Next, confirm whether the target pattern or mark comes to the optical center of the SEM (Scanning Electron Microscope). When there is an offset, the coordinate relationship between the optical centers is corrected. That is, it is confirmed whether the target pattern or mark is located at the center of the optical microscope and the center of the TDI image frame of the image optical system and the center of the SEM image, and it is confirmed and determined to be within the allowable value. After confirming that the target pattern or mark has entered the allowable value from the center of the TDI image frame and the SEM image of the optical microscope or the imaging optical system, the distance between the positions of the optical centers is determined and stored in the memory or the like. The allowable value can be set to 1 μm or less. Moreover, the adjustment of high precision can also be set to an allowable value of 0.1 μm or less.

D) In other respects, if the detector used in the imaging optical system is a CCD or an EB-CCD in which electrons are directly incident on the surface of the sensor to form an image, since the platform acquires an image at a stationary state, it comes to the optical center. Its static image center is adjusted and decided to make adjustments - decided to optical microscopy The EB-CCD image of the mirror and the imaging optical system, the optical center of the SEM image, and the pattern or mark enter the allowable value. The program or permissible value is the same as above.

As described above, the present invention is characterized by having: A) the step of aligning with an optical microscope, and B) aligning the direction of the detector sensor (y direction: the cumulative direction of the sensor) by the imaging optical system. Steps, C) Obtain the optical center of the optical microscope-image optical system-SEM and memorize the relevant coordinates. Further, when the TDI image of the image optical system is used, the step of obtaining the center of the frame of the TDI image is a necessary step in the step B).

Further, in the step C), when the precision of the SEM image is further determined, the following method can be performed. Adjust and determine the x and y directions of the image frame that forms the SEM image. The direction of the target is that the SEM image of the pattern or mark is acquired at the end of the above A) and B), and the offset in the x and y directions is extracted and corrected. The means of correction is to make the offset directions of x and y consistent with a small amount of continuous and uninterrupted. In order to achieve this, the deflector required to form the SEM image (i.e., required for scanning) can be used with more than 8 poles. Thereby, the control of the deflection angle can be adjusted from 1/1000 to 1/100,000 rad. The content of this job can also be done in step C).

In the use example of the apparatus system of the present invention, after the step of the pre-inspection operation is completed, the defect inspection is performed using the TDI image in the imaging optical system. Then, the patch image and the coordinate value of the inspection result are outputted and stored in a memory or the like. Secondly, when the SEM is used for detailed inspection, the portion where the defect is detected by the TDI image is photographed in detail by SEM to perform defect or Virtual decision.

At this time, the image processing device can perform defect determination and virtual determination through the SEM image. Regarding this step, there are the following methods.

First, the SEM images are compared with each other: by SEM and reference image comparison.

Second, a comparison of the SEM image and the TDI image (differential image). The difference image is a defect image obtained during the inspection, and is cut out from the vicinity of the defect detected by the TDI scan image and stored in the memory. It is usually carried out at around 50 to 200 Px. When the value is the long side of the image, it is more suitable to use 1 to 1/3 of the short side.

In particular, in the second method, when the x and y directions of the SEM image are shifted, image processing such as pattern matching cannot be performed smoothly, and therefore it is necessary to adjust the x and y directions of the SEM image in advance to determine the operation. For example, pattern misalignment can occur due to the positional shift of the pattern.

Furthermore, in the x and y direction offset correction of the SEM image, the method is also performed by image processing. It is also feasible to obtain the correction amount of x and y of the SEM image in advance, and to perform correction comparison when comparing with TDI image.

(Example 2)

In the present embodiment, when the detailed inspection is performed by SEM, the adhered dirt can be removed.

It is known that there is a phenomenon in which dirt such as carbon adheres when performing SEM observation. When the SEM is observed in detail, the dirt itself is generated in the form of a defect due to the adhesion of the dirt. Especially in the end region of the beam scanning, a lot of dirt is generated.

To solve this problem, the present invention uses two methods.

First, in the SEM observation, a method of purifying the surface of the sample while simultaneously performing SEM observation by introducing a dirt reaction gas is performed.

Second, after SEM observation, a method of purifying by a mapping optical system. Among them, there may be cases where a gas which is reactive with dirt is introduced and a case where it is not introduced.

In the first method, a fluorine gas such as oxygen, oxygen, Ar or the like, or a fluorine-based gas such as SF6 is introduced, and the SEM is observed in detail. According to this method, although dirt is generated in the SEM observation, the gas is introduced into the gas, and the generated gas reacts with the generated gas to be removed in a gas state such as sublimation. At this time, it is important to adjust the gas introduction amount to the left and right which have no influence on the resolution of the SEM image. Its role is described below. The residual gas particles in the vacuum vessel are excited by the electron beam, and are polymerized by C or H to cause dirt such as carbon or DLC to adhere to the surface. The amount produced is grown in a relationship proportional to the amount or time of irradiation of the beam. In this case, when the gas such as oxygen which is reactive with dirt is introduced, the gas particles are excited by the irradiation of the electron beam to become active gas particles, and are formed into a gas reactive with dirt such as oxygen radicals. particle. Then, the reactive gas particles react with the dirt to become, for example, CO, CO.₂It is removed by the gas particles. As a result, SEM observation can be performed in a state where the amount of dirt adhesion is small.

The second method described above is a purification method for achieving soil removal at a high speed after detailed observation by SEM. As the electron beam irradiation, beam irradiation is performed by a mapping optical system that is irradiated with a surface beam. At this time, the reactive gas is introduced, and the gas particles are irradiated by surface beam to make it into a gas. As the active gas particles, gas particles such as oxygen radicals are formed to remove the dirt. The dirt removal effect is the same as described above. The advantage of using this face beam is that it can be processed at high speed. As the surface beam, for example, a beam of 200 × 200 μm, when the stage is moved at a speed of 30 mm/s, an area of 100 mm square can be irradiated for about 30 minutes. In this manner, the soil removal treatment can be performed at a high speed of 2 to 3 times or more of the SEM type. At this time, in order to perform the dirt removal efficiently, it is effective to remove the dirt in the boundary portion where the SEM has been observed in detail. In this region, since a large amount of dirt is generated, the dirt removal in the region can be greatly reduced, and the defect of the dirt can be greatly reduced. Practically, the purification of the region is often sufficient.

The optical microscope, the imaging optical system, and the SEM described above are provided as a whole in a semiconductor inspection apparatus in one room, and are shown in Fig. 43 which has been described above. The detailed composition has been separately described, and the description thereof allows the use of this configuration. As described above, the definition of the location can be obtained with high precision, and the alignment adjustment is very simple. Moreover, the inspection of the ultra-small pattern of 100 nm or less and its determination-classification can be performed very efficiently and at high speed. Further, this embodiment can also be applied to the device configurations of the above-described first to eleventh embodiments. It is very effective to install the SEM and the optical microscope of the present embodiment in the same apparatus system for inspection and detailed observation of the apparatus having the same imaging optical system.

(Embodiment 13)

-Particle countermeasures -

With the advancement of the micronization technology, the influence of the particles is increasing, and the countermeasures for preventing the adhesion of the particles are becoming increasingly strong. In the implementation as in this technology pair 5 to At the same time as the device for defect inspection of foreign matter or pattern specifications of 30 nm, it is necessary to have the same level of particle adhesion prevention specifications in terms of preventing adhesion of particles. In this regard, any of the conventional devices is insufficient. However, this can be achieved in accordance with the present invention. The countermeasures for the particles used in the inspection apparatus and the inspection method of the present invention will be described with reference to Figs. 64 and 65. The following is only stated in terms of changes that are different from the conventional countermeasures.

- Ceiling cover installation -

(efficacy)

Protection of particles falling from the ceiling

The ceiling cover is mounted in such a manner as to cover the moving area of the sample of the platform. Thereby, it is possible to obtain the effect that the particles from the tribe down to the main chamber having the sample do not adhere to the sample. Further, a conductive conduction cover is formed around the surface of the sample such as a mask to form the same potential as the surface of the mask. Further, the thickness of the end portion of the conduction cover is about 10 μm to 300 μm, and the back surface is in contact with and is conducted to a region having a conductive film on the surface of the sample. This is to set the above thickness in order to minimize the influence of the potential change of the surface of the mask in the vicinity of the conduction cover. At this time, the width of the conduction cover is about 10 to 30 mm in the outward direction from the sample. With this configuration, by reducing the distance between the conduction cover and the ceiling cover, the probability of particles entering the space colliding-attaching to the conduction cover and the ceiling cover is increased, and the particles are prevented from intruding into the surface of the sample. Moreover, by covering the surface of the sample to which the RTD voltage is applied by the ceiling cover with the GND potential, it is possible to prevent the particles above the ceiling cover from being attracted by the electric field.

- Dust collector -

(efficacy)

A dust collector is provided, and an electrode to which the same voltage is applied is disposed around the sample to which the RTD voltage is applied, so that the particles coming from the periphery can be attracted-attached. In the example of the dust collector, there are cases where the electrode is one segment and a case where there are a plurality of segments. For example, in the case of the two-stage, different voltages can be applied by the inner-outer electrodes to increase the dust collecting efficiency. For example, any voltage having a potential higher than the surface potential of the sample or a low potential is applied, so that particles having a positive charge are adsorbed to the dust collecting electrode having a lower applied voltage, and particles having a negative charge are adsorbed to the applied voltage. Higher dust collecting electrode. In this way, the particles are not allowed to reach the surface of the sample.

-Supersonic motor cover -

(feature)

The movable member used in the vacuum chamber, for example, an ultrasonic motor or the like, is covered with a cover to cover the vacuum chamber as a dust generation source. Further applying a voltage to the cover allows the particles to be actively attracted-adhered by the cover.

The particles attached to the moving plate can be captured by arranging the moving plate cover at the movable portion of the moving plate.

- Platform cable -

(feature)

By using a flat cable (Teflon (registered trademark)) for the platform cable that operates in the vacuum chamber, the particles generated by the friction between the cables can be reduced.

A cable holder (Teflon (registered trademark)) is provided on the flat cable contact surface to eliminate friction between the cable and the metal. This is because the usual cable has a round profile The core wire is mostly covered with a resin. When there are many such cables, most of the cables are bundled together and bundled with a INSULOK. At this time, since the cable in the bundled state is moved-deformed as the platform moves, the cables are rubbed against each other to generate particles. Further, the present embodiment can also be applied to the devices of the above-described first to twelfth embodiments.

(Embodiment 14)

(Special axis of photoelectron image)

The specific axial adjustment of the optoelectronic image of the inspection apparatus and the inspection method of the present invention will be described.

For example, Figures 26 and 27 and the embodiments are applicable and are incorporated by reference. The surface of the sample was irradiated with a DUV laser to generate photoelectrons from the surface of the sample, and the image was magnified and imaged by a secondary optical system, and a two-dimensional photoelectron image thereof was taken. In the imaging condition setting of the optoelectronic image, the surface potential is positively charged by photoelectron emission for the RTD condition of the secondary optical system, for example, -4000V. Because of this surface potential, the RTD will shift from -4000V, so the offset is corrected by changing the potential of the RTD.

(simultaneous illumination of different wavelengths of light)

In the inspection apparatus and the inspection method of the present invention, an example in which the sample W is simultaneously irradiated with light having different wavelengths will be described.

In the case of a sample belonging to an EUV mask and having a concave-convex structure pattern, in the case where the outermost layer of the convex portion is TaBO and the concave portion is Ru, the contrast of TaBO and Ru becomes TaBO depending on the wavelength. Ru (λ = 266 nm), or TaBO &lt; Ru (λ = 244 nm). That is, when the two wavelengths are simultaneously irradiated to the surface of the sample patterned by TaBO and Ru, the pattern It disappears, and a portion other than the pattern constituting material (for example, foreign matter) is detected. This method can improve the detection sensitivity of the particle inspection or the defect (foreign matter) of the pattern concave portion (Ru).

When the light having different wavelengths is simultaneously irradiated, the quantum efficiency (Quantum efficiency) of each of them is not the same, and in order to eliminate the contrast of TaBO and Ru, the DUV light intensity or the DUV laser intensity of each is adjusted. necessary. At this time, in the case of laser irradiation, the polarizing filter is used, and the angle of the polarizing surface of the laser and the polarizing surface of the polarizing filter are adjusted, that is, the laser irradiation intensity to the surface of the sample is adjusted. In addition, depending on the wavelength of the light, the quantum efficiency will be different, so the amount of change in the surface potential varies depending on the material of the surface of the sample.

At the aforementioned RTD-4000V, when the surface potential of the detector is GND, the energy of photoelectrons is 4000 eV. When the lens condition of the secondary optical system at this time is optimized by the adjustment of the RTD, in order to match the material to be observed as a white signal (the amount of photoelectron generation is relatively large), other materials are largely shifted from the lens condition. The wavelength of the illumination and the wavelength of the pre-inspection wavelength (Laser) can increase the contrast difference and enhance the signal strength. Taking the 33rd (a) and 35th to 42th as the target examples, the photo-surface wafer is irradiated with a laser, and the photoelectrons generated here are transmitted to the surface of the sample through the primary optical system as a primary beam pair. The surface of the sample was subjected to photoelectron irradiation, and the electrons emitted from the surface of the sample were subjected to magnification and imaging by a secondary optical system through a secondary optical system. At this time, it is preferable that the laser light irradiated on the photoelectric surface wafer has a work wavelength larger than a work function of a material constituting the photoelectric surface, and the energy is infinitely close. This is to reduce the amount of kinetic energy that photoelectrons have to the extreme. small. That is, the energy width of the primary electron beam can be reduced in a conventional manner. Thereby, the energy dispersion of the secondary electron emission or the mirror reflection electron generated by irradiating the surface of the sample once with a beam (photoelectron) is also small, and in the imaging system, the image quality with small aberration and sharpness can be obtained. . In the photovoltaic surface wafer structure, the one-side surface of the substrate having a good laser permeability is coated with a photovoltaic material. For example, Ru, Au, Ag, and the like. The laser is irradiated from the back side of the coated surface, and photoelectrons are generated from the coated surface, and the photoelectrons are emitted in the opposite direction of the laser irradiation. By doing so, it is not necessary to form an angle between the laser irradiation axis and the optoelectronic axis, and the structure of the Gun can be simplified.

(Embodiment 15)

- Atmospheric handling for handling 2 layers of POD -

The air handling of transporting two layers of POD in the inspection apparatus and inspection method of the present invention will be described with reference to Fig. 66.

(action - condition)

The two-layer POD refers to a case where a box covering a sample such as a mask forms a two-layer structure. Since the inner layer POD has a gap or a hole, the outer layer POD can be opened during the load lock. At this time, the inner layer POD is composed of a lower plate and an upper plate. The EUV mask pattern surface is close to the lower plate setting. At this time, the action flow is as follows. Outer POD opener<img file="TW201634917A_D0003.tif" wi="59" he="44" img-format="tif" img-content="character" orientation="portrait" inline="no" />Inner POD opener<img file="TW201634917A_D0004.tif" wi="62" he="44" img-format="tif" img-content="character" orientation="portrait" inline="no" />Rotating unit<img file="TW201634917A_D0005.tif" wi="57" he="44" img-format="tif" img-content="character" orientation="portrait" inline="no" />Reversal unit<img file="TW201634917A_D0006.tif" wi="62" he="52" img-format="tif" img-content="character" orientation="portrait" inline="no" />Power removal unit<img file="TW201634917A_D0007.tif" wi="59" he="47" img-format="tif" img-content="character" orientation="portrait" inline="no" />Pallet carrying unit<img file="TW201634917A_D0008.tif" wi="59" he="47" img-format="tif" img-content="character" orientation="portrait" inline="no" />Load the lock chamber.

(effect - efficacy - advantages)

This embodiment is based on a two-layer POD. However, in the operation flow, the inner layer POD opener takes out the mask in a single manner. Therefore, one layer of POD and two layers of POD can be used in combination. In addition, this embodiment can also be applied. In the first embodiment to the fourteenth embodiment.

(Embodiment 16)

-Relationship with small SEM -

A small SEM used in the inspection apparatus and inspection method of the present invention will be described with reference to Fig. 67.

In this embodiment, the detected defects are inspected at a high magnification by a scanning electron microscope (hereinafter referred to as SEM) in the same chamber as the defect detecting device (image type electron beam inspecting device).

(efficacy)

Since detailed inspection can be performed in the same room, there is no sample movement between the devices, and the time efficiency is excellent, and there is no contamination of the sample.

This embodiment is characterized by determining the actual defect or the virtual defect of the detected defect based on the result of detailed observation by the above SEM.

(efficacy)

Improve the reliability of defect detection results by performing detailed defect classification (ADC) using detailed image data.

The SEM used for detailed observation is characterized in that all of the electron lenses are composed of electrostatic lenses.

The SEM composed of the electrostatic lens can be used for the observation of the alignment mark even when the distortion of the image is small even when viewed at a low magnification.

(efficacy)

In the case of an electromagnetic lens, in order to obtain a low-distortion image at a low magnification, a system for correcting the scanning shape of the beam is required, but an electrostatic lens is not required, and the system can be simple. Moreover, in the case of an electromagnetic lens, there is a current The coils constituting the lens are circulated to generate a magnetic field and form a lens field. Therefore, it takes about 1 to 5 hours from the time the device is used until the temperature is stabilized and the resistance is stabilized, that is, until the stable lens operation state is formed. On the other hand, when a fixed voltage power supply is used for an electrostatic lens, if a predetermined voltage output setting is performed, it will reach a stable state in about several minutes. From this, it can be seen that the correspondence between conditional stability and change is excellent. Further, the present embodiment can be applied to the above-described first to fifteenth embodiments.

(Embodiment 17)

-Laser beam adjustment method -

The method of adjusting the laser beam used in the inspection apparatus and inspection method of the present invention will be described. This adjustment method is applied to Figs. 26 to 31 and related embodiments, and these embodiments are taken as examples.

In order to improve the resolution of photoelectrons emitted from the sample surface, it is necessary to set the voltage of the lens of the secondary system to an optimum value, that is, a so-called optical axis adjustment operation. The optical axis adjustment is usually performed by observing the image of the initial low magnification, making a rough axis adjustment, and slowly increasing the magnification to perform high-precision axis adjustment.

It contributes to the amount of photoelectrons observed, and the higher the magnification, the less. That is, if the photoelectron density is constant and the magnification is higher, the Px specification is smaller, and the amount of photoelectrons per 1 Px decreases as the Px specification decreases. Therefore, when the power is not changed, when the low magnification is changed to the high magnification, the amount of the signal is insufficient and only a dark image can be obtained. Conversely, when the low magnification is maintained at a high magnification while maintaining the optimum laser power, the number of photoelectrons is too much and saturated, and the desired contrast cannot be obtained. Therefore, the laser must be adjusted The power of light. As far as the laser itself is concerned, there are many situations where power adjustment is not possible. In this case, the laser beam power reaching the sample surface can be adjusted by using an optical element such as a variable beam splitter, an attenuator, a polarizing element, or a lens.

The variable beam splitter utilizes a flat optical element that adjusts the angle relative to the laser beam to change the ratio of transmitted light.

The attenuator can integrate optical components such as most variable beam splitters into one body, which is easy to handle.

The polarizing element changes the transmittance according to the polarization state of the beam, or changes the phase to change the polarization state. There are polarizing plates or wave plates, polarizing plates, and the like. By combining these components, the power of the light can be controlled by simply passing a specific polarization state.

Regarding the lens, by changing the position of the plano-convex lens for concentrating, for example, the focal length is changed, and the profile of the power is changed, so that the local power density of the observation range is changed, and the amount of photoelectrons is controlled more or less. In addition, the thickness or the amount of the light-transmitting material such as a lens or quartz glass may be adjusted, or the number of reflections of the mirror may be increased or decreased to adjust the power.

The method of adjusting the laser power has the following methods in addition to the use of the elements as described above. For example, when the angle of the mirror is changed to change the irradiation position of the beam, the power density of the observation surface changes. When a portion having a higher power density is irradiated to the observation portion, a large amount of photoelectrons can be obtained, and when a portion having a lower power density is irradiated to the observation portion, a small amount of photoelectrons can be obtained.

A method of using two or more types of light sources having different powers may also be considered. example For example, in the case of low magnification observation, a mercury xenon lamp having a low power in the ultraviolet region is used, and a YAG4 double-wave solid laser or an Ar ion laser double-wave gas laser can be used for high magnification observation. At this time, the mercury xenon lamp can be introduced into the vacuum by an optical fiber, and then irradiated to the sample surface directly from the fiber exit or via an optical element such as a mirror.

The laser beam emitted from the laser is introduced into the vacuum chamber through the synthetic quartz viewing window by using the first mirror, the collecting lens, and the second mirror. After passing through the viewing window, it is reflected by a triangular mirror disposed near the center of the axis of the cylinder, and is irradiated on the sample surface at an angle of about 0.1 to 30 degrees from the center of the axis of the electron beam.

The triangular mirror is a mirror for enabling the electron beam to pass through the center of the shaft with a hole having a diameter of 0.5 to 5.0 mm. The surface of synthetic quartz or phosphor bronze is coated with aluminum. The potential of the triangular mirror is set to be the same as the ground potential of the ground potential, so that the electron beam passing through the hole of the triangular mirror is not bent by the electric field. When a mirror is made of synthetic quartz, the hole must also be coated with aluminum to ensure electrical conductivity.

In order to adjust the power of the laser, a polarizing element, a beam splitter, an attenuator or the like may be disposed, for example, in the rear of the lens.

The above describes a method in which a mirror or a lens is disposed on the atmosphere side, and these lenses or mirrors may all be disposed in a vacuum chamber.

(Embodiment 18)

- An inspection device for controlling the incident angle θ of the beam toward the sample surface and the incident direction α -

The method for controlling the incident angle θ and the incident direction α of the primary electron beam toward the sample surface by the inspection apparatus and the inspection method of the present invention will be described.

(differences from prior art)

In the image projection type electron beam apparatus, the LE of the electron beam is used as a transition region including both the mirror reflection electrons and the secondary electron emission, and the NA position is optimized, and the electrons emitted from the conductive material are taken out and high. Techniques for contrast line and space (Line And Space) or contact plugs are disclosed in International Publication No. WO 2009/125603.

In addition, the paragraph [0205] of the Japanese Patent Application No. 2010-091297 also discloses a method of optimizing the NA position while observing the electron distribution on the NA surface by imaging the NA surface on the lens condition of the detector. In the present case, a method of controlling the incident angle of the primary beam and the incident direction toward the sample surface to further obtain a high-contrast image was found.

Figure 6 (B) of the Japanese Patent Application No. 2010-091297 shows a schematic view of the NA surface (Fig. 68).

The inventors have learned that in the polycondensation (CO) image of the optical condition of the NA imaging, even if the LE of the primary beam is set as the transition region of the electron emission and the mirror reflection electron twice, the distribution of the ec is encountered. In addition to the distribution of the second electrons (CO), there is a distribution (MC) of mirror-reflected electrons that bounce back without hitting the sample (Fig. 69).

The image obtained by changing the LE under the above-described NA imaging conditions corresponds to the schematic diagram of Fig. 3 of the Japanese Patent Application No. 2010-091297, and the schematic diagram of the NA imaging is shown in Fig. 70. At this time, the incident energy with a lower LEA is the mirror reflection electron region, the LEALELEB is the transition region between the mirror reflection electron and the second discharge electron, and the LEB&lt;LE is the second discharge electron region. An example of LEA=0eV is given below. It. Since the energy distribution in the primary beam, for example, the LaB6 wafer, there will be about 2 eV, so when the incident energy LE = 0 eV, there are many electrons with negative LE. For example, in Fig. 70, CO and MC are sometimes found at the same time.

An example of this is shown in Figure 70. The rightmost graph shows the CO profile of the electron emission twice, because as the shift to the left increases the amount of electrons emitted by the mirror, and the amount of electrons emitted twice decreases, the CO becomes smaller, and the MC appears and gradually increases its intensity. When LE becomes negative, only MC exists. The size of this electron distribution is sometimes the distribution of the angle of the electrons representing the mirror reflection electrons or the secondary electron emission.

In the case of mirror-reflected electrons, sometimes one beam is parabolically drawn according to the potential of the sample surface and the orbit is changed in the opposite direction, and MC is formed on the NA surface. That is, the position of the MC can be controlled according to the incident angle of the beam.

On the other hand, the two electrons emitted when the sample is touched are different from the primary beam (electron) and are not affected by the incident angle.

The method of changing only the incident angle without changing the irradiation position can be realized by using two-stage deflectors (BA1, 2) placed on the primary cylinder (Fig. 71).

The data of the voltage and incident angle of the 2-stage deflector in the X direction or the Y direction is shown in Fig. 72.

In order to increase the resolution and to reduce the NA diameter, the MC position becomes important for CO.

In order to obtain a high resolution and arrange the NA at the CO center, in the case where the primary beam is normally incident, the MC mainly forms an image by NA (Fig. 73). MC can't get a good picture because there is no information on the sample surface. image.

Therefore, it is attempted to obtain a high-resolution image by changing the incident angle of the primary beam and shifting the MC to the top, bottom, left, and right.

When the L&amp;S pattern is observed, in the case of the vertical arrangement, the convex (line) portion of the horizontal pattern is likely to emit light due to the influence of the light of the MC, and the electron density of the convex (line) portion of the vertical pattern tends to be high in the case of the left and right arrangement. . Moreover, the contrast between the vertical pattern and the horizontal pattern is easily uneven. When the beam incident angle θ is shifted by the same angle at both X and Y, and the incident direction is 45°, the MC position is also in the 45° direction in the NA plane, and the vertical and horizontal patterns are all the same contrast. Good image.

In the example of Fig. 74, the incident direction is set to 0° in the X+ direction to define the angle in a counterclockwise rotation manner.

In Fig. 72, the voltage is set to a deflector voltage which is a desired incident angle. Figure 74 is a schematic diagram showing the schematic of NA imaging and the orbital of mirror-reflected electrons in orbit A.

Sometimes, when the MC is close to NA, the SN of the pattern will increase, but the contrast will deteriorate. Conversely, when it is far away from the NA, the SN will deteriorate, but the contrast will decrease. It has been experimentally confirmed that the MC has an optimum positional relationship in the distance of 2 to 3 times the NA diameter from the center.

The MC position adjustment method can be implemented simply and surely while observing the NA imaging.

Further, it has been experimentally confirmed that when the particles on the sample are observed, a higher sensitivity can be obtained in the vertical direction (X direction) than in the left-right direction (X direction).

In addition, when the MC is placed on the left or right (X direction), the left and right sides of the image become sharper, and when scanning in the Y direction, one of the left and right sides becomes more vivid. At this time, it is preferable to arrange the MC 2 to 3 times the NA diameter from the CO center.

For the advantages of using EB-CCD or EB-TDI in these observations and inspections, reference is made to paragraphs [0403] and [0404] of International Publication WO2009/125603. Further, the present embodiment can also be applied to the above-described Embodiments 1 to 17.

(Example 1)

[Step 1]

Sample: Cu/SiO₂Wiring pattern (refer to Figure 75)

Acceleration voltage: -4005 [V]

Sample surface potential: -4002.6 [V]

LE=2.4[eV]

Current density: 1 [mA/cm²〕

2 times optical system NA imaging TL2-2; 5550 [V]

The detector used an EB-CCD camera to observe MC and CO under NA imaging conditions.

The values of the aligners BA1 and 2 are set so that the incident surface angle θ = 100 [mrad] is incident from the irradiation direction α = 45 [°] (Fig. 76).

[Step 2]

The magnification of the electron optical system is set to 29 nm (pixels)

The NA aperture is set to 30 to 100 um to observe the 25nmL&amp;S pattern on the wafer.

The BA1, 2, and NA positions are fine-tuned so that the L&amp;S (line and space) patterns are the same.

[Step 3]

The MC position was set to 45° on the sample surface, and the incident angle was varied in the range of 20 to 200 [mrad] to determine the contrast and S/N of the L&amp;S pattern. The S/N is defined by the result obtained by dividing the contrast by the standard deviation σ of the average darkness of the W face.

The relative position of the MC with respect to the NA diameter is shown in Fig. 77. From Fig. 77, an image with good contrast and good S/N around the relative position of 2.5 can be obtained.

The correlation between MC relative position and contrast, S/N is shown in Fig. 78.

[Step 4]

Platform speed: 1 to 20mm/s

Information transfer rate: 50 to 1000 MPPS

Using a TDI camera, the inspection operation on the wafer is performed while scanning, and as a result, defect detection equivalent to 25 nm is generated.

(Example 2)

[Step 1]

Sample: About φ 30nm φ W contact plug / SiO₂For the structure, please refer to Figure 7 of the application of Japanese Patent Application No. 2010-091297.

Accelerating voltage: -4005 [eV]

Sample surface potential: -4002 [eV]

LE=2.4[eV]

Current density: 1 [mA/cm²〕

Optical system NA imaging TL2-2; 5550 [V]

The mirror-reflected electrons (MC) and polycondensation (CO) of the NA imaging were confirmed by an EB-CCD camera, and the two-stage deflectors BA1 and 2 were arranged as follows (refer to Fig. 79).

The values of BA1 and 2 were adjusted, and the sample surface was incident from the 0 [°] Y direction at an incident angle of 100 [mrad] (refer to Fig. 80).

[Step 2]

The magnification of the electron optical system is set to 29 nm (pixels)

The NA aperture was set to 30 to 100 um, and the W plug on the wafer was observed, and the correlation between the contrast of the above [Step 3] and SN was measured. In this case, the best image can also be obtained in the vicinity of the relative position of 2.5.

[Step 3]

Platform speed: 1 to 20mm/s

Information transfer rate: 50 to 1000 MPPS

Using a TDI camera, the plug inspection on the wafer is performed while scanning and scanning, and as a result, a plug structure defect inspection of φ 30 nm can be achieved. The high contrast check can be further achieved by using the beam dose of FIG. 19 in the application of Japanese Patent Application No. 2010-091297 or the contrast inversion principle of FIG.

(Example 3)

[Step 1]

Sample: Particles on Si

The incident angle was adjusted under the same conditions as the [first step] of the second embodiment.

[Step 2]

The magnification of the electron optical system is set to 100 nm (pixels)

The NA aperture was set to 30 to 100 um, and the particles on the wafer were observed, and the correlation between the contrast of the above [Step 3] and SN was measured. In this case, the best image can also be obtained in the vicinity of the relative position of 2.5.

[Step 3]

Platform speed: 1 to 20mm/s

Information transfer rate: 50 to 1000 MPPS

Under this condition, the TDI camera is used to perform the inspection of the particles on the wafer while scanning and shooting, and as a result, the inspection of the φ 10 to 30 nm specification can be achieved.

(Embodiment 19)

The high voltage power supply used in the inspection apparatus and inspection method of the present invention is shown in Fig. 81.

Fig. 81 (a) and Fig. 81 (b) show an overlap type high voltage generating device and an application example. In the high voltage generating unit of Fig. 81(a), the high voltage generating unit receives an alternating current signal of 20 kHz with a high voltage insulating transformer so that different potentials can overlap. The output side of the insulating transformer is converted to DC by a rectifier circuit. The 20 kHz AC signal is ideal for sinusoidal or non-immediately rising rectangular waves in the sense of preventing transitional noise. The DC output is the energy source required to obtain a high voltage output and is converted to an AC signal by a converter. Here, as described above, it is preferable to use a sine wave or a rectangular wave that does not rise sharply in the sense of preventing the occurrence of transitional noise. After being converted into an AC signal, it is converted into a number kV by a step-up transformer, and converted into a DC voltage that is boosted to a target voltage by a boost rectifier circuit. Boost rectifier circuit It is generally a Cockcroft Walton type circuit. The boosted DC high voltage output is divided by voltage, and compared with the voltage command value in the block of the voltage control function, and the feedback value feedback control is minimized, and the output voltage is kept constant. The voltage command value and the monitor output are photoelectrically converted and input or output through an optical fiber. The method of photoelectric conversion is converted into a frequency proportional to the voltage value, processed into a binary signal of light, and is reversely converted on the receiving side to return to the voltage value, which has been used in the past. However, this method belongs to the generalized A/D and D/A conversions, so other known means can be applied.

In the case where the high voltage power supply and the cylinder are separated, the connection requires a high voltage connector, a high voltage cable, and a high voltage vacuum introduction terminal. Since all of these elements need to have withstand voltage in the atmosphere, it is necessary to increase the distance from the surface to be large in proportion to the use voltage. Therefore, it becomes a huge obstacle to the miniaturization of the device. Incidentally, a conventional commercially available connector requires a length of 200 mm and a diameter of about 50 mm in terms of a withstand voltage of 30 kV. On the other hand, at 1×10^-4In the vacuum pressure below Pa, the density of the gas molecules as the medium is very thin, and the withstand voltage of the space is remarkably improved as compared with the atmosphere. In the present invention, this advantage is utilized in that a block which generates a high voltage is disposed in a space adjacent to the optical system, and a high voltage is generated and supplied in a high vacuum. Based on the fact that the optical system requires higher vacuum and extremely suspicion of dirt adhesion, the optical system and the high voltage generating portion are separated by the partition wall, the high voltage is supplied through the partition wall, and the exhaust system is exhausted by the respective exhaust system. . For the high voltage generating unit, the low voltage AC or DC voltage for high voltage generation and the voltage control are supplied from the atmosphere side. Although the optical fiber or the signal cable used in the communication system needs to be connected, the potential is remarkably lowered, and it is about several tens of V or less. Therefore, it is possible to reduce the size of the device without any problem by using a small vacuum introduction terminal. Further, the present embodiment can also be applied to the above-described first to eighth embodiments.

(Embodiment 20)

-EO correction -

An EO correction example of the inspection apparatus and the inspection method of the present invention will be described.

(summary)

Although a wafer is used as an example of a sample, it is not limited to a wafer, and an exposure mask, an EUV mask, a mask for a nano-imprint, and a template can be used in the same manner.

TDI shooting with the beam from the wafer, although the position of the wafer must be positioned correctly, but is in fact the wafer on the XY stage, the in mechanical positioning system embodiment, it is actually the number of precision is to 100μm The number is 10 nm, and the response speed is from several seconds to several ms.

Furthermore, since the design rule is refined toward a number of 10 nm, it is only dependent on the above-mentioned mechanical positioning to take a picture, and the response time and the level of the positioning precision and the design rule and the level of the shooting precision are too different, so that the correctness is obtained. For example, it has become a significant obstacle.

The program of shooting is implemented by a combination of steps (x-axis) and constant-speed scanning (y-axis), while comparing dynamic control (y-axis), the control residual is usually large, if the image is blurred. The meaning of the considerations requires a higher degree of control.

In view of these requirements, of course, there is a high-precision and responsive XY platform, and in order to achieve the precision and speed requirements of the beam control of the imaging unit, the functions of the EO correction are also innovated. It cannot be achieved by the platform.

Basically, the wafer position on the platform is delayed by a few microseconds (Sub-nm order) by a laser interference meter system and a special quartz mirror (Bar mirror) placed on the xy axis. The position is correctly identified, and the mechanical actuator is driven by the automatic control loop, and the time delay and the residual side are located at the target position. The control residual using the control positioning result can be obtained by the difference between the target position generated inside the control device and the current position obtained by the laser interference meter system. On the other hand, after passing through a plurality of electrodes, the beam is guided to the imaging device via the correction biasing electrode. The correction deflecting electrode has a sensitivity that can be biased to a distance on the wafer of about several tens of μm, and by applying a voltage thereto, the beam can be deflected to a two-dimensional arbitrary position. The control residual is converted into a voltage by a D/A converter after being calculated by the arithmetic unit, and is applied to the correction deflecting electrode in the direction of offsetting the residual. With the above configuration, it is possible to perform correction of the resolution close to the laser interference meter.

In other ways, an x-axis (step direction) is also developed by the above means, and the Y-axis (scanning direction) is used as a way of synchronously transmitting the TDI transmission clock of the imaging element and the moving speed of the platform. .

Figure 82 shows the concept of EO correction. An indication of the output from 1 to the target position is transmitted to the control feedback loops 2, 4, 5 containing the mechanical actuators. This part is equivalent to the platform. Driven to produce a knot after positional change If the feedback is implemented by the position detector of 5, the positional change of the drive system will continue to converge to the target position of the position indication, but the gain of the control system is limited, so residuals will occur. The position output system using the current position 6 (here, a laser interference meter) is detected in the next nm level, and the difference detector is used to detect the difference between the position indicating device of 1 and the high voltage high speed amplifier using 7 At the biasing electrodes of 9, 10, a voltage is applied in the direction of the offset residual. If there is no such function, the function of reducing the amount of fluctuation generated as shown in 13 to 14 is obtained.

Figure 83 shows the specific device configuration.

The XY stage system uses the X-axis drive servo motor 2 and the 1-encoder to perform X-axis drive, approximate position, and speed detection to achieve smooth servo characteristics. In this example, a servo motor is used, but the same configuration may be employed for an actuator such as a linear motor or an ultrasonic motor. 20 is a power amplifier that drives the motor. The precise position information of the X-axis can be realized by a combination of a 7 mirror, an 11-interference meter, a 12-receiver, a 13-laser light source, and a 14-interference meter board to realize a position detection function having a second nm resolution.

The Y-axis is also the same function of the linear motion, but consists of 10, 22, 4, 5, and 6.

The 19X-Y platform controller controls these devices through integration to enable the platform to perform two-dimensional operations, achieving precision from 100μ to 10nm and response speeds from seconds to ms. On the other hand, the X reference value and the Y reference value are output from 19 to the 23EO corrector, and the position information output from the 32-bit binary form of 14 is corrected by the 23 EO buffer board (Buffer board). Receive the current location. Inside After the calculus, voltage amplification is applied by the high-voltage high-voltage amplifiers of 25 and 24, and then applied to the deflection electrodes composed of 28, 29, 30, and 31, and the deflection is corrected to correct the residual difference, and the positional offset has been corrected to a minimum. The image information electron beam is led to 26TDI (shooting element). 27 will be described later, which is the portion that generates the timing signal for determining the transmission speed of 26.

Secondly, the function of the target position of the scanning direction of the device is stated.

The EO correction is a function of determining the difference between the target position and the actual position to offset the electron beam and correcting the position, and the correction range is limited to a range of several tens of μm.

This is determined by the sensitivity of the electrode, the dynamic range of the high-voltage high-speed amplifier, the noise level, and the number of bits of the D/A converter. However, the actual position of the platform at the time of scanning, compared with the stop when the gain of the control loop is limited, causes a large offset with respect to the target position. When moving at 20mm/s, the deviation from the target position is about 400μm. Even if the difference is directly calculated and output, the system will be saturated by greatly exceeding the correction range.

In order to prevent this phenomenon, in the present apparatus, the following means are employed to avoid this problem.

This concept is shown in Figure 84. 1 is the target position of the platform, and the movement is constant speed when scanning, so the linearity increases with time. On the other hand, the platform mechanical position 3 of the actual control result includes a mechanical vibration of several micrometers, and a fixed deviation of about 400 μm. As a means for removing the fixed deviation, it is possible to consider the use of the filter, and the position at the time of actual movement The smoothing of the signal, but at this time due to the filter time constant, there will be a delay, if the time constant that can ignore the left and right of the chopping, the measurement start range will be greatly limited, which has the disadvantage of a large increase in the overall measurement time. . Therefore, in the present case, in order to detect the fixed deviation, at least the difference between the current position and the target position at the time point of the previous scan is accumulated at least 2 to the 16th power, and the value is divided by the number of samples to obtain the target. The average value of the fixed deviation of the position and the current position is 5, and the operation is performed at this scan, and 5 is subtracted from the target position 4 as the compositing target position 6, and the fifth embodiment shown in FIG. 5 can be performed in the dynamic range. The composition of the EO correction.

Figure 86 shows a block diagram. The target value 1 system and the current position 2 are subtracted, and the previous cumulative calculation is performed in the block of 3 during scanning. On the other hand, the average value of the fixed deviation obtained in the same manner as the previous time is output from 4 to 3. The subtraction unit 5 subtracts 4 from 1 as the compositing target position 6, and subtracts the value from the current position 7 from the jammer to realize the EO correction data 8 with no response delay or chopping.

Fig. 87 is a diagram showing the construction of the block differential averaging detection regarding 3 of Fig. 86. The cumulative calculation is performed at 3 and 4, and the value of the information selector of 5 is selected by the value of the tired counter of 6 to perform equal division, and the output of the fixed deviation average value is realized.

Figure 88 depicts an example of a TDI transmission clock. The TDI system connects the photoelectric elements in multiple stages in the scanning direction, and transmits the charges of the respective imaging elements to subsequent elements for the purpose of improving the sensitivity and reducing the irregular noise. As shown in Figure 88, the subject on the platform and TDI It is very important that the pixels form a one-to-one correspondence. When this relationship collapses, the image will be blurred. In the synchronization relationship, it is 1-1, 1-2, 2-1, and 2-2, and when it deviates from synchronization, it is 3-1, 3-2, 4-1, and 4-2. Since the transmission system of the TDI and the pulse from the outside are transmitted to the next segment in synchronization, the platform moves 1 pixel, and if the transmission pulse is generated, the purpose can be achieved.

However, the positional information output of the mainstream laser jammer is now in the form of a 32-bit binary output synchronized with the 10 MHz internal clock, so it cannot be directly realized. Further, when the resolution is tens of nanometers, the precision of the transmission pulse is also important, and high-speed and high-precision digital processing is required. The way this piece is designed is revealed in H-1.

Carry out the description of Fig. 89. The position information of the laser interference meter and the synchronization signal 10M are introduced into the circuit from 1. The 10th MCLOCK (clock) of 2 generates a synchronized 100MHz clock through the PLL of 4, supplies it to each circuit, and adopts a method of performing arithmetic processing according to each state of the synchronization signal 10. This time the location information is kept at 22, and the previous value is maintained at 24. The difference between the two is calculated at 26, and the position difference of each state of the sync signal 10 is output from 27. The difference value is loaded as a parallel value (Parallel serial converter) 14 and the difference is output as 16 as the number of series pulses in synchronization with the 100 MHz clock. 15 also has the same function, but is a combination of 11, 12, 13, 6, 7, and 8, and can be immediately operated in each state of the sync signal 10. As a result, the series pulse corresponding to the position difference of every 10M is output from the sum circuit of 16 to the counter 17. The resolution of the laser interference meter is set to 0.6 nm, and when 1 pixel is set to 48 nm, if the comparator 18 is preset at 80, the counter is equal to 1 pixel. The timing output pulse 19. By using this signal as a transmission pulse from outside the TDI, even if the platform speed changes, it can be synchronized with it, and it is possible to prevent blurring and shaking of the image.

Figure 90 shows a timing diagram.

1 is the interference meter coordinate (position) information, and the number is represented by the position. 2 is a 100 MHz synchronous signal made by PLL. Group A (Bank) is the operation sequence of the series converter (Fig. 89, 14) series, and Group B is the action sequence of Figure 89. After the latch timing 7 of the memory location information, the differential calculation sequence 8 is executed, and the value is loaded into the serial converter (Fig. 89, 14), and the time of the next 10M clock 3 is used to execute 4 Output. In the B group, the same operation is performed at the timing of the delay of 1 cycle of the 10M clock 3, and the generation of the clock 6 is smoothly achieved. Further, the present embodiment can also be applied to the above-described Embodiments 1 to 19.

(Embodiment 21)

An example of a tray used in the inspection apparatus and inspection method of the present invention will be described with reference to Fig. 91.

(constitution, structure)

[Bottom plate] made of aluminum or titanium. Electrostatic chuck adsorption surface (requires flatness). It has an RTD contact section and a mask support pin.

[Transportation unit] made of aluminum or titanium. The position is moved when using the atmosphere and the vacuum robot.

[Frame] Made of phosphor bronze or titanium. Cover the mask from the end to about 1.2mm.

[Apply pin] Phosphor bronze. Apply the RTD to the mask. Wiring Pull from the application portion of the base plate to the application pin.

(action conditions)

When a mask is placed in the tray, the frame mechanism is moved up and down using another drive mechanism.

(effect advantage)

When the mask unit is not used, the following effects can be obtained when the tray is used.

When the tray of the present invention is used, a portion of the back surface of the mask is supported without being in contact with the pattern forming portion. Therefore, it is possible to prevent the electrostatic chuck from being damaged by the mask and attaching foreign matter. That is, the mask is directly disposed on the electrostatic chuck mechanism, which may adversely affect the back surface of the mask or the adhesion of foreign matter. In particular, when EUV masks or nano-imprinted masks are required, high-precision setting conditions are required, so particle size must also be prevented from adhering to small particles. For example, the back surface of the EUV mask is prevented from being 30 to 50 nm, and since the exposure is poor due to the imprinting, the foreign matter must be prevented from 3 to 20 nm.

The tray of the present invention has a contact mechanism for performing voltage application on the mask surface at a portion of the inner peripheral portion of the frame portion. Therefore, the frame portion is set to have the same electric potential as the mask applied voltage, and the potential uniformity of the mask end portion can be achieved, and a stable voltage can be applied to the mask surface (the surface conductive film) through the arrangement of the frame portion. That is, a stable RTD voltage can be applied from the top of the mask.

The operation of the vacuum is eliminated by making the frame portion have the effect of the correction ring. That is, by the arrangement of the tray, it is possible to move in a vacuum, and the mask and its setting members are not moved, so that friction does not occur, and it is effective for preventing foreign matter from adhering. Further, this embodiment can also be applied to the above-described first to twenty first embodiments.

(Embodiment 22)

Method of applying a high voltage to a vacuum movable portion

The method of applying a high voltage to the vacuum movable portion used in the inspection apparatus and the inspection method of the present invention will be described with reference to Fig. 92.

- Apply -

The voltage applied to the sample on the imaged optical electron beam inspection device is -4000 [V] ± 5 [V], and the voltage is applied to the sample in the vacuum chamber by a voltage applied to the sample. In the example of Fig. 42, a high voltage power source of a voltage of -20 to -50 [kV] applied to the sample is applied to the movable portion. At this time, a high-voltage power supply is provided on the platform of the vacuum chamber (for example, on the x, y platform) to avoid using a high-voltage line as a movable cable, that is, applying a high voltage to the sample with an immovable cable wire. . Furthermore, by limiting the electrical lead-in line from the atmosphere side to the low-voltage signal line, the high voltage can be easily used. The wiring cable from the high-voltage power supply is very thick, and it can not be moved, which not only reduces the generation of particles, but also does not require a large feedthrough. The design can be made more efficient and the cost can be reduced.

Further, the high-voltage power source applied to the sample as described above is provided on the movable portion such as the platform, and the high-voltage wiring does not exist in the movable portion. Further, a form in which the low-voltage power source is superposed on the high voltage applied to the sample from the outside is formed, and the voltage value control of the high voltage can be performed here. Further, this embodiment can also be applied to the above-described first to twenty-first embodiments.

(Embodiment 23)

- Method for improving defect detection rate of defect inspection device -

The method of raising the defect detection rate in the defect inspection apparatus using the inspection apparatus and the inspection method of the present invention will be described with reference to Figs. 93 and 94. In each of the figures, (a) shows an electron micrograph and (b) shows a schematic view. This electron microscope photograph is obtained by photographing the beam shape at the NA position in the image projection mode. This is a photographing by forming a condition in which the image of the beam shape at the NA position is imaged on the detector surface by a lens located between the NA and the detector. Thereby, information such as the shape change center position of the beam can be obtained, the desired beam position (mirror reflected electrons) can be adjusted to a desired position, and the NA can be set to an optimum position.

The defect inspection device using the image projection method is a method for improving the defect detection sensitivity by changing the incident angle of the primary electron beam that is perpendicularly incident under the condition that images are acquired using both the electron emission and the mirror reflection electrons. .

In the above device, the method of changing the incident angle of the primary electron beam that is normally incident is changed so that the defect projected to the detector is larger than the actual defect size to improve the defect detection sensitivity.

The center position of the mirror, which is the center of the beam of the specular reflected electrons, is adjusted in the range of 100 to 800 μm toward the scanning direction side with respect to the center of the polycondensation when the NA image is obtained.

When the position of the center of the mirror and the center of the polycondensation are aligned, if the center of the polycondensation is adjusted in the direction away from the scanning direction, the high-luminance region emitted from the defect is continuously amplified.

In the case where no defect occurs, electrons are emitted twice, and in the case where the defective portion forms a mirror-reflecting electron, the mirror is reversed by performing the above adjustment. The electron emission area is enlarged, and the detection sensitivity of the minute defects is improved by the enlargement of the high-luminance area.

(Embodiment 24)

- The surface potential of the sample is uniformly and stably supplied -

An example in which the surface potential of the sample of the inspection apparatus and the inspection method of the present invention is uniformly and stably supplied is described with reference to Fig. 95, Fig. 96A, and Fig. 96B.

In the defect inspection device using the image projection method, it is necessary to apply a voltage to the surface of the sample.

The appearance of the apparent defect of the surface state of the sample is adjusted by changing the voltage supplied to the surface of the sample.

That is, if the voltage distribution on the surface of the sample is different, the condition changes due to the difference in voltage distribution, and a problem of reproducibility is formed.

Therefore, in order to make the voltage distribution on the surface of the sample uniform, there is a proposal for an application method.

Currently, a contact portion is provided on the surface of the mask. By connecting the output from the high voltage power source to the portion, a high voltage is applied to the surface of the sample to expand the area of the contact sample.

The portion to which the sample application electrode is attached is referred to as a frame portion, and the sample can be carried in and out by moving up and down.

When the frame portion is lowered, the sample application electrode contacts the surface of the sample to uniformly supply a voltage to the sample (see Fig. 96A).

Furthermore, if other frame structures are used, it is effective for uniform and stable application. An example of this is shown in Figure 96B. Reference is made to Figure 96B. The frame bottom view (FIG. 96B (b)) and the top view (FIG. 96B (c)) are shown, which is a frame structure of a smooth finished surface having no protrusions thereon, for example, 195×195 mm Made of titanium or phosphor bronze, it has a 146 × 146mm hole inside. Further, as shown on the back side, there are projections at three locations. The protrusion height of the protrusion is about 10 to 200 μm. The front end of the protrusion may also be pointed. A voltage of a predetermined value is applied to the surface layer of the mask by the frame portion (cover). In the present invention, the mask is provided on the tray, and the tray has a mask support pin on which an exposure mask such as an EUV mask is provided. The mask supports parts with a small degree of particle generation. The material to be used is a metal member coated with a resin such as polyimide, Teflon (registered trademark) or fluororesin or a member itself is a resin. The masking position of the support pin can make contact on the outer side of the mask 142 x 142 mm. If it is placed inside, when a mask is provided in a unit such as an exposure device, if foreign matter or particles adhere to each other, the mask may be affected by tilting. Therefore, the area is set to prevent foreign matter or particles from adhering. In addition, the support pin may be fixed to the corner formed by the side and the bottom of the mask. In this case, the contact portion has a surface structure that is inclined at a predetermined angle. Further, in order to prevent the position of the mask from changing when the platform is moved, a cover fixing guide pin for fixing the position may be provided to fix the contact state.

It is assumed that the EUV mask is set in this way. Usually, the EUV mask has an insulating film on the outermost surface and a conductive film on the lower portion. Therefore, in order to apply a stable and uniform voltage application to the mask surface, it is necessary to damage the outermost insulating film to be applied to the conductive film. At this time, the frame portion (cover) having the projection shown in Fig. 96B has its effect. The frame portion has a predetermined voltage to be applied to the surface of the mask. Moreover, as shown in Fig. 96A, the frame portion is above the mask. Settings. At this time, the protruding portion can pierce the insulating film to reach the lower conductive film to perform stable voltage application. When the projection is provided, since the portion becomes an application portion to the mask, the application portion is defined, that is, the application portion can be controlled. Moreover, since the three-point contact is formed, the advantage that the parallelism of the upper surface of the mask and the frame portion can be set with good precision can be obtained. Since the frame portion is tilted at the time of setting at 2 o'clock, when it is set at 4 or more points, it is difficult to define which one of the protrusions breaks the insulating film to apply a voltage to the conductive film. Further, similarly, when there is no protrusion, it is difficult to define which part is in contact with the mask. Therefore, each time the mask is replaced, there is a possibility of forming different contact states. At this time, since the thickness of the insulating film of the EUV mask is usually 10 to 20 nm, the weight of the frame portion suitable for breaking the insulating film may be set in advance.

Furthermore, when the frame portion is in contact, it is necessary to reduce the step difference between the mask surface and the frame portion. This is because the electric field distribution is uneven due to the step difference. When the end portion of the mask, that is, the portion close to the frame portion, is inspected, the electronic track is shifted due to the unevenness of the electric field distribution, and the center position of the coordinate and the electronic image is shifted. Therefore, the step difference between the frame portion and the mask surface must be minimized. In the present invention, a structure in which the inhibition is 10 to 200 μm is formed. It is preferable that the step difference of 10 to 100 μm is preferable. Further, it is also possible to adopt a method of thinning the thickness of the sheet in the vicinity of the mask contact surface of the frame portion. Further, this embodiment can also be applied to the above-described first to twenty-third embodiments.

(Embodiment 25)

- Formation of photoelectrons using light/laser illumination -

Using light/laser irradiation in the inspection apparatus and inspection method of the present invention The technique for forming a photoelectron image is described below. The sample to be used has samples having irregularities as shown in Fig. 53 (a), (b), and 57. These samples include exposure masks, EUV masks, masks and samples for nano-imprinting, and semiconductor wafers.

(Feed through + use of a triangular mirror to illuminate the sample)

The drawing as a reference is an embodiment in which the form of Figs. 26 to 32 can be applied. The description of the triangular mirror in vacuum and other mirror-lens in the atmosphere has been described above. In addition to the light source, there is also a form that is placed in a vacuum. When light with high energy such as UV, DUV, EUV, or X-ray is irradiated, if there is air, dust, dust, or the like in the path, light scattering or flicker occurs, and the irradiation system is increased in instability. In addition, it may also cause deterioration of the surface of an optical element such as a lens or a mirror. That is, the formation of an oxide film on the surface of the element, deterioration of the film quality at the light contact portion, and deterioration of the surface due to adhesion of dust and dust occur. SUMMARY OF THE INVENTION The present invention is a method and apparatus for substantially reducing these problems by placing an optical member in a vacuum. At this time, the optical system is placed in another chamber that is in contact with the main column, and after the optical axis adjustment is performed in the atmosphere, vacuum evacuation of the optical system is performed. If only the optical system is located in another room, atmospheric open/vacuum exhaust can be easily performed. When the electron optical system and the cylinder are vacuum-conducted, the main column and the main chamber must also be opened together in the atmosphere, and the unrelated parts cannot be vacuum-exhausted and the atmosphere is open. These jobs can result in significant cost and time lost.

(Adjusting the optical axis of the atmospheric side mirror lens system and adjusting the irradiation specification using the lens position)

In the mirror-lens system outside the main cylinder, it can be carried out in the main cylinder The adjustment of the optical axis and the position of the light irradiation position. Before the light is introduced into the main cylinder, the optical axes are adjusted by two or more mirrors, that is, the shifting and tilting are performed. The adjustment is made so that the center of the illumination light comes to the irradiation position of the sample. At this time, adjustment of the angle with respect to the vertical direction of the sample surface (that is, the incident angle) can also be performed.

In addition, the illumination light specification can be adjusted by changing the lens position toward the optical axis direction. For example, if the focal length is f300 mm, the distance from the lens position to the sample surface position can be set to be shorter or longer than f, and the specification of the illumination light can be adjusted from φ 5 to φ 1000 μm, which is a thunder. It shoots about 1/200 to ×1 times the light path.

(The light output adjustment corresponding to the 2nd system magnification is performed by the attenuator, and the optical system adjustment is performed twice)

In the adjustment of the optical axis of the main cylinder, that is, in the adjustment of the axis of the photoelectron, when the light is irradiated to use the photoelectrons, the amount of irradiation of the light must be controlled in accordance with the optical axis magnification of the main cylinder. This is because when the density (amount) of the same light is used, in the case of a low magnification, the amount of electrons reaching the detector increases, and in the case of a high magnification, the amount of electrons decreases. When the high-density optical density irradiation is performed, in the case of low magnification, the amount of detected electrons becomes excessive and the detector is saturated. Therefore, in order to change the density of the illumination light, an attenuator or a beam splitter may be provided in the optical path outside the main cylinder to adjust the power of the illumination light. The light source itself, in order to stabilize the action, does not make the output change better. Moreover, it is particularly effective when the adjustment is performed without changing the irradiation specifications. Furthermore, in order to change the density of the illumination light, the illumination specification of the light can be changed. At low magnification, the light irradiation density is reduced to adjust, at high magnification, Then, the specification of the illumination is reduced to increase the optical density to perform image adjustment. In addition, the adjustment of the introduced light output and the specification control of the illumination light can also be used for image adjustment.

(λ&lt;264nm: Ru white signal, TaBO black signal, λ&gt;264nm: Ru black signal, TaBO white signal)

The work function WF of the material is the intrinsic value of the light emitted by the material when the material is irradiated with energy greater than its value. The wavelength of light corresponding to the WF is set to λ WF . Among them, there is a method of selecting a wavelength in which the energy is higher than the WF of the sample, a method of generating a photoelectron amount by a large amount, and a method of photographing a state in which the amount of photoelectrons is smaller than a wavelength at which the energy is lower than Wf.

Further, in the sample in which a plurality of materials are mixed, the present invention can use a photographing method having the following characteristics. At this time, for example, when there are two kinds of materials 1 and 2, the work functions are WF1, λ WF1, WF2, and λ WF2, respectively. The wavelength of the illumination light is set to λ.

A: λ&lt;λ WF1&lt;λ WF2

On material 1 and material 2, it can be observed that the higher the efficiency of photoelectron emission is the white signal, and the lower one is the contrast of the black signal.

B: λ WF1 &lt; λ &lt; λ WF2

In the case of the material 1, since the light energy &lt;WF1, the photoelectron from the material 1 is greatly reduced. Therefore, it can be observed that the material 1 is black and the material 2 is white (the number of electrons is relatively large is called white signal, and the lesser is called black signal).

C: λ WF1 &lt; λ WF2 &lt; λ

Since the light energy is &lt;WF1 and WF2, the amount of photoelectrons in both the material 1 and the material 2 is greatly reduced. Therefore, both sides are difficult to distinguish between white and black. situation.

In this way, by selecting the wavelength of the light, the energy function of the material with a large amount of energy is high light or laser, and the contrast due to the difference in the amount of photoelectrons can be obtained, and the energy width of the photoelectron is compared with that of the second electron. Very small (for example, 1/5 to 1/20), so the aberration is small and high resolution can be achieved.

In the case of the above A, if the material 1 selects a material having a high photoelectron emission efficiency, and the material 2 selects a material having a low photoelectron emission efficiency, it can be photographed with high contrast (contrast 0.5 to 1.0).

In the case of the above B, the material 1 will provide a lower wavelength than the λ WF1, so the photoelectron amount of the material 1 is lower, and the material 2 irradiates the wavelength light (or laser) higher than the WF2, so the material 2 will A large amount of photoelectrons are generated. Therefore, the high-contrast photographing pattern in which the material 2 is a white signal and the material 1 is a black signal can be relatively.

In the case of the above C, the material 1 is also irradiated with a wavelength lower than the work function of the material 2, so that the photoelectron is in a state of being small. At this time, if there is a defect such as a foreign matter having a low work function, the defect generates photoelectrons and can be detected.

In this way, for a sample made of a variety of materials such as a mask, an EUV mask, a semiconductor wafer, a nano-imprinted mask, etc., the wavelength of the irradiated light or the laser is selected for the work function of each material (the energy of the laser is selected) ), it is very efficient and effective to select a wavelength at which a large amount of photoelectrons will occur. High-resolution image capture and high-sensitivity foreign object shooting are possible. At this time, in the case of two kinds of materials, it is effective to use a method in which the amount of photoelectrons is relatively large/small. Furthermore, if it is three or more types, if only one is formed A state in which a large amount of photoelectrons is effective is effective. At this time, in the case of a sample such as the above-mentioned mask, EUV mask, semiconductor wafer, or nano-imprinted mask, there are many cases in which a concavo-convex structure is formed. This is especially effective. These samples have only the top layer of the top layer to form a large amount of photoelectrons, which can be used for photographing or defect inspection. At this time, the shape of the pattern can be clearly shot, so it is effective for shooting and checking the pattern defects. Further, when the condition that the amount of photoelectrons of the material of the bottom surface of the concave portion is large is selected, it is very effective to observe or inspect the defect of the concave portion. In this case, it is particularly effective to further make pattern specifications such as line/space, half-pitch (hp) specifications, or hole shape specifications smaller than the wavelength of light. For example, in the case of a photodetection device, the scattered light of the irradiated light is used to observe or inspect defects such as patterns, and the resolution is greatly reduced due to the wavelength limit, and the pattern is resolved by the scattered light. The degree is greatly reduced, making it difficult to observe or inspect the pattern. On the other hand, in the aspect in which photoelectrons are used in the present invention, even when the pattern size is shorter than the wavelength, the pattern can be observed or inspected with high resolution. The reason is as follows: 1) the wavelength is selected for the work function of the material to improve the contrast; and 2) the photoelectron can be generated by the formation of the near field even in the case of a pattern having a smaller wavelength. A pattern-shaped photoelectron is generated, and high contrast and high resolution can be achieved. For example, if a state in which only a large amount of photoelectrons is formed in the top layer, a pattern in which the top portion (that is, the line portion (convex portion)) is white can be formed. Further, in order to form a condition in which only a large amount of photoelectrons are formed on the bottom surface of the concave portion, the concave portion (space portion) can be observed or inspected with a white signal and a high resolution. To form these conditions, the above methods A, B, and C can be selected. In addition, if there are more than three materials at this time, for example When there are a top layer, a concave portion, or a wall portion, the same condition can be formed only for the top layer to have a large amount of photoelectrons. Only the concave portion (space portion) is in a state in which the amount of photoelectrons is large, and only the wall portion is in a state in which the amount of photoelectrons is large.

In addition, it is more effective when the shooting and inspection of these states are used in combination. For example, when only the top layer (convex portion) has a large amount of photoelectrons and only the recessed (space) photoelectrons are observed or inspected, and when the defects are extracted from the results of both, the pattern shape defects can be detected and located in the concave portion. Pattern defects such as minute foreign matter defects or film abnormalities can be observed or inspected without any omission. Further, it is possible to simultaneously form a state in which the above-mentioned top photoelectrons are large and a state in which the photoelectrons in the concave portion are large to perform observation or inspection. At this time, since it becomes a combined image, the resolution of the pattern is deteriorated, and the state of gray (the hue of the intermediate color of white and black) is exhibited. Moreover, the defective portion is largely deviated from the hue and becomes white or blackened, so that it can only be prepared for detecting foreign matter or pattern defects. In particular, the foreign matter defect on the pattern is easy to detect if it is a foreign matter that falls on the top layer, but it is very effective for the minute foreign matter in the concave portion of the concave-convex pattern. Because only this tiny foreign object will turn white or black, it can be detected with high contrast. When the materials are different, the effect of the photoelectron is different due to the difference in work function, and if it is a homogenous material, the potential distribution in the concave portion changes due to the tiny foreign matter, and the orbit of the photoelectron changes, and the photoelectron arrives. When there are more, the relative white signal is observed, and when it is less, it is observed with the black signal. In addition, when the amount of photoelectrons of the minute foreign matter is small, it is observed by a black signal, and when it is large, it is observed with a white signal. Also, these characteristics can be used for inspection.

For example, in the case of an EUV mask, the top layer is TaBO and the recess is Ru. The construction of the situation is often used. Therefore, Ru has a work function of 4.7 eV and a comparable wavelength of 264 nm.

At this time, by selecting or combining the wavelength λ of the illumination light or the laser according to the conditions of the above A, B, and C, it is possible to select a case where the top layer is observed or inspected by the white signal and the concave portion by the white signal.

It is possible to observe or inspect, for example, in a state where λ &lt; 264 nm and Ru of the concave portion have a large number of optical signals (that is, white signals). Moreover, in the case of λ&gt;264 nm, the top layer of TaBO can be observed and inspected by white signals.

At this time, in the irradiation of light or (and) laser light, by irradiating from a plurality of directions, an image having an objectivity can be formed on a line/space pattern having vertical, horizontal, and oblique directions. At this time, irradiation or the like can be performed from four directions or eight directions. Moreover, it is very effective when the T/M direction is selected or combined to perform light or laser irradiation.

This is because the distribution of the near-field (the energy field transmitted in the pattern specification area smaller than the wavelength) formed on the pattern surface is greatly different depending on the direction of the pattern, the direction of the light or the direction of the laser light, and the direction of the polarization. You can choose to use conditions that give you high contrast.

That is to say, it is also possible to select a condition of a strong near field at the top layer or a condition of obtaining a strong near field in the recess, and observe or inspect.

At this time, if a single-wavelength laser is used, there is a case where it is affected by interference or diffraction. Because of these effects, there are cases where an unnecessary white/black pattern is formed. In order to avoid this effect, the wavelength width is increased. For example, ±1 to 2 nm. In addition, it is also effective to simultaneously irradiate lasers having different wavelengths.

Furthermore, regarding this problem, the direction of polarization is not set to a single direction. It is very effective when set in multiple directions. Can greatly reduce the impact of interference or diffraction. It is also effective to continuously change the direction of polarization while shooting. When the polarization direction is changed during the shooting time of one frame, the image completed in the exposure time is an integrated image, and an image having little influence of the diffraction interference can be obtained. Further, the present embodiment can also be applied to the above-described first to twenty-fourth embodiments.

- Inspection method device for scatter light generated by EUV light irradiation -

The same method and apparatus as described in Figures 1 through 25 of the device system can be used to perform an EUV mask inspection. In this example, the EUV light is irradiated to the sample. Then, the pattern shape formed by the light (or EUV light) reflected from the concave-convex pattern of the EUV mask is imaged by the mirror system and detected by a detector. Defect detection can be performed in subsequent signal processing procedures. The inspection can be a unit cell/unit cell, a grain/grain, a grain/database, and the like. The EUV exposure uses light having a wavelength of 13.5 nm. However, it is not limited thereto, and the same method and device system can be used for the wavelength of the soft X-ray region.

Since EUV has a short wavelength, it is highly lost when it is transmitted and irradiated in the atmosphere, and when the image is formed, and it is inefficient due to the formation of ions by atmospheric particles, which causes contamination or increase in noise. Therefore, light transmission-irradiation-image formation must be performed in the vacuum device system. At this time, the vacuum apparatus system of the present invention can be applied.

The EUV irradiation is the same as the embodiment described with reference to Figs. 8, 9, and 26 to 32. The difference from the embodiment set forth above is that the pattern using light or EUV light reflected or scattered from the EUV mask is formed using a primary optical system. Since the 2nd system is not mirror reflection The electrons are emitted twice or twice, so that the magnifying optical system of the EUV mirror is used instead of the electrostatic lens or the electromagnetic lens for electrons, and the optical system is used to image and enlarge the projection on the detector. The optical magnification is about 500 to 3000 times, and at this time, 3 to 4 Px/line can be achieved for, for example, the L/S pattern. The advantage of EUV light irradiation is that the concave portion of the concave-convex pattern has a high reflectance. Further, in the EUV light, the side wall portion or the convex portion becomes an absorbing layer, so that the reflected light intensity of the concave portion is high, and the reflected light intensity of the convex portion is low, so that high contrast and S/N can be achieved. At this time, an EB-TDI or EB-CCD can be used as the detector. Because of its high energy, EUV light has its sensitivity even if it is not incident on electrons. Furthermore, the adjustment of the optical system or image formation conditions can be performed by the EB-CCD.

In addition, at the time of inspection, a continuous inspection image can be formed using EB-TDI, and inspection can be performed at high speed. At this time, the platform and the EB-TDI system perform the interlocking operation, and the amount of light acquired can be increased by the cumulative calculation of the image to perform the inspection of the height S/N and the high-speed inspection. The EB-TDI and EB-CCD are provided with a sensor unit in a vacuum. However, it is not limited to the detectors, as long as it is a detector that can be disposed in a vacuum in the sensor portion, and is applicable to an element that can form a two-dimensional image. For example, scintillator + TDI, etc. Further, in the primary optical system which is an optical system for EUV reflected light, in order to utilize EUV reflection or photoelectron generation to reduce noise, TaBN or carbon film coating may be performed on the surface of the cylinder or the surface of the member other than the mirror.

Further, when the EUV mask has an oxide film on the top layer, in order to suppress charging of the surface of the EUV mask caused by EUV light irradiation, a control electrode may be provided in the vicinity of the sample of FIG. 192 to control the potential of the surface of the sample.

For example, when the EUV mask itself is grounded at GND, if EUV light is applied to the oxide film on the top layer, photoelectrons are emitted from the oxide film to positively charge the oxide film. In order to suppress this, a negative electric field is generated in the vicinity of the mask, that is, the mask is set to a positive potential. Thereby, the generated photoelectrons are returned to the oxide film to suppress charging. Further, the potential of the EUV mask surface is set to about 3 to 10 V, and the method of applying a negative potential in advance is also effective. Further, in the above-described first to twenty-fifth embodiments, the present embodiment can be applied similarly except for the secondary optical system.

(Method of adjusting the illumination optical axis by using a jig)

This method is characterized by using a light or a laser irradiation mechanism of the first aspect used for photographing or inspection, and a light or laser for adjusting the second aspect other than the mechanism. The light or laser system of the second aspect is used for adjustment. At this time, the light of the second embodiment is incorporated in a lens or a cylinder, and the irradiation position is fixed, and the irradiation position is measured in advance, and the irradiation light is located at the center of the substantially electro-optical system. Further, depending on the mechanical assembly precision of the introduction mechanism, the position is, for example, within ±100 μm. At this time, the specification of the light or the laser beam is larger than the set error. For example, the assembly precision is about +200 to 2000 μm. In the use example, it is φ 1 mm (may also be a circle, a circle of 1 × 1.5 mm, etc.). At this time, an introduction mechanism made of an optical fiber and a lens can be used. According to this method, the irradiation position can be determined according to the mechanical precision of the setting, the operation is easy, and the reproducibility of the installation position is good. When the light of the second aspect is introduced into the system for adjustment, it is very efficient.

Adjustment 1: Optical axis adjustment for electro-optical systems. Photoelectrons are emitted from the surface of the sample, and the photoelectrons are guided to the inspection through the I-order electron optical system. Detector. In this case, the light beam or the laser beam is irradiated, and the two-dimensional planar photoelectrons are emitted, and then guided to the detector through the electron optical system twice. The planar photoelectron forms a two-dimensional photoelectron image, and the image is magnified and projected onto the detector through a secondary optical system. At this time, the photoelectron image can be formed by light introduction of the second embodiment, and the optical axis center condition of the secondary optical system can be obtained. The optical axis condition and the center position of the objective lens (object position) are obtained in advance. Thereby, a mark is set at this position (a pattern having a characteristic, a Faraday cup, etc. can be used). This determines the center position of the objective lens. Next, the light or laser introduction system of the first aspect is adjusted so that the beam is irradiated to the position of the mark. In this case, the system is composed of a mirror, a lens, and a light source, and the irradiation position and specifications can be adjusted by transmitting two or more mirrors and lenses. In this case, since the objective lens center mark obtained in the first aspect is provided, it is only necessary to perform the light of the second aspect or the laser light. By having this mark, the optical axis of the first light or laser introduction system can be adjusted very efficiently. In the case where the second optical system is not provided, the optical axis adjustment must be performed in the first optical introduction system after the first optical system is assembled. Therefore, the position is initially found in a state where the objective lens center is unknown. That is, the condition of the visible photoelectron image is roughly formed, and the optical system is adjusted twice in this state, that is, the center position of the objective lens is obtained, the mark is set there, and then the optical axis adjustment of the first light introduction system is performed. . Therefore, it is necessary to obtain the objective optical axis from the general situation and the optical axis adjustment of the light introduction system of the first aspect, and perform the axial adjustment of the final two optical systems and the light of the first form while alternately performing the axial adjustment. Import the axis adjustment of the system. Therefore, through the light introduction system of the second aspect, the optical system is formed twice in advance. It is very efficient to complete the adjustment of the axis.

In addition, the axis adjustment of the first light introduction system can also be performed using a jig. Set the open-ended guide and set the power meter under it. The optical axis adjustment of the first light introduction system is performed to maximize the amount of holes passing through the guide. At this time, the coordinates of the position of the hole of the guide plate to the center of the objective lens are obtained in advance.

(light + EB illumination)

It is stated that there are implementations when there are two primary systems.

Image formation by a combination of two times of emitting electrons and/or mirror-reflecting electrons (in the case of mirror-reflected electrons and non-mirror-reflecting electrons) generated by photoelectron or laser irradiation The practice is also very effective. Here, the secondary electron emission refers to a part of a secondary electron, a reflected electron, and a backscattered electron, or a mixed state. Especially when it is low LE, it is difficult to distinguish these electrons.

In the present embodiment, the form of the light or laser irradiation sample is shown in Figs. 7 to 9 and Fig. 26 to Fig. 31, and the electron emission in the first system of Fig. 33 (a) to Fig. 42. The form of the beam irradiation sample is combined with the one. Examples of the embodiments are shown in Fig. 196, Fig. 197, and Fig. 198. An example of a case where the sample is a concave-convex shape will be described below.

This example is an example in which laser irradiation (or light) and electron beam irradiation are simultaneously performed as a primary beam. In the irradiation mode, it is possible to alternately irradiate at the same time or in time. The characteristics of the laser irradiation and the electron beam irradiation at this time are respectively stated, and the effects of the fusion are stated.

When laser irradiation is performed, the amount of photoelectrons in the top layer (convex portion) is large and white signals are emitted. When the electron beam is irradiated, the electrons are emitted more from the top layer twice. In the case of a white signal, the combination of the photoelectron image and the two-shot electronic image can increase the amount of electrons on the top layer (photoelectron white + 2 times electron emission and/or mirror reflection electron white), that is, the top layer (protrusion) can be formed. The white and concave parts are black and the contrast and S/N are increased.

On the contrary, when it is observed that there are many photoelectrons in the concave portion, the concave portion has a white signal, and in the case where the electrons in the concave portion of the electron emission are large, and the concave portion is in the white signal, if the laser irradiation and the electron beam irradiation are simultaneously performed (combination ), it is possible to increase the concave portion to white (photoelectron white + 2 times emitting electrons and/or mirror reflecting electron white), and the top layer (convex portion) is the contrast and S/N of the image formed by black. At this time, the so-called white signal means that the number of detected electrons is larger than that of other parts, and the brightness is relatively high, that is, the white signal can be taken.

As shown in Fig. 33(a), when the electron beam is used, since the separation of the second beam is performed (using the Wien filter condition for making the second beam straight forward), it is necessary to emit electrons such as E×B. Beam separator. Therefore, in the form in which the electron beam is combined with the laser or the beam, an electron beam splitter is also required. This example is shown in Fig. 196, Fig. 197, and Fig. 198.

The differences between the 196th, 197th and 198th are as follows. In the 196th and 197th drawings, there is a laser (or light) introduction mechanism on the sample side compared to E×B; and the 198th image has a laser on the detector side compared to E×B (or Light) introduction mechanism. For example, in the 196th and 197th drawings, the cathode lens is provided with a laser introducing hole, and the laser beam can be irradiated to the sample in a state where the outdoor portion is aligned and adjusted by a mirror or the like, or the optical fiber can be used. A lens or the like is introduced into the cathode lens to perform laser irradiation or the like. Moreover, Figure 198 is the same reflection as that described in Figure 26 in the cylinder of the secondary system. The mirror member can be introduced from the outside of the cylinder to illuminate the sample with a laser (or light). Fig. 198 shows a case where the amount of electrons of the convex portion is large (white signal) by laser irradiation and electron beam irradiation. Conversely, in the case where the amount of electrons in the concave portion is large (white signal), it can be performed in the same manner as in Fig. 196.

Further, it is more effective to use the electron beam described in the embodiment shown in Figs. 35 to 41 for the electron beam of the primary system. Since a large current can be used to illuminate an electron beam of a narrow range of energy, the energy range of the emitted electrons or the mirror-reflected electrons formed twice is narrow, and a high resolution image with low aberration and ambiguity can be realized. Further, since the range of the photoelectron energy by the laser irradiation is narrower than that of the secondary electron emission, even if fusion/combination is performed, the state of the narrow range of energy can be maintained, and the electron amount is increased but the energy width is not expanded. This advantage is very effective and useful when the irradiation laser or the electron beam is increased in order to increase the throughput without deteriorating the image quality.

In addition, conversely, it is also possible that the photoelectrons are white and the electrons are blacked out twice. In this case, the combined image is gray, that is, the intermediate color of white and black, and the resolution and contrast of the pattern are lowered. At this point, it can be observed that only the white signal of the defect becomes stronger or the black signal becomes stronger. At this time, for example, if it is a defect having high sensitivity to light irradiation, a white or black signal can be formed by increasing or decreasing the amount of photoelectrons. Further, in the case of a defect having high sensitivity to electron irradiation, a white or black signal can be formed by increasing or decreasing the amount of electrons emitted from the electrons twice.

Furthermore, it is also possible to use a combination in which the photoelectrons are black and the electrons are emitted twice. In the case of an EUV mask, the TaBO and the concave relative to the top layer The Ru of the department can be combined as follows.

A combination of a Ru white/TaBO black photoelectron image and a secondary electron emission and/or mirror reflection electron image, a Ru black/TaBO white photoelectron image and two electron emission and/or mirror reflection electrons. combination)

Thereby, high contrast and S/N can be achieved, and pattern defect inspection and foreign matter inspection with high sensitivity can be performed.

For the low LE image, the oxidation film potential is stabilized by light irradiation. This is very effective for a low LE image having an electron irradiation energy of -5 eV &lt; LE &lt; 10 eV, particularly when the material of the top layer is an oxide film. When the top layer is an oxide film, the oxide film generates a negative voltage by irradiation with a low LE electron beam. Because of its influence, image quality deterioration occurs, and the current density cannot be increased. At this time, light or laser irradiation such as UV, DUV, EUV, or X-ray can be performed to control the potential of the vaporized film. By illuminating these light, by generating photoelectrons, a positive charge can be generated. Therefore, the potential of the oxide film can be controlled to a certain extent by performing low-LE and simultaneous or intermittent exposure of these lights or lasers. Further, the potential of the oxide film is kept constant, and the image quality is stabilized, and even if the current density is increased, a stable image can be formed, and the throughput can be increased.

(Photoelectron cathode 1st system)

An example for the case where the reference voltage is not GND but a high voltage is disclosed in Fig. 37. In this example, the reference voltage is +40000V. In order to make the reference voltage uniformly form an electric field in the cylinder, a cylindrical tube is provided. This tube was set to tube 1. Then, 40,000 V was applied to form a reference voltage. Furthermore, the proximity to the photoelectron surface is parallel to the equipotential line (distribution) photocathode, and the lens is thus a magnetic field lens. At the same time, the aligner uses an electromagnetic pair Standard device. In addition, the NA and other apertures are at the reference potential and are arranged in a tube configuration. The tube 1 has a tube 2 on its outer side due to the application of a high voltage. This tube 2 is set to GND and can be connected as a device to form a GND connection. Tube 1 and tube 2 are insulated with a voltage-resistant insulator to maintain the necessary applied voltage. It is not mentioned here that the reference voltage of the primary system is controlled so that the reference voltage of the secondary optical system becomes a high voltage. Therefore, the secondary optical system is similar to the primary optical system in that the tubular body is formed into a two-layered column. The inner tube is applied with a high voltage and the outer tube is at GND. The voltage difference is kept the same as that of the primary system. Further, the tube 1 is a conductor, and the outer peripheral portion of the tube 1 may be coated with a resin material such as polyimide or epoxy resin. Further, the outer peripheral portion of the resin material is coated with a conductive material, and the coated conductive material may also be GND. Thereby, the inside of the resin material is a high voltage type reference voltage, and the outer side is GND, and other GND connection and GND components can be assembled. Moreover, the tube 2 on the outer side thereof may also be a conductor type shield tube. The tube 2 is a magnetic body of Permalloy or pure iron, which blocks an external magnetic field. Further, the present embodiment can also be applied to the above-described first to twenty-fifth embodiments and the embodiment in which the number is not assigned.

(Embodiment 26)

-EO correction -

The EO correction example used in the inspection apparatus and inspection method of the present invention will be described.

A. Summary

When shooting a beam from a wafer with TDI, the position of the wafer must be correctly positioned, but in reality the wafer is tied to the XY stage, mechanically In terms of positioning, the actual precision value is from several hundred μ to several 10 nm, and the response speed is from several seconds to several ms.

On the other hand, the design rule is being refined in the number of 10 nm. Therefore, it is necessary to inspect a wiring having a line width of 10 nm or a hole having a diameter of 10 nm, and to perform detection of a shape defect or an electrical defect of these line holes and a diameter of 10 nm. Detection of dirt. However, when relying on the above-mentioned mechanical positioning for shooting, the response time and the level of positioning precision will be too different from the design rules and the shooting precision level, causing significant obstacles in obtaining the correct image.

The order of shooting is performed by a combination of step (x-axis) and fixed-speed scanning (y-axis), while in dynamic control (y-axis), the control residuals are generally large, preventing blurring of the image. In the sense, it requires a higher degree of control.

In view of these considerations, it is a matter of course to have an XY stage with high precision and excellent responsiveness. Moreover, in order to achieve precise control and speed of the beam relative to the imaging unit, it is necessary to have an EO correction function, but This kind of function is not what the platform can do.

Basically, the wafer position on the platform is correctly identified by a laser interference system and a special quartz mirror placed on the xy axis in a sub-nm order with a delay of several microseconds. Position and drive the mechanical actuator with an automatic control loop, with the time delay and residual side being at the target position. The control residual using the control to perform the positioning can be obtained by the difference between the target position generated inside the control device and the current position obtained by the laser interference meter system. On the other hand, after passing through a plurality of electrodes, the beam is guided to the beat via the correction biasing electrode. Camera. The correction deflection electrode has sensitivity and can be converted to a distance on the wafer, and can be biased to about several hundred μm or less, preferably 100 μm or less, more preferably several tens of μm or less, and the beam can be biased to the second by applying a voltage thereto. Any position of the dimension. The control residual is converted into a voltage by a D/A converter after being calculated by the arithmetic unit, and is applied to the correction deflecting electrode in the direction of offsetting the residual. With the above configuration, it is possible to perform correction similar to the resolution of the laser interference meter.

In other ways, an x-axis (stepping direction) system has been developed, and the Y-axis (scanning direction) is used to synchronously transmit the moving speed of the TDI transmission block and the platform as the imaging element.

Figure 97 shows the concept of EO correction. The indication 951 to the target position is transmitted to the control feedback loop 952 containing the mechanical actuator. This part is equivalent to the platform. After being driven, the result of the position change is transmitted through the position detector 953, and the positional change of the drive system continues to converge to the target position from the position indication. However, due to the limited gain of the control system, the residual is generated. difference. The current position is detected by the position output system 954 (here, a laser interference meter) at the next nm level, and the difference between the position detector 951 and the position indicating device 95 is detected by the residual detector 955, and the high voltage is used. The high speed amplifier 95 is applied to the deflection electrode 957, and a voltage is applied in the direction of the offset residual. If there is no such function, it has a function of reducing the amount of fluctuation generated as in 958 to 959.

Figure 98 shows the specific device configuration. The XY stage 961 uses the X-axis drive servo motor 962 and the encoder 963 to drive the X-axis and the approximate position and speed to realize a smooth servo. Sex. In this example, a servo motor is used, but the same configuration can be employed for an actuator such as a linear motor or an ultrasonic motor. 966 is the power amplifier that drives the motor. The precise position information of the X-axis can be achieved with a combination of mirror 967, interference meter 968, receiver 969, laser source 9610, interference meter board 9611 with sub-nm resolution Position detection function.

The function of the Y-axis is similar to that of the orthogonal X-axis, and is composed of a servo motor 9612, an amplifier 9613, a mirror 9614, an interference meter 9615, and a receiver 9616. The XY stage controller 9617 can control these devices in a comprehensive manner so that the platform can perform two-dimensional operations, achieving a precision of 1000 μm to 1 nm, preferably a precision of 100 μm to 2 nm, and more preferably a precision of 1 μm to 2 nm. More preferably, the degree of precision is from 0.1 μm to 2 nm, and the response speed is 1000 ms or less, preferably 10 ms or less, more preferably several ms or less. On the other hand, the X reference value and the Y reference value are output from the XY stage controller 9617 to the EO corrector 9618, and the position information output from the interference meter 9611 in 32-bit binary form is transmitted via the cache. Plate 9619, the current position is received by the EO modifier 9618. After the internal calculation, the voltage is amplified by the high-voltage high-speed amplifiers 9620 and 9621, and then applied to the deflection electrode 9622, and the deflection is performed to correct the residual difference, and the image information electron beam having a very small positional shift is obtained. Boot to TDI (shooting component) 9623. 96.24 will be stated later as part of the timing signal used to generate the transmission speed for determining TDI 9623.

Secondly, the production of the target position of the scanning direction of the device is stated. can. The EO correction is to obtain the difference between the target position and the actual position, and to bias the electron beam to correct the position and offset the difference. The correction range is limited to the range of several tens of μm. This range is determined by the sensitivity of the electrode, the dynamic range of the high-voltage high-speed amplifier, the noise level, and the number of bits of the D/A converter. However, the actual position of the platform at the time of scanning is significantly offset from the target position as compared to when the control loop has a limited gain. When moving at 20mm/s, the deviation from the target position is about 400μm. Even if the difference is directly calculated and output, the system will be saturated by greatly exceeding the correction range. In order to prevent this, the device employs the following means to avoid this problem. Figure 99 illustrates this concept.

971 is the target position of the platform. Because it is moving at the same speed during scanning, it increases linearly with time. On the other hand, the platform mechanical position 972 of the actual control result includes mechanical vibration of several micrometers, and has a fixed deviation of 97.3 of about 400 μm. As a means for removing the fixed deviation, it is possible to use a filter to smooth the position information during actual movement. However, at this time, due to the filter time constant, a delay is always generated, and if it is made to have a chopping wave (it becomes miscellaneous) When the time constant of the degree of voltage fluctuation is large, the measurement start range is greatly limited, and there is a disadvantage that the overall measurement time is greatly increased. Therefore, in the present case, in order to detect the fixed deviation, at least the difference between the current position and the target position of the previous scanning time point is accumulated by at least about 16 to the power of 16, and the value is divided by the number of sampling times to obtain the target. The average value of the fixed deviation between the position and the current position is 97.84. In this scan, the calculation is performed, and the average value 97.74 is subtracted from the target position 975 as the composite target position 97.6, and the 98th figure is realized. 1 The composition shown in the EO correction can be performed in the dynamic range. In addition, the cumulative number may be obtained as long as the target precision is obtained, and is not limited to this value, and may be a smaller number of accumulated segments.

Figure 101 shows a block diagram. The target value of 99. 1 is subtracted from the current position of 99.2, and the previous cumulative calculation is performed in the box of 993 during scanning. On the other hand, the average value of the fixed deviation obtained in the same manner as the previous time is output from 994 to 99.83. Subtract 999.2 from 991 using the subtractor 995 as the composite target position 996. This value is subtracted from the current position of the interferometer, 997, to achieve EO correction information without response delay or chopping.

Figure 102 is a diagram showing the construction of the block differential averaging detection of 99.83 in Fig. 101. The cumulative calculation is performed at 1001 and 1002, and the character selector 1004 is selected by the value of the cumulative counter 1003, and the output of the fixed deviation average value is realized by performing the equalization.

Figure 103 illustrates the concept of the transmission clock for TDI. The TDI is a multi-segment connection in which the photoelectric elements are connected in the scanning direction, and the imaging element is designed to achieve an increase in sensitivity and a reduction in irregular noise by transmitting the charge of each imaging element to subsequent elements. As shown in Fig. 101, it is important that the object on the platform and the pixels on the TDI have a one-to-one correspondence. If the relationship is broken, blurring of the image will occur. The case when there is a synchronization relationship is shown in 1-1, 1-2, 2-1, and 2-2, and the case of deviating from the synchronization relationship is shown in 3-1, 3-2, 4-1, and 4-2. Since the transmission operation of the TDI is transmitted to the next stage in synchronization with the pulse from the outside, the purpose can be achieved if the transmission pulse can be generated when the platform moves by 1 pixel.

However, the positional information output form of the mainstream laser jammer now outputs a 32-bit binary output synchronized with 10 MHz of its own internal clock, so it cannot be directly implemented. Further, when the resolution is set to several tens of nm, the precision of the transmission pulse is also important, and high-speed and high-precision digital processing is required. The method developed by this article is shown in Figure 104. In the figure, the position information of the jammer and the 10 MHz sync signal are introduced into the circuit from the buffer 1021. The 10 MHz clock 1022 generates a synchronized 100 MHz clock through PLL 1023 and transmits it to each circuit. The calculation processing is executed every 10 states of the synchronization signal 1024, so that the current location information is maintained at 1025, and the previous value is maintained at 1026. The difference between the two is calculated at 1027, and the position difference for every 10 states is output from 1028. The difference value is loaded as a parallel value and connected to the serial converter 1029, and synchronized with the 100 MHz clock to output the difference as the number of series pulses from 10210. 10211 is also the same function, but in combination with 10212, 10213, it can be continuously operated in every 10 states without interruption. As a result, the series pulse corresponding to each 10 MHz and the position difference is output from the sum circuit 10210 to the counter 10214. When the resolution of the laser interference meter is set to 0.6 nm and the 1 pixel is set to 48 nm, the counter can output a pulse of 19 at a timing of approximately 1 pixel as long as the comparator 10215 is preset to 80. By using this signal as a transmission pulse from the outside of the TDI, even if the platform speed is changed, it can be synchronized with it, and the image can be prevented from being blurred or shaken.

Figure 105 shows the timing diagram. 1 is the interference meter coordinate (position) information, and the number shows the position example. 2 is 100MHZ synchronization made through PLL Signal. Among them, the A group is the action sequence of the parallel converter 1029, and the B group is the action sequence of the same 10211. After the latch timing 7 of the memory location information, the operation of the differential operation timing 8 is performed, and the value is loaded into the parallel converter 1029, and the output of 4 is performed using the cycle time of the next 10M clock 3. In the B group, the same operation was performed according to the timing of the 10M clock 3 delay of one cycle, and the pulse of 6 was smoothly generated. Further, the present embodiment can also be applied to the above-described first to twenty-sixth embodiments and the embodiment in which the number is not assigned.

(Embodiment 27)

- Foreign matter (particle) adhesion prevention method and electron beam inspection device -

The foreign matter adhesion preventing method of the inspection apparatus and the inspection method of the present invention will be described.

Embodiments of the present invention will be described in detail below with reference to the drawings (Figs. 106 and 107). In the following examples, a rectangular mask or a circular semiconductor crystal having a surface layer having a conductive film such as Si (including doping), Cr, TaN, TaBN, CrN, Ru, Ta, W, or Cu is used. The circle is used as a sample to prevent foreign matter such as particles from adhering to the surface of the sample. The outermost surface of the film may also have TaBO, TaO or SiO₂Insulation film. For example, a thin film is formed on a quartz or quartz substrate, and a circuit pattern film structure for LSI is formed on the Si wafer as a mask. In the following examples, the same or corresponding components are denoted by the same reference numerals, and the description thereof will not be repeated.

Fig. 106 is a longitudinal sectional front view showing an outline of an important part of an electron beam inspection apparatus according to an embodiment of the present invention, and Fig. 107 is a cross-sectional plan view taken on line 106. As shown in Figures 106 and 107, the electron beam inspection equipment The vacuum chamber 12 is provided with a vacuum evacuation chamber. The vacuum chamber 12 is provided with an XY stage 14 that is movable in the X direction and the Y direction. Above the XY stage 14, in the present example, a holding box 18 is provided by the electrostatic chuck 20, in which a sample 16 made of a rectangular mask is held.

The XY stage 14 has a stroke movement area of the effective distance of the sample 16 + the assist distance (checking the highest speed * speed stabilization time), and the sample (mask) 16 can perform shooting or defect inspection in the effective area. For example, when the effective distance of the sample 16 in the X direction and the Y direction is 300 mm and the assist distance is 100 mm/s × 0.5 s = 50 mm, the XY stage 14 has a stroke movement area of 400 mm.

A dust collecting electrode 22 which is continuously extended into a rectangular frame shape and has a rectangular cross section is disposed at a predetermined interval from the sample 16 provided on the XY stage 14 so as to surround the entire circumference of the sample 16. Further, a gap control plate 24 having a through hole 24a in the center thereof is disposed horizontally above the sample (mask) 16 and the dust collecting electrode 22 provided on the XY stage 14, and the gap control plate and the vacuum chamber 12 are disposed horizontally. There is a slight gap in the circumference. The optical system element 26 of the electron beam inspection apparatus is positioned in the through hole 24a, and the electron beam is irradiated onto the surface of the sample 16 provided on the XY stage 14 through the optical system element 26. The size of the through hole 24a is set to be slightly larger than the outer shape of the optical system element 26.

The dust collecting electrode 22 is made of a non-magnetic material such as phosphor bronze or Ti in order to eliminate electron beam bending or orbital change due to a magnetic field. The electron beam includes an electron beam irradiated once in the system, an electron beam emitted from the sample 16 twice, and a mirror reflected electron beam reflected in the vicinity of the sample 16.

The gap control plate 24 is made of a flat plate such as phosphor bronze, Ti or SUS material, and has a plate thickness of, for example, 0.3 to 5 mm. In order to stabilize the potential and prevent contamination, the gap control plate 24 may be coated with Au, Pt, Ru or Os. The size of the gap control plate 24 is set such that even if the XY stage 14 moves within its moving area, the dust collecting electrode 22 can be covered to an area not beyond the outer side of the gap control board 21. Thereby, when the XY stage 14 is moved to the position where the sample 16 provided on the XY stage 14 is also moved to the most biased position in the vacuum chamber 12, the electric field distribution is prevented from being broken and the orbit of the particles is changed, and the particles are prevented from flying to the test. Sample 16 was attached to the surface of sample 16. Further, the gap control panel 24 is not necessarily required. This is the same in the following examples.

In this example, as shown in Fig. 107, a dust collecting electrode 22 continuous in a rectangular frame shape is used, and the dust collecting electrode 22 integrally surrounds the entire circumference of the sample 16 provided on the XY stage 14, Thus, it is possible to prevent the electric field uneven portion generated by the gap generated from the position along the longitudinal direction of the dust collecting electrode 22, that is, the so-called electric field gap, from invading into the inside surrounded by the collecting electrode 22.

The dust collecting electrode 22 does not necessarily surround the entire circumference of the sample 16, as long as the electric field formed by the collecting electrode 22 can surround the periphery of the sample 16. For example, as shown in FIG. 108, the dust collecting electrode 22a extending linearly may be disposed to extend substantially along the entire length of each side of the sample 16, and the dust collecting electrode 22a may surround the substantially entire periphery of the sample 16. the way. Further, although not shown, the dust collecting electrodes extending in a straight line may be separated from each other in the middle of formation, and in this case, electric field distortion may occur between the dust collecting electrodes adjacent to each other, but as long as the collecting electrode can be obtained The required potential distribution is sufficient. E.g, When considering two dimensions, the width of the dust collecting electrode is D, and when the distance between the electrodes of the dust collecting electrode is L, there is no problem as long as it is D/L 4. This is the same in the following examples.

In the above example, a rectangular mask was used as the sample 16. When a circular semiconductor wafer is used as the sample 16a, as shown in Fig. 109, the sample (semiconductor wafer) 16a held by the circular holding case 18a is provided on the XY stage 14 and is passed around the sample 16a. The dust collecting electrode 22b which is continuously formed into a circular ring shape can integrally surround the entire circumference of the sample 16a. In this case, as shown in FIG. 110, one of the semicircular shapes may be disposed opposite to each other in a manner of forming a perfect circle, and a test in which the dust collecting electrode 22c is to be disposed on the XY stage 14 may be formed. A manner in which substantially the entire circumference of the sample (semiconductor wafer) 16a is surrounded. Further, although not illustrated, a method in which a plurality of dust collecting electrodes are extended in the circumferential direction and separated from each other may be formed.

Fig. 111 is an enlarged view showing the sample 16, the dust collecting electrode 22, and the gap control plate 24. As shown in Fig. 111, the sample 16 is connected to a first power source 28 for applying a predetermined voltage to the surface of the sample 16, and the dust collecting electrode 22 is connected to a second power source 30 for applying a predetermined voltage to the dust collecting electrode 22. The thickness of the dust collecting electrode 22 is, for example, 0.1 to 5 mm. Although the width W1 of the dust collecting electrode 22 is as wide as possible, the larger the width, the larger the volume of the dust collecting electrode 22 in the vacuum chamber 12 is, so it is generally 5 to 50 mm. The distance L1 between the sample 16 and the dust collecting electrode 22 is preferably used in a range from the relationship with the width W1 of the dust collecting electrode 22 to satisfy, for example, a relationship of 0.5 L1 &lt; W1 &lt; 5 L1.

In this example, the surface of the sample 16 is passed through the first power source 28. A voltage of, for example, -1 to -5 kV is applied, and a voltage is applied to the dust collecting electrode 22 through the second power source 30, and the polarity of the voltage is the same as the voltage applied to the sample 16, and the absolute value thereof is applied to the sample 16 The voltage is as large as 0.5 to 5 kV, for example. That is, for example, when a voltage of -3 kV is applied to the sample 16, a voltage of -3.5 to -8 kV, for example, a voltage of -5 kV, is applied to the dust collecting electrode 22.

The vacuum chamber 12 is made of a metal material such as iron or aluminum and has a ground potential. Further, when foreign matter such as particles existing inside the vacuum chamber 12 is charged by static electricity or the like, when the potential of the sample 16 is negative, foreign matter such as positively charged particles is attracted by the electric field and flies to the sample 16.

According to this example, the entire circumference of the sample 16 to which the negative potential is applied is surrounded by the dust collecting electrode 22, and a negative voltage which is larger than the voltage applied to the sample electrode 16 is applied to the dust collecting electrode 22, and is attracted by the electric field. Most of the foreign matter such as the flying particles are trapped by the dust collecting electrode 18, so that the probability that the foreign matter such as particles flies to the sample 16 and adheres to the surface of the sample 16 is greatly reduced. Thereby, it is possible to prevent foreign matter from adhering to the surface of the sample 16.

In this example, a gap control plate 24 for preventing foreign matter such as particles from adhering to the surface of the sample 16 by passing off the track of the dust collecting electrode 22 is provided. When the gap control plate 24 is provided as described above, the attraction force of the dust collecting electrode 22 to foreign matter such as particles that are deviated from the orbit of the dust collecting electrode 22 is lowered. Therefore, the probability that the foreign matter such as particles is captured by the dust collecting electrode 22 is inversely proportional to the distance. Reduced ground. Therefore, when a negative voltage is applied to the sample 16, the electric field intensity A between the sample 16 and the dust collecting electrode 22 is made negative (A &lt; 0), whereby the attraction force of the dust collecting electrode 22 can be enhanced, and the particles can be improved. The probability that the foreign matter is caught by the dust collecting electrode 22. Moreover, the electric field between the gap control plate 24 and the dust collecting electrode 22 is made strong. The degree (absolute value) B has a relationship of 0.1 B (absolute value) 10 kV/mm, and the probability that foreign matter such as particles is caught by the dust collecting electrode 22 is further improved.

For example, a negative voltage of -1 to -5 kV is applied to the sample 16, and a negative voltage of -1.5 to -10 kV which is -0.5 to -5 kV which is a large negative value of the negative voltage applied to the sample 16 is applied to the dust collecting electrode 22. When the gap control board 24 is at the ground potential, the distance between the sample 16 and the dust collecting electrode 22 is set to L1 = 10 mm, the distance between the gap control plate 24 and the dust collecting electrode 22 is Z1 = 8 mm, and the sample 16 and the dust collecting electrode 22 are The electric field strength A between the gaps is negative (A &lt; 0), and the electric field strength (absolute value) between the gap control plate 24 and the dust collecting electrode 22 is B = 19.19 to 1.25 kV/mm (= 1.5 to 10 kV / 8 mm), in particular When a voltage of -5 kV is applied to the dust collecting electrode 22, the electric field intensity (absolute value) becomes B = 0.625 kV/mm (= 5 kV / 8 mm), which is an effective condition. At this time, as the withstand voltage of the space is transmitted so as not to exceed 10 kV/mm, it is possible to prevent the occurrence of spatial discharge.

Figure 112 shows the detailed construction of the XY stage 14. As shown in FIG. 112, the XY stage 14 is formed by laminating the X platform 32 and the Y platform 34, and an ultrasonic motor 36 is interposed between the X platform 32 and the Y platform 34. The first dust cover 40 whose upper end reaches the dust collecting electrode 22 is disposed at an outer position of the dust collecting electrode 22 that surrounds the upper surface of the XY stage 14, and the outer side of the ultrasonic motor 36 is disposed to block the ultrasonic motor 36. The second dust cover 42 at the open end of the storage portion.

By providing the first dust cover 40 in this manner, it is possible to prevent foreign matter such as particles from scattering on the surface of the sample 16 and adhering to the surface. Further, by disposing the second dust cover 42 on the outside of the ultrasonic motor 36 serving as the particle generation source, it is possible to prevent foreign matter such as particles scattered from the ultrasonic motor 36 from scattering to the vacuum chamber. 12 inside. In this way, it is possible to prevent the foreign matter such as particles from scattering from the generation source into the inside of the vacuum chamber 12, and it is particularly effective when a motor is driven by a frictional wall surface such as a piezoelectric actuator (Piezoactuator).

In this example, as shown in detail in FIG. 113, a sealed structural wiring box 50 is disposed inside the vacuum chamber 12. The wiring box 50 is for preventing foreign matter such as particles generated from the cable from being scattered into the vacuum chamber 12 due to bending or friction of the cable. In this example, the portions where the cable 52 is bent accompanying the movement of the XY stage 14 or the like are all included in the wiring box 50. That is, one end of the cable 52 is connected to the fixing plate 54 fixed to the XY stage 14, and the cable 52 linearly extends from the XY stage 14 toward the wiring box 50, and is reached by the slit 50a provided in the wiring box 50. The inside of the wiring box 50 is then bent 180 degrees downward and then reversed. The other end of the cable 52 is connected to a moving plate 58, which is provided with a terminal block 56 disposed inside the wiring box 50. Thereby, when the XY stage 14 moves in the X direction, only the cable 52 is bent in the meandering portion 52a in the wiring box 50.

Inside the wiring box 50, a guide wheel 60 that extends in the Y direction and guides the cable 52 is disposed. When the XY stage 14 moves in the Y direction, the moving plate 58 moves in the Y direction along the guide wheel 60. Thereby, the stress in the Y direction does not act on the cable 52 before the moving plate 58. The cable extending from the terminal block 56 is connected to the feedthrough provided in the vacuum chamber 12 through a wiring hole provided in the wiring box 50, but is not shown.

Thus, when the bent portions of the cable 52 are all located in the wiring box 50, since the holes connected to the outside of the wiring box 50 are small, the inside of the wiring box 50 is produced. The probability that foreign matter such as raw particles fly out of the wiring box 50 is greatly reduced, and most of them are attached to the inner wall of the wiring box 50. In this example, the wiring box dust collecting electrode 62 is disposed inside the wiring box 50, and a voltage for trapping foreign matter such as particles is applied to the dust collecting electrode 62 for the wiring box, thereby causing foreign matter such as particles to be removed. The probability that the wiring box 50 is scattered to the outside is greatly reduced.

In addition, by applying: (1) straightening the cable length, (2) fixing and correcting the cable with a cable strap (bundle), and (3) using a flat cable as a cable, etc., it is possible to reduce friction due to most cables. The case of producing particles. That is, if the length of most cables is straightened and fixed, the cable bundles will be integrated. When the XY platform moves, the cables will bend, but the friction of the cables at this time can be reduced, and the foreign matter such as particles can be reduced. produce. Furthermore, by using a flat cable as a cable, most of the wiring can be set as a cable, and the friction of the cables can be eliminated. Further, when it is not possible to directly use a flat cable having a large number of wirings, it is effective to combine the above measures (1) and (2).

In the above example, the dust collecting electrode 22 having a rectangular cross section is used, but the dust collecting electrode 22d having a circular cross section may be used as shown in Fig. 114. The diameter D of the dust collecting electrode 22d is preferably used in a range of a relationship of 0.5 L2 &lt; D &lt; 5 L2 in the relationship of the distance L2 between the sample 16 and the dust collecting electrode 22d. If the diameter D of the dust collecting electrode 22d is smaller than this range, the capturing probability of the collecting electrode 22d is lowered. If the range is larger, the capturing probability of the collecting electrode 22d does not change, but the excess catching particles is caused. Wait for foreign matter.

Further, in the above example, the dust collecting electrode 22 is disposed at a position spaced apart from the sample 16 by a predetermined interval, and its polarity is the same as the voltage applied to the sample 16, and its absolute value is more than the voltage applied to the sample 16. A large voltage is applied to the dust collecting electrode 22, but as shown in Fig. 115, the inner peripheral edge portion is placed in contact with the outer peripheral portion of the sample 16 while surrounding the entire periphery of the sample 16 A dust collecting electrode 22e having a rectangular frame shape and a rectangular cross section is formed, and a voltage equal to the voltage applied to the sample 16 by the first power source 28 is applied to the dust collecting electrode 22e via the second power source. The dust collecting electrode 22e has a thickness of, for example, 0.1 to 5 mm, and the width W2 is the same as the dust collecting electrode 22 described above, for example, 5 to 50 mm.

In this example, the dust collecting electrode 22e having a smaller inner shape than the sample 16 is used, and the inner peripheral edge portion of the dust collecting electrode 22e is in contact with the outer peripheral portion of the sample 16, but it may be the same as the 116th. As shown in the figure, for example, a dust collecting electrode 22f having a shape slightly larger than that of the sample 16 is used, and the dust collecting electrode 22f is surrounded by the entire circumference of the sample 16 in such a manner as to retain a minute gap S. The dust collecting electrode 22f. This gap S is, for example, 1 to 500 μm.

In this example, a negative voltage of, for example, -1 to -5 kV is applied to the sample 16 by the first power source 28, and the same voltage as that applied to the sample 16 is applied to the dust collecting electrode 22e by the second power source 30. That is, for example, when a voltage of -3 kV is applied to the sample 16, a voltage of -3 kV is applied to the dust collecting electrode 22e.

Similarly to the above, when the potential of the sample 16 is negative, foreign matter such as positively charged particles is attracted by the electric field and flies to the sample 16. According to this example, the dust collecting electrode 22e having the same potential as that of the sample 16 is disposed at a position surrounding the entire periphery of the sample 16, so that foreign matter such as particles flying from the electric field is attracted by the electric field. Most of it is captured by the dust collecting electrode 22e. In this manner, by collecting most of the foreign matter such as particles by the dust collecting electrode 22e disposed around the sample 16, the foreign matter such as particles flying on the surface of the sample 16 and adhering to the surface is reduced, thereby preventing the sample 16 from being prevented. Foreign matter adheres to the surface.

In this example, when the distance between the dust collecting electrode 22e and the gap control plate 24 is Z2, it is particularly effective when W2 &gt; 4Z2 in relation to the width W2 of the dust collecting electrode 22e. In addition, when the magnitude (absolute value) of the voltage density B between the dust collecting electrode 22e and the gap control plate 24 is larger than 0.1 kV/mm (B (absolute value)&gt;0.1 kV/mm), it is more effective.

Fig. 117 is a view showing still another example in which the above-described example mainly shown in Fig. 111 and the example shown in Fig. 115 are combined. In this example, the first dust collecting electrode 70 which is continuous in a rectangular frame shape and has a rectangular cross section is disposed such that the inner peripheral edge portion is in contact with the outer peripheral edge portion of the sample 16 and surrounds the entire periphery of the sample 16 and In a manner of surrounding the entire circumference of the first dust collecting electrode 70, a second dust collecting electrode 72 continuous in a rectangular frame shape and having a rectangular cross section is disposed at a position spaced apart from the first dust collecting electrode 70 by a predetermined interval. Further, the second power source 74 is connected to the first dust collecting electrode 70, and the third power source 76 is connected to the second dust collecting electrode 72.

Further, in the same manner as described above, the second dust collecting electrode extending in a straight line may be arranged to extend over substantially the entire length of each side of the first dust collecting electrode, and the first collecting electrode may be surrounded by the second dust collecting electrode. The entire outer circumference or the second dust collecting electrodes extending in a straight line may be separated from each other in the middle.

In this example, as in the above, a voltage of, for example, -1 to -5 kV is applied to the sample 16 by the first power source 28, and the first dust collecting electrode 70 is applied to the first dust collecting electrode 70. When a voltage applied to the sample 16 is the same voltage, for example, when -3 eV is applied to the sample 16, -3 eV is applied. Further, the second dust collecting electrode 72 is applied with the same polarity as that applied to the sample 16 and the absolute value is higher than the voltage applied to the sample 16 by, for example, a voltage of 0.5 to 5 kV. That is, for example, when a voltage of -3 kV is applied to the sample 16, a voltage of -3.5 to -8 kV, for example, -5 kV is applied to the second dust collecting electrode 72.

In this example, similarly to the example disclosed in the above-mentioned FIG. 111 and the like, when a negative voltage is applied to the sample 16, the electric field intensity A between the sample 16 and the second dust collecting electrode 72 is made negative (A). In the manner of &lt;0), the suction force of the second dust collecting electrode 72 is increased, and the probability that foreign matter such as particles is caught by the second dust collecting electrode 72 can be improved. Further, by setting the electric field intensity (absolute value) B between the gap control plate 24 and the second dust collecting electrode 72 to have a relationship of 0.1 B (absolute value) 10 kV/mm, foreign matter such as particles can be caused. The probability of capturing the second collecting electrode 72 is further improved.

Similarly to the dust collecting electrode 22e shown in Fig. 115, the first dust collecting electrode 70 has a thickness of, for example, 0.1 to 5 mm and a width W3 of, for example, 5 to 50 mm. Further, the second dust collecting electrode 72 has a thickness of, for example, 0.1 to 50 mm and a width W4 of, for example, about 5 to 50 mm, similarly to the dust collecting electrode 22 shown in Fig. 111.

Then, for example, a negative voltage of -1 to -5 kV is applied to the sample 16 and the first dust collecting electrode 70, and a negative voltage applied to the second dust collecting electrode 72 is applied higher than the negative voltage applied to the sample 16 and the first dust collecting electrode 70. A negative value of -0.5 to -5 kV becomes a negative voltage of -1.5 to -10 kV. When the gap control plate 24 is at the ground potential, the distance between the gap control plate 24 and the second dust collecting electrode 72 is set to Z8=8 mm, and the sample is sampled. The electric field intensity A between the 16 and the second dust collecting electrode 72 becomes a negative value (A &lt; 0), and the electric field intensity (absolute value) between the gap control plate 24 and the second dust collecting electrode 72 becomes B = 0.19 to 1.25. kV/mm (=1.5 to 10 kV/8 mm), especially when a voltage of -5 kV is applied to the dust collecting electrode 22, the electric field intensity (absolute value) is B = 0.625 kV/mm (= 5 kV / 8 mm), and becomes effective. condition. At this time, it is possible to prevent the occurrence of spatial discharge by making the withstand voltage of the space not exceed 10 kV/mm.

Figure 118 is a schematic view showing another embodiment of the electron beam inspection apparatus. In this example, the XY stage 16 on which the sample 16 is placed is disposed inside the vacuum chamber 12, and the imaging optical inspection device 80, the SEM inspection device 82, and the optical microscope 84 are disposed in the vacuum chamber 12, and in the vacuum chamber 12 The sample 16 provided on the XY stage 14 can be observed and inspected by both the optical inspection apparatus 80 and the SEM inspection apparatus 82.

According to this example, since the sample 16 is mounted on the XY stage 14 common to the optical inspection apparatus 80 and the SEM inspection apparatus 82, the sample 16 moves between the optical inspection apparatus 80 and the SEM inspection apparatus 82. The coordinate relationship is obtained in an unambiguous manner, and it is easy to define the same part with high precision.

That is, when the movement of the sample is performed between the separated individual inspection devices, since the samples must be placed on the respective platforms, the alignment of the samples must be performed separately. In this way, even if the alignment of the sample is performed, the definition error of the same portion is 5 to 10 μm or more. In particular, when the sample has no pattern, the error is naturally greater because the position reference cannot be defined.

According to this example, even in the imaging optical inspection device 80 and SEM In the case where the movement of the sample 16 is performed between the type inspection devices 82, the same portion can be defined with high precision, so that high-precision portion definition can be achieved, and the precision can be, for example, 1 μm or less. Therefore, when the image optical inspection apparatus 80 performs inspection of patterns and pattern defects, it is also effective to perform the definition and detailed observation of the detection defects by the SEM inspection apparatus 82. That is, since the position can be defined, not only can the presence or absence of the defect be judged (if it is not, the virtual detection is performed), and the determination of the correct size and shape of the defect can be performed at a high speed. In the case of individual devices, the detection and definition of pattern defects takes a lot of time.

As described above, by preventing the foreign matter such as particles from adhering to the surface of the sample 16 and using the imaging optical inspection apparatus and the SEM type inspection apparatus to mount the apparatus in the same chamber, ultrafine patterns of 100 nm or less are particularly used. The inspection, its determination and classification are carried out in a good and efficient manner.

As shown in Fig. 119, in the equal electric field generated between the electrodes formed by the parallel plates (q+=q-), even if particles are formed by the insulator, the particles may be electrostatically induced by the electric field. Polarized, but not scattered. However, when the electric field is an unequal electric field, the charge generated by the polarization of the medium causes the particles to scatter. Similarly, as shown in Fig. 120, in the unequal electric field between one pair of electrodes on the one side (q+q-), if particles formed by the insulator exist, the particles will be electrostatically induced by the electric field. The polarization is scattered, but as shown in Fig. 121, in the equal electric field (q+=q-) generated between one pair of electrodes on one side, even if there are particles formed by the insulator, these particles may be caused by The electric field is electrostatically induced and polarized, but does not scatter.

That is, as shown in Fig. 120 and Fig. 121, foreign matter such as particles is scattered. The probability of being considered to be greatly affected by the initial charge of foreign matter such as particles before the polarization of the medium₀Around. The residue has an initial charge q₀It can be considered that it is mainly given by static electricity generated by the air flow during vacuum evacuation.

Figure 122 shows the other vacuum chambers 12a provided in the electronic inspection device. An XY stage 14 on which the sample 16 is placed is disposed inside the vacuum chamber 12a. Two vacuum pumps 90a and 90b are connected to the vacuum chamber 12a, and the two vacuum pumps 90a and 90b are connected to the common dry pump 92. In addition, in order to prevent the foreign matter (residue) such as particles that cannot be removed by the cleaning of the vacuum chamber 12a from being charged by static electricity, the static elimination device 94 is provided inside the vacuum chamber 12a, and the gas is ionized by soft X-ray or UV rays. The ionized gas electrostatically removes the surface of the object in the ionized gas.

According to this example, the static elimination device 94 can be operated while the vacuum evacuation is started in the vacuum chamber 12a, or before the vacuum evacuation is started, or in the vacuum evacuation in the vacuum chamber 12a. Device 94 continues to operate. That is, the static eliminating device 94 is continuously operated until the air flow in the vacuum chamber 12a disappears and static electricity is not generated due to the air flow. In this way, it is possible to prevent the foreign matter (residue) such as particles in the vacuum chamber 12a from being charged, and to pass through the initial charge q.₀=0 (refer to Fig. 121), the probability of scattering due to polarization of the dielectric in the unequal electric field is reduced.

Further, foreign matter such as particles remaining in the vacuum chamber cannot be removed by cleaning in the vacuum chamber, and even if it is minute and light, it is precipitated in the plane structure in the vacuum chamber by gravity.

Fig. 123 is a perspective view showing an example of a wall for constituting a vacuum chamber 12 shown in Fig. 106 or the like or a vacuum chamber shown in Fig. 122. The plane constructor of 12a. Figure 124 is a cross-sectional view of Fig. 123. As shown in Fig. 123 and Fig. 124, the wall structure for constituting, for example, the vacuum chamber 12 or 12a (refer to Figs. 106 and 122, etc.) is a wall body having a plurality of dot-like holes 96a on the inner surface. 96 is composed. In this manner, by providing a plurality of dot-shaped holes 96a on the inner surface of the wall body 96, the foreign matter P such as particles remaining in the vacuum chamber is deposited on the bottom of the hole 96 by gravity. Through the electrostatic shielding effect of the dot matrix holes 96, the electric field does not enter the bottom of the lattice-shaped holes 96a. Therefore, the foreign matter (residue) P deposited to the bottom of the lattice-shaped hole 96 is not affected by the electrostatic attraction force and does not scatter. Therefore, it is possible to prevent foreign matter such as particles remaining in the vacuum chamber 12 or 12a from adhering to the surface of the sample 16 disposed inside the vacuum chamber 12 or 12a, for example.

Fig. 125 is a perspective view showing another example of a wall for constituting a plane constructor such as the vacuum chamber 12 shown in Fig. 106 or the vacuum chamber 12a shown in Fig. 122. Figure 126 is a cross-sectional view of Figure 125. As shown in Figs. 125 and 126, the wall structure constituting the vacuum chamber 12 or 12a (refer to Figs. 106 and 122, etc.) is separated from the wall 98 by a flat wall 98 and a predetermined interval from the wall 98. The flat plate 100 having a mesh structure in parallel is formed, and the flat plate 100 having a mesh structure is connected to an independent power source 102.

Thereby, the foreign matter P such as particles remaining in the vacuum chamber 12 or 12a can be passed through the mesh 100 of the mesh structure to the surface of the wall body 98 by its gravity. Since the wall 98 is covered by the mesh 100 of the mesh structure, the electric field is shielded by the mesh 100 of the mesh structure without reaching the surface of the wall 98. Therefore, the foreign matter (residue) P reaching the surface of the wall body 98 is not scattered by the electrostatic attractive force. Thereby, it is possible to prevent residual in the vacuum chamber 12 or 12a Foreign matter such as particles inside adheres to, for example, the surface of the sample 16 disposed inside the vacuum chamber 12 or 12a.

In particular, by applying a voltage to the flat plate 100 of the mesh structure independently, the foreign matter P such as particles remaining in the vacuum chamber is actively attracted to the mesh structure type flat plate 100, and the gravity of the foreign matter P can be used. The foreign matter P is precipitated and adhered to the surface of the wall body 98 constituting a planar structure such as the vacuum chamber 12 or 12a.

Although an embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various embodiments may be implemented within the scope of the technical idea, and it is needless to say. Further, the present embodiment can also be applied to the above-described first to twenty-seventh embodiments and the embodiment in which the number is not assigned.

(Embodiment 28)

- A substrate mounting device for mounting a substrate on a tray and a positioning method of the substrate relative to the tray -

Hereinafter, in the inspection apparatus and the inspection method of the present invention, a substrate mounting device in which a substrate is mounted on a tray and a positioning method of the substrate with respect to the tray will be described.

Hereinafter, a substrate mounting device according to an embodiment of the present invention will be described with reference to the drawings.

In the present embodiment, the substrate is, for example, a mask used in an EUV exposure apparatus, and the substrate mounting apparatus is mounted on, for example, a substrate inspection apparatus for a mask.

Fig. 127 and Fig. 128 show an inspection apparatus including a substrate mounting device. Before explaining the substrate mounting device in detail, an outline of the inspection device will be described.

Fig. 127 is a view of the inspection apparatus 1 as viewed from above. As shown in FIG. 127, the inspection apparatus 1 is largely divided into an atmospheric conveying unit 3 and a vacuum conveying unit 5. The atmospheric transport unit 3 processes the substrate in the air, and the vacuum transport unit 5 processes the substrate in a vacuum. The atmospheric conveying unit 3 and the vacuum conveying unit 5 are separated by a switchable partition wall.

The atmospheric transport unit 3 is also referred to as a small environmental chamber. A standard mechanical interface (SMIF) cartridge 7 is provided adjacent to the atmospheric transport unit 3. Further, the atmospheric transport unit 3 is provided with an atmospheric transfer robot 9, a substrate rotation reversing unit 11, a substrate mounting unit 13, a neutralization unit 15, and a fan filter unit (FFU, Fan Filter Unit).

The standard mechanical interface box 7 is a structure for holding a substrate (mask) before and after the inspection. The atmospheric transfer robot 9 is a robot for transporting substrates in the atmosphere. The substrate rotation reversing unit 11 can receive the substrate from the atmospheric transfer robot 9 and rotate and reverse it. The board mounting unit 13 mounts the board on the tray. The substrate mounting unit 13 is the substrate mounting device of the embodiment. The static eliminating unit 15 performs a static elimination process of the substrate before and after the inspection. Although not shown, the fan filter unit (FFU) is provided only in the upper part of the small environment room of the air transport unit 3. More specifically, the FFU is installed in the ceiling or in the vicinity of the atmospheric transfer robot 9, the substrate rotation reversing unit 11, the substrate mounting unit 13, the static eliminating unit 15, and the like.

Further, the vacuum transfer unit 5 is provided with a load lock chamber 17, a transfer chamber 19, a first turbo molecular pump 21, a main chamber 23, an inspection lens barrel 25, and a second turbo molecular pump 27.

Two CCD cameras 29 are disposed in the load lock chamber 17. CCD camera The machine 29 is used for positioning of the substrate as will be described later. The transfer chamber 19 is a chamber for transporting the substrate from the load lock chamber 17 to the main chamber 23. The transfer chamber 19 is provided with a vacuum transfer robot 31. The vacuum transfer robot 31 is a robot that transports substrates in a vacuum. Further, the first turbo molecular pump 21 sets the load lock chamber 17 and the transfer chamber 19 to a vacuum state. The main chamber 23 and the inspection lens barrel 25 are configured to irradiate a substrate with a charged particle beam to inspect the structure of the substrate. The second turbo molecular pump 27 sets the main chamber 23 and the inspection lens barrel 25 to a vacuum state.

Fig. 128 is a view of the main chamber 23 and the inspection lens barrel 25 as seen from the lateral direction. The main chamber 23 is provided with a platform 33. The tray holding the substrate is placed on the platform 33. The platform 33 is configured to move the tray in the horizontal direction for moving the tray in the X, Y, and θ directions. The X and Y directions are directions along mutually orthogonal axes, and θ is an angle of rotation about the axis of rotation, that is, also rotationally.

The inspection lens barrel 25 is connected to the upper side of the main chamber 23. The inspection lens barrel 25 includes an electron gun 35, a primary lens system 37, a secondary lens system 39, and a detector 41. The electron gun 35 is a charged particle beam source. The electron gun 35 and the primary lens system 37 are electron beam irradiation systems for illuminating the electron beam toward the substrate. The electron beam is deflected by the Wien filter 43 and is incident on the substrate through the objective lens system 45. When the electron beam is irradiated onto the substrate, the substrate emits a signal having the substrate information. The signal is, for example, two discharges of electrons (secondary electrons, reflected electrons, backscattered electrons) or mirror-reflected electrons. The signal reaches the detector 41 through the objective lens system 45, the Wien filter 43, and the secondary lens system 39, and is detected by the detector 41.

The detector 41 is connected to the image processing unit 47, and supplies the detected signal to the image processing unit 47. The image processing unit 47 is configured by a computer having an image processing function to perform defect inspection. In other words, the image processing unit 47 forms an image of the sample from the signal detected by the detector 41, and further processes the image of the sample to detect and determine the defect.

Further, as shown in Fig. 128, the inspection apparatus 1 is provided with a control unit 49. The control unit 49 is configured by a computer and is configured to control the entire inspection apparatus 1 to perform inspection. As shown in the figure, the control unit 49 controls the main chamber 23, the inspection lens barrel 25, and the image processing unit 47. Thereby, the control unit 49 moves the substrate (tray), irradiates the electron beam onto the substrate, and generates the substrate image in the image processing unit 47.

The control unit 49 can control the inspection conditions, specifically, the energy, the magnification, the dose, and the like of the electron beam. The beam energy, specifically, the incident energy when the electron beam is irradiated onto the substrate.

In the present embodiment, the inspection apparatus 1 is a map projection type inspection apparatus. In the image projection type inspection apparatus, the electron beam has a beam size (beam diameter) corresponding to the two-dimensional pixel group, that is, has a certain specification. The illuminated area on the sample also has an area corresponding to the two-dimensional pixel group. The signal detected by the detector 41 also corresponds to the two-dimensional pixel group. Further, the detector 41 has a detection capability corresponding to a two-dimensional pixel group, for example, a CCD having a two-dimensional detection surface.

The image projection type inspection device and the SEM type inspection device are compared. In the SEM, the electron beam is thinner and corresponds to 1 pixel. In the SEM, Scanning with an electron beam, repeating the measurement of 1 pixel, and accumulating the measured values to obtain an image of the sample. The electron beam has a beam size of one pixel with respect to the electron beam in the SEM inspection apparatus, and the electron beam of the image projection type inspection apparatus has a beam specification corresponding to a plurality of pixel groups. The image projection type inspection device can inspect fine defects. Further, the image projection type inspection apparatus is not limited to inspection of pattern defects, and various inspections can be performed. For example, the image projection type inspection apparatus can also be used for inspection of foreign matter such as particles, and can also be used for defect inspection in a multilayer film.

Returning to Fig. 127, the overall operation of the inspection apparatus 1 will be described. The atmospheric transfer robot 9 takes out the substrate from the standard mechanical interface box 7 and transports it to the static eliminating unit 15. The static eliminating unit 15 performs power removal of the substrate. Then, the substrate is transported to the substrate rotation inverting unit 11 through the atmospheric transfer robot 9, and the substrate is reversed and rotated as needed. Next, the atmospheric transfer robot 9 transports the substrate to the substrate mounting unit 13. In the board mounting unit 13, the board is mounted on a tray prepared in advance.

When the substrate is mounted on the tray, the atmospheric transfer robot 9 holds the tray and transports it to the load lock chamber 17. At this time, the partition walls of the air transport unit 3 and the vacuum transport unit 5 are opened. In the load lock chamber 17, the mark of the substrate is photographed by the CCD camera 29. This detects the position of the mark. The vacuum transfer robot 31 transports the tray from the load lock chamber 17 to the main chamber 23, and places the tray on the platform 33 of the main chamber 23. At this time, the substrate is positioned according to the result of the mark detection.

When the substrate is mounted on the tray, the substrate is mounted by the substrate mounting unit 13. Next, when it is mounted on the platform 33 as described above, that is, according to CCD imaging The detection result of the machine 29 positions the substrate. The positioning of the former can be called "temporary positioning", and the positioning of the latter can be called "formal positioning". In the actual substrate inspection process, after the formal positioning, the optical microscope can be used for further positioning. The optical microscope system and the electron beam inspection apparatus are disposed together in the main chamber 23, and the substrate and the tray are positioned together using the optical image of the optical microscope, and then the substrate is inspected by electron beam. Here, the substrate is positioned such that the electron beam is irradiated to the defect position detected by the optical microscope.

In the main chamber 23, as described using the Fig. 128, the substrate is inspected by electron beam irradiation. The substrate after the inspection is carried by the vacuum transfer robot 31 from the main chamber 23 to the load lock chamber 17 through the transfer chamber 19.

Next, the atmospheric transfer robot 9 transports the substrate from the load lock chamber 17 to the substrate mounting unit 13. The substrate mounting unit 13 removes the substrate from the tray. The substrate is further transported to the substrate rotation inverting unit 11, and rotated and reversed as needed. Next, the substrate is transported to the static eliminating unit 15 for power removal. Thereafter, the substrate is returned to the standard mechanical interface box 7 by the atmospheric transfer robot 9.

The overall configuration and operation of the inspection apparatus 1 have been described above. The above configuration is a typical system configuration example, and the above operation is a typical operation example. Therefore, the inspection apparatus unit described above may be replaced with the primary beam of the ninth diagram as light, and the secondary beam using the photoelectron inspection apparatus. The configuration and operation of the inspection apparatus are not limited to the above examples as long as they are within the scope of the invention.

"Substrate mounting device"

Next, the substrate mounting device according to the embodiment will be described in detail. The substrate mounting device corresponds to the substrate mounting unit of the aforementioned 127th drawing 13. The system and the atmospheric transport robot 9 function together to mount the substrate on the tray. The substrate is, for example, a mask, and more specifically, for example, a glass mask having a square length of 6 , and a thickness of 6.35 mm.

The 129th to 138th drawings show the substrate-mounted device of the embodiment. FIGS. 129 to 132 show the substrate loading device 51 in the trayless state, and FIGS. 133 to 136 show the substrate mounting device 51 and the tray. Fig. 129 is a plan view of the substrate mounting device 51, Fig. 130 is a view of the substrate mounting device 51 of Fig. 129 as seen from the direction of the arrow A, and Fig. 131 is a substrate mounting device 51 of Fig. 129 viewed from the direction of the arrow B. FIG. 132 is a view of the substrate mounting device 51 of FIG. 129 as seen from the direction of the arrow C, and is also taken along a diagonal line of the tray T and the substrate S. Similarly, Fig. 133 is a plan view of the substrate mounting device 51, and Fig. 134 is a view of the substrate mounting device 51 of Fig. 133 as seen from the direction of the arrow D. Fig. 135 is a view of the substrate mounting device 51 of Fig. 133 as seen from the direction of the arrow E. Fig. 136 is a view of the substrate mounting device 51 of Fig. 133 as seen from the direction of the arrow F, and is also taken along a diagonal line of the tray T and the substrate S. In addition, FIGS. 137 and 138 are schematic diagrams showing the substrate mounting device 51 for the sake of explanation.

The substrate mounting device 51 is a device for mounting the substrate S on the tray T. Roughly speaking, the substrate loading device 51 is a platform 53, a lifting mechanism 55, a clamping mechanism 57, a tray holding mechanism 59, and<img file="TW201634917A_D0009.tif" wi="59" he="61" img-format="tif" img-content="character" orientation="portrait" inline="no" />The frame tribe is composed of 61 institutions. Hereinafter, the configuration of the tray T will be described first, and then various configurations of the substrate loading device 51 will be described.

"Tray T"

The configuration of the tray T will be described with reference to Figs. 133 to 138. As shown in FIG. 137 and the like, the tray T is constituted by the tray main body 71 and the frame portion 73. The tray T is made of, for example, ceramic.

The tray body 71 has a flat shape and has a substantially square shape corresponding to the square substrate S. The plurality of substrate mounting pins 75 protrude from the tray main body 71, and the substrate S is supported by the substrate mounting pins 75. The number of pins sold is four. With this configuration, the substrate S is in contact with the substrate-mounting pin 75 with a small contact area, and is held in the suspended state in the tray body 71 without direct contact. Further, the substrate mounting pin 75 has a surface on which the substrate S is hard to slide, thereby preventing the substrate S during transportation from shifting.

The frame portion 73 is supported by the tray body 71 and surrounds the substrate S. The frame portion 73 is configured to apply a potential to the upper surface of the substrate S at the time of inspection, and is provided with a potential via the terminal portion 77. Further, the through-frame portion 73 functions as a virtual substrate around the substrate S. The equipotential surface tortuosity of the end portion of the substrate S is reduced, and the potential in the vicinity of the end portion is made uniform. As a result, the potential of the entire substrate including the end portion can be made uniform, and the inspection precision can be improved. Further, the frame portion 73 can be raised and lowered with respect to the tray body 71. As shown in FIG. 137, when the frame portion 73 is lowered, the surface of the frame portion and the surface of the substrate are at the same height, and the frame portion 73 surrounds the substrate S. As shown in FIGS. 134 and 135, when the frame portion 73 is raised, an insertion port 79 is formed between the tray main body 71 and the frame portion 73. The insertion port 79 is a gap or an opening between the tray body 71 and the frame portion 73, and may also be referred to as an insertion gap. The insertion port 79 allows the substrate S to be inserted and removed, and is accessible for access to the substrate S.

The configuration of the frame portion 73 will be described in more detail. The frame portion 73 has a frame body 81. The plurality of frame legs 83 extending from the frame body 81 to the lower portion.

As shown in FIGS. 133 and 138, the frame body 81 is a quadrangular plate and has a quadrangular opening 85. The material of the frame body 81 may be an insulator such as ceramic or the like. The opening 85 has a size corresponding to the substrate S. The substrate S is disposed in the opening 85, whereby the substrate S is surrounded by the frame portion 73. The frame body 81 and the substrate S are not in direct contact. The frame body 81 and the substrate S form a substantially constant gap over the entire circumference.

A frame cover 86 is provided on the upper surface of the frame portion 73. Referring to Fig. 134, the upper end of the frame portion 73 is slightly wider than the frame body 83, and the upper end portion of the frame portion 73 is slightly protruded from the inner side, and the upper end portion corresponds to the frame cover 86. Figure 143 shows the frame cover 86 in a schematic manner. As shown in Fig. 143, the frame cover 86 is a thin flat plate. The frame cover 86 is bent downward in the L shape at the outer end portion, and is fixed to the side surface of the frame main body 81 by screws or the like as shown in the drawing (the frame cover 86 may be fixed to the upper surface of the frame main body 81, etc.). Other parts). The material of the frame cover 86 is a conductive material such as copper. The outer peripheral shape of the frame cover 86 is substantially the same as the frame main body 81. However, since the opening of the frame cover 86 is smaller than the opening 85 of the frame main body 81, the frame cover 86 protrudes inside from the frame main body 81. When the substrate S is mounted, the inner peripheral edge of the frame cover 86 overlaps with and contacts the outer periphery of the substrate S. In order to create such an overlap, the opening specifications of the frame cover 86 are specified. Therefore, the frame cover 86 covers the gap between the frame portion 73 (the frame portion main body 81) and the substrate S when viewed from above. By providing such a frame cover 86, the equipotential surface buckling of the edge portion of the substrate S is alleviated. The frame cover 86 is omitted in the drawings of Figs. 137 and 138.

The frame body 81 is provided with two terminal portions 77. The terminal portion 77 is protruded from Opening 85. When the substrate S is disposed in the opening 85, the terminal portion 77 is in contact with the upper surface of the substrate S. As described above, the terminal portion 77 is for applying a potential to the upper surface of the substrate S in order to inspect the substrate. Further, the frame cover 86 covers the gap between the substrate S and the frame portion 73 in addition to the terminal portions 77 over the entire circumference of the frame portion.

The number of the frame leg portions 83 is four, and the four frame leg portions 83 are disposed in the vicinity of the four corners of the frame portion body 81, respectively. As shown in Fig. 137, when descending, the frame leg portion 83 is supported by the tray main body 71, and the frame portion main body 81 is positioned above the tray main body 71, and the upper surface of the frame portion and the upper surface of the substrate S on the substrate mounting pin 75 are provided. At substantially the same height (as shown in FIG. 143, in detail, the upper surface of the frame body 81 is at the same height as the upper surface of the substrate. The upper surface of the frame cover 86 is located at substantially the same height as the upper surface of the substrate, but in detail, The thickness of the cover is slightly higher than the top of the substrate. The same applies below). When the frame portion 73 is raised, an insertion port 79 is formed between the tray main body 71 and the frame portion body 81, and the substrate S can be taken in or out toward the substrate S.

Further, as shown in FIGS. 134 and 135, in order to guide the lifting operation of the frame portion 73, a lifting guide 87 is provided. The lift guide 87 is a guide rod that protrudes from the tray body 71. The lifting guide 87 is inserted into the guide hole of the frame leg portion 83. Thereby, the frame portion 73 can only move in the vertical direction with respect to the plane of the tray.

Further, as shown in FIG. 133 and the like, the tray body 71 has a plurality of protruding edge portions 89. In the present embodiment, the four protruding edge portions 89 are respectively provided at four corners of the tray main body 71. The protruding edge portion 89 is a portion that protrudes outward from the frame portion 73. More specifically, the corners of the tray body 71 are provided with wall portions, and the protruding edge portions 89 project outward from the upper end of the wall portion. As described later, the protruding edge The portion 89 is for supporting the tray T by the transport robot, and holds the tray body 71 when the frame portion 73 is raised.

"platform"

The platform 53 is configured to hold the tray T. As shown in FIG. 129 and FIG. 130, etc., the platform 53 has a platform bottom plate 91. The platform bottom plate 91 is provided with a plurality of platform columns 93. In the present embodiment, four platform posts 93 are provided at positions corresponding to the four corners of the tray T. The four corners of the tray body 71 are supported by the four platform posts 93. The tray T is held so as to be movable in the horizontal direction, and can be moved when positioned by the chucking mechanism 57 described later.

As shown in FIG. 129 and the like, the platform 53 is provided with a substrate presence detecting sensor 95, a tray presence detecting sensor 97, a substrate tilt detecting sensor 99, and a tray tilt detecting sensor 101. The sensors are mounted to the platform floor 91.

"lifting mechanism"

The elevating mechanism 55 is configured to move the frame portion 73 and the substrate S in the vertical direction with respect to the plane of the tray, and to raise and lower the frame portion 73 and the substrate S.

As shown in Fig. 130, Fig. 134, Fig. 137 and the like, the elevating mechanism 55 has a lifting plate 111 belonging to a flat lifting member. The lift plate 111 is driven by the lift cylinder 113 used by the lift mechanism 55, and is lifted and lowered within a predetermined range below the tray T.

The plurality of frame portion holding pins 115 and the plurality of substrate holding pins 117 protrude upward from the lift plate 111. The substrate holding pin 117 and the frame holding pin 115 are substantially the same height.

The frame portion holding pin 115 corresponds to the frame portion holding member of the present invention, and is disposed at a position below the frame portion 73 and is raised and lowered together with the lift plate 111. In the present embodiment, the four frame portion holding pins 115 are disposed at positions corresponding to the four frame leg portions 83 of the frame portion 73. As shown in FIGS. 134 and 135, when the elevating mechanism 55 is raised, the frame holding pin 115 passes through the hole of the tray main body 71 and abuts against the underside of the frame portion 83, lifting the frame portion 73 upward, and An insertion port 79 is formed between the tray body 71 and the frame portion 73.

Further, the substrate holding pin 117 corresponds to the substrate holding member of the present invention, and is disposed at a position below the substrate S3 at the time of positioning, and is lifted and lowered together with the lift plate 111. As shown in Fig. 133, in the present embodiment, the four substrate holding pins 117 are arranged to be offset from the four substrate mounting pins 75 of the tray T. Further, as shown in FIGS. 134 and 135, when the elevating mechanism 55 is raised, the substrate holding pin 117 is also raised through the hole of the tray main body 71 in the same manner as the frame holding pin 115. The front end of the substrate holding pin 117 is slightly above the front end of the substrate mounting pin 75 of the tray T. Thereby, the substrate S is held by the substrate holding pin 117 before being mounted on the substrate mounting pin 75.

Here, the position of the substrate S in the height direction when the tray T is mounted on the substrate mounting pin 75 is referred to as a substrate mounting height. Further, the position of the substrate S in the height direction when held by the substrate holding pin 117 of the elevating mechanism 55 is referred to as a substrate holding height. As described above, the substrate holding height is higher than the substrate mounting height. Further, the substrate holding height is at a height corresponding to the above-described insertion port 79. Before the substrate holding pin 117 reaches the substrate holding height and the substrate S is mounted on the substrate mounting pin 75, the substrate holding pin 117 is held at the substrate holding height, and then the substrate S is sandwiched as will be described later.

Further, the substrate holding pin 117 is made of a material that is easy to slide, such as polychlorotrifluoroethylene (PCTFE, registered trademark). Thereby, the positioning of the substrate S is easy.

Further, as shown in FIG. 130 and the like, the elevating mechanism 55 has a frame holding height adjusting mechanism 119 for adjusting the height of the frame holding pin 115, and a substrate holding height adjustment for adjusting the height of the substrate holding pin 117. Agency 121. The frame portion holding height adjusting mechanism 119 has a threaded configuration. A male screw is provided on the outer circumference of the frame holding pin 115. The male screw is screwed to the female thread of the lifting plate 111, and the height of the pin can be adjusted by rotating the frame holding pin 115. The substrate holding height adjustment mechanism 121 also has the same thread configuration. With this configuration, the frame holding height adjusting mechanism 119 and the substrate holding height adjusting mechanism 121 can independently adjust the heights of the frame holding pin 115 and the substrate holding pin 117.

"clamping mechanism"

The clamping mechanism 57 is configured to position the substrate S relative to the tray T. In the present embodiment, the holding mechanism 57 is moved in a direction parallel to the plane of the tray, and then both the substrate S and the tray T are sandwiched by one clamping operation. The details of the clamping mechanism 57 will be described below.

As shown in Figs. 133, 136, 137, and 138, the holding mechanism 57 is composed of a plurality of holders. In the present embodiment, the number of the holders is two, and the fixed side holder 131 and the passive side holder 133 are provided. The fixed side holder 131 and the passive side holder 133 are disposed to face each other along diagonal lines of the substrate S and the tray T.

The fixed side holder 131 and the passive side holder 133 are respectively coupled to The fixed side pressure cylinder 135 and the passive side pressure cylinder 137 are provided. The fixed-side holding body 131 is linearly driven by the fixed-side pressure cylinder 135, and the passive-side holding body 133 is linearly driven by the passive-side pressure cylinder 137. Thereby, the clamping mechanism 57 is opened and closed. The fixed side pressure cylinder 135 and the passive side pressure cylinder 137 correspond to the clamp moving mechanism of the present invention.

The fixed side holder 131 has two tray holding arms 139 on the fixed side and two substrate holding arms 141. The tray holding arm 139 and the substrate holding arm 141 extend toward the tray T. The height of the tray holding arm 139 corresponds to the height of the tray body 71. The height of the substrate holding arm 141 corresponds to the substrate holding height when the substrate holding pin 117 of the lifting mechanism 55 is supported. Therefore, the substrate holding arm 141 is positioned above the tray holding arm 139.

The two tray holding arms 139 are horizontally separated as shown in the drawing, and each tray holding arm 139 has a tray holding portion 143 in the vicinity of the front end. The tray holding portion 143 is constituted by a holding pin which is a portion for contacting the tray T when being gripped for positioning. The moving direction of the clamping mechanism 57 is along the diagonal of the tray T, and therefore, it is inclined with respect to the contact surface. Therefore, the pin of the tray holding portion 143 protrudes obliquely from the tray holding arm 139 toward the diagonal of the tray.

The substrate holding arm portion 141 also has the same configuration as the tray holding arm 139. That is, the two substrate holding arms 141 are separated in the horizontal direction, and each of the substrate holding arms 141 has the substrate holding portion 145 in the vicinity of the front end. The substrate holding portion 145 is constituted by a holding pin which is a portion for contacting the substrate S when being sandwiched for positioning. The moving direction of the clamping mechanism 57 is along the substrate The diagonal line of S, therefore, forms an oblique state with respect to the contact surface. Therefore, the pin of the substrate holding portion 145 protrudes obliquely from the substrate holding arm 141 toward the diagonal of the substrate.

Further, the outer shape of the substrate S is smaller than the tray T. Therefore, as shown in Fig. 132, the substrate holding arm 141 protrudes much more than the tray holding arm 139.

Next, the passive side holder 133 will be described. The passive side holder 133 has a shape having a narrow width compared to the fixed side holder 131.

The passive side holder 133 and the fixed side holder 131 have a tray holding arm 147 protruding toward the tray T and a substrate holding arm 149. The height of the tray holding arm 147 corresponds to the height of the tray body 71. The height of the substrate holding arm 149 corresponds to the height of the substrate S when supported by the substrate holding pin 117 of the lifting mechanism 55. The number of the tray holding arms 147 and the fixed side are also two. On the other hand, the number of the substrate holding arms 149 is different from that of the fixed side, and there is only one.

The two tray holding arms 147 have the same configuration as the fixed side except for the point where the arm spacing is narrow and the arm length is short. That is, the two tray holding arms 147 are separated in the horizontal direction, and each tray holding arm 147 has a tray holding portion 151 near the front end. The tray holding portion 151 is constituted by a holding pin, and contacts the tray T during clamping. The pin of the tray holding portion 151 protrudes obliquely toward the diagonal of the tray.

The substrate holding arm 149 is one as described above. A substrate holding portion 153 is provided at the front end of the arm. The substrate holding portion 153 is a concave portion formed by recessing the tip end of the arm. When sandwiching the substrate holding portion 153, the both side edges of the concave portion abut against the two end portions of the end surface of the substrate S.

In the present embodiment, the tray holding arm 147 and the substrate holding arm 149 are disposed, and the substrate holding portion 153 is brought into contact with the substrate S before the tray holding portion 151 on the passive side contacts the tray T. Further, as shown in Fig. 132, the substrate holding arm 149 is disposed on the linear motion guide 155 so as to be linearly movable in the moving direction of the clamping. Further, a spring-type pusher 157 is disposed behind the substrate holding arm 149. The spring type pusher 157 is equivalent to the spring pusher of the present invention. The substrate holding arm 149 and the substrate holding portion 153 are elastically urged toward the substrate S by the spring pusher 157. When the substrate holding portion 153 is pushed toward the substrate S by the sandwiching reaction force of the substrate S, the substrate holding portion 153 can be retracted on the passive side holding body 133 while being elastically pressed toward the substrate S. Regarding the function of the above elastic structure, it is stated later.

Further, in the clamp mechanism 57, the fixed-side holder 131 is provided with a tray holding position adjusting mechanism 159 and a substrate holding position adjusting mechanism 161.

The tray holding position adjusting mechanism 159 is a mechanism provided in each of the tray holding portions 143 for adjusting the amount by which the tray holding portion 143 protrudes toward the tray T. The tray clamping position adjustment mechanism 159 has a threaded configuration. The outer circumference of the holding pin of the tray holding portion 143 is provided with a male screw which is screwed to the female thread of the tray holding arm portion 139, and the height of the pin can be adjusted by rotating the holding pin. The substrate holding position adjusting mechanism 161 is a mechanism provided in each of the substrate holding portions 145 for adjusting the amount of protrusion of the substrate holding portion 145 to the substrate S. The substrate holding position adjusting mechanism 161 also has the same screw configuration as the tray holding position adjusting mechanism 159. With this configuration, the amount of protrusion of the tray holding portion 143 and the substrate holding portion 145 of the fixed side holder 131 can be independently adjusted.

Further, the passive side holding body 133 also has a tray holding position adjusting mechanism 163 similarly to the fixed side holding body 131. The tray holding position adjusting mechanism 163 is provided in each of the tray holding portions 151, and has the same screw structure as the fixed side.

Further, in the present embodiment, the tray T is sandwiched at four locations, and the substrate S is also sandwiched at four locations. However, the number of nip points is not limited to the above. The minimum number of nips required varies depending on the shape of the tray and the substrate. For example, when the substrate is circular, the substrate can be held at three points.

"Tray retention mechanism"

As shown in Figs. 130, 131, 134, and 137, the tray holding mechanism 59 is a driving mechanism that is perpendicular to the plane of the tray. The tray holding mechanism 59 has a plurality of tray holding pins 171 that are lowered toward the tray body 71 and abut against the tray body 71. The tray holding pin 171 corresponds to the tray holding member of the present invention. As shown in Fig. 131, the tray holding pin 171 is attached to the pin attachment arm 173, and the pin attachment arm 173 is coupled to the lift cylinder 175 for the tray holding mechanism 59, and the tray holding pin 171 is lifted and lowered by the lift cylinder 175.

The tray holding mechanism 59 has a function of preventing the tray body 71 from rising when the frame portion 73 is raised. In other words, when the elevating mechanism 55 lifts the frame portion 73, the tray holding mechanism 59 lowers the tray holding pin 171 and abuts against the tray main body 71, thereby restraining the tray main body 71 from rising.

In the present embodiment, the number of the tray holding pins 171 is four. The four tray holding pins 171 are disposed so as to correspond to the protruding edge portions 89 of the four corners of the tray main body 71, and abut against and press the protruding edge portion 89 when descending. Thus, the upper surface of the protruding edge portion 89 has a function as a contact portion of the tray holding pin 171. Further, the lower surface of the protruding edge portion 89 also has a function of a robot supporting surface when the tray is conveyed.

Further, as shown in Fig. 131, the tray holding mechanism 59 has a height adjusting mechanism 177 for adjusting the height of the tray holding pin 171. The height adjustment mechanism 177 has the same thread configuration as the adjustment mechanism of the lift mechanism 55. That is, the outer circumference of the tray holding pin 171 is provided with a male screw which is screwed to the female thread of the pin mounting arm 173, and the height of the pin is adjusted by rotating the tray holding pin 171.

"Boxing the Horde below"

As shown in Fig. 130, Fig. 131, Fig. 134, and Fig. 137, the frame lower body mechanism 61 is a drive mechanism that is perpendicular to the plane of the tray. The frame lower arm mechanism 61 has a plurality of sub-frames 181 which are lowered toward the frame portion 73 and urge the frame portion 73. The box tribe sells 181 series, which is equivalent to the lower part of the frame tribe of the present invention. As shown in Fig. 131, the frame tribe is sold 181 in the pin mounting arm 183, and the pin mounting arm 183 is connected to the lifting cylinder 185 for the frame lower mechanism 61 to use the lifting cylinder 185 to sell the frame. 181 lifts.

In the present embodiment, the number of the frame-selling pins 181 is four, and is disposed above the frame portion 73. When the frame portion is lifted and released after the positioning by the gripping mechanism 57 is completed, the frame lower arm mechanism 61 lowers the frame lower body pin 181, and presses the frame portion 73 so that the frame portion 73 is surely seated on the tray body 71. The seating of the frame portion 73 can be detected to ensure that the descending action is true.

In addition, as shown in Fig. 131, the frame lower body mechanism 61 has a The box tribe sells a height adjustment mechanism 187 of height 181. The height adjustment mechanism 187 also has the same thread configuration as the adjustment mechanism of the lifting mechanism 55. That is, a male thread is provided on the outer circumference of the frame tribe under the pin 181, and the male thread is screwed with the female thread of the pin mounting arm 183, and the height of the pin is adjusted by the lower frame pin 181 of the rotating frame.

The configuration of each unit of the substrate mounting device 51 of the present embodiment will be described above. Next, the operation of the substrate mounting device 51 will be described.

FIGS. 139 and 140 are schematic views showing the operation of the substrate mounting device 51. As shown in FIG. 139, the tray T is placed on the stage 53 before the substrate S is mounted. The frame portion 73 is lowered and supported by the tray body 71 by the frame leg portion 83.

The elevating mechanism 55 is lowered, and the frame holding pin 115 and the substrate holding pin 117 are also located below. When the clamping mechanism 57 is opened, the fixed side holder 131 and the passive side holder 133 are retracted and are in a predetermined retracted position. Further, the tray holding mechanism 59 and the dropping mechanism 61 are raised, and the tray holding pin 171 and the frame lowering pin 181 are positioned above the tray T.

When the mounting operation is started, first, the tray holding mechanism 59 lowers the tray holding pin 171. The tray holding pin 171 abuts against the protruding edge portion 89 of the tray body 71 to press the tray body 71.

Next, the elevating mechanism 55 raises the frame holding pin 115 and the substrate holding pin 117. The frame holding pin 115 passes through the hole of the tray body 71 and abuts against the lower surface of the frame leg portion 83 to lift the frame portion 73. At this time, the frame portion 73 is guided by the direct-motion type lifting guide 87 (FIG. 135) provided in the frame leg portion 83, and is raised in the vertical direction.

Further, the substrate holding pin 117 also rises through the hole of the tray main body 71 similarly to the frame portion holding pin 115. The front end of the substrate holding pin 117 is slightly above the front end of the substrate mounting pin 75 of the tray T.

The frame portion 73 is lifted up by the frame holding pin 115, and an insertion port 79 is formed between the tray main body 71 and the frame portion body 81. The substrate S is inserted from the insertion port 79 by a robot belonging to the substrate conveyance unit (substrate conveyance means), and placed on the substrate holding pin 117 of the substrate holding height. The substrate S is inserted from the direction of the arrow D in Fig. 133. Since the substrate holding pin 117 protrudes more than the substrate mounting pin 75 of the tray T, the substrate S is held on the substrate holding pin 117 instead of the substrate mounting pin 75. The above robot is the atmospheric transfer robot 9 of Fig. 127. The transport robot has, for example, a fork arm that supports the substrate S with the arm front end and extends the arm to insert the substrate S.

Here, in the present embodiment, the height of the front end of the frame holding pin 115 and the height of the front end of the substrate holding pin 117 are substantially the same. However, the lower surface of the frame leg portion 83 is located below the lower surface of the substrate S. Before the substrate holding pin 117 abuts against the lower surface of the substrate S, the frame holding pin 115 has first abutted against the lower surface of the frame portion 83. Therefore, the elevating mechanism 55 can raise the height of the frame portion 73 by the frame portion 73. As a result, the frame body 81 reaches a level higher than the substrate holding height of the front end of the substrate holding pin 117, and a sufficiently large insertion port 79 can be formed.

Next, as shown in Fig. 140, the clamping mechanism 57 is used to position the substrate S relative to the tray T. In the present embodiment, the substrate T is correctly positioned by sandwiching the tray T and the substrate S by the chucking mechanism 57.

Figure 141 is a diagram illustrating the clamping action. As shown in the figure, before the start of the clamping, the fixed side holder 131 and the passive side holder 133 are separated from the tray T and the substrate S. At the time of starting the clamping (at the start of positioning), first, the fixed-side holding body 131 is driven by the fixed-side pressure cylinder 135 (Fig. 136), and is stopped until it is moved to the predetermined fixed-side clamping position.

Then, the passive side holder 133 is driven by the passive side cylinder 137 to abut against the end portion of the tray T and the end portion of the substrate S, and the tray T and the substrate S are pressed against the fixed side holder 131. The tray holding portion 151 on the passive side presses the tray main body 71 toward the tray holding portion 143 on the fixed side, and the substrate holding portion 153 on the passive side presses the substrate S toward the substrate holding portion 145 on the fixed side.

Thereby, the tray main body 71 and the substrate S can be clamped by one clamping operation. The tray main body 71 and the substrate S are positioned in contact with the fixed side holder 131 to determine the relative positional relationship between the substrate S and the tray T. Therefore, the substrate S is positioned relative to the tray T.

When the gripping operation is completed, the passive side holder 133 is retracted and separated from the substrate S and the tray T. Then, the fixed side holder 131 is retracted and separated from the substrate S and the tray T, and the clamping operation is released.

According to the above aspect, according to the present embodiment, the tray T and the substrate S are simultaneously positioned in one clamping operation. The fixed-side holding body 131 has a configuration in which the tray holding portion 143 and the substrate holding portion 145 are integrally provided, and thus the positional relationship is fixed. The tray T and the substrate S are pressed by the fixed side holders 131. As a result, the individual positions of the tray T and the substrate S and the position of the substrate S relative to the tray T correspond to the tray holding portion 143 and the substrate holding The positional relationship of the part 145 is determined. Thus, high positioning precision can be obtained.

Further, in the present embodiment, the fixed side holder 131 is advanced as described above, and then the passive side holder 133 is advanced. By constituting the holding mechanism 57 in this manner, the advantage that the amount of movement of the tray can be reduced can be obtained.

Further, in the nip operation, the fixed side pressure cylinder 135 and the passive side pressure cylinder 137 are controlled such that the clamping force on the fixed side is smaller than the clamping force on the passive side. The fixed side clamping force is a force by which the fixed side pressure cylinder 135 fixes the fixed side clamp body 131, and the passive side clamping force is the force of the passive side pressure cylinder 137 moving the passive side clamp body 133. By the setting of the clamping force, the offset of the fixed-side holder 131 can be prevented, and high positioning precision can be obtained.

Further, in the present embodiment, as described above, the substrate holding pin 117 protrudes above the substrate mounting pin 75 of the tray T. Therefore, as shown in Fig. 140, the substrate S is held on the substrate holding pin 117 while being positioned, instead of being held on the substrate mounting pin 75. The substrate holding pin 75 has a surface that is less likely to slide in consideration of subsequent conveyance, and the substrate holding pin 117 has a surface that is easy to slide. Therefore, the substrate S is placed in a state of being easily moved in the horizontal direction. Thereby, damage during positioning of the substrate S can be prevented. Moreover, since the substrate S is surely moved to an appropriate position, the positioning precision is also improved.

Further, as described above, in the present embodiment, the passive side holding body 133 is provided with the substrate holding arm 149 on the linear motion guide 155, and the spring type pushing member 157 is disposed behind the substrate holding arm 149. side. This configuration produces a function in the following manner when clamped.

Referring to Figure 142, in the present embodiment, the tray holding arm 147 and The substrate holding arm 149 is disposed such that the substrate holding portion 153 first contacts the substrate S before the tray holding portion 151 on the passive side contacts the tray T.

Therefore, when the passive side holding body 133 continues to move, the substrate holding portion 153 initially abuts against the substrate S and is pressed. The opposite side of the substrate S abuts against the fixed-side holder 131, and the spring-type pusher 157 is contracted by the reaction force from the substrate S, and the substrate-clamping arm 149 is retracted on the linear guide 155. At this point of time, the substrate S is held by the elastic force of the spring-type pusher 157. Then, the tray holding portion 151 abuts against the tray T and is pressed. Next, both the substrate S and the tray T are clamped and positioned.

Here, it is assumed that the linear motion guide 155 and the spring type pusher 157 are not provided. At this time, due to the dimensional error, one of the tray holding portion 151 and the substrate holding portion 153 cannot be brought into contact with the object, and thus the positioning precision is lowered. In the present embodiment, the tray holding portion 151 and the substrate holding portion 153 can reliably abut against the tray T and the substrate S, and the positioning precision is improved.

Further, in the present embodiment, an elastic structure is used on the substrate S instead of the tray T. The weight of the substrate S is smaller than that of the tray T, and the substrate S is held by the substrate holding pin 117 of the elevating mechanism 55 when positioned. Therefore, the substrate S is easy to move, and even when pressed by a spring or the like, it can be surely moved to achieve the purpose of positioning, so that the positioning precision can be further improved.

Go back to Figure 140. When the positioning is completed and the nip is released, the elevating mechanism 55 is lowered, and the frame holding pin 115 and the substrate holding pin 117 are also lowered. The frame portion 73 is lowered, the frame leg portion 83 is supported by the tray body 71, and the frame portion 73 is Go back to the original location. Further, the substrate S is lowered and supported by the substrate mounting pin 75. The upper surface of the substrate S and the upper surface of the frame portion 73 are at the same height (as described above, in detail, the upper surface of the frame portion 81 is at the same height as the upper surface of the substrate, and the upper surface of the frame portion 86 is located above the substrate. Slightly high position). The terminal portion 77 of the frame portion 73 is in contact with the upper surface of the substrate S. Further, the frame cover 86 covers the gap between the frame portion 73 and the substrate S along the entire circumference of the substrate S (excluding the portions of the two terminal portions 77).

When the lifting mechanism 55 descends, the frame lower tribe mechanism 61 causes the box tribe to sell 181 down. The frame tribe sells 181 and abuts on the frame 73 and pushes it. Thereby, the frame lower body mechanism 61 will assist the frame portion 73 to descend. The frame portion 73 is also used by the frame tribe to sell 181 in addition to its own weight, and is surely lowered to the original position.

Secondly, the frame lower agency 61 will cause the frame tribe to sell 181, and the tray holding mechanism 59 will cause the tray holding pin 171 to rise. In this way, a series of piggybacking actions are completed. When the mounting operation is completed, the robot transports the tray T on which the substrate S is mounted. The robot has an arm that supports the underside of the protruding edge portion 89 of the tray body 71 with the arm portion. As described above, the robot is the atmospheric transport robot 9 of Fig. 127, and the tray T is transported to the load lock chamber 17.

The above description will be given of an operation in which the substrate S is positioned while being placed on the tray T. Next, the operation when the substrate S is removed from the tray T will be described.

In the present embodiment, the substrate mounting device 51 is provided with an inspection device. At the end of the inspection, the tray S is transported by the robot and placed on the platform 53 of the substrate loading device 51. At this time, the substrate mounting device 51 is mounted and mounted as described above. The same state at the time of completion. That is, in the substrate mounting device 51, the elevating mechanism 55 is lowered, and the frame holding pin 115 and the substrate holding pin 117 are also located below. The clamping mechanism 57 is opened, and the fixed side holder 131 and the passive side holder 133 are retracted and are in a predetermined retracted position. Further, the tray holding mechanism 59 and the dropping mechanism 61 are in an ascending state, and the tray holding pin 171 and the frame lowering pin 181 are positioned above the tray T.

First, the tray holding mechanism 59 lowers the tray holding pin 171 and presses the tray body 71. Next, the elevating mechanism 55 raises the frame holding pin 115 and the substrate holding pin 117. The frame portion 73 also rises, and an insertion port 79 is formed between the tray main body 71 and the frame portion body 81. The terminal portion 77 of the frame portion 73 is separated from the upper surface of the substrate S. The frame cover 86 also moves above the substrate S. Further, the substrate S is lifted by the substrate holding pin 117 of the elevating mechanism 55. Thereby, the substrate S is detached from the substrate mounting pin 75 on the tray and floated up.

Next, the robot extends the arm, reaches the substrate S from the insertion port 79, and supports the lower surface of the substrate S, and takes out the substrate S from the insertion port 79. The substrate S is taken out in the opposite direction to the arrow D in Fig. 133.

Here, the repositioning can be performed before the substrate taking operation described above. Specifically, the positioning is performed by the clamp mechanism 57 after the frame portion is raised. The positioning action can be the same as the positioning of the substrate mounting process. Then, through the robot, the substrate S is taken out. By repositioning in this way, the accuracy of handling can be increased.

When the substrate S is taken out, the elevating mechanism 55 is lowered, and the frame portion 73 is returned to the original position. At this time, the box tribe lower mechanism 61 lowers the frame tribe down 181 to assist the lowering of the frame portion 73. Then, under the framework of the Horde, The frame tribe is sold 181, and the tray holding mechanism 59 raises the tray holding pin 171. Thereby, a series of substrate removal operations are completed. Further, the present embodiment is applicable to the above-described first to twenty-eighth embodiments and the embodiment in which the number is not assigned.

"Other implementations"

The above is a description of the substrate mounting device of the present invention. Other embodiments of the present invention are described below.

A mask inspection apparatus or method for inspecting a mask of a charged particle beam by irradiating a mask mounted on a tray with the above-described substrate mounting apparatus or method.

A mask manufacturing apparatus or method for irradiating a mask mounted on a tray with a charged particle beam to perform inspection of a mask forming step using the substrate mounting apparatus or method described above.

The mask inspected by the above-described mask inspection device or method. A mask manufactured by the above-described mask manufacturing apparatus or method.

A semiconductor manufacturing apparatus or method for manufacturing a semiconductor device using the above-described mask.

A semiconductor device fabricated using the above-described mask. A semiconductor device manufactured by the above semiconductor manufacturing apparatus or method.

The type of the mask is, for example, a Cr mask, an EUV mask, or a mask for a nano-imprint. The Cr mask is used for light exposure, the EUV mask is used for EUV exposure, and the nanoprint mask is used to form a resist pattern by nanolithography. Regarding each of the masks, a mask formed with a pattern can be an object to be inspected. Further, a mask (material) in a state in which a film is formed before pattern formation can be an inspection object.

The following describes a method of manufacturing a semiconductor device to which the mask obtained by the above embodiment can be applied. The manufacturing method includes the following steps (1) to (5).

(1) Wafer manufacturing steps for manufacturing wafers (or wafer preparation steps for preparing wafers)

(2) a mask manufacturing step of manufacturing an exposure mask (or a mask preparation step of preparing a mask)

(3) Wafer processing steps for performing necessary processing on the wafer

(4) A wafer assembly step in which one wafer of a wafer formed on a wafer is cut out and allowed to operate

(5) Check the wafer wafer inspection steps

In the wafer processing step of the above (3), the designed circuit patterns are sequentially stacked on the wafer to form a plurality of wafers that operate as a memory, an MPU (Micro Processing Unit), or the like. The wafer processing step includes the following majority steps.

(A) Forming a thin film forming step (using CVD, sputtering, etc.) for forming a dielectric thin film, a wiring portion, and an electrode portion as a metal thin film

(B) an oxidation step for oxidizing the film layer and the wafer substrate

(C) Photolithographic imprinting step using a mask to form a resist pattern for selectively processing a thin film layer, a wafer substrate, or the like

(D) an etching step of processing the thin film layer and the substrate according to the resist pattern (using, for example, dry etching)

(E) ion impurity implantation diffusion step

(F) Repellent stripping step

(G) Steps for inspecting processed wafers

The wafer processing step is repeated only for the necessary number of layers. The lithography imprinting step of (C) is as follows.

(a) a resist coating step of applying a resist on a wafer on which a circuit pattern has been formed in the previous step

(b) Step of exposing the resist

(c) an imaging step of developing the exposed resist to obtain a resist pattern

(d) an annealing step for stabilizing the developed resist pattern

The preferred embodiments of the present invention have been described above. As described above, according to the present invention, the chucking mechanism holds both the tray and the substrate in one clamping operation. This clamping operation is an action of pressing the tray and the substrate from different directions by a plurality of holding pieces. Since the chucking mechanism that moves in parallel with the plane of the substrate is used, positioning can be performed without tilting the substrate. Further, since the tray and the substrate are held in one operation by using the holder in which the tray holding portion and the substrate holding portion are integrally formed, the substrate can be aligned with respect to the tray in accordance with the positional relationship between the tray holding portion and the substrate holding portion. Positioning is performed, and the positional relationship between the tray and the substrate is correctly determined. Thus, the substrate can be positioned with high precision with respect to the tray.

Further, according to the present invention, since the substrate and the tray are held in one operation as described above, it is possible to obtain not only the positioning of the substrate but also the positioning of the tray at the same time.

Furthermore, according to the present invention, since the substrate is held close to the substrate, contact with the substrate can be avoided as much as possible. It is ideal for releasing the grip after the positioning is completed, and reducing the contact time. Therefore, according to the present invention, positioning can be appropriately performed without contacting the substrate as much as possible.

In particular, in the above-described embodiment, the substrate is a mask, and the substrate mounting device is provided in a charged particle type inspection device (especially an image projection type inspection device). In this case, there are cases in which both sides of the mask are inspected in turn, and it is necessary to avoid contact with both sides of the mask as much as possible. Therefore, the substrate is mounted well, and the position of the mask is also good, and it is required to perform with high precision, and try not to touch the mask. According to the present invention, it is possible to appropriately respond to such a mask inspection request.

Further, the mask used in the present invention is, for example, a glass mask having a square of 6 inches and a thickness of 6.35 mm, which is relatively heavy compared to a wafer, and is covered by the substrate mounting device described above. The cover can also be properly positioned.

Further, in the inspection apparatus to which the present invention is applied, the mask inspection is performed in the vacuum chamber as described in the above embodiment. However, in this case, the present invention can appropriately position the mask.

Further, in the above-described embodiment, the substrate-mounted device is temporarily positioned by the inspection substrate as a whole. Formal positioning will be done in the latter paragraph. That is, the mark of the substrate is detected by the CCD camera in the load lock chamber, and the set position of the tray in the main chamber is controlled according to the mark position, whereby the official positioning can be performed. The accuracy of this formal positioning is affected by the temporary positioning accuracy. For example, when the visibility of the CCD camera is enlarged due to the low temporary positioning accuracy, the magnification of the CCD camera is reduced, and the formal positioning accuracy is lowered. According to the present invention, the reduction in the accuracy of the official positioning can be prevented, and the inspection accuracy can be improved.

Further advantages of the invention are described. According to the invention, the fixed side holder and the passive side holder can be provided as a plurality of holders. First, the fixed side The holder may be disposed at a predetermined fixed side clamping position, and then the passive side holder is moved to push the tray and the substrate toward the fixed side holder. The positional relationship between the tray holding portion of the fixed-side holding body and the substrate holding portion can be fixed at the time of clamping. Thereby, the fixed-side holding body and the passive-side holding system are connected, and the substrate is accurately positioned at a predetermined position defined by the positional relationship between the tray holding portion and the substrate holding portion of the fixed-side holding body. Further, after the fixed-side holding body advances, the passive-side holding body advances, so that the amount of movement of the tray during positioning can be reduced.

Further, in the passive side holder, the tray holding portion and the substrate holding portion may be disposed such that the substrate holding portion first contacts the substrate before the tray holding portion contacts the tray, and the clamping portion is provided The elastic portion that the substrate holding portion elastically presses against the substrate. The biasing portion may be configured such that the substrate holding portion can be retracted when the substrate holding portion on the passive side is pressed by the substrate due to the clamping reaction force of the substrate. The elastic portion is, for example, an elastic member, and can be disposed behind the substrate holding portion on the passive side. The elastic element is for example a spring. With this configuration, it is possible to prevent the positioning accuracy caused by the dimensional error of the tray holding portion and the substrate holding portion from being lowered, and to accurately position the substrate. Further, by applying an elastic structure to a substrate which is easy to move compared to a tray having a small weight, the positioning accuracy can be further improved.

Moreover, according to the present invention, the holder may have: a tray holding position adjusting mechanism for adjusting the amount of protrusion of the tray holding portion to the tray; and a substrate holding for adjusting the amount of protrusion of the substrate holding portion to the substrate Position adjustment mechanism. The amount of protrusion of the tray holding portion and the amount of protrusion of the substrate holding portion can be independently adjusted. With this configuration, the tray holding portion and the substrate of the holder can be adjusted. The amount of protrusion of the clamping portion can improve the positioning accuracy.

Further, according to the present invention, the substrate mounting device may have a plurality of substrate holding members that are disposed at positions corresponding to the substrates at the time of positioning and that are movable up and down. The chucking mechanism can hold the substrate while the substrate is held at a height higher than the substrate mounting height of the tray by the plurality of substrate holding members. Thereby, the substrate can be held in a state where the substrate is detached from the tray, so that the sliding of the tray and the substrate at the time of positioning can be avoided. Therefore, positioning can be performed in a state where the substrate is not touched as much as possible, and positioning accuracy can be prevented from being lowered due to sliding, and damage to the substrate due to sliding can be further prevented.

Moreover, according to the present invention, the tray may include: a tray body; and a frame portion that can be raised from the tray body and surround the substrate. The substrate mounting device may have a plurality of frame holding members that are disposed at positions corresponding to the frame portions and that are movable up and down. The frame portion is lifted by the frame holding member, and an insertion hole and an insertion opening required for inserting the holder can be formed between the frame portion and the tray body. Thereby, a potential can be applied to the upper surface of the substrate from the frame portion around the substrate. Further, by surrounding the substrate with the frame portion, the potential in the vicinity of the edge portion of the substrate is uniformized. It is preferable that the frame portion is appropriately in contact with the edge portion of the substrate, whereby the potential can be appropriately uniformized. In the above embodiment, the frame cover is provided to cover the gap between the frame portion and the substrate, and the frame cover is in contact with the edge portion of the substrate. Further, in the present invention, since the frame portion is lifted up, even when a frame portion is provided to impart a potential to the upper surface of the substrate, interference between the clamp mechanism and the frame portion can be prevented, and the positioning of the clamp can be appropriately performed.

More specifically, the elevating mechanism can be provided as described in the above embodiment. The lifting mechanism may have a plurality of frames arranged at positions corresponding to the frame portion a portion holding member; and a plurality of substrate holding members disposed at positions corresponding to the substrate, and the plurality of frame holding members and the plurality of substrate holding members are movable up and down. Further, the elevating mechanism can lift the frame portion by raising the frame holding member, and the insertion hole and the insertion opening required for inserting the holder are formed between the frame portion and the tray main body. Further, the elevating mechanism can raise the substrate holding member so that the substrate holding member protrudes above the substrate mounting height of the tray and the substrate holding height corresponding to the height of the insertion opening protrudes. The substrate can be inserted through the insertion opening while maintaining the height of the substrate held by the substrate holding member. The holder can also be inserted smoothly and without interference through the insertion opening.

With this configuration, the frame portion can be appropriately raised in order to form the insertion opening. Moreover, the substrate can be properly held in an upper position that is detached from the tray when positioned. Therefore, the advantages of the present invention described above can be obtained. Further, by providing the frame holding member and the substrate holding member in a common lifting mechanism, the configuration can be easily completed.

Furthermore, according to the invention, the frame portion may have a frame body that surrounds the substrate, and a frame portion that extends from the frame body to the lower portion. The following block leg portion of the substrate may be located below the relatively more below the surface. The frame holding member can be disposed at a position corresponding to the frame portion. Through the support frame leg portion, the frame portion body can be positioned above the substrate holding height held by the substrate holding member to form an insertion opening. According to this configuration, even when the substrate holding member and the frame holding member are simultaneously raised by the same distance, the frame main body can be lifted to a position higher than the substrate holding height, and the substrate can be appropriately formed between the frame main body and the tray main body. The insertion port for clamping. Therefore, the organization of the lifting mechanism can be simply completed.

Moreover, according to the present invention, the substrate mounting device can be provided with: a frame holding height adjusting mechanism for the height of the frame holding member; and a substrate holding height adjusting mechanism for adjusting the height of the substrate holding member. The height of the frame holding member and the height of the substrate holding member can be independently adjusted. With this configuration, the height of the frame holding member and the height of the substrate holding member can be adjusted, and the positional relationship between the frame portion and the substrate can be appropriately adjusted.

Further, according to the present invention, a tray holding mechanism may be provided. The tray holding mechanism can configure the tray holding member at a position corresponding to the tray body to raise and lower the tray holding member. When the lifting mechanism lifts the frame portion, the tray holding mechanism can lower the tray holding member and abut against the tray body to prevent the tray body from rising. According to this configuration, it is possible to prevent the tray main body from rising as the frame portion rises, and the tray and the substrate are simultaneously clamped and positioned appropriately.

Moreover, according to the present invention, the tray body may have a protruding edge portion that protrudes more outward than the frame portion, and the upper surface of the protruding edge portion may be an abutting surface of the tray holding member, and the lower surface of the protruding edge portion may be a handling robot for carrying the tray. Support surface. With this configuration, the protruding edge portion of the tray main body can be used for both the tray holding when the substrate is mounted and the tray transport after the mounting is completed. These two functions can be implemented with a simple configuration.

Furthermore, according to the present invention, it is also possible to provide a sub-framework. The lower arm of the frame tribe can be equipped with the lower part of the frame tribe at the position corresponding to the frame, so that the lower part of the frame tribe can be raised and lowered. When the frame is lifted and lowered, the lower frame of the frame tribe can lower the lower part of the frame tribe to push the frame. According to this configuration, the frame portion that is lifted up to prevent interference with the substrate and the nip can be surely returned to the position surrounding the substrate.

Further, in the present invention, the substrate may be a mask for semiconductor fabrication. Moreover, the substrate can be quadrangular. In the prior art, the positioning mechanism of the circular substrate is generally used, and the positioning of the quadrangular mask is not applicable. According to the present invention, it is suitable for positioning a quadrangular mask. However, within the scope of the present invention, the substrate is not limited to the mask, and may be a wafer. Further, the shape of the substrate is not limited to a square shape, and may be a circular shape. Fig. 144 is an example of a substrate mounting device in which the shape of the substrate is circular, and the circular substrate is, for example, a wafer. As shown in the figure, the shape of the tray is also changed to a circular shape in conjunction with the substrate. The substrate mounting device 201 of Fig. 144 may have the same configuration as that of the above-described embodiment, except for the change required to adapt the shape of the substrate and the tray. The substrate loading device 201 of Fig. 144 can similarly have the above-described various configurations of the present invention, that is, a platform, a lifting mechanism, a clamping mechanism, a tray holding mechanism, a dropping mechanism, and the like. The configuration (including the configuration of the cross section) viewed from the lateral direction may be substantially the same as the embodiment of the above-described quadrangular substrate.

The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above-described embodiments, and those skilled in the art can change the above-described embodiments within the scope of the invention, and it goes without saying. Further, the present embodiment is also applicable to the above-described first to twenty-eighth embodiments and the embodiments in which the number is not assigned.

In view of the above, the present invention is very useful in terms of a substrate mounting technique for positioning a substrate with respect to a tray with high precision.

(Embodiment 29)

- sample observation method and device, and inspection method and device using the same

An example of a sample observation method and apparatus for an inspection apparatus and an inspection method of the present invention will be described.

[1st point of view]

The first viewpoint is about the observation of foreign matter, in particular, the technique of inspecting foreign matter.

An object of the present invention is to provide an electron beam inspection method and an electron beam inspection apparatus capable of detecting foreign matter on the surface of a sample at high speed and reliably.

In the electron beam inspection method of the present invention, the surface of the sample is irradiated with a photographing electron beam having a predetermined irradiation area, and the reflected electrons are detected by a detector to obtain a foreign matter on the surface of the sample and the surface of the sample. An electron beam inspection method according to the aspect of the invention includes a foreign matter charging step of charging a precursor foreign matter by irradiation with a charged electron beam, and forming a potential distribution different from a surface of the sample around the foreign matter; and an enlarged image obtaining step And reflecting the electrons from the foreign matter by the irradiation of the electron beam, and detecting the electrons of the detector by the orbit of bending by the potential distribution to obtain the magnification of the foreign matter. A magnified image of the aforementioned foreign matter having an increased magnification of the surface.

Thereby, since the electron beam inspection is performed using the electron beam having the predetermined irradiation area, the wide area can be inspected at high speed. Further, since the magnified image in which the image of the foreign object is enlarged compared with the surface of the surrounding sample can be obtained, the foreign matter can be surely detected.

Further, in the above-described foreign matter charging step, the foreign matter may be negatively charged by the irradiation of the electron beam for charging, and the amplified image obtaining step may set the incident energy of the photographing electron beam to 10 eV. Hereinafter, mirror-reflected electrons reflected in front of the foreign matter can be detected to obtain the enlarged image of the foreign matter.

Thereby, electrons that are easily generated by the mirror generated in the low incident energy band can be used to surely detect the enlarged image of the foreign matter.

Furthermore, in the present invention, the foreign matter charging step may increase the absolute value of the potential of the foreign matter by the irradiation of the electron beam for charging.

Thereby, the potential difference between the surface of the sample and the foreign matter as the background can be increased, and the contrast of the foreign matter magnified image can be improved, and the electron beam inspection can be facilitated.

Further, in the invention, the incident energy of the charged electron beam may be larger than the incident energy of the photographing electron beam.

Thereby, the absolute value of the negative potential of the foreign matter can be increased by the irradiation of the charged electron beam with a higher incident energy. Therefore, when the electron beam is irradiated, the mirror can easily generate electrons.

Further, in the invention, the incident energy of the charged electron beam may be smaller than the incident energy of the photographing electron beam.

This configuration is suitable for the case where the appropriate incident energy of the electron beam is taken. According to the above configuration, when the magnified image of the foreign matter is obtained by capturing the electron beam, it is possible to prevent the potential change of the surface of the foreign object from increasing. Therefore, it is possible to surely detect the enlarged image.

Furthermore, in the present invention, the charged electron beam and the captured electron beam have the same incident energy and the dose may be different.

Thereby, the charge of the foreign matter can be controlled by the dose without changing the incident energy of the electron beam. Therefore, it is possible to detect a magnified image of a foreign object with simple control.

Moreover, the method of the present invention can inject the aforementioned electron beam into the surface of the sample in a non-perpendicular manner.

Thereby, the incident angle of the photographing electron beam can be appropriately adjusted to obtain a foreign matter magnified image with a higher resolution.

Furthermore, in the above-described enlarged image obtaining step, the incident energy of the photographing electron beam can be set to 10 eV or more, and electrons emitted from the foreign matter and reflected twice can be detected to obtain an enlarged image of the foreign matter.

Thereby, the electrons can be generated from the foreign matter twice, and the magnified image of the foreign matter can be obtained by releasing the electrons twice to perform the electron beam inspection.

Further, in the present invention, the incident energy of the electron beam to be photographed may be equal to or higher than the highest incident energy when all of the electrons reflected from the surface of the sample become mirror-reflected electrons, and all of the electrons reflected from the surface of the sample become electrons emitted twice. The lowest incident energy at the time plus the incident energy below the value obtained by 5 eV.

In other words, in the present invention, the incident energy LE of the aforementioned photographing electron beam can be set to LEALE(LEB+5eV). Here, LEA is the highest incident energy when all of the electrons reflected from the surface of the sample become mirror-reflected electrons, and LEB is the lowest incident energy when all the electrons reflected from the surface of the sample become electrons emitted twice.

Thereby, the electron beam inspection can be performed using the incident energy of the foreign matter and the surface of the surrounding sample having a large difference in the degree of darkness. Therefore, electron beam inspection can be performed easily and surely by obtaining an image with a large contrast. Here, the degree of shading indicates the brightness of the image, and the difference in the degree of shading indicates the difference in brightness.

Further, in the present invention, the incident energy of the photographing electron beam is in an incident energy band when the electrons reflected from the surface of the sample are a mixture of specular reflected electrons and secondary emitted electrons or only two electrons are emitted. The electrons reflected from the foreign matter are incident energy bands when the mirror electrons and the electrons are mixed twice, and are set to an incident energy when the difference between the image of the surface of the sample and the enlarged image of the foreign matter is maximized. .

Thereby, the difference between the surrounding background and the foreign matter is maximized. Therefore, the detection of the foreign matter can be performed in a state where the foreign matter is easily detected.

An electron beam inspection apparatus according to the present invention includes: a stage for placing a sample; a primary optical system for generating an electron beam having a predetermined irradiation area, and irradiating the electron beam to the sample; and 2 times An optical system having a detector for detecting electrons reflected from the sample to obtain an image of a predetermined field of view of the sample; and the primary optical system is configured to charge the foreign matter by irradiation of a charged electron beam. The potential distribution of the foreign matter is different from the surface of the sample, and then the electron beam is irradiated onto the sample; the secondary optical system is reflected from the foreign matter and reaches the track curved by the action of the potential distribution. The electrons of the detector are detected to obtain an enlarged image of the foreign matter whose magnification of the foreign matter is larger than the magnification of the surface of the sample.

Thereby, the entire surface of the sample can be quickly inspected by using an electron beam having an irradiation area of a predetermined size. Moreover, the image of the foreign object can be enlarged to be larger than the surrounding image to reliably detect the foreign matter.

Furthermore, in the present invention, the primary optical system can charge the foreign matter by irradiation of the electron beam for charging, and then irradiate the sample. The incident electron beam having an incident energy of 10 eV or less is detectable by the mirror in which the mirror reflected electrons reflected in front of the foreign matter are detected by the detector to obtain an enlarged image of the foreign matter.

Thereby, it is possible to use a low incident energy to cause foreign matter to form a state in which mirror-reflected electrons are easily generated. By using a mirror to reflect electrons, it is easy to obtain a magnified image of a foreign object. Therefore, the detection of foreign matter can be more certain.

Further, in the present invention, at least one of a Faraday cup, a reference sample wafer, and an EBCCD may be placed on the stage.

Thereby, the change profile of the electron beam can be directly detected, and the adjustment of the electron beam can be appropriately performed.

Further, in the present invention, the reference sample wafer may be placed on the stage, and the reference sample wafer may have any one of a circular, a cross or a rectangular shape.

Thereby, the beam profile of the electron beam can be adjusted, and the mirror reflection electrons can be generated. The mirror-reflected electrons are suitable for the detection of a foreign matter magnified image, and the above configuration can appropriately generate mirror-reflected electrons.

Further, in the present invention, the primary optical system can set the incident energy of the imaging electron beam to 10 eV or more, and the secondary optical system can detect and emit electrons from the foreign matter and reach the detector twice. A magnified image of the foreign matter is obtained.

Thereby, even if electrons are emitted twice from the foreign matter, the foreign matter can be detected.

Further, in the present invention, the aforementioned secondary optical system may have an EB-CCD exchangeable with an NA stop.

Thereby, the profile of the change can be directly determined for the secondary electron beam passing through the secondary optical system. Therefore, appropriate adjustments can be made.

Furthermore, in the present invention, the secondary optical system may have an NA diaphragm. The NA stop can be arranged such that the intensity center of the mirror-reflected electrons coincides with the center position of the pupil.

Thereby, the position of the NA pupil can be appropriately configured to smoothly detect the mirror reflection electronic signal while relatively reducing the detection amount of the electron emission twice. Therefore, a high contrast image can be obtained.

Further, in the present invention, the secondary optical system may have an NA diaphragm, and the shape of the NA diaphragm may be an elliptical shape having a long axis in a direction corresponding to a longitudinal direction of the mirror reflection electron intensity distribution.

Thereby, it is possible to use a circular aperture which is consistent with the mirror reflection electron intensity distribution. Therefore, more mirror reflection electronic signals can be detected to obtain a high contrast image.

Further, in the present invention, the secondary optical system may have an NA diaphragm having a plurality of apertures, and the NA apertures may be arranged such that the plurality of apertures are located around the center of the reflected electron intensity of the mirror.

Here, the NA aperture is an aperture element, and the complex aperture is a plurality of openings provided in the aperture element. According to the above configuration, the arrangement of the pupil can be performed in accordance with the scattering direction of the mirror reflection electrons. It is thus possible to perform appropriate detection of mirror-reflected electrons depending on the application or nature.

Furthermore, in the present invention, the secondary optical system may include an NA diaphragm having a plurality of apertures, and the NA aperture may be disposed such that any one of the plurality of apertures coincides with an intensity center of the mirror reflected electrons. the way.

Here, the NA aperture is a pupil element, and the plurality of apertures are plural openings provided in the aperture element. According to the above configuration, it is possible to effectively inspect foreign matter having characteristics in the scattering direction. Inspections that are useful for foreign matter classification can also be performed.

Further, in the present invention, the secondary optical system may include a moving mechanism for moving the NA diaphragm.

Thereby, the position adjustment of the NA diaphragm can be easily performed using the moving mechanism.

Further, in the present invention, the primary optical system and the secondary optical system may be optical systems that have been subjected to sensitivity correction using microspheres having a known size dispersed on the sample.

Thereby, high-accuracy sensitivity correction can be performed. Therefore, the image can be obtained under good conditions.

Furthermore, the electron beam inspection apparatus of the present invention may have: a chamber for accommodating the platform; and an SEM type inspection device provided in the chamber. The platform can be moved based on the position information of the foreign object magnified image obtained by the detector, and the foreign matter can be inspected in detail by the SEM type inspection device.

Thereby, the detailed inspection of the foreign matter can be performed quickly and accurately, and the foreign matter inspection can be performed at a high speed and with high precision.

As described above, according to the present invention, it is possible to quickly perform foreign matter inspection while reliably and easily detecting foreign matter.

The invention will be further described in detail below. However, the following detailed description and drawings are not intended to limit the invention. In contrast, the scope of the invention is defined by the scope of the patent pending application.

Fig. 145A shows an image obtained by the electron beam inspection method of the present embodiment. The schematic principle of the present invention will now be described with reference to Figure 145.

Fig. 145A shows an image 80 of the foreign matter 10 obtained by the image projection method of the present embodiment. The foreign matter specification is 40 [nm]. In the image of Fig. 145A, the size of the foreign matter 10 is about the extent of satisfying the area of the pixel size of 2 × 2 [μm]. Here, the term "pixel specification" refers to the actual specification on the sample corresponding to one pixel of the detector. The pixel specification means the smallest unit of the specimable sample size. Therefore, in Fig. 145A, the actual foreign matter specification is 40 [nm], and the displayed image 80 is enlarged to a level close to 2 × 2 [μm]. In other words, even if the pixel size is about 1 [μm] or 1.5 [μm], a foreign matter 10 of about 40 [nm] can be found.

In Fig. 145A, the incident energy of the electron beam for photographing is 1 [eV]. The pixel size is 100 [nm]. The conventional technique assumes that the actual size of the foreign matter is 40 [nm], and the pixel size must be smaller than 40 [nm]. On the other hand, in the present embodiment, it is possible to obtain an enlarged image of the foreign matter 10 that is enlarged larger than the optical magnification.

Fig. 145B shows an image 280 of the foreign matter 10 obtained by a conventional SEM type (Scanning Electron Microscope) foreign matter inspection device. The foreign matter specification is 40 [nm]. In Fig. 145B, the pixel size is 2 × 2 [μm] which is the same as that of Fig. 145A. However, as compared with Fig. 145A, the image size of the foreign matter 10 in Fig. 145B becomes quite small.

Thus, the electron beam inspection method and the conventional SEM method of the present embodiment When the equations are compared, an image in which the size of the foreign matter 10 has been greatly increased can be obtained. That is, the detection signal from the foreign matter 10 can be amplified to be larger than the optical magnification. For foreign objects with ultra-small specifications, high sensitivity can also be achieved. Moreover, not only that, but also a larger pixel size than actual foreign matter can be used to detect foreign matter.

Fig. 145C is a side view showing a state in which the foreign matter 10 is present on the sample 20. In Fig. 145C, the surface of the foreign matter 10 is spherical. Therefore, electrons reflected from the surface do not pass through the vertical track, but change the track in an expanded manner. This is due to the following reasons. Since the foreign matter 10 has a spherical surface shape, the potential distribution of the foreign matter 10 is in a state different from the sample surface 21. Therefore, when the sample surface 21 is viewed from a macroscopic point of view, the potential distribution of the portion of the foreign matter 10 is distorted. Thereby the orbit of the electron changes. Details on this point.

Figures 146A and 146B show conventional electron beam inspection methods for comparison purposes. Figure 146A shows an electron beam inspection method of a conventional light mode. In the light mode, the foreign matter 10 is detected by a so-called dark-field scattering method. That is, the sample surface 21 of the sample 20 is irradiated with light, a laser, and the light is scattered and detected by the detector 170. However, in the conventional light mode, the detection sensitivity is lowered in the case where the foreign matter 10 has a size of 50 to 100 [nm] or less of ultrafine foreign matter or adhesion of an organic substance. It is therefore increasingly difficult to apply. The main cause of the decrease in sensitivity is considered to be that the foreign matter 10 is smaller than the wavelength of the light, and the S/N is lowered.

Figure 146B shows an electron ray inspection method using a conventional SEM method. In the SEM method, the electron beam is contracted to reduce the pixel size. It is possible to detect ultra-fine pattern defects 22 and the like. Since the pixel size is smaller than the target foreign matter specification, for example, the inspection of the foreign matter 10 can be performed with high resolution. However, since the pixel size is small, the inspection time becomes very lengthy, and it is difficult to check with a realistic time, which is not practical.

As described above, the conventional technique does not have a foreign matter inspection method and a foreign matter inspection device that can realize high sensitivity, high speed, and high throughput for foreign matter inspection of ultra-fine specifications of 50 to 100 [nm] or less.

Figs. 147A and 147B show an example of the degree of gradation of the enlarged image 80 of the foreign matter 10 obtained by the foreign matter inspection method and the enlarged image. Here, the degree of darkness indicates the brightness of the image, and the difference in the degree of darkness is the difference in brightness. The greater the degree of darkness, the greater the brightness. The 147A is an example of the enlarged image 80. More specifically, the central white area is the enlarged image 81 of the foreign object 10, and the black area indicates the surface image 82 of the sample 20. Here, the foreign matter specification (diameter) was 40 [nm], and the optical magnification was 300 times. At this time, according to the conventional foreign matter inspection method, the image of the image of the foreign matter 10 is 40 [nm] × optical magnification 300 = 12 [μm]. In the present embodiment of Fig. 147A, the size of the enlarged image 81 of the foreign matter 10 is 190 [μm], and the pixel size of the detector is 15 [μm].

Figure 147B shows the degree of shading of the pixel position. Among them, the horizontal axis is the pixel position coordinate, and the vertical axis is the degree of the depth of the profile. In Fig. 147B, the triangular mark (?) indicates a portion of a mountain shape (convex shape). This portion is a region with a relatively deep degree, which corresponds to the portion of the white magnified image 81 of Fig. 147A. That is, the horizontal width (triangle mark Δ) of the enlarged image 81 on the image 80 is 190 [μm].

Here, the pixel size of the detector 65 is 15 [μm]. Therefore, if According to the conventional method, the foreign matter specification is displayed on the image 80 by 12 [μm]. Therefore, the image of the foreign matter 10 is a signal of 1 pixel or less. However, the foreign matter 10 cannot be correctly represented by 1 pixel.

On the other hand, according to the foreign matter inspection method of the present embodiment, the enlarged image 81 of the foreign matter 10 can be detected by forming an image of the number of pixels = 12.7. Therefore, it is possible to shoot at a larger pixel size of a lower magnification. If the image can be taken with a large pixel size, the sample surface 21 as a whole can be inspected at a high speed. Therefore, high-speed and high-throughput foreign matter inspection can be realized. For example, when the foreign matter size is 10 to 30 [nm], the pixel size may be 100 to 1000 [nm]. In this way, it is possible to use a pixel specification that is larger than the foreign matter specification, and a high-speed foreign matter inspection can be obtained.

An electron beam inspection apparatus to which the electron beam inspection method of the present embodiment is applied has an electron beam column (primary optical system) of a projection projection method. In the SEM method, the electron beam is contracted. The spot size of the electron beam is a pixel specification of 1 pixel. On the other hand, in the image projection method, the electron beam has a predetermined area including a plurality of pixels. This sample was irradiated with the electron beam. The detector simultaneously detects the electrons corresponding to the complex pixels. An image of a plurality of pixels is formed and obtained as an image signal. As such, the image projection optical system has an electron irradiation system that irradiates electrons on the surface 21 of the sample; an optical system that forms an image of electrons reflected from the sample surface 21 as a magnification; a detector 70; and will come from the detector 70. The signal is applied to the processed image processing device system.

Figure 148A shows the relationship between the incident energy of the electron beam irradiated to the sample and the electrons emitted from the sample. In more detail, Figure 148A shows The amount of electron emission generated twice when the electron beam was irradiated onto the sample 20 while changing the incident energy was shown.

In Fig. 148A, the horizontal axis represents the incident energy LE [keV], and the vertical axis represents the ratio of the amount of electron emission generated twice to the amount of incident electrons.

In Fig. 148A, when the amount of electron emission generated by two times is greater than 1, the amount of electrons emitted is larger than the amount of electrons injected. Therefore, the sample is positively charged. In Fig. 148A, the positively charged region is a region where the incident energy LE is 10 [eV] or more and 1.5 [keV] or less.

Conversely, in the region where the secondary electron emission amount is less than 1, the amount of electrons injected into the sample 20 is larger than the amount of electrons emitted from the sample 20. Therefore, the sample 20 is negatively charged. In Fig. 148A, the negatively charged region is a region where the incident energy LE is 10 [eV] or less, and a region where the incident energy LE is 1.5 (keV) or more.

Figure 148B shows the mirror reflected electrons. In Fig. 148B, foreign matter 10 is present on the surface 21 of the sample, and the foreign matter 10 is electrically charged with a negative polarity. When the electron beam is irradiated to the sample 20 under certain conditions, the electrons of the electron beam do not collide with the foreign matter 10, but are redirected until they reach the foreign matter. In this way, the electrons that do not collide with the illuminating object and bounce back before reaching are called mirror-reflecting electrons. The electrons to be irradiated become mirror-reflected electrons depending on the potential distribution (charge state) of the foreign matter 10 and the incident energy of the electron beam irradiated to the foreign matter 10. For example, when the foreign matter 10 is in a state of being negatively charged, and the incident energy is not so high, the electron beam is bounced by the negative electric field of the foreign matter 10, and is reflected without colliding with the foreign matter 10, and becomes a mirror-reflecting electron.

Figure 148C shows the electron emission twice. In Fig. 148C, the electron beam is irradiated onto the sample 20 and collides with the sample surface 21, and as a result, electrons are emitted twice from the sample. Similarly, in the case of the foreign matter 10, the electron beam collides with the foreign matter 10, and the foreign matter 10 emits the electrons twice.

In the electron beam inspection method of the present embodiment, the foreign matter 10 present on the sample surface 21 is detected using mirror-reflected electrons and secondary electron emission.

FIGS. 149A and 149B show an example of the relationship between the incident energy LE of the electron beam irradiated on the sample 20 and the foreign matter 10 and the signal intensity/average shading degree of the electron reflected from the sample 20. Here, "reflection" means that electrons from the sample 20 or the foreign matter 10 are returned in a direction substantially opposite to the electron beam by the irradiation of the electron beam. Therefore, the "reflection" includes both electrons that are reflected without colliding with the sample 20 or the foreign matter 10, and two electrons that are reflected and collided with the sample 20 or the foreign matter 10 and are reflected and emitted twice. Moreover, the same applies to "release" and "produce".

Figure 149A shows an example of the relationship between the incident energy LE of the irradiated electron beam and the signal intensity/average shading of the reflected electron. In Fig. 149A, the horizontal axis represents the incident energy LE of the electron beam, and the vertical axis represents the signal intensity/average shading. The average darkness indicates the brightness of the image and corresponds to the signal strength. Figure 149 shows the characteristics of the incident energy LE in the vicinity of 0 [eV], which shows the characteristics of the energy band which is much lower than that of Fig. 148. In Fig. 149A, a region where the incident energy is LE = 10 [eV] or less is a region in which a signal (white) generated by the specular reflected electrons can be obtained. On the other hand, a region where the incident energy is LE=10 [eV] or more is a region in which a signal (black) generated by electron emission can be obtained twice. In the region where the electrons are reflected by the mirror, the incident energy LE decreases. The signal strength increases.

Fig. 149B shows an example different from Fig. 149A. Figure 149B also shows the relationship between the incident energy of the irradiated electron beam and the signal intensity/average shading of the reflected electrons. In Fig. 149B, the region where the incident energy is LE=5[eV] or less is the region where the signal (white) generated by the mirror electrons can be obtained, and the region where the incident energy is LE=5[eV] or more is the second emission. The area of the signal (black) produced by the electron.

The characteristic line of Fig. 149B and the characteristic line of Fig. 149A differ in the case where the signal of the specular reflected electron and the boundary of the signal of the second discharge of electrons have an incident energy LE of 5 [eV]. The boundary between the specular reflected electrons and the incident energy LE of the emitted electrons twice changes depending on the characteristics of the sample 20 or the change profile of the electron beam, and various values can be obtained. In the electron beam inspection method and the electron beam inspection apparatus of the present embodiment, an example of the case of Fig. 149A (an example of the incident energy LE of the boundary is 10 [eV]) will be described. However, it is not intended that the invention be limited thereto. As shown in Fig. 149B, the present invention is applicable to a case where the boundary incident energy is 10 [eV] or less, for example, the boundary incident energy may be 5 [eV].

Further, in FIGS. 149A and 149B, the region where the incident energy is below the boundary corresponds to the transition region of the present invention, and the form in which the mirror-reflected electrons and the two-shot electrons are mixed exists. Further, the region where the incident energy is above the boundary corresponds to the secondary electron emission region of the present invention. As described above, the boundary incident energy is 10 [eV] in the example of Fig. 149A, and 5 [eV] in the example of Fig. 149B.

Figure 150 shows the foreign matter 10 present on the sample surface 21 of the sample 20. status. As shown, electrons are generated by the illumination of the electron beam. When the incident energy is LE 10 [eV], the foreign matter 10 is negatively charged. When the electron beam is incident on the foreign matter 10, the electrons of the electron beam become the mirror reflection electron me. Thus, the electrons do not collide with the foreign matter 10, but reflect and reach the detector 70. On the other hand, in the normal portion (sample surface 21) where the foreign matter 10 does not exist, the electrons se are emitted twice by the irradiation of the primary electron beam.

Here, the term "two-time discharge electron se" means any one of secondary electrons, reflected electrons, and backscattered electrons. When these electrons are mixed, it is equivalent to "2 times of discharging electrons".

Regarding such a secondary electron emission, the emission rate η is usually small. In particular, when the incident energy LE is about 50 [eV] or less, the discharge rate is η &lt; 1.0. The closer the incident energy LE is to 0, the lower the release rate is. When the incident energy LE=0, the release rate is approximately zero.

In addition, the angle of electron emission is also distributed. For example, the secondary electrons are distributed according to the Cosine law. Therefore, the transmittance of electrons reaching the detector 70 is several tens or less in the imaging optical projection system.

On the other hand, the specular reflection electron me is the one in which incident electrons are reflected before the moment the foreign object 10 is hit. The mirror-reflecting electrons are reflected from the foreign matter 10 at an angle substantially symmetrical with the angle of the incident electron beam and are incident on the lens system of the secondary system. Therefore, the scattering or radiation distribution is small, and the mirror reflection electron me reaches the detector 70 at a transmittance of about 100%.

Fig. 151A shows an image 80 of the foreign matter 10 on the sample surface 21 obtained when the incident energy LE is 10 [eV] or less. Figure 151B shows the darkness value of image 80.

Refer to Figure 151A. In the images of the sample surface 21 and the foreign matter 10, the enlarged image 81 of the foreign matter 10 is indicated by a white area, and the surface image 82 of the sample surface 21 is indicated by a black area. In this case, the portion where the electrons are reflected by the mirror is obtained, and the brightness (darkness) is very high.

Fig. 151B is an example of the relationship between the direction cross-sectional position and the darkness value on the image 80 of the detector 70. The magnified image 81 of the foreign matter 10 is contained in the range of the y direction. As shown in Fig. 151B, the depth of the electrons reflected by the mirror is as high as about 3 times as compared with, for example, the portion where the mirror reflection electron me is not obtained. Therefore, high brightness and high S/N can be achieved.

In the example of Fig. 151B, the portion where the electrons are reflected by the mirror is displayed, and the darkness value DN of about 3 times is displayed as compared with the portion which is not obtained. However, the relationship between the degree of shading will vary depending on conditions and the like. The value of the darkness of the electron portion of the mirror can be obtained by about 2 to 10 times.

Fig. 152 shows a state in which the mirror 10 reflects the electrons me from the foreign matter 10 by the irradiation of the foreign matter 10 by the electron beam. Due to the shape of the foreign matter 10, a reflection point shift of the mirror reflection electron me or a non-uniformity of the charging voltage occurs. Therefore, the mirror reflection electron me will have a weak offset in the orbit and energy. As a result, when the mirror reflection electron me passes through the lens of the secondary system, the beam filter, or the like, the size of the signal region becomes large.

In Fig. 152, due to the influence of the surface potential of the foreign matter 10, the reflection direction of the mirror reflection electron me is radially expanded. As a result, on the signal of the foreign matter 10 reaching the detector 70, the signal specification is amplified larger than the optical magnification of the electro-optical system. The magnification is, for example, 5 to 50 times.

For example, suppose that there is a two-time system with an optical magnification of 100 times. About coming The secondary electrons from the foreign matter 10 are calculated in an ideal manner, and the signal size at the detector 70 is 100 times × 0.1 [μm] = 10 [μm].

On the other hand, the signal specification of the mirror reflection electron me of the foreign matter 10 can be amplified by, for example, 30 times. Therefore, the signal size incident on the detector 70 can reach 300 [μm]. This phenomenon is equivalent to an amplifying optical system which simply amplifies 100 [nm] (0.1 [μm]) to 300 [μm]. That is, a 3000x magnification optical system can be achieved. This means that a larger pixel size than the foreign object 10 can be used. If the foreign matter 10 is 100 [nm], the pixel size can be larger than 100 [nm], and a pixel specification of 300 to 1000 [nm] can be used.

By using a pixel size larger than the foreign matter of the object, the large area of the sample surface 21 of the sample 20 can be inspected once. Therefore, it is very effective in high-speed inspection. For example, the inspection speed of the pixel size 300 [nm] can be as high as 9 times as compared with the case where the pixel size is 100 [nm]. When the pixel size is 500 [nm], the inspection speed can be 25 times. That is, it is assumed that the conventional technique takes 25 hours for one inspection, and according to this embodiment, the inspection can be completed in one hour. On the other hand, the SEM method must be photographed with a pixel specification that is smaller than the foreign matter specification. This is because the SEM method is a method of forming a high-precision shape image and detecting the foreign matter by comparing the image with the normal portion.

In this way, the image projection optical system can not only increase the luminance difference (contrast) of the mirror reflection electron me and the secondary discharge electron se, but also achieve high speed.

Further, in the case where the incident energy is LE 10 [eV], pre-charging can be suitably used. The precharge system is realized by irradiating an electron beam for charging before photographing.

Pre-charging can be used to increase the charging voltage of the foreign matter 10. And, pre-charge It can be used to reduce the potential change of the foreign matter 10 when shooting. In the foreign matter inspection method, the amount of fluctuation in the charging voltage can be controlled by the incident energy LE1 of the charged beam. For example, foreign matter 10 having various specifications and various capacities exists. At this time, the charged foreign matter 10 charged to a certain voltage or lower can be detected by using mirror-reflected electrons. In addition, the trajectory of the reflected electrons of the mirror becomes appropriate by the difference between the charged voltage and the surrounding sample voltage. Thereby, a state in which the transmittance of the mirror-reflected electrons is high can be formed. This will be explained in detail later.

Next, a description will be given of a method of precharging. There are 3 ways to precharge.

[Precharge-1]

Fig. 153A and Fig. 153B are diagrams for explaining the first precharge mode (precharge-1). Here, the incident energy of the charged electron beam is set to LE1, and the incident energy of the photographing electron beam is set to LE2. The precharge-1 system sets the incident energy at LE2&lt;LE1, whereby the mirror reflection electrons are easily generated.

In Fig. 153A, a foreign matter 10 is present on the sample surface 21, and a charged electron beam having an incident energy of LE1 is irradiated thereto to perform precharging. The precharged incident energy LE1 is greater than the incident energy LE2 of the captured electron beam. Thereby, the charged voltage of the foreign matter 10 is made large, and at the time of photographing, electrons easily become mirror-reflected electrons. That is, by increasing the absolute value of the negative potential of the foreign matter 10, it is possible to form a reflection point of the electric field distribution by charging before the foreign matter 10 is instantaneous. Therefore, the incident imaging electron beam is reflected by the mirror reflection electron me before colliding with the foreign matter 10.

In Fig. 153B, the state in which the foreign matter 10 of the sample surface 21 has been irradiated with the electron beam is shown. In Figure 153B, the foreign body 10 is negatively charged, with There is a negative voltage potential distribution. The photographing electron beam has incident energy LE2 as described above. The incident electron is subjected to the surface potential of the foreign matter 10, does not collide with the foreign matter 10, and becomes a mirror reflection electron me to reflect before the collision. On the other hand, electrons se are emitted twice from the sample surface 21.

Thus, according to the configuration shown in FIGS. 153A and 153B, the incident energy LE1 of the charged electron beam can be set to be larger than the incident energy LE2 of the photographing electron beam. Thereby, it is suitable to generate the mirror reflection electron me from the photographing electron beam irradiated on the foreign matter 10, and to obtain the magnified image 81 of the foreign matter 10.

[Precharge-2]

Fig. 154 is a diagram for explaining the second precharge mode (precharge-2). In Precharge-2, the incident energy LE2 of the electron beam is set to be larger than the incident energy LE1 of the electron beam for charging. In the foreign matter inspection method, photographing can be performed while appropriately changing the potential at the time of shooting.

In Fig. 154, the horizontal axis represents the incident energy of the electron beam, and the vertical axis represents the surface potential of the foreign matter 10. The incident energy LE1 of the charged electron beam is smaller than the incident energy LE2 of the captured electron beam. The surface potential of the foreign matter 10 varies between LE1 and LE2. As shown, the potential difference ΔV is small.

The precharge-2 of Fig. 154 is suitable for the case where the incident energy LE2 of the appropriate electron beam is predicted at the time of shooting. When the photographing electron beam of the appropriate incident energy LE2 is simply used for photographing, the surface potential of the foreign matter 10 fluctuates during photographing, and the possibility of correcting the magnified image 81 is not obtained. This situation is avoided in pre-charge-2. The precharge-2 is configured to control the surface potential of the foreign matter 10 through precharging until the value near the optimum point. until. In this way, the potential change ΔV of the surface potential of the foreign matter 10 can be small at the time of shooting.

[Precharge-3]

Fig. 155 is a diagram for explaining the third precharge mode (precharge-3). In precharge-3, the incident energy LE1 of the charged electron beam is set to be equal to the incident energy LE2 of the photographing electron beam. Moreover, the dose of the electron beam for charging is different from that of the photographing electron beam. In Fig. 155, the horizontal axis represents the irradiation dose, and the vertical axis represents the surface potential of the foreign matter 10.

The pre-charging -3 pair stabilizes the charged voltage of the foreign matter 10, and is effective for achieving stable shooting and sensitivity. Fig. 155 shows that the surface potential of the foreign matter 10 varies depending on the irradiation dose. Precharge is performed in order to give a dose D1 close to the necessary dose. Thereafter, the dose D2 was again administered and photographing was performed. This composition is very effective. Thereby, the potential variation ΔV at the surface of the foreign matter photographed at the dose D2 can be suppressed to be small. Therefore, stable image quality (shape, focal length, etc.) can be achieved.

In the three types of precharging of FIGS. 153 to 155, the beam source of the precharged electron beam can be the same as the beam source of the photographing electron beam, and the conditions of the beam source are performed in order to perform the above precharging. Can be controlled. In addition, the pre-charging unit for pre-charging can also be set independently. Thereby, the inspection throughput can be increased.

The precharge unit may use a cathode composed of, for example, LaB6, a hot wire, a hollow cathode, a carbon nanotube, or the like. The precharge unit may also use a Veneta pole for extracting an electron beam, or an extraction electrode or an illumination area control lens. The beam specification of the pre-charging unit and the beam gauge normally illuminated in the cylinder system The grid is equal or slightly larger. The incident energy of the electron beam is determined by the voltage difference between the cathode and the sample. For example, it is assumed that a negative voltage of -3000 [V] is applied to the sample 20, and the incident energy of the electron beam is set to 10 [eV]. At this time, if a cathode voltage of -3010 [V] is applied to the cathode, an electron beam can be generated.

"Other inspection methods (in the case of LE&gt;10[eV])"

Figure 156 shows an image 80a taken by the detector 70 when the incident energy LE of the electron beam is greater than 10 [eV]. In Fig. 156, the enlarged image 81a of the foreign matter 10 is indicated by a black signal, and the surface image 82a of the sample 20 is indicated by a white signal.

Figs. 157A to 157C show a case where the discharge electrons 2 are discharged from the foreign matter 10 by the irradiation of the photographing electron beam.

Fig. 157A shows the behavior of discharging the electron se twice in a state where the foreign matter 10 is charged and the potential difference between the foreign matter 10 and the surrounding sample surface 21 is increased. In Fig. 157A, the foreign matter 10 is negatively charged, and the track from the foreign matter 10 which emits the electron se twice has been bent. Therefore, the transmittance (the ratio of electrons reaching the detector 70) is extremely lowered. As a result, in the observed image, the brightness of the foreign matter portion is lower than that of the surroundings, that is, the foreign matter 10 is detected as a black signal.

Fig. 157B shows the behavior in the state where the potential difference of the foreign matter 10 and the surrounding sample surface 21 is small in two times. In Fig. 157B, since the potential difference between the foreign matter 10 and the periphery is small, electrons are generated almost uniformly from the foreign matter 10 or from the sample surface 21. Therefore, the foreign matter 10 is difficult to distinguish from the surroundings. That is, it is difficult to detect the foreign matter 10 from the acquired image. This situation can be avoided. So even from foreign objects 10 When the electron se is released twice, it is suitable to charge the foreign matter 10 by irradiation with a charged electron beam. After charging, the detection of the foreign matter 10 can be facilitated by the irradiation of the photographing electron beam as described above.

Figure 157C shows the behavior of discharging the electron se twice in the positively charged region. In the positively charged region, once the electrons se are released by the foreign matter 10, the upward orbit is formed. As shown in the figure, due to the influence of the potential distribution of the foreign matter 10, the orbit of the electrons se is released twice, and the number of electrons reaching the detector 70 is reduced. This phenomenon is the same as in Figure 157A. Therefore, even when it is positively charged, the same phenomenon can be observed, and the enlarged image 81a of the foreign matter 10 can be obtained in the form of the image of the black signal.

Further, in the foreign matter inspection method and the foreign matter inspection device of the present embodiment, in order to further increase the throughput, an electron beam projection method is used. By using the imaging optical system, the electrons se or the mirror-reflecting electrons me are used twice from the surface 21 of the sample, so that foreign matter such as wafers and masks can be detected with high speed and high throughput, and it is suitable for implementing, for example, a sample. Foreign matter detection after washing. As described above, since the detection signal from the foreign matter 10 can be amplified to be larger than the optical magnification, the signal of the ultra-fine foreign matter 10 can be obtained in a large pixel size, whereby high-speed and high-detection throughput can be realized.

For example, the specification of the foreign matter signal can be scaled up to 5 to 50 times the actual specification. It is possible to apply a pixel specification of 3 times or more of the foreign matter specification of the detection target. This is effective for a foreign matter 10 having a specification of 50 to 100 [nm] or less. The foreign matter 10 of such a specification is difficult to detect by light. In addition, the SEM method must use a pixel specification that is smaller than the foreign material specification. Therefore, when a small foreign matter is to be detected, the detected throughput is significantly lowered. According to this embodiment In the electron beam inspection method, the foreign matter 10 on the wafer in the middle of the process can be detected at a high speed by using the image projection method. Further, the foreign matter 10 can be reliably detected by the obtained magnification images 81 and 81a.

"Electronic inspection device"

Fig. 158 is a view showing the configuration of an electron beam inspection apparatus to which the present invention is applied. In the above, the principle part of the foreign matter inspection method is mainly described. Here, a foreign matter inspection device that is applied to the above-described foreign matter inspection method will be described. Therefore, all of the above-described foreign matter inspection methods can be applied to the foreign matter inspection device described below.

The object to be inspected by the electron beam inspection apparatus is the sample 20. The sample 20 is a tantalum wafer, a glass mask, a semiconductor substrate, a semiconductor pattern substrate, or a substrate having a metal film. The electron beam inspection apparatus of this embodiment is for detecting the presence or absence of foreign matter 10 on the surface of the sample 20 formed of the substrates. The foreign matter 10 is an insulator, a conductive material, a semiconductor material, a composite of these, or the like. The type of the foreign matter 10 is particles, a cleaning residue (organic matter), a reaction product on the surface, and the like. The electron beam inspection device may be an SEM mode device or a map projection device. In this example, the present invention is applied to a projection projection inspection apparatus.

The electron beam inspection apparatus of the image projection system includes a primary optical system 40 for generating an electron beam, a sample 20, a stage 30 for setting a sample, and electrons or mirror reflection electrons from the sample twice. a secondary optical system 60 for magnifying image formation; a detector 70 for detecting these electrons; an image processing device 90 (image processing system) for processing signals from the detector 70; an optical microscope for aligning positions 110; and detailed inspection SEM120. In the present invention, the detector 70 can be included in the secondary optical system 60. Further, the image processing device 90 may be included in the image processing unit of the present invention.

The primary optical system 40 is configured to generate an electron beam and irradiate the sample 20. The primary optical system 40 has an electron gun 41, a lens (42, 45), a diaphragm (43, 44), an E x B filter 46, lenses (47, 49, 50), and a diaphragm 48. An electron beam can be generated by the electron gun 41. Lenses 42, 45 and apertures 43, 44 are used to shape the electron beam while controlling the direction of the electron beam. In addition, in the E x B filter 46, the electron beam is affected by the magnetic field and the electric field. The electron beam is incident from the oblique direction into the E x B filter 46 and is deflected vertically downward toward the sample 20 side. The lenses 47, 49, 50 control the direction of the electron beam and perform appropriate deceleration to adjust the incident energy LE.

The primary optical system 40 irradiates the electron beam toward the sample 20. As described above, the primary optical system 40 is an irradiation of both the charged electron beam and the captured electron beam that are precharged. According to the experimental results, the difference between the precharged incident energy LE1 and the incident electron beam LE2 of the photographing electron beam is preferably 5 to 20 [eV].

In this regard, when there is a potential difference between the foreign matter 10 and the surroundings, it is irradiated as a precharged incident energy LE1 in the negatively charged region. The charging voltage will vary depending on the value of LE1. Because the relative ratio of LE1 and LE2 will change (as mentioned above, LE2 is the incident energy of the electron beam). When LE1 is large, the charging voltage becomes high, whereby the position above the foreign matter 10 (closer to the position of the detector 70) forms a reflection point. According to the position of the reflection point, the orbit and transmittance of the mirror reflected electrons change. Thus, according to the reflection point, the most Good charging voltage conditions are determined. In addition, when LE1 is too low, the formation efficiency of mirror-reflected electrons is lowered. The present inventors have found that the difference between LE1 and LE2 is preferably from 5 to 20 [eV]. Further, the value of LE1 is preferably from 0 to 40 [eV], more preferably from 5 to 20 [eV].

Further, in the primary optical system 40 of the image projection optical system, the E x B filter 46 is particularly important. By adjusting the conditions of the electric and magnetic fields of the E x B filter 46, the electron beam angle can be determined once. For example, the condition of the E x B filter 46 can be set such that the electron beam of the primary system and the electron beam of the secondary system are incident substantially perpendicularly to the sample 20. In order to further increase the sensitivity, for example, it is effective to tilt the electron beam of the primary system to the incident angle of the sample 20. 0.05 to 10 degrees is a suitable inclination angle, preferably about 0.1 to 3 degrees.

In Fig. 159, the foreign matter 10 present on the surface 21 of the sample is irradiated with the system electron beam once. The tilt angle of the electron beam is θ. The angle θ may be, for example, in the range of ±0.05 to 10°, and preferably in the range of ±0.1 to ±3°.

Thus, by irradiating the electron beam with respect to the foreign matter 10 with a predetermined angle θ, the signal from the foreign matter 10 can be reinforced. Thereby, it is possible to form a condition that the orbit of the mirror-reflecting electrons does not shift from the center of the optical axis of the secondary system. Therefore, the transmittance of the mirror reflected electrons can be improved. Therefore, when the foreign matter 10 is charged and the mirror is guided to reflect electrons, it is very advantageous to use a tilted electron beam.

Go back to Figure 158. The platform 30 is a means for placing the sample 20, and is movable in the horizontal direction and the θ direction of xy. In addition, the platform 30 can also be moved in the z direction as needed. The surface of the platform 30 can also be set with an electrostatic chuck. Sample fixing mechanism.

The platform 30 has a sample 20 on which a foreign matter 10 is placed. The primary optical system 40 irradiates an electron beam onto the sample surface 21 at an incident energy of LE-5 to -10 [eV]. The foreign matter 10 is charged, and the incident electrons of the primary optical system 40 are bounced back without touching the foreign matter 10. Thereby, the mirror reflected electrons are guided to the detector 70 through the secondary optical system 60. At this time, the electron emission system is discharged twice from the sample surface 21 in the expanding direction. Therefore, the transmittance of electrons emitted twice is a low value, for example, about 0.5 to 4.0%. On the other hand, the direction of the mirror-reflected electrons does not form a scattering, so that the mirror-reflected electrons can achieve a high transmittance of approximately 100%. The mirror-reflected electrons are formed in the foreign matter 10. Therefore, only the signal of the foreign matter 10 can generate high luminance (a state in which the number of electrons is large). The ratio of the difference in brightness between the emitted electrons and the surrounding two times becomes large, and high contrast can be obtained.

Further, as described above, the image of the mirror-reflected electron can be amplified at a magnification larger than the optical magnification. The magnification is 5 to 50 times. In typical conditions, the magnification is often 20 to 30 times. At this time, even if the pixel size is three times or more of the foreign matter specification, the foreign matter can be detected. Thus, high speed and high throughput can be achieved.

For example, when the size of the foreign matter 10 is 20 [nm], the pixel size may be 60 [nm], 100 [nm], 500 [nm], or the like. As shown in this example, it is possible to perform photographing and inspection of foreign matter using a pixel size of three times or more of a foreign matter. Compared with the SEM method and the like, this method is a remarkably excellent feature for the purpose of improving throughput.

The secondary optical system 60 guides the electrons reflected from the sample 20 to the inspection The means of the detector 70. The secondary optical system 60 has a lens (61, 63), an NA diaphragm 62, an aligner 64, and a detector 70. The electrons are reflected from the sample 20 and passed through the objective lens 50, the lens 49, the aperture 48, the lens 47, and the E x B filter 46. The electrons are then directed to the secondary optical system 60. In the secondary optical system 60, electrons are collected by the lens 61, the NA stop 62, and the lens 63. The electrons are shaped at the aligner 64 and detected by the detector 70.

The NA stop 62 has a function of specifying the transmittance aberration of the secondary system. The specification and position of the NA aperture 62 can be selected to increase the difference between the signal from the foreign object 10 (mirror reflected electrons, etc.) and the surrounding (normal portion) signal. Alternatively, the size and position of the NA stop 62 can be selected to increase the ratio of the signal from the foreign object 10 to the surrounding signal. Thereby, S/N can be improved.

It is assumed that the NA stop 62d can be selected in the range of, for example, φ 50 to φ 3000 [μm], and it is assumed that the detected electrons are mixed with the mirror reflection electrons and the second discharge electrons. In order to increase the S/N of the mirror reflection electron image under such conditions, it is advantageous to select the aperture specification. At this time, it is suitable to select the specification of the NA stop 62 to reduce the transmittance of the electrons emitted twice, and to maintain the transmittance of the reflected electrons of the mirror.

For example, when the incident angle of the primary electron beam is 3°, the reflection angle of the mirror reflection is about 3°. In this case, it is suitable to select the specification of the NA stop 62 to which the orbit of the mirror reflection electron can pass. For example, a suitable specification is φ 250 [μm]. Since it is limited to the NA aperture (diameter φ 250 [μm]), the transmittance of electrons emitted twice will decrease. Thus, the S/N of the mirror-reflected electronic image can be increased. For example, when the pupil diameter is set from φ 2000 to φ 250 [μm], The background level (noise level) can be reduced to less than 1/2.

The foreign matter 10 may be composed of any kind of material such as a semiconductor, an insulator, a metal, or the like. Figs. 160A and 160B show a foreign matter 10a of a metal material on the surface 21 of the sample. Fig. 160B is an enlarged view of the metal material foreign matter 10a. In Fig. 160B, the foreign matter 10a may be a metal or a semiconductor or the like, or a mixture of these. As shown in the figure, the natural oxide film 11 or the like is formed on the surface of the foreign matter, so that the foreign matter 10 is covered with the insulating material. Therefore, even if the material of the foreign matter 10 is metal, charging occurs due to the oxide film 11. This type of charging is well suited for use in the present invention.

Go back to Figure 158. The detector 70 is a means for detecting electrons guided by the secondary optical system 60. The detector 70 has a plurality of pixels on its surface. The detector 70 can be applied to various types of two-dimensional sensors. For example, the detector 70 can apply a CCD (Charge Coupled Device) and a TDI (Time Delay Integration)-CCD. These devices are sensors that perform signal detection after converting electrons into light. Therefore, means such as photoelectric conversion is required. Therefore, electrons can be converted into light using a photoelectric conversion or an attenuator. The light image information passes the light to the detected TDI, in which way the electrons are detected.

Here, an example in which the detector 70 applies EB-TDI will be described. EB-TDI does not require a photoelectric conversion mechanism optical transmission mechanism. The electronics are directly injected into the EB-TDI sensor surface. Therefore, without deterioration of the resolution, a high MTF (Modulation Transfer Function) and contrast can be obtained. In the prior art, the detection of small foreign matter 10 is unstable. In contrast, if EB-TDI is used, the S/N of the weak signal of the small foreign object 10 can be mentioned. high. Thus, higher sensitivity can be obtained. The S/N increase will be 1.2 to 2 times.

Furthermore, in addition to EB-TDI, an EB-CCD can also be provided. EB-TDI and EB-CCD can be exchanged and can be switched at will. It is also effective to use this configuration. For example, the usage method shown in Fig. 161 can be applied.

Figure 161 shows the EB-TDI72 and EB-CCD71 switchable detectors 70. The two sensors can be exchanged according to the application, and both sensors can be used.

In Fig. 161, the detector 70 includes an EB-CCD 71 and an EB-TDI 72 provided in a vacuum container 75. EB-CCD71 and EB-TDI72 are electronic sensors that accept electron beams. The electron beam e is directly incident on the detection surface. In this configuration, the EB-CCD 71 is used to adjust the optical axis of the electron beam, and is used for adjustment and optimization of image capturing conditions. On the other hand, when the EB-TDI 72 is used, the EB-CCD 71 moves to a position away from the optical axis through the moving mechanism M. Then, the conditions obtained by applying the EB-CCD 71 are used or referenced, and the EB-TDI 72 is used for photographing, and the image is used for evaluation or measurement. Further, the moving mechanism M is not only moved in the direction (X direction) of the EB-CCD 71, but also configured to be movable in three axes (for example, X, Y, and Z directions), or may be configured to center the EB-CCD 71 with respect to the electro-optical system. The center of the optical axis is finely adjusted.

In the detector 70, foreign matter detection of the semiconductor wafer can be performed using the EB-TDI 72 using or referring to the electro-optical conditions obtained by applying the EB-CCD 71.

After EB-TDI72 for foreign body inspection, EB-CCD71 can be used. For detailed inspection, defects can be evaluated for foreign object types or foreign material specifications. In EB-CCD71, it is possible to accumulate images. Through accumulation, noise can be reduced. Therefore, detailed inspection shooting of the defect detection portion can be performed with high S/N. Moreover, the pixels of EB-CCD71 are smaller than the pixels of EB-TDI72. That is, the number of pixels of the imaging element can be increased with respect to the signal specification amplified by the image projection optical system. Therefore, an image with higher resolution can be obtained. This image is used for inspection and classification determination of defect types and the like.

The EB-TDI 72 has a configuration in which pixels are arranged two-dimensionally, and has a rectangular shape, for example. Thereby, the EB-TDI 172 can directly receive the electron beam e to form an electronic image. The pixel specification is, for example, 12 to 16 [μm]. On the other hand, the pixel specification of the EB-CCD 71 is, for example, 6 to 8 [μm].

Further, EB-TDI 72 is in the form of a package. The package itself is used to achieve the effect of a feed-through FT. A pin 73 of the package is connected to the camera 74 on the atmospheric side.

The configuration shown in Fig. 161 can eliminate various disadvantages. The disadvantages of the release are the optical conversion loss caused by FOP (Fiber Optical Plate), optical glass for sealing, optical lens, etc., aberration and distortion during light transmission, and thus deterioration of image resolution. Poor detection, high cost, large size, etc.

Fig. 162A and Fig. 162B are diagrams for explaining the method of determining the electron beam orbit condition with good efficiency. This method is very effective in obtaining a mirror-reflected electron image. The electron beam orbital conditions are the lens conditions of the lenses (42, 45, 47, 49, 50, 61, 63) of the primary optical system 40, the secondary optical system 60, and the aligner conditions of the aligner 64.

Fig. 162A shows a configuration in which a laminated structure of a polycrystalline germanium layer 23 and a ceria film 24 is provided on the sample surface 21 of the tantalum substrate sample 20. The seam of the laminated structure is formed with a groove 25. In Fig. 162B, a ceria layer 24a is formed on the sample surface 21 of the ruthenium substrate sample 20. The seam of the layer is formed with a groove 25a.

Figure 162A shows the signal intensity distribution map of the mirror reflection electron me. The incident energy is set in a region where the mirror reflection electron me is generated, and the orbit of the incident electron is easily bent, and the mirror reflection electron me is easily generated at the edge portion 26 of the pattern, so that the signal intensity of the edge portion 26 of the groove 25 is increased.

Figure 162B shows the orbit of the electron beam EB incident, which is reflected by the mirror reflection electron me. The electrons are incident on the sample 20, and are reflected by the one edge portion 26a to travel substantially horizontally, and then moved to the opposite side of the groove 25a, and are reflected and raised at the edge portion 26a on the opposite side. Thus, mirror-reflecting electrons are likely to occur at the edge portion of the groove 25a.

This phenomenon is particularly remarkable in the case of a concave symmetrical configuration. The symmetrical configuration is, for example, a Faraday cup or a cross groove configuration. At this time, the symmetry of the mirror-reflected electrons generated at the edge portions 26, 26a affects the resolution of the image. In order to make the difference in the degree of darkness between the two edges of the image to be less than ±5%, the symmetry of the degree of darkness can be achieved. The degree of darkness is the brightness of the image, and the difference in the degree of darkness is the difference in brightness. When the lens conditions and aligner conditions are adjusted to achieve this symmetry, the lens and aligner conditions using mirror-reflecting electrons can be optimized. Thereby, a mirror-reflected electronic image excellent in resolution can be realized. Compared with when the adjustment method is not used, the S/N can be increased by 10 to 30%, and the adjustment time can be shortened by about 10 to 5%.

Figure 163 is a side cross-sectional view showing the Faraday cup 31. The Faraday cup 31 is provided with an opening 32 of a conductor and a metal electrode 33 of a cup shape. The Faraday cup 31 uses an ammeter 34 to measure the amount of electrons passing through the opening 32. The opening 32 may have a size of, for example, about 30 [μm] in diameter. Since the Faraday cup 31 has a groove shape, it is easy to generate mirror-reflecting electrons at the edge portion as described above. Thus, the Faraday Cup 31 can be helpful for making adjustments.

Next, an example in which the foreign matter inspection method of the present invention is applied to the foreign matter inspection device of Fig. 158 will be described.

The aforementioned Fig. 148A shows the correlation of "2 electron yield (Yield)" - "incident energy LE". This correlation shows the mechanism (Machanism) for detecting foreign matter 10 using an electron beam of LE &gt; 10 [eV]. According to the incident energy LE irradiated to the foreign matter 10, the secondary electron emission rate will be different. Therefore, a negatively charged state and a positively charged state are formed. For example, the insulator is SiO₂The following charged states are visible.

50[eV]LE: negatively charged

50&lt;LE1500[eV]: positively charged

1500[eV]&lt;LE: negatively charged

In either case, when the foreign matter 10 is charged, the foreign matter and the potential around it become different values, and the potential distribution around the foreign matter also forms a distorted state. The electric field of the distortion greatly bends the secondary electron orbits from the foreign matter 10 to lower the transmittance. Therefore, the number of electrons reaching the detector from the foreign matter becomes extremely small as compared with the periphery of the foreign matter. Therefore, the brightness of the foreign matter is smaller than the surrounding (black signal), and the foreign matter 10 can be detected with high contrast. The specification of the black signal of the foreign object can be enlarged to be larger than the optical magnification. Can capture 5 to 20 times magnification Foreign object signal. This phenomenon and detection can also be achieved in the above three energy regions.

Next, an example of an electron beam column system using an image projection method of an electron beam will be disclosed. The sample 20 may be a wafer, an exposure mask, a recording medium, or the like. When the sample is a wafer, the circuit pattern in the middle of the LSI process can be formed on the wafer of 8 to 12 Å. Moreover, the wafer can also be unpatterned. The wafer can be in a patterned state after film formation. Further, the wafer may be in a state of being subjected to planarization after polishing or CMP. Further, the wafer may be a Si substrate in a state before the treatment such as film formation.

The sample 20 is placed on the control platform 30 of x, y, and θ. The electron beam is emitted from the electron gun 41, and the beam irradiation region and the irradiation energy are controlled by the lens 42, the aperture (43, 44), the quadrupole lens 45, the E x B filter 46, and the like, and the electron beam is irradiated. The surface of the sample. The beam diameter is, for example, φ 300 [μm] (or a circle of about 270 × 80 [μm]). The imaging optical system images the image of the emitted electrons from the sample surface 21 at a magnification of 50 to 500 times at the detector 70. Sample 20 was applied with a negative voltage. The main surface potential of the first lens 50 of the primary optical system 40 is positive. Therefore, a positive electric field is formed in the vicinity of the sample 20. The positive electric field can be, for example, 1 to 3 [kV/mm]. The detector 70 is composed of an MCP (Micro Channel Plate), a fluorescent plate, a FOP (Fiber Optical Plate), and a TDI. (For the internal structure, see, for example, Figure 219). The MCP system doubles the amount of detected electrons, and the fluorescent plate converts the electrons into optical signals. The two-dimensional optical signal is transmitted through the FOP, and an image is formed by the TDI sensor, and the signal is detected. In the case of using TDI, the sample is continuously moved while acquiring a two-dimensional image signal. thus The image signal can be obtained at high speed. The image processing mechanism processes the signal from the TDI and performs electronic image formation, foreign matter detection, and foreign matter classification discrimination.

Next, an example of performing inspection of the foreign matter 10 on the sample 20 using such an electron beam cylinder system will be described. The incident energy LE of the primary system electron beam irradiated to the sample 20 is set at 2 [eV]. The incident energy LE is the difference between the cathode voltage of the electron gun 41 of the primary optical system 40 and the sample voltage (applied voltage). The foreign matter 10 is charged by the irradiation of the electron beam. Then, only the beam irradiated to the foreign matter 10 forms a mirror-reflecting electron. The mirror reflected electrons are directed to the detector 70 by the secondary optical system 60. In the normal portion where no foreign matter 10 is present, electrons emitted by the beam irradiation are also guided to the detector 70. The electrons emitted twice are secondary electrons, reflected electrons, or backscattered electrons. It can also be a mixture of these electrons.

Here, the closer the LE is to 0, the lower the release rate η of the electrons released twice. Moreover, the direction in which the electrons are emitted from the surface is shown to be distributed (the distribution of the second electrons follows, for example, the law of cosine). Therefore, when the design calculation of the secondary electron emission to the detector 70 is performed by the secondary optical system 60, the arrival rate of the electrons discharged twice is about several %. Accordingly, the arrival rate of the mirror-reflected electrons is high, and the electron arrival rate and the release rate of the surrounding portions are low. Therefore, a relatively large electron ratio is generated, that is, a luminance difference is generated. Thus, a large contrast ratio and S/N can be obtained. For example, when the pixel size is 100 [nm] and the diameter of the foreign matter 10 is φ = 20 [nm], the S/N can be 5 to 10. Usually S/N3, it can be fully tested and inspected. Therefore, according to the present invention, the inspection of the extremely small foreign matter 10 like the above example can be made larger than the foreign matter specification. The pixel specifications are implemented.

An example of using a precharged charged electron beam in the above device system will be described.

LE1 is the incident energy of the precharged charged electron beam, and LE2 is the incident energy of the electron beam at the time of shooting and inspection. Under the conditions of LE1=14[eV] and LE2=1[eV], the foreign matter 10 of the insulator can be inspected with good efficiency. Si, SiO₂Foreign matter 10 on the surface of the film, metal film, SOI, glass mask, etc. can be inspected. In this procedure, the electron beam for charging can be irradiated to the inspection area by LE1 = 14 [eV]. Next, the electron beam is irradiated with LE2=1[eV], and the photographing and inspection of the foreign matter 10 is performed. The implementation of this step depends on how much time the precharge effect can be maintained. In general, the time for maintaining the precharge effect is about 10 to 30 hours, and in some cases, 150 hours or more, without applying a static elimination treatment or the like.

The effect of forming mirror-reflected electrons can be increased in comparison with the case where such pre-charging is performed and the case where pre-charging is not performed. Moreover, S/N can be increased by about 3 to 10 times.

When the incident energy is LE 10 [eV], particularly in the region of LE 0 [eV], mirror-reflected electrons can be formed in the normal portion. Even if the condition has been set, the present invention can form that the mirror-reflected electrons from the foreign matter 10 reach the detector 70, and the mirror-reflected electrons of the normal portion do not reach the state of the detector 70, and the foreign matter 10 can be inspected with a high S/N. . More specifically, the sample surface 21 is flat and the electron beam is incident substantially perpendicularly. The incident beam of the normal portion is decelerated on the sample surface 21. Therefore, the orbit of the electrons is bent and deviated from the center of the secondary optical system 60. As a result, this phenomenon will make it positive The number of electrons that are normally directed to the detector 70 is reduced. On the other hand, the mirror-reflected electrons from the foreign matter 10 rise from the curved surface or the inclined surface of the foreign matter 10, and are guided to the detector 70 through the track near the center of the secondary optical system 60. Thus, the mirror-reflected electronic signal from the foreign object 10 is directed to the detector at a high transmittance. Moreover, a high S/N can be achieved. In this regard, it will be explained in detail using FIG.

Fig. 164 is an explanatory diagram of filtering action when mirror-reflecting electrons are generated from the foreign matter 10 and the normal portion of the periphery. In Fig. 164, the electron beam is irradiated in a state where the foreign matter 10 is present on the sample 20, and mirror-reflected electrons are generated by being reflected from both the foreign matter 10 and the sample surface 21. In this case, the present invention causes the mirror-reflected electrons reflected from the foreign matter 10 to reach the detector 70, and the mirror-reflected electrons from the sample surface 21 of the normal portion do not reach the detector 70. That is, the foreign matter 10 is charged, and a potential difference is generated between the foreign matter and the surrounding normal portion (the sample surface 21). Thereby, the specularly reflected electrons from the foreign matter 10 and the specularly reflected electrons from the sample surface 21 of the normal portion can be separated.

For example, as illustrated in Fig. 159, the incident angle of the primary electron beam is slightly inclined from the vertical and deviated from the center. Thereby, the condition that the orbit of the mirror-reflected electron passes through the vicinity of the center of the secondary optical system 60 can be formed. In the flat normal portion, the orbit of the mirror-reflecting electrons is deviated, and the orbit of the mirror-reflecting electrons from the normal portion is deviated from the central portion of the secondary optical system 60, and as a result, the number and probability of electrons reaching the detector 70 are lowered. . Alternatively, the mirror-reflected electrons from the normal portion may become stray electrons or the like due to collision with the cylinder of the secondary optical system 60. Thus, the foreign matter 10 and the surrounding sample surface 21 There is a difference in the number of electrons or electron density reaching the detector 70. Thereby, a large degree of darkness (i.e., contrast) can be formed.

At this time, the main factors affecting the deviation of the track are the intensity of the lens (47, 49, 50, 61, 63), the focus, and the E x B filter 46 and the NA stop 62. Regarding the lenses 47, 49, 50, 61, 63, the focus and intensity are adjusted, and the condition that the mirror-reflected electron orbit from the foreign matter 10 passes through the center of the secondary optical system 60 is obtained. The mirror incident electrons from the normal portion (sample surface 21) and the mirror-reflected electrons from the foreign matter 10 have different lens incident angles and focal points. Therefore, the mirror-reflecting electrons from the normal portion pass through the orbits deviating from the center of the secondary optical system 60. Furthermore, the NA stop 62 will block the mirror reflection electrons passing through the off-center track, thereby reducing the amount of arrival and arrival probability of the electrons toward the detector 70. Moreover, when the mirror-reflected electrons pass through the E x B filter 46, the E x B filter 46 is adjusted so that the mirror-reflected electrons from the foreign matter 10 pass through the track of the rear-stage NA stop 62 and the detector 70. Thereby, the mirror-reflected electrons are appropriately adjusted as they pass through the E×B filter 46. In the mirror electrons from the foreign matter 10 and the mirror electrons from the surrounding normal portion (sample surface 21), the incident angle to the E×B filter 46 and the energy in the axial direction (Z-axis direction) are not the same. . Therefore, the mirror-reflected electrons reflected from the sample surface 21 of the normal portion deviate from the center of the NA pupil 62, the lenses 61, 63 in the subsequent stage. Thus, the probability of being incident on the detector 70 is reduced.

The LE area that can be effectively used is usually -30 to 0 [eV]. However, when the angle between the optical axis of the secondary optical system 60 and the sample surface is vertically shifted, even if LE is 0 [eV] or more, mirror-reflected electrons may be formed. and, In the sample having fine irregularities on the surface of the patterned wafer or the like, even if LE is 0 [eV] or more, mirror-reflected electrons may be formed. For example, in the LE area of -30 to 10 [eV], there is a possibility of forming such a situation.

Further, by effectively utilizing the precharge, the electron beam inspection method of the present invention can be applied even in the SEM. For example, even in the SEM, foreign matter inspection can be performed by performing pre-charging under the following conditions and performing photographing and inspection.

Precharge LE1: 0 to 30 [eV]

Shoot LE2: -5 to 20 [eV]

For example, shooting is performed under the conditions of pre-charging LE1=25 [eV] and shooting LE2=5 [eV]. Then, the foreign matter (insulator, or an object containing the insulator) is charged, and the surface potential becomes negatively charged (for example, -7 V). Then, the photographing electron beam (LE2 = 5 [eV]) is irradiated. Thereby, mirror-reflected electrons are formed only at the charged portion of the foreign matter, and the mirror-reflected electrons can be obtained at the detector 70. In the normal portion where no foreign matter 10 is present, electrons are emitted twice (two electrons are emitted, two electrons are reflected, and electrons are scattered, or a mixture of these electrons may be used). Since the emission rate of the electrons discharged twice is low, the brightness of the normal portion is low. The difference in brightness (contrast) between the specular reflected electrons of the foreign matter 10 and the secondary emitted electrons of the normal portion is large, so that the foreign matter 10 can be detected with high sensitivity.

In order to perform pre-charging efficiently, a pre-charging device can be provided before the imaging unit.

Further, when pre-charging is not performed in the SEM method, it is conceivable that the following disadvantages occur. Usually, in the SEM formula, in order to properly pattern or The image formation and shape change of the foreign matter 10 are set, and the spot size of the electron beam is set to be smaller than the object size of the pattern defect or the foreign matter specification to be detected. Therefore, the locality of the foreign matter 10 and the temporally charged potential change occur due to the difference between the spot size of the beam and the foreign matter specification. Therefore, a stable signal cannot be obtained, or it is difficult to obtain a stable mirror reflection electron. Therefore, it is important to stabilize the surface potential state of the foreign matter 10 by pre-charging or to stabilize the charged state and potential of the foreign matter 10, and then to perform photographing.

Further, in the conventional SEM type, since the beam scanning is performed, the incident angle of the beam with respect to the sample 20 largely changes depending on the scanning position. When a beam of mirror-reflected electrons is formed, the angle of reflection of the beam will vary depending on the angle of incidence. As a result, there is a disadvantage that the probability of electrons entering the detector 70 is greatly different depending on the scanning position. Therefore, it is difficult to obtain an image that is uniform and excellent in accuracy. In order to overcome this deficiencies, the alignment of the aligner and the lens voltage is appropriately adjusted so that the incident angle of the electron beam with respect to the sample is nearly vertical.

Thus, the electron beam inspection method of the present invention can be applied to the SEM type through appropriate setting conditions.

Figure 165 shows an electron beam inspection apparatus to which the present invention is applied. Here, an example of a holistic system configuration will be described.

In Fig. 165, the foreign matter inspection device includes a sample carrier 190, a small environment device 180, a load lock chamber 162, a transport chamber 161, a main chamber 160, an electron beam column system 100, and an image processing device 90. The small environment device 180 is provided with an atmospheric transfer robot, a sample alignment device, a clean air supply mechanism, and the like. The transport chamber 161 is provided with a transport robot in a vacuum. Always set Since the robot is disposed in the transfer chamber 161 in a vacuum state, particles or the like generated due to pressure fluctuations can be minimized.

The main chamber 160 is provided with a platform 30 that moves in the x-direction, the y-direction, and the θ (rotation) direction, and an electrostatic chuck is disposed above the platform 30. The sample 20 itself is placed on an electrostatic chuck. Alternatively, the sample 20 is held in the electrostatic chuck in a state of being placed on a tray or a jig.

The main chamber 160 is controlled by the vacuum control system 150 to maintain the chamber in a vacuum state. Further, the main chamber 160, the transfer chamber 161, and the load lock chamber 162 are placed on the vibration isolation table 170 so that vibration from the floor is not transmitted.

Furthermore, the main chamber 160 is provided with an electron column 100. The electron column 100 includes a primary optical system 40 and a cylindrical cylinder of the secondary optical system 60, and a detector 70 for detecting secondary electron emission or mirror reflection electrons from the sample 20. The signal from detector 70 is passed to image processing device 90 for processing. Both on-time signal processing and off-time signal processing can be performed. The on-time signal processing is performed during the inspection. When the off-time signal processing is performed, only the image is acquired, and the signal processing is performed afterwards. The data processed by the image processing device 90 is stored in a recording medium such as a hard disk or a memory. And it is necessary to display the data on the monitor of the console as needed. The displayed data is, for example, an inspection area, a foreign object number map, a foreign matter specification distribution/map, a foreign matter classification, a difference image, and the like. In order to perform such signal processing, system software 140 is provided. Further, an electro-optical system control power source 130 is provided to supply power to the electronic cylinder system. Further, the main chamber 160 may be provided with an optical microscope 110 or an SEM inspection apparatus 120.

Fig. 166 is a view showing an example of a configuration in which the electronic column 100 and the SEM inspection apparatus 120 of the optical inspection apparatus are provided in the same main chamber 160. As shown in Fig. 166, it is advantageous when the optical inspection apparatus and the SEM inspection apparatus 120 are provided in the same chamber 160. The sample 20 is mounted on the same stage 30, and the sample 20 can be observed or inspected by both the image method and the SEM method. The methods of use and advantages of this configuration are as follows.

First, since the sample 20 is mounted on the same stage 30, when the sample 20 is moved between the electron column 100 of the image type and the SEM inspection apparatus 120, the coordinate relationship can be obtained without any ambiguity. Therefore, when the detection site or the like of the foreign matter is to be defined, the two inspection apparatuses can easily define the same portion with high precision.

When the above configuration is not applicable, for example, the image type optical inspection apparatus and the SEM type inspection apparatus 120 are configured as separate devices, and the sample 20 is moved between the separate devices. At this time, since the sample 20 must be placed on each of the platforms 30, the alignment of the sample 20 must be performed separately for the two devices. Further, when the alignment of the sample 20 is performed separately, the specific error of the same position is 5 to 10 [μm]. In particular, in the case where the sample 20 has no pattern, since the position reference cannot be defined, the error is larger.

On the other hand, in the present embodiment, as shown in Fig. 166, in the two types of inspection, the sample 20 is provided on the stage 30 of the same chamber 160. Even in the case where the stage 30 is moved between the electronic cylinder 100 of the image type and the SEM inspection apparatus 120, the same position can be defined with high precision. Therefore, even when the sample 20 has no pattern, the position can be defined with high precision. For example, the position can be defined with an accuracy of 1 [μm] or less.

This high definition is very beneficial for the following situations. First, the foreign matter inspection of the unpatterned sample 20 can be performed by imaging. Then, the definition and detailed observation of the detected foreign matter 10 can be performed in the SEM inspection apparatus 120. Since the correct position definition can be made, not only the presence or absence of the foreign matter 10 but also the virtual inspection of the foreign matter 10 can be judged, and the specification or shape of the foreign matter 10 can be observed at a high speed.

As described above, when the electronic column 100 for foreign matter detection and the SEM type inspection device 120 for detailed inspection are separately provided, the definition of the foreign matter 10 takes a lot of time. Moreover, in the case of a patternless sample, the degree of difficulty increases. This problem can be solved by this embodiment.

As described above, in the present embodiment, the ultra-fine foreign matter 10 is inspected with high sensitivity by using the imaging conditions of the foreign matter 10 of the imaging optical method. Further, the optical column 100 and the SEM inspection device 120 of the imaging optical system are mounted in the same chamber 160. Thereby, in particular, the inspection of the ultra-fine foreign matter 10 of 30 [nm] or less and the determination and classification of the foreign matter 10 can be performed very efficiently and efficiently. Further, the present embodiment can be applied to the above-described first to twenty-eighth embodiments and the embodiment in which the number is not assigned.

Next, another example of inspection using both the image projection type inspection apparatus and the SEM will be described.

In the above, foreign matter was detected by a projection projection type inspection device, and detailed inspection was performed by SEM. However, the invention is not limited to this type. Other inspection methods can also be applied to the two inspection devices. It is also possible to combine the characteristics of each inspection device for effective inspection. Other inspection methods are, for example, those described below.

In this inspection method, the image projection type inspection device and the SEM system do not perform Inspection of the same area. Further, the image projection type inspection apparatus employs a "cell to cell" inspection, and the SEM applies "die to die" inspection, thereby achieving high-precision inspection with good efficiency as a whole.

More specifically, the image projection type inspection apparatus performs "cell-to-cell" inspection of a region having a large number of repeating patterns in a crystal grain. The SEM checks the "grain-to-grain" for areas with fewer repeat patterns. The test results of the two parties are combined to obtain one test result. The "grain-to-grain" is an inspection that compares images of two crystal grains sequentially obtained. The "cell to cell" is a comparison of the images of the two unit cells obtained in sequence. The unit cell is part of the grain.

In the above-described inspection method, the high-speed inspection is performed on the pattern of the repeating pattern by the image projection method, and on the other hand, the inspection is performed on the region where the number of the repeating patterns is small with high precision with a small SEM. SEM is not suitable for high speed inspection. However, the area with less repeating pattern is relatively narrow, so the SEM inspection time can be completed without a long time, and the overall inspection time can be suppressed to a short time. This inspection method maximizes the advantages of the two inspection methods and allows high-precision inspections in a short inspection time.

Next, returning to Fig. 165, the handling mechanism of the sample 20 will be described.

A sample 20 such as a wafer or a mask is transported from the loading cassette to the small environment device 180, and an alignment operation is performed therein. The sample 20 is transported to the load lock chamber 162 by the atmospheric transfer robot. The load lock chamber 162 is vented from the atmospheric state to the vacuum state through the vacuum pump. When the pressure reaches a certain value (about 1 [Pa] or less), the sample 20 is transported from the load lock chamber 162 to the main chamber 160 by the transfer robot placed in the transfer chamber 161. The sample 20 is then placed on an electrostatic chuck mechanism on the platform 30.

Figure 167 shows the interior of the main chamber 160 and the electronic cylinder system 100 disposed in the upper portion of the main chamber 160. The same components as those in FIG. 158 are denoted by the same reference numerals as those in FIG. 158, and the description thereof will be omitted.

The sample 20 is placed on a platform 30 that is movable in the x, y, z, and θ directions. High precision alignment is achieved through the platform 30 and the optical microscope 110. Then, the image projection optical system performs the foreign matter inspection and the pattern defect inspection of the sample 20 using the electron beam. Here, the potential of the sample surface 21 is very important. In order to measure the surface potential, the main chamber 160 is equipped with a measuring device capable of measuring the surface potential in a vacuum. The surface potential measuring device measures the two-dimensional surface potential distribution on the sample 20. According to the measurement result, focus control is performed on the secondary optical system 60a that forms an electronic image. The focus map of the two-dimensional position of the sample 20 is produced based on the potential distribution. Use this map to check while changing the focus in the control check. As a result, image blur or distortion due to a change in the surface circular potential due to the portion is reduced, and image acquisition and inspection with excellent accuracy and stability can be performed.

Here, the secondary optical system 60a is configured to measure the detection current of electrons incident on the NA stop 62 and the detector 70, and is configured to be provided with an EB-CCD at the position of the NA stop 62. . This composition is very beneficial and efficient. In Fig. 167, the NA stop 62 and the EB-CCD 65 are provided in a body type holding member 66 having openings 67, 68. Further, the secondary optical system 60a is provided with a mechanism capable of independently performing current absorption of the NA stop 62 and image acquisition of the EB-CCD 65. In order to realize this mechanism, the NA aperture 62 and the EB-CCD 65 are arranged in a vacuum X, Y platform 66. Therefore, position control and positioning of the NA diaphragm 62 and the EB-CCD 65 can be performed. Furthermore, the platform 66 is provided with openings 67, 68 so that the mirror reflects electrons and the secondary electrons can pass through the NA stop 62 or the EB-CCD 65.

Next, the operation of the secondary optical system 60a having such a configuration will be described. First, the EB-CCD 65 detects the spot shape of the secondary electron beam and its center position. Then, the voltage adjustment of the aberration compensator, the lenses 61, 63, and the aligner 64 is performed in such a manner that the spot shape is circular and the diameter is minimized. In this regard, the adjustment of the spot shape and the astigmatic aberration at the position of the NA stop 62 by the prior art cannot be directly performed. This direct adjustment is possible in this embodiment, and it is also possible to perform high-precision correction of astigmatic aberration.

Moreover, the center position of the beam spot is easy to detect. Therefore, the position of the NA stop 62 can be adjusted, and the center of the hole of the NA stop 62 is placed at the beam spot position. In this regard, conventional techniques cannot directly adjust the position of the NA stop 62. In the present embodiment, the position adjustment of the NA diaphragm 62 can be directly performed. Thereby, the NA diaphragm can be positioned with high precision, the aberration of the electronic image can be reduced, and the uniformity is improved. Moreover, the transmittance uniformity is improved, and an electronic image having a high resolution and a uniform degree of lightness can be obtained.

Further, in the inspection of the foreign matter 10, it is important to obtain the mirror-reflected electronic signal from the foreign matter 10 with good efficiency. The position of the NA aperture 62 limits the transmission and aberration of the signal, so it is very important. The electron emission twice is released according to the cosine law in a wide range of angles from the surface of the sample, and reaches an evenly wide area (for example, φ 3 [mm]) at the NA position. Thus, the electron emission twice is insensitive to the position of the NA stop 62. In contrast, in the case of a mirror-reflecting electron, the angle of reflection on the surface of the sample, the degree and the degree of electricity The incident angle of the sub-beams is the same. Therefore, the mirror-reflected electrons show a small spread and reach the NA stop 62 with a small beam path. For example, the extended area of the specular reflected electrons is 1/20 or less of the extended area of the secondary electrons. Thus, the mirror reflected electrons are very sensitive to the position of the NA stop 62. The extended region of the mirror-reflected electron at the NA position is usually in the region of φ 10 to 100 [μm]. Therefore, it is very advantageous to determine the position at which the mirror reflection electron intensity is the highest, and to arrange the center position of the NA stop 62 at the determined position.

In order to achieve the purpose of setting the NA stop 62 to such an appropriate position, in a preferred embodiment, the NA stop 62 can be moved in the x, y direction with an accuracy of about 1 [μm] in the vacuum of the electron column 100. The signal intensity is measured while moving the NA stop 62. Then, the position with the highest signal intensity is obtained, and the center of the NA stop 62 is set at the obtained coordinate position.

It is advantageous to use the EB-CCD 65 for signal strength measurement. Because, by this, you can know the two-dimensional information of the beam. The number of electrons incident on the detector 70 can be determined, so that the signal intensity can be quantitatively evaluated.

Alternatively, in order to achieve a conjugate relationship between the position of the NA stop 62 and the position of the detection surface of the detector 70, the pupil arrangement can be specified, and the condition of the lens 63 between the aperture and the detector can be set. This composition is also very advantageous. Thereby, the beam image at the position of the NA stop 62 can be imaged on the detection surface of the detector 70. Thus, the detector 70 can be used to observe the beam variation profile of the position of the NA stop 62.

In addition, the NA specification (light path) of the NA diaphragm 62 is also important. As described above, since the signal area of the mirror-reflected electrons is small, the effective NA specification is about 10 to 200 [μm]. And the NA specification is better relative to The beam diameter is a large specification of +10 to 100 [%].

In this regard, the image of the electron is formed by mirror-reflecting electrons and electron emission twice. Through the setting of the above-mentioned aperture specifications, the proportion of mirror-reflected electrons can be higher. Thereby, the contrast of the mirror-reflected electrons can be increased, that is, the contrast of the foreign matter 10 can be improved.

If it is explained in more detail, when the aperture of the aperture is reduced, the electron emission twice will be inversely proportional to the area of the pupil. Therefore, the degree of darkness of the normal portion becomes small. However, the mirror reflection signal does not change, and the degree of darkness of the foreign matter 10 does not change. Therefore, the contrast of the foreign matter 10 becomes larger in accordance with the reduction in the degree of darkness of the surroundings, and a higher S/N can be obtained.

Further, in order to adjust the position of the pupil not only in the x and y directions but also in the z-axis direction, a diaphragm or the like can be reconstructed. This composition is also advantageous. The aperture is preferably disposed at a position where the mirror reflection electrons can be most concentrated. Thereby, the reduction of the mirror reflection electron aberration and the reduction of the secondary emission electrons can be performed very effectively. Thus a higher S/N can be obtained.

As mentioned above, mirror-reflected electrons are very sensitive to NA specifications and shapes. Therefore, in order to obtain a high S/N, it is very important to appropriately select the NA size and shape. A configuration example for selecting such an appropriate NA specification and shape will be described below. Here, the shape of the aperture (hole) of the NA aperture 62 will also be described.

Here, the NA stop 62 is an element (component) having a hole. In general, a part is sometimes called a diaphragm, and sometimes a hole is called a diaphragm. In the following description relating to the aperture, referring to FIGS. 168 to 172, in order to distinguish the component (component) and the hole provided therein, the component is referred to as an NA aperture, and the aperture in the component is referred to as Light. In the following description, symbols 62, 62a to 62d is a NA stop, and symbols 169, 69, 69a, 69b are apertures (holes). The shape of the aperture generally refers to the shape of the aperture.

Figure 168 is a reference example showing the conventional aperture 169. As shown in Fig. 168, the conventional technique places the circular aperture 169 in a fixed position. Therefore, the above-mentioned appropriate NA specifications and shapes cannot be selected.

On the other hand, in the sample inspection device of the present embodiment, the position of the NA stop 62 can be moved in two or three dimensions, and the position can be set. The movement of the NA stop 62 can be performed using the XY stage 66 described in accordance with Figure 167. Moreover, the applicable aperture can be appropriately selected from the complex apertures and positioned in parallel. Additionally, an NA diaphragm 62 can be provided with a plurality of apertures 69. In order to select one of the apertures, the NA aperture 62 can be moved (this configuration is also equivalent to selecting from a plurality of apertures). Other moving mechanisms can also be used. For example, a linear motor can be utilized to move the NA stop 62 in place of the XY stage 66. Furthermore, the NA stop 62 can also be supported by a rotating support member, which can also be used to perform positional movement of the NA stop 62. Hereinafter, a specific example of the shape of the hole of the NA stop 62 will be described.

Fig. 169 shows an example of the shape of one of the apertures 69. In Fig. 169, the aperture 69 has a circular hole shape. The shape of the hole is set to match the intensity distribution of the mirror reflected electronic signal. In this example, the intensity distribution is a long circular circle in the y direction in the measurement result of the reflected electron intensity distribution of the pupil. Here, the y direction is a direction in which the E×B filter 46 is deflected. The y-direction system and the optical axis of the primary electron beam are in the same direction. That is, the reason why the y-direction is rounded is considered to be the biasing component of the E×B filter 46. Therefore, in order to capture mirror-reflected electrons with good efficiency, the y direction has a long axis. The shape of the pupil is very advantageous. Thereby, the yield of the mirror-reflected electrons can be improved as compared with the prior art, and a high S/N (for example, ×2 or more) can be obtained. For example, the intensity distribution of the electron beam emitted twice is set to 100 [μm] in the y direction and 50 [μm] in the x direction (these values are half values of the full width). The circular aperture 69 can be selected within a range of positive to (Plus) 10 to 100 [%] with respect to the secondary emission electron beam diameter. For example, the diaphragm can be selected in such a manner that the aperture specification is 150 [μm] in the y direction and 75 [μm] in the x direction.

Next, the configuration of the NA stop 62 having the complex aperture 69 will be described using Fig. 170 to Fig. 173. Here, the NA apertures 62a to 62c are diaphragm members, and the aperture 69a is an opening provided in the aperture member.

Fig. 170 shows an example of the configuration of one of the NA apertures 62a having a plurality of apertures 69a. In Fig. 170, the NA stop 62a has two circular apertures 69a. In this example, the two holes are arranged at positions offset from the ±y direction based on the intensity center of the mirror-reflected electrons. The amount of deviation is, for example, about 50 [μm]. This configuration can capture mirror-reflected electrons from both the +y side and the -y side scattered by the foreign matter 10. Therefore, the configuration can expand the difference between the scattered mirror reflected electronic signal and the background two-time discharge electronic signal amount, and a high S/N can be obtained. The reasons for this are as follows. In the case where electrons are emitted twice, the amount of scattering in the scattering direction is limited to a small amount. Therefore, the amount of background electrons is reduced, and the S/N can be relatively increased.

Fig. 171 shows an example of the configuration of one of the NA apertures 62a having four apertures 69a. In Fig. 171, four circular apertures 69a are symmetrically arranged on the x-axis and the y-axis. That is, the two apertures 69a are arranged on the x-axis, the two apertures 69a are arranged on the y-axis, and the four apertures 69a are located equidistant from the center (origin). At the office. In other words, the four apertures 69a are arranged at equal intervals around the origin. More simply, the four apertures 69a are arranged in a diamond shape. Thereby, even when there are mirror-reflected electrons scattered from the foreign matter 10 toward both the x direction and the y direction, electrons can be obtained at a high S/N.

Figure 172 shows an NA stop 62c with four apertures 69a. The configuration of Fig. 172 is an example different from the configuration of Fig. 171. In Fig. 172, four circular apertures 69a are respectively arranged in the first quadrant to the fourth quadrant in the xy plane. In this example, the four apertures 69a are symmetrically arranged on the x-axis and the y-axis, and are arranged equidistant from the center (origin). In other words, the four apertures 69a are arranged at equal intervals around the origin. Even in the NA aperture 62c of such a shape, the aperture 69a can be set at a position where the intensity of the mirror reflection electronic signal is increased, and a high S/N signal can be obtained.

As shown in Figures 171 and 172, the number of apertures 69a is the same, and the configurations may be different. Thereby, appropriate NA apertures 62b, 62c can be used depending on the application. Thereby, high S/N can be obtained for various uses.

Fig. 173 is a view showing an example of the configuration of one of the NA pupils 62d having eight apertures 69b. As shown in Fig. 173, the number of pupils 69d can also be more than four. In the NA stop 62d shown in Fig. 173, a plurality of apertures 69b are arranged at equal intervals around the circumference of the center of the electron reflection intensity of the mirror. This configuration is advantageous when the position of the pupil 69b at a certain position on the circumference has a mirror-reflecting electron having a strong scattering specificity. The mirror reflection electrons can be properly captured.

In addition, in Fig. 170 to Fig. 173, in the relationship between the intensity of the specular reflected electron signal intensity and the pupils 69a, 69b, the pupil position and intensity The heart deviates. However, the present invention is not limited thereto, and the pupil position may coincide with the intensity center. That is, it can be set such that one pupil 69a, 69b and the mirror reflection electron intensity center coincide. In this case, the other apertures 69a, 69b capture the reflected electrons of the scatter mirror. These electron systems and the mirror-reflecting electrons of the intensity center are contained together in the electron image, and this composite image is obtained in the detector 70. In this way, a composite image of a strong mirror-reflected electron and a specifically scattered mirror-reflected electron can be obtained. Thus, a high S/N can be obtained while effectively detecting the foreign matter 10 having characteristics in the scattering direction. Moreover, the characteristics of the scattering direction can also contribute to the classification of the foreign matter 10.

Further, according to the present embodiment, the apertures 69, 69a, and 69b having an appropriate shape can be selected for the incident energy LE to be used. This choice can also provide very beneficial effects. The intensity distribution of the mirror reflected electrons varies depending on the incident energy LE. Therefore, the inspection apparatus according to the present embodiment can be configured to operate the apertures 69, 69a, and 69b corresponding to the incident energy LE used in the specification and shape. Thereby, it is very advantageous to adjust the aperture in accordance with the intensity distribution. For example, it is possible to consider the case where the mirror-reflected electrons have a long circular intensity distribution in the y direction. At this time, it is assumed that shooting or inspection is performed under two different conditions. For example, in the first imaging inspection condition, the incident energy is set to the first value, that is, LE=3 [eV]. In the second imaging inspection condition, the incident energy is set to the second value, that is, LE = 2 [eV]. Here, if the incident energy LE is small, the intensity of the specular reflected electron intensity becomes large at the positions of the NA pupils 62, 62a to 62d. The NA stop 62, 62a to 62d can be appropriately selected to accommodate this distribution change. For example, when the first incident energy is used, a circular light having a y direction of 100 [μm] and an x direction of 50 [μm] can be selected. 69. When the second incident energy is used, the intensity of the specular reflected electrons is about twice as large. Therefore, a circular aperture 69 having a y direction of 200 [μm] and an x direction of 100 [μm] can be used. By selecting the aperture in this way, the mirror reflection electrons can be detected very efficiently.

Next, the configuration of the Faraday cup and the like described in Fig. 162 will be described again. These configurations can be set in the electron beam inspection apparatus of Fig. 167.

Figure 174 shows the platform 30 of Figure 167. The stage 30 is provided with a Faraday cup 31, a reference sample wafer 26 having grooves 25 and 25a, and an EB-CCD 37. Thereby, the uniformity of the primary electron beam and the irradiation position can be monitored with high precision, and the fluctuation of the electron beam with time can be monitored with high precision.

In this regard, conventional techniques do not directly monitor the primary electron beam. Therefore, the conventional technique periodically mounts the Faraday cup 31 on a plurality of points on the same sample 20, and obtains an image of the electron beam irradiation using the Faraday cup 31. This image can be used for beam evaluation and adjustment. However, the conventional technique can only obtain an image obtained by superimposing the fluctuations of the optical system 40 and the secondary optical system 60a once. The operation of separating and evaluating and adjusting the main factors of the two optical systems is complicated and the precision is also inferior. This embodiment can solve such a problem.

Further, according to the present embodiment, the current density distribution of the primary electron beam can also be measured with high precision. For the lens (42, 45) of the primary optical system, the aligner, and the electronic release control system of the electron gun 41, it is possible to perform feedback with excellent precision. Thus, a more uniform beam variation profile can be formed. For example, in the conventional current density distribution measurement, the diameter is about φ 30 [μm] Faraday Cup. Then, about 5 points were measured at a distance of 30 [μm]. This measurement limits the resolution due to the hole size of the Faraday cup 31. Moreover, it takes a lot of time because the measurement is performed little by little. Therefore, the instantaneous distribution of the electron beam irradiation cannot be measured.

According to the foreign matter inspection apparatus of the present embodiment, the beam change profile of the electron beam can be directly measured. Then, the electron beam can be appropriately adjusted once according to the measurement result.

Further, in the adjustment of the primary electron beam, in order to obtain the relationship between the specification of the foreign matter 10, the signal intensity, and the S/N, the present embodiment can be used to produce a standardized sample. Through the use of such samples, great advantages can be obtained. For example, a normalized microsphere whose specifications are known is spread on a single film of a sample. This type of sample is well suited for sensitivity correction.

Figure 175 shows a sample 20 with sample 15 interspersed. Sample 15 is a schematic replacement for foreign matter 10. Therefore, it is preferable to use a sample having a material close to the foreign matter 10 and close to the material of the foreign matter 10. For example, sample 15 is a normalized microsphere made of PSL (polystyrene latex, Polystyrene Latex). Ultrafine particles can also be used. The sample 20 may be a semiconductor wafer such as Si. A film can be formed on a semiconductor wafer. The sample 20 may also be a glass substrate on which a film is formed. The film on the sample 20 may be either a conductive film or an insulating film. For example, the film on the semiconductor wafer can be SiO_2、A film of Ta, Cu, Al, W, or the like. Further, the film on the glass substrate may be, for example, a film of Cr, CrN, Ta, TaN, TaBN, TaBO, Si, Al, Mo or the like.

In Figure 175, the size of sample 15 is known. Therefore, by obtaining the image of the sample 15, the specification and signal intensity or S/N of the sample 15 can be obtained. Relationship.

Fig. 176 is a view showing the measurement results obtained when the image of the sample 15 shown in Fig. 175 was obtained. Figure 176 is an example of the relationship between sample 15 and signal intensity. In Fig. 176, the horizontal axis is the specification of the sample 15, and the vertical axis is the signal intensity. The vertical axis can also be S/N. By varying the specifications of the sample 15, the signal intensity corresponding to the sample size can be obtained. As shown in Figure 176, a plot can be made from the signal strength. Thereby, the relationship between the specification of the foreign matter 10 and the signal strength or S/N can be grasped.

In the above, a microsphere was used as the sample 15. In particular, a suitable sphere size is 100 [nm] or less. That is, it is advantageous to use a microsphere of φ 1 to φ 100 [nm].

As described above, the electron beam inspection apparatus and the electron beam inspection method of the present embodiment have sensitivity to the nano-scale ultra-fine foreign matter 10. The minute sample 15 like the above is particularly advantageous for inspecting the minute foreign matter 10.

In this regard, in the conventional light type foreign matter inspection method, the resolution is limited by the wavelength of light, and the detection of a small size foreign matter 10 of less than 100 [nm] is difficult. According to the electron beam inspection apparatus and the electron beam inspection method of the present embodiment, sufficient sensitivity can be obtained to detect the small foreign matter 10.

Next, an embodiment in which the incident energy is appropriately set will be further described with reference to Fig. 177.

Fig. 177 is a view showing the degree of darkness of the incident energy with respect to the beam in the electron beam inspection method of the present embodiment. The foreign body inspection method can be adapted It is used for the sample 20 having a plain plane (a plain plane means a non-patterned surface, the same applies hereinafter) or a patterned surface. This embodiment is characterized in that the characteristics shown in Fig. 177 are obtained, and the region of the incident energy LE is selected using the characteristics of Fig. 177. The degree of darkness (change in the degree of darkness relative to the incident energy LE) is related to the type of electron to be detected. The following shows the types of electronics:

LE&lt;LEA: Mirror reflection electron

LEALELEB: 2 times the state of electron and mirror reflection electrons mixed

LEBLE: 2 times to release the electronics

Here, by setting LE in the area of LEALELEB+5[eV], it is possible to obtain a high S/N image, and it is possible to realize high-sensitivity defect inspection and foreign matter inspection. The reason for this setting is explained. For example, assume that there is a foreign matter 10 on the plane of Si or W. In the present embodiment, the foreign matter 10 is charged to form mirror-reflected electrons. At this time, it is desirable that the prime plane (the surface without the pattern) as the background is less shaded. Because S/N will go high. In order to reduce the degree of shading of the prime plane, the energy conditions for discharging the electron region and the mixed region twice are set to appropriate values. The mixed area is an area where the mirror reflects electrons and the electrons are mixed twice. The mixed region is between the secondary electron emission region and the mirror reflection electron region, and corresponds to the transition region.

The mixing area is LEALELEB of Figure 177. In this region, it is considered that the mirror-reflected electrons are generated from the foreign matter 10, and the electrons are emitted twice from the background sample 20. In the mirror-reflected electron region of LE&lt;LEA, electrons are also reflected from the background. Therefore, the degree of background is increased, and the difference between the degree of darkness of the foreign matter 10 and the degree of background is small. That is, S/N Become smaller. Moreover, an energy region in which the LE is much larger than the LEB also emits electrons twice from the foreign matter 10. In this case, S/N also becomes smaller.

In order to easily detect the foreign matter 10, the magnified image 81 of the foreign matter 10 and the surface image 82 of the background sample surface 21 are poorly shaded to the maximum. The difference in depth is dependent on the degree of darkness relative to the incident energy LE shown in Fig. 177. Furthermore, in Figure 177, a characteristic curve is shown. On the other hand, in the present embodiment, two characteristic curves such as a characteristic curve of the foreign matter 10 and a characteristic curve of the pure sample 20 are used as appropriate. Comparing these two characteristics, in the present embodiment, the incident energy LE having the largest difference in the degree of darkness can be used. Thereby, the incident energy is appropriately determined.

Regarding the above, depending on the combination of the characteristic curve of the foreign matter 10 and the characteristic curve of the sample surface 21, the energy band having a large difference in the degree of darkness changes. Therefore, the incident energy can be appropriately set using the characteristic curve of the inspection object.

In addition, based on the experience of previous experiments, the LE of the LEALELEB+5[eV] region can be used very advantageously and can achieve great effects. The method and configuration of applying the energy region are applied to any of the methods and configurations described above to the extent possible. Thereby, a high S/N can be obtained to realize high-sensitivity and high-speed defect inspection and foreign matter inspection.

Next, using Fig. 178, the efficiency of the primary system electron beam incident energy LE in the detection or inspection of the foreign matter 10 will be described in more detail. Figure 178 shows the relationship between the incident energy LE of the primary electron beam and the degree of darkness of the image. In Fig. 178, the relationship between the darkness of the sample 20 and the degree of the foreign matter 10 is shown to show the relationship between the sample 20 and the foreign matter 10.

As mentioned in the description of Fig. 177, the region where the incident energy LE is smaller than the LEA represents the mirror reflection electron region. The mirror-reflecting electron region is an energy region in which only substantially all of the mirror-reflected electrons are detected from the normal portion where the foreign matter 10 is not present on the sample 20.

Furthermore, the region where the incident energy LE is larger than LEB is shown as the secondary electron emission region. The secondary electron emission area was an area where only substantially all of the electrons were detected from the normal portion of the sample 20. As described above, the electron emission twice includes two electrons, reflected electrons, and backscattered electrons.

Further, the region where the incident energy LE is equal to or higher than LEA and below LEB is a mixed region. The mixed region showed a region in which both the specular reflected electrons and the two emitted electrons were mixed from the normal portion of the sample 20. The mixed region is a transition region between the mirror-reflecting electron region and the secondary electron-emitting region.

As described above, the electron beam incident energy LE of the primary system to be irradiated is preferably an energy region set to LEALELEB or LEALELEB+5[eV]. This point is explained in more detail using Figure 178.

Figure 178 is a graph showing the change in the degree of DN of the incident energy LE of the normal portion of the foreign matter 10 and the sample 20 with respect to the primary system electron beam. The DN (Digital Number) corresponds to the number of electrons detected by the detector 70. When the contact resistance between the foreign matter 10 and the sample 20 is high, or when the foreign matter 10 is charged, the foreign matter 10 exhibits a different degree of change from the normal portion around it. This is because the foreign matter 10 generates a potential change, and it becomes easy to generate mirror-reflected electrons. According to the findings of the inventors, it has been confirmed many times that the range of LEA to LEB is -5 [eV] to +5 [eV]. Moreover, as described above, the foreign matter 10 is even in the primary system as compared with the normal portion. In the state where the incident energy LE of the electron beam is high, mirror-reflected electrons are also generated. (Here, the mirror-reflected electrons can be mixed with the two-shot electrons). Therefore, when photographing or inspecting the foreign matter 10, LEA to LEB+5 [eV] is suitable as the incident energy LE region to be used. For example, assume that LEA to LEB is -5 [eV] to +5 [eV]. At this time, the incident energy LE region is preferably -5 [eV] to +10 (= 5 + 5) [eV].

In addition, the incident energy range "LEA to LEB+5[eV]" is effective for all types of substrates regardless of the material of the substrate. For example, the incident energy range "LEA to LEB+5 [eV]" is effective for a substrate on which a pattern or the like is formed, or a substrate on which a foreign matter is present on the surface. Moreover, the LE range is effective irrespective of the material of the substrate or foreign matter. For example, in the observation of a glass substrate, the incident energy range "LEA to LEB+5 [eV]" is very useful. Thereby, a good image can be obtained.

Here, the reason why the foreign matter 10 can be photographed with high contrast can be understood from Fig. 178. As shown in Fig. 178, the foreign matter 10 and the surrounding normal portion have different brightness variations. Secondly, the foreign matter 10 produces mirror-reflected electrons with a high incident energy LE (= LEB + 5 [eV]) compared to the normal portion. Therefore, as shown in the figure, it is possible to obtain a difference ΔDN between the large foreign matter 10 and the normal portion. For example, assume that the DN of the normal portion is 50 DN, and the brightness variation (noise) of the normal portion is 3 DN. Moreover, it is assumed that the degree of DN of the foreign matter 10 is 100 DN. At this time, the difference in depth is ΔDN=50DN (=100DN-50DN), and S/N is 50/3=16.7. In this way, a high S/N value can be obtained. This is indeed a phenomenon occurring in the LE region of the LEA to LEB+5 [eV] incident energy LE. By using this phenomenon, Shoot or check with high contrast. In the other incident energy LE region, it is not possible to form the mirror-reflected electron generation state only by the foreign matter 10. Therefore, as described above, the contrast between the foreign matter 10 and the surrounding normal portion cannot be improved. Therefore, in the detection of the foreign matter 10, it is preferable to perform detection in the range of LEALELEB+5[eV]. Further, with regard to the matters stated in this embodiment, the adjustment methods set forth in the embodiments related to FIGS. 45 to 50 are used, that is, the mirror reflection electrons (mirror reflection center MC) and the NA pupil position ( The x, y direction) is a method of adjusting and controlling the position of the relative position of the electron-polycondensation center position with respect to the position of the NA pupil. Thereby, it is possible to efficiently and efficiently obtain high contrast, and S/N of defects.

The above description of the preferred embodiments of the invention is now contemplated. However, it should be understood that the present embodiment can be modified in various forms, and various changes in the true spirit and scope of the present invention should be included in the appended claims.

The present invention can be applied to an electron beam inspection apparatus that inspects the presence or absence of foreign matter on a sample such as a semiconductor wafer using an electron beam or inspects for the presence or absence of defects. Further, the present embodiment can be applied to the above-described first to twenty-eighth embodiments and the embodiment in which the number is not assigned.

(Embodiment 30)

-Platform-

The example of the stage of the inspection apparatus and the inspection method of the present invention will be described.

In the present embodiment, an example in which the inspection apparatus and the inspection method of the present invention are used for a mask (EUV mask, NIL mask) is different from the inspection apparatus and inspection method of the present invention for wafers in the above embodiment. section (Reverse rotation, rotation unit, tray mounting unit, and static elimination unit) will be described centering.

Refer to Figure 179 and Figure 180. The mask placed on the card such as SMIF is transported by an atmospheric transport robot. Check the surface of the mask and either side of the back. Also, choose to set it in the direction of the platform and check it. If it is the same as a certain state in the cassette, it does not need to be reversed.<img file="TW201634917A_D0010.tif" wi="54" he="56" img-format="tif" img-content="character" orientation="portrait" inline="no" />. When the direction of the mask is changed by 90 degrees, 180 degrees, etc., the direction is selected by the rotation unit. If it is in the same direction as the cassette, no work is required. Then, the robot is placed on the tray loading unit, and the mask is placed on the tray and transported together with the tray. When the tray is mounted, there is a positioning mechanism. For example, a direction adapter or the like that determines the direction of the angle provided on the diagonal is used. Thereby, the general direction is determined, for example, ±1 to 10 mrad (radian). The de-energizing unit is used to uncharge the mask. In the procedures and inspections of static electricity or pre-steps, if it is affected by the electrical properties of light or electrons, it will remain charged on the surface. In order to release it, the surface potential state is always maintained to perform stable inspection, and the gas is removed in the atmospheric transportation section. As the static elimination system, an X-ray irradiation type, a UV irradiation type, an ion irradiation type, or the like is used. After that, it is transferred to the LL room. Vacuum is applied to the LL chamber to form a vacuum state. At this time, the slow exhaust is performed at the initial stage, and then the reduction of the adhering particles and the shortening of the vacuum forming time are performed by increasing the exhaust velocity. In the LL room, a CCD camera is provided in order to obtain the alignment correction amount of the mask. It is thereby determined whether there is any degree of deviation in the direction of rotation to determine the amount of correction of the rotating table on the platform. If the correction amount exceeds the correction amount, return to the tray loading unit again and reset the mask. The vacuum handling robot is in the operating room. Through the vacuum state Room to prevent particles from adhering to the robot and preventing dust from the robot. When the LL chamber reaches the specified degree of vacuum, the gate valve is opened, and the tray and the mask are carried by the vacuum transfer robot, and then placed on the platform of the main chamber. The platform can perform 3-axis control of X, Y, and θ. In order to maintain accuracy, the angle of rotation of the θ platform is small, with a maximum of ±1 to 3 degrees. This angle can be within the range that can be achieved by pre-alignment. Thereby, the accuracy and rigidity of the θ platform can be improved. When the angle of rotation is large, a large clearance is required. Because the angle control accuracy and rigidity will deteriorate. The trays placed on the platform are fixed by electrostatic chucks. There are Faraday cups, calibration samples and standard samples on the platform. The platform position detection system is the same as the wafer, and the position detection and control is performed by using a laser interference meter. Further, the present embodiment can also be applied to the above-described first to twenty-ninth embodiments and the embodiments in which the number is not assigned.

(Embodiment 31)

Further examples of other foreign matter inspections of the inspection apparatus and the inspection method of the present invention will be described.

(EO adjustment method)

As described in the above-described pattern EO adjustment method, the EO condition for foreign matter inspection can also be performed by measuring the distribution of the beam coming to the NA position, and performing position control of the mirror reflection electron to realize high-sensitivity foreign matter inspection. . In the embodiment of the pattern inspection, reference is made to Fig. 181 in which the beam observation at the NA position is performed and the incident angles θ and α are adjusted. The situation of foreign matter inspection is mostly the case of a non-patterned plane. Therefore, accuracy is not required for α. However, it must be able to align the coordinates of the defect. At this time, the correlation with NA can be determined in the same manner as the pattern check.

In particular, in order to increase the sensitivity to small foreign matter having a size of 5 to 30 nm, it is effective to set the mirror reflection electron position at a distance from the CO center of the electron emission twice. That is, it is effective to expand the incident electron angle θ. (The vertical axis is set from the sample to the z-axis, 0 degrees). This is because when incident at a larger angle than normal incidence, it is susceptible to surface potential non-uniformity. That is, the surface potential difference with respect to the energy in the z direction affects, but the closer to the horizontal incidence than the normal incidence, the velocity component in the z direction is easily affected, and thus the difference from the electron orbit of the peripheral normal portion occurs greatly. . Another important factor is that the unevenness of foreign matter charging is prone to occur. In short, it is easy to cause the beam to be bright, and the potential difference between the irradiated portion and the unirradiated portion is large, and a sharp potential distribution change is formed in the vicinity of the foreign matter. Therefore, it will be affected by it, and there will be cases where the track is easy to change. Through this action, the sensitivity is improved. The conceptual diagram is shown in Figure 182. From the test results, it is found that when the inspection is performed with high sensitivity in the detection of minute foreign matter having a specification of 5 to 30 nm, the incident angle is preferably about 10 to 30 degrees. Further, at this time, it is preferable to take the mirror reflection electron signal from the normal portion as much as possible, and determine the NA position in order to obtain only the mirror electrons from the foreign matter. Specifically, the NA is set to be in the middle of the CO center and MC position of the electron discharge twice. Further, it is preferable to set the NA at an extreme position that does not reach the MC position (mirror reflected electron position). Further, it is preferable that the distance between the end portion of the MC and the end of the NA is from 1 to 100 μm, more preferably from 10 to 50 μm. When such a distance is inspected for a large area, the MC position changes due to various important factors. Therefore, in order to stably achieve the foreign matter detection with high S/N at this time, the above is maintained. The distance is very important. In addition, although the direction of the MC can be in any direction, it is easier to generate signals in this direction. That is, in the case of a spherical foreign object, a round signal can be obtained in this direction. This means that the signal can be obtained by amplification. That is, since the signal can be obtained in a specification larger than the foreign matter specification, it is possible to inspect the Px specification which is larger than the foreign matter. This is very effective for the throughput of particularly small foreign objects. For example, when the foreign matter specification is 10 nm, the present invention can also be detected at 100 nm Px. This produces a ×100-fold throughput difference compared to the case of using 10 nm Px. In the present invention, a Px specification of about 5 to 50 times the minimum foreign matter size to be detected can be used. Further, especially in the case of ultra-micro foreign matter which is difficult to detect: 5 to 30 nm, it is effective to use ×2 to ×10. When the foreign matter is detected by the SEM method, the Px specification must be 1/3 to 1/10 of the foreign matter, and the difference from the present invention produces a difference of ×6 to ×100 only in the Px specification, resulting in a large throughput. difference. Further, in the light type, the specification of the 1/2 wavelength is the foreign matter detection limit, and it is impossible to detect the minute foreign matter as the object in the present invention.

Further, regarding the direction, if MC and NA are set in the Y direction or the X direction, a signal which is symmetrical with respect to the x and y axes can be obtained, and the oblique direction becomes asymmetrical. These directions are used separately depending on the sample or foreign matter to be used as the target. Further, the present embodiment can be applied to the above-described first to third embodiments and the embodiments which are not numbered.

(Embodiment 32)

This embodiment describes an example of NA (numerical aperture) for using the inspection apparatus and inspection method of the present invention.

(NA shape)

In pattern inspection and foreign matter inspection, it is more effective to use the NA of the shape shown in Fig. 183 and Fig. 184 as compared with the generally used circular aperture NA. It achieves high contrast, S/N, and electron count, which increases sensitivity and increases throughput.

In the pattern inspection, there are cases where the contrast ratios of the vertical and horizontal patterns are different in the y direction and the x direction. At this time, if the cross hole shape is used, the information of the electronic signal having a strong vertical contrast and the electronic signal having a strong horizontal contrast can be synthesized, and the amount of electrons can be increased with high contrast, and high S/N can be obtained.

When using a slit, it is effective to obtain electronic information that is more powerful in obtaining the information of the vertical signal or in the direction of obtaining more horizontal information. In the case of a pattern, there may be a case where the polycondensation points (COx, COy) in the x and y directions are shifted in the z direction, for example, by E×B or the like. In this case, a slit having a long hole in the y direction can be used to define the amount and aberration of the electronic information in the y direction using the COx in the x direction, and the slit for the long hole in the y direction can be defined by the slit for the COy in the y direction. The amount and aberration of electronic information in the x direction. In this way, the electronic quantity and aberration control can be performed, high contrast and S/N can be obtained, and the amount of electrons can be increased, and the sensitivity can be improved and the throughput can be increased with respect to the circular hole.

According to the above NA, the sensitivity can be increased by ×1.4 to ×5, and the amount of electrons can be increased by ×1.5 to ×10, which is very effective.

Further, the NA shape shown in Fig. 184 is particularly effective at the foreign matter inspection, that is, the shape A of the concave curved portion and the NA of the slit type. In the foreign matter inspection, it is effective to set the NA to a position where the reflected electrons are not reflected, but it is placed at a position close to the MC where the mirror reflection electron intensity is high. At times, S/N of higher foreign matter signals is sometimes obtained. At this time, it is very effective to use the NA as shown in Fig. 184. In the case of a circular hole, since it is a convex hole, it is impossible to obtain many electronic information near the MC, and it is easy to be affected by the change of the MC position. As a method for solving this problem, there is a shape shown in Fig. 184. If it is a slit and has a shape of a concave curved portion, it is easy to approach the MC, and many electronic information near the MC can be obtained. Further, with respect to the change in the position of the MC, the concave shape along the C shape or the shape of the slit (linear) hole is close to the side of the MC, so that the fluctuation of the MC position is set even without being affected by it. The distance will also obtain more peripheral electronic information than the circular shape of the convex hole. This is the use of the mirror-reflecting electronic signal of the foreign object and the electronic information mixed with the secondary emission electronic signal of the normal part. Only the foreign object will increase the brightness due to the electron reflected by the mirror. At this time, if the electronic information of the MC is mixed, the mirror-reflected electrons are added to the entire image area, so that the difference between the foreign matter and the surrounding area (the difference in brightness and the difference in the amount of electrons) is lowered, and the S/N is also lowered.

The lower one shown in Fig. 184 is a conceptual diagram showing such a state, which shows an example of the setting state of MC and NA. The MC is usually circular. At this time, each of the NA systems is disposed such that its hole end portion is located at the same distance L1 from the MC end portion. Moreover, the width and diameter (L2) of the holes are assumed to be the same. At this time, compared with the circular shape, the area of the concave curved shape and the slit type is large and the area close to the MC is large, so that an electronic signal close to the MC can be obtained.

In addition, FIGS. 185 and 186 are examples of the positional relationship between the MC and the NA. This positional relationship can be used for pattern inspection and foreign matter inspection. Furthermore, regarding the electron beam conditions or pre-charging conditions of the irradiation, etc. The use form of the above-described embodiment 29 is very effective.

(Embodiment 33)

This embodiment is a modification of the embodiment of the inspection apparatus of the present invention shown in Fig. 8. Except that the electrode 725 is further provided, the rest is the same as the inspection apparatus of the present invention shown in Fig. 8, and the repeated description is omitted. As shown in FIG. 187, when the wafer W has a via b, the voltage applied to the electrode 725 is 4000V to -400V, and the electric field of the electron beam irradiation surface of the wafer can be set at 2.0kV to -0.2kV. /mm (- indicates that the W side of the wafer is high). In this way, the electric field strength (from the surface to the vertical direction, the Z-axis direction) of the surface of the sample can be raised or lowered. That is, when the sample is easily discharged, the electric field strength can be lowered so that no discharge occurs. In this state, no discharge occurs between the objective lens system 724 and the wafer W, and the defect inspection of the wafer W can be performed, but the detection efficiency of photoelectrons is somewhat lowered. Therefore, a series of operations of irradiating light and detecting photoelectrons are performed, for example, four times, and the detection results of the obtained four times are subjected to processing such as cumulative addition and averaging to obtain a predetermined detection sensitivity.

Furthermore, when the wafer has no through holes b, a higher electric field strength can be used. Even if the voltage applied to the electrode 725 is set to +3000 V, no discharge occurs between the objective lens system 724 and the wafer, and defect inspection of the wafer W can be performed. At this time, since the starting electric field can be increased by the voltage applied to the electrode 725, and the aberration of the objective lens can be reduced, the resolution can be improved, and high contrast and S/N can be achieved. Therefore, it can be inspected with high sensitivity and high throughput.

(electrode)

The objective lens 724 and the wafer W are arranged to be irradiated with respect to the electron beam. The electrode 725 is substantially axially symmetrical with respect to the optical axis. Figs. 188 and 189 show an example of the shape of one of the electrodes 725.

FIGS. 188 and 189 are perspective views of the electrode 725, and FIG. 188 is a perspective view showing the electrode 725 in an axisymmetric cylindrical shape, and FIG. 189 is a view showing the electrode 725 in an axisymmetric disc shape. Oblique view.

In the present embodiment, as shown in Fig. 188, the electrode 725 is described as a cylindrical shape. However, the electrode 725 may be substantially axisymmetrical to the optical axis of the electron beam, and may have a disk shape as shown in Fig. 189.

Further, the electrode 725 is applied with a predetermined voltage (negative potential) through the power source 726, and in order to generate an electric field for preventing discharge between the objective lens 724 and the wafer W, the predetermined voltage is higher than the voltage applied to the wafer W (this embodiment) In the form, it is in a grounded state, so the potential is 0V) is low. The potential distribution between the wafer W and the objective lens 724 at this time will be described with reference to FIG.

Figure 190 shows a graph showing the voltage distribution between the wafer W and the objective lens 724.

In Fig. 190, the voltage distribution from the wafer W to the position of the objective lens 724 is displayed with the position of the electron beam irradiation optical axis as the horizontal axis.

In the conventional electron beam apparatus in which the electrode 725 is not provided, the voltage distribution from the objective lens 724 to the wafer W is the maximum value of the voltage applied to the objective lens 724, and the wafer W is smoothly changed until it is grounded ( "Practical Technology" shown in Figure 190).

On the other hand, in the electron beam apparatus of the present embodiment, the electrode 725 is disposed between the objective lens 724 and the wafer W, and the electrode 725 is applied through the power source 726. A predetermined voltage (negative potential) lower than the voltage applied to the wafer W is applied to weaken the electric field of the wafer W. ("Electrode" shown in Figure 190)

Therefore, in the electron beam apparatus of the present embodiment, the electric field is not concentrated in the vicinity of the through hole b of the wafer W, and a high electric field is not formed. Further, even if the electron beam is irradiated to the through hole b and the electrons are emitted twice, the emitted secondary electrons are not accelerated to the extent that the residual gas is ionized, so that discharge between the objective lens 724 and the wafer W can be prevented.

Further, since discharge between the objective lens 724 and the through hole b can be prevented, there is no possibility that the pattern of the wafer W or the like is broken by the discharge.

Further, in the above embodiment, the discharge between the objective lens 724 and the wafer W having the through hole b can be prevented. However, since the counter electrode 725 is applied with a negative potential, the detector 761 can be made to have a negative potential. The case where the sensitivity of the second electronic detection is lowered. Therefore, when the detection sensitivity is lowered, as long as the electron beam is irradiated a plurality of times and a series of operations of the second electrons are detected as described above, the obtained plurality of detection results are subjected to cumulative addition and averaging processing to obtain a predetermined detection. Sensitivity (S/N ratio of signal) is sufficient.

In the present embodiment, the detection sensitivity is described as a signal-to-noise ratio (S/N ratio) by way of an example.

The above-described second electronic detecting operation will be described with reference to Fig. 191.

Figure 191 is a flow chart showing the second electronic detection operation of the electron beam apparatus.

First, the secondary electrons from the sample to be inspected are detected by the detector 761 (step 1). Next, a determination is made as to whether the signal-to-noise ratio (S/N ratio) is equal to or greater than a predetermined value (step 2). In step 2, the signal-to-noise ratio is at a predetermined value. In the above case, since the second-time electronic detection performed by the detector 761 is sufficient, the secondary electronic detection operation is completed.

On the other hand, in step 2, when the signal-to-noise ratio is less than the predetermined value, the electron beam is irradiated to detect a series of secondary electrons for 4N times, and the averaging process is performed (step 3). Here, since the initial value of N is set to "1", in step 3, the detection operation of the secondary electrons is performed four times for the first time.

Next, the N addition "1" is counted (step 4), and in step 2, the judgment of whether the signal-to-noise ratio is equal to or greater than a predetermined value is performed again. Here, when the signal-to-noise ratio is less than the predetermined value, the process proceeds to step 3 again, and this time, the detection operation of the second electron is performed 8 times. Then, the number of N is incremented until steps 2-4 are repeated until the signal-to-noise ratio is above a predetermined value.

Further, in the present embodiment, a predetermined voltage (negative potential) lower than the voltage applied to the wafer W is applied to the counter electrode 725 to prevent the wafer W having the through hole b from being discharged, but there are two electrons. The detection efficiency is reduced.

Therefore, when the sample to be inspected is a wafer having no through-hole or the like, and the sample to be inspected such that discharge is less likely to occur between the objective lens 724, the voltage applied to the electrode 725 can be controlled to cause the detector to be used. The electrical detection efficiency of the 761 is improved twice.

Specifically, in the case where the sample to be inspected is grounded, the voltage applied to the electrode 725 is set to a predetermined voltage higher than the voltage applied to the sample to be inspected, for example, at +10 V. Moreover, at this time, the distance between the electrode 725 and the sample to be inspected is set between the electrode 725 and the sample to be inspected. The distance from the discharge.

At this time, the secondary electrons generated by irradiating the electron beam to the sample to be inspected are accelerated toward the electron beam source 721 side by the electric field generated by the voltage applied to the electrode 725. Further, since the electric field generated by the voltage applied to the objective lens 724 is accelerated toward the electron beam source 721 side and is converged, a large amount of secondary electrons can be incident on the detector 761 to improve the detection efficiency.

Further, since the electrode 725 is axisymmetric, it also has a lens action for converging electron beams that are irradiated onto the sample to be inspected. Therefore, the primary electron beam can be contracted to be finer by the voltage applied to the electrode 725. Further, since the primary electron beam can be narrowed by the electrode 725, it can be combined with the objective lens 724 to form an objective lens system having a lower aberration. To the extent that such a lens may function, the electrode 725 may be substantially axisymmetric.

According to the electron beam apparatus of the above embodiment, the electrode to be inspected and the objective lens are provided with an electrode which is substantially axisymmetric with respect to the electron beam irradiation axis to control the electron beam irradiation surface of the sample to be inspected. The electric field strength controls the electric field between the sample to be inspected and the objective lens.

Further, an electrode is provided between the sample to be inspected and the objective lens, and is substantially symmetrical with respect to the electron beam irradiation axis to weaken the electric field intensity of the electron beam irradiation surface of the sample to be inspected, thereby eliminating the Check the discharge between the sample and the objective.

Further, since the secondary electrons can be efficiently passed through the objective lens without changing the voltage applied to the objective lens, the detection efficiency is improved, and a signal having a good S/N ratio can be obtained.

Furthermore, it can be controlled to weaken the inspection according to the type of the sample to be inspected. The voltage of the electric field strength of the electron beam irradiation surface of the sample.

For example, when the sample to be inspected is a sample to be inspected which is easily discharged from the objective lens, the electric field intensity of the electron beam irradiation surface of the sample to be inspected is further weakened by changing the voltage of the electrode, and discharge can be prevented.

Moreover, the voltage applied to the electrodes can be changed depending on whether or not the semiconductor wafer has a through hole, that is, the voltage for weakening the electric field intensity of the electron beam irradiation surface of the semiconductor wafer is changed.

For example, when the sample to be inspected is a sample to be inspected which is easily discharged from the objective lens, the electric field intensity of the electron beam irradiation surface of the sample to be inspected is further weakened by changing the electric field generated by the electrode, thereby preventing the special It is the discharge around the through hole or through hole.

Further, since the discharge between the through hole and the objective lens can be prevented, there is no possibility that the pattern of the semiconductor wafer or the like is broken by the discharge.

Further, since the electric potential supplied to the electrode is lower than the electric charge supplied to the sample to be inspected, the electric field intensity of the electron beam irradiation surface of the test sample can be weakened, and discharge to the test sample can be prevented.

Further, since the potential supplied to the electrode is set to a negative potential and the sample to be inspected is grounded, the electric field intensity of the electron beam irradiation surface of the sample to be inspected can be weakened, and discharge to the sample to be inspected can be prevented. Further, the present embodiment can also be applied to the above-described first to third embodiments and embodiments in which no number is given.

(Embodiment 34)

In the present embodiment, a modified embodiment of the inspection apparatus of the present invention shown in Fig. 8 will be described. Figure 192 shows a schematic display of this book. An imaging device of a pre-charging unit of an embodiment. The imaging device includes a primary optical system 72, a secondary optical system 74, a detection system 76, and a charge control means 840 for equalizing or reducing the charge of the object. In the present embodiment, the same configurations as those of the above-described first embodiment will be omitted.

The charge control means 840 for homogenizing or reducing the charge carried by the object, in this embodiment, is provided with an electrode 841 between the object W and the electrostatic lens 724 of the primary optical system 72 closest to the object W, The switch 842 is electrically connected to the electrode 841; the voltage generator 844 is electrically connected to one of the switches 842; and the charge detector 846 is electrically connected to the switch 842. One terminal 845. Charge detector 846 has a high impedance. A timing generator 849 is formed to issue an operation timing command to the CCD 762 and the image processing unit 763 of the detection system 76, the changeover switch 842 of the charge reduction means 840, the voltage generator 844, and the charge detectors 846 and 848. Further, the present embodiment can also be applied to the above-described first to third embodiments and embodiments in which no number is given.

(Embodiment 35)

- Device manufacturing method -

Next, an embodiment of a method of manufacturing a semiconductor device of the present invention will be described with reference to FIGS. 193 and 194.

Figure 193 is a flow chart showing a method of fabricating a semiconductor device in accordance with an embodiment of the present invention. The manufacturing steps of this embodiment comprise the main steps described below.

(1) A wafer manufacturing step of manufacturing a wafer (or a wafer preparation step of preparing a wafer) (step 1400)

(2) a mask manufacturing step of manufacturing an exposure mask (or a mask preparation step of preparing a mask) (step 1401)

(3) A wafer processing step of performing necessary processing on the wafer (step 1402)

(4) A wafer assembly step of cutting out one piece of the wafer formed on the wafer to make it work (step 1403)

(5) A wafer inspection step of inspecting the completed wafer (step 1404)

In addition, each of the above main steps is further comprised of several steps.

Among these main steps, the wafer processing step of (3) is decisive for the performance of the semiconductor device. In this step, the designed circuit patterns are sequentially stacked on the wafer to form a plurality of wafers that operate as a memory or a microprocessor. The wafer processing step includes the following steps.

(A) a thin film forming step (using CVD, sputtering, etc.) as a dielectric thin film or wiring portion of an insulating layer or a metal thin film or the like as an electrode portion

(B) an oxidation step for oxidizing the film layer or the wafer substrate

(C) Photolithographic imprinting step for forming a resist pattern by using a reticle for selective processing of a thin film layer or a wafer substrate or the like

(D) an etching step of processing a thin film layer or substrate in accordance with a resist pattern (using, for example, dry etching)

(E) ion impurity implantation diffusion step

(F) Repellent stripping step

(G) Steps to check the processed wafer

In addition, the wafer processing step repeats only the necessary number of layers to fabricate a semiconductor device that can operate according to design requirements.

Figure 194A is a flow chart showing the lithographic imprinting step as the core of the wafer processing step of Figure 193. The lithography imprinting step includes the following steps.

(a) a resist coating step of applying a resist on the wafer on which the circuit pattern is formed in the previous step (step 1500)

(b) a step of exposing the resist (step 1501)

(c) a developing step of developing the exposed resist to obtain a resist pattern (step 1502)

(d) an annealing step for stabilizing the developed resist pattern (step 1503)

The above-described semiconductor device manufacturing steps, wafer processing steps, and lithography imprinting steps are well-known techniques and need not be further described.

When the defect inspection method and the defect inspection device of the present invention are used in the inspection step of the above (G), the semiconductor device having the fine pattern can also be inspected with a good throughput, or can be inspected in full, and the product yield can be improved. , the prevention of defective product delivery.

- inspection procedures -

The inspection procedure statement in the inspection step of the above (G) is as follows.

The defect inspection device that generally uses electron ray is very expensive, and the throughput is lower than other processing devices. Therefore, at this stage, it is an important step (such as etching, film formation, or CMP) that is considered to be the most inspectable. The defect inspection device is used only after mechanical grinding), flattening, etc.).

After the wafer to be inspected is placed on the ultra-precision XY stage by the atmospheric handling system and the vacuum handling system, the electrostatic chuck mechanism is utilized. It is fixed, and then the defect inspection is performed according to the procedure (Fig. 194B). First, the position of each crystal grain or the height of each part is detected by an optical microscope as needed, and it is memorized. The optical microscope acquires an optical microscope image of a portion to be observed such as other defects, and is also used for comparison with an electron beam image. Next, after the processing manual information input device corresponding to the type of wafer (after which step or wafer specification is 20 cm or 30 cm, etc.), the following inspection site designation, electro-optical system setting, inspection condition setting, etc. are usually performed. Performs an immediate defect check while performing image acquisition. The comparison between the cells and the grain comparison is performed by a high-speed information processing system having an algorithm, and the result is output to a CRT or the like as needed, and the memory is performed to the memory. Defects include particle defects, abnormal shape (pattern defects), and electrical defects (such as disconnection or poor conduction of wiring or through holes). It is also possible to distinguish these defects and automatically classify the defect size, fatal defect (the wafer reaches a major defect that cannot be used, etc.). The detection of electrical defects can be achieved by detecting contrast irregularities. For example, a portion having poor conductivity can be positively charged by electron beam irradiation (about 500 eV), and the contrast is lowered, so that it can be distinguished from the normal portion. In this case, the electron beam irradiation means means a low-potential (energy) type electron beam generating means (produces hot electrons, UV) which is provided in addition to the electron beam irradiation means for normal inspection. /photoelectron). The electron beam of the low potential (energy) is generated and irradiated before the inspection target region is irradiated with the electron beam for inspection. When the electron beam for illumination inspection itself is a positively charged image projection method, it is not necessary to separately provide a low-potential electron beam generation means depending on the form. In addition, through the wafer and other samples The potential is positive or negative with respect to the reference potential (because the flow is easily changed depending on the forward or reverse direction of the element), and the contrast is different, and defect detection can be performed. It can also be used for line width measuring devices and alignment accuracy measurement.

All of the above-described embodiments can be applied when performing the inspection and inspection procedures necessary for the above processes. Furthermore, it can also be applied to all device systems having the features of the functional mechanisms of Figs. 1 to 25 as stated in the device system. Thereby, a very efficient inspection can be carried out in the wafer or mask process steps. Further, the present embodiment can also be applied to the above-described first to third embodiments and the embodiment in which the number is not assigned.

(Embodiment 36)

HDD (hard disk drive) substrate, magnetic head component inspection

The present invention is not limited to wafers and exposure masks, and can also be applied to inspection of substrates for HDDs. Although the application examples are set forth below, they are the same as those used for semiconductor wafers or masks.

For example, in the substrate for HDD, the top layer of the magnetic layer of the glass substrate or the aluminum substrate is usually a magnetic body having a thin lubricating layer thereon. Two types of inspections can be performed on the substrate. One is the inspection of foreign matter, particle adhesion and scratches during substrate manufacture. The other is an examination of the abnormality of the film formed on the surface.

When the aluminum or glass substrate itself has scratches or foreign matter-particles adhered after being produced or washed, the magnetic film formed on the upper portion cannot be normally formed. Because the flatness will be deteriorated or unevenness will occur. However, in recent years, high-density media, the amount of floating of the substrate and the magnetic head has reached about 5 nm, and the degree of unevenness must be smaller than the amount of floating. That is, it is necessary to prevent foreign matter adhering to a specification of 5 nm. The same is true for the bumps. Thus, the inspection device of the present invention can be used It is applied to the inspection of such foreign objects or scars, and the inspection is performed at high speed and high sensitivity. Its principle function is as described above.

Further, when a problem occurs in the formation of the substrate, for example, when the protective film has pinholes or a magnetic film component is not uniform, the potential distribution on the film may be negatively uniform. For example, when a certain amount of electric charge is supplied to the surface of the substrate to the extent that no damage is caused, a uniform surface potential is formed if it is a homogeneous magnetic layer. However, when the protective film has pinholes or the magnetic layer is not homogeneous, the surface potential is not uniform. At this time, by performing an inspection by increasing the resolution (reducing the Px specification), it is possible to observe and measure the ratio of the component of the mirror-reflected electron and the secondary electron-emitting electron corresponding to the surface potential. That is, if the unit cell/cell inspection is applied, if there is uniformity, the component ratio of the electronic signal generated by the part will be small. If there is no uniformity, the component ratio will be different, and it can be used as a defect. Test it.

Regarding the method, the method and apparatus of the present invention can be used to perform defect inspection of the HDD substrate, but the invention is not limited thereto. Compared with the prior art, the method and apparatus of the present invention can perform inspections with high speed and high sensitivity. Because the object defects have reached an ultra-fine level, the use of conventional optical inspection devices is not sensitive enough, and the use of a higher resolution SEM takes a lot of time.

Moreover, similarly, defect inspection can also be performed in the head process. In the magnetic head process, the same procedure as that of the semiconductor wafer is also employed, so that the high-efficiency defect inspection as described above can be performed for shape defects and poor film quality. Further, the present embodiment can also be applied to the first to third embodiments and the embodiment in which the number is not assigned.

(Embodiment 37)

- Platform device -

The platform device used in the inspection apparatus and inspection method of the present invention will be described.

Figures 195A and 195B show the construction of the platform device.

As shown in FIG. 195A and FIG. 195B, the platform apparatus 1 of the inspection apparatus of the present invention includes a Y-axis bottom plate 2 disposed on the bottom wall of the casing 4, and a Y platform 5 disposed in parallel in the Y-axis direction. The guide rail 3 on the Y-axis floor plate 2 is guided to move in the Y-axis direction; and the mask plate 8 is mounted on the X-platform 7 by an X-platform 7 that is movable in the XY-axis direction, and the X-platform 7 is The X guide rail 6 disposed on the Y stage 5 in parallel with the X-axis direction perpendicular to the Y-axis direction is guided to move in the X-axis direction. The main function of the platform device 1 is to inspect the mask 22 on which the mask 8 is mounted, and to check the light 26 with respect to the electron beam irradiation from the cylinder 21 of the electro-optical system device, so that the mask 22 is used. The X-axis direction scanning movement of the X stage 7 and the repetitive motion of the Y-axis direction step movement of the Y stage 5 are moved in a predetermined area. The X platform 7 is scanned in the X-axis direction in accordance with the speed and the moving distance with a predetermined moving direction, and the Y-platform 5 is stepped in the Y-axis direction in accordance with a predetermined moving direction and moving distance. Here, the mask 22 is fixed to a tray (not shown), and the tray is fixed by an electrostatic chuck (not shown) provided on the mask 8. Further, the casing 4 on which the platform device 1 is placed is transmitted through the surface of the flat plate 24 which is supported by the four portions of the vibration isolation table 23, and the influence of external vibration from the floor panel 25 is reduced. Furthermore, the platform device 1 is covered by the casing 4, And at 10^-4Pa is used in a surrounding atmosphere with a vacuum around. Therefore, in order to prevent generation of gas, heat, or dust as much as possible, the X servo motor 9 and the Y servo motor 11 of the drive system are disposed on the outer side of the casing 4. The X platform 7 and the Y platform 5 are driven by the side wall of the casing 4 and the X-drive shaft 10 and the Y-drive shaft 12 provided with a vacuum seal. The control of the X-servo motor 9 uses the encoder 27, similarly. The control of the Y servo motor 11 uses the encoder 28. Further, the position measurement of the mask plate 8 on which the mask 22 is mounted is provided with a laser interference meter system which is composed of an X-platform mirror 19 and X interference disposed on the X-axis side of the mask plate 8. An interferometer 13; an X interferometer stage 15 supporting the X interferometer 13; a Y stage mirror 20 and a Y interferometer 14 disposed on the Y-axis side; and a Y interferometer stage 16 supporting the Y interferometer 14; It is composed of an optical element such as a laser head (not shown) and an AXIS board (not shown) for performing photoelectric signal conversion, and the X platform is measured with high accuracy by measuring X long light 17 and Y long light 18 7, Y platform 5 each location. For the X servo motor 9 and the Y servo motor 11 of the drive system, the platform control system (not shown) performs the feedback control of each axis by the XY axis direction position signal obtained by the laser interference meter system. High-precision platform positioning control at micron level.

In this embodiment, the scanning movement of the mask inspection is set to the X-axis direction, and the step movement is set to the Y-axis direction. However, the scanning movement can be set to the Y-axis direction in accordance with the mask inspection direction, and the step movement is set to the X-axis direction. In addition, the driving system of the platform device 1 can also utilize a high-non-magnetic air pressure driving mechanism, or maintain true, from the manner of processing electron beam inspection light. In the differential venting mode of the air atmosphere, the platform rail is set as a gas hydrostatic bearing, and each platform is set as a non-contact platform device 1.

(Embodiment 38)

- Laser irradiation position control -

For example, in the configuration shown in Fig. 35, it is necessary to irradiate the center of the spot of the laser light at a predetermined position on the photoelectron surface 2021. This is because electrons (photoelectrons) are generated from the position of the spot, and the position is made into an electron generating position. The electrons (photoelectrons) generated from this position are irradiated onto the sample surface through the primary system. At this time, if an electron (photoelectron) is to be incident on the nearest lens, it must pass through the center position of the lens. If deviated from the center of the lens, the orbit of the electron (photoelectron) will be distorted. When the distortion of the track is large, the electron (photoelectron) may hit the wall of the column or exceed the track correction range formed by the aligner (bias), and the track correction may not be obtained. When there is no aligner between the lens and the photoelectron generating portion, the passing trajectory in the lens is determined by the position of the photoelectron generating portion. That is, when the laser illuminates the position of the deviation and generates photoelectrons from the off position, the electron beam does not pass through the center of the lens.

In the present embodiment, in order to solve this problem, a configuration example in which the center of the laser spot can be irradiated to a predetermined position of the optoelectronic surface 2021 shown in Fig. 35 is explained using Fig. 199. As shown in the schematic cross-sectional view of the photoelectron surface 2021 of FIG. 199(a), the optoelectronic surface 2021 is provided with a substrate material 20211, an optoelectronic material 20212, a conductive material 20213, a holding member 20214, and a laser irradiation aperture 20215. The matrix material 20211 is a light transmissive member such as quartz, quartz glass, quartz glass, or magnesium fluoride glass. The optoelectronic material 20212 uses a material having a lower work function such as ruthenium or gold (photoelectron generation) An efficient material) is suitable and is covered by the substrate 20211. The conductive material 20213 is preferably made of a material having low conductivity such as chromium. The holding member 20214 is made of a conductive material for holding the substrate 20211 and the like. As shown in Figure 199(a). The optoelectronic material 20212, the conductive material 20213, and the holding member 20214 are electrically connected. The laser irradiation aperture 20215 can also be electrically connected to these elements.

The laser irradiation aperture 20215 is preferably made of a reflective material such as molybdenum or tantalum, and is provided on the laser incident side of the substrate 20211. It is preferable that the surface of the laser irradiation aperture 20215 is excellent in surface roughness, and the reflection intensity of the laser is stronger (higher) than the optoelectronic material 20212. For example, mirror polishing or surface roughness Ra1 μm or less is preferable. Further, in this example, as shown in the upper diagram of the laser irradiation diaphragm 20215 in Fig. 199(b), the laser irradiation aperture 20215 is at the center of the disk-shaped member having a diameter d1 (about 3 to 5 mm). The portion is provided with an inner diameter region 20216 of diameter d2. When the irradiated laser is reflected by the laser irradiation aperture 20215, the reflected intensity of the reflected light is stronger than that of the reflected light reflected by the optoelectronic material 20212 which will be described later. On the other hand, when the laser passes through the inner diameter region 20216 and is reflected by the optoelectronic material 20212, the reflected light intensity is weaker than that reflected by the laser irradiation aperture 20215. Further, the intensity of the reflected light may be measured by a light meter provided on the optical path of the reflected light.

The laser is a DUV laser, and for example, a laser having a wavelength of 266 nm or a wavelength of 244 nm can be used. Solid or gas lasers can also be used. It is not limited to lasers, but it can also use light that emits wavelengths below 270 nm. In the case of a solid laser, a harmonic such as a 4x wave or a 5x wave of a YAG laser can be used. Shoot. Moreover, in the case of gas lasers, an Ar ion laser or a quasi-molecular laser can be used.

When the laser passes through the inner diameter region 20216 of the pupil 2015 by the laser, it reaches the optoelectronic material 20212 to generate photoelectrons. The laser irradiation position is changed from the position in the inner diameter region 20216 (for example, first, the laser irradiation position is appropriately changed, and the laser irradiation position is changed to the position of the inner diameter region 20216 from the change in the reflected light intensity shown below. ), the optical system such as the mirror is controlled by the control system to move the irradiation position in the +X direction. When the laser irradiation position reaches the end portion of the inner diameter region 20216 (the inner diameter side end portion of the laser irradiation aperture 20215), as shown in Fig. 199(c), the reflected light intensity measured by a light meter or the like Will rise. The position of the end (x1, y1) can be recorded using a control system. The optical system such as the mirror is controlled by the control system to move the irradiation position in the -X direction. When the laser irradiation position reaches the opposite end portion of the inner diameter region 20216 (the inner diameter side end portion of the laser irradiation aperture 20215), the intensity of the reflected light measured by the light amount meter or the like increases. The position of the end (x2, y2) can be recorded using a control system. The amount of mirror angular movement when moving from (x1, y1) to (x2, y2) by this action can be recorded by the control system. In the example shown in Fig. 199(c), (x1, y1) corresponds to PL, and (x2, y2) corresponds to PR. The coordinate movement amount of each minimum memory unit (minimum adjustment amount or control amount) of the mirror movement adjustment can be calculated by the control system. The x direction is Δx, and the y direction is Δy. For example, each minimum adjustment amount (memory unit) is 5 μm or the like.

At this time, the coordinates of the portion on the inner diameter side end portion 4 of the laser irradiation diaphragm 20215 can be recorded by the control of the control system. For example, up, down, left and right Coordinates such as (PL, PR, PU, PD) can be recorded. Thereby, the center C(0, 0) of the inner diameter region 20216 can be determined. Further, even if it is not the coordinates of PL, PR, PU, or PD, if the coordinates of the end portion 2 when the irradiation position is moved in the x direction and the position of the end portion 2 when the irradiation position is moved in the y direction can be obtained, the center C(0,0) can be determined. The coordinates of the two parts in the x direction are (xa, y0) and (xb, y0), and when the coordinates of the two parts in the y direction are (x0, ya) and (x0, yb), the center C can be determined as (( Xa+xb)/2, (ya+yb)/2).

Thereafter, the control system can identify the inner diameter region 20216, that is, the position within the four coordinates, the electron beam passes through the laser irradiation position at the center of the lens, that is, the photoelectron material 20212 on the central axis of the lens. The coordinates of the position P(x, y). In this way, the control system can grasp the position of the laser irradiation, that is, the coordinates of the irradiation position are obtained and recorded. Thereby, even when the positional relationship of the laser, the mirror, and the lens (photoelectric generating device 2020) is changed, the control system can again irradiate the laser to the position P(x, y). The control of the laser irradiation position is performed before the inspection of the sample. Further, the present embodiment can also be applied to the first to third embodiments and the embodiment in which the number is not assigned.

Further, the optoelectronic surface 2021 may be formed as another configuration example as shown in FIG. The side of the position of the optoelectronic material 20212 of the substrate material 20211 shown in Fig. 199(a) has a shape which is maintained by the holding member 20214, but in the example of Fig. 200 of the present embodiment, it is formed. It is a plane. Further, the holding member 20214 holds and holds the substrate 20211 and the like from both sides by means of an element 20217 such as a screw.

(Embodiment 39)

-1 axis adjustment of the system -

In the above-described Embodiment 38, when the laser irradiation position is adjusted to the position P (x, y), the method of setting the orbit of the electron beam through the lens center will be described. For example, in the photoelectric generating device 2020 shown in Fig. 35, when the orbit of the electron beam passes through the center of the lens, as shown by EB1 in Fig. 201, even if the power (lens power) of the lenses 2022, 2023, 2024 is changed, the passage is passed. The electron beam orbit behind the lens will not change. On the other hand, when the orbit of the electron beam passes the position deviated from the center of the lens, if the lens power is changed, as shown by EB2 and EB3 in Fig. 201, the electron beam orbit passing through the lens changes. The configuration of this embodiment utilizes this feature.

The photoelectric generating device 2020 of the present embodiment shown in Fig. 201 has the same configuration as that of Fig. 35. The measurement is performed using the aperture 2040 and the aligner 2030 as a measurement. One of the aligners 2030 can be used, such as 2031 or 2032, 2033. (At this time, the numerical aperture 2025 can also use a large-sized aperture that does not cause a problem in measurement, for example, φ 500 to φ 2000 μm). The measurement aperture 2040 is configured to measure the absorption current generated by the electron beam irradiation result.

In order to determine the offset vector (for example, bias voltage or bias current required) of the electron beam trajectory from the hole end portion of the measurement pupil 2040 to the opposite end portion using the aligner 2030 (for example, 2031 in FIG. 35), The control system will control it. That is, as shown in Fig. 201, the bias voltage of the aligner 2030 is changed to continuously shift the electron beam orbit, and the measurement beam 2040 is hit from the electron beam orbit (the electron beam does not pass through the measuring beam). State of the hole 2040 (EB3) electron beam orbit passes through the state of the hole of the measuring aperture 2040 (EB1, EB2) The state in which the electron beam orbitals hit the measurement aperture 2040 (EB4), and the absorption current of the measurement aperture 2040 with respect to the deflection voltage of the aligner 2030 is measured. When this measurement is performed, as shown in Fig. 202, the absorption current of the measurement aperture 2040 is measured, and the "electron beam full absorption (absorption current is large)" "electron beam due to the passage hole" is formed. Reduced absorption (small current absorption)" - "full absorption of electron beam (large current absorption)". The lens power (GL power) was changed a plurality of times to perform measurement.

As shown in Fig. 202(a), in the case of large and small GL power, the partial vector with the largest decrease in absorption current (the BA voltage (bias voltage, the same below) that absorbs the minimum current) is the electron beam. By measuring the orbit of the center of the hole of the aperture 2040. On the other hand, as shown in Fig. 202(b), in the case where the GL power is large or small, when the BA voltage at which the sink current reaches the minimum is different, the track system is shifted from the center of the lens. The control system changes the position of the laser irradiation while obtaining the position where the BA voltage at which the GL power is large and the minimum absorption current reaches the same voltage, that is, the electron beam orbit passes through the laser irradiation position at the center of the lens, and The coordinates are recorded as the electron beam passing through the laser irradiation position P(x, y) at the center of the lens. Further, the present embodiment can also be applied to the first to third embodiments and the embodiment in which the number is not assigned.

(Embodiment 40)

- Laser irradiation specification control -

As described in the above-described Embodiments 38 and 39, in addition to the control of the laser irradiation position irradiated on the optoelectronic surface 2021, the laser irradiation specification (light spot diameter) is an important parameter that affects the specification of the electron beam irradiated to the sample. In the light spot In terms of adjustment, sometimes the laser output from the light source can only be adjusted to the desired specifications using only the lens and the mirror. The spot diameter 2 ω 0 is represented by 2 ω 0 = (4 λ / π) (F / D). Here, λ is the wavelength of light, F is the focal length of the lens, and D is the laser diameter at the lens position. From this equation, the spot diameter system is proportional to the focal length and inversely proportional to the laser diameter at the lens position. Therefore, in order to reduce the spot diameter, there is a method in which a laser beam from a light source is enlarged by a beam expander and then incident on a lens, and a method using a short focus lens. The method for reducing the spot diameter and the control method of the aforementioned laser irradiation position are effective for appropriately adjusting the irradiation position and the spot diameter of the laser.

An example in which a beam expander is used will be described with reference to Fig. 203. As shown in Fig. 203, the laser beam having the diameter Φ d1 output from the light source 10000 is magnified A by the beam expander 810 to become a laser having a diameter Φ d2 and incident on the lens 820 of the lens focus F1. The laser is reflected by the mirror 830 and passes through the transparent window 840 provided by the vacuum vessel 850 for maintaining the vacuum, and reaches the optoelectronic material 2012 disposed at the corresponding position of the lens focus F1. Here, the spot diameter of the laser light of the optoelectronic material 20212 is at least 2 ω 0 = (4 λ / π) (F1/Φ d2). In this example, the laser system is set to CW (Continuous Wave) of λ = 266 nm. Further, the adjustment of the laser irradiation position of Embodiment 38 can be adjusted by changing the angle of the mirror 830 or the like. Further, a moving mechanism 825 for moving the lens 820 along the laser optical axis may be provided. The laser spot diameter of the optoelectronic material 20212 can also be changed by the movement of the lens 820 by the moving mechanism 825.

Regarding the case of a beam expander in this case or the absence of a beam expander, the relationship between the focal length of the lens and the minimum spot diameter is shown in Figure 204. As shown in Fig. 204, the longer the focal length of the lens, the larger the spot diameter, and in the case of a beam expander, the spot diameter is smaller than in the case of no beam expander.

Next, an example in which a short focus lens is used between the mirror 830 and the vacuum container 850 will be described using FIG. As shown in Fig. 205, the laser beam of the diameter Φ d1 output from the light source 10000 is reflected by the mirror 830 and incident on the lens 821 of the lens focus F2. The laser passes through a transparent window 840 provided by vacuum vessel 850 for maintaining vacuum and reaches optoelectronic material 2012 disposed at a corresponding location of lens focal point F2.

Further, a moving mechanism 825 for moving the lens 821 along the laser optical axis may be provided. The laser spot diameter of the optoelectronic material 20212 can also be changed by the movement of the lens 821 by the moving mechanism 825. Moreover, because the lens 821 is a short focus lens, its position will be closer to the optoelectronic material 20212 side than the mirror 830. Therefore, when the laser irradiation position is adjusted by the mirror 830, the laser may deviate from the center of the lens 821. Thus, to correct for this deviation, the moving mechanism 825 can be configured to move the lens 821 in a plane that is normal to the optical axis. Further, the beam expander 810 (not shown in Fig. 205) described in Fig. 203 may be used in combination.

Further, a lens 822 having a focal length shorter than the lens 821 (having a lens focus of F3) can also be used. The lenses 820 and 821 are provided on the outside air side of the vacuum container 850. However, as shown in Fig. 206, a lens 822 may be provided inside the vacuum container 850. The lens 822 can also be provided with a moving mechanism 825 as previously described. In this case, the beam expander 810 (not shown in Fig. 205) described in Fig. 203 may be used in combination. In addition, this embodiment is also applicable The embodiments 1 to 39 and the embodiments in which the numbers are not assigned are applied.

(Embodiment 41)

In the photoelectric generating device 2020 shown in Fig. 35, the aligner 2060 is additionally provided in the subsequent stages of the lens groups 2022, 2023, and 2024, as shown in Fig. 207. In the present embodiment, the description of the same portions as those shown in Fig. 35 will be omitted. The aligner 2060 has a first aligner 2061 and a second aligner 2062. The first aligner 2061 and the second aligner 2062 are disposed between the third segment lens 2024 and the numerical aperture 2025, and are statically operated in the same manner as the first aligner 2031 and the second aligner 2032. Aligner. However, as described above, the first aligner 2031 and the second aligner 2032 are used to control the position of the electron beam irradiated toward the sample, and the first aligner 2061 and the second aligner 2062 are used. It is to control the electron beam through the center of the hole of the numerical aperture 2025. Further, the present embodiment can also be applied to the above-described first to tenth embodiments and the embodiments in which the numbers are not assigned.

(Embodiment 42)

- Photoelectric surface 1 system with Zoom function -

In the configuration shown in Fig. 35, Fig. 207, and the like, an amplification function for controlling the specification of the electron beam irradiated to the sample may be provided. Here, the case where the configuration of the 207th embodiment described in the 41st embodiment has an enlargement function will be described using FIG. In the present embodiment, the same components as those shown in Fig. 207 are denoted by the same reference numerals.

The configuration shown in Fig. 208 is a lens group provided between the numerical aperture 2025 and the first aligner 2031, in addition to the configuration shown in Fig. 207. 2091, 2092, 2093 (one lens is composed of three electrodes). In Figs. 35, 207, and 208, the electrodes 2022, 2023, and 2024 constitute one lens group in such a configuration. In the embodiment of FIG. 208, the lenses 2022, 2023, and 2024 are referred to as EL1 (EL is an abbreviation of Electrostatic Lens, the same applies hereinafter), and lenses 2091, 2092, and 2093 are referred to as EL2, and aligner 2060. Referring to aligner 1, aligner 2030 is referred to as aligner 2. Further, a numerical aperture may be provided between the lens 2093 and the first aligner 2031. At this time, the primary optical system 2000 has two numerical apertures.

By adopting the configuration of the magnifying lens of this embodiment, it is possible to control the specification of the electron beam irradiated to the sample. This is to control the size of the electron beam irradiated to the sample by the amplification function of EL1 and EL2 under the same conditions by the laser irradiation specification irradiated on the photoelectron surface 2021. For example, the laser specification irradiated to the optoelectronic surface 2021 can be controlled from about 0.1 to about 30.

The necessity of changing the electron beam size on the sample surface is necessary in accordance with the magnification of the secondary system (the optical system forming the electronic image of the sample). When the magnification of the system is changed twice, the specification of the field of view on the sample surface (the area where the detector is used as an electronic image) also changes. Therefore, the specification of the electron beam also needs to be changed in response to the change in the magnification. For example, when the field of view is changed from 30 × 15 μm to 200 × 100 μm, the specifications of the electron beam are also changed to the specifications to be covered. For example, an electron beam irradiated with a circular or rectangular shape of 60 × 30 μm must be changed into a circular or rectangular electron beam of 300 × 150 μm.

At this time, although the laser irradiation specification of the optoelectronic surface 2021 can also be changed. Correspondingly, in order to change the laser irradiation specifications, it is necessary to replace and adjust the laser optical system, which is time consuming. Moreover, when a small laser spot diameter corresponding to a small field of view is formed, the laser density changes, thereby generating The change in the amount of photoelectrons. Furthermore, the amount of photoelectrons is also unstable. At this time, if the amplification function of the primary system described above is used, the control of the electron beam irradiation region on the sample surface can be performed even if the same laser irradiation specification is used. Therefore, the configuration having this amplification function is very effective.

The configuration of this embodiment is an example in which two aligners such as the aligner 1 and the aligner 2 are provided. The aligner 1 is used to pass the orbit of the electron beam through the centers of the numerical apertures 2025 and EL2. For the center of the lens, there are also cases where it is used in combination with the aligner in the system twice.

An example of the voltage applied to each component shown in Fig. 208 is shown. The voltage of the photoelectron surface 2021 is V1, and the voltage of the first extraction electrode 2022 is V2, and the voltage of the second extraction electrode 2023 is V3, and the third extraction electrode is used for the voltage of the electrode constituting the extraction lens. The voltage 2024 is V4, (here, one electrostatic lens is formed by the electrodes 2022, 2023, and 2024), the voltage of the numerical aperture 2025 is V5, and the voltage of the third aligner 2033 is V6. The voltage of the lens electrode 2091 is V6, the voltage of the lens electrode 2092 is V7, the voltage of the lens electrode 2093 is V8, and the voltage of the aperture 2040 is V9. Furthermore, the wafer surface voltage (also known as the deceleration voltage) is set to RTD. In the primary optical system 2000 of the present embodiment, when the voltage V1 of the optoelectronic surface 2021 is described as a basis, a voltage is applied to each component as follows. That is, when LE is low, V1=RTD-10V to RTD+5V, V2, V4, V6, V8=V1+3000 Up to 30000V, V3, V7=V4+10000 to 30000V, V5, V9=reference potential. Further, in the optical system example of the present embodiment, the setting method is: RTD=-5000V, V1=-5005V, V2, V4, V6, V8=GND, V3=+20000V, and V7=+17000V. Through the above voltage application, high throughput can be achieved with high resolution at low LE. However, this is only an example, and the voltage applied to each component is not limited to this. Further, the present embodiment can also be applied to the first to fourth embodiments and the embodiment in which the number is not assigned.

(Embodiment 43)

- Prevent discharge separator shape -

The extraction electrode of each lens such as the lens 724 has a limited distance between electrodes. Therefore, when the wall surface of the insulator partition member sandwiched between the electrodes is planar (the cross section is linear), the pressure resistance of the creeping surface may be insufficient. At this time, it is effective to adopt the configuration shown in Fig. 209. In the example shown in Fig. 209, the partition member 7245 provided between the electrodes 7241 and 7242 is formed by connecting three string-shaped partition members, and the surface thereof is formed in a wave shape. Moreover, the spacer 7245 has a surface resistance of 10⁸To 10¹²It is composed of an insulator such as Ω·cm ceramic, and the charging is reduced due to the leakage of a small amount of leakage current. Further, the string-shaped spacers are not limited to three, and may be more or less, and may be used in the case of 4 to 12 or so.

In the case where the pressure resistance of the surface is insufficient, when the value is 1 kV/mm or more (for example, when the potential difference between the electrodes 7241 and 7242 is 20 kV or less, D = 20 mm or less), the shape is separated by the shape of Fig. 209 (a). Item 7245. The side of the partition member is formed into a concave-convex shape, and the electric field at the creeping distance is 1 kV/mm. under. At this time, the connection surface with the electrodes 7241, 7242, the spacer connector is the spacer member 7245 shown in Fig. 209 (a), and the spacer connector is the spacer member 7246 shown in Fig. 209 (b). .

At this time, the separation resistance 7245 which is connected by the concave portion and the electrode, and the partitioning member 7426 which is connected by the convex portion and the connection portion are greatly different in discharge resistance. The superior is the partition 7245 shown in Fig. 209 (a). For example, when discharge occurs at point a, if L/d is large, the discharge of the concave portion converges, and the probability of occurrence of discharge on the outer side decreases. This is because the electric field fluctuation in the concave portion is small, that is, the electrons are not easily scattered to the outside because they belong to the same potential space. Therefore, it is stable in a state where discharge between electrodes is hard to occur. On the other hand, in the spacer 7246 shown in Fig. 209 (b), since the projections are connected to the electrodes 7241 and 7242, discharge occurs at the point b, and discharge is easily generated outside the spacer 7246. This is because the outer space of the partition member 7246 is immediately adjacent to the side thereof, and the probability of electrons scattering into the outer space is increased as compared with the partition member 7245 shown in Fig. 209 (a). Further, even if discharge occurs in the vicinity from point b, the generated electrons are scattered around, and secondary factors such as discharge are generated. In general, a portion where the potential fluctuates greatly, that is, a point and b point at this time, the possibility of discharge is high. In the case of the partition member 7245 shown in Fig. 209 (a), in particular, a larger L/d is preferable. For example, when L/d 4, the discharge resistance is improved, and L/d 4 to 10 is very suitable for the possibility of manufacturing.

The lens using the spacer 7245 is used, for example, as shown in Fig. 209 (c). Further, the present embodiment can also be applied to the above-described first to fourth embodiments and embodiments in which no number is given.

(Embodiment 44)

- Prevent pollution -

As shown in the above-described Embodiments 13 and 27, the composition for preventing the particles may be obtained from other configurations. For example, as shown in Fig. 210, a voltage shielding barrier VB is provided around the lens 724 with an interval G required to prevent discharge. Fig. 210(a) is a cross-sectional view showing a plane passing through the central axis of the lens 724 in a schematic manner, and Fig. 210(b) is a view showing the configuration of the lens 724 from the sample W. In the configuration of the twenty-first embodiment, when a high voltage is applied to the lens 724, the surface of the sample W (contaminated area CA) corresponding to the portion G of the gap may be contaminated by foreign matter.

As a method of preventing this contamination, two configurations (FIG. 211, FIG. 212) have been described in the present embodiment. FIGS. 211 (a) and 212 (a) are cross-sectional views showing the plane passing through the central axis of the lens 724 in a schematic manner, and FIGS. 211 (b) and 212 (b) are views showing the sample W. Each constituent view of the direction of the lens 724.

As shown in FIG. 211, in the first configuration example, the gap G is closed by a disk-shaped insulator shield IS1 having a central opening. Insulator IS1 is made of ceramic, SiO₂The configuration is such that the gap G is sealed by being attached to the voltage shielding barrier VB. In this example, the insulator IS1 is attached to the sample W side of the voltage shielding barrier VB. Further, it is also possible to completely block the gap without using the insulator IS1, and to provide a gap between the spacer and the insulator IS1, thereby narrowing the gap G. In this manner, by providing the insulator shield IS1 such that the gap G is sealed or narrowed, foreign matter accumulation in the contaminated area CA of the sample W can be eliminated or reduced.

As shown in FIG. 212, the second configuration example is a configuration in which a cylindrical insulator shield IS2 that forms a side surface of the surrounding lens 724 is provided. Insulator shield IS2 is made of ceramic, SiO₂And the like, and is fixed to the lens 724. The thermal expansion rates of the insulator shield IS2 and the lens 724 are different. By utilizing this difference in thermal expansion rate, the insulator shield IS2 and the lens 724 can be fixed by first cooling the lens and inserting the insulator shield IS2, and then returning to the ooling fit operation at a normal temperature. When it is fixed by a screw or the like, the positioning accuracy is lowered due to the existence of the tolerance, but the alignment of the lens central axis is facilitated by the adhesion and fixation as in the present embodiment.

According to this configuration, when the foreign matter reaches the sample W due to the influence of the electric field from the lens 724, it can be electrically suppressed by the presence of the insulator shield IS2 to eliminate or reduce the foreign matter accumulation in the contaminated area CA of the sample W. . Moreover, the surface resistance of both the IS1 and IS2 through the insulator shield is 10⁸To 10¹²Ωcm can reduce the leakage current or electrification of the surface. Further, the present embodiment can also be applied to the above-described first to thirteenth embodiments and the embodiments in which the numbers are not assigned.

(Embodiment 45)

- Prevent discharge -

When the positional control of the aforementioned platform device 50 is performed using the laser interference ranging device, the configuration shown in Fig. 213 is utilized. The positional relationship between the respective configurations of the lens 724 and the stage device 50 is shown in Fig. 213 (a) when viewed from the side direction, and is shown in Fig. 213 (b) when viewed from the upper direction. The platform device 50 is provided with a laser interference meter mirror 510x for performing position control in the x-axis direction, and for performing y The laser interference detector for the positional control in the axial direction is a mirror 510y. For the laser jammer 510x, the laser is illuminated from the laser jammer 511x. For the laser interference meter mirror 510y, the laser is irradiated from the laser interference meter 511y.

It is preferable that the laser interferometer mirrors 510x and 510y reflect the laser at the same height as the sample W (the surface position of the sample W). This is because, as shown by the Abbe principle, the difference in the height of the sample W is larger, and the error in the measurement position is larger. As shown in Fig. 214, when the laser interferometer 510x is tilted by the angle Φ, the error at the position of the laser b becomes the difference between the height Hb of the laser b and the height Hw of the sample W (Hb-Hw) × Tan Φ. If Φ is very small, tan Φ Φ. Therefore, in terms of the error, the larger the difference (Hb-Hw) between the height Hb of the laser b and the height Hw of the sample W, the larger the error. Therefore, the difference (Hb-Hw) of the height Hb of the laser b and the height Hw of the sample W is 0, that is, the laser a is preferably irradiated to the laser interference detector 510x at the height of the sample W.

Therefore, the laser interference meter mirrors 510x, 510y must be higher than the height Hw of the sample W. The reason for this is that, in order to improve the assembly accuracy, when the ceramics are used for the laser interference mirrors 510x and 510y, the ceramic surface is mirror-finished to form a mirror. At this time, the range of several mm (for example, about 3 mm) from the upper end portion is required as an outer region of the specification due to the processing requirements. As shown in FIG. 215, the height Hr of the laser interference mirrors 510x, 510y is longer than the height Hw (the height of the sample W (the position of the surface of the sample W)) irradiated by the laser, for example, as shown in FIG. 3mm. Moreover, the distance between the surface of the sample W and the lowermost electrode 72D of the lens barrel is optically designed. The determined distance is 4 mm in this embodiment. Therefore, in the present embodiment, the distance between the barrel lowermost electrode 72D and the laser interference meter mirrors 510x and 510y is 1 mm. Therefore, depending on the position of the stage device 50, the upper end portions of the laser interference meter mirrors 510x and 510y may be discharged too close to the lens 724 to which the high voltage is applied. Therefore, in order not to generate such a discharge, the platform unit 50 is necessary for positioning.

As shown in the schematic view from the side in Fig. 216, the lowermost electrode 72D of the barrel and the lens 724 to which a high voltage (20 kV in this example) is applied are fixed via an insulator IS and grounded (GND). In the present embodiment, the design value of the insulation withstand voltage is set to 4 kV/mm. Therefore, as shown in FIG. 216, the lens 724 to which the voltage of 20 kV is applied is 5 mm from the lower end portion, and it is necessary to prevent the stage device 50, the sample W, the laser interference meter mirrors 510x, 510y, etc. (especially The laser jammer enters with a mirror.

Therefore, as shown in the schematic view from above in Fig. 217, from the lower end portion of the lens 724 to the laser interferometer mirrors 510x and 510y, the horizontal interval Gd is not preset to be separated by 4.58 mm or more. There will be a situation in which an electric field of 4 kV/mm or more is generated. Therefore, the movable range of the stage device 50 is limited by the range in which the interval Gd is 4.58 mm or more. The platform unit 50 can control its position within the movable range through the control system. The example shown in Fig. 217 is a state in which the platform device 50 is closest to the laser interference meters 511x and 511y in the range where the interval Gd is 4.58 mm or more. That is, when the sample W is transferred from the loading chamber 40 to the stage device 50 located in the main casing 30, or when the sample W is transferred from the platform device 50 located in the main casing 30 to the loading chamber 40, Position control to move the platform unit 50 to the laser The position where the upper part of the interference meter is not discharged by the lens (the position of the aforementioned movable range). Since it is necessary to perform the transfer of the sample W at the position of the stage device 50 shown in FIG. 217 (and the wall side of the main casing 30 opposite to the wall side on which the laser interference meter 511x or 511y is disposed), as the pair of loading chambers The closing device 45 of the entrance and exit of 40 is necessary to be provided at any of the two locations shown in Fig. 217 (and the wall side of the main casing 30 opposite to the wall side on which the laser interference meter 511x or 511y is disposed). . Further, the present embodiment can also be applied to the first to fourth embodiments and the embodiment in which the number is not assigned.

(Embodiment 46)

The type of the light source 10000 that illuminates the light on the optoelectronic surface 2021 has been described above, but may be another light source. For example, FUV lamps, excimer lamps, xenon lamps, xenon lamps, and the like. Further, an LD excitation light source lamp in which an LD (Laser Diode, Laser Diode) is condensed to form a spot plasma and whose excitation light is used as a light source can also be used. The excitation light may be introduced into the optoelectronic surface 2021 using at least one of a lens and a mirror. Further, the excitation light may be introduced into the optical fiber by at least one of a lens and a mirror, and then introduced into the optoelectronic surface 2021 from the optical fiber. Furthermore, the magnetic field can also be used for plasma control. Further, the present embodiment can also be applied to the first to fourth embodiments and the embodiment in which the number is not assigned.

(Embodiment 47)

In the description of Fig. 161, an example in which EB-TDI is used in the detection system 70 will be described, and another configuration example in the case of using TDI will be described. In the present embodiment, the description of the same portions as those in the above-described first embodiment is omitted.

First, the detection system 70 shown in Fig. 161 is configured such that when the EB-TDI 72 is used, the EB-CCD 71 is moved to a position deviated from the optical axis by the moving mechanism M. In the first example, as shown in FIG. 218, the moving mechanism M can be coupled to the rotating shaft S. In Fig. 218 (a), one end of the rotating shaft S is coupled to one end of a flat EB-CCD 71 in which a necessary circuit or substrate is built, and the other end of the rotating shaft S is coupled to the moving mechanism M. Fig. 218 (b) and (c) are views showing the configuration shown in Fig. 218 (a) from the viewpoint of the moving mechanism M. When the EB-CCD 73 is used, as shown in Fig. 218 (b), in order to cause the electron beam e to enter the EB-CCD 73, the sensor surface of the EB-CCD 73 moves to be perpendicular to the electron beam e. On the other hand, when the EB-TDI 72 is used, as shown in Fig. 218 (c), the rotating shaft 21 is rotated by the moving mechanism M, and the EB-CCD 73 is moved parallel to the optical axis of the electro-optical system. Therefore, the electron beam e is not incident on the EB-CCD 73 but is incident on the EB-TDI 72.

The moving mechanism M by rotation shown in Fig. 218 has an advantage that the specification and weight can be reduced to, for example, 1/2 to 1/10, compared with the moving mechanism which is moved in the one-axis direction described in Fig. 161.

In the second example, the detection system 70 is not the EB-TDI 72, but as shown in FIG. 219, the TDI sensor 721, the FOP (fiber plate) 722, and the fluorescent plate 723 formed in one package may be used. And the composition of the MCP (microchannel plate) 724. The output lead of the TDI sensor 721 is connected to the lead 73 of the feedthrough FT by other connecting means such as wire bonding. In this case, as described above, the MCP 724 performs a multiplication of the detected electron amount, and the fluorescent panel 723 converts the electron into an optical signal. The two-dimensional optical signal is transmitted through the FOP 722, formed into an image by the TDI sensor 721, and detects the signal. In the second embodiment, the moving mechanism M is described in parallel with the case a in which the EB-CCD 71 is rotated and the case b in which the position is shifted from the optical beam axis. However, any one of them may be used. (Fig. 219) (b) and (c) are views showing a configuration shown in Fig. 219 (a) when the moving mechanism M for rotating the EB-CCD 71 is viewed from the viewpoint of the moving mechanism M. In addition, when electronic amplification is not required, as shown in FIG. 220, the detection system 70 may not use the MCP 724.

Further, the detection system 70 may be configured to be switched between the configuration using the EB-TDI 72 shown in Fig. 221 (a) and the configuration using the EB-CCD 71 shown in Fig. 221 (b). At this time, the detection system 70 may be configured by using the configuration shown in FIG. 222(a) or the configuration shown in FIG. 222(b) instead of the configuration using EB-TDI72 shown in FIG. 2(a).

Next, the operation of the EB-TDI 72 will be described. Figure 223 shows the pixel P in the sensor surface 72S of the EB-TDI 72₁₁To P_IjFloor plan. In the figure, the arrow T1 indicates the cumulative direction of the sensor face 72S, and T2 indicates the direction perpendicular to the cumulative direction T1, that is, the continuous moving direction of the platform device 50. In this embodiment, the pixel of the EB-TDI 72 pixel P₁₁To P_IjIn the cumulative direction T1, 500 segments (the cumulative number of segments i=500) are arranged, and 4000 (j=4000) are arranged in the continuous moving direction T2 of the platform device 50.

Figure 224 is a diagram schematically showing the positional relationship between EB-TDI 72 and secondary charged particles. In Fig. 224, when the secondary charged particles EB discharged from the sample W are released from the same portion of the sample W at a certain time, the secondary charged particles EB are accompanied by the continuous movement of the stage device 50 to the projection projection type optical A series of parts a, b, c, d, e, and i on the system MO are sequentially injected from a to i. Into the image projection type optical system MO incident The secondary charged particles EB are sequentially discharged from a series of parts A, B, C, D, E, , and I on the image projection type optical system MO. At this time, if the cumulative charge of the incident charge (Charge) in the cumulative direction T1 of the EB-TDI 72 is synchronized with the continuous movement of the platform device 50, the parts A, B, C, D, E of the image projection type optical system MO are The second charged particle EB emitted by I will be incident on the same part of the sensor surface 72S in sequence, and the incident charge will be accumulated according to the predetermined cumulative number of segments i. In this way, each pixel P of the sensor face 72S₁₁To P_IjYou can get more radioactive signals. Thereby, a high S/N ratio can be achieved, and a two-dimensional electronic image can be obtained at high speed. The image projection type optical system MO has a magnification of, for example, 300 times.

The EB-CCD and EB-TDI described above have the following features.

(A) The gain is determined unambiguously based on the incident energy of the electron

(B) When the incident energy of the electron rises, the sensor gain also rises.

(C) An effective sensor thickness (easily accumulating the thickness of electrons) is formed with respect to the incident energy band of the electron. When the thickness is too small, the amount of electrons accumulated is small, and when it is too thick, electrons are hard to accumulate.

(D) A sensor that is directly incident on the electron.

(E) It is possible to use not only a surface illumination type sensor but also a back side (inside) illumination type sensor.

(F) A voltage (GND or fixed voltage) can be applied to the surface of the sensor.

(G) There may be a noise blocking cover around the sensor.

(H) The voltage of at least one of the sensor and the camera can be set to a floating type (which is configured to control the reference potential from the outside).

(I) Sensor gain = maximum accumulated charge amount / maximum acquired electron number

Further, the present embodiment can also be applied to the first to fourth embodiments and the embodiment in which the number is not assigned.