Pattern characteristic-detection apparatus for photomask and pattern characteristic-detection method for ph0tomask

0001The present invention relates to a photomask characteristic detection device and a photomask characteristic detection method.

0002In the fields of semiconductor devices, flat panel displays, MEMS (Micro Electro Mechanical Systems), circuit boards, optical components, mechanical components, and the like, structures having a pattern on the surface are manufactured. Then, in the manufacture of such a structure, defects such as an abnormality in the pattern shape, an abnormality in the pattern size, and the presence of foreign matter are inspected. As a method for inspecting a defect of a pattern formed on the surface of a structure, an inspection method called a die-to-database method is known. In this inspection method, inspection data obtained by forming a magnified optical image of a pattern on a light receiving surface such as a CCD sensor (Charge Coupled Device Image Sensor) and design data (CAD data) used when designing the pattern. ) Etc. are compared with the reference data created from, etc., and the discrepancy between the two is detected as a defect.

0003Here, a pattern inspection device is known that detects a missize defect or a misalignment defect of a contact hole pattern, which is a fine pattern, based on the sum of the brightness values of the transmitted light transmitted through the inspection target region (for example,). See Patent Document 1). Further, in a photomask inspection device used in the manufacture of a semiconductor device, a defect may be inspected and a line width of a fine line and space pattern may be detected. Further, in order to evaluate the transferability of the photomask on the wafer surface transferred by the exposure apparatus, a magnified optical image of the photomask pattern is imaged with a CCD sensor or the like using an optical system that imitates the optical system of the exposure apparatus. Devices for detecting the transmittance and line width after transfer are also known.

0004However, in a device capable of evaluating transferability, the technique for detecting the transmittance of these contact hole patterns and the line width of line-and-space patterns has a limited detection target area, and detection is performed at predetermined intervals. Although it could be done, it was not possible to detect the entire area of the photomask with high resolution. On the other hand, in the photomask inspection device, there is a device having a function of obtaining a line width distribution, but it is not possible to obtain a line width distribution in consideration of transferability. Therefore, it was not possible to detect the characteristics and characteristic distribution of the photomask (for example, the distribution of transmittance and the distribution of line width in consideration of transferability) with high resolution over the entire mask. Here, in recent miniaturized patterns, it is possible to improve the detection sensitivity of the contact hole pattern and accurately evaluate the quality of the photomask and the factors that reduce the process margin due to the photomask. It is becoming more and more demanded. However, the conventional technology for detecting transmittance and line width cannot detect the characteristics and characteristic distribution of the photomask. For example, there is an abnormality that does not lead to a defect, the process margin is lowered, and the yield is low. There was a risk that the cause could not be identified if it decreased.

<p num="0005"><patcit num="1"><text>Japanese Unexamined Patent Publication No. 7-128248</text></patcit></p>

<p num="0006"> The present invention provides a photomask characteristic detection device capable of detecting the characteristics of a photomask and a photomask characteristic detection method.</p>

<p num="0007"> According to one aspect of the present invention, the detection data creation unit that creates detection data based on the optical image of the pattern formed on the object to be detected and the reference data related to the pattern are used to develop the design data of the pattern. The reference data creation unit to be created, the reference pattern corresponding to the pattern to be detected of the characteristic, the position information of the reference pattern, the extraction unit for extracting from the reference data, and the detection data based on the reference pattern. In, a first area setting unit that sets a region for detecting the characteristic and extracts a pattern to be detected of the characteristic from the detection data based on the position information, and a first region setting unit.<u style="single">Before</u>The extraction unit includes a detection unit that detects the characteristics of the pattern to be detected in the region for detecting the characteristics, and an aggregation unit that aggregates the detected characteristics. The extraction unit includes pixels that form a template. Provided is a photomask characteristic detection device characterized in that a variable template capable of arbitrarily setting the logic in is provided.</p><p num="0008"> Further, according to another aspect of the present invention, the detection data creation unit that creates detection data based on the optical image of the pattern formed on the object to be detected and the reference data related to the pattern are used.<u style="single">By expanding the design data of the pattern</u>The detection data is performed by performing an inverse conversion process with a reference data creation unit to be created, a reference pattern corresponding to the pattern to be detected of the characteristic, and an extraction unit that extracts the position information of the reference pattern from the reference data. Based on the inverse conversion unit that obtains the pattern formed on the photomask, the transfer pattern calculation unit that obtains the transfer pattern from the pattern formed on the photomask, and the reference pattern.<u style="single">In the transfer pattern</u>A second area setting unit that sets a region for detecting the characteristic and extracts a pattern to be detected of the characteristic from the transfer pattern based on the position information, and a detection of the characteristic in the region for detecting the characteristic. It is provided with a detection unit that detects the characteristics of the target pattern and an aggregation unit that aggregates the detected characteristics.<u style="single">, The extraction unit is provided with a variable template in which the logic of the pixels constituting the template can be arbitrarily set.</u>A photomask characteristic detection device is provided.</p><p num="0009"> Further, according to another aspect of the present invention, detection data is created based on an optical image of a pattern formed on the object to be detected, and reference data relating to the pattern is used.<u style="single">By expanding the design data of the pattern</u>The reference pattern corresponding to the pattern to be detected of the characteristic and the position information of the reference pattern are extracted from the reference data.<u style="single">The extraction of the reference pattern is performed using a variable template in which the logic in the pixels constituting the template can be arbitrarily set, and the logic is set based on the detection recipe.</u>Based on the reference pattern<u style="single">In the detection data</u>A region for detecting the characteristic is set, a pattern to be detected of the characteristic is extracted from the detection data based on the position information, and the characteristic of the pattern to be detected of the characteristic in the region for detecting the characteristic is determined. Provided is a method for detecting characteristics of a photomask, which comprises detecting and aggregating the detected characteristics.</p><p num="0010"> Further, according to another aspect of the present invention, detection data is created based on an optical image of a pattern formed on the object to be detected, and reference data relating to the pattern is used.<u style="single">By expanding the design data of the pattern</u>The reference pattern corresponding to the pattern to be detected of the characteristic and the position information of the reference pattern are extracted from the reference data.<u style="single">The extraction of the reference pattern is performed using a variable template in which the logic in the pixels constituting the template can be arbitrarily set, and the logic is set based on the detection recipe.</u>By performing the inverse conversion process, a pattern formed on the photomask is obtained from the detection data, a transfer pattern is obtained from the pattern formed on the photomask, and the transfer pattern is obtained based on the reference pattern.<u style="single">In the transfer pattern</u>A region for detecting the characteristic is set, a pattern to be detected for the characteristic is extracted from the transfer pattern based on the position information, and the characteristic of the pattern to be detected for the characteristic in the region for detecting the characteristic is determined. Provided is a method for detecting characteristics of a photomask, which comprises detecting and aggregating the detected characteristics.</p>

<p num="0011"> According to the present invention, there is provided a photomask characteristic detection device capable of detecting the characteristics of a photomask and a photomask characteristic detection method.</p>

0012<figref num="1">It is a block diagram for exemplifying the characteristic detection apparatus of the photomask which concerns on this embodiment.</figref><figref num="2">It is a schematic diagram for exemplifying the extraction part.</figref><figref num="3">It is a block diagram for exemplifying a variable template.</figref><figref num="4">It is a block diagram for exemplifying the detection and aggregation of a characteristic.</figref><figref num="5">It is a block diagram for exemplifying the case where the variable template is provided.</figref><figref num="6">It is a block diagram for exemplifying the inverse conversion process and the calculation of a transfer pattern.</figref><figref num="7">It is a block diagram for exemplifying a conversion part.</figref>

0013Hereinafter, embodiments of the present invention will be illustrated with reference to the drawings. In each drawing, similar components are designated by the same reference numerals and detailed description thereof will be omitted as appropriate. FIG. 1 is a block diagram for exemplifying a photomask characteristic detection device according to the present embodiment. As shown in FIG. 1, the characteristic detection device 1 is provided with a detection data creation unit 2, a reference data creation unit 3, a characteristic detection unit 4, an extraction unit 5, and a display unit 6.

0014The detection data creation unit 2 is provided with a light source 21, an illumination optical system 22, a mounting unit 23, an imaging optical system 24, a detection unit 25, and a conversion unit 26. The detection data creation unit 2 creates detection data based on the optical image of the pattern formed on the object to be detected 100. The light source 21 emits the detection light 21a. As the light source 21, various light sources that emit white light, monochromatic light, coherent light, or the like can be used. In this case, in order to detect a fine pattern, it is preferable that the detection light 21a having a short wavelength can be emitted. As such, for example, a YAG laser light source that emits detection light 21a having a wavelength of 266 nm can be exemplified. However, the present invention is not limited to the laser light source, and can be appropriately changed according to the size of the pattern and the like.

0015The illumination optical system 22 guides the detection light 21a emitted from the light source 21 to the detection region of the object to be detected 100 and controls the size of the irradiated portion. The imaging optical system 24 guides the detection light 21a from the object to be detected 100 to the light receiving surface of the detection unit 25 and forms an image on the light receiving surface.

0016The illumination optical system 22 and the imaging optical system 24 can be composed of, for example, optical elements such as a lens as illustrated in FIG. However, the optical elements and their arrangements are not limited to those illustrated, and can be changed as appropriate. Further, for example, other optical elements such as a mirror, an aperture, a beam splitter, a magnification changer, and a zoom mechanism can be appropriately provided. Further, the imaging optical system 24 illustrated in FIG. 1 guides the detection light 21a transmitted through the object to be detected 100 to the detection unit 25, and detects the detection light 21a reflected by the object to be detected 100. It can also lead to 25.

0017The mounting portion 23 is for mounting and holding the detected body 100. Further, the mounting portion 23 is provided with a moving means (not shown), and the position where the detection is performed can be changed by moving the position of the detected object 100 mounted on the mounting portion 23. ing. It should be noted that it is not always necessary to provide a moving means (not shown) in the mounting portion 23, and it is sufficient that the position where the detection is performed changes relatively. For example, the positions of the illumination optical system 22, the imaging optical system 24, the detection unit 25, and the like may be changed by a moving means (not shown).

0018The detection unit 25 photoelectrically converts the optical image formed on the light receiving surface. Examples of the detection unit 25 include a CCD (Charge Coupled Device) sensor, a CCD line sensor, and a TDI (Time Delay and Integration) sensor. However, the present invention is not limited to these, and those capable of photoelectrically converting the light of the formed optical image can be appropriately selected. The conversion unit 26 A / D-converts the electric signal output from the detection unit 25. In addition, detection data is created by graphically interpreting the A / D converted electrical signal.

0019The reference data creation unit 3 is provided with a data storage unit 31, a data expansion unit 32, and a data creation unit 33. The reference data creation unit 3 creates reference data based on the design data and the like stored in the data storage unit 31. That is, the reference data creation unit 3 creates reference data regarding the pattern formed on the detected object. The data storage unit 31 stores drawing data used for pattern formation, design data before conversion to drawing data, and the like.

0020The data expansion unit 32 expands the design data and the like provided by the data storage unit 31 into two-dimensional data. The data creation unit 33 creates reference data by graphically interpreting the data expanded into the two-dimensional data. At this time, reference data is created according to the resolution of the detected data. That is, the data expanded by the data expansion unit 32 is converted into reference data having the same resolution as the optical image data (detection data) acquired by the detection unit 25. The extraction unit 5 extracts a pattern having the same shape and size as the pattern to be detected of the characteristic (for example, transmittance) from the reference data. Then, the signal (valid flag) related to the extracted pattern (reference pattern) is output to the area setting unit 43. At this time, the position information of the extracted pattern (reference pattern) (position information of where the extracted pattern (reference pattern) is located in the photomask) is also output. That is, the extraction unit 5 extracts the reference pattern corresponding to the pattern for which the characteristic is to be detected and the position information of the reference pattern from the reference data. Then, the signal (valid flag) related to the extracted reference pattern and the position information of the reference pattern are output to the area setting unit 43. The details of the extraction unit 5 will be described later.

0021The characteristic detection unit 4 is provided with a first characteristic detection unit 4a and a second characteristic detection unit 4b. In this case, a plurality of characteristic detection units provided with the first characteristic detection unit 4a and the second characteristic detection unit 4b may be provided. The first characteristic detection unit 4a detects the characteristics (for example, transmittance) of the pattern of the detection data corresponding to the reference pattern extracted by the extraction unit 5, and aggregates the detected characteristics.

0022The first characteristic detection unit 4a is provided with an area setting unit 43, a detection unit 44a, and a totaling unit 45a. The area setting unit 43 sets an area for detecting the characteristics of the pattern (for example, transmittance) based on the signal (valid flag) from the extraction unit 5. Further, based on the position information from the extraction unit 5, the pattern to be detected of the characteristic is extracted from the detection data. That is, the area setting unit 43 sets the area for detecting the characteristic based on the reference pattern, and extracts the pattern to be detected of the characteristic from the detection data based on the above-mentioned position information.

0023The detection unit 44a detects the characteristic of the pattern based on the region set by the region setting unit 43 and the pattern for which the characteristic is to be detected. That is, the detection unit 44a detects the characteristic of the pattern to be detected of the characteristic in the region for detecting the characteristic. The aggregation unit 45a aggregates the characteristics detected by the detection unit 44a. In this case, the characteristics in the entire photomask can be aggregated based on the detected characteristics. It is also possible to create information about the characteristic distribution (for example, transmittance distribution) based on the detected characteristics and the position information. The second characteristic detection unit 4b obtains the pattern formed on the photomask from the detection data by performing the inverse conversion process. Next, based on the pattern formed on the photomask, a pattern (transfer pattern) to be transferred to the surface of the object to be detected 100 (for example, a wafer) is obtained by simulation. Then, the characteristics of the transfer pattern (for example, transmittance) corresponding to a specific position of the reference pattern are detected, and the detected characteristics are aggregated. Details regarding the inverse conversion process and the calculation of the transfer pattern will be described later.

0024The second characteristic detection unit 4b is provided with an inverse conversion unit 41, a transfer pattern calculation unit 42, an area setting unit 43, a detection unit 44b, and a totaling unit 45b. The inverse conversion unit 41 obtains the pattern formed on the photomask from the detection data by performing the inverse conversion process. In this case, the inverse transformation processing includes the optical imaging characteristics of the imaging optical system 24, that is, the point spread function (PSF), the image forming function indicating the sensitivity distribution of each pixel of the detection unit 25, and the image forming function. Is done using.

0025The transfer pattern calculation unit 42 obtains a transfer pattern from the pattern formed on the photomask obtained by the inverse conversion unit 41. The area setting unit 43 sets an area for detecting the characteristics of the pattern (for example, transmittance) based on the signal (valid flag) from the extraction unit 5. Further, based on the position information from the extraction unit 5, the pattern to be detected of the characteristic is extracted from the transfer pattern obtained by the transfer pattern calculation unit 42. That is, the region setting unit 43 sets the region for detecting the characteristic based on the reference pattern, and extracts the pattern to be detected for the characteristic from the transfer pattern based on the above-mentioned position information. The detection unit 44b detects the characteristic of the pattern based on the region set by the region setting unit 43 and the pattern to be detected of the characteristic. That is, the detection unit 44b detects the characteristic of the pattern to be detected of the characteristic in the region for detecting the characteristic. The aggregation unit 45b aggregates the characteristics detected by the detection unit 44b. In this case, the characteristics in the entire photomask can be aggregated based on the detected characteristics. It is also possible to create information about the characteristic distribution (for example, transmittance distribution) based on the detected characteristics and the position information. Details regarding the setting of the area for detecting the characteristics, the detection of the characteristics, the aggregation of the characteristics, and the like will be described later.

0026The display unit 6 visualizes the characteristic data aggregated by the aggregation units 45a and 45b. For example, a characteristic distribution map (for example, a transmittance distribution map) can be displayed. As described above, the aggregation unit 45a and 45b can create the information about the characteristic distribution, but the display unit 6 can also create the information. It is also possible to switch the type of pattern to be displayed (for example, the shape and size of the pattern), and to switch between the characteristics from the first characteristic detection unit 4a and the characteristics from the second characteristic detection unit 4b. it can. In addition, the display range, distribution intensity, and the like can be arbitrarily set. The display unit 6 is not always necessary and may be provided as appropriate. For example, a storage unit (not shown) for storing data such as characteristics aggregated by the aggregation units 45a and 45b may be provided instead of the display unit 6.

0027Next, the extraction unit 5 will be further illustrated. FIG. 2 is a schematic diagram for exemplifying the extraction unit. Note that FIG. 2 (a) is a block diagram for exemplifying the extraction unit, FIGS. 2 (b) to 2 (d) are schematic diagrams for exemplifying the state of template matching, and FIG. 2 (e) is for detecting pattern dimensions. A schematic diagram for exemplifying the position, FIG. 2 (f) is a schematic diagram for exemplifying the detection of the pattern dimension, and FIG. 2 (g) is a schematic diagram for exemplifying the state of matching by the pattern area. As shown in FIG. 2A, the extraction unit 5 is provided with a first collation unit 35, a second collation unit 36, and a logical product calculation unit 39. In the present embodiment, the pattern selected by the first collation unit 35 is further pattern-matched by the second collation unit 36. Therefore, the extraction accuracy of the target pattern can be improved.

0028The first collation unit 35 selects a pattern having the same shape and size as the pattern to be detected of the characteristic from the reference data. Sorting can be performed by template matching as shown in FIGS. 2 (b) to 2 (d). For example, the pattern 101 as shown in FIG. 2 (b) is selected by performing template matching as shown in FIG. 2 (d) by using a template using the threshold values 102 and 103 as shown in FIG. 2 (c). can do. Here, the threshold value 102 is used for matching a region smaller than a predetermined light transmission amount, and the threshold value 103 is used for matching a region larger than a predetermined light transmission amount. The template shown in FIG. 2C is a fixed template in which a region for matching according to the threshold value 102 is provided around the region for matching according to the threshold value 103.

0029The second collation unit 36 is provided with a pattern dimension calculation unit 37 and a pattern area calculation unit 38. The pattern dimension calculation unit 37 performs pattern matching by calculating the pattern dimensions. For example, pattern matching is performed by calculating pattern dimensions in directions orthogonal to each other as shown in FIG. 2 (e). In this case, as shown in FIG. 2 (f), the pattern dimension can be calculated from the light transmission amount profile 104 and the predetermined threshold value 105. For example, the size 105 of the profile 104 at a predetermined threshold 105 can be used as the pattern size.

0030The pattern area calculation unit 38 performs pattern matching by calculating the pattern area. For example, as shown in FIG. 2 (g), pattern matching is performed by calculating the area of a portion of a predetermined area 106 that is equal to or larger than a predetermined threshold value. In this case, the area 106 is set to be the same range as the area where matching is performed by the threshold value 102 illustrated in FIG. 2 (c). The logical product calculation unit 39 performs a logical product calculation on the matching result output from the pattern dimension calculation unit 37 and the pattern area calculation unit 38. In this way, a reference pattern having the same shape and size as the pattern for which the characteristic is to be detected is extracted from the reference data.

0031Further, although the template illustrated in FIG. 2C is a fixed template, it can also be a variable template in which the matching area can be arbitrarily set. The extraction unit 5 may be provided with a variable template in which the logic of the pixels constituting the template can be arbitrarily set. FIG. 3 is a block diagram for exemplifying the variable template. Note that FIG. 3 is a block diagram for exemplifying the variable template 50 provided in the first collation unit 35. As shown in FIG. 3, the variable template 50 is provided with a delay unit 51, a buffer unit 52, a binarization unit 53, a matching unit 54, and a logical product calculation unit 57. The delay unit 51 creates a certain time delay in transmission without changing the waveform of the electrical signal of the reference data. The buffer unit 52 stores the reference data input via the delay unit 51 as N × N pixel data.

0032The binarization unit 53 is provided with conversion units 53a and 53b. The conversion units 53a and 53b perform binarization using different threshold values. As the threshold value used, for example, the above-mentioned threshold value 102 and threshold value 103 can be exemplified. The N × N pixel data stored in the buffer unit 52 is binarized by the binarization unit 53 using different threshold values, and the pixel matching unit 54 corresponding to each pixel of the N × N pixels.₀~54_N-1Provided to each. The threshold value can be changed arbitrarily.

0033The matching unit 54 includes a pixel matching unit 54 corresponding to each pixel of N × N pixels.₀~54_N-1Is provided. In addition, each pixel matching unit 54₀~54_N-1Is provided with a logical operation unit 55a and 55b and a logical product operation unit 56, respectively. The logic in the logical operation units 55a and 55b can be set arbitrarily. That is, by arbitrarily setting the logic in the logical operation units 55a and 55b, a variable template in which the logic in the pixels constituting the template can be arbitrarily set is configured. For example, in the logical operation unit 55a, any one of "value of the pixel&gt; threshold value 103, value of the pixel threshold value 103, calculation unnecessary" is selected and set. Further, in the logical operation unit 55b, any one of "value of the pixel&gt; threshold value 102, value of the pixel threshold value 102, calculation unnecessary" is selected and set. Then, the data provided by the binarization unit 53 is determined by the set logic.

0034In this case, "calculation not required" can be selected and set when it is clear what the value of the pixel is. For example, when it is clear that the pixel is "bright" like a pixel located in the hole part (transparent part) of the hole pattern, "1" in the figure is selected and set to always "ON (bright)". "Can be determined.

0035The logical product calculation unit 56 outputs a matching result in the pixel by performing a logical product operation on the determination results output from the logical operation units 55a and 55b. The logical product calculation unit 57 is a pixel matching unit 54.₀~54_N-1The matching result of the logical product calculation unit 56 in the above is further ANDed, and when matching is performed, the "template matching result" is output. The "template matching result" is output to the area setting unit 43 as an "valid flag".

0036In this way, it is possible to take the logical product of the matching results of all N × N pixels according to the logical setting of the set template, so that the template matching can be performed by an arbitrary logical setting. Further, if template matching by arbitrary logical setting becomes possible, a template matching the shape of the pattern for detecting the characteristic can be easily set from the "detection recipe" or the like.

0037Next, the detection and aggregation of characteristics (for example, transmittance) will be further illustrated. FIG. 4 is a block diagram for exemplifying the detection and aggregation of characteristics. Note that FIG. 4 illustrates a case where the transmittance is detected and aggregated as an example. As shown in FIG. 4, the area setting unit 43 sets an area for detecting the transmittance of the pattern based on the signal (valid flag) from the extraction unit 5. At this time, based on the signal (valid flag) from the extraction unit 5, a template matching the shape and size of the pattern for which the transmittance is to be detected is selected. Further, based on the position information from the extraction unit 5, the pattern to be detected for the transmittance is extracted from the detection data.

0038The detection unit 44a detects the transmittance of the pattern based on the area set by the area setting unit 43 and the pattern to be detected for the transmittance. For example, the transmittance of the pattern is detected by obtaining the sum of the data of the portion equal to or larger than the predetermined threshold value in the set region. The aggregation unit 45a aggregates the transmittance detected by the detection unit 44a. In this case, the transmittance in the entire photomask can be aggregated, and information on the transmittance distribution can be created based on the detected transmittance and the position information. Although the case of the first characteristic detection unit 4a has been described as an example in FIG. 4, the same can be applied to the case of the second characteristic detection unit 4b regarding the detection and aggregation of the transmittance. Further, although the case where the fixed template is provided in the area setting unit 43 is illustrated, a variable template may be provided.

0039FIG. 5 is a block diagram for exemplifying the case where the variable template is provided. As shown in FIG. 5, the third characteristic detection unit 4c is provided with a buffer unit 60, a variable template 61, a calculation unit 63, a logical product unit 64, and an aggregation unit 65. The buffer unit 60 stores the input detection data as N × N pixel data. The variable template 61 is provided with a selection unit 62. Further, in the selection unit 62, the pixel selection unit 62a corresponds to each pixel of N × N pixels.₀~ 62a_N-1Is provided. In addition, the pixel selection unit 62a₀~ 62a_N-1In, it is possible to set (selecting a pixel) whether or not to select data related to the pixel. That is, the selection unit 62 can set an area for detecting the characteristic by arbitrarily selecting the pixels constituting the detection data. Then, the total sum of the detected data in the set area is calculated by the calculation unit 63.

0040That is, the detection data is the pixel selection unit 62a provided corresponding to each pixel of the N × N pixels via the buffer unit 60 that stores the data of the N × N pixels.₀~ 62a_N-1Is entered in. Pixel selection unit 62a₀~ 62a_N-1Is designed to operate by setting whether or not to select the corresponding pixel, so that the sum total of the values in the arbitrarily set pixels can be calculated by the calculation unit 63. In this way, the pixels to be set for the detected data can be arbitrarily selected and the total can be obtained. The logical product unit 64 takes the logical product of the total value of the detected data calculated by the calculation unit 63 and the valid flag from the extraction unit 5. In this way, the total value (characteristic value) of the pattern to be detected of the characteristic corresponding to the reference pattern can be calculated.

0041In this way, since it is possible to obtain the total value (characteristic value) of all N × N pixels according to the logical setting of the set variable template 61, it is possible to detect the characteristic by an arbitrary logical setting. Further, if the characteristic can be detected by an arbitrary logical setting, a template matching the shape of the pattern for detecting the characteristic can be easily set from the "detection recipe" or the like.

0042Next, further examples will be given regarding the inverse conversion process and the calculation of the transfer pattern. FIG. 6 is a block diagram for exemplifying the inverse conversion process and the calculation of the transfer pattern. As shown in FIG. 6, the inverse transform unit 41 described above is provided with an inverse Fourier transform unit 41a and an inverse transform calculation unit 41b. The inverse Fourier transform unit 41a performs an inverse Fourier transform on the detected data and converts it into the detected data on the Fourier plane. The inverse transformation calculation unit 41b converts the image into a Fourier plane by using the optical imaging characteristics of the imaging optical system 24 and the image forming function of the detection unit 25 considering the point spread function (PSF). The pattern formed on the photomask is obtained from the detected detection data. Even if the point image distribution (optical imaging characteristics) of the optical system and the image formation function of the detection unit 25 are theoretically derived from the optical characteristics and the characteristics of the detection unit 25 (so-called sensor characteristics, etc.). It may be obtained by experiments or the like.

0043Here, a further example will be given with respect to obtaining a pattern formed on a photomask from the detection data. According to the theory of image formation, the output from the detection unit 25 can be expressed by the following equation (1). That is, the output from the detection unit 25 performs a composition (convolution) calculation using the point image distribution (optical imaging characteristics) of the optical system and the image formation function which is the sensitivity distribution of each pixel of the detection unit 25. Then, the sampling is performed using the comb-shaped function comb corresponding to the discrete sampling of the detection unit 25.

0044<maths num="1"><img id="000002" he="8" wi="159" file="JP5198397B2_D0001.tif" img-format="tif" img-content="drawing" /></maths> Here, i (x) is the output from the detection unit 25 (detection data), d (x) is the image formation function of the detection unit 25 (function of the sensitivity distribution of the sensor pixels), and psf (x) is the point of the optical system. Image distribution (optical imaging characteristics), o (x)<u style="single">object</u>The image, comb (x) is a function indicating the pixel position (array) of pitch p, and * is a pollution operation.

0045Then, Eq. (2) is derived by Fourier transforming Eq. (1). <maths num="2"><img id="000003" he="9" wi="159" file="JP5198397B2_D0001.tif" img-format="tif" img-content="drawing" /></maths> Here, I (u) is the Fourier transform of the output (detection data) from the detection unit 25, and D (u) is the Fourier transform of the image formation function (function of the sensitivity distribution of the sensor pixels) of the detection unit 25. , OTF (u) is the Fourier transform of the point image distribution (optical imaging characteristics) of the optical system, and O (u) is<u style="single">object</u>The image is Fourier transformed, comb (u) is the Fourier transform of the function indicating the pixel position (array) of pitch p, and * is the pollution operation.

0046Next, Eq. (3) is derived by inverse Fourier transforming Eq. (2). That is, the pattern formed on the photomask on the Fourier plane can be obtained from the output signal of the detection unit 25.

0047<maths num="3"><img id="000004" he="8" wi="159" file="JP5198397B2_D0001.tif" img-format="tif" img-content="drawing" /></maths> In such a Fourier transform, the band is limited by the pixel pitch p of the detection unit 25. In such a case, the band on the Fourier plane can be expanded by performing zero padding when the output (detection data) from the detection unit 25 is Fourier transformed.

0048Next, returning to FIG. 6, the transfer pattern calculation unit 42 will be illustrated. The transfer pattern calculation unit 42 described above is provided with a positive Fourier transform unit 42a and an imaging calculation unit 42b. The imaging calculation unit 42b obtains a transfer pattern based on the pattern (pattern formed on the photomask) obtained by the inverse transformation calculation unit 41b. That is, the transfer pattern on the Fourier plane is obtained by transfer simulation in consideration of the light source intensity distribution of the exposure apparatus, the pupil function of the optical system, the aberration characteristics, and the like. Since the transfer simulation can be performed using a known optical simulation for lithography, an arithmetic unit, or the like, the description thereof will be omitted. The positive Fourier transform unit 42a obtains a pattern (transfer pattern in real space) transferred to the surface of the object to be detected 100 (for example, a wafer) by Fourier transforming the transfer pattern on the Fourier plane.

0049Next, the operation of the photomask characteristic detection device 1 and the photomask characteristic detection method will be illustrated. First, the object to be detected 100 is placed on the mounting unit 23 by a transport device (not shown), an operator, or the like. Next, the detection light 21a is emitted from the light source 21. The detection light 21a emitted from the light source 21 is guided to the detection region of the object to be detected 100 by the illumination optical system 22, and the size of the irradiated portion is controlled. Then, the position where the detected body 100 mounted on the mounting portion 23 is detected is relatively changed by a moving means (not shown) or the like.

0050The detection light 21a from the object to be detected 100 is guided on the light receiving surface of the detection unit 25 by the imaging optical system 24 and is imaged on the light receiving surface. The light of the optical image formed on the light receiving surface is photoelectrically converted by the detection unit 25. Then, the electrical signal photoelectrically converted by the detection unit 25 is A / D converted by the conversion unit 26, and the graphic interpretation is performed to create the detection data.

0051On the other hand, as the detection of the detected object 100 progresses, the data storage unit 31 provided in the reference data creation unit 3 provides the data expansion unit 32 with design data and the like, and the data is expanded into two-dimensional data. The data expanded into the two-dimensional data is graphically interpreted by the data creation unit 33 to create reference data. Then, a pattern (reference pattern) having the same shape and size as the pattern to be detected of the characteristic is extracted from the reference data by the extraction unit 5. Then, a signal (valid flag) related to the extracted pattern is output to the area setting unit 43. At this time, the position information of the extracted pattern, that is, the position information of where the extracted pattern is in the photomask is also output. The reference pattern can be extracted by using the fixed template or variable template described above.

0052Next, the characteristic detection unit 4 detects the characteristics of the pattern to be detected and aggregates the detected characteristics. At this time, the area setting unit 43 sets an area for detecting the characteristics of the pattern based on the signal (valid flag) from the extraction unit 5. Further, based on the position information from the extraction unit 5, the pattern to be detected of the characteristic is extracted from the detection data.

0053In this case, the characteristics of the pattern to be detected are detected by the first characteristic detection unit 4a based on the detection data, and the detected characteristics are aggregated. On the other hand, the second characteristic detection unit 4b obtains a transfer pattern from the detection data, detects the characteristics of the pattern to be detected based on the transfer pattern, and aggregates the detected characteristics. Then, the aggregated characteristic data is visualized by the display unit 6. At this time, it is also possible to display the characteristic distribution over the entire photomask.

0054That is, in the photomask characteristic detection method according to the present embodiment, detection data is created based on the optical image of the pattern formed on the detected object 100, and reference data regarding the pattern formed on the detected object 100 is created. Is created, the reference pattern corresponding to the pattern to be detected of the characteristic and the position information of the reference pattern are extracted from the reference data, the area for detecting the characteristic based on the reference pattern is set, and the position information is used. Based on this, the pattern to be detected of the characteristic is extracted from the detection data, the characteristic of the pattern to be detected of the characteristic in the region where the characteristic is detected is detected, and the detected characteristic is aggregated.

0055Further, the detection data is created based on the optical image of the pattern formed on the detected object 100, the reference data on the pattern formed on the detected object 100 is created, and the reference corresponding to the pattern to be detected of the characteristic is created. The pattern and the position information of the reference pattern are extracted from the reference data, and the pattern formed on the photomask is obtained from the detected data by performing the inverse conversion processing, and the transfer pattern is obtained from the pattern formed on the photomask. A region for detecting characteristics is set based on the reference pattern, a pattern to be detected for characteristics is extracted from the transfer pattern based on position information, and the characteristics of the pattern to be detected for characteristics in the region for detecting characteristics are determined. Detect and aggregate the detected characteristics.

0056It is also possible to create information on the characteristic distribution based on the detected characteristics and the position information. Further, the reference pattern is extracted by using a variable template in which the logic in the pixels constituting the template can be arbitrarily set, and the logic can be set based on the "detection recipe". ..

0057In the above-exemplified ones, the transmittance is mentioned as a characteristic of the photomask, but the present invention is not limited to this. For example, the line width (light-shielding portion) can be detected in the same manner as the transmittance (translucent portion). In addition, it is possible to detect a design danger point, a transmittance in a specific area near the danger point, and the like. In addition, information on the characteristic distribution can be created for these other characteristics as well. Therefore, it is also possible to know the characteristic distribution over the entire photomask for other characteristics. In addition, these characteristics can be combined as appropriate for detection, aggregation, and the like. For example, the line width of the transferred image can be detected from the transfer pattern together with the transmittance, and the transmittance distribution and the line width distribution over the entire photomask can be aggregated. It can also be combined with photomask inspection and the like. For example, it is possible to inspect defects in a photomask and detect a transmittance distribution such as a hole pattern that attracts attention over the entire photomask. Therefore, since it is possible to know the abnormality that does not lead to a defect, it is possible to accurately evaluate the quality of the photomask and the factors that reduce the process margin due to the photomask.

0058According to this embodiment, it is possible to detect characteristics (for example, transmittance, line width, etc.) and characteristic distribution (for example, transmittance distribution, line width distribution, etc.) over the entire photomask. Therefore, even if there is a slight abnormality that does not lead to a defect and the process margin is lowered or the yield is lowered, the cause can be accurately identified. In addition, it is possible to evaluate the quality of the photomask such as uniformity and fluctuation of transmittance and line width, and the process margin caused by the photomask, so that the yield in the lithography process can be improved. Can be planned.

0059In addition, the template setting, the parameter setting of the extracted pattern (for example, the parameter related to the shape and size of the pattern), the setting of the area for detecting the transmittance and the line width, etc. can be arbitrarily performed by the "detection recipe" or the like. Can be done. That is, the template used for pattern extraction or the like can be arbitrarily set from the "detection recipe" or the like.

0060Next, another embodiment that can be applied when obtaining a pattern formed on a photomask from the detection data will be illustrated. FIG. 7 is a block diagram for exemplifying the conversion unit. As shown in FIG. 7, the conversion unit 141 is provided with a shape change unit 142, a convolution calculation unit 143, a correlation calculation unit 144, and an optimization unit 145. The shape changing portion 142 changes the shape and size of the pattern stepwise. The convolution calculation unit 143 performs a composition (convolution) calculation using the point image distribution (optical imaging characteristic) of the optical system and the image formation function which is the sensitivity distribution of each pixel of the detection unit 25. The correlation calculation unit 144 performs a correlation calculation between the pattern data that has been calculated (convolution) and the detection data. The optimization unit 145 performs optimization based on the result of the correlation calculation.

0061In the inverse transform unit 41 illustrated in FIG. 6, the detection data is inverse-Fourier-transformed and converted into a Fourier plane, and the detection data converted into the Fourier plane using the optical imaging characteristics and the image formation function is converted into a photomask. I try to find the formed pattern. On the other hand, in the present embodiment, the pattern that most correlates with the detected data is obtained by the correlation calculation, and this is used as the pattern formed on the photomask.

0062For example, the shape and size of the pattern in the design data input to the shape changing unit 142 are changed by a predetermined value. As for the changed pattern, the pattern formed in the photomask is obtained by performing the composition (convolution) calculation in the convolution calculation unit 143. Then, the correlation calculation between the obtained pattern and the detected data is performed. If the correlation is low, the optimization unit 145 performs optimization so that the correlation is high. Then, the shape changing unit 142 changes the line width and size of the pattern based on the optimization data. After that, by repeating this series of operations, the pattern in which the most correlation with the detected data is obtained is obtained as the pattern formed in the photomask. According to this embodiment, even when the inverse Fourier transform is difficult, the pattern formed on the photomask can be obtained from the detection data. Therefore, even when the inverse Fourier transform is difficult, the characteristics related to the transfer pattern can be detected. In addition to the contact hole pattern, a specific line-and-space pattern, a location of a danger point in the pattern design where defects are likely to occur in the pattern, or a transmittance distribution in a specific area in the vicinity thereof can be obtained in the same manner. It can be measured and displayed.

0063Next, an example will be given of a method for manufacturing a photomask according to the present embodiment. In the photomask manufacturing method according to the present embodiment, the characteristic distribution (for example, transmittance distribution and line width distribution) over the entire photomask is used by using the photomask characteristic detection device 1 and the photomask characteristic detection method described above. Etc.), and modify the pattern layout (exposure pattern data) in consideration of the result. Then, a photomask is created based on the pattern layout (exposure pattern data) modified in this way. In this case, the photomask can be made by using an etching method. It is also possible to create control information regarding exposure conditions without modifying the pattern layout (exposure pattern data). For example, control information can be created so that the exposure conditions can be changed according to the characteristic distribution.

0064According to the photomask manufacturing method according to the present embodiment, it is possible to obtain a photomask having a uniform characteristic distribution over the entire photomask. In addition, it is possible to know the abnormality before it leads to a defect. Therefore, the quality of the photomask can be improved. In addition, the template used for pattern extraction and the like can be arbitrarily set from the "detection recipe" and the like. Therefore, it is possible to improve the productivity, quality, yield and the like of the photomask. In addition, by creating control information that changes the exposure conditions according to the characteristic distribution, it is possible to improve the yield of the photomask and also improve the quality and yield of the product.

0065Next, an example will be given of a method for manufacturing an electronic device according to the present embodiment. As an example, a method of manufacturing a semiconductor device will be described as an example. The semiconductor device manufacturing method includes a process of forming a pattern on a wafer by film formation, resist coating, exposure, development, etching, resist removal, etc., an inspection process, a cleaning process, a heat treatment process, an impurity introduction process, a diffusion process, and flattening. It is carried out by repeating a plurality of steps such as a step. Then, in such a method for manufacturing a semiconductor device, a photomask having uniform characteristics is manufactured by using the above-mentioned photomask manufacturing method, and exposure is performed using the manufactured photomask. Further, since the characteristic distribution over the entire photomask can be known in advance, it is possible to control the exposure conditions in the exposure process according to the characteristic distribution. At this time, the control can be performed based on the above-mentioned "control information for changing the exposure condition according to the characteristic distribution". Since known techniques of each step can be applied to methods other than the above-mentioned photomask manufacturing method, their description will be omitted.

0066Further, as an example, a method of manufacturing a semiconductor device has been described as an example of a method of manufacturing an electronic device according to the present embodiment, but the present invention is not limited to this. For example, it can be widely applied to the manufacture of electronic devices using photolithography technology, such as pattern formation in the manufacture of flat panel displays (for example, pattern formation in liquid crystal color filters, array substrates, etc.).

0067According to the method for manufacturing an electronic device according to the present embodiment, a circuit pattern or the like can be formed by using a photomask having uniform characteristics. Further, since the characteristics over the entire photomask can be known in advance, the exposure conditions in the exposure process can be controlled according to the characteristics. Therefore, deterioration of electrical characteristics due to deformation of the circuit pattern, bridges and disconnections of the circuit pattern, etc. can be suppressed, so that product yield and quality can be improved.

0068The present embodiment has been illustrated above. However, the present invention is not limited to these descriptions. With respect to the above-described embodiment, those skilled in the art with appropriate design changes are also included in the scope of the present invention as long as they have the features of the present invention. For example, the shape, dimensions, arrangement, number, and the like of each element included in the characteristic detection device 1 are not limited to those illustrated, and can be changed as appropriate. In addition, the elements included in each of the above-described embodiments can be combined as much as possible, and the combination thereof is also included in the scope of the present invention as long as the features of the present invention are included.

00691 characteristic detection device, 2 detection data creation unit, 3 reference data creation unit, 4 characteristic detection unit, 4a first characteristic detection unit, 4b second characteristic detection unit, 4c third characteristic detection unit, 5 extraction unit, 6 Display unit, 21 Light source, 25 Detection unit, 26 Conversion unit, 31 Data storage unit, 32 Data expansion unit, 33 Data creation unit, 35 1st collation unit, 36 2nd collation unit, 37 Pattern dimension calculation unit, 38 Pattern area calculation unit, 39 AND calculation unit, 41 Inverse conversion unit, 41a Inverse Fourier conversion unit, 41b Inverse conversion calculation unit, 42 Transfer pattern calculation unit, 43 Area setting unit, 44a detection unit, 44b detection unit, 45a Aggregation Part, 45b Aggregation part, 50 Variable template, 51 Delay part, 52 Buffer part, 53 Binification part, 54 Matching part, 57 Logical product calculation part, 60 Buffer part, 61 Variable template, 63 Calculation part, 64 Logical product part , 65 Aggregate, 100 Detected, 141 Converter, 142 Shape change, 143 AND, 144 Correlation, 145 Optimize