Control for electron microprobe

[57] ABSTRACT

An electron microprobe has a holder for a sample to be analyzed. The holder is movable along an X-axis and a Y-axis, each at right angles to the electron beam directed at the sample. Drives, each including an electric motor, are operatively connected to the holder to adjust the position of the holder along the respective axis. There are also electron beam position controls for positioning the beam in a path oriented as desired with respect to the X. and Y axes. An electron beam current control enables the quantity of electrons in the beam to be adjusted. A third drive includes an electric motor, and is connected to a reflector to position the reflector with respect to the X-rays emitted from the sample, and with respect to X-ray detectors. The controls and drives are connected to a control matrix through which signals are supplied to the respective controls and drives, as dictated by a computer. Each of the controls and drives produces an analogue feedback signal indicative of the setting thereof. These feedback signals are sent to a selector matrix from which a specific signal is selected, as dictated by the computer, with the selected signal being delivered to the computer through an analogue to digital converter. The X-ray detectors also produce signals which are ultimately delivered to the computer either directly or through the analogue to digital converter. Uy appropriately programming the computer, the apparatus automatically analyzes the sample for constituents as desired.

Claims, 3 Drawing Figures

<img file="US3733484A_D0001.tif" />

PATENTED MAY151373

3,733,484

SHEET 1 OF 3

<img file="US3733484A_D0002.tif" />

l BEAM CURRENT CONTROL

I 58

<img file="US3733484A_D0003.tif" />

<img file="US3733484A_D0004.tif" />

CONTROL

<img file="US3733484A_D0005.tif" />

<img file="US3733484A_D0006.tif" />

selector NIATRl'A

<img file="US3733484A_D0007.tif" />

PATENTED 51373

3.733,484

SHEET 2 OF 3

<img file="US3733484A_D0008.tif" />

PATENTED HAY 151973

3,733,484

SHEET 3 OF 3

<img file="US3733484A_D0009.tif" />

<img file="US3733484A_D0010.tif" />

<img file="US3733484A_D0011.tif" />

3,733,484

CONTROL FOR ELECTRON MICROPROBE axes are identified herein as X and Y axes, but these are

BACKGROUND OF THE INVENTION

Electron microprobes are employed to make composition analyses of various samples, i.e. small areas of a 5 metal surface, occlusions in a polished section, small areas in thin sections, free particles mounted on a substrate, etc. Under the prior art practices, these analyses require constant attention by a qualified operator. Once the raw data is obtained by the operator, the op- 10 erator must spend a substantial amount of time reducing the raw data to useful analytical answers. Following these practices, a skilled operator cannot be expected to make complete quantitative analysis of more than about three to four samples per day. The electron mi- 15 croprobe is an expensive piece of equipment and the cost of the time of a skilled operator is not insubstantial. Thus, not only is the amount of analytical output relatively small, but the cost per analysis is relatively high. 20

Utilizing the present invention, the operator need only work about an hour a day to do 10 to a hundred times as much work as he previously accomplished in 8 hours. The operator can analyze 32 samples per 8 hour day using the spectrometers of the microprobe, 25 and can analyze as many as 400 a day using the nondispersive detector. Furthermore, during an eight hour day during which analyses are being made, perhaps seven-eighths of the operator’s time can be devoted to activities other than overseeing the analyses that are tak- ³⁰ing place. It is possible for the apparatus to be set up in the evening to carry out analyses overnight. In the morning when the operator returns to work, the results of those analyses (the raw data which has been reduced to useful analytical answers) is waiting for him. ³⁵

SUMMARY OF THE INVENTION

The present invention relates to a unique combination of a microprobe and controls therefor which will automatically carry out analyses of samples and produce refined analytical answers.

DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagrammatic representation of an electron microprobe having the automatic control apparatus of the present invention;

FIG. 2 is a schematic drawing of the control and feedback apparatus for one of the sample holder drives; and

FIG. 3 is a partial schematic diagram illustrating the control and feedback for one of the electron beam po- ⁵⁰sitioning controls.

DESCRIPTION OF SPECIFIC EMBODIMENT

Essentially, an electron microprobe comprises a cathode 15 from which an electron beam 16 is emitted, ⁵⁵with the electrons being driven towards a sample 17. The impact of the electrons on the sample produce a secondary emission of X-rays which are indicated on FIG. 1 by the dot-dash line 18. The X-rays are picked up and directed by a reflector 19 to X-ray detectors 20 ⁶⁰incorporated in a probe 21.

The position at which the beam 16 impacts on the sample 17 is controlled in two ways. A relatively coarse control is obtained by reason of the fact that the holder 24, on which the sample 17 is mounted, is movable along an X-axis and along a Y-axis, both at right angles to the line of beam 16. As an aid to illustration, various solely to facilitate understanding of the apparatus. Thus, as indicated by arrow 25, holder 24 is movable along an X-axis. It is also movable along a Y-axis as indicated by arrow 26. The fine control of the position at which the beam strikes the sample is obtained by using magnetic fields to align the position of the electron beam 16. Thus, there are a pair of deflecting coils 27 positioned to produce a magnetic field, when energized, to align the beam 16 along the X-axis. There are also similar coils 28 which produce a magnetic field for aligning the beam along the Y-axis.

The angle at which the X-ray beam 18 will be emitted from the sample 17 depends upon the element upon which the beam impacts. Thus, for one element being struck by the electrons, the X-rays emitted therefrom will be in the form of an inverted cone having a first specific central angle; while, for another element, the central angle of the cone will be a second specific angle. The illustrated line 18 from the sample 17 to the reflector 19 is but one line extending from the apex of one of the cones and there are a multitude of similar lines at other positions about that one cone. The reflector 19 is movable with respect to these cones, as indicated by the arrow 29, so that it will pick up and direct to probe 21 X-rays of the cone having a particular central angle. Thus, for a specific setting of the position of reflector 19, it will be known that the probe is receiving X-rays produced by a particular element at the point of impact on sample 17.

The quantity of that element present at the particular point will be roughly indicated by the current flowing between the sample and ground, and can be measured by a sample current detecting means 32. In effect, this detector is nothing more than a refined current measuring apparatus (e.g. meter). A more accurate measurement of the quantity of the element present is obtained by measuring the quantity of X-rays along path 18 by 40 means of the X-ray detectors 20. These detectors take two forms, often simultaneously incorporated in a single probe; namely, (1) a ratemeter and (2) a scaler. The scaler produces a digital output signal in relation to time. The ratemeter produces a voltage signal. Both, 45 of course, have values indicative of the quantity of Xrays received.

An actual spectrometer will have various other components which are inconsequential so far as the explanation of the present invention is concerned. For example, the electron beam 16 will be in an enclosure in which there is a relatively low air pressure maintained so as not to interfere with the movement of the electron beam; various provisions will be made for accelerating the electron beam; etc. Also, in the conventional spectrometer, there are three probes 21. Each probe is associated with a respective reflector 19, individually movable. These are each positioned at different radial positions about the axis of beam 16 and thus pick up various portions of the cones of emitted X-rays. The purpose of using three, rather than only one illustrated, is merely to permit simultaneous examination for different elements at the point of impact of the electron beam. So far as the present invention is concerned, a ₆₅ description of the apparatus in connection with one of these probes and its associated reflector will fully explain the manner in which the invention is employed in connection with each of the two others.

3,733,484

The exact manner in which the holder 24 is movably mounted and traversed is not consequential. The mechanism for doing this includes a lead screw 33 (threaded rod) operatively connected to holder 24 so that a rotation of the lead screw in one direction or the other will move holder 24 in one direction or the other along the X-axis. Similarly, a lead screw 34 is operatively connected so that its rotation will move the holder along the Y-axis. A third lead screw 35 is operatively connected to reflector 19 so that its rotation will move the reflector to various positions transversely to the various cones of the X-rays to obtain the desired positioning of the reflector to reflect the X-rays in a particular cone from a specific element.

In the present invention an X-axis drive apparatus, generally 36, is connected to lead screw 33. A Y-axis drive apparatus, generally 37, is connected to lead screw 34, and a reflector drive apparatus, generally 38, is connected to lead screw 35. These drives 36, 37, 38 position the respective lead screws in response to input signals on lines 41, 42 and 43 respectively. Each drive apparatus includes a mechanism (hereinafter described) for producing feedback signals indicative of the then position of the drive. These feedback signals appear on output lines 44 46 respectively.

A beam control apparatus, generally 49, is connected to deflecting coils 28. A similar beam control apparatus, generally 50, is connected to deflecting coils 27. Each of these controls adjusts the current flow through the respective coils so as to obtain the desired positioning of the electron beam 16. The control of this current by controls 49 and 50 is dictated by a signal on input lines 51 and 52 respectively. Each control apparatus also includes means for providing a feedback signal indicative of the current control setting. These feedback signals appear on output lines 53 and 54 respectively. Similarly, there is a beam current control apparatus, generally 56, operatively connected to cathode 15 to regulate the magnitude of electron beam 16. This magnitude is dictated by an input signal on line 57. The setting of the control is indicated by a feedback signal on output line 58.

Each of the input lines 41,42,43,51,52,57 are connected to a control matrix, generally 60, which produces signals to the respective apparatus as dictated by a computer, generally 61. Through line 59, computer 61 dictates what signal and to which apparatus the control matrix 60 shall send a signal. The feedback signals on lines 44,45,46,53,54,58 are sent to a selector matrix, generally 62. The selector matrix also receives a signal from the sample current means 32 over output line 63. It receives a voltage signal from the ratemeter in probe 21 over output line 64. Under the dictation of computer 61, as indicated by line 65, the selector matrix sends the feedback signal from one of these components from which is receives feedback signals to an analogue to digital converter (sometimes called a digital voltmeter) generally 66, as indicated by connecting line 67. The converter changes the analogue feedback signal it receives to a digital signal and sends it to a computer on line 68. The digital signal from the scaler in probe 21 is delivered directly to the computer 61 on line 69. The probe also includes a timer in conjunction with the scaler and the time signal therefrom is sent to the computer 61 on line 70, whereby the computer has the time base against which the scaler digital signal is read. Lines 68, 69 and 70 are informational inputs for the computer.

As illustrative of each of the lead screw drive devices 36,37, 38, and the matrixes connected thereto, FIG. 2 illustrates the structure and operation of the X-axis drive 36. This drive apparatus is divided into two components one of which (36α) does the driving, and the other of which (36b) provides the feedback signal.

The drive component 36α comprises a reversible synchronous motor 73, the output shaft of which is connected to a brake 74. The motor shaft is also connected to a clutch 75 and then through a two-speed transmission 76 to lead screw 33. The brake is employed to obtain the quick stopping of the output shaft when the motor 73 is deenergized. The clutch 75 is employed to connect and disconnect the motor from the, transmission 76. The transmission 76 can be set to obtain rapid rotation or a comparatively slow rotation of lead screw 33 for either a quick traversal of holder 24 or a slow traversal thereof.

The control matrix 60 comprises a plurality of substantially identical control segments, only one of which 60α is illustrated in FIG. 2. The number of such control segments that are employed will depend upon the number of control functions to be achieved. In actual embodiments of the present invention, nine such control segments are used providing a total of 81 input signals to various devices. Not all of these input signals are actually used, however.

The control matrix segment 60α comprises four bistable flip-flops 80-83, which receive control signals from computer 61 through lines 59α - S9d respectively. There is also a clock 89 which receives a signal from the computer over line S9e. Line 91 carries signals from clock 89 to each of the flip-flops and to the computer. The output signals from the flip-flops proceed to decoder 92 over lines 93-96. Decoder 92 has nine output lines 98-106, each leading to an amplifier 107. The nine amplifiers have nine output lines 110-118 leading to the solenoids 120α - 128α of relays which also include switches 120b — 128b actuated by the respective solenoid. Preferably, these are reed relays, but various forms of relays or solid state switches could be employed as an alternative. The input line 41, illustrated in FIG. 1, is actually made up of a number of separate lines (41α - 41/in FIG. 2) corresponding to the number of functions or types of actuations incorporated in the drive apparatus 36. Lines 41α and 41b are connected to switches 120b and 121b respectively to energize the motor for forward or reverse operation respectively. The switches are also connected to one side of the source of alternating current 130, the other side being grounded. Line 41c is connected to switch 122b; line 41d is connected to switch 123b; line 41e is connected to switch 124b; and line 41/ is connected to switch 125b. The other side of switches 122b - 125b are connected to a direct current source 131. Thus, for example, when switch 122b is closed brake 74 will be energized from source 131; etc.

A signal on line 59α to flip-flop 80 is indicative of the number 1; a signal on line 59b to flip-flop 82 is indicative of the number 2; a signal on line 59c to flip-flop 82 is indicative of the number 4; and a signal on line 59d to flip-flop 83 is indicative of the number 8. If two or more flip-flops simultaneously receive signals from the computer, their respective numbers are added by decoder 92 to produce a signal on the line that equals the

3,733,484

4' sum of the flip-flop numbers. Thus, for example, if brake 74 were to be energized, coincident signals would be received by flip-flops 80 and 81. At the appropriate time Clock 89 would be actuated to tell flipflops 80 and 81 to release their respective signals to decoder 92 and to clear themselves. The simultaneous signals on lines 93 and 94 would cause decoder 92 to produce an output signal on line 100 indicative of the sum 3. Through the respective amplifier 107, solenoid 122α would be energized to close switch 1226 and energize brake 74.

For purposes of simplification, all of these relays have been shown as single acting relays. In some instances, however, they would be of the latching type with two coils, one of which was energized to latch the relay switch closed and the other of which was energized to unlatch the relay and open the switch. The respective coils would be connected to separate outputs on decoder 92 (through respective amplifiers) to be actuated independently when signals were received from the decoder.

Like the control matrix 60, the selector matrix 62 is divided into a number of segments, one of which 62α is illustrated in FIG. 2. Again, there are four outputs 65a — 65d to the four flip-flops 139—142 respectively. A clock 144 receives a signal from the computer over line 65«? and delivers a signal to the flip-flops and to the computer over line 146. Each flip-flop has an output line to decoder 147, the four lines being numbered 148-150 respectively. There are nine output lines 3° 155-163 from decoder 147 to nine amplifiers 107. In turn the nine amplifiers 107 have nine output lines 165-173 to the nine relay solenoids 175α - 183α. Each solenoid is part of a relay including switches 1756 1836 respectively. One side of each of the switches is connected to line 67. The other side of each of the switches is connected to receive one of the analogue feedback signals from a control apparatus.

In FIG. 2, that portion of the drive apparatus 36 that provides the feedback signal is identified as 366. This includes a potentiometer 185 having a slider or pick-off 186. Slider 186 is coupled to lead screw 33 so that the slider is moved along the potentiometer resistance in accordance with the rotation of the lead screw. Assume that the lead screw 33 is effective in moving holder 24 ⁴⁵through the length D of the lead screw, then slider 186 is coupled so that it moves the full length of, the resistance of potentiometer 185 during the course of the effective operation of the lead screw through the dimension D. Thus, for example, if it takes 270 turns of the lead screw 33 to move the holder 24 through the dimension D, then slider 186 moves one two-hundredseventieth of the length of the resistance of the potentiometer 185 during each rotation of the lead screw. A wire 188 connects one side of the potentiometer resistance to a direct current source of electric power (e.g. 131). The other side of the potentiometer and the source of power are grounded. Thus a voltage will be obtained at slider 186 (in relation to ground), which is indicative of the position of the slider on the potentiometer, and thus indicative of the position of the holder 24 along the lead screw 33.

To obtain even more accurate resolution of the position of the holder along the lead screw, a second potentiometer 190 has a slider 191 which is coupled through the gearing arrangement 192 to the lead screw 33. Potentiometer 190 is capable of continuous rotation and the slider 191 thereof is geared to lead screw 33 in a way such that it turns much more rapidly in relation to the rotation of lead screw 33 than is the movement of slider 186 for a given amount of rotation of the lead screw. For example, slider 191 could rotate through two revolutions of the potentiometer resistance for each revolution of lead screw 33. A continuously rotating potentiometer does have a gap in its path of movement, at which gap the slider is not in contact with the potentiometer resistance. To overcome this problem in actual practice, two such potentiometers 190 are ganged together but aligned so that their gaps are offset from each other. Thus it is possible to avoid a loss of signal at the gap. For the purposes of simplifying the illustration, only a single potentiometer 190 is shown in FIG. 2.

Again, wire 188 connects to one side of potentiometer resistance 190 so that a voltage is impressed across the potentiometer. Thus the voltage appearing at line 44b (in relation to ground) is indicative of the position of slider 191 on the resistance. Utilizing this arrangement, an accuracy in one part in four thousand as to the position of lead screw 43 is obtained without having to use gearing connecting the potentiometers to the lead screw which has an accuracy greater than one percent. In an actual embodiment, the potentiometer 185 was a 10-tum potentiometer and was gear connected to lead screw 33.

To “read” the position of lead screw 33, the computer first sends out a signal to flip-flop 139. Upon the actuation of clock 144, a signal indicating the number 1 is received by decoder 147. In turn, the decoder sends out a signal on its number 1 line, i.e. line 155. Solenoid 175 is thereby energized to close switch 1756. Over line 67 a voltage is obtained by the converter 66 indicative of the position of the slider 186 on the “coarse” potentiometer 185. Converter 66 changes this voltage signal into a digital signal and sends it to computer 61 over line 68. Next the computer reads the “fine” potentiometer 190. To do this, a signal is sent by the computer over line 65b to flip-flop 140. In turn, the decoder 147 sends a signal over lines 156 and 166 to solenoid 176α. This results in a closing of switch 1766. Now a voltage is obtained at line 67 (switch 1756 having opened in the meantime), indicative of the position of the slider 191 on its resistance.

FIG. 3 illustrates the construction and operation of the X-axis beam control apparatus 50 in relation to the computer. The same arrangement would be used for the Y-axis beam control 49, and a similar arrangement for the electron beam current control 56.

The beam control apparatus 50 includes a potentiometer 200 having a slider 201 connected to motor 202 through a gear box 203. A comparable (or actually the same one) gear box 204 connects a slider 205 of potentiometer 206 to motor 202. The resistance of both of the potentiometers are connected to a suitable source of direct current, e.g. source 131, so that a voltage is impressed thereacross. Thus the voltage at the sliders 201 and 205 will vary, depending upon the position of the slider on the potentiometer resistance.

Slider 201 is connected through an amplifier 210 to the deflection coils 27. Amplifier 210 adjusts the magnitude of the current flow through coils 27 in relation to the input potential of the amplifier as picked off by slider 201. Slider 205 is connected by line 54 to switch 1836 of the selector matrix section 62α. Lines 52α and

3,733,484

52b connect the reversible motor 202 to switches 126b and 127b respectively of control matrix section 60α. Thus motor 202 is energized from source 130 in a forward or reverse direction, depending upon the closing of switches 126b and 127b. If, for example, slider 201 is to be moved in one direction along its resistance by closing switch 126b, a signal indicative of the number 7 is sent from the computer to the flip-flops (i.e. signals sent out by the computer on lines 59α, 59b and 59c). Upon the clock 89 directing that these signals be transmitted to the decoder 92, signals are sent to the decoder on lines 93, 94 and 95. In response thereto, the decoder 92 sends out a signal on line 104, through the respective amplifier 107 and through line 116 to solenoid 126α. This closes switch 126b to energize motor 202 for rotation in the desired direction.

In response to the actuation of the motor moving slider 201, a corresponding movement of slider 205 results. Thus, a potential will be obtained on line 54 indicative of the position of sliders 205 and 201 on their respective resistances (and corresponding to the current flow through coils 27). To read this signal the computer sends out signals to the decoder 147 which sends a signal on its number 9 line to energize relay solenoid 183α. The voltage picked off by slider 205 then is transmitted through closed switch 183b to converter 66. Converter 66 sends a corresponding signal in digital form to computer 61.

Assume that it was desired to see what transition elements were located at a particular area on a sample and then to find all other areas of the sample that were similar in composition over a 2 X 2 millimeter area with a resolution of ten micrometers. The initial area of the sample would have been initially selected by an individual looking at the sample through an optical microscope. This microscope would have a holder identical to 24 so that the coordinate would be the same for both holders when the sample was transferred from one to the other. Assume that the initially selected area had coordinates of X-1999 and Y-2999. The program (in english) to the computer would be substantially as follows:

Store X position of 1999.

Read in the actual X position on X-axis drive 36b. Match to the X position in store.

If different, move the drive (lead screw 33) fast if the difference is greater than 10 units, or slow if the difference is less than 10 units, until a match is obtained with the X in the store.

Repeat the foregoing for Y position.

Analyze for the various transition elements by moving the reflector 19 to the required position for each and reading in and storing the results from each position as detected by the probe 21.

As to each of these, recall the reflector position for the particular element. Move the reflector to that position. Read the feedback from the reflector drive apparatus. If match is obtained, then read the ratemeter reading from probe 21 and store. Then move the reflector 19 to the position for obtaining a background reading on the same element. Check the feedback signal from the reflector drive apparatus 38 to make sure that the reflector is in the correct position. If match is obtained, read the ratemeter reading from probe 21 and subtract from the ratemeter reading obtained for the element as initially read. Store the difference. Repeat the same procedure for each of the other elements.

Recall the readings for each of the intensities and if any are above a predetermined minimum, store the element and quantity.

Step the holder in the X direction 10 micrometers. 5 Read the feedback to assure that the new position is correct.

Reposition reflector 19 for each of the elements, and take ratemeter readings (as outlined above). Store the readings for only those positions where the element 10 readings match.

Repeat, in a plurality of steps, the X movements of 10 micrometers and make the same readings at each stop until a total travel of 2 millimeters has been accomplished.

¹⁵ Step the Y-axis drive 37 ten micrometers. Read in the feedback signal from drive 37 to make sure that the right position has been obtained.

Now repeat the stepping of the X-axis movements and take readings at each stop (as previously outlined) ²⁰ until a 2 millimeter travel has been achieved.

Repeat the last two actions until the total coverage over the entire 2 millimeter area (in 10 micrometer increments) has been achieved.

₂₅ Type out each position in readout 216 where a match was obtained corresponding to the readings of the initially selected position.

Various refinements, not illustrated, can be incorporated into the apparatus. Such as, for example, limit 30 switches on the movement of the holder in relation to the lead screws so that exceeding the desired limit of movement will shut down the apparatus. A further refinement would be to read the actual electron beam current and if it dropped below a given minimum, as for 35 example due to filament burnout, the apparatus would be shut down.