JPH1050779A - Electron beam device and method of using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To more accurately obtain the corresponding relationship between a position on an image and a position on a layout design drawing by determining a value of a parameter so that a specific equation becomes the minimum. SOLUTION: A parameter value is determined so that a specific equation becomes the minimum. A weight function g (|q-qi|) becomes large as an interval |q-qi| between a position q and a position qi is shortened and r is a constant. The more the position q is close to the position qi, the more |pi-f(qi)|<r> contributes to an evaluation value S(q). On the other hand, since an expression of coordinate transformation is determined so that the evaluation value becomes the minimum, the corresponding relationship between a position on an image of a sample and a position on a layout design drawing can be accurately obtained more than a case of using ordinary least square. Consequently, the operability of an electron beam device can be improved.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an electron beam apparatus such as an electron microscope and an electron beam tester, and a method of using the same.

[0002]

2. Description of the Related Art An electron beam tester can measure a voltage waveform of a fine wiring pattern on an LSI chip without applying a probe. The electron beam tester has the function of a scanning electron microscope. Before voltage waveform measurement, the electron beam is scanned on the chip while detecting secondary electrons to detect SEM (scanning electron microscopy).
py) Obtain an image, display it on the screen, and specify a measurement point on the screen. However, one side of the SEM image is 50
Since the field of view is narrow, it is difficult to determine which position on the chip is to be observed.

[0003] Therefore, 3
Stage positions p1, p corresponding to points q1, q2, q3
2, p3 is obtained, and a coordinate conversion equation p = f (q) for obtaining the stage position p from the position q on the layout design drawing is obtained in advance. Then, a measurement point is specified on the layout design drawing, a stage position p corresponding to the specified position q is calculated using the coordinate conversion equation p = f (q), and the stage is moved by the position p. Had acquired. The stage position p corresponds to the intersection between the optical axis of the electron beam tester and the surface of the chip mounted on the stage.

FIG. 6 shows an LSI 10 with an open lid as a sample of an electron beam tester. This LSI1
In the package 11 of No. 0, a concave portion 12 corresponding to the lid is formed, a concave portion 13 is further formed inside the concave portion 12, and a semiconductor chip 14 is disposed in the concave portion 13. Step 1
5 is slightly higher than the surface of the semiconductor chip 14, and the inner end of the inner lead 16 is arranged on the step 15. The inner end of the inner lead 16 and a pad (not shown) on the semiconductor chip 14 are connected by a bonding wire 17, and the inner lead 16 is formed integrally with the pin 18.

[0005]

With respect to the semiconductor chip 14 of 6 mm × 12 mm, the position q on the layout design drawing
Is obtained by a coordinate conversion equation p = f (q), an electron beam is scanned around the obtained position p to obtain an SEM image, and the actual position p corresponding to the position q is actually obtained. When the displacement Δp = pp−p ′ of the calculated position p in the X direction and the Y direction with respect to the position p ′ on the semiconductor chip 14 is actually measured, the displacement greatly differs depending on the positions p = p1 to p14, and the displacement Δpx in the X direction is The deviation Δpy in the Y direction was −16 to 34 μm, and was −7 to 33 μm. Arrows in the figure are vectors in the X and Y directions of the displacement at the positions p1 to p14.

The cause of such irregular displacement is that the stepped portion 15 of the insulator is charged by the scattered electrons and ions to form a local electric field, and the electron beam is deflected by this electric field. Conceivable. Since the position p is deviated from the target position p ′, the operator visually checks the SEM image to determine the position p ′.
And a position p ′ is designated for the electron beam tester.

As described above, the positional deviation of the SEM image with respect to the field size is relatively large, and the positional deviation varies greatly depending on the position on the chip, and there is no regularity. Therefore, an accurate corresponding point p 'on the SEM image is found. This is not easy, and the operability is poor. In order to automatically correct the positional deviation, a method of obtaining a positional deviation by performing pattern matching between an SEM image and a layout design corresponding thereto has been used (Japanese Patent Laid-Open No. 5-101166).

However, in the pattern matching, the process of calculating the coincidence every time one of the two images is shifted by one pixel is repeated until the matching is successful. Therefore, if the displacement is large, the time required for the pattern matching becomes long. Also, pattern matching may fail. In this case, the operator has to search for the corresponding position p ′ on the SEM image and specify it, and thus the operability is poor.

SUMMARY OF THE INVENTION In view of the above problems, it is an object of the present invention to improve the operability by more accurately obtaining the correspondence between the position on the SEM image of the sample and the position on the layout design drawing. It is an object of the present invention to provide an electron beam apparatus and a method of using the same.

[0010]

In the first invention, an electron beam apparatus having a function as an electron microscope is used, and the SEM including a predetermined position on a sample is used by the function.
In order to make the position at the time of obtaining an image correspond to the position on the layout design drawing of the sample, the other p is calculated from one q of the two positions based on a coordinate conversion equation p = f (q) including parameters. In the method of using the electron beam apparatus, before the calculation, the positions pi, i = 1 to m of the other m points corresponding to the positions qi, i = 1 to m of the one m points (m ≧ 4), respectively. Is actually obtained, and for the value of q, Is determined so that the value of the parameter becomes minimum. Here, the weight function g (| q-qi |) is such that the function value increases as the interval | q-qi | between the position q and the position qi decreases. , R are constants.

According to the first aspect, the position q is equal to the position qi.
Is closer to, the contribution of | pi-f (qi) | ^{r to} the evaluation value S (q) is larger. On the other hand, the coordinate transformation formula is determined so that this evaluation value is minimized. Compared with the case where the sample is used, the correspondence between the position on the SEM image of the sample and the position on the layout design drawing can be more accurately obtained, and the operability of the electron beam device is improved.

In the first aspect of the first invention, r = 2, g (|
q-qi |) = 1 / {1 + α | q-qi |}, where α is a constant. According to the first aspect, the calculation is simplified. In a second aspect of the first invention, a sample is mounted on a moving stage, a position q of the sample on a layout design drawing is specified, and the coordinate conversion equation p = f that defines the parameters by the method according to claim 1. Based on (q), a position p of the moving stage corresponding to the position q is calculated, the moving stage is moved based on the position p, and an SEM including the position p is calculated.
Get an image.

In a third aspect of the first invention, the electron beam device is an electron beam tester, and the SEM image is acquired by the method according to claim 3, and the SEM image corresponding to the SEM image is obtained based on the position q. The layout design data of the sample is read, the layout design data is converted into a layout design drawing corresponding to the SEM image, and a displacement of the position p is obtained by pattern matching between the SEM image and the layout design drawing. The electron beam irradiation position shift is corrected based on the above, and the position on the sample corresponding to the position q is irradiated with the electron beam to measure the voltage waveform at the irradiation point.

When the maximum value of the matching degree of the pattern matching is equal to or less than the set value, it is determined that the matching has failed, and the operator must specify an accurate position. According to the third aspect, the deviation is small. Therefore, the matching failure rate is reduced as compared with the related art, and the operability is improved. In addition, since the deviation is reduced, the matching processing time is shortened, and the effect of improving the throughput is obtained.

In the second invention, a function as an electron microscope is provided, and in order to make the position when obtaining an SEM image including a predetermined position on the sample correspond to the position on the layout design drawing of the sample by the function. , A coordinate transformation expression p including parameters
= F (q), an electron beam apparatus provided with a calculating means for calculating one of the two positions from the other q, the position qi of one of the m points (m ≧ 4), i = 1 to m, and Position q
i, the position p of the other m point corresponding to each of i = 1 to m
i, i = 1 to m are stored, and the calculating means calculates the value of the one q before the calculation. Is determined so that the value of the parameter becomes minimum. Here, the weight function g (| q-qi |) is such that the function value increases as the interval | q-qi | between the position q and the position qi decreases. , R are constants.

In the first embodiment of the second invention, r = 2, g (|
q-qi |) = 1 / {1 + α | q-qi |}, where α is a constant. According to a second aspect of the second invention, there is provided a moving stage on which a sample is mounted, input means for designating a position q of the sample on a layout design drawing, and image storage means. The position p of the moving stage corresponding to the position q is calculated on the basis of the coordinate conversion equation p = f (q) that defines the following formula, and the moving stage is moved based on the position p to move the position p. SEM containing
Control means for storing an image in the image storage means.

According to a third aspect of the second invention, there is provided design data storage means for storing the layout design data of the sample, wherein the control means stores the layout design data of the sample based on the position q. The layout design data is read from the design data storage unit, the layout design data is converted into a layout design drawing corresponding to the SEM image, a shift of the position p is obtained by pattern matching between the SEM image and the layout design drawing, and based on the shift, To correct the electron beam irradiation position deviation, and irradiate the electron beam to a position on the sample corresponding to the position q for voltage waveform measurement.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows a schematic configuration of an electron beam tester according to one embodiment of the present invention. The electron beam tester main body 30 includes an electron gun 32, an alignment coil 33, a blanking electrode 34, an aperture 35, a converging magnetic field lens 36, a deflector 37, and an objective magnetic field lens 38 in a column 31.
Are disposed, and an energy analysis electrode 40, a control electrode 41, and an extraction electrode 42 are attached to a cylinder 39 disposed inside the objective magnetic field lens 38. A moving stage 44 is disposed in the sample chamber 43 of the electron beam tester main body 30, and a DUT board 45 is mounted at a predetermined position on the moving stage 44. The LSI 10 in FIG.
It is inserted into the socket of the UT board 45.

The electron beam EB emitted from the electron gun 32
1 is convergently irradiated on the semiconductor chip 14 and secondary electrons EB2 emitted from the irradiation point are detected by a secondary electron detector 46 disposed between the deflector 37 and the objective magnetic lens 38. The detected amount changes depending on the irradiation point potential on the semiconductor chip 14 and the potential of the energy analysis electrode 40, and the irradiation point potential is measured based on the relationship and the output of the secondary electron detector 46. Can be.

The components in the electron beam tester main body 30 are controlled by an electron beam controller 51 based on instructions from a computer 50. The movement of the moving stage 44 is controlled by the stage control device 52 based on an instruction from the computer 50. The position of the moving stage 44 is detected by a laser interferometer provided in the stage controller 52 and supplied to the computer 50.

The signal input device 53 includes a secondary electron detector 4
6 and an electron beam deflection amount target value from the electron beam controller 51 are supplied. The signal input device 53 is SE
When an M image is obtained, the output of the secondary electron detector 46 is amplified, converted into a digital value, a storage address is obtained based on the target value of the amount of electron beam deflection, and this is stored in the SEM image frame memory 54. The secondary electron detection value is stored in the address. The contents of the SEM image frame memory 54 are displayed on the image display device 55.

When measuring the voltage waveform at the electron beam irradiation point, the signal input device 53 supplies the secondary electron detection value to the measurement waveform processing device 56. Measurement waveform processing device 56
Measures the potential of the electron beam irradiation point based on this value and the energy analysis voltage set value from the electron beam control device 51 to determine its waveform, and causes the image display device 57 to display the waveform.

On the other hand, the external storage device 58 stores CAD data (layout design data) of a pattern on the semiconductor chip 14. The computer 50 reads data in a range corresponding to the size of the SEM image stored in the SEM image frame memory 54 from the external storage device 58, converts the data into layout design drawing data for display, and stores the data in the layout diagram frame memory 59. I do. The contents of the layout diagram frame memory 59 are displayed on the image display device 55.

The computer 50 has a manual operation input device 6
0 and the position memory 61 are connected. The manual operation input device 60 includes a pointing device for inputting the designated position q on the layout design drawing of the semiconductor chip 14 and the position shift Δp on the SEM image corresponding to the designated position q to the computer 50. And a keyboard for giving other instructions to the computer 50. In the position memory 61, the position q and the position p where the deviation has been corrected are stored correspondingly.

Next, the operation of the electron beam tester configured as described above will be described with reference to FIGS. (S1) An initial value 1 is substituted for a repetition number identification variable k. (S2) By operating the manual input device 60, a point qk on the layout design drawing of the semiconductor chip 14 is designated. This is, for example, by specifying a point on the layout design drawing with a digitizer that is one of the manual operation input devices 60, or by specifying an enlargement ratio with a keyboard that is one of the manual operation input devices 60,
The layout design drawing of this enlargement ratio is displayed together with the cursor on the image display device 55, and a point on the layout design drawing is designated by a mouse which is one of the manual operation input devices 60. Point q1
To qm are designated points for coordinate transformation formula determination, and points qr to qn
(R = m + 1) is a designated point for voltage waveform measurement. m and the designated points q1 to qm are stored in the position memory 61.

FIG. 4 shows, as a simplified example, m = 6, n
The designated points q1 to q12 on the layout design drawing of the semiconductor chip 14 when = 12 are shown. (S3) The layout design data of the SEM image display range centered on the position qk is read from the external storage device 58, and converted into layout design drawing data for display on the image display device 55, and the layout diagram frame memory 59
To be stored.

(S4) Thus, the layout design drawing display on the screen of the image display device 55 is updated. As shown in FIG. 5, the layout design drawing 59a is displayed on the left half of the screen 55a on the image display device 55. The position qk coincides with the center point of the layout design drawing 59a. (S5) If k <4, proceed to step S6, if k = 4, proceed to step S14, and if k> 4, proceed to step S15.

(S6) The position pk = f (qk) of the moving stage 44 corresponding to the position qk on the layout design drawing is calculated and supplied to the stage controller 52. The linear coordinate conversion equation p = f (q) expresses the coordinates of the positions p and q as p = (px,
py) and q = (qx, qy) are represented by the following equations. px = A11 · qx + A12 · qy + A13 (1) py = A21 · qx + A22 · qy + A23 (2) This coordinate conversion formula includes translation, rotation, and reduction. In the present embodiment, the coefficients A11, A12, A1
3, A21, A22 and A23 are used as parameters, and by updating the values of the parameters, the coordinate conversion equation p = f
Update (q).

For k = 1 to 3, the moving stage 44
From the position of the semiconductor chip 14 above, the size of the semiconductor chip 14 and the layout design enlargement ratio, the coordinate conversion equation p = f
The parameter of (q) is obtained in advance. Alternatively, the position p is obtained by manually moving the moving stage 44. (S7) In response, the stage control device 52 moves the moving stage 44 to the position pk. With this movement,
In an ideal case, the intersection between the optical axis of the electron beam tester and the surface of the chip mounted on the stage corresponds to the position qk, but there is actually a deviation Δpk.

(S8) The semiconductor chip 14 is scanned with the electron beam EB1. At this time, the secondary electron EB2 is detected by the secondary electron detector 46, and the SEM image frame memory 54
Stores an SEM image. (S9) Thereby, the SE on the screen of the image display device 55 is
The display of the M image is updated. As shown in FIG.
As shown in the figure, the SEM image 54a is displayed on the right half of the screen 55a. There is a one-to-one relationship between the position of the moving stage 44 and the center position of the SEM image 54a, and both positions are set to the same position p.
Expressed by k. The position qk actually corresponds to pk ′ in the wiring pattern 70A in the SEM image 54a. The wiring pattern 70A corresponds to the wiring pattern 70 in the layout design drawing 59a.

(S10) If k ≦ m, step S11
Otherwise, the process proceeds to step S16. (S11) The pk 'on the SEM image 54a is specified by operating the manual operation input device 60, for example, a mouse. (S12) The computer 50 calculates the deviation Δpk = pk′−
pk is obtained, corrected as pk ← pk + Δpk, and the corrected position pk is stored in the position memory 61 in association with the position qk.

(S13) The variable k is incremented by one. The processing of steps S2 to S13 is repeated three times,
In the next fourth step S5, it is determined that k = 4, and the process proceeds to step S14. If the number of repetitions k is greater than 4, the process proceeds to step S15. (S14) Using the three sets of positions (p1, q1), (p2, q2) and (p3, q3) stored in the position memory 61, the parameters of the linear coordinate conversion equation p = f (q) are obtained. Thus, the initial coordinate conversion formula is updated. This equation is used for k = 4 to m and is more accurate than the original equation, so that the designation of the position pk ′ in step S14 is k
= 1 to 3 becomes easier.

Next, the process proceeds to step S6. (S15) Using m pairs (pi, qi) of positions stored in the position memory 61, i = 1 to m, and when q = qk,
The next evaluation value S (q), The above parameters of the coordinate conversion equation p = f (q) are determined so that is minimized. Here, the weight function g (| q-qi
|) Is such that the smaller the interval | q-qi | between the position q and the position qi becomes, the larger the function value becomes. For example, g (| q-qi |) = 1 / ｛1 + α | q-qi | ^r ｝ (4). α and r are constants. The value of r is, for example, 2. The value of α is, for example, α = 100 / (length of one side of the chip).

Using the coordinate transformation equation determined in this manner, as the position q is closer to the position qi, | pi-f (q
i) The contribution of | ^{r to} the evaluation value S (q) is large, and the coordinate transformation formula is determined so that this evaluation value is minimized. Δpk decreases, and the position pk ′ on the SEM image 54a
Is easier to specify than before.

Next, the processing of steps S6 to S9 is performed,
If k ≦ m in step S10, the process proceeds to step S11, and if k> m, the process proceeds to step S16. (S16) S stored in the SEM image frame memory 54
The computer 50 performs matching between the EM image and the layout design diagram stored in the layout diagram frame memory 59. That is, for each of the X-axis projection luminances of both images, the degree of coincidence is calculated each time one of the images is shifted by one pixel in the X direction.
The shift amount when the degree of coincidence is maximized is represented by X at the stage position.
It is obtained as a direction shift Δpx or −Δpx. The same applies to the Y axis perpendicular to the X axis.

If the maximum value of the degree of coincidence is equal to or less than the set value, it is determined that matching has failed, and the operator must specify pk 'on the SEM image 54a as in step S14. However, as described in step S15, the deviation Δp
Since k is small, the matching failure rate is lower than in the past, and the operability is improved. Further, since the shift Δpk becomes small, the matching processing time is shortened, and the throughput is improved.

(S17) Position qk on layout design drawing
So that the position pk ′ on the semiconductor chip 14 corresponding to the above is located on the optical axis of the electron beam tester main body 30 or the electron beam EB1 is irradiated at this position by deflecting the electron beam EB1 by the deflector 37. Stage position deviation Δp
Is corrected. (S18) The position pk 'on the semiconductor chip 14 is irradiated with the electron beam EB1, and the voltage waveform at the irradiation point is measured. Next, the process proceeds to step S13.

The present invention also includes various modifications. For example, instead of step S11, step S
16 is executed, and only when the matching fails, the step S1 is executed.
1 may be performed. Further, after the processing in step S17, a set of the position qk and the corrected position pk may be added for the parameter determination.

The coordinate conversion equation may be a nonlinear equation including, for example, a quadratic term. In step S2, the semiconductor chip 14
Point q1 to qn on the layout design drawing may be designated, and the process may return from step S13 to step S3. Further, contrary to the above embodiment, the SEM image 5
To indicate where 4a is on the layout plan,
A configuration may be used in which a position on the layout design drawing is calculated from the position of the moving stage 44 based on the coordinate conversion formula, and this position is taught to the operator.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of an electron beam tester according to an embodiment of the present invention.

FIG. 2 is a flowchart showing an operation of the electron beam tester of FIG.

FIG. 3 is a flowchart showing a continuation of FIG. 2;

FIG. 4 is a diagram showing designated points on a layout design drawing of a semiconductor chip.

FIG. 5 is a diagram showing an image in which a layout design drawing and an SEM image are displayed in correspondence.

FIG. 6 is an explanatory diagram of a problem in the related art.

[Explanation of symbols]

 Reference Signs List 14 semiconductor chip 30 electron beam tester main body 32 electron gun 37 deflector 38 objective magnetic field lens 40 energy analysis electrode 44 moving stage 50 computer 51 electron beam controller 52 stage controller 53 signal input device 54 SEM image frame memory 55, 57 image display Device 56 Measured waveform processing device 58 External storage device 59 Layout diagram frame memory 60 Manual input device 61 Position memory

Claims (8)

[Claims] 1. Using an electron beam apparatus having a function as an electron microscope, the position when obtaining an SEM image including a predetermined position on a sample by the function corresponds to the position on the layout design drawing of the sample. In the electron beam apparatus using method for calculating the other p from one q of both positions based on a coordinate transformation equation p = f (q) including a parameter, the one m-point (m ≧ 4) The positions pi, i = 1 to m of the other m points corresponding to each of the positions qi, i = 1 to m are actually obtained, and for the value of the one q, Is determined so that the value of the parameter becomes minimum. Here, the weight function g (| q-qi |) is such that the function value increases as the interval | q-qi | between the position q and the position qi decreases. , R is a constant, The method of using an electron beam device. 2. r = 2, g (| q-qi |) = 1 / ｛1
2. The method according to claim 1, wherein + α | q−qi |｝, where α is a constant. 3. A sample is mounted on a moving stage, a position q on the layout design drawing of the sample is designated, and the coordinate conversion equation p = f (q) that defines the parameters by the method according to claim 1. Calculating a position p of the moving stage corresponding to the position q, moving the moving stage based on the position p, and acquiring an SEM image including the position p. How to use the device. 4. The electron beam apparatus is an electron beam tester, acquires the SEM image by the method according to claim 3, and, based on the position q, generates layout design data of the sample corresponding to the SEM image. The layout design data is read and converted into a layout design drawing corresponding to the SEM image, a shift of the position p is obtained by pattern matching between the SEM image and the layout design drawing, and an electron beam irradiation position is determined based on the shift. Correcting the displacement and irradiating an electron beam to a position on the sample corresponding to the position q to measure a voltage waveform at an irradiation point. 5. A function as an electron microscope is provided, and a parameter is set in order to make the position at the time of obtaining an SEM image including a predetermined position on the sample by the function correspond to the position on the layout design drawing of the sample. An electron beam apparatus provided with an arithmetic means for calculating the other p from one q of both positions based on a coordinate transformation equation p = f (q) including the position qi, i = 1 to m and the position pi, i = 1 to m of the other m point corresponding to each of the positions qi, i = 1 to m. Where, for the value of the one q, Is determined so that is minimized, where the weighting function g (| q-qi |) is determined by the position q and the position qi
The function value increases as the interval | q−qi | decreases, and r is a constant. 6. r = 2, g (| q-qi |) = 1 / ｛1
The electron beam apparatus according to claim 5, wherein + α | q-qi |｝, where α is a constant. 7. A moving stage on which a sample is mounted, an input unit for designating a position q of the sample on a layout design drawing, and an image storage unit, wherein the calculating unit sets the parameter. A position p of the moving stage corresponding to the position q is calculated based on a coordinate transformation equation p = f (q), and further, the moving stage is moved based on the position p and an SEM image including the position p 7. An electron beam apparatus according to claim 5, further comprising: control means for storing in the image storage means. 8. Design data storage means for storing layout design data of the sample, wherein the control means reads layout design data of the sample from the design data storage means based on the position q, The layout design data is converted into a layout design drawing corresponding to the SEM image, a shift of the position p is obtained by pattern matching between the SEM image and the layout design drawing, and an electron beam irradiation position shift is determined based on the shift. 8. The electron beam apparatus according to claim 7, wherein the electron beam is corrected and a position on the sample corresponding to the position q is irradiated with an electron beam for measuring a voltage waveform.