JP2002352764A - Method for calibrating electron microscope, and standard sample for calibrating the electron microscope - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a calibration method of an electron microscope, which can fully expand the range, where the calibration can be made by using standard samples. SOLUTION: Samples A to E, consisting of a plurality of particles having different dimensions in steps, are dispersed and are held on a standard microscale M. First, the dimensions of a single particle among the plurality of particles is calibrated to define a second standard sample by using the pattern of the standard microscale M, and then the dimensions of another single particle among the plurality of particles is calibrated to define a third standard sample, by using the second standard sample. Subsequently, the same process is repeated about the other particles to define standard samples from the second to n-th sample. After the pattern of the reference sample is added as the first standard sample on the standard samples from the second to the n-th, calibration is implemented by using one of these standard samples from the first to the n-th.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for calibrating the magnification and dimensions of an electron microscope using a sample having a guaranteed pattern of known dimensions as a reference, and more particularly to a method for observing a scanning electron microscope. The present invention relates to a method for calibrating an electron microscope and a standard sample for calibrating an electron microscope suitable for a case where the size of an unknown sample becomes wide.

[0002]

2. Description of the Related Art Various electron microscopes, such as a scanning electron microscope, sometimes require calibration (calibration) of magnification and dimensions when operating them. A reference sample having the following pattern is used.

At this time, dimensions measured by an electron microscope
The (magnification display) changes according to various conditions such as an acceleration voltage, a magnification, a lens condition, a working distance of a sample, and a state of an electron source. Therefore, in order to calibrate the magnification and dimensions of the unknown sample, it is necessary to measure the dimensions of the standard sample under the same conditions as the measurement conditions of the unknown sample, and use that as a reference.

[0004] Here, as a reference sample, a JQ using the pitch of a lattice (striped) pattern of silicon as a dimensional reference is used.
There is a standard microscale certified by A (Japan Quality Assurance Organization). Therefore, in the prior art, the standard microscale is used as a reference sample to calibrate magnification and dimensions.

[0005]

The prior art described above does not take into account that the magnification that can be calibrated is determined by the standard microscale, and cannot accurately cope with a wide range of observation magnification. There was a problem.

That is, in the prior art, since the pitch of the standard micro-scale grid pattern is compared with the unknown sample, considerable accuracy can be expected when the sizes are close to each other, but the accuracy increases as the size increases. Since it is reduced, the range of application is limited, and the accuracy is also limited.

More specifically, in the case of the standard micro-scale, the dimension of the pattern is one pitch (0.24).
0 μm ± 0.001 μm), an observation magnification of 300,000 to 100,000 is in an appropriate range for maintaining high accuracy. If the magnification is out of this magnification range, the pattern pitch cannot be correctly determined. In addition, the calibration cannot be performed.

[0008] Here, the invention according to the application of Japanese Patent Application No. 2000-144573 proposes a method of directly dispersing target microparticles of unknown dimensions on a standard microscale and calibrating the dimensions of the microparticles.

However, also in this case, there is no difference in that the fine particles dispersed on the standard micro-scale are calibrated in comparison with the pitch of the grid pattern of the standard micro-scale. Does not mean that it can be expanded.

[0010] Particularly recently, the International Standard for Quality Management and Quality Assurance established by the International Organization for Standardization (ISO)
The need for improving the reliability of measurement and analysis results has been strongly pointed out by the 9000 series) and the laboratory accreditation system (ISO Guido 30), which is the basis of international mutual recognition of measurement and analysis results.

Therefore, even in a scanning electron microscope, a highly reliable measurement method is not limited to a certain magnification (dimension) range but a wide observation magnification, that is, a measurement method capable of clearly indicating "accuracy" (= uncertainty). The development of a method is desired, but in the prior art, it was specified by the standard microscale and could not respond to this.

An object of the present invention is to provide a dimensional calibration method and a standard sample for electron microscope calibration, which can sufficiently expand the range that can be calibrated by a standard sample.

[0013]

SUMMARY OF THE INVENTION An object of the present invention is to provide a method for calibrating an electron microscope using a reference sample having a pattern of a known size and using the pattern as a standard sample, wherein the size is stepwise different from the known size. A plurality of particles are dispersed and supported on the observation surface of the reference sample, and first, one of the plurality of particles is calibrated according to the pattern of the reference sample to obtain a second standard sample. The other one of the particles is calibrated with the second standard sample to obtain a third standard sample, and thereafter, the same process is repeated for the remaining particles, so that the second to the second particles are obtained. n, and after including the pattern of the reference sample as the first standard sample in the second to n-th standard samples, using any of the first to n-th standard samples. To calibrate Ri is achieved.

At this time, the unknown sample may be supported on the observation surface of the reference sample, and the unknown sample has a size outside the calibration range of the first standard sample, and The calibration may be performed using the second to n-th standard samples.

Another object of the present invention is to disperse and support a plurality of particles having different sizes stepwise from the known size on an observation surface of a reference sample having a pattern of a known size to obtain a standard sample for electron microscope calibration. Is achieved by

It is still another object of the present invention to disperse and support a plurality of particles having different sizes stepwise from the known size on an observation surface of a reference sample having a pattern of a known size, and to support the unknown sample on the observation surface. This is achieved by using the standard sample for electron microscope calibration as described above.

[0017]

Hereinafter, a method for calibrating an electron microscope according to the present invention will be described in detail with reference to the illustrated embodiments. FIG. 1 shows an example of a reference sample R used in an embodiment of the present invention, in which unknown samples A to E made of particles are provided on an observation surface of a standard microscale M.

First, the standard microscale M is
According to the above-mentioned JQA certification, it has a pattern P composed of a plurality of grooves arranged at a pitch of 0.240 μm.

Next, as shown in the drawing, the samples A to E have different sizes (sizes and dimensions) from each other, and a range of magnifications at which dimensional comparison measurement can be performed by each of the samples A to E. Is a combination that overlaps the range of magnification that can be measured by at least one of the standard microscale M and other samples among the samples A to E.

Here, the reference sample R is prepared, for example, as follows. First, samples having various particle sizes, for example, latex particles are prepared. Then, the particles are dropped and sprayed on the observation surface of the standard microscale M, or a liquid in which the particles are suspended is dropped, and then the remaining particles are blown off to adhere the remaining particles in a dispersed state. A to E.

At this time, the sizes of these samples A to E are as follows:
It is desirable to be able to handle all magnifications of the applied electron microscope.

FIG. 2 shows an example of a scanning electron microscope to which one embodiment of the present invention is applied.
The electron beam 4 generated from the electron gun (filament) 1 by the condenser lens 5 and the objective lens 6
The sample placed on the sample stage 7 on the sample stage 8, for example, is converged on the surface of a reference sample R.

For this reason, a voltage controlled by the high voltage control unit 13 is applied to the electron gun 1, the extraction electrode 2 and the acceleration electrode 3, and the condenser lens 5 and the objective lens 6 are controlled by the lens control unit 15. Current is supplied.

An X-direction scanning coil 9 and a Y-direction scanning coil 10 are provided in the path of the electron beam 4. A predetermined deflection current is supplied to these coils from the magnification control unit 14 according to an arbitrary set magnification. And thereby the electron beam 4
Is deflected to scan the surface of the sample on the sample stage 7.

Here, the high-voltage application control condition, the lens control condition, the magnification control condition and the like until the electron beam 4 scans the surface of the sample are controlled by the length measurement condition control unit 17.
As a result of the electron beam 4 being converged on the surface of the sample, for example, the reference sample R, the secondary electrons 11
Occurs.

The secondary electrons 11 are collected by the detector 12 to become an image signal, and this image signal is stored in the memory of the image monitor / memory unit 16 together with the measurement conditions at this time. Then, using the data stored in the image monitor / memory unit 16, the dimension measurement is performed by the length measurement control unit 18.

The image signal at this time is transmitted to the monitor (display device such as CRT or liquid crystal) of the image monitor / memory section 16.
, Where it is displayed as a scanning electron microscope image and made available for observation by an operator.

Next, the dimension measurement and magnification calibration by the length measurement control unit 18 will be described. First, acceleration voltage, extraction voltage, emission current,
The length measurement conditions such as the excitation of the condenser lens, the magnification, the sample working distance, and the scan speed are given by providing the length measurement condition control unit 17 with a corresponding memory table for at least one of the dimensional calibration coefficient and the magnification correction coefficient. .

The measurement conditions stored in the memory table at this time need not necessarily be limited to the above conditions, and need not be all of the above conditions, and affect the magnification and size of the scanning electron microscope. Any measurement condition may be used. For example, when other conditions are constant and the dimension does not change so much in a certain magnification range, the coefficient input field may be previously set to one column in a certain magnification range. .

Then, using the reference sample R shown in FIG.
The length is measured in advance under the measurement conditions of the unknown sample that is predicted by the method described later with reference to FIG. 7, and the magnification correction and the dimensional calibration coefficient are determined, and are stored in the memory table.
Then, the length measurement control unit 18 determines whether or not the memory table under the current length measurement conditions is satisfied.
It is displayed on the monitor of the memory unit 16.

On the other hand, when an unknown sample is measured under the condition that the dimensional calibration coefficient and the magnification correction coefficient are not determined, the measurement conditions are stored as necessary conditions for length measurement calibration, read out, and re-read to the measurement condition control unit 17. After that, the length is measured using the reference sample R in the same manner as described later with reference to FIG. 7, and the dimensional calibration coefficient and the magnification correction coefficient are determined and stored in the memory table.

Here, the magnification correction of the memory image may be performed by the following method. Change the display magnification or size on the image. Change the display size of the image. When performing magnification correction of an image in real time, the length measurement control unit 18 changes the magnification control condition. Then, when performing the dimension measurement, the dimension is calibrated by the calibration coefficient.

Next, the operation of this embodiment will be described with reference to FIG.
This will be described with reference to the flowchart of FIG. In this embodiment,
When the dimensions of the unknown samples A to E are measured with high precision using the reference sample R, the dimensions and magnification are first calibrated using the pattern of the standard microscale M as a standard sample, and then the unknown samples A to E are obtained under the same observation conditions. E needs to be measured.

When the processing of FIG. 3 is started, first, observation conditions for an unknown sample are set (processing 301). For example, when measuring the dimensions of the unknown sample A (FIG. 1), the observation conditions are predicted and set so that the magnification is appropriate for the unknown sample A, that is, the state shown in FIG. is there. At this time, save the measurement data.

Next, it is determined whether or not the dimensions of the standard microscale M can be measured at this observation magnification (step 302).
If it is determined in step 302 that measurement is possible, step 303
Proceed to the standard sample (standard microscale M pattern)
To calibrate the magnification of the electron microscope (step 303).

Next, it is checked whether or not the measurement data of the unknown sample (in this case, unknown sample A) has already been taken (process 3).
04). If it is determined in step 304 that there is data, the data is called up (step 306), and then the data is calibrated (step 307). On the other hand, when there is no data, step 305 is executed instead of step 306.

Here, the relationship between the observation magnification of the standard sample and that of the unknown sample can be represented schematically as shown in FIG. The observation conditions other than the magnification are the same, and the size relationship between the standard sample and the unknown sample can be expressed in the same way even if they are opposite.

First, in the case of FIG. 4, the magnification is in the region b in FIG.
In the range. Therefore, in this case, within the magnification range (area a) in which the dimensions of the standard sample can be measured, the processing of FIG. It means that dimensional calibration of A is possible.

Therefore, in this case, it is not necessary to exchange the sample or search for the visual field, and it is not necessary to perform a difficult operation such as setting the same observation conditions, and it is possible to measure the dimensions with high accuracy and high throughput. Here, according to ISO Guide 30,
The "uncertainty" of a measurement in a scanning electron microscope (the range over which a true value is presumed to exist) can be used to quantitatively represent "accuracy".

The "uncertainty" is represented by the following equation as a general equation. _{Uc = √ {(U 1)} 2 + (U 2) 2 + (U 3) 2} Uc: Synthesis uncertainty (%) U _1: uncertainty of the standard sample U _2: Uncertainty of magnification calibration U _3: sample Measurement uncertainty (instrument-dependent and sample-dependent uncertainties)

Therefore, in the above case, pitches at different positions of the standard microscale M having a pitch of 0.240 μm and an error of 2σ = 0.0006 μm were measured 20 times, and the average value was set. The diameter was measured 10 times, the average particle diameter was 663.4 nm, and the error was 3σ (n−1) = 3.
Assuming that 5 nm is obtained, in this case, U ₁ = (0.6 / 2) /240×100=0.125 (%) U ₂ = 0.5 × t (α, n−1) S / √N t (α, n−1): t distribution Sampling number N = 20, confidence interval 95% = 0.234S (nm).

Here, the magnification calibration uncertainty S = 1.677n
If m is used, U ₂ = 0.234 × 1.677 = 0.390 (nm) = 0.390 / 240 × 100 = 0.163 (%) U ₃ = 3.5 / 3 / 663.4 × 100 = 0.176 (%) Uc = √ {(0.125) ² + (0.163) ² + (0.17
6) ² ｝ = 0.262 (%), and the combined uncertainty is calculated to be 0.262%.

The above is based on the following document. "High Resolution Electron Microscopy for Material Evaluation" Daisuke Endo Kenji Hiraga Co-authored 1996 Kyoritsu Shuppan

By the way, the standard microscale M used for the reference sample R has a fixed pitch width as shown in FIG. 5A, and there is a magnification range suitable for measurement at this pitch width. .

Therefore, if the observation is made at a high magnification, for example, at a high magnification of 100,000 or more, as shown in FIG. 5B, one pitch of the pattern cannot enter the field of view on the monitor surface. The exact pitch cannot be measured.

On the other hand, at low magnification, as shown in FIG. 5C, the pitch interval of the pattern is too narrow, so that the pitch cannot be measured accurately, and calibration cannot be performed in any case. In such a case, in the flowchart of FIG.
The result at 2 is No (No).

Therefore, in this case, it is then examined whether or not the samples that can be measured at both the observation magnification of the unknown sample and the magnification at which the dimension of the standard sample can be measured are present in the unknown samples A to E. (Process 308).

Now, for example, if the object to be measured is an unknown sample F, in this case, the magnification at which the sample F can be measured (region e in FIG. 7) and the magnification at which the standard sample can be measured (FIG. 7) In both regions a), there is a sample E that can be measured (observable region: region c in FIG. 7).

Therefore, in this case, first, at the magnification (region b in FIG. 7) shown in FIG. 6A, the dimension calibration coefficient of the standard sample (pattern) is determined, and then the dimension of the unknown sample E is measured. (Step 309). The dimension-corrected unknown sample E is used as a second standard sample.

Next, as shown in FIG. 6B, the observation conditions were reset to a magnification (region d in FIG. 7) at which both unknown samples E and F can be measured, and then the second standard sample was set. Is measured, and the dimension calibration constant is determined, and then the dimension of the unknown sample F is measured.
(Process 310).

The uncertainty of the measured value at this time is calculated in the same manner from the above equation by _first obtaining the uncertainty of the sample E as the second standard sample by the above-described method and using this value as U _1. it can.

Therefore, according to this embodiment, even when calibrating from the second standard sample, "uncertainty" can be specified, and as a result, it is possible to appropriately cope with the expansion of the calibratable range by the standard sample. .

Next, if the unknown sample F is an unknown sample having a different size from the unknown sample F and the second standard sample of the unknown sample E is used, the dimension cannot be measured with sufficient accuracy.
When the determination result in the processing 308 of No. is No, as shown in FIG. 7 again, it is examined whether there is a combination of samples that can be measured at both the observation magnification of the unknown sample and the dimensional measurement magnification of the standard sample. Then, the size of the unknown sample that can be measured with the standard sample is set as the second standard sample (process 312).

The same operation is repeated n times until a standard sample capable of measuring an unknown sample is obtained (Step 313), and the dimensional calibration is performed until the n-th standard sample is finally obtained. Execute and measure the dimensions of the unknown sample each time.
(Processes 314 and 315) When a standard sample from which an unknown sample can be measured is obtained, the process proceeds to Process 304.

The “uncertainty” at this time can also be calculated from the above equation by obtaining the uncertainty up to the sample serving as the n-th standard sample by the above-described method and using this value as U ₁ . Therefore, according to this embodiment, the range that can be calibrated by the standard sample can be sufficiently expanded.

Here, assuming that the allowable range of the uncertainty is 5%, the value of n generally changes depending on the dimensions of the standard samples after the second standard sample. Is appropriate.

In the above embodiment, the case where the unknown sample whose size is to be measured is mounted on the standard micro-scale M as a sample substrate has been described.
If the dimensions or magnification can be calibrated by a combination of a substrate and a sample having a different size, the present invention is not limited to the case where an unknown sample is placed on a standard microscale.

[0058]

According to the present invention, by mounting a sample that can be measured at a magnification different from that of the standard sample on the standard sample, it is possible to easily measure the dimensions of an unknown sample that cannot be calibrated with the standard sample alone. . As a result, factors that lower work efficiency, such as selection processing of a sample that can be an appropriate standard and loading and unloading of a standard sample, can be eliminated, thereby greatly contributing to an improvement in throughput.

In the prior art, it may be difficult to maintain the measurement accuracy due to difficulty in observing the same field of view, a change in the current value, and a human error that occurs when setting the conditions, which may be caused by inserting and removing the sample many times. However, according to the present invention, the number of sample insertions can be suppressed to about 1 to 2 times, and it is not necessary to set sample observation conditions each time, and dimensional calibration can be performed in the same field of view. The uncertainty of the measured value can be reduced, and the time required for measuring an unknown sample can be reduced at the same time as the accuracy of dimensional calibration is improved.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing an example of a standard sample in an embodiment of a method for calibrating an electron microscope according to the present invention.

FIG. 2 is a schematic configuration diagram illustrating an example of a scanning electron microscope to which an embodiment of the present invention is applied.

FIG. 3 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 4 is an explanatory diagram showing an example of an unknown sample placed on a standard sample.

FIG. 5 is an explanatory diagram showing an example of observation of a standard sample at various magnifications.

FIG. 6 is an explanatory diagram showing an example of dimension measurement of an unknown sample according to an embodiment of the present invention.

FIG. 7 is a schematic diagram showing a relationship between magnifications of a standard sample and an unknown sample in one embodiment of the present invention.

[Explanation of symbols]

 A-E Unknown sample composed of particles M Standard micro-scale P Pattern R Reference sample 1 Electron gun 2 Extraction electrode 3 Acceleration electrode 4 Electron beam 5 Condenser lens 6 Objective lens 7 Sample stage 8 Sample stage 9 X direction scanning coil 10 Y direction Scanning coil 11 Secondary electron 12 Detector 13 High voltage application control unit 14 Magnification control unit 15 Lens control unit 16 Image monitor / memory unit 17 Measurement condition control unit 18 Length measurement control unit

Continued on the front page (72) Inventor Mine Nakagawa, 882 Ma, Ichiki, Hitachinaka City, Ibaraki Prefecture Inside Hitachi Science Systems, Ltd. (72) Inventor Kaname Takahashi 882, Ichimo, Hitachinaka-shi, Ibaraki F-term within the Hitachi Measuring Instruments Group (reference) 2G001 AA03 BA07 CA03 FA06 HA13 MA04 SA12 SA30 2G052 AA39 GA35 5C001 AA08 CC04 5C033 FF10

Claims (5)

[Claims] 1. A method of calibrating an electron microscope using a reference sample having a pattern of a known size and using the pattern as a standard sample, the method comprising the steps of: First, one of the plurality of particles is calibrated according to the pattern of the reference sample to obtain a second standard sample.
The other one of the plurality of particles is calibrated with the second standard sample to obtain a third standard sample, and thereafter, the same process is repeated for the remaining particles, whereby the second particle is obtained. To the n-th standard sample, and after including the pattern of the reference sample as the first standard sample in the second to n-th standard samples, any of the first to n-th standard samples A method for calibrating an electron microscope, wherein the method is used to perform calibration. 2. The method according to claim 1, wherein an unknown sample is supported on an observation surface of the reference sample. 3. The invention according to claim 1, wherein the unknown sample has a size outside a calibration range of the first standard sample, and the unknown sample is calibrated by the second to n-th standard samples. A method for calibrating an electron microscope, characterized in that the method is performed by using a method. 4. A standard sample for electron microscope calibration, wherein a plurality of particles having different sizes in a stepwise manner from the known size are dispersed and supported on an observation surface of a reference sample having a pattern of a known size. 5. A method for dispersing and supporting a plurality of particles having different sizes stepwise from the known dimensions on an observation surface of a reference sample having a pattern of a known size, wherein the unknown sample is supported on the observation surface. Characteristic standard sample for electron microscope calibration.