JPH11271459A - Beam measurement method - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To realize a beam measurement method capable of correctly measuring a beam size, a beam position, and a focus state by a curve fitting method in a short time. SOLUTION: In a curve fitting process, a detection signal of a Faraday cup 6 stored in a waveform memory 8 is read out by a control CPU 9, and the control CPU 9
A fitting process between the detection signal and the modeled signal is performed. At this time, a beam position measured in advance is used as an initial value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring a charged particle beam in an apparatus using a charged particle beam, such as an electron beam drawing apparatus and an ion beam apparatus, and a laser beam in a laser beam apparatus.

[0002]

2. Description of the Related Art For example, in an electron beam writing apparatus, prior to an actual writing operation, the size and position of an electron beam used for writing or the focus state of the electron beam are measured, and based on the measurement result, the electron beam is drawn. Has been adjusted. FIG. 1 shows an example of an apparatus used for measuring such an electron beam, and 1 is an electron beam to be measured. Although not shown, the electron beam 1 has a rectangular cross section formed by two rectangular slits and a deflector provided between the two rectangular slits.

An electron beam 1 is focused by a final lens 2 and further deflected by an electrostatic deflector 3.
A knife edge member 4 is arranged below the deflector 3, and the knife edge member 4 is provided with a rectangular opening, and the inside thereof is formed thin and linearly. At the lower part of the knife edge member 4, an aperture 5 for cutting the scattered electron beam is provided.
A Faraday cup 6 for detecting a current amount of the electron beam is provided.

In the above configuration, when a sawtooth-shaped deflection signal shown in FIG.
Are deflected in the X direction. By electron beam deflection,
The electron beam is gradually blocked by the knife edge member 4, and the amount of the electron beam incident on the Faraday cup 6 decreases. When the electron beam 1 is completely shielded by the knife edge member 4, the detection current of the Faraday cup 6 becomes zero.
Becomes

FIG. 2B shows a detection current of the Faraday cup 6. When the detection current signal is differentiated once, a signal shown in FIG. 2C is obtained. Further, when the signal of FIG. 2C is differentiated again, the signal of FIG. 2D is obtained.
In FIG. 2D, the horizontal axis represents the scanning position of the electron beam.
The size of the electron beam is determined based on the distance between the two peaks of the signal. Further, the position of the electron beam is determined based on the intermediate position between the two peak positions. Further, the peak value of the peak indicates the focus state of the electron beam. Various adjustments of the electron beam are performed based on the beam size, the beam position, and the focus state obtained as described above, and then a normal drawing operation is performed.

[0006]

In the measurement of the size of the electron beam and the like, the detection signal shown in FIG. 2B usually contains a noise component. FIG. 3A shows a detection signal waveform of a Faraday cup containing a noise component. When a signal containing such a noise component is differentiated once, a signal shown in FIG. 3B is obtained. The signal resulting from the second differentiation is as shown in FIG. This figure 3
In the signal (c), two peaks are buried in the noise peak, and it becomes impossible to accurately measure the beam size, position, and focus state.

[0007] Therefore, a signal smoothing process is performed. FIG. 4A shows a detection signal containing a noise component, and when this signal is differentiated, a signal shown in FIG. 4B is obtained. At this time, the primary differential signal has been subjected to noise removal smoothing processing. After performing this processing, when differentiation is performed again, a signal shown in FIG. 4C is obtained. In this signal, although the peak position can be obtained correctly, the peak becomes dull due to the smoothing process, so that the peak value of the peak that reflects the focus state of the beam is not correct, and the focus state is substantially correct. It cannot be measured.

For this purpose, a beam is scanned across a member having a linear edge, and a curve fitting is performed on a change in the signal of the beam detected in accordance with the scanning to measure a beam size and a position. It is considered to perform.

FIG. 5 shows an example of a configuration for performing this curve fitting. The same reference numerals as those in the apparatus shown in FIG. 1 denote the same components. With this configuration, the Faraday cup 6
Is converted into a digital signal by the AD converter 7 and then supplied to the waveform memory 8. The signal supplied to and stored in the waveform memory 8 is read out by the control CPU 9 and subjected to a curve fitting process. The control CPU 9 controls a deflection circuit 10 for supplying a scanning signal of the electron beam 1 to the electrostatic deflector 3. The operation of such a configuration will now be described.

When measuring the size, position and focus state of the electron beam 1, the control CPU 9 controls the deflection circuit 10 to apply a sawtooth-shaped deflection signal to the electrostatic deflector 3. With the application of this deflection signal, the rectangular electron beam 1
To be deflected in the direction. Due to the deflection of the electron beam, the electron beam is gradually shielded by the knife edge member 4, and the amount of the electron beam incident on the Faraday cup 6 decreases. When the electron beam 1 is completely shielded by the knife edge member 4, the detection current of the Faraday cup 6 becomes zero.

FIG. 6A shows a detection signal waveform of the Faraday cup 6 obtained by deflecting the electron beam 1, and this detection signal is stored in the waveform memory 8. The detection signal stored in the waveform memory 8 is read out by the control CPU 9, and the first differentiation is performed. Here, the primary differential waveform is modeled in advance as shown in FIG. In FIG. 7, a is a half of the beam size, and b is a beam position.

The basic idea in curve fitting is to perform fitting processing on a waveform obtained by modeling a detection signal waveform. The signal waveform shown in FIG. 6A is first-order differentiated and is shown in FIG. By performing fitting with the modeled signal waveform shown, the signal in FIG. 6B is obtained. The noise component is removed from the signal subjected to the fitting processing, and the signal of FIG. 6C is obtained by further differentiating the signal of FIG. 6B.

The signal shown in FIG. 6C has been subjected to fitting processing, so that noise components have been removed.
Furthermore, since the smoothing processing is not performed, the signal component based on the edge of the knife edge remains clearly without dulling, so that the size, position, and focus state of the electron beam 1 can be accurately measured. .

Next, a more specific fitting process will be described. First, the detection signal of the Faraday cup 6 stored in the waveform memory 8 is read out by the control CPU 9,
First derivative processing is performed. The control CPU 9 performs fitting processing on the primary differential signal. This fitting process is performed using an appropriate evaluation function. For example, if a is １ ／ of the beam size, b is the beam position, and c is the focus information, the following evaluation function can be used.

Fi (a, b, c) = Tanh ｛(ia
−b) / c｝ −Tanh ｛(i + ab) / c｝ where i is the beam scanning position (i = 1, 2,...)
.., N). In the fitting, the parameters a, b, and c are determined such that the sum of squares of the difference between the n data Ai of the primary differential signal and the evaluation function is minimized.
That is, the parameters are determined using the following equation.

[0016]

(Equation 1)

In the above-described example, the detection signal of the Faraday cup 6 is first-order differentiated, and then fitting processing is performed. In this case, the present invention can be applied to a case where the SN ratio of the detection signal is relatively excellent. it can. When the SN ratio of the detection signal of the Faraday cup 6 is relatively poor, the square sum of the difference between the n data Bi (i = 1, 2,..., N) of the waveform memory 8 and the evaluation function Fi

[0018]

(Equation 2)

Parameters a, b, c so that
Can be determined. In this case, as the evaluation function, for example, the following function using the nonlinear least squares method can be used.

Fi (a, b, c) = Log [Cosh
{(I + ab) / c}]-Log [Cosh} (-i
+ A + b) / c}] The above is the measurement of the beam in the X direction, but the measurement of the beam in the Y direction is performed in the same manner. After measuring the beam size, the beam position, and the focus information of the electron beam in this way, the adjustment of the beam size, the correction of the beam position, and the adjustment of the focus are performed, and then the normal drawing operation is started.

In the above-described curve fitting method, a non-linear least squares method is used. That is, the parameters of the signal waveform model are adjusted so that the sum of squares of the residual between the signal waveform model and the measurement data is minimized.
Therefore, when the initial value of the parameter deviates from the true value, it takes a long time to calculate. Particularly, regarding the beam position, the position tends to change due to deflection distortion and beam drift, which has been a problem.

The present invention has been made in view of such a point, and an object of the present invention is to provide a beam measuring method capable of correctly measuring a beam size, a beam position, and a focus state by a curve fitting method in a short time. To be realized.

[0023]

According to a first aspect of the present invention, there is provided a beam measuring method for scanning a beam across a member having a linear edge, and detecting a signal of the beam detected along with the scanning. In a beam measurement method in which a curve fitting is performed for the change to measure at least one of the beam size, the beam position, and the focus information of the charged particle beam, the beam position is measured in advance, and the measured beam position is measured. Is set as an initial value to perform a curve fitting process.

According to the first aspect of the present invention, the position of the beam is measured in advance, and a curve fitting process is performed using the measured beam position as an initial value, and each information is accurately measured in a short time without the influence of noise.

According to a second aspect of the present invention, a charged particle beam is scanned across a member having a linear edge, and a signal of the charged particle beam detected in accordance with the scanning is first differentiated. Then, n pieces of data Ai (i is a scanning position, i = 1, 2,..., N) of the primary differential signal and a
Is the beam size, b is the beam position, and c is the focus information. The following evaluation function Fi (a, b, c) = Tanh {(iab) / c}
−Tanh {(i + ab) / c}, and a beam measurement method in which parameters a, b, and c are determined such that the sum of squares of the difference between the data Ai and the evaluation function Fi is minimized. Is characterized in that the beam position is measured in advance, and the parameters a, b, and c are determined using the measured beam position as an initial value.

According to the second aspect of the present invention, the parameters a (1/2 of the beam size) and b are set so that the sum of squares of the difference between the n data Ai of the primary differential signal and the evaluation function Fi is minimized.
In determining (beam position) and c (focus information), the beam position is measured in advance, and parameters a, b, and c are determined using the measured beam position as an initial value.

[0027]

Embodiments of the present invention will be described below in detail with reference to the drawings. The beam measurement according to the present invention is performed, for example, by the configuration shown in FIG. 8, and the same parts as those in FIG. 5 are denoted by the same reference numerals. That is,
When measuring the size, position, and focus state of the electron beam 1, the control CPU 9 controls the deflection circuit 10 and applies a sawtooth-shaped deflection signal to the electrostatic deflector 3.

With the application of the deflection signal, the rectangular electron beam 1 is deflected in the X direction. Due to the deflection of the electron beam, the electron beam is gradually shielded by the knife edge member 4, and the amount of the electron beam incident on the Faraday cup 6 decreases. When the electron beam 1 is completely shielded by the knife edge member 4, the detection current of the Faraday cup 6 becomes zero.

The detection signal of the Faraday cup 6 is converted into a digital signal by an AD converter 7 and then recorded in a waveform memory 8. After that, the control CPU 9 reads out the waveform memory data from the waveform memory 8, obtains the approximate beam position as a pre-process of the curve fitting, and performs the curve fitting process. The obtained approximate beam position is stored in the memory 15.

Curve fitting (curve fitting by the nonlinear least squares method) is performed using the obtained beam position, the beam size specified as the initial parameter, and the value of the edge sharpness as initial values. From the result of this fitting, the beam size, beam position, and focus information (edge sharpness) can be accurately obtained in a short time.

Here, the measurement of the beam position as a pre-process of the curve fitting and the subsequent curve fitting process will be described in more detail. To measure the beam position, the beam is scanned once across a specific edge of the knife edge 4. By this scanning, the Faraday cup 6
Ai (i = 1, 2, 3,...)
n) is stored in the waveform memory 8.

Based on the n pieces of data Ai stored in the waveform memory 8, the control CPU 9 obtains a position bmid which is an intermediate value between the maximum value (or initial value) and the minimum value (or final value) of the data. 15 is stored. FIG. 9 shows the signal Ai. FIG. 9A shows the case where the beam position is almost at the center, and FIG. 9B shows the case where the beam position is shifted.

Since the amount of calculation for obtaining the intermediate position bmid is small, the approximate position of the beam can be obtained at high speed. The curve fitting process uses the nonlinear least-squares method to obtain n data in the waveform memory, Ai = 1, 2, 3,..., N, and the signal waveform model formula Fi (a, b, c) = Log [Cosh ｛(i + a-
b) / c｝]-Log [Cosh ｛(-i + a + b) /
c}] i = 1,2,3,..., n
b and c are determined. That is, the parameters are determined using the following equation.

[0034]

(Equation 3)

In the above description, a is 1/1 of the beam size.
2, b represents the beam position, and c represents the focus information (edge sharpness). Also, bmid uses bmid, and a and c use specified values as initial values of parameters.

Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment. For example, the input signal S
If / N is excellent, fitting may be performed after the differentiation processing. The following equation can be used as an example of the signal waveform model in that case.

Fi (a, b, c) = Tanh ｛(x−a
−b) / c｝ −Tanh ｛(x + ab) / c｝ x = 1, 2, 3,..., N Further, the fitting calculation processing is not performed by the control CPU to improve the speed, but is performed separately. May be used. Further, the evaluation function is not limited to the above equation as long as it can represent the beam size, the beam position, and the focus information. If the evaluation function is changed, the present invention can be applied to not only a rectangular beam but also a spot beam.

Although the above embodiment has been described using an electron beam, the present invention can be applied to a laser beam device of the ion beam device 7. Further, the amount of the electron beam which is shielded by the knife edge member 4 and enters the Faraday cup 6 is detected. A material having a linear edge and a high emission coefficient of secondary electrons and reflected electrons is used. A configuration may be employed in which a charged particle beam is scanned across the material to detect secondary electrons and reflected electrons from the material.

[0039]

As described above, according to the first aspect of the present invention, the position of the beam is measured in advance, and the curve fitting process is performed using the measured beam position as an initial value. Each information can be measured accurately without the influence of. In addition, since the detection signal is not smoothed, focus information can be measured more accurately.

According to the second aspect of the present invention, the parameters a (1/2 of the beam size) and b are set so that the sum of squares of the difference between the n data Ai of the primary differential signal and the evaluation function Fi is minimized.
In determining (beam position) and c (focus information), the position of the beam is measured in advance, and parameters a, b, and c are determined using the measured beam position as an initial value. Each information can be measured accurately without the influence of noise. In addition, since the detection signal is not smoothed, focus information can be measured more accurately.

[Brief description of the drawings]

FIG. 1 is a diagram showing an example of an apparatus used for conventional electron beam measurement.

FIG. 2 is a waveform chart for explaining basic signal processing for electron beam measurement.

FIG. 3 is a diagram showing various waveforms obtained by conventional signal processing.

FIG. 4 is a diagram showing various waveforms by conventional signal processing accompanied by smoothing processing.

FIG. 5 is a diagram showing an example of an apparatus for performing curve fitting.

FIG. 6 is a diagram showing various waveforms obtained by signal processing accompanied by fitting processing.

FIG. 7 is a diagram showing a modeled first derivative signal.

FIG. 8 is a diagram showing an example of an apparatus for implementing the present invention.

FIG. 9 is a diagram showing a detection signal waveform by beam scanning for measuring a beam position.

[Explanation of symbols]

 Reference Signs List 1 electron beam 2 final stage lens 3 electrostatic deflector 4 knife edge member 5 aperture 6 Faraday cup 7 AD converter 8 waveform memory 9 control CPU 10 deflection circuit 15 memory

Claims (2)

[Claims] A beam is scanned across a member having a linear edge, and a curve change is applied to a change in a signal of the beam detected in accordance with the scanning, and a beam size and a beam size of the charged particle beam are adjusted. Position, in the beam measurement method to measure at least one of the focus information, the position of the beam is measured in advance,
A beam measurement method, wherein a curve fitting process is performed using a measured beam position as an initial value. 2. A charged particle beam is scanned across a member having a linear edge, and a signal of the charged particle beam detected in accordance with the scanning is first-differentiated.
Data Ai (i is a scanning position, i = 1, 2,...,
n), a is １ ／ of the beam size, b is the beam position,
The next evaluation function where c is focus information Fi (a, b, c) = Tanh {(iab) / c}
−Tanh {(i + ab) / c}, and a beam measurement method in which parameters a, b, and c are determined such that the sum of squares of the difference between the data Ai and the evaluation function Fi is minimized. 3. The beam measuring method according to 2, wherein the beam position is measured in advance, and parameters a, b, and c are determined using the measured beam position as an initial value.