JP2004266182A - Electron beam lithography system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an electron beam lithography system which can minimize fogging at the periphery of an alignment mark caused by beam irradiation upon detection of the positioning mark or the pollution of the mark caused by resist removal during alignment of the electron beam lithography system. <P>SOLUTION: In beam irradiation conditions during the detection of the alignment mark, a necessary accuracy can be secured by evaluating a detection reproduction accuracy using the actual alignment mark and can be realized by optimizing a necessary minimum quantity of irradiation. This can be attained by providing the algorithm of optimizing the alignment mark detection conditions and a unit for controlling the alignment-mark detection conditions as its implementation means. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an electron beam lithography apparatus, and in particular, to a fine pattern forming method that needs to draw a pattern of an electronic circuit on a sample surface with high precision, the position of a pattern to be drawn and a pattern previously formed on a wafer substrate. When performing alignment, so-called alignment mark detection for wafer position detection during alignment, with the same beam as drawing, resist exposure due to alignment mark detection, which was a disadvantage, is minimized and highly accurate. The present invention relates to an alignment mark detection method that enables pattern drawing.
[0002]
[Prior art]
In recent years, semiconductor devices have surpassed the competition for prototypes of devices having a size of 100 nm or less. As the manufacturing apparatus, there are extreme ultraviolet lithography (EUV) and electron projection lithography (EPL), which is called next generation lithography technology NGL (Next Generation Lithography).
[0003]
A representative candidate for NGL is an electron beam lithography system. The electron beam lithography system forms an electron beam using an electron lens and a shaping aperture, controls the beam to an arbitrary size, and prints a pattern directly on a wafer. Unlike an ordinary exposure apparatus, a mask serving as an original image is not required. Therefore, the range of use of the apparatus has attracted attention because it covers many generations from the trial production of fine patterns to the mass production of maskless QTAT devices.
[0004]
An electron beam lithography apparatus currently in practical use will be described with reference to FIG. The electron beam 9 emitted from the electron source 1 passes through the first mask 2, passes through the molded lens 4, and reaches the second mask 6. When irradiating the second mask 6, a molding dimension is designated to the molding deflection control circuit 15 by the control computer 22 in the molding deflector 3, a voltage is applied to the molding deflector 3 by the molding deflection control circuit, and the designated dimension is set on the wafer 12. Such a rectangular electron beam transmits through the second mask 6. The electron beam 9 transmitted through the second mask is reduced by the reduction lens 7, and the deflection signal specified by the control computer 22 is set in the positioning deflection control circuit 16 to set a deflection signal in the positioning deflector 8. The electron beam 9 is deflected to a designated position and is irradiated on the wafer 12 on the sample stage 13 through the objective lens 10.
[0005]
The sample stage 13 is accurately positioned by the control computer 22 by the sample stage positioning mechanism 14, the sample stage position measuring circuit 20, and the sample stage position control circuit 18. Although FIG. 9 shows only one axis, it actually has at least two axes, and the X direction, the Y direction, and the yawing component are accurately measured and positioned. By controlling the electron beam 9 and the sample stage 13 as described above, a desired drawing pattern is drawn on the chips 31 arranged on the wafer 12.
[0006]
Next, the procedure of alignment will be described. An alignment mark 19 is formed in advance in a chip 31 on the wafer 12, and the reflected electrons 23 of the electron beam 9 radiated on the alignment mark 19 are reflected by the reflected-electron detector 11 for alignment mark detection. The position of the alignment mark on the sample table 13 is measured by the detection and signal processing by the alignment mark detection circuit 17, and the control computer 22 obtains accurate position information of the wafer 12.
[0007]
FIG. 2 shows a specific alignment execution procedure. FIG. 2 is an enlarged view of one chip 31 on the wafer 12, and alignment positioning marks 19 a to 19 d are formed in advance at four corners of the chip 31. In the alignment, first, as shown in FIG. 2-1, one of the XY alignment marks 19a (the X direction in FIG. 2) is scanned with the beam 9 to obtain the X direction position information.
[0008]
Next, as shown in FIG. 2-2, scanning is performed in the other direction (Y direction in FIG. 2) to obtain Y direction position information. Thus, XY position information of the alignment mark 19a is obtained. In the same manner, the position coordinates of the alignment marks 19b, 19c, and 19d are calculated, and finally the coordinates and distortion of the base pattern 30 formed on the chip are calculated, and the position to be drawn is accurately obtained. . Based on the obtained beam drawing coordinates, a beam is sequentially irradiated as shown in FIG. 2-3, and drawing is performed in a state where a desired drawing pattern is accurately aligned with the base pattern 30.
[0009]
Here, the specific operation of the alignment is described in Japanese Patent Application Laid-Open No. 08-274003 (Patent Document 1). As described in Japanese Unexamined Patent Publication No. 59-723134 (Patent Document 2), there is one in which signal processing is devised.
[0010]
However, none of the alignment mark irradiation conditions is directed to at least reducing the irradiation amount. According to the prior art, the condition for irradiating the alignment mark 19 for electron beam irradiation for alignment is executed under the condition registered in the memory file of the control computer 22 in advance. The above-mentioned registration condition is registered in advance under the condition whose accuracy has been confirmed before the lot processing. However, in actuality, no consideration is given to minimizing the dose amount for alignment in order to give consideration to necessary accuracy.
[0011]
The problem on the device due to the dose (photosensitivity) of the alignment mark part in the alignment is the effect of process contamination and fogging near the alignment mark. Since the alignment mark detection is performed using the same electron beam as the drawing, the alignment mark portion and its peripheral portion are exposed by the alignment operation.
[0012]
Usually, the alignment mark is provided in an area to be finally cut called a scribe of the device pattern, but is not limited thereto. In some cases, there is an alignment mark in the vicinity of the circuit pattern. In this case, the exposure near the alignment mark becomes the same as the pre-dose before drawing, and this becomes an error factor for originally desired dimensional accuracy in drawing a fine pattern. Even when the alignment mark is in the scribe area, if a resist residue or the like occurs in the photosensitive portion, there is a case where a problem is caused because contamination occurs in a post-process.
[0013]
In order to solve this problem, there is off-axis alignment (Off-Axis) in which alignment mark detection is performed using a beam or light having a different wavelength from drawing. In this case, a wavelength at which the resist is not exposed is selected. The problem with Off-Axis is that the distance between the axis of the writing beam and the axis of the alignment beam (base line) must always be kept constant because of the influence on the alignment accuracy. In fact, Off-Axis requires complicated control including temperature management, and although it has been used in steppers, it has not yet been put to practical use in an electron beam lithography apparatus.
[0014]
[Patent Document 1]
JP 08-274003 A [Patent Document 2]
JP-A-59-723134
[Problems to be solved by the invention]
The TTL on-axis alignment method (TTL On-Axis), in which alignment is performed by the same beam as writing, is a very effective method as a high-precision alignment means because it does not require baseline management like the above-described Off-Axis. The problem of the resist exposure accompanying the detection of the alignment mark can be improved or solved by controlling the dose (the amount of exposure) at the time of detecting the alignment mark to a minimum.
[0016]
When alignment (positioning mark detection) is performed using the same light or electron beam as in drawing, it is practically unavoidable that the irradiated portion (near the positioning mark) is exposed when the positioning mark is detected.
[0017]
SUMMARY OF THE INVENTION It is an object of the present invention to provide an electron beam lithography apparatus which has good detection accuracy of an alignment mark and has a small amount of light exposure near the alignment mark.
[0018]
[Means for Solving the Problems]
The present invention irradiates a focused beam on an alignment mark on a sample surface, which is previously formed in the vicinity of a base area of a pattern including an electronic circuit for drawing with a focused beam, and captures the reflected focused beam. In an electron beam lithography system equipped with an alignment function that measures the position of the sample surface, the photosensitive mark that can accurately detect and reproduce the alignment mark calculated from the captured focused beam and can be positioned near the alignment mark by irradiating the focused beam The method is characterized in that a focused beam is irradiated at the time of detecting the alignment mark under the condition that the exposure amount becomes small.
[0019]
As a result, alignment can be performed with the minimum necessary irradiation.
[0020]
Another object of the present invention is to determine the minimum dose required for alignment and to automate the determination. To achieve this, the alignment accuracy, that is, the alignment mark position detection accuracy is evaluated using the actual alignment mark on the wafer, and the parameters for determining the irradiation amount are changed to directly evaluate the detection accuracy of the alignment mark position. The solution can be solved by finding the condition with the minimum irradiation amount while securing the desired accuracy from the evaluation result. Automation can be solved by developing software or hardware that automatically controls the above-mentioned judgment algorithm.
[0021]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 is a schematic view of an electron beam writing apparatus for realizing the present invention.
The basic operations of alignment and drawing are the same as in the conventional example described with reference to FIG. An embodiment of the automatic control of the alignment mark detection condition for alignment according to the present invention will be described in detail below.
[0022]
When the first wafer 12 at the beginning of the lot process is set on the sample stage 13, first, the wafer 12 is moved to the position of the first alignment mark 19. Further, the alignment mark detection condition control unit 21 is activated by a command from the control computer 22, and automatically controls the alignment operation thereafter.
[0023]
FIG. 3 shows actual alignment positioning marks and specific beam irradiation conditions. As the alignment mark 19, a cross alignment mark or a cross position alignment mark is usually used to obtain positional information in the X and Y directions. This figure shows an example of a cross alignment mark, but the shape of the alignment mark is not particularly limited.
[0024]
Main parameters and contents at the time of detecting the alignment mark will be described with reference to FIG. The shot size, which is the size of the focused beam, is specified by width W and height h. The specified shot moves as needed at the size specified by the shot pitch SP and moves within the area specified by the full length scan width SW. The shot while moving in the horizontal direction (X direction) is called a shot scan.
[0025]
Assuming that the above-described operation is one-shot scan, the amount specified by the line pitch LP is then moved in the vertical direction, and the next shot scan is performed. The line pitch LP is a line feed interval which is a vertical movement.
[0026]
This shot scan is repeated for each line feed in the vertical direction to read the alignment mark. The alignment mark of the focused beam reflected by the shot scan is captured, and the position of the data surface is measured (calculated) using the read information of the alignment mark as a parameter.
[0027]
The line pitch LP (line feed interval) can of course be set even when LP = 0. The above-described scanning operation is repeated the number of times specified by the number of scans SN, and finally the alignment mark detection signals obtained in all the scans are combined to obtain the alignment mark detection signal as shown in FIG. Since the obtained signal has position information on the horizontal axis and detection signal intensity on the vertical axis, predetermined signal processing is performed to accurately calculate the position coordinates of the alignment mark.
[0028]
Next, with respect to the above parameters, an algorithm for securing desired alignment accuracy with a minimum necessary dose (irradiation amount), which is an object of the present invention, will be described with reference to FIG.
[0029]
First, each parameter is initialized 50. In FIG. 5, among the parameters, (1) shot size and (2) scan width are fixed values. (3) The shot pitch is set to an initial constant SP ', (4) The line pitch is set to an initial constant LP', and (5) The number of scans is set to an initial constant SN '. Each initial constant is registered in advance in a file of the control computer or the like, and can be arbitrarily changed. Next, (3) optimization 60 of the shot pitch is performed.
[0030]
The optimization flow of the shot pitch will be described with reference to FIG. According to Box 61, the items other than (3) shot pitch SP 'fix the above initial constant, and (3) SP' value is used as an initial value as a variable for shot pitch. Next, the number of evaluation conditions n is set to 0 as in box 62, and the alignment mark detection is repeatedly executed in box 63. The number of times of execution is specified by N times, but is usually about 30 times. Next, the detection reproducibility is confirmed by the box 64, and the result is statistically processed.
[0031]
Next, it is determined whether all the evaluation conditions specified in advance in box 65 have been completed. If not completed, the number of evaluation conditions is set to n = n + 1 in box 66, and the shot pitch SP 'is added by the specified pitch ΔSP in box 67. I do. Thereafter, processing is performed in the same manner as n = 0, and finally all evaluation conditions are completed in box 65. From the reproducibility accuracy results under all evaluation conditions obtained in Box 68, the SP 'conditions with the best reproducibility are determined by summarizing as shown in the graph, and the optimum solution SP is obtained.
[0032]
Next, (4) line pitch optimization 70 in FIG. 5 will be described with reference to FIG. According to the Box 71, constants are fixed for all items other than (4) line pitch LP '. Of course, the SP value optimized by the box 60 is set for the above-mentioned (3) shot pitch. Thereafter, the procedure proceeds in substantially the same manner as the shot pitch optimization. The box 72 sets the number of evaluation conditions n to 0, and the box 73 repeatedly executes the alignment mark detection. The number of times of execution is specified by N times, but is usually about 30 times.
[0033]
Next, the detection reproduction accuracy is checked in box 74, and the result is statistically processed. Next, it is determined in box 75 whether or not all the evaluation conditions specified in advance have been completed. If not completed, the number of evaluation conditions is set to n = n + 1 in box 76, and the line pitch LP ′ is added in box 77 by the specified pitch ΔLP. I do. Thereafter, the processing is performed in the same manner as n = 0, and finally all evaluation conditions are completed in box 75. From the reproducibility accuracy results under all evaluation conditions obtained in Box 78, the results are summarized as shown in the graph, and the LP 'condition with the best reproducibility is determined to find the optimal solution LP.
[0034]
Next, the (5) scan number optimization 80 in FIG. 5 will be described with reference to FIG. According to Box 81, constants are fixed for items other than (5) the number of scans SN ', and constants SP and LP values optimized as described above are set for (3) shot pitch and (4) line pitch. Next, the number of evaluation conditions n is set to 0 as in box 82, and the alignment mark detection is repeatedly executed in box 83. The number of times of execution is specified by N times, but is usually about 30 times. Next, the detection reproduction accuracy is checked in box 84, and the result is statistically processed.
[0035]
Next, it is determined whether all the evaluation conditions specified in advance are completed in box 85. If not completed, the number of evaluation conditions is set to n = n + 1 by box 86, and the number of scans SN ′ is added by box 87 by the specified pitch ΔSN. I do. Thereafter, the processing is performed in the same manner as n = 0, and finally all evaluation conditions are completed in box 85. Based on the reproducibility results under all evaluation conditions obtained in Box 88, the results are summarized as shown in the graph.
[0036]
It is known that the number of scans, which is greatly related to the dose to the resist, is advantageous in terms of accuracy by increasing the number of scans, so optimization means finding the minimum number of scans required to satisfy the desired accuracy. Is in the thing. From the results obtained as shown in the graph shown in Box 88, the condition that ensures the desired accuracy and minimizes the number of scans SN ′ is determined as the optimum solution SN. Thus, the optimization of the alignment mark detection condition is completed. In other words, under the condition that the desired accuracy with good reproduction accuracy is satisfied, the exposure amount of the photosensitive portion formed near the alignment mark is small and the actual scan value is obtained, thereby optimizing the alignment mark detection condition. Can be found.
[0037]
It should be noted that there is no particular bottleneck in automatically executing the above algorithm by a dedicated control unit.
[0038]
In the automatic optimization process of the detection condition of the present invention, specifically, each parameter is automatically calculated in the first ware of the lot process (the condition can also be used for a wafer), and the obtained optimized condition is obtained. Usually, lot processing is performed.
[0039]
Further, in the present description, the fixed shot size and scan width can be optimized by a similar algorithm. The optimized parameters are automatically registered in the control computer 22, and the alignment operation is performed under the optimized conditions for detecting the alignment marks of the second and subsequent wafers of the first wafer of the lot and for the treatment of the second and subsequent wafers of the lot. I do.
[0040]
Also, it is of course possible to start the alignment mark detection condition control unit 21 from the control computer 22 at an arbitrary timing to execute the optimization. It is also possible to confirm the current alignment accuracy using the alignment mark detection condition control unit 21. By displaying the results of the optimization process on the screen, the state can be confirmed.
[0041]
According to the above-described embodiment of the present invention, the amount of exposure (dose) to the resist, which is a photosensitive agent, can be controlled to a necessary minimum when detecting an alignment mark for alignment. Line width fluctuation and process contamination due to fogging around the alignment mark caused by exposure to the alignment mark can be prevented. It is also useful as a tool for confirming the accuracy of the alignment mark detection, and the device can always be used in a normal state, thereby improving the reliability of the device. That is, the electron beam lithography apparatus can always be operated with stable performance, and the yield in device manufacturing can be improved.
[0042]
【The invention's effect】
As described above, according to the present invention, it is possible to minimize the light exposure at the peripheral portion (near) of the alignment mark.
[Brief description of the drawings]
FIG. 1 is an electron beam drawing apparatus according to an embodiment of the present invention.
FIG. 2 is an enlarged view of the inside of a chip showing an alignment execution procedure.
FIG. 3 is an explanatory diagram of specific detection conditions and contents of alignment mark detection according to the embodiment of the present invention.
FIG. 4 is a signal waveform diagram showing a result of detecting an alignment mark according to the embodiment of the present invention.
FIG. 5 is a flowchart according to the embodiment of the present invention, showing a main requirement, that is, optimization of alignment mark detection conditions.
FIG. 6 relates to an embodiment of the present invention, and is a flow for obtaining an optimum value of a shot pitch SP which is a main element.
FIG. 7 relates to the embodiment of the present invention, and is a flow for obtaining an optimum value of a line pitch LP which is a main element.
FIG. 8 is a flowchart according to the embodiment of the present invention, which finds an optimum value of the number of scans SN, which is a main element.
FIG. 9 is an electron beam drawing apparatus according to a conventional example.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Electron source, 2 ... rectangular shaping diaphragm, 3 ... shaping deflector, 4 ... shaping lens, 5 ... blanker, 6 ... variable shaping aperture, 7 ... reduction lens, 8 ... positioning deflector, 9 ... electron beam, 10 ... Objective lens, 11: reflection electron detector for detecting alignment marks, 12: wafer, 13: sample stage, 14: sample stage positioning mechanism, 15: shaping deflection control circuit, 16: positioning deflection control circuit, 17: alignment mark Detection circuit, 18: sample stage position control circuit, 19: alignment mark, 20: sample stage position measurement circuit, 21: alignment mark detection condition control unit, 22: control computer, 23: backscattered electron, 30: ground Pattern, 31 ... chip, 50-88 ... Algorithm flow for explaining the optimization of the alignment mark detection condition, 60-68 ... Flow for finding the optimal value of shot pitch, 70- 8 ... flow for determining the optimum value of the line pitch, the flow for determining the optimum value of 80 to 88 ... scan number.

Claims (4)

An electron source for generating an electron beam, an electron optical system for focusing the electron beam, and guiding the focused beam to a sample surface such as a semiconductor wafer, and deflecting the focused beam to the sample surface. A deflection control system for applying the focused beam, and irradiating the focused beam to an alignment mark on the sample surface, which is previously applied to the vicinity of a base area of a pattern including an electronic circuit for writing with the focused beam, and reflects the focused beam. In an electron beam lithography apparatus equipped with an alignment function for measuring the position of the sample surface by capturing the incoming focused beam,
Under the conditions that the detection and reproduction accuracy of the alignment mark calculated by the captured focused beam is high, and the exposure amount of the photosensitive portion formed near the alignment mark by irradiation of the focused beam is small, the alignment mark is An electron beam lithography apparatus for irradiating a focused beam upon detection. An electron source that generates an electron beam, an electron optical system that focuses the electron beam, and guides the focused focused beam to irradiate a sample surface such as a semiconductor wafer, and deflects the focused beam that irradiates the sample surface. A deflection control system for applying the focused beam, and irradiating the focused beam to an alignment mark on the sample surface, which is previously applied to the vicinity of a base area of a pattern including an electronic circuit for writing with the focused beam, and reflects the focused beam. In a focused beam irradiation method of an electron beam lithography apparatus equipped with an alignment function for capturing a focused beam and reading a position of an alignment mark and measuring a position of the sample surface based on the read information,
The reading of the alignment mark is performed by repeatedly performing a shot scan of a focused beam irradiated while moving in a horizontal direction on the alignment mark, every vertical line feed,
In the shot pitch, which is the interval between shots of the shot scan, an optimum shot pitch value at which the detection and reproduction accuracy of the alignment mark is best is obtained, and the obtained shot pitch value is maintained while maintaining the obtained shot pitch value. From the line pitch, which is the line feed interval, to determine the optimal line pitch value at which the detection and reproduction accuracy of the alignment mark is the best,
Under the condition that the obtained line pitch value is satisfied, the line pitch value that has already been obtained is satisfied from among the number of scanning lines that is the number of line feeds in the vertical direction, and alignment is performed by irradiating a focused beam. The light exposure of the photosensitive area that can be formed near the mark is small.
A focused beam irradiation method for an electron beam lithography apparatus, comprising: irradiating a focused beam at the time of detecting an alignment mark under the conditions of the obtained actual scan value. A focused beam irradiation method for an electron beam lithography apparatus according to claim 2,
The focused beam irradiation method, wherein the scan pitch value, the line pitch value, and the actual scan value are obtained by repeatedly overlapping focused beam shots. A focused beam irradiation method for an electron beam lithography apparatus according to claim 2 or 3,
The determination of the line pitch value, the line pitch value, and the actual scan value regarding the detection reproducibility of the alignment mark is performed at the beginning of each time a lot of a product such as a semiconductor wafer to be drawn is changed. A focused beam irradiation method for an electron beam lithography apparatus, comprising:

Images