JP3453004B2 - Charged particle beam exposure method and apparatus - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a charged particle beam exposure method and apparatus, and more particularly to shaping a charged particle beam into a rectangular shape of an arbitrary size or an arbitrary shape and converging and deflecting it by using a deflector or an electromagnetic lens. Thus, the present invention relates to a charged particle beam exposure method and apparatus for forming a fine pattern on a sample.

With the recent increase in the density of integrated circuits, the exposure technique for forming a pattern on a semiconductor wafer has changed from photolithography technique, which has been the mainstream for many years, to charged particle beam exposure technique using a charged particle beam such as an electron beam. It is transitioning. The charged particle beam exposure technique includes a variable rectangular exposure technique and a block exposure technique depending on the pattern shape that can be generated at one time. According to the block exposure technique, a mask having a repeating basic unit pattern transmits the charged particle beam, and even if the pattern is complicated, the unit pattern is
Expose with a shot. Therefore, the block exposure technique is effective for the repetition of the basic pattern, such as a 256 Mbit DRAM pattern, which has a small but most exposed area. In addition, recently, a pair of electrodes is provided on each of the apertures arranged in a matrix on the blanking aperture array (BAA), and the voltage applied to the electrodes is turned on / off to obtain it through each aperture. An exposure technique called BAA for forming a pattern by turning on / off an electron beam has also been proposed.

[0003]

22 is a schematic diagram showing an example of a conventional electron beam exposure apparatus using block exposure. The electron beam exposure apparatus includes an electron gun 101, an electron lens system L1a,
Rectangular aperture plate 102, electron lens system L1b, beam shaping deflector 103, first mask deflector MD1, dynamic mask stigmator DS, second mask deflector MD
2, a dynamic mask focus coil DF, an electron lens system L2a, a mask stage 105 on which the block mask 104 is mounted, an electron lens system L2b, a third mask deflector MD3, a blanking deflector 106, a fourth mask deflector MD4, Reduction electron lens system L3, circular aperture 107 functioning as a diaphragm, projection electron lens system L4, main deflector (electromagnetic deflector) 108, sub deflector (electrostatic deflector)
109, a projection electron lens system L5, a wafer stage 111 on which the wafer 110 is mounted, and a control system. The control system includes a central control unit (CPU) 121, a clock unit 122 that generates various clocks including an exposure clock, a buffer memory 123, a control unit 124, a data correction unit 125, a mask memory 126, and a main deflector setting unit 127. . A CPU 121 for controlling the operation of the entire electron beam exposure apparatus, and a clock unit 122.
, The mask memory 126, and the main deflector setting unit 12
7 is connected via a bus 128. Incidentally, FIG.
In FIG. 2, for convenience, the data correction unit 125 and the main deflector setting unit 127 are illustrated as including the functions of a digital-analog converter and an amplifier. Further, the laser interferometer for measuring the position of the wafer stage 111 and the stage moving mechanism for moving the wafer stage 111 are
Since they are known from U.S. Pat. No. 5,173,582 and U.S. Pat. No. 5,194,741, respectively, their illustration and description will be omitted.

The electron beam emitted from the electron gun 10 is
The first and second deflectors MD pass through the rectangular aperture plate 102.
1, MD2, and it passes through a desired pattern portion on the block mask 104. The electron beam having a patterned cross-sectional shape is returned to the optical axis by the converging action of the electron lens systems L2a and L2b and the deflecting actions of the third and fourth deflectors MD3 and MD4. After that, the cross section of the electron beam is reduced by the reduction electron lens system L3, and the electron beam is irradiated on the wafer 110 by the projection electron lens systems L4 and L5.
The desired pattern is exposed.

The buffer memory 123 contains the wafer 110.
The exposure pattern data regarding the pattern to be exposed, the block data regarding the mask pattern on the block mask 104, and the like are stored. The exposure pattern data also includes main deflection data supplied to the main deflector 108. Further, the mask memory 126 stores the relationship between the mask pattern position and the deflection data measured in advance before the exposure, the correction data for correcting the deflection data supplied to the dynamic mask stigmator DS and the dynamic mask focus coil DF, and the like. It is stored.

The exposure pattern data fetched by the CPU 121 and stored in the buffer memory 123 includes a pattern data code PDC indicating which mask pattern on the block mask 104 is used for exposure. The control unit 124 uses the pattern data code PDC to output deflection data for deflecting the electron beam to the position of the mask pattern to be used in the mask memory 1.
It is read out from 26 and supplied to the first to fourth deflectors MD1 to MD4 for pattern selection. Also, the mask memory 12
The deflection data read from 6 is also supplied to the data correction unit 125. The deflection data is read from the mask memory 126 based on the exposure clock generated by the clock unit 122.

On the other hand, the main deflector setting unit 127 reads the main deflection data of the main deflector 108 from the buffer memory 123 based on the clock from the clock unit 122 and supplies it to the main deflector 108. In addition, the sub deflector 1
The deflection data of 09, the deflection data of the beam shaping deflector 103, and the deflection data of the blanking deflector 106 are decomposed into shot data by the control unit 124 according to the data stored in the buffer memory 123, and the data correction unit 125. It is supplied to the corresponding sub-deflector 109, beam shaping deflector 103 and blanking deflector 106 via. That is, the control unit 124 obtains the size of the electron beam and the deflection position of the electron beam on the block mask 104 in the case of performing the variable rectangular exposure according to the data stored in the buffer memory 123, and supplies the electron correction size to the data correction unit 125. To do. Data correction unit 125
Corrects each deflection data of the electron beam according to the pattern to be exposed based on the correction data read from the mask memory 126. The deflection data of the beam shaping deflector 103 determines the variable rectangular size of the electron beam, and the deflection data of the blanking deflector 106 is set for each exposure shot.

FIG. 23 is a diagram showing an example of the block mask 104. As shown in FIG. 9A, the block mask 104 includes a substrate 104a made of a semiconductor such as silicon or a metal, and a plurality of deflection areas 104-1 to 104-12 provided on the substrate 104a. . A plurality of mask patterns are formed in each of the deflection areas 104-1 to 104-12. In an electron beam exposure apparatus that uses block exposure, the range of mask patterns that can be selected by deflecting an electron beam around a position of a certain mask stage 105 is determined, and each deflection area 104-1 to
104-12 is a range of, for example, 5 mm □ corresponding to the range of this selectable pattern. For example, the deflection area 10
When the mask pattern in 4-8 is selected and the pattern is exposed, the mask stage 105 is moved so that the electron optical axis of the electron beam exposure apparatus substantially coincides with the center of the deflection area 104-8.

FIG. 23B shows the structure of the deflection area 104-8. The number of block patterns that can be arranged in the deflection area 104-8 is, for example, 48, and each of them is recognized by the pattern data code PDC. That is, the pattern data code PDC is a mark for reading the contents of the mask memory 126 corresponding to each mask pattern based on the exposure clock from the clock unit 122, which has a maximum of 10 MHz, for example. As described above, the mask memory 126 stores the relationship between the mask pattern position and the deflection data for deflecting the electron beam to each mask pattern position, the correction data supplied to the dynamic mask stigmator DS and the dynamic mask focus coil DF, and the like. It is stored. These data are stored in the mask memory 126 by adjusting the electron beam in advance before the exposure and obtaining the deflection data and the correction data for the deflection area to be used.

In order to perform higher-speed exposure, it is necessary to reduce the shot waiting time, that is, the time required to deflect the electron beam. Although not shown in FIG. 22, the deflection data actually supplied to the main deflector 108 and the like is supplied via an amplifier. Therefore, the settling time of such an amplifier greatly affects the waiting time of the shot. Therefore, it is desired to shorten the waiting time of shots by shortening the settling time of such an amplifier, and reduce the dead time of writing in the electron beam exposure apparatus to improve the throughput.

FIG. 24 is a block diagram showing an example of a driving portion of the main deflector 108. In the figure, the deflection data from the main deflector setting unit 127 includes digital / analog converter (DAC) 201 and current / voltage (IV) converter 2.
02 and the resistor 203, and is supplied to the drive system 210. The drive system 210 includes a resistor 20 connected as shown.
4, 205, the differential amplifier 206, and the capacitor 207.
Consists of. The main deflector 108 comprises a coil 208 which is an inductive load.

[0012]

However, the main deflector 10
Since 8 is composed of the coil 208 and has a large feedback response delay, for example, the input signal shown in FIG.
25, the output signal at the node B when the capacitor 207 is not provided is shown in FIG.
As shown in (b), ringing occurs.

Therefore, in order to speed up the settling time of the response waveform, a capacitor 207 for preventing ringing is connected in parallel with the feedback resistor 204 to lower the band of the differential amplifier 206. That is, the capacitor 20
7 is provided, the frequency characteristic of the differential amplifier 206 is degraded and the slew rate is reduced.

However, by providing the capacitor 207, the change of the input of the differential amplifier 206 becomes extremely slow, and the output signal at the node B becomes as shown in FIG. There was a problem that the settling time of was long. For this reason, the waiting time for shots must be set to a large value according to the long settling time,
This has hindered the improvement of the throughput of the electron beam exposure apparatus.

Therefore, the present invention reduces the amplifier settling time and the shot waiting time by suppressing the ringing of the amplifier output without lowering the band of the amplifier that supplies the deflection data to the deflector. It is an object of the present invention to provide a charged particle beam exposure apparatus capable of performing the above.

[0016]

The above-mentioned problem is solved by claim 1.
Charged particle beam exposure method for forming a pattern by irradiating a charged particle beam from a beam generation source onto a pattern plate having a pattern of transmission holes, and irradiating a charged particle beam transmitted through the pattern plate onto a sample to form a pattern In the first step of deflecting the charged particle beam by a deflector having an inductive load on the downstream side of the beam generation source and on the upstream side of the sample based on the deflection data, and the deflection data via an amplifier. When the first undulation is applied to the deflector to cancel the first undulation of the ringing generated in the output of the amplifier, a pulse-shaped signal having a polarity opposite to that of the first undulation is used as one deflection data.
A second step of adding to the input of the amplifier only once per second.

According to a second aspect of the present invention, in the first aspect, the second step has a pulse width corresponding to a time equal to or less than a time between the start of the first undulation and the start of the second undulation of the ringing. The pulse signal is added to the input of the amplifier. According to a third aspect of the present invention, in the first or second aspect, at least one of a pulse generation timing, a pulse amplitude, and a pulse width of a pulse signal can be variably set in the second step.

In a fourth aspect of the present invention, in any one of the first to third aspects, in the second step, each of the pulse generation timing, pulse amplitude and pulse width of the pulsed signal is changed independently of each other. It can be set. According to a fifth aspect of the present invention, in the first aspect, the second step generates the pulsed signal based on a pulse parameter relating to the pulsed signal, which is stored in advance in a memory.

According to a sixth aspect of the present invention, in the fifth aspect, the method further includes a third step of pre-storing the pulse parameter relating to the pulsed signal in the memory. The invention according to claim 7 is the method according to claim 6, wherein in the third step, a pulse parameter that causes a differential value of the output of the amplifier to be 0 is obtained earliest after the output of the amplifier is changed, and is stored in the memory. Store.

According to an eighth aspect of the present invention, in the sixth aspect, the third step is a pulse in which the time from when the output of the amplifier starts to change to when the output of the amplifier falls within a predetermined range is minimum. The parameters are obtained and stored in the memory. According to a ninth aspect of the present invention, in the sixth aspect, the third step generates a detection signal indicating the movement of the charged particle beam on the sample, and the detection signal of the detection signal is the fastest after the movement of the charged particle beam. A pulse parameter having a differential value of 0 is obtained and stored in the memory.

According to a tenth aspect of the present invention, in any one of the first to ninth aspects, the first step uses a coil having a multi-stage structure in which each stage has at least one coil, and each stage has a coil. Deflection data is supplied via independent amplifiers, and the directions of the currents driving the coils of the adjacent stages are set opposite to each other.

[0022] The invention of claim 11, wherein, in claim 10, wherein the first step is Ru with a plurality out of phase with respect to the optical axis of the charged particle beam coils of the multi-stage configuration.

[0023] The above problems, according to claim 12, irradiated to the pattern plate transmission hole pattern is formed with the charged particle beam from the beam source, irradiating a charged particle beam transmitted through the pattern plate on the sample A charged particle beam exposure apparatus for forming a pattern by means of a deflection means having an inductive load for deflecting the charged particle beam on the downstream side of the beam generation source and on the upstream side of the sample based on deflection data;
When the deflection data is supplied to the deflection means through the amplifier, a pulse-shaped signal having a polarity opposite to that of the first undulation for canceling the first undulation of ringing occurring at the output of the amplifier is provided for each deflection data. It can also be achieved by a charged particle beam exposure apparatus having circuit means for adding only once to the input of the amplifier.

According to a thirteenth aspect of the present invention, in the twelfth aspect , the circuit means has a pulse signal having a pulse width corresponding to a time equal to or less than a time from the start of the first undulation to the start of the second undulation of the ringing. To the input of the amplifier. The invention according to claim 14 is the invention according to claim 12 or 13.
In the above, the circuit means can variably set at least one of a pulse generation timing, a pulse amplitude and a pulse width of a pulsed signal.

The invention of claim 15 is the same as claims 12 to 1 .
In any one of 4 above, the circuit means can variably set the pulse generation timing, the pulse amplitude, and the pulse width of the pulsed signal independently of each other. Claim 16
According to a twelfth aspect of the present invention, the circuit means comprises:
The pulse-shaped signal is generated based on the pulse parameters regarding the pulse-shaped signal that are stored in advance in the memory.

According to a seventeenth aspect of the present invention, in the sixteenth aspect , the present invention further comprises a memory and a control means for storing pulse parameters relating to a pulse-shaped signal in the memory in advance. According to an eighteenth aspect of the present invention, in the seventeenth aspect , the control means obtains a pulse parameter that makes the differential value of the output of the amplifier become 0 fastest after the output of the amplifier changes, and stores it in the memory. .

According to a nineteenth aspect of the present invention, in the seventeenth aspect , the control means sets a pulse parameter that minimizes a time from when the output of the amplifier starts to change until the output of the amplifier falls within a predetermined range. It is obtained and stored in the memory. According to a twentieth aspect of the invention, in the seventeenth aspect , the control means generates the detection signal indicating the movement of the charged particle beam on the sample, and the detection signal of the detection signal is the fastest after the movement of the charged particle beam. Means for obtaining a pulse parameter having a differential value of 0 and storing it in the memory.

The invention according to claim 21 is the invention according to claims 12 to 2.
0, the deflection means has a multi-stage coil each stage having at least one coil,
Deflection data is supplied to the coils of each stage through independent amplifiers, and means for setting the directions of the currents driving the coils of the adjacent stages in mutually opposite directions is further provided.

According to a twenty-second aspect of the present invention, in the twenty-first aspect , the deflecting means is provided with a plurality of coils having a multi-stage structure with their phases shifted with respect to the optical axis of the charged particle beam.
It

[0030]

[0031]

According to the first aspect of the invention, by adding the pulsed signal to the input of the amplifier, the response delay fed back to the amplifier can be canceled and the ringing can be suppressed, so that the frequency of the amplifier can be suppressed. An amplifier for an inductive load with a short settling time can be realized without sacrificing characteristics.

According to the present invention, the ringing of the output of the amplifier can be surely suppressed. According to the sixth to ninth aspects of the invention, since the pulse signal can be generated based on the pulse parameter having the optimum value stored in advance, the ringing of the output of the amplifier can be suppressed with a relatively simple configuration. You can

According to the inventions of claims 10 and 11,
With the multi-stage coil, the load can be substantially reduced and the drift of the output of the amplifier can be substantially canceled, so the settling time of the amplifier can be shortened.
It

According to the twelfth aspect of the invention, by adding the pulsed signal to the input of the amplifier, the response delay fed back to the amplifier can be canceled and the ringing can be suppressed, so that the frequency of the amplifier can be suppressed. An amplifier for an inductive load with a short settling time can be realized without sacrificing characteristics.

According to the thirteenth to sixteenth aspects of the present invention, ringing of the output of the amplifier can be reliably suppressed. According to the seventeenth to twentieth aspects of the present invention, since the pulse signal can be generated based on the pulse parameter having the optimum value stored in advance, the ringing of the output of the amplifier can be suppressed with a relatively simple configuration. You can

According to the inventions of claims 21 and 22 ,
With the multi-stage coil, the load can be substantially reduced and the drift of the output of the amplifier can be substantially canceled, so the settling time of the amplifier can be shortened.
It

Therefore, according to the present invention, the ringing of the amplifier output is suppressed without lowering the band of the amplifier that supplies the deflection data to the deflector, thereby shortening the amplifier settling time and the shot waiting time. can do.

[0038]

1 is a block diagram showing a main part of a first embodiment of a charged particle beam exposure apparatus according to the present invention. 24, those parts which are the same as those corresponding parts in FIG. 24 are designated by the same reference numerals. In this embodiment, the present invention is applied to an electron beam exposure apparatus, and the first embodiment of the charged particle beam exposure method according to the present invention is used.

FIG. 1 shows a driving portion of the main deflector 108. In the figure, the deflection data from the main deflector setting unit 127 is the digital-analog converter (DAC) 20.
1, is supplied to the drive system 210 via the current / voltage (IV) converter 202 and the resistor 203. The drive system 210 includes resistors 204 and 205 and a differential amplifier 206 which are connected as shown in the figure. The main deflector 108 comprises a coil 208 which is an inductive load. The node N has a resistor 203,
The resistor 204 and a resistor 5 described later are connected to the inverting input terminal of the differential amplifier 206. The non-inverting input terminal of the differential amplifier 206 is grounded. Further, the node D has a resistance 204
And the resistor 205 are connected, and the coil 208 is connected to the differential amplifier 2
It is connected between the output terminal of 06 and the node D.

Further, the pulse generation circuit 1 is based on the deflection data from the main deflector setting unit 127, and at least the first waviness for canceling at least the first waviness of the ringing appearing in the output of the differential amplifier 206. Generates a pulsed signal of opposite polarity and supplies it to the differential amplifier 206 via the resistor 5. The pulse width of this pulsed signal is
It corresponds to less than the time between the start of the first swell of ringing and the start of the second swell. Pulse generation circuit 1
Is composed of a pulse parameter output circuit 2, a DAC 3, and a current / voltage converter 4. The pulse parameter output circuit 2 outputs a pulse parameter for suppressing ringing of an output signal at a node D, which will be described later, based on the deflection data from the main variation late setting unit 127. The pulse parameters include the pulse delay time, pulse width, amplitude, etc. of the pulse-shaped signal output from the pulse generation circuit 1, and D
Applied to node C via AC3 and IV converter 4. It is desirable that at least one of the pulse parameters can be variably set, and each pulse parameter can be variably set independently of each other.

FIG. 2A shows an input signal applied to the node A in FIG. 1, and FIG. 2B shows an output signal obtained from the node D when the pulse generating circuit 1 is not provided. Show. 2C shows a pulse-shaped signal output from the pulse generation circuit 1 and applied to the node C, and FIG. 2D is obtained from the node D because the pulse generation circuit 1 is provided. The output signal is shown. That is, in the present embodiment, the pulse-shaped signal obtained through the node C is added to the signal obtained through the node A and applied to the inverting input terminal of the differential amplifier 206, whereby the differential amplifier 205. The response delay that is fed back to is canceled out, and the ringing shown in FIG. 2B is suppressed. Therefore, it is possible to realize the differential amplifier 206 for inductive loads with a short settling time without sacrificing the frequency characteristics of the differential amplifier 206.

Next, a second embodiment of the charged particle beam exposure apparatus according to the present invention will be described with reference to FIG. FIG. 3 is a block diagram showing a main part of the second embodiment of the charged particle beam exposure apparatus. In the figure, those parts which are the same as those corresponding parts in FIG. 1 are designated by the same reference numerals, and a description thereof will be omitted. In this embodiment, the present invention is applied to an electron beam exposure apparatus, and the second embodiment of the charged particle beam exposure method according to the present invention is used.

In FIG. 3, the deflection data is generated from the pattern generator (PG) 127A and supplied to the DAC 201 and the pulse generation circuit 11. Pulse generation circuit 11
Is a data timing adjustment circuit 12 and memories 13, 1
4, a clock generation circuit 15, a DAC 3, and an IV converter 4. DAC 201 and clock generation circuit 1
The clock supplied to 5 is obtained from the clock unit 122 shown in FIG. 22, for example. The clock from the clock generation circuit 15 is supplied to the DAC 3. Further, the output signal from the node D is supplied to the control unit 16 via an analog / digital converter (ADC) 17 which constitutes a feedback adjustment system. This control unit 16
Corresponds to the control unit 124 shown in FIG. 22, for example.

The change of the nth to n + 1th data from the PG 127A appears as a change of the output of the differential amplifier 206. Therefore, the change in the voltage at the node D
By feeding back to the control unit 16 via, the control unit 16 can monitor the output change of the differential amplifier 206. The control unit 16 is an ADC
The data fed back via 17 is differentiated to find the pulse parameter whose differential value becomes 0 the fastest, that is, the pulse parameter which suppresses the first undulation of the ringing of the signal obtained from the node D. Specifically, while the control unit 16 changes the pulse parameter, the AD unit
The data obtained from C17 is differentiated, and the optimum value of the pulse parameter where the differential value becomes 0 fastest is obtained. When obtaining the optimum values of the pulse parameters, the control unit 16 supplies the parameters related to the pulse delay and the pulse width among the pulse parameters to the clock generation circuit 15 and the parameters related to the pulse amplitude to the DAC 3.

In this way, the optimum value of the pulse parameter obtained by the control unit 16 uses the nth to (n + 1) th data from PG127A as the address, and the memory 13 as the parameter of the DAC 3 for the output change.
14 is stored. Of the optimum values of pulse parameters,
The optimum values of the parameters regarding the pulse delay and the pulse width are stored in the memory 13, and the optimum values of the parameters regarding the pulse amplitude are stored in the memory 14. The operation of storing the optimum pulse parameter values in the memories 13 and 14 is performed for three types of data changes. The optimum values of the pulse parameters for other data changes are obtained from the optimum values of the pulse parameters for the three kinds of data changes using the following approximation function δpara.

Δpara = P (Xn + 1-X
n) × Q (Xn) In the above equation, P is a function depending on the jump amount of data,
Q is a linear interpolation function. That is, the pulse parameter is proportional to the change of the input data, that is, the difference between the data or the jump amount, and is affected by the dependency term of the jump start point Xn as a correction in the form of a product. B (Xn) is the control unit 1
The point monitored by 6 is a linear interpolation function that coincides with the monitoring point. Based on such an approximate function δpara, the change of all input data, that is, the optimum value of the pulse parameter with respect to the jump amount is obtained. Then, once again, based on the optimum values of these pulse parameters, the final optimum values of the pulse parameters are finely adjusted by the same method as that for obtaining the optimum values of these pulse parameters. Therefore, the optimum values of the pulse parameters for all the changes of the input data can be obtained in a relatively short time.

Next, the operation at the time of exposure will be described. PG12
The deflection data corresponding to the (n + 1) th pattern data from 7A includes the DAC 201, the IV converter 202, and the resistor 2.
It is supplied to the differential amplifier 206 via 03 as a change from the nth data to the (n + 1) th data. On the other hand,
The data timing adjustment circuit 12 adjusts the timing of the nth data and the n + 1th data from the PG 127A, and supplies the nth data and the n + 1th data to the memories 13 and 14 as addresses at the same time.
As a result, the parameters regarding the pulse delay and the pulse width among the pulse parameters are read from the memory 13 and supplied to the clock generation circuit 15. Further, of the pulse parameters, the parameter relating to the pulse amplitude is read out from the memory 13 and supplied to the DAC 3.

FIG. 4 is a block diagram showing an embodiment of the data timing adjusting circuit 12 together with the memories 13 and 14. In the figure, for convenience of description, the memory 13 and the memory 14
Are shown as one memory. The data timing adjustment circuit 12 has a register 12A, and the nth data is once stored in the register 12A and then supplied to the memories 13 and 14 at the same timing as the (n + 1) th data.

The clock generation circuit 15 generates a clock in response to the clock synchronized with the (n + 1) th data from the clock unit 122 based on the parameters regarding the pulse delay and the pulse width read from the memory 13, and the DAC 3 Supply to. The parameter regarding the pulse amplitude read from the memory 14 is supplied to the DAC 3 as weighting data. The output of the DAC 3 is supplied to the node via the IV converter 4 and the resistor 5, and the DAC 201 obtained via the IV converter 202 and the resistor 203.
Is added to the output of the differential amplifier 206 before being supplied to the inverting input terminal of the differential amplifier 206.

FIG. 5A shows a clock supplied from the clock unit 122 to the DAC 201 and the clock generation circuit 15, and FIG. 5B shows the clock obtained from the node D when the pulse generation circuit 11 is not provided. 2 shows the output signal that can be output. Further, FIG. 5C shows a clock supplied from the clock generation circuit 15 to the DAC 3, and FIG.
The correction pulse signal output from the IV converter 4 is shown.

FIG. 6 is a diagram for explaining the output voltage obtained by the simulation. FIG. 6A shows the output voltage obtained from the node D when the pulse generation circuit 11 is not provided and the correction pulse signal is not applied to the node N. Further, FIG. 6B shows an output voltage obtained from the node D when the pulse generation circuit 11 is provided and the correction pulse signal is applied to the node N. Figure 6
As is clear from the comparison between (a) and FIG. 6 (b), according to the present embodiment, the ringing of the output voltage is suppressed well.

Next, the operation for obtaining the optimum value of the pulse parameter in FIG. 3 will be described with reference to FIG. In FIG. 7, in step S1, the PG 127A outputs deflection data and inputs it to the amplifier 206 via the DAC 201, the IV converter 202 and the resistor 203. In step S2, the pulse amplitude and the pulse delay are set from the control unit 16 to the pulse generation circuit 11. In step S3, the pulse generation circuit 11 outputs a correction pulse signal based on the pulse parameter set in step S2. In step S4, the control unit 16 outputs from node D
The output signal obtained via the DC 17 is monitored, and in step S5, the settling time of the output signal monitored by the control unit 16 is measured. In step S6, the control unit 16 determines whether or not the correction pulse signal has been generated for all pulse amplitudes and pulse delays in the pulse generation circuit 11, and if the determination result is NO, the process returns to step S2. .

On the other hand, if the decision result in the step S6 is YES, in a step S7, the control unit 16 obtains a pulse amplitude and a pulse delay that minimize the settling time. In step S8, it is determined whether or not the control unit 16 has acquired enough data to fit the approximate function δpara, and if the determination result is NO, the process proceeds to step S8.
Return to 1. On the other hand, if the decision result in the step S8 is YES, the control unit 16 applies the acquired data to the approximate function δpara in a step S9. Further, in step S10, the control unit 16 obtains pulse parameters regarding pulse amplitude, pulse delay and pulse width for all deflection data (deflection patterns). Step S1
In 1, the pulse parameters obtained by the control unit 16 are written in the memories 13 and 14, and the process ends.

Next, a third embodiment of the charged particle beam exposure apparatus according to the present invention will be described with reference to FIG. FIG. 8 is a block diagram showing a main part of a third embodiment of the charged particle beam exposure apparatus. In the figure, those parts which are the same as those corresponding parts in FIG. 3 are designated by the same reference numerals, and a description thereof will be omitted. In this embodiment, the present invention is applied to an electron beam exposure apparatus, and a charged particle beam exposure method according to the third embodiment of the present invention is used.

In FIG. 8, the feedback adjustment system is
It comprises an error band setting circuit 22 for setting the width of the error band based on an instruction from the control unit 16 and a window comparator 21. The window comparator 21 compares the output signal obtained from the node D with the error band width obtained from the error band setting circuit 22,
A signal is provided to the control unit 16 indicating whether the output signal falls within the error bandwidth. As a result, the control unit 16 minimizes the time from when the output signal (voltage) at the node D starts to change to within the error bandwidth based on the signal from the window comparator 21.
The optimum value of the pulse parameter is obtained by changing the pulse parameter so as to obtain the parameter of AC3 and continuing the feedback by the window comparator 21. Other operations are the same as in the case of the second embodiment described with reference to FIG.

Next, the operation for obtaining the optimum value of the pulse parameter in FIG. 8 will be described with reference to FIG. In FIG. 9, in step S21, the control unit 16 sends an instruction to the error band setting circuit 22 to set the width of the error band. In step S22, the PG 127A outputs the deflection data, and the DAC 201, the IV converter 202 and the resistor 2 are output.
It is input to the amplifier 206 via 03. Step S23
Then, the control unit 16 sets the pulse amplitude and the pulse delay in the pulse generation circuit 11. In step S24, the pulse generation circuit 11 outputs a correction pulse signal based on the pulse parameter set in step S23. In step S25, the control unit 16 monitors the signal obtained from the window comparator 21, and in step S26, the settling time (setting time) of the signal monitored by the control unit 16, that is, the output signal from the node D is set. Measure the settling time to fall within the error bandwidth. In step S27, the control unit 16
However, it is determined whether or not the correction pulse signal is generated for all the pulse amplitudes and the pulse delays in the pulse generation circuit 11, and if the determination result is NO, the process is step S2.
Return to 3.

On the other hand, the decision result in the step S27 is YES.
In this case, in step S28, the control unit 16 obtains the pulse amplitude and the pulse delay that minimize the settling time. In step S29, it is determined whether or not the control unit 16 has acquired enough data to fit the approximate function δpara. If the determination result is NO, the process returns to step S22. On the other hand, if the decision result in the step S29 is YES, the control unit 16 applies the acquired data to the approximate function δpara in a step S30.
Further, in step S31, the control unit 16 obtains pulse parameters regarding pulse amplitude, pulse delay, and pulse width for all deflection data (deflection patterns).
In step S32, the pulse parameters obtained by the control unit 16 are written in the memories 13 and 14, and the process is completed.

Next, a fourth embodiment of the charged particle beam exposure apparatus according to the present invention will be described with reference to FIG. Figure 10
It is a block diagram which shows the principal part of 4th Example of a charged particle beam exposure apparatus. In the figure, those parts which are the same as those corresponding parts in FIG. 3 are designated by the same reference numerals, and a description thereof will be omitted. In this embodiment, the present invention is applied to an electron beam exposure apparatus, and the fourth embodiment of the charged particle beam exposure method according to the present invention is used.

In FIG. 10, the feedback adjustment system amplifies a backscattered electron detector 31 for detecting backscattered electrons in the electron beam irradiated on the wafer 110 and a detection signal from the backscattered electron detector 31. Amplifier 32
And an ADC 33 that converts the output of the amplifier 32 into digital data and supplies it to the control unit 16. The control unit 16 can monitor the movement of the electron beam based on the digital data obtained from the ADC 33. Therefore, in the same manner as in the second embodiment, the control unit 16 changes the position of the electron beam,
The digital data obtained from the ADC 33 is differentiated, and the optimum value of the pulse parameter whose differential value becomes 0 fastest is obtained. When obtaining the optimum values of the pulse parameters, the control unit 16 supplies the parameters related to the pulse delay and the pulse width among the pulse parameters to the clock generation circuit 15 and the parameters related to the pulse amplitude to the DAC 3. Other operations are the same as in the case of the second embodiment described with reference to FIG.

Next, the operation for obtaining the optimum value of the pulse parameter in FIG. 10 will be described with reference to FIG. Figure 11
In step S41, the PG 127A outputs deflection data and inputs it to the amplifier 206 via the DAC 201, the IV converter 202, and the resistor 203. Step S4
In 2, the control unit 16 sets the pulse amplitude and the pulse delay in the pulse generation circuit 11. Step S43
Then, the pulse generation circuit 11 outputs the correction pulse signal based on the pulse parameter set in step S2. In step S44, the control unit 16 detects the backscattered electrons with the backscattered electron detector 31 while changing the position of the electron beam. In step S45, the control unit 16
Is an amplifier 32 that detects the detection signal output from the backscattered electron detector 31.
Also, the settling time of the detection signal monitored by the control unit 16 is measured in step S46. In step S47, the control unit 16 determines whether or not the correction pulse signal has been generated for all pulse amplitudes and pulse delays in the pulse generation circuit 11, and if the determination result is NO, the process returns to step S42. .

On the other hand, the decision result in the step S47 is YES.
In this case, in step S48, the control unit 16 obtains the pulse amplitude and the pulse delay that minimize the settling time. In step S49, it is determined whether or not the control unit 16 has acquired enough data to fit the approximate function δpara. If the determination result is NO, the process returns to step S41. On the other hand, if the decision result in the step S49 is YES, the control unit 16 applies the acquired data to the approximate function δpara in a step S50.
Further, in step S51, the control unit 16 obtains pulse parameters regarding pulse amplitude, pulse delay, and pulse width for all deflection data (deflection patterns).
In step S52, the pulse parameters obtained by the control unit 16 are written in the memories 13 and 14, and the process is completed.

In each of the above-mentioned embodiments, the case where the deflector is an electromagnetic deflector such as a main deflector has been described as an example. However, the deflector to which the present invention is applicable is a main deflector or an electromagnetic deflector. It is not limited to vessels. By the way, in FIG. 22, the dynamic mask stigmator DS is, for example, as shown in FIG.
It has the configuration shown in. In FIG. 12, (a) shows the configuration of the portion of the dynamic mask stigmator DS with respect to the x-axis, and (b) shows the dynamic mask stigmator DS.
The structure of the part with respect to the y-axis of FIG.

In FIG. 12 (a), the stigmator coil portion for the x-axis of the dynamic mask stigmator DS has, for example, coils LX1 to LX1 each having 40 turns.
With LX4. For example, 12-bit data is DAC4
1 and a coil LX1 connected in series via an amplifier 42
To LX4. Further, in FIG. 12B, the stigmator coil portion with respect to the y axis of the dynamic mask stigmator DS has coils LY1 to LY4 each having 40 turns. For example, 12-bit data is supplied to the coils LY1 to LY4 connected in series via the DAC 51 and the amplifier 52. Amplifier 4 for supplying data to the dynamic mask stigmator DS
The settling time of 1,51 is determined by the load of the dynamic mask stigmator DS. That is, the amplifier 4
The settling times of 1,51 are the diameters of the coils LX1 to LX4, LY1 to LY4 of the dynamics mask tigmeter DS, the number of turns (the number of windings) that determines the self-inductance, and the mutual inductance that determines the interaction between the coils. Depends on.

However, since the magnitude of the load is predetermined to a certain extent by the beam deflection system and the lens system, it is not possible to reduce the load and shorten the settling time of the amplifiers 41 and 51. FIG. 13 is a diagram illustrating ringing of the output signal of the amplifier 41 according to the number of turns of each of the coils LX1 to LX4, for example. In the figure, (a) shows the output signal of the amplifier 41 when the number of turns is 0, that is, when the coil is short-circuited. Further, in the figure, (b) shows the output signal of the amplifier 41 when the number of turns is 10, (c) shows the number of turns of 20, and (d) shows the output signal of the amplifier 41 when the number of turns is 30, respectively. As is apparent from FIG. 13, as the number of turns increases, the load increases and the ringing of the output signal of the amplifier 41 also increases.

Further, in FIG. 22, the dynamic focus coil DF has the structure shown in FIG. 14, for example.
In FIG. 14, the focus coil portion of the dynamic focus coil DF has, for example, a coil LF having 40 turns each. For example, 12-bit data is supplied to the coil LF via the DAC 61 and the amplifier 62. The settling time of the amplifier 61 that supplies data to the dynamic focus coil DF is generally determined by the load of the dynamic focus coil DF. That is, the settling time of the amplifier 61 depends on the dynamic focus coil D
It depends on the diameter of the coil LF of F, the number of turns (the number of windings) that determines the self-inductance, the mutual inductance that determines the interaction between the coils, and the like.

However, since the magnitude of the load is determined in advance by the beam deflection system and the lens system, it is not possible to reduce the load and shorten the settling time of the amplifier 61. Therefore, by adding the correction pulse signal to the input of the amplifier as in each of the above embodiments, the amplifiers 41, 51, 61 are
It is possible to suppress the ringing of the output signal of and reduce the settling time.

As a method of shortening the settling time of the amplifier, it is also possible to reduce the substantial load of the coil as in the following embodiments. Next, a fifth embodiment of the charged particle beam exposure apparatus according to the present invention will be described with reference to FIG. FIG. 15 is a view showing the main parts of the fifth embodiment of the charged particle beam exposure apparatus. 12, those parts which are the same as those corresponding parts in FIGS. 12 and 22 are designated by the same reference numerals, and a description thereof will be omitted. In this embodiment, the present invention is applied to an electron beam exposure apparatus, and the fifth embodiment of the charged particle beam exposure method according to the present invention is used.

FIG. 15 shows the stigmator coil portion DS of the dynamic mask stigmator DS with respect to the x-axis.
x1 to DSxN, and the stigmator coil portion DS with respect to the y axis of the dynamic mask stigmator DS
Illustration of y1 to DSyN is omitted because it is substantially the same as the stigmator coil portions DSx1 to DSxN with respect to the x axis. Dynamic Mask Stigmeter DS x
The stigmator coil portion DSx1 for the axis is, for example, coils LX1-1 to LX4 each having 40 turns.
-N (N is an integer). For example, 12-bit data is supplied to the coils LX1-1 to LX4-1 connected in series via the DAC 41-1 and the amplifier 42-1. Similarly, the 12-bit data is connected in series via the DAC 41-2 and the amplifier 42-2, and the coils LX1-2 to LX4- form the stigmator coil portion DSx2.
2 and the 12-bit data is connected in series via the DAC 41-N and the amplifier 42-N to form the stigmator coil portion DSxN.
Supplied to 4-N.

In FIG. 15, the amplifier 4 connected to each stage
Even if the same amplifier is used for 1-1 to 41-N, a slight drift occurs in the output of each amplifier depending on the temperature. That is, there is a slight drift in the output of the amplifier due to a change in ambient temperature, a change in temperature due to heat generation of the amplifier itself, and the like. Therefore, when a plurality of stages of stigmator coil portions are arranged on the optical axis, fluctuations in the current density of the electron beam passing through the circular aperture 107 and blurring of the electron beam occur. FIG. 16 is a diagram showing variations in the current density of the electron beam passing through the circular aperture 107 due to the position shift of the electron beam caused by the drift of the output of the amplifier in this case. As is clear from FIG. 16, it is necessary to position the electron beam with extremely high accuracy in order to keep the fluctuation of the current density within 1%. Therefore, in this embodiment, as shown in FIG. 15, two adjacent stigmator coil portions (for example, stigmator coil portion D) are used.
Between Sx1 and DSx2), the current driving directions are set to be opposite to each other.

FIG. 17 is a diagram showing the drift of the output of the amplifier. As shown in the figure, for the current in the positive direction, the drift increases in the negative direction with time. On the other hand, for a negative current, the drift increases negatively with time. Therefore, the drift of the output of the amplifier can be substantially canceled by setting the driving directions of the currents to be opposite to each other between the two adjacent stigmator coil portions as described above.

FIG. 18 is a diagram showing the arrangement of the stigmator coil portion when N = 6. Each of the stigmator coil portions DSx1 to DSx6 has, for example, a diameter of 15 mm, a length on the optical axis of the electron beam exposure apparatus of 4 mm, and is arranged at a pitch of 0.8 mm. The total distance on the optical axis of the electron beam exposure apparatus from the stigmator coil portions DSx1 to DSx6 is 28 mm.

The configuration of the stigmator coil portions DSy1 to DSyN with respect to the y axis of the dynamic stigmator DS is similar to that of the stigmator coils DSx1 to DSxN, and the stigmator coils DSx1 to DSx1 respectively.
For example, the phase is shifted by 45 degrees with respect to the optical axis of the electron beam exposure apparatus with respect to DSxN.

In the fifth embodiment, the coil is provided with four poles, but even if the number of poles of the coil is 2 or 4 or more, the magnetic field generated by each coil is at the center position between the coils. It goes without saying that it is sufficient if they are connected so as to offset each other. Next, a sixth embodiment of the charged particle beam exposure apparatus according to the present invention will be described with reference to FIG. FIG. 19 is a diagram showing a main part of a sixth embodiment of the charged particle beam exposure apparatus.
14, those parts which are the same as those corresponding parts in FIGS. 14 and 22 are designated by the same reference numerals, and a description thereof will be omitted. In this embodiment, the present invention is applied to an electron beam exposure apparatus, and the sixth embodiment of the charged particle beam exposure method according to the present invention is used.

FIG. 19 shows the focus coil portion of the dynamic focus coil DF. The focus coil portion of the dynamic focus coil DF has coils LF1 to LFN (N is an integer) each having, for example, 40 turns. For example, 12-bit data is DAC61
-1 and the amplifier 62-1 are supplied to the coil LF1. Similarly, 12-bit data is converted to DAC61-
2 and the amplifier 62-2 are supplied to the coil LF2 forming the focus coil portion, and 12-bit data is supplied to the coil LFN forming the focus coil portion via the DAC 61-N and the amplifier 62-N.

In FIG. 19, the amplifier 6 connected to each stage
Even if the same amplifier is used for 1-1 to 61-N, a slight drift occurs in the output of each amplifier depending on the temperature. That is, there is a slight drift in the output of the amplifier due to a change in ambient temperature, a change in temperature due to heat generation of the amplifier itself, and the like. Therefore, when a plurality of stages of focus coil portions are arranged on the optical axis, fluctuations in the current density of the electron beam passing through the circular aperture 107 and blurring of the electron beam occur. FIG. 20 is a diagram showing variations in the current density of the electron beam passing through the circular aperture 107 due to the position shift of the electron beam caused by the drift of the output of the amplifier in this case. As is clear from FIG. 20, it is necessary to position the electron beam with extremely high accuracy in order to keep the fluctuation of the current density within 1%. Therefore, in the present embodiment, as shown in FIG. 19, the drive directions of the currents are set to be opposite to each other between two adjacent focus coil portions (for example, stigmator coil portions DF1 and DF2).

FIG. 21 is a diagram showing the arrangement of the focus coil portion when N = 5. In the figure, (a) shows the focus coil portion of the dynamic focus coil DF of the electronic exposure apparatus, and (b) shows the arrangement of each focus coil of the dynamic focus coil DF in an enlarged manner. Further, in FIG. 7B, the magnetic field B on the optical axis of the lens diameter L2a is also shown. The focus coils LF1 to LF5 have a diameter φ of 5 mm, for example, and are arranged at a pitch P of 5 mm. The total distance TL on the optical axis of the electron beam exposure apparatus from the focus coils LF1 to LF5 is 20 m.
m. Here, it is sufficient if P ≧ φ.

In the fifth and sixth embodiments, the correction pulse signal may be added to the input of each amplifier as in the first to fourth embodiments. In this case, it is possible to reduce the adverse effect of the output drift of the amplifier and the adverse effect of the ringing of the output of the amplifier.

The present invention is not limited to electron beam exposure, but can be widely applied to charged particle beam exposure. Although the present invention has been described above with reference to the embodiments, it is needless to say that the present invention is not limited to these embodiments and various modifications and improvements can be made within the scope of the present invention.

[0079]

According to the first aspect of the present invention, by adding the pulsed signal to the input of the amplifier, the response delay fed back to the amplifier can be canceled and the ringing can be suppressed. An amplifier for an inductive load having a short settling time can be realized without sacrificing the frequency characteristic of.

According to the invention described in claims 2 to 5, ringing of the output of the amplifier can be surely suppressed. According to the sixth to ninth aspects of the invention, since the pulse signal can be generated based on the pulse parameter having the optimum value stored in advance, the ringing of the output of the amplifier can be suppressed with a relatively simple configuration. You can

According to the inventions of claims 10 and 11,
With the multi-stage coil, the load can be substantially reduced and the drift of the output of the amplifier can be substantially canceled, so the settling time of the amplifier can be shortened.
It

According to the twelfth aspect of the present invention, by adding the pulsed signal to the input of the amplifier, the response delay fed back to the amplifier can be canceled and the ringing can be suppressed. An amplifier for an inductive load with a short settling time can be realized without sacrificing characteristics.

According to the thirteenth to sixteenth aspects of the present invention, ringing of the output of the amplifier can be surely suppressed. According to the seventeenth to twentieth aspects of the present invention, since the pulse signal can be generated based on the pulse parameter having the optimum value stored in advance, the ringing of the output of the amplifier can be suppressed with a relatively simple configuration. You can

According to the inventions of claims 21 and 22 ,
With the multi-stage coil, the load can be substantially reduced and the drift of the output of the amplifier can be substantially canceled, so the settling time of the amplifier can be shortened.
It

Therefore, according to the present invention, the ringing of the amplifier output is suppressed without lowering the band of the amplifier that supplies the deflection data to the deflector, so that the amplifier settling time is shortened and the shot waiting time is shortened. can do.

[Brief description of drawings]

FIG. 1 is a block diagram showing a main part of a first embodiment of a charged particle beam exposure apparatus according to the present invention.

FIG. 2 is a timing chart showing signals in each part shown in FIG. 1 with and without a pulse generation circuit.

FIG. 3 is a block diagram showing a main part of a second embodiment of the charged particle beam exposure apparatus.

FIG. 4 is a block diagram showing an embodiment of a data timing adjustment circuit together with a memory.

FIG. 5 is a timing chart showing signals in each unit shown in FIG.

FIG. 6 is a diagram illustrating an output voltage obtained by simulation.

FIG. 7 is a flowchart illustrating an operation of obtaining an optimum value of a pulse parameter in FIG.

FIG. 8 is a block diagram showing a main part of a third embodiment of the charged particle beam exposure apparatus.

9 is a flowchart illustrating an operation of obtaining an optimum value of a pulse parameter in FIG.

FIG. 10 is a block diagram showing a main part of a fourth embodiment of the charged particle beam exposure apparatus.

FIG. 11 is a flowchart illustrating an operation for obtaining an optimum pulse parameter value in FIG. 10;

FIG. 12 is a diagram showing a configuration of a dynamic mask stigmator.

FIG. 13 is a diagram illustrating ringing of an output signal of an amplifier according to the number of turns of each coil, for example.

FIG. 14 is a diagram showing a configuration of a dynamic focus coil.

FIG. 15 is a diagram showing a main part of a fifth embodiment of the charged particle beam exposure apparatus.

FIG. 16 is a diagram showing fluctuations in the current density of an electron beam passing through a circular aperture due to a position shift of the electron beam caused by a drift of an output of an amplifier for a stigmator coil.

FIG. 17 is a diagram showing the drift of the output of the amplifier.

FIG. 18 is a diagram showing an arrangement of stigmator coil portions when N = 6.

FIG. 19 is a diagram showing a main part of a sixth embodiment of the charged particle beam exposure apparatus.

FIG. 20 is a diagram showing fluctuations in the current density of the electron beam passing through the circular aperture due to the displacement of the electron beam caused by the drift of the output of the amplifier, for the focus coil.

FIG. 21 is a diagram showing an arrangement of focus coil portions when N = 5.

FIG. 22 is a schematic view showing an example of a conventional electron beam exposure apparatus using block exposure.

FIG. 23 is a diagram showing an example of a block mask.

FIG. 24 is a block diagram showing an example of a driving portion of a main deflector.

FIG. 25 is a timing chart showing signals in each unit shown in FIG. 22.

[Explanation of symbols]

1,11 pulse generator 2 pulse parameter output circuit 3,41,51,61,201 DAC 4,202 IV converter 5,203-205 resistance 12 Data timing adjustment circuit 12A register 13,14 memory 15 Clock generation circuit 16 control unit 17,33 ADC 21 Window collator 22 Error band setting circuit 32, 42, 52, 62 amplifier 101 electron gun L1a electronic lens system 102 rectangular aperture plate L1b electronic lens system 103 Beam shaping deflector MD1 First mask deflector DS Dynamic Mask Stigmator MD2 Second mask deflector DF Dynamic mask focus coil L2a electronic lens system 104 block mask 105 mask stage L2b electronic lens system MD3 Third mask deflector 106 Blanking deflector MD4 Fourth mask deflector L3 reduction electron lens system 107 circular aperture L4 projection electron lens system 108 Main deflector (electromagnetic deflector) 109 Sub deflector (electrostatic deflector) L5 projection electron lens system 110 wafers 111 wafer stage 121 Central control unit (CPU) 122 clock unit 122 123 buffer memory 124 control unit 124 125 data correction unit 126 mask memory 127 Main deflector setting unit 127A PG 128 buses 206 differential amplifier 207 capacitor 208 coil 210 drive system A, B, C, D, N nodes

─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Hiroshi Yasuda 1015 Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture, Fujitsu Limited (72) Inventor Yoshihisa Ohba, 1015, Kamiodanaka, Nakahara-ku, Kawasaki, Kanagawa Prefecture, Fujitsu Limited ( 72) Inventor Tomohiko Abe 1015 Kamiodanaka, Nakahara-ku, Kawasaki-shi, Kanagawa Within Fujitsu Limited (56) Reference JP 61-59727 (JP, A) JP 5-144912 (JP, A) JP Hei 2-246316 (JP, A) JP-A-6-177023 (JP, A) JP-A-6-151283 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) H01L 21/027 G03F 7/20 H01J 37/147 H01J 37/305

Claims (22)

(57) [Claims] 1. A charged particle beam for forming a pattern by irradiating a charged particle beam from a beam generation source onto a pattern plate having a transmission hole pattern, and irradiating the charged particle beam transmitted through the pattern plate onto a sample. In the exposure method, a first step of deflecting a charged particle beam by a deflector having an inductive load on the downstream side of the beam generation source and on the upstream side of the sample based on the deflection data, and the deflection data amplifier when supplying to the deflector through, for canceling the first wave of ringing at the output of the amplifier, the first swell only the amplifier once per deflection data pulsed signals of opposite polarity Charged particle beam exposure method, the method further comprising: 2. The second step adds to the input of the amplifier a pulse signal of a pulse width corresponding to a time between the start of the first undulation and the start of the second undulation of the ringing. The charged particle beam exposure method according to claim 1. 3. The charged particle beam exposure method according to claim 1, wherein in the second step, at least one of pulse generation timing, pulse amplitude and pulse width of a pulsed signal can be variably set. . 4. The second step can variably set the pulse generation timing, pulse amplitude, and pulse width of a pulse-shaped signal independently of each other.
The charged particle beam exposure method according to claim 1. 5. The charged particle beam exposure method according to claim 1, wherein in the second step, the pulsed signal is generated based on a pulse parameter relating to the pulsed signal, which is stored in advance in a memory. 6. The charged particle beam exposure method according to claim 5, further comprising a third step of pre-storing pulse parameters relating to a pulsed signal in said memory. 7. In the third step, the differential value of the output of the amplifier is 0 fastest after the output of the amplifier changes.
7. The charged particle beam exposure method according to claim 6, wherein a pulse parameter to be obtained is obtained and stored in the memory. 8. The third step obtains a pulse parameter that minimizes the time from when the output of the amplifier begins to change until the output of the amplifier falls within a predetermined range, and stores the pulse parameter in the memory. Item 7. The charged particle beam exposure method according to Item 6. 9. The third step is to generate a detection signal indicating the movement of the charged particle beam on the sample, and a pulse parameter that makes the differential value of the detection signal be 0 fastest after the movement of the charged particle beam. The charged particle beam exposure method according to claim 6, further comprising: 10. The first step uses a coil having a multi-stage structure, each stage including at least one coil,
10. The charging according to claim 1, wherein the deflection data is supplied to the coils of each stage through an independent amplifier, and the directions of the currents driving the coils of the adjacent stages are set to be opposite to each other. Particle beam exposure method. 11. The charged particle beam exposure method according to claim 10, wherein in the first step, a plurality of multi-stage coils are used with their phases shifted with respect to the optical axis of the charged particle beam. 12. A charged particle beam from a beam source
The pattern plate on which the through hole pattern is formed is irradiated and the pattern
Irradiate the charged particle beam that passed through the turn plate onto the sample.
In a charged particle beam exposure apparatus that forms a pattern , based on deflection data, the downstream side of the beam generation source, and
The inductive property of deflecting the charged particle beam on the upstream side of the sample
A deflection means having a load, and when supplying the deflection data to the deflection means via an amplifier
The first swell of ringing that occurs at the output of the amplifier
A pulse with the opposite polarity to the first swell to cancel
Signal to the input of the amplifier only once per deflection data
Charged particle beam exposure apparatus having circuit means for adding
Place 13. The circuit means controls the ringing.
From the start of the first swell to the start of the second swell
The amplifier outputs a pulse signal having a pulse width corresponding to time or less
13. The charged particle beam dew of claim 12, which is added to the input of
Light equipment. 14. The circuit means comprises a pulse-shaped signal pattern.
Pulse generation timing, pulse amplitude and pulse width
13. It is possible to variably set one, if not at all.
3. The charged particle beam exposure apparatus according to item 3. 15. The circuit means comprises a pulse-shaped signal pattern.
The timing of pulse generation , pulse amplitude and pulse width
The variable setting is possible independently of each other.
The charged particle beam exposure apparatus according to claim 1. 16. The circuit means is stored in advance in a memory.
Pulse parameters for pulsed signals that are
The pulsed signal is generated based on the pulse signal.
Charged particle beam exposure system. 17. A memory and a pulse signal related signal.
A control hand that stores the loose parameters in the memory in advance.
The charged particle beam dew of claim 16, further comprising a step.
Light equipment. 18. The output of the amplifier is controlled by the control means.
After the change, the differential value of the output of the amplifier becomes 0 fastest.
The pulse parameters to be stored and stored in the memory.
The charged particle beam exposure apparatus according to claim 17. 19. The control means outputs the output of the amplifier.
The output of the amplifier stays within the specified range after it starts to change
To find the pulse parameter that minimizes the time to
18. The charged particle beam dew of claim 17, stored in a memory.
Light equipment. 20. The control unit is a charged particle on the sample.
A means for generating a detection signal indicating the movement of the daughter beam;
The fastest differentiation of the detection signal after the particle beam starts moving
The pulse parameter with a value of 0 is obtained and stored in the memory.
18. The charged particle bee according to claim 17, further comprising:
Exposure device. 21. The deflecting means has at least stages.
Having a multi-stage coil consisting of one coil,
Deflection data is supplied to the coil via an independent amplifier
The directions of the currents that drive the coils of adjacent stages
21. The method according to claim 12, further comprising means for setting in reverse.
The charged particle beam exposure apparatus according to claim 1. 22. The deflecting means is a multi-stage coil.
Providing more than one with the phase shifted with respect to the optical axis of the charged particle beam
22. A charged particle beam exposure apparatus according to claim 21, wherein
Place