JP2010072180A - Correction system and electron beam drawing apparatus - Google Patents

Abstract

A pattern is accurately drawn on a substrate by correcting the irradiation position of the electron beam irradiated on the substrate.
A rotation error of a rotary table is measured based on a detection signal from a first detector and a detection signal from a second detector. Then, by calculating using the detection signal from the first detector and the detection signal from the second detector, the incident position error of the electron beam is calculated, and the electron beam is calculated based on the calculation result. The incident position is corrected. Thereby, it becomes possible to make an electron beam enter into a desired position accurately.
[Selection] Figure 1

Description

  The present invention relates to a correction system and an electron beam drawing apparatus, and more particularly to a correction system for correcting an incident position of an electron beam and an electron beam drawing apparatus for drawing a pattern on a sample.

  In recent years, with the digitization of information, there has been an increasing demand for larger capacity of optical disks and hard disks. Instead of conventional optical disks such as CD (Compact Disk) and DVD (Digital Versatile Disk), for example, the wavelength is about 400 nm. Research and development of next-generation type optical discs that record and reproduce information by ultraviolet light and patterned media capable of high-density recording are actively conducted.

  In the manufacturing process of next-generation optical disc masters (stampers) and patterned media recording media, the pattern formed on the recording layer is finer than that of conventional optical discs. For example, an electron beam drawing apparatus capable of drawing a spiral or concentric fine pattern with ultrafine wires of 0.05 μm or less on the surface of the substrate by irradiation is used.

  In this type of apparatus, the electron beam must be accurately incident on a target position on the substrate. However, the incident position of the electron beam may deviate from the target position depending on the rotation error of the rotary table on which the substrate is placed, the rotational speed unevenness, or the like. Therefore, various methods for causing an electron beam to accurately enter a target position on a substrate have been proposed.

JP 2005-50084 A JP 2002-92977 A NTN TECHNICICAL REVIEW No. 69

  In the apparatus described in Patent Document 1, the phase error is detected by comparing the phase of the output signal from the rotary encoder and the output signal from the VCO in the write clock PLL circuit. However, with this device, the output value from the phase meter is output as a voltage value. However, since it is necessary to set the full scale of the output voltage within the range of the phase difference ± 2π, it is difficult to increase the phase detection sensitivity. is there. When a 16-bit AD converter is used, the resolution (1LSB) is about 200 μrad, which is equivalent to a little less than 10 μm when converted on the recording surface. Furthermore, the phase conversion noise is about 50 μrad, and it is not possible to generate a correction signal for rotation unevenness in the order of nm.

  Moreover, in the apparatus described in Patent Document 2, the rotational shake is measured using the side surface of the turntable on which the sample is mounted. However, even if the shape error value of the turntable is accurately measured in advance, the shape error changes when the turntable is actually attached to the spindle. For this reason, it is extremely difficult to measure the rotational shake of the turntable in the order of nm taking into account the shape error of the turntable. In addition, the rotational shake must be measured in consideration of the accuracy of the spindle that rotates the turntable, and it is difficult to accurately measure the rotational shake in the order of nm with the conventional method.

  The present invention has been made under the circumstances described above, and a first object thereof is to provide a correction system capable of accurately correcting the incident position of an electron beam incident on a sample.

  A second object of the present invention is to provide an electron beam drawing apparatus that can accurately draw a pattern on a sample by correcting the incident position of the electron beam.

  According to a first aspect of the present invention, there is provided a correction system for correcting fluctuations in the incident position of an electron beam due to a rotation error of a rotary table, and detecting two rotation angles of different rotary axes provided on the rotary table. A first detector, a second detector, and a measuring device for measuring the rotation error based on detection signals from the first detector and the second detector; A deflecting device that deflects the electron beam based on a measurement result of the measuring device and corrects an incident position of the electron beam on a sample placed on the rotary table.

  According to this, the rotation error of the rotary table is measured based on the detection signal from the first detector and the detection signal from the second detector. Then, the calculation using the detection signal from the first detector and the detection signal from the second detector is performed, so that the influence of the mounting error between the first detector and the second detector is affected. The excluded rotation error is measured, and the incident position of the electron beam is corrected based on the measurement error. Therefore, the electron beam can be incident on the desired position with high accuracy.

  According to a second aspect of the present invention, there is provided an electron beam drawing apparatus that draws a pattern on the sample by irradiating the rotating sample with an electron beam, the rotary table holding the sample rotatably; An electron beam drawing apparatus comprising: a rotation mechanism that rotates the rotation table around a rotation axis of the rotation table; and the correction system of the present invention.

  According to this, since the electron beam is incident on a desired position on the sample by correcting the incident position of the electron beam by the correction system of the present invention, a pattern is accurately drawn on the sample as a result. It becomes possible.

Hereinafter, an embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows a schematic configuration of an electron beam drawing apparatus 200 according to the present embodiment. The electron beam drawing apparatus 200 irradiates an electron beam to a substrate W coated with a resist material in an environment where the degree of vacuum is about 10 ⁻⁴ Pa, for example, and draws a fine pattern on the drawing surface of the substrate W. A line drawing device.

  As shown in FIG. 1, the electron beam drawing apparatus 200 accommodates an irradiation apparatus 10 that irradiates a substrate with an electron beam, a rotary table unit 30 that includes a rotary table 31 on which the substrate is placed, and a rotary table unit 30. A vacuum chamber 40 and a control system for controlling each of the above parts are provided.

  The vacuum chamber 40 is a rectangular parallelepiped hollow member, and a circular opening is formed on the upper surface.

  The rotary table unit 30 is disposed on the bottom wall surface inside the vacuum chamber 40. The rotary table unit 30 includes a rotary table 31 on which the substrate W is placed, a spindle motor 35 that rotates the rotary table 31 at a predetermined rotational speed around a rotary shaft 31a substantially parallel to the vertical axis, and a rotary table 31. A moving stage 34 that is movably held in the X-axis direction, a slide unit 33 that drives the moving stage 34 in the X-axis direction with a predetermined stroke, and a lower end portion and an upper end portion of a rotating shaft 31 a formed on the rotary table 31. A first encoder 36A and a second encoder 36B are provided for measuring the rotation angle, respectively.

  The irradiation device 10 includes a casing 11 whose longitudinal direction is the Z-axis direction, and an electron gun 12, a magnetic lens 13, a blanking electrode 14, and an aperture member 15, which are sequentially arranged from the inside to the bottom of the casing 11. , A scanning electrode 16 and an objective lens 17.

  The casing 11 is a cylindrical casing that is open at the bottom, and is fitted into an opening formed on the upper surface of the vacuum chamber 40 from above without any gap. And the part located in the inside of the vacuum chamber 40 is a taper shape in which the diameter becomes small toward -Z direction.

  The electron gun 12 is disposed inside the casing 11. The electron gun 12 is a thermal field emission type electron gun that emits electrons extracted from the cathode by heat and an electric field, and emits, for example, an electron beam having a diameter of about 20 to 50 nm downward (−Z direction).

  The magnetic field lens 13 is an annular lens disposed below the electron gun 12, and acts on the electron beam emitted downward from the electron gun 12 in a focusing direction.

  The blanking electrode 14 has a pair of rectangular plate-like electrodes arranged so as to face each other at a predetermined interval in the X-axis direction, and according to a voltage applied by the control device 70, a magnetic lens The electron beam that has passed through 13 is deflected in the + X direction as indicated by the dotted line in the figure.

  The aperture member 15 is a plate-like member provided with an opening through which an electron beam passes in the center. The aperture member 15 is arranged so that the opening is located near the point where the electron beam that has passed through the blanking electrode 14 converges.

  The scanning electrode 16 is disposed below the aperture member 15. The scan electrode 16 has a pair of electrodes arranged so as to face each other in the X-axis direction and a pair of electrodes arranged so as to face each other in the Y-axis direction. Depending on the voltage, the electron beam that has passed through the aperture member 15 is deflected in the X-axis direction or the Y-axis direction.

  The objective lens 17 is disposed below the scanning electrode 16 and converges the electron beam that has passed through the scanning electrode 16 onto the surface of the substrate placed on the rotary table 31.

  In the irradiation apparatus 10 configured as described above, the electron beam emitted from the electron gun 12 is focused by passing through the magnetic lens 13 and is near the opening (hereinafter referred to as a crossover point) provided in the aperture member 15. ) Is once crossed. Next, the beam diameter of the electron beam that has passed the crossover point is shaped by passing through the aperture member 15 while diverging. Then, the light is converged on the surface of the substrate W placed on the rotary table 31 by the objective lens 17.

  In the irradiation apparatus 10, the blanking electrode 14 is controlled in parallel with the above operation to deflect the electron beam in the X-axis direction, thereby shielding the electron beam with the aperture member 15, and Blanking can be done. Further, the irradiation position of the electron beam on the substrate W can be adjusted by controlling the voltage applied to the scanning electrode 16 and deflecting the electron beam in the X-axis direction or the Y-axis direction. Yes.

  The control system includes a basic clock generation circuit 100, a formatter 101, a track direction beam irradiation position correction circuit 102, a feed direction beam irradiation position correction circuit 103, a pulse generation system 104, a blanker control circuit 105, a track direction deflection control circuit 106, The feed direction deflection control circuit 107, the rotation error measurement circuit 108, the rotation drive control circuit 109, the lateral feed drive control circuit 110, and the like are configured.

  The basic clock generation circuit 100 generates a basic clock signal CLK having a predetermined frequency that defines the operation of each part constituting the control system.

  The blanker control circuit 105 generates a blanking signal that is a voltage signal based on the signal supplied from the formatter 101 and supplies the blanking signal to the blanking electrode 14. Thereby, the electron beam is modulated based on the pattern information drawn on the substrate W.

  The pulse generation system 104 generates a signal Fclk and a signal dFclk including error information with respect to the signal Fclk, a signal Tclk, and the signal Tclk for forming a CAV format information track on the surface of the substrate W in the CLV driving state. A signal dTclk including error information with respect to, a signal Sclk, and a signal dSclk including error information with respect to the signal dSclk are generated and output, respectively. FIG. 2 is a block diagram illustrating an example of the pulse generation system 104. As shown in FIG. 2, the pulse generation system 104 includes a frequency division data latch circuit 120, a frequency division circuit 121, a delay circuit 122, a delay pulse selection circuit 123, and a delay pulse selection data latch circuit 124. . In the pulse generation system 104 configured as described above, for example, the basic clock CLK is frequency-divided by the frequency divider circuit 121, and the frequency-divided pulse obtained as a result is input to the delay circuit 122. Then, a delay pulse group having a minute time resolution is generated by the delay circuit 122 and supplied to the delay pulse selection circuit 123. The delay pulse selection circuit 123 selects a pulse closest to a desired generation time from the delay pulse group as an output pulse, and outputs it as a signal Fclk, a signal Tclk, and a signal Sclk. Further, the frequency division data and the delay pulse selection data are updated for each output pulse.

  The formatter 101 generates a signal including pattern information drawn on the substrate W based on the basic clock signal CLK and the signal Fclk from the pulse generation system 104 and outputs the signal to the blanker control circuit 105.

  The rotational drive control circuit 109 measures the rotational speed of the lower end portion of the rotary shaft 31a formed on the rotary table 31 via the first encoder 36A, and determines the spindle based on the signal Tclk from the pulse generator system 104. The motor 35 is driven. Thereby, the substrate W is rotated at a desired number of rotations.

  The lateral feed drive circuit 110 measures the position of the rotary table 31 in the X-axis direction via the interferometer 37 and moves the moving stage 34 via the slide unit 33 based on the signal Sclk from the pulse generator system 104. Is moved in the X-axis direction. Thereby, the rotary table 31 moves in the X-axis direction so that the deviation between the target position of the rotary table 31 included in the position command information and the measured position of the rotary table 31 measured via the interferometer 37 becomes zero. Is done. Further, the lateral feed drive circuit 110 outputs deviation information including a deviation between the target position of the rotary table 31 and the actually measured position to the feed direction beam irradiation position correction circuit 103.

  As shown in FIG. 3, the rotation error measurement circuit 108 includes a first high-speed clock counter 108a, a second high-speed clock counter 108b, a first latch circuit 108c, a second latch circuit 108d, and a correction amount calculation circuit 108e. It is configured.

  The first high-speed clock counter 108a receives the basic clock signal CLK from the basic clock generation circuit 100 and the pulse signal Tfb from the first encoder 36A. FIG. 4 shows the basic clock signal CLK and the pulse signal Tfb. The first high-speed clock counter 108a synchronizes with the origin signal ORG from the first encoder 36A, and calculates the number of pulses of the clock signal CLK in units of one cycle of the pulse signal Tfb from the rising of the pulse signal Tfb. Counting is performed, and data ΔT1 (n) including a difference between the count number and the reference count number is supplied to the first latch circuit 108c. Note that the period of the pulse signal increases and decreases depending on, for example, the inclination and the degree of eccentricity of the rotating shaft 31a. For example, when the period A in FIG. 5 is the reference period, the number of counts decreases in the period B that is shorter than the reference period. In the period C, which is longer than the reference period, the number of counts increases.

  Similarly, the second high speed clock counter 108b receives the basic clock signal CLK from the basic clock generation circuit 100 and the pulse signal Mfb from the second encoder 36B. Then, the second high-speed clock counter 108b counts the number of pulses of the clock signal CLK in units of one cycle of the pulse signal in synchronization with the origin signal ORG from the first encoder 36A. The data ΔT2 (n) including the difference is supplied to the second latch circuit 108d.

  The first latch circuit 108c and the second latch circuit 108d store the data ΔT1 (n) and ΔT2 (n), respectively, and sequentially supply them to the correction amount calculation circuit 108e.

  The correction amount calculation circuit 108e performs a calculation using the data ΔT1 (n) and ΔT2 (n), thereby correcting the correction amount r · Δθx in the X-axis direction (feed direction) of the electron beam incident on the sample, Error data for generating a correction amount r · Δθy in the Y-axis direction (track direction) is calculated.

  The rotation error data caused by the spindle motor included in the outputs from the first encoder 36A and the second encoder 36B is data in which the rotation shaft shake error and the rotation speed unevenness error are mixed. Among these, the rotational axis shake error causes an incident position shift in the track direction (Y-axis direction) and the feed direction (X-axis direction), and the rotational speed unevenness error causes an incident position shift in the track direction. The correction amount calculation circuit 108e generates data for calculating a rotation axis shake error, that is, a correction amount in the feed direction, and data for calculating a correction amount in the track direction.

<Measurement of rotation error>
First, when measuring the rotation error, before starting drawing on the substrate W, read error data based on the eccentricity of the first encoder 36A and the second encoder 36B is created and stored in the memory. Further, the position of the rotational center P of the rotational axis shake motion of the rotational shaft 31 a formed in advance on the rotary table 31 is measured. FIG. 5 is a diagram for explaining how to obtain the rotation center P of the axis collapse of the rotation shaft 31a. As shown in FIG. 5, the rotating shaft 31a continues to rotate in a state where the rotating shaft 31a is tilted by a predetermined tilt angle.

  The deviation ΔT1 (n) and the deviation ΔT2 (n) between the period of the pulse signal from each of the first encoder 36A and the second encoder 36B and the reference time Tcb are converted into angles to obtain the deviation ΔT1 (n). The rotation center P can be obtained from the amplitude B and the amplitude A of the curve indicating the change of the deviation ΔT2 (n). For example, when the distances between the two encoders and the rotation center P are S and R, respectively, the relationship between the amplitudes B and A of the measurement rotary encoder is expressed by the following equation (1).

A: B = R: S (1)

  Thus, if the rotation center P is measured in advance, the amplitude ratio between the two encoders 36A and 36B can be grasped in advance. Further, by arranging the two encoders 36A and 36B symmetrically with respect to the rotation center, the phase difference between the output waveforms from the two rotary encoders can be set to 180 degrees.

  Next, the first encoder 36A and the second encoder 36B are aligned. In this alignment, the detection heads of the encoders 36A and 36B are finely adjusted so that the rotation angle of the rotary shaft 31a can be detected simultaneously by the first encoder 36A and the second encoder 36B.

  In the measurement of the rotation error, it is necessary to detect the direction of the rotation shaft 31a and the rotation center of the taste mashing motion for each rotation angle of the rotation shaft 31a. Regarding the detection of the rotation shaft runout, data ΔT1 (n) (n = 1 to N) calculated based on the pulse signal Tfb and the pulse signal Mfb from the first encoder 36A and the second encoder 36B, and ΔT2 ( n) (n = 1 to N) is used. Note that the measurement data of the rotational speed unevenness measured at the same time is detected only by the same value by both encoders 36A and 36B. In other words, the rotational speed unevenness component can be canceled by taking the difference between the two measurement data. Based on the rotation center measured in advance, only the component of the rotational shaft runout can be separated and extracted.

  Next, a position error in the track direction and a position error in the feed direction are calculated from the separated rotation shaft run-out error and rotation speed unevenness error.

  FIG. 6A is a diagram showing a change in data ΔT1 (n), and FIG. 6B is a diagram showing a change in data ΔT2 (n). When the data ΔT1 (n) and the data ΔT2 (n) are supplied, the correction amount calculation circuit 108e first subtracts the data T1 (n) from the data ΔT2 (n) to thereby obtain the data shown in FIG. ΔT12 (n) is calculated.

  Since the first encoder 36A and the second encoder 36B simultaneously measure the angle of the rotating shaft 31a, the rotational speed unevenness components that enter the measurement value are temporally equal. Therefore, by taking the difference between the data ΔT1 (n) and the data ΔT2 (n), the influence of the speed unevenness can be eliminated. Furthermore, since the phase difference between the data ΔT1 (n) and the data ΔT2 (n) is 180 degrees, the amplitude of the curve indicating the change in the data ΔT2 (n) is represented by A + B.

Next, the correction amount calculation circuit 108e calculates the amplitude A of the data ΔT2 (n) and generates error data in the track direction. The distance R and the distance S from the measurement position of the rotary shaft 31a by the first encoder 36A and the second encoder 36B to the axis fall rotation center P of the rotary shaft 31a are known. Since the ratio between the amplitude of the change curve of data ΔT1 (n) and the amplitude of the change curve of data ΔT2 (n) is known, the sum A + B of the amplitudes of both change curves is expressed by the following equation (2). That is, by dividing the data ΔT12n shown in FIG. 7A by (1 + S / R), the rotational axis shake error data ΔT2 ₁ (n) in the track direction can be generated. The value of the error data ΔT2 ₁ (n) changes as shown by the change curve in FIG.

A + B = A + S × A / R = A (1 + S / R) (2)

Next, the correction amount calculation circuit 108e generates rotation axis shake error data ΔT2 ₂ (n) in the feed direction. The magnitude of the rotation axis run-out error in the feed direction is the same amplitude as that of the rotation axis run-out error in the track direction, that is, the change curve of the error data ΔT2 ₁ (n), and the waveform is delayed by 90 degrees in time. Therefore, the change curve of the error data ΔT2 ₂ (n) is as shown in FIG.

In addition, the rotational speed unevenness (cogging) of the rotating shaft 31a changes as shown by a curve shown in FIG. This curve is obtained by subtracting ΔT2 ₁ (n) from ΔT2 (n). However, this data is not actually used for generating the correction signal.

The error data ΔT2 ₁ (n) and error data ΔT2 ₂ (n) calculated as described above are supplied to the track direction beam irradiation position correction circuit 102 and the feed direction beam irradiation position correction circuit 103, respectively.

The track direction beam irradiation position correction circuit 102 calculates a correction amount r · Δθy in the Y-axis direction (track direction) based on the error data ΔT2 ₁ (n) supplied from the rotation error measurement circuit 108, and calculates this calculation. A control signal including the result is output to the track direction deflection control circuit 106. The correction amount r · Δθy in the Y-axis direction is obtained by multiplying the data ΔT2 ₁ (n) by the rotation angular velocity of the rotation shaft 31a and the distance r from the rotation center of the rotation table 31 to the incident position of the electron beam. Is calculated by The correction amount r · Δθy does not take into account the asynchronous error. However, when the asynchronous error is taken into consideration, the asynchronous amount ΔY in the Y-axis direction is added to the correction amount r · Δθy to obtain the correction amount. It is good. Asynchronous errors can be detected, for example, by the method disclosed in Japanese Patent Application Laid-Open No. 2005-274550. The correction amount r · Δθy is calculated for each rotation angle of the rotation shaft 31a.

  The track direction deflection control circuit 106 generates and scans a signal Bt for deflecting the electron beam in the track direction (Y-axis complementary direction) based on the control signal output from the track direction beam irradiation position correction circuit 102. Output to the electrode 16. As a result, the electron beam is deflected in the Y-axis direction, and the incident position of the electron beam on the substrate W is corrected in the Y-axis direction by an amount corresponding to the signal Bt.

The feed direction beam irradiation position correction circuit 103 calculates the (feed direction) correction amount r · Δθx in the X-axis direction based on the error data ΔT2 ₂ (n) supplied from the rotation error measurement circuit 108, and calculates this calculation. A control signal including the result is output to the feed direction deflection control circuit 107. The correction amount r · Δθx in the X-axis direction is obtained by multiplying the data ΔT2 ₂ (n) by the rotation angular velocity of the rotation shaft 31a and the distance r from the rotation center of the rotation table 31 to the incident position of the electron beam. Is calculated by The correction amount r · Δθx does not take into account the asynchronous error. However, when the asynchronous error is taken into consideration, the asynchronous amount ΔX in the X-axis direction is added to the correction amount r · Δθy to obtain the correction amount. It is good. Furthermore, the feed error information ΔS from the lateral feed drive control circuit 110 may be added to the correction amount r · Δθx, and this may be used as the correction amount. The correction amount r · Δθy is calculated for each rotation angle of the rotation shaft 31a.

  The feed direction deflection control circuit 107 generates a signal Bs for deflecting the electron beam in the feed direction (X-axis complementary direction) based on the control signal supplied from the feed direction beam irradiation position correction circuit 103, and performs scanning. Output to the electrode 16. As a result, the electron beam is deflected in the X-axis direction, and the incident position of the electron beam on the substrate W is corrected in the X-axis direction by an amount corresponding to the signal Bs.

  In the electron beam drawing apparatus 200 described above, for example, when a drawing start command is received from a user or a host device, the moving stage 34 is driven by the lateral feed drive control circuit 110 of the control system, and the rotation table is rotated by the rotation drive control circuit 109. 31 is rotated at a predetermined rotational speed. Then, a concentric or spiral pattern is formed on the surface of the substrate W by irradiating the surface of the substrate W with the electron beam modulated based on the drawing pattern from the irradiation device 10.

  While the substrate W is being irradiated with the electron beam, the rotation error measurement circuit 108, the track direction beam irradiation position correction circuit 102, and the feed direction beam irradiation position correction circuit 103 allow the X-axis direction and the Y-axis of the electron beam. The incident position deviation in the direction is calculated, and the incident position of the electron beam is corrected based on the calculation result. Therefore, a pattern is drawn on the substrate W with high accuracy.

  In general, the error component resulting from the rotation error of the rotation shaft 31a includes a rotation shaft run-out error, rotation speed unevenness (cogging), and asynchronous rotation error.

《Rotational shaft runout error》
As the spindle motor bearing structure, a hydrostatic bearing is generally used because high rotational accuracy can be obtained. However, the rotational accuracy largely depends on the processing accuracy of the bearing, and a rotational shaft runout error occurs. Moreover, the magnitude | size changes with load inertia and there exists rotational speed dependence. As the number of rotations increases, the rotation shaft tends to tilt more and the rotation shaft deflection error tends to increase. In particular, if there is an imbalance in the moment of inertia of the rotating body, the inclination of the rotating shaft also increases. This exercise is sometimes called miso soup exercise. As a result, the miso-scrubbing motion changes the direction of the tilt of the sample surface in synchronism with the rotation period of the spindle, so that the electron beam is drawn to meander on the sample.

  FIG. 9A is a diagram illustrating the axis collapse of the rotating shaft. The rotation axis continues to rotate with the tilt angle α tilted at the rotation center P with respect to the center axis. FIG. 9B shows the locus of the axis on the rotary table surface during one rotation of the rotary shaft. The axis locus is circular with a diameter of about 0.2 μm, and the rotation axis makes one round on this locus while the rotation axis makes one rotation. FIG. 9C shows the trajectory of the electron beam drawn on the substrate.

《Uneven rotation speed (cogging)》
When an air spindle is used, an AC servo motor is generally used as a drive motor. In the AC servo motor, it is difficult to completely eliminate the rotational speed unevenness due to the influence of the number of poles of the motor unit. FIG. 10 is a diagram for explaining the influence of cogging on uneven rotation speed. In the case of the present embodiment, the rotational speed unevenness occurs at a cycle of 4 peaks per rotation. If the rotation speed is uneven, the drawing is performed four times in one rotation of the spindle at the drawing position in the track direction.

《Asynchronous rotation error》
A non-repetitive component rotation error occurs due to an unsteady component of the gas flow in the hydrostatic bearing gap, vibration from the motor, lack of dynamic angular rigidity of the bearing, and the like. As described above, when a non-rotation synchronization error occurs, an asynchronous drawing position shift of the drawing position occurs in the track direction and the feed direction.

  FIG. 11 shows an error correction control system in which error data of the rotary encoder is incorporated so as to correct the feedback signal, and is disclosed in Reference 3. In addition to a normal PLL control system, a memory in which error data is recorded, a reading circuit that extracts error data corresponding to the rotational position, and a subtracter that subtracts the error data from the output of the deviation detection circuit are added. ing. The phase comparator outputs a pulse train having a width corresponding to the phase difference between the reference pulse and the output pulse of the rotary encoder. This pulse output is converted into a voltage proportional to the phase difference by the deviation detection circuit. As a result, it is possible to obtain a signal from which an error due to eccentricity is removed from the measured value of the control rotary encoder used as a feedback signal for controlling the rotation of the spindle, so that the rotation accuracy of the spindle can be improved.

  Further, the measurement error due to the eccentric eccentricity of the rotary encoder can be measured by the reference comparison method or the time method disclosed in Patent Document 3. FIG. 12 is a diagram showing an encoder error measurement method. As shown in the figure, the reference comparison method is a method in which a high-precision rotary encoder (reference encoder) whose accuracy is guaranteed in advance is attached to the top of the turntable via a coupling, and sequentially compared with the readings of the control encoder. It is a way to go. The time method is a method in which a constant current is supplied to the motor in an open loop and the output pulse of the encoder is measured in a state where the rotation is stable. The fact that the spindle motor can be regarded as rotating at a constant speed under open-loop control is utilized due to the moment of inertia of the turntable. In the present embodiment, the measurement error caused by the eccentricity of the first encoder 36A and the second encoder 36B can be measured in advance and stored in the storage means.

  As described above, the correction system of the present invention is suitable for correcting the incident position of the electron beam. The electron beam drawing apparatus of the present invention is suitable for drawing a pattern on a substrate.

It is a figure which shows schematic structure of the electron beam drawing apparatus 200 which concerns on one Embodiment of this invention. 2 is a block diagram of a pulse generation system 104. FIG. 3 is a block diagram of a rotation error measurement circuit 108. FIG. It is a figure which shows a clock signal and the pulse signal from encoder 36A, 36B. It is a figure for demonstrating how to obtain | require the rotation center P of the axis fall of the rotating shaft. 6A and 6B are diagrams (part 1 and part 2) for explaining the operation of the correction amount calculation circuit 108e. FIGS. 7A to 7C are diagrams (Nos. 3 to 5) for explaining the operation of the correction amount calculation circuit 108e. It is a figure for demonstrating the rotational speed nonuniformity of the rotating shaft 31a. FIG. 9A is a diagram illustrating the axis collapse of the rotating shaft. FIG. 9B is a diagram showing the locus of the axial center on the rotary table surface. FIG. 9C is a diagram showing the trajectory of the electron beam drawn on the substrate. It is a figure for demonstrating the influence which it has on rotation speed nonuniformity by cogging. It is a figure which shows an example of an error correction control system. It is a figure for demonstrating the error measuring method of an encoder.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Irradiation device, 11 ... Casing, 12 ... Electron gun, 13 ... Magnetic lens, 14 ... Blanking electrode, 15 ... Aperture member, 16 ... Scanning electrode, 17 ... Objective lens, 30 ... Rotary table unit, 31 ... Rotary table 31a ... rotating shaft, 33 ... slide unit, 34 ... moving stage, 35 ... spindle motor, 36A ... first encoder, 36B ... second encoder, 37 ... interferometer, 40 ... vacuum chamber, 100 ... basic clock generation circuit, DESCRIPTION OF SYMBOLS 101 ... Formatter, 102 ... Track direction beam irradiation position correction circuit, 103 ... Feed direction beam irradiation position correction circuit, 104 ... Pulse generation system, 105 ... Blanker control circuit, 106 ... Track direction deflection control circuit, 107 ... Feed direction deflection Control circuit 108 ... Rotation error measurement circuit 108a ... First high speed Lock counter, 108b ... second high-speed clock counter, 108c ... first latch circuit, 108d ... second latch circuit, 108e ... correction amount calculation circuit, 109 ... rotation drive control circuit, 110 ... cross feed drive control circuit, 120 ... minutes Peripheral data latch circuit 121... Frequency divider circuit 122... Delay circuit 123. Delay pulse selection circuit 124. Delay pulse selection data latch circuit 200.

Claims (6)

A correction system for correcting an incident position of an electron beam due to a rotation error of a rotary table,
A first detector and a second detector for detecting rotation angles at two different rotation axes provided on the rotary table;
A measuring device for measuring the rotation error based on detection signals from the first detector and the second detector;
A deflection system comprising: a deflection device that deflects the electron beam based on a measurement result of the measurement device to correct an incident position of the electron beam on a sample placed on the rotary table. The measuring device is
A basic clock generator for generating a basic clock signal;
A first counter circuit for counting the number of clocks of the basic clock signal per cycle of the pulse signal from the first detector;
A second counter circuit for counting the number of clocks of the basic clock signal per cycle of the pulse signal from the second detector;
And a calculation circuit for calculating the rotation error from data obtained by comparing a count result of the first counter circuit and a count result of the second counter circuit with a reference value. Item 2. The correction system according to Item 1.   The first detector detects a rotation angle at a position near the rotary table on one side of the rotation shaft, and the second detector detects a rotation angle at a position on the other side of the rotation shaft. The correction system according to claim 1 or 2, wherein   The correction system according to any one of claims 1 to 3, wherein the rotation error includes a rotation shaft fluctuation and a rotation speed unevenness of the rotary table. An electron beam drawing apparatus for irradiating a rotating sample with an electron beam and drawing a pattern on the sample,
A rotary table for rotatably holding the sample;
A rotation mechanism for rotating the rotary table about the rotation axis of the rotary table;
An electron beam drawing apparatus comprising: the correction system according to claim 1. A moving stage for holding a sample rotated by the rotating mechanism;
The electron beam drawing apparatus according to claim 5, further comprising: a moving mechanism that moves the moving stage in a predetermined direction.