Multi-charged particle beam writing apparatus and multi-charged particle beam writing method

The multi-charged particle beam lithography apparatus addresses defective blanking control circuits in multi-beam systems by using a diagnostic section to identify and restrict defective registers, ensuring accurate beam control and preventing pattern errors.

JP2024157192AInactive Publication Date: 2024-11-07NUFLARE TECH INC
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2023071375
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-25
Publication Date
2024-11-07
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In multi-beam lithography systems, defective blanking control circuits in the blanking aperture array mechanism can lead to uncontrollable beams, resulting in pattern errors due to defective beams that are always ON or OFF, which are difficult to correct.

Method used

A multi-charged particle beam lithography apparatus and method that includes a blanking aperture array substrate with control electrodes and counter electrodes, a register for inputting blanking control signals, and a diagnostic section to detect defective registers, generating flag signals to restrict the operation of defective circuits, ensuring beams are either always OFF or controlled properly.

Benefits of technology

Prevents defective beams from entering the sample, maintaining pattern accuracy by ensuring proper blanking control even when individual beam control circuits are defective.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024157192000001_ABST
    Figure 2024157192000001_ABST
Patent Text Reader

Abstract

To prevent a defective beam from being incident on a sample even when a circuit for taking in transferred blanking control data is defective in a blanking aperture array mechanism in multi-beam writing.SOLUTION: A multi-charged particle beam writing device includes a plurality of control circuits 41 each having a register for inputting a blanking control signal and an amplifier for applying a beam ON potential or a beam OFF potential to a corresponding one of a plurality of control electrodes in accordance with the blanking control signal input to the register, a deflection control circuit for holding a flag signal including an indication of whether the register is a normal register or a defective register, and a plurality of registers 45 for inputting each generated flag signal via a data path at a different time from the blanking control signal, and for restricting the operation of the control circuit in which the corresponding register is located when the input flag signal is a defective flag signal.SELECTED DRAWING: Figure 11
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a multi-charged particle beam drawing apparatus and a multi-charged particle beam drawing method, and more particularly to an irradiation circuit method for a defective beam in multi-beam drawing. [Background technology]

[0002] Lithography technology, which is responsible for the advancement of miniaturization of semiconductor devices, is an extremely important process in semiconductor manufacturing that is the only one that generates patterns. In recent years, with the increasing integration of LSIs, the circuit line width required for semiconductor devices has become finer year by year. Here, electron beam (electron beam) drawing technology has inherently excellent resolution, and drawing on wafers and the like is carried out using electron beams.

[0003] For example, there is a lithography device that uses multiple beams. Compared to lithography using a single electron beam, the use of multiple beams allows many beams to be irradiated at once, greatly improving throughput. In such a lithography device using the multi-beam method, for example, an electron beam emitted from an electron gun is passed through a mask with multiple holes to form multiple beams, each of which is blanked, and each beam that is not blocked is reduced in size by an optical system and deflected by a deflector to be irradiated at a desired position on a sample.

[0004] In multi-beam lithography, the dose of each beam is individually controlled by the irradiation time. In order to control the dose of each beam with high precision, it is necessary to perform blanking control, which turns the beam on and off, at high speed. In a multi-beam lithography device, a blanking control circuit for each beam is installed in the blanking aperture array mechanism in which each blanker of the multi-beam is arranged.

[0005] Here, in the blanking aperture array mechanism, since the LSI circuit is arranged between the beam passing holes, degradation of the LSI circuit occurs due to radiation caused by X-rays, scattered electrons, etc. In addition, there may be cases where the circuit is malfunctioning due to manufacturing defects. As a data transfer means to the LSI circuit for each beam of the blanking aperture array mechanism, for example, a matrix scan method or a shift register method in which a plurality of shift registers arranged for each row of the beam array are connected in series to transfer data can be mentioned. In either method, a register is used as a circuit to capture the ON / OFF data of the transferred beam in the LSI circuit for each beam. A typical mode of circuit degradation due to radiation is a change in the switching threshold voltage of a transistor. When such degradation occurs, the switching speed slows down, and an error occurs in setting data to a register composed of transistors and in reading data. If the circuit that captures data specifying the ON / OFF (blanking) state of the beam in the LSI circuit is defective, it becomes difficult to control the operation of the circuit that applies the blanking potential. As a result, defective beams that are always OFF, defective beams with uncontrollable blanking, or defective beams that are always ON occur, causing pattern errors on the written sample. Of these defective beams, defective beams with uncontrollable blanking and defective beams that are always ON are difficult to correct when they are incident on the sample.

[0006] Here, although it is not a technology of a blanking aperture array mechanism, a microprocessor that isolates a faulty register has been disclosed. In such a microprocessor, information of a faulty register among n registers is input from a control circuit and stored in a storage device, and a selection signal of the n registers is switched to a redundant register selection signal according to the value stored in the storage device (for example, see Patent Document 1). [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Publication No. 4-010033 Summary of the Invention [Problem to be solved by the invention]

[0008] One aspect of the present invention provides a multi-charged particle beam drawing apparatus and a multi-charged particle beam drawing method that can prevent defective beams that are unable to be blanked and defective beams that are always ON from being incident on a sample, even if a circuit that incorporates transferred blanking control data in a control circuit for individual beams is faulty in a blanking aperture array mechanism in multi-beam drawing. [Means for solving the problem]

[0009] According to one aspect of the present invention, there is provided a multi-charged particle beam writing apparatus, a multi-beam forming mechanism for forming multiple charged particle beams; a blanking aperture array substrate having a plurality of openings formed at positions through which the multi-charged particle beam passes; a plurality of control electrodes disposed on a blanking aperture array substrate in the vicinity of the plurality of openings; a plurality of counter electrodes disposed on a blanking aperture array substrate so as to face one of the plurality of control electrodes across different openings among the plurality of openings; a plurality of control circuits each including a register arranged on the blanking aperture array substrate and inputting a blanking control signal transmitted via a data path in the blanking aperture array substrate, and an amplifier applying a beam-on potential or a beam-off potential to a corresponding one of the plurality of control electrodes in accordance with the blanking control signal inputted to the register; a storage unit for storing flag signals, the flag signals being generated for each of the plurality of control circuits and including either a normal flag signal for identifying a register as a normal register or a defective flag signal for identifying a register as a defective register; a plurality of operation limiting circuits which input the generated flag signals through the same data path through which the blanking control signal is transmitted at a time different from that of the blanking control signal, and which limit the operation of a control circuit in which a register corresponding to the defective flag signal is disposed when the input flag signal is a defective flag signal; a limiting aperture substrate that blocks a beam that is blanked and deflected by applying a beam-OFF potential to a control electrode; A multi-charged particle beam writing apparatus comprising:

[0010] Also, a diagnostic unit for diagnosing the register; a flag data generating unit that generates a normal flag signal or a defective flag signal using a diagnosis result of the register; It is preferable that the sensor further comprises:

[0011] In addition, when a defect flag signal is input, each of the multiple operation limiting circuits preferably fixes the output of a beam amplifier that uses a register corresponding to the defect flag signal to a beam OFF potential at all times.

[0012] It is also preferable that each of the plurality of operation limiting circuits receives the flag signal via the same data path through which the blanking control signal is transmitted and via a register of a corresponding one of the plurality of control circuits.

[0013] In addition, a shift register is used as the register, The multi-charged particle beam is divided into multiple groups of multiple beams, Multiple shift registers for multiple beams in the same group are connected in series, It is preferable that the circuit further comprises a selection circuit which connects, among the multiple serially connected shift registers, the shift register in its own control circuit to the shift register in the adjacent control circuit downstream in the data transfer direction, and which connects to the shift register in the adjacent control circuit upstream in the data transfer direction and the shift register in the adjacent control circuit downstream in the data transfer direction so as to skip the shift register in its own control circuit, and which selects whether data transfer skips the shift register in its own control circuit or not in response to a flag signal.

[0014] The blanking aperture array substrate has a multi-beam passing region at a central portion through which the multi-charged particle beam passes, and an outer peripheral region at an outer periphery, The control circuits are disposed in the multi-beam passing region; The plurality of operation limiting circuits are preferably arranged in the outer periphery region.

[0015] Another aspect of the present invention is a multi-charged particle beam writing apparatus, a multi-beam forming mechanism for forming multiple charged particle beams; a blanking aperture array substrate having a plurality of openings through which the multi-charged particle beam passes; a plurality of control electrodes disposed on a blanking aperture array substrate in the vicinity of the plurality of openings; a plurality of counter electrodes disposed on a blanking aperture array substrate so as to face one of the plurality of control electrodes across different openings among the plurality of openings; a plurality of control circuits, each of which includes a first register and a second register arranged in parallel and which input a blanking control signal transmitted through a data path in the blanking aperture array substrate, and an amplifier which applies a beam-on potential or a beam-off potential to a corresponding one of the plurality of control electrodes in accordance with the blanking control signal input to either one of the first register or the second register; a storage unit for storing flag signals, the flag signals being generated for each of the plurality of control circuits, the flag signals including either a normal flag signal for identifying that the first register is a normal register or a defective flag signal for identifying that the first register is a defective register; a plurality of switching circuits that input the generated flag signal at a different time from the blanking control signal through the same data path through which the blanking control signal is transmitted and through the first register of a corresponding one of the plurality of control circuits, and when the input flag signal is a defective flag signal, switch a connection to the amplifier side from the first register that input the flag signal to a second register; a limiting aperture substrate that blocks a beam that is blanked and deflected by applying a beam-OFF potential to a control electrode; The present invention is characterized by comprising:

[0016] A multi-charged particle beam writing method according to one aspect of the present invention includes the steps of: a step of diagnosing a plurality of registers, which are arranged in a blanking aperture array substrate for blanking and deflecting the multiple charged particle beams, and which input blanking control signals for individual beams of the multiple charged particle beams, to detect a defective register; a step of generating a flag signal including either a normal flag signal for identifying a register that is not a defective register or a defective flag signal for identifying a defective register, inputting the flag signal at a time different from that of the blanking control signal through the same data path through which the blanking control signal is transmitted, and restricting the operation of an amplifier for a beam that uses a register corresponding to the defective flag signal when the input flag signal is a defective flag signal; a step of applying a beam-on potential or a beam-off potential from an individual amplifier for each beam to a control electrode arranged in the vicinity of an opening formed in the blanking aperture array substrate through which the beam passes, in accordance with a blanking control signal input to an individual register for each beam transmitted through a data path in the blanking aperture array substrate, shielding the blanking-deflected beam due to a potential difference between the control electrode and an opposing electrode opposing the control electrode as a result of the beam-off potential being applied to the control electrode, and drawing a pattern on a sample using a group of beams among the multi-charged particle beams that have not been shielded; The present invention is characterized by comprising: Effect of the Invention

[0017] According to one aspect of the present invention, in a blanking aperture array mechanism for multi-beam writing, even if a circuit that imports transferred blanking control data within a control circuit for an individual beam is faulty, it is possible to prevent defective beams that are out of blanking control and defective beams that are always ON from being incident on a sample. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 is a conceptual diagram showing a configuration of a drawing device in a first embodiment. [Diagram 2] 2 is a conceptual diagram showing a configuration of a shaping aperture array substrate in the first embodiment. FIG. [Diagram 3] 2 is a cross-sectional view showing a configuration of a blanking aperture array mechanism in the first embodiment. FIG. [Figure 4] 4 is a top conceptual diagram showing a part of the configuration within a membrane region of the blanking aperture array mechanism in the first embodiment. FIG. [Diagram 5] FIG. 2 is a diagram illustrating an example of an individual blanking mechanism according to the first embodiment. [Figure 6] FIG. 2 is a diagram showing an example of a connection configuration of a shift register in the first embodiment. [Figure 7]FIG. 2 is a conceptual diagram for explaining an example of a drawing operation in the first embodiment. [Figure 8] 3 is a diagram showing an example of a region irradiated with multiple beams and a pixel to be written in the first embodiment. FIG. [Figure 9] FIG. 2 is a diagram for explaining an example of a multi-beam writing method according to the first embodiment. [Figure 10] FIG. 2 is a diagram showing an example of divided shots of a multi-beam in the first embodiment. [Figure 11] FIG. 2 is a diagram showing an example of a control circuit for individual beams arranged inside a blanking aperture array mechanism in the first embodiment. [Figure 12] FIG. 11 is an example of a timing chart illustrating the beam ON / OFF switching operation of beam 1 for each divided shot in the first embodiment. [Figure 13] FIG. 2 is a flow chart showing main steps of an example of a writing method according to the first embodiment. [Figure 14] 10 is an example of a timing chart illustrating an operation of a control circuit for beam 1 regarding flag signal transfer in the first embodiment. [Figure 15] 11 is a diagram for explaining a data flow in a case where a shift register in its own control circuit is not skipped in the first embodiment. FIG. [Figure 16] 11 is a diagram for explaining a data flow when a shift register in its own control circuit is skipped in the first embodiment. FIG. [Figure 17] FIG. 11 is a diagram showing an example of a control circuit for individual beams arranged inside a blanking aperture array mechanism in the second embodiment. [Figure 18] FIG. 13 is an example of a timing chart illustrating the beam ON / OFF switching operation of the beam 1 for each divided shot in the second embodiment. [Figure 19] 13 is an example of a timing chart illustrating the operation of the control circuit for beam 1 regarding flag signal transfer in the second embodiment. [Figure 20]FIG. 11 is a diagram showing an example of a control circuit for individual beams and an example of an operation limiting circuit arranged inside a blanking aperture array mechanism in the third embodiment. [Figure 21] 13 is a diagram for explaining a control circuit in which an amplifier output is limited by a defect flag signal in the third embodiment and the comparative example. FIG. [Figure 22] FIG. 13 is a diagram showing an example of a control circuit for individual beams arranged inside a blanking aperture array mechanism in the fourth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] In the following, in the embodiments, a configuration using an electron beam as an example of a charged particle beam will be described. However, the charged particle beam is not limited to an electron beam, and a beam using charged particles such as an ion beam may be used. In addition, in the embodiments, a configuration using a drawing apparatus as an example of an exposure apparatus will be described.

[0020] Embodiment 1 FIG. 1 is a conceptual diagram showing a configuration of a drawing apparatus in the first embodiment. In FIG. 1, the drawing apparatus 100 includes a drawing mechanism 150 and a control circuit 160. The drawing apparatus 100 is an example of a multi-charged particle beam drawing apparatus and an example of a multi-charged particle beam exposure apparatus. The drawing mechanism 150 includes an electron lens barrel 102 (electron beam column) and a drawing chamber 103. In the electron lens barrel 102, an electron gun 201, an illumination lens 202, a shaping aperture array substrate 203, a blanking aperture array mechanism 204, a reduction lens 205, a limiting aperture substrate 206, an objective lens 207, a main deflector 208, a sub-deflector 209, and a common blanking deflector 212 are arranged. In the drawing chamber 103, an XY stage 105 is arranged. On the XY stage 105, a sample 101 such as a mask that becomes a substrate to be drawn during drawing (exposure) is arranged. The sample 101 includes an exposure mask used in manufacturing a semiconductor device, a semiconductor substrate (silicon wafer) on which a semiconductor device is manufactured, and the like. The sample 101 also includes a mask blank on which a resist is applied and on which nothing is yet drawn. A mirror 210 for measuring the position of the XY stage 105 is further disposed on the XY stage 105. A Faraday cup 106 is further disposed on the XY stage 105.

[0021] The control system circuit 160 includes a control computer 110, a memory 112, a deflection control circuit 130, a logic circuit 131, digital-to-analog conversion (DAC) amplifiers (units) 132 and 134, a lens control circuit 136, an amplifier 137, a stage control mechanism 138, a stage position measurement device 139, and storage devices 140 and 142 such as magnetic disk devices. The control computer 110, the memory 112, the deflection control circuit 130, the lens control circuit 136, the amplifier 137, the stage control mechanism 138, the stage position measurement device 139, and the storage devices 140 and 142 are connected to each other via a bus (not shown). The deflection control circuit 130 is connected to the logic circuit 131, the DAC amplifiers (units) 132 and 134, and a blanking aperture array mechanism 204. The sub-deflector 209 is composed of four or more electrodes, and each electrode is controlled by the deflection control circuit 130 via the DAC amplifier 132. The main deflector 208 is composed of four or more electrodes, and each electrode is controlled by the deflection control circuit 130 via a DAC amplifier 134. The common blanking deflector 212 is composed of two or more electrodes, and is connected to the logic circuit 131. The stage position measurement device 139 measures the position of the XY stage 105 based on the principle of laser interferometry by receiving reflected light from the mirror 210. The Faraday cup 106 is connected to an amplifier 137.

[0022] In the control computer 110, a defective register detection unit 60, an irradiation time calculation unit 62, a blanking data generation unit 64, a flag data generation unit 66, a clock control unit 70, an amplifier operation control unit 72, a detection unit 76, a transfer control unit 79, and a drawing control unit 80 are arranged. Each of the "~ units" such as the defective register detection unit 60, the irradiation time calculation unit 62, the blanking data generation unit 64, the flag data generation unit 66, the clock control unit 70, the amplifier operation control unit 72, the detection unit 76, the transfer control unit 79, and the drawing control unit 80 has a processing circuit. Such a processing circuit includes, for example, an electric circuit, a computer, a processor, a circuit board, a quantum circuit, or a semiconductor device. Each of the "~ units" may use a common processing circuit (the same processing circuit) or may use different processing circuits (separate processing circuits). Information input / output to / from the defective register detection unit 60, the irradiation time calculation unit 62, the blanking data generation unit 64, the flag data generation unit 66, the clock control unit 70, the amplifier operation control unit 72, the detection unit 76, the transfer control unit 79, and the drawing control unit 80, as well as information being calculated, are stored in the memory 112 each time.

[0023] Furthermore, drawing data is input from outside the drawing device 100 and stored in the storage device 140 .

[0024] 1 shows the configuration necessary for explaining the embodiment 1. The drawing device 100 may also include other configurations that are normally required.

[0025] FIG. 2 is a conceptual diagram showing the configuration of the shaping aperture array substrate in the first embodiment. In FIG. 2, holes (openings) 22 are formed in a matrix of p columns (y direction)×q columns (x direction) (p, q≧2) at a predetermined arrangement pitch in the shaping aperture array substrate 203. In FIG. 2, for example, 512×512 columns of holes 22 are formed in the vertical and horizontal directions (x, y directions). Each hole 22 is formed as a rectangle of the same size and shape. Alternatively, it may be a circle of the same diameter. A part of the electron beam 200 passes through each of these multiple holes 22, thereby forming a multibeam 20. In other words, the shaping aperture array substrate 203 is an example of a multibeam forming mechanism, which forms a multibeam. Here, an example in which holes 22 are arranged in two or more rows in both the vertical and horizontal directions (x, y directions) is shown, but this is not limited to this. For example, it is also possible to have multiple rows in either the vertical or horizontal direction (x, y direction) and only one row in the other direction. Furthermore, the arrangement of holes 22 is not limited to the case where the holes are arranged vertically and horizontally in a lattice pattern as in Fig. 2. For example, the holes in the kth column in the vertical direction (y direction) and the k+1th column may be arranged to be shifted by a dimension a in the horizontal direction (x direction). Similarly, the holes in the k+1th column in the vertical direction (y direction) and the k+2th column may be arranged to be shifted by a dimension b in the horizontal direction (x direction).

[0026] FIG. 3 is a cross-sectional view showing a configuration of a blanking aperture array mechanism in the first embodiment. FIG. 4 is a top view conceptual diagram showing a part of the configuration in the membrane area of ​​the blanking aperture array mechanism in the first embodiment. In FIG. 3 and FIG. 4, the positional relationship between the control electrode 24, the counter electrode 26, the control circuit 41, and the pad 343 is not shown to be consistent. In the blanking aperture array mechanism 204, as shown in FIG. 3, a blanking aperture array substrate 31 using a semiconductor substrate made of silicon or the like is placed on a support base 333. The central part of the blanking aperture array substrate 31 is thinly scraped from the back side, for example, and processed into a membrane area 330 (first area) with a thin film thickness h. The periphery surrounding the membrane area 330 becomes an outer peripheral area 332 (second area) with a thick film thickness H. The upper surface of the membrane area 330 and the upper surface of the outer peripheral area 332 are formed to be at the same height position or substantially at the same height position. The substrate 31 is held on the support base 333 at the back side of the outer peripheral area 332. The center of the support base 333 is open, and the membrane region 330 is located in the open region of the support base 333 .

[0027] In the membrane region 330 of the blanking aperture array substrate 31, a plurality of openings are formed at positions through which the multibeams pass. Specifically, in the membrane region 330 of the blanking aperture array substrate 31, passage holes 25 (openings) for passing each beam of the multibeams are opened at positions corresponding to the holes 22 of the shaping aperture array substrate 203 shown in FIG. 2. In other words, in the membrane region 330 of the blanking aperture array substrate 31, a plurality of passage holes 25 through which the corresponding beams of the multibeams using electron beams pass are formed in an array. Then, on the membrane region 330 of the blanking aperture array substrate 31, a plurality of electrode pairs each having two electrodes are disposed at positions facing each other across the corresponding passage hole 25 among the plurality of passage holes 25. A plurality of control electrodes 24, which are one of the two electrodes, are disposed on the blanking aperture array substrate 31 near different passage holes 25 among the plurality of passage holes 25. A plurality of counter electrodes 26, which are the other of the two electrodes, are disposed on the blanking aperture array substrate 31 so as to face one of the plurality of control electrodes 24 across different ones of the plurality of through holes 25. Specifically, as shown in Figs. 3 and 4, pairs of blanking deflection control electrodes 24 and counter electrodes 26 (blankers: blanking deflectors) are disposed on the membrane region 330, across the corresponding through holes 25 in the vicinity of the respective through holes 25. Also, inside the blanking aperture array substrate 31 (in the blanking aperture array substrate), a control circuit 41 (logic circuit; cell) that applies a deflection voltage to the control electrode 24 for each through hole 25 is disposed in the vicinity of each through hole 25 on the membrane region 330. The counter electrodes 26 for each beam are connected to ground.

[0028] As shown in FIG. 4, each control circuit 41 is connected to n-bit (for example, 1-bit) parallel wiring for control signals. In addition to n-bit parallel wiring for blanking control signals (data), each control circuit 41 is connected to wiring for clock (shift clock) signals, load (preload, leg load, and flag load) signals, shot signals, and power supplies. A part of the parallel wiring may be used for the wiring for the clock (shift clock) signals, load (preload, leg load, and flag load) signals, shot signals, and power supplies. An individual blanking mechanism 47 is formed for each beam constituting the multi-beam by the control electrode 24, the counter electrode 26, and the control circuit 41. In the example of FIG. 3, the control electrode 24, the counter electrode 26, and the control circuit 41 are arranged in a membrane region 330 having a thin film thickness of the blanking aperture array substrate 31. However, this is not limited to this. In the first embodiment, a shift register method is used as a data transfer method. In the shift register method, the multi-beam is divided into a plurality of groups for each of the plurality of beams, and a plurality of shift registers for the plurality of beams in the same group are connected in series. Specifically, a plurality of control circuits 41 formed in an array in the membrane region 330 are grouped, for example, in the same row or the same column at a predetermined pitch. The control circuits 41 in the same group are connected in series as shown in FIG. 4. Then, a signal from a pad 343 arranged for each group is transmitted to the control circuits 41 in the group. Specifically, a shift register, which will be described later, is arranged in each control circuit 41, and, for example, the shift registers in the control circuits 41 for beams in the same row among p×q multi-beams are assigned at a predetermined pitch, grouped, and connected in series. Then, for example, the control signals for q beams in the same row of p (vertical)×q (horizontal) multi-beams are transmitted in series, and the control signals for each row are further divided into q' parallel signals. Then, for example, the control signals for each beam are stored in the corresponding control circuits 41 by q / q' clock signals.

[0029] FIG. 5 is a diagram showing an example of the individual blanking mechanism of the first embodiment. In FIG. 5, an amplifier 46 (an example of a switching circuit) is arranged in the control circuit 41. In the example of FIG. 5, a CMOS (Complementary MOS) inverter circuit is arranged as an example of the amplifier 46. The CMOS inverter circuit is connected to a positive potential (Vdd: blanking potential: first potential) (for example, 5V) (beam OFF potential) and a ground potential (GND: second potential) (beam ON potential). An output line (OUT) of the CMOS inverter circuit is connected to a control electrode 24. On the other hand, a ground potential is applied to the counter electrode 26. A plurality of control electrodes 24 to which a blanking potential and a ground potential are switchably applied are arranged on the blanking aperture array substrate 31 at positions facing the corresponding counter electrodes 26 of the plurality of counter electrodes 26 across the corresponding passing holes 25 of the plurality of passing holes 25.

[0030] To the input (IN) of the CMOS inverter circuit, either an L (low) potential (for example, ground potential) lower than the threshold voltage or an H (high) potential (for example, 1.5V) equal to or higher than the threshold voltage is applied as a control signal. In the first embodiment, when an L potential is applied to the input (IN) of the CMOS inverter circuit, the output (OUT) of the CMOS inverter circuit becomes a positive potential (Vdd), and the corresponding beam 20 is deflected by an electric field due to a potential difference with the ground potential of the counter electrode 26, and is controlled so that the beam is turned OFF by blocking with the limiting aperture substrate 206. On the other hand, when an H potential is applied to the input (IN) of the CMOS inverter circuit (active state), the output (OUT) of the CMOS inverter circuit becomes a ground potential, and there is no potential difference with the ground potential of the counter electrode 26, so the corresponding beam 20 is not deflected, and the beam is controlled so that it passes through the limiting aperture substrate 206 and is turned ON.

[0031] The electron beam 20 passing through each passage hole is deflected by voltages applied to two independent pairs of control electrodes 24 and counter electrodes 26. Blanking control is performed by such deflection. Specifically, the pairs of control electrodes 24 and counter electrodes 26 individually blank and deflect corresponding beams of the multi-beams by potentials switched by CMOS inverter circuits serving as corresponding switching circuits. In this way, the multiple blankers perform blanking deflection of the corresponding beams among the multi-beams that have passed through the multiple holes 22 (openings) of the shaping aperture array substrate 203.

[0032] FIG. 6 is a diagram showing an example of a connection configuration of a shift register in the first embodiment. The control circuits 41 for each beam are formed in an array in the membrane region 330. For example, for each of the multiple control circuits 41 (x direction) arranged in the same row, as shown in FIG. 6, the control circuit 41 columns in each row are sequentially distributed into, for example, eight groups and grouped. For example, in the case of a multi-beam configuration of 512 columns x 512 rows, the control circuits 41 for the 1st to 512th beams in each row form a data sequence 1 (group) at 1st, 9th, 17th, 25th, . . . , and every 8 beam pitches. Similarly, the control circuits 41 for the 2nd, 10th, 18th, 26th, . . . , and every 8 beam pitches form a data sequence 2 (group). Similarly, the control circuits 41 in each group are formed as data sequence 3 (group) to data sequence 8 (group). Then, the control circuits 41 in each group are connected in series. Then, the signal for each row is divided from an I / O circuit (not shown) and transmitted in parallel to each group. Then, the signal of each group is transmitted to the control circuits 41 connected in series in the group. Specifically, a shift register 11 is arranged in each control circuit 41, and the shift registers 11 in the control circuits 41 of the same group are connected in series. In the example of FIG. 6, 64 shift registers 11 are connected in series for each data column (group). Therefore, the blanking control signal (data) for each beam is transmitted to the shift register 11 for each beam by 64 clock signals. In the example of FIG. 6, a case where a plurality of groups are formed for each row (x direction) at a predetermined pitch is shown, but this is not limited to this. For example, it is preferable to divide the same row into a left half and a right half, and form a plurality of groups for each control circuit 41 for the beam in the left half of the same row at a predetermined pitch, and form another plurality of groups for each control circuit 41 for the beam in the right half of the same row at a predetermined pitch. As a result, 32 shift registers 11 are connected in series for each data column (group), and the number of clocks for data shifting can be halved. Alternatively, a plurality of groups may be formed for each of a plurality of control circuits 41 (y direction) arranged in the same row. Alternatively, a plurality of control circuits 41 arranged in the same row (or column) may form one group.In that case, for example, if the device is configured with a multi-beam of 512 columns x 512 rows, 512 clock cycles will be required to transfer the data required for one beam shot, but this can be applied if it is acceptable to sacrifice operating speed.

[0033] Next, a specific example of the operation of the drawing mechanism 150 will be described. An electron beam 200 emitted from an electron gun 201 (emission source) illuminates the entire shaping aperture array substrate 203 almost perpendicularly by an illumination lens 202. A plurality of rectangular holes 22 (openings) are formed in the shaping aperture array substrate 203, and the electron beam 200 illuminates an area including all of the plurality of holes 22. Each part of the electron beam 200 irradiated to the position of the plurality of holes 22 passes through each of the plurality of holes 22 of the shaping aperture array substrate 203, thereby forming, for example, a rectangular multi-beam (a plurality of electron beams) 20a to e. The multi-beam 20 passes through each corresponding blanker (first deflector: individual blanking mechanism 47) of the blanking aperture array mechanism 204. The blankers individually perform blanking control on the passing beams so that the beams are in an ON state for a set drawing time (irradiation time).

[0034] The multi-beam 20 that has passed through the blanking aperture array mechanism 204 is reduced by the reduction lens 205 and travels toward the central hole formed in the limiting aperture substrate 206. Here, the electron beams deflected by the blanker of the blanking aperture array mechanism 204 are displaced from the central hole of the limiting aperture substrate 206 and are blocked by the limiting aperture substrate 206. On the other hand, the electron beams that have not been deflected by the blanker of the blanking aperture array mechanism 204 pass through the central hole of the limiting aperture substrate 206 as shown in FIG. 1. In this way, the limiting aperture substrate 206 blocks each beam that has been deflected by the individual blanking mechanism 47 to be in the beam OFF state. Then, each beam for one shot is formed by the beams that have passed through the limiting aperture substrate 206 and that have been formed from when the beam is turned ON until when the beam is turned OFF. The multi-beams 20 that have passed through the limiting aperture substrate 206 are focused by the objective lens 207 to become a pattern image with a desired reduction ratio, and the entire multi-beams 20 that have passed through the limiting aperture substrate 206 are deflected in the same direction by the main deflector 208 and the sub-deflector 209, and each beam is irradiated onto its respective irradiation position on the sample 101. Also, for example, when the XY stage 105 is moving continuously, the main deflector 208 performs tracking control so that the irradiation position of the beam follows the movement of the XY stage 105. The multi-beams 20 irradiated at one time are ideally arranged at a pitch obtained by multiplying the arrangement pitch of the multiple holes 22 in the shaping aperture array substrate 203 by the desired reduction ratio described above.

[0035] FIG. 7 is a conceptual diagram for explaining an example of the writing operation in the first embodiment. As shown in FIG. 7, the writing region 30 of the sample 101 is virtually divided into a plurality of stripe regions 32 each having a predetermined width in the y direction. First, the XY stage 105 is moved to adjust the irradiation region 34 that can be irradiated with one shot of the multi-beam 20 to be located at the left end of the first stripe region 32 or at a position further to the left, and writing is started. When writing the first stripe region 32, the XY stage 105 is moved, for example, in the -x direction to relatively advance writing in the x direction. The XY stage 105 is moved continuously at a constant speed, for example. After the writing of the first stripe region 32 is completed, the stage position is moved in the -y direction to adjust the irradiation region 34 to be located at the right end of the second stripe region 32 or at a position further to the right in the y direction, and then the XY stage 105 is moved, for example, in the x direction to similarly perform writing in the -x direction. The writing time can be reduced by alternately changing the direction of writing, such as writing in the x direction in the third stripe region 32 and writing in the -x direction in the fourth stripe region 32. However, writing in the same direction may be performed in each stripe region 32, and is not limited to alternately changing the direction of writing. In one shot, a plurality of shot patterns, up to the same number as the number of holes 22, are formed at once by the multi-beams formed by passing through the holes 22 of the shaping aperture array substrate 203.

[0036] FIG. 8 is a diagram showing an example of a multibeam irradiation region and a pixel to be written in the first embodiment. In FIG. 8, the stripe region 32 is divided into a plurality of mesh regions in a mesh shape by the beam size of the multibeam, for example. Each mesh region is a pixel 36 to be written (unit irradiation region, irradiation position, or writing position). The size of the pixel 36 to be written is not limited to the beam size, and may be an arbitrary size regardless of the beam size. For example, the size may be 1 / n (n is an integer equal to or greater than 1) of the beam size. The example of FIG. 8 shows a case where the writing region of the sample 101 is divided into a plurality of stripe regions 32 in the y direction, for example, with a width substantially equal to the size of the irradiation region 34 (writing field) that can be irradiated by one irradiation of the multibeam 20. The size of the rectangular irradiation region 34 in the x direction can be defined by the number of beams in the x direction x the beam pitch in the x direction. The size of the rectangular irradiation region 34 in the y direction can be defined by the number of beams in the y direction x the beam pitch in the y direction. The width of the stripe region 32 is not limited to this. It is preferable that the size is n times (n is an integer of 1 or more) that of the irradiation region 34. In the example of FIG. 8, for example, the illustration of 512×512 rows of multi-beams is abbreviated to 8×8 rows of multi-beams. In addition, a plurality of pixels 28 (beam drawing positions) that can be irradiated by one shot of the multi-beam 20 are shown in the irradiation region 34. In other words, the pitch between adjacent pixels 28 is the pitch between each beam of the multi-beam. In the example of FIG. 7, one sub-irradiation region 29 is formed by a square region surrounded by four adjacent pixels 28 and including one pixel 28 out of the four pixels 28. In other words, one sub-irradiation region 29 is formed by a rectangular region including the pixel 28 and having a size of the beam pitch. In the example of FIG. 8, each sub-irradiation region 29 is formed by 4×4 pixels.

[0037] FIG. 9 is a diagram for explaining an example of a multi-beam writing method in the first embodiment. FIG. 9 shows a part of the sub-irradiation area 29 written by each beam of coordinates (1,3), (2,3), (3,3), . . . , (512,3) in the third row in the y direction among the multi-beams that write the stripe area 32 shown in FIG. 8. The example of FIG. 9 shows a case where, for example, four pixels in the same sub-irradiation area 29 are written (exposed) while the sub-deflector 209 shifts the irradiation position (pixel 36) in sequence while the XY stage 105 moves a distance of eight beam pitches. While writing (exposing) these four pixels, the main deflector 208 deflects the entire multi-beam 20 collectively so that the relative position of the irradiation area 34 with respect to the sample 101 does not shift due to the movement of the XY stage 105, thereby making the irradiation area 34 follow the movement of the XY stage 105. In other words, tracking control is performed. In the example of FIG. 9, one tracking cycle is performed by drawing (exposing) four pixels while moving a distance of eight beam pitches. The remaining pixels in the sub-irradiation area 29 are drawn by other beams. When one tracking cycle is completed, tracking is reset and the system returns to the previous tracking start position. Since drawing of the first pixel row from the right in each sub-irradiation area 29 is completed, the sub-deflector 209 first deflects (shifts) the drawing position of the beam so as to draw the second pixel row from the right in each sub-irradiation area 29 after tracking reset in the next tracking cycle. In the example of FIG. 9, for example, the beam 1 draws four pixels in the first row from the right in the sub-irradiation area 29. Next, the beam 9, which is eight beam pitches away, draws four pixels in the second row from the right in the sub-irradiation area 29. Next, the beam 17 (not shown), which is eight beam pitches away, draws four pixels in the third row from the right in the sub-irradiation area 29. Next, a beam 25 (not shown) eight beam pitches away draws four pixels in the fourth column from the right of the sub-irradiation area 29. This means that all 4×4 pixels in the sub-irradiation area 29 have become the drawing targets. Drawing is similarly performed in the other sub-irradiation areas 29 using multiple beams.By repeating this operation while drawing the stripe region 32, the position of the irradiation region 34 moves in sequence from irradiation regions 34a to 34o as shown in FIG.

[0038] FIG. 10 is a diagram showing an example of split shots of a multi-beam in the first embodiment. In FIG. 10, the maximum irradiation time Ttr of one shot for each pixel 36 is divided into, for example, n sub-irradiation times with different irradiation times. Then, for example, from the n split shots with different sub-irradiation times, a combination of multiple split shots is selected according to an arbitrary irradiation time of each shot. First, a gradation value Ntr is determined by dividing the maximum irradiation time Ttr by a quantization unit Δ (gradation value resolution). For example, when n=10, the shot is divided into 10 split shots. When the gradation value Ntr is defined as a binary value with n digits, it is preferable to set the quantization unit Δ in advance so that the gradation value Ntr=1023. This results in the maximum irradiation time Ttr=1023Δ. Then, as shown in FIG. 10, the n split shots are divided into two digits k'=0 to 9. k’ In other words, 512Δ(=2 9 Δ), 256Δ(=2 8 Δ), 128Δ(=2 7 Δ), 64Δ(=2 6 Δ), 32Δ(=2 5 Δ), 16Δ(=2 4 Δ),8Δ(=2 3 Δ), 4Δ(=2 2 Δ),2Δ(=2 1 Δ),Δ(=2 0In other words, one multi-beam shot is divided into a divided shot with an irradiation time tk' of 512Δ, a divided shot with an irradiation time tk' of 256Δ, a divided shot with an irradiation time tk' of 128Δ, a divided shot with an irradiation time tk' of 64Δ, a divided shot with an irradiation time tk' of 32Δ, a divided shot with an irradiation time tk' of 16Δ, a divided shot with an irradiation time tk' of 8Δ, a divided shot with an irradiation time tk' of 4Δ, a divided shot with an irradiation time tk' of 2Δ, and a divided shot with an irradiation time tk' of Δ.

[0039] Therefore, one shot of any irradiation time t (=NΔ) for each pixel 36 is, instead, 512Δ (=2 9 Δ), 256Δ(=2 8 Δ), 128Δ(=2 7 Δ), 64Δ(=2 6 Δ), 32Δ(=2 5 Δ), 16Δ(=2 4 Δ),8Δ(=2 3 Δ), 4Δ(=2 2 Δ),2Δ(=2 1 Δ), and Δ(=2 0 This can be satisfied by performing a combination of at least one divided shot selected from a set of divided shots having a sub-exposure time defined by Δ) and having a non-zero exposure time.

[0040] Here, in the blanking aperture array mechanism 204, as described above, degradation of the LSI circuit occurs due to radiation caused by X-rays, scattered electrons, etc. When the control circuit 41 is installed in the membrane region 330 as shown in FIG. 3, degradation of the LSI circuit due to radiation is likely to occur, but even when the control circuit is installed in the outer peripheral region 332, the possibility of degradation of the LSI circuit due to radiation remains. In addition, there may be a case where a circuit malfunctions due to a manufacturing defect or the manufacturing life of the LSI. In the shift register system shown in FIG. 6 as a data transfer means to the control circuit 41 for each beam in the blanking aperture array mechanism 204, the shift register 11 that takes in data deteriorates. If the shift register 11 is defective, it becomes difficult to control the operation of the amplifier 46 that applies a blanking potential such as a beam ON potential (GND) or a beam OFF potential (Vdd). As a result, a defective beam that is always OFF, a defective beam that cannot be blanked, or a defective beam that is always ON occurs, and a pattern error occurs on the written sample 101. Among these defective beams, if a defective beam that cannot be blanked or a defective beam that is always ON among them enters a sample, it is difficult to correct the beam. Therefore, in the first embodiment, a defective shift register 11 is detected, and in the control circuit 41 in which the defective shift register 11 is arranged, the amplifier 46 is configured to apply a beam OFF potential to the control electrode 24 at all times, regardless of the contents of the blanking control signal received by the shift register 11.

[0041] FIG. 11 is a diagram showing an example of a control circuit for individual beams arranged inside the blanking aperture array mechanism in the first embodiment. In FIG. 11, in each control circuit 41, a shift register 11, two stages of registers 42 and 43 serving as buffers, an execution register 44, a register 45, an amplifier 46, a selection circuit 48, and an AND circuit 49 are arranged. In FIG. 11, the output of the shift register 11 is connected to the input of the register 42 and the input of the register 45. The output of the register 42 is connected to the input of the register 43. The output of the register 43 is connected to the input of the register 44. The output of the register 44 is connected to the input of the AND circuit 49. In addition, the output of the register 45 is connected to the input of the AND circuit 49. The input of the AND circuit 49 is further connected to a shot signal transmitted from the deflection control circuit 130. The output of the AND circuit 49 is connected to the input of the amplifier 46. The output of the amplifier 46 is connected to the control electrode 24 as shown in FIG. 5.

[0042] 11, the shift register 11 reads and stores data in synchronization with a shift clock signal transmitted from the deflection control circuit 130. Also, the stored data is transferred in synchronization with the shift clock signal to the shift register 11 in the next control circuit 41 of the same data string. The register 42 reads and stores data in synchronization with a preload 1 signal transmitted from the deflection control circuit 130. The register 43 reads and stores data in synchronization with a preload 2 signal transmitted from the deflection control circuit 130. The register 44 reads and stores data in synchronization with a leg load signal transmitted from the deflection control circuit 130. The register 45 reads and stores data in synchronization with a flag load signal transmitted from the deflection control circuit 130.

[0043] In the example of FIG. 11, it is preferable to provide the selection circuit 48, but it may be omitted. In the case where the selection circuit 48 is provided, in the example of FIG. 11, the output of the shift register 11 is further connected to one of two inputs of the selection circuit 48. The other of the two inputs of the selection circuit 48 is connected to the upstream data line of the shift register 11. The output of the selection circuit 48 is connected to the input of the next shift register 11 connected in series. Also, the selection circuit 48 selects one of the two inputs according to the signal stored in the register 45. Also, it goes without saying that when the selection circuit 48 is omitted, the output of the shift register 11 is connected to the input of the next shift register 11 connected in series.

[0044] As described above, each of the multiple control circuits 41 has a shift register 11 and an amplifier 46. The shift register 11 is disposed on the blanking aperture array substrate 31, and inputs a blanking control signal transmitted via a data path in the blanking aperture array substrate 31. The amplifier 46 then applies a beam-on potential or a beam-off potential to a corresponding one of the multiple control electrodes 24 in accordance with the blanking control signal input to the shift register 11. A specific description will be given below.

[0045] 12 is an example of a timing chart showing the beam ON / OFF switching operation of the beam 1 for each divided shot in the first embodiment. In the example of FIG. 12, one shot with the maximum irradiation time Ttr is, for example, 32Δ(=2 5 Δ), 16Δ(=2 4 Δ),8Δ(=2 3 Δ), 4Δ(=2 2 Δ),2Δ(=2 1 Δ), and Δ(=2 0 In the example of FIG. 12, the shot order of the divided shot groups is, for example, 32Δ(=2 5 Δ) split shot, Δ(=2 0 Δ) split shot, 16Δ(=2 4Δ) split shot, 2Δ(=2 1 Δ) split shot, 8Δ(=2 3 Δ) split shots, and 4Δ(=2 2Δ) in the order of division shots. The blanking control signal (data) of each beam transmitted through the data path in the blanking aperture array substrate 31 is transferred in synchronization with the shift clock signal transmitted from the deflection control circuit 130 by the number of the shift registers 11 connected in series in the same group, and is stored in the desired shift register 11. The blanking control signal (data) is binary controlled, and is an H signal or an L signal. The blanking control signal (data) stored in the shift register 11 is read into the register 42 in synchronization with the preload 1 signal transmitted from the deflection control circuit 130 and stored. The blanking control signal (data) stored in the register 42 is read into the register 43 in synchronization with the preload 2 signal transmitted from the deflection control circuit 130 and stored. After the data of the next division shot is stored in the shift register 11 in synchronization with the shift clock signal transmitted by the number of the shift registers 11 connected in series, the blanking control signal (data) stored in the shift register 11 is read into the register 42 in synchronization with the preload 1 signal transmitted from the deflection control circuit 130 and stored. Therefore, in a state where the blanking control signal (data) of the first divided shot is stored in the register 43, the blanking control signal (data) of the second divided shot is stored in the register 42. The blanking control signal (data) stored in the register 43 is read and stored in the register 44 in synchronization with the leg load signal transmitted from the deflection control circuit 130. As a result, the blanking control signal (data) stored in the register 44 is input to the input of the AND circuit 49. For example, an H signal if it is a beam ON signal, or an L signal if it is a beam OFF signal is input from the register 44 to the AND circuit 49. On the other hand, for example, an H signal if it is a normal flag signal (enable) that identifies a normal beam, or an L signal if it is a defective flag signal (disable) that identifies a defective beam is input to the AND circuit 49 from the register 45.Then, the AND circuit 49 outputs a beam ON signal (H) to the amplifier 46 only during the period when the AND circuit 49 receives a shot signal (enable; H signal) transmitted from the deflection control circuit 130, if both the signals from the register 44 and the register 45 are H signals. Otherwise, the AND circuit 49 outputs a beam OFF signal (L) to the amplifier 46. The amplifier 46 applies a beam ON potential (GND) to the control electrode 24 only during the period when the AND circuit 49 outputs a beam ON signal (H). Otherwise, the amplifier 46 applies a beam OFF potential (Vdd) to the control electrode 24. This executes the first divided shot. Data for the next divided shot is sent to each register in sequence, and after the first divided shot is completed, the AND circuit 49 inputs the next shot signal (enable; H signal) to execute the second divided shot. Thereafter, each divided shot is executed in sequence.

[0046] Here, the AND circuit 49 outputs a beam-on signal (H) only during the period when the H signal is input from both the register 44 and the register 45 and the shot signal (enable; H signal) is input. In other words, the beam-on signal (H) is output only during the period when all three signals are H signals. Therefore, when the shift register 11 of the target control circuit 41 is defective, the AND circuit 49 inputs an L signal from the register 45, so that the amplifier 46 can constantly apply a beam-off potential (Vdd) to the control electrode 24 regardless of the signal from the register 44 and the shot signal. In order to perform such control, it is necessary to transfer individual flag signals for each beam, such as a normal flag signal (enable) or a defective flag signal (disable), from the deflection control circuit 130 to the individual registers 45 for each beam. In order to transfer the individual flag signals for each beam, it is difficult to arrange a data path connected to each control circuit 41 separately within the blanking aperture array mechanism 204 in a limited space. On the other hand, for the defective shift register 11, the data can be read and transferred accurately by slowing down the speed of the shift clock used when transferring the blanking control signal (data). Alternatively, the drive voltage of the control circuit 41 can be temporarily made higher than that during normal operation, so that the data can be read and transferred accurately. However, since increasing the drive voltage may speed up the progression of circuit degradation, a method of slowing down the shift clock speed is more preferable.

[0047] Therefore, in the first embodiment, either a normal flag signal for identifying that the shift register 11 being used is a normal register or a defective flag signal for identifying that the shift register 11 is a defective register is input at a different time from the blanking control signal (data) through the same data path through which the blanking control signal (data) is transmitted to each control circuit 41 within the blanking aperture array mechanism 204. Specifically, the operation is as follows.

[0048] The flag signal of each beam transmitted through the same data path as the blanking control signal (data) in the blanking aperture array substrate 31 is transferred in synchronization with a slow shift clock signal transmitted from the deflection control circuit 130 by the number of shift registers 11 connected in series in the same group, and then stored in the desired shift register 11. Alternatively, as described above, the flag signal may be transferred in synchronization with a shift clock signal having the same speed as the shift clock used when transferring the blanking control signal (data) while the driving voltage of the control circuit 41 is temporarily made higher than that during normal operation. After being transferred with the driving voltage increased, the flag signal is stored in the desired shift register 11. The flag signal stored in the shift register 11 is read into the register 45 in synchronization with a flag load signal transmitted from the deflection control circuit 130 and stored. This makes it possible to eliminate the need to arrange a separate data path in the blanking aperture array mechanism 204 for transferring individual flag signals for each beam.

[0049] Fig. 13 is a flow chart showing the main steps of an example of the writing method in the first embodiment. In Fig. 13, the writing method in the first embodiment performs a series of steps including a defective register detection step (S102), a flag data generation step (S104), a clock switching step (S106), a flag transfer step (S108), an amplifier disabling step (S110), a beam disabling confirmation step (S112), a register selection step (S114), a blanking data generation step (S116), a clock switching step (S118), and a writing step (S120). Note that when the arrangement of the selection circuit 48 in Fig. 11 is omitted, the register selection step (S114) is also omitted.

[0050] In the defective register detection step (S102), the defective register detection unit 60 (diagnostic unit) diagnoses the shift register 11. In other words, the defective register detection unit 60 diagnoses a plurality of shift registers 11 that are arranged in a blanking aperture array substrate 31 that blanks and deflects the multibeam 20 and input blanking control signals of individual beams of the multibeam 20, and detects a defective register. Specifically, the operation is as follows. The defective register detection unit 60 detects a defective register from among a plurality of shift registers 11 that are arranged in a blanking aperture array substrate 31 that blanks and deflects the multibeam 20 and input blanking control signals of individual beams of the multibeam 20. In other words, the defective register detection unit 60 detects a defective shift register 11 from among the shift registers 11 for each beam of the multibeam 20. Specifically, the drawing control unit 80 transfers a data string in which evaluation blanking control signals (data) for one shot (one divided shot) for each beam are arranged to the deflection control circuit 130. The data string in which the blanking control signals (data) for evaluation for each beam are arranged may be stored in the storage device 140 in advance. The deflection control circuit 130 outputs the data string in which the blanking control signals (data) for evaluation are arranged to the blanking aperture array mechanism 204, and transfers it to the shift register 11 for each beam using a shift clock signal during normal drawing processing. Next, using a shift clock signal that is sufficiently slower than the shift clock signal during normal drawing processing, the data string stored in each chain of the serially connected shift registers 11 is read out in order from the last shift register 11 connected in order. The read out data string is output to the defective register detection unit 60. The defective register detection unit 60 detects the presence or absence and position of erroneous data in the read out data string for each chain of the serially connected shift registers 11. The defective register detection unit 60 determines that the shift register 11 in which erroneous data is stored is defective.

[0051] Here, if a shift register 11 in the middle of the multiple shift registers 11 connected in series is defective, the data transferred to the subsequent normal shift registers 11 may also be incorrect. In other words, the data is unreliable. Therefore, when the selection circuit 48 is omitted, the defective register detection unit 60 detects the defective shift register 11 and the subsequent normal shift registers 11 as defective registers. On the other hand, when the selection circuit 48 is arranged, when the defective register detection unit 60 detects multiple serially connected shift registers 11 storing incorrect data, it first detects the most upstream shift register 11 as a defective register. Next, the most upstream shift register 11 is removed from the data transfer order using the selection circuit 48, and in a skipped state, a data string in which blanking control signals (data) for evaluation for each beam are arranged is similarly transferred to the shift registers 11 for each beam using the shift clock signal during normal drawing processing. At that time, it is preferable to transfer a data string in which the data for the skipped shift register 11 is omitted. Then, using a shift clock signal that is sufficiently slower than the shift clock signal during normal drawing processing, the data string stored in each chain of the serially connected shift registers 11 is read out in sequence, starting from the last shift register 11 in the connection order. For each chain of the serially connected shift registers 11, the defective register detection unit 60 detects the upstream shift register 11 in which incorrect data is stored as a defective register from the data strings read out. The same operation is repeated until the serially connected shift registers 11 storing incorrect data are no longer detected. This makes it possible to distinguish between defective shift registers 11 in the middle and normal shift registers 11 thereafter, among the serially connected shift registers 11.

[0052] In the flag data generating step (S104), the flag data generating unit 66 uses the diagnosis result (detection result of defective register) of the shift register 11 to generate, for each shift register 11, either a normal flag signal for identifying that the shift register 11 is a normal register or a defective flag signal for identifying that the shift register 11 is a defective register. Then, a data string is generated in which flag signals for one shot (one divided shot) for each beam are arranged. For example, an "H" signal is generated as the normal flag signal, and an "L" signal is generated as the defective flag signal. The flag data generating unit 66 may also function as the defective register detecting unit 60 described above.

[0053] The generated flag signals, which are generated for each of the multiple control circuits 41 and include either a normal flag signal for identifying that the shift register 11 is a normal register or a defective flag signal for identifying that the shift register 11 is a defective register, are temporarily stored in the memory 112 and are also transferred to and stored in the deflection control circuit 130. Specifically, the flag signals are stored in a storage device (not shown) in the deflection control circuit 130.

[0054] As a clock switching step (S106), the clock control unit 70 switches from a shift clock signal for normal drawing processing to a shift clock signal for flag transfer that is sufficiently slower than the shift clock signal for the drawing processing. Instead of the clock switching step (S106), a driving voltage switching step of switching the driving voltage to a higher level may be performed.

[0055] In the flag transfer step (S108), the transfer control unit 79 transfers to the deflection control circuit 130 a data string in which flag signals for one shot (one divided shot) for each beam are arranged.

[0056] FIG. 14 is an example of a timing chart showing the operation of the control circuit for beam 1 with respect to flag signal transfer in the first embodiment. The deflection control circuit 130 outputs a data string in which flag signals are arranged to the blanking aperture array mechanism 204. The blanking aperture array mechanism 204 transfers the flag signals to the shift registers 11 for each beam through the same data path through which the blanking control signal is transmitted, at a different time from the blanking control signal during normal drawing processing. The flag signals of each beam transmitted through the data path in the blanking aperture array substrate 31 are stored in the desired shift register 11 by being transferred in synchronization with the shift clock signal transmitted from the deflection control circuit 130 by the number of shift registers 11 connected in series in the same group, as shown in FIG. 14. Here, the shift clock signal transmitted from the deflection control circuit 130 is a sufficiently slowed shift clock signal. Or a high driving voltage is used. Since the shift clock signal is sufficiently slowed, the flag signals can be accurately read and transferred not only from the normal shift register 11 but also from the defective shift register 11.

[0057] As an amplifier disabling step (S110), under the control of the amplifier operation control section 72, the deflection control circuit 130 outputs a flag load signal to each control circuit 41. Each register 45 of the multiple registers 45 (operation limiting circuit) inputs and holds a flag signal, which is generated for each of the multiple control circuits 41 and includes either a normal flag signal for identifying that the shift register 11 is a normal register (a register that is not a defective register) or a defective flag signal for identifying that the shift register 11 is a defective register, at a time different from that of the blanking control signal via the same data path through which the blanking control signal is transmitted. In particular, in the first embodiment, each register 45 of the multiple registers 45 (operation limiting circuit) inputs the flag signal via the same data path through which the blanking control signal is transmitted and via the shift register 11 of the corresponding control circuit 41 among the multiple control circuits 41. Specifically, the flag signal stored in the shift register 11 is read into the register 45 in synchronization with the flag load signal transmitted from the deflection control circuit 130 and stored therein, as shown in FIG. 14. As a result, either a normal flag signal for identifying that the shift register 11 in the control circuit 41 is a normal register or a defective flag signal for identifying that the shift register 11 is a defective register is input to the register 45 in each control circuit 41. In the first embodiment, the flag signal is transferred to each control circuit 41 via the existing shift register 11, so that the flag signal can be transferred using the shift register method without adding a data path. In addition, by transferring the flag signal via the shift register 11, a transfer method such as the shift register method can be used without changing anything other than the clock speed, etc.

[0058] When the input flag signal is a defective flag signal, the register 45 restricts the operation of the control circuit 41 in which the shift register 11 corresponding to the defective flag signal is disposed. Here, when the input flag signal is a defective flag signal, the register 45 restricts the operation of the beam amplifier 46 using the shift register 11 corresponding to the defective flag signal. Specifically, the operation is as follows. When a flag signal is stored in each register 45, for example, an H signal is input from the register 45 to the AND circuit 49 if it is a normal flag signal (enable) that identifies a normal beam, and an L signal is input if it is a defective flag signal (disable) that identifies a defective beam. The AND circuit 49 outputs a beam ON signal (H) only during a period when it receives H signals from both the register 44 and the register 45 and also receives a shot signal (enable; H signal). In other words, it outputs a beam ON signal (H) only during a period when all three signals are H signals. Therefore, when a defective flag signal is input, each register 45 of the multiple registers 45 (operation restriction circuit) constantly fixes the output of the beam amplifier 46 using the shift register 11 corresponding to the defective flag signal to a beam OFF potential. Specifically, a failure flag (disable) signal is stored in the register 45 in the control circuit 41 in which the defective shift register 11 is arranged, so that the AND circuit 49 can output a beam OFF signal (L) to the amplifier 46. As a result, the operation of the amplifier 46 can be restricted so that the amplifier 46 always outputs a beam OFF potential (Vdd) as shown in FIG. 14. On the other hand, a normal flag signal (enable) is stored in the register 45 in the control circuit 41 in which the normal shift register 11 is arranged, so that the AND circuit 49 inputs an H signal from the register 45. Therefore, the output of the AND circuit 49 is controlled by the signal from the register 44 and the shot signal (enable; H signal), and it is possible to avoid restrictions from the register 45.

[0059] In the above example, the register 45 has both the function of a storage device that holds a flag signal and the function of an operation limiting circuit that limits the operation of the control circuit 41. However, the present invention is not limited to this. It is also preferable to provide the function of the storage device and the function of the operation limiting circuit separately. In other words, for example, a storage device (storage unit) such as a register may be arranged in the control circuit 41 in addition to the register 45 (operation limiting circuit). In this case, a storage device other than the register 45 holds flag signals generated for each of the multiple control circuits 41, including either a normal flag signal that identifies the shift register 11 as a normal register or a defective flag signal that identifies the shift register 11 as a defective register. Then, each of the multiple registers 45 inputs the generated flag signals through the same data path through which the blanking control signal is transmitted at a different time from the blanking control signal, and when the input flag signal is a defective flag signal, the register 45 restricts the operation of the control circuit 41 in which the shift register 11 corresponding to the defective flag signal is arranged.

[0060] In this way, the beam controlled by the control circuit 41 in which the defective shift register 11 is arranged can be invalidated.

[0061] As a beam invalidation confirmation step (S112), the drawing mechanism 150 confirms whether the beam controlled by the control circuit 41 in which the defective shift register 11 is arranged is turned OFF. Specifically, the blanking aperture array mechanism 204 controls the target beam to be turned ON and controls all other beams to be turned OFF. Then, the target beam is irradiated onto the Faraday cup 106 on the XY stage 105. For example, the beam with the maximum irradiation time is repeatedly shot. The output of the Faraday cup 106 is amplified by the amplifier 137 and output to the detection unit 76 as a digital signal. The detection unit 76 detects the amount of current measured by the Faraday cup 106. If the detected amount of current is zero or the same amount of current as when all beams are controlled to be OFF, it can be confirmed that the beam has been invalidated.

[0062] In the register selection step (S114), when the selection circuit 48 is arranged, the selection circuit 48 connects the shift register 11 in its own control circuit 41 to the shift register 11 in the control circuit 41 adjacent downstream in the data transfer direction among the multiple shift registers 11 connected in series. Furthermore, the selection circuit 48 is connected to the shift register 11 in the control circuit 41 adjacent upstream in the data transfer direction and the shift register in the control circuit 41 adjacent downstream in the data transfer direction so as to skip the shift register 11 in its own control circuit 41. In this way, the input of the selection circuit 48 has two systems, a first system connected to the shift register 11 in its own control circuit 41 and a second system connected to the shift register 11 in the control circuit 41 adjacent upstream in the data transfer direction. The output of the selection circuit 48 is connected to the shift register 11 in the control circuit 41 adjacent downstream in the data transfer direction. Then, the selection circuit 48 selects whether the data transfer skips the shift register 11 in its own control circuit 41 or does not skip the shift register 11 in the control circuit 41 according to the flag signal.

[0063] FIG. 15 is a diagram for explaining a data flow in the case where the shift register in the control circuit of the first embodiment is not skipped. FIG. 16 is a diagram for explaining the flow of data when skipping the shift register in the control circuit of the selection circuit 48 in the first embodiment. The selection circuit 48 reads out a flag signal from the register 45. For example, if the selection circuit 48 receives a normal flag signal (enable) for identifying a normal beam, an H signal is input to the selection circuit 48 from the register 45, and if the selection circuit 48 receives a defective flag signal (disable) for identifying a defective beam, an L signal is input to the selection circuit 48. When the selection circuit 48 receives a normal flag signal, as shown in FIG. 15, the selection circuit 48 selects the shift register 11 in the control circuit 41 of the first system among the two inputs of the selection circuit 48. Thus, normal operation is performed. On the other hand, when the selection circuit 48 receives a defective flag signal, the selection circuit 48 selects the shift register 11 in the control circuit 41 adjacent to the upstream side in the data transfer direction of the second system as shown in FIG. 16. As a result, the shift register 11 in the control circuit 41 of the selection circuit 48 is removed from the series connection among the multiple shift registers 11 connected in series in the same group, and the data to be transferred skips the shift register 11 in the control circuit 41 of the selection circuit 48 and is transferred to the next shift register 11.

[0064] In the blanking data generating step (S116), the blanking data generating unit 64 calculates an irradiation amount D for each pixel 36 to be irradiated to the pixel 36. The irradiation amount D may be calculated, for example, as a value obtained by multiplying a preset reference irradiation amount Dbase by a proximity effect correction irradiation coefficient Dp and a pattern area density ρ. In this way, it is preferable to obtain the irradiation amount D in proportion to the area density of the pattern calculated for each pixel 36. Regarding the proximity effect correction irradiation coefficient Dp, the drawing region (here, for example, the stripe region 32) is virtually divided into a plurality of proximity mesh regions (mesh regions for calculating proximity effect correction) in a mesh shape with a predetermined size. It is preferable to set the size of the proximity mesh region to about 1 / 10 of the range of influence of the proximity effect, for example, about 1 μm. Then, the drawing data is read from the storage device 140, and for each proximity mesh region, the pattern area density ρ' of the pattern to be arranged in the proximity mesh region is calculated.

[0065] Next, a proximity effect correction irradiation coefficient Dp for correcting the proximity effect is calculated for each proximity mesh region. Here, the size of the mesh region for calculating the proximity effect correction irradiation coefficient Dp does not need to be the same as the size of the mesh region for calculating the pattern area density ρ'. Also, the correction model of the proximity effect correction irradiation coefficient Dp and its calculation method may be the same as the method used in the conventional single beam writing method.

[0066] Then, the irradiation time calculation unit 62 calculates, for each pixel 36, the irradiation time t of the electron beam for making the calculated irradiation amount D incident on the pixel 36. The irradiation time t can be calculated by dividing the irradiation amount D by the current density J.

[0067] Next, the blanking data generating unit 64 selects, for each pixel 36, a combination of divided shots that matches the irradiation time t for the pixel 36 from a set of divided shots (shots) corresponding to a plurality of preset exposure times shown in FIG. 10. If the irradiation time for the pixel 36 is not zero, at least one divided shot will be selected. When selecting a divided shot, it is preferable to select from the divided shot side with the longest exposure time (irradiation time). The blanking data generating unit 64 generates, for example, an ON / OFF signal (blanking data) in binary so that the selected divided shot can be identified. In the example of FIG. 10, for example, for the pixel 36 with an irradiation time of 1023Δ, the blanking data for the pixel is "1111111111" because all divided shots are selected. Next, the blanking data generating unit 64 rearranges the blanking data in the shot order according to the drawing sequence, and also rearranges the blanking data in the data transfer order taking into account the arrangement order of the shift registers 11 of each group for each divided shot.

[0068] Here, the writing control unit 80 sets the writing sequence assuming that the disabled beam is not used. Also, when the selection circuit 48 is arranged, if there is data for the shift register 11 skipped by the selection circuit 48, the number of shift clocks and the number of shift registers 11 connected in series will not match, and data will not be stored in the desired shift register 11. Therefore, when the selection circuit 48 is arranged, the blanking data generating unit 64 shortens the data string by the amount of the skipped shift register 11 for the chain in which the shift register 11 skipped by the selection circuit 48 is located.

[0069] In the clock switching step (S118), the clock control unit 70 switches from the delayed shift clock signal for flag transfer to a shift clock signal for normal drawing processing. Note that, when a drive voltage switching step is performed instead of the clock switching step (S106), a drive voltage switching step of switching to a lower drive voltage is performed instead of the clock switching step (S118).

[0070] In the drawing step (S120), the drawing mechanism 150 draws a desired pattern on the sample 101 by irradiating the sample 101 with the multibeam 20 according to the blanking data. Here, a required dose is applied to each pixel 36 by a combination of a plurality of divided shots that are successively irradiated to the same pixel 36. At this time, for example, in each control circuit 41 of a plurality of control circuits 41 arranged in the blanking aperture array substrate 31, the shift register 11 inputs blanking data (blanking control signal) transmitted via a data path in the blanking aperture array substrate 31 to control blanking of the multibeam 20. Then, for each beam transmitted via the data path in the blanking aperture array substrate 31, a beam ON potential or beam OFF potential is applied from an individual amplifier 46 for each beam to a control electrode 24 arranged in the vicinity of a passage hole 25 formed in the blanking aperture array substrate 31 through which the beam passes, according to the blanking data input to the individual shift register 11. In other words, the amplifier 46 applies a beam-on potential or a beam-off potential to the corresponding one of the plurality of control electrodes 24 in accordance with the blanking data input to the shift register 11 .

[0071] The drawing mechanism 150 shields the beams that have been blanked and deflected by the potential difference between the control electrode 24 and the counter electrode 26 opposed to the control electrode 24 as a result of the beam-off potential being applied to the control electrode 24, and draws a pattern on the sample 101 using the beam group of the multi-beams 20 that have not been shielded. Specifically, it operates as follows. The multi-beams 20 that have passed through the blanking aperture array mechanism 204 are reduced by the reduction lens 205 and proceed toward the central hole formed in the limiting aperture substrate 206. Here, the beams that have been blanked and deflected by the beam-off potential being applied to the control electrode 24 are displaced from the central hole of the limiting aperture substrate 206 and are shielded by the limiting aperture substrate 206. The disabled beams are also blanked and deflected, and are therefore shielded by the limiting aperture substrate 206. On the other hand, the electron beams that are not deflected by the blanker of the blanking aperture array mechanism 204 and the common blanking deflector 212 pass through the hole in the center of the limiting aperture substrate 206 as shown in FIG. 1. Then, the beams that have passed through the limiting aperture substrate 206 from when the beam is turned on until when it is turned off form each beam for one shot. The multi-beams 20 that have passed through the limiting aperture substrate 206 are focused by the objective lens 207 to become a pattern image with a desired reduction ratio, and are imaged on the surface of the sample 101. At this time, the entire multi-beams 20 that have passed through the limiting aperture substrate 206 are deflected in the same direction by the main deflector 208 and the sub-deflector 209, and each beam is irradiated at its respective irradiation position on the sample 101.

[0072] In embodiment 1, an example is shown in which blanking control of each beam is performed using both beam ON / OFF control by each control circuit 41 for individual blanking control and beam ON / OFF control by a logic circuit 131 for common blanking control that collectively controls blanking of the entire multi-beam.

[0073] When each control circuit 41 for individual blanking control is arranged in the blanking aperture array mechanism 204, the arrangement space is narrow. Therefore, the amount of information of the control signal can be reduced to reduce the installation area of ​​the control circuit. For example, a 1-bit control signal is input and output. However, in each control circuit 41, it is difficult to arrange a circuit for high-speed response from the viewpoint of installation space. On the other hand, the logic circuit 131 for common blanking does not perform multiple different controls at the same time, but only one circuit for ON / OFF control is required, so a circuit for high-speed response can be arranged.

[0074] In the logic circuit 131, the ON / OFF switching of the individual blanking mechanism is performed after the voltage stabilization time (settling time) of the amplifier 46 has elapsed. This makes it possible to prevent beam irradiation at an unstable voltage at the start-up of the individual amplifier 46. Then, the logic circuit 131 turns OFF when the irradiation time of the target kth shot has elapsed. As a result, when both the individual amplifier 46 and the amplifier in the logic circuit 131 are ON, the actual beam turns ON and is irradiated onto the sample 101. Therefore, the ON time of the amplifier in the logic circuit 131 is controlled to be the irradiation time of the actual beam. However, by the logic circuit 131 controlling each control circuit 41 for individual blanking control, it is possible to control without using the common blanking deflector 212.

[0075] As described above, according to the first embodiment, in the blanking aperture array mechanism 204 in multi-beam writing, even if the circuit that takes in the transferred blanking data in the control circuit 41 for individual beams is faulty, it is possible to prevent a defective beam with uncontrollable blanking or a defective beam that is always ON from being incident on the sample 101.

[0076] Embodiment 2 In the first embodiment, a configuration using a shift register method in which a plurality of shift registers 11 arranged, for example, for each row of the beam array are connected in series for each group to transfer data as a data transfer means to the control circuit 41 for each beam of the blanking aperture array mechanism 204 has been described. In the second embodiment, a configuration using, for example, a matrix scan method will be described as a data transfer means to the control circuit 41 for each beam of the blanking aperture array mechanism 204. The configuration of the drawing device 100 in the second embodiment is the same as that in FIG. 1. Also, the flowchart of the drawing method in the second embodiment is the same as that in FIG. 13. However, in the second embodiment, the register selection step (S114) is omitted. Hereinafter, points that are not particularly described may be the same as those in the first embodiment.

[0077] FIG. 17 is a diagram showing an example of a control circuit for individual beams arranged inside the blanking aperture array mechanism in the second embodiment. In FIG. 17, two stages of registers 42 and 43 serving as buffers, an execution register 44, a register 45, an amplifier 46, and an AND circuit 49 are arranged in each control circuit 41. In FIG. 17, the output of the register 42 is connected to the input of the register 43 and the input of the register 45. The output of the register 43 is connected to the input of the AND circuit 49. The output of the register 44 is connected to the input of the AND circuit 49. The input of the AND circuit 49 is further connected to a shot signal transmitted from the deflection control circuit 130. The output of the AND circuit 49 is connected to the input of the amplifier 46. The output of the amplifier 46 is connected to the control electrode 24 as shown in FIG. 5. In the example of FIG. 17, each control circuit 41 is arranged in a matrix shape similar to the arrangement of the multi-beams 20. The input of the odd-numbered register 42 in each row is connected to data k-1 of the two data lines, and the input of the even-numbered register 42 is connected to data k-2 of the two data lines. k indicates the row number. In the example of FIG. 17, the input of the odd-numbered register 42 in the upper row is connected to data 1-1, and the input of the even-numbered register 42 is connected to data 1-2 of the two data lines. The input of the odd-numbered register 42 in the lower row is connected to data 2-1, and the input of the even-numbered register 42 is connected to data 2-2 of the two data lines. In addition, a preload 1 signal is connected to each group of registers 42 in each column. In addition, a flag load signal is connected to each group of registers 45 in each row. In addition, a shot signal is connected to each group of AND circuits 49 in each row.

[0078] FIG. 18 is an example of a timing chart showing the beam ON / OFF switching operation of the beam 1 for each divided shot in the second embodiment. In the example of FIG. 18, one shot having the maximum irradiation time Ttr is, for example, 32Δ(=2 5 Δ), 16Δ(=2 4 Δ),8Δ(=2 3 Δ), 4Δ(=2 2 Δ),2Δ(=21 Δ), and Δ(=2 0 In the example of FIG. 18, the shot order of the divided shot groups is, for example, 32Δ(=2 5 Δ) split shot, Δ(=2 0 Δ) split shot, 16Δ(=2 4 Δ) split shot, 2Δ(=2 1 Δ) split shot, 8Δ(=2 3 Δ) split shots, and 4Δ(=2 2 Δ) in the order of divided shots.

[0079] In FIG. 18, data is sent serially from the deflection control circuit 130 to the data line of each row in synchronization with the clock for the number of the connected registers 42. In other words, the data output to the data line is switched in synchronization with the clock for the number of the connected registers 42. Also, each register 42 reads and stores data in synchronization with the preload 1 signal sent from the deflection control circuit 130 to each column, which is synchronized with the clock. For example, the first register 42 column from the upstream side in the data transfer direction of the data line of each row is sent preload 1 signals for the number of the registers 42 connected to the data line. As a result, the last data of the data string is finally stored. The second register 42 column from the upstream side is sent preload 1 signals for the number of the registers 42 connected to the data line minus 1. As a result, the penultimate data of the data string is finally stored. Thereafter, data is stored in each register 42 by the same operation. By arranging two data lines for each row, the data transfer time can be halved. The irradiation time control signal (data) is binary controlled and therefore H or L. Register 43 reads and stores data in synchronization with the preload 2 signal transmitted from the deflection control circuit 130. With the irradiation time control signal (data) for the first divided shot stored in register 43, the irradiation time control signal (data) for the second divided shot is stored in register 42. Register 44 reads and stores data in synchronization with the leg load signal transmitted from the deflection control circuit 130.

[0080] As a result, the blanking control signal (data) stored in the register 44 is input to the input of the AND circuit 49. For example, an H signal is input to the AND circuit 49 from the register 44 if it is a beam ON signal, and an L signal is input to the AND circuit 49 from the register 44 if it is a beam OFF signal. On the other hand, for example, an H signal is input to the AND circuit 49 from the register 45 if the register 42 is a normal register, and an L signal is input to the AND circuit 49 if it is a defective register, from the register 45 if it is a defective register. Then, the AND circuit 49 outputs a beam ON signal (H) to the amplifier 46 if the signals from the registers 44 and 45 are both H signals only during the period in which the shot signal (enable; H signal) transmitted from the deflection control circuit 130 is being input. Otherwise, the amplifier 46 outputs a beam OFF signal (L) to the amplifier 46. The amplifier 46 applies a beam ON potential (GND) to the control electrode 24 only during the period in which the beam ON signal (H) is being output from the AND circuit 49. Otherwise, the amplifier 46 applies a beam OFF potential (Vdd) to the control electrode 24. This completes the first divided shot. Data for the next divided shot is sent to each register in sequence, and after the first divided shot is completed, the AND circuit 49 inputs the next shot signal (enable; H signal) to complete the second divided shot. Thereafter, each divided shot is performed in sequence. The register 45 reads and stores data in synchronization with a flag load signal sent from the deflection control circuit 130.

[0081] As described above, the AND circuit 49 outputs the beam ON signal (H) only during the period when H signals are input from both the register 44 and the register 45 and the shot signal (enable; H signal) is input. In other words, the beam ON signal (H) is output only during the period when all three signals are H signals. Therefore, when the register 42 of the target control circuit 41 is defective, an L signal is input from the register 45, causing the amplifier 46 to constantly apply the beam OFF potential (Vdd) to the control electrode 24, regardless of the signal from the register 44 and the shot signal.

[0082] In the matrix scan method, if the register 42 becomes defective, the data can be read and transferred accurately by lengthening the data hold time of the register 42. The data hold time can be lengthened by lengthening the interval of the preload 1 signal, and thus by slowing down the clock speed. Alternatively, the drive voltage of the control circuit 41 can be temporarily made higher than during normal operation, allowing the data to be read and transferred accurately. However, since increasing the drive voltage can speed up the progression of circuit degradation, a method of slowing down the clock speed is more preferable.

[0083] Therefore, in the second embodiment, similarly to the first embodiment, either a normal flag signal for identifying that the register 42 used in the individual register 45 for each beam is a normal register or a defective flag signal for identifying that the register is a defective register is input at a different time from the blanking control signal (data) via the same data path through which the blanking control signal (data) is transmitted to each control circuit 41 within the blanking aperture array mechanism 204. In the second embodiment, by transferring the flag signal to each control circuit 41 via the existing register 42, the flag signal can be transferred using the matrix scan method without adding a data path.

[0084] In the defective register detection step (S102), the defective register detection unit 60 detects a defective register from among the multiple registers 42 that are arranged in the blanking aperture array substrate 31 that blanks and deflects the multibeam 20 and that input blanking control signals for individual beams of the multibeam 20. In other words, the defective register detection unit 60 detects a defective register 42 from among the registers 42 for each beam of the multibeam 20. In the matrix scan method, unlike the shift register method, the multiple registers 42 are not connected in series, so it is difficult to sequentially read out the data string stored in each chain from the register 42 that is the last in the connection order. Therefore, for example, a shielding plate (not shown) having a through hole that allows only one beam to pass is arranged to select the multibeams 20 one by one. Then, the drawing mechanism 150 irradiates the multibeams 20 one by one onto the Faraday cup 106 on the XY stage 105. Therefore, if the amount of current is detected during the beam OFF control, it can be determined that the beam is always ON or that the beam is defective and blanking control is difficult. Then, by estimating that the resistor 42 is defective as the cause of the defective beam, the defective resistor is detected.

[0085] Alternatively, it is also preferable to provide a diagnostic circuit (not shown) for diagnosing each control circuit 41 of the blanking aperture array mechanism 204, and to diagnose the registers 42 individually to detect defective registers.

[0086] The contents of the flag data generating step (S104) are the same as those in embodiment 1. In the clock switching step (S106), the clock for switching the preload 1 signal and data output is switched to a clock that is sufficiently slower than that in the normal drawing process.

[0087] In the flag transfer step (S108), the transfer control unit 79 transfers to the deflection control circuit 130 a data string in which flag signals for one shot (one divided shot) for each beam are arranged.

[0088] 19 is an example of a timing chart showing the operation of the control circuit of the beam 1 with respect to the flag signal transfer in the second embodiment. The deflection control circuit 130 outputs a data string in which flag signals are arranged to the blanking aperture array mechanism 204. The blanking aperture array mechanism 204 outputs a data string of flag signals switched by a clock sufficiently slower than that in the case of normal drawing processing to a data line, and transfers the flag signals to each register 42 in a matrix scan manner via the same data path as the blanking data, using a preload 1 signal transmitted by a clock sufficiently slower than that in the case of normal drawing processing. Here, since the preload 1 signal transmitted from the deflection control circuit 130 and the clock for switching the data output are sufficiently slowed, the flag signals can be accurately read and transferred not only in the normal register 42 but also in the defective register 42.

[0089] As an amplifier disabling step (S110), each register 45 of the multiple registers 45 (operation limiting circuits) inputs either a normal flag signal for identifying that the register 42 is a normal register (a register that is not a defective register) or a defective flag signal for identifying that the register 42 is a defective register, at a different time from the blanking control signal, via the same data path through which the blanking control signal is transmitted. In particular, in the second embodiment, each register 45 of the multiple registers 45 (operation limiting circuits) inputs the flag signal via the same data path through which the blanking control signal is transmitted and via the register 42 of the corresponding control circuit 41 among the multiple control circuits 41. Specifically, the flag signal stored in the register 42 is read and stored in the register 45 in synchronization with a flag load signal transmitted from the deflection control circuit 130, as shown in FIG. 19. After the flag signal is transferred to all columns, the flag signal is stored in the register 45 simultaneously for all columns. As a result, either a normal flag signal for identifying that the register 42 in the control circuit 41 is a normal register or a defective flag signal for identifying that the register 42 is a defective register is input to the register 45 in each control circuit 41. In the second embodiment, the flag signal is transferred to each control circuit 41 via the existing register 42, so that the flag signal can be transferred using the matrix scan method without adding a data path.

[0090] When the input flag signal is a defective flag signal, register 45 restricts the operation of control circuit 41 in which register 42 corresponding to the defective flag signal is arranged. Here, when the input flag signal is a defective flag signal, register 45 restricts the operation of beam amplifier 46 that uses register 42 corresponding to the defective flag signal. The restriction method is the same as in embodiment 1. This allows the beam controlled by control circuit 41 in which the defective register is arranged to be constantly turned off and disabled.

[0091] The contents of the beam invalidation confirmation step (S112) and the blanking data generation step (S116) are the same as those of the first embodiment.

[0092] In the clock switching step (S118), the clock for switching the preload 1 signal and data output is switched from the slower clock to a clock with a speed for normal drawing processing. The contents of the drawing step (S120) are the same as those in the first embodiment. In this case, the shift register 11 is replaced with a register 42.

[0093] As described above, according to the second embodiment, in the matrix scan method, even if the circuit that takes in the transferred blanking data in the control circuit 41 for individual beams is defective, it is possible to prevent defective beams that are out of blanking control and defective beams that are always ON from being incident on the sample 101.

[0094] Embodiment 3 In each of the above-mentioned embodiments, the register 45, which serves as an operation limiting circuit for limiting the operation of the amplifier 46, is disposed inside each control circuit 41, but this is not limited thereto. In the third embodiment, a case where the register 45 is disposed outside the control circuit 41 will be described. The configuration of the drawing device 100 in the third embodiment is the same as that in FIG. 1. Also, the flowchart of the drawing method in the third embodiment is the same as that in FIG. 13. In the third embodiment, the clock switching step (S106) and the clock switching step (S118) may be omitted. Points not specifically described below may be the same as in the first embodiment.

[0095] FIG. 20 is a diagram showing an example of a control circuit for individual beams and an example of an operation limiting circuit arranged inside the blanking aperture array mechanism in the third embodiment. In FIG. 20, a shift register 11, two stages of registers 42 and 43 serving as buffers, an execution register 44, an amplifier 46, a selection circuit 48, and an AND circuit 49 are arranged inside each control circuit 41. In FIG. 20, the output of the shift register 11 is connected to the input of the register 42. The output of the register 42 is connected to the input of the register 43. The output of the register 43 is connected to the input of the register 44. The output of the register 44 is connected to the input of the AND circuit 49. The input of the AND circuit 49 is connected to the output of the register 45 arranged outside the control circuit 41 and a shot signal transmitted from the deflection control circuit 130. The output of the AND circuit 49 is connected to the input of the amplifier 46. The output of the amplifier 46 is connected to the control electrode 24 as shown in FIG. 5.

[0096] 20, shift register 11 reads and stores data in synchronization with a shift clock signal transmitted from deflection control circuit 130. Also, shift register 11 transfers the stored data in synchronization with the shift clock signal. Register 42 reads and stores data in synchronization with a preload 1 signal transmitted from deflection control circuit 130. Register 43 reads and stores data in synchronization with a preload 2 signal transmitted from deflection control circuit 130. Register 44 reads and stores data in synchronization with a leg load signal transmitted from deflection control circuit 130.

[0097] As shown in FIG. 3, the blanking aperture array substrate 31 has a membrane region 330 with a thin film thickness h, which is a central multi-beam passing region through which the multi-beams 20 pass, and an outer peripheral region 332 surrounding the membrane region 330. As shown in FIG. 20, a plurality of control circuits 41 are arranged in an array in the membrane region 330, which is a multi-beam passing region. Meanwhile, a plurality of registers 45 (operation limiting circuits) are arranged in the outer peripheral region 332. As shown in FIG. 6, for example, in the case of a multi-beam configuration of 512 columns x 512 rows, the control circuits 41 for the 1st to 512th beams in each row form a data sequence 1 (group) at 8 beam pitches, i.e., beams 1, 9, 17, 25, . . . Similarly, a data sequence 2 (group) is formed at 8 beam pitches, i.e., beams 2, 10, 18, 26, . . . Similarly, data sequence 3 (group) to data sequence 8 (group) are formed. The control circuits 41 in each group are connected in series. Then, data columns 1 to 8 of each group in the same row are arranged in a series. For example, as row data for the kth row of a beam array, data columns 1 to 8 of the kth row of a beam array are arranged in a series. Then, when each row data is transferred in parallel from the deflection control circuit 130 to the blanking aperture array mechanism 204, it is transmitted in parallel to the membrane region 330 via a data path arranged in the outer periphery region 332 of the blanking aperture array substrate 31. Then, the row data is divided into data columns 1 to 8 for each row by an I / O circuit arranged near the periphery of the membrane region 330, and transmitted in parallel to each group.

[0098] In the third embodiment, for example, eight groups for each row, one register 45 (operation limiting circuit) is arranged for each group. For example, eight registers 45 (a to h) for each row are connected in series. The output of the register 45 for each group is connected to the input of, for example, eight AND circuits 49 in the same group. The input of the first register 45 connected in series for each row is connected to a data path for blanking data arranged in the outer peripheral region 332 of the blanking aperture array substrate 31. Therefore, a flag signal can be transferred to the register 45 via the data path for blanking data without adding a data path for data input to the register 45.

[0099] Each register 45 reads and stores data in synchronization with a flag load signal transmitted from the deflection control circuit 130. In the example of Fig. 20, since multiple registers 45 for each row are connected in series, it is necessary to transfer flag signals using a shift register method. Therefore, a desired flag signal can be transferred to each register 45 by, for example, eight flag load signals.

[0100] In the example of FIG. 20, it is preferable to provide the selection circuit 48, but it may be omitted. In the case where the selection circuit 48 is provided, in the example of FIG. 20, the output of the shift register 11 is further connected to one of two inputs of the selection circuit 48. The other of the two inputs of the selection circuit 48 is connected to the upstream data line of the shift register 11. The output of the selection circuit 48 is connected to the input of the next shift register 11 connected in series. Also, the selection circuit 48 selects one of the two inputs according to the signal stored in the register 45. Also, it goes without saying that when the selection circuit 48 is omitted, the output of the shift register 11 is connected to the input of the next shift register 11 connected in series.

[0101] As described above, the AND circuit 49 outputs the beam ON signal (H) only during the period when H signals are input from both the register 44 and the register 45 and when a shot signal (enable; H signal) is input. In other words, the beam ON signal (H) is output only during the period when all three signals are H signals. Therefore, when the shift register 11 of the target control circuit 41 is defective, an L signal is input from the register 45, causing the amplifier 46 to constantly apply a beam OFF potential (Vdd) to the control electrode 24, regardless of the signal from the register 44 and the shot signal.

[0102] In the example of FIG. 20, for example, eight groups for each row, one register 45 (operation limiting circuit) is arranged for each group, so that the operations of, for example, eight amplifiers 46 in the same group are simultaneously limited.

[0103] FIG. 21 is a diagram for explaining a control circuit in which an amplifier output is limited by a failure flag signal in the third embodiment and a comparative example. The examples of FIG. 21(a) and FIG. 21(b) show a case where, for example, in a beam array of 16 columns (x direction)×8 rows (y direction), the control circuits 41 (cells) are grouped for every 4 beam pitches in each row. In the third embodiment, one register 45 (operation limiting circuit) is arranged for each group, so that flag signals for each row are sufficient as many as the number of groups. Therefore, as shown in FIG. 21(a), 4×8 flag signals may be generated. The groups of failure flag signals are shown as filled squares. The example of FIG. 21(a) shows a case where, for example, a failure flag signal is applied to the group in the second row from the top and the second column from the left. By applying this failure flag signal, the beams controlled by a total of four cells, the cell in the second row from the top and the second column from the left, and the three cells arranged every four pitches on the left side, are invalidated. On the other hand, the example of FIG. 21(b) shows a comparative example in which a flag signal is generated for each cell. In the comparative examples (eg, the first and second embodiments), the beams controlled by the cells to which the failure flag signal is applied are invalidated individually, not for each group.

[0104] In the flag data generating step (S104), the flag data generating unit 66 generates 4×8 flag signals as shown in FIG. 21(a). For example, if there is even one defective register among the multiple shift registers 11 in a group, a defective flag signal is generated for that group. For groups with no defective registers, a normal flag signal is generated. Then, a data string of flag signals for one shot (one divided shot) for each beam, arranged in series by row, is generated. For example, an "H" signal is generated as the normal flag signal, and an "L" signal is generated as the defective flag signal.

[0105] In the third embodiment, since the flag signal is not transferred via the defective shift register 11, there is no need to slow down the clock, and the clock switching step (S106) may be omitted.

[0106] In a flag transfer step (S108), the transfer control unit 79 transfers a data string in which flag signals for one shot (one divided shot) for each beam are arranged to the deflection control circuit 130. In synchronization with the flag load signal, each register 45 sequentially reads and transfers the flag signals. By transmitting a flag load signal for the number of groups per row from the deflection control circuit 130, the desired flag signals are stored in each register 45.

[0107] In the amplifier disabling step (S110), each flag signal is read into the register 45 and stored, so that each register 45 restricts the operation of the control circuit 41 in which the shift register 11 in the group corresponding to the defective flag signal is disposed when the input flag signal is a defective flag signal. Here, when the input flag signal is a defective flag signal, the operation of all amplifiers 46 using the shift register 11 in the group corresponding to the defective flag signal is restricted. Specifically, the operation is as follows. When a defective flag signal is input, each register 45 fixes the output of all amplifiers 46 using the shift register 11 in the group corresponding to the defective flag signal to a constant beam OFF potential. Specifically, by storing a defective flag (disable) signal in the register 45, it is possible to cause the AND circuit 49 to output a beam OFF signal (L) to the amplifier 46. As a result, as shown in FIG. 20, the operation of the amplifier 46 can be restricted so that the amplifier 46 outputs a constant beam OFF potential (Vdd). On the other hand, a normal flag signal (enable) is stored in the register 45 for the group without the defective shift register 11, so that the AND circuit 49 inputs an H signal from the register 45. Therefore, the output of the AND circuit 49 is controlled by the signal from the register 44 and the shot signal (enable; H signal), and is not subject to restrictions from the register 45.

[0108] In this manner, the beam controlled by the group including the control circuit 41 in which the defective shift register 11 is arranged can be disabled.

[0109] In the example of FIG. 20, one register 45 is arranged for each group, but this is not limited thereto. One group may be divided into a plurality of subgroups, and one register 45 may be arranged for each subgroup. In addition, in the example of FIG. 20, a plurality of registers 45, the number of which is equal to the number of groups, are connected in series for each row, but this is not limited thereto. One group may be divided into a plurality of subgroups, and a register 45 may be connected in series for each subgroup. This can reduce the number of clocks of the flag load signal for transferring the flag signal to each register 45.

[0110] The contents of each step after the beam invalidation confirmation step (S112) may be the same as those in the first embodiment.

[0111] According to the third embodiment, the operation limiting circuit can be disposed in the outer periphery region 332, which can secure a larger arrangement space than the membrane region 330.

[0112] Embodiment 4 In the above-mentioned embodiments, the case where the operation of the amplifier 46 is restricted is described, but the present invention is not limited to this. In the fourth embodiment, a configuration in which a defective register itself is separated and switched to a normal register is described. The configuration of the writing device 100 in the fourth embodiment is the same as that in FIG. 1. Also, the flowchart of the writing method in the fourth embodiment is the same as that in FIG. 13, except that the amplifier disabling step (S110) and the beam checking step (S112) are deleted. Points that are not particularly described below may be the same as those in the first embodiment.

[0113] FIG. 22 is a diagram showing an example of a control circuit for individual beams arranged inside the blanking aperture array mechanism in the fourth embodiment. In FIG. 22, two shift registers 11a and 11b, two-stage registers 42 and 43 serving as buffers, an execution register 44, a register 45, an amplifier 46, a selection circuit 48, and an AND circuit 49 are arranged in each control circuit 41 arranged in the blanking aperture array substrate 31. In FIG. 22, two shift registers 11a (used normally) (first register) and a shift register 11b (used for reserve) (second register) are arranged in parallel. Of the two shift registers 11a and 11b, for example, the shift register 11a is used normally. The shift register 11b is used as a reserve in case the shift register 11a becomes defective. In the fourth embodiment, when a plurality of shift registers of the same group are connected in series, two parallel shift registers 11a and 11b are arranged for each control circuit 41. The output of the shift register 11a is connected to one of the two inputs of the selection circuit 48. The output of the shift register 11b is connected to the other of the two inputs of the selection circuit 48. The output of the shift register 11a is further connected to the input of the register 45. The output of the selection circuit 48 is connected to the inputs of the two shift registers 11a and 11b of the next control circuit 41, and is also connected to the input of the register 42. The output of the register 42 is connected to the input of the register 43. The output of the register 43 is connected to the input of the register 44. The output of the register 44 is connected to the input of the AND circuit 49. The input of the AND circuit 49 is further connected to a shot signal transmitted from the deflection control circuit 130. The output of the AND circuit 49 is connected to the input of the amplifier 46. The output of the amplifier 46 is connected to the control electrode 24, as shown in FIG. 5.

[0114] As described above, each of the multiple control circuits 41 has a shift register 11a, a shift register 11b, and an amplifier 46. The shift register 11a and the shift register 11b are arranged in parallel and are arranged in the blanking aperture array substrate 31, and input a blanking control signal transmitted via a data path in the blanking aperture array substrate 31. The amplifier 46 applies a beam-on potential or a beam-off potential to a corresponding one of the multiple control electrodes 24 in accordance with the blanking control signal input to either the shift register 11a or the shift register 11b. A specific description will be given below.

[0115] In FIG. 22, two shift registers 11a and 11b arranged in parallel read (input) and store a blanking control signal (data) transmitted through a data path in the blanking aperture array substrate in synchronization with a shift clock signal transmitted from the deflection control circuit 130. Also, the stored data is transferred in synchronization with the shift clock signal. An amplifier 46 applies a beam-on potential or a beam-off potential to a corresponding control electrode 24 among the plurality of control electrodes 24 according to the blanking control signal input to one of the two shift registers 11a and 11b. A specific description will be given below. The register 42 reads and stores data in synchronization with a preload 1 signal transmitted from the deflection control circuit 130. The register 42 reads data from one of the two shift registers 11a and 11b whose connection is selected by the selection circuit 48. The register 43 reads and stores data in synchronization with a preload 2 signal transmitted from the deflection control circuit 130. The register 44 reads and stores data in synchronization with the leg load signal transmitted from the deflection control circuit 130. As a result, the blanking control signal (data) stored in the register 44 is input to the input of the AND circuit 49. For example, an H signal is input from the register 44 to the AND circuit 49 if it is a beam ON signal, and an L signal is input if it is a beam OFF signal. Then, the AND circuit 49 outputs a beam ON signal (H) to the amplifier 46 if the signals from the register 44 and the register 45 are both H signals only during the period when the shot signal (enable; H signal) transmitted from the deflection control circuit 130 is being input. Otherwise, the amplifier 46 outputs a beam OFF signal (L) to the amplifier 46. The amplifier 46 applies a beam ON potential (GND) to the control electrode 24 only during the period when the beam ON signal (H) is being output from the AND circuit 49. Otherwise, the amplifier 46 applies a beam OFF potential (Vdd) to the control electrode 24.

[0116] The defective register detection step (S102), the flag data generation step (S104), and the clock switching step (S106) are similar to those in the first embodiment.

[0117] In the flag transfer step (S108), the register 45 (storage unit) holds flag signals, which are generated for each of the multiple control circuits 41 and include either a normal flag signal for identifying that the shift register 11a is a normal register or a defective flag signal for identifying that the shift register 11a is a defective register. Specifically, the operation is as follows. The register 45 inputs either a normal flag signal for identifying that the shift register 11a is a normal register or a defective flag signal for identifying that the shift register 11a is a defective register, at a different time from the blanking data, via the same data path through which the blanking data is transmitted and via the shift register 11a of the corresponding control circuit 41 among the multiple control circuits 41. For the defective shift register 11a, the speed of the shift clock used when transferring the blanking control signal (data) can be slowed down to allow data to be read and transferred accurately. Alternatively, the drive voltage of the control circuit 41 can be temporarily increased to be higher than that during normal operation, to allow data to be read and transferred accurately. However, since increasing the drive voltage may increase the rate at which circuit deterioration progresses, a method of slowing down the shift clock speed is more preferable. At this time, the register 45 reads and stores the data in synchronization with a flag load signal transmitted from the deflection control circuit 130. As a result, the flag signal stored in the register 45 is input to the selection circuit 48.

[0118] As a register selection step (S114), each of the multiple selection circuits 48 (one example of a switching circuit) inputs the generated flag signal at a different time from the blanking control signal through the same data path through which the blanking control signal is transmitted and through, for example, the shift register 11a of the corresponding control circuit 41 among the multiple control circuits 41, and when the input flag signal is a defective flag signal, switches the connection from the shift register 11a that input the flag signal to the shift register 11b on the amplifier 46 side. Specifically, it operates as follows. The selection circuit 48 selects one of two inputs according to the output of the register 45. In other words, the selection circuit 48 switches the connection from one of the two inputs to the other according to the output of the register 45. For example, an H signal is input to the selection circuit 48 if the flag signal is normal, and an L signal is input to the selection circuit 48 if the flag signal is defective. Usually, the shift register 11a is normal, so the selection circuit 48 connects the shift register 11a to the amplifier 46 side. However, when the input flag signal is a defective flag signal (L), the selection circuit 48 switches the connection to the amplifier 46 side from the shift register 11a to which the flag signal is input to the spare shift register 11b. As a result, the defective shift register 11a can be disconnected and switched to the normal shift register 11b.

[0119] The contents of each step after the blanking data generating step (S116) may be the same as those in the embodiment 1. Also, it goes without saying that a step of checking the beam may be carried out after the register selecting step (S114).

[0120] As described above, according to the fourth embodiment, even if the circuit that takes in the transferred blanking data in the control circuit 41 for individual beams in the blanking aperture array mechanism 204 in multi-beam writing is defective, it is possible to prevent defective beams due to uncontrollable blanking and defective beams that are always ON from being incident on the sample 101 without disabling the amplifier 46.

[0121] Although the embodiments have been described above with reference to specific examples, the present invention is not limited to these specific examples.

[0122] Although the description of the device configuration, control method, and other parts not directly necessary for the explanation of the present invention have been omitted, the required device configuration and control method can be appropriately selected and used. For example, the description of the control unit configuration that controls the drawing device 100 has been omitted, but it goes without saying that the required control unit configuration can be appropriately selected and used.

[0123] In addition, all other multi-charged particle beam writing apparatuses and multi-charged particle beam writing methods that include the elements of the present invention and that can be appropriately modified by those skilled in the art are included in the scope of the present invention. [Explanation of symbols]

[0124] 11 Shift Registers 20 Multibeam 22 holes 24 Control electrode 25 Passing hole 26 Counter electrode 28,36 pixels 29 Sub-irradiation area 30 drawing area 31 Blanking aperture array board 32 Stripe Area 34 Irradiation area 41 Control circuit 42, 43, 44, 45 Registers 46 Amplifier 47 Individual Blanking Mechanism 48 Selection circuit 49 And Circuit 60 Bad register detection unit 62 Irradiation time calculation unit 64 Blanking data generator 66 Flag Data Generation Unit 70 Clock control section 72 Amplifier operation control section 76 Detection unit 79 Transfer control section 80 Drawing control section 100 drawing device 101 Sample 102 Electron Tube 103 Drawing room 105 XY Stage 106 Faraday Cup 110 Control computer 112 Memory 130 Deflection control circuit 131 Logic Circuits 132,134 DAC amplifier unit 136 Lens control circuit 137 Amplifier 138 Stage Control Mechanism 139 Stage Position Measuring Instrument 140,142 Storage device 150 Drawing mechanism 160 Control Circuits 200 Electron Beam 201 Electron Gun 202 Lighting lens 203 Shaped aperture array substrate 204 Blanking Aperture Array Mechanism 205 Reduction Lens 206 Limiting Aperture Board 207 Objective Lens 208 Main deflector 209 Sub deflector 210 Mirror 212 Common blanking deflector 330 Membrane Region 332 Outer area 343 Pad

Claims

1. a multi-beam forming mechanism for forming multiple charged particle beams; a blanking aperture array substrate having a plurality of openings formed at positions through which the multi-charged particle beam passes; a plurality of control electrodes disposed on the blanking aperture array substrate in the vicinity of the plurality of openings; a plurality of counter electrodes disposed on the blanking aperture array substrate so as to face one of the plurality of control electrodes across different openings among the plurality of openings; a plurality of control circuits each including a register arranged on the blanking aperture array substrate and inputting a blanking control signal transmitted via a data path in the blanking aperture array substrate, and an amplifier applying a beam-on potential or a beam-off potential to a corresponding one of the plurality of control electrodes in accordance with the blanking control signal inputted to the register; a storage unit for storing flag signals, which are generated for each of the plurality of control circuits and include either a normal flag signal for identifying that the register is a normal register or a defective flag signal for identifying that the register is a defective register; a plurality of operation limiting circuits that input each of the generated flag signals through the same data path through which the blanking control signal is transmitted at a different time from the blanking control signal, and limit the operation of a control circuit in which a register corresponding to the defective flag signal is disposed when the input flag signal is a defective flag signal; a limiting aperture substrate for blocking the beam deflected by applying a beam-OFF potential to the control electrode; A multi-charged particle beam writing apparatus comprising:

2. A diagnostic unit for diagnosing the register; a flag data generating unit that generates the normal flag signal or the defective flag signal by using a diagnosis result of the register; 2. The multi-charged particle beam writing apparatus according to claim 1, further comprising:

3. 3. A multi-charged particle beam drawing apparatus according to claim 1, wherein each of the plurality of operation limiting circuits, when receiving a fault flag signal, fixes an output of a beam amplifier using a register corresponding to the fault flag signal to a constant beam OFF potential.

4. 3. A multi-charged particle beam drawing apparatus as claimed in claim 1, wherein each of the plurality of operation limiting circuits inputs the flag signal via the same data path through which the blanking control signal is transmitted and via the register of a corresponding one of the plurality of control circuits.

5. A shift register is used as the register, The multi-charged particle beam is divided into a plurality of groups each of a plurality of beams, the shift registers for the beams in the same group are connected in series; 3. The multi-charged particle beam drawing apparatus according to claim 1, further comprising a selection circuit which connects, among the plurality of serially connected shift registers, a shift register in its own control circuit to a shift register in an adjacent control circuit on the downstream side in the data transfer direction, and which is connected to a shift register in an adjacent control circuit on the upstream side in the data transfer direction and a shift register in an adjacent control circuit on the downstream side so as to skip the shift register in its own control circuit, and which selects whether data transfer skips the shift register in its own control circuit or not in response to the flag signal.

6. the blanking aperture array substrate has a multi-beam passing region at a central portion through which the multi-charged particle beam passes, and an outer peripheral region at an outer periphery; the plurality of control circuits are disposed in the multi-beam passing region; 3. The multi-charged particle beam drawing apparatus according to claim 1, wherein the plurality of operation limiting circuits are arranged in the outer circumferential area.

7. a multi-beam forming mechanism for forming multiple charged particle beams; a blanking aperture array substrate having a plurality of openings through which the multi-charged particle beam passes; a plurality of control electrodes disposed on the blanking aperture array substrate in the vicinity of the plurality of openings; a plurality of counter electrodes disposed on the blanking aperture array substrate so as to face one of the plurality of control electrodes across different openings among the plurality of openings; a plurality of control circuits, each of which includes a first register and a second register arranged in parallel and which input a blanking control signal transmitted through a data path in the blanking aperture array substrate, and an amplifier which applies a beam-on potential or a beam-off potential to a corresponding one of the plurality of control electrodes in accordance with the blanking control signal input to one of the first register and the second register; a storage unit for storing flag signals, the flag signals being generated for each of the plurality of control circuits and including either a normal flag signal for identifying that the first register is a normal register or a defective flag signal for identifying that the first register is a defective register; a plurality of switching circuits that input the generated flag signal at a different time from the blanking control signal, via the same data path through which the blanking control signal is transmitted and via the first register of a corresponding one of the plurality of control circuits, and when the input flag signal is a defective flag signal, switch a connection to the amplifier side from the first register that input the flag signal to the second register; a limiting aperture substrate for blocking the beam deflected by applying a beam-OFF potential to the control electrode; A multi-charged particle beam drawing apparatus comprising:

8. a step of diagnosing a plurality of registers, which are arranged in a blanking aperture array substrate for blanking and deflecting the multiple charged particle beams, and which input blanking control signals for individual beams of the multiple charged particle beams, to detect a defective register; generating a flag signal including either a normal flag signal for identifying a register that is not the defective register or a defective flag signal for identifying the defective register, inputting the flag signal at a time different from that of the blanking control signal through the same data path through which the blanking control signal is transmitted, and restricting operation of an amplifier for a beam that uses a register corresponding to the defective flag signal when the input flag signal is the defective flag signal; a step of applying a beam-ON potential or a beam-OFF potential from an individual amplifier for each beam to a control electrode disposed in the vicinity of an opening formed on the blanking aperture array substrate through which the beam passes, in accordance with the blanking control signal input to an individual register for each beam transmitted through a data path in the blanking aperture array substrate, shielding a blanking-deflected beam due to a potential difference between the control electrode and an opposing electrode opposing the control electrode as a result of the beam-OFF potential being applied to the control electrode, and drawing a pattern on a sample using a group of beams among the multi-charged particle beams that have not been shielded; A multi-charged particle beam writing method comprising:

Citation Information

Patent Citations

  • Microprocessor containing redundant register

    JP1992010033A