Methods, apparatus and programs for depicting multiple charged particle beams

Abstract

A method, apparatus and program for multi-charged particle beam mapping are provided. When multiple beams are used to map the sample, regardless of whether there are beams with specific current densities, the yield reduction and quantization error can be suppressed. One aspect of the multi-charged particle beam mapping method of the present invention involves setting one of a plurality of weighting coefficients for each beam of the multi-charged particle beam according to the arrangement position of the multi-charged particle beam. When multiple mappings are performed on the same position of the sample to be mapped by a plurality of beams with different arrangement positions and two or more weighting coefficients set in the multi-charged particle beam, for each beam of the plurality of beams irradiating that position, the irradiation amount of the plurality of beams is corrected by using two or more weighting coefficients of the plurality of beams irradiating that position. For each position of the sample, the position is multiple mapped by the plurality of beams using the corrected irradiation amount of each beam of the plurality of beams, thereby mapping a pattern on the sample using multi-charged particle beams.

Description

Technical Field

[0001] One aspect of the present invention is a multi-charged particle beam drawing method, a multi-charged particle beam drawing device and a program, for example, a method of using multiple beams with current density distribution to correct the irradiation time in multiple drawing.

[0002] This application claims priority based on Japanese Patent Application No. 2024-013619 (filing date: January 31, 2024), and the entire contents of the basic application are incorporated herein by reference. Prior Art

[0003] Lithography, the single most important process for pattern generation in semiconductor manufacturing, is responsible for the advancement of semiconductor device miniaturization. With the increasing integration of LSIs in recent years, the required circuit line widths for semiconductor devices have become increasingly smaller. Electron beam lithography, a technique with inherently superior resolution, has long been used for patterning wafers and other applications.

[0004] For example, there are multi-beam patterning devices. Compared to patterning with a single electron beam, using multiple beams allows for more beams to be irradiated at once, significantly increasing throughput. In this multi-beam patterning device, for example, the electron beam emitted from an electron gun passes through a mask with multiple apertures to form multiple beams. Each beam is then masked. The unmasked beams are optically reduced and deflected by a deflector to irradiate the desired location on the sample.

[0005] The current density of each beam used in multi-beam mapping is not uniform, but rather results in a current density distribution. Therefore, if mapping is performed without correction, the required dose cannot be achieved, resulting in reduced mapping accuracy. To address this issue, a conventional approach has been to use the ratio of the ideal current density divided by the current density of each beam when irradiating the sample, and to correct the designed dose. This correction is applied on a shot-by-shot basis, so beams with unique current densities may require a different irradiation time than other shots. In multi-beam mapping, the firing cycle is set so that the entire irradiation time matches the maximum irradiation time. Therefore, shots with unique irradiation times increase the overall mapping time, resulting in reduced yield.

[0006] Therefore, a method has been devised: when a sample is subjected to multiple imaging, the current density of the beam used in each pass is averaged, and the ratio of the ideal current density divided by the averaged current density is used as a correction factor to correct the designed dose for each pass (see Japanese Patent Application Laid-Open No. 2023-056384). This method averages the specific current density, thereby reducing the specific irradiation time.

[0007] In this case, the same correction factor is applied to multiple beams irradiating the same position, resulting in the same irradiation time in each pass. As a result, quantization errors accumulate in the same way, causing a discrepancy between the actual dose and the actual dose. Summary of the Invention

[0008] One aspect of the present invention provides a multi-charged particle beam profiling method, a multi-charged particle beam profiling device, and a program. When a sample is subjected to multiple profiling using multiple beams, regardless of the presence or absence of beams with specific current densities, the method can suppress yield reduction and reduce quantization errors.

[0009] A multi-charged particle beam drawing method according to one aspect of the present invention, For each beam of the multi-charged particle beam, one of the plurality of weight coefficients is set according to the arrangement position of the multi-charged particle beam. When the same position of a sample to be mapped is mapped multiple times using a plurality of beams arranged at different positions and having two or more weighting coefficients set in a multi-charged particle beam, the irradiation dose of each of the plurality of beams that irradiate the position is corrected using the two or more weighting coefficients of the plurality of beams that map the position. For each position of the sample, a plurality of beams are used to make multiple depictions of the position by correcting the irradiation amount of each beam of the plurality of beams, thereby using multiple charged particle beams to draw patterns on the sample.

[0010] A multi-charged particle beam imaging device according to one aspect of the present invention comprises: a setting circuit for setting one of a plurality of weight coefficients for each beam of the multiple charged particle beams according to an arrangement position of the multiple charged particle beams; a correction circuit for correcting, when the same position of a sample to be mapped is multiple-mapped by a plurality of beams arranged at different positions and having two or more weight coefficients set thereto in a multi-charged particle beam, for each of the plurality of beams irradiating the position, a previously determined irradiation dose of the plurality of beams using the two or more weight coefficients of the plurality of beams mapping the position; and The drawing mechanism uses multiple charged particle beams to draw a pattern on the sample by using multiple beams to draw multiple patterns on each position of the sample by using the irradiation amount of each beam of the multiple beams after correction.

[0011] A program according to one aspect of the present invention causes a computer to execute: For each beam of the multiple charged particle beams, one of a plurality of weight coefficients is set according to the arrangement position of the multiple charged particle beams; The process of storing the weight coefficients set for each beam in a memory device; and A processing method is provided for reading weight coefficients from a memory device, and when the same position of a sample to be depicted is multiplexed by a plurality of beams arranged in different positions and having two or more weight coefficients set thereto among a multi-charged particle beam, for each of the plurality of beams irradiating the position, correcting the irradiation dose of the plurality of beams, which has been previously calculated, by applying the two or more weight coefficients of the plurality of beams depicting the position. Simple diagram description

[0012] [Figure 1] is a conceptual diagram illustrating the structure of a drawing device in embodiment 1. [Figure 2] is a conceptual diagram illustrating the structure of the formed aperture array substrate in embodiment 1. [Figure 3] is a cross-sectional view illustrating the structure of the masking aperture array mechanism in Embodiment 1. [Figure 4] is a conceptual diagram used to illustrate the drawing action in embodiment 1. FIG5 is a diagram illustrating an example of a multi-beam irradiation area and pixels to be drawn in the first embodiment. [Figure 6] is a diagram used to illustrate an example of the multi-beam imaging operation in embodiment 1. [Figure 7] is a diagram illustrating an example of current density distribution in Embodiment 1. FIG. 8 is a diagram for explaining a method for correcting variations in irradiation time due to differences in current density in Comparative Example 1 of Embodiment 1. FIG. FIG. 9 is a diagram for explaining a method for correcting variations in irradiation time due to differences in current density in Comparative Example 2 of Embodiment 1. FIG. [Figure 10] is a diagram for explaining an example of multiple rendering of multiplicity 2 in embodiment 1. [Figure 11] is an example of a flowchart illustrating the main steps of the drawing method in embodiment 1. [Figure 12] is a diagram for explaining an example of multiple rendering with a multiplicity of 4 in embodiment 1. [Figure 13] is a diagram illustrating an example of a group area of ​​weight coefficients in Implementation 1. [Figure 14] is a diagram illustrating an example of the block structure within the beam array in embodiment 1. [Figure 15] is a diagram illustrating another example of the group area of ​​the weight coefficient in Implementation Example 1. [Figure 16] is a diagram used to illustrate the method of multiple drawing in variant example 1 of implementation method 1. [Figure 17] is a diagram illustrating an example of correction coefficient distribution and quantization error in Comparative Example 1 of Implementation Example 1. [Figure 18] is a diagram illustrating an example of correction coefficient distribution and quantization error in Comparative Example 2 of Implementation 1. [Figure 19] is a diagram illustrating an example of correction coefficient distribution and quantization error in embodiment 1. [Figure 20] is a diagram used to illustrate modification example 1 of the weight coefficient in embodiment 1. [Figure 21] is a diagram used to illustrate a second modification of the weight coefficient in implementation mode 1. Implementation Method

[0013] The following embodiments provide a method and apparatus for characterizing a sample using multiple beams, which can suppress yield reduction and reduce quantization error regardless of the presence of beams with specific current densities.

[0014] In the following embodiments, a configuration using an electron beam as an example of a charged particle beam is described. However, the charged particle beam is not limited to an electron beam and may also be a beam using charged particles such as an ion beam. Implementation method 1.

[0015] FIG1 is a conceptual diagram illustrating the configuration of a drawing apparatus in Embodiment 1. FIG1 shows drawing apparatus 100, which includes a drawing mechanism 150 and a control system circuit 160. 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. Drawing mechanism 150 includes an electron column 102 (electron beam column) and a drawing chamber 103. Within electron column 102 are located 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, and a secondary deflector 209.

[0016] An XY stage 105 is placed within the drawing chamber 103. A sample 101, such as a photomask, which will serve as the substrate to be drawn during drawing (exposure), is placed on the XY stage 105. Sample 101 includes exposure masks used in semiconductor device manufacturing or semiconductor substrates (silicon wafers) used in semiconductor device manufacturing. Sample 101 also includes mask blanks that have been coated with resist but have not yet been subjected to any drawing. A mirror 210 is also placed on the XY stage 105 for measuring its position.

[0017] In addition, a Faraday cup 106 is arranged on the XY stage 105 .

[0018] The control system circuit 160 includes a control computer 110, a memory 112, a deflection control circuit 130, digital-to-analog converter (DAC) amplifier units 132 and 134, a lens control circuit 136, a stage control mechanism 138, a stage position detector 139, and memory devices 140 and 142, such as disk drives. The control computer 110, the memory 112, the deflection control circuit 130, the lens control circuit 136, the stage control mechanism 138, the stage position detector 139, and the memory devices 140 and 142 are interconnected via a bus (not shown). The deflection control circuit 130 is connected to the DAC amplifier units 132 and 134 and the blanking aperture array mechanism 204. The auxiliary 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, each of which is controlled by the deflection control circuit 130 through a DAC amplifier 134. The lens group, including the illumination lens 202, the reduction lens 205, and the objective lens 207, is controlled by the lens control circuit 136.

[0019] The position of the XY stage 105 is controlled by driving motors (not shown) on each axis controlled by the stage control mechanism 138. The stage position detector 139 receives reflected light from the mirror 210 and measures the position of the XY stage 105 using the principle of laser interferometry.

[0020] The control computer 110 includes a gridding processing unit 50 , a firing data generating unit 52 , a current density distribution creating unit 56 , a weight coefficient setting unit 58 , a correction coefficient calculating unit 60 , a correction unit 62 , a drawing control unit 72 , and a transfer processing unit 74 . Each of the "units"—the gridding processing unit 50, the firing data generating unit 52, the current density distribution creating unit 56, the weighting coefficient setting unit 58, the correction coefficient calculating unit 60, the correction unit 62, the rendering control unit 72, and the transfer processing unit 74—has a processing circuit. This processing circuit may include, for example, an electronic circuit, a computer, a processor, a circuit board, a quantum circuit, or a semiconductor device. Each "unit" may use a common processing circuit (the same processing circuit) or different processing circuits (individual processing circuits). Information input and output by the gridding processing unit 50, the firing data generating unit 52, the current density distribution creating unit 56, the weighting coefficient setting unit 58, the correction coefficient calculating unit 60, the correction unit 62, the rendering control unit 72, and the transfer processing unit 74, as well as information currently being calculated, is stored in the memory 112 at all times.

[0021] The drawing operation of the drawing device 100 is controlled by the drawing control unit 72. In other words, the drawing control unit 72 (an example of a control circuit) controls the drawing mechanism 150. Furthermore, the transfer processing of the irradiation time data of each shot to the deflection control circuit 130 is controlled by the transfer processing unit 74.

[0022] Furthermore, drawing data (chip data) is input from outside the drawing device 100 and stored in the memory device 140. The chip data defines information about the multiple graphics patterns that make up the chip pattern. Specifically, for each graphic pattern, the coordinates of each vertex are defined, for example, in the order in which the graphics are formed. Alternatively, for example, a graphic code, coordinates, and dimensions are defined for each graphic pattern.

[0023] 1 shows the necessary configuration for explaining Embodiment 1. The drawing device 100 may also include other necessary configurations.

[0024] FIG2 is a conceptual diagram illustrating the structure of the shaped aperture array substrate in Embodiment 1. FIG2 shows a shaped aperture array substrate 203 having p rows in the horizontal (x-direction) and q rows in the vertical (y-direction) (p, q ≥ 2) of apertures (openings) arranged in a matrix at a predetermined pitch. The example in FIG2 illustrates a case where 512 × 512 rows of apertures 22 are formed in the horizontal and vertical (x, y-directions). The number of apertures 22 is not limited to this. For example, 32 × 32 rows of apertures 22 are also possible. Each aperture 22 is formed in a rectangular shape of the same size. Alternatively, it may be circular with the same diameter. Portions of the electron beam 200 pass through each of these multiple apertures 22, thereby forming multiple beams 20. In other words, the shaped aperture array substrate 203 forms multiple beams 20.

[0025] FIG3 is a cross-sectional view illustrating the structure of the blanking aperture array mechanism in Embodiment 1. As shown in FIG3 , the blanking aperture array mechanism 204 comprises a blanking aperture array substrate 31, a semiconductor substrate made of silicon or the like, mounted on a support 33. In the thin film region 330 in the center of the blanking aperture array substrate 31, passage holes 25 (openings) for each of the multiple beams 20 to pass through are formed at positions corresponding to the apertures 22 of the shaped aperture array substrate 203 shown in FIG2 . Furthermore, a pair of control electrodes 24 and opposing electrodes 26 (blanking devices or blanking deflectors) are disposed at positions facing each corresponding aperture 25 within the plurality of passage holes 25. Furthermore, within the blanking aperture array substrate 31 adjacent to each passage hole 25, a control circuit 41 (logic circuit) is disposed to apply a bias voltage to the control electrodes 24 for each passage hole 25. The opposing electrodes 26 for each beam are grounded.

[0026] An amplifier (an example of a switching circuit) (not shown) is configured within the control circuit 41. As an example of an amplifier, a CMOS (Complementary MOS) inverter circuit is configured as a switching circuit. A control signal is applied to the input (IN) of the CMOS inverter circuit, either an L (low) potential (e.g., ground potential) below the threshold voltage, or an H (high) potential (e.g., 1.5V) above the threshold voltage. In Embodiment 1, when an L potential is applied to the input (IN) of the CMOS inverter circuit, the output (OUT) of the CMOS inverter circuit applied to the control circuit 41 becomes a positive potential (Vdd). The electric field generated by the potential difference from the ground potential of the counter electrode 26 deflects the corresponding beam and is shielded by the aperture limiting substrate 206, thereby controlling the beam to be OFF. 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 ground potential. The potential difference with the ground potential of the counter electrode 26 disappears, and the corresponding beam is not deflected. Therefore, it passes through the limiting aperture substrate 206, thereby controlling the beam to be on. This deflection allows for blanking control.

[0027] Next, a specific example of the operation of the drawing mechanism 150 will be described. Electron beam 200 emitted from electron gun 201 (emission source) illuminates the entire aperture array substrate 203 nearly vertically through illumination lens 202. The aperture array substrate 203 is formed with a plurality of rectangular apertures 22 (openings), and electron beam 200 illuminates an area encompassing all of these apertures 22. Portions of electron beam 200 irradiated at the locations of the plurality of apertures 22 pass through the aperture array substrate 203, forming, for example, rectangular multi-beams (plural electron beams) 20 (beam array). These multi-beams 20 pass through corresponding blankers of the blanking aperture array mechanism 204. Each blanker controls the blanking of the individual beams passing through it, ensuring that the beams remain in the ON state for the set drawing time (irradiation time).

[0028] The multiple beams 20 that have passed through the blanking aperture array mechanism 204 are reduced by the reduction lens 205 and travel toward the hole formed in the center of the limiting aperture substrate 206. Here, the electron beams deflected by the blanker of the blanking aperture array mechanism 204 are shifted from the hole in the center of the limiting aperture substrate 206 and blocked by the limiting aperture substrate 206. Meanwhile, the electron beams not deflected by the blanker of the blanking aperture array mechanism 204 pass through the hole in the center of the limiting aperture substrate 206, as shown in Figure 1. Thus, the limiting aperture substrate 206 blocks the beams deflected by the blanker of the blanking aperture array mechanism 204 into the beam-off state. The beams formed by passing through the limiting aperture substrate 206 from the time of beam-on to the time of beam-off form the beams of a single shot. The multiple beams 20 that have passed through the limiting aperture substrate 206 are focused by the objective lens 207 to form a pattern image at the desired reduction ratio. The multiple beams 20 that have passed through the limiting aperture substrate 206 are then collectively deflected in the same direction by the main deflector 208 and the sub-deflector 209, irradiating the sample 101 at their respective irradiation positions. Furthermore, for example, when the XY stage 105 is continuously moving, tracking control is performed by the main deflector 208 so that the irradiation positions of the beams follow the movement of the XY stage 105. Ideally, the multiple beams 20 irradiated at a time are aligned at a distance determined by multiplying the arrangement pitch of the plurality of apertures 22 of the aperture array substrate 203 by the desired reduction ratio.

[0029] FIG4 is a conceptual diagram illustrating the drawing operation in Embodiment 1. As shown in FIG4 , a drawing area 30 (bold lines) of a sample 101 is hypothetically divided into a plurality of stripe regions 32 with a predetermined width in the y-direction. FIG4 illustrates the example of a drawing area 30 of a sample 101 being divided into a plurality of stripe regions 32 with a width substantially equal to the size of a designed irradiation area 34 (drawing field; beam array area) that can be irradiated by a single multi-beam 20 irradiation.

[0030] In the example of Figure 4, a multi-pass drawing with a multiplicity of 2 is shown. A first stripe layer, consisting of a plurality of stripe regions 32 formed by dividing the drawing area 30, is set for the first drawing process. Furthermore, a second stripe layer, consisting of a plurality of stripe regions 32 shifted in the y direction relative to the first stripe layer, is set for the second drawing process. During multi-pass drawing, the amount of shift in the x and y directions between each pass is set, for example, based on the multiplicity. For example, at a multiplicity of N, a suitable shift between each pass is 1 / N of the width of the stripe region 32. The multiplicity is not limited to 2 and can be 3 or higher. A multi-pass drawing process refers to the stage movement when repeatedly moving the stage to draw multiple stripe regions 32 with the same number. In addition, in multiple drawing, there are also cases where the same pixel is drawn multiple times in the same pass, that is, in one platform movement, thereby performing multiple drawing.

[0031] In addition, while the example in Figure 4 illustrates position shifting in both the x and y directions, this is not limiting. Position shifting may be performed only in the x direction between multiple rendering passes. Alternatively, position shifting may be performed only in the y direction between multiple rendering passes. Next, an example of rendering operation will be described.

[0032] First, the XY stage 105 is moved and adjusted so that the irradiation area 34 of the multi-beam 20 is located at or further to the left of the first stripe area 32 of the first stripe layer. Then, when drawing the first stripe area 32, the XY stage 105 is moved, for example, in the -x direction, thereby gradually advancing the drawing in the x direction. The XY stage 105 is moved continuously, for example, at a constant speed.

[0033] After the first stripe region 32 of the first stripe layer is drawn, the stage position is moved in the -y direction by a dimension equal to 1 / N of the width of the stripe region 32. This shifts the drawn stripe region 32 in the y direction by, for example, 1 / N of the width of the stripe region 32. In the example of FIG4 , which illustrates a two-pass multi-drawing process, the shift in the y direction is by, for example, 1 / 2 of the width of the stripe region 32.

[0034] Next, the irradiation area 34 of the multi-beam 20 is adjusted so that it is positioned at or further to the left of the first stripe area 32 of the second stripe layer. The XY stage 105 is then moved, for example, in the -x direction, to gradually advance the patterning process in the x direction. This progresses to patterning the first stripe area 32 of the second stripe layer. After the patterning of the first stripe area 32 of the second stripe layer is completed, the patterning of the second stripe area 32 of the first stripe layer is performed. In this manner, the corresponding stripe areas 32 of each stripe layer are patterned sequentially. This process is then repeated to complete the patterning of all stripe areas 32 of each stripe layer.

[0035] Furthermore, while the example in Figure 4 illustrates the process of drawing each stripe region 32 in the same direction, this is not limiting. For example, the next stripe region 32 to be drawn following the stripe region 32 already drawn in the x direction can be drawn in the -x direction by moving the XY stage 105, for example, in the x direction. This alternating direction change during drawing can reduce stage movement time, and ultimately, drawing time. In a single shot, the multiple beams formed by passing through the apertures 22 of the shaped aperture array substrate 203 can simultaneously produce a maximum of a plurality of shot patterns, equal to the number of apertures 22.

[0036] FIG5 illustrates an example of a multi-beam irradiation area and target pixels for mapping in Embodiment 1. In FIG5 , a stripe area 32 is divided into a plurality of mesh regions, for example, based on the beam size of the multi-beam. Each mesh region serves as a target pixel 36 (beam unit irradiation area; irradiation position; position). The size of target pixel 36 is not limited to the beam size and can be any size regardless of the beam size. For example, it can be 1 / n (n is an integer greater than or equal to 1) the beam size. The example in FIG5 illustrates a case where the mapping area of ​​sample 101 is divided into a plurality of stripe regions 32, for example, with a width in the y-direction substantially equal to the width of the irradiation area 34 (the mapping field) that can be irradiated by a single irradiation of the multi-beam 20. The x-direction dimension of the rectangular irradiation area 34 can be defined as the number of beams in the x-direction × the inter-beam spacing in the x-direction. The y-dimensional dimension of the rectangular irradiation area 34 can be defined as the number of beams in the y-dimensional direction × the inter-beam spacing in the y-dimensional direction. In the example of Figure 5 , for example, the illustration of 512×512 rows of multi-beams is abbreviated to 8×8 rows of multi-beams. Furthermore, within the irradiation area 34 , the plurality of pixels 28 (beam mapping positions) that can be irradiated by a single shot of the multi-beam 20 are shown. The spacing between adjacent pixels 28 is the inter-beam spacing of the multi-beams. A sub-irradiation area 29 (pitch unit area) is formed by a rectangular area enclosed by the inter-beam spacing in the x and y directions. In the example of Figure 5 , each sub-irradiation area 29 is composed of, for example, 4×4 pixels.

[0037] FIG6 is a diagram illustrating an example of a multi-beam imaging operation in Embodiment 1. The example in FIG6 illustrates imaging within each sub-irradiation area 29 using four different beams. Furthermore, FIG6 illustrates an imaging operation in which the XY stage 105 continuously moves at a speed corresponding to a distance L equal to exactly 8 beam pitches while imaging a quarter (one-fourth of the number of beam channels used for irradiation) of each sub-irradiation area 29. The imaging operation illustrated in FIG6 illustrates, for example, that while the XY stage 105 moves a distance L equal to 8 beam pitches, the sub-deflector 209 sequentially shifts the irradiation position (pixels 36) while firing the multi-beam 20 four times with a firing period T, thereby imaging (exposing) four different pixels within the same sub-irradiation area 29. During the drawing (exposure) of these four pixels, the main deflector 208 deflects all of the multi-beams 20 simultaneously, allowing the irradiation area 34 to track the movement of the XY stage 105. This prevents the irradiation area 34 from shifting relative to the sample 101 due to the movement of the XY stage 105. In other words, tracking control is performed. At the end of a tracking cycle, a tracking reset is performed, returning to the previous tracking start position. Furthermore, since the drawing of the first pixel row from the right of each sub-irradiation area 29 has been completed, after the tracking reset, the sub-deflector 209 first deflects the beam drawing position in the next tracking cycle to align (shift) it to the undrawn pixel row of each sub-irradiation area 29, such as the second pixel row from the right. In the depiction of the stripe region 32, by repeating this action, the position of the irradiation region 34 of the multi-beam 20 gradually moves in sequence as shown in the irradiation regions 34a, 34b, 34c, ... 34o in the lower figure of FIG. 4, and the depiction is gradually performed.

[0038] For example, in the example shown in Figure 6, when the 32×32 multi-beam 20 is used for patterning, each pixel 36 is patterned once during one stage pass. When the same operation is performed using a 512×512-row multi-beam 20, each sub-irradiation area 29 is composed of 16×16 pixels. Furthermore, while the XY stage 105 moves a distance L equal to 32 beam pitches, the sub-deflector 209 sequentially shifts the irradiation position (pixels 36) while firing the multi-beam 20 16 times with a firing cycle T, thereby patterning (exposing) 16 different pixels within the same sub-irradiation area 29. After tracking reset, the irradiation position is aligned with the row next to the row in each sub-irradiation area 29 where patterning has been completed. By repeating this operation, each pixel 36 is patterned once during one stage pass (per pass).

[0039] Figure 7 illustrates an example of current density distribution in Embodiment 1. The current density of each of the multiple beams 20 that make up the beam array is not uniform. For example, as shown in Figure 7, the current density decreases radially outward compared to the central beam group. In the example shown in Figure 7, the current density of the central beam group is 100%, while the current density is 97% outward, 95% even further outward, and the current density decreases further outward. Therefore, even when beams are irradiated on the sample surface for the same irradiation time, the dose incident on the sample will still vary.

[0040] FIG8 illustrates a method for correcting irradiation time variations caused by current density differences in Comparative Example 1 of Embodiment 1. In Comparative Example 1, for beams with low current density, the current density error is corrected using a correction factor to achieve an ideal current density. Specifically, the ratio (J 0 / J (i, j)) of the current density J (i, j) of each beam to the ideal current density J 0, for example, 100%, is used as the correction factor to correct the irradiation time t (i, j). (i, j) represents an index indicating the arrangement position of each beam within the beam array. The dose dose (i, j) can be defined by the following equation (1).

[0041]

[0042] As shown in equation (1), the dose dose(i,j) is calculated by multiplying the current density J(i,j) by the irradiation time. Therefore, the designed dose dose(i,j) can be obtained by multiplying the product of the pre-correction irradiation time t(i,j) calculated from the ideal current density and the current density J(i,j) by the correction factor (J 0 / J(i,j)). The drawing process is controlled by the irradiation time, so the corrected irradiation time t'(i,j) is simply multiplied by the correction factor (J 0 / J(i,j)) to obtain the corrected irradiation time t'(i,j). However, in Comparative Example 1, if a beam with a uniquely low current density occurs within the beam array, the correction factor for that beam will uniquely increase. In multi-beam mapping, the firing cycle is set to match the longest exposure time. Therefore, if a unique correction factor exists, the firing cycle for all shots will increase, leading to a longer mapping time. This results in reduced yield.

[0043] FIG9 is a diagram illustrating a method for correcting the deviation in irradiation time caused by differences in current density in Comparative Example 2 of Implementation 1. As shown in FIG9 , multiple traces are performed while shifting the position of the beam array. The example in FIG9 illustrates a case where multiple traces are performed while shifting the beam array halfway in the x-direction. In this case, each pixel on the sample surface is irradiated by a plurality of beams arranged at different positions. Furthermore, in Comparative Example 2, the current density j(i, j) of the plurality of beams arranged at different positions used for the same pixel is averaged in each pass of the multiple traces. The dose dose(i, j) in Comparative Example 2 can be defined by the following formula (2).

[0044]

[0045] As shown in formula (2), the ratio of the ideal current density J0 divided by the averaged current density (ΣJ(i,j) / n) is used as a correction factor to correct the designed dose for each pass. n represents the number of averaged current densities. In other words, n represents the multiplicity N. This allows the specific current density to be averaged, thus suppressing the specific irradiation time.

[0046] Figure 10 illustrates an example of multi-pass drawing with a multiplicity of 2 in Embodiment 1. The example in Figure 10 illustrates a case of multiplicity N = 2. The second pass is drawn by shifting the stripe region 32 in both the x and y directions from the first pass. Each stripe region 32 is repeatedly drawn under the same beam conditions in a plurality of rectangular regions 35, each divided in the x direction by the same dimensions as the beam array's irradiation area 34. Therefore, when performing multi-pass drawing with a shift of 1 / 2 the width of the stripe region 32 in both the x and y directions, two beam groups with the same markings, each representing four blocks formed by dividing the beam array's irradiation area into 2×2 patterns, draw the same area of ​​the stripe region 32 in different passes of the multi-pass drawing. Of the two blocks A, the lower left and upper right, the area drawn by the beam group in the upper right block A in the first pass is drawn by the beam group in the lower left block A in the second pass. In other words, the pixels irradiated by the beams in the upper right block A are irradiated by the beams in the corresponding arrangement positions in the lower left block A.

[0047] Here, in Comparative Example 2, the correction coefficients of the two beams irradiating the same pixel are the same due to the averaging shown in formula (2). It is a common practice to set the design irradiation time of each pass in multiple drawing to the same value. Therefore, there is a high probability that the corrected irradiation time will also be the same. As a result, the quantization error generated when the irradiation time is digitized in a predetermined quantization unit will cause the error of the dose controlled by the irradiation time to act in the same direction. When the reference irradiation time is set to 1 and standardized, the irradiation time of each pass of a certain pixel is assumed to be 1.5, for example. Assume that the quantization error in this case is, for example, +0.1. In this case, since multiple drawing is repeated with the same irradiation time, the quantization error will accumulate. For example, if the multiplicity is 2, a quantization error of +0.2 will occur, and if the multiplicity is 4, a quantization error of +0.4 will occur. Therefore, if the multiplicity is 4, a dose error of 0.4× the current density will occur in the direction of excessive increase of the dose. Conversely, assume that the quantization error is, for example, -0.1. In this case, since multiple imaging is repeated with the same irradiation time, the quantization error accumulates. For example, a multiplicity of 2 results in a quantization error of -0.2, while a multiplicity of 4 results in a quantization error of -0.4. Therefore, a multiplicity of 4 results in a dose error of 0.4 times the current density, which leads to insufficient dose.

[0048] In view of this, in the first embodiment, two or more weightings are applied to the multiple beams irradiating each pixel during multi-pass rendering, thereby staggering the irradiation time in each pass. This is described in detail below.

[0049] FIG11 is an example of a flow chart illustrating the main steps of the drawing method according to Embodiment 1. In FIG11 , the drawing method according to Embodiment 1 includes a series of steps: a current density distribution creation step ( S102 ), a weight coefficient setting step ( S104 ), a correction coefficient calculation step ( S110 ), an irradiation time data generation step ( S120 ), an irradiation time correction step ( S130 ), and a drawing step ( S140 ).

[0050] In the current density distribution creation step (S102), the current density of each beam in the multi-beam 20 is first measured under the control of the drawing control unit 72. For example, the beams other than the target beam are controlled to be turned off, and the target beam is incident on the Faraday cup 106. This allows the current value of the beam to be measured. The measurement results for each beam are output to the control computer 110 via a detection circuit (not shown). The current density distribution creation unit 56 creates a current density distribution that defines the current density of each beam as an element. For example, a current density map is created. The current density can be calculated by dividing the measured current value by the cross-sectional area of ​​the beam. In addition, if the diameters of the apertures (openings) 22 of the aperture array substrate 203 vary, the current amount of each beam can be defined as an element, taking the variation in aperture diameter into account. The created current density distribution (current amount distribution) is stored in the memory device 142. Here, the current density of each beam is described as an example of the current amount of each beam.

[0051] As a weight coefficient setting step ( S104 ), the weight coefficient setting unit 58 sets one of a plurality of weight coefficients α for each beam of the multi-beam 20 according to the arrangement position (i, j) of the multi-beam 20 .

[0052] Figure 12 illustrates an example of multiple-pass drawing with a multiplicity of 4 in Embodiment 1. The example in Figure 12 illustrates a case where the multiplicity N = 4. Each pass is drawn by shifting the stripe region 32 in both the x and y directions from the previous pass by 1 / 4 of its width. Each stripe region 32 is repeatedly drawn under the same beam conditions within a plurality of rectangular regions 35, each of the same size as the beam array's irradiation area 34, divided in the x direction. Therefore, when multiple-pass drawing is performed by shifting the stripe region 32 in both the x and y directions by 1 / 4 of its width, the beam groups within the 16 blocks A through D, each of which is a 4×4 division of the beam array's irradiation area, will draw the same region of the stripe region 32.

[0053] Therefore, the beam groups within the four blocks marked A within the plurality of blocks within the beam array depict the same area of ​​stripe region 32 in different passes within each block. In other words, beams at corresponding arrangement positions within the block with the same mark (e.g., A) illuminate the same pixel. If the correction coefficients of the four beams illuminating the same pixel were averaged and set to the same value as in Comparative Example 2, the aforementioned quantization error would result. Therefore, the weight coefficient setting unit 58 sets a weight coefficient for each beam so that at least two of the passes have different correction coefficients.

[0054] In the first embodiment, the same pixel 36 of the sample 101 is repeatedly mapped using a plurality of beams in the multi-beam 20, each of which is assigned two or more weighting coefficients α(i, j) and arranged at different positions. In this case, the dose (dose(i, j)) of each beam in each pass can be defined using the two or more weighting coefficients α(i, j) for mapping the pixel 36 and the current amount, such as the current density J(i, j), of each beam in the plurality of beams. The dose (dose(i, j)) of each beam in each pass can be defined using the following equation (3). While equation (3) uses the current density J(i, j) of each beam, the current amount I(i, j) of each beam can also be used instead of the current density J(i, j).

[0055]

[0056] The weight coefficient α(i, j) set for each beam irradiating the same pixel in equation (3) is independently set regardless of the beam current, such as the current density J(i, j). Furthermore, the weight coefficient α(i, j) is set for each beam so that the total dose (irradiation amount) of the multiple beams irradiating each pixel 36 of the sample 101 reaches the designed value. The total value is allowed to differ from the designed value to a degree that does not affect the imaging accuracy.

[0057] Furthermore, in equation (3), when averaging the current density J between passes, the values ​​obtained by multiplying the current density J of each beam by the weight coefficient for that beam are summed. This sum is then divided by the weight coefficient for that beam. Multiplying the averaged correction coefficient by the weight coefficient alone, as described in Comparative Example 2, would result in a deviation from the designed dose. Therefore, when applying the weight coefficient for correction, as in equation (3), by correlating it with the current density J, the designed dose can be achieved. A specific example is described below.

[0058] Figure 13 illustrates an example of a group area for weight coefficients in Embodiment 1. The example in Figure 13 illustrates a multi-level mapping operation with a multiplicity of 4, shifting the position in both the x and y directions by 1 / 4 of the width of stripe region 32. The example in Figure 13 also illustrates a scenario where different weight coefficients are assigned to multiple beams irradiating the same position on sample 101. This is explained in detail below. Furthermore, the term "same position" does not necessarily mean the exact same position; it also allows for beam position deviation or design-based shifting of the mapping (irradiation) position, as long as the beams overlap at least partially.

[0059] In the example shown in Figure 13, the beam array is divided into four group regions in the y direction. Then, starting from the bottom region, the beam groups within that region are defined as group region G1, group region G2, group region G3, and group region G4. For a multi-beam array 20 with 512 × 512 rows, the arrangement position (i, j) of each beam is defined as 0 to 511 for i and j, respectively.

[0060] The weighting coefficient α is preferably a number that is exactly the same as the multiplicity N of the multi-rendering. The multiple weighting coefficients may also be designed to include 1. Furthermore, the multiple weighting coefficients may include two or more different values. For example, in the case of a multiplicity of 4, four weighting coefficients, 0.98, 0.99, 1.01, and 1.02, may be used as α. In this manner, four completely different values ​​are preferably set. However, this is not limiting. α may include 1, such as 0.97, 1.00, 1.01, and 1.02. Alternatively, as long as there are two or more values, some of the same values ​​may be used. In the case of a multiplicity of 2, two weighting coefficients, 0.99 and 1.01, may be used as α. While the range of weighting coefficients is not particularly limited, considering quantization error, it is preferable that the difference between multiple weighting coefficients be on the scale of 1 / M (where M is the number of gradations in the irradiation dose). For example, it is suitable to set the weighting coefficients within the range of 0.95 to 1.05, with the weighting coefficients being divisible by 1 / 100.

[0061] In either case, the total value of the weight coefficients applied to the same pixel in each pass is set to match the multiplicity. In the above example, if α is 0.98, 0.99, 1.01, or 1.02, the total becomes 4, which matches the multiplicity of 4. This allows the total value of the dose (irradiation amount) of the multiple beams irradiating each pixel 36 to be set as the design value.

[0062] A plurality of weight coefficients α may be predetermined for each multiplicity. The weight coefficient set data for each multiplicity may be pre-stored in the memory device 140 .

[0063] In the example of FIG. 13 , the weight coefficient setting unit 58 defines the beam group (beams in the first region from the bottom) whose arrangement positions are (0 to 511, 0 to 127) as the group region G1, and sets a weight coefficient α0 = 0.99 for each beam in the group region G1.

[0064] The weight coefficient setting unit 58 defines the beam group (beams in the second region from the bottom) whose arrangement positions are (0 to 511, 128 to 255) as the group region G2, and sets the weight coefficient α1 = 1.01 for each beam in the group region G2.

[0065] The weight coefficient setting unit 58 defines the beam group with arrangement positions (0 to 511, 256 to 383) (the beams in the third area from the bottom) as the group area G3, and sets the weight coefficient α2 = 0.98 for each beam in the group area G3.

[0066] The weight coefficient setting unit 58 defines the beam group (beams in the fourth region from the bottom) whose arrangement positions are (0-511, 384-511) as group region G4, and sets a weight coefficient α3 = 1.02 for each beam in group region G4.

[0067] As described above, during multi-pass mapping using multiple beams, the same area is repeatedly mapped for each block of the beam array. Therefore, by setting different weighting coefficients for blocks with the same symbol, the weighting coefficients for the multiple beams irradiating the same position in each pass can be set to different values. In the example of Figure 12, the position is shifted in the x-direction with each pass, resulting in four blocks A-D in the x-direction. However, simply changing the weighting coefficients for the four group areas divided in the y-direction is sufficient. By setting weighting coefficients for the group areas encompassing multiple beams, the number of weighting coefficients can be reduced compared to setting weighting coefficients for each beam individually. Alternatively, a weighting coefficient can be set for each beam for a greater number of blocks than the number of group areas, resulting in a different correction coefficient. Alternatively, a weighting coefficient can be set for each beam for each sub-block formed by further subdividing each block. For example, dividing each block into 4×4 sub-blocks is also suitable. Alternatively, a weighting coefficient may be set for each beam so as to be a different correction coefficient.

[0068] In the example shown in Figure 12, in the first pass, a subregion of stripe region 32 is irradiated by the beam group of the topmost block A among the four blocks A. Therefore, α3 = 1.02 is applied to the beam group of the topmost block A. In the second pass, the same subregion is irradiated by the beam group of the second block A from above. Therefore, α2 = 0.98 is applied to the beam group of the second block A from above. In the third pass, α1 = 1.01 is applied. And in the fourth pass, α0 = 0.99 is applied. Therefore, although the same area is irradiated using different multi-drawing passes in the four blocks A, the weighting coefficients vary with each pass. Alternatively, at least two different weighting coefficients are applied across the four passes.

[0069] As a correction coefficient calculation process (S110), when the same pixel 36 (position) of the sample 101 as the depiction object is multiple-depicted by a plurality of beams with different arrangement positions of two or more weight coefficients α set in the multi-beam 20, the correction coefficient calculation unit 60 calculates a correction coefficient K for each of the plurality of beams irradiating the pixel 36 using the two or more weight coefficients α of the plurality of beams depicting the pixel 36 and the current density J(i, j) of each of the plurality of beams.

[0070] As shown in Figure 13, the correction coefficient K can be defined by the following formula (4) using the weight coefficient α(i, j) of the beam used in each pass, the ideal current density J0, and the current density J(i, j) of the beam used in each pass, as shown in formula (3).

[0071]

[0072] The data of the calculated correction coefficients of each beam are stored in advance in the memory device 142 .

[0073] Figure 14 illustrates an example of the block configuration within the beam array in Embodiment 1. The example in Figure 14 illustrates a multiplicity N = 4. Each pass is performed by shifting the stripe region 32 in both the x and y directions from the previous pass. Each stripe region 32 is repeatedly mapped under the same beam conditions within a plurality of rectangular regions 35, each of the same size as the beam array's irradiation area 34, divided in the x direction. Therefore, when multiple mapping is performed by shifting the stripe region 32 in both the x and y directions by 1 / 4 of its width, the beam groups within the 16 blocks A through D, each divided into 4×4 sections of the beam array's irradiation area, map the same area of ​​the stripe region 32.

[0074] Therefore, the beam groups within the four blocks labeled A (A1-A4) within the plurality of blocks within the beam array will depict the same area of ​​stripe region 32 at different passes within each block. In other words, beams at corresponding arrangement positions within the block with the same label (e.g., A) illuminate the same pixel. This is similar to the situation in Figure 12.

[0075] In the example of Figure 14 , the weighting coefficients are varied for each group area. Therefore, even for blocks with the same symbol A, different correction coefficients can be generated for blocks A1, A2, A3, and A4. Therefore, even if quantization errors occur, they are not accumulated in a single direction due to multiple rendering, but can be averaged. Similarly, even for blocks with the same symbol B, different correction coefficients can be generated for blocks B1, B2, B3, and B4. Therefore, even if quantization errors occur, they are not accumulated in a single direction due to multiple rendering, but can be averaged. Similarly, even for blocks with the same symbol C, different correction coefficients can be generated for blocks C1, C2, C3, and C4. Therefore, even if quantization errors occur, they are not accumulated in a single direction due to multiple rendering, but can be averaged. Similarly, even for blocks with the same symbol D, different correction coefficients can be generated for blocks D1, D2, D3, and D4. Therefore, even if quantization errors occur, they are not accumulated in a single direction due to multiple renderings, but are averaged out.

[0076] In the example of FIG. 13 described above, although the beam array is divided in the y direction into a plurality of group regions, the present invention is not limited thereto.

[0077] Figure 15 illustrates another example of group regions for weight coefficients in Embodiment 1. In addition to the example in Figure 13 , Figure 15 also features segmentation in the x-direction. In the example in Figure 15 , segmentation in the x-direction is performed into four group regions, and the beam array is divided into 16 group regions G11 to G44 in 4×4 rows. Weight coefficients are then set for each group region. At this point, weight coefficients are adjusted for groups depicting the same region in different passes through multiple rendering. Specifically, weight coefficients are adjusted for group regions G11, G22, G33, and G44 (corresponding to blocks A1, A2, A3, and A4) depicting the same region. Similarly, weight coefficients are adjusted for group regions G41, G12, G23, and G34 (corresponding to blocks B1, B2, B3, and B4) depicting the same region. Similarly, weight coefficients are adjusted for group regions G21, G32, G43, and G14 (corresponding to blocks C1, C2, C3, and C4) depicting the same region. Similarly, in the group areas G31, G42, G13, and G24 (equivalent to the blocks D1, D2, D3, and D4) that depict the same area, the weight coefficients are adjusted.

[0078] In the example of FIG. 15 , the weight coefficient setting unit 58 defines the beam group whose arrangement positions are (0 to 127, 0 to 127) as a group region G11 and sets a weight coefficient α11 = 0.99 for each beam in the group region G11.

[0079] The weight coefficient setting unit 58 defines the beam group with arrangement positions (128 to 255, 0 to 127) as a group region G21, and sets a weight coefficient α21 = 0.99 for each beam in the group region G21.

[0080] The weight coefficient setting unit 58 defines the beam group with arrangement positions (256 to 383, 0 to 127) as a group region G31, and sets a weight coefficient α31 = 1.02 for each beam in the group region G31.

[0081] The weight coefficient setting unit 58 defines the beam group with arrangement positions (384 to 511, 0 to 127) as a group region G41, and sets a weight coefficient α41 = 0.99 for each beam in the group region G41.

[0082] The weight coefficient setting unit 58 defines the beam group with arrangement positions (0 to 127, 128 to 255) as a group region G12, and sets a weight coefficient α12 = 1.02 for each beam in the group region G12.

[0083] The weight coefficient setting unit 58 defines the beam group with arrangement positions (128 to 255, 128 to 255) as a group region G22, and sets a weight coefficient α22 = 1.01 for each beam in the group region G22.

[0084] The weight coefficient setting unit 58 defines the beam group with arrangement positions (256 to 383, 128 to 255) as a group region G32, and sets a weight coefficient α32 = 1.01 for each beam in the group region G32.

[0085] The weight coefficient setting unit 58 defines the beam group with arrangement positions (384 to 511, 128 to 255) as a group region G42, and sets a weight coefficient α42 = 0.98 for each beam in the group region G42.

[0086] The weight coefficient setting unit 58 defines the beam group with arrangement positions (0 to 127, 256 to 383) as a group region G13, and sets a weight coefficient α13 = 1.01 for each beam in the group region G13.

[0087] The weight coefficient setting unit 58 defines the beam group with arrangement positions (128 to 255, 256 to 383) as a group region G23, and sets a weight coefficient α23 = 0.98 for each beam in the group region G23.

[0088] The weight coefficient setting unit 58 defines the beam group with arrangement positions (256 to 383, 256 to 383) as a group region G33, and sets a weight coefficient α33 = 0.98 for each beam in the group region G33.

[0089] The weight coefficient setting unit 58 defines the beam group with arrangement positions (384 to 511, 256 to 383) as a group region G43, and sets a weight coefficient α43 = 0.98 for each beam in the group region G43.

[0090] The weight coefficient setting unit 58 defines the beam group with arrangement positions (0 to 127, 384 to 511) as a group region G14, and sets a weight coefficient α14 = 1.02 for each beam in the group region G14.

[0091] The weight coefficient setting unit 58 defines the beam group with arrangement positions (128 to 255, 384 to 511) as a group region G24, and sets a weight coefficient α24 = 0.99 for each beam in the group region G24.

[0092] The weight coefficient setting unit 58 defines the beam group with arrangement positions (256 to 383, 384 to 511) as a group region G34, and sets a weight coefficient α34 = 1.01 for each beam in the group region G34.

[0093] The weight coefficient setting unit 58 defines the beam group with the arrangement positions (384 to 511, 384 to 511) as a group region G44, and sets a weight coefficient α44 = 1.02 for each beam in the group region G44.

[0094] FIG16 illustrates the multi-image method in Variation 1 of Embodiment 1. The example in FIG16 illustrates multi-image processing within the same pass. The example in FIG16 illustrates multi-image processing using the left and right halves of the beam array. For example, when imaging using 512×512 rows of multi-beams 20, each sub-irradiation area 29 is configured with, for example, 16×16 pixels. In this case, while the XY stage 105 moves a distance L equal to 16 beam pitches, the sub-deflector 209 sequentially shifts the irradiation position (pixels 36) while firing the multi-beam 20 16 times with a firing period T, thereby imaging (exposing) 16 different pixels within the same sub-irradiation area 29. This allows the area already imaged by the right half of the beam array to be multi-imaged by the left half of the beam array.

[0095] The example in Figure 16 illustrates a case where the multiplicity N = 8. Four passes, with two passes within each pass, totaling eight multiple rendering passes. Each pass shifts the stripe region 32 in the y direction by exactly one-quarter of its width and in the x direction by exactly one-eighth of its width from the previous pass. Each stripe region 32 is repeatedly rendered under the same beam conditions within a plurality of rectangular regions 35, each of the same size as the beam array's irradiation area 34, divided in the x direction. Furthermore, in Figure 16, within each rectangular region 35, both the left and right halves are repeatedly rendered under the same beam conditions. Therefore, when multiple depictions are performed by shifting the beam array by exactly 1 / 4 of the width of the stripe region 32 in the y direction and by exactly 1 / 8 in the x direction, the beam groups within the 32 blocks A to D with the same markings, which are divided into 8×4 areas, will depict the same area of ​​the stripe region 32.

[0096] Therefore, the beam groups within the eight blocks marked A within the plurality of blocks within the beam array will depict the same area of ​​stripe region 32 in different passes or within the same pass within each block. In other words, beams at corresponding arrangement positions within the blocks with the same mark (e.g., A) illuminate the same pixel. Regarding the correction coefficients of the eight beams illuminating the same pixel, the weight coefficient setting unit 58 sets a weight coefficient for each beam so that at least two of the passes have different correction coefficients. Alternatively, the weight coefficients may be set for each beam for each pass. Alternatively, the weight coefficients may be set for each beam for each block. Alternatively, the weight coefficients may be set for each beam for each beam.

[0097] In the irradiation time data generation step ( S120 ), the gridding processing unit 50 first reads the wafer pattern data (drawing data) for each stripe region 32 from the memory device 140 and performs gridding processing. Specifically, the pattern density (pattern area density) is calculated for each pixel 36 .

[0098] Next, the firing data generation unit 52 calculates the irradiation dose D for each pixel 36. The irradiation dose D can be calculated, for example, by multiplying a pre-set base irradiation dose Dbase by the proximity effect correction irradiation coefficient Dp and the pattern area density ρ. Thus, the irradiation dose D is preferably calculated proportionally to the pattern area density calculated for each pixel 36. To correct the proximity effect irradiation coefficient Dp, the drawing area (here, for example, the stripe area 32) is virtually divided into a plurality of adjacent mesh areas (mesh areas for proximity effect correction calculation) of a predetermined size. The size of the adjacent mesh areas is preferably set to approximately 1 / 10 of the range of the proximity effect, for example, approximately 1 μm. The drawing data is then read from the memory device 140, and for each adjacent mesh area, the pattern density ρ' (pattern area density) of the pattern arranged within that adjacent mesh area is calculated.

[0099] Next, a proximity effect correction exposure coefficient Dp is calculated for each adjacent mesh area to correct for the proximity effect. The mesh area size for calculating the proximity effect correction exposure coefficient Dp does not necessarily need to be the same as the mesh area size for calculating the pattern density ρ'. Furthermore, the correction model and calculation method for the proximity effect correction exposure coefficient Dp can be the same as those used in conventional single-beam profiling methods.

[0100] Next, the shot data generating unit 52 first calculates, for each pixel 36, the irradiation time t of the electron beam for incident upon the calculated irradiation dose D. The irradiation time t can be calculated by dividing the irradiation dose D by the current density J. This creates a dose map (actually, an irradiation time map containing the irradiation time data as an element) that defines the irradiation time data (shot data) for each pixel 36.

[0101] Then, when performing multiple drawing passes, a dose map (actually, an irradiation time map) is generated for each drawing pass. In other words, a dose map (actually, an irradiation time map) is generated for each stripe layer. The generated irradiation time data is stored in the memory device 142.

[0102] In the irradiation time correction step (S130), when the same location on the sample 101 is repeatedly imaged by a plurality of beams arranged at different positions and having two or more weighting coefficients set in the multi-beam 20, the correction unit 62 (an example of an irradiation time calculation unit) corrects the previously calculated irradiation dose of each of the plurality of beams irradiating the location by applying the two or more weighting coefficients of the plurality of beams that image the location. As described above, the irradiation dose of each beam is obtained by multiplying the current of each beam by the irradiation time. Here, for example, the irradiation time of each beam is corrected. In other words, when the same pixel 36 (position) of the sample 101 is multiply mapped using a plurality of beams with different arrangement positions and two or more weighting coefficients set in the multi-beam 20, the correction unit 62 (an example of an irradiation time calculation unit) calculates, for each of the plurality of beams irradiating the pixel 36, the corrected irradiation time t'(i, j) using the two or more weighting coefficients α(i, j) used to map the pixel 36 and the current density J(i, j) of each of the plurality of beams. Specifically, the correction unit 62 reads the correction coefficient K of the target beam and the pre-correction irradiation time t(i, j) from the memory device 142, multiplies the pre-correction irradiation time t(i, j) by the correction coefficient K for the beam, and calculates the corrected irradiation time t'(i, j). The generated corrected irradiation time data is stored in the memory device 142 in the order of firing.

[0103] The above example illustrates a configuration in which a correction coefficient K is first calculated for each beam at each arrangement position and pre-stored in the memory device 142. The correction coefficient K is then read from the memory device 142 to correct the pre-correction irradiation time t(i, j). However, this is not limiting. Instead of pre-calculating the correction coefficient K, the corrected irradiation time t'(i, j) can be directly calculated using the following equation (5) using two or more weighting coefficients α(i, j) for the multiple beams representing the pixel 36 and the current density J(i, j) of each of the multiple beams.

[0104]

[0105] Furthermore, it is not necessary to correct the irradiation time but the irradiation dose; the current density (current amount) can also be corrected. In this case, the correction factor K can be multiplied by the current density (current amount) or by a factor multiplied by a reference current density (current amount).

[0106] In the drawing step (S140), the drawing mechanism 150 uses the multi-beam 20 to draw a pattern on the sample 101, using a plurality of beams to draw multiple images of each pixel 36 of the sample 101 using the corrected irradiation doses of each beam. In other words, the drawing mechanism 150 uses the multi-beam 20 to draw a pattern on the sample 101, using a plurality of beams to draw multiple images of each pixel 36 of the sample 101 using the calculated individual irradiation times t'(i, j).

[0107] As described above, the irradiation dose of each beam is obtained by multiplying the current of each beam by the irradiation time. For each of the multiple beams, the irradiation time of each beam is corrected using two or more weighting coefficients and the current of each beam, and each beam is irradiated using the corrected irradiation time.

[0108] FIG. 17 is a diagram illustrating an example of correction coefficient distribution and quantization error in Comparative Example 1 of Embodiment 1. FIG. FIG. 18 is a diagram illustrating an example of correction coefficient distribution and quantization error in Comparative Example 2 of Embodiment 1. FIG. FIG19 is a diagram illustrating an example of correction coefficient distribution and quantization error in the first embodiment. In Figures 17, 18, and 19, the upper section shows an example of the in-plane distribution of the correction coefficient. The middle section shows an example of a schematic diagram of the variation of the correction coefficient for each block. The lower section shows an example of the quantitative error for each design dose.

[0109] In Comparative Example 1, as shown in the upper graph of Figure 17, the correction coefficient for the peripheral beam with low current density increases, prolonging the irradiation time. Furthermore, the irradiation dose remains uniform. However, as shown in area A in the middle graph of Figure 17, a uniquely large correction coefficient may be generated. As a result, the irradiation time at the specific point increases, lengthening the overall firing cycle. This results in longer drawing times and reduced yield. Furthermore, as shown in the lower graph of Figure 17, the quantization error is larger in pixels with low doses. Consequently, controllability of the pattern edge position tends to deteriorate.

[0110] In Comparative Example 2, as shown in the upper graph of Figure 18, averaging results in a nearly uniform correction coefficient value. Consequently, the irradiation time also becomes uniform. Furthermore, as shown in the middle graph of Figure 18, the nearly uniform correction coefficient shortens the maximum irradiation time. Consequently, the rendering time is shortened. However, as shown in the lower graph of Figure 18, the irradiation time of each beam tends to be the same, so quantization errors accumulate in the same direction, increasing the total quantization error.

[0111] In contrast, in Embodiment 1, as shown in the upper and middle graphs of Figure 19 , the correction coefficients do not become uniquely large, and can be made uniform. This prevents the firing cycle from becoming longer. Furthermore, as shown in the lower graph of Figure 19 , the correction coefficients do not become uniform, allowing the quantization errors of each beam to be averaged. As a result, the magnitude of the total quantization error can be improved.

[0112] FIG20 illustrates a first variation of the weighting coefficients in Embodiment 1. Coulomb force-induced blooming is a common occurrence during electron beam mapping. To mitigate the effects of the Coulomb force, dose levels can be varied between adjacent beams irradiated simultaneously. Therefore, as shown in FIG20 , the dose levels of each beam in the multi-beam 20 can be set to form a checkerboard pattern, for example. Therefore, the weighting coefficient α(i, j) is defined by the product of a plurality of independent weighting coefficient elements. In the example of FIG20 , the weighting coefficient α(i, j) set for each beam is preferably the product of the independent weighting coefficients α(i, j) (first weighting coefficient) and the weighting coefficient β(i, j) for Coulomb effect suppression (second weighting coefficient). Furthermore, the weighting coefficient β(i, j) for Coulomb effect suppression, for example, is set to 1.5 for one of the adjacent beams and 0.5 for the other in the first pass, for a multiplicity of 2. Then, for example, in the second pass, the Coulomb effect can be reduced by setting the value of one of the adjacent beams to 0.5 and the value of the other to 1.5.

[0113] FIG21 illustrates a second variation of the weighting coefficients in Embodiment 1. Among the multiple beams 20, there may be defective beams whose dose cannot be controlled or are constantly off. Therefore, the weighting coefficients for these defective beams are set to zero to achieve a corrected irradiation time of zero. Therefore, as shown in FIG21 , the weighting coefficient α(i, j) set for each beam is preferably the product of the independently described weighting coefficients α(i, j) (first weighting coefficient) and the weighting coefficient β(i, j) for defective beam determination (second weighting coefficient). The weighting coefficient β(i, j) for defective beam determination can be set to 1 for normal beams and 0 for defective beams. Furthermore, in this defect determination, constantly on beams cannot be controlled and are therefore eliminated. Furthermore, the design dose required for beams with β(i, j) = 0 can be allocated to beams in other passes.

[0114] As described above, according to the first embodiment, when the sample 101 is subjected to multiple patterning using the multi-beam 20, it is possible to suppress a decrease in yield and reduce quantization errors regardless of the presence or absence of beams with specific current densities.

[0115] Furthermore, the program for causing a computer to execute the processing described in the above-mentioned embodiment can also be designed to be stored in a non-transitory tangible readable recording medium such as a magnetic disk device.

[0116] Furthermore, although descriptions of device configurations and control methods that are not directly necessary for the description of the present invention are omitted, necessary device configurations and control methods can be appropriately selected and used. For example, descriptions of the control unit configuration for controlling the rendering device 100 are omitted, but of course, necessary control unit configurations can be appropriately selected and used. In addition, although the above-mentioned embodiments describe a drawing device, drawing method, and program using a charged particle beam, these are not limited to charged particles and are also applicable to a drawing device, drawing method, and program using a laser.

[0117] All other multi-charged particle beam drawing methods, multi-charged particle beam drawing devices, and programs that have the elements of the present invention and can be appropriately modified by those skilled in the art are included in the scope of the present invention.

[0118] 20:Multi-beam 22: Hole 24: Control electrode 25:Through hole 26: Opposite electrodes 29: Sub-irradiation area 30: Delineate area 32: Stripe area 34: Irradiation area 35: Rectangular area 36: Pixels 41: Control circuit 42: Pattern 50: Grid processing unit 52: Firing data generation unit 56: Current density distribution production unit 58: Weight coefficient setting unit 60: Correction coefficient calculation unit 62: Correction Department 72: Drawing control unit 74: Transfer Processing Department 100:Depicting device 101: Sample 102:Electron tube 103: Drawing Room 105:XY Platform 106: Faraday Cup 110: Control computer 112: Memory 130: Bias control circuit 132,134:DAC amplifier unit 136: Lens control circuit 138: Platform Control Mechanism 139: Platform position detector 140,142:Memory device 150:Depicting the Organization 160: Control system circuit 200:Electron beam 201:Electron Gun 202: Lighting lens 203: forming aperture array substrate 204: Submerged aperture array mechanism 205: Zooming out lens 206: limiting aperture substrate 207:Objective lens 208: Main deflector 209: auxiliary deflector 210: Mirror 330: Film area

Claims

1. A method for depicting multiple charged particle beams, wherein for each beam of a multiple charged particle beam, one of a plurality of weighting coefficients is set according to the arrangement position of the multiple charged particle beams. When multiple positions of a sample to be depicted are made multiple times using multiple beams of the multiple charged particle beams with different arrangement positions having two or more weighting coefficients set, the irradiation dose of the multiple beams irradiating the position is corrected by using two or more weighting coefficients of the multiple beams depicting the position. For each position of the sample, the position is made multiple times using the multiple beams with the corrected irradiation dose of each of the multiple beams, thereby depicting a pattern on the sample using the multiple charged particle beams.

2. The method for depicting multiple charged particle beams as described in claim 1, wherein, The irradiation amount of each of the aforementioned beams is obtained by multiplying the current of each of the aforementioned beams by the irradiation time. For each of the aforementioned plurality of beams, the irradiation time of each of the aforementioned beams is corrected by using the aforementioned two or more weighting coefficients and the aforementioned current of each of the aforementioned plurality of beams, and the aforementioned beams are irradiated by the corrected irradiation time.

3. The method for depicting multiple charged particle beams as described in claim 2, wherein, The weighting coefficients set for each of the aforementioned beams are set independently of the current of that beam.

4. The method for depicting multiple charged particle beams as described in claim 1, wherein, Weighting coefficients are assigned to each of the aforementioned beams so that the total amount of irradiation by the plurality of beams irradiating each position of the aforementioned sample becomes the design value.

5. The method for depicting multiple charged particle beams as described in claim 1, wherein, Different weighting coefficients are assigned to the aforementioned plurality of beams irradiating the same location of the aforementioned sample.

6. The method for depicting multiple charged particle beams as described in claim 1, wherein, As the weighting coefficients set for each beam, the product of the independent first weighting coefficient and the second weighting coefficient is used.

7. A multi-charged particle beam drawing apparatus, comprising: a setting circuit for setting one of a plurality of weighting coefficients for each beam of a plurality of charged particle beams according to the arrangement position of the plurality of charged particle beams; a correction circuit for correcting a pre-determined irradiation dose of each of the plurality of beams irradiating that position by using the two or more weighting coefficients of the plurality of beams irradiating that position when multiple drawings are made on the same position of a sample to be drawn by using a plurality of beams with different arrangement positions for drawing that position; and a drawing mechanism for drawing a pattern on the sample by using the plurality of charged particle beams to multiplely draw the position on each position of the sample by multiple drawings using the corrected irradiation dose of each of the plurality of beams.

8. A program for instructing a computer to perform: a process of setting one of a plurality of weighting coefficients for each beam of a plurality of charged particle beams according to the arrangement of the plurality of charged particle beams; a process of storing the weighting coefficient set for each beam in a memory device; and a process of reading the weighting coefficient from the memory device, and when multiple depictions of the same position of a sample to be depicted are performed by a plurality of beams of the plurality of charged particle beams with different arrangement positions having two or more weighting coefficients set, a process of correcting the pre-determined irradiation dose of the plurality of beams for each of the plurality of beams irradiating that position using the two or more weighting coefficients of the plurality of beams depicting that position, and outputting the result.

Images