JP5918478B2 - Charged particle beam drawing apparatus and charged particle beam drawing method - Google Patents

Description

  The present invention relates to a charged particle beam drawing apparatus and a charged particle beam drawing method, and for example, relates to an alignment method for drawing a pattern of a plurality of layers.

  Lithography technology, which is responsible for the progress of miniaturization of semiconductor devices, is an extremely important process for generating a pattern among semiconductor manufacturing processes. In recent years, with the high integration of LSI, circuit line widths required for semiconductor devices have been reduced year by year. In order to form a desired circuit pattern on these semiconductor devices, a highly accurate original pattern (also referred to as a reticle or a mask) is required. Here, the electron beam (electron beam) drawing technique has an essentially excellent resolution, and is used for producing a high-precision original pattern.

FIG. 12 is a conceptual diagram for explaining the operation of the variable shaped electron beam drawing apparatus.
The variable shaped electron beam (EB) drawing apparatus operates as follows. In the first aperture 410, a rectangular opening for forming the electron beam 330, for example, a rectangular opening 411 is formed. Further, the second aperture 420 is formed with a variable shaping opening 421 for shaping the electron beam 330 having passed through the opening 411 of the first aperture 410 into a desired rectangular shape. The electron beam 330 irradiated from the charged particle source 430 and passed through the opening 411 of the first aperture 410 is deflected by the deflector, passes through a part of the variable shaping opening 421 of the second aperture 420, and passes through a predetermined range. The sample 340 mounted on a stage that continuously moves in one direction (for example, the X direction) is irradiated. That is, the drawing area of the sample 340 mounted on the stage in which the rectangular shape that can pass through both the opening 411 of the first aperture 410 and the variable shaping opening 421 of the second aperture 420 is continuously moved in the X direction. Drawn on. A method of creating an arbitrary shape by passing both the opening 411 of the first aperture 410 and the variable shaping opening 421 of the second aperture 420 is referred to as a variable shaping method (VSB method).

  With the recent miniaturization of patterns, improvement in resolution in photolithography is required. One method for solving this is a phase shift method. Since the phase shift mask requires a two-layer pattern of a light shielding pattern layer and a halftone pattern layer, the alignment accuracy when these patterns are superimposed is important. For example, a cross mark pattern for alignment is created when the first layer pattern is formed, and the drawing position of the second layer pattern is adjusted using the position of the cross mark pattern. Here, since the cross mark pattern formed in the first layer for alignment is generally difficult to arrange in the actual pattern (main pattern) in the first layer, it is arranged around the main pattern. It will be. Therefore, it is often arranged near the limit position of the accuracy compensation area of the drawing accuracy or outside the limit position. For this reason, the cross mark pattern is more likely to have poor positional accuracy than the main pattern formed at the center. In addition, when a contamination or the like is attached to the first cross mark pattern, an incorrect position may be measured. When alignment for drawing the second layer is performed using such a cross mark pattern with poor positional accuracy, there is a problem that the overlay accuracy of the first and second layer patterns deteriorates.

  Here, although it is not an alignment method when two-layer patterns are overlaid by a drawing apparatus, a technique is disclosed in which position information having a predetermined accuracy is selected from position information of a plurality of alignment marks. (For example, refer to Patent Document 1).

JP 2005-11980 A

  Here, as described above, when alignment is performed using a cross mark pattern with poor positional accuracy, there is a problem in that the overlay accuracy of the first and second layer patterns deteriorates. Therefore, it is also considered to simply eliminate the mark with poor position accuracy, but it may not be possible to obtain the required alignment accuracy simply by removing the mark with poor position accuracy.

  In view of the above, an object of the present invention is to overcome the above-described problems and to provide a drawing apparatus and method for realizing a higher-precision alignment when drawing a plurality of layer patterns.

A charged particle beam drawing apparatus according to one embodiment of the present invention includes:
A stage for mounting a substrate on which a plurality of marks are formed together with a first layer pattern;
A mark position measuring unit for measuring from a result obtained by scanning the positions of a plurality of marks with a charged particle beam in a state where the substrate is placed on the stage;
An acquisition unit that obtains an approximate expression by fitting an amount of deviation from a desired position of a plurality of measured marks by an Nth order polynomial;
A correction unit that corrects the amount of positional deviation of the N-order component at the positions of the plurality of marks using such an approximate expression acquired based on the measured positions of the plurality of marks;
An abnormal mark detector that detects an abnormal mark among the positions of the corrected multiple marks,
Among the detected abnormal marks, a mark removing unit that removes positions of abnormal marks within a predetermined number,
An alignment calculation unit that performs alignment calculation using the positions of the remaining marks from which abnormal marks within a predetermined number have been removed, and
A drawing unit for drawing the second layer pattern at the aligned position;
It is provided with.

In addition, it further includes a sort processing unit that sorts the detected deviation amount of the abnormal mark in descending order,
It is preferable that the mark removing unit is configured to remove up to the predetermined number of abnormal marks in order from the larger deviation amount among the detected abnormal marks.

In addition, a variation amount calculation unit that calculates a variation amount for including the plurality of remaining marks described above at a predetermined rate with respect to the normal distribution of the positional deviation amounts of the plurality of remaining marks from which the abnormal marks have been removed. Further comprising
It is preferable that the drawing process is stopped when the variation amount exceeds the threshold value.

A storage unit for storing the positions of the plurality of measured marks;
An update unit that updates the positions of the plurality of marks stored in the storage unit to the positions of the remaining plurality of marks from which the abnormal marks have been removed when the variation amount does not exceed the threshold;
Further comprising
The alignment calculation unit preferably performs the alignment calculation using the positions of the plurality of marks stored in the updated storage unit.

The charged particle beam drawing method of one embodiment of the present invention includes:
A step of measuring from the result obtained by scanning the positions of the plurality of marks with a charged particle beam in a state where the substrate is placed on the stage on which the substrate on which the plurality of marks are formed together with the first layer pattern is placed; ,
Fitting an amount of deviation of the measured positions of the plurality of marks from a desired position by an Nth order polynomial to obtain an approximate expression;
Correcting the amount of misalignment of the N-order component of the positions of the plurality of marks using the approximate expression obtained based on the measured positions of the plurality of marks;
Detecting an abnormal mark among the positions of a plurality of marks after correction;
A step of removing positions of abnormal marks within a predetermined number of detected abnormal marks;
A step of performing an alignment calculation using the positions of the remaining marks from which abnormal marks within a predetermined number have been removed;
Drawing a second layer pattern at the aligned position;
It is provided with.

  According to one embodiment of the present invention, more accurate alignment can be realized when a plurality of layers of patterns are drawn in an overlapping manner. As a result, a pattern of a plurality of layers can be drawn at a highly accurate overlay position.

1 is a conceptual diagram illustrating a configuration of a drawing apparatus according to Embodiment 1. FIG. FIG. 5 is a flowchart showing main steps of the mask forming method in the first embodiment. FIG. 5 is a process cross-sectional view of the mask forming method in the first embodiment. FIG. 3 is a top view showing an example of a sample on which a first layer main pattern and a plurality of cross mark patterns around it are formed in the first embodiment. FIG. 3 is a conceptual diagram for explaining a measurement technique in the first embodiment. 6 is a diagram showing an example of a positional deviation amount and an example of a positional deviation component obtained from the measurement result of the cross mark pattern position in the first embodiment. FIG. 6 is a diagram illustrating an example of a mark position in which up to an N component error is corrected in the first embodiment. FIG. 6 is a diagram illustrating an example of mark positions subjected to sorting processing in Embodiment 1. FIG. 6 is a conceptual diagram for explaining removal of an abnormal mark in Embodiment 1. FIG. FIG. 6 is a diagram for explaining alignment calculation in the first embodiment. FIG. 5 is a process cross-sectional view of the mask forming method in the first embodiment. It is a conceptual diagram for demonstrating operation | movement of a variable shaping type | mold electron beam drawing apparatus.

  Hereinafter, in the embodiment, a configuration using an electron beam will be described as an example of a charged particle beam. 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. Further, a variable shaping type drawing apparatus will be described as an example of the charged particle beam apparatus.

Embodiment 1 FIG.
FIG. 1 is a conceptual diagram illustrating a configuration of a drawing apparatus according to the first embodiment. In FIG. 1, the drawing apparatus 100 includes a drawing unit 150 and a control unit 160. The drawing apparatus 100 is an example of a charged particle beam drawing apparatus. In particular, it is an example of a variable shaping type drawing apparatus. The drawing unit 150 includes an electron column 102 and a drawing chamber 103. In the electron column 102, an electron gun 201, an illumination lens 202, a first aperture 203, a projection lens 204, a deflector 205, a second aperture 206, an objective lens 207, a main deflector 208, a sub deflector 209, And a detector 212 are arranged. An XY stage 105 is disposed in the drawing chamber 103. On the XY stage 105, a sample 101 such as a mask to be a drawing target substrate at the time of drawing is arranged. The sample 101 includes an exposure mask for manufacturing a semiconductor device. Further, the sample 101 includes mask blanks to which a resist is applied and nothing is drawn yet. Alternatively, a mask in which a resist is applied on the already formed first layer pattern is included.

  The control unit 160 includes a control computer 110, a memory 112, an interface circuit 114, a deflection control circuit 120, an amplifier 214, and storage devices 140, 142, and 144 such as a magnetic disk device. The control computer 110, the memory 112, the interface circuit 114, the deflection control circuit 120, the amplifier 214, and the storage devices 140, 142, and 144 are connected to each other via a bus (not shown).

  In the control computer 110, there are a measurement unit 10, a recording unit 12, a reading unit 14, correction units 16, 18, a fitting processing unit 20, a correction unit 22, a detection unit 24, a sort processing unit 26, a removal unit 28, and a 3σ calculation unit. 30, a determination unit 32, an output unit 34, an update unit 36, an alignment calculation unit 38, a drawing data processing unit 40, and a drawing control unit 42 are arranged. Measurement unit 10, recording unit 12, reading unit 14, correction units 16, 18, fitting processing unit 20, correction unit 22, detection unit 24, sort processing unit 26, removal unit 28, 3σ calculation unit 30, determination unit 32, output Each function such as the unit 34, the update unit 36, the alignment calculation unit 38, the drawing data processing unit 40, and the drawing control unit 42 may be configured by hardware such as an electric circuit, a program for executing these functions, or the like. It may be configured by software. Alternatively, it may be configured by a combination of hardware and software. Measurement unit 10, recording unit 12, reading unit 14, correction units 16, 18, fitting processing unit 20, correction unit 22, detection unit 24, sort processing unit 26, removal unit 28, 3σ calculation unit 30, determination unit 32, output Information that is input to and output from the unit 34, the update unit 36, the alignment calculation unit 38, the drawing data processing unit 40, and the drawing control unit 42 is stored in the memory 112 each time.

  In the deflection control circuit 120, a deflection position correction unit 44 and a deflection amount calculation unit 46 are arranged. Each function such as the deflection position correction unit 44 and the deflection amount calculation unit 46 may be configured by hardware such as an electric circuit, or may be configured by software such as a program for executing these functions. Alternatively, it may be configured by a combination of hardware and software. Information input / output to / from the deflection position correction unit 44 and the deflection amount calculation unit 46 and information being calculated are stored in a memory (not shown) each time.

  Here, FIG. 1 shows a configuration necessary for explaining the first embodiment. The drawing apparatus 100 may normally have other necessary configurations. For example, the main deflector 208 and the sub-deflector 209, which are the main and sub two-stage multi-stage deflectors, are used for position deflection, but the position deflection is performed by one stage deflector or three or more stages of multi-stage deflectors. It may be the case.

  In the storage device 140 (storage unit), drawing data of the actual pattern of the first layer for forming the phase shift mask and the cross mark pattern for alignment are input from the outside and stored. In addition, the storage device 140 stores drawing data of a real pattern of the second layer for forming the phase shift mask from the outside.

  FIG. 2 is a flowchart showing main steps of the mask forming method according to the first embodiment. 2, the mask forming method in the first embodiment includes a first layer drawing step (S102), a developing step (S104), an etching step (S106), a mark position measuring step (S108), and mark position recording. Step (S110), mark information reading step (S112), deflection distortion correction step (S114), random distortion correction step (S116), N-order component fitting step (S118), and N-order component correction step (S120) ), Abnormal mark detection step (S122), sort processing step (S124), abnormal mark removal step (S126), remaining mark 3σ calculation step (S128), determination step (S130), and update step (S132). ), Alignment calculation step (S134), deflection position correction step (S136), second layer drawing step (S136), rinse and An image step (S140), performing a series of steps of the etching step (S142). Among these steps, the steps other than the developing step (S104), the etching step (S106), the rinsing and developing step (S140), and the etching step (S142) are performed in the drawing apparatus 100, for example.

  FIG. 3 is a process sectional view of the mask forming method according to the first embodiment. FIG. 3 shows from the first layer drawing step (S102) to the preparation stage of the mark position measuring step (S108).

  First, the sample 101 includes a glass substrate 60, a halftone film 62, a light shielding film 64, and a resist film 66. Specifically, a halftone film 62 is formed on the glass substrate 60. For example, a MoSi film is preferably used as the halftone film 62. A light shielding film 64 is formed on the halftone film 62. For example, a chromium (Cr) film is preferably used as the light shielding film 64. On the light shielding film 64, an electron beam resist film 66 is formed. At this stage, a mask blank state in which no pattern is drawn is obtained.

  In FIG. 3A, as the first layer drawing step (S102), the main pattern of the first layer, which becomes an actual pattern, is drawn at the center (actual pattern region) of the sample 101. Then, a plurality of cross mark patterns are drawn in the mark area around the first layer main pattern.

  Drawing of these patterns is performed in the drawing apparatus 100. First, the sample 101 in which the halftone film 62, the light-shielding film 64, and the resist film 66 are formed in this order on the glass substrate 60 is transported and placed on the XY stage 105.

  Then, the drawing data processing unit 40 reads the drawing data of the first layer from the storage device 140, performs a plurality of stages of data conversion processing, and generates shot data of the pattern of the first layer. In order to draw a graphic pattern by the drawing apparatus 100, it is necessary to divide each graphic pattern defined in the drawing data into a size that can be irradiated with one beam shot. Therefore, the drawing data processing unit 40 generates shot figures by dividing each figure pattern into a size that can be irradiated with one shot of a beam in order to actually draw. Then, shot data is generated for each shot figure. In the shot data, for example, graphic data such as a graphic type, a graphic size, and an irradiation position are defined. Further, irradiation amount data is also generated. The generated shot data and irradiation amount data are stored in the storage device 142.

  The deflection amount calculation unit 46 reads shot data from the storage device 142 and calculates the deflection amount to be applied to each deflector. In FIG. 1, the configuration indicating connection to other than the main deflector 208 is omitted, but a deflection amount is calculated for each deflector, and a deflection voltage corresponding to the deflection amount applied to the corresponding deflector is applied. It goes without saying that it will be. Then, the drawing unit 150 draws a first layer main pattern and a plurality of alignment cross mark patterns on the sample 101 in accordance with the shot data and the dose data. Specifically, it operates as follows.

  The electron beam 200 emitted from the electron gun 201 (emission unit) illuminates the entire first aperture 203 having a rectangular hole, for example, a rectangular hole, by the illumination lens 202. Here, the electron beam 200 is first formed into a rectangle, for example, a rectangle. Then, the electron beam 200 of the first aperture image that has passed through the first aperture 203 is projected onto the second aperture 206 by the projection lens 204. The deflector 205 controls the deflection of the first aperture image on the second aperture 206, and can change (variably shape) the beam shape and dimensions. The electron beam 200 of the second aperture image that has passed through the second aperture 206 is focused by the objective lens 207, deflected by the main deflector 208 and the sub deflector 209, and continuously moved. The desired position of the sample 101 arranged in the above is irradiated. FIG. 1 shows a case in which multi-stage deflection of main and sub two stages is used for position deflection. In such a case, the electron beam 200 of the corresponding shot is deflected while following the stage movement to the reference position of the subfield (SF) in which the drawing region is virtually divided by the main deflector 208, and each sub-deflector 209 deflects each electron beam 200 in the SF. What is necessary is just to deflect the beam of the shot concerning an irradiation position.

  In FIG. 3B, first, as a developing step (S104), the sample 101 conveyed from the drawing apparatus 100 is developed. Thereby, for example, the resist in the region irradiated with the electron beam 200 is removed, and a resist pattern is formed.

  Next, as an etching step (S106), the exposed light shielding film 64 and halftone film 62 are removed by etching using the resist pattern as a mask. Then, by removing the resist film by ashing or the like, the first-layer main pattern 152 and a plurality of cross mark patterns 151 around it can be formed.

  In FIG. 3C, a resist film 67 and an antistatic film 69 are formed on the sample 101 on which the first layer main pattern 152 and a plurality of cross mark patterns 151 around it are formed. A sample 101 for drawing is prepared.

  FIG. 4 is a top view showing an example of a sample in which the first layer main pattern and a plurality of cross mark patterns around it are formed in the first embodiment. In FIG. 4, in the actual pattern region 50, a first layer actual pattern (main pattern) is formed. A plurality of cross mark patterns 52 are formed around the main pattern so as to surround the main pattern. Here, as shown in FIG. 4, among the plurality of cross mark patterns 52, an abnormal mark 54 indicating a positional deviation exceeding a threshold value may be formed. The number of abnormality marks 54 may be one or plural. In FIG. 4, three abnormality marks 54 are shown as an example. In reality, it is assumed that there are more abnormal marks.

  Here, in the first embodiment, when performing alignment for overlaying the second layer pattern, the alignment mark accuracy is calculated by excluding the abnormal mark pattern from the plurality of cross mark patterns up to the set removal number. To improve. Further, in the first embodiment, it is also evaluated whether or not the alignment accuracy can maintain the allowable accuracy by using the position information of the remaining cross mark pattern after removing the abnormal mark pattern. In the following, description will be given in order.

  As the mark position measurement step (S108), first, the sample 101 for drawing the pattern of the second layer shown in FIG. 3 is transferred onto the XY stage 105 of the drawing apparatus 100. From the result obtained by the measurement unit 10 scanning the positions of all the cross mark patterns 52 including the abnormal mark 54 with the electron beam 200 while the sample 101 (substrate) is placed on the XY stage 105. taking measurement. The measurement unit 10 is an example of a mark position measurement unit.

  FIG. 5 is a conceptual diagram for explaining the measurement technique in the first embodiment. The mark position is scanned (beam scan) by deflecting the electron beam 200 using the main deflector 208 and / or the sub-deflector 209 in the drawing apparatus 100. In addition to the antistatic film 69 and the resist film 67, the glass substrate 60 is formed at the position of the cross mark pattern. On the other hand, a light shielding film 64 is formed around the periphery. Therefore, except for the cross mark pattern, the electron beam 200 passes through the antistatic film 69 and the resist film 67 and is reflected by the light shielding film 64 to generate reflected electrons. On the other hand, in the cross mark pattern, since the light shielding film 64 and the halftone film 62 do not exist, the electron beam passes through the glass substrate 60 as it is. Therefore, the position of the cross mark pattern can be measured by detecting the reflected electrons. Alternatively, the position of the cross mark pattern can be measured by the contrast between the reflected electrons reflected by the light shielding film 64 and the reflected electrons reflected by the glass substrate 60. The reflected electrons are detected by the detector 212, the analog output is amplified by the amplifier 214, and is output to the measurement unit 10 as a digital signal. Further, when measuring the position of each mark, the XY stage 105 may be moved for each mark to a position where the mark to be measured can be scanned with the electron beam 200.

  As the mark position recording step (S110), the recording unit 12 records the measured position information (mark information) of each mark in the storage device 144.

  FIG. 6 is a diagram illustrating an example of a positional deviation amount and an example of a positional deviation component obtained from the measurement result of the cross mark pattern position in the first embodiment. The measured cross mark pattern is measured with the position shifted. FIG. 6A shows an example of a coordinate system misalignment including the misalignment amount of the cross mark pattern and the misalignment amount of the first layer main pattern. As a factor that causes a positional shift in such a coordinate system, in other words, a cross mark pattern, first, there is a deflection distortion of an electron beam caused by the optical system of the drawing apparatus 100. Therefore, in the drawing apparatus 100, the amount of such deflection distortion is obtained in advance by experiments before drawing. For example, a mark (not shown) on the XY stage 105 is scanned while the main deflector 208 deflects the electron beam. It is only necessary to measure at a plurality of positions in the main deflection region and obtain the amount of deflection distortion depending on the deflection position. Then, the obtained plurality of deflection distortion amounts are fitted with polynomials, and as shown in FIG. 6B, the deflection distortion amount f (x, y) with the coordinates (x, y) of the deflection position as variables. ) Here, there is an amount of deflection distortion at a part that cannot be defined by such f (x, y). As for the amount of random deflection distortion that cannot be defined by f (x, y), as shown in FIG. 6C, the deflection distortion amount g based on the map correlated with the coordinates (x, y) of the deflection position. Find (x, y). Thus, the information of f (x, y) and g (x, y) obtained in advance before drawing is input from the outside into the drawing apparatus 100 as part of parameters.

  Further, as shown in FIG. 6D, there is an N-order component error L (x, y) at coordinates (x, y) when the positional deviation amount of the cross mark pattern is fitted with a polynomial. For example, an arrangement position shift (shift in the x and y directions and a rotation shift) when the sample 101 is transported to and placed on the XY stage 105, a scale error when measuring the position of the cross mark pattern, and the like are also included. is there. There are zero-order components such as deviations in the x and y directions and errors of primary components such as rotational deviation and scale.

  In addition, there is an error p (x, y) due to variations in the beam scan itself as shown in FIG. 6E and other correction residual component q (x, y) as shown in FIG. 6F.

  Therefore, first, as the mark information reading step (S112), the reading unit 14 reads each position information of the plurality of cross mark patterns measured from the storage device 144.

  In the deflection distortion correction step (S114), the correction unit 16 corrects the deflection distortion amount f (x, y) for each cross mark pattern.

  Subsequently, as a random distortion correction step (S116), the correction unit 18 corrects the deflection distortion amount g (x, y) for each cross mark pattern.

In the Nth-order component fitting step (S118), the fitting processing unit 20 obtains an approximate expression by fitting the deviation amounts of a plurality of mark positions from a desired position (for example, a design position) with an Nth-order polynomial. The fitting processing unit 20 is an example of an acquisition unit. Specifically, the fitting processing unit 20 performs the Nth-order component for each mark position in which the deflection distortion amount f (x, y) and the random deflection distortion amount g (x, y) are corrected. In order to correct the error L (x, y), fitting is performed with an Nth order polynomial. For example, fitting is performed using a first-order polynomial (1) shown below. Here, the N-order component error in the x direction is denoted by Δx, and the N-order component error in the y direction is denoted by Δy.
(1) Δx = a ₀ + a ₁ x + a ₂ y, Δy = b ₀ + b ₁ x + b ₂ y

  In the N-order component correction step (S120), the correction unit 22 corrects the N-order component error L (x, y) (position shift amount of the N-order component) of the positions of the plurality of marks using the obtained approximate expression. To do.

  FIG. 7 is a diagram showing an example of mark positions corrected up to the N component error in the first embodiment. From the amount of misalignment in the state before correction, which is the result of measuring each cross mark pattern shown in FIG. 7B, the deflection distortion amount f (x, y) shown in FIG. 7C and FIG. By subtracting the deflection distortion amount g (x, y) shown in d) and the Nth-order component error L (x, y) shown in FIG. 7E, an error p (as shown in FIG. 7A) is obtained. x, y) and the corrected residual component q (x, y) can be corrected to a state in which they remain. In the first embodiment, the above-described deflection distortion amount f (x, y), random deflection distortion amount g (x, y), and N-order component error L ( After correcting x, y), an abnormal mark pattern formed at a defective position is detected from a plurality of cross mark patterns. By using the mark position after correcting the deflection distortion amount f (x, y), the random deflection distortion amount g (x, y), and the Nth-order component error L (x, y), an abnormal mark having a poor position is obtained. The pattern can be detected correctly.

  As the abnormal mark detection step (S122), the detection unit 24 detects an abnormal mark among the positions of the corrected marks described above. The detection unit 24 is an example of an abnormal mark detection unit. Specifically, a threshold value is set, and the determination is made based on whether or not the cross mark pattern to be inspected is relatively shifted from the surrounding other cross mark pattern by more than the threshold value. For example, the determination is made based on whether or not the deviation is 30 nm or more. For example, when a plurality of cross mark patterns are arranged in order in the same direction, the determination can be made based on whether or not a deviation from the direction line by a threshold value or more. Here, it is assumed that other cross mark patterns in the vicinity are also displaced, but can be averaged by increasing the number of parameters of the other cross mark patterns in the periphery used for the determination. Therefore, it is preferable to determine by the amount of deviation from the average position.

  Here, a plurality of abnormal marks may be detected by such detection. However, if the position information of all detected abnormal marks is removed as it is from the subsequent calculations, the number of parameters of the remaining cross mark pattern is reduced. Therefore, in the first embodiment, the number of abnormal marks to be removed is limited.

  As the sort processing step (S124), the sort processing unit 26 sorts the detected abnormal mark deviation amounts in descending order.

  FIG. 8 is a diagram illustrating an example of mark positions subjected to sorting processing in the first embodiment. As shown in FIG. 8, the detected abnormality marks are arranged from the larger deviation amount to the smaller deviation amount.

  As the abnormal mark removal step (S126), the removal unit 28 removes the positions of the abnormal marks within the set number of removals (predetermined number) from the detected abnormal marks. The removal unit 28 is an example of a mark removal unit. Specifically, the removal unit 28 removes the abnormal marks from the detected abnormal marks up to the maximum number of removals set in order from the larger deviation amount. The removal number may be set in advance in the drawing apparatus 100 as a part of the parameter.

  FIG. 9 is a conceptual diagram for explaining removal of abnormal marks in the first embodiment. In FIG. 9A, a first main pattern is formed in the actual pattern region 50 in the sample 101, and a plurality of cross mark patterns 52 including an abnormal mark 54 formed in the first layer around the actual pattern region 50. It is shown. On the other hand, FIG. 9B shows a cross mark pattern 52 that is adopted after the abnormal mark 54 is removed in the abnormal mark removing step (S126). Needless to say, the abnormality mark 54 is not physically removed from the sample 101 but the position information of the abnormality mark 54 is excluded from the calculation in the subsequent steps.

  As the remaining mark 3σ calculation step (S128), the variation amount calculation unit includes the remaining marks at a predetermined rate of probability with respect to the normal distribution of the positional deviation amounts of the remaining marks from which the abnormal marks have been removed. The amount of variation to be calculated is calculated. Here, as an example, the 3σ calculation unit 30 calculates 3σ of the positional deviation amounts of the remaining marks. 3σ is a variation amount (positional deviation amount from the average value) for including the remaining plurality of marks with a probability of 97%. Alternatively, the maximum positional deviation amount of the remaining plurality of marks may be obtained.

  As the determination step (S130), the determination unit 32 determines whether or not the variation amount (3σ) exceeds the threshold value α. If the threshold value α is exceeded (3σ> α), the drawing control unit 42 stops the drawing process. Then, the output unit 34 outputs the result via the interface circuit 114. For example, it outputs to a monitor (not shown). Or it is good to output to the display lamp etc. which are not illustrated. Further, the determination unit 32 determines whether or not the variation amount (3σ) exceeds another threshold value β even if the variation amount (3σ) is equal to or less than the threshold value α. Even when the threshold value α is equal to or less than the threshold value α, when the threshold value β exceeds another threshold value β (β <3σ ≦ α), the output unit 34 outputs the result as warning information via the interface circuit 114. For example, it outputs to a monitor (not shown). Or it is good to output to the display lamp etc. which are not illustrated. For example, α = 30 nm, β = 20 nm

  As described above, in the first embodiment, it is evaluated whether or not the accuracy of the remaining cross mark pattern from which the abnormal mark is eliminated can be used for the alignment for the second layer pattern. If the allowable accuracy is not satisfied, useless machine time can be reduced by stopping the drawing process. Further, it is possible to reduce the cost required for drawing a wasteful second layer. In addition, when a certain amount of positional deviation occurs, a warning can be output to allow the user to select whether or not to continue the drawing process.

  Here, if a part of the cross mark pattern group around the main pattern has been displaced in the same direction, the above-described abnormal mark removal step (S126) detects the relative position, so the cross It is difficult to detect an abnormal mark from the mark pattern group. On the other hand, the larger the number of parameters of the remaining plurality of marks, the more the average amount of positional deviation is averaged. Therefore, by limiting the number of removed abnormal marks in the abnormal mark removing step (S126), the population parameter can be kept above a certain level. Then, by calculating 3σ with the parameter maintained at a certain level or more, it is possible to evaluate the cross mark pattern groups that have been displaced in the same direction. Therefore, it is possible to improve the evaluation accuracy of the remaining cross mark patterns by limiting the number of abnormal mark removals.

  As the update step (S132), the update unit 36 determines the positions of the plurality of marks stored in the storage device 144 as the positions of the remaining marks from which the abnormal marks have been removed when the variation amount does not exceed the threshold value α. Update to Here, the positional information (mark information) after correcting the deflection distortion amount f (x, y) and the random deflection distortion amount g (x, y) described above is used as the position information of the remaining marks. It is preferable.

  By configuring as described above, it is possible to exclude the position information of the abnormal mark pattern within a predetermined number and evaluate whether the position accuracy of the remaining cross mark patterns is more than a certain level. Therefore, by using the position information of the remaining cross mark pattern that has passed the evaluation, it is possible to perform alignment calculation with higher accuracy.

  As the alignment calculation step (S134), the alignment calculation unit 38 performs alignment calculation using the positions of the remaining marks from which the abnormal marks within the set number of removals have been removed. Specifically, the alignment calculation unit 38 performs alignment calculation using the positions of the plurality of marks stored in the updated storage device 144. For example, the alignment calculation unit 38 performs fitting with an Nth order polynomial to correct the Nth order component error N (x, y) for each updated mark position. For example, fitting is performed using the above-described equation (1). As a result, it is possible to obtain each coefficient of the equation (1) for obtaining the N-order component error in the x direction by Δx and the N-order component error in the y direction by Δy. The obtained coefficient is output to the deflection position correction unit 44.

  FIG. 10 is a diagram for explaining alignment calculation in the first embodiment. From the mark measurement result shown in FIG. 10A, the deflection distortion amount f (x, y) shown in FIG. 10B and the deflection distortion amount g (x, y) shown in FIG. 10C are subtracted. It is preferable to perform alignment calculation based on the mark information after this. In other words, from the mark information in the state in which the Nth-order component error L (x, y), the error p (x, y), and the corrected residual component q (x, y) shown in FIG. A polynomial coefficient is obtained to correct the alignment N-order component error N (x, y) shown in (e).

  On the other hand, in parallel with the above-described process for evaluating the cross mark pattern, the drawing data conversion process for the second layer pattern is performed. Specifically, the drawing data processing unit 40 reads the drawing data of the second layer from the storage device 140, performs a plurality of stages of data conversion processing, and generates shot data of the pattern of the second layer. Further, irradiation amount data is also generated. The generated shot data and irradiation amount data are stored in the storage device 142.

  As the deflection position correction step (S136), the deflection position correction unit 44 reads out shot data from the storage device 142, inputs each coefficient of equation (1), and aligns each position of the shot data by equation (1). Correction is performed with correction amounts (Δx, Δy). In addition, each position of the shot data may be corrected with the above-described deflection distortion amount f (x, y) and random deflection distortion amount g (x, y) by another configuration (not shown). Alternatively, the deflection position correction unit 44 also corrects the above-described deflection distortion amount f (x, y) and random deflection distortion amount g (x, y) from each position of the shot data. good.

  FIG. 11 is a process sectional view of the mask forming method according to the first embodiment. FIG. 11 shows the second layer drawing process (S136) to the etching process (S142).

  In FIG. 11A, as the second layer drawing step (S136), the deflection amount calculation unit 46 determines the deflection amount to be applied to each deflector so that the electron beam is irradiated to the position of the aligned shot data. Calculate. In FIG. 1, the configuration indicating connection to other than the main deflector 208 is omitted, but a deflection amount is calculated for each deflector, and a deflection voltage corresponding to the deflection amount applied to the corresponding deflector is applied. It goes without saying that it will be. Then, the drawing unit 150 draws the actual pattern (main pattern) of the second layer at the aligned position on the sample 101 according to the shot data and the dose data.

  In FIG. 11B, first, as a rinsing and developing step (S140), the antistatic film 69 is first washed away with water by rinsing the sample 101 transported from the drawing apparatus 100, and then developing processing is performed. Thereby, the antistatic film 69 and, for example, the resist in the region irradiated with the electron beam 200 are removed, and a resist pattern is formed.

  Next, as an etching step (S142), the exposed light shielding film 64 is removed by etching using the resist pattern as a mask. Then, the second-layer main pattern can be formed by removing the resist film by ashing or the like. As described above, the phase shift mask can be formed.

  As described above, according to the first embodiment, more accurate alignment can be realized when a plurality of layers of patterns are overlaid. As a result, it is possible to reduce deterioration in drawing accuracy due to misalignment. Therefore, a pattern of a plurality of layers can be drawn at a highly accurate overlay position. Therefore, mask loss due to misalignment can also be reduced. In addition, since the accuracy of the remaining marks is evaluated, useless drawing time can be reduced.

  The embodiments have been described above with reference to specific examples. However, the present invention is not limited to these specific examples.

  In addition, although descriptions are omitted for parts and the like that are not directly required for the description of the present invention, such as a device configuration and a control method, a required device configuration and a control method can be appropriately selected and used. For example, although the description of the control unit configuration for controlling the drawing apparatus 100 is omitted, it goes without saying that the required control unit configuration is appropriately selected and used.

  In addition, all charged particle beam writing apparatuses and methods that include 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.

DESCRIPTION OF SYMBOLS 10 Measurement part 12 Recording part 14 Reading part 16, 18 Correction part 20 Fitting process part 22 Correction part 24 Detection part 26 Sort process part 28 Removal part 30 3 (sigma) calculation part 32 Determination part 34 Output part 36 Update part 38 Alignment calculation part 40 Drawing Data processing unit 42 Drawing control unit 44 Deflection position correction unit 46 Deflection amount calculation unit 50 Actual pattern region 52 Cross mark pattern 54 Abnormal mark 60 Glass substrate 62 Halftone film 64 Light shielding film 66, 67 Resist film 69 Antistatic film 100 Drawing apparatus 101, 340 Sample 102 Electronic column 103 Drawing chamber 105 XY stage 110 Control computer 112 Memory 114 Interface circuit 120 Deflection control circuit 140, 142, 144 Storage device 150 Drawing unit 151 Cross mark pattern 152 Main pattern 160 Control unit 200 201 Electron gun 202 Illumination lens 203, 410 First aperture 204 Projection lens 205 Deflector 206, 420 Second aperture 207 Objective lens 208 Main deflector 209 Sub deflector 212 Detector 214 Amplifier 330 Electron beam 411 Opening 421 Variable shaping Aperture 430 Charged particle source

Claims (5)

A stage for mounting a substrate on which a plurality of marks are formed together with a first layer pattern;
A mark position measuring unit for measuring from a result obtained by scanning the positions of the plurality of marks with a charged particle beam in a state where the substrate is placed on the stage;
An acquisition unit for acquiring an approximate expression by fitting the measured amount of deviation from the desired position of the plurality of marks by the N-th order polynomial,
A correction unit that corrects a positional deviation amount of the N-order component at the positions of the plurality of marks using the approximate expression acquired based on the measured positions of the plurality of marks;
An abnormal mark detection unit for detecting an abnormal mark among the positions of the plurality of marks after correction;
Among the detected abnormal marks, a mark removing unit that removes positions of abnormal marks within a predetermined number,
An alignment calculation unit that performs an alignment calculation using the positions of the remaining marks from which the abnormal marks within the predetermined number have been removed;
A drawing unit for drawing the second layer pattern at the aligned position;
A charged particle beam drawing apparatus comprising: It further includes a sort processing unit that sorts the detected deviation amount of the abnormal mark in descending order,
2. The charged particle beam drawing apparatus according to claim 1, wherein the mark removing unit removes up to the predetermined number of abnormal marks in order from a larger deviation amount among detected abnormal marks. It further includes a variation amount calculation unit for calculating a variation amount for including the remaining plurality of marks at a predetermined rate with respect to the normal distribution of the positional deviation amounts of the remaining plurality of marks from which the abnormal marks have been removed. ,
The charged particle beam drawing apparatus according to claim 1, wherein the drawing process is stopped when the variation amount exceeds a threshold value. A storage unit for storing the measured positions of the plurality of marks;
An update unit that updates the positions of the plurality of marks stored in the storage unit to the positions of the remaining plurality of marks from which the abnormal mark has been removed when the variation amount does not exceed a threshold;
Further comprising
The charged particle beam drawing apparatus according to claim 1, wherein the alignment calculation unit performs alignment calculation using a plurality of mark positions stored in the updated storage unit. Measured from a result obtained by scanning the positions of the plurality of marks with a charged particle beam in a state where the substrate is placed on a stage on which a substrate on which a plurality of marks are formed together with the first layer pattern is placed. Process,
A step of acquiring a by fitting approximation formula by N order polynomial amount of deviation from the measured desired position of the plurality of marks,
Correcting the amount of positional deviation of the N-order component at the positions of the plurality of marks using the approximate expression acquired based on the measured positions of the plurality of marks;
Detecting an abnormal mark among the positions of the plurality of marks after correction;
A step of removing positions of abnormal marks within a predetermined number of detected abnormal marks;
A step of performing alignment calculation using the positions of the remaining marks from which the abnormal marks within the predetermined number have been removed;
Drawing a second layer pattern at the aligned position;
A charged particle beam drawing method comprising:

Images