JP2008233686A - Lithography simulation method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a simulation method ensuring high-precision lithography simulation in a simple manner. <P>SOLUTION: The simulation method includes step S2 of applying bias processing to a mask pattern to be simulated with reference to a reference table wherein relations between pattern sizes and pattern bias amounts are regulated and step S3 of obtaining an optical image of the mask pattern subjected to the bias processing. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a lithography simulation method and a program.

  With the miniaturization of the mask pattern (mask layout), it has become difficult to ensure the accuracy of lithography simulation.

  For example, a large difference has arisen between the simulation result obtained by the mask thin film approximation model and the simulation result obtained by rigorous calculation reflecting the mask topography effect (for example, Non-patent document 1). Therefore, in order to perform a high-accuracy simulation, it is necessary to perform a strict calculation reflecting the mask stereo effect, that is, a strict calculation considering an electromagnetic field near the mask. However, if a strict calculation is performed using a physical model reflecting the mask stereoscopic effect, there is a problem that the amount of calculation becomes enormous.

Therefore, conventionally, it has been difficult to perform high-precision lithography simulation in consideration of the mask solid effect by a simple method.
Proc.SPIE2005, vol.5754, p.383-394, March 2005

  An object of this invention is to provide the simulation method etc. which can perform a highly accurate lithography simulation by a simple method.

  The lithography simulation method according to the first aspect of the present invention includes a step of performing bias processing on a mask pattern to be subjected to lithography simulation with reference to a reference table in which a relationship between a pattern dimension and a pattern bias amount is defined. And obtaining an optical image of the mask pattern subjected to the bias process.

  The program according to the second aspect of the present invention is a program applied to lithography simulation, and the relationship between the pattern dimension and the pattern bias amount is defined for a mask pattern to be subjected to lithography simulation. The procedure for performing the bias processing with reference to the reference table and the procedure for obtaining the optical image of the mask pattern subjected to the bias processing are executed.

  According to the present invention, high-precision lithography simulation can be realized by a simple method.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a flowchart showing a basic procedure of a lithography simulation method according to an embodiment of the present invention.

  First, a reference table in which the relationship between the pattern dimension and the pattern bias amount considering the mask topography effect is defined is prepared (S1). Details will be described below.

  2 and 3 are explanatory diagrams showing the mask stereoscopic effect. 2A and 3A are diagrams schematically showing a photomask, in which 50 is a photomask, 51 is a light shielding portion, and 52 is a light transmitting portion. FIGS. 2B and 3B are diagrams showing the electric field distribution of light passing through the photomask 50, where P is the actual electric field distribution based on the mask steric effect, and Q is the mask thin film approximation (thin film). approximation) is an electric field distribution calculated using a model (a model assuming that the thickness of the pattern on the photomask is zero).

  The mask three-dimensional effect is roughly classified into an edge scattering effect, an oblique incidence effect, and a waveguide effect. The edge scattering effect is a phenomenon in which the electric field distribution is scattered at the boundary (edge) between the light shielding portion 51 and the light transmitting portion 52 of the photomask 50 as shown in FIG. It can happen. As shown in FIG. 3, the oblique incidence effect and the waveguide effect are phenomena in which light passing through the light transmitting part (opening part) 52 sandwiched between the light shielding parts 51 is attenuated, and the opening width ( As the space width becomes narrower, the oblique incidence effect and the waveguide effect increase. The oblique incidence effect is a phenomenon in which the light intensity is attenuated at the boundary (edge) between the light shielding portion 51 and the light transmitting portion 52 due to oblique incident light (light incident on the photomask 50 from an oblique direction). . The waveguide effect is also a phenomenon in which the intensity of light passing through the light transmitting portion (opening) 52 is attenuated, and the waveguide effect becomes particularly apparent when the opening width (space width) of the light transmitting portion 52 is narrowed. Here, the oblique incidence effect and the waveguide effect are defined as a mask stereo effect in a narrow sense.

  FIG. 4 is a diagram showing a calculation result of an optical image by simulation for a line and space pattern. As shown in FIG. 5, the optical image 14 is defined based on the light intensity distribution of the light that has passed through the projection optical system 13 obtained when the illumination light is irradiated onto the mask pattern 12 on the photomask 11. In FIG. 4, (a) is a calculation result of an optical image calculated using a strict physical model reflecting the mask stereoscopic effect. (B) is a calculation result of an optical image calculated using a mask thin film approximate model (a model in which the thickness of the pattern on the photomask is assumed to be zero).

  The line and space pattern used in the simulation has a line width L and a space width S of 50 nm. The mask used for the simulation is an attPSM (attenuated phase shift mask) with a magnification of 4. The simulation conditions are: exposure light wavelength: 193 nm, NA: 1.3, illumination: annular illumination.

  As can be seen from FIG. 4, there is an optical image (a) calculated using a strict physical model reflecting the mask stereo effect and an optical image (b) calculated using a mask thin film approximate model. There is a big error.

  A simulation similar to the above-described simulation was performed by changing the line width L and the space width S of the line and space pattern. That is, (L = 48 nm, S = 52 nm), (L = 49 nm, S = 51 nm), (L = 51 nm, S = 49 nm), (L = 52 nm, S = 48 nm), and so forth. A simulation using a mask thin film approximate model was performed by changing the width S. The change width of the line width L and the space width S was 0.25 nm. As a result, when (L = 53.75 nm, S = 46.25 nm), the simulation result (line width L = 50 nm, space width S = 50 nm) using a strict physical model reflecting the mask steric effect is the most. Near simulation results were obtained.

  FIG. 6 is a diagram showing a calculation result of an optical image by the above-described simulation. In FIG. 6, (a) is a calculation result of an optical image calculated using a strict physical model reflecting the mask stereo effect (line width L = 50 nm, space width S = 50 nm). (B) is the calculation result of the optical image calculated using the mask thin film approximate model (line width L = 53.75 nm, space width S = 46.25 nm). As shown in FIG. 6, by increasing the line width L by 3.75 nm and decreasing the space width S by 3.75 nm, a simulation result using a strict physical model is obtained by simulation using a mask thin film approximate model. It can be seen that it can be approximated very well.

  As can be seen from the above, by changing the line width L and the space width S of the pattern, the mask three-dimensional effect can be reflected in the lithography simulation in a pseudo manner. In other words, by changing the pattern bias amount, the mask stereoscopic effect can be reflected in the lithography simulation in a pseudo manner. In the example described above, the bias amount per edge, that is, the edge movement amount is 1.875 nm (= 3.75 nm / 2).

  As already described, as the space width of the opening portion becomes narrower, the mask three-dimensional effect (oblique incidence effect and waveguide effect) increases, and the attenuation factor of light passing through the opening portion increases. Therefore, the bias amount depends on the pattern dimensions such as the line width L and the space width S. Therefore, if a reference table that prescribes the relationship between the pattern dimension and the pattern bias amount is created in advance, the mask stereoscopic effect is reflected in the simulation in a simulated manner without using a strict physical model that reflects the mask stereoscopic effect. It is possible.

  FIG. 7 is a diagram illustrating an example of a reference table. In the example shown in FIG. 7, the space width S is used as the pattern dimension. Further, as shown in FIG. 8, the bias amount corresponds to the edge movement amount B of the opposing pattern 20 that defines the space width S. As shown in FIG. 7, in the reference table, the bias amount is defined according to the pattern dimension. For example, when the space width S is 50 nm, the bias amount is 1.625 nm for each of the opposing patterns that define the space width S. Therefore, in the lithography simulation, the calculation is performed assuming that the space width S ′ is 46.75 nm (= 50 nm− (1.625 × 2) nm).

  FIG. 9 is a flowchart showing a procedure for creating a reference table.

  First, a test pattern is prepared (S11). For this test pattern, an optical image is calculated using a strict physical model reflecting the mask stereo effect (that is, a physical model including the thickness of the pattern as a variable) (S12). For example, if the test pattern has a space width S of 50 nm, the exact calculation is performed with the space width S set to 50 nm. Specifically, the electric field distribution near the mask is obtained by numerical calculation based on the Maxwell equation. For example, numerical calculation is performed by FDTD (finite region time difference method) or RCWA (strict coupling wave analysis method) using the optical constant of the actual mask material and the incident angle of light as parameters. Then, using the electric field distribution in the vicinity of the mask obtained by numerical calculation, the intensity distribution of the light that has passed through the projection optical system is calculated, and the dimensions are calculated by slicing at a certain threshold.

  Next, the test pattern is perturbed (S13). That is, a bias amount B is given to the test pattern. Subsequently, an optical image is calculated using a mask thin film approximation model for the test pattern subjected to the perturbation process (bias process) (S14).

  Next, the optical image obtained in step S12 is compared with the optical image obtained in step S14 (S15). Further, based on the comparison result, it is determined whether or not the difference between the two satisfies a predetermined standard (whether or not the difference between the two is within a predetermined range) (S16). If the predetermined standard is satisfied, the bias amount B with respect to the space width S is determined (S17). If the predetermined standard is not satisfied, the process returns to step S13, and the bias amount B is changed to execute steps S14 to S16. In this way, steps S13 to S16 are repeated until a predetermined standard is satisfied.

  As described above, for a certain test pattern (for example, a test pattern having a space width S of 50 nm), the relationship between the pattern dimension and the bias amount is obtained. Similarly, for other test patterns (for example, test patterns with different space widths S), the relationship between the pattern dimension and the bias amount is obtained. A reference table is created based on the relationship between the pattern dimension and the bias amount obtained in this way.

  As described above, a reference table in which the relationship between the pattern dimension and the pattern bias amount is defined is prepared (S1, see FIG. 1).

  In the lithography simulation, a bias process is executed with reference to the reference table for the mask pattern to be simulated (S2). For example, for a pattern with a space width S of 50 nm, the bias amount is set to 1.625 nm with reference to the reference table of FIG. Therefore, the space width S after the bias processing is 46.75 nm (= 50 nm− (1.625 × 2) nm).

  Next, simulation is performed on the mask pattern subjected to the bias process, and an optical image of the mask pattern subjected to the bias process is obtained (S3). That is, an optical image is calculated by simple calculation using a mask thin film approximate model without using a strict physical model reflecting the mask stereoscopic effect. As can be seen from the above, the optical image obtained in this way is a simulated reflection of the mask stereo effect and is calculated using a strict physical model that reflects the mask stereo effect. It is in good agreement with the optical image.

  Hereinafter, evaluation results of the lithography simulation method of this embodiment will be described.

  FIG. 10 is a diagram showing a mask pattern (mask layout) used for evaluation. The line width of the pattern is L1, and the space width is S1 and S2. The dimension of L1 and S1 was fixed to 50 nm, and the dimension of S2 was made into four types, 50 nm, 70 nm, 100 nm, and 200 nm.

  FIG. 11 shows the evaluation results. (A) is an evaluation result when the lithography simulation method of this embodiment is applied. That is, it is an evaluation result when a simulation is performed using a mask thin film approximate model on a mask pattern subjected to bias processing. A dimension difference between the dimension of S1 calculated using the lithography simulation method of the present embodiment and the dimension of S1 calculated using a strict physical model reflecting the mask stereo effect is shown. (B) is an evaluation result when the lithography simulation method of the comparative example is applied. In the comparative example, the simulation is performed simply using the mask thin film approximation model without performing the bias process. A dimension difference between the dimension of S1 calculated using the lithography simulation method of the comparative example and the dimension of S1 calculated using a strict physical model reflecting the mask stereoscopic effect is shown.

  In the case of the comparative example, the dimensional difference is relatively small when the dimension of S2 is 70 nm and 100 nm, but the dimensional difference is very large when the dimension of S2 is 50 nm and 200 nm. On the other hand, in the present embodiment, the dimensional difference is small regardless of whether the dimension of S2 is 50 nm, 70 nm, 100 n, or 200 nm. Therefore, simulation accuracy can be improved by using the method of this embodiment.

  As described above, in this embodiment, a reference table in which the relationship between the pattern dimension and the pattern bias amount is defined is created in advance, and the bias is referred to the mask pattern to be subjected to lithography simulation with reference to the reference table. Apply processing. By performing such a bias process, an optical image of the mask pattern can be obtained by a simulation in which the mask stereoscopic effect is reflected in a pseudo manner. Therefore, according to the present embodiment, high-precision lithography simulation can be performed by a simple method.

  In the embodiment described above, the space width is assumed as the pattern dimension, but the line width may be assumed as the pattern dimension. Moreover, you may assume both a line width and a space width as a pattern dimension.

  Further, the lithography simulation method described in the above-described embodiment can be applied to a method for manufacturing a semiconductor device. FIG. 12 is a flowchart showing an outline of a method for manufacturing a semiconductor device.

  First, design data is prepared (S21), and lithography simulation is performed by the method described in the above embodiment (S22). Subsequently, mask data is generated from the design data based on the guideline obtained by the lithography simulation (S23). Further, a photomask is produced based on the generated mask data (S24). The pattern formed on the photomask thus fabricated is transferred to the photoresist on the semiconductor wafer (S25). Further, the photoresist is developed to form a photoresist pattern (S26), and etching is performed using the photoresist pattern as a mask to form a pattern on the semiconductor wafer (S27).

  Further, the method described in the above-described embodiment can be realized by a computer whose operation is controlled by a program in which the procedure of the method is described. The program can be provided by a recording medium such as a magnetic disk or a communication line (wired line or wireless line) such as the Internet.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the present invention. Furthermore, the above embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining the disclosed constituent elements. For example, even if several constituent requirements are deleted from the disclosed constituent requirements, the invention can be extracted as an invention as long as a predetermined effect can be obtained.

It is the flowchart which showed the basic procedure of the lithography simulation method which concerns on embodiment of this invention. It is explanatory drawing shown about the mask solid effect. It is explanatory drawing shown about the mask solid effect. It is the figure which showed the calculation result of the optical image by simulation. It is the figure shown about the optical image. It is the figure which showed the calculation result of the optical image by simulation. FIG. 6 is a diagram illustrating an example of a reference table according to the embodiment of the present invention. FIG. 6 is a diagram illustrating a space width and a bias amount according to the embodiment of the present invention. 6 is a flowchart illustrating a procedure for creating a reference table according to the embodiment of the present invention. It is the figure which showed the mask pattern used for evaluation of the lithography simulation method which concerns on embodiment and comparative example of this invention. It is the figure which showed the evaluation result of the lithography simulation method which concerns on embodiment and comparative example of this invention. 5 is a flowchart illustrating an outline of a method for manufacturing a semiconductor device according to an embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 ... Photomask 12 ... Mask pattern 13 ... Projection optical system 14 ... Optical image 20 ... Opposite pattern 50 ... Photomask 51 ... Light-shielding part 52 ... Translucent part

Claims (5)

A step of performing bias processing on a mask pattern to be subjected to lithography simulation with reference to a reference table in which a relationship between a pattern dimension and a pattern bias amount is defined;
Obtaining an optical image of the mask pattern subjected to the bias treatment;
A lithography simulation method comprising: The lithography simulation method according to claim 1, wherein an optical image of the mask pattern subjected to the bias processing is obtained using a mask thin film approximate model. In the reference table, the difference between the optical image of the mask pattern obtained using a physical model including the thickness of the pattern as a variable and the optical image of the mask pattern subjected to bias processing satisfies a predetermined criterion. The lithography simulation method according to claim 1, wherein a relationship between a pattern dimension and the pattern bias amount is defined. The lithography simulation method according to claim 1, wherein the pattern dimension includes at least one of a line width and a space width. A program applied to lithography simulation,
On the computer,
A procedure for performing bias processing on a mask pattern to be subjected to lithography simulation with reference to a reference table in which a relationship between a pattern dimension and a pattern bias amount is defined,
A procedure for obtaining an optical image of the mask pattern subjected to the bias processing;
A program for running

Images