JP2009092792A - Method for evaluating photomask transfer characteristic - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for evaluating transfer characteristics of a photomask pattern by which objectivity and reliability are enhanced, problems such as aberration in an optical system, an error in light reception of a CCD or pixel resolution are not caused, a data amount is reduced upon simulation, and the load of analysis is reduced. <P>SOLUTION: In the method for evaluating transfer characteristics of the photomask, the transfer pattern is formed by irradiating the photomask through an illumination optical system with illumination light from a light source to project the pattern of the photomask onto a wafer to expose through a projection optical system. The transfer characteristics are evaluated by analyzing the diffracted light by the photomask pattern and calculating contrast and a mask error enhancing factor (MEEF) on the wafer. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a method for evaluating transfer characteristics of a photomask pattern used in an optical transfer technique of semiconductor lithography.

  The formation of a circuit pattern of a semiconductor element or the like generally requires a process called a photolithography technique. In this process, a photomask (hereinafter also simply referred to as a mask) pattern is usually applied to a semiconductor wafer or the like using a projection exposure apparatus. A method of transferring onto an exposed substrate is employed. A photosensitive photoresist is applied on the substrate to be exposed, and a circuit pattern is transferred to the photoresist in accordance with the pattern shape of the mask pattern.

  In order to realize high integration and ultra-miniaturization of semiconductor elements that progress to a half pitch of 45 nm on the wafer, in the photolithography technology, the numerical aperture of the projection lens has been increased as a high resolution technology in the exposure apparatus. Developments such as a high NA technique, an immersion exposure technique in which exposure is performed with a high refractive index medium interposed between the projection lens and the object to be exposed, and an exposure technique with a modified illumination are being rapidly developed.

  On the other hand, resolution improvement measures for photomasks used in photolithography technology include phase reduction using light interference along with miniaturization and high accuracy of conventional binary masks that consist of light-transmitting parts and light-shielding parts. Development and commercialization of phase shift masks that improve resolution by the shift effect are in progress.

  In photolithography technology, the minimum dimension (resolution) that can be transferred by a projection exposure apparatus is proportional to the wavelength of light used for exposure and inversely proportional to the numerical aperture (NA) of the lens of the projection optical system. With the demand for the reduction in wavelength, the exposure light has been shortened and the projection optical system has a higher NA. However, there is a limit to satisfying this requirement only by the shorter wavelength and the higher NA.

Therefore, in order to increase the resolution, a super-resolution technique for miniaturization by reducing the value of process constant k ₁ (k ₁ = resolution line width × numerical aperture of projection optical system / exposure light wavelength) has recently been proposed. Has been. As such a super-resolution technique, a mask pattern is optimized by giving an auxiliary pattern or a line width offset to the mask pattern in accordance with the characteristics of the exposure optical system, or a modified illumination method (also referred to as an oblique incidence illumination method). There is a method called. In the modified illumination method, usually, annular illumination using a pupil filter, dipole illumination using a dipole (also referred to as Dipole) and quadrupole (quadrupole: Cquad) are used. ) Quadrupole illumination using a pupil filter is used.

  When the half pitch is 50 nm or less, the photomask pattern transfer characteristic has a great influence on the production of the semiconductor element. In recent years, as a method for evaluating transfer characteristics of a photomask, a process window evaluation using wafer dimension measurement based on exposure amount variation, a mask error enhancement factor using wafer dimension measurement based on mask size variation (Mask Error Enhancement Factor, hereinafter also referred to as MEEF). ) A method such as evaluation is performed (for example, see Patent Document 1).

The contrast indicating the photomask transfer performance and MEEF are important factors for evaluating the transfer characteristics. Conventionally, the contrast is calculated by the following equation (8) by obtaining the light intensity on the wafer. As shown in FIG. 13, the top of the light intensity on the wafer is I _top , and the bottom value of the light intensity is I _bottom . When the contrast is high (max. 1), the exposure margin is widened, the line edge roughness is improved, and the yield of the photolithography process is improved.

The mask error increasing factor (MEEF) is conventionally calculated by the following formula (9), and is represented by the ratio of the pattern dimension change amount (ΔWaferCD) on the wafer to the mask dimension change amount (ΔMaskCD). CD indicates a critical dimension of a mask or wafer. Numerical value 4 in the formula (9) is a reduction ratio of the mask, and exemplifies a case where a 4 × mask is used. As shown in Equation (9), when the MEEF value is lower (near 1), the mask pattern is faithfully transferred to the wafer pattern, and when the MEEF is lowered, the wafer manufacturing yield is improved. Also, the yield of mask production used for wafer production is improved.

Conventionally, in a simulation or a lithography simulation microscope (aerial image measurement system, hereinafter also referred to as AIMS: Aerial Image Measurement System), contrast and MEEF are evaluated by evaluating the light intensity distribution on the imaging surface.
JP-A-9-236906

  However, in the conventional photomask pattern evaluation method described above, the method of actual transfer onto the wafer depends on the process factors such as the performance of the exposure apparatus, resist characteristics, process stability, and dimensional measurement accuracy. There was a problem of lack of reliability. Also, the conventional photomask pattern evaluation method that measures and evaluates the light intensity distribution on the imaging plane by AIMS is said to depend on the aberration of the optical system, the light receiving sensitivity error of the image sensor CCD, the pixel resolution, and the noise reduction effect. There was a problem.

  As an example in which the pixel resolution and noise reduction in AIMS are adversely affected, FIG. 14 shows the results of an investigation of the contrast of half pitch (hp) 40 nm to 70 nm on the wafer. As a result, as shown in FIG. 14, the conventional evaluation method for measuring and evaluating the light intensity distribution on the imaging plane (rectangular points in the figure) by AIMS performs simulation with a dense line and space (L & S) profile ( There was a problem that the deviation from the curve in FIG.

  On the other hand, simulation is suitable as a prior evaluation method for actual semiconductor manufacturing, and it can be predicted with high accuracy depending on the accuracy of the model. However, the conventional photomask pattern evaluation method that calculates and evaluates the light intensity distribution on the imaging plane uses a Fourier imaging simulation for evaluating transfer characteristics in addition to the conventional electromagnetic field analysis simulation output results. There is a problem that the amount of data of the light intensity distribution on the image plane is accumulated, the amount of data to be stored and analyzed becomes enormous, and the analysis load increases further.

  Therefore, the present invention has been made in view of the above problems. That is, the object of the present invention is high in objectivity and reliability, does not cause problems such as optical system aberration, CCD light receiving sensitivity error, pixel resolution, etc., and reduces the amount of data and the analysis load in simulation. Another object is to provide a method for evaluating transfer characteristics of a photomask pattern.

  In order to solve the above problems, a photomask pattern transfer characteristic evaluation method according to the invention of claim 1 irradiates the photomask with illumination light emitted from a light source via an illumination optical system, and A method for evaluating a transfer characteristic of a photomask pattern when a pattern is projected onto a wafer via a projection optical system to form a transfer pattern, wherein diffracted light by the photomask pattern is analyzed and contrast on the wafer is analyzed And the mask error increasing factor (MEEF) are calculated to evaluate the transfer characteristics.

A photomask pattern transfer property evaluation method according to a second aspect of the present invention is the photomask pattern transfer property evaluation method according to the first aspect, wherein the photomask pattern is a line and space pattern, and the diffracted light is It is at least two-beam interference light of 0th-order light and 1st-order light, and the sum of the light intensities of diffracted light that can be taken in by the projection optical system is I _sum , the amplitude of _0th- order light is A _0th , and the amplitude of primary light is A _1st , phase shift is φ, CD of photomask pattern is MCD, target CD of transfer pattern on wafer is hp, light intensity on wafer when it is target CD is I _slice , mask dimension is wafer When the change ratio of the sum of the diffracted light intensities when the scale is changed by 1 nm is ΔDose, the contrast (Contrast) is expressed by Equation (1), Click error enhancement factor (MEEF) is characterized in that calculated by Equation (2) and Equation (3).

According to a third aspect of the present invention, there is provided a photomask pattern transfer characteristic evaluation method according to the second aspect, wherein a dipole pupil filter is used in the illumination optical system. When the diffraction angles of the 0th-order light and the first-order light are the same, the light intensity of the _0th- order light of the diffracted light is I _0th , and the light intensity of the first-order light is I _1st , the total light intensity of the diffracted light I _sum is expressed by Formula (4).

According to a fourth aspect of the present invention, there is provided a photomask pattern transfer characteristic evaluation method according to the second aspect, wherein a quadrupole pupil filter is used in the illumination optical system. When the light intensity of the _0th- order light of the diffracted light is I _0th , the light intensity of the primary light is I _1st , and the light intensity of the diffracted light that does not contribute to resolution is I _YTE , the total light intensity I _sum of the diffracted light is It is characterized by the expression (5).

  The photomask pattern transfer property evaluation method according to claim 5 is the photomask pattern transfer property evaluation method according to any one of claims 1 to 5, wherein the analysis by the diffracted light is performed by lithography simulation. The measurement is performed using a microscope and / or an ellipsometer.

  According to the photomask pattern transfer characteristic evaluation method of the present invention, it is not necessary to measure and analyze the light intensity distribution on the imaging surface as in the prior art, and the photomask pattern transfer characteristic can be evaluated only by analyzing the diffracted light. As a result, it is possible to obtain a result of transfer characteristics independent of the aberration of the optical system, the light receiving sensitivity of the CCD, and the pixel resolution. By measuring the transmitted 0th-order light in advance on a glass part without a mask pattern, the aberration of the optical system and the light receiving sensitivity error of the CCD can be canceled, and the problem of pixel resolution and noise reduction is the measurement of diffracted light. There is an advantage that there is no influence in nature. In the simulation, it is possible to reduce the data amount and the analysis load by using only the output result of the electromagnetic field analysis simulation. The photomask pattern transfer characteristic evaluation method of the present invention can be used during diffracted light measurement and simulation of an ellipsometer, AIMS, etc., and it is possible to easily evaluate transfer characteristics.

  Hereinafter, a method for evaluating transfer characteristics of a photomask pattern according to an embodiment of the present invention will be described in detail with reference to the drawings.

(First embodiment)
The photomask pattern transfer characteristic evaluation method of the present invention analyzes and evaluates diffracted light from a photomask pattern. FIG. 1 is a schematic diagram of diffracted light generated by a mask pattern when a dipole pupil filter is used in an illumination optical system as a first embodiment of the present invention. When the mask pattern is a general line and space (L & S) pattern as a semiconductor element, and the illumination light is irradiated onto the photomask, the diffracted light by the photomask pattern is a two-beam interference light of 0th order light and 1st order light. Shows the case.

In FIG. 1, the light intensity of the _0th- order light of the diffracted light is I _0th , the light intensity of the 1st-order light is I _1th , the amplitude of the _0th- order light of the diffracted light is A _0th , the amplitude of the primary light is A _1st , and the phase difference Assuming that the phase difference is φ, the contrast on the wafer (Contrast) is expressed by the following equation (6). Here, I = A2 and φ = 0 ^th phase-1 ^st phase. The light intensity of the diffracted light is measured by providing a CCD camera or the like on the light receiving side, and the amplitude can be calculated from the light intensity by a known method.

Next, in the first embodiment shown in FIG. 1, the photomask pattern CD is MCD, the target CD targeted for the transfer pattern on the wafer is hp (half pitch), and diffraction is performed when the target CD is the target CD. Assuming that the light intensity of light is I _slice , and the change ratio of the total diffracted light intensity when the mask dimension is changed by 1 nm on the wafer scale is ΔDose, in this embodiment, the mask error increasing factor (MEEF) is expressed by Equation (2) and It is calculated by Expression (3). Formula (6) is used for Contrast of Formula (2). In the above target CD, target CD = pattern pitch on the wafer / 2 = hp.

  In the present invention, in calculating ΔDose in the above formula (3), the mask dimension change amount dMCD is preferably set to a smaller value in order to increase calculation accuracy. Normally, dMCD is a numerical value range of about 1 to 4 nm. Is preferable.

(Second Embodiment)
FIG. 2 is a schematic diagram of diffracted light generated by a mask pattern when a quadrupole (Cquad) pupil filter is used in the illumination optical system as a second embodiment of the present invention. In the present embodiment, when the mask pattern is a line and space (L & S) pattern, and the illumination light is irradiated onto the photomask, the diffracted light by the photomask pattern is the two-beam interference light of the zeroth order light and the first order light. The case of light that does not contribute to resolution is shown. FIG. 3 is an explanatory diagram of contrast with the wafer position (wafer-position) on the horizontal axis and the light intensity (Intensity) on the vertical axis.

In FIG. 3, when the top of the light intensity on the wafer is I _top and the bottom value of the light intensity is I _bottom , the conventional contrast is (I _top −I _bottom ) / (I _top + I _bottom ).
In FIG. 2, the light intensity of the _0th- order light of the diffracted light is I _0th , the light intensity of the 1st-order light is I _1th , the light intensity of the diffracted light that does not contribute to resolution is I _YTE , and the amplitude of the 0th-order light of the diffracted light is When A _0th , the amplitude of the primary light is A _1st , and the phase difference is φ, I _top + I _bottom = I _0th + I _1st + I _YTE , I _top −I _bottom = A _0th A _1st cos (φ) The contrast on the wafer (Contrast) is expressed by the following formula (7). Here, I = A2 and φ = 0 ^th phase-1 ^st phase. The light intensity of the diffracted light is measured by providing a CCD camera or the like on the light receiving side, and the amplitude can be calculated from the light intensity by a known method.

Next, in the second embodiment shown in FIG. 2, the photomask pattern CD is MCD, the target CD targeted for the transfer pattern on the wafer is hp (half pitch), and diffraction is performed when the target CD is the target CD. Assuming that the light intensity of light is I _slice and the change ratio of the total diffracted light intensity when the mask dimension is changed by 1 nm on the wafer scale is ΔDose, in this embodiment, the mask error increasing factor (MEEF) is the same as in the first embodiment. Are calculated by the above formulas (2) and (3). However, in the case of the present embodiment, Expression (7) is used for Contrast of Expression (2). In the above target CD, target CD = pattern pitch on the wafer / 2 = hp.

Equations (6) and (7) showing the contrast (Contrast) of the first and second embodiments are expressed by the following equation (1) when the total light intensity of the diffracted light is summarized as I _sum. be able to.

When the above formula (1) is used, the sum I _sum of the diffracted light intensity that can be taken in by the projection optical system is expressed by the following formula (4) when a dipole pupil filter is used in the illumination optical system. When a quadrupole (Cquad) pupil filter is used for the illumination optical system, it is expressed by the following equation (5).

  FIG. 4 is a schematic plan view of a mask pattern for explaining a mask CD error (MCD), and FIG. 5 is a diagram showing a change in intensity of mask transmitted light due to the mask CD error. As shown in FIG. 5, the change in the light intensity due to the mask CD error can be replaced with the change ratio (ΔDose) of the total diffracted light intensity. In the present invention, ΔDose [% / nm] indicates a dose change amount (%) when the mask CD changes by 1 nm in terms of dimensions on the wafer, and is expressed by Equation (3). FIG. 5 exemplifies a case in which the change (represented by a chain line) is replaced with a decrease in the dose amount y% due to a mask CD error.

  As shown in the above equation (2), the mask error increasing factor (MEEF) is inversely proportional to the contrast (Contrast) and proportional to ΔDose. As shown in Equation (2) and Equation (3), the photomask pattern transfer characteristic evaluation method of the present invention can predict MEEF only by analyzing diffracted light.

  In pattern formation of 1: 1 lines and spaces with a half pitch of 50 nm or less on a wafer, the influence of resolution performance and mask manufacturing errors on transfer changes rapidly due to the influence of the three-dimensional effect of the mask. According to the photomask pattern evaluation method of the present invention, the contrast (Contrast) and the mask error increasing factor (MEEF) can be obtained by measuring and analyzing the diffracted light in experiments and simulations, thereby facilitating transfer characteristics. Can be evaluated.

(Example)
Two types of masks, molybdenum silicide halftone masks and chromium binary masks, which are frequently used as photomasks, were prepared. Next, the transfer characteristics of the two types of photomasks were evaluated using a lithography simulation microscope (AIMS) and simulation. FIG. 6 is a schematic diagram of an AIMS optical system that has analyzed the diffracted light of the present invention. As shown in FIG. 6, the intensity of diffracted light was measured with a CCD 65 using a σ aperture 61 with a monopole pupil filter in the illumination system. On the other hand, EM-Suite (trade name: manufactured by Panoramic Technology) was used as simulation software for estimating the transfer characteristics of the mask pattern. Further, a non-constant scattering coefficient model was used for the three-dimensional electromagnetic field simulation of the photomask by the FDTD method (also referred to as a time domain difference method or a finite difference time domain method) by TEMPESTpr2 (EM-Suite option).

  FIG. 7 and FIG. 8 show the results of diffracted intensity (Diffraction intensity) with respect to the mask CD (maskCD) by AIMS and simulation. The mask CD is converted into a dimension (nm) on the wafer. FIG. 7 shows the results of a molybdenum silicide halftone mask, and FIG. 8 shows the results of a chromium binary mask. The intensity of diffracted light is normalized to 1 by previously measuring a glass portion without a mask pattern. Therefore, the aberration of the optical system and the light receiving sensitivity error of the CCD are cancelled. The curves in FIGS. 7 and 8 are the results of the diffracted light intensities obtained by simulation. The first-order light, the zero-order light, and the diffracted light that does not contribute to resolution are respectively expressed as sim 1st, sim 0th, sim. Indicated by YTE. In addition, each rectangular point in the figure in contact with the curve is a result of AIMS measurement, and the first-order light, the 0th-order light, and the diffracted light that does not contribute to the resolution are sequentially expressed as aims 1st, aim 0th, and aim YTE. Show. As shown in FIG. 7 and FIG. 8, the result by AIMS and the result by simulation are almost the same, indicating that the reliability of the diffracted light intensity measurement of the mask pattern using AIMS is high.

  Next, on the basis of the results shown in FIGS. 7 and 8, a wafer as a mask pattern transfer characteristic evaluation item using the above mathematical formulas (1), (2), (3) and (5). The upper contrast (Contrast) and mask error enhancement factor (MEEF) were determined.

  FIG. 9 shows the result of contrast with respect to the mask CD (maskCD), and shows both a molybdenum silicide halftone mask and a chromium binary mask. The curves in FIG. 9 are simulation results, and the halftone mask is represented by HT-M sim and the binary mask is represented by Bi-M sim. In addition, each rectangular point in the figure that is substantially in contact with the curve is a numerical value obtained from the result of AIMS measurement, and the halftone mask is represented by HT-M aim and the binary mask is represented by Bi-M aim. The result of contrast obtained from the diffracted light analysis by AIMS and the result of the simulation are almost the same, which shows that the reliability of the method by the diffracted light analysis of the mask pattern is high.

  FIG. 10 shows the result of the mask error enhancement factor (MEEF) for the mask CD (maskCD), and shows both a molybdenum silicide halftone mask and a chromium binary mask. The curves in FIG. 10 are simulation results, and the halftone mask is indicated by HT-M sim and the binary mask is indicated by Bi-M sim. In addition, each rectangular point in the figure that is substantially in contact with the curve is a numerical value obtained from the result of AIMS measurement, and the halftone mask is represented by HT-M aim and the binary mask is represented by Bi-M aim. The result of MEEF obtained from the diffracted light analysis by AIMS and the result of the simulation are almost the same, indicating that the reliability of the method of the present invention by the diffracted light analysis of the mask pattern is high.

(Comparative example)
For comparison, a conventional photomask pattern evaluation method for measuring and evaluating the light intensity distribution on the imaging surface by AIMS was performed.
An optical system similar to that shown in FIG. 6 was used as the optical system for the AIMS, but the Bertra lens was omitted in order to measure the light intensity distribution on the imaging surface. The simulation is the same as described above, but in addition to the conventional electromagnetic field analysis simulation output result, data of the light intensity distribution on the imaging surface using Fourier imaging simulation for evaluating transfer characteristics was accumulated and used.

  Based on the results of the AIMS measurement, the above-described conventional equations (8) and (9) are used to obtain the contrast (Contrast) on the wafer and the mask error enhancement factor (MEEF), which are the mask pattern transfer characteristics evaluation items. It was. FIG. 11 shows the contrast results, and FIG. 12 shows the MEEF results. The meanings of the alphabetical symbols in FIGS. 11 and 12 are the same as those in FIGS. 9 and 10 described above.

  As is apparent from FIGS. 11 and 12, the contrast and MEEF results obtained by the conventional transfer characteristic evaluation method based on the light intensity distribution measurement on the imaging surface of the AIMS are different from the simulation results for both the binary mask and the halftone mask. Is shown.

It is a schematic diagram of the diffracted light generated by the mask pattern when the dipole pupil filter is used as the first embodiment of the present invention. It is a schematic diagram of diffracted light generated by a mask pattern when a quadrupole pupil filter is used as the second embodiment of the present invention. It is explanatory drawing of contrast. It is a top view of the mask pattern explaining a mask CD error. It is a figure which shows the intensity | strength change of the mask transmitted light by a mask CD error. It is a schematic explanatory drawing of the optical system of the lithography simulation microscope (AIMS) used for the diffraction light analysis of this invention. It is a figure which shows the diffracted light intensity with respect to mask CD of the halftone mask by AIMS and simulation. It is a figure which shows the diffracted light intensity with respect to mask CD of the binary mask by AIMS and simulation. It is the figure which calculated | required the contrast with respect to mask CD by the transfer characteristic evaluation method of the mask pattern of this invention. It is the figure which calculated | required MEEF with respect to mask CD by the transfer characteristic evaluation method of the mask pattern of this invention. It is the figure which calculated | required the contrast with respect to mask CD by the transfer characteristic evaluation method of the conventional mask pattern. It is the figure which calculated | required MEEF with respect to mask CD by the transfer characteristic evaluation method of the conventional mask pattern. It is explanatory drawing of the light intensity on the wafer for calculating | requiring contrast by the conventional method. It is the contrast investigation result of the half pitch (hp) 40-70 nm on the wafer by the conventional method.

Explanation of symbols

61 σ aperture 62 mask 63 NA aperture 64 Bertrand lens 65 CCD

Claims (5)

Illumination light emitted from a light source is irradiated onto a photomask via an illumination optical system, and a pattern on the photomask is projected onto a wafer via a projection optical system to form a transfer pattern. A transfer property evaluation method comprising:
Analyzing the diffracted light by the photomask pattern and calculating the transfer characteristic by calculating the contrast on the wafer and a mask error increasing factor (MEEF) . The photomask pattern is a line and space pattern, the diffracted light is at least a zero-order light and a first-order two-beam interference light, and the sum of the light intensities of the diffracted light that can be taken in by the projection optical system is I _sum The amplitude of the zero-order light is A _0th , the amplitude of the primary light is A _1st , the phase difference is φ, the CD of the photomask pattern is MCD, the target CD of the transfer pattern on the wafer is hp, and the target CD is When the light intensity on the wafer is I _slice and the change rate of the total diffracted light intensity when the mask dimension is changed by 1 nm on the wafer scale is ΔDose, the contrast (Contrast) is expressed by Equation (1), 2. The photomask pattern transfer according to claim 1, wherein the mask error increasing factor (MEEF) is calculated by Equations (2) and (3). Gender evaluation method.
When a dipole pupil filter is used in the illumination optical system, the diffraction angles of the zero-order light and the first-order light are the same, and the light intensity of the zero-order light of the diffracted light is _expressed as I _0th , 3. The method for evaluating transfer characteristics of a photomask pattern according to claim 2, wherein the sum I _sum of the diffracted light intensity is expressed by Equation (4), where I _1st is the light intensity of the photomask.
When a quadrupole pupil filter is used in the illumination optical system, the light intensity of the zero-order light of the diffracted light is I _0th , the light intensity of the first-order light is I _1st , and the light intensity of the diffracted light that does not contribute to resolution 3. The photomask pattern transfer characteristic evaluation method according to claim 2, wherein the sum I _sum of the diffracted light intensity is expressed by Equation (5), where I _YTE is:
  6. The photomask pattern transfer characteristic evaluation method according to claim 1, wherein the analysis using the diffracted light is performed using a lithography simulation microscope and / or an ellipsometer.