JPH06267833A - Electron beam lithography - Google Patents

Abstract

PURPOSE:To prevent forward scattering in resist and deterioration of pattern dimension precision due to a development process, by a method wherein only the contribution of backward scattering electrons in each lithography position is calculated, and the maximum dose or irradiation time corresponding with the contribution of the backward scattering electrons is determined by referring to a table for reference. CONSTITUTION:An electron beam outputted from an electron gun 21 is turned ON-OFF by a deflector 23 for blanking. By adjusting the irradiation time in this case, the dose is changed according to the irradiation position. Firstly, input data having the contribution U of backward scattering electrons as a parameter are formed at each position of a lithography pattern. Secondly, the data are inputted in an EB device, and a table for reference which is to be used is selected. From the table, the EB device referrs to the irradiation time (t) corresponding with U at each irradiation position, and determines (t). A pattern is written with the determined dose (t).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electron beam writing technique, and more particularly to an electron beam writing method for reducing the proximity effect.

[0002]

2. Description of the Related Art In recent years, an electron beam drawing apparatus has been used for drawing a fine pattern on a sample such as a semiconductor wafer or a mask substrate. In this apparatus, the backscattered electrons cause the pattern to become thick or thin. The influence of so-called proximity effect becomes a problem.

As a method of reducing the proximity effect, there is a dose correction method in which the dose is adjusted depending on the location based on the size and density of the pattern. Conventionally, when determining the dose by this dose correction method, a method using a matrix (M.Parik
h, J.Appl.Phys.p4371, p4378, p4383 (1979)), method using approximate formula (T.Abe, S.Yoshikawa, T.Takigawa, JJ)
AP vol.30,3B, p521-531 (1991), JM, Pavkovich, J.Va
c. Sci.Technol.B4,159 (1986)) and the like are used. Although there is a difference in the method used for determining the dose, these methods determine the optimum dose that keeps the total absorbed energy absorption of the exposed portion in the resist constant regardless of the density of the pattern,
By drawing with this irradiation amount, the effect of the proximity effect is reduced.

However, these methods have the following problems. That is, when the pattern size is miniaturized, the blur of the incident electron beam and the amount of forward scattering in the resist become nonnegligible as compared with the pattern size, and the pattern dimensional accuracy is degraded.

Further, since the amount of energy absorbed in the unexposed portion differs depending on the density of the pattern, the optimum development time of each pattern also varies. For this reason, when the patterns are miniaturized, the patterns on the same wafer have a large difference in formation size due to the difference in pattern density under the same development conditions.

These problems cannot be solved by a conventional correction method in which the total absorbed energy absorption amount of the exposed portion in the resist is constant regardless of the density of the pattern. In addition, many parameters are required for an ideal correction formula that takes these effects into consideration, which increases the time required for correction calculation.

[0007]

As described above, in the conventional electron beam writing method, the size of the blur of the incident electron beam and the amount of forward scattering in the resist cannot be ignored compared with the pattern size. If it has, or if the ratio of the dissolution rate of the exposed portion of the resist and the dissolution rate of the non-exposed portion is not sufficiently large, and the film loss or line thinning of the pattern that occurs during development cannot be ignored compared to the pattern size,
It has been difficult to accurately expose a fine pattern.

The present invention has been made in view of the above circumstances, and its object is to reduce the pattern dimensional accuracy due to the proximity effect, the blurring of the incident electron beam, the forward scattering in the resist, and the developing process. An object of the present invention is to provide an electron beam drawing method capable of preventing the above and significantly reducing the time required for correction calculation.

[0009]

The essence of the present invention is to provide a reference table of the optimum irradiation amount or irradiation time corresponding to the contribution of backscattered electrons in advance, and set the optimum irradiation amount for each drawing position in the pattern. When calculating, only the contribution of backscattered electrons at each drawing position is calculated, and the optimal irradiation amount or irradiation time according to the contribution of the backscattered electrons is determined by referring to a reference table, and the proximity effect or It is intended to prevent deterioration of pattern dimensional accuracy due to blurring of an incident electron beam, forward scattering in a resist, and a development process, and to significantly reduce a time required for determining a dose.

That is, according to the present invention (Claim 1), prior to irradiating an electron beam on a sample to draw a desired pattern,
Find the optimal dose for each position in the pattern to be drawn,
In the electron beam drawing method for drawing each pattern with this optimum irradiation amount, the steps (1) and (2) below are performed as the steps for determining the optimum irradiation amount. (1) Obtain the contribution of backscattered electrons from itself and the surroundings at each drawing position in the pattern to be drawn. (2) The optimum irradiation amount or irradiation time for each drawing position is determined using the reference table of the optimum irradiation amount or irradiation time corresponding to the contribution of the backscattered electrons which is created in advance.

Further, the present invention (claim 2) is characterized in that the steps (1) to (3) below are performed as a step of creating the reference table. (1) Draw line & space patterns with several different pattern densities and find the optimum dose of each pattern. (2) Create an interpolation formula corresponding to the pattern density, that is, the contribution of backscattered electrons, from the obtained optimum dose. (3) Create a lookup table corresponding to the contribution of backscattered electrons from the interpolation formula.

[0012]

According to the present invention, a reference table for the optimum irradiation amount or irradiation time corresponding to the contribution of backscattered electrons is provided in advance, the contribution of backscattered electrons at each drawing position is calculated, and the contribution is calculated. By determining the optimum irradiation dose or irradiation time according to the table, it is possible to prevent deterioration of pattern dimensional accuracy due to proximity effect, blur of incident electron beam, forward scattering in resist, and development process. It is possible to significantly reduce the time required to determine the dose.

Here, as an example, it is considered that a plurality of types of line & space patterns having different pattern densities are formed by using a negative resist by using a variable shaped beam type electron beam drawing apparatus. First, the reason why the conventional dose correction method causes a large difference in the formation dimension between these patterns on the same wafer under the same developing condition will be described.

In the conventional method, the influence of the proximity effect is reduced by obtaining the optimum irradiation amount that makes the energy absorption amount of the exposed portion in the resist constant irrespective of the density of the pattern and drawing with this irradiation amount. However, when the pattern is miniaturized, a large dimensional error occurs due to the density of the pattern as shown in FIG. Here, the following two can be cited as the causes of the dimensional error. (1) Absorption energy distribution (beam blur, forward scattering in resist) (2) Difference in energy absorption amount in unexposed area where development progresses First, the decrease in pattern dimensional accuracy due to the absorption energy distribution will be described. Figure 2 shows that the pattern size is larger than the spread of incident beam blur and forward scattering in the resist.
The absorbed energy distribution in the resist by the conventional correction method when miniaturized to the same degree is shown. In the figure, 1 indicates an energy distribution due to contribution of forward scattered electrons, and 2 indicates an energy distribution due to contribution of back scattered electrons.

The energy distribution due to the contribution of forward scattered electrons in the resist can be represented by a Gaussian distribution having a half-width of the desired size of the electron beam. The size of the full width at half maximum depends on the acceleration voltage and the resist film thickness, and does not depend on the magnitude of the irradiation amount. When 4 in the drawing indicates the half-value width of the energy distribution in the central portion, 6 in the drawing corresponds to the half-value width at the end portion, resulting in a difference in the amount of energy absorption. Comparing the widths of the energy distributions with the same energy absorption amount (3 in the figure), the central part (4 in the figure)
Is wider than the end (5 in the figure). That is, the difference in the widths of 5 and 6 in the drawing causes the dimensional accuracy to be lowered.

Next, the difference in the energy absorption amount of the unexposed portion due to the density of the pattern and the decrease in the pattern dimension accuracy due to this difference will be described with reference to FIG. FIG. 3 shows changes in the resist shape from the start of development to the end of development for the isolated remaining pattern, isolated isolated pattern, and line & space pattern. Further, it is assumed that the exposure is subjected to proximity effect correction by the conventional method. FIG. 3A shows the optimum development time of the isolated remaining pattern, FIG. 3B shows the optimum development time of the line & space pattern, and FIG. 3C shows the resist shape of each pattern at the optimum development time of the isolated remaining pattern. Shown respectively.

In the case of a negative resist, an unexposed portion is immediately developed in a portion having a low pattern density in which contribution of backscattered electrons is small. The development time is generally set according to the place where the contrast of the energy absorption amount between the exposed portion and the unexposed portion is the worst. That is, in the example of FIG. 3, since the development time is set according to the isolated removal pattern, the exposed portion that should be left originally is exposed to the developing solution for a long time in a portion having a low pattern density such as an isolated leaving, and line thinning occurs. Occur. This line thinning amount cannot be ignored when the pattern size is small.

As described above, according to the two reasons described with reference to FIGS. 2 and 3, the conventional dose correction method in which the energy absorption amount of the exposed portion in the resist is constant regardless of the density of the pattern. Patterns of different densities on the same wafer have a large difference in formation size under the same developing conditions.

Next, by determining the optimum dose according to the present invention, it is possible to prevent the proximity effect, the blurring of the incident electron beam, the forward scattering in the resist, and the deterioration of the pattern dimensional accuracy due to the development process. The reason why the time required to determine the quantity can be greatly reduced will be described.

In the present invention, an interpolation formula corresponding to the contribution of backscattered electrons is created. However, the optimum dose of the line-and-space pattern used when creating the formula is already the blur of the beam, forward scattering, and development. Includes the effects of the process. Therefore, by performing drawing with the optimum dose derived from the interpolation formula, it is possible to prevent the proximity effect, the blurring of the incident electron beam, the forward scattering of electrons in the resist, and the deterioration of the pattern dimensional accuracy due to the development process. It becomes possible.

When an effect similar to the above interpolation formula is to be obtained by using an approximate expression, the developing process must be taken into consideration in addition to the absorbed energy distribution. Therefore, the approximation formula must be complicated. On the other hand, the interpolation formula used in the present invention can be obtained relatively easily, and the optimum dose can be determined with an accuracy equal to or higher than that of a complicated approximate expression.

Further, in the conventional method, it takes a huge amount of time to calculate the optimum dose for each drawing position. However, in the present invention, only the contribution of backscatter for each drawing position is obtained, and the optimum irradiation amount is determined by referring to the table. Generally, reading the corresponding data from the selected memory table enables faster processing than performing the calculation. Similarly, even when an approximate expression is used to determine the optimum dose, it is possible to reduce the time required for correction calculation by creating the reference table in advance. In addition to this, when an approximate expression is used, it is possible to construct a system that can flexibly respond to changes in formulas.

[0023]

Embodiments of the present invention will be described below with reference to the drawings. (Embodiment 1) FIG. 1 is a schematic configuration diagram showing an electron beam drawing apparatus used in the method of the first embodiment of the present invention. In the figure, 10 is a sample chamber, 11 is a target (sample), 12 is a sample stand, 20 is an electron optical lens barrel, 21 is an electron gun, and 22a to 22.
Reference numeral e is various lens systems, 23 to 26 are various deflection systems, 27a is a blanking plate, and 27b and 27c are beam forming aperture masks. Further, 31 is a sample table drive circuit unit, 32 is a laser length measurement system, 33 is a deflection control circuit unit, and 34 is a deflection control circuit unit.
Is a variable shaped beam size control circuit unit, 35 is a blanking control circuit unit, 36 is a buffer memory and control circuit, 37
Is a control computer, 38 is a data conversion computer, and 39 is CA.
1 shows a D system.

The electron beam emitted from the electron gun 21 is turned on and off by the blanking deflector 23. This device makes it possible to change the irradiation amount according to the irradiation position by adjusting the irradiation time at this time.
The beam that has passed through the blanking plate 27 a is deflected by the beam forming deflector 24 and the beam forming aperture mask 27.
The beam is shaped into a rectangular beam by b and 27c, and the size of the rectangle is variable. Then, the shaped beam is deflected and scanned on the target 11 by the scanning deflectors 25 and 26, and the target 11 is drawn in a desired pattern by this beam scanning. The standard acceleration voltage of the electron beam in this device is 50 kV, and the maximum size of the variable shaped beam that can be generated is 2 μm in height,
It is a rectangle with a width of 2 μm.

The current density distribution of the electron beam of the drawing apparatus used in this example is shown in FIG. 4 (a). This is shown in Figure 4.
It is the current density distribution of the cross section when the beam is shaped so that the first aperture image is projected onto the second aperture as shown in (b) to form a rectangle of 0.25 μm × 0.25 μm on the sample surface. Reference numeral 7 in FIG. 4A indicates the half-value width of the energy distribution, and its size is 0.25 μm. The divergence at the beam edge is peculiar to the electron optical system of the drawing apparatus used in the examples, and the divergence is such that the current density is 10% and 90%.
%, It reaches to 0.1 μm.

Further, FIG. 5 shows a case where the electron scattering distribution in this embodiment is approximated by two Gaussian functions of forward scattering and backward scattering respectively. If the spread of scattering is the distance at which the intensity is 1 / e, the spread of forward scattering is about 0.
At 1 μm, the backscatter spread is ˜7 μm.

Next, an electron beam drawing method using the above apparatus, particularly a proximity effect correcting method will be described with reference to FIGS.

First, the accelerating voltage of the apparatus was standard 40 kV, and the current density was 30 A / cm ² . Also, a resist SAL601 coated on a Si substrate with a film thickness of 0.5 μm
(Manufactured by Shipley Co., Ltd.) was used as a test pattern, and a line and space pattern of 0.25 μm was used at several different pattern densities. The process conditions used in this example were pre-baking at 125 ° C. for 1 minute, post-exposure baking at 115 ° C. for 2 minutes.
Development is carried out for 2 minutes with a 0.36N alkaline developer.

The optimum irradiation amount or optimum irradiation time at each irradiation position is determined as shown in FIG. First, prior to the correction calculation, a reference table in which the optimum dose corresponding to the contribution of backscattering is described in advance is created.

Specifically, the resist is spin-coated on a Si substrate with a film thickness of 0.5 μm, and a plurality of 0.25 μm line & space patterns from isolated removal to remaining isolated are drawn (S1). Then, the optimum dose of each pattern was experimentally obtained (S2). Further, the obtained data points were fitted to create an empirical formula (interpolation formula) of the optimum dose corresponding to the pattern density, that is, the contribution of backscattering (S3). FIG. 6 shows the relationship of the optimum dose corresponding to the contribution of backscattering obtained in this example.

The reference table was created using the above empirical formula. It is not necessary to create this reference table each time correction calculation is performed. It suffices to create a device corresponding to the accelerating voltage of the device, the substrate type, and the resist process.

Drawing is performed as follows. First, input data DATA-1 having the contribution U of backscattered electrons at each position of the drawing pattern as a parameter is created using a computer (P1). Next, a reference table TABL for inputting and using the above data in the EB device
Select E-1. The EB device determines the irradiation time t corresponding to U for each irradiation position from the table (P
2). Then, a pattern is drawn with the determined irradiation amount t.

The control up to the determination of the optimum dose in the EB apparatus is processed in the part surrounded by the dotted line in FIG. FIG. 7 shows a specific structure of this portion.

First, DATA-1 is sent from the control computer 37 to the buffer memory M1, the selected reference table TABLE-1 is stored in the buffer memory M2, and the circuit A1 is started.

In this state, the reference irradiation amount is 11.0 μc / c
Drawing is performed as m ² . At this time, the circuit A1 performs the following work. (1) Read the contribution U of backscattered electrons described for each drawing position in the drawing data DATA-1. (2) The optimum irradiation time t corresponding to the contribution U of backscattered electrons is read from the buffer memory M2. (3) The optimum irradiation time t is added to the drawing data instead of the contribution U of backscattered electrons, and the data is stored in the buffer memory M3.

In the present embodiment, the irradiation dose of each pattern was 11.0 μc / cm ^{2 for the} isolated remaining pattern, 16.0 μc / cm ^{2 for} the line & space, and ^{2 for} the isolated removal pattern.
It was 7.1 μc / cm ² .

When the drawing is performed as described above, FIG.
The patterns shown in (a) to (c) were obtained. Figure 9 (a)
For comparison, SEM photographs when the proximity effect correction is performed by the conventional method are shown in (c). About the result of dimension measurement,
A comparison with the conventional method is shown in FIG. FIG. 10A shows the dimensional accuracy of the 0.25 μm pattern in this embodiment.
0 (b) shows the dimensional accuracy by the correction of the conventional method. In this embodiment, the pattern dimension could be controlled with an accuracy of ± 0.02 μm. (Embodiment 2) In the first embodiment described above, the reference table was obtained experimentally, but instead of this, the reference table may be created using an approximate expression. Even in this case, since it is not necessary to perform the calculation for calculating the irradiation amount for each drawing position, the time for determining the optimum irradiation amount can be significantly shortened. In this example, the following equation was used.

D = D / (1 + 2ηU) In this equation, the energy is 1 in the energy distribution due to the contribution of both the current density distribution peculiar to the electron beam drawing apparatus and the forward scattering energy distribution when the electron beam enters the sample. The optimum irradiation amount is given so that the total absorbed energy amount including the backscattered energy distribution at the drawing position of / 2, that is, the half width is constant regardless of the density of the pattern. In other words, it can be said that this is an approximate expression in which the energy absorption amount at the pattern edge is constant regardless of the density of the pattern.

In the drawing experiment using the above approximate expression, the drawing data to be inputted to the EB drawing device can be created at a higher speed than the conventional method, and further, the same correction effect as that of the first embodiment can be obtained. It was

The present invention is not limited to the above embodiment. Although the variable shaped beam is used in the embodiment, it is also applicable to a drawing apparatus of a method other than this, for example, an electron beam drawing apparatus of a character projection method. Further, the present invention does not limit the purpose of use of the electron beam drawing apparatus. For example, it can be used not only for the purpose of directly forming a resist pattern on a wafer but also for producing an X-ray mask, an optical stepper mask, a reticle, and the like. In addition, various modifications can be made without departing from the scope of the present invention.

[0041]

As described above in detail, according to the present invention, when the optimum dose is determined in the proximity effect correction of the dose correction method, the optimum dose or irradiation time corresponding to the contribution of backscattered electrons is determined in advance. A reference table is provided, the contribution of backscattered electrons of each pattern is calculated, and the optimum irradiation amount or irradiation time according to the contribution is determined by referring to the table, which causes the proximity effect. It is possible to prevent not only the decrease in pattern size accuracy but also the decrease in pattern size accuracy due to the blur of the incident electron beam and the forward scattering of electrons in the resist, and it is a fundamental problem in the dose correction of the unexposed portion. It is possible to prevent dimensional accuracy deterioration during development. Further, since it is not necessary to perform the correction calculation for each irradiation position, it is possible to shorten the time required for determining the irradiation amount.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram showing an electron beam drawing apparatus used in a method according to a first embodiment of the present invention.

FIG. 2 is a diagram for explaining the absorbed energy distribution in the conventional method and a decrease in pattern dimension accuracy due to the distribution.

FIG. 3 is a diagram for explaining a difference in optimal development time due to pattern density and a decrease in pattern dimension accuracy due to the difference.

FIG. 4 is a diagram showing a current distribution in the electron beam writing apparatus used in the first embodiment.

FIG. 5 is a diagram for explaining an absorbed energy distribution in electron beam exposure.

FIG. 6 is a diagram showing a relation of an interpolation formula used when creating a reference table in the embodiment.

FIG. 7 is a block diagram specifically showing a main part of FIG.

FIG. 8 is a photograph showing a fine pattern formed on a substrate according to an example.

FIG. 9 is an SEM photograph of a line-and-space pattern obtained by correction using a conventional method, in which a 0.25 μm size is isolated and left alone.

FIG. 10 is a diagram showing correction accuracy according to the present invention.

FIG. 11 is a diagram showing a method for determining an optimum irradiation amount or an optimum irradiation time at each irradiation position.

[Explanation of symbols]

 1 ... Energy distribution due to contribution of forward-scattered electrons 2 ... Energy distribution due to contribution of back-scattered electrons 3 ... Same energy absorption amount 4 ... Central portion 5 ... Edge 6 ... Half width at end 7 ... Half width of current density distribution 10 Specimen chamber 11 Target (specimen) 12 Specimen stage 20 Electron optical column 21 Electron gun 22a-22e Various lens systems 23 26 26 Various deflection systems 27a Blanking plates 27b, 27c Beam forming apertures Mask 31 ... Sample stage drive circuit section 32 ... Laser length measurement system 33 ... Deflection control circuit section 34 ... Variable shaping beam size control circuit section 35 ... Blanking control circuit section 36 ... Buffer memory and control circuit 37 ... Control computer 38 ... Data Conversion computer 39 ... CAD system

Claims (2)

[Claims] 1. An electron for irradiating an electron beam on a sample to draw a desired pattern, obtaining an optimum dose for each position in the pattern to be drawn, and drawing each pattern with this optimum dose. In the beam drawing method, a reference table for the optimum irradiation amount or irradiation time corresponding to the contribution of backscattered electrons is provided in advance, and when determining the optimum irradiation amount for each drawing position in the pattern,
An electron beam drawing method characterized in that only the contribution of backscattered electrons is calculated, and the optimum dose or irradiation time according to the contribution of backscattered electrons obtained by the calculation is determined by referring to the table. 2. The electron beam drawing according to claim 1, wherein an approximation formula or an interpolation formula obtained from optimum doses of several types of lines and spaces having different pattern densities is used in creating the reference table. Method.