JPS5932128A - Correcting method for beam emitting position in charged beam exposure apparatus - Google Patents

Abstract

PURPOSE:To obtain the drawing of a highly accurate pattern by employing a memory which stores Coulomb's effect correcting amount on the basis of some beam areas, obtaining a Coulomb's effect correcting amount corresponding to the actual area data, and obtaining the correction coefficient of the beam emitting position in response to the obtained correcting amount, thereby calculating the correction of the position. CONSTITUTION:The correcting amount of the deflecting distortion of an electro-optical system is varied in response to the exciting conditions, and the drawing of a pattern is enhanced in accuracy irrespective of the correcting operation of Coulomb's effect. For this purposed, an apparatus is composed of a pattern data 1 for determining the beam emitting position and the beam area, a block 2 for calculating the correction of a main deflection, a memory 3 for storing the coefficient for correcting the sub deflection for each deflection data, and a block 5 for calculating the correction of the sub deflection. Further, an electro-optical system 6, a main deflector 7, an electrostatic lens 8 for correcting the Coulomb's effect, a sub deflector 9, a height detector 10, a wafer 11 of a sample, and a memory 12 for storing the Coulomb's effect correcting amount such as a beam area data are provided, thereby accurately correcting the displacement of the emission of the beam.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is used for manufacturing semiconductor devices such as LSIs. This article relates to a beam irradiation position correction method that allows highly accurate beam irradiation even when a large beam current is used in a charged beam exposure apparatus that draws a pattern using a variable shaped beam. "Improved productivity" was required. For this purpose, it is necessary to improve the amount of beam current. However, when the amount of beam current is increased, the beam becomes blurred due to the repulsion of the Coulomb forces between the charged beams (hereinafter referred to as the Coulomb effect). The excitation conditions for the system were set. In particular, in an exposure apparatus using a variable shaped beam, the amount of beam current changes each time the beam is irradiated, so an excitation condition setting operation (hereinafter referred to as Coulomb effect correction) is required for each irradiation. However, since the beam irradiation position changes when the excitation conditions are changed, such an apparatus has the disadvantage that accurate pattern drawing is impossible.

In order to eliminate this drawback, the present invention changes the correction amount of the deflection distortion of the electron optical system according to the excitation conditions, and the purpose is to achieve highly accurate pattern writing regardless of the Coulomb effect correction operation. It is something that will be realized.

The features of the present invention are to achieve the above object.

A process of extracting the Coulomb effect correction value C corresponding to the actual beam area data from a memory in which Coulomb effect correction amounts are stored in advance based on some beam area data, and beam irradiation according to the extracted correction amount C. A method for correcting a beam irradiation position includes a process of obtaining a correction coefficient for a position correction calculation, and a process of performing a correction calculation of a beam irradiation position using the correction coefficient.

Nine embodiments of the present invention will be described below with reference to FIGS. 1 and 2. In Figure 1, 1 is the turn data that determines the beam irradiation position, beam area, etc.

2 is a block that executes the main deflection correction calculation (the main deflection can largely deflect the beam, but the operation is slow), and B is the block that stores in advance the coefficients αi, j, and β for sub-deflection correction for each main deflection data. 4 is the secondary deflection distortion A block for calculating coefficients, and a block 5 for performing sub-deflection correction calculations.

6 is an electron optical system, 7 is a main deflector, 8 is an electrostatic lens for correcting the Coulomb effect, 9 is a sub deflector, 10 is a height detector,
Reference numeral 11 denotes a memory in which Coulomb effect correction 11iC is recorded for each sample h (sample) and 12ζ beam area data.

FIG. 2 is a diagram for explaining how to obtain various correction coefficients in the present invention, (a) is an overall view, 15 is the main deflection area, 16 is a representative grid point, and (b) is a certain representative grid point. This is an enlarged view of the lattice points, where 17 is a mark and 18 is a sub-deflection area.

First, assuming that the necessary correction coefficients have been obtained, the correction operation for the beam irradiation position will be described. In pattern drawing, first the main deflection data X, Y (X and Y coordinate values in the orthogonal coordinate system set on the irradiation surface) are sent.

A correction calculation for correcting the main deflection is performed in block 2 on the main deflection data X, Y, and main deflection correction is performed in the main deflector 7 based on the calculation result.

Block 6 uses the main deflection data X and Y as addresses, selects and takes out pre-stored coefficients α-cutting and β-field for sub-deflection correction, and sends them to block 4.

Next, from pattern data 1, beam area data W and sub-deflection data x, y are sent out for each beam irradiation operation. The memory 12 outputs the Coulomb effect correction amount C stored in advance using the beam area data W as an address. This correction amount C is sent to the electrostatic lens 8 to correct the Coulomb effect.

Block 4 includes correction coefficients α and β1 . j and the Coulomb effect correction amount C output from the memory 12, the coefficients a□ and bi necessary for the sub-deflection correction calculation executed in block 5 are calculated by the following formula.

However, i20. t 2.3. Block 5 is 2
al + bj obtained by equation (1) and sub-deflection data x
, y, the sub-deflection correction calculation is performed according to the following equation.

X'l Y' obtained by equation (2) is sent to the sub-deflector 9.

Pattern writing is performed by deflecting the beam by a predetermined amount.

Since the correction operation is as described above, highly accurate pattern drawing can be performed as long as the following conditions are satisfied.

(a) All the coefficients necessary for the correction calculation described in - have been obtained.

(b) By correction of Coulomb effect. The change in the beam position can be sufficiently corrected by setting the sub-deflection correction coefficient using the linear expression of the correction amount C.

Here, how to obtain the coefficients for the second condition (a) will be described later. Regarding condition (b), the change in beam position due to correction of the Coulomb effect is small, about 1 μm or less.

It is sufficient to provide a correction amount for the position change using a linear expression of the correction amount C for each area to be drawn using the sub-deflection. The above formula (1
) means that a more precise correction amount is given.

Below, we will explain how to find the coefficients for the two conditions (a).

1) The correction amount C is obtained based on some beam area data W so that the beam blur is minimized.

Write to the memory at the address corresponding to the beam area data W. 12 in FIG. 1 is this memory.

2) Obtain the correction coefficients ai and bl for the sub-deflection distortion at all the representative grid points 16 in FIG. 2 by the following method. That is,
By positioning the beam at the grid point 16 with the main deflection, the sub-deflection region 18 is set. Next, the position of the mark 17 is detected within this sub-deflection area 18 using two mark detectors (not shown). The position of the mark 17 is accurately measured in advance, and the sub-deflection distortion correction coefficient "l+b1" can be obtained by a well-known method described below.

In this embodiment, the number of representative grid points 16 is 25.

As a result, we obtain the following data. ``Similarly, the above data is obtained under two conditions each of the minimum correction amount Cm1n and the maximum correction amount Cmax.

3) Using the above data, express "I+bl" with a cubic equation of the main deflection data X and Y. That is.

Under the conditions of Cm1n Under the condition of Cmax Find α1ljlk and β'+J+k.

The method for determining αi+Lk and β'+J+ in equation (3) is shown below.

If we substitute the data obtained in section 2) above into equation (3).

We get the formula below. Here, () indicates a vector or a matrix.

However, the subscript t indicates the transposed matrix, and (ai) = (ai, t"°゛, a1, 25) t(
βl ) = (βi, +, o βsu, 1 °゛°βi
, t9)'.

Applying the least squares method to Equation (5), [α] and (β1) are obtained. That is.

(α1. ) = ((X)'(X7')・(>'l
'(aI)(βi+=((Xl'・(Xt1)・
Perform the calculation (X)1・(bi).

4) Set data in the sub-deflection correction coefficient memory 3. That is, by substituting the coefficients α and β obtained in the above 3) and the main deflection data X and Y corresponding to the address of the memory 5 into equation (3),
By substituting αi,1 and β1,1 into equation (4) again, αi,
2 and βl,2 are calculated and written into the memory 3.

From the above, all the coefficients necessary for the above-mentioned correction calculation were obtained.

Therefore, accurate pattern writing can be performed while correcting the Coulomb effect every time the beam is irradiated.

Now, in order to draw a pattern with high precision on a wafer with 2-curve or the like, it is necessary to measure the deformation of the wafer in the height direction and change the deflection strain coefficient based on this measured value. To deal with this, increase the storage capacity of memory 3 and α1,3
and βi, 3, and then use the output of the height detector 10 in the same way as the Coulomb effect correction amount C to perform the same correction calculation as described above. Regarding the calculation of αi3 and β13, we used two types of marks at different heights.

It can be determined by the method described above, and in this case, most of the calculation programs for coefficient calculation can be shared.

Increase the sub-deflection correction coefficient a Ir b 1 and use equation 2 (2)
It is also possible to use a quadratic expression containing X and Y. however.

In this case, the number of marks in the sub-deflection area must be at least six.

As explained above, according to the present invention, it is possible to accurately correct the deviation of the beam irradiation position accompanying the Ron effect correction operation, which is indispensable when using a large current, resulting in high productivity and high productivity. This has the effect of making it possible to draw patterns with high precision.

[Brief explanation of drawings]

FIG. 1 is a detailed explanatory diagram of the present invention, and FIG. 2 is a diagram explaining how to obtain various correction coefficients in the present invention. Description of symbols 1...Pattern data 2...Main deflection correction calculation block 3...Sub deflection correction coefficient memory 4...Sub deflection distortion coefficient calculation block 5...Sub deflection correction calculation block 6...・Electron optical system 7... Main deflector 8...
Electrostatic lens 9...Sub deflector 10...Height detector 11...Wafer (sample) 12...Coulomb effect correction amount memory 15...Main deflection area 16...Representative grid point 17
...Mark 18... Sub-deflection area patent applicant Junnosuke Nakamura, patent attorney representing Nippon Telegraph and Telephone Public Corporation

Claims (1)

[Claims] (+) Pattern drawing using a charged beam is provided with a Coulomb effect correction mechanism that corrects the Coulomb effect, which causes blurring of the beam due to Coulomb repulsion between the charged beams, by changing the excitation conditions of the electron optical system. In a charged beam exposure apparatus that performs a process, there is a process of extracting a Coulomb effect correction amount C corresponding to actual beam area data from a memory in which the Coulomb effect correction amount is stored in advance based on some beam area data, and a process of extracting the Coulomb effect correction amount C corresponding to actual beam area data. A charged beam exposure apparatus comprising: a step of determining a correction coefficient for correcting a beam irradiation position according to a correction amount C; and a step of performing a correction calculation of a beam irradiation position using the correction coefficient. Beam irradiation position correction method. (2) The process of extracting the Coulomb effect correction amount C is performed using at least two correction amounts, a correction amount Cm1n that corrects the Coulomb effect to a minimum and a correction amount Cmax that corrects the Coulomb effect to a maximum, based on the actual beam area data 1. Cm1n,
The correction of the 1mI beam irradiation position is made by the process without taking Cmax, and the correction of the 1mI beam irradiation position is performed using the above-mentioned at least two correction amounts Cm1n, Cm
2. A beam irradiation position correction method in a charged beam exposure apparatus according to claim 1, wherein correction coefficients for ax are obtained and the correction calculation is performed using these coefficients.