JP2005108981A - Exposure method and exposure apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an exposure method and an exposure apparatus which can compensate with higher accuracy, for non-linear distortions generated in reticle during the exposure. <P>SOLUTION: Locations of all sub-fields and locations of marks for detecting locations formed in the circumference of the sub-field region are measured with a coordinate measuring apparatus (S11), by defining the center of the sub-field region of reticle as the reference location. Locations of the same marks are measured by loading the reticle to the exposure apparatus (S12). Thereafter, displacement data of each mark in the exposure apparatus are calculated from the location data of mark measured in the steps S11 and S12 (S13). Linear deformation of reticle is estimated with the method of least squares (S14). Non-linear displacement of each mark is calculated (S15). Linear displacement of each sub-field is estimated (S16). Non-linear displacement of each sub-field is obtained from a length R<SB>si</SB>of the half-line, extending from the reference location to the original location of the sub-field and the reference location, as a function of the ratio of the line connecting the adjacent marks and the length R<SB>m</SB>up to the intersecting point of this half-line (S17). Exposure is conducted by using the location change data of each sub-field (S18), while correcting compensating for the location of sub-field. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an exposure method and an exposure apparatus used for lithography such as a semiconductor integrated circuit. In particular, the present invention relates to an exposure method and an exposure apparatus that can correct reticle distortion with high accuracy.

  In recent years, with the development of batch transfer systems, reticles have also been used for charged particle beam lithography. Since these reticles are often formed of a thin film such as silicon, the reticle is distorted due to the stress of the film and the weight of the reticle. However, since the reduction ratio at the time of pattern transfer has been large in the past, reticle distortion as described above has not been a serious problem in semiconductor lithography. The cause of reticle distortion is not only the reticle's own weight and film stress, but also the distortion caused by the holding force of the chuck and reticle holder when it is fixed in the exposure apparatus, and the thermal expansion of the reticle due to the temperature inside the apparatus. Distortion caused by the above also occurs.

  As a method for inspecting / correcting reticle distortion as described above, an inspection apparatus such as a coordinate measuring apparatus different from the exposure apparatus is provided by providing a position measurement mark on the beam portion of the reticle (see reference numeral 45 in FIG. 5). In this method, the mark position is measured to obtain position data, and the data is used to correct the pattern transfer position during exposure. Alternatively, there is a method of measuring and correcting the mark position in the exposure apparatus. Japanese Patent Application Laid-Open No. 2004-133620 proposes a method of measuring linear distortion in an exposure apparatus, measuring nonlinear distortion with an inspection apparatus, and superimposing both to estimate reticle position data in the exposure apparatus.

JP 2002-319533 A

In the reticle distortion correction method described above, for non-linear distortion caused by chuck holding force etc., the reticle distortion is measured by measuring all points every time exposure is performed, and the distortion is corrected during exposure. It was a method of exposing while. However, it is impossible to measure all points of the reticle during exposure, and there is a problem that the exposure accuracy is deteriorated when the nonlinear distortion is large.
In view of the above points, an object of the present invention is to provide an exposure method and an exposure apparatus that can correct nonlinear distortion generated in a reticle during exposure with high accuracy.

In order to solve the above-described problems, an exposure method of the present invention divides a device pattern to be transferred onto a sensitive substrate into small regions (subfields), and forms the reticle on each of the subfields. By illuminating with energy rays, projecting and imaging energy rays that have passed or reflected through the sub-fields onto appropriate positions on the sensitive substrate, and connecting the pattern images of the sub-fields on the sensitive substrate. An exposure method for transferring the entire device pattern, wherein A): a position detection mark is formed around a region (subfield region) where the subfield group exists on the reticle, and B): reference position in the center of the sub-field range _(X 0, _{Y 0)} Place, C): the position of all of the subfields in the coordinate measuring device _{(X si, Y} si) Position of the fine the mark _{(X mi, Y} mi) was measured, D): the position of the mark after mounting the reticle to the exposure apparatus _{(X mi ', Y} mi' measured), E): C) , D) to obtain position deviation data (ΔX _mi = X _mi ′ −X _mi , ΔY _mi = Y _mi ′ −Y _mi ) in the exposure apparatus from the position data of the mark measured in (D), F): the position From the deviation data, the linear deformation (magnification difference S, orthogonality difference ω, rotation difference θ, translation difference O) with respect to the deformation accompanying the mounting of the reticle on the exposure apparatus is estimated, and G): position of each mark mounting deviation (ΔX _{mi, ΔY mi)} from linear position displacement ([Delta] X _mi of the respective marks due to linear deformation of the reticle estimated by F) ^L, pull the [Delta] Y _mi ^L), to the exposure apparatus of each mark nonlinear position shift of accompanying ([Delta] X _mi ^L, to obtain a ^{_{ΔY mi NL), H) :(}} X si, Y si) from the positional deviation due to the mounting of the exposure apparatus of each sub-field due to linear deformation of the reticle ^{_{(ΔX si L, ΔY si L}} ) I): Non-linear positional deviation (ΔX _si ^NL , ΔY _si ^NL ) of each subfield after the exposure apparatus is mounted from (X _si , Y _si ) and (ΔX _si ^L , ΔY _si ^L ) A nonlinear positional deviation before being mounted on the exposure apparatus is obtained, and further, I1): a line drawn from the reference position (X ₀ , Y ₀ ) to the original position (X _si , Y _si ) of each subfield is The position detection marks around the field area are extended until they intersect with a line connecting adjacent ones of the position detection marks, and the non-linear position deviation of the intersection of I2): I1) is determined from the non-linear position deviation of the marks adjacent to the intersection. Interpolate, 3): reference position of the line I1) _(X 0, _{Y 0)} from the original position of the sub-fields _{(X si,} the length _{R si} to _{Y si),} from the reference position _(X 0, _{Y 0)} as a function of the ratio of the length _{R m} to the intersection, estimating a change of the nonlinear positional deviation of each sub-field after exposure apparatus mounting ^{_{(ΔX si NL, ΔY si NL}} ), J): I) obtained in Further, exposure is performed while correcting the position of the subfield using the position change data of each subfield.

In the above I3), the non-linear position deviation dx = (ΔX _Rm ^NL , ΔY _Rm ^NL ) at the intersection obtained by interpolation from the non-linear position deviations of the marks adjacent to the intersection and r = R _m Substituting dx = r ^N A (N is an arbitrary number) to obtain a matrix A = (A _x , A _y ), substituting the matrix A and r = R _si into the formula dx = r ^N A, It is possible to estimate the non-linear positional deviation dx = (ΔX _si ^NL , ΔY _si ^NL ) of the subfield.

The exposure apparatus of the present invention divides a device pattern to be transferred onto a sensitive substrate into small regions (subfields) and forms them on a reticle, and illuminates the reticle with energy rays for each of the subfields. Exposure to transfer the entire device pattern by projecting and forming an energy beam that has passed or reflected through the field onto an appropriate position on the sensitive substrate, and joining the images of the patterns of the subfields on the sensitive substrate. A position detection mark formed around a region where the group of subfields exists (subfield region), means for detecting the position of the position detection mark, and coordinates position _{(X si, Y} si) of all of the subfields was measured by the measuring device and the position _{(X mi, Y} mi) of the marked exposure apparatus Position of the mark measured after mounting the reticle _{(X mi ', Y mi'} ) because obtained by calculating, linear positional deviation of the marks ^{_{(ΔX mi L, ΔY mi L}} ), positional deviation of the marks Using the non-linear positional deviation (ΔX _mi ^NL , ΔY _mi ^NL ) and the linear positional deviation (ΔX _si ^L , ΔY _si ^L ) of each subfield from the reference position (X ₀ , Y ₀ ) The non-linear positional deviation of the intersection between the line drawn to the position (X _si , Y _si ) and the line extended until it intersects with a line connecting adjacent ones of the position detection marks around the subfield area, The length R _si from the reference position (X ₀ , Y ₀ ) of the line to the original position (X _si , Y _si ) of the line is interpolated from the non-linear positional deviation of the marks on both sides of the intersection , Standard position As a function of the ratio of the length _{R m} of _(X 0, _{Y 0)} from to the intersection, nonlinear positional deviation ^{_{(ΔX si NL, ΔY si NL}} ) of each sub-field and means for estimating a, to have a Features.

  As described above, according to the present invention, in the exposure apparatus, it is possible to predict the displacement of all the subfields only by measuring the positions of the position detection marks formed around the subfield area. it can. Therefore, high-precision exposure position control can be realized by position measurement of a small number of points without performing all-point measurement as in the prior art, and high-throughput exposure can be performed.

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
First, an outline of an electron beam exposure method which is one of exposure methods for semiconductor devices and the like will be described.
FIG. 4 is a schematic diagram showing an imaging relationship in the entire optical system of the divided transfer type electron beam exposure apparatus.
An electron gun 1 is disposed in the uppermost stream of the optical system (upward in the figure). The electron gun 1 irradiates an electron beam downward in the figure. Below the electron gun 1, two-stage condenser lenses 2 and 3 are provided. The electron beam is converged by these condenser lenses 2 and 3, and crossover C.D. O. Is imaged.

  An illumination beam shaping aperture 4 is disposed below the condenser lens 3. The image of the opening 4 is formed on the reticle 10 by the lens 9.

  A blanking deflector 5 is disposed below the illumination beam shaping opening 4. The deflector 5 deflects the illumination beam as necessary and hits the non-opening portion of the blanking opening 7 so that the beam does not hit the reticle 10.

An illumination beam deflector 8 is disposed below the blanking opening 7. The deflector 8 sequentially scans the illumination beam in the X-axis direction to illuminate each subfield of the reticle 10 in the field of view of the optical system. At this time, the illumination beam is scanned while performing position correction based on the position shift data calculated in advance (the procedure for calculating position shift data will be described later).
An illumination lens 9 is disposed below the deflector 8. The illumination lens 9 images the illumination beam shaping aperture 4 on the reticle 10.

The reticle 10 extends in a plane (XY plane) perpendicular to the optical axis, and has a number of subfields (see FIG. 5 and will be described in detail later). On the reticle 10, a device pattern constituting one semiconductor chip as a whole is formed. Note that a device pattern constituting one chip may be divided and formed on a plurality of reticles.
The reticle 10 is installed on a movable reticle stage 11. By moving the reticle stage 11 in the XY direction, each subfield on the reticle 10 that extends over a wider range than the field of view of the illumination optical system is displayed. Can be illuminated. A position detector (interferometer) 12 is attached to the reticle stage 11 so that the position of the reticle stage 11 can be accurately grasped.

  Projection lenses 15 and 19 and a deflector 16 are provided below the reticle 10. The electron beam that has passed through one subfield of the reticle 10 is imaged at a desired position on the wafer 23 by the projection lenses 15 and 19 and the deflector 16.

  An appropriate resist is applied on the wafer 23. An electron beam dose is given to the resist, and the device pattern on the reticle 10 is reduced and transferred.

  At a position where the reticle 10 and the wafer 23 are internally divided by a reduction ratio, a crossover C.I. O. The contrast opening 18 is provided at this position. This opening 18 blocks the electron beam scattered by the non-pattern part of the reticle 10 from reaching the wafer 23. At this time, most of the scattered electron beams can be blocked, but a very small amount of scattered electron beams reach the wafer 23 without being blocked.

  The wafer 23 is disposed on a wafer stage 24 that can move in the X-axis and Y-axis directions via an electrostatic chuck (not shown). By moving the reticle stage 11 and the wafer stage 24 synchronously in opposite directions, a device pattern that extends beyond the field of view of the optical system can be sequentially exposed and transferred. The wafer stage 24 also includes a position detector 25 as in the reticle stage 11 described above.

  A backscattered electron detector 22 is disposed immediately above the wafer 23. The backscattered electron detector 22 detects the amount of electrons reflected by the exposure surface of the wafer 23 and the mark on the wafer stage 24. For example, the relative positional relationship between the reticle 10 and the wafer 23 can be known by scanning the mark on the wafer 23 with a beam that has passed through the mark pattern on the reticle 10 and detecting electrons reflected by the mark.

  The lenses 2, 3, 9, 15, 19 and the deflectors 5, 8, 16 are respectively connected to the corresponding coil power supply control units 2a, 3a, 9a, 15a, 19a and 5a, 8a, 16a. Controlled by the controller 31. The reticle stage 11 and the wafer stage 24 are also controlled by the controller 31 via corresponding stage control units 11a and 24a.

  The stage position detectors 12 and 25 send signals to the controller 31 via the interfaces 12a and 25a including amplifiers and A / D converters. Similarly, the backscattered electron detector 22 sends a signal to the controller 31 via the interface 22a.

  The controller 31 grasps the alignment error of the stage and corrects the error by the image position adjusting deflector 16. Thereby, the reduced image of the subfield on the reticle 10 is accurately transferred to the target position on the wafer 23. Then, the images of the subfields are joined on the wafer 23, and the entire device pattern on the reticle 10 is transferred onto the wafer.

Next, a specific example of the reticle 10 will be described.
FIG. 5 is a diagram schematically showing an example of a reticle used in electron beam exposure. 5A is a plan view of the whole, FIG. 5B is a partial perspective view, and FIG. 5C is a plan view of one small membrane region. Such a reticle can be manufactured, for example, by drawing on a silicon wafer using an electron beam and then performing etching.

  Device patterns are divided and arranged on the reticle 10 shown in FIG. A large number of squares 41 shown in FIG. 5A are small membrane regions each including a device pattern corresponding to one subfield. For example, the small membrane region 41 has a thickness of 0.1 μm to several μm.

As shown in FIG. 5C, the small membrane region 41 includes a central subfield 42 and a skirt 43 surrounding the periphery thereof in a frame shape.
The subfield 42 is, for example, a square having a side of 0.5 to 5 mm. If the reduction ratio at the time of projection is 1/5, the image of the subfield 42 projected onto the wafer 23 is a square having one side of 0.1 to 1 mm. An alignment mark (not shown) is provided around each subfield 42.
As a form to form such a subfield pattern, there is a scattering stencil type in which a perforated portion is provided in a scattering membrane through which an electron beam is scattered and a pattern layer comprising a high scatterer that scatters an electron beam well. There is a continuous membrane type in which is formed on a very thin membrane as an electron transmission body.

  The skirt 43 is a frame-like part (with a width of 0.05 mm in one example) where no pattern is formed, and the edge of the illumination beam hits it.

  A lattice-shaped portion (support layer) around the small membrane region 41 in FIGS. The glage 45 is a beam (thickness 0.5 to 1 mm, width 0.1 mm, for example) for maintaining the mechanical strength of the reticle 10.

As shown in FIG. 5A, a large number of small membrane regions 41 are arranged in the X-axis direction in FIG. 5 to form one group (electrical stripe 44), and a large number of electrical stripes 44 in the Y-axis direction in FIG. One mechanical stripe 49 is formed side by side.
The length of the electrical stripe 44 (the width of the mechanical stripe 49) is limited by the width of the field of view that can be covered by the deflection of the optical system.
In addition, a method in which a non-pattern region such as a skirt or a grudge is not provided between adjacent subfields in one electrical stripe 44 has been studied.

  A plurality of mechanical stripes 49 are arranged in the X-axis direction of FIG. A portion between adjacent mechanical stripes 49 is a strut 47. The strut 47 is a beam that is slightly thicker than the grage 45 and reduces the deflection of the reticle 10. The strut 47 and the grage 45 are integrated. In the reticle 10 shown in FIG. 5, two mechanical stripes 49 are formed, and a position detection mark region (may be a subfield) in which position detection marks are formed so as to surround the two mechanical stripes 49. 46 is provided (see FIG. 3).

  In the exposure of the reticle 10, a column of small membrane regions 41 (hereinafter referred to as subfields 41) of the electrical stripe 44 is sequentially exposed by deflecting an electron beam in the X-axis direction of FIG. 5. On the other hand, exposure is sequentially performed in the Y-axis direction of FIG. 5 by continuous stage scanning.

Here, the types of distortion generated in the reticle will be described.
FIG. 6 is a diagram showing the types of distortion (error) generated in the reticle. 6A is a diagram showing a rotation error, FIG. 6B is a diagram showing an orthogonality error, and FIG. 6C is a diagram showing a magnification error.
In the figure, the positions of the pattern region 40 and the alignment mark 53 before the distortion occurs are indicated by broken lines.

In FIG. 6A, the positions of the pattern region 40 ′ and the alignment mark 53 ′ after the occurrence of the rotational error distortion are indicated by solid lines. FIG. 6A shows a state in which the pattern region 40 ′ is rotated clockwise by the angle θ from the original position 40 around the center point. Assuming that the coordinates of an arbitrary point on the original pattern area 40 are (x, y), the coordinates of the corresponding point (X, Y) on the pattern area 40 ′ after the occurrence of rotational error distortion are
(X, Y) = (x · cos θ−y · sin θ, x · sin θ + y · cos θ) (a)
It becomes.

In FIG. 6B, the positions of the pattern region 40 ′ and the alignment mark 53 ′ after the occurrence of the orthogonality error distortion are indicated by solid lines. FIG. 6B shows a state in which the orthogonality of the original pattern region 40 is distorted by an angle ω and the pattern region 40 ′ has a parallelogram shape. If the coordinates of an arbitrary point on the original pattern area 40 are (x, y), the coordinates of the corresponding point (X, Y) on the pattern area 40 ′ after the occurrence of the orthogonality error distortion is
(X, Y) = (xy · tan ω, y) (b)
It becomes.

In FIG. 6C, the positions of the pattern region 40 ′ and the alignment mark 53 ′ after the magnification error distortion are shown by solid lines. FIG. 6C shows a state where the original pattern region 40 is enlarged due to thermal expansion or the like. When the magnification error in the X direction is S _x and the magnification error in the Y direction is S _y , the size of the pattern region 40 ′ after the magnification error distortion is (1 + S _x ) × (1 + S _y ).

In the exposure apparatus, as described above, since the subfield 41 is supported by the grunge 45, distortion generated in the subfield 41 is reduced.
However, the reticle 10 as a whole has a very large distortion of about several tens of nanometers. In order to correct such distortion, only correction of rotation and magnification is insufficient to realize required exposure accuracy. Further, it is impossible to perform all-point position measurement as is done in the inspection apparatus in order to correct distortion in the exposure apparatus.
In the present invention, subfield arrangement correction due to distortion in the entire reticle (mask) is considered. Therefore, a position detection mark region 46 is formed around the subfield region (two mechanical stripes 49) where the exposure pattern is formed, and the electron beam is used to form the position detection mark region 46 in the exposure apparatus. By detecting the positions of the position detection marks, deviations from the design values of all subfield positions are obtained.

レチクルの線型変形に起因する各サブフィールドの位置ずれ（ΔＸ_ｓｉ ^Ｌ，ΔＹ_ｓｉ ^Ｌ）を、式（１）を用いて推定する（Ｓ１６）。 Below, the exposure method of this invention is demonstrated.
FIG. 1 is a flowchart showing an exposure method according to an embodiment of the present invention.
First, the reference position (X ₀ , Y ₀ ) is placed at the center of the subfield area of the reticle. Then, the position (X _si , Y _si ) of all the subfields and the position (X _mi , Y _mi ) of the position detection mark (see FIGS. 3 and 5) are measured by the coordinate measuring device (S11). Here, distortion data at the time of coordinate measurement is obtained.
Next, after the reticle is mounted on the exposure apparatus, the position (X _mi ′, Y _mi ′) of the position detection mark is measured (S12), and the position detection mark position data measured in S11 and S12 is transferred to the exposure apparatus. Displacement data (ΔX _mi = X _mi ′ −X _mi , ΔY _mi = Y _mi ′ −Y _mi ) associated with the mounting of the mark is calculated (S13).
Then, using equation (1), the linear deformation (magnification difference S, orthogonality difference ω, rotation difference θ, translation difference O) accompanying the mounting of the reticle on the exposure apparatus is estimated by the least square method (S14). In equation (1), matrix (A) is a rotation error distortion (see equation (a) above), matrix (B) is an orthogonality error distortion (see equation (b) above), and matrix (C) is a magnification error. It is a matrix representing distortion (see FIG. 6). Further, (O _x , O _y ) indicates a deviation (translation error) of the reference position of the subfield.
Subtracting the linear displacement (ΔX _mi ^L , ΔY _mi ^L ) of each mark resulting from the linear deformation of the reticle estimated in S14 from the positional displacement (ΔX _mi , ΔY _mi ) of each mark (formula (2)), Non-linear positional deviations (ΔX _mi ^NL , ΔY _mi ^NL ) of positional deviations associated with mounting of each mark on the exposure apparatus are calculated (S15).

The displacement (ΔX _si ^L , ΔY _si ^L ) of each subfield due to the linear deformation of the reticle is estimated using equation (1) (S16).

Here, non-linear distortion data (expected value, mark position (ΔX _mi ^PNL , ΔY _mi ^PNL ) and subfield position (ΔX _si ^PNL , ΔY _si ^PNL )) at the time of exposure, which can be predicted from the distortion data at the time of coordinate measurement, is calculated. To do. Then, the non-linear position shift (ΔX _mi ^NL , ΔY _mi ^NL ) of the position shift of each mark obtained in S15 and the non-linear distortion data (predicted value) of the mark position at the time of exposure (ΔX _mi ^PNL , ΔY _mi ^PNL ) , And if the nonlinear distortion is predicted with sufficient accuracy ((ΔX _mi ^NL , ΔY _mi ^NL ) ≈ (ΔX _mi ^PNL , ΔY _mi ^PNL )), the predicted value (ΔX _si ^PNL , ΔY _si) ^PNL ) and the linear positional deviation data estimated in S16, and exposure is performed while correcting (S18).

If the non-linear distortion cannot be predicted with sufficient accuracy, the non-linear displacement (ΔX _si ^NL , ΔY _si ^NL ) of each subfield is estimated by the following procedure (S 17).
Here, the procedure (S17) for estimating the non-linear positional deviation (ΔX _si ^NL , ΔY _si ^NL ) of each subfield will be described.
FIG. 2 is a flowchart showing a method of estimating the non-linear positional deviation of the subfield. FIG. 3 is a diagram illustrating a method of estimating the non-linear positional deviation of the subfield. 3A is a plan view of the reticle, and FIG. 3B is an enlarged plan view showing a part of the position detection mark.
In the following, the position of the point on the subfield is expressed using polar coordinate display with the reference position 100 (X ₀ , Y ₀ ) as the origin. Therefore, the position 101 (X _si , Y _si ) measured by the coordinate measuring device of each subfield is written as (R _si , θ ₀ ).
First, as shown in FIG. 3A, a half line l is drawn from the reference position 100 to the original position 101 (R _si , θ ₀ ) of each subfield. The half line l passes between the position detection marks 461 and 462 in the position detection mark area 46 around the subfield area 49. Then, an intersection 102 (R _m , θ ₀ ) between the line connecting these position detection marks 461 and 462 and the half line l is obtained (FIG. 3B). The non-linear positional deviation (ΔX _Rm ^NL , ΔY _Rm ^NL ) at the intersection 102 is estimated by interpolation from the non-linear positional deviations of the marks adjacent to the intersection 102 (S 171). Examples of this interpolation method include linear interpolation and spline interpolation.

Then, the non-linear positional deviation (ΔX _si ^NL , ΔY _si ^NL ) of each subfield is estimated (S172). Here, the non-linear positional shift in θ ₀ (on the half line 1) is the length R _si from the reference position 100 to the original position 101 (R _si , θ ₀ ) of the subfield, and the reference position 100 (X ₀ , Y ₀ ) to the intersection point 102 as a function of the ratio to the length R _m . For example, the following calculation is performed.
First, non-linear positional deviation in _{R m} determined in ^{_{S171 (ΔX Rm NL, ΔY Rm}} NL) a and r = _{R m} into Equation (3), obtaining the matrix A. Next, the matrix A and r = R _si are substituted into Equation (3) to obtain nonlinear positional deviations (ΔX _si ^NL , ΔY _si ^NL ). In the formula (3), N is an arbitrary number (practically N = 2 to 3).

Finally, position change data for each subfield:
(ΔX _si , ΔY _si ) = (ΔX _si ^L , ΔY _si ^L ) + (ΔX _si ^NL , ΔY _si ^NL )
The exposure is performed while correcting the position of the subfield using (S18).

It is a flowchart which shows the exposure method which concerns on one Embodiment of this invention. It is a flowchart which shows the method of estimating the nonlinear position shift of a subfield. It is a figure which shows the method of estimating the nonlinear position shift of a subfield. (A) It is a top view of a reticle. (B) It is a top view which expands and shows a part of position detection mark. It is a schematic diagram showing an imaging relationship in the entire optical system of a split transfer type electron beam exposure apparatus. It is a figure which shows typically an example of the reticle used in electron beam exposure. (A) It is the whole top view. (B) It is a partial perspective view. (C) It is a top view of one small membrane area | region. It is a figure which shows the kind of distortion (error) which arises in a reticle. (A) It is a figure which shows a rotation error. (B) It is a figure which shows an orthogonality error. (C) It is a figure which shows a magnification error.

Explanation of symbols

10 reticle 46 position detection mark area 461, 462 position detection mark 49 mechanical stripe 100 reticle reference position 101 subfield position 102 half-line l extending from reticle reference position 100 to subfield position 101, position detection mark 461, Intersection of line connecting 462 1 Electron gun 2, 3 Condenser lens 4 Illumination beam shaping aperture 5 Blanking deflector 7 Blanking aperture 8 Illumination beam deflector 9 Illumination lens 11 Reticle stage (chuck)
12, 25 Position detector 15, 19 Projection lens 16 Deflector 18 Contrast aperture 22 Backscattered electron detector 23 Wafer 24 Wafer stage 2a, 3a, 5a, 8a, 9a, 15a, 16a, 19a Coil power controller 31 Controller 41 Small Membrane area 42 Subfield 43 Skirt 44 Electrical stripe 45 Grage 47 Strut 50 Holding part

Claims (3)

The device pattern to be transferred onto the sensitive substrate is divided into small areas (subfields) and formed on the reticle.
Illuminating the reticle with energy rays for each subfield,
Project and image the energy rays that have passed or reflected through the subfield at the appropriate positions on the sensitive substrate,
On the sensitive substrate, an exposure method for transferring the entire device pattern by joining the pattern images of the subfields,
A): A position detection mark is formed around a region (subfield region) where the group of subfields of the reticle exists,
B): A reference position (X ₀ , Y ₀ ) is placed at the center of the subfield area,
C): Measure the position (X _si , Y _si ) of all subfields and the position of the mark (X _mi , Y _mi ) with a coordinate measuring device,
D): measuring the position (X _mi ′, Y _mi ′) of the mark after mounting the reticle on the exposure apparatus;
E): The positional deviation data (ΔX _mi = X _mi ′ −X _mi , ΔY _mi = Y _mi ′ −Y _mi ) in the exposure apparatus is obtained from the mark position data measured in C) and D).
F): From the positional deviation data, a linear deformation (magnification difference S, orthogonality difference ω, rotation difference θ, translation difference O) with respect to the deformation accompanying the mounting of the reticle on the exposure apparatus is estimated,
G): From the positional deviations (ΔX _mi , ΔY _mi ) of the marks, the linear positional deviations (ΔX _mi ^L , ΔY _mi ^L ) of the marks due to the linear deformation of the reticle estimated in F) are subtracted. Obtaining a non-linear positional deviation (ΔX _mi ^NL , ΔY _mi ^NL ) of the positional deviation accompanying mounting of the mark on the exposure apparatus;
H): From (X _si , Y _si ), estimate the displacement (ΔX _si ^L , ΔY _si ^L ) of each subfield due to the linear deformation of the reticle,
I): With respect to the non-linear positional deviation (ΔX _si ^NL , ΔY _si ^NL ) of each subfield, the nonlinearity before mounting on the exposure apparatus from each (X _si , Y _si ) and (ΔX _si ^L , ΔY _si ^L ) Find the misalignment, and
I1): A line drawn from the reference position (X ₀ , Y ₀ ) to the original position (X _si , Y _si ) of each subfield is adjacent to position detection marks around the subfield area. Extend until crossing the connecting line,
I2): Interpolate the non-linear misalignment of the intersection of I1) from the non-linear misalignment of the marks adjacent to the intersection;
I3): reference position of the line I1) _(X 0, _{Y 0)} from the original position of the sub-fields _{(X si,} the length _{R si} to _{Y si),} from the reference position _(X 0, _{Y 0)} As a function of the ratio to the length R _m to the intersection, a change in non-linear misalignment (ΔX _si ^NL , ΔY _si ^NL ) of each subfield after mounting the exposure apparatus is estimated,
J): An exposure method wherein exposure is performed while correcting the position of each subfield using the position change data of each subfield obtained in I). The non-linear position deviation dx = (ΔX _Rm ^NL , ΔY _Rm ^NL ) at the intersection obtained by interpolation from the non-linear position deviations of the marks adjacent to the intersection, and r = R _m are expressed by the formula dx = r ^N A (N Is an arbitrary number) to obtain a matrix A = (A _x , A _y ),
2. The matrix A and r = R _si are substituted into an expression dx = r ^N A to estimate the non-linear positional deviation dx = (ΔX _si ^NL , ΔY _si ^NL ) of each subfield. The exposure method as described. The device pattern to be transferred onto the sensitive substrate is divided into small areas (subfields) and formed on the reticle.
Illuminating the reticle with energy rays for each subfield,
Project and image the energy rays that have passed or reflected through the subfield at the appropriate positions on the sensitive substrate,
On the sensitive substrate, an exposure apparatus for transferring the entire device pattern by joining the image of the pattern of each subfield,
A position detection mark formed around the area where the group of subfields exists (subfield area);
Means for detecting the position of the position detection mark;
Positions of all the subfields was measured by the coordinate measuring device _{(X si, Y} si) and the position of the mark _{(X mi, Y} mi) and the position _{(X mi} of the marks was measured after mounting the reticle in the exposure apparatus ′, Y _mi ′), and the linear position shift (ΔX _mi ^L , ΔY _mi ^L ) of each mark, the non-linear position shift (ΔX _mi ^NL , ΔY _mi ^NL ) of the position shift of each mark, and A line drawn from the reference position (X ₀ , Y ₀ ) to the original position (X _si , Y _si ) of each subfield using the linear position shift (ΔX _si ^L , ΔY _si ^L ) of each subfield, Interpolate the non-linear positional deviation of the intersection of the position detection marks around the subfield area with the line extending until it intersects the adjacent lines from the non-linear positional deviation of the marks adjacent to the intersection. Te, the original position of the sub-fields from the reference position _(X 0, _{Y 0)} of the line _{(X si, Y} si) and the length _{R si} to, from the reference position _(X 0, _{Y 0)} to the intersection Means for estimating the non-linear misregistration (ΔX _si ^NL , ΔY _si ^NL ) of each subfield as a function of the ratio to the length R _{m of}
An exposure apparatus comprising:

Images