JP2594660B2 - Charged particle beam exposure method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Overview] Regarding a charged particle beam exposure method, there is provided a charged particle beam exposure method capable of accurately calibrating a beam deflection coordinate value and a mask coordinate system and improving the exposure performance. A mask plate having a plurality of aperture patterns, and irradiating one of the plurality of aperture patterns with a charged particle beam by deflecting the charged particle beam by an aperture pattern selecting deflector. In a charged particle beam exposure method for forming a particle beam and irradiating the sample on a sample, a calibration pattern formed on the mask plate and having a predetermined size and arrangement position is deflected by the aperture pattern selecting deflector. Irradiation with the charged particle beam, and then further charged the charged particle beam by the aperture pattern selecting deflector Deflection, detecting a deflection condition at the boundary where the current value of the charged particle beam reaching the sample becomes positive to zero, and based on the deflection condition and the size and arrangement position of the calibration pattern, The position and size of the charged particle beam at are calculated.

[Industrial applications]

The present invention relates to a charged particle beam exposure method, and more particularly, to a charged particle beam exposure method for forming a fine pattern with a charged particle beam, for example, an electron beam.

In recent years, as the density of integrated circuits has increased, instead of photolithography, which has been the mainstream of fine pattern formation methods for many years,
A new exposure method using an electron beam is being studied and actually used.

[Conventional technology]

A conventional electron beam exposure apparatus is a drawing apparatus that deflects and scans an electron beam on a sample wafer using a variable rectangular beam to draw a pattern.

Such an apparatus is an apparatus having a pattern generation function of creating hardware called a pattern from pattern data that is software.However, since a pattern is drawn by connecting rectangular shots, the unit becomes generally smaller as the pattern size becomes smaller. The number of exposure shots per area increases,
There is a problem that the throughput is reduced.

Therefore, a block exposure method has been proposed to address this problem and obtain a realistic throughput even in the exposure of an ultrafine pattern.

That is, most semiconductor devices that require an ultra-fine pattern are, for example, 64MDRAMs, which are repetitions of a basic pattern that has a small but almost all exposed area. If a basic pattern, which is a unit of a repetitive pattern, can be generated in one shot regardless of its own complexity, it becomes possible to expose such a pattern at a constant throughput regardless of the fineness.

Therefore, a method in which a basic pattern as described above is held on a transmission mask and irradiated with an electron beam to generate a basic pattern in one shot, and the pattern is connected to repeatedly expose the pattern is a block exposure method. .

One example of such an exposure method is IEEE TRANS.ON ELECTR
ON DEVICE vol.ED-26 (1979) 663,
The configuration is as shown in FIG. In the figure, an electron beam 2 emitted from an electron gun 1 passes through a first rectangular shaping aperture 3 and is converged by a lens 4 and then deflected by a pattern selecting deflector 5 placed at an image point of a crossover. Irradiation is performed on an arbitrary pattern portion on the mask 6. On the stencil mask 6, a variable rectangular aperture 6A, an adjustment mark pattern 6B, and a basic pattern for repetition 6C are formed, and a lens 7 is arranged close to the stencil mask 6. Therefore, after the electron beam 2 has its cross section patterned by the stencil mask 6, it is returned to the optical axis by the convergence of the lens 7, passes through the deflector 8 for returning, and is reduced in cross section by the reduction lens 9, and the projection lens 10. Then, the light passes through the deflection system 11 and is exposed on the wafer 12. Thus, for example, in the case of a memory cell, several cells are exposed in one shot.

However, in this method, since the electrons 2 deflected from the optical axis are returned to the optical axis only by the contraction of the lens 7, the electrons 2 follow different orbits in the magnetic lens 7 depending on which pattern is selected. Come. Also, in order to arrange as many patterns as possible on the stencil mask 6, it is necessary to make the electron trajectory pass through a portion as far away from the optical axis as possible. The effect may be significant.

Therefore, the applicant of the present invention has previously proposed an apparatus having an electron optical system to solve this problem. The configuration of this device is the same as that of the later-described embodiment, and will not be described. In this apparatus, two lenses are provided vertically above and below a stencil mask, and an electron beam is deflected above and below the stencil mask by an aperture pattern selecting deflector, and the electron beam is turned back to pass the center of the upper and lower lenses. You have selected a pattern.

[Problems to be solved by the invention]

Incidentally, in the block pattern transfer type electron beam exposure apparatus according to the above-mentioned prior application, the electron optical system is set so that the image of the first rectangular shaping aperture is formed on the stencil mask, and the incident side mask deflection is performed. By changing the image position on the stencil mask, the desired block pattern position is selected and the beam is passed through, or the beam is passed on the same block pattern to change the beam cross-sectional shape on the wafer. And the size of the stencil mask, the relationship between the input data to the incident-side mask deflector and the image position (beam passage position) of the aperture on the stencil mask can be calibrated with high accuracy. It is necessary to determine the size of the aperture image, which is difficult in practice, and ultimately improves the exposure performance. There is a problem that had.

For example, if the shot size is specified on the wafer with a precision of 0.01 μm and the reduction ratio from the mask to the wafer is 1/100, it is necessary to set the aperture image position (beam passage position) on the mask with a precision of 1 μm. is there. For this calibration, a calibration mark pattern is formed on the mask. The calibration mark pattern is usually a rectangular pattern of about 300 μm square in consideration of the reduction ratio, and the size of the mark pattern is accurately determined when the mask pattern is formed. on the other hand,
The size of the aperture image on the stencil mask is
The size is about 300 μm □, and this size varies depending on the lens conditions, and cannot be specified accurately. That is,
As shown in FIG. 7, a beam whose cross-sectional size is not determined passes through a calibration mark pattern (rectangular hole) 21 whose size L (up to 300 μm) and arrangement position (X ₀ , Y ₀ ) are known in advance. Therefore, it is necessary to determine the beam position and the size with an accuracy within 1 μm. For example, the aperture image 22 on the stencil mask is represented by the position of the representative point B at the lower left of the beam and the beam size x, y, and the calibration of the deflector is finally performed in this coordinate system. In the beam calibration, conditions for the maximum beam to pass through the calibration mark pattern portion are usually determined and adjusted.

However, in such a method, unless the size of the calibration mark pattern and the beam size on the stencil mask are exactly equal, the beam passing conditions cannot be set with sufficient accuracy, and the exposure performance deteriorates. .

That is, FIG. 8 shows an example in which the size of the beam 23 is larger than the size of the calibration mark pattern 21. Although the beam positions are different in FIGS. 8A and 8B, the passing beam amounts are almost the same. Cannot be distinguished, so that the beam passage conditions cannot be set with sufficient accuracy. FIG. 9 shows an example in which the size of the beam 23 is smaller than the size of the calibration mark pattern 21. In this case, the beam positions are different between FIGS. Therefore, since the two cannot be distinguished from each other, the beam passage conditions cannot be set with sufficient accuracy.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a charged particle beam exposure method which can calibrate a deflection coordinate value of a beam and a mask coordinate system with high accuracy and enhance exposure performance.

[Means for solving the problem]

The charged particle beam exposure method according to the present invention includes a mask plate having a plurality of opening patterns to achieve the above object,
By irradiating one of the plurality of opening patterns by deflecting the charged particle beam by an opening pattern selecting deflector, the charged particle beam exposure method of shaping the charged particle beam and irradiating the sample with a sample is performed. A pattern for calibration formed on the mask plate and having a predetermined size and arrangement position is irradiated with a charged particle beam deflected by the aperture pattern selecting deflector, and then the charged particle beam is irradiated with the charged particle beam. The deflector is further deflected by the aperture pattern selecting deflector, and the deflection condition at the boundary where the current value of the charged particle beam reaching the sample becomes from positive to zero is detected. The deflection condition and the size and arrangement position of the calibration pattern are detected. The position and size of the charged particle beam on the mask plate are obtained based on the above.

[Action]

In the present invention, the current value associated with the irradiation of the charged particle beam is measured on the sample side, whereby the passing state of the charged particle beam in the calibration mark pattern whose size and arrangement position formed on the mask plate are known in advance is known. A beam is passed through the calibration mark pattern under at least two or more conditions capable of recognizing the pattern, and the mask is determined from the deflection conditions of the mask plate for each condition and the size and arrangement position of the calibration mark pattern known in advance. The beam position and size on the plate are determined.

Therefore, since the mask coordinate system is known in advance,
The deflection coordinate value of the beam can be calibrated with high accuracy in the mask coordinate system.

〔Example〕

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

1 to 5 are views showing one embodiment of a charged particle beam exposure method according to the present invention. FIG. 1 is a configuration diagram of an electron beam exposure apparatus to which the exposure method of the present invention is applied. In this figure, 31 is an electron gun, 32 is an anode, 33 is an aperture,
34 is a first lens, 35 is a deflector, 36 is a second lens, 37 is a third lens, 38A and 38B are incident mask deflectors, 38C and 38D
Is an emission mask deflector, 39 is a stencil mask, 40 is a deflector drive circuit that gives a deflection condition to each deflector,
41 is a blanking electrode, 42 is a reduction lens, 43 is a diaphragm aperture, 44 is a first projection lens, 45 is a second projection lens, 46
Is a deflector (corresponding to irradiation deflection means), 47 is a wafer, 48 is an electron beam, and 49 is a current measuring device (corresponding to current measuring means) for measuring a current value of the wafer 47.

Second lens 36 and third lens 37 (corresponding to lens system)
Between, the electron beam 48 is almost a parallel beam,
Incident mask deflectors 38A and 38B for selecting aperture patterns
The electron beam 48 is oscillated above and below the stencil mask 39 by the emission mask deflectors 38C and 38D, and is turned back to select a mask pattern on the stencil mask 39 under the condition of passing through the lens centers of the lenses 36 and 37.

The stencil mask (corresponding to a mask plate) 39 has a structure as shown in FIG. 2, and a semiconductor substrate such as Si or a metal plate is used as a mask substrate. 2A is a plan view of the stencil mask 39, and FIG. 2B is a stencil mask 39.
FIG. The pattern forming portion of the stencil mask 39 is thinned as shown in the cross-sectional view, and a punched pattern is formed on the thinned portion using an etching technique. The stencil mask 39 has an electron beam irradiable area 39A and a pattern formable area 39B, and a non-repeated pattern opening 39C and a pattern 39D are formed in the area 39B.

A mask deflector 38 is provided on the stencil mask 39.
A plurality (a plurality of sets) of pattern groups (sets of patterns) which are a group of patterns selectable from A to 38D are prepared, and each set of patterns is arranged in the vicinity of the electron optical axis by the X and Y stages supporting the mask. Can be moved.
Further, on the stencil mask 39, a mark pattern 61 for calibration described later is formed. In addition,
In order to load the mask 39 on this stage, a mask loading sub-chamber is provided which can be separated from the column body by a gate valve.

The mask deflectors 38A to 38D constitute a block deflecting unit 51, and the block deflecting unit 51 transmits the beam to a desired opening in the pattern group based on the output of the deflector driving circuit 40, and the beam incident side of the stencil mask 39. Then, while deflecting the beam from the optical axis determined by the lens systems 36 and 37, and irradiating the beam to a desired aperture in the pattern group,
On the beam exit side of the stencil mask 39, the beam is returned to the optical axis again after exiting the mask. In the present embodiment, the wafer 47
By measuring the beam current value by the current measuring device 49 on the side, the deflection coordinate values of the mask deflectors 38A to 38D are calibrated in the mask coordinate system.

In the above configuration, in this embodiment, the sample (wafer) 47 is used.
On the side, the current value accompanying the irradiation of the electron beam 48 is measured by the current measuring device 49, and thereby, the passage of the electron beam 48 in the calibration mark pattern whose size and arrangement position formed on the stencil mask 39 are known in advance. A beam is passed through the calibration mark pattern under at least two or more conditions that can recognize the situation, the deflection condition of the stencil mask 39 for each condition and the size and arrangement position of the calibration mark pattern known in advance. , The beam position and size on the stencil mask 39 are determined.

This will be specifically described as follows. Reference numeral 61 shown in FIG. 3 denotes a calibration mark pattern formed on the stencil mask 39, and its size L1, L2
(Up to 300 μm) and the arrangement position (X ₀ , Y
₀ ) is known in advance when the mask is created. Where L1 ≠ L
2 is acceptable. The position of the beam origin in the mask coordinate system and the beam sizes x and y must be obtained by passing the electron beam 48 through the mark pattern 61.

Calibration I Therefore, first, the beam current value is observed on the wafer 47 side, and a situation is set in which the beam passes through the mark pattern 61 under conditions such that the beam is irradiated on the beam patterns 62 and 63 shown in the position in FIG. This means that each of the electron beams 48 is slightly + X
A condition may be obtained in which the current value becomes zero when the current value is shifted to the + Y side. When the deflection coordinates of the block deflecting means 51 (coordinates at which the block deflecting means 51 deflects the beam) at this time are (X ₁ , Y ₁ ) and (X ₂ , Y ₂ ), the corner P of the mark pattern 61 The deflection coordinates of the corresponding block deflection means 51 can be determined as (X ₁ , Y ₂ ). Therefore, since the coordinates of the point P in the mask coordinate system are known in advance,
If the relationship between the two is used, the deflection coordinate value of the block deflection means 51 can be calibrated in the mask coordinate system.

Calibration II Further, the beam current value is observed on the wafer 47 side, and a situation is set in which the beam passes through the mark pattern 61 under conditions such that the beam irradiates the beam patterns 64 and 65 shown in the position of FIG. This means that the electron beam 48 is slightly shifted on the + X side, -X
A condition may be obtained in which the current value becomes zero when shifted to the side. If the deflection coordinates of the block deflecting means 51 at this time are (X ₃ , Y ₃ ) and (X ₄ , Y ₄ ), respectively, the difference X ₃ −
Become equal to X ₄ is a amount added to the size L1 size x in the X direction of the beam of the mark pattern 61. The deflection coordinate of the block deflecting means 51 and the mask coordinate system are correlated in the calibration I, and the size L1 or L2 of the mark pattern 61 is known in advance. be able to.

Calibration III Further, the beam current value is observed on the wafer 47 side, and a situation is set in which the beam passes through the mark pattern 61 under conditions such that the beam is irradiated on the beam patterns 65 and 66 shown in the position in FIG. This means that the electron beam 48 is slightly shifted on the + Y side and-
It is sufficient to find a condition that the current value becomes zero when the current value is shifted to the Y side. If the deflection coordinates of the block deflecting means 51 at this time are (X ₅ , Y ₅ ) and (X ₆ , Y ₆ ), the deflection amount difference Y ₅
Become equal to -Y ₆ becomes the amount added to the size L2 size y of the y direction of the beam of the mark pattern 61. The deflection I of the block deflecting means 51 and the mask coordinate system are correlated in the calibration I, and the size L1 or L2 of the mark pattern 61 is known in advance. be able to.

As described above, since the coordinates of the point P in the mask coordinate system are known in advance, the deflection coordinate value of the block deflecting unit 51 can be calibrated in the mask coordinate system by using the relationship between them. Therefore, calibration between the deflection coordinate value of the beam and the mask coordinate system can be performed with high accuracy, and the exposure performance can be improved.

〔The invention's effect〕

According to the present invention, the calibration between the deflection coordinate value of the beam and the mask coordinate system can be performed with high accuracy, and the exposure performance can be improved.

[Brief description of the drawings]

1 to 5 are diagrams showing an embodiment of a charged particle beam exposure method according to the present invention. FIG. 1 is a configuration diagram of an electron beam exposure apparatus to which the present invention is applied, and FIG. FIG. 3 is a diagram illustrating the structure, FIG. 3 is a diagram illustrating the method of the calibration I, FIG. 4 is a diagram illustrating the method of the calibration II, FIG. 5 is a diagram illustrating the method of the calibration III, 6 to 9 are views showing a conventional electron beam exposure apparatus, FIG. 6 is a block diagram of the apparatus, FIG. 7 is a view for explaining the calibration method, and FIG. FIG. 9 shows a case where the beam size is smaller than the mark pattern size. 31 ... electron gun, 32 ... anode, 33 ... aperture, 34 ... first lens, 35 ... deflector, 36 ... second lens (lens system), 37 ... third lens (lens system) , 38A, 38B: Incident mask deflector, 38C, 38D: Outgoing mask deflector, 39: Stencil mask (mask plate), 40: Deflector drive circuit, 41: Blanking electrode, 42: Reduction lens, 43 ... Aperture aperture, 44 ... First projection lens, 45 ... Second projection lens, 46 ... Deflector (irradiation / deflection means), 47 ... Wafer (sample), 48 ... Electron beam, 49 ... Current Measuring device (current measuring means), 51: Block deflecting means, 61: Mark pattern for calibration, 62 to 66: Beam pattern.

Claims (1)

(57) [Claims] A mask plate having a plurality of aperture patterns, wherein the charged particle beam is deflected by an aperture pattern selecting deflector to irradiate one of the plurality of aperture patterns with the charged particle beam. In a charged particle beam exposure method of shaping a beam and irradiating the sample on a sample, a calibration pattern formed on the mask plate and having a predetermined size and arrangement position is deflected by the opening pattern selecting deflector. Then, the charged particle beam is further deflected by the aperture pattern selecting deflector, and a deflection condition at a boundary where the current value of the charged particle beam reaching the sample becomes from positive to zero is detected. Charging on the mask plate based on the deflection condition and the size and arrangement position of the calibration pattern. A charged particle beam exposure method characterized by determining the position and size of the child beam.