JPS63284815A - Projecting optical equipment - Google Patents

Abstract

PURPOSE:To measure the image focusing characteristics of a projecting optical system with a high accuracy and a high reproducibility by specifying the period of the arrangement of bright parts and dark parts of a pattern. CONSTITUTION:A 1st pattern 100 which has bright parts and dark parts and a 2nd pattern 110 which has a light transmitting part are provided. When the image focusing characteristics of a projecting optical system are measured by projecting one of the two patterns onto the other by the projecting optical system and detecting the extent of the overlapping of both the pattern and the projected image, if the period of the arrangement of the bright parts 106 and the dark parts 102 is defined as (2a) and the multiplying factor of the image of the 1st pattern projected onto the 2nd pattern is defined as beta (>0), (a) is selected in such a manner that the intensity distribution of the projected image of the 1st pattern does not contain Fourier components with periods 2betaa/(2m+1) [wherein (m) denotes a positive integer] and the width of the light transmitting part 112 of the 2nd pattern 110 is selected to be (betaa). With this constitution, the signal of the projected image detected photoelectrically becomes a sine wave so that the position of the image can be measured with a high accuracy and a high reproducibility.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application in Industry A] The present invention relates to a projection optical device, and particularly relates to an improvement in a method for measuring distortion or magnification of a projection optical system.

[Prior Art] A typical device for measuring the imaging characteristics of a projection optical system that projects a predetermined pattern in a projection optical device is shown in FIG. 6, for example.

In this figure, measurement light having the same wavelength as the exposure wavelength output from a light source 10 is condensed by a condenser lens 12, and is made to enter a reticle or mask 14, which is an object to be illuminated.

A measurement mask pattern 16 is formed on the mask 14, and an image of this mask pattern 16 is projected onto the projection optical system 1.
8 is projected onto the detection surface 20.

This detection surface 20 is provided with an appropriate slit 22, and the measurement light transmitted through this slit 22 is sent to a detector 26, such as a photomultiplier, arranged in a stage 24 on which the detection surface 22 is formed. The total amount of light transmitted through the slit 22 is thereby converted into an electrical signal.

The stage 24 is configured to be movable in a required direction by a motor 28, and its coordinate position is measured by an interferometer 30.

Additionally, a detector 26, a motor 28, and a pedometer 30
are respectively connected to the control device 32, and the detector 2
A light quantity signal outputted from the interferometer 6 and a device signal constituted by the interferometer 30 are inputted. The motor 28 is driven based on commands from the control device 32.

Next, the operation of the above device will be described. The mask pattern 16 is illuminated by the measurement light output from the light source 10, and its image is projected onto the detection surface 20 by the operation of the projection optical system 18.

When the motor 28 is driven based on a command from the control device 32, the stage 24 moves. That is, the slits 22 are arranged eight times with respect to the projected image, and the projected image is scanned by the detector 26.

FIG. 7 shows the change in the amount of light input to the detector 26 with respect to the stage position. This example is an example where an aberration exists in the projection optical system 18 that is the object to be measured.

For example, if this graph is sliced at an appropriate level v1 and the midpoint P1 is found, this becomes the center position of the projected image.

Next, the center position of the projected image when it is assumed that there is no aberration in the projection optical system 18 can be determined in advance from the design data.

By comparing this designed center with the measured center as shown in FIG. 7, the imaging characteristics of the projection optical system 18 can be obtained.

[Problems to be Solved by the Invention] However, as shown in FIG. 7, if the projection optical system 18 has aberrations, the signal becomes asymmetrical.

Therefore, when determining the center position of a projected image, the value will differ depending on where in the signal the signal is sliced.

For example, if the center is found by slicing at level 2 in the figure, the result will be P2, which will result in a deviation of Δ compared to the case described above. This deviation directly results in measurement error.

Also, the slice level to find the center of the projected image is
It is not constant in the strict sense, but varies within a predetermined range.

Such fluctuations also cause measurement errors, resulting in a disadvantage that the reproducibility of measurements decreases.

The present invention has been made in view of the above, and an object of the present invention is to provide a projection optical device that can measure the imaging characteristics of a projection optical system with good accuracy and good reproducibility. be.

[Means for Solving the Problems] The present invention provides a first pattern having a bright portion and a dark portion, and a second pattern having a transparent portion, respectively, with a projection optical system in between. When measuring the imaging characteristics of the projection optical system by projecting one pattern onto another pattern using a projection optical system and detecting the degree of overlap between the two; the arrangement period of the bright and dark areas of the first pattern; is 2a, and the imaging magnification of the second pattern with respect to the first pattern by the projection optical system is β(〉
0); the a is set so that the intensity distribution of the projected image of the first pattern does not include a Fourier component with a period of 2βa/(2m+1) (m is a positive integer); the second pattern The width of the transparent portion is set to βa.

According to one aspect of the invention, the second pattern is
Two or more of the transparent parts are arranged in parallel with a period of 2nβa (Ω is a positive integer).

According to another aspect of the invention, a is any of the following, or, when the numerical aperture of the projection optical system on the first pattern is NA, the sigma value is σ, and the wavelength of illumination light is λ. is set to satisfy the following.

According to still another aspect of the present invention, the first pattern has a width of both a bright part and a dark part; a is a numerical aperture of NA on the first pattern of the projection optical system. ,
When the sigma value is σ and the wavelength of the illumination light is λ, it is set to satisfy either of the following.

[Operation] According to the present invention, the intensity distribution of the projected image of the first pattern is set so as not to include a Fourier component with a period of 2βa/(2m+1) (m is a positive integer). Based on this condition, the period of the first pattern and the width of the transparent portion of the second pattern are set.

Under such conditions, the photoelectric signal obtained when detecting the degree of pattern overlap becomes a sine wave, and no matter which one novel is sliced and its center determined, it will match.

[Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Examples of mask patterns according to an embodiment of the present invention are shown in FIGS. 1A and 1B, respectively. In these figures, the mask 100 includes a light shielding portion 102.10.
4 are formed and arranged at appropriate intervals using a material such as chromium. Its period is 2a.

The mask pattern is composed of the light-shielding parts 102, 104 arranged periodically as described above, and the light-transmitting parts 106, 108 formed between them.

Note that the widths of the light shielding parts 102 and 104 are different between the one shown in FIG.

Next, FIGS. 2A and 2B show examples of slits according to the embodiment. First, the same figure (A)
A slit 112 whose width is represented by βa is formed in the detection surface 110 shown in FIG. 6, where the imaging magnification of the projection optical system 18 (see FIG. 6) is β (>0).

Further, on the detection surface 110 shown in FIG. 3B, slits 114 and 116 each having a width βa and a distance between their centers of 4βa are shown.

The mask 10σ having the mask pattern as described above and the detection surface 110 having slits are applied instead of the mask 14 and the detection surface 20 of the apparatus shown in FIG.

Operation of the arrangement Next, the operation of the embodiment configured as described above will be explained.

As mentioned above, the width of the slits 112-116 is βa. Therefore, the intensity distribution of the projected image of the mask pattern on the detection surface 110 has a period of 2 as shown in FIG. 3(A).
If it is a sinusoidal wave of βa, it is connected to the slits 112 to 11.
6, the photoelectric signal obtained by the detector 26 when scanning in the direction of arrow F in FIG. 6(A) also becomes a sine wave with a similar period as shown in FIG. 6(B).

Also, suppose that in the Fourier component of the above projected image intensity distribution,
The period as shown in the same figure (C) is 1/2.1/4 of 2βa
．． Even if objects of 1/2 m... exist, the energy after passing through the slit is as shown in the same figure (°
As shown in D), it becomes a component that does not change even if the slit is moved, that is, a DC component. Therefore, even in such a case, the total signal is a signal that changes in a sinusoidal waveform as in the case described above.

In this way, when the photoelectric signal detected by the detector 26 becomes sinusoidal, as shown in FIG. 4, the center position will be the same no matter what level the slice is sliced at, improving measurement accuracy and reproducibility.

Elimination of B by Mathematical Formulas of Functions Next, the functions of the above embodiments will be explained in detail below using mathematical formulas.

First, a case where the slit 112 shown in FIG. 2(A) is used will be described.

The intensity distribution of the projected image on the detection surface 110 described above is expressed by I (x) with respect to the position [X].

When r (x) is decomposed into Fourier components, we get the following (1
) is as follows.

At I (X)°, 4-I (4m) exp[i rt
h X] ,,, (1) Here, the period of the image is 2
Since βa, n*-27E k/2aβ,
I(y+k)-I"(-nk). Also, * means complex conjugation.

Next, if the slit 112 with width βa is scanned and its center position is t, the total amount of light E (t) detected at that time is ・(2/ηh)・5in(y+k(βa/2))−
(2aβ/ πk)・sin (π to/2)Oo
~ ・Σ (I (n J ・exp[i n kt] + r
(-ηk)・・・・・・(2) However, when k=2+n (m is a positive integer), sin
(rt k/2)-0, and since there is no Fourier component with a period of 2βa/(2m+1), there is also no term k=2m+1 (m is a positive integer).

Therefore, equation (2) is E (t)-aβI(0) +1(π/aβ)exp[iπt/aβ](2aβ/
π)-I (-re /aβ)exp[-i rt t
/aβ1(2aβ/π)=A+Bcos((2πt/
2aβ)+C1・−ψ・(3), and the signal becomes a sine wave signal. A here.

B and C are appropriate constants.

Therefore, no matter which level on the signal strength graph is used,
The center positions are determined to be the same value, improving the reproducibility of measurements and making them more resistant to noise.

Furthermore, as shown in FIG. 2(B), the period 2Lβa(I
(L is a positive integer) and may be formed by a plurality of slits each having a width βa.

In this case, if the number of slits is L, the signal strength Et
, (t) is expressed as -L-E(t), and the same result as when there is a single slit can be obtained. In this case, since the signal strength is multiplied by L, it is possible to perform measurements with better reproducibility and resistance to noise.

21 Blood Emperor Difference Next, the conditions under which the above-mentioned sinusoidal signal can be obtained are generally determined as follows.

According to "Fourier imaging theory" (written by Teruji Kose, Kyoritsu Shuppan), the intensity I (x) of the projected image by the projection optical system 18 at position X is the amplitude transmittance of the mask pattern (see Figure 1). The Fourier component is set to 0 (ν) for the spatial frequency ν.
Assuming that the homopolarized light source is 5E(S), the number of pupil planes of the projection optical system 18 is G(ν), and the proportionality constant is α, it is expressed by the following equation.

1(X)・α・SO(ν)・0″(υ°)・G(v+
5)-G"(v"+5)・5c(S)・exp[
i (v −v')x/β]dvdv"ds...
...(4) Although the mask pattern and slits are shown in a -dimensional representation here, they are both periodic patterns in one direction, as shown in Figures 1 and 2. That's enough.

By the way, since the basic period of the mask pattern is 2a, the spatial frequency ν takes a discrete value υ, and becomes 1゛2・k(k#f regularity λ...(5).ν
The interference term between kl and υ with 2 is a ・j dsO(+ on υ) O'(v k2)・G(
Shear 1◆S)G”(ν2S>SE (S)・exp
[i (υ, -shi, 2)・×/β]+ (t ・5
dso(ij k2) 0 blame 1'kl)・G(shi, 2
◆5) G responsibility νkI+S) SE(S), ・exp[
i (V kz-v m+) ・x/β], a'-c
os((k+-2)・X/2aβ], but I k
If the term +-kal-2m+1 (m is a positive integer) becomes zero, the signal becomes a sine wave.

(1) The first case that satisfies this condition is that when the difference between the orders of two diffracted waves is "3", both do not pass through the projection optical system 18 at the same time.

In FIG. 5(A), the above conditions are generally shown as an optical path diagram between the mask pattern and the projection optical system.

The condition for preventing the +3rd order light from entering the projection optical system 18 with respect to the order diffracted light is determined by assuming that the numerical aperture of the projection optical system 18 on the mask pattern side is NA, 3λ -> 28A a . 3λ aku□ ・・・・・・・・・(6)NA becomes.

(2) Next, the second sufficient condition is that diffracted light of a higher order than the secondary light does not pass through the projection optical system 18.

Such a case is shown in FIG. 5(B). From this figure, the conditions for blocking ±secondary light are the projection optical system! The numerical aperture of the transmitted light source 120 at position 8 with respect to the mask pattern is σ
As NA, 2λ □−σNA>NA a, and λ a<(1°.)NA (7).

(3) In addition, as shown in FIG. 5(C), when the light-transmitting part and the light-blocking part of the mask pattern have the same width a, ±
There is no second-order diffracted light. For this reason, it is only necessary to consider a condition in which the ±3rd-order diffracted light is blocked. Therefore, in this case, 3λ □ - σNA > NA a , and 3λ a × 2 (1°.) NA ... ...(8
).

In any of the above cases, as shown in FIG. 4, the change in position of the amount of incident light on the detector 26 becomes sinusoidal, and the center position can be determined satisfactorily.

Numerical Examples Next, specific numerical examples will be explained. For example, N
A = 0.35. G = 0.5. λ=0.
When it is 4358 μm, the period of the mask pattern is a<0
j93 μm is sufficient, and when the widths of both the light transmitting portion and the light shielding portion are a, it is sufficient that a<1.25 μm.

Next, N A = 0.35. σ=0.2. λ
= 0.4358 μm, it is sufficient that a<1.04 μm, and when both the light transmitting portion and the light shielding portion are a, it is sufficient that a<1.55 μm.

11 cases of enemy 9j Next, other embodiments will be described.

(i) In the embodiments described above, the mask pattern is projected onto the slit side by a predetermined light source, but according to the reciprocity theorem, the same effect can be obtained even if the light source and the detector are interchanged. Therefore, when this invention is applied to an exposure apparatus used in integrated circuit manufacturing, etc., it is possible to reverse the direction of illumination during exposure and when measuring imaging characteristics. It goes without saying that there is.

That is, the slit is illuminated with a sufficiently large numerical aperture so that the σ value due to the numerical aperture of the detector relative to the mask pattern is the same as the σ value due to the numerical aperture of the exposure illumination, and the light passing through the mask pattern is detected. do it. At this time,
The illuminated part is a slit having a width βa or a periodic one (period 2nβa), and the pattern on the mask is a pattern having a period 2a.

(if) Measurement is also possible by using the entire optical system in reverse. That is, a slit with a width a or a periodic slit with a period of 2na is placed on the mask that is illuminated during exposure, and a slit with a period of 2β is placed at the position where an image is created during exposure.
Place pattern a. Further, during exposure, illumination is performed from the side where the image is formed with the same σ value as during exposure, and the light passing through the pattern on the mask is detected by a detector having a sufficiently large frontage angle.

In this case, the numerical aperture NA' of the projection lens at the position where the image is created during exposure is NA' = NA/β, and the pattern period 2a' = 2βa, so the σ values are the same, so (6), ( 7) and (8) are similarly a
' , NA' holds true.

(iii) In addition, a slit of width a or one that opens periodically with a period of 2na is placed on the mask that is illuminated during exposure, and a slit with a period of 2βa is placed at the position where an image is formed during exposure.
Leave a pattern. Then, the pattern on the mask is illuminated with a sufficiently large beam A, and the amount of light that passes through the pattern at the position where the image will be formed during exposure is determined by the numerical aperture of the detector. Detection is performed using a detector whose σ value is equal to the value on the mask.

In this case as well, similarly to (ii), the numerical aperture N of the projection lens
If A' = NA/β and pattern period 2a' = 2βa, equations (8), (7), and (8) are similarly established.

Further, the mask pattern and the shape of the slit are not limited to those shown in the above-mentioned embodiments, and various forms exist.

In particular, if the mask pattern has three or more periods, it can be approximately regarded as a periodic object, and the measurement reproducibility will be good.

Furthermore, as a detector, a CCD or the like may be used in addition to a photomultiplier.

[Effects of the Invention] As explained above, according to the present invention, since the signal of the projected image that is photoelectrically detected is a sine wave, there is an effect that the image position can be measured with high precision and good reproducibility. .

[Brief explanation of drawings]

Fig. 1 is a perspective view showing a mask pattern in an embodiment of the present invention, Fig. 2 is a perspective view showing a slit in the same embodiment, Fig. 3 is a graph showing the effect of the embodiment, and Fig. 4 is an embodiment. Graph showing an example of the detected photoelectric signal in the fifth
The figure is an explanatory diagram showing a mode for obtaining a sine wave signal, Figure 6 is a block diagram showing the general configuration of an apparatus for measuring the imaging characteristics of a projection optical system, and Figure 7 is a detection signal of a conventional apparatus. It is a graph showing an example. [Explanation of symbols of main parts] 10...Measurement light source, 14, 100...Mask, 16
. . . Mask pattern, 18 . . . Projection optical system, 20.
110...Detection surface, 22, 112, 114°116.
...Slit, 26...Detector, 102°104
... Light-shielding part, 106.108... Light-transmitting part.

Claims (4)

[Claims] (1) A first pattern having a bright part and a dark part and a second pattern having a transparent part are arranged with the projection optical system in between, and either one pattern is applied to the other pattern by the projection optical system. In a projection optical device that measures the imaging characteristics of the projection optical system by projecting images and detecting the degree of overlap between the two, the arrangement period of the bright portions and dark portions of the first pattern is set to 2a, and the projection optical system comprises: When the imaging magnification of the second pattern relative to the first pattern by the system is β (>0), the intensity distribution of the projected image of the first pattern is 2βa/
A projection optical device characterized in that the projection optical device is set so as not to include a Fourier component with a period of (2m+1) (m is a positive integer), and the width of the transparent portion of the second pattern is βa. (2) The second pattern has a transparent part with a period of 2nβa
2. The projection optical device according to claim 1, wherein two or more projection optical devices (n is a positive integer) are arranged in parallel. (3) When the numerical aperture of the projection optical system on the first pattern is NA, the sigma value is σ, and the wavelength of the illumination light is λ, a<λ/{(1+σ)NA} or , a<3λ/4NA. (4) The width of both the bright and dark parts of the first pattern is a, and a is the numerical aperture of the projection optical system on the first pattern, NA is the sigma value, and σ is the illumination light. The projection optical device according to claim 1 or 2, which satisfies either a<3λ/{2(1+σ)NA} or a<3λ/4NA, where the wavelength of is λ.