JP2015007725A - Optical imaging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an optical microscope that provides high contrast of images, suppresses the leakage of light, and enables observation of light from a minute area while the light being distinguished from light from the adjacent area.SOLUTION: There is provided an optical microscope having a pupil filter. In the pupil filter, complex transmittance of the pupil surface of the optical microscope is a function of a radius r; an amplitude part is represented by a function obtained by multiplying the SINC function by the Hanning window function; and a phase part is constant or represented by a function obtained by approximating its amplitude part by a step function or a smooth function, and has the maximum in a peripheral part.

Description

  The present invention relates to an optical imaging apparatus.

  Microscopic systems are widely used in lithography exposure apparatuses for semiconductor microfabrication and the like and in the observation of microstructures in the industrial and biotechnology fields. In the two systems, light passes through the same type of structure in the opposite direction. First, take a microscopic system as an example.

  FIG. 1 shows a normal microscope system or an immersion microscope system. Light emerging from the focal plane travels through a liquid such as air or water to the microscope objective. In the microscope, there is a “pupil plane” onto which the two-dimensional distribution of light on the focal plane is mapped as a two-dimensional Fourier transform function. Further, the image on the pupil plane is observed by a person through an eyepiece lens or mapped onto a photosensitive surface through a photographic apparatus. The image on the focal plane is said to be conjugate with the focal plane, and the conjugate planes are equivalent to each other except for the enlargement / reduction of the image.

  In a thin lens, the exit pupil plane exists behind the lens by a focal length. In the high-magnification objective lens shown in FIG. 2 as an example, the exit pupil plane is often located inside the group lens and may be located in the glass of the lens. In the microscopic system, there are usually one or more planes conjugate to the exit pupil plane.

Of the light emitted from the focal point, light having a small angle with the optical axis of the microscope can be easily collected, but in order to increase the resolution, light having a maximum angle is used. The maximum angle is θ, the refractive index of the medium between the focal plane and the objective lens is n, and the wavelength of the light to be used is λ. On the other hand, in an exposure apparatus that is paired with a microscope, the optical path is opposite in the exposure apparatus. Usually, output light from a laser is used as a light source for exposure. Let that wavelength be λ. Light travels in the opposite direction through the objective lens, passes through the pupil plane, and travels toward the focal plane. In an objective lens configured by combining a single lens or a plurality of bonded lenses, the surface closest to the focal point is usually a flat surface. Between that plane and the focal plane, there is a liquid-filled one and a dry one that is not.
The resolution in the microscope and the minimum spot diameter in the exposure apparatus are both several times the quantity λ _m expressed as λ _m = λ / nsin θ. The smaller λ, the larger n, and the larger θ are advantageous for increasing the resolution.

In the microscope system, light emitted equally from one point in all directions of the space enters the objective lens, passes through the pupil plane, and forms an image through an eyepiece lens or the like. The image is given by the two-dimensional Fourier transform of the electromagnetic field on the pupil plane. The electromagnetic field of light uniform in the angular direction is simplified to the Hankel transformation. In this case, an electromagnetic field having a uniform amplitude and a uniform phase is induced on the pupil plane, and a two-dimensional Fourier transform or Hankel transform image having a uniform distribution is formed. FIG. 3 shows an equivalent projection of the image onto the focal plane of the objective lens. Centered in amplitude maximum, until far have then ripple becomes zero at a radius away about 1 times half of lambda _m (ripple). It is desirable for the image to have high resolution and low crosstalk with other light spots, but it has a small distance to the first zero (defined as the radius of the light beam) and the amplitude ratio of the maximum ripple to the main beam (ripple Rate) is understood to be small. In the above sense, the image quality induced by the uniform excitation in the pupil plane is represented by a small circle in FIG.

  Next, image formation in the exposure apparatus will be described. The purpose is to form a sharp image on the focal plane of the objective lens. Light from the laser light source is guided to the pupil plane, and an image is formed on the focal plane by the objective lens. The electromagnetic field distribution at the pupil plane determines the quality of the imaging at the focal plane. Among the various pupil plane excitation methods, the standard is the same uniform excitation as above, and the distribution of the beam in the focal plane is represented by the curve in FIG.

型である。ただし波はz方向に向かうものとし、x, yはそれと直交する座標である。 Excitation by Gaussian wave The first category of pupil plane excitation is excitation by Gaussian wave. The light beam is usually the thinnest at the plane called the beam waist, and the beam diameter gradually increases as it moves away from it. A laser pointer or the like is typical, and the diameter of the beam waist is on the order of mm or cm even if the diameter of the beam waist propagates from several meters to several hundred meters. In contrast to this, the exposure apparatus uses a beam waist with a wavelength order, and the entire beam forms a drum shape with a large apex angle. In both cases, the field distribution at the beam waist is Gaussian, ie

It is a type. However, the wave goes in the z direction, and x and y are coordinates orthogonal to it.

Due to the magnitude relationship between the (physical) pupil plane radius and w in the above equation, when w >> physical radius: substantially uniform excitation on the pupil plane. The focal plane beam is sharp, but the utilization efficiency of laser light is low.
As W becomes smaller, utilization efficiency increases, but the focal plane beam radius increases. At the same time, the ripple rate is improved. This is shown by the curve shown by the broken line in FIG. This is understood by the following consideration.

The field distribution on the pupil plane (written as pupil plane function for simplicity) and the field distribution on the focal plane (focal plane function) are connected to each other by two-dimensional Fourier transform; that is, if the pupil plane function is two-dimensional Fourier transformed A focal plane function is obtained, and a pupil plane function can be obtained by two-dimensional Fourier transform of the focal plane function. Similarly, the Fourier transform of a two-dimensional Gaussian function is a two-dimensional Gaussian function. As a characteristic of the Fourier transform, if the original function is spatially K times (expanded if K> 1 and reduced if K <1), the transformed function will be spatially 1 / K times. In order to reduce the beam diameter of the focal plane in a Gaussian light beam, it is necessary to expand the Gaussian function in the pupil plane. The geometric maximum radius of the pupil plane is equivalent to an amount expressed as n sin θ in terms of the maximum light receiving angle and refractive index of the optical system. In order to have a sharp peak in the Gaussian function on the focal plane, it is inevitable that the bottom of the Gaussian function on the pupil plane is blocked by the maximum radius of the pupil plane.

Excitation through pupil filter It is important to reduce the radius of the main beam in a system intended for microfabrication. For this reason, it has recently been found that it is effective to darken the central part of the pupil surface and brighten the peripheral part (Patent Document 1). There are two degrees of freedom: area division and inner transmittance. As shown in FIG. 5-1, the pupil plane is set to the inside / outside radius ratio α: (1-α), the outside transmittance = 1, the inside transmittance is b, b is a semi-fixed parameter, and a is variable. If the focal plane beam radius and ripple rate are analyzed as parameters, the result in Fig. 5-2 is obtained. If the pupil plane excitation is indented with respect to a two-dimensional variable, the radius of the main beam can be reduced, and the ripple rate is degraded (increased) as a price.
The pupil filter has been developed for an exposure system, but is not used for a microscope because of a high ripple rate.

JP 2009-290206 A

Shinya Inoue, Kenneth R .; Spring; Video microscope, its basics and usage SECOND EDITION, Susumu Terakawa, Seiji Ichie, Akira Watanabe, Kyoritsu Publishing Co., Ltd., 2001, 2.2 Max Born, Emil Wolf; Principles of Optics 7th (expanded) edition, CAMBRIDGE UNIVERSITY PRESS, 1999, 4.8.2, 8.2.2. H. Kogelnik, On the Propagation of Gaussian Beams of LightThrough Lenslike Media Inclusion this a Loss or Gain Op., Applied. 1562-1569, December 1965.

In industrial technology fields such as optical microscopes and lithography, it is essential to observe light from a minute area separately from light from an adjacent area, or to collect light from other light sources in a fine area. is important. It is well known that light of a certain wavelength (denoted by λ) can be imaged in a small region several times as long as λ. The competitiveness of microscopes and lithography equipment
・ One is to reduce the wavelength used (short wavelength), increase the allowable angle of incidence, increase the refractive index of the immersion medium, and the second is to reduce the proportional factor of “several times”. Yes.

  In particular, it is important that the contrast of the image is high and the leakage light is small. This is the same as searching for an optical design in which the operating point can be selected on the left and down on the two-dimensional map shown in FIGS.

A first aspect of the present invention relates to an optical microscope having a pupil filter.
This side pupil filter is, for example,
The complex transmittance of the pupil plane of the optical microscope is a function of the radius r,
It is a pupil filter whose amplitude is a monotonically decreasing function of r in the central part, has a maximum in the peripheral part, and makes its phase part constant.

に幅が

以上のＨａｎｎｉｎｇ窓関数を乗じて得られる関数Ａ（ｒ）にＨａｎｋｅｌ変換

を行って得られる関数Ｆ（β）で表され，
位相部分は一定ならしめるである瞳フィルタであるか，又はその振幅部分を階段関数または滑らかな関数で近似した瞳フィルタであってもよい。 The pupil filter on this side is
The complex transmittance of the pupil plane is a function of the radius β,
The amplitude part is a function

Width is

Hankel transform into function A (r) obtained by multiplying the above Hanning window function

Is expressed by a function F (β) obtained by performing
The phase portion may be a pupil filter that is constant, or may be a pupil filter that approximates its amplitude portion with a step function or a smooth function.

  The second aspect of the present invention relates to an exposure apparatus having a pupil filter.

This side pupil filter is, for example,
After the excitation light incident on the pupil plane passes through the pupil filter, the amplitude part is a monotonically decreasing function of r in the central part, and has a maximum in the peripheral part.
This is a pupil filter that makes the phase portion constant.

に幅が

以上のHanning窓関数を乗じて得られる関数Ａ（ｒ）にHankel変換

を行って得られる関数Ｆ（β）で表され，
位相部分は一定ならしめる瞳フィルタ，あるいはその振幅部分を階段関数または滑らかな関数で近似した瞳フィルタであってもよい。 The pupil filter on this side has a complex transmittance of the pupil surface as a function of the radius β,
The amplitude part is a function

Width is

Hankel transform to function A (r) obtained by multiplying the above Hanning window function

Is expressed by a function F (β) obtained by performing
The phase portion may be a pupil filter that makes the phase constant, or a pupil filter that approximates the amplitude portion with a step function or a smooth function.

  As described above, a microscope, an optical imaging apparatus, and an exposure apparatus with a narrow main beam and a small ripple were obtained.

FIG. 1 shows a normal microscope system or an immersion microscope system. High magnification objective lens FIG. 3 shows the parameters in the focal plane. The relationship between beam diameter and ripple rate is shown. Definition of pupil plane Focal plane beam radius, ripple rate Field distribution in the image plane for various values of a (in units of λm) Indicates the meaning of G (β) Pupil function 2D map Example of complex transmittance of pupil plane Example of complex transmittance of pupil plane Optical microscope Approximate characteristics of pupil filter Exposure equipment Approximate characteristics of pupil filter

A first aspect of the present invention relates to an optical microscope having a pupil filter.
This side pupil filter is, for example,
The complex transmittance of the pupil plane of the optical microscope is a function of the radius r,
It is a pupil filter whose amplitude is a monotonically decreasing function of r in the central part, has a maximum in the peripheral part, and makes its phase part constant.

に幅が

以上のHanning窓関数を乗じて得られる関数Ａ（ｒ）にＨａｎｋｅｌ変換

を行って得られる関数Ｆ（β）で表され，
位相部分は一定ならしめるである瞳フィルタであるか，又はその振幅部分を階段関数または滑らかな関数で近似した瞳フィルタであってもよい。 The pupil filter on this side is
The complex transmittance of the pupil plane is a function of the radius β,
The amplitude part is a function

Width is

Hankel transform into function A (r) obtained by multiplying the above Hanning window function

Is expressed by a function F (β) obtained by performing
The phase portion may be a pupil filter that is constant, or may be a pupil filter that approximates its amplitude portion with a step function or a smooth function.

  The second aspect of the present invention relates to an exposure apparatus having a pupil filter.

This side pupil filter is, for example,
After the excitation light incident on the pupil plane passes through the pupil filter, the amplitude part is a monotonically decreasing function of r in the central part, and has a maximum in the peripheral part.
This is a pupil filter that makes the phase portion constant.

に幅が

以上のHanning窓関数を乗じて得られる関数Ａ（ｒ）にHankel変換

を行って得られる関数Ｆ（β）で表され，
位相部分は一定ならしめる瞳フィルタ，あるいはその振幅部分を階段関数または滑らかな関数で近似した瞳フィルタであってもよい。 The pupil filter on this side has a complex transmittance of the pupil surface as a function of the radius β,
The amplitude part is a function

Width is

Hankel transform to function A (r) obtained by multiplying the above Hanning window function

Is expressed by a function F (β) obtained by performing
The phase portion may be a pupil filter that makes the phase constant, or a pupil filter that approximates the amplitude portion with a step function or a smooth function.

  When considering the distribution represented by the following equation in the plane where the beam is most strongly focused (beam waist) for any one orthogonal coordinate component of the electromagnetic field

k = 2π / λ _m = 2π _n sin θ / λ

  This field has the property that the maximum value of the transverse wave number of the Fourier component is k, that is, all the Fourier components pass through the pupil plane considered in the present application.

FIG. 6 shows the field distribution in the image plane for various values of a (in units of λ _m ). When a is small, the main beam is narrow (advantageous) and the ripple is large (disadvantageous). This is shown by the solid line in FIG.

反対にａがλ_ｍに対して無視できるときは

に近づく。主ビームが細く、周囲の円環の相対的高さが小さいことが望ましい。種々のビーム、種々の集光方式において、「与えられたリップル率（最大のリップル高さの主ビーム高さに対する比）において主ビームの極小半径の小ささ」を比較したところ今回導かれた解析解が優れていることがわかった。 The field represented by this function has a main beam having one peak and an annular ripple around the main beam that oscillates in a radial direction with a smaller amplitude. When α is larger than λ _m

Conversely, when a is negligible for λ _m

Get closer to. It is desirable that the main beam is thin and the relative height of the surrounding annulus is small. Comparison of “smallness of the minimum radius of the main beam at a given ripple ratio (ratio of the maximum ripple height to the main beam height)” for various beams and various condensing methods It turns out that the solution is excellent.

  In addition, when the electromagnetic field of the pupil plane corresponding to the cross section where the beam diameter is the thinnest in the field of the analytical solution, that is, the focal plane is analyzed, there is a very characteristic distribution with a sharp peak near the outermost periphery of the pupil plane I discovered that. That is, when β is smaller than k, where β is the radial wavenumber in the x, y plane

と表される。

It is expressed.

FIG. 7 shows the meaning of the above formula. Regardless of the size of A, it has a sharp peak on the outermost periphery of the pupil plane. The value of A is less (less 0.3λ _m) In substantially flat at the center, to no 0.35Ramuda _m when more 0.4Ramuda _m large has a smooth peaks in the central portion. Means for realizing the above expression with a high degree of approximation on the pupil plane will be described in an embodiment.

In the optical microscope, immersion type (water, refractive index 1.33), maximum acceptance angle 70 degrees, NA = 1.25
This objective lens is used. As shown in FIG. 11, a pupil filter is inserted into the pupil plane conjugate with the objective exit pupil. The transmittance at each radius of the pupil surface of the pupil filter is determined as follows:

The imaginary shift α is selected to be 0.35λ _m from the thinness of the main beam and the small ripple. In order to obtain an effect equivalent to the sharp pupil function of Equation (5) with a high degree of approximation, the following design was performed.

Hanning window function H (r) = (1 + cosπr / W) / 2 (r <W) in order to approximate the focal plane function (1) well in the vicinity of the central axis (z-axis).
= 0 (r> W)

Is multiplied by (1) to make a Hankel transform, which is used as an approximate function of the pupil function. FIG. 8 illustrates the pupil function thus obtained (2nW = 200, 100, 40, 20, 10 in the order of a solid line that changes sharply, an alternate long and short dash line, a broken line, and a solid line that changes gently. Using the approximate function, the focal plane function slightly changes from the ideal formula (1): The transition from the ideal curve (solid line) in the two-dimensional map in Fig. 9 is a series of small circles (slightly square circles). Indicated.
Thus, it can be seen that even if the sharp pupil function is replaced with a more gentle pupil function, the change in the focal plane function is sufficiently small.

  Further, if a gentle pupil function is replaced with a piecewise uniform pupil function, a pupil filter can be easily created by a known method. FIGS. 10-1 and 10-2 are examples. The image function by this pupil filter matches the circle in FIG. 9 (third from the left) on the drawing.

  Such a pupil filter is created by a two-dimensional array of fine black spots on the glass surface with density controlled for each location. It is also created by a computer generated hologram.

As a result, a high resolution microscope with a narrow main beam and low ripple can be realized.
Although the pupil filter can be installed on the pupil plane of the objective lens, it is advantageous to select and mount a place that can be installed more easily and inexpensively, such as the pupil plane of an eyepiece.

Exposure system The design policy is almost the same as that of a microscope. However, in order to use a thinner beam, the imaginary shift α was selected to be 0.31λ _m . Hanning as in Example 1 to make it easier to design pupil filters
A window function was used (2nW = 200, 100, 40, 20, 10). The change of the focal plane function is as shown in the two-dimensional map of FIG. 9, and the change is sufficiently small for 40 or more. In the exposure apparatus shown in FIG. 13, a pupil filter is inserted at a position conjugate with the objective pupil plane. Considering the convenience of creation, a pupil filter is used in which the curve for 2nW = 100 is approximated by the polygonal line (piecewise uniform function) in FIG. The nature of the image function matches the third on the left of the triangular point in FIG.
Such pupil filters are made by well-known computer hologram methods.

  The present invention can be used in the field of optical instruments.

Claims (4)

An optical microscope having a pupil filter,
The pupil filter is
The complex transmittance of the pupil plane of the optical microscope is a function of the radius r,
An optical microscope whose amplitude part is a monotonically decreasing function of r in the central part, has a maximum in the peripheral part, and makes its phase part constant. An optical microscope having a pupil filter,
The pupil filter is
The complex transmittance of the pupil plane is a function of the radius β,
The amplitude part is a function

Width is

Hankel transform into function A (r) obtained by multiplying the above Hanning window function

Is expressed by a function F (β) obtained by performing
An optical microscope having a pupil filter whose phase portion is constant or whose amplitude portion is approximated by a step function or a smooth function. An exposure apparatus having a pupil filter,
The pupil filter is
After the excitation light incident on the pupil plane passes through the pupil filter, the amplitude part is a monotonically decreasing function of r in the central part, and has a maximum in the peripheral part.
It is a pupil filter that makes the phase part constant,
Exposure device. An exposure apparatus having a pupil filter,
The pupil filter is
The complex transmittance of the pupil plane is a function of the radius β,
The amplitude part is a function

Width is

Hankel transform to function A (r) obtained by multiplying the above Hanning window function

Is expressed by a function F (β) obtained by performing
The phase filter is a pupil filter that makes the phase constant, or the pupil filter that approximates the amplitude part with a step function or a smooth function.
Exposure device.

Images