JP3866849B2 - Ellipsometer - Google Patents

Description

【００３８】
位相差板についても同様にミュラーマトリックスＭｒが求められる。偏光方向Ｐ_m の場合のストークスベクトルＳ_i はＩ₀ （１，cos Ｐ_m ，sin Ｐ_m ，０）^T である。したがって検光子に入射する光のストークスベクトルＳo は下記式で求められる。
ケース１の場合；Ｓo ＝ＭｃＳ_i
ケース２の場合：Ｓo ＝ＭｃＭｒＳ_i
結論として４種類（あるいはそれ以上）のＳo 、つまりＳom（ｍ＝１，２，３，４，‥‥）が求められた。下記式で検出器側のストークスベクトルを求めることができる。
【００３９】
Ｓom＝Ｉ₀ Ｍｃ（Ψ_c ，Δ_c ）Ｍ（Ｐ_m ）（１，１，０，０）^T
Ｍ（Ｐ_m ）は、座標回転のミュラーマトリックスである。このようにして検出系に入射する光の４種（あるいはそれ以上）の偏光状態のストークスベクトルＳom＝（Ｓ0 ，Ｓ1 ，Ｓ2 ，Ｓ3 ）^T が計算で求められる。
４種（又はそれ以上）の偏光状態Ｓomの各々で４つの光束の測定強度をＩｍ＝（Ｉ₁ ｍ，Ｉ₂ ｍ，Ｉ₃ ｍ，Ｉ₄ ｍ）^T とする。
【００４０】
Ｉｍ＝Ｃ Ｓom
４種類（またはそれ以上）の設定条件の結果は連立方程式になりｍ行４列の変換マトリクスＣ（装置定数）が求められる。ｍが４以上の場合にはＣは最小二乗法で求めることが出来る。Ｃは各波長ごとに求めることが必要である。
これにより装置のキャリブレーションは完了する。
(4) ストークスメータの場合の測定
未知試料を所定の場所に装着し、偏光子の角度Ｐ_m と位相差板の方位角度を設定してある偏光を試料に照射し、４光束の強度Ｉを測定すると、
Ｉ＝ＣＳo
未知試料のミュラーマトリックスをＭ_S とすると、
Ｉ＝Ｃ Ｍ_s ＭｒＳ_i
ここでＩは測定値、Ｃは装置定数、Ｍｒはすでに求められているので、試料のミュラーマトリックスのみが未知数となる。したがって、ストークスメータの測定目的にあわせた特定の偏光を試料に照射すると上記式から、ミュラーマトリックスの要素からなる目的としたパラメータを得ることが可能である。
【００４１】
【発明の効果】
以上のように本発明の偏光解析装置によれば、従来ならば偏光素子を回転させて複数回測定するところを、同時に測定することができるので、測定時間が短くできる。したがって、高速測定が可能となり、特に試料が経時変化を起こす場合にも対応できる。
【００４２】
また、偏光素子を回転させる機構が必要ないので、装置の構造が簡単になるとともに、測定精度もよくなる。
【図面の簡単な説明】
【図１】本発明に係る試料透過型の分光偏光解析装置の構成を示す図である。
【図２】光束分離集光光学系の詳細を示す図である。
【図３】２次元検出器の検出面におけるスペクトルの結像位置を説明する図である。
【図４】本発明に係る試料反射型の分光偏光解析装置の構成を示す図である。
【図５】光束分離集光光学系の詳細を示す図である。
【図６】従来の試料透過型の偏光解析装置を示す図である。
【図７】従来の試料反射型の偏光解析装置を示す図である。
【符号の説明】
２ 白色光源
３ 集光光学系
４ 偏光子
５ 試料
６ａ〜６ｄ 検光子
７ 光束分離集光光学系
８ａ〜８ｄ 光ファイバ
９ ２次元分光器
１０ ２次元検出器
１１ 演算処理装置
１２ａ〜１２ｄ 検光子
１３ 光束分離集光光学系[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a polarization analyzer that analyzes a polarization characteristic of a sample by irradiating the sample with polarized light and measuring a polarization state of transmitted light or reflected light.
[0002]
[Prior art]
An ellipsometer, which is a type of ellipsometer, is a device that irradiates a sample with light of a certain polarization, obtains ellipso parameters from the polarization state of the transmitted or reflected light, and analyzes the polarization characteristics of the sample.
Ratio tanΨ and phase difference Δ (or Ψ and Δ) between the reflectance of the s-wave, which is a component perpendicular to the incident surface (the surface through which the incident light beam and the reflected light beam pass), and the p-wave reflectance, which is a parallel component. ) Is called an ellipso parameter, and the ellipso parameter can be used to measure the film thickness and the physical properties of the thin film material (see Japanese Patent Laid-Open Nos. 5-133811 and 5-142141).
[0003]
There is also known a Stokes meter that irradiates a sample with light of a certain polarization and obtains a Stokes parameter from the polarization state of transmitted light or reflected light.
In this specification, the ellipsometer and the Stokes meter are collectively referred to as “polarization analyzer”.
In particular, a spectroscopic ellipsometer capable of measuring the wavelength dependence of parameters is optimal for analyzing complex films such as multilayer films.
[0004]
The configuration of a conventional ellipsometer is shown in FIGS.
FIG. 6 shows a sample transmission type ellipsometer. In this sample transmission type ellipsometer, the light from the white light source 2 is applied to the sample 5 through the condensing optical system 3, the rotating polarizer 4, and the rotatable phase difference plate 21, and the light transmitted through the sample 5 is rotated by the rotating analyzer. 14, it is put into the optical fiber 8 through the condensing optical system 15. The light transmitted through the optical fiber 8 enters the spectroscope 16, and its spectrum is detected by the detector 17. The detected spectrum signal is input to the arithmetic processing unit 18 where the ellipsometric calculation is performed.
[0005]
FIG. 7 shows a sample reflection type ellipsometer. The difference between this sample reflection type ellipsometer and the sample transmission type ellipsometer is that the light reflected from the sample 5 is measured, and the rest is the same as the sample transmission type ellipsometer.
In addition, when using as an ellipsometer, the phase difference plate 21 between the rotating polarizer 4 and the sample 5 is not required.
[0006]
[Problems to be solved by the invention]
By the way, the conventional ellipsometer needs to rotate the polarizer and the analyzer (half rotation or more) when measuring, and the rotation requires a certain time or more. Therefore, high-speed measurement is difficult, and analysis cannot be performed particularly when the sample changes with time. In addition, since a mechanism for rotating the polarizer and analyzer is also required, the structure becomes complicated, the optical axis of the optical element is not adjusted accurately, the processing accuracy of the optical element is insufficient, and the linearity of the central axis of the optical element is insufficient. There is a possibility that the optical axis may move during measurement due to a defect or the like, which has been a cause of poor measurement accuracy.
[0007]
Therefore, realization of an ellipsometer capable of measuring in a short time and having a simple structure is required.
[0008]
[Means for Solving the Problems]
The ellipsometer according to the present invention passes light that spreads in different directions from a sample, respectively, and passes through each of three or more types of analyzers in combination with three or more types of analyzers. The beam separation and condensing optical system for reflecting the reflected light by a mirror and condensing it at different focal points, the spectroscope that respectively divides the plural lights collected at the different focal points, and detecting the plural spectra a detector capable of, Ru der having an arithmetic processing unit for performing ellipsometry calculation.
[0009]
The ellipsometric calculation is performed using the device constant C or the gain factor G.
According to the ellipsometer of the above-described configuration, since the analyzing means combining three or more kinds of analyzers is used for measurement, the light that has passed through each analyzer can be measured simultaneously, and thus the measurement time is shortened. A rotating part is also unnecessary.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
FIG. 1 shows the configuration of a sample transmission type spectroscopic ellipsometer according to the present invention.
This sample transmission-type spectroscopic ellipsometer analyzes light from a white light source 2 with a condensing optical system 3, a linear polarizer (hereinafter simply referred to as “polarizer”) 4, and a rotatable phase difference plate (hereinafter simply referred to as “phase difference plate”). The light that has passed through the sample 5 through the sample 5 and passed through the sample 5 is passed through a plurality (four in this example) of linear analyzers (hereinafter simply referred to as “analyzers”) 6a to 6d to separate the light beam. 7, a plurality (four in this example) of optical fibers 8a to 8d are inserted. The light transmitted through each of the optical fibers 8a to 8d enters a two-dimensional spectrometer 9 such as a CCD, and the spectrum thereof is detected independently by the two-dimensional detector 10. The detected spectrum signal is input to the arithmetic processing unit 11 where the ellipsometric calculation is performed.
[0011]
The polarizer 4 needs to be a rotating type when used as a Stokes meter for measurement and calibration and as an ellipsometer for calibration, but cannot be rotated when used as an ellipsometer for measurement. But you can. The phase difference plate 20 is necessary when used as a Stokes meter for measurement and calibration, and is not absolutely necessary when used as an ellipsometer for measurement and calibration. Further, the phase difference plate 20 may be positioned after the sample 5 as long as it is between the polarizer 4 and the analyzers 6a to 6d.
[0012]
The white light source 2 only needs to have a wavelength range that allows the spectrum of the transmitted light of the sample 5 to be measured within a necessary range. For example, a lamp using a rare gas such as an Xe lamp, a halogen lamp, a mercury lamp, etc. Can be used.
The condensing optical system 3 is a condensing optical system configured using a lens or a mirror.
[0013]
The analyzers 6a to 6d are a combination of a plurality of analyzers 6a to 6d having different polarization directions. In the example of FIG. 1, the squares are arranged to be vertically and horizontally divided into four. The reason why the analyzers 6a to 6d are arranged in this way is to make the distance from the optical axis D shown in FIG. 1 uniform and minimum for each analyzer 6a to 6d.
[0014]
The polarization directions of the plurality of analyzers 6a to 6d may be any angle that does not overlap at an arbitrary angle from 0 ° to 180 ° (where 180 ° or more is an angle obtained by subtracting an integral multiple of 180 ° from the angle). To do). For example, the angle may be shifted by 0 °, 45 °, 90 °, 135 °, and 45 °, respectively, with an arbitrary angle as a reference.
Although the number of analyzers 6a to 6d is four, the ellipso parameters can be measured with at least three analyzers.
[0015]
As will be described later, the light beam separating and condensing optical system 7 is an optical system that collects the light exiting from the sample 5 and passing through the plurality of analyzers 6a to 6d at the incident ends 81 of the plurality of optical fibers 8a to 8d, respectively. is there.
The two-dimensional spectrometer 9 is a spectrometer that separates each light beam emitted from the plurality of optical fibers 8 a to 8 d and forms an image as a spectrum at different detection positions of the two-dimensional detector 10.
[0016]
In the above spectroscopic ellipsometer, the condensing point (focal point) of the light emitted from the white light source 2 is the observation point of the sample 5, the incident end 81 of the optical fibers 8a to 8d, the slit of the two-dimensional spectrometer 9, and 2 It will be located on the detection surface of the dimension detector 10, respectively.
In FIG. 1, the polarizer 4 and the analyzers 6a to 6d may be interchanged. The condensing optical system 3 and the light beam separating condensing optical system 7 may be interchanged. Further, the optical fibers 8a to 8d may be omitted and the light may be directly input to the two-dimensional spectrometer 9.
[0017]
Details of the light beam separating and condensing optical system 7 are shown in FIG. In FIG. 2, the light that has passed through the sample 5 and has passed through the analyzers 6 a to 6 d is reflected by the four plane mirrors 71 a to 71 d, further reflected by the concave mirror 72, and condensed at the four optical fiber end faces 81, respectively. Is done. The plane mirrors 71a to 71d are not placed on the same plane, but have different installation angles. This is because the light emitted from the sample 5 is focused on the four optical fiber end faces 81 separately.
[0018]
In the configuration of FIG. 2, the optical fiber end faces 81 are arranged vertically and horizontally, but the present invention is not limited to this, and four optical fiber end faces 81 may be arranged in a straight line. The plane mirrors 71a to 71d and the concave mirror 72 may be interchanged. Further, it is possible to make the optical system compact by inserting a plane mirror in the middle. It is also possible to omit the concave mirror 72 by using the plane mirrors 71a to 71d as concave mirrors.
[0019]
It is also possible to place a condensing optical system or a beam expanding optical system using a lens, a mirror, or a combination thereof between the sample 5 and the analyzers 6a to 6d. In this way, the light beam can be guided to the detector even when the arrangement of the sample 5 and the detection optical system is limited. FIG. 3 is a view for explaining the spectral imaging position on the detection surface 10 a of the two-dimensional detector 10. The light beams emitted from the plurality of optical fibers 8a to 8d are respectively dispersed and imaged separately on the detection surface 10a of the two-dimensional detector 10 in the vertical direction of FIG. The horizontal direction in FIG. 3 corresponds to the wavelength. If a channel for background correction is set in the region between the spectra, there is an advantage that a spectrum from which the stray light of the spectrometer is removed can be obtained by subtracting from the spectrum of the light beam.
[0020]
FIG. 4 shows a configuration of a sample reflection type spectroscopic ellipsometer according to the present invention.
Only the difference from the sample transmission type spectroscopic ellipsometer shown in FIG. 1 will be described. Light emitted from the white light source 2 strikes the sample 5 obliquely and the reflected light is detected, and a plurality of analyzers 12a. ˜12d are arranged in a horizontal row, and the configuration of the light beam separating and condensing optical system 13 is different.
[0021]
The reason why the plurality of analyzers 12a to 12d are arranged in a horizontal row is to make the angle θ formed by the reflected light with the perpendicular B shown in FIG. 4 uniform for each of the analyzers 12a to 12d.
Details of the light beam separating and condensing optical system 13 are shown in FIG. In FIG. 5, the light reflected from the sample 5 and transmitted through the analyzers 12a to 12d is reflected by the four plane mirrors 13a to 13d, further reflected by the concave mirror 131, and condensed at the four optical fiber end surfaces 81, respectively. Is done. The plane mirrors 13a to 13d are not placed on the same plane, but have different installation angles. This is because the light that has come out of the sample 5 and transmitted through the analyzers 12a to 12d is focused separately at the optical fiber end face 81. In the configuration of FIG. 5, the optical fiber end faces 81 are arranged in a straight line, but the present invention is not limited to this, and the optical fiber end faces 81 may be arranged vertically and horizontally. The plane mirrors 13a to 13d and the concave mirror 131 may be interchanged. Further, it is possible to make the optical system compact by inserting a plane mirror in the middle. It is also possible to omit the concave mirror 131 by using the plane mirrors 13a to 13d as concave mirrors. In FIG. 4, the polarizer 4 and the analyzers 12a to 12d may be interchanged. It is also possible to place a condensing optical system or a beam expanding optical system using a lens, a mirror, or a combination thereof between the sample 5 and the analyzers 12a to 12d.
[0022]
Calibration and measurement methods in the above-described sample transmission type and sample reflection type spectroscopic ellipsometers will be described. However, the following method can be modified and is not limited to this method, and other procedures may be adopted.
(1) Equipment calibration for ellipsoparameter measurement Sample transmission type is without sample, and sample reflection type is a calibration sample such as quartz plate or glass plate, and ellipsoparameter is one of the analyzers. (Or a plurality) is measured by the rotating polarizer method.
[0023]
First, the polarization characteristics of the calibration sample are measured.
The detection intensity I is a function of the azimuth angle P of the polarizer 4.
I = I ₀ (1 + αcos 2P + βsin 2P)
The intensity I is obtained with respect to P by the rotating polarizer method, and α and β are obtained from the above equations. The function of the ellipso parameter of the calibration sample is obtained by the following equation.
[0024]
tan Ψ _c exp (j Δ _c ) = tan A / {[β ± j (1-α ² −β ² ) ^1/2 ] / (1 + α)}
Here, A is the azimuth angle of the analyzer. Since the above equation is a complex number, the ellipso parameter tan Ψ _c and the phase difference Δ _c of the calibration sample can be obtained from this equation.
[0025]
At this time, if the phase difference plate 20 is rotated simultaneously with the polarizer 4, an ellipso parameter of the calibration sample is obtained, and if the phase difference plate 20 is fixed, an ellipso parameter to which the characteristics of the phase difference plate 20 are added is obtained.
When using the intensity of a plurality of light beams, the polarization state χ ₀ , tan Ψ _c or cos Δ _c obtained from the intensity of each light beam is averaged.
[0026]
From the obtained tan Ψ _c and Δ _c , the apparatus constant, that is, the gain factor of the detection system for each of the four light beams is obtained in the following order.
When the azimuth angle of the polarizer 4 is set to P _m , the polarization characteristic of the light entering the detection optical system when the calibration sample is placed is given by the following equation.
χ ₀ = tanP _m / {tanΨ _c exp (jΔ _c )}
The detected light quantity of each light beam is expressed by the following equation.
[0027]
Ia = GaI ₀ (1 + α′cos 2Aa + β′sin 2Aa)
Ib = GbI ₀ (1 + α′cos 2Ab + β′sin 2Ab)
Ic = GcI ₀ (1 + α′cos 2Ac + β′sin 2Ac)
Id = GdI ₀ (1 + α′cos 2Ad + β′sin 2Ad)
Here, Ga to Gd are gain factors (equipment constants) that arise from the fact that the light receiving efficiency varies depending on the light flux, and Aa to Ad are azimuth angles of the respective analyzers.
[0028]
When there are three optical systems (three light beams) to be separated, Gc is obtained from the three gain factors Ga. α ′ and β ′ are given by the following equations.
α ′ = (1−χ ₀ χ ₀ ^* ) / (1 + χ ₀ χ ₀ ^* )
β ′ = (1 + α ′) Re (χ ₀ )
Where χ ^* is the conjugate complex number of χ. Re represents the real part. Accordingly, gain factors GaI _{0 to} GdI ₀ are obtained by calculation. When a certain gain factor, for example, Ga is 1, gain factors that are all device constants can be obtained. Since these gain factors are wavelength-dependent, it is necessary to obtain them for each wavelength.
(2) Measurement of Ellipsoparameter of Sample Next, the measurement sample is mounted and the intensities Ia to Id of the respective light beams are measured. In the following equation, the unknowns are I ₀ , α ″, β ″, and there are many equations, so the unknowns (α ″, β ″) are obtained by the least square method from the following equation.
[0029]
Ia / Ga = I ₀ (1 + α ″ cos 2Aa + β ″ sin 2Aa)
Ib / Gb = I ₀ (1 + α ″ cos 2Ab + β ″ sin 2Ab)
Ic / Gc = I ₀ (1 + α ″ cos 2Ac + β ″ sin 2Ac)
Id / Gd = I ₀ (1 + α ″ cos 2Ad + β ″ sin 2Ad)
Using the obtained α ″ and β ″, the ellipso parameters Ψ and Δ of the sample in the case of reflection measurement are obtained from the following equation.
[0030]
tanΨ exp (jΔ) = tanP _m / {[β ″ ± j (1-α ″ ² −β ″ ² ) ^1/2 ] / (1 + α ″)}
When the optical axis direction of a sample having phase difference characteristics in transmission measurement is set to the reference direction of the azimuth angle, the ratio of the transmittance of polarized light (corresponding to Ψ) and the phase difference (in Δ) Equivalent) is obtained from the same equation.
[0031]
When the number of light beams is three, the above equation has three unknown parameters, and since there are three equations, a solution can be obtained. Therefore, the ellipso parameters Ψ and Δ can be obtained even in the case of three light beams.
(3) Equipment calibration in the case of a Stokes meter In the case of the sample transmission type, the sample is rotated, and in the case of the sample reflection type, rotation is performed using a calibration sample such as a quartz plate, a glass plate or a silicon substrate with a thin oxide film. The polarization characteristics of the sample are measured by the polarizer method.
[0032]
(1) The azimuth angle of the phase difference plate placed between the sample 4 and the sample 5 is simultaneously rotated so that it is the same as one of the azimuth angles of the analyzer, and one (or more) luminous flux signal is generated. Detect and measure the characteristics of the calibration sample by the rotating polarizer method. In this case, the light applied to the sample is linearly polarized light. The detected intensity I is a function of the azimuth angle P of the polarizer.
[0033]
I = I ₀ (1 + αcos 2P + βsin 2P)
Using the polarizer, a graph of intensity I is obtained by the rotating polarizer method, and α and β are obtained from the above equations. The polarization characteristic of the calibration sample is obtained by the following formula.
tan Ψ _c exp (j Δ _c ) = tan A / {[β ± j (1-α ² −β ² ) ^1/2 ] / (1 + α)}
Here, A is the azimuth angle of the analyzer of the selected light flux. However, the sign of Δ _c (that is, ± in the above equation) is estimated from the characteristics of the mounted retardation plate and the calibration sample.
[0034]
(2) Next, the azimuth angle of the phase difference plate placed between the sample 4 and the sample 5 is fixed to 0 degree or 90 degrees (any azimuth angle is also possible). Using one or more light flux signals, the polarization characteristic of a calibration jig including a retardation plate is measured by a rotating polarizer method. The characteristics of the phase difference plate are measured from the Fourier coefficients α ′ and β ′ of the intensity at this time.
[0035]
tan ψ _rc exp (jΔ _rc ) = 1 / {tan ψ _c exp (j Δ _c )} × tan A / {[β ′ ± j (1-α ′ ² −β ′ ² ) ^1/2 ] / (1 + α ′ )}
Here, the values obtained in (3) (1) are used for Ψ _c and Δ _c .
▲ 3 ▼ then set the azimuth P _m of the polarizer 4 four or more (m = 1,2,3,4, ‥‥). The azimuth angle of the phase difference plate is assumed to be the case 1 when rotating simultaneously with the polarizer 4, and the case 2 being fixed to 0 degree or 90 degrees. The four types of measurements require at least one case 1 and one case 2 to be included.
[0036]
Since the polarization characteristic of the calibration sample is known, the following equation is given.
χ _i / χ ₀ = {tan Ψ _c exp (j Δ _c )}
Therefore, a Mueller matrix Mc (Ψ _c , Δ _c ) corresponding to this equation is obtained. This matrix is a Mueller matrix of a sample having ellipso parameters Ψ and Δ, and is given by the following equation.
[0037]
[Expression 1]

[0038]
Similarly, the Mueller matrix Mr is obtained for the retardation plate. The Stokes vector S _{i in} the case of the polarization direction P _m is I ₀ (1, cos P _m , sin P _m , 0) ^T. Therefore, the Stokes vector So of the light incident on the analyzer can be obtained by the following equation.
Case 1; So = McS _i
Case 2: So = McMrS _i
In conclusion, four types (or more) of So, that is, Som (m = 1, 2, 3, 4,...) Were obtained. The detector-side Stokes vector can be obtained by the following equation.
[0039]
_{Som = I 0 Mc (Ψ c} , Δ c) M (P m) (1,1,0,0) T
M (P _m ) is a Mueller matrix of coordinate rotation. In this way, the Stokes vectors Som = (S0, S1, S2, S3) ^T of the four types (or more) of polarization states of the light incident on the detection system are obtained by calculation.
The measured intensity of the four light beams in each of the four (or more) polarization states Som is Im = (I ₁ m, I ₂ m, I ₃ m, I ₄ m) ^T.
[0040]
Im = C Som
The result of the four types (or more) of the setting conditions is a simultaneous equation, and a m-by-4 conversion matrix C (device constant) is obtained. When m is 4 or more, C can be obtained by the least square method. C must be obtained for each wavelength.
This completes the calibration of the apparatus.
(4) In the case of a Stokes meter, an unknown sample is mounted at a predetermined location, and the sample is irradiated with polarized light in which the angle P _{m of the} polarizer and the azimuth angle of the retardation plate are set, and the intensity I of the four beams is obtained. When measuring
I = CSo
_Let M _{S be} the Mueller matrix of an unknown sample.
I = C M _s MrS _i
Here, since I is a measured value, C is an apparatus constant, and Mr is already obtained, only the Mueller matrix of the sample is an unknown. Therefore, when the sample is irradiated with a specific polarized light in accordance with the measurement purpose of the Stokes meter, it is possible to obtain a target parameter composed of elements of the Mueller matrix from the above formula.
[0041]
【The invention's effect】
As described above, according to the ellipsometry apparatus of the present invention, the measurement time can be shortened because it is possible to measure a plurality of times by rotating the polarizing element conventionally. Therefore, high-speed measurement is possible, and in particular, it is possible to cope with a case where the sample changes with time.
[0042]
Further, since a mechanism for rotating the polarizing element is not necessary, the structure of the apparatus is simplified and the measurement accuracy is improved.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of a sample transmission type spectroscopic ellipsometer according to the present invention.
FIG. 2 is a diagram showing details of a light beam separating and condensing optical system.
FIG. 3 is a diagram for explaining an imaging position of a spectrum on a detection surface of a two-dimensional detector.
FIG. 4 is a diagram showing a configuration of a sample reflection type spectroscopic ellipsometer according to the present invention.
FIG. 5 is a diagram showing details of a light beam separating and condensing optical system.
FIG. 6 is a diagram showing a conventional sample transmission type ellipsometer.
FIG. 7 is a diagram showing a conventional sample reflection type ellipsometer.
[Explanation of symbols]
2 White light source 3 Condensing optical system 4 Polarizer 5 Samples 6a to 6d Analyzer 7 Light beam separating and condensing optical systems 8a to 8d Optical fiber 9 Two-dimensional spectrometer 10 Two-dimensional detector 11 Arithmetic processing units 12a to 12d Analyzer 13 Beam separation and condensing optical system

Claims (4)

In a polarization analyzer that separates sample light into a plurality of different polarization components and measures each polarization component,
Analyzing means combining three or more types of analyzers, each of which passes light spreading in different directions from the sample,
A light beam separating and condensing optical system for reflecting the light that has passed through each of the three or more analyzers of the light analyzing means by a mirror and condensing them at different focal points;
A spectroscope that separates each of the plurality of lights collected at the separate focal points;
A detector capable of detecting multiple spectra;
An ellipsometry apparatus comprising: an arithmetic processing unit that performs ellipsometry computation.   The ellipsometer according to claim 1, further comprising a condensing optical system for condensing the light from the light source and entering the sample. The ellipsometer according to claim 2 , wherein a polarizer is disposed between the condensing optical system and the sample. The ellipsometer according to claim 2 , wherein the light source is a white light source.

Images