CN104281011B - A kind of detection method of high-NA imaging system Polarization aberration - Google Patents

Abstract

The invention discloses a kind of detection method of high-NA imaging system Polarization aberration, adopt the physics pupil characteristic manner of Polarization aberration, and different Ze Nike decomposition methods is carried out to different aberration compositions, eventually through the test map migration of different test mask, focal plane translation and feature size error, the full detail of imaging system Polarization aberration can be obtained in real time, exactly, be applicable to the measurement for any high NA imaging system Polarization aberration; The method can detect the full detail that the Polarization aberration of the whole pupil of imaging system comprises, and does not need to carry out extra measurement, has the advantage that accuracy of detection is high, detection speed is fast; Adopt test mask conventional in imaging system as test badge, do not need to design special detecting element, there is the advantage that structure is simple, testing cost is low.

Description

A kind of detection method of high-NA imaging system Polarization aberration

Technical field

The present invention relates to high-resolution imaging system aberration detection technique field, particularly relate to a kind of detection method of high-NA imaging system Polarization aberration.

Background technology

Current high resolution microscope, telescope and the etching system for the preparation of VLSI (very large scale integrated circuit), all adopt high-NA (NA) imaging technique.In order to improve imaging system resolving power, need to utilize immersion (NA > 1) imaging system.Research shows, under high NA and polarization illumination condition, the Polarization aberration of imaging system has become the key factor affecting image quality.The Polarization aberration detecting imaging system is quickly and accurately the important prerequisite effectively controlling Polarization aberration, significant for raising image quality.

Patents (Chinese patent CN103197512A) discloses a kind of method obtaining lithographic projection system Polarization aberration.Adopting line orientations in the method along the alternating phase-shift mask of X and Y-axis as resolution chart, by measuring focal plane translation and the map migration of resolution chart, obtaining the Polarization aberration of lithographic projection system, Polarization aberration wherein adopts Pauli pupil to characterize.But the method can only detect the phase term of Pauli pupil Section 1 and the imaginary part of Section 2 in theory, the full detail of optical projection system Polarization aberration can not be detected, thus can not be used for the measurement of actual lithographic projection system Polarization aberration, in this invention, also not provide the Polarization aberration measurement result adopting the method.

Summary of the invention

In view of this, the invention provides a kind of detection method of high-NA imaging system Polarization aberration, the full detail of imaging system Polarization aberration can be obtained exactly, be applicable to the measurement of any high NA imaging system Polarization aberration.

In order to solve the problems of the technologies described above, the present invention is achieved in that

The detection method of a kind of high-NA imaging system Polarization aberration of the present invention, comprises the steps:

Step 1, the test mask arranged in high-NA imaging system: described test mask comprises three groups of test badges: first group is the binary mask of optical grating construction, second group is the alternating phase-shift mask of phase shift 90 °, and the 3rd group is the alternating phase-shift mask of phase shift 180 °; Three groups of test masks are intensive linear, and the dutycycle of opaque lines and transparent lines is 1: 1;

Step 2, the polarization illumination mode of the lighting source in high-NA imaging system is arranged linearly polarized light at 45 °, and adopt quadrupole illuminating mode, obtain the aerial image distribution at the desirable focal plane place of first group of test mask, and then obtain the phase-shift phase of single order intensity distributions at desirable focal plane place then phase-shift phase is set up and the relational expression between the orientation zernike coefficient of the zernike coefficient of the strange item of wave aberration and the strange item of two-way delay:

Wherein, for the zernike coefficient of the strange item of wave aberration, with for the orientation zernike coefficient of the strange item of two-way delay, for the sensitivity coefficient of the strange item correspondence of wave aberration, with sensitivity coefficient for the strange item correspondence of two-way delay:

$\begin{array}{c}{S}_{\mathrm{no}}^W=\mathrm{\π}\frac{{\mathrm{\Θ}}_{\mathrm{no}}}{\∫\∫Q(f,g)\·\mathrm{Bdfdg}}\\ S_{\mathrm{no}}^{\mathrm{ret}}=\frac{S_{1\_o}^{\mathrm{ret}}+{S_{1\_o}^{\mathrm{ret}}}^{\′}}{2\∫\∫Q(f,g)\·\mathrm{Bdfdg}}\\ S_{-\mathrm{no}}^{\mathrm{ret}}=\frac{S_{2\_o}^{\mathrm{ret}}+{S_{2\_o}^{\mathrm{ret}}}^{\′}}{2\∫\∫Q(f,g)\·\mathrm{Bdfdg}}\end{array}---\left(2\right)$

Q (f, g) represents the intensity of lighting source; (f, g) represents pupil plane coordinate; f _-1and f ₁under representing normal incidence respectively, the negative one-level of test mask diffraction and the coordinate of positive one-level frequency spectrum point in x-axis;

$S_{1\_o}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\mathrm{dfdg},$

$S_{2\_o}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\mathrm{dfdg},$

${S_{1\_o}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\mathrm{dfdg},$

${S_{2\_o}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\mathrm{dfdg},$

$R_{11}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g),$

$R_{12}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f+f_1,g)+R_{\mathrm{no}}(f+f_1,g),$

$R_{13}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g),$

$R_{14}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f+f_1,g)-R_{\mathrm{no}}(f+f_1,g),$

$R_{23}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}+1}(f+f_1,g),$

$R_{22}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g),$

$R_{23}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}+1}(f+f_1,g),$

$R_{24}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)-R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g),$

${\mathrm{\Θ}}_{\mathrm{no}}=\∫\∫Q(f,g)\·B\·[R_{\mathrm{no}}\left(f,g\right)-R_{\mathrm{no}}(f+f_{-1},g)-R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}}(f,g)]\mathrm{dfdg}.$

Wherein, R _no(f, g) represents the strange item in zernike polynomial; No value is 2,3,7,8,10,11,14,15,19,20,23,24,26,27,30,31,34,35;

B＝B ₁+B ₂+B ₃+B ₄

$B_1=0.5\·[T_{\mathrm{xx}}\·T_{\mathrm{xx}}^{*}+T_{\mathrm{yx}}\·T_{\mathrm{yx}}^{*}+T_{\mathrm{zx}}{T}_{\mathrm{zx}}^{*}]\·H\·H^{*},$

$B_2=0.5\·[T_{\mathrm{xy}}\·T_{\mathrm{xy}}^{*}+T_{\mathrm{yy}}\·T_{\mathrm{yy}}^{*}+T_{\mathrm{zy}}{T}_{\mathrm{zy}}^{*}]\·H\·H^{*},$

$B_3=0.5\·[T_{\mathrm{xx}}\·T_{\mathrm{xy}}^{*}+T_{\mathrm{yx}}\·T_{\mathrm{yy}}^{*}+T_{\mathrm{zx}}{T}_{\mathrm{zy}}^{*}]\·H\·H^{*},$

$B_4=0.5\·[T_{\mathrm{xy}}\·T_{\mathrm{xx}}^{*}+T_{\mathrm{yy}}\·T_{\mathrm{yx}}^{*}+T_{\mathrm{zy}}{T}_{\mathrm{zx}}^{*}]\·H\·H^{*};$

T _xx, T _xy, T _yx, T _yy, T _zxand T _zybe respectively component in the pupil transformation matrix T of imaging system; H is the pupil function of imaging system, and * represents complex conjugate;

Change the coherence factor of lighting source and the cycle of first group of test mask, carry out the measurement under M different test condition, obtain phase-shift phase then the sensitivity coefficient of the strange item correspondence of wave aberration under the different test condition of M group is obtained respectively according to formula (2) and the sensitivity coefficient of the strange item correspondence of two-way delay with matrix composed as follows:

$S_1=\left[\begin{array}{ccc}{S}_{\mathrm{no}}^{W1}& S_{\mathrm{no}}^{\mathrm{ret}1}& S_{-\mathrm{no}}^{\mathrm{ret}1}\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ S_{\mathrm{no}}^{\mathrm{WM}}& S_{\mathrm{no}}^{\mathrm{retM}}& S_{-\mathrm{no}}^{\mathrm{retM}}\end{array}\right]$

Wherein, the value of M is the multiple of 54;

By the matrix S obtained under test conditions different for M group ₁be updated in formula (1), obtain the solution of the zernike coefficient of the strange item of M/54 group wave aberration and the orientation zernike coefficient of the strange item of two-way delay;

Step 3, the polarization illumination mode of the lighting source in high-NA imaging system is arranged linearly polarized light at 45 °, and adopt quadrupole illuminating mode, obtain the aerial image distribution at the out of focus face place of first group of test mask, and then obtain the phase-shift phase of test mask aerial image intensity distributions on out of focus face that jointly causes of strange item of wave aberration, scalar apodization, two-way delay and two-way decay four kinds of aberrations

Wherein, represent the phase-shift phase of the out of focus face place aerial image that wave aberration and two-way delay cause,

$\begin{array}{c}{{\mathrm{\Θ}}_{\mathrm{no}}^W}^{\′}=\∫\∫Q(f,g)\·B\·\left\{\right[R_{\mathrm{no}}(f,g)-R_{\mathrm{no}}(f+f_{-1},g)]\·\cos \left(D(f+f_{-1},g;f,g)\right)-\\ [R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}}(f,g)]\·\cos \left(D(f,g;f+f_1,g)\right)\}\mathrm{dfdg}\end{array};$

represent the phase-shift phase of the single order intensity distributions in the out of focus face caused by scalar apodization and two-way decay;

Set up phase-shift phase and the relational expression between the orientation zernike coefficient of the zernike coefficient of the strange item of scalar apodization and the strange item of two-way decay:

Wherein, for the zernike coefficient of the strange item of scalar apodization, with for the orientation zernike coefficient of the strange item of two-way delay, for the sensitivity coefficient that the zernike coefficient of the strange item of scalar apodization is corresponding, with for the sensitivity coefficient that the orientation zernike coefficient of the strange item of two-way decay is corresponding;

$S_{\mathrm{no}}^A=\frac{{\mathrm{\Θ}}_{\mathrm{no}}^A}{\∫\∫Q(f,g)\·(B\·\cos \left(D(f+f_{-1},g;f,g)\right)+B\·\cos \left(D(f,g;f+f_1,g)\right))\mathrm{dfdg}};$

$S_{\mathrm{no}}^{\mathrm{dia}}=\frac{S_{d1\_o}^{\mathrm{dia}}+{S_{d1\_o}^{\mathrm{dia}}}^{\′}}{\∫\∫Q(f,g)\·(B\·\cos \left(D(f+f_{-1},g;f,g)\right)+B\·\cos \left(D(f,g;f+f_1,g)\right))\mathrm{dfdg}}$

$S_{-\mathrm{no}}^{\mathrm{dia}}=\frac{S_{d2\_o}^{\mathrm{dia}}+{S_{d2\_o}^{\mathrm{dia}}}^{\′}}{\∫\∫Q(f,g)\·(B\·\cos \left(D(f+f_{-1},g;f,g)\right)+B\·\cos \left(D(f,g;f+f_1,g)\right))\mathrm{dfdg}}---\left(4\right)$

$S_{d1\_o}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\·\cos \left(D(f,g;f+f_1,g)\right)\mathrm{dfdg}$

$S_{d2\_o}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\·\cos \left(D(f,g;f+f_1,g)\right)\mathrm{dfdg}$

${S_{d1\_o}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\·\cos \left(D(f+f_{-1},g;f,g)\right)\mathrm{dfdg}$

${S_{d2\_o}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\·\cos \left(D(f+f_{-1},g;f,g)\right)\mathrm{dfdg}$

$\begin{array}{c}{\mathrm{\Θ}}_{\mathrm{no}}^A=\∫\∫Q(f,g)\·B\·\left\{\right[R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}}(f,g)]\·\sin \left(D(f,g;f+f_1,g)\right)+\\ [R_{\mathrm{no}}(f+f_{-1},g)+R_{\mathrm{no}}(f,g)]\·\sin \left(D(f+f_{-1},g;f,g)\right)\}\mathrm{dfdg}\end{array}$

$S_{d1\_o}^{\mathrm{dia}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{dia}}+B_2{R}_{12}^{\mathrm{dia}}+B_3{R}_{13}^{\mathrm{dia}}+B_4{R}_{14}^{\mathrm{dia}})}_{(f,g;f+f_1,g)}\·\sin \left(D(f,g;f+f_1,g)\right)\mathrm{dfdg}$

$S_{d2\_o}^{\mathrm{dia}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{dia}}+B_2{R}_{22}^{\mathrm{dia}}+B_3{R}_{23}^{\mathrm{dia}}+B_4{R}_{24}^{\mathrm{dia}})}_{(f,g;f+f_1,g)}\·\sin \left(D(f,g;f+f_1,g)\right)\mathrm{dfdg}$

${S_{d1\_o}^{\mathrm{dia}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{dia}}+B_2{R}_{12}^{\mathrm{dia}}+B_3{R}_{13}^{\mathrm{dia}}+B_4{R}_{14}^{\mathrm{dia}})}_{(f+f_{-1},g;f,g)}\·\sin \left(D(f+f_{-1},g;f,g)\right)\mathrm{dfdg}$

${S_{d2\_o}^{\mathrm{dia}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{dia}}+B_2{R}_{22}^{\mathrm{dia}}+B_3{R}_{23}^{\mathrm{dia}}+B_4{R}_{24}^{\mathrm{dia}})}_{(f+f_{-1},g;f,g)}\·\sin \left(D(f+f_{-1},g;f,g)\right)\mathrm{dfdg}$

$R_{11}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}+1}(f+f_1,g)$

$R_{12}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f+f_1,g)-R_{\mathrm{no}}(f+f_1,g)$

$R_{13}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}+1}(f+f_1,g)-R_{\mathrm{no}}(f+f_1,g)$

$R_{14}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f+f_1,g)+R_{\mathrm{no}}(f+f_1,g)$

$R_{21}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g)$

$R_{22}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}+1}(f+f_1,g)$

$R_{23}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g)$

$R_{24}^{\mathrm{dia}}(f,g;f+f_1,g)={-R}_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)+R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g)$

Wherein, represent the phase place change that defocusing amount d causes, k is wave vector;

Change the coherence factor of lighting source and the cycle of first group of test mask, carry out the measurement under M different test condition, obtain phase-shift phase respectively then the sensitivity coefficient of the different strange item correspondence of test condition subscript quantity apodization of M group is obtained respectively according to formula (4) and the sensitivity coefficient of the strange item correspondence of two-way decay with matrix composed as follows:

$S_2=\left[\begin{array}{ccc}{S}_{\mathrm{no}}^{A1}& S_{\mathrm{no}}^{\mathrm{dia}1}& S_{-\mathrm{no}}^{\mathrm{dia}1}\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ S_{\mathrm{no}}^{\mathrm{AM}}& S_{\mathrm{no}}^{\mathrm{diaM}}& S_{-\mathrm{no}}^{\mathrm{diaM}}\end{array}\right]$

By the matrix S obtained under test conditions different for M group ₂be updated in relational expression (3), obtain the solution of the zernike coefficient of the strange item of M/54 group scalar apodization and the orientation zernike coefficient of the strange item of two-way decay;

Step 4, the polarization illumination mode of the lighting source in high-NA imaging system is arranged linearly polarized light at 45 °, and adopt traditional lighting mode, obtain the aerial image distribution at the desirable focal plane place of second group of test mask, and then adjacent peak intensity difference Δ I on the aerial image obtaining desirable focal plane place, set up the relational expression between the zernike coefficient of adjacent peak intensity difference Δ I and the even item of wave aberration and the orientation zernike coefficient of the even item of two-way delay:

$\begin{array}{c}\mathrm{\ΔI}={\mathrm{\Φ}}^{\′}+\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^W\·S_{\mathrm{ne}}^W-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^{\mathrm{ret}}\·S_{\mathrm{ne}}^{\mathrm{ret}}-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{-\mathrm{ne}}^{\mathrm{ret}}\·S_{-\mathrm{ne}}^{\mathrm{ret}}\\ ={\mathrm{\Φ}}^{\′}+4\mathrm{Re}\left[M\left(0\right)M^{*}\left(f_1\right)\right]\·[\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^W\·2\mathrm{\π}\·{\mathrm{\Θ}}_{\mathrm{ne}}\\ -\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^{\mathrm{ret}}\·(S_{1\_e}^{\mathrm{ret}}+{S_{1\_e}^{\mathrm{ret}}}^{\′})-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{-\mathrm{ne}}^{\mathrm{ret}}\·(S_{2\_e}^{\mathrm{ret}}+{S_{2\_e}^{\mathrm{ret}}}^{\′})]\end{array}---\left(5\right)$

Wherein, Φ '=-8Im [M (0) M ^*(f ₁)] (∫ ∫ Q (f, g) Bdfdg) be constant term, M (0) and M (f ₁) represent the amplitude of test mask zero level and positive first-order diffraction light respectively, for the zernike coefficient of the even item of wave aberration, with represent the orientation zernike coefficient of the even item of two-way delay, for the sensitivity coefficient that the zernike coefficient of the even item of wave aberration is corresponding, with for the sensitivity coefficient that the orientation zernike coefficient of the even item of two-way delay is corresponding;

$S_{\mathrm{ne}}^W=4\mathrm{Re}\left[M\left(0\right)M^{*}\left(f_1\right)\right]\·2\mathrm{\π}\·{\mathrm{\Θ}}_{\mathrm{ne}}$

$S_{\mathrm{ne}}^{\mathrm{ret}}=4\mathrm{Re}\left[M\left(0\right)M^{*}\left(f_1\right)\right]\·(S_{1\_e}^{\mathrm{ret}}+{S_{1\_e}^{\mathrm{ret}}}^{\′})$

$S_{-\mathrm{ne}}^{\mathrm{ret}}==4\mathrm{Re}\left[M\left(0\right)M^{*}\left(f_1\right)\right]\·(S_{2\_e}^{\mathrm{ret}}+{S_{2\_e}^{\mathrm{ret}}}^{\′})---\left(6\right)$

Ne value is 4,5,6,9,12,13,16,17,18,21,22,25,28,29,32,33,36,37;

$S_{1\_e}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\mathrm{dfdg}$

$S_{2\_e}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\mathrm{dfdg}$

${S_{1\_e}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\mathrm{dfdg}$

${S_{2\_e}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\mathrm{dfdg}$

Θ _ne＝∫∫Q(f，g)·B·[R _ne(f，g)-R _ne(f+f _-1，g)-R _ne(f+f ₁，g)+R _ne(f，g)]dfdg

R _ne(f, g) represents the even item in zernike polynomial;

Change the coherence factor of lighting source and the cycle of second group of test mask, carry out the measurement under M different test condition, obtain adjacent peak intensity difference Δ I respectively; Then the sensitivity coefficient of the even item correspondence of wave aberration under the different test condition of M group is obtained respectively according to formula (6) and the sensitivity coefficient of the even item correspondence of two-way delay with matrix composed as follows:

$S_3=\left[\begin{array}{ccc}{S}_{\mathrm{ne}}^{W1}& S_{\mathrm{ne}}^{\mathrm{ret}1}& S_{-\mathrm{ne}}^{\mathrm{ret}1}\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ S_{\mathrm{ne}}^{\mathrm{WM}}& S_{\mathrm{ne}}^{\mathrm{retM}}& S_{-\mathrm{ne}}^{\mathrm{retM}}\end{array}\right]$

By the matrix S under test conditions different for M group ₃be updated in formula (5), obtain the solution of the zernike coefficient of the even item of M/54 group wave aberration and the orientation zernike coefficient of the even item of two-way delay;

Step 5, the polarization illumination mode of the lighting source in high-NA imaging system is arranged linearly polarized light at 45 °, and adopt traditional lighting mode, obtain the aerial image distribution at the desirable focal plane place of the 3rd group of test mask, and then feature size error Δ CD on the aerial image obtaining desirable focal plane place, set up the relational expression between the zernike coefficient of feature size error Δ CD and the even item of scalar apodization and the zernike coefficient of two-way decay even number:

$\frac{1}{\mathrm{\ΔCD}}=\frac{M\left(f_1\right)\·M^{*}\left(f_1\right)}{I_{\mathrm{th}}}\·[\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^A\·S_{\mathrm{ne}}^A-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^{\mathrm{dia}}\·S_{\mathrm{ne}}^{\mathrm{dia}}-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{-\mathrm{ne}}^{\mathrm{dia}}\·S_{-\mathrm{ne}}^{\mathrm{dia}}]---\left(7\right)$

Wherein, I _thfor obtaining the intensity distributions threshold value of feature dimension of interest during aberrationless, M (f ₁) for participating in the frequency spectrum of the first-order diffraction light of interfering, * represents complex conjugate, for the zernike coefficient of the even item of scalar apodization, with for the orientation zernike coefficient of the even item of two-way decay, for the sensitivity coefficient that the zernike coefficient of the even item of scalar apodization is corresponding, with for the sensitivity coefficient that the orientation zernike coefficient of the even item of two-way decay is corresponding;

$\begin{array}{c}{S}_{\mathrm{ne}}^A=\∫\∫Q(f,g)\·{\left(B\right)}_{(f+f_{-1},g;f+f_1,g)}\·[R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)]\mathrm{dfdg}\\ S_{\mathrm{ne}}^{\mathrm{dia}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{dia}}+B_2{R}_{12}^{\mathrm{dia}}+B_3{R}_{13}^{\mathrm{dia}}+B_4{R}_{14}^{\mathrm{dia}})}_{(f+f_{-1},g;f+f_1,g)}\mathrm{dfdg}\\ S_{-\mathrm{ne}}^{\mathrm{dia}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{dia}}+B_2{R}_{22}^{\mathrm{dia}}+B_3{R}_{23}^{\mathrm{dia}}+B_4{R}_{24}^{\mathrm{dia}})}_{(f+f_{-1},g;f+f_1,g)}\mathrm{dfdg}\end{array}---\left(8\right)$

$R_{11}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)+R_{\mathrm{ne}+1}(f+f_1,g),$

$R_{12}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}+1}(f+f_{-1},g)-R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_1,g)-R_{\mathrm{ne}}(f+f_1,g),$

$R_{13}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_1,g)-R_{\mathrm{ne}}(f+f_1,g),$

$R_{14}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}+1}(f+f_{-1},g)-R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_1,g)+R_{\mathrm{ne}}(f+f_1,g),$

$R_{21}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)-R_{\mathrm{ne}+1}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)-R_{\mathrm{ne}+1}(f+f_1,g),$

$R_{22}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)+R_{\mathrm{ne}+1}(f+f_1,g),$

$R_{23}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)-R_{\mathrm{ne}+1}(f+f_{-1},g)-R_{\mathrm{ne}}(f+f_1,g)-R_{\mathrm{ne}+1}(f+f_1,g),$

$R_{24}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)={-R}_{\mathrm{ne}+1}(f+f_{-1},g)-R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)-R_{\mathrm{ne}+1}(f+f_1,g),$

Change the coherence factor of lighting source and the cycle of the 3rd group of test mask, carry out the measurement under M different test condition, obtain feature size error respectively; Then the sensitivity coefficient of the even item correspondence of the different test condition subscript quantity apodization of M group is obtained respectively according to formula (8) and the sensitivity coefficient of the even item correspondence of two-way decay with matrix composed as follows:

$S_4=\left[\begin{array}{ccc}{S}_{\mathrm{ne}}^{A1}& S_{\mathrm{ne}}^{\mathrm{dia}1}& S_{-\mathrm{ne}}^{\mathrm{dia}1}\\ .& .& .\\ .& .& .\\ .& .& .\\ S_{\mathrm{ne}}^{\mathrm{AM}}& S_{\mathrm{ne}}^{\mathrm{diaM}}& S_{-\mathrm{ne}}^{\mathrm{diaM}}\end{array}\right]$

By the matrix S under test conditions different for M group ₄be updated in relational expression (7), obtain the solution of the zernike coefficient of the even item of M/54 group scalar apodization and the orientation zernike coefficient of the even item of two-way decay;

Optimum solution in the M/54 group solution that step 6, respectively selecting step 2,3,4 and 5 obtain, is then updated in following four formulas respectively:

$A(f,g)=1-\underset{n=1}{\overset{37}{\mathrm{\Σ}}}{z}_n^A{R}_n(f,g)---\left(9\right)$

$W(f,g)=\underset{n=1}{\overset{37}{\mathrm{\Σ}}}{z}_n^W{R}_n(f,g)---\left(10\right)$

$\begin{array}{c}{J}_{\mathrm{dia}}(f,g)=N+\underset{n=1}{\overset{37}{\mathrm{\Σ}}}[z_n^{\mathrm{dia}}O{Z}_n(f,g)+z_{-n}^{\mathrm{dia}}{\mathrm{OZ}}_{-n}(f,g)]\\ =N+\underset{n=1}{\overset{37}{\mathrm{\Σ}}}\left[\begin{array}{c}{z}_n^{\mathrm{dia}}\left(\begin{array}{cc}{R}_n(f,g)& R_{n+1}(f,g)\\ R_{n+1}(f,g)& -R_n(f,g)\end{array}\right)+z_{-n}^{\mathrm{dia}}\left(\begin{array}{cc}{R}_n(f,g)& -R_{n+1}(f,g)\\ -R_{n+1}(f,g)& -R_n(f,g)\end{array}\right)\end{array}\right]\end{array}---\left(11\right)$

$\begin{array}{c}{J}_{\mathrm{ret}}(f,g)\≈N-i\underset{n=1}{\overset{37}{\mathrm{\Σ}}}[z_n^{\mathrm{ret}}O{Z}_n(f,g)+z_{-n}^{\mathrm{ret}}{\mathrm{OZ}}_{-n}(f,g)]\\ =N-i\underset{n=1}{\overset{37}{\mathrm{\Σ}}}\left[\begin{array}{c}{z}_n^{\mathrm{ret}}\left(\begin{array}{cc}{R}_n(f,g)& R_{n+1}(f,g)\\ R_{n+1}(f,g)& -R_n(f,g)\end{array}\right)+z_{-n}^{\mathrm{ret}}\left(\begin{array}{cc}{R}_n(f,g)& -R_{n+1}(f,g)\\ -R_{n+1}(f,g)& -R_n(f,g)\end{array}\right)\end{array}\right]\end{array}---\left(12\right)$

Wherein, A (f, g) represents scalar apodization, and W (f, g) represents wave aberration, J _dia(f, g) represents two-way decay, J _ret(f, g) represents two-way delay, N representation unit matrix, with be respectively the zernike coefficient representing scalar apodization and wave aberration, with and with represent the orientation zernike coefficient of two-way decay and two-way delay respectively; R _n(f, g) is zernike polynomial, OZ _n(f, g) and OZ _-n(f, g) represents orientation zernike polynomial;

The value of the formula obtained (9), (10), (11) and (12) is updated in formula (13),

J(f，g)＝A(f，g)·e ^iW(f，g)·J _dia(f，g)·J _ret(f，g)(13)

Obtain the Polarization aberration of high-NA imaging system.

Preferably, the optimum solution system of selection in described step 6 is: utilize least square method, one group of solution that Select Error is minimum, as optimum solution.

The present invention has following beneficial effect:

1), the method that relates in the present invention can detect the full detail that the Polarization aberration of the whole pupil of imaging system comprises, and do not need to carry out extra measurement, has the advantage that accuracy of detection is high, detection speed is fast;

2), the present invention to adopt in imaging system conventional test mask as test badge, do not need to design special detecting element (such as polaroid, wave plate and interferometer etc.), has the advantage that structure is simple, testing cost is low;

3), the present invention is by the analysis to polarization item difference mechanism of production, different lighting systems and mask is adopted to test, obtain respectively for zernike coefficient corresponding to characterize Polarization aberration four parts and orientation zernike coefficient, thus can accurate characterization Polarization aberration;

4), the present invention by changing lighting source parameter and mask parameters, repetitive measurement obtains zernike coefficient and orientation zernike coefficient, makes the Polarization aberration that obtains more accurate;

5), method of the present invention is applicable to, in any high NA polarization (vector) imaging system, include but are not limited to high resolution microscope, telescope and the etching system for the preparation of VLSI (very large scale integrated circuit).

Accompanying drawing explanation

Fig. 1 is the structural representation of the high NA imaging system that the present invention adopts.

Fig. 2 is types of illumination and the test mask structural representation of the employing of high NA imaging system.

Fig. 3 is the Polarization aberration and component thereof that input in the present embodiment.

Fig. 4 is the zernike coefficient of Polarization aberration in corresponding diagram 3 and the zernike coefficient contrast of this method detection.

Embodiment

To develop simultaneously embodiment below in conjunction with accompanying drawing, describe the present invention.

Etching system for preparation VLSI (very large scale integrated circuit) is example, the projection objective Polarization aberration pick-up unit that the present invention uses as shown in Figure 1, this device comprises the quasi-molecule laser source producing illumination light, the lighting source that can carry out beam collimation, produce any illumination shape, carry out even polarization illumination, test mask, carry the mask platform of mask, to the high-NA optical projection system of mask imaging, the sensor of record test mask aerial image intensity distribution, carry the work stage of image-position sensor, data handling system.

As shown in Figure 2, the lighting source polarization type adopted for this method and test mask mark.Wherein 201 represent quadrupole illuminating mode, 202 represent conventional part coherent illumination mode, the binary mask of 203 expression optical grating construction types, namely the mask mark that adopts of strange item of the strange item of wave aberration in Polarization aberration, the strange item of scalar apodization, the strange item of two-way delay and two-way decay is measured, 204 represent that phase shift is the alternating phase-shift mask of 90 °, and the mask that the even item of the even item and two-way delay of namely measuring wave aberration in Polarization aberration adopts marks; 205 represent that phase shift is the alternating phase-shift mask of 180 °, and the mask that the even item of the even item and two-way decay of namely measuring Polarization aberration acceptance of the bid quantitative change mark adopts marks.

According to the definition of physical decomposition, Polarization aberration can be decomposed into scalar apodization, wave aberration, two-way decay and two-way delay four parts hardly loss of accuracy, that is:

$J(f,g)=\left[\begin{array}{cc}{J}_{\mathrm{xx}}(f,g)& J_{\mathrm{xy}}(f,g)\\ J_{\mathrm{yx}}(f,g)& J_{\mathrm{yy}}(f,g)\end{array}\right]=A(f,g)\·e^{\mathrm{iW}(f,g)}\·J_{\mathrm{dia}}(f,g)\·J_{\mathrm{ret}}(f,g)---\left(1\right)$

Wherein, A (f, g) represents scalar apodization, and W (f, g) represents wave aberration, J _dia(f, g) represents two-way decay, J _ret(f, g) represents two-way delay, and (f, g) is pupil plane coordinate; In above formula, A and W is the form of scalar, the form of scalar zernike polynomial can be adopted to decompose it, that is:

$A(f,g)=1-\underset{n=1}{\overset{37}{\mathrm{\Σ}}}{z}_n^A{R}_n(f,g)---\left(2\right)$

$W(f,g)=\underset{n=1}{\overset{37}{\mathrm{\Σ}}}{z}_n^W{R}_n(f,g)---\left(3\right)$

Wherein, with be respectively the zernike coefficient representing scalar apodization and wave aberration, R _n(f, g) is zernike polynomial;

J _diaand J _rettwo is the form of vector, and each element is wherein the matrix of 2 × 2.According to the definition of orientation Ze Nike, J _diaand J _retcan be expressed as respectively:

$\begin{array}{c}{J}_{\mathrm{dia}}(f,g)=N+\underset{n=1}{\overset{37}{\mathrm{\Σ}}}[z_n^{\mathrm{dia}}{\mathrm{OZ}}_n(f,g)+z_{-n}^{\mathrm{dia}}{\mathrm{OZ}}_{-n}(f,g)]\\ =N+\underset{n=1}{\overset{37}{\mathrm{\Σ}}}[z_n^{\mathrm{dia}}\left(\begin{array}{cc}{R}_n(f,g)& R_{n+1}(f,g)\\ R_{n+1}(f,g)& -R_n(f,g)\end{array}\right)+z_{-n}^{\mathrm{dia}}\left(\begin{array}{cc}{R}_n(f,g)& -R_{n+1}(f,g)\\ -R_{n+1}(f,g)& -R_n(f,g)\end{array}\right)]\end{array}---\left(4\right)$

$\begin{array}{c}{J}_{\mathrm{rat}}(f,g)\≈N-i\underset{n=1}{\overset{37}{\mathrm{\Σ}}}[z_n^{\mathrm{rat}}{\mathrm{OZ}}_n(f,g)+z_{-n}^{\mathrm{rat}}{\mathrm{OZ}}_{-n}(f,g)]\\ =N-i\underset{n=1}{\overset{37}{\mathrm{\Σ}}}[z_n^{\mathrm{rat}}\left(\begin{array}{cc}{R}_n(f,g)& R_{n+1}(f,g)\\ R_{n+1}(f,g)& -R_n(f,g)\end{array}\right)+z_{-n}^{\mathrm{rat}}\left(\begin{array}{cc}{R}_n(f,g)& -R_{n+1}(f,g)\\ -R_{n+1}(f,g)& -R_n(f,g)\end{array}\right)\end{array}---\left(5\right)$

Wherein, N representation unit matrix, with and with represent the orientation zernike coefficient of two-way decay and two-way delay respectively, OZ _n(f, g) and OZ _-n(f, g) represents orientation zernike polynomial;

As can be seen from above formula, scalar apodization and wave aberration can be characterized with zernike coefficient, characterize two-way decay and two-way delay with orientation zernike coefficient.4 parts in Polarization aberration all can resolve into odd, even two parts according to the form of zernike polynomial, and represent the item number of strange aberration with subscript no, and subscript ne represents the item number of idol difference.In first 37 in zernike polynomial, Section 1 is constant term, and strange item no value is 2,3,7,8,10,11,14,15,19,20,23,24,26,27,30,31,34,35; Even item ne value is 4,5,6,9,12,13,16,17,18,21,22,25,28,29,32,33,36,37.

Method of the present invention comprises following concrete steps:

Step 1, the test mask arranged in high-NA imaging system: described test mask comprises three groups of test badges: first group is the binary mask of optical grating construction, second group is the alternating phase-shift mask of phase shift 90 °, and the 3rd group is the alternating phase-shift mask of phase shift 180 °; Three groups of test masks are intensive linear, and the dutycycle of opaque lines and transparent lines is 1: 1.

The sensitivity coefficient of the strange item of zernike coefficient of wave aberration and the strange item correspondence of the orientation zernike coefficient of two-way delay in step 2, calculating Polarization aberration:

Setting polarization illumination mode is 45 ° of linearly polarized lights, and adopt quadrupole illuminating mode, as shown in Fig. 2 201, the interior coherence factor variation range of quadrupole illuminating is 0.47-0.67, and the variation range of the external coherence system factor is 0.62 ~ 0.82, and ring width is 0.15.Adopt the first group echo in test mask, i.e. the binary mask of optical grating construction, as shown in Fig. 2 203, the mechanical periodicity scope of test mask is 100nm-200nm.With this understanding, the strange item of wave aberration and the strange item of two-way delay can cause the map migration of test mask aerial image intensity distributions on desirable focal plane.By carrying out fourier series decomposition to the intensity distributions of aerial image, the map migration effect of aerial image is also converted into the phase shift of its single order intensity.In order to set up aberration with the relation between map migration, need first to utilize vector imaging theory to calculate the intensity distributions of aerial image.Under Three-beam Interfere condition, carry out fourier progression expanding method to the intensity distributions of test mask aerial image, its single order intensity distributions is:

Wherein, x _irepresent the lateral coordinates of image planes, I ₁(f ₁) for the first-order spectrum of aerial image distributes, its form is

I ₁(f ₁)＝TCC _v(f _-1，0)·M(f _-1)·M ^*(0)+TCC _v(0，f ₁)·M(0)·M ^*(f ₁)

Wherein, f _-1, 0, f ₁represent respectively under normal incidence, the negative one-level of mask diffraction, zero level and the positive coordinate of one-level frequency spectrum point in x-axis, and f ₁=-f _-1.M (f _-1), M (0) and M (f ₁) represent that the amplitude of one-level, zero level and positive first-order diffraction light born by mask respectively.TCC _v(0, f ₁) represent transmitance transmission (TCC) function of zero level and the first-order diffraction interference of light, correspondingly TCC _v(f _-1, 0) and represent the TCC function of zero level and the negative first-order diffraction interference of light, the form of TCC function is:

TCC _v＝T·J·H·E·T ^*·J ^*·H ^*·E ^*

Wherein, E is the Jones vector representing incident light polarization state, and J is Polarization aberration $J=\left(\begin{array}{cc}{J}_{\mathrm{xx}}& J_{\mathrm{xy}}\\ J_{\mathrm{yx}}& J_{\mathrm{yy}}\end{array}\right),$ H is the pupil function of imaging system, and * represents complex conjugate, and T is the pupil transformation matrix of imaging system:

$T=\left(\begin{array}{cc}{T}_{\mathrm{xx}}& T_{\mathrm{xy}}\\ T_{\mathrm{yx}}& T_{\mathrm{yy}}\\ T_{\mathrm{zx}}& T_{\mathrm{zy}}\end{array}\right)$

And for the transposition mutually of single order intensity, its definition is:

Like this, the map migration that aberration causes just is converted into the transposition mutually of single order intensity change.Under 45 ° of linearly polarized light illuminations, $E={\left[\begin{array}{cc}\frac{\sqrt{2}}{2},& \frac{\sqrt{2}}{2}\end{array}\right]}^T,$ By the TCC function TCC of matrix form _vbe expressed as the form of numerical value, the value obtaining TCC function is:

$\begin{array}{c}\mathrm{TCC}={\mathrm{TCC}}_x+{\mathrm{TCC}}_y+{\mathrm{TCC}}_z\\ =J_{\mathrm{xx}}\·J_{\mathrm{xx}}^{*}\·B_1+J_{\mathrm{yy}}\·J_{\mathrm{yy}}^{*}\·B_2+J_{\mathrm{xy}}\·J_{\mathrm{yx}}^{*}\·B_3+J_{\mathrm{yx}}\·J_{\mathrm{xy}}^{*}\·B_4\end{array}$

Wherein

$B_1=0.5\·[T_{\mathrm{xx}}\·T_{\mathrm{xx}}^{*}+T_{\mathrm{yx}}\·T_{\mathrm{yx}}^{*}+T_{\mathrm{zx}}{T}_{\mathrm{zx}}^{*}]\·H\·H^{*},$

$B_2=0.5\·[T_{\mathrm{xy}}\·T_{\mathrm{xy}}^{*}+T_{\mathrm{yy}}\·T_{\mathrm{yy}}^{*}+T_{\mathrm{zy}}{T}_{\mathrm{zy}}^{*}]\·H\·H^{*},$

$B_3=0.5\·[T_{\mathrm{xx}}\·T_{\mathrm{xy}}^{*}+T_{\mathrm{yx}}\·T_{\mathrm{yy}}^{*}+T_{\mathrm{zx}}{T}_{\mathrm{zy}}^{*}]\·H\·H^{*},$

$B_4=0.5\·[T_{\mathrm{xy}}\·T_{\mathrm{xx}}^{*}+T_{\mathrm{yy}}\·T_{\mathrm{yx}}^{*}+T_{\mathrm{zy}}{T}_{\mathrm{zx}}^{*}]\·H\·H^{*}.$

By vector imaging theory, the real part and the imaginary part that obtain aerial image intensity distribution represent:

$\begin{array}{c}\mathrm{Re}\left[I_1\left(f_1\right)\right]=2\mathrm{Re}[M\left(f_{-1}\right)\·M^{*}\left(0\right)]\·\∫\∫Q(f,g)\·(B_1+B_2+B_3+B_4)\mathrm{dfdg}\\ +\mathrm{Im}[M\left(f_{-1}\right)\·M^{*}\left(0\right)]\·\left[\underset{\mathrm{no}}{\mathrm{\Σ}}\right[z_{\mathrm{no}}^{\mathrm{ret}}\·(S_{1\_o}^{\mathrm{ret}}+{S_{1\_o}^{\mathrm{ret}}}^{\′})+z_{-\mathrm{no}}^{\mathrm{ret}}\·(S_{2\_o}^{\mathrm{ret}}+{S_{2\_o}^{\mathrm{ret}}}^{\′})\left]\right]\\ +\mathrm{Im}[M\left(f_{-1}\right)\·M^{*}\left(0\right)]\·2\mathrm{\π}[\underset{\mathrm{no}}{\mathrm{\Σ}}{z}_{\mathrm{no}}^W\·(S_1^W-S_2^W)]\end{array}$

$\begin{array}{c}\mathrm{Im}\left[I_1\left(f_1\right)\right]=\mathrm{Re}[M\left(f_{-1}\right)\·M^{*}\left(0\right)]\·\left[\underset{\mathrm{no}}{\mathrm{\Σ}}\right[z_{\mathrm{no}}^{\mathrm{ret}}\·(S_{1\_o}^{\mathrm{ret}}+{S_{1\_0}^{\mathrm{ret}}}^{\′})+z_{-\mathrm{no}}^{\mathrm{ret}}\·(S_{2\_o}^{\mathrm{ret}}+{S_{2\_o}^{\mathrm{ret}}}^{\′})\left]\right]\\ +\mathrm{Re}[M\left(f_{-1}\right)\·M^{*}\left(0\right)]\·2\mathrm{\π}\·[\underset{\mathrm{no}}{\mathrm{\Σ}}{z}_{\mathrm{no}}^W\·(S_2^W-S_1^W)]\end{array}$

In conjunction with the definition of aerial image single order phase shift, can derive and represent the phase shift with desirable focal plane place aerial image single order intensity distributions of the zernike coefficient of wave aberration strange item and the zernike coefficient of the strange item of two-way delay pass be:

Wherein, for the sensitivity coefficient of the strange item correspondence of wave aberration, with sensitivity coefficient for the strange item correspondence of two-way delay:

$S_{\mathrm{no}}^W=\mathrm{\π}\frac{{\mathrm{\Θ}}_{\mathrm{no}}}{\∫\∫Q(f,g)\·\mathrm{Bdfdg}}$

$S_{\mathrm{no}}^{\mathrm{ret}}=\frac{S_{1\_o}^{\mathrm{ret}}+{S_{1\_0}^{\mathrm{ret}}}^{\′}}{2\∫\∫Q(f,g)\·\mathrm{Bdfdg}}---\left(7\right)$

$S_{-\mathrm{no}}^{\mathrm{ret}}=\frac{S_{2\_o}^{\mathrm{ret}}+{S_{2\_o}^{\mathrm{ret}}}^{\′}}{2\∫\∫Q(f,g)\·\mathrm{Bdfdg}}$

Q (f, g) represents the intensity of lighting source, B=B ₁+ B ₂+ B ₃+ B ₄;

$S_{1\_o}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\mathrm{dfdg},$

$S_{2\_o}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\mathrm{dfdg},$

${S_{1\_o}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\mathrm{dfdg},$

${S_{2\_o}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\mathrm{dfdg},$

$R_{11}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g),$

$R_{12}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f+f_1,g)+R_{\mathrm{no}}(f+f_1,g),$

$R_{13}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g),$

$R_{14}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f+f_1,g)-R_{\mathrm{no}}(f+f_1,g),$

$R_{21}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}+1}(f+f_1,g),$

$R_{22}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g),$

$R_{23}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}+1}(f+f_1,g),$

$R_{24}^{\mathrm{ret}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)-R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g),$

Θ _no＝∫∫Q(f，g)·B·[R _no(f，g)-R _no(f+f _-1，g)-R _no(f+f ₁，g)+R _no(f，g)]dfdg.

Wherein, R _no(f, g) represents the strange item in zernike polynomial;

As can be seen from above formula, sensitivity coefficient and lighting source intensity Q (f, g) relevant with parameter B, and parameter B is relevant with each component of Polarization aberration J and transmitance transport function TCC, the change of test mask parameter result in the change of each component of Polarization aberration J and transmitance transport function TCC, therefore, sensitivity coefficient is only relevant with lighting condition with test mask parameter, therefore different test masks and different lighting conditions is adopted, according to formula (7), the sensitivity coefficient under different test condition can be obtained, these sensitivity coefficients are expressed as the form of matrix, that is:

$S_1=\left[\begin{array}{ccc}{S}_{\mathrm{no}}^{W1}& S_{\mathrm{no}}^{\mathrm{ret}1}& S_{-\mathrm{no}}^{\mathrm{ret}1}\\ .& .& .\\ .& .& .\\ .& .& .\\ S_{\mathrm{no}}^{\mathrm{WM}}& S_{\mathrm{no}}^{\mathrm{retM}}& S_{-\mathrm{no}}^{\mathrm{retM}}\end{array}\right]$

Wherein, represent the sensitivity coefficient measuring the strange item correspondence of wave aberration obtained for the 1st time to the M time; with represent the sensitivity coefficient measuring the strange item correspondence of two-way delay obtained for the 1st time to the M time; Each part aberration due to Polarization aberration has 18 strange items and 18 even items, and strange item in two-way delay and two-way decay and even item comprise positive and negative two parts, therefore, calculate 54, at least will obtain 54 groups of data, the equation formed represented by 54 formulas (6) just can solve zernike coefficient, therefore the M value in the present invention is the multiple of 54, M value is larger, then data are more, and the precision of the zernike coefficient obtained is higher.By the matrix S obtained under test conditions different for M group ₁be updated in formula (6), M equation will be set up, and obtain the solution of the zernike coefficient of the strange item of M/54 group wave aberration and the orientation zernike coefficient of the strange item of two-way delay.

The sensitivity coefficient of step 3, the strange item of zernike coefficient of calculating Polarization aberration acceptance of the bid quantitative change mark and the strange item correspondence of the orientation zernike coefficient of two-way decay:

Setting polarization illumination mode is 45 ° of linearly polarized lights, and adopt quadrupole illuminating mode, as shown in Fig. 2 201, the interior coherence factor variation range of quadrupole illuminating is 0.47-0.67, and the variation range of the external coherence system factor is 0.62-0.82, and ring width is 0.15.Adopt the first group echo in test mask, i.e. the binary mask of optical grating construction, as shown in Fig. 2 203, the feature size variations scope of mask graph is 100nm-200nm, and the dutycycle of mask is always 1: 1.This step is with the difference of step 2, and this step obtains the aerial image of optical projection system out of focus position.Because the strange item of wave aberration and the strange item of two-way delay can cause the map migration of test mask aerial image intensity distributions on out of focus face, therefore, the map migration now obtained is the coefficient effect of strange item of wave aberration, scalar apodization, two-way delay and two-way decay four kinds of aberrations.Because wave aberration and the map migration that two-way decay causes on out of focus face are identical with in step 2, therefore from result, this part map migration can be rejected.

According to the derivation similar to step 2, utilize vector imaging theory, can derive and represent the phase shift with out of focus face place aerial image single order intensity distributions of the zernike coefficient of scalar apodization strange item and the zernike coefficient of the strange item of two-way decay between relation:

Wherein, represent the phase-shift phase jointly caused by the strange item of wave aberration, scalar apodization, two-way delay and two-way decay four kinds of aberrations;

represent the map migration of the out of focus face place aerial image that the aberration of step 2 causes, can determine according to the zernike coefficient in step 2 and following formula, that is:

Wherein, $\begin{array}{c}{{{\mathrm{\Θ}}_{\mathrm{no}}^W}^{\′}=\∫\∫Q(f,g)\·B\·\left\{\right[R_{\mathrm{no}}(f,g)-R_{\mathrm{no}}(f+f_{-1},g)]\·\cos \left(D(f+f_{-1},g;f,g)\right)-}^{}\\ [R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}}(f,g)]\·\cos \left(D(f,g;f+f_1,g)\right)\}\mathrm{dfdg}\end{array}$

$S_{d1\_o}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\·\cos \left(D(f,g;f+f_1,g)\right)\mathrm{dfdg}$

$S_{d2\_o}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\·\cos \left(D(f,g;f+f_1,g)\right)\mathrm{dfdg}$

${S_{d1\_o}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\·\cos \left(D(f+f_{-1},g;f,g)\right)\mathrm{dfdg}$

${S_{d2\_o}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\·\cos \left(D(f+f_{-1},g;f,g)\right)\mathrm{dfdg}$

$\begin{array}{c}{{\mathrm{\Θ}}_{\mathrm{no}}^A=\∫\∫Q(f,g)\·B\·\left\{\right[R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}}\left(f{,}_{}g\right)]\·\sin \left(D(f{,}_{}g;f+f_1,g)\right)+}^{}\\ [R_{\mathrm{no}}(f+f_{-1},g)+R_{\mathrm{no}}(f,g)]\·\sin \left(D(f+f_{-1},g;f,g)\right)\}\mathrm{dfdg}\end{array}$

$S_{d1\_o}^{\mathrm{dia}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{dia}}+B_2{R}_{12}^{\mathrm{dia}}+B_3{R}_{13}^{\mathrm{dia}}+B_4{R}_{14}^{\mathrm{dia}})}_{(f,g;f+f_1,g)}\·\sin \left(D(f,g;f+f_1,g)\right)\mathrm{dfdg}$

$S_{d2\_o}^{\mathrm{dia}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{dia}}+B_2{R}_{22}^{\mathrm{dia}}+B_3{R}_{23}^{\mathrm{dia}}+B_4{R}_{24}^{\mathrm{dia}})}_{(f,g;f+f_1,g)}\·\sin \left(D(f,g;f+f_1,g)\right)\mathrm{dfdg}$

${S_{d1\_o}^{\mathrm{dia}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{dia}}+B_2{R}_{12}^{\mathrm{dia}}+B_3{R}_{13}^{\mathrm{dia}}+B_4{R}_{14}^{\mathrm{dia}})}_{(f+f_{-1},g;f,g)}\·\sin \left(D(f+f_{-1},g;f,g)\right)\mathrm{dfdg}$

${S_{d2\_o}^{\mathrm{dia}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{dia}}+B_2{R}_{22}^{\mathrm{dia}}+B_3{R}_{23}^{\mathrm{dia}}+B_4{R}_{24}^{\mathrm{dia}})}_{(f+f_{-1},g;f,g)}\·\sin \left(D(f+f_{-1},g;f,g)\right)\mathrm{dfdg}$

$R_{11}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}+1}(f+f_1,g)$

$R_{12}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f+f_1,g)-R_{\mathrm{no}}(f+f_1,g)$

$R_{13}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}+1}(f+f_1,g)-R_{\mathrm{no}}(f+f_1,g)$

$R_{14}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f+f_1,g)+R_{\mathrm{no}}(f+f_1,g)$

$R_{21}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g)$

$R_{22}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)+R_{\mathrm{no}+1}(f,g)+R_{\mathrm{no}}(f+f_1,g)+R_{\mathrm{no}+1}(f+f_1,g)$

$R_{23}^{\mathrm{dia}}(f,g;f+f_1,g)=R_{\mathrm{no}}(f,g)-R_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g)$

$R_{24}^{\mathrm{dia}}(f,g;f+f_1,g)={-R}_{\mathrm{no}+1}(f,g)-R_{\mathrm{no}}(f,g)+R_{\mathrm{no}}(f+f_1,g)-R_{\mathrm{no}+1}(f+f_1,g)$

Wherein, represent the phase place change that defocusing amount d causes, k is wave vector, for the sensitivity coefficient that the zernike coefficient of the strange item of scalar apodization is corresponding, with for the sensitivity coefficient that the zernike coefficient of the strange item of two-way decay is corresponding, that is:

$\begin{array}{c}{S}_{\mathrm{no}}^A=\frac{{\mathrm{\Θ}}_{\mathrm{no}}^A}{\∫\∫Q(f,g)\·(B\·\cos \left(D(f+f_{-1},g;f,g)\right)+B\·\cos \left(D(f,g;f+f_1,g)\right))\mathrm{dfdg}};\\ S_{\mathrm{no}}^{\mathrm{dia}}=\frac{S_{d1\_o}^{\mathrm{dia}}+{S_{d1\_o}^{\mathrm{dia}}}^{\′}}{\∫\∫Q(f,g)\·(B\·\cos \left(D(f+f_{-1},g;f,g)\right)+B\·\cos \left(D(f,g;f+f_1,g)\right))\mathrm{dfdg}}\\ S_{-\mathrm{no}}^{\mathrm{dia}}=\frac{S_{d2\_o}^{\mathrm{dia}}+{S_{d2\_o}^{\mathrm{dia}}}^{\′}}{\∫\∫Q(f,g)\·(B\·\cos \left(D(f+f_{-1},g;f,g)\right)+B\·\cos \left(D(f,g;f+f_1,g)\right))\mathrm{dfdg}}\end{array}---\left(9\right)$

Because sensitivity coefficient is only relevant with mask parameters and lighting condition, therefore different test masks and different lighting conditions is adopted, the sensitivity coefficient under different test condition can be obtained according to formula (9), these sensitivity coefficients are expressed as the form of battle array, that is:

$S_2=\left[\begin{array}{ccc}{S}_{\mathrm{no}}^{A1}& S_{\mathrm{no}}^{\mathrm{dia}1}& S_{-\mathrm{no}}^{\mathrm{dia}1}\\ .& .& .\\ .& .& .\\ .& .& .\\ S_{\mathrm{no}}^{\mathrm{AM}}& S_{\mathrm{no}}^{\mathrm{diaM}}& S_{-\mathrm{no}}^{\mathrm{diaM}}\end{array}\right]$

Wherein, represent the sensitivity coefficient measuring the strange item correspondence of scalar apodization obtained for the 1st time to the M time; with represent the sensitivity coefficient measuring the strange item correspondence of two-way decay obtained for the 1st time to the M time; By the matrix S obtained under test conditions different for M group ₂be updated in relational expression (8), obtain the solution of the zernike coefficient of the strange item of M/54 group scalar apodization and the orientation zernike coefficient of the strange item of two-way decay.

The sensitivity coefficient of the even item of zernike coefficient of wave aberration and the even item correspondence of the orientation zernike coefficient of two-way delay in step 4, calculating Polarization aberration:

Setting polarization illumination mode is 45 ° of linearly polarized lights, and adopt traditional lighting mode, as shown in Fig. 2 202, the coherence factor variation range of traditional lighting is 0.08-0.21.Adopt the second group echo in test mask, namely phase shift is the alternating phase-shift mask of 90 °, and as shown in Fig. 2 204, the feature size variations scope of mask graph is 52nm-70nm, and the dutycycle of mask is always 1: 1.For the alternating phase-shift mask of 90 °, its aerial image can be expressed as:

I(x _i)＝Φ ₁+Φ ₂sin(2πf ₁x _i)+Φ ₃sin(4πf ₁x _i)+Φ ₄，

Wherein,

Φ ₁＝TCC _v(0；0)·M(0)·M ^*(0)+TCC _v(f ₁；f ₁)·M(f ₁)·M ^*(f ₁)

+TCC _v(f _-1；f _-1)·M(f _-1)·M ^*(f ₁)

Φ ₂＝-2{Im[TCC _v(0；f ₁)]+Im[TCC _v(0；f _-1)]}·Re[M(0)·M ^*(f _-1)]

-4Re[TCC _v(0；f ₁)]·Im[M(0)·M ^*(f ₁)]

Φ ₃＝-2Im[TCC _v(f ₁；f _-1)]·[M(f ₁)·M ^*(f _-1)]，

Φ ₄＝2Re[TCC _v(f ₁；f _-1)]·[M(f ₁)·M ^*(f _-1)].

TCC functional form in formula is identical with step 2, and just independent variable is different, and now adjacent intensity can be approximated to be:

I _max1＝Φ ₁+Φ ₂+Φ ₄；I _max2＝Φ ₁-Φ ₂+Φ ₄。

Due under this condition, other aberrations can't cause the adjacent peak difference of test mask intensity distributions on desirable focal plane, therefore by vector imaging theory, series expansion is carried out to TCC function, and get finite term, the zernike coefficient can deriving zernike coefficient and the even item of two-way delay representing the even item of wave aberration with the pass of adjacent peak intensity difference Δ I on resolution chart aerial image is:

$\begin{array}{c}\mathrm{\ΔI}=2{\mathrm{\Φ}}_2={\mathrm{\Φ}}^{\′}+\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^W\·S_{\mathrm{ne}}^W-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^{\mathrm{ret}}\·S_{\mathrm{ne}}^{\mathrm{ret}}-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{-\mathrm{ne}}^{\mathrm{ret}}\·S_{-\mathrm{ne}}^{\mathrm{ret}}\\ ={\mathrm{\Φ}}^{\′}+4\mathrm{Re}\left[M\left(0\right)M^{*}\left(f_1\right)\right]\·[\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^W\·2\mathrm{\π}\·{\mathrm{\Θ}}_{\mathrm{ne}}\\ -\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^{\mathrm{ret}}\·\left(S_{1\_e}^{\mathrm{ret}}+{S_{1\_e}^{\mathrm{ret}}}^{\′}\right)\underset{\mathrm{ne}}{-\mathrm{\Σ}}{z}_{-\mathrm{ne}}^{\mathrm{ret}}\·(S_{2\_e}^{\mathrm{ret}}+{S_{2\_e}^{\mathrm{ret}}}^{\′})]\end{array}---\left(10\right)$

Wherein, Φ '=-8Im [M (0) M ^*(f ₁)] (∫ ∫ Q (f, g) (B ₁+ B ₂+ B ₃+ B ₄) dfdg), be constant term, for the sensitivity coefficient that the zernike coefficient of the even item of wave aberration is corresponding, with for the sensitivity coefficient that the orientation zernike coefficient of the even item of two-way delay is corresponding, that is:

$S_{\mathrm{ne}}^W=4\mathrm{Re}\left[M\left(0\right)M^{*}\left(f_1\right)\right]\·2\mathrm{\π}\·{\mathrm{\Θ}}_{\mathrm{ne}}$

$S_{\mathrm{ne}}^{\mathrm{ret}}=4\mathrm{Re}\left[M\left(0\right)M^{*}\left(f_1\right)\right]\·(S_{1\_e}^{\mathrm{ret}}+{S_{1\_e}^{\mathrm{ret}}}^{\′})---\left(11\right)$

$S_{-\mathrm{ne}}^{\mathrm{ret}}==4\mathrm{Re}\left[M\left(0\right)M^{*}\left(f_1\right)\right]\·(S_{2\_e}^{\mathrm{ret}}+{S_{2\_e}^{\mathrm{ret}}}^{\′})$

$S_{1\_e}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\mathrm{dfdg}$

$S_{2\_e}^{\mathrm{ret}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f,g;f+f_1,g)}\mathrm{dfdg}$

${S_{1\_e}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{ret}}+B_2{R}_{12}^{\mathrm{ret}}+B_3{R}_{13}^{\mathrm{ret}}+B_4{R}_{14}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\mathrm{dfdg}$

${S_{2\_e}^{\mathrm{ret}}}^{\′}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{ret}}+B_2{R}_{22}^{\mathrm{ret}}+B_3{R}_{23}^{\mathrm{ret}}+B_4{R}_{24}^{\mathrm{ret}})}_{(f+f_{-1},g;f,g)}\mathrm{dfdg}$

Θ _ne=∫∫Q(f，g)·B·[R _ne(f，g)-R _ne(f+f _-1，g)-R _ne(f+f ₁，g)+R _ne(f，g)]dfdg

Utilize optical patterning model, adopt different test masks and different lighting conditions, according to formula (11), the sensitivity coefficient under different test condition can be obtained, these sensitivity coefficients are expressed as the form of matrix, that is:

$S_3=\left[\begin{array}{ccc}{S}_{\mathrm{ne}}^{W1}& S_{\mathrm{ne}}^{\mathrm{ret}1}& S_{-\mathrm{ne}}^{\mathrm{ret}1}\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ S_{\mathrm{ne}}^{\mathrm{WM}}& S_{\mathrm{ne}}^{\mathrm{retM}}& S_{-\mathrm{ne}}^{\mathrm{retM}}\end{array}\right]$

Wherein, represent the sensitivity coefficient measuring the even item correspondence of the wave aberration obtained for the 1st time to the M time; with represent the sensitivity coefficient measuring the even item correspondence of the two-way delay obtained for the 1st time to the M time; By the matrix S under test conditions different for M group ₃be updated in formula (10), obtain the solution of the zernike coefficient of the even item of M/54 group wave aberration and the orientation zernike coefficient of the even item of two-way delay.

The sensitivity coefficient matrix of step 5, the even item of calculating Polarization aberration acceptance of the bid quantitative change mark and the even item correspondence of two-way decay:

Setting polarization illumination mode is 45 ° of linearly polarized lights, and adopt traditional lighting mode, as shown in Fig. 2 202, the coherence factor variation range of traditional lighting is 0.12-0.3.Adopt the second group echo in test mask, namely phase shift is the alternating phase-shift mask of 180 °, and as shown in Fig. 2 205, the feature size variations scope of mask graph is 55nm-80nm, and the dutycycle of mask is always 1: 1.For the alternating phase-shift mask of 180 °, the amplitude of its zero order diffracted light equals zero, and therefore the intensity distributions of aerial image is:

$\begin{array}{c}I\left(x_i\right)={\mathrm{TCC}}_v(f_1;f_1)\·M\left(f_1\right)\·M^{*}\left(f_1\right)+{\mathrm{TCC}}_v(f_{-1};f_{-1})\·M\left(f_{-1}\right)\·M^{*}\left(f_{-1}\right)\\ +{\mathrm{TCC}}_v(f_{-1},f_{-1})M\left(f_{-1}\right)\·M^{*}\left(f_1\right)e^{j4\mathrm{\π}{f}_1{x}_i}+{\mathrm{TCC}}_v(f_1,f_{-1})M\left(f_1\right)\·M^{*}\left(f_{-1}\right)e^{-j4\mathrm{\π}{f}_1{x}_i}\\ {=\mathrm{TCC}}_v(f_1;f_1)\·M\left(f_1\right)\·M^{*}\left(f_1\right)+{\mathrm{TCC}}_v(f_{-1};f_{-1})\·M\left(f_{-1}\right)\·M^{*}\left(f_{-1}\right)\\ +2\mathrm{TC}{C}_v(f_1;f_{-1})\·M\left(f_1\right)\·M^{*}\left(f_{-1}\right)\·\cos \left(4\mathrm{\π}{f}_1{x}_i\right)\end{array}.$

Suppose that the aerial image that alternating phase-shift mask is formed in aberrationless imaging system is I in threshold value _thtime, can obtain desirable characteristic dimension, namely Δ CD equals 0.So for the imaging system that there is aberration, because other aberrations can't cause the feature size error of test mask aerial image distribution on desirable focal plane, therefore by vector imaging theory, the even zernike coefficient of item of scalar apodization can be derived and the zernike coefficient of two-way decay even number with the pass of feature size error Δ CD on resolution chart aerial image is:

$\frac{1}{\mathrm{\ΔCD}}=\frac{M\left(f_1\right)\·M^{*}\left(f_1\right)}{I_{\mathrm{th}}}\·[\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^A\·S_{\mathrm{ne}}^A-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{\mathrm{ne}}^{\mathrm{dia}}\·S_{\mathrm{ne}}^{\mathrm{dia}}-\underset{\mathrm{ne}}{\mathrm{\Σ}}{z}_{-\mathrm{ne}}^{\mathrm{dia}}\·S_{-\mathrm{ne}}^{\mathrm{dia}}]---\left(12\right)$

Wherein, I _thfor obtaining the intensity distributions threshold value of feature dimension of interest during aberrationless, M (f ₁) for participating in the frequency spectrum of the first-order diffraction light of interfering, * represents complex conjugate, for the sensitivity coefficient of the even item correspondence of scalar apodization zernike coefficient, with for the sensitivity coefficient of the even item correspondence of the orientation zernike coefficient of two-way decay.

$\begin{array}{c}{S}_{\mathrm{ne}}^A=\∫\∫Q(f,g)\·{\left(B\right)}_{(f+f_{-1},g;f+f_1,g)}\·[R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)]\mathrm{dfdg}\\ S_{\mathrm{ne}}^{\mathrm{dia}}=\∫\∫Q(f,g)\·{(B_1{R}_{11}^{\mathrm{dia}}+B_2{R}_{12}^{\mathrm{dia}}+B_3{R}_{13}^{\mathrm{dia}}+B_4{R}_{14}^{\mathrm{dia}})}_{(f+f_{-1},g;f+f_1,g)}\mathrm{dfdg}\\ S_{-\mathrm{ne}}^{\mathrm{dia}}=\∫\∫Q(f,g)\·{(B_1{R}_{21}^{\mathrm{dia}}+B_2{R}_{22}^{\mathrm{dia}}+B_3{R}_{23}^{\mathrm{dia}}+B_4{R}_{24}^{\mathrm{dia}})}_{(f+f_{-1},g;f+f_1,g)}\mathrm{dfdg}\end{array}---\left(13\right)$

$R_{11}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)+R_{\mathrm{ne}+1}(f+f_1,g),$

$R_{12}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}+1}(f+f_{-1},g)-R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_1,g)-R_{\mathrm{ne}}(f+f_1,g),$

$R_{13}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_1,g)-R_{\mathrm{ne}}(f+f_1,g),$

$R_{14}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}+1}(f+f_{-1},g)-R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_1,g)+R_{\mathrm{ne}}(f+f_1,g),$

$R_{21}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)-R_{\mathrm{ne}+1}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)-R_{\mathrm{ne}+1}(f+f_1,g),$

$R_{22}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}+1}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)+R_{\mathrm{ne}+1}(f+f_1,g),$

$R_{23}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)=R_{\mathrm{ne}}(f+f_{-1},g)-R_{\mathrm{ne}+1}(f+f_{-1},g)-R_{\mathrm{ne}}(f+f_1,g)-R_{\mathrm{ne}+1}(f+f_1,g),$

$R_{24}^{\mathrm{dia}}(f+f_{-1},g;f+f_1,g)={-R}_{\mathrm{ne}+1}(f+f_{-1},g)-R_{\mathrm{ne}}(f+f_{-1},g)+R_{\mathrm{ne}}(f+f_1,g)-R_{\mathrm{ne}+1}(f+f_1,g),$

Utilize optical patterning model, adopt different test masks and different lighting conditions, the sensitivity coefficient under different test condition can be obtained according to formula (13), these sensitivity coefficients are expressed as the form of matrix, that is:

$S_4=\left[\begin{array}{ccc}{S}_{\mathrm{ne}}^{A1}& S_{\mathrm{ne}}^{\mathrm{dia}1}& S_{-\mathrm{ne}}^{\mathrm{dia}1}\\ .& .& .\\ .& .& .\\ .& .& .\\ S_{\mathrm{ne}}^{\mathrm{AM}}& S_{\mathrm{ne}}^{\mathrm{diaM}}& S_{-\mathrm{ne}}^{\mathrm{diaM}}\end{array}\right]$

Wherein, represent the sensitivity coefficient measuring the even item correspondence of the scalar apodization obtained for the 1st time to the M time; with represent the sensitivity coefficient measuring the even item correspondence of the two-way decay obtained for the 1st time to the M time; By the matrix S under test conditions different for M group ₄be updated in relational expression (12), obtain the solution of the zernike coefficient of the even item of M/54 group scalar apodization and the orientation zernike coefficient of the even item of two-way decay.

Step 6, obtain the Polarization aberration of high-NA imaging system:

Utilize least square method, the solution that error respectively in the M/54 group solution that obtains of calculation procedure 2,3,4 and 5 is minimum, as optimum solution, then, the strange item of the zernike coefficient of one of optimum group of scalar apodization and even item are substituted in formula (2), obtains the expression formula of scalar apodization aberration; The strange item of the zernike coefficient of one of optimum group of wave aberration and even item are substituted in formula (3), obtains the expression formula of wave aberration; The strange item of the orientation zernike coefficient of one of optimum group of two-way delay and even item are substituted in formula (4), obtains the expression formula of two-way delay aberration; The strange item of the orientation zernike coefficient of one of optimum group of two-way decay and even item are substituted in formula (4), obtains the expression formula of two-way decay aberration; Finally the expression formula of above-mentioned 4 aberrations is all updated in formula (1), obtains the corrugated distribution of Polarization aberration, realize the measurement of the Polarization aberration of high-NA imaging system thus.

As shown in Figure 3, choose in emulation by the input quantity of the aberration of any visual field point of the lithographic projection system of lab design as emulation.301 represent the wave aberration of these visual field points, and 302 ~ 309 is 8 components of Jones's pupil form Polarization aberration of this visual field point.302,303 J is respectively _xxreal part and imaginary part.304,305 J is respectively _xyreal part and imaginary part.306,307 J is respectively _yxreal part and imaginary part.308,309 J is respectively _yyreal part and imaginary part.

As shown in Figure 4, the wave aberration of 301 is input in lithographic projection system and carries out imaging, obtain phase-shift phase respectively phase-shift phase after the difference DELTA I of adjacent peak and feature size error Δ CD, method according to the present invention calculates wave aberration, scalar apodization, the zernike coefficient of two-way delay and two-way decay or orientation zernike coefficient, as measured value;

Carry out physical decomposition and orientation Ze Nike to the Polarization aberration shown in Fig. 3 to decompose, the zernike coefficient of wave aberration, scalar apodization, two-way delay and two-way decay or orientation zernike coefficient, as input value.As shown in 401-406: 401 represent that the input value of the zernike coefficient of wave aberration and measured value contrast; 402 represent that the input value of the zernike coefficient of scalar apodization and measured value contrast; 403,404 the input value of two orientation zernike coefficients of two-way decay and measured value contrast is respectively; 405,406 the input value of two orientation zernike coefficients of two-way delay and measured value contrast is respectively.Can find measured value and input value difference seldom from figure, therefore, the Polarization aberration detection method based on method establishment of the present invention can obtain the full detail of Polarization aberration, and has higher accuracy of detection.

In sum, these are only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (2)

1. a detection method for high-NA imaging system Polarization aberration, is characterized in that, comprises the steps:

Step 1, the test mask arranged in high-NA imaging system: described test mask comprises three groups of test masks: first group of test mask is the binary mask of optical grating construction, second group of test mask is the alternating phase-shift mask of phase shift 90 °, and the 3rd group of test mask is the alternating phase-shift mask of phase shift 180 °; Three groups of test masks are intensive linear, and the dutycycle of opaque lines and transparent lines is 1:1;

Step 2, the polarization illumination mode of the lighting source in high-NA imaging system is arranged linearly polarized light at 45 °, and adopt quadrupole illuminating mode, obtain the aerial image distribution at the desirable focal plane place of first group of test mask, and then obtain the phase-shift phase of single order intensity distributions at desirable focal plane place then phase-shift phase is set up and the relational expression between the orientation zernike coefficient of the zernike coefficient of the strange item of wave aberration and the strange item of two-way delay:

Wherein, for the zernike coefficient of the strange item of wave aberration, with for the orientation zernike coefficient of the strange item of two-way delay, for the sensitivity coefficient of the strange item correspondence of wave aberration, with sensitivity coefficient for the strange item correspondence of two-way delay:

Q (f, g) represents the intensity of lighting source; (f, g) represents pupil plane coordinate; f _-1and f ₁under representing normal incidence respectively, the negative one-level of mask diffraction and the coordinate of positive one-level frequency spectrum point in x-axis;

Θ _no＝∫∫Q(f,g)·B·[R _no(f,g)-R _no(f+f _-1,g)-R _no(f+f ₁,g)+R _no(f,g)]dfdg,

Wherein, R _no(f, g) represents the strange item in zernike polynomial; No value is 2,3,7,8,10,11,14,15,19,20,23,24,26,27,30,31,34,35;

B＝B ₁+B ₂+B ₃+B ₄

T _xx, T _xy, T _yx, T _yy, T _zxand T _zybe respectively component in the pupil transformation matrix T of imaging system; H is the pupil function of imaging system, and * represents complex conjugate;

Change the coherence factor of lighting source and the cycle of first group of test mask, carry out the measurement under M different test condition, obtain phase-shift phase then the sensitivity coefficient of the strange item correspondence of wave aberration under the different test condition of M group is obtained respectively according to formula (2) and the sensitivity coefficient of the strange item correspondence of two-way delay with matrix composed as follows:

Wherein, the value of M is the multiple of 54;

By the matrix S obtained under test conditions different for M group ₁be updated in formula (1), obtain the solution of the zernike coefficient of the strange item of M/54 group wave aberration and the orientation zernike coefficient of the strange item of two-way delay;

Step 3, the polarization illumination mode of the lighting source in high-NA imaging system is arranged linearly polarized light at 45 °, and adopt quadrupole illuminating mode, obtain the aerial image distribution at the out of focus face place of first group of test mask, and then obtain the phase-shift phase of test mask aerial image intensity distributions on out of focus face that jointly causes of strange item of wave aberration, scalar apodization, two-way delay and two-way decay four kinds of aberrations

Wherein, represent the phase-shift phase of the out of focus face place aerial image that wave aberration and two-way delay cause,

represent the phase-shift phase of the single order intensity distributions in the out of focus face caused by scalar apodization and two-way decay;

Set up phase-shift phase and the relational expression between the orientation zernike coefficient of the zernike coefficient of the strange item of scalar apodization and the strange item of two-way decay:

Wherein, for the zernike coefficient of the strange item of scalar apodization, with for the orientation zernike coefficient of the strange item of two-way delay, for the sensitivity coefficient that the zernike coefficient of the strange item of scalar apodization is corresponding, with for the sensitivity coefficient that the orientation zernike coefficient of the strange item of two-way decay is corresponding;

Wherein, represent the phase place change that defocusing amount d causes, k is wave vector;

Change the coherence factor of lighting source and the cycle of first group of test mask, carry out the measurement under M different test condition, obtain phase-shift phase respectively then the sensitivity coefficient of the different strange item correspondence of test condition subscript quantity apodization of M group is obtained respectively according to formula (4) and the sensitivity coefficient of the strange item correspondence of two-way decay with matrix composed as follows:

By the matrix S obtained under test conditions different for M group ₂be updated in relational expression (3), obtain the solution of the zernike coefficient of the strange item of M/54 group scalar apodization and the orientation zernike coefficient of the strange item of two-way decay;

Step 4, the polarization illumination mode of the lighting source in high-NA imaging system is arranged linearly polarized light at 45 °, and adopt traditional lighting mode, obtain the aerial image distribution at the desirable focal plane place of second group of test mask, and then adjacent peak intensity difference Δ I on the aerial image obtaining desirable focal plane place, set up the relational expression between the zernike coefficient of adjacent peak intensity difference Δ I and the even item of wave aberration and the orientation zernike coefficient of the even item of two-way delay:

Wherein, Φ '=-8Im [M (0) M ^*(f ₁)] (∫ ∫ Q (f, g) Bdfdg) be constant term, M (0) and M (f ₁) represent the amplitude of test mask zero level and positive first-order diffraction light respectively, for the zernike coefficient of the even item of wave aberration, with represent the orientation zernike coefficient of the even item of two-way delay, for the sensitivity coefficient that the zernike coefficient of the even item of wave aberration is corresponding, with for the sensitivity coefficient that the orientation zernike coefficient of the even item of two-way delay is corresponding;

Ne value is 4,5,6,9,12,13,16,17,18,21,22,25,28,29,32,33,36,37;

Θ _ne＝∫∫Q(f,g)·B·[R _ne(f,g)-R _ne(f+f _-1,g)-R _ne(f+f ₁,g)+R _ne(f,g)]dfdg

R _ne(f, g) represents the even item in zernike polynomial;

Change the coherence factor of lighting source and the cycle of second group of test mask, carry out the measurement under M different test condition, obtain adjacent peak intensity difference Δ I respectively; Then the sensitivity coefficient of the even item correspondence of wave aberration under the different test condition of M group is obtained respectively according to formula (6) and the sensitivity coefficient of the even item correspondence of two-way delay with matrix composed as follows:

By the matrix S under test conditions different for M group ₃be updated in formula (5), obtain the solution of the zernike coefficient of the even item of M/54 group wave aberration and the orientation zernike coefficient of the even item of two-way delay;

Step 5, the polarization illumination mode of the lighting source in high-NA imaging system is arranged linearly polarized light at 45 °, and adopt traditional lighting mode, obtain the aerial image distribution at the desirable focal plane place of the 3rd group of test mask, and then feature size error Δ CD on the aerial image obtaining desirable focal plane place, set up the relational expression between the zernike coefficient of feature size error Δ CD and the even item of scalar apodization and the zernike coefficient of two-way decay even number:

Wherein, I _thfor obtaining the intensity distributions threshold value of feature dimension of interest during aberrationless, M (f ₁) for participating in the frequency spectrum of the first-order diffraction light of interfering, * represents complex conjugate, for the zernike coefficient of the even item of scalar apodization, with for the orientation zernike coefficient of the even item of two-way decay, for the sensitivity coefficient that the zernike coefficient of the even item of scalar apodization is corresponding, with for the sensitivity coefficient that the orientation zernike coefficient of the even item of two-way decay is corresponding;

Change the coherence factor of lighting source and the cycle of the 3rd group of test mask, carry out the measurement under M different test condition, obtain feature size error respectively; Then the sensitivity coefficient of the even item correspondence of the different test condition subscript quantity apodization of M group is obtained respectively according to formula (8) and the sensitivity coefficient of the even item correspondence of two-way decay with matrix composed as follows:

By the matrix S under test conditions different for M group ₄be updated in relational expression (7), obtain the solution of the zernike coefficient of the even item of M/54 group scalar apodization and the orientation zernike coefficient of the even item of two-way decay;

Optimum solution in the M/54 group solution that step 6, respectively selecting step 2,3,4 and 5 obtain, is then updated in following four formulas respectively:

Wherein, A (f, g) represents scalar apodization, and W (f, g) represents wave aberration, J _dia(f, g) represents two-way decay, J _ret(f, g) represents two-way delay, N representation unit matrix, with be respectively the zernike coefficient representing scalar apodization and wave aberration, with and with represent the orientation zernike coefficient of two-way decay and two-way delay respectively; R _n(f, g) is zernike polynomial, OZ _n(f, g) and OZ _-n(f, g) represents orientation zernike polynomial;

The value of the formula obtained (9), (10), (11) and (12) is updated in formula (13),

J(f,g)＝A(f,g)·e ^iW(f,g)·J _dia(f,g)·J _ret(f,g)(13)

Obtain the Polarization aberration of high-NA imaging system.

2. the detection method of a kind of high-NA imaging system Polarization aberration as claimed in claim 1, it is characterized in that, the optimum solution system of selection in described step 6 is: utilize least square method, one group of solution that Select Error is minimum, as optimum solution.