-
反射式斯托克斯椭偏仪采用DOAP结构作为光偏振态斯托克斯参量测量模块。其测量光偏振态的方程为:I=X·S,其中I为测量到的4路电流信号的4×1矩阵,S为待测光偏振态的4个斯托克斯参量的4×1矩阵,X为4×4的仪器矩阵,它的质量好坏直接影响光偏振态斯托克斯参量的测量精度,间接限制椭偏仪的测量精确度。为了实现仪器矩阵的优化,在传统DOAP的分光棱镜后的透射及反射光路上分别引进一块波片。改进后的DOAP光路结构如图 1所示。图 1中,i表示入射光, t和r分别表示透射光和反射光,BS为分光棱镜,Pt和Pr分别表示透射和反射光路的波片,WP1和WP2分别表示透射和反射光路的沃拉斯顿棱镜,D0, D1, D2和D3是同一种型号的光电探测器,i0, i1, i2和i3是光电探测器经过光电转换后的电信号。
令T和R分别表示分光棱镜透射和反射的米勒矩阵,用Mt和Mr分别表示透射和反射光路中波片的米勒矩阵,W代表沃拉斯顿棱镜米勒矩阵的第1行,α1和α2分别是透射和反射光路中沃拉斯顿棱镜方位角,则测量系统的仪器矩阵的表达式X为:
$ \mathit{\boldsymbol{X = }}\left[ {\begin{array}{*{20}{c}} \mathit{\boldsymbol{W}}({\alpha _1}){\mathit{\boldsymbol{M}}_{\rm{t}}}\mathit{\boldsymbol{T}}\\ \mathit{\boldsymbol{W}}({\alpha _1} + \frac{{\rm{ \mathsf{ π} }}}{2}){\mathit{\boldsymbol{M}}_{\rm{t}}}\mathit{\boldsymbol{T}}\\ \mathit{\boldsymbol{W}}({\alpha _2}){\mathit{\boldsymbol{M}}_{\rm{r}}}\mathit{\boldsymbol{R}}\\ \mathit{\boldsymbol{W}}({\alpha _2} + \frac{{\rm{ \mathsf{ π} }}}{2}){\mathit{\boldsymbol{M}}_{\rm{r}}}\mathit{\boldsymbol{R}} \end{array}} \right] $
(1) 以(1)式的第1行为例分析。设分光棱镜的透射椭偏参量为ψt和δt,则分光棱镜透射光路的米勒矩阵T为:
$ \begin{array}{*{20}{c}} {\mathit{\boldsymbol{T}} = {t_0} \times }\\ {\left[ {\begin{array}{*{20}{c}} 1&{ - \cos \left( {2{\psi _t}} \right)}&0&0\\ { - \cos \left( {2{\psi _t}} \right)}&1&0&0\\ 0&0&{\sin \left( {2{\psi _t}} \right)\cos {\delta _{\rm{t}}}}&{\sin \left( {2{\psi _t}} \right)\sin {\delta _{\rm{t}}}}\\ 0&0&{ - \sin \left( {2{\psi _t}} \right)\sin {\delta _{\rm{t}}}}&{\sin \left( {2{\psi _t}} \right)\cos {\delta _{\rm{t}}}} \end{array}} \right]} \end{array} $
(2) 式中,t0为分光棱镜的透射系数。令A1=-cos(2ψt), B1=sin(2ψt)cosδt, C1=sin(2ψt)sinδt,则T可以简化为:
$ \mathit{\boldsymbol{T}} = {t_0}\left[ {\begin{array}{*{20}{c}} 1&{{A_1}}&0&0\\ {{A_1}}&1&0&0\\ 0&0&{{B_1}}&{{C_1}}\\ 0&0&{ - {C_1}}&{{B_1}} \end{array}} \right] $
(3) 令βt和Φt分别为透射光路波片的偏振方位角和相位延迟量,则透射光路上的波片的传输矩阵Mt为:
$ \begin{array}{l} {\mathit{\boldsymbol{M}}_{\rm{t}}} = \\ \left[ {\begin{array}{*{20}{c}} 1&0&0&0\\ 0&{{{\cos }^2}\left( {2{\beta _t}} \right) + {{\sin }^2}\left( {2{\beta _t}} \right)\cos {\mathit{\Phi }_{\rm{t}}}}&{\left( {1 - \cos {\mathit{\Phi }_{\rm{t}}}} \right)\sin \left( {2{\beta _t}} \right)\cos \left( {2{\beta _t}} \right)}&{ - \sin \left( {2{\beta _t}} \right)\sin {\mathit{\Phi }_{\rm{t}}}}\\ 0&{\left( {1 - \cos {\mathit{\Phi }_{\rm{t}}}} \right)\sin \left( {2{\beta _t}} \right)\cos \left( {2{\beta _t}} \right)}&{{{\sin }^2}\left( {2{\beta _t}} \right) + {{\cos }^2}\left( {2{\beta _t}} \right)\cos {\mathit{\Phi }_{\rm{t}}}}&{\cos \left( {2{\beta _t}} \right)\sin {\mathit{\Phi }_{\rm{t}}}}\\ 0&{\sin \left( {2{\beta _t}} \right)\sin {\mathit{\Phi }_{\rm{t}}}}&{ - \cos \left( {2{\beta _t}} \right)\sin {\mathit{\Phi }_{\rm{t}}}}&{\cos {\mathit{\Phi }_{\rm{t}}}} \end{array}} \right] \end{array} $
(4) 为了方便分析,设α1与α2为45°,此时沃拉斯顿棱镜的米勒矩阵第1行为:
$ \mathit{\boldsymbol{W}}\left( {45{\rm{^\circ }}} \right) = \frac{1}{2}\left[ {\begin{array}{*{20}{c}} 1&0&1&0 \end{array}} \right] $
(5) 把(3)式、(4)式和(5)式代入(1)式, 可算得(1)式第1行的结果为:
$ \begin{array}{*{20}{c}} {{\mathit{\boldsymbol{X}}_1} = \frac{1}{2} \times }\\ {{{\left[ {\begin{array}{*{20}{c}} {{t_0}\left[ {1 + {A_1}\left( {1 - \cos {\mathit{\Phi }_{\rm{t}}}} \right)\sin \left( {2{\beta _t}} \right)\cos \left( {2{\beta _t}} \right)} \right]}\\ {{t_0}\left[ {{A_1}\left( {1 - \cos {\mathit{\Phi }_{\rm{t}}}} \right)\sin \left( {2{\beta _t}} \right)\cos \left( {2{\beta _t}} \right)} \right]}\\ {{t_0}{B_1}\left[ {{{\sin }^2}\left( {2{\beta _t}} \right) + {{\cos }^2}\left( {2{\beta _t}} \right)\cos {\mathit{\Phi }_{\rm{t}}}} \right] - {t_0}{C_1}\cos \left( {2{\beta _t}} \right)\sin {\mathit{\Phi }_{\rm{t}}}}\\ {{t_0}{C_1}\left[ {{{\sin }^2}\left( {2{\beta _t}} \right) + {{\cos }^2}\left( {2{\beta _t}} \right)\cos {\mathit{\Phi }_{\rm{t}}}} \right] + {t_0}{B_1}\cos \left( {2{\beta _t}} \right)\sin {\mathit{\Phi }_{\rm{t}}}} \end{array}} \right]}^{\rm{T}}}} \end{array} $
(6) 式中,T是矩阵的转置。同理可算出系统的仪器矩阵X的其它行。计算得到优化后DOAP的仪器矩阵X为:
$ \begin{array}{l} \mathit{\boldsymbol{X = }}\frac{1}{2}\\ \left[ {\begin{array}{*{20}{c}} {{t_0}\left[ {1 + {A_1}{D_1}\left( {1 - \cos {\mathit{\Phi }_{\rm{t}}}} \right)} \right]}&{{t_0}\left[ {{A_1} + {D_1}\left( {1 - \cos {\mathit{\Phi }_{\rm{t}}}} \right)} \right]}&{{t_0}\left( {{B_1}{E_1} - {C_1}{F_1}} \right)}&{{t_0}\left( {{C_1}{E_1} + {B_1}{F_1}} \right)}\\ {{t_0}\left[ {1 - {A_1}{D_1}\left( {1 - \cos {\mathit{\Phi }_{\rm{t}}}} \right)} \right]}&{{t_0}\left[ {{A_1} - {D_1}\left( {1 - \cos {\mathit{\Phi }_{\rm{t}}}} \right)} \right]}&{{t_0}\left( {{C_1}{F_1} - {B_1}{E_1}} \right)}&{ - {t_0}\left( {{C_1}{E_1} + {B_1}{F_1}} \right)}\\ {{r_0}\left[ {1 + {A_2}{D_2}\left( {1 - \cos {\mathit{\Phi }_{\rm{r}}}} \right)} \right]}&{{r_0}\left[ {{A_2} + {D_2}\left( {1 - \cos {\mathit{\Phi }_{\rm{r}}}} \right)} \right]}&{{r_0}\left( {{B_2}{E_2} - {C_2}{F_2}} \right)}&{{r_0}\left( {{C_2}{E_2} + {B_2}{F_2}} \right)}\\ {{r_0}\left[ {1 + {A_2}{D_2}\left( {1 - \cos {\mathit{\Phi }_{\rm{r}}}} \right)} \right]}&{{r_0}\left[ {{A_2} - {D_2}\left( {1 - \cos {\mathit{\Phi }_{\rm{r}}}} \right)} \right]}&{{r_0}\left( {{C_2}{F_2} - {B_2}{E_2}} \right)}&{ - {r_0}\left( {{C_2}{E_2} + {B_2}{F_2}} \right)} \end{array}} \right] \end{array} $
(7) 式中,r0是分光棱镜的反射系数,A2=-cos(2ψr), B2=sin(2ψr)cosδr, C2=sin(2ψr)sinδr, D1=sin(2βt)×cos(2βt), D2=sin(2βr)cos(2βr), E1=sin2(2βt)+cos2(2βt)cosΦt, E2=sin2(2βr)+cos2(2βr)cosΦr, F1=cos(2βt)sinΦt, F2=cos(2βr)sinΦr。其中,ψr和δr分别是分光棱镜的反射椭偏参量,βr和Φr分别为反射光路上波片的偏振方位角和相位延迟量。当Φt=Φr=0°时,(7)式对应于参考文献[15]中未加波片时的仪器矩阵,是一种特例。
由(7)式可知,仪器矩阵与波片的方位角和相位延迟量有关,这为利用波片优化系统仪器矩阵提供了依据。
-
为了验证优化DOAP系统后的测量准确度及重复性精度,组建了相应的斯托克斯椭偏仪测量系统,如图 4所示,选择仅在DOAP的透射光路插进一块λ/4波片。从He-Ne激光器出射的632.0nm激光束通过方位角为45°或135°的偏振片后以70°入射角入射到薄膜样品,反射后的偏振光进入优化后的DOAP。4路输出电流信号被放大后由计算机采样。选择4块标准薄膜样品,采用在位四区定标方法测量系统的仪器矩阵[18-19]并计算其条件数。表 1是插入一块λ/4波片前后测量3块不同条件镀制的硅衬底表面SiO2薄膜的实验结果。采用定标过的美国L116S300型STOKES椭偏仪测量结果作为参考值或标准值。分析表 1可知:(1)条件数越小,测量薄膜厚度及折射率的精确度越高;(2)系统插入波片,调节其方位角为90°时,测量系统处于最佳状态,测量薄膜厚度及折射率的重复性精度分别达到0.1nm和0.001。通过插入波片,可以实现椭偏测量系统的优化。这些结论与理论分析完全一致(见图 3a)。
Table 1. Measurement results of Stokes ellipsometer before and after the insert of wave-plate
no wave-plate quarter wave-plate at 45° quarter wave-plate at 90° reference value condition number 156.53 171.08 32.63 — sample 1 thickness/nm 95.3±0.9 95.1±0.9 102.1±0.1 101.84±0.04 refractive index 1.550±0.006 1.528±0.009 1.454±0.001 1.460±0.001 sample 2 thickness/nm 144.7±0.9 142.8±0.9 162.9±0.1 162.03±0.05 refractive index 1.565±0.007 1.571±0.011 1.460±0.001 1.463±0.001 sample 3 thickness/nm 109.0±1.1 113.2±1.2 184.9±0.1 185.19±0.04 refractive index 1.931±0.009 1.775±0.014 1.455±0.001 1.454±0.001
斯托克斯光偏振态测量系统的优化
Optimization of Stokes optical polarization measurement system
-
摘要: 为了满足光偏振态分振幅测量模块(DOAP)对分光棱镜复杂且严格的加工要求,采用在经典DOAP透射光路及反射光路各引入一块波片的方法,组成改进后的光偏振态测量模块。推导了新的仪器矩阵表达式,通过分析波片参量对仪器矩阵条件数的影响,得到了最佳波片的参量及其关系。结果表明,优化后的斯托克斯椭偏仪测量薄膜样品的厚度和折射率的标准差分别为0.1nm和0.001。通过选择波片的最佳方位角或相位延迟量可以实现斯托克斯椭偏仪仪器矩阵的优化,从而提高系统的测量稳定性及可靠性。Abstract: Division-of-amplitude photopolarimeter(DOAP) was the key part of Stokes ellipsometer. To reach the complex and strict processing demands of beam splitter in DOAP, two waveplates are inserted respectively in the reflection light path and transmittion light path of classic DOAP to form an improved DOAP. The instrument matrix of new DOAP was derived and influence of waveplate parameters on the condition number of instrument matrix was also analyzed. From theoretical analysis and experimental verification, the optimal parameters and their relationships of waveplates were gotten. The results show that, measurement standard errors of film thickness and refractive index of the optimized Stokes ellipsometer are 0.1nm and 0.001 respectively. By choosing the optimal azimuthal angles or phase retardations of waveplates, the instrument matrix can be optimized to improve the accuracy and stability of measurement.
-
Table 1. Measurement results of Stokes ellipsometer before and after the insert of wave-plate
no wave-plate quarter wave-plate at 45° quarter wave-plate at 90° reference value condition number 156.53 171.08 32.63 — sample 1 thickness/nm 95.3±0.9 95.1±0.9 102.1±0.1 101.84±0.04 refractive index 1.550±0.006 1.528±0.009 1.454±0.001 1.460±0.001 sample 2 thickness/nm 144.7±0.9 142.8±0.9 162.9±0.1 162.03±0.05 refractive index 1.565±0.007 1.571±0.011 1.460±0.001 1.463±0.001 sample 3 thickness/nm 109.0±1.1 113.2±1.2 184.9±0.1 185.19±0.04 refractive index 1.931±0.009 1.775±0.014 1.455±0.001 1.454±0.001 -
[1] XIAO G H, HUANG Z H, LIU Ch N, et al. Analysis of ellipsometric data processing about transparent film on the transparent substrate[J]. Laser Technology, 2010, 34(2):247-249(in Chinese). [2] CHENG M X, HE Zh J, HUANG Z H. A method of ellipsometry based on Stokes parameters with a division-of-amplitude polarimeter(DOAP) system[J]. Opto-Electronic Engineering, 2009, 36(7):100-106(in Chinese). [3] LÜ X, JIANG Zh Q, LI Y Ch. Study on the measurement of SOP with high-speed and real-time[J]. Optical Technique, 2010, 36(6):866-870(in Chinese). [4] QIN Z Y, CHENG Z G, ZHANG Z P, et al. High-speed and real-time measurement of Stokes parameters[J]. Acta Optica Sinica, 2007, 27(4):659-662(in Chinese). [5] LIN H H, PHAN Q H, LO Y L. Characterization of voltage-driven twisted nematic liquid crystal cell by dynamic polarization scanning ellipsometry[J]. Optics Express, 2015, 23(8):10213-10223. doi: 10.1364/OE.23.010213 [6] HALL S A, COVERT P A, BLINN B R, et al. Rapid and sensitive polarization measurement for characterizing protein adsorption at the solid-liquid interface[J]. Journal of Physical Chemistry, 2013, C117(4):1796-1803. [7] LO Y L, HSIEH W H, CHUNG Y F, et al. An approach for measuring the ellipsometric parameters of isotropic and anisotropic thin films using the stokes parameter method[J]. Journal of Lightwave Technology, 2012, 30(14):2299-2306. doi: 10.1109/JLT.2012.2196977 [8] CHENG M X, HE Z J, HUANG Z H. Measurement and application of stokes parameters of polarized light[J]. Infrared and Laser Engineering, 2006, 35(s2):109-115(in Chinese). [9] XU X S. The design of an experimental study of polarimeter system of Stokes parameters based on division-of-amplitude[D]. Hangzhou: Zhejiang University, 2011: 44-52(in Chinese). [10] AZZAM R M A. Division-of-amplitude photopolarimeter (DOAP) for the simultaneous measurement of all four Stokes parameters of light[J]. Optica Acta, 1982, 29(5):685-689. doi: 10.1080/713820903 [11] AZZAM R M A. Division-of-amplitude photopolarimeter based on conical diffraction from a metallic grating[J]. Applied Optics, 1992, 31(19):3574-3576. doi: 10.1364/AO.31.003574 [12] AZZAM R M A, SUDRADJAT F F. Single-layer-coated beam splitters for the division-of-amplitude photopolarimeter[J]. Applied Optics, 2005, 44(2):190-196. doi: 10.1364/AO.44.000190 [13] TAYA S A, EI-AGEZ T M, ALKANOO A A. Ellipsometric configurations using a phase retarder and a rotating polarizer and analyzer at any speed ratio[J]. Chinese Physics, 2012, B21(11):110701. [14] YUAN W J, SHEN W D, ZHANG Y G, et al. Dielectric multilayer beam splitter with differential phase shift on transmission and reflection for division-of-amplitude photopolarimeter[J]. Optics Express, 2014, 22(9):11011-11020. doi: 10.1364/OE.22.011011 [15] AZZAM R M A. Beam splitters for the division-of-amplitude photopolarimeter[J]. Optica Acta, 1985, 32(11):1407-1412. doi: 10.1080/713821665 [16] AZZAM R M A. Optimal beam splitters for the division-of-amplitude photopolarimeter[J]. Journal of the Optical Society of America, 2003, A20(5):955-958. [17] REN Sh F, WANG X X. Polarization states of linearly polarized light transmitting through several wave-plates with arbitrary thickness[J]. Laser Technology, 2014, 38(3):394-397(in Chinese). [18] ZHANG Y, HUANG Z H, ZHAO Z T, et al. In-place calibration of Stokes ellipsometer's instrument matrix[J]. Acta Optica Sinica, 2013, 33(2):0212002(in Chinese). doi: 10.3788/AOS [19] WANG Y H, ZHENG C L, ZHAO Z T. Muti-point calibration method based on Stokes ellipsometry system[J]. Chinses Journal of Lasers, 2012, 39(11):1108013(in Chinese). doi: 10.3788/CJL