衍射法X射线激光分束光栅平行度测量系统_英文

衍射法X射线激光分束光栅平行度测量系统_英文 ----------------------- Page 1----------------------- 第 15卷第1期 功能材料与器件学报 V ol15, No1 2009年2月 JOURNA L OF FUNCT IONA L MATER IALS AND DEV ICES Feb., 2009 文章编号: 1007- 4252 2009) 01- 0061- 05 Perform ance of a soft x- ray splitter grating parallelism m easuring system by diffraction m ethod TAN X in, LIU Y ing, L IU Zheng- kun, FU Shao- jun National Synchrotron Radiation Laboratory, University of Science and Technology of China, Hefei 230029, China) Abstract: W e present new progress in the d iffraction grating interferom eter being pre - aligned used a double frequency grating. To m easure the parallelism of the double frequency grating to a nicety before built in the interferom eter, a dev ice based on D iffraction Technique for m easuring the parallelism of the double frequency grating is designed. It is built of a sem iconductor laser, a collmi ator, gratings, a preci sion turnplate, a beeline workbench, a redressal shelves, a ccd detector. The system error of the dev ice is analyzed in this paper, and the parallelism of the doub le frequency grating ism easured by th is dev ice. The results dem onstrate that the diffraction m easuring dev ice su its the parallelism m easured the d iffraction grat ing interferom eter based on the double frequency grating of that parallelism can attain a h igh pre- aligning precision. K ey w ords: Double frequency grating; parallelism; d iffraction techn ique CLC: TN214 Docum ent code: A 衍射法X射线激光分束光栅平行度测量系统 谭鑫, 刘颖, 刘正坤, 付绍军 中国科学技术大学国家同步辐射实验室, 合肥230029) 摘要: 介绍了软X 射线光 分束镜M ach- Zehnder干涉等离子体诊断系统调整用双频 光 线条平 行度的衍射法测量系统。该系统主要由激光器, 准直镜, 待测光 , 精密转台 含角度测 量仪), 直 线工作台, 光 调节架和探测器组成。 分析 定性数据统计分析pdf销售业绩分析模板建筑结构震害分析销售进度分析表京东商城竞争战略分析 了系统中各种误差对测量精度的影响其中包 括距离测 量误差、波长误差、光 准直误差即光 刻线与入射光和反射光组成的平面不垂直、光 转 动过程 的误差、光 表面面型误差、探测器误差, 经计算得到系统的绝对误差为m inute。计算 表明, 该系统 的测量精度满足软X 射线M ach- Zehnder干涉系统对双频调整光 的性能要求。 关键词: 双频光 ; 平行度; 衍射法 0 Introduction laser- produced plasm a, interferom etry has m any ad vantages in the accuratem easurem ent of the plasm a e A s aw idely used d iagnosing tool in the research of 收稿日期: 2008- 03- 25; 修订日期: 2008- 05- 15 基金项目: NationalNature Science Foundation No. 10676032) . 作者简介: TAN X in, graduate student, E- mail: d ippen@ m ai.l ustc. edu. cn); FU Shaojun, professor, E - m ai:l sjfu@ ustc. edu. cn). ----------------------- Page 2----------------------- 62 功能材料与器件学报 15卷 lectron density because it d irectly gives refraction index ted in figure 1. The parallelism is indirectly calcu lated m apping from the interference pattern processing. Over by m easuring the vertical position error of the diffracted years, v isib le and UV interferom etry have been app lied beam s. in the study of laser plasm a characteristics such as pro The precision of the dev ice lies on the m easuring file steepening, filam entation and others. precision of the vertical position error and the aligned The electron density grad ient and electron density precision of the dev ice. The double frequency grating of larger var iety p lasm a w ill be probed by using the soft is deposited to incept the beam s of the sem iconductor X- ray diffraction grating M ach- Zehnder interferom e laser. Then w e can obtain the vertical position error of ter DG I). Hereby, it is very mi portant to the Plasm a the tw o m diffracted order beam s m 2 ). W e can D iagnosis, X- ray Laser and so on. calculate the angle of the lines of the double The diffraction grating in terferom eter must be pre - aligned by a v isible light like a sem iconductor la [ 1, 2] ser , because softX - ray laser is a sightless light. To ensure that the SXRL and the sem iconductor laser beam s follow the sam e path, the d iffraction gratings w ere ruled w ith tw o vertically separated regions of d if ferent line densities. There is a large difference in the line densities for the soft X - ray laser and the sem i conductor laser 50 mu ltip les upw ards). W e choose the double frequency grating DFG ) instead of two F ig. 1 M easuring principium gratings, because both of the parallelism and p laneness of the double frequency grating is better. To achieve a frequency grating by Equation 1): high precision of the pre - align ing, w e must m easure h tan= 1) the parallelism of the double frequency grating to a n i L cety before built in. H is the vertical position error perpendicu lar to To m easure the parallelism of the double frequen the XY p lane in figure1) and L is the horizontal dis cy grating to a n icety before built in the interferom eter, tance from the zero d iffracted order beam s to them eas a dev ice based on D iffraction Techn ique[ 5] for m easur uring poin t of the m diffracted order beam s parallel to ing the parallelism of the double frequency grating is the XY p lane in figure1). If the lines of the double designed. It is bu ilt of a sem iconductor laser, a colli frequency grating have high parallelism, the diffracted m ator, gratings, a precision turnplate, a beeline work beam s w ill spread in theXY plane. bench, a redressal shelves, a ccd detector. The system W e can easily calculate the by m easuring the h error of the dev ice is analyzed in this paper, and the andL. Tom easure thew holeDFG, w e used a beeline parallelism of the double frequency grating is m easured w orkbench. To ensure the precision, w em easured one DFG tmi e after tmi e and took m ore d iffracted orders to by this dev ice bem easured. 1 M easuring the parallelism of the B. M easuring m ethod doub le frequency gratings based on d if To ensure the precision of the h, w e must ascer fraction techn ique tain the benchm ark position of the DFG. The bench A. M easuring principium m ark position is a plane w hich is perpend icularity to he princip ium of our m easuring dev ice is illustra the XY p lane and the high frequency grating of the ----------------------- Page 3----------------------- 1期 谭鑫, 等: 衍射法X 射线激光分束光栅平 行度测量系统 63 DFG is in. by the ccd detector. Calculate the facu las positions 1 1 a. Ensuring the benchm ark position h and h . W e can calcu late h by: h = h - h . m m m m M easure the h ighnessof the zero diffracted order The error ofh ism ade up of the benchm ark posi 1 beam s and the highnessof the m d iffracted order beam s tion error and m easuring error ofhm . of the h igh frequency lines by the precision turnplate. D ifferential equation of the equation above: !h = M odu late the highness error bewt een and to the preci 1 !h - !h . m m sion of the detector is) . Then w e get the benchm ark The benchm ark position error is brought on by the position. p lane error of the precision turnplate and the detector b. Ensuring L the horizontal distance from the error. Grating collmi ation error can be ignore, because zero d iffracted order beam s to the m easuring poin t of to ensure the benchm ark position w e m odu late the them d iffracted order beam s) h ighness error bewt een h0 and hm to 0 ! 0. 05?m. The W e m easured by the beam s in norm al incidence, p lane error of the precision turnp late is ! 0. 001mm. so w e can calculate the position of them d iffraction or The facu las positions error is ! 0. 05?m. The bench der and L by: N m = sin , N m = sin and L = 1 1 2 2 m ark position error is ! 1. 05?m. L sin - L sin = L (N -N )m 1 0 2 0 1 0 2 1 The m easuring error ofhm is the detector error, e N 1 is the frequency of the h igh frequency grating qual to ! 10- 3 ?m. In case of there is no relation of the of the DFG. N 2 is the frequency of the other grating two k inds errors, !h = ! 1. 051?m. Due to the size of that used w ith the h igh frequency grating to bring a low the detector, the m ax mi um of h is 5. 09mm. H of our frequency grating. L0 is the d istance betw een the de DFG is betw een 0. 0628mm and 0. 0175mm L= tector and the incidence point of the DFG. is the 0120mm) . Relative error of h is !h /h = 0. 024 ~ 0. w avelength of the laser. 086. c. M easuring h the vertical position error) Record the facu las of them d iffracted order beam s by the ccd detector. Calculate the facu las s positions. B. M easuring error ofL0 Then com e at h. The m easuring error of L0 is brought on by the reading error and the grating collmi ation error. The 3 Perform ance of them easuring system reading error is ! 0. 1mm. The grating collmi ation er There are m any k inds of errors to influence the ror is the angle of the lines of the high frequency grat precision of the dev ice, such as distance error, w ave ing and the p lane form ed by the incidence and diffrac length error, grating collmi ation error, turn ing error, ted beam s. W e presum e the angle #= 0. 5. grating shape error, detector error. The relative m easuring error ofL0 is calculated by W e can calcu late it by Equation 4) : E quation 5) : h !L0 L - L L - L2 - (L cos sin #) 2 tan= L0 (N 2 -N 1 )m 2) L0 = 0 L0 f a ct = 0 0 L0 0 2 ( 5) h = 1- 1- ( cos sin #) 2 = arctan 2 L0 (N 2 - N 1 )m The h igh frequency grating is ru led 1000 /mm and (N -N )L !h- (N -N )L !h- (N -N ) h!L 2 1 0 2 1 0 2 2 0 the low frequency grating is ru led 16/mm. = 18. != 2 2 2 [ (N -N )L ] + h - 5 2 1 0 1399?. So !L0 /L0 = 3. 44 # 10 . 4) The grating collmi ation error can be ignored be cause it far less than the read ing error in the dev ice A. The error of h w ithL0 less than 1000mm. It can be easily detect that Record the facu las of them d iffracted order beam s the m easuring error of the dev ice w ill be largened if the ----------------------- Page 4----------------------- 64 功能材料与器件学报 15卷 read ing ofL0 is less than the tr mi size. ! 1 and ! 2 diffracted orders beam s for ten tmi es. W e m ake the detector to record one data every 0. 2s. W e C. O ther errors obtain 40 datum every tmi e for one point. The datum processed by m atlab in figure 2. ? + show s the paral Thew avelength error is about 10~ 7nm, it can be ignored. The dispersion of the laser doesn t affect the lelism. The mulilines are the parallelism of each posi tion. The broken lines are the average parallelism of detector receiv ing the m diffracted order beam s, and - 3 each position. the precision of the detector is 10 ?m, so it can be W e calculate the average parallelism of each DFG ignored. The stab ility of the laser is good, so w e can by m atlab and show it in figure3. ? + show s the par ignore it. allelism. The mu lilines are the parallelism of each The w avelength of the laser in the parallelism m easured dev ice is 632. 8nm. L = 120mm. The abso DFG. The broken lines are the average parallelism of 0 - 4 each DFG. lute error of the dev ice is ! 8. 22 # 10 m inute. Then The absolute error of the d iffraction m easuring de w e can see that the parallelism of the DFG wh ich is - 4 v ice is ! 8. 22 # 10 m inu te. It is far less than the m easured by th is dev ice is creditab le parallelism, so w e can use this dev ice. S ix pieces of double frequency gratings had been m easured by th is 4 Results of the dfg s parallelism and dev ice. The m ax mi al scope of the parallelism is be the pre- aligned precision of the soft x- tw een 0. 60m inute and 1. 35m inute. The m ax mi al ray laser interferom eter scope of the average parallelism is betw een 0. 89m inute W e process the data w e got to obtain the parallel and 0. 98m inute. The m ax mi al undulation scope of the ism of the DFG. Then w e use it tom odify the pre- a parallelism of these gratings is 0. 05m inute. In these ligned error of the soft X- ray laser interferom eter. W e six pieces of doub le frequency gratings, w e found the obtain the pre- aligned precision of the softX - ray la best. Its parallelism m ax mi al scope is betw een 0. ser interferom eter the parallelism of DFG on ly). 60m inute and 0. 65m inute. If w e optmi ize our condi tion, w e can do better. A. D ata processing F igure3 Average parallelism of the DFG Figure 2 Parallelism of the DFG B. Precision of pre- align ing the d iffraction grating M ach- Zehnder interferom eter There are five points on the DFG w hich is ru led The DG I likes figure 4, its precision is affected 25 ! 2mm w idth to be m easured. M easure each of the not on ly by the parallelism of theDFG, but also by the ----------------------- Page 5----------------------- 1期 谭鑫, 等: 衍射法X 射线激光分束光栅 平行度测量系统 65 assemblage precision, the rectify ing precision, the col The pre- aligning precision show ing above is outclass lmi ation precision of theDFG, the position precision of the reality, because the size of the DGI is sm aller. the m irror and so much other uncertain factor. The Therefore, the DFG is satisfiab le and the D iffraction conclusion after is calcu lated w hen all the factor is ig Techn ique for m easuring the parallelism of theDFG in nored besides the parallelism. troduced above is precise and conven ien N itrogen - doped T iO2 nanotubelike arrays electrodes w ere pre pared by treating T iO2 nanotubelike arrays electrodes in %%. XRD and XPS resu lts indicated that % %. References: [ 1] Eason G, Noble B, Sneddon I N. On certain integrals of Lipsch itz- H ankel type involv ing products of B essel func tions[ J]. Ph il T rans Roy Soc L ondon 1955, A 247: 529 - 55 1. Fig. 4 Princip le ofM ach- Zehnder in terferom eter based on d if [ 2] C lerk M axwell J. A T reatise on E lectricity and M agnetism fraction grating [ M ]. Oxford: C larendon, 1892. 68- 73. [ 3] Jacobs I S, Bean C P. F ine particles, thin film s and ex A s show ing in figure 4, the angle of incidence is change an isotropy[ J] . M agnetism, 1963, 271- 350. [ 3, 4] [ 4] E lissa K. T itle of paper if know n, unpub lished. 75?~ 85? , the w avelength is 13. 9nm, the angle from the 0 order to the - 1 order is about 3。4716?~ [ 5] N icoleR. T itle of paperw ith only first w ord capitalized[ J]. 7. 0869?. The length from m irror L1 to m irror L2 is J Nam e Stand Abbrev in press. [ 6] Y orozu Y, H irano M, Oka K, et al. E lectron spectroscopy 100mm. The ligh t length from theDFG to them irror is stud ies on m agneto - opticalm edia and plastic substrate in about 808. 99mm ~ 1650. 67mm, so the ligh t length terface[ J]. IEEE T ran sl JM agn Japan 1987, 2: 740 - from the DFG G 1 to the DFG G2 is about 16 18mm ~ 74 1. 3301mm. Introducing the parallelism of the DFG, the [ 6]M Y oung. T he T echn icalW riter sH andbook[ M ]. M ill V al prealign ing precision is about 0. 2094mm ~ 0. ley, CA: University Science, 1988. 4706mm. The pre - align ing precision in the position of G2 is about 0~ 0. 9412mm.