201105949 六、發明說明： 【發明所屬之技術領域】 本方法及裝置係關於光伏打薄膜特徵量測之領域，及特 定言之係關於包括吸收薄膜之光伏打薄膜堆疊物的參數之 量測。 【先前技術】 光伏打面板係一種環境友好型電能源，其係藉由吸收入 射太陽能輻射並將其轉化成光電流而運作。薄膜光伏打 (pv)面板呈現複數個在可撓性或剛性基板上沉積於彼此上 方之薄膜（TF)。某些薄膜係可收集由具有光伏打特性之薄 膜所產生之電流的接點。該TF PV面板藉由在某層或某些 層中吸收入射太陽能輕射並將其轉化成光電流而產生電 流。該等具有光伏打特性之薄膜層亦被稱為吸收薄膜或 層。某些形成該等接點之薄膜係透明的及某些係不透明的 或至少部份不透明。 為提供太陽能至光電流之高轉化效率及長期穩定性，鹿 使吸收薄膜之光學吸收在太陽能輻射或照明之光譜中最佳 化。 最佳化方法之一包括適當地設計吸收膜之能隙（EG)使大 約與太陽能光谱之長波長邊緣一致。（在本揭示案中，對 於有時在文獻中使用術語光學帶隙、能帶邊緣或吸收邊緣 之情況’吾人亦使用術語能隙）。此對於大部份包括非晶 矽、微晶矽及碲化鎘（CdTe)及三元及四元材料（諸如土晒 化銅铜（CIS)及·一砸化銅铜錄（CIGS)以及其他）之ργ·材料係 146632.doc 201105949 有關的會衫響eg之待控制之材料特性包括：化學計 量、晶相、結晶度、摻雜濃度、雜質濃度、陷阱濃度、粒 度Ba粒取向、晶粒形狀、晶界純化、空隙密度及其他。 可使用光譜量測（諸如反射率、透射率及光譜橢圓偏振 測量法）於分析該等吸收膜之特性。可使用薄膜及薄膜堆 疊物之光學模型來提供可擬合至量測光譜之相應的理論光 4。光學模型可係基於光傳播通過材料堆疊物之計算，其 中該等層體可經參數（諸如厚度、折射率及消光係數）及材 料特性中之界面、散射、空間梯度之效應等等加以定義。 通常進行擬合程序，藉此改變光學模型之參數，直到在 理論光譜與實際量測光譜之間達到高度一致性或匹配為 止。該理論光譜特性與實際量測的薄膜特性相對應。 可信賴的光譜量測需要特定量值的反射或透射照明，其 在堆疊物包括一或多個吸收或不透明膜時並非始終可得。 已認知需要一種包括至少一吸收或部份不透明膜之薄膜 堆疊物之特性的可信賴量測方法。 【發明内容】 本發明係關於一種用於量測薄膜光伏打面板之吸收層特 性之方法，其中該面板包括其中至少一薄膜為吸收膜之薄 膜堆疊H頻照明照在堆疊物之要發生特性量測之位 置上。一具有内建式檢測器之分光計量測及確定反射或透 射照明之光譜。薄膜參數係、藉由經量測的薄膜光譜與對應 於堆#物的光學模型及堆疊&的介電函數模型之薄膜參數 的關連而定義。 146632.doc 201105949 基於光譜量測，可建立堆疊物之至少某些層之介電函數 模型及將其用於製程控制。所測定之堆疊物參數可包括薄 膜特性，諸如能隙、能隙梯度.、結晶度、結晶度梯度、吸 收膜之吸收及吸收梯度。光譜量測可在光伏打面板之單元 區域、劃線區域及於薄膜堆疊物之至少一層中形成的特定 形成目標開孔上進行。 【實施方式】 為瞭解該裝置及方法及明白其可如何實踐，現將僅藉由 非限制性實例’參照附圖闡述若干示範性實施例。 圖1係典型薄膜堆疊物之示意圖。該等堆疊物係針對使 用薄膜的不同應用製造及特定言之係用於光伏打面板中。 面板100包括經沉積於透明或不透明基板108上以及經吸收 或部份不透明膜112塗覆之導電膜1〇4，例如金屬戍 tco(透明導電氧化物）。在光伏打面板中，該吸收薄膜 一般為半導體材料，如矽（si)、碲化鎘（CdTe)、二硒化銅 銦鎵（CIGS)或類似物。另一導電（TC〇或不透明金屬）膜ιΐ6 覆蓋該吸收膜112。層104與116之至少一層係透明的。此 典型光伏打薄膜結構被用於太陽能面板製造中。該吸收膜 112包含p-n接面及膜1〇4與116充當此等接面之導電接點。 圖2A-2C中描述一種典型的薄膜太陽能面板結構。面板 200—般係經由劃線區域202切割成被稱為單元且串聯連接 之多個條狀物208，藉此提供一組合高電壓輸出。此串聯 連接單元之方法通常被稱為單石積體。該等單元一般係藉 由在製造過程中於一定數量（一般為三個）之站處穿過某些 I46632.doc 201105949 層將材料劃線(即移除-狹窄條狀物)而形成。圖2B闡述一 劃線區域2〇2，其一般為一組通常標記為PI、P2及P3之三 個劃線2 0 4。圖2 C關Hi — ^ 聞尤一在元成该三個劃線之後的典型劃 線區域之截面圖，其顯示在接觸層104、吸收層112及接觸 層116中之切割，以使得其串聯連接個別單元。 光-曰里測(諸如反射率、透射率及光譜橢圓偏振量測法) 可用於定性分析該等吸收膜。該等光谱量測可作為控制材 料，儿積條件之方案之—部份實施，以維持高量子轉化效 率》該等受控制的沉積製程參數可為至少某些以下參數： 原材料動速率 '源材料流動比率、周圍溫度、周圍壓 力、基板溫度、基板移動速度、製程時間及其他。在多步 驟製程中，可控制上述參數及不同步驟的參數之間的關 係。舉例而言’在一類CIGS多步驟沉積製程中，其中藉由 添加第四種組分藉由擴散（硒）將三元CIG(銅銦鎵）材料轉化 為四元材料（CIGS)，需要控制沉積步驟時間與擴散步驟時 間之間的關係及溫度。 基於光譜量測，可建立該堆疊物之至少某些層的介電函 數模型並將其用於製程控制中。該介電函數模型可係基於 若干已知模型，諸如勞倫兹振盪器（L〇rentz 〇sciHat〇r)、 Tauc、Drude ' 有效介 f 近似appr〇ximaU〇n) 、Cauchy及其他或其組合0 可發展該等介電函數分散模型’以在寬廣波長範圍内準 喊地擬合層之反射率、透射率及光譜橢圓偏振儀信號。對 於藉此部份光譜完全被吸收於材料中之層，無法直接量測 146632.doc 201105949 全光譜範圍的介電函數。此係若干太陽能面板類型之情 況，因為該等層係經設計為基本上吸收該波長範圍的地面 太陽能輻射。 以下闡述在製造太陽能面板期間在吸收波長範圍内進行 "電函數之特性分析之方法。製備一組具有全範圍期望製 程變化之標稱厚度的太陽能面板樣品。可在至少某些以下 參數中發生變化：源材料流動速率、源材料流動比率、周 圍溫度、㈣壓力、基板溫度、基板移動速度、製程時間 及其他。當該材料可以比用於製造太陽能面板之標稱厚度 低的厚度沈積而不會明顯影響該等材料特性時，可製備第 ’、’蓴且基本上透明的樣品。此在整個所要求波長範圍内 至少部份透明之組係用於獲得一全範圍的纟電函數之更準 確的模型。此第二組可藉由以減少厚度沉積及/或藉由諸 如蝕刻或化學機械拋光(CMP)之方法控制材料之移除而製 備在校準過程中進行光譜量測，進行光學模型之擬合並 ，於所有樣σσ s十算介電函數。藉由將來自該兩組的相應樣 ⑽之介電函數匹配’可證實該介電函數對於全範圍製程變 的適用n。如此校準該等介電函數模型，以能夠在製造 期間藉由擬合非吸收波長範圍内之介電函數及執行介電函 數之外插至吸收波長範圍内，而於全吸收波㈣圍内控制 材料特性。 在材料特性與厚度相關的情況下，除標稱厚度組外再製 一土 透月材料之更薄的第二組。在其整個透明範圍内 疋性分析此二組。在校準過程中，進行光譜量測，進行光 146632.doc 201105949 學模型之擬合並計算所右 丹所有樣品之介電函數。基於在重疊非 吸收性範圍中之介雷说制_ 、一 電ώ數’製備關連厚組樣品與最接近樣 口口或缚組樣品的對癖矣+ ’心表或對應函數。若需要，可製備具有 額外製程變化的額外墟姐。 ^ 4樣。0以加寬覆蓋度，或者提供内 插使得標稱組之介電函數基本上完全延伸至吸收性波 長？圍内。因此校準該等介電函數模型，以能夠在製造期 間藉由在非吸收性波長範圍内擬合該介電函數及將該介電 “數卜插至及收性波長範圍内而於全吸收性波長範圍内控 制材料特性。 由於在太陽能面板製造中控制量子效率（QE)的重要性， 因而除先則所述之介電函數校準外，可使用一種獲得 的光谱相依性與吸收體層之吸收的光譜相依性之間的關係 的技術。此可用於製造中以基於在吸收體層沉積之後（即 在形成完整電可測試太陽能面板之前的大量製程步驟）所 進仃的光學量測提供關於所期望的製程末端（end-of-line) QE之資訊。 在吸收體層之吸收的光譜相依性與完整太陽能面板之量 子效率（QE)的光譜相依性之間存在一種關連。可能減少此 關連之因素包括：其他層中之吸收、光散射、光產生載體 之再結合及其他。吸收光譜之某些區域具有與QE較高的 關連’例如接近與EG相關的波長處《關連的減少亦可受 歸因於製程變化之吸收體層中之變化影響。囊括全範圍製 程變化之吸收體層中之變化可經由若干效應而引起關連之 變化’例如：隨後熱預算之不同效應、能帶偏移之變化、 146632.doc 201105949 界面狀態密度變化、層間擴散之變化及其他。 製備一組具有全範圍預期製程變化的太陽能面板樣品。 其後使用該等樣品（或等效組）於形成完成的操作性太陽能 電池，並對該QE之光譜相依性進行特性分析。對每個樣 品進行介電函數的特性分析，及若需要則如前所述將其 外插至吸收性波長範圍内，並獲得相應的光譜吸收相依 性。對於每個樣品，藉由計算QE對吸收層的吸收之比率 而測得波長相依關係。如此對於預期製程條件的範圍獲得 Q對吸收的波長相依比率。因此，可藉由以QE作為待最 佳化之目標來控制製程中之變化。 ，為使獨立的光譜資訊量最大化，及藉此增強介電函數模 ^'建立的準確性，較佳在不同的量測條件下進行大量的獨 立光4 Ϊ:測。此等包括諸如垂直入射透射率及反射率、斜 向入射透射率及反射率、及’或光譜橢圓偏振測量的量 測。亦可使用利用可變收㈣、可變數值孔徑及傅立葉 (oirner)平面空間渡波器之量測，其包括利用積分球之廣 角收集除此之外，可選擇寬頻照明以包括㈠列如㈣照 明、可見照明及IR照明之照明範圍或光學賴射範圍中的至 少一者。 在製程流程内進行所有此等量測有困難，因為許多TF pv製程係利用不透明基板或係㈣在支承透明基板上方之 不透明導電材料之初始塗層進行。 可藉由在最大預定間隔的位置附近進行量測而獲得歸因 ㈣同位置之通過不同量測而得之來自多個量測通道之資 146632.doc 201105949 訊。歸因於TF PV製程的典型大變化長度尺度，此可為十 毫米量級的間隔，即單元寬度之尺度。因此，可利用不同 的光學系統進行不同類型的量測及隨後組合量測進行分 析。 在利用於支承透明基板上方之不透明導體的初始薄膜塗 層的製程中’可藉由形成無不透明材料的區域而進行經沉 積於不透明層上之吸收體層的可信賴透射率量測。在一示 範性實施例中’此可藉由在經形成為用於分離太陽能面板 之單元的劃線或劃線區域中量測薄膜特性，而不利用額外 的製造程序或操作步驟達到。圖3係在於接觸層中形成劃 線3 04之後經’’儿積於接觸層3 〇8之頂部上的吸收薄膜3 12之 不意圖。此劃線一般係稱為ρι，於某些情況下該接觸層係 由不透明材料組成。此劃線3〇4係相對窄的條狀物，其中 導體308係在吸收體312薄膜沉積之前移除、可在移除導體 層及將吸收層直接沉積於透明基板3 16上之後直接在劃線 中進订量測。舉例而言，在其中初始將充當導體層3〇8之 ㈣厚『透明層沉積於玻璃基板上之CK5S太陽能面板製程 中’㈣劃線移除I目層’而允許進行隨後沉積之CIGs吸收 膜的透射率量測。 在自勺400 nm至最多至少1000 nm的波長範圍内於其中 層隹且物係於鉬上之CIGS的單元區域内進行反射率量 測。可在其中之層堆曼物係於玻璃316上之CIGS的劃線3〇4 上進行其他量測。此第二量測可為反射率或透射率，且可 在於單元區域内之㈣中開孔之量測目標上進行。在 146632.doc 201105949 施例中，一具有位於面板上方的光學系統及大於劃線寬度 的光斑尺寸及約400 nm至約1700 nm的波長範圍之裝置在 單元區域及劃線上進行個別的反射率量測。該劃線信號藉 由以下所述之權重方法自單元信號中分離出。此兩個別的 單元及劃線層堆疊物藉由具有共同層但不同基板的兩種光 千模型擬合’猎此k供在800 nm以上的非吸收範圍中CIGS 層堆疊物之增強資訊。 邊方法可應用於在製程期間分析非晶石夕（a Si)之EG、吸 收及厚度之特性，其中在圖3B中，吸收體層33〇係沉積於 透明導電氧化物（TCO)3 3 4上並在某種程度上重複其沉積於 其上之層之分佈。在許多製程中，使用具有高度表面粗糙 度338的TCO，其亦造成隨後沉積於其上之層之粗糙度。 该粗糙度會引起光散射，藉此扭曲透射、反射及光譜橢圓 偏振篁測中的光學信號。此扭曲抑制對a_si吸收體層之光 學參數準確建立模型及量測之能力。在已移除TC〇之劃線 342上進行量測可自a_Si層獲得更清晰及更具代表性的信 號，藉此可更準確地量測該層之特性。類似地，在TC〇層 中實施量測目標可獲得更清晰的量測。 可在具有大量測光斑之劃線上進行量測，藉此藉由以下 所述之權重量測方法將該劃線信號自該單元信號分離。在 某些情況下，可使用考慮由該層粗糙度所引起的散射之近 似光學模型於在粗糙區域上進行直接量測。藉由權重量測 方法自劃線分離信號可增強該層特性分析的量測準確度。 為特性分析該a-Si吸收體層，藉由具有共同層但不同基板 146632.doc 201105949 及粗糙度效應之兩個光學模型來擬合兩個別的單元與劃線 層堆疊物，額外的量測資訊藉此提供該a_si層特性之增強 的準確度。此量測可藉由在約400 nm高至約10〇〇 nm之波 長範圍中的反射率、透射率或光譜橢圓偏振量測進行。當 可在製程中實施時，可以於單元區域之TCO層中開孔的量 測目標替代劃線量測。直接在劃線上量測之優勢在於能夠 不需在製程中實施變化下進行特性分析。 在圖4A及4B中所示之另外的實施例中，可橫跨該太陽 能面板於預定位置處形成特定的接觸材料移除區域（量測 目標）。該材料係在劃線製程步驟期間且在吸收體層沉積 之前通常藉由雷射進行移除。隨後將此等目標塗覆吸收體 材料’並可進行包括透射率的光譜量測》 實施量測目標可有利於不透明與透明導電層二者。TC〇 層具有分散入射及透射照明且負面影響量測結果之粗糙表 面。藉由移除該等膜所形成的量測目標可減少表面粗糙度 影響並可在隨後沉積之層上進行更高品質的量測。 目標404之區域比單元208之區域小，其不會明顯影響該 單元之輸出。因此，若該目標404之區域相對於單元2〇8之 區域而言小，在某種水平上小於太陽能面板中不同單元之 光電流的統計學變化時，則太陽能面板能量效率之降級將 可忽略。目標404之形狀可為（例如）經封圍在單元内之正方 形島狀物《或者，其可作為劃線之橫向延伸添加至於隨後 製程步驟中未添加額外劃線之面上。目標404之大小可遠 大於劃線204之寬度，因此放寬對光學系統之光斑尺寸及 146632.doc •12- 201105949 對準的要求。 在某些情況下，製程並不允許該等量測目標的實施（例 如若使用基於針狀物之機械劃線器），則可在劃線中進行 透射率量測《為減少劃線之邊緣及底面的可能粗糙度的影 響，可在沿該線的若干位置中量測透射信號並取平均值。 或者可利用大照明光斑在劃線區域上進行量測，藉此收集 透射信號。 ^ 圖5係於層508中製造之示範劃線之示意圖。若劃線5〇4 之底面500由於（例如）由脈衝雷射劃線所引起的基板516中 之凹口 512而係不平坦（如圖5所示），則沿劃線5〇8之光譜特 性可大幅度改變。-般而言，沿著該劃線之特性可隨著劃 線雷射之每個步進52〇變化及該等變化可係沿著及橫跨該 劃線（該劃線之寬度與厚度二者）。若位於劃線中之薄層的 厚度【化係為大於數十奈米的量級，則無論係藉由反射、 透射或其他類型的光譜信號量測，來自高及低區域之光譜 °最大值的位置中具有顯著差異。利用足夠大以包含該201105949 VI. Description of the Invention: [Technical Field of the Invention] The method and apparatus relate to the field of photovoltaic filming feature measurement, and in particular to the measurement of parameters of a photovoltaic film stack comprising an absorbing film. [Prior Art] A photovoltaic panel is an environmentally friendly electric energy source that operates by absorbing solar radiation and converting it into photocurrent. Thin film photovoltaic (pv) panels exhibit a plurality of films (TF) deposited on top of each other on a flexible or rigid substrate. Some films collect joints of current generated by a film having photovoltaic characteristics. The TF PV panel generates current by absorbing incident solar light in a layer or layers and converting it into photocurrent. These thin film layers having photovoltaic properties are also referred to as absorbent films or layers. Some of the films forming the contacts are transparent and some are opaque or at least partially opaque. To provide high conversion efficiency and long-term stability of solar to photocurrent, deer optimizes the optical absorption of the absorbing film in the spectrum of solar radiation or illumination. One of the optimization methods involves appropriately designing the energy gap (EG) of the absorbing film to conform approximately to the long wavelength edge of the solar spectrum. (In the present disclosure, the term "energy gap" is also used in the case where the term optical band gap, band edge or absorption edge is sometimes used in the literature. This is mostly for amorphous germanium, microcrystalline germanium and cadmium telluride (CdTe) and ternary and quaternary materials (such as soil copper (CIS) and copper and copper (CIGS) and others. Ργ·Materials 146632.doc 201105949 The material properties to be controlled of the singularity include: stoichiometry, crystal phase, crystallinity, doping concentration, impurity concentration, trap concentration, grain size, grain orientation, grain size Shape, grain boundary purification, void density and others. Spectral measurements, such as reflectance, transmittance, and spectral ellipsometry, can be used to analyze the properties of the absorbing films. An optical model of the film and film stack can be used to provide a corresponding theoretical light 4 that can be fitted to the measurement spectrum. Optical models can be based on the calculation of light propagation through a stack of materials, which can be defined by parameters such as thickness, refractive index, and extinction coefficient, as well as interfaces, scattering, spatial gradient effects, and the like in the material properties. A fitting procedure is usually performed whereby the parameters of the optical model are changed until a high degree of agreement or match between the theoretical spectrum and the actual measured spectrum is achieved. This theoretical spectral characteristic corresponds to the actual measured film properties. Reliable spectral measurements require a specific amount of reflected or transmitted illumination that is not always available when the stack includes one or more absorbing or opaque films. There is a known need for a reliable measurement method that includes the characteristics of a thin film stack of at least one absorbing or partially opaque film. SUMMARY OF THE INVENTION The present invention is directed to a method for measuring the characteristics of an absorber layer of a thin film photovoltaic panel, wherein the panel includes a film stack in which at least one film is an absorber film. The location of the measurement. A spectroscopic measurement with a built-in detector and the determination of the spectrum of the reflected or transmitted illumination. The film parameters are defined by the correlation of the measured film spectrum with the film parameters corresponding to the optical model of the stack and the dielectric function model of the stack & 146632.doc 201105949 Based on spectral measurements, a dielectric function model of at least some of the layers of the stack can be created and used for process control. The measured stack parameters may include film properties such as energy gap, energy gap gradient, crystallinity, crystallinity gradient, absorption of the absorption film, and absorption gradient. Spectral measurements can be made on the cell area of the photovoltaic panel, the scribe line area, and the particular target opening formed in at least one of the film stacks. [Embodiment] To understand the apparatus and method, and to understand how it can be practiced, several exemplary embodiments will now be described by way of non-limiting example. Figure 1 is a schematic representation of a typical film stack. These stacks are manufactured for different applications using films and in particular for use in photovoltaic panels. The panel 100 includes a conductive film 1〇4, such as a metal tantalum tco (transparent conductive oxide), deposited on a transparent or opaque substrate 108 and coated by an absorbing or partially opaque film 112. In photovoltaic panels, the absorbing film is typically a semiconductor material such as bismuth (si), cadmium telluride (CdTe), copper indium diselenide (CIGS) or the like. Another conductive (TC 〇 or opaque metal) film ι 6 covers the absorbing film 112. At least one of layers 104 and 116 is transparent. This typical photovoltaic film structure is used in solar panel manufacturing. The absorbing film 112 includes a p-n junction and the films 1 〇 4 and 116 serve as conductive contacts for the junctions. A typical thin film solar panel structure is depicted in Figures 2A-2C. Panel 200 is typically cut through scribe region 202 into a plurality of strips 208, referred to as cells, connected in series, thereby providing a combined high voltage output. This method of connecting the units in series is commonly referred to as a single stone. These units are typically formed by scribing (i.e., removing - narrow strips) the material through a number of I46632.doc 201105949 layers at a number (usually three) of stations during the manufacturing process. Figure 2B illustrates a scribe region 2〇2, which is typically a set of three scribe lines 2 0 4, generally designated PI, P2, and P3. Figure 2 C is a cross-sectional view of a typical scribe line region after the three scribe lines, which shows the cut in the contact layer 104, the absorbing layer 112 and the contact layer 116, so that they are connected in series. Connect individual units. Light-in-the-middle measurements (such as reflectance, transmittance, and spectral ellipsometry) can be used to qualitatively analyze such absorbing films. These spectral measurements can be used as control materials, part of the implementation of the condition to maintain high quantum conversion efficiency. These controlled deposition process parameters can be at least some of the following parameters: Raw material rate 'source material Flow ratio, ambient temperature, ambient pressure, substrate temperature, substrate movement speed, process time, and others. In a multi-step process, the relationship between the above parameters and the parameters of the different steps can be controlled. For example, in a CIGS multi-step deposition process, where a ternary CIG (copper indium gallium) material is converted to a quaternary material (CIGS) by diffusion (selenium) by adding a fourth component, controlled deposition is required. The relationship between the step time and the diffusion step time and the temperature. Based on spectral measurements, a dielectric function model of at least some of the layers of the stack can be created and used in process control. The dielectric function model can be based on several known models, such as the Lorentz oscillator (L〇rentz 〇sciHat〇r), Tauc, Drude 'effective media f appr〇ximaU〇n), Cauchy and others or combinations thereof 0 These dielectric function dispersion models can be developed to fit the reflectivity, transmittance, and spectral ellipsometer signals of the layer in a wide range of wavelengths. For the layer in which part of the spectrum is completely absorbed in the material, the dielectric function of the full spectral range of 146632.doc 201105949 cannot be directly measured. This is the case for several solar panel types because the layers are designed to substantially absorb ground solar radiation in this wavelength range. The method of performing the characteristic analysis of the electric function in the absorption wavelength range during the manufacture of the solar panel is explained below. A set of solar panel samples having a nominal thickness of the full range of desired process variations was prepared. Changes can occur in at least some of the following parameters: source material flow rate, source material flow ratio, ambient temperature, (iv) pressure, substrate temperature, substrate movement speed, process time, and others. When the material can be deposited at a lower thickness than the nominal thickness used to fabricate the solar panel without significantly affecting the material properties, the first, '莼, and substantially transparent sample can be prepared. This at least partially transparent set of wavelengths throughout the required wavelength range is used to obtain a more accurate model of the full range of tantalum functions. This second set can be used to perform spectral measurements during the calibration process by controlling the removal of the material by reducing the thickness deposition and/or by means such as etching or chemical mechanical polishing (CMP), and fitting the optical model and , calculate the dielectric function for all samples σσ s. The dielectric function can be verified to be applicable to the full range process by matching the dielectric function of the corresponding sample (10) from the two groups. The dielectric function models are calibrated in such a way that they can be controlled within the fully absorbed wave (four) by fitting a dielectric function in the non-absorbed wavelength range and performing a dielectric function outside the absorption wavelength range during fabrication. Material properties. In the case of material properties related to thickness, a thinner second set of soil-permeable materials is produced in addition to the nominal thickness set. The two groups were analyzed in the entire transparent range. During the calibration process, spectral measurements were taken, the light was fitted to the model and the dielectric function of all samples of the right Dan was calculated. A pair of 癖矣+ 'heart tables or corresponding functions of the associated thick sample and the closest sample or set of samples are prepared based on the number of enthalpy in the overlapping non-absorbent range. Additional cashiers with additional process variations can be prepared if needed. ^ 4 samples. 0 to widen the coverage, or to provide interpolation so that the dielectric function of the nominal set extends substantially completely to the absorptive wavelength? Inside. The dielectric function models are therefore calibrated to enable full-absorption during the manufacturing process by fitting the dielectric function in the non-absorptive wavelength range and inserting the dielectric into the range of wavelengths. Controlling material properties in the wavelength range. Due to the importance of controlling quantum efficiency (QE) in solar panel fabrication, an acquired spectral dependence and absorption of the absorber layer can be used in addition to the dielectric function calibration described earlier. A technique for the relationship between spectral dependencies. This can be used in manufacturing to provide the desired information based on optical measurements taken after deposition of the absorber layer (ie, a number of process steps prior to forming a fully electrically testable solar panel). End-of-line Information on QE There is a correlation between the spectral dependence of the absorption in the absorber layer and the spectral dependence of the quantum efficiency (QE) of the complete solar panel. Factors that may reduce this association include: Absorption in other layers, light scattering, recombination of light-generating carriers, and others. Some regions of the absorption spectrum have a high correlation with QE. For example, near the wavelength associated with EG, the reduction in correlation can also be affected by changes in the absorber layer due to process variations. Variations in the absorber layer that encompass the full range of process variations can cause related changes via several effects, for example. : Subsequent thermal budget effects, changes in band offset, 146632.doc 201105949 Interface state density changes, interlayer diffusion changes, and others. Prepare a set of solar panel samples with full range of expected process variations. The sample (or equivalent group) is formed into a completed operational solar cell, and the spectral dependence of the QE is characterized. A characterization of the dielectric function is performed for each sample, and if necessary, as described above. Extrapolation into the absorption wavelength range and obtaining the corresponding spectral absorption dependence. For each sample, the wavelength dependence is measured by calculating the ratio of QE absorption to the absorption layer. Thus obtaining Q for the range of expected process conditions The wavelength dependence ratio of the absorption. Therefore, the change in the process can be controlled by using QE as the target to be optimized. In order to maximize the amount of independent spectral information and thereby enhance the accuracy of the establishment of the dielectric function module, it is preferred to perform a large number of independent light measurements under different measurement conditions. Vertical incident transmittance and reflectivity, oblique incident transmittance and reflectivity, and 'or spectral ellipsometry measurements. Variable-receiving (four), variable numerical aperture, and oirner planar space ferrites can also be used. The measurement includes wide-angle collection using an integrating sphere. In addition, the wide-band illumination may be selected to include (a) at least one of (b) illumination, visible illumination, and IR illumination illumination range or optical range. It is difficult to perform all of these measurements in the process because many TF pv processes are performed using an opaque substrate or system (4) on the initial coating of an opaque conductive material overlying the transparent substrate. Attribution can be obtained by measuring near the position of the maximum predetermined interval. (4) The same position is obtained from different measurement channels by different measurements. 146632.doc 201105949. Due to the typical large variation length scale of the TF PV process, this can be an interval of the order of ten millimeters, i.e., the dimension of the cell width. Therefore, different optical systems can be used for different types of measurements and subsequent combined measurements for analysis. In the process of utilizing an initial film coating for supporting an opaque conductor over a transparent substrate, a reliable transmittance measurement of the absorber layer deposited on the opaque layer can be performed by forming a region free of opaque material. In an exemplary embodiment, this can be achieved by measuring film characteristics in a scribe line or scribe line region formed as a unit for separating solar panels, without the use of additional manufacturing procedures or operational steps. Figure 3 is a schematic view of the absorbing film 3 12 which is deposited on the top of the contact layer 3 〇8 after the scribe line 3 04 is formed in the contact layer. This scribe line is generally referred to as ρι, which in some cases consists of an opaque material. The scribe line 3〇4 is a relatively narrow strip in which the conductor 308 is removed prior to film deposition of the absorber 312, and can be directly drawn after the conductor layer is removed and the absorber layer is deposited directly on the transparent substrate 316. Ordering in the line. For example, a CIGs absorbing film that allows for subsequent deposition in a CK5S solar panel process in which a (four) thick "transparent layer" of a conductor layer 3〇8 is deposited on a glass substrate is initially used. Transmittance measurement. Reflectance measurements were made in the cell region of the CIGS in which the layer was germanium and on the molybdenum in the wavelength range from 400 nm up to at least 1000 nm. Other measurements can be made on the scribe line 3〇4 of the CIGS in which the layer of the manned material is attached to the glass 316. This second measurement can be reflectance or transmittance and can be performed on the measurement target of the opening in (4) in the cell region. In the example of 146632.doc 201105949, an apparatus having an optical system located above the panel and a spot size larger than the scribe line width and a wavelength range of about 400 nm to about 1700 nm performs individual reflectance amounts on the cell area and the scribe line. Measurement. The scribe signal is separated from the unit signal by a weighting method as described below. The two other cell and scribing layer stacks are fitted by two light-mode models with a common layer but different substrates to provide enhanced information for the CIGS layer stack in the non-absorption range above 800 nm. The edge method can be applied to analyze the characteristics of EG, absorption and thickness of amorphous silicon (a Si) during the process, wherein in FIG. 3B, the absorber layer 33 is deposited on transparent conductive oxide (TCO) 3 3 4 And to some extent repeat the distribution of the layers on which it is deposited. In many processes, a TCO having a high surface roughness 338 is used which also causes the roughness of the layer subsequently deposited thereon. This roughness causes light scattering, thereby distorting the optical signals in transmission, reflection, and spectral ellipsometry. This distortion suppresses the ability to accurately model and measure the optical parameters of the a_si absorber layer. Measurements on the removed TC scribe line 342 provide a clearer and more representative signal from the a_Si layer, thereby more accurately measuring the characteristics of the layer. Similarly, a more realistic measurement can be obtained by implementing a measurement target in the TC layer. The measurement can be performed on a line having a large number of light spots, whereby the line signal is separated from the unit signal by the weight measurement method described below. In some cases, a near optical model that takes into account the scattering caused by the roughness of the layer can be used for direct measurement on the roughened area. Separating the signal from the scribe by weight weighting method enhances the measurement accuracy of the characterization of the layer. For the characterization of the a-Si absorber layer, two additional units and the scribe layer stack were fitted by two optical models with a common layer but different substrates 146632.doc 201105949 and roughness effects, additional measurements The information thereby provides an enhanced accuracy of the a_si layer characteristics. This measurement can be performed by reflectance, transmittance, or spectral ellipsometry measurements in the wavelength range from about 400 nm up to about 10 〇〇 nm. When it can be implemented in the process, the measurement target of the opening in the TCO layer of the unit area can be used instead of the scribe line measurement. The advantage of measuring directly on the line is the ability to perform characterization without the need to implement changes in the process. In still other embodiments shown in Figures 4A and 4B, a particular contact material removal area (measurement target) can be formed at a predetermined location across the solar panel. The material is typically removed by laser during the scribing process step and prior to deposition of the absorber layer. Subsequent application of such targets to the absorber material' and spectral measurements including transmission can be performed to achieve a measurement target that can facilitate both opaque and transparent conductive layers. The TC layer has a rough surface that disperses incident and transmitted illumination and negatively affects the measurement results. By measuring the target formed by the removal of the films, surface roughness effects can be reduced and higher quality measurements can be made on subsequently deposited layers. The area of target 404 is smaller than the area of unit 208, which does not significantly affect the output of the unit. Therefore, if the area of the target 404 is small relative to the area of the unit 2〇8 and is less than a statistically significant change in the photocurrent of the different elements in the solar panel at a certain level, the energy efficiency degradation of the solar panel will be negligible. . The shape of the target 404 can be, for example, a square island enclosed within the unit. Alternatively, it can be added as a lateral extension of the scribe line to the surface where no additional scribe lines are added in subsequent processing steps. The size of the target 404 can be much larger than the width of the scribe line 204, thus relaxing the spot size of the optical system and the alignment requirements of 146632.doc • 12-201105949. In some cases, the process does not allow the implementation of the measurement targets (for example, if a needle-based mechanical scriber is used), the transmittance measurement can be performed in the scribe line. And the effect of the possible roughness of the bottom surface, the transmission signal can be measured and averaged over several locations along the line. Alternatively, a large illumination spot can be utilized to measure over the scribe region, thereby collecting the transmitted signal. ^ Figure 5 is a schematic illustration of an exemplary scribe line made in layer 508. If the bottom surface 500 of the scribe line 5 〇 4 is not flat due to, for example, the notch 512 in the substrate 516 caused by the pulsed laser scribe line (as shown in FIG. 5 ), the spectrum along the scribe line 5 〇 8 Features can vary dramatically. In general, the characteristics along the scribe line can vary with each step of the scribe line laser and the change can be along and across the scribe line (the width and thickness of the scribe line) By). If the thickness of the thin layer in the scribe line is on the order of more than tens of nanometers, the spectral maximum from the high and low regions is measured by reflection, transmission or other type of spectral signal measurement. There are significant differences in the location. Use large enough to contain this
等變化厚度之區域的I 匕坺的先斑尺寸之量測可導致在量測信號中 由於來自不同區域的彳士 ^的彳5唬之平均而使預期的光譜震盪減弱 或者甚至消除。 使用小光斑尺寸όΓ ;隹> π ^广 寸了進仃不同厚度區域之局部採樣。由面 板相對於光學系# β、* Λ* Μ、，，之連續移動，因此需要相當短的量測採 樣時間來限制以上的平 .b| + 十均化效應。該量測可藉由多個短量 測（可他使用閃控照明I · 、、 進仃’其隨後經分離成類似特徵之 群。信號雜訊之作號 。虎了#由在每組内平均而改善。 146632.doc 201105949 方法所而之光斑尺寸約為劃線寬度或更小。可將此等作號 分組（諸如）以在劃線内提供關於最大及最小厚度之資訊， 或者若該光斑尺寸夠小，則提供關於劃線之底面之輪廓的 半連續資訊。 m 一般而言’劃線寬度係在25與50微米之間，因此1〇至2〇 微米的光斑尺寸可適用於劃線内的量測。若使用設定為 1 〇〇 Hz的重複率及丨微秒的有效脈衝長度之氙氣閃光燈（例 如Hamamatsu L9455系列），且該太陽能面板以3米/分鐘的 速度移動，則由於相對運動而產生之光斑尺寸的漏光將為 可忽略的50奈米以及沿著劃線之量測間的間隔可為5〇〇微 米。 或者，可應用光學移動系統以補償樣品相對於光學系統 的相對移動。可將該光學系統置於移動系統上或可使用任 何技藝中已知的將在配合面板移動的方向掃描的光學掃描 方法此可Γ/東結」影像並藉此可在任何預定位置上有效 地進行靜態量測。移動之範圍係由量測時間需求界定。基 於以上實例，對於10毫秒的量測時間將需要0.5 mm的掃描 範圍。 圖6A係利用大照明光斑6〇4量測薄膜參數的本發明量測 方法之不範性實施例的示意圖。在此實施例中，經接收或 檢測之信號強度係藉由將照明光斑6〇4特定成形為在平行 於劃線204(圖2)的方向上排成一線的長條而增強。該照明 光斑係經配置為與劃線具有相對大的重疊及增加透射信 號。該條狀照明光斑亦可用於對透射及反射量測二者量測 146632.doc •14- 201105949 單元區域。為使光學系統簡單化以及消除主動調整量測位 置的橫向佈置之需求’可將光斑604寬度界定為劃線2〇4寬 度之特定倍數，其取決於劃線位置的橫向對準變化。該橫 向對準變化係受劃線204之位置相.對於太陽能面板邊緣之 變化及太陽能面板相對於光學系統之橫向佈置的變化影 響。若此相對於光斑寬度之變化較大，則光學系統可能需 在量測之前進行光斑位置相對於太陽能面板的橫向對準， 以提供光斑與劃線的適宜重疊。為改善自劃線量測得的信 號品質，當來自單元區域之信號不可忽略時，可僅在單元 區域内進行額外的量測。在藉由光斑尺寸内之單元對劃線 區域之比率正確權重該單元量測之後，可將其自劃線量測 中減去以及將該信號再正規化以獲得純劃線信號。在透射 量測的情況下，若單元區域係至少部份透射則此係有用 的。此技術可藉由反射量測用於大多數情況。 在另一示範性實施例（圖6B)中，可使用尺寸至少為單元 寬度之大照明光斑608，藉此用於收集透射光之光學元件 可自至少一單兀寬度的大區域收集光，由此可經由至少兩 條劃線進行透射量測。使用較大的光斑尺寸可簡化光學系 統並減少主動橫向移動量測位置的需求。 用於透射量測之光學系統之進一步簡化可藉由使用單一 廣區域照明(圖7)系統702而達成’該系統沿著太陽能面板 704的整個寬度照明條狀物，同時將一系列的收集光學元 件系統708没置於面板之相對側上，其中每個收集器自少 數劃線接收其信號。亦可使用更小數量的收集光學元件系 146632.doc -15- 201105949 統，以及在進行量測的同時將其沿著照明區域掃描。 圖8係圖2B之放大圖及係劃線區域之示意圖，其顯示在 不同位置之量測及不同特徵對測量信號的權重貢獻，以使 得權重量測方法能夠分開該等特徵的貢獻，其中術語特徵 可包括至少某些單元區域，P1劃線、p2劃線及p3劃線。量 測係在製程之稍後階段在於劃線區域中形成兩個或更多劃 線（例如用於隔離接觸層i 〇4之p丨以及用於分離吸收層i丄2 區域之P2)之後進行。為自單元之信號分離劃線之信號及 亦使劃線信號之間分離，需進行至少兩個及經常甚至多於 兩個量測。若光斑尺寸之邊緣的過渡明確且小於單一劃線 區域内不同劃線之間的距離，則在單元（s〇)、單元及第— 劃線P1(量測信號S0*(l-a)+Sl*a)及在單元在第—ρι與第二 P2劃線二者上（量測信號s〇*(l-b-c)+Sl*b+S2*c)上進行多 個量測，其中SO、S1及S2分別係該單元、p】、及？2之純量 測信號，以及a、b或c係決定每個區域對信號的相對貢獻 之係數。 根据測量光斑尺寸隨波長之變化，係数a、^及c可隨波 長而變化。可進行額外的量測以自不同的特徵掏取不同比 率的信號。若由單元區域貢獻給量測光譜之信號可忽略不 計，例如歸因於由粗糙表面所引起的強散射或在其中之單 元區域係高度吸收性之透射量測中’則僅需要後者的劃線 量測6若在光斑邊緣之強度過渡並不急劇（即大於劃線之 間的距離），則需要藉由在其等之間垂直於劃線之受控移 動進行一系列量測。 146632.doc -16* 201105949 此糸列之量測應包括至少—個僅單元之量測（量測㈣ s〇)、-具有來自第-劃線之部份貢獻之量測（量測信號 s〇*(1，a)+sl*a)、_具有來自第—劃線之部份貢獻及來自 第—劃線之部份貢獻之量測（量測信號s〇*(1_b_c)+sl*b+S2*c) 。可進行額外的量測以自不同的特徵操取不同比率的作 號。基於量測點之相對橫向位置的資料，可將該等量測如 圖中所述擬合至與兩劃線之貢獻迴旋之橫向光斑強度分 佈之函數。若該光斑輪廓對於不同的波長係不同的，則此 可針對光譜的每個波長分別進行。可分別藉由不同的方式 (諸如掃描刀口或刃口鏡或窄縫)測定此光斑橫向光譜強度 分佈。 在擬合劃線貢獻至迴旋信號後，可將此等貢獻分離並分 析其各別的光譜特性，以擬合光學模型並計算每個劃線中 之各別層堆疊物之參數。此方法可使用反射、透射或光譜 橢圓偏振量測及其組合進行。 透射里測亦可在光學系統與太陽能面板之間的相對移動 』門於劃線甲進行。在—實施例中，纟中面板係置於連續 移動的輸送器上，該光學系統係對準於劃線上方且該面板 係在基本上平行於劃線的方向移動。如此，可延長量測時 間，藉此改善信號對雜訊比以及平緩劃線中可能局部粗糙 度之效應。可將該光學系統在製造線中預對準以重疊一特 定之劃線。可設置多個光學系統以同時橫跨該面板的寬度 取樣多個劃線’而提供映射能力及可控制製程的橫向均勻 度。在另一實施例中，可將單一光學系統對準以在單一劃 146632.doc •17· 201105949 線^進行量測，及以等於單元寬度之倍數的步進移動，藉 此叙跨面板寬度連續取樣多個劃線以提供劃線映射能力。 在某些條件下，太陽能面板相對於移動方向之旋轉大，可 月匕*要主動橫向對準量測光斑位置之系統來維持量測光斑 與劃線之適宜重疊。若使用多個光學系統，則其可同時移 動以追蹤劃線並維持量測光斑之適宜重疊。可將該等多個 光子系統附接至一共同的機械介面以允許使用單一運動機 制的。 在薄膜係經沉積於不透明基板上之薄膜光伏打面板製程 中…法進行透射量測。在此情況可藉由在不同的入射角 量則反射率及/或光譜橢圓偏振量測而獲得增強的資訊。 亦可藉由在該單元區域内及在劃線中進行該吸收體層之個 別的反射率量測而在該層上添加資訊。此在含有效不同基 ，之層堆#物上提供兩組獨立的量測。可在劃線步驟期間 實施$測目標（如前所述），藉此可放寬對於在劃線層堆疊 物上量測之照明光斑尺寸及對準的要求。 圖9係裝備有用於量測薄膜光伏打面板參數的本發明系 統之典型薄膜製造線的示意圖。纟層係在適宜的沉積工作 站/儿積且輸送器9〇〇在該等站之間推進面板2〇〇(圖2)。繼吸 收體層沉積之後，太陽能面板2〇〇通常沿著輸送器系統9〇〇 刖進並通過由多個光學系統（藉由916及92〇示例）組成之裝 置904。可將該等光學系統設置於該面板之上、該面板之 下及面板之上與之下。該等光學系統係經預對準以使其設 置於移動面板900的特定區域之上及/或之下。可定位至少 146632.doc 201105949 某些光學系統（例如系統916)，藉以使其在單元208區域中 進行量測’可將某些系統（例如系統92〇)定位於劃線上以及 可將某些定位以在量測目標4〇4(圖4)上進行量測。可操作 視需要選用之一或多個位置控制模組9〇8，以提供橫向及 縱向定位移動（如箭頭91〇及912所示）以補償光學系統相對 於面板900及劃線204之位置之定位。可進一步操作位置控 制模組908以提供旋轉並補償光學系統相對於面板9〇〇及劃 線204之位置之定位。該定位模組可旋轉整個裝置9〇4或分 別旋轉每個光學系統9丨6及92〇。一或多個檢測器928感測 相對於光學系統916及920取向的面板9〇〇之取向。控制器 932控制系統之所有單元的操作並使其同步化。 圖1 〇係闡述利用本發明系統之薄膜光伏打面板參數量測 方法的示範性製程之流程圖。#接收到面板接近的輸入信 號時（方塊_)，裝置開始量測順序。面㈣達之信號係 自輸送器㈣_接收，或係由置於該等光㈣統上游之 特定檢測器928產生。檢測器928感測相對於標稱位置及標 稱角度之面板_的取向(方塊1〇〇4)、及劃線2〇4之橫向偏 移及角度(方塊1〇〇8)。此訊息經發送至系統控制器，該控 制Is計算校正動作及發送校 仪彳0唬至控制多個光學系統之 心向位置的定位模組0該定位 疋位楨、、且校正多個光學系統之橫 向位置（方塊1012)，使得經預定A 、 s θ 疋马在特疋特徵（諸如劃線、 早几或置測目標）上進行量測 ,..,...Ώ # 尤予糸統對各別特徵適宜 ““ 偏移以追蹤該等特徵之可能對 角移動。控制器932使基於預定量 $叫δ十畫之不同光學系統 146632.doc •19· 201105949 916及920之量測同步化（方塊1〇16)。視需要，檢測器928亦 可經定位以能夠在移動面板上檢測量測目標404之存在及 縱向位置’隨後將信號（除該面板取向資訊外）發送至控制 器932 °該控制器計算在量測目標上進行量測所需之計 時’及其後所需的光學系統經適宜啟動以在目標上進行同 步量測（涵蓋於方塊1 〇 16中）。 在沿著面板長度之多個位置處重複量測以產生面板特性 之圖像。來自所有量測通道之資料經控制器接收（方塊 1020)並經轉移至一資料系統。將來自一預定距離（通常類 似於單元寬度）内的多個位置之量測組合成一延伸資料 組。藉由將該組之每個量測與適宜光學模型比較而分析該 延伸資料組。每個光學模型係由至少一個薄膜或薄膜層堆 疊物組成，其係用於計算經擬合至量測光譜之相應的理論 光谱。光學模型係基於光傳播通過材料堆疊物的計算，其 中该等層係藉由參數（諸如厚度、折射率及消光係數）以及 材料特性中之界面、散射、空間梯度效應及其他而界定。 進行擬合程序，藉此改變該光學模型之參數直至在理論 光谱與實際量測光譜之間達到高度一致性或匹配為止。該 理論光譜特性係對應於實際量測的薄膜特性。 可將對至少吸收體層之薄膜特性所測得的資訊連通至設 置於上游及下游的製造設備，且其中該等連通參數使得可 對吸收層形成製程進行控制。所控制的製程參數可為至少 某些以下參數：源材料流動速率、源材料流動比率、周圍 皿度周圍壓力、基板溫度、基板移動速度、製程時間及 146632.doc -20. 201105949 其他。 為利於利用多個變化參數量測層及層堆疊物，需收集盡 可能多的獨立資訊。擴大所收集光之波長範圍可提高分離 ' ^中不同參數變化之效應的能力。組合紫外線、可見及紅 料三個波長範圍中的至少兩個可有利於此種增強。舉例 而。，組合 VIS-NIR(400 nm-i000 nm)&IR(95〇 nml7〇〇 紐)之光4感測器可產生涵蓋太陽能面板之操作範圍以及 其中該等材料變得透明之更低能量IR範圍之{學系統。 該等用於增強太陽能面板之QE的技術之一係基於藉由 控制該等材料特性之深度相依性的變化而設計材料。舉例 而s ，在沉積期間改變材料之化學計量以形成EC}之分 級，藉此改善該單元之電流及電壓特徵。為控制此種程 序，需使藉由量測所收集的資訊足夠廣泛，以能夠建立 EG作為深度（而不僅只是層堆疊物之平均值）之函數的模 型。為利於多層或分級層吸收體堆疊物之特性分析，使用 單一層堆疊物光學模型於在低於EG(圖11A)之完全吸收波 長範圍（通常在400 nm至約800 nm的範圍或圖11C中所指的 吸收區域）内擬合所量測的反射率信號，更長的波長係由 所量測吸收材料之EG來決定。對自約8〇〇 nm起的過渡及 - 透明區域（圖11B)使用完全層堆疊物光學模型，其中該完 整堆疊物模型之頂層1104係與該單一層堆疊物相同及其餘 的完全層堆疊物1106係經建立為分級或逐步式。兩個波長 區域間之邊界波長係取決於吸收體材料之類型及其標稱厚 度。根據EG之值，該層堆疊物通常自約波長1 〇〇〇 nm及以 146632.doc 21 201105949 上係透明的（圖11C)。此利用不同光學模型分離成波長區 域，其儘管具有共同的參數，但能夠進行更快的計算收 敛。 於相同層堆疊物上進行之透射量測未在完全吸收波長範 圍中產生任何信號（圖11C)。然而，該信號取決於EG在自 約800 nm起之過渡範圍中開始增加。在約1〇〇〇 nm及以上 的透明範圍中，反射率及透射光譜二者皆包含取決於該等 層（尤其係吸收層）之厚度及介電函數之震盪。將光學模型 擬合至以下四種量測之至少兩種（其中至少一種為反射率 量測）：來自單元區域之反射率、來自單元區域之透射 率、來自劃線區域之反射率及來自劃線區域之透射率，能 夠進行吸收層介電函數、梯度及層厚度的特性分析。材料 特性中梯度之存在可藉由透明範圍中震盪之振幅變化來識 另J層中梯度的方向可藉由擬合表面層11〇4之介電函數的 至少某些參數（如藉自完全吸收波長範s中的反射率所量 測）而界定（圖1 1C)。 以上技術可當在於吸收層之頂部上存在額外、基本上透 明的層時實施。此種情況下，該額外層係經添加至每個層 模型之頂部’即共用層之頂部。 一種提咼CIGS太陽能面板之效率的已知技術係藉由在 CIGS層之組合物中形成梯度，藉此引起EG梯度並形成能 夠增加量子效率的内部電場。該使用至少單元反射率量測 之刀析可提供關於該等EG梯度及吸收梯度之資訊。可對 CIGS製程實施以上所述用$分離不同;皮長範圍内所用模型 146632.doc •22- 201105949 之技術。 為藉由更加有⑨的榻取太陽能光譜來提高該太陽能面板 之效率，利用經串聯沉積之具不同EG之多個吸收體層製 造串聯式或多接面結構以形成-堆疊物。本方法亦可應用 於特性分析串聯式結構（諸如於非_層上之微晶石夕一 Si))之層的EG"及收、結晶度及厚度（其中該等吸收體層係 沉積於通常粗糙的透明導電氧化物（TC〇)上）。該^卜“層 一般係由嵌入a - S i中之不同濃度的結晶矽之小顆粒所組 成。結晶度值通常係以體積分率之單位定義。實施劃線或 量測目標f測大大提高了分別特性分析a_s〜c_si層並提 供pc-Si層内之結晶度及結晶度梯度的詳細資料之能力。 自約400 nm高至約1000 0111的波長範圍中之反射率量測係 在單元區域上進行，其中該層堆疊物由上至下依次為pc_ Si a Si TCO。可在劃線區域上進行額外量測，其中該 層堆疊物由上至下依次為u_Si、a_Si、玻璃。此量測可藉 由反射率、透射率或光譜橢圓偏振量測在自約4〇〇 nm高至 約1000 nm的波長範圍内進行。當可在製程中實施時，可 用於單元區域中之勘層中孔的量測目標替代劃線量 測。直接在劃線上量測之優勢在於可不需要於製程中實施 改變而進行特性分析的能力。該使用至少單元反射率量測 之分析可提供關於構成層之EG及EG梯度、吸收及吸收梯 度及結晶度及結晶度梯度之資訊。 該方法亦可用於在製程内特性分析cdTe之吸收及E(3， 其中該吸收體層係沉積於CdS層上。 146632.doc -23- 201105949 已發現將鈉併入CIGS材料之多晶結構中係提供高光轉 化效率的-關鍵因素。Na原子在材料中經歷擴散並使晶界 處的表面狀態鈍化。所需Na濃度之範圍—方面係由為接近 忒材料之所有晶界表面區域的最小值界定，同時最大值係 CIGS層之黏合開始劣化的值。該Na濃度亦會影響cIGS層 之表面形態。該（：1(}8層之反射率會受Na加入含量及濃度 的衫響《因此，除能隙外，介電函數模型可提供關於1^3濃 度的育訊，同時亦可經模型化之表面粗糙度的變化提供關 於Na濃度的額外資訊。 已闡述若干實施例。然而，應瞭解可在不脫離本方法及 電極結構的主旨及範疇下進行各種修改。因此，其他實施 例係涵蓋於以下申請專利範圍之範疇内。 【圖式簡單說明】 圖1係典型薄膜堆疊物之示意圖； 圖2A係典型薄膜光伏打面板結構之示意圖； 圖2B至2C係提供圖2的典型薄膜光伏打面板結構之額外 洋述的放大截面； 圖3A及3B係適用於量測吸收薄膜（其係於劃線上沉積於 接觸層之頂部）參數之位置的示範性實施例之示意圖； 圖4A及4B係適用於量測經沉積於量測目標區域中之吸 收薄膜參數的位置的另一示範性實施例之示意圖； 圖5係在沉積層中製造之示範劃線之示意圖； 圖6A及6B係用於利用大照明光斑量測薄膜參數之本發 明量測方法的示範性實施例之示意圖； 146632.doc • 24· 201105949 圖7係利用單一廣區域照明系統的本發明量測方法的示 範性實施例之示意圖； 圖8係圖2B之放大圖且係劃線區域之示意圖； 圖9係裝備有用於量測薄膜光伏打面板之本發明系統的 典型薄膜製造線之示意圖； 圖10係闡述利用本發明系統之薄膜光伏打面板參數量測 方法的示範性製程之流程圖；及 圖11A至11C係CIGS吸收層之透射及反射率之示意圖。 【主要元件符號說明】 100 面板 104 導電膜 108 基板 112 吸收薄膜 116 導電膜 200 面板 202 劃線區域 204 劃線 208 條狀物單元 304 劃線 308， 接觸層 312 吸收薄臈 316 透明基板 330 吸收體層 334 透明導電氧彳4 146632.doc •25- 201105949 338 342 404 500 504 508 512 516 520 604 608 702 704 708 900 904 908 910 912 916 920 928 932 1104 表面粗縫度 劃線 目標 底面 劃線 劃線 凹口 基板 雷射之步進 照明光斑 照明光斑 廣區域照明系統 太陽能面板 收集光學元件系統 輸送器 裝置 位置控制模組 箭頭 箭頭 光學系統 光學系統 檢測器 控制器 完整堆疊物模型之頂層 146632.doc -26- 201105949 1106 層堆疊物 1111 吸收體表面層 1112 吸收體分級層 1113 導電層 1114 基板 PI 劃線 P2 劃線 P3 劃線 146632.doc - 27 -The measurement of the first spot size of the I 等 in the region of varying thicknesses may result in the expected spectral oscillations being attenuated or even eliminated in the measurement signal due to the average of the 彳 5 来自 of the gentlemen from different regions. Use small spot size όΓ ; 隹 > π ^ to enlarge the local sampling of different thickness areas. Since the panel moves continuously with respect to the optical system #β, * Λ* Μ, ,, it requires a relatively short measurement sampling time to limit the above flat bb| + ten-homogenization effect. The measurement can be performed by a plurality of short measurements (he can use the flashing illumination I · , , and then the group that is separated into similar features. The signal noise is made. The tiger is # in each group. Average improvement. 146632.doc 201105949 The method has a spot size of about the width of the line or less. These numbers can be grouped (such as) to provide information about the maximum and minimum thickness within the line, or The spot size is small enough to provide semi-continuous information about the contour of the underside of the scribe line. m In general, the scribe line width is between 25 and 50 microns, so a spot size of 1 〇 to 2 〇 microns is suitable for padding. In-line measurement. If you use a xenon flash (for example, Hamamatsu L9455 series) set to a repetition rate of 1 〇〇 Hz and an effective pulse length of 丨 microseconds, and the solar panel moves at 3 m/min, The light leakage of the spot size generated by the relative motion will be negligible 50 nm and the interval between the measurements along the scribe line may be 5 〇〇 micrometer. Alternatively, an optical movement system may be applied to compensate the sample relative to the optical system. phase The optical system can be placed on a mobile system or an optical scanning method that scans in the direction in which the panel is moved can be used, as known in the art, and can be imaged at any predetermined location. The static measurement is performed effectively. The range of movement is defined by the measurement time requirement. Based on the above example, a measurement range of 0.5 mm will be required for a measurement time of 10 milliseconds. Figure 6A is a measurement of the film using a large illumination spot 6〇4 A schematic diagram of an exemplary embodiment of the measuring method of the present invention. In this embodiment, the received or detected signal strength is formed by specifically shaping the illumination spot 6〇4 parallel to the scribe line 204 (Fig. 2 The direction of the illumination is enhanced by a line of lines that are configured to have a relatively large overlap with the scribe line and increase the transmission signal. The strip illumination spot can also be used for both transmission and reflection measurements. 146632.doc •14- 201105949 Unit area. To simplify the optical system and eliminate the need for lateral adjustment of the active adjustment measurement position, the width of the spot 604 can be defined as a scribe line 2〇4 width. A particular multiple, which depends on the lateral alignment change of the scribe line position. This lateral alignment change is affected by the positional phase of the scribe line 204. The variation of the edge of the solar panel and the variation of the solar panel relative to the lateral arrangement of the optical system. If the change with respect to the spot width is large, the optical system may need to perform lateral alignment of the spot position with respect to the solar panel before the measurement to provide a suitable overlap of the spot and the scribe line. Signal quality, when the signal from the cell area is not negligible, additional measurements can be made only in the cell area. After the unit is measured by the ratio of the cell to the scribe area within the spot size, It is subtracted from the scribe measurement and the signal is renormalized to obtain a pure scribe signal. In the case of transmission measurements, this is useful if the cell region is at least partially transmissive. This technique can be used for most situations by reflection measurements. In another exemplary embodiment (Fig. 6B), a large illumination spot 608 having a size at least a unit width may be used, whereby the optical element for collecting transmitted light may collect light from a large area of at least one single turn width, This can be measured by transmission via at least two scribe lines. The use of larger spot sizes simplifies the optical system and reduces the need for active lateral movement measurement locations. Further simplification of the optical system for transmission measurement can be achieved by using a single wide area illumination (Fig. 7) system 702 to 'illuminate the strip along the entire width of the solar panel 704 while a series of collection optics The component systems 708 are not placed on opposite sides of the panel, with each collector receiving its signal from a few scribe lines. A smaller number of collection optics systems can also be used, as well as scanning along the illumination area while measuring. Figure 8 is a magnified view of Figure 2B and a schematic view of a scribe region showing the weighting contributions of different features to the measurement signals at different locations such that the weight measurement method can separate the contributions of the features, wherein the terms Features may include at least some of the cell regions, P1 scribe, p2 scribe, and p3 scribe. The measurement system is performed at a later stage of the process by forming two or more scribe lines in the scribe line region (for example, p 用于 for isolating the contact layer i 〇 4 and P 2 for separating the absorbing layer i 丄 2 region). . Separating the lines of the signal from the unit signal and also separating the line signals requires at least two and often more than two measurements. If the transition of the edge of the spot size is clear and smaller than the distance between different scribe lines in a single scribe line region, then the cell (s〇), the cell and the dash line P1 (measurement signal S0*(la)+Sl*) a) and performing multiple measurements on both the first and second P2 lines (measurement signal s〇*(lbc)+Sl*b+S2*c), where SO, S1 and S2 Are the units, p], and ? The quaternary signal of 2, and a, b or c are the coefficients that determine the relative contribution of each region to the signal. The coefficients a, ^, and c may vary with wavelength depending on the measurement spot size as a function of wavelength. Additional measurements can be taken to extract signals of different ratios from different features. If the signal contributed by the cell region to the measurement spectrum is negligible, for example due to strong scattering caused by a rough surface or in a transmission measurement in which the cell region is highly absorptive, then only the latter is required. Measurement 6 If the intensity transition at the edge of the spot is not sharp (i.e., greater than the distance between the scribe lines), then a series of measurements are required by controlled movement perpendicular to the scribe line between them. 146632.doc -16* 201105949 The measurement of this queue shall include at least one measurement of only the unit (measurement (four) s〇), - with a contribution from the partial contribution of the first line (measurement signal s 〇*(1,a)+sl*a), _ has a contribution from the partial contribution of the first line and the contribution from the part of the first line (measurement signal s〇*(1_b_c)+sl* b+S2*c). Additional measurements can be taken to manipulate different ratios from different features. Based on the data of the relative lateral position of the measurement points, the measurements can be fitted to a function of the lateral spot intensity distribution of the contribution cyclotrons of the two scribe lines as described in the figure. If the spot profile is different for different wavelengths, this can be done separately for each wavelength of the spectrum. The lateral spectral intensity distribution of the spot can be determined by different means, such as scanning edge or edge mirror or slit. After the fitted scribe lines contribute to the gyro signal, these contributions can be separated and their respective spectral characteristics analyzed to fit the optical model and calculate the parameters of the individual layer stacks in each scribe line. This method can be performed using reflection, transmission or spectral ellipsometry and combinations thereof. Transmission measurement can also be performed on the line between the optical system and the solar panel. In an embodiment, the mid-span panel is placed on a continuously moving conveyor that is aligned above the score line and the panel is moved in a direction substantially parallel to the score line. In this way, the measurement time can be extended to improve the signal-to-noise ratio and the effect of possible local roughness in the grading. The optical system can be pre-aligned in the manufacturing line to overlap a particular scribe line. Multiple optical systems can be provided to simultaneously sample multiple scribe lines across the width of the panel to provide mapping capabilities and controllable lateral uniformity of the process. In another embodiment, a single optical system can be aligned to be measured at a single stroke 146632.doc • 17·201105949 line and moved in steps equal to a multiple of the cell width, thereby continuing across the panel width. A plurality of scribe lines are sampled to provide scribe mapping capabilities. Under certain conditions, the rotation of the solar panel relative to the direction of movement is large, and the system of actively measuring the position of the spot to actively maintain the proper overlap of the spot and the scribe line. If multiple optical systems are used, they can be moved simultaneously to track the scribe lines and maintain a suitable overlap of the measurement spots. The plurality of optical subsystems can be attached to a common mechanical interface to allow for the use of a single motion mechanism. Transmittance measurements were made in a thin film photovoltaic panel process in which the film was deposited on an opaque substrate. In this case, enhanced information can be obtained by measuring reflectance and/or spectral ellipsometry at different incident angles. Information can also be added to the layer by performing a separate reflectance measurement of the absorber layer in the cell region and in the scribe line. This provides two independent measurements on a layer of material containing effective different bases. The $target (as previously described) can be implemented during the scribing step, thereby relaxing the requirements for illumination spot size and alignment measured on the scribing layer stack. Figure 9 is a schematic illustration of a typical film manufacturing line equipped with the system of the present invention for measuring thin film photovoltaic panel parameters. The raft layer is carried at a suitable deposition station/carrier and the conveyor 9 is advanced between the stations 2 (Fig. 2). Following the deposition of the bulk layer, the solar panel 2 is typically advanced along the conveyor system 9 and passes through a device 904 comprised of a plurality of optical systems (exemplified by 916 and 92 。). The optical systems can be disposed above the panel, below the panel, and above and below the panel. The optical systems are pre-aligned to be placed above and/or below a particular area of the mobile panel 900. At least 146632.doc 201105949 may be positioned to allow certain optical systems (eg, system 916) to be metered in the area of unit 208 'some systems (eg, system 92〇) may be positioned on the scribe line and some may be positioned The measurement was performed on the measurement target 4〇4 (Fig. 4). Optionally, one or more position control modules 9〇8 are selected to provide lateral and longitudinal positioning movements (as indicated by arrows 91〇 and 912) to compensate for the position of the optical system relative to the panel 900 and the scribe line 204. Positioning. The position control module 908 can be further operated to provide rotation and compensate for the positioning of the optical system relative to the position of the panel 9 and the line 204. The positioning module can rotate the entire device 9〇4 or rotate each of the optical systems 9丨6 and 92〇, respectively. One or more detectors 928 sense the orientation of the panel 9A that is oriented relative to the optical systems 916 and 920. Controller 932 controls the operation of all of the units of the system and synchronizes them. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a flow chart showing an exemplary process for a thin film photovoltaic panel parameter measurement method utilizing the system of the present invention. # When receiving the input signal close to the panel (box _), the device starts the measurement sequence. The signal from the face (4) is received from the conveyor (4)_ or is generated by a specific detector 928 placed upstream of the light (four) system. Detector 928 senses the orientation of panel _ relative to the nominal position and nominal angle (block 1 〇〇 4), and the lateral offset and angle of scribe 2 〇 4 (block 1 〇〇 8). The message is sent to the system controller, and the control Is calculates the corrective action and sends the calibration device to the positioning module 0 that controls the cardiac position of the plurality of optical systems. The positioning position is 桢, and the plurality of optical systems are corrected. The lateral position (block 1012) is such that the predetermined A, s θ 疋 在 在 在 在 疋 疋 疋 疋 疋 在 在 在 在 在 在 , , , , , , , , , , , , , , , , , , , , , , , , , Appropriate "" offsets are appropriate for individual features to track possible diagonal movements of the features. The controller 932 synchronizes the measurements based on a predetermined amount of different optical systems 146632.doc • 19·201105949 916 and 920 (blocks 1〇16). Detector 928 can also be positioned to detect the presence and longitudinal position of measurement target 404 on the mobile panel as needed, and then send a signal (in addition to the panel orientation information) to controller 932. The timing required to measure on the target' and the optical system required thereafter are suitably initiated to perform simultaneous measurements on the target (covered in Box 1 〇 16). The measurements are repeated at a plurality of locations along the length of the panel to produce an image of the panel characteristics. Data from all measurement channels is received by the controller (block 1020) and transferred to a data system. The measurements from a plurality of locations within a predetermined distance (typically similar to the cell width) are combined into an extended data set. The extended data set is analyzed by comparing each measurement of the set to a suitable optical model. Each optical model consists of at least one film or film layer stack that is used to calculate the corresponding theoretical spectrum that is fitted to the measurement spectrum. Optical models are based on the calculation of light propagation through a stack of materials defined by parameters such as thickness, refractive index, and extinction coefficient, as well as interface, scattering, spatial gradient effects, and others in material properties. A fitting procedure is performed whereby the parameters of the optical model are changed until a high degree of agreement or match is achieved between the theoretical spectrum and the actual measured spectrum. The theoretical spectral characteristics correspond to the actual measured film properties. Information measured for at least the film properties of the absorber layer can be communicated to manufacturing equipment disposed upstream and downstream, and wherein the communication parameters allow control of the absorber formation process. The process parameters controlled may be at least some of the following parameters: source material flow rate, source material flow ratio, ambient pressure around the substrate, substrate temperature, substrate movement speed, process time, and 146632.doc -20. 201105949 Other. In order to facilitate the use of multiple variable parameter measurement layers and layer stacks, it is necessary to collect as much independent information as possible. Increasing the wavelength range of the collected light increases the ability to separate the effects of different parameter variations in '^. Combining at least two of the three wavelength ranges of ultraviolet, visible, and red may facilitate such enhancement. For example. A combination of VIS-NIR (400 nm-i000 nm) & IR (95〇nml7〇〇 New) Light 4 sensors produces a range of operating energy covering solar panels and lower energy IR where the materials become transparent The scope of the {scientific system. One of these techniques for enhancing the QE of solar panels is based on designing materials by controlling changes in the depth dependence of the properties of the materials. For example, s, the stoichiometry of the material is changed during deposition to form a fraction of EC}, thereby improving the current and voltage characteristics of the unit. In order to control such a procedure, the information collected by the measurements needs to be sufficiently broad to be able to model the EG as a function of depth (and not just the average of the layer stacks). To facilitate characterization of multilayer or graded layer absorber stacks, a single layer stack optical model is used at a full absorption wavelength range below EG (Fig. 11A) (typically in the range of 400 nm to about 800 nm or in Figure 11C) The measured reflectance signal is fitted within the indicated absorption region. The longer wavelength is determined by the EG of the measured absorbent material. A full layer stack optical model is used for the transition from about 8 〇〇 nm and the transparent region (Fig. 11B), wherein the top layer 1104 of the complete stack model is identical to the single layer stack and the remaining full layer stack 1106 is established as a grading or stepwise. The boundary wavelength between the two wavelength regions depends on the type of absorber material and its nominal thickness. Depending on the value of EG, the layer stack is typically transparent from about 1 〇〇〇 nm and 146632.doc 21 201105949 (Fig. 11C). This is separated into wavelength regions using different optical models, which, despite having common parameters, enable faster computational convergence. Transmission measurements made on the same layer stack did not produce any signal in the full absorption wavelength range (Fig. 11C). However, this signal depends on the EG starting to increase in the transition range from approximately 800 nm. In the transparent range of about 1 〇〇〇 nm and above, both the reflectance and the transmission spectrum contain oscillations depending on the thickness of the layers (especially the absorber layer) and the dielectric function. The optical model is fitted to at least two of the following four measurements (at least one of which is reflectance measurement): reflectance from the cell region, transmittance from the cell region, reflectance from the scribe region, and The transmittance of the line region enables analysis of the characteristics of the dielectric function, gradient, and layer thickness of the absorber layer. The presence of a gradient in the material properties can be determined by the amplitude variation of the oscillation in the transparent range to identify the direction of the gradient in the J layer by fitting at least some of the parameters of the dielectric function of the surface layer 11〇4 (eg, by full absorption) Defined by the reflectance in the wavelength range s) (Fig. 1 1C). The above technique can be practiced when there is an additional, substantially transparent layer on top of the absorbent layer. In this case, the additional layer is added to the top of each layer model, i.e., at the top of the common layer. One known technique for improving the efficiency of CIGS solar panels is by forming a gradient in the composition of the CIGS layer, thereby causing an EG gradient and forming an internal electric field that increases quantum efficiency. The use of at least unit reflectance measurements can provide information about the EG gradients and absorption gradients. The techniques described above for the CIGS process can be implemented using the $ separation method; the model used in the skin length range 146632.doc • 22-201105949. In order to increase the efficiency of the solar panel by taking a solar radiation spectrum of 9 more, a series or multi-junction structure is formed by a plurality of absorber layers deposited in series with different EGs to form a stack. The method can also be applied to the EG" and the acceptance, crystallinity and thickness of the layer of the characteristic analysis tandem structure (such as the microcrystalline Si-Si on the non-layer) (where the absorber layers are deposited in the usual roughness) Transparent conductive oxide (TC〇)). The layer is generally composed of small particles of different concentrations of crystalline germanium embedded in a - S i . The crystallinity value is usually defined in units of volume fraction. The implementation of scribing or measurement target f measurement is greatly improved The ability to separately analyze the a_s~c_si layer and provide detailed information on the crystallinity and crystallinity gradient in the pc-Si layer. The reflectance measurement in the wavelength range from about 400 nm up to about 1000 0111 is in the cell region. The above process is performed, wherein the layer stack is pc_Si a Si TCO from top to bottom. Additional measurement can be performed on the scribe region, wherein the layer stack is u_Si, a_Si, glass in order from top to bottom. The measurement can be performed in the wavelength range from about 4 〇〇 nm up to about 1000 nm by reflectance, transmittance or spectral ellipsometry. It can be used in the layer in the cell area when it can be implemented in the process. The measurement target of the hole replaces the measurement of the scribe line. The advantage of directly measuring on the scribe line is the ability to perform characterization without the need to implement changes in the process. The analysis using at least the unit reflectance measurement can provide information about the constituent layers. EG and EG ladder , absorption and absorption gradients and crystallinity and crystallinity gradient information. This method can also be used to analyze the absorption of cdTe and E (3, where the absorber layer is deposited on the CdS layer. 146632.doc -23- 201105949 It has been found that the incorporation of sodium into the polycrystalline structure of the CIGS material provides a key factor in providing high light conversion efficiency. The Na atom undergoes diffusion in the material and passivates the surface state at the grain boundary. The range of Na concentration required - aspect It is defined by the minimum value of all grain boundary surface regions close to the tantalum material, and the maximum value is the value at which the adhesion of the CIGS layer begins to deteriorate. The Na concentration also affects the surface morphology of the cIGS layer. (:1(}8 layer The reflectivity will be affected by the amount of Na added and the concentration of the shirt. Therefore, in addition to the energy gap, the dielectric function model can provide information on the concentration of 1^3, and can also provide a description of Na by modeling the change in surface roughness. Additional Information on Concentrations Several embodiments have been described. However, it should be understood that various modifications can be made without departing from the spirit and scope of the method and the electrode structure. BRIEF DESCRIPTION OF THE PARTS: BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic view of a typical thin film stack; FIG. 2A is a schematic view of a typical thin film photovoltaic panel structure; FIGS. 2B to 2C are diagrams showing the typical thin film photovoltaic panel structure of FIG. 3A and 3B are schematic views of an exemplary embodiment suitable for measuring the position of the absorbing film (which is deposited on the top of the contact layer on the scribe line) parameters; FIGS. 4A and 4B are applicable to A schematic diagram of another exemplary embodiment of measuring the position of an absorbing film parameter deposited in a measurement target region; FIG. 5 is a schematic diagram of an exemplary scribe line fabricated in a deposited layer; FIGS. 6A and 6B are for utilizing large Schematic diagram of an exemplary embodiment of a measurement method of the present invention for illuminating a spot measurement film parameter; 146632.doc • 24· 201105949 FIG. 7 is a schematic illustration of an exemplary embodiment of a measurement method of the present invention utilizing a single wide area illumination system; Figure 8 is an enlarged view of Figure 2B and is a schematic view of a scribe line region; Figure 9 is a representation of a typical film manufacturing line equipped with the system of the present invention for measuring thin film photovoltaic panels. ; Figure 10 a flow diagram illustrating based systems utilizing the present invention, a thin film photovoltaic panel parameter measurement method of an exemplary process; and a schematic view of the transmittance and reflectance of the layers 11A to 11C based CIGS absorber. [Main component symbol description] 100 panel 104 conductive film 108 substrate 112 absorption film 116 conductive film 200 panel 202 scribe region 204 scribe line 208 strip unit 304 scribe line 308, contact layer 312 absorption thin 316 transparent substrate 330 absorber layer 334 Transparent Conductive Oxygen 4 4 146632.doc •25- 201105949 338 342 404 500 504 508 512 516 520 604 608 702 704 708 900 904 908 910 912 916 920 928 932 1104 Surface rough seam marking target bottom underline marking Port substrate laser stepping illumination spot illumination spot wide area illumination system solar panel collection optics system conveyor device position control module arrow arrow optical system optical system detector controller complete stack model top 146632.doc -26- 201105949 1106 Layer stack 1111 absorber surface layer 1112 absorber layer grading layer 1113 conductive layer 1114 substrate PI scribe line P2 scribe line P3 line 146632.doc - 27 -