201135962 六、發明說明： 【發明所屬之技術領域】 本發明的實施例大體上關於太陽能電池的製造，特別 是關於矽太陽能電池的鈍化。 【先前技術】 太陽能電池是將太陽光直接轉換成電力的光伏元件。 最一般的太陽能電池材料是矽（Si)，其為單晶、多晶、複 晶基材或非晶膜型式。世人正致力於減少製造太陽能電 池的成本及所得電池的成本，同時維持或増加所生產的 太陽能電池之總效能。 太陽能電池的效能可透過使用鈍化層而增強，該鈍化 層亦作用如形成太陽能電池的矽基材t射極（emitter)區 域上的抗反射塗層（ARCp當光從一個介質傳到另一個介 質（例如從空氣到玻璃或從玻璃到矽），一些光線可能從 兩個介質之間的界面反射。所反射的光比例是兩個介質 之間的折射率差異之函數，其中兩個相鄰介質的折射率 差愈大，會造成更高比例的光從該二介質之間的界面反 射。 太陽能電池將入射光能量轉換成電能的效能會受許多 因素不利地影響，該等因素包括從太陽能電池反射且在 電池結構（諸如鈍化層）中吸收的入射光之比例，以及 太陽忐電池中電子與電洞的重組速率。每一次電子電洞 201135962 對重組，會消除電荷載子，因而減少太陽能電池的效能。 重··:且可月b發生在基材的巨量石夕（buik silicon)中，為巨量 矽中缺陷的數目之函數；或者發生於基材表面上，為懸 空鍵（dangling bond，即未終結的化學鍵）存在多寡的函 數。 透過使用鈍化層徹底鈍化太陽能電池大體上藉由減少 重組速率而改善太陽能電池的效能，然，折射率（η)需要 隨裱繞的層調整，以減少光反射，同時亦維持太陽能電 池期望的光吸收能力。一般而言，薄的透明膜具有固有 的消光係數（k)(其量值為膜吸收的光線量之指標）以及 折射率（η)(該量值為當光線從一介質傳入另一介質時彎 折的程度指標）。 諸如貫用於鈍化的SiN的膜中，η與k數值的量值相 關聯，右一者高，則另一者同樣地也高。因為鈍化膜的 折射率乾圍受限於其所夾疊（sandwich)的材料，在先前技 述的操作内，所得的k值之範圍亦因而受限，故可見可 接受的折射率造就了無法接受的高k值。 此外’冗積速率乃至於基材可在時間的設定期間内接 收期望膜層的終極數目對折射率與k值產生影響，並且 影響膜的物理性質，諸如膜中的晶界與晶粒長度或尺 寸大型晶粒與其所造成的長晶界造成污染通過鈍化膜 進入矽的途徑，導致電池失效。因此，用於在每單位時 較大里太陽能電池的更高的沉積速率（能用於電 漿’儿積製程者）造成更大的晶粒以及因此造成針孔。 201135962 因此’而要-種形成鈍化層的改善方法，其具有結合 功能性與光學梯度的性質，減少電荷載子的表面重組、 改善所形成的太陽能電池效能，並且造成實質上無針孔 且有期望光學與鈍化性質的鈍化層。 【發明内容】 根據上文所述，本發明的實施例大體上提供用於塗層 的方法該塗層可作為尚品質純化與ARC層以用於太陽 能電池。根據一個實施例的方法包括形成多層鈍化與 ARC塗層，其透過以下步驟達成：將—第—製程氣體混 合物流進一處理腔室内的一製程空間；在該處理腔室中 於超過0.65 W/Cm2的功率密度生成電漿；在該製程空間 中沉積一含氮化矽界面次層於一太陽能基材上；將一第 二製程氣體混合物流進該製程空間；以及沉積一含氮化 碎巨量次層於該含氮化矽界面次層上。 另一實施例中揭露一種用於偵測形成於一太陽能電池 上的一鈍化層中的針孔之方法。該方法包括以下步驟： 將具有一含氮化矽鈍化層形成於其上的一太陽能電池浸 潤於一銅電解液中；施加電流通過該太陽能電池的金屬 覆蓋的背側’以鍍覆任何從該鈍化層的一外表面延伸到 該太陽能電池的一摻雜區域之針孔；以及偵測鍍覆於任 何針孔中的任何長出的銅。 尚有另一實施例，揭露一種太陽能電池，其包括一基 201135962 材，該基材具有一接面區域以及一鈍化的抗反射層於該 基材之一表面上。該鈍化的抗反射層包括一含氮化矽界 面次層’以及直接在該界面次層上的一含氮化石夕巨量次 層’該界面次層具有大於該巨量次層的折射率（n)，且其 中該鈍化層貫質上無完全通過該界面次層與該巨量次層 二者的針孔。 另一實施例中，揭露一種用於形成一膜於一太陽能電 池上的系統。該系統包括：一電漿處理腔室，.其用於在 該處理腔室的一處理空間内形成一鈍化/ARC層於一太 陽能電池基材上’該鈍化/ARC層包含一含氮化石夕界面次 層與一含氮化矽巨量次層，該含氮化矽界面次層是使用 由一第一製程氣體混合物在超過〇.65 W/cm2之功率密度 下生成的電漿形成在該太陽能電池基材上，而該含氮化 矽巨量次層是使用由一第二製程氣體混合物在超過〇 65 W/cm2之功率密度下生成的電漿形成在該界面次層上。 該系統亦包括一系統控制器，其與該電漿處理腔室通 訊，該系統控制器設以控制電漿功率密度、第一製程氣 體混合物流率以及第二製程氣體混合物流率，使得該界 面次層的折射率（η)大於該所得的巨量次層的折射率，且 該界面次層與該巨量次層二者皆具有從Q到Q i的消光 係數（k值）。 【實施方式】 201135962 本發明大體上提供形成高品質鈍化層的方法，以形成 尚效能太陽能電池元件。可受惠於本發明的太陽能電池 基材包括具有含單晶石夕、複晶石夕、多晶石夕、與非晶石夕之 主動區域（即薄膜單元）的基材’但亦可用於包含鍺 (Ge)、砷化鎵（GaAs)、碲化鎘（CdTe)、硫化鎘（Cds)、銅 銦鎵硒化物（CIGS)、銅銦硒化物（CuInSe2)、鎵銦磷化物 (CalnP2)、有機材料、以及異接面單元，諸如201135962 VI. Description of the Invention: TECHNICAL FIELD OF THE INVENTION [0001] Embodiments of the present invention generally relate to the fabrication of solar cells, and more particularly to passivation of germanium solar cells. [Prior Art] A solar cell is a photovoltaic element that directly converts sunlight into electricity. The most common solar cell material is germanium (Si), which is a single crystal, polycrystalline, polycrystalline substrate or amorphous film type. The world is working to reduce the cost of manufacturing solar cells and the cost of the resulting batteries while maintaining or increasing the overall performance of the solar cells produced. The efficacy of a solar cell can be enhanced by the use of a passivation layer that also acts as an anti-reflective coating on the emitter region of the germanium substrate forming the solar cell (ARCp is transmitted from one medium to another) (for example, from air to glass or from glass to enamel), some light may be reflected from the interface between the two media. The proportion of light reflected is a function of the difference in refractive index between the two media, where two adjacent media The greater the refractive index difference, the higher the proportion of light reflected from the interface between the two media. The efficiency with which solar cells convert incident light energy into electrical energy can be adversely affected by many factors, including solar cells. The ratio of incident light that is reflected and absorbed in the cell structure (such as the passivation layer), and the rate of recombination of electrons and holes in the solar cell. Each electron hole 201135962 recombines, eliminating charge carriers, thus reducing solar cells The efficiency of .··: and the monthly b occurs in the huge amount of buik silicon in the substrate, which is a function of the number of defects in the giant crucible. Or occurs on the surface of the substrate as a function of the amount of dangling bonds (ie, unfinished chemical bonds). Passivating the solar cell thoroughly by using a passivation layer generally improves the performance of the solar cell by reducing the rate of recombination, The refractive index (η) needs to be adjusted with the winding layer to reduce light reflection while maintaining the desired light absorption capacity of the solar cell. In general, a thin transparent film has an inherent extinction coefficient (k) (the magnitude of which is The index of the amount of light absorbed by the film) and the refractive index (η) (the magnitude is an indicator of the degree of bending when light is transmitted from one medium to another). For example, in a film of SiN used for passivation, η and The magnitude of the k value is associated, the right one is high, and the other is equally high. Because the refractive index of the passivation film is limited by the material it is sandwiching, within the operation of the prior art The range of k values obtained is thus limited, so that an acceptable refractive index results in an unacceptably high k value. Furthermore, the redundancy rate is such that the substrate can receive the desired film over a set period of time. The ultimate number affects the refractive index and the k value, and affects the physical properties of the film, such as the grain boundaries and grain lengths in the film or the size of the large grains and the resulting long grain boundaries causing contamination through the passivation film into the enthalpy. This leads to battery failure. Therefore, it is used for a higher deposition rate of solar cells per unit time (which can be used for plasma 'storage processors) to cause larger grains and thus pinholes. 201135962 Therefore' An improved method of forming a passivation layer, which combines the properties of functional and optical gradients, reduces surface recombination of charge carriers, improves solar cell performance, and results in substantially pinhole-free and desired optical and optical Passivation Layer of Passivation Properties [Invention] In accordance with the above, embodiments of the present invention generally provide methods for coatings that can be used as a quality purification and ARC layer for solar cells. A method according to one embodiment includes forming a multilayer passivation and ARC coating by: flowing a -process gas mixture into a process space within a processing chamber; in the processing chamber exceeding 0.65 W/cm2 a power density to generate a plasma; depositing a layer containing a tantalum nitride interface on a solar substrate in the process space; flowing a second process gas mixture into the process space; and depositing a massive amount of nitride The sublayer is on the sublayer containing the tantalum nitride interface. Another embodiment discloses a method for detecting pinholes formed in a passivation layer on a solar cell. The method comprises the steps of: impregnating a solar cell having a passivation layer containing a tantalum nitride thereon in a copper electrolyte; applying a current through the back side of the metal cover of the solar cell to plate any An outer surface of the passivation layer extends to a pinhole of a doped region of the solar cell; and any elongate copper plated in any pinhole is detected. In yet another embodiment, a solar cell is disclosed that includes a substrate 201135962 having a junction region and a passivated anti-reflective layer on one surface of the substrate. The passivated anti-reflective layer comprises a tantalum nitride-containing sublayer and a nitride-containing sub-layer directly on the interfacial sub-layer. The interfacial sub-layer has a refractive index greater than the giant sub-layer ( n), and wherein the passivation layer does not have a pinhole that passes completely through both the interface sublayer and the macrosecond layer. In another embodiment, a system for forming a film on a solar cell is disclosed. The system includes: a plasma processing chamber for forming a passivation/ARC layer on a solar cell substrate in a processing space of the processing chamber. The passivation/ARC layer comprises a nitride-containing layer The interface sublayer and a massive sublayer containing tantalum nitride are formed by using a plasma generated by a first process gas mixture at a power density exceeding 〇.65 W/cm 2 . The solar cell substrate is formed on the interface sublayer using a plasma generated by a second process gas mixture at a power density exceeding W65 W/cm2. The system also includes a system controller in communication with the plasma processing chamber, the system controller configured to control the plasma power density, the first process gas mixture flow rate, and the second process gas mixture flow rate such that the interface The refractive index (η) of the sublayer is greater than the refractive index of the resulting macrolayer, and both the interface sublayer and the macrolayer have an extinction coefficient (k value) from Q to Q i . [Embodiment] 201135962 The present invention generally provides a method of forming a high quality passivation layer to form a still efficient solar cell component. A solar cell substrate which can be favored by the present invention includes a substrate having a single crystal, a polycrystalline stone, a polycrystalline stone, and an amorphous region (ie, a thin film unit), but can also be used for Contains germanium (Ge), gallium arsenide (GaAs), cadmium telluride (CdTe), cadmium sulfide (Cds), copper indium gallium selenide (CIGS), copper indium selenide (CuInSe2), gallium indium phosphide (CalnP2) , organic materials, and out-of-plane units, such as
GalnP/GaAs/Ge或ZnSe/GaAs/Ge基材’其用於將太陽光 轉化成電力。 大體而&，鈍化層將具有光學性質以盡量減少光反射 與光通過鈍化層的吸收，並且具有功能性質以「表面」 鈍化所配置於其上的表面、「巨量（bulk)」鈍化相鄰區域 及基材表面並且儲存正電荷於鈍化層中、或「場」鈍化。 因此，鈍化層含有期望的氫濃度以復原基材表面上所見 的淺缺陷。鈍化層能夠藉以執行該等功能的機制包括例 如所形成的鈍化層成為用於校正基材區域中缺陷的氫 (H+)來源的能力’以及所形成的層能夠緊包⑴d基材 表面處懸空鍵的物理及/或化學特性。 平衡太陽能電池的鈍化層之期望性質具有挑戰性，特 別是當純化層亦作用為拍·;5 Ma 寶"卞用马抗反射塗層時。當使用氮化矽 (SixNy ’亦縮寫為SiN )膜做為純化層時，挑戰增加’ 這是因為達成期望的膜性質需要平衡詩形成具特殊光 學或功能品質的鈍化層的製程參數之競爭。例如，告尋 求改善純化層的光學梯度性質時，其代價通常是犧^ 201135962 能性質’諸如基材的表面、巨量、及場純化。 有時，甚至難以平衡一個區域内的性質。例如，在過 去太陽能工業企圖實行鈍化層的光學梯度性質卻失敗， 因為膜中難以獲得低消光係數性質與一般製程中的 高折射率（η)。在使用典型膜形成方法時，生成具有高折 射率（η)的膜亦意味著生成具有高消光係數的臈。換言 之，可變的η與k彼此相互反映，其中於根據習知方法 形成膜時η與k大體上一起上升或下降。|^與n值的量 值之間的獨立性提供結合期望光學與功能性質至鈍化層 的能力，即能有較低的k且因此較少光線損失，且同時 能有較高的η因而降低反射率。應注意，所測量的n與 k值依賴其所受測量所處的頻率（即光波長）。在此討論 的k與η值是個別在4〇〇 nm與633 nm測量。 膜的表面重組速率（surface rec〇mbinati〇n vel〇city, SRV)是另一項鈍化抗反射層的困難性質，其需要與所 有其他性質平衡。SRV是基材表面處自由電子與電洞重 組因而彼此中和的速率。此外，為了在一個區域達成期 望的膜性質（諸如與期望的光學性質相對的期望功能性 質），可能需要減少膜沉積速率而因此降低處理量與生 產。 因此，大體上難以在不折衷的情況下形成結合功能性 與光學性質的氮化矽（SixNy )鈍化層，其造成個別為次 適（suboptimal)的性質，因為SiN材料無法提供這些性質 而無視另一性質。 201135962 本發明的實施例大體上提供一 樘升^成鈍化/抗反射層 的方法’其提供結合期望功能 a呈力此性質與光學梯度性質的能 力’其中避免了先前技術中發規的兮 货兄的5亥專性質之聯結性的 來源。發明人已發現，相較 平又孓使用正常功率範圍（一般 是2000-3000瓦（w)之間），锈 透k使用較向功率沉積氮化 # η傾向更具獨立性。高電|功率密度使得期望 膜性質的形成得以呈現光學與功能性梯度性質二者，豆 非基於傳統上高k值搭配高η值及低k值與低η值等折 衷，透過使用多層沉積㈣於形成鈍化層，避免了針孔 延伸穿過整個膜層。 因此，本發明之實施例提供形成具有高折料但低消 光係數00的鈍化/ARC層之方法。透過特定地修改製程 化學條件以用於膜形成，多層鈍化膜中的每一次層可具 有一起結合與形成具期望光學與功能性質的鈍化/arc 層的特定性質。 在一個實施例中，鈍化層可包含一層或多層，或漸變 區域（graded regi〇ns)，其具有不同組成、不同物理性質 及/或不同的電子性質，以提供鈍化效應與光學性質。例 如，在一個實施例中，鈍化層12〇含有含矽氮的界面次 層121與形成在界面次層121上的含矽氮的巨量次層 122，如第ic圖至第1F圖所示。 鈍化層形成製程 第1A圖至第if圖說明太陽能電池基材11〇在處理序 201135962 列中不同階段的概略剖面視圖，該處理序列用於形成純 化/ARC層120於太陽能電池1〇〇的表面上（例如，頂表 面1〇5)。第2圖說明用於形成純化層於太陽能電池基材 no上的製程序列2〇〇。第2圖中的序列對應帛1A圖至 第1F圖中所描繪的階段。太陽能電池100的一個實施例 中，P型基材110 (具有底表面1〇6並且包含結晶矽）具 有基底區g m以及形成於其上的n型摻雜射極區域 102，其一般是透過一摻雜與擴散/退火製程而形成，然 可使用其他包括離子佈植的製程。基材11〇亦包括p_n 接面區域103,該區域配置在太陽能電池的基底區域 與射極區域1 02之間，且其為太陽能電池1 00受到入射 光線的光子照明時，電子電洞對生成之區域。 雖下文中的論述主要論及用於處理形成在p型基底區 域上具有η型射極區域的基材，然申請人不希望此組態 限制此述的發明範嘴，因為鈍化層亦可形成於η型基底 區域、ρ型射極的太陽能電池組態上。 參考第2圖，在製程序列2〇〇期間，基材11〇的表面 經受多個製程’該等製程用於形成界面次層121與巨量 -人層122於基材表面上。接著是製程2〇12〇5的範例， 其可在類似於處理腔室300 (第3圖）的處理腔室中執 行。一個實施例中’所有在製程序列200中執行的製程 疋在一或多個系統400中所設的處理腔室431-437 (第4 圖）中執行。 用於形成純化層於太陽能電池基材110上的製程序列 201135962 200大體上始於將原生氧化物層下伏基材移除，如第2 圖的製程20L所示。正常處理太陽能電池元件期間，原 生氧化物層U5將形成於基材11〇的一個或多個表面 上。在製程201 ’基# 110的表面受到清潔而移除氧化 物層115 (第1A圖）。一個實施例中，清潔製程可 使用乾式清潔製程執行，在該製程中基材㈣暴露至反 應性電漿蝕刻製程，以移除氧化物層115。一個實施例 中’在製程20卜於處理腔室（如第3圖的腔室3〇〇)中 配置-個或多個基材110後，原生氧化物層ιΐ5暴露至 反應性氣體，該氣體可包含氮、氟與氫。接著，與反應 性氣體反應的氧化物層115受到熱處理，以將之從基材 表面移除。一些實施例中，熱處理可為在處理腔室300 或系統400中所設的另一相鄰腔室中執行的退火製程。 實例中彳期望確疋基材不暴露至氧以延長時間 歷時。因此，本發明-些實施例中，期望在無氧的，隋性 及/或真空環境（諸如群集工具或系統4〇〇 (第*圖）的 真空製程空間）中執行製程2G3_2G8的每—者，如此， 基材不會在製程203-208之間暴露於氧。 一個實施例中’於配置在基材載具425上的—批基材 110上執行製帛2〇1後，該等基材隨後定位在處理腔室 中，使得在202-206執行的製程可在基材上執行。接著， 如第2圖與第1B圖所示，含氮化矽的界面次層⑵形成 於乾淨的、已移除氧化物的基材表面1〇5上。一個實施 例中’界面次層121可介於約5〇 A至約A厚，諸如 12 201135962 150 A厚。一個實施例中，使用化學氣相沉積（CV+D)、電 聚強化化學氣相沉積（PECVD)、或物理氣相沉積（pVD) 技術將界面次層121形成覆於頂表面ι〇5上。 在製程202,於一個態樣中，形成界面次層121的方 法包括將第一製程氣體混合物流進處理腔室内的製程空 間306。在製程203’電漿於製程空間3〇6生成，而在製 程204，製程空間306中含氮化矽界面次層121沉積在 太陽能電池基材110上。 接著’在第2圖與第1C圖至第1D圖所示的製程2〇5 中’使用電漿強化化學氣相沉積（PECVD)製程將含氮化 矽巨量次層122形成於界面次層121上，因而形成多層 純化抗反射塗層120。或者，可使用化學氣相沉積（CVD) 或物理氣相沉積（PVD)技術將巨量次層122形成覆於界 面次層m上。PVD製程可用於在氫氣氛（atm〇sphere) 内反應性濺鍍，以形成多層鈍化抗反射層。例如，矽標 把可在氮與氫氣氛中以氬濺射，以沉積各siN層。一個 實施例中’巨量次層122可為約400 A至約700 A厚， 諸如600 A厚。一個範例中，基材暴露至13 56 MHz的 RF電漿，以形成界面次層12ι與巨量次層122二者。 一個實施例中，第一與第二製程氣體混合物包含含矽 刖驅物與含氮前驅物。例如，第一製程氣體混合物可包 含矽烷（SiH4)、氮（NO及/或氨（NH3)。第二製程氣體混合 物可包含矽烷與氮、矽烷與氨（NH3)、或者是矽烷、氨與 氮。表1詳述可用於藉由PECVD形成界面次層121與巨 13 201135962 量次層122的製程條件。表1列出VR 〃 、 率，其基於每公升製程空間。♦ 1 ,、 4氮及/或氨的μ 如Ν2)對含矽前驅物（例3 : 1亦包括含氮前驅物（例 積劁铲沾Α玄A 1 4)的流率比率、每一沉 積製％的功率密度、喷頭早母 及每一次層所需的竹接* 何支撐件之間的間距、以 121的皆她/丨士政 J如，用於形成界面次層 121的貫施例中，氮流率為每 ..v v / 升氣程空間約77.30標準 立万a h (seem)，而矽烷流率為 千兩母A升製程空間約5.25 seem。界面次層m可 方 了以從每分鐘1〇〇〇 Λ至3000人的 速率沉積，例如每分鐘〗1〇〇 Α， 印达量次層1 22可以每 为鐘超過3000 Α之速率沉積。 另實施例中，第一製程氣體混合物亦可包括氮氣（出) 稀釋劑纟可以每公升腔室體積J i 0 sccm的流率添加到 如表1所示的矽烷、氨、與氮的流率中。雖然表丨中未 示，但在巨量次層製程配方中，氨對矽烷流率的比率可 為約0.90。相信小心地控制矽烷氣流助於達成期望的膜 光學與功能性質。在巨量次層製程條件中，製程氣流大 體上較南。具通常技術者能夠取決於沉積製程的功率、 壓力、間距、及溫度而成功地修改氣流比率。 N2流率 (sccm/1) NH3流率 (sccm/1) SiHU流率 (sccm/1) N2/S1H4 比率 功率密度 (W/cm2) 間距 (mils) 時間 (Sec.) 界 面 77.30 0.00 5.25 14.70 .65-1.0 800 9 14 201135962 次 層 巨 量 次 層 77.30 8.40 9.20 8.35 .65-1.0 1100 15 表1 一個實施例中’用於界面與巨量次層沉積製程二者的 基材溫度可維持在”(^至40(rc，諸如從38〇。(：至39〇 c。各種用於維持基材溫度的手段可為負載鎖定加熱、 基材支撐件加熱、電漿加熱等。該二次層可於約1.5τ〇ΓΓ 的腔室壓力下沉積。 用於母人層的電漿可由RF功率提供，相較於典型 SiN沉積製程的2〇〇〇_3〇〇〇 w之範圍，該rf功率在兄 MHZ之頻率下是介於約4350 W至約6700 W，諸如約 5000 W。可提供RF功率至喷頭31〇及/或基材支撐件 33〇。用於界面次層與巨量次層沉積二者的RF功率密度 可為基材表面的約〇.65 w/em2以上以生成電漿。例如， 一些實施例中，RF功率密度可為丨〇〇 w/cm2。另一實施 例中，RF功率密度可為約〇 75 w/cm2。功率密度可盡可 月咼因為較咼的功率使界面與巨量次層中正電荷增 加。因此，較高的功率提供較佳的場鈍化並且降低SRV。 15 201135962 另-實施例中，界面次層與巨量次層之間的分界可更 清楚地透過停止及重啟電漿（同時製程氣體從界面次層 配方轉變成巨量次層配方）而以突麸 八…、的轉變區界定。此 轉變可以各種方式發生。例如，右筮_ & 在第一製程氣體混合物 導入處理腔室之前，可停止第一贺起#祕 乐 ^私乳體混合物的流 動。另一實施例中’僅停止矽院氣片拉 70札流時，其餘第一製程 氣體混合物的前驅物繼續流動。你丨如 . j w ’在關掉功率的期 間，用於界面次層的氣流配方可轉變成巨量次層配方。 氣流將不提供立即的完美氣體混合。因此，電^沉積製 程中的此斷裂將使得用於巨量鈍化層配方的氣體適當地 在電漿功率重啟之前混合。於製程204後執行製程的「斷 裂」可持續約2秒，其容許在第二製程氣體混合物流進 腔室前，實質上從腔室沖淨第一製程氣體混合物。 在製程配方的轉變期間，可公升高矽烷流以將其均等 地分配通過腔室，之後再度供應功率給噴頭以重新點燃 電漿’因而結束形成純化層12〇。界面次層沉積後關掉 電漿功率及重啟電漿以用於巨量次層沉積藉由製做更理 想且突然轉變的層而增加了膜密度以及最終效能。斷裂 期間’基材溫度大體上維持在約38(rc至約39(rc的溫 度。其他實施例中，功率可公升高至巨量次層沉積製程 期間其最終功率設定。在一些測試十’停止及重啟功率 可使膜的開路電壓（V〇c)增加3 mV (毫伏）。 另一實施例中，巨量次層的形成可透過在不「斷裂」 下’將氣流從界面次層配方轉變到巨量次層配方（如表 16 201135962 1所示）而發生’因而創造無界定化學計量的轉變層。 氣體流率及/或氣體混合物組成經改變以從一個製程配 方轉變到下一個製程配方’同時開啟電漿功率。此實施 例中的轉變層可為3-5 nm厚，或低於巨量次層厚度的 10% ’但為7-8%的總沉積鈍化層厚度。 所得的界面次層121與巨量次層122(如概略說明於 第1C圖至第1D圖者）形成鈍化/arc層120覆於p型 掺雜區域的頂表面105上。一個實施例中，期望形成多 層鈍化ARC層120，其含有期望量的捕捉的正電荷以提 供期望的p型區域之表面鈍化。另一實施例中，捕捉的 正電荷提供期望的具η型摻雜區域的η型基材之表面純 化。一個實施例中，多層鈍化塗層12〇中捕捉的正電荷 之總量Q!與捕捉的負電荷之總量Q2的總和具有足夠的 捕捉電％以達成約lxl〇12 Ccmlombs/cm2 (每平方公分的 庫倫數）以上的電荷密度，諸如約lxl0i2c〇ul〇mbs/cm2 至約 lxlO14 Coulombs/cm2 之間，或約 2xl〇i2A GalnP/GaAs/Ge or ZnSe/GaAs/Ge substrate' is used to convert sunlight into electricity. In general, & the passivation layer will have optical properties to minimize light reflection and absorption of light through the passivation layer, and have functional properties to "surface" passivate the surface on which it is disposed, "bulk" passivation phase Adjacent areas and substrate surfaces and store positive charges in the passivation layer, or "field" passivation. Therefore, the passivation layer contains the desired hydrogen concentration to restore the shallow defects seen on the surface of the substrate. The mechanism by which the passivation layer can perform such functions includes, for example, the ability of the passivation layer formed to be a source of hydrogen (H+) for correcting defects in the region of the substrate, and the layer formed can tightly wrap the (1)d dangling bond at the surface of the substrate. Physical and/or chemical properties. It is challenging to balance the desired properties of the passivation layer of the solar cell, especially when the purification layer also acts as a photographic film; 5 Ma Bao " When using a tantalum nitride (SixNy' also abbreviated as SiN) film as a purification layer, the challenge increases because the achievement of the desired film properties requires competition for the process parameters of the passivation layer having a special optical or functional quality. For example, when seeking to improve the optical gradient properties of a purified layer, the cost is usually to sacrifice the properties of the substrate such as the surface of the substrate, bulk, and field purification. Sometimes it is even difficult to balance the nature of an area. For example, attempts to implement the optical gradient properties of passivation layers in the solar industry have failed because of the low extinction coefficient properties and high refractive index (η) in the general process. When a typical film formation method is used, the formation of a film having a high refractive index (?) also means that a ruthenium having a high extinction coefficient is generated. In other words, the variable η and k reflect each other, wherein η and k generally rise or fall together when the film is formed according to a conventional method. The independence between the values of ^ and n provides the ability to combine the desired optical and functional properties into the passivation layer, ie, can have a lower k and therefore less light loss, and at the same time have a higher η and thus lower Reflectivity. It should be noted that the measured values of n and k depend on the frequency at which the measurement is taken (i.e., the wavelength of light). The k and η values discussed here are measured individually at 4 〇〇 nm and 633 nm. The surface recombination rate of the membrane (surface rec〇mbinati〇n vel〇city, SRV) is another difficult property of passivating the antireflective layer, which needs to be balanced with all other properties. SRV is the rate at which free electrons and holes at the surface of the substrate recombine and thus neutralize each other. Moreover, in order to achieve desired film properties in one region (such as desired functional properties as opposed to desired optical properties), it may be desirable to reduce the film deposition rate and thus the throughput and production. Therefore, it is generally difficult to form a nitride-nitride (SixNy) passivation layer that combines functional and optical properties without compromise, which results in sub-optimal properties because SiN materials cannot provide these properties and ignore the other. A nature. 201135962 Embodiments of the present invention generally provide a method of swelling a passivation/anti-reflective layer that provides the ability to combine this property with optical gradient properties in conjunction with a desired function a, which avoids the prior art The source of the brother's 5 Hai special nature. The inventors have found that the use of a normal power range (typically between 2000 and 3000 watts (w)) is more independent of the tendency of the rust to be used. High power|power density enables the formation of desired film properties to exhibit both optical and functional gradient properties. Beans are not based on traditional high-k values with high η values and low-k values and low η values, through the use of multilayer deposition (4) In forming the passivation layer, pinholes are prevented from extending through the entire film layer. Accordingly, embodiments of the present invention provide a method of forming a passivation/ARC layer having a high refractive index but a low extinction coefficient of 00. By specifically modifying the process chemistry for film formation, each of the layers of the multilayer passivation film can have the specific properties of bonding and forming passivation/arc layers with desired optical and functional properties. In one embodiment, the passivation layer may comprise one or more layers, or graded regi ns, having different compositions, different physical properties, and/or different electronic properties to provide passivation and optical properties. For example, in one embodiment, the passivation layer 12A contains a niobium-containing interfacial sub-layer 121 and a niobium-nitrogen-containing macro-sublayer 122 formed on the interfacial sub-layer 121, as shown in Figures ic to 1F. . The passivation layer forming process 1A to IF illustrate a schematic cross-sectional view of the solar cell substrate 11 at different stages in the process sequence 201135962, which is used to form the surface of the purification/ARC layer 120 on the solar cell 1 Upper (for example, top surface 1〇5). Figure 2 illustrates the procedure for forming a purification layer on a solar cell substrate no. The sequence in Figure 2 corresponds to the stages depicted in Figures 1A through 1F. In one embodiment of solar cell 100, P-type substrate 110 (having a bottom surface 1 〇 6 and comprising crystalline germanium) has a base region gm and an n-type doped emitter region 102 formed thereon, which is typically transmitted through a It is formed by doping and diffusion/annealing processes, but other processes including ion implantation can be used. The substrate 11A also includes a p_n junction region 103 disposed between the base region of the solar cell and the emitter region 102, and which is generated when the solar cell 100 is illuminated by photons of incident light. The area. Although the following discussion mainly deals with processing a substrate having an n-type emitter region formed on a p-type substrate region, the applicant does not want this configuration to limit the invention as described above because the passivation layer can also be formed. In the n-type substrate region, p-type emitter solar cell configuration. Referring to Fig. 2, during the preparation of the column 2, the surface of the substrate 11 is subjected to a plurality of processes. The processes are used to form the interface sub-layer 121 and the macro-man layer 122 on the surface of the substrate. This is followed by an example of a process 2〇12〇5, which can be performed in a processing chamber similar to process chamber 300 (Fig. 3). The process executed in all of the program queues 200 in one embodiment is performed in processing chambers 431-437 (Fig. 4) provided in one or more of the systems 400. The procedure for forming a purification layer on a solar cell substrate 110 201135962 200 generally begins with the removal of the underlying oxide layer underlying substrate, as shown in Process 20L of Figure 2. During normal processing of the solar cell elements, the native oxide layer U5 will be formed on one or more surfaces of the substrate 11〇. The surface of the process 201' base #110 is cleaned to remove the oxide layer 115 (Fig. 1A). In one embodiment, the cleaning process can be performed using a dry cleaning process in which the substrate (4) is exposed to a reactive plasma etch process to remove the oxide layer 115. In one embodiment, after the process 20 is disposed in the processing chamber (such as the chamber 3 of FIG. 3), the native oxide layer ι 5 is exposed to a reactive gas. It may contain nitrogen, fluorine and hydrogen. Next, the oxide layer 115 reactive with the reactive gas is subjected to heat treatment to remove it from the surface of the substrate. In some embodiments, the heat treatment can be an annealing process performed in another adjacent chamber provided in the processing chamber 300 or system 400. In the examples, it is desirable to ensure that the substrate is not exposed to oxygen for an extended period of time. Thus, in some embodiments of the invention, it is desirable to perform each of the processes 2G3_2G8 in an oxygen-free, inert and/or vacuum environment such as a vacuum tooling space of a cluster tool or system (Fig. Thus, the substrate is not exposed to oxygen between processes 203-208. In one embodiment, after the 帛2〇1 is performed on the batch substrate 110 disposed on the substrate carrier 425, the substrates are then positioned in the processing chamber such that the process performed at 202-206 can be performed. Performed on a substrate. Next, as shown in Fig. 2 and Fig. 1B, the interface layer (2) containing tantalum nitride is formed on the surface 1〇5 of the clean, oxide-removed substrate. In one embodiment, the interface sub-layer 121 can be between about 5 Å A and about A thick, such as 12 2011 35962 150 A thick. In one embodiment, the interface sub-layer 121 is formed over the top surface ι 5 using chemical vapor deposition (CV+D), electro-convexified chemical vapor deposition (PECVD), or physical vapor deposition (pVD) techniques. . In process 202, in one aspect, the method of forming interface sub-layer 121 includes flowing a first process gas mixture into process space 306 within the processing chamber. The process 203' plasma is generated in the process space 3〇6, and in the process 204, the process layer 306 containing the tantalum nitride interface sublayer 121 is deposited on the solar cell substrate 110. Then, in the process 2〇5 shown in FIG. 2 and FIG. 1C to FIG. 1D, a plasma-enhanced chemical vapor deposition (PECVD) process is used to form a giant sub-layer 122 containing tantalum nitride at the interface sublayer. 121, thus forming a multilayer purified anti-reflective coating 120. Alternatively, a macrosecond layer 122 can be formed over the interface sublayer m using chemical vapor deposition (CVD) or physical vapor deposition (PVD) techniques. The PVD process can be used for reactive sputtering in a hydrogen atmosphere (atm〇sphere) to form a multilayer passivation antireflective layer. For example, the target can be sputtered with argon in a nitrogen and hydrogen atmosphere to deposit each siN layer. In one embodiment, the macro-sublayer 122 can be from about 400 A to about 700 A thick, such as 600 A thick. In one example, the substrate is exposed to an RF plasma of 13 56 MHz to form both the interface sublayer 12i and the bulk sublayer 122. In one embodiment, the first and second process gas mixtures comprise a ruthenium containing ruthenium and a nitrogen containing precursor. For example, the first process gas mixture may comprise decane (SiH4), nitrogen (NO and/or ammonia (NH3). The second process gas mixture may comprise decane and nitrogen, decane and ammonia (NH3), or decane, ammonia and nitrogen. Table 1 details the process conditions that can be used to form the interface sub-layer 121 and the giant 13 201135962 sub-layer 122 by PECVD. Table 1 lists the VR 、, rate, which is based on the space per liter. ♦ 1 , 4 nitrogen and / or ammonia μ such as Ν 2) for the ruthenium-containing precursor (Example 3: 1 also includes the nitrogen-containing precursor (such as the accumulation of shovel Α Α Α A1 4) flow rate ratio, the power density of each deposition system %, The distance between the early mother and the required bamboo joints of each layer*, the spacing between the supports, and the ratio of 121 to her, such as her, is used to form the interface sub-layer 121, the nitrogen flow rate. For each ..vv / liter air space is about 77.30 standard ah (seem), while the decane flow rate is about 2.25 seem. The interface sub-layer m can be taken from 1 每 per minute. A rate deposition of up to 3,000 people, for example, 1 minute per minute, can be deposited at a rate of more than 3000 每 per minute. In another embodiment, the first process gas mixture may also include a nitrogen (out) diluent, which may be added to the flow rate of decane, ammonia, and nitrogen as shown in Table 1 at a flow rate of liter chamber volume J i 0 sccm. Although not shown in the table, the ratio of ammonia to decane flow rate can be about 0.90 in a large number of sub-layer process formulations. It is believed that careful control of the decane gas flow helps achieve the desired optical and functional properties of the film. In the sub-layer process conditions, the process gas flow is generally souther. The average technician can successfully modify the gas flow ratio depending on the power, pressure, spacing, and temperature of the deposition process. N2 flow rate (sccm/1) NH3 flow rate (sccm/1) SiHU flow rate (sccm/1) N2/S1H4 Ratio power density (W/cm2) Spacing (mils) Time (Sec.) Interface 77.30 0.00 5.25 14.70 .65-1.0 800 9 14 201135962 Sublayer massive Sublayer 77.30 8.40 9.20 8.35 .65-1.0 1100 15 Table 1 In one embodiment, the substrate temperature for both the interface and the bulk sublayer deposition process can be maintained at (^ to 40 (rc, such as from 38〇) (: to 39〇c. Various means for maintaining the temperature of the substrate can be For load lock heating, substrate support heating, plasma heating, etc. The secondary layer can be deposited at a chamber pressure of about 1.5 τ. The plasma used for the parent layer can be provided by RF power, as compared to The range of 2 〇〇〇 _ 3 〇〇〇 w of a typical SiN deposition process, which is between about 4350 W and about 6700 W, such as about 5000 W, at the frequency of the MHZ. RF power can be supplied to the showerhead 31 and/or the substrate support 33A. The RF power density for both the interface sub-layer and the bulk sub-layer deposition may be about 65.65 w/em2 or more of the surface of the substrate to form a plasma. For example, in some embodiments, the RF power density can be 丨〇〇 w/cm 2 . In another embodiment, the RF power density can be about w 75 w/cm 2 . The power density can be as long as the enthalpy power increases the positive charge in the interface and the massive sub-layer. Therefore, higher power provides better field passivation and reduces SRV. 15 201135962 In another embodiment, the boundary between the interface sub-layer and the massive sub-layer can be more clearly achieved by stopping and restarting the plasma (while the process gas is converted from the interface sub-layer formulation to the giant sub-layer formula) Bundle eight..., the transformation zone is defined. This transition can occur in a variety of ways. For example, right 筮 _ & The flow of the first escaping # secret music mixture can be stopped before the first process gas mixture is introduced into the processing chamber. In another embodiment, the precursor of the remaining first process gas mixture continues to flow when only the stagnation gas sheet pull 70 is stopped. For example, j w ’ during the power-off period, the airflow formulation for the interface sub-layer can be converted into a huge sub-layer formula. The airflow will not provide an immediate perfect gas mix. Therefore, this break in the electrodeposition process will cause the gas for the bulk passivation layer formulation to be properly mixed before the plasma power is restarted. The "break" of the process after the process 204 is continued for about 2 seconds, which allows the first process gas mixture to be substantially flushed from the chamber before the second process gas mixture flows into the chamber. During the transition of the process recipe, the decane stream can be raised to distribute it equally through the chamber, after which power is again supplied to the showerhead to reignite the plasma' thus ending the formation of the purification layer 12'. Turning off the plasma power after the interface sublayer deposition and restarting the plasma for massive sublayer deposition increases film density and ultimate performance by making layers that are more desirable and suddenly transitioning. During the break, the substrate temperature is generally maintained at about 38 (rc to about 39 (rc temperature. In other embodiments, the power can be raised to the final power setting during the massive sublayer deposition process. In some tests ten stops) And restarting the power to increase the open circuit voltage (V〇c) of the film by 3 mV (millivolts). In another embodiment, the formation of a large number of sublayers can be passed through the interface sublayer formulation without "breaking" Transition to a large sub-layer formulation (as shown in Table 16 201135962 1) occurs - thus creating a transition layer with no defined stoichiometry. The gas flow rate and / or gas mixture composition is changed to change from one process recipe to the next The formulation 'turns on the plasma power at the same time. The transition layer in this embodiment can be 3-5 nm thick, or less than 10% of the thickness of the giant sublayer' but is 7-8% of the total deposited passivation layer thickness. The interface sub-layer 121 and the macro-sub-layer 122 (as schematically illustrated in Figures 1C-1D) form a passivation/arc layer 120 overlying the top surface 105 of the p-doped region. In one embodiment, formation is desired Multi-layer passivated ARC layer 120 containing a desired amount of capture The positive charge is provided to provide surface passivation of the desired p-type region. In another embodiment, the captured positive charge provides surface purification of the desired n-type substrate having an n-type doped region. In one embodiment, the multilayer passivation coating The sum of the total amount of positive charges Q! captured in layer 12〇 and the total amount of captured negative charges Q2 has a sufficient amount of trapped electricity to achieve a charge of about 1×10 12 Ccmlombs/cm 2 (Coulomb number per square centimeter) or more. Density, such as between about lxl0i2c〇ul〇mbs/cm2 to about lxlO14 Coulombs/cm2, or about 2xl〇i2
Coulombs/cm2 至約 4xl〇n Coulombs/cm2 之間。另一實施 例中，總捕捉正電荷可從約5xl〇ii c〇ul〇mbs/cm2至約 lxlO13 Coulombs/cm2 » 在期望的電荷密度數前缺少負號僅欲意味表面ι〇5所 經歷的電荷是正對負，而因此Qi的絕對值大於h的絕 對值。在界面次層121與巨量次層122各含有正電荷與 負電何的情況中，在此論及的Qi及Q2的值為電荷的淨 值，或各層中正電荷總量的絕對值減去負電荷總量的絕 17 201135962 :值之總和。大體而言’正電荷量愈高，太陽能電池的 壽命愈長’而SRV愈低。在-些實施例中，期望將捕捉 的電街Qi的巨罝部定位在離表面1〇5在1〇〇埃（A)以 下處，以確保捕捉的電荷將具有期望的場強度，以除拒 表面105處或表面105下方的電洞，因為除拒電洞的能 力將隨捕捉的正電荷Q 1與電洞之間的距離平方分之一 (Ι/d2)變化。 表2與表3詳述根據本發明實施例形成的界面次層i2i 與巨量次層122各物理及電子性質。表2顯示每一次層 中的原子氫百分比、折射率（n)與消光係數（k)值的範圍、 與膜密度。界面次層121可具有氫（H+)濃度，以助基材 110的巨量與表面鈍化於諸如12%(原子百分比）左右。 巨里次層122可具有諸如18% (原子百分比）左右的H+ 濃度。沉積鈍化/ARC層120後於基材上執行的燒成或退 火製程期間，巨量次層122中的氫將被汲取進入基材以 亦提供基材的巨量鈍化。界面與巨量次層中的H+離子儲 存區可汲引到基材與矽基材的頂表面，以鈍化任何基材 結構中的空位或晶體缺陷。因界面次層i 2丨是基材1【〇 頂表面105上的唯一層，其可具有低於1〇 cm/sec的表面 組速率（SRV) ’然而巨量次層不接觸頂表面ι〇5，因此巨 量次層的SRV不重要。 膜性質 H% η k 密度（g/cm3) 界面次層 5%-15% 2.4-2.6 <0.04 2.5-3.0 18 201135962 巨量次層 2.3-2.9Coulombs/cm2 to approximately 4xl〇n Coulombs/cm2. In another embodiment, the total positive charge can be captured from about 5xl〇ii c〇ul〇mbs/cm2 to about lxlO13 Coulombs/cm2 » lack of a negative sign before the desired charge density number is only intended to mean the surface ι〇5 experienced The charge is positive and negative, and therefore the absolute value of Qi is greater than the absolute value of h. In the case where the interface sub-layer 121 and the macro-sub-layer 122 each contain a positive charge and a negative charge, the values of Qi and Q2 referred to herein are the net value of the charge, or the absolute value of the total amount of positive charges in each layer minus the negative charge. The total amount of the absolute 17 201135962: the sum of the values. In general, the higher the positive charge, the longer the life of the solar cell, and the lower the SRV. In some embodiments, it is desirable to position the giant crotch portion of the captured electric street Qi at a distance from the surface 1 〇 5 below 1 〇〇 (A) to ensure that the captured charge will have a desired field strength to divide The hole at the surface 105 or below the surface 105 is rejected because the ability to remove the hole will vary with the square of the distance between the captured positive charge Q1 and the hole (Ι/d2). Tables 2 and 3 detail the physical and electronic properties of the interface sub-layer i2i and the macro-sub-layer 122 formed in accordance with an embodiment of the present invention. Table 2 shows the atomic hydrogen percentage, the range of the refractive index (n) and extinction coefficient (k) values, and the film density in each layer. The interface sub-layer 121 may have a hydrogen (H+) concentration to assist the bulk and surface passivation of the substrate 110 to be, for example, about 12% (atomic percent). The giant sub-layer 122 may have an H+ concentration such as about 18% (atomic percent). During the firing or annealing process performed on the substrate after deposition of the passivation/ARC layer 120, hydrogen in the bulk sub-layer 122 will be drawn into the substrate to also provide substantial passivation of the substrate. The interface and the H+ ion reservoir in the macrolayer can be directed to the top surface of the substrate and the tantalum substrate to passivate vacancies or crystal defects in any substrate structure. Since the interface sublayer i 2丨 is the only layer on the substrate 1 [the dome surface 105, it may have a surface group velocity (SRV) of less than 1 〇cm/sec.] However, the giant sublayer does not contact the top surface ι〇 5, so the huge sub-layer SRV is not important. Membrane properties H% η k Density (g/cm3) Interface sublayer 5%-15% 2.4-2.6 <0.04 2.5-3.0 18 201135962 Massive sublayer 2.3-2.9
1〇%-25% 2.05-2.15 表2 沉積的界面次層121可具有大於巨量次層122的折射 率⑻’且界面次4 121與巨量次層122二者可具有從。 到0.1的k值。大體而言，折射率是經選擇以用於具有 純化/ARU 12G的所得電池的封裝上，其接觸用於在形 成製程結束時包覆太陽能電池的黏結材料。當的黏結 材料的-些|a例包括乙基乙酸乙烯酯⑽心㈣“⑽价， EVA)與聚乙烯醇縮丁酸（p〇lyWnyi㈤㈣，pvB)。因此，’ 由於光通過玻璃(n=1·5)、黏結材料(η=1·5)、包括巨量次 層122與界面次層121的鈍化/ARC層12〇、及矽基材 (η=3·0)’當光通過每一介質時從每一層界面反射的光量 將會減少，因每一連續介質之間的折射率差大體上是小 的，故減少了光從玻璃下方、射極區域1〇2上方的膜反 射的量。例如界面次| 121與巨量次層122的η值可 各為2.4與2.08。 表3顯示每一次層的N_H/Si_H比率範圍、膜應力、當 於膜的400 A處測量時的平帶電壓（Vfb)、以及膜厚度。 各層所測量的厚度為如在紋理化表面上測量的厚度。因 此，這些厚度是從每一層的表面到表面。例如，若使用 研磨過的基材於與監視產品基材同時的製程，則產品基 材上的膜層121、122的所得厚度會低於研磨過的基材一 已知因數，有時稱為空間因數（space factor)。空間因數 19 201135962 可為70%到90%左右 即產品基材上的層厚可為約7〇%1〇%-25% 2.05-2.15 Table 2 The deposited interface sub-layer 121 may have a refractive index (8)' greater than the giant sub-layer 122 and both the interface sub-4111 and the macro-sub-layer 122 may have. The k value to 0.1. In general, the index of refraction is selected for use on a package of the resulting cell having a purification/ARU 12G that contacts the bonding material used to coat the solar cell at the end of the forming process. When the bonding material - a | a case includes ethyl vinyl acetate (10) heart (four) "(10) valence, EVA) and polyvinyl butyric acid (p〇ly Wnyi (five) (four), pvB). Therefore, 'because light passes through the glass (n = 1·5), bonding material (η=1·5), passivation/ARC layer 12〇 including giant sub-layer 122 and interface sub-layer 121, and germanium substrate (η=3·0)' when light passes through each The amount of light reflected from each interface at a medium will be reduced because the refractive index difference between each successive medium is substantially small, thereby reducing the reflection of light from the film below the glass and above the emitter region 1〇2. For example, the η values of the interface times | 121 and the giant sub-layers 122 may each be 2.4 and 2.08. Table 3 shows the N_H/Si_H ratio range of each layer, the film stress, and the flatness measured at 400 A of the film. With voltage (Vfb), and film thickness. The thickness measured by each layer is the thickness as measured on the textured surface. Therefore, these thicknesses are from the surface of each layer to the surface. For example, if a ground substrate is used, By monitoring the simultaneous process of the product substrate, the resulting thickness of the film layers 121, 122 on the product substrate will be lower than the ground basis. A known factor, sometimes referred to as the space factor (space factor). 19201135962 space factor may be about 70% to 90% of the thickness of the product i.e. the substrate may be from about 7〇%
N-H/Si-H鍵比率對修改次層121、122的光學性質而 言是重要的。透過在沉積鈍化層120期間增加電漿功 率相彳s相較於先前技術所用的較低之電漿功率設定（例 如2000-3000 w之間）所沉積的鈍化層，所得的沉積膜 將具有增加的折㈣及純的U ^在具有氮切型鈍 化層的矽太陽能電池中發現的至少兩種類型的鍵結引發 光的吸收：S-Η鍵與Si-Si鍵。然而，S-Η鍵與Si-Si鍵 並非氮化矽材料（理論上是全部為Si#4，有時是指化學 當量的氮化矽）# 一部份。然而，僅$叫队的化學當 量氮化矽臈是低劣的太陽能ARC材料，因為其中無氫， 會造成低劣的總太陽能電池效能，因化學當量的氮化矽 的折射率為1 ·9左右。因此，需要添加氫到氮化矽ARC 層’以進一步強化其抗反射與鈍化性質。 然而’當氫添加到矽基材上的氮化矽膜時，一歧鱼 201135962 以形成N-H鍵與Si-H鍵。Si_H材料吸收uv範圍邊緣 的光，並且貢獻總k值次要的部份，然而材料吸收 可見光’因此貢獻總k值主要的部份。為了使折射率至 太陽能電池期望的層級’需要額外的石夕。然而，該額外 的夕不與其他⑦鍵結。相信在沉積純化層期間使用高 電毁密度抑制形成S“Si鍵（雖然不必然完全防止）並且 增強队N及/或Si_H鍵在生長膜中的形成。換言之，高 電聚功率密度減少生長膜中及/或最初在基材表面處的 矽原子與含碎前驅物氣體中的矽原子鍵結。透過減少 Si-Si鍵結及增加生長膜中的⑽與㈣鍵結的百分 比’沉積的氮化石夕膜之k值可受到調整及/或控制。在一 些組態中’期望調整電|功率密度以修改或調整所形成 的氮切層之功率層級可為先前用於形成氮化石夕 層之正常層級的兩倍，以斷裂期望的鍵結並JL使通常直 接成正比的！^與k之間的關係解開（dec〇uple)。 因此，相較於使用先前技術方法形成的SiN鈍化膜， 與石夕基材接觸的界面次I 121具有高折射率及最適光穿 透品質，並且鈍化了 Si_SiN界面的缺陷。沉積在界面次 層121頂部上的巨量次層122可經修改以用於電荷儲 存、氫留駐、最適應力與高沉積速率（相較於界面次層 121)。組份層的厚度可經挑選以減少光的吸收性損失， 同時亦維持從射極區域上方的層反射的光量至最小值。 第1D圖是鈍化/八狀層12〇的特寫剖面視圖，說明各 位於巨量次層122與界面次層121中的針孔i3〇、i3i。 201135962 所描繪的針孔有筆直側壁，儘管一般為不規則、非線性 形狀的針孔。化學氣相沉積製程透過成核啟動膜生長， 接著是晶粒生長。當鄰近的晶粒繼續生長至晶粒聚結點 以形成連續膜時，整體積材區域被膜覆蓋。然而，由於 透過成核與晶粒聚結的一般膜生長模式之本質之故，有 很高的機會不連續（諸如小間隙、空洞、及孔洞）會導 入聚結晶粒的邊界。於該等邊界形成的膜中缺陷（諸如 針孔）使得污染（諸如氧或濕氣）抵達下伏的摻雜區域 102，導致壽命減少及太陽能電池的失效率增加。一旦諸 域孔之通道或空洞狀缺陷纟pECVD帛生長的最初階 段導入，針孔會在膜增厚時持續抵達膜表面。 氨氣與石夕院氣體-般用於PECVD a化石夕膜的高速尋 沉積H料沉積料A體上造成膜的残續，^ :咖製程期間的針孔狀缺陷。一項減少此類缺㈣ 是減少透過使用氛、氣、氣及錢之膜的化, 製程條Γ積速率。藉由小"控制氨、氮與㈣前驅物穿 A)在美广相對較低的沉積速率（諸如每分鐘100 土上生長近乎無針孔的pecvd。但，對給定 度的SiN而言，較^^ '° 大想上，㈣與二 長的生產時間。 但也増加的増加會增加SiN沉積速率， s扪針孔產生的可能性。 然而，若沉積多層Si為膜，任 :能…準另-層一狀缺陷I:::: 造成此類缺陷的高沉因此已知固有地 積速率可能仍會用於形成構成鈍化 22 201135962 /ARC層120的次層。因此，在頂表面（即巨量次層i22 的外表面）與底表面（即界面次層121與摻雜區域 之間的界面）之間延伸的針孔的可能性會大幅減少。如 第1D圖所示，鈍化/ARC層12〇實質上無完全通過界面 與巨里次層二者的針孔13〇、131。針孔13〇、131大體 上不對準，並且創造出「曲折路徑（t〇rtu〇us path)」，即， 從頂表面至底表面的路徑會是曲折的路徑，因為針孔並 不在兩個表面間形成大體上筆直的通道或直接的途徑。 換σ之針孔仍可形成於各次層121、122中.，但該結合 的層了 a質上無完全通過該二層的針孔，完全通過即從 巨量次層的頂表面至界面次層的底表面。 本發明的另一實施例中，提供一種於c Si (結晶矽） 太陽能電池上針孔偵測的方法。針孔尺寸取決於膜晶粒 尺寸，般直徑/寬度的範圍是從數十奈米（nm)至數百奈 米。然而偵測此類針孔並非直截了當地簡單，除非針孔 豐存。 使用電鍍技術的技術可用於揭發太陽能電池基材上鈍 化層（例如c_Si太陽能電池基材上的pECVD SiN)中的 針孔狀缺陷。此技術的優點是其能應用至完成的太陽能 電池。具有鈍化層（諸如PECVD SiN)的太陽能電池浸 满在電解液中’太陽能電池的背側有適當絕緣，同時透 過太陽％電池的金屬覆蓋（例如鋁）之背側供應電流。 功率施加到石夕，且若電解液接觸矽，則其形成電流，觸 發針孔的鍍覆。 23 201135962 使用此方法’針孔從鈍化/ARC層1 20延伸到摻雜石夕區 域102之處’來自鍍覆溶液的金屬（例如在電解液包括 CuS〇4時’其為銅）將會鍍覆到針孔上，並且開始生長 覆於針孔上’ 一旦其完全填滿針孔則最後會向外鍍覆。 然而’因為在針孔延伸至摻雜層丨〇2處將會形成通過石夕 與電解液的迴路’引發銅從電解液鍍覆於針孔中，所以 諸如銅之類的金屬將不會鍍覆於不完全延伸通過鈍化 /ARC層120的針孔中。鍍覆的銅在基材面上呈現亮圓 點’其尺寸比針孔開口大得多’並且視鍍覆時間而定， 其在尺寸上從數微米變化到數百微米。鍍覆的銅的閃斑 使得針孔得以在光學顯微鏡下被偵測到。 形成含氮化矽巨量次層後，太陽能電池可進一步受處 理’以提供其他保護層，如第1E圖與第1F圖所示，其 說明兩種元成的太陽能電池100的不同實施例。第1E圖 描繪具有多層鈍化抗反射塗層120的太陽能電池1〇〇, 黏結材料層124與玻璃基材126放置在在該塗層上。黏 結材料層124包覆太陽能電池基材以及任何其上形成的 特徵結構，以與玻璃基材126形成保護層。 另一實施例中，欲形成保護層於鈍化層上，可將第三 製程氣體流入製程空間以沉積含矽、氧、氮層（諸如氮 氧化矽（SiON))於巨量次層122，如製程2〇7_2〇8所指示 及第1F圖所示。此實施例一般是經製做以用於實驗室研 究與一般研究與測試。氮氧化矽層可具有19_18的η 值。第三製程氣體混合物可含有矽烷、Α甲基環狀四矽 24 201135962 氧烷（〇Ctamethylcycl〇tetrasil〇xane, 〇MCTS)、矽酸四乙 酯（tetraethyl 〇rth〇Silicate，TE〇s)、〇2、〇3、n2〇、N〇2、 ΝΑ、Η?及N2之至少一者。此舉可在無背玻璃蓋片或基 材用在太陽能最終結構中之實施例中完成。 太陽能電池基材可在尖峰燒成（spike firing)製程中於 850 C退火1秒，如209所示。氮化矽沉積後高溫燒成太 陽能電池改善巨量氮化矽層品質，並且將氫汲引進入基 材。δ亥燒成製程可包括各種升溫加熱與冷卻，其可取決 於太1% Bb電池中所用的糊狀物之類型。另一實施例中， 退火可發生在 >儿積含氣化石夕巨量次層之後，如21 〇所 示。諸如EVA或PVB的黏結材料隨後置於巨量次層上， 之後’背玻璃基材配置在黏結材料上，且隨後層壓以完 成太陽能基材製造製程，如211及212所示。然而，無 論太陽能電池上是否執行燒成製程，皆可達成表面鈍化。 硬體组態 設以處理大面積基材的電漿強化化學氣相沉積 (PECVD)能夠在高沉積速率下沉積具卓越的膜均勻性的 SiN層。此對於平行板、高頻PECvd系統而言特別真實， 該系統中一個或多個基材在電漿腔室中定位在兩個實質 上平行的電極之間。腔室的氣體分配板大體上做為第一 電極，而腔室的基材支撐件做為第二電極。前驅物氣體 混合物導入腔室，藉由施加射頻（RF)功率至該等電極之 一而被賦能至電漿態’並且橫越基材表面流動以造成期 25 201135962 望材料層沉積。腔室的幾何結構與在沉積平面正上方的 氣體喷頭最適於以南處理量創造多重漸變層，而不對系 統增加貫質的成本、尺寸或複雜度。 難以用其他類型的處理腔室達成氮化矽鈍化膜的高η 與低k二者性質，因難以在那些處理腔室中形成高電漿 功率密度層。例如，其他處理腔室可能使用具許多電子 連接在一起的基材之碳舟（carb〇n boat)系統。碳舟系統中 的每一基材形成陽極與陰極，相信其阻止了此述的處理The N-H/Si-H bond ratio is important for modifying the optical properties of the sub-layers 121, 122. The resulting deposited film will have an increase by increasing the plasma power phase s during deposition of the passivation layer 120 compared to the lower plasma power set (eg, between 2000 and 3000 w) of the passivation layer deposited prior to the prior art. Folding (4) and pure U^ at least two types of bonding found in tantalum solar cells with a nitrogen-cut passivation layer induce light absorption: S-Η bonds and Si-Si bonds. However, the S-Η bond and the Si-Si bond are not part of the tantalum nitride material (theoretically all are Si#4, sometimes referred to as chemically equivalent tantalum nitride)#. However, only the chemical cesium nitride of the team is a poor solar ARC material because there is no hydrogen, which will result in inferior total solar cell performance, because the chemical equivalent of lanthanum nitride has a refractive index of about 1.9. Therefore, it is necessary to add hydrogen to the tantalum nitride ARC layer' to further enhance its anti-reflection and passivation properties. However, when hydrogen is added to the tantalum nitride film on the tantalum substrate, a fish 201135962 is formed to form an N-H bond and a Si-H bond. The Si_H material absorbs light at the edge of the uv range and contributes a minor portion of the total k value, whereas the material absorbs visible light' thus contributing the major portion of the total k value. In order to bring the refractive index to the desired level of the solar cell, an additional stone eve is required. However, this extra eve is not tied to the other 7 keys. It is believed that the use of high electrical destruction density during the deposition of the purification layer inhibits the formation of S"Si bonds (although not necessarily completely prevented) and enhances the formation of the N and/or Si_H bonds in the growth film. In other words, the high electropolymer power density reduces the growth film. Neutral and/or initially ruthenium atoms at the surface of the substrate are bonded to ruthenium atoms in the gas containing the precursor. By reducing Si-Si bonding and increasing the percentage of (10) and (4) bonds in the growth film. The k value of the fossil film can be adjusted and/or controlled. In some configurations, the power level desired to adjust the power density to modify or adjust the formed nitrogen layer can be previously used to form the nitride layer. Twice the normal level, to break the desired bond and JL to make the relationship between ^ and k, which is usually directly proportional, dec〇uple. Therefore, compared to SiN passivation film formed using prior art methods The interface sub-I 121 in contact with the stone substrate has a high refractive index and an optimum light transmission quality, and passivates the defects of the Si_SiN interface. The giant sub-layer 122 deposited on the top of the interface sub-layer 121 can be modified to be used. Charge storage Hydrogen retention, most adaptability, and high deposition rate (compared to interface sub-layer 121). The thickness of the component layer can be selected to reduce the loss of light absorption while maintaining the amount of light reflected from the layer above the emitter region to The first dimension is a close-up cross-sectional view of the passivation/octagonal layer 12〇, illustrating the pinholes i3〇, i3i located in the macrosecond layer 122 and the interface sublayer 121. The pinholes depicted in 201135962 have straight sidewalls Although generally irregular, non-linear shaped pinholes, the chemical vapor deposition process grows through the nucleation promoter film, followed by grain growth. When adjacent grains continue to grow to the grain coalescence point to form a continuous film The entire volume area is covered by the film. However, due to the nature of the general film growth mode through nucleation and grain coalescence, there is a high chance that discontinuities (such as small gaps, voids, and pores) will be introduced into the polycrystal. The boundaries of the particles. Defects (such as pinholes) in the film formed at the boundaries cause contamination (such as oxygen or moisture) to reach the underlying doped region 102, resulting in reduced lifetime and increased failure rate of the solar cell. Once the channels or void-like defects of the pores are introduced into the initial stage of growth, the pinholes will continue to reach the surface of the membrane as the membrane thickens. Ammonia gas and Shixiyuan gas are generally used for PECVD a fossil membrane High-speed smear deposition of H material deposits on the body A to cause film residue, ^: pinhole-like defects during the coffee process. One reduction of such defects (four) is to reduce the use of membranes, air, gas and money. Process bar hoarding rate. By small "controlling ammonia, nitrogen and (iv) precursors through A) at a relatively low deposition rate in the United States and Guangxi (such as growing almost pinhole-free pecvd on 100 soils per minute. For the determined SiN, it is better than ^^ '°, (4) and two long production times. However, the addition of yttrium increases the deposition rate of SiN, and the possibility of pinhole generation. However, if a plurality of layers of Si are deposited as a film, any one of the defects can be caused by a high level of such defects. Therefore, it is known that the inherent rate of accumulation may still be used to form a passivation 22 201135962 /ARC The sublayer of layer 120. Therefore, the possibility of pinholes extending between the top surface (i.e., the outer surface of the giant sub-layer i22) and the bottom surface (i.e., the interface between the interface sub-layer 121 and the doped region) is greatly reduced. As shown in Fig. 1D, the passivation/ARC layer 12 is substantially free of pinholes 13A, 131 that completely pass through both the interface and the giant sub-layer. The pinholes 13〇, 131 are substantially misaligned and create a “t〇rtu〇us path”, ie the path from the top surface to the bottom surface will be a tortuous path because the pinholes are not in two A substantially straight passage or direct path is formed between the surfaces. The pinhole for σ can still be formed in each of the sub-layers 121, 122. However, the bonded layer does not completely pass through the pinhole of the two layers, and completely passes through the top surface of the massive sub-layer to the interface. The bottom surface of the secondary layer. In another embodiment of the invention, a method of pinhole detection on a cSi (crystalline germanium) solar cell is provided. The pinhole size depends on the film grain size, which ranges from tens of nanometers (nm) to hundreds of nanometers. However, detecting such pinholes is not straightforward, unless the pinholes are abundant. Techniques using electroplating techniques can be used to uncover pinhole defects in passivation layers on solar cell substrates, such as pECVD SiN on c_Si solar cell substrates. The advantage of this technology is that it can be applied to completed solar cells. A solar cell having a passivation layer (such as PECVD SiN) is immersed in the electrolyte. The back side of the solar cell is suitably insulated while current is supplied through the back side of the metal cover (e.g., aluminum) of the solar cell. Power is applied to Shi Xi, and if the electrolyte contacts 矽, it forms an electric current that triggers the plating of the pinhole. 23 201135962 Using this method 'pinholes extend from passivation/ARC layer 1 20 to doped day zone 102' metal from plating solution (eg copper when the electrolyte includes CuS〇4) will be plated Covers the pinhole and begins to grow over the pinhole'. Once it has completely filled the pinhole, it will eventually be plated outward. However, 'Because the pinhole extends to the doped layer 丨〇2 will form a loop through the stone and the electrolyte' induces copper to be plated from the electrolyte into the pinhole, so a metal such as copper will not be plated. Covering the pinholes that do not extend completely through the passivation/ARC layer 120. The plated copper presents a bright dot on the substrate surface 'which is much larger than the pinhole opening' and varies in size from a few microns to hundreds of microns depending on the plating time. The flashing of the plated copper allows the pinhole to be detected under an optical microscope. After the formation of a massive sub-layer containing tantalum nitride, the solar cell can be further processed to provide additional protective layers, as shown in Figures 1E and 1F, which illustrate different embodiments of the two-component solar cell 100. Figure 1E depicts a solar cell 1 having a multilayer passivation anti-reflective coating 120 on which a layer of bonding material 124 and a glass substrate 126 are placed. The layer of bonding material 124 encases the solar cell substrate and any features formed thereon to form a protective layer with the glass substrate 126. In another embodiment, to form a protective layer on the passivation layer, a third process gas may be flowed into the process space to deposit a layer containing germanium, oxygen, nitrogen (such as cerium oxynitride (SiON)) in the macrosecond layer 122, such as The process is indicated by 2〇7_2〇8 and shown in Figure 1F. This embodiment is generally made for laboratory research and general research and testing. The hafnium oxynitride layer may have an η value of 19-18. The third process gas mixture may contain decane, hydrazine methyl cyclic tetracycline 24 201135962 oxane (〇Ctamethylcycl〇tetrasil〇xane, 〇MCTS), tetraethyl 〇rth〇Silicate (TE〇s), 〇 2. At least one of 〇3, n2〇, N〇2, ΝΑ, Η? and N2. This can be done in embodiments where the back glass cover sheet or substrate is used in the final structure of the solar energy. The solar cell substrate can be annealed at 850 C for 1 second in a spike firing process, as indicated at 209. The high-temperature firing of the solar cell after the deposition of tantalum nitride improves the quality of the massive tantalum nitride layer and introduces hydroquinone into the substrate. The delta firing process can include a variety of elevated temperature heating and cooling, depending on the type of paste used in a 1% Bb battery. In another embodiment, the annealing may occur after the > gas-bearing fossil-rich sub-layer, as shown in Figure 21. A bonding material such as EVA or PVB is then placed over a massive sub-layer, after which the back glass substrate is placed over the bonding material and subsequently laminated to complete the solar substrate fabrication process, as indicated by 211 and 212. However, surface passivation can be achieved regardless of whether or not a firing process is performed on the solar cell. Hardware Configuration Plasma-enhanced chemical vapor deposition (PECVD) designed to process large-area substrates can deposit SiN layers with excellent film uniformity at high deposition rates. This is especially true for parallel plate, high frequency PECvd systems in which one or more substrates are positioned between two substantially parallel electrodes in a plasma chamber. The gas distribution plate of the chamber is generally used as the first electrode, and the substrate support of the chamber serves as the second electrode. The precursor gas mixture is introduced into the chamber and is energized to the plasma state by application of radio frequency (RF) power to one of the electrodes and flows across the surface of the substrate to cause deposition of the material layer. The geometry of the chamber and the gas jet directly above the deposition plane are best suited to create multiple graded layers with a south throughput without adding cost, size or complexity to the system. It is difficult to achieve both high η and low k properties of tantalum nitride passivation films with other types of processing chambers due to the difficulty in forming high plasma power density layers in those processing chambers. For example, other processing chambers may use a carb〇n boat system with a plurality of substrates that are electronically coupled together. Each substrate in the carbon boat system forms an anode and a cathode, which is believed to prevent the treatment described herein.
腔室中生成的類似類型的電漿。然而，以此述的pECVD 工具，相信平行板類型的處理腔室之電容式耦合能致使 較面電漿功率密度用以形成具有期望光學與功能性質的 鈍化膜 第3圖是電漿強化化學氣相沉積（pECVD)腔室3〇〇之 一個實施例的概略側視圖’該腔室中，可執行與第2圖 一併說明及論述的一個或多個製程。類似組態的電漿強 化化學氣相沉積腔室可購自位於美國加州sanuc⑽的 應用材料公司。應考量到可利用包括購自其他製造商的 其他沉積腔室可用於操作本發明。 相信設於處理腔室300由认泰a*上 至中的電漿處理組態在用於執行 第2圖所述之一個或多個劁 a ^ 3¾程時’具有勝於其他組態的 顯著優點。-個實施例中’ pEcvD腔室則設以一次處 理多個基材。-個實施例中，電聚強化化學氣相沉積 (PECVD)腔室300適於m拉♦ 、Π時處理以平面陣列排列在基材 載具425上的多個基材llf)r 签何Π0 (如第4圖所示），其與基材 26 201135962 的處理垂直堆疊（例如，諸如在垂直爐系統中執行的卡 匣中堆疊的數批基材）呈對比。處理數批以平面陣列排 列的基材容許每一批次中的基材直接且同時暴露所生成 的氣體、輻射熱、及/或處理氣體。因此，平面陣列中的 每一基材在處理腔室的製程空間中類似地受處理，而此 不仰賴擴散類型製程及/或習知設置的批次中一系列能 量轉移至所有受處理的基材，諸如堆疊或背對背設置的 一批基材。 一個組態中’ PECVD腔室300適於接收基材載具425 (第3圖與第4圖），該基材載具設以在傳送與處理基材 時固持一批基材。一個實施例中，基材載具425具有約 10000 cm2以上的表面積，諸如約4〇〇〇〇 cm2以上或約 55000 cm2以上，其設以支撐處理期間配置於其上的數個 基材no之平面陣列。一個實施例中，基材載具425固 持約4個至約49個之間的太陽能電池基材，該等基材尺 寸上為156mmX156mmX〇.3mm，呈面向上或面向下之組 態。-個組態中’一批太陽能電池基材在真空或惰性環 境（例如第4圖的傳送腔室42〇) +於基材載i化上 在數個處理腔室之間傳送，以減少污染機會並且改善處 丹度參考第 3〇_ 圃’腔室3〇0大體上包括界定製程空ft 的壁3〇2、底部取、喷頭則、以及基材切件别 過閥308存取製程空間3〇6’使得配 上的基材可傳送進出腔室。基材支料33〇載包括: 27 201135962 收表面3 3 2盘心飪·3。z ^ ^ 34 ’其耦接舉升系統（諸如舉升銷 338)以抬升及（I备_ 土材支撐件3 3 0。遮蔽框3 3 3 (維持 於腔室300内）可視情況置於基材載具425周邊之上。 舉升銷338以可移動式配置成穿過基材支撑件330以將 土材載具425移至、移出接收表面332。基材支樓件別 亦可包括嵌入的加熱及’或冷卻元件338，以將基材支撐 牛〇.准持在期望的溫度。基材支撐件33〇亦可包括接 也帶331以提供基材支撐件330周邊的rf接地。一個 實施例中|材支樓件33G具有輕接電極（圖中未示） 的RF源（圖中未示）’該電極嵌在基材支撐件330中， 使得RF偏壓能夠施加到配置在基材支撐彳33◎上的基材 110 ° 喷頭310透過懸垂件314輕接背襯板312於其周邊。 喷頭310亦可透過一或多個中央支撐件316耦接背襯 板，以助防止垂弛及/或控制噴頭31〇的筆直度/彎曲度。 氣體源320耦接背襯板312’以提供氣體通過背襯板312 並且通過喷頭310的通道311到基材接收表面332。真 空泵309耦接腔室300以控制製程空間3〇6於期望壓 力。RF功率源322耦接背襯板312及/或喷頭310以提 供RF功率給喷頭；no ’使得電場在喷頭與基材支撐件之 間生成，因此使用在喷頭310與基材支撐件33〇之間配 置的氣體生成電容式耦合的電漿。可使用各RF頻率，諸 如約〇·3 MHz至約100 MHz之間的頻率。相信電容式耦 合電漿直接接觸基材110的處理表面110A (第3圖）具 28 201135962 有勝於不直接暴露所有基材至電漿的設計之優點，因為 腔至300組態能夠在處理期間直接提供具能量及/或離子 化物料給處理表面11〇A的所有部份。 一個實施例中，加熱及/或冷卻元件339可經設定以在 沉積期間提供基材支撐溫度約4〇(rc以下諸如約1〇〇。 C至約40(TC之間。沉積期間配置於基材載具425上的 基材頂表面（配置於基材接收表面322上）與喷頭31〇 之間的間距可為300 mil至約11〇〇㈤之間。例如，沉 積含氮化矽界面次@ 121期間的間距可在沉積含氮化矽 巨量次層期間介於約8〇〇mil至約u〇〇mii之間。 另實施例中，揭路用於形成膜於太陽能電池上的系 統。該系統包括電漿處理腔室（諸如腔室3〇〇)用於在 處理腔室的製程空間306内形成鈍化/ARc層12〇於太陽 能電池基材110上。系統控制器（例如電腦440)與電 聚處理腔室通訊，並且其設以㈣用於形成純化/arc層 120的界面次層121與巨量次層122的處理條件與配方。 電腦440透過控制包含第一製程氣體混合物的氣體流率 與氣體組成、以及由第一氣體混合物生成的電漿之功率 密度’可啟動形成界面次層121的製程條件。在某沉積 時間或其他事件後，電腦44G可使用任何前述㈣變類 型（諸如「斷裂」轉變）執行從第—製程氣體混合物至 第二製程氣體混合物的轉變，以形成巨量次層122 ^電 腦440因此控制處理腔室中的製程條件，以形成每一次 層，使得界面次層與巨量次層具有前述之修改的光學與 29 201135962 鈍化性質。 第4圖疋製程系統4〇〇的一個實施例的頂部概略視 圖’ s亥系統具有複數個處理腔室43丨_437，諸如第3圖的 PECVD腔室300或其他能夠執行與第2圖一併描述的製 程之適合的腔室。製程系統4〇〇包括傳送腔室42〇，其 耦接裝載鎖定腔室410與製程腔室431_437。裝載鎖定腔 室410使基材得以在系統外的周圍環境與傳送腔室42〇 及處理腔室43 1-43 7内的真空環境之間傳送。裝載鎖定 腔室410包括一個或多個可抽空區域，其設以固持一或 多個設以支撐複數個基材11〇的基材載具425。可抽空 區域在將基材輸入系統400期間以泵抽降壓，且在從系 統400輸出基材時通氣。傳送腔室420具有至少一個真 空機器人422配置於其中，該機器人適於在裝載鎖定腔 室410與處理腔室43 1-437之間傳送基材載具425與基 材。第4圖中顯示七個製程腔室；然而，系統4〇〇可具 有任何適當數目的製程腔室。 系統400的一個實施例中，第一處理腔室43 1設以執 行製程201 ’第二處理腔室432設以執行製程202-206, 第三處理腔室433設以執行製程207-208,而第四處理腔 室434設以執行製程209或210於基材上。其他實施例 可使用系統400的處理腔室431-437的各種組合，以執 行製程201 -21 0。尚有系統400的另一實施例，處理腔室 431-437中至少一者設以執行多數該等製程（諸如 201-210)於基材上。 30 201135962 使用此述的Λ施例，含矽氮的鈍化層1 可以實 I上較先則技術製程快速的速率沉積，同時大體上提供 各種鈍化優點而不負面地影響太陽能電池鈍化層之品 質。例如’界面次層含有一些氫與矽自由基，其能與矽 基材的懸空鍵反應以鈍切基材。形成用於界面次層的 沉積的«化學物質之類型具冑比巨量次層@電聚化學 物質更多的氫與矽自由基。 相信當使用其他類型的CVD腔室（諸如遠端電漿沉積 腔室）執行電t沉積日夺，用於製作氛化石夕層的成份以正 確的混合物流入腔室，以產生正確的化學當量比率，但 疋，一旦其沉積於基材上，其不容許對膜有太多控制。 然而，在直接電漿製程系統（諸如pECVD)中，增加的 功率斷裂較弱的鍵結（諸如以⑸鍵），以形成較強的鍵 結’同時膜沉積於基材上。 具有水平噴頭（如第3圖所示）的沉積工具中，喷頭 開口或通道311直接面向基材，混合氣體與電漿直進下 降於基材上。因此，顯示於第3圖的工具具有處理期間 以高沉積速率於飛行中快速改變臈層的能力，其透過改 變進入腔室的氣體混合物而達成。因此，本發明的實施 例一項優點是單一腔室能用於沉積多重氮化矽鈍化層， 同時提供變化氣體流進腔室的能力。 刖述者疋導向本發明的貫施例，本發明其他與進一步 的實施例可在不背離其基本範疇的情況下設計，而本發 明之範疇由隨後的申請專利範圍確定。 31 201135962 【圖式簡單說明】 參考某些繪示於附圖的實施例，可得到之前簡要總括 發月之更详細之描述，如此，可詳細瞭解之前陳述 的本發明的特色。但應注意，附圖只繪示本發明的典型 只施例’因本發明允許其他同等有效的實施例，故不視 為其範圍限制。 第1A圖至第1F圖繪示根據說明於第2圖的製程各階 段的基材一部分的剖面視圖。 第2圖描繪鈍化層形成製程的製程流程圖，該製程是 根據本發明一個實施例於太陽能電池基材上執行。 第3圖是用於執行本發明實施例的平行板pEcvD系 之概略側視圖。 第4圖是具有複數個製程腔室的製程系統之一個 例的頂部概略視圖。 施 為有助瞭解，如可能則使用單一元件符號以指定此、 於該等圖式之單一元件。應考量到一實施例中的元：通 特徵可有利地結合其他實施例而無需附加描述。及 【主要元件符號說明】 103 ρ-η接面區域 1〇5頂表面 1 〇6底表面 100太陽能電池 101基底區域 102 η型摻雜射極區域 32 201135962 110A處理表面 3 12 110 太陽能電池基材 3 14 115 原生氧化物層 3 16 120 鈍化/ARC層 320 121 界面次層 322 122 巨量次層 330 124 黏結材料層 33 1 126 玻璃基材 332 130 、131 針孔 333 200 製程序列 334 201- 212製程 338 300 處理腔室 339 302 壁 400 304 底部 410 306 製程空間 420 308 閥 422 309 真空泵 425 310 喷頭 431 311 通道 440 Qi ' Q2電荷 背槪板 懸垂件 中央支撐件 氣體源 RF功率源 基材支撐件 接地帶 基材接收表面 遮蔽框 心柱 舉升系統 冷卻元件 製程系統 裝載鎖定腔室 傳送腔室 真空機器人 基材載具 -437處理腔室 電腦 33A similar type of plasma is produced in the chamber. However, with the pECVD tool described above, it is believed that the capacitive coupling of a parallel plate type processing chamber can result in a planar plasma power density for forming a passivation film having desired optical and functional properties. Figure 3 is a plasma enhanced chemical gas. A schematic side view of one embodiment of a phase deposition (pECVD) chamber 3' in the chamber, one or more processes illustrated and discussed in conjunction with FIG. 2 can be performed. A similarly configured plasma enhanced chemical vapor deposition chamber is available from Applied Materials, Inc., Sanuc, California. It is contemplated that other deposition chambers, including those purchased from other manufacturers, may be utilized to operate the invention. It is believed that the plasma processing configuration provided in the processing chamber 300 from the top to the middle of the tamper a* is used to perform one or more of the processes described in FIG. 2' with significant advantages over other configurations. advantage. In one embodiment, the 'pEcvD chamber is configured to process a plurality of substrates at a time. In one embodiment, the electro-convex-enhanced chemical vapor deposition (PECVD) chamber 300 is adapted to m-pulse and process a plurality of substrates arranged in a planar array on the substrate carrier 425. (as shown in FIG. 4), which is vertically stacked with the processing of substrate 26 201135962 (eg, such as several batches of substrates stacked in a cassette performed in a vertical furnace system). Processing a plurality of batches of substrates arranged in a planar array allows the substrates in each batch to directly and simultaneously expose the generated gases, radiant heat, and/or process gases. Thus, each substrate in the planar array is similarly processed in the process space of the processing chamber, and this does not rely on a series of energies in the batch of the diffusion type process and/or the conventional settings to transfer to all of the treated substrates. A batch of substrates such as stacked or back-to-back. In one configuration, the 'PECVD chamber 300 is adapted to receive a substrate carrier 425 (Figs. 3 and 4) that is configured to hold a batch of substrates while transporting and processing the substrate. In one embodiment, the substrate carrier 425 has a surface area of greater than about 10,000 cm2, such as greater than about 4 cm2 or greater than about 55,000 cm2, which is configured to support a plurality of substrates disposed thereon during processing. Planar array. In one embodiment, the substrate carrier 425 holds between about 4 and about 49 solar cell substrates having a size of 156 mm X 156 mm X 〇 .3 mm in an upward or downward facing configuration. In one configuration, 'a batch of solar cell substrates are transferred between several processing chambers in a vacuum or inert environment (such as transfer chamber 42 in Figure 4) + on substrate loading to reduce contamination Opportunity and improvement of the Dandan reference to the third 〇 圃 腔 〇 〇 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 大体上 308 The space 3〇6' allows the associated substrate to be transported into and out of the chamber. The substrate support 33 includes: 27 201135962 Receiving surface 3 3 2 disc heart cooking · 3. z ^ ^ 34 ' is coupled to a lifting system (such as lift pin 338) to raise and (I reserve _ soil support 3 3 0. The shadow frame 333 (maintained in chamber 300) can be placed as appropriate Above the substrate carrier 425. The lift pins 338 are movably configured to pass through the substrate support 330 to move the soil carrier 425 to and from the receiving surface 332. The substrate support may also include An embedded heating and/or cooling element 338 is provided to support the substrate at a desired temperature. The substrate support 33 can also include a strap 331 to provide rf grounding around the substrate support 330. In one embodiment, the material member 33G has an RF source (not shown) that is connected to an electrode (not shown). The electrode is embedded in the substrate support 330 so that the RF bias can be applied to the configuration. Substrate 110° on substrate support ◎ 33 The shower head 310 is lightly attached to the periphery of the backing plate 312 through the suspension member 314. The shower head 310 can also be coupled to the backing plate through one or more central supports 316 to Helps prevent sagging and/or control the straightness/bending of the nozzle 31. The gas source 320 is coupled to the backing plate 312' to provide gas through the back The plate 312 passes through the passage 311 of the showerhead 310 to the substrate receiving surface 332. The vacuum pump 309 is coupled to the chamber 300 to control the process space 3〇6 at a desired pressure. The RF power source 322 is coupled to the backing plate 312 and/or the showerhead 310. To provide RF power to the showerhead; no' causes an electric field to be generated between the showerhead and the substrate support, thus creating a capacitively coupled plasma using the gas disposed between the showerhead 310 and the substrate support 33A. Each RF frequency can be used, such as a frequency between about 3 MHz and about 100 MHz. It is believed that the capacitively coupled plasma directly contacts the processing surface 110A of the substrate 110 (Fig. 3) with 28 201135962 better than not directly exposed The advantages of all substrate-to-plasma design, as the cavity-to-300 configuration is capable of providing all of the energy and/or ionized material directly to the processing surface 11A during processing. In one embodiment, heating and/or Or the cooling element 339 can be configured to provide a substrate support temperature of about 4 Torr (such as about 1 Torr. C to about 40 (between TC) and a substrate disposed on the substrate carrier 425 during deposition) during deposition. Top surface (disposed on substrate receiving surface 322 The spacing between the nozzle 31 and the nozzle 31 can be between 300 mil and about 11 〇〇 (f). For example, the spacing during the deposition of the tantalum nitride-containing interface @121 can be deposited during the deposition of the nano-layer containing tantalum nitride. Between about 8 mils and about u〇〇mii. In another embodiment, the system is used to form a film on a solar cell. The system includes a plasma processing chamber (such as a chamber 3). A passivation/ARc layer 12 is formed over the solar cell substrate 110 in a process space 306 of the processing chamber. A system controller (e.g., computer 440) is in communication with the polymerization processing chamber and is provided with (4) processing conditions and recipes for forming the interface sub-layer 121 and the macro-sub-layer 122 of the purification/arc layer 120. The computer 440 can initiate the process conditions for forming the interface sub-layer 121 by controlling the gas flow rate and gas composition comprising the first process gas mixture and the power density of the plasma generated by the first gas mixture. After a deposition time or other event, computer 44G may perform any transition from the first process gas mixture to the second process gas mixture using any of the aforementioned (four) variants (such as a "fracture" transition) to form a massive sub-layer 122 ^ 440 thus controls the process conditions in the processing chamber to form each layer such that the interface sub-layer and the macro-sub-layer have the aforementioned modified optical and 29 201135962 passivation properties. 4 is a top schematic view of an embodiment of a process system 4', having a plurality of processing chambers 43A-437, such as the PECVD chamber 300 of FIG. 3 or other capable of performing with FIG. And describe the suitable chamber for the process. The process system 4A includes a transfer chamber 42A coupled to the load lock chamber 410 and the process chamber 431_437. The load lock chamber 410 allows the substrate to be transferred between the ambient environment outside the system and the transfer chamber 42 and the vacuum environment within the process chambers 43 1-43. The load lock chamber 410 includes one or more evacuatable regions that are configured to hold one or more substrate carriers 425 that are configured to support a plurality of substrates 11A. The evacuatable region is pumped down during the input of the substrate into the system 400 and vented when the substrate is output from the system 400. The transfer chamber 420 has at least one vacuum robot 422 disposed therein that is adapted to transfer the substrate carrier 425 and the substrate between the load lock chamber 410 and the process chambers 43 1-437. Seven process chambers are shown in Figure 4; however, system 4 can have any suitable number of process chambers. In one embodiment of system 400, first processing chamber 43 1 is configured to execute process 201 'second processing chamber 432 is configured to execute processes 202-206, and third processing chamber 433 is configured to perform processes 207-208, The fourth processing chamber 434 is configured to perform a process 209 or 210 on the substrate. Other Embodiments Various combinations of processing chambers 431-437 of system 400 can be used to execute process 201-210. In still another embodiment of system 400, at least one of processing chambers 431-437 is configured to perform a majority of such processes (such as 201-210) on a substrate. 30 201135962 Using the described embodiment, the niobium-containing passivation layer 1 can be deposited at a faster rate than the prior art process, while generally providing various passivation advantages without negatively affecting the quality of the solar cell passivation layer. For example, the interface sublayer contains some hydrogen and ruthenium radicals that react with the dangling bonds of the ruthenium substrate to blunt the substrate. The type of chemical substance that forms the deposition of the interfacial layer has more hydrogen and helium radicals than the giant sublayer @electropolymer. It is believed that when other types of CVD chambers, such as remote plasma deposition chambers, are used to perform the electrical t deposition, the components used to make the fossilized layer flow into the chamber with the correct mixture to produce the correct stoichiometric ratio. However, once it is deposited on the substrate, it does not allow much control over the film. However, in direct plasma processing systems, such as pECVD, the increased power breaks weaker bonds (such as with (5) bonds) to form stronger bonds' while the film is deposited on the substrate. In a deposition tool having a horizontal nozzle (as shown in Figure 3), the nozzle opening or channel 311 is directly facing the substrate, and the mixed gas and plasma are directly lowered onto the substrate. Thus, the tool shown in Figure 3 has the ability to rapidly change the ruthenium layer during flight at high deposition rates during processing by varying the gas mixture entering the chamber. Thus, an advantage of embodiments of the present invention is that a single chamber can be used to deposit multiple tantalum nitride passivation layers while providing the ability to vary the flow of gas into the chamber. The scope of the present invention is defined by the scope of the following claims, and the scope of the present invention is defined by the scope of the appended claims. 31 201135962 [Schematic Description of the Drawings] Referring to certain embodiments illustrated in the accompanying drawings, a more detailed description of the foregoing brief summary is provided, so that the features of the present invention as set forth in the foregoing can be understood in detail. It should be noted, however, that the drawings are only illustrative of the typical embodiments of the invention, and that the claims Figs. 1A to 1F are cross-sectional views showing a part of a substrate according to each stage of the process illustrated in Fig. 2. Figure 2 depicts a process flow diagram of a passivation layer formation process performed on a solar cell substrate in accordance with one embodiment of the present invention. Fig. 3 is a schematic side view of a parallel plate pEcvD system for carrying out an embodiment of the present invention. Figure 4 is a top plan view of an example of a process system having a plurality of process chambers. It is helpful to understand that a single component symbol is used, if possible, to designate a single component in the drawings. It should be considered that the elements in one embodiment can be advantageously combined with other embodiments without additional description. And [Main component symbol description] 103 ρ-η junction region 1〇5 top surface 1 〇6 bottom surface 100 solar cell 101 base region 102 n-type doped emitter region 32 201135962 110A treatment surface 3 12 110 solar cell substrate 3 14 115 Primary oxide layer 3 16 120 passivation/ARC layer 320 121 interface sublayer 322 122 macro sublayer 330 124 bonding material layer 33 1 126 glass substrate 332 130 , 131 pinhole 333 200 program line 334 201- 212 Process 338 300 Processing Chamber 339 302 Wall 400 304 Bottom 410 306 Process Space 420 308 Valve 422 309 Vacuum Pump 425 310 Head 431 311 Channel 440 Qi ' Q2 Charge Backing Plate Suspension Member Central Support Gas Source RF Power Source Substrate Support Piece Grounding Substrate Receiving Surface Screening Frame Heart Column Lifting System Cooling Element Process System Loading Locking Chamber Transfer Chamber Vacuum Robot Substrate Carrier - 437 Processing Chamber Computer 33