201023370 六、發明說明： 【發明所屬之技術領域】 本發明係關於透明導電性/傳導性氧化物薄膜，特別係 關於其應用於薄膜矽光伏（photovoltaic)太陽能電池、太陽 能模組或類似之光伏裝置作爲電極材料之用途。 【先前技術】 太陽能電池也稱作爲光伏電池乃可將電池能諸如光輻 0 射或太陽能輻射直接轉換成電力之半導體。此等半導體係 以其價電子帶與其傳導電子帶間之能帶隙，因而電子通常 無法存在或留在此帶隙爲特徵。但當光被光伏電池的特徵 性材料吸收時，佔據低能態的電子被激勵，跳過帶隙至未 被佔據的較高能態。如此，當半導體之價帶吸收來自於太 陽能輻射之光子之足夠能量時，電子跳過帶隙至較高能傳 導帶。 已被激勵至較高能態之電子留下未被占用的低能位 © 置，稱作爲電洞。此等電洞可於晶格中由一個原子遷移至 —個原子，如同於傳導帶之游離電子，電洞於價帶中作爲 電荷載子來促成晶體的導電性。大部分被吸收於半導體中 的電子產生此種電子-電洞對。此等電子-電洞對產生光 流’及於內建場存在下，產生太陽能電池的光電壓。 由光所產生之電子電洞對最終將複合且轉成熱或光 子’除非被防止避免如此進行。爲了防止此種現象，於半 導體中經由摻雜或界面不同材料產生空間電荷層而形成局 201023370 部電場。空間電荷層隔開用作爲電荷載子之電洞及電子。 一旦被分開，收集的電洞及電子電荷載子產生空間電荷， 結果導致跨接面之電壓’亦即光電壓。若此等分開的電洞 及電荷載子被允許流經外部負載，則將組成光電流。 薄膜太陽能電池通常包括第一電極、一個或多個半導 體薄膜p-i-n或n-i-p接面，及一個第二電極。此等電極係 連續堆疊於一基材上。各個p-i-n接面或薄膜光電變換單元 ^ 包括i型層夾置於P型層及η型層間（p型=正摻雜，η型= ❹ 負摻雜）。1型層爲實質上特性半導體層，佔據薄膜p-i-n接 面厚度之大部分。光電變換主要係發生於此i型層。 先前技術第1圖顯示一種簡單光伏電池40包含一透明 基材41，例如有一層透明導電性氧化物（TCO) 42沉積於其 上之玻璃。此層也稱作爲前接點F/C且係作爲光伏元件之 第一電極。下一層43係作爲活性光伏層且包含形成p-i-n 接面之三層次層。該層43包含經氫化之微晶矽、奈米結晶 ❹ 矽或非晶矽或其組合物。與TCO前接點42相鄰之次層44 爲正摻雜，相鄰次層45爲特有層（intrinsic layer)，及最終 次層46爲負摻雜。於另一個具體實施例中，所描述之層順 序P-i-n可顛倒成爲n-i-p，則層44識別爲η層，層45再 度爲特有層 '層46爲ρ層。 最後’電池包括一後接點層47 (也稱作爲後接點B/C) 其可由例如氧化鋅、氧化錫或ΙΤΟ製成，及一反射層48。 另外可實現金屬後接點，其可組合後反射器48與後接點47 201023370 之物理性質。用於舉例說明，箭頭指示入射光。 晚近由於TCO層對PV電池之總功效有重大貢獻，故 努力目標集中於採用TCO層之PV薄膜矽太陽能電池。TC〇 層用作爲PV太陽能電池系統中之前及後接點或前及後電 極之一種極爲有展望的材料爲氧化鋅ZnO。適合用於PC應 用之氧化鋅層例如爲歐瑞康（Oerlikon)太陽能TCO 1 200量 產系統，本系統爲水平線上沉積系統。可用於前接點及後 接點的量產。該系統有不同的模組負責基材的運送、預熱 響 及處理。於標準配置中，建立一裝載站、一裝載閘、四個 製程模組（4PM)、一卸載閘、一卸載站及一回送軌。使用此 種標準4PM系統，可獲得產率高於95%每年大於1 5 0,000 片基材之產出量。系統可裝配六個製程模組，如此系統之 產出量更增高》 裝載閘爲第一模組，影響沉積程序因而影響層性質。 裝載閘首先將製程室由大氣壓減壓至製程壓力。第二項任 〇 務係於減壓期間於可調整之時間以內預熱基材。藉由可獨 立控制之輻射加熱器集合而實現加熱。達到期望之基材溫 度後，熱基材移動至第一製程模組。於製程模組PM中， 進行低壓化學氣相沉積LPCVD程序。各個製程模組結合有 不同的控制式加熱區之熱板及專用的製程氣體配送系統來 確保穩定的氣體供應。玻璃鋪在熱板上來於沉積期間控制 基材溫度。每個製程模組可以3.5奈米/秒之沉積速率沉積 氧化鋅。全部相關製程化學品係使用該等化學品於其中氣 201023370 化之外部氣體箱而呈氣相遞送入製程模組內。適當泵送係 由對稱性幫浦歧管提供。 透明導電性氧化物（TCO)爲薄膜太陽能電池的主要部 分。TCO層例如係作爲透明前窗且需具有高導電性及高透 光性。進一步，TCO層必須具有於TCO電池界面散射入射 光的能力。此種散射促成於（光）活性層中之光徑延長，如 此增加吸光效率及模組效率。除了對TCO電池粗度之良好 控制之外，界面允許於電池與前接點界面與後接點界面間 ❹ 之光獲得多次反射，如此進一步增加通過吸收劑的光徑。 此種光捕捉機制由於光吸收係數不良，故用於微晶矽薄膜 太陽.能電池特別重要。 藉LPCVD技術沉積之摻硼氧化鋅層顯示作爲薄膜太陽 能技術之良好候選者。其原料組成分廣泛易得，成本低且 無害。此外，LPCVD技術極爲適合用於大規模裝置製造。 此外，LPCVD氧化鋅層用於「如所生長的（as-grown)」太陽 Q 能電池製造，而無需額外處理例如後處理或蝕刻。 因此’氧化鋅廣用於薄膜太陽能電池作爲後接點及前 接點。已經摻雜之氧化鋅薄膜具有高導電性及高透光性。 用於薄膜太陽能電池之TC0層之電性質及光性質顯著影響 模組效率。因此本發明之目的係對適合用於光伏薄膜電池 之TC0層配置之較佳具體實施例界定製程窗。 【發明內容】 發明槪要 201023370 一種適合用於光伏薄膜電池之TCO層配置之製程窗’ 採用於一沉積系統之一製程室內，藉低壓化學氣相沉積 (LPCVD)於一基材上由至少二乙基鋅、水、及乙硼烷製造一 透明導電性氧化鋅層，該方法包含下列步驟： -以0.87至1.3之氣體流量比提供二乙基鋅及水， -維持乙硼烷/二乙基鋅之氣體流量比爲〇· 〇5至0.4’ -如此於基材上沉積摻硼導電性氧化鋅層， 具有於600奈米之漫射透射比對總透射比之比値測量 ❿ 得氧化鋅層之濁度爲10%至25%。 藉前述低壓化學氣相沉積（LPCVD)製程室所沉積之個 別透明導電性氧化鋅層將具有於400奈米至800奈米範圍 之特性層透射比大於93%及400奈米至1100奈米之特性層 透射比大於92%。於1.4平方米大小之整個基材上之片電阻 將小於10歐姆/平方（Ω/口）。 薄膜光伏電池可裝配有呈堆叠配置之此種氧化鋅層作 〇 爲第一電極及/或第二電極如下：基材-第一基材個或多 個半導體薄膜p-i-n接面或n-i-p接面-第二電極。 【實施方式】 TC0層之性質係由多項參數諸如基材溫度、製程氣體 混合物及壓力決定。第2圖顯示典型氧化鋅層表面之掃描 電子顯微鏡（SEM)相片。特定表面型態乃獲得最佳光捕捉之 關鍵因素中之一者。氧化鋅光散射效率係使用濁度定量， 濁度容易於600奈米波長於10%至2 5%間微調。 201023370 TCO 1200量產系統於1.4平方米大小之全部基材上製 造具有厚度均勻度低於20%及片電阻小於1〇 Ω/□之氧化鋅 層。此外，如於第2圖之片電阻可知，層品質如基材數目 之函數維持恆定。 ❹ 用作爲前接點，該層透射比扮演重要角色。層透射比 於400奈米至800奈米之範圍係高於93%，而於400奈米 至1100奈米之範圍係高於92 %。第3圖顯示於不同玻璃基 材上於不同TC0層間之透射比測量値之比較。顯示使用 TC0 1200製造系統所沉積的氧化鋅於極爲寬廣之波長範圍 具有最高透射比。 實驗結果 TC ◦性質可容易且精準地微調乃太陽能電池製造商極 爲期望的特性，原因在於其允許修改根據太陽能設計之層 性質。 藉由增加TC0層厚度，可改良片電阻，但透明度減低。 〇 理想的折衷可確保最佳模組效能明確依據吸收劑類型決 定。厚度可藉沉積時間控制，厚度也可藉作用參數諸如沉 積期間之溫度、壓力及製程氣體流量決定（參考第16圖頂 圖）。總而言之，厚度可被改變來確保獲得良好層均勻度。 第4圖顯示1 1 00x 1 300平方毫米V5玻璃之厚度及片電阻測 量値。已經對玻璃上均勻分布的143點進行厚度及片電阻 測量。 另一項相關參數爲摻雜濃度。氧化鋅層摻雜硼（硼來源 201023370 爲乙硼烷B2H6，市面上以於氫之2%稀釋液獲得）俾便提高 層導電性。摻雜濃度容易藉該製程之輸入而改變但無需作 硬體變化。此點係與物理氣相沉積（PVD)相反，於PVD中， 摻雜濃度主要係由濺鍍標靶之化學組成決定。但摻雜對濁 度及對透射度造成負面影響。藉霍爾（Hall)移動性測量値， 可估算保證獲得最佳折衷的摻雜濃度。於第5圖中，霍爾 移動性及濁度（濁度係定義爲漫射透射比對總透射比之比201023370 VI. Description of the Invention: [Technical Field] The present invention relates to a transparent conductive/conductive oxide film, particularly for use in a thin film photovoltaic solar cell, a solar module or the like. Used as an electrode material. [Prior Art] A solar cell, also referred to as a photovoltaic cell, is a semiconductor that can directly convert a battery such as light radiation or solar radiation into electricity. These semiconductors are characterized by an energy band gap between their valence band and their conduction electron band, and thus electrons are usually not present or remain in this band gap. However, when light is absorbed by the characteristic material of the photovoltaic cell, the electrons occupying the low energy state are excited, skipping the band gap to the unoccupied higher energy state. Thus, when the valence band of the semiconductor absorbs sufficient energy from the photons radiated by the solar energy, the electrons skip the band gap to the higher energy conduction band. The electrons that have been excited to the higher energy state leave the unoccupied low energy level ©, called the hole. These holes can migrate from one atom to one atom in the crystal lattice, like the free electrons in the conduction band, and the holes act as charge carriers in the valence band to promote the conductivity of the crystal. Most of the electrons absorbed in the semiconductor produce such electron-hole pairs. These electron-hole pairs generate optical currents and, in the presence of built-in fields, produce the photovoltage of the solar cells. The pair of electron holes generated by the light will eventually recombine and turn into heat or photons' unless it is prevented from doing so. In order to prevent this phenomenon, a space charge layer is formed in a semiconductor via doping or a different interface material to form an electric field of 201023370. The space charge layer separates the holes and electrons used as charge carriers. Once separated, the collected holes and electron charge carriers generate space charge, resulting in a voltage across the junction, which is the photovoltage. If such separate holes and charge carriers are allowed to flow through an external load, they will constitute a photocurrent. Thin film solar cells typically include a first electrode, one or more semiconductor thin films p-i-n or n-i-p junctions, and a second electrode. These electrodes are continuously stacked on a substrate. Each p-i-n junction or thin film photoelectric conversion unit ^ includes an i-type layer sandwiched between a p-type layer and an n-type layer (p-type = positive doping, n-type = ❹ negative doping). The type 1 layer is a substantially characteristic semiconductor layer occupying most of the thickness of the p-i-n junction of the film. The photoelectric conversion mainly occurs in this i-type layer. Prior art Figure 1 shows a simple photovoltaic cell 40 comprising a transparent substrate 41, such as a glass having a layer of transparent conductive oxide (TCO) 42 deposited thereon. This layer is also referred to as the front contact F/C and serves as the first electrode of the photovoltaic element. The next layer 43 serves as the active photovoltaic layer and comprises a three-layer layer forming a p-i-n junction. This layer 43 comprises hydrogenated microcrystalline germanium, nanocrystalline germanium or amorphous germanium or a combination thereof. The sub-layer 44 adjacent to the TCO front contact 42 is positively doped, the adjacent sub-layer 45 is an intrinsic layer, and the final sub-layer 46 is negatively doped. In another embodiment, the described layer order P-i-n may be reversed to n-i-p, layer 44 is identified as an η layer, and layer 45 is again a unique layer 'layer 46 is a ρ layer. Finally, the battery includes a back contact layer 47 (also referred to as a back contact B/C) which may be made of, for example, zinc oxide, tin oxide or antimony, and a reflective layer 48. In addition, a metal back contact can be realized which combines the physical properties of the back reflector 48 and the back contact 47 201023370. For purposes of illustration, the arrows indicate incident light. Due to the significant contribution of the TCO layer to the overall efficacy of PV cells, efforts have focused on PV films and solar cells using TCO layers. The TC layer used as a highly promising material for the front and rear contacts or the front and rear electrodes in a PV solar cell system is zinc oxide ZnO. The zinc oxide layer suitable for PC applications is, for example, the Oerlikon solar TCO 1 200 mass production system, which is a horizontal line deposition system. It can be used for mass production of front and rear contacts. The system has different modules responsible for substrate transport, preheating and handling. In the standard configuration, a loading station, a loading gate, four process modules (4PM), an unloading gate, an unloading station, and a return rail are established. Using this standard 4PM system, a yield of greater than 95% of substrate per year greater than 1,500,000 substrates can be obtained. The system can be equipped with six process modules, so that the output of the system is increased. The load gate is the first module, which affects the deposition process and thus affects the layer properties. The loading gate first decompresses the process chamber from atmospheric pressure to process pressure. The second task is to preheat the substrate within an adjustable time period during decompression. Heating is achieved by a collection of radiant heaters that can be independently controlled. After the desired substrate temperature is reached, the thermal substrate moves to the first process module. In the process module PM, a low pressure chemical vapor deposition LPCVD process is performed. Each process module incorporates a hot plate with different controlled heating zones and a dedicated process gas distribution system to ensure a stable gas supply. The glass is spread on a hot plate to control the substrate temperature during deposition. Each process module can deposit zinc oxide at a deposition rate of 3.5 nm/sec. All relevant process chemicals are delivered to the process module in a gaseous phase using these chemicals in the external gas box of the gas 201023370. The appropriate pumping system is provided by the symmetrical pump manifold. Transparent conductive oxide (TCO) is a major part of thin film solar cells. The TCO layer is, for example, a transparent front window and is required to have high conductivity and high light transmittance. Further, the TCO layer must have the ability to scatter incident light at the TCO cell interface. Such scattering contributes to the extension of the optical path in the (light) active layer, thereby increasing the light absorption efficiency and module efficiency. In addition to good control of the TCO cell thickness, the interface allows for multiple reflections of light between the cell and the front contact interface and the back contact interface, thus further increasing the optical path through the absorber. Such a light trapping mechanism is particularly important for use in a microcrystalline germanium film because of its poor light absorption coefficient. The boron-doped zinc oxide layer deposited by LPCVD technology is a good candidate for thin film solar technology. Its raw material composition is widely available, low cost and harmless. In addition, LPCVD technology is extremely suitable for large scale device manufacturing. In addition, the LPCVD zinc oxide layer is used in "as-grown" solar Q-energy cells without additional processing such as post-processing or etching. Therefore, zinc oxide is widely used as a thin film solar cell as a back contact and a front contact point. The already doped zinc oxide film has high conductivity and high light transmittance. The electrical and optical properties of the TC0 layer used in thin film solar cells significantly affect module efficiency. It is therefore an object of the present invention to customize a process window for a preferred embodiment of a TC0 layer configuration suitable for use in a photovoltaic thin film battery. SUMMARY OF THE INVENTION Summary of the Invention 201023370 A process window suitable for use in a TCO layer configuration of a photovoltaic thin film battery is employed in a process chamber of one deposition system by low pressure chemical vapor deposition (LPCVD) on a substrate by at least two A transparent conductive zinc oxide layer is prepared from ethyl zinc, water, and diborane. The method comprises the following steps: - providing diethyl zinc and water at a gas flow ratio of 0.87 to 1.3, - maintaining diborane / diethyl The gas flow ratio of the zinc group is 〇·〇5 to 0.4' - thus depositing a boron-doped conductive zinc oxide layer on the substrate, having a ratio of diffuse transmittance to total transmittance at 600 nm, ❿ oxidation The turbidity of the zinc layer is from 10% to 25%. The individual transparent conductive zinc oxide layer deposited by the low pressure chemical vapor deposition (LPCVD) process chamber will have a characteristic layer transmittance of greater than 93% and 400 nm to 1100 nm in the range of 400 nm to 800 nm. The characteristic layer transmittance is greater than 92%. The sheet resistance on the entire substrate of 1.4 square meters will be less than 10 ohms/square (Ω/□). The thin film photovoltaic cell can be equipped with such a zinc oxide layer in a stacked configuration as the first electrode and/or the second electrode as follows: substrate - first substrate or a plurality of semiconductor film pin junctions or nip junctions - Second electrode. [Embodiment] The properties of the TC0 layer are determined by a number of parameters such as substrate temperature, process gas mixture, and pressure. Figure 2 shows a scanning electron microscope (SEM) photograph of the surface of a typical zinc oxide layer. A particular surface type is one of the key factors in achieving optimal light capture. Zinc oxide light scattering efficiency is quantified using turbidity, and turbidity is easily fine-tuned between 10% and 25% at a wavelength of 600 nm. The 201023370 TCO 1200 mass production system produces a zinc oxide layer with a thickness uniformity of less than 20% and a sheet resistance of less than 1 〇 Ω/□ on all substrates of 1.4 m2. Further, as can be seen from the sheet resistance of Fig. 2, the layer quality as a function of the number of substrates is maintained constant. ❹ Used as a front contact, the transmittance of this layer plays an important role. The layer transmittance is higher than 93% in the range of 400 nm to 800 nm, and higher than 92% in the range of 400 nm to 1100 nm. Figure 3 shows a comparison of the transmittance measurements between different TC0 layers on different glass substrates. Zinc oxide deposited using the TC0 1200 manufacturing system is shown to have the highest transmittance over a very wide wavelength range. EXPERIMENTAL RESULTS TC ◦ properties can be easily and accurately fine-tuned as a highly desirable feature for solar cell manufacturers because they allow modification of the properties of layers based on solar design. By increasing the thickness of the TC0 layer, the sheet resistance can be improved, but the transparency is reduced.理想 The ideal compromise ensures that the optimum module performance is clearly determined by the type of absorbent. The thickness can be controlled by the deposition time, and the thickness can also be determined by the action parameters such as the temperature, pressure and process gas flow during the deposition (refer to the top view of Figure 16). In summary, the thickness can be varied to ensure good layer uniformity. Figure 4 shows the thickness and sheet resistance measurement of 1 1 00 x 1 300 mm 2 V5 glass. Thickness and sheet resistance measurements have been made on 143 points evenly distributed over the glass. Another related parameter is the doping concentration. The zinc oxide layer is doped with boron (boron source 201023370 is diborane B2H6, commercially available as a 2% dilution of hydrogen) to improve layer conductivity. The doping concentration is easily changed by the input of the process but does not require a hard change. This point is in contrast to physical vapor deposition (PVD), where the doping concentration is primarily determined by the chemical composition of the sputter target. However, doping has a negative impact on turbidity and on transmission. With the Hall mobility measurement, it is possible to estimate the doping concentration that guarantees the best compromise. In Figure 5, Hall mobility and turbidity (turbidity is defined as the ratio of diffuse transmittance to total transmittance)
値）顯示爲乙硼烷/DEZ比之函數。 LPCVD ZnO沉積方法之寬廣製程窗允許對製程參數安 定性較少限制而不會對層性質造成影響。舉例言之，藉壓 力變化（±15%)或藉氣體流量（絕對及相對）之些微改變，讓 TCO層性質維持相同。此種LPCVD方法之安定性允許儘管 於製程上有微小變化，工具擁有者仍然可獲得已塗覆玻璃 之穩定輸出。如此兩種主要前驅物水及二乙基鋅（DEZ)之流 量比可於寬廣範圍改變而氧化鋅層並無重大變化（參考第 〇 16 圖）》 除了透明度及導電度之外，影響PV模組效能之另外一 項TCO特性爲於TCO-電池界面之光散射。矽之吸收係數 相當低，此點對微晶矽特別爲真。因此期望增加通過吸收 劑的光徑。控制光散射的主要參數爲濁度。結果顯示根據 本揭示成長之LPCVD ZnO層，相較於大部分市售氧化錫 TC0玻璃，具有優異濁度（參考第6圖頂圖）。 藉由對TCO表面結構之更準確控制，可於具有前接點 201023370 及後接點之吸收劑界面感應引發多次總內反射之光之顯著 部分。如此有助於顯著增加吸收。爲了測量及控制氧化鋅 表面之此等關鍵性質，發展出角解析散射（ARS)設備。第6 圖底圖指示如所生長之LPCVD ZnO層具有極佳光散射性 質。値) is shown as a function of the diborane/DEZ ratio. The wide process window of the LPCVD ZnO deposition method allows for less restriction on process parameter stability without affecting layer properties. For example, the TCO layer properties remain the same by a change in pressure (±15%) or by a slight change in gas flow (absolute and relative). The stability of this LPCVD method allows the tool owner to obtain a stable output of the coated glass despite minor changes in the process. The flow ratio of the two main precursors of water and diethylzinc (DEZ) can be varied over a wide range without significant changes in the zinc oxide layer (see Figure 16). In addition to transparency and conductivity, the PV mode is affected. Another TCO characteristic of group performance is light scattering at the TCO-battery interface. The absorption coefficient of ruthenium is quite low, which is especially true for microcrystalline enamel. It is therefore desirable to increase the optical path through the absorbent. The main parameter controlling light scattering is turbidity. The results show that the LPCVD ZnO layer grown according to the present disclosure has excellent haze compared to most commercially available tin oxide TC0 glasses (refer to the top view of Fig. 6). By more precise control of the TCO surface structure, a significant portion of the light that initiates multiple total internal reflections can be induced at the absorber interface with the front contact 201023370 and the back contact. This helps to significantly increase absorption. In order to measure and control these key properties of zinc oxide surfaces, angular resolution scattering (ARS) devices have been developed. Figure 6 is a bottom view showing that the grown LPCVD ZnO layer has excellent light scattering properties.
實驗結果II 如所指示，對薄膜太陽能電池裝置，光捕捉爲關鍵特 ^ 性俾便提高電池效率與顯著減低光伏電池成本》因此前 〇 TC0性質係於高效能實驗室TCO (Asahi-U)與場內製造的 LPCVD ZnO間做比較。對兩種TC0已經對單一接面1平方 厘米a-Si:H電池之性質呈i層厚度之函數做詳細硏究。於 A sahi-U上獲得最大安定效率爲8.6%。特別，兩種接受硏 究之基材之差異爲裝置內he値，此項觀察於TCO測量得 之光參數間良好吻合。進一步製程發展硏究中，LPCVD ZnO 之安定效率可改良至9.1 %之顯著數値。使用高效能a-Si:Η φ 方法於LPCVD-ZnO，製造若干迷你模組（10x10平方厘米） 及經過光浸漬。於伊斯普拉（Ispra)之〗RC之ESTI實驗室證 實最佳LPCVD-ZnO迷你模組之孔隙效率爲8.32%。基於對 a-Si:H單一接面裝置所得知識，實現微型銜接電池及迷你 模組。電池顯示高初始效率（大於11.8%)，及獲得具有安定 效率接近10%之迷你模組。 a-Si:H吸收劑層厚度之縮小對製造產出量及裝置安定 性（光浸漬後）二者皆有利。爲了達成此項目的，使用粗糙 -10- 201023370 TCOs用於提升裝置內部之光捕捉。有效光捕捉結果導致光 徑提升數倍，允許吸收劑層厚度縮小。此外，於微型銜接 裝置中，由於pc-Si:H之光吸收程度比a-Si:H更低，故TC0 之光捕捉性質增強甚至更爲重要。此等光捕捉性質於可見 光範圍且特別於近紅外光範圍特別令人感興趣。最後，優 異之光捕捉能力、高透射性及高導電性爲用於薄膜矽太陽 能電池之TCOs之重要面相。Asahi-U二氧化錫具有良好光 性質及電性質，結果導致a-Si:H p-i-n電池之優異效能。 藉低壓化學氣相沉積（LPCVD)製造之氧化鋅（ZnO)被視 爲具有用於製造優異的薄膜矽太陽能電池之潛力（由於其 具有傑出的光捕捉性質）。後文揭示使用p-i-n a-Si:H電池 及迷你模組於Asahi-U及LPCVD-ZnO前接點上達成最佳效 能。後文示例說明LPCVD-ZnO獲得優異初始電池效能及穩 定a-Si:H電池效能的潛力。最後，比較延伸至微型銜接電 池及迷你模組。 φ 所呈現之P-i-n a-Si:H太陽能電池係沉積於歐瑞康太陽 能公司（Oerlikon Solar)製造之R&D單室KAI-M系統（52x41 平方厘米基材尺寸）。爲了提升沉積效率，PECVD方法自適 應於40.6 8 MHz之激勵頻率。KAI反應器之清潔係基於原 位電漿方法’且係於各電池操作之後執行。特性裝置品質 a-Si:H之沉積速率爲3.35埃/秒，pc-Si:Hi層（於微型銜接） 係於高達5埃/秒之速率沉積。供比較目的，p-i_n a-Si:H電 池之二系列i層厚度分別沉積於LPCVD-ZnO、Asahi-U二氧 -11- 201023370 化矽上。各基材類型之最佳p-i-n a-Si:H電池係用作爲微型 銜接製造的頂電池；最後也比較其效能。場內製造之 LPCVD-ZnO也組合白反射器（WR)施用於全部電池及迷你模 組上作爲後接點。氧化鋅層之沉積參數經最佳化來於該裝 置之前接點及後接點二者上獲得有效光捕捉、光透明度及 導電性。 爲了小心決定特徵，如瑞士聯邦度量衡局（MET AS )核 處 准，全部測試電極皆以雷射製作圖案至1.0029平方厘米之 © 經明確界定的面積來結構化。藉施加雷射雕刻用於單晶系 列連結而實現10x10平方厘米尺寸的迷你模組。迷你模組 之孔隙面積也經過METAS核准。a-Si:H電池之I (V)特性係 於 25°C 於 AM 1 .5 照明（瓦康（Wacom) WXS-155S-L2 雙源模 擬器）下測量。光浸潤實驗係以1000小時時間於50°C於接 近AM 1.5之照明下進行。經過最佳光浸潤之a-Si:H迷你模 組送至義大利伊斯普拉之】RC之ESTI實驗室進行獨立決定 φ 特徵。 場內氧化鋅係藉LPCVD方法於低於200t之溫度製 造。前TCO層之片電阻係低於10 Ω/□。多晶薄膜係由大型 晶粒組成，其末端係以大型金字塔出現於成長中的表面（參 考第7b圖）。如所成長之粗糙面質地提供有效散射光漫射 入矽裝置，如第7圖所示。 LPCVD-ZnO具有優異的光學性質可供應用於薄膜矽太 陽能電池中作爲前TC0。該層於可見光範圍之總透射率係 -12- 201023370 遠高於80%及Asahi-U。此外，由第8 a圖及第8 b圖之漫 射透射率資料評估，其光散射性質甚至比Asahi-U之光散 射能力更佳。 p-i-η電池效能之整體表現如下··於p層對兩種TC 0s (LPCVD-ZnO及Asahi-U)最佳化之後，對V〇c値觀察得類似 的數値及趨勢。於初始態，V。。値大致上與i層厚度無關， 而於經光浸漬態，對兩個TCOs而言V。。略爲隨著厚度的增 ©加而減低。如前文已知，ZnO上之初始FF値比較Asahi-U 爲略低。但於光浸漬後，於兩個TCOs上的FF可相媲美。 與光學資料相吻合，對給定之i層厚度而言，LPCVD-ZnO 上之；he値遠比Asahi-U上的Jsc値更高。例如於300奈米 發現LPCVD-ZnO上之Jsc値（於安定化狀態）比Asahi-U更高 1毫安培/平方厘米。 最後獲得結論，於Asahi-U上，裝置之i層的最佳厚度 約爲200奈米或以下。另一方面，於LPCVD-ZnO上，吸收 〇 劑層須具有240奈米至300奈米之範圍之厚度來獲得最高 穩定後之效率。於本硏究中，於LPCVD ZnO上，最高效率 爲8.8%’而Asahi-U上之最高效率爲8.6%。根據此等新發 現’發明人驗證藉發明人之場內氧化鋅成功地置換Asahi_U TC0可獲得具有最高安定化效率之單—接面a_Si:H裝置。 於第11圖中’顯示於第3圖及第4圖之硏究獲得電池 之相對效率降級。根據此等結果，對兩個TCOs以類似方 式’裝置效率之相對降級取決於i層厚度，即使初始效率 -13- 201023370 及安定化之效率之絕對値不同亦如此。 沉積方案 效率範圍（已安定化）[%] 1 8.9-9.1 2 8.8-8.9 3 9.0-9.1 4 8.9-9.0 根據先前對沉積於LPCVD-ZnO上之P-i-n a-Si:H裝置 之發現，最佳i層厚度係於240奈米至300奈米間。基於 該項結果，裝置之製程進一步改良，找到四項沉積方案（參 考下表；方案1至4)可將安定化效率甚至提高至9%。該製 程強勁之驗證爲使用不同製程窗可達成約9 %之穩定化效 率 〇 _ 表：對四種不同最佳化p-i-n a-Si:H沉積方案（各方案 測量4至13個個別1平方厘米電池）之於LPCVD ZnO前TC0 上所達成之穩定效率。 〇 若干製程微調之嘗試也經硏究來改良於Asahi-U上之 效率但未能進一步進展。但於LPCVD-ZnO上，發明人顯著 改良電氣效能高達上表及第12圖所示程度》至目前爲止於 不同前TCOs上所得最佳a-Si:H電池之I (V)曲線顯示於第 12圖。EXPERIMENTAL RESULTS II As indicated, for thin-film solar cell devices, light capture is a key feature that improves cell efficiency and significantly reduces photovoltaic cell cost. Therefore, the former TC0 property is in the high-performance laboratory TCO (Asahi-U) and Comparisons were made between LPCVD ZnO fabricated in the field. A detailed study of the properties of the i-layer thickness for the properties of a single junction 1 square centimeter a-Si:H cell has been investigated for both TC0. The maximum stability efficiency achieved on A sahi-U was 8.6%. In particular, the difference between the two substrates subjected to the study was the helium in the device, and this observation was in good agreement with the light parameters measured by the TCO. In further process development, the stability of LPCVD ZnO can be improved to a significant number of 9.1%. A number of mini-modules (10 x 10 cm2) were fabricated using LP-ZnO as a high-performance a-Si: φ φ method and subjected to light impregnation. The best efficiency of the LPCVD-ZnO mini-module was 8.32% at the ESTI laboratory of RC in Ispra. Based on the knowledge gained from the a-Si:H single junction device, the micro-connected battery and mini-module are realized. The battery shows high initial efficiency (greater than 11.8%) and a mini module with a stability efficiency close to 10%. The reduction in the thickness of the a-Si:H absorber layer is advantageous for both the manufacturing throughput and the device stability (after light immersion). In order to achieve this project, the use of rough -10- 201023370 TCOs is used to enhance the light trapping inside the unit. The effective light capture results in a multiple of the optical path, allowing the thickness of the absorber layer to be reduced. In addition, in the micro-coupling device, since the light absorption degree of pc-Si:H is lower than that of a-Si:H, the light capturing property of TC0 is enhanced and even more important. Such light trapping properties are of particular interest in the visible light range and particularly in the near infrared light range. Finally, superior light capture, high transmission and high conductivity are important aspects of TCOs used in thin-film solar cells. Asahi-U tin dioxide has good optical and electrical properties, resulting in excellent performance of the a-Si:H p-i-n battery. Zinc oxide (ZnO) manufactured by low pressure chemical vapor deposition (LPCVD) is considered to have the potential to produce an excellent thin film tantalum solar cell (due to its excellent light capturing properties). The use of p-i-n a-Si:H batteries and mini-modules to achieve the best results at the Asahi-U and LPCVD-ZnO front contacts is disclosed later. The following examples illustrate the potential of LPCVD-ZnO to achieve excellent initial battery performance and to stabilize a-Si:H battery performance. Finally, the comparison extends to micro-connected batteries and mini-modules. The P-i-n a-Si:H solar cell presented by φ was deposited on the R&D single chamber KAI-M system manufactured by Oerlikon Solar (52 x 41 cm 2 substrate size). In order to improve deposition efficiency, the PECVD method is adaptive to the excitation frequency of 40.6 8 MHz. The cleaning of the KAI reactor is based on the in situ plasma method' and is performed after each battery operation. Characteristic Device Quality The deposition rate of a-Si:H was 3.35 Å/sec, and the pc-Si:Hi layer (in micro-coupling) was deposited at a rate of up to 5 Å/sec. For comparison purposes, the thickness of the second series of i-layers of the p-i_n a-Si:H battery was deposited on LPCVD-ZnO and Asahi-U dioxane-11-201023370. The best p-i-n a-Si:H battery type for each substrate type was used as a top cell for micro-joining; the final performance was also compared. In-field manufactured LPCVD-ZnO is also combined with a white reflector (WR) applied to all cells and mini-modules as a back contact. The deposition parameters of the zinc oxide layer are optimized to achieve effective light capture, light transparency, and electrical conductivity both before and after the device. In order to carefully determine the characteristics, such as the Swiss Federal Bureau of Weights and Measures (MET AS), all test electrodes are laser-patterned to 1.0029 cm 2 © structured area. A mini module of 10 x 10 cm2 size is realized by applying laser engraving for single crystal series connection. The pore area of the mini module is also approved by METAS. The I (V) characteristics of the a-Si:H battery are measured at 25 ° C under AM 1.5 lighting (Wacom WXS-155S-L2 dual source simulator). The light infiltration experiment was carried out at 1000 ° C for 50 hours under illumination of AM 1.5. After the best light infiltration, the a-Si:H mini module was sent to the ESTI laboratory of RC, Italy, to determine the φ characteristics independently. In-situ zinc oxide is produced by LPCVD at temperatures below 200t. The sheet resistance of the front TCO layer is less than 10 Ω/□. The polycrystalline film consists of large grains with a large pyramid appearing on the growing surface (refer to Figure 7b). The rough surface texture as grown provides effective diffused light diffusing into the device, as shown in Figure 7. LPCVD-ZnO has excellent optical properties and can be used as a front TC0 in a thin film solar cell. The total transmittance of this layer in the visible range is well above -12-201023370 and is higher than 80% and Asahi-U. In addition, from the diffuse transmittance data of Figures 8a and 8b, the light scattering properties are even better than those of Asahi-U. The overall performance of the p-i-η battery performance is as follows: After the p layer is optimized for two TC 0s (LPCVD-ZnO and Asahi-U), similar numbers and trends are observed for V〇c値. In the initial state, V. .値 is generally independent of the thickness of the i-layer, while in the light-impregnated state, V for two TCOs. . Slightly decrease with increasing thickness © ©. As previously known, the initial FF ZnO on ZnO is slightly lower than Asahi-U. However, after immersion in light, the FF on the two TCOs is comparable. Consistent with the optical data, for a given thickness of the i layer, on the LPCVD-ZnO; he値 is much higher than the Jsc値 on the Asahi-U. For example, at 300 nm, Jsc値 (in the stabilized state) on LPCVD-ZnO was found to be 1 mA/cm 2 higher than Asahi-U. Finally, it was concluded that on Asahi-U, the optimal thickness of the i-layer of the device is about 200 nm or less. On the other hand, on LPCVD-ZnO, the absorbing agent layer must have a thickness in the range of 240 nm to 300 nm to obtain the most stable efficiency. In this study, the highest efficiency was 8.8% on LPCVD ZnO and 8.6% on Asahi-U. According to these new discoveries, the inventors have verified that the single-junction a_Si:H device with the highest stabilization efficiency can be obtained by successfully replacing the Asahi_U TC0 with zinc oxide in the inventor's field. The relative efficiency degradation of the battery was obtained in the Fig. 11 and the graphs shown in Figs. 3 and 4. Based on these results, the relative degradation of the efficiency of the two TCOs in a similar manner depends on the thickness of the i-layer, even if the initial efficiency is -13-201023370 and the absolute efficiency of the stabilization is different. Deposition scheme efficiency range (already stabilized) [%] 1 8.9-9.1 2 8.8-8.9 3 9.0-9.1 4 8.9-9.0 According to the previous discovery of Pin a-Si:H device deposited on LPCVD-ZnO, the best The thickness of the i layer is between 240 nm and 300 nm. Based on this result, the process of the device was further improved. Finding four deposition schemes (refer to the table below; schemes 1 to 4) can even increase the stabilization efficiency to 9%. The robust verification of the process allows for a stabilization efficiency of approximately 9% using different process windows 表 Table: Four different optimized pin a-Si:H deposition schemes (4 to 13 individual 1 cm 2 for each scenario) Battery) The stable efficiency achieved on LPCVD ZnO pre-TC0.尝试 Attempts to fine-tune several processes have also been improved to improve the efficiency on Asahi-U but have not progressed further. However, on LPCVD-ZnO, the inventors significantly improved the electrical performance up to the extent shown in the above table and Figure 12. The I (V) curve of the best a-Si:H battery obtained on different pre-TCOs has been shown in the first 12 pictures.
使用200奈米之i層厚度（Jsc = 14.8毫安培/平方厘米， FF = 66.6%，及V〇c = 873毫伏特，參考第12圖）可獲得8.6% 安定化之電池效率（於Asahi-U)上。於LPCVD-ZnO前TCO -14-, 201023370 之情況下，使用240奈米之i層厚度（Jsc=15.6毫安培/平方 厘米，FF = 66.7%，及V〇c= 876毫伏特’參考第1 2圖）可獲得 之最高穩定化電池效率爲9.1%。 未使用任何抗反射塗覆層（ARC)獲得此項結果’該結果 係與帶有ARC (藉NREL證實）以及使用LPCVD-ZnO所獲得 之全球冠軍電池之9.47%可相媲美。 應注意本9.1%安定的3-5丨：11電池係配置於市售1^八1-\1 反應器內。 ❹ 於1平方厘米電池大小上只發現可用來將LPCVD-ZnO 上之迷你模組最佳化。最佳經過光浸漬之迷你模組係藉於 伊斯普拉之JRC之ESTI實驗室分別決定特徵（參考第13 圖）。於LPCVD-ZnO上所得穩定化效率8.32%係比先前記錄 的As ahi-U上迷你模組之穩定化效率更高0.5 % (絕對效率）。Use a 200 nm layer thickness (Jsc = 14.8 mA/cm2, FF = 66.6%, and V〇c = 873 mV, see Figure 12) to obtain 8.6% stabilized cell efficiency (in Asahi- U). In the case of LPCVD-ZnO pre-TCO-14-, 201023370, a thickness of 240 nm is used (Jsc = 15.6 mA/cm2, FF = 66.7%, and V〇c = 876 mV). 2) The highest stabilized battery efficiency available is 9.1%. This result was obtained without using any anti-reflective coating (ARC). This result is comparable to 9.47% of the global champion battery with ARC (confirmed by NREL) and obtained with LPCVD-ZnO. It should be noted that this 9.1% stable 3-5 丨:11 battery system is placed in a commercially available 1^81-\1 reactor.只 Only found on the 1 cm2 battery size can be used to optimize the mini-module on LPCVD-ZnO. The best light-impregnated mini-modules are determined by the ESTI laboratory of the JRC of Ispra (see Figure 13). The stabilization efficiency of 8.32% on LPCVD-ZnO is 0.5% higher than the stability of the previously recorded As ahi-U mini-module (absolute efficiency).
Asahi二氧化錫上及LPCVD-ZnO上之最佳化a-Si:H頂 電池已與底電池組合來實現微型銜接裝置。第14 φ 圖顯示發明人於兩種TCOs上獲得最佳結果之I (V)曲線。 目前正在硏究於LPCVD-ZnO上最大化微型裝置效率之 多種方法，其中兩種方法繪圖於第12圖亦即P1及P2。一 方面可製造具有高；he値之電池（P1)，另一方面也可製造具 有高Vm値之裝置（P2)。因此至目前爲止於LPCVD-ZnO所 得最佳效率仍略低於於Asahi二氧化錫上所達成之最佳效 率。有明確文獻記載維持高V〇c及FF而未減低：he之方式 係於i層沉積期間使用砂烷濃度側寫資料。方法2(P2)係遵 -15- 201023370 循此項途徑。 經由於微型迷你模組中實施最佳a-Si:H電池於Asahi 二氧化錫上作爲頂電池，可實現撇取最高10%之穩定化孔 隙效率（第9圖）。 經由微調底電池之i層及復合接面可進一步最佳化微 型裝置效率。此外，於此等裝置中頂電池與底電池間並無 倂入反射器，預期可更進一步改良效率。 義 要言之，a-Si:H單一接面p-i-n裝置係於Asahi-U及 LPCVD-ZnO前TC0上硏究。於對各個TC0之p層最佳化 後，於發明人之R & D ΚΑΙ Μ反應器中沉積p-i-n電池之i 層厚度系列。本硏究顯示於Asahi-U上之i層厚度減低可獲 得穩定化效率增高（對200奈米之i層厚度高達8.6%)。該 種情況對LPCVD-ZnO前TC0而言爲不同，此處本硏究指出 對240奈米至300奈米範圍之i層獲得最高穩定效率。於 此TC0上進一步對裝置進行最佳化之後，最高穩定效率達 〇 9.1%。未施用抗反射塗覆層（ARC)至前玻璃基材上即可獲得 此項結果。 【圖式簡單說明】 第1圖：先前技術，薄膜矽PV電池之層堆疊。 第2圖：片電阻及其均勻度呈基材數目之函數之安定 性。 第3圖：玻璃上不同TC0層之光透射比》 第4圖：厚度及片電阻之表面作圖。 -16- 201023370 第5圖：對具有不同厚度之兩層呈B2H*/DEZ比之函 數，移動性及於波長600奈米之濁度（底圖）。 第6圖：濁度及對不同濁度値之總透射率（頂圖）。對 LPCVD氧化鋅之濁度値比較市售經二氧化錫塗覆之玻璃顯 著較高。角度解析散射測量値指示LPCVD氧化鋅比經二氧 化矽塗覆玻璃更有效率散射光。 第 7a、7b 圖：Asahi-U 層（a)及典型 LPCVD-ZnO 層（b) A 之表面之掃描電子顯微鏡（SEM)顯微相片。 第8a、8b圖：如用於本硏究之Asahi-U及LPCVD-ZnO 之總透射（使用指數匹配液體測量）及漫射透射；ZnO爲上曲 線。 第9圖：於初始態及光浸漬態，、Jsc、FF及效率 呈i層厚度之函數（由200奈米至3 00奈米，於3.35埃/秒）。 Asahi-U用作爲p-i-n電池之前TCO。爲了改良統計數字， 考慮每個i層厚度5至10個電池。 〇 第10圖：於初始態及光浸漬態中，V〇c、Jsc、FF及效 率呈i層厚度之函數。LPCVD-ZnO用作爲前TCO。i層厚度 係由180奈米變化至400奈米（於3.35埃/秒）。爲了改良統 計數字，考慮4個至7個電池。 第11圖··已經p-i-na-Si:H電池沉積之LPCVD-ZnO及 Asahi-U之光浸漬分解所導致之相對效率變化。 第1 2圖：於LPCVD-ZnO及Asahi-U上所得至目前爲止 爲最佳之a-Si:H電池之I(V)曲線，ZnO爲外曲線。 -17- .201023370 第13圖：於LPCVD-ZnO上最佳p-i-n a-Si:H (經光浸 漬）10x10平方厘米迷你模組之I (V)曲線。孔隙面積係經 過瑞士聯邦度量衡局認證。 第14圖：於Asahi二氧化錫前TCO及LPCVD-ZnO前 TCO上至目前爲止所得最佳1平方厘米微型銜接電池之 I(V)曲線。已經開發若干方法（此處爲P1及P2)用於沉積 gc-Si:H i 層於 LPCVD-ZnO 前 TC◦上。 ^ 第15圖：穩定化（1000小時，1日光，50°C )微型10x10 平方厘米迷你模組之9.94%效率之人1^1.51-¥特性。作爲前 TC0，已經施用Asahi二氧化錫。 第16圖：對若干乙砸烷/DEZ比之厚度層DEZ/水比及 總透射比之函數。 【主要元件符號說明】 40 光 伏 電 池 4 1 透 明 基 材 42 透 明 導 電 性 氧 化 物（TCO)、TCO前接點 43 活 性 光 伏 層 44 次 層 45 相 鄰 次 層 46 最 末 次 層 47 後 接 點 層 48 反 射 層 、 後 反 射 器 -18 *Optimization of Asahi Tin Oxide and LPCVD-ZnO The a-Si:H top cell has been combined with a bottom cell to implement a micro-coupling device. The 14th φ plot shows the inventor's I (V) curve for the best results on both TCOs. A number of methods for maximizing the efficiency of microdevices on LPCVD-ZnO are currently being investigated, two of which are plotted in Fig. 12, namely P1 and P2. In one aspect, a battery (P1) having a high height can be manufactured, and on the other hand, a device (P2) having a high Vm can be manufactured. Therefore, the best efficiency achieved by LPCVD-ZnO so far is still slightly lower than that achieved by Asahi tin dioxide. There is a clear document that maintains high V〇c and FF without reducing: he is based on the side of the i-layer deposition using shale concentration side data. Method 2 (P2) follows -15-201023370. By using the best a-Si:H battery in the microminiature module as the top cell on Asahi tin dioxide, it is possible to achieve a stabilizing hole efficiency of up to 10% (Fig. 9). The efficiency of the microdevice can be further optimized by fine tuning the i-layer and the composite junction of the bottom cell. In addition, there is no intrusion reflector between the top cell and the bottom cell in such devices, and it is expected that the efficiency can be further improved. In a nutshell, the a-Si:H single junction p-i-n device is based on Asahi-U and LPCVD-ZnO pre-TC0. After optimizing the p layer of each TC0, the i-layer thickness series of the p-i-n battery was deposited in the inventor's R & D ΚΑΙ Μ reactor. This study shows that the reduction in the thickness of the i layer on the Asahi-U results in an increase in stabilization efficiency (up to 8.6% for the 200 nm layer). This situation is different for LPCVD-ZnO pre-TC0, where the study indicates that the highest stability efficiency is obtained for the i layer in the range of 240 nm to 300 nm. After further optimizing the device on this TC0, the maximum stable efficiency is 9.1 9.1%. This result can be obtained without applying an anti-reflective coating (ARC) to the front glass substrate. [Simple description of the diagram] Figure 1: Prior art, layer stacking of thin film 矽PV cells. Figure 2: The stability of the sheet resistance and its uniformity as a function of the number of substrates. Figure 3: Light transmittance of different TC0 layers on glass. Figure 4: Surface plot of thickness and sheet resistance. -16- 201023370 Figure 5: The function of the B2H*/DEZ ratio for two layers with different thicknesses, mobility and turbidity at a wavelength of 600 nm (basemap). Figure 6: Turbidity and total transmittance for different turbidity (top). The turbidity of LPCVD zinc oxide is significantly higher than that of commercially available tin oxide coated glass. Angle-resolved scatterometry 値 indicates that LPCVD zinc oxide scatters light more efficiently than bismuth dioxide-coated glass. Figures 7a, 7b: Scanning electron microscopy (SEM) micrographs of the surface of the Asahi-U layer (a) and the typical LPCVD-ZnO layer (b) A. Figures 8a, 8b: Total transmission (measured using exponential matching liquid) and diffuse transmission of Asahi-U and LPCVD-ZnO used in this study; ZnO is the upper curve. Figure 9: In the initial state and the photo-impregnated state, Jsc, FF and efficiency are a function of the thickness of the i-layer (from 200 nm to 300 nm at 3.35 Å/sec). Asahi-U is used as a TCO before the p-i-n battery. In order to improve the statistics, consider 5 to 10 cells per layer thickness. 〇 Figure 10: V〇c, Jsc, FF and efficiency are a function of the thickness of the i layer in the initial state and the light immersion state. LPCVD-ZnO is used as the front TCO. The thickness of the i layer varies from 180 nm to 400 nm (at 3.35 Å/sec). To improve the statistics, consider 4 to 7 batteries. Figure 11 · Relative efficiency changes caused by light immersion decomposition of LPCVD-ZnO and Asahi-U deposited by p-i-na-Si:H cells. Figure 1 2: The I(V) curve of the a-Si:H battery which is the best obtained on LPCVD-ZnO and Asahi-U until now, and ZnO is an external curve. -17- .201023370 Figure 13: I (V) curve of the best p-i-n a-Si:H (light immersed) 10x10 cm2 mini module on LPCVD-ZnO. The pore area is certified by the Swiss Federal Bureau of Weights and Measures. Figure 14: I(V) curve of the best 1 cm2 micro-connected cell obtained from Asahi tin dioxide pre-TCO and LPCVD-ZnO pre-TCO. Several methods (here P1 and P2) have been developed for depositing a gc-Si:H i layer on the LPCVD-ZnO front TC◦. ^ Figure 15: Stabilization (1000 hours, 1 daylight, 50 °C) 1.10% efficiency of the miniature 10x10 cm2 mini module 1^1.51-¥ characteristics. As the former TC0, Asahi tin dioxide has been applied. Figure 16: A function of the DEZ/water ratio and total transmittance for a thickness layer of several decane/DEZ ratios. [Main component symbol description] 40 Photovoltaic cell 4 1 Transparent substrate 42 Transparent conductive oxide (TCO), TCO front contact 43 Active photovoltaic layer 44 Sublayer 45 Adjacent sublayer 46 Last layer 47 Rear contact layer 48 Reflective layer, back reflector-18 *