201220514 四、 指定代表圖： (一） 本案指定代表圖為：第（1)圖。 (二） 本代表圖之元件符號簡單說明： 1 〇〜薄膜太陽電池； 11〜基板； 12〜金屬電極層（第1電極層）； 13〜第1半導體層； 14〜以c-si層（純半導體層）； 15〜第2半導體層； 16〜透明電極（TC0)層（第2電極層）； 3 0〜結晶粒界； 131〜η型a-Si區域（η型半導體區域）； 132〜Si Οχ區域（第1絕緣體區域）； 151〜ρ型a-si區域（ρ型半導體區域）； 152〜SiOx區域（第2絕緣體區域）。 五、 本案若有化學式時，請揭示最能顯示發明特徵的化學式： 無0 六、發明說明： 【發明所屬之技術領域】 本發明係關於薄膜太陽電池以及具有與上述薄膜太陽 電池相同構造的半導體裝置。 201220514 【先前技術】 山近年來，以暖化等環境問題的觀點來看，不排出二氧 化碳的乾淨能源-太陽電池受到注目。太陽電池中有各種種 類.形態，但光能轉換為電力的光電轉換效率高的太陽電 池，，知的為使用高純度石夕單結晶的單結晶石夕太陽電池。 單結晶石夕太陽電池，-般使用稱作pn接合的構造，係 摻雜P型雜質的p層與摻雜Μ雜質的n層相接。又分 別安裝電極至ρ層與η層。 ρ型半導體與η型半導體接合時’η型半導體側的電子 擴散至電子密度低的ρ型半導體區域，同樣地電洞係相反 於電子的產生。如此的載子移動持續至維持熱平衡狀態， 結果’ρη接合界面附近η區域中的電子不足，出現帶正電 的空間電荷，ρ側相反地產生帶負電的空間電荷，引起相 當於接合前的費米（Fermi)準位差的能量電位障壁。此電位 差稱作pn接合的擴散電位，擴散電位存在的區域係稱作ρ 往η的遷移區域。由於此遷移區域的載子密度比表體……㈧ 少，也稱作空乏層。入射空乏層的光能量在空乏層產生載 子時，由於擴散電位形成的電場，電子往η層、電洞往ρ 層移動。這些光激載子由兩層安裝的電極取出至外部，藉 此太陽電池動作。如此一來，ρη接合中空乏層作用為發電 層（光電轉換層）。 單結晶矽太陽電池如上所述地光電轉換效率優異。不 過，由於使用高純度矽單結晶作為半導體基板，有生產成 本變高的問題。又，由於矽單結晶的基板係鑄塊切片製造 201220514 的，不容易薄膜化.大面積化。 解決薄膜化、大面積化、低成本化的各種問題的太陽 電池，非晶（amorphous)矽太陽電池實用化。非晶矽太陽電 池中，一般使用稱作pin接合的構造，係在pn接合中預先 設置相當於空乏層的層。在此pn接合中，p層與n層之間 所夾的非摻雜的純半導體層（i層）作用為發電層。 非晶矽太陽電池，雖然光電轉換效率比單結晶矽太陽 電池低，由於可以利用化學氣相成長法（CVD法）製造容 易薄膜化.大面積化。又，可以削減原料的使用量，具有 可以以低成本製造的優點。不$，非晶石夕太陽電池對太陽 光光譜的光吸收特性狹小，而且由於有光入射的光電對換 效率下降（Stabler-Wronski效應）的光惡化問題，持續長 期間供給穩定電力有困難。 為了解決這些光吸收特性、光惡化問題，近年來奈米 到微米尺寸的矽微結晶構成的薄膜用於i層的微結晶矽太 陽電池正在開發。由於微結晶矽與非晶矽同樣地可以以電 漿CVD法製造，薄膜化、大面積化、低成本化容易的同時， 由於構成成分為結晶質，可以改善上述光吸收性、光惡化 的問題。又，光激載子的移動度還具有比非晶矽大的特長。 [先行專利文件] [非專利文件1 ] ”非晶矽/薄膜結晶矽堆疊型太陽電 池’’ [on line(上線）]，三洋電機株式會社[平成21年12 月 14 日檢索]，網路〈url : http//sanyo. com/technical review/jp/n〇75/pdf/7504.pdf> 4 201220514 【發明内容】 [發明所欲解決的課題] 儘管具有這些特長，目前製造的微結晶石夕太陽電池的 光電轉換效率只不過是與非晶⑦太陽電池同程度。微結晶 石夕太陽電池的光電轉換效率低的主因有幾個要考慮，其中 個例如微結晶石夕薄膜内形成的結晶粒界中發生漏電流 (非專利文件1 )。 例如pin接合的太陽電池中，為了使光激載子效率良 好移動電層的i層的p層側面與n層側面之間要進行 效率佳的電荷分離，產生足夠的電位差很重要。因此，丄 層必須具有用以維持此電位差的膜電阻。 不過，微結晶石夕薄膜，在製膜過程中，結晶粒子往膜 成長方向柱狀成長，結果，這些柱狀晶之間形成往相同方 :延伸的結晶粒界。如此的柱狀晶多數排列構造(以下為 柱狀構造」）中，在柱狀晶間的結晶粒界中容易產生電流 漏泄’由於維持膜成長方向的電位差變得困難，有 換效率下降的可能性。 得 ::明所欲解決的課題，係在具有柱狀構造的薄膜太 呈中’提供可以抑制漏電流薄膜太陽電池。又，對於 :有同樣的結晶構造的發光二極體等的半導體裝置，提供 以抑制漏電流引起的性能下降的半導體裝置。 [用以解決課題的手段] 用以解決上述課題而形成的本發明的薄膜太陽電池， 5 201220514 具有第1電極層、第1半導體層、純半導體層、帛2半導 體層、第2電極層依序堆疊的構造，其特御在於：上述第 1半導體層由面内方向互相離間形成的複數的n型半導體 區域、以及這些η型半導體區域之間形成的第丨絕緣區域 所構成；上述純半導體層具有往堆疊方向延伸的柱狀晶排 列的柱狀構造；上述第2半導體層由面内方向互相離間形 成的複數的ρ型半導體區域、以及這些？型半導體區域之 間形成的第2絕緣區域所構成；各層平行的平面上，投影 上述ρ型半導體區域以及上述η型半導體區域之際一 ρ 型半導體區域與最接近上述ρ型半導體區域的半導體 區域之間的距離在上述純半導體層的柱狀晶的上述平面上 的徑平均值以上。 又，本發明的柱狀晶係從下層到上層貫通的單一結晶 粒子。因此，鄰接的柱狀晶間連接上層與下層，形成結晶 粒界。又，本發明的2區域間的「距離」’係其投影平面 内的2區域之間的長度最小值。 用以解決上述課題而構成的本發明的半導體裝置 具有第1電極層、第B導體層、中間半導體層、第2半 導體層、第2電極層依序堆疊的構造，其特御在於：上述 第1半導體層由面内方向互相離間形成的複數的η型半導 體區域、以及這些η型半導體區域之間形成的帛1絕緣區 域所構成；上述中間半導體層具有往堆疊方向延伸的柱狀 晶排列的柱狀構造；上述第2半導體層由面内方向互相離 間形成的複數的ρ型半導體區域、以及這些ρ型半導體區 201220514 .域之門形成的第2絕緣區域所構成；各層平行的平面上， 才又:上述P型半導體區域以及η型半導體區域之際，一 p 型半導體區域與最接近上述P型半導體區域的η型半導體 區域之間的距離在上述中間半導體層的柱狀晶的上述平面 上的徑平均值以上。 [發明效果] 本發明的薄膜太陽電池的構造，係在第1半導體層與 第2半導體層的一部分設置絕緣體區域，並在純半導體層 的柱狀晶之間形成的膜成長方向延伸的結晶粒界的一方或 兩方的端部配置上述絕緣體區域。因此，由於膜成長方向 的結晶粒界可以防止電流漏泄，可以維持純半導體層的第 1半導體層側的面與第2半導體層的面之間的電位差，可 以提高薄膜太陽電池的光電轉換效率。 又’本發明的薄膜太陽電池構造可用於具有柱狀晶構 造半導體層的其他半導體裝置。例如，由於光電二極體與 太陽電池在構造、功能方面相同’本發明構造的薄膜太陽 電池可以原封不動地應用於光電二極體。雷射二極體、發 光二極體等的半導體裝置中，與太陽電池間的不同，只是 光轉換為電或電轉換為光的功能面的差異，構造本身與太 陽電池相同。因此，本發明構造的半導體裝置也可以適人 用於太陽電池以外的用途，可以對提高半導體裝置的功能 作貢獻。 【實施方式】 7 201220514 首先，使用第5〜7圖略概說明習知的pin接合型薄膜 太陽電池。 第5圓係顯示pin接合型薄臈太陽電池的一般構造的 概略縱剖面圖。此薄膜太陽電池2G具有在基板21上依序 堆疊金屬電極層22、η型非晶矽（a-Si )層23、微結晶矽（α c-Si)層24、1)型a-Si層25、以及透明電極層26的構造。 此構造的薄膜太陽電池中，n型^以層23及1)型8_“層 25分別對應η層及ρ層，# c_Si層24對應i層。 第5圖的薄媒太陽電池2〇可以以例如電漿cm法製 造。因此，可以以比使用高純度矽單結晶（cSi)的c_si太 陽電池低的成本製&，而且可以易於實行大面積化、薄膜 化。又，由於發電層使用Mc — Si，可以抑制因a_si太陽電 池而成為問題的光惡化。 不過，/ic-Si中有源於柱狀構造的結晶粒界形成的問 題。第6圖的y C-Si薄膜的剖面TEM像中顯示，从中 形成結晶粒界（圖中的白條紋）往膜成長方向延伸。薄膜太 陽電池20的構造中 此結晶粒界30中由於漏電流流動而 性能下降。利用第7圖說明。 如第7圖所示，Pin構造的薄膜太陽電池20中，由於 光的入射在i層的”—。層24中產生光激载子(電子電洞 對）3卜此光激載子31中，電子31八在a_si層μ /， 電洞31B在a-Si層25中，由於 、μ c bi 層 24 的 p 型 an 層23側的面與η型a-Si層μ彳目，丨沾二 , a功側的面之間的電位差（電壓） 所形成的電場而移動。此時，右關 c . 有關β c-Si層24的電壓低 201220514 的話，電荷（電子與雷洞、 、 未充分分離而再結合，光電轉換 效率下降。 、 於疋，有關//C-Si居9/1 層24的電壓係愈高愈好。不過， /zc-Si層24中，由於报士， 由於形成如第6圖所示往膜成長方向延 伸的結晶粒界30，電壓變其& ^ ^ ^ 电《跫间時，與電流的流動方向32相 反’通過結晶粒界30流過漏雷法η七a糾t恭〜 构€ /爪（逆方向飽和電流）3 3。因 此’ 1層中使用/zc-Si等的微結晶半導體的薄膜太陽電池 中電壓不太高，由於電荷分離不充分，考慮到光電轉換效 率下降。 ' 要使光激載子移動效率佳，發電層的“-Si f 24的n 型“i層23側的面與p型a_Si@ 25側的面之間必須得 到充分的電位差’因此"c_Si層24必須具有高的膜電阻。 如上述，假設“-Si層24的膜電阻下降因結晶粒界3。發 生時’往膜成長方向施加電壓的情況與往面内方向施加電 壓的情況，本申請書發明者預測以Si Μ的電阻率不 同0 驗 為 了證實此預測’本中請書發明者進行以下所示的實 首先’為了測量膜成長方向的電阻率，你 ^ _ 干作成第8(a)圖 所不的構造的半導體裝置。此半導體裝置， 仕厚度4英寸 (inch)的玻璃基板50上，以濺鍍法形成厚度2〇〇毫微米（⑽) 的鎢獏51作為下部電極後，基板溫度加熱3〇〇它以電漿ϋ 法形成厚度2# m(微米）的β c-Si(微結晶 瞄 R〇， )m 52° U c-Si 膜52上，一邊使用蒸鍍光罩，一邊以真 為鍍法形成直徑 201220514 lmm(毫米）、厚谇ςηΛ ，… 度5〇〇nm的鋁臈53作為上部電極 鎢膜51與鋁膜53 电極於疋， Μ之間施加直流電壓 的膜成長方向的雷阳扣械， 里# c iu膜52 门的電阻。根據此測量值與"c_s 的積’算出臈成長方向的體積電阻率。 a 的半導俨奘番 、使用第8(b)圓所示構造 的+導體裝置。此半導體裝置，在厚 璃基板5〇卜η. at- 央寸（lnch)的玻 上，基板溫度加熱㈣以電聚m法形成厚度 2/zm(微米）的“ c_sj膜52。於是，r 犋w於疋，以四端子法測 膜52的表面電阻a，廿招M ll 士 ^ 51 並根據此表面電阻率的測量值盥 膜厚2“的積，算出面内方向的體積電阻率。’、 膜成長（面直立）方向的體積電阻率與面内方向的體積 電阻率的測量結果顯示於第9(a)圖。根據此圖，看出膜成 長方向與面内方向之間體積電阻率相差接近3位數。又， 第9(a)圖的橫軸顯示冑52的上的面内測量位置， 以某點為基準，此測量位置冑〇關，從此開始測量位置如 第9(b)圖所示往A方向變化的情況以及往與其垂直的8方 向變化的情況。 根據這些貫驗結果，本申請書發明者，藉由對習知的 薄膜太陽電池改變P層與„層的構造，找出解決上述問題 的方法。以下，進行說明本發明的薄膜太陽電池。 [第一實施例] 本發明的薄膜太陽電池在第丨圖中顯示第一實施例的 概略縱剖面圖。 本實施例的薄膜太陽電池10，在基板U上具有金屬 201220514 電極層（第1電極層）12、第1半導體層13、yc_Si層（純 半導體層）14、第2半導體層15、透明電極（tc〇)層（第2 電極層）ΐδ依序堆疊的構造。在此，第1半導體層13由面 内方向互相離間形成的複數的η型a-Si區域（η型半導體 區域）131、以及這些η型a-Si區域131之間形成的si〇x 區域（第1絕緣體區域）132所構成。又，上述第2半導體 層15由面内方向互相離間形成的複數的？型&_以區域 型半導體區域）151、以及這些口型a_Si區域151之間形成 的SiOx區域（第2絕緣體區域）152所構成。 本實施例的薄膜太陽電池1〇中，n型a_Si區域131 與P型a-Si區域151，夾住；zc-s^ 14不互相重疊配置， 又如第1圖所示，"C-S…4内形成的結晶粒界3〇的一 方或兩方的端部’具有配置絕緣構件的Si〇x形成區域（第 1絕緣層132及第2絕緣層152)的構造。由於採用如此的 構造，η型a-Si區域131與"a_Si區域151之間，可 以防止發生電流短路。 不過’事先知道B-Si層14中何處形成結晶粒界3〇 是困難的。因此，實際製作的薄膜，求得基板面上 投影的柱狀晶的徑(大小)的平均值(以下為「平均柱徑」）， 據此決定η型a-Si區域131與㈣心區域i5i的位置。201220514 IV. Designated representative map: (1) The representative representative of the case is: (1). (b) The symbol of the symbol of this representative diagram is briefly described: 1 薄膜 ~ thin film solar cell; 11 ~ substrate; 12 ~ metal electrode layer (first electrode layer); 13 ~ first semiconductor layer; 14 ~ with c-si layer ( Pure semiconductor layer); 15 to 2nd semiconductor layer; 16 to transparent electrode (TC0) layer (second electrode layer); 3 0~ crystal grain boundary; 131 to n-type a-Si region (n-type semiconductor region); ~Si Οχ region (first insulator region); 151 to p-type a-si region (p-type semiconductor region); 152 to SiOx region (second insulator region). 5. If there is a chemical formula in this case, please disclose the chemical formula that best shows the characteristics of the invention: None. 6. Description of the Invention: Technical Field of the Invention The present invention relates to a thin film solar cell and a semiconductor having the same structure as the above-described thin film solar cell. Device. 201220514 [Prior Art] In recent years, from the viewpoint of environmental issues such as warming, solar cells, which do not emit carbon dioxide, have attracted attention. There are various types of solar cells in the solar cell, but the solar energy is converted into a solar cell with high photoelectric conversion efficiency. It is known that a single crystal stone solar cell using high-purity single crystal is used. In a single crystal solar cell, a structure called a pn junction is generally used, and a p-layer doped with a P-type impurity is in contact with an n-layer doped with yttrium impurities. The electrodes are also mounted to the p layer and the n layer, respectively. When the p-type semiconductor is bonded to the n-type semiconductor, electrons on the n-type semiconductor side are diffused to the p-type semiconductor region having a low electron density, and similarly, the hole system is opposite to the generation of electrons. Such carrier movement continues until the thermal equilibrium state is maintained, and as a result, the electrons in the η region near the interface of the ρη junction are insufficient, and a positively charged space charge occurs, and the ρ side reversely generates a negatively charged space charge, causing a charge equivalent to the junction. Fermi's quasi-position energy potential barrier. This potential difference is called the diffusion potential of the pn junction, and the region where the diffusion potential exists is called the transition region of ρ to η. Since the carrier density of this migration region is less than that of the surface (8), it is also called a depletion layer. When the light energy of the incident depletion layer generates a carrier in the depletion layer, electrons move toward the η layer and the hole to the ρ layer due to the electric field formed by the diffusion potential. These photo-active carriers are taken out to the outside by two layers of electrodes, whereby the solar cell operates. As a result, the ρη junction hollow layer acts as a power generation layer (photoelectric conversion layer). The single crystal germanium solar cell is excellent in photoelectric conversion efficiency as described above. However, since high-purity germanium single crystal is used as the semiconductor substrate, there is a problem that the production cost becomes high. In addition, since the substrate-based ingot slicing of the single crystal is manufactured in 201220514, it is not easy to be thinned and has a large area. A solar cell that solves various problems of thin film formation, large area, and low cost, and an amorphous solar cell is put into practical use. In the amorphous tantalum solar cell, a structure called pin bonding is generally used, and a layer corresponding to the depletion layer is provided in advance in the pn junction. In this pn junction, the undoped pure semiconductor layer (i layer) sandwiched between the p layer and the n layer functions as a power generation layer. The amorphous germanium solar cell has a lower photoelectric conversion efficiency than a single crystal germanium solar cell, and can be easily thinned by a chemical vapor phase growth method (CVD method) to have a large area. Further, it is possible to reduce the amount of raw materials used, and it is advantageous in that it can be manufactured at low cost. The amorphous Austenitic solar cell has a narrow light absorption characteristic to the solar spectrum, and it is difficult to supply stable power for a long period of time due to the problem of light deterioration of the photoelectric conversion efficiency (Stabler-Wronski effect) in which light is incident. In order to solve these problems of light absorption characteristics and light deterioration, in recent years, a film composed of nano-sized micro-crystals of cerium to micro-crystals for the i-layer is being developed. In the same manner as the amorphous ruthenium, the microcrystalline ruthenium can be produced by a plasma CVD method, and it is easy to form a thin film, a large area, and a low cost, and the composition is crystallized, thereby improving the problem of light absorbing property and light deterioration. . Moreover, the mobility of the photoexcited carrier also has a feature larger than that of the amorphous crucible. [Private Patent Document] [Non-Patent Document 1] "Amorphous 矽/Thin Crystal 矽 Stacked Solar Cell" 'on line (on line), Sanyo Electric Co., Ltd. [Search on December 14, 2011], network <url: http//sanyo.com/technical review/jp/n〇75/pdf/7504.pdf> 4 201220514 [Summary of the Invention] [Problems to be Solved by the Invention] Despite these advantages, the currently produced microcrystalline stone The photoelectric conversion efficiency of the solar cell is only the same as that of the amorphous 7 solar cell. There are several main reasons for the low photoelectric conversion efficiency of the microcrystalline stone solar cell, one of which is crystallized in the microcrystalline film. Leakage current occurs in the grain boundary (Non-Patent Document 1). For example, in a pin-bonded solar cell, in order to make the photoexciter efficient, it is efficient to move the p-layer side and the n-layer side of the i-layer of the electric layer. It is important to separate the charge and generate a sufficient potential difference. Therefore, the tantalum layer must have a membrane resistance to maintain this potential difference. However, in the microcrystalline crystal film, the crystal particles grow in a column shape in the film growth direction. As a result, a crystal grain boundary extending to the same side is formed between the columnar crystals. In such a columnar crystal majority arrangement structure (hereinafter, a columnar structure), a current is easily generated in the crystal grain boundary between the columnar crystals. Leakage 'Because it is difficult to maintain the potential difference in the film growth direction, there is a possibility that the switching efficiency is lowered. It is a problem that the problem to be solved is that a film having a columnar structure is too large to provide a solar cell capable of suppressing leakage current. Further, a semiconductor device such as a light-emitting diode having the same crystal structure is provided with a semiconductor device for suppressing performance degradation due to leakage current. [Means for Solving the Problem] The thin film solar cell of the present invention formed to solve the above problems, 5 201220514, has a first electrode layer, a first semiconductor layer, a pure semiconductor layer, a 帛2 semiconductor layer, and a second electrode layer. The structure of the sequential stack is characterized in that the first semiconductor layer is composed of a plurality of n-type semiconductor regions formed by in-plane directions and a second insulating region formed between the n-type semiconductor regions; the pure semiconductor The layer has a columnar structure in which columnar crystals are arranged to extend in the stacking direction; and the plurality of p-type semiconductor regions in which the second semiconductor layer is formed apart from each other in the in-plane direction, and these? a second insulating region formed between the semiconductor regions; a p-type semiconductor region and a semiconductor region closest to the p-type semiconductor region when the p-type semiconductor region and the n-type semiconductor region are projected on a plane parallel to each layer The distance between them is equal to or larger than the average value of the diameter of the columnar crystal of the pure semiconductor layer. Further, the columnar crystal system of the present invention is a single crystal particle penetrating from the lower layer to the upper layer. Therefore, the adjacent columnar crystals connect the upper layer to the lower layer to form a crystal grain boundary. Further, the "distance" between the two regions of the present invention is the minimum length between the two regions in the projection plane. The semiconductor device of the present invention configured to solve the above problems has a structure in which a first electrode layer, a B-th conductor layer, an intermediate semiconductor layer, a second semiconductor layer, and a second electrode layer are sequentially stacked, and the above-described a semiconductor layer comprising a plurality of n-type semiconductor regions formed between the in-plane directions and a 帛1 insulating region formed between the n-type semiconductor regions; the intermediate semiconductor layer having a columnar crystal arrangement extending in a stacking direction a columnar structure; the second semiconductor layer is composed of a plurality of p-type semiconductor regions formed by in-plane directions and a second insulating region formed by the gates of the p-type semiconductor regions 201220514; and the layers are parallel to each other Further, in the case of the P-type semiconductor region and the n-type semiconductor region, a distance between a p-type semiconductor region and an n-type semiconductor region closest to the P-type semiconductor region is at a plane of the columnar crystal of the intermediate semiconductor layer Above the average value of the diameter. [Effect of the Invention] The structure of the thin film solar cell of the present invention is a crystal grain in which an insulator region is provided in a part of the first semiconductor layer and the second semiconductor layer, and a film growth direction is formed between the columnar crystals of the pure semiconductor layer. The insulator region is disposed at one or both ends of the boundary. Therefore, the crystal grain boundary in the film growth direction can prevent current leakage, and the potential difference between the surface on the first semiconductor layer side of the pure semiconductor layer and the surface of the second semiconductor layer can be maintained, and the photoelectric conversion efficiency of the thin film solar cell can be improved. Further, the thin film solar cell structure of the present invention can be applied to other semiconductor devices having a columnar crystal structure semiconductor layer. For example, since the photodiode is identical in structure and function to the solar cell, the thin film solar cell constructed by the present invention can be applied to the photodiode as it is. In a semiconductor device such as a laser diode or a light-emitting diode, the difference from the solar cell is only a difference in the functional surface in which light is converted into electricity or electricity, and the structure itself is the same as that in the solar cell. Therefore, the semiconductor device of the present invention can be suitably used for applications other than solar cells, and can contribute to improvement of the function of the semiconductor device. [Embodiment] 7 201220514 First, a conventional pin junction type thin film solar cell will be briefly described using Figs. The fifth circular system shows a schematic longitudinal cross-sectional view of a general structure of a pin-joined thin tan solar cell. The thin film solar cell 2G has a metal electrode layer 22, an n-type amorphous germanium (a-Si) layer 23, a microcrystalline germanium (α c-Si) layer 24, and a) type a-Si layer sequentially stacked on the substrate 21. 25. The construction of the transparent electrode layer 26. In the thin film solar cell of this structure, the n-type layer 23 and the 1) type 8_"the layer 25 correspond to the η layer and the ρ layer, respectively, and the #c_Si layer 24 corresponds to the i layer. The thin-film solar cell of FIG. 5 can be For example, it can be produced by the plasma cm method. Therefore, it can be made at a lower cost than the c_si solar cell using high-purity bismuth single crystal (cSi), and can be easily enlarged and thinned. Also, since the power generation layer uses Mc. —Si, it is possible to suppress the deterioration of light which is a problem due to the a_si solar cell. However, the problem of the formation of crystal grain boundaries which are active in the columnar structure in /ic-Si. The cross-sectional TEM image of the y C-Si film in Fig. 6 It is shown that the crystal grain boundary (the white streak in the figure) is formed to extend in the film growth direction. In the structure of the thin film solar cell 20, the performance of the crystal grain boundary 30 is degraded due to the leakage current flow, which is illustrated by Fig. 7. As shown in Fig. 7, in the thin-film solar cell 20 of the Pin structure, since the light is incident on the i-layer "-. In the layer 24, a photoexciter (electron hole pair) is generated. In the photoexcited carrier 31, the electron 31 is in the a_si layer μ /, the hole 31B is in the a-Si layer 25, and the μ c bi layer The surface on the side of the p-type an layer 23 of 24 moves with the electric field formed by the potential difference (voltage) between the n-type a-Si layer and the surface of the a-side. At this time, the right turn c. When the voltage of the β c-Si layer 24 is low 201220514, the electric charge (electron and the mine hole are not sufficiently separated and recombined, and the photoelectric conversion efficiency is lowered., Yu Yu, about //C-Si) The higher the voltage of the 9/1 layer 24 is, the better. However, in the /zc-Si layer 24, due to the formation of the crystal grain boundary 30 extending toward the film growth direction as shown in Fig. 6, the voltage becomes & ^ ^ ^ Electric "when the daytime, contrary to the flow direction of the current 32" through the crystal grain boundary 30 through the leakage ray method η seven a correction t Gong ~ structure / claw (reverse direction saturation current) 3 3 In a thin film solar cell using a microcrystalline semiconductor such as /zc-Si in the first layer, the voltage is not too high, and the charge separation efficiency is insufficient due to insufficient charge separation. 'To make the photoexcited carrier move efficiently, the power generation layer A sufficient potential difference must be obtained between the n-type "i-layer 23 side surface of the -Si f 24" and the p-type a_Si@25 side surface. Therefore, the c_Si layer 24 must have a high film resistance. As described above, "The film resistance of the -Si layer 24 is lowered due to the grain boundary 3. When it occurs, the voltage is applied to the film growth direction." In the case where a voltage is applied in the in-plane direction, the inventors of the present invention have predicted that the resistivity of Si 不同 is different. In order to confirm the prediction, the inventor of the present invention performs the following first, in order to measure the resistance of the film growth direction. Rate, you ^ _ dry into a semiconductor device of the structure of Figure 8 (a). This semiconductor device, a thickness of 4 inches (inch) of the glass substrate 50, by sputtering method to form a thickness of 2 〇〇 nanometer ( (10)) After the tungsten crucible 51 is used as the lower electrode, the substrate temperature is heated by 3 Torr. It is formed by plasma ϋ method to form β c-Si (microcrystalline target R 〇, ) with a thickness of 2 # m (micrometer) m 52 ° U c- On the Si film 52, an aluminum crucible 53 having a diameter of 201220514 lmm (mm), a thickness 谇ςηΛ, ... 5 〇〇 nm was formed as a top electrode tungsten film 51 and an aluminum film 53 by a true plating method using a vapor deposition mask. The electrode is applied to the 成长, Μ 施加 直流 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 雷 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 a semi-conducting of a, using the + conductor of the structure shown in the 8th (b) circle The semiconductor device is formed on a glass substrate of a thick glass substrate, and the substrate temperature is heated (4) to form a "c_sj film 52 having a thickness of 2/zm (micrometer) by an electropolymerization method. Thus, r 犋w疋, the surface resistance a of the film 52 is measured by a four-terminal method, and the volume resistivity of the in-plane direction is calculated based on the product of the surface resistivity measured by the film thickness of 2". The measurement results of the volume resistivity in the film growth (face upright) direction and the volume resistivity in the in-plane direction are shown in Fig. 9(a). According to this figure, it is seen that the volume resistivity between the film growth direction and the in-plane direction differs by approximately three digits. Further, the horizontal axis of Fig. 9(a) shows the in-plane measurement position on the 胄52, and the measurement position is off based on a certain point, and the measurement position is started as shown in Fig. 9(b). The case where the direction changes and the case where it changes in the direction perpendicular to it. Based on these results, the inventors of the present application have found a method for solving the above problems by changing the structure of the P layer and the layer of the conventional thin film solar cell. Hereinafter, the thin film solar cell of the present invention will be described. [First Embodiment] A thin film solar cell of the present invention is shown in a first perspective view in the first embodiment. The thin film solar cell 10 of the present embodiment has a metal 201220514 electrode layer (first electrode layer) on the substrate U. 12) The first semiconductor layer 13, the yc_Si layer (pure semiconductor layer) 14, the second semiconductor layer 15, and the transparent electrode (tc) layer (second electrode layer) ΐ δ are sequentially stacked. Here, the first semiconductor The layer 13 has a plurality of n-type a-Si regions (n-type semiconductor regions) 131 formed by in-plane directions, and a si〇x region (first insulator region) 132 formed between the n-type a-Si regions 131. Further, the second semiconductor layer 15 has a plurality of SiO2 regions formed between the plurality of ?-type regions and 151 regions formed by the in-plane directions, and the SiOx regions formed between the gate-type a_Si regions 151 (2nd) Insulator area In the thin film solar cell of the present embodiment, the n-type a_Si region 131 and the p-type a-Si region 151 are sandwiched; zc-s^14 are not overlapped with each other, and as shown in Fig. 1 The "one or both ends of the crystal grain boundary 3" formed in the "CS...4" has a structure in which the Si〇x formation region (the first insulating layer 132 and the second insulating layer 152) in which the insulating member is disposed. With such a configuration, a current short circuit can be prevented between the n-type a-Si region 131 and the "a_Si region 151. However, it is difficult to know in advance where the crystal grain boundary 3 is formed in the B-Si layer 14. The film actually produced is obtained by the average value of the diameter (size) of the columnar crystal projected on the substrate surface (hereinafter referred to as "average column diameter"), and accordingly, the n-type a-Si region 131 and the (four) core region i5i are determined. position.
Si層内的各柱狀晶的平均柱徑可以經由適當 製膜條件來控制。在此，對於某製膜條件的^^層Η :的均柱輕為R，在各層的平行平面P上，考慮投影η '區域131與？型a-Si區域HI的的情況。如上述， 11 201220514 由於η型a-Si區域131與！）型a —Si區域151，夾住"c_si 層14不互相重疊配置（第2(a)圖的上圖），平面p上投影 的η型a-Si區域131A與p型a-Si區域i51a，存在於互 相離間的位置上（第2(a)圖的下圖）。要使此平面p上最接 近的η型a-Si區域131A與p型a-Si區域15u的距離L 為L2 R’分別配置第1半導體層13内的n型a_Si區域ι3ι 與第2半導體層15内的p型a-Si區域151的話，# c_Si 層14中形成的結晶粒界3 0的一方或兩方端部可以成為第 1絕緣層132及第2絕緣層152配置的構造。 又，η型a-Si區域131與p型a-si區域151中，第 2(a)圖的下圖所示的帶狀之外，可以利用如第2(b)圖的島 狀形狀（η型a-Si區域131B與p型a-Si區域151B)。 其次，使用電漿CVD法的薄膜太陽電池丨〇的製造順 序，參照第3圖來說明。 首先，基板11上，依序堆疊保護層（未圖示）、第j電 極層（金屬薄膜）12、以及只有η型a-Si構成的第1半導體 層13(第3(a)及（b)圖）。於是，第1半導體層13上形成光 阻圖案40 (第3(c)圖），施行含氧電漿的氧化處理。藉此， 形成η型a-Si區域131及Si 〇χ區域132(第3(d)圖）。之 後，除去光阻40(第3(e)圖）。 其次’第1半導體層13上依序形成yc-si層14、以 及只有P型a-Si構成的第2半導體層15(第3(f)及（g) 圖）。形成光阻41以及氧化處理產生的第2半導體層15内 的P型a-Si區域151及Si Οχ區域152(第3(h)及（i)圖）。The average column diameter of each columnar crystal in the Si layer can be controlled by appropriate film formation conditions. Here, for a certain film formation condition, the uniform column light is R, and on the parallel plane P of each layer, the projection η 'region 131 and ? The case of the type a-Si region HI. As mentioned above, 11 201220514 due to the n-type a-Si area 131 and! ) a-Si region 151, sandwiched "c_si layer 14 does not overlap each other (top view of Fig. 2(a)), n-type a-Si region 131A and p-type a-Si region projected on plane p I51a exists at a position apart from each other (the lower diagram of Fig. 2(a)). The n-type a_Si region ι3ι and the second semiconductor layer in the first semiconductor layer 13 are disposed such that the distance L between the closest n-type a-Si region 131A and the p-type a-Si region 15u on the plane p is L2 R'. In the p-type a-Si region 151 in the 15th, one or both end portions of the crystal grain boundary 30 formed in the #c_Si layer 14 may have a structure in which the first insulating layer 132 and the second insulating layer 152 are disposed. Further, in the n-type a-Si region 131 and the p-type a-si region 151, in addition to the strip shape shown in the lower diagram of the second (a) diagram, an island shape as in the second (b) diagram can be used ( The n-type a-Si region 131B and the p-type a-Si region 151B). Next, the manufacturing procedure of the thin film solar cell using the plasma CVD method will be described with reference to Fig. 3. First, a protective layer (not shown), a jth electrode layer (metal thin film) 12, and a first semiconductor layer 13 made of only n-type a-Si are stacked on the substrate 11 (3(a) and (b) )))). Then, a photoresist pattern 40 (Fig. 3(c)) is formed on the first semiconductor layer 13, and an oxidation treatment of the oxygen-containing plasma is performed. Thereby, the n-type a-Si region 131 and the Si 〇χ region 132 are formed (Fig. 3(d)). Thereafter, the photoresist 40 is removed (Fig. 3(e)). Next, the yc-si layer 14 and the second semiconductor layer 15 composed of only P-type a-Si are sequentially formed on the first semiconductor layer 13 (Figs. 3(f) and (g)). The photoresist 41 and the P-type a-Si region 151 and the Si Οχ region 152 in the second semiconductor layer 15 which are generated by the oxidation treatment are formed (Figs. 3(h) and (i)).
12 201220514 之後，除去光阻41(第3(j)圖），形成透明電極（TC0)層 16 (第3 ( k )圖）。藉此，製造本實施例的薄膜太陽電池丄〇。 [第二實施例] 第一實施例所示的pin接合的薄膜太陽電池的構造， 也可以應用於Pn接合的薄膜太陽電池。第4圖中，第-實 施例的薄膜太陽雷#庙 防電池應用於pn接合，顯示第二實施例的薄 膜太陽電池的概略縱剖面圖。 #實k例係n型以c_Si層63與p型#層65互接 的叩接合型的薄膜太陽電池，夾住這些的2電極層中，使 用本發月的概必。即’本實施例的第】電極層Μ由往面内 方向互相離間形成的複數的金屬電極區域、以及這些 區域間形成的第1絕緣體區域622構成，第2電極層66由 " 向互相離間形成的複數的透明電極區域661、以 及這些區域間形成的第2絕緣體區域622構成。又，投影 在基板61的面上最接近的金屬電極區域621與透明電極區 ::1的距離，係以n型…i層63及…c_Sl“5 的平均柱徑以上配置。 由於採取如此的構造，即B ., L P使疋Pn接合的薄膜太陽電 池，也可以達到第一實施例同樣的效果。 又，各實施例所示的薄膜太陽電池構造，也可以庳用 於光雷-炻挪外， 愿用 尤1： 一極體、發光二極體等 ^ . '、他牛導體裝置。例如，發 尤一極體中’具 * , , 啕依序堆疊下部電極、η型包 覆（clad)層、活性層（中間半 電極的槿1 千導體層）、P型包覆層、上極 的構…直以來，活性層中使用例如_氮化銦 201220514 鎵）等的單結晶，但由於與太陽電池同樣地單結晶半導體成 本高且大面積化困難，近年來漸漸使用微結晶半導體代替 單結晶半導體。不過’即使使用微結晶半導體，由於發生 與太陽電池同樣的問題，藉由η型包覆層與p型包覆層中 使用與上述第1半導體層及第2半導體層相同的構造，可 以提高發光二極體的發光效率。 又’上述各實施例中，具有柱狀構造的層為微結晶半 導體的情況以範例說明，但本發明不限定於此。例如像多 釔SB半導體，即使結晶的粒徑大，具有柱狀晶構造的半導 體的話，與上述各實施例同樣地也適用本發明。 【圖式簡單說明】 [第1圖]係顯示本發明薄膜太陽電池的第一實施例構 造的概略縱剖面圖； 圖](a)〜（b)係顯示第一實施例的薄膜太陽電池 的P型半導體區域盥 ^ ^ 、 1 +導體區域在各層平行平面投影 之際的位置關係模式圖； [第 3 ϋ1r a)~(k)係顯示第一實施例的薄膜太陽電池 的製造順序概略剖面圖； [第4圖]係顯干太旅姐# 造的概略膜太陽電池的第二實施例構 面圖； [第5圖1位_ . ’、4不習知的薄膜太陽電池構造的概略 9[第6圖]係微妹曰 0日日矽溥膜的剖面TEM (透過型電 縱剖 子顯微 14 201220514 鏡）像； [第 式圖； 7圖]係顯示習知的12 201220514 After that, the photoresist 41 (Fig. 3(j)) is removed to form a transparent electrode (TC0) layer 16 (Fig. 3 (k)). Thereby, the thin film solar cell cartridge of the present embodiment was fabricated. [Second Embodiment] The configuration of the pin-bonded thin film solar cell shown in the first embodiment can also be applied to a Pn-bonded thin film solar cell. In Fig. 4, the film solar cell of the first embodiment is applied to a pn junction, and a schematic longitudinal cross-sectional view of the film solar cell of the second embodiment is shown. #实k例 The n-type tantalum-type thin-film solar cell in which the c_Si layer 63 and the p-type #65 are connected to each other, and the two-electrode layer sandwiching these are used, and it is necessary to use this month. That is, the "electrode layer" of the present embodiment is composed of a plurality of metal electrode regions formed to be apart from each other in the in-plane direction, and a first insulator region 622 formed between these regions, and the second electrode layer 66 is separated from each other by " The plurality of transparent electrode regions 661 formed and the second insulator region 622 formed between these regions are formed. Further, the distance between the metal electrode region 621 and the transparent electrode region::1 which are projected closest to the surface of the substrate 61 is set to be equal to or larger than the average column diameter of the n-type i layer 63 and ... c_S1 "5. The structure, that is, B., LP, can achieve the same effect as the first embodiment by the thin-film solar cell in which the 疋Pn is bonded. Moreover, the thin-film solar cell structure shown in each embodiment can also be used for the light-ray-to-beam movement. In addition, I would like to use a special one: a polar body, a light-emitting diode, etc. ', and his cattle conductor device. For example, in the hair of a pole, 'with *, , 堆叠 stack the lower electrode, n-type coating ( The clad) layer, the active layer (the 半1 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ In the same way as a solar cell, a single crystal semiconductor is costly and large in area, and in recent years, a microcrystalline semiconductor has been gradually used instead of a single crystal semiconductor. However, even if a microcrystalline semiconductor is used, the same problem as that of a solar cell occurs, and an n type is used. Coating and p-type cladding The same structure as the first semiconductor layer and the second semiconductor layer described above can be used to improve the luminous efficiency of the light-emitting diode. Further, in the above embodiments, the case where the columnar structure is a microcrystalline semiconductor is exemplified. However, the present invention is not limited to this. For example, in the case of a semiconductor having a columnar crystal structure even if the crystal grain size is large, such as a multi-turn SB semiconductor, the present invention is also applied in the same manner as in the above embodiments. [Fig. 1] is a schematic longitudinal cross-sectional view showing the structure of a first embodiment of a thin film solar cell of the present invention; Fig. 1(a) to (b) are views showing a P-type semiconductor region of the thin film solar cell of the first embodiment. ^, 1 + a pattern of positional relationship of conductor regions projected on parallel planes of each layer; [3rd ϋ1r a)~(k) shows a schematic sectional view of the manufacturing sequence of the thin film solar cell of the first embodiment; [Fig. 4] [Photograph of the second embodiment of the schematic film solar cell made by the system; [Fig. 5, 1 position _. ', 4 unobvious schematic structure of the thin film solar cell 9 [Fig. 6] The section of the micro-sister TEM (transmission type electron longitudinal sectional microphotograph 14201220514 mirror) image; [Formula of FIG.; 7] the conventional display system
Pln接合薄臈太陽電池動作模 [第8圖](a)〜（b)係用以測 膜成長方向與面内方向的電阻率 略縱剖面圖；以及 里具有結晶粒界的薄膜 所使用的半導體裝置的 的 概 [第9圖]係顯示具有結晶粒界的薄膜中測量位置的變 化與膜成長方向及面内方向的體積電阻率的變化關係圖 (a) ’以及顯示薄膜上的測量位置的概略圖（b)。 【主要元件符號說明】 1 〇〜薄膜太陽電池； 11〜基板； 12〜金屬電極層（第1電極層）； 13〜第1半導體層； W〜# c-Si層（純半導體層）； 15〜第2半導體層； 16〜透明電極（TC0)層（第2電極層）； 20〜薄膜太陽電池； 21〜基板； 22〜金屬電極層； 23 η型非晶碎（a_si)層； 24〜/z C-Si層（純半導體層）； 25〜P型a-Si層； 15 201220514 26〜透明電極層（第2電極層）； 3〇〜結晶粒界； 31光激載子（電子電洞對）； 31A〜電子·， 31B〜電洞； 32〜電流流動的方向； 33漏電流（逆方向飽和電流）流動的方向. 40、41〜光阻（光阻圖案）； ， 50〜破璃基板； 51〜鎢膜； 5 2〜以c - S i (微結晶石夕）膜； 53〜鋁膜； 61〜基板； 62〜第1電極層； 63〜η型"C-Si層； 65〜p型# c — si層； 66〜第2電極層； 131、131A、131B〜π型a-Si區域（n型半導體區域）； 132〜SiOx區域（第1絕緣體區域）； 151、151A、151B〜p型a-Si區域（p型半導體區域）； 152〜SiOx區域（第2絕緣體區域）； 621〜金屬電極區域； 622〜第1絕緣體區域； 661〜透明電極區域； 16 201220514 662〜第2絕緣體區域。 17Pln bonded thin solar cell operating mode [Fig. 8] (a) to (b) are longitudinal cross-sectional views of resistivity for measuring film growth direction and in-plane direction; and films used for crystal grain boundaries The outline of the semiconductor device [Fig. 9] shows the relationship between the change in the measurement position in the film having the crystal grain boundary and the volume resistivity in the film growth direction and the in-plane direction (a) 'and the measurement position on the display film. Schematic diagram (b). [Description of main component symbols] 1 〇 ~ thin film solar cell; 11 ~ substrate; 12 - metal electrode layer (first electrode layer); 13 ~ first semiconductor layer; W~# c-Si layer (pure semiconductor layer); ~ 2nd semiconductor layer; 16~ transparent electrode (TC0) layer (2nd electrode layer); 20~ thin film solar cell; 21~ substrate; 22~ metal electrode layer; 23 η-type amorphous broken (a_si) layer; /z C-Si layer (pure semiconductor layer); 25~P type a-Si layer; 15 201220514 26~transparent electrode layer (second electrode layer); 3〇~crystal grain boundary; 31 photoexcited carrier (electron Hole pair); 31A~Electrical, 31B~ hole; 32~ direction of current flow; 33 direction of leakage current (reverse current saturation current). 40, 41~ photoresist (resistance pattern); , 50~ broken Glass substrate; 51~tungsten film; 5 2~ with c-S i (microcrystalline lithium) film; 53~aluminum film; 61~substrate; 62~1st electrode layer; 63~η type"C-Si layer 65~p type #c - si layer; 66~2nd electrode layer; 131, 131A, 131B~π type a-Si region (n-type semiconductor region); 132~SiOx region (first insulator region) 151, 151A, 151B to p-type a-Si region (p-type semiconductor region); 152 to SiOx region (second insulator region); 621 to metal electrode region; 622 to first insulator region; 661 to transparent electrode region ; 16 201220514 662 ~ 2nd insulator area. 17