201140613 六、發明說明： c ^'明戶斤屬老^椅々貝超^ 3 發明領域 本發明係有關於一種適合於藉由焙燒貫通法所形成的 太陽電池電極用之導電性糊組成物。 【先前冬好；3 發明背景 舉例言之’一般的矽系太陽電池係具有以下構造，即： 透過n+層’於屬於P型多結晶半導體之矽基板之上面具備防 止反射膜及受光面電極，同時透過p+層，於下面具備裏面 電極（以下，在未區分該等時僅稱作「電極」。）者，又，構 成為經由電極，取出藉由受光而於半導體之卯接合中所產 生的電力。前述防止反射膜係用以維持充分之可見光透過 率並減低表面反射率而提高受光效率，且由Si3N4、Ti〇2、 Si02等之薄膜所構成。 由於前述防止反射臈之電阻值高，因此，會構成欲有 效地取出於半導體之pn接合中所產生的電力之阻礙。故， 太陽電池之受光面電極係例如藉由被稱作焙燒貫通之方法 來形成。於該電極形成方法中，舉例言之，在將前述防止 反射膜設置於n+層上之全面後，使用例如網版印刷法，於 該防止反射膜上以適當之形狀塗佈導電性糊，並施行焙燒 處理。前述導電性糊係例如將銀粉末、玻璃料(將玻璃原料 炼融並急冷後依需要進行粉碎的薄片狀或粉末狀之玻璃碎 片）、有機質載體及有機溶劑作為主成分，且於焙燒過程 201140613 _ ’由於該導電性糊中的玻璃成分會弄破防止反射膜因 此’藉由導電性糊巾的㈣成讀n+層，形成歐姆接觸(例 如參照專利文獻1。）。若藉由前述電極形成方法則相較 於局部地除去防止反射膜而於該除去部分形成電極者’步 驟會變得簡單，且亦不會產生除去部分與電極形成位置之 錯位問題。 於此種太陽電池之受光面電極形成中，以提升培燒貫 通性而改善歐姆接觸，進而提高曲線肝㈣或能量轉換 效率等為目的，迄今揭示有各種提案。舉例言之，包括： 藉由於導電性糊中添加P、V、Bi等之5族元素，促進玻璃及 銀相對於防止反射膜之氧化還原作用，並提升培燒貫通性 者(例如參照前述專利文獻卜）。又，包括：於導電性糊中 、氣化物、，臭化物或氣化物，藉此，該等添加物係輔助 破場及銀弄破防止反射敎仙岐善歐賴觸者（例如 參照專利文獻2。）。 又包括.在由SisN4或Si〇2等所構成的防止反射膜上 燒固電極材料，且該電極材料係含#Ag粉末、有機載體、 ^場料及 Ti、Bi、Co、Zn、Zr、Fe、CrU》、1H 奴取传安定之歐姆接觸及_接著強度者（例如參照專利 2獻3 ° )。Τι、Βι等係相對於Agl〇〇重量份而宜為〇 〇5重量 伤至5重量份。雖未顯示取得前述效果的理由，然而，若含 i Bi等，則於電極材料之焙燒過程中，該等會熔解於 破續中’其後，電極材料係作用於防止反射膜，因此，相 心於未含有Ti、Bi等者，與防止反㈣之反應會安定化。 201140613 又’揭示有一種厚膜導電性組成物，且該厚膜導電性 組成物係將Ag粉末、粒徑為7(11111)至ι〇〇(ηιη)之ZnO及軟化 點為300( C)至600(。〇之範圍内的玻璃料分散於有機溶劑 中（例如參照專利文獻4。）。該厚膜導電性組成物係用以形 成太陽電池之受光面電極，且藉由添加Zn，改善導電性與 焊料接著性。又，在相同之目的下，亦揭示有—種使用Μη02 以取代ZnO者(例如參照專利文獻5。）。 先行技術文獻 專利文獻 〔專利文獻1〕日本專利公開公報特公平03-046985號 公報 〔專利文獻2〕日本專利第3707715號公報 〔專利文獻3〕特開2001-313400號公報 〔專利文獻4〕特開2006-302890號公報 〔專利文獻5〕特開2006_3〇2891號公報 〔專利文獻6〕特開昭58-096777號公報 〔專利文獻7〕特開2008-192921號公報 〔專利文獻8〕特開2〇05-347276號公報 〔專利文獻9〕特開2008-226816號公報 〔專利文獻10〕特開平08-097417號公報 〔專利文獻11〕特開2〇〇4-323910號公報 【發明内容】 發明概要 發明欲解決之課題 201140613 不過’在製造太陽電池棋組時，會包括接著單元與焊 料之步驟，而减組之可靠財面，«的是無論在剛接 著後或經久後皆可確保其接著強度。於前述專利文獻卜專 利文獻2情揭不之導電性糊係焊料之接著強度仍低，且宜 進一步地提兩接著強度。另 心…^祖 另—方面，於前述專利文獻3至專 利文獻5中所揭示之導電性糊中，㈣會 而’由於接觸電阻增大，因此會有轉換效^；:、 又，於太陽電池中’受光面電極係阻擔入射之太: ^而=域電池單元之能量會依照該受 面積而減少。對此，目前掂_ 變細而使受光面積增大者，線寬 之線寬作成則㈣以下1而，線 ^知13〇(㈣ 好之歐姆翻，且接觸 、19 _以取得良 降低，因此，轉換效率反進^起電流密度之 焊料接著強度之導電I::降 因應細線化之要求是更加困難的。‘ ’因此，要 因此又所==係由於少數栽子之再結合速度高， 斤生成的電子與電騎再結合 =變成熱，熱損失)。特別是由於半== ’種如下所述者，即：蕤一 薄片電阻而使η層變薄，同時降低:質濃電池基板之 擴散距離變大，抑顧 & 〇數載子之 長的太陽… 成的熱損失，並使短波 %先有助於發電（被稱作淺結發射極或薄接面單 6 201140613 元）。藉此’電流會增大而可提高轉換效率。然而，若表面 附近之雜質濃度降低，則歐姆接觸會變差，因此，接觸電 阻會提高而電流密度降低，且轉換效率反而降低。故，由 於業已改善焊料接著強度之導電性糊原本就有接觸電阻高 的問題，因此’要應用在如前述淺結發射極是更加困難的。 本發明係以前述情形為背景而完成，其目的在提供一 種可提高焊料接著強度且無損電氣特性之太陽電池用導電 性糊組成物。 用以欲解決課題之手段 為了達成前述目的，本發明之要旨在於：一種太陽電 池用導電性糊組成物，係含有導電性銀粉末、玻璃料及有 機介質，並用以形成矽系太陽電池之電極者，又，含有平 均板為10(nm)至l〇〇(nm)範圍内的抓及抓〇之至少一者。 發明效果 若依此作成，則由於導電性糊組成物係含有微粉之Ni 或ΝιΟ(以下稱作Νι等），因此，若於n型矽基板上設置受光 面電極等之導體膜’則該導體膜會具有冑焊料接著強度。 由於在石夕基板與導體膜中的銀之間適當地形成導電通 道，因此’絲板與導體膜之接觸電阻會降低，並具有盘 未添加轉時同等以上之電氣特性，&，細線化會變得容 易’且於應用在受光面電極時，可加大受光面積，因此， 可取得同等以上之光電轉換效率。故，若藉由本發明，則 I製得-魏提高焊料⑽強度且無損電氣特性之太陽電 '用導電性糊組成物，此種導電性糊組成物係適合於太陽 201140613 電池之受光面電極。 另’若於形成自未含有Ni之導電性糊的電極進行焊 接’則電極中的Ag會與焊料中的Sn熔解而生成金屬化合 物，藉此，於電極與焊料之界面產生應力，故，一般認為 此會造成焊料接著強度之降低。若使用含有Ni之導電性 糊，則由於Ni與Sn之反應速度低，因此會抑制八§與如之反 應，故，可抑制金屬化合物之生成，且一般認為會進而改 善接著強度。又，若Ni及NiO之平均粒徑小，則由 於谷易凝結，因此分散性亦差，且處理困難。又，若平均 粒徑大於100(nm) ’則相較於未添加時，光電轉換效率反而 降低。該等平均粒徑係例如自使用SEM影像而藉由目視所 測定之值來算出。另，於導電性糊中，亦可含有其他之Ni 化合物，例如NiC等。 附帶一提’迄今進行的是在太陽電池用導電性糊中添 加Ni ’舉例言之，於前述專利文獻6中揭示有一種在太陽電 池之受光面側之電極使用含有5(wt%)至丨〇(wt%)之Ni的Ag 糊者。於該專利文獻6中，Ni之粒徑並未作任何揭示，且亦 未揭示使用微粉者或利用其之效果。另，Ni添加量係明顯 變多’然而，由於在專利文獻6中所揭示之太陽電池為低輸 出’因此’ 一般認為因大量添加Ni所造成的效率降低並不 成問題。故，即便使用此種Ag糊，由於Ni過多，因此，接 觸電阻反而增大，且導電性降低，進而降低FF值。 又’於前述專利文獻7中揭示有一種在形成太陽電池之 裏面A1電極時，使用將Ag粉末、平均粒徑Ο.Ι(μηι)至Ι.Ο(μηι) 201140613 之帅末、玻補、載體，作成相對於續末則5(wt%) 至2.〇(Wt%)之比例含有Ni粉末之組成的導電性糊者。於該 技術中，Ni含量亦明顯變多，“，細加人&，則會構 成受體，因此1在由η層所構柄表面側❹此種糊，則 會牽涉到效率降低，然而’ ^裏面側，騎使从變多亦 無妨。若使用較大粒徑之Ni，則為了享受所期待之效果， 必須含量會變多，然而，使用在裏面側咖不成問題。 又’於前述專散獻8t揭示有—種導電性糊，且該導 電性糊係適合用以燒貫通法，形成添加有仙等之氣 化物或演化物的太陽電池之受光面電極。該技術係藉由添 加Ni氟化物“提高純貫紐，並降低接觸電阻，且為 了不讓容Μ化之Ni作成氧化膜，讀化物錢化物來添 加。提升接著強度者並未作任何考慮，且未特別考慮见化 合物之粒徑或添加量。 又，應用在不同構造之太陽電池用電極，於前述專利 文獻9中揭示有一種電極形成用組成物，且該電極形成用組 成物係金屬奈米粒子分散於分散介質，並含有選自於 PVP、PVP之共聚物、PVA、纖維素醚中的有機高分子。金 屬奈米粒子係含有一次粒子徑為1〇(nm)至5〇(nm)之銀奈米 粒子75(wt%)以上；及合計為0.02(wt%)至25(wt%)之金鎳 等。此係使用在具備透明導電膜的覆板型太陽電池之裏面 電極，目的係作成不讓透明導電膜與裏面電極之接合界面 形成空氣層等之空間等，提高接著強度者等則未作任何考慮。 又，藉由添加鎳而降低接觸電阻者包括··在製造如前 201140613 述專利文獻10所MMIS型場效電㈣時，細料成膜於 晶圓上而進行熱處理者。利用僅成膜於㈣露出面接合之 領域的高炫點金屬膜以自相容方式變化成高炫點金屬石夕化 物膜者，將雜質擴散領域與閘極之表面同時地以自相容方 式石夕化物化而低電阻化。錄係防止雜質朝妙中擴散，:般 認為會有助於良好之歐姆接觸。如前所述，於專利文獻6至 專利文獻10中所揭示之技術係分別使用抓或抑化合物藉 此，提咼導電性或降低接觸電阻，於藉由焙燒貫通法所^ 成的太陽電池之表面側電極中宜改善接著強度者，或是4 藉由添加Νι等而將其改善者等則未作任何揭示。 在此’較為理想的是前述Ni及Ni〇係相對於糊組成物全 體而含有〇.5(wt%)以下之比例。Ni等之添加量越多，越會 提南接者強度’然而’右過量添加，則生成自導電性糊之 導體膜本身之電阻值會增大，同時導體膜與單元之接觸電 阻會增大而有轉換效率降低之傾向。故，Ni等之添加量宜 停留在0.5(wt%)以下。依據本發明，由於使用極微細之Ni， 因此，即使添加量微量，亦可充分地取得焊料接著強度之 改善效果，故，依據前述理由，在可看出接著強度改善之 範圍，Ni添加量宜為較少者。 另’Ni粉及NiO粉之製造方法或物性並無特殊之限制， 舉例言之’可自各種市售品中使用適當者。又，舉例言之， 製造方法可列舉如以下方法，即：如前述專利文獻η所示， 藉由多元醇法’使鎳鹽粉體或鎳氫氧化物粉體懸浮，並將 其加熱而還原成鎳粉，且將所生成的凝結鎳粉進行粉碎處 201140613 理者。若藉由該方法，則可製得分散性優異之微粉鎳粉。 又，較為理想的是於前述太陽電池用導電性糊組成物 中’前述玻璃料係使用軟化點位於300(°C；)至6〇〇(^c)之範圍 内者。若依此作成，則可製得一種能適當地使用在以下情 形之太陽電池用導電性糊組成物，即：藉由焙燒貫通法弄 破防止反射膜，並於太陽電池形成受光面電極者。即，藉 由使用具有前述溫度範圍之軟化點的玻璃料，太陽電池用 導電性糊組成物係具有良好之焙燒貫通性，且即使於η層薄 的淺結發射極等中，亦難以產生因玻璃所造成的ρη接合之 破壞。若軟化點小於3〇〇(。〇 ’則糊組成物之浸蝕性會變得 過強’因此，容易破壞ρη接合，另一方面，若軟化點大於 600(°C)’則會難以對防止反射膜浸蝕而無法取得歐姆接觸。 又，較為理想的是前述玻璃料係平均粒徑（D5〇)為 〇·3(μπι)至3_0(μιη)之範圍内，且相對於糊全體而含有 至20(V〇1%)之範圍内之比例。若玻璃料之平均粒徑過小， 則電極之培燒時’熔合會過快’因此’電氣特性會降低，然 而，若為0·3(μιπ)以上，則可取得適度之熔合性故，可進 -步地提高電氣特性，由於難以產生凝結，因此，在 糊調製時可取得更加良狀分散性。1玻_之平均粒 徑明顯大於«性銀粉末之平均粒_，亦會降低粉末全 體之分散性n若為3._m)以下，則可取得更加良好 之分散性，且可取得破璃之進—步之_性。X，若玻璃 量為―m可進—步地提高防止反射膜之溶合 又，若玻璃量為 性，因此，可取得更力σ良好之歐姆接觸。 201140613 20(v〇l%)以下，則難以進一步地形成絕緣層，因此，可取 得更高之導m，為了取得更加良好之歐姆接觸，宜 同時滿足前述平均粒徑以及在糊中的比例。在糊中的比例 特別宜為2(vol%)至 l〇(vol°/0)。 另，前述玻璃料之平均粒徑係依據空氣透過法之值。 空氣透過法係以下方法，即：自相對於粉_^_0 空氣)之透過性，測定粉體之比表面積。構成該測定方法之 基礎者係顯示構成粉體層之全粒子之潤濕表面積與通過該 處的流體之流速及壓力降之關係的柯辰尼.卡曼 (Kozeny-Carmarm)之方程式，且於依照裝置來訂定之條件 下’測定相對於所填充的粉體層之流速與壓力降而求取試 料之比表面積。該方法係將所填充的粉體粒子之間隙當作 細孔’並求取在空氣流中構成阻力的粒子群之潤濕表面 積’通常係顯示小於藉由氣體吸附法所求取的比表面積之 值。自所求取的前述比表_及粒子密度，可算出業已假 設為球形粒子之平均粒徑。 又’較為理想的是前述導電性銀粉末係平均粒徑(D50) 為〇.3_)至3·〇(μηι)之範圍内。又，若銀粉末之平均粒徑為 3·0(μηι)以下’則可取得更加良好之分散性，因此，可取得 更南之導電性。又’若為〇3(μηι)以上，則可抑制凝結，並 取㈣加良好之分散性。另，由於小於ϋ 3(μηι)之銀粉末係 月顯问價目此’若由製造成本方面來看亦宜為〇鄭爪） 以上。又’右導電性銀粉末、玻璃料皆為平均粒徑3 〇(㈣ 以下’則具有即使在藉由細顧案印卿錢極時亦難以 12 201140613 產生/1 且塞之優點。 ^另，前述銀粉末並無特殊之限制，可為球狀或鱗片狀 等任何-種形狀之粉末。然:而，使用球狀粉時印刷性優異， 同時在塗佈膜中的銀粉末之填充率會提高，因此，相較於 使用鱗片狀等其他形狀之銀粉末者，生成自該塗佈膜的電 極之導電率會提高，故，可在業已確保必要之導電性之狀 態下將線寬進一步地變細，因此特別理想。 又，較為理想的是前述太陽電池導電性糊組成物在 25(C)-20(rpm)中的黏度為 150(Pa . s)至 250(Pa · s)之範圍 内’且黏度比（即，[在10(rpm)中的黏度]/[在10〇(rpm)中的 黏度])為3至8。藉由使用具有此種黏度特性之糊，在擠壓時 可適當地低黏度化而透過網眼，且於該透過後恢復成高黏 度而抑制印刷寬度之擴展，因此，可輕易地透過網篩而不 會產生阻塞等，在業已維持印刷性之狀態下輕易地製得細 線圖案。糊組成物之黏度更宜為160(pa . s)至200(Pa . s)之 範圍’且黏度比更宜為3.2至6_0之範圍。又，在設計線寬為 1〇〇(μιη)以下之細線化時，宜為黏度比4至6。 另’舉例言之，即使將線寬變細亦可維持截面積而將 膜厚加厚者，亦可為將印刷製版之乳劑厚度加厚者；提高 張力者；將線徑變細而擴大開口徑者等。然而，若將乳劑 厚度加厚，則脫版會變差，因此，無法取得印刷圖案形狀 之安定性。又，若提高張力或將線徑變細，則網眼容易拉 伸，因此會難以維持尺寸.形狀精度，同時會有印刷製版 之耐久性降低之問題。又’由於以寬幅設置，因此，無需 13 201140613 將膜厚加厚之匯流排亦會變厚，故，亦會有材料之浪費變 多之問題。 又，前述玻璃料並無特殊之限制，可使用有鉛玻璃及 無鉛玻璃中之任一者。舉例言之，可使用：以氧化物換算 而含有 PbO 為 46(mol%)至 57(m〇i〇/0)、B2〇3 為 1(m〇1%)至 7(mol%)' Si02為38(mol0/〇)至53(mol%)之範圍内之比例的有 叙j玻璃’以氧化物換舁而含有Li2〇為〇.6(mol%)至 18(mol°/。）' PbO為 20(mol%)至 65(mol%)、B2〇3 為 l(m〇i〇/0) 至18(mol%)、Si02為20(mol%)至65(mol%)之範圍内之比例 的含Li有鉛玻璃；以氧化物換算而含有別2〇3為10(111〇丨％)至 29(mol°/〇) ' ZnO為 15(mol%)至30(mol%)、Si02為 0(mol%)至 20(mol%)、B2〇3為 20(mol%)至33(mol%)、Li20、Na20、K2〇 之合計量為8(mol%)至21 (mol%)之範圍内之比例的無鉛玻 璃等。 於前述有鉛玻璃中，PbO係使玻璃之軟化點降低之成 分，且為用以能進行低溫焙燒所必須。為了取得良好之焙 燒貫通性，PbO宜為46(mol%)以上且為57(mol°/〇)以下。PbO 量更宜為49(mol%)以上，且更宜為54(mol%)以下。即，更 為理想的是49(mol%)至54(mol%)之範圍。 又，於前述有鉛玻璃中，B2〇3係形成玻璃氧化物（即， 作成玻璃之骨架之成分），且為用以降低玻璃之軟化點所必 須之成分。為了取得良好之焙燒貫通性，B2〇3宜為l(mol%) 以上且為7(mol%)以下。b2〇3量更宜為3(m〇l%)以上，且更 宜為5(mol°/〇)以下。即，更為理想的是3(mol%)至5(mol%) 201140613 之範圍。 又，於前述有鉛玻璃中，Si02係形成玻璃氧化物，且 為用以提高玻璃之耐化學性所必須之成分。為了取得良好 之焙燒貫通性，Si02宜為38(mol°/。）以上且為53(mol%)以 下。Si〇2量更宜為43(mol%)以上，且更宜為48(mol%)以下。 即，更為理想的是43(mol%)至48(mol%)之範圍。 又，前述有鉛玻璃可於無損其特性之範圍含有其他各 種玻璃構成成分或添加物。舉例言之，亦可含有AhZr、 Na、Li、Ca、Zn、Mg、K、Ti、Ba、Sr等。該等可含有例 如合計10(mol%)以下之範圍。 又，於前述含Li有鉛玻璃中，除了 PbO、B2〇3、Si〇2 外，LisO係必須成分。Li20係使玻璃之軟化點降低之成分， 且為了取得良好之焙燒貫通性，Li20宜為0.6(mol%)以上且 為18(mol°/〇)以下。若1^2〇小於〇.6(mol%)，則軟化點會變得 過高，進而使得對防止反射膜之浸蝕性容易變得不足。另 一方面’若大於18(mol%)，則浸触性會變得過強，因此’ 反而會有電氣特性降低之傾向。附帶一提，由於Li係促進 擴散’因此，一般對半導體而言係屬於雜質，且由於具有 使特性降低之傾向’因此，在半導體用途中宜加以避免。 特別是通常在Pb量多時，若含有Li，則浸蝕性會變得過強 而有控制困難之傾向。然而，在如前述太陽電池用途中， 會使用含有Li之玻璃而無法看出特性之降低，反而藉由適 量含有，可改善焙燒貫通性，且可看出特性之提升。Li係 施體元素’且亦可降低接觸電阻。又，藉由作成含有Li之 15 201140613 組成，可看出能取得良好焙燒貫通性之玻璃的組成範圍會 擴大。不過，於太陽電池用途中，若過量含有，則浸蝕性 亦會變得過強，並有電氣特性降低之傾向^ Li2〇量更宜為 6(mol%)以上，且更宜為12(mol%)以下。即，更為理想的是 6(mol%)至l2(m〇l%)之範圍，又’特別理想的是6(m〇i%)。 又，於前述含Li有鉛玻璃中，為了取得良好之焙燒貫 通性，PbO量宜為20(mol%)以上且為65(m〇1%)以下。若pb〇 量小於20(mol%)，則軟化點會變得過高，因此，會難以對 防止反射膜次触’進而難以取得良好之歐姆接觸。另一方 面’若大於65(mol%) ’則軟化點會變得過低，因此，浸蝕 性會變得過強而產生容易破壞pn接合等問題ePb〇量更宜為 22.4(mol%)以上，且更宜為5〇_8(mol%)以下。即，更為理押 的是22.4(mol。/。)至50_8(mol%)之範圍，又，特別理想的是 30(mol°/。)至40(mol%)。 又，於前述含Li有鉛玻璃中，為了取得良好之焙燒貫 通性’ B2〇3里宜為1 (mol%)以上且為18(m〇l%)以下。若β q 量小於l(mol°/。）’則軟化點會變得過高，因此，會難以對防 止反射膜浸蝕’進而難以取得良好之歐姆接觸，同時耐濕 性亦有降低之傾向。特別是在玻璃中含有]^之態樣中若 B2〇3未含有l(mol%)以上，則會明顯變得難以炫化。另—方 面，若大於18(mol%)，則軟化點會變得過低，因此，浸钮 性會變得過強而產生破壞pn接合等問題。B2〇3量更宜為 2.8(mol%)以上，且更宜為i2(m〇l%)以下。即，更為理想的 是2_8(mol%)至12(mol%)之範圍，又，特別理想的是6(m〇i%) 16 201140613 至 12(mol%)。 又，於前述含Li有鉛玻璃中，為了取得良好之焙燒貫 通性，Si〇2量宜為20(mol%)以上且為65(mol°/〇)以下。若Si02 量小於20(mol%)，則耐化學性不足，同時會有玻璃形成困 難之傾向’另一方面’若大於65(mol%)，則軟化點會變得 過高而難以對防止反射膜浸钱，進而會有難以取得良好之 歐姆接觸之傾向。Si02量更宜為27.0(mol%)以上，且更宜為 48.5(mol%)以下。即，更為理想的是27.0(mol%)至48.5(mol%) 之範圍，又，特別理想的是30(mol%)至35(mol%)。 又，前述含Li有鉛玻璃可於無損其特性之範圍含有其 他各種玻璃構成成分或添加物。舉例言之，亦可含有A1、 Zr、Na、Ca、Zn、Mg、K、Ti、Ba、Sr等。A1係用以取得 玻璃之安定性有效之成分，因此，雖然幾乎不會影響特性， 但是宜加以含有。該等可含有例如合計30(mol%)以下之範 圍。舉例言之，A1及Ti宜分別為6(mol%)以下，且更宜為 3(111〇1%)以下。又，211宜為30(111〇1%)以下，且更宜為15(111〇1%) 以下。藉由作成適量含有該等Al、Ti、Zn之組成，可提升 並聯電阻Rsh，進而提升開放電壓Voc及短路電流I%，因 此，可取得更高的電氣特性。 又，於前述無鉛玻璃中，為了取得良好之焙燒貫通性， B2〇3量宜為20(mol%)以上且為33(mol°/。）以下。若小於 20(mol%)，則會有軟化點過高之傾向，若大於33(m〇1%), 則會有太陽電池之電氣特性不足之傾向。B2〇3越少，軟^匕 點越會上升’另一方面，B2〇3越多’電氣特性越會降低(舉 17 201140613 例言之，於矽系太陽電池中，一般認為起因於與屬於基板 材料之Si的反應性提高者）’因此，其比例宜考慮所期望之 軟化點與電氣特性來決定，舉例言之，宜為30(m〇l%)以下。 又，於前述無船玻璃中，Bi2〇3係使玻璃之軟化點降低 之成分，且為了能進行低溫培燒，宜加以含有。若小於 10(mol%) ’則會有軟化點過高之傾向，若大於29(m〇1%)， 則會有太陽電池之電氣特性不足之傾向。為了取得儘量高 的電氣特性，Bi2〇3量宜為較少者，且更為理想的是停留在 20(mol%)以下。又，為了充分地降低軟化點，Bi2〇3量宜為 較多者，且宜為15(mol°/〇)以上。即，特別理想的*15(m〇1%) 至20(mol°/〇)之範圍。 又，於前述無鉛玻璃中’ Zn〇係降低玻璃之軟化點同 時提高财久性（即’長期可靠性)之成分，若小於15(—%)， 則軟化點會構成過咼之值，同時耐久性亦會不足。另一方 面，若大於30(mol%) ’則雖然與其他成分之平衡亦會有所 影響，然而，玻璃會變得容易結晶化^ Zn〇量越少，軟化 點越會上升，同時耐久性亦會降低，另一方面，Zn〇量越 多，越容易結晶化，因此，更宜為2〇(111〇丨％)以上，且更宜 為30(mol%)以下。即，特別理想的是2〇加〇1%)至3〇加〇1%) 之範圍。 又，於削述無鉛玻璃中，鹼金屬成分Li2〇、Na2〇 、K20 係使玻璃之軟化點降低之成分，若合計量小於8(则1%)，則 軟化點會構成過高之值，若纽21(mGl%)，社陽電池之 電氣特性會變得不足。衫屬成分量越少，軟化點越會上 18 201140613 升，另一方面，驗金屬成分量越多，電氣特性越會降低， 因此，更宜為10(mol%)以上，且更宜為20(mol%)以下。即， 特別理想的是10(mol%)至20(mol°/〇)之範圍。 又，於前述無鉛玻璃中，Si02係形成玻璃氧化物，且 於無鉛玻璃中，亦具有提升玻璃安定性之效果，因此，雖 然並非必須成分，但宜加以含有。然而，Si02越多，軟化 點越會上升，因此，必須停留在20(mol%)以下。為了取得 充分之安定性，更宜為4(mol%)以上，且為了將軟化點停留 在夠低的值’更宜為11 (mol%)以下。即，特別理想的是 4(mol%)至 ll(mol%)。 又，前述玻璃料可於前述組成範圍由各種可玻璃化之 原料來合成，舉例言之’可列舉如：氧化物、碳酸鹽、確 酸鹽等’又’舉例言之，可使用二氧化矽Si〇2作為Si源，使 用硼酸B2〇3作為B源，使用鉛丹Pt>3〇4作為pb源，使用氧化 銀作為Bi源，使用氧化鋅作為Zn源’使用碳酸链作為Li源， 使用碳酸鈉作為NWi，使用碳酸鉀作為K源。 又，於有錄玻璃、無錯玻璃中之任一者中，在除了主 要成分外並含有其他成分時，可使用該等之氧化物、氫氧 化物、碳酸鹽、硝酸鹽等。 又，如前所述，本發明之導電性糊組成物係於業已維 持電氣特性之狀態下改善接著強度，因此，細線化容易， 故，可適當地使用在藉由培燒貫通法取得歐姆接觸的石夕系 太陽電池之受光面電極。 圖式簡單說明 19 201140613 第1圖係I員示本發明之一實施例之電極用糊應用在受 光面電極之形成的太陽電池之截面構造模式圖。 第2圖係顯示第1圖之太陽電池之受光面電極圖案之一 例之圖。 【實施冷式3 較佳實施例之詳細說明 以下，參照圖式，詳細地說明本發明之—實施例。另， 於以下實施例中，圖式係適當地簡化或變形，各部之尺寸 比及形狀專未必會正確地描繪。 第1圖係以模式方式顯示本發明之導電性糊組成物之 一實施例的電極用糊使用在受光面電極2 〇之形成的石夕系太 陽電池10之截面構造圖。於第1圖中，太陽電池1〇具備：例 如屬於p型多結晶半導體之梦基板12 ;分別形成於其上下面 之η層14及p層16，形成於該η層14上之防止反射膜is及受 光面電極20 ;及形成於該ρ+層16上之裏面電極22 ^前述石夕 基板12之厚度尺寸係例如1〇〇(μηι)至200(μηι)。 前述η層14及ρ+層16係藉由於石夕基板12之上下面形成 雜質濃度高之層體來設置，且該高濃度層之厚度尺寸係11層 14例如為70(nm)至100(nm) ’ ρ+層16例如為500(nm)。於一般 的矽系太陽電池中，η層14係100(nm)至200(nm)，然而，於 本實施例中會比此更薄，並構成被稱作淺結發射極之構 造。另，包含於η層14之雜質係n型摻雜物，例如磷(p)，包 含於ρ+層16之雜質係Ρ型摻雜物，例如鋁(Α1)或硼(Β)。 又，前述防止反射膜18係例如由氮化矽Si3N4等所構成 20 201140613 之薄膜，且藉由以例如可見光波長的1/4之光學厚度，例如 80(nm)來設置’構成為1 〇(%)以下，例如2(%)之極低反射率。 又，前述受光面電極20係由例如一樣厚度尺寸之厚膜 導體所構成，且如第2圖所示，藉由以下平面形狀，設置於 受光面24之大略全面，即.構成具有多數條細線部之梳狀 者。前述厚膜導體係由將Ag作為導體成分且含有78(wt%) 至99(wt%)之範圍並以〇.5(wt%)以下之範圍含有Ni的厚膜 銀所構成’且厚膜導體中的玻璃成分係含Li有鉛玻璃，該 含Li有船玻璃係以業經氧化物換算之值，分別含有以下比 例’即：PbO為20(mol%)至65(mol%)之範圍内，例如 22.4(mol%) ; B2〇3為 l(mol%)至18(mol%)之範圍内，例如 9.0(mol%) ; Si02 為 20(mol%)至65(mol%)之範圍内，例如 35.6(mol%) ; Al2〇3為〇(mol%)至6(mol%)之範圍内，例如 3.0(mol%) ; Li20為0.6(mol0/〇)至 18(mol%)之範圍内，例如 12.0(mol%) ; Ti02 為0(mol%)至 6(mol%)之範圍内，例如 3.0(mol%) ; ZnO 為 0(mol%)至 30(mol%)之範圍内，例如 l5.0(mol%)。又，前述導體層之厚度尺寸係例如2〇(μιη)至 30(μιη)之範圍内，例如25(μηι)，且細線部各自之寬度尺寸 係例如80(μηι)至130(μηι)之範圍内，例如ΐ〇〇(μΓη)，並具有 夠尚的導電性。 又，前述Ni係含有若干量之NiC，然而，如後述製造步 驟中所示，由於形成導體層時之加熱處理係於氧化環境氣 體下進行，因此，一般認為導體膜中的Ni會構成Ni〇的可能 性高。 21 201140613 又，前述裏面電極22係由以下電極所構成即：全面 電極26 ’係形成為於p+層处略呈全面地塗佈將紹作為導 體成分之賴材料者；及帶狀電㈣，係形成為於該全面 電極26上呈帶狀地塗佈且由賴銀所構成者。該帶狀電極 28係為了能將導線等焊接於襄面電極加設置。 本實施例之太陽電池1〇係由於受光面電極2〇如前所述 般由厚膜銀所構成，且該厚膜銀係、含有屬於低電 阻之η型半 導體的導電性鋅氧化物，因此，儘管線寬變細至⑽(㈣， 亦可在與η層14fa1轉&好之歐姆接觸，並降低接觸電阻。 如前述受光面電極20係例如使用由導體粉末、玻璃 料、導電性鋅氧化物、載體及溶劑所構成的電極用糊，並 藉由熟知之焙燒貫通法來形成。以下，說明包括該受光面 電極形成的太陽電池10之製造方法之一例。 首先’製作玻璃料。在使用由前述含Li有敍玻璃所構 成之玻璃料時’分別準備碳酸鋰LifO3作為Li源、二氧化 矽Si〇2作為Si源、硼酸Βζ〇3作為b源、鉛丹pb3〇4作為pb源、 氧化紐八丨2〇3作為A1源、氧化鈦Ti〇2作為Ti源、氧化鋅211〇 作為Zn源，並秤量、調和成構成前述範圍内之適當組成。 將其投入坩堝中，並藉由因應組成的900(。〇至1200(。〇之 範圍内之溫度’溶融30分鐘至1小時，且藉由急冷而使其玻 璃化°使用行星研磨機或球磨機等適當之粉碎裝置，將該 玻璃進行粉碎。粉碎後之平均粒徑(D5〇)係例如0·3(μιη)至 3·〇(μιη)。 亦可取代前述玻璃料，使用有鉛玻璃或無鉛玻璃等， 22 201140613 且m述有鉛玻璃係以氧化物換算而含有以下範圍内之比 例，即：PbO 為 46(mol%)至 57(mol%)、B2〇3 為 l(mol%)至 7(mol%)、Si〇2為38(mol%)至53(mol%)，且前述無鉛玻璃係 以氧化物換算而含有以下範圍内之比例，即：Bi2〇3為 l〇(mol%)至29(mol%)、ZnO為 15(mol%)至30(mol%)、Si02 為 〇(mol%)至 2〇(mol%)、B2〇3 為 20(mol%)至 33(mol%)、 Li20、Na2〇、K20之合計量為 8(mol0/。)至21(mol%)。在使用 前述有鉛玻璃時，舉例言之，除了分別使用氧化鈉Na20作 為Na源、氧化鋰Li2〇作為Li源、碳酸鉀作為K源之外，作成 與製造前述含Li有鉛玻璃時相同而使其玻璃化。又，在使 用無鉛玻璃時，除了分別使用氧化鉍作為Bi源、NH4H2P〇4 作為P源、氧化鈣CaO作為Ca源、BaC03作為Ba源之外，作 成與前述含Li有鉛玻璃及前述有鉛玻璃相同而製造玻璃 料。表1係顯示在本實施例中所使用的玻璃組成。表1中， Νο.1、Νο.3、Νο·4係含Li有鉛玻璃，No.2係有鉛玻璃，Νο·5 係無鉛玻璃。於本實施例中，可使用該等中之任一者。 〔表1〕 〔玻璃組成〕 玻璃料 組成(mol%)201140613 VI. Description of the invention: c^'Minghu's old chair, mussels, superb 3 FIELD OF THE INVENTION The present invention relates to a conductive paste composition suitable for use in a solar cell electrode formed by a firing penetration method. [Previously good winter; 3 BACKGROUND OF THE INVENTION] A general lanthanide solar cell system has a structure in which an anti-reflection film and a light-receiving surface electrode are provided on a top surface of a ruthenium substrate belonging to a P-type polycrystalline semiconductor through an n+ layer. At the same time, the p+ layer is provided with a back electrode (hereinafter, simply referred to as an "electrode" when it is not distinguished), and is configured to take out an element generated by light reception and semiconductor bonding. electric power. The anti-reflection film is formed of a film of Si3N4, Ti2, SiO2 or the like for maintaining sufficient visible light transmittance and reducing surface reflectance to improve light-receiving efficiency. Since the above-mentioned anti-reflection 臈 has a high resistance value, it constitutes an obstacle to effectively extracting electric power generated in the pn junction of the semiconductor. Therefore, the light-receiving surface electrode of the solar cell is formed, for example, by a method called baking. In the electrode forming method, for example, after the anti-reflection film is provided on the n+ layer, the conductive paste is applied to the anti-reflection film in an appropriate shape by, for example, screen printing. The roasting treatment is carried out. The conductive paste is, for example, a silver powder, a glass frit (a flaky or powdery glass cullet which is pulverized after being cooled and quenched, an organic carrier and an organic solvent), and is used as a main component in the baking process 201140613 _ 'Because the glass component in the conductive paste breaks the antireflection film, the ohmic contact is formed by reading the n+ layer of the conductive paste (see, for example, Patent Document 1). According to the electrode forming method described above, the step of forming the electrode in the removed portion as compared with the partial removal of the antireflection film becomes simple, and the problem of the displacement of the removed portion and the electrode forming position does not occur. In the formation of the light-receiving surface electrode of such a solar cell, various proposals have been made so far for improving the ohmic contact, improving the curve liver (IV), energy conversion efficiency, and the like in order to improve the patency of the smear. For example, by adding a group 5 element of P, V, Bi or the like to the conductive paste, promoting the redox effect of the glass and the silver relative to the antireflection film, and improving the penetration of the burnt film (for example, refer to the aforementioned patent) Literature b). Further, it includes: in the conductive paste, the vapor, the odor, or the vapor, whereby the additives are used to assist in the breakage and the silver breakage to prevent the reflection of the 敎 岐 岐 欧 ( (see, for example, Patent Document 2) .). Also included. The electrode material is baked on the anti-reflection film composed of SisN4 or Si〇2, and the electrode material contains #Ag powder, organic carrier, field material, and Ti, Bi, Co, Zn, Zr, Fe, CrU. , 1H slave pass stability and ohmic contact and _ subsequent strength (for example, refer to patent 2 for 3 °). Τι, Βι, etc. are preferably 〇 重量 5 parts by weight to 5 parts by weight relative to the weight of Agl 。. Although the reason for obtaining the above effect is not shown, if i Bi or the like is contained, the electrode material is melted in the process of firing during the firing of the electrode material, and thereafter, the electrode material acts on the antireflection film, and therefore, the phase If the heart does not contain Ti, Bi, etc., the reaction with the anti-(4) will be stabilized. 201140613 Further discloses that there is a thick film conductive composition, and the thick film conductive composition is Ag powder, ZnO having a particle diameter of 7 (11111) to ι〇〇(ηιη) and a softening point of 300 (C) The glass frit in the range of 600% is dispersed in an organic solvent (for example, refer to Patent Document 4). The thick film conductive composition is used to form a light-receiving surface electrode of a solar cell, and is improved by adding Zn. Conductivity and solder adhesion. Further, for the same purpose, it is also known to use Μη02 instead of ZnO (for example, refer to Patent Document 5). PRIOR ART DOCUMENT Patent Document [Patent Document 1] Japanese Patent Publication No. [Patent Document 2] Japanese Patent Publication No. 3707715 (Patent Document 3) JP-A-2006-313400 (Patent Document 4) JP-A-2006-302890 (Patent Document 5) Japanese Laid-Open Patent Publication No. 2008-192921 (Patent Document No. 2). Bulletin No. -226816 [Patent Document 10] Japanese Laid-Open Patent Publication No. Hei. No. Hei. No. Hei. No. Hei. No. Hei. No. Hei. No. Hei. Then, the steps of the unit and the solder are reduced, and the reliability of the group is reduced, so that the bonding strength can be ensured immediately after the last or after the long-term. Then, the strength is still low, and it is preferable to further increase the strength of the two. In addition, in the conductive paste disclosed in the aforementioned Patent Documents 3 to 5, (4) may be due to increase in contact resistance. Large, so there will be conversion effect ^;:, in the solar cell, 'the light-receiving electrode system resists the incident too: ^ and = the energy of the domain cell will decrease according to the area. For this, _ If the light-receiving area is increased and the light-receiving area is increased, the line width of the line width is made to be (4) 1 or less, and the line is known to be 13 〇 ((4) Good ohmic turn, and contact, 19 _ to achieve a good reduction, therefore, the conversion efficiency is reversed ^ Current density The conductivity of the solder followed by the strength I:: The requirement for thinning is more difficult. ' Therefore, therefore, == because of the high recombination speed of a few plants, the electrons generated by the kilogram are combined with the electric rider = It becomes heat, heat loss.) Especially because half == 'species as follows: 蕤 a sheet resistance to make the η layer thin, and at the same time reduce: the diffusion distance of the dense battery substrate becomes larger, and the & The heat of the long sun of the carrier... and the short-wave% help to generate electricity first (called the shallow junction emitter or thin junction single 6 201140613 yuan). Thereby, the current is increased to increase the conversion efficiency. However, if the impurity concentration near the surface is lowered, the ohmic contact is deteriorated, so that the contact resistance is increased and the current density is lowered, and the conversion efficiency is rather lowered. Therefore, since the conductive paste which has improved the solder joint strength has a problem of high contact resistance, it is more difficult to apply it to the shallow junction emitter as described above. The present invention has been made in view of the foregoing circumstances, and an object thereof is to provide a conductive paste composition for a solar cell which can improve solder joint strength without impairing electrical characteristics. Means for Solving the Problems In order to achieve the above object, the present invention is directed to a conductive paste composition for a solar cell comprising a conductive silver powder, a glass frit, and an organic medium, and used to form an electrode of a lanthanide solar cell. And, in addition, at least one of the grasping and grasping of the average plate in the range of 10 (nm) to 1 。 (nm). According to the above, the conductive paste composition contains Ni or ΝιΟ (hereinafter referred to as Νι) of the fine powder. Therefore, if a conductor film such as a light-receiving surface electrode is provided on the n-type ruthenium substrate, the conductor is provided. The film will have a solder joint strength. Since the conductive path is appropriately formed between the stone substrate and the silver in the conductor film, the contact resistance between the wire plate and the conductor film is lowered, and the electrical characteristics of the disk are equal to or higher than when the disk is not added, and thinning is performed. It is easy to be used, and when applied to a light-receiving surface electrode, the light-receiving area can be increased. Therefore, the photoelectric conversion efficiency of the same or higher can be obtained. Therefore, according to the present invention, a conductive paste composition for solar electric power which enhances the strength of the solder (10) and which does not impair electrical properties is obtained, and such a conductive paste composition is suitable for the light-receiving surface electrode of the solar cell 201140613. In addition, if the electrode is formed by welding an electrode containing a conductive paste containing no Ni, Ag in the electrode melts with Sn in the solder to form a metal compound, whereby stress is generated at the interface between the electrode and the solder. This is believed to cause a decrease in the strength of the solder joint. When a conductive paste containing Ni is used, since the reaction speed of Ni and Sn is low, the reaction between the eight and the § is suppressed, so that the formation of the metal compound can be suppressed, and it is generally considered that the bonding strength is further improved. Further, when the average particle diameter of Ni and NiO is small, since the grains are easily coagulated, the dispersibility is also poor, and handling is difficult. Further, when the average particle diameter is larger than 100 (nm)', the photoelectric conversion efficiency is rather lowered as compared with the case where it is not added. The average particle diameter is calculated, for example, from a value measured by visual observation using an SEM image. Further, other conductive Ni compounds such as NiC may be contained in the conductive paste. Incidentally, the addition of Ni to the conductive paste for solar cells has been described. For example, the above-mentioned Patent Document 6 discloses that the electrode used on the light-receiving side of the solar cell contains 5 (wt%) to 丨. 〇 (wt%) of Ni's Ag paste. In this Patent Document 6, the particle size of Ni is not disclosed, and the effect of using the fine powder or using it is not disclosed. In addition, the amount of addition of Ni is significantly increased. However, since the solar cell disclosed in Patent Document 6 has a low output, it is considered that the decrease in efficiency due to the large addition of Ni is not a problem. Therefore, even if such an Ag paste is used, since the Ni is excessively large, the contact resistance is increased, the conductivity is lowered, and the FF value is further lowered. Further, in the aforementioned Patent Document 7, there is disclosed an Ag powder having an average particle diameter when forming an A1 electrode in a solar cell. Ι(μηι) to Ι. Ο(μηι) 201140613 The end of the handsome, glass-filled, carrier, made 5 (wt%) to 2. The ratio of 〇 (Wt%) contains a conductive paste of the composition of Ni powder. In this technique, the Ni content is also significantly increased, ", fine addition of people &, will constitute the receptor, so 1 in the η layer of the handle surface side of the paste, it will involve a reduction in efficiency, however ' ^ Inside the side, riding more than no change. If you use a larger particle size of Ni, in order to enjoy the desired effect, the content must be increased, however, the use of the inside side is not a problem. Dedicated to 8t reveals a kind of conductive paste, and the conductive paste is suitable for the method of burning through to form a light-receiving surface electrode of a solar cell to which a vapor or an evolution of a fairy or the like is added. The technique is added by adding The Ni fluoride "improves the pure bond and lowers the contact resistance, and is added in order to prevent the Ni which is formed into an oxide film from being formed. The improvement of the strength of the adhesion was not taken into consideration, and the particle size or addition amount of the compound was not particularly considered. Further, in the solar cell electrode of a different structure, the above-mentioned Patent Document 9 discloses a composition for forming an electrode, and the electrode-forming composition metal nanoparticles are dispersed in a dispersion medium and are selected from PVP. , PVP copolymer, PVA, organic polymer in cellulose ether. The metal nanoparticles have a silver nanoparticle having a primary particle diameter of 1 〇 (nm) to 5 Å (nm) of 75 (wt%) or more; and a total of 0. 02 (wt%) to 25 (wt%) of gold nickel and the like. In this case, the inside electrode of a cover-type solar cell having a transparent conductive film is used, and the purpose is to prevent a space such as an air layer from being formed at the joint interface between the transparent conductive film and the back electrode, and the adhesion strength is not considered. . Further, when the contact resistance is lowered by the addition of nickel, when the MMIS type field effect electric power (4) of the patent document 10 of the above-mentioned 201140613 is manufactured, the fine material is formed on the wafer and heat-treated. The self-consistent manner is adopted in the impurity diffusion field and the surface of the gate simultaneously by using a high-focus metal film formed only in the field of (4) exposed surface bonding to change into a high-point metallization film in a self-compatible manner. Shi Xi is chemicalized and has low resistance. The recording system prevents impurities from spreading to the other side: it is believed to contribute to good ohmic contact. As described above, the techniques disclosed in Patent Documents 6 to 10 respectively use a scratching or suppressing compound to improve the conductivity or reduce the contact resistance, and the solar cell is formed by the firing method. It is preferable to improve the adhesion strength in the surface side electrode, or to improve the adhesion by adding Νι or the like. Here, it is preferable that the Ni and Ni lanthanum are contained in the entire composition of the paste composition. 5 (wt%) ratio below. The more the addition amount of Ni and the like, the more the strength of the south receiver is added, but the right excess is added, the resistance value of the conductor film itself generated from the conductive paste increases, and the contact resistance between the conductor film and the unit increases. There is a tendency to reduce conversion efficiency. Therefore, the addition amount of Ni and the like should stay at 0. 5 (wt%) or less. According to the present invention, since extremely fine Ni is used, even if the amount of addition is small, the effect of improving the solder joint strength can be sufficiently obtained. Therefore, for the above reasons, it is possible to see that the amount of Ni added is preferably in the range of the subsequent strength improvement. For the lesser. The manufacturing method or physical properties of the 'Ni powder and NiO powder are not particularly limited. For example, 'the product can be used from various commercial products. Further, for example, the production method may be a method of suspending a nickel salt powder or a nickel hydroxide powder by a polyol method as shown in the aforementioned Patent Document η, and heating it to reduce it. Nickel powder is formed, and the generated condensed nickel powder is pulverized at 201140613. According to this method, fine powder nickel powder excellent in dispersibility can be obtained. Further, it is preferable that the above-mentioned glass frit uses a softening point in the range of 300 (° C;) to 6 〇〇 (^c) in the conductive paste composition for a solar cell. According to this, it is possible to obtain a conductive paste composition for a solar cell which can be suitably used in a case where the antireflection film is broken by a baking penetration method and a light receiving surface electrode is formed on a solar cell. In other words, by using a glass frit having a softening point in the above temperature range, the conductive paste composition for a solar cell has good baking penetration properties, and it is difficult to cause a cause even in a shallow junction emitter or the like which is thin in the η layer. The destruction of the ρη joint caused by the glass. If the softening point is less than 3 〇〇 (.〇', the etchability of the paste composition becomes too strong'. Therefore, it is easy to break the ρη joint. On the other hand, if the softening point is larger than 600 (°C), it is difficult to prevent The reflective film is etched to obtain an ohmic contact. Further, it is preferable that the average frit size (D5〇) of the glass frit is in the range of 〇·3 (μπι) to 3_0 (μιη), and is contained in the entire paste. The ratio of the range of 20 (V〇1%). If the average particle size of the glass frit is too small, the fusion will be too fast when the electrode is fired, so the electrical characteristics will decrease, however, if it is 0·3 (μιπ) Above, it is possible to obtain moderate fusion properties, and it is possible to further improve electrical characteristics, and it is difficult to cause condensation. Therefore, more favorable dispersibility can be obtained in paste preparation. The average particle diameter of 1 glass is significantly larger than « The average particle size of the silver powder will also reduce the dispersibility of the powder as a whole. _m) below, it can achieve better dispersion, and can achieve the progress of the glass. X, if the amount of glass is -m, the anti-reflection film can be improved in a stepwise manner. If the amount of glass is a property, an ohmic contact with a better force σ can be obtained. In the case of 201140613 20 (v〇l%) or less, it is difficult to further form the insulating layer, and therefore, a higher conductivity m can be obtained, and in order to obtain a more favorable ohmic contact, it is preferable to satisfy the aforementioned average particle diameter and the ratio in the paste. The proportion in the paste is particularly preferably from 2 (vol%) to l〇 (vol°/0). Further, the average particle diameter of the above glass frit is based on the value of the air permeation method. The air permeation method measures the specific surface area of the powder by the following method, that is, from the permeability with respect to the powder _^_0 air. The basis of the measurement method is the Kochenni showing the relationship between the wetted surface area of the whole particles constituting the powder layer and the flow velocity and pressure drop of the fluid passing therethrough. The equation of Kozeny-Carmarm, and determining the specific surface area of the sample relative to the flow rate and pressure drop of the filled powder layer, is determined under the conditions set by the apparatus. The method uses the gap between the filled powder particles as the pores 'and the wetted surface area of the particle group forming the resistance in the air flow' is generally smaller than the specific surface area determined by the gas adsorption method. value. From the above-mentioned ratio table and particle density, the average particle diameter of the spherical particles can be calculated. Further, it is preferable that the average particle diameter (D50) of the conductive silver powder is 〇. 3_) to 3·〇(μηι). Further, if the average particle diameter of the silver powder is 3·0 (μηι) or less, more excellent dispersibility can be obtained, and thus further souther conductivity can be obtained. Further, if it is 〇3 (μηι) or more, coagulation can be suppressed, and (4) good dispersibility can be obtained. In addition, since the silver powder of less than ϋ 3 (μηι) is the price of the moon, it is preferable to be more than 〇 Zheng claws in terms of manufacturing cost. In addition, the right-handed silver powder and glass frit have an average particle size of 3 〇 ((4) or less, which has the advantage of being able to produce /1 and plugging even if it is handled by the case.) The silver powder is not particularly limited, and may be a powder of any shape such as a spherical shape or a scaly shape. However, when the spherical powder is used, the printing property is excellent, and at the same time, the filling rate of the silver powder in the coating film is high. Therefore, the conductivity of the electrode formed from the coating film is improved as compared with the use of other shapes of silver powder such as a scaly shape, so that the line width can be further improved while ensuring the necessary conductivity. Further, it is particularly preferable that the solar cell conductive paste composition has a viscosity of 25 (C) - 20 (rpm) of 150 (Pa. s) to the range of 250 (Pa · s) and the viscosity ratio (i.e., [viscosity in 10 (rpm)] / [viscosity in 10 rpm) is 3 to 8. By using a paste having such a viscosity characteristic, it can be appropriately low-viscosity and can be transmitted through the mesh at the time of extrusion, and can be restored to a high viscosity after the transmission to suppress the expansion of the printing width, so that the mesh can be easily passed through the mesh. Without causing clogging, etc., a fine line pattern can be easily produced while maintaining printability. The viscosity of the paste composition is preferably 160 (pa. s) to 200 (Pa. s) range ' and the viscosity ratio is more preferably 3. Range of 2 to 6_0. Further, when designing a thin line having a line width of 1 〇〇 (μιη) or less, it is preferable to have a viscosity ratio of 4 to 6. In addition, for example, even if the line width is reduced, the cross-sectional area can be maintained and the film thickness is increased, and the thickness of the emulsion for printing and plate-making can be increased; the tension is increased; the wire diameter is thinned and enlarged. Calibers, etc. However, if the thickness of the emulsion is increased, the release is deteriorated, so that the stability of the shape of the printed pattern cannot be obtained. Moreover, if the tension is increased or the wire diameter is made thinner, the mesh is easily stretched, so that it is difficult to maintain the size. The shape accuracy and the durability of the printing plate are reduced. Moreover, since it is arranged in a wide width, it is not necessary to use 13 201140613 to thicken the bus bar which is thickened in film thickness, so there is also a problem that material waste is increased. Further, the glass frit is not particularly limited, and any of leaded glass and lead-free glass can be used. For example, it can be used: in terms of oxide, PbO is 46 (mol%) to 57 (m〇i〇/0), and B2〇3 is 1 (m〇1%) to 7 (mol%) 'Si02 The ratio of the range of 38 (mol0 / 〇) to 53 (mol%) of the J. J glass 'exchange with oxide and Li 2 〇 is 〇. 6 (mol%) to 18 (mol ° / .) 'PbO is 20 (mol%) to 65 (mol%), B2 〇 3 is l (m〇i 〇 / 0) to 18 (mol%), SiO 2 is Li-containing leaded glass in a ratio ranging from 20 (mol%) to 65 (mol%); in the conversion of oxide, other 2〇3 is 10 (111〇丨%) to 29 (mol°/〇) ZnO is 15 (mol%) to 30 (mol%), SiO 2 is 0 (mol%) to 20 (mol%), B2〇3 is 20 (mol%) to 33 (mol%), Li20, Na20, K2〇 The total amount of lead-free glass or the like in the range of 8 (mol%) to 21 (mol%). Among the lead-containing glasses described above, PbO is a component which lowers the softening point of the glass and is required for enabling low-temperature baking. In order to obtain good baking penetration, PbO is preferably 46 (mol%) or more and 57 (mol/min) or less. The amount of PbO is more preferably 49 (mol%) or more, and more preferably 54 (mol%) or less. That is, it is more desirable to be in the range of 49 (mol%) to 54 (mol%). Further, in the lead-containing glass, B2〇3 forms a glass oxide (i.e., a component which forms a skeleton of glass) and is a component necessary for lowering the softening point of the glass. In order to obtain good baking penetration, B2〇3 is preferably 1 (mol%) or more and 7 (mol%) or less. The amount of b2〇3 is more preferably 3 (m〇l%) or more, and is more preferably 5 (mol ° / 〇) or less. That is, it is more desirable to be in the range of 3 (mol%) to 5 (mol%) 201140613. Further, in the lead-containing glass, SiO 2 forms a glass oxide and is a component necessary for improving the chemical resistance of the glass. In order to obtain good baking penetration, SiO 2 is preferably 38 (mol Å / or more) and 53 (mol %) or less. The amount of Si〇2 is more preferably 43 (mol%) or more, and more preferably 48 (mol%) or less. That is, it is more preferably in the range of 43 (mol%) to 48 (mol%). Further, the lead-containing glass may contain various other glass constituents or additives in a range in which the properties are not impaired. For example, AhZr, Na, Li, Ca, Zn, Mg, K, Ti, Ba, Sr, and the like may also be contained. These may contain, for example, a total of 10 (mol%) or less. Further, in the Li-containing lead-containing glass, in addition to PbO, B2〇3, and Si〇2, LisO is an essential component. Li20 is a component which lowers the softening point of the glass, and in order to obtain good baking penetration, Li20 is preferably 0. 6 (mol%) or more and 18 (mol ° / 〇) or less. If 1^2〇 is less than 〇. When 6 (mol%), the softening point becomes too high, and the corrosion resistance of the antireflection film tends to be insufficient. On the other hand, if it is larger than 18 (mol%), the thixotropy becomes too strong, and as a result, electrical characteristics tend to decrease. Incidentally, since Li promotes diffusion, it is generally an impurity for a semiconductor and has a tendency to lower the characteristics. Therefore, it is preferable to avoid it in semiconductor applications. In particular, when Li is contained in a large amount, if Li is contained, the etching property tends to be too strong, and control tends to be difficult. However, in the solar cell application as described above, the glass containing Li is used, and the deterioration of the characteristics cannot be seen. On the contrary, the baking property can be improved by an appropriate amount, and the improvement in characteristics can be seen. Li is a donor element' and can also reduce contact resistance. Further, by forming the composition of Li 2011-1540613, it can be seen that the composition range of the glass which can achieve good baking penetration is expanded. However, in solar cell applications, if it is contained in excess, the etching property will become too strong, and the electrical properties will be lowered. The amount of Li2 is preferably 6 (mol%) or more, and more preferably 12 (mol). %)the following. Namely, it is more preferably in the range of 6 (mol%) to l2 (m〇l%), and it is particularly desirable to be 6 (m〇i%). Further, in the Li-containing lead-containing glass, in order to obtain good baking performance, the amount of PbO is preferably 20 (mol%) or more and 65 (m〇1%) or less. If the amount of pb 小于 is less than 20 (mol%), the softening point becomes too high, so that it is difficult to prevent the reflection film from being sub-touched, and it is difficult to obtain a good ohmic contact. On the other hand, if it is larger than 65 (mol%), the softening point becomes too low, so that the etching property becomes too strong and the problem of easily breaking the pn junction is more likely. 4 (mol%) or more, and more preferably 5 〇 8 (mol%) or less. That is, the more pledged is 22. 4 (mol. /.) to the range of 50_8 (mol%), and particularly preferably 30 (mol / /) to 40 (mol%). Further, in the Li-containing lead-containing glass, in order to obtain good baking performance, 'B2〇3' is preferably 1 (mol%) or more and 18 (m〇l%) or less. If the amount of β q is less than 1 (mol ° /.)', the softening point becomes too high, so that it is difficult to prevent the reflection film from being etched, and it is difficult to obtain a good ohmic contact, and the moisture resistance tends to be lowered. In particular, when B2〇3 does not contain l (mol%) or more in the glass containing the film, it becomes apparent that it is difficult to condense. On the other hand, if it is more than 18 (mol%), the softening point becomes too low, so that the buttoning property becomes too strong and the problem of pn bonding is broken. The amount of B2〇3 is more preferably 2. 8 (mol%) or more, and more preferably i2 (m〇l%) or less. Namely, it is more preferably in the range of 2_8 (mol%) to 12 (mol%), and particularly preferably 6 (m〇i%) 16 201140613 to 12 (mol%). Further, in the Li-containing lead-containing glass, in order to obtain good baking performance, the amount of Si〇2 is preferably 20 (mol%) or more and 65 (mol/?) or less. If the amount of SiO 2 is less than 20 (mol%), the chemical resistance is insufficient and the glass tends to be difficult to form. On the other hand, if it is more than 65 (mol%), the softening point becomes too high and it is difficult to prevent reflection. The film is immersed in money, and there is a tendency that it is difficult to obtain good ohmic contact. The amount of Si02 is more preferably 27. 0 (mol%) or more, and more preferably 48. 5 (mol%) or less. That is, more ideal is 27. 0 (mol%) to 48. The range of 5 (mol%), and particularly preferably 30 (mol%) to 35 (mol%). Further, the Li-containing lead-containing glass may contain various other glass constituents or additives in a range not impairing its characteristics. For example, it may contain A1, Zr, Na, Ca, Zn, Mg, K, Ti, Ba, Sr, and the like. A1 is a component that is effective for obtaining the stability of glass. Therefore, although it does not affect the characteristics, it should be contained. These may contain, for example, a total of 30 (mol%) or less. For example, A1 and Ti are preferably 6 (mol%) or less, and more preferably 3 (111〇1%) or less. Further, 211 is preferably 30 (111 〇 1%) or less, and more preferably 15 (111 〇 1%) or less. By forming an appropriate amount of the composition of the Al, Ti, and Zn, the parallel resistance Rsh can be increased, and the open voltage Voc and the short-circuit current I% can be increased, so that higher electrical characteristics can be obtained. Further, in the lead-free glass, in order to obtain good baking penetration, the amount of B2〇3 is preferably 20 (mol%) or more and 33 (mol%) or less. If it is less than 20 (mol%), the softening point tends to be too high, and if it is more than 33 (m〇1%), the electrical characteristics of the solar cell tend to be insufficient. The less B2〇3, the more the soft point will rise. On the other hand, the more B2〇3, the more the electrical characteristics will decrease. (17 201140613 For example, in solar cells, it is generally considered to be caused by The reactivity of Si of the substrate material is improved. Therefore, the ratio should be determined in consideration of the desired softening point and electrical characteristics. For example, it is preferably 30 (m〇l%) or less. Further, in the above-mentioned shipless glass, Bi2〇3 is a component which lowers the softening point of the glass, and is preferably contained in order to enable low-temperature burning. If it is less than 10 (mol%), the softening point tends to be too high, and if it is more than 29 (m〇1%), the electrical characteristics of the solar cell tend to be insufficient. In order to obtain the highest possible electrical characteristics, the amount of Bi2〇3 should be less, and it is more desirable to stay below 20 (mol%). Further, in order to sufficiently lower the softening point, the amount of Bi2〇3 is preferably a large amount, and is preferably 15 (mol/min) or more. That is, a particularly desirable range of *15 (m 〇 1%) to 20 (mol ° / 〇). Further, in the lead-free glass, the component which reduces the softening point of the glass while improving the softening point of the glass (that is, the 'long-term reliability') is less than 15 (-%), and the softening point constitutes a value of excessive enthalpy. Durability will also be insufficient. On the other hand, if it is more than 30 (mol%), the balance with other components will be affected. However, the glass will become easily crystallized. The smaller the amount of Zn, the higher the softening point and the durability. On the other hand, the more the amount of Zn is, the more easily it crystallizes. Therefore, it is more preferably 2 Å (111 〇丨%) or more, and more preferably 30 (mol%) or less. That is, it is particularly desirable to have a range of 2% plus 3% to 3% plus 1%). Further, in the lead-free glass, the alkali metal components Li2, Na2, and K20 are components which lower the softening point of the glass, and if the total amount is less than 8 (1%), the softening point is excessively high. If the New York 21 (mGl%), the electrical characteristics of the social Yang battery will become insufficient. The smaller the amount of the genus, the higher the softening point will be 18 201140613 liters. On the other hand, the more the metal component is, the lower the electrical characteristics are. Therefore, it is more preferably 10 (mol%) or more, and more preferably 20 (mol%) below. That is, it is particularly desirable to have a range of 10 (mol%) to 20 (mol/min). Further, in the lead-free glass, SiO 2 forms a glass oxide, and in the lead-free glass, it also has an effect of improving the glass stability. Therefore, although it is not an essential component, it is preferably contained. However, the more SiO 2 , the higher the softening point and therefore must stay below 20 (mol%). In order to obtain sufficient stability, it is more preferably 4 (mol%) or more, and it is more preferably 11 (mol%) or less in order to keep the softening point at a sufficiently low value. That is, it is particularly desirable that it is 4 (mol%) to ll (mol%). Further, the glass frit may be synthesized from various vitrifiable raw materials in the above composition range, and may be exemplified by, for example, oxides, carbonates, acid salts, etc., and by way of example, cerium oxide may be used. Si〇2 is used as the Si source, boric acid B2〇3 is used as the B source, lead dan Pt>3〇4 is used as the pb source, silver oxide is used as the Bi source, and zinc oxide is used as the Zn source. Sodium carbonate is used as NWi, and potassium carbonate is used as the K source. Further, in any of the recorded glass and the error-free glass, when other components are contained in addition to the main component, such oxides, hydroxides, carbonates, nitrates and the like can be used. Further, as described above, the conductive paste composition of the present invention improves the adhesion strength while maintaining the electrical properties. Therefore, the thinning is easy, and therefore, it is possible to suitably use the ohmic contact by the penetration method. The light-emitting surface electrode of the solar cell of the Shi Xi. Brief Description of the Drawings 19 201140613 Fig. 1 shows a cross-sectional structural view of a solar cell in which an electrode paste according to an embodiment of the present invention is applied to a light-receiving electrode. Fig. 2 is a view showing an example of the light-receiving surface electrode pattern of the solar cell of Fig. 1. [Detailed Description of the Preferred Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Further, in the following embodiments, the drawings are appropriately simplified or modified, and the dimensional ratios and shapes of the respective portions are not necessarily correctly depicted. Fig. 1 is a cross-sectional structural view showing the use of the electrode paste of the embodiment of the conductive paste composition of the present invention in which the electrode for the light-receiving surface electrode 2 is formed. In the first embodiment, the solar cell module 1 includes, for example, a dream substrate 12 belonging to a p-type polycrystalline semiconductor; an n-layer 14 and a p-layer 16 respectively formed on the upper and lower surfaces thereof, and an anti-reflection film formed on the n-layer 14 The is and the light-receiving electrode 20; and the inner electrode 22 formed on the p+ layer 16. The thickness of the stone substrate 12 is, for example, 1 μ (μηι) to 200 (μηι). The n-layer 14 and the p+ layer 16 are disposed by forming a layer body having a high impurity concentration on the upper and lower sides of the substrate, and the thickness of the high-concentration layer is 11 layers 14 of, for example, 70 (nm) to 100 ( Nm) The ρ+ layer 16 is, for example, 500 (nm). In a general lanthanide solar cell, the η layer 14 is 100 (nm) to 200 (nm), however, it is thinner than this in this embodiment, and constitutes a structure called a shallow junction emitter. Further, the impurity included in the n layer 14 is an n-type dopant such as phosphorus (p), and an impurity-based dopant of the p + layer 16, such as aluminum (Α1) or boron (Β). Further, the anti-reflection film 18 is made of, for example, tantalum nitride Si3N4 or the like, and is formed of a thin film of 20 201140613, and is set to be 1 〇 by an optical thickness of, for example, 1/4 of a wavelength of visible light, for example, 80 (nm). %) Below, for example, 2 (%) extremely low reflectance. Further, the light-receiving surface electrode 20 is formed of, for example, a thick film conductor having the same thickness, and as shown in Fig. 2, is provided on the light-receiving surface 24 in a substantially comprehensive manner by the following planar shape, that is, A comb having a plurality of thin line portions is formed. The foregoing thick film guiding system has Ag as a conductor component and contains a range of 78 (wt%) to 99 (wt%) and is 〇. 5 (wt%) or less of the range consisting of thick film silver of Ni' and the glass component of the thick film conductor contains Li lead glass, and the Li-containing ship glass system has the following values in terms of oxides, respectively The ratio 'is: PbO is in the range of 20 (mol%) to 65 (mol%), for example 22. 4 (mol%); B2〇3 is in the range of l (mol%) to 18 (mol%), for example 9. 0 (mol%); Si02 is in the range of 20 (mol%) to 65 (mol%), for example, 35. 6 (mol%); Al2〇3 is in the range of 〇 (mol%) to 6 (mol%), for example 3. 0 (mol%); Li20 is 0. Within the range of 6 (mol0/〇) to 18 (mol%), for example 12. 0 (mol%); Ti02 is in the range of 0 (mol%) to 6 (mol%), for example 3. 0 (mol%); ZnO is in the range of 0 (mol%) to 30 (mol%), for example, l5. 0 (mol%). Further, the thickness of the conductor layer is, for example, in the range of 2 μ (μιη) to 30 (μηη), for example, 25 (μηι), and the width dimension of each of the thin line portions is, for example, a range of 80 (μηι) to 130 (μηι). Inside, for example, ΐ〇〇 (μΓη), and has enough conductivity. Further, the Ni system contains a certain amount of NiC. However, as shown in the manufacturing step described later, since the heat treatment in forming the conductor layer is performed under an oxidizing atmosphere, it is considered that Ni in the conductor film constitutes Ni〇. The possibility is high. 21 201140613 Further, the back electrode 22 is composed of the following electrode: the full electrode 26' is formed by slightly coating the p+ layer as a material for the conductor component; and the strip electrode (four) It is formed in a strip shape on the entire electrode 26 and is made of Laiyin. The strip electrode 28 is provided for welding a lead wire or the like to the tantalum electrode. The solar cell 1 of the present embodiment is composed of thick film silver as described above for the light-receiving surface electrode 2, and the thick film is made of silver and contains conductive zinc oxide which is a low-resistance n-type semiconductor. Although the line width is reduced to (10) ((4), it is also possible to make an ohmic contact with the η layer 14fa1 and reduce the contact resistance. For example, the above-mentioned light-receiving surface electrode 20 is made of, for example, a conductor powder, a glass frit, and a conductive zinc. An electrode paste composed of an oxide, a carrier, and a solvent is formed by a well-known baking penetration method. Hereinafter, an example of a method of manufacturing the solar cell 10 including the light-receiving surface electrode will be described. First, a glass frit is produced. When a glass frit composed of the aforementioned Li-containing glass is used, 'Lithium carbonate LifO3 is separately prepared as a Li source, cerium oxide Si〇2 is used as a Si source, strontium borate 3 is used as a b source, and lead dan pb3〇4 is used as a pb source. Oxidized New Barley 2〇3 is used as the A1 source, titanium oxide Ti〇2 is used as the Ti source, and zinc oxide 211〇 is used as the Zn source, and is weighed and blended to form an appropriate composition within the above range. 9 by response 00 (. 〇 to 1200 (the temperature in the range of 〇 is melted for 30 minutes to 1 hour, and is vitrified by quenching). The glass is pulverized using a suitable pulverizing apparatus such as a planetary mill or a ball mill. The average particle diameter (D5〇) after the pulverization is, for example, 0·3 (μιη) to 3·〇 (μιη). Instead of the glass frit, a lead glass or a lead-free glass may be used, 22 201140613 and a lead glass is described. It is contained in the following range in terms of oxide, that is, PbO is 46 (mol%) to 57 (mol%), B2〇3 is 1 (mol%) to 7 (mol%), and Si〇2 is 38. (mol%) to 53 (mol%), and the lead-free glass contains a ratio in the following range in terms of oxide, that is, Bi2〇3 is from 10% by mol to 29% by mol, and ZnO is 15%. (mol%) to 30 (mol%), SiO 2 is 〇 (mol%) to 2 〇 (mol%), B2 〇 3 is 20 (mol%) to 33 (mol%), Li20, Na2 〇, K20 The amount is from 8 (mol0 /.) to 21 (mol%). When using the aforementioned leaded glass, for example, except that sodium oxide Na20 is used as the Na source, lithium oxide Li2 is used as the Li source, and potassium carbonate is used as the K source. In addition to making and manufacturing the aforementioned lead-containing glass containing Li In the case of the use of the lead-free glass, in addition to the use of yttrium oxide as the Bi source, NH4H2P〇4 as the P source, calcium oxide CaO as the Ca source, and BaC03 as the Ba source, The glass frit was produced in the same manner as the lead-containing glass containing Li and the lead-containing glass described above. Table 1 shows the composition of the glass used in the present example. In Table 1, Νο. 1, Νο. 3, Νο·4 series containing Li lead glass, No. 2 series have lead glass, Νο·5 series lead-free glass. In the present embodiment, any of these may be used. [Table 1] [Glass composition] Glass frit Composition (mol%)
No. PbO BoO, Si02 Al2〇3 Li20 Ti02 ZnO Bi203 9* 3 2* 9* 4 0 8 7 2·4·o o 2 5 5 4 6 0 4 3 5-30·2· 3 4 4 3 3 2 o o o - 6·6·17 o 2 另一方面，導體粉末係準備例如平均粒徑（D50)為 〇_3(μιη)至3.0(μηι)之範圍内的市售球狀銀粉末。藉由使用此 種平均粒徑夠小的銀粉末，可提高在塗佈膜中的銀粉末之 23 201140613 填充率，進而提高導體之導電率。又，前述載體係使有機 結合劑溶解於有機溶劑中而調製，有機溶劑係使用例如丁 基卡必醇醋酸酯，有機結合劑係使用例如乙基纖維素。載 體中的乙基纖維素之比例係例如15(wt%)。又，有別於載體 而添加的溶劑係例如丁基卡必醇醋酸酯。即，並不限於此， 亦可為與載體中所使用者相同之溶劑，該溶劑之添加目的 係調整糊之黏度。 又，另外準備微粉Ni。如前述專利文獻η中所揭示， 微粉Ni可利用多元醇法等來製造，然而，於本實施例中， 使用例如平均粒徑為10(nm)至91(nm)之三井金屬礦業（股） 製造之微粉鎳。 分別準備以上糊原料，並种量例如平均粒徑為1 ·6(μπι) 之球狀Ag粉末77(wt%)至88(wt%);選自於前述表1所示中的 破項料l(wt%)至10(wt%);前述微粉Ni〇.〇l(wt%)至 〇.5(wt%);載體4(wt%)至 14(wt%);溶劑2(wt%)至8(wt%)之 比例，且使用攪拌機等進行混合後，藉由例如三輥輥磨機 來進行分散處理，藉此，製得電極用糊。另，於本實施例 中，為了將試樣間之印刷性作成同等，調整為2〇(rpm)_25(〇c) 之黏度會構成160(Pa · 3)至180(1>3 . s),且印刷製版係使用 SlJS325、線徑23(μηι)、乳劑厚度2〇(μιη)之篩孔。 依前述作成而調製電極用糊，另一方面，舉例言之， 藉由熱擴散法或離子植入等熟知之方法，使雜質擴散或注 入適當之矽基板而形成前述11層14及{)+層16，藉此，製作前 述矽基板12。其次，藉由例如pE_CVD(電漿CVD)等適當之 24 201140613 方法，於此形成氮化矽薄膜，並設置前述防止反射膜18。 其次，藉由前述第2圖所示之圖案，將前述電極用糊網 版印刷於前述防止反射膜18上。於印刷時，將印刷條件設 定成網格線之焙燒後寬度尺寸會構成丨⑼仏叫。藉由例如 150(°C)將其乾燥’再於近紅外爐中以74〇(。〇至900(。〇之範 圍内之溫度施行培燒處理，藉此，於該培燒過程中，電極 用糊中的玻璃成分會將防止反射膜18溶化，且該電極用糊 會弄破防止反射膜18，因此，可取得電極用糊中的導體成 分，即，銀與η層14之電連接，且如前述第丨圖所示，可取 得矽基板12與受光面電極20之歐姆接觸。受光面電極加係 依此而形成。 另，前述裏面電極22可於前述步驟後形成，然而，亦 可與受光面電極则0#地騎錢而賴。在形成裏面電 極22時，會藉由網版印刷法等，將例如鋁糊塗佈於前述矽 基板12之裏面全面，並藉由施行錢處理，形成由紹厚膜 所構成的前述全面電極26。再者，使㈣版印刷法等，將 前述電極用糊呈帶狀地塗佈於該全面電極26之表面，並施 行焙燒處理，藉此，形成前述帶狀電極28。藉此，形成由 覆蓋裏面全面之全面電極26及於其表面之__部分設置成慨 狀之帶狀電極28所構成的裏面電極22，並製得前述太陽: 池Η)。於前述步驟中’在藉由同時培燒來製造時，會在我 光面電極20之焙燒前施行印刷處理。 又 將電極用糊之調合組成進行各種變更，並依據前述製 造步驟製造太陽電池1G，錢用市售之太陽模《測定其 25 201140613 輸出而評價曲線因子FF值，同時將端子焊接於受光面電極 20，並藉由拉伸試驗機測定剝離強度之結果(N〇丨至n〇 23) 係與各試樣之調合規格一併地與比較例(No.24至No.28)同 時地顯示於表2至表5中。於各表中，添加劑欄係分別顯示 所添加的Ni或Ni〇之粒徑、種類，即，Ni、NiO別、相對於 糊100重量份之添加量。又，玻璃No.欄係使用前述表丨所示 之玻料編號，顯示所使用的玻璃種類。玻璃添加量係全部 為3重量份。又，表中雖未顯示，然而，Ag$係全部將平均 粒徑1.6(nm)之球狀粉作成84重量份之添加量。又，有機成 分，即，載體量係全部作成13重量份。#，將Ag粉、玻璃 料及載體之α 量作成咖重量份。又表2至表4係依所著 眼要素個別地歸納實_，相互重複之實驗資料係附上同 一實施例Νο_而再度揭示。 〔表2〕 [Ni粒子徑〕 實施例 No.粒徑 Qim) 1 2 3 4 5 6 7 8 9 10 11 12 10 10 22 22 22 22 22 22 52 52 91 91 劑 ^^nnnnnnnnnnnn_ f0'020·10·010·020·060·10·20·50·020·10·020·1) ^N0No. PbO BoO, Si02 Al2〇3 Li20 Ti02 ZnO Bi203 9* 3 2* 9* 4 0 8 7 2·4·oo 2 5 5 4 6 0 4 3 5-30·2· 3 4 4 3 3 2 ooo - 6·6·17 o 2 On the other hand, the conductor powder is prepared, for example, from a commercially available spherical silver powder having an average particle diameter (D50) in the range of 〇_3 (μιη) to 3.0 (μηι). By using such a silver powder having an average particle diameter small enough, the filling rate of the silver powder in the coating film can be increased, thereby increasing the electrical conductivity of the conductor. Further, the carrier is prepared by dissolving an organic binder in an organic solvent, for example, butyl carbitol acetate is used as the organic solvent, and ethyl cellulose is used as the organic binder. The proportion of ethyl cellulose in the carrier is, for example, 15 (wt%). Further, a solvent added in addition to the carrier is, for example, butyl carbitol acetate. That is, it is not limited thereto, and may be the same solvent as the user of the carrier, and the purpose of the addition of the solvent is to adjust the viscosity of the paste. Further, fine powder Ni was additionally prepared. As disclosed in the aforementioned Patent Document η, the fine powder Ni can be produced by a polyol method or the like, however, in the present embodiment, for example, Mitsui Metals Mining Co., Ltd. having an average particle diameter of 10 (nm) to 91 (nm) is used. Made of micronized nickel. Preparing the above paste raw materials separately, and cultivating, for example, spherical Ag powder 77 (wt%) to 88 (wt%) having an average particle diameter of 1-6 (μm); and the breakage material selected from the above Table 1 l (wt%) to 10 (wt%); the aforementioned fine powder Ni〇.〇l (wt%) to 〇.5 (wt%); carrier 4 (wt%) to 14 (wt%); solvent 2 (wt% When the ratio is 8 (wt%) and the mixture is mixed using a stirrer or the like, the dispersion treatment is carried out by, for example, a three-roll mill, whereby an electrode paste is obtained. Further, in the present embodiment, in order to make the printability between the samples equal, the viscosity adjusted to 2 rpm _25 (〇c) may constitute 160 (Pa · 3) to 180 (1 > 3 . s). And the printing plate is a sieve hole of SlJS325, wire diameter 23 (μηι), and emulsion thickness of 2 〇 (μιη). The paste for electrode is prepared as described above. On the other hand, for example, a known method such as thermal diffusion or ion implantation is used to diffuse or implant an impurity into a suitable substrate to form the 11 layers 14 and {)+. The layer 16 is thereby formed into the above-described ruthenium substrate 12. Next, a tantalum nitride film is formed by a suitable method of, for example, pE_CVD (plasma CVD), and the anti-reflection film 18 is provided. Next, the electrode paste is screen printed on the anti-reflection film 18 by the pattern shown in Fig. 2 described above. At the time of printing, the printing conditions are set such that the width dimension after the firing of the grid lines constitutes a 丨 (9) squeak. It is dried by, for example, 150 (° C.) and then subjected to a calcination treatment in a near-infrared furnace at a temperature of 74 Torr (to 900 900 ( , , , , , , , , , , , 电极The glass component in the paste melts the anti-reflection film 18, and the electrode paste breaks the anti-reflection film 18, so that the conductor component in the electrode paste, that is, the electrical connection between the silver and the n layer 14 can be obtained. As shown in the above figure, the ohmic contact between the germanium substrate 12 and the light-receiving surface electrode 20 can be obtained. The light-receiving surface electrode is formed in this manner. Further, the back electrode 22 can be formed after the foregoing steps, however, When the inner electrode 22 is formed, for example, an aluminum paste is applied to the inside of the ruthenium substrate 12 by screen printing or the like, and the money is processed by the method. The total electrode 26 made of a thick film is formed. Further, the electrode paste is applied to the surface of the total electrode 26 in a strip shape by a (four) plate printing method or the like, and a baking treatment is performed. Forming the aforementioned strip electrode 28. Thereby, the formation is covered by The full-scale electrode 26 and the __ portion of the surface thereof are disposed as the inner electrode 22 formed by the strip-shaped electrode 28 of the generous shape, and the aforementioned sun: pool is prepared. In the foregoing step, 'by simultaneous burning At the time of manufacture, the printing process is performed before the baking of the smoothing electrode 20. The composition of the electrode paste is adjusted to various changes, and the solar cell 1G is manufactured according to the above-described manufacturing steps, and the commercially available solar mode "measures 25". 201140613 Output and evaluate the curve factor FF value, and solder the terminal to the light-receiving surface electrode 20, and measure the peel strength by the tensile tester (N〇丨 to n〇23) together with the blending specifications of each sample. The ground and the comparative examples (No. 24 to No. 28) are simultaneously shown in Tables 2 to 5. In each of the tables, the additive column shows the particle diameter and type of Ni or Ni〇 added, that is, Ni. And the amount of addition of NiO to 100 parts by weight of the paste. In addition, the glass No. column uses the glass number shown in the above table to display the type of glass used. The amount of glass added is all 3 parts by weight. Although not shown in the table, however, Ag$ is all The spherical powder having an average particle diameter of 1.6 (nm) was added in an amount of 84 parts by weight. Further, the organic component, that is, the amount of the carrier was all made into 13 parts by weight. #, the amount of Ag of Ag powder, glass frit and carrier was made into coffee. In addition, Tables 2 to 4 are individually summarized according to the elements of the eye, and the experimental data which are repeated with each other are attached to the same embodiment Νο_ and are revealed again. [Table 2] [Ni particle diameter] Example No. Particle size Qim) 1 2 3 4 5 6 7 8 9 10 11 12 10 10 22 22 22 22 22 22 52 52 91 91 Agent ^^nnnnnnnnnnnn_ f0'020·10·010·020·060·10·20·50· 020·10·020·1) ^N0
FF
定 判 FF 定 判 焊料接著強度(N) 剛接著後 經久後 判定 7 6 t^ 77666655 7 7 1 7 7 7 7 7 7 7 7 7The FF rule is determined. The solder joint strength (N) is determined immediately after the long-term determination. 7 6 t^ 77666655 7 7 1 7 7 7 7 7 7 7 7 7
良好良良良良好好好好好好I 常<〇k常常常常l I I良4^1 : Lll LF tr hr I «1^ 7^ CI^ J ΟΛ^ 2· X 2. λλ X X 2. 2. 2. 好 好好好好好 好-ok好好-okl良好好好 良常良良常常常常常良1^ fcr br"· :t "TTV* TTV- TTT·· TTV·. 26 201140613〔表3〕 〔Ni添加量〕Good good, good, good, good, good, I often <〇k often often l II good 4^1 : Lll LF tr hr I «1^ 7^ CI^ J ΟΛ^ 2· X 2. λλ XX 2. 2. 2 Good and good - ok well - okl good good good often good often often good 1^ fcr br"· :t "TTV* TTV- TTT·· TTV·. 26 201140613 [Table 3] [Ni addition 〕
〔表4〕〔Table 4〕
焊料接著強度(N) 剛接著後 經久後 判定 4.2 4.4 4.5 4.1 5.5 4.1 4.5 3.5 2.8 2.2 2.9 4.9 2.9 3.9 常 •二 好 良好好 非常良好 非常良好 非常良好 〔表5〕 〔比較例〕The solder joint strength (N) is judged after a long time. 4.2 4.4 4.5 4.1 5.5 4.1 4.5 3.5 2.8 2.2 2.9 4.9 2.9 3.9 Chang • 2 Good Good Good Very good Very good Very good Very good [Table 5] [Comparative example]
No.IT 25 26 27 28 添加劑 破場 添加量(份)No.IT 25 26 27 28 Additives Breaking the field Adding amount (parts)
22 195 19S 1.0 0.02 0.1 FF判定 判定 76 良好 76 良好 71 不可 73 不可 70 不可 焊料接著強度(N) 剛接著後 經久後 判定 3.6 3.3 4.0 3.8 3.7 0.9 不可 0.5 不可 3.1 非常良好 2.5 良好 2.4 良好 至表5中，FF判定攔係揭示所測定之FF 值(％)與好壞判定。, L 野概匈定係將小於75(%)作成「不可」，將75(/〇)以上作成「于」，特別是將FF值77(〇/〇)作成「非常 良好」。該好壞判定係遵财場上要求7邪)以上之職之 27 201140613 貫情。又，焊料接著強度欄係顯示在剛接著後及經久後分 別所測定之接著強度’同時將經久後之強度小於1(N)作成 「不可」’將1(N)至3(N)作成「良好」，將大於3(N)者作成「非 常良好」。經久後之接著強度係於焊接後一星期保管於 85(°C)之高溫槽後進行測定。太陽電池係要求在例如2〇年以 上之長期間可維持品質之耐久性，前述接著強度之評價係 用以判斷有無該耐久性之加速試驗。強度之判斷基準係作 成遵循市場上所要求之實情。 前述表2係將Ni粒子徑於l〇(nm)至91(nm)間進行各種 變更而評價對FF值及接著強度之影響。於说粒子徑為1〇(nm) 至91(11〇1)之實施例价.1至1^〇.〖2中，可取得75(%)至77(%)之 良好或非常良好之FF值，相對於此，於Ni粒子徑大到i95(mn) 之比較例Νο.27、Νο·28(參照表5)中，FF值係停留在73(%) 以下且不足。焊料接著強度在經久後皆為2.2(N)以上且良 好。又’於未添加Ni之比較例Νο·24、No.25(參照表5)中， 可取得76(%)之良好之FF值，然而，經久後之接著強度係明 顯低到0_9(N)。若對照該等，則可得知若添加Ni，則無論粒 瓜白可取付接著強度之提升效果，然而，若粒徑大於 l〇〇(nm) ’則FF值會降低而變得不足。又，依據將22(nm) 以下之N i粉添加0 · 1重量份以下之較少量的實施例n 〇. 1至 No.6(除了 n〇_2之外)之評價結果，可得知藉由微量添加微細 之Ni，可取得FF值提升之效果。一般認為取得此種FF值提 升效果者係由於所添加的Ni會在與單元之Si間形成鎳矽化 物，藉此，可取得並聯電阻Rsh之提升、漏電流減少、接觸 28 201140613 電阻減低等之效果。 又，前述表3係針對分別使用2種玻璃料丨、玻璃料2之 情形’將Νι添加量於〇.〇1重量份至〇5重量份之範圍進行各 種’炎更而》平彳貝對FF值及接著強度之影響。川粉係全部使用 平均粒徑為22(nm)者。在使用玻璃料丨時，於犯添加量為〇」 重量份以下之範圍’ FF值會高到77(%)而可取得非常良好之 結果，在使用玻璃料丨且添加Ni〇.2重量份至〇5重量份時， 以及在使用玻璃料2時，可取得FF值為75(%)以上之良好之 結果。又，焊料接著強度在使用玻璃料丨時，當奶添加量為 〇.〇6重量份以上時’可取得大於稱之非f良好之結果在 其他情形時亦可取得2(N)以上之良好之結果。另—方面， 於价添加量為U重量份之比較例N〇_26(參照表5)中，雖然 焊料接著強度會高到3.1(N)而可取得非f良好之結果，然 F值卻明顯低到71(%)且不足。依據該等評價結果，22 195 19S 1.0 0.02 0.1 FF judgment judging 76 good 76 good 71 not 73 not 70 no soldering strength (N) just after the long time after the judgment 3.6 3.3 4.0 3.8 3.7 0.9 not 0.5 not very good 3.1 good 2.4 good 2.4 good to table 5 In the middle, the FF decision block reveals the measured FF value (%) and the good or bad judgment. In the case of L, the Hungarian system is less than 75 (%) and is made "not", and 75 (/〇) or more is made to "Yes". In particular, the FF value of 77 (〇/〇) is made "very good". The judgment of good or bad is in accordance with the requirements of the bank for more than 7 evils. In addition, the solder strength column shows the subsequent strength measured immediately after and after the long-term, and the intensity after the long-term is less than 1 (N) is made "not possible" and 1 (N) to 3 (N) is made " Good, and those who are greater than 3 (N) are made "very good." The adhesive strength after long-term storage was measured in a high temperature bath at 85 (° C.) one week after the welding. The solar cell system is required to maintain the durability of the quality for a period of, for example, 2 years or more, and the evaluation of the subsequent strength is used to determine whether or not the durability test is performed. The basis for determining the strength is to follow the truth as required by the market. In the above Table 2, the influence of the Ni particle diameter between 10 Å (nm) and 91 (nm) was evaluated to evaluate the influence on the FF value and the subsequent strength. For example, the particle diameter is from 1 〇 (nm) to 91 (11 〇 1). The price of the example is from 1 to 1 〇. In 〖2, 75 (%) to 77 (%) can be obtained as good or very good FF. In contrast, in the comparative examples Νο.27 and Νο·28 (see Table 5) in which the Ni particle diameter is as large as i95 (mn), the FF value stays below 73 (%) and is insufficient. The solder joint strength was 2.2 (N) or more and was good after a long period of time. Further, in the comparative examples Νο.24 and No.25 (see Table 5) in which Ni was not added, a good FF value of 76 (%) was obtained, however, the post-elongation strength was significantly lower to 0_9 (N). . According to these comparisons, it is understood that when Ni is added, the effect of improving the strength can be obtained regardless of the whiteness of the granules. However, if the particle diameter is larger than l 〇〇 (nm) ′, the FF value is lowered and becomes insufficient. Further, according to the evaluation results of adding less than 0.1 parts by weight of the N i powder of 22 (nm) or less to a smaller amount of the examples n 〇. 1 to No. 6 (except n 〇 2), It is known that the effect of increasing the FF value can be obtained by adding a fine amount of Ni. It is considered that the effect of obtaining such an increase in FF value is due to the fact that Ni is added to form a nickel telluride between the Si and the Si, whereby the parallel resistance Rsh can be improved, the leakage current can be reduced, and the contact 28 201140613 can be reduced in resistance. effect. Further, in the above Table 3, in the case where two types of glass frit and frit 2 are used, respectively, the amount of Νι is added in the range of from 1 part by weight to 5 parts by weight of 玻璃. The effect of FF value and subsequent strength. All of the Sichuan powder systems used an average particle size of 22 (nm). When glass frit is used, the amount of addition is 〇"% by weight", the FF value is as high as 77 (%), and very good results can be obtained. In the case of glass frit, Ni 〇. 2 parts by weight is added. When the amount is 5 parts by weight and when the glass frit 2 is used, a good result of an FF value of 75 (%) or more can be obtained. In addition, when the amount of the milk is 〇. 〇 6 parts by weight or more when the amount of the milk is increased, the result of the solder can be more than 2 (N). The result. On the other hand, in Comparative Example N〇_26 (refer to Table 5) in which the amount of addition is U by weight, although the solder joint strength is as high as 3.1 (N), a non-f good result can be obtained, but the F value is Significantly as low as 71 (%) and insufficient. Based on the results of these evaluations,
Nl =加量越多’接著強度越高，然而，FF值會看出以0〇5 重量份為峰值而降低之傾向，若將添加量增加至〇.5重量 伤則會降低至容許下限值Μ㈤，若為重量份，則會 明顯降低。故，Ni添加量宜停留在〇5重量份以下。另，表 3中雖未顯示’然而，若添加量為2重量份以上，則可看出 接著強度之降低。__般認為此係由於若见添加 量，則會妨礙Ag之燒結之故。 成過 旦另，並無特殊之Ni添加量之下限值，即使為極少之添 加里’亦可取得提升接著強度之效果，然而，添加量越少， 越難以確保糊之均質性，因此，於前述評價中，並未評價 29 201140613 少於0_01重量份之情形。 又’於前述表3所示之評價結果中，在使用破壤料 a夺於0.06重量伤至〇丨重量份之範圍添加粒徑项⑽ 粉的版5、％种,值為77(%)而非常良好，且接著強二 亦可取付3·4(Ν)至3.5(n)之非常良好之結果。依據該等姓 果，玻璃宜使用玻璃料卜且Ni微粉之粒徑22(nm)是最為^ 想的，其添加量為G綱量份至(U«份之範討說是最 為理想的。 又’前述表4係歸納以下結果，即：在如前述般可取得 最為理想之結果的添加粒徑22(nm)2Ni微粉()」重量份之系 統中，使用玻璃料1至玻璃料5來進行評價者，又，使用⑽ 作為添加物以取代见之他.22、N。23亦一併進行評價。依 據該評價結果’在使用玻璃料卜玻璃料5時，可取得77(%) 之非常高之FF值’同時可取得35(N)之非常高之接著強 度，因此，相較於玻璃料2至玻璃料4，一般認為該等是較 為理想的。特別是在使用玻璃料5之實施例^^〇 21中，在經 久後可取得4·9(Ν)之極高之接著強度，於此次進行評價之範 圍中，其係最為理想的。又，針對Ni〇，僅實施使用玻璃料 1之評價’於〇.〇2重量份之添加中，可取得砰值為77(%)、 接著強度為2.9(N)之結果，於0.1重量份之添加中，可取得 FF值為77(%)、接著強度為3.9(N)之結果。於Ni〇之情形時， 一般認為亦宜將添加量作成0.1重量份。 另，若對照前述表2至表5之實施例ν〇·ι至No.23及比較 例1^〇.26至><1〇.28與比較例>^〇.24、1\1〇.25，則剛接著後之強 30 201140613 度音顯示較高之值，然而，於添加Ni之前者中，在經久後 亦具有高接著強度，相對於此，於未添加Ni之後者中，經 久後之強度會明顯降低。於任一者中皆可看出強度因老化 而降低之傾向，然而’若添加Ni，則可得知其降低程度會 明顯緩和。 依據前述評價結果，藉由於將Ag粉作為導體成分之電 極用糊中添加微細之Ni或NiO，若於n型矽基板12上設置受 光面電極20，則該受光面電極2〇會具有高焊料接著強度。 又，由於在矽基板12與受光面電極20中的Ag之間適當地形 成導電通道，因此，石夕基板12與受光面電極2〇之接觸電阻 會降低，並具有與未添加Ni等時同等以上之電氣特性，故， 細線化會變得容易，且可加大受光面積，因此，可取得同 等以上之光電轉換效率。故，可製得一種能提高焊料接著 強度且無損電氣特性並適合於太陽電池10之t光面電極2〇 的電極用糊。 以上，參照圖式詳細地說明本發明，然而，本發明亦 可進—步地於其他態樣中實施’且可於未脫離其主旨之範 圍施加各種變更。 舉例言之’於前述實施例中’防止反射膜18係由氮化 夕膜所構$ $而，其構成材料並無特殊之限制，同樣地 可使用由-般在太陽電池巾所使㈣二氧化鈦τ叫等盆他 之各種材料所構成者。 又，於實施例中，說明本發明係應用在石夕系太陽電池 1〇之情形，然'而’本發明係只要是與本實施例相同，電極 31 201140613 中的導電性鋅氧化物之能階會構成電極中的導體成分及基 板之能階之中間的大小關係者，則不限於矽系而可加以應 用，又，只要是可藉由焙燒貫通法形成受光面電極之太陽 電池，則應用對象之基板材料並無特殊之限制。 c圖式簡單說明3 第1圖係顯示本發明之一實施例之電極用糊應用在受 光面電極之形成的太陽電池之截面構造模式圖。 第2圖係顯示第1圖之太陽電池之受光面電極圖案之一 例之圖。 【主要元件符號說明】 10.. .太陽電池 12.. .碎基板 14·_·η層 16.··ρ+層 18.. .防止反射膜 20.. .受光面電極 22.. .襄面電極 24.. .受光面 26.. .全面電極 28.. .帶狀電極 32Nl = the more the amount, the higher the strength, however, the FF value will show a tendency to decrease with a peak of 0〇5 parts. If the amount is increased to 〇.5, the weight will decrease to the lower limit. The value Μ (five), if it is part by weight, will be significantly reduced. Therefore, the amount of Ni added should preferably stay below 5 parts by weight. Further, although not shown in Table 3, however, when the amount added is 2 parts by weight or more, the subsequent decrease in strength can be seen. __ It is thought that this is due to the addition of the amount, which will hinder the sintering of Ag. In addition, there is no special lower limit of the amount of addition of Ni, and even if it is added in a very small amount, the effect of improving the strength can be obtained. However, the smaller the amount of addition, the more difficult it is to ensure the homogeneity of the paste. In the foregoing evaluation, the case where 29 201140613 is less than 0_01 parts by weight is not evaluated. Further, in the evaluation results shown in the above Table 3, the version 5 and the % of the powder of the particle size (10) powder were added in a range of 0.06 by weight to the weight part by weight of the ground material a, and the value was 77 (%). Very good, and then strong two can also pay very good results from 3.4 (Ν) to 3.5 (n). According to the surname, the glass should be made of glass frit and the particle size of 22 (nm) of Ni fine powder is the most desirable, and the addition amount is G-class fraction to (U« part of the specification is most desirable. Further, in the above Table 4, the following results are summarized, that is, in the system of adding the particle size 22 (nm) 2Ni fine powder () by weight as the above-mentioned optimum result, the glass frit 1 to the frit 5 are used. For the evaluator, (10) was used as an additive instead of the other. 22 and N. 23 were also evaluated together. According to the evaluation result, when the glass frit 5 was used, 77 (%) was obtained. A very high FF value 'at the same time can achieve a very high bond strength of 35 (N), therefore, compared to frit 2 to frit 4, it is generally considered to be preferable. Especially in the use of frit 5 In the example ^^〇21, the extremely high bonding strength of 4·9 (Ν) can be obtained after a long period of time, and it is most desirable in the range of evaluation of this time. Further, for Ni〇, only the use is performed. Evaluation of frit 1 'In addition to 2 parts by weight of 〇. 可, the 砰 value is 77 (%), and the subsequent strength is 2.9 (N) As a result, in the addition of 0.1 part by weight, the result of the FF value of 77 (%) and the subsequent strength of 3.9 (N) can be obtained. In the case of Ni 〇, it is considered that the addition amount is preferably made 0.1 part by weight. Further, if the examples ν〇·ι to No. 23 and Comparative Example 1^〇.26 to >1〇.28 and Comparative Example> 1〇.25, just after the strong 30 201140613 treble shows a higher value, however, in the former before adding Ni, it also has a high bonding strength after the long-term, in contrast, after the addition of Ni After a long time, the strength will be significantly reduced. In either case, the tendency of the strength to decrease due to aging can be seen. However, if Ni is added, the degree of reduction can be significantly alleviated. When fine Ni or NiO is added to the electrode paste of the Ag powder as a conductor component, when the light-receiving surface electrode 20 is provided on the n-type ruthenium substrate 12, the light-receiving surface electrode 2〇 has a high solder joint strength. A conductive path is appropriately formed between the germanium substrate 12 and the Ag in the light-receiving surface electrode 20, and therefore, the stone substrate 12 The contact resistance of the light-receiving electrode 2〇 is lowered, and the electrical characteristics are equal to or higher than those of the case where no Ni or the like is added. Therefore, the thinning is facilitated and the light-receiving area can be increased. Therefore, the photoelectric conversion of the same or higher can be obtained. Therefore, an electrode paste which can improve solder joint strength and has no loss of electrical characteristics and is suitable for the t-plane electrode 2 of the solar cell 10 can be obtained. The present invention will be described in detail with reference to the drawings, however, the present invention It is also possible to carry out the steps in other aspects and to apply various changes without departing from the spirit of the invention. For example, in the foregoing embodiments, the anti-reflection film 18 is constructed of a nitride film. Further, the constituent material is not particularly limited, and a material composed of various materials such as titanium dioxide τ in the solar cell towel can be used in the same manner. Further, in the embodiment, the present invention is applied to the case of the Shih-tung solar cell, but the present invention is the same as the present embodiment, and the conductivity of the conductive zinc oxide in the electrode 31 201140613. The relationship between the conductor component in the electrode constituent electrode and the energy level of the substrate is not limited to the lanthanide system, and the solar cell can be used as long as it can form the light-receiving surface electrode by the firing-through method. The substrate material of the object is not particularly limited. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic cross-sectional structural view showing a solar cell in which an electrode paste according to an embodiment of the present invention is applied to a light-receiving electrode. Fig. 2 is a view showing an example of the light-receiving surface electrode pattern of the solar cell of Fig. 1. [Description of main component symbols] 10.. Solar cell 12.. Fragmented substrate 14·_·η layer 16.··ρ+ layer 18.. Anti-reflection film 20... Light-receiving electrode 22.. Surface electrode 24: light receiving surface 26.. full electrode 28.. strip electrode 32