JP2004031435A - Method for forming thin film, forming device, and manufacturing method of thin film photoelectric converter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device and a method for forming a film of high quality with high productivity in order to manufacture a photoelectric converter unit of polycrystalline thin films for a hybrid photoelectric converter module having satisfying solar cell characteristics. <P>SOLUTION: A plurality of bar electrodes are arranged in parallel in a single plane to form a discharge region, both sides of which are arranged with film-forming surfaces of a substrate, respectively. A plurality of such planar discharge regions are allowed to exist in one film-forming chamber. <P>COPYRIGHT: (C)2004,JPO

Description

【００７３】
【表２】

【００７４】
基板２の割れの原因は基板内の温度分布によるものであるが、バックプレート４４の材料として熱伝導が良く均熱性の高いカーボンを用いることで温度分布が改善される。
【００７５】
図１１（ａ）に示すように、製膜室内に、棒状電極２６による製膜空間と冷却板として機能するプレート４０による冷却空間を別々に設けることにより、製膜と冷却とを、真空を破ることなく繰り返すことが可能である。
【００７６】
また、図１１（ｃ）のように棒状電極２６と冷却板として機能するプレート４０とを交互に配置し、更に基板２を揺動する事により、高速製膜と基板温度の最適化を同時に計ることができる。プレート４０の機能としては、基板２の加熱、冷却、ガスの導入、排気が可能である。
【００７７】
特に、図１１（ｄ）のように１本の棒状電極２６の２本の直線部分の中央にプレート４０を配置した場合には、効果的に基板２の加熱、冷却、ガスの導入、排気ができる。
【００７８】
さらに、図１１（ｂ）のようにカート４１による搬送を２枚の基板２の単位で実施する方式の場合には、バックプレート４４の後ろに設けられるプレート４０により基板の加熱、冷却、ガスの導入、排気を行うことができる。
【００７９】
さらに、また、図１３（ｅ）のように搬送を１枚の基板２の単位で実施できる方式の装置にした場合には、棒状電極２６の導入を製膜装置の上下から交互に行うことができ、基板の面内でより均一に製膜可能となる。
【００８０】
以上の図１１（ａ）〜（ｅ）の方式を組み合わせて実施するとより効果的である。
【００８１】
製膜条件としては、基板の製膜面と棒状電極との間の距離３４、圧力、ガス流量、温度、電源周波数、放電パワー等がある。
【００８２】
本装置では、棒状電極２６等の比較的容易に物理的な力で変形する突起物がＰＣ室にあり、また、カート４１が移動するために、図６の様に、ＰＣ内の製膜領域の下部に、カート４１の移動及びメンテナンスのための空間を設ける必要がある。
【００８３】
本装置では、特に４０ＭＨｚ以下の周波数帯の高周波電力を使用する場合、反応ガスとしてＳｉＨ４系ガスとその１０倍以上の流量比のＨ２が必要となる。ＳｉＨ４系ガスと水素の粘性は大きく異なるため、棒状電極２６のような複雑な形状の流路を通じて上記混合ガスが流れる場合、個々のアンテナの僅かな寸法誤差や部品組み付けの状態によって、放電領域３８に到達するガスの実質的な混合比が変化する場合がある。このため、安定生産のためには一平面上に並んだ複数の棒状電極２６群の周辺に発生する各々の放電領域３８に対して、独立に反応ガスの流量制御を行う必要がある。具体的には、各々の放電領域３８に対して独立にマスフローコントローラが備えられていることが好ましい。
【００８４】
また、本装置においては棒状電極２６上にシリコン膜が堆積するため安定生産のためには、棒状電極２６やプレート４０等が、着脱が容易で、堆積膜の除去がしやすく、また製膜中の膜の脱落が発生しにくい構造となっていること重要である。具体的には、棒状電極２６やプレート４０等の材質はアルカリによるシリコンエッチング除去に耐え得るＳＵＳが望ましく、さらに製膜中の脱膜防止のため、表面がサンドブラスト等により微少な凹凸状になっていることが望ましい。
【００８５】
このような本装置を用いて光電変換ユニット４ｂを形成する場合には、以上のように、高速で大面積に多数枚同時に安定して製膜することができる。
【００８６】
すなわち、本発明によると、安価な非晶質シリコン／多結晶シリコンのタンデム型光電変換素子モジュール及びその製造方法が提供される。
【００８７】
本発明においては、棒状のアンテナ電極を有するＣＶＤ装置により、多結晶シリコンユニットセルの光電変換層を、高速で大面積に多数枚同時に安定して製膜することができる。
【００８８】
（実施例）
以下に示す手順で、図１、２に示す薄膜光電変換モジュール１を、図５〜１５の装置を用いて、表３、表４の条件で製膜し作製し、表５、表６の結果を得た。
【００８９】
まず、一方の主面に厚み８００ｎｍのＳｎＯ２膜３を有する９１０ｍｍ×９１０ｍｍ×５ｍｍｔのガラス基板を準備した。このガラス基板のヘイズ率は１３％、シート抵抗は１２Ω／□であった。
【００９０】
【表３】

【００９１】
【表４】

【００９２】
【表５】

【００９３】
【表６】

【００９４】
次に、ＹＡＧ　ＩＲパルスレーザを用いて基板２にＳｎＯ_２膜側からレーザビームを照射しスキャンすることにより、　ＳｎＯ_２膜３を複数の帯状パターンへと分割する幅４０μｍの分離溝２１を形成した。
【００９５】
この分離溝２１は、ＳｎＯ_２膜３を複数の帯状パターンに分割する。スクライブライン間の距離は８．９ｍｍであり、１００個の幅８．９ｍｍ長さ９０４ｍｍの帯状パターンが基板２の中央に形成される。
【００９６】
さらに、薄膜光電変換モジュール１の集積方向５０に平行な基板２の２つの辺から５ｍｍ内側においてレーザースクライブにより周辺絶縁溝１１を形成する。図３に示すように、この位置は、基板２を基板面保持板３６の上にセットした際、基板保持板３６の縁から約１ｍｍ基板２の内側の位置となる。この周辺絶縁溝１１は複数本で形成することにより絶縁性を確実にすることが好ましい。具体的には、第１周辺絶縁溝１１−１から内側に約０．７ｍｍの距離にレーザースクライブにより幅約１００μｍの第２周辺絶縁溝１１−２を形成し、さらに第２周辺絶縁溝１１−２から内側に約０．７ｍｍの距離にレーザースクライブにより幅約１００μｍの第３周辺絶縁溝１１−３を形成した。
【００９７】
この時、図４に示すように３本の周辺絶縁溝１１を挟んで約８ｍｍの間隔を空けてメガテスターの探針５１を当てて、２５０Ｖの電圧を印可し測定したときに０．５ＭΩ以上の絶縁性を得ることができた。これら第１〜第３の周辺絶縁溝１１−１〜３はいずれも分離溝２１の端部から基板２の内側に形成されており、ＳｎＯ_２膜３を周縁部とセル集積領域に絶縁する。
【００９８】
その後、洗浄及び乾燥を行い、さらに、プラズマＣＶＤ法により、　ＳｎＯ_２膜３上に膜厚２００ｎｍ〜３５０ｎｍの薄膜光電変換ユニット４ａを成膜した。なお、この光電変換ユニット４ａは、光電変換層としてノンドープの非晶質シリコン層を有しており、ｐ−ｉ−ｎ接合を形成している。光電変換ユニット４ａの製膜は、製膜室構成を図５（ｃ）に示す多室のインライン装置を用いて、その製膜室の側面図を図５（ａ）に示す装置を用いて１３．５６ＭＨｚの周波数のＲＦ電源から平行平板電極２５に電力を供給し２枚の基板２を同時に製膜した。
【００９９】
この光電変換ユニット４ａ製膜後に、ＳｎＯ_２膜３及び薄膜光電変換ユニット４ａを貫通して周辺絶縁溝１１を形成しても良い。
【０１００】
続いて、ＣＶＤ法により、薄膜光電変換ユニット４ａ上に薄膜光電変換ユニット４ｂを成膜した。なお、この光電変換ユニット４ｂは、光電変換層としてノンドープの多結晶シリコン層を有している。場合によっては、４ａの製膜終了後に引き続いて４ｂ層のｐ層又はｎ層を製膜し、以下の図６、図７に示す製膜装置では４ｂ層のｉ層から製膜する場合もある。
【０１０１】
続いて、ＹＡＧ　ＳＨＧパルスレーザを用いて基板２にガラス面側からレーザビームを照射しスキャンすることにより、これら薄膜光電変換ユニット４ａ及び薄膜光電変換ユニット４ｂのスクライブを行い、それらを複数の帯状パターンへと分割する幅６０μｍの接続溝２３を形成した。なお、接続溝２３と分離溝２４との中心間距離は１００μｍとした。
【０１０２】
その後、薄膜光電変換ユニット４ｂ上に、スパッタリング法により、ＺｎＯ膜（図示せず）及びＡｇ膜を順次製膜して裏面電極層５を形成した。次いで、ＹＡＧ　ＳＨＧパルスレーザを用いて基板１にガラス面側からレーザビームを照射しスキャンすることにより、薄膜光電変換ユニット４ａ、４ｂ、　及び裏面電極層５を複数の帯状パターンへと分割する幅６０μｍの分離溝２２を形成した。なお、分離溝２２と接続溝２３との中心間距離は１００μｍとした。
【０１０３】
続いて、ＹＡＧ　ＩＲパルスレーザを用いて基板２の周囲に沿ってレーザスキャンすることにより、ＳｎＯ_２膜３、薄膜光電変換ユニット４ａ、４ｂ、裏面電極層５に溝を形成し、さらに、　ＹＡＧ　ＳＨＧパルスレーザを用いて基板２の周囲に沿ってレーザスキャンすることにより、　薄膜光電変換ユニット４ａ、４ｂ、　及び裏面電極層５に溝を形成して発電領域を確定した。
【０１０４】
以上のようにして、それぞれ８．９ｍｍ×８９０ｍｍのサイズを有するハイブリッド型薄膜光電変換セル１０が、基板２の集積方向５０に１００段直列接続された構造を形成した。
【０１０５】
その後、基板２に一対の電極バスバー１２を取り付けた。さらに、基板２のハイブリッド型薄膜光電変換セル１０を形成した面に、封止樹脂層６としてＥＶＡシートを、その上に、有機保護層７として黒色のフッ素樹脂系シート（商標名：テドラー）を載せ、真空ラミネート法によりラミネートすることにより図１及び図２に示す薄膜光電変換モジュール１を得た。
【０１０６】
太陽電池の初期特性は、光源としてキセノンランプ及びハロゲンランプを用いた放射照度１００ｍＷ／ｃｍ^２、ＡＭ１．５のソーラーシュミレータを用いて出力特性を調べた。なお測定温度は２５℃とした。
【０１０７】
膜厚分布は、分離溝２２形成後でハイブリッド型薄膜光電変換セル１０を形成した面への封止樹脂層６及び有機保護層７のラミネートまでに、レーザ顕微鏡により分離溝２２の段差から薄膜光電変換ユニット４ａ、４ｂ、裏面電極層５の総計の膜厚を測定し、比較的均一で薄い薄膜光電変換ユニット４ａ及び裏面電極層５の膜厚を予め設定した値で一定とし、測定結果からこの一定値を引くことで、薄膜光電変換ユニット４ｂの膜厚とした。
【０１０８】
測定点は、基板の端から１、２５、５０、７５、１００本目の分離溝２２の、基板の端から５５、２５５、４５５、６５５、８５５ｍｍの位置とし、基板面内で２５箇所測定し（最大値−最小値）／（最大値＋最小値）×１００％を膜厚分布とした。
【０１０９】
各実施例の光電変換ユニット４ｂの製膜は、図６、図７に示す装置を用いて製膜した。比較例の光電変換ユニット４ｂの製膜は、図５に示す従来のＣＶＤ装置を用いて、光電変換ユニット４ｂのｐｉｎ層をｐ層、ｉ層、ｎ層を別々の製膜室で製膜した。
【０１１０】
比較例１では２５―１の構造の電極とした装置で、比較例２ではイクスターナル電極２５−２とした装置で、各々２枚の基板２に同時に製膜した。
【０１１１】
光電変換ユニット４ｂはｐ層、ｉ層、ｎ層の３層からなるが、各実施例ではガスを切り替えて導入することにより１室で各層を、６枚の基板２を同時に製膜した。
【０１１２】
実施例１〜１０では、アンテナ電極２６−１として直線部の長さが約１２０ｃｍで全長が約２４０ｃｍで１６ｍｍφステンレス管をブラスト処理したものを用いた。アンテナ電極２６−１の一方の端は同軸ケーブルで高周波電源に接続されており、他方の端は接地されている。この状態で、電源側と接地側が交互になるようにアンテナは配置されている。また棒状電極中心軸間距離３７は１６ｃｍとした。また、アンテナとして断面が扁平な楕円のものを用い、基板面に対して扁平な面を向けて配置し、更に、基板面に対しない曲率が大きな面に複数のガス吹き出し孔を設けた構造のアンテナ電極２６−１を用い、このガス吹き出し孔から製膜ガスであるシランと水素を製膜室に導入した。
【０１１３】
アンテナ電極２６−１に供給する高周波電力の位相は、分波器からのケーブル長さを調節することで、隣り合うアンテナ電極２６−１で互いに逆位相にした。分派器では、高周波アンプからの電力を均等に１６分割した。装置内のアンテナ電極２６−１への電力供給はＣＶＤ装置壁２８のガイシ付フランジ２９を介して行った。
【０１１４】
アンテナ電極２６−１の接地は、図１０に示すように、ＣＶＤ装置壁２８のガイシ付きフランジ２９を介して一旦装置外に導出してから、再びＣＶＤ装置壁２８に広表面積導体３０を介してねじ留めすることにより実施した。
【０１１５】
実施例１１では、長さが約１２０ｃｍで、６ｃｍφの石英管３２、１ｃｍφのステンレス管３１、５ｃｍφのステンレス管３３から構成されるマイクロ波電極２６−３を使用した装置で光電変換ユニット４ｂの製膜を行った。製膜ガスの製膜室への導入はＣＶＤ装置壁２８から行った。また棒状電極中心軸間距離３７は１６ｃｍとした。
【０１１６】
実施例で使用した装置は、図６、図７に示されるように基板加熱機構が無い装置で、放電による加熱のみで製膜を実施した。また、実施例では図１０（ａ）の室構成のインライン方式の装置を使用した。さらに、実施例での基板２の着脱及び搬送は図１２、図１３に示す方法を用いた。
【０１１７】
アンテナ電極２６−１へ供給する高周波には、高周波アンプにパルス信号を加えることにより１ｋＨｚでデューティー比１：１のＯＮ　ＯＦＦ変調を加えた。また、マイクロ波電極２６−３にも同様の変調を加えた。
【０１１８】
さらに、各実施例では基板２を棒状電極中心軸間距離３７の半分の８ｃｍを８秒で往復する揺動を実施した。
【０１１９】
製膜室（ＰＣ）の真空度が２×１０−３Ｐａ以下となる状態でカート４１の投入が可能な状態となる。基板２を６枚セットした状態でカート４１はＬＣへ導入される。カート４１が導入された状態でＬＣ内は１×１０−２Ｐａまで真空排気され、その後、ゲートバルブの開閉操作を経て、ＨＣにカート４１が移動する。
【０１２０】
カート４１が導入されたＨＣ内は１２０℃まで昇温され、次にＰＣにカート４１が移動する。ＰＣでの製膜条件として表３の条件を用いた。製膜終了後ＰＣ内は真空度が２×１０−３Ｐａ以下となるまで真空引きれ、カート４１はＵＣへ導入される。カート４１が導入されたＵＣ内に窒素ガス導入され、減圧下での強制対流伝熱により１００℃以下の基板温度とした後、大気開放され、カート４１は装置の外に搬出される。
【０１２１】
以上説明した方法で各条件６枚のモジュール１を製造し、太陽電池の初期特性の測定を行った。それら結果を、実施例は表５に、比較例は表６に示す。
【０１２２】
また、実施例９Ａのサンプルについては、断面ＴＥＭ（透過電子顕微鏡）観察、ラマン測定、ＳＩＭＳによる不純物分析を実施した。
【０１２３】
図１６は実施例９Ａから切り出したサンプルの断面ＴＥＭ像である。基板側から製膜方向に繊維状構造の多結晶シリコンが製膜されている。
【０１２４】
図１７は実施例９Ａ、比較例１、及び単結晶シリコンのラマンスペクトルである。実施例９Ａ及び比較例１のラマン測定用のサンプルは薄膜光電変換ユニット４ｂ製膜後、金属裏面電極層５の製膜を実施せず基板２から小面積に切り出したものである。ラマン測定の浸入深さは１μｍ程度なので、最上部の多結晶シリコン薄膜の結晶性がこの測定により評価できる。図１７（ａ）に示すように５２０ｃｍ^−１付近のシリコン結晶によるピークは比較例１に比べ、実施例９Ａの方が強い。図１７（ｂ）に示すように非晶質のＬＡ（３２０ｃｍ^−１）、ＬＯ（４２０ｃｍ^−１）、ＴＯ（４９０ｃｍ^−１）のピークは実施例９Ａに比べ、比較例１の方が強い。従って、実施例９Ａの方が非晶質成分が少なく結晶子サイズが大きいといえる。
【０１２５】
図１９は、上記のラマン測定と同様のサンプルでＳＩＭＳによる不純物分析を実施例９Ａについて実施した結果である。多結晶シリコン薄膜中の不純物量は窒素が１Ｅ＋１７、炭素が２Ｅ＋１８、酸素が５Ｅ＋１８ａｔｏｍｓ／ｃｃ程度であった。ガラス基板２上に表面が凹凸なＳｎＯ_２膜を製膜し、その上にシリコン薄膜を製膜しているので、ＳＩＭＳにより深さ方向に掘って分析した場合、界面部分でスペクトルの立ち上がり（立ち下がり）はどうしてもシャープにならずブロードになってしまう。
【０１２６】
製膜条件４イ（実施例９Ａの製膜条件）で単結晶シリコンウェハ上に多結晶シリコンを７００ｎｍ製膜し、ＩＲ（赤外吸収）測定とＴＤＳ（昇温脱理ガス）分析を実施した。
【０１２７】
図１８（ａ）、（ｂ）はこのＩＲ測定の結果である。水素化シリコン薄膜の赤外吸収スペクトルには、通常、Ｓｉ−Ｈ結合に対応する２０００ｃｍ^−１の吸収と、Ｓｉ−Ｈ_２結合に対応する２１００ｃｍ^−１の吸収とがあるが、この領域でピークは１つであった。
【０１２８】
ＴＤＳ測定では、水素イオンを打ち込んだ単結晶シリコンをリファレンスとして多結晶シリコン中の水素の絶対濃度を測定した。その結果、結合水素濃度は６．７％であった。
【０１２９】
各実施例では、比較例と比較し、基板１枚当たりの製膜時間が１／３〜１／１０倍と短縮され製膜装置の生産性が大幅に向上しているにも関わらず特性は同等以上の結果となった。
【０１３０】
また、各実施例では、放電周波数が高くなるにつれて、最適な水素／シランの流量比が小さくなっており、比較例に比べて基板１枚当たり少ない水素流量で高品質膜の製膜が可能であった。
【０１３１】
また、各実施例では、放電周波数が高くなるにつれて、最適な圧力は低くなり、最適な基板の製膜面と棒状電極との間の距離３４（Ｅ／Ｓ）は狭くなり、膜厚分布は大きくなるが、製膜速度は大きくなることが判った。
【０１３２】
棒状電極２６の長手方向と集積方向５０とが平行な場合と直角な場合とでは、表５の実施例９Ａと実施例９Ｂのように、平行になるように基板２がセットされた場合に高い出力が得られた。これは本製膜装置において発生する棒状電極２６からの距離に応じた透明導電膜ＳｎＯ_２膜３の分離溝２１との相互干渉が、平行な場合には防止され、面内の膜厚分布によるユニットセル１０毎の光電流の発生量のばらつきが小さくなるためである。。
【０１３３】
薄膜光電変換ユニット４ａ製膜前に周辺絶縁溝１１を形成した場合と、薄膜光電変換ユニット４ａ製膜後で光電変換ユニット４ａ製膜前に周辺絶縁溝１１を形成した場合と、製膜前には周囲絶縁溝１１を形成せず、全製膜終了後に周囲絶縁溝１１を形成した場合の膜厚分布等の比較を表３の条件３イ〜３ハに、太陽電池特性の比較を表５実施例７イ〜ロに示す。周囲絶縁溝の形成により膜厚分布が小さくなり、太陽電池特性が向上している。
【０１３４】
基板２の揺動の有無及びＯＮ　ＯＦＦ変調の有無の比較を表３の条件４イ〜４ニに示す。基板揺動と変調により膜厚分布が小さくなり、基板の割れの発生が防止されている。
【０１３５】
【発明の効果】
以上説明したように、本発明においては、棒状電極を具備するＣＶＤ装置による多結晶シリコンの製膜により、充分な太陽電池特性を有するハイブリッド型光電変換素子モジュールが高い生産性で製造可能である。
【０１３６】
すなわち、本発明によると、安価な非晶質シリコン／多結晶シリコンのハイブリッド型光電変換素子モジュール及びその製造方法が提供される。
【図面の簡単な説明】
【図１】ハイブリッド型薄膜光電変換モジュールを概略的に示す平面図。
【図２】ハイブリッド型薄膜光電変換モジュールを概略的に示す断面図。
【図３】周辺絶縁溝の形成位置を概略的に示す断面図。
【図４】周辺絶縁溝の絶縁性を測定する方法を示す平面図。
【図５】平行平板型電極を具備するＣＶＤ装置の概念図。
【図６】棒状電極を具備するＣＶＤ装置の概念図。
【図７】直線状の棒状電極を具備するＣＶＤ装置の概念図。
【図８】アンテナ電極への電力供給及び接地方法を示す概念図。
【図９】マイクロ波電極の構造を示す概念図。
【図１０】製膜装置の室構成を示す概念図。
【図１１】種々の態様の棒状電極を具備するＣＶＤ装置の概念図。
【図１２】カートの構造示す概念図。
【図１３】カートへの基板の脱着方法を示す概念図。
【図１４】カートの別な構造示す概念図。
【図１５】別な構造のカートへの基板の脱着方法を示す概念図。
【図１６】実施例９ＡのＴＥＭ像。
【図１７】実施例９Ａ、比較例１及び単結晶シリコンのラマンスペクトル。
【図１８】製膜条件４イで製膜したサンプルのＩＲスペクトル。
【図１９】実施例９ＡのＳＩＭＳデプスプロファイル。
【符号の説明】
１　薄膜光電変換モジュール
２　透明基板
３　透明前面電極層
４ａ，４ｂ　薄膜光電変換ユニット
５　金属裏面電極層
６　封止樹脂層
７　有機保護層
１０　薄膜光電変換セル
１１　周辺絶縁溝
１２　電極バスバー
１３　モジュール周辺領域
１４，２１，２２　分離溝
２３　接続溝
２５　平行平板電極
２６　棒状電極
２６−１　アンテナ電極
２６−２　ワイヤ電極
２６−３　マイクロ波電極
２７　基板加熱ヒータ
２８　ＣＶＤ装置壁
２９　Ｏリングまたはガイシ付フランジ
３０　広表面積導体
３１　導波管
３２　誘電体管
３３　導電体
３４　基板の製膜面と棒状電極との間の距離
３５　基板枠
３６　基板面保持板
３７　棒状電極中心軸間距離
３８　放電領域
３９　放電点火装置
４０　プレート
４１　カート
４２　レール
４３　蝶番
４４　バックプレート
４５　受け皿
４６　レールガイド
５０　集積方向
５１　メガテスターの探針[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a thin-film photoelectric conversion module, and in particular, to a hybrid amorphous silicon / polycrystalline silicon photoelectric conversion element module in which a polycrystalline silicon thin-film layer is formed by a CVD apparatus having a rod-shaped antenna-type electrode.
[0002]
[Prior art]
In general, when manufacturing a thin film photoelectric conversion module, the cost of a film forming apparatus depending on the film forming speed of the photoelectric conversion layer occupies a large proportion of the module manufacturing cost. In particular, in the case of a tandem type photoelectric conversion element module of amorphous silicon / polycrystalline silicon, not only the amorphous cell but also a polycrystalline cell requiring a photoelectric conversion layer thickness 5 to 10 times that of the amorphous cell is manufactured. It is necessary to form a film, and an apparatus for simultaneously and stably forming a large number of sheets on a large area at high speed is essential.
[0003]
In general, a thin-film photoelectric conversion module has a structure in which a plurality of thin-film photoelectric conversion cells are connected in series on a glass substrate. Each thin film photoelectric conversion cell is generally formed by sequentially forming and patterning a front transparent electrode layer, a thin film photoelectric conversion unit, and a metal back electrode layer on a glass substrate.
[0004]
Such a thin-film photoelectric conversion module is required to improve photoelectric conversion efficiency. The tandem structure has a structure in which a plurality of thin-film photoelectric conversion units having different absorption wavelength ranges are stacked between a front transparent electrode layer and a metal back electrode layer, and is known as a structure that makes it possible to more effectively use incident light. Has been.
[0005]
In the hybrid type structure, which is one kind of the tandem type structure, the crystallinity of the photoelectric conversion layer, which is the main part of the thin film photoelectric conversion unit, is different between the thin film photoelectric conversion units. For example, in a thin film photoelectric conversion module having a hybrid structure, an amorphous silicon layer having a wider band gap is used as a photoelectric conversion layer of the thin film photoelectric conversion unit on the light incident side (or the front side), and the thin film on the back side is used. As the photoelectric conversion layer of the photoelectric conversion unit, a polysilicon layer having a narrower band gap is used.
[0006]
[Problems to be solved by the invention]
The present invention has been made in view of the above problems, and has as its object to provide an inexpensive amorphous silicon / polycrystalline silicon tandem-type photoelectric conversion element module and a method of manufacturing the same.
[0007]
[Means for Solving the Problems]
In order to solve the above-mentioned problems, the present inventors have studied a polycrystalline silicon film formation using a CVD apparatus having a rod-shaped antenna electrode. As a result, they have found that a hybrid photoelectric conversion element module having sufficient solar cell characteristics can be manufactured with high productivity.
[0008]
That is, according to the present invention, an inexpensive amorphous silicon / polycrystalline silicon hybrid photoelectric conversion element module and a method of manufacturing the same are provided.
[0009]
The term “crystalline” used herein includes polycrystals and microcrystals. Further, the terms “polycrystalline” and “microcrystalline” also mean those that partially contain amorphous.
[0010]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in more detail with reference to the drawings. In each of the drawings, similar members are denoted by the same reference numerals, and redundant description will be omitted.
[0011]
FIG. 1 is a plan view schematically showing a hybrid photoelectric conversion element module according to an embodiment of the present invention.
[0012]
The hybrid photoelectric conversion element module shown in FIG. 1 will be described in more detail.
[0013]
As shown in FIG. 2, the thin-film photoelectric conversion cell 10 of the module 1 includes a transparent front electrode layer 3 on a transparent substrate 2, a first thin-film photoelectric conversion unit 4a having an amorphous photoelectric conversion layer, and a crystalline photoelectric conversion unit 4a. It has a structure in which a second thin-film photoelectric conversion unit 4b having a conversion layer, a metal back electrode layer 5, a sealing resin layer 6, and an organic protective layer 7 are sequentially laminated. That is, in the module 1, the light incident from the transparent substrate 2 side is photoelectrically converted by the photoelectric conversion units 4a and 4b forming the hybrid type structure.
[0014]
Next, each component of the hybrid photoelectric conversion element module will be described.
[0015]
As the transparent substrate 2, for example, a glass plate or a transparent resin film can be used. As a glass plate, a large-area plate can be obtained at low cost and has high transparency and insulation properties. ₂ , Na ₂ Float glass panes whose main surfaces are mainly O and CaO can be used.
[0016]
The transparent front electrode layer 3 is made of an ITO film, SnO ₂ It can be composed of a film or a transparent conductive oxide layer such as a ZnO film. In general, conductivity can be represented by sheet resistance (Ω / □). The transparent front electrode layer 3 may have a single-layer structure or a multilayer structure. The transparent front electrode layer 3 can be formed by a known vapor deposition method such as a vapor deposition method, a CVD method, or a sputtering method.
[0017]
It is preferable to form a surface texture structure including fine irregularities on the surface of the transparent front electrode layer 3. By forming such a texture structure on the surface of the transparent front electrode layer 3, the efficiency of light incidence on the photoelectric conversion unit 4 can be improved. The method of forming the surface texture structure is not particularly limited, and various known methods can be used. The degree of the surface texture of the transparent overall electrode layer 3 is represented by a haze ratio measured with a C light source in a state where the transparent front electrode layer 3 is formed on the transparent substrate 2 having both smooth main surfaces.
[0018]
The thin-film photoelectric conversion unit 4a includes an amorphous photoelectric conversion layer. For example, a p-type silicon-based semiconductor layer, a silicon-based photoelectric conversion layer, and an n-type silicon-based semiconductor layer are sequentially stacked from the transparent front electrode layer 3 side. Having a structure. Each of the p-type semiconductor layer, the amorphous photoelectric conversion layer, and the n-type semiconductor layer can be formed by a plasma CVD method.
[0019]
On the other hand, the thin-film photoelectric conversion unit 4b includes a crystalline photoelectric conversion layer. For example, a p-type silicon-based semiconductor layer, a silicon-based photoelectric conversion layer, and an n-type silicon-based semiconductor layer are sequentially stacked from the thin-film photoelectric conversion unit 4a side. It has the following structure. All of the p-type semiconductor layer, the crystalline photoelectric conversion layer, and the n-type semiconductor layer can be formed by a plasma CVD method.
[0020]
The p-type semiconductor layers constituting these thin-film photoelectric conversion units 4a and 4b are formed by doping, for example, silicon or a silicon alloy such as silicon carbide or silicon germanium with p-conductivity determining impurity atoms such as boron or aluminum. be able to. In addition, the amorphous photoelectric conversion layer and the crystalline photoelectric conversion layer can be formed using an amorphous silicon-based semiconductor material and a crystalline silicon-based semiconductor material, respectively. Silicon hydride) or silicon carbide such as silicon carbide and silicon germanium. In addition, as long as the photoelectric conversion function is sufficiently provided, a weak p-type or weak n-type silicon-based semiconductor material containing a trace amount of impurities for determining conductivity type can be used. Further, the n-type semiconductor layer can be formed by doping silicon or a silicon alloy such as silicon carbide or silicon germanium with an n-conductivity determining impurity atom such as phosphorus or nitrogen.
[0021]
The absorption wavelength ranges of the thin film photoelectric conversion unit 4a and the thin film photoelectric conversion unit 4b configured as described above are different from each other. For example, when the thin-film photoelectric conversion layer of the thin-film photoelectric conversion unit 4a is made of amorphous silicon and the thin-film photoelectric conversion layer of the thin-film photoelectric conversion unit 4b is made of crystalline silicon, a light component of about 550 nm is added to the former. The light can be absorbed most efficiently, and the latter can most efficiently absorb a light component of about 900 nm.
[0022]
The thickness of the thin film photoelectric conversion unit 4a is preferably in the range of 0.01 μm to 0.5 μm, and more preferably in the range of 0.1 μm to 0.3 μm.
[0023]
On the other hand, the thickness of the thin-film photoelectric conversion unit 4b is preferably in the range of 0.1 μm to 10 μm, and more preferably in the range of 0.1 μm to 5 μm. That is, the thickness of the thin-film photoelectric conversion unit 4b is preferably about several times to about 10 times the thickness of the thin-film photoelectric conversion unit 4a. This is because the crystalline photoelectric conversion layer has a smaller light absorption coefficient than the amorphous photoelectric conversion layer.
[0024]
The metal back electrode layer 5 not only has a function as an electrode, but also reflects light incident on the photoelectric conversion units 4a and 4b from the transparent substrate 2 and arriving at the metal back electrode layer 5 to enter the photoelectric conversion units 4a and 4b. It also has a function as a reflective layer for re-entering. The metal back electrode layer 5 can be formed to a thickness of, for example, about 200 nm to 400 nm using silver, aluminum, or the like by an evaporation method, a sputtering method, or the like.
[0025]
Note that a transparent conductive thin film (not shown) made of a nonmetallic material such as ZnO is provided between the metal back electrode layer 5 and the photoelectric conversion unit 4b, for example, in order to improve the adhesiveness between the two. Can be provided.
[0026]
Each thin-film photoelectric conversion cell 10 formed on the transparent substrate 2 of the module 1 is sealed with an organic protective layer 7 via a sealing resin layer 6. As the sealing resin layer 6, a resin capable of bonding an organic protective layer to these cells 10 is used. As such a resin, for example, EVA (ethylene / vinyl acetate copolymer), PVB (polyvinyl butyral), PIB (polyisobutylene), and a silicone resin can be used. Further, as the organic protective layer 7, a fluororesin-based film such as a polyvinyl fluoride film (for example, Tedlar film (registered trademark)) or an insulating film excellent in moisture resistance and water resistance such as a PET film is used. Can be The organic protective layer may have a single-layer structure or a laminated structure in which these are laminated. Further, the organic protective layer may have a structure in which a metal foil made of aluminum or the like is sandwiched between these films. Since a metal foil such as an aluminum foil has a function of improving moisture resistance and water resistance, by using such a structure of the organic protective layer, it is possible to more effectively protect the thin film photoelectric conversion cell 10 from moisture. . The sealing resin layer 6 and the organic protective layer 7 can be simultaneously attached to the back surface of the thin-film photoelectric conversion module 1 by a vacuum laminating method.
[0027]
As shown in FIG. 2, the thin-film photoelectric conversion module 1 is provided with first and second separation grooves 21 and 22 for dividing the thin film and a connection groove 23. The first and second separation grooves 21 and 22 and the connection groove 23 are parallel to each other and extend in a direction perpendicular to the paper surface. The boundary between the adjacent cells 10 is defined by the second separation groove 22.
[0028]
The first separation groove 21 divides the transparent front electrode layer 3 corresponding to each cell 10, has an opening at the interface between the transparent front electrode layer 3 and the thin-film photoelectric conversion unit 4 a, and The surface is the bottom. The first separation groove 21 is filled with an amorphous material constituting the thin-film photoelectric conversion unit 4a, and electrically insulates the adjacent transparent front electrode layers 3 from each other.
[0029]
The second separation groove 22 is provided at a position separated from the first separation groove 21. The second separation groove 22 divides the thin-film photoelectric conversion units 4 a and 4 b and the metal back electrode layer 5 corresponding to each cell 10, and forms an interface between the metal back electrode layer 5 and the resin sealing layer 6. And the surface of the transparent front electrode layer 3 is a bottom surface. The second separation groove 22 is filled with the sealing resin layer 6, and electrically insulates the metal back electrode layers 6 between the adjacent cells 10.
[0030]
The connection groove 23 is provided between the first separation groove 21 and the second separation groove 22. The connection groove 23 divides the thin film photoelectric conversion units 4a and 4b, has an opening at the interface between the thin film photoelectric conversion unit 4b and the metal back electrode layer 5, and has the surface of the transparent front electrode layer 3 as the bottom surface. The connection groove 23 is filled with a metal material constituting the metal back electrode layer 5, and electrically connects the one metal back electrode layer 5 and the other transparent front electrode layer 3 of the adjacent cell 10. I have. That is, the connection groove 23 and the metal material filling the groove serve to connect the cells 10 juxtaposed on the substrate 1 in series.
[0031]
The film formation of the photoelectric conversion unit 4a is performed by using a multi-chamber in-line apparatus as shown in FIG. 5C and a side view of the film forming chamber as shown in FIG. 5A. Filmed. This film forming chamber is a so-called vertical CVD apparatus, and is a capacitively coupled CVD apparatus having a parallel plate electrode 25 capable of simultaneously processing two vertically installed substrates. In FIG. 5, LC is a loading chamber, HC is a heating chamber, p-layer C is a p-layer deposition chamber, i-layer C is an i-layer deposition chamber, n-layer C is an n-layer deposition chamber, and UC is an unloading chamber. . Hereinafter, the film forming chamber including the p layer C, the i layer C, the n layer C, and the like is referred to as PC.
[0032]
In the capacitive coupling type CVD apparatus including the parallel plate electrodes 25, the substrate heating heater 27 is installed at the center of the film forming chamber, and the parallel plate electrodes 25 connected to the high frequency power supply and introducing gas are installed on both sides. . At this time, the parallel plate electrode 25 may be present separately from the wall of the film forming chamber as shown at 25-1 in FIG. 5, but is also used as the so-called external electrode 25-2 as the wall of the film forming chamber. There is also.
[0033]
Conventionally, the film formation of the photoelectric conversion unit 4b has also been performed by a capacitively coupled CVD apparatus having the parallel plate electrodes 25 shown in FIG. 5, but only two film formation surfaces can be simultaneously processed. There is a problem that the film formation speed of the photoelectric conversion unit 4b, which needs to be stacked thicker than that of the photoelectric conversion unit 4a, cannot be sufficiently increased.
[0034]
Therefore, in the present invention, the film formation of the photoelectric conversion unit 4b was performed using the apparatus shown in FIG. This film-forming chamber is a so-called vertical CVD apparatus which can process a plurality of vertically installed substrates at the same time. Specifically, it can process six substrates simultaneously in three film-forming spaces. This is an inductance-coupled CVD apparatus including a rod-shaped electrode 26 to which power is supplied from a power source having a frequency selected from a plurality of frequencies of 80 MHz to 3 GHz.
[0035]
13.56, it is preferable that the frequency of VHF and MW can be changed. In addition, the quality of the film can be optimized by changing the frequency or performing the film formation using another rod-shaped electrode 26 of the present invention during the film formation. By using the vertical CVD, even when a large-area substrate having a side of 1 m is processed, a film can be formed without being affected by bending of the substrate and the antenna. Further, even if powder is generated in the apparatus during film formation, the powder can be prevented from dropping and adhering to the film forming surface of the substrate.
[0036]
6 is a U-shaped antenna electrode 25-1. In the apparatus of FIG. 6, the film formation of the photoelectric conversion unit 4b may use the wire electrode 26-2 instead of the antenna electrode 25-1. As the wire electrode 26-2, a U-shaped wire electrode is used similarly to the antenna electrode 26-1.
[0037]
Further, when microwaves of 1 GHz or more are introduced, a device in which a plurality of microwave electrodes 26-3 are linearly arranged vertically as rod electrodes 26 as shown in FIGS. 9 (a) and 9 (b) is used. Can be
[0038]
A blasted stainless steel rod is used as the antenna electrode 26-1. One end of the antenna is connected to a high frequency power supply by a coaxial cable, and the other end is grounded. In this state, the antenna is arranged so that the power supply side and the ground side alternate.
[0039]
The cross-sectional shape of the antenna electrode 26-1 is generally a perfect circle, but a flat ellipse is used and the flat surface faces the substrate surface to reduce the film thickness distribution. Can be. Further, a hollow pipe can be used as the antenna electrode 26-1, and can be used for supplying a film forming gas. In this case, the difference in film thickness between the substrates can be reduced by using a hollow pipe having an elliptical cross-sectional shape and providing a plurality of gas blowing holes on a surface having a large curvature that is not relative to the substrate surface.
[0040]
When forming a film with the film forming apparatus shown in FIG. 6B, it is important that the phases of the high-frequency power supplied to the antenna electrode 26-1 be opposite to each other between the adjacent antenna electrodes 26-1. Power is supplied from a single power supply to a plurality of adjacent antenna electrodes 26-1 or from a plurality of power supplies to a plurality of adjacent antenna electrodes 26-1. By controlling the phase between the two, the adjacent power supply units have opposite phases.
[0041]
FIG. 8 shows a method of supplying power to the antenna electrode 26-1. The high frequency generated by the function generator is current-amplified by a high-frequency amplifier, and is split into a predetermined number of power supplies by a splitter. The high frequency wave separated by the duplexer passes through the O-ring or the flange 29 with the insulator of the CVD apparatus wall 28 via the circulator for guiding the reflected wave from the antenna electrode 26-1 to the dummy load, and the antenna electrode 26- in the apparatus. 1 is supplied.
[0042]
It is possible to apply amplitude modulation or ON / OFF modulation to the high frequency supplied to the antenna electrode 26-1 by adding a pulse signal to the high frequency amplifier.
[0043]
By inserting a phase shifter between the function generator and the high-frequency amplifier, it is possible to adjust the phase between the plurality of antenna electrodes 26-1.
[0044]
The function generator, the phase shifter, the high-frequency amplifier, the duplexer, the circulator, the dummy load, the O-ring or the flange 29 with the insulator are connected by a coaxial cable.
[0045]
The other end of the antenna electrode 26-1 is electrically connected to the CVD device wall 28 and grounded. Specifically, as shown in FIG. 8, it is once led out of the CVD device wall 28 via the O-ring of the CVD device wall 28 or the flange 29 with a insulator, and is again electrically connected to the CVD device wall 28. Grounded. In this case, it is possible to control the discharge at the antenna electrode 26-1 to a certain extent for each electrode by grounding via the impedance of a capacitor, a coil, or the like. In such a case, a capacitor having a capacitance of 100 pF to 10 μF is used. Grounding of the CVD apparatus wall 28 is performed by screwing a copper or SUS high surface area conductor 30 electrically connected to the cable to the apparatus wall.
[0046]
When the microwave is supplied to the rod-shaped electrode 26, the rod-shaped electrode 26 becomes the microwave electrode 26-3 having the structure shown in FIG. Further, it is possible to install a conductor 33 between the dielectric tube 32 and the waveguide 31 to control the spread of discharge. A vacuum or nitrogen gas flows in the dielectric tube 32. Usually, a quartz tube is used as the dielectric tube 32, and a stainless steel tube is used as the waveguide 31 and the conductor 33.
[0047]
Power is supplied to the microwave electrode 26-3 by applying a microwave generated by a microwave oscillator to the microwave electrode 26-3 via a microwave distributor, an isolator, a stub tuner, a ball antenna, a waveguide, and the like. Is performed by propagating the light through the waveguide 31.
[0048]
As the material of the dielectric tube 32, other dielectric materials such as alumina ceramics, polytetrafluoroethylene, borosilicate glass, aluminum oxide, magnesium oxide, and silicon oxide can be used instead of quartz. Further, the plasma-sprayed Al is formed on the outside of the tubular dielectric material. ₂ O ₃ , BN, AlN, ZrSiO ₄ , TiO ₂ , Cr ₂ O ₃ It is also possible to use a material to which a ceramic material such as MgO or MgO is adhered.
[0049]
As a material of the waveguide 31 and the conductor 33, other materials such as Al, Fe, Ni, Cr, Ti, Mo, Au, In, Nb, Te, V, Pt, and Pb or a metal such as these are used instead of stainless steel. Alloys, especially Inconel, may be used.
[0050]
When the microwave electrode 26-3 is introduced from above and below the film forming apparatus in the arrangement of FIG. 7B, the microwave electrode 26 with the waveguide 31 cut off at the center as shown in FIG. 9B. -3.
[0051]
Since the radiation of microwaves in the direction in which the conductor 33 exists is limited, the film is formed with the substrates disposed on the left and right sides of the cross-sectional view of FIG. Although the conductor 33 has an elongated hole in the axial direction, it may be a partially closed ring to maintain the strength of the conductor tube 33 or to control the radiation of microwaves into the film forming chamber. Further, instead of a tube as shown in FIG. 9B, two elongated metal plates may be used. In particular, by using a part of a tube having a curvature opposite to that of FIG. 9A as two elongated metal plates, it is possible to improve the radiation of microwaves into the film forming chamber and to improve the thickness and quality of the substrate. Can improve the uniformity.
[0052]
As shown in FIG. 6, the distance 34 (ES) between the film-forming surface of the substrate 2 and the rod-shaped electrode 26 is adjusted by adjusting the thickness of the substrate frame 35 and changing the distance of the substrate-surface holding plate 36 from the rod-shaped electrode 26. By doing so, it can be varied between 0 cm and the distance 37 between the rod-shaped electrode central axes.
[0053]
As the wire electrode 26-2, a wire or plate-shaped transition metal wound in a straight or coil shape and bent into a U-shape is used. In this case, the arrangement is as shown in FIG. As shown in FIG. 7B, a linear wire electrode 26-2 may be used to supply power from above and below the device. The wire electrode 26-2 is connected to a current introduction port of the CVD apparatus wall 28, and is supplied with power from a power supply outside the apparatus. Each wire electrode 26-2 in the device is connected to a power source in parallel or in series with each other. Examples of the metal material used for the wire electrode 26-2 include tantalum, tungsten, molybdenum, and the like. The film formation of the polycrystalline silicon thin film on the substrate 2 by the wire electrode 26-2 is carried out by heating the wire electrode 26-2 to a temperature of usually 1000 ° C. to 2000 ° C., preferably 1300 ° C. to 1700 ° C. by energizing the hydrogen. This is performed by decomposition of silane gas.
[0054]
For example, in an apparatus having an LC-PC configuration, a method in which the substrate is heated by the wire electrode 26-2 and a film forming gas is introduced at the initial stage to form an initial film is conceivable. After the temperature rise and the completion of the initial film formation, the power supply to the wire electrode 26-2 is stopped, and the film formation by plasma discharge is started.
[0055]
In order to reliably perform discharge at each rod-shaped electrode 26, a discharge ignition device 39 can be provided in each rod-shaped electrode 26 or each discharge region 38 as shown in FIG. In the case of forming a film at a relatively high pressure, the discharge can be reliably started by irradiating the discharge region 38 with laser light.
[0056]
As the chamber configuration of the present CVD apparatus, a system such as an in-line, an inter-back, and a single wafer can be considered. For example, as for the in-line configuration, as shown in FIG. 10A, considering the order of LC (loading chamber), HC (heating chamber), PC (film forming chamber), and UC (loading chamber), three sets of substrates are considered. The cart in which is disposed moves the chambers of the CVD apparatus in this order, so that evacuation, temperature rise, film formation, introduction into the atmosphere, and the like are performed in order.
[0057]
In this CVD apparatus, since a silicon film adheres to the rod-shaped electrode 26 and polysilane powder generated in plasma adheres to the inner wall of the PC, CF4, CFmHn (m + n = 4), ClF ₃ , C ₂ F ₆ , SF ₆ , NF ₃ , ClF ₃ Cleaning using a mixed gas of these or the like is effective, and in that case, plasma discharge can be utilized. O if necessary ₂ , H ₂ , Ar, N ₂ , He may be mixed with one or more gases. In particular, when the chamber configuration is a ULC (load-in / out room) -HC-PC in FIG. 10B or an ULHC (load-in / out heating chamber) -PC in FIG. By performing the plasma cleaning on the PC, the manufacturing tact time can be reduced and the maintenance cycle can be extended. Further, when the film is formed by a CVD apparatus having the microwave electrode 26-3, a stable film can be formed by removing the silicon film or the powder attached to the surface of the dielectric tube 32 by cleaning, and particularly, It is valid.
[0058]
As shown in FIG. 10D, the present CVD apparatus can be connected in-line with the CVD apparatus of the photoelectric conversion unit 4a without breaking vacuum to increase the productivity as a manufacturing apparatus.
[0059]
In the present apparatus, control of cart movement, pump drive, gas valve opening / closing, film forming power supply, heater power supply, and the like are performed by a recipe previously input to a computer.
[0060]
In the present apparatus, exhaust by a pump such as a rotary pump, a mechanical booster pump, a turbo pump, and a dry pump is performed from a peripheral portion of the chamber, the rod-shaped electrode 26, a plate 40 described later, and the like. In the present CVD apparatus, a large amount of film forming gas is flowed to simultaneously form films on six substrates. Therefore, it is necessary to take measures against powder generated in the apparatus. For example, a method is conceivable in which a filter for trapping powder is provided in the exhaust pipe, two or more exhaust pipes are provided in parallel, and the filter is periodically replaced.
[0061]
The transfer of the substrate 2 is performed by moving a cart 41 in which six substrates are set. The cart 41 is transported along a rail 42 provided above, below, or above and below. The carts 41 need not always be transported in groups of six, and may be of a type capable of transporting an arbitrary number of sheets between one and six.
[0062]
In this apparatus, the film forming gas is introduced through a hole provided in the rod-shaped electrode 26. As shown in FIGS. 11 (c) and 11 (d), a plate 40 may be provided in the plane of the U-shaped rod-shaped electrode 26, this may be a straight pipe, and gas may be blown out from a hole provided in the pipe. In an apparatus using a linear rod-shaped electrode 26 as shown in FIGS. 7A and 7B, a plate 40 is provided in the plane of the rod-shaped electrode 26, and gas can be supplied in the same manner. Further, the gas may be blown out from a gas blowout hole provided in the side wall of the chamber.
[0063]
12 to 15 show the structure of the cart 42 used in the present CVD apparatus. At the time of attaching / detaching the substrate, as shown in FIG. With this structure, the substrate 2 and the back plate 44 can be easily detached. Further, a tray 45 may be provided at the lower part of the cart 42 to receive the powder generated at the time of film formation, and replace it with the tray 45 after cleaning at the time of attaching and detaching the substrate 2 to prevent scattering of the powder into the apparatus. .
[0064]
Further, a cart 41 having rail guides 46 at the top and bottom as shown in FIG. 14 may be used. In this case, the cart 41 is transported by a rail guide 46 along a rail 42 provided in the film forming apparatus. When the substrate 2 is detached, the cart 41 slides in the rail direction as shown in FIG. 15, so that the substrate 2 and the back plate 44 can be detached easily. The cart 41 may have a structure combining FIG. 12 and FIG.
[0065]
The substrate 2 is inserted into the opening surrounded by the substrate frame 35 so that the end of the film forming surface contacts the substrate surface holding plate 36, and the back plate 44 covers the non-film forming surface of the substrate 2 thereon. After that, it is set in the cart 41 by taking measures to prevent it from falling off.
[0066]
In this apparatus, a carbon back plate 44 is set on the non-film-forming surface of the substrate 2 in order to prevent the deposition of a thin film on the non-film-forming surface of the substrate due to the wraparound of the film-forming atmosphere. As a material of the back plate 44, glass, stainless steel or the like is used in addition to carbon.
[0067]
The temperature inside the HC into which the cart 41 is introduced is raised to 180 ° C., and then the cart moves to the PC room. The temperature was raised by forced convection heat transfer at substantially atmospheric pressure by the atmosphere, followed by decompression by a vacuum pump, thereby shortening the temperature raising time and improving the uniformity. Other methods for raising the temperature include a method using radiant heat transfer by lamp heating, a method using conduction, radiation, and convection heat transfer using a rod-shaped heater, and a method using conduction, radiation, and convection heat transfer using a heating plate. In the method using a heating plate, it is desirable to use a heating plate that can be driven in the direction of the substrate to make contact with the substrate.
[0068]
When the film is formed by the rod-shaped electrodes 26 as in this apparatus, the in-plane discharge intensity distribution becomes non-uniform compared to the case where the film is formed by the parallel plate electrodes 25. For this reason, it is preferable that the substrate 2 be oscillated monolithically during film formation by transporting the cart 41 in parallel with the arrangement of the rod-shaped electrodes 26 in the same discharge region 38. If the amplitude of the swing at this time is set to half or １ ／ of the distance 37 between the rod-shaped electrode center axes, the in-plane non-uniformity can be sufficiently reduced.
[0069]
Further, there is a case where the substrate 2 is locally heated due to non-uniformity of discharge during film formation. In particular, when microwave power is applied to the rod-shaped electrode 26, the effect is remarkable, and the substrate 2 may be thermally cracked. The swing of the substrate 2 is also a measure for preventing the occurrence of thermal cracks due to such local heating.
[0070]
When the film is formed by the rod-shaped electrode 26 as in this apparatus, the density distribution of the active species contributing to the film is more biased than when the film is formed by the parallel plate electrode 25. By controlling the generation of the active species in consideration of the lifetime of the active species and the diffusion length in the plasma, it is possible to make the distribution of the film thickness and the film quality uniform. Specifically, it is preferable to apply modulation such as ON / OFF to input of power to the rod-shaped electrode 26.
[0071]
Further, when the input power to the rod-shaped electrode 26 is large, the duty ratio of the ON / OFF modulation is not fixed, and the temperature ratio of the substrate 2 is controlled by decreasing the duty ratio with the lapse of time from the start of film formation. You may need to. For example, a method of attenuating the duty ratio exponentially as shown in Table 1 is effective. Further, the temperature rise may be suppressed by slightly shifting the pulse timing between the rod-shaped electrodes 26. Table 2 shows how to shift the pulse timing, for example. In this example, the power supplied to the rod-shaped electrode 26 is shifted by a half cycle between the center and the end in the device.
[0072]
[Table 1]

[0073]
[Table 2]

[0074]
The cause of the cracks in the substrate 2 is due to the temperature distribution in the substrate, but the temperature distribution is improved by using carbon having good thermal conductivity and high uniformity as the material of the back plate 44.
[0075]
As shown in FIG. 11A, by separately providing a film forming space by the rod-shaped electrode 26 and a cooling space by the plate 40 functioning as a cooling plate in the film forming chamber, the film forming and the cooling are broken. It is possible to repeat without.
[0076]
Further, as shown in FIG. 11 (c), the rod-like electrodes 26 and the plates 40 functioning as cooling plates are alternately arranged, and the substrate 2 is further swung to simultaneously perform high-speed film formation and optimization of the substrate temperature. be able to. The function of the plate 40 is to heat and cool the substrate 2 and to introduce and exhaust gas.
[0077]
In particular, when the plate 40 is arranged at the center of two linear portions of one rod-shaped electrode 26 as shown in FIG. 11D, the heating, cooling, gas introduction, and exhaust of the substrate 2 can be effectively performed. it can.
[0078]
Further, in the case of a system in which the transport by the cart 41 is performed in units of two substrates 2 as shown in FIG. 11B, the substrate 40 is provided behind the back plate 44 to heat the substrate, cool the substrate, and remove the gas. Inlet and exhaust can be performed.
[0079]
Further, in the case where the apparatus is such that the transfer can be performed in units of one substrate 2 as shown in FIG. 13E, the introduction of the rod-shaped electrodes 26 can be performed alternately from above and below the film forming apparatus. As a result, the film can be formed more uniformly in the plane of the substrate.
[0080]
It is more effective to combine the above-described methods of FIGS. 11A to 11E.
[0081]
The film forming conditions include the distance 34 between the film forming surface of the substrate and the rod-shaped electrode, pressure, gas flow rate, temperature, power supply frequency, discharge power, and the like.
[0082]
In the present apparatus, there is a projection in the PC room, such as the rod-shaped electrode 26, which is relatively easily deformed by physical force. Also, since the cart 41 moves, as shown in FIG. It is necessary to provide a space for the movement and maintenance of the cart 41 in the lower part of.
[0083]
In the present apparatus, particularly when high-frequency power in a frequency band of 40 MHz or less is used, SiH 4 -based gas and H 2 having a flow ratio of 10 times or more thereof are required as a reaction gas. Since the viscosity of the SiH4-based gas and that of hydrogen are greatly different, when the mixed gas flows through a flow path having a complicated shape such as the rod-shaped electrode 26, the discharge region 38 depends on a slight dimensional error of each antenna or a state of assembling parts. In some cases, the substantial mixing ratio of the gas reaching the sample may change. Therefore, for stable production, it is necessary to independently control the flow rate of the reactant gas for each discharge region 38 generated around the plurality of rod-shaped electrodes 26 arranged on one plane. Specifically, it is preferable that a mass flow controller is independently provided for each discharge region 38.
[0084]
Further, in this apparatus, since a silicon film is deposited on the rod-shaped electrode 26, for stable production, the rod-shaped electrode 26 and the plate 40 can be easily attached and detached, the deposited film can be easily removed, and during the film formation. It is important that the film has a structure that does not easily fall off. Specifically, the material of the rod-like electrode 26 and the plate 40 is preferably SUS that can withstand silicon etching removal by alkali. Further, in order to prevent the film from being removed during film formation, the surface is made to have minute unevenness by sandblasting or the like. Is desirable.
[0085]
When the photoelectric conversion unit 4b is formed using such an apparatus, a large number of sheets can be simultaneously and stably formed on a large area at a high speed as described above.
[0086]
That is, according to the present invention, an inexpensive amorphous silicon / polycrystalline silicon tandem photoelectric conversion element module and a method for manufacturing the same are provided.
[0087]
In the present invention, a large number of photoelectric conversion layers of a polycrystalline silicon unit cell can be simultaneously and stably formed in a large area by a CVD apparatus having a rod-shaped antenna electrode.
[0088]
(Example)
According to the following procedure, the thin film photoelectric conversion module 1 shown in FIGS. 1 and 2 was formed by using the apparatus shown in FIGS. 5 to 15 under the conditions shown in Tables 3 and 4, and the results shown in Tables 5 and 6 were obtained. Got.
[0089]
First, a 910 mm × 910 mm × 5 mmt glass substrate having an 800 nm thick SnO 2 film 3 on one main surface was prepared. The haze ratio of this glass substrate was 13%, and the sheet resistance was 12 Ω / □.
[0090]
[Table 3]

[0091]
[Table 4]

[0092]
[Table 5]

[0093]
[Table 6]

[0094]
Next, using a YAG IR pulsed laser, SnO ₂ By irradiating a laser beam from the film side and scanning, SnO ₂ A separation groove 21 having a width of 40 μm was formed to divide the film 3 into a plurality of strip patterns.
[0095]
This separation groove 21 is made of SnO ₂ The film 3 is divided into a plurality of strip patterns. The distance between the scribe lines is 8.9 mm, and 100 band-shaped patterns having a width of 8.9 mm and a length of 904 mm are formed in the center of the substrate 2.
[0096]
Further, a peripheral insulating groove 11 is formed by laser scribing 5 mm inward from two sides of the substrate 2 parallel to the integration direction 50 of the thin-film photoelectric conversion module 1. As shown in FIG. 3, when the substrate 2 is set on the substrate surface holding plate 36, the position is about 1 mm inside the substrate 2 from the edge of the substrate holding plate 36. It is preferable that the peripheral insulating groove 11 is formed of a plurality of grooves to ensure insulation. Specifically, a second peripheral insulating groove 11-2 having a width of about 100 μm is formed by laser scribing at a distance of about 0.7 mm inward from the first peripheral insulating groove 11-1. A third peripheral insulating groove 11-3 having a width of about 100 μm was formed at a distance of about 0.7 mm inward from 2 by laser scribing.
[0097]
At this time, as shown in FIG. 4, the probe 51 of the mega tester was applied at an interval of about 8 mm across the three peripheral insulating grooves 11, and a voltage of 250 V was applied and measured. Was obtained. Each of the first to third peripheral insulating grooves 11-1 to 11-3 is formed from the end of the separation groove 21 to the inside of the substrate 2; ₂ The film 3 is insulated from the periphery and the cell integration region.
[0098]
Thereafter, cleaning and drying are performed, and further, SnO is formed by a plasma CVD method. ₂ A thin film photoelectric conversion unit 4 a having a thickness of 200 nm to 350 nm was formed on the film 3. The photoelectric conversion unit 4a has a non-doped amorphous silicon layer as a photoelectric conversion layer, and forms a pin junction. The film formation of the photoelectric conversion unit 4a is performed using a multi-chamber in-line apparatus shown in FIG. 5C and a side view of the film forming chamber 13 using the apparatus shown in FIG. Electric power was supplied to the parallel plate electrodes 25 from an RF power supply having a frequency of .56 MHz, and two substrates 2 were simultaneously formed.
[0099]
After forming the photoelectric conversion unit 4a, SnO ₂ The peripheral insulating groove 11 may be formed through the film 3 and the thin-film photoelectric conversion unit 4a.
[0100]
Subsequently, the thin film photoelectric conversion unit 4b was formed on the thin film photoelectric conversion unit 4a by the CVD method. The photoelectric conversion unit 4b has a non-doped polycrystalline silicon layer as a photoelectric conversion layer. In some cases, after the formation of 4a, the p-layer or the n-layer of 4b is subsequently formed, and in the film forming apparatus shown in FIGS. 6 and 7 below, the i-layer of 4b may be formed. .
[0101]
Subsequently, the thin film photoelectric conversion unit 4a and the thin film photoelectric conversion unit 4b are scribed by irradiating the substrate 2 with a laser beam from the glass surface side using a YAG SHG pulsed laser, and scanning the substrate 2 with a plurality of band-shaped patterns. A connecting groove 23 having a width of 60 μm was formed. The distance between the centers of the connection groove 23 and the separation groove 24 was 100 μm.
[0102]
Thereafter, a ZnO film (not shown) and an Ag film were sequentially formed on the thin film photoelectric conversion unit 4b by a sputtering method to form the back electrode layer 5. Next, by irradiating the substrate 1 with a laser beam from the glass surface side using a YAG SHG pulse laser and scanning the substrate 1, the thin film photoelectric conversion units 4a, 4b and the back electrode layer 5 are divided into a plurality of belt-shaped patterns by a width of 60 μm. Was formed. The distance between the centers of the separation groove 22 and the connection groove 23 was 100 μm.
[0103]
Subsequently, by performing laser scanning along the periphery of the substrate 2 using a YAG IR pulse laser, SnO ₂ Grooves are formed in the film 3, the thin-film photoelectric conversion units 4a and 4b, and the back electrode layer 5, and laser scanning is performed along the periphery of the substrate 2 using a YAG SHG pulsed laser. , And a groove was formed in the back electrode layer 5 to define a power generation region.
[0104]
As described above, a structure was formed in which 100 hybrid-type thin-film photoelectric conversion cells 10 each having a size of 8.9 mm × 890 mm were connected in series in the integration direction 50 of the substrate 2.
[0105]
Thereafter, a pair of electrode bus bars 12 was attached to the substrate 2. Further, on the surface of the substrate 2 on which the hybrid type thin-film photoelectric conversion cells 10 are formed, an EVA sheet is formed as the sealing resin layer 6, and a black fluororesin-based sheet (trade name: Tedlar) is formed thereon as the organic protective layer 7. The thin film photoelectric conversion module 1 shown in FIGS. 1 and 2 was obtained by mounting and laminating by a vacuum lamination method.
[0106]
The initial characteristics of the solar cell are irradiance 100 mW / cm using a xenon lamp and a halogen lamp as light sources. ² The output characteristics were examined using a solar simulator of AM1.5. The measurement temperature was 25 ° C.
[0107]
The thickness distribution is determined by the laser microscope from the step of the separation groove 22 to the lamination of the sealing resin layer 6 and the organic protective layer 7 on the surface on which the hybrid type thin film photoelectric conversion cell 10 is formed after the separation groove 22 is formed. The total film thickness of the conversion units 4a, 4b and the back electrode layer 5 is measured, and the film thickness of the relatively uniform and thin thin film photoelectric conversion unit 4a and the back electrode layer 5 is fixed at a preset value. By subtracting a constant value, the film thickness of the thin-film photoelectric conversion unit 4b was obtained.
[0108]
The measurement points are at the positions of 55, 255, 455, 655, and 855 mm from the end of the substrate in the 1, 25, 50, 75, and 100th separation grooves 22 from the end of the substrate. The maximum value-minimum value / (maximum value + minimum value) × 100% was defined as the film thickness distribution.
[0109]
The film formation of the photoelectric conversion unit 4b of each example was performed using the apparatus shown in FIGS. The film formation of the photoelectric conversion unit 4b of the comparative example was performed by using the conventional CVD apparatus shown in FIG. 5 to form the pin layer, the i layer, and the n layer of the photoelectric conversion unit 4b in separate film forming chambers. .
[0110]
In Comparative Example 1, an apparatus having an electrode having a 25-1 structure was used, and in Comparative Example 2, an apparatus having an external electrode 25-2 was used to form films on two substrates 2 at the same time.
[0111]
The photoelectric conversion unit 4b is composed of three layers, i.e., a p-layer, an i-layer, and an n-layer. In each embodiment, six layers of the substrates 2 are simultaneously formed in one chamber by switching and introducing gases.
[0112]
In Examples 1 to 10, the antenna electrode 26-1 was formed by blasting a stainless steel tube having a length of about 120 cm, a total length of about 240 cm, and a length of about 240 cm, and a diameter of 16 mm. One end of the antenna electrode 26-1 is connected to a high frequency power supply by a coaxial cable, and the other end is grounded. In this state, the antenna is arranged so that the power supply side and the ground side alternate. The distance 37 between the center axes of the rod-shaped electrodes was 16 cm. The antenna has a structure in which a flat elliptical cross section is used, the flat surface is oriented with respect to the substrate surface, and a plurality of gas blowing holes are provided on a surface having a large curvature that is not relative to the substrate surface. Using the antenna electrode 26-1, silane and hydrogen as a film forming gas were introduced into the film forming chamber from the gas blowing holes.
[0113]
The phase of the high-frequency power supplied to the antenna electrode 26-1 was made opposite to that of the adjacent antenna electrode 26-1 by adjusting the cable length from the duplexer. In the splitter, the power from the high-frequency amplifier was equally divided into 16 parts. Power was supplied to the antenna electrode 26-1 in the apparatus through a flange 29 with a insulator on the wall 28 of the CVD apparatus.
[0114]
As shown in FIG. 10, the grounding of the antenna electrode 26-1 is once led out of the apparatus through a flange 29 with a insulator of the CVD apparatus wall 28, and is again applied to the CVD apparatus wall 28 via the large surface area conductor 30. Performed by screwing.
[0115]
In the eleventh embodiment, the photoelectric conversion unit 4b is manufactured using a microwave electrode 26-3 having a length of about 120 cm, a quartz tube 32 of 6 cmφ, a stainless tube 31 of 1 cmφ, and a stainless tube 33 of 5 cmφ. The membrane was made. The film forming gas was introduced into the film forming chamber from the wall 28 of the CVD apparatus. The distance 37 between the center axes of the rod-shaped electrodes was 16 cm.
[0116]
The apparatus used in the examples was an apparatus having no substrate heating mechanism as shown in FIGS. 6 and 7, and film formation was performed only by heating by electric discharge. In the example, an in-line type apparatus having a chamber configuration shown in FIG. 10A was used. Further, the method shown in FIGS. 12 and 13 was used for attaching, detaching and transporting the substrate 2 in the embodiment.
[0117]
The high frequency supplied to the antenna electrode 26-1 was subjected to ON / OFF modulation at 1 kHz and a duty ratio of 1: 1 by applying a pulse signal to a high frequency amplifier. The same modulation was applied to the microwave electrode 26-3.
[0118]
Further, in each example, the substrate 2 was reciprocated in 8 seconds, which is half the distance 37 between the center axes of the rod-shaped electrodes in 8 seconds.
[0119]
The cart 41 can be loaded in a state where the degree of vacuum in the film forming chamber (PC) is 2 × 10 −3 Pa or less. With the six substrates 2 set, the cart 41 is introduced into the LC. With the cart 41 introduced, the inside of the LC is evacuated to 1 × 10 −2 Pa, and thereafter, the cart 41 moves to the HC through the opening and closing operation of the gate valve.
[0120]
The temperature of the inside of the HC into which the cart 41 is introduced is raised to 120 ° C., and then the cart 41 moves to the PC. The conditions shown in Table 3 were used as film forming conditions on a PC. After the film formation, the inside of the PC is evacuated until the degree of vacuum becomes 2 × 10 −3 Pa or less, and the cart 41 is introduced into the UC. Nitrogen gas is introduced into the UC into which the cart 41 has been introduced, and the temperature of the substrate is set to 100 ° C. or lower by forced convection heat transfer under reduced pressure, and then the atmosphere is released to the atmosphere, and the cart 41 is carried out of the apparatus.
[0121]
According to the method described above, six modules 1 under each condition were manufactured, and the initial characteristics of the solar cell were measured. The results are shown in Table 5 for Examples and Table 6 for Comparative Examples.
[0122]
For the sample of Example 9A, cross-sectional TEM (transmission electron microscope) observation, Raman measurement, and impurity analysis by SIMS were performed.
[0123]
FIG. 16 is a cross-sectional TEM image of a sample cut out from Example 9A. Polycrystalline silicon having a fibrous structure is formed in a film forming direction from the substrate side.
[0124]
FIG. 17 shows Raman spectra of Example 9A, Comparative Example 1, and single crystal silicon. The samples for Raman measurement in Example 9A and Comparative Example 1 were obtained by forming a thin film photoelectric conversion unit 4b and then cutting the substrate 2 into a small area without forming the metal back electrode layer 5. Since the penetration depth in Raman measurement is about 1 μm, the crystallinity of the uppermost polycrystalline silicon thin film can be evaluated by this measurement. 520 cm as shown in FIG. ^-1 The peak due to the silicon crystal in the vicinity is stronger in Example 9A than in Comparative Example 1. As shown in FIG. 17B, the amorphous LA (320 cm ^-1 ), LO (420cm ^-1 ), TO (490cm) ^-1 The peak of ()) is stronger in Comparative Example 1 than in Example 9A. Therefore, it can be said that Example 9A has a smaller amorphous component and a larger crystallite size.
[0125]
FIG. 19 shows the result of performing impurity analysis by SIMS on the sample similar to the Raman measurement described above for Example 9A. The amount of impurities in the polycrystalline silicon thin film was about 1E + 17 for nitrogen, about 2E + 18 for carbon, and about 5E + 18 atoms / cc for oxygen. SnO with uneven surface on glass substrate 2 ₂ Since a film is formed and a silicon thin film is formed thereon, when the analysis is performed by digging in the depth direction by SIMS, the rising (falling) of the spectrum at the interface is not sharp but broad. Would.
[0126]
Under a film forming condition 4a (film forming condition of Example 9A), polycrystalline silicon was formed into a film of 700 nm on a single crystal silicon wafer, and IR (infrared absorption) measurement and TDS (thermal degassing) analysis were performed. .
[0127]
FIGS. 18A and 18B show the results of the IR measurement. The infrared absorption spectrum of a silicon hydride thin film usually has a thickness of 2000 cm corresponding to a Si-H bond. ^-1 Absorption and Si-H ₂ 2100cm corresponding to the bond ^-1 And there was only one peak in this region.
[0128]
In the TDS measurement, the absolute concentration of hydrogen in polycrystalline silicon was measured using single crystal silicon into which hydrogen ions were implanted as a reference. As a result, the concentration of bonded hydrogen was 6.7%.
[0129]
In each of the examples, the characteristics were improved in spite of the fact that the film forming time per substrate was shortened to 1/3 to 1/10 times as compared with the comparative example and the productivity of the film forming apparatus was greatly improved. The results were equal or better.
[0130]
Further, in each embodiment, as the discharge frequency increases, the optimal hydrogen / silane flow ratio decreases, and a high-quality film can be formed with a smaller hydrogen flow per substrate than in the comparative example. there were.
[0131]
In each embodiment, as the discharge frequency increases, the optimal pressure decreases, the optimal distance 34 (E / S) between the film-forming surface of the substrate and the rod-shaped electrode decreases, and the film thickness distribution increases. It was found that the film forming speed was increased, but the film forming speed was increased.
[0132]
The case where the longitudinal direction of the rod-shaped electrode 26 is parallel to the accumulation direction 50 and the case where it is perpendicular are higher when the substrate 2 is set to be parallel as in Example 9A and Example 9B of Table 5. Output was obtained. This is because the transparent conductive film SnO according to the distance from the rod-shaped electrode 26 generated in the film forming apparatus is used. ₂ This is because mutual interference with the separation grooves 21 of the film 3 is prevented when the film 3 is parallel, and the variation in the amount of photocurrent generated for each unit cell 10 due to the in-plane film thickness distribution is reduced. .
[0133]
A case where the peripheral insulating groove 11 is formed before the film formation of the thin film photoelectric conversion unit 4a, a case where the peripheral insulating groove 11 is formed before the film formation of the photoelectric conversion unit 4a after the film formation of the thin film photoelectric conversion unit 4a, and Table 3 shows the comparison of the film thickness distribution and the like in the case where the peripheral insulating groove 11 was formed after the completion of the entire film formation without forming the peripheral insulating groove 11, and the comparison of the solar cell characteristics was performed under the conditions 3a to 3c in Table 3. Examples 7 to 7 are shown. The formation of the peripheral insulating groove reduces the film thickness distribution, and improves the solar cell characteristics.
[0134]
Table 3 shows conditions 4a to 4d in which a comparison between the presence / absence of swing of the substrate 2 and the presence / absence of ON / OFF modulation is shown. The thickness distribution is reduced by the oscillation and modulation of the substrate, thereby preventing the substrate from cracking.
[0135]
【The invention's effect】
As described above, in the present invention, a hybrid-type photoelectric conversion element module having sufficient solar cell characteristics can be manufactured with high productivity by forming a polycrystalline silicon film using a CVD apparatus having rod-shaped electrodes.
[0136]
That is, according to the present invention, an inexpensive amorphous silicon / polycrystalline silicon hybrid photoelectric conversion element module and a method of manufacturing the same are provided.
[Brief description of the drawings]
FIG. 1 is a plan view schematically showing a hybrid thin-film photoelectric conversion module.
FIG. 2 is a sectional view schematically showing a hybrid thin-film photoelectric conversion module.
FIG. 3 is a cross-sectional view schematically showing a formation position of a peripheral insulating groove.
FIG. 4 is a plan view showing a method for measuring the insulating property of a peripheral insulating groove.
FIG. 5 is a conceptual diagram of a CVD apparatus having a parallel plate type electrode.
FIG. 6 is a conceptual diagram of a CVD apparatus provided with a rod-shaped electrode.
FIG. 7 is a conceptual diagram of a CVD apparatus provided with a linear rod-shaped electrode.
FIG. 8 is a conceptual diagram showing a method for supplying power to the antenna electrode and grounding.
FIG. 9 is a conceptual diagram showing the structure of a microwave electrode.
FIG. 10 is a conceptual diagram showing a chamber configuration of a film forming apparatus.
FIG. 11 is a conceptual diagram of a CVD apparatus provided with various types of rod-shaped electrodes.
FIG. 12 is a conceptual diagram showing the structure of a cart.
FIG. 13 is a conceptual diagram showing a method of attaching and detaching a substrate to and from a cart.
FIG. 14 is a conceptual diagram showing another structure of the cart.
FIG. 15 is a conceptual diagram showing a method of attaching and detaching a substrate to and from a cart having another structure.
FIG. 16 is a TEM image of Example 9A.
FIG. 17 is a Raman spectrum of Example 9A, Comparative Example 1, and single-crystal silicon.
FIG. 18 is an IR spectrum of a sample formed under film forming conditions 4a.
FIG. 19 is a SIMS depth profile of Example 9A.
[Explanation of symbols]
1 Thin-film photoelectric conversion module
2 Transparent substrate
3 Transparent front electrode layer
4a, 4b Thin-film photoelectric conversion unit
5 Metal back electrode layer
6 sealing resin layer
7 Organic protective layer
10 Thin-film photoelectric conversion cell
11 Peripheral insulation groove
12. Electrode bus bar
13 Module peripheral area
14,21,22 Separation groove
23 Connection groove
25 parallel plate electrode
26 rod electrode
26-1 Antenna electrode
26-2 Wire electrode
26-3 Microwave electrode
27 Substrate heater
28 CVD equipment wall
29 O-ring or flange with insulator
30 large surface area conductor
31 Waveguide
32 dielectric tube
33 Conductor
34 Distance between film-forming surface of substrate and rod-shaped electrode
35 Board frame
36 Board holding plate
37 Bar-shaped electrode center axis distance
38 Discharge area
39 Discharge ignition device
40 plates
41 cart
42 rails
43 Hinge
44 Back plate
45 saucer
46 Rail Guide
50 Accumulation direction
51 Mega Tester Tip

Claims (6)

A method for producing a polycrystalline silicon thin film from a reaction gas containing hydrogen and a silane-based gas using a reaction vessel comprising an exhaust means, a gas introduction means, and a power supply means, wherein the vertical A plurality of rod-shaped electrodes are arranged in parallel in the same plane to form a discharge region by supplying power, a substrate is arranged on both sides of the discharge region to form a film, and the planar discharge region is A method for forming a polycrystalline silicon thin film, wherein a plurality of polycrystalline silicon thin films are present in a reaction chamber. A method for manufacturing a tandem-type photoelectric conversion element, comprising: stacking a photoelectric conversion unit having an amorphous silicon thin film as a photoelectric conversion layer and a photoelectric conversion unit including a polycrystalline silicon thin film formed by the method according to claim 1. 2. The method for producing a polycrystalline silicon thin film according to claim 1, wherein the frequency of power supplied to said rod-shaped electrode is variable between 13.56 MHz and 3 GHz. Membrane method. 2. The polycrystalline silicon thin film forming method according to claim 1, wherein one end of the rod-shaped electrode is connected to a high-frequency power source, and the other end is connected to a wall surface of the reaction chamber and grounded. Silicon thin film forming method. 3. The method for manufacturing a tandem-type photoelectric conversion element according to claim 2, wherein a plurality of strip-shaped photoelectric conversion elements are connected in series on one substrate, and a direction of the series connection and a longitudinal direction of the rod-shaped electrode. Are in a parallel positional relationship with each other. A film forming apparatus comprising an exhaust unit, a gas introducing unit, and a power supply unit, and forming a polycrystalline silicon thin film from a reaction gas containing a silane-based gas and hydrogen. A plurality of rod-shaped electrodes are arranged in parallel in a plane to form a discharge region by supplying power, substrates are arranged on both sides of the discharge region to form a film, and the planar discharge region is formed by the film formation. An apparatus for forming a polycrystalline silicon thin film, wherein a plurality of such apparatuses are present in the apparatus.

Images