JP4502534B2 - Photovoltaic device manufacturing method - Google Patents

Description

【００２９】
次に、本発明の実施形態である下地層の表面近傍の膜中水素量を下地層のバルクの膜中水素量に比べて相対的に少なくした場合の効果を検討した。ここで、下地層の表面近傍とは、最表面から深さ５０Å以下の領域をいう。
【００３０】
図４は、基板温度１２０℃、ＲＦパワー１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍに固定して、Ｈ₂流量を０〜５０００ｓｃｃｍまで変化させて水素希釈率（Ｈ₂／ＳｉＨ₄）を変化させた際のａ−Ｓｉ：Ｈｐ層の膜中水素量の変化を示す。図１から図３までに用いたｐ層はＨ₂／ＳｉＨ₄＝２５にて形成した。これに対して、図４に示すものは、初期の１７０ÅをＨ₂／ＳｉＨ₄＝２５、すなわち水素量２１ａｔ．％の条件、残りの３０ÅをＨ₂／ＳｉＨ₄＝６０、すなわち水素量１６ａｔ．％の条件にて形成し、発電層を形成する下地層として用いた。このようにして形成された下地層は、表面近傍の水素膜中濃度が他のバルクの水素膜中濃度より相対的に少なくなる。
【００３１】
次に、発電層を先ほどの基板温度１２℃、１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて膜厚１５０Åおよび１５００Å形成し膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した。その結果を図５、図６に夫々示す。図５は、発電層として膜厚１５０Åを形成したもの、図６は、発電層として膜厚１５００Åを形成したものである。
【００３２】
図５より、下地ｐ層の発電層側の表面に低水素領域、発電層側のバルク発電層の水素量より相対的に水素含有量の多い高水素量含有発電層が存在していることが確認され、目的通りの構造が発電層の初期形成時に確認された。
【００３３】
更に、図６の如く発電層膜厚が厚く形成される過程で熱による水素拡散が進行し、水素濃度の分布が大幅に改善されていることが明らかになった。すなわち、プラズマＣＶＤの初期放電に起因した光起電力装置の発電層の下地層との界面領域での高水素量含有発電層の水素が低水素量下地層に拡散することにより高水素量含有発電層の発生を抑制できた。
【００３４】
次に、光起電力装置を形成し、図２で水素プロファイルを評価した構造すなわち均質ｐ層上にＲＦパワー１５０ｍＷ／ｃｍ²にて発電層を形成した場合の特性により規格化を行った結果を表２に示す。表２より、狙い通り初期のＦ．Ｆ．、変換効率が大幅に改善され、更に光劣化率の低減により光劣化後効率では約１５％の改善が実現できた。
【００３５】
【表２】

【００３６】
次に、基板温度１２０℃、ＲＦパワー１５０ｍＷ／cm²、圧力１０Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍに固定して、Ｈ₂流量：１２６０ｓｃｃｍにて１００Åのｐ層に対して、その表面をＲＦパワー３００ｍＷ／cm²、圧力１００Ｐａ、Ａｒ流量：１０００ｓｃｃｍにてＡｒプラズマ処理を行った際のＡｒプラズマ処理時間と表面から５０Åの領域の膜中水素量の関係を図７に示す。
【００３７】
図７よりＡｒプラズマ処理時間の増加に伴い表面近傍領域の水素量が低下することが分かる。この知見を利用して、Ａｒプラズマ処理時間２分の条件を適用し、先ほど同様に発電層を基板温度１２０℃、１５０ｍＷ／cm²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて膜厚１５００Å形成し、膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した。その結果を図8に示す。
【００３８】
図８より、図2に示した従来例に比べて光起電力装置の発電層の下地層との界面領域での高水素量含有発電層の発生を抑制できた。
【００３９】
更に、光起電力装置を形成し、図２で水素プロファイルを評価した構造すなわち均質ｐ層上にＲＦパワー１５０ｍＷ／cm²にて発電層を形成した場合の特性により規格化を行った結果を表３に示す。
【００４０】
【表３】

【００４１】
表３より、狙い通り初期のＦ．Ｆ．、変換効率が大幅に改善され、更に光劣化率の低減により光劣化後効率では約１４％の改善が実現できた。尚、同様の効果が、Ａｒ以外の希ガスプラズマ処理でも得られることも確認した。
【００４２】
次に、基板温度１２０℃、ＲＦパワー１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍに固定してＨ₂流量：１２５０ｓｃｃｍにて２００Åのｐ層に対して、その表面をＲＦパワー３００ｍＷ／ｃｍ²、圧力１００Ｐａ、Ｈ₂流量：１０００ｓｃｃｍにてＨ₂プラズマ処理を行った際のＨ₂プラズマ処理時間と表面から５０Åの領域の膜中水素量、および５０から２００Åのバルクｐ層の膜中水素量の関係を図９に示す。
【００４３】
図９よりＨ₂プラズマ処理時間の増加に伴いバルクの膜中水素量は増加するが、表面近傍領域の水素量は低下することが分かる。これは、バルクの水素量増加は水素打ち込み効果により、表面近傍では水素プラズマによる表面へのエネルギー付与と水素濃度の増加によりａ−Ｓｉ：Ｈの構造緩和に伴った水素脱離反応が促進されているためであると考えられる。
【００４４】
この知見を利用して、Ｈ₂プラズマ処理時間３分の条件を適用し、先ほど同様に発電層を基板温度１２０℃、１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて膜厚１５００Å形成し、膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した。その結果を図１０に示す。図１０より、図２に示した従来例に比べて光起電力装置の発電層の下地層との界面領域での高水素量含有発電層の発生を抑制できた。
【００４５】
更に、光起電力装置を形成し、図２で水素プロファイルを評価した構造すなわち均質ｐ層上にＲＦパワー１５０ｍＷ／ｃｍ²にて発電層を形成した場合の特性により規格化を行った結果を表４に示す。
【００４６】
【表４】

【００４７】
表４より、狙い通り初期のＦ．Ｆ．、変換効率が大幅に改善され、更に光劣化率の低減により光劣化後効率では約１６％の改善が実現できた。表２、３の実施形態に比べてＩｓｃの増加が大きいのはｐ層のバルクヘの水素打ち込みによるワイドギャップ化に起因した光吸収ロス低減によると考えられる。
【００４８】
ここで、表２〜３で評価した光起電力装置と図２で評価した光起電力装置の光起電力装置基板上での水素濃度プロファイルを比較した結果を図１１、図１２に示す。図１１は図２にて評価した従来の条件、図１２は図８で評価した本発明の実施形態の条件における光起電力装置構造でのＳＩＭＳにより評価した水素プロファイルである。太陽竜池では凹凸基板を使用している為に水素プロファイルの深さ方向の精度は劣ると考えられるが、従来例に比べて本発明の実施形態を用いた場合は、ｐ／ｉ界面近傍のｉ側での高水素量含有発電層の抑制効果は一目瞭然であった。
【００４９】
次に、基板温度１２０℃、４５０ｍＷ／ｃｍ²、圧力３００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００００ｓｃｃｍ、２％ボロンドープにて形成した微結晶ｐ層上に、基板温度１２０℃、４５０ｍＷ／ｃｍ²、圧力３００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００００ｓｃｃｍにて形成した微結晶ｉ層を１５００Å形成し膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した結果を図１３に示す。微結晶の場合にも成膜初期に高水素領域が存在し、断面ＴＥＭ写真よりａ―Ｓｉ：Ｈであることが確認された。
【００５０】
更に、本発明の実施形態であるＲＦパワー３００ｍＷ／ｃｍ²、圧力１００Ｐａ、Ｈ₂流量：１０００ｓｃｃｍにてＨ₂プラズマ処理を行った後に前記条件にて微結晶発電層を形成した際のＳＩＭＳによる膜中水素量の評価結果を図１４に示す。
【００５１】
図１４より、図１３にみられた高水素量含有発電層の発生が抑制されていることが明らかとなった。
【００５２】
更に、光起電力装置を作成して特性を比較した結果を表５に示す。表５は、本発明の実施形態の太陽電池特性を従来例の太陽電池特性にて規格化した値である。表５より、ａ―Ｓｉ：Ｈ発電層の場合同様、微結晶Ｓｉ発電層においても、本発明が有効であることが確認された。
【００５３】
【表５】

【００５４】
以上から明らかなように、本発明によるところのｐｉｎ型光起電力装置において発電層を形成する下地層の表面近傍の膜中水素量を下地層のバルクの膜中水素量に比べて相対的に少なくすることにより、プラズマＣＶＤの初期放電に起因した光起電力装置の発電層の下地層との界面領域での高水素量含有発電層の水素が低水素量下地層に拡散することにより高水素量含有発電層の発生を抑制でき、光劣化前後において高効率非晶質光起電力装置を得ることが可能であることが明らかとなつた。
【００５５】
上記した実施形態においては、ｐ層上にｉ層を直接形成する際に、下地になるｐ層の表面近傍の水素量をバルクに比べて少なくした場合につき説明した。同様に、ｐ層上にバッファ層を形成し、このバッファ層上にｉ層を形成する場合にも下地となるバッファ層のｉ層との界面の表面近傍の水素量をバルクに比べて少なくすることで、同様の効果が得られることは勿論のことである。
【００５６】
さらに、上記した実施形態においては、ｐｉｎのシングル構造の光起電力装置にこの発明を適用した場合につき説明した。同様に、ｐｉｎ構造の半導体層を数段階積層した所謂タンデム構造の光起電力装置にこの発明は適用できる。即ち、ｉ層の下地になる層の表面近傍の水素量をバルクに比べて少なくするとよい。
【００５７】
また、上記した実施形態においては、ｐｉｎ構造の光起電力装置の場合につき説明したが、ｎｉｐ型構造の光起電力装置にもこの発明は適用できる。このｎｉｐ構造の場合には、下地となるｎ層の表面近傍の水素量をバルクに比べて少なくすればよい。このように、下地となるｎ層表面近傍の水素量をバルクに比べて少なくすることで、発電層の水素含有プロファイルが平坦化し、長波長側の感度を上げることができる。
【００５８】
【発明の効果】
以上から明らかなように、本発明によれば、発電層を形成する下地層の表面近傍の膜中水素量を下地層のバルクの膜中水素量に比べて相対的に少なくすることにより、プラズマＣＶＤの初期放電に起因した光起電力装置の発電層の下地層との界面領域での高水素量含有発電層の水素が低水素量下地層に拡散することにより高水素量含有発電層の発生を抑制し、高効率非晶質光起電力装置を得ることできる。
【図面の簡単な説明】
【図１】基板温度１２０℃、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて発電層成膜時のＲＦパワーを５０ｍＷ／ｃｍ²にて膜厚１５００Åの発電層を形成した際の、下地ｐ層と発電層近傍の２次イオン質量分析法（ＳＩＭＳ）により評価した水素濃度の深さ方向分布を示す特性図である。
【図２】基板温度１２０℃、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて発電層成膜時のＲＦパワーを１５０ｍＷ／ｃｍ²にて膜厚１５００Åの発電層を形成した際の、下地ｐ層と発電層近傍の２次イオン質量分析法（ＳＩＭＳ）により評価した水素濃度の深さ方向分布を示す特性図である。
【図３】ｐ層上にパワー密度２５ｍＷ／ｃｍ²にて初期の５０Åを形成し、その後５０ｍＷ／ｃｍ²にて残り１４５０Åの発電層を形成した場合のＳＩＭＳによる水素濃度評価結果を示す特性図である。
【図４】基板温度１２０℃、ＲＦパワー１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍに固定してＨ₂流量を０〜５０００ｓｃｃｍまで変化させて水素希釈率（Ｈ₂／ＳｉＨ₄）を変化させた際のｐ層の膜中水素量の変化を示す特性図である。
【図５】図４に示す下地層上に、発電層を基板温度１２０℃、１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、８ｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて膜厚１５０Å形成し、膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した結果を示す特性図である。
【図６】図４に示す下地層上に、発電層を基板温度１２０℃、１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて膜厚１５００Å形成し膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した結果を示す特性図である。
【図７】下地層を形成後、その表面をＲＦパワー３００ｍＷ／ｃｍ²、圧力１００Ｐａ、Ａｒ流量：１０００ｓｃｃｍにてＡｒプラズマ処理を行った際のＡｒプラズマ処理時間と表面から５０Åの領域の膜中水素量の関係を示す特性図である。
【図８】図７に示す下地層上、発電層を基板温度１２０℃、１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて膜厚１５００Å形成し膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した結果を示す特性図である。
【図９】下地層を形成後、その表面をＲＦパワー３００ｍＷ／ｃｍ²、圧力１００Ｐａ、水素流量：１０００ｓｃｃｍにて水素プラズマ処理を行った際のＡｒプラズマ処理時間と表面から５０Åの領域の膜中水素量の関係を示す特性図である。
【図１０】図９に示す下地層上、発電層を基板温度１２０℃、１５０ｍＷ／ｃｍ²、圧力１００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００ｓｃｃｍにて膜厚１５００Å形成し膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した結果を示す特性図である。
【図１１】図２にて評価した従来の条件おける光起電力装置構造でのＳＩＭＳにより評価した水素プロファイルである。
【図１２】図８で評価した本発明の実施形態の条件における光起電力装置構造でのＳＩＭＳにより評価した水素プロファイルである。
【図１３】基板温度１２０℃、４５０ｍＷ／ｃｍ²、圧力３００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００００ｓｃｃｍ、２％ボロンドープにて形成した微結晶ｐ層上に、基板温度１２０℃、４５０ｍＷ／ｃｍ²、圧力３００Ｐａ、ＳｉＨ₄流量：５０ｓｃｃｍ、Ｈ₂流量：２００００ｓｃｃｍにて形成した微結晶ｉ層を１５００Å形成し、膜中水素量の膜厚方向の変化をＳＩＭＳにより評価した結果を示す特性図である。
【図１４】下地層に、ＲＦパワー３００ｍＷ／ｃｍ²、圧力１００Ｐａ、Ｈ₂流量：１０００ｓｃｃｍにてＨ₂プラズマ処理を行った後に、微結晶発電層を形成した際のＳＩＭＳによる膜中水素量の評価結果を示す特性図である。
【図１５】ｐｉｎ接合を有するｐｉｎ型ａ−Ｓｉ光起電力装置の構造を示す断面図である。
【符号の説明】
１ ガラス基板
２ 透明電極
３ ｐ型ａ−ＳｉＣ層
４ ｉ型ａ−Ｓｉ層（発電層）
５ ｎ型微結晶シリコン
７ 裏面金属電極[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method for manufacturing a photovoltaic device, and more particularly to a method for manufacturing a photovoltaic device using an amorphous thin film semiconductor for a photovoltaic layer.
[0002]
[Prior art]
Conventionally, a photovoltaic device mainly made of amorphous silicon (hereinafter referred to as a-Si) formed by glow discharge decomposition of a raw material gas has a feature that it is easy to increase the thickness and area of a thin film. Expected to be a cost photovoltaic device.
[0003]
As this type of photovoltaic device, a pin-type a-Si photovoltaic device having a photoelectric conversion layer composed of a pin junction is known. FIG. 15 shows the structure of such a photovoltaic device. On a glass substrate 1, a transparent electrode 2, such as tin oxide (SnO ₂ ), a p-type a-SiC layer 3, an i-type a-Si layer 4, n It is formed by sequentially laminating a type microcrystalline silicon (hereinafter referred to as μc-Si) layer 5 and a back metal electrode 7. In the photovoltaic device, photovoltaic power is generated by light incident through the glass substrate 1. In FIG. 15, a transparent conductive layer 6 such as ZnO or ITO is provided between the back metal electrode 7 and the μc-Si layer 5 in order to suppress alloying or the like.
[0004]
By the way, in the pin-type photovoltaic device using amorphous silicon (a-Si: H) in which light is incident from the above glass substrate, power generation is generally located on the light incident side and has a high photoinduced carrier density. It is known that the film characteristics on the p-layer side of the layer (i-layer) greatly influence the solar cell characteristics.
[0005]
[Problems to be solved by the invention]
However, in the plasma CVD technique generally used for forming a semiconductor layer, the amount of hydrogen in the film is relatively high compared to the bulk at the beginning of discharge, the SiH ₂ / SiH bond ratio is large, and the film characteristics are inferior. An a-Si: H film is formed, which causes deterioration of solar cell characteristics not only in the initial stage but also after light deterioration.
[0006]
Furthermore, the development of high-speed film formation technology for power generation layers has been energetically performed in response to the recent demand for cost reduction, but under high-speed film formation conditions, the amount of hydrogen in the power generation layer in the initial generation discharge has increased. The deterioration of the film quality becomes remarkable.
[0007]
The present invention solves these problems and suppresses the generation of a power generation layer containing a high hydrogen content in the interface region between the power generation layer of the photovoltaic device and the underlying layer, thereby obtaining a highly efficient photovoltaic device. With the goal.
[0008]
[Means for Solving the Problems]
The present invention relates to a photovoltaic device manufacturing method in which a photoelectric conversion layer made of a thin film semiconductor having a semiconductor junction therein is formed by chemical vapor deposition, and a base semiconductor layer on which the power generation layer of the photoelectric conversion layer is formed A step of performing plasma treatment with hydrogen gas so that the hydrogen concentration in the film near the surface is relatively lower than the hydrogen concentration in the bulk of the base semiconductor layer; Forming a power generation layer on the surface, and forming a power generation layer having a certain amount of hydrogen concentration profile .
[0009]
According to the above configuration, the amount of hydrogen in the film near the surface of the underlayer that forms the power generation layer is relatively small compared to the amount of hydrogen in the bulk of the underlayer, resulting in the initial discharge of plasma CVD. A hydrogen concentration profile of a certain amount can be formed by diffusing hydrogen in the high hydrogen content-containing power generation layer into the low hydrogen content base layer in the interface region between the power generation layer and the base layer of the photovoltaic device.
[0010]
In addition, the present invention is preferably applied to a photovoltaic device in which the photoelectric conversion layer has a pin junction and light is incident from the p-type semiconductor layer side.
[0011]
Due to the initial discharge of plasma CVD, the amount of hydrogen in the film near the surface of the underlayer that forms the power generation layer in the pin type photovoltaic device is relatively small compared to the amount of hydrogen in the bulk of the underlayer. The high hydrogen content power generation layer diffuses into the low hydrogen content base layer in the interface region between the photovoltaic device power generation layer and the base layer. An amorphous photovoltaic device can be obtained.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings.
First, the inventor has intensively studied the hydrogen concentration in the i-type a-Si: H film serving as a power generation layer formed by the plasma CVD method described above. For this study, a photovoltaic device was created as follows.
[0015]
A p-layer, an i-layer serving as a power generation layer, and an n-layer were formed using a known RF plasma CVD (13.56 MHz) on a substrate in which SnO ₂ was formed as a transparent conductive film on a glass substrate. The formation temperature of i-type a-Si: H serving as a power generation layer is 100 to 300 ° C., the reaction pressure is 5 to 100 Pa, and the RF power is 1 to 500 mW / cm ² .
[0016]
This power generation layer has a single junction structure with an optical gap Eopt of 1.60 eV and a film thickness of 1000 to 3000 mm.
[0017]
The p layer and the n layer are also formed by using known RF plasma CVD, the doping amount (boron atom / silicon atom in the p layer, phosphorus atom / silicon atom in the n layer) is 1%, and the film thickness of the p layer is constant to 200 mm. The film thickness of the n layer was 100 mm.
[0018]
1 and 2 show a base p layer when a power generation layer (i layer) is formed at a substrate temperature of 120 ° C., a pressure of 100 Pa, a monosilane (SiH ₄ ) flow rate: 50 sccm, and a hydrogen (H ₂ ) flow rate: 200 sccm. 2 is a characteristic diagram showing a depth direction distribution of hydrogen concentration evaluated by secondary ion mass spectrometry (SIMS) in the vicinity of the power generation layer. FIG. 1 shows that when a power generation layer (i layer) having a film thickness of 1500 Å is formed at an RF power of 50 mW / cm ² , FIG. 2 shows a power generation layer (i) having a film thickness of 1500 Å at an RF power of 150 mW / cm ² . In this case, the layer is formed.
[0019]
Unless otherwise specified, in this specification, when a hydrogen profile is confirmed, an underlayer and an i layer are formed on a single crystal silicon (c-Si) substrate having a flat surface. This is to improve the measurement accuracy in the depth direction of the hydrogen profile.
[0020]
Moreover, the cross-sectional TEM (transmission electron microscope) photograph was used for the film thickness of an underlayer / i layer, and interface specification.
[0021]
1 and 2, there is a high hydrogen content power generation layer having a hydrogen content relatively higher than that of the bulk power generation layer in the vicinity of the interface between the base p layer and the power generation layer. When it was high, an increase in the hydrogen concentration peak and an increase in the thickness of the power generation layer containing a high hydrogen content were confirmed. This is thought to be due to the generation of higher-order radicals (SiH ₂ radicals) in the plasma due to the initial discharge.
[0022]
The high hydrogen region also exists on the p layer side of the p / i interface specified by the cross-sectional TEM photograph is considered to be due to hydrogen implantation during the formation of the i layer and hydrogen diffusion during the formation of the power generation layer. .
[0023]
Further, an initial 50 mm is formed at a minimum power density of 25 mW / cm ² capable of stable discharge in a plasma CVD apparatus using low power of the initial discharge, which is generally used as a countermeasure for initial discharge, and then 50 mW / cm. the hydrogen concentration evaluation results by SIMS in the case of forming a power generation layer consisting of i layer remaining 1450Å at ² shown in FIG.
[0024]
As shown in FIG. 3, by reducing the initial power of 50 layers of the i layer, the hydrogen concentration peak of the high hydrogen content-containing power generation layer near the interface between the base p layer and the power generation layer is reduced, and the thickness of the high hydrogen content power generation layer is reduced. Has been confirmed, but has not been completely resolved.
[0025]
Further, in FIG. 3, when the RF power was continuously increased from 25 mW / cm ² to 50 mW / cm ² , formation of a high hydrogen content-containing power generation layer having a hydrogen concentration higher than that of the bulk was observed.
[0026]
Table 1 shows the solar cell characteristics when the RF power during power generation film formation is constant at 50 mW / cm ² and 150 mW / cm ² in terms of the solar cell characteristics under the condition that the initial 50 mm of the i layer is reduced in power. Normalized IV and normalized degradation efficiency (500 mW / cm ² , 25 ° C., 160 min) are shown. All of the _samples prepared under the condition of constant RF power are I _sc , _F.I. F. This is considered to be due to an increase in carrier recombination in the vicinity of the p / i interface.
[0027]
In the case of 50 mW / cm ² to 150 mW / cm ² , I _sc , _F.I. F. A significant difference was also confirmed in the photodegradation rate. It has been confirmed from the film properties that there is no significant difference in the bulk film quality (conductivity, photosensitivity, defect density) between the two, and the hydrogen concentration in the power generation layer with a high hydrogen content near the interface between the underlying p layer and the power generation layer This is thought to be related to an increase in peak and an increase in the thickness of the power generation layer containing a high hydrogen content.
[0028]
[Table 1]

[0029]
Next, the effect of reducing the amount of hydrogen in the film near the surface of the underlayer, which is an embodiment of the present invention, relative to the amount of hydrogen in the bulk of the underlayer was examined. Here, the vicinity of the surface of the underlayer refers to a region having a depth of 50 mm or less from the outermost surface.
[0030]
FIG. 4 shows a hydrogen dilution rate (H ₂ / SiH ₄ ) with a substrate temperature of 120 ° C., RF power of 150 mW / cm ² , pressure of 100 Pa, SiH ₄ flow rate: 50 sccm, and H ₂ flow rate changed from 0 to 5000 sccm. The change of the amount of hydrogen in the film of the a-Si: Hp layer when changing is shown. The p layer used in FIGS. 1 to 3 was formed with H ₂ / SiH ₄ = 25. On the other hand, in the case shown in FIG. 4, the initial 170% is H ₂ / SiH ₄ = 25, that is, the hydrogen amount is 21 at. % Condition, the remaining 30% is H ₂ / SiH ₄ = 60, that is, the amount of hydrogen is 16 at. %, And used as a base layer for forming a power generation layer. The underlayer formed in this way has a hydrogen film concentration in the vicinity of the surface that is relatively lower than other bulk hydrogen film concentrations.
[0031]
Next, the power generation layer is formed with a substrate temperature of 12 ° C., 150 mW / cm ² , a pressure of 100 Pa, a SiH ₄ flow rate of 50 sccm, and an H ₂ flow rate of 200 sccm. Changes were evaluated by SIMS. The results are shown in FIGS. 5 and 6, respectively. FIG. 5 shows a power generation layer with a thickness of 150 mm, and FIG. 6 shows a power generation layer with a thickness of 1500 mm.
[0032]
From FIG. 5, it can be seen that there is a high hydrogen content-containing power generation layer having a relatively high hydrogen content in the low hydrogen region and the power generation layer side bulk power generation layer on the surface of the base p layer on the power generation layer side. It was confirmed that the intended structure was confirmed at the initial formation of the power generation layer.
[0033]
Furthermore, it became clear that hydrogen diffusion by heat progressed in the process of forming the power generation layer thick as shown in FIG. 6, and the hydrogen concentration distribution was greatly improved. In other words, high hydrogen content power generation due to diffusion of hydrogen in the high hydrogen content power generation layer to the low hydrogen content base layer in the interface region between the power generation layer of the photovoltaic device and the base layer due to the initial discharge of plasma CVD. Generation of layers could be suppressed.
[0034]
Next, a photovoltaic device was formed, and the result of normalization according to the characteristics obtained when the hydrogen profile was evaluated in FIG. 2, that is, when the power generation layer was formed at a RF power of 150 mW / cm ² on the homogeneous p-layer was obtained. It shows in Table 2. From Table 2, the initial F.D. F. As a result, the conversion efficiency was greatly improved, and the post light degradation efficiency was improved by about 15% by reducing the light degradation rate.
[0035]
[Table 2]

[0036]
Next, the substrate temperature is 120 ° C., the RF power is 150 mW / cm ² , the pressure is 10 Pa, the SiH ₄ flow rate is fixed to 50 sccm, and the surface is applied to the p layer of 100 mm at the H ₂ flow rate: 1260 sccm with the RF power of 300 mW / FIG. 7 shows the relationship between the Ar plasma processing time and the amount of hydrogen in the film in the region 50 mm from the surface when Ar plasma processing is performed at cm ² , pressure 100 Pa, Ar flow rate: 1000 sccm.
[0037]
FIG. 7 shows that the amount of hydrogen in the vicinity of the surface decreases as the Ar plasma treatment time increases. Using this knowledge, the condition of Ar plasma treatment time of 2 minutes was applied, and the power generation layer was similarly set to the substrate temperature of 120 ° C., 150 mW / cm ² , pressure of 100 Pa, SiH ₄ flow rate: 50 sccm, and H ₂ flow rate: 200 sccm. A film thickness of 1500 mm was formed, and the change in the film thickness direction of the amount of hydrogen in the film was evaluated by SIMS. The results are shown in FIG.
[0038]
From FIG. 8, compared with the prior art shown in FIG. 2, the generation of the high hydrogen content power generation layer in the interface region between the power generation layer of the photovoltaic device and the base layer could be suppressed.
[0039]
Further, the results of normalization by the characteristics when a photovoltaic device was formed and the structure in which the hydrogen profile was evaluated in FIG. 2, that is, when the power generation layer was formed on the homogeneous p layer at an RF power of 150 mW / cm ² are shown. 3 shows.
[0040]
[Table 3]

[0041]
From Table 3, the initial F.D. F. The conversion efficiency was greatly improved, and the post-light degradation efficiency was improved by about 14% by reducing the light degradation rate. It was also confirmed that the same effect can be obtained by a rare gas plasma treatment other than Ar.
[0042]
Next, the substrate temperature is 120 ° C., the RF power is 150 mW / cm ² , the pressure is 100 Pa, the SiH ₄ flow rate is fixed to 50 sccm, the H ₂ flow rate is 1250 sccm, and the surface is RF power 300 mW / cm with respect to the 200-cm p layer. ^2. H ₂ plasma treatment time when H ₂ plasma treatment was performed at a pressure of 100 Pa and H ₂ flow rate: 1000 sccm, the amount of hydrogen in the film in the region of 50 mm from the surface, and the hydrogen in the film of the bulk p layer of 50 to 200 kg The quantity relationship is shown in FIG.
[0043]
As can be seen from FIG. 9, the amount of hydrogen in the bulk film increases as the H ₂ plasma treatment time increases, but the amount of hydrogen in the vicinity of the surface decreases. This is because the increase in the amount of hydrogen in the bulk is due to the hydrogen implantation effect, and in the vicinity of the surface, the hydrogen desorption reaction accompanying the structural relaxation of a-Si: H is promoted by the application of energy to the surface by hydrogen plasma and the increase in the hydrogen concentration. It is thought that this is because.
[0044]
Using this knowledge, the condition of 3 minutes of H ₂ plasma treatment time was applied, and the power generation layer was similarly subjected to the substrate temperature of 120 ° C., 150 mW / cm ² , pressure of 100 Pa, SiH ₄ flow rate: 50 sccm, H ₂ flow rate: 200 sccm. The film thickness was 1500 mm and the change in the film thickness direction of the hydrogen content in the film was evaluated by SIMS. The result is shown in FIG. From FIG. 10, compared with the conventional example shown in FIG. 2, generation | occurrence | production of the high hydrogen content content electric power generation layer in the interface area | region with the base layer of the electric power generation layer of a photovoltaic apparatus was able to be suppressed.
[0045]
Furthermore, the results of normalization by the characteristics when a photovoltaic device is formed and the structure in which the hydrogen profile is evaluated in FIG. 2, that is, when a power generation layer is formed on a homogeneous p layer at an RF power of 150 mW / cm ² are shown. 4 shows.
[0046]
[Table 4]

[0047]
From Table 4, the initial F.D. F. The conversion efficiency was greatly improved, and the post-light degradation efficiency was improved by about 16% by reducing the light degradation rate. It can be considered that the increase in Isc is larger than that in the embodiments of Tables 2 and 3 due to the reduction in light absorption loss due to the wide gap due to hydrogen implantation into the bulk of the p layer.
[0048]
Here, the hydrogen concentration profile on the photovoltaic device substrate of the photovoltaic device evaluated in Tables 2 to 3 and the photovoltaic device evaluated in FIG. 2 is compared and shown in FIGS. FIG. 11 shows the hydrogen profile evaluated by SIMS in the photovoltaic device structure under the conventional conditions evaluated in FIG. 2, and FIG. 12 shows the conditions of the embodiment of the present invention evaluated in FIG. Although the solar dragon pond uses an uneven substrate, it is considered that the accuracy of the hydrogen profile in the depth direction is inferior. However, when the embodiment of the present invention is used as compared with the conventional example, it is near the p / i interface. The suppression effect of the high hydrogen content power generation layer on the i side was obvious.
[0049]
Next, a substrate temperature of 120 ° C., 450 mW / cm ² , a pressure of 300 Pa, a SiH ₄ flow rate of 50 sccm, a H ₂ flow rate of 20000 sccm, and a microcrystalline p layer formed by 2% boron doping, a substrate temperature of 120 ° C. and 450 mW / cm. ^2, pressure 300 Pa, SiH ₄ flow rate: 50 sccm, H ₂ flow rate: the results of the change in the film thickness direction of 1500Å formed Makuchu hydrogen amount microcrystalline i layer formed was evaluated by SIMS at 20000sccm 13. Even in the case of microcrystals, a high hydrogen region was present at the initial stage of film formation, and a-Si: H was confirmed from a cross-sectional TEM photograph.
[0050]
Furthermore, a film by SIMS when a microcrystalline power generation layer is formed under the above conditions after performing H ₂ plasma treatment with RF power of 300 mW / cm ² , pressure of 100 Pa, H ₂ flow rate: 1000 sccm, which is an embodiment of the present invention The evaluation result of the amount of medium hydrogen is shown in FIG.
[0051]
From FIG. 14, it became clear that generation | occurrence | production of the high hydrogen content containing electric power generation layer seen by FIG. 13 is suppressed.
[0052]
Further, Table 5 shows the result of comparing the characteristics of the photovoltaic device. Table 5 is a value obtained by normalizing the solar cell characteristics of the embodiment of the present invention with the solar cell characteristics of the conventional example. From Table 5, it was confirmed that the present invention is effective in the microcrystalline Si power generation layer as in the case of the a-Si: H power generation layer.
[0053]
[Table 5]

[0054]
As is clear from the above, in the pin type photovoltaic device according to the present invention, the amount of hydrogen in the film near the surface of the underlayer forming the power generation layer is relatively smaller than the amount of hydrogen in the underlayer bulk. By reducing the amount of hydrogen in the power generation layer containing the high hydrogen content in the interface region with the base layer of the power generation layer of the photovoltaic device due to the initial discharge of plasma CVD, the hydrogen is diffused into the low hydrogen content base layer. It has been clarified that generation of a quantity-containing power generation layer can be suppressed and a highly efficient amorphous photovoltaic device can be obtained before and after photodegradation.
[0055]
In the above-described embodiment, the case where the amount of hydrogen in the vicinity of the surface of the p layer serving as the base is reduced compared to the bulk when forming the i layer directly on the p layer has been described. Similarly, when a buffer layer is formed on the p layer and an i layer is formed on the buffer layer, the amount of hydrogen in the vicinity of the surface of the interface with the i layer of the buffer layer serving as the base is reduced as compared with the bulk. Of course, the same effect can be obtained.
[0056]
Further, in the above-described embodiment, the case where the present invention is applied to a photovoltaic device having a single pin structure has been described. Similarly, the present invention can be applied to a so-called tandem photovoltaic device in which several pin-structure semiconductor layers are stacked. That is, the amount of hydrogen in the vicinity of the surface of the layer serving as the base of the i layer may be smaller than that of the bulk.
[0057]
In the above-described embodiment, the case of a photovoltaic device having a pin structure has been described. However, the present invention can also be applied to a photovoltaic device having a nip type structure. In the case of this nip structure, the amount of hydrogen in the vicinity of the surface of the n layer serving as the base may be reduced compared to the bulk. In this way, by reducing the amount of hydrogen in the vicinity of the surface of the n layer serving as the base as compared with the bulk, the hydrogen-containing profile of the power generation layer is flattened, and the sensitivity on the long wavelength side can be increased.
[0058]
【The invention's effect】
As is clear from the above, according to the present invention, the amount of hydrogen in the film near the surface of the underlayer that forms the power generation layer is relatively reduced compared to the amount of hydrogen in the bulk of the underlayer. Generation of a high hydrogen content power generation layer by diffusion of hydrogen in the high hydrogen content power generation layer to the low hydrogen content base layer in the interface region between the power generation layer of the photovoltaic device and the base layer due to the initial discharge of CVD And a highly efficient amorphous photovoltaic device can be obtained.
[Brief description of the drawings]
FIG. 1 shows a case where a power generation layer having a thickness of 1500 mm is formed with a substrate temperature of 120 ° C., a pressure of 100 Pa, an SiH ₄ flow rate of 50 sccm, an H ₂ flow rate of 200 sccm and an RF power during film formation of a power generation layer of 50 mW / cm ² . It is a characteristic view which shows the depth direction distribution of the hydrogen concentration evaluated by the secondary ion mass spectrometry (SIMS) of the base p layer and the electric power generation layer vicinity.
FIG. 2 shows a case where a power generation layer having a thickness of 1500 mm is formed with a substrate temperature of 120 ° C., a pressure of 100 Pa, an SiH ₄ flow rate of 50 sccm, an H ₂ flow rate of 200 sccm and an RF power during film formation of a power generation layer of 150 mW / cm ² . It is a characteristic view which shows the depth direction distribution of the hydrogen concentration evaluated by the secondary ion mass spectrometry (SIMS) of the base p layer and the electric power generation layer vicinity.
[3] The initial 50Å was formed by the power density 25 mW / cm ² on the p layer, then characteristic diagram showing the hydrogen concentration evaluation results by SIMS in the case of forming the power generation layer of the remaining 1450Å at 50 mW / cm ² It is.
FIG. 4 shows a substrate temperature of 120 ° C., RF power of 150 mW / cm ² , pressure of 100 Pa, SiH ₄ flow rate: 50 sccm, H ₂ flow rate is changed from 0 to 5000 sccm, and hydrogen dilution rate (H ₂ / SiH ₄ ) is changed. It is a characteristic view which shows the change of the hydrogen amount in the film | membrane of p layer at the time of making it change.
5 is formed on the underlayer shown in FIG. 4 by forming a power generation layer with a substrate temperature of 120 ° C., 150 mW / cm ² , pressure of 100 Pa, 8iH ₄ flow rate: 50 sccm, H ₂ flow rate: 200 sccm, and a film thickness of 150 mm. It is a characteristic view which shows the result of having evaluated the change of the film thickness direction of hydrogen amount by SIMS.
6 is formed on the underlayer shown in FIG. 4 by forming a power generation layer with a substrate temperature of 120 ° C., 150 mW / cm ² , pressure of 100 Pa, SiH ₄ flow rate: 50 sccm, H ₂ flow rate: 200 sccm and a film thickness of 1500 μm. It is a characteristic view which shows the result of having evaluated the change of the film thickness direction of quantity by SIMS.
FIG. 7 shows that after forming the underlayer, the surface is subjected to Ar plasma treatment at an RF power of 300 mW / cm ² , a pressure of 100 Pa, and an Ar flow rate of 1000 sccm, and an Ar plasma treatment time and a film in a region of 50 mm from the surface. It is a characteristic view which shows the relationship of hydrogen amount.
FIG. 8 shows that the power generation layer is formed on the base layer shown in FIG. 7 with a substrate temperature of 120 ° C., 150 mW / cm ² , pressure of 100 Pa, SiH ₄ flow rate: 50 sccm, H ₂ flow rate: 200 sccm and a film thickness of 1500 Å. It is a characteristic view which shows the result of having evaluated the change of the film thickness direction of this by SIMS.
FIG. 9 shows that after forming the underlayer, the surface is subjected to hydrogen plasma treatment at an RF power of 300 mW / cm ² , a pressure of 100 Pa, and a hydrogen flow rate of 1000 sccm. It is a characteristic view which shows the relationship of hydrogen amount.
10 is formed on the base layer shown in FIG. 9 by forming a power generation layer with a substrate temperature of 120 ° C., 150 mW / cm ² , pressure of 100 Pa, SiH ₄ flow rate: 50 sccm, H ₂ flow rate: 200 sccm and a film thickness of 1500 Å. It is a characteristic view which shows the result of having evaluated the change of the film thickness direction of this by SIMS.
FIG. 11 is a hydrogen profile evaluated by SIMS in the photovoltaic device structure under the conventional conditions evaluated in FIG.
12 is a hydrogen profile evaluated by SIMS in a photovoltaic device structure under the conditions of the embodiment of the present invention evaluated in FIG.
FIG. 13 shows a substrate temperature of 120 ° C., 450 mW / cm ² , a pressure of 300 Pa, a SiH ₄ flow rate of 50 sccm, a H ₂ flow rate of 20000 sccm, and a microcrystalline p layer formed by 2% boron doping. Characteristic diagram showing the result of evaluating the change in the film thickness direction of the amount of hydrogen in the film by SIMS after forming 1500 μm of microcrystalline i layer formed at cm ² , pressure 300 Pa, SiH ₄ flow rate: 50 sccm, H ₂ flow rate: 20000 sccm It is.
FIG. 14 shows the amount of hydrogen in a film by SIMS when a microcrystalline power generation layer was formed after performing an H ₂ plasma treatment on an underlayer at an RF power of 300 mW / cm ² , a pressure of 100 Pa, and an H ₂ flow rate: 1000 sccm. It is a characteristic view which shows an evaluation result.
FIG. 15 is a cross-sectional view showing the structure of a pin-type a-Si photovoltaic device having a pin junction.
[Explanation of symbols]
1 glass substrate 2 transparent electrode 3 p-type a-SiC layer 4 i-type a-Si layer (power generation layer)
5 n-type microcrystalline silicon 7 Back side metal electrode

Claims (2)

In a method for manufacturing a photovoltaic device, in which a photoelectric conversion layer made of a thin film semiconductor having a semiconductor junction therein is formed by chemical vapor deposition,
Hydrogen is added to the surface of the base semiconductor layer on which the power generation layer of the photoelectric conversion layer is formed so that the concentration of hydrogen in the film near the surface is relatively smaller than the hydrogen concentration in the bulk of the base semiconductor layer. Applying plasma treatment with gas;
Forming a power generation layer on the surface where the concentration of hydrogen is reduced;
And forming a power generation layer having a certain amount of hydrogen concentration profile .   The method for manufacturing a photovoltaic device according to claim 1, wherein the photoelectric conversion layer has a pin junction, and light is incident from the p-type semiconductor layer side.

Images