本篇論文之實驗是利用電子迴旋共振化學氣相沉積系統(ECRCVD)成長微晶矽薄膜，探討其薄膜性質，並應用在p-i-n型薄膜太陽能電池之本質主動層。ECRCVD系統具有的優點為工作壓力低（10-4~10-1torr）、蒸鍍溫度低、離子能量低、氣體解離度高、電漿密度高（1011~1013cm-3）、可獨立控制離子通量與離子能量及對基板低離子轟擊等，無需電極以避免其他雜質汙染。實驗中利用調變製程參數如製程氣體流量、氣體比例、磁場線圈電流大小、製程腔體壓力等探討微晶矽薄膜之特性。
由實驗結果得知，氫氣稀釋比、製程壓力為影響薄膜結晶率最主要因素，於ECRCVD製程中，結晶率極限約80%。利用提高氣體總流量，可提升微晶矽薄膜沉積速率最高至11.2Å/sec，高於傳統RF-PECVD系統約只有1~3 Å/sec許多，且於ECRCVD系統中，微晶矽製程不需大量氫氣稀釋即可獲得結晶相之薄膜，氫氣稀釋比最低約2.4，遠低於PECVD製程所需約20~50以上。在應用於太陽能電池本質層中，以薄膜結晶率為35%之電池效率最佳，效率為0.95%，可見微晶矽薄膜在微晶相接近非晶相之過渡區，最為適合做為太陽能電池之本質層，然而其效率稱不上優良，應為氫電漿密度過高，易侵蝕TCO(透明導電膜)基板導致透光率不足並在沉積p-i-n各層時造成各層界面的損傷。

Experiments of this paper is the use of electron cyclotron resonance chemical vapor deposition system (ECRCVD) for microcrystalline silicon thin film growth to explore its film properties and application in pin-type thin-film solar cell intrinsic active layer. ECRCVD system has the advantages that low working pressure (10-4~10-1torr), low evaporation temperature, low ion energy, high gas dissociation and plasma density (1011~1013cm-3), low ion bombardment of the substrate, and ions flow as well as ion energy can be separately controlled. In experiments we use modulation process parameters such as process gas flow, gas ratio, the magnetic field coil current and the process chamber pressure to investigate the characteristics of microcrystalline silicon thin films.
According to the experimental results, the hydrogen dilution ratio, process pressure is the important factor affecting the film crystallization rate. And we found that the crystallization rate limit is about 80%. By raising the total gas flow rate, we can enhance the microcrystalline silicon thin film deposition rate up to 11.2Å/sec. And in ECRCVD system, we could obtain crystalline phase without large amount of hydrogen dilution. The hydrogen dilution ratio minimum is about 2.4, well below the PECVD process required that is about 20 to 50 or more. For intrinsic layers used in solar cells, as the film crystallinity is 35%, we can obtain the cell efficiency of 0.95%, showing that the mix of microcrystalline and amorphous phase would be the most suitable for the intrinsic layer. Yet its efficiency cannot be good due to the high hydrogen plasma density. It corrodes TCO substrate and results in insufficient light transmittance caused by layers interface damage.

[1] Sterling, H. F. and R. C. G. Swann, Solid-State Electron. 8, 653 (1965)
[2] Veprek, S., Marecek, V., ‘‘The preparation of thin layers of Ge and Si by chemical hydrogen plasma transport’’ (1986), Solid State Electronics, Vol. 11, pp. 683-684.
[3] Usui, S., Kikuchi, M., ‘‘Properties of heavily doped gd-Si with low resistivity’’ (1979), Journal of Non-Crystalline Solids, Vol. 34, pp. 1-11.
[4] M. Tsuda, S. Oikawa, K. Saito, J. Chem. Phys. 91 (1989) 6822.
[5] J. Perrin, O. Leroy, M.C. Bordage, Contrib. Plasma Phys. 36(1996) 3.
[6] Yamamoto, K., Materials Research Society Symp. Proc. 507 (1998)
[7] Wagner, H., Beyer, W., Solid-State Communications, Vol. 48, pp. 585-587 (1983)
[8] Bhattacharya, E., Mahan, A. H., Applied Physics Letters, Vol. 52, pp. 1587-1589 (1988)
[9] Langford, A. A., M. L. Fleet, B. P. Nelson, W.A. Lanford, and N. Maley, Physical Review B, Vol.45, 13367 (1992)
[10] Tauc, J., in: Optical properties of Solids, ed. by F. Abeles, p.277 (1972)
[11] Schropp, R. E. I., and M. Zeman, Amorphous and Microcrystalline Silicon Solar Cells, Kluwer Academic Publishers, Boston, P.47 (1998)
[12] Akihisa Matsuda, Journal of Non-Crystalline Solids, Vol. 338, pp. 338-340 (2004)