臺灣博碩士論文加值系統

溼熱老化試驗(Damp Heat Test, DH)是國際認證測試IEC 61215其中一項測試，透過將模組放在85℃/85%相對濕度下1000小時，藉此確保模組在戶外對溫度和濕度的抵抗能力，此外還能分辨何種設計的模組較能抵抗溫度與濕度的影響。然而近年來，由於製程與封裝技術進步，太陽能模組能夠輕易通過DH 1000小時試驗並且無任何衰減，需要延長測試時間達到數千小時，才能看到模組差異，相當耗費時間。因此太陽能產業需要更為嚴苛的試驗，使能夠在短時間內評估不同設計或是材料的模組可靠度，例如壓力鍋試驗(Pressure Cooker Test, PCT)。PCT過去應用在測試IC封裝，透過比DH更高的溫度與溼度，再加上超過一大氣壓的壓力，這會強迫水氣侵入至樣品，造成樣品快速失效，但是過去PCT並沒有應用在測試太陽能模組可靠度。因此本論文希望建立合適的PCT測試條件。在第一部份探討模組在PCT環境下失效機制，並比較是否與DH相同。在第二部分，將實驗數據套用在佩克模型(Peck Model)，並用此模型預測樣品在不同環境壓力下的功率表現。由於壓力鍋機台大小無法放入一般太陽能模組，因此本實驗將多片太陽能電池封裝成單片太陽能模組，並在外部包上鋁箔代替鋁框，以模仿實際模組情況。之後分別放置在DH 和PCT環境下長時間試驗，並透過電致發光影像(Electroluminescence, EL)、光致發光影像(Photoluminescence, PL)、掃描式電子顯微鏡(SEM)和能量色散X-射線光譜(EDX)分析失效模式。在這研究中，我們發現PCT的失效原因是由於傳導金屬銀氧化造成，在高溫與高濕，配合上封裝材料乙烯/醋酸乙烯酯共聚物(Ethylene-vinyl acetate, EVA)水解所產生的醋酸催化的情況下，銀電極形成銀氧化物，造成電子收集困難，串聯電阻上升，最後導致功率下降，此情形與DH測試下的樣品失效模式相同。
在本論文的PCT條件中，以PCT121℃/100%相對溼度的環境條件最為適合區分不同模組對於濕熱的可靠度。藉由修改後的派克模型，我們成功描述單片模組在DH和PCT條件試驗下的功率變化，並且可以預估樣品在不同溫度、濕度下的功率隨時間表現，透過模型計算，PCT121℃/100%衰減速率比DH的14.8倍。

Damp Heat (DH) test is part of the qualification tests in IEC 61215. The test is done in a climatic chamber that is kept at 85°C and relative humidity (R.H.) of 85%. It can investigate the PV modules’ resistance to temperature and humidity. Recently, due to the progress in manufacturing processes and packaging technology, most of the PV modules can easily pass the DH 1000 test without degradation. It therefore needs to extend test duration up to thousands of hours in order to qualify the module’s capability to withstand long-term temperature and humidity variation. The process becomes very time consuming. As a result, more stringent accelerated aging protocol, such as Pressure Cooker Tests (PCT) has been proposed. With the temperature more than 105 centigrade and 100% relative humidity, the pressure in the chamber is more than 1 atm. It can force moisture to penetrate into the sample, and shorten the testing time. However, PCT has not been used for the PV industry. Therefore, we hope to establish standard testing procedures of PCT for PV modules. In the first part of this work, we study the failure modes of PCT to ensure it is similar to DH. In the second part, we modified the Peck model, and use this model to evaluate the PV module performance under different environment conditions.
Because the PCT in the past is used on IC package test, the size of PCT chamber is not big enough for PV modules. We have to make single-cell PV modules instead. Unlike regular solar modules, single-cell PV modules do not include the Al frame. We therefore replace it by the Al foil. After preparation, the samples are aged in different environments. We study the failure mechanism of samples using IV test, Electroluminescence (EL) and Photoluminescence (PL) images, scanning electron microscope (SEM), and Energy-dispersive X-ray spectroscopy (EDX).
We found that the failure mechanism of PCT is due to the formation of silver oxide, enhanced by the high temperature, humidity and acetic acid. Acetic acid is produced by EVA at high temperature and humidity. The formation of silver oxide impedes the electron collection, leading to power degradation. Similar failure mechanism has also been found in the DH test. Besides, our result suggests the 121 centigrade condition of the PCT test is most appropriate to test the reliability of modules. Furthermore, by modifying the Peck model, we can describe the power degradation curves induced by the DH and PCTs successfully. The modified Peck model suggests the degradation rate of PCT 121 is 14.8 times faster than the DH degradation rate.

摘要 II
ABSTRACT IV
致謝 VI
目錄 VII
圖目錄 VIII
表目錄 X
第一章 緒論 1
1.1 太陽能發展背景與成本 1
1.2 加速老化測試和認證測試介紹 3
1.3 研究動機與目的 7
第二章 研究理論基礎與量測分析技術 12
2.1 太陽能電池架構與工作原理(SOLAR CELL , PHOTOVOLTAIC CELL) 12
2.2 太陽能模組(SOLAR MODULE, PHOTOVOLTAIC MODULE) 13
2.3 濕熱測試(DAMP HEAT TEST) 16
2.4 壓力鍋測試(PRESSURE COOKER TEST, PCT) 17
2.5 阿瑞尼斯模型(ARRHENIUS MODEL) 18
2.6 派克模型(PECK MODEL)[23] 18
2.7 量測分析技術 19
第三章 實驗流程與實驗儀器操作 25
第四章 實驗結果與分析 28
4.1 濕熱測試與壓力鍋測試失效模式 28
4.2 模型分析 42
第五章 結論與未來展望 48
第六章 參考文獻 49
附錄 52
附錄-A 53
附錄-B 57

[1] T. Kenning. "Solar installations hit 59GW in 2015, 64GW to come in 2016 – GTM," http://www.pv-tech.org/news/global-solar-installations-hit-59gw-in-2015-gtm.
[2] "2015 Taiwan power system installed capacity diagram, Bureau of Energy," https://web3.moeaboe.gov.tw/ECW/populace/content/wHandMenuFile.ashx?menu_id=1002.
[3] "Taiwan power plants status, Bureau of Energy," https://web3.moeaboe.gov.tw/ECW/populace/content/Content.aspx?menu_id=997.
[4] "Cost of electricity in Taiwan, Taiwan Power Company," http://www.taipower.com.tw/content/new_info/new_info-a02.aspx.
[5] K. Zweibel, “Should solar photovoltaics be deployed sooner because of long operating life at low, predictable cost?,” Energy policy, vol. 38, no. 11, pp. 7519-7530, 2010.
[6] R. Ross Jr, and M. Smokler, Flat-Plate Solar Array Project: Final Report: Volume 6, Engineering Sciences and Reliability, Jet Propulsion Lab., Pasadena, CA (USA), 1986.
[7] J. Wohlgemuth, "IEC 61215: What it is and isn’t."
[8] IEC, “IEC 61215, 2nd,” 2005.
[9] G. TamizhMani et al., "Failure analysis of design qualification testing: 2007 vs. 2005." pp. 1-4.
[10] B. Ketola, and A. Norris, “Degradation mechanism investigation of extended damp heat aged PV modules,” 26th EUPVSEC, 2011.
[11] A. Masuda, N. Uchiyama, and Y. Hara, “Degradation by acetic acid for crystalline Si photovoltaic modules,” Japanese Journal of Applied Physics, vol. 54, no. 4S, pp. 04DR04, 2015.
[12] "Ethylene-vinyl_acetate," https://en.wikipedia.org/wiki/Ethylene-vinyl_acetate.
[13] "Ethylene-vinyl_acetate appearance," http://ouyuan.51pla.com/productshow_4187987.htm.
[14] "Dupont backsheet," http://www.dupont.com/.
[15] D. Otth, and R. Ross Jr, "Assessing photovoltaic module degradation and lifetime from long term environmental tests." pp. 121-126.
[16] R. G. Ross, “PV reliability development lessons from JPL's flat plate solar array project,” IEEE Journal of Photovoltaics, vol. 4, no. 1, pp. 291-298, 2014.
[17] "Accelerated Moisture Resistance -Unbiased Autoclave," http://ihome.ust.hk/~epack/member%20download%20area/22a102c.pdf.
[18] "Environment chamber pressue," http://www.espec.co.jp/english/inquiry/catalog/ehs.pdf.
[19] "Arrhenius equation," https://en.wikipedia.org/wiki/Arrhenius_equation.
[20] E. Bumiller, and C. Hillman, “A review of models for time-to-failure due to metallic migration mechanisms,” White paper, DfR Solutions, 2006.
[21] D. S. Peck, "Comprehensive model for humidity testing correlation." pp. 44-50.
[22] "Energy-dispersive X-ray spectroscopy," https://en.wikipedia.org/wiki/Energy-dispersive_X-ray_spectroscopy.
[23] E. Wang et al., “Determination of Acetic Acid Content in Ethylene-Vinyl Acetate (EVA) Based PV Modules ”.
[24] A. Kraft et al., “Investigation of Acetic Acid Corrosion Impact on Printed Solar Cell Contacts,” IEEE Journal of Photovoltaics, vol. 5, no. 3, pp. 736-743, 2015.
[25] T. H. Kim, N. C. Park, and D. H. Kim, “The effect of moisture on the degradation mechanism of multi-crystalline silicon photovoltaic module,” Microelectronics Reliability, vol. 53, no. 9, pp. 1823-1827, 2013.
[26] H. Gong, G. Wang, and M. Gao, "Degradation mechanism analysis of damp heat aged PV modules." pp. 3518-3522.
[27] J. S. S. T. Association, “Failure mechanisms and models for semiconductor devices,” JEDEC Publication JEP122-B, 2003.
[28] O. Tebbi, F. Guerin, and B. Dumon, "Comparative study of Accelerated Testing models, applications in Mechanics." pp. 2099-2104.
[29] F. Dia et al., “Model Associated with the Study of the Degradation Based on the Accelerated Test: A Literature Review,” Open Journal of Applied Sciences, vol. 6, no. 01, pp. 49, 2016.
[30] B. Braisaz et al., “PV Aging Model Applied to Several Meteorological Conditions,” 29th EU PVSEC.
[31] S. Y. Gregory M. Kimball, Ajay Saproo, “Global acceleration factors for damp heat tests of PVC modules,” IEEE 39th Photovoltaic Specialists Conference, 2016.