283253 Life Cycle Assessment of Thin-Film CdTe Photovoltaics

Friday, November 2, 2012: 9:42 AM
327 (Convention Center )
Yuan Yao and Fengqi You, Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL

Growing concern to adverse health effects and severe atmospheric pollution caused by the operation of fossil-fuel-burning power plants has fostered the research on more sustainable energy resources. Solar energy is one of the most promising alternatives for large-scale electricity production because of its abundance.[1] In the past 12 years, photovoltaics (PV) experienced an average growth of approximately 45% per year that was affected only slightly by the recent global financial crisis.[2] Among a variety of photovoltaic technologies, thin film is relatively new but developing rapidly for low cost, large flexibility[3] and small amount of materials used in modules. CdTe is one of the thin film solar cells that have the great potential to be produced in large scale with low cost.[4] However, any energy production, no matter how “green” they look like, will lead to environmental impacts in their entire life cycle.[5] Thus, it is necessary to apply Life Cycle Assessment (LCA) tools to CdTe photovoltaics to assess their environmental performance, including GHG and heavy metal emissions. Currently, most assessments are conducted based on the data from OECD countries (e.g., United States, Canada and Germany).[6, 7] However, most of solar cell companies prefer to move their manufacturing plants to non-OECD countries (e.g., China, India and Malaysia) because of low cost and abundant raw materials. Current LCA results of CdTe PV fail to provide decision makers with appropriate references on whether it is environmentally sustainable to have CdTe PV manufactured abroad, especially in none-OECD countries. It is of great importance to conduct comparative LCAs in two cases, where CdTe solar cells are manufactured domestically and oversea respectively.

There are several challenges to conduct this comparative study. The first one is to figure out all possible differences between two cases in each stage of CdTe PV life cycle. Since cadmium and tellurium are byproducts of zinc and copper production, LCAs should start with zinc and copper ore mining. Different grades of ores, efficiencies of metallurgical operation, electricity mixes, and some other factors may contribute to the difference of emission factors of life cycle stages between CdTe PV made in OECD and none-OECD countries. Another challenge is that the data for none-OECD countries are not as accessible as those for OECD members. A comprehensive understanding of material flows and emissions, and a deep data digging for non-OECD countries in all stages are required. 

In this paper, a “Cradle to Gate” LCA, which includes raw material exploitation, material processing and PV manufacturing, of CdTe PV in two case studies is carried out. In the first case study, China is chosen as a representative of none-OECD countries. A CdTe PV manufacture plant is built in Sichuan Province where the first Chinese CdTe PV manufacturing line was located, [8] and CdTe solar panels are shipped to venders in the United States. In the second case, a local plant is built in Chicago, United States. The geographically representative data mainly come from industrial reports, literatures and Gabi.[9] GHG and heavy metal emissions in all life cycle stages are estimated in two case studies. Total life-cycle GHG emissions of CdTe PV made in China are 113 kgCO2-eq/m2 and heavy metal emissions are 1.17 mg Cd/m2. While GHG emissions and heavy metal emissions of CdTe PV made in the United States are 74 kgCO2-eq/m2 and 0.37 mg Cd/m2 respectively, which are much smaller than that in China. Emissions of other heavy metals, such as arsenic, lead, mercury and nickel are also estimated in two case studies. In order to integrate heavy metal and GHG emissions to the corresponding environmental impacts, Eco-indicator 99 [10] is applied to two case studies. The total indicator of CdTe PV in the United States is 4.32 points/m2; while the indicator in China is 8.67 points/m2, which indicates that the environmental damages caused by investing CdTe PV manufacturing in China is twice as severe as that in the United States. Thus CdTe PV made in the United States are more sustainable. Energy payback times (EPBT) are calculated based on 10.7% module efficiency, 80% performance ratio, 30 years life time and 1825 kWh/m2/year that is forecast of radiation for Chicago area from National Solar Radiation Data Base by using time-series methods with Oracle Crystal ball.[11]  EPBT for two case studies are very close, 0.7 years for CdTe PV in China and 0.62 years for that in the United States. This suggests that CdTe PV made in the United States spend relatively shorter time to generate the same amount of energy used in total life-cycle of CdTe PV. Based on LCAs results, CdTe PV made in the United States obtain a relatively smaller EPBT with much less environmental damages. Investing CdTe PV plants in non-OECD is not able to get a faster energy payback, while generates more emissions and causes severer environmental damages. Therefore, domestic manufacturing in OECD countries is a more environmentally sustainable approach suitable for CdTe PV.


[1]        S. B. Darling, F. You, T. Veselka, and A. Velosa, "Assumptions and the levelized cost of energy for photovoltaics," Energy & Environmental Science, vol. 4, pp. 3133-3139, 2011.

[2]        V. M. Fthenakis, "Sustainability metrics for extending thin-film photovoltaics to terawatt levels," presented at the MRS Bulletin, 2012.

[3]        K. L. Chopra, P. D. Paulson, and V. Dutta, "Thin-film solar cells: An overview," Progress in Photovoltaics, vol. 12, pp. 69-92, 2004.

[4]        K. Kato, T. Hibino, K. Komoto, S. Ihara, S. Yamamoto, and H. Fujihara, "A life-cycle analysis on thin-film CdS/CdTe PV modules," Solar Energy Materials and Solar Cells, vol. 67, pp. 279-287, 2001.

[5]        V. M. Fthenakis and E. Alsema, "Photovoltaics energy payback times, greenhouse gas emissions and external costs: 2004–early 2005 status," Progress in Photovoltaics: Research and Applications, vol. 14, pp. 275-280, 2006.

[6]        V. M. Fthenakis, W. M. Wang, and H. C. Kim, "Life cycle inventory analysis of the production of metals used in photovoltaics," Renewable & Sustainable Energy Reviews, vol. 13, pp. 493-517, 2009.

[7]        V. M. Fthenakis and H. C. Kim, "Photovoltaics: Life-cycle analyses," Solar Energy, vol. 85, pp. 1609-1628, 2011.

[8]        (Apr. 10th ). Chengdu COE Apollo Solar Co ., Ltd. Available: http://www.coeapollo.com/en/about.asp

[9]        Gabi, "Software and database contents for Life Cycle Engineering.," 2011 ed. Stuttgart: PE INTERNATIONAL AG.

[10]      M. Goedkoop, R. Spriensma et al, "The Eco-indicator 99  a damage oriented method for life cycle impact assessment. Methodology report.," PRé-Consultants, Amersfoort2001.

[11]      (24th, March). Oracal Crystal Ball. Available: http://www.oracle.com/us/products/applications/crystalball/index.html

Extended Abstract: File Not Uploaded
See more of this Session: Sustainability Metrics At the Process and Product Level
See more of this Group/Topical: Environmental Division