(387c) Hybrid Life Cycle Assessment Model of Silicon Photovoltaics
Research on sustainable energy resources is receiving increasing attention because of growing concern about severe pollution associated with fossil-fuel power plants. Solar energy is one of the most promising alternatives. Photovoltaic (PV) technology is gradually becoming competitive in power generation market for its abundant resources and affordable electricity. Among all solar technology, silicon PV technology has been successfully commercialized and has the greatest market share. However, any energy production will also consume energy, generate pollutants and result in some environmental impacts . In order to assess the energy sustainability and environmental impacts of silicon PVs, a lot of work has been done on applying Life Cycle Assessment (LCA) tools to solar energy. Currently, the majority of assessments adopt process-based LCA which is a bottom-up method that is limited by system boundary variations, difficulty for replicating results, and inability to provide assessments for future products due to frequent use of proprietary data. Constrained by incomplete system boundary, results of process-based LCA studies often fail to evaluate the entirety of energy and environmental impacts of solar technology on the whole society and ecosystem. Another tool for LCA studies is Economic Input-output LCA (EIO-LCA) model which has a more comprehensive system boundary, allows for comparisons in systems-level and capacity of assessing future product developments. However, EIO-LCA has time lag for current practices, highly depends on government statistical data, and suffers from sectorial aggregations.
In this paper, a hybrid LCA model for silicon PVs, which takes advantage of both process-based method and EIO-LCA, was developed. This hybrid model expanded the system boundary to all supply chain materials of silicon PV by matrix operation (e.g. usage of petroleum coke to produce silicon carbide for wire sawing, electricity generation and usage of all upstream materials). Another strength of this hybrid LCA model is that the geographical shift of PV manufacturing, such as expanding Chinese silicon PV production, can be easily estimated by using publicly accessible input-output data of that country. Since silicon PV has been commercialized for several decades, the variation of technology parameters caused by geographic shift is not significant. However, the difference of sustainability performance of PV made in different countries can still be considerable because of diverse economic structure and energy mix. This difference can be captured and assessed by this hybrid LCA model.
For model demonstration, a case study of U.S. multicrystalline PV production was conducted. Results show that the GHG emissions and energy intensity are 41 g CO2-eq/kWh (based on 1,800 kWh/m2/year, 14% module efficiency, 0.75 performance ratio and 30 years lifetime) and 5811 MJ/m2 (balance of system is included), respectively. The Energy Payback Time (EPBT) is calculated to be 2.54 year. Results of this study are 30 – 60% higher than previous studies using process-based LCA model after harmonization. This is caused by the comprehensive boundary of hybrid model, covering more upstream materials in the supply chain and thus more GHG emissions and energy considerations in research system. Sensitivity analysis was conducted to filter dominant factors influencing the sustainability of multicrystalline PV made in the U.S. from a more comprehensive view, which can provide useful recommendations for future research. Given that this model also shows specific energy consumption and GHG emissions of all sectors in U.S. economy caused by multicrystalline PV production and operation, study findings provide policy maker and general public with a more holistic view of the impacts of PV technology over the whole society and ecosystem. More case studies for silicon PV made in other countries, such as China, will be shown in this paper to give a comparison and show the fluctuation of the sustainability performance of silicon PV caused by geographic shift.
 D. D. Hsu, P. O’Donoughue, V. Fthenakis, G. A. Heath, H. C. Kim, P. Sawyer, et al., "Life Cycle Greenhouse Gas Emissions of Crystalline Silicon Photovoltaic Electricity Generation," Journal of Industrial Ecology, vol. 16, pp. S122-S135, 2012.
 "Solar Generation 6," European Photovoltaic Industry Association (EPIA)2011.
 EPA, "Annual Energy Outlook 2013," U.S. Energy Information Administration2013.
 C. T. Hendrickson, L. B. Lave, and H. S. Matthews, Environmental Life Cycle Assessment of Goods and Services: An Input-Output Approach: RFF Press, 2006.
 Y. Chang, R. J. Ries, and Y. Wang, "The quantification of the embodied impacts of construction projects on energy, environment, and society based on I–O LCA," Energy Policy, vol. 39, pp. 6321-6330, 2011.