You are on page 1of 8

Article pubs.acs.


Energy Balance of the Global Photovoltaic (PV) Industry - Is the PV Industry a Net Electricity Producer?
Michael Dale*,† and Sally M. Benson‡
† ‡

Global Climate and Energy Project, Stanford University, United States, and Energy Resources Engineering, Stanford University, United States
S Supporting Information *

ABSTRACT: A combination of declining costs and policy measures motivated by greenhouse gas (GHG) emissions reduction and energy security have driven rapid growth in the global installed capacity of solar photovoltaics (PV). This paper develops a number of unique data sets, namely the following: calculation of distribution of global capacity factor for PV deployment; meta-analysis of energy consumption in PV system manufacture and deployment; and documentation of reduction in energetic costs of PV system production. These data are used as input into a new net energy analysis of the global PV industry, as opposed to device level analysis. In addition, the paper introduces a new concept: a model tracking energetic costs of manufacturing and installing PV systems, including balance of system (BOS) components. The model is used to forecast electrical energy requirements to scale up the PV industry and determine the electricity balance of the global PV industry to 2020. Results suggest that the industry was a net consumer of electricity as recently as 2010. However, there is a >50% that in 2012 the PV industry is a net electricity provider and will “pay back” the electrical energy required for its early growth before 2020. Further reducing energetic costs of PV deployment will enable more rapid growth of the PV industry. There is also great potential to increase the capacity factor of PV deployment. These conclusions have a number of implications for R&D and deployment, including the following: monitoring of the energy embodied within PV systems; designing more efficient and durable systems; and deploying PV systems in locations that will achieve high capacity factors.

INTRODUCTION The amount of sunlight energy that flows through an area the size of Connecticut (15,000 km2) each second is roughly equal to the average energy demand of the whole of human society (15 TW). Given the magnitude of the solar energy flow, the direct conversion of sunlight to electricity via photovoltaic (PV) systems provides an opportunity for human society to meet a significant amount of its energy needs from a low carbon, renewable source. The global installed capacity of PV grew at an average rate of 40% per year in the period 2000−2010.1−3 Global installed capacity in 2010 was around 40 GW, split between six main PV technologies: single-crystal (sc-Si) and multicrystalline (mc-Si) silicon (which together comprise around 90% of the market), amorphous silicon (a-Si), ribbon silicon, cadmium telluride (CdTe), and copper indium gallium diselenide (CIGS) as shown in Figure 1. Growth has been particularly rapid (over 100% per year in the period 2008−2010) in thin film technologies, especially CdTe and CIGS. Because PV-generated electricity emits few greenhouse gases (GHG) during its lifetime operation, governments worldwide promote PV as a GHG emissions reduction strategy (other reasons include energy security, health, and other environmental concerns). Many governments have policies that
© 2013 American Chemical Society

support renewable energy technologies through mandated proportions of renewable electricity generation, such as California’s Renewables Portfolio Standard (Senate Bill X102) of 33% by 2020;4 financial incentives for system installation, such as the California Solar Initiative;5 or guaranteed prices paid on electricity generated and sold back to the grid, such as the Feed-In Tariffs Scheme (FITs) in Germany and the UK.6 Such policies are aimed at providing a market for PV systems to stimulate production and drive down financial costs of PV deployment. The success of these policies is typically judged by monitoring the financial cost of PV modules and the installed capacity of PV systems, not by their success at reducing GHG emissions. Primary energy ‘consumption’, with associated GHG emissions, occurs throughout the full life-cycle of a PV system from extraction of raw materials, through panel and BOS manufacture, system operation and maintenance, and finally disposal. This life-cycle primary energy consumption is termed the cumulative energy demand (CED).7 The time taken for a
Received: Revised: Accepted: Published:

September 25, 2012 February 15, 2013 February 26, 2013 February 26, 2013 | Environ. Sci. Technol. 2013, 47, 3482−3489

r [%/yr] of an energy production industry based on deployment of that technology as f = rEPBR = r (CED)/(Ė g). The term energy subsidy is used to define this energy input. A rapidly growing industry may deploy new systems before previous deployment has ‘paid off’ its initial investment. This article first explores the general relationship between energy inputs and energy production at the system level and the energy balance of an industry with installed capacity of a number of such systems.13 While it is clear that individual PV systems can generate a positive net energy yield. both the timing and magnitude of energy inputs and output are important to determine the energy balance of the whole industry. In order to track this evolution requires a dynamic perspective that goes beyond the static measures often 3483 ENERGY BALANCE AND INDUSTRY GROWTH Inputs and Output from a Single Energy System and an Energy Industry. Figure 2 depicts energy flows into and out of a single energy production device or system. requiring a total energy input to construction of Econ. An energy flow.e.e. A model is presented to track the change in electrical energy costs of PV system deployment and is applied to historic data for the PV industry between 2000 and 2010. Compiled using data from refs 1−3. A number of useful definitions are outlined in Table 1. Ė op. both in terms of growing installed capacity and decreasing energy inputs to PV system deployment. is required for decommissioning. energy flows are levelized over the full lifecycle of the energy system. thereby driving the whole industry into deficit. If f < 100%. Grimmer (1981) defines the relationship between the fractional reinvestment. Construction begins at time. the present analysis focuses on the industry-scale issue by addressing the question “Is the global PV industry a net electricity producer?”. copper indium gallium diselenide (CIGS).e. multicrystalline silicon (mc-Si). or (3) some combination of (1) and (2). to distinguish from the energy investment at the system level. An industry composed of individual systems. implications for R&D and technology deployment are discussed.20 Two ratios used to characterize the scale of energy inputs and outputs at the system level are the EPBT and EROI (both defined in Table 1). Ė con. Ed. Gross and net energy output from this industry growing logistically up to some limit are plotted in Figure 3. This burden may even stifle growth of the PV industry. Figure S1 shows the relationship between EPBT. consumes more energy than it produces. 3482−3489 ■ employed in PV technology assessment. could still be a net energy consumer. industry growth rate.1021/es3038824 | Environ. is required to operate and maintain the system. ribbon silicon. The PV industry is changing rapidly.with energy produced by the system. some energy. then the industry is a net energy sink. At the end of the system lifetime. Growth of the industry normally proceeds exponentially at rate r. i. If the industry grows too quickly. and the growth rate. A fractional reinvestment. (2) decrease the rate of growth. Hence. the industry must eventually deliver more energy to society than it consumes. Turning instead to the industry-scale. 47. and other (mainly organic/polymer-based systems). there is a large benefit to the PV industry in reducing energetic costs of PV deployment. it must produce a positive net energy yield. Technol.11 A rapidly growing PV industry may even temporarily exacerbate the problem of GHG emissions if operating at a net energy loss. Rapid exponential growth of dx. comparing energy inputs . i. Ė d. Flows may be in either primary energy or electricity but must be consistent between inputs and outputs. such that r ≥ (Ė g)/(Ė con). system to ‘pay back’ its CED is termed the energy payback time (EPBT). when it grows rapidly. Energy must be imported from outside the industry. f = 100%. f [%]. and 15. Global cumulative installed capacity [GW] of PV disaggregated by the major technologies: single crystal silicon (scSi).14. Once the project starts producing energy it is assumed to produce a constant gross flow of energy at rate Ė g over the whole lifetime tL. cadmium telluride (CdTe). To compute these ratios. marks the breakeven threshold.investment .org/10. assumed to flow at a constant rate. 2013. If f > 100%. each with a positive net energy contribution over their lifetimes. Raw data for this figure are provided in Table S1. This model is then used to forecast electrical energy costs of future PV system deployment in order to determine the gross and net electricity production of the PV industry to 2020 to determine when the PV industry will become a net electricity provider.12 There has been a long-standing myth that PV systems are unable to generate a positive net energy return over their lifetime. i.doi. Issues with definition and usage of EPBT are discussed in refs 8−10. If an industry is a net energy sink. the industry is a net energy sink and goes into deficit (see SI Section Derivation of breakeven threshold conditions). This relationship is then explored for the specific case of the global PV industry by looking at the energy consumed in manufacture and installation of PV systems using an analogy of the CED metric. .21 Using this relationship. An energy production industry is based on sequential installation of many of these energy production systems. If solar PV is to make an important contribution to the global energy system. t−tc. consuming more energy each year in manufacturing and installing PV systems than is produced by systems in operation. then there are three means by which it may cross the breakeven threshold: (1) decrease the EPBT of system production by either reducing CED or by increasing the energy output of the system. again assumed to flow at a constant rate. then this means the industry is a net energy producer. and fractional reinvestment.Environmental Science & Technology Article Figure 1. Finally. Sci. the effects of its being a net energy consumer or provider are negligible. Energy technologies with long EPBT (due to high CED or low capacity factor) may constrain industry growth and rapid transition to a low GHG emission energy sector. When the PV industry is a small proportion of the overall energy sector. the EPBT [yrs] of a technology. Previous research has addressed the material constraints on PV system deployment. As the industry grows the burden on the overall energy sector becomes greater. amorphous silicon (a-Si).

Technol. Definitions Article Figure 2. 3482−3489 . Energy inputs and outputs to a single energy production system.Environmental Science & Technology Table 1.1021/es3038824 | Environ.doi. Figure 3. Sci. net output is shown with the dashed line. and energy production (yellow) is shown above the line. 2013. Energy inputs (blue) are above the horizontal line. Gross output is shown as a bold line. 3484 dx. Energy inputs and outputs for an energy production industry growing asymptotically to some upper limit.

a-Si.doi. which is equivalent to a CEeD of 2. the industry produces enough positive net energy to “pay back” the energy subsidy required for its early growth.11. At the global average growth rate between 2000 and 2010 of 40%. EePBT for a system is proportional to the CEeD of that system assuming a fixed capacity factor. The growth trajectories of the PV technologies are depicted on a biannual basis in Figure 7. The year in which this occurs is termed the payback year. they dx. Average values are highly weighted toward the capacity factor of Germany.23 as illustrated in Figure S2. As shown in Table 1. the definition of EROI is the ratio of electricity production to primary energy inputs.3 we have calculated the distribution in capacity factor for the global PV industry. Learning curves have been used extensively to track financial costs in a variety of energy technologies. mc-Si. In order to make the results commensurate with the metrics developed in our analysis. is shown on a log−log plot in Figure 5 for sc-Si.11. mc-Si.5%. a-Si. This indicates that not all financial cost reduction will result in commensurate energetic cost reductions. after which the industry may then cross the breakeven threshold and thereafter produce a positive net energy output. Financial and Energy Costs of PV Systems. for the PV industry as a whole to be a positive net electricity provider. ribbon. Distribution of estimates of CEeD [kWhe/Wp] of PV systems for a variety of technologies (sc-Si. All of the PV technologies started at or below the breakeven threshold in the year 2000. as shown in Figure 3. CdTe. A number of alternative models were evaluated and are discussed in SI Section Alternative models of decreasing CEeD. In general thin film technologies (a-Si. Rates of decrease are comparable to those for financial costs within the PV industry.5 yrs. 47. we have developed the electricity-return-on-investment (EeROI) as the ratio of electricity outputs divided by the primary energy inputs converted into electrical energy equivalents assuming a grid conversion efficiency of ηgrid = 0. EeROI = 41−82). have shown that the EROI of electricity production from a single PV system is at least as high as oil-fired electricity (EROI = 4−11.1021/es3038824 | Environ.17 While the device level EROI shows that over its lifetime a PV system will produce between 19 and 38 times the electrical energy equivalents needed to produce it. which has the lowest energetic requirements for system manufacture and deployment of all PV technologies. which makes up nearly 40% of global installed capacity. per unit of peak capacity as a function of cumulative installed capacity [MWp]. and installation of PV modules and installed PV systems. The curves demonstrate a clear reduction in the CEeD of PV system manufacture and installation. CdTe has the lowest CEeD of all of the technologies. this metric does not consider the implications of having over 90% of the energetic cost “paid up front”. We combine rates of decrease in CEeD from Figure 5 with EePBT and technology specific growth rates shown in Figure 1 to determine the historical electricity balance of each of the PV technologies between 2000 and 2010. After some time.xlsx.Environmental Science & Technology the industry entails its consuming more energy in system construction and deployment than it produces on an annual basis. 2013. (2012) have compared the energy return on investment (EROI) of PV systems with fossil fuelbased electricity generating technologies.15. we now turn to the specific case for the PV industry.50 We present here an analogous model tracking CEeD of PV system manufacturing and installation. We convert all inputs to PV system manufacture into equivalents of electricity (discussed in SI section Meta-analysis) to determine the cumulative electricity demand (CEeD) [kWhe/Wp] which allows direct comparison of energy inputs into the PV system with electrical output from the system. based on an average capacity factor of 11. 3485 Article Figure 4. EeROI = 12−35).7. ribbon.3. Cumulative Electricity Demand and Growth of the PV Industry.5% for global PV deployments as a whole. and for CdTe. and the EROI of A number of such energy analyses have determined the CED for the materials.5 kWhe/Wp. Technol. Using national data for PV installed capacity and electricity generation between 2005 and 2008. the industry’s net energy output will pass through a minimum value.23−49 These studies were compared in a meta-analysis to allow comparison between studies (see SI Section Meta-analysis for methodological details). 3482−3489 .23−49 In general. As their rates of growth increased over time. CdTe and CIGS) have lower CEeD. including BOS costs. is within the range of coal-fired electricity (EROI = 12−25. thus requiring an energy subsidy. Wafer technologies (sc-Si and mc-Si) have CEeD between 2 and 16 kWhe/Wp. Raugei et al. The CEeD of the production of PV modules (left column) and systems (right column).7. shown in Figure 6. Sci. Underlying data are presented in SI PV CED. and the median for the UN data (light blue) is 12.2. the EePBT must be below 2. Continued rapid growth drives the industry further into deficit. Raugei et al. Sizes of circles represent the installed capacity of each of the PV technologies. ranging between 1 and 7 kWhe/Wp. The median for the EIA data (dark blue) is 11.22 This stage accounts for only 13% of energy inputs. manufacturing. and CIGS) from sources. Having explored the general case. The majority of energetic costs (57%) involve the extraction and purification of poly silicon and the production of PV wafers. A breakdown of cost inputs into different stages in the multicrystalline silicon (mc-Si) PV system production process indicates that one-third of financial costs occur in the last stage involving manufacture of the balance of system (BOS) components and system installation. This crossover happens in the breakeven year. Distribution of estimates for CEeD of the six PV technologies that make up over 99% of the global PV installed capacity are shown in Figure 4. If either the industry growth rate or the CED of systems being deployed decrease sufficiently. The CEeD can be used to determine the electricity payback time (EePBT) for a PV system. thin film technologies have a lower CEeD than wafer technologies.

47. the PV industry should “pay back” electricity consumed in its early growth only by the end of this decade. since they are spread over the full 25 year lifetime of the PV system. despite reduction in their CEeD.11. We carried out a Monte Carlo analysis (full details of which can be found in SI Section: Monte Carlo simulation) to test the sensitivity of the model outputs to varying technology growth rates. maintenance. This perspective does not incorporate the distribution in capacity factor nor any uncertainty in the CEeD.000 TWh/yr (right axis).doi. electrical energy inputs due to O&M activity will be less than 1% of total annual electricity inputs to the PV industry. The installed capacity of PV in the year 2000 was 1. 3482−3489 ■ .1 Using historical installed capacity (Figure 1) and projecting average growth rates from 2005 to 2010 to estimate installed capacity beyond 2010. As of 2010 − the latest year of historical data − the only technology producing a positive net electricity yield was ribbon silicon. operation. Our analysis found that the PV industry was a net electricity consumer as recently as 2010. we modeled the net electricity production of the PV industry to 2020.Environmental Science & Technology Article Figure 5.4 GW. the PV industry could achieve even greater net electricity yields and dx. mc-Si. we undertook an analysis using a system dynamics model outlined in SI Section Model Structure. all moved into the negative net electricity domain. Decommissioning electrical energy inputs lag construction inputs by the lifetime of the system. and in 2008 the PV industry consumed 75% more electricity than it produced. To incorporate these factors. Figure 6. Thus.23−49 plotted as a function of cumulative production [MW] of each technology. This highlights the obvious benefit of technologies with lower CEeD.1021/es3038824 | Environ. Curves are fitted to the data. We also assume that the PV industry received until the year 2000 a total electricity subsidy of 10 TWh from the existing electricity system. after accounting for electricity to construct and install new capacity. This assumption is discussed in more detail in SI Section O&M and disposal costs. The CEeD of PV system production and installation has decreased since the year 2000. 2013. Figure 8 shows that the payback year has a 50% likelihood of occurring between 2012 and 2015 with the probability increasing to almost certainty by 2020. Technol.7 As such. and decommissioning costs account for less than 5% of total lifecycle electrical energy costs for PV systems. CEeD and capacity factor of PV installations. so remain negligible over the modeled time period. a-Si. in the ‘worst case’ scenario. With further reduced CEeD. Estimates of CEeD [kWhe/Wp] of PV modules (left column) and installed PV systems (right column) for a variety of technologies (sc-Si. The fractional reinvestment of CdTe in 2010 was similar to that of sc-Si at around 150%. coupled with the curves tracking reduction in CEeD (Figure 5). and CdTe) from sources7. In reality. shows that deployment of most PV technologies was operating at a net electricity loss for most of the decade 2000−2010. despite CdTe having a much higher growth rate (160% compared with 63%). Distribution in global PV capacity factors using data from refs 2 and 3. There is a 50% probability that the PV industry stopped being a net consumer of electricity by 2011. The net output is plotted in different shades for both 50% and 100% confidence intervals about the median (dark red line). We assume a continuation of growth rates from the period 2005−2010 and historical rates of reduction of CEeD. The gross output (green line) represents the median value from the sensitivity analysis. Sci. with the exception of ribbon silicon. Figure 8 (top) shows that. We assume a system lifetime of 25 years and zero operation and maintenance and decommissioning electricity costs. Figure 7 3486 RESULTS AND DISCUSSION Figure 8 shows the estimated gross and net annual power [TWh/yr] (top) and cumulative electricity [TWh] (bottom) output of the PV industry in both absolute terms (left axis) and annual output as a fraction of the 2010 global annual electricity production of 20.9 GW in 2010. Historical gross power output [TWh/yr] is plotted as blue rising to 22. the PV industry does not become a net electricity producer until 2015.

dx. for example obtaining pure silicon from silica.g. The size of the circles represents the amount of installed capacity [GW] of each technology on a logarithmic basis. hot-carrier cells. As such. the PV industry continues the historical rate of decreasing CEeD. Consequently.51 The other major factor in how well the PV industry performs going forward is the capacity factor that new installations can achieve. and green lines indicate positive net electricity yield. corresponding to an might not be beneficial for the energy sector as a whole. Red lines indicate negative net electricity yield. low-carbon energy sector.1021/es3038824 | Environ. a model has been developed to track the evolution of electrical energy inputs to PV system deployment. then maintaining high PV industry growth rates is beneficial. If.Environmental Science & Technology Article Figure 7. this would entail an electrical energy input of around 2937 TWh/yr. 2013.. the input would be 264 TWh/yr. A number of factors could decrease the fractional reinvestment rate: reducing the growth rate of the PV industry. Doubling the capacity factor. such as the minimum chemical exergy that must be provided in order to purify a material from its usual ore state. the PV industry will benefit from further reductions in the CEeD and deployment in areas with higher capacity factors. without increasing net electricity yield (by either decreasing CEeD or improving PV performance).5% capacity factor (bottom axis) for a number of fractional reinvestment rates [%] (diagonal lines). This is equivalent to the CE e D of some existing polymer PV systems. reducing the CEeD or increasing the capacity factor of PV system deployment. Sci. Deployment of PV in regions with high insolation and capacity factors provides greater overall benefits for the energy system. If no decrease in CEeD of PV occurs. R&D can decrease the amount and purity of materials required in order to maximize the utilization of a given material feedstock (e. There are fundamental thermodynamic limits to our ability to reduce the CEeD. upconversion. policies aimed solely at rapid deployment in areas with low insolation. or thermophotonics. In summary. ceteris paribus. thus have the potential to offset more GHG emissions. however. A number of nascent technologies exist to boost PV efficiency beyond the Shockley-Queisser limit (which defines the maximum theoretical efficiency of a single p-n junction solar cell at 34%) such as multijunction and multiple-exciton cells. by 2020 production could reach 3300 TWh/yr (approximately 17% of current global annual electricity production). a fractional reinvestment of 8%. 47. halves the EePBT. Technology development can reduce the CEeD of PV by decreasing the material requirements per unit of capacity [kg/ Wp] (or alternatively increasing the device efficiency per unit of material used [Wp/kg]) and by decreasing the energy required 3487 to process the materials. If the PV industry continues at its current growth rate.2 kWhe/Wp. It is therefore important to simultaneously increase net energy production by also increasing system efficiency and durability. If the goal is a rapid transition to a sustainable. silicon). 3482−3489 . representative of a fractional reinvestment of 89%. Consequently.doi. mean CEeD of 0. it is important to continue to track and reduce the CEeD of the PV industry. The trajectories of CEeD (or EePBT) and annual growth rates for PV technologies are depicted on a biannual basis between 2000 and 2010. Technol. Growth rate [%/yr] as a function of CEeD) [kWhe/Wp] (top axis) and electricity payback time (EePBT) [yrs] assuming an 11.

Can distributed generation offer substantial benefits in a Northeastern American context? A case study of small-scale renewable technologies using a life cycle methodology. The benefits of this model have been demonstrated by modeling both historic evolution of the PV industry between 2000 and 2010 and also by forecasting PV industry performance out to 2020. (3) UN. (4) Renewables Portfolio Standard Senate Bill X1-02. Wagner. solar thermal energy conversion or energy storage) as well as energy conservation strategies ( Mark The importance of reducing CEeD and increasing net energy should also be recognized as explicit goals in their own right. R. 2011. Lesage. R. A.. The energy return on energy investment (EROI) of photovoltaics: Methodology and comparisons with fossil fuel life cycles. pp 283−289. 281− B. Pineau. Materials availability expands the opportunity for large-scale photovoltaics deployment. (19) Peet. Elsevier: New York. P. Electrical energy inputs to and electricity production by the PV industry should be monitored to judge the success of such a strategy. This analysis is. Adam Brandt. Energy pay-back time: methodological caveats and future scenarios. 2002. Figures S1−S3. 32. (5) California Solar Initiative. 2010. 2010.1249. (11) N. Photovoltaics 2012.Life Cycle Inventories and Life Cycle Assessments of Photovoltaic Systems. Sci. 3488 ■ S Supporting Information * ASSOCIATED CONTENT (1) EPIA. 3482−3489 . N. (10) Fthenakis. Energy Research Centre of the Netherlands ECN. Energy Policy 2004. Judging the impact of a GHG emissions reduction strategy should be based on the actual GHG emissions reduction achieved by PV deployment considering the full life-cycle of emissions. UN Energy Statistics Database. and Mark Schwarz at Stanford University for their invaluable advise.-J. P. 2011. 2013. 2007. 2010. Solar Generation 6. Steve Eglash. 2004.. however. 2012). Fthenakis. insulation. (9) Explorer. Technol. Renewable Sustainable Energy Rev.doi. H. 2012). Dick Swanson.. http://www. (2) EIA. E-mail: mikdale@ stanford. P. 2012.. H. (6) Sijm. M. Fax: 650-723-9190. V. J.. K. M.. Energy Analysis: A Review of Theory and Applications. and text. Kim. The net electricity output is plotted as median (dark red line). or fuel-efficient vehicles) that require large. Fullana-i Palmer. Permanently dispelling a myth of photovoltaics via the adoption of a new net energy indicator. 11. Tables S1−S3. 14.htm (accessed February 5.g. Notes The authors declare no competing financial interest. 2072−2077. Alivisatos. P. M. Mike McGehee. B. M. (20) Herendeen..eia. not limited to the PV industry but applies equally to any rapidly growing alternative energy industry (e. DOI: Photovoltaics Power Systems Programme . V. (7) Amor.. Roy Johnson. http://data. Renewable Sustainable Energy Rev. (12) Wadia. A. 47. Bansal. Confidence intervals of one and two quartiles about the median across all simulations are plotted in different colors. wind. Sci. The authors are also indebted to Richard Sassoon. The performance of feed-in tariffs to promote renewable electricity in European countries. Power Generation from Coal: Measuring and Reporting Efficiency Performance and CO2 Emissions.. J. Watt. Solar Energy 2009. Energy Policy 2012.cpuc. more specifically the change in CEeD as a function of cumulative installed capacity of PV. 162−172. 2012. This material is available free of charge via the Internet at http://pubs. Photovoltaics: Life-cycle analyses. D. Cutler. Filippidou. 1986. This requires supplementing economic and market analyses with energy and material flow analysis. low-energy buildings. New Zealand Energy Research and Development Committee. Annual electricity production (top) and cumulative electricity production (bottom).. The electricity balance of the PV industry is critical to the success of any GHG emissions reduction strategy based on PV deployment. 2009.acs.. 45. Environ. 2010. 2885−2895. J. 576−582.1021/es3038824 | Environ. Thanks are also due to the reviewers for a number of interesting and useful suggestions. Photovoltaics Power Systems Programme . (8) IEA. http://www. 45. REFERENCES Figure 8. (14) DOE..g. 2012). (16) Botsaris. 85. Technol. (13) Richards.. N. cpuc. Charlie Barnhart. 1609−1628.. dx. 2008 Solar Technologies Market Report. Prog. M. http://www. Samson. Net Energy Analysis: Concepts and Methods. (15) IEA. ACKNOWLEDGMENTS This work was made possible with financial assistance from GCEP. John Weyant.1002/pip. C.-O. International Energy energy/solar (accessed February 5. J. Kammen.cfm (accessed February 5. P. 166−175. up-front investment.Environmental Science & Technology ■ ■ ■ Article AUTHOR INFORMATION Corresponding Author *Phone: 650-725-8579. (17) Raugei. Estimation of the energy payback time (EPR) for a PV module installed in North Eastern Greece.Methodology Guidelines on Life Cycle Assessment of Photovoltaic Electricity. countries/data.aspx?d=EDATA (accessed February 5. Solar Energy 2011. 2012). Gross electricity output (green line) represents the median across all simulations. 43. C. (18) IEA/CIAB. Dynamic energy analysis to assess maximum growth rates in developing power generation capacity: case study of India..

A. P. (29) Knapp. Ex-ante environmental and economic evaluation of polymer photovoltaics.. (30) Mathur. 165− 172. (40) Pacca. Younes. 47. J. S. 95− Energy 2010. 33. P. and CIS modules. Ho.Environmental Science & Technology (21) Grimmer. K. 49−53. Inst.. Future concepts for photovoltaic energy conversion. E. Jester.. 517−524. K. 14. Energy 2009. Kichimi. E.-J. Mater. Renewable Sustainable Energy Rev. A. 71. C. Dynamic hybrid life cycle assessment of energy and carbon of multicrystalline silicon photovoltaic systems. 220. Keoleian. (27) Alsema. 387 (accessed February 5. Global Emissions Model for Integrated Systems. Life cycle assessment of photovoltaic electricity generation. greenhouse gas emissions and external costs: 2004 . 952−967. Solar energy breeders. (47) Nishimura. Alsema. (34) Battisti. (31) GEMIS. D. T.. A. M. R. Hayashi. Prog. R. A. M. J. Osman. A.... Energy pay-back time and CO2 emissions of PV systems. Energy Sources 2002. E. (41) Raugei. Future Outlook and Implications for the Calculation of External Costs. 35. 2011. (43) Roes. (39) Mohr.early 2005 status. P. dx. Hu. 3316−3326. M. Life cycle assessment study of solar PV systems: An example of a 2.7 kW(p) distributed solar PV system in Singapore.. Application of life-cycle energy analysis to photovoltaic module design. Schermer. M. Lambert. 44. Mech. Appl. Energy Policy 2001. Corrado. G. 29.. (35) Fthenakis. Bansal. T. D. R. P. K. Frankl. Photovoltaics 2006. Williams. K. A. Prog.. 372−393. Update of environmental indicators and energy payback time of CdTe PV systems in Europe. Solar Energy 2001. (44) Fthenakis. K. 3489 Article (45) Raugei.. 15. Learning rates for energy technologies. M. Eng. Ulgiati. G. 2797−2807. K. Life cycle analysis to estimate the environmental impact of residential photovoltaic systems in regions with a low solar irradiation. a-Si. L. R. Wagner. 255−261. Energy 2007. Photovoltaics energy payback times.. (28) Frankl. Technol... Kim.. Cells 1997. Prog.. 19−26.. A. Part A 2006.. (26) Keoleian. (38) Kannan. Schrattenholzer. An evaluation on the life cycle of photovoltaic energy system considering production energy of offgrade silicon. M. Dewulf.. 17.. J. 87. (51) Guillemoles. 2009. S.. M. 80. Ilg. (37) Mason. Life cycle assessment of a medium-sized photovoltaic facility at a high latitude location. Prog. Sol. Energy analysis of solar photovoltaic module production in India. M. N. 29. K. K. (33) Krauter. 17... 17−25. Sol. 2010. Photovoltaics 1997. Lewis. (22) Swanson.. Energy 2008. Energy Environ. CdTe. 7950−7955. J. Sivaraman. A comparative study on cost and life-cycle analysis for 100aA ̂ ̆ L’MW very large-scale PV (VLS-PV) systems in deserts using m-Si. S. 163−179.. Kato. Hirota. 5. Leong. L. Kato. Held. 47. P. 15. 24. K. (50) McDonald. 14. Technol... A... 2012). 238. C. Kim. V. Energy 2005. Blok. Krones. H. 34. Environ. Considerations for the calculation of greenhouse gas reduction by photovoltaic solar energy. Sci. Araki.. Patel. Sci. Energy Sources 1995. Ito. V.. 2002.stanford. M. Bansal. S.. Energy and environmental correlation for renewable energy systems in India.. K. Energy Policy 2007. Life cycle assessment and evaluation of energy payback time on high-concentration photovoltaic power generation system. C.. http://energyseminar. Life cycle assessment of thin-film GaAs and GaInP/GaAs solar modules. (46) Zhai.. (23) Stoppato. 32. Komoto. Update of PV energy payback times and life-cycle greenhouse gas emissions. Murata. J.doi.. Photovoltaics 2011.. Photovoltaics 2009. Proc. S.. Raugei. Prog. C. Energy 2004. Meijer... (36) Muneer. 345−355. Y. M. J. K.. Energy payback and life-cycle CO2 emissions of the BOS in an optimized 3Â u°5aA ̂ ̆ L’MW PV installation. Life Cycle Assessment of Renewables: Present Issues. Fundam. 392−399.. A. L. Reijnders. 614−626. H. Energy Mater. M. Bargigli. K. 287−300. Solar Energy 1981. M. Tanaka. 179− 190. Prog. Empirical investigation of the energy payback time for photovoltaic modules. N. (25) Kato. D. (49) Held.. Fthenakis. R. Solar Energy 2006.. Sustainability 2011. A. Prog. 2011. R. H. 3482−3489 . Albrecht. V. 1310−1318.. Parameters affecting the life cycle performance of PV technologies and systems.-J.. Photovoltaics 2000. 267−281.. 30. Wagner. G. H. 275−280. 19. Photovoltaics 2006. K. J. 26.1021/es3038824 | Environ. 2013. 2001. N. (48) Laleman. 8. 555−563.. M. Prog. J.. Tso. T. Cumulative energy demand and cumulative emissions of photovoltaics production in Europe. Life cycle impacts and costs of photovoltaic systems: Current state of the art and future outlooks. Huijbregts. A. E.. N. E. Kurokawa. T. A. Alsema. J. Photovoltaics 2007. (32) Gürzenich. Hansen. 29. D. K. Evaluation of technical improvements of photovoltaic systems through life cycle assessment methodology.. J. Life cycle assessment and energy pay-back time of advanced photovoltaic modules: CdTe and CIS compared to poly-Si. 605−613. 17−30. R. 224−232. Solar's Learning Curve Paves Way to Competitive Costs. 2297−2303. Ruther. J.. E. R. (24) Prakash. S.. Kubie. (42) Ito. H. Renewable Energy 2004. Sakuta.. Photovoltaics 2008. J.