Life cycle assessments and external cost estimates of photovoltaics have been often based on old data that do not reflect the extensive technological progress made over the past decade. Our assessment uses current (2004-early 2005) manufacturing data, from twelve European and US photovoltaic companies, and establishes the Energy Payback Times (EPBT), Greenhouse Gas (GHG) emissions and external environmental costs of current commercial PV technologies. Estimates of external costs are about 70% lower than those in recent high-impact publications which were derived from the old data. Copyright # 2006 John Wiley & Sons, Ltd. INTRODUCTIONI t is well understood that production of energy by burning of fossil fuels generates a number of pollutants and carbon dioxide. What is less known is that any anthropogenic means of energy production, including solar, generate pollutants when their entire life cycle is accounted for. A life cycle starts from the mining and processing of materials that comprise solar cells, modules and balance of system, and ends to their final decommissioning, disposal and/or recycling. Costs associated with the environmental, health and societal impacts that are not included in the direct cost of electricity, are called external costs of electricity production. While societal external costs are difficult to quantify, external costs associated with environmental and health protection or damage have been quantified in monetary terms. Perhaps the most well-known effort to quantify environmental and health damages due to electricity production, is the European Union's series of ExternE (External Costs of Energy) projects. The ExternE methodology starts from emissions generated at specific sources and follows their impact to receptors through atmospheric dispersion and dose-response functions. In general, this type of environmental impact assessment is well accepted, although assumptions related to Broader Perspectives the monetary valuation of estimated impacts, especially green-house related impacts, are debateable. 'The ExternE methodology has been applied in a large number of European and national studies to give advice for environmental, energy and transport policies.' 1 Photovoltaic installations in Germany were presented in the latest ExternE report to the European Commission 1 as having 30% higher health impacts than natural gas and GHG emissions of 180 g CO 2 -eq./kWh which would be 10 times higher than those for the nuclear fuel cycle (Figure 1). These results were based on 15-years old PV systems and even older data on module production technology. z Also based on outdated PV technology data a life cycle-based comparison of energy technologies in Australia 2 showed that PV emits about 100 g CO 2 /kWh during its life cycle (Figure 1). The results from these two studies were widely circulated and especially the ExternE publication with its official status is likely to have influenced policy decisions with regard to energy technology. More recent (i.e., 2000) data are included in the Ecoinvent...

Officer and his technical guidance is gratefully acknowledged. John Viren is a private consultant working under contract to Brookhaven National Laboratory. We thank several outside reviewers for their comments. This reportdoes not necessarilyreflect the views and policies of the U.S. Department of Energy nordoes mention of trade names or commercial productsconstituteendorsementor recommendationfor use.

No abstract

Title III of the 1990 Clean Air Act Amendments (CAAA) mandated that the U.S. Environmental Protection Agency (EPA) evaluate the need to regulate mercury emissions from electric utilities. In support of this forthcoming regulatory analysis the U.S. DOE, sponsored a risk assessment project at Brookhaven (BNL) to evaluate methylmercury (MeHg) hazards independently. In the U.S. MeHg is the predominant way of exposure to mercury originated in the atmosphere. In the BNL study, health risks to adults resulting from Hg emissions from a hypothetical lo00 MW coal-fEed power plant were estimated using probabilistic risk assessment techniques. This study showed that the effects of emissions of a single power plant may double the background exposures to MeHg resulting from consuming fish obtained from a localized area near the power plant. Even at these more elevated exposure levels, the attributable incidence in mild neurological symptoms @aesthesia) was estimated to be quite small, especially when compared with the estimated background incidence in the population. This research was performed under the auspices of the United States Department of Energy under Contract 7) NO. DE-ACE-76CHWI16

No abstract

One of the most promising materials for low-cost thin film photovoltaic cells is copper indium selenide (CulnSg or CIS). As with any new material, successful commercialization of CIS photovoltaic (PV) technology will require attention to environmental issues related to the sources of raw materials, their usage, ana the disposal andlor recycling of products at the end of their useful life. This paper focuses on three specific environmental issues related to CIS technology: (i) Economics of the use and re-use of materials: (ii) regulations on environmental disposal and waste hapdling, and (iii) logistics and economics of recycling and disposing of products by industries faced with comparable environmental issues. COST AND MATERIAL AVAILABILITY OF CIS PVEstimates of the production costs for large scale 10-20 MWlyr) CIS PV modules range from $0.5-13a. The materials in CIS modules cost 3545% of the total projected manufacturing costs. Glass for the substrates and cover sheets accounts for about one third of this amount, and the active thin-film electrodes and semiconductor junction materials account for an additional one quarter to one third. Indium is the most costly of the thin-film constituents, accounting for 2.5-5% of the total projected cost of a module. The absence of any key recoverable material makes it unlikely that the availability of materials will economically warrant recovering any materials from worn-out mcdules.Overall material supply is an important issue for the long-term contribution of CIS PV technology. In the long-term, large-scale production of CIS PV modules may be limited by the availability of indium and selenium io levels of ca. 200 GWp. Furthermore, other new and growing industries may compete with the PV industry for these materials; for example, the flat-panel display industry may compete for In. These reasons suggest that we should consider options for minimizing the consumption of these materials, and for recovering these materials from end-of-life modules.

Title III of the 1990 Clean Air Act Amendments (CAAA) mandated that the U.S. Environmental Protection Agency (EPA) evaluate the need to regulate mercury emissions from electric utilities. In support of this forthcoming regulatory analysis the U.S. DOE, sponsored a risk assessment project at Brookhaven (BNL) to evaluate methylmercury (MeHg) hazards independently. In the U.S. MeHg is the predominant way of exposure to mercury originated in the atmosphere. In the BNL study, health risks to adults resulting from Hg emissions from a hypothetical lo00 MW coal-fEed power plant were estimated using probabilistic risk assessment techniques. This study showed that the effects of emissions of a single power plant may double the background exposures to MeHg resulting from consuming fish obtained from a localized area near the power plant. Even at these more elevated exposure levels, the attributable incidence in mild neurological symptoms @aesthesia) was estimated to be quite small, especially when compared with the estimated background incidence in the population. This research was performed under the auspices of the United States Department of Energy under Contract 7) NO. DE-ACE-76CHWI16

Plasma etching in the photovoltaic industries involves several inorganic and organic fluorides, chlorides and bromides, some of which are toxic, corrosive, or flammable. Nitrogen fluoride, a relatively new etching material, presents special concerns because it may form self‐ignitable mixtures with some flammable gases. In addition, non‐toxic compounds (e.g. Freon), can form toxic by‐products in a plasma environment. Another safety concern is failure of vacuum pumps caused by corrosive effluents, and occupational safety issues associated with cleaning traps and reactors. These issues are discussed in this paper and the applicable safety and environmental control options are outlined.