Electricity generation is a key contributor to global emissions of greenhouse gases (GHG), NO x and SO 2 and their related environmental impact. A critical review of 167 case studies involving the life cycle assessment (LCA) of electricity generation based on hard coal, lignite, natural gas, oil, nuclear, biomass, hydroelectric, solar photovoltaic (PV) and wind was carried out to identify ranges of emission data for GHG, NO x and SO 2 related to individual technologies. It was shown that GHG emissions could not be used as a single indicator to represent the environmental performance of a system or technology. Emission data were evaluated with respect to three life cycle phases (fuel provision, plant operation, and infrastructure). Direct emissions from plant operation represented the majority of the life cycle emissions for fossil fuel technologies, whereas fuel provision represented the largest contribution for biomass technologies (71% for GHG, 54% for NO x and 61% for SO 2 ) and nuclear power (60% for GHG, 82% for NO x and 92% for SO 2 ); infrastructures provided the highest impact for renewables. These data indicated that all three phases should be included for completeness and to avoid problem shifting. The most critical methodological aspects in relation to LCA studies were identified as follows: definition of the functional unit, the LCA method employed (e.g., IOA, PCA and hybrid), the emission allocation principle and/or system boundary expansion. The most important technological aspects were identified as follows: the energy recovery efficiency and the flue gas cleaning system for fossil fuel technologies; the electricity mix used during both the manufacturing and the construction phases for nuclear and renewable technologies; and the type, quality and origin of feedstock, as well as the amount and type of co-products, for biomass-based systems. This review demonstrates that the variability of existing LCA results for electricity generation can give rise to conflicting decisions regarding the environmental consequences of implementing new technologies.

23Life cycle assessment (LCA) has been used extensively within the recent decade to 24 evaluate the environmental performance of thermal Waste-to-Energy (WtE) 25 technologies: incineration, co-combustion, pyrolysis and gasification. A critical review 26 was carried out involving 250 individual case-studies published in 136 peer-reviewed 27 journal articles within 1995 and 2013. The studies were evaluated with respect to 28 critical aspects such as: i) goal and scope definitions (e.g. functional units, system 29 boundaries, temporal and geographic scopes), ii) detailed technology parameters (e.g. 30 related to waste composition, technology, gas cleaning, energy recovery, residue 31 management, and inventory data), and iii) modeling principles (e.g. energy/mass 32 calculation principles, energy substitution, inclusion of capital goods and uncertainty 33 evaluation). Very few of the published studies provided full and transparent descriptions 34 of all these aspects, in many cases preventing an evaluation of the validity of results, 35 and limiting applicability of data and results in other contexts. The review clearly 36suggests that the quality of LCA studies of WtE technologies and systems including 37 energy recovery can be significantly improved. Based on the review, a detailed 38 overview of assumptions and modeling choices in existing literature is provided in 39 conjunction with practical recommendations for state-of-the-art LCA of waste-to-40 energy. 41 42

In Europe, about 20% of municipal solid waste is incinerated. Large differences can be found between northern and southern Europe regarding energy recovery efficiencies, flue gas cleaning technologies and residue management. Life-cycle assessment (LCA) of waste incineration often provides contradictory results if these local conditions are not properly accounted for. The importance of regional differences and site-specific data, and choice of LCA model itself, was evaluated by assessment of two waste incinerators representing northern and southern Europe (Denmark and Italy) based on two different LCA models (SimaPro and EASEWASTE). The results showed that assumptions and modelling approaches regarding energy recovery/ substitution and direct air emissions were most critical. Differences in model design and model databases mainly had consequences for the toxicity-related impact categories. The overall environmental performance of the Danish system was better than the Italian, mainly because of higher heat recovery at the Danish plant. Flue gas cleaning at the Italian plant was, however, preferable to the Danish, indicating that efficient flue gas cleaning may provide significant benefits. Differences in waste composition between the two countries mainly affected global warming and human toxicity via water. Overall, SimaPro and EASEWASTE provided consistent ranking of the individual scenarios. However, important differences in results from the two models were related to differences in the databases and modelling approaches, in particular the possibility for modelling of waste-specific emissions affected the toxicity-related impact categories. The results clearly showed that the use of site-specific data was essential for the results.

No abstract