1 Department of Management Engineering, Technical University of Denmark
The present report results as part of a ph.d. project aiming at developing general rules for LCA of basic chemicals. Part 1 reviews the state of the art framework for life-cycle assessment and part 2 relates that framework to the chemical industry and gives some proposals for LCAs of basic chemicals. A survey of the use of LCA in the European chemical industry has been conducted and is reported in appendix 1. Life cycle assessment (LCA) is a tool/process to evaluate the potential environmental impacts of a system (most often a product) through its whole life cycle by collecting input and output data (together interventions) from all unit processes in the system and assessing the potential environmental impact of these interventions. The LCA framework comprise four phases which are iteratively interlinked: · Goal and scope definition · Inventory · Impact assessment · Interpretation In the goal definition the purpose of the study is defined. The intended application of the results is very important to determine the extent of the study. The scope definition concerns defining how the study is to be performed, e.g. how is the function of the system defined, what part of the system should be included or excluded, which parameter should be evaluated, how should the study be validated etc. The inventory analysis is the phase where all the needed data are collected and it is the most laborious phase. Preferably data from the actual process should be used, but the data are most often collected from a variety of sources and will constitute a mix of actual measurements, literature data and estimates. Actually, during the case studies and from other studies, data availability was identified as a major problem. Therefore, data sources for process data from the chemical industry were investigated. The primary data sources are technical chemical encyclopaedias. However, as these works rarely include emissions, a review of several methodologies to estimate emission were made and some of the approaches tested in the case studies. It was found that in general the emission factors provided by these sources provided a good, although at times somewhat conservative estimate of the emissions. A special problem occurs when there are two or more product produced in the same process, a situation frequently encountered in chemical manufacturing. In such cases the environmental interventions of the process have to be allocated between the products from that process. In the chemical industry this is most often done in relation to the mass of the products. To provide a reliable result of the LCA it is further important to analyse the uncertainties of the study. However, most often the data does not support a quantitative uncertainty analysis and a sensitivity analysis must be performed as a check for validity. The impact assessment aims to evaluate the contribution of the interventions to different environmental impact categories, e.g. global warming, toxicity to humans, acidification etc. The assessment of toxicological impacts have been given special attention because the chemical industry typically will be significant contributor to this impact category. Several aspect of the LCA framework inflict the possibilities the assess actual impacts and especially for local impacts like toxicity this is a cause of limited accordance between the potential impact predicted in the LCA and the occurrence of actual impacts. The interpretation phase of the LCA checks the completeness and consistency of the results in relation to the goal and scope. LCA is mainly applied for internal purposes in the chemical industry. Typically, the most important application are in research and development to improve the environmental performance of processes and products, and as support for strategic decision making. However, the coherence between LCA and decision making needs to be addressed more specifically because most companies in the survey did not think LCA is sufficiently fit to support strategic decision making. Because, the chemical industry is a major supplier to other product systems the major incentive to perform LCAs has been to comply with customer requirements. An LCA may not always need to be very detailed to fulfil the goal, a simplified approach may in some cases serve as well. However, because of the diverse nature of chemical manufacturing it was not possible to establish general rules for simplification. Therefore a simplification procedure is suggested, where a semiquantitative principle is used to identify focal point for simplified LCAs.