Chemical Networks: A Methodology to Rapidly Assess the Environmental Impact of Chemical Processes : Applying graph theory principles to chemical industry data enables early-stage decision making for optimum decarbonisation solutions

Abstract

As the chemicals industry transitions towards a net zero future, rapid assessment of the sustainability metrics of different process results will be essential to support investment decisions in innovation and deployment. Life cycle analysis (LCA) offers the gold standard for process assessment, but LCA can take weeks or months to complete, with incomplete databases and inflexibility in comparing different chemical pathways. In this study, we demonstrate an alternative and complementary methodology. By simplifying the metrics used to describe chemical processes, each process may be linked to another by its feedstocks and products. This generates a network of the chemical industry, which may be investigated using graph theory principles. A case study of the plastics industry is provided, using publicly available information to quantitatively compare with a more formalised and detailed LCA approach. This methodology proves useful for quickly estimating the carbon intensity and water footprint of thousands of routes. Further development, such as including Scope 3 emissions and additional industrial data, may further improve the methodology.