(455d) Data Envelopment Analysis Coupled with Thermodynamic and Life Cycle Assessment Metrics for Solvent Screening: Application to CO2 Capture

The growing trend towards more sustainable products and processes calls for advanced decision-making tools for the assessment and optimisation of the economic, environmental and social impact of chemicals and chemical processes. The pressure for more sustainable processes is particularly strong in the chemical industry, in which thousands of products are developed and launched at a very fast pace, raising the question (before scaling them up) of whether they are more sustainable than the existing ones or not.

Particularly, in green chemistry applications, engineers and scientists are confronted with the task of screening chemicals (e.g. mainly solvents, additives, etc.) according to sustainability principles. In practice, it is quite hard to find a single chemical/process performing best in the whole range of sustainability metrics, as inherent trade-offs typically arise between conflictive thermodynamic, environmental and safety metrics. There has been a substantial amount of research in the field of multi-criteria assessment of chemical processes.^1-3 In contrast, the problem of screening chemicals according to several criteria has received much less attention to date. In the absence of any rigorous and systematic screening method, practitioners assess them by defining, in either an implicit or explicit manner, subjective weights that are attached to every category of interest (e.g. viscosity, density, solubility, etc.). These weights enable the calculation of aggregated indicators on the basis of which the chemicals can be ranked.^3-5While it is straightforward to implement, this approach is highly sensitive to the weights defined and provides in turn no insight into the underlying multi-criteria problem.

In this work we propose a new methodology to screen and select chemicals according to the extent to which they adhere to sustainability principles that is based on Data Envelopment Analysis (DEA), a technique developed in economics for efficiency assessment. Essentially, DEA is a non-parametric analytical tool for quantifying the relative efficiency of a set of decision making units (DMUs) that makes use of linear programming techniques.^6,7DEA serves two purposes: i) it identifies, among a set of DMUs - each producing multiple outputs from multiple inputs - the efficient ones (i.e. those with the best ratio desired outputs to required inputs); and ii) it establishes efficient targets for the inefficient units that if achieved would make them efficient. DEA has recently found many applications in environmental studies, but to the best of our knowledge has never been applied to assess chemicals in green chemistry applications. In essence, we combine here DEA with a set of thermodynamic and life cycle assessment indicators so as to identify the chemicals that perform best according to technical (economic), environmental and social aspects simultaneously.

The capabilities of the DEA-based approach in the context of green chemistry are demonstrated through the screening of 125 conventional amine-based solvents for CO2 capture according to 10 performance indicators. Our approach eliminates 36% of the solvents for being inefficient, establishing quantitative targets and clear guidelines on how to improve them. The DEA models are capable of identifying the practical downsides of amines as main sources of inefficiency through the calculation of clear improvement targets for properties such as vapour pressure, acute toxicity and several life cycle impacts. The final aim of the DEA-based analysis is to facilitate the selection of more sustainable chemicals in the transition towards a more sustainable chemical industry.

References

1. C. Capello, U. Fischer and K. Hungerbühler, Green Chem., 2007, 9, 927.
2. G. Wernet, S. Papadokonstantakis, S. Hellweg and K. Hungerbühler, Green Chem., 2009, 11, 1826.
3. S. Hellweg, U. Fischer, M. Scheringer and K. Hungerbühler, Green Chem., 2004, 6, 418.
4. M. Tobiszewski, S. Tsakovski, V. Simeonov, J. NamieÅ?nik and F. Pena-Pereira, Green Chem., 17(10), 4773-4785.
5. A. I. Papadopoulos, S. Badr, A. Chremos, E. Forte, T. Zarogiannis, P. Seferlis, S. Papadokonstantakis, C. S. Adjiman, A. Galindo and G. Jackson, Chem. Eng. Trans., 2014, 39, 211â??216.
6. W. D. Cook and L. M. Seiford, Eur. J. Oper. Res., 2009, 192, 1â??17.
7. P. Zhou, B. W. Ang and K. L. Poh, Eur. J. Oper. Res., 2008, 189, 1â??18.