Design in the chemical industry increasingly aims not only at economic but also at environmental targets. Environmental targets are usually best quantified using the standardized, holistic method of life cycle assessment (LCA). The resulting life cycle perspective poses a major challenge to chemical engineering design because the design scope is expanded to include process, product, and supply chain. Here, we first provide a brief tutorial highlighting key elements of LCA. Methods to fill data gaps in LCA are discussed, as capturing the full life cycle is data intensive. On this basis, we review recent methods for integrating LCA into the design of chemical processes, products, and supply chains. Whereas adding LCA as a posteriori tool for decision support can be regarded as established, the integration of LCA into the design process is an active field of research. We present recent advances and derive future challenges for LCA-based design. 203 Annu. Rev. Chem. Biomol. Eng. 2020.11:203-233. Downloaded from www.annualreviews.org Access provided by 44.224.250.200 on 07/08/20. For personal use only.

We present COSMO-CAMPD, a framework for integrated design of molecules and processes based on COSMO-RS.

No abstract

The kinetics of chemical reactions in the liquid phase are often strongly determined by the reaction solvent. Consequently, the choice of the optimal solvent is an important task in chemical process design. Because of the vast number of potential solvents, experimental testing of all candidates is infeasible. To explore the design space of possible reaction solvents, computer-aided molecular design (CAMD) methods have been developed. However, state-of-the-art CAMD methods for reaction solvent design consider usually only a limited molecular design space and rely on simplified models fitted to experimental data to predict solvent performance. To overcome these limitations, we here propose Rx-COSMO-CAMD as the method for the design of reaction solvents. Rx-COSMO-CAMD combines CAMD using the genetic optimization algorithm LEA3D with sound prediction of reaction kinetics based on transition-state theory and advanced quantum chemical methods. Thereby, no experimental data are required. The predictions are shown to be computationally efficient and not limited to certain structural groups. Thus, large and diverse molecular design spaces can be explored. To demonstrate the proposed Rx-COSMO-CAMD method, we successfully design solvents, enhancing the reaction kinetics of a Menschutkin reaction and a chain propagation reaction for the production of polymers and microgels. The method is shown to identify promising solvents for significant enhancement of reaction rates. Rx-COSMO-CAMD is therefore a powerful, fully predictive tool for the identification of optimal reaction solvents.