Development of a methodology for the sustainability evaluation of industrial pharmaceutical processes

This work extends a general framework for the prediction of industrial sustainability, i.e. Process Sustainability Prediction (PSP) Framework proposed by Fermeglia et al. [1], by implementing features for treating the sustainability estimation of batch pharmaceutical processes. The resulting procedure allows performing evaluation of different industrial batch plants alternatives using representative indicators and metrics that cover the three aspects of sustainability: economy, environment and society. Two kinds of sustainability indicators are adopted: three dimensional indicators (3D-indicators) and one dimensional indicators (1D-indicators). The methodology proposed by Martins et al. [2], containing 3D-indicators, has been updated, modified and extended including the concept of time, related to discontinuous processes as pharmaceutical ones. The methodology takes into account E-factor (proposed by Sheldon [3]), energy intensity, potential chemical risk and potential environmental impact modified by considering the cycle time of batch reactors. The 3D-indicators are calculated using the results obtained from process simulation coupled with the toxicological database, containing data from EPA, other sources, and an in-house procedure for the estimation of toxicological properties based on molecular modelling. Whenever a more sophisticated evaluation of the impact is needed, one may resort to the evaluation of 1D-indicators. One dimensional indicators are focused on single contribution to sustainability, e.g. Waste Reduction Algorithm (WAR) [4] focused on environmental issues, Profit Intensity, an index based on Materials Intensity [2] and modified to inspect economic aspects, and HTP (Hazard Toxicity Potential) developed to define the hazard risk for employers inside the chemical plant. In this paper, we have modified the original framework of WAR by including the time of charge, reaction and discharge from reactor and the introduction of new chemicals for rinsing for taking into account more closely the industrial operation in this sector. Some other changes have been done on environmental impact categories to obtain more reliable data on environmental issues only. Process data are extracted from process simulation software, coupled with toxicological data and with a molecular modelling procedure based on quantum chemistry (COSMO-RS) for the prediction of toxicological properties.