In silico design of solvents for carbon capture with simultaneous optimisation of operating conditions

This paper presents an in silico optimisation methodology that simultaneously targets an optimal solvent as well as optimal process conditions for a pre-combustion carbon capture process. A mixed integer non-linear programming (MINLP) method is formulated to minimise the energy requirement for capturing and compressing carbon dioxide. The optimisation is founded on a Computer Aided Molecular Design (CAMD) framework and is firstly carried out to obtain an optimal pure solvent, while an optimal binary solvent blend is targeted through a second separate optimisation. When targeting a solvent blend, a 9% decrease in capture energy requirement is achieved with simultaneous optimisation of solvent and operating conditions over the case with operating conditions optimisation only. Moreover, a 2% decrease in energy requirement is achieved when a binary solvent mixture is optimised as opposed to a pure solvent.