Relative contribution and sensitivity analyses for decision-making in the environmental assessment of bio-based phase change materials

This paper aims to investigate the environmental impact of the production of bio-based phase change materials to contribute to the advancement and development of energy systems with a low environmental impact. It uses relative contribution, local, and global sensitivity analysis to target critical inputs for decision-making. The materials selected are jojoba oil, coconut oil, lauric, myristic, and stearic acids, xylitol, and adipic acid, with melting temperatures varying from 14 °C to 152 °C. The methodology follows the ISO guidelines for the life cycle assessment of all materials, for which the boundary system has considered three unit processes: farming, transportation, and manufacturing. It was found that to produce 1 kg of adipic acid, the global warming potential is 13 kg CO_{2 eq}, while for jojoba oil, it is only 0.96 kg CO_{2 eq}. For phase change materials like jojoba oil, coconut oil, myristic acid, and stearic acid, farming dominates the global warming potential, while manufacturing takes over the major share of this environmental indicator for the remaining materials. The local sensitivity analysis results show that jojoba oil's global warming potential can increase by nearly 38 % if irrigation water reaches its upper bound. On the other hand, global sensitivity analysis shows the Sobol index by which inputs contribute to output uncertainty. For instance, the jojoba oil global warming uncertainty could be affected by 60 % by irrigation water. Results show that the higher melting temperature phase change materials have a higher footprint than the lower melting temperature ones, except for coconut oil, as a consequence of more process inputs and production process complexity. Overall, relative contribution analysis helps locate critical unit processes and inputs, while local sensitivity analysis allows for identifying how critical input variation affects the targeted indicator, and the global sensitivity analysis is useful to understand the critical inputs in terms of uncertainty.