High pressure equilibrium data of CO2/cyclohexene oxide and CO2/limonene oxide systems in the context of polycarbonate synthesis using CO2 as a co-monomer

Abstract

Polycarbonates are a class of high-performance polymers with a wide range of industrial applications. Traditionally, polycarbonate production involves the use of bisphenol A (BPA) and phosgene (COCl₂), which have raised concerns due to their high toxicity and the overall environmental impact of the process. To address these issues, more sustainable methods such as ring-opening copolymerization (ROCOP) of epoxides and carbon dioxide (CO₂) under supercritical CO₂ (sCO₂) are being developed. Recent research has focused on the synthesis of poly(cyclohexene carbonate) (PCHC) and poly(limonene carbonate) (PLC) from cyclohexene oxide (CHO) and limonene oxide (LO), respectively. Limonene oxide has attracted particular interest due to its non-toxic properties and its derivation from limonene, an available bio-based terpene. Reaction performance is strongly influenced by the initial physical state of the mixture, in particular the composition of the liquid and vapor phases. Therefore, access to this thermodynamic information, represented by phase diagrams, is essential for understanding and predicting reaction behavior. However, there is a lack of equilibrium data for these two systems. The aim of this study is to investigate the phase equilibria of the CO₂/CHO and CO₂/LO binary mixtures, thereby contributing to the advancement of greener polycarbonate production. The experiments were performed in a variable volume view-cell, covering a temperature range from 338.15 K to 363.15 K for the CO₂/CHO mixture and from 303.15 K to 343.15 K for the CO₂/LO system. The molar fraction of CO₂ was varied between 0.189 and 0.967 for the CO₂/CHO case, and between 0.227 and 0.997 for the CO₂/LO mixture. Vapor-liquid bubble point (VLE-BP) and dew point (VLE-DP) were determined for both mixtures, along with mixture critical points for the temperatures studied. The phase behavior was modeled using several equations of state (EoS). In particular, the Peng-Robinson (PR) equation of state with the volume translated Peng-Robinson (VTPR) complex mixing rule and the Wilson model for the activity coefficient provided a highly accurate description of the experimental results.