Solubilities of CO2, CH4, C2H6, CO, H2, N2, N2O, and H2S in commercial physical solvents from Monte Carlo simulations

Q. Chen, Mahinder Ramdin, Thijs J.H. Vlugt*

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

91 Downloads (Pure)


The removal of acid gas impurities from synthesis gas or natural gas can be achieved using several physical solvents. Examples of solvents applied on a commercial scale include methanol (Rectisol), poly(ethylene glycol) dimethyl ethers (Selexol), n-methyl-2-pyrrolidone (Purisol), and propylene carbonate (Fluor solvent). Continuous Fractional Component Monte Carlo (CFCMC) simulations in the osmotic ensemble were used to compute the Henry coefficients of the pure gases CO (Formula presented.), CH (Formula presented.), C (Formula presented.) H (Formula presented.), CO, H (Formula presented.), N (Formula presented.), N (Formula presented.) O, and H (Formula presented.) S in the aforementioned solvents. The predicted Henry coefficients are in good agreement with the experimental results. The Monte Carlo method correctly predicts the gas solubility trend in these physical solvents, which obeys the following order: H (Formula presented.) S > CO (Formula presented.) > C (Formula presented.) H (Formula presented.) > CH (Formula presented.) > CO > N (Formula presented.) > H (Formula presented.). The gas separation selectivities for the precombustion process and the natural gas sweetening process are calculated from the pure gas Henry coefficients. The CO (Formula presented.) /N (Formula presented.) O analogy is verified for the solubility in these solvents.

Original languageEnglish
Pages (from-to)1341-1349
JournalMolecular Simulation
Issue number13-14
Publication statusPublished - 2023


  • CO /N O analogy
  • commercial physical solvents
  • gas solubilities
  • osmotic ensemble
  • separation selectivities


Dive into the research topics of 'Solubilities of CO2, CH4, C2H6, CO, H2, N2, N2O, and H2S in commercial physical solvents from Monte Carlo simulations'. Together they form a unique fingerprint.

Cite this