Chemical potentials of water, methanol, carbon dioxide and hydrogen sulphide at low temperatures using continuous fractional component Gibbs ensemble Monte Carlo. Issue 5 (24th March 2018)

• Record Type: Journal Article
• Authors: Rahbari, Ahmadreza; Poursaeidesfahani, Ali; Torres-Knoop, Ariana; Dubbeldam, David; Vlugt, Thijs J. H.
• Is Part Of: Molecular simulation. Volume 44:Issue 5(2018)
• Page Start: 405
• Page End: 414
• Publication Date: 2018-03-24
• Subjects: CFCMC -- Gibbs ensemble -- chemical potential -- phase equilibrium -- water -- hydrogen bonding; Molecular dynamics -- Computer simulation -- Periodicals; Statistical mechanics -- Computer simulation -- Periodicals; 539.6
• DOI: 10.1080/08927022.2017.1391385
• Languages: English
• ISSNs: 0892-7022

Abstract

Abstract : Chemical potentials of coexisting gas and liquid phases for water, methanol, hydrogen sulphide and carbon dioxide for the temperature range K to K are computed using two different methodologies: (1) Widom's test particle insertion (WTPI) method in the conventional Gibbs Ensemble (GE), and (2) the Continuous Fractional Component Gibbs Ensemble Monte Carlo (CFCGE MC) method. It is shown that the WTPI method fails to accurately compute the chemical potentials of water and methanol in the liquid phase at low temperatures, while accurate chemical potentials in the liquid phase are computed using CFCGE MC method. For the CFCGE MC method, the statistical uncertainty for computed chemical potentials of water and methanol in the liquid phase are considerably smaller compared to the WTPI method. For the water models considered in this study (SPC, TIP3P-EW, TIP4P-EW, TIP5P-EW), computed excess chemical potentials based on three-site models are in better agreement with the chemical potentials computed from an empirical equation of state from the NIST database. For water, orientational biasing is applied during test particle insertion to check whether certain orientations of test particle are energetically unfavourable. A two-dimensional Overlapping Distribution Method (ODM) in the NVT ensemble is derived for this purpose. It is shown that failure of the WTPI method for systems with a strong hydrogen bonding network does not depend on orientation of the test molecule in that system. For all systems in this study, the WTPI method breaks down when the void fraction of the system drops below approximately 0.50.