UW-Madison

  UW-Madison · Engineering  
  Chem & Biol Engr · MTSM  
   

UW-Madison

 

Alumnus
Amadeu K. Sum

Education

Research

Application of Quantum Chemical Ab Initio Methods for Phase Equilibria Predictions

Dr. Sum's research used ab initio quantum mechanics methods to study interactions (hydrogen-bonding and van der Waals) of molecules in a small molecular cluster. The computations performed for this research required high-performance computers in order to study realistic systems. The calculated interaction energies between molecular pairs were then utilized in activity coefficient models (e.g. UNIQUAC and Wilson) as energy parameters in order to predict the phase behavior of systems. Activity coefficients play a central role in phase equilibria calculations. Currently, UNIFAC is the best predictive model for activity coefficients; however, it is limited by availability of parameters that have been regressed from experimental data, and by the uncertain accuracy of the method. In addition, as a result of technological advances and environmental concerns, thermodynamicists are faced with the need to predict the phase behavior of more complex and/or hazardous systems for which experimental data may not exist. This work presented a new and unique approach in which phase equilibrium predictions were based solely on the use of ab initio quantum mechanics methods to determine the parameters in activity coefficient models without the use of experimental data. In particular, the intermolecular energies that appear in activity coefficient models are the result of electronic and nuclear interactions, and these were computed by solving the Schrödinger equation. This was done by using ab initio methods to determine a minimum energy geometry of a cluster of molecules. Then, from this cluster (typically of eight molecules), the interaction energies of like and unlike pairs of molecules were calculated and used as input parameters for activity coefficient models such as UNIQUAC. Vapor-liquid equilibria predictions using this method has been successfully obtained for several binary aqueous systems at both low and high pressures.

Dr. Sum also looked at macroscopic properties predictions based on intermolecular potentials derived from ab initio calculations. Intermolecular potentials can be obtained from ab initio methods by calculating the interaction energy of several hundreds of molecular configurations between molecular pairs. These interaction energy can then be fitted to a potential function suitable to molecular simulations. This was another area of Dr. Sum's research, using the developed ab initio potential in Gibbs ensemble Monte Carlo (GEMC) simulations in order to predict thermodynamic properties of the system. In addition, Dr. Sum also looked at the deficiencies of the ab initio derived potentials and applied correction to account for multibody interactions, which is explictly excluded from the pair potentials, since these are not effective potentials as most commonly used force-fields. Multibody interactions are added with a polarizable model, where the induced dipole of a molecule generates an electric field that affects all other molecules in the system. The inclusion of multibody effects with a polarizable model adds substantially to the computational load of the simulations, therefore, Dr. Sum also implemented parallel libraries to our code in order to perform the computations as efficiently and timely as possible. The parallelization of our code is being done with the MPI library routines, which are widely used and become almost as a standard in scientific parallel computing.

Publications

  1. Amadeu K. Sum, Roland Faller, and Juan J. de Pablo.
    "Molecular Simulation Study of Phospholipid Bilayers and Insights of the Interactions with Disaccharides,".
    Biophysical Journal 85(5):2830-2844, November 2003. [BiophysJ]
  2. Amadeu K. Sum and Juan J. de Pablo.
    "Molecular Simulation Study on the Influence of Dimethylsulfoxide on the Structure of Phospholipid Bilayers".
    Biophysical Journal 85(6):3636-3645, December 2003. [BiophysJ]
  3. Amadeu K. Sum, Mary J. Biddy, Juan J. de Pablo, and Michael J. Tupy.
    "Predictive Molecular Model for the Thermodynamic and Transport Properties of Triacylglycerols".
    Journal of Physical Chemistry B 107(51):14443--14451, December 2, 2003. [doi:10.1021/jp035906g] [ACS]
  4. Rajesh Khare, Amadeu K. Sum, Shyamal K. Nath, and Juan J. de Pablo.
    "Simulation of Vapor-Liquid Phase Equilibria of Primary Alcohols and Alcohol-Alkane Mixtures".
    Journal of Physical Chemistry B 108(28) 10071--10076, June 10, 2004. [doi:10.1021/jp048144d] [ACS]
  5. L. de Pablo, M. L. Chávez, Amadeu K. Sum, and Juan J. de Pablo.
    "Monte Carlo molecular simulation of the hydration of Na-montmorillonite at reservoir conditions".
    Journal of Chemical Physics 120(2):939--946, January 8, 2004. [doi:10.1063/1.1631440] [AIP/JCP]
  6. Emmanouil Doxastakis, Amadeu K. Sum, and Juan J. de Pablo.
    "Modulating Membrane Properties: The Effect of Trehalose and Cholesterol on a Phospholipid Bilayer".
    Journal of Physical Chemistry B 109(50):24173--24181, November 25, 2005. [doi:10.1021/jp054843u] [ACS]

Contact

Dr. Amadeu K. Sum
Department of Chemical Engineering
Colorado School of Mines
331 Alderson Hall
Golden, Colorado 80401
U.S.A.
email: asum@mines.edu
http://www.mines.edu/~asum/
tel: +1 303 273-3873
fax: +1 303 273-3730