(329g) Study of Phase Transitions By 2-D Density of States Simulation

Authors: 
Gai, L., Vanderbilt University
Iacovella, C. R., Vanderbilt University
Cummings, P. T., Vanderbilt University
McCabe, C., Vanderbilt University



We have developed a hybrid Wang-Landau (WL)/statistical temperature molecular dynamics (STMD) method to simulate a 2D joint density of states (JDOS). The 2D JDOS [1] of continuous models is an important quantity as the free energy can be calculated not only as a function of temperature, but also as a function of concentration or pressure. However, determining the JDOS requires significant simulation time and is impractical with the original Wang-Landau (WL) method [2]. Previously, we have examined the phase behavior of self-assembling lipids as a function of temperature, via the statistical temperature molecular dynamics (STMD) [3] and demonstrated improved efficiency compared to the original WL method. STMD is an extension of the WL method that is suitable for simulations with large energy bins and can generate a deterministic MD trajectory to extend the WL method to complicated systems beyond the scope of effective MC moves. Building on earlier work [4-5], we have developed a hybrid WL /STMD method to simulate the 2D JDOS of energy and lipid concentration in the semi-grand canonical ensemble. The lipid model analyzed consists of 3 coarse-grained segments solvated by water. The phase behavior of the lipids with respect to bilayer formation has been characterized through the calculation of the heat capacity as a function of both temperature and lipid concentration. The method enables simulations to be performed to very low temperatures, with the whole system exhibiting well-ordered structures. A full view of the self-assembling bilayer phases are observed within the temperature range studied along each concentration, including gel phase bilayers with frozen water, with mixed water (i.e., frozen and liquid water), and with liquid water and a more fluid bilayer with liquid water.  The results obtained at selected concentration are compared to those from STMD with independent runs at the same concentration, and are consistently in excellent agreement with each other.

1.  C. Zhou, T.C. Schulthess, S. Torbrügge, and D.P. Landau, “Wang-Landau algorithm for continuous models and joint density of states,” Phys. Rev. Lett. 96 (2006) 120201

2. F. Wang and D. P. Landau, “Efficient, Multiple-Range Random Walk Algorithm to Calculate the Density of States,” Phys. Rev. Lett. 86 (2001) 2050–2053.

3.  J. Kim, J. E. Straub and T. Keyes, “Statistical-Temperature Monte Carlo and Molecular Dynamics Algorithms,” Phys. Rev. Lett. 97 (2006) 050601

4. L. Gai, K. Maerzke, P.T. Cummings and C. McCabe, “A Wang-Laudau study of Bilayer formation on a lattice model,” J. Chem. Phys. 137 (2012) 144901

5. L. Gai, T. Vogel, K. Maerzke, C. Iacovella, D.P. Landau, P.T. Cummings, and C. McCabe, “Examining the Phase transition behavior of amphiphilic lipids in solution using Statistical Temperature Molecular Dynamics and Replica-Exchange Wang-Landau Methods,” J. Chem. Phys. submitted.