TOC graphic from Lara Patel’s paper on MTBE<\/a><\/p>\n We are applying these methods in united-atom simulations of surfactants (both neutral and ionic) and simulations of surfactant-peptide interactions.<\/p>\n TOC graphic from Xiaokun and Jorge’s paper on octyl phosphocholine<\/a><\/p>\n More information on a collaboration with the Bester-Rogac group at the University of Ljubljana is here<\/a>.<\/p>\n Snapshot of a partially folded peptide embedded in an octyl phosphocholine micelle, work by Jorge Arce.<\/p>\n <\/p>\n Methodological extensions that we are targeting include careful analysis of non-ideal effects and a hybrid PEACH-Bennett Acceptance Ratio (BAR) method for slowly exchanging clusters.<\/p>\n <\/p>\n Grand Canonical Monte Carlo for Condensed Phases<\/strong><\/p>\n From Ziwei’s paper in Langmuir<\/a><\/p>\n Monte Carlo simulations in the grand canonical ensemble are performed with a variable number of particles of one or more type. Trial moves performed over the course of the simulation generate new structures in which one or more particles are added, removed, or exchanged for particles of a different type. The probability for accepting one of these trial moves depends both on the resulting change in the system’s potential energy (as in canonical MC) and on the chemical potentials (represented by the symbol \u00b5) of all particles involved.<\/p>\n There are a few reasons why grand canonical Monte Carlo (GCMC) simulations can be a useful alternative to simulations that fix the number of particles. One reason is that one learns how \u00b5 depends on the system density and composition. This is particularly useful when simulating a system whose composition is determined by exchange of particles with a reservoir \u2013 for instance, molecules sticking to a surface or embedded in pores, where the surface coverage or pore loading is dictated by the concentration of molecules in the gas or solution above the surface or in contact with the pores. A discontinuity in composition with changing \u00b5 is also a useful indication of a first-order phase transition. Another reason is that exchanging particles with the imaginary reservoir leads to equilibration of composition across all of the microenvironments in the system being studied, and in some cases this equilibration is more efficient than the equilibration that happens through particles diffusing across the system.<\/p>\n Conventional GCMC involves attempts to insert a particle to a random position in the simulation box. In a dense system (for instance, a liquid or solution) these moves are very unlikely to succeed because a randomly chosen position is likely to be occupied by another particle, and the energy cost to add two particles very close to each other is very high. This energy cost can be lower when a particle is exchanged for a particle of a different type (in a “semi-grand canonical” type of simulation), but only if the particles are sufficiently similar in size, shape, and interactions.<\/p>\n Our goal is to develop GCMC methods that allow a particle (and if necessary, its local solvent shell) to be exchanged for a number of solvent molecules. A key challenge for these methods is finding ways to pack solvent into a cavity that are efficient and that satisfy detailed balance. In a “proof of principle” study, Solvent Repacking Monte Carlo was relatively successful for exploring the phase behavior of a 2-dimensional binary hard disk mixture.\u00a0 (See here<\/a>.) We are working both to apply this method to related systems and understand ordering transitions of hard particles on curved surfaces, and to extend the range of practical applicability. These techniques have been applied to 3-d Lennard-Jones mixtures<\/a>\u00a0and to study ordering transitions in colloidal monolayers<\/a>. Ultimately our goal is to be able to perform GCMC on solutes of moderate size in explicit water.<\/p>\n <\/p>\n <\/p>\n <\/p>\n <\/p>\n Peptide-lipid interactions<\/b><\/p>\n We are using MD simulations to study the dynamics of folding and insertion of antimicrobial peptides at lipid bilayers, in collaboration with the Dyer group<\/a>.<\/p>\n <\/p>\n Phase behavior of lipid bilayers<\/strong><\/p>\n Lipid bilayers undergo phase transitions between ordered (gel or ripple) and disordered (fluid) structures as temperature is varied. \u00a0Earlier work in the group addressed methods to study demixing effects associated with ordering transitions in bilayer mixtures (here<\/a> and here<\/a>), to identify the actual transition temperature of a simulation model (here<\/a>), and to characterize how lipids are arranged within the gel phase (here<\/a>). \u00a0 We have been working on several questions related to the structure and properties of bilayers near their transition temperatures:<\/p>\n Anomalous permeability<\/em>. \u00a0Molecules pass through the bilayer much more easily in the fluid state than in the gel state. \u00a0Experiments dating back to the early 1970’s have shown that the permeability is greatest near the transition temperature. \u00a0Lewen Yang has been working on simulations that have helped us develop a new understanding for this anomaly – inserting a molecule into a fluid region that is surrounded by ordered lipids will make the fluid domain shrink, relieving some interfacial energy. \u00a0For more, see here<\/a> and here<\/a><\/p>\n <\/p>\n Melting dynamics of unilamellar vesicles. <\/i>\u00a0We have been working to understand the dynamics of vesicles subject to an ultrafast temperature jump that changes their phase, as studied in experiments<\/a>\u00a0by the Dyer group<\/a>\u00a0here at Emory. \u00a0Images from Ph.D. student Lara Patel’s research are shown below.\u00a0<\/i>For more, see here (pdf)<\/a>\u00a0or here (link)<\/a>.<\/p>\n Rheology of telechelic hydrogels<\/strong>\u00a0 The mechanical properties of self-assembled transient networks are characterized by a relaxation time; roughly speaking, they behave solid-like over time periods shorter than this relaxation time\u00a0and liquid-like over longer time periods. \u00a0The stiffness of the solid-like behavior is described by a plateau shear modulus, which is sensitive to the degree of connectivity within the network.\u00a0\u00a0Through simulation and theory, group alumna Ana West uncovered an interesting relationship between the plateau shear modulus and the shear relaxation time in self-assembled transient networks. \u00a0The model fits (some, not all) experimental data from the literature rather nicely. \u00a0 The results are described here<\/a>.<\/p>\n <\/p>\n Kindt Group on Youtube:\u00a0Molecular animations as teaching tools<\/b><\/p>\n![\"\"](\"https:\/\/scholarblogs.emory.edu\/kindtgroup\/files\/2015\/12\/TOC.png\")
<\/a><\/p>\n![\"\"](\"https:\/\/scholarblogs.emory.edu\/kindtgroup\/files\/2019\/02\/FPE-TOC.png\")
<\/a><\/p>\n![\"\"](\"https:\/\/scholarblogs.emory.edu\/kindtgroup\/files\/2019\/02\/CM15-embedded-in-OPC.png\")
<\/a><\/p>\n![\"\"](\"https:\/\/scholarblogs.emory.edu\/kindtgroup\/files\/2018\/08\/Figure-8.png\")
<\/a><\/p>\n![\"d3-5e34A-box\"](\"http:\/\/scholarblogs.emory.edu\/kindtgroup\/files\/2015\/12\/d3-5e34A-box.png\")
\u00a0 \u00a0![\"d3-detail\"](\"http:\/\/scholarblogs.emory.edu\/kindtgroup\/files\/2015\/12\/d3-detail.png\")
<\/a>Snapshot and detail from Solvent Repacking GCMC simulations of a bidisperse hard-disk mixture, with large disks (d=3) shown in white and small disks (d=1) color-coded according to complex value of the S6<\/sub>\u00a0order parameter. \u00a0(Intensity of color correlates to degree of local hexagonal order; hue correlates to the local packing direction.)<\/p>\n![\"graphic\"](\"http:\/\/scholarblogs.emory.edu\/kindtgroup\/files\/2015\/12\/graphic.png\")
<\/a><\/p>\n![\"small-vesicle-melting\"](\"http:\/\/scholarblogs.emory.edu\/kindtgroup\/files\/2015\/12\/small-vesicle-melting.png\")
<\/a><\/p>\n