Alkane/Water Partition Coefficient Calculation Based on the Modified AM1 Method and Internal Hydrogen Bonding Sampling Using COSMO-RS.

We introduce a physics-based model for calculating partition coefficients of solutes between water and alkanes, using a combination of a semi-empirical method for COSMO charge density calculation and statistical sampling of internal hydrogen bonds (IHBs). We validate the model on the experimental partition data (∼3500 molecules) of small organics, drug-like molecules, and statistical assessment of modeling of proteins and ligand drugs. The model combines two novel algorithms: a bond-correction method for improving the calculation of COSMO charge density from AM1 calculations and a sampling method to deal with IHBs. From a comparison of simulated and experimental partition coefficients, we find a root-mean-square deviation of roughly one log 10 unit. From IHB analysis, we know that IHBs can be present in two states: open (in water) and closed (in apolar solvent). The difference can lead to a shift of as much as two log 10 units per IHB; not taking this effect into account can lead to substantial errors. The method takes a few minutes of calculation time on a single core, per molecule. Although this is still much slower than quantitative structure-activity relationship, it is much faster than molecular simulations and can be readily incorporated into any screening method.

Coarse-Grained Models for Automated Fragmentation and Parametrization of Molecular Databases.

We calibrate coarse-grained interaction potentials suitable for screening large data sets in top-down fashion. Three new algorithms are introduced: (i) automated decomposition of molecules into coarse-grained units (fragmentation); (ii) Coarse-Grained Reference Interaction Site Model-Hypernetted Chain (CG RISM-HNC) as an intermediate proxy for dissipative particle dynamics (DPD); and (iii) a simple top-down coarse-grained interaction potential/model based on activity coefficient theories from engineering (using COSMO-RS). We find that the fragment distribution follows Zipf and Heaps scaling laws. The accuracy in Gibbs energy of mixing calculations is a few tenths of a kilocalorie per mole. As a final proof of principle, we use full coarse-grained sampling through DPD thermodynamics integration to calculate log POW for 4627 compounds with an average error of 0.84 log unit. The computational speeds per calculation are a few seconds for CG RISM-HNC and a few minutes for DPD thermodynamic integration.

Method of moments for computational microemulsion analysis and prediction in tertiary oil recovery.

We discuss the application of Helfrich's surface torque density concept to microemulsion design and analysis from three different angles: (i) from the point of view of coarse-grained molecular simulations, using Dissipative Particle Dynamics, including charge interactions and added salt, (ii) using an approximate double-film model for the surface, and (iii) comparison with formulation approaches. The simulations use that the surface torque can be calculated unambiguously from the stress profile, provided the surface is tensionless. Very good agreement is found on predicting optimal salinity (or the absence of that) for a range of surfactants: dioctyl sodium sulfosuccinate, various twin-tailed sulfonates and sodium dodecyl sulfate. The simulations are very fast, on par with times for experiments, thus they could lead to a practical tool for discovery of more efficient surfactants, although much remains to be done with respect to other important variables: oil composition, surfactant mixtures, aggregation in solution, and so on. The microscopic model (second approach) is highly approximate: it is essentially based on two opposing swelling tendencies, that are both of osmotic nature. In accordance with the model, the tails are swollen by the oil and the charged head groups are confined in a salty layer in Donnan equilibrium with the salt solution. In this way, the surface interactions are purely entropic. The comparison of the film model with existing formulation approaches (third approach) covers the interfacial tension minimum, Winsor R theory, quantitative structure property relations (QSPR), hydrophilic-lipophilic deviation (HLD), HLD-net average curvature, and temperature coefficients. Using the surface torque analysis, we succeed in deriving in an ab initio way QSPR empirical coefficients that have been known for decades, but until now, have been obscure in origin.

A complete multiscale modelling approach for polymer-clay nanocomposites.

We present an innovative, multiscale computational approach to probe the behaviour of polymer-clay nanocomposites (PCNs). Our modeling recipe is based on 1) quantum/force-field-based atomistic simulation to derive interaction energies among all system components; 2) mapping of these values onto mesoscopic bead-field (MBF) hybrid-method parameters; 3) mesoscopic simulations to determine system density distributions and morphologies (i.e., intercalated versus exfoliated); and 4) simulations at finite-element levels to calculate the relative macroscopic properties. The entire computational procedure has been applied to two well-known PCN systems, namely Nylon 6/Cloisite 20A and Nylon 6/Cloisite 30B, as test materials, and their mechanical properties were predicted in excellent agreement with the available experimental data. Importantly, our methodology is a truly bottom-up approach, and no "learning from experiment" was needed in any step of the entire procedure.

Calculation of Diffusion Coefficients through Coarse-Grained Simulations Using the Automated-Fragmentation-Parametrization Method and the Recovery of Wilke-Chang Statistical Correlation.

We introduce a model for the calculation of diffusion coefficients using dissipative particle dynamics coarse-grained molecular simulations. We validate the model on experimental diffusion data of small organics and drug-like molecules in water. The new model relies on our automated-fragmentation-parametrization protocol for cutting molecules into fragments, which are calibrated using the COSMO-RS thermodynamic model ( J. Chem. Inf. MODEL: 2016 , 56 ( 12 ), 2361 - 2377 , DOI: 10.1021/acs.jcim.6b00003 ). By simulations over the entire CULGI database of more than 11000 molecules, we recover the decades-old empirical Wilke-Chang correlation between diffusion coefficient and molar volume. We believe this is the first demonstration of the correlation by simulation or theory. From a comparison of simulated and experimental diffusion coefficients, we find that one full time unit of coarse-grained simulation equals 64 ± 13 ps real time.

Accuracy Test of the OPLS-AA Force Field for Calculating Free Energies of Mixing and Comparison with PAC-MAC.

We have calculated the excess free energy of mixing of 1053 binary mixtures with the OPLS-AA force field using two different methods: thermodynamic integration (TI) of molecular dynamics simulations and the Pair Configuration to Molecular Activity Coefficient (PAC-MAC) method. PAC-MAC is a force field based quasi-chemical method for predicting miscibility properties of various binary mixtures. The TI calculations yield a root mean squared error (RMSE) compared to experimental data of 0.132 kBT (0.37 kJ/mol). PAC-MAC shows a RMSE of 0.151 kBT with a calculation speed being potentially 1.0 × 104 times greater than TI. OPLS-AA force field parameters are optimized using PAC-MAC based on vapor-liquid equilibrium data, instead of enthalpies of vaporization or densities. The RMSE of PAC-MAC is reduced to 0.099 kBT by optimizing 50 force field parameters. The resulting OPLS-PM force field has a comparable accuracy as the OPLS-AA force field in the calculation of mixing free energies using TI.

Force-Field Based Quasi-Chemical Method for Rapid Evaluation of Binary Phase Diagrams.

We present the Pair Configurations to Molecular Activity Coefficients (PAC-MAC) method. The method is based on the pair sampling technique of Blanco (Fan, C. F.; Olafson, B. D.; Blanco, M.; Hsu, S. L. Application of Molecular Simulation to Derive Phase Diagrams of Binary Mixtures. Macromolecules 1992, 25, 3667-3676) with an extension that takes the packing of the molecules into account by a free energy model. The intermolecular energy is calculated using classical force fields. PAC-MAC is able to predict activity coefficients and corresponding vapor-liquid equilibrium diagrams at least 4 orders of magnitude faster than molecular simulations. The accuracy of the PAC-MAC method is tested by comparing the results with experimental data and with the results of the COSMO-SAC model (Lin, S.-T.; Sandler, S. I. A Priori Phase Equilibrium Prediction from a Segment Contribution Solvation Model. Ind. Eng. Chem. Res. 2002, 41, 899-913). PAC-MAC (using the OPLS-aa force field) is shown to be comparable in accuracy to COSMO-SAC, at the considerable advantage that PAC-MAC in principle does not require quantum calculation, provided proper force fields to be available.

Electric field versus surface alignment in confined films of a diblock copolymer melt.

The dynamics of alignment of microstructure in confined films of diblock copolymer melts in the presence of an external electric field was studied numerically. We consider in detail a symmetric diblock copolymer melt, exhibiting a lamellar morphology. The method used is a dynamic mean-field density functional method, derived from the generalized time-dependent Ginzburg-Landau theory. The time evolution of concentration variables and therefore the alignment kinetics of the morphologies are described by a set of stochastic equations of a diffusion form with Gaussian noise. We investigated the effect of an electric field on block copolymers under the assumption that the long-range dipolar interaction induced by the fluctuations of composition pattern is a dominant mechanism of electric-field-induced domain alignment. The interactions with bounding electrode surfaces were taken into account as short-range interactions resulting in an additional term in the free energy of the sample. This term contributes only in the vicinity of the surfaces. The surfaces and the electric field compete with each other and align the microstructure in perpendicular directions. Depending on the ratio between electric field and interfacial interactions, parallel or perpendicular lamellar orientations were observed. The time scale of the electric-field-induced alignment is much larger than the time scale of the surface-induced alignment and microphase separation.