United Atom Model Research Articles

Coarse–grained molecular modeling of the microphase structure of polyurea elastomer

We present a structure–matching coarse–grained model of polyurea, similar to united atom models, in which hydrogen atoms are implicitly represented. The model was trained using iteration Boltzmann inversion and a new heuristically–determined, distance–dependent scaling function that dramatically reduces the iterations required. With its reduced complexity and accelerated dynamics, the coarse–grained model can simulate microphase separation with hard domain spacing of 5 nm, comparable to x-ray scattering measurements of similar polyurea elastomers. An analysis of the morphology of two model systems shows a large, interconnected hard domain within a multiblock system, compared to an interrupted hard phase composed of separate smaller ribbon–shaped domains in a diblock system. To analyze the topology of soft segment connectivity, we calculated their end–to–end distribution, revealing that soft segments are composed of a large population of bridge–like segments and a smaller population of loop–like segments.

Prediction of PEKK properties related to crystallization by molecular dynamics simulations with a united-atom model

Semicrystalline polyetherketoneketone (PEKK) is widely used as the matrix in carbon-fiber composites. Understanding and predicting the crystallization thermodynamics and kinetics of this class of polymers is, thus, of great interest. This paper uses molecular dynamic (MD) simulations using Dreiding united-atom (UA) potentials to characterize a wide range of thermos-physical properties of PEKK. We characterized the effect of terephthaloyl chloride to isophthaloyl chloride (T/I) on an extensive set of properties, including the lattice parameters and stability of PEKK crystal structures, glass transition temperature, melting temperature, crystal/amorphous interfacial energy, and enthalpy of fusion. We find good overall agreement between predicted properties and experimental values and the simulations help clarify inconsistencies in the literature. In combination with classical nucleation theory, nucleation barriers and critical nucleus size at different temperature are predicted.

Pair and many-body interactions between ligated Au nanoparticles.

We report the results of molecular dynamics simulations of the properties of a pseudo-atom (united atom) model of dodecane thiol ligated 5-nm diameter gold nanoparticles (AuNPs) in a vacuum as a function of ligand coverage and particle separation in three states of aggregation, namely, the isolated AuNPs, the isolated pair of AuNPs, and a square lattice of four AuNPs. Our calculations show that the ligand density along a radius emanating from the core of an isolated AuNP has the same gross features for all values of the coverage; it oscillates around a constant value up to a distance along the chain corresponding to the position of the fourth pseudo-atom and then smoothly decays to zero, reflecting both the restricted conformations of the chain near the core surface and the larger numbers of conformations available further from the core. Interaction between two AuNPs generates changes in the ligand distributions of each. We examine the structure and general shape of the ligand envelope as a function of the coverage and demonstrate that the equilibrium structure of the envelope and the deformation of that envelope generated by interaction between the NPs are coverage-dependent so that the shape, depth, and position of the minimum of the potential of mean force display a systematic dependence on the ligand coverage. We propose an accurate analytical description of the calculated potential of mean force as a function of a set of parameters that scale linearly with the ligand coverage. Noting that the conformational freedom of the ligands implies that multiparticle induced deviations from additivity of the pair potential of mean force are likely important; we define and calculate a "bond stretching" effective pair potential of mean force for a square lattice of particles that contains, implicitly, both the three- and four-NP contributions. We find that the bond stretching effective pair potential of mean force in this cluster has a different minimum and a different well depth from the isolated pair potential of mean force. Previous work has found that the three-particle contribution to deviation from pair additivity is monotonically repulsive, whereas we find that the combined three- and four-particle contributions have an attractive well, implying that the three- and four-particle contributions are of comparable magnitude but opposite sign, thereby suggesting that even higher order correction terms likely play a significant role in the behavior of dense assemblies of many nanoparticles.

Origin of the Difference in Ion-Water Distances Determined by X-ray and Neutron Diffraction Measurements for Aqueous NaCl and KCl Solutions

Abstract Experimental evidence has been presented on the difference in intermolecular ion-water distances obtained from X-ray and neutron diffraction methods. Simultaneous least squares fitting procedures were performed for X-ray and neutron interference terms observed for (NaCl)x(*H2O)1−x, (x = 0, 0.02, 0.05, and 0.098) and (KCl)x(*H2O)1−x, (x = 0, 0.02, 0.05, and 0.075) solutions at 25 °C, respectively. The null-water mixture was employed for neutron diffraction measurements for these solutions to eliminate structural contribution from hydrogen atoms. It has been revealed that the hydration numbers of Na+ and K+ are concentration dependent and the values for lower-concentration limit are 5 and 6, respectively. The nearest neighbor Na+⋯H2O and K+⋯H2O distances are obtained to be 2.36–2.37 and 2.75–2.82 Å, respectively. In order to examine the effect of the separate treatment of interactions between ion-oxygen and ion-hydrogen atoms in the X-ray model function, simultaneous fitting procedures were carried out for X-ray and neutron diffraction data observed for 9.8 mol% NaCl and 7.5 mol% KCl solutions by employing the individual atom model for the X-ray interference term. Obtained Na+⋯O and K+⋯O distances are ca. 0.02 Å shorter than those determined by the simultaneous fit employing the usual united model for water molecules. The nearest neighbor Cl−⋯O distance derived from the simultaneous fit by means of the individual atom model exhibits ca. 0.1 Å shorter than that obtained from the fit using the united atom model. The present Cl−⋯O distance agrees with that obtained from neutron diffraction measurements on 35Cl/37Cl isotopically substituted aqueous 5 mol% Na*Cl solutions in D2O. The simultaneous fitting analyses employing X-ray model function with the united and individual atom models of water molecule have revealed that the ion-oxygen (water) internuclear distance is significantly shorter than the average separation of electron clouds between ion and neighboring water molecule. The present results indicate that the difference in ion-water distance observed from X-ray and neutron diffraction studies mainly arises from the united atom model of X-ray diffraction data analysis assuming a spherical electron density around oxygen atom within the water molecules.

Molecular Dynamics Study on Transmission Mechanism of Torsional Deformation in Cellulose Nanofibers with Hierarchical Structure

Cellulose nanofiber (CNF) is a fibrous and nano-sized substance produced by decomposition of bulk-type cellulose which is a main component of plants. It has high strength comparable to steel, and it shows low environmental load during a cycle of production and disposal. Besides it has many excellent properties and functions such as high rigidity, light-weight, flexibility and shape memory effect, so it is expected as a next-generation new material. Usually it is composed of many cellulose micro fibrils (CMFs) in which molecular chains of cellulose are aggregated in a crystal structure, the knowledge of mechanical properties for each CMF unit is important. Since actual fibrils are complicatedly intertwined, it is also crucial to elucidate the transmission mechanism of force and deformation not only in one fibril but also in between fibrils. How the dynamic and hierarchical structure composed of CMFs responds to bending or torsion is an interesting issue. However, little is known on torsional characteristics (shear modulus, torsional rigidity, etc.) concerning CMF. In general, in a wire-like structure, it is difficult to enhance torsional rigidity and strength, compared with tensile ones. Therefore, in this study, we try to build a hierarchical model of CNF by multiplying CMF fibers and to conduct molecular dynamics simulation for torsional deformation, by using hybrid model between all-atom and united-atoms model. First, shear modulus was estimated for one CMF fibril and it showed a value close to the experimental values. Also, we assume a state in which two CMFs are ideally arranged in parallel, and create a hierarchical structure. We evaluate the dependence on the temperature for the bond strength and toughness in the hierarchical structures. Furthermore, we mentioned the transmission mechanism between components of a hierarchical structure.

Molecular models for Sodium Dodecyl Sulphate in aqueous solution to reduce the micelle time formation in molecular simulation

In this work two Sodium Dodecyl Sulphate (SDS) molecular models are proposed; one based on the Explicit-Atom (EA) approach while the other on the United-Atom (UA) approach. A SDS aqueous solution was simulated using the TIP4P/ε potential for water. With the EA model the micellar aggregate spent less than 1 ns of simulation time to be formed. This is a quite low simulation time compared to the 23 ns spent by existing SDS models to attain the same micellar formation. When both proposed models are compared, micellar formation from EA model is 10 times faster than the UA model. The time reduction is due to the implementation of two repulsive sites in the carbon atoms on the hydrophobic tails. To validate the models, density, radial distribution functions, radius of gyration and shear viscosity were calculated; results were in good agreement with those reported in literature.

Self-diffusion coefficient and viscosity of methane and carbon dioxide via molecular dynamics simulations based on new ab initio-derived force fields

In the present study, the self-diffusion coefficient and viscosity of methane (CH4) and carbon dioxide (CO2) were studied by molecular dynamics (MD) simulations in the superheated vapor, gaseous, and supercritical state. The parameters used for the MD simulations are based on new molecular force fields (FFs), which were derived from previously developed, highly accurate pair potentials based on ab initio-calculated interaction energies in the limit of zero density. Using these optimized rigid all-atom FFs, multiple MD simulation runs in the order of several μs were performed to evaluate the dynamic viscosity from the plateau of the time integral of the pressure autocorrelation function and the self-diffusion coefficient from the linear Einstein regime. For the latter property, it is shown that the Yeh-Hummer correction accounting for effects of the finite box size for dense liquid systems in the hydrodynamic regime can be transferred consistently to gaseous systems at various densities. A comparison of the simulated dynamical properties obtained from our ab initio-derived FFs with those obtained from established literature FFs showed that our suggested approach is superior to the other rigid all-atom approaches and in particular to flexible and united-atom models over density ranges corresponding to pressures between 0.1 and 10 MPa. For temperatures of 295, 325, and 355 K, our simulation results for the product of self-diffusion coefficient and density as well as for the dynamic viscosity with average expanded statistical uncertainties (k = 2) of 0.6% as well as 8.0%, respectively, represent the density dependency of both properties and agree with the few simulation and experimental data available in the literature.

Monte Carlo Algorithm Based on Internal Bridging Moves for the Atomistic Simulation of Thiophene Oligomers and Polymers

We introduce a powerful Monte Carlo (MC) algorithm for the atomistic simulation of bulk models of oligo- and polythiophenes by redesigning MC moves originally developed for considerably simpler polymer structures and architectures, such as linear and branched polyethylene, to account for the ring structure of the thiophene monomer. Elementary MC moves implemented include bias reptation of an end thiophene ring, flip of an internal thiophene ring, rotation of an end thiophene ring, concerted rotation of three thiophene rings, rigid translation of an entire molecule, rotation of an entire molecule, and volume fluctuation. In the implementation of all moves we assume that thiophene ring atoms remain rigid and strictly coplanar; on the other hand, inter-ring torsion and bond bending angles are left fully flexible subject to suitable potential energy functions. Test simulations with the new algorithm of an important thiophene oligomer, α-sexithiophene (α-6T), at a high enough temperature (above its isotropic-to-nematic phase transition) using a new united atom model specifically developed for the purpose of this work provide predictions for the volumetric, conformational, and structural properties that are remarkably close to those obtained from detailed atomistic molecular dynamics (MD) simulations using an all-atom model. The new algorithm is particularly promising for addressing the rich (and largely unexplored) phase behavior and nanoscale ordering of very long (also more complex) thiophene-based polymers which cannot be accessed by conventional MD methods due to the extremely long relaxation times characterizing chain dynamics in these systems.

On the Structure Factors of Aqueous Mixtures of 1‐Propanol and 2‐Propanol: X‐Ray Diffraction Experiments and Molecular Dynamics Simulations

The structure factor of pure 1‐propanol, 2‐propanol, and mixtures of 1‐propanol/water and 2‐propanol/water, as a function of composition, has been determined experimentally and by molecular dynamics simulations. The primary aim is to find interatomic potentials that reproduce measured structural data at the highest possible level. For this reason, various alcohol potential models have been employed, including united atom (UA) and all atom (AA) types, in combination with a TIP4P‐based model for water. In order to improve agreement with experimental values of the dielectric constant and mass density, a new UA force field for the alcohols has also been constructed. In terms of structural properties, the AA model reproduces experimental results better than any of the UA models for all compositions.

SAFT-γ Force Field for the Simulation of Molecular Fluids. 5. Hetero-Group Coarse-Grained Models of Linear Alkanes and the Importance of Intramolecular Interactions.

The SAFT-γ Mie group-contribution equation of state [ Papaioannou J. Chem. Phys. 2014 , 140 , 054107 ] is used to develop a transferable coarse-grained (CG) force-field suitable for the molecular simulation of linear alkanes. A heterogroup model is fashioned at the resolution of three carbon atoms per bead in which different Mie (generalized Lennard-Jones) interactions are used to characterize the terminal (CH3-CH2-CH2-) and middle (-CH2-CH2-CH2-) beads. The force field is developed by combining the SAFT-γ CG top-down approach [ Avendaño J. Phys. Chem. B 2011 , 115 , 11154 ], using experimental phase-equilibrium data for n-alkanes ranging from n-nonane to n-pentadecane to parametrize the intermolecular (nonbonded) bead-bead interactions, with a bottom-up approach relying on simulations based on the higher resolution TraPPE united-atom (UA) model [ Martin ; , Siepmann J. Phys. Chem. B 1998 , 102 , 2569 ] to establish the intramolecular (bonded) interactions. The transferability of the SAFT-γ CG model is assessed from a detailed examination of the properties of linear alkanes ranging from n-hexane ( n-C6H14) to n-octadecane ( n-C18H38), including an additional evaluation of the reliability of the description for longer chains such as n-hexacontane ( n-C60H122) and a prototypical linear polyethylene of moderate molecular weight ( n-C900H1802). A variety of structural, thermodynamic, and transport properties are examined, including the pair distribution functions, vapor-liquid equilibria, interfacial tension, viscosity, and diffusivity. Particular focus is placed on the impact of incorporating intramolecular interactions on the accuracy, transferability, and representability of the CG model. The novel SAFT-γ CG force field is shown to provide a reliable description of the thermophysical properties of the n-alkanes, in most cases at a level comparable to the that obtained with higher resolution models.

The effect of alumina nanoparticles on the thermal properties of PMMA: a molecular dynamics simulation

ABSTRACTMetal oxides, as one of the most promising flame retardant additives, improve the fire retardant and the thermal stability properties of polymers. In the present study, molecular dynamics (MD) simulations based on the united atom model were applied to study the effect of alumina nanoparticles on the density, thermal conductivity, heat capacity, and thermal diffusivity of isotactic poly(methyl methacrylate) (is-PMMA). Thermal diffusivity of PMMA and PMMA/alumina nanocomposite were investigated through calculating thermal conductivity, density and heat capacity in the range of 300–700 K. Heat capacity can be calculated using fluctuations properties of energy. Thermal conductivity was calculated through the nonequilibrium molecular dynamics (NEMD) simulation by Fourier’s law approach. Our results show that the addition of alumina nanoparticles decreases the heat capacity and increases the glass transition temperature (Tg), thermal conductivity and thermal diffusivity of the PMMA. Therefore, the addition of alumina nanoparticles to PMMA improves the fire retardancy of the polymer. In addition, we illustrate the links between the intermolecular and bulk properties of PMMA in the presence of the alumina nanoparticles.

Conformational and Dynamic Properties of Poly(ethylene oxide) in BMIM+BF4–: A Microsecond Computer Simulation Study Using ab Initio Force Fields

The behavior of polymers in complex solvents is interesting from a fundamental perspective and of practical importance from the standpoint of polymer processing. There has been recent interest in the conformational and dynamic properties of polymer in room temperature ionic liquids, with conflicting predictions from computations using models with different resolutions and conflicting results of experiments from different groups. In this work, we develop a first-principles, nonpolarizable united atom (UA) force field for a mixture of poly(ethylene oxide) (PEO) in the ionic liquid BMIM+BF4–. The UA force field is benchmarked against ab initio calculations, and the PEO atomic charges are parametrized to implicitly capture the polarization contribution to the solvation energy of a single PEO molecule in BMIM+BF4–. The UA model allows one to perform multi-microsecond molecular dynamics simulations. This is necessary because the conformational relaxation correlation times are of the order of 100 ns. The simulations predict that the radius of gyration, Rg, scales with molecular weight, Rg ∼ Mwν with ν ≈ 0.56 in the temperature range 300–600 K, consistent with experiment, seemingly in between a self-avoiding walk and an ideal chain. An examination of the snapshots of the polymer demonstrates, however, that the polymer conformations are composed of ringlike and linear segments, with ringlike parts of the chain wrapped around cations of the ionic liquid. The slow dynamics arises from the barrier to unwrapping the ringlike segments of the polymer. The mean-square displacement shows three regimes which we interpret as confinement, Zimm, and diffusive. The simulations emphasize the importance of accurate force fields and microsecond simulations in obtaining reliable results for polymers and elucidate important correlation effects for polymers in strongly interacting solvents.

The dependence of the size of confined water fluid molecules on the radius of carbon nanotube

A recently-derived perturbation-theory-based equation of state (EoS) has been used for describing the p-v-T behavior of water confined inside a carbon nanotube (CNT). The one-center united atom model (OCUA) was employed for modeling of water molecules inside the CNT with temperature-dependent effective diameter, σeff. The interaction between two nearest-neighbor molecules was taken into account as an effective pair potential (EPP) which is an average effective extended Lennard-Jones (12, 6, 3) (AEELJ) pair interactions in the framework of thermodynamic perturbation theory (TPT). A new concept as effective molecular diameter tensor (EMDT) was introduced to consider the anisotropic character of compression factor in the EoS. The values of σeff,zz, and σeff,xy of the EMDT for the axial and radial directions, respectively, in both filled and unfilled states of the CNT, were estimated by using the simulated p-v-T data in the EoS. The results mirror the fact that the σeff,zz, and σeff,xy values of water molecule increase with the radius of CNT (R). Moreover, the values of σeff,zz for the different radii of CNT are nearly close to those of σeff,xy. Also, the results show that the layers of water molecule formed in the filled state are of less space with respect to those of the unfilled state. Our results showed that the EMDT is of the importance to follow the changes of molecular shape of the nanoconfined fluid with the CNT's size.

Langevin dynamics simulation of crystallization of ring polymers.

We have studied the crystallization of ring polymers using Langevin dynamics simulations with a coarse-grained united atom model. We show that there are marked differences in the crystallization of single ring polymers in comparison to single linear polymers. Contrary to expectations from equilibrium thermodynamics, ring polymers melt at lower temperatures than linear polymers. An analysis of the early stage crystallization mechanism shows that ring and linear polymers crystallize through the birth of baby nuclei with their coarsening depending uniquely on their topology. The single ring polymers nucleate faster than the single linear analogs and into several metastable lamellar thicknesses, although the motion of the monomers in both cases is comparable. Additionally, using multiple polymer molecules, we find that the secondary nucleation of ring polymers proceeds with free energy barriers, as opposed to linear polymers where no barriers are found. Our results are in qualitative agreement with some experiments, while in disagreement with some other experiments, indicating additional roles by chemistries of ring and linear polymers. Our simulations are designed to explore only the topological effects without any consideration of non-universal chemical effects for our particular model.

Simulations of 1-Butyl-3-methylimidazolium Tetrafluoroborate + Acetonitrile Mixtures: Force-Field Validation and Frictional Characteristics.

To temper their prohibitively high viscosities, ionic liquids are commonly mixed with polar cosolvents to retain favorable physical properties and make them suitable for industrial applications. Molecular dynamics simulations of 1-butyl-3-methylimidazolium tetrafluoroborate ([Im41][BF4]) mixed with acetonitrile (CH3CN) are conducted primarily to test the accuracy of a composite force field (FF) and also to provide some insights into the solvation and frictional characteristics of this mixture. The FF combines the united-atom model for imidazolium ionic liquids of Zhong et al. [ J. Phys. Chem. B 2011, 115, 10027] with the acetonitrile model of Nikitin and Lyubartsev [ J. Comput. Chem. 2007, 28, 2020]. Comparison of simulated properties such as mixture densities, viscosities, electrical conductivities, and component diffusion coefficients to experimental data at 298 K shows that this combined FF provides reasonable accuracy for both static and dynamic properties. Component rotational dynamics, as well as those of a dilute benzene solute, probed via new 2H NMR T1 measurements, are also reasonably reproduced. Simulated coordination numbers reveal virtually random mixing between the [Im41][BF4] and CH3CN components in this system. Comparison of translational diffusion coefficients and rotation times to simple hydrodynamic predictions indicates that the friction on most motions becomes increasingly decoupled from solution viscosity as the [Im41][BF4] concentration increases.

Many Body Effects and Icosahedral Order in Superlattice Self-Assembly.

We elucidate how nanocrystals "bond" to form ordered structures. For that purpose we consider nanocrystal configurations consisting of regular polygons and polyhedra, which are the motifs that constitute single component and binary nanocrystal superlattices, and simulate them using united atom models. We compute the free energy and quantify many body effects, i.e., those that cannot be accounted for by pair potential (two-body) interactions, further showing that they arise from coalescing vortices of capping ligands. We find that such vortex textures exist for configurations with local coordination number ≤6. For higher coordination numbers, vortices are expelled and nanocrystals arrange in configurations with tetrahedral or icosahedral order. We provide explicit formulas for the optimal separations between nanocrystals, which correspond to the minima of the free energies. Our results quantitatively explain the structure of superlattice nanocrystals as reported in experiments and reveal how packing arguments, extended to include soft components, predict ordered nanocrystal aggregation.

Effects of ion concentration and solvent composition on the properties of water-methanol solutions of NaCl. NPT molecular dynamics computer simulation results

Isothermal-isobaric molecular dynamics simulations are used to examine the microscopic structure and other properties of a model solution consisting of NaCl salt dissolved in water-methanol mixture. The SPC/E water model and the united atom model for methanol are combined with the force field for ions by Dang [J. Amer. Chem. Soc., 1995, 117, 6954] to describe the entire system. Our principal focus is to study the effects of two variables, namely, the solvent composition and ion concentrations on the solution's density, on the structural properties, self-diffusion coefficients of the species and the dielectric constant. Moreover, we performed a detailed analysis of the first coordination numbers of the species. Trends of the behaviour of the average number of hydrogen bonds between solvent molecules are evaluated.

New Force Field Parameters for the Sodium Dodecyl Sulfate and Alpha Olefin Sulfonate Anionic Surfactants.

Molecular dynamics simulations were carried out to obtain new force field parameters of two most commonly used anionic surfactants: sodium dodecyl sulfate and alpha olefin sulfonate. The present united atom models, of those surfactants, fail to reproduce one of the most important thermodynamic properties, the surface tension. Therefore, by scaling the Lennard-Jones parameters, the potential well (ϵ) and the length (σ), we were able to fit the experimental data. The correct micelle structure of the surfactants was also captured with the new set of parameters. The new proposed united atom models of both surfactants were tested with two different water models, TIP4P/ϵ and SPC/E, and good agreement with actual experiments was found.

Microscopic picture of heat conduction in liquid ethylene glycol by molecular dynamics simulation: Difference from the monohydric case

The present study investigates the molecular-scale heat transfer in the liquid of ethylene glycol, which is widely used as heat transfer media. First, by combining existing molecular models, we developed a new united atom model of ethylene glycol, and showed that this model reasonably reproduces the experimental thermal conductivity. Using the non-equilibrium molecular dynamics simulations with this model, we characterized the heat transfers due to different kinds of inter- and intramolecular interactions on the basis of a picture that a single pair interaction is a path of heat transfer. These characteristics were compared with those of ethanol (Matsubara et al., 2017) to elucidate the molecular mechanism which realizes an enhanced thermal conductivity because of an additional hydroxylation on ethanol. The results indicate that the thermal conductivity enhancement occurs because the additional heat paths provided by the second hydroxyl group increases the amount of heat conduction owing to all of the van der Waals, Coulomb, and covalent interactions. In particular, the increase in the number of the paths associated with the intermolecular Coulomb interaction between the non-hydrogen bonding hydroxyl groups is prominent and consequently the Coulomb interaction, which is an efficient heat carrier, performs the largest amount of heat conduction in ethylene glycol. Although the second hydroxyl group also increases the number of hydrogen bonds, the direct heat transfer via the hydrogen bonds accounts for only a small portion of the total heat conduction. On the other hand, this augmentation of hydrogen bond, since it keeps a dense molecular packing against the increase in molecular volume, is indispensable in increasing the density of heat paths.

Coil-Globule Collapse of Polystyrene Chains in Tetrahydrofuran-Water Mixtures.

We study the coil and globule states of a single polymer chain in solution by performing molecular dynamics simulations with a united atom model. Specifically, we characterize the structural properties of atactic polystyrene chains with N = 20-150 monomers in tetrahydrofuran-water mixtures at varying mixing ratios. We find that the hydrophobic polymers form rather open coils when the mole fraction of water, XW, is roughly below 0.25, whereas the chains collapse into globules when XW ≳ 0.75. We confirm the theoretically expected scaling laws for the radius of gyration, Rg, in these regimes, i.e., Rg ∝ N3/5 and Rg ∝ N1/3 for good and poor solvent conditions, respectively. For poor solvent conditions with XW = 0.75, we find a sizable fraction of residual tetrahydrofuran trapped inside the collapsed polymer chains with an excess amount located at the globule surface, acting as a protective layer between the hydrophobic polystyrene and the surrounding water-rich mixture. These findings have important implications for nanoparticle fabrication techniques where solvent exchange is exploited to drive polymer aggregation, since residual solvent can significantly influence the physical properties of the precipitated nanoparticles.