A multiscale modeling approach is used to develop an implicit-solvent intermediate resolution model, LIME (Lipid Intermediate Resolution Model), to simulate the behavior of phospholipids in solution. The LIME geometric and energetic parameters are obtained by collecting data from atomistic simulations of a system composed of 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) molecules and explicit water and then grouping the atoms on DPPC into 14 coarse-grained sites with unique properties. The interactions between coarse-grained sites are represented by hard sphere and square well potentials, as opposed to Lennard Jones potentials, thereby allowing us to use discontinuous molecular dynamics (DMD) , a fast alternative to traditional molecular dynamics. DMD simulations on systems containing 256 DPPC molecules show spontaneous formation of a bilayer from a random solution in less than 4 hours. The structural properties of the DPPC bilayer are accurately reproduced including the area per lipid, bilayer thickness, bond order and mass density profiles. DMD/LIME simulations are also applied to aqueous bilayers composed of two types of phospholipids: those containing phosph-L-serine (PS) head groups and those containing phosphatidylcholine (PC) head groups. Our simulations predict that this system phase-separates in agreement with the results of our collaborator Stavorula Sofou ( Rutgers University) on a similar system, DPPC and 1,2-distearoyl-sn-glycero-3-phospho-L-serine (DSPS).

Gold nanoparticles (AuNPs) have recently emerged as a flexible nanoscale platform for applications in drug delivery, biosensing, and bioimaging. The surface properties of AuNPs can be tuned for desired applications by grafting a protecting ligand monolayer to the AuNP surface. A typical strategy for biological applications is to graft alkanethiol ligands to the surface with some or all of the ligands end-functionalized with hydrophilic end groups to confer aqueous solubility. Recently, gold nanoparticles (AuNPs) protected by a binary mixture of purely hydrophobic and anionic end-functionalized linear alkanethiols were observed to spontaneously penetrate cell membranes via a non-disruptive, non-endocytotic mechanism (1). The propensity for penetration was shown to depend on both the composition and morphology of the mixed ligand monolayer. The observation of non-disruptive penetration is unusual given that the monolayers are highly anionic, unlike cationic cell-penetrating peptides or cationic AuNPs that are known to penetrate cell membranes driven by electrostatic interactions. To the best of our knowledge, no mechanism has yet emerged to explain membrane penetration by anionic, monolayer-protected AuNPs.
In this work, we use a combination of both implicit and explicit simulation models, supported by experiments, to demonstrate that these same AuNPs fuse with single component lipid bilayers. The fusion process leaves the AuNPs embedded within the hydrophobic core of the bilayer, resembling transmembrane proteins. To avoid strong energetic penalties associated with the exposure of charged end groups to the hydrophobic bilayer core, the charged ligands “snorkel” to allow charges to access nearby aqueous interfaces while the hydrophobic alkane backbones of the ligands remain in favorable positions within the bilayer. Particle insertion is thus driven by the hydrophobic effect rather than electrostatic interactions. Free energy calculations indicate that fusion depends on the size of the AuNPs, with a preferred particle size and cutoff size for favorable insertion that are both a function of the particle composition. Atomistic molecular dynamics simulations provide further insight into the kinetics of particle insertion, showing the pathway by which insertion proceeds as a function of particle composition. Finally, experiments show a size dependence for the penetration of AuNPs into cell membranes that is similar to the size dependence predicted for bilayer fusion, implying that fusion may be a critical step for cell penetration. We believe this work improves our understanding of AuNP-bilayer interactions and will enable the design of nanoparticles for drug delivery and biosensing applications based on the relationships between AuNP composition and function outlined here.
(1) A. Verma et al, “Surface-structure-regulated cell-membrane penetration by monolayer-protected nanoparticles,” Nature Materials 7 (2008).

The entropic force between two fluctuating fluid membranes is critically reexamined both analytically and through systematic Monte Carlo simulations. A recent work by Freund [1] has questioned the validity of a well-accepted result derived by Helfrich. We find that, in agreement with Freund, for small inter-membrane separations d, the entropic pressure scales as 1/d, in contrast to Helfrich's result: 1/d^3. For intermediate distances, our calculations agree with that of Helfrich and finally, for large inter membrane separations, we observe an exponentially decaying behavior.
[1] L. B. Freund. PNAS, 110(6): 2047, 2013.

Using computer simulations, we probe how to design a synthetic vesicle that can act as an artificial phagocyte by selectively capturing and encapsulating solutes from the surrounding solvent. Our synthetic phagocyte consists of two components: a stimuli-sensitive microgel particle and a lipid membrane enclosing the microgel. Under the action of an appropriate stimulus, microgel swells and perforates the membrane enabling transport of solutes into the vesicle interior. When the target solutes bind to the microgel surface and the stimulus is removed, microgel deswells, which in turn restores the membrane integrity. As a result, captured solutes are isolated inside the vesicle. In our simulations performed using dissipative particle dynamics, we probe the operation of this artificial phagocyte and examine the conditions leading to the membrane pore formation and closing. Our results will be useful for developing new methods for drug delivery and in-vivo sampling.

Our objective is to design effective controlled-release vehicles for lab-on-chip devices for applications in the areas of bionanotherapeutics, sensing and catalysis. In this presentation we present our results on the role of molecular architecture on the morphology and mechanical properties of bio-inspired soft materials. We focus on four lipid architectures with variations in the head group shape and the hydrocarbon tail length. Each lipid species is composed of a hydrophilic head group and two hydrophobic tails. We characterize the soft material via the thickness, interfacial tension, molecular diffusion, area compression modulus, bending modulus, and total energy of the lipid bilayer. We demonstrate the equilibrium morphology of these bilayer membranes to be determined by factors such as the average area of the lipid, lipid head group orientation, and lipid tail length (Koufos et al., 2013.) We use a Molecular Dynamics-based mesoscopic simulation technique called Dissipative Particle Dynamics that captures the molecular details of the components through soft-sphere coarse-grained models and reproduces the hydrodynamic behavior of the system over extended time scales. The DPD method is highly effective in resolving the dynamics of complex fluids over extended spatiotemporal scales due to the soft-repulsive interaction forces.

We generate bilayers in a large variety of geometries, which resemble unusual cellular shell shapes, by mixing cationic and anionic amphiphiles [1]. We use pH to control the degree of ionization of the amphiphiles and hence their intermolecular electrostatic interactions. At low and high pH, we observe closed faceted vesicles with 2D hexagonal molecular arrangements, while at intermediate pH we observe ribbons with rectangular-C packing of the amphiphiles. That is, pH acts as a switch to control the morphology of the ionic bilayers via transitions in the crystalline lattice. Atomistic simulations explain the observed bilayer interdigitation when the charge fraction of both, cationic and anionic head groups is large, and reveal hexagonally packed tails forming bilayers at low pH, when the fraction of charged cationic head groups is low. Mesoscale geometry coarse-grained simulations at low pH unveil faceted vesicles composed of crystalline facetted faces and liquid-like boundaries. These simulations explain that the observed low symmetry polyhedral vesicle result from the possibility of adjusting the local density of hexagonally packed tails at the polyhedral edges to allow for sharp bends.
[1] Leung et al., ACS Nano, 2012, 6 (12), pp 10901-10909.

We develop a computational model to design and characterize multi-component nanostructured soft biomaterials using the Molecular Dynamics simulation technique. Our objective is to generate a stable vesicle through the self-assembly of amphiphilic lipid molecules using implicit solvent coarse-grained molecular models. The amphiphilic lipid molecules are composed of a hydrophilic head group and two hydrophobic tails. We investigate the dependence of mechanical properties as well as morphological stability of the lipid vesicles on temperature. We show that these lipid vesicles are morphologically stable upon changes in the shape and mechanical properties due to temperature. We have extended the model to investigate two component lipid mixtures to enable the formation of hybrid lipid vesicles through the self-assembly. We find that the degree of phase segregation between the two lipid species is tunable via their distinct effective chemical specificity and molecular architecture. We characterize phase segregation process by analyzing the distribution of the pure species domain populations. We furthermore illustrate that the degree of phase segregation has effects on the structure and mechanical properties of the lipid vesicle. Our findings can be used to design new soft biomaterials for various applications in medicine, sensing and energy.

Remora fish have a unique dorsal pad capable of fast, reversible adhesion to a large range of natural and artificial surfaces. The effectiveness of adhesion is due in part to the pad&’s ability to dynamically conform and adapt to the geometry of its host. Simulations based on measured material properties and geometry can provide useful design metrics for biologically inspired design, and furthermore, serve as platform for virtual experiments. The pad itself consists of a lamellar, composite structure composed of mineralized and soft tissue. In this work, finite element models based on microCT scans and measured viscoelastic material properties elucidate the pad&’s complex moduli frequency spectrum and response to different loading configurations.

We combine the string method with self-consistent field theory to compute the most probable transition pathway, i.e. the minimum free energy path, for the insertion of Janus and protein-like nanoparticles into a polymer membrane bilayer. The method makes no assumptions in the reaction coordinate and overcomes the long timescales challenge associated with simulating rare events. Our study suggests that one approach to building functional polymer-nanoparticle composite membranes with oriented nanoparticles is through electrostatic interactions. In particular, hydrophobic Janus nanoparticles with an asymmetric charge distribution can be made to directionally insert into charged membranes. This process is kinetically driven, and involves overcoming a thermally-surmountable activation barrier, which requires favorable interactions between the nanoparticle and the hydrophilic block of the membrane. In contrast, the insertion of protein-like nanoparticles with alternating hydrophilic- hydrophobic-hydrophilic domains into polymer membranes does not occur as a thermally activated event.
Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy&’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

Droplet spreading is a fundamental physical process that is subject to intensive scientific research and of interest in many technical applications. Particularly interesting phenomena in this context are the spreading of droplets of pure trisiloxane surfactants and the spreading of droplets of their aqueous solution. Pure droplets of trisiloxane surfactants form precursor films with complex shapes that depend on the critical surface tension of the substrate: small variations can lead to double layers, terraced layers, or sandpiles. The spreading velocity of these droplets is controlled by the humidity of the surrounding air. Droplets of aqueous surfactant solutions show the unique and fascinating behavior of superspreading, the greatly enhanced spreading of droplets of aqueous solution on hydrophobic substrates. Despite extensive experimental research, the underlying physics of spreading and superspreading processes are still not well understood.
We perform large-scale molecular dynamics simulations with advanced algorithms and optimized force fields to study the spreading of droplets of pure trisiloxane surfactants and of aqueous solutions on amorphous polymer substrates. The simulation approach, and especially the ability to resolve molecular-scale features, provides insights that cannot be obtained from experiments or other simulation methods. We identify the molecular mechanisms of superspreading, the formation of precursor films, and the influence of humidity on spreading velocities. We address open questions from the scientific literature and relate our findings to theories of precursor formation, spreading, and superspreading.

In conventional theoretical or experimental approaches, DNA-grafted colloids are allowed to self-assemble and then the resulting structures examined. While this empirical approach is useful for a-posteriori understanding the factors governing assembly, it does not allow one to a-priori design DNA-grafted colloids that will assemble into desired structures. Here we present an evolutionary optimization framework for designing the building blocks that will self-assemble into desired crystal structures. The proposed ``design" methodology is validated using known experimental results, promising a rational materials design paradigm with broad potential applications.

The capture of solar radiation for energy production and storage is one of the grand challenges of the next decades. Among the many photo-physical processes involved in the developed solar energy conversion devices, it is the first one, namely the light harvesting (LH) step, which has probably received less attention from the scientific community. Nonetheless, light harvesting represents a key step in natural photosynthetic systems: natural LH proteins allow tuning the absorption spectra of the system and an efficient energy funneling towards the reaction centers, while keeping up an effective photo-protection mechanism. To exploit sunlight efficiently we could integrate these antenna systems into photonic devices, e.g. by combining LH proteins with plasmonic metal nanoparticles (MNPs).
We present a newly developed QM/MM/continuum approach which is designed to model all the different electronic processes (light absorption, inter-pigment and pigment-MNP excitation energy transfers, and emission) quantum-mechanically, always including mutual polarization effects among all the QM (the pigments) and the classical parts (the protein and the MNPs).
The QM description of the pigments is of paramount importance, since it allows us to capture excitonic effects, which are behind the most peculiar features of biological systems: any form of semi-empirical or classical description of pigments, e.g. in terms of point-like dipoles, while working properly for simpler artificial devices, would inevitably miss most of the important physics of bio-hybrid devices.
Applications of the model to the case study of the PCP light harvesting protein on a silver islands film are presented. Our multiscale simulations not only reproduce and interpret the experimental findings but they also allow for suggesting general rules to design novel bio-hybrid devices.

Understanding the recognition and interactions at the interface between titania and peptides is crucial for a wide range of applications ranging from the development of biocompatible materials for medical implants to the design and fabrication of nanomaterials for biotechnological and nanotechnological applications [1-4]. Despite the fact that considerable effort has been directed toward a general understanding of peptide recognition and binding to inorganic materials, a deep understanding of the mechanistic features of recognition and binding at the molecular or atomistic level is far from being accomplished [5]. Although several molecular dynamics (MD) simulation studies of peptide adsorption at the aqueous titania interface have been reported [3,4], efficient conformational sampling of adsorbed peptides at the interface is a great challenge. Advanced sampling techniques are pivotal to the understanding, advancement and optimization of interfacial properties. A recently developed version of the promising Replica Exchange with Solute Tempering (REST2)[6] method has shown great improvement in sampling efficiency of peptides adsorbed at aqueous interfaces, at a fraction of the computational cost [7]. Using this technique, for the first time, we have studied the binding and interaction of two titania-binding dodecameric peptides at the aqueous titania (110) surface. Metadynamics simulations have been used to calculate the free energy of adsorption of the two peptides to the titania surface. The results reveal critical insights into the role of peptide sequence and conformation in the binding to titania surface.
[1] Carravetta V. and Monti S., J. Phys. Chem. B, 2006, 110, 6160-6169.
[2] Evans J. S. et al., MRS Bull., 2008, 33, 514-518.
[3] Schneider J. and Ciacchi L. C., J. Am. Chem. Soc., 2012, 134, 2407-2413.
[4] Skelton A. A. et al., ACS Appl. Mater. Interfaces, 2009, 1(7), 1482-1491.
[5] Puddu V. and Perry C. C., ACS NANO, 2012, 6(7), 6356-6363.
[6] Terakawa T. et al., J. Comput. Chem., 2011, 32, 1228-1234.
[7] Wright L. B. and Walsh T. R., Phys. Chem. Chem. Phys., 2013, 15, 4715-4726.

The chemical bonding of dithiolate molecule on the Au surface has been widely studied for applications in molecular electronics. We perform classical molecular simulations by combining grand canonical Monte Carlo (GCMC) sampling with molecular dynamics (MD) simulation to explore the dynamic gold nanojunctions in a benzenedithiol (BDT) or a Alkenedithiol (ADT) solvent. A model system has been built according to the experimental setup. The system involves the long-range elasticity of the gold electrodes in BDT or ADT solvent. The total number of gold atoms in the system is 832. The two layers at the far ends of the gold tips are kept rigid, and the atoms between them are dynamic. The upper gold tip is connected to a driving support through a spring and follows a driven dynamics motion due to the pulling of the driving support. Combining Grand Canonical Monte Carlo (GCMC) and Molecular Dynamics (MD) Simulations In order to reflect the experimental environment, in our simulation we propose a method of combining GCMC and MD techniques alternatively to keep the BDT or ADT density at a bulk value during the pulling process. Temperature is controlled at 298K. The equilibrium absorption configuration obtained from GCMC run then is taken as the input for the MD simulations. After the 0.3 nano meter retraction MD simulation run is finished, MD is switched back to GCMC for a new equilibrium adsorption run based on the configuration of the system from MD. These two procedures proceed alternatively until the breakage of gold single atomic contact begins to form. According to the experiment, we remove all the nonbonded BDT or ADT molecules and stretching without the insertion and deletion of new BDT or ADT molecules. We observe the formation mechanism of single Au-BDT-Au junction.

A large number of practical processes in nanotechnology involve the interactions of metals and peptides(1). However, the current understanding of the atomistic behavior of the aqueous peptide-metal interface is far from complete. In this work, we present atomistic simulations of the aqueous peptide-gold interface. The recently developed GolP-CHARMM force field(2) is employed to model the peptide-metal interactions due to its ability to capture the dynamic polarization of gold atoms, chemisorbing and the hybridized carbon-gold atoms. These features result in an accurate representation of the peptide binding in gold. Intensive Replica Exchange Solute Tempering(3) (REST) simulations were performed for a set of 12 different peptides(4). From the simulations, clustering analysis, contact motifs, Ramachandran plots, and the most probable conformations are obtained. These results expand the current understanding of the single residue binding and show the relevance of the contact residues and the role of entropy in peptide binding. The simulations reveal that the total binding strength is more than simply the sum of multiple anchoring residues, and that the position of the anchoring residues has a significant impact on the conformational entropy. These findings provide guidelines in the development of new peptide sequences that efficiently bind to gold interfaces.
(1) Sarikaya, M.; Tamerler, C.; Jen, A.K.Y.; Schulten, K.; Baneyx, F., Nat. Mater. 2003, 2, 577.
(2) Wright, L. B.; Rodger, P. M.; Corni, S.; Walsh, T. R., J. Chem. Theory Comput. 2013, 9, 1616.
(3) Terakawa, T.; Kameda, T.; Takada, S., J. Comput. Chem. 2011, 32, 1228.
(4) Tang, Z.; Palafox-Hernandez, J. P.; Law, WC.; Swihart, M. T.; Prasad, P. N.; Knecht, M. R.; Walsh, T. R., ACS Nano, under review.

The synthesis of biologically-inspired macromolecules with well-defined sequence, dispersity and assembly function has large potential for applications ranging from delivery vehicles of medical therapeutics, sensing applications, to scaffolds for tissue regeneration. Here we report the progress of our work in exploiting the ability of a specific DNA polymerase, terminal deoxynucleotidyl transferase (TdT), to polymerize long chains of single strand DNA (ssDNA) and to incorporate unnatural nucleotides with useful functional groups into the growing polynucleotide chain. Specifically we report on a method that allows the facile synthesis of amphiphilic ssDNA biomacromolecules (0.6 to 2.7 kilobases) via enzymatic polymerization of mononucleotides. The amphiphilicity arises from the synthesis of hydrophilic homopolynucleotides and the modification of hydrophobic, unnatural nucleotides into the extended ssDNA chain. We synthesize these ssDNA polynucleotides by a two-step enzymatic polymerization reaction in which the MW and composition are controlled by the ratio of monomer to initiator concentration and the type of nucleotides, respectively. The hydrophobicity, resulting from incorporation of unnatural nucleotides, can be tuned to drive the assembly of these polynucleotides into star-like, hairy micelles. We used high resolution AFM and cryo-TEM imaging to directly visualize the star-like micelle morphologies with a condensed core and hairy corona. In addition, we investigated the hydrodynamic radius (Rh) of homopolynucleotides and polynucleotides amphiphiles that form micelles by dynamic light scattering (DLS). Finally we report on simulations in which a coarse-graining computational method was used to represent the amphiphilic polynucleotides, and we show that they form star-like micelles with relatively condensed core and hairy corona.

Cyclic peptide nanotubes (CPNs) are self-assembled supramolecular systems with outstanding mechanical properties (Ruiz et al. Nanotechnology, 2013). CPNs are excellent candidates as biomimetic synthetic nanopores that could serve as artificial ion channels because they offer precise control of the pore diameter and the nanotube outer surface chemistry, as well as interior polarity through nonstandard mutations presented in the amino acid sequence (Hourani et al. JACS 2011).
One successful strategy to generate CPN based thin nanoporous membranes for applications such as carbon capture or low energy water desalination is through co-assembly of polymer conjugated CPs with block copolymers (Xu et al. ACS Nano 2013). Experimental results suggest that the conjugated polymer helps to prevent lateral aggregation of CPNs and increase processability in solution and in polymer thin films. However, our simulations validated by experiments also show that polymer conjugation introduces a free energy penalty to the binding energy of the CPs, that if not well-tuned, can hamper the nanotube formation. Building on this finding, here we investigate whether the free energy penalty associated with the polymer conjugation can be leveraged to direct the self-assembly of functionalized CPs towards a defined stacking sequence. To illustrate this point, we present a theoretical framework and molecular dynamics simulations to explain the self-assembly of conjugated-CPs and show how the conjugation degree can indeed be used to control the self-assembly stacking sequence. We use coarse-grained molecular MD and a single energy barrier kinetic model to quantify the free energy penalty induced by the polymer conjugates. Replica exchange simulations are implemented to enhance the sampling of the energy space of a coarse-grain model of the conjugated-CPs, thereby reproducing the stacking sequences that corresponds to thermodynamic and kinetic limits. Our results provide a theoretical framework for precisely controlling the self-assembled stacking sequence of CPs with different polarity, paving the way for generating artificial transmembrane peptide assemblies with tunable interiors and exteriors.

Ionomer melts have unique mechanical and electrical properties due to the formation of nanoscale ionic aggregates in the polymer matrix. However, the relationships between ionomer chemistry, morphology, and ion transport are poorly understood. To elucidate the ionic aggregate morphology, we have performed molecular dynamics simulations of a series of polyethylene-based model ionomer melts, in which the spacing between functional groups is precisely controlled. We vary the polymer-bound ion chemistry (methylimidazolium, sulfonate and carboxylate), the counterion type, the neutralization level, and the length of the spacer. The simulations provide new insights into the shape, size and composition of ionic aggregates. In particular, we observe a wide variety of aggregate morphologies, ranging from small spherical aggregates to string-like shapes and large percolated networks. The structure factors calculated from simulation agree well with available X-ray scattering data. Depending on the morphology, the simulation and experimental scattering curves can be well fit with a modified hard sphere or modified hard cylinder model. These fits, combined with the simulation data, provide the first (indirect) experimental evidence of string-like aggregate morphologies in ionomer melts.
This work was performed, in part, at the Center for Integrated Nanotechnologies, an Office of Science User Facility operated for the U.S. Department of Energy (DOE) Office of Science. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy&’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

The design and manipulation of mesoscale assemblies is a promising area of emerging research, which requires the ability to understand and harness the van der Waals-London dispersion (vdW-Ld) interactions that arise in these systems. This is achieved by measuring the interband optical properties of materials, computing their vdW-Ld interaction potentials, and establishing an open-source software platform for the calculation and distribution of these results.
Previous work enabled us to understand vdW-Ld interactions between objects with simple geometries, such as solid substrates, rods and spheres, as well as more complex anisotropic systems such as carbon nanotubes. This understanding laid a foundation for our current research on a range of biological and biologically relevant materials, including collagen, AlPO₄, and SiO₂. We have focused on AlPO₄ because of the strong electron localization in the phosphate complex ion, which provides insights into the role of phosphates in the electronic structure of biological systems; SiO₂, which has a similar electronic structure, is presented for comparison.
The interband optical properties of AlPO₄ and SiO₂ have been derived from Vacuum Ultraviolet (VUV) spectroscopy and spectroscopic ellipsometry. Ab initio calculations, based on the local density approximation of density functional theory, and further analyses, based on Lifshitz theory, give us insights into the electronic structures, interband transitions and vdW-Ld interaction potentials for SiO₂, AlPO₄, and collagen. These results, along with the new interaction geometries, have been incorporated into our open-source Gecko Hamaker software.
Further studies will apply this same methodology to bovine serum albumin and DNA oligonucleotides, including the determination of interband optical properties through VUV spectroscopy and ab initio methods, as well as direct determination of interaction strengths via static light scattering measurements of biomolecules in solution. These results will facilitate the mesoscale material design.

DNA template can trigger the self-assembly of metal or semiconductor nanoparticles into programmable molecular architectures. The performance of DNA-nanoparticle system strongly depends on the size and ligand chemistry of nanoparticles. We performed molecular dynamics simulations to investigate the effect of colloidal gold nanoparticle (GNP) ligands charge and polarity on the ability to bind DNA molecules. We tailored the surface of GNP by introducing different terminal functionality to thiolated ligands, such as hydrophobic, polar and charged groups. We found that uncharged GNPs and GNPs with cationic ligand charge density of less than 10% can only bind to the minor groove of DNA. Whereas GNPs with ligands charge density of higher than 10% can bind to major or minor groove. Binding to major groove result in significant distortion and wrapping of DNA around the GNP. The distortions of the DNA helical structure strongly depends on the ligand charge density. Also at higher nanoparticle concentration and low charge densities, the ligand hydrophobicity can disrupt the hydrogen bonding between base pairs of DNA strands and leads to unwinding of DNA helix. We observed that by tuning the cationic charge density and concentration of GNP we can control the binding modes and DNA structural modifications.

Via computational modeling, we investigate the mechanism of strain recovery in dual cross-linked polymer grafted nanoparticle networks. The individual nanoparticles are composed of a rigid spherical core and a corona of grafted polymers that encompass reactive end groups. With the overlap of the coronas on adjacent particles, the reactive end groups form permanent or labile bonds, and thus form a “dual cross-linked” network. We consider the strain recovery of the material after it is allowed to relax from the application of a tensile force. Notably, the existing labile bonds can break and new bonds can form in the course of deformation. Hence, a damaged material could be “rejuvenated” both in terms of the recovery of strain and the number of bonds, if the relaxation occurs over a sufficiently long time. We show that this rejuvenation depends on the fraction of permanent bonds, strength of labile bonds, and maximal strain. Specifically, we show that while an increase in the labile bond energy leads to formation of a tough material, it also leads to delayed strain recovery. Further, we show that an increase in the fraction of permanent bonds not only enables faster recovery but also yields improved recovery even after multiple stretch-relaxation cycles.

Recently, hydrogel have found to be promising biomaterials since their porous structure and hydrophilicity enables them to absorb a large amount of water. In this study the role of water on the mechanical properties of hydrogel are studied using ab-initio molecular dynamics (MD) and coarse-grained simulations. Condensed-Phased Optimized Molecular Potential (COMPASS) and MARTINI force fields are used in the all-atom atomistic models and coarse-grained simulations, respectively. The crosslinking process is modeled using a novel approach by cyclic NPT and NVT simulations starting from a high temperature, cooling down to a lower temperature to model the curing process. Radial distribution functions for different water contents (20%, 40%, 60% and 80%) have shown the crosslinks atoms are more hydrophilic than the other atoms. Diffusion coefficients are quantified in different water contents and the effect of crosslinking density on the water diffusion is studied. Elasticity parameters are computed by constant strain energy minimization in mechanical deformation simulations. It is shown that an increase in the water content results in a decrease in the elastic moduli. Finally, using the knowledge of nanomechanical behavior, continuum hyper elastic model of contact lens were used for three different loading scenarios.

Using molecular dynamics simulation, we investigated constant strain rate deformation behavior of a model hydrogel consisting of a single polymer network in which the ratio between the number of physical and chemical cross-links was varied. The physical cross-links were modeled as clusters of acrylic acid (AA) groups connected by hydrophilic flexible chains. Chemical cross-links were represented as permanent bonds between selected segments of neighboring chains and were added after the physical cross-links were formed. The structural parameter used was the number of chemical cross-links related to the number of physical clusters (cpc). Slightly cross-linked networks up to cpc=0.7 exhibitted no change in their mechanical response compared to the pure physically cross-linked gel. A moderately chemically cross-linked network gel with cpc = 1.58 exhibitted deformation behavior similar to a double network hydrogel. The strongly crosslinked networks with cpc > 3.02 behaved as purely chemically cross-linked network gel over the entire deformation range. The strongest effect of chemical cross-links on the non-linear deformation response of the hydrogels, was obtained for physical and chemical cross-links set as far apart as possible. The addition of 64 crosslinks to the simulation box composed of 1600 non-solvent atomic groups, switched the character of the deformation response from that for a physical to that for a chemically cross-linked network gel.
The molecular mechanism of the hybrid cross-linked hydrogel deformation consisted of (i) hopping of AA groups between neighbouring clusters, (ii) cluster reorientation and (iii) combination of the two. The AA groups hops exhibited relaxation times of 10-5 - 10-4 ns and the reorientation was slower ranging from 10-3.5 to 10-3 ns. The relaxation times for the third process resulting in a significant remodeling of the original network structure, comprising of both the hops and reorientation, ranged from 1 ns to the maximum simulation time of 100 ns. The redistribution of the force between chemical cross-links was more efficient than that between purely physical cross-links. .
In conclusion, we believe that the analysis of our model hydrogel could be instructive for the preparation of tough single chemistry hydrogels. The model can provide leads for designing extent and spatial arrangement of chemical cross-links of reversible bond hydrogels with prescribed deformation behavior similar to that obtained for double network hydrogels.
Acknowledgements
J. Zidek, and J. Jancar acknowledge European Regional Development Fund (CEITEC, CZ.1.05/1.1.00/02.0068), A. M. acknowledges support by CECAM at MPIP during his stay at the Institute.

A new class of porous polyarylamide polymer films that contain s-dibenzocyclooctadiene (DBCOD) moieties can generate unconventional thermal contraction that is completely reversible under low-energy stimulation. They possess an extremely large negative thermal expansion coefficient (NTE) of approximately -1200 ppm/K under ambient conditions, much higher than the best previously known NTE materials at these operating conditions. Significantly, the NTE value remains unchanged after extended cycling. Density-functional theory calculations revealed that the spectroscopic change with temperature and macroscopic thermal contraction is due to the conformational changes of the DBCOD moiety, from the thermodynamic global minimum (twist-boat) to a local minimum (chair). This contention is further supported by the calorimetry analysis indicating that the conformational change is the origin of this abnormal thermal shrinkage. The observed giant and stable NTE that can be triggered by low and penetrating energy sources have the potential to generate high mechanical energy output otherwise unattainable. This newly identified thermally agile molecular subunit opens a new pathway for creating NIR-based macromolecular switches and motors, as well as enabling ambient thermal and photothermal energy storage and conversion.

Liquid crystal elastomers (LCE) are functional soft materials consisting of weakly crosslinked polymer networks with embedded liquid crystalline (mesogenic) molecules. Consequently, LCE are characterized by a pronounced coupling between macroscopic strain and orientational mesogenic order. As the latter can be controlled by external stimuli such as temperature, electric field, or ultraviolet light, LCE have great potential for application as sensors and actuators [1]. Here we present large-scale molecular simulations of swollen main-chain LCE [2,3]. The simulated experiments include temperature scans, stress-strain runs, and the application of an external electric field. Our isostress Monte Carlo simulations are capable of reproducing isotropic, nematic, and smectic phases, as well as a stress-induced isotropic-to-nematic transition [2]. Moreover, a transversal electric field is seen to induce nematic director rotation resulting in orientational stripe domains [3]. We use our simulation output to connect to typical experimental observables, such as sample dimensions, specific heat, deuterium magnetic resonance spectra, and scattered X-ray patterns.
[1] M. Warner and E. M. Terentjev, Liquid Crystal Elastomers, Oxford University Press, Oxford, 2003.
[2] G. Ska#269;ej and C. Zannoni, Soft Matter 7, 9983 (2011).
[3] G. Ska#269;ej and C. Zannoni, Proc. Natl. Acad. Sci. USA 109, 10193 (2012).

There has been a growing need for synthetic polymers with optimized mechanical responses and energy dissipation characteristics that can be implemented as tissue simulants. Such "soft materials" can serve as test media for examining effects of ballistics on soft-tissues, and enable design of devices and protective garments that minimize blunt impact trauma. The limited mechanical tunability of current tissue simulant gels to replicate the energy dissipation capacity and rates of biological soft-tissues highlights the importance of further investigations. Furthermore, experimental measurements are inefficient and insufficient as stand-alone approaches to match impact responses of new materials to those of biological tissues. Here, we validated a multi-scale finite element model against impact experiments for synthetic polymer gels intended as candidates for mechanical simulants of soft-tissue. Importantly, using a bi-layer system of dissipative and nondissipative materials, we decoupled the two design metrics of impact resistance and energy dissipation capacity. These computational predictions provide promising prospects for tunable design of soft-tissue simulant materials.

Recently developed theoretical approach is presented in this work with the aim to design stimuli-responsive hydrogels and optimize their swelling properties. Among many possible interesting applications of such gels is a mucoadhesive drug delivery. The structure and as a result delivery properties of these gels are modulated by external impulses.
Using molecular theory and molecular dynamics we also demonstrate in this work that the presented approach allows us to apply it on understanding the barrier properties of mucus towards sexually transmitted diseases. Mucus is a gel like substance that consists of water, electrolytes, amino acids and highly glycosylated proteins called mucins. It acts as a sensor towards incoming pathogens, and signals to epithelium in response to sensing such dangers. The mucus layer can naturally modify its barrier properties. For example, cervical mucus properties change regularly during the menstrual cycle, providing variable resistance towards viruses.
Mucus structure varies a lot for different parts of the body. Since there is very little and not enough information about exact structure of mucus we begun from modeling mucus by a structure of stimuli responsive hydrogels such as poly(methacrylic acid-grafted-poly(ethylene glycol)) gels (P(MAA-gEG)) which were previously proposed as a good protein and drug delivers[1].
Here we present results of our recent studies: the variations of binding properties of synthetic stimuli responsive hydrogels as a result of environment changes like pH, salt concentration and volume fraction of the gel.
1. L. Serra, J. Doménech and Nicholas A. Peppas, European Journal of Pharmaceutics and Biopharmaceutics 71 (2009) 519-528.

Nematic Liquid Crystal Elastomers (NLCE) are lightly cross-linked polymeric materials that exhibit rubber elasticity and liquid-crystalline orientional order. We investigated the thermal response and microstructure of several samples of side-chain NLCEs which had varying chemical compositions. Real-time synchrotron wide-angle X-ray scattering (WAXS) during thermal treatment showed that the materials were highly anisotropic with a fiber-like diffraction pattern. The isotropic to nematic transition temperature showed a dependence on the chemical composition. Transition temperatures from WAXS were correlated with thermal properties, using differential scanning calorimetry, and with optical properties, using polarizing optical microscopy and transmission ellipsometry. In the low temperature nematic phase, we investigate how the stress-strain behavior depends on the chemical composition and thermo-mechanical history of the samples.

Inclusion of colloidal particles into a nematogenic material functionalizes defects as carriers of cohesion, self-assembly and manipulation in the resulting composite material. One of the commonly used particle types are spheres, treated to enforce homeotropic surface alignment -- they form structures with just enough degrees of freedom to allow for complex entangled structures, but still follow models that are easy to formulate. Depending on the geometry of the confinement and arrangement of the dispersed spheres, entangled particle clusters can be assembled, ranging from disordered clusters, one-dimensional wires, two-dimensional regular arrays and finally opal-like 3D colloidal crystals.
We focus on structures that enforce disclination lines with a constant director field cross section with a winding number of -1/2. In these systems, the main phenomenological building elements are so-called ``rewiring sites'', where the disclinations are free to choose the way they connect to each other, while away from these sites, the disclination geometry is enforced by the confining surfaces of in the inclusions. The rewiring sites appear in highly symmetric tetrahedral and cubic forms and their enumeration is an elementary question of elementary geometry and symmetry groups. As a result, they can be used to predict the possible structures, use them as templates for numerical modeling, and lead the process of experimental assembly.
We review the rewiring sites from topological and geometric point of view, and demonstrate the numerical models of their appearance in the nematic colloids.

We study propagation of elastic waves in layered media undergoing finite deformation. Applied deformation influences the wave propagation in two ways: first, as a material undergoes deformation, its microstructure evolves; second, the presence of stress can influence wave propagation. By application of Bloch-Floquet technique we have derived an exact analytical solution for elastic wave propagation in flat layers. However, when layered materials are compressed along the layer direction, a buckled shape may developed upon attending a critical load. Further compression beyond the critical strain may lead to the wrinkling of the interfacial layer. Thus, a system of periodic scatterers can be created. These scatterers reflect and interfere with waves. To analyze the wave propagation in the media with wrinkled interfacial layers, a finite element model was developed. By means of the numerical analysis we demonstrate that the topology of wrinkling interfacial layers can be controlled by deformation and used to produce band-gaps and to filter frequencies. Since the microstructure change is reversible, the mechanism can be used for tuning and controlling wave propagation by deformation.

References
[1] Kushwaha, M.S. and Halevi, P. and Dobrzynski, L. and Djafari-Rouhani, B. Acoustic band structure of periodic elastic composites. Phys. Rev. Lett., 71(13):2022-2025, 1993.
[2] Li, Y. and Kaynia, N. and Rudykh, S. and Boyce, M. C. Wrinkling of Interfacial Layers in Stratified Composites. Adv. Eng. Mater., 2013.

Transport of nano-scale particulate materials in polymeric network is the key phenomenon that determines their efficacy as delivery system. Nano-rheological measurements detect the movement of a particle through polymeric systems and providing direct way to probe local polymer dynamics. We study sub-diffusive motion of probe particles and coupling between particle transport properties in polymer solutions at various concentrations analytically and numerically. Applied Diffusing Wave Spectroscopy technique allows characterizing dynamic mobility of particles with the size range from tens nanometers to microns in various natural polymeric systems that represent biological matrices. It was shown that well known Stokes - Einstein model cannot be applied to describe the mobility of particle in semi-diluted polymer solutions and gels because the diffusivity of particle is not inversely proportional to the bulk viscosity. Our results show that there are several distinct regimes of particle mobility corresponding to the various arrangements of polymeric molecules in the solutions and melts. Measured mobility coefficients for the embedded nanoparticles are in good correspondence with De Genes scaling theory. Based on obtained data, we propose simple criteria that allow rational design of the various nanocomposite materials.

Via computer simulations, we investigate the linear and non-linear viscoelastic response of polymer grafted nanoparticle networks subject to oscillatory shear at various amplitudes and frequencies. The individual nanoparticles are composed of a rigid spherical core and a corona of grafted polymers that encompass reactive end groups. With the overlap of the coronas on adjacent particles, the reactive end groups form permanent or labile bonds, and thus form a “dual-crosslinked” network. Notably, the existing labile bonds between particles can break and reform depending on bond rupture rate, extent of deformation and the frequency of oscillation. We study how the viscoelastic behavior of the material depends on the energy of the labile bonds and identify the network characteristics that give rise to the observed viscoelastic response. We observe that with an increase in labile bond energy, the storage modulus increases while the loss modulus shows a more complex response depending on the labile bond energy. Specifically, in the case of the weaker labile bond samples, the loss modulus increases monotonically, while for the stronger labile bond samples, it exhibits a minimum with increase in frequency. We show that an increase in storage modulus corresponds to an enhancement in the average number of bonds in the samples and the loss modulus characteristics depend on the collective rearrangements of the particles. Furthermore, we determine that the effective contribution of the bonds to the storage modulus decreases with increase in strain amplitude. In particular, while bond formation at small amplitude drives an increase in storage modulus, at large amplitudes it promotes clustering and formation of voids leading to strain softening. Our simulations provide a mesoscopic picture of how the nature of labile bonds affects the performance of cross-linked polymer-grafted nanoparticle networks.

Polyimides (PI) are a class of polymers that have excellent mechanical strength and high thermal resistance. Composites of PI with carbon nanotubes (CNTs) as the reinforcement are hence promising multifunctional materials for extreme thermal conditions. An understanding of the molecular level interactions between PI and CNTs is necessary to engineer high performance materials. We performed molecular dynamics (MD) simulations to investigate the interfacial characteristics of the polyimide/CNT systems. As a function of the number of CNT walls, we report molecular details of the pullout process, and predict the critical loading rate required to obtain a complete pullout for each CNT. As a function of varying volume fractions, stress-strain curves of composites with different volume fraction were drawn after 5% strain was applied on both ends of CNTs. The traditional volume fraction equation was modified by regarding the interface as voids, so the volume fraction of CNTs, matrix and interface can be accurately calculated, and we also developed an indirect approach to predict the degree of order of the PI matrix with the addition of the CNT reinforcements. The results indicate that the strength and modulus increase with the volume fraction of CNTs, and the degree of ordering is improved when volume fraction increases, especially at the interface. The results support the assumption mentioned above and are consistent with the experimental observation.

Structural changes in soft matter systems induced by changes of state occur on time and length scales that reach far beyond atomistic scales. Multiscale simulation methods and coarse-grained models designed to bridge these scales have been developed in recent years but the models usually exhibit poor transferability owing to the use of effective pair potentials describing noncovalent interactions between the coarse-grained units [1]. Limited transferability manifests a severe drawback, preventing the use of several of these models in coarse-grained simulations of processes induced by changes in thermodynamic state point.
We have recently developed a free-energy-based coarse graining methodology - coined the Conditional Reversible Work (CRW) method [2] - that obtains transferable pair potentials for soft matter systems. The method has been developed with an eye to minimizing strongly state-point-dependent average multi-body contributions in the effective pair potentials. The CRW method will be discussed and the examples presented will include CRW potentials for aliphatic groups, which can be used as coarse-grained building blocks for macromolecular systems. It will be show that CRW models are chemically transferable to coarse-grained linear alkanes and thermodynamically transferable in the liquid phase branch of the phase diagram (1 atm.) between the melting and boiling points. We further discuss recent multiscale simulations of glassy polystyrene surfaces performed with CRW pair potentials [3].
References:
[1] E. Brini et al. Soft Matter 9, 2108 (2013)
[2] E. Brini, N.F.A. van der Vegt, J. Chem. Phys. 137, 154113 (2012)
[3] V. Marcon, D. Fritz, N.F.A. van der Vegt, Soft Matter 8, 5585 (2012)

Poly(N-isopropylacrylamide) (PNIPAM) is a thermosensitive polymer that is well-known for its lower critical solution temperature (LCST) around 305K. Below the LCST, PNIPAM is soluble in water, and above this temperature, polymer chains collapse and transform into a globule-state. In this study, we have carried out MD simulations of PNIPAM polymer chains consisting of 60 monomer units grafted on a gold nanoparticles, of 6 and 10 nm diameter, to study the effect of curvature and temperature on the polymer conformations. Additionally, We have also studied effect of grafting density on the coil-to-globule transition shown by PNIPAM above the LCST. Studied system consisted of > ~3 million atoms. All the simulations were carried out below and above the LCST of PNIPAM, namely, at 275 and 325 K. Simulation trajectories were analyzed for structural and dynamical properties such as radius of gyration of PNIPAM chains, hydrogen bond life-times, and residence time probability of water molecules.

We discuss recent results on designing robust and compliant nanoscale sheets and shells for prospective delivery vehicles and self-rolled microstructures. Ultrathin films from synthetic polymers and natural proteins were assembled through a layer-by-layer (LbL) assembly and were subjected variable chemical environment. These changes resulted in variable osmotic pressure and interfacial stresses and caused changes in shape and dimensions of compliant microstructures. A mesoscale computational model was developed to examine the properties of LbL constructs and to predict their responsive behavior. The local deformation, buckling behavior, and reconfigurable behavior under variable chemical environment have been studied using a combination of both experimental and simulation approaches.

In this work we study the difference in the macroscopic behavior of composites with hierarchical stochastic microstructures, and those of same filler size and volume fraction, but in which inclusions are randomly distributed in the matrix. To this end, we study numerically the elastic-plastic and damping response of random microstructures, and of microstructures with exponential and power law spatial correlations. We show that as the range of spatial correlations increases, gains are observed in most macroscopic properties. Significant improvements are observed in the damping behavior. These results are important for the mescoscale design of nanocomposites. The implications of the presence of “interphase” zones in polymer nanocomposites with this type of mesoscale structure will be discussed.

Stimuli-responsive polymers are an important class of smart materials, with potential applications in aerospace, civil engineering, and biomedical devices. The past decade has witnessed remarkable advances in the development of active and shape memory polymers. Recent progress has been toward stimuli-responsive polymer-based nanocomposites that allow for enhancing and broadening the properties and applications of shape memory polymers. These nanocomposites represent a promising potential for new generation of smart materials. On the other hand, graphene has attracted much attention due to its two-dimensional structure and desirable physical properties with possible applications in electronics, energy storage and conversion.
In this talk we describe a computational study of graphene/epoxy bi-layer structures designed to engineer strain and shape. We use molecular dynamics simulations to study curing of a thin epoxy (EPON862 cured with DETDA) layer over a graphene ribbon. The volumetric shrinkage during polymerization and mismatch in thermal expansion coefficient induces a significant curvature and strain in the system. The responsive deformation of graphene/epoxy composites due to temperature changes and the temperature-dependent curvature of graphene ribbons with different lengths and different resin thicknesses are characterized. The MD simulations provide insight into graphene/epoxy nanomaterials including ranges of curvature achievable and strains.

We have conducted theory and molecular simulation studies to understand effects of polydispersity on the morphology in polymer nanocomposites with polymer grafted nanoparticle as fillers. In polymer nanocomposites containing monodisperse polymer (brush) grafted particles in a monodisperse polymer matrix with chemically similar graft and matrix polymers, the grafted particles are well dispersed in the polymer matrix only when the grafted polymer chains are much longer than the matrix polymer chains. By introducing polydispersity in the grafted polymers [Ref 1], one can stabilize dispersions of polymer grafted nanoparticles in a polymer matrix even if the average grafted polymer chain length is less than the matrix chain length. This is because the polydispersity in grafted polymers relieves monomer crowding in the grafted polymer layer on the particle, and improves wetting of the grafted layer by the matrix chains. Furthermore, these dispersion-stabilizing effects of polydispersity hold even in the presence of strong (~10kT) particle-particle attraction. By introducing polydispersity in the matrix chains, i.e. by using a matrix consisting of short and long chains, with monodisperse polymer grafted particles, we find that the grafted polymer layer is preferentially wet by the short matrix chains, even when the graft polymer chain length is less than the short matrix chain length. In summary, these theory and simulation results show that polydispersity, either in the grafted polymers or in the matrix polymers, can be a design parameter that stabilizes dispersed morphology in composites consisting of polymer grafted particles in a polymer matrix.
[1] Martin, T. B.; Dodd, P. M.; Jayaraman, A.; Polydispersity for Tuning the Potential of Mean Force between Polymer Grafted Nanoparticles in a Polymer Matrix. Phys. Rev. Lett. 2013, 110 (1), 018301.

Unlike synthetic composites in which one quality is sacrificed over the other, stiffness, strength and toughness are almost perfectly balanced in biomaterials such as nacre, bone and spider silk. The key to this superior mechanical behavior can be attributed to the unique mechanism of shear load transfer between the thin building blocks. This has imparted an increasing interest for synthesizing “brick-and-mortar” structures with analogous mechanical behavior.
Majority of previous studies focus on shear load transfer between tablets with identical properties and dimensions. Herein, we present a analytical model coupled to numerical simulations to study the elastic and plastic shear load transfer between dissimilar tablets with varying dimensions.
Our elastic formulations show that how tuning the elastic moduli and the thickness of either of the tablets and the interfacial shear modulus dramatically changes the axial stress distributions in the tablets as wells as the interfacial shear stress distributions. In addition, the efficiency of the shear load transfer is investigated via identifying the maximum strain energy density of the composite, which translates into an overlap length. While for certain ratios of elastic moduli and thicknesses of the tablets, the elastic strain energy can be maximized, we show that for specific combination of tablets there is no optimum strain energy density and overlap length. In such composites, the elastic strain energy density will decrease monotonically with increasing the overlap length. Finally, we extend these analyses to the plastic regime where we explore the interplay between the different tablet strengths and maximum interfacial shear strength. Overall, our analyses provide guiding hypothesis for designing platelet-matrix composite structures with predetermined thicknesses and material properties to achieve stronger and more durable characteristics.

We assess the accuracy and efficiency of mesoscopic simulation methods, namely Brownian Dynamics (BD), Stochastic Rotation Dynamics (SRD) and Dissipative Particle Dynamics (DPD), for flows of complex fluids, such as polymer solutions, through microfluidic geometries. Since both SRD and DPD use soft or weakly interacting solvent “beads” to carry momentum, both methods contain unwanted particle inertial effects, and high fluid compressibility. To assess these effects, we compare the speed and accuracy of predictions of SRD and DPD for computing flow in both a straight channel with periodic slip boundary conditions, representing a periodic electroosmotic flow, and flow through periodic contractions. We find that SRD is roughly 10 to 30 times faster than DPD in predicting the flow field, with similar accuracy at low Reynolds number. However, SRD has more severe problems with compressibility effects than does DPD, which limits the Reynolds numbers attainable in SRD at reasonable computational cost to Re < 100. In a periodic channel, we find that there is an elasto-hydrodynamic drift mechanism that allows polymers to be size segregated based on molecular weight, and that this mechanism can be most efficiently simulated by BD simulations. The results help determine the range of conditions for which BD, SRD or DPD is preferable for mesoscopic simulations of flows in microfluidic geometries of complex fluids.

The morphology of blends of conjugated polymers (electron donors) and fullerene derivatives (electron acceptors) strongly affects the charge transport, charge separation and the overall efficiency of organic photovoltaic devices. In this talk we will present high-throughput, coarse-grained molecular simulation studies to understand how molecular-level features such as alkyl side chain length, alkyl side chain spacing along thiophene polymer backbone and fullerene functionalization (and in turn miscibility with the conjugated polymer) affect the blend morphology. Our coarse-grained models reproduce neat polymer (without acceptors) morphologies observed in experiments, such as lamellae and hexagonally packed cylinders. Furthermore, for blends of conjugated polymers and fullerene derivatives, this work shows how conjugated polymer architecture and acceptor miscibility can be tuned to obtain new blend morphologies with features that are known to enhance efficiency of organic solar cells.

We report the results of a computational study of the effect of introducing a small amount of strong polymer chains into networks of cross-linked polymer-grafted nanoparticles (PGNs) swollen in water. We take advantage of the hierarchical structure of the PGNs by considering the essential features at each level of the multi-scale problem. The model consists of rigid cores with coronas of grafted polymer chains, which overlap and allow reactive groups on chain ends to form dual cross-links composed of weaker labile bonds, which can reform after breakage, and stronger bonds that rupture irreversibly. We show that the network can be reinforced by the introduction of very strong chains; the added strong chains are treated as unbreakable for simplicity. The network is subject to extension at constant strain-rate to measure its strain at break and toughness. We show that the introduction of unbreakable chains produces a distinctive mechanical response. Specifically, we demonstrate a greater frequency of samples with larger strain at break and toughness for networks that encompass a fraction of the unbreakable chains. Furthermore, we reveal that samples exhibit two contrasting modes of failure. One in which ruptures occur in areas devoid of unbreakable chains at small strains, or samples that form long threads of unbreakable chains enabling a larger strain at break. The findings provide guidelines for creating PGN networks that show remarkable strength and ductility.

An expanding area of green technology is in the wastewater treatment for heavy metal ions. Adsorption of the fluid cations to solids has proven to be successful and recent research has investigated the improvement of this method through novel materials such as grafting dendrimers onto a solid support. We develop a reduced representation of the system via the flow of a fluid composed of solvent and metal cations through a channel of parallel surfaces grafted with dendrimers. The system is modeled after polyamidoamine (PAMAM) dendrimers with functional groups that promote the adsorption of Pb (II) ions. Coarse-grained molecular dynamics simulations using the LAMMPS package are utilized to predict the nanoscale behavior of the system. The goal is to observe and characterize the adsorption of the Pb (II) ions to the branched PAMAM dendrimers through the measurement of fluid properties and dendrimer dynamics. Our project entails investigating the effects of electrostatic interactions, solvent quality, and surface architecture. These simulations can identify the role of molecular scale properties on the adsorption process, thereby allowing predictions for bulk analysis including the continuum phase. These investigations have the ability to be applied to other chemical processes that make use of polymers grafted onto a support.

The addition of nanoparticles (nanospheres, nanotubes, nanoplatelets) with dimensions on the nanometer scale to a polymer can result in materials with significantly improved properties. Even though polymer nanocomposites have been extensively investigated, many basic questions regarding the molecular origin of their properties are not solved. Understanding the polymer dynamics is particularly important when designing fabrication methods for nanocomposites. Recently, polymer tracer in PS/ silica nanocomposites [1] and PMMA/silica nanocomposites [2] has been measured. The polymer diffusion is observed to be significantly slower than that predicted by the Maxwell model.
The structure, dynamics and primitive path network of polymer nanocomposites containing spherical nanoparticles is investigated by means of molecular simulations for both unentangled and entangled polymers. In particular, we performed molecular dynamics simulations on a bead-spring model of polymer melts and spherical nanoparticles in order to identify the influence of nanoparticle size, loading, and polymer-nanoparticle interactions on the overall polymer configuration, polymer dynamics and free volume, below and above the percolation threshold. We found that the overall polymer configuration, as characterized by the radius of gyration is perturbed due to the presence of the nanoparticles for polymer radius of gyration greater than the nanoparticle radius. Moreover, we determine the primitive path network [3] by using topological algorithms [4] and observe that the entanglement length decreases with the addition of nanoparticles, and is altered by the nanoparticle size and polymer-nanoparticle interactions.
[1] S. Gam, J.S. Meth, S.G. Zane, C. Chi, B.A. Wood, M.E. Seitz, K.I. Winey, N. Clarke, R.J. Composto, Macromolecules, 44, 3494 (2011).
[2] C. C. Lin, S. Gam, J. S. Meth, N. Clarke, K. I. Winey, R.J. Composto, Macromolecules, 46, 4502 (2013).
[3] A. Karatrantos, N. Clarke, R.J. Composto, K.I.Winey, Soft Matter, 9, 3877 (2013).
[4] M. Kroger, Comput. Phys. Commun. 168, 209 (2005).

This presentation will describe the use of polymer capsules, especially emulsion droplets, for the encapsulation of nanoparticles, selective deposition of nanoparticles on substrates, and pickup of nanoparticles from substrates. We attempt to understand the key interactions of nanoparticles resting on a substrate and their affinity for attaching onto, or entering into, droplets that come into contact with them. The wetting characteristics of the substrate, particles, and droplet or capsule wall are expected to be key to successful manipulation of the nanoparticles. Moreover, we find that integrating functionality into the capsule is critically important for promoting strong interactions with the nanoparticles.

Quite recently, the multiphase lattice Boltzmann method has been extended to account for the effect of thermal fluctuations thus opening the way for a wide range of applications from the micro- to nano-fluidic phenomena [1-3]. In this talk, I briefly present theoretical foundations of the method and address a number of physical problems where the method has successfully been applied to uncover new interesting phenomena. This includes nanodrop spreading, surface roughening [4] and critical dynamics of an isothermal fluid [5,6]. In all the cases investigated, the reliability of the approach is underlined by comparison with theoretical predictions [1-6].
[1] M. Gross, R. Adhikari, ME Cates, F. Varnik, Phys. Rev. E 82, 056714 (2010).
[2] M. Gross, ME Cates, F. Varnik, R. Adhikari, J. Stat. Mech. P03030 (2011).
[3] M. Gross, R. Adhikari, ME Cates, F. Varnik, Phil. Trans. R. Soc. A 369, 2274 (2011).
[4] M. Gross, F. Varnik, Phys. Rev. E 87, 022407 (2013).
[5] M. Gross, F. Varnik, Phys. Rev. E 85, 056707 (2012).
[6] M. Gross, F. Varnik, Phys. Rev. E 86, 061119 (2012).

Tethered Nanoparticles (TNPs) are a unique class of building blocks with assembly behavior analogous to copolymers, surfactants, and some liquid crystals. In all such polymeric systems, complex network morphologies, like the Double Gyroid, appear for suitable choices of building block architecture. These morphologies are highly sought after for applications ranging from photonics, where band gaps can be tuned based on the length-scales of the network, to microfluidics, for their potential to subdivide space into non-interacting regions. We discuss the presence of these and other complex morphologies in TNP Telechelics - two nanoparticles of different chemical specificity joined by a polymer tether - simulated via Langevin Dynamics at varying nanoparticle size ratios, polymer tether lengths, temperatures, and packing fractions. By drawing comparisons to other types of systems, particular focus is placed on how introducing a polymer tether as an assembly strategy affects the types of ordered structures that emerge. We further discuss how TNP systems can be adapted to create structures with symmetries not traditionally observed in polymeric systems.

Monolayer-protected gold nanoparticles (AuNPs) have recently emerged as a prominent tool for studying biophysical interactions. In recent work, we have shown that monolayer-protected AuNPs can spontaneously insert into lipid bilayers depending on the composition of the protecting monolayer and diameter of the particle, embedding the AuNP in a conformation resembling a transmembrane protein (1). A key behavior essential to the function of many transmembrane proteins are interactions with the surrounding lipid matrix, where mechanical deformation of nearby lipids may be coupled to protein function. For example, the open probability of bacterial mechanosensitive channels depends on the surface tension of the surrounding lipid layer, which itself is modulated by the extent of hydrophobic mismatch between protein and lipids (2). In general, many embedded membrane proteins exert an influence on surrounding lipids that lead to deformation and membrane-mediated interactions that may be essential for function. Similar to these membrane proteins, embedded AuNPs will also induce membrane deformations related to the same physicochemical forces, namely the desire for hydrophobic ligand backbones to match to the hydrophobic tails of the lipids. Unlike many transmembrane proteins, however, the AuNPs are also highly charged which may exert preferential interactions on surrounding lipid head groups. The ability to control the extent of membrane deformation by tailoring AuNP surface properties may thus lead to the ability to manipulate protein function, induce aggregation via membrane-mediated forces, or encourage preferential interactions between AuNPs and membrane-bound molecules for sensing applications.

In this work, we use atomistic molecular dynamics simulations to show that the membrane around the embedded particles may experience local thinning, headgroup reorientation, and an increase in lipid density depending on the size and surface composition of the embedded AuNP. We quantify the extent of these deformations and illustrate the complex interplay between lipid tail group and head group interactions that go beyond pure thickness deformations that may be expected from coarse-grained or continuum models. We also show that the deformation properties of such AuNPs may induce favorable aggregations with other biological molecules, such as cholesterol. This work thus suggests guidelines for the design of particles that spontaneously partition into lipid bilayers and influence local membrane mechanical properties in a targeted manner.
(1) R. C. Van Lehn et al, “Effect of particle diameter and surface composition on the spontaneous fusion of monolayer-protected gold nanoparticles with lipid bilayers,” submitted.
(2) R. Phillips et al, “Emerging roles for lipids in shaping membrane-protein function,” Nature 459 (2009).

An understanding of the statistical and time-dependent features of cluster formation is essential for the application and control of the dispersion quality of colloidal nanoparticles (CNPs). We performed computational and theoretical studies on the formation of clusters in CNPs, focusing on the scaling behavior of the growth of the cluster size and size distribution, with analysis of the fractal dimension. For the study, we employed a kinetic Monte Carlo (KMC) algorithm in which NPs are moved by self-avoiding diffusive jumping with a random walk. To describe diffusion of the NPs in a colloidal environment, the diffusivity was modeled as a configuration-dependent function of the interacting potential of the clusters. To verify the computational analysis, a kinetic model based on rate theory (RT) was used to analyze the temporal evolution of the concentrations of the monomer and clusters. The KMC simulations agreed well with the predictions from RT in terms of the description of the scaling behaviors. In particular, we observed that the scaling exponents for the average cluster size and weight are smaller than the conventional predictions, although the fractal dimension of the cluster was comparable to that observed in the typical reaction-limited aggregation of particles. We provided a semi-empirical explanation of how the scaling exponent of the cluster size and weight should be reduced depending on the scaling behavior of the monomer concentration. We also provided a model to explain the dependence of the induction time for cluster formation on the initial monomer concentration; the model is supported by the KMC simulation and RT calculation. The results of this study can be used to design and control the colloidal quality of NP dispersions by understanding the cluster growth behavior and its dynamics.

Translocation of polyelectrolytes, such as DNA, through a nanopore or nanochannel with small diameter d requires application of very high electric field E. The electrostatic energy gain of a small chain section of size d upon entering nanochannel has to be higher than free energy cost due to confinement of this section in the nanochannel. Finding ways to lower the threshold electric filed E as well as to precisely control the location and conformation of molecules is highly desirable especially for narrow nanopores and nanochannels. We propose a conceptually different way of handling polyelectrolytes in confined spaces using funnels of specially designed shapes. The electric field is focused in the funnel and is pushing the whole molecule towards the narrow part of the funnel where the entrance into the nanochannel is located. Outer parts of the polyelectrolyte located downfield are compressing polymer sections located in the narrower part of the funnel creating high osmotic pressure. We call this phenomenon - the electro-osmotic piston. At lower electric field polyelectrolyte is localized in a particular place within the funnel. Variation of electric field moves the molecule up or down the funnel. These electro-osmotic tweezers can move the molecule towards the nanochannel, preparing it for translocation in a precisely controlled way. The critical threshold filed required for DNA translocation through the channel of the same diameter d is orders of magnitude smaller than in the absence of the funnel because it is pushed in by the electro-osmotic pressure of the whole molecule rather than sucked in by the field acting on a small section of size d. We develop analytical theory of electro-osmotic tweezers and pistons that utilizes novel concepts of electro-hydrodynamic force acting on localized charged objects and modifies the ideas of nano-confinement of compressed chains and gels as well as the effects of marginal solvent. We also perform molecular dynamics simulations and conduct experiments on double-stranded lambda; and T-4 DNA in nanofunnels connected to nanochannels. The predictions of the theory are in good agreement with results of simulations and experiments.

With increasing density imposed by external osmotic pressure, DNA in monovalent salt solutions (e.g., NaCl) is known to go through a set of ordered mesophases, eventually crystallizing into an orthorhombic crystal. While the transition from the (chiral) nematic to the line-hexatic phase has been observed before, it has remained unclear whether the transition is of second order or weak first order. In multivalent salt solutions (e.g., CoHex = Co(NH3)6 Cl3), DNA is known to collapse into an ordered aggregate under osmotic pressure, or even at zero osmotic pressure at large enough concentrations of the multivalent salt.
We use the small but accurately measurable temperature dependence of the osmotic pressure of a PEG solution to fine-regulate the osmotic stress with which it acts on the DNA subphase. This allows us to set the osmotic pressure to an accuracy never achieved before. This advance in experimental methodology allows us then to detect small but nevertheless finite changes in the density of DNA as it goes through the ordering transitions. In this way, we first determine experimentally the small density change that occurs at the (chiral) nematic to line-hexatic phase transition. For Na-DNA, this density change can be translated into a ~1.5 Angstrom change in the interaxial spacing. This density discontinuity at the phase transition either does not depend on the NaCl concentration or varies by less than 0.1 Angstrom as the salt concentration is varied.
Further, we establish that this small density discontinuity of Na-DNA is substantially increased when a multivalent salt, such as CoHex, is added to the solution. For example, in 0.1M NaCl with [CoHex]=0mM, the transition takes place at osmotic pressure Posm~10 atm, and the concurrent interaxial spacing changes from 38 to 36.5 Angstrom. When the polyvalent salt concentration is increased to [CoHex] = 1mM, the transition osmotic pressure is Posm ~4atm and the interaxial spacing changes from 35 to 32.5 Angstrom. There is thus a continuity of states that depends on the concentration of the polyvalent salt. It connects the (chiral) nematic to the line hexatic phase transitions in monovalent salts to the DNA condensation transition in multivalent salts. Increasing CoHex concentration finally leads to a phase separation at zero imposed osmotic pressure. The DNA condensate then shows a finite density corresponding to interaxial spacing ~29.5 Angstrom for [CoHex] = 2mM , and ~28.2 Angstrom for [CoHex] = 3mM, after which it does not vary any more. Establishing a continuity of thermodynamic states for these two ordering transition, thought to be completely unrelated before, represents an important advance in our understanding of DNA polymorphism in electrolyte solutions. It is liberating and essential to be able to see unexpected details of chameleon possibilities in DNA morphology in vitro that are likely used for action in vivo.

Ionic liquids (ILs) are currently being investigated as a viable solvent medium to dissolve cellulose. Many of their inherent properties make them desirable in environmentally-friendly processes converting cellulosic biomass into usable resources such as fuels, and plastics. However, many practical considerations still remain. One main obstacle is the lack of understanding surrounding the dissolution of cellulose. Though it has generally been accepted that the ability to dissolve cellulose rests with the hydrogen bond accepting ability of the anions, much less is known about the role of the cation. Interestingly, certain cations can render the anion ineffective, and it has also been reported that the imidazolium chlorides exhibit a peculiar "odd-even" effect with respect to the length of the alkyl tail attached to the imidazolium ring.
We systematically examine the pairings of three different anions with 4 imidazolium-based cations of differing tail length for a total of 12 different ILs. Using both molecular dynamics simulation and experiments we are able to demonstrate how the anions and cations work in unison to solvate cellulose. We find that the tail length affects the coupling of the anion with cellulose and ultimately the overall interaction energy. Further, this coupling of the anions with cellulose contorts the overall structure, facilitating the weaker interactions between the cations and cellulose. A deeper understanding of the interplay between the anions and cations could aid in the search for more effective cellulose solvents.

Silk and elastin are exemplary protein materials that exhibit excellent material properties. Silk is uniquely strong, exhibiting strength values that surpass those of engineering materials such as Kevlar and steel, while elastin displays exquisite flexibility and has a reversible ability to fold into a more structured form upon increasing temperatures at which many proteins unfold and denature, a phenomenon termed the inverse temperature transition (ITT). The ITT is a reversible, controllable process, via sequence modification and environmental factors, suggesting applications in drug delivery and biomimetic devices. Here, we design various silk-elastin-like protein polymers (SELPs), which combine repeating silk and elastin blocks, introduced as biologically-inspired biomaterials that combine the distinctive properties of the composing parts to achieve strong and extensible, mutable biomaterials. We use Replica Exchange Molecular Dynamics, an accelerated sampling method, to derive the structure of various SELP sequences. Combining molecular dynamics simulations and experiment, we study the effect of silk to elastin ratio in SELPs as well as the effect of elastin block modification on the ITT and molecular structure. To validate our methods, we develop models of elastin-like sequences, identifying a previously observed ITT, which suggests that our models are capable of capturing this transition. Our findings confirm the applicability of SELPs as smart materials for biomimetics and medicine.

Genetically engineered peptides for inorganic solids (GEPI) are of a broad interest due to their ability to control the functionalization of nanomaterials. Peptides are either biocombinatorially selected and de novo tailored with a explicit amino acid sequence that can specifically and selectively bind to an inorganic solid surface. This versatility is due to the biochemical interaction between GEPI and the surface and the strength of the interaction can be adjustable with the choice of the sequence with respect to specific morphology, size, crystallography or surface stereochemistry. Among the inorganic solid surfaces, graphene has special attention with its extraordinary physical and chemical properties such as large surface area, high intrinsic mobility, high young modulus and thermal conductivity as well as its optical transmittance. Diffusion and self-assembly of peptides on solid materials are significant because of the linker function of these peptides bridges biology and engineered solids. Although some peptides are known to bind to carbon-based solids, systematic study how a given peptide could be made, at will, binding, unbinding, and self-assembling have not been achieved. Using the surface-specific affinity and molecular recognition, followed by surface diffusion and ordered assembly of a dodecapeptide have been interrogated using molecular dynamics and mechanics (MD/MM) simulations. The computational modeling results have been compared with commensurate experimental observations of a generic model for genetically selected dodecapeptide sequence and its rationally-designed mutations, i.e., GrBP5-(Wild Type-WT, and mutations, M1, M2 and M3) adsorbed on graphene surface. In particular, we present here the effects of peptide mutation on physical and chemical specificity and selectivity of surface recognition and assembly. The importance of specific amino acids and their spatial distribution upon folding subsequent to the surface binding emphasizes the strong correlation between amino acid sequence and the dynamics of the molecular architecture followed by a sequence of surface phenomena (e.g., diffusion, intermolecular interaction, ordering). The energetic analysis by modeling at room temperature strongly correlates with the experimental results obtained by surface characterization, e.g., atomic force microscopy. The model, therefore, reproduces experimentally observed features such as diffusion which can also help to analyze the initial stages of clustering or ordered self-assembly of the peptides. The computational model developed herein would be applicable to a given peptide/2D atomic solid, allowing the rational design and tailoring of these molecularly hybrid materials for a wide range of devices, from bioelectronics to biosensors.

The size of colloidal metal nanocrystals prepared by reductive synthesis from solution precursors can be tailored in the sub-4 nm range using biological capping agents; particularly peptides selected via combinatorial approaches and their mutants. Through accessing this size range, the surface-to-volume ratio as well as specific atomistic details of the surface structure may be customized to achieve superior functionality in many applications such as carbon-carbon coupling reactions. However, dominating size-tuning mechanisms and effect of peptide coverage on catalytic activity of nanoparticles still remain ambiguous. This ambiguity is mainly related to inaccessibility of the peptide-particle interfaces to most of the advanced detection techniques and complexity of the multi-phase systems. In this study, we carry out a critical molecular dynamics simulation on adsorption strength, wrapping geometry, and coverage of nanoparticle surface facets in the presence of a palladium specified peptide (Pd4: TSNAVHPTLRHL) and its mutants to understand the likely mechanism of size control in aqueous solutions. In addition, we present a detailed examination of surface facets, atom types, and accessibility of leachable atoms to understand the influence of peptide coverage during oxidative addition and potential atom leaching during Stille carbon-carbon coupling reactions. Extensive molecular dynamics calculations using the INTERFACE force field on thermodynamically stable particles in sub-4 nm size range demonstrate that the particle size is independent upon the global peptide binding strength although a critical minimum adsorption is necessary to block particle growth. Pd4 and its mutants are capable to extend across several surface facets, i.e. {111}, {100}, and {110}, when adsorbed on the particles because of their relatively long backbone chain. Pd4 and its mutants show the same behavior on particles with similar size and surface structure. The wrapping geometry depends on the length of the peptide backbone, the particle size and surface structure. The relative coverage of surface by peptides is likely responsible for fine-tuning the particle size upon in-situ reductive synthesis. The surface coverage correlates with inverse of Pd particle size, i.e. the higher the surface coverage the smaller will be the achievable average particle size due to reduced access of particles surface for atom deposition. Visual analysis of peptide-particle interactions ultimately show that accessibility of leachable atoms and computed relative atom abstraction rates (relative TOFs if partial atom abstraction limits the rate) are not significantly influenced by nanoparticles surface coverage but rather by the particle morphology and the type of peptides itself (e.g. thiol versus non-thiol). The presented relationships are a step toward rational design of peptide directed palladium nanostructures with catalytic functionality.

The interaction of biomolecules with noble metal (Ag/Au) surfaces and/or nanoparticles has potential applications in a wide range of areas including: materials synthesis, biosensing and nano-medicine.[1] These applications have prompted significant interest in the field, however, to fully realize these potential applications, a deeper understanding of these interfacial interactions at the molecular level is required. A number of recent studies have reported use of molecular dynamics (MD) to investigate the interaction of peptides with gold surfaces.[2,3] These investigations have provided a better understanding of the factors at play at the atomistic level in these system. In contrast, there has been relatively little work investigating the interaction of peptides with silver surfaces. The AgP-CHARMM force-field has been specifically developed to model the interaction of amino acids/peptides with the Ag(111) and Ag(100) surfaces.[4] This force-field has been parametrized against the results of PW-DFT calculations, using the vdW-DF functional,[5] of small molecules at a silver interface. In addition, it has been designed to be compatible with the GolP-CHARMM force-field[6] used for the simulation of peptides with gold surfaces, allowing cross-comparison between the behaviors of peptides on the two different metal surfaces to be made. We have used AgP-CHARMM to investigate the interaction of a variety of amino acids at the aqueous Ag(111) and Au(111) interfaces, and have compared with similar data obtained for Au(111) and Au(100), respectively. The free energy of adsorption of the amino acids to the metal surfaces has been calculated using meta-dynamics. Our results show that despite the similarity of the two metals there are some important differences in the nature of the interaction of amino acids (and thus peptides) with the two surfaces.
[1] Grey, J.J., Curr. Opin. Struct. Biol., 2004, 14, 110-115.
[2] Feng, J. et al., Small, 2012, 8, 1049-1059
[3] Tang, Z., et al., “Biomolecular Recognition Principles for Bionanocombinatorics: An Integrated Approach to Elucidate Enthalpic and Entropic Factors”,, In Submission.
[4] Hughes, Z.E., Wright, L.B. and Walsh, T.R., “AgP-CHARMM: Extension of the first-principles based GolP-CHARMM force-field to Ag(111) and Ag(100) surfaces.”, In Preparation.
[5] Dion, M., et al., Phys. Rev. Lett., 2004, 92, 246401.
[6] Wright, L.B., Rodger, P.M., Corni, S. and Walsh, T.R. J. Chem Theory Comput. 2013, 9, 1616-1630.

Bio-composites and bio-minerals provide inspiration for the design and manufacture of other high-performance composites. For example in nacre (the mineral-rich material common to mollusk shells), proteins play an important role in exerting polymorph selection of CaCO3 minerals, stabilizing the aragonite polymorph of calcium carbonate[1] . How this regulation works at the molecular level is unknown, because we do not know how the protein structure(s) relate to the function in this context. Determination of these protein structures in solution and whilst adsorbed at the mineral interface, is now possible, but difficult to accomplish in practice[2] . As the first steps towards identifying the surface-adsorbed structures, we have focussed initially on probing the structure(s) of the N-terminal half of a known aragonite-stabilizing protein, n16. The resulting peptide (30 a.a.) is denoted n16N. Previous work via solution-state NMR1 has probed the conformational ensemble of n16N in solution both in pure water and in response to the presence of Ca2+ ions. These authors found in both cases that n16N exhibited a large degree of conformational lability, featuring an ensemble of random-coil structures; and thus n16N has been proposed as an intrinsically disordered peptide (IDP). IDP&’s present a challenge to simulation, due to their lack of a single well defined native state requiring extensive and exhaustive sampling of conformational space. In this work, we present results from our molecular dynamics simulations using the advanced sampling technique Replica Exchange with Solute Tempering(REST)[3-5] in determining this ensemble of structures. We compare our findings against the existing experimental NMR data[1], reveal evidence for different sub-domains within n16N, and, determine the key intra-peptide interactions involved in stabilising the ensemble of conformations featured by n16N[6].

1. S. Collino and J. S. Evans, Biomacromolecules, 9, 1909 (2008)
2. P. A. Mirau, R. R Naik and P. Gehring, J. Amer. Chem. Soc., (2011), 133, p18243
3. Moors, S.L.C. et al., J. Chem. Theory Comput., (2011), 7, p231
4. Terakawa, T. et al., J. Comput. Chem., (2011), 32, p1228
5. Wang, L. et al, J. Phys. Chem. B, (2011), 115, p9431
6. A. H. Brown, P.M.Rodger, T.R.Walsh, In preparation, 2013

Our goal is to develop a model for nanostructured biomaterials that form through the self-assembly of two amphiphilic lipid species. Individual lipid molecules are represented by a hydrophilic head group and two hydrophobic tails. The lipid species can differ in terms of specific chemical properties of the polar head groups and the hydrocarbon tail groups. We study the self-organization of the lipid species in pre-assembled stable hybrid vesicle or membrane bilayers. The bilayers are formed via the self-assembly of the two amphiphilic lipid species in the presence of a hydrophilic solvent. We use a Molecular Dynamics-based mesoscopic simulation technique called Dissipative Particle Dynamics which simultaneously resolves the structure and dynamics of the nanoscopic building blocks and the hybrid aggregate. We will present our characterization of the morphology and dynamics of the hybrid nanostructures with focus on the phase segregation behavior of the system and the kinetics of the coarsening dynamics. The morphological analysis of the hybrid aggregates and their material characterization can be combined to predict the structural and dynamical properties of other hybrid soft materials.

The assembly of sequence-specific synthetic polymers into bilayer nanosheets [Sanii et al, J. Am. Chem. Soc. 133 20808 (2011)] demonstrates how nonequilibrium protocols can direct assembly of information-rich building blocks into complex structures. However, it also challenges our ability to model and theoretically inform the dynamic and large-scale mechanisms underlying such assembly processes. In our experimental system, sequence-specific peptoid polymers first form a monolayer at the air-water interface then buckle into ordered bilayers as the interface is compressed. Elucidating the multi-scale and dynamic mechanisms underlying these processes requires a simulation approach that captures the behavior of polymers on length scales ranging from the monomer to the polymer and beyond. Here, we present a novel coarse-grained modeling approach that combines accuracy with computational speed to faithfully model the surface adsorption, compression, and buckling of the polymers. We parameterize detailed interactions among anisotropic coarse-grained sites to match experimental and all-atom computational data for their molecular analogs. We arrive at a model that reproduces the available experimental data (compression isotherms and X-ray diffraction patterns) while providing a real-space picture of the microscopic behavior inaccessible to in-situ experimental methods.

Current research has attempted to mimic the production of silk fibers due to their desirable mechanical performance. However, the natural archetypes performance has not yet been matched. In spiders, silk fibers assemble under shear flow in the spinning duct. This flow creates a tough polymer nanocomposite consisting of small and strong beta-sheet nanocrystals embedded in a flexible semi-amorphous phase. Previous NMR spectroscopy studies of the spidroin sequence MaSp1, e.g. from Nephila Clavipes dragline silk, has revealed a primary structure that is highly repetitive. In the present study, a large-scale Replica Exchange molecular dynamics simulation in implicit solvent is used to determine equilibrium structures of the spidroin sequences with varying repetition length. We equilibrate in explicit solvent in the presence of salt at physiologically relevant concentrations. We then determine the critical shear force needed to enforce an alpha-helix to beta-sheet transition via steered molecular dynamics experiments. We estimate the shear force needed in the spider spinning duct to create the highly aligned fiber structure. This study may inform the design of bioinspired processing units that mimic the natural spinning process.

Fibrillar collagens are common tissue scaffolds, and are known to assemble into ordered arrays in matrices such as bones, tendon, and cartilage. They have also been observed to self-assemble in vitro, and the resulting scaffolds have a variety of biological and biotechnological applications. The variety of self-assembled collagen structures, often hierarchical, can lead to a variety of applications and to fundamental understanding of the assembly processes. Such processes can be highly complex, and can benefit from even the most basic models that capture the essential interactions between collagen molecules in solution which leads to assembly. Here, we propose a coarse grained model in which the collagen strands are treated as strings of beads connected via springs, with well defined interactions between beads that belong to different strings. Molecular dynamics simulations employing this coarse-grained description indicate that the morphology of self-assembled collagen on a flat substrate is governed by the interplay of two competing factors, namely the intermolecular collagen-collagen (c-c) interaction and the collagen-substrate (c-s) interaction. The utility of this model lies in its ability to vary each of these interactions independently, since such a decoupling is difficult to achieve in the laboratory. Our simulations clearly demonstrate that the c-c interactions promote surface diffusion favoring the formation of three-dimensional collagen bundles, while strong c-s interactions lead to random monolayer networks. Furthermore, the configurations generated by this model are in excellent agreement with atomic force microscopy experiments.

The elastin-like peptide (ELP) is temperature sensitive biopolymer which undergoes conformational transition at critical temperature. The design of elastin-like peptides is motivated by their ability to self-assemble in response to clinically relevant intrinsic stimuli; however the understanding of how sequence and length of the peptide controls self-assembly processes is unknown. The hydrophobic peptide with repeating Val-Pro-Gly-Val-Gly sequence, (VPGVG)n, is a widely used model of peptides. While this peptide is soluble in water at cold temperatures, it can undergo an inverse self-assembly into fibril like structures at physiological temperatures. Recent experimental observations show that the order of amino acids in the ELP sequence also plays an important role since (VGPVG)n and (VPGVG)n peptides show the difference in thermo-transition behavior. In order to elucidate the factors driving the conformational transitions of ELPs we performed molecular dynamics simulations of (VPGVG)n and (VGPVG)n peptides with different length and at a range of temperatures. Comprehensive analyses of the hydrophobicity, shape, size, hydration and dynamics of (VPGVG)n and (VGPVG)n suggest that individual peptides are in a globular conformation at all temperatures, but exhibit change in hydrophobicity due to rotations of side chains with increase in temperature. The temperature dependent self-assembly is a collective phenomenon that is driven by this increase in hydrophobicity of individual peptides. The sequence dependent difference for ELP peptides, such as (VPGVG)n and (VGPVG)n, is attributed to the different level of side chains rotations and show different hysteresis behaviors. The results from our study provide an atomic-level description of the thermo-responsive conformational properties of elastin-like peptides and explain experimental observations. Support for this research was provided by the NSF's Research Triangle MRSEC (DMR-1121107).

When designing materials from a bottom-up approach, the desired structure should not only have a low free energy as compared to competing structures. It should also have a favourable kinetic pathway, as a molecular system can easily be trapped in a thermodynamically unfavourable, but kinetically accessible state. Another possibility is that there is a multi-step nucleation process, in which the structure first transforms to an intermediate metastable state, from which it transforms further to the stable one.
These pathways quickly become very complex and in order to understand the possibilities one has to resort to molecular simulations. Another challenge is that the time scales accompanied with these transitions can be beyond that attainable by means of brute force methods. Recently advanced transition path sampling techniques have proven to be successful for tackling this problem in the case of simple monatomic and binary blend models. We have extended such modelling to be able to study more complex models that mimic small organic molecules.
We report results from molecular dynamics simulations, acquired by a combination of brute force, transition path sampling techniques and free energy methods to shed light on the process of forming a crystal from a melt of semiflexible short-chain molecules. Both thermodynamic and kinetic properties related to the growth will be discussed. We find that the chains have a tendency to aggregate in clusters that grow into 2-dimensional structures. We furthermore show that the anisotropy in growth rates, caused by the anisotropy of the molecules, determine the nanostructure of the resulting crystalline material.

Polymer thin films are usually fabricated using solvent-based thin-film deposition technologies (e.g., spin coating, drop casting). Depending on the specifics of the polymer blend and processing conditions, different morphologies are typically formed in the thin film. It is of paramount importance to understand how phase separation is initiated and how it evolves to form the final morphology. Such an understanding will enable rational design of thin films with desired structure and thus tailored properties. However, it is challenging to experimentally visualize morphology evolution during processing (processes involved are highly dynamic at low scale and typical components do not show high contrast). Consequently, details of morphology evolution during solvent-based thinning deposition technique from initiation stage to the final morphology are still under debate with three mechanisms being hypothesized.
Using an experimentally validated computational framework we investigate the competing effect of kinetics and thermodynamics on morphology evolution in a typical thin film organic system using. We perform a high-throughput computational analysis by varying processing parameters (evaporation rates, blend ratio) to observe morphology evolution. We identify four mechanisms through which the morphology evolves and phase-separation is initiated -- (i) from the top surface, (ii) homogeneously along the thickness, (iii) from the bottom surface, and (iv) initiation from middle zone. These mechanisms have been reported in earlier experimental studies. We unravel the origin of this behavior by identifying processing conditions necessary for each mechanism to operate. Specifically, we utilize a linear stability analysis to identify which mechanism of phase-separation is chosen for a given processing condition. This allows us to construct a phase-diagram of the processing condition phase-space. We further elucidate the spatial and temporal heterogeneity of these conditions and the interplay between the kinetics and thermodynamics for these four mechanisms. Finally, we utilize this analysis on a several realistic systems PS:PMMA, PFB:PCBM, PFB:F8BT and and methodically investigate phase space of evaporation rate and blend ratio.

Liquid crystalline (LC) mesophases of polymeric semiconductors present significant interest for technological applications, leading to materials with improved properties, e.g., charge-carrier mobility. We present a strategy [1,2] for modeling LC mesophases of such materials, with uniaxial or biaxial symmetry. For this purpose, we combine a description of polymer architecture based on coarse-graining of atomistic single-chain conformations with non-bonded interactions captured by soft pair-wise directional potentials. The latter can be motivated by functionals of collective variables (local densities and orientational tensors). Achieving biaxial structuring, i.e., “planar orientation” of polymer chains is essential for describing effects of strong anisotropy in interactions (e.g., pi-pi stacking) and chain architecture (e.g., side chains) on LC morphologies.
Inspired by experiments [3] reporting LC mesophases in poly(3-alkylthiophenes) we address poly(3-hexylthiophene) (P3HT) polymers as a test case. We demonstrate that the approach offers a realistic description of material properties of nematic polymeric LC&’s, such as the effects of molecular weight on phase behavior and the Frank elastic constants.
Capturing biaxial nematic P3HT mesophases offers insights into how collective orientation and planarization of polymer chains affect properties related to charge transport. These are considered from the scope of the distribution of the lengths of conjugated segments, defined via conjugation-breaking torsional defects. For moderate chain lengths these defects are found to concentrate towards the ends of the polymer. Relating the distribution of lengths of the conjugated segments to the density of states (DOS) through a simple estimation, neglecting effects of electrostatic interactions, we observe that orientational correlations in biaxial morphologies alone are sufficient for generating a spatially correlated, non-Gaussian DOS. [1] K. Ch. Daoulas, V. Ruehle, K. Kremer, J. Phys.: Condens. Matter 2012, 24, 284121 [2] P. Gemünden, C. Poelking, K. Kremer, D. Andrienko, K.Ch. Daoulas submitted [3] V. Ho, B. Boudouris, R. Segalman, Macromolecules 2010, 43, 7895.

The phase behavior of material systems which are termed thermodynamically small has been the subject of intensive theoretical study over the past two decades. These finite systems consist of a very small number (10-100) of interacting particles and, as such, they are far removed from the infinite limit of traditional macroscopic thermodynamics. In these systems, the nature of thermodynamically stable phases and the phase transitions between them are of particular interest. Developing an understanding of the phase behavior of these systems has direct application to the self and directed assembly of colloidal particles into structures within materials and devices. In this presentation, we report results of a systematic investigation of phase behavior in two thermodynamically small systems, finite assemblies of colloidal particles interacting via a hard-core and a depletion-attraction potential and the 38-member Lennard-Jones (LJ38) cluster. In the colloidal system, we have studied order-to-disorder phase coexistence over a broad range of two parameters, the system size, expressed by the number of particles, N, in the assembly, and the inter-particle interaction strength, which is controlled by the depletant osmotic pressure. In the LJ38 cluster, we have focused on a single parameter, the system temperature, and studied its effects on polymorphic solid-to-solid and melting solid-to-fluid phase transitions.
To provide a description of the phase behavior of these systems, we have constructed free-energy landscapes (FELs) based on Monte Carlo umbrella sampling (MC-US). For the implementation of MC-US and the construction of meaningful FELs, a coarse-grained representation of the system is required. This coarse graining reduces the dimensionality of the system from the full 3N-dimensional representation into a coarse description with a much lower dimensionality. To perform this dimensionality reduction, we have applied the diffusion mapping approach to both systems of interest, with data sets provided by Brownian- or molecular-dynamics simulations for the colloidal and LJ38 clusters, respectively. In the colloidal particle assemblies, we find that for very small clusters only a single fluid-like phase is stable. However, as the cluster size increases, a second ordered phase emerges in coexistence with the fluid-like phase; this second stable phase at increased cluster size is a crystalline phase. The onset of stability of this crystalline phase corresponds to the size of a critical crystalline nucleus and marks the onset of crystallization in colloidal particle assemblies. In the LJ38 cluster, we find that, at very low temperatures, the system only samples its minimum-energy, octahedral, configuration. However, as the temperature increases, the system undergoes a polymorphic transition between two different solid phases followed by an order-to-disorder, solid-to-fluid transition at even higher temperatures.

The glass crossover occurs when crystallization is avoided by increasing size polydispersity of the particles. Polydisperse systems are common in colloidal suspensions, and polydispersity effects on nearly hard sphere colloids have been investigated by many experiments, theoretical and computational studies for polydisperse hard sphere fluids. However, a hard sphere system has a purely repulsive interparticle potential and the only liquid-solid transition occurs. Further studies are required to reveal the effects of size polydispersity on a system with interparticle attractions associated with the gas-liquid and liquid-solid transitions. The Lennard-Jones (LJ) potential is standard model with short-range attractions. The phase diagram of monodisperse LJ systems is well determined. However, the effect of size polydispersity on LJ systems is still unclear. We perform generalized replica exchange method (gREM) computer simulations [1] on model glass formers, 12-6 polydisperse LJ systems with a Gaussian size polydispersity, denoted standard deviation s, at constant pressure. The effect of size polydispersity on the gas-liquid and liquid-crystal transitions, and on the liquid-glass crossover are discussed between the triple point and critical point [2].
At high temperatures in the gas phase, there are no polydispersity effects on the observables. Size polydispersity plays an increasingly important role with decreasing temperature. With decreasing temperature, the enthalpy, internal energy and volume decrease with increasing polydispersity. The gas-liquid transition point also shifts to higher temperature. The dependence of enthalpy upon polydispersity is linked to that of the internal energy change. Enthalpy and internal energy therefore show similar behavior over the whole temperature region. On the other hand, volume shows different polydispersity dependences in the liquid, crystal, and glass states. At low temperature, sufficient polydispersity causes the liquid-crystal transition to disappear. The pressure dependence of polydispersity effects are also discussed, and a preliminary gas-liquid-crystal-glass phase diagram of polydisperse LJ systems is obtained. Small size polydispersity does not affect the liquid-crystal transition point at constant pressure. Colloids up to s=0.05 size polydispersity are usually treated as monodisperse in experiments. However, the absolute values of enthalpy and internal energy on the liquid and crystal branches are strongly affected by small size polydispersity Our results suggests that quantitative observations related to enthalpy and internal energy require more careful attention [2].
We will also discuss the effect of size polydispersity above the critical point in the meeting.
[References]
[1] J. Kim, T. Keyes, and J. E. Straub, J. Chem. Phys. Vol.132, 224107(2010).
[2] Y. Terada, T. Keyes, J. Kim, and M. Tokuyama, AIP Proceedings Vol.1518, p.776(2013).

Solubility plays a significant role in drug design, synthesis, pharmaceutical formulation, absorption and bio-distribution. It is a critical parameter in drug effectiveness, efficiency and overall drug development. Since most of these drugs are in nanoscale and many have been nanoencapsulated with polymers, the systems for which they are synthesized can be assumed to be colloidal in nature. Theoretically, solubility can be predicted based on the forces of interaction among colloidal particles like nanoparticles in a suspension, and these interactions are critical in determining the nanoparticle behavior and properties, hence characterizing its biological applications. There are several proposed models to determine the solubility correlation and prediction based on these forces of attraction. However, most of them are based on non-electrolytes in a solvent mixture, and non-polar solvents, therefore restricting its applicability to non- ionizable compounds and non-polar solvents. In this work, dielectric constant, which is a dominant factor of solubility, is used to predict cosolvency of nanoparticle in polar and non-polar solvents and evaluate their solubilization behavior. Although, a new model and method to determine solubility based on dielectric constants in nonhomogenous system is proposed in this study, other solubility models both DVLO and non-DVLO based are also presented for comparison. Modeling and simulation of the colloidal system is also done to measure nanoparticle stability and aggregation.

Polymer materials are used for polarizer protecting films of liquid crystal displays (LCDs). However, birefringence of polarizer protecting films changes the polarization state of the incident light and consequently degrades the contrast of images of LCDs. Therefore, birefringence needs to be eliminated. Our group has proposed a random copolymerization method for eliminating birefringence [1], [2]. The birefringence of copolymer is designed by using three equations which describe the relationship between birefringence properties (intrinsic birefringence and photoelastic coefficient) and weight fraction of monomers. In previous research, a zero-zero-birefringence polymer (ZZBP) which exhibits no orientational birefringence and no photoelastic birefringence was demonstrated in a system consisting methacrylate monomers by solving the equations. The ZZBP has the advantage that it can maintain the polarization state of the incident light. On the other hand, improving glass transition temperature (Tg) of ZZBPs is required because Tg of the ZZBP is not enough to apply to polarizer protecting films. Therefore, we propose N-methyl maleimide (MeMI) as a candidate for a material of ZZBPs for the first time because of its high heat resistance. The purpose of this article is to clarify birefringence properties of poly(MeMI) and to design a novel ZZBP using MeMI.
We synthesized poly(methyl methacrylate (MMA) /MeMI) in various composition ratios and measured birefringence properties of the copolymer films. Then, we estimated birefringence properties of poly(MeMI) from those of the copolymer films. The results show that the intrinsic birefringence and the photoelastic coefficient of poly(MeMI) are positive and positive, respectively. Based on the results, we calculated composition of monomers in a system of MMA, benzyl methacrylate (BzMA), and MeMI for designing a ZZBP. Then, we synthesized poly(MMA/BzMA/MeMI) with the optimal composition and evaluated the birefringence properties and the Tg of the copolymer film. The results show that synthesized polymer is free of two types of birefringence. Also, the Tg of the ZZBP is 120 degrees C that is higher than those of ZZBPs in previous works [1], [2].
[ACKNOWLEDGEMENT]
This research is supported by the Japan Society for the Promotion of Science (JSPS) through its “Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program)”.
[1] A. Tagaya, H. Ohkita, T. Harada, K. Ishibashi, and Y. Koike, Macromolecules, 39, 3019-3023 (2006).
[2] S. Iwasaki, Z. Satoh, H. Shafiee, A. Tagaya, and Y. Koike, Polymer, 53, 3287-3296 (2012).

Recent studies have shown that mesoscale-aligned nanoparticles are of tremendous interest for a broad range of important applications including energy conversion, optoelectronics, biological/chemical sensors, and nanophotonics. The first step towards understanding these novel plasmonic structures is the theoretical and computational modeling of their plasmonic properties in interplay with each other and their surroundings. This interplay can include basic interactions such as the presence of a substrate in close contact with the nanoparticle, which breaks the symmetry of the local surface plasmon resonance mode. More sophisticated modes are obtained by the addition of more particles and through more complex aggregations which result in plasmonic enhancement. These specific aggregations of nanoparticles are inspired by experiments such as nanoparticle aggregation through porous alumina membranes and through template-assisted methods. We found that the role of nanoparticle shape is often underestimated in the literature. Furthermore, we demonstrate close agreement with the experimental results, especially in the formation of novel plasmonic resonances of nanoparticles embedded in soft responsive materials.

We discuss the controlled sub-diffraction modulations of photosensitive polymer films which are induced by surface plasmon interference in striking contrast in well-known conventional microscopic gratings. The near-field light intensity patterns were generated at nano-slits fabricated in a silver layer with a photosensitive polymer film placed above. We observed that the topographical modulations can be excited only when the polarization is perpendicular to the nano-slits. Moreover, we have shown that light with certain wavelengths resulted in a characteristic topographical pattern with the periodicity three times smaller than the wavelength of incoming light. A combination of experimental observations with simulation modeling showed that the unique sub-diffraction topographical patterns are caused by constructive interference between two counter-propagating surface plasmon waves generated at neighboring nano-slits in the metal layer beneath the photosensitive polymer film. Light intensity distribution simulations demonstrated strong dependency upon the slit array periodicity as well as the wavelength and polarization of incoming light. Furthermore, the theoretical modeling shows that the thin photosensitive polymer acts as a dielectric waveguide with significantly different modes depending on the energy levels present.

With the increased tendency for miniaturization of microelectronic devices, the search for new candidate macromolecular assemblies with non-conventional magnetic and electric properties has become an object of intense research over the past few years. The major goal in the design of one (1-D) or two (2-D) dimensional devices is not only to find the proper organic building blocks but also to align these blocks properly in order to obtain the desired magnetic ordering or electric conductivity [1]. In this work we present theoretical results for the magnetic properties of modeled 2-D Langmuir-Blodgett (LB) films consisting of different types of stable organic π-monoradicals [2]. Their energy spectra and magnetic characteristics are investigated using the many-body band theory of magnetism in π-electron approximation. The main factors which determine the magnitude and character of the magnetic interaction in the 2-D molecular arrangements are thoroughly studied. The proposed models for LB films are potential candidates for new 2-D ferromagnetic materials which are characterized by a substantial increase of the critical temperatures in comparison to the now existing pure organic ferromagnetic materials.
[1] J. Veciana, H. Iwamura, MRS Bulletin 25 (2000) 41-51
[2] N. Drebov, N. Tyutyulkov, F. Dietz, Langmuir 29 (2013) 873-880

The hairpin loop for nucleic acids is an important structural motif that has many possible practical applications.[1-4] One such application of interest is the use of a DNA hairpin as a linker between two structures, where the hairpin conformation will change with temperature. While there is a lot of information on DNA hairpins tethered onto a gold surface,[5] currently there is very little information regarding how the presence of a gold surface affects the structure of the DNA hairpin. It is of interest to know what possible effects gold will have on the conformation of a possible DNA hairpin linker. Molecular Dynamics (MD) simulations at the atomistic level were used to model the DNA hairpin sequence (5&’)GGATAATTTTTTATCC(3&’) alone in solution and also in solution on the Au(111) surface. The DNA in solution simulations were performed as a reference for the later DNA on gold simulations where the GolP-CHARMM Au(111) force field[6] was used to describe the DNA/water/gold interface. As a measure of the degree of order/disorder in the hairpin structure, we have calculated the root mean square deviation (RMSD) in atomic positions, the number of hydrogen bonds between base pairs and native stacking numbers[7] as well as the number of bases in close contact with the gold surface. These analyses indicate that the presence of the gold surface creates an earlier onset of disorder for the DNA hairpin, with respect to temperature. Our findings also suggest that there is a strong interaction between the bases and the gold surface.
References
[1] C. M. Strohsahl, B. L. Miller and T. D. Krauss, Nat. Protocols, 2007, 2,
2105-2110.
[2] G. Zauner, Y. Wang, M. Lavesa-Curto, A. MacDonald, A. G. Mayes, R. P.
Bowater and J. N. Butt, Analyst, 2005, 130, 345-349.
[3] V. Lavalley, A. Laurent, A. Zebda, J. E. Mendez and V. Stambouli, Sensor
Actuat B-Chem, 2007, 124, 564-571.
[4] C. Huang, T. Stakenborg, Y. Cheng, F. Colle, T. Steylaerts, K. Jans, P. Van
Dorpe and L. Lagae, Biosens Bioelectron, 2011, 26, 3121-3126.
[5] O.-S. Lee, V. Y. Cho and G. C. Schatz, J Phys Chem. B, 2012, 116,
7000-7005.
[6] L. B. Wright, P. M. Rodger, S. Corni and T. R. Walsh, J Chem Theor Comp, 2013.
[7] G. Portella and M. Orozco, Ang Chemie Int Edit, 2010, 49, 7673-7676.
[8] Drew, Palafox-Hernandez and Walsh, in preparation 2013.

Using the density functional theory, we have studied the relaxed structure and electronic properties of 4-hydroxy-N-desmethyl-tamoxifen. The calculations were performed with a linear combination of atomic orbitals method using pseudopotentials and the van der Waals approximation for the exchange-correlation potential. We found and analyzed the atomic and electronic structure. The 4-hydroxy-N-desmethyl-tamoxifen properties were compared with tamoxifen properties, both molecules currently are used for treatment of breast cancer. Theoretical and experimental researches report that 4-hydroxy-N-desmethyl-tamoxifen is more favorable to bind to alpha estrogenic receptor than tamoxifen.
Acknowledgements
This research was supported by FOMIX 2011-09 No.: 170297

Intrinsically disordered peptides (IDPs) defy conventional views of the protein/peptide structure-property paradigm, by showing the ability to confer function in the absence of a well-defined secondary or tertiary structure.¹
An interesting sub-class of IDPs comprise of peptides that can reversibly switch conformation from random-coil to a well-defined structure, in the presence of an external stimulus.² To understand this behavior, bioinformatics approaches have previously sought to correlate sequence motifs and individual residues of the IDPs to their structural preferences based on environmental influences.^{3 4}
JAK1 is a de novo designed IDP, comprising Ala-rich segments believed to favour helical structure at lower temperatures⁵ as well as the incorporation of Gla (gamma carboxy-glutamic acid), a strong chelator to calcium. It has been proposed that the Gla-rich regions, with their highly-charged side chains, make helical structures unfavourable in the absence of calcium ions. Based on circular dichroism spectroscopy (CD) results JAK1 is thought to present a random coil structure in Ca2+-free solution but an alpha helical structure when either adsorbed at the aqueous hydroxyapatite (HA) interface, or in the presence of a strong concentration of free Ca2+ ions in solution. A control peptide, cJAK1 (where Gla → Glu), showed no such behavior, and did not bind to HA. Molecular simulation can provide complementary, in-depth molecular-level insights into this reversible switching mechanism.⁶
Here, we present results of large-scale Replica Exchange with Solute Tempering
(REST) molecular dynamics simulations of JAK1 and cJAK1 in buffer solution, as well as in Ca+ saturated solution.⁷ Using Ramachandran plots to investigate secondary structure preferences, and residue-residue contact maps to determine key intra-peptide interactions, we provide molecular level insight into how environmental influences can mediate the conformational switching of JAK1 through side chain interactions.
(1) Uversky, V. N. Int. J. Biochem. 2011, 43, 1090

(2) Dunker, A. K.; Oldfield, C. J.; Meng, J.; Romero, P.; Yang, J. Y.; Chen, J. W.; Vacic, V.; Obradovic, Z.; Uversky, V. N. IEEE BIBE 2007. 2008,

(3) Delak, K.; Collino, S.; Evans, J. S. Biochemistry 2009, 48, 3669

(4) Delak, K.; Harcup, C.; Lakshminarayanan, R.; Sun, Z.; Fan, Y.; Moradian-Oldak, J.; Evans, J. S. Biochemistry 2009, 48, 2272

(5) Marqusee, S.; Robbins, V. H.; Baldwin, R. L Proc. Natl. Acad. Sci U.S.A 1989, 86, 5286

(6) Capriotti, L. A.; Beebe, T. P; Schneider J. P. J. Am. Chem. Soc. 2007, 129, 5281

(7) Terakawa, T.; Kameda, T.; Takada, S. J. Comp. Chem. 2010 32, 1228

Being a key component in modern electronics and electric power systems, the search for new high energy density dielectric materials has become a major scientific challenge. The current standard material for capacitive energy storage, polypropylene, has high electrical breakdown strength and low dielectric loss, but its dielectric constant is low.
We are involved in a search for new classes of polymers with dielectric constant and band gap superior to polypropylene using first principles computations. Essential to this search are (i) strategies to efficiently navigate through the polymer chemical space (ii) schemes to identify promising building block motifs using our `high-throughput' first principles computations1, and (iii) identification of the stable crystal structures for these promising polymers. The present work involves aspect (iii). From our high-throughput first principles computations, it has been already determined that systems containing at least one aromatic group and at least one of the following moieties: -NH-, -CO-, or -O-, is desirable.1
In order to predict the stable structure of these polymeric materials, we used a specifically designed constrained evolutionary algorithm2-4 consisting of well-defined molecular units or motifs that can successfully predict polymeric crystal structures from first principles quantum mechanical computations. The new constrained evolutionary algorithm based computational approach, which has been implemented in the USPEX code,2-4 is applied to a set of polymers for which reliable crystal structure information is already available experimentally, namely polyethylene (PE), polyacetylene (PA), poly(glycolic acid) (PGA), poly(phenylene oxide) (PPO), poly(oxymethylene) (POM), poly (p-phenylene sulfide) (PPS), two forms of poly(vinyldene fluoride) (β -PVDF and δ -PVDF), poly(tetrafluoroethylene) (PTFE), and poly(vinyl chloride) (PVC). The excellent agreement between predicted and experimentally known crystal structures implies that this approach can play a crucial role in new polymer materials design. Next, we apply our methodology to predict the crystal structures for the identified new class of polymeric materials with high dielectric constant and high band gap. The dielectric properties computed for these new polymers in the optimized crystal structures indicate that the identified polymers are of sufficient technological importance, which are hence currently going through a synthesis cycle.
(i) R. G. Lorenzini, W. M. Kline, C. C. Wang, R. Ramprasad, G. A. Sotzing, Polymer, (2013) Article in press
(ii) A.R. Oganov (Editor). Modern Methods of Crystal Structure Prediction. Wiley-VCH, Berlin. (2010).
(iii) A.R. Oganov, C.W. Glass, J. Chem. Phys., 124 (2006), p. 244704; A.O. Lyakhov, A.R. Oganov, M. Valle, Comput. Phys. Comm., 181 (2010) 1623
(iv) Q. Zhu, A.R. Oganov, C.W. Glass, H.T. Stokes, Acta Cryst. B 68, (2012) 215.

Semi-crystalline structures are prepared from different processing conditions. Bidirectional pre-oriented melt were crystallized at 375 K and atmospheric pressure for 10 nanoseconds (ns), to generate a lamellar semi-crystalline structure. Similar structures are also prepared from deformation of a cubic amorphous initial structure isothermally at 375 K. For comparison, two different thermostats, the constant stress (NPT) and constant volume (NVT) conditions are applied to the system during 10 ns of crystallization. The semi-crystalline samples shared common morphological features such as in the crystallinity, crystal orientation, lamellar thickness, and density distribution, etc. However, during the subsequent uniaxial tensile deformation simulations to a strain level of 0.5, different stress-strain behaviors are demonstrated. The microstructural evolutions of samples under crystallization and deformation are characterized by Hermans&’ orientation and two-point correlation statistics on the nanometer scale. By combining the observations of morphologies during deformation simulations and analysis of the stress-strain curves, conclusions are made that the effectiveness of the network has a strong influence on the mechanical property and strain hardening behavior. The oriented network from the constant stress crystallization, owing to the taut chains, give rise to optimal mechanical response with substantial strain-hardening. Cavitation is observed in the early stage of tensile test of constant volume crystallized sample, which reduces the performance and causes strain-softening. Insights are given into the microstructural-property relations to guide polymer architecture and processing design.

A computational algorithm has been developed to simulate the transport properties of oriented and un-oriented thin film nanocomposites of isotactic Polypropylene (iPP) and carbon nanotubes (CNT) with increasing CNT concentration. Our goal is to be able to design materials with optimal properties using simulations. We use cellular automata approach in Matlab simulation environment. The percolation threshold is reproduced in the simulations, matching experimental data. Upon percolation, the thermal transport in the films increases sharply, more so for the electrical than for the thermal conductivity, due to the larger difference in the electric conductivities of the CNTs and the polymer. To verify the simulation, the thin-film samples were sheared in the melt at 200 0C at 1 Hz in a Linkan microscope shearing hot stage. The thermal and electrical conductivity measurements were performed on the same cell arrangement with the transport perpendicular to the thin-film plane using a DC method. The thermal and electrical conductivity are higher for the un-sheared as compared to the sheared samples with stronger temperature dependence for the latter as compared to the former. Our cellular automata simulations provide information about the microstructure-macroscopic property relation in the thin film nanocomposites and can be extended to simulations of other important materials.

Background: The inherent compliance in soft material robotic systems enables capabilities and task versatility not found in traditional rigid-bodied robotic systems. Elastomeric actuators powered by pressurized fluid (i.e. pneumatics or hydraulics) offer several desirable features including robust, lightweight structures, inexpensive development, proven fabrication methods, and simple as well as complex motion paths with one or multiple inputs. Furthermore, these actuators are capable of providing compliance, fast actuation speeds and safe human interaction. In this study the design of a single integrated channel in a soft actuator is proposed. The asymmetrical structure creates articulation upon fluid pressurization where fiber reinforcements embedded in the structure of the actuator predefine the motion path.
Method: The actuators are fabricated following a multistage molding process, which offers complete control over every aspect of the assembled soft actuator including geometry, material properties, and fiber reinforcements. The fabrication process uses 3D printed molds to define the actuator shape and RTV silicone for the actuator body. These fiber reinforced actuators can operate at high pressures (550 kPa) and are highly compliant, able to fully bend with less than 1 N of force when un-actuated but exert significant higher forces when pressurized. Bending curvature and force response of these actuators are investigated using geometrical analysis and a finite element model (FEM). A dedicated experimental evaluation platform has been developed integrating pneumatic monitoring and control, bending angle measurement and force sensing to validate the various modeling approaches.
Results: A series of experiments that mechanically characterized the actuators have been carried out using the dedicated evaluation platform. The actuators are able to bend 360 degrees utilizing 250 kPa of air pressure. The actuators could also generate 3N of bending force measured at the tip with an input air pressure of 170 kPa. The experimental data is compared to results obtained from the analytical model and the FEM simulations showing good agreement.
Conclusions: This study provides a tool for analyzing and predicting the performance of bending fiber reinforcement soft elastomeric actuators. It enables complete control at every stage of development and contributes to new fundamental modeling tools for simulating fluid-interaction with soft robot designs that have complex geometry and non-linear material properties. This will be a valuable tool, not only in visualizing the behavior of these soft devices, but also in providing quantitative metrics regarding the interaction of the soft robot designs with the environment that can inform certain actuator capabilities with respect to a given set of design specifications.

Ligand functionalized nanoparticles are widely used in nanotechnology. The identity and polarity of surface ligands can influence properties such as cellular uptake of AuNPs or molecule binding. In this study, we use atomistic molecular dynamics to study interactions of AuNPs with single-stranded DNA (ssDNA). We use alkyl ligands with terminal groups of differing polarity, charge, and hydrophobicity to elucidate the effect of ligand identity with respect to folding dynamics of ssDNA. Our results indicate that the presence of AuNPs generally increases the folding rate of ssDNA. We observe the greatest effect with positively charged NPs because the backbone of DNA is negatively charged. The extent of ssDNA wrapping into the ligand corona is found to depend on the number of polar ligands in NP corona. A greater number of ligand polar terminal groups leads to ssDNA association with end-groups, while ssDNA prefers to associate with alkyl chains for weakly polar NPs. We also evaluate structural transitions of folded ssDNA caused by the presence of AuNPs. Our results show that NPs can be used to control the structure and dynamics of ssDNA through tailoring of NP ligands.

Successful insertion of microneedles into human skin is dependent on optimising a range of parameters such as the microneedle geometry, the polymer&’s constituent materials and insertion method; each of which can be a limiting factor in penetrating the skin. Therefore, microneedle design and analysis are necessary steps needed to successfully fabricate microneedle structures that can penetrate the skin without breaking.
This paper describes a finite element model used to simulate the insertion of a microneedle into human skin. A contact axisymmetrical model comprising a microneedle, and a two layer skin tissue was proposed. The model was validated against experimental data and showed good general agreement. The performance of various materials constituting a microneedle was determined. The model was used to investigate the effects of varying material properties as well as the microneedle&’s tip geometry on failure forces and insertion; and suggested that half cone-half cylinder microneedles would be an optimum design.

In this presentation, we will elucidate impact of ligand attachment on PbS QD in terms of absorption coefficient using ab initio density functional theory (DFT) calculation. Midgap states are generally considered as trap states which prevent to transport charges and also act as charge recombination centers in traditional view. They are only impediments to QD solar cell photo conversion efficiency. In such a point, researchers always attempt to passivate bare QD with ligands to prevent midgap states. In this work, however, we find that midgap states can play a positive role in optical absorption which is especially useful for solar cell application. According to analysis of solar spectrum, almost all the energy of photons which reach to ground is less than 4eV. If something can increase the absorption of light in 4eV or less, it would be good for solar cell efficiency. According to our findings, midgap states of QD act to absorb light that QDs will not be able to absorb if midgap states don&’t exist.
We compute DFT-RPA optical absorption of stoichiometric bare Pb32S32 QD. The DFT-RPA optical absorption can be expressed as an imaginary part of frequency dependent dielectric function, i.e. extinction coefficient ε2. We calculate absorption coefficient from it. We also calculate absorption coefficient of one iodine ligand attached Pb32S32 QD. Iodine is attached to bare QD because iodine is usually used atomic ligand from high efficient QD solar cell. Surprisingly, orders of magnitude enhancement in absorption coefficient is observed, in particular in the range 0-2.5eV which shows the peak intensity in solar spectrum. We calculate an iodine ligand absorption coefficient to verify that ligand itself make enhancement in the range of 0-2.5eV. From calculated results, an iodine ligand atom absorbs light having energy of 2.7eV or more. So iodine ligand absorption does not affect absorption enhancement of light from 0eV to 2.5eV. As we showed in our previous paper, Pb32S32 bare QD has no midgap state because of its stoichiometry. When we attach an iodine atom at stoichiometric bare QD, midgap states are appeared. We calculate oscillator strength in electronic band structure transitions to verify what cause light absorption enhancement from 0eV to 2.5eV. Calculated oscillator strengths show clearly that light absorption enhancements between 0eV and 2.5eV stem from midgap states. We believe that midgap states are reasons that cause absorption coefficient enhancement in IR and visible. Electrons which are filled in midgap states below Fermi level are excited by sun light. Midgap states can play a positive role in optical absorption although they influence negative effect on charge transport property.

Over the last decade, a growing percentage of the world&’s drinking water has been obtained from seawater and brackish waster involving desalination by some means. Moreover, increasing population and the predicted changes to the environment and climate make it likely that desalination will become even more widely used.[1] While there are a number of desalination techniques in use, reverse osmosis (RO) is the method that is currently the most popular. In RO saltwater is filtered through a membrane at high pressure; while the water is able to pass through the membrane the salt ions are not. Although there has been interest in using a number of materials as RO membranes, including zeolites[2] and carbon nanotubes[3,4] in practice most modern commercial RO membranes are made from polyamides (PAs).
Despite the importance of desalination technology, there remain many unanswered questions regarding the structure and behavior of PA membranes at the molecular level. Molecular dynamics (MD) simulations of PA membranes can help elucidate the behavior of these systems in the presence of water, salt ions and foulant molecules.[5,6] Insights obtained in this way may assist in the development of new, more efficient membranes with a better flow rate, higher salt rejection and/or greater resistance to fouling.
The interaction of a polyamide membrane with both water and salt ions has been investigated using MD. The rates of diffusion of different species within the PA membrane has been determined and umbrella sampling methods have been used to determine the free energy required for salt ions to enter the membrane. In addition, the interactions of glucose and phenol, which are two of the major constituent fragments of humic and alginic acids (two of the most important foulant species), with the RO membrane have been investigated. The results of the simulations reveal that while both molecules reduce the rate of flow of water molecule across the membrane-aqueous interface they interact with the membrane differently.
[1] Fritzmann , C., et al., Desalination, 2007, 216, 1-76.
[2] Hughes, Z.E. et al., J. Phys. Chem. C, 2011, 115, 4063-4075.
[3] Corry, B. Energy Environ. Sci., 2011, 4, 751-759.
[4] Hughes, Z.E. et al., J. Phys. Chem. C, 2012, 116, 24943-24953.
[5] Hughes, Z.E and Gale, J.D., J. Mater. Chem., 2010, 20, 7788-7799.
[6] Hughes, Z.E and Gale, J.D., J. Mater. Chem., 2012, 22, 175-184.

It has been seen from experimental studies[1] that polypropylene and polyethylene with a small fraction of -OH functional groups attached on the sides show a marked increase in the dielectric permittivity of the polymer. We have attempted to capture this increase using first principles computations on polyethylene chains containing -OH groups. Based on the effectiveness of the hydroxyl group, we can also expect other functional groups to cause a similar increase in the dielectric constant.
In this work, density functional theory (DFT) was used to study the effects of different functional groups, namely hydroxyl (-OH), amine (-NH2), ammonium chloride (-NH3Cl), nitro (-NO2), thiol (-SH) and carboxyl (-COOH) on the dielectric properties of polyethylene (PE). Each system was considered both in the presence and absence of trapped moisture in order to clearly isolate the effect of the water molecules. The dielectric constants were calculated using a novel single chain-based approach, in which effective medium theory is applied to obtain the polymer dielectric constant from that of the supercell in which the polymer is present.
The dielectric constant was indeed seen to increase for the PE-X system (where X is a particular functional group) as compared with the system containing just the PE chain. In the case of PE-OH, there was a further increase when interactions with trapped water molecules come into the picture. This can be explained by the hydrogen bonding between -OH groups and trapped water molecules in the system that would lead to an increased ionic contribution to the dielectric constant. The trends of improvement in dielectric permittivity with the different functional groups were mapped out and certain groups, like -OH and -NH2, were found to be more effective than the others. Furthermore, molecular dynamics simulations on the same systems were found to support our trends. Our finding, namely that functionalization of polyolefins using many of the groups presently studied leads to an increase in the dielectric constant values, can be exploited in improving the energy density of polymer based capacitors.
References
[1] X. Yuan, Y. Matsuyama and T.C. Chung, Macromolecules 43, 4011 (2010).
[2] C.C. Wang, G. Pilania, R. Ramprasad, Manish Agarwal, Mayank Misra et al, APL 102, 152901 (2013).
[3] C.C. Wang, G. Pilania, R. Ramprasad, Phys. Rev. B 87, 035103 (2013).
[4] S. Grimme, J. Comput. Chem. 27, 1787 (2006).

In numerous industrial applications of polymeric materials, modeling structure-property relationships requires equilibrated configurations of melts of long, entangled chains described with microscopic detail. We present a strategy for obtaining such configurations through sequential fine-graining of “snapshots” corresponding to a coarse-grained (CG) description based on a soft model.
Here for the microscopic description of polymer melts a generic model is employed, retaining, however, the two main features perplexing their equilibration: a) chain connectivity and b) hard excluded volume interactions. The former is described within FENE potentials while the latter are captured through repulsive Lennard Jones interactions. This microscopic representation is mapped on a CG model considering the polymers as chains of soft spheres with fluctuating size [1,2]. The spheres represent Gaussian density distributions of Nb segments of underlying microscopic sub-chains and the nonbonded interactions are obtained from the overlap of these density clouds.
The choice of Nb sets the resolution and we will discuss how the CG parameters can be obtained as a function of Nb from microscopic configurations with relatively short chains which can be still equilibrated by MD simulations. Microscopic configurations of melts with high polymerization degrees, N, are obtained through sequential fine-graining. Firstly a multi-step procedure is employed to back-map CG configurations equilibrated at crude resolution (e.g., Nb=100) on CG melts described with high detail (e.g., Nb =25). Microscopic details are then re-introduced using constraining pseudo-Hamiltonians and gradual activation of excluded volume interactions. In this way microscopic configurations of melts with polymerization degrees beyond the scope of current brute-force simulation techniques (e.g., N = 5000) can be obtained. Extensive evidence of equilibration will be provided by monitoring key structural and conformational properties. [1] T. Vettorel, G. Besold, K. Kremer, Soft Matter 2010, 6, 2282; [2] G. Zhang, K. Ch. Daoulas, K. Kremer, Macromol. Chem. Phys. 2013, 214, 214.

Water plays a central role in many fields of science ranging from biology to electrochemistry and corrosion. Molecular dynamics (MD) simulations are a valuable tool to study processes involving water at the atomic level. However, in order to perform such simulations a reliable representation of the potential-energy surface is required. Artificial neural networks (NNs) provide an unbiased way to construct accurate interatomic potentials based on electronic structure calculations for a wide range of systems [1,2]. The obtained NN potentials can then be evaluated several orders of magnitude faster than the underlying electronic structure data allowing to perform accurate MD simulations on extended length and time scales.
Previously, we have shown that very accurate NN potentials for water clusters can be constructed based on reference density-functional theory data [3-5]. Augmenting our potentials with van der Waals correction terms resulted in an improved description of these clusters with binding energies close to high-level quantum chemical calculations [5].
Here, we report a universal NN potential for bulk water, able to describe water under a wide range of conditions: from high-pressure ice polymorphs to the liquid phase. We demonstrate the capability of the NN potential to accurately reproduce results from ab initio MD simulations by a comparison of various structural and dynamical properties of water at different temperatures and pressures. Finally, we use the NN potential to study proton-transfer reactions in liquid water.
[1] J. Behler, Phys. Chem. Chem. Phys. 13, 17930 (2011).
[2] N. Artrith, T. Morawietz, and J. Behler, Phys. Rev. B 83, 153101 (2011).
[3] T. Morawietz, V. Sharma, and J. Behler, J. Chem. Phys. 136, 064103 (2012).
[4] T. Morawietz and J. Behler, Z. Phys. Chem., accepted (2013).
[5] T. Morawietz and J. Behler, J. Phys. Chem. A, DOI: 10.1021/jp401225b (2013).

The rearrangement of the polymer molecules within the film matrix during film drying, which defines its internal and surface structure, has an important role in the control release, optical and mechanical properties of the therapeutical oral films. Via a coarse-grained Molecular Dynamics-based simulation technique called Dissipative Particle Dynamics we investigate changes in the molecular conformational dynamics and morphology of polymer chains in films during the drying process. Dissipative Particle Dynamics is a multiscale mesoscopic simulation method that simultaneously captures the molecular details of the components through soft-sphere coarse-grained models and reproduces the hydrodynamic behavior of the system over extended time scales. We have developed a bead-spring model of individual polymer molecules where each bead or soft sphere is a coarse-grained representation of a group of atoms with two consecutively bonded beads along a polymer chain linked by a spring. The solvent is modeled by soft spheres which are coarse-grained representation of multiple solvent molecules. To capture the drying process, we simulate the polymer chain dynamics for different concentrations of the polymer, for a constant number of solvent and polymer soft spheres in the system. We characterize the polymer chain dynamics and conformation at the various stages of the drying process, or solvent concentration, via the correlation between measurements of the radius of gyration and end-to-end distance with the inter- and intra-polymer chain interactions. Obtained computational data are in good agreement with experimentally observed conformation changes in the polymeric matrix determined by means of Raman spectroscopy and diffusing wave spectroscopy. It is also confirmed that DPD methodology can provide a good prediction for the critical points of drying
process.

Flexible piezoelectric materials have drawn wide attention as a new type of materials for soft electronic devices which require self-generation of electricity and easiness to carry. By generating electricity with piezoelectric materials, low-power devices can operate without external power sources. As well as the generation of electricity, flexibility is one of the important factors for soft electronic devices. Some organic polymers such as polyvinylidene fluoride (PVDF) have piezoelectricity and flexibility. Flexible PVDF is an organic fluoropolymer composed of carbon, hydrogen, and fluorine with several phases. The beta phase of PVDF is dipolar, and the polymer chains are entangled to make 50-60% crystalline bulk PVDF. While piezoelectric oxides and organic polymers can be synthesized in thin film form, such materials actually have three dimensions on the atomic scale. By selectively controlling hydrogenation and fluorination of graphene, the inversion symmetry of a graphene layer can be broken, inducing the spontaneous polarization or piezoelectric properties. For the hydrogenated and fluorinated graphene (C₂HF), which hydrogen atoms are attached from the top of a graphene layer and fluorine atoms from the bottom of the layer in an ideal condition, four different C₂HF conformations are generated by attached atoms, which are referred to as chair, boat, zigzag, and armchair conformations. We examine the stability of these C₂HF conformations by calculating the formation energy of them, and confirm that the four conformations are energetically stable and the chair conformation is the most stable of them all. For the chair C₂HF layer, the spontaneous polarization along the direction perpendicular to the layer is computed as 47.3 pC/m in terms of the Berry phase calculations. Similar piezoelectric properties can be observed in chemically modified hexagonal boron nitride. The chair form of hydrogenated hexagonal boron nitride has polarization of 48.6 pC/m. All calculations have been performed within density-functional theory formalism.

The ferroelectric phases of poly(vinylidene fluoride) (PVDF) have applications in many fields including energy storage, energy generation, sensors, memory, and biomedical devices. For a many of these applications it is advantageous to use high quality thin films of PVDF (<1 micron) in the device. While there are different ways to deposit thin polymer films, spin coating is particularly attractive due to its affordability and its common use in industry. One of the major challenges of spin coating high quality ferroelectric PVDF is the porosity observed in the film. In this presentation we use in-situ 2-d light scattering observations to show that the porosity in the film can be understood from a thermodynamic phase separation due to the absorption of water (non-solvent) from the atmosphere into the film. We then continue to show that by monitoring the phase separation and appropriately raising the temperature of the film by using an in-situ rapid thermal annealing set-up we are able to make high quality thin film PVDF. This principle is quite different from post-spin annealing since we are briefly raising the temperature of the film to cause remixing instead of evaporating a large amount of solvent. We measure the transmission of optical light as a measure of films quality. Our films have transmissions of over 90% indicating very high quality films. We use Fourier Transform Infrared Spectroscopy to investigate the phase of the films. We find the films have 0% alpha phase (non-ferroelectric) crystallites and the film consists of amorphous polymer and ferroelectric crystallites only. We measure the electrical properties of the films and show that there is a significant polarization response.

The elastic deformation and failure of cellulose nanocrystals (CNCs) are studied using molecular dynamics simulation. Four molecular structures, single cellulose chains, single and double hydrogen-bonding planes and full CNCs, are investigated to predict and understand their mechanical properties. The simulation results show that the elastic modulus and ultimate tensile strength of the model CNCs are comparable to related experimental measurements. Further, both elastic properties and failure mechanisms are affected by intra- and inter-molecular hydrogen bonding. The inter-molecular hydrogen bonds enhance the elastic modulus of CNCs while the hydrogen bonds between adjacent molecules in the same hydrogen-bonding plane restrict the rotational motion of single molecules and decrease the ultimate strength.

Single-walled carbon nanotubes (SWCNTs) suspensions in aqueous solutions of sodium dodecyl sulfate (SDS) and saturated fatty acids (Cn) as co-dispersants are studied. The quality of the dispersions is analyzed through photoluminescence spectroscopy (PL) as a function of the Cn chain length. Nanocomposite powders prepared by adding SWNTs to ethanol solutions of Cn's followed by ultrasonication and evaporation of the solvent were also studied. Resonance Raman scattering (RRS) measurements and Molecular Dynamics (MD) simulations were performed to study the effect of the surrounding medium on SWCNTs properties in suspensions and in nanocomposite powders.
The populations of individualized SWCNTs are drastically decreased with the addition of short chain fatty acids to the SDS nanotube suspensions, as seen in PL and in RRS spectra. The suspensions approximately recover the original SDS-dispersed SWCNT populations with C12, the fatty acid that has the same number of carbon atoms than SDS. Upon addition of longer chain Cn's the PL peaks show increased intensities as compared to SDS suspensions, with the peaks corresponding to (9,4), (7,6) and (8,4) nanotubes being especially bright. MD simulations show a tendency of short Cn's to cluster together in water and compete for SDS, which is consistent with a decrease in individualized nanotubes concentration.
In nanocomposites, radial breathing mode (RBM) Raman frequency of SWCNTs bundles interacting with Cn's were measured. All the RBM values found exhibit a blue-shift in the frequency. Remarkably, nanotubes with diameters smaller than 1.0 nm show blue-shifts in the range 2.5 to 4.5 cm-1 while those having diameters larger than 1.0 nm have blue-shifts in the range 6.5 to 8.0 cm-1. MD simulations showed that large diameter nanotubes admit Cn's and ethanol molecules into their inner cavity while small diameter nanotubes can only accommodate a line of ethanol molecules along the tube axis, thus justifying the observed differences in the hardening of the RBM mode.

Novel organic dyes with N,N-dimethylaniline (DMA) moiety (electron donor) and cyanoacrylic acid moiety (electron acceptor) were designed for dye-sensitized solar cells (DSSCs). The molecular structure and electron distribution of the metal-free sensitizers were determined by density function theory (DFT) calculations with the B3LYP method and the 6-31+g(d) basis set. The optimized structures were classified as local minima on their respective potential energy surfaces according to the vibration frequencies. Electronic states in different environment were calculated by means of the Tomasi's Polarized Continum Model (PCM). The electron density of HOMO would be localized mainly on the DMA moiety and is extended along the π-bridge to the cyanoacrylic acid moiety. Under light illumination (LUMO), the intramolecular charge transfer would occur from electron donor to electron acceptor. We expect that the designed sensitizers would have light harvesting ability for DSSCs.

There is considerable interest in designing materials capable of self-assembling into ordered structures with characteristic dimensions in the 5 to 100 nm regime. Such materials are of interest in a wide variery of technologies, ranging from fabrication of integrated circuits to design of plasmonic devices and optical sensors. Developing models and methods with which to describe such systems presents a unique opportunity for computational materials design, but several important challenges must be overcome. On the one hand, it is essential to identify and preserve the key molecular features that lead to self assembly. On the other hand, the relevant length and time scales are too large to adopt a purely molecular representation in terms of atomistic models.
In this presentation, I will describe recently proposed theoretically informed coarse-grained (TICG) descriptions of self-assembling materials that adopt a molecular-level representation and enable description of relatively large domains. The implementation of such methods will be described in the context of two classes of materials. In the first, TICG models are proposed to describe the thermodynamics and dynamics of morphology formation in block polymers and their mixtures with solvents. The models are then used to design directed assembly processes for fabrication of circuit layouts, and their predictions are compared to experimental data for a variety of systems. In the second application, a TICG approach is developed to study the formation of ordered phases in liquid crystalline systems. The formalism is then used to design liquid crystalline droplets that can be used for biological sensing or for preparation of highly ordered, patchy particles. The predictions of the model are also compared to experimental results for nematic liquid crystals.

In typical diblock copolymers, two blocks of chemically dissimilar monomers A and B are covalently bonded together. Depending on the fraction of A monomers and the quantity chi;N, the A and B blocks microphase segregate to form various ordered morphologies, where chi; is a measure of how unfavorable the A and B interactions are and N is polymer length. Therefore, increasing N (such as to improve material properties) can simultaneously also affect the microphase segregated state. Of special interest here is that the double gyroid phase, which is desirable for certain applications because of its bicontinuous morphology, becomes less accessible at high chi;N.
This has sparked interest in tapered block copolymers, in which a tapered region separates the A block from the B block. This taper has a linear composition gradient from pure A to pure B (or from B to A for inverse-tapered systems). These systems allow a degree of independent control over N and the microphase segregated state by adjusting the ratio of the tapered region to the total polymer length. To show the details of how this strategy works, we map the phase diagrams of model tapered and inverse tapered polymers. We use self-consistent field theory (SCFT) and a multi-block model that approximates the tapered region with alternating A and B blocks of appropriate lengths. We find that the ordered phases shift to higher chi;N for tapered systems, and the shift increases as the taper length increases. Inverse tapers shift the phase diagram to even higher chi;N. Interestingly, the double gyroid region widens significantly for direct tapered systems versus the pure diblock—for large tapers the gyroid phase even becomes the dominant nonlamellar phase at high chi;N. We also implement fluid density functional theory (fDFT) of a similar model to reveal additional monomer scale details of the morphology. Preliminary results of the fDFT calculations will be discussed.

Fabricating arbitrary circuit-like features at the sub 10 nm length scale is the next frontier for semiconducting manufacturers. In order to achieve features at these length scales, new methods of lithography are necessary as the cost and feasibility of traditional photolithographic methods becomes insurmountable. Block copolymer directed self-assembly is the current forerunning method to try to achieve these length scale features. This method has allowed for the creation of periodic features, but creating arbitrary complex patterns with features such as bends, junctions, contact holes, and terminations relevant to semiconductor devices remains an active problem.
In this presentation, we use a random optimization process and self-consistent field theory (SCFT) simulation that finds the templates necessary for forming a target structure with arbitrary features and to explore the parameter space where such solutions exist. The model uses hard wall field boundary conditions surrounded by block-selective attractive boundary conditions to represent lithographically fabricated and chemically functionalized nanoscale posts used in experiment that direct the self-assembly of a thin film of a block copolymer. A block copolymer density field is created using traditional SCFT and used with an optimization randomization process algorithm to find the post configuration that minimizes the free energy of the target structure under the post constraints. A final traditional SCFT simulation is then performed to confirm that the post configuration found does indeed produce the target structure from a random initial field seeding.
To explore the parameter space where these inverse solutions are feasible, several target structures were chosen with various features of interest. The minority polymer volume fraction f, Flory-Huggins interaction parameter chi;, and number of posts were all varied to find the optimal conditions to produce a given target structure. The free energy of the solutions as a function of these parameters was compared and analyzed to understand the physics that allows the target structure to be feasible for some combinations of parameters and explain why no solution exists for other parameters. Additionally, these studies produce a phase diagram of where solutions are feasible under different conditions and thus can be a design tool in determining the parameters necessary to create a given target structure under whatever design constraints are imposed. These solutions are compared with results observed in experiments with a polydimethylsiloxane-b-polystyrene cylinder-forming diblock copolymer templated by posts made from an electron-beam resist to demonstrate the success of the method as a design tool for creating arbitrary structures. The method thus has great potential in being used in minimizing the experimentation necessary for designing complex arbitrary patterns.

Triblock copolymers are known to self-assemble via microphase segregation. While strongly segregating block copolymers were widely studied using self-consistent mean field theory approach, little is known about the microphase segregation and resulting chain conformations in moderately or weakly segregating systems. The identification of microdomains and chains conformations in molecular systems that do not segregate strongly poses a significant challenge in particle dynamics simulations. We have developed an approach to post-process particle simulation data in order to estimate the clusters formed by the endblock particles in a triblock copolymer system. We used this approach to investigate ABA copolymer melts using dissipative particle dynamics (DPD) simulations. The estimated bridging, looping and dangling end fractions in the triblock copolymer melts are discussed as a function of the copolymer composition (fA/B), copolymer chain length and the incompatibility between the two blocks (chi;AB). Furthermore, the molecular network formation is quantified as a function of the system temperature. Our results reveal the nature of copolymer chain conformation (looping/bridging or dangling ends) across a broad range of incompatibility strengths for the symmetric and asymmetric triblock copolymers.

In this presentation, we describe multi-scale modeling method combining PROLITH lithography simulation with Self-Consistent Field Theory (SCFT) computation of the block copolymer Directed Self-Assembly (DSA). Such simulations allow us to explore the feasibility of using patterned photoresist features as a graphoepitaxy template. Because the photoresist opening size and wall angles vary with exposure dose and focus offset, these template variations can create additional challenges for successful DSA. Within this method, we utilize PROLITH to predict the shape of a lithographic feature as function of process conditions. The results of that calculation are then used as input into SCFT simulation to predict the distribution of the matrix and etchable blocks of the DSA polymers (such as PS-b-PDMS or PS-b-PMMA) inside that feature. We applied this method to several simple cases (e.g., rectangular trench, cylindrical contact hole, and clover-leaf contact hole), and investigated the self-assembly of various polymers as function of their compositions. For the contact hole case, we study how changing litho focus and exposure impacts the feature shape and subsequently the resulting DSA. The new tool could therefore be applied to rapidly design and screen lithographic process conditions together with polymers used to shrink or rectify the features within the DSA technology.

In this talk, I will discuss our recent work in the context of simulations of ion conductivities of structured polymeric matrices. While much earlier work has focused on these issues in the context of glassy homopolymeric membranes, there have been only a few isolated studies on the issues specific to structured polymeric matrices, such as those arising in block copolymers and nanocomposite membranes. Modeling and understanding the mechanistic features underlying the transport of ions and penetrants through such heterogeneous matrices is of interest to a wide range of applications, including, water purification, polymer batteries and fuel cell membranes. In our work, we have used a combination of multiscale simulations and nonequilibrium statistical mechanics models to shed light on the parameters influencing such transport properties. In this talk, I will present results in the context of (i) Polymer nanocomposites; and (ii) Block copolymer electrolytes, and discuss the ramifications for the design of batteries and polymer separation membranes.

Conductivity in polymer electrolytes has long been regarded as occurring in the amorphous region, and being closely related to the mobility of the polymer. Several studies have demonstrated a contradictory view. The ionic conductivity can be greater in the ordered crystalline phase, in which a static, continuous conduction pathway is provided to promote ion transport. Similar to this idea, we found that chain-like ion aggregation also improves conductivity. This conduction mechanism involves a charge transfer between two chain ends (conduction sites): a cation hopping to one chain end and the cation at the other end hopping to a nearby site. This allows long range positive charge transfer while the cations only move locally. We use molecular dynamics simulation to systematically study these chain-like structures and how they impact the conductivity. The results suggest that the conduction mechanism depends on the length and lifetime of the chain aggregates. While a long chain reduces the overall number of conduction sites, a short chain prevents long range charge transfer. If the lifetime of an aggregate is shorter than the hopping time for cations, the hopping will not occur and the conductivity is limited to ions&’ self-diffusion. Our work introduces a different charge transport mechanism which does not depend on the motion of the PEO but rather on the arrangement of the ions. This suggests the possibility of enhancing charge transport by rational design of ion aggregates.

The adapted Su-Schrieffer-Heeger Hamiltonian is developed in this work to predict and design the optical bandgaps of thousands of different π-conjugated systems. The accuracy of our predicated optical bandgaps with respect to the corresponding experimental measurements exceeds the accuracy of the time-dependent density functional theory. Insights on the structure-property relationship of these π-conjugated systems lead to guiding rules for the photovoltaic materials design. New π-conjugated systems with designed optical bandgaps from far infrared to ultraviolet are suggested and discussed. The charge transfer mechanisms and the exciton and charge carrier mobilities are computed and compared for various bulk-heterojunction structures to improve the overall power convention efficiency.

Using a dispersion corrected density functional theory (DFT/B97D) approach, we have performed bulk solid-state calculations to investigate the influence of side-chain length on the molecular packing and optical properties of poly (9, 9-di-n-alkylfluorene-alt-benzothiadiazole) or FnBT where n is the number of CH2 units in the alkyl side-chains. Our results indicate that due to the strong intermolecular interactions between the side-chains for the structures with longer side-chains, the packing of these polymers forms a lamellar structure in their most stable configuration. For the FnBT with shorter side-chains, two nearly degenerate stable crystal structures with hexagonal symmetries are possible. These different packing structures can be attributed to the microphase separations between the flexible side-chains and the rigid backbones and are in agreement with previous investigations for other hairy-rod polymers. In addition, as a result of the efficient inter-chain interactions for the structures with longer side-chains, the dihedral angle between the F and BT units is reduced by about 30 degrees providing a more planar configuration for the backbone. As a consequence of a more planar backbone, the band gap in the lamellar structure is decreased by about 0.2 eV and 0.3 eV in comparison to the gas and the nearly hexagonal phases respectively. Time-dependent DFT (TD-DFT/B3LYP) was used to study the excited states of the monomer of FnBT with various lengths of side chains. It is found that the absorption spectra for the solid-state polymers with longer side-chains are red shifted relative to the solid -state polymers with shorter side-chains and the gas phase polymers.

Organic photovoltaic cells (OPVs) are one of the promising renewable energy resources because of their low production costs, mechanical flexibility, and light weight comparing with their inorganic counterparts. The photoactive layer of OPV - namely, the bulk heterojunction (BHJ) layer - comprises an interpenetrating network of electron donor and acceptor materials. Electron donor materials are usually semiconducting polymers such as P3HT or small molecules (CuPc), whereas electron acceptor materials are usually materials with high electron affinity such as fullerenes and their derivatives. The nanomorphologies of the BHJ layer are critical for the efficiency of OPVs and are very sensitive to the device fabrication conditions; therefore, comprehensive insights into the correlations between the BHJ nanomorphologies and BHJ film forming processes are important for optimization of device fabrication protocols. However, experimental characterization of the nanomorhologies of the BHJ layer is never trivial; therefore, computer simulations can effectively help fill the gap between device fabrication conditions and BHJ nanomorphologies. In this study, by constructing a multiscale molecular simulation framework based on coarse-grained molecular dynamics (CGMD) simulations, for the first time, we are able to reveal the nanomorphologies of BHJ OPVs with system sizes compatible with experiments (Lee, Pao, and Chen, Energy & Environmental Science 2013). We have constructed CG models for polymer-fullerene, polymer-inorganic nanocrystal hybrid, and small molecule BHJ OPVs, and simulated the nanomorphology evolution in the BHJ layer during thermal annealing, solvent evaporation, and vacuum co-deposition processes. Our results reveal that the multiscale molecular simulation framework we constructed is a powerful tool in studying the nanomorphologies of the OPVs under various fabrication conditions, potentially helpful for experimental teams to develop next-generation OPV devices with superior performances.