Mesoscale Simulations of Structure Formation in Polyacrylonitrile Nascent Fibers Induced by Binary Solvent Mixture.

Polyacrylonitrile (PAN) is widely used as a raw material for the production of high-modulus carbon fibers, the internal structure of which is directly affected by the spinning of the precursor. Although PAN fibers have been studied for a long time, the formation of their internal structure has not been sufficiently investigated theoretically. This is due to the large number of stages in the process and the parameters controlling them. In this study, we present a mesoscale model describing the evolution of nascent PAN fibers during the coagulation. It is constructed within the framework of a mesoscale dynamic density functional theory. We use the model to study the influence of a combined solvent of dimethyl sulfoxide (DMSO, a good solvent) and water (a non-solvent) on the microstructure of the fibers. A porous structure of PAN is formed as a result of the microphase separation of the polymer and the residual combined solvent at a high water content in the system. The model shows that one of the possible ways to obtain the homogeneous fiber structure is to slow down the coagulation by increasing the amount of good solvent in the system. This result is in agreement with the existing experimental data and confirms the efficiency of the presented model.

Self-assembly in systems based on L-cysteine-silver-nitrate aqueous solution: multiscale computer simulation.

We use fully atomistic, quantum mechanics and mesoscopic simulations to investigate multiscale structure formation in a supramolecular system based on aqueous solutions of silver nitrate with L-cysteine (CSS). Fully atomistic modeling reveals that silver mercaptide clusters are formed in solution at the stage of aging, which has a pronounced "core-shell" structure. The core is formed due to the bonding of SAg groups of silver mercaptide (SM) zwitterions while the shell consists of NH3+ and C(O)O- groups. Self-assembly of large-scale aggregates in CSS occurs due to the interaction of SM functional groups located on the surface of the clusters, which allows them to be considered supramonomers. Quantum-mechanical calculations reveal additional insight into the intermolecular interaction of L-cysteine with the components of the system. The data on the structure and properties of supramonomers are used to develop and parameterize a mesoscopic CSS model supplemented with allowance for salt concentration. In the mesoscopic model, supramonomers are presented as "sticky spheres", the interaction between which is determined by short-range and screened Coulomb potentials. Depending on the salt concentration, all structural transitions typical of CSS are observed: the formation of a stabilized colloidal dispersion, the filamentary aggregates of a gel network, the formation of large-scale unbound aggregates, and precipitation. These stages qualitatively reproduce the experimentally observed behavior of a real solution.

Additive-induced ordered structures formed by PC71BM fullerene derivatives.

We report the results of an experimental and theoretical study of structure formation in mixtures of phenyl-C71-butyric acid methyl ester (PC71BM) with high boiling octane based solvent additives 1,8-octanedithiol (ODT), 1,8-dibromooctane, and 1,8-diiodooctane obtained by evaporation of a host-solvent (chlorobenzene). Experimental studies by DSC, SAXS and WAXS methods found evidence of crystallization of fullerenes in the presence of the high boiling additives in the mixtures. A molecular dynamics simulation of a PC71BM/ODT mixture revealed the self-assembly of fullerenes into sponge-like network structures.

Effect of Nanoparticles Surface Bonding and Aspect Ratio on Mechanical Properties of Highly Cross-Linked Epoxy Nanocomposites: Mesoscopic Simulations.

The paper aims to study the mechanical properties of epoxy resin filled with clay nanoparticles (NPs), depending on their shapes and content on the surface of a modifying agent capable of forming covalent bonds with a polymer. The cylindrical clay nanoparticles with equal volume and different aspects ratios (disks, barrel, and stick) are addressed. The NPs' bonding ratio with the polymer (RGC) is determined by the fraction of reactive groups and conversion time and varies from RGC = 0 (non-bonded nanoparticles) to RGC = 0.65 (more than half of the surface groups are linked with the polymer matrix). The performed simulations show the so-called load-bearing chains (LBCs) of chemically cross-linked monomers and modified nanoparticles to determine the mechanical properties of the simulated composites. The introduction of nanoparticles leads to the breaking of such chains, and the chemical cross-linking of NPs with the polymer matrix restores the LBCs and strengthens the composite. At small values of RGC, the largest value of the elastic modulus is found for systems filled with nanoparticles having the smallest surface area, and at high values of RGC, on the contrary, the systems containing disk-shaped particles with the largest surface area have a larger elastic modulus than the others. All calculations are performed within the framework of a mesoscopic model based on accurate mapping of the atomistic structures of the polymer matrix and nanoparticles into coarse-grained representations, which, if necessary, allow reverse data mapping and quantitative assessment of the state of the filled epoxy resin. On the other hand, the obtained data can be used to design the functional materials with specified mechanical properties based on other practically significant polymer matrices and nanofillers.