Self-assembly of amphiphilic hyperbranched multiarm copolymers (HMCs) has shown great potential for preparing all kinds of delicate supramolecular structures in all scales and dimensions in solution. However, theoretical studies on the influencing factors for the self-assembly of HMCs have been greatly lagging behind. The phase diagram of HMCs in selective solvents is very necessary but has not been disclosed up to now. Here, the self-assembly of HMCs with different hydrophilic fractions in various solvents was studied systematically by using dissipative particle dynamics (DPD) simulations. Three morphological phase diagrams are constructed and a rich variety of morphologies, ranging from spherical micelles, worm-like micelles, membranes, vesicles, vesosomes, small micellar aggregates (SMAs), and aggregates of spherical and worm-like micelles to helical micelles, are obtained. In addition, both the self-assembly mechanisms and the dynamic processes for the formation of these self-assemblies have been systematically investigated. The simulation results are consistent with available experimental observations. Besides, several novel structures, like aggregates of spherical and worm-like micelles, vesosomes and helical micelles, are firstly discovered for HMC self-assembly. We believe the current work will extend the knowledge on the self-assembly of HMCs, especially on the control of supramolecular structures and on fabricating novel self-assemblies.

We propose a facile inverse design strategy to generate three-dimensional (3D) nanopatterns by using either block copolymers or a binary homopolymer blend via dissipative particle dynamics simulations. We find that the composition window of block copolymers to form a specific 3D morphology can be expanded when the self-assembly of block copolymers is directed by templates. We also find that a binary homopolymer blend can serve as a better candidate in the inverse templating design, since they have similar performances on recovering the target pattern, with much lower cost. This strategy is proved efficient for fabricating templates with desired topographical configuration, and the inverse design idea sheds lights on better control and design of materials with complex nanopatterns.

By controlling the feed ratio of CMS/styrene and the polymerization time, a series of hyperbranched copolystyrenes (HBCPS) were synthesized with comparable weight-averaged molecular weights (Mw) but different degree of branching (DB) through atom transfer radical self-condensing vinyl copolymerization (ATR-SCVCP) with CuBr/2,2′-bipyridyl as the catalyst. The resulting HBCPS samples were used to investigate the effect of branching architecture on their glass transition behavior. With the DB increased, the glass transition temperatures (Tg) of HBCPS samples measured by DMA and DSC both decreased. Their spin-lattice relaxation times (1H T1r) of protons displayed the same downtrend with increasing DB. Besides, a correlation between the Tgs and the DB was well established by all-atom molecular dynamics (MD) simulations. The values of MD-determined Tgs are little higher than the corresponding experimental ones. However, the dependence of Tgs on DB is in good agreement with the experimental results, i.e., Tg decreases both in experiments and simulations with increasing DB.

Hyperbranched multiarm copolymers (HMCs) have shown great potential to be excellent precursors in self-assembly to form various supramolecular structures in all scales and dimensions in solution. However, theoretical studies on the self-assembly of HMCs, especially the self-assembly dynamics and mechanisms, have been greatly lagging behind the experimental progress. Herein, we investigate the effect of degree of branching (DB) on the self-assembly structures of HMCs by dissipative particle dynamics (DPD) simulation. Our simulation results demonstrate that the self-assembly morphologies of HMCs can be changed from spherical micelles, wormlike micelles, to vesicles with the increase of DBs, which are qualitatively consistent with the experimental observations. In addition, both the self-assembly mechanisms and the dynamic processes for the formation of these three aggregates have been systematically disclosed through the simulations. These self-assembly details are difficult to be shown by experiments and are very useful to fully understand the self-assembly behaviors of HMCs.

Developing a simple and efficient method to organize nanoscale building blocks into ordered superstructures, understanding the mechanism for self-assembly and revealing the essential collective properties are crucial steps toward the practical use of nanostructures in nanotechnology-based applications. In this study, we showed that the high-yield formation of ZnO nanoparticle chains with micrometer length can be readily achieved by the variation of solvents from methanol to water. Spectroscopic studies confirmed the solvent effect on the surface properties of ZnO nanoparticles, which were found to be critical for the formation of anisotropic assemblies. Quantum mechanical calculations and all atom molecular dynamic simulations indicated the contribution of hydrogen bonding for stabilizing the structure in water. Dissipative particle dynamics further revealed the importance of solvent–nanoparticle interactions for promoting one-dimensional self-assembly. The branching of chains was found upon aging, resulting in the size increase of the ensembles and network formation. Steady-state and time-resolved luminescent spectroscopes, which probed the variation of defect-related emission, revealed stronger Forster resonance energy transfer (FRET) between nanoparticles when the chain networks were formed. The high efficiency of FRET quenching can be ascribed to the presence of multiple energy transfer channels, as well as the short internanoparticle distances and the dipole alignment.Keywords: anisotropic structure; chains; hydrogen bonding; molecular dynamics; nanoparticles; one-dimensional; self-assembly; ZnO;

In this paper, we present the coil-to-globule (CG) transitions of homopolymers and multiblock copolymers with different topology and stiffness by using molecular dynamics with integrated tempering sampling method. The sampling method was a novel enhanced method that efficiently sampled the energy space with low computational costs. The method proved to be efficient and precise to study the structural transitions of polymer chains with complex topological constraint, which may not be easily done by using conventional Monte Carlo method. The topological constraint affects the globule shape of the polymer chain, thus further influencing the CG transition. We found that increasing the topological constraint generally decreased CG transition temperature for homopolymers. For semiflexible chains, an additional first-order like symmetry-broken transition emerged. For block copolymers, the topological constraint did not obviously change the transition temperature, but greatly reduced the energy signal of the CG transition.

Although an analogy has been drawn between them, organic molecular amphiphiles (MAMs) and inorganic nanoparticle (NP) amphiphiles (NPAMs) are significantly different in dimension, geometry, and composition as well as their assembly behavior. Their concurrent assembly can synergetically combine the inherent properties of both building blocks, thus leading to new hybrid materials with increasing complexity and functionality. Here we present a new strategy to fabricate hybrid vesicles with well-defined shape, morphology, and surface pattern by coassembling MAMs of block copolymers (BCPs) and NPAMs comprising inorganic NPs tethered with amphiphilic BCPs. The assembly of binary mixtures generated unique hybrid Janus-like vesicles with different shapes, patchy vesicles, and heterogeneous vesicles. Our experimental and computational studies indicate that the different nanostructures arise from the delicate interplay between the dimension mismatch of the two types of amphiphiles, the entanglement of polymer chains, and the mobility of NPAMs. In addition, the entropic attraction between NPAMs plays a dominant role in controlling the lateral phase separation of the two types of amphiphiles in the membranes. The ability to utilize multiple distinct amphiphiles to construct discrete assemblies represents a promising step in the self-assembly of structurally complex functional materials.

The self-assembly processes of miktoarm star-like block copolymers (Ax1)y-C-(Bx2)y, in which y homopolymer chains of type A with chain length x1 as well as y chains of type B with chain length x2 are connected to a center core C, are investigated by using Brownian dynamics simulations. We focus on the selective solvent condition, i.e., the solvent is poor for components B and C, but varies from good to poor for component A. The miktoarm star-like block copolymers with A and B arms of equal length can form spherical micelles when the solvent is good for component A. In the same solvent conditions, the micelles contain fewer miktoarm star-like block copolymers as the arm length increases. When the solvent is poor for component A, the miktoarm star-like block copolymers can self-assemble into cylindrical or disk-like micelles with decreasing solvent quality. For the miktoarm star-like block copolymers with longer B arms but shorter A arms, self-assembly into spherical vesicles is possible in a wide range of solvent conditions.

Herein, we report a novel Janus particle and supramolecular block copolymer consisting of two chemically distinct hyperbranched polymers, which is coined as Janus hyperbranched polymer. It is constructed by the noncovalent coupling between a hydrophobic hyperbranched poly(3-ethyl-3-oxetanemethanol) with an apex of an azobenzene (AZO) group and a hydrophilic hyperbranched polyglycerol with an apex of a β-cyclodextrin (CD) group through the specific AZO/CD host–guest interactions. Such an amphiphilic supramolecular polymer resembles a tree together with its root very well in the architecture and can further self-assemble into unilamellar bilayer vesicles with narrow size distribution, which disassembles reversibly under the irradiation of UV light due to the trans-to-cis isomerization of the AZO groups. In addition, the obtained vesicles could further aggregate into colloidal crystal-like close-packed arrays under freeze-drying conditions. The dynamics and mechanism for the self-assembly of vesicles as well as the bilayer structure have been disclosed by a dissipative particle dynamics simulation.

A coarse-grained model, with three sets of effective pair potentials for 1-butyl-3-methylimidazolium hexafluorophosphate ([Bmim][PF6]) ionic liquid, is introduced and used to study the structural and dynamical properties over extended length and time scales. Three sets of effective pair potentials between coarse-grained beads are obtained using the Newton Inversion and the Iterative Boltzmann Inversion methods, respectively, with different treatment of electrostatic interactions. The coarse-grained simulation results are compared systematically with corresponding atomistic simulation results on several thermodynamical and structural quantities together with charge density distributions. In addition, the scattering and dynamical properties are also calculated and compared to both atomistic simulation results and experimental measurements. While all three sets of the effective potentials provide perfect agreement with the atomistic simulation for radial distribution functions, our analysis shows that in coarse-grained simulations, the long-range electrostatic interactions between ionic groups are important and should be treated explicitly to improve the reliability of other simulation results. With explicit incorporation of electrostatic interactions derived from the Newton Inversion, the coarse-grained potentials provide the most consistent description of thermodynamical, scattering and dynamical properties including their temperature dependence as compared to atomistic simulations. We conclude also that the current atomistic force field should be further improved to meet specific requirements for studying the dynamical properties of the [Bmim][PF6] system over a large temperature range.

Microscopic structures, orientational preferences together with mass, number and electron density distributions of 1-butyl-3-methylimidazolium hexafluorophosphate ([BMIM][PF6]) ionic liquid (IL) have been studied on a neutral hydrophobic graphene surface, and at the IL–vacuum interface using atomistic Molecular Dynamics simulations. At the IL–graphene interface, distinct mass, number and electron density distributions are observed oscillating into the bulk region with several compact structural layers. The imidazolium ring of [BMIM] cations lies preferentially flat on the graphene surface, with its methyl and butyl side chains elongated along the graphene surface. At the IL–vacuum interface, however, the distributions of [BMIM][PF6] ion pairs are strongly influenced by the thickness of IL film. With the increase of IL film thickness, the orientations of [BMIM] cations at the IL–vacuum interface change gradually from dominant flat distributions along the graphene surface to orientations where the imidazolium rings are either parallel or perpendicular to the IL–vacuum interface with tilted angles. The outmost layers are populated with alkyl groups and imparted with distinct hydrophobic character. The calculated radial distribution functions suggest that ionic structures of [BMIM][PF6] ion pairs in IL–graphene and IL–vacuum interfacial regions are significantly different from each other and also from that in bulk regions.

Molecular dynamics simulations are useful tools to unveil molecular mechanisms of polymer phase separation, self-assembly, adsorption, and so on. Due to large molecular size and slow relaxation of the polymer chains, a great amount of issues related to large-distance chain displacement cannot be tackled easily with conventional molecular dynamic simulations. Systematic coarse-graining and enhanced sampling methods are two types of improvements that can boost spatiotemporal scales in polymer simulations. We present two typical ways to obtain the coarse-graining potential either by fitting to correct liquid structures or by fitting to available thermodynamic properties of polymer systems. The newly proposed anisotropic coarse-grained particle model can be used to describe aggregation and assembly of polymeric building blocks from disk-like micelles to Janus particles. We also present a stochastic polymerization model combined with coarse-grained simulations to investigate the problems strongly influenced by the coupling of polymerization and excluded volume effects. Finally, a facile implementation of integrated tempering sampling method is illustrated to be very efficient on bypassing local energy minima and having access to true equilibrium polymer structures.

Comprehensive dissipative particle dynamics simulations are performed to investigate the effect of A-block polydispersity on the phase behavior of AB-diblock copolymers. The experimental results of Lynd and Hillmyer on polydispersity induced domain spacing expansion at different segregations (Macromolecules 2005, 38, 8803) are well reproduced and explained in terms of interfacial free energy and molecular stretching in domain center. Short block copolymers in polydisperse system are found to preferably accumulate at A/B interface as compatibilizers causing a reduction in interfacial free energy. At the same time, domain centers are found mainly occupied by long blocks which in turn determining the spacing of the corresponding domain. More importantly, increase in polydispersity will introduce more long blocks into system and therefore will enhance the molecular packing and stretching in the domain center. Such chain stretching in domain center and reduction in interfacial free energy are found to be more severe at stronger segregation, which will lead to more severe domain spacing expansion with increasing polydispersity.

Coarse-grained models for β-cyclodextrin (β-CD) and adamantane (ADA) are proposed by fitting to their experimental host–guest complex equilibrium constant in solution. By using Brownian dynamics simulations, we suggest a simple supramolecular route for synthesizing multiblock copolymers (MBCs) via forming complexes between β-CD and ADA groups terminated at the chain ends of diblock copolymers (DBCs). The chain length distribution of the resulted MBC is found to follow the statistics of Flory formula for typical linear condensation polymerization process. Therefore, the proposed supramolecular route can be viewed as a novel linear condensation polymerization process with DBCs as reactive monomers. Due to the complex formations between head and tail (β-CD and ADA), ring-shaped MBCs are also observed in our simulations, which will reduce the yield of the MBC. Because we are using a generic model for DBC, the proposed route of building MBCs are applicable for all synthetic DBCs with two ends terminated by either β-CD or ADA groups.

The phase behavior of polydisperse ABA triblock copolymers is studied using dissipative particle dynamics simulations, focusing on the emergence and property of bicontinuous structures. Bicontinuous structures are characterized by two separate, intermeshed nanoscopic domains extending throughout the material. The connectivity of polymeric bicontinuous structures makes them highly desirable for many applications. For conventional monodisperse diblock and triblock copolymers, regular bicontinuous structures (i.e., gyroid and Fddd) can be formed over a narrow composition window of 3%. We demonstrate that the composition window for the formation of bicontinuous structures can be regulated by designed polydispersity distributions of ABA triblock copolymers. In particular, introducing polydispersity in both A and B blocks can lead to a significant enhancement of the composition window of bicontinuous structures with both continuous A and B domains. The mechanism of the bicontinuous structure enhancement is elucidated from the distribution of the long and short blocks. Furthermore, it is shown that the polymeric bicontinuous structures from polydisperse ABA triblock copolymers possess good continuity throughout the sample, making them ideal candidates for advanced applications.

We study living polymerization initiated from concave surfaces. We clarify that, depending on different criteria for ceasing the reaction, different relationships between grafted chain polydispersity index and the grafting surface curvature can be categorized. The average molecular weight of the grafted chains monotonically decreases as the grafting surface curvature increases. These results shed light on better control and design of functional porous materials for use in bioimplanting or chemical sensors.

We study the self-assembly of symmetric star-like block copolymers (Ax)y(Bx)yC in dilute solution by using Brownian dynamics simulations. In the star-like block copolymer, incompatible A and B components are both solvophobic, and connected to the center bead C of the polymer. Therefore, this star-like block copolymer can be taken as a representative of soft and deformable Janus particles. In our Brownian dynamics simulations, these “soft Janus particles” are found to self-assemble into worm-like lamellar structures, loose aggregates and so on. By systematically varying solvent conditions and temperature, we build up the phase diagram to illustrate the effects of polymer structure and temperature on the aggregate structures. At lower temperatures, we can observe large worm-like lamellar aggregates. Upon increasing the temperature, some block copolymers detach from the aggregate; this phenomenon is especially sensitive for the polymers with less arms. The aggregate structure will be quite disordered when the temperature is high. The incompatibility between the two parts in the star-like block copolymer also affects the self-assembled structures. We find that the worm-like structure is longer and narrower as the incompatibility between the two parts is stronger.

Dissipative particle dynamics simulations are used to study the specific binding structures of polyamidoamine (PAMAM) dendrimers on amphiphilic membranes and the permeation mechanisms. Mutually consistent coarse-grained (CG) models both for PAMAM dendrimers and for dimyristoylphosphatidylcholine (DMPC) lipid molecules are constructed. The PAMAM CG model describes correctly the conformational behavior of the dendrimers, and the DMPC CG model can properly give the surface tension of the amphiphilic membrane. A series of systematic simulations is performed to investigate the binding structures of the dendrimers on membranes with varied length of the hydrophobic tails of amphiphiles. The permeability of dendrimers across membranes is enhanced upon increasing the dendrimer size (generation). The length of the hydrophobic tails of amphiphiles in turn affects the dendrimer conformation, as well as the binding structure of the dendrimer–membrane complexes. The negative curvature of the membrane formed in the dendrimer–membrane complexes is related to dendrimer concentration. Higher dendrimer concentration together with increased dendrimer generation is observed to enhance the permeability of dendrimers across the amphiphilic membranes.

The reorganization energy and the charge transfer integral between the initial and final states are two key parameters of charge transport. In this study, we find that the internal reorganization energies of molecules can be reduced effectively by cyanation on the tetracene molecule, which is helpful to improve the carrier mobilities of investigated molecules. On the basis of the polymorph predictor, the appreciated crystal structures of 5-cyanotetracene (1CT), 5,11-dicyanotetracene (2CT), and 5,6,11,12-tetracyanotetracene (4CT) with π–π stacking favorable to enhance the charge transfer integral were obtained. Benefiting from a low internal reorganization energy and dense packing structure, high hole motilities (6.0 and 6.2 cm2 V–1 s–1 for 1CT-9 and 4CT-2, respectively) were achieved. Both 1CT-9 and 4CT-2 are promising candidates for high carrier mobility organic optoelectronic functional materials. The process that constructs crystal structures from single molecules provides a rational way for the search of high-performance organic photovoltaic materials.

Atomistic molecular dynamics simulations have are used to investigate the liquid crystal systems based on [4-pentyl-(1-cyclohexenyl)]-(4-cyanophenyl)diazene (5CPDCN) and 4-cyano-4′-pentylazobenzene (5AZCN). The results show the growth process of a nematic phase from a disordered phase. Then the phase transition caused by isomerization reaction is studied based on a temporary modification of the dihedral potential. The properties of 5AZCN and 5CPDCN are compared, showing that the orientation of trans-5CPDCN is more highly ordered than trans-5AZCN. This can be attributed to the more extended dihedral angles φ2 (i.e. the dihedral angle between the ring system and the terminal chain) in trans-5CPDCN enhance the rod-like conformation of the molecules. The orientational correlation functions gl(r) (l = 1, 2) are also calculated, by which we find that both 5CPDCN and 5AZCN systems in nematic phase present parallel and anti-parallel dipole correlations. The anti-parallel dipole correlation is localized for the 5CPDCN system; on the contrary, the parallel dipole correlation is weakly localized for the 5AZCN system.

Vesicles are membrane-enclosed capsules that can store or transport substances. Their structures and the corresponding structural transitions are important to fulfill specific functions. Using dissipative particle dynamics method, we study the complex structure transitions of vesicles that are spontaneously formed by A6(B2)3 type comb-like block copolymers. In the simulations, the interaction parameters between different components are tuned to mimic the variations of amphiphilicity of the block copolymers and the selectivity of the solvent which are experimentally tractable by, for example, a temperature quench. Complex vesicle structures are found in this research; their transitions, such as fission and reversal, are studied in detail with this dynamic simulation method. We find that the line tension plays a decisive role on the vesicle fission pathways. Moreover, the tube-like vesicles tend to transform to a special layered micelle structure, whereas the onion-shape vesicles tend to transform to reverse onion-shape vesicles when vesicle reversal takes place due to the variation of solvent selectivity.

A novel mesoscopic simulation model is proposed to study the liquid crystal phase behavior of the anisotropic rodlike particles with a soft repulsive interaction, which possesses a modified anisotropic conservative force type used in dissipative particle dynamics. The influences of the repulsion strength and the particle shape on the phase behavior of soft rodlike particles are examined. In the simulations, we observe the formation of the nematic phase and smectic-A phase from the initially isotropic phase. Moreover, we find that shorter soft rodlike particles with anisotropic repulsive interactions can form a stable smectic-B phase. Our results demonstrate that the soft anisotropic purely-repulsive potential between the rodlike particles can reflect the interaction nature between soft rodlike particles in a simple way and is sufficient to produce a range of ordered LC-like mesophases.

We develop a mesoscale nonequilibrium simulation model to study the effect of steady shear on the hierarchical self-assembly of soft disklike particles in dilute solutions. By properly tuning shear rates and solvent conditions, soft disklike particles can self-assemble into flexible threads and bundle-like structures along the flow direction. Shear flow facilitates the self-assembly of soft disklike particles into one-dimensional long threads along the flow direction; however, it suppresses the formation of flexible bundles from the threads while decreasing the solvent quality. The relatively well-defined bundle structures along the flow direction can only be obtained when the solvent condition becomes even worse. Our study elucidates how the solvent condition and shear rate can be utilized to control the shear-induced self-assembled structures, which would enable designed nanofabrication.

We introduce a reaction model for use in coarse-grained simulations to study the chemical reactions in polymer systems at mesoscopic level. In this model, we employ an idea of reaction probability in control of the whole process of chemical reactions. This model has been successfully applied to the studies of surface initiated polymerization process and the network structure formation of typical epoxy resin systems. It can be further modified to study different kinds of chemical reactions at mesoscopic scale.

We develop a novel mesoscale simulation model in order to study the hierarchical self-assembly of soft disklike particles in dilute solutions. In suitable solvent conditions, the soft anisotropic disklike particles first self-assemble into one-dimensional flexible threads, in accord with experiments. Then, intriguingly, the threads reversibly pack into flexible hexagonal bundles by decreasing the solvent quality. Hierarchical self-assembly of this type may be important to provide a strategy to create bundle structures by bottom-up self-assembly with a single type of soft and flexible building block and mimic the bundles commonly found in biological structures.

The surface-initiated polymerization with different initiator densities and different polymerization rates is investigated using molecular dynamics simulation method. We find that the initiator density, together with the polymerization rate, greatly determines the polymer brush structure, the initiation efficiency, and the graft density, especially when the initiator density is high. The excluded volume effect also plays a crucial role in the system when the chains are densely grafted. By tuning the initiator density and modifying the polymerization rate, we can obtain the polymer brushes with different degrees of polydispersity. This study partially emphasizes the importance of considering the effects of polymerization rate in further investigations.

Amphiphilic plasmonic micelle-like nanoparticles (APMNs) composed of gold nanoparticles (AuNPs) and amphiphilic block copolymers (BCPs) structurally resemble polymer micelles with well-defined architectures and chemistry. The APMNs can be potentially considered as a prototype for modeling a higher-level self-assembly of micelles. The understanding of such secondary self-assembly is of particular importance for the bottom-up design of new hierarchical nanostructures. This article describes the self-assembly, modeling, and applications of APMN assemblies in selective solvents. In a mixture of water/tetrahydrofuran, APMNs assembled into various superstructures, including unimolecular micelles, clusters with controlled number of APMNs, and vesicles, depending on the lengths of polymer tethers and the sizes of AuNP cores. The delicate interplay of entropy and enthalpy contributions to the overall free energy associated with the assembly process, which is strongly dependent on the spherical architecture of APMNs, yields an assembly diagram that is different from the assembly of linear BCPs. Our experimental and computational studies suggested that the morphologies of assemblies were largely determined by the deformability of the effective nanoparticles (that is, nanoparticles together with tethered chains as a whole). The assemblies of APMNs resulted in strong absorption in near-infrared range due to the remarkable plasmonic coupling of Au cores, thus facilitating their biomedical applications in bioimaging and photothermal therapy of cancer.

A coarse-grained model, with three sets of effective pair potentials for 1-butyl-3-methylimidazolium hexafluorophosphate ([Bmim][PF6]) ionic liquid, is introduced and used to study the structural and dynamical properties over extended length and time scales. Three sets of effective pair potentials between coarse-grained beads are obtained using the Newton Inversion and the Iterative Boltzmann Inversion methods, respectively, with different treatment of electrostatic interactions. The coarse-grained simulation results are compared systematically with corresponding atomistic simulation results on several thermodynamical and structural quantities together with charge density distributions. In addition, the scattering and dynamical properties are also calculated and compared to both atomistic simulation results and experimental measurements. While all three sets of the effective potentials provide perfect agreement with the atomistic simulation for radial distribution functions, our analysis shows that in coarse-grained simulations, the long-range electrostatic interactions between ionic groups are important and should be treated explicitly to improve the reliability of other simulation results. With explicit incorporation of electrostatic interactions derived from the Newton Inversion, the coarse-grained potentials provide the most consistent description of thermodynamical, scattering and dynamical properties including their temperature dependence as compared to atomistic simulations. We conclude also that the current atomistic force field should be further improved to meet specific requirements for studying the dynamical properties of the [Bmim][PF6] system over a large temperature range.

Self-assembly processes play a key role in the fabrication of functional nano-structures with widespread application in drug delivery and micro-reactors. In addition to the thermodynamics, the kinetics of the self-assembled nano-structures also play an important role in determining the formed structures. However, as the self-assembly process is often highly heterogeneous, systematic elucidation of the dominant kinetic pathways of self-assembly is challenging. Here, based on mass flow, we developed a new method for the construction of kinetic network models and applied it to identify the dominant kinetic pathways for the self-assembly of star-like block copolymers. We found that the dominant pathways are controlled by two competing kinetic parameters: the encounter time Te, characterizing the frequency of collision and the transition time Tt for the aggregate morphology change from rod to sphere. Interestingly, two distinct self-assembly mechanisms, diffusion of an individual copolymer into the aggregate core and membrane closure, both appear at different stages (with different values of Tt) of a single self-assembly process. In particular, the diffusion mechanism dominates the middle-sized semi-vesicle formation stage (with large Tt), while the membrane closure mechanism dominates the large-sized vesicle formation stage (with small Tt). Through the rational design of the hydrophibicity of the copolymer, we successfully tuned the transition time Tt and altered the dominant self-assembly pathways.

Dissipative particle dynamics simulations are used to study the specific binding structures of polyamidoamine (PAMAM) dendrimers on amphiphilic membranes and the permeation mechanisms. Mutually consistent coarse-grained (CG) models both for PAMAM dendrimers and for dimyristoylphosphatidylcholine (DMPC) lipid molecules are constructed. The PAMAM CG model describes correctly the conformational behavior of the dendrimers, and the DMPC CG model can properly give the surface tension of the amphiphilic membrane. A series of systematic simulations is performed to investigate the binding structures of the dendrimers on membranes with varied length of the hydrophobic tails of amphiphiles. The permeability of dendrimers across membranes is enhanced upon increasing the dendrimer size (generation). The length of the hydrophobic tails of amphiphiles in turn affects the dendrimer conformation, as well as the binding structure of the dendrimer–membrane complexes. The negative curvature of the membrane formed in the dendrimer–membrane complexes is related to dendrimer concentration. Higher dendrimer concentration together with increased dendrimer generation is observed to enhance the permeability of dendrimers across the amphiphilic membranes.

We study the self-assembly of symmetric star-like block copolymers (Ax)y(Bx)yC in dilute solution by using Brownian dynamics simulations. In the star-like block copolymer, incompatible A and B components are both solvophobic, and connected to the center bead C of the polymer. Therefore, this star-like block copolymer can be taken as a representative of soft and deformable Janus particles. In our Brownian dynamics simulations, these “soft Janus particles” are found to self-assemble into worm-like lamellar structures, loose aggregates and so on. By systematically varying solvent conditions and temperature, we build up the phase diagram to illustrate the effects of polymer structure and temperature on the aggregate structures. At lower temperatures, we can observe large worm-like lamellar aggregates. Upon increasing the temperature, some block copolymers detach from the aggregate; this phenomenon is especially sensitive for the polymers with less arms. The aggregate structure will be quite disordered when the temperature is high. The incompatibility between the two parts in the star-like block copolymer also affects the self-assembled structures. We find that the worm-like structure is longer and narrower as the incompatibility between the two parts is stronger.

Atomistic molecular dynamics simulations have are used to investigate the liquid crystal systems based on [4-pentyl-(1-cyclohexenyl)]-(4-cyanophenyl)diazene (5CPDCN) and 4-cyano-4′-pentylazobenzene (5AZCN). The results show the growth process of a nematic phase from a disordered phase. Then the phase transition caused by isomerization reaction is studied based on a temporary modification of the dihedral potential. The properties of 5AZCN and 5CPDCN are compared, showing that the orientation of trans-5CPDCN is more highly ordered than trans-5AZCN. This can be attributed to the more extended dihedral angles φ2 (i.e. the dihedral angle between the ring system and the terminal chain) in trans-5CPDCN enhance the rod-like conformation of the molecules. The orientational correlation functions gl(r) (l = 1, 2) are also calculated, by which we find that both 5CPDCN and 5AZCN systems in nematic phase present parallel and anti-parallel dipole correlations. The anti-parallel dipole correlation is localized for the 5CPDCN system; on the contrary, the parallel dipole correlation is weakly localized for the 5AZCN system.

Microscopic structures, orientational preferences together with mass, number and electron density distributions of 1-butyl-3-methylimidazolium hexafluorophosphate ([BMIM][PF6]) ionic liquid (IL) have been studied on a neutral hydrophobic graphene surface, and at the IL–vacuum interface using atomistic Molecular Dynamics simulations. At the IL–graphene interface, distinct mass, number and electron density distributions are observed oscillating into the bulk region with several compact structural layers. The imidazolium ring of [BMIM] cations lies preferentially flat on the graphene surface, with its methyl and butyl side chains elongated along the graphene surface. At the IL–vacuum interface, however, the distributions of [BMIM][PF6] ion pairs are strongly influenced by the thickness of IL film. With the increase of IL film thickness, the orientations of [BMIM] cations at the IL–vacuum interface change gradually from dominant flat distributions along the graphene surface to orientations where the imidazolium rings are either parallel or perpendicular to the IL–vacuum interface with tilted angles. The outmost layers are populated with alkyl groups and imparted with distinct hydrophobic character. The calculated radial distribution functions suggest that ionic structures of [BMIM][PF6] ion pairs in IL–graphene and IL–vacuum interfacial regions are significantly different from each other and also from that in bulk regions.