Juan Manuel Castillo

On the Mechanism Behind the Instability of Isoreticular Metal-Organic Frameworks (IRMOFs) in Humid Environments

Increasing the resistance to humid environments is mandatory for the implementation of isoreticular metal-organic frameworks (IRMOFs) in industry. To date, the causes behind the sensitivity of [Zn(4)(μ(4)-O)(μ-bdc)(3)](8)(IRMOF-1; bdc=1,4-benzenedicarboxylate) to water remain still open. A multiscale scheme that combines Monte Carlo simulations, density functional theory and first-principles Born-Oppenheimer molecular dynamics on IRMOF-1 was employed to unravel the underlying atomistic mechanism responsible for lattice disruption. At very low water contents, H(2)O molecules are isolated in the lattice but provoke a dynamic opening of the terephthalic acid, and the lattice collapse occurs at about 6% water weight at room temperature. The ability of Zn to form fivefold coordination spheres and the increasing basicity of water when forming clusters are responsible for the displacement of the organic linker. The present results pave the way for synthetic challenges with new target linkers that might provide more robust IRMOF structures.

Characterization of Alkylsilane Self-Assembled Monolayers by Molecular Simulation

Self-assembled monolayers (SAM) of dodecyltrichlorosilane (DTS) and octadecyltrichlorosilane (OTS) on silica are studied by molecular dynamics simulations at 298 K and 1 bar. The coverage (number of alkylsilane molecules per surface area) is systematically varied. The results yield insight into the properties of the alkylsilane SAMs, which complement experimental studies from the literature. Relationships are reported between thickness, tilt angle, and coverage of alkylsilane SAMs, which also hold for alkylsilanes other than DTS and OTS. They are interpreted based on the information on molecular ordering in the SAMs taken form the simulation data. System size and simulation time are much larger than in most former simulation works on the topic. This reduces the influence of the initial configuration as well as the periodic boundary conditions and hence minimizes the risk of artificial ordering. At the same time, more reliable statistics for the calculated properties can be provided. The evaluation of experimental data in the field is often based on strongly simplified models. The present simulation results suggest that some of these lead to errors, concerning the interpretation of experimental results, which could be avoided by introducing more realistic models.

Acceleration of Monte-Carlo molecular simulations on hybrid computing architectures

Markov-Chain Monte-Carlo (MCMC) methods are an important class of simulation techniques, which execute a sequence of simulation steps, where each new step depends on the previous ones. Due to this fundamental dependency, MCMC methods are inherently hard to parallelize on any architecture. The upcoming generations of hybrid CPU/GPGPU architectures with their multi-core CPUs and tightly coupled many-core GPGPUs provide new acceleration opportunities especially for MCMC methods, if the new degrees of freedom are exploited correctly. In this paper, the outcomes of an interdisciplinary collaboration are presented, which focused on the parallel mapping of a MCMC molecular simulation from thermodynamics to hybrid CPU/GPGPU computing systems. While the mapping is designed for upcoming hybrid architectures, the implementation of this approach on an NVIDIA Tesla system already leads to a substantial speedup of more than 87× despite the additional communication overheads.

Influence of bidisperse self-assembled monolayer structure on the slip boundary condition of thin polymer films

Alkylsilane self-assembled monolayers (SAMs) are often used as model substrates for their ease of preparation and hydrophobic properties. We have observed that these atomically smooth monolayers also provide a slip boundary condition for dewetting films composed of unentangled polymers. This slip length, an indirect measure of the friction between a given liquid and different solids, is switchable and can be increased [R. Fetzer et al., Phys. Rev. Lett. 95, 127801 (2005); O. Bäumchen et al., J. Phys.: Condens. Matter 24, 325102 (2012)] if the alkyl chain length is changed from 18 to 12 backbone carbons, for example. Typically, this change in boundary condition is affected in a quantized way, using one or the other alkyl chain length, thus obtaining one or the other slip length. Here, we present results in which this SAM structure is changed in a continuous way. We prepare bidisperse mixed SAMs of alkyl silanes, with the composition as a control parameter. We find that all the mixed SAMs investigated show an enhanced slip boundary condition as compared to the single-component SAMs. The slip boundary condition is accessed using optical and atomic force microscopy, and we describe these observations in the context of X-ray reflectivity measurements. The slip length, varying over nearly two orders of magnitude, of identical polymer melts on chemically similar SAMs is found to correlate with the density of exposed alkyl chains. Our results demonstrate the importance of a well characterized solid/liquid pair, down to the angstrom level, when discussing the friction between a liquid and a solid.

Molecular Modelling and Simulation of Electrolyte Solutions, Biomolecules, and Wetting of Component Surfaces

Massively-parallel molecular dynamics simulation is applied to systems containing electrolytes, vapour-liquid interfaces, and biomolecules in contact with water-oil interfaces. Novel molecular models of alkali halide salts are presented and employed for the simulation of electrolytes in aqueous solution. The enzymatically catalysed hydroxylation of oleic acid is investigated by molecular dynamics simulation taking the internal degrees of freedom of the macromolecules into account. Thereby, Ewald summation methods are used to compute the long range electrostatic interactions. In systems with a phase boundary, the dispersive interaction, which is modelled by the Lennard-Jones potential here, has a more significant long range contribution than in homogeneous systems. This effect is accounted for by implementing the Janecek cutoff correction scheme. On this basis, the HPC infrastructure at the Steinbuch Centre for Computing was accessed and efficiently used, yielding new insights on the molecular systems under consideration.