Nikolaos Asproulis

A hybrid molecular continuum method using point wise coupling

Over the past decade, advances in micro and nanofluidics, have influenced a range of areas spanning from chemistry to semiconductor design. The phenomena observed at micro- and nano-scales are characterised by their inherent multiscale nature. Accurate numerical modelling of these phenomena is the cornerstone to enhance the applicability of micro and nanofluidics in the industrial environment. In this paper a novel multiscale approach, in the hybrid continuum-molecular framework, is presented. In this approach molecular models are employed as refinement for calculating data required by the continuum solver. The method has been applied to a number of test cases including Couette flows with slip boundary conditions, Couette flows with roughness and Poiseuille flows of polymeric fluids.

Multi-scale Computational Modelling of Flow and Heat Transfer

Purpose – The purpose of this paper is to present different approaches for applying macroscopic boundary conditions in hybrid multiscale modelling.Design/methodology/approach – Molecular dynamics (MD) was employed for the microscopic simulations. The continuum boundary conditions were applied either through rescaling of atomistic velocities or resampling based on velocity distribution functions.Findings – The methods have been tested for various fluid flows with heat transfer scenarios. The selection of the most suitable method is not a trivial task and depends on a number of factors such as accuracy requirements and availability of computational resource.Originality/value – The applicability of the methods has been assessed for liquid and gas flows. Specific parameters that affect their accuracy and efficiency have been identified. The effects of these parameters on the accuracy and efficiency of the simulations are investigated. The study provides knowledge regarding the development and application of b...

Mesoscale flow and heat transfer modelling and its application to liquid and gas flows

Advances in micro and nanofluidics have influenced technological developments in several areas, including materials, chemistry, electronics and bio-medicine. The phenomena observed at micro and nanoscale are characterised by their inherent multiscale nature. Accurate numerical modelling of these phenomena is the cornerstone for enhancing the applicability of micro and nanofluidics in the industrial environment. We investigated different strategies for applying macroscopic boundary conditions to microscopic simulations. Continuous rescaling of atomic velocities and velocity distribution functions, such as Maxwell-Boltzmann or Chapman-Enskog, were investigated. Simulations were performed for problems involving liquids and gases under different velocity and temperature conditions. The results revealed that the selection of the most suitable method is not a trivial issue and depends on the nature of the problem, availability of computational resource and accuracy requirement.

Molecular dynamics modelling of mechanical properties of polymers for adaptive aerospace structures

The features of adaptive structures depend on the properties of the supporting materials. For example, morphing wing structures require wing skin materials, such as rubbers that can withstand the forces imposed by the internal mechanism while maintaining the required aerodynamic properties of the aircraft. In this study, Molecular Dynamics and Minimization simulations are being used to establish well-equilibrated models of Ethylene-Propylene-Diene Monomer (EPDM) elastomer systems and investigate their mechanical properties.

HPC Parallelisation of Boundary Conditions in Multiscale Methods

This paper investigates two numerical implementations of continuum boundary conditions in parallel high performance computing (HPC) systems in conjunction with multiscale modelling comprising molecular dynamics (MD) and computational fluid dynamics (CFD) methods. The multiscale method provides the best compromise in terms of accuracy and computational cost in mesoscale regimes, however, there are still algorithmic challenges preventing the practical application of these methods. The present study investigates some of these challenges, namely different domain decompositions of the momentum transferred from the continuum domain to the atomistic region in conjunction with HPC parallelisation.

Thermal behaviour of nanofluids confined in nanochannels

This work investigates the effect of spatial restriction on the thermal properties of nanofluids. Using Molecular Dynamics simulations, a Copper-Argon nanofluid is restricted within idealized walls. The thermal conductivity of the suspension is calculated using the Green-Kubo relations and is correlated with the volume fraction of the copper particles within the system as well as the channel width. The thermal conductivity is further broken down into its individual components in the three dimensions, revealing anisotropy between the directions parallel and normal to the channel walls. The observed thermodynamic patterns are justified by considering how the spatial restriction affects the liquid structure around the nanoparticle.

Surrogate Models for Aerodynamic Shape Optimisation

The main challenges in full-scale aerospace systems development are related to the level of our understanding with respect to the systems behaviour. Computational modelling, through high-fidelity simulations, provides a viable approach towards efficient implementation of the design specifications and enhancing our understanding of the system’s response. Although high-fidelity modelling provides valuable information the associated computational cost restricts its applicability to full-scaled systems. This chapter presents a Computational Fluid Dynamics optimisation strategy based on surrogate modelling for obtaining high-fidelity predictions of aerodynamic forces and aerodynamic efficiency. An Aerodynamic Shape Optimisation problem is formulated and solved using Genetic Algorithm with surrogate models in the place of actual computational fluid dynamics algorithms. Ordinary Kriging approach and Hammersley Sequence Sampling plan are used to construct the surrogate models.

Characterization of CO2 Flow Through Charged Carbon Nanotubes

The equilibrium transport of CO2 through infinitely long charged and uncharged single-walled carbon nanotubes has been examined by employing molecular dynamics simulations. It has been shown that the molecular transport concludes into a Fickian diffusion for all the examined cases. Ballistic and single-file diffusion mechanisms have been met especially in cases where the degree of loading is low since at higher pressures the electrostatic interactions derived by the charge of the nanotubes is canceled by the stronger SWNT-CO2 interactions.

Carbon Dioxide Transport in Carbon Nanopores

In the present study atomistic simulations are employed for investigating the CO2 transport properties through pores that are of slit, cylindrical and scroll configuration. We aim to investigate any possible differences in the adsorption process among the three different geometries focusing on the adsorption type that occurs along with pore filling and pore emptying mechanisms. Carbon-slit pores are of widths between 0.8 and 2.0 nm. The simulated Single-Walled Carbon Nanotubes (SWNTs) are of 1.08 and 2.17 nm ((8,8) and (16,16) respectively) while for the Carbon Nanoscrolls (CNSs) the inner diameter corresponds to a (6,6) SWNT with an intralayer distance of 0.4 to 1.0nm.

Comparison of High‐order Finite Volume and Discontinuous Galerkin Methods on 3D Unstructured Grids

The paper presents a direct comparison of convergence properties of finite volume and discontinuous Galerkin methods of the same nominal order of accuracy. Convergence is evaluated on tetrahedral grids for an advection equation and manufactured solution of Euler equations. It is shown that for the test cases considered, the discontinuous Galerkin discretisation tends to recover the asymptotic range of convergence on coarser grids and yields a lower error norm by comparison with the finite volume discretisation.