Three parallel algorithms for classical molecular dynamics are presented. The first assigns each processor a fixed subset of atoms; the second assigns each a fixed subset of inter-atomic forces to compute; the third assigns each a fixed spatial region. The algorithms are suitable for molecular dynamics models which can be difficult to parallelize efficiently—those with short-range forces where the neighbors of each atom change rapidly. They can be implemented on any distributed-memory parallel machine which allows for message-passing of data between independently executing processors. The algorithms are tested on a standard Lennard-Jones benchmark problem for system sizes ranging from 500 to 100,000,000 atoms on several parallel supercomputers--the nCUBE 2, Intel iPSC/860 and Paragon, and Cray T3D. Comparing the results to the fastest reported vectorized Cray Y-MP and C90 algorithm shows that the current generation of parallel machines is competitive with conventional vector supercomputers even for small problems. For large problems, the spatial algorithm achieves parallel efficiencies of 90% and a 1840-node Intel Paragon performs up to 165 faster than a single Cray C9O processor. Trade-offs between the three algorithms and guidelines for adapting them to more complex molecular dynamics simulations are also discussed.

Fast parallel algorithms for short-range molecular dynamics

Understanding and controlling the flow of water confined in nanopores has tremendous implications in theoretical studies and industrial applications. Here, we propose a simple model for the confined water flow based on the concept of effective slip, which is a linear sum of true slip, depending on a contact angle, and apparent slip, caused by a spatial variation of the confined water viscosity as a function of wettability as well as the nanopore dimension. Results from this model show that the flow capacity of confined water is 10 −1 ∼10 7 times that calculated by the no-slip Hagen–Poiseuille equation for nanopores with various contact angles and dimensions, in agreement with the majority of 53 different study cases from the literature. This work further sheds light on a controversy over an increase or decrease in flow capacity from molecular dynamics simulations and experiments.

https://www.pnas.org/content/pnas/114/13/3358.full.pdf

Wettability effect on nanoconfined water flow

We report high magnetic field scanning tunneling microscopy and Landau level spectroscopy of twisted graphene layers grown by chemical vapor deposition. For twist angles exceeding ~3° the low energy carriers exhibit Landau level spectra characteristic of massless Dirac fermions. Above 20° the layers effectively decouple and the electronic properties are indistinguishable from those in single-layer graphene, while for smaller angles we observe a slowdown of the carrier velocity which is strongly angle dependent. At the smallest angles the spectra are dominated by twist-induced van Hove singularities and the Dirac fermions eventually become localized. An unexpected electron-hole asymmetry is observed which is substantially larger than the asymmetry in either single or untwisted bilayer graphene.

Single-Layer Behavior and Its Breakdown in Twisted Graphene Layers

Hydrodynamic and hydromagnetic stability

Motivated by experiments in which a polynucleotide is driven through a proteinaceous pore by an electric field, we study the diffusive motion of a polymer threaded through a narrow channel with which it may have strong interactions. We show that there is a range of polymer lengths in which the system is approximately translationally invariant, and we develop a coarse-grained description of this regime. From this description, general features of the distribution of times for the polymer to pass through the pore may be deduced. We also introduce a more microscopic model. This model provides a physically reasonable scenario in which, as in experiments, the polymer's speed depends sensitively on its chemical composition, and even on its orientation in the channel. Finally, we point out that the experimental distribution of times for the polymer to pass through the pore is much broader than expected from simple estimates, and speculate on why this might be.

Driven Polymer Translocation Through a Narrow Pore

Fluid biomarkers extracted from many types of body fluids provide significant information that serve as indicators of the underlying physiological and pathological conditions of the human body. Analysis of multiple fluid biomarkers could help improve the early identification and progression of comorbid conditions to enhance the diagnostic accuracy, which can help in developing patient-specific treatment plans. In this work, an attempt has been made to differentiate the co-occurrence of diabetes, hypertension and cardiovascular disease (comorbid conditions) from non-comorbid using multiple fluid biomarkers. Fluid biomarkers are obtained from a public dataset under comorbid ([Formula: see text]) and non-comorbid ([Formula: see text]) conditions. Five features, such as serum creatinine, serum sodium, platelet count, creatine phosphokinase and ejection fraction, are extracted for further analysis. Machine learning algorithms namely, [Formula: see text]-nearest neighbor and linear discriminant analysis (LDA) are used to classify comorbid and non-comorbid conditions. The results show an increase in platelet count in comorbid subjects. This feature also exhibits significant difference ([Formula: see text]) between both the conditions. This study also uses the random undersampling technique to reduce bias associated with data imbalance. LDA classifier yields a maximum accuracy of 54.30% in classifying these two conditions. Further study can be carried out to improve the accuracy and might be helpful in clinical practice for prediction of comorbid conditions. 

A framework for the analysis of comorbid conditions using intelligent extraction of multiple fluid biomarkers

Liquid structuring of confined organic coolant in nanochannels. A comparative study with water

A new method is proposed to model the thermal boundary resistance (TBR) at the nanoscale, solid-liquid interface from macroscopic observables that characterize a nanoscale interface. We correlated the TBR with thermodynamic state variables, material properties, and geometric parameters to derive a generalized relationship with the help of data-driven heuristic algorithms. The results show that TBR can be expressed in terms of physical observables of the systems and material-specific parameters. We investigated the mutual independence of descriptor variables and quantified the weightage for each observable parameter in the TBR models. The interfacial liquid layering has a robust correlation with TBR. However, for systems with phonon size effects and under extreme thermodynamic conditions, the work of adhesion and system geometry also affects the variation in TBR. The data-driven approach followed in this study helps us gain better insight into the mechanism of TBR at nanoscale solid-liquid interfaces and shows significant improvement in our knowledge about interfacial thermal transport.

A data driven approach to model thermal boundary resistance from molecular dynamics simulations.

Flow-induced electrohydrodynamics in a nanochannel with finite temperature gradient

The anisotropic heat transport in graphene-CNT based materials provoked the development of three-dimensional pillared-graphene (PG) systems. In this study, we performed non-equilibrium molecular dynamics simulations to analyze PG thermal conductivity and thermal boundary conductance. For the first time, we have considered the influence of pillar chirality and the temperature effect on PG heat transport. We analyzed the influence of pillar chirality and pillar length on the in- and out-of-plane transport properties. For the temperature-dependent analysis, the chosen temperatures were in the range of 100 K to 500 K. To elucidate the mechanism underlying the heat transport, we investigated the phonon density of states (DOS) in the different regions of PG systems. The overlap factor was calculated to quantify the mismatch in the phonon DOS profiles. Across the pillar region, the overlap factor correlates directly with the thermal boundary conductance. When heat is transported in an out-of-plane direction, the zig-zag PG system performs better than the armchair PG system. The atomic arrangement at the graphene-CNT interface plays an inevitable role in limiting heat transport in PG systems. The calculated phonon energy in the zig-zag PG interface is higher than that in the armchair PG interface.

Sarith P. Sathian

Papers

A framework for the analysis of comorbid conditions using intelligent extraction of multiple fluid biomarkers

Liquid structuring of confined organic coolant in nanochannels. A comparative study with water

A data driven approach to model thermal boundary resistance from molecular dynamics simulations.

Flow-induced electrohydrodynamics in a nanochannel with finite temperature gradient

Modeling the effect of chirality on thermal transport in a pillared-graphene structure.