Molecular Dynamics Simulation Of Water Friction On Two-dimensional Nanosurface

Friction is one of the most common physical phenomena in life,which is closely related to our production and life.Inseparable from walking on the road,operating mechanical equipment,or playing musical instruments,friction is an essential factor.However,friction can also be harmful in many cases.Worldwide,the annual loss of primary energy due to friction reaches 30%.In addition,up to 70%of mechanical equipment damage is caused by various forms of abrasion.Therefore,it is of great practical importance to explore the internal mechanism of friction,to realize rational use of friction or reduce friction.When the friction process goes down to the nanoscale,its behavior is very different from that of the macroscopic case.Many studies indicate that factors such as temperature,humidity,and sliding direction all have important effects on nanoscale friction.The study of nano-friction is of great significance to the design of nano-devices such as nano-friction generators and nano-oscillators and to reduce energy loss.In addition,two-dimensional materials provide an excellent research object for nano-tribology due to their large specific surface area,lattice periodicity,and enhanced interface effects.With the development of high-performance computers,molecular dynamics simulation has become an important way to study nano-friction.In this paper,the water friction on a variety of graphene-like nanosurfaces was studied by using molecular dynamics simulation,and the correlation between the friction coefficient and the free energy of water molecules or probability parameters was investigated.The main findings and conclusions are as follows:(1)The frictional properties of water on six types of graphene-like nanomaterials:graphene(GRA),hexagonal boron nitride(h-BN),C3N,BC3,C2N,and C3N4 were investigated.The friction coefficients were calculated by the Green-Kubo formula,and the relationship between the friction coefficients of the six materials was as follows:GRA<h-BN<C3N<BC3<C2N<C3N4.Among the six models,the water/GRA has the smallest friction coefficient λ=(1.4 ± 0.1)×04 N s/m3,and the water/C3N4 has the largest friction coefficient λ=(11.08±0.28)×105 Ns/m3.Then,based on the probability distribution function of water molecules and referring to the definition of free energy,the calculation method of the maximum corrugation of free energy is improved.We found that the relationship between the friction coefficients of six nanomaterials and the maximum corrugation of free energy can be fitted with exponential expressions.The C2N and C3N4 have large friction coefficients and maximum corrugation of free energy,arising from the strong adsorption from nitrogenous pores.(2)To verify the validity of the free energy model in more systems,we use five raw material models:GRA,h-BN,C3N,BC3,and C2N.By scaling up the original charge,we get 24 material models whose charge were changed.The range of the friction coefficient extends to(1.4±0.1)× 104 N s/m3～(1.74±0.03)×106 N s/m3.We find that although the friction coefficient is positively correlated with the maximum corrugation of free energy in a larger range of friction coefficients,the friction coefficient is not a single exponential function of the maximum corrugation of free energy now.Next,we introduce a probability parameter P(d,t),where d represents the distance of water molecules from the 2D nano-surface,and t represents the movement time of water molecules in a water layer with thickness d.To study the correlation between λ and P(d,t),we calculate the Pearson correlation coefficient between the probability parameter P(d,t)and the friction coefficient λ for different d and t.We find that there are several high correlation regions between λ and P(d,t)in the phase space(d,t),where the Pearson correlation coefficient is close to±1.To explain the appearance of different correlation regions,we selected four high-correlation regions to analyze the water molecular motion patterns in each region.Our calculations indicate that the four highly correlated regions represent four types of motion modes of water molecules,and each motion mode can be expressed as a linear correlation function of λ.Therefore,P(d,t)can be used as a simple and effective parameter to accurately predict the friction at the liquid-nanostructure interface.In addition,we found that nanomaterials can be essentially divided into hydrophilic and hydrophobic materials based on the friction coefficient,and the boundary is between(1.35±0.08)×105 N s/m3 and(2.61±0.24)×105 N s/m3.This work reveals the molecule mechanism of friction between the nanosurface and water to a certain extent.The probability parameter P(d,t)introduced in this paper has an important enlightening reference value for the study of nanofriction in other systems.Since solid-liquid friction is an important constraint on fluid transport at the nanoscale,our work also has reference significance for nanofluidics-related applications,such as energy conversion,desalination,and water filtration to a certain extent.