Searcharxiv⌕ Search

arXiv subjects

Walter Kob

Publications and source records attributed to Walter Kob.

At least 127 records · Page 7Linked to original sources

Molecular Dynamics Simulations

A tutorial introduction to the technique of Molecular Dynamics (MD) is given, and some characteristic examples of applications are described. The purpose and scope of these simulations and the relation to other simulation methods is discussed, and the basic MD algorithms are described. The sampling of intensive variables (temperature T, pressure p) in runs carried out in the microcanonical (NVE) ensemble (N= particle number, V = volume, E = energy) is discussed, as well as the realization of other ensembles (e.g. the NVT ensemble). For a typical application example, molten SiO2, the estimation of various transport coefficients (self-diffusion constants, viscosity, thermal conductivity) is discussed. As an example of Non-Equilibrium Molecular Dynamics (NEMD), a study of a glass-forming polymer melt under shear is mentioned.

cond-mat.dis-nn↗

Structure and diffusion in amorphous aluminium silicate: A molecular dynamics computer simulation

The amorphous aluminium silicate (Al2O3)2(SiO2) [AS2] is investigated by means of large scale molecular dynamics computer simulations. We consider fully equilibrated melts in the temperature range 6100K >= T >= 2300K as well as glass configurations that were obtained from cooling runs from T=2300K to 300K with a cooling rate of about 10^12K/s. Already at temperatures as high as 4000K, most of the Al and Si atoms are four-fold coordinated by oxygen atoms. Thus, the structure of AS2 is that of a disordered tetrahedral network. The packing of AlO4 tetrahedra is very different from that of SiO4 tetrahedra in that Al is involved with a relatively high probability in small-membered rings and in triclusters in which an O atom is surrounded by four cations. We find as typical configurations two-membered rings with two Al atoms in which the shared O atoms form a tricluster. On larger length scales, the system shows a microphase separation in which the Al-rich network structure percolates through the SiO2 network. The latter structure gives rise to a prepeak in the static structure factor at a wavenumber q=0.5Å^{-1}. The comparison of experimental X-ray data with the results from the simulation shows a good agreement for the structure function. The diffusion dynamics in AS2 is found to be much faster than in SiO2. We show that the self-diffusion constants for O and Al are very similar and that they are by a factor of 2-3 larger than the one for Si.

cond-mat.stat-mech↗

Statics and dynamics of the ten-state nearest-neighbor Potts glass on the simple-cubic lattice

We present the results of Monte Carlo simulations of two different Potts glass models with short range random interactions. In the first model a \pm J-distribution of the bonds is chosen, in the second model a Gaussian distribution. In both cases the first two moments of the distribution are chosen to be J_0=-1, Delta J=+1, so that no ferromagnetic ordering of the Potts spins can occur. We find that for all temperatures investigated the spin glass susceptibility remains finite, the spin glass order parameter remains zero, and that the specific heat has only a smooth Schottky-like peak. These results can be understood quantitatively by considering small but independent clusters of spins. Hence we have evidence that there is no static phase transition at any nonzero temperature. Consistent with these findings, only very minor size effects are observed, which implies that all correlation lengths of the models remain very short. We also compute for both models the time auto-correlation function C(t) of the Potts spins. While in the Gaussian model C(t) shows a smooth uniform decay, the correlator for the \pm J model has several distinct steps. These steps correspond to the breaking of bonds in small clusters of ferromagnetically coupled spins (dimers, trimers, etc.). The relaxation times follow simple Arrhenius laws, with activation energies that are readily interpreted within the cluster picture, giving evidence that the system does not have a dynamic transition at a finite temperature. Hence we find that for the present models all the transitions known for the mean-field version of the model are completely wiped out. Finally we also determine the time auto-correlation functions of individual spins, and show that the system is dynamically very heterogeneous.

cond-mat.stat-mech↗

The vibrational dynamics of vitreous silica: Classical force fields vs. first-principles

We compare the vibrational properties of model SiO_2 glasses generated by molecular-dynamics simulations using the effective force field of van Beest et al. (BKS) with those obtained when the BKS structure is relaxed using an ab initio calculation in the framework of the density functional theory. We find that this relaxation significantly improves the agreement of the density of states with the experimental result. For frequencies between 14 and 26 THz the nature of the vibrational modes as determined from the BKS model is very different from the one from the ab initio calculation, showing that the interpretation of the vibrational spectra in terms of calculations using effective potentials can be very misleading.

cond-mat.stat-mech↗

The structural relaxation of molten sodium disilicate

We use molecular dynamics computer simulations to study the relaxation dynamics of Na2O-2(SiO2) in its molten, highly viscous state. We find that at low temperatures the incoherent intermediate scattering function for Na relaxes about 100 times faster than the one of the Si and O atoms. In contrast to this all coherent functions relax on the same time scale if the wave-vector is around 1AA^-1. This anomalous relaxation dynamics is traced back to the channel-like structure for the Na atoms that have been found for this system. We find that the relaxation dynamics for Si and O as well as the time dependence of the coherent functions for Na can be rationalized well by means of mode-coupling theory. In particular we show that the diffusion constants as well as the alpha-relaxation times follow the power-law predicted by the theory and that in the beta-relaxation regime the correlators obey the factorization property with a master curve that is described well by a von Schweidler-law. The value of the von Schweidler exponent $b$ is compatible with the one found for the mentioned power-law of the relaxation times/diffusion constants. Finally we study the wave-vector dependence of f_s(q) and f(q), the coherent and incoherent non-ergodicity parameters. For the Si and O atoms these functions look qualitatively similar to the ones found in simple liquids or pure silica, in that the coherent function oscillates (in phase with the static structure factor) around the incoherent one and in that the latter is approximated well by a Gaussian function. In contrast to this, f(q) for Na-Na is always smaller than f_s(q) for Na and the latter can be approximated by a Gaussian only for relatively large q.

cond-mat.stat-mech↗

Cooperative motion and growing length scales in supercooled confined liquids

Using molecular dynamics simulations we investigate the relaxation dynamics of a supercooled liquid close to a rough as well as close to a smooth wall. For the former situation the relaxation times increase strongly with decreasing distance from the wall whereas in the second case they strongly decrease. We use this dependence to extract various dynamical length scales and show that they grow with decreasing temperature. By calculating the frequency dependent average susceptibility of such confined systems we show that the experimental interpretation of such data is very difficult.

cond-mat.stat-mech↗

Simulation of models for the glass transition: Is there progress?

The glass transition of supercooled fluids is a particular challenge for computer simulation, because the (longest) relaxation times increase by about 15 decades upon approaching the transition temperature T_g. Brute-force molecular dynamics simulations, as presented here for molten SiO_2 and coarse-grained bead-spring models of polymer chains, can yield very useful insight about the first few decades of this slowing down. Hence this allows to access the temperature range around T_c of the so-called mode coupling theory, whereas the dynamics around the experimental glass transition is completely out of reach. While methods such as ``parallel tempering'' improve the situation somewhat, a method that allows to span a significant part of the region T_g\leq T\leq T_c is still lacking. Only for abstract models such as the infinite range 10-state Potts glass with a few hundred spins this region can be explored. However this model suffers from very strong finite size effects thus making it difficult to extrapolate the results obtained for the finite system sizes to the thermodynamic limit. For the case of polymer melts, two different strategies to use lattice models instead of continuum models are discussed: In the first approach, a mapping of an atomistically realistic model of polyethylene to the bond fluctuation model with suitable effective potentials and a temperature-dependent time rescaling factor is attempted. In the second approach, devoted to a test of the entropy theory, moves that are artificial but which lead to a faster relaxation (``slithering snake'' algorithm) are used, to get at least static properties at somewhat lower temperatures than possible with a ``realistic'' dynamics. The merits and shortcomings of all these approaches are discussed.

cond-mat.stat-mech↗

Evidence against a glass transition in the 10-state short range Potts glass

We present the results of Monte Carlo simulations of two different 10-state Potts glasses with random nearest neighbor interactions on a simple cubic lattice. In the first model the interactions come from a \pm J distribution and in the second model from a Gaussian one, and in both cases the first two moments of the distribution are chosen to be equal to J_0=-1 and Delta J=1. At low temperatures the spin autocorrelation function for the \pm J model relaxes in several steps whereas the one for the Gaussian model shows only one. In both systems the relaxation time increases like an Arrhenius law. Unlike the infinite range model, there are only very weak finite size effects and there is no evidence that a dynamical or a static transition exists at a finite temperature.

cond-mat.stat-mech↗

The dynamics of sodium in sodium disilicate: Channel relaxation and sodium diffusion

We use molecular dynamics computer simulations to study the dynamics of amorphous (Na_2O)2(SiO_2). We find that the Na ions move in channels embedded in a SiO_2 matrix. The characteristic distance between these channels gives rise to a prepeak in the structure factor at around q=0.95 A^-1. The dynamics of sodium is given by a fast process which can be seen in the incoherent scattering function and a slow process which is seen in the coherent function. The relaxation time of the latter coincides with the alpha-relaxation time of the matrix. The Kohlrausch exponent of the fast process for q>1.6 A^1 is the same as the von Schweidler exponent for the slow one, demonstrating that the two processes are closely related.

cond-mat.stat-mech↗

Dynamical Heterogeneities Below the Glass Transition

We present molecular dynamics simulations of a binary Lennard-Jones mixture at temperatures below the kinetic glass transition. The ``mobility'' of a particle is characterized by the amplitude of its fluctuation around its average position. The 5% particles with the largest/smallest mean amplitude are thus defined as the relatively most mobile/immobile particles. We investigate for these 5% particles their spatial distribution and find them to be distributed very heterogeneously in that mobile as well as immobile particles form clusters. The reason for this dynamic heterogeneity is traced back to the fact that mobile/immobile particles are surrounded by fewer/more neighbors which form an effectively wider/narrower cage. The dependence of our results on the length of the simulation run indicates that individual particles have a characteristic mobility time scale, which can be approximated via the non-Gaussian parameter.

cond-mat.stat-mech↗

Quantitative tests of mode-coupling theory for fragile and strong glass-formers

We calculate for a binary mixture of Lennard-Jones particles the time dependence of the solution of the mode-coupling equations in which the full wave vector dependence is taken into account. In addition we also take into account the short time dynamics, which we model with a simple memory kernel. We find that the so obtained solution agrees very well with the time and wave vector dependence of the coherent and incoherent intermediate scattering functions as determined from molecular dynamics computer simulations. Furthermore we calculate the wave vector dependence of the Debye-Waller factor for a realistic model of silica and compare these results with the ones obtained from a simulation of this model. We find that if the contribution of the three point correlation function to the vertices of the memory kernel are taken into account, the agreement between theory and simulation is very good. Hence we conclude that mode coupling theory is able to give a correct quantitative description of the caging phenomena in fragile as well as strong glass-forming liquids.

cond-mat.stat-mech↗

Classical and ab initio molecular dynamic simulation of an amorphous silica surface

We present the results of a classical molecular dynamic simulation as well as of an ab initio molecular dynamic simulation of an amorphous silica surface. In the case of the classical simulation we use the potential proposed by van Beest et al. (BKS) whereas the ab initio simulation is done with a Car-Parrinello method (CPMD). We find that the surfaces generated by BKS have a higher concentration of defects (e.g. concentration of two-membered rings) than those generated with CPMD. In addition also the distribution functions of the angles and of the distances are different for the short rings. Hence we conclude that whereas the BKS potential is able to reproduce correctly the surface on the length scale beyond approx 5 Angstroems, it is necessary to use an ab initio method to predict reliably the structure at small scales.

cond-mat.stat-mech↗

Channel diffusion of sodium in a silicate glass

We use classical molecular dynamics simulations to study the dynamics of sodium atoms in amorphous Na$_2$O-4SiO$_2$. We find that the sodium trajectories form a well connected network of pockets and channels. Inside these channels the motion of the atoms is not cooperative but rather given by independent thermally activated hops of individual atoms between the pockets. By determining the probability that an atom returns to a given starting site, we show that such events are not important for the dynamics of this system.

cond-mat.stat-mech↗

Statics and Dynamics of the 10-state mean-field Potts glass model: A Monte Carlo study

We investigate by means of Monte Carlo simulations the fully connected p-state Potts model for different system sizes in order to see how the static and dynamic properties of a finite model compare with the, exactly known, behavior of the system in the thermodynamic limit. Using p=10 we are able to study the equilibrium dynamics for system sizes as large as N=2560. We find that the static quantities, such as the energy, the entropy, the spin glass susceptibility as well as the distribution of the order parameter P(q) show very strong finite size effects. From P(q) we calculate the forth order cumulant g_4(N,T) and the Guerra parameter G(N,T) and show that these quantities cannot be used to locate the static transition temperature for the system sizes investigated. Also the spin-autocorrelation function C(t) shows strong finite size effects in that it does not show a plateau even for temperatures around the dynamical critical temperature T_D. We show that the N-and T-dependence of the α-relaxation time can be understood by means of a dynamical finite size scaling Ansatz. C(t) does not obey the time-temperature superposition principle for temperatures around T_D, but does so for significantly lower T. Finally we study the relaxation dynamics of the individual spins and show that their dependence on time depends strongly on the chosen spin, i.e. that the system is dynamically very heterogeneous, which explains the non-exponentiality of C(t).

cond-mat.stat-mech↗

Growing length scales in a supercooled liquid close to an interface

We present the results of molecular dynamics computer simulations of a simple glass former close to an interface between the liquid and the frozen amorphous phase of the same material. By investigating F_s(q,z,t), the incoherent intermediate scattering function for particles that have a distance z from the wall, we show that the relaxation dynamics of the particles close to the wall is much slower than the one for particles far away from the wall. For small z the typical relaxation time for F_s(q,z,t) increases like exp(Delta/(z-z_p)), where Delta and z_p are constants. We use the location of the crossover from this law to the bulk behavior to define a first length scale tilde{z}. A different length scale is defined by considering the Ansatz F_s(q,z,t) = F_s^{bulk}(q,t) +a(t) exp[-(z/xi(t))^{beta(t)}], where a(t), xi(t), and beta(t) are fit parameters. We show that this Ansatz gives a very good description of the data for all times and all values of z. The length xi(t) increases for short and intermediate times and decreases again on the time scale of the alpha-relaxation of the system. The maximum value of xi(t) can thus be defined as a new length scale xi_max. We find that tilde{z} as well as xi_max increase with decreasing temperature. The temperature dependence of this increase is compatible with a divergence of the length scale at the Kauzmann temperature of the bulk system.

cond-mat.stat-mech↗

The High Temperature Dynamics of a mean field Potts glass

We use Monte Carlo simulations to investigate the dynamical properties of the infinite range 10 states Potts glass. By analyzing the spin autocorrelation function for system sizes up to N=2560, we show that strong finite size effects are present around the predicted dynamical transition temperature. The autocorrelation function shows strong self-averaging at high temperatures, whereas close to the dynamical transition they show the lack of self-averaging.

cond-mat.stat-mech↗

The relaxation dynamics of a viscous silica melt: II The intermediate scattering functions

We use molecular dynamics computer simulations to study the relaxation dynamics of a viscous melt of silica. The coherent and incoherent intermediate scattering functions, F_d(q,t) and F_s(q,t), show a crossover from a nearly exponential decay at high temperatures to a two-step relaxation at low temperatures. Close to the critical temperature of mode-coupling theory (MCT) the correlators obey in the alpha-regime the time temperature superposition principle (TTSP) and show a weak stretching. We determine the wave-vector dependence of the stretching parameter and find that for F_d(q,t) it shows oscillations which are in phase with the static structure factor. The temperature dependence of the alpha- relaxation times tau shows a crossover from an Arrhenius law at low temperatures to a weaker T-dependence at intermediate and high temperatures. At the latter temperatures the T-dependence is described well by a power law. We find that the exponent gamma of the power law for tau are significantly larger than the one for the diffusion constant. The q-dependence of the alpha-relaxation times for F_d(q,t) oscillates around tau(q) for F_s(q,t) and is in phase with the structure factor. Due to the strong vibrational component of the dynamics at short times the TTSP is not valid in the beta- relaxation regime. We show, however, that in this time window the shape of the curves is independent of the correlator and is given by a functional form proposed by MCT. We find that the value of the von Schweidler exponent and the value of gamma for finite q are compatible with the expression proposed by MCT. We conclude that, in the temperature regime where the relaxation times are mesoscopic, many aspects of the dynamics of this strong glass former can be rationalized very well by MCT.

cond-mat.stat-mech↗