Session B53: Disordered and Glassy Systems II

Nanoscale dynamics of binary metallic glass Cu$_x$Hf$_{1-x}$ films

Scanning probe microscopy provides a valuable tool for investigating nanoscale structure of thin films. Less commonly it can be applied to study the low speed dynamical behavior of these systems as well. Presented here are scanning tunneling microscope investigations of sputtered glass Cu$_x$Hf$_{1-x}$ films which reveal the nanocrystalline structure of the films as well as hopping dynamics of crystal clusters on the surface. A correction for limited bandwidth and a range of activation energies is developed in the context of an Arrhenius process to allow extraction of the average energy barrier for cluster hopping. Concentration of the component metals in the films was varied allowing observation of the change in cluster size as well as the transition to the amorphous state. A second form of dynamics, more diffusive in character, was found for amorphous samples.   

Isothermal Pressurizations and Glass Transition Dynamics in the Intermediate Glass-Forming Liquid Glycerol

Brillouin scattering data along both a 75 \r{ }C and 100 \r{ }C isotherm to pressures as high 6 GPa are reported for glycerol an intermediate strength glass forming liquid. This represents the highest pressure data of any type reported for glycerol, and enables us to probe directly the alpha relaxation process at these high pressures. Acoustic mode frequencies and linewidths are obtained from fits to the spectra. These frequency shifts and linewidths are fit for each isotherm with an iterative technique in which parameters are adjusted until self consistency is obtained. The Tait equation of state along with a complex expression for the dynamical longitudinal modulus, M($\omega )$, and quantitative models for other physical quantities such as the adiabatic index are used in our analysis. A Cole-Davidson function is used to model the dynamical modulus, and self-consistent fits indicate that the stretching parameter, $\beta $, is pressure independent with a value of 0.37 consistent with other low pressure acoustic results in the literature. Final values for the pressure dependent dynamical longitudinal modulus and relaxation time are obtained. In contrast to the results of recent pressure-dependent dielectric studies, there does not appear to be a second process that obscures the alpha process.   

Fast Scanning Calorimetry study of non-equilibrium relaxation in 2-Ethyl-1-Hexanol

Fast scanning calorimetry (FSC), capable of heating rates in excess of 1000000 K/s, was combined with vapor deposition technique to investigate non-equilibrium relaxation in micrometer thick ultraviscous of 2-Ethyl-1-Hexanol (2E1H) films under high vacuum conditions. Rapid heating of 2E1H samples prepared at temperatures above approximately 145 K (standard glass transition temperature of 2E1H, Tgs), resulted in well manifested dynamic glass transitions at temperatures tens of degrees higher than Tgs. Furthermore, strong and complex dependence of dynamic glass transition temperature on the sample's initial state, i.e., the starting temperature of FSC scan was also observed. We discuss implications of these results for contemporary models of non-equilibrium relaxation in glasses and supercooled liquids.   

Fluctuating Relaxation Times in Glass-forming Liquids

The presence of fluctuating local relaxation times, $\tau(\vec{r},t)$ has been used for some time as a conceptual tool to describe dynamical heterogeneities~\cite{Ediger-arpc-2000}. Here we report on a new method for determining the local phase field, $\phi(\vec{r},t)\equiv\int^{t}\frac{dt'}{\tau(\vec{r},t')}$ from snapshots $\{\vec{r}(t_i)\}_{i=1...M}$ of the positions of the particles in a system, and we apply it to extract $\phi(\vec{r},t)$ from simulations of glass forming models. By studying how the phase field depends on the number of snapshots, we find that it is a well defined quantity. By studying fluctuations of the phase field, we find that they describe heterogeneities well at long distance scales. We also determine how the stretching exponent $\beta$ depends on the coarse graining volume, in order to test the hypothesis that relaxation in small regions is exponential and it only becomes non-exponential when considering large regions of the system. \\[4pt] [1] M.~D. Ediger, 2000 Annu. Rev. Phys. Chem. \textbf{51} 99   

Mapping dynamical heterogeneity in structural glasses to correlated fluctuations of the time variables

Dynamical heterogeneity is believed to play an important role in the dynamical behavior of slowly relaxing disordered materials. In this work, we test one hypothesis for its origin, namely that it emerges from soft (Goldstone) modes associated with a broken continuous symmetry under time reparametrizations. We do this by constructing coarse grained observables and decomposing the fluctuations of these observables into transverse components, which are associated with the postulated time-fluctuation soft modes, and a longitudinal component, which is unrelated to them. We perform our test on data obtained in simulations of four models of structural glasses. We find that as temperature is lowered and timescales are increased, the time reparametrization fluctuations become increasingly dominant. In particular, the ratio between the strengths of the transverse fluctuations and the longitudinal fluctuations grows as a function of the dynamical susceptibility $\chi_4$, which represents the strength of the dynamical heterogeneity; and the correlation volumes for the transverse fluctuations are approximately proportional to those for the dynamical heterogeneity, while the correlation volumes for the longitudinal fluctuations remain small and approximately constant.   

High Pressure Brillouin Scattering in the Fragile Glass Former Cumene

In recent years full-spectrum analysis in light-scattering has been utilized to explore the liquid-glass transition at variable temperature and ambient pressure. In this study we present temperature- and pressure-dependent Brillouin scattering results for the fragile glass-former cumene. Both equal-angle forward scattering and depolarized backscattering geometries are used, and high pressures are attained by the use of a diamond anvil cell mounted in a custom temperature-controlled housing. Opening up the variable pressure regime to full-spectrum analysis will allow more stringent tests of mode-coupling theory as well as greater insight into the behavior of glass-forming systems.   

Molecular modeling of ultra-stable vapor deposited glasses

Recent studies have shown that physical vapor deposition can be used to prepare glasses of small organic molecules with remarkably high kinetic stability and low enthalpy, particularly when compared to ordinary glasses prepared by cooling the supercooled liquid. The thermophysical properties of these new ultra-stable glasses are equivalent to those of common glasses after thousands of years of aging. However, experimental studies have so far been limited to relatively few types of molecules. We propose a molecular modeling scheme to prepare stable glasses that mimics the experimental procedure of vapor deposition. For simple disaccharides, such as trehalose, the thermophysical properties of our simulated glasses are consistent with those measured experimentally. We also prepare stable glasses of trehalose and glycerol mixtures, which are of interest for their use in stabilization of biomolecules in the glass state. Results for model binary Lennard-Jones glasses, which have been studied extensively in the literature, are also discussed. We find that the most stable glasses formed by vapor deposition are equivalent to ordinary glasses formed by cooling at a rate approximately 10 orders of magnitude slower than those accessible by ordinary cooling methods.   

Dynamic heterogeneity above and below the mode-coupling temperature

We study the temperature dependence of the spatial extend of the dynamic heterogeneity in a soft sphere system near the so-called mode-coupling temperature $T_c$. We utilize a recently introduced procedure\footnote{E. Flenner and G. Szamel, Phys. Rev. Lett. \textbf{105}, 217801 (2010)} to calculate the ensemble independent dynamic susceptibility $\chi_4(\tau_\alpha)$ and the dynamic correlation length $\xi(\tau_\alpha)$ at the alpha relaxation time $\tau_\alpha$. Above $T_c$, we find that $\chi_4(\tau_\alpha) \sim \xi(\tau_\alpha)^3$ and $\xi(\tau_\alpha) \sim \ln(\tau_\alpha)$, which is the same behavior found in a binary hard-sphere system. We track these relationships below $T_c$ to examine the recently reported non-monotonic temperature dependence of dynamic correlations found in the same system\footnote{W. Kob, S. Roland-Vargas and L. Berthier, Nat. Phys. DOI:10.1038/NPHYS2133}. Finally, we examine the relationship between dynamic susceptibilities that can be determined from experiments and the dynamic correlation length $\xi(\tau_\alpha)$.   

Fast Scanning Calorimetry studies of glassy and supercooled water

Despite intense efforts, development of a comprehensive system of relationships between various condensed phases of water remains an illusive goal. The lack of consensus on the nature of supercooled and glassy water is due primarily to the lack of kinetic and thermodynamic data at temperatures from 150 to 235 K. Because supercooled water undergoes rapid crystallization near 235 K, application of standard experimental methods is virtually impossible. With the objective of gaining insights into properties of water, we have developed an experimental approach which relies on rapid (1000000 K/s) heating of micro- and mesoscopic aqueous samples prepared by vapor deposition in vacuum at cryogenic temperatures. Due to high heating rates, this Fast Scanning Calorimetry approach, makes it possible to bypass crystallization and to obtain new data on molecular kinetics and thermodynamics in glassy water in previously inaccessible temperature interval. We will report the results of our FSC studies and discuss their impact on fundamental and applied research areas where glassy and supercooled water plays significant role.   

Volume and structural analysis of super-cooled water under high pressure

Motivated by recent experimental study of super-cooled water at high pressure [1], we performed atomistic molecular dynamic simulations study on bulk water molecules at isothermal-isobaric ensemble. These simulations are performed at temperatures that range from 40 K to 380 K using two different cooling rates, 10K/\textit{ns} and 10K/5\textit{ns}, and pressure that ranges from 1atm to 10000 atm. Our analysis for the variation of the volume of the bulk sample against temperature indicates a downward concave shape for pressures above certain values, as reported in [1]. The same downward concave behavior is observed at high pressure on the mean-squared-displacements (MSD) of the water molecules when the MSD is plotted against time. To get further insight on the effect of the pressure on the sample we have also performed a structural analysis of the sample.\\[4pt] [1] O. Mishima, J. Chem. Phys. 133, 144503 (2010);   

A field theory approach to the dynamics of classical particles

For nearly 30 years, mode-coupling theory (MCT) has been regarded as the \textit{de facto} theoretic description of dense fluids and the transition from the fluid to glassy state. But MCT is limited by its ad hoc construction and lacks a mechanism to institute corrections. We present a new fundamental theory for the kinetics of systems of classical particles which represents a unification of kinetic theory, Brownian motion and field theory. It is developed from first principles via a self-consistent perturbation in terms of an effective two-body potential, and we use this theory to investigate the existence of ergodic-nonergodic (ENE) transitions near the liquid-glass transition. After a brief introduction of the theory, we will address the development of a kinetic equation of the memory function form. The memory function kernel (or self-energy) determined by the theory shares properties with the MCT form, however our theory provides the crucial advantage of well-defined, perturbative corrections.   

N.Q.R measurements of low energy Chiral structures in powdered glassy As$_{2}$Se$_{3}$

Experimental and theoretical work on the As-chalcogen glasses have shown that in the glassy state the local cylindrical symmetry associated with the elemental pyramidal unit is preserved. Here we introduce a local paracrystalline model of glassy As$_{2}$Se$_{3}$. This model is based on a tight binding calculation of the electric field gradient (EFG) at the core of an As atom located at the apex of the pyramidal structure. This EFG is shown to be hyper sensitive to the bond angles and bond lengths the As atom forms with the chalcogen nearest neighbors, as well as the hybrid angle formed with second neighbor As atoms. A continuous variation of the bonding parameters produces a unique set of these pyramidal units which are shown to fit the NQR data for powdered glassy samples. The best fit to the NQR data indicates that the pyramidal units organize themselves into Chiral structures in the glass. A plot of the electronic energy per molecular site shows that the chiral structures have on average a lower electronic energy than a random configuration.   

Slow relaxations in glasses: full aging and beyond

Experiments performed in the last years demonstrated slow relaxations and aging in the conductance of a large variety of materials. Here, we present experimental and theoretical results for conductance relaxation and aging for the case-study example of porous silicon. The relaxations are experimentally observed even at room temperature over time scales of hours, and when a strong electric field is applied for a time $t_w$, the ensuing relaxation depends on $t_w$. We derive a theoretical curve and show that all experimental data collapse onto it with a single time scale as a fitting parameter. This time scale is found to be of the order of thousands of seconds at room temperature. The generic theory suggested is not fine-tuned to porous silicon, and thus we believe the results should be universal, and the presented method should be applicable for many other systems manifesting memory and other glassy effects. Reference: Phys. Rev. Lett. 107, 186407 (2011)   

Voids and molecuar hydrogen in hydrogenated amorphous silicon

Nuclear magnetic resonance (NMR) and Infrared (IR) spectroscopy experiments show that hydrogen microstructure consists of clustered and diluted hydrogen atoms as well as voids and hydrogen molecules in hydrogenated amorphous silicon. Several theoretical studies have also attempted that whether the microstructure incorporates voids and hydrogen molecules or not, by introducing hydrogen atoms within artificially created cavities, after relaxing the models of hydrogenated amorphous silicon. However, no theoretical study, up until now, has conclusively demonstrated that the voids and molecular hydrogen are built-in features of the microstructure. We generate several realistic models of hydrogenated amorphous silicon at different hydrogen concentrations by developing an information-based inverse method. The models not only satisfy structural and electronic properties but also provide correct NMR line spectra as compare to NMR experiments. The microstructure at high ($>$15\%) hydrogen concentration shows the presence of voids and some hydrogen molecules within the voids. The voids with molecular hydrogen are built-in configurations of the microstructure because they evolve themselves while relaxing the models via the first-principles density functional method.   

Structural and microscopic relaxations in glycerol: an IXS study

We present an Inelastic X Ray Scattering study of the THz dynamics of room temperature glycerol at pressures spanning the 0.66-3 Kbar range. We propose a comparison with ultrasound absorption results available in literature, which leads to infer the presence of two distinct relaxation phenomena, a slow and a fast one. Although the former relaxation has been thoroughly studied in glycerol by lower frequency spectroscopic techniques, no experimental evidences of the latter were so far reported in literature. A line-shape modeling based upon the memory function formalism allows us to observe that the characteristic timescale of the fast relaxation ranges in the sub-picosecond, tends to decrease with increasing the wave-vector and is rather insensitive to pressure changes. More in general, the observed phenomenology definitely reveals the microscopic, single particle, nature of this additional relaxation process.