The existence of a “crossover region” in glass-forming liquids has long been considered as a general phenomenon that is as important as the glass transition. One potential origin for the crossover behavior is a liquid-to-liquid phase transition (LLPT). Although a LLPT is believed to exist in all forms of liquids, structural evidence for this, particularly in supercooled liquids, is scarce, elusive, and in many cases controversial. A key challenge to the search for a LLPT in a supercooled liquid is the interference of crystallization during cooling. Crystallization induces major structural changes, which can overwhelm and therefore mask the more subtle changes associated with a LLPT. Here, we report the results of an in-situ containerless synchrotron study of a metallic-glass-forming liquid (Zr₅₇Nb₅Al₁₀Cu_15.4Ni_12.6) that show distinct changes in the liquid structure at ~1000 K, a temperature well below the melting temperature of the liquid and 150 K above the crystallization temperature. The structural transition is characterized by growing short- and extended-range order below the transition temperature, and is accompanied by a concurrent change in density. These data provide strong evidence for a LLPT in the supercooled metallic liquid, particularly in the light of a recent computer simulation study.

The control of phase hierarchy and morphological selection during the devitrification in metallic alloys requires a full understanding of the origin of chemical and topological order in liquids and amorphous solids. Any possible medium-range and nanoscale local order in the liquid phase plays a critical role in dictating crystallization pathways. Ab-initio simulation techniques have been used for a quite long time and they accurately describe the short-range order of liquids. However, the small size of simulation cell and highly short time steps limit this technique to represent a true undercooled liquid structure. A combination of state-of-art experimental technique should be correlated with simulation method to develop a comprehensive structural description of the molten and glassy phase. In this respect, we have analyzed the as-quenched and molten state of Al-RE binary marginal glass forming systems in their glass forming compositional ranges. Al-RE binary marginal glass forming alloys are well known for their unusual primary crystallization behavior, having number density of nanocrystals on the order of 10²³ - 10²⁵ m^-3 upon thermal devitrification. As many hypotheses were developed based on both experimentation and modelling, still no consensus has been reached to explain this enigmatic behavior. In this study, we have used in-situ high energy XRD (HEXRD), atom probe tomography (APT), ab-initio and Reverse Monte Carlo (RMC) modelling to describe the glassy and molten states. It is known that in Al-RE metallic glasses, solute atoms are highly coordinated with Al atoms, which reveal itself as satellite peaks, i.e. pre & side peaks, accompanying diffuse amorphous scattering. We have shown that diffraction patterns of liquid alloys also showed satellite peaks, indicating the structural heterogeneity in amorphous state is inherited from the liquid state. Furthermore, from HEXRD experiments, it was observed that with a 100 K increment above the equilibrium melting point, the intensity of satellite peak decreases, leading to a more disordered structure. This is also supported with the corresponding RMC models constrained with ab-initio results. The structures of liquid and solid amorphous phases were studied through Voronoi and Warren-Cowley cluster analyses. These techniques have proven themselves to describe the SRO however, extending these methods to describe MRO can be problematic. In that sense, we analyzed the distribution and mutual interactions of clusters and compared them with possible metastable and stable atomic configurations. The structural heterogeneity will be discussed in details though the evolution of SRO and MRO clusters in liquid and amorphous solid.

Experimental measurements of the viscosity, eta;, of several metallic liquids are presented, demonstrating that log(eta;/eta;_o) as a function of T_A/T gives a universally scaled curve. Here, eta;_o is the viscosity extrapolated to infinite temperature and the temperature T_A corresponds to the onset of cooperative motion and is strongly correlated with T_g (T_A/T_g asymp; 2), with the fragility factor, m, embedded in the precise ratio. These results are corroborated by our molecular simulations (MD) results for Cu₆₄Zr₃₆. Taken together, they indicate that the processes underlying the glass transition actually first appear in the high temperature liquid. Structural and MD studies demonstrate that metallic glass-forming liquids develop significant short- and medium-range order with increased supercooling, which influence the physical properties of the liquid and its phase transitions. A structural signature that correlates with fragility is, therefore, expected. Our recent experimental studies of the X-ray structure factor, S(q), as a function of temperature have confirmed this. The magnitudes of the heights of the first peaks in S(q) increase abruptly upon approaching the glass transition temperature in fragile liquids, but change more continuously with decreasing temperature in strong liquids. It is often suggested that fragility is related to the nature of chemical bonding, raising the question that the changes observed in S(q) might actually reflect chemical ordering rather than overall topological ordering. This question is explored by combining data obtained from high-energy X-ray scattering studies with elastic neutron scattering data obtained using a unique electrostatic levitation facility designed to take advantage of the high flux at the Spallation Neutron Source at Oak Ridge National Laboratory. *Research was partially supported by the National Science Foundation (DMR-1206707, DMR- 1207141, DMR-1106293, DMR-141122) and NASA (NNX10AU19G).

The enthalpy-entropy compensational law (or rule), or Meyer-Neldel rule, is a very widespread thermodynamic law in nature, which indicates a simple linear relationship between activation entropy and enthalpy of a specific chemical, physical, or even biological process. It was firstly discovered in chemistry, and was then extended to biology, condensed matter physics, plasticity of crystals in mechanics, as well as material sciences. Despite decades of studies of this effect in crystals, the validity of it remains unknown in amorphous solids. In this work, for the first time, we report a universally well-defined enthalpy-entropy compensation in the deformation of metallic glasses via both computer simulations and experiments. We generalize this rule by experiments and modellings on glassy alloys with various chemical compositions. We thus extend the enthalpy-entropy compensation rule from crystals to amorphous, although the microstructure and plastic deformation units of these two phases differ sharply from each other. The compensation temperature is found to be closely associated with critical temperatures in metallic glass

We show that in deeply supercooled liquids, structural relaxation proceeds via the accumulation of Eshelby events, i.e. local rearrangements that create long-ranged and anisotropic stresses in the surrounding medium. Such events must be characterized using tensorial observables and we construct an analytical framework to probe their correlations using local stress data. By analyzing numerical simulations, we then demonstrate that events are power-law correlated in space, with a time-dependent amplitude which peaks at the alpha relaxation time. This effect, which becomes stronger near the glass transition, results from the increasingly important role of local stress fluctuations in facilitating relaxation events. Our finding precludes the existence of any length scale beyond which the relaxation process decorrelates.

The studies on dynamics and deformation in glassy materials are particularly challenging because of their strongly disordered atomic structure. Here by probing the atomic displacements and stresses changes on the potential energy landscape (PEL), two deformation modes—localized process and cascade process—have been observed. The localized deformation involves less than 30 atoms, appears in both systems. Its size is invariant of cooling history and consistent with the reported size of STZs in previous studies. On the other hand, however, the cascade deformation is more frequently observed in the fast quenched system than in the slowly quenched system. The size distribution of cascade deformation exhibits a power-law decay with the exponent of 1.6, suggesting a typical self-organized criticality in 3D system. The origin of the cascade process in the fast quenched system is attributed to the higher density of local minima on the underlying PEL.

The study of plasticity using molecular dynamics is generally limited to a predominantly athermal stress-driven regime of deformation. For material systems which are known to exhibit thermally activated plastic flow, this is a well known shortcoming of the atomistic simulation technique. Here we use ultra-long molecular dynamics simulation runs to follow thermally driven structural transitions in model glasses. Via carefully chosen thermal protocols, spanning over four orders of magnitude in temperature ramping time-scales, thermally activated processes are studied under zero load and fixed applied strain giving insight into both thermally activated plasticity and optimal simulation sample preparation strategies. The stress relaxation simulations at fixed applied strain reveal both local and collective plasticity and give tentative evidence for emergent collective behaviour and eventual shear-band activity.

We study the distribution of displacement and strain in elasto-plastic automata (EPAs) based on the picture of local shear transformation zones (STZs).
We show that the mean squared displacement and mean squared strain are both sensitive to both the initial distribution of stresses and the particular form of the kernel which couples the elementary plastic inclusions.
We find that in all cases, regardless of the initial stresses and interaction kernel, the mean squared displacement has an anomalous system size dependence, while the mean squared strain is independent of system size, in analogy with atomistic simulations.

Bulk metallic glasses are typically “fragile” liquids, with a viscosity that rises rapidly as the temperature approaches the glass transition. The long time scale associated with stress relaxations prevents the use of direct molecular dynamic simulations to determine the temperature-dependent viscosity at values much greater than one poise. We present two new approaches for examining fragility: one based upon thermodynamics, which relates to seminal work by Kauzmann [1], one based upon the development of structural order, based on recent experimental evidence [2]. We use two different CuZr potentials, and demonstrate that both predict “strong” behavior in the undercooled liquid, in contrast to experiments. The results suggest that the strong/fragile behavior is directly related to the development of chemical short-range order, and may provide important information necessary for developing better potentials.
Research supported by the U.S. Department of Energy (DOE), Basic Energy Sciences (BES), Materials Sciences and Engineering Division.
[1] W. Kauzmann, Chemical Reviews 43, 219 (1948).
[2] N. A. Mauro, A. J. Vogt, M. L. Johnson, J. C. Bendert, and K. F. Kelton, Applied Physics Letters 103, 021904 (2013).

Understanding and properly describing atomic-level structure in metallic glasses and other amorphous materials represents a long-standing and significant scientific problem. Metallic glasses have been shown to exhibit anomalous non-cubic scaling in volume with respect to their first diffraction peak position, q₁, with a power law exponent in the range of ~2.3-2.5.^{1, 2} This scaling is a measure of the dimensionality of the material, and the observed range of exponents is characteristic of a fractal. However, the nature of this underlying fractal structure is ambiguous. Our in-situ x-ray tomography measurements of the sample volume along with corresponding x-ray diffraction data shows a shift in this power law exponent with measurements of the second diffraction peak, q₂, as a function of hydrostatic pressure. We also show a crossover in the scaling behavior from exponent ~2.5 (fractal) to ~3 (homogeneous) that occurs at the second and third nearest neighbor positions, r₂, r₃, in the real space radial distribution functions for three distinct simulated metallic glasses. These results are explained using continuum percolation theory where the atomic structure has a correlation length, xi;. Experiments, simulations, and theory on multiple glass compositions corroborate that the atomic structure is well described by a specific class of fractal, the percolation cluster, demonstrating a unifying picture of how long-range structure may organize without order in metallic glasses. The long-range structural detail afforded by this model may have significant implications on the physical properties of glasses as well as the origin of the glass transition phenomenon.
1. Ma, D.; Stoica, A. D.; Wang, X. L. Nat Mater 2009, 8, (1), 30-34.
2. Zeng, Q. S.; Kono, Y.; Lin, Y.; Zeng, Z. D.; Wang, J. Y.; Sinogeikin, S. V.; Park, C.; Meng, Y.; Yang, W. G.; Mao, H. K.; Mao, W. L. Phys Rev Lett 2014, 112, (18).

An innovative model based on spinodal-like diffusional decomposition of supercooled multi-component liquids was developed in order to predict the Glass Forming Ability (GFA) of ternary alloys. The up-hill diffusion in the initially-homogeneous freezing ternary liquid was considered a necessary condition for solidification of crystalline phases. New generalized criteria for spinodal decomposition of ternary alloys for time/space-correlated fluctuations were formulated. These criteria take into account both the thermodynamic and kinetic properties of the system. The temperature dependent amplification factor of concentration fluctuations was calculated based on the available thermodynamic and kinetic data for a number of Mg-based ternary alloys. Time-dependent amplitudes of compositional fluctuations growing during rapid cooling were determined by integration of the amplification factor over the entire cooling temperature range. The introduced [GFA] parameter, inversely proportional to the integrated fluctuation amplitude, was suggested for GFA evaluation of different ternary compositions. Identical or opposite growth directions of spontaneous compositional fluctuations were considered as a criterion for the prediction of a phase crystallization sequence. Melt-spun ternary and quaternary Mg-based alloys with additions of Y, La, Zn and Nd were investigated by X-Ray Diffraction and Differential Scanning Calorimetry. The calculated [GFA] parameter was found to provide adequate GFA ranging of the investigated alloys. The model is expected to predict the GFA of any envisaged composition for which required thermodynamic and kinetic data are available, thereby avoiding cumbersome experimental efforts.

Time-dependent phenomena such as stress relaxation avalanches [1] and strain localization [2] are of interest in understanding amorphous plasticity and the effects of strain rate. We study tensile strain evolution in a bulk metallic glass by atomistic simulation in which transition-state pathways across an energy landscape with multiple minima and saddle points are sampled using a meta-dynamics algorithm [3]. Three stages of creep behavior, transient (logarithmic in time), steady-state (linear), and tertiary (diverging) are obtained over a time interval of several orders of magnitude. The corresponding atomic configurations at each stage reveal the role of local shear transformations, from independent nucleation events during the early stage to correlated shear band formation and growth in the later stage. Implications of the present simulation findings, as extensions of a previous atomistic study of creep [4], and as complementary results to current theoretical and experimental studies on metallic glasses [1, 2] will be discussed, along with considerations of other phenomena of slow dynamics, notably cement setting [5], stress corrosion cracking [5], and irradiation void swelling [6]. Together they illustrate a class of mesoscale materials science problems relevant to this Symposium.
References
[1] J. Antonaglia, Phys. Rev. Lett. 112, 155501 (2014).
[2] J-O Krisponeit et al., Nature Communications 5:3616 (2014).
[3] P. Cao, M. Li, R. J. Heugle, H. S. Park, and X. Lin, Phys. Rev. E86, 016710 (2012).
[4] T. Lau, A. Kushima, S. Yip, Phys. Rev. Lett. 104, 177501 (2010).
[5] S. Yip and M. P. Short, Nature Materials 12, 774 (2013).
[6] M. P. Short and S. Yip, Current Opinion in Solid State and Materials Science (2015, doi:10.1016/j.cossms.2014.12.005).

Onset of cooperative dynamics has been observed in the metastable regime of many molecular liquids, colloids, and granular materials approaching their respective glass or jamming transition points. It is also considered to play a significant role in the emergence of the slow dynamics. However, the nature of such dynamical cooperativity remains elusive in multicomponent metallic liquids that are characterized by the complex many-body interactions and high mixing entropy. Herein, we report indications of the onset of cooperative dynamics in an equilibrium glass-forming metallic liquid (LM601: Zr₅₁Cu₃₆Ni₄Al₉). We performed Quasi-Elastic Neutron Scattering (QENS) measurements of the mean diffusion coefficient in the generalized hydrodynamic regime (Q: 0.5 - 1.6 Å^-1). It follows an Arrhenius temperature dependence at high temperatures, but shows a deviation from the Arrhenius behavior below T_o asymp; 1300 K. Guided by Molecular Dynamics (MD) simulations, we identify T_o as the onset of cooperative dynamics, i.e., a crossover from uncorrelated dynamics above T_o to landscape-influenced correlated dynamics below T_o. The onset/crossover temperature in this metallic glass-forming liquid is found approximately twice larger than its calorimetric glass transition temperature (T_g asymp; 697 K) and in its stable liquid phase, unlike many molecular liquids. Furthermore, we show the presence of such a dynamical onset phenomenon in ten other glass-forming metallic liquids, universally occurring at approximately twice of their calorimetric glass transition temperatures and in their stable liquid phases.

Different metallic-glass-forming supercooled liquids exhibit very different temperature (T) dependence of the viscosity (fragility) and excess specific heat (Cp), profoundly influencing their glass forming ability. Using molecular dynamics simulations in different model liquids, we explain these observations from a structural perspective. We demonstrate that the contrasting Cp and relaxation/viscosity behavior has an underlying origin in their different local ordering upon undercooling. We also investigate the T-dependent evolution of the first peak position/shape in pair distribution functions of liquids. Making use of constant-volume simulations in the absence of thermal expansion and change in inherent structure, we demonstrate that the previously reported anomalous shift of the peak maximum (contraction versus thermal expansion) can be a result of the asymmetric shape of the peak, as the asymmetry increases with temperature-induced spreading of neighboring atoms to shorter and longer distances due to the anharmonic nature of the interatomic interaction potential.

We have recently developed electron correlation microscopy (ECM). ECM is the electron scattering equivalent of photon correlation spectroscopy. It uses the autocorrelation function of a time series of coherent electron nanodiffraction patterns to measure the structural relaxation time tau; of a sample. The combination of a large elastic scattering cross section and good optics means that ECM can be performed with nanometer spatial resolution using a focused probe. We will report ECM measurements with 2 nm spatial resolution of tau; and the stretching exponent β for Pd₄₀Ni₄₀P₂₀ metallic glass samples heated into the supercooled liquid region above T_g inside the electron microscope. Near T_g, we observe t of ~90 s, in reasonable agreement with previous bulk measurements. tau;(T) decreases with temperature up to T_g + 35 K, then saturates at ~10 s. β exhibits a transition from greater than one to less than one with increasing temperature. ECM experiments on supercooled Pt_57.5Cu_14.7Ni_5.3P_22.5 using a fast direct electron detection camera with 2 ms time resolution will also be discussed.

Aqueous carbonic acid(H₂CO₃) is the centerpiece of both the global carbon cycle and physiological buffer systems, yet it remains poorly characterized despite enormous effort. This reflects the fact that carbonic acid is intrinsically unstable upon contact with even a single water molecule, reacting in via a proton chain mechanism to ultimately form aqueous bicarbonate and carbonate anions and hydrated protons, which comprises the reversible mechanism of dissolution of CO₂ gas. While solid and gaseous carbonic acid have been studied in some detail, only two spectroscopic studies of the aqueous form have been successful. We report the detection and characterization of both aqueous H₂CO₃ and dissolved CO₂ by X-ray absorption spectroscopy in a liquid microjet designed to effect rapid mixing of reactant solutions, wherein protonation of a bicarbonate solution continuously generates the short-lived acid under ambient conditions. Accompanying first principles calculations of the carbon K-edge transitions facilitate spectral assignment and interpretation.
Lam, R. K., England, A. H., Smith, J. W., Rizzuto, A. M., Shih, O., Prendergast, D., Saykally, R. J., "The hydration structure of dissolved carbon dioxide from X-ray absorption spectroscopy" Chem. Phys. Lett., 633, 214-217 (2015).
Lam, R. K., England, A. H., Sheardy, A. T., Shih, O., Smith, J. W., Rizzuto, A. M., Prendergast, D., Saykally, R. J., "The hydration structure of aqueous carbonic acid from X-ray absorption spectroscopy" Chem. Phys. Lett., 614, 282-286 (2014) *Cover Article.
*A. H. England, A. M. Duffin, C. P. Schwartz, J. S. Uejio, D. Prendergast, and R. J. Saykally "On the Hydration and Hydrolysis of Carbon Dioxide," Chem. Phys.

Hydrophobic interactions have been recognized as a key driving force for water-mediated self-assembly processes such as protein folding and micelle formation. Such interactions are known to be dependent on the topology of the solute and the ionic environment. However, the molecular mechanisms of these phenomena are not yet fully understood. Here, we provide molecular insights into the effects of proximally immobilized ions and solute curvature on the hydrophobic hydration. For the immobilized ion effects, we measure the hydrophobic interactions between model amphiphilic nano-rods and a reference non-polar plate. Inspired by recent chemical force microscopy experiment using oligo-β-peptide, we study both globally amphiphilic (GA) and non-globally amphiphilic (iso-GA) nano-rods. Our study unambiguously reveals that the hydrophobic interaction can be switched on and off by choosing different chemical nano-patterns on the surface of the nano-rods. We further demonstrate that for the GA nano-pattern, the strength of the hydrophobic interaction can be tuned by using different immobilized ion species. Interestingly, we find that the effect of immobilized ions do not necessarily rank into an order that has been identified for free ions, or the so-called Hofmeister series. To study the curvature-dependent hydrophobic solvation, we measure the solvation force experienced by nano-tubes with varying diameters in water, methanol and hexane. While both water and methanol can form hydrogen bond networks, they exhibit distinct thermodynamic profiles in solvating the nano-tubes. Analysis of the solvation structures shows that methanol can form ordered interfacial layer near flat non-polar surface to avoid breaking hydrogen bonds. Without hydrogen bonding, solvation in hexane (used as a reference) is less sensitive to temperature, which implies that the structure of solvating hexane does not change as much as the ones of hydrogen-bonded liquids. Our results demonstrate that the interfacial structure of hydrogen-bonded liquid is determined by the interplay between the topology of the solute and the molecular shape of the solvent that optimizes the free energy of the hydrogen bond network.

In most reviews on the Glass Transition phenomenon, it is evoked that no remarkable changes happen on the local structure of a liquid or a melt as the glass transition temperature is crossed; the static structure factor varies only smoothly with the temperature, whereas the dynamical properties evolve dramatically. We propose here to revisit this point of view with some non trivial behaviors of the relaxation properties in the time scale of pico-to-nanosecond, thanks to quasi elastic neutron scattering experiments that is related to structural analysis performed by elastic scattering, both under pressure. The systems considered here are respectively by decreasing their complexity: mono-alcohols, where the molecular self-association driven by the presence of H-bonds is counterbalanced by the steric hindrance of the alkyl chains, aromatic quasi planar and spherical like van der Waals systems. While their ‘fragility&’ at the glass transition remains pretty similar, their strongest differences can be found in the high temperature or Arrhenius regime not far from their melting temperature, where an effective activation energy E_yen; of already several k_BT can be related to intermolecular interactions and specific motions due to the molecular shape.

Glass dynamics can be induced by light irradiation. We used scanning tunneling microscopy to study surface glassy dynamics of amorphous silicon carbide irradiated with above-bandgap 532 nm light. Surface clusters of ~ 3-5 glass forming unit in diameter hop mostly in a two-state fashion, both with and without irradiation. A very long (~1 day) movie of individual clusters with varying laser power density provides direct evidence for changes in photoinduced dynamics of surface clusters. The average surface hopping increases by a factor of 3 upon photoexcitation, mainly by recruiting immobile clusters to hop. No significant heating was observed. We propose two electronic mechanisms: saturation and cooperativity for the photoenhanced surface dynamics. Such mechanisms could underline the macroscopic photoinduced phenomena in opaque glasses including photorelaxation and photoinduced aging.

Glasses play an important role in technology as a result of their macroscopic homogeneity (e.g., the clarity of window glass) and the ability to tune properties through composition changes. A problem with liquid-cooled glasses is they exhibit marginal kinetic stability and slowly evolve towards lower energy glasses and crystalline states. We have used physical vapor deposition and the mobility of glassy surfaces to prepare what are likely the most stable glasses on the planet. Our materials have the properties expected for “million-year-old” glasses, including high density, low enthalpy, and high mechanical moduli. Surprisingly, these glasses “melt” like crystals, with a constant velocity transformation front. Recent work has shown that these efficiently packed glasses suppress the amplitude of the beta relaxation process and also the number of tunneling two-level systems. We have used ellipsometry and GIWAXS to characterize the structure of organic glasses prepared by vapor deposition. By controlling the substrate temperature during deposition, molecular orientation can be tuned from being parallel to perpendicular to the substrate, while maintaining high kinetic stability. Anisotropic molecular packing can also be tuned by changing the substrate temperature. These effects can be reproduced in computer simulations and understood in terms of surface mobility and the molecular orientation present at the equilibrium liquid surface. These developments present significant opportunities to expand our understanding of amorphous packing and to design new anisotropic solids for applications such as organic electronics.

Many people are aware that neutron crystallography can help in the structure-based design of inhibitors using static protein targets. However, recently it has been shown that computational drug design protocols benefit considerably from a dynamic, rather than just a static, description of the protein to be modulated. We describe how small-angle and dynamic neutron scattering, when combined with computer simulation, provide useful information on the motions involved [1-7]. The use of a dynamic description in supercomputer-based virtual high-throughput screening is then outlined, with reference to recent results in our lab in discovering lead compounds for antibiotic resistance, thrombosis and hyperphosphatemia [8].
Hong et al PRL, 107, 148102 (2011); 108, 238102 (2012); 110, 028104 (2013); PRL 112, 158102 (2014).
Noe et al PNAS, 108, 4822 (2011).
Godec et al PRL, 107, 267801 (2011); PRL 111, 127801 (2013).
Lindner et al JCP 139, 175101 (2013).
Zheng et al JCP 139, 175102 (2013)
Hong et al Biophys. J 107 393 (2014).
Moritsugu et al JPCB, 118, 8559 (2014)
Ellingson et al J. Comp Chem 34 2212 (2013); Mol. Simul. 40, 848 (2014); JPCB 119 1026 (2015).

The poor solubility of the active pharmaceutical ingredients can significantly reduce the efficacy of medications. Amorphous drug forms often have enhanced solubility, faster dissolution rates, and higher bioavailability compared to their crystalline counterparts. Here, we report the structural characterization of several amorphous drugs obtained using conventional vitrification techniques, the quenching of solutions suspended in an acoustic levitator and by spray drying. The results demonstrate that acoustic levitation serves as a useful screening tool for characterizing the structural evolution as the droplets dry in a controlled environment. Diffraction results combined with 13C NMR data are discussed for several pharmaceutical systems. Neutron and X-ray pair distribution function analyses were applied to the glassy materials, and the results compared to those from their crystalline counterparts. The two diffraction techniques yielded similar results in most cases and identified distinct intra- and inter-molecular correlations. A multi-parameter data fitting technique has been developed to separate out the intra-molecular scattering contribution from the measured structure factors.
Example studies include Probucol and Itraconazole: (1) Probucol is used to reduce cholesterol in blood, so it&’s effectiveness depends strongly on it&’s solubility which is very low. The glassy form was found to be stable with a shelf life of at least 8 months. Probucol is known to have at least two conformational polymorphs where the packing is determined by van der Waals interactions. 13C NMR spectra confirmed that in the glassy state the Probucol molecule retains an extended chain conformation. The intra-molecular fit for the glassy PDF reveals only very weak broad inter-molecular interactions are observed at distances of 6.3 and 11 Å. (2) Itraconazole is an effective antifungal drug, but it exhibits a relatively poor dissolution rate and low bioavailability after oral administration. Itraconazole has a rod-like molecular structure and DSC measurements have revealed two glass transition temperatures at 74⁰C and 90⁰C. Upon cooling from the melt a color change occurs from transparent to white at 90⁰C and is most likely associated with an isotropic to nematic transition. Significant ordering is observed in the diffraction data with the transition at 74⁰C which can be explained by a nematic to smetic A transition.
Future directions on slow release pharmaceuticals contained in mesoporous silicas chacaterized by combined small and wide angle scattering extended range PDF techniques, as well as amorphization through the incorporation into polymer complexes will also be discussed.

In connection to the possibility of exploring the so-called 'no-man's land', a large number of studies of supercooled water in nanoconfined geometry or around macromolecules of biological interest like proteins has characterized the scientific literature of the last years.
Here we show a detailed microscopic study of protein hydration water done with molecular dynamics simulations on an aqueous solution containing a Lysozyme protein. The translational dynamics has been investigated from 300 K down to 200 K. We found that hydration water shows two distinct structural relaxations: one is the structural alpha relaxation typical of glass formers and present also in bulk water upon supercooling. The second relaxation can be ascribed to a dynamic coupling of hydration water motion to protein surface motion. Both relaxation times slow down consistently upon cooling the system. Static properties and hydrogen bond network distortion of hydration water are also analyzed.
References
Camisasca G. and Gallo P. 2015, in preparation.

Molecular dynamics simulations of TIP4P and TIP4P/2005 are presented upon supercooling along different isochores close to the liquid liquid critical point.
We study the MCT behaviour that holds in the region of mild supercooling, the fragile to strong transition and its connection to the Widom line, and the hopping phenomena that appear in the region of strong supercooling.
The Widom line is the line of maxima of thermodynamics response functions that emanates from a critical point. In particular we studied densities ranging from 1.03 to 0.95 gr/cm³ and temperatures spanning from 300 K to 190 K.
We calculated the intermediate scattering functions and the translational relaxation times and we studied the onset of hopping for supercooled water through the van Hove self correlation functions. In particular hopping appears more favored on the low density side of the Widom Line.
Besides the connection between the fragile to strong transition and the hopping phenomenon we also explore the link to the Widom line for these dynamic features [1].
[1] M. De Marzio, M. Rovere and P. Gallo, in preparation (2015).

In supercooled liquid region, good glass formers show high stability against devitrification [1, 2]. In this paper, we would like to illustrate the major issues concerning phase stability of supercooled metallic-glass forming liquids from the aspect of the ordering kinetics and local atom rearrangement. Here, we experimentally compare the atomic structural evolutions during amorphous-to-crystalline transformation between a good glass former, Zr₄₆Cu₄₆Al₈, and an average glass former, Zr₅₆Cu₃₆Al_8, upon isothermal annealing in the supercooled liquid region. In-situ synchrotron pair distribution function (PDF) measurements were conducted at Advanced Photo Source (APS) of Argonne National Laboratory (ANL) with a data rate 1 Hz. The PDF analyses show that the initial states of both two glasses are nearly the same. During devitrification, the nearest new bond in Zr₄₆Cu₄₆Al₈ formed at much higher correlation length than that in Zr₅₆Cu₃₆Al₈ did. Furthermore, the incubation time of the newly-forming bonds in Zr₄₆Cu₄₆Al₈ varies widely, while all the correlation shells of Zr₅₆Cu₃₆Al₈ started to develop at almost the same time. In addition, Zr₄₆Cu₄₆Al₈ finally crystallized into a crystalline phase with confined order, in contrast, Zr₅₆Cu₃₆Al₈ crystallized into a crystalline phase with much more extended ordering. Our results suggest that medium-range order plays a unique role on frustrating devitrification in the metallic liquids with good glass-forming ability.
[1] P.G. Debenedetti, F.H. Stillinger. Supercooled liquids and the glass transition. Nature 410, 259, 2001.
[2] S. Lan, X. Wei, J. Zhou, Z. Lu, X. Wu, M. Feygenson, J. Neuefeind, X.-L. Wang. In-situ study of crystallization kinetics in ternary bulk metallic glass alloys with different glass forming abilities. Applied Physics Letters 105, 201906, 2014.

The ST2 interaction potential has been used in a large number of simulation studies to explore the possibility of a liquid-liquid phase transition (LLPT) in supercooled water. Using umbrella sampling Monte Carlo simulations of ST2 water, we evaluate the free energy of formation of small ice nuclei in the supercooled liquid in the vicinity of the Widom line, the region above the critical temperature of the LLPT where a number of thermodynamic anomalies occur. Our results show that in this region there is a substantial free-energy cost for the formation of small ice nuclei, demonstrating that the thermodynamic anomalies associated with the Widom line in ST2 water occur in a well-defined metastable liquid phase. On passing through the Widom line, we identify changes in the free energy to form small ice nuclei that illustrate how the thermodynamic anomalies associated with the LLPT may influence the ice nucleation process. We also present preliminary results of simulations close to the critical point of the LLPT, as well as results for larger systems to test for finite-size effects.

We conduct simulations of a metamagnet below the tricritical temperature in order to study the nucleation process from a metastable phase to a stable phase in the presence of an additional metastable phase. At fixed temperature below the tricritical temperature, the phase diagram as a function of the direct and staggered magnetic fields containes a triple point at which three phases coexist, two antiferromagnetic phases and a ferromagnetic phase. We examine the nucleation process from one antiferromagnetic phase to the other as a function of the direct field. We find that the decrease in the nucleation time as the direct field increases can be understood in terms of the influence of the proximity of the metastable ferromagnetic phase in the free energy surface of the system. Our results have implications for understanding nucleation in systems with multiple metastable phases, such as may occur in deeply supercooled water.

Metallic glasses are inherent metastable from a thermodynamic point of view and therefore crystallize upon annealing. In the present work a CuZrAl-based BMG was heated at a wide range of rates between 0.08-180 K s-1 in order to investigate the influence of the heating rate on crystallization, which involves nucleation and growth. Immediate quenching is applied to the BMG after heating it. We can induce partial crystallization and preserve the appendant microstructure. By applying high heating rates kinetic constraints are effective not only on the phase formation, but also on the crystallization kinetics. Direct observation of crystallite sizes and number allows us to explore these metastable kinetics. The nucleation rate at very fast heating significantly diverges from the steady state. The phase formation changes from eutectic to polymorphic crystallization, and finely distributed metastable shape-memory crystals precipitate in the glass. The so-obtained bulk metallic glass composites exhibit a strongly enhanced room-temperature deformability. Here we show that fast heating not only permits tailoring of BMG composites to overcome the intrinsic brittleness of monolithic BMGs, but also permits studying the fundamental crystallization processes in highly undercooled melts.

Results of experiments using three scattering methods will be presented here:
1) Elastic neutron diffraction measurements of D2O using SPINS at NCNR giving the density profile of supercooled water at certain range of pressuresthat shows a Hysteresis phenomenon. This gives the locus of LLPT line in the two-phase region.
2) QENS measurements of H2O to study the dynamic crossover (DC) phenomenon using Backscattering spectrometers at NCNR and SNS. The DC is related to the existence of the Widom line which implies that there is a LLCP and the associated one-phase region.
3) Inelastic neutron scattering experiments (INS) at SNS on H2O show that: The emergence of the LDL and HDL phases as the result of thermodynamic continuation of the LDA and HDA amorphous ices into the liquid state.
Both the peak frequency and the width of the Boson peak exhibit abrupt changes as the water transforms between LDL and HDL phases. This allows us to rescale all the measured Boson peaks into two distinct peaks, one belonging to the LDL phase and the other to the HDL phase.

To see the low temperature behavior of supercooled water, thermodynamic (ρ-p-T relation and isochoric heat capacity) and transport properties (self-diffusivity, shear viscosity, and thermal conductivity) of four water models (TIP3P, SPC/E, TIP4P/Ew, and TIP4P/2005) are calculated with the use of equilibrium molecular dynamics method. The study is conducted over a very broad phase space (temperature from 145 to 363 K and density from 0.865 to 1.3 g/cm³), hence enables a thorough study to the physics in the deeply supercooled regime. First, the water anomalies of the three transport properties are shown to occur across all models, indicating their existenc e to be not model dependent. Second, the structural, dynamic, and thermodynamic anomalies are arranged in a hierarchy fashion from high to low temperatures. Particularly, the thermal conductivity anomaly is buried between isochoric heat capacity anomaly and locus of compressibility maxima. We show that these anomalies to be consistent with the two-fluid picture (LDL/HDL), and the physical property changes of water can be explained through the change in their LDL/HDL composition.

The dynamic properties of water in confined geometry, such as chromatography and bio membrane, are highly needed for understanding the unique properties and processes which differ from those in bulk water. In the present study, the dynamic properties of low-temperature water in mesoporous organosilica are investigated by neutron spin echo (NSE) and dielectric relaxation spectroscopy (DRS). The periodic mesoporous organosilica (PMO) used have independent cylindrical channels of pore diameter ~3.0 nm whose wall consists of silica matrix embedded with phenyl groups (Ph-PMO) and the silanol groups (Si-OH) further modified by trimethylsilane (TMS-Ph-PMO) to increase the hydrophobicity of the wall. Water confined in Ph-PMO and TMS-Ph-PMO is frozen at ~230 K by differential scanning calorimetry. The intermediate scattering functions of water in both PMOs above the freezing point are analyzed with the Kohlrausch-Williams-Watts function. The relaxation time obtained of confined water show non-Arrhenius behavior in both PMOs. The diffusion coefficient of a water molecule in TMS-Ph-PMO is larger than in Ph-PMO, i.e. the motion of a water molecule is less retarded with an increase in hydrophobicity of the pore wall. The DRS data taken below the freezing point reveal that there are three types of confined water: ice in the center of pore, unfrozen interfacial water on the pore wall, and intermediate unfrozen water between them. Interestingly, the relaxation times of the intermediate water confined in Ph-PMO, TMS-Ph-PMO, and MCM-41, whose wall is hydrophilic, fall on a similar line in their Arrhenius plots. Furthermore, an inflection point is observed at ~180 K, probably due to a crossover from the a-relaxation to the b-relaxation of the deeply supercooled water. The relaxation time of the interfacial water strongly depends on the nature of pore wall and decreases in the order of TMS-Ph-PMO > MCM-41 > Ph-PMO.

The dynamics of molecular structures in methanol and methanol/water mixtures was investigated using quasielastic neutron scattering (QENS). These samples are notable cases of the many liquids^[1] which, although homogeneous at the macroscopic level, present significant structuring at the nanoscopic level. By measuring the structure factor from samples differing only by their isotopic composition, structuring between methyl groups and hydroxyl groups (of both methanol and water) is put in evidence. Such structuring takes place over nano length scales which are however larger than the intermolecular distance. The dynamics of such molecular organization was studied by QENS taking advantage of the same isotopic substitution techniques. The data reveal that the molecular arrangement is transient but slower than the structural relaxation. In the pure methanol case the results can be understood within the Transient Chain Model (TCM).^[2] For methanol/water mixtures, the obtained dynamics over length scales of several nanometers is diffusive and significantly slower than the single particle diffusion of either methanol or water in the system. These results further validate the hypothesis of the existence of some structuring in the system^[3] which is dynamic in character.
[1] Mesostructure and Dynamics in Liquids and Solutions ; Faraday Discussion; 2013; Vol. 167.
[2] Gainaru, C.; et al.; Phys. Rev. Lett. 2010, 105, 258303.
[3] Dixit, S.; et al.; Nature 2002, 416, 829.

We discuss recent and literature data on the relaxation times (the primary t_α), viscosity and self-diffusion in water-glycerol and water-methanol mixtures in a wide temperature interval from the stable water phase to the deep supercooled regime (373 - 100K). In particular, we have studied the mixture in terms of the water molar fraction (X) by exploring the entire composition range at fixed temperatures with steps of 5K.
In such a way we observe a marked deviation from the ideal thermodynamic behavior of the transport functions. A deviation that is strongly T and X dependent spanning values that range from two orders of magnitude at the highest temperature to more than five in the deep supercooled regime (more precisely at about 200K).
The analysis of these deviations , in terms of the difference from the measured values and the ideal ones, reveals the dominant role of the hydrogen-bonding water network in the system properties up to X = 0.3.
Furthermore, an Arrhenius plot of the maximum excess value (Δt_α(T) v.s. 1/T) reveals linear behaviors with two significant slope changes: one at the dynamical crossover temperature (T_L~ 225K, i.e. the locus of the Widom line); and one at T ~ 315K (the temperature of the minimum of the water isothermal compressibility, chi;_T.

This talk will describe recent x-ray spectroscopy and scattering measurements, using both synchrotron radiation at SSRL and the x-ray laser at LCLS, at temperatures from deep supercooling to boiling. The results shows that the liquid can be described as fluctuations between two types of local hydrogen bonded structures driven by incommensurate requirements for minimizing enthalpy and maximizing entropy. The connection of these results to low and high density water and the 2^nd critical point model will be discussed.

Abstract:Abstract: Works of art and artifacts that constitute our cultural heritage are subject to deterioration. Their surfaces interacting with the environment are the most prone to aging and decay; accordingly, soiling is a prime factor in the degradation of surfaces, chemical and mechanical degradation are often associated to soiling and lead to the disfigurement of a piece of art. The effects of these processes are usually strongly amplified in the presence of protective coatings (mainly acrylic and vinyl polymers), applied in previous restoration treatments. In the past years we pioneered the synthesis and the application of several advanced systems for the consolidation and the cleaning of works of art, as hydroxides nanoparticles, microemulsions and chemical/physical gels. Small-angle neutron scattering (SANS) with contrast variation and SAXS have been used to infer a detailed picture of the structure of these complex fluids. We found that the composition and the structure at the nanoscale determine the capability of removing a broad range of different polymer coatings from porous materials. In this talk examples from self assembled systems for the cleaning or the removal of coatings from pictorial surfaces will be highlighted. Micellar solutions and microemulsions constitute very efficient systems for the removal of acrylic, vinyl and alkyd polymers or grime/soil. These systems (as well as neat solvents used in "traditional" conservation) can be confined into chemical and physical gels having proper nano-domains for the upload or the delivery of compounds from/to the work of art. For example, a fine control of the cleaning procedure can be obtained even for challenging cleanings as water sensitive works of art, where the cleaning can be achieved by using water confined into gels, and leaving no residues on the works of art.
Keywords: Micelle, Microemulsions, Physical Gel, Chemical Gel, Consolidation, Cleaning, Wall Paintings, Easel Paintings
References
1) P. Baglioni, D. Chelazzi, Nanoscience for the Conservation of Works of Art, Royal Society of Chemistry, 2013.
2) D. Chelazzi, G. Poggi, Y.Jaidar, N. Toccafondi, R. Giorgi, P. Baglioni, Journal Colloid Interface Sci. 2013, 392, 42-49.
3) R. Giorgi, M. Baglioni, D. Berti, P. Baglioni, Acc. Chem. Res. 2010, 43, 695-704.
4) Baglioni, M.; Rengstl, D.; Berti, D.; Bonini, M.; Giorgi, R.; Baglioni, P. 2010, 2, 1723-1732.
5) Baglioni, M.; Giorgi, R.; Berti, D.; Baglioni, P. Nanoscale 2012, 4, 42-53.
6) Baglioni, M.; Berti, D.; Teixeira, J.; Giorgi, R.; Baglioni, P. Langmuir, 2012, 28, 15193-15202.
7) Carretti, E.; Fratini, E.; Berti, D.; Dei, L.; Baglioni, P. Angew. Chem. Int. Ed. 2009, 48, 8966minus;8969.
@No kinship is present among the authors

Owing to their unique topology, macromolecular rings without chain ends are subject of intense interdisciplinary research. In biology among others, plasmids, cyclic polysaccharides, peptides, RNA or DNA are investigated. Employing blends of different polymer architectures including rings, material science and engineering aim for polymer materials with emerging properties. Recently in particular simulation science led to new inshy;sight into ring dynamics on a coarse grained level. One of the important and fundamental goals of this broad approach is the understanding of entangled polymer dynamics far beyond the state#8208;of#8208;the-shy;art. For cyclic polymers reptation as well as the related relaxation through contour length #64258;uctuations (CLF) and constraint release (CR) are fully suppressed and the tube concept is not applicable. Instead a lattice animal like conformation may describe the topological interactions in large ring melts. We present neutron scattering data (i) on the structure and dynamics of melts from polyehtylenoxide rings with molecular weights up to ten times the entanglement mass of the linear counterpart [1] and (ii) address the ring dynamics in ring linear blends. In ring melts the ring polymers assume very compact conformations displaying a structure approaching a mass fractal as hypothesized by recent simulation work. The center of mass motion is sub diffusive up to a mean square displacement of several times the radius of gyration. The segmental dynamics is characterized by fast Rouse relaxation of sub units (loops) and slower larger scale dynamics displaying a lattice animal like loop displacement. The loop size is an intrinsic property of the ring architecture and is independent of molecular weight. In ring#8208;linear blends at low ring concentration the segmental ring dynamics is slaved by the host and differs qualitatively from that of the ring melts. The ring topology suppresses ring segmental motion beyond the entanglement constraints imposed by the host. In this way the fundamental length scale of all tube theories is directly and uniquely accessed #8208; the lateral tube size corresponds to the unrelaxed plateau modulus. In moderately entangled matrices, where on the time scale of the experiment chain end effects play a role, the ring motion directly reveals that CLF contributions are weaker than assumed in state-shy;of-shy;the-shy;art rheological theories and show that at short times CR effects are negligible.
This experimental observation of the space time evolution of segmental motion in ring polymers illustrates the dramatic dynamic consequences of their topology that is unique among all polymeric systems of any other known architecture.
[1] Physical Review Letters 131, 168302 (2014)

Polymer melts are glass-forming systems where the structural units are macromolecules. These systems display not only the typical dynamic processes of glass-formers in general (α and β relaxations) but also large-scale macromolecular chain motions, which are unique for polymer melts and which drive the viscoelastic properties of such materials. The Rouse model is nowadays the accepted theoretical framework for chain dynamics of unentangled polymers - melts of macromolecules of molecular weight lower than a certain limit: the ‘entanglement mass&’. The Rouse model describes the chain motions in terms of the so-called Rouse modes, X_p(t), of wavelength N/p (N being the number of ‘beads&’ in the macromolecular chain and p the mode number p=1,2hellip;N-1) and the Rouse modes correlators C_p(t)=< X_p(t), X_p(0)>. Orthogonality of the Rouse modes and exponential behavior of C_p(t) are the two main predictions of the Rouse model and the basis of the derivation of the magnitudes probing chain dynamics. These predictions are a consequence of neglecting, respectively, the spatial and time correlation of the random forces acting on a polymer segment in a polymer melt. Although the Rouse model works rather well for homopolymer melts at temperatures well above the glass-transition (T_g), strong deviations are nowadays well documented in the case of the so-called dynamically asymmetric polymer blends (miscible polymer systems where the two components display different α-relaxation time) when the chain dynamics of the fast component is monitored in the presence of the slow component. These experimental results have been confirmed by MD-simulations of atomistic systems and generic ‘bead-spring&’ models. The simulation results have demonstrated that in these systems, C_p(t) are non exponential, even in the long wavelength (N/p) limit where local intra-chain potentials and chain stiffness do not play any role. Recently (S. Arrese-Igor et al., Phys. Rev. Lett. 113 (2014) 078302) it has been experimentally shown that the long wavelength C_p(t) is also non-exponential in the case of homopolymers at low temperatures approaching T_g. In fact, the values of the non-exponential parameter (β of a stretched exponential description of C_p(t)) corresponding to different systems (homopolymers and blends) and different temperatures collapse on a ‘universal&’ curve when they are represented as a function of the ratio tau;_R/tau;_α, where tau;_R is the Rouse time (p=1) and tau;_α the characteristic time of the relevant density fluctuations in the system. These results can be quantitatively explained in the framework of a theoretical approach based on: (i) a generalized Langevin equation formalism and (ii) a memory function, which takes into account the interplay between intra-chain dynamics and collective dynamics. This approach opens the way of understanding the relationship between α-relaxation and chain dynamics - a general unsolved problem for glass-forming polymers.

Block copolymers provide a remarkably versatile platform for achieving desired nanostructures by self-assembly, with lengthscales varying from a few nanometers up to several hundred nanometers. In particular, block copolymer micelles in selective solvents are of great interest across a range of technologies, including drug delivery, imaging, catalysis, lubrication, and extraction. While block copolymers generally adopt the morphologies familiar in small molecule surfactants and lipids (i.e., spherical micelles, worm-like micelles, and vesicles), one key difference is that polymeric micelles are very rarely at equilibrium. The primary reason for this is the large number of repeat units in the insoluble block, N_core, which makes the thermodynamic penalty for extracting a single chain (“unimer exchange”) prohibitive. As a consequence, the critical micelle concentration (CMC) is rarely accessed experimentally. However, in the proximity of a critical micelle temperature (CMT) equilibration is possible. We describe time-resolved SANS measurements of unimer exchange in selected block copolymer systems, including diblocks and triblocks, at various molecular weights and concentrations. The results are interpreted in terms of current models of block copolymer dynamics.

Block copolymer soft solids are formed by driving a disordered triblock copolymer solution through a thermoreversible phase transition to close packed micelles. For specific chain architectures, the crystal structure of the soft solid further undergoes an order-order transition (OOT) from spherical micelles packed in a cubic crystal structure to close packed cylindrical micelles. The soft solids are aligned to near single crystals with oscillatory shear. The crystal structure is characterized using small angle neutron scattering (SANS) with contrast variation while mechanical properties are probed with rheology. Three different projections of the aligned structure are characterized to relate structure and rheology. We further characterize the impact of nanoparticles dispersed in the soft solids on the nanoscale structure and flow behavior of the micellar phase. An aqueous block copolymer solution with hydrophilic particles, specifically globular proteins, segregated into the interstitial spaces between micelles provides a system with a wide variety of structures; including BCC and FCC phases as well as hexagonally packed cylinders. Changes in structure of the cubic and cylindrical micellar crystals are observed with dispersed protein. The flow behavior of both phases is also observed and quantified. The observations and connection to macroscopic rheology are discussed.

In spite of the central role of aqueous interfaces in most of chemical science and engineering, they are poorly understood at the molecular scale. We will discuss recent progress and new challenges in predicting the electronic [1] and vibrational [2] properties of water [3] and simple aqueous solutions [4] interfaced with semiconducting and insulating surfaces, at ambient conditions and under pressure [5]. We will present results obtained by coupling ab initio molecular dynamics calculations with many body and density functional perturbation theory.
[1] T. Anh Pham, Cui Zhang, Eric Schwegler, and Giulia Galli, Phys. Rev. B 89, 060202(R) (2014); T. Anh Pham, Donghwa Lee,Eric Schwegler, and Giulia Galli, JACS 136, 17071 (2014).
[2] Quan Wan, Leonardo Spanu, Giulia Galli and Francois Gygi, J. Chem. Theo. Comp. 9, 4124 (2013); Patrick Huang, T. Anh Pham, Giulia Galli, and Eric Schwegler , J. Phys. Chem. C 118, 8944 (2014).
[3] Alex P. Gaiduk, Francois Gygi, and Giulia Galli, J. Phys. Chem. Lett. 2015 (in press).
[4] Alex P. Gaiduk, Cui Zhang, Francois Gygi, and Giulia Galli, Chem. Phys. Lett. 604, 89 (2014); Quan Wan, Leonardo Spanu, Francois Gygi and Giulia Galli J. Phys. Chem. Lett. 5(15), 2562 (2014); Cui Zhang, Tuan Anh Pham, Francois Gygi and Giulia Galli, J. Chem. Phys- Comm. 138, 181102 (2013).
[5] Ding Pan, Quan Wan, and Giulia Galli, Nat. Comm. 5, 3919 (2014).

In nature ambient liquid water is denser than its solid counterpart, crystalline ice, an anomalous property of water which has enormous impact on the environment and life. However, semi-local (GGA) density functional theory (DFT) based ab initio molecular dynamics fail to predict this important property of water. We have found that this discrepancy arises due to the lack of van der Waals (vdW) interactions in GGA-DFT. In fact, an accurate description of vdW interactions is crucial in stabilizing liquid water configurations with an increased population of molecules in the interstitial region (i.e., the region between the first two coordination shells), thereby increasing the density by ~15% with respect to ambient liquid water at the GGA-DFT level of theory. Furthermore, by simultaneously accounting for both exact exchange and vdW interactions, the computed equilibrium densities of both liquid water and crystalline ice were found to be within a few percent of the experimentally observed densities—a level of accuracy that has been difficult to achieve from DFT simulations of water to date. With such accurate potentials, we further attempt to predict and characterize the anomalous density maximum that is observed in liquid water just above the melting temperature.

To accurately define the oxygen-oxygen radial distribution function g_OO(r), a model of the electron density is required to accurately weight the H_OO(Q) component relative to the intramolecular and oxygen-hydrogen correlations from the total intensity observable. In this talk I will carefully define the current tractable electron density model, and its underlying assumptions, and more explicitly utilize two restraints on the allowable g_OO(r) functions which must conform to both very low experimental errors at high-Q and the need to satisfy the isothermal compressibility at low-Q. Although highly restrained by these conditions, the underdetermined nature of the problem is such that I will present a family of g_OO(r)&’s that provide equally good agreement with the high-Q intensity and compressibility restraints and with physically correct behavior at small r.

Identifying theoretical concepts that allow to accurately and computationally efficiently reproduce the complex atomic and electronic structure of liquid water on the microscopic level is crucial to describe and understand properties of and processes occurring in electro-chemical and biological systems. Performing extensive ab-initio molecular dynamics simulations we study the solvation of charged and neutral (Zn, Cl, I, Br) ions in water. Analyzing the structural correlations, electronic properties and formation energies of the studied ions we discuss the quality of the description obtained at different levels of theoretical modelling (classical models, DFT using different exchange-correlations functionals and implicit solvation models) and compare to experimental observations. Aligning the water band edges on an absolute scale allows us to compare the calculated charge transfer-to-solute states (CTTS)/transition levels for the halide ions in water with photo-emission experiments.

Neutron scattering study shows that water molecules confined within the 5 A channels of mineral beryl undergo quantum tunneling at low temperatures. The obtained inelastic neutron scattering spectra display a number of low energy peaks having the characteristic temperature and neutron momentum transfer dependence of quantum tunneling. The momentum distribution of the water protons in beryl measured with deep inelastic neutron scattering (DINS) shows oscillations or interference fringes, revealing that in the ground state the water protons are coherently delocalized over the possible positions in the channel. The calculated from DINS data average proton kinetic energy, EK=106 meV, is much smaller than in bulk water (~150 meV), in spite of the observed highest possible internal O-H stretching modes. A combination of ab initio density functional theory and path integral molecular dynamics simulations of beryl shows that water protons are laterally spread over 1.2 A, supporting the neutron scattering data. The calculations also demonstrate water duality in beryl: for some time it stays trapped/localized with the structure similar to almost free water molecule, then it shows a phonon-assisted tunneling. In the tunneling state water exhibits a new state of molecule, which has significantly delocalized protons and consequently small proton kinetic energy.

Studies of the glass transition usually neglect any quantum effects in structural relaxation. However, our analysis shows [1] that liquids of light molecules with low Tg might have significant contribution of quantum effects. They should lead to a significant decrease of T_g with respect to the melting temperature T_m, and can lead to a sub-Arrhenius temperature dependence of structural relaxation time (instead of usual super-Arrhenius behavior). Combining neutron scattering and dielectric relaxation spectroscopy we show that quantum fluctuations are not negligible in deeply supercooled bulk water obtained as Low-Density Amorphous (LDA) or vapor deposited form [2,3]. We suggest that quantum tunneling might be the origin of its unusually low fragility, m~14 [2,3]. The discovered anomalously large isotope effect in Tg of water [2] is consistent with the quantum effects dominating structural relaxation of water at these temperatures. However, analysis of dynamics in strongly confined water revealed reasonable fragility, m~18-19, and much weaker isotope effect. These results suggest that quantum fluctuations in confined water lead to the reduction of the activation energy barrier (due to zero-point energy), but no tunneling. Thus our studies demonstrate that indeed quantum effects are not negligible for dynamics of water molecules at low temperatures.
1. V. N. Novikov, A. P. Sokolov, PRL110, 065701, (2013).
2. C. Gainaru, et al, PNAS 111, N49, 17402 (2014).
3. A. L. Agapov, et al., Phys. Rev. E 91, 022312 (2015).

Water confined on a scale of 20 Å has been shown through Debye Waller, INS and Neutron Compton Scattering measurements to be in a non-molecular ground state in which protons are coherently delocalized over a distance of .2-.3 Å. The electron distribution of the water has been shown, using X-ray Compton scattering, to be anomalous as well. The scale of changes in the electron momentum distribution is 17 times that due to the disordering of the hydrogen bond as water is heated from 5^oC to 95^o C. This state persists up to room temperature.
We will describe the experiments that led to the recognition of this state and what is known of its proton transport and electronic excitation properties. In as much as 20Å is the scale of the distance between the elements of biological cells, this anomalous state of water is the one that is relevant to the cells functioning.

Polymers are a class of materials that are increasingly utilized in energy generation and conversion, energy storage and lightweight materials, chemical separations, gas and liquid purification. Because their properties depend sensitively upon intra- and inter-polymer chain interactions, new methods to manipulate and understand these interactions are continually being pursued. In this talk I will give a brief overview of how large-scale simulations of polymer structure and dynamics can be used in concert with experimental synthesis and characterization to develop improved understanding of the underlying interactions, structural and dynamical processes to enable subsequent rational materials design/optimization. Emphasis will be on understanding how/if isotopic substitution (deuterium) modifies structure, morphology, and optoelectronic properties of model conjugated polymer composites with an integrated approach that combines the synthesis of deuterated materials, optoelectronic properties measurements, theoretical simulations, and neutron scattering.

Using ternary blends of polystyrene (PS), poly(methyl methacrylate) (PMMA), and Janus particles (JPs) with symmetric PS and PMMA hemispheres, we demonstrate the stabilization of dispersed and bicontinuous phase-separated morphologies by the interfacial adsorption of Janus particles during demixing upon solvent removal. The resulting blend morphology could be varied by changing the blend composition and JP loading. Increasing particle loading decreased the size of phase-separated domains, while altering the mixing ratio of the PS/PMMA homopolymers produced morphologies ranging from PMMA droplets in a PS matrix to PS droplets in a PMMA matrix. Notably, bicontinuous morphologies were obtained at intermediate blend compositions, marking the first report of highly continuous domains obtained through demixing in a blend compatibilized by Janus particles. The JPs were found to assemble in a densely packed monolayer at the interface, allowing for the stabilization of bicontinuous morphologies in films above the glass transition temperature by inhibiting coarsening and coalescence of the phase-separated domains. The rate of solvent evaporation from the drop-cast films and the molecular weights of the homopolymers were found to greatly affect blend morphology.

Conjugated polymer films, nanofibers, and networks can be ideal materials for the design of the next generation of efficient photovoltaic devices, batteries, thermoelectric cells, light emitting diodes and other technologies. It is also recognized that the structure and dynamics of organic semiconductor materials correlate strongly with large changes in optical, electronic and mechanical properties so that their control and manipulation is essential to the advancement of the field. This presentation outlines the use of neutron scattering techniques in the development of structure-property relationships for conjugated polymer nanomaterials. We also highlight how quasi-elastic neutron scattering (QENS) is used along with MD simulations to understand the nature of nanoscale motions and to optimize performance metrics in these materials. Future prospects for using nanostructured phases of organic semiconductors in energy applications will also be discussed.

The introduction of proteins into organic solvent systems provides new avenues that allow for the development of next generation technologies and chemistries. Unfortunately, preservation of protein structure and functionality remains a restrictive factor. By analyzing the surface of model proteins, complimentary random heteropolymer surfactants can be developed that optimizes protein-surfactant interactions and provides enhanced stabilization in nonpolar media. Our recent studies showed that rationally designed random copolymer surfactants provides prolonged retention of protein activity in organic solvent in comparison to both a small molecule surfactant and amphiphilic block copolymers. The modularity of this technique allows us to adapt random heteropolymer surfactants for proteins that hold more pertinent functionality, such as proteins for medical therapeutics, functional tissue scaffolds, catalysis, and biomolecular sensors.

We will discuss (a) the interplay between RNA and Mg²⁺ at atomic resolution¹ through explicit solvent molecular dynamics simulations of SAM-I riboswitch. We find that glass-like magnesium influences RNA tertiary structure dynamics. Simulations reveal that outer sphere coordinated Mg²⁺ ions fluctuate on the same time scale as the folded RNA. The outer sphere Mg²⁺ ions responsible for the effects account for 80% of the Mg²⁺ in the simulations. Preliminary diffusion analysis suggests transient outer sphere Mg²⁺ dynamics are glassy; (b) the characteristics of glassy state in a protein. Due to coupling of solvent and protein dynamics, the slope of the Debye-Waller factor as a function of temperature is negative. We predict the slope of the Debye-Waller factor using a generalization of Adam-Gibbs model. We will compare the results with molecular dynamics simulations of the Debye-Waller factor of ice with different proton-disordered configurations.
1. In collaboration with R. Hayes, J. Noel, P. Whitford, S. Hennelly, J. N. Onuchic, and K. Sanbonmatsu.

There are recent research interests in colloidal suspensions with both a short-range attraction (SA) and a long-range repulsion (LR). These competing interaction features are ubiquitous for charged protein solutions at low salt conditions. Using purified lysozyme proteins, we have investigated concentrated protein solutions at a very wide range of concentrations and temperatures. The localized glassy like behavior is observed at the short time scale for highly concentrated solutions at low temperatures using a neutron spin echo while the long time properties of the same samples are found to behave like Newtonian fluid for the investigated range of the share rates using a microcapillary viscometer. The inter-protein structure investigated by small angle neutron scattering revealed that this interesting behavior is due to the intermediate range order (IRO) caused by the competition of the SA and LR interaction. The IRO causes the localized heterogeneous density distribution only at the intermediate range length scale and the same samples are uniform at larger length scales. This is distinct from commonly studied colloidal gel systems where the heterogeneous density distribution tends to extend to the whole system. Our results demonstrated the importance of the structure at the IRO length scale on gelation/glass transitions.

The compression of polyelectrolyte microgel particles in a salt-free highly compressed colloid due to osmotic pressure outside of the particles due to counterions located there is studied for a model based on a quasi-analytic solution of the Poisson-Boltzmann equation and a model for the gel elasticity based on counterion osmotic pressure inside the particles and polymer elasticity (of entropic origin). It is found that for particles of radius of the order of a tenth of a micron, the counterion osmotic pressure should play a significant role in the compression of the particles, especially particles which do not have a corona (i.e., nonlinked polymer chains attached to their surface). The presence of a corona of monomer density smaller than that of the core of the microgel reduces the contribution of the osmotic pressure due to counterions outside of the microgel. It is also demonstrated that counterion osmotic pressure outside the particles can provide a significant contribution to the lubrication of the interface between the particles and a surface along which the compressed colloid is made to slide, for sufficiently slow velocities.

I will summarize recent results from computer simulations addressed to the important issues related to the connection between the Widom lines and the dynamical crossovers in water.
The Widom line is the line of convergence of the maxima of the thermodynamic response functions upon approaching the critical point. The characterization of this line is very important both in the supercooled and in the supercritical state and it can be found upon approaching respectively the liquid-liquid and the liquid gas critical points.
I will show how in the supercritical state this line is connected to a dynamic crossover from a liquid-like to a gas-like behaviour of the transport coefficients both in experiments and in simulations of many all atoms popular water models [1,2].
In the supercooled state the well known dynamic fragile to strong transition is a feature of many glass forming systems. The fragile behaviour can be interpreted by means of the Mode Coupling Theory.
I will show that in supercooled water this dynamic crossover can also be connected to the crossing of the Widom line not only in the bulk ST2 and TIP4P models of water [3,4] but also in experiments and simulations in confined water [3,5,6], in simulations of aqueous solutions of electrolytes [7] and in hydrophobic aqueous solutions based on a primitive model [8].
Both aqueous solutions and confinement are important to shed light on the behaviour of water in the supercooled region as they ease supercooling and still retain, up to a certain extent, the phenomenology of the bulk.
Finally I will show an analysis of the appearance of hopping phenomena in bulk supercooled water both in TIP4P and TIP4P/2005 [9]. In particular I will analyze the connection of hopping to the crossing of the Widom line and to the fragile to strong transition.
[1] P. Gallo, D. Corradini and M. Rovere, Nat. Commun. 5:5806 doi: 10.1038/ncomms6806 (2014).
[2] D. Corradini, M. Rovere and P. Gallo in preparation (2015)
[3] L. Xu, P. Kumar, S.V. Buldyrev, S.-H. Chen, P.H. Poole, F. Sciortino, H.E. Stanley, Proc. Natl Acad. Sci. USA 102, 16558 (2005).
[4] P. Gallo and M. Rovere, J. Chem. Phys., 137 , 164503 (2012).
[5] L. Liu, S.-H. Chen., A. Faraone, C.-W. Yen, C.-Y. Mou, Phys. Rev. Lett. 95, 117802 (2005).
[6] P. Gallo, M. Rovere and S.-H. Chen J. Phys. Chem. Lett. 1, 729 (2010).
[7] P. Gallo, D. Corradini and M. Rovere, J. Chem. Phys. 139, 204503 (2013).
[8] D. Corradini, P. Gallo, S.V. Buldyrev and H.E. Stanley, Phys. Rev. E 85, 051503 (2012).
[9] M. De Marzio, M. Rovere, P. Gallo in preparation (2015).

The OH-stretch vibration is an excellent probe of local environment in water primarily due to the sensitivity of its vibrational frequency to local hydrogen-bonding condition. In the literature, the OH-stretch vibrational spectroscopy is dominated by IR and Raman, which are determined not only by the vibrational density of states (VDOS) but also by the optical selection rules (e.g., transition dipole for IR). An alternative technique, inelastic incoherent neutron scattering (IINS), has been recently applied to estimate the OH-stretch VDOS for many condensed phases of water due to the availability of high-energy neutron sources. In this work, we have extended the mixed quantum/classical approach, originally developed by Skinner and co-workers to simulate OH-stretch IR and Raman spectra, to the calculation of IINS spectra with small k values for ambient and supercooled water and ice Ih. The agreement between theory and experiment in the limit k=0 is reasonable. We also further show that the differences between IR and IINS for neat H2O water are due to the optical selection rule (i.e., transition dipoles), as well as the coherent contribution from distinct OH stretches in IR.

The fast secondary relaxation in aqueous systems and other liquids is understood, even if somewhat simplistically, as rattling in the transient cage of the nearest neighbors on the shorter time scale followed by the main cage-breaking structural relaxation on the longer time scale. The spatial characteristics of this fast relaxation can be probed using inelastic neutron scattering [1] with the characteristic energy resolution provided by cold time-of-flight or backscattering neutron spectrometers. Adopting the same methodological approach, we use much higher energy resolution neutron spin-echo spectrometry to investigate the slow secondary relaxation (or excess wing) in aqueous systems. We have found [2] that at sufficiently low temperatures water molecules in an aqueous solution engage in the center-of-mass dynamics that are different from both the main structural relaxation and the aforementioned fast structural relaxation. While the fast relaxation motions are known to take place on the picosecond time scale and Angstrom length scale, the slower motions that we have investigated are observed on the nanosecond time scale and nanometer length scale [3]. They are associated with the slow secondary relaxation, or excess wing dynamics, in glass-forming liquids. Our approach, utilizing the unique neutron scattering strength of the signal dependence on the scattering momentum transfer, can be applied to probe the characteristic length scale of the dynamic entities associated with slow dynamics in glass-forming liquids, which presently cannot be studied by other experimental techniques.
[1] J. Qvist, H. Schober, and B. Halle, J. Chem. Phys. 134, 144508 (2011).
[2] E. Mamontov and M. Ohl, Phys. Chem. Chem. Phys.15, 10732 (2013).
[3] E. Mamontov, P. Zolnierczuk, and M. Ohl, Phys. Chem. Chem. Phys.17, 4466 (2015).

Ice formation is ubiquitous in nature, with important consequences in a variety of systems and environments, including biological cells, soil, aircraft, transportation infrastructure and atmospheric clouds. However, its intrinsic kinetics and microscopic mechanism are difficult to discern with experiments. Molecular simulations of ice nucleation are also challenging, mostly because of the slow structural relaxation of supercooled water, and because of the large nucleation barriers that separate the supercooled liquid and crystalline basins. Indeed, direct calculations of nucleation rates have only been performed [1-3] for mW [4], a monoatomic coarse-grained model of water. For the more realistic molecular models such as the TIP4P family, only indirect estimates have been obtained by assuming the validity of the classical nucleation theory (CNT) [5], and the problem of directly computing the nucleation rate for a molecular model is generally considered as one of the most challenging open problems in computational studies of water. Here, we use a path sampling approach known as forward-flux sampling [6] to perform the first direct calculation of the homogeneous nucleation rate. We use the TIP4P/Ice model [7], which is the most accurate water model for studying ice polymorphs. By using a novel topological order parameter for distinguishing different polymorphs, we are able to identify a freezing mechanism that involves a competition between cubic and hexagonal ice polymorphs in the early stages of nucleation. In this competition, the cubic polymorph takes over since the addition of new topological structural motifs consistent with cubic ice leads to the formation of more compact crystallites. This is not true for topological hexagonal motifs that give rise to elongated crystallites that are not able to grow. This leads to transition states that are rich in cubic ice, and not the thermodynamically stable hexagonal polymorph [8].
[1] Li, Donadio, et al., Phys. Chem. Chem. Phys. 13, 19807 (2011).
[2] Li, Donadio, et al., Nat. Comm. 4, 1887 (2013).
[3] Haji-Akbari, DeFever, et al., arXiv:1409.0953.
[4] Molinero, Moore, J. Phys. Chem. B 113, 4008 (2009).
[5] Sanz, Vega, et al. J Am. Chem. Soc. 135, 15008 (2013).
[6] Allen, Frenkel, et al., J. Chem. Phys. 124, 194111 (2006).
[7] Abascal, Sanz, et al., J. Chem. Phys. 122, 234511 (2005).
[8] Haji-Akbari, Debenedetti, in press Proc. Natl. Acad. Sci. (arXiv: http://arxiv.org/abs/1505.01126).

I will discuss a simple extension of the well-known ST2 model for water that allows for a continuous modification of the hydrogen bond angular flexibility. I will show that the bond flexibility affects the relative thermodynamic stability of the liquid and of the hexagonal (or cubic) ice. On increasing flexibility, the liquid-liquid critical point, which in the original ST2 model is located in the no-man's land (i. e. the region where ice is the thermodynamically stable phase) progressively moves to a temperature where the liquid is more stable than ice. The study definitively proves that the liquid-liquid transition in ST2 is a genuine phenomenon, of high relevance in all tetrahedral network-forming liquids, including water.
Related References:
Frank Smallenburg, Laura Filion and Francesco Sciortino
Erasing no-man's land by thermodynamically stabilizing the liquid-liquid transition in tetrahedral particles Nature Physics 27 JULY 2014 | DOI: 10.1038/NPHYS3030.
Frank Smallenburg and Francesco Sciortino Sciortino
Tuning the Liquid-Liquid Transition by Modulating the Hydrogen-Bond Angular Flexibility in a Model for Water Phys. Rev. Letts. 115, 015701 (2015) DOI: 10.1103/PhysRevLett.115.015701

The ST2 interaction potential has been used in a large number of simulation studies to explore the possibility of a liquid-liquid phase transition (LLPT) in supercooled water. A comprehensive phase diagram of the ST2 model has recently been determined [1]. Here, we present simulation results on ST2 water in which we evaluate the free energy of ice formation in the vicinity of the Widom line [2], as well as approaching the critical point of the LLPT itself [3]. Our results allow us to draw conclusions about the stability of the metastable liquid state near the Widom line, and to discern how the thermodynamic anomalies that extend outward from the LLPT may influence ice nucleation. Using a simple lattice model, we also discuss in more general terms how the nucleation barrier behaves in a metastable system in the presence of fluctuations to both another metastable phase as well as to the stable phase, with the aim of improving our understanding of how to use ice nucleation as a "probe" to test for the presence or absence of a LLPT in experiments on supercooled water [4].
[1] F. Smallenburg, P.H. Poole, and F. Sciortino, Molecular Physics, in press (2015).
[2] C.R.C. Buhariwalla, R.K. Bowles, I. Saika-Voivod, F. Sciortino, and P.H. Poole, Eur. Phys. J. E 38, 39 (2015).
[3] See poster of S. Morris, et al., this meeting.
[4] See poster of D. James, et al., this meeting.

The physical properties of supercooled water have been a source of continued interest since the pioneering work of Speedy and Angell, who reported sharp increases in the response functions upon isobaric cooling [1]. One intriguing hypothesis that has been formulated to explain this behavior is the existence of a metastable liquid-liquid transition at deeply supercooled conditions [2]. The preponderance of experimental evidence is consistent with this hypothesis (e.g., [3], [4]), although no definitive proof exists to date. Computational studies have played an important role in this area [2], [5]-[13]. State-of-the-art free energy techniques provide clear evidence of a liquid-liquid transition in the ST2 model [14] of water [15], including the identification of three phases at the same, deeply supercooled thermodynamic conditions: two metastable liquids in equilibrium, and a stable crystal [15]. Recent calculations on tunable tetrahedral models [16, 17] provide definitive confirmation of the free energy analysis [15].
A coarse-grained implementation of the forward-flux sampling technique [18] allows the direct calculation of the rate of homogeneous ice nucleation in the TIP4P/Ice model of water [19] and provides new insights into the mechanism of ice formation at deeply supercooled conditions [20]. The early stages of nucleation are characterized by a competition between cubic and hexagonal ice. Transition states are rich in cubic ice, rather than in the thermodynamically stable hexagonal ice.
Modern computational methods provide important insights into the thermodynamics and kinetics of deeply supercooled water, complementing ongoing experimental studies.
[1] Speedy, R.J., Angell, C.A. J. Chem. Phys., 65, 851 (1976).
[2] Poole, P.H., Sciortino, F., Essmann, U., Stanley, H.E. Nature, 360, 324 (1992).
[3] Mishima, O., Stanley, H.E. Nature, 392, 164 (1998).
[4] Amann-Winkel, K., Gainaru, C., Handler, P.H., Seidl, M., Nelson, H., Böhmer, R., Loerting, T. PNAS, 110, 17720 (2013).
[5] Liu, Y., Panagiotopoulos, A.Z., Debenedetti, P.G. J. Chem. Phys., 131, 104508 (2009).
[6] Moore, E.B., Molinero, V. Nature, 479, 506 (2011).
[7] Limmer, D.T., Chandler, D. J. Chem. Phys., 135, 134503 (2011).
[8] Liu, Y., Palmer, J.C., Panagiotopoulos, A.Z., Debenedetti, P.G., J. Chem. Phys., 137, 214505 (2012).
[9] Poole, P.H., Bowles, R.K., Saika-Voivod, I., Sciortino, F. J. Chem. Phys., 138, 034505 (2013).
[10] Overduin, S.D., Patey, G.N. J. Chem. Phys., 138, 184502 (2013).
[11] Limmer, D.T., Chandler, D. J. Chem. Phys., 138, 214504 (2013).
[12] Kesselring, T.A., Lascaris, E., Franzese, G., Buldyrev, S.V., Stanley, H.E. J. Chem. Phys., 138, 244506 (2013).
[13] Li, Y.P., Li, J.C., Wang, F. PNAS, 110, 12209 (2013).
[14] Stillinger, F.H., Rahman, A. J. Chem. Phys., 60, 1545 (1974).
[15] Palmer, J.C., Martelli, F., Liu, Y., Car, R., Panagiotopoulos, A.Z., Debenedetti, P.G. Nature, 510, 385 (2014).
[16] Smallenburg, F., Fillon, L., Sciortino, F. Nature Phys., 10, 653 (2014).
[17] Smallemburg, F., Sciortino, F. Phys. Rev. Lett., in press (2015).
[18] Allen, R.J., Warren, P.B. ten Wolde, P.R. Phys. Rev. Lett., 94, 018104 (2005).
[19] Abascal, J.L.F., Sanz, E., Fernández, R.G., Vega, C. J. Chem. Phys., 122, 234511 (2005).
[20] Haji-Akbari, A., Debenedetti, P.G. PNAS, in press (2015).

It has been determined by experiments and simulations that a ubiquitous route to the formation of physical gels involves the kinetic arrest of the process of phase separation of homogeneous fluids quenched inside its thermodynamically unstable ‘spinodal&’ region [1]. Understanding the fundamental basis of this process of arrested spinodal decomposition and its connection with the formation of (repulsive and attractive) glasses is a relevant challenge, which can only be faced from the perspective of a non-equilibrium statistical thermodynamic theory. The Non-equilibrium Self-consistent Generalized Langevin Equation (NE-SCGLE) theory of irreversible relaxation [2] has been shown to provide a detailed theoretical scenario of the equilibration and aging processes in glass-forming liquids [3]. In this work we describe our recent advances in the application of the same theoretical framework to the description of the processes of arrested spinodal decomposition [4]. Among the main features of the resulting scenario is the prediction that the spinodal curve, besides being the threshold of thermodynamic stability of homogeneous states, is also the borderline between the regions of ergodic and non-ergodic homogeneous states. The theory also predicts the existence of a glass-glass transition line buried inside the low-temperature part of the non-ergodic region, which is the continuation to the interior of the spinodal, of the “ordinary” liquid-glass transition line that extends to higher densities to connect with the hard-sphere glass transition. Within the region bounded from below by this low-temperature glass-glass transition and from above by the spinodal dynamic arrest line we can recognize two kinetically distinct regions: the upper, shallow-quench region, where the kinetics of the arrest process is slow enough to allow for full phase separation, and the deeper-quench region, whose faster dynamic arrest leads to the formation of gel structures by interrupted spinodal decomposition.
[1] E. Zaccarelli, J. Phys.: Condens. Matter 19, 323101 (2007); [2] P. Ramírez-González and M. Medina-Noyola, Phys. Rev. E 82, 061503 (2010); [3] L.E. Sánchez-Díaz, P. Ramírez-González and M. Medina-Noyola, Phys. Rev. E 87, 052306 (2013); [4] J. M. Olais-Govéa, L. Loacute;pez-Flores, and M. Medina-Noyola, arXiv:1505.00390[cond-mat.soft] (2015).

Science strategy in the 21^st century is based around understanding tomorrow&’s world by deconstructing complexity. In this quest, neutrons continue to play an important role as they are able to give key insights into the properties and behaviour of complex systems. In this talk I will concentrate on examples that illustrate the possibilities offered by a combination of neutron techniques for both applied and fundamental science cases: organic (polymer-based) electronics and bio-systems.

Polymer electrolytes have potential for use in next generation lithium and sodium batteries. Replacing the liquid electrolyte currently used has several advantages: it allows use of high energy density solid lithium as the anode, removes toxic solvents, improves safety, and eliminates the need for heavy casings. Despite their advantages, the conductivity of solid polymer electrolytes is not sufficient for use in batteries. As a result, considerable effort towards improving conductivity and understanding mechanisms of lithium transport has taken place over the last 30 years. This talk considers the interplay of conductivity, crystallinity, local coordination and polymer dynamics in solid polymer electrolytes. Using a combination of experimental and computational technique, we propose the possibility of high charge mobility using ion aggregates and percolated nanofiller networks.

The structural changes in liquid ⁴He through the lambda transition were previously studied by neutron scattering using 2-axis spectrometer. The temperature dependence of the pair distribution function indicated a decrease in the peak corresponding to nearest distance. This behavior was interpreted that fraction of the atoms occupied uniformly condensate as in ideal Bose gas and did not contribute to spatial correlations. These experiments probed so called “static structure function” - the integrated over all energies S(Q,w). To clarify the physical meaning of this delocalization we carried out energy resolved study of the atomic correlation function as a function of temperature using the dynamic PDF approach. In this method correlation function is obtained for different energy ranges and can be used to study atomic dynamics through the transition. Inelastic neutron scattering measurements were carried out at the CNCS spectrometer at SNS/ORNL using three incident energies 3.63 , 15 and 50 meV in the temperature range from 2.85 to 1.82 K. The normalized data were processed to obtain structure function at given energy slice and then Fourier transformed to obtain corresponding dynamic PDF. The results suggested that below the lambda transition there was a shift in the correlation intensity to lower energies and shorter interatomic distances.

Single-chain technologies are a new paradigm for the development of future soft, smart nano-devices [1]. Single-chain nano-particles (SCNPs) can be envisioned as a key element of this innovative field. Current SCNPs are unimolecular soft nano-objects consisting in individual polymer chains collapsed to a certain degree [2]. Many of the potential applications of SCNPs relay on their particular molecular architecture. Claims and drawings about the compact globular structure of SCNPs in solution are still common in the literature. However, we provide SANS and SAXS results --supported by MD-simulations-- indicating that this is not the case in solution [3], where SCNPs&’ morphology resembles that of intrinsically disordered proteins. Furthermore, SCNPs act as ‘fillers&’ in all-polymer nano-composites, if they are dispersed within a linear chain polymer matrix. There, the conformation of the SCNP as revealed by SANS is rather compact, similar to that of a crumpled globule [4]. In addition, we have investigated by QENS (backscattering, time-of-flight and neutron spin echo) the dynamics of isotopically labelled nano-composites where SCNPs based on PMMA [poly(methyl methacrylate)] are mixed with PEO [poly(ethylene oxide)] [5,6,7]. Here, we focus on the PEO dynamics in a nano-composite with PEO as majority component [5,7]. Compared to the bulk behavior, we find (i) a slowing down of the Rouse dynamics and (ii) much more pronounced long-time decays of the single chain dynamic structure factor. These translate in a striking increase of the tube diameter representing the entanglement confinements, namely from 5.3 nm in bulk PEO to about 9 nm in the nano-composite. For comparison, we have also studied a blend of PEO with the PMMA component present as linear chains (i. e., as precursors of the SCNPs prior to the internal cross-linking process). While the Rouse dynamics is rather similar for PEO in the linear blend and in the nano-composite, the tube diameter for PEO in the linear blend is, within the uncertainties, the same as in bulk. This implies that the entanglement dilution has to be induced by the particular topology of the SCNPs. We emphasize that this is the first direct experimental observation of entanglement tube dilation in polymer nano-composites. We explore the influence of topology and dynamics of the SCNPs with the aid of coarse-grained MD- simulations.
[1] M. Ouchi, N. Badi, J. F. Lutz, M. Sawamoto, Nat. Chem. 3, 917 (2011).
[2] A. Sanchez-Sanchez, I. Perez-Baena, J. A. Pomposo, Molecules 18, 3339 (2013).
[3] J. A. Pomposo, I. Perez-Baena, F. Lo Verso, A. J. Moreno, A. Arbe, J. Colmenero, ACS Macro Letters 3, 767 (2014).
[4] J. Colmenero et al, in progress.
[5] D. Bhowmik, J. A. Pomposo, F. Juranyi, V. García-Sakai, M. Zamponi, Y. Su, A. Arbe, J. Colmenero, J., Macromolecules 47, 304 (2014).
[6] D. Bhowmik J. A. Pomposo, F. Juranyi, V. García-Sakai, M. Zamponi, A. Arbe, J. Colmenero, Macromolecules 47, 3005 (2014).
[7] A. Arbe et al, in progress.

To determine the nature of excellent glass-forming ability (GFA) of alloys, it is desirable to investigate crystallization kinetics in their supercooled liquid region [1]. Although considerable amount of works have been reported, most of them were carried out on multicomponent alloys with complex chemistry [2]. In order to access the physics of crystallization, it is essential to simplify the chemistry [3]. In this paper we report a time-resolved neutron diffraction study of crystallization kinetics in five ternary supercooled metallic liquids (SMLs), Zr₅₆Cu₃₆Al₈, Zr₅₄Cu₃₈Al₈, Zr₄₈Cu₄₅Al₇, Zr₄₅Cu₄₉Al₆ and Zr₄₆Cu₄₆Al₈, with gradually enhanced GFA. The measurements were carried out using NOMAD at the Spallation Neutron Source, Oak Ridge National Laboratory (ORNL)_. The structural evolution in short- and medium- range order (SRO and MRO) during the amorphous-to-crystalline transformation for these SMLs has been well resolved with a time-resolution of ~ 2 minute. According to the property of their crystalline products and their crystallization pathways, these liquid metals can be divided into two types: Cu₁₀Zr₇-type and Zr₂Cu-type. For Zr₂Cu-type liquid metals, i.e. Zr₅₆Cu₃₆Al₈ and Zr₅₄Cu₃₈Al₈, there is just one crystallization temperature (e.g. ~751 K for Zr₅₆Cu₃₆Al₈) on their Differential Scanning Calorimetry (DSC) curves, and total neutron diffraction patterns show highly-ordered crystalline phase after divitrification. For Cu₁₀Zr₇-type liquid metals, i.e. Zr₄₈Cu₄₅Al₇, Zr₄₅Cu₄₉Al₆ and Zr₄₆Cu₄₆Al₈, two crystallization temperatures (e.g. ~ 776 K and ~ 778 K for Zr₄₆Cu₄₆Al₈) were identified, much like that in Zr_52.5Cu_17.9Ni_14.6Al₁₀Ti₅ alloy (BAM-11) [2], an excellent glass former. The crystalline phase of devitrified Cu₁₀Zr₇-type liquid metals, with better GFA, is poorly ordered, also similar to BAM-11. These results suggest that the unique crystallization pathway for Cu₁₀Zr₇-type SMLs would be a distinctive feature of alloys with excellent GFA.
[1] P.G. Debenedetti, F.H. Stillinger. Supercooled liquids and the glass transition. Nature 410, 259, 2001.
[2] X. L. Wang, J. Almer, C. T. Liu, et al.. In situ synchrotron study of phase transformation behaviors in bulk metallic glass by simultaneous diffraction and small angle scattering. Physical Review Letters 91: 265501, 2003.
[3] S. Lan, X. Wei, J. Zhou, Z. Lu, X. Wu, M. Feygenson, J. Neuefeind, X.-L. Wang. In-situ study of crystallization kinetics in ternary bulk metallic glass alloys with different glass forming abilities. Applied Physics Letters 105, 201906, 2014.

Glass-forming metallic alloys are of wide interest for both practical applications and fundamental studies of structure-property relationships in the liquid and amorphous solid states. Metallic liquids, which exhibit short- and even medium range order (SRO and MRO), serve as interesting model systems for understanding the evolution of structural order and dynamics of the supercooled liquid. There is great interest in understanding the relationship between chemical and topological structure of the liquid, local atomic dynamics, bulk dynamical properties, and glass-forming ability (GFA).
One major obstacle to experimental studies of these high-temperature alloy liquids is constructing a furnace with a suitable sample container. Containerless sample environments, however, address several challenges. They offer a chemically pure environment by eliminating possible reactions with a crucible, reduce scattering effects from the sample environment, and remove heterogeneous nucleation sites that hinder access to the supercooled liquid regime. We present results from some of the first neutron scattering studies using the Neutron ElectroStatic Levitator (NESL), a containerless sample environment optimized for in situ scattering studies at the Spallation Neutron Source (SNS) located at Oak Ridge National Laboratory (ORNL). Combined X-ray and neutron scattering studies of glass-forming, Zr-based liquids and plans for future inelastic neutron scattering measurements are discussed.

Recently, it has been theoretically suggested that quantum fluctuations and topological features may induce glassy states, that we call a spin jam, in defect-free spin systems.[1] Very recently our experimental results show the evidence for existence of a spin jam state in SrCr_9pGa_12-9pO₁₉ (SCGO(p)) in which the magnetic Cr³⁺ (3d³) ions form distorted quasi-two-dimensional triangular lattice of bi-pyramids.[2] Most recently, we successfully synthesized a new isostructural system BaCr_9pGa_12-9pO₁₉ (BCGO(p)) with , which is probably another system associated with a spin jam state. Here, we present the results of structure, bulk magnetic susceptibility and neutron scattering of BCGO(p). Neutron powder diffraction measurements and refinements show that BCGO(p) is isostructural as SCGO(p) but with a much larger c lattice constant indicating that BCGO(p) is more two dimensional. BCGO(p) exhibits similar glassy behaviors as SCGO(p) but with higher freezing temperature T_f. Very high Curie-Weiss temperature and frustration index were also observed in BCGO(p) suggesting that BCGO(p) is one of the most frustrating system. From inelastic neutron scattering measurement, we observed dispersionless magnetic excitations, which arise from 4f_vi-4f_vi spin dimers, at an energy of 16.5(1) meV. Similar as SCGO(p),[3] the 4f_vi-4f_vi spin dimers in BCGO(p) are adjacent spins in neighboring triangular lattice planes which separate two-dimensional sublattices of interacting kagomé-triangle-kagomé layers. These results indicate that BCGO(p) is another good candidate for observation of spin jam state.
References:
[1] I. Klich, S.-H. Lee and K. Iida, From frustration to glassiness via quantum fluctuations and random tiling with exotic entropy, Nature Comm. 5, 3497 (2014).
[2] J. Yang, A. M. Samarakoon, S. E. Dissanayake, H. Ueda, I. Klich, K. Iida, D. Pajerowski, N. Butch, Q. Huang, J.R.D. Copley, S.-H. Lee, A spin jam state of a frustrated magnet, submitted (2014).
[3] S.-H. Lee, C. Broholm, G. Aeppli, T. G. Perring, B. Hessen, ans A. Taylor, Isolated Spin Pairs and Two-Dimensional Magnetism in SrCr_9pGa_12-9pO₁₉. Phys. Rev. Lett. 76, 4424-4427 (1996).

In spite of their distinctively different material properties, RNA and Nanodiamond (ND) both have applications in nano-biotechnology. The architecturally flexible RNA with catalytic functionality forms nanocomposites that can treat life-threatening diseases like cancer. The non-toxic ND has excellent mechanical and optical properties and functionalizable high surface area and thus is actively considered for bactericidal materials and cancer treatment. Therefore, integrating the attractive properties of these two organic and inorganic molecules into a nanocomposite is a rational approach toward developing a potential new generation, in vivo, drug-delivery system. The enhanced material properties can only be achieved if and only if the fundamental understanding of the structural and dynamical behavior of this system at the nanoscale is fully understood. To understand the fundamentals, we used quasi-elastic neutron scattering (QENS) and Molecular Dynamics (MD) simulation techniques to compare the dynamics of freestanding RNA with the dynamics on the ND surface. We observed faster RNA motion on the ND surface compared to a ‘freestanding&’ RNA. Large-scale MD simulations helped understand the underlying physics for the faster dynamics; the faster RNA motion in the nanocomposite is due to de-confinement of RNA on the ND surface. The freestanding RNA shows a “jammed” phase with heterogeneous dynamics that is responsible for the slow motion of the ‘freestanding&’ RNA. We also use our in-house workflow to optimize the force-field parameters that helps model the RNA-ND system at exact experimental environment.

We present results of packing and assembly simulations of hard particles, as well as simulations of nucleation and growth of crystals from oscillating pair potentials, to address the problem of the glass transition. We construct the densest packing and ground state landscapes and compare ground state structures to finite density and finite temperature fluids and their assemblies. Through this comparison, we provide an alternative view of the underlying cause of the glass transition in soft matter systems.

Surfactants and block copolymers can form liquid crystals. The commonality among these mesogens is the existence in the same molecule of incompatible groups attached by a chemical bond. Here we demonstrate that binary mixtures of spherical particles that interact through a combination of simple isotropic one-length scale interactions can form the same mesophases as surfactants and block copolymers. We present the phase diagram of the mesogenic mixtures and the regions of stability of the lamellar, gyroid and hexagonal liquid crystal phases, and provide insights on the mechanisms of nucleation of these phases.

The talk will describe nonequilibrium molecular dynamics (MD) simulations of flow in disordered systems, following the evolution of stress-strain curves, conformational changes and spatial correlations as systems are quenched from simple fluids through the glass transition and to athermal solids. First the shear-thinning response of simple binary Lennard-Jones fluids, short chain molecules and the lubricant squalane are followed over more than 5 orders of magnitude in strain rate while the system is driven through the glass transition by varying temperature or pressure. Changes in molecular orientation and positional correlations are correlated with changes in viscosity. When the full range of rates is considered one finds inconsistencies with simple time-temperature superposition or power law shear-thinning. The low rate response is qualitatively consistent with Eyring type models with Newtonian behavior crossing over to an activated logarithmic rate dependence as rate increases. The results are important to models of elastohydrodynamic lubrication and are compared to high rate and pressure experiments on squalane. The second part of the talk will describe the extreme athermal limit where the system evolves through mechanical instability rather than thermal activation. The stress and correlation length exhibit power law scaling with strain rate that is related to a power law distribution of avalanches in the quasistatic limit.The connection to the high rate response near the glass transition will be discussed as well as anomalous diffusion due to power law spatial correlations.

A coordinated study is carried out to investigate the microscopic origin of viscosity in soft matter. We focus on charged colloidal systems with strong inter-colloidal screened Coulomb interactions to elucidate the flow mechanism and the origin of viscosity in terms of the colloid-colloid interactions through a combination of neutron scattering and computer simulations. Both affine and nonaffine flows are observed by Brownian dynamics simulations, with greater development of nonaffine deformation as shear rate increases. The average size of the affine deformed zone obtained from small angle neutron scattering is correlated with the nonlinear rheological responses to the applied shear in the shear thinning regime. This research leads to an understanding of the microscopic perspective of the origin of viscosity, allowing this property to be controlled in the design of novel soft materials.

We study the form of the velocity and strain rate fields in simlpe particle-based models of soft glassy solids (emulsions/microgel suspensions) using Durian's bubble model.
Our results focus on the velocity and finite time displacement field in steady flow.
At low rate, surprisingly, the velocity field has correlations consistent with Eshelby transformations oriented \emph{isotropically} and not preferentially aligned with the imposed flow.
At higher rate, regardless of orientation, the Eshelby fields are cut off beyond a characteristic length, $\xi$ which diverges with vanishing shear rate, $\dot{\gamma}$, as $\xi\sim\dot{\gamma}^{-1/3}$.
The finite time displacement fields behave differently.
In the low $\dot{\gamma}$ regime, and for small time windows, $\Delta t$, the displacement fields show the same orientation-independent Eshelby correlations as the velocities.
At larger $\Delta t$, the displacement fields are composed of lines of slip aligned along the flow and gradient direction, precisely as in Lennard-Jones glasses.
At finite rate, these slip lines are cut off beyond the same rate-dependent characteristic length scale as the instantaneous velocity correlations.
The rate dependence of the shear stress and diffusion coefficient also obeys the same $\dot{\gamma}^{-1/3}$ scaling as the $\xi$ observed in the velocity and displacement fields.
The observation of long ranged (essentially infinite) Eshelby correlations in the instantaneous velocity fields during quasi-static shear is in sharp contrast with previous results on Lennard-Jones glasses and indicates that plasticity in amorphous solids where inertia is important, such as granular materials and metallic glasses, should have qualitatively different rate dependence than systems without inertia such as emulsions, suspensions, or foams.

Particle dispersion, surfactant and block copolymer self-assembly are investigated in room temperature ionic liquids. In this lecture I will show how we combine molecular thermodynamics & colloid and polymer science with neutron scattering and rheology measurements to explore the formation of surfactant mesophases, including spontaneous vesicles, hierarchically self-assembled block-copolymer micelle phases, and stable nanoparticle dispersions in ionic liquids. A cationic surfactant (DDAB) is shown to self-assemble into spontaneous vesicles by ion exchange between the surfactant and ionic liquid. At higher concentrations, this surfactant system shows a sponge-to lamellar transition, with a stable L₃ phase over a broad range of compositions and temperatures. Pluronic® block copolymers with varying solvophilic blocks and constant solvophobic blocks are shown to self-assemble into mesophases with structures varying systematically with the relative block lengths in qualitative accordance with expectations from packing considerations. Shear flow is shown to order these phases and an interesting stick-slip behavior is observed under time-dependent flows, where the microstructure is strongly coupled to the flow. Finally, shear thickening dispersions of colloidal particles in ionic liquids are being developed for possible use to improve the ballistic, puncture and abrasion resistance of space suits and micrometeorite and orbital debris (MMOD) shielding for spacecraft. Ionic liquids are proposed as the solvent phase of STFs formulation for space application because of their stability over the broad range of temperatures and low volatility. We successfully disperse nanoparticles into the ionic liquid [C₄mim][BF₄] by modifying the surface of silica nanoparticles with fluorocarbon 1H, 1H, 9H-hexadecafluoro-1-nonanol. A combination of rheology, dynamic light scattering, and small angle neutron scattering are employed to test an unproven hypothesis in literature, namely that stabilization of particles in this ionic liquid is a consequence of a solvation layer induced by hydrogen bonds between the cation and the surface groups. This work also has potentially important implications for environmental and energy engineering as ionic liquids are leading candidates for remediation, separation, and recycling of nuclear wastes.

Over the past several decades, there has been a significant emphasis on trying to elucidate the transport mechanism in amorphous polymers. One pervasive theme is that free volume facilitates transport. Free volume theories argue that the water molecules reside in dynamically fluctuating cages that are free to move around with thermal activation. In this limit, they predict that diffusion increases exponentially with the free volume. The underlying premise is that packing defects or density heterogeneities influence transport.
To better understand these relationships, we quantify the water transport properties in a series of epoxy networks with systematic variations in both their structural rigidity and cross-link density, as shown previously.[1] In these earlier studies we performed positron annihilation lifetime spectroscopy measurements to characterize the packing defects, often referred to as “free volume,” as a function of temperature and then correlated the positron free volume to both the equilibrium moisture uptakes [1] and effective diffusion coefficients [2]. These measurements revealed a positive correlation between the free volume fraction of the material at absolute zero (static packing defects in the absence of thermal motion) and the equilibrium uptake for networks of nominally similar polar group content across a wide temperature range. However, the polar group content of the network was deemed more important in that packing defects in establishing equilibrium uptake than the free volume fraction. However, these measurements also revealed a mild dependence on the effective diffusion coefficient of water on the average positron packing defect cavity size at temperatures below 50 C, but a cavity size independence for temperatures above 50 C.
What&’s surprising is that these observations contradict the conventional free volume concepts of transport where the diffusivity increases exponentially with free volume. To address this we explore the relationship between polymer dynamics and water transport by quantifying the mean square atomic displacement (MSD) fluctuations of these same epoxy networks with fixed window scans on the NIST High Flux Back Scattering Spectrometer. These measurements reveal a stronger correlation between the MSD and the transport kinetics than the positron free volume fractions. We quantitatively contrast the correlations between the positron free volume parameters and the moisture transport properties with the direct MSD measurements of matrix dynamics at the ns to ps time scale. These comparisons reveal a stronger correlation between polymer dynamics and moisture transport that seems to challenge previous theories suggesting that free volume governs transport. The nuances of these differences are discussed in detail.
1. Soles, C.L., et al, J Poly Sci Part B: Poly Phys, 1998. 36(17): p. 3035-3048.
2. Soles, C.L., et al, J Poly Sci Part B: Poly Phys, 2000. 38(5): p. 776-791.