Molecular Reorientation Dynamics Govern the Glass Transitions of the

Letter pubs.acs.org/JPCL

Molecular Reorientation Dynamics Govern the Glass Transitions of the Amorphous Ices J. J. Shephard and C. G. Salzmann* Department of Chemistry, University College London, 20 Gordon Street, London WC1H 0AJ, United Kingdom ABSTRACT: The glass transitions of low-density amorphous ice (LDA) and high-density amorphous ice (HDA) are the topic of controversial discussions. Understanding their exact nature may be the key to explaining the anomalies of liquid water but has also got implications in the general context of polyamorphism, the occurrence of multiple amorphous forms of the same material. We first show that the glass transition of hydrogendisordered ice VI is associated with the kinetic unfreezing of molecular reorientation dynamics by measuring the calorimetric responses of the corresponding H2O, H218O, and D2O materials in combination with X-ray diffraction. Well-relaxed LDA and HDA show identical isotopic-response patterns in calorimetry as ice VI, and we conclude that the glass transitions of the amorphous ices are also governed by molecular reorientation processes. This “reorientation scenario” seems to resolve the previously conflicting viewpoints and is consistent with the fragile-to-strong transition from water to the amorphous ices.

L

reorientation dynamics that lack the translational diffusion characteristics of the liquid state.8,27,28 For HDA, it has also been suggested that a glass-to-liquid transition takes place upon heating under pressure29−31 and even at ambient pressure if the sample is well relaxed,32 yet overall it is fair to say that our understanding of the glass transition of HDA is less advanced compared with LDA. Clarifying the exact mechanisms of the glass transitions of the amorphous ices may be of fundamental importance with respect to not only understanding the anomalies of liquid water but also the phenomenon of polyamorphism, the occurrence of different amorphous forms, in general. What is clear is that water is certainly not a simple glass-forming liquid, as illustrated, for example, by the unusual observation that its heat capacity increases in the supercooled regime and that a change from fragile to strong relaxation behavior according to Angell’s classification33 is observed as liquid water turns into LDA.7,34 Like LDA, HDA also seems to display strong relaxation characteristics.35 We aim to gain new insights into the mechanisms of the glass transitions of LDA, HDA, and also the ice VI phase by comparing the calorimetric responses of the corresponding H2O, H218O, and D2O materials. This analysis will be assisted by X-ray diffraction data recorded at the same heating rate as the calorimetric data. The overall ambition is to describe the glass transitions of LDA and HDA in a fashion that resolves the apparently conflicting statements in the literature. It is important to note that the hydrogen-disordered phases of ice, including ices I, IV, V, VI, and XII, also show glass transitions at low temperatures.36−41 The oxygen atoms in these phases of ice exhibit long-range crystalline order, but

iquid water is one of the most familiar everyday substances, and its existence on our planet is intimately linked to the evolution of life and the origins of our civilization.1 From the scientific point of view, water is a highly anomalous and in some aspects still poorly understood liquid.2 For example, the way liquid water forms a glass at low temperatures has been the topic of longstanding debates.3−9 The fundamental obstacle in forming glassy water is the simple fact that water displays a strong tendency for crystallization below its melting point. Small droplets and cooling rates greater than 107 K s−1 are required to prevent crystallization.10−13 The resulting material at low temperatures is amorphous, as seen by diffraction, and classified as low-density amorphous ice (LDA). Water-vapor deposition onto cold substrates produces a structurally very similar LDA.14−17 High-density amorphous ice (HDA) on the other hand can be obtained by pressure amorphization of the ordinary ice, ice Ih, or by compression of LDA at low temperatures.18,19 The discovery of two distinct amorphous forms of ice as well as a first-order like transition between the two19 has captured the attention of the scientific community. It has been suggested that the equilibrium line between LDA and HDA terminates in a second critical point in the phase diagram and that this singularity could be the origin of the many anomalies of liquid water.20−22 A question of paramount importance in this context is whether LDA and HDA are glassy materials that undergo glass-to-liquid transitions upon heating and are therefore thermodynamically linked with the liquid state. A weak endothermic step observed in calorimetry at ∼136 K upon heating LDA at 30 K min−1 and ambient pressure has been associated with a transition from a glass to the deeply supercooled liquid;23−26 however, arguments have also been made that this feature, which occurs just before crystallization to ice I, is associated with a “shadow glass transition”,5 an order−disorder transition,7 or that it may be caused by © XXXX American Chemical Society

Received: April 24, 2016 Accepted: May 31, 2016

2281

DOI: 10.1021/acs.jpclett.6b00881 J. Phys. Chem. Lett. 2016, 7, 2281−2285

Letter

The Journal of Physical Chemistry Letters there is disorder with respect to the orientations of the hydrogen-bonded water molecules. The glass transitions of these materials have been associated with the unfreezing of the reorientation dynamics upon heating.28,36−41 Strong evidence of this understanding is that doping with hydrochloric acid, which dramatically accelerates molecular reorientations dynamics in ices V, VI, and XII,42,43 leads to the complete disappearance of the endothermic steps, and the corresponding hydrogen-ordered phases of ice can be prepared at temperatures below the glass transitions of the pure hydrogendisordered phases.40,41 The glass transitions of the hydrogendisordered ices illustrate nicely that the glass-to-liquid transition is only one type of glass transition, and others, such as the glass transitions of glassy crystals and glassy liquid crystals, need to be considered as well.44 The former display translational order and orientational disorder, whereas the latter are defined by translation disorder and orientational order. Kauzmann in his seminal paper on glass transitions already pointed out that the phenomenon of the glass transition is not restricted to liquids but is, in principle, possible for all noncrystalline materials;45 that is, all materials that display residual entropy at 0 K. The increases in heat capacity, ΔCp, of the glass transitions of the hydrogen-disordered ices and also of hydrogen-disordered clathrate hydrates and inorganic hydrates are typically ∼1 J K−1 mol−1, and therefore very similar as observed for LDA and HDA.23−26,32,46 Before investigating the amorphous ices in detail, we first confirm that the glass transitions of the hydrogen-disordered ices are indeed associated with the unfreezing of the reorientation dynamics. For this, we measure the differences in the calorimetric responses of H2O, H218O, and D2O ice VI at ambient pressure. Changing from H2O to H218O or D2O increases the molecular weight from 18 to 20 g mol−1 in both cases. For dynamic unfreezing processes involving translational motion, H218O and D2O materials would therefore be expected to display different behavior compared with the corresponding H2O material.47 For molecular reorientation processes on the contrary, predominately involving movements of hydrogen atoms, D2O materials are expected to display delayed thermal onsets of their glass transitions compared with their H2O and H218O counterparts, which are expected to behave similarly. Figure 1a shows that the differential scanning calorimetry (DSC) scans of H2O and H218O ice VI are very similar with glass-transition temperatures, Tg, of 133.8 ± 0.1 K and 133.9 ± 0.8 K, respectively. The Tg of D2O ice VI, however, is significantly shifted toward higher temperatures and located at 138.1 ± 0.6 K.41 Following the endothermic steps, all three samples show a small “plateau” region before the irreversible and exothermic crystallization to stacking-disordered ice I (ice Isd). Consistent with the DSC scans of pressure-quenched ices IV, V, and XII, ice VI also shows a so-called undershoot effect before the heat capacity increase. Such features are generally associated with unrelaxed samples that begin to relax before the glass-transition temperatures are reached.48 It is noteworthy that the onset temperatures of the undershoot features show the same isotopic shift pattern as the glass-transition temperatures. In line with previous suggestions,28,38−41 the isotopic shift pattern of ice VI clearly shows that its glass transition is associated with molecular reorientation dynamics and that the kinetic unfreezing event has no translational diffusion elements associated with it. Translational diffusion would destroy the crystalline oxygen framework of ice VI. To show that this is not

Figure 1. DSC scans of H2O, H218O, and D2O (a) ice VI, (b) LDA, and (c) eHDA recorded upon heating at 10 K min−1. The LDA samples were annealed at 130 K for 90 min prior to recording the DSC scans.24 The glass-transition temperatures were determined from the intersections of the gray dashed lines. The margins of error of the Tg values determined from at least four individual measurements are indicated.

the case we have recorded X-ray diffraction patterns upon heating H2O ice VI at ambient pressure at the same heating rate as for the DSC experiments. The diffraction data in Figure 2a confirm that the glass transition of ice VI is not visible in its diffraction patterns, which display strong Bragg peaks, and that the ice VI transforms at higher temperatures directly to ice Isd without any intermediate stages. This means that the oxygen atoms stay in fixed positions in ice VI beyond the glasstransition temperature and until the irreversible phase transition to ice Isd takes place. Having established the calorimetric response pattern upon H/D and 16O/18O isotopic substitution for the kinetic unfreezing of the molecular reorientation dynamics in hydrogen-disordered ice VI, we now turn to LDA and HDA. To make the glass transitions of these amorphous materials clearly observable it is important that well-relaxed samples are used. In the case of LDA, the samples were therefore annealed at 130 K for 90 min24 before recording the DSC scans shown in Figure 1b. The Tg values of H2O, H218O, and D2O LDA upon heating at 10 K min−1 are 131.8 ± 0.8, 131.0 ± 0.2, and 135.6 ± 0.1 K, respectively. The glass transition of HDA at ambient pressure is 2282

DOI: 10.1021/acs.jpclett.6b00881 J. Phys. Chem. Lett. 2016, 7, 2281−2285

Letter

The Journal of Physical Chemistry Letters

Figure 3. Ratios of the glass-transition temperatures of H218O and D2O ice VI, LDA, and eHDA with the glass-transition temperatures of the corresponding H2O materials.

respect to the Tg values of the corresponding H2O materials. Remarkably, not only do all Tg values shift upward in temperature upon going from H2O to D2O, but there is also quantitative agreement in that the D2O/H2O ratios of the Tg values of ice VI, LDA, and eHDA have the same value of ∼1.03. This illustrates that the glass transitions of LDA and HDA have the same underlying mechanism as for the hydrogen-disordered ice VI. The low-temperature glass transitions of hydrogendisordered and amorphous ice materials are governed by the thermal unfreezing of the motions of hydrogen atoms and are therefore related to changes in the orientations of the water molecules. The isotopic shift in Tg from H2O to D2O LDA as well as the differences in their dielectric relaxation times have been recently discussed, and it has been suggested that quantum effects, possibly connected with tunneling processes, contribute to the observed differences.51,52 This view is entirely consistent with our conclusion that reorientation dynamics, which involve the movement of light hydrogen atoms only, are responsible for the glass transitions of LDA and, as shown here, of eHDA and the hydrogen-disordered ices. The exact mechanism of the reorientation dynamics will require further studies, but it is most conceivable that the movements of the hydrogen atoms involve either ionic or Bjerrum-type point defects within the fully hydrogen-bonded networks of the amorphous and hydrogen-disordered ices.8,27,28 This implies that the resulting reorientation processes are cooperative and involve a large number of water molecules. The cooperative rearrangement of hydrogen atoms via defect migration is thought to be a different mechanism to those governing diffusion processes in the ambient temperature liquid phase.27,53,54 In our opinion, this difference in diffusion mechanisms explains the different relaxation behaviors of the fragile liquid phase and the strong amorphous ices, which are, like the hydrogen-disordered ices, much more dominated by the network character of their structures. We believe this study resolves much of the controversy surrounding the glass transitions of the amorphous ices. The calorimetric data of the amorphous ices display all of the typical characteristics of glass transitions. So the initial assignments to the glass-to-liquid transitions have been a logical first step.23−26,32 We also believe that the descriptions of the glass transition of LDA as a “shadow glass transition” of the liquid5 or an order−disorder transition7 are fully consistent with the “reorientation scenario” outlined here. A question that remains, however, is how translational diffusion elements begin to appear above the glass transitions of the amorphous ices. This could take place gradually or, if crystallization could be suppressed, it could perhaps be seen, in line with Angell’s suggestion,4,5 in the form

Figure 2. X-ray powder diffraction patterns (Cu Kα1) recorded upon heating at 10 K min−1 of H2O (a) ice VI, (b) LDA, and (c) eHDA. Panels b and c show the square roots of the recorded intensities to emphasize lower intensity features. Asterisks in panels b and c indicate small amounts of ice Ih, which resulted from vapor condensation during the sample transfers.

only observed for the well-relaxed expanded HDA (eHDA) state,49 which can be obtained by decompression of very highdensity amorphous ice (vHDA).32,50 Figure 1c shows the glass transitions of H2O, H218O, and D2O eHDA with Tg values of 113.7 ± 1.1, 112.90 ± 0.8, and 117.7 ± 1.2 K, respectively. The X-ray diffraction data of the H2O materials corresponding to the DSC traces in Figure 1b,c are shown in Figure 2b,c. The stable position of the diffraction maximum of eHDA upon heating illustrates the well-relaxed nature of the eHDA state, which transforms abruptly to LDA, as has been previously observed in DSC.32,50 The isotopic shift pattern upon H/D and 16O/18O substitution for LDA and eHDA is the same as previously observed for the hydrogen-disordered ice VI. The DSC scans of the H218O materials are within error identical with the H2O materials, and pronounced shifts of Tg are observed for the D2O materials. Figure 3 shows the ratios of the Tg values with 2283

DOI: 10.1021/acs.jpclett.6b00881 J. Phys. Chem. Lett. 2016, 7, 2281−2285

Letter

The Journal of Physical Chemistry Letters

(9) Amann-Winkel, K.; Böhmer, R.; Fujara, F.; Gainaru, C.; Geil, B.; Loerting, T. Colloquium: Water’s Controversial Glass Transitions. Rev. Mod. Phys. 2016, 88 (1), 011002. (10) Brüggeller, P.; Mayer, E. Complete Vitrification in Pure Liquid Water and Dilute Aqueous Solutions. Nature 1980, 288, 569−571. (11) Mayer, E.; Brüggeller, P. Vitrification of Pure Liquid Water by High-pressure Jet-freezing. Nature 1982, 298, 715−718. (12) Mayer, E. New Method for Vitrifying Water and Other Liquids by Rapid Cooling of Their Aerosols. J. Appl. Phys. 1985, 58, 663−667. (13) Kohl, I.; Bachmann, L.; Hallbrucker, A.; Mayer, E.; Loerting, T. Liquid-like Relaxation in Hyperquenched Water at £ 140 K. Phys. Chem. Chem. Phys. 2005, 7 (17), 3210−3220. (14) Burton, E. F.; Oliver, W. F. The Crystal Structure of Ice at Low Temperatures. Proc. R. Soc. London, Ser. A 1935, 153, 166−172. (15) Burton, E. F.; Oliver, W. F. X-ray Diffraction Patterns of Ice. Nature 1935, 135, 505−506. (16) Jenniskens, P.; Blake, D. F. Structural Transitions in Amorphous Water Ice and Astrophysical Implications. Science 1994, 265, 753−756. (17) Bowron, D. T.; Finney, J. L.; Hallbrucker, A.; Kohl, I.; Loerting, T.; Mayer, E.; Soper, A. K. The Local and Intermediate Range Structures of the Five Amorphous Ices at 80 K and Ambient Pressure: A Faber-Ziman and Bhatia-Thornton analysis. J. Chem. Phys. 2006, 125, 194502. (18) Mishima, O.; Calvert, L. D.; Whalley, E. ’Melting Ice’ I at 77 K and 10 kbar: A New Method of Making Amorphous Solids. Nature 1984, 310, 393−395. (19) Mishima, O.; Calvert, L. D.; Whalley, E. An Apparently Firstorder Transition Between Two Amorphous Phases of Ice Induced by Pressure. Nature 1985, 314, 76−78. (20) Poole, P. H.; Sciortino, F.; Essmann, U.; Stanley, H. E. Phase Behavior of Supercooled Water. Nature 1992, 360, 324−328. (21) Stanley, H. E.; Angell, C. A.; Essmann, U.; Hemmati, M.; Poole, P. H.; Sciortino, F. Is There a Second Critical Point in Liquid Water? Phys. A 1994, 205, 122. (22) Sciortino, F.; Poole, P. H.; Essmann, U.; Stanley, H. E. Line of Compressibility Maxima in the Phase Diagram of Supercooled Water. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1997, 55 (1), 727−737. (23) Johari, G. P.; Hallbrucker, A.; Mayer, E. The Glass-liquid Transition of Hyperquenched Water. Nature 1987, 330, 552−553. (24) Hallbrucker, A.; Mayer, E.; Johari, G. P. Glass-Liquid Transition and the Enthalpy of Devitrification of Annealed Vapor-Deposited Amorphous Solid Water: A Comparison with Hyperquenched Glassy Water. J. Phys. Chem. 1989, 93, 4986−4990. (25) Hallbrucker, A.; Mayer, E.; Johari, G. P. The Heat Capacity and Glass Transition of Hyperquenched Glassy Water. Philos. Mag. B 1989, 60, 179−187. (26) Johari, G. P. Liquid State of Low-Density Pressure-Amorphized Ice above Its Tg. J. Phys. Chem. B 1998, 102, 4711−4714. (27) Fisher, M.; Devlin, J. P. Defect Activity in Amorphous Ice from Isotopic Exchange Data: Insight into the Glass Transition. J. Phys. Chem. 1995, 99, 11584−11590. (28) Salzmann, C. G.; Radaelli, P. G.; Slater, B.; Finney, J. L. The Polymorphism of Ice: Five Unresolved Questions. Phys. Chem. Chem. Phys. 2011, 13, 18468−18480. (29) Mishima, O. The Glass-to-liquid Transition of the Emulsified High-density Amorphous Ice Made by Pressure-induced Amorphization. J. Chem. Phys. 2004, 121, 3161−3164. (30) Andersson, O. Glass−liquid Transition of Water at High Pressure. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 11013−11016. (31) Seidl, M.; Elsaesser, M. S.; Winkel, K.; Zifferer, G.; Mayer, E.; Loerting, T. Volumetric Study Consistent with a Glass-to-liquid Transition in Amorphous Ices under Pressure. Phys. Rev. B: Condens. Matter Mater. Phys. 2011, 83, 100201. (32) Amann-Winkel, K.; Gainaru, K.; Handle, P. H.; Seidl, M.; Nelson, H.; Böhmer, R.; Loerting, T. Water’s Second Glass Transition. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 17720−17725.

of additional glass transitions. According to our analysis, these would correspond to the kinetic unfreezing of translational diffusion. An intriguing possibility is, of course, that the unfreezing of translational diffusion is the strong-to-fragile transition.

■

EXPERIMENTAL METHODS D2O was obtained from Sigma-Aldrich (99.9% D) and H218O from Cambridge Isotopes (97% 18O). The various ice samples were prepared using a piston cylinder setup and following literature procedures. Specifically, ice VI was obtained by heating ice Ih at 1.0 GPa to 260 K41 and LDA by annealing uHDA at 130 K for 90 min.24 For making eHDA, vHDA was prepared by annealing uHDA at 1.1 GPa and 165 K, followed by decompression to 0.3 GPa at 140 K.50 A few small pieces of the ices were then transferred into stainless-steel capsules with screwable lids under liquid nitrogen. These were quickly transferred into a precooled PerkinElmer DSC 8000 advanced double-furnace differential scanning calorimeter before heating at 10 K min−1. The moles of ice in the DSC capsules were determined by recording the endothermic melting at 0 °C and using the corresponding enthalpies of melting of ice Ih. For X-ray diffraction measurements, the ice samples were transferred under liquid nitrogen into a purpose-built Kapton window sample holder mounted on a Stoe Stadi P diffractometer with Cu Kα1 radiation at 40 kV, 30 mA and monochromated by a Ge 111 crystal. Data were collected at a fixed angle using a Mythen 1K area detector, and the samples were heated at 10 K min−1 using an Oxford Instruments CryojetHT.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the Royal Society (UF100144) and the Leverhulme Trust (RPG-2014-04) for financial support, M. Vickers for help with the XRD measurements, and J. K. Cockcroft for access to the Cryojet.

■

REFERENCES

(1) Ball, P. H2O A Biography of Water; Weidenfeld & Nicolson, 1999. (2) Ball, P. Water - an Enduring Mystery. Nature 2008, 452, 291− 292. (3) Smith, R. S.; Kay, B. D. The Existence of Supercooled Liquid Water at 150 K. Nature 1999, 398, 788−791. (4) Velikov, V.; Borick, S.; Angell, C. A. The Glass Transition of Water, Based on Hyperquenching Experiments. Science 2001, 294, 2335−2338. (5) Yue, Y.; Angell, C. A. Clarifying The Glass-transition Behaviour of Water by Comparison with Hyperquenched Inorganic Glasses. Nature 2004, 427, 717−720. (6) Souda, R. Kinetics of the Glass−liquid Transition of Water. Chem. Phys. Lett. 2005, 415, 146−149. (7) Angell, C. A. Insights into Phases of Liquid Water from Study of Its Unusual Glass-forming Properties. Science 2008, 319, 582−587. (8) Shephard, J. J.; Evans, J. S. O.; Salzmann, C. G. Structural Relaxation of Low-Density Amorphous Ice upon Thermal Annealing. J. Phys. Chem. Lett. 2013, 4, 3672−3676. 2284

DOI: 10.1021/acs.jpclett.6b00881 J. Phys. Chem. Lett. 2016, 7, 2281−2285

Letter

The Journal of Physical Chemistry Letters (33) Angell, C. A. Structural Instability and Relaxation in Liquid and Glassy Phases Near the Fragile Liquid Limit. J. Non-Cryst. Solids 1988, 102 (1−3), 205−221. (34) Angell, C. A. Water II is a ″Strong″ Liquid. J. Phys. Chem. 1993, 97 (24), 6339−6341. (35) Angell, C. A. Liquid Fragility and the Glass Transition in Water and Aqueous Solutions. Chem. Rev. 2002, 102 (8), 2627−2650. (36) Haida, O.; Matsuo, T.; Suga, H.; Seki, S. Calorimetric Study of the Glassy State X. Enthalpy Relaxation at the Glass-transition Temperature of Hexagonal Ice. J. Chem. Thermodyn. 1974, 6, 815− 825. (37) Suga, H.; Matsuo, T.; Yamamuro, O. Thermodynamic Study of Ice and Clathrate Hydrates. Pure Appl. Chem. 1992, 64, 17−26. (38) Salzmann, C. G.; Kohl, I.; Loerting, T.; Mayer, E.; Hallbrucker, A. The Low-temperature Dynamics of Recovered Ice XII as Studied by Differential Scanning Calorimetry: A Comparision with Ice V. Phys. Chem. Chem. Phys. 2003, 5, 3507−3517. (39) Salzmann, C. G.; Mayer, E.; Hallbrucker, A. Thermal Properties of Metastable Ices IV and XII: Comparison, Isotope Effects and Relative Stabilities. Phys. Chem. Chem. Phys. 2004, 6, 1269−1276. (40) Salzmann, C. G.; Radaelli, P. G.; Finney, J. L.; Mayer, E. A Calorimetric Study on the Low Temperature Dynamics of Doped Ice V and Its Reversible Phase Transition to Hydrogen Ordered Ice XIII. Phys. Chem. Chem. Phys. 2008, 10, 6313−6324. (41) Shephard, J. J.; Salzmann, C. G. The Complex Kinetics of the Ice VI to Ice XV Hydrogen Ordering Phase Transition. Chem. Phys. Lett. 2015, 637, 63−66. (42) Salzmann, C. G.; Radaelli, P. G.; Hallbrucker, A.; Mayer, E.; Finney, J. L. The Preparation and Structures of Hydrogen Ordered Phases of Ice. Science 2006, 311, 1758−1761. (43) Salzmann, C. G.; Radaelli, P. G.; Mayer, E.; Finney, J. L. Ice XV: A New Thermodynamically Stable Phase of Ice. Phys. Rev. Lett. 2009, 103, 105701. (44) Suga, H.; Seki, S. Thermodynamic Investigation on Glassy States of Pure Simple Compounds. J. Non-Cryst. Solids 1974, 16 (2), 171−194. (45) Kauzmann, W. The Nature of the Glassy State and the Behavior of Liquids at Low Temperatures. Chem. Rev. 1948, 43 (2), 219−256. (46) Handa, Y. P.; Klug, D. D. Heat Capacity and Glass Transition Behavior of Amorphous Ice. J. Phys. Chem. 1988, 92, 3323−3325. (47) Mehrer, H. Diffusion in Solids; Springer: Berlin, Germany, 2007. (48) Scherer, G. W. Relaxation in Glass and Composites; John Wiley & Sons: Corning, NY, 1986. (49) Nelmes, R. J.; Loveday, J. S.; Strässle, T.; Bull, C. L.; Guthrie, M.; Hamel, G.; Klotz, S. Annealed High-density Amorphous Ice Under Pressure. Nat. Phys. 2006, 2, 414−418. (50) Winkel, K.; Mayer, E.; Loerting, T. Equilibrated High-Density Amorphous Ice and Its First-Order Transition to the Low-Density Form. J. Phys. Chem. B 2011, 115, 14141−14148. (51) Gainaru, C.; Agapov, A. L.; Fuentes-Landete, V.; AmannWinkel, K.; Nelson, H.; Köster, K. W.; Kolesnikov, A. I.; Novikov, V. N.; Richert, R.; Böhmer, R.; et al. Anomalously Large Isotope Effect in the Glass Transition of Water. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 17402−17407. (52) Agapov, A. L.; Kolesnikov, A. I.; Novikov, V. N.; Richert, R.; Sokolov, A. P. Quantum Effects in the Dynamics of Deeply Supercooled Water. Phys. Rev. B 2015, 91 (2), 022312. (53) Qvist, J.; Schober, H.; Halle, B. Structural Dynamics of Supercooled Water from Quasielastic Neutron Scattering and Molecular Simulations. J. Chem. Phys. 2011, 134 (14), 144508. (54) Qvist, J.; Mattea, C.; Sunde, E. P.; Halle, B. Rotational Dynamics in Supercooled Water from Nuclear Spin Relaxation and Molecular Simulations. J. Chem. Phys. 2012, 136 (20), 204505.

2285

DOI: 10.1021/acs.jpclett.6b00881 J. Phys. Chem. Lett. 2016, 7, 2281−2285