pubmed.ncbi.nlm.nih.gov

Origin of the low-temperature endotherm of acid-doped ice VI: new hydrogen-ordered phase of ice or deep glassy states? - PubMed

  • ️Mon Jan 01 2018

. 2018 Oct 10;10(2):515-523.

doi: 10.1039/c8sc03647k. eCollection 2019 Jan 14.

Affiliations

Origin of the low-temperature endotherm of acid-doped ice VI: new hydrogen-ordered phase of ice or deep glassy states?

Alexander Rosu-Finsen et al. Chem Sci. 2018.

Abstract

On the basis of a low-temperature endotherm, it has recently been argued that cooling acid-doped ice VI at high pressures leads to a new hydrogen-ordered phase. We show that the endotherms are in fact caused by the glass transitions of deep glassy states related to ice VI. As expected for such endothermic overshoot effects, they display a characteristic dependence on pressure and cooling rate, they can be produced by sub-T g annealing at ambient pressure, and they can be made to appear or disappear depending on the heating rate and the initial extent of relaxation. It is stressed that the existence of a new crystalline phase, as it has been suggested, cannot depend on the heating rate at which it is heated. X-ray diffraction shows that samples for which the low-temperature endotherm is present, weak or absent, as observed at a heating rate of 5 K min-1, are structurally very similar. Furthermore, we show that the reported shifts of the (102) Bragg peak upon heating are fully consistent with our scenario and also with our earlier neutron diffraction study. Deuterated acid-doped ice VI cooled at high pressure also displays a low-temperature endotherm and its neutron diffraction pattern is consistent with deep glassy ice VI. Accessing deep glassy states of ice with the help of acid doping opens up a fascinating new chapter in ice research. Compared to pure ice VI, the glass transition temperature is lowered by more than 30 K by the acid dopant. Future work should focus on the deep glassy states related to all the other hydrogen-disordered ices including the 'ordinary' ice Ih.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1. Schematic illustration of the various processes that can take place upon cooling pure and HCl-doped hydrogen-disordered ice VI leading to glassy, partially hydrogen-disordered or fully hydrogen-ordered states, respectively. The inset shows how the formation of glassy states at the glass transition temperature of HCl-doped ice VI is affected by the cooling rate. Adapted from ref. 31.
Fig. 2
Fig. 2. (a) DSC scans of HCl-doped H2O ice VI/XV samples recorded after quenching at ∼40 K min–1 at 1.0, 1.4 or 1.8 GPa (1a, 2a, 3a), after slow-cooling at 2.5 K min–1 at 1.0, 1.4 and 1.8 GPa (1b, 2b, 3b), and after cooling from 138 K at 5 K min–1 at ambient pressure (1c, 2c, 3c). All scans were recorded upon heating at 5 K min–1. The dashed black lines indicate the boundary between endothermic and exothermic processes for each of the scans. (b) Schematic illustration of how glassy states with different degrees of enthalpy relaxation behave upon heating as the dynamic equilibrium line is crossed. The dashed black line shows how the experimentally observed glass-transition temperature increases for deeper glassy states. Adapted from ref. 39.
Fig. 3
Fig. 3. (a) Effect of sub-Tg annealing on the DSC scans of HCl-doped H2O ice VI/XV quenched at 1.4 GPa. Scan (1) shows the unannealed sample whereas scans (2) and (3) were recorded after annealing at 93 K for 2 and 6 hours, respectively. The dashed black lines indicate the boundary between endothermic and exothermic processes for each of the scans. All scans were recorded upon heating at 5 K min–1. (b) Schematic illustration of the effect of sub-Tg annealing on the subsequent heating. Adapted from ref. 39 and 41.
Fig. 4
Fig. 4. (a) Effect of heating rate on the appearance of the ice VI/XV low-temperature endotherm. The DSC scans were recorded upon heating HCl-doped H2O ice VI/XV quenched at 1.0, 1.4 or 1.8 GPa with a range of different heating rates as indicated. The dashed black lines indicate the boundary between endothermic and exothermic processes for each of the scans. (b) Schematic illustration of the effect of heating a glassy state at different heating rates. Adapted from ref. 39 and 41.
Fig. 5
Fig. 5. X-ray diffraction patterns of HCl-doped H2O ice VI/XV quenched at 1.0, 1.4 or 1.8 GPa recorded at 95 K and ambient pressure. The inset shows the region where weak Bragg peaks characteristic for ice XV appeared. The tickmarks indicate the expected peak position for ice VI and ice XV.,
Fig. 6
Fig. 6. Changes of the position of the (102) Bragg peak upon heating HCl-doped H2O and DCl-doped D2O ice VI/XV samples. The data for the HCl-doped H2O sample was taken from Fig. 10(c) in ref. 30 whereas the position of the (102) peak of the DCl-doped D2O sample was calculated from the lattice constants shown in Fig. 4 of ref. 26.
Fig. 7
Fig. 7. Calorimetry and neutron diffraction of DCl-doped D2O ice VI. (a) DSC scans recorded upon heating DCl-doped ice VI/XV quenched at 1.8 GPa at either 10 or 30 K min–1. (b) Powder neutron diffraction patterns at ambient pressure of DCl-doped ice VI at 80 K after quenching at 1.8 GPa. The experimental data is shown as a thick grey line, the Rietveld fit as a thin red line, and the differences between the experimental and fitted data as a thin black line. Tickmarks indicate the positions of the Bragg peaks of ice VI. A small peak due to vanadium at ∼2.1 Å is marked with a “V”.

Similar articles

Cited by

References

    1. Bridgman P. W. Proc. Am. Acad. Arts Sci. 1912;47:441–558.
    1. Kamb B. Science. 1965;150:205–209. - PubMed
    1. Kuhs W. F., Finney J. L., Vettier C., Bliss D. V. J. Chem. Phys. 1984;81:3612–3623.
    1. Kuhs W. F., Ahsbahs H., Londono D., Finney J. L. Phys. Rev. B: Condens. Matter Mater. Phys. 1989;156–157:684–687.
    1. Johari G. P., Whalley E. J. Chem. Phys. 1976;64:4484–4489.