Metastable Melting Lines for H 2 O and the Liquid-Liquid Phase Transition Hypothesis
- PDF / 473,754 Bytes
- 9 Pages / 414.72 x 648 pts Page_size
- 56 Downloads / 203 Views
ABSTRACT When ice Ih in an emulsion is compressed below 250 K, it melts to supercooled liquid water, avoiding the formation of other crystal phases. Here, we create emulsified high-pressure ices under high pressure and low temperature, and measure their temperature while these ices are decompressed at a constant rate at different temperatures. We detect metastable melting points of high-pressure ices, and identify their melting lines. We find what could be possibly two new ice phases, and discuss the relationship between decompression-induced melting and decompression-induced amorphization. Finally, we discuss briefly the analysis of experimental data and simulation results that are consistent with the hypothesized "second critical point" with temperature and pressure coordinates of approximately 200 K and 100 Mpa. When liquid water is supercooled below the homogeneous nucleation temperature, TH, crystal phases nucleate homogeneously, and the liquid freezes spontaneously to the crystalline phase. When amorphous solid ice is heated, it crystallizes above the crystallization temperature, Tx. Therefore, amorphous forms of H 2 0 do not exist between TH and Tx (Fig. 1). When we compress the crystalline ice Ih at low temperatures (Figs. 1 and 2), it is transformed to supercooled liquid (L) on its metastable melting line above TH; between TH and Tx, to a high-pressure crystalline ice (X) at the smoothly extrapolated melting line; and below Tx, it is amorphized to the high-density amorphous ice (HDA) at a pressure higher than the smoothly extrapolated melting line [1]. To avoid the complication of the usual crystalcrystal transformations interrupting the melting process, we use an ice emulsion (1-10rm ice particles in oil [2]). In this study, we create 1 cm 3 emulsified high-pressure ices in a piston-cylinder apparatus, decompress the sample at a constant rate of 0.2 GPa/min, and observe their transitions by detecting a change in the sample temperature by an attached clomel-alumel thermocouple during the decompression. Then, we determine melting pressures at different temperatures (Fig. 3). The obtained melting curves agree with previously-reported data [3,4], which confirms the accuracy of this method. Moreover, we can determine the location of metastable melting lines to much lower temperatures (Figs. 4 and 5). Unexpectedly, we find what appear to be two possible new phases (PNP) of solid H 20, denoted PNP-XIII and PNP-XIV. This nomenclature is perhaps not optimal, and hence, we will replace these terms by any better suggestion. The location of the PNP-XIV line (the XIV line in Figs. 1 and 5) is slightly different from that of ice VIII reported by Klug et
al. [5].
443 Mat. Res. Soc. Symp. Proc. Vol. 499 01998 Materials Research Society
When we compress the crystalline ice Ih (Fig. 1), it is transformed to supercooled liquid (L) on its metastable melting line above TH; between TH and Tx, to a high-pressure crystalline ice (X) at the smoothly extrapolated melting line; and below Tx, it is amorphized to the high-density
Data Loading...
