Boson Peak in Deeply Cooled Confined Water: A Possible Way to Explore the Existence of the Liquid-to-Liquid Transition in Water

Abstract

The boson peak in deeply cooled water confined in nanopores is studied with inelastic neutron scattering. We show that in the (P, T) plane, the locus of the emergence of the boson peak is nearly parallel to the Widom line below ∼1600 bar. Above 1600 bar, the situation is different and from this difference the end pressure of the Widom line is estimated. The frequency and width of the boson peak correlate with the density of water, which suggests a method to distinguish the hypothetical "low-density liquid" and "highdensity liquid" phases in deeply cooled water. DOI: 10.1103/PhysRevLett.112.237802 PACS numbers: 61.20.Lc, 63.50.−x The boson peak is a broad peak observed at low frequencies (∼2-10 meV) in the inelastic neutron, x-ray, and Raman scattering spectra of many amorphous materials and supercooled liquids [1][2][3]. Preliminary measurements on the boson peak in deeply cooled water confined in nanopores [4] and protein surfaces [5] under ambient pressure have been reported. The measured spectra include a boson peak at about 6 meV, which only emerges below ∼230 K. A similar result was also observed in a computer simulation study [6]. Since this temperature is close to the ambient-pressure Widom line temperature T W ∼224 K (the Widom line is the locus of specific heat maxima [7]), the authors tentatively explained this phenomenon as due to transformation of the local structure of water from a predominantly high-density liquid (HDL) to a predominantly low-density liquid (LDL) form as T W is crossed from above. However, this experimental result is insufficient. First, it is clear that this emergence of the boson peak in the raw spectra depends on the energy resolution of the experimental facility. Second, and more importantly, the situation at high pressures has, heretofore, remained unknown. Therefore, without quantitative analysis of the entire spectra, including quasielastic and inelastic components at high pressure, definite conclusions cannot be drawn.In this Letter, we have measured the boson peak in deeply cooled water with a series of inelastic neutron scattering (INS) experiments in the temperature (T) range from 120 to 230 K and the pressure (P) range from 400 to 2400 bar employing the Disk Chopper Spectrometer (DCS) [8] at the National Institute of Standards and Technology (NIST) Center for Neutron Research and the Cold Neutron Chopper Spectrometer (CNCS) [9] at the Spallation Neutron Source of the Oak Ridge National Laboratory (ORNL). The DAVE software was used for the data reduction [10]. We find that below ∼1600 bar, the emergence of the boson peak is correlated (but does not overlap) with the Widom line and with the fragile-to-strong crossover (FSC) in deeply cooled water [11]. Above ∼1600 bar, the locus of the emergence of the boson peak in the (P, T) plane has a different slope as compared with its behavior below ∼1600 bar, and the end pressure of the Widom line is estimated by determining where the slope begins to change. Moreover, the (P, T) dependences of the shape of the boson peak are f...

X-ray and Neutron Scattering of Water

“…Even though experimental evidence has been provided in the past for the presence of a Boson peak in HDA ice 152 , more recent neutron time-of-flight and backscattering spectroscopy results seem to rule out this possibility 153 . On the other hand, a Boson peak has been found in experiments on and simulations of protein hydration water 154,155 , simulations of supercooled water 156 and supercooled confined water 134 .…”

“…Very often, StokesEinstein relation describes remarkably well such relation between self-diffusion and viscosity through a characteristic length. However, QENS measurements in undercooled 134 and high-density water 14 show that this relation breaks down in water, where the local structure and the available free volume for diffusion are determined by the HB network rearrangement under variation of the thermodynamic parameters.…”

The Boson peak in confined water: An experimental investigation of the liquid-liquid phase transition hypothesis

“…We report a BP data analysis of INS spectra reported in our two previous papers [38,39]. The INS experiments were carried out in the following P -T ranges: 120 K < T < 230 K and 1 bar < P < 2400 bar (large symbols in Figure 1).…”

“…Under this confinement, water remains in the liquid state at temperatures much lower than that of homogeneous nucleation. Details on the experimental procedure and sample preparation are reported elsewhere [38,39]. The measured energy spectra for a fixed wavevector Q (the magnitude of the momentum transfer of the incident neutron) are indicated as S m (Q, E).…”

“…In fact, we study the BP at the water FSDC in order to verify if there is any link between the BP and the Widom line as recently hypothesized for bulk water by an MD simulation study [37] and observed in confined water [38,39]. In other words, the aim of the present work, by using inelastic neutron scattering (INS), is to verify in the (P, T ) plane whether deeply cooled water confined in nanopores can provide confirmation about the appearance of LLPT.…”