Evaluation of Pressure Transmitting Media by63Cu-NQR of Cu2O

High pressure technique up to about 3GPa is now widely used in studies of condensed matter physics. This is due to the development of hybrid cylinder clamp cells, especially CuBe/NiCrAl and CuBe/MP35N. Further high pressures can be achieved by Bridgman anvil cell (up to about 10GPa), cubic anvil cell (up to about 20GPa) and diamond anvil cell (up to about 400GPa), and so on. Optical measurements and electric resistivity measurement with the use of the diamond anvil cell are now conventional methods for high pressure measurements beyond about 100GPa. However, other major measurements such as neutron diffraction, nuclear magnetic resonance (NMR), nuclear quadrupole resonance (NQR), and muon spin relaxation ( SR) have technical difficulty since large sample space is essentially needed for these methods. Of these methods, NMR/NQR under high pressure is relatively accessible. We indeed reported Cu-NQR measurements of Cu2O up to about 10GPa by using a modified Bridgman anvil cell which is designed for general-purpose pressure cell. In our previous NQR measurements of Cu2O, the spectral broadening was observed and was nearly proportional to pressure P. This was ascribable to the existence of inhomogeneous pressure distribution inside the cell and/or local shear of the crystalline lattice, induced by the uniaxial stress of the inhomogeneous pressure. A reason of the pressure inhomogeneity except for the problem, originating from the Bridgman anvil cell itself, is that the pressure transmitting medium is already solid at high pressures. Generally, the solid pressure transmitting media make the pressure distribution inside the cell inhomogeneous. However, it is difficult to avoid solidification of the pressure media at high pressures. Hence, it is vital to choose the most appreciate medium for high pressure measurements. The NMR/NQR technique has an advantage to investigate the suitability of the pressure media because the pressure inhomogeneity inside pressure cell is strongly related to the broadening of NMR/NQR spectra. In this paper, we report Cu-NQR measurement of Cu2O with the currently representative liquid pressure-transmitting media. As the first step of the investigation, we demonstrated the NQR experiments under pressure up to about 3GPa at low temperature and room temperature. There are some reports on the liquid pressure-transmitting media, but few appeared in refereed journals. Moreover, to the best of our knowledge, there is no report of the evaluation and the comparison of the pressure media from the viewpoint of NMR/NQR technique. Therefore, we evaluated the pressure quality of the representative pressure-transmitting media by NQR. The Cu-NQR measurements of Cu2O were performed using a phase-coherent pulsed NQR spectrometer in the resonance frequency range 26–28MHz. We used commercial Cu2O powder (99.9%). We performed the NQR measurements at 4.2K and room temperature (about 300K). The obtained spin echo data were Fourier transformed to NQR spectra. In most cases, the line shape at a pressure was obtained at an excitation resonance frequency. We used conventional Cu-Be/NiCrAl hybrid piston cylinder cell. Single-layer coil wound about 20 turns was used in our measurements ( 1⁄4 3mm, l 1⁄4 5mm). The diameter of the initial sample space was 4.5mm and its length was 18mm. We used the following four representative liquid pressure-transmitting media: a 1 : 1 mixture of Fluorinert 70 and 77 (Fluorinert), Daphne 7373 oil (Daphne), a 1 : 1 mixture of n-pentane and isopentane (Pentane), and glycerin (Glycerin). We did not try a mixture of methanol and ethanol since this mixture sometimes dissolves the epoxy which is used for the pressure seal of the piston cylinder cell. Note that the configuration of the sample and the sample coil except for the pressure media was always the same inside the cell in order to compare the contribution of only the pressure media to the NQR spectra. The frequency f dependence of Cu-NQR resonance frequency of Cu2O up to about 9GPa can be used for a pressure manometer. In our previous NQR measurements of Cu2O, we revealed that the calibration formula reported by Reyes et al. is basically consistent with our extended calibration formula. Because our calibration formula was obtained only at low temperature, we used the calibration formula reported by Reyes et al. in this study, which covers the temperature range between 4.2 and 350K and the pressure range up to about 2GPa. We determined the pressure beyond 2GPa by the extrapolation of this formula. We clarified its appropriateness in our previous study. The calibration formulae are f ðP; 4:2KÞ 1⁄4 26:82þ 0:402P 1:496 10 P and f ðP; 300KÞ 1⁄4 25:99þ 0:355P 6:738 10 P. Here, f and P are in MHz and in GPa, respectively. With these formulae and the spectral center f of NQR of Cu2O, we derived pressure inside the cell. In Fig. 1, we show the P dependence of the full width at half maximum (FWHM) of Cu-NQR spectra of Cu2O with the various pressure transmitting media, Fluorinert, Daphne, Pentane, and Glycerin. The data are obtained at room temperature [about 300K, Fig. 1(a)] and 4.2K [Fig. 1(b)]. The accuracy of the data is within 5%. In the inset of Fig. 1(b), we show the Cu-NQR spectra of Cu2O obtained at room temperature under the pressures of 2.15GPa (Fluorinert) and 2.29GPa (Pentane). The most obvious feature of the results is the P dependence of the FWHM of the Fluorinert. The FWHM of the Fluorinert linearly increases with increasing P. This is clear in the inset of Fig. 1(b), which shows that the line shape of the Fluorinert is much broader than that of the E-mail: hideto@nmr.s.chiba-u.ac.jp Journal of the Physical Society of Japan Vol. 76, No. 12, December, 2007, 125001 #2007 The Physical Society of Japan