Abstract

Y4A1209 has a monoclinic crystal structure, space group P21/c [1]. The atomic parameters of this structure were refined by single crystal X-ray and neutron diffraction and by Rietveld analysis for the X-ray synchrotron powder diffraction pattern [2, 3]. Recently, we revealed the high-temperature phase transformation of Y4AIzO9 at 1377 °C by high-temperature differential scanning calorimetry (DSC), high-temperature dilatometry and hightemperature X-ray powder diffraction (XRD) [4, 5]. The low-temperature phase transformed to the high-temperature phase with a small change in endothermic enthalpy and small volume shrinkage. Thermal hysteresis and athermal phenomena were observed at this phase transition. Judging from these results, the phase transition was thought to be martensitic. Takizawa et al. reported a stress induced phase transition from a tetragonal structure to a monoclinic structure for (Hol_xLax)4Al209 and (Yl_xLax)4AlzO9 at room temperature [6]. In our previous study, peaks in the XRD of the high-temperature phase were indexed with a monoclinic cell [4]. However, the crystal structure of the high-temperature phase has not been clarified. This letter describes the thermogravimetric analysis of Y4AI209 and analysis of the crystal structure of the lowand high-temperature phases by the Rietveld method for XRD patterns obtained at room temperature, 1300 °C and 1400 °C. Y4A1209 was prepared from powders of Y203 (99.9% purity) and A1203 (99.9% purity). Stoichiometric amounts of powders were weighed and mixed in an agate mortar. The mixture was pressed into pellets and heated on a Pr 60%-Rh 40% plate in a molybdenum silicide furnace at 1600 °C for 10 h. The obtained pellets were powdered for thermogravimetry (TG) and XRD. TG was carried out with a high-temperature thermobalance (TG-DTA 2200, Mac Science). The rate of heating was 10 °Cmin -1. The XRD pattern was obtained at room temperature using CuKo~ radiation with a pyrolitic graphite monochromator and a diffractometer (RAD-C, Rigaku; tube voltage and current: 40kV and 20 mA). High-temperature XRD patterns were obtained with a graphite monochromator using a diffractometer with a high-temperature furnace attached (MXP18, Mac Science, tube voltage and current: 45 kV and 360 mA). The temperature was measured with a Pt-Rh13% thermocouple set into a Pt sample holder. Since the phase transition had the thermal hysteresis [4, 5], high-temperature XRD was performed at 1300 °C on heating and at 1400 °C on cooling f rom 1500 °C. The intensity data were collected from 10 to 70 ° in 20 with a sampling step width of 0.04 °. Rietveld analysis of the XRD data was carried out using the RIETAN program [7]. A weight loss of about 0.18% was detected for the powdered Y4A1209 sample from room temperature to 350 °C. This was probably due to detachment of adsorbed water. Above this temperature to 1500 °C, the weight change of the sample was less than 0.005%. This result indicated that oxygen atoms and other elements did not transfer from or into the sample within this weight range. Fig. 1 shows the profile fit and difference patterns of Rietveld analysis for the XRD data of Y4AI209 at room temperature, 1300 °C and 1400 °C. The solid lines are calculated intensity profiles and the dots crossing the profiles are observed intensities. The short vertical lines show the positions of possible Bragg reflections. The differences between observed and calculated intensities are plotted below the profiles. Table I lists the refined lattice p0rameters of the primitive monoclinic cells and R faqtors of the Rietveld analysis [8]. In the Rietveld refinement for the XRD pattern at room temperature, the calculated pattern fits the observed one well. The lattice parameters of the present study were almost equal to those reported by Lehmann et al. [2] for the a-axis and /3, and were slightly smaller than those for the band c-axes. The relative intensities of the main diffraction peaks of the high-temperature phase at 1400 °C were not so different from those of the low-temperature phase at room temperature and 1300 °C (Fig. 1). Since the peaks overlapped at higher Bragg angles, the relative intensities of the peaks at 20 from 10 to 25 ° were compared. For the high-temperature phase, peaks of 10 0 and i 12 became very small, and the relative intensities of 1 10 and i 02 were larger than those of the low-temperature phase. Based on the peaks of the powder diffraction pattern of the high-temperature phase, we were not able to adopt any space group except P21/c. Therefore, we analysed its crystal structure with the space group of P21/c. The lattice parameters of the face-centred cell are also listed in Table I; they were calculated from the refined primitive lattice parameters with the conversion proposed by Brandle and Steinfink [9]. The/3' angle of the face-centred cell was close to 90 ° but the lattice of the high-temperature phase did not become orthorhombic.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call