Incomplete Crystallization of Skutterudites
- PDF / 364,695 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 73 Downloads / 224 Views
0886-F09-12.1
Incomplete Crystallization of Skutterudites Arwyn L. E. Smalley1 and David C. Johnson Department of Chemistry, 1253 University of Oregon, Eugene, OR 97403-1253, U.S.A. 1 Department of Chemistry, Norwich University, 158 Harmon Dr. Northfield, VT 05663, U.S.A. ABSTRACT A series of CexCo4Sb12 samples were synthesized, with x = 0 to 2. The crystallinity of the samples was studied using quantitative Rietveld analysis, and revealed that the samples did not crystallize completely. These data correlated with changes in the lattice parameter, DSC peak temperature, and DSC peak area. The crystallinity was confirmed using electron backscatter diffraction (EBSD), which showed crystallites with a background of an amorphous matrix. This unusual morphology may improve thermoelectric properties by decreasing thermal conductivity. INTRODUCTION In our research synthesizing metastable skutterudites, we have consistently formed samples that are single-phase based on X-ray diffraction, even when the stoichiometry of the skutterudite formed did not match the stoichiometry of the sample. For example, in a series of cerium-filled CoSb3 samples it was necessary to add excess cerium to the samples in order to achieve more complete filling of the void sites [1,2]. Despite this, even in samples with twice the stoichiometric amount of cerium required to fill all the void sites, no crystalline cerium or other cerium-containing impurity phases were observed in the X-ray diffraction patterns. Magnetic susceptibility data indicated that more than one phase was present in the compound [2]. This prompted us to use Rietveld analysis of X-ray diffraction data to quantify the amount of filling and the percent of crystalline skutterudite present. The incomplete crystallization of the samples was confirmed using using electron back-scatter diffraction (EBSD). EXPERIMENT All samples were synthesized in a custom-built, ultra-high vacuum, elemental deposition system, described elsewhere [3]. Samples were synthesized in a 10-6-10-7 torr atmosphere. Elemental sources (Co, Ru, Ni, Fe) were alternately deposited from an electron beam gun at a rate of 0.2 or 0.5 Å/sec, except Sb, which was deposited from an effusion cell at a rate of ~1 Å/sec. A computer-controlled quartz crystal monitoring system was used to control elemental layer thickness, resulting in repeat thicknesses of 10-15 Å, and total thicknesses of 2000-4000 Å, producing 5-30 mg of sample. Samples were deposited on a PMMA-coated silicon wafer on which had been affixed a silicon chip. The samples were soaked in acetone to dissolve the PMMA, and then the powder collected by vacuum filtration. X-ray diffraction studies were conducted on the silicon chip and the powder sample using a Philips X’Pert diffractometer and a Scintag XDS-2000 θ-2θ diffractometer. Samples were annealed for one hour at a time in a Thermolyne 1500 box furnace under nitrogen atmosphere, at
0886-F09-12.2
temperatures sufficient to crystallize the samples (detailed elsewhere) [2]. The powder sample was mixed with si
Data Loading...
