In situ monitoring of crystallographic changes in Pd induced by diffusion of D

Crystallographic changes in a palladium wire cathode were monitored in situ, as deuterium was electrochemically deposited on the surface and diffused radially into the wire. Initially, the wire was pure Pd. A constant electrolysis current density of 1 ${\mathrm{m}\mathrm{A}/\mathrm{c}\mathrm{m}}^2$ was maintained and D slowly diffused into the wire. As the D concentration increased, the wire transformed from pure Pd, to the α phase, and finally into the β phase. This reversible phase transformation begins on the surface and progresses radially inward. During the experiment, x-ray-diffraction data were collected from a volume element of about 180 pl. This volume element was systematically moved in 50-μm steps from the edge to the center of a 1.0 mm diameter Pd wire. Throughout the course of the experiment, the bulk value of x in ${\mathrm{PdD}}_x,$ as determined from simultaneous measurements of the electrical resistivity, increased from 0 to ∼0.72. For each setting of the volume element, a monotonic increase in the volume of the α phase was observed, until the material entered the two-phase region. Once the β phase appeared, the volumes of both phases decreased slightly with continued loading. The integrated intensities of diffraction peaks from each phase were used in conjunction with the known phase diagram to estimate the rate of compositional change within the volume element. The diffusion rate for the solute atoms was estimated to be $57\pm 8\mathrm{n}\mathrm{m}/\mathrm{s},$ based on the temporal and spatial dependence of the integrated intensities of the diffraction peaks from each phase. These data also were used to evaluate the time dependence of the concentration of the solute atoms $\partial c/\partial t$ and their diffusivity D. The value of $\partial c/\partial t$ increased linearly from $6.2\times {10}^{-5}{\mathrm{s}}^{\mathrm{-}1}$ at the surface, to $7.6\times {10}^{-5}{\mathrm{s}}^{\mathrm{-}1}$ midway into the wire. D was computed to be $(3.1\mathrm{}\pm 1.0)\times {10}^{-11}{\mathrm{m}}^2/\mathrm{s}$ when the transition began at $r=250\mathrm{}\mu \mathrm{m};$ 2 ks later it had decreased to $(2.1\mathrm{}\pm 0.3)\times {10}^{-11}{\mathrm{m}}^2/\mathrm{s}.$ This may be due to the fact that the volume of the β lattice also decreased during this period.