P. Hjort et al., RESISTIVITY AND HYDROGEN UPTAKE MEASUREMENTS IN EVAPORATED MG FILMS AT 350 K, Journal of alloys and compounds, 234(2), 1996, pp. 11-15
Resistivity measurements were performed during hydriding-decomposition
cycles of UHV-prepared Mg films. The hydrogen exposures were made in
situ at p(H2) = 0.6 torr and 350 K. To obtain measurable uptake rates
at these conditions a thin layer of Pd (approximate to 7 nm) was depos
ited on top of the typically 380 nm thick Mg films. Mg and Pd film thi
cknesses and hydrogen uptakes were determined by the frequency shift o
f a quartz crystal microbalance (QCM). Both the QCM and the quartz sub
strate onto which the Mg resistor film was deposited were mounted on t
he same sample holder. The p, T conditions were chosen so that hydroge
n was distributed nearly homogeneously through the sample. The total h
ydrogen uptakes were in most cases kept small (<40 at.% H) to minimize
the morphology changes. The resistivity vs. H concentration relation
is, within experimental uncertainty, linear up to x = (H/Mg) approxima
te to 0.4 according to the formula rho(x) = rho(0)(1 + c . x), c = 1.0
8 +/- 0.25. The relatively large uncertainty is caused by a spread in
the uptake data and by the irreversible resistance changes during hydr
iding-decomposition cycles. Both these uncertainties could be attribut
ed to morphological changes induced by the volume expansion-contractio
n accompanying the cycles. Exploratory measurements were also made at
much larger H uptakes. These measurements reveal a rapid reversible re
sistivity increase at (H/Mg) approximate to 1.2, which can be qualitat
ively attributed to the metal-insulator transition expected for the Mg
-H-x system when the hydride phase becomes continuous. The measurement
s and their quantitative interpretation were complicated by the morpho
logical changes of the sample(s) caused by volume expansion-contractio
n.