Tc measurements on KHg are summarized in Table 
. The hydrogenation had very little effect on the
superconductivity of pink specimens, which initially had a
narrow transition and Tc 1.5 K. The transition of
a pink C4KHg sample before and after hydrogenation
is shown in Figure b). On the other hand,
hydrogenation had two major effects on the gold samples. The
first effect was to narrow the superconducting transition,
decreasing Delta Tc/ Tc by as
much as a factor of two. In addition, Tc in the
hydrogenated gold samples was about 1.5 K, just as in the
pink specimens. Superconducting transitions before and after
hydrogenation for a gold sample are shown in Figure
a). The effect of hydrogen on a specimen with an initially
broad transition but a fairly high Tc is shown in
Figure c). Notice that the 3
transitions before hydrogenation are quite different from one
another, whereas afterward the three all have Tc
1.5 K and Delta Tc/ Tc on the
order of 10-2.
A valid question about these experiments is whether a small
fraction of the sample volume might account for the changes
seen in the superconducting transition. This concern is
especially important because the inductive measurements are
sensitive to a transition in as little as 2% of the typical
sample volume (see Section
). Also, the small size of the hydrogen uptake might lead one
to suspect that only a tiny portion of the sample was
hydrogenated. Because of the sensitivity of the transition
height to the sample's position within the secondary coil,
and the lack of information about the sample's exact
dimensions after intercalation, it is not possible to say
with any certainty what the actual areal fraction of
superconductivity before and after hydrogenation is. However,
the superconducting transitions measured both before and
after hydrogenation were always at least 2 muV high, a
considerable fraction of the 5 muV expected for the
typical sample cross-sectional area. (See the discussion of
Figure for more details.) Most
of the transitions were on the order of 4 to 8 muV
high. The Tc's quoted in Table
can be said with certainty to correspond to at least 40% of
the areal fraction, and probably they represent true bulk
superconductivity. Therefore the hydrogen-induced
Tc enhancement in C4KHg is undoubtedly
a bulk effect.
Table: Effect of hydrogenation on
Tc in KHg-GIC's. All the Tc's are for
C4KHg, except for the blue 1.879 K sample, which
was C8KHg. dagger indicates the
second hydrogenation of a previously hydrogenated sample.
* indicates that deuterium rather than hydrogen
was added. S indicates that the
remeasurement of a transition one year after
hydrogenation.
Figure: Superconducting transitions before
and after hydrogenation in three types of C4KHg
samples. a) A gold sample. Tc increases from 0.88
K to 1.54 K, and Delta Tc/ Tc
decreases from 7.3× 10-2 to 7.8×
10-2. b) A pink sample. Tc is almost
constant; Delta Tc/ Tc decreases
from 4.7× 10-2 to 2.2×
10-2. c) A copper-colored sample. Tc
increase from 1.32 K to 1.50 K; Delta Tc/
Tc decreases from 0.138 to 6.47×
10-2.
Another issue related to sample homogeneity is the question
of multiple superconducting transitions. It is reasonable to
ask whether the hydrogenated Tc = 1.54 K specimen
(whose transition appears in Figure
a)) did not still have a transition at 0.88 K, its
pre-hydrogenation Tc. The answer is no, within the
limits of the sensitivity of the apparatus. A second,
lower-temperature transition was always checked for and never
found in samples whose Tc was increased by
hydrogenation. The lack of a second transition is additional
evidence that hydrogen increased the bulk Tc.
A related hypothesis is that the transition narrowing is merely due to suppression of the superconductivity of the lower- Tc phase. This assertion cannot be correct since Tc enhancement was also seen in GIC's with initially narrow transitions.
Because of the slow kinetics of hydrogen uptake in HOPG-based
GIC's, one might wonder whether Tc would increase
above 1.5 K through further hydrogen additions. As is shown
in Table , a second 5-minute
hydrogen exposure had almost no effect on Tc or
Delta Tc/ Tc. As another check
of time-dependent effects, the superconducting transition of
one hydrogenated GIC was remeasured after a wait of one year.
The Tc increased only from 1.52 K to 1.54 K, which
is within the experimental uncertainty. The (00l)
x-rays also did not show any change over this period.
In order to get more clues about the origin of the
Tc enhancement, one copper-colored
C4KHg was exposed to deuterium rather than
hydrogen. Deuterium appears to have the same effect as
hydrogen, as is shown in Table .
The lack of any isotope effect associated with the
hydrogenation suggests that the Tc enhancement is
not due to optic phonons, the mechanism responsible for the
Tc enhancement in the transition metals.[68]
Table also shows that one
C4KHg specimen showed only a small Tc
increase as a result of hydrogenation, from 0.72 K to 0.94 K.
The reason why the final Tc of this sample was not
1.5 K is not known. The fact that this sample had the lowest
Tc of any of the measured samples suggests that it
may have been too disordered to improve substantially upon
hydrogenation. Or alternatively, the hydrogen uptake rate in
this specimen may have been especially slow.
In light of the results on C4KHg, the apparent decrease of Tc in C8KHg by hydrogen addition is quite surprising. It would be a mistake to make very much of this result since hydrogen was added to only one stage 2 sample. The hydrogenation of C8KHg deserves further investigation.
The data in Table tend to underline the
differences between the pink and gold C4KHg GIC's
that were summarized in Table .
These experiments show that adding hydrogen to a gold sample
seems to change its superconducting properties into those of
a pink sample. Hydrogenation of a sample with a broad
transition seems to turn the lower- Tc material
that forms the foot of the transition into Tc =
1.5 K material. Hydrogen appear to have no effect on regions
that initially have Tc = 1.5 K. The striking
aspect of these results is thus not that hydrogen raises
Tc but that hydrogen appears to be transforming
one type of C4KHg into another since the end state
appears to be always the same.
By what means does hydrogen transform the lower- Tc material into Tc = 1.5 K material? And what is the essential difference between the lower and higher- Tc material? The next section contains some speculative attempts at answers to these questions.