The first consideration when undertaking critical field measurements on anisotropic samples is how to mount the specimens. High anisotropy, the property of central interest, unfortunately presents the investigator with special problems. In a nutshell, the larger the anisotropy, the greater are the errors induced by small misalignments. Since most of the experiments in the present work were done on HOPG-derived polycrystals, the only concern relevant to alignment was the careful orientation of the crystalline c-axis with respect to the applied magnetic field. Due to the samples' air sensitivity, they had to be preserved inside sealed glass tubes. Therefore alignment in practice meant fixing the orientation of the GIC with respect to the glass tube. Early critical field results were found to be unreliable due to the fact that the samples were mobile inside the glass tubes.
The obvious tack to take in mounting the samples was to use the technique employed by Iye and Tanuma.[120] These workers wrapped their GIC's in Parafilm, a kind of wax paper manufactured by Union Carbide, and took the sample thus wrapped out of their glovebox directly to a coldfinger, where it was immediately mounted and cooled down. Unfortunately in the present case the available cryostat and glovebox were separated by a great distance, so this transfer method was thought to be infeasible. Also, this method presents difficulties for preserving the sample upon warming to room temperature, and it is desirable to save the samples for repetition of experiments. Therefore it was decided that the specimens would have to be always encapsulated in glass tubes for handling and transfer. The idea of wrapping the sample in Parafilm in order to fix its position within the glass tube did not seem to be practical.
Several methods were tried in an effort to fix the samples without degrading them. First, it was attempted to hold the samples in place by using a gas torch to make an indentation of the encapsulating tube around the sample's position, trapping it in one orientation. This method was unsuccessful due to the deintercalation that the heat of the torch caused. The next idea, that of gluing the samples to a holder, was discarded because of the extreme chemical reactivity of alkali metal compounds.
Success was finally achieved by mounting the specimens on
specially made metal holders whose general features are
sketched in Figure 
. In this method, sheets
of copper or brass were cut into pieces of such a width that
they fit tightly into the appropriate size of glass tube.
Copper and brass were chosen because their high thermal
conductivity should prevent the development of a temperature
gradient across the samples at the very low temperatures used
in the superconductivity studies. Also, these metals are
neither superconducting nor ferromagnetic, and so produce no
temperature-dependent inductive signal to compete with that
from the superconducting transition. After the metal pieces
were cut, they were cleaned with a weak solution of
hydrochloric acid, dipped in a bicarbonate solution to stop
the acid reaction, and then rinsed in de-ionized water. The
metal holders were then heated under to vacuum to remove any
residual water, and taken into the inert environment of the
glovebox. (See general comments on handling samples in the
glovebox in Chapter .)
Figure: A sketch of the sample holder used
in the critical field measurements. The dimension d of
the metal piece was chosen to be the inner diameter of the
sample tube so that the holder would be centered and fixed
inside the tube. A careful effort was made to orient the
carbon (graphene) planes parallel to the holder's surface.
The GIC's were affixed to the metal pieces with Apiezon N
grease.
When the metal holders were first used, the samples were
attached to them by pressure applied with a folded-over flap
of metal. This flap was squeezed over the sample with
tweezers after the sample was placed on top of the metal.
Half the time this technique was very successful: the sample
after mounting showed no degradation of its superconducting
transition or of its (00l) x-ray diffractogram. (See
Chapter for a discussion of
staging characterization of GIC's by (00l) x-ray
diffraction.) The other half of the time, the sample's
transition temperature was reduced, and its x-ray spectrum
was either attenuated or completely suppressed, presumably
due to fracture of the carbon (graphene) planes resulting
from excess pressure.
Because the sample attrition rate from this method of mounting was unacceptably high, it was decided to try attaching the samples to the metal holders with vacuum grease. Vacuum grease is considerably more inert than commercially available glues, and Apiezon N grease was used successfully in mounting C8K by Sano and coworkers.[214] In the end, the best method of mounting the samples turned out to be dabbing grease on one side with a piece of wire, and then using the grease to adhere the samples to the metal holder. The grease was left overnight in an open container in the glovebox before use so that it would outgas. Mounting with vacuum grease allowed specimens to be fixed in such a way that their Tc and x-ray spectrum were unaffected, and in addition satisfied the goal of keeping the GIC stationary even when the sample tube was shaken. The GIC's should have been firmly fixed at low temperatures where the grease can freeze them to the holder.
Once the GIC's were fixed in their glass tubes, the tubes were fixed inside the inductance bridge's primary coil with GE 7031 varnish, taking care that the tube's axis was parallel to the modulation field direction. This alignment is thought to be good to within a degree or so since the walls of the tubing and the inside of the coil are both flat. The result was that the modulation field was applied in the direction perpendicular to the graphite c-axis, where it had the least effect on the critical field measurements.