The C4KHg specimens used in these studies were synthesized using the two-zone vapor-phase intercalation technique pioneered for these materials by El Makrini et al. [21] The starting amalgams ranged in composition from K2Hg3 to K5Hg4. No systematic effects were seen as a function of the starting alloy composition. The amalgam temperature during intercalation was varied between 200°C and 260°C. In addition, the graphite temperature was independently varied from 0°C to 10°C above the amalgam temperature. The intercalation reaction was always allowed to proceed for at least two weeks. The intercalation ampoules were withdrawn from the furnace after it had cooled nearly to room temperature.
Timp found that the application of a 10°C temperature difference between the graphite and the amalgam during intercalation increased disorder and lowered Tc.[53] He reported a gold color for low-Tc disordered specimens and a pink color for ordered Tc = 1.5 K ones.[53] Timp[53] also reported that a temperature difference between the graphite and the amalgam increases the amount of the (sqrt3 × 2)R(30°, 0°) phase as compared with the majority (2 × 2)R0° phase.[53] The (2 × 2)R0° ordering is associated with a 10.24 Å repeat distance and has been called the alpha phase.[20] The (sqrt3 × 2)R(30°, 0°) ordering has been positively identified with the repeat distance Ic = 10.83 Å by Kamitakahara et al. [31] The Ic = 10.83 Å phase is called the beta phase.[20] Kim and coworkers[34] report that application of a temperature difference during intercalation increases the amount of the beta phase, in agreement with the finding of Timp.[53]
In the present study, X-ray and neutron diffraction were used to characterize the C4KHg specimens for staging fidelity and phase purity. The x-ray characterization was performed at MIT using Mo Kalpha radiation in a theta-2theta geometry. Neutron diffraction (00l) spectra were taken at the National Institute of Science and Technology research reactor using the BT-4 triple-axis neutron diffractometer with thermal neutrons of wavelength 1.528 Å. Neutron diffraction spectra of pink (single-phase) and gold (mixed-phase) C4KHg specimens are shown in Fig. 1. Interestingly, neutron diffraction revealed the presence of about 10% beta phase in a sample which showed only alpha-phase peaks in x-ray diffraction spectra. This inconsistency is undoubtedly due to the limited penetration depth of x-rays in C4KHg (about 35 µm). Obviously caution is advisable when using x-ray diffraction to interpret the results of experiments that measure the bulk properties of C4KHg.
In all, over twenty batches of C4KHg were prepared and characterized. The assertion that a 5-10°C temperature difference increases the amount of the beta phase was strongly confirmed. However, the beta phase can also be formed under nominally isothermal intercalation conditions, sometimes even in the same ampoule with pure alpha-phase specimens. It seems likely that variables that are difficult to control precisely (such as the condition of the edges of the graphite flake) may affect the speed of the intercalation reaction and impact the final ratio of the two phases in a sample. Pure beta-phase specimens are never produced, as previously noted by Lagrange.[38] Pure alpha-phase specimens were routinely produced and found to be pink in color, while mixed (alpha + beta) phase specimens were variously gold or copper in color.
In order to derive quantitative structural information from the neutron diffraction data, the integrated intensities of the diffraction patterns were fit to a 5-layer model[39] of the unit cell. A gradient-least squares optimization algorithm[6] was employed to obtain the residual-minimizing parameters.[24] The Lorentz factor was corrected for the specimen's mosaic spread using a standard procedure.[5] Further details of the analysis are given in Ref. [10]. Only the majority alpha-phase intensities were fit because the number of observed beta-phase peaks was too small. The results of the analysis are compared to the parameters obtained by previous investigators in Table I.
Careful examination of Table I shows that the structural and stoichiometric differences between the majority phases of the gold and pink specimens are small. The parameters in Table I obtained from the structure-factor fits are likewise in good agreement with those of Ref. [60]. In conjunction with the results of wet chemical analysis on samples whose Tc was known,[10] the results in Table I contradict the assertion[45,46] that the gold mixed-phase C4KHg samples are mercury-deficient. The general conclusion from the neutron diffraction experiments is that the majority-phase material in the gold and pink samples appears to be the same. The only reproducibly observed difference in the normal-state properties of the gold and pink C4KHg samples is that the gold (Tc = 0.8 K) samples are mixed-phase, while the pink (Tc = 1.5 K) samples are pure alpha phase. The superconducting properties of the single-phase and mixed-phase samples are compared in detail below.