We find that the carrier concentration in Stage II AsF5- graphite can be varied over a wide range, f≃.30–.46. By studying the Fermi surfaces vs. f via magnetooscillations, we find that the band parameters are essentially unchanged from pure graphite. Thus suggests that all stage two compounds should display a universal behavior as a function of f, and we adduce evidence of this from the literature. We further begin an analysis of extra oscillation frequencies. In AsF5-graphite, mixing frequencies are probably caused by Fermi-level modulation (Vinter-Overhauser effect), while a low frequency (β) is associated with a superlattice.

Tight binding crystal orbital calculations on infinite layers of graphite have been performed for charge transfer (q) values ranging from −0.15 to +0.15 e/carbon atom. The rC-C carbon-carbon bond lengths have been optimized at several q values. The change of the calculated rC-C values as a function of q fits very well with the experimentally observed variations of the C-C bond lengths of both acceptor and donor compounds of intercalated graphite. The asymmetry of the variation of rC-c with respect to the sign of the charge transfer is related to the slightly antibonding nature (at the level of second neighbor interactions) of the π-electrons around the Fermi level of pristine graphite, similar to those in polyacetylene.

Recent results on the temperature and stage dependence of the thermal conductivity of various graphite intercalation compounds are reported and briefly analyzed. In the lowest temperature range the in-plane thermal conductivity which is mainly electronic, is higher than that of pristine graphite. Around room temperature, in spite of an increased electronic conduction, a decrease in the total thermal conductivity relative to pristine graphite is observed. This shows that the dominant effect is the decrease of the lattice contribution due to increased phonon scattering by lattice defects. The c-axis conduction is attributed entirely to phonons and the anisotropy ratio is of order of 102 in the liquid helium range and increases up to several hundreds around room temperature. Some results on the temperature variation of the in-plane and c-axis thermopower for donor and acceptor compounds are also presented and discussed.

Temperature variations of the thermopower (TEP) of acceptor graphite intercalation compounds (GIC) are very different from that of pristine graphite. At low temperatures the TEP increases monotonically with T, then levels off above 150 K. This behavior is ascribed to the phonon drag effect. In the region where the TEP is nearly constant, phonon relaxation is mainly controlled by the Rayleigh scattering due to point defects or impurities. This process leads to T-independent phonon drag TEP. The importance of Rayleigh scattering is due to the large cross section diameter of the Fermi surface in GIC. At low temperatures where the boundary scattering becomes important, the TEP is proportional to T3 . Detailed calculations are carried out by solving the phonon-carrier coupled Boltzmann equation.

C-axis electrical resistivities were measured from 2-300K on SbCl5-graphite intercalation compounds, using both HOPG and single crystals as host materials. The resistivity of these compounds is up to an order of magnitude higher (˜10−2Ωm) than that of pure HOPG graphite. Yet, for stage n≤3 (in the case of HOPG), a metallic temperature dependence of ρC is observed throughout the entire investigated range. Higher stage compounds show metallic behavior at high temperatures, but exhibit a crossover to an activated dependence as the temperature is lowered. All samples show hysteretic behavior with cycling. A model, based on defect-mediated short-circuiting channels, is proposed to account for the existence of a metallic temperature dependence in conjunction with large resistivity.

Temperature-dependent c-axis resistivity measurements on LiC6, KC8, RbC8, CsC8 with and without substitutional boron in the starting material, and KHgC4 are presented. The overall behavior is metallic except for the MC8 compounds below 50-100K where Pc increases with decreasing T. This effect appears to be an intrinsic property of stage 1 materials with commensurate 2×2 superlattices.

Calculations of the magnetoresistance of graphite acceptor compounds are made using a tight binding model for the carrier dispersion proposed by Blinowski et al, and measured values of the zero-field resistivity. It is shown that, if a reasonable physical model is used for the mobilities, the magnetoresistance cannot be fitted with two- or three-carrier models. Suggestions for the origin of the magnetoresistance are made.

In first stage graphite intercalation compounds with cylindrical Fermi surfaces and isotropic relaxation times, no magnetoresistance occurs. Holzwarth pointed out that trigonal band warping provides a finite magnetoresistance in first stage compounds. A calculation of the galvanomagnetic coefficients of stage-2 acceptor compounds is performed including the effect of trigonal warping on the bands. In the present problem a perturbation calculation provides a good approximation. Without trigonal warping, the stage-2 compounds exhibit a finite magnetoresistance since they have two kinds of holes. Retaining terms up to order (γ3/γ0)2, the weak field magnetoresistance is calculated in limit H→0; Δρ/ρoH2 ≍ (e/h2c)2 (τ/EF)2 (3γ0b/2)4[1/4(γ1/EF)2+57/2(γ3/γ0)2] where b = 1.42 Å is the nearest C-C distance, EF the Fermi energy and τ the carrier relaxation time. The first term in curly brackets is due to the existence of the two kinds of holes. This formula indicates that the term due to trigonal warping is much larger than the other term.

Superconductivity in graphite intercalation compounds, C4nMHg (M:K, Rb, n=1,2) and C4KTl1.5 is studied in ambient as well as elevated pressures. It is argued that the major role in the superconductivity of this class of materials is played by electrons in the intercalant bands rather than those in the graphite bands. It is experimentally demonstrated that the two-dimensional anisotropy of the superconductivity is controlled by stage and pressure.

Measurements of the pressure (P) dependence of the superconducting transition temperature Tc of KHgC4, KHgC8, KTℓ1.5C4 and KC8 are reported. KC24 was found to be nonsuperconducting to a limiting temperature T ≈ 0.07 K for P ≤ 10 kbar, and RbC8 was found to be nonsuperconducting to T ≈ 1.30 K for P ≤ 22 kbar. The Tc of KHgC8 was found to monotonically decrease at a rate dTc/dP = −6.5 × 10−5 K/bar from a zero pressure Tc = 1.850 K. The Tc of KTℓ1.5C4 initially decreases at a rate dTc/dP≤ −4.5 × 10−5 K/bar from a zero pressure Tc = 2.54 K. KHgC4 displayed a broad zero pressure superconducting transition with an onset at T ≈ 1.42 K, followed by an abrupt narrowing of the transition and rapid decrease of Tc at a rate dTc/dP = −5 × 10−5 K/bar at p > 1 kbar. The Tc of KC8 was observed to jump from a zero pressure value of 0.13 K to a broad transition with an onset at ˜ 1.7 K for P > 2 kbar.

We report on the structural, lattice and electronic properties of stage 1,2,3 potassium-amalgam GIG and the relation of these properties to the observed superconducting transition temperature. The ultramicrostructure of the intercalant domains is investigated using lattice fringe imaging. Our observation of macroscopic (√3×√3),(2×2) and (√3×2) domains is consistent with the basal plane zonefolded phenomena observed in the Raman spectra. Shubnikovde Haas measurements of the Fermi surface show high frequency (>1000 T) oscillations in stage 2, identified with the intercalant band.

Precise measurements of the temperature dependence of the magnetic susceptibility for stage 2 and 5 graphite-CoCl2 intercalation compounds are reported. Comparison of the experimental results with theoretical calculations based on a two-dimensional planar model show agreement with theory, suggesting that graphite intercalation compounds represent ideal two-dimensional magnetic systems.

The heat capacity Cp of graphite-CoC12 (stage 2) is measured at zero and high (up to 14 T) magnetic fields (H applied in-plane). By suppressing the magnetic contribution to Cp at the highest fields, we are able to decompose Cp into its electronic, lattice, and magnetic contributions. The magnetic heat capacity CM is seen to have a broad peak at ⋍ 9.1 K. The shape of this peak is consistent with the reported Monte Carlo calculations of CM based on a twodimensional xy model.

Dissolved cobalt(0) complexes of the type [L3L'Co] (L = phosphanes, e.g. PMe3, L' = olefins, e.g. C2H4) are reversibly reduced by alkali metals A (A = Li, K, Rb, Cs)

n[L3L'Co] + A ⇋ A[L3L'Co]

and hence can be used as A-carriers. These carrier complexes A[L3L'Co]n are even soluble in apolar solvents like pentane.

Action of [L3L'Co] plus A in pentane solution on graphite yields binary intercalation compounds ACn. By contrast, conventional ambient temperature A-transfer reagents (e.g. solutions of A in naphthalene-ether or in NH3) require strongly polar solvents and yield ternary intercalation compounds A(solv)yCn.

The “reducing power” of the alkali cobaltates is close to that of free A: alkali-rich phases like 1st stage KC8 or LiC6 or highly doped polyacetylenes (e.g. K(CH)5) are readily prepared. If intercalation of solvated species is unlikely, the A-transfer reactions may also be performed in polar solvents like ethers.

Gas-phase intercalation of graphite by nitric acid is a one-step process. The weight uptake of the sample is a function of nitric acid vapor pressure. A pure second stage compound was formed when the HNO3 reservoir was maintained at 17°C. Only a fourth stage compound was formed when the acid was kept at 0°C. The product gases due to intercalation and gas species evolved during deintercalation were analyzed by mass spectrometry. NO2 and H2O were the major components detected from the intercalation product gases. A small amount of oxygen was also present. The existence of O2 is probably due to the photo-chemical decomposition of nitric acid. As such, the photodecomposition of nitric acid is not a contributory factor in the intercalation chemistry.

Graphite intercalation compounds can be used as electrode materials in batteries owing to the high electronic conductivity of carbon layers and the easy diffusion of ions between the layers. Moreover intercalation has the beneficial effect of impeding some parasitic reactions detrimental to good electrode reversibility. After a brief review of recent proposals, emphasis is given to transition metal chlorides. It is shown that, besides the nickel chloride-graphite compound proposed two years ago as a cathode material for alkaline batteries, other metal chlorides intercalated into graphite are excellent cathode or anode materials.

The discharge reaction mechanism for two types of graphite fluorides, (CF)n and (C2F)n cathodes of lithium cells utilizing a 1 M LiClO4-propylene carbonate (PC) system were studied by means of x-ray diffraction, ESCA, NMR and transmission electron microscopy. It has been revealed that the discharge reaction proceeds through the formation of an intermediate solvated ternary compound of (C·F·Li)·PC or (C2· F·Li)·PC, and then they decompose to graphite with low crystallinity and LiF. The decomposition process is not involved in the electrochemical reaction and this discharge reaction infers the open-circuit voltage of cells. New graphite intercalation compounds, CxF(MFn)y(MFn=AIF3, MgF2) were prepared under a fluorine atmosphere of 1 atm in the temperature range between 20 and 400°C. These compounds are highly stable even in moist air and have high electrical conductivity. The discharge potentials of CxF(MFn)y-lithium cells are 2.5–2.8 V at current densities 40–400μA, which are higher than those of (CF)n or (C2F)n-lithium cells below 300μA.

The equations of nonequilibrium thermodynamics are used to investigate how temperature gradients may be utilized to enhance performance of solid state batteries. One result is that the efficiency of a battery with doped cerium oxide as the electrolyte may be increased by imposition of a temperature gradient. This may lead to greater use of electrolytes of appreciable electronic conductivity. Moreover, temperature gradients may be used to increase mobile ion diffusivity in intercalation electrodes during battery discharge and recharge.

In situ Raman scattering studies of the high-frequency carbon intralayer modes in graphite-H2SO4 have been carried out during the electrochemical intercalation of graphite in sulfuric acid. The data show that within an optical depth of ∼ 200 Å the entire c-face undergoes a sudden transition to the next lower stage (n−1) at the moment when the bulk has just completed forming stage n. Raman data collected in the “overcharge” and “overoxidation” regimes of the electrochemical synthesis indicate rapid shifting of the peaks during these times, indicating a significant increase in the oxidation of the carbon layers.

First-order transitions to dilute stage one from stages 2,3 and 4 and from mixed stage 1+2 are observed in Li-graphite as predicted by Safran. The upper phase boundary is asymmetric in concentration and is sharply peaked about x ∼ 0.4. The phonons in the dilute stage 1 have energies between those of LiC6 and graphite. Unlike the stage 2 compounds, the compressibility does not scale in a simple way with intercalant density.