Specific heat anomalies and spin-spin interactions in carbons: A review

In the introductory part of the paper the classification of carbons according to mode of preparation and structure into soft and hard, carbon blacks, pyrocarbons, evaporated films, and whiskers, and their response to heat treatment, including the kinetics of carbonization and graphitization, are briefly described. The changes occurring in electronic properties as the hopping conductivity of raw carbons transforms into band conductivity upon departure of organics and the further evolution of bands toward graphitic structure, as well as various doping techniques, are then briefly discussed. In the second part, after a brief review of the early research on low-temperature specific heat of carbon materials heat treated to graphitizing temperatures and corresponding efforts at interpretation, information gathered through the last ten years on specific heat anomalies in variously heat-treated and neutron-irradiated soft and hard carbons is reviewed. In soft carbons two specific heat peaks at about 0.3 and 0.65 K, and a large, linear term of nonconduction origin, are observed at all stages of carbonization up to quite high degrees of graphitization or when disorder is introduced by neutron irradiation. In this latter case a spectacular disappearance of the anomalies by subsequent thermal anneal to 300°C has been observed. The peaks and the linear term are shown to have a magnetic origin, the specific heat varying greatly with applied magnetic field. In hard carbons no clear specific heat peaks were found—only a small, broad hump at around 0.3–0.7K and continuously increasing values of C/T down to lowest temperature (0.1 K), and again a large linear term. As in soft carbons the anomalies are greatly diminished by heat treatment to 2700°C or above, but are brought back by neutron irradiation. Since all the facts point toward a localized free electronic spin origin of the anomalies, in the third part a condensed discussion of the electron spin effects in carbons is presented. After a short description of the ESR of conduction carriers in well-graphitized materials, the technique of separation of the localized spin from conduction carrier contribution and the mixing of g- values and of the widths by exchange in carbons are discussed. Information on the heat treatment dependence, neutron irradiation, and thermal anneal and on doping effects for various types of carbons heat treated above 1500°C follows. Next, the formation of localized spins in carbonization and condensation processes of organic materials, and their subsequent gradual decay after appearance of conductivity, with an exchange narrowing of the ESR line somewhat after the concentration crosses its peak, as well as production of artificial spin centers by gas reactions, all in temperature range below 1000°C, are described. Finally the difficulties and uncertainties in obtaining data, caused by broadening effects in the heat treatment range 1000–1500°C, are pointed out. In the last part of this review the spin concentration data supplied by ESR studies are compared with specific heat data. It turns out that only about a quarter of all localized spins are condensing in the two peaks, the rest of them (or a part) being involved in the linear term anomaly. Whereas a general picture of the specific heat anomalies can be outlined, none of the known exchange interactions can explain the values at which the fractions condense into presumably an antiferromagnetic phase, nor the high values necessary for the conflict areas to be responsible for the linear term. Thus the most basic questions remain unanswered and further experimental studies, coupled with some theoretical attempts, are badly needed.