Our calorimetry measurements have revealed a series of unexpected behaviours along the transition line separating the metal from the magnetic quantised phases:

-1- The specific heat jump at the metal-FISDW transition, for low magnetic fields, is comparable with the value predicted by the BCS model of superconductivity [1] (weak coupling), while for strong fields the jump rises toward a very strong coupling (beyond that of lead) (Fig. 1).

FIG. 1. Electronic specific heat Cel/T as a function of the temperature, for two field values: 5.75 and 10 teslas. The horizontal dashed line corresponds to the g  value of the normal metal. Note the BCS shape of the low-field metal/FISDW transition (DC/g Tc~1.4), and the sharp departure from the BCS behaviour at high fields (DC/g Tc~5.7) [publi 10].

-2- The jump at the transition oscillates along the quantised FISDW phases (Fig. 2). This result initiated a theoretical calculation in the framework of the "quantised nesting" model [2]. For strong fields, the jump rises to a "strong coupling" value that is NOT explained by the available theoretical models.

FIG. 2. The specific heat jump at the transition oscillates ("jumps") at each FISDW transition (arrows) [publi 10]. The "jump of the jump" amplitude stays moderate in the low field region (where the jump DC/g Tc lies around the BCS value). However the last jump exhibits a very strong magnitude around 9 teslas ("jump of the jump of the jump"). It corresponds to the "strong coupling" transition Fig. 1.

[1] Theory of Superconductivity, J. Bardeen, L.N. Cooper and J.R. Schrieffer, Phys. Rev. 108, 1175 (1957).
[2] The specific heat jump at the magnetic-field-induced metal-spin-density-wave transition in quasi-one-dimensional conductors, G. Montambaux, J. Phys. C20, L-327 (1987).

-3- Our simultaneous specific heat and thermal conductivity measurements provided spectacular information about these critical phenomena. Fixed field specific heat measurements indicate the presence of two successive transitions (Fig. 3). Thermal conductivity strangely displays that only one of them exhibits a noticeable anomaly: the lowest temperature one. This difference in criticality behaviour is not explained.

FIG. 3. As two lines are crossed in a fixed field (here B=7T), Cp exhibits two jumps (which can be deconvolved), while thermal conductivity along the c* axis only exhibits one anomaly (vertical arrows) [publi 15].

-4- In strong fields, close to the "strong coupling" jump, other phenomena are revealed: two specific heat jumps of same amplitude may yield very different thermal conductivity anomalies, whereas symmetrically two very different jumps may give rise to similar amplitude anomalies [publi 21]. The magnetocaloric effect (see the following) leads in the same field range to an irreversible behaviour, while the effect is reversible on crossing the low field phase transitions [publi 8].