Specific heat of Ce3Bi4Pt3 at 60 T
Introduction
Kondo insulator materials (KI) [1] have attracted considerable attention during recent years in part because the synthesis of quality single-crystal samples has allowed the measurement of the temperature dependence of the spin gap with high-resolution photoemission techniques [2], [3], [4]. Another way to study the nature of the spin gap in KI is to measure the magnetic field dependence of their transport properties. The electrical resistivity [5], [6], Hall effect [6], and magnetization [7] were measured in cubic Ce3Bi4Pt3 (Fig. 1) as a function of the magnetic field in DC and capacitor-bank driven pulsed magnets. A rather large negative magnetoresistance was observed at low temperatures in single-crystal samples together with a recovery of the charge carrier concentration in fields up to 60 T, evidence that strongly suggest that the spin gap can be closed in such magnetic fields [6]. More recent low-temperature/high-field magnetization measurements in powder samples showed no indication of insulator-to-metal crossover [7], in apparent contradiction with transport results. Measurements of the specific heat in magnetic field directly probe the evolution of the excitation spectrum, and Kondo gap in particular, and can therefore provide the key to understanding the physical origins of the very distinctive ground-state properties of KI.
The 60 T long pulse (60 TLP) magnet, at the National High Magnetic Field Laboratory-Los Alamos Pulsed Field Laboratory, is driven by a 1.4 GW synchronous power generator and produces a flat top field for a period of 100+ ms at 60 T and for longer time at lower fields. We have built a calorimeter out of plastic materials that enables us to perform heat capacity measurements at temperatures between 1.4 and 20 K in this magnet. We have also used the relaxation technique to measure the heat capacity of our sample at higher temperatures in a 20 T superconducting magnet, in a standard calorimeter.
Section snippets
Low magnetic fields
The presence of the spin gap in Ce3Bi4Pt3 should be evident in the temperature dependence of the specific heat at temperatures comparable to the gap as the conduction band is depopulated, if the phonon contribution is not too large and the sensitivity of the experiment is high. For magnetic fields available in DC superconducting (SC) magnets, we expect the value of the gap to remain high, and therefore, the experimental temperature range must be extended to values comparable with the zero-field
High magnetic fields
For high enough magnetic fields the spin gap should be totally suppressed and as the sample changes from insulator to metal, the low-temperature specific heat should change accordingly. During the magnetic field pulse produced by the 60 TLP magnet, which lasts for about 2 s, our plastic calorimeter can be regarded as thermally isolated, i.e. in an adiabatic condition. We used a heat pulse method, where a known amount of heat is delivered to the calorimeter using a chip resistor, to measure the
Discussion
In order to tell, whether our results indicate the recovery of the metallic Kondo state in high fields, the increase in γH observed in Ce3Bi4Pt3 needs to be put in perspective. Magnetic susceptibility and the high temperature neutron quasielastic line width measurements can be used to estimate a zero-field Kondo temperature of TK0=240–320 K. In turn, the Sommerfeld coefficient for a metal with such TK can be estimated using the expression for a single-impurity Kondo system [9]
Conclusions
These results of thermodynamic measurements add new and essential information to what we know about Kondo insulators from transport measurements in high fields, and will aid in the construction of a theoretical model for this important class of strongly correlated compounds. Additionally, in the course of these studies we have demonstrated the feasibility of direct specific heat measurements in the extreme conditions of the pulsed magnetic fields produced by the 60 TLP magnet at the NHMFL-LANL
Acknowledgments
We thank Z. Fisk, J.D. Thompson and P. Schlottmann for discussions; J. Kim for his assistance with the thermometry calibration in the 30 T d.c. magnet at the NHMFL/Tallahassee; D. Rickel, C. Mielke, J. Betts, J. Schillig, M. Gordon, J. Sims and M. Pacheco for technical assistance and operation of the 60 TLP magnet.
References (12)
Physica B
(1993)Physica B
(1995)- et al.
Comment. Cond. Matt. Phys.
(1992) Phys. Rev. Lett.
(1999)Phys. Rev. Lett.
(1999)Euro. Phys. Lett.
(1998)
Cited by (7)
Chapter 11 Kondo insulators
2008, Handbook of Metal PhysicsPhysics in high magnetic fields
2004, Reports on Progress in PhysicsDevelopment of reliable 70 T pulsed magnets
2003, Measurement Science and TechnologyRipple Suppression of High Magnetic Field by Eddy Currents in Metal Tube
2023, 2023 26th International Conference on Electrical Machines and Systems, ICEMS 2023Flat-Top Pulsed High Magnetic Field Technology and Its Application
2022, Diangong Jishu Xuebao/Transactions of China Electrotechnical SocietyHeat-pulse measurements of specific heat in 36 ms pulsed magnetic fields
2013, Measurement Science and Technology