(a) Schematic diagram of the Peltier element formed of - and -type Bi-Te alloys used to modulate the sample temperature . The sinusoidal voltage from the output of a lock-in amplifier (Ref. 17) is fed into the Peltier element. (b) Schematic diagram of the Peltier ac calorimeter (view from top). The voltages read on three type E Chromel-Constantan thermocouples measure the following temperature differences: between the sample and the top plate of the Peltier cell, between the sample and the heat sink (or main bath), and between the top plate of the Peltier cell and the main bath (see text). (c) Platform holding two independent calorimeters with their Peltier cells. The smaller one, (as used for the measurements presented in this article), uses thermocouples, whereas the larger one, , uses thermocouples.
Schematic diagram of the calorimeter, where the shaded area represents the top plate of the Peltier element which can be considered as a modulated thermal bath of temperature , is the sample temperature and the temperature of the main thermal bath. The heat conductance is determined by the legs of the thermocouple cross between the sample and the point where they are thermally anchored to the top plate of the Peltier element. is the heat conductance between the top and base plate of the Peltier element. The double-sided arrows show the temperature differences , and as measured by the three thermocouples (see Fig. 1).
(a) Frequency dependence of the amplitude of the thermocouples measuring , and for a (Bi-2212) sample at 90 K (measured without preamplifier or transformer). Taking into account phase shifts, the complex relation is obeyed. (b) Frequency dependence of the amplitudes and divided by the excitation . The lines are one-parameter fits using Eqs. (5) and (11). The crossing point of the two lines represents the cutoff frequency . The data at high frequencies of the thermocouple fall below the fit, possibly due to an asymmetry of heat flow through the and thermocouples. The data taken above 4 Hz are excluded from the analysis (see text for details). (c) Dependence of the amplitudes , and on the voltage (effective value) driving the temperature modulation of the top plate of the Peltier element. In this test conducted at and , a Bi-2212 sample is attached to the thermocouple cross.
Total heat capacity of a pure (99.999%) Ag sample including the addenda (top curve) compared to reference data for the sample alone (dotted line) (Ref. 19). The difference allows one to estimate the addenda heat capacity of the thermocouple cross and the glue [GE7031 (Ref. 13)] used to mount the sample (see text for details).
(a) Temperature dependence of the amplitude and of the thermocouple signals. The data are taken at , slightly below the cutoff frequency, to be analyzed in the modulated-bath scheme. (b) Temperature dependence of the amplitude and of the thermocouple signals. Here the data are taken at , about five times the cutoff frequency, to be analyzed within the ac method.
(a) Specific heat of a single crystal of Bi-2212 with a mass of in fields of 0 T, 1 T, 2 T, 8 T, and 14 T measured with the present modulated-bath calorimeter. The data are taken at and analyzed within the ac method. One gram-atom is 59.4 g . The anomaly at 90 K is the superconducting transition. In the insert a background given by the smoothed 14 T data has been subtracted. (b) Zero-field specific-heat data of the Bi-2212 sample compared to literature data (Ref. 21).
Residual noise in the specific-heat data after subtracting a smooth polynomial from the 14 T data in Fig. 6. The root-mean-square value of the scatter is or 0.023% of the sample specific heat.
Article metrics loading...
Full text loading...