SearcharxivSearch

arXiv subjects

Aaron A. Pesetski

Publications and source records attributed to Aaron A. Pesetski.

3 recordsLinked to original sources

Demonstration of a scalable all-solid-state refrigerator exploiting diffusion geometries and limiting interfacial conductances at temperatures below 1 kelvin

Solid-state refrigerators using Normal-metal/Insulator/Superconductor (NIS) junctions have previously demonstrated excellent electron cooling but limited ability to cool phonons. The energy gap of the superconductor is used as an energy filter to allow higher than average energy electrons to preferentially tunnel from the normal-metal through the insulator into the superconductor where they travel as quasi-particles. Typically, the heat is moved and work is done to deposit hot quasi-particles into a normal-metal quasi-particle trap for rejection to the next refrigeration stage. Realizing that (1) the quasi-particles flow diffusively, driven by a concentration gradient in the electric field-free superconductor, and (2) that the undesirable backwards leaking of heat from the hot-side trap can be reduced by engineering the geometry and materials at the superconductor-to-trap interface, enhanced cooling can be achieved. Fabrication of the refrigerator was accomplished using a tungsten and titanium-tungsten alloy as the cold-side normal-metal, aluminum oxide as the insulator, aluminum as the superconductor, and gold as the trap, with the cold-side NIS portion being attached to the hot-side gold trap by bump bonding. The refrigerator consisted of 1121 junction pairs, each pair being an SINIS unit, all electrically connected in series. Using this we have measured the effective phonon temperature of a 3.9 mm x 3.9 mm x 0.65 mm silicon chip driven down to 70 mK from a bath temperature of 120 mK, and down to 174 mK from a 271 mK rejection temperature (a cooling of -97 mK). This is the first demonstration of the sub 1 K cooling of an entire silicon chip using NIS junctions.

cond-mat.supr-con

Experimental Study of the Inductance of Pinned Vortices in Superconducting YBa2Cu3O7-d Films

Using a two-coil mutual inductance method, we have measured the complex resistivity, rho_v(T,Be), of pinned vortices in c-axis pulsed laser deposited YBa2Cu3O7-d films with magnetic field Be applied perpendicular to the film. At low frequencies, (<100 kHz), rho_v is inductive and is inversely proportional to the Labusch parameter, the average vortex pinning force constant, kappa_exp. The observed weakening of kappa_exp with Be is consistent with a simple model based on linear pinning defects. Adding classical thermal fluctuations to the model in a simple way describes the observed linear T dependence of rho_v, below ~15 K and provides reasonable values for the effective radius (.3 nm to >.8 nm) of the defects and the depth of the pinning potential. The success of this model implies that thermal supercurrent (phase) fluctuations have their full classical amplitude down to 5 K for frequencies below the characteristic depinning frequency. To date, no sufficient theory exists to explain the data between ~15 K and the vortex glass melting temperature.

cond-mat.supr-con

Effect of Thermal Phase Fluctuations on the Inductances of Josephson Junctions, Arrays of Junctions, and Superconducting Films

We calculate the factor by which thermal phase fluctuations, as distinct from phase-slip fluctuations, increase the inductance, LJ, of a resistively-shunted Josephson junction (JJ) above its mean-field value, L0. We find that quantum mechanics suppresses fluctuations when T drops below a temperature, TQ = h/kBGL0, where G is the shunt conductance. Examination of the calculated sheet inductance, LA(T)/L0(T), of arrays of JJ's reveals that 2-D interconnections halve fluctuation effects, while reducing phase-slip effects by a much larger factor. Guided by these results, we calculate the sheet inductance, LF(T)/L0(T), of 2-D films by treating each plasma oscillation mode as an overdamped JJ. In disordered s-wave superconductors, quantum suppression is important for LF(0)/LF(T) > 0.14, (or, T/TC0 < 0.94). In optimally doped YBCO and BSCCO quantum suppression is important for l2(0)/l2(T) > 0.25, where l is the penetration depth.

cond-mat.supr-con