Thursday, July 5, 2012

1112.3698 (Allan Bayntun et al.)

Finite Size Scaling in Quantum Hallography    [PDF]

Allan Bayntun, C. P. Burgess
At low temperatures observations of the Hall resistance for Quantum Hall systems at the interface between two Hall plateaux reveal a power-law behaviour, dR_xy/dB ~ T^(-p) (with p = 0.42 +/- 0.01); changing at still smaller temperatures, T < T_s, to a temperature-independent value. Experiments also show that the transition temperature varies with sample size, L, according to T_s ~ 1/L. These experiments pose a potential challenge to the holographic AdS/QHE model recently proposed in arXiv:1008.1917. This proposal, which was motivated by the natural way AdS/CFT methods capture the emergent duality symmetries exhibited by quantum Hall systems, successfully describes the scaling exponent p by relating it to an infrared dynamical exponent z with p = 2/z. For a broad class of models z is robustly shown to be z = 5 in the regime relevant to the experiments (though becoming z = 1 further in the ultraviolet). By incorporating finite-size effects into these models we show that they reproduce a transition to a temperature-independent regime, predicting a transition temperature satisfying T_s ~ 1/L or ~ 1/L^5 in two separate regions of parameter space, even though z = 5 governs the temperature dependence of the conductivity in both cases. The possibility of a deviation from naive z = 5 scaling arises because the brane tension introduces a new scale, which alters where the transition between UV and IR scaling occurs, in an L-dependent way. The AdS/CFT calculation indicates the two regimes of temperature scaling are separated by a first-order transition, suggesting new possibilities for testing the picture experimentally. Remarkably, in this interpretation the gravity dual of the transition from temperature scaling to temperature-independent resistance is related to the Chandrashekar transition from a star to a black hole with increasing mass.
View original: http://arxiv.org/abs/1112.3698

No comments:

Post a Comment