Hae-Young Kee: "Electronic Nematic Fluid and Metamagnetic transition in Sr3Ru2O7"

Hae-Young discussed a theory of electronic nematic order (also known as Pommeranchuk distortions), which, she argued rather convincingly, explains the metamagnetic transitions in the bilayer strontium ruthenate (the title compound).

Hae-Young began by pointing out that the monolayer ruthenate is isostructural to La2CuO4. She then moved on to a discussion of the experimental results of Andy Mackenzie’s group on this material. When a field is applied along the c-axis at low temperature the resistivity is more or less constant until a critical field, Hc1, where it rather rapidly doubles in size. As the field is increases above Hc1 the resistivity traces out a dome before return to its low initial value for H>Hc2. Peaks in the imaginary part of the susceptibility and jumps in the magnetisation are observed at both Hc1 and Hc2. From this data Hae-Young sketched a H-T phase diagram with first order transitions at Hc1 and Hc2 and a dome of second order transitions connecting them.

Throughout this discussion John Mydosh wanted to know why no specific heat experiments had been performed. John pointed out that this is important to rule out the possibility of a spin glass. Hae-Young stressed that this is an itinerant system and so one should not expect a spin glass. These two then debated the gradient of the first order line with Hae-Young saying that it was vertical and John arguing that it concave.

Hae-Young then wrote down a Hamiltonian with hopping terms, as well as Hubbard U and V and a correlated hopping term, tc, (which is where most of the physics appears to arise from). Andrey Chubukov asked about the relative magnitude of these terms and Hae-Young said that the size of U was not very important, but that tc>V. She then wrote down her order parameter, which describes a d-wave Fermi surface distortion. Hae-Young spared us the details of her calculations and simply sketched the dependence of the order parameter on the chemical potential. This looked remarkably like the experimentally derived phase diagram she had sketched earlier, except for the fact that the x-axis was chemical potential rather than field. However, Hae-Young quickly pointed out that the Zeeman term of the field acts exactly like a spin dependant chemical potential and so the down spins may undergo a nematic transition if they hit a van Hove singularity while the up spins are simply spectators.

Hae-Young then argued that the formation of domains is responsible for the anomalies seen in the resistivity. (The two domains correspond to elongating the Fermi surface in either the x or y directions.) Then she moved on to discuss the experimental observation that tilting the field removes the resistance anomaly. She argued that this is because of the bilayer structure, which allows for circulating interlayer currents, which pick out one domain over the other. Claudio Castelani then asked whether one could use this hysteresis effect as a test of the theory. Hae-Young thought that one could but that the experiment had not been performed.

In question time John asked whether de Haas-van Alphen experiments had been performed as these could look directly for the Fermi surface distortion. Hae-Young said that they had and that although nice quantum oscillations could be seen below Hc1 and above Hc2 nothing could be seen in the intermediate region. She interpreted this as evidence for domains.

Andrey asked if there is direct evidence for the van Hove singularity that is required for her thesis. Hae-Young said that Takagi et al. had seen evidence of this in STM experiments.

Your humble blogger asked what was known about the effect of disorder on Hae-Young’s nematic phase, as the experimental anomalies appear to be strongly suppressed by disorder. Hae-Young replied that although there are not any definitive calculations, arguments have been supplied by Kivelson, Fradkin and others that suggest that the nematic phase is suppressed by disorder.