Following my recent post An end to a frustrating materials search, Radu Coldea brought to my attention a 2011 PRL which reports thermodynamic studies of the analogous Cu compound, Ba3CuSb2O9.
The authors suggest this material is also described by the antiferromagnetic Heisenberg model on the isotropic triangular lattice. They estimate the exchange interaction J ~ 32 K, but observe no magnetic ordering down to 0.2 K. This leads the authors to suggest the compound has a spin liquid ground state. It should be stressed that this absence of magnetic ordering is inconsistent with the material being described by a model with purely nearest neighbour coupling.
They observe a large Schottky anomaly in the specific heat and suggest that 5% of the Cu2+ ions are on Sb sites.
It is not clear to me
-what affect these large impurity contributions have to the subtractions which are performed to get the pure lattice magnetic contribution to the susceptibility and specific heat.
-whether the disorder present would be sufficient to prevent magnetic ordering.
It would be nice to see comparisons of the experimental data with calculations of the temperature dependence of the specific heat for the Heisenberg model on the triangular lattice [see this example this PRB by Bernu and Misguich]. They show the specific heat should have a maximum at a temperature of about 0.8J. Experimentally, it is at about 10 K, suggesting a much smaller value of J~12 K than that deduced from the experimental susceptibility.
Regardless of the above issues discovery of this new material is an exciting development.