Does classical spin ice Dy2Ti2O7 order or keep its Pauling entropy?
In plain words
Ice-rule states are so numerous that spin ice keeps a leftover entropy at low temperature, as water ice does. Long-range dipolar forces predict an ordered state below about 0.2 K, but experiments struggle to reach equilibrium there.
Precise statement
In Dy2Ti2O7, determine whether thermal equilibrium below 0.5 K approaches the ordered state predicted by the dipolar spin-ice model (transition near 0.18 K) with loss of the Pauling entropy $(R/2)\ln(3/2)$ per mole Dy, or retains a disordered ice manifold. An answer is the equilibrium entropy and any ordering wavevector as $T \to 0$.
What would settle it
Equilibrium heat capacity and neutron scattering below 0.3 K after controlled long annealing, compared with Monte Carlo of the fitted dipolar model.
Status in the literature
Pomaranski et al. (Nature Physics 2013) found the residual entropy is lost after long equilibration without detecting the predicted order.