 |
Center for Nonlinear Studies |
We show that artificial spin ice systems exhibit an imperfect return point memory when subject to drives below full saturation. We show that the loss of return point memory is correlated with the presence and movement of grain boundaries in the square ice. We argue that the hexagonal ice regains full return point memory at a faster rate than the square ice because the lattice is not bipartite and cannot support real grain boundaries. We also show that substrate disorder can immobilize the grain boundaries and help regaining return point memory in these systems.
Host: T-4