Artificial spin ice is a class of lithographically created arrays of interacting ferromagnetic nanometre-scale islands. It was introduced to investigate many-body phenomena related to frustration and disorder in a material that could be tailored to precise specifications and imaged directly. Because of the large magnetic energy scales of these nanoscale islands, it has so far been impossible to thermally anneal artificial spin ice into desired thermodynamic ensembles; nearly all studies of artificial spin ice have either treated it as a granular material activated by alternating fields or focused on the as-grown state of the arrays. This limitation has prevented experimental investigation of novel phases that can emerge from the nominal ground states of frustrated lattices. For example, artificial kagome spin ice, in which the islands are arranged on the edges of a hexagonal net, is predicted to support states with monopolar charge order at entropies below that of the previously observed pseudo-ice manifold. Here we demonstrate a method for thermalizing artificial spin ices with square and kagome lattices by heating above the Curie temperature of the constituent material. In this manner, artificial square spin ice achieves unprecedented thermal ordering of the moments. In artificial kagome spin ice, we observe incipient crystallization of the magnetic charges embedded in pseudo-ice, with crystallites of magnetic charges whose size can be controlled by tuning the lattice constant. We find excellent agreement between experimental data and Monte Carlo simulations of emergent charge-charge interactions.
Bibliographical noteFunding Information:
Acknowledgements This project was fundedby the USDepartment of Energy, Office of Basic Energy Sciences, Materials Sciences and Engineering Division under grant no. DE-SC0005313. Lithography was performed with the support of the National Nanotechnology Infrastructure Network. The work of C.N. and G.-W.C. was carried out under the auspices of the US Department of Energy at LANL under contract no. DE-AC52-06NA253962. Work at the University of Minnesota was supported by the NSF MRSEC under award DMR-0819885 and EU Marie Curie IOF project no. 299376. Certain theory elements were supported by the NSF MRSEC under award DMR-0820404.