| Dowman P. Varn Max-Planck-Institut fur Physik Komplexer Systeme Nothnitzer Strasse 38 01187 Dresden, Germany |
Geoffrey S. Canright Department of Physics and Astronomy University of Tennessee Knoxville, Tennessee 37996, USA and Telenor Research and Development 1331 Fornebu, Olso, Norway |
James P. Crutchfield Center for Computational Science and Engineering Physics DepartmenEngineering University of California, Davis One Shields Avenue Davis, CA 95616, USA |
ABSTRACT: Previously we detailed a novel algorithm, epsilon-machine spectral reconstruction theory (eMSR), that infers pattern and disorder in planar-faulted, close-packed structures directly from X-ray diffraction spectra [Varn, Canright and Crutchfield, to appear in Acta Crystallographica A. Here we apply eMSR to simulated diffraction spectra from five close-packed crystals. We find that for stacking structures with a memory length of three or less, eMSR reproduces the statistics of the stacking structure; the result being in the form of a directed graph called an epsilon-machine. For stacking structures with a memory length larger than three, eMSR returns a model that captures many important features of the original stacking structure. These include multiple stacking faults and multiple crystal structures. Further, we find that eMSR is able to discover stacking structurein even highly disordered crystals. In order to address issues concerning the long range order observed in many classes of layered materials, we define several length parameters calculable from the epsilon-machine, and discuss their relevance.