Machine-learning-assisted insight into spin ice Dy2Ti2O7

Abstract

Complex behavior poses challenges in extracting models from experiment. An example is spin liquid formation in frustrated magnets like Dy2Ti2O7. Understanding has been hindered by issues including disorder, glass formation, and interpretation of scattering data. Here, we use a novel automated capability to extract model Hamiltonians from data, and to identify different magnetic regimes. This involves training an autoencoder to learn a compressed representation of three-dimensional diffuse scattering, over a wide range of spin Hamiltonians. The autoencoder finds optimal matches according to scattering and heat capacity data and provides confidence intervals. Validation tests indicate that our optimal Hamiltonian accurately predicts temperature and field dependence of both magnetic structure and magnetization, as well as glass formation and irreversibility in Dy2Ti2O7. The autoencoder can also categorize different magnetic behaviors and eliminate background

First Experiments: New Science Opportunities at the Spallation Neutron Source Second Target Station

“…Figure 4.3 shows a three-dimensional (3D) diffuse scattering data set collected on a spin liquid material using the CORELLI diffractometer at the FTS. Machine learning was used to train a neural network to interpret the data and identify the experimental features and phases over five independent dimensions of model interactions/parameters [Samarakoon et al, 2019]. This example demonstrates how new science that goes beyond current theory and analysis can be realized utilizing such new approaches.…”

Principal component analysis of diffuse magnetic neutron scattering: a theoretical study

“…Formally, PCA is equivalent to a linear autoencoder [24] and is a suitable initial step for a wide range of classification problems. In recent years PCA and auto-encoders have been applied to data obtained through numerical simulation of many-body systems [25,26]. It has been shown that, when provided with detailed information on large representative samples of microstates of such systems, such algorithms are capable of "discovering" important features in their phase diagrams, including order parameters and phase transitions [25].…”

“…Recently an autoencoder-based approach to magnetic diffuse neutron scattering data on the "spin-ice" material Dy 2 Ti 2 O 7 has been demonstrated [26]. The autoencoder is trained on a set of simulated neutron-scattering images.…”

Thermal expansion and phase transformation in the rare earth di-titanate (R2Ti2O7) system

“…The R 2 Ti 2 O 7 pyrochlore structure is well known from studies of magnetic monopoles in frustrated magnetic, spin ice states (Bramwell et al, 2009;Bramwell & Gingras, 2001;Sosin et al, 2005;Samarakoon et al, 2020). These structures also show high ionic conductivity and have been investigated for use as oxygen electrodes and electrolytes in solid oxide fuel cells (Moon & Tuller, 1988;Farmer et al, 2014;Subramanian et al, 1983).…”