Search

Deep neural networks (NNs) provide flexible frameworks for learning data representations and functions relating data to other properties and are often claimed to achieve 'super-human' performance in inferring relationships between input data and desired property. In the context of inelastic neutron scattering experiments, however, as in many other scientific scenarios, a number of issues arise: (i) scarcity of labelled experimental data, (ii) lack of uncertainty quantification on results, and (iii) lack of interpretability of the deep NNs. In this work we examine approaches to all three issues. We use simulated data to train a deep NN to distinguish between two possible magnetic exchange models of a half-doped manganite. We apply the recently developed deterministic uncertainty quantification method to provide error estimates for the classification, demonstrating in the process how important realistic representations of instrument resolution in the training data are for reliable estimates on experimental data. Finally we use class activation maps to determine which regions of the spectra are most important for the final classification result reached by the network., (Creative Commons Attribution license.)

We have investigated the magnetic properties of a polycrystalline sample of NdOs 2 Al 10 using neutron diffraction and inelastic neutron scattering, magnetic susceptibility and heat capacity measurements. The magnetic susceptibility data reveal an antiferromagnetic transition at T N1 = 2.2 K while the heat capacity data which extend to lower temperatures show a further transition at T N2 = 1.1 K. The INS measurements detect four well-resolved crystal field excitations at 7.4, 12.4, 17.6 meV and 19.2 meV at 5 K. The positions and intensities of the observed CEF excitations and the temperature dependence of previously published single crystal susceptibility data are analysed based on the orthorhombic CEF model of the J = 9/2 ground state multiplet of Nd 3+ ion. The neutron diffraction study reveals the magnetic structure at T = 1.6 K to be of sine wave type ( κ = [0, 0.723, 0]) with an ordered state moment of μ Nd 3+ = 1.59(1) μ B aligned along the a -axis in agreement with the single ion crystal field anisotropy. This indicates that the magnetism of NdOs 2 Al 10 is governed by the RKKY interactions in agreement with other RT 2 Al 10 ( R = Nd, Sm, Tb, T = Fe, Ru and Os), but different to that of R = Ce based compounds where anisotropic two ions interaction as well as Kondo interaction govern the anomalous direction of the ordered state moment., (© 2021 IOP Publishing Ltd.)

Direct engineering of material properties through exploitation of spin, phonon, and charge-coupled degrees of freedom is an active area of development in materials science. However, the relative contribution of the competing orders to controlling the desired behavior is challenging to decipher. In particular, the independent role of phonons, magnons, and electrons, quasiparticle coupling, and relative contributions to the phase transition free energy largely remain unexplored, especially for magnetic phase transitions. Here, we study the lattice and magnetic dynamics of biferroic yttrium orthochromite using Raman, infrared, and inelastic neutron spectroscopy techniques, supporting our experimental results with first-principles lattice dynamics and spin-wave simulations across the antiferromagnetic transition at T N ∼ 138 K. Spectroscopy data and simulations together with the heat capacity ( C p ) measurements, allow us to quantify individual entropic contributions from phonons (0.01 ± 0.01 k B atom -1 ), dilational (0.03 ± 0.01 k B atom -1 ), and magnons (0.11 ± 0.01 k B atom -1 ) across T N . High-resolution phonon measurements conducted in a magnetic field show that anomalous T -dependence of phonon energies across T N originates from magnetoelastic coupling. Phonon scattering is primarily governed by the phonon-phonon coupling, with little contribution from magnon-phonon coupling, short-range spin correlations, or magnetostriction effects; a conclusion further supported by our thermal conductivity measurements conducted up to 14 T, and phenomenological modeling., (© 2021 IOP Publishing Ltd.)

Multiferroic materials have attracted much interest due to the unusual coexistence of ferroelectric and (anti-)ferromagnetic ground states in a single compound. They offer an exciting platform for new physics and potentially novel devices. BiFeO3 is one of the most celebrated multiferroic materials and has highly desirable properties. It is the only known room-temperature multiferroic with TC ≈ 1100 K and TN ≈ 650 K, and exhibits one of the largest spontaneous electric polarisations, P ≈ 80 µC cm(-2). At the same time, it has a magnetic cycloid structure with an extremely long period of 620 Å, which arises from competition between the usual symmetric exchange interaction and the antisymmetric Dzyaloshinskii-Moriya (DM) interaction. There is also an intriguing interplay between the DM interaction and single ion anisotropy K. In this review, we have attempted to paint a complete picture of bulk BiFeO3 by summarising the structural and dynamic properties of both the spin and lattice parts and their magneto-electric coupling.

The solid state exhibits a fascinating variety of phases, which can be stabilized by the variation of external parameters such as temperature, magnetic field and pressure. Until recently, numerical analysis of magnetic and/or orbital phases with collective excitations on a periodic lattice tended to be done on a case-by-case basis. Nowadays dynamical matrix diagonalization (DMD) has become an important and powerful standard method for the calculation of dispersive modes. The application of DMD to the interpretation of inelastic neutron scattering (INS) data on dispersive magnetic excitations is reviewed. A methodical survey of calculations employing spin-orbit and intermediate coupling schemes is illustrated by examples. These are taken from recent work on rare earth, actinide and transition metal compounds and demonstrate the application of the formalism developed.

We report inelastic neutron scattering measurements and random phase approximation calculations of the dispersive crystal field excitations of UPd(3). The measured spectra at lower energies agree with those calculated using quadrupolar interaction parameters deduced from bulk and x-ray scattering measurements. The more intense excitations arising from the hexagonal sites were used to obtain exchange parameters which proved to be anisotropic.

We present the results of x-ray and ultraviolet photoelectron spectroscopy of NpPd(3) and PuPd(3). The spectra indicate that for both compounds, the 5f electrons are well localized on the actinide sites. Comparison with bulk measurements indicates that for NpPd(3) the electrons have a valence of Np(3+) and thus a ground state 5f(4) with a Hund's rules (5)I(4) configuration. Similarly for PuPd(3), we find a Pu(3+) valence, 5f(5) ground state and a Hund's rules (6)H(5/2) configuration.