Session J04: Condensed Matter Physics III

Understanding Geometrically Frustrated Magnets Without Getting Frustrated

In geometrically frustrated magnets, the positions of the magnetic moments within the crystal lattice prevent the simultaneous satisfaction of competing magnetic interactions. As a result, magnetic ordering is suppressed and a variety of exotic phenomena may ensue, such as macroscopic ground-state degeneracy, extreme sensitivity to perturbations, and the formation of a quantum spin liquid. Experimental studies of the magnetic correlations in frustrated magnets have historically been very challenging due to the suppression of long-range magnetic order, limiting our ability to understand the consequences of geometrical frustration using conventional techniques. Here, I will introduce atomic and magnetic pair distribution function (PDF) analysis, a method of analyzing neutron scattering data to study short-range atomic and magnetic correlations. I will then discuss the case of the frustrated triangular antiferromagnet NaMnO$_2$ as an example of how PDF methods provide unique insight into the unusual behavior of frustrated magnets. This example teaches us that geometrically frustrated magnets need not be so frustrating after all, as long as we use the right tools.   

Bayesian Approach to Uncertainty Quantification of Interatomic Models in OpenKIM Database

Interatomic models (IMs) are used in molecular modeling to predict material properties of interest. The development of a single IM can take anywhere from several months to years and relies on expert intuition, and yet these potentials are usually only valid for a particular application of interest. Extending existing IMs to new applications is an active area of research. Quantifying the uncertainty of an IM can tell us how much we can trust the predictions it makes. I take a Bayesian approach to uncertainty quantification. Using Monte Carlo methods, I sample from the posterior distribution of the parameters when trained on data. I demonstrate this method on Lennard-Jones and Morse potentials fit to triclinic crystal configurations from the OpenKIM database. These examples illustrate several subtleties related to the selection of Bayesian prior and choice of model parameterization. In particular, IMs are often sloppy, i.e., have likelihood surfaces with long, narrow canyons and broad, flat plateaus. Because of these features, the posterior can depend strongly on the prior and model parameterization. I discuss implications of sloppiness for uncertainty quantification in molecular modeling.   

Evolution of Radiative Heat Transfer Between Graphene Nanodisks Over Time

Radiative heat transfer is one mechanism by which objects thermalize with one another and their environment. This involves the emission of photons with wavelengths determined by the temperature of the object, and is, at the macroscale, accurately described by Planck's law. On the other hand, when the dimensions of a system are shrunk to below the thermal wavelength, the emergence of strong near-field modes, which can be further enhanced by optical resonances, can result in radiative transfer that is very different from the predictions of this law. This could enable the realization of a diverse range of technologies, such as improved thermophotovoltaics and more efficient thermal management in nanoscale electronics. Specifically, materials whose optical responses can be actively tuned, such as graphene nanostructures, are an ideal platform to achieve these goals. In this work, we study the time evolution of the temperature of graphene nanodisks in different arrangements, taking advantage of their active tunability to achieve dynamical control over the heat exchanged between them. This allows for the design of exotic heat transfer scenarios not possible with passive materials and can therefore help to inspire new technologies requiring active control over nanoscale energy transfer.   

Anharmonic Coupling in Beta Barium Borate Using Two-dimensional Spectroscopy.

High-field, ultrafast pulses of terahertz (THz) light allow the extreme excitation of atomic motion that allow us to probe regions of the underlying potential energy surface (PES). The PES, which dictates the material properties of a sample, has been studied with one-dimensional (1D, single pump pulse) measurements. At times, 1D measurements can be challenging to interpret. More information is available using 2D Terahertz spectroscopy to investigate the specific nonlinear couplings between excitations in solid materials. Here we use 2D Terahertz spectroscopy to gain insight into how phonon modes interact at large oscillation amplitudes to study $\beta $-barium borate (BBO), a commonly used nonlinear optical crystal. BBO has 16 phonon modes that are both Raman and IR active within our pump excitation bandwidth (1-6 THz), and the interaction between these modes leads to a feature rich 2D spectrum. We study the BBO sample responses, as a function of THz polarization with respect to the sample, to learn about the modes that nonlinearly interact to contribute to the 2D spectrum. Modeling the anharmonic coupling of the vibrational modes allows us to disentangle the 2D spectrum and gain insights into the complex underlying PES.~   

Semiconductor Dielectric Function Modeling Using the Tanguy Analytical Expression for the Hulthen Exciton

Tanguy\textsc{\char13}s analytical expression\footnote{C. Tanguy, \textbf{Phy. Rev. B} 60, 10660 (1999).} for the complex dielectric function associated with a Hulthen exciton in a semiconductor is used to obtain second derivatives that can be used for fast fits of spectroscopic ellipsometry data. Since the Hulthen potential provides an excellent description of a screened exciton, the expressions obtained are ideally suited to investigate the possible excitonic origin of the \textit{ad hoc} phase factors that are often needed to match dielectric function theory with ellipsometry data. A detailed analysis is presented with applications to elemental semiconductors such as Ge.   

Connecting Molecular Dynamics to Experiment using Machine Learning

Current resources allow for investigation of large parameter spaces of materials and material properties or processes using atomistic simulations much more quickly than by experiment. An additional advantage to simulation is the level of detail available in the results, namely atom positions, which can be difficult to recover from experimental results. We propose a process to connect experimental images, such as STEM, with similar atomistic simulation results; focusing specifically on defect crystal structures.