skip to main content

Title: Shear properties of MgO inferred using neural networks
Abstract. Shear properties of mantle minerals are vital for interpreting seismic shear wave speeds and therefore inferring the composition and dynamics of a planetary interior. Shear wave speed and elastic tensor components, from which the shear modulus can be computed, are usually measured in the laboratory mimicking the Earth's (or a planet's) internal pressure and temperature conditions. A functional form that relates the shear modulus to pressure (and temperature) is fitted to the measurements and used to interpolate within and extrapolate beyond the range covered by the data. Assuming a functional form provides prior information, and the constraints on the predicted shear modulus and its uncertainties might depend largely on the assumed prior rather than the data. In the present study, we propose a data-driven approach in which we train a neural network to learn the relationship between the pressure, temperature and shear modulus from the experimental data without prescribing a functional form a priori. We present an application to MgO, but the same approach works for any other mineral if there are sufficient data to train a neural network. At low pressures, the shear modulus of MgO is well-constrained by the data. However, our results show that different experimental results are inconsistent even at room temperature, seen as multiple peaks and diverging trends in probability density functions predicted by the network. Furthermore, although an explicit finite-strain equation mostly agrees with the likelihood predicted by the neural network, there are regions where it diverges from the range given by the networks. In those regions, it is the prior assumption of the form of the equation that provides constraints on the shear modulus regardless of how the Earth behaves (or data behave). In situations where realistic uncertainties are not reported, one can become overconfident when interpreting seismic models based on those defined equations of state. In contrast, the trained neural network provides a reasonable approximation to experimental data and quantifies the uncertainty from experimental errors, interpolation uncertainty, data sparsity and inconsistencies from different experiments.  more » « less
Award ID(s):
Author(s) / Creator(s):
; ; ; ;
Date Published:
Journal Name:
European Journal of Mineralogy
Page Range / eLocation ID:
45 to 58
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. In silico property prediction based on density functional theory (DFT) is increasingly performed for crystalline materials. Whether quantitative agreement with experiment can be achieved with current methods is often an unresolved question, and may require detailed examination of physical effects such as electron correlation, reciprocal space sampling, phonon anharmonicity, and nuclear quantum effects (NQE), among others. In this work, we attempt first-principles equation of state prediction for the crystalline materials ScF3 and CaZrF6, which are known to exhibit negative thermal expansion (NTE) over a broad temperature range. We develop neural network (NN) potentials for both ScF3 and CaZrF6 trained to extensive DFT data, and conduct direct molecular dynamics prediction of the equation(s) of state over a broad temperature/pressure range. The NN potentials serve as surrogates of the DFT Hamiltonian with enhanced computational efficiency allowing for simulations with larger supercells and inclusion of NQE utilizing path integral approaches. The conclusion of the study is mixed: while some equation of state behavior is predicted in semiquantitative agreement with experiment, the pressure-induced softening phenomenon observed for ScF3 is not captured in our simulations. We show that NQE have a moderate effect on NTE at low temperature but does not significantly contribute to equation of state predictions at increasing temperature. Overall, while the NN potentials are valuable for property prediction of these NTE (and related) materials, we infer that a higher level of electron correlation, beyond the generalized gradient approximation density functional employed here, is necessary for achieving quantitative agreement with experiment.

    more » « less
  2. Robots are active agents that operate in dynamic scenarios with noisy sensors. Predictions based on these noisy sensor measurements often lead to errors and can be unreliable. To this end, roboticists have used fusion methods using multiple observations. Lately, neural networks have dominated the accuracy charts for perception-driven predictions for robotic decision-making and often lack uncertainty metrics associated with the predictions. Here, we present a mathematical formulation to obtain the heteroscedastic aleatoric uncertainty of any arbitrary distribution without prior knowledge about the data. The approach has no prior assumptions about the prediction labels and is agnostic to network architecture. Furthermore, our class of networks, Ajna, adds minimal computation and requires only a small change to the loss function while training neural networks to obtain uncertainty of predictions, enabling real-time operation even on resource-constrained robots. In addition, we study the informational cues present in the uncertainties of predicted values and their utility in the unification of common robotics problems. In particular, we present an approach to dodge dynamic obstacles, navigate through a cluttered scene, fly through unknown gaps, and segment an object pile, without computing depth but rather using the uncertainties of optical flow obtained from a monocular camera with onboard sensing and computation. We successfully evaluate and demonstrate the proposed Ajna network on four aforementioned common robotics and computer vision tasks and show comparable results to methods directly using depth. Our work demonstrates a generalized deep uncertainty method and demonstrates its utilization in robotics applications.

    more » « less
  3. The compression behavior of the hexagonal AlB2 phase of Hafnium Diboride (HfB2) was studied in a diamond anvil cell to a pressure of 208 GPa by axial X-ray diffraction employing platinum as an internal pressure standard. The deformation behavior of HfB2 was studied by radial X-ray diffraction technique to 50 GPa, which allows for measurement of maximum differential stress or compressive yield strength at high pressures. The hydrostatic compression curve deduced from radial X-ray diffraction measurements yielded an ambient-pressure volume V0 = 29.73 Å3/atom and a bulk modulus K0 = 282 GPa. Density functional theory calculations showed ambient-pressure volume V0 = 29.84 Å3/atom and bulk modulus K0 = 262 GPa, which are in good agreement with the hydrostatic experimental values. The measured compressive yield strength approaches 3% of the shear modulus at a pressure of 50 GPa. The theoretical strain-stress calculation shows a maximum shear stress τmax~39 GPa along the (1−10) [110] direction of the hexagonal lattice of HfB2, which thereby can be an incompressible high strength material for extreme-environment applications. 
    more » « less
  4. Abstract

    Investigating water worlds presents a unique opportunity to understand the fundamental processes of planetary formation and evolution. One key aspect is characterizing the interactions between water and rock under the pressures and temperatures present within these worlds. Investigating the conditions for the homogeneous mixing of these materials is imperative to characterizing bulk properties and evolution of water‐rich exoplanets. Here we use density functional molecular dynamics simulations to study MgO‐H2O mixtures at high pressure–temperature conditions where H2O occurs in solid, superionic, or liquid form. MgO, the representative rocky material, can be either solid or liquid. We start from 500 K at 120 GPa, increasing the temperature step by step up to 8000 K. By inspection, we determine the temperature at which MgO‐HO homogeneously mix in our simulations. At 6000 K and 174 GPa is when we find the system to homogeneously mix. This heat‐until‐it‐mixes approach provides us with an upper bound on the temperature for the mixing of MgO and H2O. We find that homogeneous mixing occurs at sufficiently low temperatures to be relevant for the collisional growth of a water‐rich planet.

    more » « less
  5. Abstract

    The thermal equation of state (TEOS) for solids is a mathematic model among pressure, temperature and density, and is essential for geophysical, geochemical, and other high pressure–temperature (high P–T) researches. However, in the last few decades, there has been a growing concern about the accuracy of the pressure scales of the calibrants, and efforts have been made to improve it by either introducing a reference standard or building new thermal pressure models. The existing thermal equation of state,P(V,T) = P(V,T0) + Pth(V,T), consists of an isothermal compression and an isochoric heating, while the thermal pressure is the pressure change in the isochoric heating. In this paper, we demonstrate that, for solids in a soft pressure medium in a diamond anvil cell, the thermal pressure can neither be determined from a single heating process, nor from the thermal pressure of its calibrant. To avoid the thermal pressure, we propose to replace the thermal pressure with a well-known thermal expansion model, and integrate it with the isothermal compression model to yields a Birch–Murnaghan-expansion TEOS model, called VPT TEOS. The predicted pressure of MgO and Au at ambient pressure from Birch–Murnaghan-expansion VPT TEOS model matches the experimental pressure of zero (0) GPa very well, while the pressure prediction from the approximated Anderson PVT TEOS exhibit a big deviation and a wrong trend.

    more » « less