-
Evaluating the diversity and utility of materials proposed by generative models
Authors:
Alexander New,
Michael Pekala,
Elizabeth A. Pogue,
Nam Q. Le,
Janna Domenico,
Christine D. Piatko,
Christopher D. Stiles
Abstract:
Generative machine learning models can use data generated by scientific modeling to create large quantities of novel material structures. Here, we assess how one state-of-the-art generative model, the physics-guided crystal generation model (PGCGM), can be used as part of the inverse design process. We show that the default PGCGM's input space is not smooth with respect to parameter variation, mak…
▽ More
Generative machine learning models can use data generated by scientific modeling to create large quantities of novel material structures. Here, we assess how one state-of-the-art generative model, the physics-guided crystal generation model (PGCGM), can be used as part of the inverse design process. We show that the default PGCGM's input space is not smooth with respect to parameter variation, making material optimization difficult and limited. We also demonstrate that most generated structures are predicted to be thermodynamically unstable by a separate property-prediction model, partially due to out-of-domain data challenges. Our findings suggest how generative models might be improved to enable better inverse design.
△ Less
Submitted 9 August, 2023;
originally announced September 2023.
-
Closed-loop machine learning for discovery of novel superconductors
Authors:
Elizabeth A. Pogue,
Alexander New,
Kyle McElroy,
Nam Q. Le,
Michael J. Pekala,
Ian McCue,
Eddie Gienger,
Janna Domenico,
Elizabeth Hedrick,
Tyrel M. McQueen,
Brandon Wilfong,
Christine D. Piatko,
Christopher R. Ratto,
Andrew Lennon,
Christine Chung,
Timothy Montalbano,
Gregory Bassen,
Christopher D. Stiles
Abstract:
The discovery of novel materials drives industrial innovation, although the pace of discovery tends to be slow due to the infrequency of "Eureka!" moments. These moments are typically tangential to the original target of the experimental work: "accidental discoveries". Here we demonstrate the acceleration of intentional materials discovery - targeting material properties of interest while generali…
▽ More
The discovery of novel materials drives industrial innovation, although the pace of discovery tends to be slow due to the infrequency of "Eureka!" moments. These moments are typically tangential to the original target of the experimental work: "accidental discoveries". Here we demonstrate the acceleration of intentional materials discovery - targeting material properties of interest while generalizing the search to a large materials space with machine learning (ML) methods. We demonstrate a closed-loop ML discovery process targeting novel superconducting materials, which have industrial applications ranging from quantum computing to sensors to power delivery. By closing the loop, i.e. by experimentally testing the results of the ML-generated superconductivity predictions and feeding data back into the ML model to refine, we demonstrate that success rates for superconductor discovery can be more than doubled. In four closed-loop cycles, we discovered a new superconductor in the Zr-In-Ni system, re-discovered five superconductors unknown in the training datasets, and identified two additional phase diagrams of interest for new superconducting materials. Our work demonstrates the critical role experimental feedback provides in ML-driven discovery, and provides definite evidence that such technologies can accelerate discovery even in the absence of knowledge of the underlying physics.
△ Less
Submitted 22 December, 2022;
originally announced December 2022.
-
Minimum and maximum conductance of a thin film layer bridged interface: the role of anharmonicity and layer thickness
Authors:
**gjie Zhang,
Rouzbeh Rastgarkafshgarkolaei,
Carlos A. Polanco,
Nam Q. Le,
Keivan Esfarjani,
Pamela M. Norris,
Avik W. Ghosh
Abstract:
We study the role of anharmonicity at interfaces with an added intermediate layer designed to facilitate interfacial phonon transport. Our results demonstrate that while in the harmonic limit the bridge may lower the conductance due to fewer available channels, anharmonicity can strongly enhance the thermal conductance of the bridged structure due to added inelastic channels. Moreover, we show tha…
▽ More
We study the role of anharmonicity at interfaces with an added intermediate layer designed to facilitate interfacial phonon transport. Our results demonstrate that while in the harmonic limit the bridge may lower the conductance due to fewer available channels, anharmonicity can strongly enhance the thermal conductance of the bridged structure due to added inelastic channels. Moreover, we show that the effect of anharmonicity on the conductance can be tuned by varying temperature or the bridge layer thickness, as both parameters change the total rate of occurrence of phonon-phonon scattering processes. Additionally, we show that the additive rule of thermal resistances(Ohms law) is valid for bridge layer thickness quite shorter than the average bulk MFP, beyond the regime it would be expected to fail.
△ Less
Submitted 16 October, 2019;
originally announced October 2019.
-
Effects of bulk and interfacial anharmonicity on thermal conductance at solid/solid interfaces
Authors:
Nam Q. Le,
Carlos A. Polanco,
Rouzbeh Rastgarkafshgarkolaei,
**gjie Zhang,
Avik W. Ghosh,
Pamela M. Norris
Abstract:
We present the results of classical molecular dynamics simulations to assess the relative contributions to interfacial thermal conductance from inelastic phonon processes at the interface and in the adjacent bulk materials. The simulated system is the prototypical interface between argon and "heavy argon" crystals, which enables comparison with many past computational studies. We run simulations i…
▽ More
We present the results of classical molecular dynamics simulations to assess the relative contributions to interfacial thermal conductance from inelastic phonon processes at the interface and in the adjacent bulk materials. The simulated system is the prototypical interface between argon and "heavy argon" crystals, which enables comparison with many past computational studies. We run simulations interchanging the Lennard-Jones potential with its harmonic approximation to test the effect of anharmonicity on conductance. The results confirm that the presence of anharmonicity is correlated with increasing thermal conductance with temperature, which supports conclusions from prior experimental and theoretical work. However, in the model Ar/heavy-Ar system, anharmonic effects at the interface itself contribute a surprisingly small part of the total thermal conductance. The larger fraction of the thermal conductance at high temperatures arises from anharmonic effects away from the interface. These observations are supported by comparisons of the spectral energy density, which suggest that bulk anharmonic processes increase interfacial conductance by thermalizing energy from modes with low transmission to modes with high transmission.
△ Less
Submitted 4 October, 2016;
originally announced October 2016.
-
Role of crystal structure and junction morphology on interface thermal conductance
Authors:
Carlos A. Polanco,
Rouzbeh Rastgarkafshgarkolaei,
**gjie Zhang,
Nam Q. Le,
Pamela M. Norris,
Patrick E. Hopkins,
Avik W. Ghosh
Abstract:
We argue that the relative thermal conductance between interfaces with different morphologies is controlled by crystal structure through $M_{min}/M_c > 1$, the ratio between the {\it minimum mode} count on either side $M_{min}$, and the {\it conserving modes} $M_c$ that preserve phonon momentum transverse to the interface. Junctions with an added homogenous layer, "uniform", and "abrupt" junctions…
▽ More
We argue that the relative thermal conductance between interfaces with different morphologies is controlled by crystal structure through $M_{min}/M_c > 1$, the ratio between the {\it minimum mode} count on either side $M_{min}$, and the {\it conserving modes} $M_c$ that preserve phonon momentum transverse to the interface. Junctions with an added homogenous layer, "uniform", and "abrupt" junctions are limited to $M_c$ while junctions with interfacial disorder, "mixed", exploit the expansion of mode spectrum to $M_{min}$. In our studies with cubic crystals, the largest enhancement of conductance from "abrupt" to "mixed" interfaces seems to be correlated with the emergence of voids in the conserving modes, where $M_c = 0$. Such voids typically arise when the interlayer coupling is weakly dispersive, making the bands shift rigidly with momentum. Interfacial mixing also increases alloy scattering, which reduces conductance in opposition with the mode spectrum expansion. Thus the conductance across a "mixed' junction does not always increase relative to that at a "uniform" interface.
△ Less
Submitted 15 July, 2015;
originally announced July 2015.
