arXiv subjects

Alpha A Lee

Publications and source records attributed to Alpha A Lee.

6 recordsLinked to original sources

arXiv2024

Pushing the Pareto front of band gap and permittivity: ML-guided search for dielectric materials

Materials with high-dielectric constant easily polarize under external electric fields, allowing them to perform essential functions in many modern electronic devices. Their practical utility is determined by two conflicting properties: high dielectric constants tend to occur in materials with narrow band gaps, limiting the operating voltage before dielectric breakdown. We present a high-throughput workflow that combines element substitution, ML pre-screening, ab initio simulation and human expert intuition to efficiently explore the vast space of unknown materials for potential dielectrics, leading to the synthesis and characterization of two novel dielectric materials, CsTaTeO6 and Bi2Zr2O7. Our key idea is to deploy ML in a multi-objective optimization setting with concave Pareto front. While usually considered more challenging than single-objective optimization, we argue and show preliminary evidence that the $1/x$-correlation between band gap and permittivity in fact makes the task more amenable to ML methods by allowing separate models for band gap and permittivity to each operate in regions of good training support while still predicting materials of exceptional merit. To our knowledge, this is the first instance of successful ML-guided multi-objective materials optimization achieving experimental synthesis and characterization. CsTaTeO6 is a structure generated via element substitution not present in our reference data sources, thus exemplifying successful de-novo materials design. Meanwhile, we report the first high-purity synthesis and dielectric characterization of Bi2Zr2O7 with a band gap of 2.27 eV and a permittivity of 20.5, meeting all target metrics of our multi-objective search.

cond-mat.mtrl-sci↗

arXiv2019

Molecular Transformer - A Model for Uncertainty-Calibrated Chemical Reaction Prediction

Organic synthesis is one of the key stumbling blocks in medicinal chemistry. A necessary yet unsolved step in planning synthesis is solving the forward problem: given reactants and reagents, predict the products. Similar to other work, we treat reaction prediction as a machine translation problem between SMILES strings of reactants-reagents and the products. We show that a multi-head attention Molecular Transformer model outperforms all algorithms in the literature, achieving a top-1 accuracy above 90% on a common benchmark dataset. Our algorithm requires no handcrafted rules, and accurately predicts subtle chemical transformations. Crucially, our model can accurately estimate its own uncertainty, with an uncertainty score that is 89% accurate in terms of classifying whether a prediction is correct. Furthermore, we show that the model is able to handle inputs without reactant-reagent split and including stereochemistry, which makes our method universally applicable.

physics.chem-ph↗

arXiv2016

Capacitance-Power-Hysteresis Trilemma in Nanoporous Supercapacitors

Nanoporous supercapacitors are an important player in the field of energy storage that fill the gap between dielectric capacitors and batteries. The key challenge in the development of supercapacitors is the perceived trade-off between capacitance and power delivery. Current efforts to boost the capacitance of nanoporous supercapacitors focus on reducing the pore size so that they can only accommodate a single layer of ions. However, this tight packing compromises the charging dynamics and hence power density. We show via an analytical theory and Monte Carlo simulations that charging is sensitively dependent on the affinity of ions to the pores, and that high capacitances can be obtained for ionophobic pores of widths significantly larger than the ion diameter. Our theory also predicts that charging can be hysteretic with a significant energy loss per cycle for intermediate ionophilicities. We use these observations to explore the parameter regimes in which a capacitance-power-hysteresis trilemma may be avoided.

cond-mat.mtrl-sci↗

arXiv2015

Degenerate Mobilities in Phase Field Models are Insufficient to Capture Surface Diffusion

Phase field models frequently provide insight to phase transitions, and are robust numerical tools to solve free boundary problems corresponding to the motion of interfaces. A body of prior literature suggests that interface motion via surface diffusion is the long-time, sharp interface limit of microscopic phase field models such as the Cahn-Hilliard equation with a degenerate mobility function. Contrary to this conventional wisdom, we show that the long-time behaviour of degenerate Cahn-Hilliard equation with a polynomial free energy undergoes coarsening, reflecting the presence of bulk diffusion, rather than pure surface diffusion. This reveals an important limitation of phase field models that are frequently used to model surface diffusion.

cond-mat.soft↗

arXiv2014

Are Room Temperature Ionic Liquids Dilute Electrolytes?

An important question in understanding the structure of ionic liquids is whether ions are truly "free" and mobile which would correspond to a concentrated ionic melt, or are rather "bound" in ion pairs, that is a liquid of ion pairs with a small concentration of free ions. Recent surface force balance experiments from different groups have given conflicting answers to this question. We propose a simple model for the thermodynamics and kinetics of ion pairing in ionic liquids. Our model takes into account screened ion-ion, dipole-dipole and dipole-ion interactions in the mean field limit. The results of this model suggest that almost two thirds of the ions are free at any instant, and ion pairs have a short lifetime comparable to the characteristic timescale for diffusion. These results suggest that there is no particular thermodynamic or kinetic preference for ions residing in pairs. We therefore conclude that ionic liquids are concentrated, rather than dilute, electrolytes.

cond-mat.soft↗

arXiv2014

Unravelling Nanoconfined Films of Ionic Liquids

The confinement of an ionic liquid between charged solid surfaces is treated using an exactly solvable 1D Coulomb gas model. The theory highlights the importance of two dimensionless parameters: the fugacity of the ionic liquid, and the electrostatic interaction energy of ions at closest approach relative to thermal energy, in determining how the disjoining pressure exerted on the walls depends on the geometrical confinement. Our theory reveals that thermodynamic fluctuations play a vital role in the "squeezing out" of charged layers as the confinement is increased. The model shows good qualitative agreement with previous experimental data, with all parameters independently estimated without fitting.

cond-mat.soft↗