Thermodynamics Of Mixtures Research Articles

Differentiable modeling and optimization of non-aqueous Li-based battery electrolyte solutions using geometric deep learning

Electrolytes play a critical role in designing next-generation battery systems, by allowing efficient ion transfer, preventing charge transfer, and stabilizing electrode-electrolyte interfaces. In this work, we develop a differentiable geometric deep learning (GDL) model for chemical mixtures, DiffMix, which is applied in guiding robotic experimentation and optimization towards fast-charging battery electrolytes. In particular, we extend mixture thermodynamic and transport laws by creating GDL-learnable physical coefficients. We evaluate our model with mixture thermodynamics and ion transport properties, where we show improved prediction accuracy and model robustness of DiffMix than its purely data-driven variants. Furthermore, with a robotic experimentation setup, Clio, we improve ionic conductivity of electrolytes by over 18.8% within 10 experimental steps, via differentiable optimization built on DiffMix gradients. By combining GDL, mixture physics laws, and robotic experimentation, DiffMix expands the predictive modeling methods for chemical mixtures and enables efficient optimization in large chemical spaces.

Kinetic model of multi-component polyatomic gas mixture considering internal energy excitation

In this paper, a kinetic Boltzmann model equation with internal degrees of freedom is established for thermodynamic non-equilibrium multi-component polyatomic gas mixture by using continuous energy levels to deal with rotational energy and vibrational energy. The normalization and conservativeness of the kinetic model are analyzed and proved. Then, numerical algorithm and wall boundary conditions for solving the model equation are given. Then, normal shock-wave structure flow problem and pressure/temperature gradient driven micro-channel flow problem for the two-component gas mixture are used to verify the reliability of the model. The simulation results are in good agreement with those obtained by other methods, which verifies the reliability of the model and algorithm in this paper. Finally, taking 25 N attitude control engine two-dimensional profile as the object, the study for two-dimensional profile nozzle internal and external mixed flow problem is carried out, and the influence of different inlet rarefaction levels on the molecular transports of mixed flow field and the molecular transport phenomena of mixed flow field with different number of components are discussed.

On thermodynamic processes, state equations and critical phenomena for homogeneous mixtures

In this paper, we study the thermodynamics of homogeneous mixtures in equilibrium. From the perspective of thermodynamics, substances are understood as Legendre submanifolds, which are equipped with a Riemannian structure in addition. We refer to these as Legendre-Riemannian manifolds. This Legendre structure reflects the law of conservation of energy, while the Riemannian structure corresponds to the second central moment of measurement of extensive quantities, indicating that we only consider stable states. Thermodynamic processes, such as chemical reactions, correspond to contact vector fields that preserve the law of energy conservation, or are contact. The presence of a Riemannian structure distinguishes between three classes of processes: positive, which increase the metric; neutral, which preserve the metric; and negative, which decrease the metric. We provide a detailed description of the processes and suggest a method for finding state equations for a homogeneous mixture in mechanical or chemical equilibrium.

Impacts of Short-Chain Alcohols on the Cloudy Development and Thermodynamics of Triton X-100 and Metformin Hydrochloride Drug Mixture

Impacts of Short-Chain Alcohols on the Cloudy Development and Thermodynamics of Triton X-100 and Metformin Hydrochloride Drug Mixture

On immiscibility preservation conditions of material interfaces in the generic five-equation model

Interfaces separating pure materials and mixtures tend to be severely smeared with interface-capturing methods for compressible multi-material flows, necessitating the requirement of various interface-sharpening techniques. However, these techniques have various problems related to consistency, conservation, and thermodynamic compatibility. In this work, we derive a general theoretical formulation of interface-sharpening techniques for the generic five-equation model. This theoretical formulation is not only conservative in mass, momentum, and total energy but also asymptotically compatible with the thermodynamic mixture laws of the mixture model upon which it is constructed, and is independent of various specific numerical algorithms. We further propose a general numerical method to solve this theoretical formulation. The proposed method is consistent and conservative, and it prevents spurious errors at the interfaces. Examples of one- and two-dimensional multimaterial compressible flow problems, including shocks and interfaces, are considered to verify the analysis and demonstrate the efficiency of the method.

Thermodynamics and energy properties of the lowest eutectic mixture of TNBA/MTNP

In light of the challenges posed by the low density, diminished energy levels, and shrinkage holes in the loading arise of the current cast-cured explosive liquid vehicle, TNT, the investigation of low-melting eutectic materials as potential substitutes for TNT has emerged as a focal point of research. By employing the solvent-nonsolvent method, the lowest eutectic mixture of TNBA(2,4,6-trinitro-3-bromoanisole)/MTNP(1-methyl-3,4,5-trinitropyrazole) was synthesized, distinguished by a moderate melting point and markedly elevated energy levels as compared to TNT. Comprehensive SEM, EDS, IR, XRD, and XPS analyses were conducted on the lowest eutectic mixture, substantiating the absence of any chemical interaction between TNBA and MTNP and affirming their exceptional compatibility. Moreover, an exploration into the thermal decomposition behavior and mechanism of the lowest eutectic mixture was undertaken, its mechanical sensitivity was tested, and its detonation performance was evaluated by EXPLO-5 software. The thermal decomposition parameters of the lowest eutectic mixture closely paralleled those of the precursor materials, with the thermal decomposition kinetics model function being identified as the Jander function. The impact sensitivity of the lowest eutectic was 48.1 cm and the friction sensitivity was 12%. Furthermore, the density of the lowest eutectic mixture surpassed 1.8 g/cm3, and its detonation velocity approached 8000 m/s, markedly surpassing the explosive performance of TNT.

A detailed kinetic submechanism for OH[formula omitted] chemiluminescence in hydrogen combustion revisited. Part 1

In this series of two papers, we present a novel physically sound kinetic submechanism aimed at modeling the formation and decay of chemiluminescent (electronically excited) OH∗ molecules in the ignition and combustion of hydrogen and hydrogen-based mixtures over a wide range of thermodynamic parameters and mixture compositions. The present part of this series describes the preparatory stages of the work, such as the aggregation of the data set for the OH∗ chemiluminescent emission (including the OH∗ emission profiles obtained in our shock-tube experiments), the choice of the basic kinetic model of hydrogen oxidation, and, importantly, quantum chemical calculations and theoretical estimates for the rate constant of the nonadiabatic preassociation reaction H + O + M ⇆ OH∗ + M, recognized as a major source of excited OH∗ molecules in hydrogen-air flames. These rate constant estimates are carried out in a wide range of temperatures (200–4000 K) and pressures (10−3−100 bar). It is shown that the rate constant of this key reaction exhibits distinct non-Arrhenius temperature behavior and, moreover, has a complex fall-off pressure dependence, with the transition from a third to a second-order, high-pressure regime occurring at pressures about 10 atm. The obtained dependence on temperature and pressure is in excellent agreement with the known experimental data and can be recommended (as part of the appropriate reaction mechanisms) for further kinetic modeling of OH∗ chemiluminescence accompanying the high-temperature oxidation of hydrogen and other fuels.Novelty and Significance StatementAlthough ultraviolet OH∗ chemiluminescence as an optical signature of the combustion process and the underlying chemistry is known to provide unique capabilities for combustion diagnostics, the development of the detailed reaction mechanisms intended for quantitative interpretation of the OH∗ emission measurements has now almost stagnated. Indeed, existing kinetic models contain a discouragingly small number of elementary processes and, apparently, do not account for the full variety of possible reaction pathways of OH∗ formation and depletion. Therefore, in this serial work, we have revised the current ideas about the kinetics of OH∗ production and consumption during ignition and combustion and focused first on the OH∗ kinetics in a reacting H2/O2/Ar/N2 mixture. The fundamentally significant and novel results of this Part of the present series are (i) the rate constant of the nonadiabatic preassociation reaction H + O + M ⇆ OH∗ + M (the primary source of OH∗ in hydrogen combustion), theoretically obtained for the first time as a function of temperature and pressure, and (ii) the OH∗ emission profiles obtained in our shock-tube experiments. The other aspects described here, such as the aggregation of the literature-based data set for the post-shock OH∗ chemiluminescent emission and the choice of the basic kinetic model of hydrogen oxidation, are also of interest and relevance.

Comparison of the symmetric hyperbolic thermodynamically compatible framework with Hamiltonian mechanics of binary mixtures

How to properly describe continuum thermodynamics of binary mixtures where each constituent has its own momentum? The Symmetric Hyperbolic Thermodynamically Consistent (SHTC) framework and Hamiltonian mechanics in the form of the General Equation for Non-Equilibrium Reversible-Irreversible Coupling (GENERIC) provide two answers, which are similar but not identical, and are compared in this article. They are compared both analytically and numerically on several levels of description, varying in the amount of detail. Namely, a reduction to a more common one-momentum setting is shown, where the effects of the second momentum translate into diffusive fluxes. Both SHTC and GENERIC can thus be interpreted as a method specifying diffusive flux in standard theory. The GENERIC equations, stemming from the Liouville equation, contain terms expressing self-advection of the relative velocity by itself, which lead to a vorticity-dependent diffusion matrix after the reduction. The SHTC equations, on the other hand, do not contain such terms. We also discuss the possibility to formulate a theory of mixtures with two momenta and only one temperature that is compatible with the Liouville equation and possesses the Hamiltonian structure, including Jacobi identity.

The Effect of CuO on the Thermal Behavior and Combustion Features of Pyrotechnic Compositions with AN/MgAl

Ammonium nitrate (AN) is of considerable interest to researchers in developing new types of energetic mixtures due to the release of environmentally benign gaseous products during burning and thermal decomposition. However, poor ignition and a low burning rate require special additives to speed up this process. The advantage of this research is the use of high-energy aluminum-based alloys as fuel to compensate for the disadvantages of AN. In addition, the effect of copper oxide (CuO) on the burning kinetics and thermodynamics of the energetic mixture based on ammonium nitrate–magnesium–aluminum alloys (AN/MgAl) is investigated. Alloys based on aluminum were created through a process of high-temperature diffusion welding, conducted in an environment of argon gas. The structure and thermal characteristics of alloys are determined by X-ray diffraction, scanning electron microscopy, and DTA-TG analyses. It has been found that CuO has significant effects on the thermal decomposition of an AN/MgAl-based energetic mixture by shifting the decomposition temperature from 269.33 °C to 261.34 °C and decreasing the activation energy from 91.41 kJ mol−1 to 89.26 kJ mol−1. Adding CuO reduced the pressure deflagration limit from 2 MPa to 1 MPa, and the linear burning rate of the AN/MgAl energetic mixture increased approximately twice (rb = 6.17 mm/s vs. rb = 15.44 mm/s, at a chamber pressure of P0 = 5 MPa).

Expanding coarse-grained model for lipids to investigate the thermodynamics of mixtures

Expanding coarse-grained model for lipids to investigate the thermodynamics of mixtures

HANNA: hard-constraint neural network for consistent activity coefficient prediction.

We present the first hard-constraint neural network model for predicting activity coefficients (HANNA), a thermodynamic mixture property that is the basis for many applications in science and engineering. Unlike traditional neural networks, which ignore physical laws and result in inconsistent predictions, our model is designed to strictly adhere to all thermodynamic consistency criteria. By leveraging deep-set neural networks, HANNA maintains symmetry under the permutation of the components. Furthermore, by hard-coding physical constraints in the model architecture, we ensure consistency with the Gibbs-Duhem equation and in modeling the pure components. The model was trained and evaluated on 317 421 data points for activity coefficients in binary mixtures from the Dortmund Data Bank, achieving significantly higher prediction accuracies than the current state-of-the-art model UNIFAC. Moreover, HANNA only requires the SMILES of the components as input, making it applicable to any binary mixture of interest. HANNA is fully open-source and available for free use.

Application of ERAS-model and PFP theory on excess and deviation properties for binary systems of ionic liquid + amine derivatives

Thermodynamics of mixtures includes ionic liquid 2-hydroxyethylammonium propionate [2-HEAPr] + 1-propanol, 3-amino-1-propanol [AP], 1-amino-2-propanol [MIPA], and monoethanolamine [MEA] were studied. Density ρ, speed of sound u, and refractive index nD of pure and mixtures were measured at T = (298.15, 308.15, 318.15) K. Excess molar volume VmE, excess partial molar volume Vm,iE, excess thermal expansion coefficient αE, isothermal coefficient of excess molar enthalpy (∂Hm∂P)T,x, deviations in isentropic compressibility Δ ks, and refractive index Δ nD were calculated and data correlated with Redlich-Kister polynomial relation. The Prigogine-Flory-Patterson (PFP) and Extended Real Associated Solution (ERAS) models were applied to analyze intramolecular and intermolecular interactions between like and unlike molecules. The accuracy of theoretical models were found to be in agreement with empirical results. Finally, the study confirmed the structures of the ILs using FT-IR and HNMR analysis.

Parameters of State in the Global Thermodynamics of Binary Ideal Gas Mixtures in a Stationary Heat Flow.

In this paper, we formulate the first law of global thermodynamics for stationary states of the binary ideal gas mixture subjected to heat flow. We map the non-uniform system onto the uniform one and show that the internal energy U(S*,V,N1,N2,f1*,f2*) is the function of the following parameters of state: a non-equilibrium entropy S*, volume V, number of particles of the first component, N1, number of particles of the second component N2 and the renormalized degrees of freedom. The parameters f1*,f2*, N1,N2 satisfy the relation (N1/(N1+N2))f1*/f1+(N2/(N1+N2))f2*/f2=1 (f1 and f2 are the degrees of freedom for each component respectively). Thus, only 5 parameters of state describe the non-equilibrium state of the binary mixture in the heat flow. We calculate the non-equilibrium entropy S* and new thermodynamic parameters of state f1*,f2* explicitly. The latter are responsible for heat generation due to the concentration gradients. The theory reduces to equilibrium thermodynamics, when the heat flux goes to zero. As in equilibrium thermodynamics, the steady-state fundamental equation also leads to the thermodynamic Maxwell relations for measurable steady-state properties.

Phase-Field DeepONet: Physics-informed deep operator neural network for fast simulations of pattern formation governed by gradient flows of free-energy functionals

Recent advances in scientific machine learning have shed light on the modeling of pattern-forming systems. However, simulations of real patterns still incur significant computational costs, which could be alleviated by leveraging large image datasets. Physics-informed machine learning and operator learning are two new emerging and promising concepts for this application. Here, we propose “Phase-Field DeepONet”, a physics-informed operator neural network framework that predicts the dynamic responses of systems governed by gradient flows of free-energy functionals. Examples used to validate the feasibility and accuracy of the method include the Allen–Cahn and Cahn–Hilliard equations, as special cases of reactive phase-field models for nonequilibrium thermodynamics of chemical mixtures. This is achieved by incorporating the minimizing movement scheme into the framework, which optimizes and controls how the total free energy of a system evolves, instead of solving the governing equations directly. The trained operator neural networks can work as explicit time-steppers that take the current state as the input and output the next state. This could potentially facilitate fast real-time predictions of pattern-forming dynamical systems, such as phase-separating Li-ion batteries, emulsions, colloidal displays, or biological patterns.

Jacobson’s intermolecular free lengths in liquid mixtures – A thermodynamic approach

Jacobson’s intermolecular free length is a quantity related to molar volume and the internal surface of a liquid. A difference between intermolecular free length in a mixture and that calculated from the free lengths in its pure components results from molecular interactions. Commonly, the mole-fraction-weighted average of the intermolecular free lengths is a reference function in such calculations. The reference state lacking physical meaning leads to questionable conclusions. A new, correct formula for the intermolecular free length in thermodynamic ideal mixture was derived and discussed in this paper. The free lengths of the components in a thermodynamic ideal mixture are additive in a new “internal surface fraction” concentration scale. Examples illustrated significant differences between the deviations calculated from the two formulae for model systems and real binary mixtures. Since the values of the intermolecular free length deviations and the excess volumes have the same sign, the conclusions about intermolecular interactions based on these functions will be the same.

Thermodynamics of liquid and fluid mixtures from the kinetic Monte Carlo viewpoint.

In this study a binary mixture is modelled in a uniform simulation cell at various temperatures using an extended version of the grand canonical kinetic Monte Carlo (GC-kMC) method. The main goal of this study is to consider the thermodynamic properties of binary liquids, gases, and gas-liquid mixtures from a more general point of view than that applied to the particular case of vapour-liquid equilibrium when the pressure and partial chemical potentials in coexisting phases are the same. Particular attention is paid to thermodynamic functions such as chemical potentials, Gibbs free energy and entropy. For the pair potential of unlike molecules a more universal scheme is proposed in comparison with the Lorentz-Berthelot combining rule. The approach is tested on an Ar-Kr mixture in a wide range of temperatures. In all cases, the obtained values of chemical potentials, pressure and internal energy for the entire set of component densities and temperature fully satisfy the Gibbs-Duhem equation with a high degree of accuracy. For the case of vapour-liquid equilibrium, the developed approach made it possible to reproduce the experimental pressure-composition diagrams with the highest accuracy ever achieved in the literature. Despite the fact that the Ar-Kr mixture, according to Raoult's law, is close to an ideal system, it was found that the partial pressures in the liquid phase or in a dense supercritical gas mixture are non-linear functions of the composition, and the partial pressure of the heavier component (Kr) can even be negative.

Modeling asphaltene deposition in vertical oil wells

Asphaltene deposition in oil wells is a challenging flow assurance phenomenon that affects well production, project economics, and operational safety. While asphaltene precipitation is governed by hydrocarbon mixture thermodynamics, asphaltene deposition is governed by complex hydrodynamic behavior and characteristics. This study aims to evaluate and compare the performance of existing asphaltene deposition models based on transport mechanisms, and to improve the current theoretical understanding and prediction of asphaltene deposition phenomenon in single-phase liquid upward vertical flow. A large experimental database is collected from open literature, which falls into two categories, namely aerosol (air/metal particles), and crude oil system (oil/asphaltene particles). The aerosol transport coefficient database is divided into diffusion, diffusion-inertia, and impaction flow mechanisms, which are used to evaluate the transport coefficient models in each region. A statistical error analysis revealed that Beal (1970), Kor and Kharrat (2016), and Friedlander and Johnstone (1957) are the most accurate models in predicting transport coefficient in diffusion, diffusion-inertia, and impaction regions, respectively. Furthermore, a simplified asphaltene sticking probability model is proposed in this study, which is curve fitted using crude oil system asphaltene deposition flux data. The proposed sticking probability model is not only physically sound, but also requires fewer input data than existing models. A validation study of the proposed model slightly over-predicted the experimental data with an absolute average error of 9.8% and standard deviation of 21.8% outperforming existing model. The significance of this work is to improve current theoretical understanding, and propose predictive asphaltene deposition model in pipes that can be incorporated in an integrated asphaltene deposition model to prevent, mitigate, and manage oil field asphaltene deposition.

Evaluation of possible reactivation of undetected faults during CO[formula omitted] injection

Geologic storage of carbon dioxide can efficiently contribute to reduce greenhouse gas emissions to the atmosphere. Two major hazards of this technology are leakage towards the ground surface and fluid-induced seismicity. While major faults may be detected and avoided during site characterization, undetected subseismic faults could be encountered once injection has started. This paper investigates leakage and reactivation of undetected faults through coupled thermal–hydraulic–mechanical modeling. The simulations are performed using a recently developed sequential simulator, TOUGH-Pylith, that allows to accurately account for the thermodynamics of brine-CO2 mixtures, and to model faults as surfaces of discontinuity using state-of-the-art fault friction laws. The simulator is benchmarked against well-known analytical solutions and subsequently applied to investigate two cases of CO2 injection close to undetected faults under normal and strike-slip faulting regimes. Although the scenarios are generic and represent unfavorable conditions, they suggest that leakage could occur and that undetected faults could trigger minor seismic events. Therefore, careful site characterization and continuous monitoring during operations should be always performed.

Ethanol in Aqueous Solution Studied by Microjet Photoelectron Spectroscopy and Theory.

By combining results and analysis from cylindrical microjet photoelectron spectroscopy (cMJ-PES) and theoretical simulations, we unravel the microscopic properties of ethanol-water solutions with respect to structure and intermolecular bonding patterns following the full concentration scale from 0 to 100% ethanol content. In particular, we highlight the salient differences between bulk and surface. Like for the pure water and alcohol constituents, alcohol-water mixtures have attracted much interest in applications of X-ray spectroscopies owing to their potential of combining electronic and geometric structure probing. The water mixtures of the two simplest alcohols, methanol and ethanol, have generated particular attention due to their delicate hydrogen bonding networks that underlie their structural and thermodynamic properties. Macroscopically ethanol-water seems to mix very well, however microscopically this is not true. The aberrant thermodynamics of water-alcohol mixtures have been suggested to be caused by energy differences of hydrogen bonding between water-water, alcohol-alcohol and alcohol-water molecules. These networks may perturb the local character of the interaction between X-rays and matter, calling for analysis that go beyond the normally applied local selection and building block rules and that can combine the effects of light-matter, intra- and intermolecular interactions. However, despite decades of ongoing research there are still controversies of the precise nature of hydrogen bonding networks that underlie the mixing of these simple molecules. Our combined analysis indicates that at low concentration ethanol molecules form a film at the surface since ethanol at the surface can expose its hydrophobic part to the vacuum retaining its two (or three) possible hydrogen bonds, while water at the surface cannot retain all its four possible hydrogen bonds. Thus, ethanol at the surface becomes energetically favorable. Ethanol molecules show a tilting angle variation of the C-C axis with respect to the surface normal as large as 60° at very low concentration. In bulk, around ca. ten %, the ethanol oxygen atoms tend to make a third acceptor hydrogen bond to water molecules. At ca. 20 %, there is a U-shaped change in the CH3 to CH2OH binding energy (BE) shift indicating the presence of ring-like agglomerates called clathrate structures. At the surface, between 5 and 25%, ethanol forms a closely packed layer with the smallest C-C tilting angle variation down to ∼20°. Above 25% and below the azeotrope at the surface, ethanol shows an increase in the tilting angle variation, while at very high ethanol concentrations water tends to move to the surface so giving a microscopic explanation of the azeotrope effect. This migration is connected to the presence of longer (shorter) ethanol chains in the bulk (surface). A brief comparison with discussions and predictions from other spectroscopic techniques is also given. We emphasize the execution of an integrated approach that combines molecular structural dynamics with quantum predictions of the core electronic chemical shift, so establishing a protocol with considerable interpretative as well as predictive power for cMJ-PES measurements. We believe that this protocol can valorize cMJ-PES for studies of properties of other alcohol mixtures as well as of binary solutions in general.

(Invited) Multifunctional Lithium Ion Battery Separators through Polymerization-Induced Phase Separation

In state-of-the-art lithium ion batteries, separators (microporous membranes) play a passive yet critical role – hosting liquid electrolyte and maintaining physical separation of the electrodes. However, as the demands on lithium ion batteries increase, with an emphasis on greater energy density, longevity (cycle/calendar life), and safety, engineering separators to take a on more active role in the cell (electro)chemistry is expected to be an important strategy. Myriad membrane materials and separator designs have been developed to impart additional functionality, for example, acid and/or transition metal scavenging, temperature responsiveness, enhanced thermal stability, increased ion dissociation, combustion suppression, and mechanical strength.In this talk, I will preset my group’s approach to additive manufacturing of next-generation lithium ion battery separators. Our approach is based on polymerization induced phase separation (PIPS), wherein polymerizable monomers (or prepolymer resins) are mixed with porogen. Through rapid, low-cost, readily scalable photopolymerization, the monomers are converted to a crosslinked polymer network, which results in the porogen becoming immiscible and phase separating through spinodal decomposition. By tuning the thermodynamics of the polymer-porogen mixture and photopolymerization kinetics, the porosity and pore size of the resulting polymeric phase can be tuned. We have shown that ethylene carbonate (EC) mixed with common acrylate monomers, such as 1,4-butanediol diacrylate, is an effective porogen. Most importantly because EC is an indispensable component in liquid electrolytes, it does not need to be extracted from the separator prior to incorporation into the electrochemical cell.By controlling the ratio of the 1,4-butanediol diacrylate (BDDA) monomer to EC, monolithic microporous membranes are readily prepared with 25 µm thickness and pore sizes and porosities ranging from 6.8 to 22nm and 15.4% to 38.54%, respectively. The optimal poly(1,4-butanediol diacrylate) (pBDDA) separator has a porosity of 38.5% and average pore size of 22 nm; uptakes 127% liquid electrolyte by mass, and has an ionic conductivity of 1.98 mS/cm, which is higher than that of Celgard 2500. Lithium ion battery half cells consisting of LiNi0.5Mn0.3Co0.2O2 cathodes and pBDDA separators were shown to undergo reversible charge/discharge cycling with an average discharge capacity of 142 mAh/g and a capacity retention of 98.4% over 100 cycles - comparable to cells using state-of-the-art separators. Furthermore, the pBDDA separators were shown to be thermally stable to 400°C, lack low temperature thermal transitions that can compromise cell safety, and exhibits no thermal shrinkage up to 150°C. I will also discuss my group’s efforts to engineer separators with additional functionality to improve cell performance under abuse conditions.