Over time, the chemicals that make up a battery’s electrolyte will degrade, especially when exposed to high temperatures. This is a major challenge in battery R&D, since electrolyte decomposition can significantly affect a battery’s cycle life and calendar life. In addition, some of the chemical compounds formed as a result of the electrolyte decomposition can be toxic, corrosive, or generate gases that can increase the risk of an explosion.  

In the past decade, the demand for batteries with higher energy storage capacity and faster charging rates has grown massively. Lithium-ion batteries are currently the gold standard — they're commonly found in electric vehicles, laptops, smartphones, and many other electronic devices.  

However, lithium-ion batteries face challenges with electrolyte degradation, often due to the lithium salts in the electrolyte, which are the most chemically and thermally unstable compounds in the battery. Common commercially available formulations of lithium-ion batteries can start decomposing at temperatures as low as 50 °C in some cases, leading to potential toxicity and fire hazards in addition to a rapid reduction of the battery’s performance.  

To develop safer and better performing batteries, studying and optimizing electrolyte degradation is a fundamental area of battery R&D.

‍

Use case: Simulating electrolyte decomposition in QuantistryLab

Studying the process of electrolyte decomposition can be quite challenging because of the large number of possible reactions happening among the multiple chemical compounds that make up the electrolyte, which are in turn also affected by external conditions such as the temperature. Identifying all the compounds that can potentially emerge from the degradation of a given electrolyte formulation requires numerous analytical techniques, making this essential step of the R&D process lengthy and costly.  

Chemical simulations are emerging as an alternative approach to study and predict the outcome of complex chemical reactions. In this use case, QuantistryLab’s nanoreactor feature was used to investigate the thermal reactions occurring during the decomposition process of a commercially available electrolyte formulation.

The quantum nanoreactor can simulate chemical reactions and identify the most likely products of the reaction. This virtual tool can complement experimental studies and accelerate the exploration of chemical compounds and their reactions.  

QuantistryLab’s nanoreactor feature enables an unguided search for chemical reactions using density functional theory (DFT)-based metadynamics simulations within a spherical cavity or wall potential, confining the reactants. Molecules or molecular clusters can be placed in the nanoreactor and heated to trigger chemical reactions, and the reaction products are determined automatically. Quantum chemical methods are employed to simulate the reactions, and a machine learning tool analyzes the complex reaction paths to provide further insights into the products and by-products formed.  

The first step to run the quantum nanoreactor simulation is to create a model of a fresh, brand-new electrolyte formulation. With just a few clicks, QuantistryLab allows the user to select the desired chemicals from the compound library and set their ratio. For this use case, the electrolyte formulation was created to replicate the composition of a commercial electrolyte containing ethylene carbonate (EC) and dimethyl carbonate (DMC) as solvents in an equal weight ratio, and lithium hexafluorophosphate (LiPF₆) as the lithium salt.

The next step is to extract the lithium solvation shells from this liquid electrolyte preparation. The solvation shell represents the lithium atoms in the electrolyte and the layer of molecules from the solvent that are directly in contact with them; Their distribution determines how the electrolyte reacts over time when exposed to high temperatures.

Once the lithium solvation shells have been extracted, a simulation workflow can be started using the QuantistryLab nanoreactor feature with just a few clicks. Acetone was selected as a solvent to replicate the environment in a liquid electrolyte mixture with a dielectric constant of 20.7. To study the effects of thermal decomposition on the electrolyte, the temperature was set to 600 °K (~327 °C).

The simulation yields a histogram listing the most likely products of the electrolyte decomposition reaction. Among them numerous compounds can be found that have been previously reported in the literature as typical products of electrolyte degradation, such as carbon dioxide, fluoromethane, formaldehyde, hydrogen and methane.  

Many of the main products of the reaction are gases, which can contribute to the swelling of battery cells at high temperatures. Among the reaction products are also multiple flammable chemicals that are known to be responsible for battery cells exploding and catching fire, including hydrocarbons such as ethylene and propane. Corrosive compounds like hydrofluoric acid were also found, which further contribute to the degradation of the battery cell and increase the fire and explosion hazard.  

The nanoreactor simulation also identified several compounds containing bonds between phosphorus and fluorine atoms, which often pose a toxic hazard and have been reported as a product of electrolyte decomposition. Among them was methylphosphonyl difluoride, which is known to have corrosive and neurotoxic properties.

This use case demonstrates how QuantistryLab can investigate electrolyte decomposition and characterize reaction products to identify potentially hazardous compounds. The nanoreactor tool can be integrated into battery R&D workflows to study the decomposition products of novel electrolyte formulations, optimizing performance while minimizing safety risks. Understanding the by-products formed from a specific formulation provides crucial insights, allowing researchers to refine and design formulations before lab testing, potentially avoiding costly analyses. Essentially, the quantum nanoreactor enables the exploration of electrolyte formulations that may perform better under specific conditions.

As the demand for lithium-ion batteries with more capacity and longer duration keeps growing, there is an increasing interest in studying the mechanisms behind electrolyte aging and degradation. Since the electrolyte is a key component determining a battery’s performance, optimizing electrolyte formulations is essential for the development of new battery materials.  

QuantistryLab offers a full range of computational solutions to study the properties of electrolyte formulations that go beyond characterizing decomposition reactions, which includes simulating key properties of the electrolyte such as thermal stability and viscosity.

‍

‍

‍

‍