Solvent and Salt Screening
Rank carbonate, ether, phosphate, nitrile, sulfone, ionic-liquid, and alternative solvent systems with sodium salts using physicochemical, transport, and stability descriptors.
Design sodium-compatible electrolyte systems by connecting solvent and salt selection with Na+ solvation, ion transport, electrochemical stability, and electrode-specific interphase requirements. CD ComputaBio builds decision-focused computational workflows for hard-carbon, layered-oxide, Prussian blue analogue, polyanionic, and other sodium-ion battery chemistries.
The larger ionic radius and different Lewis acidity of Na+ change coordination structure, desolvation, mobility, and interfacial reaction preferences. A solvent or additive that performs well in a lithium-ion cell may therefore produce weak transport, excessive gas, unstable cathode electrolyte interphase, or poor hard-carbon compatibility in a sodium-ion system.
Our studies translate these coupled effects into a ranked formulation strategy, with assumptions, uncertainty, and experimental validation priorities made explicit.
Select individual modules or combine them into a staged screening program from broad chemical space to experimentally testable candidates.
Rank carbonate, ether, phosphate, nitrile, sulfone, ionic-liquid, and alternative solvent systems with sodium salts using physicochemical, transport, and stability descriptors.
Characterize coordination number, solvent–anion competition, contact ion pairs, aggregates, residence times, and concentration-dependent speciation.
Evaluate self-diffusion, correlated transport, viscosity-related behavior, ionic association, and temperature-sensitive mobility using molecular simulation and data-driven models.
Assess reductive stability, adsorption, decomposition tendencies, and additive strategies relevant to irreversible capacity loss and stable SEI formation.
Prioritize electrolyte systems for layered oxides, Prussian blue analogues, polyanionic compounds, and other sodium cathodes under the intended voltage window.
Integrate calculated descriptors and experimental results through mixture models, active learning, or Bayesian optimization to propose the next formulations to test.
The most useful electrolyte is the one that satisfies the cell-level constraints together. We compare candidates across a transparent set of criteria rather than optimizing one descriptor in isolation.
Diffusion, association, viscosity-related descriptors, conductivity trends, and temperature response.
Oxidation/reduction susceptibility, decomposition pathways, gas-forming risk, and solvent–salt reactivity.
Adsorption, early decomposition products, SEI/CEI hypotheses, and compatibility with coatings or binders.
Concentration, miscibility, flash-point or safety considerations, cost boundaries, and available raw materials.
Each modeling layer is selected according to the formulation decision, available evidence, and level of mechanistic resolution required.
Broad candidate spaces are screened first, while higher-cost simulations are reserved for candidates where additional mechanistic evidence can change the experimental shortlist.
Prioritized solvent–salt–additive combinations with reasons for inclusion and exclusion.
Solvation, transport, redox, adsorption, interphase, and concentration-dependent interpretations.
Structures, descriptors, simulation outputs, plots, calculated values, and method documentation.
Recommended controls, concentration ranges, electrochemical readouts, and next-round rules.
Yes. A study can begin from a defined molecular or formulation space, a literature-supported baseline, and clear cell constraints. The workflow can later incorporate proprietary cycling or characterization data as it becomes available.
Yes. The electrolyte decision can be evaluated against the intended cathode and anode pair, including hard carbon, layered oxides, Prussian blue analogues, polyanionic materials, and other specified systems.
The output depends on the available data and modeling level. We can calculate transport descriptors and trends, build calibrated predictive models when suitable reference data exist, and clearly distinguish calculated values from model-based estimates.
Yes. New cycling, impedance, spectroscopy, or physical-property measurements can be used to update ranking models and select the next formulations through an active-learning workflow.
Share your electrode chemistry, baseline formulation, candidate list, available data, and target operating conditions.
Submit your project details below, and our team will respond within 24 hours.
Talk to our technical team about your project!
I Want To Talk