Gas Separation Membranes
Screen polymeric, mixed-matrix and microporous membrane candidates for permeability, selectivity, mixed-gas behavior, plasticization, aging and manufacturability.
Explore service →AI for Materials
AI for Membranes and Solvent Systems
Physics-informed AI and molecular modeling for membrane transport, ion conduction, green solvent selection and application-specific solvent replacement.
Start Your ProjectService overview
Connect molecular design with membrane and solvent performance
Membrane and solvent development spans two coupled design spaces: the chemistry and morphology that control transport through a selective layer, and the liquid environment that controls solubility, speciation, viscosity, conductivity, reaction compatibility and recovery. We build project-specific workflows that translate operating conditions into measurable targets, combine experimental evidence with molecular simulation and AI, and prioritize candidates that remain credible beyond a single idealized property.
MembranesPermeability, selectivity, conductivity and stability
SolventsSolvation, phase behavior, viscosity and compatibility
TranslationCandidate ranking, trade-offs and validation plans
Specialized services
Five pathways for transport and liquid-system design
Each service uses representations, property models and validation criteria matched to the material family and intended process rather than applying one generic screening model.
Ion-Exchange and Energy Membranes
Relate fixed-charge chemistry, hydration, morphology and reinforcement to ion conductivity, selectivity, swelling and chemical durability.
Explore service →Green Solvent Screening
Rank solvents for solubility, selectivity, reaction or extraction performance while integrating volatility, hazard, sourcing and recovery constraints.
Explore service →Solvent Replacement
Identify feasible substitutes for restricted or undesirable solvents using functional equivalence, process compatibility and change-risk assessment.
Explore service →Ionic Liquids and Deep Eutectic Solvents
Explore component identities and ratios for target solvation, gas capture, extraction, catalysis or electrochemical performance.
Explore service →Integrated approach
From target operating window to testable formulations
The workflow connects chemistry, morphology and process conditions while preserving the distinction between intrinsic properties and device- or process-level performance.
✓
Define the functional target
Translate feed composition, temperature, pressure, water content, electrochemical conditions or unit operation into quantitative property and exclusion criteria.
✓
Represent chemistry and structure
Encode polymer repeat units, free volume, ionic groups, morphology, solvent molecules, ion pairs, mixture composition and hydrogen-bond networks.
✓
Combine physics and data
Integrate transport models, MD, quantum chemistry, COSMO-based methods, curated measurements and calibrated ML surrogates at the fidelity required.
✓
Rank with practical constraints
Balance performance with stability, viscosity, swelling, toxicity, flammability, recovery, material availability and fabrication feasibility.
Method selection
Methods selected for the relevant transport or solvent question
Model fidelity is chosen according to the endpoint, data coverage and required confidence; screening predictions are separated from claims that require mixed-feed, long-term or process-scale validation.
| Project need | Typical methods | Key inputs | Decision output |
|---|---|---|---|
| Gas-transport screening | QSPR/graph ML, solution–diffusion descriptors, sorption and diffusion models | Polymer or framework chemistry, gas pair, temperature, pressure and available measurements | Permeability/selectivity estimates, uncertainty and trade-off position |
| Mixed-gas and stability risk | Multicomponent sorption, MD/GCMC as appropriate, plasticization and aging models | Feed composition, condensable species, film history, thickness and test conditions | Risk flags, credible operating window and validation matrix |
| Ion-conducting membranes | Molecular dynamics, morphology descriptors, conductivity and swelling surrogates | Fixed-charge chemistry, ion form, hydration, temperature and reinforcement variables | Conductivity/selectivity balance, hydration window and structural design rules |
| Solubility and liquid–liquid selectivity | DFT/COSMO-RS, activity-coefficient models, MD and data-driven property prediction | Solute, solvent/mixture identity, composition, temperature and phase constraints | Solubility, partitioning, selectivity and phase-behavior shortlist |
| Green solvent or replacement search | Similarity/QSPR, Hansen or solvatochromic descriptors, hazard and process scoring | Functional role, incumbent solvent, required properties, EHS and equipment constraints | Substitution matrix, trade-offs, change risks and test plan |
| Ionic liquid and DES design | Component/ratio descriptors, COSMO-based screening, ML, MD and viscosity models | Cation/anion or HBA/HBD candidates, ratios, water content and target application | Ranked formulations with viscosity, conductivity, solvation and stability estimates |
| Process-level comparison | Mass/energy balances, membrane module or solvent-recovery surrogates, sensitivity analysis | Flux or equilibrium data, stage/cycle design, utilities, recovery and purity targets | Process KPIs, operating window and value-of-information priorities |
Model boundary: ideal-film permeability, neat-solvent properties or infinite-dilution predictions are not treated as direct guarantees of mixed-feed, thin-film, recycled-solvent or full-process performance. Morphology, defects, impurities, water, viscosity, aging and recovery conditions are included as explicit validation risks.
Project workflow
A decision-led route from requirements to validation
Frame the operating window
Define separation, electrochemical, reaction or extraction conditions and the metrics that determine success.
Audit evidence and constraints
Review structures, formulations, test protocols, units, uncertainty, prohibited chemistries and manufacturing boundaries.
Build calibrated models
Select molecular, continuum and data-driven methods; benchmark against relevant materials or solvent systems.
Explore trade-offs
Search chemistry, composition and process variables under multi-objective performance, safety and feasibility constraints.
Design validation
Prioritize membrane fabrication, mixed-feed testing, conductivity, solubility, viscosity, cycling or recovery experiments.
Typical inputs
- Target separation, reaction, extraction or electrochemical application
- Feed/solute composition, impurities, water content and operating window
- Required permeability, selectivity, flux, conductivity, solubility or recovery
- Candidate membrane chemistries, solvent lists, formulations or synthesis variables
- Permeation, sorption, conductivity, swelling, phase-equilibrium or viscosity data
- EHS restrictions, regulatory screens, materials compatibility and equipment limits
- Cost, sourcing, recyclability and preferred validation budget
Typical deliverables
- Curated, condition-aware dataset and evidence-quality assessment
- Documented molecular/material representations and modeling protocol
- Predictions with uncertainty and applicability-domain analysis
- Permeability–selectivity, conductivity–swelling or solvent-performance trade-off maps
- Ranked membrane chemistries, solvent candidates or IL/DES formulations
- EHS, compatibility and recovery-aware substitution matrix
- Mechanistic and sensitivity analysis for key design variables
- Recommended fabrication, property-testing and process-validation plan
- Technical report, figures and agreed reusable data/model outputs
Representative engagements
Projects built around distinct development decisions
Mixed-gas membrane down-selection
Goal: identify polymer candidates that retain useful selectivity under pressure and condensable-gas exposure. Workflow: condition-aware data curation, chemistry descriptors, calibrated transport models, plasticization risk and thin-film validation design.
Electrolyte membrane optimization
Goal: improve conductivity without unacceptable swelling or loss of selectivity. Workflow: fixed-charge and hydration descriptors, morphology-informed modeling, multi-objective ranking and targeted humidity/temperature testing.
Restricted-solvent replacement
Goal: replace an incumbent solvent while preserving dissolution and process behavior. Workflow: functional-property profile, solubility and similarity models, EHS/recovery filters, compatibility risks and staged bench tests.
FAQ
Frequently asked questions
Can you work with a small proprietary membrane or solvent dataset?
Yes, after feasibility review. Small-data projects may use chemically meaningful descriptors, transfer learning, Gaussian processes, literature anchoring and targeted simulation. We separate interpolation from extrapolation and specify the minimum additional experiments when the endpoint is under-supported.
How do you address the permeability–selectivity trade-off?
We treat it as a multi-objective problem rather than optimizing one property in isolation. Candidate position relative to relevant material families is combined with uncertainty, operating conditions, plasticization, aging and film-formation considerations.
Can the workflow screen mixed solvents or variable IL/DES ratios?
Yes. Mixture composition, water content and temperature can be explicit variables. Depending on the system, we combine component descriptors, COSMO-based calculations, activity models, ML and selected molecular simulations, then recommend experimental checks for viscosity, phase stability and performance.
Does "green solvent" mean a solvent is automatically safe?
No. Low volatility or renewable origin alone is insufficient. Screening considers hazard, exposure potential, flammability, persistence, sourcing, process mass intensity, recovery and disposal, with the criteria tailored to the application and jurisdiction.
Can you predict membrane module or solvent-recovery performance?
Intrinsic-property screening can be extended to module or recovery models when the required geometry, mass-transfer, equilibrium, energy and operating data are available. These process outputs are scoped separately from material-level predictions.
Are membrane fabrication and solvent testing included?
The core service provides computational design and validation recommendations. Customer-generated data can be integrated iteratively; fabrication, analytical testing or coordinated experimental validation can be added as a separate work package.
Online Inquiry
Submit your project details below, and our team will respond within 24 hours.
x
Need help getting the data you need?
Talk to our technical team about your project!
I Want To Talk