Precursor and Activation Design
Model coal, coconut shell, wood, biomass, polymer, or residue feedstocks against carbonization temperature, heating rate, residence time, steam/CO₂ activation, chemical activation, washing, and burn-off constraints.
Activated Carbons Screening and Design
Activated Carbons Screening and Design
AI-guided optimization of precursors, activation, pore structure, surface chemistry, adsorption performance, and fixed-bed processes for validation-ready carbon materials.
Start Your ProjectOverview
Design activated carbon around the feed, adsorbate, and operating cycle
Activated carbon is not a single, structure-defined material. Adsorption performance emerges from precursor composition, carbonization and activation history, pore-size distribution, surface oxygen and heteroatom chemistry, mineral matter, particle form, moisture, and the conditions used to measure it. High BET area alone cannot establish working capacity, selectivity, kinetics, mechanical durability, or performance in a competitive feed.
Our AI for Porous Adsorbent Materials Screening and Design workflow connects heterogeneous production data, pore and surface characterization, adsorption experiments, molecular descriptors, interpretable machine learning, fixed-bed modeling, and validation planning. The result is a material–process shortlist—not an isolated capacity prediction.
Core Services
From precursor variability to bed-scale performance
Pore Architecture Optimization
Relate ultramicropore, micropore, mesopore, and transport-pore volumes to adsorbate size and phase using gas-sorption data, pore-size distributions, density, and fit-for-purpose structural descriptors.
Surface Chemistry and Ash Control
Assess oxygen functionality, pH at point of zero charge, heteroatom doping, hydrophobicity, ash, leachable metals, and deashing or post-treatment options for the intended water or gas matrix.
Adsorption and Competition Modeling
Analyze isotherms, kinetics, heats, selectivity, moisture effects, pH/ionic-strength sensitivity, dissolved organic matter, and co-adsorbates without treating single-solute batch capacity as universal performance.
Granule, Pellet, and Bed Engineering
Connect particle-size distribution, apparent density, hardness, attrition, binder, mass-transfer zone, empty-bed contact time, head loss, breakthrough, and vessel constraints.
Regeneration and Lifecycle Screening
Evaluate thermal or other regeneration routes, heel formation, pore blockage, mass loss, reactivation conditions, contaminant destruction requirements, replacement frequency, and lifecycle trade-offs.
Integrated Workflow
From application specification to validation-ready carbon
| Stage | Key Activities | Decision Output |
|---|---|---|
| 1. Define use case | Fix feed composition, target and competing adsorbates, phase, humidity, pH, temperature, flow, cycle, vessel, compliance limits, cost, and disposal or regeneration route. | Performance specification and constraints |
| 2. Audit material and test data | Harmonize precursor, activation basis, burn-off, yield, washing, particle form, BET/pore analysis, surface chemistry, isotherm, kinetic, column, and regeneration metadata. | Curated dataset and evidence gaps |
| 3. Learn process–structure links | Use interpretable ML and response surfaces with precursor-aware validation to relate manufacturing variables to yield, pore hierarchy, chemistry, density, and strength. | Feasible production windows with uncertainty |
| 4. Screen adsorption performance | Combine target-specific descriptors, equilibrium/kinetic models, competitive-feed corrections, matrix effects, and uncertainty-aware ranking. | Candidate portfolio and failure-risk map |
| 5. Translate to fixed-bed operation | Estimate working capacity, mass-transfer zone, breakthrough, empty-bed contact time, pressure drop, carbon usage rate, guard-bed needs, and regeneration frequency. | Material–bed Pareto shortlist |
| 6. Validate and update | Specify carbon preparation or sourcing, characterization, batch/isotherm tests, RSSCT or pilot columns, humidity/matrix challenge, attrition, cycling, and model recalibration. | Executable validation plan |
Manufacturing and adsorption are optimized together
Activation severity can open micropores but reduce yield, density, strength, or regeneration stability. Mesopores may improve access for larger molecules while consuming carbon volume that could otherwise contribute to small-gas uptake.
We therefore rank production windows and finished-carbon forms against the actual adsorbate, matrix, contact time, mechanical duty, and lifecycle—not against surface area alone.
Deliverables
Decision-ready files for carbon selection and scale-up
Curated Carbon Dataset
Precursor, carbonization, activation, washing, yield, burn-off, particle form, pore, surface, adsorption, column, and regeneration data with provenance and exclusions.
Process–Property Model
Validated relationships between controllable manufacturing variables and yield, pore hierarchy, surface chemistry, density, hardness, and target performance.
Candidate and Supplier Matrix
Ranked carbon grades or development conditions with confidence, applicability limits, critical quality attributes, sourcing assumptions, and rejection rationale.
Adsorption and Breakthrough Package
Isotherm/kinetic fits, competitive-feed scenarios, working capacities, breakthrough estimates, mass-transfer assumptions, sensitivity ranges, and data-quality flags.
Bed and Regeneration Window
Particle-size and contact-time recommendations, pressure-drop constraints, carbon usage estimates, change-out logic, regeneration options, and capacity-loss scenarios.
Experimental Validation Plan
Prioritized samples, characterization, matrix-matched batch tests, RSSCT/pilot design, analytical endpoints, attrition/cycling tests, and acceptance criteria.
Applications
Activated carbon projects across water, gas, and recovery operations
Drinking Water and PFAS Treatment
Carbon selection for trace organic micropollutants, natural organic matter competition, short-chain breakthrough risk, RSSCT design, and carbon-use rate.
Wastewater Polishing
PAC or GAC screening for color, COD, pharmaceuticals, pesticides, and industrial contaminants under realistic pH, ionic strength, and background organics.
Air and VOC Control
Vapor-phase capacity, humidity competition, ignition and desorption risk, cartridge or vessel sizing, solvent recovery, and guard-bed strategies.
Gas Purification and Carbon Capture
Ultramicropore and surface-chemistry tuning for CO₂, H₂S, siloxanes, trace impurities, biogas upgrading, and regenerable cyclic adsorption.
Food, Beverage, and Process Purification
Decolorization, odor removal, impurity polishing, low-leachable carbon selection, powdered-versus-granular form, and filtration compatibility.
Solvent and Resource Recovery
Adsorption–desorption balance, working capacity, steam or thermal regeneration, product recovery, heel formation, and long-term cycle economics.
Scientific Evidence
Open evidence for data-driven production and adsorber design
An open-access study demonstrates the use of machine-learning models to predict adsorption behavior for waste-derived biochar-activated carbon.1 Published GAC research also shows that carbon properties, contaminant descriptors, and water-quality variables can support early-breakthrough prediction, while retaining substantial uncertainty that warrants pilot validation.2
Screening models are most useful when precursor class, activation route, analytical method, adsorbate domain, matrix, and operating conditions are represented in the training data. Predictions outside that domain are flagged for targeted testing.
EPA guidance describes contaminant adsorption on both outer and inner GAC surfaces, fixed-bed treatment, spent-carbon replacement, and high-temperature regeneration as core practical elements of GAC operation.3
1 Chang, J.; Lee, J.-Y. Machine Learning-Based Prediction of the Adsorption Characteristics of Biochar from Waste Wood by Chemical Activation. Materials 2024, 17, 5359. https://doi.org/10.3390/ma17215359. Distributed under Open Access license CC BY 4.0.
2 Koyama, Y.; Fasaee, M. A. K.; Berglund, E. Z.; Knappe, D. R. U. Machine Learning Models to Predict Early Breakthrough of Recalcitrant Organic Micropollutants in Granular Activated Carbon Adsorbers. Environmental Science & Technology 2024, 58, 17114–17124. https://doi.org/10.1021/acs.est.4c01316.
3 U.S. Environmental Protection Agency. Community Guide to Granular Activated Carbon Treatment, EPA 542-F-25-003, 2025, p. 1. https://semspub.epa.gov/work/HQ/401595.pdf. Publicly available, with modification.
Project Strategy
Recommendations tied to measurable carbon quality attributes
We keep precursor lot, activation basis, wash state, particle form, pore-analysis method, surface chemistry, feed matrix, model version, uncertainty, and validation handoffs visible throughout the project. To discuss a carbon grade, precursor route, fixed-bed system, regeneration problem, or internal dataset, please Contact Us or submit the Online Inquiry below.
Related Services
Connected porous adsorbent capabilities
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