CBMS-Sim is an industrial emissions control simulation platform designed to model, analyze, and validate a novel coral-inspired biomineralization technology for capturing multiple pollutants in a single integrated process.
Copyright (c) 2026 Sahil Khutey. All Rights Reserved. Licensed under the Business Source License 1.1 (BSL 1.1). See the LICENSE file for terms.
The Coral-Inspired Biomineralization Multi-Pollutant Solidification System (CBMS) represents a paradigm shift in industrial emissions control for hard-to-abate sectors.
- Low-CAPEX Turnkey Retrofit: CBMS requires ~40% of the CAPEX of conventional amine-scrubbing, electrocatalytic, or Metal-Organic Framework (MOF) based carbon capture systems.
- Dual Revenue Model:
- CCTS Credits: Earn compliance or voluntary carbon credits under the Indian Carbon Market (CCTS) at a baseline price of ₹1,850/tCO₂.
- Circular Solid Products: Compaction of captured carbonate sludge into construction-grade gypsum and biomineralized bricks, eliminating geological storage costs and creating a local revenue stream.
- Indigenous Supply Chain: Uses biowaste chitosan sourced from Indian seafood processors, protecting the supply chain from geopolitical volatility.
CBMS-Sim is structured as a monorepo consisting of high-performance scientific solvers, web services, and an interactive digital twin visualization dashboard.
┌──────────────────────────┐
│ Client Layer (React) │
│ - Digital Twin DCS │
│ - Results Dashboards │
└────────────┬─────────────┘
│ WebSocket / REST
┌────────────▼─────────────┐
│ API Layer (FastAPI) │
│ - Tenant auth, CORS │
└──────┬─────────────┬─────┘
│ │
┌────────────▼─────┐ ┌─────▼──────────────┐
│ Workers (Celery) │ │ sim-core (Numba) │
│ - Report, PDF │ │ - Stiff ODE Solver │
└──────────────────┘ └────────────────────┘
packages/sim-core: Numba-accelerated scientific kernels for chemical kinetics, mass balance, Wiener process simulations, and Latin Hypercube sampling.packages/api: FastAPI application implementing user authentication, role-based access control, CRUD endpoints, and WebSockets.packages/workers: Celery tasks for heavy report generation and background simulations.packages/web: React 19, TypeScript, Tailwind CSS, and Vite frontend for user configuration.
Ensure you have Python 3.12, Node.js 20, pnpm, and Poetry installed.
- Python dependencies:
poetry install
- Node.js dependencies:
pnpm install --ignore-scripts
- Run Tests:
poetry run pytest
- Build Frontend:
pnpm build:web
The scientific core of CBMS-Sim models the accelerated capture of
- Stiff Kinetic Solver: Models a 9-species ordinary differential equation (ODE) system capturing chemical equilibria in viscous media.
- Biocatalytic Kinetics: Employs Michaelis-Menten kinetics modeling carbonic anhydrase (CA) enzyme deactivation over time using Arrhenius equations.
-
Chitosan-Templated Mineralization: Simulates the nucleation and crystal growth of
$\text{CaCO}_3$ on chitosan polymer scaffolds. - Uncertainty Quantification: Runs parallel Monte Carlo simulations and Latin Hypercube Sampling (LHS) paired with Sobol global sensitivity analysis to map key parameters.
The mathematical models and calibration pipelines are structured for multi-stage empirical validation:
-
Literature-Derived Benchmarks: Initial model equations and core physical constants are derived from and aligned with peer-reviewed literature (e.g.,
$\text{CO}_2$ hydration rates within 5% of Mirjafari et al. 2007, gypsum equilibrium comparisons to PHREEQC within 2%). - Parameter Confidence Tiers: As documented in provenance.md, currently approximately 10% of simulation parameters are directly measured in-house, with the remaining 90% literature-derived or initial design placeholders.
-
Validation Pipeline: The calibration pipeline is currently validated using synthetic assays (e.g., mock CE-1 rates). Phase 1 research targets transitioning the codebase to
$\ge 60%$ in-house measured parameters as experimental data from physical CE-1, CE-2, and CE-3 trials becomes available.
The mathematical framework, stochastic modeling methods, and parameter registries are detailed in the peer-reviewed canonical manuscript:
CBMS-Sim is released under the Business Source License 1.1 (BSL 1.1). See LICENSE for the full text.
| If you want to... | You need... |
|---|---|
| 👀 Read the code | ✅ Nothing — it's source-available |
| 🔬 Run it for academic research | ✅ Nothing — non-production use is free |
| 🧪 Run a pilot (<100 hrs/month or <10 t CO₂/month) | ✅ Nothing — within Additional Use Grant |
| 🛠️ Contribute code | ✅ Sign the CLA (see CONTRIBUTING.md) |
| 📚 Cite it in a paper | ✅ Use the CITATION.cff format |
| 🏭 Run it in production (>100 hrs/month) | 💰 Commercial license — email licensing@cbms.in |
| 💼 Fork for competing product | 💰 Commercial license required |
| 🔒 Share our trade secrets | ❌ Don't. They're protected. |
✅ Freely shareable under BSL 1.1:
- All source code in this repository
- Documentation (
docs/,manuscript/) - Validation cases against published data
- API specifications
- Scientific methodology (published parts)
- Customer-specific simulation results
- Pilot operational data from named partners
- Detailed cost models with vendor-specific pricing
- Investor due diligence materials (under NDA)
❌ Never share — protected as trade secrets:
- Proprietary reaction kinetics coefficients (beyond published literature)
- Cost-optimized reagent formulations with specific suppliers
- Pilot plant tuning parameters
- Unpublished bench-scale experimental data
- Internal financial projections and cap table
- Customer pipeline and sales data
If you're unsure whether something is shareable, ask first:
- 📧 General IP: ip@cbms.in
- 📧 Commercial licensing: licensing@cbms.in
- 📧 Security disclosures: security@cbms.in
By submitting a pull request, you agree that:
- You retain copyright to your contributions
- You grant CBMS Technologies a non-exclusive license to use your contributions
- Your contributions are licensed under BSL 1.1 (the same as the project)
We use CLA Assistant to automate this. The CLA bot will comment on your first PR with a link to sign.