Carbon Capture Modeling and Simulation Platform: A Coupled Microalgal Bioreactor-Yeast Fermentation Approach for Bioethanol

This paper presents an Electron/React-based desktop simulation platform that integrates Monod, Logistic, and Luedeking-Piret kinetic models to digitally twin a coupled microalgal bioreactor and yeast fermentation system for optimizing CO2 capture and bioethanol production.

The Gist

Imagine you have a factory that spews out a lot of smoke (Carbon Dioxide, or CO₂), which is bad for our planet. Usually, we just let that smoke float away into the sky. But what if, instead of letting it go, we could catch it, feed it to some tiny living things, and turn it into a fuel that powers our cars?

That is exactly what this paper is about. The researchers at Cyprus International University built a virtual "video game" or simulator that helps engineers figure out how to do this efficiently without having to build expensive, messy real-life factories first.

Here is the simple breakdown of how their system works and what they built:

1. The "Two-Step Dance" of Nature

The system relies on a team of two microscopic workers who pass a baton back and forth:

• Worker A: The Algae (The Chef)
  Imagine a tiny green plant called Chlorella vulgaris. It lives in a special glass tank (a bioreactor) filled with light. It eats the CO₂ smoke, drinks water, and uses sunlight to grow. As it grows, it turns that smoke into sugar (carbohydrates). Think of it as a chef turning raw ingredients into a delicious cake.
• Worker B: The Yeast (The Baker)
  Once the algae has made enough sugar, it gets harvested and passed to a second tank. Here, a different microbe, Saccharomyces cerevisiae (the same yeast used to make bread and beer), eats the sugar. As it digests the sugar, it produces bioethanol (a clean fuel) and releases CO₂ as a byproduct.

The Magic Loop:
Here is the clever part: The CO₂ that the yeast releases isn't thrown away. It is caught and sent back to the algae tank to be eaten again. It's like a recycling loop where the waste of one worker becomes the food for the other, creating a nearly closed circle that saves money and reduces pollution.

2. The Problem: Real Life is Messy and Expensive

In the real world, building these tanks, testing different temperatures, and trying to find the perfect balance of light and food is incredibly expensive and time-consuming. If you get the math wrong, you might end up with a tank full of dead algae and no fuel.

Commercial software that does this already exists, but it's like buying a Ferrari to drive to the grocery store: it's too expensive, too complicated, and requires a special license to use.

3. The Solution: A "Digital Twin" App

This paper introduces a new computer program (a desktop app) that acts as a "Digital Twin" of the real factory.

• What it does: You can type in numbers like "How much light?" or "How much yeast?" into the app.
• The Engine: Behind the scenes, the app uses three famous mathematical rules (called Monod, Logistic, and Luedeking-Piret models). Think of these as the "laws of physics" for how fast these tiny creatures grow and eat.
• The Result: In just a few seconds, the app tells you: "If you do X, you will get Y amount of fuel." It also shows you pretty 3D pictures of the tanks so you can see what's happening inside.

4. How They Built It (The Tech Stuff)

The team wanted to make sure anyone could use this, not just people with expensive computers.

• They took the heavy math code (which used to take 2 minutes to run on old software) and rewrote it in a faster language (Python). Now, it runs in 4 seconds.
• They wrapped it in a user-friendly interface (like a modern website) that works on Windows, Mac, and Linux.
• They added 3D models so you can spin the virtual bioreactor around and look at it from different angles, making it great for students to learn.

5. Did It Work?

They tested the app with three different "scenarios":

1. The Perfect Day: They gave the system plenty of time and the right amount of food. The app predicted a 93% efficiency, meaning it almost perfectly converted the smoke into fuel.
2. The Industrial Scale: They simulated a huge factory running for a long time. It still worked very well (92% efficiency).
3. The "Oops" Moment: They tried to run it for only 1 hour with very little food. The app correctly predicted that almost nothing would happen (1% efficiency), because the microbes didn't have time to wake up and start working.

Why This Matters

This platform is like a flight simulator for bio-fuel engineers.

• For Students: It makes learning about biology and engineering fun and visual.
• For Scientists: It lets them test thousands of ideas quickly and cheaply before building a real lab.
• For the Planet: It helps us design better systems to turn our pollution into clean energy, fighting climate change with biology.

In short, they built a free, easy-to-use tool that helps us figure out how to turn "bad air" into "good gas" using tiny plants and yeast, all without spending a fortune on trial and error.

Technical Summary

1. Problem Statement

The paper addresses the critical environmental challenge of rising anthropogenic CO₂ levels and the need for scalable, economically viable Carbon Capture and Utilization (CCU) strategies. While biological routes coupling microalgal CO₂ fixation with fermentative biofuel production show promise, there is a significant gap between theoretical process design and accessible computational tools.

• Limitations of Existing Tools: Commercial simulators (e.g., Aspen Plus, SuperPro Designer) are expensive, require specialized licensing, and demand high expertise. Purely mathematical implementations often lack user-friendly interfaces for rapid scenario exploration.
• The Gap: There is a need for a cost-effective, accessible "digital twin" that integrates rigorous bioprocess mathematics with intuitive visualization to allow researchers and students to evaluate integrated CO₂-to-bioethanol systems without heavy computational overhead or software costs.

2. Methodology

The authors developed a desktop application that simulates a two-stage biological system: CO₂ fixation by Chlorella vulgaris (microalgae) followed by bioethanol production by Saccharomyces cerevisiae (yeast), with a closed-loop CO₂ recycling subsystem.

A. Biological System Design

• Stage 1 (Photochemical Bioreactor): C. vulgaris utilizes CO₂, light, and nutrients to produce carbohydrate-rich biomass. The system uses an airlift bioreactor with LED lighting and controlled pH/temperature.
• Stage 2 (Fermenter): Biomass is hydrolyzed to glucose, which is fermented by S. cerevisiae into ethanol and CO₂.
• Recycling Loop: CO₂ off-gas from fermentation is captured, purified, and recycled back into the algal bioreactor to enhance carbon utilization efficiency.

B. Mathematical Modeling

The simulation engine integrates three established kinetic models:

1. Monod Model: Describes substrate-limited growth (CO₂ for algae, glucose for yeast).
   • $\mu = \mu_{max} \cdot S / (K_s + S)$
2. Logistic Growth Model: Accounts for carrying capacity ($X_{max}$) and density-dependent growth limitations, generating sigmoidal growth curves.
   • $dX/dt = \mu_{max} \cdot X \cdot (1 - X/X_{max})$
3. Luedeking-Piret Model: Describes product formation (ethanol) as a combination of growth-associated ($\alpha$) and non-growth-associated ($\beta$) components.
   • $dE/dt = \alpha \cdot (dX/dt) + \beta \cdot X$

The model also includes mass balances for CO₂ consumption and environmental control algorithms (temperature, pH, light intensity) that act as correction factors on kinetic rates.

C. Software Architecture & Implementation

• Frontend: Built with React (v19.1.0) and Electron (v35.1.5) for a cross-platform desktop application (Windows, macOS, Linux). It features interactive 3D bioreactor visualizations using Three.js and @react-three/fiber.
• Backend: The simulation engine was migrated from MATLAB to Python 3.13 (using NumPy for integration and Matplotlib for plotting).
  • Performance Gain: The Python implementation reduced execution time from 2 minutes (MATLAB) to **4 seconds**, a 30-fold improvement.
  • Deployment: Compiled into a portable executable using PyInstaller, eliminating the need for users to install scientific computing environments.
• Workflow: Users input parameters via a React interface; the Electron main process invokes the Python backend via IPC; results (graphs, PDF reports, numerical data) are returned and visualized in real-time.

3. Key Contributions

1. Integrated Simulation Platform: A novel desktop application that couples validated kinetic models with an interactive 3D visualization environment, bridging the gap between complex bioprocess engineering and user accessibility.
2. Closed-Loop Carbon Modeling: The platform explicitly models a partially closed-loop system where fermentation off-gas is recycled to the bioreactor, allowing for the assessment of carbon utilization efficiency.
3. Performance Optimization: The strategic shift from MATLAB to a compiled Python/Electron stack significantly improved computational speed and removed licensing barriers, making the tool accessible to a broader academic and research community.
4. Multi-Parameter Efficiency Metrics: The system calculates three distinct efficiency metrics:
   • Molar Efficiency: Mole-to-mole conversion of CO₂ to ethanol.
   • Mass Yield Efficiency: Grams of ethanol per gram of CO₂ fed.
   • Overall Process Efficiency: Percentage of the theoretical maximum yield (0.348 g ethanol/g CO₂) achieved.

4. Results

The platform was validated through three representative simulation scenarios with varying initial conditions and durations:

• Run 1 (Moderate, 5 hours): Demonstrated the transition from exponential to stationary growth. Achieved 93.78% of theoretical maximum efficiency, validating the model's ability to resolve near-complete substrate utilization.
• Run 2 (Industrial Scale, 24 hours, High Inoculum): Simulated a prolonged industrial run. Produced 6.64 g/L ethanol with 92.99% efficiency. The slight efficiency drop was attributed to higher CO₂ production from increased biomass, consistent with theoretical expectations.
• Run 3 (Minimal, 1 hour): Simulated a system in the lag phase. Resulted in 1.09% efficiency and negligible ethanol production, correctly modeling the behavior of an under-inoculated or prematurely terminated system.

Key Findings:

• The platform successfully resolves the coupled dynamics of algae and yeast, accurately predicting biomass growth, substrate consumption, and ethanol yield.
• Efficiency metrics ranged from 1.09% to 93.78%, proving the tool's sensitivity to operational parameters.
• The software execution time of ~4 seconds allows for rapid "what-if" scenario analysis, which was not feasible with the previous MATLAB implementation.

5. Significance and Future Outlook

• Educational Value: The 3D visualization and interactive interface make complex bioprocess concepts (kinetics, mass balance, closed-loop systems) accessible to students and non-experts.
• Research Utility: Provides a low-cost pre-feasibility tool for researchers to screen operating conditions before committing to expensive experimental campaigns.
• Technological Impact: Demonstrates that high-fidelity bioprocess simulation can be achieved using open-source technologies (Python, React) rather than proprietary commercial software.

Limitations & Future Work:

• The current model is deterministic and does not account for stochastic variability (e.g., contamination, microbial evolution).
• Energy consumption (lighting, compression) is not yet explicitly modeled for techno-economic analysis.
• Future versions aim to incorporate stochastic modeling, real-time sensor integration (Digital Twin capabilities), and expansion to other biofuel pathways (biodiesel, biogas).

In conclusion, this platform represents a significant step toward democratizing the design and optimization of sustainable carbon capture and biofuel production systems, offering a robust, fast, and accessible alternative to traditional industrial simulators.