Week 1 HW: Final Project Ideas

Coal mining landscape in Silesia Coal mining landscape in Silesia

Fashion Feeds on Pollution

Starting point

I grew up in Silesia, a region whose landscape, economy and sense of itself were built by coal. As Poland moves away from coal, I find myself less interested in how an industry ends than in what it leaves behind, and whether any of that residue can be put to work.

Coal mines keep emitting methane long after extraction stops. In 2023, Polish hard-coal mines generated roughly 752 million m³ of methane, of which around 551 million m³ went straight into the atmosphere (Borowski et al., 2025). Abandoned mines can continue venting for years after closure (Hadro et al., 2025; European Union, 2024). That is a very large, very local waste stream with almost no productive use attached to it.

I am currently based in Barcelona, near a research ecosystem that has just demonstrated something relevant: methane-eating bacteria can convert methane into protein-rich extracellular polymers (Aranda et al., 2026). A biological capability in Catalonia, a waste stream in Silesia, and a material question that belongs to my own practice in fashion and biomaterials. This project is an attempt to connect the three.

From material idea to innovation hypothesis

Reading Asin-Garcia et al. (2025), one argument stayed with me: there is a persistent gap between what biomanufacturing makes theoretically possible and what is technically and economically feasible, and engineering biology tends to generate hype well before industrial viability has been shown.

That reframed how I wanted to pose the project. Rather than starting from methane + bacteria = sustainable fashion, I want to start from a hypothesis that can fail:

What if methane-derived extracellular polymers become a flexible fashion material, and under what technical, material, economic and environmental conditions would that actually make sense?

Four questions follow from it. Would fashion need this material (desirability)? Can biology and material processing produce it (feasibility)? Could it become economically realistic (viable)? Would its lifecycle genuinely improve on what it replaces (responsibility)? Together, these form what I think of as the project’s bio-techno-economy, and I would rather establish early that the answer to one of them is no than arrive at a beautiful swatch with no system behind it.

What already exists

Innovation here does not mean inventing without precedent. It means finding an existing capability, understanding its limits, and recombining technologies and contexts.

Newlight Technologies’ AirCarbon is the clearest precedent: microorganisms transform greenhouse-gas carbon into PHB, part of the PHA family of biopolyesters, and the resulting material has been developed into sheets, fibres, leather-like materials and consumer goods (Newlight Technologies, 2026). What matters is not that I want to reproduce it, but that it proves an emission can be metabolically restructured into a new polymer rather than simply captured or burned.

Mango Materials is the closer precedent. The company co-locates production near waste methane sources and uses methanotrophs to manufacture P3HB, which can be formulated into fibres for textiles. The pathway from methane to microorganism to biopolymer to fashion already exists as a technological direction.

This is useful because it rules out the obvious framing. “Make fashion from methane” is not a research gap. The more productive question is where the current system still has friction, and that question led me to polymer recovery.

The bottleneck: intracellular PHA

PHAs such as PHB accumulate inside microbial cells. Manufacturing therefore, requires growing cells, harvesting them, breaking the biomass, recovering the polymer and purifying it. That downstream chain adds high cost and complexity.

The CH4-BIOPOL project at the Catalan Institute for Water Research (ICRA) addresses this directly by asking whether methane could instead produce extracellular polymers - material the organism secretes outside the cell, potentially shortening the recovery chain. From an innovation-strategy perspective, this is interesting because it changes not only the material but the manufacturing architecture. The most important innovation may not be a new polymer at all, but a simpler route to producing and recovering one.

In 2026, researchers associated with ICRA published work on Methylotuvimicrobium alcaliphilum 20Z showing that the organism converts methane into extracellular polymeric substances (EPS), and that the resulting material is consistently protein-rich (Aranda et al., 2026). Across fourteen combinations of pH, temperature and methane concentration, they found that pH strongly influenced EPS production, that the pH 6–8.5 range supported higher yields, and that greater methane availability increased EPS accumulation.

What the study demonstrates is a biological capability. What it leaves open is what the polymer can become. That gap is where I want to position the project. The question shifts from can biology make something to can biological production become a material system with a meaningful application.

1. What I want to biomanufacture

A standalone flexible material for fashion, made from methane-derived EPS, not a coating applied to an existing textile, but a self-supporting bio-skin that can be cut, stitched, draped, bonded, moulded, heat-formed or digitally fabricated. The applications I have in mind are those currently dependent on flexible synthetics: PU, TPU, PVC, synthetic leather-like sheets and flexible polymer surfaces.

I want to work backwards from a material brief rather than forwards from the biology. The material should be flexible, self-supporting, thin, cuttable, joinable, drapable, durable in use, and designed for biological end-of-life. That gives me a target against which the biology can be judged rather than simply described.

The first success criterion is not a garment but a 10 × 10 cm self-supporting swatch that behaves like a material rather than a laboratory residue. Initial evaluation would be qualitative: hand-feel, drape, thickness, transparency, foldability, stitchability, bonding, heat and water response; before moving to: tensile strength, elongation, tear resistance, water uptake, ageing and degradation. The distinction matters: a biological system can be perfectly capable of producing a polymer that is of no interest whatsoever as a fashion material.

As for why EPS rather than a copy of AirCarbon, Newlight and Mango validate that methane-derived PHA can become commercially relevant, which is evidence that the opportunity space is real. My question is whether secreted polymer offers a lower-complexity recovery route. I am not assuming that it does. That is precisely the hypothesis requiring techno-economic analysis rather than optimism.

2. The living system

The proposed chassis is Methylotuvimicrobium alcaliphilum 20Z, a methanotroph that uses methane as its carbon and energy source. Methane, oxygen and mineral nutrients go in; methane metabolism drives extracellular polymer production; EPS comes out, to be recovered and formulated into material. Carbon transformation happens through microbial metabolism rather than petrochemical polymerisation in a chemical plant.

The organism is haloalkaliphilic, which means it is adapted to salty, alkaline conditions. Ye et al. (2023) argue that extremophiles may offer real advantages for next-generation industrial biotechnology, including lower contamination risk and reduced dependence on expensive sterile fermentation. The appeal is that the organism itself could lower the infrastructure burden.

There is a tension here worth naming rather than smoothing over. Aranda et al. found EPS production was strongest around pH 6–8.5 and strongly inhibited at pH 11. The conditions that make an extremophile attractive for cheap industrial fermentation are not necessarily the conditions that maximise the product I want. This is exactly the kind of trade-off that only becomes visible when an idea moves from biological possibility toward techno-economic feasibility.

Geography as infrastructure

The project has three locations, and I want to treat them as one distributed system rather than three separate sites.

Silesia supplies the problem and the feedstock. My preferred input is methane that would otherwise be emitted, particularly from closing or abandoned mines, a deliberate constraint, since the project must not create any incentive to continue extraction.

Catalonia supplies the scientific capability, with ICRA in Girona having demonstrated the methane-to-EPS process. What I could contribute is an application pull rather than a technology push: instead of asking what else the organism can produce, ask what material performance would make this capability relevant to fashion.

London could supply the translation layer. LifeFabs Institute works across biofabrication, materials, fashion, automation and digital fabrication, with Opentrons Flex and OT-2 systems, bioprinters, incubators, microscopy and digital fabrication equipment, and has previously used automated liquid handling for polymer formulation research.

Asin-Garcia et al. (2025) argue that better coordination between research infrastructures, knowledge and data could accelerate the bioeconomy. My project does not need every capability in one building: feedstock and environmental data in Silesia, microbiology and fermentation data in Catalonia, material and fabrication data in London, and could function as exactly the kind of meta-workflow they describe.

3. Automation

Petzold and Mukhopadhyay (2025) identify the sheer size of the experimental design space as a major challenge in microbial biomanufacturing. My system has two stacked design spaces: pH, temperature, methane concentration, nutrients, salinity and time on the biological side, determining EPS yield and composition; then EPS concentration, plasticiser, crosslinking, drying and thickness on the material side, determining performance. Testing these manually becomes unmanageable very quickly.

I would structure development as a simplified Design–Build–Test–Learn cycle, and Petzold and Mukhopadhyay describe automation as particularly valuable at the test stage, where automated cultivation, sampling and analysis reduce error and improve reproducibility.

An Opentrons Flex would not handle methane gas - cultivation needs dedicated gas-handling equipment, but it could automate media preparation, nutrient dosing, sampling, EPS formulation, plasticiser and crosslinker screening, and sample preparation. A realistic first experiment: four EPS concentrations × four plasticiser levels × three drying conditions gives 48 formulations, systematically produced rather than hand-mixed one at a time.

The longer-term ambition is not automation but learning. Structured high-throughput experimentation generates datasets that can later support machine learning, predictive models and metabolic modelling, turning the output from “I made a biomaterial recipe” into a model of how biological and material variables drive performance.

This connects to the interoperability argument in Asin-Garcia et al. (2025), and I would organise the work around a small data lake rather than scattered lab notes: feedstock data (mine, location, CH₄ concentration, gas flow, impurities, mine status, existing utilisation); biology data (pH, temperature, CH₄, salinity, nutrients, culture time, EPS yield and composition); material data (formulation, plasticiser, crosslinker, thickness, strength, elongation, drape, water response, surface); end-of-life data (degradation conditions and time, mass loss, residual additives, reuse and recycling potential); and techno-economic data (energy, reactor volume, methane consumed, recovery yield, processing steps, labour, material yield, estimated €/kg).

The goal is eventually to be able to ask: given a particular methane source and a desired fashion property, which combination of biological and processing conditions should I choose? That turns a recipe into a decision-making system.

The techno-economic layer

A biologically interesting material is not automatically a viable innovation. It has to clear technical feasibility, material desirability, economic viability and genuine environmental improvement, and Asin-Garcia et al. (2025) specifically warn about the distance between what is theoretically possible and what is realistic.

I want to be prepared for this to fail. If producing 1 kg of EPS demands enormous methane input, extremely dilute fermentation, intensive purification, high-energy drying and expensive additives, then the responsible conclusion is not to scale this pathway. Establishing that early would still be a successful research outcome.

4. How much, and how long

How much

Since methane-derived EPS has not been demonstrated as a standalone fashion material, starting from a collection makes no sense. My first meaningful scale is one experimental wearable demonstrator: approximately 0.5 m² of material at 0.5 mm thickness.

  • 0.5 m² = 5,000 cm²; 0.5 mm = 0.05 cm
  • Volume: 5,000 × 0.05 = 250 cm³
  • Density is currently unknown; assuming 1 g/cm³ as a placeholder: 250 cm³ × 1 g/cm³ = ~250 g dry EPS-based material

Earlier methane-biomanufacturing research reported extracellular polymer concentrations of approximately 2.6 g/L (Cantera et al., 2018). This was a different methane-EPS system, so I treat it strictly as an order-of-magnitude reference rather than a prediction.

  • At 2.6 g/L: 250 g ÷ 2.6 g/L ≈ 96 L culture
  • Assuming 70% recovery: 250 ÷ (2.6 × 0.70) ≈ 137 L, rounded to ~140 L culture per demonstrator

That calculation immediately exposes the central problem: the polymer is very dilute. Increasing volumetric productivity may matter more than almost any other variable, which is exactly why automation, strain, process optimisation, and techno-economic analysis become load-bearing rather than decorative.

How long

Aranda et al. (2026) cultivated M. Alcaliphilum 20Z for approximately seven days when investigating EPS production. Adding two to three days for recovery, formulation, and drying gives roughly 9–10 days per production cycle.

At this stage, the realistic target is grams rather than hundreds of grams. The first milestone is simply whether a self-supporting flexible sample can exist at all.

5. What I don’t know yet

Is the feedstock compatible? The Catalan experiments use controlled laboratory methane. Real Silesian mine gas carries variable CH₄ concentration, CO₂, oxygen, humidity, and contaminants. The process could work beautifully on clean methane and fail on mine gas, and that needs testing before anything larger is built.

Is the polymer actually a material? Aranda et al. demonstrate that EPS exists. They do not demonstrate that it behaves like a fashion material. I do not yet know whether it is brittle, flexible, sticky, water-soluble, weak, strong or thermally processable. The opportunity exists precisely because that application space is untested.

How much modification does it need? If wearability requires large quantities of synthetic plasticiser, coating, stabiliser or crosslinker, the biological origin loses much of its meaning. The target is minimum intervention for the required performance.

Does extracellular actually mean cheaper? This is the strategic assumption most in need of challenge. Avoiding intracellular extraction is not the same as reducing total complexity; EPS may be too dilute, hard to purify, expensive to dry or chemically inconsistent. The real question is whether extracellular production reduces overall manufacturing complexity or simply relocates it.

Reflection: Why work on Earth

Averesch et al. (2023) examine microbial biomanufacturing for space exploration. The paper is technically fascinating, and it provoked a reaction I did not expect: biomanufacturing has been discussed for decades as a way to sustain human life where conventional supply chains cannot reach. I kept wanting to redirect that ambition toward repairing the material systems already around us.

Silesia is my example. It is a landscape shaped by extraction, energy, employment, pollution and now transition. I would rather explore how biotechnology could build a different material economy in this one than manufacture in another.

The historical logic runs extract, use, pollute, and abandon. The alternative I want to test runs identify waste, capture, metabolise, manufacture, use, recover, or biodegrade. That is not material substitution; it is potentially system innovation. Which means the interesting object of the project is not really the swatch, it is the relationship between regional waste, biological manufacturing, research infrastructure, data, fashion, economics, and end-of-life.

Could Silesia’s industrial past become biological feedstock for its post-coal future, without creating the next pollution problem? That is the hypothesis I want to test.

References

Aranda, A., Primo-Catalunya, D., Sánchez-Melsió, A., Balcázar, J. L., & Pijuan, M. (2026). Exploring the capabilities of Methylotuvimicrobium alcaliphilum 20Z to produce exopolymeric substances from methane. Enzyme and Microbial Technology, 199, 110893. DOI: 10.1016/j.enzmictec.2026.110893.

Asin-Garcia, E., Fawcett, J. D., Batianis, C., & Martins dos Santos, V. A. P. (2025). A snapshot of biomanufacturing and the need for enabling research infrastructure. Trends in Biotechnology, 43(5), 1000–1014. DOI: 10.1016/j.tibtech.2024.10.014.

Averesch, N. J. H., Berliner, A. J., Nangle, S. N., Zezulka, S., Vengerova, G. L., Ho, D., Casale, C. A., Lehner, B. A. E., Snyder, J. E., Clark, K. B., Dartnell, L. R., Criddle, C. S., & Arkin, A. P. (2023). Microbial biomanufacturing for space-exploration—what to take and when to make. Nature Communications, 14, 2311. DOI: 10.1038/s41467-023-37910-1.

Borowski, M., Zwolińska-Glądys, K., & Cheng, J. (2025). Methane emissions from coal mines: Quantification, capture, and utilization strategies for atmospheric impact mitigation—A case study from Poland. Atmosphere, 16(2), 174. DOI: 10.3390/atmos16020174.

Cantera, S., Lebrero, R., Rodríguez, E., García-Encina, P. A., & Muñoz, R. (2017). Continuous abatement of methane coupled with ectoine production by Methylomicrobium alcaliphilum 20Z in stirred tank reactors: A step further towards greenhouse gas biorefineries. Journal of Cleaner Production, 152, 134–141.

Cantera, S., Sánchez-Andrea, I., Lebrero, R., García-Encina, P. A., Stams, A. J. M., & Muñoz, R. (2018). Multi-production of high added market value metabolites from diluted methane emissions via methanotrophic extremophiles. Bioresource Technology, 267, 401–407. DOI: 10.1016/j.biortech.2018.07.057.

European Union. (2024). Regulation (EU) 2024/1787 on the reduction of methane emissions in the energy sector. Official Journal of the European Union.

Hadro, J., Jureczka, J., Suszka, G., & Strzemińska, K. (2025). The most important factors contributing to the effective development of abandoned mine methane resources in the Upper Silesian Coal Basin. Geological Quarterly.

ICRA — Catalan Institute for Water Research. (2022–2025). CH4-BIOPOL: Valorizing methane — biorefinery based on methanotrophs for the production of extracellular polymers. Project TED2021-129501B-I00.

Petzold, C. J., & Mukhopadhyay, A. (2025). From bench to biofactory: High-throughput technologies and automated workflows to accelerate biomanufacturing. Current Opinion in Biotechnology, 94, 103320. DOI: 10.1016/j.copbio.2025.103320.

Ye, J.-W., Lin, Y.-N., Yi, X.-Q., Yu, Z.-X., Liu, X., & Chen, G.-Q. (2023). Synthetic biology of extremophiles: A new wave of biomanufacturing. Trends in Biotechnology, 41(3), 342–357. DOI: 10.1016/j.tibtech.2022.11.010.

Industry and research precedents

LifeFabs Institute. Biofabrication R&D Studio and facilities, London.

Mango Materials. Methane-to-PHA biomanufacturing and textile applications.

Newlight Technologies. AirCarbon methane/greenhouse-gas-derived PHB technology.

AI disclosure

AI clients: ChatGPT (OpenAI, GPT-5.6 Sol) and Claude (Anthropic) Date: September 2026

I used both as research and ideation assistants while developing this proposal, and the process was iterative. I initially explored bacterial cellulose, textile-waste-to-PHA pathways and methane-derived PHBV before recent Catalan research on methane-derived extracellular polymers shifted the project toward EPS.

My prompts focused less on generating concepts than on stress-testing them through an innovation-strategy lens. Representative examples:

  • “What if pollution from Silesia could become a feedstock for fashion?”
  • “How is this different from AirCarbon and Mango Materials?”
  • “What existing technology validates the opportunity, and what gap is still unresolved?”
  • “How could recent Catalan methane-to-EPS research become a material innovation rather than remaining only a biological result?”
  • “What are the desirability, feasibility, viability and end-of-life risks of this system?”
  • “How could Silesia, Catalonia and London become parts of one distributed innovation ecosystem?”
  • “What techno-economic assumptions could make this project fail?”

AI outputs were used as research starting points. Scientific and factual claims were checked against the publications and primary sources listed above.