Gas fermentation for food protein

Gas fermentation grows microbial biomass on gases rather than sugar. For food protein the main route uses hydrogen-oxidising bacteria — chemolithoautotrophs such as Cupriavidus necator — which fix CO₂ into biomass through the Calvin cycle, using hydrogen as the electron donor and oxygen as the terminal electron acceptor. A second route uses methanotrophs, which grow on methane.

The appeal is that neither route needs arable land, sunlight or a crop. What it needs instead is energy, and the honest way to describe it is not land-free food but electricity-derived food.

The reaction, and why it is dangerous

The energy-yielding reaction is the oxidation of hydrogen by oxygen — the knallgas reaction — with the released energy driving CO₂ fixation. The organism therefore requires hydrogen and oxygen simultaneously, in the same vessel, in the same aqueous phase.

Hydrogen and oxygen mixtures are flammable across a famously wide composition range, and hydrogen ignites on very little energy. Keeping the gas phase outside that range while supplying both gases at the rates the culture demands is the defining engineering constraint of the platform. It drives the design toward gas-loop reactors with controlled recycle, tight composition monitoring and containment, and it is the reason this is not simply a conventional fermenter with a different sparge gas.

Mass transfer is the throughput limit

Oxygen is poorly soluble in water and is the usual ceiling on aerobic fermentation. Hydrogen is less soluble still. Volumetric productivity is therefore set by how fast hydrogen can be transferred from bubble to liquid, not by how fast the cells could grow if fed. Improving transfer means more agitation, higher pressure or better gas recycle — each of which costs energy and interacts with the flammability constraint above.

The efficiency chain has several steps

The realistic accounting runs: electricity → hydrogen by electrolysis → microbial metabolism → biomass → dried protein. Each step has a conversion efficiency below one, and they multiply. The overall electricity requirement per kilogram of protein is therefore substantial, and it is the single number that decides whether the route makes environmental sense.

That makes the environmental claim entirely dependent on the electricity source. On low-carbon power the case is strong, and the land-use comparison against soy or animal protein is genuinely favourable, because the process’s land footprint is essentially the plant plus its power generation. On grid average power in a fossil-heavy system, the case weakens sharply. Published figures for this sector diverge mostly because they assume different power mixes, and a figure quoted without its assumed grid is not interpretable.

The nutritional constraints are the familiar ones

The product is single-cell protein, so the nucleic-acid ceiling applies: rapidly growing bacteria are RNA-rich, dietary purines raise uric acid, and either an RNA-reduction step or a limit on inclusion level is required. Amino-acid profile is generally good. Methanotroph routes carry an additional framing question — methane is a fossil or biogenic feedstock depending on source, and that choice, not the fermentation, determines the carbon accounting.