Making oil is easy. Making it cheap is not.
Why fermentation derived ingredients fail at the price sheet, not in the lab
Prof. Dr. Jeroen Hugenholtz, CTO and co-founder, NoPalm Ingredients
Every few months, another company announces that it has produced a fat or an oil through fermentation. There is a photograph of a flask. There is a press release with the word breakthrough in it.
The biology is almost never the breakthrough.
Oleaginous yeast has been documented for decades. Any competent microbiologist can accumulate lipid in a strain and show you the result. What the announcement rarely mentions is what the material costs, and that is the number that decides whether something becomes an industry or stays a demonstration.
This is the central confusion in fermentation derived ingredients. The field celebrates feasibility, then acts surprised when commercialisation stalls. Feasibility was never the question.
The cost structure the industry inherited
Industrial biotechnology grew up serving pharmaceuticals and high value specialities. In that world the product sells by the gram, purity is everything, and the cost of the equipment barely registers against the value of the output. The engineering conventions that emerged make complete sense for insulin.
They are ruinous for a commodity.
Oils and fats trade by the tonne against published benchmarks. A buyer comparing your material against an incumbent specialty fat is not interested in the elegance of your process. Carry a pharmaceutical cost base into that market and the gap is not a few percent. It is a multiple.
Most fermentation ventures inherit that cost base without noticing, because it does not arrive as a decision. It arrives embedded in the equipment they buy or the feedstock they default to. Then they spend years trying to optimise their way out of a choice they made in their first six months.
Four levers, and when they close
There are only four places the cost of a fermentation derived ingredient actually sits.
What you feed the organism. How much product it makes from that feed. What the equipment costs to build and to run. What it takes to recover and refine the output to specification.
The trap is that these do not stay open for the same length of time.
Yield is the one that gets attention, because yield is what a research environment rewards and what looks good in a data set. It is also the lever that can be improved later, continuously, for the life of the company.
Feedstock and equipment class cannot. Change either one and you are not optimising a process, you are redeveloping it. Those decisions close early, quietly, and usually before anyone has thought about cost per tonne.
Our argument is that a company in this category should design backwards from the price it eventually has to hit, and accept a harder biology problem in exchange for a structurally lower cost base.
What that meant for us
We took three decisions early, and we took them together.
Upcycled side streams instead of purified sugar. Purified sugar is a clean, consistent feed. It is also expensive, and it competes with food. We feed our yeast side streams from the food industry instead. The cost is variability: batch to batch composition moves, and consistency becomes our problem rather than our supplier's. We have spent years learning to run a stable process on a feed that is not stable. What we get is a feed that is significantly cheaper, and one that carries nutrients purified sugar does not, which the organism can use. We take on the variability and we get a better and less expensive feed.
Non GMO strains instead of engineered ones. Engineering the strain is the obvious route to higher yield. We gave it up. European customers ask for non GMO and many will not run a trial without it, development is faster when there is no construct to manage, and the regulatory path is lighter, which matters when the ingredient already faces a Novel Food process in Europe. The cost is yield. That is precisely where our know how sits: everything we do to recover it without touching the genetics, from strain selection to feed strategy to how the fermentation is run. We surrender the shortcut and compensate in the fermenter.
Standard biomass fermentation equipment instead of pharmaceutical grade kit. It is a less controlled environment. It is also dramatically cheaper to build and dramatically cheaper to run, and the equipment already exists, so nobody has to invent a vessel for us. When your product sells by the tonne, capex and opex decide whether there is a business at all. Control is worth having. It is not worth that much.
Every one of the three has a price. Every one pays for itself somewhere else.
The argument against, stated honestly
Stack those decisions and you have removed three safety nets at the same time. You are solving a harder biology problem than a competitor running an engineered strain on clean sugar in good equipment, and that competitor will show better yield data than you for years.
That was the argument against , and it was a serious one. It remains the main risk in our approach. The most challenging decision at NoPalm Ingredients was never any single one of the three. It was doing all of them at once.
Our answer is that yield data is not the product. None of the three is remarkable alone. A non GMO strain on purified sugar in expensive equipment is a science project. Cheap equipment running an engineered strain is a different company's business model. The combination is the only route we have found to a fat that competes against incumbent fats on cost.
Free is not cheap
One correction worth publishing, because we see others making it.
We spent real time on feedstocks that were free. They were free because they required heavy pre treatment before yeast could use them, and building that pre treatment would have cost more than the feed saved. We dropped them.
The cost of a feedstock is not its price. It is its price plus the process you have to invent to use it. Free material with a development burden attached is one of the most expensive things you can put in a fermenter.
What the category should be measured on
If you are assessing a company in this field, as an investor, a customer or a potential partner, the flask tells you close to nothing. Four questions tell you most of it.
What does the organism eat, and what does that feed cost at industrial volume.
What class of equipment does the process require.
How many separate facilities does one production run touch.
What is the cost per tonne at a stated scale, with the scale stated.
A company that cannot answer those has not yet started on the hard part. A company that can is no longer running a science project.
NoPalm Ingredients produces REVÓLEO™, a specialty fat made through biomass fermentation of oleaginous yeast on upcycled agrifood side streams.