Skip to main content

Yeast Management and Propagation on Gluten-Free Wort

There are only three ways to put yeast in a gluten-free fermenter without a barley pathway somewhere behind it. Commercial production used all three, and the most common one was the cheapest: repitch from the last gluten-free batch.

Yeast Selection covers why the propagation medium is an ingredient. This page covers what you actually do about it, batch after batch.

The three supply routes​

Set out in writing in 2019, in response to a direct question from a gluten-free drinker about how a brewery can call wet yeast gluten-free:

Route 1 — propagate it yourself, from a single cell. Start from a single cell and step it up on your own gluten-free wort. It never sees barley wort at any stage. This was the preferred method.

Route 2 — buy pitchable quantities propagated to order. Over twelve years, two different yeast suppliers were willing to sell pitchable volumes grown on sugar, or on the brewery's own sorghum malt extract supplied to them. Note what that means practically: the medium is negotiable if you ask, and you may need to ship your own extract to the lab.

Route 3 — repitch from the previous gluten-free batch. This was the most common route in practice, and it is the cheapest. The first pitch has to be clean; after that, the population carries forward.

The conclusion drawn at the time is worth stating plainly, because it contradicts a common assumption: wet yeast is not the problem. Liquid yeast is only a gluten risk because of what it was grown on. Control the medium and the format stops mattering.

The observation: yeast run on sorghum ferments better​

This is field experience, stated as experience and not as a finding.

The reason Route 1 was preferred was not only gluten control. The working belief across production was that the yeast has time to learn how to eat sorghum rather than barley — and that after a few generations propagated on sorghum, or on whichever gluten-free grain is the target, fermentation ran more robustly.

This is not a requirement. A first-generation pitch onto sorghum wort ferments. The observation is that it tends to run better once the population has been carried on that substrate for a few generations.

A plausible mechanism, offered as hypothesis only

Serial repitching in any medium applies selection pressure: the cells that handle that particular sugar spectrum, nitrogen level, and osmotic condition best are the ones that flocculate, get harvested, and dominate the next pitch. A sorghum wort built with external enzymes differs from barley wort in exactly the ways that would matter — a glucose-forward sugar spectrum, roughly half the free amino nitrogen, and a different mineral and lipid background.

A population repeatedly selected in that environment would plausibly outperform one arriving straight from a barley-wort or molasses propagation, without requiring any genetic change at all — simple enrichment of the fittest sub-population is enough to explain it.

We have no controlled data for this, and we are not going to pretend otherwise. What we have is twelve years of commercial practice and a consistent impression. The hypothesis above is untested reasoning, included because it is falsifiable rather than because it is established.

Research wanted: does generation count change fermentation performance on sorghum wort?

This is a clean, cheap experiment. Take one strain, split it, and run one line propagated on sorghum wort and one on a conventional medium. Pitch both into identical sorghum worts across several generations, and track lag time, fermentation rate, final RDF, and sensory outcome by generation. Two or three independent reports would turn a twenty-year impression into something the field can use. Send results.

Repitching, in practice​

Repitching is the default route, so it deserves the detail:

  • The first pitch sets the chain. Every later generation inherits its gluten status from generation zero. Document the original propagation medium and keep it with the yeast records, not just the batch record.
  • Count generations and write the number down. "Generation" is a record, not a feeling. It belongs in the batch record alongside pitch rate.
  • Harvest from a clean, healthy fermentation. A stressed fermentation makes a worse pitch, and on a low-nitrogen wort stress is easier to reach — see Yeast Nutrition.
  • Set a generation limit and a reason for it. Most breweries cap repitching somewhere and start fresh. Pick a number, watch performance and sensory against it, and move the number based on your own data rather than someone else's rule.
  • A repitch cannot outrun contamination. Serial repitching carries whatever is in the slurry forward, including anything that should not be there. Micro checks matter more as generations accumulate, not less.

Where the gluten chain can still break​

Even on a disciplined programme, three places quietly reintroduce risk:

  1. The emergency pitch. A stuck or lost fermentation at 2 a.m. is exactly when someone grabs whatever yeast is on hand. That pitch is an ingredient decision, and it needs the same evidence as any other.
  2. A supplier changing its medium without telling you. The statement you got in writing was true when you got it. Re-qualify it, the way you re-qualify a grain supplier.
  3. Nutrient and processing aids added at pitch. These are ingredients too — see Nutrients and Processing Aids.

What to record​

  • Strain, supplier, and lot
  • Propagation medium of generation zero, with the supplier's statement on file
  • Generation number of this pitch
  • Pitch rate, and the cell count or slurry volume it came from
  • Harvest source — which tank, which batch
  • Micro results, if run
  • Any emergency substitution, flagged as such

Limits of this page​

The three routes and the preference between them are one company's documented practice over roughly twelve years of commercial gluten-free production. The generation observation is field experience without a controlled comparison behind it. Neither is a specification for your brewery, and the generation limit in particular is something every operation has to establish from its own data.

Source and Validation Notes

The three supply routes, the preference for single-cell propagation on the brewery's own gluten-free wort, the twelve-year span with two suppliers willing to propagate on sugar or on supplied sorghum malt extract, repitching as the most common route in practice, and the "wet yeast is not the problem" conclusion all come from a written statement by Craig Belser dated 20 November 2019, held in the Gluten Free Brewer source archive. It was written as a direct answer to a gluten-free drinker asking how a brewery could describe wet yeast as gluten-free; the correspondent is a private individual and is not identified here.

The "learn to eat sorghum rather than barley" observation is from that same statement and is Craig's commercial field experience, not a controlled comparison. The selection-pressure mechanism offered above is untested reasoning by this site, labelled as hypothesis, and should not be cited as a finding. No published study of serial repitching on sorghum wort was located in the literature surveyed for this page.

Supplier names are deliberately omitted: which laboratories were willing to propagate on a supplied medium in that period is commercially specific, likely to have changed, and not ours to publish. Current supplier positions are tracked separately on Yeast Suppliers' Gluten-Free Statements.