Coffee processing biotech

A coffee cherry contains two seeds surrounded by a sticky layer called mucilage, itself inside a parchment husk and skin. Post-harvest processing exists to remove all of that without spoiling the seed, and the way it is removed is the largest processing influence on the final cup.

Mucilage is a pectin gel

Mucilage is roughly 85 percent water, with pectin — a polymer of galacturonic acid — as its structural component, along with sugars. Pectin chains cross-link through calcium bridges and hydrogen bonding into a hydrated gel that adheres tenaciously to the parchment. It cannot simply be washed off.

Pectinases depolymerise it: polygalacturonase hydrolyses the galacturonic acid backbone, pectin lyase cleaves it by elimination, and pectin methylesterase removes methyl groups and changes how the chains interact with calcium. Once the backbone is cut, the gel loses integrity and the residue washes away.

Three processes, one variable

In washed processing, the skin and most mucilage are removed mechanically, then the coffee is held in tanks for a day or so. Naturally occurring yeasts and bacteria — on the fruit, on the equipment, in the water — secrete pectinases that finish the job, after which the coffee is washed and dried. Adding commercial pectinase shortens this from a day to a few hours and reduces water use substantially.

In natural processing the whole cherry is dried intact, and the fruit ferments slowly around the seed as it dries. In honey processing the skin is removed but some mucilage is left on during drying.

The single variable is how much sugar-rich fruit tissue is in contact with the seed, for how long, while it is metabolically active.

The part that is genuinely contested

Two accounts of fermentation’s role coexist, and the honest position is that they are not resolved.

One holds that fermentation is subtractive: it removes mucilage, nothing more, and cup differences between processes come from the drying regime and from sugars and acids diffusing from fruit into seed, not from microbial products.

The other holds it is generative: microbial metabolites — organic acids, esters, alcohols, aldehydes — penetrate the seed and become flavour precursors that survive roasting.

Evidence exists on both sides. Inoculated fermentation with defined yeast strains does produce measurable, reproducible cup differences in controlled trials, which supports the generative account. But process variables are hard to isolate, sensory panels are noisy, and a fermentation that runs too long produces recognised defects — sour, oniony and fermented off-notes from acetic and propionic acid and from over-extended microbial activity. The defect side of the argument is much better established than the benefit side, and the honest summary is that microbial contribution to positive flavour is demonstrated in specific controlled conditions and not well characterised in general practice.

Roasting is where most flavour is made

Whatever processing contributes, the large majority of coffee flavour compounds are created in roasting: Maillard reactions between reducing sugars and amino acids, Strecker degradation generating aldehydes, caramelisation, and the thermal breakdown of chlorogenic acids into quinic and caffeic acids, which drives much of the bitterness and astringency. Post-harvest processing supplies precursors; roasting builds the flavour from them.