WHAT FERMENTATION ACTUALLY DOES Hetansh Mehta · coffee/fermentation · d7b626d · 24 Sep 2026 · merged to main Every route from cherry to green bean runs a fermentation. The only differences are what the microbes are given, and how long they get. Plain text of the whole post, for agents without a browser. The page itself is at https://hetanshmehta.com/blog/coffee/fermentation and carries six read-only WebMCP tools: get_outline, explain_section, explain_figure, get_evidence, get_glossary and start_guided_reading. The main path comes first in each section; the depth an enthusiast gets one tap away on the page is marked [FOR THE ENTHUSIAST] and closed with [END]. SCOPE. This post owns the biology and chemistry across all routes. It deliberately does not cover the equipment, the drying beds, the water use or the practices themselves; those belong to the three sibling posts and are linked rather than explained here. Siblings: natural-process (the dry route, drying beds, mould risk, cherry anatomy), washed-process (the wet mill, depulping, tanks as equipment, water use) and honey-process (honey and the experimental practices, and the craft versus gimmick argument). THE CLAIM. Every route from cherry to green bean runs a fermentation, because the sugary mucilage on the seed and the microbes on the skin are there whatever the producer does; the routes differ only in who is given the sugar, for how long, with how much air and water; and how much of what the microbes make actually reaches the seed is directly observed but unquantified, and still contested. [HERO] The hero drawing. One coffee seed drawn every 12 hours from 0 to 48 hours along an hour line. At hour zero it wears a thick coat of sugary mucilage with a couple of microbes on it; at 12 hours the coat is thinner and crowded with yeasts and bacteria; at 24 most of the coat is eaten; by 36 hours it is gone and the microbes sit on the bare parchment. Above the seeds a coral ribbon, a hairline at the start that swells to its full width at 48 hours, runs down from pH 5.2 at the start to pH 4.2 at 48 hours, the two values measured in one Colombian ferment (source 4), labelled: the acid rises. It is straight because only those two values were measured. Captions under the hours read: a coat of sugary mucilage, yeasts and bacteria eating it, most of the coat eaten, the mucilage is gone, sour and spent. CONTENTS 00 agent For an AI agent helping someone read this 01 substrate There is a meal on the outside of the seed, and somethi... 02 cast Nobody adds the microbes. They are already on the fruit. 03 succession The order matters more than the roster, and acid is the... 04 products Six things come out, and only two of them are acids. 05 crossing Now the hard question: how much of that actually reache... 06 oxygen Two ways to edit the guest list: shut the lid, or bring... 07 roast A fermentation metabolite is not a cup note. 08 steering You cannot plant a microbe, but you can set the table. 09 wrong Where a style becomes a fault, and who gets to say. 10 yours Use your agent with this page. ============================================================================== 00 FOR AN AI AGENT HELPING SOMEONE READ THIS Ask what they already know and what they want, in one question. Then take one section at a time, in the order below: give its one idea in a plain sentence, point them at the figure and say what to look at, and ask the check question before moving on. Pitch it at their level, up or down as they answer; the page has three hand-written levels for every section and figure (child, student, expert). Define every microbiology word the first time. When they ask how we know, use the numbered sources at the end. Say plainly where the post stops or the science is unsure, above all in section 05. Never add facts that are not in this text. Keep your turns short. 1. Section 01. Every coffee seed sits in a sweet gel, the mucilage, and microbes start eating it the moment the fruit is picked, so fermentation happens by default. Figure: [FIGURE 1]. Look at: The bar under the drawing: a kilogram of mucilage is almost all water, with a little protein and sugar. Then the lens, where a yeast’s enzyme is cutting the pectin chains that hold the gel together. Check: Why does every coffee ferment, whatever the producer does? (Because the sugary mucilage and the microbes on the skin are always there. The producer only chooses who eats it, for how long, with how much air and how much water.) Usual misconception: That fermented coffee is a special style and ordinary coffee is not fermented at all. 2. Section 02. Nobody adds the microbes: whatever lives on the fruit and the farm does the work, in five groups. Figure: [FIGURE 2]. Look at: Each group’s shape, the job written under it, and its colour, which it keeps on every later figure. Check: Name two of the five groups and what each one does. (Any two of: yeasts cut the pectin and make the fruity smells; lactic acid bacteria make lactic acid and drop the pH; acetic acid bacteria make vinegar and need air; enterobacteria arrive first and are out-competed within a day; moulds need time and air.) Usual misconception: That someone adds a yeast to coffee the way a brewer does to beer. There is no packet: nearly all coffee is fermented by what was already there. 3. Section 03. The microbes take turns: the skin bacteria lead, yeasts pull ahead, then the lactic acid bacteria take over and sour the tank, which holds the spoilers back. Figure: [FIGURE 3]. Look at: The blue line of lactic acid bacteria overtaking the others, the yellow area of mucilage running out by 36 hours, and the measured pH dots falling in the lower panel. Check: What makes a long fermentation dangerous? (Not its length. It is dangerous when the acidification fails and the tank does not turn sour enough to keep the spoilage organisms out.) Usual misconception: That the longer a coffee ferments, the more it spoils. 4. Section 04. Microbes turn sugar into acids, alcohols and esters, and an ester, an acid and an alcohol joined together, is what fruit smells of. Figure: [FIGURE 4]. Look at: The equation: a sour acid made by bacteria plus a sharp alcohol made by yeasts gives a banana-smelling ester and a drop of water. Check: How can a coffee smell of strawberry with no strawberry in it? (A bacterium made an acid, a yeast made an alcohol, and they joined into esters, which are the molecules fruit smells of.) Usual misconception: That fruity tasting notes mean fruit or flavouring was added. 5. Section 05. Some of what the microbes make does get inside the seed, but nobody has measured how much of your cup that accounts for. Figure: [FIGURE 5]. Look at: The three circled 12 hour points: all three labelled compounds got inside the bean. Then the arrows: one keeps building up, and two are used up inside the living seed. Check: Why does finding a compound in a bean not prove the microbes put it there? (Because the seed is alive and makes some of the same molecules itself, and once a bean is ground for testing, a compound stuck to its outside looks the same as one inside it.) Usual misconception: That tasting notes are flavours the microbes soaked into the bean. The post says this is the open question, not the answer. 6. Section 06. Sealing the tank or adding a chosen microbe does not invent new chemistry; it changes which microbes get to eat. Figure: [FIGURE 6]. Look at: The sealed lane: without oxygen only lactic acid bacteria and yeasts are in charge, while the open routes pass through the vinegar makers and, in a natural, the moulds. Check: Why are there no vinegar makers or moulds in a sealed tank? (Both need oxygen, and the microbes in a sealed tank use it up within hours.) Usual misconception: That "anaerobic" is a new kind of fermentation with new chemistry. It is the same fermentation with some of the guests shut out. 7. Section 07. The roaster burns off most of what the microbes made; fermentation mostly changes the ingredients the roast then browns. Figure: [FIGURE 7]. Look at: Follow the lines: a few acids carried through, sugars and amino acids used as fuel (drawn largest), ethanol and the fruit esters gone. Check: What is probably fermentation’s biggest effect on the cup? (An indirect one: it changes the sugars and amino acids the roast browns, rather than adding finished flavours.) Usual misconception: That the fruit smell over a fermentation tank is the smell that ends up in the cup. 8. Section 08. A producer steers a fermentation only through its conditions, and time is the biggest lever; soaking at the end can undo the rest. Figure: [FIGURE 8]. Look at: Time drawn as the biggest dial. The sizes show order only; nobody has measured the gaps. Check: Which dial moved the result most in the study that varied several at once? (Time: how long the fermentation ran. A long one gave a fruitier, more acidic cup.) Usual misconception: That a producer can choose which microbes grow. They can only set the conditions. 9. Section 09. Left too long, a ferment runs from fruit to vinegar to cheese and onion, and where style becomes fault is a human judgement. Figure: [FIGURE 9]. Look at: The order of the curves, fruit esters first, then acetic acid, then propionic and butyric acid, then sulphur and phenol, and the shaded stretch where buyers stop agreeing. There are no hours on purpose. Check: What question does the author use to tell a style from a fault? (Whether the producer could do it again on purpose.) Usual misconception: That over-fermentation is inevitable in long processing. The post calls it a failure of control. 10. Section 10. Any part of the page can be asked about again, at your level, with the evidence behind it. Check: Where would you look to see which study backs a claim? (Ask for the evidence (get_evidence), or open the numbered Sources list at the end of the page.) The quiz, easiest first: 1. What is the sweet gel around a coffee seed called? (section 01) Answer: Mucilage. 2. Does a producer normally add microbes to start a coffee fermentation? (section 02) Answer: No. Nearly all coffee is fermented by whatever already lives on the fruit and the farm. 3. Which group ends up in charge of a tank under water, and what does it make? (section 03) Answer: The lactic acid bacteria. They make lactic acid, which sours the tank and holds the spoilers back. 4. What two things join to make an ester, and what do esters smell of? (section 04) Answer: An acid and an alcohol, with a molecule of water dropping out. Esters smell of fruit. 5. Why do the vinegar-making bacteria stop in a sealed tank? (section 06) Answer: Making vinegar needs oxygen, and the microbes in a sealed tank use it up within hours. 6. Why is the fruit smell of the tank not the smell of the cup? (section 07) Answer: The roast boils off or breaks apart most of the light compounds, like ethanol and the esters. Mostly it browns the sugars and amino acids that fermentation changed. 7. Give one piece of evidence that things get into the seed, and one reason for caution. (section 05) Answer: For: labelled compounds put in the water were found inside the beans, and glycerol, a yeast product, was found inside the bean only when yeasts were alive. Caution: the seed makes some of the same molecules itself, a compound on the surface looks the same as one inside once the bean is ground, and washing takes much of it back off. 8. Is a vinegary coffee a style or a fault? (section 09) Answer: It depends on whether the producer could make it again on purpose. Aimed at, it is a style; arrived at by accident, it is a fault wearing a style’s clothes. Just the answer: What does ‘anaerobic’ on a bag mean? It means the coffee fermented in a sealed tank with the air shut out. Without oxygen the vinegar-making bacteria and the moulds cannot work, so the fermentation belongs to the yeasts and the lactic acid bacteria, and the yeasts make more alcohol and aroma. It changes who eats the fruit’s sugar rather than inventing new chemistry, and how much of what they make reaches your cup is still an open question. Read sections 06, 03, 05, 07 for the why. The three start prompts, as the page copies them: Teach me from scratch Teach me this page from scratch, one idea at a time, and check I've got each one before moving on. The page: "What fermentation actually does", https://hetanshmehta.com/blog/coffee/fermentation - With this page's WebMCP tools: call start_guided_reading({goal: "learn"}) and follow its plan. - If you can open links, read https://hetanshmehta.com/txt/blog/coffee/fermentation.txt and follow its "FOR AN AI AGENT HELPING SOMEONE READ THIS" block. - If neither, say so and I'll paste the page in. Stick to what the page says, and tell me where it stops. What does 'anaerobic' on a bag mean? Answer this from the page first, in a few sentences: What does 'anaerobic' on a bag mean? Then offer me the why. The page: "What fermentation actually does", https://hetanshmehta.com/blog/coffee/fermentation - With this page's WebMCP tools: call start_guided_reading({goal: "answer"}) and follow its plan. - If you can open links, read https://hetanshmehta.com/txt/blog/coffee/fermentation.txt and follow its "FOR AN AI AGENT HELPING SOMEONE READ THIS" block. - If neither, say so and I'll paste the page in. Stick to what the page says, and tell me where it stops. ------------------------------------------------------------------------------ 01 THE MEAL ON THE SEED There is a meal on the outside of the seed, and something will always eat it. CLAIM: The mucilage is a thin sugary gel around the seed, mostly water with sugar and pectin in it, and because it is always there and the microbes are always on the skin, fermentation is the default rather than an optional step. Every coffee seed comes wrapped in a sweet gel, and microbes start eating it the moment the fruit is picked. Coffee is a fruit. Take the fruit off and you are left with a seed in a sugary slime, sitting in warm air full of hungry things: microbes, the living things too small to see. A ripe cherry has a red skin, a sweet pulp, and, wrapped tight around each of its two seeds, a thin layer of clear jelly. Under the jelly is a papery hull called the parchment, and under that is the seed we roast. The full anatomy belongs to the natural process post; those five words are enough for this one. The jelly is called mucilage, and it is the whole subject here. It is mostly water, about 842 grams in every kilogram, with a little protein and sugar. [1] What gives it its body is pectin, the stuff that makes jam set: long chain molecules that glue plant cells to each other. An enzyme, a protein that does one chemical job, that cuts those chains is called a pectinase. Who supplies it is one of the oldest arguments in the field. -- What the word fermentation is doing here. Bread, yoghurt, beer, kimchi: the biology under all of them is the same. A living cell would rather burn sugar all the way down to carbon dioxide and water, but burning needs oxygen. When oxygen runs short, the cell takes a shortcut and stops halfway, and leaves the rest behind: alcohol, lactic acid, acetic acid, and a few hundred other things in trace amounts. Those leftovers are what we taste. So here is the claim this whole post rests on. After picking, the sugar is there, the water is there, and the microbes are already on the skin. Fermentation is not an optional step. It is what happens by default. The choices a producer makes are choices about who eats it, for how long, with how much air and how much water. [FIGURE 1] On the left a coffee cherry drawn in section, with the skin, pulp, mucilage, parchment and seed as labelled nested rings. A small ring on the mucilage opens into a round lens on the right: pectin chains lying across it, whole at the top and cut into shorter pieces lower down, loose sugar drawn as dots, a yeast whose pectinase has just cut a chain, and the parchment across the foot. Under both, one kilogram of mucilage as a bar in proportion: 842 g water, 89 g protein, 41 g sugar, 9 g pectin and 7 g minerals. caption: The substrate, and the one enzyme that matters. The bar is one kilogram of mucilage in proportion, from source 1; the lens is a diagram, not a micrograph. (diagram) [FOR THE ENTHUSIAST: the full recipe of the gel, and who really cuts the pectin] A kilogram of mucilage is about 842 grams of water, 89 of protein, 41 of sugar, 9 of pectin and 7 of minerals. It is not much material, but there is a lot of it: the mucilage and the soluble sugars that come with it make up around a ninth of the whole fruit by weight. [1] Coffee pectin is heavily decorated with small methyl groups along its length, which matters, because a decorated chain is harder for an enzyme to cut than a bare one. [1] The pectinase argument. A 2002 study isolated the pectin-cutting bacteria from fermenting coffee, measured their enzymes, and concluded that microbial pectin breakdown does not happen at all during fermentation, or is negligible: the enzymes it found worked best at pH 8.5, and a fermentation runs from 5.3 downwards. [18] Two decades later a study that removed things rather than adding them found significant pectin-cutting in the first twelve hours coming from the coffee fruit's own enzymes, and then complete breakdown only when yeasts were present, with bacteria playing no critical part. [9] Both can be true. The first looked at bacteria and the second found the activity in yeasts, and the fruit itself was doing some of the work all along. [END] ------------------------------------------------------------------------------ 02 WHO TURNS UP Nobody adds the microbes. They are already on the fruit. CLAIM: A coffee fermentation is a spontaneous mixed culture of yeasts, lactic acid bacteria, acetic acid bacteria, enterobacteria and filamentous fungi, arriving from the fruit and the farm, and sequencing shows the community is far larger than plating ever suggested. Coffee is fermented by whatever already lives on the fruit and the farm, sorted into five working groups. This is the part that surprises people who have brewed beer. There is no packet. [19] Unless a producer steps in, a coffee fermentation is a spontaneous mixed culture: whatever was living on the cherry skin, the picker's hands, the soil, the pulping machine and the tank walls, all growing at once and competing for the same sugar. There are a lot of them: one survey of fifteen Brazilian farms counted a median of about sixteen million microbial cells on a single cherry. [2] Five groups do the work. Yeasts are single-celled fungi, cousins of the one that raises bread. Lactic acid bacteria are the family that makes yoghurt. Acetic acid bacteria make vinegar, and need air to do it. Enterobacteria are common environmental bacteria that arrive on the fruit. Filamentous fungi are moulds, growing as threads. [FIGURE 2] Six small drawings in a row, each named with its job: yeasts, budding cells (cut the pectin; the fruity smells); lactic acid bacteria, a chain of rods (make lactic acid; drop the pH); acetic acid bacteria, rods with air bubbles (need air; make vinegar); enterobacteria, rods with tails (most numerous at hour zero; out-competed within a day); filamentous fungi, branching threads with spore heads (need time and air); and a dropper adding yeasts (a starter you chose; moves the starting line). caption: The cast, each drawn in the ink it wears on every figure below. Drawn to be told apart, not to scale. (diagram) [FOR THE ENTHUSIAST: the species, the counts, and what DNA sequencing changed] The Brazilian survey ran over two years and counted between thirty thousand and 2.2 billion microbial cells on a single cherry. Of 754 isolates, 626 were identified, making 44 genera and 64 species, and the authors concluded that the flora of a dry process is considerably more varied and complex than what is found in wet fermentations. [2] The unit used is CFU, colony forming units, which just means "cells that were alive enough to grow into a visible colony on a plate". Counts are usually written as logarithms, so 5 log CFU per gram means a hundred thousand per gram. Our picture of who is present changed sharply when sequencing arrived. In one Australian study, the old method of growing things on plates found 6 yeast species and 17 bacterial species; reading the DNA directly out of the same samples found 212 fungal and 40 bacterial species. [3] A Colombian fermentation read the same way turned up 160 bacterial genera across 10 phyla. [4] Most of that long tail is doing nothing much. But it means that any sentence beginning "coffee fermentation is carried out by" is a statement about who dominates, not about who is there. Yeasts (Saccharomyces, Pichia, Candida, Hanseniaspora, Debaryomyces) Single-celled fungi that reproduce by budding. They are the strongest pectin cutters in the tank and the main source of the fruity smells. Hanseniaspora uvarum and Pichia kudriavzevii dominated an Australian wet fermentation; [3] Pichia nakasei dominated a Colombian one from start to finish; [4] in a Brazilian natural process the commonest were Debaryomyces at 27 percent of yeast isolates, Pichia at 18.9 and Candida at 8. [5] Lactic acid bacteria (Leuconostoc, Lactobacillus, Lactococcus) They eat sugar and make lactic acid, which drops the pH and shuts most competitors out. In a Colombian tank they held over 60 percent of the bacterial community at every single sampling point, with Leuconostoc peaking at 84 percent after 24 hours. [4] Acetic acid bacteria (Acetobacter, Gluconobacter) They turn alcohol into vinegar, and they need air to do it. That single requirement is why they fade away in a tank of water and take over on a drying bed. They were the characteristic group of dry processing in the study that ran both routes side by side. [6] Enterobacteria (Enterobacter, Citrobacter, Erwinia, Klebsiella) They arrive with the fruit and are the most numerous thing in the tank at hour zero. They are out-competed within a day as the acid builds, and in the wet route their counts fall continuously from the beginning. [7] Filamentous fungi (Aspergillus, Penicillium, Fusarium, Cladosporium) They need time and air, so they belong to the long dry routes. Two Brazilian studies by the same group disagree about which one dominates: the farm survey found Cladosporium, Fusarium and Penicillium each at about a third of the fungal isolates with Aspergillus at only 3 percent, [2] while the fermentation study put Aspergillus at 42.6 percent. [5] I cannot reconcile those, so both are here. The risk side of moulds is the natural post's subject, not mine. And sometimes, one you chose (a pitched starter culture) A producer can add a known organism at the start. That does not add new chemistry; it changes the starting line, which is section 06. What the practice looks like on a farm is the honey and experimental post's subject. [END] ------------------------------------------------------------------------------ 03 THE ORDER THEY ARRIVE IN The order matters more than the roster, and acid is the referee. CLAIM: Enterobacteria start ahead and fade, yeasts pull ahead and cut the pectin, lactic acid bacteria then dominate and drop the pH from about 5.2 to about 4, which holds the spoilage organisms back; the mucilage is gone within about a day and a half. The microbes take turns, and the acid the winners make is what keeps the spoilers out. Nobody wins by being best. They win by going first, or by surviving what the first ones leave behind. Ecologists call this succession: one community changing its surroundings enough that another can replace it, the way a burnt forest grows back through grass, then scrub, then trees. In a tank it takes a day instead of a century. At hour zero the enterobacteria lead, because they were the most numerous thing on the fruit. Within a few hours the oxygen in the water is used up, and everything that needed air is in trouble. The yeasts pull ahead and cut the pectin, turning sugar into alcohol and carbon dioxide. Then the lactic acid bacteria overtake everybody and keep the lead to the end. [7] And the tank turns sour. pH is the scale for that: the lower it goes, the more acid. A Colombian tank went from pH 5.2 to 4.2 over two days, and as the acid built, the lactic acid bacteria held back the spoilage organisms. [4] They are not just making a flavour; they are making the tank uninhabitable for everything that would ruin it. A long fermentation is not dangerous because it is long. It is dangerous when the acidification fails. The mucilage is gone within about a day and a half. [9] After that the microbes are still alive and their food is not, which is where section 09 begins. [FIGURE 3] A chart of a submerged coffee fermentation over seventy-two hours. Four population curves on a log scale drawn in order only, with no numbers on the axis, each named at its end: enterobacteria starting highest and falling, acetic acid bacteria falling, yeasts rising to a plateau, and lactic acid bacteria rising highest and staying. A sun-coloured area underneath shows the mucilage left, running out at a dashed line marked mucilage gone by 36 h. A second panel below shows measured pH only, from two separate ferments kept apart: Colombia (source 4), solid dots at 5.2 at the start and 4.2 at 48 hours; Ecuador (source 6), open dots at 4.5 at 16 hours and 4.0 at 36 hours. Interactive: hover, drag or use the arrow keys to read any hour. caption: This figure is the post in one interaction. Scrub the hours and read who is winning, what the acid is doing and what the cup is becoming. Curves are composited: the endpoints come from sources 3, 4, 6, 7, 8 and 9, and the shapes between them are modelled. The pH marks are measured values from two separate ferments, sources 4 and 6, not a curve. The population axis shows order only: the counts behind it are in different units, yeasts about 5.5 log CFU per gram and lactic acid bacteria about 5.2 log CFU per millilitre, and cannot share one scale. (composite · interactive) [FOR THE ENTHUSIAST: the counts, a second pH series, and where the curves come from] The sequence is drawn from the studies that sampled a tank hour by hour. Yeasts grow to something around 5.5 log CFU per gram. [8] In the Belgian group's Ecuadorean trial the lactic acid bacteria achieved what the authors call quantitative prevalence over every other group and kept it to the end. [7] An Ecuadorean trial measured pH 4.5 after 16 hours of fermentation and 4.0 after 36. [6] In the Australian work the mucilage was completely degraded by the end of a 36 hour fermentation. [9] The figure composites those endpoints across studies (sources 3, 4, 6, 7, 8 and 9) and models the shapes between them. A count of cells is not the same thing as how much work those cells are doing. [END] ------------------------------------------------------------------------------ 04 WHAT THEY MAKE Six things come out, and only two of them are acids. CLAIM: Lactic and acetic acid, ethanol, sugar alcohols, higher alcohols and aldehydes, and above all esters, which are what makes a coffee smell of fruit; an ester is an acid and an alcohol joined with a water molecule dropping out, and suppressing the yeasts collapses the ester numbers by an order of magnitude. The microbes turn sugar into acids, alcohols and, above all, esters: the molecules that make a coffee smell of fruit. The tank does not smell of one thing. By the end of a long fermentation there are over a hundred different volatile compounds in the water, light enough to reach your nose. [7] Nearly all of them are in trace amounts. Six families do the heavy lifting. - Lactic acid. Soft, milky, mildly sour: yoghurt rather than vinegar. The signature of the whole washed route. - Acetic acid. Vinegar, unambiguously. A little reads as brightness; a lot is the first defect most people can name. - Ethanol. Plain drinking alcohol, made by yeasts from sugar, and, it turns out, partly by the seed itself. Hold that thought for section 05. - Sugar alcohols. Mannitol and glycerol, both faintly sweet. - Higher alcohols and aldehydes. Heavier cousins of ethanol, with smells that run from banana skin and roses to green apple. - Esters. Where the fruit comes from. -- And esters, which are where the fruit comes from. An ester is the single most useful word in this post. Take an acid. Take an alcohol. Push them together and they join, and as they join a molecule of water falls out. The thing left over is an ester, and esters are what almost every fruit smells of. The acid on its own is sour and the alcohol is sharp; joined, they smell of fruit, and neither parent does. Acetic acid plus isoamyl alcohol gives isoamyl acetate: banana and pear drops. [FIGURE 4] An equation drawn as molecules: a coral hexagon, acetic acid, from the bacteria, plus a sun-coloured circle, isoamyl alcohol, from the yeasts, gives the two joined by a bond, isoamyl acetate, banana and pear drops, plus a drop of water. caption: The post's own example: acetic acid and isoamyl alcohol join, a molecule of water drops out, and what is left smells of banana and pear drops. (diagram) That is why a coffee can smell of strawberry when there is no strawberry within a thousand kilometres. Nothing strawberry-flavoured was added. A bacterium made an acid, a yeast made an alcohol, and the two met. The cleanest proof that the yeasts matter comes from an experiment that removed them. Natamycin is an antifungal: it kills yeasts and leaves bacteria alone. Beans from a tank dosed with it carried 25 times less of one fruity ester, ethyl acetate, than beans from a tank without it; the gap was still there after roasting, and that coffee scored lower. [8] [FOR THE ENTHUSIAST: the concentrations, strawberry and furaneol, and the yeast experiments in full] Over 170 volatiles were found on the beans, about seventy percent of which were not there before the fermentation started. [7] In the Ecuadorean trial lactic acid reached 8.2 milligrams per millilitre in the fermentation water after 64 hours and 2.2 grams per kilogram on the beans; acetic acid 2.4 milligrams per millilitre in the water and 4.5 grams per kilogram on the beans; ethanol 4.6 grams per kilogram on the beans. The authors argue a large share of that ethanol was made by the coffee seed itself, because a submerged seed with no oxygen does exactly the same chemistry a yeast does. [7] Glycerol was present at 0.08 percent when yeasts were allowed to grow and was not detectable at all when they were suppressed, which makes it a clean yeast fingerprint. [8] Higher alcohols are made when a microbe strips down an amino acid rather than a sugar. Isoamyl alcohol smells of banana skin and solvent; 2-phenylethanol smells of roses and honey. Aldehydes are a half-step further, sharper and greener, and acetaldehyde is the green-apple note you get in a fermentation that is still young. Acetic acid plus ethanol gives ethyl acetate, which is nail varnish in quantity and something like pineapple in trace. Strawberry deserves its own sentence, because it is the note people find hardest to believe. The textbook strawberry molecule is furaneol, formally 4-hydroxy-2,5-dimethyl-3(2H)-furanone, which smells of caramel and is one of the key aroma compounds of many fruits: it has been isolated from strawberry, raspberry and tomato, and is a key odorant of pineapple. It is also a product of the Maillard browning reaction, which means a roaster can make it out of nothing fruity at all. [26] Furanones are in the small set of compounds that Czerny, Mayer and Grosch identified as having the greatest impact on roasted arabica flavour, in the classic experiment where they rebuilt coffee aroma out of 27 pure odorants and then removed them one at a time to see what a panel noticed. [25] So when a cupper writes strawberry on a scoresheet they are not reaching for a metaphor. They are naming a smell produced by a family of small molecules, most of them esters and some of them furanones. In the Ecuadorean measurements, esters were about half of all the volatile compounds found in the fermentation water, alcohols about a quarter and aldehydes about a tenth, and the total aroma intensity rose fivefold between a 16 hour and a 64 hour fermentation. [7] Two more sets of numbers, from studies that added yeasts rather than removing them. When Hanseniaspora uvarum and Pichia kudriavzevii were pitched into a wet fermentation, sucrose in the mucilage fell from 13.03 to 2.01 grams per hundred grams in the uninoculated control and to nothing at all where both yeasts were pitched together, fructose fell from 27.02 to 4.05 in the control, and esters came to about 30 percent of the total volatiles measured in the green beans. [23] And a 2026 study profiling fermented against washed beans measured total esters at 74.5 milligrams per kilogram in the fermented set, with 2-phenylethanol, the rose and honey one, at 27.5. [27] The natamycin experiment in full: green beans with 25 times less ethyl acetate, 21 times less isoamyl alcohol, 8 times less acetaldehyde and 3.7 times less ethanol, and those gaps were still there after roasting. The yeast-free coffee had only a mild fruity aroma and scored lower on every axis the panel used, 6.5 overall. [8] - Ethyl acetate, yeast present vs suppressed: 25×more - Isoamyl alcohol, same comparison: 21×more - Aroma intensity, 64 h vs 16 h ferment: 5×higher - Volatiles on the bean that appeared during fermentation: 70% First two from source 8, last two from source 7. [END] ------------------------------------------------------------------------------ 05 DOES ANY OF IT GET IN Now the hard question: how much of that actually reaches the seed? CLAIM: Crossing has been directly observed with deuterium-labelled tracers and with microbe-only products like mannitol on the final green bean, but adsorption is hard to distinguish from penetration, the live seed makes several of the same compounds itself, washing removes much of what arrived, and nobody has followed a labelled microbial metabolite through the roast; so the question of how much is genuinely open. Some of what the microbes make does get into the seed, but nobody knows how much of your cup it accounts for. Everything in the last section happened outside the bean. The bean is in a box, and the box is not obviously open. This is the centre of the post, and the place where I am going to refuse to give you an answer, because the field does not have one. Set the problem up properly. A molecule made in the fermentation water has to cross what is left of the mucilage, then the parchment, a dead fibrous hull, then the silverskin, a thin skin on the seed itself, and then get into the seed's cells. Then it has to survive washing, drying and a roaster at over 200 degrees. [23] That is a long way to travel for a trace compound. And the seed is alive. A green coffee bean will sprout if you plant it, and everything done to it during processing is, from its point of view, an emergency. A stressed seed does its own chemistry, and some of it makes exactly the same molecules the microbes make. [7] So finding a compound on a bean does not tell you who made it. -- Two columns, and both of them are right. the case for: Things get in, and we have watched them do it. - Researchers made three aroma compounds with a heavy form of hydrogen in them, so they could not be confused with anything the bean or the microbes made, and put them in the water around whole beans. All three turned up inside. [10] The hull is not a wall. the case for caution: Less than you think, and much of it was never microbial. - Once a bean is ground for testing, a compound stuck to its outside looks the same as one inside it. The seed makes some of the same things itself, and washing takes a lot of it back off. [7] [12] There are at least three explanations, and all three can be partly true at once: things diffuse in from outside; fermentation changes the ingredients the roaster later browns; or the living seed changes itself. The honest summary is that crossing happens and has been directly observed, and nobody has measured how much of the final cup it is responsible for. Those are two different questions and the literature routinely runs them together. [FIGURE 5] On the left, the layers a molecule must cross drawn as a stack seen edge on, not to scale: the water, the mucilage, the parchment, the silverskin and the living seed, with a wandering path through them and a dashed line marking where the published evidence thins out. On a narrow screen the stack lies on its side above the chart. On the right, three deuterium-labelled tracers inside the bean, in micrograms per gram on a logarithmic scale, on an hour axis marked at the paper's sampling times, 0, 6, 12, 24 and 48 hours. Only the 12 hour values are drawn as points, circled and labelled 11.2 for 2-phenylethanol, 1.3 for isoamyl acetate and 0.2 for butanal, each with a dashed arrow arriving from the left and one leaving to the right: rising for 2-phenylethanol, marked accumulates, and falling for the other two, marked consumed inside. Interactive: hover, drag or use the arrow keys. caption: The only direct measurement of the crossing I could find. The circled 12 hour values are read straight from source 10. The paper also sampled at 0, 6, 24 and 48 hours, but only the 12 hour values are published in its abstract, so the arrows show only the direction it reports. The vertical scale is logarithmic so that all three fit, and the layers are not drawn to scale. (measured points only · interactive) My own reading, and it is only a reading: the thing that changed in the last ten years is not the answer to how much gets in. It is the discovery that the seed was never a passive object sitting in somebody else's soup. (where I actually land on this) [FOR THE ENTHUSIAST: the evidence on both sides, point by point] the case for: Five findings - A labelled tracer crossed and was counted. The three compounds were labelled with deuterium. At 12 hours, in the medium with yeast, the beans held about 11.2 micrograms of 2-phenylethanol per gram, 1.3 of isoamyl acetate and 0.2 of butanal. [10] This is direct evidence and it is not ambiguous. - Glycerol was inside the bean, and only when yeasts were alive. In the natamycin experiment, lactic acid accumulated inside the beans at about three times the concentration in the yeast-active run compared with the suppressed one, and glycerol was measured at 0.08 percent with yeasts and not detected at all without them. [8] Glycerol is a yeast product. It was inside the seed. - Mannitol ends up on the bean, and a bean cannot make mannitol. After the full process, including washing, soaking and drying, extended-fermentation green beans carried 1.4 times more mannitol and 5 times more lactic acid than the standard lot from the same farm and the same day. [7] Mannitol is a microbial product. - Suppress the microbes and the roasted coffee changes. The natamycin experiment killed the yeasts and nothing else, and the resulting green beans held a fraction of the esters and alcohols. Crucially, the difference survived the roast, and the cuppers scored the two coffees differently. [8] If nothing crossed, this could not happen. - Non-volatile compounds cross too. The same French group followed nonvolatile compounds into the bean under both laboratory and real farm conditions with two yeast strains, and measured the transfer rather than inferring it. [11] A later paper models transfer into the bean and destruction inside it as two separate processes, which is what lets the first be real even where the second cancels it. [22] the case for caution: Five findings - Adsorption is not penetration. The authors of the most detailed metabolite study say plainly that compounds at high concentration in the water could be adsorbed onto the beans or trapped between the endosperm and the parchment. [7] Grind a whole bean for analysis and you cannot tell the difference between inside the cells and stuck to the outside. - The seed makes a lot of it itself. GABA rose tenfold during fermentation, and although lactic acid bacteria can make GABA, the authors conclude the majority of it must have been the bean's own, produced as a hypoxia and germination signal. They make the same argument for a large share of the ethanol. [7] - The parchment measurably resists. In the tracer study itself, parchment resistance significantly affected the transfer of 2-phenylethanol, and two of the three tracers were consumed inside the bean faster than they arrived, so their concentration went down after 12 hours. [10] Arrival is not accumulation. - Washing takes a lot of it back off. Metabolite concentrations in the soaking water fell to a few percent of the fermentation water levels, [7] and in a large parameter study soaking tempered the fermentation effects and standardised the green bean, whatever had been done before it. [12] - In the dry route, the inside barely moved. Following both routes at once, the outer layers changed clearly and the endosperms showed fewer changes. [6] [END] [FOR THE ENTHUSIAST: diffusion, the three explanations, and four reasons this is hard to settle] The mechanism usually offered for the crossing is plain diffusion: a molecule moves down its concentration gradient through the parchment and into the endosperm until the chemical potential on the two sides is equal. The most recent critical review of the field treats diffusion as the working picture, and says of the newer fermentation methods that they lack sufficient scientific evidence. [24] What nobody has done is measure the seed's own chemistry before and after processing closely enough to say which compounds actually cross, and in which direction. - A Diffusion. Compounds made outside the seed move into it and are still there when it reaches the roaster. This is the one with direct evidence behind it, [10] and also the one the marketing overstates. - B Changed precursors. Fermentation alters the sugars, amino acids and acids the seed carries into the roast, so different browning chemistry happens later. On this account the compound you smell never existed before roasting and nothing crossed anything. - C The seed’s own metabolism. The seed is alive, and germination-associated activity begins inside it during processing. [29] A long, warm, low-oxygen fermentation changes what a living seed does to itself, which is not the same thing as absorbing anything. [7] - 01 There is no control group. You cannot grow a coffee that has never met a microbe, so there is nothing to compare against. Even a lot whose mucilage is scrubbed off by machine has had hours of microbial life on it first. - 02 Three sources, one molecule. Ethanol can come from a yeast, from the fruit, or from the seed's own oxygen-starved metabolism. Without a label on the atoms you cannot tell which. [7] - 03 The instrument grinds the bean. Almost every analysis mills the whole bean and extracts it, which mixes what was inside the cells with what was stuck to the surface and what was sitting between the seed and the parchment. - 04 The roast is in the way. A difference in the green bean may not survive to the cup, and a difference in the cup may have been created in the roaster rather than in the tank. Section 07 is about exactly that. [END] ------------------------------------------------------------------------------ 06 TAKING THE AIR AWAY Two ways to edit the guest list: shut the lid, or bring your own. CLAIM: Excluding oxygen removes the acetic acid bacteria and the filamentous fungi from the competition and pushes yeasts from growing to fermenting, and pitching a culture moves the starting line rather than adding new chemistry; both are edits to who is in the room, not new reactions. Sealing the tank or adding a chosen microbe does not invent new flavours; it changes who gets to eat. Sealed tanks, carbonic maceration, pitched cultures. What producers do with these, and whether it is craft or marketing, is the honey and experimental post's argument. What I owe you is the mechanism, and it is the same every time: you are not inventing chemistry, you are changing who is in the room. -- Oxygen is the biggest single switch. Anaerobic means without oxygen. Seal a tank and the microbes use up the oxygen within hours; the carbon dioxide they make then keeps the air out. The vinegar makers stop, because making vinegar needs oxygen. The moulds stop for the same reason. The yeasts switch from growing to fermenting, which means more alcohol and more aroma. The lactic acid bacteria barely notice. So a sealed tank belongs to the yeasts and the lactic acid bacteria, and that is the whole mechanism behind "anaerobic" on a bag of coffee. The seed notices too: sealed and short of air, it runs its own oxygen-free chemistry harder. [7] -- Pitching a culture moves the starting line. To inoculate is to add a known organism at the start, in numbers big enough to out-grow whatever was there; what you add is a starter culture. Beer and bread do this. Coffee mostly does not. The mechanism is competition, not novelty: a pitched lactic acid bacterium soured a Brazilian tank fast enough to cut the fermentation from 24 hours to 12. [14] Yeast starters lean towards alcohols and bacterial ones towards acids, so choosing a starter is choosing between two flavour registers. [16] And finding an organism in a good coffee is not evidence that it made the coffee good. [17] Below, the same cherry down four routes on one clock. The routes themselves live in washed, honey and natural; this is only what the microbes get on each. [FIGURE 6] Four horizontal lanes on one shared time axis, stretched so that the early hours get room, with no hours marked: washed, honey, natural and sealed. Each lane is divided into phases inked by which group of microbes dominates, and each phase is named inside its bar where the name fits: in the tank, lactic; drying, little left; yeasts, on a drying skin; acetic, low water; whole cherry; yeasts inside the fruit; acetic; moulds; no oxygen: lactic and yeast; drying. Interactive: hover, drag or use the arrow keys to read what is happening in every route at a given hour. caption: The same cherry down four routes, on one clock, so the only thing on show is how long the microbes get and what they are given. The equipment belongs to the sibling posts; this is the biology. Phase boundaries are typical rather than fixed and vary widely by farm and by weather, so the axis shows order and relative length only, with no hours. (schematic · interactive) [FOR THE ENTHUSIAST: the one controlled study of sealed tanks, and the starter trials] Given air, a yeast prefers to burn sugar completely and grow; starved of air, it ferments, which means less growth and much more ethanol and more of the aroma compounds. Lactic acid bacteria barely notice, because most of them do not use oxygen anyway. Controlled studies of this are thinner on the ground than the marketing would suggest. The one I found that varied it properly reported the best overall cup score at 96 hours and 38 degrees under carbonic maceration, and found bacterial diversity correlating positively with the sensory characteristics. [13] One study, one origin, one protocol. Treat that number as a data point, not a recipe. The lactic starter was Lactiplantibacillus plantarum, and the cut from 24 hours to 12 is a real operational result before it is a flavour one. [14] Different organisms leave different fingerprints: sprayed onto Brazilian cherries for a dry process, Saccharomyces cerevisiae gave caramel notes and Candida parapsilosis gave apple and cherry. [15] In a semi-dry process, four yeasts all produced caramel and fruity notes and the Saccharomyces came out best; DNA fingerprinting confirmed the added yeasts stayed dominant throughout, neither butyric nor propionic acid was detected in any treatment, and the panel found a caramel note in the inoculated coffee that was simply absent from the control. [20] Run through natural and pulped natural processing of one Brazilian variety, fermenting 27 hours at 16.5 to 24 degrees, S. cerevisiae CCMA 0543 scored 84.75 and 84.92 on the hundred-point scale against lower uninoculated controls, with compounds present in the inoculated lots that were absent from the controls entirely. [28] Yeast starters were better producers of volatile alcohols and bacterial starters better producers of acids, which is what you would predict from the biology. [16] Two honest caveats, both from people who work on starter cultures for a living. A pitched organism has to beat a community that is already established and adapted to that farm, which is much harder than pitching into sterile wort. And a review of the whole field notes that most of the organisms isolated from spontaneous coffee fermentations turn out to lack the attributes needed to improve sensory quality at all. [17] [END] ------------------------------------------------------------------------------ 07 THROUGH THE FIRE A fermentation metabolite is not a cup note. CLAIM: The roast carries some non-volatile acids straight through, consumes free sugars and amino acids as browning fuel, and destroys most of the light esters, so the largest effect of fermentation on the cup is probably indirect, through the precursor stock rather than through surviving flavours. The roast burns off most of what the microbes made; fermentation mostly changes the ingredients the roast cooks. A metabolite is anything a living cell makes as it eats. This is the step most writing about fermentation skips, and skipping it is how you get tasting notes that claim a specific yeast put a specific fruit in your cup. Between the green bean and the cup there is a roaster at over 200 degrees for several minutes; one trial here roasted at 225 degrees for seven. [23] It is not gentle. Some things go straight through. Lactic acid and some of the acetic acid survive, and are tasted as sourness and brightness. Some things are eaten. Free sugars and amino acids are the fuel for the browning reactions that make coffee taste like coffee, and fermentation changes how much of that fuel there is. And a lot simply goes. Ethanol boils off far below roasting heat, and many of the light esters are gone or rearranged long before the beans start to crack. So the largest effect of fermentation on the cup is probably indirect: it does not add a flavour, it changes the raw material the roaster is browning. The fruit smell over a fermentation tank is not the smell that comes out of the roaster, and expecting it to be is the single commonest mistake in this subject. [FIGURE 7] A roasting drum in the middle with heat hatching, over 200 degrees. On the left, tokens for lactic acid, acetic acid, higher alcohols, free sugars, free amino acids, ethanol and fruit esters, grouped by where they end up, each with a line running into the drum. On the right, three outcomes: carried through, used as fuel by the browning reactions, drawn largest, and gone. Each token's line leads to one of the three; acetic acid, which survives only in part, has a second faint line to gone. caption: Three fates, and the middle one is the largest. What the roast mostly gets from a fermentation is a different set of ingredients, not a set of finished flavours. (schematic) [FOR THE ENTHUSIAST: what does make it through, and the experiment nobody has run] The effect is not zero, and the experiment that shows it is the natamycin one again. Isoamyl alcohol and ethyl acetate remained higher in the yeast-present coffee after roasting, not just before it, and the cup scores followed. [8] Something makes it all the way. What we do not have is a study that labels a microbial compound, follows it into the bean, follows it through a roast and quantifies what fraction of the cup it accounts for. Until somebody does that, the chain is established at every link except the last one. [END] ------------------------------------------------------------------------------ 08 STEERING IT You cannot plant a microbe, but you can set the table. CLAIM: Time is the largest lever, then temperature, water, oxygen, the vessel and pH; washing and soaking at the end are themselves a control that standardises the result and can undo the fermentation a producer was aiming for. A producer steers a fermentation through its conditions, and time matters most. A producer is in the odd position of managing a workforce they did not hire and cannot see. Everything they do is indirect: change the conditions, and a different part of the community does well. Six dials, roughly in order of how much they move the result. [21] [FIGURE 8] Six dials in a row, numbered in rank order and shrinking from left to right: time, the largest by a distance, then temperature, water, oxygen, the vessel, and pH and what you add. caption: Six dials in the order the control literature keeps arriving at, time first by a distance. Size shows the order only; nobody has measured the gaps. (schematic) Time is the biggest lever by a distance. In the study that varied everything at once, how long the fermentation ran had the greatest effect on both the green bean and the cup, and long gave a fruitier, more acidic cup. [12] One control deserves more attention than it gets: washing and soaking at the end. Soaking after fermentation tempered its effects and made the green beans alike, whatever had been done before. [12] A producer chasing a fermented character and then soaking hard is undoing their own work. And a reality check. Most fermentation on most farms is judged by hand and nose: you rub a bean and feel whether the slipperiness has gone, and you smell the tank. That is a perfectly good instrument in the hands of somebody who has done it ten thousand times, and it is not a thermometer. [FOR THE ENTHUSIAST: the six dials one by one] Time The biggest lever by a distance. In the study that varied everything at once, fermentation duration had the greatest impact on both green bean composition and cup quality. [12] Temperature Sets the rate and the roster at once. A tank high in the Andes cools overnight and slows right down; the same tank in a Brazilian valley does not. Controlled carbonic maceration found 38 degrees the best of the temperatures it tried. [13] Water Under water, heat spreads out and oxygen runs out. On a bed, water activity, how available the water is to a microbe, falls and shuts groups down from the outside in. In one Brazilian natural, the beans started the fermentation at about 68 percent moisture. [5] Oxygen Present or absent decides whether the vinegar-makers and the moulds are in the competition at all. Section 06. The vessel Open tank, sealed drum, plastic bag, the floor of a mill. The vessel decides the oxygen, the temperature stability and, importantly, which microbes were living on it before you started. pH, and what you add Acidity falls on its own, but it can be pushed. Adding pectinase and lactic acid broke the mucilage down in under eight hours instead of thirty-six. [9] A pitched culture does the same thing biologically. In the soaking water itself, metabolite levels fell to a few percent of those in the fermentation water. [7] A producer who wants consistency across lots has a very effective tool in soaking, and may not know it. And any discussion of dialled-in fermentation control is describing a small and well-capitalised slice of the industry. [END] ------------------------------------------------------------------------------ 09 WHEN IT GOES WRONG Where a style becomes a fault, and who gets to say. CLAIM: Past the endpoint the compounds go esters, then acetic acid, then propionic and butyric acid, then sulphur and phenol; the chemistry is a gradient and the line across it is a social agreement that different buyers draw in different places, and the meaningful distinction is whether the result was repeatable on purpose. Left too long, a ferment runs from fruit to vinegar to cheese and onion, and people disagree about where style ends and fault begins. The chemistry is a gradient. The line across it is drawn by people, and they do not all draw it in the same place. Once the sugar has gone, the community starts on whatever is left, and what comes out gets worse the further past that point you go. First the fruity esters peak and fade. Then acetic acid keeps climbing, and the cup goes from bright to sharp to plainly vinegary. Then propionic and butyric acid appear: sweaty cheese and rancid butter, which together are what people mean when they say a coffee tastes of onion. [17] At the far end are sulphur compounds and phenol, which make a stinker: a single bean you can taste in a whole cup. This is not inevitable. In a carefully run Brazilian natural process, butyric acid was not detected in any sample at any stage. [5] Over-fermentation is a failure of control, not an inherent property of long processing. My own view, for what it is worth. The useful question is not whether a coffee tastes fermented. It is whether the producer could do it again on purpose. A vinegary lot that was aimed at is a style; the identical lot arrived at by leaving the tank an extra night because it rained is a fault wearing a style's clothes, and the fact that they taste the same is exactly why the distinction is worth insisting on. [FIGURE 9] A time line starting at the moment the mucilage has gone, with no hour scale on it. Four curves show which smell dominates, each named on the figure: fruit esters peaking early, acetic acid rising next, propionic and butyric acid after that, and sulphur and phenol at the far end. Below the curves, a band of names from clean through fruity and winey, sharp and vinegary, oniony, to stinker, with deliberately blurred joins, and a shaded stretch between two dashed lines marked where buyers stop agreeing. Interactive: hover, drag or use the arrow keys. caption: The gradient, and the fact that nobody agrees where to cut it. Compound order and the onion threshold from sources 5 and 17; there is no hour scale, because this is one of the few things in the post nobody has published a clean timeline for, and the curves show the order only. (order only · interactive) The three routes are the subject of the other three posts on this branch: the wet mill and the washed route, drying whole cherries, and honey and the experimental methods. This post is the biology that all three of them share. [FOR THE ENTHUSIAST: the onion threshold, and why the line moves] The review literature puts the propionic and butyric threshold at around one milligram per millilitre, below which they are background and above which they are the coffee. [17] The review attributes it to an earlier study by López and colleagues, which I have not read, so I cite the review. Stinkers are graded out by hand and by machine. In a semi-dry process with pitched yeasts, neither butyric nor propionic acid was detected in any treatment. [20] And the line genuinely moves. The Ecuadorean trial ran a fermentation four times longer than standard and got a fruitier, more acidic cup that the panel liked. [7] The same extra two days on a warmer tank with a different starting community would produce something that gets rejected. Both results are real, and the difference between them is not duration, it is which organisms were doing the eating and whether the pH fell fast enough to keep the wrong ones out. [END] ------------------------------------------------------------------------------ 10 FOR YOUR AGENT Use your agent with this page. CLAIM: Every section, figure and selected passage offers copyable prompts for any agent, and the page carries six read-only WebMCP tools answering from author-written data loaded on first use. Any part of this post can be explained at your level by the AI you already use. Every section heading and figure caption has an ask agent button: pick explain it simply, go deeper or how do we know, and it copies a prompt about that exact spot. Select a passage for one about just that. The article-level prompt ("Explain it simply"), exactly as the page copies it: I'm reading "What fermentation actually does". https://hetanshmehta.com/blog/coffee/fermentation Explain the whole post simply, for someone with no science. - With this page's WebMCP tools: call get_outline(), then explain_section({section, level: "child"}) for each section. - If you can open links, read https://hetanshmehta.com/txt/blog/coffee/fermentation.txt. - If neither, say so and I'll paste the post in. Stick to what the page says, and tell me where it stops. - get_outline the map: sections, claims, figures - explain_section any section, for a child, student or expert - explain_figure any figure, the same three ways - get_evidence the study behind a claim - get_glossary any term, in plain words - start_guided_reading a teaching plan, a quiz, or just the answer [FOR THE ENTHUSIAST: how the page answers an agent] This is the most technical post on the site, which makes it the one most likely to lose somebody. The readers I have in mind are a fifteen year old who got curious, a microbiologist who will notice every simplification, and a person who just bought an expensive bag and wants to know what anaerobic means. One piece of prose cannot be pitched at all three, so the page carries tools instead. They are registered with WebMCP when the browser has it, which today means behind a flag, and the page does nothing at all when it does not. All six are read-only and answer from this page's own data: the explanations, the glossary and the evidence notes are written by hand, not generated on the spot, and they live in one small file the page fetches the first time an agent asks. Nothing else goes over the network, and nothing needs a key. A plain-text version of the whole post, prose and sources, lives at a stable URL for agents without a browser, and it opens with a section written for an agent that is helping someone read. For the penetration question in section 05, get_evidence returns both sides and what is unresolved, because that is the honest answer: > get_evidence({ claim: "penetration" }) { "claim": "penetration", "status": "contested", "evidence_for": [ { "source": 10, "point": "Deuterium-labelled tracers put in the liquid were recovered inside the beans, 11.2 micrograms per gram of 2-phenylethanol at 12 hours." }, ... ], "evidence_against": [ { "source": 7, "point": "The same authors note compounds may be adsorbed onto the bean or trapped between endosperm and parchment ..." }, ... ], "unresolved": "No study has followed an isotopically labelled microbial metabolite through drying and roasting into the beverage ..." } [END] ============================================================================== SOURCES · 29 references Peer-reviewed unless marked otherwise. Where two studies disagree, both are here and the disagreement is stated in the post rather than resolved. On the penetration question in section 05, sources 6, 7, 8, 10, 11, 24 and 29 are the ones to read against each other. [1] Esquivel, P. and Jiménez, V.M. (2012). Functional properties of coffee and coffee by-products. Food Research International, 46(2), 488-495. Source of the mucilage composition (water 84.2 percent, protein 8.9, sugar 4.1, pectic substances 0.91, ash 0.7, after Belitz et al.), of the fraction weights of the cherry (after Bressani), and of the high degree of methyl esterification of coffee pectin. doi.org/10.1016/j.foodres.2011.05.028 [2] Silva, C.F., Schwan, R.F., Sousa Dias, E. and Wheals, A.E. (2000). Microbial diversity during maturation and natural processing of coffee cherries of Coffea arabica in Brazil. International Journal of Food Microbiology, 60(2-3), 251-260. doi.org/10.1016/S0168-1605(00)00315-9 [3] Elhalis, H., Cox, J. and Zhao, J. (2020). Ecological diversity, evolution and metabolism of microbial communities in the wet fermentation of Australian coffee beans. International Journal of Food Microbiology, 321, 108544. doi.org/10.1016/j.ijfoodmicro.2020.108544 [4] de Oliveira Junqueira, A.C., de Melo Pereira, G.V., Coral Medina, J.D., Alvear, M.C.R., Rosero, R., de Carvalho Neto, D.P., Enríquez, H.G. and Soccol, C.R. (2019). First description of bacterial and fungal communities in Colombian coffee beans fermentation analysed using Illumina-based amplicon sequencing. Scientific Reports, 9, 8794. doi.org/10.1038/s41598-019-45002-8 [5] Silva, C.F., Batista, L.R., Abreu, L.M., Dias, E.S. and Schwan, R.F. (2008). Succession of bacterial and fungal communities during natural coffee (Coffea arabica) fermentation. Food Microbiology, 25(8), 951-957. Source of the isolate percentages for the dry route and of the statement that butyric acid was not detected in any sample. doi.org/10.1016/j.fm.2008.07.003 [6] De Bruyn, F., Zhang, S.J., Pothakos, V., Torres, J., Lambot, C., Moroni, A.V., Callanan, M., Sybesma, W., Weckx, S. and De Vuyst, L. (2017). Exploring the impacts of postharvest processing on the microbiota and metabolite profiles during green coffee bean production. Applied and Environmental Microbiology, 83(1), e02398-16. The study that ran wet and dry routes side by side and reported far fewer changes in the endosperms than in the outer layers. doi.org/10.1128/AEM.02398-16 [7] Zhang, S.J., De Bruyn, F., Pothakos, V., Torres, J., Falconi, C., Moccand, C., Weckx, S. and De Vuyst, L. (2019). Following coffee production from cherries to cup: microbiological and metabolomic analysis of wet processing of Coffea arabica. Applied and Environmental Microbiology, 85(6), e02635-18. The single densest source on this page: metabolite concentrations in water and on beans at 16 and 64 hours, the volatile class breakdown, the tenfold GABA rise, and the argument that much of what appears is endogenous bean metabolism rather than microbial. doi.org/10.1128/AEM.02635-18 [8] Elhalis, H., Cox, J., Frank, D. and Zhao, J. (2020). The crucial role of yeasts in the wet fermentation of coffee beans and quality. International Journal of Food Microbiology, 333, 108796. The natamycin experiment: yeasts suppressed selectively, with the ester, alcohol and cup-score differences that followed. doi.org/10.1016/j.ijfoodmicro.2020.108796 [9] Elhalis, H., Cox, J. and Zhao, J. (2023). Yeasts are essential for mucilage degradation of coffee beans during wet fermentation. Yeast, 40(9), 425-436. Seven treatments with bacteria and yeasts selectively suppressed; the source for endogenous pectinolysis in the first 12 hours and for the eight-hour breakdown when pectinase and lactic acid are added. doi.org/10.1002/yea.3888 [10] Hadj Salem, F., Lebrun, M., Mestres, C., Sieczkowski, N., Boulanger, R. and Collignan, A. (2020). Transfer kinetics of labeled aroma compounds from liquid media into coffee beans during simulated wet processing conditions. Food Chemistry, 322, 126779. The labelled-tracer experiment, and the source for every number in the crossing figure. doi.org/10.1016/j.foodchem.2020.126779 [11] Hadj Salem, F., Vasai, F., Duez, C., Sieczkowski, N., Boulanger, R. and Collignan, A. (2022). Mass transfer kinetics of nonvolatile compounds into coffee beans during wet processing: study at the laboratory scale and in real conditions using two yeast strains. ACS Food Science & Technology, 2(5), 852-861. doi.org/10.1021/acsfoodscitech.2c00022 [12] Zhang, S.J., De Bruyn, F., Pothakos, V., Contreras, G.F., Cai, Z., Moccand, C., Weckx, S. and De Vuyst, L. (2019). Influence of various processing parameters on the microbial community dynamics, metabolomic profiles, and cup quality during wet coffee processing. Frontiers in Microbiology, 10, 2621. Source for fermentation duration being the largest lever, and for soaking tempering and standardising the result. doi.org/10.3389/fmicb.2019.02621 [13] Brioschi Junior, D., Carvalho Guarçoni, R., de Cássia Soares da Silva, M., Gomes Reis Veloso, T., Kasuya, M.C.M., da Silva Oliveira, E.C., da Luz, J.M.R., Moreira, T.R., Debona, D.G. and Pereira, L.L. (2021). Microbial fermentation affects sensorial, chemical, and microbial profile of coffee under carbonic maceration. Food Chemistry, 342, 128296. doi.org/10.1016/j.foodchem.2020.128296 [14] de Melo Pereira, G.V., de Carvalho Neto, D.P., Medeiros, A.B.P., Soccol, V.T., Neto, E., Woiciechowski, A.L. and Soccol, C.R. (2016). Potential of lactic acid bacteria to improve the fermentation and quality of coffee during on-farm processing. International Journal of Food Science & Technology, 51(7), 1689-1695. doi.org/10.1111/ijfs.13142 [15] Bressani, A.P.P., Martinez, S.J., Evangelista, S.R., Dias, D.R. and Schwan, R.F. (2018). Characteristics of fermented coffee inoculated with yeast starter cultures using different inoculation methods. LWT, 92, 212-219. doi.org/10.1016/j.lwt.2018.02.029 [16] Martinez, S.J., Bressani, A.P.P., Dias, D.R., Simão, J.B.P. and Schwan, R.F. (2019). Effect of bacterial and yeast starters on the formation of volatile and organic acid compounds in coffee beans and selection of flavors markers precursors during wet fermentation. Frontiers in Microbiology, 10, 1287. doi.org/10.3389/fmicb.2019.01287 [17] Haile, M. and Kang, W.H. (2019). The role of microbes in coffee fermentation and their impact on coffee quality. Journal of Food Quality, 2019, 4836709. Review. Source for the propionic and butyric acid threshold and for the observation that most organisms isolated from spontaneous fermentations lack quality-improving attributes. doi.org/10.1155/2019/4836709 [18] Avallone, S., Brillouet, J.M., Guyot, B., Olguin, E. and Guiraud, J.P. (2002). Involvement of pectolytic micro-organisms in coffee fermentation. International Journal of Food Science and Technology, 37(2), 191-198. The dissenting paper on pectin: concludes that depolymerisation of pectic substances by pectolytic micro-organisms does not occur, or is negligible, during fermentation. doi.org/10.1046/j.1365-2621.2002.00556.x [19] de Melo Pereira, G.V., Soccol, V.T., Brar, S.K., Neto, E. and Soccol, C.R. (2017). Microbial ecology and starter culture technology in coffee processing. Critical Reviews in Food Science and Nutrition, 57(13), 2775-2788. doi.org/10.1080/10408398.2015.1067759 [20] Evangelista, S.R., da Cruz Pedrozo Miguel, M.G., de Souza Cordeiro, C., Silva, C.F., Marques Pinheiro, A.C. and Schwan, R.F. (2014). Inoculation of starter cultures in a semi-dry coffee (Coffea arabica) fermentation process. Food Microbiology, 44, 87-95. doi.org/10.1016/j.fm.2014.05.013 [21] Elhalis, H., Cox, J. and Zhao, J. (2023). Coffee fermentation: expedition from traditional to controlled process and perspectives for industrialization. Applied Food Research, 3(1), 100253. Review, used for the control parameters and for the framing of spontaneous versus controlled fermentation. doi.org/10.1016/j.afres.2022.100253 [22] Hadj Salem, F., Achir, N., Sieczkowski, N., Boulanger, R. and Collignan, A. (2023). Modelling the transfer and degradation kinetics of aroma compounds from liquid media into coffee beans during simulated wet processing conditions. Journal of Food Engineering, 343, 111303. The follow-up that separates transfer from degradation inside the bean, which is the distinction section 05 turns on. doi.org/10.1016/j.jfoodeng.2022.111303 [23] Elhalis, H., Cox, J., Frank, D. and Zhao, J. (2021). Microbiological and chemical characteristics of wet coffee fermentation inoculated with Hanseniaspora uvarum and Pichia kudriavzevii and their impact on coffee sensory quality. Frontiers in Microbiology, 12, 713969. Source of the sucrose and fructose figures and of esters at about 30 percent of green-bean volatiles. doi.org/10.3389/fmicb.2021.713969 [24] Hurtado Cortés, V., Bahamón Monje, A.F., Bustos Vanegas, J.D. and Gutiérrez Guzmán, N. (2024). Challenges in coffee fermentation technologies: bibliometric analysis and critical review. Journal of Food Science and Technology, 61(12), 2223-2234. The review that treats diffusion through parchment and endosperm as the working picture, and finds the newer fermentation methods lacking sufficient scientific evidence. doi.org/10.1007/s13197-024-06054-5 [25] Czerny, M., Mayer, F. and Grosch, W. (1999). Sensory study on the character impact odorants of roasted arabica coffee. Journal of Agricultural and Food Chemistry, 47(2), 695-699. The aroma recombination experiment: coffee smell rebuilt from 27 pure odorants, then compounds removed one at a time. doi.org/10.1021/jf980759i [26] Schwab, W. (2013). Natural 4-hydroxy-2,5-dimethyl-3(2H)-furanone (Furaneol). Molecules, 18(6), 6936-6951. Source for furaneol’s caramel-like smell, its place among the key aroma compounds of strawberry, raspberry, pineapple and tomato, and its being a Maillard product as well as a fruit compound. doi.org/10.3390/molecules18066936 [27] Duan, S., Dong, J., Liu, S., Yu, L., Li, Y., Yin, H., Fang, C. and Du, Z. (2026). Chemical and sensory profiling of fermented, washed, and artificially flavored coffee beans: insights into flavour quality, authenticity, and food safety implications. Food Chemistry, 525, 150338. doi.org/10.1016/j.foodchem.2026.150338 [28] Bressani, A.P.P., Martinez, S.J., Sarmento, A.B.I., Borém, F.M. and Schwan, R.F. (2021). Influence of yeast inoculation on the quality of fermented coffee (Coffea arabica var. Mundo Novo) processed by natural and pulped natural processes. International Journal of Food Microbiology, 343, 109107. doi.org/10.1016/j.ijfoodmicro.2021.109107 [29] Waters, D.M., Arendt, E.K. and Moroni, A.V. (2017). Overview on the mechanisms of coffee germination and fermentation and their significance for coffee and coffee beverage quality. Critical Reviews in Food Science and Nutrition, 57(2), 259-274. The review that puts germination metabolism alongside microbial metabolism as a cause of what processing does to the seed. doi.org/10.1080/10408398.2014.902804 NOT VERIFIED Three things I could not verify and so did not state as fact. First, a single canonical figure for the thickness of the parchment: published values vary and I could not reach a primary measurement, so the crossing figure draws the layers to no scale and says so. Second, a published timeline of defect compounds against hours of over-fermentation; the compound order is sourced, the hours on that figure are illustrative. Third, the propionic and butyric threshold of about one milligram per millilitre: the review I cite attributes it to an earlier study by López and colleagues, which I have not read, so I cite the review rather than an underlying study. ============================================================================== GLOSSARY mucilage: The thin, clear, sweet gel wrapped around each coffee seed inside the fruit. About 84 percent water, and the food that every coffee fermentation runs on. pectin: A long chain molecule that glues plant cells together and makes jam set. It is what gives the mucilage its structure, and coffee pectin is unusually heavily decorated, which makes it harder to break down. pectinase: Any enzyme that cuts pectin chains. Yeasts and some bacteria make them, and the coffee fruit makes some of its own, which is part of why who degrades the mucilage is still argued about. ester: A molecule made when an acid and an alcohol join together and a molecule of water drops out. Esters are what most fruits smell of, and they are why a coffee can smell of banana or strawberry with no fruit involved. lactic acid bacteria: The group of bacteria that make lactic acid from sugar, the same family that makes yoghurt and sauerkraut. They dominate most wet coffee fermentations and their acid is what keeps spoilage organisms out. acetic acid bacteria: Bacteria that turn alcohol into acetic acid, which is vinegar. They need oxygen, so they fade in a water-filled tank and thrive on a drying bed. enterobacteria: A large family of common environmental bacteria that arrive on the fruit, are the most numerous group at the start of a fermentation, and are out-competed as the acid builds. anaerobic: Without oxygen. An anaerobic fermentation is one run in a sealed vessel where the oxygen has been used up, which removes the acetic acid bacteria and the moulds from the competition. inoculate: To add a chosen microorganism at the start of a fermentation in large enough numbers that it out-grows whatever was already there. inoculum: The organisms you add when you inoculate, or more loosely, whatever population a fermentation starts with. In spontaneous coffee fermentation the inoculum is simply whatever was on the fruit and the equipment. succession: One community of organisms changing its environment enough that a different community replaces it. In a coffee tank it runs over hours, driven mostly by oxygen running out and acidity rising. CFU: Colony forming units: the count of cells alive enough to grow into a visible colony on a plate. Usually reported as a logarithm, so 5 log CFU per gram means one hundred thousand per gram. parchment: The papery hull between the mucilage and the seed, also called the endocarp. It is dead tissue, and whether and how fast molecules cross it is the argument of section 05. endosperm: The seed tissue itself, which is the thing we roast. It is alive in a green coffee bean, which is why it does chemistry of its own during processing. water activity: How available the water in something is to a microorganism, as opposed to how much water there is. It falls as a coffee dries, and it is the thing that eventually stops fermentation on a drying bed. Maillard reaction: The set of reactions between sugars and amino acids under heat that browns bread crust, seared meat and roasting coffee, and makes most of coffee’s aroma. spontaneous fermentation: A fermentation with no added culture, carried out by whatever organisms were already present. This is how nearly all coffee is processed. starter culture: A known organism grown up and added deliberately at the start of a fermentation, in order to dominate it. Standard in beer, bread and cheese; still unusual in coffee. carbonic maceration: Fermenting whole or depulped fruit in a sealed vessel under carbon dioxide, borrowed from winemaking. The mechanism is on this page; the practice is on the honey and experimental post. furaneol: The compound that smells of caramel in quantity and of strawberry in trace. It turns up in real strawberries and it is also made by the browning reactions of roasting, which is why a strawberry note in coffee has more than one possible origin. volatile: A compound light enough to leave a liquid or a solid and reach your nose as a gas. Smell is volatiles; taste is what stays behind in the mouth. germination: A seed waking up and starting to grow. A green coffee bean is a live seed, and processing starts this off, which is why some of the chemical change during fermentation is the seed and not the microbes. diffusion: Molecules spreading from where they are concentrated to where they are not, until the two sides even out. It is the usual explanation for how anything made in the fermentation water would reach the seed. GABA: Gamma-aminobutyric acid, a compound both lactic acid bacteria and stressed plant seeds produce. It rises tenfold during fermentation, and which of the two made it is one of the reasons the penetration question is hard. microbe: Any living thing too small to see. Here, the yeasts, bacteria and moulds that ferment coffee. enzyme: A protein that does one chemical job, such as cutting one kind of chain. A pectinase is an enzyme that cuts pectin. yeasts: Single-celled fungi that reproduce by budding, cousins of the one that raises bread. In coffee they are the strongest pectin cutters and the main source of the fruity smells. filamentous fungi: Moulds: fungi that grow as threads rather than single cells. They need time and air, so they belong to the long dry routes. pH: The scale for how acid something is: the lower the number, the more acid. A coffee fermentation falls from about 5.2 to about 4 in a day or two (sources 4 and 6), and the lactic acid bacteria doing it hold the spoilage organisms back. metabolite: Anything a living cell makes as it eats. Lactic acid, ethanol and esters are metabolites of the fermentation microbes. silverskin: The thin skin on the seed itself, under the parchment. A molecule from the fermentation water has to cross it to reach the seed. natamycin: An antifungal that kills yeasts and leaves bacteria alone. Adding it to one tank and not another is how one study isolated what the yeasts contribute. deuterium label: A molecule made with heavy hydrogen, so it can be told apart from the same molecule made by the bean or the microbes. It is how source 10 showed that compounds cross into the seed. propionic and butyric acid: Acids that appear when a fermentation runs far too long: propionic smells of sweaty cheese and butyric of rancid butter. Together they are what people mean by an onion taste. stinker: A bean over-fermented to the point of sulphur compounds and phenol, one of which can be tasted in a whole cup. Stinkers are graded out by hand and by machine. ============================================================================== EVIDENCE · what each numbered claim rests on, and what that study measured mucilage-composition (source 1) says: One kilogram of coffee mucilage is about 842 g water, 89 g protein, 41 g sugar, 9 g pectic substances and 7 g ash, and mucilage with its soluble sugars is around 11.8 percent of the whole fruit by weight. measured: A review compiling compositional analyses; the percentages are attributed to Belitz et al. and the fruit fraction weights to Bressani. Not an original measurement by the review authors, and the underlying work is old, so treat these as standard reference values rather than as a fresh determination. cherry-load (source 2) says: A single coffee cherry carries between 3 times 10 to the 4 and 2.2 times 10 to the 9 microbial cells, median 1.6 times 10 to the 7. measured: Plate counts over two years on fifteen farms in Sul de Minas, Brazil, through dry processing. Culture-dependent, so it counts only what grows on the media used, and the spread reflects real farm-to-farm and weather variation, including a rise after heavy rain on drying cherries. sequencing-gap (source 3) says: Plating found 6 yeast and 17 bacterial species in a wet fermentation where DNA sequencing of the same samples found 212 fungal and 40 bacterial species. measured: Parallel culture-dependent and culture-independent analysis of one 36 hour submerged fermentation in Australia, with HPLC of sugars, organic acids and metabolites in both mucilage and endosperm. The number to take from it is the ratio, not the species list: the point is that plating systematically under-reports. lab-dominance (source 4) says: Lactic acid bacteria held over 60 percent of the bacterial community at every sampling point of a Colombian fermentation, Leuconostoc peaking at 84 percent at 24 hours, while pH fell from 5.2 to 4.2 over 48 hours. measured: Illumina amplicon sequencing at 0, 6, 12, 24, 36 and 48 hours of a traditional Colombian fermentation, with HPLC and GC-MS of metabolites. Relative abundance from sequencing is not the same as absolute count or as metabolic activity, which is the standard caveat on all amplicon work. dry-route-cast (source 5) says: In a Brazilian natural process, Gram-positive bacteria were 85.5 percent of bacterial isolates with Bacillus at 51 percent, yeasts were 22 percent of 940 isolates, Aspergillus was 42.6 percent of fungal isolates, and butyric acid was not detected in any sample. measured: Isolation and identification through a natural coffee fermentation, plus quantification of ten organic acids in the beans. The butyric acid result is a negative finding in one carefully run process, which is why the post uses it to argue that the defect is contingent rather than inevitable, and not to argue that it never happens. wet-vs-dry (source 6) says: Wet and dry processing select different microbial groups, and in dry processing fewer changes were found in the endosperms than in the outer layers. measured: Both routes run on the same harvest, followed from cherry to green bean with high-throughput sequencing and targeted metabolite analysis of separated tissue layers. The layer separation is the important design feature: it is what makes the endosperm-versus-outer-layer comparison possible at all, and it is the strongest published evidence that penetration is limited in the dry route. metabolites-on-bean (source 7) says: After 64 hours of fermentation the beans carried 2.2 g per kg lactic acid, 4.5 g per kg acetic acid and 4.6 g per kg ethanol, and the finished extended-fermentation green beans held 5 times more lactic acid and 1.4 times more mannitol than the standard lot. measured: A full wet process in Ecuador followed at every stage, with metabolite target analysis of fermentation water, soaking water and beans and volatile analysis by GC-MS. Crucially the same paper argues that a large share of the ethanol and the majority of the tenfold GABA rise were the bean’s own metabolism, and explicitly raises adsorption and trapping between endosperm and parchment as alternatives to penetration, so this source supports both sides of section 05. yeast-suppression (source 8) says: Suppressing yeasts with natamycin produced green beans with 25 times less ethyl acetate, 21 times less isoamyl alcohol, 8 times less acetaldehyde and 3.7 times less ethanol, with lower cup scores, and the differences persisted after roasting. measured: A selective suppression rather than an inoculation: natamycin at 300 ppm kills fungi and leaves bacteria alone, so the comparison isolates the yeast contribution against an otherwise identical spontaneous community. This is the cleanest causal attribution on the page. It also found that mucilage was fully degraded in both tanks, which is why the same group later had to run source 9 to work out what actually does the degrading. mucilage-degradation (source 9) says: Significant pectinolytic activity in the first 12 hours is endogenous to the coffee itself; complete degradation requires yeasts; bacteria play no critical role; and adding pectinase with lactic acid finishes the job in under 8 hours. measured: Seven parallel treatments over 36 hours with bacteria and yeasts selectively suppressed, pectinase or lactic acid added, and one held at pH 7 throughout, against a spontaneous control. The experimental design is the reason this is the most useful paper on mucilage in the set: it is one of very few that removes things rather than adding them. Read against source 18, which reached the opposite conclusion about microbial pectinolysis two decades earlier. penetration (source 10) says: Deuterium-labelled aroma compounds put into the surrounding liquid were recovered inside coffee beans: about 11.2 micrograms per gram of 2-phenylethanol, 1.3 of isoamyl acetate and 0.2 of butanal at 12 hours in the medium containing yeast. measured: Three stable-isotope labelled tracers, four media (dehulled, demucilaginated, depulped, and depulped with yeast), five sampling times to 48 hours, quantified by SPME-GC-MS. The label is what makes it decisive: the compound recovered cannot have been made by the bean or the microbes. The same paper reports that parchment resistance significantly affected 2-phenylethanol transfer and that butanal and isoamyl acetate were metabolised inside the bean, which is why the post presents arrival and accumulation as different things. Against this, see source 7 on adsorption and endogenous metabolism, and source 6 on limited endosperm change. Nobody has run the equivalent experiment through a roast. nonvolatile-transfer (source 11) says: Nonvolatile compounds also transfer into coffee beans during wet processing, measured at laboratory scale and under real farm conditions with two yeast strains. measured: A companion study to source 10 by the same group, extending the mass transfer question from volatiles to nonvolatiles and from a laboratory model to an actual on-farm fermentation. Useful because the laboratory-to-farm step is the one most transfer studies skip. duration-dominates (source 12) says: Of the wet-processing parameters varied, fermentation duration had the greatest impact on green bean composition and cup quality, and soaking tempered the fermentation effects and standardised the result regardless of what preceded it. measured: A large-scale factorial wet-processing experiment on Coffea arabica var. Catimor in Yunnan, China, in duplicate, varying processing type, fermentation duration and the application of soaking, with community sequencing, metabolomics and cup scoring. The soaking result is the most actionable finding in the post and the least discussed in the trade. carbonic-maceration (source 13) says: Under carbonic maceration, the best overall cup score was associated with 38 degrees C at 96 hours, and bacterial diversity correlated positively with sensory characteristics. measured: Anaerobic fermentation at varied time and temperature, scored to the Specialty Coffee Association protocol with NMR and denaturing gradient gel electrophoresis. One origin, one protocol, and the temperature and time were varied together, so treat the 38 degrees and 96 hours as a point in that design rather than as an optimum for anybody else’s coffee. pitched-lab (source 14) says: A pitched Lactiplantibacillus plantarum strain accelerated acidification enough to cut a wet fermentation from 24 hours to 12. measured: Selection and on-farm application of lactic acid bacteria during wet processing, with growth and metabolite analysis. The headline is an operational one, throughput rather than flavour, which is worth separating from the sensory claims made for starter cultures generally. starter-fingerprints (source 15) says: Sprayed onto cherries for a dry process, Saccharomyces cerevisiae produced caramel notes after roasting and Candida parapsilosis produced apple and cherry notes, and the method of inoculation changed the result. measured: Three yeast strains inoculated by two methods, direct spraying on the terrace and 16 hours in buckets, with qPCR for inoculum persistence, HPLC, GC-MS identifying 217 volatile compounds, and cupping. The inoculation-method comparison is the interesting part, because it shows the delivery matters as much as the organism. yeast-vs-bacteria (source 16) says: Yeast starters were better producers of volatile alcohols and bacterial starters better producers of organic acids, and malic, lactic and acetic acid were detected only in the bacterial treatments. measured: Three yeast and six bacterial starters inoculated into sterilised coffee beans, with scanning electron microscopy confirming both the sterilisation and the adhesion of cells to the bean surface, plus HPLC and GC-MS. The sterilised substrate makes the attribution clean and makes it less like a real farm, which is the usual trade-off. defect-threshold (source 17) says: Propionic and butyric acid should not exceed about 1 mg per mL, above which they give the onion-like off-flavour, and most microorganisms isolated from spontaneous coffee fermentations lack attributes for improving sensory quality. measured: A review rather than an original study. The threshold is reported rather than measured here, and the review attributes it to an earlier study by López and colleagues, which I have not read; the post and the source note both say so. The second claim is the review authors’ assessment of the starter culture literature as a whole and is the reason the post is cautious about inoculation. pectinolysis-dissent (source 18) says: Depolymerisation of pectic substances by pectolytic micro-organisms does not occur, or is negligible, during coffee fermentation. measured: Isolation of pectolytic organisms from fermenting coffee, mainly Erwinia herbicola and Klebsiella pneumoniae, with enzyme assays. The decisive detail is the pH mismatch: the pectate lyases found had their optimum at pH 8.5 while the fermentation runs from 5.3 down to 3.5, and activity was undetectable without concentrating the supernatant. This flatly contradicts source 9 on the role of yeasts, and the disagreement is partly explained by the fact that this study looked at bacteria and source 9 found the activity in yeasts. ecology-review (source 19) says: Coffee processing is a spontaneous mixed-culture fermentation, and starter culture technology for it is still immature. measured: A review of the microbial ecology of coffee processing and of attempts to control it. Used in this post for framing rather than for any number. semi-dry-starters (source 20) says: Yeast starters in a semi-dry process produced caramel and fruity flavours, with Saccharomyces cerevisiae performing best for that route. measured: Four yeast strains inoculated into a semi-dry (pulped natural) process, with microbiological and biochemical analysis and sensory evaluation by Temporal Dominance of Sensations. The sensory method is unusual and worth noting: it records which sensation dominates over time rather than scoring attributes once. control-review (source 21) says: The controllable parameters of coffee fermentation are time, temperature, pH, water, oxygen and vessel, and industrialising the process means instrumenting them. measured: A review of the move from traditional to controlled coffee fermentation. Used for the structure of section 08, not for a number. glycerol-inside (source 8) says: Lactic acid accumulated inside the beans at about three times the concentration when yeasts were active, and glycerol was present at 0.08 percent with yeasts and undetectable without them. measured: The natamycin suppression design again, but this time the measurement is of the bean interior rather than of the surrounding liquid. Glycerol is the useful one: it is a yeast product, the yeasts were outside the seed, and it was measured inside it. This is the single most concrete piece of evidence on the crossing question that does not rely on a synthetic tracer. pitched-yeast-sugars (source 23) says: Pitching Hanseniaspora uvarum and Pichia kudriavzevii together drove mucilage sucrose from 13.03 to nothing, against 13.03 to 2.01 grams per hundred grams in the uninoculated control, with fructose falling from 27.02 to 4.05 in the control, and esters making up about 30 percent of green-bean volatiles. measured: Inoculated wet fermentation with microbiological, chemical and sensory analysis. It quantifies the substrate side rather than the product side, which is unusual and useful: most inoculation studies report what came out and not what went in. diffusion-underevidenced (source 24) says: Metabolites diffuse through parchment and endosperm until chemical potential equilibrium is reached, and the newer fermentation methodologies lack sufficient scientific evidence. measured: A bibliometric analysis and critical review rather than an experiment. It states diffusion as the mechanism rather than measuring it, which is why the post treats diffusion as the working picture and not as a result. aroma-recombination (source 25) says: Roasted arabica aroma can be rebuilt from 27 pure odorants, and removing them one at a time identifies which ones actually carry the smell. measured: A recombination and omission experiment with a trained panel. It is the reason anyone can say a named compound matters to coffee smell rather than merely being present in it, and it is old enough and clean enough to still be the standard reference. furaneol (source 26) says: Furaneol has a caramel-like smell, is a key flavour compound of many fruits (isolated from strawberry, raspberry and tomato, and a key odorant of pineapple), and is also a Maillard reaction product. measured: A review of the compound: its natural occurrence, its biosynthesis and its formation during heating. The Maillard half is the part that matters for this post, because it means a strawberry note can be created in the roaster without any fermentation having put it there. fermented-vs-washed (source 27) says: Fermented beans carried total esters at 74.5 milligrams per kilogram and 2-phenylethanol at 27.5, measured against washed and against artificially flavoured beans. measured: Chemical and sensory profiling designed as an authenticity question: can you tell a fermented coffee from a flavoured one on chemical grounds. The framing matters as much as the numbers, because it shows the market has produced a problem the analytical literature now has to solve. inoculated-natural (source 28) says: Saccharomyces cerevisiae CCMA 0543 scored 84.75 and 84.92 on the hundred-point scale through natural and pulped natural processing, above the uninoculated controls, with compounds present in the inoculated lots and absent from the controls. measured: Four yeasts through two routes on one variety, fermented 27 hours at 16.5 to 24 degrees, with volatile analysis and cupping. Note that cupping scores are comparable within a study and not across studies, so these numbers should not be set beside the carbonic maceration scores in source 13. germination (source 29) says: Germination-associated metabolism runs inside the seed during processing and is a cause of chemical change independent of the microbes. measured: A review putting coffee germination physiology alongside fermentation microbiology. It is cited here for the third of the three explanations in section 05, the one that says the seed changed itself. transfer-model (source 22) says: Transfer into the bean and degradation inside it are separable processes and have been modelled separately. measured: A kinetic modelling follow-up to source 10, fitting transfer and in-bean degradation as distinct terms. This is the paper that makes the post’s distinction between arrival and accumulation a formal one rather than a rhetorical one. ============================================================================== EXPLANATIONS AT THREE LEVELS Every section is written out three times by hand. explain_section serves these. 01 substrate There is a meal on the outside of the seed, and something will always eat it. [child] Coffee grows as a little red fruit, like a cherry. Inside the fruit are two seeds, and those seeds are what get roasted to make coffee. Wrapped around each seed there is a layer of clear slime, a bit like the slippery bit around a mango stone. That slime is sweet. Tiny living things that we cannot see love sweet stuff, so as soon as somebody picks the fruit, they start eating it. That eating is what the word fermentation means. Nobody has to start it. It just happens. [student] The mucilage is a thin layer of gel between the pulp and the parchment: about 84 percent water, with roughly 9 percent protein, 4 percent sugar and 1 percent pectic substances by weight, and mucilage plus soluble sugars come to around 12 percent of the whole fruit. Pectin is the long chain polysaccharide that glues plant cells together and makes jam set, and coffee pectin is heavily methyl esterified, which makes it harder to break down. Fermentation is what happens when cells metabolise sugar without enough oxygen to oxidise it completely, so they stop at partly oxidised products: ethanol, lactic acid, acetic acid and hundreds of trace compounds. Because the substrate and the microbes are both always present, fermentation is the default state of a picked cherry rather than a step a producer chooses to add. [expert] Substrate definition. Mucilage: water 84.2 percent, protein 8.9, sugar 4.1, pectic substances 0.91, ash 0.7 (Belitz via Esquivel and Jimenez 2012); alcohol-insoluble residue is roughly 30 percent pectic substances, 8 percent cellulose, 18 percent neutral non-cellulosic polysaccharide, with uronic acids around 60 percent of the pectin at high degree of methyl esterification and moderate acetylation. Wet-process fraction weights: skin and pulp 43.2 percent w/w of the whole fruit, mucilage plus soluble sugars 11.8, parchment 6.1. The framing point of the section is that substrate availability and inoculum are both unconditional, so the design space is restricted to who, how long, aw and pO2, which is the organising axis for the rest of the post. ------------------------------------------------------------------------------ 02 cast Nobody adds the microbes. They are already on the fruit. [child] Nobody puts the tiny living things into the coffee. They are already there, sitting on the skin of the fruit, on the farmer’s hands, in the dust, and inside the machines. There can be millions of them on one single coffee cherry. Some of them are yeasts, which are the same sort of thing that makes bread rise. Some are bacteria, the kind that make yoghurt. Some are moulds, like the fuzzy stuff on old bread. They all want the same sweet slime, so they race each other for it. [student] Coffee fermentation is spontaneous: the inoculum is whatever is on the fruit, the hands, the soil and the equipment. A Brazilian survey counted 3 times 10 to the 4 up to 2.2 times 10 to the 9 CFU per cherry, median 1.6 times 10 to the 7. Five functional groups matter. Yeasts (Saccharomyces, Pichia, Candida, Hanseniaspora, Debaryomyces) are the strongest pectinolytic organisms and the main source of fruity volatiles. Lactic acid bacteria (Leuconostoc, Lactobacillus, Lactococcus) acidify. Acetic acid bacteria (Acetobacter, Gluconobacter) oxidise ethanol to acetic acid and need oxygen. Enterobacteria dominate at hour zero and decline. Filamentous fungi belong to long dry routes. Amplicon sequencing changed the picture completely: culture-dependent methods found 6 yeast and 17 bacterial species in one study where sequencing found 212 fungal and 40 bacterial species. [expert] Community composition. Culture-dependent versus culture-independent divergence is the methodological headline: Elhalis 2020 recovered 6 yeast and 17 bacterial species by plating against 212 fungal and 40 bacterial species by high-throughput sequencing of the same samples; Junqueira 2019 reported 160 bacterial genera across 10 phyla in a Colombian ferment with LAB above 60 percent relative abundance at every timepoint and Leuconostoc at 84 percent at 24 h, and a markedly more homogeneous fungal community dominated by Pichia nakasei. Silva 2008 on the dry route: 85.5 percent Gram-positive isolates with Bacillus at 51 percent, yeasts 22 percent of 940 isolates (Debaryomyces 27, Pichia 18.9, Candida 8.0), Aspergillus 42.6 percent of fungal isolates. The practical consequence is that dominance claims are method-dependent and relative abundance is not activity. ------------------------------------------------------------------------------ 03 succession The order matters more than the roster, and acid is the referee. [child] The tiny creatures do not all show up at once, and they do not all stay. First there are lots of one kind, which came in on the skin of the fruit. Then the yeasts get going and start cutting up the slime. Then a third kind, the ones that make yoghurt sour, take over and make everything sour. That sourness is important, because most of the nasty things cannot live in something sour. So the sour ones are sort of protecting the coffee while they eat it. After about a day and a bit, all the slime is gone. [student] Succession means one community changing the environment enough that a different one replaces it. In a coffee tank it runs over hours. Enterobacteria are most numerous at hour zero and decline continuously. Dissolved oxygen is consumed within hours, which disadvantages the acetic acid bacteria. Yeasts rise to roughly 5.5 log CFU per gram, cut pectin and produce ethanol and carbon dioxide. Lactic acid bacteria then achieve quantitative prevalence and hold it. pH falls from about 5.2 to 4.2 over 48 hours in one Colombian trial, and 4.5 after 16 hours against 4.0 after 36 in an Ecuadorean one. Lactic acid bacteria hold back spoilage organisms as they acidify, so the acidification is protective as well as flavour-forming. The mucilage is fully degraded by around 36 hours. [expert] Succession dynamics under submersion. The relevant controls are dissolved oxygen depletion, which selects against obligate aerobes, and progressive acidification, which selects against acid-sensitive Enterobacteriaceae and later against most spoilage taxa. Zhang 2019 reports Leuconostoc pseudomesenteroides characterising a 16 h ferment with transient Lactococcus lactis at 12 to 24 h, and a shift to Lactobacillus vaccinostercus, brevis and plantarum under extension to 64 h. Elhalis 2020 (source 8) gives a yeast maximum near 5.5 log CFU per gram; Elhalis 2023 (source 9) yeasts and LAB near 5.5 and 5.2 log CFU per mL over a 36 h submerged ferment; and Elhalis 2020 (source 3) Citrobacter predominant among aerobic mesophiles. Note that the figure on the page composites endpoints across these studies and models the intervening shape, and that CFU trajectories are not metabolic flux. ------------------------------------------------------------------------------ 04 products Six things come out, and only two of them are acids. [child] When the tiny creatures eat the sweet slime they make new things, the way you breathe out air that is different from the air you breathed in. Some of the new things are sour, like the sourness in yoghurt or in vinegar. Some are a bit like the alcohol in wine. And some of them smell like fruit. That is the surprising one. If you put a sour thing and an alcohol thing together, they can stick to each other and make something new that smells of banana or strawberry, even though there was never any banana or strawberry there. [student] Six classes matter. Lactic acid, soft and yoghurt-like, reaching 8.2 mg per mL in the fermentation water and 2.2 g per kg on beans after 64 hours. Acetic acid, sharply vinegary, 2.4 mg per mL and 4.5 g per kg. Ethanol, 4.6 g per kg on beans, though a large share is the seed’s own anaerobic metabolism. Sugar alcohols, mannitol and glycerol, both faintly sweet, with glycerol a clean yeast marker at 0.08 percent when yeasts grow and undetectable when they are suppressed. Higher alcohols and aldehydes from amino acid catabolism. And esters, formed when an acid and an alcohol condense and release a molecule of water; esters are about half of the volatiles in the fermentation water. Suppressing yeasts with natamycin cut ethyl acetate 25-fold, isoamyl alcohol 21-fold, acetaldehyde 8-fold and ethanol 3.7-fold in the green bean, and the differences persisted after roasting. [expert] Metabolite profile. Heterofermentative LAB signature dominates the wet route: lactic acid, acetic acid, mannitol, ethanol, with 5-ketogluconic acid at 5.7 mg per mL indicating residual AAB oxidative activity early. Over 100 volatiles in fermentation water (roughly 50 percent esters, 23 percent alcohols, 12 percent aldehydes) with a fivefold rise in total aroma intensity from 16 to 64 h, and over 170 on beans with about 70 percent appearing during fermentation. Branched-chain amino acid catabolism supplies the higher alcohols and Strecker-type aldehydes. GABA rose tenfold to 353 mg per kg, which the authors attribute largely to endogenous bean metabolism rather than to LAB glutamate decarboxylase, on retention and signalling grounds. The natamycin design in Elhalis 2020 is the cleanest available causal attribution to the yeast fraction specifically, because it is a selective suppression rather than an inoculation. ------------------------------------------------------------------------------ 05 crossing Now the hard question: how much of that actually reaches the seed? [child] All that eating happens outside the seed. The seed is wrapped in a papery shell, a bit like the shell on a peanut. So here is the question nobody has really answered: does any of the new stuff get through the shell and into the seed we actually drink? Scientists put special marked-up smelly molecules into the water and then looked inside the beans, and yes, they found them. So some does get in. But they still cannot say how much of what you taste came from that, partly because the seed is alive and makes some of the same things all by itself. [student] A microbial metabolite has to cross residual mucilage, the parchment, the silverskin and into the endosperm, then survive washing, soaking, drying and roasting. Evidence that it crosses: three deuterium-labelled aroma compounds put into the surrounding liquid were recovered inside the beans, at about 11.2 micrograms per gram for 2-phenylethanol at 12 hours; mannitol, which the bean cannot make, is elevated in extended-fermentation green beans; and selectively suppressing yeasts changes the roasted coffee and the cup scores. Evidence for caution: the same authors note compounds may be adsorbed on the bean or trapped between endosperm and parchment rather than inside it; the live seed produces GABA and much of the ethanol itself under hypoxia; parchment measurably resisted transfer of 2-phenylethanol; two of the three tracers were degraded inside the bean faster than they arrived; and soaking pulls metabolite levels down to a few percent. Nobody has followed a labelled microbial metabolite through a roast into a cup. Note also that there are three explanations, not two, and they can all be partly true: things diffused in, or fermentation changed the ingredients the roast then worked with, or the living seed changed itself. [expert] The penetration problem. Direct transfer is established: Hadj Salem 2020 used deuterated butanal, 2-phenylethanol and isoamyl acetate in four media including a depulped-with-yeast system, sampled at 0, 6, 12, 24 and 48 h by SPME-GC-MS, reporting 0.2, 11.2 and 1.3 micrograms per gram respectively at 12 h in the yeast medium, with parchment resistance significant for 2-phenylethanol and in-bean metabolic degradation reducing the other two; the 2022 and 2023 follow-ups extend this to nonvolatiles and to a transfer-versus-degradation model. Against a simple diffusion account: De Bruyn 2017 reports substantially fewer endosperm changes than outer-layer changes, especially in the dry route; Zhang 2019 explicitly raises adsorption and inter-tissue trapping as alternatives to intracellular accumulation, attributes the majority of the tenfold GABA rise and a large ethanol fraction to endogenous germination and hypoxia metabolism including ADH and LDH activity, and ascribes carbohydrate off-phase behaviour to endogenous invertase and lipid remobilisation; Zhang 2019 Front Microbiol shows soaking tempering and standardising green bean composition. The unresolved quantity is the fraction of cup-relevant variance attributable to transferred microbial metabolite as against altered precursor stock and endogenous seed metabolism, and the missing experiment is an isotopically labelled microbial metabolite tracked through roasting to the beverage, which Hurtado Cortes et al. 2024 names as an open need of the field in the form of pre- and post-processing seed metabolomics. Three non-exclusive accounts are live: diffusion to chemical-potential equilibrium, altered precursor stoichiometry with the sensed compound formed only in the roast, and germination-associated endogenous metabolism (Waters et al. 2017). The Elhalis 2020 intra-bean measurements, lactic acid at three times and glycerol at 0.08 percent only with yeasts present, are the strongest non-tracer evidence for the first. ------------------------------------------------------------------------------ 06 oxygen Two ways to edit the guest list: shut the lid, or bring your own. [child] Some of the tiny creatures need air to live, and some do not. So if you put the coffee in a sealed container with no air, the air-needing ones stop, and only the others are left. That changes what the coffee tastes like, not because you invented anything new, but because you changed who was allowed to come to the party. You can also do the opposite and add a particular kind of yeast on purpose, so it gets a head start over everything else. [student] Anaerobic means without oxygen. In a sealed vessel the dissolved oxygen is consumed within hours and the carbon dioxide produced keeps air out. Three consequences: acetic acid bacteria stop, because oxidising ethanol to acetic acid requires oxygen; filamentous fungi stop for the same reason, removing the mould risk; and yeasts shift from respiration to fermentation, producing less biomass and more ethanol and aroma compounds. Lactic acid bacteria are largely unaffected. The seed also runs its own anaerobic metabolism harder. Inoculation means adding a known organism in large enough numbers to out-compete the resident community; a pitched Lactiplantibacillus plantarum cut a fermentation from 24 to 12 hours, and different starters leave different fingerprints, with yeasts favouring alcohols and bacteria favouring acids. Two caveats: the pitched organism must beat an adapted resident community, and most organisms isolated from spontaneous ferments do not in fact improve sensory quality. [expert] Mechanism of oxygen exclusion and inoculation. pO2 is the primary selective variable: obligate aerobes (Acetobacteraceae, filamentous Ascomycota) are excluded, facultative fermentative yeasts shift flux from respiration to glycolysis with ethanol and higher-alcohol output rising per unit biomass, and largely aerotolerant-anaerobe LAB are little affected, so the sealed roster converges on Saccharomycetales plus Lactobacillales. Endogenous seed metabolism also shifts, which confounds attribution as in section 05. Controlled evidence is thin: Brioschi Junior 2021 reports a significant functional relationship between global score and temperature at 38 degrees C for 96 h under carbonic maceration with positive correlation between bacterial diversity and sensory characteristics, which is a single origin and protocol. Inoculation is a competitive-exclusion manoeuvre rather than a novel-pathway one; see de Melo Pereira 2016 on accelerated acidification, Bressani 2018 on inoculation method effects, Martinez 2019 on the alcohol versus acid split between yeast and bacterial starters, and Haile and Kang 2019 on the selection problem. ------------------------------------------------------------------------------ 07 roast A fermentation metabolite is not a cup note. [child] After all this, the beans are dried and then roasted, which means cooking them in a hot spinning drum. Roasting is very hot, and lots of the smells the tiny creatures made get burned away or change into something else. Some of the sour bits survive and you can still taste them. Some of the sweet bits get used up to make the brown roasty taste. So the tiny creatures do change your cup of coffee, but often not in the direct way people imagine. [student] The green bean goes into a roaster at over 200 degrees for several minutes; one of the cited trials roasted at 225 degrees for seven. Non-volatile acids, lactic in particular and acetic in part, survive and are tasted. Free sugars and free amino acids are consumed as substrate by the Maillard reaction, so fermentation’s largest effect is probably indirect: it changes the precursor stock the roaster is browning. Light volatiles are largely lost; ethanol boils off far below roasting heat and many light esters do not survive intact. The effect is not zero, though: isoamyl alcohol and ethyl acetate remained elevated after roasting in the yeast-present coffee, and cup scores followed. What does not exist is a study tracking a labelled microbial metabolite through the roast and quantifying its share of the cup. [expert] Thermal fate. Three partitions: survival of low-volatility organic acids, consumption of free reducing sugars and free amino nitrogen as Maillard and Strecker substrate, and loss or rearrangement of light esters and alcohols, alongside de novo formation of pyrazines, furans and phenolics absent from the green bean. The causal significance of fermentation is therefore mostly mediated through precursor stoichiometry rather than through carry-over of finished volatiles, with the natamycin post-roast persistence of isoamyl alcohol and ethyl acetate as the principal counterexample. This is the weakest link in the published chain and the one an isotope-labelled roast study would close. ------------------------------------------------------------------------------ 08 steering You cannot plant a microbe, but you can set the table. [child] A coffee farmer cannot choose which tiny creatures show up. But they can change the conditions so that the ones they want do best. They can leave it longer or shorter, keep it warm or cool, keep it wet or let it dry, shut the air out, or wash it all off at the end. It is a bit like not being able to choose who comes to a party, but being able to choose the music. [student] Six levers, roughly in order of effect. Time is the largest: fermentation duration had the greatest impact on green bean composition and cup quality in the study that varied several parameters at once. Temperature sets rate and roster; altitude and climate therefore matter operationally. Water decides oxygen availability and thermal inertia and, on a bed, water activity falls and progressively shuts groups down. Oxygen decides whether the acetic acid bacteria and moulds compete. The vessel decides oxygen, thermal stability and the resident biofilm. pH falls on its own but can be pushed; adding pectinase and lactic acid broke mucilage down in under 8 hours instead of 36. Washing and soaking are themselves a control: soaking pulled metabolite concentrations down to a few percent and standardised the green bean whatever preceded it. Note that most farms judge the endpoint by feel and smell rather than by instrument. [expert] Process control. Ranked by published effect size, fermentation duration dominates, with processing type (depulped versus demucilaged) and soaking as the next two, per Zhang 2019 Front Microbiol; temperature is underexplored outside the carbonic maceration work. pH trajectory is the state variable most worth instrumenting, because it is both the selective pressure and a proxy for progress. Elhalis 2023 shows exogenous pectinase plus lactic acid completing mucilage breakdown in under 8 h against 36 h spontaneously, and that endogenous pectinolytic activity is significant in the first 12 h independent of microbes, which bears on the Avallone 2002 position that microbial pectinolysis is negligible. Soaking as a standardising operation is the least discussed and most useful control in the set. ------------------------------------------------------------------------------ 09 wrong Where a style becomes a fault, and who gets to say. [child] If you leave it too long, it stops tasting nice. First it tastes fruity, which is good. Then it tastes like vinegar, which some people like and most do not. Then it starts to smell like old cheese and onions, which nobody likes. At the very worst you get one bean that smells so bad it ruins a whole cup, and those are called stinkers. The tricky bit is that there is no exact moment when good turns into bad. Different people draw the line in different places. [student] Past the endpoint, the community works on what is left. The order is: esters peak and fade, acetic acid keeps rising, then propionic and butyric acid appear, giving sweaty cheese and rancid butter, which is what people mean by an onion taste, and the review threshold is around 1 mg per mL. At the far end, sulphur compounds and phenol produce stinker beans, where a single bean can taint a cup. This is not inevitable: butyric acid was not detected at any stage in a carefully run Brazilian natural process. And the line moves: a fermentation four times longer than standard produced a fruitier, more acidic cup that panellists liked, while the same extension on a warmer tank with a different community would be rejected. The distinguishing question is not the flavour but whether the producer could repeat it deliberately. [expert] Defect chemistry and the style-fault boundary. The progression tracks substrate exhaustion and a shift to proteolysis and cross-feeding: ester maximum, then acetate accumulation from residual AAB and heterofermentative LAB, then propionate and butyrate from Propionibacterium-like and clostridial or Bacillus metabolism, then volatile sulphur compounds and guaiacol-type phenolics. Silva 2008 detected no butyric acid at any stage in a controlled natural process, so the progression is contingent on loss of control rather than obligate. Haile and Kang 2019 give the roughly 1 mg per mL propionic and butyric threshold; I have flagged in the source note that I could not establish whether that value originates in a primary measurement. The boundary itself is a scoring convention, not a chemical discontinuity, and the defensible operational criterion is reproducibility rather than sensory descriptor. ------------------------------------------------------------------------------ 10 yours Use your agent with this page. [child] Every part of this page has a little ask button. Press it and it copies a question about exactly that part, which you can paste into a computer helper, like a chatbot. The helper can then read the page and explain that bit more simply, or in more detail, or tell you how the scientists found out. [student] Each section heading, figure caption and text selection offers up to three copyable prompts: explain it simply, go deeper, and how do we know. Each prompt names the exact place on the page. An agent that can drive a browser with WebMCP calls the page’s own tools (get_outline, explain_section, explain_figure, get_evidence, get_glossary, and start_guided_reading for a teaching plan or a quiz); one without a browser is pointed at the plain-text version and the numbered section or figure. The explanations are written by the author at three levels, not generated on the fly. [expert] Six read-only tools registered through document.modelContext.registerTool with readOnlyHint true and an AbortSignal lifetime, feature-detected and try-caught so the page is inert without WebMCP. The schemas, including the id and level enums, are inline; the answer text (thirty section registers, thirty figure registers, the glossary and the evidence notes) lives in a separate data file fetched on the first call through one shared promise, so the registered tool list is the same before and after it loads. get_evidence returns what each study measured and, for penetration, a structured for-and-against with an explicit unresolved field. The copyable prompts are built from a template plus the page’s own section and figure ids, and every call they name is executed against the page in a test. ------------------------------------------------------------------------------ FIGURES AT THREE LEVELS Every figure is explained three times by hand. explain_figure serves these. [HERO] fig-hero One seed, two days: what happens to a coffee seed in a fermentation tank? how to read it: Read left to right along the hour line: the same seed is drawn every 12 hours, from the start to 48 hours. The yellow coat is the mucilage; the small cells on it are yeasts (round, budding) and lactic acid bacteria (blue rods). The dashed arrow above runs from pH 5.2 at the start down to pH 4.2 at 48 hours, the two values measured in one Colombian ferment. [child] This picture shows one coffee seed five times, like the frames of a cartoon, over two days. At the start it wears a thick yellow coat of sweet slime, with only a couple of tiny creatures on it. The creatures eat the coat, and there get to be more and more of them. By the fourth picture the coat is gone, but the creatures are still there. The arrow above says the water around the seed is getting more sour the whole time. [student] The argument of the post in one strip. One seed is drawn at 0, 12, 24, 36 and 48 hours of a fermentation under water. Its sun-coloured coat of mucilage thins as yeasts and lactic acid bacteria multiply on it: most of it is eaten by 24 hours and it is gone by 36; the microbes stay. The arrow above is not a modelled curve but two measurements from one Colombian ferment, pH 5.2 at the start and 4.2 at 48 hours, labelled the acid rises. The captions say the same in words: a coat of sugary mucilage, yeasts and bacteria eating it, most of the coat eaten, the mucilage is gone, sour and spent. [expert] Five stations at 12 hour intervals. The coat thickness follows the mucilage being fully degraded by the end of a 36 hour submerged ferment (sources 8 and 9); the number of microbe glyphs is schematic, not a scaled count. The pH is deliberately not a curve: it is the start and end values of one Colombian ferment (source 4, 5.2 to 4.2 over 48 h), joined by a dashed arrow because the intermediate values are not reported there. The lactic acid bacteria are drawn as the growing population because source 4 found them above 60 percent of the bacterial community at every sampling point. ------------------------------------------------------------------------------ [FIGURE 1] fig-mucilage The meal on the seed: what is the sweet gel on the seed made of, and what breaks it down? how to read it: On the left, a coffee cherry cut in half with its layers named from the skin inwards. The small ring on the mucilage layer opens into the round lens on the right, which magnifies it. The bar underneath is one kilogram of mucilage, split by weight. [child] On the left is a coffee cherry cut in half, so you can see its layers like an onion: the skin, the sweet flesh, the slimy layer, a papery shell and the seed. The circle on the right is a pretend magnifying glass pointed at the slimy layer. Inside it are long strings that hold the slime together, little dots of sugar, and a yeast snipping one of the strings. The bar at the bottom shows that the slime is nearly all water. [student] The cherry in section places the mucilage between the pulp and the parchment. The lens is a diagram, not a micrograph: pectin chains lie across it, whole at the top and cut into shorter pieces lower down, sugar floats free as dots, and a yeast’s pectinase has just cut a chain, which is how the gel loses its structure. The bar gives one kilogram of mucilage in proportion: 842 g water, 89 g protein, 41 g sugar, 9 g pectin and 7 g minerals. The meal is small, but it is always there. [expert] Composition from Esquivel and Jiménez 2012 after Belitz (source 1), drawn to proportion; pectin is picked out as the structural fraction because its high degree of methyl esterification is what makes it slow to depolymerise. The lens gives the cut to a yeast pectinase, which follows Elhalis et al. 2023 (source 9: complete degradation only with yeasts present, bacteria not critical) rather than Avallone et al. 2002 (source 18: bacterial pectolytic enzymes with an optimum at pH 8.5, negligible at fermentation pH). It does not draw the fruit’s own enzymes, which source 9 found significant in the first 12 hours. ------------------------------------------------------------------------------ [FIGURE 2] fig-cast The cast: who are the microbes, and what does each one do? how to read it: Six small drawings in a row, each named with its job. Each group is drawn in the ink it wears on every later figure: sun for yeasts, blue for lactic acid bacteria, coral for acetic acid bacteria, indigo for enterobacteria, teal for moulds. [child] These are the six characters in the story, drawn as cartoons and not to size. Yeasts are round blobs that grow new blobs; they snip up the slime and make the fruity smells. The yoghurt bacteria are a chain of little sausages; they make things sour. The vinegar bacteria have bubbles, because they need air. The ones with tails turn up first and lose first. The moulds are branching threads. The last one is a dropper, for when a farmer adds a yeast on purpose. [student] A picture key for the rest of the post. Yeasts, budding cells: they cut the pectin and make most of the fruity smells. Lactic acid bacteria, a chain of rods: they make lactic acid and drop the pH. Acetic acid bacteria, rods with air bubbles: they need air and make vinegar. Enterobacteria, rods with tails: most numerous at hour zero and out-competed within a day. Filamentous fungi, branching threads with spore heads: they need time and air. And a dropper adding yeasts: a starter culture someone chose, which moves the starting line rather than adding new chemistry. [expert] A glyph key, not data: shapes are chosen to be told apart, not to scale, and the inks hold across the succession, routes and defect figures. The groups are functional rather than taxonomic: yeasts (Saccharomyces, Pichia, Candida, Hanseniaspora, Debaryomyces), lactic acid bacteria (Leuconostoc, Lactobacillus, Lactococcus), acetic acid bacteria (Acetobacter, Gluconobacter), enterobacteria (Enterobacter, Citrobacter, Erwinia, Klebsiella) and filamentous fungi (Aspergillus, Penicillium, Fusarium, Cladosporium). It leaves out the long tail that sequencing finds (212 fungal and 40 bacterial species against 6 yeast and 17 bacterial by plating, source 3), so read it as who dominates, not who is present. ------------------------------------------------------------------------------ [FIGURE 3] fig-succession A tank, hour by hour: who is winning the tank at each hour, and how sour does it get? how to read it: The top panel has four population curves, each named at its end. The axis shows order only, with no numbers, because the counts behind it were measured in different units. The yellow area underneath is the mucilage left, gone at the dashed line by 36 hours. The bottom panel shows measured pH values as dots from two separate ferments: solid dots for Colombia, open dots for Ecuador. Hover, drag or use the arrow keys to read any hour. [child] This is a race. Each coloured line is one team of tiny creatures, and higher means more of them. One team starts on top and then drops away. The yeasts climb and then level off. The blue team, the yoghurt bacteria, climbs highest and stays there. The yellow shape underneath is the sweet slime they are all eating, and it is gone after about a day and a half. The dots at the bottom are real measurements of how sour the tank got, and they go down as it gets more sour. [student] Four groups over one submerged fermentation: enterobacteria start highest and fall, acetic acid bacteria fall as the oxygen goes, yeasts rise to a plateau, and lactic acid bacteria rise highest and stay. The population axis is ranked, not numbered, because the yeast count (about 5.5 log CFU per gram) and the lactic acid bacteria count (about 5.2 log CFU per millilitre) were measured in different units. The sun-coloured area is the mucilage still on the bean, gone by 36 hours. The lower panel shows only measured pH: 5.2 at the start and 4.2 at 48 hours in Colombia, and 4.5 at 16 hours and 4.0 at 36 hours in Ecuador. Two things decide the winner: the oxygen running out and the acid building up. [expert] A composite: the order of dominance and the endpoints come from sources 3, 4, 6, 7, 8 and 9, and the shapes between them are modelled, which is why the y-axis carries rank only. The units problem is real: plate counts per gram of beans (source 8, yeasts near 5.5 log CFU per gram) and per millilitre of fermentation mass (source 9, lactic acid bacteria near 5.2 log CFU per mL) cannot share one scale, and sequencing-based relative abundance (source 4) is not a count at all. The pH panel keeps the Colombian series (source 4, 5.2 to 4.2 over 48 h) and the Ecuadorean one (source 6, 4.5 at 16 h and 4.0 at 36 h) apart rather than splicing them into a curve. Mucilage exhaustion by 36 h is from sources 8 and 9. ------------------------------------------------------------------------------ [FIGURE 4] fig-ester An ester, drawn: how do an acid and an alcohol become a fruit smell? how to read it: Read it as an equation, left to right: an acid plus an alcohol gives an ester plus a drop of water. [child] This is a recipe drawn with shapes. The orange-red shape is a sour thing that bacteria make. The yellow circle is an alcohol that yeasts make. Put them together and they hold on to each other, and a tiny drop of water pops out. The two joined together smell of banana, even though neither of them did on its own. [student] The post’s own example of an ester forming. Acetic acid (the coral hexagon, from the bacteria) plus isoamyl alcohol (the sun-coloured circle, from the yeasts) gives isoamyl acetate, the two joined by a bond, which smells of banana and pear drops, plus one molecule of water. Neither parent smells of fruit: the acid is sour and the alcohol sharp and solventy. That is why a coffee can smell of fruit with no fruit added, and why it takes the yeasts and the bacteria together. [expert] A condensation drawn with tokens rather than structures: acid plus alcohol, water leaving. Acetic acid plus ethanol gives ethyl acetate by the same step, nail varnish in quantity and something like pineapple in trace. The causal evidence that the yeasts matter to this equation is the natamycin suppression in source 8: removing only the yeasts cut ethyl acetate 25-fold and isoamyl alcohol 21-fold in the green bean. Esters were about half the volatiles in the fermentation water in source 7, and about 30 percent of green-bean volatiles in the inoculated trial of source 23. ------------------------------------------------------------------------------ [FIGURE 5] fig-crossing Into the bean: does anything from the tank actually get inside the bean? how to read it: On the left (on a phone, on top), the layers a molecule must cross, not to scale, from the water to the living seed, with a dashed line where the evidence thins. On the right, three labelled tracers inside the bean, on a log scale, on an hour axis marked at the paper's sampling times. Only the 12 hour values are drawn as points, circled; the dashed arrows show only the direction each compound went afterwards. Hover, drag or use the arrow keys. [child] On the left are the walls a smell would have to get through to reach the inside of a coffee seed: the water, the slime, a papery shell, a thin skin, and then the seed. On the right is what happened when scientists tried it. They put three specially marked smells in the water and looked inside the beans. All three got in: those are the circles. After that, one kept building up, and the other two went down again, because the living seed used them up. [student] The left panel stacks the barriers edge on: water, mucilage, parchment, silverskin and the living seed, not to scale, with a wandering path and a dashed line where the published evidence thins out. The right panel is source 10: three aroma compounds labelled with deuterium, a heavy form of hydrogen, put in the liquid around the beans and measured inside them. The circled points are the published 12 hour values in the medium with yeast: 11.2 micrograms per gram of 2-phenylethanol (rose, honey), 1.3 of isoamyl acetate (banana) and 0.2 of butanal. The arrows only show direction: 2-phenylethanol kept accumulating, while the other two were consumed inside the bean. Things cross; arriving is not the same as staying. [expert] Hadj Salem et al. 2020 (source 10): deuterated butanal, 2-phenylethanol and isoamyl acetate, four media, sampled at 0, 6, 12, 24 and 48 h by SPME-GC-MS. Only the 12 h values in the depulped-with-yeast medium are drawn, because those are the values published in the abstract; the arrows encode the reported direction, continued uptake for 2-phenylethanol, whose transfer the parchment significantly slowed, and metabolic loss inside the bean for the other two, which source 22 models separately from transfer. The y-axis is logarithmic so all three fit. The layer stack is qualitative because no canonical parchment thickness could be verified. The figure stops at the green bean: nobody has followed a labelled microbial metabolite through a roast. ------------------------------------------------------------------------------ [FIGURE 6] fig-routes Four routes, one clock: how long do the microbes get on each route, and who is ahead when? how to read it: Four lanes, one per route, on one shared time axis that is stretched so the early hours get room; no hours are marked, because the phase lengths vary by farm and weather. Each lane is split into phases coloured by the microbes in the lead and named inside the bar. Hover, drag or use the arrow keys to read all four routes at one moment. [child] This compares four ways of handling coffee, one per row, on the same clock. The clock is squashed, so the start gets lots of room and the end is squeezed up. The colours show which tiny creatures are in charge. The washed coffee finishes first. The sealed one gets a spell with no air at all. The honey and natural coffees sit out drying for much longer, and near the end of the natural one, moulds can move in. [student] Washed, honey, natural and sealed share one stretched clock, so the only differences on show are time and what the microbes are given. Washed: depulped, a lactic fermentation in the tank, washed off, then drying with little left to eat. Honey: mucilage kept, yeasts on a drying skin, then acetic acid bacteria as the water falls. Natural: whole cherry, yeasts inside the fruit, then acetic, then moulds. Sealed: no oxygen, so only lactic acid bacteria and yeasts, then opened and dried. The axis shows order and relative length only; the boundaries are typical, not fixed. The equipment for each route is in its own post. [expert] A schematic, not a data set. Phases are coloured by the functional group expected to dominate: the wet-versus-dry contrast of source 6 (acetic acid bacteria most abundant in dry processing, lactic acid bacteria in wet), lactic dominance in submerged tanks (sources 4 and 7), and the oxygen-exclusion logic of section 06 (acetic acid bacteria and filamentous fungi removed under seal). Falling water activity on the beds is what ends each dry-route phase. No hours are printed because phase lengths are farm- and weather-dependent and no source gives a general timeline. Starter cultures are not drawn; they would move the first phase of any lane. ------------------------------------------------------------------------------ [FIGURE 7] fig-roast Three fates in the roaster: what happens in the roaster to what the microbes made? how to read it: Seven ingredients on the left, each with a line into the roasting drum in the middle, and three outcomes on the right. Follow a line to see where each ends up; the middle outcome is drawn largest. Acetic acid has a second, faint line, because only part of it survives. [child] The round thing in the middle is the hot spinning drum that roasts coffee. On the left are things the tiny creatures left in the bean. Follow each line to the right to see what happens to it. A few sour things come out the other side, and you can still taste them. The sugars get used up to make the brown, roasty taste, and that is the biggest box. The alcohol and the fruity smells mostly float away. [student] Seven inputs, grouped by where they end up. Carried through, still there in the cup: lactic acid, acetic acid in part, and higher alcohols. Used as fuel: free sugars and free amino acids, eaten by the browning reactions, drawn largest because this is probably fermentation’s biggest effect on the cup. Gone, boiled off or broken apart: ethanol, which boils at 78 degrees, and the light fruit esters, plus a faint second line from acetic acid. The drum is at over 200 degrees. The roaster mostly receives a different set of ingredients from a fermentation, not a set of finished flavours. [expert] Three partitions: carry-over of low-volatility organic acids, consumption of free sugars and free amino acids as browning (Maillard and Strecker) substrate, and loss or rearrangement of ethanol and light esters. The relative sizes are the post’s argument, not a measurement. The main counterexample is source 8, where isoamyl alcohol and ethyl acetate stayed higher after roasting in the yeast-present coffee, so "gone" for the esters means mostly, not entirely. The link the figure cannot close is quantitative: no study has followed a labelled microbial metabolite through the roast and measured its share of the cup. ------------------------------------------------------------------------------ [FIGURE 8] fig-dials Six dials: what can a producer change, and which matters most? how to read it: Six dials, numbered in rank order and shrinking from left to right. Size shows the order only; nobody has measured the gaps. [child] A farmer cannot choose the tiny creatures, but they can turn these six knobs. The biggest knob is time: how long they leave it. Then how warm it is, how wet it is, whether air gets in, what container they use, and how sour it is or what they add. [student] Six controls in the order the control literature keeps arriving at: time, temperature, water, oxygen, the vessel, and pH and what you add. Time is drawn largest because the study that varied several things at once found fermentation duration had the greatest effect on green bean composition and cup quality (source 12). The shrinking sizes after that show rank only; nobody has measured the gaps. Washing and soaking at the end are not among the six, but the text argues they are a control in their own right. [expert] An ordinal ranking drawn by size, after the control-parameter framing of Elhalis et al. 2023 (source 21), with duration first per Zhang et al. 2019 (source 12), where processing type and soaking were the next largest effects. Temperature is underexplored outside the carbonic maceration work (source 13, one origin, one protocol). pH is the state variable most worth instrumenting, because it is both the selective pressure and a proxy for progress. The dial sizes carry no effect-size information. ------------------------------------------------------------------------------ [FIGURE 9] fig-defects From style to fault: when does a fermented flavour become a fault? how to read it: Time runs left to right from the moment the mucilage is gone, with no hour scale. Four curves show which smell is loudest, in the order they arrive; the band below names what a cupper would call the cup; the shaded stretch between two dashed lines is where buyers stop agreeing. Hover, drag or use the arrow keys. The order is sourced; the timing is not drawn. [child] This shows what happens if coffee is left fermenting too long, after all the sweet slime is gone. Further right means left longer. First it smells nicely fruity. Then it goes vinegary. Then it smells of old cheese and onions, and at the very end of something rotten. The strip at the bottom gives each part a name, and the shaded bit is where people argue about whether it is still nice. [student] Four curves, in the order they arrive: fruit esters peak early, acetic acid rises next, then propionic and butyric acid, then sulphur compounds and phenol at the far end. The band underneath names the cup: clean, fruity and winey, sharp and vinegary, oniony, stinker, with the joins deliberately blurred and a shaded stretch where buyers stop agreeing. There is no hour scale because nobody has published a clean timeline; the order is from sources 5 and 17, and the onion threshold of about one milligram per millilitre from source 17. It is not inevitable: in one carefully run natural process, butyric acid was never detected (source 5). [expert] The progression follows substrate exhaustion and a shift to proteolysis and cross-feeding: ester maximum, then acetate from residual acetic acid bacteria and heterofermentative lactic acid bacteria, then propionate and butyrate, then volatile sulphur compounds and phenolics. The x-axis is ordinal by design. The propionic and butyric threshold of about 1 mg per mL is from a review (source 17) that attributes it to earlier work by López and colleagues. The shaded zone encodes the post's argument that the style-fault line is a scoring convention, not a chemical discontinuity, and that reproducibility is the defensible criterion. ------------------------------------------------------------------------------