{ "id": "anaerobic-fermentation", "url": "https://coffearium.org/entries/anaerobic-fermentation/", "license": "CC-BY-4.0", "title": "Anaerobic fermentation", "summary": "Fermenting whole cherries or pulped coffee in closed, oxygen-poor vessels before drying. It changes the acids, alcohols and aroma of the coffee, and it carries its own risks for the seed and the cup.", "domain": "post-harvest-processing", "kind": "Method", "stage": "process", "topics": [ "science" ], "topicSections": {}, "lang": "en", "status": "verified", "reviewer": "Coffearium verification pass", "lastVerified": "2026-10-21T00:00:00.000Z", "updated": "2026-10-21T00:00:00.000Z", "added": "2026-10-21T00:00:00.000Z", "brief": [ "In self-induced anaerobic fermentation (SIAF), coffee is closed in a tank with an airlock; microbes use up the oxygen, and the carbon dioxide they make escapes through the valve.2,6", "Whole cherries or pulped coffee can be fermented this way, with or without added water.1", "In trials the pH fell from about 5.1–5.8 to 3.7–4.4 within 72–96 hours, and lactic acid was the main acid formed.1,5,6", "Several trials scored anaerobic coffees above conventional controls, but a 2025 Colombian trial found no significant effect of the fermentation condition on cup score.2,4,5", "Long runs bring risks – falling seed viability, more mycotoxin-producing fungi and phenolic cups, especially in open tanks.2,3" ], "quickFacts": [ { "label": "Vessel", "value": "Closed tank or bucket with an airlock; vacuum-sealed or CO₂-filled bags in some trials" }, { "label": "Raw material", "value": "Whole cherries or pulped coffee, with or without water" }, { "label": "Trial durations", "value": "24 h to 192 h; 7 days in one trial" }, { "label": "pH course in trials", "value": "From about 5.1–5.8 to 3.7–4.4" }, { "label": "Temperature difference (ΔT), room at 20 °C", "value": "0.5–3.7 °C sealed; 5.1–17.6 °C open tank (one trial)" }, { "label": "Main acid formed", "value": "Lactic acid" }, { "label": "Cup scores in trials", "value": "78.51–87.50 points, apart from one defective sample; gains over controls not consistent" } ], "aka": [ "SIAF", "Self-induced anaerobic fermentation", "Anaerobic processing", "Anaerobic natural", "Anaerobic washed" ], "linkTerms": [ "anaerobic coffee", "anaerobic coffees", "anaerobically fermented", "sealed tank" ], "related": [ "fermentation", "microorganisms", "carbonic-maceration", "co-fermentation", "natural-process", "washed-process", "honey-process", "drying", "seed-physiology", "processing-defects", "ochratoxin-a" ], "contested": [ "Effect on cup score: in several trials anaerobic or SIAF coffees scored above conventionally processed coffee (86.50 for yeast-inoculated SIAF against 84.25 for conventional pulped coffee in Brazil; 87.50 against 83.30 for natural coffee in another Brazilian trial; 87.38 against 84.00 for washed coffee and 82.50 for the control sample in Peru)4,5,11, while a 2025 Colombian trial found that the fermentation condition, the raw material and the time had no significant effect on cup score.2", "Seed viability under SIAF: in a Colombian trial embryo viability stayed above 80% for up to 120 hours of SIAF2, while in a Brazilian trial with 180 hours of SIAF it fell from 85–87.5% in fresh coffee to 52.5–70% after drying.4 The trials differ in duration, variety and when viability was measured.", "Which microbes lead: in one Brazilian SIAF of pulped coffee, enterobacteria made up about 85% of the bacteria in the first 24 hours and acetic acid bacteria stayed below 1%1; in a Colombian trial the bacteria were dominated by lactic acid bacteria (mainly Weissella), acetic acid bacteria and enterobacteria, and acetic acid bacteria were more abundant in open tanks, where they made up 20–57% of the community.2" ], "corrections": [ { "date": "2026-10-21T00:00:00.000Z", "change": "The 86.50 score of the Brazilian 2023 trial is now attributed to yeast-inoculated SIAF (Contested, score chart); the paper gives no score for uninoculated SIAF.", "reason": "Jimenez et al. 2023 report 86.50 for pulped coffee inoculated with T. delbrueckii." }, { "date": "2026-10-21T00:00:00.000Z", "change": "Peru 2023: washed coffee (84.00) is no longer called a control; the study's own control sample (82.50) is added, and the chart notes that the anaerobic method is not described.", "reason": "The abstract of Antezana & Luna-Mercado 2023 lists four methods and a separate control sample." }, { "date": "2026-10-21T00:00:00.000Z", "change": "Colombian microbiota: the community is described as dominated by lactic acid bacteria, acetic acid bacteria and enterobacteria together; acetic acid bacteria were more abundant (20–57%) in open tanks, not present only there.", "reason": "Peñuela-Martínez et al. 2025 name all three groups as dominant and Weissella as the dominant lactic acid bacterium." }, { "date": "2026-10-21T00:00:00.000Z", "change": "Citrus and berry notes are attributed to whole cherries in open tanks, not to open tanks in general.", "reason": "Peñuela-Martínez et al. 2025 report these notes for fruit fermentations under semi-anaerobic conditions." }, { "date": "2026-10-21T00:00:00.000Z", "change": "Cup-score range in Quick facts widened to 78.51–87.50 (one defective sample excluded); 'filamentous fungi' became 'mycotoxin-producing fungi'; the open-tank warming is given as the paper's temperature difference (ΔT).", "reason": "The Conilon trial scored down to 78.51; Buitrago-Zuluaga et al. 2025 report more mycotoxigenic genera with time; the paper does not define ΔT." }, { "date": "2026-10-21T00:00:00.000Z", "change": "Smaller fixes: lactic acid citations moved from da Silva Vale et al. (who name lactic and acetic acid together) to Rocha et al.; Yunnan drying no longer says 'racks'; Brazilian 2024 bioreactors 'in the sun' rather than 'outdoors'; the 2,3-butanediol pathway sentence limited to the bacterium studied; Fusarium 'predominated' rather than 'rose'.", "reason": "Wording in the cited sources." } ], "sources": [ { "id": "s1", "title": "Understanding the Effects of Self-Induced Anaerobic Fermentation on Coffee Beans Quality: Microbiological, Metabolic, and Sensory Studies", "authors": "da Silva Vale, A.; Balla, G.; Rodrigues, L. R. S.; de Carvalho Neto, D. P.; Soccol, C. R.; de Melo Pereira, G. V.", "year": 2023, "publisher": "Foods 12(1), 37", "url": "https://www.mdpi.com/2304-8158/12/1/37", "doi": "10.3390/foods12010037", "tier": "A", "quote": "All the coffee beverages scored ≥ 84 and were classified as specialty coffees.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s2", "title": "Development of Fermentation Strategies for Quality Mild Coffee Production (Coffea arabica L.) Based on Oxygen Availability and Processing Time", "authors": "Peñuela-Martínez, A. E.; Osorio-Giraldo, C. V.; Buitrago-Zuluaga, C.; Medina-Rivera, R. D.", "year": 2025, "publisher": "Foods 14(17), 3001", "url": "https://publicaciones.cenicafe.org/index.php/pex/article/download/4301/5973/4657", "doi": "10.3390/foods14173001", "tier": "A", "quote": "Under SIAF, viability remained above 80% for up to 120 h.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s3", "title": "Identification of Filamentous Fungi Present in Prolonged Fermentations of Coffea arabica L. var. Castillo", "authors": "Buitrago-Zuluaga, C.; Osorio-Giraldo, C. V.; Peñuela-Martínez, A. E.", "year": 2025, "publisher": "Applied Microbiology 5(4), 114", "url": "https://publicaciones.cenicafe.org/index.php/pex/article/download/4556/5969/4653", "doi": "10.3390/applmicrobiol5040114", "tier": "A", "quote": "SIAF showed no OTA production and lower abundances of OTA-producing genera compared with the SA method.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s4", "title": "Influence of anaerobic fermentation and yeast inoculation on the viability, chemical composition, and quality of coffee", "authors": "Jimenez, E. J. M.; Martins, P. M. M.; Vilela, A. L. de O.; Batista, N. N.; da Rosa, S. D. V. F.; Dias, D. R.; Schwan, R. F.", "year": 2023, "publisher": "Food Bioscience 51, 102218", "url": "https://www.alice.cnptia.embrapa.br/alice/bitstream/doc/1148954/1/Influence-of-anaerobic-fermentation-and-yeast-inoculation-on-the-viability.pdf", "doi": "10.1016/j.fbio.2022.102218", "tier": "A", "quote": "The fermentation time for both natural and pulped coffee was 180 h.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s5", "title": "Enhancing Sensory Quality of Coffee: The Impact of Fermentation Techniques on Coffea arabica cv. Catiguá MG2", "authors": "Silva, L. C. F.; Pereira, P. V. R.; da Cruz, M. A. D.; Costa, G. X. R.; Rocha, R. A. R.; Bertarini, P. L. L.; do Amaral, L. R.; Gomes, M. S.; Santos, L. D.", "year": 2024, "publisher": "Foods 13(5), 653", "url": "https://mdpi-res.com/d_attachment/foods/foods-13-00653/article_deploy/foods-13-00653.pdf", "doi": "10.3390/foods13050653", "tier": "A", "quote": "All bioreactors were hermetically sealed and equipped with an S-type Airlock with a seal for gas exchange.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s6", "title": "Evaluation of Arabica Coffee Fermentation Using Machine Learning", "authors": "Rocha, R. A. R.; da Cruz, M. A. D.; Silva, L. C. F.; Costa, G. X. R.; Amaral, L. R.; Bertarini, P. L. L.; Gomes, M. S.; Santos, L. D.", "year": 2024, "publisher": "Foods 13(3), 454", "url": "https://mdpi-res.com/d_attachment/foods/foods-13-00454/article_deploy/foods-13-00454.pdf", "doi": "10.3390/foods13030454", "tier": "A", "quote": "This setup allowed the carbon dioxide generated during fermentation to escape while preventing outside air from entering.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s7", "title": "Metabolome analysis of processing methods regulating flavor precursors in Arabica coffee beans and correlation with sensory attributes", "authors": "Fu, X.; Yu, H.; Dong, W.; Bi, X.; Li, Z.; et al.", "year": 2026, "publisher": "Beverage Plant Research 6, e029", "url": "https://www.maxapress.com/article/doi/10.48130/bpr-0026-0010", "doi": "10.48130/bpr-0026-0010", "tier": "A", "quote": "Vacuum extraction is used to remove air from the plastic bags, thereby ensuring an anaerobic environment.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s8", "title": "Changes in the Chemical and Sensory Profile of Coffea canephora var. Conilon Promoted by Carbonic Maceration", "authors": "Gomes, W. d. S.; Pereira, L. L.; Filete, C. A.; Moreira, T. R.; Guarçoni, R. C.; da Silva Oliveira, E. C.; Moreli, A. P.; Guimarães, C. V.; Simmer, M. M. B.; Júnior, V. L.; et al.", "year": 2022, "publisher": "Agronomy 12(10), 2265", "url": "https://biblioteca.incaper.es.gov.br/digital/bitstream/item/5318/1/Changes-in-the-Chemical-and-Sensory-Prof.pdf", "doi": "10.3390/agronomy12102265", "tier": "A", "quote": "The study was implemented in anaerobic conditions with different fermentation times (24, 48, 72, 96, and 120 h) and temperatures (18, 28, and 38 °C).", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s9", "title": "Effects of postharvest processing on aroma formation in roasted coffee – a review", "authors": "Cao, X.; Wu, H.; Viejo, C. G.; Dunshea, F. R.; Suleria, H. A. R.", "year": 2023, "publisher": "International Journal of Food Science and Technology 58(3), 1007–1027", "url": "https://eprints.whiterose.ac.uk/id/eprint/195860/", "doi": "10.1111/ijfs.16261", "tier": "A", "quote": "It was introduced to the coffee industry recently and is designed to leave the intact coffee beans sealed in an enclosed environment, with expelled O2 and filled with CO2.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s10", "title": "The altitude of coffee cultivation causes shifts in the microbial community assembly and biochemical compounds in natural induced anaerobic fermentations", "authors": "Martinez, S. J.; Simão, J. B. P.; Pylro, V. S.; Schwan, R. F.", "year": 2021, "publisher": "Frontiers in Microbiology 12, 671395", "url": "https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2021.671395/full", "doi": "10.3389/fmicb.2021.671395", "tier": "A", "quote": "Fermentation lasted 72 h, and sub-samples of approximately 100 g were taken after 48 h.", "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s11", "title": "Effect of processing methods (washed, honey, natural, anaerobic) of catimor coffee on physical and sensory quality in Alto Inambari, Peru", "authors": "Antezana, R. N. V.; Luna-Mercado, G. I.", "year": 2023, "publisher": "Coffee Science 18, e182111", "url": "https://coffeescience.ufla.br/index.php/Coffeescience/article/view/2111", "doi": "10.25186/.v18i.2111", "tier": "A", "quote": "The results showed that yield, the main attribute of physical quality, grouped the treatments into two different groups: (1) control 71.17%, washed 75.83%, honey 75.24% and (2) natural 44.45%, anaerobic 45.70%; regarding to the sensory quality, of the sensory attributes only fragrance and body are significantly different, the total cup score was: 82.50 (control), 84.00 (washed), 82.88 (honey), 85.75 (natural) and 87.38 (anaerobic).", "abstractOnly": true, "accessed": "2026-10-21T00:00:00.000Z" }, { "id": "s12", "title": "Metabolic engineering for the production of acetoin and 2,3-butanediol at elevated temperature in Parageobacillus thermoglucosidasius NCIMB 11955", "authors": "Sheng, L.; Madika, A.; Lau, M. S. H.; Zhang, Y.; Minton, N. P.", "year": 2023, "publisher": "Frontiers in Bioengineering and Biotechnology 11, 1191079", "url": "https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2023.1191079/full", "doi": "10.3389/fbioe.2023.1191079", "tier": "A", "quote": "The route to 2,3-BDO from pyruvate involves the intermediate metabolites acetolactate and acetoin, catalyzed by ALS, ALD and BDH.", "accessed": "2026-10-21T00:00:00.000Z" } ], "markdown": "## What anaerobic means in coffee\n\nAnaerobic means “without oxygen”. In coffee processing, the word describes fermentation in a closed vessel that keeps outside air away from the coffee. Producers and researchers use three set-ups:\n\n- **Self-induced anaerobic fermentation (SIAF)** – the coffee is closed in a tank or bucket. Microbes consume the oxygen that is inside, and carbon dioxide (CO₂) slowly builds up.2,4 An airlock lets the gas escape and stops outside air from coming in.1,6\n- **Air removed at the start** – the coffee is packed in plastic bags and the air is drawn out with a vacuum pump.7\n- **Air replaced with CO₂** – after the vacuum step, pure CO₂ is pumped into the bags.8 Studies using this set-up often call it carbonic maceration, which has its own entry.8,9\n\nThe opposite is an open tank. A 2025 Colombian study calls fermentation in open containers “semi-anaerobic”.2 In practice, “anaerobic” means low in oxygen, not free of it. The closed vessel limits oxygen and favours microbes that can live with or without it.4\n\n
\n
\n
Three ways to keep air away from fermenting coffee, next to an open tank for comparison. Schematic, not to scale.2,6,7,8
\n
\n\n## Whole cherries or pulped coffee\n\nAnaerobic fermentation can start from whole cherries or from pulped coffee. Both can be fermented dry or with added water.1 Two Brazilian trials, for example, compared whole and pulped coffee, each without water or with water making up 30% of the volume.5,6\n\nThe choice changes what the microbes can feed on. Whole cherries keep their sugary pulp. In one trial, the juice of whole cherries started at 23.65 °Brix, against 7.33 °Brix for pulped coffee.6 °Brix is a measure of dissolved solids, mostly sugars.\n\nVessel sizes in the trials range from 20-litre buckets with lids to 200-litre tanks.10,5 Runs lasted from one day to eight days.1,2\n\n## What happens in the closed tank\n\nThe steps resemble those of ordinary fermentation, which has its own entry. Sugars are used up, acids and alcohols form, and the pH falls.\n\n- **pH falls.** In a Brazilian tank of pulped coffee, the pH fell from 5.8 to 4.4.1 In another trial, it fell from about 5.1 to as low as 3.70 within 72 hours.6\n- **Lactic acid builds up.** Lactic acid was the main acid formed in several trials.4,5,6 In a Colombian comparison, the sealed tanks used more glucose and made more lactic acid than open tanks.2\n- **Alcohol builds up.** Green coffee fermented for 72 hours held about six times more ethanol than coffee fermented for 24 hours.1\n- **The mass stays cooler.** In a room held at 20 °C, the temperature differences (ΔT) reported were only 0.5–3.7 °C for sealed tanks, against 5.1–17.6 °C for open tanks.2\n\n## Drying afterwards\n\nAfter the tank, the coffee is dried like any other coffee. Whole cherries can be dried straight away, as in natural coffee.7 Other trials wash the coffee to stop fermentation before drying.5,7 In the trials, coffee was dried on raised beds or racks to about 11–12% moisture.1,4,5,6\n\nProducers often name the result after both steps, for example “anaerobic natural” or “anaerobic washed”. One 2026 study called anaerobic fermentation the most common of the newer processing methods.7\n\n## Effect on the cup and the risks\n\nMost anaerobic coffees in the trials scored as specialty coffee. In one trial, all SIAF coffees scored 84 points or more.1 Some trials found gains over conventional coffee, but a 2025 Colombian trial did not.2,4,5 In that trial, sealed tanks gave more tropical fruit notes, and whole cherries in open tanks more citrus and berry notes.2\n\nLong fermentation has costs. The seed is alive, and it can die in the tank.2 Mycotoxin-producing fungi became more common as fermentation went on.3 One open-tank sample had a phenolic defect.2\n\n\n\n## Oxygen, carbon dioxide and the vessel\n\n**How the oxygen disappears.** In SIAF, nothing is pumped in or out. Microorganisms consume the available oxygen as part of their metabolism,2 and CO₂ from microbial metabolism builds up gradually, limiting oxygen and favouring facultative microorganisms.4 The vessel is closed with an airlock that lets CO₂ escape and keeps air out.1,5,6 None of the five SIAF trials opened for this entry that describe their vessels reports a measured oxygen concentration;1,4,5,6,10 the oxygen status is inferred from the set-up.\n\n**Air removed or replaced.** In a 2026 trial in Yunnan, China, Catimor cherries were sealed in plastic bags and the air was removed by vacuum; the bags were kept closed for 7 days.7 In a 2022 trial with Conilon (*C. canephora*), the air was removed from 40 × 50 cm bags with a vacuum sealer, CO₂ of 99.9% purity was injected, and the bags were held at 18, 28 or 38 °C for 24–120 hours.8 The authors called this carbonic maceration.8 A 2023 review describes carbonic maceration as intact coffee sealed in an enclosed space with oxygen expelled and CO₂ added.9 How this differs from the wine technique of the same name is treated under carbonic maceration; adding starter cultures or other ingredients is treated under co-fermentation.\n\n**Open versus sealed.** A 2025 Colombian trial compared open containers (semi-anaerobic, SA) and sealed hermetic tanks with an airlock (SIAF), each with whole cherries and with pulped coffee.2 It used 30-litre high-density polyethylene bioreactors in a room held at 20 ± 1 °C and opened one bioreactor every 24 hours for up to 192 hours.2\n\n**Layers in a tall tank.** In a tank 82 cm high and 47 cm wide, holding about 175 kg of pulped coffee in 30 L of water, samples from taps at the bottom, middle and top differed significantly in their metabolites.1 Lactic acid at 72 hours reached 5.06, 4.70 and 2.18 g/L in the bottom, middle and top layers.1 The authors suggest stirring the tank to even out microbial metabolism.1\n\n## Microbial pathways and metabolites\n\nThe groups of microbes involved are described under microorganisms. This section covers what the closed tank changes.\n\n**Lactic fermentation.** Lactic acid was the most abundant acid in several trials that measured it.4,5,6 In the Colombian comparison, SIAF notably promoted glucose consumption and lactic acid production compared with open tanks.2 After 180 hours of SIAF without yeast inoculation and drying, natural coffee held 13.20 g/kg lactic acid; none was found at the start.4\n\n**Alcoholic fermentation.** Ethanol accumulates in the seed with time: after 72 hours of SIAF, beans held about six times more ethanol than after 24 hours.1 Total alcohols were highest in whole cherries fermented without water, at 62.05 mg/g after 96 hours.5\n\n**Acetic acid.** In one SIAF trial, acetic acid appeared after 18 hours, although acetic acid bacteria stayed below 1% of the bacterial reads; the authors attribute it to lactic acid bacteria.1 In the Colombian comparison, acetic acid bacteria (mainly *Acetobacter*) made up 20–57% of the community in open tanks and were less abundant in sealed ones.2 In another trial, acetic acid peaked at 48 hours and then fell.6\n\n**2,3-Butanediol.** In bacteria, 2,3-butanediol is made from pyruvate through acetolactate and acetoin; in the bacterium studied, it is one of several fermentative routes used under oxygen limitation to recycle reduced cofactors.12 In green coffee from a SIAF trial, 2,3-butanediol was detected at all three fermentation times.1 Which organisms made it in the coffee tank was not tested.\n\n**Undesirable acids.** Butyric acid is named among the undesirable products of prolonged fermentation.3 In two Brazilian SIAF trials, butyric and propionic acids were not detected in any treatment.5,6 A 2023 review lists sulphur and phenolic compounds and acetic acid among the defective volatiles that over-fermentation can produce.9\n\n**Esters and altitude.** A 2023 review states that fermentation favours the formation of esters such as ethyl acetate.9 In whole-cherry SIAF of one variety from one Brazilian region, coffee from lower altitudes had more volatile alcohols, and coffee from higher altitudes more esters, aldehydes and phenolics.10\n\n## Reference protocol\n\nThe table brings together the conditions reported in the trials opened for this entry. It shows the range used in research, not a recommended recipe.\n\n| Parameter | Values reported | Source |\n|---|---|---|\n| Vessel | 20 L buckets with lids10; 30 L HDPE bioreactors, 20 kg coffee2; 50 L bioreactors, 40 L coffee4; 200 L polypropylene bioreactors5,6; 82 × 47 cm tank, about 175 kg pulped coffee1 | Martinez et al., Peñuela-Martínez et al., Jimenez et al., Silva et al., Rocha et al., da Silva Vale et al. |\n| Closure | Screw cap or seal with airlock1,5,6; vacuum-sealed bags7; vacuum then CO₂ (99.9%)8 | da Silva Vale et al., Silva et al., Rocha et al., Fu et al., Gomes et al. |\n| Raw material | Whole cherries or pulped coffee2,5,6 | Peñuela-Martínez et al., Silva et al., Rocha et al. |\n| Water | None (solid state) or 30% of volume (submerged)5,6; 30 L for 175 kg pulped coffee1 | Silva et al., Rocha et al., da Silva Vale et al. |\n| Duration | 24–72 h1,6; 24–96 h5; 180 h4; up to 192 h2; 7 days7 | as cited |\n| Room temperature | 20 ± 1 °C (controlled room)2; in the sun at ambient temperature5 | Peñuela-Martínez et al., Silva et al. |\n| Mass temperature | 18 °C at start, 21.6–25 °C at 48 h (whole cherries)10; warming above room 0.5–3.7 °C sealed vs 5.1–17.6 °C open2 | Martinez et al., Peñuela-Martínez et al. |\n| °Brix at start | 23.65 (whole cherries); 7.33 (pulped)6 | Rocha et al. |\n| pH at start | 5.8 (pulped, with water)1; 5.06 whole, 5.21 pulped5; 5.11 whole, 5.08 pulped6 | da Silva Vale et al., Silva et al., Rocha et al. |\n| pH at end | 4.4 at 72 h1; 3.91–4.20 at 96 h5; lowest 3.70 at 72 h6 | da Silva Vale et al., Silva et al., Rocha et al. |\n| Monitoring | pH in the mass with a wireless meter; glucose with reflectometric test strips; temperature with dataloggers2; temperature and soluble solids followed until stable4 | Peñuela-Martínez et al., Jimenez et al. |\n| Stopping and drying | Washed, then raised bed to 11%5; raised beds to about 12%1,6; sun on suspended terraces to 11%4; whole cherries dried directly, or peeled coffee placed in water for 12 h before drying7 | as cited |\n\nMoisture values are given as reported (basis not stated).\n\n
\n
\n
Temperature differences (ΔT) reported for the fermenting mass in a room held at 20 °C, in open containers (semi-anaerobic) and sealed tanks with an airlock (SIAF), over up to 192 hours in a Colombian trial. The paper does not state the reference point of ΔT and gives one range per condition, without separating whole cherries and pulped coffee.2
\n
\n\n## Monitoring\n\nThe trials measured the same quantities as ordinary fermentation; the end-point methods are described under fermentation. A closed vessel needs a sampling route: the Brazilian tank had three taps at different heights,1 and the Colombian trial sacrificed one sealed bioreactor per sampling time.2\n\n- **pH** – measured directly in the fermenting mass with a wireless meter with temperature compensation.2 Starting and end values are in the reference protocol above.\n- **Temperature** – logged with dataloggers.2 Open containers warmed far more than sealed tanks; whole cherries in open containers stayed above 30 °C from 48 hours onward.2\n- **Sugars** – glucose by test strip,2 or soluble solids (°Brix) followed until stable.4\n- **Time** – trials used fixed times rather than a texture end point.1,2,5,6\n\n## Chemistry and sensory results\n\nResults from controlled trials, by the variable each one tested:\n\n- **Fermentation time.** In pulped coffee, beverages from 24, 48 and 72 hours of SIAF all scored 84 points or more; time changed aroma, flavour, acidity, body, overall and balance, but not sweetness, clean cup or uniformity.1\n- **Whole versus pulped, water versus none.** Scores ranged from 78.95 to 87.50; the highest, 87.50, came from whole cherries fermented without water for 96 hours, against 83.30 for the natural control and 83.60 for the pulped control.5\n- **Yeast starters in SIAF.** Over 180 hours, the best scores were 86.50 (pulped, *Torulaspora delbrueckii*) and 85.90 (natural, *Candida parapsilosis*); conventional coffees scored lowest, at 83.75 (natural) and 84.25 (pulped).4\n- **Open versus sealed.** Apart from one phenolic sample, scores ranged from 80.75 to 85.00 SCA points, and the treatments did not significantly affect sensory quality.2 Sealed tanks gave more tropical notes (yellow fruits, mango, pineapple, peach); whole cherries in open containers more citrus and berry notes.2\n- **Four methods on one lot.** In Peru, Catimor coffee scored 87.38 anaerobic, 85.75 natural, 84.00 washed and 82.88 honey, against 82.50 for a control sample (the abstract does not describe the anaerobic method or the control).11 In Yunnan, anaerobic-then-washed and anaerobic-then-natural coffees scored higher than conventional coffees for aroma, flavour, aftertaste, acidity, body and balance; washed coffee was cleaner.7\n- **CO₂-filled bags (Conilon).** Scores ranged from 78.51 to 83.25 and were highest at 38 °C for 120 hours; scores rose significantly with temperature at 24, 72 and 96 hours.8\n\n
\n
\n
Total cup scores reported in five trials of anaerobic fermentation (spans where the paper gives a range, dots for single scores), with conventionally processed coffee where reported. The Peruvian abstract does not describe its anaerobic method, and the Colombian span includes open containers. Trials used different varieties, protocols and panels, so the values are not directly comparable.2,4,5,8,11
\n
\n\n## Risks and defects\n\n- **Seed death.** In the Colombian trial, embryo viability stayed above 80% for up to 120 hours in sealed tanks but fell to as low as 25% by 192 hours in open tanks.2 After 180 hours of SIAF and drying, viability measured by the tetrazolium test fell from 85–87.5% in fresh coffee to 52.5–70%, while conventionally processed natural coffee kept 90%.4 The role of the living seed is covered under seed physiology.\n- **Physical quality.** The share of healthy beans was markedly lower under open-tank conditions, and the authors advise against open-tank fermentation of whole cherries beyond 24 hours.2\n- **Phenolic cup.** Of all samples in the Colombian trial, only one – whole cherries in an open tank for 72 hours – had a phenolic defect.2\n- **Fungi and ochratoxin A.** Longer fermentation brought a greater presence of mycotoxin-producing fungal genera. *Aspergillus* was most abundant in open tanks at 192 hours, and *Fusarium* predominated in sealed tanks of pulped coffee.3 Ochratoxin A stayed below 0.8 µg/kg, and SIAF showed no ochratoxin A production.3\n- **Uneven tanks.** Metabolites differed between the top, middle and bottom of a tall tank.1\n- **Off-flavours of over-fermentation.** Butyric acid, sulphur and phenolic compounds and excess acetic acid are named as undesirable products of prolonged or excessive fermentation.3,9 The classic fermentation defects are described under fermentation.\n\n## Open questions\n\n- **Real oxygen levels.** The SIAF trials opened for this entry describe the vessel but report no oxygen measurement, so how anaerobic the tanks become, and how fast, is not established.1,4,5,6,10\n- **Score gains.** Whether closed-vessel fermentation itself raises cup scores differs between trials (see Contested).2,4,5\n- **Origin of 2,3-butanediol.** It was detected in SIAF green coffee,1 but which organisms made it was not tested.\n- **Fungi in long runs.** *Fusarium* predominated in long SIAF runs even though ochratoxin A stayed low; what this means for safety and flavour is not settled.3\n\n## Key studies\n\n
\n

da Silva Vale et al., 2023 – SIAF in a tall tank. Follows pulped Catuaí coffee in Brazil through 24, 48 and 72 hours of SIAF, sampling three heights of one tank. Enterobacteria and filamentous fungi were prominent, lactic acid differed between layers, ethanol in the beans rose about sixfold from 24 to 72 hours, and every cup scored 84 points or more.

\n

Foods 12(1), 37 · doi:10.3390/foods12010037

\n
\n\n
\n

Peñuela-Martínez et al., 2025 – Oxygen availability and time. Compares open containers and sealed SIAF tanks, with whole cherries and pulped coffee, for up to 192 hours in Colombia. Sealed tanks stayed cooler, made more lactic acid and kept the embryo alive longer; cup scores did not differ significantly between treatments.

\n

Foods 14(17), 3001 · doi:10.3390/foods14173001

\n
\n\n
\n

Buitrago-Zuluaga et al., 2025 – Fungi in prolonged fermentation. Identifies filamentous fungi in the same long open and sealed fermentations. Mycotoxin-producing genera became more common with time, Aspergillus led in open tanks and Fusarium in sealed tanks of pulped coffee, while ochratoxin A stayed below 0.8 µg/kg.

\n

Applied Microbiology 5(4), 114 · doi:10.3390/applmicrobiol5040114

\n
\n\n
\n

Jimenez et al., 2023 – SIAF, yeasts and seed viability. Runs 180-hour SIAF of natural and pulped coffee with and without yeast starters in Brazil. Inoculated SIAF coffees scored up to 86.50 against 83.75–84.25 for conventional coffee, but every SIAF treatment lowered seed viability.

\n

Food Bioscience 51, 102218 · doi:10.1016/j.fbio.2022.102218

\n
\n\n
\n

Silva et al., 2024 – Whole or pulped, dry or submerged. Ferments Catiguá MG2 coffee in sealed 200-litre bioreactors for up to 96 hours. Whole cherries fermented without water gave the highest score, 87.50, and butyric and propionic acids were not detected.

\n

Foods 13(5), 653 · doi:10.3390/foods13050653

\n
" }