Cup taints and faults
The off-flavours cuppers find in the cup – potato, Rio, phenolic, fermented, mouldy, aged and roast-made – where each starts, which compounds mark them, and how cupping forms record and penalise them.
Across the chain: Science
In brief
- The Specialty Coffee Association’s Coffee Value Assessment (CVA) allows three sensory defects – potato, mouldy and phenolic – and deducts 4 points per defective cup.1
- The 2004 SCAA cupping form split defects into taints (intensity 2) and faults (intensity 4), multiplied by the number of cups affected.1,4
- 2,4,6-Trichloroanisole (TCA) is the main marker of the Rio defect; in coffee brew it was tasted at 1–2 parts per trillion in one study.9,10
- In roasted East African coffees, 2-isopropyl-3-methoxypyrazine (IPMP) separated potato-defective from clean samples.12
- In one mouldy-earthy Mexican coffee, geosmin, 2-methylisoborneol and TCA were the main causes of the off-flavour.11
Contents
What a taint or fault is#
A cup taint or fault is an off-flavour: a smell or taste that most tasters would call wrong for the coffee. The Specialty Coffee Association (SCA) standard for affective cupping defines a sensory defect as a smell or taste property “broadly seen as negative”.1 Sensory defects are found by tasting. Physical defects, such as black or insect-damaged beans, are found by looking at the green coffee.1,2
The two often go together, but not always. Some beans that spoil the cup look sound. The International Trade Centre (ITC) guide for coffee exporters says such phenolic beans cannot be detected by visual inspection.6 Counting physical defects is covered under processing defects and green coffee grading; this entry is about what reaches the cup.
The older cupping protocol of the Specialty Coffee Association of America (SCAA) named two grades of off-flavour. A taint is noticeable but not overwhelming, and usually found in the aroma.4 A fault is usually found in the taste and is either overwhelming or makes the sample unpalatable.4
How cupping forms record them#
The 2004 form. The cupper first classed a defect as a taint or a fault, then described it, for example “sour”, “rubbery”, “ferment” or “phenolic”.4 The number of cups affected was noted, and the intensity was recorded as 2 for a taint or 4 for a fault.4 The two numbers were multiplied and subtracted from the total score.4 At least five cups of each sample were prepared.4
The CVA (2024). The SCA’s Coffee Value Assessment limits sensory defects to three types: potato, mouldy and phenolic.1 Phenolic has a wide range of expression, which includes Rio-like and medicinal flavours.2 The cupper must name the defect specifically and avoid catch-all terms.1 The SCA’s rule of thumb: “if you can’t identify it as a defect, it’s probably not a defect.”2
On the form, the cupper ticks the cups that carry the defect and the defect type. If either field is missing, the coffee is not counted as defective.1 Every defective cup is also marked non-uniform, unless all cups are evenly defective.1 The score loses 4 points per defective cup and 2 per non-uniform cup.1 One defective cup among five therefore costs 6 points (our arithmetic). The full scoring rules are under cupping.
Some notes once treated as defects, such as overripe notes or isovaleric acid, are now called controversial by the SCA. Natural, honey and more intensely fermented coffees have brought them into specialty markets.2
Where the main off-flavours come from#
The sources link most off-flavours to the farm, the mill, drying or storage, and some to the roaster.3,6,8,11
- Potato. A flavour defect specific to East African coffee.13 Its main marker is 2-isopropyl-3-methoxypyrazine (IPMP).12 The ITC guide links it to bacterial infection after insects sting the cherries on the tree.6 Rwanda covers the field studies.
- Rio. A strong medicinal, phenolic or iodine-like off-flavour, known above all from Brazilian coffees.9,10 Its main marker is 2,4,6-trichloroanisole (TCA).9,10
- Phenolic and chemical. In a bulletin of Cenicafé, Colombia’s national coffee research centre, phenolic is a chemical, damp note. Its causes include pesticides, solvents and moulds.8
- Fermented, sour and onion. Sour, unpleasant and onion-like flavours come mainly from faults in processing, according to Cenicafé.8 Stinker beans are over-fermented beans from badly cleaned pulpers, tanks and channels.6
- Mouldy, musty and earthy. The cup smells of mould, earth and damp. Cenicafé names poor moisture control during drying as the cause.8
- Aged, baggy and woody. “Baggy” is a taint like the smell of a jute bag, found in coffee stored long under poor conditions.6 Cenicafé ties aged (reposado) coffee to poor storage.8 Green coffee storage gives the conditions.
- Roast-made. A roast that is too fast or too slow gives underdeveloped, burnt or baked notes.3 These are covered under roast defects.
How tasters find and confirm a taint#
Cupping can reveal defects carried by a few beans. Each cup is ground and brewed from its own portion of beans.2 The number of defective cups therefore shows how common the defective beans are in the sample.2 For the affective assessment, the SCA requires five cups per coffee, so that uniformity can be judged consistently.3
The pattern across cups is therefore a clue: a taint in one cup out of five points to a few affected beans (our reading).2 The ITC guide says Rio flavour usually runs through a whole lot, which makes sorting more or less impossible.6 It adds that a Rioy or phenolic taste cannot be hidden by blending.6
Sorting by eye may miss tainted beans. A 2024 study notes, citing earlier work, that the potato taste defect can occur in beans with and without visible insect damage.13
Cuppers also rule out the roast. A roast batch with underdeveloped, burnt or baked notes should not be cupped; the SCA asks for the roast to be repeated.2 Reference samples help tasters put a name to a taint. The World Coffee Research (WCR) Sensory Lexicon gives references for musty, mouldy and phenolic notes.5
Old and new names for off-flavours#
Trade words for off-flavours are old. In 1922, a trade handbook described a taster judging whether a brew was “Rioy”, “musty, groundy, woody, or grassy”, or “rank, hidey (sour), muddy, or bitter”.7 It wrote that Rio coffee has “a peculiar, rank flavor”.7 Coffees with a Rio flavour went only into the cheaper blends.7
The 2011 ITC guide keeps several of these words in its glossary of cup terms: baggy, earthy, musty or mouldy, fermented, onion, Rioy or phenolic, and woody.6 It describes Rioy or phenolic as a medicinal, slightly iodised taste.6
The WCR Sensory Lexicon of 2017 is value-neutral. It describes flavours without judging them and is not a tool for evaluating defective coffees.5 It still names many off-flavours: stale, papery and cardboard; musty/earthy, musty/dusty, mouldy/damp and phenolic; medicinal, rubber, petroleum and skunky; fermented and overripe.5 It has no potato attribute; peapod, a raw, beany green note, is the closest (our reading).5
The CVA descriptive form uses broader boxes. Off-notes are ticked as “sour/fermented” or under “other”: chemical, musty/earthy and woody.2 How these words fit the flavour wheel is explained under flavour language.
Reference values#
| Parameter | Value | Basis |
|---|---|---|
| CVA sensory defect types | Potato, mouldy, phenolic | SCA 104-20241 |
| CVA deductions | 4 points per defective cup; 2 per non-uniform cup; defective cups also count as non-uniform unless all cups are evenly defective | SCA 104-20241 |
| Cups per coffee, affective assessment | 5 | SCA 102-20243 |
| 2004 form, taint | Intensity 2, times the cups affected | SCAA protocol4 |
| 2004 form, fault | Intensity 4, times the cups affected | SCAA protocol4 |
| 2004 form, cups per sample | At least 5 | SCAA protocol4 |
| TCA in Rio green coffee | 1–100 ppb (nine samples) | One study9 |
| TCA lost in roasting | Less than 50% | One study9 |
| TCA threshold, coffee brew | 8 ppt by direct smell; 1–2 ppt by mouth | One study9 |
| Odour thresholds listed for water | 2-Methylisoborneol 2–20; geosmin 4–20; TCA 1–8; 2-methoxy-3-isopropylpyrazine 2–10; 2-methoxy-3-sec-butylpyrazine 1; 2-methoxy-3-isobutylpyrazine 2–20 ppt | Table of one study, compiled from earlier work; its text differs for some values11 |
| TCA in 22 green arabicas, five origins | Detected in the 4 Rio-defective Brazilian samples only | One study10 |
| Mouldy-earthy green coffee vs reference | Geosmin 1,000 vs 130 ppt; 2-methylisoborneol 100 vs <20 ppt; TCA 300 (text; table: <300) vs <50 ppt | One study, one pair of samples11 |
| IPMP in roasted coffee by cupper class | Clean 0–3.1; mild 1.6–72.4; medium 4.9–79.8; strong 4.1–529.9 ng/g | One study, 56 samples13 |
| Cup defects in Colombian warehouse analyses, 2013–2014 | 4.4% on average (base not stated); of these, fermented 41.9%, chemical and phenol 33.5%, aged 14.8%, mouldy 9.9% | Cenicafé8 |
Units as the sources give them: ppb (parts per billion) in green coffee is µg/kg; ppt (parts per trillion) is ng/kg in beans and about ng/L in water or brew. The potato values are per gram of roasted coffee.9,11,13
Marker compounds and odour thresholds#
The marker compounds of the main taints are smelled at a few parts per trillion.9,11 The thresholds below are for detection, in the medium named. Values in water come from the table of a 2001 study, headed “determined in water”, which compiles them from earlier work; the TCA values in coffee brew come from a 1990 study that added TCA to fresh brew.9,11
Data table
| Point | ppt (ng/L) | Source | |
|---|---|---|---|
| Water · 2-Methylisoborneol (MIB) | 2–20 | 2–20 | 11 |
| Water · Geosmin | 4–20 | 4–20 | 11 |
| Water · 2,4,6-Trichloroanisole (TCA) | 1–8 | 1–8 | 11 |
| Water · 2-Methoxy-3-isopropylpyrazine (IPMP) | 2–10 | 2–10 | 11 |
| Water · 2-Methoxy-3-sec-butylpyrazine | 1 | 1 | 11 |
| Water · 2-Methoxy-3-isobutylpyrazine | 2–20 | 2–20 | 11 |
| Coffee brew · TCA, direct smell | 8 | 8 | 9 |
| Coffee brew · TCA, by mouth | 1–2 | 1–2 | 9 |
Two cautions apply. The 2001 study gives other values in its text than in its table: 10–50 ppt for geosmin, 2–20 ppt for the isopropyl methoxypyrazine and below 10 ppt for 2-methylisoborneol.11 For TCA, its text gives 1–8 ppt “depending on the medium”, so the table heading “determined in water” may not hold for every value.11 A 2022 study cites thresholds for TCA in water from 0.03 to 1–2 ng/L, from other work.10 No study consulted for this entry measured thresholds for IPMP, geosmin or 2-methylisoborneol in brewed coffee.
The concentrations found in defective coffee sit far above these thresholds. Rio green coffee held 1–100 ppb TCA, that is 1,000–100,000 ppt, and roasting removed less than half.9 No study consulted for this entry measured how much of it passes into the cup.
Potato taste defect#
Marker. In roasted East African arabica grouped by severity into clean, mild, medium and strong classes, IPMP discriminated between defective and non-defective samples; 2-isobutyl-3-methoxypyrazine did not.12 In a follow-up with 56 samples, IPMP ranged from 0–3.1 ng/g in clean coffee to 4.1–529.9 ng/g in strongly defective coffee.13 The ranges of the defect classes overlap (our reading of the reported values).13 Two-dimensional gas chromatography found 359 compounds that changed significantly between the classes. Compounds with roasted, nutty, cocoa and fruity aromas were less abundant in the strongly defective samples.13
Cause. The 2024 study notes, citing earlier work, that the defect has been linked to an antestia bug, which it names as Antestiopsis orbitalis, and that it can also occur in beans without visible insect damage.13 The EPPO Global Database treats A. orbitalis and A. lineaticollis as synonyms of A. thunbergii; see Antestia bugs. In a 2025 experiment, mechanical damage to the berries alone caused the defect.14 The authors conclude that the bug is not absolutely required. They propose that any wound, including bug feeding, lets in a bacterium of the genus Pantoea, which they hold responsible for the defect and for IPMP.14 Field studies and control are covered under Rwanda.
Rio flavour and haloanisoles#
Marker. In a 1990 study of nine Rio-tainted green coffees from Brazil and Puerto Rico, TCA was identified as the most likely key compound.9 TCA added to fresh brew gave the same off-flavour as Rio coffee.9 The authors found 2,4,6-trichlorophenol (TCP), the probable precursor of TCA, in most samples, but no other chlorophenols or chloroanisoles.9 They described strong Rio coffee as smelling musty and cellar-like, rather than medicinal.9
Origin. The Rio beans were heavily infested with fungi, among them Aspergillus, Fusarium and Penicillium.9 A 2022 study proposes, drawing on findings in wine, that TCA forms when filamentous fungi methylate TCP.10 It screened 22 green arabicas from Brazil, Colombia, Ethiopia, Guatemala and Nicaragua. TCA was found only in the four Rio-defective Brazilian samples.10 The authors read this as strongly suggesting a possible microbial origin.10 The ITC guide instead links Rio flavour to very high humidity during the growing season (see Contested).6
Extent. The 1990 study states that about 20% of Brazil’s coffee production showed the Rio defect.9 The 2022 study repeats the figure.10
Mouldy, musty and earthy taints#
A 2001 study compared a mouldy-earthy green arabica from Chiapas, Mexico, with a clean coffee from the same origin.11 Using gas chromatography with sniffing, it found six compounds behind the off-flavour.11 Geosmin, 2-methylisoborneol and TCA were the main culprits; three methoxypyrazines added to an earthy, green undertone.11 Geosmin stood at 1,000 ppt, about eight times the reference.11 The authors tentatively linked the defect to post-harvest drying.11 The study rests on one pair of samples.11
TCA thus marks both Rio and mouldy-earthy taints.9,11 The WCR lexicon reflects the overlap: its musty/dusty attribute lists a geosmin reference, and its mouldy/damp attribute a 2,4,6-trichloroanisole reference.5
Moulds that taint coffee can also form ochratoxin A. That food-safety question is covered under ochratoxin A and contaminants and food safety.
Cup defects in Colombian practice#
Cenicafé, Colombia’s coffee research centre, describes four groups of cup defects. Fermented coffee has a sour, unpleasant aroma, acidity and flavour, with onion-like notes.8 Phenolic is more chemical and damp, and counts as a contaminated product.8 The bulletin attributes the phenol defect both to pesticides and solvents and to moulds such as Aspergillus and Penicillium.8 Mouldy coffee smells and tastes of mould, earth and damp; aged coffee has strong aged aromas and high, unpleasant bitterness.8
Data table
| % of cup defects | Source | |
|---|---|---|
| Fermented (fermento) | 41.9% | 8 |
| Chemical and phenol (químico y fenol) | 33.5% | 8 |
| Aged (reposo) | 14.8% | 8 |
| Mouldy (moho) | 9.9% | 8 |
In Colombian warehouse analyses of 2013 and 2014, cup defects averaged 4.4%; the bulletin does not say whether this is a share of samples or of coffee. Fermented made up 41.9% of these, chemical and phenol 33.5%, aged 14.8% and mouldy 9.9%.8 It does not give the number of samples.8
Aged coffee and roast faults#
Aged coffee. In a German storage study, green coffee kept warm and moist gained vinylphenols and methyl esters. Aroma-active breakdown products of unsaturated fatty acids did not form.15 Whether fat oxidation causes the aged flavour of green coffee is therefore open; green coffee storage sets out the evidence. The WCR lexicon names the family stale/papery and gives a papery reference made from soaked paper filters.5 Staling of roasted coffee is a separate process, covered under freshness and storage.
Roast faults. SCA 102-2024 defines roasting problems as roasts outside the target range or roasted too fast or too slow; they give underdeveloped, burnt or baked notes.3 They are not among the three CVA defect types.1 In a 2016 study, one marker compound was chosen for each roast defect: indole for light, 4-ethyl-2-methoxyphenol for scorched, phenol for dark, maltol for baked and 2,5-dimethylfuran for underdeveloped.16 A scorched or dark roast can thus carry phenolic compounds without any fault in the green coffee (our reading).16 Roast defects covers causes and cures.
Open questions#
- Thresholds in coffee. No study consulted for this entry measured thresholds for IPMP, geosmin or 2-methylisoborneol in brewed coffee; only TCA has a measured threshold in brew.9
- Phenolic markers. Apart from TCA in Rio coffee, no study consulted for this entry identified a marker compound for phenolic or medicinal taints.
- One bean, one cup. No study consulted for this entry measured how many affected beans it takes to make a cup defective.
- Potato cause. Whether antestia feeding, other wounds or bacteria alone start the defect is not settled.13,14
- Aged flavour chemistry. Which compounds give the aged, baggy or woody notes of old green coffee is not established in the sources consulted.15
Key studies#
Spadone, Takeoka and Liardon, 1990 – TCA behind the Rio defect. Nine Rio-tainted green coffees were analysed against a clean Santos. TCA was found in all of them at 1–100 ppb, survived roasting by more than half, and reproduced the off-flavour when added to fresh brew, where tasters found it at 1–2 ppt by mouth.
Cantergiani et al., 2001 – Compounds of a mouldy-earthy taint. A defective and a clean green coffee from Chiapas were compared by gas chromatography with sniffing. Geosmin, 2-methylisoborneol and TCA were the main causes, with three methoxypyrazines in a minor role; the authors tentatively linked the taint to drying.
Cain et al., 2021 – IPMP and potato severity. Roasted East African coffees, grouped by severity from clean to strong potato defect, were analysed. IPMP separated defective from clean samples, while its relative 2-isobutyl-3-methoxypyrazine did not.
Romano et al., 2022 – Screening green coffee for TCA. A fast mass-spectrometry method screened 22 green arabicas from five origins. TCA appeared only in the four Rio-defective Brazilian samples, which the authors read as pointing to a possible microbial origin.
Bigirimana et al., 2025 – Potato defect without the bug. Coffee berries were damaged mechanically, without antestia bugs. The defect still appeared, and the authors propose that any wound lets in the bacterium they hold responsible.
- Origin of the Rio flavour: the ITC guide links it to very high humidity during the growing season6; laboratory studies found heavy fungal infestation and the precursor 2,4,6-trichlorophenol in Rio beans, and a 2022 study reads its own results as strongly suggesting a possible microbial origin.9,10
- Antestia bug and potato taste defect: the ITC guide describes bacterial infection after insects sting the cherries6; a 2025 experiment caused the defect by mechanical damage alone, without the bug, and its authors conclude that the bug is not absolutely required14; a 2024 study notes, citing earlier work, that the defect can occur in beans with and without visible insect damage.13 Field evidence for a link with bug density is set out under Rwanda.
- Fermented notes as defects: the SCA’s 2024 CVA document calls overripe notes and isovaleric acid controversial now that natural, honey and more intensely fermented coffees reach specialty markets2; the 2011 ITC guide states that for most roasters there is no such thing as “just a little ferment”.6
Coffearium (2026). Cup taints and faults. Coffearium – The open coffee compendium. Retrieved 17 November 2026, from https://coffearium.org/entries/cup-taints-and-faults/
@misc{coffearium2026-cup-taints-and-faults,
author = {{Coffearium}},
title = {{Cup taints and faults}},
year = {2026},
howpublished = {Coffearium -- The open coffee compendium},
url = {https://coffearium.org/entries/cup-taints-and-faults/},
urldate = {2026-11-17},
note = {Retrieved 17 November 2026}
}Entries are updated when the evidence changes; the retrieval date tells your readers which version you used. Changes are listed in the Change log.
Change log
- – The 2024 study’s sentence on the defect in beans with and without insect damage is now given as cited from earlier work, not as its finding; “not strictly required” now reads “not absolutely required”, as in the 2025 abstract; the EPPO synonymy of A. orbitalis with A. thunbergii is noted with a pointer to Antestia bugs. Reason: Found while verifying Antestia bugs: the sentence sits in the 2024 paper’s introduction and cites its reference 9.
- – Colombian warehouse figure: “4.4% of samples” changed to an average of 4.4% cup defects with the base not stated. Reason: Cenicafé AVT 461 gives “4,4% de defectos en taza” as an average without stating whether it is a share of samples or of coffee.
- – Potato studies: “sorted by cuppers” changed to “grouped by severity” for the 2021 study; aroma decrease narrowed to the strongly defective samples; “hundreds” replaced by the 359 compounds reported. Reason: The 2021 abstract does not say who classed the samples; the 2024 study reports the decrease for strong PTD samples and 359 significant analytes.
- – 2022 TCA study: “microbial origin” changed to “possible microbial origin” (body, contested, key study). Reason: The paper writes that the result “strongly suggests the possible microbiological origin”.
- – Odour thresholds: added the 2001 study’s text values for 2-methylisoborneol (below 10 ppt) and TCA (1–8 ppt, depending on the medium), the TCA <300 table value, and that the table compiles values from earlier work. Reason: Internal differences between text and Table 3 of the 2001 study must be visible to readers, not only in the source note.
- – Antestia contested item reworded; the bug species attributed to the 2024 study; “taint” removed from linkTerms. Reason: The 2025 abstract concludes the bug is not strictly required, it does not refute a role; rwanda.md names Antestiopsis thunbergii; the bare word “taint” would link water and cargo taints in other entries to this entry.
Sources
- A/BSpecialty Coffee Association (2024). SCA Standard 104-2024 Coffee Value Assessment: Affective Assessment. Specialty Coffee Association.Quoted passage: a definition; SCA affective cupping standard (online September 2024 version); section 5.4.1 defects, 7.1 deductions.
- A/BSpecialty Coffee Association (writing team Fernández-Alduenda, M.; Ionescu, K. E.; Giuliano, P.; Rugolo, J.; Delrue, R.) (2024). A System to Assess Coffee Value: Understanding the Specialty Coffee Association’s Coffee Value Assessment (June 2024). Specialty Coffee Association.Quoted passage: a description; SCA explanatory document on the CVA, June 2024; section 8.4 defects and uniformity.
- A/BSpecialty Coffee Association (2024). SCA Standard 102-2024 Coffee Value Assessment: Sample Preparation and Tasting Mechanics. Specialty Coffee Association.Quoted passage: a recommendation; SCA sample-preparation standard (online version); cups per coffee, roasting problems.
- A/BSpecialty Coffee Association of America (2015). SCAA Protocols: Cupping Specialty Coffee. Specialty Coffee Association of America (revision of 16 December 2015).Quoted passage: a recommendation; SCAA cupping protocol, revision of 16 December 2015; defects section and cups per sample.
- BWorld Coffee Research (2017). World Coffee Research Sensory Lexicon: Unabridged Definition and References, Version 2.0. World Coffee Research.Quoted passage: a description; WCR lexicon 2.0 (local copy 2.0.1): attribute definitions and reference preparations; value-neutral.
- AInternational Trade Centre (ITC) (2011). The Coffee Exporter's Guide, Third Edition. International Trade Centre (UNCTAD/WTO), Geneva.Quoted passage: a description; ITC guide 2011, chapter 12 glossaries and section on phenolic, Rio and fermented; causes stated without data.
- BUkers, W. H. (1922). All About Coffee. The Tea and Coffee Trade Journal Company, New York (public domain; Project Gutenberg eBook 28500).Quoted passage: a description; Trade handbook, 1922; chapter XXIV cup-testing vocabulary and chapter XXV blending.
- APuerta-Quintero, G. I. (2015). Buenas prácticas para la prevención de los defectos de la calidad del café: fermento, reposado, fenólico y mohoso (Avances Técnicos 461). Cenicafé.Quoted passage: data; Cenicafé bulletin 2015; Almacafé warehouse reports, 5 months of 2013–2014, averages, base and sample number not stated.
- ASpadone, J.-C.; Takeoka, G.; Liardon, R. (1990). Analytical investigation of Rio off-flavor in green coffee. Journal of Agricultural and Food Chemistry 38(1), 226–233. doi:10.1021/jf00091a050 · other copiesQuoted passage: the authors’ own result; Nine Rio-tainted green coffees (Brazil, Puerto Rico) vs one Santos reference; brew thresholds by triangle test, 16 tasters.
- ARomano, A.; Navarini, L.; Lonzarich, V.; Bogialli, S.; Pastore, P.; Cappellin, L. (2022). 2,4,6-Trichloroanisole Off-Flavor Screening in Green Coffea arabica by a Novel Vocus NO+ CI-MS Method: A Study on Green Coffee from Different Geographical Origins. Journal of Agricultural and Food Chemistry 70, 11412–11418. doi:10.1021/acs.jafc.2c03899 · other copiesQuoted passage: the authors’ own result; 22 green arabicas from Brazil, Colombia, Ethiopia, Guatemala, Nicaragua; 4 Rio-defective Brazilian samples; thresholds cited.
- ACantergiani, E.; Brevard, H.; Krebs, Y.; Feria-Morales, A.; Amadò, R.; Yeretzian, C. (2001). Characterisation of the aroma of green Mexican coffee and identification of mouldy/earthy defect. European Food Research and Technology 212, 648–657. doi:10.1007/s002170100305 · other copiesQuoted passage: the authors’ own result; One mouldy-earthy and one reference green arabica, Chiapas, Mexico; GC-olfactometry; Table 3 thresholds compiled from earlier work.
- ACain, C. N.; Haughn, N. J.; Purcell, H. J.; Marney, L. C.; Synovec, R. E.; Thoumsin, C. T.; Jackels, S. C.; Skogerboe, K. J. (2021). Analytical Determination of the Severity of Potato Taste Defect in Roasted East African Arabica Coffee. Journal of Agricultural and Food Chemistry 69(7), 2253–2261. doi:10.1021/acs.jafc.1c00605 · other copiesQuoted passage: the authors’ own result; Roasted East African arabica grouped by potato-defect severity; abstract only, sample number not stated there.
- ACain, C. N.; Gaida, M.; Stefanuto, P.-H.; Focant, J.-F.; Synovec, R. E.; Jackels, S. C.; Skogerboe, K. J. (2024). Investigating sensory-classified roasted arabica coffee with GC × GC-TOFMS and chemometrics to understand potato taste defect. Microchemical Journal 196, 109578. doi:10.1016/j.microc.2023.109578 · other copiesQuoted passage: the authors’ own result; 56 roasted arabica samples (14 clean, 11 mild, 13 medium, 18 strong) classed by sensory panel; GC×GC-TOFMS.
- ABigirimana, J.; Mubashankwaya, I.; Niyibigira, T.; Nzeyimana, F. (2025). Mechanical Damage Causes Potato Taste Defect in Coffee Without Antestia Bug Infestations: Evidence from Experimental Data. International Journal of Fruit Science 25(1), 367–377. doi:10.1080/15538362.2025.2548292 · other copiesQuoted passage: the authors’ own result; Field experiment (abstract only): caged antestia bugs vs mechanical damage vs undamaged controls.
- AScheidig, C.; Czerny, M.; Schieberle, P. (2007). Changes in Key Odorants of Raw Coffee Beans during Storage under Defined Conditions. Journal of Agricultural and Food Chemistry 55, 5768–5775. doi:10.1021/jf070488o · other copiesQuoted passage: the authors’ own result; Green coffee stored up to 9 months under defined conditions (e.g. 40 °C, 13.5% water content); lab study.
- AYang, N.; Liu, C.; Liu, X.; Degn, T. K.; Munchow, M.; Fisk, I. D. (2016). Determination of volatile marker compounds of common coffee roast defects. Food Chemistry 211, 206–214. doi:10.1016/j.foodchem.2016.04.124 · other copiesQuoted passage: the authors’ own result; One origin roasted with six time-temperature profiles; headspace SPME GC-MS, 37 key aroma compounds.
Tier A · authoritative · B · trade and expert · C · general. Supporting quotes for every source are in the data version of this entry. If a publisher’s site refuses access, try the doi link or “other copies”.
The literature for this entry was last searched on 12 Nov 2026. A statement that no source consulted for this entry shows something refers to the sources above; newer or missing work may exist.
Discussion
A space for readers to share experience from the farm, the roastery or the bar, ask each other and offer other views. The Coffearium team does not reply here. Please keep to the community rules; to change the entry itself, use Suggest a change.