{
  "id": "processing-and-flavour-chemistry",
  "url": "https://coffearium.org/entries/processing-and-flavour-chemistry/",
  "license": "CC-BY-4.0",
  "title": "Processing and flavour chemistry",
  "summary": "How washed, natural and other processing methods change the sugars, amino acids and acids in green coffee – the precursors that roasting turns into aroma, colour and flavour in the cup.",
  "domain": "post-harvest-processing",
  "kind": "Concept",
  "stage": "process",
  "topics": [
    "science"
  ],
  "topicSections": {},
  "lang": "en",
  "status": "verified",
  "reviewer": "Coffearium verification pass",
  "lastVerified": "2026-11-13T00:00:00.000Z",
  "literatureSearched": "2026-11-13T00:00:00.000Z",
  "updated": "2026-11-16T00:00:00.000Z",
  "added": "2026-11-13T00:00:00.000Z",
  "brief": [
    "Washed arabica holds less glucose and fructose than natural arabica processed in parallel; sucrose, the main sugar, was not significantly affected.<sup><a href=\"#src-s5\">5</a></sup>",
    "The seed is alive during processing and its own metabolism shapes the precursors; the extent and nature of germination depend on the processing method.<sup><a href=\"#src-s3\">3</a></sup>",
    "GABA (γ-aminobutyric acid), a plant stress metabolite, was always markedly higher in natural than in washed arabica.<sup><a href=\"#src-s1\">1</a></sup>",
    "Lactic acid, acetic acid and mannitol made by microbes in the mucilage were also found inside the seed.<sup><a href=\"#src-s8\">8</a></sup>",
    "According to a 2023 review, sugars and amino acids feed the Maillard reaction and Strecker degradation in the roast, linking processing to aroma.<sup><a href=\"#src-s15\">15</a></sup>"
  ],
  "quickFacts": [
    {
      "label": "Main precursors",
      "value": "Sugars, free amino acids, chlorogenic acids, trigonelline"
    },
    {
      "label": "Glucose and fructose",
      "value": "Low in washed, higher in natural arabica"
    },
    {
      "label": "Sucrose",
      "value": "Not significantly changed by washed or natural processing (one study)"
    },
    {
      "label": "GABA",
      "value": "Markedly higher in natural arabica"
    },
    {
      "label": "Germination peak",
      "value": "About 2 days (washed); about 1 week (natural)"
    },
    {
      "label": "Drying heat",
      "value": "Natural seeds dried with 60 °C air did not germinate (one trial)"
    }
  ],
  "aka": [
    "Post-harvest flavour chemistry",
    "Processing chemistry of green coffee"
  ],
  "linkTerms": [
    "processing chemistry",
    "post-harvest chemistry",
    "flavour precursor profile",
    "processing and flavour"
  ],
  "related": [
    "green-coffee-composition",
    "seed-physiology",
    "washed-process",
    "natural-process",
    "honey-process",
    "pulped-natural",
    "fermentation",
    "microorganisms",
    "mucilage",
    "drying",
    "maillard-reaction",
    "strecker-degradation",
    "acids-in-coffee",
    "chlorogenic-acids",
    "sweetness-in-coffee",
    "body-and-mouthfeel",
    "green-coffee-storage",
    "cherry-to-green-conversion"
  ],
  "contested": [
    "Germination in natural processing: a 2021 review states that, in contrast to the wet method, germination is inhibited in the dry method, citing a 1980 study on inhibition of coffee seed germination by the endocarp<sup><a href=\"#src-s7\">7</a></sup>; a 2007 study measured germination markers during both methods, peaking about two days into wet processing and about one week into dry processing, with a markedly lower maximum frequency of 4C nuclei, a measure of cell-cycle activity, in dry-processed seeds.<sup><a href=\"#src-s2\">2</a></sup>",
    "Chlorogenic acids and trigonelline after processing: a comparison of 17 Brazilian arabica cultivars and progenies found significantly more of both after wet than after semi-dry processing, although for chlorogenic acids the difference vanished when the four pure arabica cultivars were compared alone<sup><a href=\"#src-s11\">11</a></sup>, while a 2021 review concludes that dry, wet and semi-dry processing did not cause significant changes in caffeine, chlorogenic acids or trigonelline.<sup><a href=\"#src-s7\">7</a></sup>",
    "Sucrose and processing: a comparison of wet and dry processing found sucrose not significantly affected<sup><a href=\"#src-s5\">5</a></sup>, while Brazilian wet-processed coffees held less sucrose than the same cultivars processed semi-dry – a difference not found in the four pure arabica cultivars alone.<sup><a href=\"#src-s11\">11</a></sup> The two studies compare different pairs of methods.",
    "Why washed coffee has less glucose and fructose: a drying study without process water attributes the gap at least in part to seed metabolism<sup><a href=\"#src-s6\">6</a></sup>, while a field study adds osmotic loss into the water during soaking.<sup><a href=\"#src-s8\">8</a></sup> The washed process and seed physiology entries set out both positions."
  ],
  "corrections": [
    {
      "date": "2026-11-13T00:00:00.000Z",
      "change": "Added that almost all sucrose is degraded in roasting (new source 18) and that no study consulted measured how much of the extra hexoses of natural coffee reach the cup; pointer to anaerobic and inoculated fermentations.",
      "reason": "External review: the sweetness statement of the 2023 review needed the roasting caveat."
    },
    {
      "date": "2026-11-13T00:00:00.000Z",
      "change": "The contested item on germination in natural processing now says that the 2021 review cites a 1980 endocarp study for its statement, not the 2007 germination study, and adds the lower 4C-nucleus frequency the 2007 study found in dry processing.",
      "reason": "The review's reference list gives a 1980 study as the source of the statement."
    },
    {
      "date": "2026-11-13T00:00:00.000Z",
      "change": "The 3.3–20.9% free amino acid figure is now attributed to the 2004 conference paper the review cites, with a note that the reference quantity is not stated.",
      "reason": "The review cites the figure; it is not the review's own finding."
    },
    {
      "date": "2026-11-13T00:00:00.000Z",
      "change": "The caffeine paragraph now gives the review's full conclusion (caffeine, chlorogenic acids and trigonelline), sets it against the 2010 Brazilian comparison in a new contested item, and no longer says the review found this across the studies it compiles.",
      "reason": "The review states this in its conclusions and cites a study that found a fall in caffeine; the 2010 comparison found differences for chlorogenic acids and trigonelline."
    },
    {
      "date": "2026-11-13T00:00:00.000Z",
      "change": "The opening no longer says green coffee has almost none of the aroma of roasted coffee; the sensory statements quote the 2023 review's wording (“cleaner and fruity”, “rich body”) instead of comparatives.",
      "reason": "The review reports about 300 volatile compounds in green coffee and does not say “fruitier” or “more body”."
    },
    {
      "date": "2026-11-13T00:00:00.000Z",
      "change": "Sugar and GABA values now taken from the original studies (full texts read): sugars read from the bar charts of Knopp et al. 2006 in g per 100 g dry matter, semi-dry processing confirmed as one of three parallel treatments; GABA from Table 1 of Bytof et al. 2005, the Tanzanian pair marked as laboratory model processing.",
      "reason": "The owner supplied both papers."
    },
    {
      "date": "2026-11-13T00:00:00.000Z",
      "change": "Chart captions and the text now state that the sugar and GABA values are compiled in a 2021 review, that the original studies' full texts were not consulted, and that “±” in the review is read as “about”.",
      "reason": "The values could be checked only against the review's Table 1."
    },
    {
      "date": "2026-11-16T00:00:00.000Z",
      "change": "Teran 2024: the sentence on Bourbon and Pacamara is replaced by the interactions the paper reports (natural Typica higher, washed Bourbon lower).",
      "reason": "The abstract's claim for Bourbon and Pacamara is not matched by a reported test in the results."
    },
    {
      "date": "2026-11-16T00:00:00.000Z",
      "change": "Duarte et al. 2010: the full text is now used. The wet vs semi-dry differences in chlorogenic acids and sucrose are reported as absent in the four arabica cultivars compared alone; the authors attribute them to the Timor Hybrid progenies.",
      "reason": "The source had been read as an abstract only; the results section qualifies the abstract's overall finding."
    }
  ],
  "sources": [
    {
      "id": "s1",
      "title": "Influence of processing on the generation of γ-aminobutyric acid in green coffee beans",
      "authors": "Bytof, G.; Knopp, S.-E.; Schieberle, P.; Teutsch, I.; Selmar, D.",
      "year": 2005,
      "publisher": "European Food Research and Technology 220(3–4), 245–250",
      "url": "https://portal.fis.tum.de/en/publications/influence-of-processing-on-the-generation-of-%CE%B3-aminobutyric-acid-/",
      "doi": "10.1007/s00217-004-1033-z",
      "tier": "A",
      "quote": "The GABA content of unwashed Arabica beans (green coffee produced by the dry processing method) was always markedly higher than that of washed Arabicas (wet processing method) as well as that of untreated seeds.",
      "quotes": [
        {
          "text": "Comparison of the concentrations of g-aminobutyric acid (GABA) and glutamic acid in differently processed green coffees from various origins.",
          "for": "Table 1: GABA in nmol per seed (Brazil batch 93 wet, 1,009 dry; Tanzania B model 140 wet, 1,860 dry)"
        }
      ],
      "scope": {
        "claim": "data",
        "context": "Arabica; Brazil (var. Acaiá) batch processing at one farm, Tanzania (var. KP 345) batch and laboratory model processing; GABA per seed in Table 1; full text read (owner's copy)",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s2",
      "title": "Transient occurrence of seed germination processes during coffee post-harvest treatment",
      "authors": "Bytof, G.; Knopp, S.-E.; Kramer, D.; Breitenstein, B.; Bergervoet, J. H. W.; Groot, S. P. C.; Selmar, D.",
      "year": 2007,
      "publisher": "Annals of Botany 100(1), 61–66",
      "url": "https://academic.oup.com/aob/article/100/1/61/2763338",
      "doi": "10.1093/aob/mcm068",
      "tier": "A",
      "quote": "The highest germination activity occurred 2 d after the onset of wet processing, whereas the corresponding maximum in the course of dry processing appeared about 1 week after the start of post harvest treatment.",
      "quotes": [
        {
          "text": "the maximum frequency of 4C nuclei was altogether markedly lower in the dry-processed beans",
          "for": "lower 4C nuclei in dry processing"
        }
      ],
      "scope": {
        "claim": "finding",
        "context": "Arabica cv. Acaiá, Brazil; germination markers (ICL, β-tubulin, 4C nuclei) during wet and dry processing; full text consulted",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s3",
      "title": "Germination of coffee seeds and its significance for coffee quality",
      "authors": "Selmar, D.; Bytof, G.; Knopp, S.-E.; Breitenstein, B.",
      "year": 2006,
      "publisher": "Plant Biology 8(2), 260–264",
      "url": "https://doi.org/10.1055/s-2006-923845",
      "doi": "10.1055/s-2006-923845",
      "tier": "A",
      "quote": "The extent and nature of the germination processes depend on the processing method.",
      "quotes": [],
      "abstractOnly": true,
      "scope": {
        "claim": "finding",
        "context": "Arabica; germination markers during standard wet and dry post-harvest treatment; abstract only consulted",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s4",
      "title": "Stress Metabolism in Green Coffee Beans (Coffea arabica L.): Expression of Dehydrins and Accumulation of GABA during Drying",
      "authors": "Kramer, D.; Breitenstein, B.; Kleinwächter, M.; Selmar, D.",
      "year": 2010,
      "publisher": "Plant and Cell Physiology 51(4), 546–553",
      "url": "https://academic.oup.com/pcp/article-abstract/51/4/546/1826697",
      "doi": "10.1093/pcp/pcq019",
      "tier": "A",
      "quote": "A decrease in water potential of the still living coffee seeds induces massive drought stress responses.",
      "quotes": [],
      "scope": {
        "claim": "finding",
        "context": "Arabica seeds during drying; dehydrin expression and GABA; drought stress interpretation is the authors'",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s5",
      "title": "Influence of processing on the content of sugars in green Arabica coffee beans",
      "authors": "Knopp, S.; Bytof, G.; Selmar, D.",
      "year": 2006,
      "publisher": "European Food Research and Technology 223, 195–201",
      "url": "https://link.springer.com/article/10.1007/s00217-005-0172-1",
      "doi": "10.1007/s00217-005-0172-1",
      "tier": "A",
      "quote": "Sucrose, the major low molecular sugar in green coffee beans, is not significantly affected by coffee processing.",
      "quotes": [
        {
          "text": "The experiments were undertaken in parallel (wet, dry and semi-dry processing) and in three repetitions (semi-dry process only in two repetitions).",
          "for": "three methods in parallel"
        },
        {
          "text": "the wet and mucilaginous parchment coffee beans were immediately dried on a separate plot of the sun terrace.",
          "for": "semi-dry = pulped, no tank fermentation"
        },
        {
          "text": "Concentration of low molecular sugars in differentially processed green coffees.",
          "for": "Fig. 2 (g/100 g dry matter)"
        }
      ],
      "scope": {
        "claim": "data",
        "context": "Arabica var. Acaiá, one farm in Minas Gerais, Brazil, 2002; wet, dry and semi-dry processing of the same batch in parallel; sugars in g/100 g dry matter shown as bar charts (Fig. 2); laboratory model processing with Tanzanian and Mexican coffee (Fig. 3); full text read (owner's copy)",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s6",
      "title": "Influence of drying on the content of sugars in wet processed green Arabica coffees",
      "authors": "Kleinwächter, M.; Selmar, D.",
      "year": 2010,
      "publisher": "Food Chemistry 119, 500–504",
      "url": "https://doi.org/10.1016/j.foodchem.2009.06.048",
      "doi": "10.1016/j.foodchem.2009.06.048",
      "tier": "A",
      "quote": "Within the first day of drying, fructose concentration decreased down to about 40 mg/100 g d.w. in the case of continuous drying, and to about 30 mg/100 g d.w. in beans dried with pauses (Fig. 1).",
      "quotes": [],
      "scope": {
        "claim": "finding",
        "context": "Wet-processed (mechanically demucilaged, dry-fermented) arabica dried at 30 °C, continuous or with pauses; mg/100 g dry weight",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s7",
      "title": "From Plantation to Cup: Changes in Bioactive Compounds during Coffee Processing",
      "authors": "Bastian, F.; Hutabarat, O. S.; Dirpan, A.; Nainu, F.; Harapan, H.; Emran, T. B.; Simal-Gandara, J.",
      "year": 2021,
      "publisher": "Foods 10(11), 2827",
      "url": "https://www.mdpi.com/2304-8158/10/11/2827",
      "doi": "10.3390/foods10112827",
      "tier": "A",
      "tierNote": "Peer-reviewed review. Its Table 1 compiles values from primary studies; the sugar values used here come from Knopp et al. (2006) and the GABA values from Bytof et al. (2005), both cited in this entry.",
      "quote": "Table 1 shows the accumulation of γ-aminobutyric acid (GABA) in the dry process, whereas only small amounts are accumulated in the wet process",
      "quotes": [
        {
          "text": "In contrast to the wet method, in which the beans undergo a germination process, in the dry method, germination is inhibited",
          "for": "germination inhibited in dry method (contested)"
        },
        {
          "text": "The difference in the number of free amino acids between wet and dry-processed coffee ranged from 3.3–20.9%",
          "for": "free amino acid range"
        },
        {
          "text": "the higher content of several free amino acids (glutamate, aspartate, and alanine) in wet processing",
          "for": "glutamate, aspartate, alanine"
        },
        {
          "text": "Dry, wet, and semi-dry processing did not cause significant caffeine, chlorogenic acid, and trigonelline changes.",
          "for": "caffeine, CGA, trigonelline conclusion"
        },
        {
          "text": "Sucrose levels are reduced due to the endogenous invertase enzyme's activity",
          "for": "invertase and sucrose"
        },
        {
          "text": "the drought stress is likewise longer, and more GABA is produced",
          "for": "longer drought stress, more GABA"
        }
      ],
      "scope": {
        "claim": "cited",
        "context": "Review; Table 1 compiles values from primary studies (sugars from Knopp et al. 2006, GABA from Bytof et al. 2005); free amino acid range cited from a 2004 ASIC conference paper; germination inhibition cited from a 1980 endocarp study; caffeine conclusion is the review's own",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s8",
      "title": "Exploring the impacts of postharvest processing on the microbiota and metabolite profiles during green coffee bean production",
      "authors": "De Bruyn, F.; Zhang, S. J.; Pothakos, V.; Torres, J.; Lambot, C.; Moroni, A. V.; Callanan, M.; Sybesma, W.; Weckx, S.; De Vuyst, L.",
      "year": 2017,
      "publisher": "Applied and Environmental Microbiology 83(1), e02398-16",
      "url": "https://journals.asm.org/doi/10.1128/aem.02398-16",
      "doi": "10.1128/AEM.02398-16",
      "tier": "A",
      "quote": "Several of the metabolites associated with lactic acid bacterial metabolism (e.g., lactic acid, acetic acid, and mannitol) produced in the mucilage were also found in the endosperm.",
      "quotes": [],
      "scope": {
        "claim": "finding",
        "context": "Typica arabica, one experimental farm in Ecuador; wet and dry processing; endosperm metabolites",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s9",
      "title": "Following coffee production from cherries to cup: microbiological and metabolomic analysis of wet processing of Coffea arabica",
      "authors": "Zhang, S. J.; De Bruyn, F.; Pothakos, V.; Torres, J.; Falconi, C.; Moccand, C.; Weckx, S.; De Vuyst, L.",
      "year": 2019,
      "publisher": "Applied and Environmental Microbiology 85(6), e02635-18",
      "url": "https://journals.asm.org/doi/10.1128/aem.02635-18",
      "doi": "10.1128/AEM.02635-18",
      "tier": "A",
      "quote": "During wet coffee processing, the interplay between microbial activities and endogenous bean metabolism results in a specific flavor precursor profile of the green coffee beans.",
      "quotes": [],
      "abstractOnly": true,
      "scope": {
        "claim": "finding",
        "context": "Arabica, wet processing on one experimental farm in Ecuador; abstract only consulted",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s10",
      "title": "Influence of various processing parameters on the microbial community dynamics, metabolomic profiles, and cup quality during wet coffee processing",
      "authors": "Zhang, S. J.; De Bruyn, F.; Pothakos, V.; Contreras, G. F.; Cai, Z.; Moccand, C.; Weckx, S.; De Vuyst, L.",
      "year": 2019,
      "publisher": "Frontiers in Microbiology 10, 2621",
      "url": "https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2019.02621/full",
      "doi": "10.3389/fmicb.2019.02621",
      "tier": "A",
      "quote": "At the same time, the endogenous coffee bean metabolism remained active at different stages along the processing, as could be seen through changes in the concentrations of carbohydrates, organic acids, and free amino acids.",
      "quotes": [],
      "scope": {
        "claim": "finding",
        "context": "Arabica (Catimor), Yunnan, China; controlled wet-processing trial",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s11",
      "title": "Chlorogenic acids and other relevant compounds in Brazilian coffees processed by semi-dry and wet post-harvesting methods",
      "authors": "Duarte, G. S.; Pereira, A. A.; Farah, A.",
      "year": 2010,
      "publisher": "Food Chemistry 118(3), 851–855",
      "url": "https://www.sciencedirect.com/science/article/pii/S0308814609006840",
      "doi": "10.1016/j.foodchem.2009.05.042",
      "tier": "A",
      "tierNote": "Full text (PDF supplied by the owner) read on 16 Nov 2026 (earlier: abstract only). The 17 samples are four arabica cultivars (Red Catuaí, Rubi, Yellow Bourbon, Topázio) and progenies of crosses with Timor Hybrid; the methods text says 11 hybrids, but numbers them 1–13 (10 Yellow Catuaí × Timor Hybrid, 2 Red Caturra × Timor Hybrid) and the statistics use n = 13. Compared alone, the four arabica cultivars showed no wet vs semi-dry difference in total chlorogenic acids or sucrose; the authors attribute the differences to the hybrids. Trigonelline p is given as < 0.01 in the abstract and < 0.0001 in the results.",
      "quote": "Coffees processed by wet method showed higher contents of chlorogenic acids (p = 0.02) and trigonelline (p < 0.01), and lower content of sucrose (p = 0.02) compared to those produced by a semi-dry method.",
      "quotes": [
        {
          "text": "when only Arabica seeds (n = 4) were compared statistically, no significant difference was observed",
          "for": "no chlorogenic-acid difference in the four arabica cultivars alone"
        },
        {
          "text": "However, when evaluating only Arabica samples, no differences were found between semi-dry and wet methods.",
          "for": "no sucrose difference in the four arabica cultivars alone"
        },
        {
          "text": "Regarding caffeine, no difference was observed between both methods.",
          "for": "caffeine did not differ"
        }
      ],
      "scope": {
        "claim": "finding",
        "context": "17 Brazilian samples (four arabica cultivars, the rest Timor Hybrid progenies), wet vs semi-dry; full text read",
        "checked": "2026-11-16T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s12",
      "title": "Impact of drying process on chemical composition and key aroma components of Arabica coffee",
      "authors": "Kulapichitr, F.; Borompichaichartkul, C.; Suppavorasatit, I.; Cadwallader, K. R.",
      "year": 2019,
      "publisher": "Food Chemistry 291, 49–58",
      "url": "https://www.sciencedirect.com/science/article/abs/pii/S0308814619306478",
      "doi": "10.1016/j.foodchem.2019.03.152",
      "tier": "A",
      "quote": "Sucrose content was higher in HP and TD than in SD green coffees.",
      "quotes": [],
      "abstractOnly": true,
      "scope": {
        "claim": "finding",
        "context": "Arabica cv. Catimor, Thailand; heat-pump (40, 45, 50 °C), tray and sun drying; abstract only consulted",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s13",
      "title": "Post-harvest effects on beverage quality and physiological performance of coffee beans",
      "authors": "Taveira, J. H. S.; Borém, F. M.; Rosa, S. D. V. F.; Oliveira, P. D.; Giomo, G. S.; Isquierdo, E. P.; Fortunato, V. A.",
      "year": 2015,
      "publisher": "African Journal of Agricultural Research 10(12), 1457–1466",
      "url": "https://www.alice.cnptia.embrapa.br/alice/handle/doc/1039291",
      "doi": "10.5897/AJAR2014.9263",
      "tier": "A",
      "quote": "The sun-drying provided coffee beans with the best physiological performance and the best drink quality.",
      "quotes": [],
      "scope": {
        "claim": "finding",
        "context": "Arabica cv. Catuaí Vermelho, Brazil; natural and washed coffee, sun vs heated-air drying; germination and cup quality",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s14",
      "title": "The Storage of Green Coffee (Coffea arabica): Decrease of Viability and Changes of Potential Aroma Precursors",
      "authors": "Selmar, D.; Bytof, G.; Knopp, S.-E.",
      "year": 2008,
      "publisher": "Annals of Botany 101(1), 31–38",
      "url": "https://academic.oup.com/aob/article/101/1/31/185681",
      "doi": "10.1093/aob/mcm277",
      "tier": "A",
      "quote": "Glucose and fructose decreased slightly in the course of storage and glutamine content declined significantly.",
      "quotes": [],
      "scope": {
        "claim": "finding",
        "context": "Stored green arabica; sugars, amino acids and viability over storage",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s15",
      "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": "Washed coffee tends to have a cleaner and fruity aroma than natural ones.",
      "quotes": [
        {
          "text": "With the mucilage intact, the natural process gives the roasted coffee a rich body and sweet characteristic",
          "for": "natural: rich body and sweetness"
        },
        {
          "text": "The dry process typically gives coffee a sweeter taste due to higher fructose and glucose contents",
          "for": "sweetness and hexoses"
        },
        {
          "text": "Under the same roasting conditions, a drier postharvest process gives higher pyrazine formation",
          "for": "pyrazines"
        },
        {
          "text": "Trigonelline, for instance, is an essential flavour precursor",
          "for": "trigonelline as precursor"
        },
        {
          "text": "the contents of reducing sugar could affect Maillard reactions",
          "for": "reducing sugars and Maillard"
        },
        {
          "text": "Around 300 VOCs have been found in raw coffee beans, while over 850 in roasted coffee",
          "for": "green coffee aroma differs"
        }
      ],
      "scope": {
        "claim": "cited",
        "context": "Review of post-harvest processing and roast aroma; sensory contrasts partly uncited, sweetness and pyrazine statements cited from primary studies",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s16",
      "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).",
      "quotes": [],
      "abstractOnly": true,
      "scope": {
        "claim": "finding",
        "context": "One Catimor lot, Alto Inambari, Peru; washed, honey, natural, anaerobic; abstract only consulted",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s17",
      "title": "Enhancement of Coffee Quality Attributes by Combining Processing Methods and Varieties",
      "authors": "Teran, E.",
      "year": 2024,
      "publisher": "Beverages 10(1), 10",
      "url": "https://www.mdpi.com/2306-5710/10/1/10",
      "doi": "10.3390/beverages10010010",
      "tier": "A",
      "tierNote": "Our reading of the full text: the abstract says Bourbon and Pacamara enhance flavour when processed natural, but the Results report no Natural/Dry × Bourbon or × Pacamara test. The reported significant interactions are Natural/Dry × Typica (higher flavour) and Washed/Wet × Bourbon (lower flavour); Pacamara has 8 records (Table A4), below the paper's own filter of more than 20 scores.",
      "quote": "specific Arabica varieties, such as Bourbon and Pacamara, enhance flavor when processed using the 'Natural/Dry' method",
      "quotes": [
        {
          "text": "the ‘Natural/Dry’ method and Typica variety significantly enhanced the flavor",
          "for": "the significant natural interaction is with Typica"
        },
        {
          "text": "a noteworthy flavor reduction was identified in the interaction of the ‘Washed/Wet’ method with the Bourbon variety",
          "for": "washed Bourbon scored lower"
        }
      ],
      "scope": {
        "claim": "data",
        "context": "Analysis of CQI cupping records (1,310 arabica, 28 robusta) by variety and processing; the Bourbon and Pacamara statement is from the abstract and is not matched by a reported test (our reading)",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    },
    {
      "id": "s18",
      "title": "The effect of coffee beans roasting on its chemical composition",
      "authors": "Diviš, P.; Pořízka, J.; Kříkala, J.",
      "year": 2019,
      "publisher": "Potravinarstvo Slovak Journal of Food Sciences 13(1), 344–350",
      "url": "https://potravinarstvo.com/journal1/index.php/potravinarstvo/article/download/1062/807",
      "doi": "10.5219/1062",
      "tier": "A",
      "quote": "The results show that sucrose degradation occurs during the roasting process.",
      "quotes": [
        {
          "text": "At the end of the roasting process, almost all sucrose has already been degraded (96 – 98%).",
          "for": "96–98% degraded"
        }
      ],
      "scope": {
        "claim": "finding",
        "context": "One Brazilian arabica, 25 g batches, home roaster, medium (210 °C, 14 min) and full city (225 °C, 19 min); mg/g dry weight.",
        "checked": "2026-11-13T00:00:00.000Z"
      },
      "accessed": "2026-11-13T00:00:00.000Z"
    }
  ],
  "markdown": "## Precursors: what the roast needs\n\nGreen coffee smells different from roasted coffee. Most of the aroma compounds of roasted coffee form in the roaster, from compounds already in the seed. These starting compounds are called precursors.<sup><a href=\"#src-s15\">15</a></sup>\n\nSugars and free amino acids are among the main precursors; trigonelline and chlorogenic acids are others.<sup><a href=\"#src-s15\">15</a></sup> Processing changes how much of each precursor the green coffee carries into the roast.<sup><a href=\"#src-s3\">3</a>,<a href=\"#src-s9\">9</a></sup> This entry follows that bridge from the mill to the cup. The methods themselves are described in their own entries.\n\n## How washed and natural coffee differ\n\n**Sugars.** Washed arabica holds only low amounts of glucose and fructose. Natural arabica processed in parallel holds significantly more.<sup><a href=\"#src-s5\">5</a></sup> Sucrose, the main free sugar in green coffee, was not significantly affected in the same study.<sup><a href=\"#src-s5\">5</a></sup>\n\n<figure>\n<div data-figure=\"processing-and-flavour-chemistry-sugars\"></div>\n<figcaption>Glucose and fructose in green arabica (var. Acaiá, one farm in Brazil) after washed, semi-dry and natural processing of the same batch, in g per 100 g dry matter. The study shows the means as bar charts only; the values are our reading of its Fig. 2, to about ±0.01. Sucrose (about 7.8–8.1 g) is not drawn.<sup><a href=\"#src-s5\">5</a></sup></figcaption>\n</figure>\n\n**Amino acids.** Processing also changes the free amino acids. GABA (γ-aminobutyric acid) is an amino acid that plants build up under stress. It was always markedly higher in natural than in washed arabica.<sup><a href=\"#src-s1\">1</a></sup>\n\n**Other precursors.** In 17 Brazilian arabica cultivars and progenies, wet-processed coffee held more chlorogenic acids and trigonelline than the same cultivars processed by the semi-dry method, as the study names it.<sup><a href=\"#src-s11\">11</a></sup> It also held less sucrose.<sup><a href=\"#src-s11\">11</a></sup> Thirteen of the 17 were progenies of crosses with Timor Hybrid; compared alone, the four arabica cultivars showed no difference in total chlorogenic acids or sucrose.<sup><a href=\"#src-s11\">11</a></sup>\n\n## Fermentation and the living seed\n\nTwo kinds of chemistry run at the same time during processing. Microbes work on the fruit layers outside the seed. Inside, the seed itself is alive and keeps metabolising.<sup><a href=\"#src-s9\">9</a>,<a href=\"#src-s10\">10</a></sup>\n\n**Microbial products.** In wet fermentation, bacteria and yeasts turn mucilage sugars into acids and alcohols. Some of these products enter the seed. Lactic acid, acetic acid and mannitol made in the mucilage were also found in the endosperm.<sup><a href=\"#src-s8\">8</a></sup> The fermentation entry covers how much enters and when.\n\n**Seed metabolism.** The living seed starts to germinate during processing.<sup><a href=\"#src-s2\">2</a></sup> In washed coffee, germination activity peaked about two days after processing began. In natural coffee, it peaked about one week after the start.<sup><a href=\"#src-s2\">2</a></sup> As the seed dries, it also reacts to drought stress.<sup><a href=\"#src-s4\">4</a></sup> The seed's metabolism changes carbohydrates, organic acids and free amino acids,<sup><a href=\"#src-s10\">10</a></sup> and its extent depends on the method.<sup><a href=\"#src-s3\">3</a></sup>\n\n## Drying heat\n\nEven gentle drying changes the precursors. In wet-processed arabica dried at 30 °C, fructose fell by more than half within the first day.<sup><a href=\"#src-s6\">6</a></sup>\n\nHot air can kill the seed. In a Brazilian trial, natural coffee dried with air at 60 °C did not germinate at all.<sup><a href=\"#src-s13\">13</a></sup> Sun-drying gave both the best physiological performance and the best cup in that trial.<sup><a href=\"#src-s13\">13</a></sup> The drying method also changed sucrose and some amino acids in a 2019 arabica study.<sup><a href=\"#src-s12\">12</a></sup>\n\n## From precursors to the cup\n\nIn the roaster, sugars and amino acids start a cascade of browning reactions, the Maillard reaction.<sup><a href=\"#src-s15\">15</a></sup> Sugar-derived compounds react further with free amino acids in Strecker degradation, which gives aldehydes.<sup><a href=\"#src-s15\">15</a></sup> Trigonelline breaks down into pyridines, and chlorogenic acids into phenols.<sup><a href=\"#src-s15\">15</a></sup> The roasting entries explain these reactions in detail.\n\n<figure class=\"iso-fig\">\n<div data-figure=\"processing-and-flavour-chemistry-pathway\"></div>\n<ol class=\"fig-legend\">\n<li data-n=\"1\"><span><b>Ripe cherry.</b> The seed inside is alive when the cherry is picked.<sup><a href=\"#src-s2\">2</a>,<a href=\"#src-s4\">4</a></sup></span></li>\n<li data-n=\"2\"><span><b>Washed path.</b> Germination activity peaks about two days into processing; glucose and fructose end up low.<sup><a href=\"#src-s2\">2</a>,<a href=\"#src-s5\">5</a></sup></span></li>\n<li data-n=\"3\"><span><b>Microbial products.</b> Lactic acid, acetic acid and mannitol from the mucilage are found in the endosperm.<sup><a href=\"#src-s8\">8</a></sup></span></li>\n<li data-n=\"4\"><span><b>Natural path.</b> Germination activity peaks about one week into processing; GABA builds up; glucose and fructose stay higher.<sup><a href=\"#src-s1\">1</a>,<a href=\"#src-s2\">2</a>,<a href=\"#src-s5\">5</a></sup></span></li>\n<li data-n=\"5\"><span><b>Drying heat.</b> Natural seeds dried with air at 60 °C did not germinate in one trial.<sup><a href=\"#src-s13\">13</a></sup></span></li>\n<li data-n=\"6\"><span><b>Green coffee.</b> The precursor profile it carries depends on processing.<sup><a href=\"#src-s3\">3</a>,<a href=\"#src-s9\">9</a></sup></span></li>\n<li data-n=\"7\"><span><b>Roasting.</b> Sugars and amino acids feed the Maillard reaction and Strecker degradation; trigonelline gives pyridines, chlorogenic acids give phenols.<sup><a href=\"#src-s15\">15</a></sup></span></li>\n<li data-n=\"8\"><span><b>Cup.</b> In a 2023 review, natural coffee is described with a rich body and sweetness, washed coffee as cleaner.<sup><a href=\"#src-s15\">15</a></sup></span></li>\n</ol>\n<figcaption>How processing shapes the precursor profile of green coffee, and where the precursors go in the roast. Sequence only: no amounts or times are drawn. Our own drawing. Schematic, not to scale.</figcaption>\n</figure>\n\nTasters do notice the method. According to a 2023 review, washed coffee “tends to have a cleaner and fruity aroma than natural ones”.<sup><a href=\"#src-s15\">15</a></sup> The same review describes natural coffee as giving a rich body and sweetness, and links the sweeter taste to its higher glucose and fructose.<sup><a href=\"#src-s15\">15</a></sup> Single studies are less uniform, as the deep dive shows.<sup><a href=\"#src-s16\">16</a>,<a href=\"#src-s17\">17</a></sup>\n\n<!-- deep-dive -->\n\n## Sugars in detail\n\n**Washed against natural.** In a 2006 study, one batch of ripe arabica (var. Acaiá) on a farm in Brazil was processed in parallel by the wet, dry and semi-dry methods; for the semi-dry coffee the pulped beans were dried with their mucilage, without tank fermentation.<sup><a href=\"#src-s5\">5</a></sup> The contents of fructose and glucose showed a close correlation with the kind of post-harvest treatment.<sup><a href=\"#src-s5\">5</a></sup> Read from the study’s bar charts, glucose was about 0.03 g per 100 g dry matter after washed processing, 0.10 after semi-dry and 0.22 after natural processing; fructose was about 0.04, 0.19 and 0.33 (our reading).<sup><a href=\"#src-s5\">5</a></sup> A 2021 review compiles almost the same values from this study.<sup><a href=\"#src-s7\">7</a></sup> Sucrose stayed at about 7.8–8.1 g per 100 g dry matter in all three and, in the study’s own conclusion, was not significantly affected by processing.<sup><a href=\"#src-s5\">5</a></sup>\n\n**Inside the seed.** A drying study removed process water from the question. Demucilaged arabica, fermented without water, was dried at 30 °C.<sup><a href=\"#src-s6\">6</a></sup> Fructose fell from nearly 110 to about 40 mg/100 g dry weight within the first day of continuous drying, and to about 30 in beans dried with pauses.<sup><a href=\"#src-s6\">6</a></sup> Sucrose, at about 6,000 mg/100 g dry weight, stayed roughly constant.<sup><a href=\"#src-s6\">6</a></sup> The authors attribute at least part of the low hexose content of washed coffee to the seed's own metabolism.<sup><a href=\"#src-s6\">6</a></sup>\n\n**In the tank.** A field study in Ecuador found fructose, glucose and sucrose in the endosperm falling during wet fermentation; fructose and glucose fell further during a 24-hour soak.<sup><a href=\"#src-s8\">8</a></sup> The review attributes the fall in sucrose during wet fermentation to the seed's own invertase, which splits sucrose into glucose and fructose.<sup><a href=\"#src-s7\">7</a></sup> How much each route contributes is not settled; see Contested.\n\n**Why hexoses matter in the roast.** Reducing sugars such as glucose and fructose take part directly in the Maillard reaction, and their content can affect it.<sup><a href=\"#src-s15\">15</a></sup> No study consulted for this entry measured whether the small hexose differences between washed and natural coffee survive the roast as measurable differences in specific aroma compounds.\n\n## Free amino acids and GABA\n\n**GABA.** In a 2005 study, GABA was present in green arabica in significantly varying amounts.<sup><a href=\"#src-s1\">1</a></sup> It was always markedly higher in natural coffee than in washed coffee or untreated seeds, and the authors proposed it as a marker of the processing method.<sup><a href=\"#src-s1\">1</a></sup> In the study’s table, a commercial batch from Brazil held 93 nmol per seed after washed and 1,009 after natural processing; a laboratory model processing of Tanzanian coffee gave 140 against 1,860.<sup><a href=\"#src-s1\">1</a></sup> A 2021 review compiles the same values.<sup><a href=\"#src-s7\">7</a></sup>\n\n<figure>\n<div data-figure=\"processing-and-flavour-chemistry-gaba\"></div>\n<figcaption>GABA in green arabica after washed and natural processing, in nmol per seed: a commercial batch from Brazil and a laboratory model processing of Tanzanian coffee. Values from Table 1 of a 2005 study.<sup><a href=\"#src-s1\">1</a></sup></figcaption>\n</figure>\n\n**Drought stress.** Falling water potential in the still-living seed induces massive drought-stress responses.<sup><a href=\"#src-s4\">4</a></sup> In a 2010 study, GABA peaks during drying coincided with germination activity and with maxima of dehydrin expression, a marker of drought stress.<sup><a href=\"#src-s4\">4</a></sup> Natural coffee dries for longer, and according to the 2021 review the longer drought stress is why it builds up more GABA.<sup><a href=\"#src-s7\">7</a></sup>\n\n**Other amino acids.** The same review reports higher contents of glutamate, aspartate and alanine in wet processing and links them to protein breakdown during germination.<sup><a href=\"#src-s7\">7</a></sup> Citing a 2004 conference paper, it gives the difference in free amino acids between wet- and dry-processed coffee as 3.3–20.9%, without saying what the percentage refers to.<sup><a href=\"#src-s7\">7</a></sup> Free amino acids are the substrate of Strecker degradation in the roast.<sup><a href=\"#src-s15\">15</a></sup>\n\n**Germination itself.** Germination markers rose and fell again in both methods; the timing differed.<sup><a href=\"#src-s2\">2</a></sup> The authors of the germination studies argue that the extent and nature of germination depend on the processing method, and that these metabolic events underlie the chemical differences between washed and natural coffee.<sup><a href=\"#src-s2\">2</a>,<a href=\"#src-s3\">3</a></sup>\n\n## Chlorogenic acids, trigonelline and caffeine\n\nFor these compounds the evidence is mixed. In 17 Brazilian arabica cultivars and progenies, the wet method gave significantly more chlorogenic acids and trigonelline and less sucrose than the semi-dry method; caffeine did not differ.<sup><a href=\"#src-s11\">11</a></sup> The authors attribute the chlorogenic-acid and sucrose differences to the Timor Hybrid progenies: in the four arabica cultivars alone, neither differed.<sup><a href=\"#src-s11\">11</a></sup> A 2021 review concludes that dry, wet and semi-dry processing did not cause significant changes in caffeine, chlorogenic acids or trigonelline.<sup><a href=\"#src-s7\">7</a></sup> The same review cites one study in which caffeine and trigonelline fell at the end of wet fermentation.<sup><a href=\"#src-s7\">7</a></sup> In a 2019 drying study, the drying method did not affect caffeine.<sup><a href=\"#src-s12\">12</a></sup>\n\nIn the roast, trigonelline is an essential flavour precursor whose breakdown products include pyridines, and phenols come mainly from the thermal breakdown of chlorogenic acids.<sup><a href=\"#src-s15\">15</a></sup> The chlorogenic acids entry follows them through the roast.\n\n## Fermentation metabolites\n\nWet processing builds a precursor profile from two sources: microbial activity and the bean's own metabolism.<sup><a href=\"#src-s9\">9</a></sup> In Yunnan, the endogenous metabolism stayed active at several stages of wet processing, with changes in carbohydrates, organic acids and free amino acids.<sup><a href=\"#src-s10\">10</a></sup> In Ecuador, lactic acid, acetic acid and mannitol produced in the mucilage were also found in the endosperm.<sup><a href=\"#src-s8\">8</a></sup>\n\nFermentation duration changes the beans too. In the Ecuadorian trial, longer-fermented coffee had more notes preferred by consumers.<sup><a href=\"#src-s9\">9</a></sup> Ethanol and the other metabolites measured in the beans, and how far they move into the seed, are described under fermentation and microorganisms.\n\n## Drying temperature\n\nDrying changes the precursors even without heat damage. In wet-processed arabica dried at 30 °C with pauses, glucose rose again from under 15 to nearly 40 mg/100 g dry weight before ending at about 20.<sup><a href=\"#src-s6\">6</a></sup> The authors read such changes as active metabolism in the drying seed.<sup><a href=\"#src-s6\">6</a></sup>\n\nHigher temperatures damage the seed. In a Brazilian trial with Catuaí Vermelho, natural coffee dried with air at 60 °C, or at 60 °C then 40 °C, did not germinate; sun-dried natural coffee reached 80%.<sup><a href=\"#src-s13\">13</a></sup> Across both processes, sun-drying gave the best cup.<sup><a href=\"#src-s13\">13</a></sup>\n\nIn the 2019 arabica study, sucrose was higher after heat-pump drying at 40–50 °C and after tray drying than after sun-drying.<sup><a href=\"#src-s12\">12</a></sup> The drying process also changed some amino acids.<sup><a href=\"#src-s12\">12</a></sup> The key odorant 2-furfurylthiol was at the same level in sun-dried and heat-pump-dried coffee and lowest after tray drying.<sup><a href=\"#src-s12\">12</a></sup> According to a 2023 review, under the same roasting conditions a drier post-harvest process gives more pyrazines.<sup><a href=\"#src-s15\">15</a></sup>\n\n## Sensory evidence\n\nA 2023 review describes a contrast. In its words, washed coffee “tends to have a cleaner and fruity aroma than natural ones”; natural coffee is described with a rich body and sweetness.<sup><a href=\"#src-s15\">15</a></sup> The review ties the sweetness to higher glucose and fructose and reports higher aldehyde content after natural processing.<sup><a href=\"#src-s15\">15</a></sup> That link runs through the roast: in one roasting study almost all sucrose was degraded by the end of the roast,<sup><a href=\"#src-s18\">18</a></sup> and no study consulted here measured how much of the extra glucose and fructose of natural coffee is left in the cup. Why black coffee can taste sweet with little sugar is discussed in Sweetness in coffee.\n\nSingle trials are less clear-cut. On one Peruvian Catimor lot, only fragrance and body differed significantly between washed, honey, natural and anaerobic processing.<sup><a href=\"#src-s16\">16</a></sup> Sealed-tank and inoculated fermentations are covered in Anaerobic fermentation and Co-fermentation. In an analysis of cupping records, variety and processing method interacted: natural-processed Typica scored significantly higher for flavour, washed Bourbon lower.<sup><a href=\"#src-s17\">17</a></sup> The abstract’s broader claim for Bourbon and Pacamara is not matched by a reported test, and Pacamara had only eight records.<sup><a href=\"#src-s17\">17</a></sup>\n\nNo study consulted for this entry traces a specific cup attribute, such as body or fruitiness, back to a single measured precursor difference between washed and natural coffee.\n\n## After processing\n\nThe precursors keep changing in the warehouse. In stored green arabica, glucose and fructose decreased slightly and glutamine declined significantly.<sup><a href=\"#src-s14\">14</a></sup> These changes were not correlated with the viability of the beans.<sup><a href=\"#src-s14\">14</a></sup> The authors suggest that Maillard reactions during storage may cause the loss of potential aroma precursors, and that loss of viability is relevant to the flattening of aroma.<sup><a href=\"#src-s14\">14</a></sup> Green coffee storage covers conditions and packaging.\n\n## Reference values\n\nValues hold for the conditions of the study named. Values from different studies are not directly comparable.\n\n| Parameter | Value reported | Source |\n|---|---|---|\n| Glucose, green arabica (Brazil, one batch) | Washed ≈0.03; semi-dry ≈0.10; natural ≈0.22 g per 100 g dry matter (our reading of the bar chart) | Knopp et al.<sup><a href=\"#src-s5\">5</a></sup> |\n| Fructose, same batch | Washed ≈0.04; semi-dry ≈0.19; natural ≈0.33 g per 100 g dry matter (our reading) | Knopp et al.<sup><a href=\"#src-s5\">5</a></sup> |\n| Sucrose, same batch | About 7.8–8.1 g per 100 g dry matter in all three (our reading); not significantly affected | Knopp et al.<sup><a href=\"#src-s5\">5</a></sup> |\n| Fructose, washed arabica dried at 30 °C | Nearly 110 → about 40 mg/100 g dry weight within day 1 (continuous); about 30 (with pauses) | Kleinwächter & Selmar<sup><a href=\"#src-s6\">6</a></sup> |\n| Sucrose, same trial | About 6,000 mg/100 g dry weight; roughly constant | Kleinwächter & Selmar<sup><a href=\"#src-s6\">6</a></sup> |\n| GABA, green arabica | Brazil, commercial batch: 93 (washed), 1,009 (natural); Tanzania, laboratory model processing: 140 (washed), 1,860 (natural) nmol per seed | Bytof et al.<sup><a href=\"#src-s1\">1</a></sup> |\n| Free amino acids | Difference between wet and dry processing 3.3–20.9% (as the review gives it, citing a 2004 conference paper; reference quantity not stated) | Bastian et al.<sup><a href=\"#src-s7\">7</a></sup> |\n| Germination peak | About 2 days (wet processing); about 1 week (dry processing) | Bytof et al.<sup><a href=\"#src-s2\">2</a></sup> |\n| Chlorogenic acids and trigonelline | Higher in wet than in semi-dry processing (p = 0.02 and p < 0.01); sucrose lower (p = 0.02); chlorogenic acids and sucrose not different in the four arabica cultivars alone | Duarte et al.<sup><a href=\"#src-s11\">11</a></sup> |\n| Caffeine | No significant change with processing method (review conclusion); no difference between wet and semi-dry (one study); not affected by drying method (one study) | Bastian et al.<sup><a href=\"#src-s7\">7</a></sup>, Duarte et al.<sup><a href=\"#src-s11\">11</a></sup>, Kulapichitr et al.<sup><a href=\"#src-s12\">12</a></sup> |\n| Germination after drying, natural (Catuaí Vermelho) | Sun-dried 80%; 60 °C 0%; 60 °C then 40 °C 0% | Taveira et al.<sup><a href=\"#src-s13\">13</a></sup> |\n| Storage | Glucose and fructose slightly lower; glutamine significantly lower | Selmar et al.<sup><a href=\"#src-s14\">14</a></sup> |\n\n## Open questions\n\n- **Precursor to attribute.** The sugar and GABA differences between washed and natural coffee are well documented,<sup><a href=\"#src-s1\">1</a>,<a href=\"#src-s5\">5</a></sup> but no study consulted for this entry links them quantitatively to specific aroma compounds or cup attributes after roasting.\n- **Fate of GABA.** Whether the extra GABA of natural coffee takes part in roasting reactions or affects the cup is not established in the sources consulted.\n- **Honey and pulped natural coffee.** Semi-dry coffee sat between washed and natural for glucose and fructose in one parallel trial,<sup><a href=\"#src-s5\">5</a></sup> but no study consulted here has timed germination or stress markers in it.\n- **Robusta.** The germination, GABA and sugar studies consulted here used arabica only.<sup><a href=\"#src-s1\">1</a>,<a href=\"#src-s2\">2</a>,<a href=\"#src-s5\">5</a></sup>\n\n## Key studies\n\n<div class=\"study\">\n<p><strong>Knopp, Bytof & Selmar, 2006 – Sugars after washed and natural processing.</strong> Compared green arabica (var. Acaiá) processed in parallel by the wet and the dry method, together with model processing in the laboratory. Glucose and fructose were low in washed coffee and significantly higher in natural coffee, while sucrose was not significantly affected.</p>\n<p class=\"meta\">European Food Research and Technology 223, 195–201 · <a href=\"https://doi.org/10.1007/s00217-005-0172-1\">doi:10.1007/s00217-005-0172-1</a></p>\n</div>\n\n<div class=\"study\">\n<p><strong>Bytof et al., 2005 – GABA as a processing marker.</strong> Measured free amino acids in differently processed green coffees. The stress metabolite GABA was always markedly higher in natural than in washed arabica, evidence of active metabolism in the seed during processing.</p>\n<p class=\"meta\">European Food Research and Technology 220(3–4), 245–250 · <a href=\"https://doi.org/10.1007/s00217-004-1033-z\">doi:10.1007/s00217-004-1033-z</a></p>\n</div>\n\n<div class=\"study\">\n<p><strong>Bytof et al., 2007 – Germination during processing.</strong> Followed germination markers in arabica seeds through wet and dry processing. Activity peaked two days into wet processing and about a week into dry processing, a timing the authors connect to the chemical differences between the two coffees.</p>\n<p class=\"meta\">Annals of Botany 100(1), 61–66 · <a href=\"https://doi.org/10.1093/aob/mcm068\">doi:10.1093/aob/mcm068</a></p>\n</div>\n\n<div class=\"study\">\n<p><strong>De Bruyn et al., 2017 – Microbes and metabolites in wet and dry processing.</strong> Followed microbial communities and metabolites through wet and dry processing in Ecuador. Lactic acid, acetic acid and mannitol produced in the mucilage were also found in the endosperm, and sugars in the endosperm fell during fermentation.</p>\n<p class=\"meta\">Applied and Environmental Microbiology 83(1), e02398-16 · <a href=\"https://doi.org/10.1128/AEM.02398-16\">doi:10.1128/AEM.02398-16</a></p>\n</div>\n\n<div class=\"study\">\n<p><strong>Cao et al., 2023 – Processing and aroma formation.</strong> Reviewed how post-harvest processing shapes the volatile compounds of roasted coffee through the Maillard reaction, Strecker degradation and related routes. Describes washed coffee as cleaner and natural coffee as having a rich body and sweetness.</p>\n<p class=\"meta\">International Journal of Food Science and Technology 58(3), 1007–1027 · <a href=\"https://doi.org/10.1111/ijfs.16261\">doi:10.1111/ijfs.16261</a></p>\n</div>"
}