{ "id": "carbohydrates-and-body", "url": "https://coffearium.org/entries/carbohydrates-and-body/", "license": "CC-BY-4.0", "title": "Carbohydrates and body", "summary": "Half of green coffee is carbohydrate, mostly cell-wall polysaccharides. Roasting burns off the sugar and breaks part of the polysaccharides down; what dissolves helps shape viscosity, crema and, by most accounts, body.", "domain": "roasting", "kind": "Concept", "stage": "roast", "topics": [ "science" ], "topicSections": {}, "lang": "en", "status": "verified", "reviewer": "Coffearium verification pass", "lastVerified": "2026-10-20T00:00:00.000Z", "updated": "2026-10-20T00:00:00.000Z", "added": "2026-10-20T00:00:00.000Z", "brief": [ "Polysaccharides make up about 50% of green coffee dry weight. Galactomannans, arabinogalactan-proteins and cellulose are the three main types.1", "Roasting degraded 12–40% of the polysaccharides in three arabica varieties, depending on roast; arabinogalactans lost up to 60%, cellulose barely changed.2", "Roasting makes galactomannans, mostly insoluble in green coffee, increasingly extractable with hot water.3", "In espresso, foam stability was related to the amount of galactomannan and arabinogalactan in the cup.11", "The link between polysaccharides and perceived body rests on textbook statements; in a 2020 multi-study analysis, roast phase timing did not affect body at all.4,15" ], "quickFacts": [ { "label": "Polysaccharides in green coffee", "value": "About 50% of dry weight" }, { "label": "Main types", "value": "Galactomannans (about half of the polysaccharides), arabinogalactan-proteins, cellulose" }, { "label": "Free sugar", "value": "Almost only sucrose, 5–12% of dry weight in mature seeds; nearly all lost in roasting" }, { "label": "Lost in roasting", "value": "12–40% of polysaccharides (three arabica varieties); arabinogalactans up to 60%" }, { "label": "Most stable", "value": "Cellulose; mannan backbones keep their molecular weight" }, { "label": "In the brew", "value": "Galactomannans and arabinogalactans, partly bound into melanoidins" }, { "label": "Linked to", "value": "Viscosity and mouthfeel (textbook); espresso foam stability (one study)" } ], "aka": [ "Coffee carbohydrates", "Coffee polysaccharides" ], "linkTerms": [ "galactomannan", "galactomannans", "arabinogalactan", "arabinogalactans", "arabinogalactan-protein", "arabinogalactan-proteins", "coffee polysaccharides", "cell-wall polysaccharides" ], "related": [ "body-and-mouthfeel", "maillard-reaction", "melanoidins", "caramelisation", "green-coffee-composition", "coffee-oil", "browning-phase", "acids-in-coffee", "espresso", "roast-loss", "glass-rubber-transition", "porosity-and-cell-structure", "extraction-yield-and-tds" ], "contested": [ "Body and roast. A roasting textbook chapter describes body as rising with roast degree to a point and then falling, so that over-roasted coffee often tastes thin14. A 2020 analysis of eight roasting studies, with colours from 46 to 99 Agtron, found that neither the time to first crack nor the development time affected body.15 The analysis was built around phase timing rather than final roast degree; in the one study that rated body across roast colours, darker roasts scored slightly higher for body, which the authors did not treat as a meaningful effect. The two statements therefore do not test exactly the same thing." ], "corrections": [ { "date": "2026-10-20T00:00:00.000Z", "change": "The contested item on body and roast no longer says the 2020 analysis varied only phase times; it now notes that in the one study that rated body across roast colours, darker roasts scored slightly higher, and the body text says the same.", "reason": "Verification read Münchow et al. 2020 (Table 4, Study 2), where roast colour correlated weakly but significantly with body (r = 0.13, p < 0.05)." }, { "date": "2026-10-20T00:00:00.000Z", "change": "The monosaccharide composition from Tian et al. 2017 is now described as that of oligosaccharides extracted from the dark-roasted coffee (weight %), not of brew oligosaccharides.", "reason": "The paper reports this composition for a hot-water extract of the roasted coffee (Figure 4), not for a brew." } ], "sources": [ { "id": "s1", "title": "Coffee carbohydrates", "authors": "Redgwell, R.; Fischer, M.", "year": 2006, "publisher": "Brazilian Journal of Plant Physiology 18(1), 165–174", "url": "https://www.scielo.br/j/bjpp/a/38pdpT9SJjNYhhPGvh93ryt/?lang=en", "doi": "10.1590/S1677-04202006000100012", "tier": "A", "quote": "The galactomannans are the predominant components in the coffee bean cell wall accounting for 50 % of the polysaccharides", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s2", "title": "Effect of roasting on degradation and structural features of polysaccharides in Arabica coffee beans", "authors": "Redgwell, R. J.; Trovato, V.; Curti, D.; Fischer, M.", "year": 2002, "publisher": "Carbohydrate Research 337(5), 421–431", "url": "https://agris.fao.org/search/es/records/65de2eb44c5aef494fdab739", "tier": "A", "quote": "The arabinogalactans and mannans were degraded up to 60 and 36%, respectively, after a dark roast.", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s3", "title": "Effect of roasting on the carbohydrate composition of Coffea arabica beans", "authors": "Oosterveld, A.; Voragen, A. G. J.; Schols, H. A.", "year": 2003, "publisher": "Carbohydrate Polymers 54(2), 183–192", "url": "https://research.wur.nl/en/publications/effect-of-roasting-on-the-carbohydrate-composition-of-coffea-arab", "doi": "10.1016/S0144-8617(03)00164-4", "tier": "A", "quote": "Cellulose remains unextractable and its solubility was not affected by the degree of roast", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s4", "title": "The Chemistry of Roasting – Decoding Flavor Formation", "authors": "Poisson, L.; Blank, I.; Dunkel, A.; Hofmann, T.", "year": 2017, "publisher": "In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 12, pp. 273–309", "url": "https://doi.org/10.1016/B978-0-12-803520-7.00012-8", "doi": "10.1016/B978-0-12-803520-7.00012-8", "tier": "B", "quote": "Polysaccharides do not seem to specifically contribute to aroma formation during roasting, but impart relevant organoleptic properties of the coffee brew, such as viscosity and mouth feel (Redgwell et al., 2002).", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s5", "title": "Formation of aliphatic acids by carbohydrate degradation during roasting of coffee", "authors": "Ginz, M.; Balzer, H. H.; Bradbury, A. G. W.; Maier, H. G.", "year": 2000, "publisher": "European Food Research and Technology 211(6), 404–410", "url": "https://doi.org/10.1007/s002170000215", "doi": "10.1007/s002170000215", "tier": "A", "quote": "Arabinose is the terminal unit on most of the side chains and due to its furanoside structure, labile to hydrolysis.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s6", "title": "Roasting Effects on Formation Mechanisms of Coffee Brew Melanoidins", "authors": "Bekedam, E. K.; Loots, M. J.; Schols, H. A.; van Boekel, M. A. J. S.; Smit, G.", "year": 2008, "publisher": "Journal of Agricultural and Food Chemistry 56(16), 7138–7145", "url": "https://research.wur.nl/en/publications/roasting-effects-on-formation-mechanisms-of-coffee-brew-melanoidi/", "doi": "10.1021/jf800999a", "tier": "A", "quote": "Furthermore, arabinogalactans seem to be relatively more involved in melanoidin formation than galactomannans.", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s7", "title": "The Brew – Extracting for Excellence", "authors": "Mestdagh, F.; Glabasnia, A.; Giuliano, P.", "year": 2017, "publisher": "In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 15, pp. 355–380", "url": "https://doi.org/10.1016/B978-0-12-803520-7.00015-3", "doi": "10.1016/B978-0-12-803520-7.00015-3", "tier": "B", "quote": "Simultaneously, coffee bed particles swell, due to the presence of water-insoluble polysaccharides in the roasted coffee, and coffee particles geometrically rearrange due to the water flow and/or pressure.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s8", "title": "Chemical Characterization of Potentially Prebiotic Oligosaccharides in Brewed Coffee and Spent Coffee Grounds", "authors": "Tian, T.; Freeman, S.; Corey, M.; German, J. B.; Barile, D.", "year": 2017, "publisher": "Journal of Agricultural and Food Chemistry 65, 2784–2792", "url": "https://pages.uoregon.edu/chendon/coffee_literature/old_literature/2017%20J.%20Agric.%20Food%20Chem.,%20Prebiotic%20sugars%20in%20brewed%20and%20roasted%20coffee.pdf", "doi": "10.1021/acs.jafc.6b04716", "tier": "A", "quote": "Mannose and galactose, respectively, at 37.1% and 42.2%, were the predominant constituents of oligosaccharides.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s9", "title": "Carboidratos na bebida do café preparado sob diferentes processos", "authors": "Martins, M. C. M.; Silva, C. O.; Buckeridge, M. S.; Vieira, C. C. J.", "year": 2005, "publisher": "Ciência e Tecnologia de Alimentos 25(2), 382–386", "url": "https://www.redalyc.org/pdf/3959/395940074034.pdf", "tier": "A", "quote": "Preparações obtidas em máquina de café expresso mostraram os menores valores de peso molecular (6 kDa)", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s10", "title": "Modulation of infusion processes to obtain coffee-derived food ingredients with distinct composition", "authors": "Lopes, G. R.; Passos, C. P.; Rodrigues, C.; Teixeira, J. A.; Coimbra, M. A.", "year": 2019, "publisher": "European Food Research and Technology 245(10), 2133–2146", "url": "https://link.springer.com/10.1007/s00217-019-03318-9", "doi": "10.1007/s00217-019-03318-9", "tier": "A", "quote": "The extraction process efficiency is more related to galactomannans extraction than arabinogalactans.", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s11", "title": "Foamability, foam stability, and chemical composition of espresso coffee as affected by the degree of roast", "authors": "Nunes, F. M.; Coimbra, M. A.; Duarte, A. C.; Delgadillo, I.", "year": 1997, "publisher": "Journal of Agricultural and Food Chemistry 45(8), 3238–3243", "url": "https://agris.fao.org/search/ar/records/65df54600f3e94b9e5d7795e", "doi": "10.1021/jf970009t", "tier": "A", "quote": "is related to the amount of galactomannan and arabinogalactan present", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s12", "title": "Neglected Food Bubbles: The Espresso Coffee Foam", "authors": "Illy, E.; Navarini, L.", "year": 2011, "publisher": "Food Biophysics 6(3), 335–348", "url": "https://link.springer.com/article/10.1007/s11483-011-9220-5", "doi": "10.1007/s11483-011-9220-5", "tier": "A", "quote": "The former was found to be composed of about 80% (on molar basis) of mannan", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s13", "title": "Crema – Formation, Stabilization, and Sensation", "authors": "Folmer, B.; Blank, I.; Hofmann, T.", "year": 2017, "publisher": "In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 17, pp. 399–417", "url": "https://doi.org/10.1016/B978-0-12-803520-7.00017-7", "doi": "10.1016/B978-0-12-803520-7.00017-7", "tier": "B", "quote": "Foam stability of espresso coffee was found to be related to the amount of the polysaccharides galactomannan and arabinogalactan.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s14", "title": "The Roast – Creating the Beans' Signature", "authors": "Schenker, S.; Rothgeb, T.", "year": 2017, "publisher": "In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 11, pp. 245–271", "url": "https://doi.org/10.1016/B978-0-12-803520-7.00011-6", "doi": "10.1016/B978-0-12-803520-7.00011-6", "tier": "B", "quote": "In a similar way the sensory attribute “body” or “mouthfeel” increases to a certain point and then decreases on continued roasting beyond this point.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s15", "title": "Roasting Conditions and Coffee Flavor: A Multi-Study Empirical Investigation", "authors": "Münchow, M.; Alstrup, J.; Steen, I.; Giacalone, D.", "year": 2020, "publisher": "Beverages 6(2), 29", "url": "https://www.mdpi.com/2306-5710/6/2/29", "doi": "10.3390/beverages6020029", "tier": "A", "quote": "By contrast, the mouthfeel attribute “body” was not affected at all by either roasting phases, which is interesting as this goes against popular beliefs linking the Maillard reaction with texture.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s16", "title": "Effect of perforated disc height and filter basket on espresso coffee carbohydrates content and composition", "authors": "Khamitova, G.; Passos, C. P.; Lopes, G. R.; Coimbra, M. A.; Mustafa, A. M.; Vittori, S.; Caprioli, G.", "year": 2022, "publisher": "European Food Research and Technology 248(5), 1217–1226", "url": "https://agris.fao.org/search/en/records/65dfdccd0f3e94b9e5dcfb44", "doi": "10.1007/s00217-022-03960-w", "tier": "A", "quote": "the increase of the height of perforated disc affected the extraction of the polysaccharides content in the cup", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" } ], "markdown": "## What carbohydrates green coffee contains\n\nCarbohydrates are sugars and the chains built from them. In green coffee they fall into two groups: free sugars, and polysaccharides, long chains of sugar units.\n\n**Polysaccharides.** They make up about half of the dry weight of green coffee.1 Most of them form the walls of the seed’s cells. A 2006 review names three main types:1\n\n- **Galactomannans.** Chains of mannose with single galactose units attached. They account for about half of the polysaccharides.1 According to a roasting chemistry chapter, they also serve as an energy store for the seed, as starch does in cereals.4\n- **Arabinogalactan-proteins.** Galactose chains with branches that end in arabinose, bound to protein.1,4\n- **Cellulose.** The rigid framework of the cell wall.1\n\n**Free sugars.** In mature seeds, sucrose makes up 5–12% of dry weight and almost all of the free sugar.1 Glucose and fructose are present only in traces.1 According to the roasting chemistry chapter, arabica holds more sucrose than robusta.4 How green coffee is composed overall is covered in [Green coffee composition](/entries/green-coffee-composition/).\n\n## What roasting does to them\n\n**Sugar is consumed.** Sucrose is almost entirely broken down in roasting.1,4 It feeds the Maillard reaction and caramelisation, and gives rise to colour, aroma, carbon dioxide and water.4 How fast this happens during a roast is shown in [Browning phase](/entries/browning-phase/).\n\n**Polysaccharides partly break down.** In a 2002 study of three arabica varieties, roasting degraded 12–40% of the polysaccharides, depending on the roast.2 Arabinogalactans lost up to 60% after a dark roast, mannans up to 36%. Cellulose was barely affected.2\n\n
\n
\n
How much of the green-coffee polysaccharide roasting breaks down, as reported. The bar shows the reported range; dots show the maximum loss after a dark roast.2
\n
\n\n**The arabinose branches go first.** The side chains of the arabinogalactans break off faster than their galactose backbone.1,3 The freed arabinose is a reactive sugar. According to the roasting chemistry chapter, most of it ends up in melanoidins, the brown polymers of roasted coffee.4\n\n**More dissolves.** Roasting cuts the long chains into shorter ones and loosens the cell wall.1 Galactomannans, mostly insoluble in green coffee, become largely extractable with hot water as roasting increases.3 Cellulose stays insoluble whatever the roast.3\n\n## From the bean to the cup\n\nThe polysaccharides that dissolve during brewing are mainly galactomannans and arabinogalactans.3,4 Part of them reach the cup bound into melanoidins.6 How much reaches the cup depends on the brew.\n\n**Espresso extracts more per millilitre.** In a 2017 study of four brew methods with one coffee, espresso held about four to six times more oligosaccharides per volume than the other brews.8 Oligosaccharides are short sugar chains, here of 3–15 units. Per serving, the small espresso and the larger cups gave comparable amounts.8\n\n
\n
\n
Oligosaccharides in four brews of the same coffee, roasted to 25 Agtron, in ppm as the paper prints them. Espresso was significantly higher than the other three.8
\n
\n\n**Temperature matters most.** In a 2019 study that varied time, temperature, ratio and grind, temperature was the main factor separating the extracts.10 How well the coffee extracted depended more on galactomannans than on arabinogalactans.10\n\n**The machine matters too.** In a 2022 espresso study, changing the height of the perforated disc under the basket changed how much polysaccharide reached the cup.16\n\n## Carbohydrates, body and crema\n\n**Viscosity and mouthfeel.** According to the roasting chemistry chapter, polysaccharides add little to aroma but give the brew viscosity and mouthfeel.4 No study opened for this entry measured the effect of polysaccharides on perceived body directly. Dissolved solids, fines and oil also shape body; they are covered in [Body and mouthfeel](/entries/body-and-mouthfeel/) and [Coffee oil](/entries/coffee-oil/).\n\n**Crema.** In a 1997 espresso study, foam stability was related to the amount of galactomannan and arabinogalactan in the cup.11 How much foam formed depended on protein, and rose with roast degree.11 More on crema is in the [espresso](/entries/espresso/) entry.\n\n**Roast and body.** Sources disagree on whether roasting changes body. A roasting textbook chapter describes body as rising with roast degree to a point and then falling.14 In a 2020 analysis of eight roasting studies, the timing of the roast phases did not affect body at all; in the one study that rated body across roast colours, darker roasts scored only slightly higher.15\n\n\n\n## Reference values\n\nValues from the sources opened for this entry. Green coffee data are % of dry matter; studies used different coffees and methods, so compare within a row, not across studies.\n\n| Parameter | Value | Conditions | Source |\n|---|---|---|---|\n| Polysaccharides, green coffee | ≈50% of dry weight | Review; galactomannans, arabinogalactan-proteins and cellulose | Redgwell & Fischer1 |\n| Galactomannans | ≈50% of the polysaccharides | Review | Redgwell & Fischer1 |\n| Galactose : mannose ratio in galactomannans | 1:7 to 1:40 | Mature grains; substitution falls during development | Redgwell & Fischer1 |\n| Sucrose, mature seeds | 5–12% of dry weight; essentially all free sugar | Arabica and robusta, developing-grain studies | Redgwell & Fischer1 |\n| Polysaccharides degraded in roasting | 12–40% | Three arabica varieties, roasts of different intensity | Redgwell et al.2 |\n| Arabinogalactans / mannans degraded | Up to 60% / up to 36% | After a dark roast | Redgwell et al.2 |\n| Oligosaccharides (DP 3–15) in brews | French press 292 ± 11; batch brewer 350 ± 13; pod 418 ± 29; espresso 1,632 ± 88 ppm | One coffee, espresso roast (25 Agtron), n = 3; ppm as printed | Tian et al.8 |\n| Sugar units in coffee oligosaccharides | Galactose 42.2%; mannose 37.1%; glucose ≈9.5%; arabinose ≈7.8% | Hot-water extract of the dark-roasted coffee; gas chromatography after methanolysis; weight % | Tian et al.8 |\n| Molecular weight of brew polysaccharides | Espresso 6 kDa; paper drip 11 kDa; electric drip machine 21 kDa | Gel filtration; espresso 15 g at 4 atm, about 50 mL | Martins et al.9 |\n| Sugar units in brew polysaccharides | Mannose 48–49%; galactose 29–30%; arabinose 14–15%; glucose 6% | Same study, all three methods | Martins et al.9 |\n| Espresso foam fraction (FFA) | ≈80% mannan, ≈20% arabinogalactan (molar basis); molecular weight 34,000–47,000 | Fraction linked to foam stability | Illy & Navarini12 |\n\nNotes. The roasting chemistry chapter gives two tables that differ: one lists more insoluble carbohydrate in arabica than in robusta, the other more total polysaccharide in robusta.4 We give the green-to-roasted figures as printed and do not compare species from them. The Tian et al. values are oligosaccharides, not polysaccharides; the paper does not define ppm further.8\n\n## The chemistry of roasting carbohydrates\n\n**Which go first.** According to the roasting chemistry chapter, sucrose and free amino acids react first, because they are immediately available.4 Arabinose branches react later, because they must first be released by depolymerisation or hydrolysis.4 Galactans, mannans and cellulose resist breakdown and join the Maillard reaction only late in the roast.4 The sequence for sucrose is described in [Maillard reaction](/entries/maillard-reaction/) and [Caramelisation](/entries/caramelisation/).\n\n**Arabinogalactan-proteins.** Arabinose sits at the end of most arabinogalactan side chains. Its furanose ring form makes it labile to hydrolysis.5 Side chains are removed faster than the galactan backbone is cut.1,3 The arabinogalactans are depolymerised already after a light roast.2 Their covalent link to protein survives roasting.2 According to the review, arabinogalactan populations with very different galactose-to-arabinose ratios can be isolated.1\n\n**Galactomannans.** Roasting reduced both the chain length and the branching of the galactomannans; the review describes this as a linearisation.1 The mannans that remain in the bean keep their molecular weight even after a dark roast.2 Roasting also made the cell-wall polymers more accessible to extraction and allowed galactomannans of higher molecular weight to dissolve.1 Sequential extraction of arabica at several roast degrees showed galactomannans moving from the insoluble to the extractable fraction.3\n\n**Small fragments.** In the same sequential study, most roasted-coffee carbohydrate was still polymeric. Oligomers and especially monomers appeared to be converted very rapidly into Maillard and pyrolysis products.3 Oligosaccharides found in brewed coffee are thought, by the authors who measured them, to derive from galactomannans and arabinogalactans; their structures were inferred, not confirmed.8\n\n**Acids.** Sucrose is the principal precursor of formic, acetic, glycolic and lactic acid in roasting.5 A model roast of arabinogalactan from larch wood, similar in structure to the coffee polymer, gave the same four acids, but with relatively more acetic acid.5 The acids themselves are covered in [Acids in coffee](/entries/acids-in-coffee/).\n\n**Melanoidins.** In a 2008 study of brews from light, medium and dark roasts, arabinogalactans appeared relatively more involved in melanoidin formation than galactomannans.6 Galactomannans were continuously incorporated into arabinogalactan-protein melanoidins as roasting went on.6 Longer roasting led mainly to melanoidins of high molecular weight.6 Their structure is discussed in [Melanoidins](/entries/melanoidins/).\n\n**The cell wall as structure.** Polysaccharides also give the bean its mechanical behaviour in the roaster. According to a roasting chapter, cell walls pass between a glassy and a rubbery state, and the bean swells while they are rubbery.14 This is covered in [Glass–rubber transition](/entries/glass-rubber-transition/).\n\n## Extraction: grind, pressure and temperature\n\n**Brew method.** In a 2005 Brazilian comparison of three brew methods with one arabica coffee, espresso gave the highest yield of soluble solids.9 Its polysaccharides were the smallest: 6 kDa, against 11 kDa for paper drip and 21 kDa for the electric machine.9 The sugar make-up of the polymers was nearly the same in all three, with mannose about half.9\n\n**Temperature and ratio.** In the 2019 extraction study, temperature was the major factor that set the extracts apart.10 Extraction efficiency followed galactomannan extraction more closely than arabinogalactan extraction.10\n\n**Pressure and viscosity.** According to a brewing chapter, an earlier study found that higher espresso pressure gave a more viscous brew with more intense body.7 The chapter does not say which compounds carried the change.\n\n**Bed swelling.** The same chapter notes that particles in an espresso bed swell on wetting, because roasted coffee contains water-insoluble polysaccharides.7 Swelling and particle rearrangement consolidate the bed, lower its porosity and slow the flow.7\n\n**Basket set-up.** A 2022 study varied the perforated-disc height and filter basket for arabica and robusta espresso at 12 and 14 g. A higher disc reduced galactomannans and arabinogalactans in the cup.16\n\n**Soluble coffee.** In industrial extraction for soluble coffee, how far the galactomannans dissolve is a critical factor in the yield of powder.1\n\n## Polysaccharides and foam\n\n**Foam stability.** The 1997 study roasted a Brazilian natural arabica and a Ugandan natural robusta to several degrees and made espresso from each.11 Foamability rose with roast degree and depended on the protein in the brew.11 Foam stability was related to galactomannan and arabinogalactan, and seemed independent of origin.11 The authors concluded that maximum foamability and maximum foam stability cannot be had in the same espresso.11\n\n**Foam fractions.** A 2011 review describes two foaming fractions isolated from espresso. One, of about 80% mannan and 20% arabinogalactan with a molecular weight of 34,000–47,000, was involved in foam stability.12 A second fraction, of lower molecular weight and higher surface activity, was linked to foamability.12 The review cites work that ascribes the loss of foam persistence at high roast to depolymerisation of polysaccharide-derived Maillard products, which lowers the viscosity of the liquid.12\n\n**Other actors.** A crema chapter adds that a strong correlation was found between foam stability and high-molecular-weight compounds, suggested to be complexes of polysaccharide, protein and phenolic compounds.13 It names 4-vinylcatechol oligomers, not polysaccharides, as key foam-active compounds in the crema of roasted coffee.13 According to the brewing chapter, water hardness lowers crema stability, possibly through cations interacting with protein–polysaccharide complexes.7\n\n## How strong the link to body is\n\n**What is stated.** A textbook chapter states that polysaccharides give the brew viscosity and mouthfeel, citing the 2002 roasting study.4 A second chapter reports that pressure raised viscosity and body in espresso.7 Both are Tier B and come from the same book.\n\n**What was measured.** The studies opened for this entry measured polysaccharides in beans and brews and their link to foam.2,3,11 None of them measured viscosity or rated body against polysaccharide content. A 2020 analysis of eight roasting studies found no effect of roast phase timing on body, against the popular idea that larger molecules from the Maillard reaction thicken the cup.15\n\n**What else counts.** Dissolved solids, fines, oil and melanoidins all bear on body. The evidence for each is set out in [Body and mouthfeel](/entries/body-and-mouthfeel/).\n\n## Measuring coffee carbohydrates\n\n- **Sequential extraction.** Hot water at 90 °C, then at 170 °C, then cold dilute alkali, compared with the green beans of the same batch.3\n- **Ethanol precipitation.** Polymeric carbohydrates in espresso were precipitated in 55% and 75% ethanol fractions.11\n- **Monosaccharide analysis.** Polymers are hydrolysed and the sugar units measured by gas chromatography.8,11\n- **Molecular weight.** Gel filtration on a Sepharose column separates polymers by size.9\n- **Antibodies.** Labelling of arabinogalactan side chains fell in both arabica and robusta even after the lightest roast.1\n\n## Open questions\n\n- **Polysaccharides and perceived body.** No study opened for this entry related polysaccharide content in the cup to viscosity or rated body. The link rests on textbook statements.4,7\n- **Cellulose.** The review notes no published work on the characteristics of coffee cellulose.1\n- **Interactions in the roast.** The review lists interactions of polysaccharides with other molecules, such as proteins and melanoidins, during roasting among the points still to be understood.1\n- **Oligosaccharide structures.** The origin of brew oligosaccharides from galactomannans and arabinogalactans is inferred, not confirmed.8\n- **Roast profile.** The studies opened here compared roast degrees, not profiles at the same colour. How time and temperature change polysaccharide breakdown at equal colour is not established.2,3\n\n## Key studies\n\n
\n

Redgwell et al., 2002 – Polysaccharides through the roast. Three arabica varieties were roasted to different degrees and their cell-wall polysaccharides analysed. Between 12% and 40% of the polysaccharides were degraded; arabinogalactans lost up to 60% and mannans up to 36% after a dark roast, while cellulose barely changed.

\n

Carbohydrate Research 337(5), 421–431 · AGRIS record

\n
\n\n
\n

Oosterveld et al., 2003 – What roasting makes extractable. Arabica beans at several roast degrees were extracted step by step with hot water and alkali and compared with the green beans. Galactomannans became largely extractable with roasting, arabinose side chains broke down faster than the galactan backbone, and cellulose stayed insoluble.

\n

Carbohydrate Polymers 54(2), 183–192 · doi:10.1016/S0144-8617(03)00164-4

\n
\n\n
\n

Nunes et al., 1997 – Espresso foam and roast degree. A natural arabica and a natural robusta were roasted to several degrees and brewed as espresso. Foamability rose with roast and followed protein; foam stability was related to galactomannans and arabinogalactans.

\n

Journal of Agricultural and Food Chemistry 45(8), 3238–3243 · doi:10.1021/jf970009t

\n
\n\n
\n

Bekedam et al., 2008 – Polysaccharides in melanoidins. Brews from green, light, medium and dark roasts were fractionated to follow how melanoidins form. Arabinogalactans were relatively more involved than galactomannans, and galactomannans kept being incorporated as roasting went on.

\n

Journal of Agricultural and Food Chemistry 56(16), 7138–7145 · doi:10.1021/jf800999a

\n
\n\n
\n

Tian et al., 2017 – Oligosaccharides by brew method. One coffee at espresso roast (25 Agtron) was brewed by French press, batch brewer, pod machine and espresso, and the short sugar chains in each brew were measured. Espresso held about four to six times more per volume, but servings delivered comparable amounts.

\n

Journal of Agricultural and Food Chemistry 65, 2784–2792 · doi:10.1021/acs.jafc.6b04716

\n
" }