{ "id": "bean-size-and-density", "url": "https://coffearium.org/entries/bean-size-and-density/", "license": "CC-BY-4.0", "title": "Bean size and density", "summary": "How large and how dense green coffee beans are, how both are measured, and what they mean for roasting – heat uptake, evenness, and the swelling that lowers density as the roast goes on.", "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": [ "Bean size is graded on screens whose numbers give the hole size in 64ths of an inch.2", "“Density” means different things – bulk density counts the air between beans, bean density only the bean with its pores – so values differ by method.3,7,11", "Bulk densities of 775–880 g/L were measured with meters in seven green coffees from two studies.5,6", "Bean density falls continuously during roasting as beans gain about 50% to almost 100% in volume.1,17", "At the same colour, a fast roast gave lower bean density than slow roasts in one study, 0.548 against 0.584–0.592 g/cm³.9,11" ], "quickFacts": [ { "label": "Size", "value": "Screen number = hole size in 1/64 in (ISO 4150)" }, { "label": "Bulk density", "value": "Mass ÷ volume a heap of beans fills, air gaps included (ISO 6669)" }, { "label": "Bean density", "value": "Mass ÷ volume of the beans themselves, measured by liquid displacement" }, { "label": "Green, bulk (meter)", "value": "775–880 g/L in seven coffees (two studies)" }, { "label": "Roasted, bean density at L* 21", "value": "0.548–0.592 g/cm³ (one arabica, four roasting processes)" }, { "label": "Volume in roasting", "value": "About +50% to almost +100%" }, { "label": "Evidence on roasting rules", "value": "Practitioner rules for dense coffees not tested in the studies opened" } ], "aka": [ "Green coffee density", "Bean density", "Coffee bean size" ], "linkTerms": [ "bean size and density", "bean density", "green coffee density", "green bean density", "density of green coffee", "free-flow bulk density", "particle density" ], "related": [ "grading-and-sorting", "green-coffee-grading", "heat-transfer", "roast-loss", "preheating-and-charge", "moisture-and-water-activity", "drying-phase", "porosity-and-cell-structure", "glass-rubber-transition", "water-in-roasting", "endothermic-and-exothermic-phases", "caffeine-and-roast-degree", "roast-degree-and-colour", "roast-defects", "peaberries-and-bean-shapes", "fruit-development" ], "contested": [ "Density values for green coffee: a 2013 review gives 0.6–0.7 g/cm³ for green arabica and robusta without naming the method4; density meters gave bulk densities of 0.775–0.880 g/mL5,6; oil displacement gave 1.154–1.235 g/cm³ for seeds without parchment during drying7; and a 2018 study cites about 1.3 for green beans.8 The methods count different volumes, so the numbers are not comparable without the method." ], "corrections": [ { "date": "2026-10-20T00:00:00.000Z", "change": "The 2023 calorimetry study's link between density and heat demand is now reported as a suggestion, and its conclusion is noted to name density and moisture together.", "reason": "Verification pass; the paper writes that the correlation \"suggests\" that denser green coffee requires more heat, and its conclusion names \"density and humidity\"." }, { "date": "2026-10-20T00:00:00.000Z", "change": "The larger share of beans on screens 18–20 at 935 m in the Mozambique study is now given as a tendency.", "reason": "Verification pass; the paper says the beans \"tended to have a greater presence\" on these screens." }, { "date": "2026-10-20T00:00:00.000Z", "change": "Removed the uncited statements that defective beans weigh less for their size, that bean density is also called particle density, and that solid density would need a gas pycnometer; the size dependence of drying rate is now attributed to an earlier model used by the 2025 study.", "reason": "Verification pass; no opened source supports these statements as written, and the 2025 study takes the size term from earlier work rather than measuring it." } ], "sources": [ { "id": "s1", "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": "The bean density decreases continuously.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s2", "title": "ISO 4150:2011 Green coffee or raw coffee — Size analysis — Manual and machine sieving", "authors": "International Organization for Standardization (ISO/TC 34/SC 15)", "year": 2011, "publisher": "ISO", "url": "https://cdn.standards.iteh.ai/samples/44602/3720908855384c8a82995149ad78e106/ISO-4150-2011.pdf", "tier": "A", "tierNote": "Standard; A for what it specifies (sieve definitions, sample, procedure); opened as the publisher's preview", "quote": "expressed in 64ths of an inch, closest to the metric dimension adopted.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s3", "title": "ISO 6669:1995 Green and roasted coffee — Determination of free-flow bulk density of whole beans (Routine method)", "authors": "International Organization for Standardization (ISO/TC 34/SC 15)", "year": 1995, "publisher": "ISO (preview pages via ANSI Webstore)", "url": "https://webstore.ansi.org/preview-pages/ISO/preview_ISO+6669-1995.pdf", "tier": "A", "tierNote": "Standard; A for what it specifies and defines. Only the cover, foreword and introduction were openly available; the ISO catalogue lists the 1995 edition as confirmed in 2018.", "quote": "Bulk density is defined as the ratio of mass to volume occupied.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s4", "title": "Quality assessment of Arabica and Robusta green and roasted coffees – A review", "authors": "Bicho, N. C.; Lidon, F. C.; Ramalho, J. C.; Leitão, A. E.", "year": 2013, "publisher": "Emirates Journal of Food and Agriculture 25(12), 945–950", "url": "https://ejfa.me/index.php/journal/article/download/1252/924", "doi": "10.9755/ejfa.v25i12.17290", "tier": "A", "quote": "apparent densities of Arabica and Robusta green coffee do not vary significantly", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s5", "title": "Analysis of coffee thermophysical changes during roasting using differential scanning colorimetry", "authors": "Mutovkina, E. A.; Bredikhin, S. A.", "year": 2023, "publisher": "Food Science and Technology (Campinas) 43, e119722", "url": "https://new.scielo.br/j/cta/a/Ffv4BQ5nwkLLprXwNLvF8rP/?lang=en", "doi": "10.1590/fst.119722", "tier": "A", "quote": "green coffee, which has a higher density, requires the use of more heat during the roasting process.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s6", "title": "The Effect of Roast Development Time Modulations on the Sensory Profile and Chemical Composition of the Coffee Brew as Measured by NMR and DHS-GC–MS", "authors": "Alstrup, J.; Petersen, M. A.; Larsen, F. H.; Münchow, M.", "year": 2020, "publisher": "Beverages 6(4), 70", "url": "https://www.mdpi.com/2306-5710/6/4/70", "doi": "10.3390/beverages6040070", "tier": "A", "quote": "The growing altitude was 1200–1800 m above sea level and the beans had a moisture content of 10%, density of 880 g/L (SINAR 6070 Grainpro) and the beans were of screen size 17.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s7", "title": "Hygroscopic equilibrium and physical properties evaluation affected by parchment presence of coffee grain", "authors": "Corrêa, P. C.; Oliveira, G. H. H.; Rodrigues, A. P. L.; Campos, S. C.; Botelho, F. M.", "year": 2010, "publisher": "Spanish Journal of Agricultural Research 8(3), 694–702", "url": "https://sjar.revistas.csic.es/index.php/sjar/article/download/1267/1211", "tier": "A", "quote": "The ρt ranged from 1,234.67 to 1,154.13 kg m–3 and from 802.78 to 1,064.99 kg m–3 for the PR and PP coffee, respectively.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s8", "title": "Time-Resolved Gravimetric Method To Assess Degassing of Roasted Coffee", "authors": "Smrke, S.; Wellinger, M.; Suzuki, T.; Balsiger, F.; Opitz, S. E. W.; Yeretzian, C.", "year": 2018, "publisher": "Journal of Agricultural and Food Chemistry 66(21), 5293–5300", "url": "https://digitalcollection.zhaw.ch/bitstream/11475/12175/2/2017_Smrke-etal_Roasted-coffe-degassing-time-resolved-gravimetry.pdf", "doi": "10.1021/acs.jafc.7b03310", "tier": "A", "tierNote": "The density values appear in the introduction as background, citing earlier work; the paper prints the unit as kg/m³.", "quote": "average density decreases from about 1.3 kg/m3 for green beans", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s9", "title": "Coffee Roasting and Aroma Formation: Application of Different Time−Temperature Conditions", "authors": "Baggenstoss, J.; Poisson, L.; Kaegi, R.; Perren, R.; Escher, F.", "year": 2008, "publisher": "Journal of Agricultural and Food Chemistry 56(14), 5836–5846", "url": "https://synapsesocial.com/papers/W1969766614", "doi": "10.1021/jf800327j", "tier": "A", "tierNote": "Quote from the abstract; methods and results (Table 4.2, bean density) read in full in the first author's 2008 ETH thesis (s11), chapter 4, where the study is reprinted in adapted form.", "quote": "Excessive roasting generally led to decreasing or stable amounts of volatile substances.", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s10", "title": "Water content of roasted coffee: impact on grinding behaviour, extraction, and aroma retention", "authors": "Baggenstoss, J.; Perren, R.; Escher, F.", "year": 2008, "publisher": "European Food Research and Technology 227(5), 1357–1365", "url": "https://agris.fao.org/search/ru/records/65df955b7c7033e84bee3b18", "doi": "10.1007/s00217-008-0852-8", "tier": "A", "tierNote": "Quote from the abstract; methods and results (Table 8.1, bulk density) read in full in the first author's 2008 ETH thesis (s11), chapter 8, where the study is reprinted in adapted form.", "quote": "Coffees with low water content were extracted more effectively than high moisture coffees, and percolation was slower.", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s11", "title": "Coffee roasting and quenching technology – formation and stability of aroma compounds (Diss. ETH No. 17696)", "authors": "Baggenstoss, J.", "year": 2008, "publisher": "ETH Zürich (doctoral thesis)", "url": "https://doi.org/10.3929/ethz-a-005666902", "doi": "10.3929/ethz-a-005666902", "tier": "B", "tierNote": "Doctoral thesis (Tier B). Its chapters 4 and 8 reprint the peer-reviewed studies s9 and s10 in adapted form; values from those chapters are cited to both.", "quote": "High temperature roasting led to lower density, higher bean volume, less roast loss, and lower moisture content compared to roasting processes at lower temperature.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s12", "title": "Physical and chemical attributes of defective crude and roasted coffee beans", "authors": "Franca, A. S.; Oliveira, L. S.; Mendonça, J. C. F.; Silva, X. A.", "year": 2005, "publisher": "Food Chemistry 90(1–2), 89–94", "url": "https://phtnet.agri.cmu.ac.th/research/download/pdf/wy353.pdf", "tier": "A", "quote": "both black and sour beans roast to a lesser degree than the other types of beans", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s13", "title": "Shade and altitude implications on the physical and chemical attributes of green coffee beans from Gorongosa Mountain, Mozambique", "authors": "Cassamo, C. T.; Mangueze, A. V. J.; Leitão, A. E.; Pais, I. P.; Moreira, R.; Campa, C.; Chiulele, R.; Reis, F. O.; Marques, I.; Scotti-Campos, P.; Lidon, F. C.; Partelli, F. L.; Ribeiro-Barros, A. I.; Ramalho, J. C.", "year": 2022, "publisher": "Agronomy 12(10), 2540", "url": "https://www.mdpi.com/2073-4395/12/10/2540", "doi": "10.3390/agronomy12102540", "tier": "A", "quote": "bean commercial homogeneity, mass, and density increased at 935 m", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s14", "title": "Insight into the effects of environmental variables on the physicochemical characteristics and biochemical composition of green Arabica coffee", "authors": "Urugo, M. M.; Tola, Y. B.; Kebede, B. T.; Ogah, O.", "year": 2024, "publisher": "Beverage Plant Research 4, e021", "url": "https://www.maxapress.com/article/doi/10.48130/bpr-0024-0021", "doi": "10.48130/bpr-0024-0021", "tier": "A", "quote": "an elevation increase corresponds to a concurrent rise in the weight of the hundred beans", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s15", "title": "Developing predictive models for determining physical properties of coffee beans during the roasting process", "authors": "Bustos-Vanegas, J. D.; Corrêa, P. C.; Martins, M. A.; Baptestini, F. M.; Campos, R. C.; de Oliveira, G. H. H.; Nunes, E. H. M.", "year": 2018, "publisher": "Industrial Crops and Products 112, 839–845", "url": "https://agris.fao.org/search/ru/records/674845816b7cc10eeb5aaa24", "doi": "10.1016/j.indcrop.2017.12.015", "tier": "A", "quote": "The bean density varied linearly with the moisture content, presenting a larger drop at a higher roasting temperature.", "abstractOnly": true, "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s16", "title": "Modeling and simulation of coffee bean heating during roasting: effect of heat generation", "authors": "Bustos-Vanegas, J. D.; Martins, M. A.; Corrêa, P. C.; Baptestini, F. M.; de Oliveira, G. H. H.", "year": 2025, "publisher": "Frontiers in Food Science and Technology 5, 1603783", "url": "https://www.frontiersin.org/journals/food-science-and-technology/articles/10.3389/frfst.2025.1603783/full", "doi": "10.3389/frfst.2025.1603783", "tier": "A", "quote": "The volume expansion leads to an increase in surface area and a decrease in the specific weight of grain", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s17", "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": "From a physical perspective, it gains from around 50% to almost 100% increased volume", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s18", "title": "Green Coffee Analytics Part IV: Density", "authors": "Kornman, C.", "year": 2016, "publisher": "Royal Coffee (green coffee importer, blog)", "url": "https://royalcoffee.com/green-coffee-analytics-part-iv-density/", "tier": "C", "quote": "First, we tend to charge with a hot drum temperature, but a low flame.", "accessed": "2026-10-20T00:00:00.000Z" }, { "id": "s19", "title": "Beyond Elevation: How Bean Density Informs Roasting and Flavor", "authors": "Lee, S.", "year": 2025, "publisher": "Roast Magazine, issue 129 (May/June 2025)", "url": "https://www.roastmagazine.com/stories/beyond-elevation", "tier": "C", "quote": "denser seeds also tend to contain more water", "accessed": "2026-10-20T00:00:00.000Z" } ], "markdown": "## Two properties of green coffee\n\nRoasters look at two physical properties of green coffee before they roast it: how large the beans are and how dense they are. A roasting textbook chapter names green bean density, moisture content, screen size and shape as factors in the roast plan.1\n\n**Size** is graded on screens. Each screen number gives the hole size in 64ths of an inch, so screen 18 has holes of 18/64 in (7.14 mm).2 How screens and sorting machines work is described under grading and sorting.\n\n**Density** is mass divided by volume. The difficulty is which volume. A heap of beans fills a container with beans and with air between them. One bean alone has its own volume, which includes small pores inside it.3,7 These two volumes give two different densities.\n\n
\n
\n
Bean density divides a bean’s mass by its own volume; bulk density divides the mass of many beans by the volume they fill, air gaps included.3,7,11 Schematic, not to scale.
\n
\n\n## How size and density are measured\n\n**Screen size.** A weighed sample is shaken through a nest of screens, and the share left on each screen is weighed. ISO 4150 describes this method for green coffee.2 The details, sample sizes and grade names are in green coffee grading standards.\n\n**Bulk density.** Beans fall freely into a container of known volume, the surface is levelled, and the beans are weighed.3 ISO 6669 standardises this routine method for whole green and roasted beans.3 Studies also report bulk density read by grain meters.5,6 Practitioners describe a simple version of the same principle: a graduated cylinder filled to a mark and weighed.18,3\n\n**Bean density.** Here the volume of the beans themselves is measured. A weighed sample is dipped into oil, and the oil it displaces gives its volume.11,7 The reading is taken quickly, so that oil does not enter the pores.11 Studies call this value bean density or, in some cases, true density.11,7\n\nAlways check which method a number comes from. In a 2010 study, coffee seeds without parchment had a bulk density of 393–534 kg/m³ but an oil-displacement density of 1,154–1,235 kg/m³ during drying.7\n\n## What makes beans larger or denser\n\nAccording to ISO 6669, botanical, horticultural, processing, storage and handling factors, including age, all influence the bulk density of green coffee.3 The value also depends on the mass, size and shape of the beans and, less, on their moisture.3\n\n**Altitude.** Two field studies link higher plantings to heavier or denser beans. On a mountain in Mozambique, beans grown at 935 m had higher mass and density than beans from 650 m.13 They also tended to sit more often on screens 18–20.13 Across Ethiopian samples, hundred-bean weight rose with elevation.14 The Mozambique study covered one cultivar over a narrow altitude band.13 No study opened for this entry tested altitude as a predictor of density across many origins.\n\n**Moisture.** Water changes both values. During drying, the oil-displacement density of seeds fell as their moisture rose.7 Practitioners report that denser seeds also hold more water.19 In one study of six green coffees, the denser ones were indeed the moister ones.5 Moisture is covered in moisture and water activity.\n\n**Defects.** Defective beans differ from sound beans in their physical properties. A 2005 study measured bean and bulk density of defective beans and could tell defect types apart by physical and chemical properties.12 Dry mills remove light beans on gravity tables, as described under grading and sorting.\n\n## Size, density and the roast\n\nPractitioners give dense coffees more heat. One importer’s roaster charges with a hot drum but a low flame, and reports that dense coffees tend to stall later in the roast.18 A trade magazine describes higher charge temperatures and more aggressive early heat for dense, high-grown coffees.19 These are practice reports. No peer-reviewed study opened for this entry tested them.\n\n**What studies show.** In a 2023 calorimetry study of six green coffees, the authors suggest that denser coffee needs more heat in roasting.5 In those samples, the denser coffees were also the moister ones (Deep dive).5 A 2025 modelling study uses a drying equation from earlier work in which smaller beans lose moisture faster.16\n\n**Mixed sizes.** A 2013 review says that green coffee must be even in size for roasting, so that smaller beans do not burn.4 Size grading at origin serves this purpose. How heat travels from the surface to the core of a bean is explained under heat transfer.\n\n## Density falls during roasting\n\nRoasting turns a dense seed into a light, porous bean. The beans gain about 50% to almost 100% in volume, while they lose weight.17,1 According to a roasting textbook chapter, bean density decreases continuously as the roast goes on.1 The weight loss is described under roast loss.\n\nRoast speed matters as well as roast degree. At the same colour, fast-roasted beans have greater volume and porosity and lower density than slow-roasted beans.1,9,11 In one study, one arabica roasted to the same colour in four ways ended between 0.548 and 0.592 g/cm³.9,11\n\n
\n
\n
Bean density of one washed Sumatran arabica roasted to the same lightness (L* 21) in a laboratory fluid bed at three time–temperature settings and in a production drum, measured by oil displacement.9,11
\n
\n\nDensity therefore says something about a roast, but not on its own, because it falls with roast degree and with roast speed.1,11 One doctoral thesis monitored its roasts by weight, colour, density, firmness, porosity and degassing together.11 Roasted coffee also fills more space: an industrial study measured 335–358 g/L, against 775–880 g/L for green coffee in other studies.10,11,5,6 What this means for dosing by volume is shown under caffeine and roast degree.\n\n\n\n## Reference values\n\nValues as reported. Methods differ; compare only values measured the same way. 1 g/cm³ = 1 g/mL = 1,000 kg/m³ = 1,000 g/L.\n\n| Parameter | Value reported | Method and conditions | Source |\n|---|---|---|---|\n| Screen number | Hole size in 64ths of an inch, closest to the metric aperture | ISO 4150 | ISO2 |\n| Bulk density, definition | Mass ÷ volume occupied | ISO 6669, free flow into a container, levelled; whole beans only | ISO3 |\n| Bulk density, green, six coffees | 775–836 g/L (arabica from Brazil, Colombia, Ethiopia, Thailand; one canephora from Brazil 836 g/L) | Lightells MD-500 meter; moisture 8.6–11.7% (basis not stated) | Mutovkina & Bredikhin5 |\n| Bulk density, green, one lot | 880 g/L | SINAR 6070 meter; washed Colombian arabica, screen 17, 1,200–1,800 m, moisture 10% (basis not stated) | Alstrup et al.6 |\n| Density, green, review | 0.6–0.7 g/cm³ for arabica and robusta, not significantly different | Method not stated (“apparent density”) | Bicho et al.4 |\n| Seed without parchment, during drying | Bulk 393–534 kg/m³; oil displacement 1,154–1,235 kg/m³ (falling with rising moisture) | Hectolitre balance; soybean-oil displacement; cv. Catuaí Amarelo | Corrêa et al.7 |\n| Average bean density, green → dark roast | About 1.3 → about 0.7 (unit printed as kg/m³; g/cm³ fits the other values, our reading); porosity about 0.5 in dark roast | Background values cited from earlier work | Smrke et al.8 |\n| Volume gain in roasting | About +50% to almost +100%; up to a factor of 2; up to 1.8 × the green volume (drum, 200–280 °C) | Literature summary; textbook; X-ray microtomography | Münchow et al.17, Schenker & Rothgeb1, Bustos-Vanegas et al. 201815 |\n| Volume gain, review | +40–69% arabica; +48–57% robusta | Review | Bicho et al.4 |\n| Bean density, roasted to L* 21 | 0.548 ± 0.001 (fluid bed, 260 °C, 160 s); 0.584 ± 0.005 (profile 180→232 °C, 900 s); 0.588 ± 0.002 (228 °C, 660 s); 0.592 ± 0.002 g/cm³ (drum, 20 kg, 1,170 s) | Peanut-oil displacement, 30 g beans; washed Sumatran arabica, green moisture 10.04 g/100 g (wb) | Baggenstoss et al.9,11 |\n| Roasted, longer roasting, review | Down to 0.4–0.3 g/cm³ | Method not stated | Bicho et al.4 |\n| Bulk density, roasted, industrial | 335–358 g/L | Tangential roaster, 100–170 kg batches, L* 22.6–23.5, after quenching to 2.3–8.8 g/100 g water (wb); method not stated | Baggenstoss et al.10,11 |\n\n
\n
\n
Densities of green and roasted coffee reported in the sources, labelled by what was measured. Bulk values sit lower than bean values because they include the air between beans. The 2018 values are printed in kg/m³ and read here as g/cm³.4,5,6,7,8,9,10
\n
\n\n## Three densities and the space between beans\n\nThree quantities share the name density. **Bulk density** counts the volume a heap of beans fills, including the gaps.3 **Bean density** counts the volume inside each bean’s outer surface, including its internal pores; it is measured by liquid displacement.11,7 Some studies call the displacement value true density.7 Others report an “apparent density” without naming the method.4,13 A third value, the density of the solid matter without pores, is not reported for coffee in any study opened for this entry.\n\nThe two measured densities give the share of volume between the beans, the intergranular porosity: ε = 1 − ρbulk ÷ ρbean.7 As an illustration only (our calculation, with values from two different studies), a green bulk density of 0.83 g/mL and a bean density of 1.2 g/cm³ give ε ≈ 0.31.5,7 Packing depends on bean shape, size spread and filling, which is why ISO 6669 fixes the free-flow filling and the levelling.3\n\nRoasting changes the inside of the bean as well. A 2018 study cites a porosity of about 0.5 in dark-roasted beans.8 How the pores form is the subject of porosity and cell structure; why the cell walls can stretch at all is explained under glass–rubber transition.\n\n## Why density falls\n\nDensity falls because volume rises while mass falls. A 2025 modelling study states that volume expansion increases surface area and decreases the specific weight of the bean.16 In a 2018 study in a drum at five temperatures from 200 to 280 °C, bean density varied linearly with moisture content and dropped more at higher roasting temperature.15 Expansion was isotropic above 220 °C, and final volume reached up to 1.8 times the green volume.15\n\nAccording to the roasting textbook chapter, bean density falls continuously, but swelling levels off at some point.1 According to the chapter, faster roasts form gas faster and expand faster, giving lower density at the same colour.1 The 2008 study is consistent with this: its fastest roast (160 s) had the lowest density, the production drum roast (1,170 s) the highest.9,11 According to the same thesis, light roasts from moister green coffee ended denser, while in dark roasts these differences disappeared.11\n\n## Size, heat and evenness: the evidence\n\n**Geometry.** For beans of the same shape, surface area per unit volume is inversely proportional to size (our calculation). A screen-18 bean is 18/15 = 1.2 times as wide as a screen-15 bean, so it has about 17% less surface per unit volume (1 ÷ 1.2 ≈ 0.83). How heat moves into the bean is covered under heat transfer. No study opened for this entry measured how much later the core of a larger bean heats in a roaster.\n\n**Drying rate.** A 2025 modelling study uses a drying-rate equation, taken from earlier work, that depends on bean size and is higher in smaller beans.16 The study modelled single beans above screen 18; it did not compare sizes itself.16\n\n**Heat demand.** The 2023 calorimetry study suggests that denser green coffee needs more heat; its conclusion names density and moisture together.5 Its two densest arabicas (831 and 834 g/L) were also the moistest (11.7 and 11.6%, basis not stated), and the lightest (775 g/L) the driest (8.6%).5 Water takes heat to warm and evaporate, so with six samples the effect of density cannot be separated from that of moisture (our reading). Calorimetry was done on 10 mg cross-sections, not on whole beans in a roaster.5\n\n**Practice.** In roasting practice, dense coffees are given a hot charge or more early heat, and they are said to stall before first crack.18,19 The importer’s roaster states that it has no solid research on density and sensory quality.18 The magazine article warns that elevation alone should not dictate roasting decisions.19 How charge temperature works is covered under preheating and charge, and the early phase under drying phase.\n\n## Measuring size and density\n\n- **Report the method.** State bulk or bean density, the instrument or container, the unit, and the moisture of the sample (with its basis).3,7\n- **Bulk density.** ISO 6669 is a routine method for whole beans that flow freely; accuracy depends on correct levelling.3 It covers whole beans only.3\n- **Bean density.** Weigh a sample of about 30 g, immerse it in oil of known density in a wire basket, and read the weight of oil displaced.11 Shake out air bubbles, but read quickly so that oil does not enter the pores.11\n- **Screen size.** Sieve a 300 g sample as in ISO 4150 and report the mass share on each screen.2\n- **Meters.** Grain meters that report bulk density are used in research; their model is named in the studies.5,6\n\n## Risks and defects\n\n- **Defective beans.** Black and sour beans roasted to a lesser degree than other beans under the same conditions in a 2005 study.12 Pale beans after roasting (quakers) are described under roast defects.\n- **Uneven size.** Small beans in a mixed lot can burn, according to a 2013 review.4\n- **Misleading numbers.** A green coffee “density” given without method may be a bulk value near 0.8 g/mL or a bean value above 1 g/cm³.5,7\n- **Quenched coffee.** Industrial bulk density was measured on coffees quenched with different amounts of water.10,11 Added water changes both mass and the reading, so compare roasted bulk densities only at known moisture.\n\n## Open questions\n\n- **Density as a roasting guide.** No peer-reviewed study opened for this entry roasted coffees that differ only in density at equal moisture and size, and measured roast course or flavour.\n- **Size and evenness.** The colour spread within a batch of mixed screen sizes has not been measured in the sources opened.\n- **Altitude and density.** The two field studies opened cover one Mozambican slope and Ethiopian samples.13,14 Whether altitude predicts density across origins, once cultivar and processing are fixed, is not established.\n- **Solid density.** No value for the density of coffee solids without pores was found in the sources opened.\n- **Density and cup quality.** Whether density predicts cup quality better than screen size is open, as noted under grading and sorting.\n\n## Key studies\n\n
\n

Baggenstoss et al., 2008 – Roasting speed and bean properties. Roasts one washed Sumatran arabica to the same colour fast and slow in a laboratory fluid bed, with a temperature profile, and in a production drum. Faster, hotter roasting gave lower bean density, higher volume and lower roast loss; the drum roast was the densest.

\n

Journal of Agricultural and Food Chemistry 56(14), 5836–5846 · doi:10.1021/jf800327j

\n
\n\n
\n

Bustos-Vanegas et al., 2018 – Physical properties during roasting. Follows arabica beans roasted in a drum at five temperatures with X-ray microtomography. Volume reached up to 1.8 times the green volume, and bean density fell linearly with moisture, faster at higher temperatures.

\n

Industrial Crops and Products · doi:10.1016/j.indcrop.2017.12.015

\n
\n\n
\n

Mutovkina & Bredikhin, 2023 – Density and heat in six coffees. Measures bulk density and moisture of six green coffees and scans slices by calorimetry. The authors link higher density to a higher heat demand; in their samples, density and moisture rose together.

\n

Food Science and Technology (Campinas) 43, e119722 · doi:10.1590/fst.119722

\n
\n\n
\n

Cassamo et al., 2022 – Altitude, shade and bean properties. Compares one Catimor cultivar grown at 650, 825 and 935 m on Gorongosa Mountain, Mozambique, under shade and full sun. Bean mass and apparent density were highest at 935 m, where beans tended to sit more often on screens 18–20.

\n

Agronomy 12(10), 2540 · doi:10.3390/agronomy12102540

\n
\n\n
\n

Corrêa et al., 2010 – Physical properties of drying coffee. Measures bulk density, oil-displacement density and porosity of arabica seeds with and without parchment during drying. Without parchment, the displacement density was 1,154–1,235 kg/m³ against a bulk density of 393–534 kg/m³.

\n

Spanish Journal of Agricultural Research 8(3), 694–702 · full text

\n
" }