CoffeariumBeta
  1. Plant
  2. Farm
  3. Process
  4. Trade
  5. Roast
  6. Brew
  7. Cup
Extraction · Fundamental

What extraction is

How water takes coffee out of the grounds – what dissolves and what is only carried along, how broken and intact cells give up their contents, and why the brew changes as extraction goes on.

Across the chain: Science

Source-checkedby Coffearium verification pass· Last source-checked 20 Oct 2026· Prepared with AI assistance· 14 sources (12 tier A)· 2 contested· Updated 20 Oct 2026
Reading depth
Contents
  • Only part of roasted coffee can dissolve in water – typically about one third of its mass, according to a 2021 model study.3
  • Each coffee particle gives up its solubles on two time scales – fast from cells broken open by grinding, slowly by diffusion out of intact cells inside.1,2
  • Compounds leave at different rates. In a 2023 espresso study, trigonelline and total dissolved solids fell fastest across the shot, then 5-CQA and caffeine.7
  • In a 2016 espresso study, the most polar aroma compounds peaked first in the flow; the least polar were still coming out after 25 s.6
  • In full immersion, extraction slows as the liquid approaches equilibrium with the grounds; a smaller concentration gradient means slower transfer.3

What leaves the grounds#

Extraction is the transfer of material from ground coffee into water. According to a coffee science textbook, the water extracts molecules selectively: not every molecule leaves the particles in the same way.4 The cup therefore tastes different from the ground coffee it came from.4

Not everything that leaves dissolves. A 2016 espresso study describes soluble compounds dissolving, while some insoluble material is washed out with the water. Depending on the brew method, this ends up in the cup as dissolved or suspended solids.6 A textbook chapter describes espresso as oil droplets dispersed in the liquid (an emulsion), solid particles held in it (a suspension) and gas bubbles forming a foam.5 How much of this a filter holds back is covered in the body and mouthfeel entry.

Only part of the coffee can dissolve at all. According to a 2021 model study, typically about one third of the mass of roasted coffee is soluble in water.3 The rest stays behind as spent grounds. Extraction yield and TDS measure the dissolved part only; that entry explains how they are measured.

Gases behave differently from solids. According to the textbook, gases dissolve less well in hotter water, so hot brewing drives aroma compounds out of the liquid into the air.4 This adds to the aroma above the cup but lowers the amount of the most volatile compounds in it.4 Carbon dioxide leaving fresh coffee is covered in the entry on CO₂ and the bloom.

How water gets coffee out of a particle#

Roasted coffee is a tissue of cells. In one 2023 microtomography study, the cells measured 25–50 µm, and the smallest ground particles were cell fragments.13 Grinding cuts through cells at every particle surface. The cells deeper inside a particle stay closed.2

Wetting. According to the textbook, water first wets the bed and dissolves the most soluble, low-molecular-weight compounds and volatile aroma compounds.4 At the same time the particles swell, because roasted coffee contains polysaccharides that do not dissolve.4 In the 2016 espresso study, the most volatile compounds showed a first small peak at 4 s, before any coffee flowed; the authors explain it by the powder already being wet and releasing these compounds into the gas phase.6

Washing out broken cells. Modelling studies describe two processes in each grain. The first is a rapid dissolution of coffee from the grain surfaces.1 A 2015 model, checked against hot-water experiments, traces this fast phase to cells damaged on the grain surface.2 Their contents lie open to the water and dissolve quickly.

Diffusion from intact cells. The second process is much slower. Coffee from the closed cells in the kernel of a grain has to diffuse through a winding network of pores to reach the surface.1,2 The larger the particle, the more of it is kernel. A 2024 espresso study states that solubles extract faster from finer particles, but also that a finer bed slows the flow, so a fine grind is not automatically the most efficient.12

Towards equilibrium. As coffee dissolves, the liquid around the particles grows stronger. In a 2021 immersion study, the brew reached an effective equilibrium in which coffee left the grounds and returned to them at equal rates.3 A weaker brew took less time to get there, because the concentration difference driving the transfer was larger.3 Extraction also changes the particles: a 2023 model describes porosity and permeability rising as coffee dissolves.14

123456
  1. Wetting. Water enters the particle, which swells.4
  2. Broken surface cells. Opened by grinding; their contents are washed out fast (solid arrows).1,2
  3. Intact cells inside. Solubles diffuse out slowly through cell walls and pores (dashed arrow).1,2
  4. Liquid at the surface. Richer in solubles than the bulk of the brew; transfer slows as the two approach equilibrium.1,3
  5. Fines. Cell fragments of about 25–50 µm; they extract fast, and some reach the cup.4,12,13
  6. Carried, not dissolved. Oil droplets and solid particles leave in suspension.5,6
One particle during extraction, as described in the sources cited. Point at a number or a line of the legend to highlight the part. Schematic, not to scale; no measured values.

What comes out first#

According to the textbook, solubility is the driving force of extraction, and it depends mainly on a compound’s polarity.4 Polar molecules mix readily with water. The textbook lists caffeine, sugars and organic acids among the highly soluble compounds extracted in the first seconds.4 Less soluble bitter and astringent compounds follow later.4 Measurements over time show the same trend, with some differences in detail:

  • Espresso, non-volatile compounds. A 2023 study collected each shot in 10 fractions. All compounds were most concentrated at the start.7 Trigonelline and dissolved solids fell fastest, then 5-CQA (a chlorogenic acid) and caffeine.7
  • Espresso, the whole shot. In a 2024 study, shots of 15 s or less already reached 17–18% extraction. That is more than 80% of the maximum extraction yield seen in the study.12
  • Espresso, aroma. The 2016 study followed 46 volatile compounds in the flow. The most polar ones peaked about 10 s after the pump started; the least polar about 20 s.6
  • Drip coffee. A 2020 study collected drip brews in fractions every 30 s. Most taste and flavour attributes decreased from one fraction to the next.9
  • Cold brew. In a 2017 study at room temperature, caffeine and 3-CQA levelled off after 6–7 h.10
1020304050605060708090100Beverage collected from 20 g coffee (g)Share of the amount in 60 g of beverage (%)Trigonelline7 TDS (all dissolved solids)7 5-CQA (a chlorogenic acid)7 Caffeine7
Data table
SeriesBeverage collected from 20 g coffee (g)Share of the amount in 60 g of beverage (%)Source
Trigonelline20727
Trigonelline40947
Trigonelline601007
TDS (all dissolved solids)20707
TDS (all dissolved solids)40937
TDS (all dissolved solids)601007
5-CQA (a chlorogenic acid)20657
5-CQA (a chlorogenic acid)40907
5-CQA (a chlorogenic acid)601007
Caffeine20627
Caffeine40897
Caffeine601007
How much of a 60 g espresso’s content was already in the first 20 g and 40 g of beverage, for four measures. Cup masses calculated by the authors from fitted extraction curves; the shares are our calculation.7 One coffee, 20 g dose, about 2 mL/s, 88 °C.

Temperature and solubility#

According to the textbook, most coffee compounds dissolve better in hotter water, though not all to the same degree.4 Less polar compounds need higher temperatures; many bitter compounds are among them.4 Hot water is also less viscous, so it enters the spaces between and inside particles more easily.4 Gases are the exception: they dissolve less well as the water gets hotter.4

In the 2016 espresso study, raising the water from 82 to 96 °C increased the extraction of volatiles. The effect was largest for the least polar compounds and in the second half of the shot.6 In a 2022 study brewing to equilibrium, yield rose from 19.55% at 4 °C to 22.28% for brews started at 92 °C.11 Brew temperature in practice has its own entry.

Under- and over-extraction#

Under- and over-extraction describe a change in what has dissolved, not only in how much. In the textbook account, the highly soluble compounds come out almost fully even from a coarse grind, while less soluble ones lag.4 According to the textbook, a low extraction yield therefore shifts the balance towards sweet and acidic compounds, and a long extraction towards bitter and astringent ones.4

Measurements support a shifting balance. In a 2017 espresso study, the ratio of trigonelline to caffeine depended on how much coffee had been extracted.8 In the 2016 study, higher temperature and pressure changed the aroma balance mainly in the last part of the shot.6 Timing alone does not explain flavour, though. In one drip study, early fractions were more bitter and more sour than late ones, and sweetness was negatively correlated with strength (see Contested).9 What the classic yield ranges mean for flavour is explained in the extraction yield and TDS entry.

Immersion and percolation#

The two main ways of brewing differ in their concentration gradient. In full immersion, all the water steeps with all the coffee, so the liquid grows stronger and extraction slows towards equilibrium.3 In percolation, fresh water keeps flowing past the particles, so the liquid around them stays weaker for longer.3 Pour-over and immersion compares the devices.

Models of extraction#

Several models describe the same physics at different scales. They agree on a fast surface phase and a slow inner phase, and on transfer that stops at an equilibrium.

Model What it describes Key result
Double-porosity bed model, 2015 Dissolution and transport in a packed bed; grains with damaged surface cells and intact kernels; parameters from measured bed properties Reproduced experimental extraction curves from a stirred suspension and a packed bed2
Well-mixed system, 2017 One grain in a stirred liquid Rapid surface dissolution plus much slower diffusion through the pore network; the liquid tends to a stable equilibrium1
Equilibrium desorption, 2021 Full immersion at equilibrium One average equilibrium constant, K = 0.717 ± 0.007, unchanged between 80 and 99 °C; maximum extractable share ≈0.33
Exponential decay, 2023 Concentration in the espresso flow against cumulative beverage mass, c = c₀·e−m/λ Centre-point λ: trigonelline 16.1 g, TDS 17.5 g, 5-CQA 20.8 g, caffeine 23.1 g7
Reduced model, 2017 Caffeine and trigonelline in the espresso flow Describes the measured data with physically meaningful parameters only8

Surface versus kernel. The 2015 and 2017 models split each grain into a surface layer of broken cells and an inner region of intact cells.1,2 The surface phase is fast because broken cells offer little resistance to mass transfer; the kernel is slow because intact cells offer much more.1 The 2017 paper notes that the cell radius and the internal surface area are difficult to measure, so the model needs fitting parameters.1

Equilibrium. In the desorption model, dissolved coffee moves between grounds and liquid until desorption and adsorption balance.3 Some species dissolve completely, others hardly at all; the model uses one average constant for all of them.3 The authors found it most surprising that this constant did not change between 80 and 99 °C.3 Temperature changed how fast the brew reached a steady TDS: the authors steeped for 60 min at 80 °C, 45 min at 85 °C, 30 min at 94 °C and 20 min at 99 °C.3

A decay constant per compound. In the 2023 espresso model, λ is the beverage mass over which a compound’s concentration in the flow falls by a factor of e (about 2.7). The authors describe it as a time constant.7 A small λ means a compound is washed out early. Trigonelline had the smallest value at the centre point, caffeine the largest.7

Beyond simple diffusion. A 2024 espresso study concludes that extraction is not only simple diffusion from particles into the liquid.12 The authors assume that the amount of fines acts mainly on the permeability of the bed, not on the extraction mechanism.12 A 2023 model adds a feedback: dissolving coffee raises porosity and permeability, so paths that carry more water open further.14 How this leads to uneven extraction is covered under particle size and fines.

What each time-resolved study measured#

Time-resolved studies measure different things. Before comparing them, check what was sampled, how and over what period.

Study Brew What was measured Sampling Main finding
Sánchez López et al., 2016 Espresso, 18 g coffee, ≈31.5 g beverage, 25 s pump time; 82–96 °C, 7–11 bar 46 volatile compounds stripped from the coffee flow (PTR-ToF-MS), as signal intensity One spectrum every 2 s Five families by time profile; polarity the main driving force6
Schmieder et al., 2023 Espresso, 20 g coffee, ≈58 g beverage, 1–3 mL/s, 80–98 °C Trigonelline, caffeine, 5-CQA and TDS 10 fractions of ≈6 g Concentration highest at the start; fastest decrease for trigonelline and TDS7
Kuhn et al., 2017 Espresso, varied particle size and tamping pressure Caffeine and trigonelline High time resolution, triplicates Finer particles gave more of both per mass collected; tamping pressure had no detectable effect8
Smrke et al., 2024 Espresso, 20 g coffee to 40 g beverage Extraction yield of shots of different length Shots of different duration 17–18% for shots of under 10 s and of 15 s, more than 80% of the maximum12
Batali et al., 2020 Drip brewer Sensory profile and monosaccharides of each fraction New carafe every 30 s Most attributes decreased with brew time; no fraction had perceptible monosaccharides9
Fuller & Rao, 2017 Cold brew, room temperature Caffeine and 3-CQA Up to 24 h Equilibrium by first-order kinetics after 6–7 h10

Flow signal versus cup amount. The aroma study recorded what was in the flow at each moment, not the running total in the cup.6 A falling signal means that less of a compound is coming out, not that the cup holds less. The 2023 study calculated cup amounts by integrating its fitted curves.7

The five aroma families. At 92 °C and 9 bar, coffee began to flow 5–6 s after the pump started, and the flow stopped 1 s after the pump at 25 s.6 Family A, the most polar compounds, peaked at 10 s and fell fastest, to below 40% of its maximum by the end.6 Its members – among them acetaldehyde, acetic acid and 2,3-butanedione – have water solubilities of 80–1,000 g/L.6 Family E, the least polar, peaked at 20 s and ended at 80–95% of its maximum.6 The authors write that an important part of the least polar compounds was still being extracted after 25 s.6 Across the families, water solubility mostly decreased and the octanol–water partition coefficient rose from A to E.6

0481216202428s after the pump startedCoffee starts to flow5–6 s6 Family A · most polar(e.g. acetaldehyde, acetic acid)peak 10 s6 Family Bplateau 10–16 s6 Family C · very volatile,first signalshoulder 4 s, before flow6 Family C · main signal10–16 s6 Family Dpeak 14–16 s6 Family E · least polar(e.g. pyrazines, guaiacol)peak 20 s6 Pump stops25 s6 Flow stops26 s6
Data table
Points after the pump startedSource
Coffee starts to flow5–6 s5–66
Family A · most polar (e.g. acetaldehyde, acetic acid)peak 10 s106
Family Bplateau 10–16 s10–166
Family C · very volatile, first signalshoulder 4 s, before flow46
Family C · main signal10–16 s10–166
Family Dpeak 14–16 s14–166
Family E · least polar (e.g. pyrazines, guaiacol)peak 20 s206
Pump stops25 s256
Flow stops26 s266
When five families of aroma compounds peaked in the flow of one espresso (18 g coffee, 92 °C, 9 bar). Times of maximum signal, not amounts in the cup.6

Fractions and cups#

The 2020 drip study shows why the order of extraction is not the whole story. Earlier fractions were systematically more bitter and more sour than later ones; later fractions were sweeter and more floral.9 The authors cite earlier work showing that TDS drops with each successive fraction.9 They conclude that perceived sweetness in drip coffee is negatively correlated with TDS.9 No fraction contained perceptible amounts of any monosaccharide; the authors attribute the sweetness to masking effects and/or sweet-associated aromas.9

A fraction is a small, separate brew. Its flavour depends on its strength as well as on its composition. The textbook statement that long extraction favours bitter compounds is about the balance in the whole cup.4 The two observations need not conflict: a late fraction is weak and can taste sweet on its own, while still adding slowly extracted compounds to the cup.4,9 No study opened for this entry measured bitter compounds such as phenylindanes or chlorogenic acid lactones fraction by fraction in filter coffee.

Temperature in detail#

Solubility. According to the textbook, water solubility generally rises with temperature but not linearly, and the ratio between compounds changes with temperature.4 Low-temperature brews therefore often lack some strength, which a much longer brew time can partly make up.4 The textbook adds a physical effect: hot water is less viscous and reaches the spaces inside particles more easily, which raises the amount of lipids in the brew.4

Gases. In line with Henry’s law, higher temperature drives dissolved gas molecules into the gas phase. According to the textbook, odorants are therefore released into the air during brewing, more so at higher temperatures.4 For volatiles, the textbook describes extraction from the particles as similar to that of non-volatiles, while their solubility in the liquid falls with temperature.4

Measured effects. In the 2016 espresso study, the extraction at 82 °C gave lower intensities for the most polar family; 92 and 96 °C did not differ for it.6 For the least polar families, differences between 82 and 96 °C appeared from 12–16 s and grew to the end of the shot.6 In the 2022 immersion study, brews held 24–36 h at 4 °C reached 19.55% yield, 12–14 h at 22 °C 20.74%, and brews started at 92 °C 22.28%.11 The equilibrium study found the same constant between 80 and 99 °C.3 The two studies cover different temperature ranges, so they do not directly conflict.

Reference values#

Parameter Value reported Context Source
Soluble share of roasted coffee ≈0.3 (about one third of the mass) Typical maximum extraction Liang et al.3
Equilibrium constant K 0.717 ± 0.007 Full immersion; unchanged between 80 and 99 °C Liang et al.3
Steeping time used to reach steady TDS 60 min at 80 °C; 45 min at 85 °C; 30 min at 94 °C; 20 min at 99 °C 1 L full immersion brews Liang et al.3
Equilibrium yield by brew temperature 19.55% (4 °C); 20.74% (22 °C); 22.28% (started at 92 °C) Full immersion to equilibrium Batali et al.11
Cell size of roasted coffee 25–50 µm Ground arabica; size of the fines peak Mo et al.13
Espresso yield, shots of 15 s or less 17–18%, more than 80% of the maximum 20 g to 40 g shots Smrke et al.12
Decay constant λ, espresso Trigonelline 16.1 g; TDS 17.5 g; 5-CQA 20.8 g; caffeine 23.1 g 20 g dose, ≈2 mL/s, ≈88 °C; centre point Schmieder et al.7
Cup content at 1:1, 1:2, 1:3 Trigonelline 75.9, 98.3, 104.9 mg; caffeine 129.0, 183.8, 207.1 mg; 5-CQA 86.5, 120.0, 133.1 mg; TDS 2.92, 3.88, 4.19 g Same centre point; calculated from fitted curves Schmieder et al.7
Share of the 1:3 content already in the cup at 1:1 ≈62% (caffeine) to ≈72% (trigonelline); TDS ≈70% Our calculation from the row above Schmieder et al.7
Yield at 1:1, 1:2, 1:3 ≈14.6%, ≈19.4%, ≈21.0% Our calculation: TDS mass ÷ 20 g dose Schmieder et al.7
Aroma peak, most polar family 10 s after pump start; below 40% of maximum at the end Espresso, 92 °C, 9 bar Sánchez López et al.6
Aroma peak, least polar family 20 s after pump start; 80–95% of maximum at the end Same Sánchez López et al.6
Cold brew, caffeine and 3-CQA Equilibrium after 6–7 h; no further significant 3-CQA after 400 min Room temperature, 21–25 °C Fuller & Rao10
Drip fractions Collected every 30 s Sensory and monosaccharide analysis Batali et al.9

Open questions#

  • Broken and intact cells. The models need the cell radius and internal surface area, which are difficult to measure, so they rely on fitting parameters.1 Whether roasting leaves the cell walls intact is itself debated; the porosity and cell structure entry sets out both views.
  • Bitter compounds over time in filter coffee. The order in which bitter compounds leave a filter bed rests on a textbook account.4 No study opened for this entry measured them fraction by fraction in filter brewing.
  • Temperature and equilibrium. One study found a temperature-independent equilibrium constant between 80 and 99 °C.3 Another found higher equilibrium yields from hot than from cold brewing.11 How the constant behaves between 22 and 80 °C was not measured in the sources opened here.
  • Caffeine timing. Caffeine counts as an early compound in the textbook account, yet came out more slowly than trigonelline in one espresso study.4,7 Whether this holds for other brew methods was not tested in the sources opened here.

Key studies#

Moroney et al., 2015 – A multiscale model of extraction. Builds a double-porosity model of a coffee bed in which each grain has damaged cells on its surface and intact cells in its kernel. Hot-water experiments in a stirred suspension and in a packed bed show a fast initial extraction followed by a slower one, and the model reproduces both.

Chemical Engineering Science 137, 216–234 · doi:10.1016/j.ces.2015.06.003

Liang et al., 2021 – Immersion as an equilibrium. Treats full immersion brewing as a balance between coffee leaving and re-entering the grounds, with a single average equilibrium constant. The constant did not change between 80 and 99 °C; hotter water reached a steady strength sooner.

Scientific Reports 11, 6904 · doi:10.1038/s41598-021-85787-1

Sánchez López et al., 2016 – Aroma over an espresso shot. Follows 46 volatile compounds in the coffee flow by on-line mass spectrometry at different temperatures and pressures. The compounds fall into five families by timing; the most polar come first, and hotter water mainly adds the least polar ones late in the shot.

International Journal of Mass Spectrometry 401, 22–30 · doi:10.1016/j.ijms.2016.02.015

Schmieder et al., 2023 – Compounds in espresso fractions. Measures trigonelline, caffeine, 5-CQA and TDS in 10 fractions of espresso shots across flow rates, grind settings and temperatures. Concentrations fall roughly exponentially with beverage mass; trigonelline and TDS fall fastest, and the beverage mass collected matters more than the other settings.

Foods 12(15), 2871 · doi:10.3390/foods12152871

Batali et al., 2020 – Drip fractions over brew time. Splits drip brews into 30 s fractions and tests them with a trained panel and mass spectrometry. Early fractions are more bitter and sour, late ones sweeter and more floral, and no fraction holds perceptible monosaccharides.

Journal of the Science of Food and Agriculture 100(7), 2953–2962 · doi:10.1002/jsfa.10323

Contested
  • Caffeine as an early compound: a coffee science textbook groups caffeine with sugars and organic acids as highly soluble compounds that are extracted efficiently in the first seconds of a brew.4 In a 2023 espresso study, the concentration of caffeine in the flow fell more slowly than that of trigonelline, total dissolved solids and 5-CQA.7 By the authors’ cup masses, the first 20 g of beverage held about 62% of the caffeine in a 60 g shot, against about 72% of the trigonelline (our calculation).7 Caffeine is early compared with the slowly extracted bitter compounds of the textbook account, but not compared with trigonelline.
  • Sourness and bitterness over a drip brew: a textbook links low extraction to a sweet and acidic cup and long extraction to bitter, harsh and astringent notes.4 In a 2020 study of drip coffee collected in 30 s fractions, the earlier fractions were systematically more bitter and more sour than the later ones, and the later ones sweeter and more floral.9 A fraction tastes of what is flowing out at that moment, at its own strength; a cup holds all fractions together. The two findings describe different things.
Share or cite this entry

Change log

  • – The 2017 extraction model is no longer said to use the cell radius and internal surface area as fitted values; it introduces fitting parameters because these are hard to measure. Reason: Verification against Moroney et al. 2017: the paper adds fitting parameters, it does not fit those two quantities directly.
  • – The espresso yield of 17–18% (Smrke et al. 2024) is now given for shots of under 10 s and of 15 s, not for 10–15 s. Reason: Verification against the paper's wording.

Sources

  1. AMoroney, K. M.; Lee, W. T.; O’Brien, S. B. G.; Suijver, F.; Marra, J. (2017). Coffee extraction kinetics in a well mixed system. Journal of Mathematics in Industry 7, 3. doi:10.1186/s13362-016-0024-6 · other copies
  2. AMoroney, K. M.; Lee, W. T.; O’Brien, S. B. G.; Suijver, F.; Marra, J. (2015). Modelling of coffee extraction during brewing using multiscale methods: An experimentally validated model. Chemical Engineering Science 137, 216–234. doi:10.1016/j.ces.2015.06.003 · other copies
  3. ALiang, J.; Chan, K. C.; Ristenpart, W. D. (2021). An equilibrium desorption model for the strength and extraction yield of full immersion brewed coffee. Scientific Reports 11, 6904. doi:10.1038/s41598-021-85787-1 · other copies
  4. BMestdagh, F.; Glabasnia, A.; Giuliano, P. (2017). The Brew – Extracting for Excellence. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 15, pp. 355–380. doi:10.1016/B978-0-12-803520-7.00015-3 · other copies
  5. BFolmer, B.; Blank, I.; Hofmann, T. (2017). Crema – Formation, Stabilization, and Sensation. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 17, pp. 399–417. doi:10.1016/B978-0-12-803520-7.00017-7 · other copies
  6. ASánchez López, J. A.; Wellinger, M.; Gloess, A. N.; Zimmermann, R.; Yeretzian, C. (2016). Extraction kinetics of coffee aroma compounds using a semi-automatic machine: On-line analysis by PTR-ToF-MS. International Journal of Mass Spectrometry 401, 22–30. doi:10.1016/j.ijms.2016.02.015 · other copies
  7. ASchmieder, B. K. L.; Pannusch, V. B.; Vannieuwenhuyse, L.; Briesen, H.; Minceva, M. (2023). Influence of Flow Rate, Particle Size, and Temperature on Espresso Extraction Kinetics. Foods 12(15), 2871. doi:10.3390/foods12152871 · other copies
  8. AKuhn, M.; Lang, S.; Bezold, F.; Minceva, M.; Briesen, H. (2017). Time-resolved extraction of caffeine and trigonelline from finely-ground espresso coffee with varying particle sizes and tamping pressures. Journal of Food Engineering 206, 37–47. doi:10.1016/j.jfoodeng.2017.03.002 · other copies
  9. ABatali, M. E.; Frost, S. C.; Lebrilla, C. B.; Ristenpart, W. D.; Guinard, J.-X. (2020). Sensory and monosaccharide analysis of drip brew coffee fractions versus brewing time. Journal of the Science of Food and Agriculture 100(7), 2953–2962. doi:10.1002/jsfa.10323 · other copies
  10. AFuller, M.; Rao, N. Z. (2017). The Effect of Time, Roasting Temperature, and Grind Size on Caffeine and Chlorogenic Acid Concentrations in Cold Brew Coffee. Scientific Reports 7, 17979. doi:10.1038/s41598-017-18247-4 · other copies
  11. ABatali, M. E.; Lim, L. X.; Liang, J.; Yeager, S. E.; Thompson, A. N.; Han, J.; Ristenpart, W. D.; Guinard, J.-X. (2022). Sensory Analysis of Full Immersion Coffee: Cold Brew Is More Floral, and Less Bitter, Sour, and Rubbery Than Hot Brew. Foods 11(16), 2440. doi:10.3390/foods11162440 · other copies
  12. ASmrke, S.; Eiermann, A.; Yeretzian, C. (2024). The role of fines in espresso extraction dynamics. Scientific Reports 14, 5612. doi:10.1038/s41598-024-55831-x · other copies
  13. AMo, C.; Johnston, R.; Navarini, L.; Suggi Liverani, F.; Ellero, M. (2023). Exploring the link between coffee matrix microstructure and flow properties using combined X-ray microtomography and smoothed particle hydrodynamics simulations. Scientific Reports 13, 16374. doi:10.1038/s41598-023-42380-y · other copies
  14. ALee, W. T.; Smith, A.; Arshad, A. (2023). Uneven extraction in coffee brewing. Physics of Fluids 35(5), 054110. doi:10.1063/5.0138998 · other copies

Tier A · authoritative · B · trade and expert · C · general. Supporting quotes for every source are in the data version of this entry. If a publisher’s site refuses access, try the doi link or “other copies”.

Suggest a change

On What extraction is. Tell us what you would change and why. A source that supports it helps most.