CoffeariumBeta
  1. Plant
  2. Farm
  3. Process
  4. Trade
  5. Roast
  6. Brew
  7. Cup
Post-harvest processing · Step 3

Mechanical demucilaging

Removing the mucilage from pulped coffee by mechanical agitation in a machine instead of by fermentation – fast, low in water, and at the heart of Colombia's ecological wet milling.

Across the chain: Sustainability · Equipment

Source-checkedby Coffearium verification pass· Last source-checked 8 Oct 2026· Prepared with AI assistance· 13 sources (11 tier A)· 2 contested· Updated 8 Oct 2026
Reading depth
Contents
  • A demucilager strips the mucilage off pulped coffee mechanically – a fast rotor puts the mass of coffee under shear while a little water carries the mucilage out through a perforated casing.1,5
  • Cenicafé’s DESLIM upflow demucilager removes more than 98% of the mucilage with less than 1 L of water per kg of dry parchment coffee; a conventional wet mill can use about 40 L/kg for the whole process.2,1
  • The parchment can go straight to the dryer; it should not be stored under water.2
  • Cup trials disagree – Colombian, Yunnan and Kenyan studies found it equal to or better than fermented coffee; a Q-grader panel in a 2021 study scored fermented coffee higher.1,6,11,7
  • The main risks are partial removal when the machine is overfed and mechanical damage when it is underfed.2
7583101246911
  1. Motor. 1
  2. Clean parchment out, to the dryer.1,2
  3. Perforated casing (screen). 1
  4. Rotor with agitators. 15 on the DESLIM 1000.1,4 Shear, not friction, strips the mucilage.1
  5. Coffee rises through the rotor. About 63 s inside the DESLIM 1000.1,4
  6. Mucilage and water out through the screen.1
  7. Washing water at the top, counter-current.1,3
  8. Demucilaging water in at mid-height.1,3
  9. Feed screw forces the coffee upwards.1
  10. Pulped coffee in, continuous feed.1,5
  11. Out to the screw conveyor, mixed with the pulp.1,3
A vertical upflow demucilager, cut open: pulped coffee enters at the bottom, is driven upwards past a rotor with agitators inside a perforated casing, and leaves clean at the top; water and loosened mucilage leave through the perforations.1,3,4,5 Schematic, not to scale.

What it does#

After pulping, the seeds are still coated in mucilage. In fermented washed coffee, microbes loosen it in a tank over many hours. A demucilager does the same job by force: it rubs the mucilage off within about a minute, so the coffee can be dried without fermenting.4,8 The FAO manual for Lao PDR puts it plainly: the mucilage is removed immediately after pulping and is not fermented.8

The idea is old. A mechanical demucilager built at the Hawaii Agricultural Experiment Station in the 1950s beat the mucilage off with rotating steel-rod blades while coffee and water travelled up a tube.5 Colombia’s research centre, Cenicafé, later built the DESLIM upflow demucilagers and combined them with water-free pulping in the Becolsub module.1,3

How the machine works#

The common design is a vertical cylinder with a rotor inside.

  • Feed – a hopper delivers pulped coffee continuously to the machine.1 In the DESLIM, a short screw at the bottom forces the coffee upwards.1 In the Hawaiian prototype, coffee and water were funnelled in together through a feeder tube below the beater.5
  • Upward flow – the coffee is pushed upwards through the cylinder; Cenicafé calls the DESLIM an upward vertical grain flow demucilager.1 Water enters at mid-height for demucilaging and at the top for washing, where grain and water flow in counter-current.1
  • Rotor and agitators – the rotor is the part that creates the shear on the mass of coffee needed to separate the mucilage from the seeds.1 The DESLIM 1000 carries 15 agitators.4 Cenicafé gives a rotor speed of 870 rpm for its machine;1 the Hawaiian beaters ran at 1,000 and 1,200 rpm.5
  • Screen – mucilage and fine material leave through the perforations of the casing.1
  • Outlets – clean parchment leaves at the top. In the Becolsub module, a screw conveyor mixes the concentrated mucilage with the pulp.1,3

Water, energy and capacity#

Water use is the main reason for the method. A conventional Colombian wet mill can use about 40 litres per kilogram of dry parchment coffee.1 Washing fermented coffee in the tank brings this down to about 4.2 L/kg, while washing and grading in open channels without recirculation uses about 39 L/kg.1 With the DESLIM, water use fell below 1.0 L/kg and could be brought down to 0.6 L/kg.1 For the DESLIM in the Becolsub module, Cenicafé specifies 0.7–1.0 L/kg.3 The FAO manual for Lao PDR gives 0.5 L per kg of fresh cherry for pulper/demucilager units.8

051015202530354045L/kg dry parchmentMechanical demucilaging(DESLIM), best casedown to 0.6 L/kg1 DESLIM in the Becolsubmodule(specification)0.7–1.0 L/kg3 Fermentation + washing inthe tankabout 4.2 L/kg1 Fermentation + washing andgrading in open channelsabout 39 L/kg1 Conventional wet mill —whole processabout 40 L/kg1
Data table
L/kg dry parchmentSource
Mechanical demucilaging (DESLIM), best casedown to 0.6 L/kg1
DESLIM in the Becolsub module (specification)0.7–1.0 L/kg3
Fermentation + washing in the tankabout 4.2 L/kg1
Fermentation + washing and grading in open channelsabout 39 L/kg1
Conventional wet mill — whole processabout 40 L/kg1
Water use in litres per kilogram of dry parchment coffee for fermentation with washing and for mechanical demucilaging, as reported by Cenicafé.1,3

Cenicafé built three DESLIM sizes: 600, 1,000 and 3,000 kg of cherry per hour, with power ratings of 0.89, 1.34 and 4.92 kW.1 The water flow is set to the coffee flow: for a 300 kg/h module it is 0.70–1.00 L/min, for a 2,500 kg/h module 5.83–8.33 L/min.3 The Hawaiian prototype handled the coffee from a pulper taking 3,000 lb (about 1,360 kg) of cherry per hour, with a 2 hp motor.5

After the machine#

Mechanically demucilaged coffee can be dried immediately, with no further treatment.2 If it must wait overnight – up to 14 hours – Cenicafé advises against holding it under water.2 The FAO manual adds that the units remove the need for fermentation tanks and washing systems and lower the risk of over-fermentation.8

The same machines appear in other processes. The FAO manual calls pulp-and-demucilage processing “semi-washed”.8 The term is not used consistently. The Codex code of practice uses “semi-washed” for the opposite case: green coffee from pulped beans dried within their mucilaginous parchment.13 And an importer lists “semi-washed” as another name for wet-hulled coffee (giling basah), a different process in which the parchment is removed before the bean is dry.12 This entry uses “semi-washed” only in the FAO sense. Where only part of the mucilage is taken off, the coffee moves towards the honey and pulped-natural styles, which have their own entries.

Other ways to remove mucilage without fermenting. Enzymes are one: in a 2025 Colombian study, mucilage was removed with a commercial pectin lyase.9 Alkali is another: in Hawaii, a 6% solution broke the mucilage down in about one minute.5 The chemistry of both is covered in the mucilage entry.

The mechanics: shear, residence time and water#

Shear. The mucilage is a gel attached to the parchment. The rotor puts the moving mass of coffee under shear – Cenicafé speaks of “cutting stresses” – and this is what separates the mucilage from the seeds.1 Cenicafé attributes the effect mainly to these shear forces, from the relative movement of fluid and grains, and not simply to friction or collisions between grains.1 It observed high mucilage detachment in the first moments of the process.1 In the Hawaiian machine, the mucilage was beaten off by the blades while the suspension travelled up the tube.5

Residence time. How long a seed stays in the rotor zone sets how much mucilage comes off and how much damage is done. Seeds spent 63 s in the DESLIM 1000.4 A related Cenicafé washer for already-fermented coffee (ECOLAV) needed only 16.9 s and 7 agitators, because the mucilage had already been loosened; its specific power was 10.9% of that of the demucilager.4 The authors read this as efficient use of mechanical energy once fermentation (or enzymes) has degraded the mucilage.4

Coffee flow. Residence time depends on the feed rate. When the flow of coffee is too high, the seeds pass through faster than recommended and come out partly demucilaged.2 When it is too low, residence time is very long and the risk of mechanical damage rises.2

Water-to-coffee ratio. Water carries the mucilage through the screen. Cenicafé ties water flow to coffee flow (see Reference values).3 Too much water is also a problem: with higher specific water flows, the pulp no longer retains the mucilage and contamination rises because the pulp is washed.3

Sensory comparisons with fermented coffee#

The evidence points in different directions.

  • Cenicafé, 1995–1996. Across 77 paired cup comparisons, 127 attribute-by-attribute results were reported: 56 favoured mechanical removal, 40 showed no difference and 31 favoured fermentation.1 Cenicafé concluded that mechanically demucilaged coffee is equal or superior to naturally fermented coffee in the cup.1
  • Cenicafé, 2009. All samples dried immediately after demucilaging scored above 5 on the scale used, against 95% of samples held dry overnight and 75% of samples held under water; no ferment or dirty defects were found.2
  • Yunnan, 2019. In a controlled Catimor trial, the demucilaged process gave a beverage comparable to the depulped, fermented one.6 The demucilaged beans were still held under clean water in a tank for 12 or 72 hours before washing.6 The fermentation waters of fermented coffee held about ten times more of the measured compounds than those of demucilaged coffee.6
  • 2021, three Q graders. Pulped coffee kept under water for 36 hours scored significantly higher in flavour, aroma, acidity, body and uniformity than mechanically processed coffee, with a noted fruity aroma.7 The fermented roasted beans held more ethanol, isoamyl alcohol, 3-methylbutanal, benzaldehyde, acetaldehyde and ethyl acetate; pyrazines, furans and other roast volatiles did not differ.7
  • Kenya. On a university farm in Nyeri, SL 28 and Ruiru 11 coffee was either pulped in an eco-pulper and passed through a demucilager, or pulped and fermented before washing, then dried. Final cup scores were close for all methods; the demucilaged coffee scored slightly higher for SL 28, and for Ruiru 11 the differences were not significant. The demucilaged coffee had more pyrazines, the fermented wet-pulper coffee more furans.11
  • Practice guidance. The FAO manual for Lao PDR states that demucilaging adds body and character to the liquor.8 The Hawaiian bulletin cites an earlier report that alkali-demucilaged coffee cupped better than fermented coffee.5

The disagreement probably reflects what each study compared. Fermentation adds fermentation volatiles;7 whether a panel rewards them depends on the coffee and the cup profile it is looking for. Another study used mechanical removal as a no-fermentation control and found significant differences in quality between treatments; its abstract does not say how the control ranked.10

Reference values#

The table brings together values from the sources. Machine settings are specific to the machine tested.

Parameter Value reported Source
Machine principle Vertical, upward flow of coffee, rotor with agitators, counter-current washing in the upper section Cenicafé1,3
Rotor speed 870 rpm (Cenicafé); 1,000 and 1,200 rpm (Hawaii, 1957) Cenicafé1, Fukunaga5
Agitators 15 (DESLIM 1000) Oliveros-Tascón et al.4
Residence time 63 s (DESLIM 1000); 16.9 s for washing fermented coffee (ECOLAV) Oliveros-Tascón et al.4
Mucilage removed More than 98% Pabón et al.2
Capacity 600, 1,000 and 3,000 kg cherry/h (DESLIM models) Cenicafé1
Motor power 0.89, 1.34 and 4.92 kW (same models) Cenicafé1
Water flow per module 0.70–1.00 L/min at 300 kg/h; 1.40–2.00 at 600 kg/h; 2.80–4.00 at 1,200 kg/h; 5.83–8.33 at 2,500 kg/h Sanz et al.3
Water, DESLIM Below 1.0 L/kg dry parchment; down to 0.6 L/kg Cenicafé1
Water, DESLIM in the Becolsub module 0.7–1.0 L/kg dry parchment Sanz et al.3
Water, pulper/demucilager units (FAO, Lao PDR manual) 0.5 L/kg fresh cherry FAO8
Water, fermentation + tank washing About 4.2 L/kg dry parchment Cenicafé1
Water, fermentation + channel washing About 39 L/kg dry parchment Cenicafé1
Water, conventional wet mill About 40 L/kg dry parchment Cenicafé1
Holding before drying Dry immediately; overnight (up to 14 h) only out of water Pabón et al.2
Mucilage retained in pulp (screw mixing) More than 50% of the “mieles” Sanz et al.3
Pollution controlled (Becolsub) More than 90% of the potential contamination Cenicafé1

Risks and defects#

  • Partial demucilaging. Overfeeding shortens residence time; the coffee comes out partly demucilaged, with stained parchment and a ferment defect in the cup.2
  • Mechanical damage. Underfeeding lengthens residence time and raises the risk of mechanical damage.2 In Cenicafé’s seed trials with the three Becolsub modules, 98.9% of demucilaged seeds remained viable and double-root damage stayed below 1%.1 In the Hawaiian trials no physical damage to the green bean was noticeable on close examination.5 Cenicafé’s washer for fermented coffee caused 0.13–0.58% mechanical damage.4
  • Holding wet coffee. Demucilaged coffee kept under water overnight scored worse than coffee dried at once.2
  • Heating. A fast rotor turns mechanical energy into heat. None of the sources opened for this entry measured the temperature rise of the coffee in a demucilager, so no value is given here.
  • Wastewater. The mucilage does not disappear – it leaves as a concentrated stream. In the Becolsub module it is mixed into the pulp by a screw conveyor, which holds back more than half of it; with too much water, the pulp is washed and pollution rises.3 Cenicafé reports that washing fermented coffee generates 30 g of chemical oxygen demand per kg of cherry.1

Open questions#

  • Cup parity. Whether mechanically demucilaged coffee matches fermented coffee depends on the study.1,6,7 No study opened here compared several machines and origins with the same panel.
  • Heat and the living seed. The temperature reached in the rotor zone, and whether it affects the seed, is not reported in the sources opened for this entry.
  • Residual mucilage and flavour. The link between the small residue of mucilage left (under 2%)2 and cup flavour has not been measured in these sources.

Key studies#

Roa et al. (Cenicafé) – Ecological processing at farm level. Describes the Becolsub system: dry pulping, an upflow demucilager and mixing of mucilage into the pulp. Reports water use below 1 L per kg of dry parchment with the DESLIM, and cup comparisons whose attribute results mostly favoured mechanical removal.

Cenicafé technical report · PDF

Pabón, Sanz & Oliveros, 2009 – Handling demucilaged coffee. Tested what to do with coffee after the DESLIM. Coffee dried immediately cupped best; holding it overnight under water lowered quality, and feeding the machine too fast or too slowly caused incomplete removal or damage.

Avances Técnicos Cenicafé 388 · PDF

Zhang et al., 2019 – Processing parameters in Yunnan. Compared depulped and fermented coffee with coffee that was mechanically demucilaged before tank holding. Demucilaged coffee had far less substrate in the tank water but a comparable cup.

Frontiers in Microbiology 10, 2621 · doi:10.3389/fmicb.2019.02621

Elhalis et al., 2021 – Wet fermentation and flavour. Compared 36-hour underwater fermentation with mechanical processing and no fermentation. Fermented beans carried more fermentation volatiles and were scored higher by a Q-grader panel.

European Food Research and Technology 247, 485–498 · doi:10.1007/s00217-020-03641-6

Oliveros-Tascón et al., 2011 – ECOLAV washer. Built an upflow washer for fermented coffee from the DESLIM design. Comparing the two shows how much residence time and energy the mechanical removal of intact mucilage needs.

Revista de Ingeniería 33, 61–67 · PDF

Contested
  • Cup quality against fermented coffee: in Cenicafé’s cup comparisons, 127 attribute results favoured mechanical removal more often than fermentation1, and trials in Yunnan and Kenya found the two comparable6,11; a 2021 study with a panel of three Q graders rated fermented coffee significantly higher in flavour, aroma, acidity, body and uniformity.7 The trials differ in origin, machine and in whether the demucilaged coffee was then held in water.
  • Water use is reported on different bases: Cenicafé gives litres per kg of dry parchment coffee (0.7–1.0 L/kg for the DESLIM in the Becolsub module)3, FAO gives 0.5 L per kg of fresh cherry for pulper/demucilager units in Lao PDR.8 The figures cannot be compared directly.
Share or cite this entry

Sources

  1. ARoa, G.; Oliveros, C. E.; Álvarez, J.; Sanz, J. R.; Ramírez, C. A.; Dávila, M. T.; Álvarez, J. R.; Zambrano, D. A.; Puerta, G. I.; Rodríguez, N. Ecological Processing of Coffee at the Farm Level: An Ecological and Profitable Case. Cenicafé (Colombian Coffee Research Centre).
  2. APabón Usaquén, J. P.; Sanz Uribe, J. R.; Oliveros Tascón, C. E. (2009). Manejo del café desmucilaginado mecánicamente (Avances Técnicos 388). Cenicafé.
  3. ASanz Uribe, J. R.; Oliveros Tascón, C. E.; Ramírez Gómez, C. A.; López Posada, U.; Velásquez Henao, J. (2011). Controle los flujos de café y agua en el módulo Becolsub (Avances Técnicos 405). Cenicafé.
  4. AOliveros-Tascón, C. E.; Sanz-Uribe, J. R.; Montoya-Restrepo, E. C.; Ramírez-Gómez, C. A. (2011). Equipo para el lavado ecológico del café con mucílago degradado con fermentación natural. Revista de Ingeniería 33, 61–67.
  5. AFukunaga, E. T. (1957). A New Mechanical Coffee Demucilaging Machine. Hawaii Agricultural Experiment Station, University of Hawaii, Bulletin 115.
  6. AZhang, S. J.; De Bruyn, F.; Pothakos, V.; Contreras, G. F.; Cai, Z.; Moccand, C.; Weckx, S.; De Vuyst, L. (2019). Influence of various processing parameters on the microbial community dynamics, metabolomic profiles, and cup quality during wet coffee processing. Frontiers in Microbiology 10, 2621. doi:10.3389/fmicb.2019.02621 · other copies
  7. AElhalis, H.; Cox, J.; Frank, D.; Zhao, J. (2021). The role of wet fermentation in enhancing coffee flavor, aroma and sensory quality. European Food Research and Technology 247(2), 485–498. doi:10.1007/s00217-020-03641-6 · other copies
  8. AWinston, E.; Op de Laak, J.; Marsh, T.; Lempke, H.; Chapman, K. (2005). Arabica coffee manual for Lao-PDR. FAO Regional Office for Asia and the Pacific, Bangkok.
  9. APeñuela-Martínez, A. E.; Osorio-Giraldo, C. V.; Buitrago-Zuluaga, C.; Medina-Rivera, R. D. (2025). Development of Fermentation Strategies for Quality Mild Coffee Production (Coffea arabica L.) Based on Oxygen Availability and Processing Time. Foods 14(17), 3001. doi:10.3390/foods14173001 · other copies
  10. APeñuela-Martínez, A. E.; Zapata-Zapata, A. D.; Durango-Restrepo, D. L. (2018). Performance of different fermentation methods and the effect on coffee quality (Coffea arabica L.). Coffee Science 13(4). doi:10.25186/cs.v13i4.1486 · other copies
  11. BKipkorir, R.; Muhoho, S.; Muliro, P.; Mugendi, B.; Frohme, M.; Broedel, O. (2015). Effects of Coffee Processing Technologies on Aroma Profiles and Sensory Quality of Ruiru 11 and SL 28 Kenyan Coffee Varieties. Asian Journal of Agriculture and Food Sciences 3(2), 178–188.
  12. BWet-Hulled (aka Semi-Washed, Sumatra Process, Giling Basah). Cafe Imports.
  13. ACodex Alimentarius Commission (FAO/WHO) (2009). Code of Practice for the Prevention and Reduction of Ochratoxin A Contamination in Coffee (CAC/RCP 69-2009). Codex Alimentarius.

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 Mechanical demucilaging. Tell us what you would change and why. A source that supports it helps most.