Roaster energy use
How much heat and electricity coffee roasting takes, where the energy goes – beans, water, hot exhaust and afterburner – and how recirculation, heat recovery and the choice of fuel change energy use and carbon dioxide.
Across the chain: Sustainability · Equipment
In brief
- Studies consulted put the heat or fuel for roasting at about 0.7–3.6 MJ per kg of coffee, mostly modelled; boundary and mass basis explain much of the spread.1,2,3,4
- In a 2023 model of a ground-coffee plant, the roaster took 66–67% of the heat and the afterburner the remaining 33–34%.1
- In the same model, a single-pass roaster without hot-air recirculation raised heat use by 146–167%.1
- A European Commission best-practice page gives savings of about 10–20% for green coffee preheating with exhaust heat in batch roasters, without showing data.9
- Roasting and grinding together contributed a median 8% of the cradle-to-grave carbon footprint of coffee in a 2024 review.7
Contents
How much energy a roast takes#
Roasting heats green coffee to high temperatures: between about 190 and 245 °C in one plant study.2 In the plants studied, the heat came from burning natural gas.2,3,4 The European Commission describes roasting as having a high demand for thermal energy.9
Studies give the energy as megajoules (MJ) per kilogram of coffee. Those consulted for this entry report about 0.7 to 3.6 MJ of heat or fuel per kg, mostly from models.1,2,3,4 A model of a UK ground-coffee plant gave the low end.1 A model of a small plant in Athens, Greece, gave the high end.2
Part of the spread comes from what is counted. Some studies count per kg of green coffee, others per kg of roasted or ground coffee.1,2,3 Some include the afterburner; one adds the heat for cooling.1,2,3 Roasted coffee weighs less than the green coffee it came from. In one plant, 300 kg of green coffee gave about 255–260 kg of roasted coffee.3 The same energy therefore looks larger per kg roasted than per kg green. The roast loss entry explains the mass change.
Data table
| kJ per kg (= MJ per tonne) | Source | |
|---|---|---|
| Ground-coffee plant, roaster plus afterburner, light to dark, model (per kg ground coffee) | 702.7–1,121.4 | 1 |
| Batch drum plant, natural gas, assumed input (per kg green) | 1,500 | 4 |
| Batch roaster with afterburner, stated without a source (per kg roasted) | ≈1,670 (1.67 GJ/t) | 3 |
| Small plant, roasting plus cooling, afterburner air recirculated, model (per kg green) | 2,872.4 | 2 |
| Small plant, roasting plus cooling, single-pass base case, model (per kg green) | 3,647.7 | 2 |
Where the heat goes#
A roaster’s heat ends up in three places, according to a modelling chapter: the beans, the gas leaving the stack and the chaff.5 In the beans, heat warms the seed and evaporates its water. The water in roasting entry covers that part.
The stack. The gas leaving the roaster is still hot. According to the modelling chapter, most of the heat input can leave through the stack.5 After an afterburner, the exhaust is at about 250–450 °C.3
The afterburner. An afterburner burns the smoke and odour in the exhaust (see Roaster emissions). It needs fuel of its own. In a 2023 model, it used a third of the plant’s heat.1 In a 2025 model, it burned about half of the natural gas (our calculation).2
Small losses. The modelling chapter calls heat lost through the roaster’s walls small for well-designed roasters.5 It finds the heat released by roasting reactions negligibly small in its balance.5
Electricity. Fans, drives and grinders run on electricity. In the 2023 model, a ground-coffee plant used far less electricity than heat; grinding and the cyclone were the main users.1
- Fuel and electricity. Gas for the burners; electricity for fans and drives.1,2,4
- Burner. Heats the process gas.5
- Drum. Heat goes into the beans: warming them and evaporating water; chaff comes off.5
- Cyclone. Separates the chaff from the gas.5
- Recirculation. Part of the hot gas returns to the burner; a single-pass roaster has no loop and needs more heat.1,5
- Afterburner. Has its own burner: a third of the heat in one model, about half of the natural gas in another.1,2
- Hot exhaust. Leaves at about 250–450 °C.3
- Stack. Hot gas through the stack is the largest loss.5
- Exhaust heat exchanger. Recovers heat from the exhaust duct.9
- Green coffee preheater. Exhaust heat warms the next batch to 80–100 °C (batch roasters only).9
- Hot afterburner air. Returned to the roaster’s burner.2
Ways to use less energy#
Recirculation. A recirculating roaster sends part of its hot gas back to the burner. The 2023 model assumed such a roaster, as common in practice.1 Without recirculation, in a single-pass roaster, the plant’s heat use rose by 146–167%.1 A single-pass roaster is one in which the gas passes through only once.5
Heat recovery. Heat can be taken from the hot exhaust. The European Commission’s best-practice page puts savings at less than 10% for exhaust heat exchangers.9 It gives about 30% for roasters that recirculate roast gas.9 The page shows no data or source for these figures.9
Green coffee preheating. Exhaust heat from the previous batch can warm the next batch of green coffee to 80–100 °C.9 This works only with batch roasters and needs no additional heat.9 The same page gives savings of about 10–20%, again without data.9
Afterburner choice. A catalytic afterburner works at a lower temperature than a thermal one.11 The modelling chapter finds much higher total energy use with thermal afterburning.5 It also proposes largely bypassing the afterburner when emissions are low enough.5
Roast degree. In the 2023 model, energy demand rose with roast degree, because darker roasts reach higher temperatures.1
Data table
| % less energy | Source | |
|---|---|---|
| Heat exchangers in the exhaust duct (EC best-practice page) | less than 10% | 9 |
| Green coffee preheating (EC best-practice page) | about 10–20% | 9 |
| Afterburner hot air recirculated, one model (our calculation) | ≈21% total energy, ≈31% natural gas | 2 |
| Roast gas recirculation machines (EC best-practice page) | ≈30% | 9 |
Fuels and carbon dioxide#
Fuels. The United States Environmental Protection Agency (EPA) describes roasters as typically fired with natural gas.11 Three plant studies and one permit consulted for this entry also name natural gas.2,3,6,12 Burning fuel produces carbon monoxide and carbon dioxide.11
Carbon dioxide per unit of fuel. The Intergovernmental Panel on Climate Change (IPCC) gives default factors for national inventories.10 Liquefied petroleum gas (LPG) emits more carbon dioxide per unit of energy than natural gas.10 For electricity, emissions depend on how it is generated; no study consulted for this entry compared electric and gas-fired roasters.
Chaff as fuel. Silverskin, the chaff that comes off the beans, can be burnt for heat. In the 2023 model, doing so lowered all environmental impacts and costs.1
Share of the footprint. Roasting is a small part of coffee’s carbon footprint. In a 2024 review, roasting and grinding together contributed a median 8% in cradle-to-grave studies.7 Inside the roastery, roasting matters more: it was a main contributor to the plant’s impacts in the 2023 model.1 The carbon footprint of coffee entry covers the whole chain; the roaster emissions entry covers smoke and its cleaning.
Reference values#
Values reported in the sources, with their basis. Energy in kJ per kg equals MJ per tonne. Conversions marked ≈ are ours.
| Parameter | Value reported | Basis and conditions | Source |
|---|---|---|---|
| Heat, ground-coffee plant | 702.7–1,121.4 MJ per tonne (≈0.70–1.12 MJ/kg), across the modelled roast degrees (light to dark) | Per tonne ground coffee; model, batch roaster with hot-air recirculation, roaster plus afterburner | Gosalvitr et al.1 |
| Afterburner share of heat | 33–34%; roaster 66–67% | Same model | Gosalvitr et al.1 |
| Electricity, ground-coffee plant | 77.8–95.8 MJ per tonne | Same model; grinding and cyclone main consumers | Gosalvitr et al.1 |
| Single-pass instead of recirculating roaster | Heat use +146–167%; no significant change in electricity | Same model | Gosalvitr et al.1 |
| Energy to roast (QSR), base | 3,534.8 kJ/kg | Per kg green; model of a 1,800 kg/day plant, Athens; single-pass roasting chamber | Kekes et al.2 |
| Total energy (QTOT), base / afterburner air recirculated | 3,647.7 / 2,872.4 kJ/kg (≈21% less) | Per kg green; same model; QTOT = roasting plus enthalpy added for cooling (113.0 kJ/kg in the base case) | Kekes et al.2 |
| Natural gas, total / afterburner, base | 37.21 / 19.23 Nm³ (afterburner ≈52%) | Same model; time basis not stated | Kekes et al.2 |
| Natural gas, total / afterburner, recirculated | 25.53 / 13.19 Nm³ (total ≈31% less than base) | Same model | Kekes et al.2 |
| Fuel, batch roaster with afterburner | ≈1.67 GJ per tonne roasted coffee (≈1.67 MJ/kg, ≈1,670 kJ/kg) | Stated without a citation in the introduction of an exergy study; not one of its results | Vučković et al.3 |
| Afterburner fuel | Up to 400% more than the roaster itself | Same introduction, without a citation | Vučković et al.3 |
| Overall exergy efficiency | 33.26% | One 300 kg batch roaster with afterburner; partly on-site data | Vučković et al.3 |
| Heat recovery potential | 1.1 MW; about EUR 60,000 a year (2016) | Roasters up to 4 t of green coffee per hour | Vučković et al.3 |
| Natural gas per kg green | 1.5 MJ/kg (1,500 kJ/kg; 375 MJ per 250 kg cycle) | Assumed input, batch drum with partial hot gas recycling, 500 kg/h | Pantaleo et al.4 |
| Stack gas temperature | 300–350 °C | Same case study | Pantaleo et al.4 |
| Afterburner exhaust temperature | 250–450 °C | Exergy study (abstract) | Vučković et al.3 |
| Whole roast-and-ground plant | 0.14 kWh electricity; 0.07 Nm³ natural gas; 1.23 kg green coffee | Per kg roasted and ground coffee; one plant in Spain, all steps from cleaning to packing | Humbert et al.6 |
| Savings, exhaust heat exchangers | Less than 10% | Batch roasting; best-practice page, no data or citation shown | JRC9 |
| Savings, roast gas recirculation | ≈30% | Same; no data or citation shown | JRC9 |
| Savings, green coffee preheating | About 10–20%; beans preheated to 80–100 °C | Batch roasters only; depends on roasting time, roast degree and exhaust treatment; no data or citation shown | JRC9 |
| Burner ratings, one permit | Roaster 0.40 MMBtu/h (≈117 kW); catalytic afterburner 0.30 MMBtu/h (≈88 kW) | 616 lb/h (≈279 kg/h) roaster; ratings, not fuel use | SDAPCD12 |
| CO₂ factor, natural gas | 56,100 kg/TJ (54,300–58,300) (≈56.1 g/MJ) | Default, 100% oxidation | IPCC10 |
| CO₂ factor, LPG | 63,100 kg/TJ (61,600–65,600) (≈63.1 g/MJ) | Same | IPCC10 |
| CO₂, batch roaster, uncontrolled / with thermal oxidiser | 180 / 530 lb per ton green coffee (≈90 / 265 kg/t); rating D | US emission factors, 1995 | EPA AP-4211 |
| Roasting and grinding, share of footprint | Median 8% | Cradle to grave, review of studies | Chéron-Bessou et al.7 |
| Roasting, share of roasting-to-cup stage | 6% of 3.05 kg CO₂e per kg green coffee; grinding 10%; consumption 70% | 2008 German pilot study, as cited | Nab & Maslin8 |
The energy balance of a roaster#
The balance. A modelling chapter computes the energy to roast a kilogram of green coffee from overall enthalpy balances.5 The heat input equals the heat taken up by the beans, the heat in the gas leaving the stack and the heat in the chaff.5 According to the chapter, the heat released by roasting reactions is negligibly small in that balance, and wall losses are small for well-designed roasters.5 The endothermic and exothermic phases entry covers the reaction heat inside the bean.
Stack gas dominates. According to the chapter, most of the heat input can leave through the stack.5 An exergy analysis points the same way. Exergy is the part of energy that can still do work; it falls as heat flows to lower temperatures. In one 300 kg gas-fired batch roaster with an afterburner, the overall exergy efficiency was 33.26%.3 Exhaust gases carried 37.62% of the exergy destroyed and lost in the system.3
Basis. The 2016 study states about 1.67 GJ of fuel energy per tonne of roasted coffee for a commercial batch roaster with an afterburner, without a citation for the figure.3 The 2025 model reports 3,647.7 kJ per kg of green coffee in its base case, of which 3,534.8 kJ/kg for roasting.2 The 2023 model reports 702.7–1,121.4 MJ of heat per tonne of ground coffee.1 The total adds about 113 kJ/kg of enthalpy for cooling to the roasting energy.2 A 2017 case study assumed 1.5 MJ of natural gas per kg of green coffee.4 The 2016 study’s own exergy analysis rests partly on on-site measurements, but its 1.67 GJ figure is not one of its results.3
A plant as a whole. One life cycle study collected primary data from a roast-and-ground coffee plant in Spain.6 Per kg of roasted and ground coffee, the plant used 0.14 kWh of electricity and 0.07 Nm³ (normal cubic metres) of natural gas, and 1.23 kg of green coffee.6 These figures cover every step from green coffee cleaning to packing, not roasting alone.6
Afterburner energy#
A separate thermal oxidiser runs hotter than the roaster. The EPA gives its desired operating temperature as 93–260 °C above the maximum temperature of most roasters.11 Some plants use the burners that heat the roaster as the oxidiser instead.11
The share of fuel the afterburner takes differs between studies (see Contested). In the 2023 model, the afterburner used 33–34% of the plant’s heat.1 In the 2025 base case, with a single-pass roasting chamber, it burned 19.23 of 37.21 Nm³ of natural gas, about 52% (our calculation).2 The 2016 study states in its introduction, without a citation, that afterburners can consume up to 400% more fuel than the roaster.3 A 2017 case study calls the afterburner quite energy intensive.4
Permits list burner ratings, not fuel use. In one San Diego permit, a roaster of 616 lb/h (≈279 kg/h) had a 0.40 MMBtu/h (≈117 kW) burner, its catalytic afterburner a 0.30 MMBtu/h (≈88 kW) burner.12
Thermal or catalytic. A catalytic oxidiser reaches the same control efficiency at a lower operating temperature.11 The modelling chapter predicts much higher total energy use with thermal than with catalytic afterburning.5 No measured comparison of the two on the same roaster was found in the sources consulted.
Recirculation, heat recovery and preheating#
Recirculation. The 2023 model assumes a batch roaster with hot-air recirculation, as commonly used in practice to raise energy efficiency.1 Switching to a single-pass system raised heat consumption for ground coffee by 146–167%, with no significant change in electricity.1 In the 2025 model, recirculating hot air from the afterburner cut total natural gas from 37.21 to 25.53 Nm³.2 That is about 31% less natural gas (our calculation).2 Total energy, which includes cooling, fell from 3,647.7 to 2,872.4 kJ/kg, about 21% less (our calculation).2 The authors report a 25% cut in greenhouse gas emissions and an 18% cut in fossil fuel use, both life-cycle indicators, for that scenario.2 Recirculation and its history are covered in Industrial roasting history and Airflow.
Exhaust heat exchangers. The European Commission page gives savings from less than 10% for exhaust heat exchangers to about 30% for roast gas recirculation machines.9 It shows no data and no citation for these figures; its literature consists mostly of roaster makers’ documents and personal communications from roasting companies.9
Green coffee preheating. In this method, exhaust gases from the previous batch warm the next batch to 80–100 °C before it enters the roaster.9 The preheater sits upstream of a batch roaster and needs no additional heat.9 The page gives savings of about 10–20%, alone or combined with other measures, depending on roasting time, roast degree and exhaust air treatment; it shows no data for this range.9 Existing roasters can be retrofitted, though the page calls this more complex than installing a preheater on a new roaster.9
Waste heat to power. In a batch plant, hot flue gas leaves in pulses. A 2017 case study found it discharged for about 15 seconds per cycle, at a stack temperature of 300–350 °C.4 The authors assessed organic Rankine cycle and micro gas turbine units to turn this heat into electricity.4 For roasters of up to 4 t/h, the 2016 study put the heat recovery potential at 1.1 MW and possible savings at about EUR 60,000 a year (2016).3
Bypassing the afterburner. The modelling chapter proposes largely bypassing the afterburner when concentrations of volatile organic compounds are low enough.5 Whether this is allowed depends on the plant’s permit; the roaster emissions entry gives examples of permit conditions.
Roast degree, batch size and electricity#
Roast degree. In the 2023 model, process energy rose with roast degree because roasting temperature rose.1 Its gate-to-gate climate change impact was 13% higher for dark-roast than for light-roast ground coffee.1 A 2024 review cites a doctoral thesis by the same first author that found a 1.6-fold difference in energy use from light to dark roasting.7
Batch size and preheating the roaster. Before the first batch, the roasting chamber and its metal must be heated. According to the modelling chapter, the heat that goes into roaster hardware, per kg of coffee, is larger for most small roasters.5 No study consulted for this entry measured warm-up or idle energy, or energy per kg across batch sizes on one roaster.
Electricity. In the 2023 model, a tonne of ground coffee took 77.8–95.8 MJ of electricity against 702.7–1,121.4 MJ of heat.1 Grinding and the cyclone were the main consumers.1 The review cites a 10-fold difference in energy use between roasting and grinding.7 One energy audit of two industrial drum roasters fitted a variable-speed drive to the exhaust fan and reported lower fan power.13
Energy in life-cycle studies#
Small share of the whole. In a 2024 review, roasting and grinding together contributed a median 8% of the cradle-to-grave footprint.7 A 2020 synthesis took its roasting-to-cup values from a 2008 German pilot study: 3.05 kg CO₂e per kg green coffee, of which consumption 70%, grinding 10% and roasting 6%.8 On those figures, roasting came to about 0.18 kg CO₂e per kg green coffee (our calculation).
Large share of the roastery. Gate to gate, the picture changes. In the 2023 model, roasting and packaging were the main contributors to the impacts and costs.1 In the 2025 model, the base plant came to 0.48 kg CO₂e per kg of green coffee.2
A worked example for natural gas
The IPCC’s default factor for natural gas is 56,100 kg CO₂ per TJ, that is 56.1 g per MJ; for LPG it is 63,100 kg per TJ.10 The factors assume 100% oxidation and apply where no country-specific factors exist.10 Worked example, our calculation: at the 1.5 MJ of natural gas per kg green coffee assumed in the 2017 case study, burning the gas releases about 84 g CO₂ per kg (1.5 MJ × 56.1 g/MJ).4,10 This assumes the 1.5 MJ is on the net calorific basis that the IPCC factors use; the case study does not state its basis.10 It counts combustion only, not gas supply chains.
The EPA’s 1995 factors for carbon dioxide from a batch roaster are 180 lb per ton of green coffee without and 530 lb per ton with a thermal oxidiser (≈90 and 265 kg per tonne).11 The EPA expects carbon monoxide and carbon dioxide from fuel combustion.11 The factors do not separate carbon dioxide from the fuel and from the beans; the roaster emissions entry gives the full table.
Measuring roaster energy use#
- Plant data. Life cycle studies collect fuel and electricity use per kg of product from plant records.6 These figures include every step in the plant, not roasting alone.
- Enthalpy balances. Models compute heat per kg of green coffee from measured or assumed temperatures and flows.2,5 The 2016 exergy study combined such a model with on-site measurements.3
- Process simulation. The 2023 study modelled the plant in process simulation software, not from plant records.1
- Basis. A value needs its basis: per kg green, roasted or ground coffee, heat or fuel, roaster alone or with afterburner.1,2,3 No standard measurement protocol for roaster energy was found in the sources consulted.
Risks and trade-offs#
- Afterburner versus energy. Burning the smoke costs fuel and adds carbon dioxide.1,11 Bypassing it saves energy only where emissions and the permit allow.5
- Intermittent heat. In batch plants, hot flue gas comes in short pulses, which makes heat recovery harder to use.4
- Cost and space. Retrofitting a preheater has to be weighed against costs and space.9 The 2016 study’s heat recovery case was calculated for large roasters of up to 4 t/h.3
- Recirculated gas. Recirculating systems must always vent part of their off-gas; the roaster emissions entry explains why.
Open questions#
- Measured energy across roasteries. Most values consulted are modelled or assumed.1,2,4 No study consulted for this entry measured fuel use per kg across many roasteries of different sizes.
- Electric roasters. No study consulted for this entry measured the energy use of electrically heated production roasters, or compared them with gas-fired roasters on a life-cycle basis.
- Drum, fluid-bed and continuous roasters. No study consulted for this entry compared the specific energy of these designs under the same conditions.
- Warm-up and idle time. How much energy a shop roaster uses before its first batch and between batches was not found in the sources consulted.
- Afterburner share. The studies consulted disagree on how much of the fuel goes to the afterburner (see Contested).
Key studies#
Gosalvitr et al., 2023 – Ground and instant coffee in the UK. A life cycle assessment and costing of coffee plants modelled in process simulation software, from light to dark roasts. In ground-coffee production, roasting took two thirds of the heat and the afterburner one third; a single-pass roaster would have raised heat use by 146–167%, and dark roasts had a higher climate impact than light ones.
Kekes et al., 2025 – Energy recovery in a small coffee plant. Modelled a plant in Athens, Greece, roasting 1,800 kg of green coffee a day with a natural-gas roaster and afterburner. Recirculating hot air from the afterburner lowered total energy per kg green coffee from 3,647.7 to 2,872.4 kJ, and greenhouse gas emissions by 25%.
Vučković et al., 2016 – Exergy of an industrial batch roaster. An exergy and cost analysis of a gas-fired 300 kg batch roaster with afterburner, partly from on-site measurements. The overall exergy efficiency was 33.26%, exhaust gases were the largest single loss, and the authors estimate a heat recovery potential of 1.1 MW for large roasters.
Pantaleo et al., 2017 – Power from intermittent roaster heat. A techno-economic case study of a 500 kg/h batch drum plant with partial hot gas recycling and afterburner, assuming 1.5 MJ of natural gas per kg green coffee. It assessed organic Rankine cycle and micro gas turbine units for exhaust heat that arrives in short pulses.
- Share of the afterburner in roasting energy: in a 2023 model of a ground-coffee plant with a recirculating batch roaster, the afterburner used 33–34% of the heat1; in a 2025 model of a small plant in Athens, Greece, with a single-pass roasting chamber in its base case, it burned 19.23 of 37.21 Nm³ of natural gas, about 52% (our calculation)2; a 2016 exergy study states in its introduction, without a citation, that afterburners can consume up to 400% more fuel than the roaster itself.3 The studies differ in plant, recirculation, oxidiser and boundary, and none of them measured the share across several roasteries.
Coffearium (2026). Roaster energy use. Coffearium – The open coffee compendium. Retrieved 17 November 2026, from https://coffearium.org/entries/roaster-energy-use/
@misc{coffearium2026-roaster-energy-use,
author = {{Coffearium}},
title = {{Roaster energy use}},
year = {2026},
howpublished = {Coffearium -- The open coffee compendium},
url = {https://coffearium.org/entries/roaster-energy-use/},
urldate = {2026-11-17},
note = {Retrieved 17 November 2026}
}Entries are updated when the evidence changes; the retrieval date tells your readers which version you used. Changes are listed in the Change log.
Change log
- – Energy range 0.7–3.6 MJ/kg no longer described as "afterburner included" (brief, quick facts, Explained text, chart title and caption); now "heat or fuel, mostly modelled", with boundary named: Kekes et al.'s QTOT is roasting plus enthalpy added for cooling (QADD 113.0 kJ/kg), and neither Kekes et al. nor Pantaleo et al. state how the afterburner enters their figure. Reason: Kekes et al. 2025, Sections 3.1 and Tables 2 and 5 (QTOT = QSR + QADD); Pantaleo et al. 2017 gives one natural-gas figure for a plant whose burner and afterburner are both gas-fired.
- – The 1.67 GJ per tonne value was labelled "partly measured" in the chart and set against "only the 2016 study rests partly on on-site measurements" in the text; now labelled as stated without a source in the study's introduction, not one of its results. The 400% afterburner statement is likewise marked as uncited introduction text (contested, table, body). Reason: Vučković et al. 2016, Introduction; neither sentence carries a reference, and the study's own on-site data concern its exergy analysis.
- – European Commission best-practice page (s9); url changed to the print version that carries the full text; savings figures (<10%, about 30%, about 10–20%) now marked in text, table, chart title, caption and brief as given without data or citation; tierNote rewritten with the tier decision; 250–450 °C no longer cited to it (the page attributes it to a roaster maker); retrofit quote corrected to the page's wording ("While it is possible …") and the text now adds that the page calls a retrofit more complex. Reason: Page print version, sections Description, Applicability and Literature; tier rules in docs/AGENT_BRIEF.md (company pages Tier C) applied to figures the page draws from roaster makers.
- – Kekes et al. 2025 base case identified as a single-pass roasting chamber (contested, Afterburner energy, table); natural gas cut with recirculated afterburner air added as about 31% (our calculation, 37.21 to 25.53 Nm³) beside the 21% fall in total energy, which includes cooling; 25% and 18% marked as life-cycle indicators; savings chart row now 21–31%. Reason: Kekes et al. 2025, Table 3 ("Single pass") and Section 3.1; arithmetic rechecked.
- – Worked example (84 g CO₂ per kg green) now shows its inputs (1.5 MJ × 56.1 g/MJ) and states the assumption that the 1.5 MJ is on the net calorific basis of the IPCC factors. Reason: 2006 IPCC Guidelines, Vol. 2, Ch. 1, Section 1.4.1.2 (NCV basis); Pantaleo et al. do not state a calorific basis.
- – Nab and Maslin values: "processing phase" changed to "roasting-to-cup stage", with consumption (70%) named, in line with carbon-footprint-of-coffee.md; EPA sentence on roaster burners as oxidisers no longer adds "by leading the exhaust back to them"; IPCC chapter author corrected to Kainou, K. Reason: Nab and Maslin 2020, Section 6.4; AP-42 9.13.2 (wording); IPCC chapter title page ("Kainou Kazunari", family name first).
Sources
- AGosalvitr, P.; Cuéllar-Franca, R. M.; Smith, R.; Azapagic, A. (2023). An environmental and economic sustainability assessment of coffee production in the UK. Chemical Engineering Journal 465, 142793. doi:10.1016/j.cej.2023.142793 · other copiesQuoted passage: the authors’ own result; Life cycle assessment and costing of ground and instant coffee made in the UK; plant modelled in Aspen Plus (batch roaster with hot-air recirculation, afterburner), not measured; per tonne of product; light to dark roasts..
- AKekes, T.; Koskinakis, S. E.; Boukouvalas, C.; Krokida, M. (2025). Enhancing Environmental Sustainability in the Coffee Processing Industry via Energy Recovery and Optimization: A Life Cycle Assessment Case Study. Sustainability 17(3), 1334. doi:10.3390/su17031334 · other copiesQuoted passage: the authors’ own result; Gate-to-gate LCA, small plant in Athens, Greece (1,800 kg green/day); natural-gas roaster (single-pass chamber in base case, Table 3) and afterburner; model, per kg green. QTOT = QSR + cooling (QADD); Nm³ time basis not stated..
- AVučković, G. D.; Stojiljković, M. M.; Vasiljević, G. M. (2016). Exergoeconomic evaluation of real processes for coffee roasting. Thermal Science 20 (Suppl. 5), S1271–S1283. doi:10.2298/TSCI16S5271V · other copiesQuoted passage: the authors’ own result; Exergy and exergoeconomic analysis of one gas-fired 300 kg batch roaster with afterburner, partly from on-site measurements (plant not named; authors in Serbia). The 1.67 GJ per tonne and the 400% afterburner statement are in the introduction without a citation..
- APantaleo, A. M.; Fordham, J.; Oyewunmi, O. A.; Markides, C. N. (2017). Intermittent waste heat recovery: Investment profitability of ORC cogeneration for batch, gas-fired coffee roasting. Energy Procedia 129, 575–582 (ORC2017). doi:10.1016/j.egypro.2017.09.209 · other copiesQuoted passage: data; Techno-economic case study (ORC, micro gas turbine) for a batch, gas-fired drum plant of 500 kg/h with partial hot gas recycling, Italy; 1.5 MJ/kg green is an assumed natural-gas input (burner and afterburner gas-fired, split not given)..
- BSchwartzberg, H. (2013). Batch Coffee Roasting; Roasting Energy Use; Reducing That Use. In: Yanniotis, S.; Taoukis, P.; Stoforos, N. G.; Karathanos, V. T. (eds), Advances in Food Process Engineering Research and Applications, Springer (Food Engineering Series), pp. 173–195 (author's text hosted by a coffee retailer). doi:10.1007/978-1-4614-7906-2_10 · other copiesQuoted passage: the authors’ own result; Enthalpy-balance model of batch, gas-fired roasters with afterburners (recirculating and single-pass), compared with a few published plant values; model, not measurement..
- AHumbert, S.; Loerincik, Y.; Rossi, V.; Margni, M.; Jolliet, O. (2009). Life cycle assessment of spray dried soluble coffee and comparison with alternatives (drip filter and capsule espresso). Journal of Cleaner Production 17(15), 1351–1358. doi:10.1016/j.jclepro.2009.04.011 · other copiesQuoted passage: data; Table 2: primary data per kg roasted and ground coffee from one plant in Spain (green coffee handling and cleaning, roasting, grinding, filling and packing, conditioning): 1.23 kg green coffee, 0.14 kWh electricity, 0.07 Nm³ natural gas..
- AChéron-Bessou, C.; Acosta-Alba, I.; Boissy, J.; Payen, S.; Rigal, C.; Setiawan, A. A. R.; Sevenster, M.; Tran, T.; Azapagic, A. (2024). Unravelling life cycle impacts of coffee: Why do results differ so much among studies?. Sustainable Production and Consumption 47, 251–266. doi:10.1016/j.spc.2024.04.005 · other copiesQuoted passage: the authors’ own result; Review of coffee life cycle assessments; median shares across cradle-to-grave studies. The light/dark and roasting/grinding ratios are cited from a 2021 doctoral thesis (Gosalvitr, University of Manchester)..
- ANab, C.; Maslin, M. (2020). Life cycle assessment synthesis of the carbon footprint of Arabica coffee: Case study of Brazil and Vietnam conventional and sustainable coffee production and export to the United Kingdom. Geo: Geography and Environment 7(2), e00096. doi:10.1002/geo2.96 · other copiesQuoted passage: a result cited from earlier work; Synthesis LCA of arabica coffee from Brazil and Vietnam to the UK; the processing-phase values are taken from the 2008 German PCF pilot study, not modelled anew..
- AEuropean Commission, Joint Research Centre Reduction of energy use through the adoption of green coffee preheating in batch coffee roasting. European Commission, Joint Research Centre, Green Best Practice Community (best environmental management practice, food and beverage manufacturing).Quoted passage: a recommendation; European Commission (JRC) best-practice page for batch coffee roasting, full text in the print version; literature 2006–2015, accessed 2013–2015; savings ranges without data or citation; references mostly roaster makers and personal communications..
- AGarg, A.; Kainou, K.; Pulles, T. (2006). 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Volume 2: Energy, Chapter 1: Introduction. Intergovernmental Panel on Climate Change (IPCC), IGES, Japan.Quoted passage: data; Table 1.4, default CO2 emission factors for combustion: natural gas 56,100 kg/TJ (54,300–58,300), liquefied petroleum gases 63,100 kg/TJ (61,600–65,600)..
- AU.S. Environmental Protection Agency (1995). AP-42, Fifth Edition, Section 9.13.2: Coffee Roasting. U.S. EPA, Compilation of Air Pollutant Emission Factors.Quoted passage: data; US emission factors (September 1995), per short ton of green coffee fed; CO2 batch roaster 180 lb/ton uncontrolled and 530 lb/ton with thermal oxidiser, rating D..
- ASan Diego County Air Pollution Control District (SDAPCD) (2024). Better Buzz 8345 AB3205 Eng Evaluation (engineering evaluation of a coffee roaster with catalytic afterburner). SDAPCD, San Diego, California (local regulation; permit evaluation signed 28 August 2024).Quoted passage: data; Permit evaluation of one drum roaster (model 'US Millennium 70 Kilo', 616 lb/h) with a catalytic afterburner, San Diego, 2024; burner ratings, not measured fuel use..
- BDimchev, I. (2026). Energy and airflow optimization in an industrial roasting system: A case study. E3S Web of Conferences 727, 01003 (EEPES 2026 proceedings). doi:10.1051/e3sconf/202672701003 · other copiesQuoted passage: the authors’ own result; Energy audit of the exhaust system of two identical industrial drum roasters with cyclone and afterburner; fan retrofit with a variable-speed drive; no fuel figures; peer review not stated..
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”.
The literature for this entry was last searched on 16 Nov 2026. A statement that no source consulted for this entry shows something refers to the sources above; newer or missing work may exist.
Discussion
A space for readers to share experience from the farm, the roastery or the bar, ask each other and offer other views. The Coffearium team does not reply here. Please keep to the community rules; to change the entry itself, use Suggest a change.