Limestone into Cement
Limestone and clay give up their carbon dioxide, roll down a turning kiln and sinter into clinker near 1,450 °C. Ground with a little gypsum, they make cement.
Scroll
Limestone and Clay
Cement starts as rock. Limestone brings the lime; clay or shale brings silica, alumina and iron oxide; where one falls short, a little sand, iron ore or bauxite makes it up (USGS). The USGS puts a tonne of clinker at about 1.7 t of raw material, about 1.5 t of it limestone or a similar rock.
Real recipes are about four-fifths limestone: a reference raw meal in a 2026 paper is 79 % limestone, about 19 % clay and sand and 2 % iron; other mixes run 83–86 % limestone. Ground together into a fine powder, they are the raw meal.
In the USGS’s typical clinker the lime is about 65 %, the silica 22 %, the alumina 6 % and the iron oxide 3 %. Three ratios steer the mix: the lime saturation factor (LSF), the lime there is against the most that the silica, alumina and iron can take up, usually 0.92–0.98 in modern clinker; the silica ratio, 2–3; the alumina ratio, about 1.6.
CaO 65 % · SiO₂ 22 % · Al₂O₃ 6 % · Fe₂O₃ 3 % · about 1.5 t of limestone per tonne of clinker
Five Stages of Cyclones
The raw meal goes first into the preheater tower. The EU’s reference document on best techniques describes four to six stages of cyclones stacked in a tower 50–120 m high, the top stage often two cyclones side by side. The hot gas from the kiln rises through them; the meal is fed into it just before the top stage.
Each cyclone flings the meal out of the gas; it drops into the duct of the stage below, is carried up again and flung out again, stage by stage, hotter each time. The meal spends only 20–90 seconds in the tower (USGS).
More stages recover more heat: in plants with a precalciner burning fossil fuels the gas leaves at about 280 °C with six stages, 310 °C with five and 350 °C with four (ECRA, 2017). The model has five, the top one a pair.
4–6 cyclone stages · a tower 50–120 m high · 20–90 s for the meal
Lime from Limestone
Limestone is mostly calcium carbonate. Heated far enough it breaks into lime and carbon dioxide: CaCO₃ → CaO + CO₂. 100 g of calcium carbonate leaves 56 g of lime; 44 g goes off as gas. The step takes in heat, about 178 kJ per mole (US National Bureau of Standards, 1969), some 1.78 MJ per kilogram (converted here).
From the same tables, the carbon dioxide over calcium carbonate reaches one atmosphere at about 885 °C (computed here); for practical burning the USGS gives about 950 °C or more.
A modern line does most of this in the precalciner, between the kiln and the cyclones. It runs at 850–950 °C and burns up to 65 % of all the fuel; its combustion air, the tertiary air, comes hot from the cooler through the duct over the kiln. The meal enters the kiln mostly calcined (75–95 % in the EU reference document).
CaCO₃ → CaO + CO₂ · about 178 kJ per mole · 850–950 °C in the precalciner
Three to Five Turns a Minute
The rotary kiln is a steel tube lined with refractory brick, sloping 2.5–4.5 % and turning 0.5–5 times a minute (EU reference document); the IEA puts it at about three to five. As it turns, the material slides and tumbles down towards the flame at the lower end.
Kilns with preheaters are typically 10 to 17 times as long as they are wide. One real kiln, at the Ste. Genevieve plant in Missouri, is 6.6 m across and 93 m long on three supports, and makes about 12,000 t of clinker a day (Global Cement, 2019). The material takes about 30–40 minutes to pass through the rotary kiln (EU reference document).
Each support is a riding ring round the shell, resting on a pair of rollers; a girth gear driven by a pinion turns the whole kiln. The model’s kiln is drawn about twice as thick as a real one and turns four times a minute.
slope 2.5–4.5 % · 0.5–5 turns a minute · one kiln: 6.6 m across, 93 m long, three supports
Only Partly Molten
At the lower end the flame is at about 2,000 °C; the material in the burning zone must be kept at 1,400–1,500 °C (EU reference document). Belite (C₂S) starts forming from about 700 °C, most of it at 1,100–1,200 °C; alite (C₃S) forms from belite and lime, and without a liquid that step is extremely slow (USGS).
Liquid appears at 1,338 °C, where the four clinker minerals and a melt can coexist. In an average clinker about 20 % is liquid near 1,340 °C and about 30 % at 1,400–1,450 °C, enough for alite to form fast; so 1,400–1,450 °C is the burning zone’s target (USGS).
The material does not melt through: partly molten, it sinters into nodules of clinker about 1–10 cm across. The burning zone is lined with basic brick such as magnesia-spinel, under a crust of clinker coating; towards the feed end the lining is high-alumina brick (RHI Magnesita). The glow in the section is drawn by temperature, as a schematic.
flame about 2,000 °C · material 1,400–1,500 °C · liquid from 1,338 °C · about 30 % liquid at 1,400–1,450 °C
Four Minerals
Clinker is mostly two calcium silicates, alite (close to C₃S, Ca₃SiO₅) and belite (close to C₂S), with aluminate (C₃A) and ferrite (close to C₄AF) between them. Cement chemists write C for CaO, S for SiO₂, A for Al₂O₃ and F for Fe₂O₃. Modern Portland cement holds about 50–70 % C₃S, 10–30 % C₂S, 3–13 % C₃A and 5–15 % C₄AF (USGS).
Polished and etched under a microscope, alite shows hexagonal outlines, belite rounded grains with a layered inside, ferrite bright white and aluminate grey (NIST). In three real clinkers the alite crystals were about 12–110 µm, the belite about 15–35 µm. The porcelain dish at the front holds nodules of clinker.
In 1929 Bogue gave a set of formulas for the minerals from the oxides; from 1940 they were in the US cement specification. By them the typical clinker of the first chapter holds about 53 % C₃S, 23 % C₂S, 11 % C₃A and 9 % C₄AF (computed here). Real minerals are not pure: in NIST’s three reference clinkers, X-rays found 5–9 points more alite than the formulas give (computed here).
C₃S · C₂S · C₃A · C₄AF · Bogue for a typical clinker: about 53 / 23 / 11 / 9 %
Cooled Fast
The clinker drops into the cooler at about 1,450 °C (EU reference document). In a grate cooler, air is blown up through the bed of clinker lying on the grate. The USGS gives about 100 °C as safe for handling; a good grate cooler gets down to 80 °C above the air around it (ECRA, 2017).
The air comes off at 600–1,000 °C and goes back into the line: some as the burner’s secondary air, some through the tertiary air duct to the precalciner (ECRA, 2017).
Cooled fast, the minerals stay as they are. Measurements cited in a 1987 paper put the start of alite’s breakdown into belite and lime, on slow cooling, between 1,210 and 1,025 °C; large belite crystals can also turn into the γ form, swell and crumble the clinker.
in at about 1,450 °C · out at about 100 °C · air back at 600–1,000 °C
A Little Gypsum
Ground together with a little gypsum, clinker becomes Portland cement. The IEA says every cement holds about 4–5 % gypsum; the USGS gives 3–7 %. Under the European standard a CEM I Portland cement is 95–100 % clinker in its main constituents, the calcium sulfate counted apart.
The gypsum is there for the aluminate. Without sulfate, C₃A reacts almost the moment it meets water and the cement takes a “flash set”; with gypsum, a shell of ettringite grows round the aluminate grains and slows them down (USGS).
Ground cement grains average about 10 µm, and 85–95 % are under 45 µm (USGS). Many cements now carry other materials too: clinker was about 71 % of the world’s cement in 2022 (IEA).
clinker + about 4–5 % gypsum · grains about 10 µm · clinker about 71 % of world cement (2022)
Not by Drying
Cement does not harden by drying. The European standard defines it as a powder that, mixed with water, sets and hardens by hydration and keeps its strength even under water; “hydraulic”, the USGS writes, means it can set and harden under water, or with water to spare, through hydration.
Alite reacts with water to give calcium silicate hydrate (C-S-H) gel and calcium hydroxide; the textbook shorthand is 2 C₃S + 6 H → C₃S₂H₃ + 3 CH, H for water (USGS), and real gel holds about 1.7–1.8 calcium to each silicon. The C-S-H is the glue. Belite reacts the same way, only slower: at 28 days about 80 % of the alite has reacted, about 40 % of the belite.
So the strength grows day by day. Standard mortar of one CEM I 42.5 R cement reached 16.8 MPa at 1 day, 28.8 MPa at 2, 43.6 MPa at 7 and 57.0 MPa at 28 (supplier’s data, in a 2020 paper). The cube at the front is hardened cement mortar.
by hydration, not drying · 2 C₃S + 6 H → C₃S₂H₃ + 3 CH · about 57 MPa at 28 days (one CEM I 42.5 R)
Carbon Dioxide Twice
Making clinker sends out carbon dioxide twice. Once from the limestone itself: a clinker of 65 % lime, all of it from calcium carbonate, gives off 0.51 t per tonne (USGS; the IPCC’s default, with the kiln dust, is 0.52 t). And once from the fuel burnt to heat it.
In the industry’s 2006 figures (GNR) a tonne of clinker gave off 866 kg on average, typically about 540 kg of it from the raw material and so about 330 kg from the fuel (converted here). A kiln’s thermal efficiency only touches the fuel’s share; the stone’s stays. A cleaner fuel still leaves the carbon in the stone to come out.
Andrew (2019) puts the world’s process emissions from cement at 1.50 billion tonnes of CO₂ in 2018, about 4 % of the emissions from fossil fuels; the IEA (2018) puts the whole cement industry at about 7 % of the world’s CO₂. How carbon dioxide warms the air is on the greenhouse page.
per tonne of clinker: about 0.54 t from the stone, about 0.33 t from the fuel (GNR 2006) · process CO₂ about 4 % of fossil-fuel CO₂
Named after a Stone
The Romans used lime and volcanic ash. Vitruvius wrote of a powder from round Mount Vesuvius that, mixed with lime and rubble, set even under the sea. That was not Portland cement.
On 21 October 1824 Joseph Aspdin was granted British patent No. 5022 for a cement “which I call Portland cement”, after the set mortar’s likeness to Portland stone. His method only called for burning until the carbonic acid was driven off, short of a sintered clinker; his son William Aspdin is generally credited with burning it hot enough to make an alite clinker from 1842, with I. C. Johnson following.
Ransome patented a rotary kiln in 1885; it was first made to work in the United States around 1890, and there, in 1892, a little gypsum was found to slow the set. The first multi-stage cyclone preheater started at Beckum in Germany in 1953; precalciners first ran in Japan in the late 1960s.
patent 1824 · clinker from 1842 · rotary kiln patent 1885 · cyclone preheater 1953
Four Billion Tonnes
The USGS estimates the world made about 4.0 billion tonnes of cement in 2024: China about 1.9 billion, close to half (47.5 %, converted here), India about 450 million, Vietnam about 110 million, the United States about 86 million. Andrew (2019) puts it at more than half a tonne per person a year.
Cement is not concrete. Concrete is cement, water, sand and gravel or crushed stone mixed together; by the USGS’s volumes, the cement powder is only about 7–15 % of it. The paste of cement and water binds the stones into an artificial rock; with sand alone, it is mortar.
A tonne of cement makes 3.4–3.8 m³ of concrete weighing 7–9 t (USGS). Along the front, left to right: limestone, raw meal, clinker, cement and a cube of hardened mortar.
about 4.0 billion t in 2024 · China about 1.9 billion t · cement powder about 7–15 % of concrete by volume
The Kiln in Parts
The preheater tower and the precalciner lift together, and the hood lifts off; the kiln’s shell parts into three sections between its riding rings and slides apart along its axis, each still cut open so its lining and bed show. The grate cooler stays.
Rest the pointer on a part and it lifts on its own, named; otherwise the parts stand apart together.
preheater tower · precalciner · hood · inlet section · middle section · burning zone · grate cooler
Heat and Lime
Computed here: the raw meal’s silica ratio (2.5) and alumina ratio (1.6) are held; its lime saturation factor is a lever, which sets the clinker’s four oxides, and Bogue’s formulas give the minerals they could make. The other lever is the burning zone’s temperature: burnt short, some lime stays uncombined, and belite that should have become alite stays belite.
How much free lime stays comes from one measured series: a raw meal of LSF about 1.00, held for 35 minutes, kept about 13.8 % at 1,250 °C, 4.3 % at 1,350 °C and 1.9 % at 1,450 °C (a 2019 paper); for other raw meals, F.L. Smidth’s burnability formula adds 0.31 % per point of LSF. The model’s alite comes out 5–6 points below that series, about Bogue’s own shortfall.
Carbon dioxide: the stone’s share from the clinker’s lime, the fuel’s held at about 0.33 t. Readouts: alite, belite, free lime and CO₂ per tonne of clinker; in the chart the temperature runs across, the lines are alite and belite, and the dots the series’ measured alite.
a model computed here · silica ratio 2.5, alumina ratio 1.6 · measured: 69 % alite at 1,450 °C with LSF 1.00
Try thisTake the burning zone from 1,450 °C down to 1,350 °C: how much alite is lost, and how much free lime stays? Then set the lime saturation factor from 0.95 to 0.90: does the CO₂ per tonne of clinker change?
What the model leaves out
This is a model of the principle, not of any one plant: the kiln is drawn about twice as thick as a real one, the preheater tower about two-thirds of its real height and the cooler short; the cyclones, ducts and meal pipes are simplified, and the raw mill, coal mill, dust filter, fans and clinker silo are not shown. The bricks, the bed and the nodules in the section are drawn larger than real, and the glow is drawn by temperature as a schematic, not a measurement. The lab is Bogue’s formulas plus one measured free-lime series: no reaction rates, no impurities in the minerals, and the fuel’s carbon dioxide held fixed.
Easy to get wrong
Cement hardens by hydration, not by drying, and hardens under water too. Clinker is sintered, only partly molten, never melted through. Cement is not concrete; it is the binder in it. The carbon dioxide from the raw material comes from the breakdown of calcium carbonate, not from burning, and a change of fuel cannot remove it. “About 4 % of fossil-fuel emissions” counts only that process share; “about 7 %” is the whole cement industry (IEA). Aspdin’s product of 1824 was not yet Portland cement as it is now made. “About four-fifths limestone” is a rough figure from a few real recipes, not a fixed ratio.
Common questions · 5
- What happens to limestone when it is heated?
Heated far enough it breaks into lime and carbon dioxide: CaCO₃ → CaO + CO₂.
03Lime from Limestone ↑- Is cement clinker melted?
The material does not melt through: partly molten, it sinters into nodules of clinker about 1–10 cm across.
05Only Partly Molten ↑- Why is gypsum added to cement?
Without sulfate, C₃A reacts almost the moment it meets water and the cement takes a “flash set”; with gypsum, a shell of ettringite grows round the aluminate grains and slows them down (USGS).
08A Little Gypsum ↑- Does cement harden by drying?
The European standard defines it as a powder that, mixed with water, sets and hardens by hydration and keeps its strength even under water; “hydraulic”, the USGS writes, means it can set and harden under water, or with water to spare, through hydration.
09Not by Drying ↑- If a cement kiln used clean fuel, would it still emit carbon dioxide?
A kiln’s thermal efficiency only touches the fuel’s share; the stone’s stays.
10Carbon Dioxide Twice ↑
Sources
- H. G. van Oss (2005): Background Facts and Issues Concerning Cement and Cement Data, USGS Open-File Report 2005-1152 (raw materials, typical clinker, minerals, kiln reactions and temperatures, the liquid, hydration, fineness, concrete, calcination CO₂)
- U.S. Geological Survey (2025): Mineral Commodity Summaries, Cement (output by country, 2024)
- European Commission JRC (2013): Best Available Techniques reference document for the production of cement, lime and magnesium oxide (preheater, precalciner, kiln, cooler, energy, emissions)
- IEA and WBCSD CSI (2018): Technology Roadmap, Low-Carbon Transition in the Cement Industry (kiln speed, cooling, gypsum, energy, the industry’s share)
- IEA (2023): Cement, Tracking Clean Energy Progress (clinker ratio, energy; archived page)
- ECRA and CSI (2017): Development of State of the Art Techniques in Cement Manufacturing (exhaust gas by number of stages, coolers)
- WBCSD CSI (2009): Cement Industry Energy and CO₂ Performance, “Getting the Numbers Right” (CO₂ per tonne of clinker, 2006)
- IPCC (2006): Guidelines for National Greenhouse Gas Inventories, vol. 3, ch. 2 (the clinker emission factor)
- R. M. Andrew (2019): Global CO₂ emissions from cement production, 1928–2018, Earth System Science Data 11
- P. Stutzman and S. Leigh (2002): NIST Technical Note 1441, and the SRM 2686–2688 reference clinker certificates (the minerals under the microscope, measured phases)
- G. Frohnsdorff (2001): Calculation of Compounds in Portland Cements, NIST SP 958 (the history of Bogue’s formulas)
- ASTM C150’s Bogue equations (reprinted in US patent 5,843,222)
- K. H. Stern and E. L. Weise (1969): NSRDS-NBS 30, High Temperature Properties and Decomposition of Inorganic Salts, Part 2, Carbonates
- Materials 19 (2026): Investigation of Solar Waste Sand as a Supplementary Raw Material for OPC Clinker (a reference raw meal, the typical ratios)
- Materials 12 (2019): Effect of the Nano-Ca(OH)₂ Addition on the Portland Clinker Cooking Efficiency (alite and free lime at three temperatures)
- US patent 8,029,618 (F.L. Smidth’s burnability formula, as quoted)
- Materials 13 (2020): Influence of the Treatment Temperature on Thermoactivated Recycled Cement (the strength of one CEM I 42.5 R)
- EN 197-1 (sample pages): the definitions of cement and of Portland cement clinker
- A. Cuesta et al. (2019): IUCrJ 6 (the calcium-to-silicon ratio of C-S-H)
- F. Puertas and F. Triviño (1987): Materiales de Construcción 37 (cooling, and alite’s breakdown)
- P. Edwards (2019): Plant report, Ste Genevieve, Global Cement Magazine (one real kiln)
- RHI Magnesita (2019): Cement Lining Concept (the bricks of each zone)
- D. Moore: the Cement Kilns site, Aspdin’s 1824 patent in full
- D. Moore: Cement Kilns, About cement (early cement)
- D. Moore: Cement Kilns, Early rotary kilns
- D. Moore: Cement Kilns, Suspension preheater kilns
- Kent Historic Environment Record: Aspdin’s kiln (William Aspdin and the clinker)
- Vitruvius: The Ten Books on Architecture, Book II, ch. 6 (pozzolana), trans. M. H. Morgan
Data
The clinker’s typical make-up, the kiln’s reactions and temperatures, the liquid, hydration, fineness, concrete and the calcination factor are from H. G. van Oss, Background Facts and Issues Concerning Cement and Cement Data (USGS Open-File Report 2005-1152); world output from the USGS Mineral Commodity Summaries (2025); the preheater, precalciner, kiln, cooler and energy figures from the EU’s Best Available Techniques reference document for cement, lime and magnesium oxide (2013), the ECRA/CSI technology papers (2017) and the IEA (2018, 2023); the carbon dioxide figures from the WBCSD CSI “Getting the Numbers Right” report (2009), the IPCC 2006 Guidelines, R. M. Andrew (Earth System Science Data, 2019) and the IEA. Bogue’s equations are ASTM C150’s, as reprinted in US patent 5,843,222; the reference clinkers and the microscope from NIST (SRM 2686–2688, TN 1441, SP 958); calcite’s dissociation from NSRDS-NBS 30 (1969); the raw meal, the ratios and the burn series from papers in Materials (2019, 2026); the burnability slope from F. L. Smidth’s formula as quoted in US patent 8,029,618; the strengths from Materials (2020) and EN 197-1; the cooling from Materiales de Construcción (1987); the history from D. Moore’s Cement Kilns site, Kent’s Historic Environment Record and Vitruvius. The lab’s model, the Bogue values and the conversions are computed here. The kiln line is a model of the principle, not any one plant. Material presets, parts and the softbox studio of the site’s model kit are ported from AirsupHQ/machine-skill (MIT, © 2026 Air Supply Technologies Inc.).
The text, pictures and models on this page are licensed CC BY-NC 4.0: free to use and adapt for teaching and other non-commercial purposes, with credit to world.lapla.si. Third-party material listed above keeps its own licence.
Cite this plate
Laplace’s Mini World, Plate 74, “Limestone into Cement”, https://world.lapla.si/en/cement/, CC BY-NC 4.0, updated 2026-11-01.

