Calcined Clay Low-Carbon Cement: 7 Plants, ~2.3 Million Tonnes of Powder, and the Polycarboxylate Gap Nobody Has Closed

Two things happened almost at once in September 2026. Holcim commissioned its dedicated calcined clay line at Čížkovice in the Czech Republic, paired with 580,000 tonnes of ECOPlanet low-carbon cement. In the same month, TCC Group disclosed that calcined clay capacity across its Portuguese and African operations had reached 1.5 million tonnes a year — a figure that had been a target ("no less than 1.5 Mt by end-2025") and is now delivered.

For the cement industry, this is a low-carbon route graduating from demonstration to scale. For the admixture industry, the plant to watch is not the cement works but the ready-mix yard: change the binder system and the polycarboxylate dosage, the slump-retention window and the entire formulation all have to be recalculated. This article uses our comparison of three mainstream industrialised calcined clay technologies as its skeleton, then adds capacity updates from the second half of 2026, the published research on how calcined clay and PCE interact, and the equipment side in China, to answer one question: where is the admixture demand on this track, how large is it, and what shape does it take.

At a glance	Key readings (with scope)
Routes	Three: D‑Pozzolan® (Dynamis / Argos — gasification pre-combustion plus a controlled-reduction rotary kiln), deOHclay® (CIMPOR / TCC — rotary kiln in oxidation, or flash suspension, either way), proximA Tech (Holcim — dedicated kiln in oxidising atmosphere, in both new-build and retrofit modes)
Plants	Seven: Rio Claro (Colombia), Sobradinho (Brazil), Abidjan (Côte d'Ivoire), Kribi (Cameroon), Saint-Pierre-la-Cour (France), Čížkovice (Czech Republic), Guayaquil (Ecuador)
Powder	Six plants with disclosed figures total about 1.81 Mt/yr; counting the French line's 500,000 t "paired" scope as well, about 2.31 Mt/yr
Group	TCC/CIMPOR Portugal plus Africa 1.5 Mt/yr (2025 sustainability report); Holcim clinker factor 70% (2025) → 65% (2030), with ECOPlanet already at 36% of net cement sales
CO₂ cut	D‑Pozzolan about 38%, deOHclay 31%–40%, proximA 30%–50% — all against CEM I, each on the operator's own basis
Cost	LC3 production cost can be up to 25% lower than OPC (Cementis estimate) — but admixture cost moves the other way: saved on clinker, spent on formulation
System	LC3-50 = 50% clinker + 30% calcined clay + 15% limestone + 5% gypsum. Workability is slightly worse than OPC and the system "lacks a chemical admixture fully matched to it" (Miao Changwen, 2018)
Demand	About 6,500–10,000 t of PCE liquid per 1 Mt of low-carbon cement (our calculation; assumptions in Section 5)
China	TCDRI's packaged kaolin suspension-calcination equipment; Sinoma International (Nanjing) ran 600 t/d industrial trials on clay tailings with a firing heat consumption of 1,800 kJ/kg — China sells the equipment, not the binder

1. Seven plants, three routes: how far industrialisation has actually run

Start with the map. All three routes overlap heavily in the 700–850°C activation window. The real differences come down to three things: atmosphere, residence time, and whether the process can be decoupled from a clinker kiln.

D‑Pozzolan® takes the long reduction route. Fuel is first gasified under oxygen starvation in a D-Gasifier pre-combustion chamber, and the reducing flue gas is fed into the kiln head to reduce Fe³⁺ in the clay to Fe²⁺ — which is how iron-rich red clay can be fired into a grey supplementary cementitious material. That is its only technical moat, and also its most expensive part. deOHclay® runs fully in oxidation and splits into two branches: a conventional rotary kiln version, and a flash suspension version that holds material in the gas stream for a few seconds to complete dehydroxylation and then quenches it to lock in reactivity. The significance of the latter is not speed — it is that a plant can be built entirely independently of a clinker kiln. proximA Tech is also fully oxidising, but packages multi-stage heat recovery and modularisation into a standard technology kit that works both for new build and as a retrofit hung off an existing clinker kiln.

Route	Owner and plant (start-up)	Clay powder ／ paired low-carbon cement	Notes
D‑Pozzolan®	Cementos Argos, Rio Claro (Colombia, 2019-04)	450,000 t/yr (1,500 t/d) LC3 / PPC	Once the world's largest dedicated calcined clay line; ~38% CO₂ reduction and ~30% energy saving; Schenck metering and blending
D‑Pozzolan®	Vicat Ciplan, Sobradinho (Brazil, 2009)	550 → 700 t/d PPC pozzolanic cement	Early demonstration plant; petcoke fuel; feed is >70% kaolinitic clay at about 25% moisture
deOHclay®	CIMPOR, Abidjan (Côte d'Ivoire, 2020-08)	300,000 t/yr LC3	Rotary kiln route; CIMPOR's first commercial line; ~31% CO₂ reduction
deOHclay®	CIMPOR, Kribi (Cameroon, 2024-07)	250,000 t/yr LC3	The world's first flash-calcination plant with no clinker kiln; cocoa-husk biomass fuel; up to 40% CO₂ reduction
proximA Tech	Holcim, Saint-Pierre-la-Cour (France, 2023-02)	Paired capacity 500,000 t/yr 500,000 t ECOPlanet	Europe's first dedicated line; 100% biomass plus waste-heat recovery; up to 50% CO₂ reduction
	Holcim, Čížkovice (Czech Republic, 2026-09)	>115,000 t/yr 580,000 t ECOPlanet	The Czech Republic's first dedicated line; ~30% CO₂ reduction
	Holcim, Guayaquil (Ecuador, H1 2025)	About 465,000 t/yr Up to 2 Mt/yr of cement	Existing clinker kiln converted; Holcim's largest calcined clay line in Latin America

Three additional signals are worth recording. First, Holcim's retrofit list is no longer limited to Ecuador: Mexico, Ecuador, Spain and Egypt are all converting existing clinker kilns to calcined clay lines. Second, TCC's 1.5 million tonnes is reported as "Portugal plus Africa" combined, which means the European base is producing too. Third — and this is the one most easily missed — Ghana has two projects running at the same time: CIMPOR's third deOHclay line, and a USD 100 million LC3 plant at Tema owned by Supacem under CBI Ghana. The direct reason the latter could start is that the Ghana Standards Authority adopted the LC3 standard GS PAS 5:2024 in 2024.

Standards are the sluice; capacity is the water. The country that writes LC3 into its standard is the country whose demand starts the clock — and that matters more to admixture companies than it does to cement companies.

2. Pull the capacity scopes apart: "powder", "low-carbon cement" and "paired" are not the same thing

The technology comparison underlying this article sets one constraint on itself very clearly: capacity must distinguish "calcined clay powder capacity" from "paired low-carbon cement capacity", and the two must never be mixed. Mixing them inflates numbers two- or threefold out of thin air — because at LC3-50's 30% substitution rate, one tonne of calcined clay powder makes more than three tonnes of cement.

Scope	What it means	Disclosed in this batch	Addable?
Calcined clay powder capacity	Tonnes of reactive SCM actually fired at the plant	Rio Claro 450k; Sobradinho 231k (700 t/d × 330 days); Abidjan 300k; Kribi 250k; Čížkovice >115k; Guayaquil 465k — about 1.81 Mt/yr	Yes, but Sobradinho is a derived figure
Paired low-carbon cement capacity	Tonnes of finished cement containing that powder	Saint-Pierre-la-Cour 500k; Čížkovice 580k; Guayaquil up to 2 Mt; TCC scope about 5 Mt/yr	Yes, but only these four are disclosed
Group-level capacity	The total a group reports externally	TCC/CIMPOR Portugal + Africa 1.5 Mt/yr	No — adding it to plant figures double counts

There is one honest blank in the source work: CIMPOR's third deOHclay line in Ghana has neither a verified capacity nor a verified start date, so no number was entered. The group scope can fill that blank with an upper bound: 1.5 Mt (Portugal plus Africa combined) minus Abidjan's 300,000 t and Kribi's 250,000 t leaves 950,000 t/yr — to be split between the Portuguese base and the Ghanaian project, with the split not disclosed. So the Ghanaian line's powder capacity cannot exceed 950,000 t/yr. That is the hardest conclusion available.

Once the scopes are pulled apart, a judgement that usually gets skipped surfaces: these seven lines total roughly 1.81–2.31 Mt of powder capacity, which does not sound like much; but the low-carbon cement they unlock is over 8 Mt. Admixtures go into concrete, not into powder — so what actually sets demand is the cement column, not the powder column.

3. Change the binder and the superplasticiser has to be rebuilt: four mechanisms

Academician Miao Changwen wrote a very direct sentence back in 2018: because of particle-size distribution and chemical adsorption effects, concrete made with a calcined clay–limestone composite binder has slightly worse workability than ordinary Portland cement concrete, and the system lacks a chemical admixture fully matched to it. Eight years on, capacity at the cement end has multiplied several times over; the technical meaning of that sentence has not changed.

Break it into four mechanisms and it becomes clear what the admixture actually has to change.

3.1 Specific surface and chemical adsorption: part of the PCE gets "eaten" first

PCE works by adsorbing onto particle surfaces and pushing particles apart through a combination of electrostatic repulsion and the steric hindrance of polyether side chains. That mechanism has one precondition: the molecules that adsorb have to stay on the cement particles. Calcined clay breaks that precondition twice.

The first break is quantity. Calcined clay is far finer than cement clinker, with a specific surface several times larger. At the same dosage, reaching the same surface coverage requires a multiple of the molecule count — a Zhejiang University team studying LC3 concrete in southern China spelled out the causal chain directly: calcined clay is much finer than OPC, so at equal water-to-binder ratio the LC3 paste is more cohesive and needs more superplasticiser to reach the same workability. The second break is strength of bonding. Metakaolin's layered structure and surface active sites adsorb PCE not just by physical coverage but by chemical adsorption, holding it more firmly than clinker particles do, so fewer free molecules remain available to disperse anything.

Li, Lei, Sui and Plank measured that curve systematically in Cement and Concrete Research (2021): with a calcined clay of roughly 50% kaolinite content, across a 0–40% clinker substitution range, PCE effectiveness falls as the substitution rate rises. Which means one thing — dosage cannot be extrapolated linearly. Trying to convert "1.0% is enough in OPC" directly into "1.3% is enough in LC3" will very likely fail.

3.2 Reactive alumina and sulfate: what changes is the shape of the time curve, not the starting point

Metakaolin brings a large amount of reactive Al₂O₃. In LC3 that alumina couples with limestone to form calcium carboaluminates, which is precisely why the calcined clay plus limestone pairing can push clinker down to 50% without losing strength. For rheology, the price is that the whole sulfate balance window shifts.

Too little gypsum and the aluminium phase consumes sulfate rapidly to form AFm, so flow loss arrives early; too much gypsum and the risks of retardation and later expansion rise. Worse, PCE's carboxyl groups and SO₄²⁻ compete for the same adsorption sites on the particle surface — change the sulfate concentration and the effective PCE adsorption capacity changes with it. So the slump-retention problem in LC3 concrete is often not "not enough of the slow-release base was added" but a change in the shape of the time curve itself: the starting point can be adjusted, the slope cannot. Flatt's group studied exactly this in the Journal of the American Ceramic Society in 2025 — the mechanism of flow loss in LC3 cannot be explained by importing the OPC framework wholesale.

3.3 Residual carbon and under-calcined clay: two kinds of "false adsorption"

This category is the most hidden, and the most easily misdiagnosed as "the customer's cement has a problem".

Residual carbon comes from the reduction route. D‑Pozzolan®'s kiln has to hold a controlled reducing atmosphere, with secondary air added at the kiln end to burn out CO — a narrow window. Residual carbon is a porous structure with very strong adsorptive capacity, and it will pull PCE out as if it were an organic compound; the mechanism is identical to that of fly ash with a high loss on ignition. So for customers on this route, a high PCE dosage may not be a formulation problem at all — it may be carbon carried in the cement.

Under-calcined clay comes from a temperature window that runs too low. Kaolinite dehydroxylation is incomplete, the layered structure survives, and PCE's polyether side chains intercalate into it — the industry's familiar "clay tolerance" problem, except that it normally shows up in dirty aggregates and has now moved into the binder. Conversely, over-calcination (above roughly 900°C) recrystallises metakaolin into mullite and destroys reactivity, but the specific surface falls at the same time, so the material becomes "less hungry for superplasticiser".

Put those two together and you get a conclusion that is bad news for sales: on the same calcination line, a small shift in the temperature window changes both reactivity and superplasticiser demand, and not necessarily in the same direction. This is the hardest kind of customer for an admixture plant — not because the formulation cannot be built, but because the cement itself is drifting.

3.4 Alkalinity and pore solution: the molecular conformation changes too

Halve the clinker and the alkali content and pH of the pore solution move down accordingly. PCE's polyether side chains provide steric hindrance by extending through their hydration layer; change the pH and ionic strength and both the degree of extension and the adsorption conformation change. Layer on top of that the slower early hydration rate of LC3 and a different rhythm of C-S-H formation, and the "floor" under early flow retention is different as well.

The practical implication is plain: for the same base, any formulation validated in OPC must be re-tested in LC3 across three data sets — paste flow, mortar flow loss over time, and concrete slump retention. There is no shortcut. And that is exactly where technical service can charge for itself.

4. The calcination method itself changes water demand

All four mechanisms above describe calcined clay as a material. There is one further layer that often gets missed: how it is fired changes particle morphology, and morphology changes water demand.

Flash (suspension) calcination produces more spherical particles with more regular morphology and lower flow resistance. A technical review from the China National Building Material system made a clear judgement: suspension calcination is favoured for its high energy efficiency and the high proportion of spherical particles, and it not only cuts energy consumption by about 40% but also reduces the water demand of the cement product. Sinoma International (Nanjing) ran a 600 t/d industrial trial on clay tailings on its new-generation suspension calcination line and produced LC3 cement from it, with a firing heat consumption of just 1,800 kJ/kg — clinker firing is a 3,000 kJ/kg-plus proposition.

Rotary kiln product, by contrast, is more angular and more densely structured, and behaves differently on specific surface. So for the same tonne of calcined clay, flash-route product is gentler on superplasticiser demand than rotary kiln product, partly offsetting the penalty that metakaolin's high specific surface would otherwise impose.

That difference directly determines the technical profile of two kinds of customer. Flash-route customers — kiln-free plants such as Kribi — have lower absolute admixture demand but are far more sensitive to batch consistency, because flash residence time is measured in seconds and any wobble in feed uniformity shows up as a wobble in reactivity. Rotary kiln customers need higher volumes but have a wider process window and more stable cement performance.

5. Run the numbers: how much PCE does 1 Mt of low-carbon cement correspond to

Set the mechanisms aside and do the money first. Every assumption below is laid out in the open; if you disagree with any of them, substitute your own and recalculate.

Assumption	Value	Basis
Binder content	350 kg/m³	Normal ready-mix range
PCE liquid dosage	1.0%–1.5% of binder mass	Normal ready-mix range; LC3 takes the upper bound, for the reasons in Section 3
Share entering ready-mix	65%	The remainder is mortar, precast and site batching
Derived result	1 Mt of low-carbon cement ≈ 6,500–10,000 t of PCE liquid per year	2.86 million m³ of concrete × 3.5–5.3 kg/m³ × 65%

Apply that coefficient to the disclosed cement figures: Holcim's three plants (France 500k + Czech Republic 580k + Ecuador 2 Mt) total 3.08 Mt, and TCC's paired scope is about 5 Mt, giving 8.08 Mt/yr combined — which corresponds to roughly 50,000–80,000 t/yr of PCE liquid. Assume all 8.08 Mt enters ready-mix and the upper bound rises to 81,000–122,000 t/yr. So the reasonable range is 50,000–120,000 t/yr.

Two cautions apply to that number. First, it is a floor: Argos and Vicat Ciplan disclose no cement-side figure and are excluded. Second, it is not today's volume but the volume at full ramp — Guayaquil's 2 Mt is a "maximum" capacity, and Čížkovice only started up in September 2026. Ramping takes time.

But the direction is settled: once these lines reach full output, these seven plants alone support an annual PCE demand pool in the tens of thousands of tonnes, approaching the hundred-thousand-tonne mark. And the nature of that pool matters — it is not new concrete volume, it is existing concrete switching binder. In other words, whoever's formulation already works with these customers keeps the business; whoever's does not, loses the tonnage to someone else.

6. The admixture profile of each route

It is the same calcined clay, but the three routes impose completely different technical demands on the customer, and the selling approach should not be the same either.

Route	Cement-side traits	Direct effect on admixtures	What to sell
D‑Pozzolan® (reduction + gasification pre-combustion)	Narrow residual-carbon window; the colour-control patent lets iron-rich red clay come out grey; the highest investment and O&M barrier	Residual carbon adsorbs PCE, so dosage drifts upward passively; architectural-finish projects with tight colour requirements are highly sensitive to formulation changes	Defoamers, sacrificial agents, carbon-tolerant base formulations — plus a stability promise that "the colour will not move"
deOHclay® (oxidation / flash, kiln-free option)	Can be built independently of a clinker kiln; local biomass feedstock; customers are mostly in clinker-short emerging markets	Lower water demand from the product, but sensitive to mineral uniformity; customers are often new grinding stations with no mature cement–admixture matching experience	A complete formulation package plus on-site commissioning, not base sold by the drum; responsiveness matters more than price
proximA Tech (oxidation, new-build and retrofit)	One plant can switch between LC3, CEM II/C and CEM IV; the highest degree of formulation freedom	One admixture has to serve several binder systems at once, and the system switches with the order book	Highly adaptable base plus fast reformulation capability; in essence, selling "formulation management"

These three profiles correspond to three completely different commercial relationships. D‑Pozzolan customers look most like traditional cement plants — asset-heavy, process-sensitive, slow to decide, but expensive to displace once you are in. deOHclay customers look most like project customers — project-based, wanting turnkey delivery and people on site. proximA customers look most like group accounts — standardised, centrally procured, and judged on global consistency.

7. The Chinese side: selling equipment, not binder

China's position in this cycle is unusual: not the definer of the binder, but the provider of the equipment.

TCDRI (Tianjin Cement Industry Design and Research Institute) has developed a proprietary packaged kaolin suspension-calcination process and equipment set, due for domestic industrial application shortly, and is working with large international cement groups to take the technology into international markets. Sinoma International (Nanjing) has completed a 600 t/d industrial trial on clay tailings on its new-generation suspension calcination line and produced LC3 cement from it. There is also a lighter path inside China: in Guangdong, shield-tunnelling spoil is being used to make LC3, which can run on existing cement lines with no major retrofit, cutting cost by up to 25% and carbon emissions by about 40%.

Put those three together and they point to one judgement: China exports "how to fire", not "what standard to fire to". The LC3 formulation system and the standards agenda sit with EPFL and IIT Delhi; cost-performance and delivery speed on the equipment sit with China.

For an admixture company, that mismatch is actually good news. Equipment exports are bundled by nature — a suspension calcination line arrives with commissioning, training, formulation and consumables attached. Overseas customers get Chinese equipment, commissioned by Chinese engineers, and the probability that they end up using a Chinese base formulation is far higher than the probability of getting them to switch to a European one. That channel does not need the LC3 standard to land first; it only needs the equipment to land.

8. Three misreadings that are easy to fall into

Misreading one: calcined clay is a fly ash substitute. It is not. Fly ash and slag were not "replaced because they performed badly" — they are already fully allocated, and supply is falling, as coal plants and steel mills close. The availability of both SCMs can only get worse. LC3's role is to fill that gap, and kaolinitic clay is the only option that can still scale. The distinction matters: your target customer is not "the people who used fly ash", but "the people who depended on fly ash and can no longer get it". Their pain point is security of supply, not cost.

Misreading two: LC3 is cheaper, so admixtures should get cheaper too. The direction is the opposite. LC3's cost advantage comes from clinker substitution (up to 25% lower), and its workability penalty has to be bought back with admixtures. The money saved at the cement end and the money spent at the admixture end sit in different line items and different budgets — the cement plant's procurement department saves, and the ready-mix yard's material cost rises. So the argument "cement got cheaper, why should the superplasticiser go up" will certainly be made, and it does not hold. This is a negotiation problem, not a technical one.

Misreading three: if you have clay, you can do it. The barrier is kaolinite content and calcination control, not the deposit. From the same pit, both under-calcination and over-calcination fail to produce reactivity — and "under-calcined" happens to be exactly the defect an admixture senses most easily: you cannot see it, but your dosage will read it out for you. Conversely, if a customer keeps complaining that dosage will not come down, asking whether their calcination temperature window has moved recently is more useful than reformulating first.

9. If you sell PCE: four things you can do now

Turn "LC3 compatible" into a table, not a claim. Paste flow, mortar flow loss over time, and concrete slump retention — three data sets, run at 30%, 40% and 50% substitution, with readings at the initial point and at 60 minutes. That table is your entry ticket to a technical conversation with the customer. Most suppliers today cannot produce it.
Watch standards, not just capacity. The Ghana Standards Authority adopted GS PAS 5:2024 in 2024, and the USD 100 million LC3 plant followed. Capacity is the water; standards are the sluice. The country that writes LC3 into its standard is the country whose demand starts the clock; before that, what you are discussing is samples, not orders.
Extend the customer list from "cement plants" to "cement plants plus grinding stations". The core selling point of the flash route is that it can be built without a clinker kiln. Those customers have no clinker line, and therefore no accumulated cement–admixture matching experience. What they lack is not cheap product; it is someone to tell them how much to use.
Sell the ability to reformulate as the cement drifts, not just the base. On the reduction route, residual carbon and reactivity move with operating conditions. A supplier who can respond to that is much harder to displace than a quote that is 5% cheaper — because replacing you means walking the same ground again.

10. One last point

Calcined clay is not just another supplementary cementitious material. It is the switch that opens up cement's formulation freedom all at once — clinker can go down to 50%, the SCM count can go from one to three, and the clay source can extend from quarries to tunnelling spoil and ceramic waste.

Once that switch is thrown, the cement plant gets a decarbonisation number and the ready-mix yard gets a new problem. And nobody has written the standard answer to that problem yet.

Miao Changwen said in 2018 that the system "lacks a chemical admixture fully matched to it". Eight years later, capacity has run from zero to more than two million tonnes of powder and over eight million tonnes of cement, and the sentence still holds. That is both the gap and the most concrete window in this cycle.

We have compiled the seven operating plants across the three mainstream industrialised routes, their capacity scopes, their decarbonisation ranges and each route's formulation characteristics, and added capacity updates from the second half of 2026 plus the equipment-side progress in China. If you need a deeper mapping of supply relationships and formulation fit for a specific group, country or route, you can reach us through the about page.