Global Thermal Phosphoric Acid Market Size, Share & Forecast 2026–2033
The acid that is pure because its feedstock was distilled
Thermal phosphoric acid is not purified. It is born pure. Phosphate rock and coke are reduced in a submerged-arc electric furnace at roughly 1,300 degrees Celsius to elemental phosphorus, at about 13,000 kilowatt-hours per tonne of P4. The phosphorus leaves the furnace as a vapour, is condensed, and is later burned and hydrated to phosphoric acid. Because the feedstock was distilled, the acid carries almost none of the iron, aluminium, magnesium, sulfate and fluorine that a wet-process plant has to take back out afterwards by solvent extraction, ion exchange or membrane filtration.
That difference in process order explains the entire commercial structure of this market. On the wet route the impurity burden is the cost centre. On the thermal route the impurity burden is absent and the power bill is the cost centre: one tonne of 100 per cent H3PO4 carries 4,109 kilowatt-hours, because a tonne of P4 yields 3.164 tonnes of acid on the stoichiometry of P4 + 5O2 → P4O10 and P4O10 + 6H2O → 4H3PO4.
The market was USD 1,180.0 million in 2025 on 1,093.0 kilo tons of product, a blended realisation of USD 1,079.6 a tonne. It reaches USD 1,857.0 million on 1,407.0 kilo tons by 2033, or USD 1,319.8 a tonne, compounding at 5.83 per cent in value and 3.21 per cent in volume. Value runs ahead of volume for a reason that is structural rather than cyclical, and it is set out below.
The inversion that defines the market
China is 89.9 per cent of the tonnage and 78.4 per cent of the value. That gap is the most useful fact in this report. Chinese material is assessed at RMB 6,760 a tonne for 85 per cent acid — Hubei 6,500 to 7,000, Yunnan around 6,600, Sichuan 6,300 to 6,800 in June 2025 — which is USD 941 a tonne at 7.18. The 110 kilotonnes made outside China realise USD 2,311 a tonne, two and a half times as much, because that is where semiconductor etching acid, pharmaceutical-grade and food-grade material is made, and where the purity of the thermal route is still paid for rather than merely specified.
This is therefore two markets sharing one molecule: a large, low-margin, price-assessed commodity concentrated in four Chinese provinces, and a small, very high-value specification business running out of Japan and a handful of other jurisdictions. The forecast treats them separately, because they fail separately.
Why 2026 re-rated the route
On 28 February 2026 the Strait of Hormuz closed. Sulfur was assessed at USD 815 to 820 a tonne FOB Middle East and USD 1,200 CFR Brazil on 28 May 2026 against roughly USD 100 before. A tonne of 85 per cent purified acid carries 0.523 tonnes of sulfur, so that cost line moved from USD 52 to USD 429 a tonne of product. The entire raw-material spread the wet route held over the thermal route on 2024 inputs was RMB 2,838 a tonne; the sulfur move alone was RMB 4,130 — 1.46 times the whole spread.
For the first time in twenty years the furnace became the cheaper way to a pure tonne. Chinese thermal acid was running at 35.46 per cent utilisation in the week of 5 September 2025, at 18,900 tonnes, against 64.75 per cent for purified wet-process acid at 60,800 tonnes. That ratio is the slack this market is now drawing on, and it is why the volume line in this forecast is not a capacity story: Chinese yellow phosphorus capacity is administratively frozen, so the only tonne available to the acid market is a tonne already inside the furnace base.
Companion analysis. The cost mechanics of the furnace are written up for a non-specialist reader in Phosphorous Acid Is Priced by a Power Bill, Not a Rock Price, which traces the same tariff from the electrode to the derivative and shows why a contract indexed to rock hedges a cost the supplier does not carry.
Companion analysis. The utilisation arithmetic behind the volume line is set out in Sixty-Three Per Cent Is Not Idle Capacity, which separates nameplate from the effective band yellow phosphorus can actually run in, and explains why 83 per cent is the number that matters rather than 63.
Table des matières
Paysage concurrentiel
Fragmented tonnage, oligopolistic purity
The structure of this market is two structures. In tonnage it is fragmented: an implied Chinese capacity base of 2.77 million tonnes spread across Yunnan, Sichuan, Guizhou and Hubei producers, many of them attached to captive hydro or to a local phosphorus chemical cluster, competing on a published weekly assessment. In purity it is oligopolistic: 110 kilo tonnes made outside China at USD 2,311 a tonne, qualified into semiconductor, pharmaceutical and food specifications that a new entrant cannot buy its way into.
Consolidation in the Chinese tonnage market is administrative rather than financial. Capacity is allocated by policy, so share moves when a quota moves, and the transaction that changes the map is a permit rather than an acquisition. That is why recent M&A activity in this market is nil even while the underlying phosphorus complex restructures around it.
Jiangsu Chengxing Phosph-Chemicals (SSE: 600078)
The clearest read on the Chinese thermal position is Chengxing’s FY2025 annual report. Revenue was RMB 3,596.3 million, up 7.16 per cent, while net profit attributable to shareholders was RMB 8.5 million against a loss of RMB 198.9 million in 2024 — a swing of 186.70 per cent in total profit on a revenue base that grew by a fourteenth. A producer with captive hydro and its own phosphorus furnace, selling into a price-assessed market, converts a 7 per cent revenue move into a near-total profit swing. That is the operating leverage of this route, in both directions.
Chengxing’s structure is the argument of this report in miniature: the plants are named Phosphorus-Electricity because the two are one business, and the group holds the Leidatan hydro station in Yunnan so that the tariff inside its own cost stack is partly its own.
Rasa Industries (TSE: 4022)
At the other end, Rasa’s FY2025 (year ended 31 March 2026) shows what the purity premium is worth: net sales JPY 47,727 million, up 5.1 per cent, operating profit JPY 6,012 million, up 26.9 per cent, with high-purity phosphoric acid at JPY 16,796 million or 35.2 per cent of sales and a stated global top share in semiconductor etching acid. Revenue growth in the mid single digits produced operating profit growth in the high twenties — the same leverage as Chengxing, pointed at specification instead of at a commodity assessment.
The scale that governs strategy
Set the market against its participants. At USD 1,180.0 million, this is RMB 8,472.4 million at 7.18 — 2.36 times Chengxing’s annual revenue. The entire global thermal acid market is worth a bit more than two years of one integrated Chinese producer’s turnover. No firm builds furnace capacity for this market; it builds it for the phosphorus, and sells the acid because the acid is what the phosphorus becomes. Any competitive analysis that treats thermal acid as a standalone prize will misread the incentives of every participant in it.
The tier below
Six producers can be identified with meaningful merchant positions: two Chinese integrated phosphorus-electricity groups, a Japanese high-purity specialist and its domestic peers, and a European and an Indian supplier serving regional food and industrial grades. The rest of the Chinese capacity base is captive or semi-captive, selling spot into the assessment when the derivative chain pays less than the acid chain.
Chaîne de valeur
Three nodes, one molecule
The chain has three nodes and each one is priced by something different. Rock is priced by a mine index. Yellow phosphorus is priced by a power tariff. Acid is priced by the sum of the two plus a conversion margin that is thinner than the tonnage suggests.
World phosphate rock mine production was 250 million tonnes in 2025 on the U.S. Geological Survey’s February 2026 estimate, and roughly 3.7 per cent of that volume leaves the fertiliser system for an electric furnace or a purification plant. Chinese yellow phosphorus output ran above 1.2 million tonnes in 2026 against a capacity base that policy forbids from growing by one net tonne, which makes the furnace base a fixed endowment and the acid market a contest over how hard that endowment is run.
The conversion margin, measured
On 5 September 2025 yellow phosphorus traded at RMB 22,500 to 22,550 a tonne in Yunnan and Sichuan and RMB 22,550 to 22,600 in Guizhou. At 3.164 tonnes of acid per tonne of P4, that is RMB 7,111 of phosphorus in every tonne of 100 per cent acid. The same week, 85 per cent thermal acid at RMB 6,760 converts to RMB 7,953 a tonne of 100 per cent product. The oxidation step therefore earned RMB 841.6 per tonne of acid before its own burning, hydration, drumming and freight costs — about 10.6 per cent of the product price for the step that turns a hazardous pyrophoric solid into a food additive. That is the whole merchant conversion business, and it explains why nobody builds a furnace to sell acid.
Where the molecule goes instead
Yellow phosphorus has competing destinations and the acid market does not get first call on it. Phosphorus trichloride for glyphosate and organophosphorus intermediates, phosphorous acid for phosphonates, and export volumes all bid for the same tonne. When the derivative chain pays more than the acid chain, the acid market is supplied last — a supply mechanism that has nothing to do with acid demand and everything to do with where the phosphorus margin sits that quarter.
Integration into power, not into rock
The defining structural move in this market is ownership of electricity rather than ownership of mines. Jiangsu Chengxing Phosph-Chemicals names its two yellow phosphorus plants for the integration itself — Xuanwei Phosphorus-Electricity and Mile Phosphorus-Electricity — and holds the Leidatan hydro station inside the group. A rock mine hedges a cost the thermal producer only pays indirectly; a power station hedges the cost that actually sets the price.
Downstream, and where this report stops
Thermal acid flows into food acidulants and beverage acidulants under E338 and FCC monographs, pharmaceutical intermediates, metal treatment and phosphating, semiconductor and display etching, flame retardants, and the phosphate salts that the phosphate rock report explicitly excludes. This report values the acid at the producer gate. It does not double-count the derivative markets sized in this segment’s own phosphorous acid and phosphonate report, and it does not claim the phosphate salt tonnage that sits under the rock report.
Activité d'investissement
No new thermal tonnage is being built
Investment in this route is maintenance capital and purity finishing, not capacity. The capacity is capped, the utilisation is historically low, and the return on a new furnace is set by a tariff the investor does not control. The capital that is being committed goes to the competing route: China commissioned 2.48 million tonnes of new purified wet-process capacity during 2026 against 4.64 million tonnes of output, a supply surplus of roughly 570,000 tonnes, financed on the LFP cathode chain rather than on food or industrial demand.
Direction: declining for tonnage, selective for specification. New money in thermal acid appears as cleanroom finishing, ppb metals capability and drumming lines that convert commodity acid into electronic-grade product, because that is the only place in this market where the price is not an assessment.
What the 2026 repricing does to the calculus
The sulfur shock improves the thermal route’s relative cost by more than the entire 2024 spread — RMB 4,130 against RMB 2,838 — and that is a genuine improvement in the return on existing furnace capacity. It does not unlock new capacity, because the cap does not move. The realistic investment response is higher utilisation of what exists and shorter payback on debottlenecking and purity upgrades, and that is exactly what the volume path in this report assumes.
Why nobody is buying anybody
Recent M&A activity in this market is nil. The reason is structural: the Chinese tonnage is allocated by permit, so a purchase buys assets without buying the right to expand them, and the high-purity assets outside China are held by groups for whom the acid is a strategic capability rather than a portfolio asset. Rasa’s electronic materials business sits alongside compound semiconductor materials and radioactive iodine adsorbents — the acid is not for sale separately.
The one scenario that changes it
A sustained sulfur price above the 2024 spread would make thermal acid structurally profitable again rather than opportunistically so, and a decade-long inversion would eventually force the capacity question. That would be a policy decision in China about the phosphorus cap, not an industrial one, and it is not in the base case.
Technologie et innovation
A mature furnace with nowhere to go on cost
The submerged-arc furnace is a mature piece of industrial equipment. Electrode consumption, off-gas recovery, slag and ferrophosphorus handling, and condenser design have all been optimised against a specific energy consumption that has sat near 13,000 kilowatt-hours a tonne of P4 for years. There is no step-change available: the thermodynamic minimum is a large fraction of the actual, and the remaining gap is engineering detail rather than a new process.
That matters commercially. When a process cannot get more efficient, its cost curve is flat and its price is set entirely by its two inputs — power and rock — which is why this market has no cost-down story to tell and why its long-run price path is a tariff path with a purity premium on top.
The innovation is on the competing route
Every process patent worth reading in phosphorus over the last decade is on the wet side: solvent extraction and MAP-DCP routes to purified acid, ion exchange, nanofiltration, and the trace-metal control that gets a wet plant to battery specification. Capital followed the patents, and the run rates prove it. In the same week of September 2025 that thermal acid ran at 35.46 per cent, purified wet-process acid ran at 64.75 per cent, and China commissioned 2.48 million tonnes of new purified capacity during 2026 against 4.64 million tonnes of output.
The thermal route’s technology position is therefore defensive. Its purity is a property of its feedstock, not of an improvement it can license, and it wins where specification is harder to reach than cost.
The last defensible premium: the etcher
Semiconductor etching is that place. Rasa Industries reports high-purity phosphoric acid sales of JPY 16,796 million in FY2025, 35.2 per cent of its JPY 47,727 million of net sales, and states a global top market share in high-purity phosphoric acid for semiconductor etching solutions. That is a thermal-route product sold on parts-per-billion metals and particle control into a qualification cycle measured in years, and it is the one segment where the wet route has not simply outbid the furnace.
The test to watch is not a new furnace design. It is whether wet purification reaches electronic grade at scale — the barrier there is not chemistry but magnetic foreign matter and trace metals at the ppb level, and every year that barrier holds, the thermal route keeps its highest-value 10 per cent of the market.
Where the next efficiency actually comes from
Not from the acid step. From the power contract and from the furnace’s ability to run at the top of its effective band. Chinese yellow phosphorus runs at 83 per cent of the effective 1.20 million tonne band rather than 63 per cent of the 1.58 million tonne nameplate, which means the headroom is 200,000 tonnes and not 580,000. Capturing that headroom is an operating and power-procurement exercise, and it is worth more than any process change available to the route.
Risque de marché
1. The reversal risk is the base case inverted
The whole improvement in this market’s position rests on sulfur. The wet route held a raw-material advantage of RMB 2,838 a tonne on 2024 inputs; the 2026 sulfur move was RMB 4,130. If sulfur normalises toward its pre-shock level, the wet route regains the spread, and thermal acid returns toward the 35.46 per cent utilisation that defined it in September 2025. Ten points of utilisation is 277 kilo tonnes — a quarter of the global base. Nothing in the volume line of this forecast is safe from that, and no producer can hedge it, because the thing being hedged is a shipping chokepoint.
2. The tariff passes straight through
At 4,109 kilowatt-hours per tonne of 100 per cent acid, a change in delivered power cost moves the floor price of this market directly and immediately. There is no substitution available inside the process and no efficiency frontier left to reach: the furnace is mature, and the remaining variable is the contract attached to it. Producers with captive hydro are structurally advantaged and structurally exposed, because their tariff is also their hydrology.
3. The cap cuts both ways
A frozen capacity base means this market cannot chase a price spike. When acid is scarce and the price doubles, the thermal producer cannot add a tonne — the tonne has to come out of a competitor’s utilisation or out of a derivative customer’s allocation. Upside is capped at the same time as downside is unhedged, which is an unusual risk profile for a commodity and is why the market carries a high overall risk rating despite its small size.
4. Concentration, and not the usual kind
One country makes 89.9 per cent of the tonnage. A policy change in Beijing, a provincial power allocation, a drought in Yunnan or a change in the capacity cap moves the world price of this acid more than any demand-side event. Geopolitical exposure is rated high on that basis alone, and no diversification strategy at the buyer level fixes a single-origin supply base.
5. Specification risk at the top of the value curve
The high-purity segment is where the money is: 110 kilo tonnes earning USD 2,311 a tonne against USD 941 in China. That premium depends on wet purification not reaching electronic grade at scale. It has not, and the technical barrier — trace metals and magnetic foreign matter at ppb — is real. But it is a barrier under sustained attack, and a single qualified wet-route electronic-grade supplier would take the best 20 per cent of this market’s value out of it without removing a single tonne of volume.
6. The measurement risk, stated plainly
The non-Chinese tonnage in the base year is a residual estimate of 110 kilo tonnes, not a reported figure, because no public series breaks thermal acid output down by country. A 50 kilo tonne error moves the global base by 4.6 per cent. It does not change the shape of the forecast or the direction of any argument in it, but it does change the base, and a reader comparing this number with another should know that the Chinese tonnage is measured while the rest is derived.
Paysage réglementaire
The binding constraint is a capacity number, not a chemical rule
The regulation that actually governs this market is Chinese and it is industrial rather than environmental: yellow phosphorus capacity is capped, and the cap is written so that the base cannot grow by one net tonne. Everything downstream inherits it. A thermal acid producer cannot respond to a price signal by adding a furnace, so supply elasticity in this market is a utilisation number between roughly 35 and 83 per cent and nothing else.
Around that cap sit the ordinary instruments: energy-intensity and energy-consumption permits for a process that is among the most electricity-intensive in the chemical industry, environmental impact assessment for furnace off-gas and slag, and provincial allocation that in Yunnan and Sichuan interacts with hydro seasonality, so that the same furnace has a different effective cost in a wet year and a dry one.
Product specifications are the real market access
On the product side, the rulebook is a stack of monographs and the grade decides the customer. Food use runs on the FCC monograph and E338 in the European Union, with limits on arsenic, lead and fluoride. Pharmaceutical use runs on USP and Ph. Eur. monographs and brings qualification and change-control obligations that make a supplier switch a project rather than a purchase. Electronic use runs on SEMI grade classifications and customer-specific metals and particle specifications that are stricter than any pharmacopoeia.
The consequence for this market is that approvals are compliance-driven rather than licence-based: there is no permit that admits a tonne to the semiconductor market, only a qualification that takes years and is lost faster than it is won.
Trade and transport
Thermal acid moves as 85 per cent solution in lined or polyethylene drums and in ISO tanks, classified as a corrosive with an exposure hazard, which puts packaging, labelling and driver certification into the delivered cost. High-purity electronics material moves in much smaller packages with much stricter handling, and its freight is a rounding error against its value — which is exactly why that segment is served from Japan to fabs worldwide rather than made next door.
What regulation does not touch
Nothing in the current rulebook prices the carbon embedded in 4,109 kilowatt-hours a tonne of acid. A border adjustment applied to phosphorus chemicals would land almost entirely on this route rather than on the wet one, because the wet route’s emissions are process and sulfate rather than electrical. That is the one regulatory change that would reprice the whole market, and it is not in the base case.