Rutile lost 22.45 per cent of its price while its supply was falling. The ore was never the reason.

Bulk rutile at a minimum of 95 per cent TiO₂, FOB Australia, cleared at USD 1,140 a tonne in 2025 — 22.45 per cent below the 2022 assessment of USD 1,470. Over the same window, world rutile mine production fell from 460,000 to 450,000 tonnes of contained TiO₂.

A supply base in structural decline was priced as though it were in surplus. That is the finding in Pheonix Research’s Global Rutile Market Report, Size, Share and Forecast 2026–2033, and it produces a growth rate that looks almost perverse for a mined commodity: the market compounds at 5.47 per cent a year, and tonnage supplies 0.45 points of it. The other 5.00 is the recovery of a price that was never about the ore.

The snapshot

The increment is USD 292.8 million, a multiple of 1.532x. It is close to the inverse of a mined commodity in growth: rutile tonnage rises 3.66 per cent across eight years, because it cannot rise faster. A new mine requires a defined heavy mineral sands province, a mining lease, a wet concentrator, a dry mill and a mineral separation plant — justified on a product that is a few per cent of what comes out of the ground. At USD 1,140 a tonne, a single tonne of rutile requires on the order of 25 to 60 tonnes of ore to be moved, concentrated and separated.

What this report counts, and what it deliberately leaves out

473,684 tonnes is natural rutile crossing a commercial boundary at the price paid, FOB Australia basis, on a minimum-95-per-cent-TiO₂ specification. The USGS reports world rutile production as 450,000 tonnes of contained TiO₂; the price is quoted per bulk tonne of a 95 per cent product. Converting at the assessment grade — 450,000 ÷ 0.95 = 473,684 tonnes — is the difference between a USD 513 million figure and the USD 550.8 million used here, and the ladder is disclosed so a reader can see exactly where the 7.4 per cent sits.

Three things are deliberately outside the number. Synthetic rutile — Iluka’s own contracted book alone is 110,000 tonnes a year, and industry capacity exceeds 240,000 tonnes — is a manufactured upgrade, not a mined one. Chloride slag, of which Tronox alone operates 410,000 tonnes of capacity, competes for the same chloride-route duty. And ilmenite, at 9,400,000 tonnes in 2025, is the base of the titanium mineral market of which rutile is 4.57 per cent of the tonnage. Add the first two and this becomes a multi-billion-dollar chloride-feedstock market with an entirely different price mechanism — driven by the acid and energy economics of upgrading rather than by reserve depletion. This report stays on the mined, depleting, grade-premium product, and says so.

Why the price fell, and why one customer moved it

Iluka Resources’ half-year to June 2026 is the clearest evidence in the sector, and it describes the opposite of a demand collapse. Rutile production fell 27.81 per cent, from 35.6 to 25.7 thousand tonnes. Rutile sales rose 34.63 per cent, from 20.5 to 27.6 thousand tonnes. Synthetic rutile production was nil, against 113.1 thousand tonnes a year earlier, and synthetic rutile sales fell 48.08 per cent to 36.5 thousand tonnes. Both kilns sit idle with restart “subject to market conditions”, and the 110 thousand tonnes contracted for 2026 will be serviced from inventory.

The company’s own explanation is not a demand collapse. It is “loss of sales to Venator and rebalancing of other contracts.” Meanwhile US apparent consumption of titanium mineral concentrates rose 10 per cent in 2025. The ore did not lose its buyers; one buyer left the register, and the tonnes that had been contracted to it had to find a home at whatever price cleared. In a market with four producers of consequence and a handful of buyers, there is no other mechanism.

The turn has already begun. Iluka’s realised rutile price ran USD 1,110 in the December 2025 quarter, 1,252 in March 2026 and 1,256 in June 2026 — 13.15 per cent in two quarters off the trough, with the supply base still depleting and both kilns still dark.

The spread that held while the price fell

Rutile’s premium over ilmenite, both FOB Australia, was 2.77x in 2022 and 2.85x in 2025. Its premium over 80–95 per cent TiO₂ chloride slag was 1.30x. Both multiples held while the underlying prices fell by roughly a fifth. The spread is set by grade and by chloride-route eligibility — rutile carries very low magnesia and lime, and those two impurities are what a chloride reactor cannot tolerate — and the chemistry behind it has not changed in three decades.

That is where the value sits. On 473,684 tonnes, the premium over chloride slag is worth USD 123.2 million a year — 22.4 per cent of the market. The premium over ilmenite is worth USD 350.5 million, or 63.6 per cent. Nearly two-thirds of the value of this market is a grade premium, which is why a producer’s reserve grade, and not its tonnage, is the number that matters.

Four producers, and far fewer buyers

World output is close to the most concentrated of any bulk mineral feedstock. Australia 200,000 tonnes, Sierra Leone 110,000, South Africa 100,000 — three countries hold 91.11 per cent of the 450,000-tonne total. Everything else combined, Canada, Kenya, Ukraine, India, Madagascar, Brazil and Mozambique, accounts for roughly 40,000 tonnes.

Two things follow. The marginal tonne is now African, not Australian: Sierra Leone and South Africa together supply 210,000 tonnes, 46.67 per cent of world output, and both carry fiscal, licensing and logistics risk that mature Australian tonnage did not. And concentration on the buying side is worse than on the selling side. The chloride-route pigment industry is a handful of plants in North America, Europe and East Asia, and one of them leaving the register moved the global price by more than a fifth. Fewer buyers than sellers, and no exchange to clear an uncontracted cargo, is the single most important fact about this market.

Even the most integrated participant is structurally short the product. Tronox holds 832,000 tonnes of titanium feedstock capacity — 182,000 tonnes rutile and leucoxene, 240,000 synthetic rutile and 410,000 slag — of which natural rutile is 21.88 per cent. It buys rutile on the market, and so does everyone else who is not Iluka.

Read the model as a spread, not a point

The forecast is built bottom-up: volume from named operations and their reserve life, price from the grade spread held constant on top of a demand-driven level. Blended price recovers from USD 1,162.80 to USD 1,718.00 by 2033, which is 5.00 per cent a year and 16.87 per cent above the 2022 assessment in nominal terms — not a boom forecast for a feedstock whose supply cannot respond, and the conservative case against a straight extrapolation of the 13.15 per cent two-quarter recovery now underway.

The downside is named and does not require a collapse.

• Substitution case — USD 808 million by 2033 at 4.91 per cent. Chloride slag and synthetic rutile take the growth tonnage, rutile volume flatlines, and the price recovery runs at 4.45 per cent rather than 5.00.
• Demand-contraction case — USD 665 million by 2033 at 2.38 per cent. Pigment demand contracts as anti-dumping duties settle and Chinese chloride capacity displaces Western plants, the price recovery stalls at 2 per cent, and volume declines 1 per cent a year as producers high-grade.

Neither of those is bolted on afterwards. Both sit inside the model.

Four things that would break the forecast

1. A second Venator-scale exit. One buyer leaving the register has already produced a 22.45 per cent price event. A repeat is a further 10–15 per cent move, and it needs no change in demand to happen.
2. The kilns relighting. Restarting Iluka’s two synthetic rutile kilns puts 110,000 to 240,000 tonnes of chloride-grade product into the pool in a single management decision. It is the largest near-term supply swing available in this market, and it is a price trigger rather than a market-existence question — the contracts already exist.
3. The trade remedies. Anti-dumping duties on Chinese TiO₂ pigment determine whether Western chloride-route plants run, and those plants are the rutile buyers. Duties that hold keep this demand intact; a settlement that lets Chinese chloride capacity take share closes the plants that buy the product.
4. The acid price, in rutile’s favour for once. Sulphuric acid is the dominant reagent cost in the sulphate route and a minor one in the chloride route. As acid gets more expensive, ilmenite-consuming plants lose competitiveness against high-grade chloride feed. It is the only swing factor in this market that transfers demand to rutile without requiring demand to grow at all.

What is not a risk, though it is sold as one

Resource depletion is not a threat to this market. It is the market’s support. Nordic Mining’s own quarterly commentary on its new Engebø operation — ilmenite and garnet, not rutile — confirms the mechanism from the opposite direction: “The supply of natural rutile continues to decline due to resource depletion and lack of new large scale projects.” A new mineral sands mine can be financed, permitted and built in this cycle. It just cannot be a rutile mine, because the world’s known large rutile provinces are already in production and their grades are falling. Capital is flowing into the substitute, not the product. Anyone modelling depletion as a downside has the mechanism backwards.

Two costs are also smaller than they look, and both matter to a trader. Licensing, rehabilitation bonding, NORM management and social-plan obligations add an estimated USD 35 to 60 a tonne — 3.1 to 5.3 per cent of the 2025 assessment. Ocean freight from Western Australia to Northwest Europe is in the order of USD 40–60 a tonne, 3.5 to 5.3 per cent of product value. The same freight on ilmenite at USD 400 a tonne is 10 to 15 per cent. The grade premium partly pays for its own logistics, which is why rutile trades globally and ilmenite trades regionally — and why a trader’s arbitrage in this product is a specification arbitrage, not a freight one.

Live links

• Report: Global Rutile Market Report, Size, Share and Forecast 2026–2033: The Price Fell 22 Per Cent and the Ore Was Not the Reason
• The price fall: Rutile Fell 22 Per Cent While Its Supply Fell. The Ore Was Not the Problem.
• The grade premium priced: The Same Tonne of Titanium Is Worth 2.85 Ilmenites
• The critical-minerals import-dependence comparison: Brazil Vanadium Pentoxide (V2O5) Market Report, Size & Forecast 2026–2033: Supply, Exports and US Import Dependence

About this report

Two research layers, and both are named.

The secondary layer carries every number in this report. It comprises the USGS Mineral Commodity Summaries series for titanium mineral concentrates, from which the production volumes, country split, FOB Australia price series and import-reliance figures are taken; Iluka Resources’ 2026 half-year results and quarterly production reports, which supply production, sales, realised price and synthetic rutile operating status; Tronox’s Form 10-K for fiscal 2025, which supplies the feedstock capacity breakdown; Nordic Mining’s quarterly interim report for independent confirmation of rutile supply decline; the published disclosures of Base Resources and Rio Tinto for asset status; and standard dry-bulk route assessments for freight.

The primary layer establishes why the price behaves as it does, and it carries no number of its own. It comprises producer, trader and technical interviews across the Australian, African and European channels; buyer-side procurement interviews in chloride pigment, welding consumables and titanium sponge; and a buying-criteria exercise establishing which specification attributes a chloride reactor will pay for and which it will reject. Where this report says a spread is structural rather than cyclical, that judgment comes from the primary layer. Where it says a number, the number comes from the secondary layer. The two are never blended without attribution.

No competitor market-research estimate is cited anywhere in this report. Where a published third-party figure conflicts with this reconstruction, it has been excluded rather than reconciled, and the reconstruction is stated with its method so a reader can test it. The report does not disclose plant-level capacity for privately held producers, does not attribute volume market share to individual companies where that share is not separately reported, does not model the pigment market, and does not forecast the outcome of the Rio Tinto divestment, which is a transaction rather than a trend.

About Pheonix Research

Pheonix Research builds bottom-up market intelligence from unit economics: volume multiplied by the price that volume actually attracts, reconciled line by line against primary disclosures rather than accepted as a headline. Our reports separate the layer being measured from the layers adjacent to it, disclose what is excluded and why, and publish the ladder back up so a reader can reconcile to a wider definition instead of taking either number on trust. Primary research establishes mechanism and carries no number.