A Quarter of the Barrels Carry 45.5 Per Cent of the Additive Bill

A refinery running 100,000 barrels a day of light sweet crude and a refinery running the same volume of heavy sour are identical on paper and unrecognisable in the chemical store. That asymmetry is the entire refinery additives market. Pheonix Research sizes it at USD 3,939.2 million in 2025, rising to USD 5,193.5 million in 2033 — a 3.52 per cent compound annual rate against 1.29 per cent volume growth, in a market whose underlying crude throughput is close to static: 83.5 million barrels a day this year, 84.5 million by 2033.

Eight years therefore buy 10.8 per cent more tonnage and 31.8 per cent more value. That gap is not price escalation. It is mix. The barrel count hardly moves; what moves is how much of it is defective in ways that require chemistry.

The full sizing, the crude-class breakdown and the six chemistry families are set out in Pheonix Research’s Global Refinery Additives Market report. What follows is the argument underneath it.

Where 45 per cent of the bill comes from

Split the global slate three ways and the arithmetic does the talking.

A quarter of the barrels carry 45.5 per cent of the money. The ratio between the two ends of the slate is 4.3 to 1 today and widens to 5.1 to 1 nominal by 2033, because heavy sour is both the faster-growing class and the one whose share of the barrel is rising. Light sweet is the only class shrinking on both counts — 30 per cent of barrels to 25, and its per-barrel rate growing at half a per cent a year, which is roughly the rate at which a mature chemical line tracks inflation and no more.

What “heavy” costs at the unit, measured rather than assumed

The per-barrel figures above are the load-bearing numbers in the report, so they were anchored to operating data rather than to a multiplier. The published practitioner benchmark from refinery chemical engineers puts crude-unit process chemicals at USD 10–20 per thousand barrels — one to two cents a barrel — and that benchmark explicitly excludes corrosion control, water treating, naphthenic acid inhibitors and antifoulants. In other words, the cheap end of this market is the part that touches the crude unit only.

The same engineering discussion carries an operator figure: Suncor’s treatment of a 160,000 barrel a day CDU/VDU running Canadian bitumen at CAD 6–7 million a year, or about CAD 0.111 a barrel — roughly seven times the conventional benchmark — on a feed the operator itself described as worst-case, with high total acid number, chlorides at 10–15 parts per million and high bottom sediment and water.

The heavy sour rate of USD 0.235 a barrel sits above that figure, and deliberately so: it includes the units downstream of the crude unit, where the consequences of a bad barrel are felt rather than merely measured. Nothing in this market scales with throughput. Everything scales with what is wrong with the feed.

The discount pays the bill, and that is the fragility

A refiner does not buy desalting chemistry because it is in a chemicals budget. It buys it because the alternative is unplanned downtime, and it can afford to because the barrel it is treating was bought cheap. That chain is quantifiable.

A high-acid crude — Alba, at a total acid number of 1.4, is the standard reference — has carried a 54 to 64 cent a barrel discount against Brent, and a 30 to 50 cent net advantage after corrosion mitigation. Set the 23.5 cent additive bill against that: the chemistry consumes 37 to 43 per cent of the discount. The rest is the refiner’s margin for running a difficult slate at all.

That single ratio is why this market is more exposed to a spread than to a demand forecast. Compress the heavy-light spread and the additive rate per barrel does not need to fall — but the reason to run the barrel does, and refiners respond by switching slate rather than by accepting a thinner margin. The chemistry follows the slate. This is also why the upstream side of the same chemistry is worth reading alongside it: the H2S scavengers market is dosed against the same defect in a different vessel, and moves on the same sulphur story.

The largest family is the slowest growing family

Sort the market by chemistry and a pattern appears that no supplier discloses but the arithmetic does.

Blend-pool fuel additives are the largest family at 23.0 per cent of the market and the slowest-growing at 1.40 per cent. The reason is mechanical: they are dosed per barrel of finished product, and finished product volumes are flat to falling as vehicle efficiency improves and the fleet electrifies — the dynamic covered in the global oil market and, from the demand-destruction side, in the new energy vehicle market.

Everything that reads a barrel’s defects rather than its volume grows more than three times faster. Demulsifiers and desalting lead at 4.68 per cent, because salt and water in the feed are the first and least negotiable problem a refinery has, and because the crude arriving at the gate is heavier and more contaminated each year. Corrosion inhibitors and amines follow at 4.48, H2S and mercaptan chemistry at 4.56, antifoulants at 4.26. FCC and catalyst additives sit in between at 2.89 per cent — a process-volume business that nonetheless levers the whole downstream chain, as the olefins market and polypropylene market both demonstrate from the other end.

Why the slate is getting heavier — and why that is not a straight line

Three forces are pushing the average barrel in the difficult direction at once.

OPEC+ has been unwinding production cuts since May 2025, and the barrels returning to the market are disproportionately medium and heavy sour. Canadian crude and condensate production is forecast at a record 4.85 million barrels a day in 2026, and two-thirds of Trans Mountain’s exports went to Asia-Pacific destinations in January to November 2025 — into Chinese refineries that are simultaneously adding cracking capacity for petrochemical yield, which is the most additive-hungry configuration a refinery can be built in.

The counterweight sits in the same base case. Arab Heavy shipments are down 280,000 barrels a day to roughly 560,000, and Venezuelan Merey has been constrained by a US export blockade. Heavy crude is being re-routed, not retired. And the scarcity that tightens physical supply of heavy sour also widens the spread that pays for the chemistry — which is the one mechanism in this market that is self-correcting in the additive supplier’s favour.

The regulatory ratchet does not just raise the dose, it changes what is being dosed

The marine fuel regime is the clearest example of a rule that redraws demand rather than simply increasing it. IMO’s 0.50 per cent global sulphur cap took effect in 2020 and cut sulphur oxide emissions by roughly 70 per cent. The Mediterranean sulphur emission control area followed at 0.10 per cent from 1 May 2025 — the fifth such area, covering a sea that carries around 20 per cent of seaborne trade — with the North-East Atlantic approved in April 2025 and the Canadian Arctic and Norwegian Sea designated in 2024.

Each ratchet raises the hydrogen and amine duty inside the refinery, which raises scavenger and corrosion chemistry demand, and it forces lower-quality feed into the FCC and hydrocracker, which raises catalyst additive demand. The effect lands on the defect-reading families, not on the volume-reading ones — the same split the chemistry table shows. It is also why the regulatory story connects this market to the wider chemicals complex rather than sitting inside refining alone: a specification change at the bunkering end becomes a catalyst purchase upstream.

The number to watch

Refinery additives is a market that grows while its underlying volume does not, and that is unusual enough to be misread. The instinct is to model it off refinery runs, or off crude prices, or off chemical industry growth, and all three produce the wrong answer. The market is a quality tax on crude, and its growth rate is the rate at which the world’s barrels are getting worse.

Which makes the single most informative variable neither throughput nor the additive price index, but the heavy-light spread. At 54 to 64 cents a barrel of discount, the 23.5 cent bill is a third of an incentive that clears. Below roughly 30 cents, the same chemistry is being bought out of margin rather than out of advantage, and the rational response is to stop buying the barrel.

Everything else in this report — the class split, the family rankings, the USD 3,939.2 million to USD 5,193.5 million path at 3.52 per cent — is downstream of that one number moving in the right direction. The full dataset, including the 2026 to 2033 year-by-year paths and the treatment of the connected oil, gas and energy complex, is in the report.