The Acid Plants Are Not Broken. They Were Switched Off.

The way this news usually arrives is not a press release. It is somebody putting their head round the door to say they have heard that two plants are down and there is a sulfuric acid shortage. That is roughly what happened to us this week, and it is worth saying up front that the person was right about the direction and wrong about the scale. The number is not two. As of the first week of August it was four facilities, the reason has almost nothing to do with anything breaking, and the most interesting part of the story is a detail about catalyst chemistry that explains why the recovery will not be as quick as the shutdown was.

We sell sulfuric acid in packaged quantities — quarts through drums and totes — in seven strengths, so we spend a lot of time explaining which market a given drum actually comes from. That turns out to be the whole point of this story. There is a real contraction going on in American acid capacity right now, and there is a great deal of loose talk about a “sulfuric acid shortage” that rolls every concentration and every grade into a single commodity moving through a single pipe. It is not one. What follows is what is genuinely down, why a fertilizer company’s margin decision removes acid capacity from the Gulf Coast, the reason an idled acid plant is a harder thing to restart than an idled boiler, and what a buyer of packaged acid should and should not conclude from any of it.

What is actually down

On 11 May 2026 Mosaic said it would cut phosphate fertilizer production by half at Faustina, Louisiana and Bartow, Florida. Bruce Bodine, the company’s president and chief executive, gave the reason plainly: “Raw material prices and availability, especially for sulfur, are forcing us to revisit our production plan.” He framed it as a waiting game rather than a retreat — “a temporary move that allows us to limit the need for incremental sulfur at today’s prices and wait until the market normalizes.”

By the second-quarter earnings call on 5 August the cuts had gone deeper than half. Louisiana was described as completely offline. Bartow was running at about 40% of capacity. New Wales and Riverview, the two large Florida plants, were in the low-to-mid 70s. Two days later, on 7 August, the company said it would curtail further, and the list grew to four sites: Bartow and Faustina, already reduced, plus Riverview and Uncle Sam. In Brazil, additional temporary curtailments and idling were under way, following an April idling of single-superphosphate production and a May halt to phosphate rock output.

It is worth being precise about what “down” means here, because it is not the usual industrial sense of the word. Nothing exploded. There was no fire, no force majeure event at the gate, no mechanical failure of a converter or an absorption tower. These are deliberate, reversible, economically motivated reductions in run rate, and the company has been consistent in describing them as temporary. That is genuinely better news than an accident would be. It is also, for reasons that come down to the physical chemistry of the catalyst bed, not quite as reversible as the word “temporary” makes it sound.

Why a fertilizer decision is an acid decision

Sulfuric acid and phosphate fertilizer are joined at the hip, and the joint is the wet process. Phosphate rock is mostly fluorapatite, and to get the phosphorus out of it in a usable form you attack the rock with sulfuric acid. The reaction produces phosphoric acid and a very large quantity of calcium sulfate, which is gypsum, along with hydrogen fluoride that has to be scrubbed:

The stoichiometry alone says five moles of sulfuric acid for every three of phosphoric, but the stoichiometry is the optimistic case. Real phosphate rock carries carbonate and other impurities that consume acid without producing anything useful, so the practical figure is higher. Published process data puts it at roughly 1,360 kg of sulfuric acid per tonne of phosphoric acid, and the working range for acid consumption is around two to three tonnes of sulfuric acid per tonne of P₂O₅ depending on the rock. Whichever number you use, the conclusion is the same: phosphate is not a side consumer of sulfuric acid. It is the main one. Roughly 56% of the world’s sulfuric acid goes into fertilizer.

That is why the demand and supply sides of this market are so awkwardly coupled. A phosphate producer that cuts output does two things at once. It stops consuming acid, which loosens the market. And, if it makes its own acid on site — which the Louisiana sites do, on a large scale — it stops producing acid, which tightens it. Which effect dominates depends on the region, the freight economics and whether those trains were selling any merchant tonnes in the first place. Anyone who tells you confidently that a phosphate curtailment is straightforwardly bullish or bearish for the acid you buy in drums is guessing.

The upstream cause of all of it is sulfur, and sulfur has been the story of 2026 since the Strait of Hormuz disruption took a large share of seaborne supply out of easy reach and export restrictions removed most of the flexible valves from the market. We wrote that up at length when it broke; if you want the feedstock side of the picture, our 2026 sulfuric acid supply crisis explainer covers where the world’s sulfur comes from and why so little of it is a free agent. The short version is that sulfur is the dominant cash cost in burning acid — estimates run from about 70% up to 80% of production cost — so an acid plant’s margin tracks the sulfur price almost one for one.

The numbers on the board explain the decision without much need for interpretation. Qatar’s sulfur contract has been settling around $805 a tonne FOB. Mosaic told investors it had locked in a significant portion of its third-quarter molten sulfur supply at $705 a tonne, which it described as considerably below spot — a useful reminder that the company is not a distressed buyer, it is a disciplined one. Against that, diammonium phosphate was assessed at about $919 a tonne and monoammonium phosphate at about $959 in early September, with DAP up around 7% year on year. Fertilizer prices are high. They are not high enough, fast enough, to make incremental tonnes of sulfur at spot worth converting, and Bodine’s summary of the global consequence was blunt: “We believe global phosphate production will fall well short of last year by up to 30 million tons.”

An idled acid plant is not a paused acid plant

Here is the part that almost never makes the market commentary, and it is the reason we would not assume a fast snap-back even if sulfur eased tomorrow.

Sulfuric acid is not extracted. It is burned into existence, in three steps. Molten sulfur is sprayed into a furnace and burns to sulfur dioxide. The dioxide is then oxidised to sulfur trioxide over a catalyst, which is the hard step and the one the whole plant is built around. The trioxide is then absorbed into strong acid:

Most people picture that catalyst as a solid: a hard ceramic pellet with vanadium in it, sitting in a bed, doing its work. That is only half right, and the other half is what makes shutdowns expensive. Under operating conditions the active vanadium phase is not solid at all. It is a molten salt — alkali pyrosulfates that dissolve the vanadium species — spread as a very thin liquid film across the surface of a porous diatomaceous-earth or silica support. The pellet is the sponge. The catalyst is the liquid in it.

A liquid catalyst has a freezing point, and this one is inconveniently high. Below roughly 400 °C the melt is no longer molten and activity collapses; reported rates fall away sharply somewhere in the 370–400 °C region depending on the formulation and the gas. In the converter itself the active phase sits as a melt across a band of about 450–610 °C, and operators generally run around 430–450 °C as the compromise between kinetics and equilibrium, which still returns 96–98% conversion per pass. An acid plant, in other words, is a machine whose working fluid must be kept above its freezing point at all times. Cold is not a neutral state for it. Cold is a phase change.

What repeated phase changes do to the bed was set out earlier this year in Sulfuric Acid Today by Aida Barberena and Jeremy Schneider of Elessent Clean Technologies, whose MECS catalysts are among the most widely used in the industry. Their argument is that the industry persistently underestimates thermal cycling, and they identify three mechanisms that compound one another.

Two details in that article deserve to be quoted rather than paraphrased. The first is a field case. A plant that went through nearly ten shutdowns in a single year — not because of anything wrong inside the acid plant, but because upstream upsets kept interrupting the hot process gas — saw the cumulative effect degrade the bed into what the authors describe as “essentially solidified dust,” forcing a premature replacement of the entire catalyst charge. The second is the sentence that ought to be pinned above every operator’s desk this year, about the temptation to rush a heat-up or admit wet gas below the acid dew point to save time: those choices “might save hours but could cost months in catalyst life.”

None of this contradicts Mosaic’s position. The company has said it is prepared to ramp back to full rates in weeks rather than months once conditions allow, and for the fertilizer trains that is entirely plausible. But it is worth holding the two clocks separately. The commercial clock is set by the sulfur price and can turn in a week. The metallurgical clock is set by how many times a bed has been taken through a cooling cycle, and it only runs one way. A plant that cycles repeatedly through a long, uncertain curtailment is accumulating a maintenance liability that does not show up in a quarterly operating rate, and the industry’s own catalyst suppliers have been unusually direct about saying so.

What this does and does not mean for your drum

Now the part that matters if you are not a phosphate producer. The honest answer is that a fertilizer-chain contraction does not price every sulfuric acid the same way, and the single most common mistake we see during a supply scare is a buyer reacting to a headline about a market they are not actually in.

Sulfuric acid is sold by concentration, and concentration is not a cosmetic difference. It changes the freezing point, the density, the corrosion behaviour toward specific metals and plastics, and what the acid is physically able to do. It also, critically, changes which production route the material came from. Bulk 93% acid off a sulphur-burning train is the grade that shares a supply chain with fertilizer production. High-purity reagent material and purified electrolyte acid are produced and finished to different specifications for different customers, and their availability and pricing do not move in lockstep with a DAP margin in Florida.

There is a related trap in the arithmetic. Substituting a different concentration to chase availability is not a like-for-like swap, because you are buying water as well as acid. A drum of 93% and a drum of 70% do not contain the same quantity of H₂SO₄, and re-basing a process on the wrong assumption is a more expensive error than a few weeks of lead time. If you are ever unsure what the equivalent volume is, our complete sulfuric acid concentration guide has the comparison laid out, and anyone diluting should remember the direction of the operation: acid into water, never the reverse, because the heat of dilution is substantial.

The practical advice we would give a packaged-acid buyer in September 2026 is unglamorous. Know the concentration and grade your process is actually specified on, and put it on the purchase order rather than relying on a generic description. Ask for the lot-specific Certificate of Analysis if your quality system needs the documented numbers rather than a nominal value. Talk to us about lead times before you are down to your last drum, because visibility is cheap and expediting is not. And treat a headline about acid plants in Louisiana as information about the commodity end of the market, which it is, rather than as a reason to panic about a laboratory reagent, which it is not.

What we are watching next

If you want to follow this without a subscription to a price service, there are a handful of indicators that will move before any announcement does.

1. The sulfur contract settlement. Qatar’s quarterly number around $805 a tonne FOB is the reference everyone else argues with. It is the input that has to fall before curtailed capacity has a reason to come back.
2. The DAP and MAP spread against it. Roughly $919 and $959 a tonne in early September. Curtailment is a margin decision, so watch the gap between fertilizer prices and sulfur cost rather than either number alone.
3. Gulf terminal acid inventories. Price reporters have been describing tightening inventories at Gulf terminals. That is the closest public proxy for merchant availability in the region that matters most to North American buyers.
4. Mosaic’s next quarterly call. The company has been explicit and specific about operating rates by site, which makes its calls the single most useful public document on American phosphate acid capacity.
5. Restart language, not restart dates. When capacity does come back, watch whether the announcements mention catalyst screening or replacement. That is the tell for whether the cycling bill has come due.

Common questions

Is there actually a sulfuric acid shortage right now?

There is a genuine contraction in American phosphate-linked acid capacity, with four Mosaic facilities curtailed or idled as of 7 August 2026, and price reporters describe tightening inventories at Gulf terminals. But “shortage” is too blunt a word for what is happening. The constraint is the cost of sulfur rather than an absence of capacity, and it affects commodity bulk acid far more directly than it affects packaged acid bought against a specification.

Which Mosaic plants are down, and by how much?

As of the second-quarter call on 5 August 2026, Louisiana was described as completely offline and Bartow, Florida as running at about 40% of capacity, with New Wales and Riverview in the low-to-mid 70s. On 7 August the company said it would curtail further, naming four sites: Bartow, Faustina, Riverview and Uncle Sam. Additional curtailments and idling were also under way in Brazil.

Why would a company switch off a plant instead of selling the acid?

Because sulfur is the dominant cash cost in making sulfuric acid — commonly estimated at 70–80% of production cost — so when sulfur is expensive enough, converting it destroys value rather than creating it. Mosaic’s chief executive described the cuts as a temporary step to limit the need for incremental sulfur at current prices while waiting for the market to normalise.

Why can’t an idled acid plant just be switched back on?

It can, but not for free. The active phase of the catalyst is a molten alkali-pyrosulfate salt holding vanadium, spread as a thin film on a porous support, and below roughly 400 °C that melt solidifies and conversion collapses. Every shutdown and restart is a thermal cycle that microcracks the support, shifts vanadium toward less reactive species, and risks leaching active salts off the pellet if the gas crosses the acid dew point. In one documented case, nearly ten shutdowns in a year degraded a bed into what the catalyst supplier called essentially solidified dust.

Does this affect the concentration I buy?

Not uniformly. Bulk 93% acid off sulphur-burning capacity is the grade most exposed to sulfur economics. Analytical ACS material, purified electrolyte acid and pre-mixed dilute battery acid are produced and sold against specifications, and their availability does not track a fertilizer margin. The correct response is to confirm your specification and lead time rather than to substitute a different concentration.

Should I switch to a different strength if my usual grade is tight?

Only with the arithmetic done first. Different concentrations contain different quantities of H₂SO₄ per unit volume, so a substitution changes your process chemistry and your dosing, and a stronger acid is not a drop-in replacement for a weaker one. Where a specification is involved, a nominally similar concentration from a different production route may not meet it at all. Talk it through before you re-base anything.

The three we would put on the bench

Sulfuric Acid 93% Technical Grade

The industrial workhorse concentration, and the one closest to the commodity story above.

Sulfuric Acid 96% ACS Grade

Bought against a published specification. A different chain from bulk commodity acid.

Sulfuric Acid 50% Electrolyte Grade

For battery and flow-battery work, where purity is the specification that matters.