The Acrylonitrile Capacity Map: Who Added Plants and Who Closed Them documents a structural shift in the propylene ammoxidation value chain, in which integrated Chinese petrochemical parks absorbed most net nameplate additions while older merchant units in Europe and Northeast Asia were retired. Acrylonitrile (C3H3N, CAS 107-13-1, molar mass 53.06 g/mol, boiling point 77.3 °C) is produced from propylene, ammonia, and oxygen over bismuth molybdate-based catalysts in fluidized-bed reactors maintained at 400–450 °C and 0.5–1.5 atm. The reaction yields acrylonitrile plus acetonitrile, hydrogen cyanide, and unconverted ammonia; recovery involves water quench, absorption, and distillation. Propylene purchase prices represent an estimated 55–65% of variable production cost, so unit competitiveness is tied directly to C3 feedstock integration.

Global ACN nameplate capacity is reported in the 7.4–7.8 million t/y range, with China accounting for approximately 3.2 million t/y after the 2020–2021 start-ups. The United States, Japan, South Korea, and Western Europe retain merchant positions, but operating utilization has diverged sharply. High-cost naphtha-based propylene in Japan and Europe has placed older units at risk of idling or permanent closure, while integrated propane dehydrogenation and refinery-grade propylene supplies in China and the US Gulf have supported new capacity and selective debottlenecks.

Acrylonitrile Capacity Map: Who Added Plants and Who Closed Them in the 2018–2024 Window

Chinese additions dominate the reported capacity map. Jiangsu Sailboat Petrochemical placed a 260 kt/y acrylonitrile unit at Lianyungang during the 2019–2020 period; the line consumes polymer-grade propylene from adjacent propane dehydrogenation and merchant sources, although exact internal propylene allocation is not disclosed in all commissioning statements. Zhejiang Petroleum & Chemical started a 260 kt/y ACN unit at Zhoushan in 2021 within a naphtha steam cracker and propane dehydrogenation complex, reducing external C3 exposure. Additional Chinese capacity was reported at Shandong Chambroad and Sinopec operations, lifting China’s ACN nameplate from roughly 2.5 million t/y in 2018 to above 3.2 million t/y by 2022.

Reported removals are concentrated in the United Kingdom and Japan. INEOS Nitriles announced the closure of its Seal Sands, UK, unit with reported capacity near 280 kt/y, citing high feedstock and energy costs. Asahi Kasei disclosed withdrawal of its Kawasaki acrylonitrile line, reported at 300 kt/y, by 2024. Mitsubishi Chemical reported removal of approximately 200 kt/y of domestic acrylonitrile capacity, with closure completed by 2024. The combined reported removals from these three sites approach 780 kt/y, roughly equal to three modern-scale Chinese units. These figures are nameplate references rather than precise operating data; some closure dates were phased across maintenance cycles.

South Korean capacity remained comparatively stable in this period, with Ulsan- and Yeosu-based ACN units reported in the 100–250 kt/y range per line. US Gulf ACN capacity was not removed permanently in the same manner, but merchant production at sites such as Cornerstone Chemical’s Fortier plant in Louisiana has experienced periodic curtailments during propylene supply tightness and turnarounds. The capacity map therefore shows a regional divergence: China reported gross additions exceeding 1.0 million t/y over the period, Japan and Europe removed roughly 780 kt/y, and the Americas remained approximately flat on a net nameplate basis.

Regional nameplate capacity is concentrated in Asia, with China alone at roughly 3.2 million t/y, Japan and South Korea together near 1.0–1.5 million t/y, and the Americas around 1.0–1.4 million t/y. Western Europe has contracted to roughly 0.4–0.5 million t/y after the Seal Sands closure. These ranges are derived from reported producer capacities and trade-press tabulations; they are not exact operating rates.

When Feedstock Rationalization Forces a Permanent Capacity Write-Down

Commercial propylene ammoxidation consumes 1.08–1.15 t of propylene and 0.42–0.50 t of ammonia per tonne of acrylonitrile. A single-pass propylene conversion above 95% and an overall acrylonitrile yield of 78–84% are typical for Sohio fluidized-bed reactors. At those consumption ratios, a 200 kt/y unit requires roughly 216–230 kt/y of propylene. A naphtha-based steam cracker that produces propylene as a co-product cannot always supply that volume without competing with polypropylene demand. When naphtha-linked propylene prices exceed the regional ACN break-even threshold, the unit enters negative operating margin; if the producer owns multiple aging lines and faces an upcoming catalyst change or re-tubing, the cost of sustaining the asset may exceed the discounted cash flow from continued operation.

Kinetic parameters vary with catalyst formulation, but commercial bismuth molybdate-based catalysts typically operate at propylene conversion above 95% and carbon selectivity to acrylonitrile of 80–85%; the remaining selectivity is distributed among acetonitrile, hydrogen cyanide, carbon monoxide, and carbon dioxide. Catalyst particles in the fluidized bed undergo attrition; fines are continuously removed by cyclones and replaced. Reactor temperature is controlled by steam generation pressure because the ammoxidation reaction is exothermic. Oxygen-to-ammonia ratios are maintained below the flammable envelope; excess oxygen increases carbon monoxide and carbon dioxide formation and reduces selectivity.

The fluidized-bed reactor in a modern ACN plant includes cyclones for catalyst recovery, vertical heat-transfer coils generating high-pressure steam, and ammonia injection below the distributor. The hot reactor effluent is cooled in a quench column and routed to an absorber where water removes acrylonitrile and acetonitrile. Recovery distillation separates acetonitrile, hydrogen cyanide, water, and heavy ends; a product column yields acrylonitrile at 99.5 wt% or higher with water below 0.5 wt%. The purified product is stabilized with monomethyl ether of hydroquinone before storage.

The absorber uses cold water to collect acrylonitrile and acetonitrile; the resulting solution is distilled in a recovery column with a side draw of hydrogen cyanide. The overhead acetonitrile stream is further purified, while the bottoms water is steam-stripped and recycled. High-purity acrylonitrile is condensed at reduced pressure to avoid thermal polymerization; inhibitor injection is maintained throughout the distillation train. The product column reboiler is a key fouling point because acrylonitrile can form low-molecular-weight polymer if inhibitor is not uniformly distributed.

The cuts at Seal Sands, Kawasaki, and Mitsubishi Chemical’s Japanese unit reflect this economic threshold rather than a disappearance of downstream demand. A unit of 280 kt/y or 300 kt/y capacity cannot be restarted without substantial fixed-cost absorption, and permanent closure eliminates future turnaround expense, catalyst inventory, ammonia storage upgrades, and emissions-control compliance. In Europe, EU Emissions Trading System allowance prices and electricity costs have increased the energy burden on acrylonitrile purification and refrigeration; in Japan, naphtha-linked propylene and power costs have produced similar pressure. Published unit-level cost data for these specific closures is limited, but trade-press statements consistently identify propylene feedstock and energy as the principal variables.

Ammonia cost contributes the second-largest variable cost after propylene. A plant consuming 0.42–0.50 t/t of ammonia is sensitive to natural-gas-linked ammonia prices. Units in regions with cheap natural gas, such as the US Gulf, gain a secondary advantage. Japanese and European units often pay higher ammonia and energy costs, compounding the propylene disadvantage.

The US ethane advantage does not automatically produce a propylene advantage for ACN. Ethane cracking yields very little propylene, so ACN producers in the US Gulf often rely on refinery propylene, propane dehydrogenation, or imported C3. When US refinery utilization falls during gasoline demand shocks or hurricane-related outages, propylene supply tightens and ACN operating rates can fall without any change in acrylonitrile demand. This structural decoupling is one reason US ACN nameplate capacity has remained flat while new Chinese capacity directly integrates with propane dehydrogenation and refinery C3 streams.

By-product credits from acetonitrile and hydrogen cyanide can offset part of fuel and catalyst cost, but their contribution is secondary compared with propylene. Typical estimates place by-product credit in the range of 5–15 USD/t of acrylonitrile, varying with acetonitrile pricing and downstream cyanide demand. That credit does not rescue a unit facing propylene cost differentials of 150–250 USD/t between integrated propane dehydrogenation propylene and merchant naphtha-grade propylene during high-price periods.

Closure of an ACN plant involves more than removing the reactor. Residual acrylonitrile and acetonitrile must be drained, inerted, and disposed; wastewater units must be decommissioned; and site environmental permits require monitoring under local regulations. These closure liabilities are part of the write-down calculation. The decision to close Seal Sands or Kawasaki rather than to restart after a turnaround reflects the high fixed cost of environmental upkeep and process safety recertification, not only current operating margin.

What Does the Present Net Capacity Map Mean for Downstream ABS and Fiber Buyers?

Downstream consumption of acrylonitrile is concentrated in ABS and SAN resins at an estimated 38–42%, acrylic fiber at 26–30%, and nitrile rubber at 3–5%, with the remainder moving into acrylamide, carbon fiber precursor, and specialty copolymers. ABS producers require acrylonitrile contents of 20–30 wt% in the SAN matrix; downstream resins may be tested by ISO 1133-1:2022 for melt flow rate and ASTM D256 for Izod impact, though those are resin specifications and not direct ACN specifications. PAN-based carbon fiber precursor requires ACN purity above 99.5% and controlled inhibitor content, because polar impurities can disrupt polymerization kinetics and influence subsequent oxidative stabilization at 200–300 °C.

Acrylonitrile monomer used in polyacrylonitrile spinning must meet low water and low inhibitor-variation limits. Polymerization for PAN fiber precursor is typically conducted in solution or aqueous suspension with redox initiators; monomer purity above 99.5% prevents chain-transfer reactions that reduce molecular weight and final fiber tensile properties. Published data for this specific configuration is limited, but the practical specification window for water and inhibitor content is narrow because both parameters affect polymer viscosity and spinning stability.

For logistics and handling, acrylonitrile is classified as a flammable liquid and toxic substance under UN 1093, Class 3, Packing Group I. The US OSHA permissible exposure limit is 2 ppm as an 8-hour TWA under 29 CFR 1910.1000 Table Z-1; storage and transfer fall under NFPA 30 flammable-liquid code. Commercial acrylonitrile is typically inhibited with monomethyl ether of hydroquinone at 35–45 ppm to prevent free-radical polymerization. If inhibitor levels deplete during long transit or heated storage, exothermic polymerization can occur; receivers should verify inhibitor concentration before tank acceptance and avoid oxygen-containing vapor space conditions that promote peroxide formation.

The net capacity map indicates that downstream buyers are not uniformly exposed to the same supply risk. A buyer sourcing from a Chinese integrated ACN site may contract with a producer that also consumes part of the ACN internally for ABS, reducing merchant availability during high-load periods. A buyer relying on US Gulf or South Korean merchant ACN may face short-term rationing during propylene supply disruptions, but also benefits from multiple independent supply points. The closures in Japan and the United Kingdom remove regional baseload supply in those markets, increasing import dependency for local ABS and acrylic fiber operations that did not backward-integrate.

Published data for this specific capacity configuration is limited because some operators report nameplate capacity only in annual securities filings and do not disclose utilization, internal transfer volumes, or by-product credit assumptions. The capacity map shows that gross additions exceeded removals in Asia, while removals were concentrated in Europe and Japan; the net global balance is less important than regional distribution. Acrylonitrile remains a fungible but hazardous monomer, and the practical supply available to any downstream site is determined by storage stability, inhibitor status, and actual operating rate rather than nameplate figure alone.