Acrylonitrile (CAS 107-13-1, EC 203-466-5, CN code 2926 10 00) is produced commercially by propylene ammoxidation in a fluidized-bed reactor. The molecule is a C3 nitrile with molar mass 53.06 g/mol and a carbon mass fraction of 0.679. EU buyers now request acrylonitrile product carbon footprint data under CBAM-related supply-chain questionnaires even though the monomer itself is not listed in Annex I to Regulation (EU) 2023/956. Acrylonitrile Product Carbon Footprint: What EU Buyers Now Ask Under CBAM now includes facility-specific direct emissions, purchased-energy emissions, and upstream feedstock emissions normalized to one tonne of technical-grade acrylonitrile. The request is no longer satisfied by regional life-cycle averages; it requires a defined system boundary, an explicit allocation method, and an assurance path consistent with product carbon footprint standards.

Where Direct Emissions Concentrate in the SOHIO Ammoxidation Train

Direct emissions from acrylonitrile production arise primarily from carbon dioxide formed by total oxidation of propylene inside the reactor and from thermal oxidation of off-gas and vent streams. The predominant chemistry is 2 C3H6 + 2 NH3 + 3 O2 → 2 C3H3N + 6 H2O. Stoichiometric feed ratios per tonne of acrylonitrile are 0.793 t propylene and 0.321 t ammonia, assuming complete conversion to the desired nitrile. Industrial operation does not achieve full selectivity; bismuth-molybdenum-based catalysts typically yield 80–84% acrylonitrile with co-production of hydrogen cyanide, acetonitrile, and carbon oxides. Carbon not retained in the product leaves the reactor as CO and CO2, with off-gas incineration converting residual organics and ammonia to CO2, H2O, and NOx. The thermal oxidizer operates at typical oxidation temperatures of 850–1,100 °C for waste-gas treatment. At sites with heat recovery, the thermal oxidizer may raise steam for distillation; the direct CO2 emission remains, but the purchased steam demand falls. Temperature excursions above the catalyst’s optimal band increase total oxidation to CO2 and reduce carbon selectivity per tonne of acrylonitrile. Published data for this specific configuration is limited, but process mass-balance records show that carbon selectivity is a stronger driver of direct emissions than absolute throughput.

After the reactor effluent passes through cyclones and quench, acrylonitrile is recovered by water absorption and purified by extractive distillation. The absorber pressure and product column reflux ratio are the main energy levers. Higher absorber pressure improves recovery but increases compressor power; lower reflux reduces steam consumption but can compromise acrylonitrile purity below technical specifications. Purchased electricity for compressors and refrigeration, plus steam for stripping and distillation, forms the indirect emissions portion of the product carbon footprint. Steam is often generated from on-site natural gas boilers or imported from an adjacent cogeneration unit; the emission factor assigned to steam must reflect actual boiler efficiency and fuel mix, not a generic grid factor. For absorbers using chilled water, electricity demand rises with ambient temperature and cooling-tower fouling. These utility inputs are frequently the largest site-to-site variable after feedstock carbon intensity.

When EU Buyers Cite CBAM Without an Annex I Listing for the Monomer

Acrylonitrile, CN code 2926 10 00, is not included in Annex I to Regulation (EU) 2023/956. The first-phase CBAM goods are limited to cement, electricity, fertilisers, iron and steel, aluminium, and hydrogen. Consequently, no Commission default value exists for acrylonitrile embedded emissions under Implementing Regulation (EU) 2023/1773, and importers cannot report acrylonitrile as a standalone CBAM good. EU buyers ask for acrylonitrile product carbon footprint data because Article 30 of Regulation (EU) 2023/956 provides a review mechanism for extending the scope to additional goods, and because downstream products containing acrylonitrile-derived polymers, fibres, or resins may be drawn into purchasing agreements that require documented carbon data. Requests typically specify a cradle-to-gate boundary, actual facility data for direct and purchased-energy emissions, and supplier data for ammonia and propylene. The absence of an Annex I listing does not remove the buyer’s leverage; incomplete data may be scored as a procurement risk in technical pre-qualification. A defensible response therefore states the regulatory status of the monomer, provides the product carbon footprint according to ISO 14067:2018, and separates data that are ready for CBAM-style reporting from data that are only life-cycle estimates.

Product carbon footprint calculations for acrylonitrile are usually structured as cradle-to-gate studies under ISO 14067:2018, using ISO 14044:2006 section 4.2.4 for system boundary definition. The functional unit should be 1 tonne of acrylonitrile at commercial purity, usually 99.0% minimum by mass. The system boundary includes propylene production, ammonia production, catalyst and chemical inputs, on-site utilities, waste-gas incineration, wastewater treatment, and internal transport up to the site gate. Downstream conversion, use, and end-of-life are excluded unless a full life-cycle environmental product declaration is requested. Cut-off criteria should be justified under ISO 14044:2006; a 1% mass and energy cut-off is common but may exclude significant trace emissions if not applied carefully. Biogenic carbon is typically not material in fossil-based acrylonitrile, but any bio-based propylene or blue or green ammonia input must be documented separately. The global warming impact is reported in kg CO2 eq per tonne acrylonitrile using IPCC GWP factors over a 100-year time horizon.

Feedstock Ammonia, Propylene, and Allocation Pressure Points

Feedstock upstream emissions usually dominate the cradle-to-gate carbon footprint of acrylonitrile. Propylene from naphtha steam cracking or refinery FCC carries a carbon burden that depends on cracker feedstock, allocation method for co-products, and refinery flare intensity. Ammonia from natural gas-based Haber-Bosch production emits roughly 1.8–2.2 t CO2 eq per tonne NH3 in published European LCA datasets. With stoichiometric ammonia use of 0.321 t/t acrylonitrile, the ammonia footprint contribution is approximately 0.58–0.71 t CO2 eq per tonne acrylonitrile before allocation for by-products. Propylene use is stoichiometrically 0.793 t/t; cradle-to-gate propylene values typically fall between 1.2 and 1.8 t CO2 eq per tonne propylene depending on upstream allocation. The resulting feedstock contribution alone can reach 1.5–2.1 t CO2 eq per tonne acrylonitrile, leaving the remainder to direct emissions and purchased utilities. Published cradle-to-gate results for European grey production commonly fall between 2.1 and 3.5 kg CO2 eq/kg; producers using green ammonia or low-carbon propylene will fall below this range, while sites with high purge losses and no heat recovery may exceed it.

By-product allocation changes the final value. Hydrogen cyanide and acetonitrile are generated in the ammoxidation reactor. If hydrogen cyanide is recovered and sold for methionine or other derivatives, the system must either be expanded to substitute production from the Andrussow process or allocate emissions by mass or economic value. ISO 14044:2006 section 4.3.4.2 provides the allocation hierarchy. System expansion tends to lower the acrylonitrile footprint because the avoided HCN production carries its own carbon burden. Allocation by mass produces a smaller credit because the HCN mass flow is low relative to acrylonitrile. EU buyers increasingly ask which allocation method was applied and whether the HCN stream is sold or incinerated.

Buyer data requests are often triggered by CBAM because the transitional reporting phase under Implementing Regulation (EU) 2023/1773 has made embedded carbon a customs-adjacent data field, even when acrylonitrile is outside the legal annex. A quarterly CBAM report for an importer of CBAM goods does not include acrylonitrile directly; however, procurement platforms now use the same template fields for scope-extension risk assessment. EU buyers ask for a product carbon footprint study that is accompanied by an assurance statement under ISO 14064-3:2019 or by a critical review opinion under ISO 14044:2006. The assurance should cover boundary, data quality, allocation choices, and recalculation policy. For site data, buyers may request the monitoring method used for natural gas and electricity meters, the source of steam emission factors, and the treatment of missing data. A public environmental product declaration should not be submitted as a CBAM report unless the underlying dataset is available; EPDs are often aggregated or use sector-average background data.

Fluidized-Bed Reactor Emissions Are Not Allocated by Unit Operations Alone

Because the fluidized-bed reactor produces acrylonitrile, hydrogen cyanide, acetonitrile, and carbon oxides simultaneously, any attempt to allocate direct emissions only to the main product violates ISO 14044:2006 section 4.3.4. The site must first define the co-products and then apply either system expansion or allocation. In many operating units, HCN is routed to a hydrogen cyanide purification system or absorbed as sodium cyanide; acetonitrile may be recovered by a heads column or incinerated. If acetonitrile is not sold and is incinerated, its carbon burden remains inside the acrylonitrile system boundary. If HCN is sold, the avoided product burden depends on the regional production route for HCN; the Andrussow process uses methane, ammonia, and air and has a distinct methane consumption per tonne of HCN. Assigning a zero-emission credit to HCN would not be consistent with either ISO 14044:2006 or the EU product environmental footprint guidance. Site mass balances should report monthly production of acrylonitrile, HCN, acetonitrile, and CO2 streams, and the product carbon footprint should state the allocation factor applied.

Published data for the propane-based ammoxidation route are limited, and facility-level variation is material. Producers should not apply propylene-based default values to propane-based processes without re-evaluating the feedstock-specific yield and reaction exotherm. A product carbon footprint based on green electricity alone does not offset direct process CO2; the two categories must remain separate for CBAM-style reporting. Where actual supplier data are unavailable, the PCF must disclose the use of secondary database values and the representativeness of the dataset. Allocation choices, steam system boundaries, and waste-gas heat recovery should be recalculated when the functional unit or production rate changes by more than 5%.