CBAM and the Petrochemical Value Chain: Product-Level CI for Chemical Exports to the EU

The European Union's Carbon Border Adjustment Mechanism entered its definitive phase in January 2026, and its scope is expanding. While the initial focus covered steel, aluminium, cement, fertilizers, hydrogen, and electricity, the regulatory trajectory points directly at the petrochemical sector. Organic chemicals, plastics, and ammonia-based fertilizer precursors are on the path to inclusion, and for the integrated refinery-petrochemical complexes that dominate global chemical production, the compliance challenge is orders of magnitude more complex than for single-product facilities.

This article examines what CBAM extension to petrochemicals means for producers, why traditional LCA approaches are inadequate for the product-level carbon intensity calculations required, and how continuous automated CI computation can address the allocation challenges inherent in multi-product chemical manufacturing.

CBAM and the Petrochemical Horizon

CBAM's core mechanism is straightforward: importers of covered goods into the EU must purchase CBAM certificates corresponding to the embedded emissions of those goods, at a price linked to the EU Emissions Trading System carbon price. The objective is to equalize the carbon cost between EU producers (who pay for emissions through the EU ETS) and non-EU producers (who may face lower or no carbon pricing in their jurisdictions).

The initial product scope was selected based on carbon intensity and trade exposure. Petrochemical products were not included in the first phase, but the European Commission has signaled their inclusion in the scope review process. Organic chemicals (HS Chapter 29), plastics in primary forms (HS Chapter 39), and certain fertilizer precursors are all candidates for the expanded scope, with legislative proposals expected by 2028 and coverage potentially beginning by 2030.

However, the smart producers are not waiting. EU importers are already requesting embedded emissions data from petrochemical suppliers as part of their Scope 3 reporting and procurement risk management. The Together for Sustainability (TfS) initiative, which counts most major chemical companies among its members, has published Product Carbon Footprint (PCF) guidelines that are becoming the de facto standard for chemical industry carbon accounting. Producers that cannot provide product-level CI data are losing tenders today, regardless of CBAM's formal timeline.

The Multi-Product Facility Challenge

Petrochemical production is fundamentally different from the industrial processes currently covered by CBAM. A steel mill produces steel. A cement plant produces clinker and cement. But a steam cracker produces ethylene, propylene, butadiene, benzene, toluene, xylenes, and a dozen other co-products and by-products simultaneously from the same feed. An integrated refinery-petrochemical complex, of the kind operated by major national oil companies and petrochemical producers across the Middle East, may produce hundreds of distinct products from shared utilities, shared feedstocks, and interconnected process units.

The allocation challenge is the central technical problem. When a single furnace consumes energy to crack naphtha into multiple products, how should those energy-related emissions be distributed? When a hydrogen plant supplies both the refinery's hydrocracker and the chemical plant's hydrogenation reactors, how should those Scope 1 emissions be assigned? When waste heat from one process unit provides steam to another, does the receiving unit get an emissions credit?

There are several established allocation methodologies, and they produce materially different results:

• Mass allocation distributes emissions proportionally by mass of each product. Simple but misleading when products have vastly different economic values and market functions.
• Economic allocation distributes emissions proportionally by revenue or market value. More reflective of economic reality but introduces volatility as commodity prices fluctuate.
• Energy allocation distributes emissions proportionally by the energy content of each product. Common in fuel-producing contexts but less meaningful for chemical intermediates.
• System expansion / substitution avoids allocation entirely by crediting co-products with the emissions they displace from alternative production routes. Technically rigorous but data-intensive and sensitive to assumptions about marginal production.

CBAM's implementing regulation follows the precedent established by the EU ETS free allocation methodology, which uses a combination of product benchmarks and fall-back approaches (heat, fuel, process emissions) for installations producing multiple products. For petrochemicals, the relevant benchmarks include those for ethylene (high-value chemicals), ammonia, and other specific products. However, the existing benchmarks do not cover the full range of petrochemical outputs, creating methodological gaps that producers must navigate.

Calculating Embedded Emissions: Ethylene, Polyethylene, and Ammonia

Consider the emissions calculation chain for polyethylene exported from a Middle Eastern integrated complex to the EU. The calculation must trace emissions through multiple conversion steps:

Ethylene production: The steam cracker consumes ethane or naphtha feed and significant quantities of fuel (typically methane or fuel gas). Emissions include combustion emissions from the cracker furnaces, electricity consumption for compression and separation, and upstream emissions from feedstock extraction and processing. The cracker also produces propylene, butadiene, and pyrolysis gasoline as co-products, requiring allocation across all outputs.

Polymerization: The polyethylene reactor consumes ethylene, electricity, catalysts, and additives. Emissions are more straightforward here since polyethylene is the primary product, but the calculation must include the embedded emissions of the ethylene feed (carrying forward the allocation decisions from the cracker stage).

For ammonia-based fertilizers, the calculation chain is equally complex. Ammonia production via steam methane reforming is one of the most carbon-intensive chemical processes, generating approximately 1.8 tonnes of CO2 per tonne of ammonia from the reforming reaction alone (where methane is converted to hydrogen and CO2), plus additional emissions from fuel combustion and electricity. When ammonia is further converted to urea, nitric acid, or ammonium nitrate, each conversion step adds its own emissions layer, and the final fertilizer product carries the cumulative embedded emissions of the entire chain.

ISCC PLUS and Circular Feedstock Certification

The emergence of circular and bio-based feedstocks adds another dimension to petrochemical carbon accounting. The International Sustainability and Carbon Certification (ISCC) PLUS scheme has become the dominant certification framework for tracking sustainable feedstocks through the petrochemical value chain.

ISCC PLUS enables mass balance accounting, which allows producers to blend certified sustainable feedstocks (such as pyrolysis oil from plastic waste, bio-naphtha from vegetable oils, or bio-ethanol) with conventional fossil feedstocks in the same processing equipment, and then allocate the sustainable attributes to specific outputs. This is critical for petrochemical facilities because dedicated processing of small volumes of alternative feedstocks is rarely economically viable; co-processing in existing crackers and reactors is the practical pathway.

However, mass balance accounting creates its own verification challenges. The producer must demonstrate an auditable chain of custody showing that the certified feedstock entered the system, that the mass balance bookkeeping correctly tracks credits and debits, and that the sustainability attributes are assigned to specific product batches without double-counting. When a cracker processes a blend containing 10 percent certified circular naphtha and 90 percent conventional naphtha, the mass balance system allows 10 percent of any output product to carry the circular certification. Tracking this accurately across continuous production runs, inventory movements, and multiple product grades requires data infrastructure that most manual systems cannot provide.

TfS and PCF Guidelines for the Chemical Industry

The Together for Sustainability initiative, founded by major chemical companies including BASF, Bayer, Evonik, Henkel, Lanxess, and Solvay, has developed Product Carbon Footprint guidelines specifically tailored to the complexities of chemical manufacturing. The TfS PCF Guideline provides standardized rules for system boundaries, allocation methods, and data quality requirements that enable comparability across producers.

Key provisions include the requirement to use operational control as the organizational boundary, the specification of allocation hierarchies (with a preference for physical relationships where possible), and detailed guidance on handling recycled content, bio-based feedstocks, and energy recovery. The TfS guidelines are aligned with ISO 14067 and GHG Protocol Product Standard but provide the chemical-industry-specific interpretive guidance that those broader standards lack.

For producers outside the TfS membership, these guidelines are still relevant. EU importers that are TfS members increasingly require their suppliers to provide PCF data calculated according to TfS methodology. Major Middle Eastern petrochemical exports to European chemical companies are subject to these data requests today, creating a commercial incentive for product-level CI capability that precedes any formal CBAM obligation.

Why Annual LCA Studies Are Insufficient

The traditional approach to product carbon footprinting in the chemical industry relies on periodic Life Cycle Assessment studies, typically conducted annually or biennially by external consultants. These studies establish a "representative" emissions profile for each product based on production data from a reference period, usually the previous calendar year.

This approach has three fundamental limitations that make it inadequate for the emerging regulatory and market requirements:

Temporal mismatch: Annual LCA studies produce static numbers that do not reflect the reality of continuous operations. A cracker's emissions profile changes when the feed composition shifts from ethane to naphtha, when a furnace is taken offline for maintenance, or when the electricity grid mix changes seasonally. An annual average obscures these variations, and a buyer receiving product in July has no way to know whether the annual CI number reflects the actual conditions under which that product was manufactured.

Allocation rigidity: Consultants make allocation decisions at the time of the study and apply them uniformly across the reference period. But the economically optimal allocation method for reporting one year may not be optimal the next, and the inability to dynamically adjust allocation as production conditions change creates both inaccuracy and missed optimization opportunities.

Verification lag: By the time an annual LCA study is completed, reviewed, and published, the data is typically 12 to 18 months old. For CBAM quarterly reporting, for real-time procurement decisions, and for ESG-linked financing covenants, this lag is unacceptable.

The Data Challenge: Cracker Feed, Energy Allocation, and Co-Products

At the operational level, continuous product-level CI computation for petrochemical facilities requires integrating data from multiple source systems that were never designed to interoperate for this purpose:

• Process control systems (DCS/SCADA): Feed rates, temperatures, pressures, and conversion data for each process unit, updated in real time.
• Laboratory information management systems (LIMS): Feed composition analysis, product quality data, and analytical results that determine actual yields and product specifications.
• Energy management systems: Steam generation and consumption by unit, electricity consumption by motor and process area, fuel gas composition and consumption rates.
• Material balance systems: Inventory movements, inter-unit transfers, and production accounting that track what enters and leaves each process boundary.
• Emissions monitoring systems (CEMS): Continuous emissions monitoring data from combustion stacks, flares, and vent systems.

Integrating these systems is the domain of Forward-Deployed Engineering. Each facility has its own unique configuration of vendors, data formats, historian systems, and naming conventions. There is no universal API for industrial carbon accounting. Instead, dedicated engineering teams must map the specific data landscape of each facility, build the integration connectors, and validate the data flows against known mass and energy balances before any CI calculation can be trusted.

How Denominator's Agents Handle Real-Time Allocation

Denominator's agent-native architecture addresses the petrochemical CI challenge through autonomous agents that operate continuously within the facility's data environment. The approach works as follows:

Data ingestion agents connect to each source system (DCS, LIMS, energy management, material balance, CEMS) through Forward-Deployed Engineering connectors. These agents normalize data formats, reconcile timestamps across systems, and flag data quality issues in real time. When a lab result conflicts with a process measurement, the agent raises an exception rather than silently propagating the error.

Allocation agents apply the configured allocation methodology continuously as production data flows in. When the cracker feed composition changes, the allocation is recalculated immediately. When a furnace trips and the remaining furnaces increase throughput, the energy allocation adjusts in real time. The agents maintain a complete audit trail of every allocation calculation, including the input data, the methodology applied, and the resulting product-level CI for every time period.

Certification agents assemble the evidence packages required for CBAM declarations, TfS PCF submissions, and ISCC PLUS mass balance audits. These packages include not just the final CI numbers but the complete provenance chain: the source data, the allocation methodology, the calculations, and the blockchain-anchored timestamps that prove when each data point was generated.

For large-scale integrated complexes, this architecture enables something that was previously impossible: a real-time view of the product-level carbon intensity of every petrochemical output, updated as production conditions change, with every calculation fully auditable and blockchain-anchored. When a European buyer requests the embedded emissions for a specific polyethylene shipment, the answer reflects the actual production conditions during the specific manufacturing period, not an annual average that may bear little resemblance to reality.

The petrochemical industry's carbon accounting challenge is not going to simplify. As CBAM scope expands, as TfS PCF requirements tighten, and as circular feedstock volumes grow, the complexity of product-level CI calculation will only increase. Producers that invest in continuous, automated, agent-native carbon intelligence infrastructure today are building the competitive foundation for a market where verified product-level emissions data is not optional; it is the price of admission.