ecoinvent v3.13 Officially Released

On 4 November 2026, ecoinvent officially released version 3.13, further expanding the industry coverage and geographical representation of its database.

In addition to the introduction of a new pharmaceutical sector, the latest version includes updates to data across chemicals and plastics, fuels, metals, electricity, textiles, agriculture, pulp and paper, waste management and recycling, construction, and batteries. Several Life Cycle Impact Assessment (LCIA) methods have also been revised.

01. Pharmaceuticals: Regional Production Data Added for 61 Substances

Pharmaceuticals is a newly introduced sector in ecoinvent v3.13. The relevant datasets were developed in collaboration with the Pharma LCA Consortium and Boehringer Ingelheim, supporting the pharmaceutical industry's efforts to harmonise Life Cycle Assessment (LCA) methodologies and align with PAS 2090, the first Product Category Rule (PCR) for pharmaceuticals.

The new datasets cover the production of 61 substances across Europe (RER), North America (RNA), and Asia (RAS). They include industry-specific materials such as amino acids, pharmaceutical excipients, cell culture media components, and pharmaceutical-grade water, as well as average pharmaceutical waste treatment and solvent recovery processes.

02. Agriculture, Fisheries and Animal Husbandry: Correction of Biogenic Carbon Allocation and Updates to Carbon Balances

Updates in this sector primarily focus on data quality corrections.

Several grain datasets covering different regions of Australia contained errors in the allocation of biogenic carbon. The relevant values had been assigned to the "Carbon dioxide, land transformation" flow instead of the correct "Carbon dioxide, non-fossil" flow, resulting in carbon imbalances. In v3.13, the elementary flow exchanges for "Carbon dioxide, non-fossil" have been updated to account for emissions from residue burning, and the descriptions of the related flows have been harmonised.

One remaining carbon balance issue in the Brazilian perennial crop land-use change datasets has not yet been resolved. During the correction in v3.12, "Carbon dioxide, to soil or biomass stock" was introduced as a balancing flow to represent carbon uptake. However, the original "Carbon dioxide, in air" flow, which should have been replaced, was not removed, resulting in double counting of carbon uptake.

03. Batteries: 24 New Battery Module and Pack Datasets

The latest version introduces 24 new battery module and battery pack datasets, covering NCA, LFP, NMC111, NMC532, NMC622, and NMC811 chemistries, including battery packs with capacities of 75 kWh and 100 kWh.

The datasets are based on the BatPaC model developed by Argonne National Laboratory. Using a consistent underlying model for inventory compilation helps improve consistency across different battery datasets.

The new battery pack datasets are primarily intended for electric vehicle modelling, with capacity specifications reflecting commonly used configurations in today's electric vehicles. The more granular battery pack data enable users to select datasets that are better suited to specific transport applications.

Updates to upstream sectors also improve battery supply chain data. Revisions to lithium production supply chains enhance technological and geographical representativeness, while the regionalisation of calcined petroleum coke (CPC) production data improves the supply chain data for graphite production.

04. Construction: Corrections to Cement Formulations, Infrastructure Inputs and Land Transformation Flows

In response to user feedback, internal reviews, and inconsistencies or implausible exchanges identified in peer-reviewed literature, v3.13 introduces several corrections to construction-related datasets.

Cement: The Swiss Portland cement (CEM I) dataset has been restored to a mass-balanced formulation compliant with EN 197-1, correcting an inconsistency introduced in v3.10.

Infrastructure: Several related datasets have been revised to adjust mechanical processing capacities and copper inputs, bringing them into line with actual operational assumptions and material intensities reported in the literature. These revisions draw on studies by Tokeda & Rouwette (2024), Wang et al. (2022), and Marinova et al. (2019).

Perlite extraction: In the quarry operation dataset, mismatched land transformation flows have been corrected, eliminating anomalous negative values in the land use indicators under EF 3.1.

These corrections improve the reliability of LCA results across construction product value chains, particularly for impact categories sensitive to mineral resource use and land transformation.

05. Chemicals and Plastics: More Detailed Production Technologies and Expanded Geographical and Product Coverage

Formaldehyde production data have been differentiated at the global level to reflect three major industrial production technologies. Average production data are now available for 23 countries, while the corresponding market datasets have also been further refined to better capture regional differences.

Geographical coverage has continued to expand for acetaldehyde and major polystyrene grades, including General Purpose Polystyrene (GPPS), High Impact Polystyrene (HIPS), and Expandable Polystyrene (EPS). These updates build on data improvements introduced in recent versions.

In addition, v3.13 introduces 12 products that were not previously covered by the database, further expanding the product coverage of the chemicals and plastics sector.

06. Electricity: Updated Electricity Mix Reference Years and Improved Market and Production Data

Electricity market datasets continue to follow the consistent methodology established in versions v3.6 to v3.12. Electricity generation data for most regions continue to rely primarily on Enerdata. Where detailed national statistics are available, the corresponding national data are used, while data for European countries are sourced from the European Commission's Eurostat database.

The reference years for electricity market mixes have been updated as follows:

• Most regions: Data for 2023.
• Brazil, Canada, China, India, Switzerland, and the United States: Updated to 2024 based on comprehensive national statistics. For India, the data correspond to the financial year from April 2023 to March 2024.
• Residual electricity mixes in European markets: Based on AIB statistics for 2025.

In addition, technical corrections have been made to the "electricity production, lignite" and "treatment of coal gas, in power plant" datasets for selected countries.

Documentation and metadata for electricity market, electricity transformation, and electricity import datasets have also been updated and harmonised.

07. Fuels: Expanded Regional and Technological Coverage of Calcined Petroleum Coke

Calcined petroleum coke (CPC) production data now cover six major regions: Brazil, China, Germany, India, Oman, and the United States, with further differentiation by production technology.

The Chinese datasets reflect the locally predominant vertical shaft kiln technology, while rotary kiln technology is applied in the other five regions. The previous Chinese dataset, which had been based on rotary kiln data from an Indian facility, has been replaced.

Market datasets have also been revised accordingly. The previous Chinese market has been replaced, and a global market alongside five regional markets has been introduced, providing downstream users with region-specific supply data.

08. Metals: Expanded Lithium Supply Chains and Updated Upstream Data for Aluminium Anodes

The lithium supply chain has been expanded with 32 new production activities and market datasets covering Argentina, Australia, Chile, China, and the United States.

Based on recent literature and data from specific production sites, the new datasets cover both brine-based and spodumene-based production routes. They include raw material extraction, concentration, and the production of lithium carbonate and lithium hydroxide monohydrate, providing a more detailed representation of geographical and technological differences in lithium production.

The new version explicitly distinguishes between battery-grade and industrial-grade lithium carbonate, reflecting differences in purity requirements and end-use applications. Corresponding downstream datasets in the chemicals and batteries sectors have also been updated to use more specific lithium products. This provides more detailed data for modelling electric mobility, grid-scale energy storage, and battery supply chains.

Some of the lithium datasets were developed in collaboration with the French Geological Survey (BRGM), including the Chinese spodumene-based production route and selected brine production sites in South America.

The regionalisation of CPC data also improves aluminium supply chain modelling. Previously, CPC inputs for aluminium anode production were linked to the Chinese market, regardless of the geographical location of production. Following the update, these inputs are connected to the newly introduced CPC markets, better reflecting regional differences in supply mixes.

In addition, v3.13 corrects the uncertainty parameter settings in one aluminium hydroxide dataset and addresses an incomplete aluminium casting alloy market in the consequential system model.

09. Pulp and Paper: Updated 2024 Data and Separate Modelling of Paperboard Production and Processing

Several European paperboard and paper production datasets have been updated.

Corrugated board production now uses data for the 2024 production year, provided through a collaboration between the European Federation of Corrugated Board Manufacturers (FEFCO) and RISE Research Institutes of Sweden.

Kraft paper and paper sack production datasets have also been updated to 2024. A new dataset, "kraft paper production, for bags and wrapping", has been introduced, covering machine-glazed and machine-finished kraft paper used for shopping bags and packaging. The data were developed in collaboration with CEPI Eurokraft, Eurosac, and RISE.

New datasets have also been introduced for the production of virgin and recycled paperboard, as well as the conversion of paperboard into folding cartons.

Previously, paperboard production and processing were combined within a single activity. In v3.13, these processes are modelled separately, providing a more detailed representation of the production and conversion stages. The paperboard production data were developed in collaboration with Pro Carton and RISE.

10. Textiles: Improved Raw Material Production and Primary Processing Data for Cotton and Flax

Version 3.13 strengthens Tier 4 data coverage for cotton and flax fibre in major producing regions, focusing on raw material cultivation and primary processing stages.

The new cotton datasets cover field cultivation of both conventional and organic seed cotton, as well as ginning processes that separate cotton lint from cottonseed.

• Conventional cotton: Coverage includes Brazil, China, India, Türkiye, and the United States, collectively representing nearly three-quarters of global cotton production.
• Organic cotton: Coverage includes Brazil, India, Peru, Türkiye, Tanzania, and the United States, collectively representing nearly three-quarters of global organic cotton production.

These datasets form part of the ongoing collaboration between ecoinvent and Textile Exchange, aimed at improving the availability and quality of textile data for the apparel and footwear industries.

Global datasets for conventional and organic seed cotton cultivation and ginning have also been updated using weighted averages based on the newly introduced regional data.

For the European flax supply chain, a new dataset covering the re-scutching of short flax fibres has been introduced. Re-scutching is an important step in the cottonisation route for flax fibres. It improves the uniformity and cleanliness of short fibres to meet the requirements of subsequent textile applications. The dataset is based on primary data collected by the European Confederation of Flax and Hemp (CELC).

The European flax supply chain represented in the database covers flax fibre products certified under the Masters of FLAX FIBRE™ scheme. As of 2026, Western Europe remains the world's leading region for fibre flax production, with France, Belgium, and the Netherlands collectively accounting for three-quarters of global cultivation and processing volumes.

11. Waste Management and Recycling: Correction of Double Counting and Introduction of Two New Waste Streams

In v3.12, the EN 15804 inventory indicators Hazardous Waste Disposed (HWD) and Non-Hazardous Waste Disposed (NHWD) were added to multiple datasets. Due to incorrect classifications in certain datasets, these indicators were double-counted for some waste flows. This issue has been corrected in v3.13.

Datasets for the disposal of coal mining residues have also been revised. Previously, some datasets underestimated the conversion of phosphorus contained in the residues into phosphate emissions. In certain datasets, the conversion of nitrogen and sulphur into nitrate and sulphate emissions was also underestimated. These issues have now been corrected.

In addition, two new waste streams and their corresponding primary and secondary end-of-life treatment processes have been introduced:

• Waste solvents: Treated through incineration.
• Wastewater from pharmaceutical production: Treated through wastewater treatment processes.

Related processes for solvent recovery and regeneration are included in the newly introduced pharmaceutical sector.

12. Life Cycle Impact Assessment (LCIA) Methods: Updated Greenhouse Gas Categories and Characterisation Factors

The IPCC 2021 method has been expanded with more detailed greenhouse gas categories, aligned with the GHG Protocol.

The IMPACT World+ footprint method has been updated from version 2.1 to version 2.2.1.

In addition, several minor corrections and improvements have been implemented, and characterisation factors have been added for newly introduced elementary flows.

#Source: Official ecoinvent v3.13 Release Notes