Emission Factors for Bioprocessing: Where to Get Them Free

August 2026 10 min read Bioprocess Engineering

Key Takeaways

Contents

  1. What an emission factor is
  2. The four open sources worth using
  3. Electricity: the factor that decides your answer
  4. Direct fuel combustion
  5. Water and waste
  6. Polymers: compute, do not look up
  7. What you cannot legally use
  8. Being honest about data quality
  9. Frequently asked questions

You do not need a commercial database to put a defensible number on a bioprocess. You need the right emission factors for bioprocessing, from sources you are allowed to use, with their vintages recorded. This article lists the open sources that cover a fermentation or cell culture process, gives the actual values, and shows the conversions that trip people up.

It also covers the one database you are probably thinking of and cannot legally embed.

What an emission factor is

An emission factor converts a unit of activity into a mass of greenhouse gas. Multiply your activity data by the factor, sum across activities, and you have a footprint.

The unit tells you what activity it expects. Electricity factors are grams of CO2e per kilowatt-hour. Fuel factors are per megajoule, per litre or per kilogram. Water is per cubic metre. Getting the unit wrong by a factor of 1,000 is the most common arithmetic error in a first footprint, usually litres against cubic metres.

Two distinctions matter. A combustion factor covers only the gas released when the fuel burns. A life cycle factor also covers extraction, refining and transport of that fuel. They differ by roughly 15 to 20% for natural gas, so mixing the two silently understates your result.

The four open sources worth using

These four emission factor sources cover the great majority of a bioprocess inventory, and all permit reuse.

Open emission factor sources usable in commercial work
SourceCoversLicenceReuse
UK Government GHG conversion factors (DESNZ) Electricity, fuels, water supply and treatment, waste, transport, materials Open Government Licence v3.0 Commercial use permitted with “© Crown copyright” attribution
US EPA eGRID Regional US electricity generation and emissions US Government, public domain Unrestricted
NETL / EPA ElectricityLCI Regionalised grid LCI, stationary fuel combustion, power plant water use CC0 1.0 (public domain dedication) Unrestricted
NREL US Life Cycle Inventory Commodity chemicals, energy carriers, transport US Government Verify per dataset
IPCC Assessment Reports Global warming potentials (the characterisation step) Published science, citable Cite the AR and horizon used
All five permit reuse. The Federal LCA Commons hosts several of these but does not state a blanket licence on its landing page, so confirm terms per repository before shipping anything.
What are you modelling? Electricity eGRID / ElectricityLCI Fuel / steam ElectricityLCI / DESNZ Water / waste UK Government (OGL) Polymer burned stoichiometry Will the number be externally verified? If yes, you need a licensed background database and a verifier. Open factors are screening-grade.
Pick the source by what you are modelling, then check whether the output needs verification. Open factors are sufficient for screening and internal decisions, not for an EPD.

Choose eGRID or ElectricityLCI for electricity, ElectricityLCI or UK Government factors for fuel and steam, UK Government factors for water and waste, and stoichiometry for burned polymer. Then check whether the result will be externally verified, because open factors are screening-grade only.

Electricity: the factor that decides your answer

For most bioprocesses electricity is the dominant contributor, so the grid emission factor matters more than every other choice combined. It also varies enormously by location.

Across the 26 eGRID subregions the range runs from 134 g CO2e/kWh in upstate New York, which is heavily hydro and nuclear, to 762 in MRO East. That is a 5.7× spread with no change to the process whatsoever. Two identical facilities, one in New York state and one in Wisconsin, will report footprints that differ by more than five-fold.

Selected US eGRID subregion carbon intensities, computed from EPA primary totals via ElectricityLCI
SubregionRegiong CO2e/kWh
NYUPUpstate New York134
CAMXCalifornia240
NEWENew England256
RFCERFC East (mid-Atlantic)346
ERCTERCOT (Texas)460
RMPARockies625
SRMWSERC Midwest736
MROEMRO East762
Values include CO2, CH4 and N2O converted on IPCC AR6 100-year GWPs. Grid factors move year to year, so refresh annually and record the vintage.

These factors, already wired into a calculator

The Bioprocess LCA Calculator ships all 26 eGRID subregions plus custom entry, and shows the source, year and licence for every factor it uses.

Open the Calculator

Direct fuel combustion

Anything you burn on site, typically natural gas for steam and sterilisation, is a Scope 1 emission and needs a combustion emission factor rather than a grid one.

Stationary combustion factors, CO2 + CH4 + N2O on AR6 GWP100
Fuelg CO2e/MJTypical bioprocess use
Natural gas50.3Steam, SIP, space heating
Propane58.5Backup / remote sites
Distillate fuel oil70.5Standby generators
Bituminous coal89.2Purchased steam in some regions
Landfill gas54.7On-site biogas
These are close to constant year on year because they are set by fuel chemistry rather than by policy or grid mix.

Worked conversion — steam to CO2e

A batch consumes 1,400 kg of steam for sterilisation. Saturated steam carries roughly 2.2 MJ/kg, and a gas boiler runs at about 80% thermal efficiency.

fuel energy = 1,400 kg × 2.2 MJ/kg ÷ 0.80 = 3,850 MJ
CO₂e = 3,850 MJ × 50.3 g/MJ ÷ 1000 = 194 kg CO₂e

Note what the boiler efficiency does. Omitting it understates the result by 20%, because you would be charging only the heat delivered rather than the gas actually burned.

For which of these lands in which reporting category, see Scope 1, 2 and 3 emissions in biomanufacturing.

Water and waste

Water carries two separate burdens and people routinely count only one. Supply is the energy to abstract, treat and pump water to you. Treatment is the energy to process it after you are done with it.

UK Government factors put supply at 0.149 kg CO2e/m³ and wastewater treatment at 0.708. Both are needed: a process that consumes 60 m³ carries 60 × (0.149 + 0.708) = 51 kg CO2e.

There is a third term these factors do not cover, and in pharmaceutical manufacture it dominates them both. Generating WFI or purified water is separate and far larger. Multi-effect distillation runs at roughly 10 to 25 kWh/m³, reverse osmosis with hot-water sanitisation at roughly 2 to 6. At 15 kWh/m³ and a 400 g/kWh grid, that same 60 m³ carries 360 kg CO2e of generation energy against 51 kg for supply and treatment combined. Count it as electricity, not as water.

Polymers: compute, do not look up

Single-use plastics carry two burdens: making the resin, and burning it at end of life. The first needs a published factor. The second does not, and you get a better answer by calculating it.

Every carbon atom in the polymer ends up as CO2. So the incineration factor is just the carbon mass fraction of the repeat unit multiplied by 44.009/12.011.

Polymer incineration factors derived from repeat-unit composition
PolymerCarbon mass fractionkg fossil CO2 per kg burned
Polyethylene (PE)0.8563.14
Polypropylene (PP)0.8563.14
Polycarbonate (PC)0.7562.77
EVA (~28% VA)0.6802.49
PVC0.3841.41
PVDF0.3751.38
PVC and PVDF look favourable here only because chlorine and fluorine displace carbon by mass. Their production burdens are higher, and halogenated combustion raises separate acidification and toxicity concerns outside the climate category.

Resin production is the part you do need a published factor for. European cradle-to-gate values are roughly 1.9 kg CO2e/kg for HDPE, 2.1 for LDPE and 2.0 for polypropylene. Speciality polymers such as polycarbonate and fluoropolymers are substantially higher and much less well characterised in open sources, which is where the uncertainty in a single-use footprint concentrates.

What you cannot legally use

ecoinvent cannot be embedded in an interactive tool

ecoinvent's own licensing documentation states that making its data available to third parties through an interactive tool, even in the background, makes you a reseller of the database. That requires a developer licence plus a sublicensing agreement priced on datasets, indicators and end-user count. Calculated results may be shared statically, for example inside a PDF report, but automated sharing through an application is not permitted under a standard single-user licence.

This is why free LCA calculators are built on open factors, and why openLCA is free while its data is not. It is a licensing boundary, not a technical one, and it is worth understanding before you design anything that other people will use.

The practical escape hatch is to let users supply their own factors. If someone holds an ecoinvent licence, they can enter values under their own terms. What you cannot do is ship the data.

Mass-based metrics need no emission factors at all, which is why process mass intensity is the industry's common benchmark.

Being honest about data quality

Open emission factors are not uniformly good, and pretending otherwise is how screening studies mislead. A simple three-tier label makes the weak spots visible.

For a bioprocess the weak tier is consistently the same: media components and speciality polymers. If those together drive a large share of your result, the honest conclusion is that you do not yet know the answer, and the next step is supplier data rather than a more elaborate model.

See the factors in a complete worked study

Our worked life cycle assessment example applies these factors end to end across all four ISO 14040 phases, with the arithmetic shown at every step.

Read the worked example

Frequently asked questions

What is an emission factor?

A conversion from a unit of activity to a mass of greenhouse gas. Electricity is grams of CO2e per kWh, fuel per MJ or litre, water per cubic metre. Multiply activity by factor and sum.

Where can I get free emission factors?

UK Government conversion factors under the Open Government Licence, US EPA eGRID and the NETL/EPA ElectricityLCI dataset (CC0), NREL's US Life Cycle Inventory, and IPCC Assessment Report tables for global warming potentials. Between them they cover electricity, fuels, water, waste and the characterisation step.

Can I use ecoinvent in my own calculator or app?

Not without a sublicensing agreement. Exposing ecoinvent through an interactive tool, even in the background, makes you a reseller under their terms. Static results in a report are fine; an application serving the data is not.

What is the emission factor for electricity?

Entirely location-dependent. Across the 26 US eGRID subregions it runs from about 134 to 762 g CO2e/kWh. Always state which grid factor you used, because it usually dominates the result.

How do I calculate CO2 from burning plastic?

Multiply the carbon mass fraction of the repeat unit by 44.009/12.011. Polyethylene at 85.6% carbon gives 3.14 kg CO2 per kg. PVDF gives only 1.38 because fluorine displaces carbon by mass.

How current do emission factors need to be?

Grid factors move fastest and should be refreshed annually. Global warming potentials change only with a new IPCC Assessment Report. Fuel combustion factors are close to stable. Record the vintage of every factor, because footprints are only comparable on matching vintages.

References

  1. UK Department for Energy Security and Net Zero. Government conversion factors for company reporting. Open Government Licence v3.0. gov.uk
  2. US EPA. Emissions & Generation Resource Integrated Database (eGRID). epa.gov/egrid
  3. NETL / US EPA. ElectricityLCI, released under CC0 1.0. github.com/NETL-RIC/ElectricityLCI
  4. US EPA. LCIA Formatter, MIT licence, source of the IPCC characterisation factors used here. github.com/FLCAC-admin/LCIAformatter
  5. Herrmann, I.T. & Moltesen, A. (2015). Does it matter which Life Cycle Assessment (LCA) tool you choose? Journal of Cleaner Production. doi:10.1016/j.jclepro.2014.08.004

Resources & Further Reading