Single-Use vs Stainless Steel: Which Has the Lower Environmental Impact?

Key differences at a glance

• Carbon: a dead heat. 3,980 kg CO2e/kg for the disposable route against 3,971 for the fixed vessel, a difference smaller than the uncertainty on either figure.
• Water and PMI: disposables win clearly. 60,000 vs 84,000 L per batch, and a process mass intensity of 19,884 vs 27,797 kg/kg.
• Solid waste: the fixed vessel wins clearly. 9.5 kg of polymer per batch against 82 kg.
• The answer flips at a grid intensity of 479 g CO2e/kWh, because cleaning is grid-sensitive and plastic is not.
• Neither choice is the lever that matters. Raising titer from 3 to 10 g/L cuts the footprint by 70%. The vessel decision moves it by 0.2%.

Everything below is generated by the free bioprocess LCA calculator using its two matched 2,000 L CHO fed-batch presets, so you can reproduce and vary any number in this article yourself. This page is about environmental impact only. For capital and operating cost, see our separate single-use vs stainless steel cost comparison, which puts the crossover around 2,000 to 5,000 L.

Side-by-side comparison

Both columns describe the same 2,000 L CHO fed-batch process at 3 g/L titer, 70% downstream yield, 14 days and a 90% success rate, in the same cleanroom, on a 400 g CO2e/kWh grid. The only things that differ are the cleaning regime and the polymer inventory.

Screening-level model. Emission factors are from open sources (US EPA eGRID via ElectricityLCI, UK Government conversion factors, published polymer eco-profiles). Your site's measured utility data takes precedence.

Single-use in detail

The disposable route replaces a cleaned vessel with a gamma-irradiated bag, and replaces the cleaning cycle with a purchase order. Environmentally, that is a trade of an operating burden for an embodied one: you stop buying steam and cleaning water, and you start buying polymer that has to be manufactured and then destroyed.

Where the burden sits

The 2,000 L configuration modelled here carries 82 kg of polymer per batch across bags, tubing, filters and connectors: 45 kg polyethylene, 18 kg polypropylene, 12 kg EVA, 4 kg polycarbonate and 3 kg PVDF. Resin production accounts for 204 kg CO2e and incineration of the whole lot for a further 243 kg, giving a plastic burden of 446 kg CO2e per batch.

Against a batch total of 12,036 kg CO2e, that is 3.7%. The number most people expect to dominate the comparison is, on this boundary, smaller than the rounding error on the cleanroom estimate. Even the end-of-life question is modest: routing all 82 kg to landfill instead of incineration would take the footprint from 3,980 to 3,900 kg CO2e/kg, an improvement of 2.0%, and it would trade carbon for a solid-waste problem that most sites cannot legally accept for biohazardous material anyway.

When disposables win

They win decisively on the mass metrics. Avoiding 24,000 L of cleaning water per batch is worth 28.6% off total water and 28.5% off process mass intensity. If your site is in a water-stressed region, or your sustainability reporting leads with water rather than carbon, that is the whole argument and it is a strong one. They also win on carbon wherever the grid is dirty, for reasons covered below.

Stainless steel in detail

A fixed vessel front-loads its environmental cost into steel that lasts twenty years and then pays a recurring toll every batch to clean and sterilise itself. That toll is what an environmental comparison has to price, and it has three parts.

The cleaning premium, itemised

Clean-in-place raises cleaning water from 2,000 to 26,000 L per batch. Steam-in-place adds 1,400 kg of saturated steam. Both then have knock-on effects, and the knock-on is bigger than the direct one:

The third row is the one teams forget. Generating purified water is energy-intensive, and at 15 kWh per cubic metre the extra 24 m³ of cleaning water costs seven times more carbon in distillation electricity than it does in supply and effluent treatment combined. Our guide to WFI and purified water systems covers why that term is so often missing from a site's model, and the cleaning validation calculator is the place to size the cycle itself.

When the fixed vessel wins

On solid waste, always: 9.5 kg of polymer per batch against 82 kg. On carbon, wherever electricity is clean enough that the distillation penalty stays small, which in this model means below 479 g CO2e/kWh. And at very large scale the vessel simply has no competitor, since the disposable route runs out of bag above roughly 2,000 L and has to be run as multiple parallel trains.

Why the two nearly cancel

Setting the shared terms aside, only five line items actually differ between the two configurations. Laid out together they explain the entire result:

Two observations follow, and both are more useful than the headline number.

First, the difference is scale-invariant per kilogram. Every term in that table scales linearly with batch volume, and so does product mass, so the 9 kg CO2e/kg gap survives unchanged whether you model 2,000 L or 20,000 L. Scaling up cuts both footprints hard, from 3,980 to 780 kg CO2e/kg at 20,000 L, because the fixed cleanroom load is spread over more product. It does not change which route wins.

Second, the difference is independent of the cleanroom. HVAC appears identically in both columns, so it cancels exactly in the subtraction. That has a counterintuitive consequence for how the comparison is presented, which is the next section.

The grid decides it, and it flips at 479 g CO2e/kWh

Look again at the decomposition. Of the three terms in the cleaning premium, one is grid-dependent: the 360 kWh of distillation electricity. The plastic premium has no grid term at all, because resin chemistry and combustion stoichiometry do not care where your electrons come from. So as the grid gets dirtier, the cleaning penalty grows and the plastic penalty stands still.

Every one of those margins is under 1%. The honest reading is not "pick disposables above 479 g CO2e/kWh"; it is that a decision this finely balanced should be made on cost, supply-chain resilience, changeover time and contamination risk, and that whichever way it goes, the environmental consequence is a rounding error. The grid figures themselves come from EPA eGRID subregions, which span 134 to 762 g CO2e/kWh across the United States alone. Our note on emission factors for bioprocessing lists the open sources behind them.

Why published studies disagree: the boundary does the arguing

Because HVAC cancels in the subtraction but not in the totals, you can change the apparent size of this comparison without changing a single physical fact. Draw the boundary tightly around the vessel and its cleaning, excluding the cleanroom that both sit in, and the same batch looks like this:

The absolute difference is identical in both rows, 9 kg CO2e per kg. Only the denominator moved. On the narrow boundary, plastic looks like a third of the footprint and the debate looks worth having; on the wide one it is under 4% and the debate is noise. A study that reports percentage savings without stating whether facility overhead is inside the boundary is not really telling you anything, and a good deal of the spread between published comparisons comes from exactly this choice. The worked life cycle assessment example walks through how the goal and scope phase pins this down before any arithmetic happens.

Pros and cons

Single-use

Stainless steel

The real environmental case for disposables has nothing to do with plastic

Everything so far charges cleanroom HVAC only for the 14 days the batch is running. That is a defensible boundary, but it flatters the route with the longer changeover, because a suite held for cleaning and release is still being heated, cooled and filtered at full rate.

Charge the cleanroom over the whole slot instead, at a 3-day changeover for the disposable route and 5 days for the fixed vessel, and the picture changes:

A 9.5% advantage is nearly fifty times the plastic-versus-cleaning difference, and it comes entirely from vacating the room two days sooner. That is the environmental argument for disposables that survives scrutiny, and it is an argument about occupancy, not about waste. It also explains why the answer tracks facility utilisation: at 280 operating days a year, the shorter cycle yields 16 batches and 48.4 kg against 14 batches and 42.3 kg. You can test the sensitivity of that to your own changeover time with the seed train planner and price the same schedule in dollars with the fermentation economics calculator.

Which should you choose?

On carbon alone the honest answer is that it does not matter much. These four scenarios cover the cases where the environmental argument is strong enough to be worth acting on.

Real-world use cases

How the comparison lands in four common configurations, all modelled on the same 400 g CO2e/kWh grid.

Lifecycle inventory and what the difference is worth

Side by side, the full per-batch inventory that the model tracks:

It is worth putting a price on the gap. At an internal carbon price of $100 per tonne, a 9 kg CO2e/kg difference is worth about $0.90 per kg of product. The harmonised cost of goods for this same batch is roughly $154,762 per kg, so the entire environmental case for one vessel type over the other is worth six ten-thousandths of one percent of manufacturing cost. Our life cycle costing guide derives that figure and the marginal abatement costs that go with it, and the short version is that every lever which raises product mass pays for itself many times over while a vessel swap does not register.

None of this argues against disposables. It argues against justifying them on carbon. Changeover speed, contamination risk, capital deferral and facility flexibility are all real and all larger. If carbon is the objective, the money belongs in titer, grid procurement and cleanroom setback, which is the ranking our Scope 1, 2 and 3 emissions guide arrives at from the reporting side.

Vendor landscape

Both camps now publish sustainability material, and the useful test is whether a claim states its system boundary and its comparator titer. Most do not.

Single-use system suppliers

• Sartorius: Biostat STR rocking and stirred platforms up to 2,000 L. Publishes film composition and extractables data, which is what a credible polymer inventory actually needs.
• Thermo Fisher Scientific: HyPerforma single-use bioreactors and mixers. Broad installed base makes its bag mass data a reasonable industry proxy.
• Cytiva: Xcellerex platforms. Its predecessor organisation authored the most-cited disposable-versus-conventional life cycle assessment, so read that work knowing the affiliation.
• Merck Millipore: Mobius bioreactors and assemblies, with a strong position in the connectors and filters that make up a surprising share of polymer mass.

Stainless steel vessel suppliers

• ABEC: Large custom fermenters and bioreactors, including the very large vessels where disposables are simply not an option.
• GEA: Fermentation systems plus the CIP and SIP skids themselves, which is where the cleaning premium in this article is actually set.
• Getinge: Applikon stirred vessels across pilot and production scale, and autoclave and sterilisation equipment.
• Pfaudler: Reactor vessels and mixing systems, with a long service life that spreads embodied steel over many campaigns.

Whichever way a supplier's material leans, the arithmetic above is reproducible on open emission factors, so you can check any claim against it in a few minutes rather than accepting it.

Related calculators and tools

Related articles

Resources and references

• Pietrzykowski et al. (2013), Journal of Cleaner Production 41:150–162. The most-cited life cycle assessment comparing disposable and conventional process technology for monoclonal antibody production. Authored from within a single-use supplier, which is worth knowing when reading its conclusions.
• Budzinski et al. (2022), New Biotechnology 68:28–36. Streamlined life cycle assessment of single-use technologies in biopharmaceutical manufacture, and a good model for how to scope a screening study.
• Budzinski et al. (2019), New Biotechnology 49:37–42. Introduces the process mass intensity metric for biologics, the basis of the PMI figures used here.
• Amasawa et al. (2021), ACS Sustainable Chemistry & Engineering 9:14012–14021. Cost-benefit analysis of monoclonal antibody cultivation scenarios covering both environmental impact and operating cost, which is the pairing this article argues for.
• UK Government GHG conversion factors for company reporting. Open Government Licence, and the source of the water supply and treatment factors used above.