Process mass intensity is the metric the biologics industry actually benchmarks on, and it is the one most process development teams could calculate this afternoon from data they already have. It is also routinely misread, because a number in the tens of thousands sounds alarming until you understand that almost all of it is water.
This article covers what PMI measures, how to calculate it, what good looks like, and the specific thing it is blind to.
What process mass intensity measures
Process mass intensity (PMI) is the total mass of everything entering a process divided by the mass of product leaving it, expressed in kg/kg. A PMI of 7,700 means 7,700 kg of input per kg of product.
The ACS Green Chemistry Institute Pharmaceutical Roundtable defines the biologics version as the total mass of water, raw materials and consumables required to make 1 kg of active pharmaceutical ingredient. Splitting it three ways matters, because the three categories behave completely differently and respond to different interventions.
The metric is deliberately simple. It requires no emission factors, no characterisation model and no background database. That is its strength as a benchmarking metric: two companies can compare results without agreeing on anything except the boundary and the three categories.
How to calculate PMI
Sum every mass input across the boundary, then divide by the product mass. The only real decisions are where the boundary sits and whether you count on a per-batch or per-campaign basis.
Worked example — 2000 L single-use CHO fed-batch
Taking the same batch used in our worked LCA: 1600 L working volume, 3.0 g/L titer, 70% downstream yield, 90% success rate, giving 3.024 kg of purified product.
media components 48 kg
single-use consumables 82 kg
─────────
total input mass 60,130 kg
PMI = 60,130 kg ÷ 3.024 kg = 19,884 kg/kg
of which water = 19,841 kg/kg (99.8%)
raw materials = 16 kg/kg
consumables = 27 kg/kg
Water is taken as 1 kg per litre. This process sits at the top of the published range, which is what you would expect at a 3 g/L titer with conventional batch chromatography.
Benchmark values for biologics
Budzinski and colleagues published the benchmarking exercise that established the reference values, covering six large pharmaceutical companies. The headline is an average of about 7,700 kg per kg of mAb, with individual processes spanning roughly 3,000 to over 20,000.
| Case | PMI (kg/kg) | Note |
|---|---|---|
| Best-in-class mAb | ~3,000 | High titer, optimised downstream |
| Industry average (6 companies) | ~7,700 | Budzinski et al. benchmarking |
| Upper end of reported range | >20,000 | Low titer, buffer-heavy purification |
| Worked example above (3 g/L) | 19,884 | Conventional batch, single-use train |
The spread is not noise. It reflects two variables: titer, which sets the denominator, and how chromatography is operated, which sets most of the numerator.
Why water dominates
Over 90% of the input mass in a biologics process is water, and in the worked example above it is 99.8%. This is the single fact that makes biologics look extreme next to small-molecule chemistry.
Water enters at media preparation, buffer preparation, chromatography, diafiltration and cleaning. Chromatography is the largest consumer, and downstream processing overall is roughly 75 percent of total process mass intensity.
Where PMI actually accumulates
Downstream processing accounts for roughly 75% of the total, and chromatography is the single largest consumer of both water and raw materials within it.
The arithmetic is unforgiving. A capture step running equilibration, load, wash, elution, strip and regeneration can consume 15 to 25 column volumes per cycle. Multiply by the cycles needed to process a harvest, add two polishing steps, and buffer volume reaches tens of times the bioreactor volume before anything has been concentrated.
This is why titer improvements pay twice. A higher titer means more product per batch, and it also means fewer chromatography cycles to process the same mass, which cuts buffer volume directly.
Calculate PMI alongside your carbon footprint
The Bioprocess LCA Calculator reports PMI split into water, raw materials and consumables using the ACS GCI definition, next to kg CO2e per kg so you can see where the two metrics disagree.
How to reduce mass intensity
In rough order of leverage for a typical mAb process:
- Raise titer. The denominator scales directly, and higher titer also reduces chromatography cycles. This is usually the largest single lever.
- Increase resin binding capacity. Higher dynamic binding capacity means fewer cycles, and fewer cycles means proportionally less buffer.
- Remove a chromatography step. Moving from three columns to two typically removes a large block of buffer volume outright.
- In-line buffer dilution. Preparing concentrates and diluting in-line cuts both preparation volume and hold-vessel cleaning.
- Continuous or connected processing. Madabhushi and colleagues compared continuous against batch for biologics and found PMI reductions, driven largely by smaller equipment and reduced hold volumes.
- Water recovery. Reusing CIP rinse water where quality requirements permit. Usually the smallest lever, and constrained by GMP.
What the metric does not tell you
The metric counts mass without weighting it by impact. A kilogram of purified water and a kilogram of an energy-intensive solvent contribute identically. That makes it an excellent benchmark and a poor optimisation target on its own.
The failure mode is specific and worth naming: a process can reduce mass intensity while increasing its carbon footprint. Switching from an aqueous wash to a solvent-based one, or replacing a water-intensive step with an energy-intensive one, improves the number and can worsen everything else.
In our worked example the two metrics point in different directions entirely. PMI says water is 99.8% of the problem. The carbon footprint of the same batch says electricity is 95.6% of the problem, and the water contributes under 0.5% of the CO2e. Both are correct. They are measuring different things, and a team optimising only one of them will make the other worse.
See both metrics on the same batch
Our worked life cycle assessment example runs the identical process through all four ISO 14040 phases, so you can see exactly where PMI and carbon disagree and why.
Frequently asked questions
What is process mass intensity (PMI)?
Total input mass divided by product mass, in kg/kg. For biologics the ACS GCI Pharmaceutical Roundtable defines it as water plus raw materials plus consumables per kg of API. A PMI of 7,700 means 7,700 kg of input per kg of product.
What is a typical PMI for a monoclonal antibody?
About 7,700 kg/kg on average across six benchmarked companies, ranging from roughly 3,000 to over 20,000. Titer and chromatography configuration explain most of the spread, so quote them alongside the number.
Why is PMI so high for biologics compared with small molecules?
Water. Over 90% of the input mass is aqueous: media, buffers, equilibration, wash, elution, diafiltration and cleaning. Low titers relative to the volumes handled mean a few grams per litre of product carries tens of thousands of litres through the process.
Which part of the process contributes most to PMI?
Downstream, at roughly 75% of the total, with chromatography the largest single consumer. Buffer volume per cycle, resin capacity and step count are the highest-leverage variables.
Is a lower PMI always better?
No. PMI is unweighted by impact, so a kilogram of water and a kilogram of solvent score the same. A process can cut PMI while raising energy use or switching to a more damaging material. Use it alongside a carbon or energy metric.
How do you reduce process mass intensity?
Raise titer first, then increase resin binding capacity, remove a chromatography step, adopt in-line buffer dilution, and consider continuous processing. Water recovery is real but usually the smallest lever and the most GMP-constrained.
References
- Budzinski, K., Blewis, M., Dahlin, P., D'Aquila, D., Esparza, J., Gavin, J. et al. (2019). Introduction of a process mass intensity metric for biologics. New Biotechnology. doi:10.1016/j.nbt.2018.07.005
- Madabhushi, S.R., Pinto, N.D.S. & Lin, H. (2022). Comparison of process mass intensity (PMI) of continuous and batch manufacturing processes for biologics. New Biotechnology. doi:10.1016/j.nbt.2022.11.002
- Bunnak, P., Allmendinger, R., Ramasamy, S.V., Lettieri, P. & Titchener-Hooker, N.J. (2016). Life-cycle and cost of goods assessment of fed-batch and perfusion-based manufacturing processes for mAbs. Biotechnology Progress. doi:10.1002/btpr.2323