Diagnosing pump-induced product damage
When a process loses yield or potency, the pump is rarely the first suspect. It should be on the list. There are four distinct ways a pump damages product, they often act together, and telling them apart is the difference between fixing the cause and chasing the symptom.
Most troubleshooting starts at the molecule: the formulation, the buffer, the membrane. Those matter. But a pump touches every milliliter of product, often hundreds of times in a recirculating step, and each pass is an exposure to whatever mechanical insult that pump architecture creates. Before you reformulate, it is worth asking a simpler question: what is the equipment doing to the molecule, and which mechanism is responsible?
01 / The four mechanisms
Four failure modes, not one
Shear is mechanical stress on the molecule, and it is not a single thing. Bulk hydrodynamic shear in the flow field is distinct from solid-solid interfacial contact at a surface, which is distinct again from the stress generated when a cavitation bubble collapses. Each has a different cause and a different fix, and conflating them is the most common error in the field.
Cavitation occurs when local pressure at the pump inlet drops below the fluid's vapor pressure, so the liquid briefly boils. The vapor bubbles then collapse violently downstream, generating localized shockwaves and heat that denature protein and rupture enveloped particles.
Pulsation is the pressure transient a pump injects with every cycle. It propagates through the whole fluid path and destabilizes anything pressure-sensitive downstream.
Hold-up (dead) volume is fluid trapped in tubing, heads, and fittings that cannot be recovered. It is direct yield loss, and a source of cross-batch carryover.
02 / Symptom to mechanism
Read the symptom first
Each mechanism leaves a different fingerprint. Use the symptom to form a hypothesis before you instrument.
Form the hypothesis from the symptom, then confirm with instrumentation. The mechanisms commonly act together.
| Symptom | Most likely mechanism |
|---|---|
| Audible knocking or rattle, worse under suction lift, high viscosity, or low tank level | Cavitation (inlet pressure below vapor pressure) |
| Subvisible particle counts that rise with run hours or total cycles | Tubing spallation and interfacial aggregation (peristaltic) |
| TMP or inline-sensor noise oscillating at a multiple of the rotor or stroke frequency | Pulsation |
| Low recovery on small volumes, cross-batch carryover, long flush requirements | Hold-up / dead volume |
03 / How to confirm it
Instrument the hypothesis
Cavitation: a high-bandwidth inlet pressure transducer shows whether suction pressure drops below vapor pressure, and whether the margin (NPSH available versus required) collapses under viscosity or backpressure.
Interfacial damage and spallation: subvisible particle counting (light obscuration or micro-flow imaging) correlated against total compression cycles, not flow rate, isolates tubing-driven damage. Aggregation that tracks cycle count points to the tubing wall.
Pulsation: take the pressure trace and run an FFT. Energy concentrated at the rotor or stroke frequency and its harmonics confirms pulsation and tells you the order, which in turn points back at the pump geometry.
Hold-up: a simple mass-balance and flush-volume study quantifies trapped, unrecoverable fluid.
04 / It is usually more than one
Mechanisms compound
The instinct to find a single villain is what makes troubleshooting slow. In practice the mechanisms stack: pulsation lowers the instantaneous inlet pressure and triggers cavitation on the trough; cavitation and interfacial contact both feed aggregation. And in any recirculating step, single-pass damage is multiplied across hundreds of passes, so a small per-pass insult becomes a large cumulative loss.
05 / What the framework points to
Trace every branch to its root
Notice that every branch of the decision tree traces back to one of three roots: negative pressure at the inlet, mechanical contact with the product, or pressure transients. That is a useful lens for selection as well as diagnosis. An architecture that floods its inlet by gravity keeps the suction pressure high and removes the cavitation branch. One that displaces fluid without a roller compressing tubing or an impeller contacting product removes the contact branch. One whose geometry cancels low-order pulsation removes the transient branch. You do not have to take any vendor's word for which pump does that; the diagnostic above lets you measure it for yourself.
Common questions
Is protein damage during pumping caused by shear?
Partly, but "shear" is not one mechanism. Bulk hydrodynamic shear, solid-solid interfacial contact at a tubing wall, and cavitation-induced stress are distinct, and peer-reviewed work shows interfacial contact at the tubing surface is a dominant driver of aggregation during peristaltic pumping. Treating all damage as bulk shear leads to the wrong fix.
How do I tell cavitation from pulsation?
Cavitation is usually audible (knocking or rattle), worsens with suction lift or viscosity, and shows on a high-speed inlet pressure trace dropping below vapor pressure. Pulsation is a periodic pressure swing visible as energy at the rotor or stroke frequency in an FFT of the pressure signal. They can occur together, since a pulsation trough can trigger cavitation.
Why does pump damage get worse at scale?
Larger volumes mean longer runs and more total cycles, and recirculating steps such as TFF expose the product to the pump hundreds of times. A per-pass insult that is invisible at bench scale accumulates into measurable loss at clinical and commercial scale, which is why a process can pass at small scale and fail later.
References
- Solid-Solid Interfacial Contact of Tubing Walls Drives Therapeutic Protein Aggregation During Peristaltic Pumping. Journal of Pharmaceutical Sciences, 2023.
- Reaching the breaking point: Effect of tubing characteristics on protein particle formation during peristaltic pumping. International Journal of Pharmaceutics, 2022.
- Particle Shedding from Peristaltic Pump Tubing in Biopharmaceutical Drug Product Manufacturing. Journal of Pharmaceutical Sciences, 2015.