Ask anyone who has evaluated monolith columns what worries them, and consistency between lots comes up before performance does. It is a fair question and it deserves a straight answer, because a monolith is made differently from a packed bed: the separation body is cast as one piece, so it cannot be re-packed if it turns out wrong, and the variation that matters arises during manufacture rather than during packing.
This page is about how that variation is detected and what you should ask a supplier to show you. It is deliberately about method rather than about any one supplier's numbers, ours included — a number without the method behind it is not worth much anyway.
Why consistency means something different on a cast body
On a packed bed, a large part of run-to-run variation comes from packing: bed height, compression, how evenly the slurry settled. The medium itself is a powder that is well characterised before it goes into a column, and a badly packed column can be unpacked and packed again.
A monolith has no packing step. The flow path is formed when the body is made, and it is fixed from that moment. That removes a whole class of variation — nobody packs it badly — and concentrates the remaining variation in manufacture. Which is why the question "how consistent are your lots" is the right question to ask a monolith supplier, and a less interesting one to ask a resin supplier.
What actually varies, and what detects it
| Property | Why it varies | What detects a change |
|---|---|---|
| Flow path uniformity | Formed during casting; the distribution of channel sizes is not a single number | A tracer step — a conductivity or UV step change — and the shape of the response curve |
| Bed integrity | A crack or a wall gap sends flow past the adsorbent | The same tracer step: a shoulder or an early breakthrough shows it immediately |
| Pressure at a defined flow | Follows from the flow path; a shift signals a change in the body or the housing | Pressure at a stated flow rate and temperature, on the same system |
| Ligand density | Functionalisation is a chemical step with its own variation | Binding capacity under stated conditions, or a direct ligand assay |
| Housing and assembly | Connectors, seals, dead volume | Hold-up volume and the symmetry of the tracer response |
The first two deserve emphasis, because they are the ones a buyer rarely asks about and a supplier can answer cheaply. A step change in conductivity, run through the column and recorded, produces a curve whose shape says a great deal: how sharp the transition is, whether it is symmetric, whether there is a shoulder. Two lots that give the same curve under the same conditions are behaving the same way hydrodynamically — before any protein is involved.
What to ask a supplier for
- A certificate of analysis per lot, not a generic specification sheet. It should name the lot, the tests run on it and the values obtained — not just "conforms".
- The conditions each value was measured under. A capacity figure without residence time, feed and breakthrough level cannot be compared to anything, as set out in how to read a dynamic binding capacity figure.
- Traceability of the parts. On a monolith the separation body, the functionalisation and the housing can each have their own lot. Ask how they are linked, because that is what makes an investigation possible if something is ever wrong.
- What happens when a lot fails. A supplier who has never rejected a lot either has a perfect process or a weak specification. The interesting answer is how failures are detected and what follows.
- Retest or expiry dating, and what it is based on.
Where we are with it
We are early, and the honest thing to say is that our own framework for this is still being built rather than finished. Publishing a control strategy now would mean describing something that looks more settled than it is, and then correcting it every quarter.
So: if lot-to-lot consistency is what decides your evaluation, ask us directly. We will tell you what we test today for the chemistry you are looking at, what we can put in front of you, and what we do not have yet. That is a more useful conversation than a paragraph on a page, and you get an answer that is true on the day you ask it.
Frequently asked questions
Why is lot-to-lot consistency a bigger question for monoliths than for resins?
Because the separation body is cast as one piece rather than packed from a characterised powder. A resin is analysed before it ever reaches a column, and a poorly packed column can be packed again. A monolith's flow path is formed during manufacture and fixed from then on, so the variation that matters sits in the manufacturing step — which is exactly where a buyer should direct the question.
How do you test consistency between monolith lots?
The most informative cheap test is a tracer step: run a conductivity or UV step change through the column and record the response. Its sharpness, symmetry and the absence of a shoulder describe the flow path and the integrity of the body. Beyond that: pressure at a defined flow rate, and binding capacity under stated conditions. All three run on a standard FPLC system.
What should a certificate of analysis for a chromatography column contain?
The lot number, the tests run, the values obtained and the conditions they were measured under — and ideally the acceptance criterion each value was judged against. A certificate that only says "conforms to specification" tells you that someone looked, not what they saw.
Is every capsule tested, or a sample of the batch?
It depends on the test, and that is true of any supplier. Some checks are non-destructive and can in principle run on every unit; a binding capacity determination consumes the unit, so it is necessarily done on a sample and reported for the lot. The useful question to put to a supplier is which of their tests falls into which category, and to ask for that in writing.
What do you do when a lot does not meet specification?
It does not ship. That is the entire point of having a specification, and it is worth asking any supplier what their threshold actually is — a limit that has never stopped anything is not yet doing work.
Still an open question?
Tell us the molecule, the feed volume and the system you run it on, and we will say whether a MonoCore™ capsule is the right starting point — or whether it is not.