Convective media — monoliths and membrane adsorbers — were both developed to escape the same constraint: the diffusion-limited mass transport that caps how fast a packed-bed resin can run without losing dynamic binding capacity. Both formats deliver short residence times and sharp impurity windows. But they take different routes to get there, and the choice between them shapes everything from buffer consumption to method-transfer effort.
This guide is aimed at downstream-processing teams choosing media for AAV polishing, mRNA / pDNA intermediate purification, and complex antibody-modality polishing. We compare the two formats on the metrics that actually decide a method: hydraulics, peak resolution, capacity, scale-up behaviour, and operating cost.
How each format works
Membrane adsorbers
Membrane adsorbers are stacks of ligand-derivatised flat-sheet membranes through which the feed is pumped: flat-stacked in lab and scale-down devices, wound or pleated in large capsules. Transport is convective along the flow path, and only the last short step to the membrane surface is diffusive — micrometres instead of the tens of micrometres inside a bead. That is why binding stays comparatively flat over a wide flow range.
The trade-off shows up in two places. First, flow distribution: in a pleated geometry the centre of a pleat sees a different velocity than its edges, which can broaden elution peaks at scale — in flat-stacked lab devices that effect is small, so this is a scale argument rather than a class argument. Second, void volume: the housing and the distribution plates add volume that dilutes the eluate.
Monolith columns
A monolith is a single, continuous porous body cast inside its housing. The feed flows through interconnected channels rather than around packed particles. Like membranes, transport is convective. Unlike membranes, the channels form a homogeneous network that delivers more uniform residence-time distribution — and therefore sharper peaks at the same throughput.
MonoCore™, Lumatix's cellulose monolith family, is built as a disc in which the feed crosses a short bed over a large cross-section. That keeps pressure drop low at high flow. The standard channel size is 15 µm, and the capsules come pre-assembled in 1.8 mL and 5.4 mL formats with UNF 10-32 ports.
Where the formats differ in practice
Pressure drop and cycle time
Both formats run at much lower back-pressure than a packed-bed resin at equivalent flow. In our experience the practical pressure regime of a cellulose monolith and a typical membrane adsorber are comparable; what changes is the velocity profile inside the device. Where membranes can develop edge-effect channelling at scale, a monolith maintains a single, continuous flow field — so cycle times stay predictable as you scale.
Peak resolution and product dilution
This is the dimension where the two formats most clearly diverge. Sharper peaks reduce co-elution of close-running impurities (HCP, fragmented or aggregated species, capsid variants in AAV) and yield a more concentrated product fraction.
A cellulose monolith with low void volume sits closer to a packed-bed resin on resolution while staying close to a membrane on throughput. That is the design point of MonoCore™ Q and S: resin-class resolution at convective throughput.
Binding capacity: measure it, do not rank it
A capacity figure is only meaningful together with the conditions it was measured under — the residence time, the feed and the breakthrough level someone chose. Two numbers from two suppliers are rarely comparable, which is why this page does not rank the formats on capacity and why the figures for each chemistry sit on the datasheet rather than in an article. What such a figure does and does not tell you is set out in how to read a dynamic binding capacity figure. Measure it on your own feed; that is the only number that decides anything.
Scale-up behaviour
Membrane scale-up multiplies identical elements, and suppliers report pressure-flow behaviour and breakthrough-curve shape that do not change with the number of cassettes; published scale-up studies still show the load density having to be re-checked on the larger device. Larger MonoCore™ capsules are factory-assembled from identical monolith discs in one housing, which keeps the residence time per disc constant. There is a limit on our side worth naming: the channels of a monolith do not exchange liquid with one another, so an uneven inlet distribution is never evened out further down. That makes the housing, not the matrix, the engineering task when the format grows. In both cases the honest claim is less re-optimisation, not none — and today's standard capsules are 1.8 mL and 5.4 mL, with larger formats built on request.
Hardware compatibility
Both formats drop into standard FPLC systems with UNF 10/32 fittings. Cleaning-in-place tolerance depends on the chemistry rather than the format, so check the NaOH limits on the datasheet of the specific ligand. Neither requires a custom skid for typical lab- and pilot-scale work.
When to choose which
Membrane adsorbers remain the right choice for very high throughput polishing where the impurity window is wide and product dilution is acceptable downstream — large-volume HCP scrubbing in mAb capture trains is a textbook fit.
Monoliths are worth evaluating when peak resolution or eluate concentration matters more than raw volume throughput — for example close-running variants in polishing, monomer/aggregate splits on antibody modalities, or any step where a diluted eluate forces an extra concentration step. MonoCore™ Q, S and HIC are built for these polishing steps.
If the technique decision is binary, the practical heuristic is: would a sharper peak save you a UF/DF step, a re-pool, or a re-run? If yes, a monolith likely pays for itself. If no, a membrane is fine.
Further reading
For the underlying mass-transport physics and why convective formats outperform diffusion-limited resins, see our explainer on convective vs diffusive transport in chromatography. For the cellulose-monolith family used as the basis of the comparison above, the MonoCore™ technology page covers the cellulose matrix, channel size, and chemistries on offer.
If you are looking for a replacement for a membrane adsorber rather than a comparison, see Alternatives to membrane adsorbers.
For what the backbone of a monolith decides — surface, topology and pressure — see What monoliths are made of.
Both formats are run at bed volumes per minute rather than at the minutes-long residence times of a packed bed, so on cycle time they behave similarly — the difference between them is resolution and void volume, not speed. If it is the packed bed you are comparing against, the chromatography run planner times a full method on both.
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.