Dynamic binding capacity is the amount of target a chromatography medium binds under flow before a defined share of it appears at the outlet, in milligrams per millilitre of bed. Most people meet it as a single number on a datasheet, and most of the time that is the right way to meet it: the manufacturer measures it, you work with the value. The difficulty is that the number only means something together with the conditions it was measured under — and those conditions are what decide whether it transfers to your process at all. Every MonoCore™ datasheet states them next to the value for that reason, and the flow rate they refer to is 4 monolith volumes per minute, not the fastest rate the capsule tolerates.
Which conditions have to match before two figures can be compared?
| Condition | Why it changes the result |
|---|---|
| Target molecule | Capacity is mass-based, so a large molecule occupies more mass per binding site. An IgG figure says little about a viral vector |
| Feed concentration | Capacity rises with titre along the adsorption isotherm and levels off once it saturates — for a mAb on Protein A typically above a few g/L |
| Buffer, pH and conductivity | Binding strength depends on the solution. For ion exchange pH usually matters more than conductivity; an affinity ligand is less sensitive, but not insensitive |
| Residence time | The one most often omitted. On porous beads capacity falls as residence time shortens, because the target has less time to diffuse into the pores |
| Bed height | The mass transfer zone has a roughly fixed length at a given residence time. On a short bed it takes up a larger share of the bed, so the same medium gives a lower figure |
| Temperature | Affects diffusion, viscosity and the binding equilibrium. A figure measured at room temperature does not simply transfer to a harvest processed in a cold room |
| Breakthrough level, and how it is read | 1 %, 5 % and 10 % give substantially different numbers. And 10 % of the UV signal is not 10 % of your product once the feed also contains host-cell protein and medium components |
The practical consequence is blunt: two capacity figures from two datasheets can only be compared when all of these match. In practice they rarely do, which is why a capacity comparison between suppliers is usually an argument about conditions rather than about media. Batch-to-batch variation in ligand density sits on top of that, and it is a property of the product rather than of the measurement.
What is the difference between static and dynamic capacity?
Static, or equilibrium, capacity is what a medium binds when it has unlimited time — measured in batch, with medium and feed simply left together. Dynamic capacity is what it binds under flow, when the target has only as long as the residence time to find a binding site. Static capacity is what a materials datasheet likes to quote; dynamic capacity is what a process design can use.
In principle static capacity is the upper bound: at the same liquid-phase concentration and on the same volume basis, nothing binds more under flow than at equilibrium. As two numbers on two datasheets they are not automatically in that order, though. A static figure from a batch experiment in which the protein was largely depleted can come out below a dynamic figure measured against a continuously supplied feed, simply because the equilibrium concentration at the end of the batch was far lower than the feed concentration. It is worth knowing which experiment produced the number before treating one as a bound on the other.
Why does capacity depend on flow — and does that differ by format?
In a porous bead most binding sites sit inside the pores, and the target reaches them by diffusion. Diffusion is slow for large molecules, so at a short residence time the target passes the bead before it has found its way in and breaks through while the medium is far from saturated. This is why packed beds are typically loaded at a residence time of two to six minutes for Protein A capture, and why their capacity falls as the flow is increased.
In a convective medium — a monolith or a membrane adsorber — the liquid moves through open channels and carries the target past the ligands, so binding does not wait on diffusion into a bead. That removes the resistance that diffusion into a pore represents, and leaves others in place. The target still has to cross the stagnant film at the channel wall, the association rate of the ligand is finite, and an uneven flow distribution or a broad channel size distribution widens the breakthrough curve, which lowers the figure at 10 % specifically. At residence times of a few seconds all three become visible. At residence times of a few seconds these become visible, so capacity on a convective medium is not independent of flow either — how flat the curve is over the range you intend to run has to be read from that product's datasheet.
None of this says which medium binds more, and this page will not: capacity belongs to a specific product under specific conditions, and it is stated in that product's datasheet rather than compared in an article. What the mechanism does say is where a figure is likely to be sensitive to how fast you run — which is the part you can check yourself. MonoCore™ capsules are supplied in 1.8 mL and 5.4 mL with 15 µm channels, and the capacity stated on each datasheet applies at 4 MV/min; the transport mechanism behind all of it is in convective vs. diffusive mass transport.
How much of the capacity should you load?
Load against the breakthrough level the figure refers to, not against the figure. A well-characterised capture step commonly runs at 80–90 % of a capacity quoted at 10 % breakthrough; a variable feed, or a figure quoted at 1 % breakthrough, justifies more margin. In an established mAb platform the loading is often deliberately aggressive, because affinity medium is the most expensive item in the process and a one to two per cent product loss is cheaper than an extra cycle.
Where service life matters, the honest basis is the end-of-life capacity rather than the new-medium value with a discount applied — capacity drifts down over a lifetime, and a margin that was meant to cover that drift has usually been spent on something else by cycle two hundred. And whatever you settle on, record the residence time next to it. A capacity figure that travels through a project without its residence time becomes folklore within two months, and the first person to run the step faster discovers the omission the expensive way.
When is it worth measuring it yourself?
Not routinely. The manufacturer measures DBC on a defined feed under defined conditions precisely so that you do not have to. These are the cases where your own measurement earns its day:
- Your feed differs substantially from the test conditions — a crude harvest against a purified protein, an unusual buffer, a titre far from the one on the datasheet.
- Your molecule is not the datasheet molecule. A bispecific, a fragment, a VLP or a viral vector is a different size and a different pI, and that is a different question from a different feed.
- You intend to run at a residence time or a bed height the datasheet does not cover. Bed height is the one people forget, and it is the usual reason a figure does not survive a scale-down.
- You are running a lifetime study, since the drift over cycles — not the starting value — is what sets the service life.
- You have to defend the loading in a regulatory filing. A supplier value is a specification, not evidence for your process.
- Something is wrong: yield down, breakthrough early, fouling suspected after a cleaning campaign.
How is a breakthrough curve measured — and where does the arithmetic go wrong?
The measurement is a frontal breakthrough run: equilibrate, establish the UV signal of the undiluted feed, load continuously at a fixed flow rate while recording UV against volume, and read the volume at which the outlet reaches your breakthrough level. Run it at two or three residence times rather than one — a single point tells you almost nothing about how the medium will behave at a different flow rate.
Three traps, in the order in which people fall into them.
- Check that the UV signal of the undiluted feed is still in the linear range of the flow cell. At normal feed concentrations UV at 280 nm saturates in a standard cell, and above saturation the C/C₀ axis is not linear — so the 10 % point is not where it appears to be. Use a shorter path length, a dilution series or a less absorbing wavelength.
- Determine the hold-up volume from a run without the column, and read it at fifty per cent of that curve rather than at the first rise. The system volume causes both a delay and a dispersion; subtracting it corrects the delay only, and on a small bed the system volume is large enough relative to the bed that getting it wrong flatters the result noticeably.
- Do the mass balance, not the rectangle. This is the common mistake and it always goes the same way: taking feed concentration times breakthrough volume counts the product that has already escaped in the flow-through as if it were bound. The bound mass is the mass fed up to the breakthrough volume minus the area under the breakthrough curve up to that point, divided by the bed volume. The error is a few per cent when the curve is sharp and ten per cent or more when it is broad — which is precisely the situation in which you are measuring it yourself.
Frequently asked questions
What is dynamic binding capacity?
The amount of target a medium binds under flow before a defined share of it breaks through at the outlet, in milligrams per millilitre of bed. It is measured at a specific residence time, bed height, feed concentration, buffer, temperature and breakthrough level, and it is only meaningful together with those conditions.
What does DBC at 10 % breakthrough mean?
It is the capacity at the point where the target concentration leaving the column reaches 10 % of the feed concentration. Ten per cent is the usual reporting point because it sits on the steeply rising part of the curve, where it can be read reproducibly, while still leaving a margin before significant product loss. Figures at 1 % or 5 % are lower for the same medium and are not interchangeable with it.
What is the difference between static and dynamic binding capacity?
Static capacity is what the medium binds given unlimited time, measured in batch; dynamic capacity is what it binds under flow, where the target has only the residence time to reach a binding site. Static capacity is the upper bound in principle, but two figures from two datasheets are not automatically in that order, because the batch experiment may have depleted the protein far below the feed concentration.
Does bed height affect dynamic binding capacity?
Yes, and it is the condition most often forgotten. The mass transfer zone has a roughly fixed length at a given residence time, so on a short bed it occupies a larger fraction of the bed and the measured capacity is lower. This is why a datasheet figure measured on a 10 to 20 cm bed does not transfer to a short guard column or to a scale-down model without checking.
Can I compare DBC values from two datasheets?
Only if the target molecule, feed concentration, buffer, residence time, bed height, temperature and breakthrough level all match. They usually do not. If a figure is quoted without its residence time it cannot be compared at all — and a capacity comparison in a sales document that omits the conditions is not a comparison.
Do I need to measure DBC myself?
Usually not. The manufacturer measures it and you work with the value plus a loading margin. Measure it yourself when your feed or your molecule differs from the test conditions, when you plan to run at a residence time or bed height the datasheet does not cover, when you are running a lifetime study, when you have to defend the loading in a filing, or when something has gone wrong.
How much of the capacity should I load?
Against the breakthrough level the figure refers to. A well-characterised capture step commonly runs at 80–90 % of a capacity quoted at 10 % breakthrough; a variable feed or a figure quoted at 1 % justifies more margin. Where service life matters, calculate against the end-of-life capacity rather than the new-medium value.
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.