A capsule filter is a filter cartridge sealed inside its own disposable housing with inlet, outlet and vent connections, so there is nothing to clean and nothing to assemble. It has taken over sterile and clarifying duty at laboratory and pilot scale for that reason, and the decisions are the same three every filtration asks: the membrane, the rating and the area.
capsule filters against a cartridge in a housing
A cartridge in a stainless housing is cheaper per square metre and needs cleaning, sanitisation and an assembly step where an operator can pinch an O-ring. A capsule arrives sterilised, is connected by tubing and discarded, which removes both the cleaning validation and the assembly error, and costs more per batch. The crossover is volume and frequency: pilot and clinical work runs on capsules, and a plant filtering the same buffer every day runs on cartridges in housings.
Membrane and rating, and what the number means
A 0.2 micrometre rating is a sterilising grade claim validated by bacterial challenge, not a pore measurement, and a 0.45 is a clarifying grade. Membrane chemistry decides binding and compatibility: polyethersulfone for low protein binding, PVDF for low binding with solvent tolerance, nylon and cellulose for aqueous work, PTFE for gases and aggressive solvents. A prefilter layer in the same capsule protects the final membrane and is what makes a difficult feed practical, and it is specified as a combination rather than added later.
Sizing area from a flow decay test
Area is sized from the feed's own fouling behavior rather than from a rule of thumb. Filter a measured volume through a small disc of the same membrane at the operating pressure, record volume against time until flow decays, and scale the throughput per square centimetre to the batch volume with margin. Two numbers come out of it: the area needed and whether a prefilter changes the answer, which it usually does. Sizing from the pump's flow alone ignores fouling and produces a filter that blocks mid-batch.
Integrity testing, and what it proves
A sterilising filter is integrity tested after use, and often before, by bubble point or diffusive flow against the manufacturer's specification for that membrane wetted with that fluid. A pass shows the membrane and its seals were intact, which is what supports the sterility claim for the filtrate; it does not show what was retained. The test needs a vent that can be operated aseptically and a record with the fluid, the temperature and the specification, which is why the capsule's vent design matters as much as its area.
A bubble point filter integrity test, and what it shows
Bubble point is the pressure at which gas first flows through a wetted membrane, and it depends on the largest pore, the wetting fluid and the temperature, which is why the specification is quoted per membrane and per fluid. A reading above the specification says the membrane and its seals were intact; a reading below says something is wrong, and the test cannot say what was retained during the filtration. Diffusive flow is the alternative for larger areas, and both need a vent that can be operated without compromising sterility.
A dialysis membrane and the cut off it is chosen by
A dialysis membrane is specified by its nominal molecular weight cut off, which is a distribution rather than a wall, so a protein near the cut off will be lost slowly over a long exchange. Membrane material decides the binding losses, and the buffer volume ratio and number of changes decide how complete the exchange is. Wetting and a rinse to remove the preservative come before the sample.
A regenerated cellulose membrane and low binding
A regenerated cellulose membrane is chosen when the sample is small and the losses on the membrane matter, since cellulose binds protein less than most synthetic materials. It is less tolerant of organic solvents than polymer alternatives, so the chemical compatibility table decides the choice as much as the pore size. For a dilute sample, a pre-rinse and a passivation step recover more than a larger membrane does.
A hydrophilic pvdf filter and the sample that decides the polymer
A hydrophilic pvdf filter is the default for aqueous samples where low protein binding matters, while the untreated polymer is used for solvents. The choice is a compatibility question first and a pore size question second. For a filtration that will be reported, the extractables data and the validation the supplier publishes matter, because a filter that leaches into the sample changes the analysis it was meant to protect.
Centrifugal filter devices and the concentration factor they can reach
Centrifugal filter devices concentrate a small volume against a membrane, and the practical limit is the point where the retentate becomes too concentrated and the membrane fouls or the protein aggregates. Spin time, temperature and the device's dead volume decide recovery, and a device left spinning to dryness loses everything. The cut off is chosen well below the protein's mass rather than close to it.
Depth filtration bioprocess steps and where they sit
In a depth filtration bioprocess the filter is a matrix rather than a surface, which is what lets it hold cells and debris after harvest without blinding immediately. It is sized by throughput per area on the actual feed rather than from a table, which is why a small scale trial precedes a purchase. Filter aid chemistry also removes some host cell protein, which is part of its value downstream.
Single use connectors and the assembly they belong to
Single use connectors are specified as part of an assembly rather than as parts: the connection type, the tubing size, the gamma dose the whole assembly is rated for and the sterile barrier claim all travel together. Aseptic connectors that make a sterile connection in an uncontrolled room are the reason these systems can be used outside an isolator, and that claim needs the supplier's validation data.
A vacuum manifold and the throughput it adds
A vacuum manifold pulls many samples through membranes or cartridges at once, which is what makes solid phase extraction and plasmid preparation practical at scale. The variable to watch is flow uniformity across positions, since an uneven manifold gives position dependent recovery that looks like sample variation. A trap between the manifold and the pump protects both the pump and the operator.
Questions people ask about capsule filter
Capsule or cartridge in a housing?
Capsules for pilot, clinical and variable work, because there is nothing to clean or assemble. Cartridges where the same filtration runs often enough for the housing's cleaning validation to pay.
How much area do I need?
From a flow decay test on the actual feed, scaled to the batch with margin, and repeated with a prefilter to see whether the combination is cheaper.
Does an integrity test prove sterility?
It proves the membrane and seals were intact against the manufacturer's specification. Sterility of the filtrate rests on that plus the process around it.