How do I choose the right membrane for a TFF system?
I’m a student getting into downstream processing and I keep getting stuck on TFF. There are so many options — UF vs MF, different MWCOs, regenerated cellulose vs PES, flat-sheet cassettes vs hollow fibers. How do I actually decide which membrane to use for a given step (say, concentrating a protein, or harvesting cells)? A simple, visual way to think about it would really help.
1 answer
Great question — this trips up almost everyone at first. The good news is that membrane selection comes down to just four decisions, taken in order. Think of the membrane as a sieve: the feed sweeps *across* its surface (that’s the “tangential” part), small stuff passes through as permeate, and anything bigger than the cut-off stays behind in the retentate. 1) UF or MF? Set by WHAT you’re separating. Microfiltration (MF, pore size ~0.1–0.65 µm) separates cells & debris from broth — used for harvest/clarification. Ultrafiltration (UF, rated by molecular-weight cut-off / MWCO, ~1–1000 kDa) concentrates and buffer-exchanges proteins, vaccines and vectors. 2) Which MWCO (for UF)? Rule of thumb: to RETAIN your product, pick a MWCO about 3–6× SMALLER than the product’s molecular weight. To PASS your product (retain something bigger), pick a MWCO ≥3× LARGER. Example: to concentrate a 150 kDa antibody, a 30–50 kDa membrane holds it back reliably while salts and buffer flow through. Too tight → slow flux and fouling; too open → product leaks into permeate (yield loss). 3) Which material? This drives protein binding (yield), flux and cleaning robustness. Regenerated cellulose (RC) = very low protein binding → best recovery, low fouling (first choice for high-value/sensitive proteins). PES = high flux, robust, wide pH/chemical range, easy to clean (great general-purpose). PVDF and others are rugged/low-binding, often for MF/harvest. Always check compatibility with your buffers and cleaning agents (e.g. 0.5 M NaOH). 4) Which format? Flat-sheet cassette = high area in a small footprint, high throughput, higher pressure — but more shear (best for robust proteins and large batches). Hollow fiber = open channels, low shear, gentle handling — the choice for cells, viral vectors and shear-sensitive product. Then size the membrane area from your target flux (LMH) and time budget, and ALWAYS confirm with a small scale-down trial before you commit. We put together an illustrated, student-friendly guide with animated diagrams that walks through exactly this — cross-flow vs dead-end, how MWCO sieving works, the material trade-offs, and cassette vs hollow fiber — here: bioprocess.bio/learn/tff-membrane-selection Short version: decide UF/MF → MWCO → material → format, then trial it. Get those four right and you’re 90% there.
Log in to post an answer.