September 1, 2014 (Vol. 34, No. 15)
Determining the Best Approach for Carrying Out Clarification Operations
Depth filtration or centrifugation? The choice is mainly a matter of economics. Depth filtration makes sense up to about 4,000 L, depending on the process, after which most biomanufacturers turn to centrifugation.
A number of factors affect economy, explains Paul Beckett, Eng.D., technology consultant at EMD Millipore. “It could be as simple as footprint. A facility may simply lack the space for the number of filters they need for their feed volumes. This is why large processes overwhelmingly use centrifugation as their primary clarification method.”
Centrifugation does not generate a fully clarified feedstream at production scale, as it does in the laboratory, which leads to what Dr. Beckett refers to as an “endless challenge.” At industrial scale, disk stack centrifugation provides around 90% clarification—not clean enough for practical membrane filtration. Thus, centrifugation-based clarification requires a secondary depth filtration step.
Even at small scale, the economics of depth filtration depend strongly on filter capacity. The 4,000 L barrier is based on an expected capacity of 150 L/m2 of filter area but, as noted, capacity may vary greatly dependent upon process stream and conditions.
Not all clarification decisions are based solely on economics. Many bioprocessors, particularly contract manufacturers, value the flexibility and speed that single-use equipment provides, and thus avoid centrifugation. “But eventually, they may reach a point where they are spending so much on filters that they might be better off with a centrifuge,” says Dr. Beckett.
For large manufacturers, a centrifuge that is already in use proves to be more economical. For smaller facilities, depth filters allow rapid validation. “Suppliers have done most of the work, and filters facilitate manufacturing speed,” continues Dr. Beckett. “Qualifying and validating a centrifuge to tight timelines can prove challenging.”
If a batch or project turns out to be larger than expected, processors can simply add more filters. When a batch is too big for a centrifuge, processors must install a much larger centrifuge or work in batches, which is expensive and time consuming and may compromise product quality. Additionally, notes Dr. Beckett, “CMOs pass filtration costs on to their clients. If they use centrifugation, the capital costs fall on them.”
Hemanth Kaligotla, marketing manager for purification technologies at Sartorius Stedim North America, is more philosophical: “Defining the problem is more productive than simply looking at the technologies’ merits and demerits.” He describes today’s high titers from fed-batch mammalian cell cultures as a “curveball” that further complicates clarification decisions: “The cell densities and the quantity of debris just keep increasing.”
Maik Jornitz, COO at G-Con Manufacturing, is perhaps less sanguine. “Both technologies are antiquated.” Centrifugation, he says, is “too complex, too tedious,” and the instrumentation is “difficult to clean.” Depth filters, by contrast, require flushing to remove leachables and extractables and are not mechanically stable. “When wetted they lose their tensile strength. If someone mistakenly raises the operating pressure, the lenticular pads can break.”
Jornitz nevertheless calls encapsulated, single-use depth filters “a big step forward. In the old days you open the stainless steel housing and all this goo comes out. It’s a mess. Single-use cartridges are easier to install and uninstall, and cleaning is obviously not an issue.”
Jornitz looks forward to the day that a novel technology will replace current clarification methods. One candidate is the Carr UniFuge centrifuge, distributed by PneumaticScaleAngelus, whose product contact surfaces are fully disposable. With a capacity of 10 to 2,000 L, the device picks up nearly at the lower volume where depth filtration and centrifugation traditionally meet. The centrifuge was first installed in 2010 at Merck and is now under evaluation at several “major biotech” companies, according to David Richardson, global manager at PneumaticScaleAngelus.
A mature, but potentially disruptive technology may bolster the case for filtration. Flocculation, a method long utilized by the wastewater industry, turns small particles that clog depth filters into larger particles that filters can more easily remove. Flocculation improves the effective capacity of specialized depth filters by up to a factor of eight, notes Dr. Beckett.
“It provides a filtrate that is capable of going through a membrane filter with only a single-stage depth filtration step. That means that depth filters can be economical, even for a 10-kL bioreactor,” he points out.
Flocculation is in the process of being validated at large scale, and bioprocessors are eager to investigate it in combination with depth filtration. Dr. Beckett expects this technique will be effective with mammalian cells and high-density/titer cell culture feeds.
Several flocculation technologies exist, including the addition of the cationic polymer polyethyleneimine, which removes negatively charged cell debris and genetic materials. Two other technologies also make depth filtration more applicable: acid precipitation resembles flocculation and dynamic body feed filtration employs a cake of diatomaceous earth that protects the filter medium from clogging.
The enemy of alternative clarification techniques that rely on particle agglomeration or precipitation is feedstream variability, which may alter the distribution of flocculants, and therefore depth filtration performance. This factor is no doubt the topic of a future discussion.