Depth Filtration

One of the more readily implemented cell separation techniques is normal-flow depth filtration. Depth filters employ the conventional filtration technique of separating cellular solids from the cell culture fluid by forcing the liquid through a porous medium in dead-end or normal-flow mode. The medium retains solids and the liquid flows through to a collection vessel. Especially well suited for smaller batch volumes (>2000 l), depth filtration can be an economical and easy-to-use primary cell removal method compared with tangential-flow microfiltration and centrifugation (Singhvi et al. 1996). Depth filtration is economical due to the lower capital costs as well as the low cost of the single-use cartridges. Ease of use is attributed to the normal-flow operation that can be performed either at constant flux with a single feed pump, or at constant pressure by pressurizing the bioreactor to drive the product into the filters. The process can be monitored simply with pressure or flow measurements, eliminating the need for a complicated control system and costly auxiliary instruments and equipment. Finally, the single-use nature of the cartridges eliminates the need for cleaning and reuse validation. Therefore depth filtration is an ideal harvest method for laboratory, pilot and smaller-scale production processes.

16.3.1.1 Depth filter media and cartridges

Depth filters are constructed from cellulosic fibres, filter aids and resins, which can impart positive charge on the media. Unlike membrane filters, depth filters are graded-density filters, meaning that the pore size gradually decreases across the thickness of the medium. The medium's thickness, positive charge and graded density combine to trap cells and adsorb negatively charged cellular debris and submicron particulates. This 'dirt-holding' capacity makes depth filtration an effective separation and clarification method for biological systems. Depth filter media are available in a range of 'grades' that reflect relative porosity as measured by clean water flow rate or permeability. Examples of typical media grades used for cell capture and debris removal are presented in Figure 16.5.

Media with higher permeabilities are used for primary cell removal and media with lower permeabilities are used for clarifying TFF filtrates and centrifuge supernatants. Although the media may be referred to by pore size, these designations should not be interpreted as absolute ratings, and care should be taken when evaluating the performance of 'equivalent' grades from multiple manufacturers, due to differences in charge and potential binding capacity.

The most commonly used form of the depth filter medium is made by layering media in flat sheets onto each side of a plastic support to form one cell. Multiple cells are then stacked together to form one cartridge and multiple cartridges of the same diameter are stacked together into stainless steel housings (Figure 16.6). The standard cartridge diameters are 8, 12 and 16 inches, each containing approximately 3.5, 18 and 50 ft2 (0.3, 1.7 and 5.1 m2) of effective filtration area. Since filter housings can be manufactured to accommodate an array of configuration options, depth-filtration harvest systems are relatively easy to customize for a particular application.

Media Grade Cellulose & Inorganic Filter Aid - DE

Cellulose & Inorganic Filter Aid - HC

Cellulose - CE

Activated Carb

Media Grade Cellulose & Inorganic Filter Aid - DE

Cellulose & Inorganic Filter Aid - HC

Cellulose - CE

Activated Carb

Millipore Filtration Depth

Figure 16.5 Depth filter retention ranges. (Reproduced with permission from Millipore Corporation.)

Figure 16.5 Depth filter retention ranges. (Reproduced with permission from Millipore Corporation.)

16.3.1.2 Depth filtration process development

One of the strategies for depth filtration harvest development is to determine the optimum configuration of media grades and filtration area to achieve the desired separation. Since some media grades are designed for trapping the large, intact cells and others are designed to clarify the smaller debris, often the optimum configuration is a multi-stage system with the more open filter upstream of the tighter-grade filter. Some manufacturers have developed cartridges containing multiple media grades to enable a dual-stage separation within one housing. These multiple-media cartridge options could lead to a reduction in the required filtration area and number of housings.

Filtration Depth

Figure 16.6 Cross-section of a Cuno Zeta Plus depth filter cartridge (left) and 8-inch, depth filter cartridges (right) (reproduced with permission from Cuno Incorporated).

Gravimetric Dust Sampling Method

12-inch and 16-inch

Figure 16.6 Cross-section of a Cuno Zeta Plus depth filter cartridge (left) and 8-inch, depth filter cartridges (right) (reproduced with permission from Cuno Incorporated).

12-inch and 16-inch

In addition to media selection, key process parameters for depth filtration harvesting are filtration flow rate and pressure drop as a function of throughput. In theory, depth filtration performance can be described by Darcy's relationship between flow rate and the pressure drop created by that flow through a porous solid (Geankoplis 1983):

dV MAP

where dV/dt is flow rate, k is bed permeability, is liquid viscosity, AP is pressure drop, L is bed thickness and A is filtration area.

One of the difficulties in applying Darcy's law to biological filtrations is the inability to calculate accurately the bed permeability or the resistance (k/L) for these complex biological fluids. While empirical relationships do exist for permeability (e.g. the Kozeny-Carman expression) and resistance, it is often more expedient to measure pressure drop experimentally as a function of flow rate for animal cell culture fluids (Carman 1937).

Small-scale experiments to determine depth filtration performance are relatively easy to conduct, and provide the most reliable data when the test feedstock closely resembles the actual feedstock in terms of solids volume, culture viability and culture density. Since flux and colloid formation are affected by temperature, small-scale studies may also need to be performed at the same temperature as the production-scale operation. The general experimental approach is to test different filter media at various flow rates to determine the optimum flux, filter area and media combination. Small-scale tests may be conducted using less than a few litres of culture fluid with 47-mm filter discs at either constant pressure (Vmax) or constant flux (Pmax) to evaluate clarity, throughput and product retention (Badmington et al. 1995; Ho & Zydney 2002). The commonly used Vmax method uses the linear form of the pore-plugging model to predict filtration capacity:

' max where t is time, V is volume, Q is initial filtrate flow rate and Vm be filtered at the test pressure before the membrane is fouled.

The experiment is performed by filtering at a constant pressure and measuring the filtered volume as a function of time. If the plot of t/v versus t is linear, then the filtration follows the gradual pore-plugging model and Vmax is calculated from the inverse of the slope. Finding the area needed at production scale is then linearly scaled from the Vmax result. If a straight line is not obtained, then the solution does not follow the pore-plugging model and the constant flux method is recommended.

Constant flux experiments are conducted by controlling the flow rate and monitoring the rise in pressure across the filter as a function of filtrate volume. In addition to monitoring the decrease in flow rate for Vmax and the increase in pressure for Pmax, it may also be important to measure product yield throughout the filtration to ensure that the target molecule is not binding to the charged filter media. For example, filtration of a target protein produced in mammalian cell culture through two different depth filter media led to product yields of 95 % and 38 % under the same processing conditions (Bender 2001).

Typically for depth filtration media, Vmax testing is a good screening tool for media selection, while the Pmax approach provides more accurate results for predicting large-scale performance of a particular filtration configuration (Yavorsky & McGee 2002). Despite the reliability of small-scale

Depth Filtration
Figure 16.7 Industrial-scale depth filtration system for cell harvest and clarification. (Reproduced with permission from Cuno Incorporated).

data, the conservative process development approach is to select an appropriate 'safety factor' and oversize the production-scale filters to account for run-to-run variabilities, fluid dynamic differences and other system effects. Large-scale depth filtration harvest systems generally also include downstream bioburden reduction and sterilizing grade filters, which must also be sized appropriately in small-scale Vmax or Pmax experiments.

As shown in Figure 16.7, the largest industrial-scale depth filtration harvest systems consist of multiple housings containing multiple cartridges. The system shown is used to harvest cells from a 15 000 l production bioreactor and includes housings for guard and sterile filters. Prior to the harvest, the depth filters are flushed with water or an appropriate buffer to remove loose par-ticulates and extractables according to the manufacturer's recommendations. Once the harvest is completed, the filters are again flushed to recover the valuable product held up in the housings. By implementing a post-use flush and ensuring no product retention losses, depth filtration harvest yields exceeding 95 % are possible.

MICROFILTRATION MEMBRANES AND DEVICES

As process volumes increase, depth filtration may no longer be the most feasible harvest method due to the large filter housings, longer processing times, increased plant space requirements and laborious cartridge installation and disposal issues. Another limiting factor for depth filtration may be the increasing amount of cellular solids produced in the bioreactor that collect within the depth filter media and housings. Consider a 10 000-l bioreactor with a solids volume of 2.5 %: 250 l of cellular solids would need to be contained within the depth filter housings without blocking or fouling the effective filter surface area. These large solids volumes make depth filtration harvesting impractical relative to tangential-flow microfiltration where the cellular solids are returned to the bioreactor, or continuous centrifugation where the solids are discharged to a separate collection or to waste.

Was this article helpful?

0 0
Healthy Chemistry For Optimal Health

Healthy Chemistry For Optimal Health

Thousands Have Used Chemicals To Improve Their Medical Condition. This Book Is one Of The Most Valuable Resources In The World When It Comes To Chemicals. Not All Chemicals Are Harmful For Your Body – Find Out Those That Helps To Maintain Your Health.

Get My Free Ebook


Post a comment