Initial Evaluation
for Ultrafiltration

---

The main components of an ultrafiltration process are the membrane material and the module format. If you choose the most appropriate components early in the development process, with consideration for the process requirements, you will increase the likelihood of success and the robustness of the final step.

Select Your Membrane Materials

Regenerated cellulose (refer to our Ultracel® composite regenerated cellulose membrane) and polyethersulfone (PES) (refer to our void-free Biomax® membrane) are two of the most common materials used for ultrafiltration membranes.

When choosing membrane materials:

Consider:	Suggested Membrane:
Does the process require organic solvents or hydrophobic antifoam agents? Is your initial process solution very dilute?	Regenerated cellulose membranesare indicated for applications where: Harsh pH conditions are not needed (pH range 2-13) Organic solvents or antifoams are present Protein loading is low (<20 g/m2 ) or the feedstock is highly fouling
Do you plan to use harsh chemicals as part of your cleaning process? Can you use bleach within your facility?	Polyethersulfone membranesare suggested for applications where: Very harsh pH conditions (pH range 1-14) are required for processing or cleaning Moderately high protein loading (>20 g/m2) is required to minimize adsorption losses

Determine Your Membrane Nominal Molecular Weight Limits (NMWLs)

Since NMWLs for ultrafiltration membranes do not indicate absolute retention/sieving ratings, these guidelines can help you determine what membrane rating is applicable for a particular process.

• The general rule of thumb is to select a membrane that has a NMWL one-third to one-fifth of the molecular weight of the product to be retained (some MAb processes have shown protein passage with the use of Biomax® 50; therefore, lower NMWL Biomax® should be evaluated). If the product must pass through the device to the filtrate, select a NMWL three to five times larger than the product.
• Generally, a minimum size difference of approximately five-fold is required between components that are being separated.
• Use a membrane that has sufficiently high retention to meet your yield goal. This is determined during optimization trials and will address the following important variables:
  
  
  • Impact of transmembrane pressure (TMP) and feed flow rates on process flux (the flow rate normalized for the area of membrane through which it is passing) and retention.
    
    

    Transmembrane Pressure: The average applied pressure from the feed to the filtrate side of the membrane. TMP [bar] = [(PF + PR)/2] - Pf

  • Impact of product concentration and buffer conditions on process flux and retention.
  • Impact of diavolumes (measures of the extent of washing that has been performed during a diafiltration step) on buffer exchange and contaminant removal.
    
    
• Since each protein feedstock and process is unique, you may need to test two or more membranes before you choose an optimal one.

Effect of NMWL and Membrane Retention on Yield: Product loss to the filtrate due to incomplete retention can be cumulative for the concentration and diafiltration sections of a process. To understand retention changes for a batch UF and constant DF process, where retention remains constant, use this calculation:

Product Loss in Permeate % = 100* (1-e(R-1)InVCF+N) )

Bear in mind that a protein’s stated molecular weight can change based on pH/ionic buffer conditions, protein-to-protein interactions and protein-contaminant interactions. During process optimization, you should characterize the feedstock solution to fully understand this potential impact on retention and membrane selection.

Choose Your Modules

The most common ultrafiltration module formats are:

• Flat plate
• Spiral wound
• Hollow fiber

Screens are often inserted into the feed and/or filtrate channels in spiral wound and flat plate modules to increase turbulence in the channels and reduce concentration polarization. The turbulence-promoted channels have higher mass transfer coefficients at lower feed flow rates, meaning that higher fluxes are achieved with lower pumping requirements. Turbulence-promoted feed channels are, therefore, more efficient than open channels. Hollow fiber modules do not have the option of turbulence-promoting channels.

Flat plate modules generally have higher packing densities (area of membrane per area of floor space), allow linear scaling and some offer the choice of open or turbulence-promoted channels. Spiral wound and hollow fiber modules are not linearly scalable. Learn about Pellicon® ultrafiltration cassettes for demanding filtration processes.

Contact the Ultrafiltration Team to explore how we can help you avoid costly errors by verifying that your design falls within Merck’s guidelines (actual process performance to be verified with optimization and process simulation trials).