very low levels, incorporate a safety
factor of at least two extra diavolumes
and test the process to ensure that
actual residual levels are acceptable.
ProFlux. M12 Benchtop TFF system with
spiral wound modules
Characterization of
Performance
Although the above discussion gives
general guidelines on how to choose
an appropriate module and operating
parameters, the performance of the
process must be tested on the actual
feedstock. One of the most important
experiments for characterizing
performance is to generate flux versus
TMP curves at several crossflow rates
(or pressure drops) and several protein
concentrations and to determine
product retention at each point. In
addition, if the process contains a
diafiltration, it is important to generate
these flux versus TMP curves in both
the starting and final buffers, since flux
and retention can change significantly
with buffer conditions. If required, the
effects of processing at different
temperatures can also be
incorporated. With a small volume of
feedstock and a single day’s work,
this experiment generates a wealth of
information about the process. The
experiment will be briefly described
here.
Typically, determine TFF
performance at approximately three
different crossflow rates that span the
range of manufacturer recommended
rates for the module being used.
Likewise, approximately three different
protein concentrations should be
tested that span the range from initial
protein concentration in the feedstock
to the highest concentration expected
in the process. Investigate at least five
transmembrane pressures for each
crossflow and protein concentration.
TMPs will vary depending on the
membrane module and the feedstock,
but will typically be in the range of 5
to 50 psid.
Perform the experiment by starting
up the module in a total recycle mode,
where both the retentate and the
filtrate lines are directed back to the
recycle tank. Set specific flow,
pressure, concentration, and
temperature conditions. After the
module has equilibrated at the
conditions, record the flows and
pressures and collect small samples of
the feed and filtrate streams for
analysis of protein concentration.
Then, apply new conditions and
repeat the procedure.
The method of startup and the order
of conditions tested can impact the
results, so take care to always begin
with the least fouling conditions and
move towards more fouling
conditions. During startup of the
operation, first slowly ramp the feed
rate (and co-flow rate, if applicable)
without any applied pressure. When
the feed rate setpoint is reached,
ramp the applied pressure to its
setpoint. Finally, if filtrate control is
being used, ramp the filtrate to its
setpoint. Shutdown of the operation
should occur in reverse order from the
startup.
When testing different flow,
concentration, and pressure points,
conditions that are least fouling are
those at low protein concentrations,
low TMPs, and high feed rates. A
good approach is to start with the
highest feed rate and lowest protein
concentration and TMP to be tested.
At constant feed rate and protein
concentration, increase the TMP until it
begins to level off. At this point, the
membrane is operating in the pressure
independent regime (see figure 9) and
higher TMPs will cause excessively
high protein concentrations within the
module without the benefit of
increased flux. Maintaining the protein
concentration constant, repeat the
TMP excursion (low to high TMP) at
each feed rate to be tested, moving
from high to low feed rates. Then,
increase the protein concentration and
repeat the entire procedure.
A sample sheet for data collection
is illustrated in figure 15. For each test
point, calculate flux, TMP, and
retention. Then, generate graphs
showing flux versus TMP at different
crossflow rates and protein
concentrations, and retention versus
TMP at different crossflow rates and
protein concentrations. From the
retention data, calculate the predicted
yield losses as described by the
equation shown in figure 6. The
collection of this data enables the
choice of successful and robust
operating conditions.
Experiment Title: Sample Experiment Lab book Reference: 10739-25 Date: 04/25/99 Operator: JCT
Objective: Determine operating parameters Feedstock Product and pool: Protein Y IEX pool Feedstock lot #: 10739-18
Membrane Material, MWCO: PLCGC Membrane Area [m2]: 0.1 Device Holder: Pellicon-mini Device lot #:
