Membranes - Introduction to Carbon Capture and Sequestration - page 307

The conclusion we gather from Figure 7.3.6 (a) through (c) is that the
pressure ratio that can be reached in practical terms places serious limits
on the separation one can achieve. For fl ue gasses from a coal-fi red
power plant the maximum separation is about 50%. Conservation of
mass is most unforgiving: all the CO 2 and all the N 2 have to pass through
the membrane. If we increase the selectivity of the membrane by decreas-
ing the permeation of N 2 , we have to increase our area to maintain the
fl ux of N 2 . Changing from using compression on the fl ue side to using
vacuum on the permeate side or changing the membrane design does
not solve this pressure ratio issue. One solution is to use two or more
membranes. This, however, would require compressing the fl ue gas or
pulling a vacuum twice. The energy requirement of such a two-stage
process would require a too large fraction of the electricity production!
Now we can do some magic. Let us look at the design in Figure 7.3.6 (d) .
Here we use some of the retentate (6%), at a reduced pressure and feed
it back to the permeate side of the membrane as a sweep gas. Why would
this be a good idea? This looks like mixing a waste stream with the prod-
uct! However, if we run the numbers we see a reduction in the total area
of the membrane required. How can this be? If we look at the design of
Figure 7.3.6 (c) , we have a permeate of about 40% CO 2 and hence 60%
N 2 . Recall that a constraint in determining the area required for mem-
branes with a high selectivity is that all the N 2 has to pass through the
membrane. If we use the retentate as a sweep gas, we have a second
source of N 2 and the N 2 mass balance can now be obeyed with a much
smaller membrane area. If we examine the design in Figure 7.3.6 (d) , we
see that we have 2-3% CO 2 at the end of the counterfl ow module; this very
small partial pressure and hence concentration of CO 2 does not affect the
driving force, but now not all the N 2 has to pass through the membrane.
Separating fl ue gasses
In the previous section, we discussed some membrane design confi gura-
tions. The main conclusion is that at present the real bottleneck in the
effi ciency of operation is not the material, but the energy required to oper-
ate the compressors or vacuum pumps. Given the enormous volume of fl ue
gas we need to process we can at best obtain a pressure ratio of 5, and
hence even for the best membrane materials the concentration of CO 2 in
the permeate will be maximally 50-60%, and thus will fail to reach the