Membrane Engineering: Progress and Potentialities in Gas Separations
291
Signifi cant efforts have been made for increasing O
2
/N
2
selectivity of the polymeric
membranes; these two molecules have a very close kinetic diameter (nitrogen, 3.64 Å ;
oxygen, 3.46 Å ), thus the use of a membrane whose selective properties are related only
on the molecular size selection was very diffi cult. As reported by Baker [7] , the fi rst
membranes used for this separation showed an O
2
/N
2
selectivity of ca. 4. Signifi cant
improvement in membrane selectivity allowed values ranging from 8 to 12 to be reached,
implying relevant reduction in compressor size and equipment costs.
Today, nitrogen separation by membrane systems is the largest GS process in use.
Membrane selectivity does not need to be high in order to produce a relatively pure
nitrogen stream, thus they became the dominant technology instead of pressure swing
adsorption (PSA) or cryogenic distillation.
At the moment, thousands of compact on - site membrane systems generating nitrogen
gas are installed in the offshore and petrochemical industry. Air Products Norway has
delivered more than 670 PRISM ® systems producing N
2
for different ship applications,
and more than 160 PRISM ® systems for offshore installations [38] . In December 2006,
Air Products started with another PRISM ® production plant in Missouri (USA) [39] .
Another new air separation unit with a capacity of 550 ton/day of oxygen was installed
by Air Liquide in Dalian (China) [40] . In Japan, Ube Industries [41] is increasing the
production of polyimide hollow fi bres for nitrogen separation to introduce a number of
ethanol refi ning plants, mainly in the USA and Europe, driven by the rapid increase in
the demand for bio - ethanol as an additive for oil products.
Concerning oxygen separation (Figure 14.4 ), great improvements in membrane per-
formances are required to produce oxygen - enriched air and, moreover, pure oxygen to be
used for chemical industries, electronic fi elds, medical fi elds, etc. The fi rst application of
membrane technology was carried out in 1980 using ethyl cellulose membranes, but the
performance was not good enough to make the process competitive [7] . A compressor on
the feed stream or a vacuum pump on the permeate side is required in order to guarantee
the necessary driving force for the permeation of the oxygen through the membrane. Both
these solutions are expensive, thus, for making this operation cheaper high fl ux mem-
branes are required. Depending on the quality of the oxygen to be obtained, the separation
process can be developed in one or two stages (Figure 14.5 ). Membranes are economically
convenient when the fi nal O
2
concentration is in the range 25 – 50% [42] ; in this case, a
single stage (Figure 14.5 a) is the most economic confi guration [43] . Pure oxygen can be
produced in two stages (Figure 14.5 b), as proposed by Baker [9] . However, many studies
Do'stlaringiz bilan baham: