Membrane Gas Separation

11.3.2
Nitric Oxides 
Combustion with air at high temperature results in the generation of nitrogen oxides, NOx, 
of which the dominant component is NO and to a lesser extent NO 
2
[25] . There appears 
to be no information on the permeability of NO in polymeric membranes reported in the 
literature. Based on kinetic diameter (Table
11.1 
), NO should diffuse faster through 
polymeric membranes than CO 
2
; however, the larger difference in critical temperature 
suggests that CO 
2
will sorb more strongly in the membrane than NO, and so this process 
should dominate in sorption selective membranes. For NO 
2
, only polytetrafl uoroethylene 
has been documented, with a permeability of 15.9 Barrer and NO 
2
/CO 
2
selectivity of 1.6 
[31] . The plasticization effects of NOx are unknown, though in the presence of water 
NOx forms nitric acid that will degrade polymeric membranes, similar to that reported 
for SOx.
11.3.3
Hydrogen Sulfi de 
Hydrogen sulfi de is present within natural gas and biogas, as well as being found in suf-
fi cient quantities in syngas to warrant removal [32,33] . The permeability of H 
2
S through 
a range of polymeric membranes and their selectivity to CO 
2
is shown in Figure 11.3 . 

---

The Effects of Minor Components on the Gas Separation Performance
213
Many of these membranes are designed for sweetening of natural gas and therefore exhibit 
high permeabilities for H 
2
S and CO 
2
with large selectivity relative to methane [34] . A 
large number of membranes demonstrate selectivity towards H 
2
S over CO 
2
, attributed to 
the higher condensability of H 
2
S (for critical temperature see Table 11.2 ) within the 
membrane, compared to CO 
2
. Similar to SO 
2
, H 
2
S has been reported to plasticize or swell 
polymeric membranes [35] .
11.3.3.1
H 
2
 S Competitive Sorption into Polydimethylsiloxane (PDMS) 
In our recent work, a PDMS membrane (thickness of
∼
10
μ
m) was exposed to a 90% 
N 
2
– 10% CO 
2
mixture with 500 ppm H 
2
S present, to simulate syngas - like conditions [36] . 
The resulting permeability of CO 
2
and N 
2
through PDMS at different temperatures and 
pressures, with and without H 
2
S present, is shown in Figures 11.4 and 11.5 respectively.
In the presence of H 
2
S, CO 
2
permeability through PDMS decreases by an average
∼
8% 
(Figure 11.4 ), implying that H 
2
S competes with CO 
2
for transport through the membrane. 
Conversely, N 
2
permeability in general increases slightly when H 
2
S is present (Figure 
11.5 ). While this increase is often within the error of the measurement, it implies that 
H 
2
S enhances either the solubility or diffusion of N 
2
within PDMS. Similar behaviour has 
been measured for H 
2
permeability through PDMS in the presence of 3% H 
2
S [37] . This 
was attributed to plasticization (swelling) of the polymer matrix by H 
2
S reducing the 
resistance to gas transport leading to greater fl ux. However, the H 
2
S concentration here 
is in comparison quite low at 0.05%.
11.3.3.2
H 
2
 S Competitive Sorption into Polysulfone and Matrimid 5218 
The infl uence of H 
2
S on the CO 
2
permeability in the glassy polymeric membranes poly-
sulfone and Matrimid 5218 (a polyimide) are shown in Figures 11.6 and 11.7 respectively, 
where the membrane was exposed to 90% N 
2
– 10% CO 
2
gas mixture with H 
2
S at 500 ppm.
Two processes are occurring within these membranes upon exposure to H 
2
S. Firstly, 
competitive sorption between H 
2
S and CO 
2
occurs within the microvoids, where H 
2
S will 
replace some adsorbed CO 
2
, and reduce the concentration of CO 
2
within the membrane. 
This correlates to a reduction in the permeability of CO 
2
, which is evident in the time -
resolved data for all three glassy membranes upon exposure to H 
2
S, at every temperature 
and partial pressures. Hence, H 
2
S strongly competes with CO 
2
for space within the 
Langmuir voids, given that H 
2
S is present in trace amounts. The second process, plasti-
cization of the membrane by H 
2
S, results in the increase in the measured permeability at 
a later time for both membranes under certain conditions, e.g. relative high temperatures 
and high H 
2
S partial pressures, as the diffusivity of gas increases. Plasticization is delayed 
because adsorbing H 
2
S requires time to build up the necessary concentration, due to 
competitive sorption, to have a signifi cant plasticization infl uence on the polymeric 
matrix. The observed plasticization is attributed to the presence of H 
2
S only, and not CO 
2
. 
This assumption is supported by the partial pressure of CO 
2
in all experiments being less 
than that require to plasticize each polymer, for polysulfone 34 bar CO 
2
at 21 ° C [39] is 
required and for Matrimid it is 14.8 bar CO 
2
at 35 ° C [40] . Secondly, each membrane is 
initially exposed to the CO 
2
/N 
2
mixture, at the beginning of the experiment for 2 hours. 
Therefore, any plasticization by CO 
2
should have occurred during this time period, and 
all later time effects are attributed to H 
2
S only. Hence, competitive sorption of H 
2
S occurs 
for both glassy membranes at the studied temperatures and pressures, but each polymer 

---