Membrane Gas Separation

CO
2
 Permeation with Pebax®-based Membranes for Global Warming Reduction
263
Table 13.2 CO
2
and N
2
permeability (  P  ), diffusion (  D  ) and sorption (  S  ) coeffi cients obtained for several Pebax ® grades by using the gas 
permeation method in time - lag mode (under an upstream pressure of 4 bars and at 25 ° C) 
Pebax ® series
11
13
31
33
Pebax ® grade
1657
1878
1074
3000 * *
1041
6100 * *
1205
Pa/PE block nature
PA6/PEO
PA6/PTMO
PA12/PEO
PA12/PTMO
PA/PE wt.%
50/50
66,7/33,3
50/50
75/25
50/50
PA/PE block M 
w
*
1.5PAr/1.5PEr
–
1.5PAr/1.5PEr
4.5PAr/1.5PEr
1PAr/1PEr
P
CO2
(a)
97.9
±
0.4
9.6
±
0.1
25.5
±
0.2
45.8
±
0.3
23.3
±
0.1
100
±
1
93.0
±
0.5
P
N2
(a)
2.04
±
0.03
2
0.583
±
0.007
2.86
±
0.09
0.7
±
0.1
2.44
±
0.07
3.43
±
0.06
 
α
 
CO2/N2
48.0
4.8
43.7
16.0
33.3
41
24.5
D
CO2
(b)
5.0
±
0.1
0.38
±
0.01
1.74
±
0.04
4.4
±
0.4
1.7
±
0.1
6.1
±
0.1
6.32
±
0.04
D
N2
(b)
4
±
3
0.24
0.4
±
0.1
5
±
1
0.11
±
0.04
9
±
6
13
±
4
 
α
 
DCO2/DN2
1.3
1.6
4.4
0.9
15.5
0.7
0.5
S
CO2
(c)
197
±
6
238
±
7
147
±
3
106
±
11
139
±
9
164
±
3
147
±
1
S
N2
(c)
8
±
6
81
16
±
4
6
±
2
72
±
23
3
±
2
2.9
±
0.8
 
α
 
SCO2/SN2
25
2.9
9.0
18
1.9
55
51
(a) Barrer or 10 
−
10
cm 
3
(STP) cm cm 
−
2
s 
−
1
cmHg 
−
1
.
(b) 10 
−
7
cm 
2
s 
−
1
.
(c) 10 
−
4
cm 
3
(STP) cm 
−
3
cmHg 
−
1
.
* Pebax ® are multiblock copolymers whose PA (polyamide) and PE (polyether) blocks alternate in the linear chain of the copolymers. The weight ratio of the PA to the PE phase 
of different grades was obtained by alternating blocks of pre - fi xed molecular weights. The relative molecular weight of the PA (PE) block of a grade is expressed here as a multiple 
of a reference molecular weight of the polyamide, PAr (or polyether PEr) block.
* * Pebax ® 3000 and 6100 are obtained by incorporating a polymer additive to the 1074 and 1041 grades, respectively (undisclosed data).  

---

264
Membrane Gas Separation
to the quasi - pure polyether or polyamide phases. It should be noted that glass transition 
of the polyamide phase was not clearly detected in the thermograms, probably because 
of the weak change in heat capacity associated to the polyamide phases. The clear detec-
tion of the separated melting of the polyether and polyamide crystalline phases, on the 
one hand, and the absence of intermediate glass transition temperatures in - between those 
of the two polymers, on the other hand, indicate that the Pebax ® fi lms consist of two 
phases, polyamide and polyether phases, without any polyether - PA blend phase. Indeed, 
it is well known that, on the one hand, the crystalline phase of a polymer consists of only 
(a)
0
0
10
20
30
40
50
60
Pebax® 1074
Pebax® 1657
Pebax® 1074
Pebax® 1657
70
20
40
60
CO
2
pressure/kPa
CO
2
concentration/
μ
mol.cm
–3
N
2
concentration/
μ
mol.cm
–3
80
100
120
(b)
0
–0.5
0
0.5
1.5
2
1
20
40
60
N
2
pressure/kPa
80
100
120
Figure 13.3 Sorption isotherms for (a) CO 
2
and (b) N 
2
on Pebax ® 1657 and 1074 at 
25 ° C

---

CO
2
 Permeation with Pebax®-based Membranes for Global Warming Reduction
265
one type of segments, since a stereo - regularity of their repetitive unit is required for the 
segment folding into lamellar crystal structure, and on the other hand, the glass transition 
temperature of an amorphous phase of two miscible polymer segments would be inter-
mediate between that of the two polymers.
The polymer crystallinity is an important parameter in permeation: the crystallites are 
not permeable to any species, and their distribution in the polymer materials defi nes the 
(a)
PA12
PA12
6100
1041
3000
1074
PA6
250
200
150
100
50
T/°C
T/°C
1657
1878
1205
PTMO
Heat Flow
Heat Flow
–50
–100
0
250
200
150
100
50
–50
–100
0
(b)
PEO
PEO
Figure 13.4 DSC thermograms (second heating scans in temperature ramps from
−
 90 to 
220 ° C) for (a) Pebax ® 1657, 1878, 1205 and (b) Pebax ® 1074, 3000, 1041, 6100 (curves 
from the bottom to the top in the graphs). The melting temperatures are approximately 
marked on the abscissa axis for comparison

---

266
Membrane Gas Separation
diffusion paths of the permeating molecules in the amorphous/molten polymer phases. 
The crystallinity of the PA phase can be calculated by assuming its melting enthalpy is 
identical to that of PA homopolymers. It falls within the 0.22 – 0.40 range for the different 
Pebax ® grades (except the 6100 grade whose crystallinity is very low due to the presence 
of an undisclosed molecular additive). Rather low crystallinity of the PA phase in Pebax ®
in comparison with that of the homopolyamide (larger than 0.4) is also consistent with 
the low molecular weight of the blocks and the less ordered crystallite structure. The 
polyether phase, whose melting point is lower than room temperature, was in the molten 
state at room temperatures, i.e. in our permeation experiments. 
AFM patterns revealed the complex nature of the Pebax ® fi lm surface. The phases 
whose stiffness was probed by the cantilever tip in friction or defl ection mode appeared 
in better contrast than in contact mode. Figure 13.5 shows the soft (molten phase) poly-
ether phase dispersed in a stiff polyamide matrix (crystallites intermingled with a glassy 
amorphous phase) for Pebax ® 1657 membrane. The Pebax ® 1657 grade appears also in 
optical microscopy images as a bi - continuous material with two phases (Figure 13.6 ). 
The polyamide phase consists of thread - like bundles (bright relief phase with high aspect 
ratio and no curvature, Figure 13.5 ) of folded lamellar PA blocks. In - between were the 
fragments of the amorphous phase of random morphology. There were phases with a 
difference in stiffness in the amorphous phase (which appear in different colours inside 
the plane phase). These two phases of the amorphous part must be the amorphous polya-
mide, and the molten polyether phase.
As the size and distribution of the three phases change with the nature and the volume 
ratio of the phases, the gas permeability could be quite different for materials of close 
chemical nature and crystallinity. Unfortunately, there is apparently no correlation 
between the gas permeability and the PA phase crystallinity. The overall PA crystalline 
phase fraction in the Pebax ® fi lms did not vary very much with the Pebax ® grades (except 
for the 6100 and 1205 grades). The apparent absence of correlations between the gas 
permeability and the PA phase crystallinity could mean that the polyamide phases (inter-
mingled amorphous and crystalline phases) did not play a crucial role in the permeation. 
Table 13.3  Values of the glass transition temperatures of PE blocks and the melting 
temperatures of PE and PA blocks for the different Pebax ® grades, determined from DSC
measurements (in the fi rst heating scans). The PA crystallinity is obtained with the values 
found in the literature for the melting enthalpy of PA 6 and PA 12 
T
m
polyamide 
block ( ° C)
T
m
polyether 
block ( ° C)
T
g
polyether 
block ( ° C)
wt.% PA 
crystallinity 
in Pebax ®
wt.% PA 
crystallinity 
in PA block
1657
208
14
−
55
20
40
1878
198
–
–
18
27
1074
173
6
−
60
19
38
3000
171
5
−
60
18
36
1041
172
5
−
60
22
29
6100
160
3
−
60
9
12
1205
147
–
−
40
11
22

---

CO
2
 Permeation with Pebax®-based Membranes for Global Warming Reduction
267
However, in addition to the three phases indicated above, there are interphase zones which 
may contribute signifi cantly to the gas permeation. The interphases that connect the 
crystalline PA phase to the amorphous PA phase and to the molten polyether phase could 
behave as a quasi - crystalline phase, or as a liquid - like phase. In polyethylenes, the con-
straints due to the crystallites reduce the molecular diffusion, and make it more selective 
[32] . In a recent study, we pointed out the signifi cant contribution of these constrained 
interphases to the permeation in nanocomposite materials with a semi - crystalline polya-
mide 12 matrix [33] . Although the existence of such an interphase can be hardly proven, 
the result analysis based on the idea of coexistence of two amorphous fractions ( ‘ real ’
(a)
0
Data type
Z range
Height
500.0 nm
5.00 
μ
m
0
500.0 nm
250.0 nm
0.0 nm
Data type
Z range
Friction
0.5000 v
5.00 
μ
m
(b)
Figure 13.5 Surface morphology of Pebax ® 1657 membrane obtained by AFM in contact 
(a) and in friction (b) modes. The contact mode images give the surface topology
0
100
m
m
Figure 13.6 Optical microscopy image of Pebax ® 1657 membrane

---

268
Membrane Gas Separation
and
‘ 
semi 
- 
ordered 
’ 
) may give a broader understanding of the relationship between 
polymer history and properties [34] . 
The permeation of gases in such a complex structure is very diffi cult to model due to 
the lack of information on the phase structures and properties, as well as the complexity 
of such modelling. Qualitatively, the reduced mobility and the chain orientation in semi -
ordered interphases due to the stiff and ordered crystallites would make the permeability 
smaller. For the Pebax ® grades with shorter polyether blocks and longer polyamide 
blocks, the tortuosity of the diffusion path will increase sharply when the polyether and 
amorphous PA phases become fi nely divided by the crystalline phase. Nevertheless, we 
tried to use the PA phase crystallinity to simulate the CO 
2
and nitrogen permeabilities in 
Pebax ® fi lms with the simple ‘ resistance model ’ [35] to estimate the infl uence of the 
Pebax ® structure on the permeability. 
We assume that there were only two phases, impermeable PA crystalline phase and 
permeable molten polyether phase, which are both ‘ continuous ’ (percolated phases). The 
model in this case consists of two continuous phases in parallel paths of permeation. The 
overall permeability for such a case in the resistance model is given by:
P
P
P
Pebax
PE
PE
PA
PA
=
+
Φ
Φ
where P
Pebax
is the Pebax ® permeability,  
Φ

i
  are the volume fractions and P 
i
  are the per-
meability of the polyether (PE) and polyamide (PA) phases. This is a simple way to 
estimate the contribution of each phase to the global permeability in a block copolymer 
[36] . 
To simplify, we supposed that the CO 
2
permeability in PA phase was at a constant 
value of ca. 0.5 Barrer (compilation of various literature data [37,38] ). The CO 
2
permeabil-
ity in PE phase was estimated by extrapolating the CO 
2
permeability obtained for different 
blends of PEG 300 – Pebax ® 1657 to 100% of PEG; we found for it a value of 275 Barrer, 
which is consistent with the literature [39] . 
The obtained simulation results are given in Table 13.4 . The values of the permeability 
calculated in taking into account the polyether and PA volume fractions indicate that the 
1657 and 1205 Pebax ® grades had a permeability of the polyether phase 30% lower than 
that of pure PEO. Such reduction in permeability of the polyether phase can be attributed 
to its confi nement in the polyamide phase. The higher permeability of the polyether phase 
in the 6100 grade compared with that of pure PEO is probably due to its additive. The 
lowest permeability of the polyether phase in the 1878 grade is easily interpreted by its 
highest content in polyamide, and its short polyether block. Its polyether phase is likely 
confi ned as very small fragments in the stiff semi - crystalline PA matrix, leading to a high 
Table 13.4 CO
2
permeability coeffi cient of different Pebax ® grades and values of the CO
2

permeability coeffi cient for the polyether phase calculated with the resistance model 
1657
1878
1074
3000
1041
6100
1205
P
CO2
(Pebax ® ) (a)
97.9
9.6
25.5
45.8
23.3
100
93
P
CO2
(polyether block) (a)
195
28
51
91
92
399
186
(a) 10 
−
10
cm 
3
(STP) cm cm 
−
2
s 
−
1
cmHg 
−
1
.

---

CO
2
 Permeation with Pebax®-based Membranes for Global Warming Reduction
269
tortuosity of the diffusion path and a lowered mobility in ‘ semi - ordered ’ interphase zones. 
However, if the polyether block is longer, its fragments, and thus the permeability, would 
be larger: this was probably the case of the Pebax ® 1041 grade, that was three times more 
permeable to CO 
2
than the 1878 grade, in spite of the lower PA content of the latter. The 
Pebax ® 1041 grade also had better diffusivity selectivity than the 1074 grade, owing to 
its higher PA content. It is diffi cult to discuss further the structure – property relationship. 
For more advanced simulations, a networked - resistance model or a molecular dynamic 
model could be used, but they required details on the morphological structure, as well as 
their permeation or molecular properties, which are not easy to determine.
The infl uence of temperature on the transport properties of CO 
2
in Pebax ® 1657 and 
6100 extruded fi lms was studied in the 20 – 40 ° C range. Their negative sorption enthalpies 
( 
−
25 and
−
5 kJ mol 
−
1
, for 1657 and 6100, respectively) can be expected from fundamentals 
of sorption thermodynamics in polymers [40] . Negative sorption enthalpy reduces the 
membrane selectivity for gas treatments at higher temperature. 
In summary, the gas transport in Pebax ® membranes depends not only on the fractions 
and the nature of the constituting chemical blocks (i.e. the polar group content) but also 
on the length of the blocks. All these parameters are likely to control the phase structure 
and properties, such as the compactness (i.e. the free volume) of the permeating phase 
and the detailed organization of the polymer segments in this phase that governs the gas 
mobility. The polyamide crystalline phase affects these structural properties, and thus the 
gas permeability, via its interfacial actions on the polymer segment in the vicinity of 
the phase interfaces and via its distribution in the membrane. The Pebax ® 1657 grade 
offered the best compromise in different structural factors for CO 
2
extraction, with a CO 
2
permeability of 100 Barrer and an ideal selectivity for carbon dioxide relative to nitrogen 
 
α
 
CO2/N2
of 50.