Structural Determination of Extem XH 1015 and Its Gas PermeabilityComparison with Polysulfone and Ultem via Molecular Simulation

Jianzhong Xia,†,‡ Songlin Liu,‡ Pramoda Kumari Pallathadka,§ Mei Lin Chng,‡ andTai-Shung Chung*,‡

NUS Graduate School for IntegratiVe Sciences and Engineering, National UniVersity of Singapore, 28 MedicalDriVe, Singapore 117456, Singapore, Department of Chemical and Biomolecular Engineering, NationalUniVersity of Singapore, 10 Kent Ridge Crescent, Singapore 119260, Singapore, and Institute of MaterialResearch and Engineering, 3 Research Link, Singapore 117602, Singapore

By employing high resolution 1H and 13C NMR spectroscopy combined with elemental analysis and FTIR-ATR, we have determined the basic chemical structure of Extem XH 1015, a new brand of polyetherimidewith good thermal, mechanical properties, and processability. Bisphenol-A dianhydride (BPADA) and diaminodiphenyl sulfone (DDS) are found to be the monomers for this newly developed polyetherimide. The gaspermeability of this new polymer is reported for the first time in the literature. Polysulfone (PSU) and Ultemare employed as reference samples for the elucidation of permeability and selectivity differences among thembecause of their structural similarities. In addition to qualitative comparison of chain rigidity and packingwith gas transport properties, computational simulations powered by Material Studio are performed at amolecular level to quantitatively investigate the relationship between the fractional accessible volume (FAV)and gas permeability. The FAV differences among these polymers increase with an increase in gas moleculesdiameters; thus these polymers have similar permeability for small gas molecules but diverse for large gasmolecules. Their selectivity differences are also discussed in terms of FAV ratio. The FAV concept is provedto be more effective than fractional free volume to analyze and predict gas separation performance.

1. Introduction

Throughout history, human beings have witnessed manyevolutions involving materials. Now it is the world of newmaterials, especially after the emergence of engineering plastics.Polyimide is one of these high performance engineering plasticsaiming at replacing metals as well as targeting other specialtyapplications such as electronics, automotives, and membranes.However, not all of the polyimide materials meet industrialspecifications because of their poor processability rendered byhigh chain rigidity and high glass transition temperatures. Oneof the exceptions is PEI (polyetherimide), which is a particularclass of polyimides that not only retains high temperaturecharacteristics but also possesses good melt processability. Ithas balanced physicochemical properties between performanceand practicability due to the introduction of ether groups. Oneof the commercialized PEIs was developed by GE (GeneralElectric) Plastics in 1986 under the trademark Ultem. It wassynthesized from bisphenol A dianhydride (BPADA) and meta-phenylene diamine (MPD) through condensation polymeriza-tion.1 This polyetherimide has already been employed in manyapplications, such as healthcare applications, pharmaceuticals,and membrane separations. A gas separation membrane madefrom Ultem in Paul’s group2 has shown impressively highselectivity for many gas pairs, especially for the helium/methanegas pair.

In recent years, Extem XH1015, a new brand of polyether-imide, was developed by GE and has been commercialized inEurope since 2007. This so-called amorphous thermoplastic

polymer offers a higher Tg (267 °C), exceptional dimensionalstability, high strength, stiffness, and creep resistance at elevatedtemperatures, outstanding flame, smoke, and low smoke-toxicityperformance without additives. Even though the Extem XH resinis not the strongest material at room temperature, it outperformsmany other high-performance amorphous and semicrystallinethermoplastics at temperatures exceeding 240 °C by retainingmost of its tensile strength and creep resistance.3 Unlike othersemicrystalline thermoplastic polyimides, this fully amorphousExtem XH does not need any postheat treatment to achieve itsdesired property.

On the basis of our previous understanding on polyetherim-ide,2 this new amorphous material may have potential to be amembrane material due to its higher Tg as compared to Ultem.However, the dense Extem membrane does not show impressivegas separation performance. Further endeavor is hindered bythe lack of information on its chemical structure to explain itsstructure relationship with gas permeability and to exploremethodologies for chemical and physical modifications toenhance its application potentials. A bundle of patents4-8 filedby GE provided few clues on this new PEI, but the detailedstructure under the trademark of Extem had never been released.

Therefore, the first objective of this study is to investigateits chemical structure with the aid of NMR, along with FTIR-ATR and elemental analysis. Scheme 1 illustrates the structurewe propose, as well as its synthesis route. Interestingly, Extem’smain chemical repeat unit is quite similar to another grade ofcommercial PEI, named Ultem XH6050, whose chemicalstructure also consists of sulfone groups.9 This so-called PEIS(polyetherimide sulfone) combines the desired properties ofpolyetherimide and polysulfone into a single resin, which includegood processability, high Tg, and good thermal stability at hightemperatures. The good melt processability may be derived fromthe unique characteristics of polyetherimde and polysulfone.Gallucci and Malinoski10 reported that the end group of PEIS

* To whom correspondence should be addressed. E-mail:chencts@nus.edu.sg.

† NUS Graduate School for Integrative Sciences and Engineering,National University of Singapore.

‡ Department of Chemical and Biomolecular Engineering, NationalUniversity of Singapore.

§ Institute of Material Research and Engineering.

Ind. Eng. Chem. Res. 2010, 49, 12014–1202112014

10.1021/ie901906p 2010 American Chemical SocietyPublished on Web 04/12/2010

plays an important role on its Tg and thermal stability. The PEIS,which has the same structure with Ultem XH6050, exhibits ahigher Tg (258 °C) by employing more anhydride as the endgroups. This suggests that the same strategy may also beemployed to improve the Tg of Extem XH1015. Their similarphysical properties indicated in ref 3 indirectly support ourhypothesis.

Since polysulfone (PSU) and Ultem contain part of the repeatunit in Extem as illustrated in Figure 1, the second and the thirdobjectives of this work are to correlate its gas permeationproperties with its chemical structure via both experiments andmolecular simulation and to further investigate its potential asa material for gas separation.

2. Background on Simulation for Gas Permeability

Polymer scientists have made significant attempts to theoreti-cally predict the intrinsic permeability of polymers accordingto monomer moieties, chain structures and rigidity, and func-tional groups.11 Salame12 developed a method named “Perma-chor” to forecast oxygen permeability in barrier-type polymersvia the calculation of chemical group’s empirical factor whichcontributes to the total permeability. Bicerano13 proposed theuse of packing density, cohesive energy, and rotational freedomof a polymer as important parameters to predict permeability.However, the method developed by Lee14 to correlate gaspermeability with free volume has received great attention andhigh remarks. The free volume concept proposed by Lee wasdefined as (V - Vo), where V was the specific volume obtainedexperimentally, while Vo was 1.3 times of Vw which is the sumof van der Waals volume of the groups calculated followingthe Bondi’s method15 or Park and Paul’s method.16 In glassypolymers, the permeability coefficient (P) can be expressed ineq 1. Extensive experiments suggest that P is much moredependent on diffusion coefficient (D) rather than on solubilitycoefficient (S).17

Though many parameters affect the diffusion coefficient, thefree volume plays a much more crucial role than others. Thus,a relationship between the permeability coefficient and the free

volume has been established through experiments18,19 despiteits oversimplifications:

where A and B are constants for a particular gas, while FFV isthe abbreviation of fractional free volume defined as following.

The FFV value can be theoretically estimated from the molecularstructure with the aid of the compass force field theory asdescribed in Chang et al.’s work.20 In Material Studio, the vander Waals volumes could be estimated accurately by employinga probe radius of 0 nm based on the Connolly task21 after thepolymer periodic amorphous cell is constructed and optimizedby molecular dynamics. The total volume of the polymeramorphous cell could also be calculated from the Connolly task.Thus, the FFV which describes the ratio of vacant volume tothe total volume of the polymer could be obtained. However,the FFV value alone could not indicate the effective space fora specific gas penetrant. In Hofmann et al.’s work,22 the sizedistribution of free volume elements was applied to overcomethe drawback of FFV. Wang et al.23 also have calculated thecavity size distribution which showed a good correlation withthe PALS (positron annihilation lifetime study) experimentalinvestigation. Some recent reports by Chang et al.20,24 onsimulation of aromatic polyimide membranes for gas separationhave already pointed out the limitation of FFV and employedfractional accessible volume (FAV) and relative FAV forreplacement. The concept of FAV which was first employedby Tung et al.25 was used to calculate the accurate informationfor a particular gas passing through the membrane. The FAVvalue can be defined as following.

The form of this equation is similar to that of FFV. However,both V and Vo are estimated from the Connolly task simulationby using different probe diameters. The mechanism of thisConnolly task is to employ a hard sphere with a particulardiameter as a probe to detect the available vacancy inside anoptimized amorphous polymer cell. The FFV concept developedfrom the Bondi’s method is universally applicable to all gases,while the Connolly approach calculates FAV for each individualpenetrant. Generally, the FAV would increase when the dimen-sion of detection probe decreases. Significant experimental andtheoretical works have validated eq 2 to correlate gas perme-ability of polymers with similar structures of relatively largermolecules, like methane, with their FAV values. Therefore, weintend to estimate Extem’s permeability and correlate it withits structure via Material Studio and to explore the fundamentalsof molecular design of polymeric materials for gas separation.

Scheme 1. Synthetic Route of Extem XH1015

Figure 1. Chemical structures of PSU, Extem, and Ultem.

P ) DS (1)

P ) A exp(-B/FFV) (2)

FFV ) (V - Vo)/V (3)

FAV ) (V - Vo)/V

Ind. Eng. Chem. Res., Vol. 49, No. 23, 2010 12015

3. Experimental Section

3.1. NMR Characterization, Elemental Analysis, andFTIR-ATR. NMR analyses were conducted using a Bruker 400FT-NMR spectrometer at a resonance frequency of 400.132MHz for 1H and 100.621 MHz for 13C, while 1H NMR spectra(1D, homonuclear decoupling) were obtained from an ExtemXH 1015 pellet sample dissolved in DMSO-d6 using a 5 mmBBO (broadband observe) probe. Other spectra were performedin CDCl3 and chemical shifts are given in parts per million units.

The elemental analysis was carried out using a Flash EA 1112Elemental Analyzer manufactured by Thermo Electron. Theanalyzer was able to determine CHNS (carbon, hydrogen,nitrogen, and sulfur) on dried samples. Oxygen content wasobtained by the subtracting the ash percent plus the total forcarbon, hydrogen, nitrogen, and sulfur.

The FTIR-ATR measurement was performed using a Perkin-Elmer FT-IR Spectrometer Spectrum 2000 with scanning time32. The Extem thin film used in FTIR-ATR was made by ringcasting.

3.2. Membrane Preparation and Gas PermeabilityMeasurements. A commercial Extem XH 1015 polymer wasobtained from the SABIC Innovative Plastics (formerly GEPlastics), MA, and was dried at 120 °C overnight under vacuumbefore use. Dicholoromethane (DCM) was purchased fromMerck and utilized without further purification. Flat dense Extemmembranes were fabricated using a solution casting methoddescribed elsewhere.26 In short, a solution was prepared bydissolving 2 wt % of the polymer sample in a solvent of DCM.The solution was then filtered through a 1 µm filter (Whatman)and cast onto a Si wafer with a metal ring on top. The wholeplate was leveled horizontally and covered with a glass plate.A small gap was left to allow slow solvent evaporation to forma dense membrane. The nascent membrane was further dried at250 °C under vacuum for 48 h to remove residual solvents.

Pure gas permeability measurements were performed usinga variable pressure constant-volume method described else-where.27 Pure gas permeability of He, H2, O2, N2, CH4, andCO2 was tested in order and replicated three times. All themeasurements were conducted at 35 °C and 10 atm except H2

at 3.5 atm. The ideal selectivity of membranes for pure gasesA and B is defined as follows:

where P is the gas permeability.

3.3. Molecular Simulation. Molecular dynamic simulationswere conduct using Material Studio 4.4 from Accelrys. Polymerchains consisting of 10 repeat units of PSU, Ultem, or Extemwere constructed via polymer build function, followed by energyminimization separately. Initiator and terminator were unknownbecause of limited publications or information on these com-mercial polymers. The isotactic configuration with randomtorsion and head-to-tail orientation was assumed prior to theamorphous cell construction. Besides, polymer chains consistingof 5 repeat units were also constructed in order to discuss thechain packing morphology. Two polymer chains of PSU, Extem,and Ultem were put together and minimized in order to discussstructural effect on chain packing. The initial density of thepolymer periodic cell was set as 0.1 g/cm3 in order to eliminatearomatic ring catenation and chain scission during amorphouscell construction as suggested by Heuchel and Hofmann.28 Fourpolymer chains were employed for amorphous cell constructionbased on compass force field calculations by the AmorphousCell module. Conformational optimization and fine convergencewith a maximum iteration of 10 000 was performed beforemolecular dynamics simulations to proceed. The detailedinformation for cell models built was listed in Table 1. Theequilibrium stage temperature was set to 308 K when MD(molecular dynamics) by Discover module was launched. Thisis because all the pure gas permeability data were collected at35 °C. A set of MD under isothermal-isobaric (NPT) modewere performed to compress amorphous cell density from 0.1g/cm3 to the experimental density. The simulated densities arelisted in Table 2. The total MD time for each polymer was 500ps in order to allow the amorphous cell to reach equilibrium.

The fractional accessible volume (FAV) was calculated fromthe free volume and occupied volume simulated by the Connollytask. The FAV distribution of each polymer was plotted by usingdifferent diameter probes to detect available volume in theamorphous cell. For one particular diameter, there is one uniqueFAV, which means only particles smaller than the probe couldgo through or occupy the volume under this definition. Thus,this so-called Connolly task corresponds to the kinetic diametersof gases employed. In this study, we choose probe diametersranging from 2.0 to 5.0 Å, not only because even the diameterof smallest helium molecules is bigger than 2.0 Å, but also thenumber of cavities larger than 5.0 Å is very small and can beignored, compared to that of the smaller cavities.

4. Results and Discussion

4.1. Extem Structure Determination. The 1H NMR and 13CNMR spectra of Extem XH 1015 are illustrated in Figures 2and 3, respectively. The 2D-NMR spectra used to assign eachcarbon and proton in the aromatic region are given in Figures4 and 5. These spectra are in complete agreement with theproposed polymer structure. In addition to NMR spectra andFTIR-ATR shown in Figure 6, the elemental analysis resultsof this polyetherimide sulfone shown in Table 3 also generallyagree with the theoretical values for the proposed structure. Very

Table 1. Atom Numbers and Cell Dimensions for PSU, Extem, andUltem Amorphous Cells

dimension (Å)

atom numbers a b c

PSU 2168 28.9 28.9 28.9Extem 3288 33.3 33.3 33.3Ultem 2768 31.3 31.3 31.3

Table 2. Gas Permeation Performance of Flat Dense PSU, Extem, and Ultem Membranes

permeability (barrera) ideal selectivity

He CO2 O2 N2 CH4 He/CH4 CO2/CH4 O2/N2 density (g/cm3) FFVc FFVd

PSU33 13 5.6 1.4 0.25 0.25 56 22.4 5.6 1.24 (1.22b) 0.143 0.186Extem 11.1 3.28 0.81 0.13 0.13 85 25.2 6.2 1.31 (1.32b) 0.131 0.171Ultem34 9.4 1.33 0.41 0.051 0.036 261 36.9 8.0 1.27 (1.28b) 0.137 0.158

a 1 barrer ) 1 × 10-10 cm3(STP) cm/(cm2 s cmHg) ) 7.5005 × 10-18 m2/(s Pa). b Density simulated from Material Studio. c FFV calculated from theBondi-Park and Paul’s method. d FFV calculated from Material Studio.

RA,B )PA

PB(4)

12016 Ind. Eng. Chem. Res., Vol. 49, No. 23, 2010

minor differences are found due to experimental error and/ordifferent element compositions of the polymer end group.

The aliphatic parts of the 1H and 13C NMR spectra of Extemcan be completely interpreted with the aid of the 1H-13CHETCOR spectrum (not shown here because its position is toofar away from the aromatic region). The adsorption of the C-12carbon at 30.9 ppm of the 13C NMR spectrum correlates wellwith a very sharp singlet of H-a protons at 1.64 ppm of the 1HNMR spectrum, revealing the existence of methyl group. C-13at 42.5 ppm is also in the aliphatic region due to a relative lowchemical shift compared to these carbons in the aromatic region

(110-170 ppm). However, the 1H-13C HETCOR spectrum doesnot show any proton connected to this carbon. The aboveevidence all strongly support a 2,2-substituted propane structurein the Extem’s backbone.

Although the area integration curve of all proton peaks isnot shown in Figure 2, the peak area ratio of a-h protons couldbe determined roughly to be 3:2:1:2:1:2:1:2, which is well-coincident with the theoretical protons ratio based on thechemical structure of Scheme 1. Fortunately, aromatic protons,

Figure 2. H NMR spectrum of Extem XH1015.

Figure 3. C NMR spectrum of Extem XH1015.

Figure 4. COSY spectrum of Extem XH1015.

Figure 5. HETCOR spectrum of Extem XH1015.

Figure 6. FTIR-ATR spectrum of Extem XH1015 thin film.

Table 3. Experimental and Theoretical Elemental Analysis

element C H N S

experimental 69.35 3.95 3.79 4.10theoretical 70.49 3.83 3.83 4.37

Figure 7. Chain morphologies of PSU, Extem, and Ultem with five repeatunits.

Ind. Eng. Chem. Res., Vol. 49, No. 23, 2010 12017

except H-g, all belong to the AX two-spin system.29 Each peakof these protons is observed as a small doublet that is split bythe proton nearby. The 1H-1H correlation (COSY) performedin CDCl3 is shown in Figure 4, revealing that H-b (7.04-7.06ppm) is split by H-d (7.28-7.30 ppm) through a 3-bondcoupling, as do the H-h (8.13-8.15 ppm) and H-f (7.72-7.74ppm) pair, as well as the H-e (7.63-7.65 ppm) and H-c(7.15-7.17 ppm) pair. The H-g at 7.18 ppm is assigned as anisolated proton because of its absence from the COSY spectrum,although it is overlapped by H-f.

Figure 3 shows the signal assignment for a 13C NMRspectrum with 1H noise decoupled during acquisition. Thesecarbon atoms at the high frequency are mainly attached toheteroatoms, making them valuable in the determination ofmonomers, especially diamino monomer. Carbon atom C-1(165.9 ppm) is assigned to the imide carbonyl group. C-5 (164.2ppm) and C-8 (152.7 ppm) constituting the BPADA ethersegment are assigned due to the direct connection to the oxygenatom. A similar carbon chemical shift is also found for C-17

(147.5 ppm) in the polysulfone 13C NMR spectrum.30 Theassignment of each peak marked in Figure 3 is assisted by thecorrelations shown in the 1H-1H COSY (Figure 4) and 1H-13CHETCOR (Figure 5) spectra to fit the structure of polyetherimidesulfone shown in Scheme 1.

In the case of FTIR-ATR spectra as shown in Figure 6, bandsat around 1781 cm-1 (attributed to CdO asymmetric stretch ofimide groups), 1717 cm-1 (attributed to CdO symmetric stretchof imide groups), and 1360 (attributed to C-N stretch of imidegroups) are characteristic imide peaks as indicated in the workof Shao et al.31 Although parts of the band at 1320 cm-1

(attributed to -SO2- asymmetric stretch) are overlapped bystrong adsorption of the imide group at 1360 cm-1, the existenceof sulfone group can still be proved by the strong peak at 1150cm-1 (attributed to a -SO2- symmetric stretch).32

4.2. Chain Morphology Comparison. Figure 7 shows thesingle chain morphologies of PSU, Extem, and Ultem containingfive repeat units at the completely extended state. Although the

Figure 8. Morphology of two polymer chains with five repeat units for PSU, Extem, and Ultem.

Figure 9. Fractional accessible volumes and relative FAV values of PSU, Extem, and Ultem, probed with different diameters.

12018 Ind. Eng. Chem. Res., Vol. 49, No. 23, 2010

morphologies are far from real conformation in our membranes,it is still valuable for the qualitative discussion to correlatemembrane gas permeation performance with their chemicalstructures. As we can see in this figure, Ultem, being the mostrigid polymer, takes a much extended chain conformation; whilePSU, the softest one, exhibits a zigzaglike morphology. Usually,an increase in chain stiffness accompanies with a decrease insegmental mobility. As a result, from PSU, Extem to Ultem,the thermal motions generating transient gaps for penetrants todiffuse decrease. These calculated chain morphologies correlatewell with their gas selectivity as shown in Table 2.

The sequence of chain rigidity in these polymers can also beelucidated from their monomers. PSU and Extem both havebisphenol A and diamino diphenyl sulfone (DDS) moieties. Theadditional heterocyclic imide group in Extem induces a planarand relatively larger ring in the backbone, which increases thechain rigidity. The introduction of heterocyclic imide group alsoenhances the interchain attractions. As shown in Figure 8, wheretwo Extem polymer chains are constructed at fully extendedstate, two heterocyclic imide rings are partially aligned parallelto each other to minimize the total energy. This interchain

attraction effect increases the chain packing efficiency, thusleading to a lower permeability. Ultem also has the samephenomenon, while PSU polymer chains could not be alignedparallel, which correlate well with its high permeability.Although the discussion is only valid in the ideal case, thepolymer chain packing will still follow the principle of minimumtotal potential energy in the real materials. Comparing Extemwith Ultem, the only difference is that the meta-phenylenediamine (MPD) moiety in Ultem is replaced by DDS in Extem.Since the DDS structure can reduce the effective chain packingdue to tetrahedral configuration of the -SO2- group, Extemhas a higher permeability and a lower selectivity than Ultem.

4.3. Molecular Simulation. The FFV values calculated byboth the Bondi-Park and Paul method16 and Material studiowere listed in Table 2 in order to compare their gas separationperformance quantitatively. PSU has the largest FFV, whichcorrelates well with experimental data of the highest perme-ability. However there is a contradiction when it comes to Extemand Ultem. The FFV value of Extem calculated by theBondi-Park and Paul method is smaller than that of Ultem,while the permeability of Extem is much larger than that ofUltem. Obviously, the FFV concept alone cannot interpret thisdiscrepancy because FFV just gives a sum of the total freevolume with no differentiation for different penetrants withdifferent dimensions. In order to explain this contradiction, thefractional accessible volume (FAV) distributions of PSU, Extem,and Ultem, as well as their relative FAV values raised by Changet al.20 in terms of PSU/Ultem and Extem/Ultem were calculatedand listed in Figure 9. In our case, the relative FAV values weredefined, respectively, as follows:

and

The selection of horizontal axis value is based on the kineticdiameter of the probe gas molecule as shown in Table 4 foreasy interpretation with permeation data. For one particulargas, three FAV values were calculated from these threepolymers. It can be seen that the FAV of PSU is always thelargest, which correlates well with its highest permeabilityof all gases as presented in Table 2. Meanwhile, the FAVvalues of Ultem are always the smallest, correlating well withits lowest gas permeability. Furthermore, for all the threepolymers, gas permeability and the reciprocal of FAV followlinear relationships for relatively larger gas molecules,especially for methane and nitrogen as shown in Figure 10,even though their polymer structures are different. In Table2, we also notice that when the gas molecule diameter issmall, the difference in their gas permeability is small, whilewhen the gas diameter increases, this difference expands. Forexample, the helium permeability coefficients of PSU, Extem,and Ultem are 13, 11.1, and 9.4 barrers with a ratio of 1.38:1.18:1. Meanwhile, the methane permeability coefficientsbecome 0.25, 0.13, and 0.036 barrers with a ratio of 6.94:3.61:1. These phenomena could be well explained using therelative FAV concept. As plotted in Figure 9, the relativeFAV values of PSU/Ultem and Extem/Ultem increase whenthe probe diameter increases. In other words, the FAVdifference between PSU and Ultem becomes larger when theprobe diameter increases. This indicates that the fractional

Figure 10. Correlation between gas permeability and 1/FAV.

Table 4. Kinetic Diameters of Various Gases

gas He H2 CO2 O2 N2 CH4

diameter (Å) 2.60 2.89 3.30 3.46 3.64 3.80

Figure 11. FAV ratios of Extem/PSU and Ultem/PSU probed by differentdiameters.

FAV(PSU) - FAV(Ultem)FAV(Ultem)

× 100% (5)

FAV(Extem) - FAV(Ultem)FAV(Ultem)

× 100% (6)

Ind. Eng. Chem. Res., Vol. 49, No. 23, 2010 12019

accessible volume is a more precise tool in analyzing thegas permeability than the total free volume.

The gas selectivity of polymer dense membranes shows agood correlation with the FAV ratio, which was defined,respectively, as follows:

and

As shown in Figure 11, the FAV ratio of Ultem/PSU is about76% for helium, while it is only 63% for methane. This means,using PSU as a reference, the fractional accessible volume forboth helium and methane transport in Ultem is less than in PSU.However, the decrease of FAV for methane is more severe thanthat for helium. Thus, the helium/methane selectivity increasesfrom 56 for PSU to 261 for Ultem. The same approach couldbe used to explain the helium/methane selectivity increase inthe Extem membrane. Another method is to investigate the slopeof a straight line connecting any two data points of interest onthe FAV curve. For any gas pair in one particular polymer, ifthe slope of the line between the selected two data points isnegative, the selectivity of this gas pair for this particularpolymer is higher than the reference polymer. For example, ifhelium and methane are chosen in the Ultem/PSU FAV ratiocurve, the slope of line between these two points is definitelynegative. Therefore, the selectivity of helium/methane is betterin Ultem than in PSU. When comparing gas selectivity of morethan two polymers, this approach is still validate by just choosingone reference polymer. For example, if there are two straightlines linking helium and methane in the Ultem/PSU and Extem/PSU FAV ratio curves, the absolute value of the slope for Ultem/PSU is obviously larger than that for Extem/PSU. Thisdemonstrates that the helium/methane selectivity differencebetween PSU (56) and Ultem (261) is larger than that betweenPSU (56) and Extem (85). This method could explain other gaspairs as well, like O2/N2. The absolute value of the slope betweenO2 and N2 in the Extem/PSU FAV ratio curve is smaller thanthat for Ultem/PSU. This correlates well with their gas selectivitydifference between PSU (5.6)/Extem (6.2) and PSU (5.6)/Ultem(8.0), respectively.

5. Conclusions

The chemical structure of Extem XH1015 has been deter-mined by NMR spectra, FTIR-ATR, and elemental analysis withthe aid of published patent information. Bisphenol-A dianhy-dride (BPADA) and diamino diphenyl sulfone (DDS) areconfirmed to be the monomers of this newly developed hightemperature polyetherimide. Gas permeability of Extem densemembranes is reported for the first time. Due to the structuralsimilarities, PSU and Ultem are used for comparison offractional free volume (FFV), fractional accessible volume(FAV), gas permeability, and selectivity. The effect of differentmonomer structure in their gas separation performance is alsodiscussed in detail. Computational simulation powered byMaterial Studio, especially the FAV value simulation, is provedas a more accurate method to analyze and predict gas separationperformance.

Acknowledgment

The authors would like to thank NUS and A*Star for fundingthis work with the grants of R-398-000-044-305 (NUS) and 092

139 0033 (A-Star), as well as the Singapore National ResearchFoundation (NRF) for the support on the project entitled“Molecular Engineering of Membrane Materials: Research andTechnology for Energy Development of Hydrogen, Natural Gasand Syngas” with grant number of R-279-000-261-281. Theauthors also thank Dr. S. K. Yen and Dr. Y. C. Xiao for theirvaluable discussion and Mr. C. H. Lau for his help on NMRexperiments.

Literature Cited

(1) Heath, D.; Wirth, J. Polyetherimides. U.S. Patent 3,847,867, 1974.(2) Barbari, T. A.; Koros, W. J.; Paul, D. R. Polymeric membranes based

on bisphenol-A for gas separations. J. Membr. Sci. 1989, 42, 69.(3) Lee, S. Extreme performance--or processability? new TP polyimide

offers both. Plast. Technol. Mag. January 2007; http://www.ptonline.com/articles/200701fa6.html.

(4) Klopfer, H. Polyetherimides with high thermal stability and solVentresistance and precursors therefor. US Patent 4,565,858, 1986.

(5) Takekoshi, T.; Klopfer, H. Method for making polyetherimides usingcarboxylic acid salts of alkali metals or zinc as catalysts. US Patent4,293,683, 1981.

(6) Gallucci, R.; Odle, R. Polyimide sulfone, method and article madetherefrom. US Patent 7,041,773, 2006.

(7) Evans, T.; Grade, M. Novel sulfur-containing polyetherimides. USPatent 4,429,102, 1984.

(8) Peters, E.; Bookbinder, D.; Cella, J. Very high heat thermoplasticpolyetherimides containing aromatic structure. US Patent 4,965,337, 1990.

(9) Odle, R.; Gallucci, R. New high heat polyetherimide resins. 61st SPEAntec Technology Conference, Nashville, TN, May 2003; p 1853.

(10) Gallucci, R.; Malinoski, J. Stabilization of polyetherimide sulfones.US Patent 7,411,014, 2008.

(11) Coleman, M. R.; Koros, W. J. Isomeric polyimides based onfluorinated dianhydrides and diamines for gas separation applications. J.Membr. Sci. 1990, 50, 285.

(12) Salame, M. Prediction of gas barrier properties of high polymers.Polym. Eng. Sci. 1986, 26, 1543.

(13) Bicerano, J. Prediction of Polymer Properties; Marcel Dekker: NewYork, 1993.

(14) Lee, W. M. Selection of barrier materials from molecular structure.Polym. Eng. Sci. 1980, 20, 65.

(15) Bondi, A. Physical Properties of Molecular Crystals. Liquids andGlasses; Wiley: New York, 1968.

(16) Park, J. Y.; Paul, D. R. Correlation and prediction of gaspermeability in glassy polymer membrane materials via a modified freevolume based group contribution method. J. Membr. Sci. 1997, 125, 23.

(17) Hellums, M. W.; Koros, W. J.; Husk, G. R.; Paul, D. R. Fluorinatedpolycarbonates for gas separation applications. J. Membr. Sci. 1989, 46,93.

(18) Park, H. B.; Kim, Y. K.; Lee, J. M.; Lee, S. Y.; Lee, Y. M.Relationship between chemical structure of aromatic polyimides and gaspermeation properties of their carbon molecular sieve membranes. J. Membr.Sci. 2004, 229, 117.

(19) Pixton, M. R.; Paul, D. R. Gas transport properties of polyarylates:Substituent size and symmetry effects. Macromolecules 1995, 28, 8277.

(20) Chang, K. S.; Tung, C. C.; Wang, K. S.; Tung, K. L. Free volumeanalysis and gas transport mechanisms of aromatic polyimide membranes:a molecular simulation study. J. Phys. Chem. B. 2009, 113, 9821.

(21) User guide, Visualizer tools section, calculating atom Volume andsurfaces, version 4.4; Acceryls Materials Studio, 2008.

(22) Hofmann, D.; Heuchel, M.; Yampolskii, Yu.; Khotimskii, V.;Shantarovich, V. Free Volume Distributions in Ultrahigh and Lower FreeVolume Polymers: Comparison between Molecular Modeling and PositronLifetime Studies. Macromolecules 2002, 35, 2129.

(23) Wang, X. Y.; Lee, K. M.; Lu, Y.; Stone, M. T.; Sanchez, I. C.;Freeman, B. D. Cavity size distributions in high free volume glassy polymersby molecular simulation. Polymer 2004, 45, 3907.

(24) Chang, K. S.; Tung, C. C.; Lin, C. C.; Tung, K. L. Residual SolventEffects on Free Volume and Performance of Fluorinated PolyimideMembranes: A Molecular Simulation Study. J. Phys. Chem. B. 2009, 113,10159.

(25) Tung, K. L.; Lu, K. T.; Ruaan, R. C.; Lai, J. Y. MD and MCsimulation analyses on the effect of solvent types on accessible free volumeand gas sorption in PMMA membranes. Desalination 2006, 192, 391.

(26) Liu, Y.; Wang, R.; Chung, T. S. Chemical cross-linking modificationof polyimide membrane for gas separation. J. Membr. Sci. 2001, 189, 231.

FAV(Extem)FAV(PSU)

× 100% (7)

FAV(Ultem)FAV(PSU)

× 100% (8)

12020 Ind. Eng. Chem. Res., Vol. 49, No. 23, 2010

(27) Lin, W. H.; Vora, R. H.; Chung, T. S. The gas transport propertiesof 6FDA-durene/1,4 phenylenediamine (pPDA) copolyimides. J. Polym.Sci. Polym. Phys. 2000, 38, 2703.

(28) Heuchel, M.; Hofmann, D. Molecular modelling of polyimidemembranes for gas separation. Desalination 2002, 144, 67.

(29) Friebolin, H. Basic one and two dimensional NMR spectroscopy,3rd ed.; Weinheim: New York, 1998.

(30) Pham, Q.; Petiaud, R. Proton and carbon NMR spectra of polymers,5th ed.; Wiley: New York, 2003.

(31) Shao, L.; Chung, T. S.; Goh, S.; Pramoda, K. Transport propertiesof cross-linked polyimide membranes induced by different generations ofdiaminobutane (DAB) dendrimers. J. Membr. Sci. 2004, 238, 153.

(32) Bolong, N.; Ismail, A. F.; Salim, M. R.; Rana, D.; Matsuura, T.Development and characterization of novel charged surface modification

macromolecule to polyethersulfone hollow fiber membrane with polyvi-nylpyrrolidone and water. J. Membr. Sci. 2009, 331, 40.

(33) McHattie, J. S.; Koros, W. J.; Paul, D. R. Gas transport propertiesof polysulfones: 1. Role of symmetry of methyl group placement onbisphenol rings. Polymer 1991, 32, 840.

(34) Barbari, T. A. Ph.D. Thesis. Permeation of pure and mixed gasesthrough polymeric membranes based on bisphenol-A. The University ofTexas at Austin, 1986.

ReceiVed for reView December 2, 2009ReVised manuscript receiVed March 15, 2010

Accepted March 30, 2010

IE901906P

Ind. Eng. Chem. Res., Vol. 49, No. 23, 2010 12021