Effect of functionality of polyhedral oligomeric silsesquioxane [POSS] on the properties of sulfonated poly(ether ether ketone) [SPEEK] based hybrid nanocomposite proton exchange membranes for fuel cell applications

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Effect of functionality of polyhedral oligomeric silsesquioxane [POSS] on the properties of sulfonated poly(ether ether ketone) [SPEEK] based hybrid nanocomposite proton exchange membranes for fuel cell applications Deeksha Gupta a,*, Anand Madhukar b, Veena Choudhary a,* a

Centre for Polymer Science & Engineering, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India b Department of Civil Engineering, Zakir Hussain College of Engineering, Aligarh Muslim University, Aligarh 202001, India

article info

abstract

Article history:

Organiceinorganic hybrid nanocomposite membranes were prepared using three different

Received 7 April 2013

types of POSS i.e., PEG POSS
cage mixture (PPOSS), trisilanol phenyl POSS
(TSP POSS) and

Received in revised form

trisulfonic acid isobutyl POSS
(SPOSS) at a fixed loading of 2% (w/w) as filler and SPEEK with

16 July 2013

degree of sulfonation (DS) 55% as polymer matrix. The influence of POSS functionality on

Accepted 20 July 2013

hybrid membrane’s thermo-mechanical properties, morphology, water uptake and proton

Available online 15 August 2013

conductivity was investigated. Thermal and mechanical stability of hybrid membranes increased upon incorporation of POSS. The size and distribution of POSS particles into SPEEK matrix was studied using transmission electron microscopy (TEM) and field emission scan-

Keywords: SPEEK

ning electron microscopy (FESEM) and it was found that TSP POSS and PPOSS based membranes showed smaller particle size and uniform distribution as compared to SPOSS based membranes which consequently affect the water uptake and proton conductivity of these

POSS Water uptake Proton conductivity

hybrid membranes. The water uptake studies were carried out at three different temperatures i.e. 30, 80, 100
C for 24 h and POSS based composite membranes showed higher water uptake and proton conductivity compared to neat SPEEK membranes. The highest proton conductivity (64.6 mS/cm) was observed for TSP POSS containing membrane which is more than double of neat SPEEK (31.3 mS/cm) membrane. The composite membrane containing TSP POSS can be considered as suitable membrane for PEMFCs applications. Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

Sulfonated aromatic polymers have been widely investigated as membranes for water purification by reverse osmosis [1]

vanadium redox flow batteries for energy storage [2] and fuel cells for energy production [3]. Among different types of fuel cells, polymer electrolyte membrane fuel cells (PEMFCs) offer high fuel utilization efficiency with environmentally

* Corresponding authors. Tel.: þ91 011 26591423; fax: þ91 011 26591421. E-mail addresses: [email protected] (D. Gupta), [email protected] (V. Choudhary). 0360-3199/$ e see front matter Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2013.07.070

12818

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

benign operation [4,5]. The properties of membranes such as proton conductivity, water uptake, mechanical, thermal and chemical stability are primary characteristics important for the fuel cell performance. Perfluorosulfonic acid (PFSA) membranes, state-of-the-art such as Nafion are the most widely used membrane materials in fuel cell applications owing to their high proton conductivity and good chemical stability. But, the problem of high cost and reduced efficiency at elevated temperatures has restricted their widespread application in PEMFC [6e8]. Therefore, the scientific community is intensively involved in finding alternative membrane materials based on different high-performance aromatic polymers functionalized with proton conductive groups [9e12]. But, the performance of most aromatic membranes is not comparable to PFSA membranes, particularly in terms of proton conductivity at low relative humidity (RH) and membrane stability. Sulfonated poly(ether ether ketone) [SPEEK] has been emerged as a potential candidate owing to have a comparable proton conductivity, superior thermal and chemical properties accompanied with lower fuel crossover. SPEEK is quite durable under fuel cell operating conditions along with a long lifetime of approximately 3000 h [13,14]. With increasing degree of sulfonation (DS), the long term stability of SPEEK is questionable because of the hydroxyl radical initiated degradation of SPEEK. On the other hand, low to medium DS of SPEEK in the range of 45e55% exhibit reasonably good thermo-chemical stability but low values of proton conductivity. SPEEK with high degree of sulfonation can be used to achieve high proton conductivity which in-turn is accompanied by excessive swelling leading to poor mechanical properties of the membrane and hence is undesirable for fuel cell applications. Therefore, a membrane showing an optimum balance of water swelling and proton conductivity would be the most suitable PEM for fuel cells. A variety of modification approaches such as cross-linking [15e18] and blending [19,20] have been explored to prepare efficient membranes based on SPEEK. The addition of inorganic filler is also one of the most widely used approaches to modify the SPEEK because inorganic additives can withstand relatively higher temperatures than virgin polymer. Different inorganic fillers such as zirconium phosphate [21], titanium [22] silica [23] and nanoclays [24] are used to alter the properties of neat polymer. The presence of functional group on inorganic filler is beneficial to improve polymer filler interaction. Few examples such as: zirconium phosphate sulfophenylene phosphonate [25,26], sulfonated titania [27], sulfonated silica [28] or organically modified clay [29] as inorganic fillers have shown better performance than their unmodified counterparts. The presence of functional groups on filler might generate a variety of interactions such as covalent, ionic, hydrogen bonding etc. with polymer matrix affecting the properties of composite. The purpose of such modifications was to suppress the methanol permeation while keeping the high values of proton conductivity. But in case of hydrogen fuel cell, the main aim to add functional filler is to retain water at higher temperatures to maintain higher values of proton conductivity. In search of potential filler for PEMFCs, polyhedral oligosilsesquioxane (POSS) has been emerged as very attractive

filler. POSS could be regarded as a hybrid nanoparticle since it has a well-defined cube-octameric siloxane skeleton (about 1e3 nm in size) with eight organic vertex groups, one or more of which are reactive or polymerizable. These particular structural features render POSS to be a versatile additive for acquiring enhanced thermo-mechanical properties, better thermal stability, oxidative resistance and abrasion resistance [30e42]. The interaction of POSS with polymer matrix usually depends on various factors such as the size of POSS cage, nature of organic periphery, number and type of reactive functional groups and the amount of POSS incorporated into polymer matrix. Further, the size and distribution of POSS particles into polymer matrix affect the organiceinorganic phase interfacial characteristics to a large extent thereby influencing the properties of nanocomposites [43]. POSS has been studied in different polymer matrix such as SPPSU [44,45] and Nafion [46] and proved to be synergistic filler for fuel cell applications. It is possible to tailor a variety of POSS/polymer composite by varying substitution on POSS [47,48]. Therefore, we found it very crucial and interesting to study the effect of POSS functionality on the properties of SPEEK nanocomposite membranes. The concentration of POSS at 2% (w/w) was optimized in our previous work.

2.

Experimental

2.1.

Materials

Victrex PEEK (150 XF ICI, USA), sulfuric acid (98% Merck), dimethyl acetamide (DMAc) (Qualigens, India), PEG POSS
Cage Mixture (PPOSS), trisulfonic acid isobutyl POSS
(SPOSS), trisilanol phenyl POSS
(TSP POSS) from Hybrid Plastics, USA were used as received without further purification. The structure of POSS particles is shown in Scheme 1(a)e(c) along with their description in Table 1.

2.2.

Synthesis of SPEEK

SPEEK with varying degree of sulfonation was prepared by varying reaction time from 2 to 3 h [49e52]. The detailed procedure of sulfonation and determination of DS is reported in our earlier publications [53,54]. For the present studies, SPEEK with DS w55% was used as polymer matrix. Ion exchange capacity IEC (expressed in meq/g) of the polymer can be calculated using the degree of sulfonation and the mean molecular mass of the repeat units (equivalent weight). For SPEEK with DS ¼ 55%, equivalent weight is EW ¼ 603, which gives IEC ¼ 1000 DS/EW ¼ 0.9.

2.3.

Membranes casting

Solution casting method was employed to prepare homogenous and defect free membranes. SPEEK with DS w55% and POSS with varying functionality in calculated amounts were dissolved separately in DMAc. The solution of SPEEK and POSS was mixed together and stirred vigorously at room temperature until homogenization and followed by ultrasonication for 30 min. The solution mixture was poured onto glass petridishes followed by solvent evaporation at 80
C for 12 h.

12819

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

The membranes were heat treated in a vacuum oven at different temperatures for definite time period i.e. at 60
C for 2 h, 80
C for 2 h, 100
C for 2 h, 120
C for 4 h and 135
C for 16 h. The composite membranes were designated as SP/ PPOSS-2.0 where SP refers to SPEEK and PPOSS-2.0 is 2% (w/ w) PPOSS, similar is true for SP/TSP POSS-2.0 and SP/SPOSS2.0.

2.4.

Characterization methods

2.4.1.

Thermal characterization

(i) Thermogravimetric analysis (TGA) Thermogravimetric (TG) and derivative thermogravimetric (DTG) traces of the membranes were recorded on a Pyris 6 TGA, PerkineElmer instrument under nitrogen atmosphere. The membrane samples were dried in vacuum oven at 100
C for 12 h prior to TG analysis. Sample weight of 6  2 mg and a heating rate of 20
C/min were used in the temperature range of 50e850
C. (ii) Dynamic mechanical analysis (DMA) TA Q800 dynamic mechanical analyzer in tension mode at an oscillation frequency of 1.0 Hz was employed to study dynamic mechanical properties of anhydrous membrane samples having dimensions of length: 25  5 mm; width: 6  0.5 mm; thickness: 100  10 mm in the temperature range from 30
C to 300
C at a heating rate of 2
C/min.

2.4.2.

Morphological characterization

(i) Transmission electron microscopy (TEM) The small membrane specimen embedded into epoxy resin was clamped in the specimen chuck of a Leica ultramicrotome and then cross-sectioned with the ultramicrotome set for 70 nm sectioning thickness. The sections were transferred onto neat copper mesh TEM sample grid. The TEM images were obtained by examining the ultrathin membrane specimens on a JEOL, 2100-F (field emission type) TEM machine operated at 200 kV [54]. (ii) Field emission scanning electron microscopy (FESEM) Scheme 1 e Structure of (a) PEG POSS
cage mixture (b) Trisulfonic acid isobutyl POSS
(c) Trisilanol phenyl POSS


FESEM images of the cryo fractured membranes were captured using Hitachi S-4800 FESEM operated at 1.5 kV. The membranes were used without any coating for SEM examination [53,54].

Table 1 e Description of different types of POSS. Product name Color Molecular formula Formula weight

PEG POSS
cage mixture

Trisulfonic acid isobutyl POSS


Trisilanol phenyl POSS


Clear [C2mþ3H4mþ7Omþ1]n(SiO1.5)n w5600 when n ¼ 8

Colorless to brown C70H132O30S3Si10 1830.83

White C42H38O12Si7 w931.34

12820

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

(iv) X-ray diffraction analysis A Panalytical X’pert PRO diffractometer (Philips X’pert PRO) with CuKa radiation source was employed to investigate the diffraction pattern of neat SPEEK and the composite membranes for 2q varying between 2
and 35
.

Neat SPEEK SP/SPOSS-2.0 SP/TSP POSS-2.0 SP/PPOSS-2.0

90

80

70

60

Weight (%)

The topographic images of freshly prepared membranes were collected using a Nanoscope IIIA Veeco Metrology group in tapping mode. A silicon nitride tip with a force constant of w50 N/m at a scan rate of 1 Hz was used to study the topography of the membrane surface.

100

Weight (%)

(iii) Atomic force microscopy (AFM)

50 Temperature ( C)

40 100

200

300

400

500

600

700

800

Temperature (oC)

(a)

Water uptake

Water uptake studies of the membranes were performed at three different temperatures i.e. at 30
C, 80
C and 100
C for 24 h in triplicate to minimize the experimental error as per method reported earlier [53,54]. The water uptake was calculated using the following formula. Water Uptake % ¼


Wwet  Wdry  100 Wdry

dTG/dT

2.4.3.

The hydration degree l was calculated using formula as given in Refs. [55,56].

SP/PPOSS-2.0 SP/TSP POSS-2.0 SP/SPOSS-2.0 Neat SPEEK




l¼

Wwet  Wdry 1000  Wdry IEC  MðwaterÞ

where, M(water) is the molecular weight of water.

2.4.4. Proton conductivity by electrochemical impedance spectroscopy Through plane proton conductivity of membranes was measured using Electrochemical Workstation IM6 (ZahnerElektrik GmbH & Co., KG, Germany) connected to the Membrane Conductivity and Single Cell Test System BT-552 (BekkTech, USA). Four-electrode geometry (inner electrodes ¼ platinum wire; outer electrodes ¼ platinum gauze), that allows to exclude membraneeelectrode contact phenomena from the bulk conductivity was used in the measurement cell. The experiments were carried out at 100
C by varying the relative humidity from 40 to 100%. A membrane sample having thickness of 30e50 mm and width of 12.5 mm was used for proton conductivity determination.

3.

Results and discussion

3.1.

Thermal characterization

(i) Thermogravimetric analysis To guarantee a long lifetime of proton exchange membranes in fuel cell, the thermal stability of membrane is a crucial property to be evaluated. Fig. 1(a) and (b) shows TG/ DTG traces for membranes based on SPEEK and SPEEK containing three different kinds of POSS at 2% (w/w)

100

200

300

400

500

600

700

800

Temperature (oC)

(b) Fig. 1 e (a)TG and (b) DTG traces of neat SPEEK and SPEEK/ POSS composite membranes.

concentration. Thermogravimetric analysis has been used in the present work to compare the thermal stability of membranes and relative thermal stability was evaluated by comparing the mass loss in different temperature ranges. All the membranes show three step mass loss pattern. The first step mass loss below 200
C is attributed to the loss of chemically and physically adsorbed water molecules. There is no sharp boundary of 200
C for the removal of total physically and chemically bound water from membrane, but, the segregation of temperature zone in TGA was done for convenience in understanding of comparative behavior of different membranes. The second mass loss in the temperature range of 200e450
C corresponds to the loss of sulfonic acid groups from SPEEK. The third mass loss observed in the temperature ranging from 450 to 800
C is due to polymer back-bone degradation. Fig. 1(a) shows that POSS containing membranes exhibit lower mass loss than neat SPEEK membrane below 200
C which could be due to (i) the entrapment of water molecules within the cage of POSS and hence they might not escape from the membrane till 200
C. (ii) As all the membranes were dried under similar conditions before TG analysis

12821

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

Table 2 e Results of storage modulus at different temperatures and Tg of membranes from DMA experiment. Storage modulus (MPa) at varying temperatures (
C)

Sample designation

Neat SPEEK SP/PPOSS-2.0 SP/TSP POSS-2.0 SP/SPOSS-2.0 Nafion-112

Tga (
C)

30

60

100

120

150

1291 2401 1265 344 405

1212 2282 1368 486 334

1244 2222 1534 661 248

1914 2230 1543 788 182

982 1667 1352 1003 65

148 177 196 218 178

a Noted as peak temperature from the plots of tan d vs. temperature plot.

so the difference in the amount of water loss below 200
C can be attributed to the varying hydrophilicity of the membranes which allows them to hold or absorb water up to different extent. Table 3 also suggests that composite membranes are more hydrophilic than neat SPEEK membranes and would have more possibility of intermolecular hydrogen bonding with water molecules therefore, inspite of absorbing more water (as evident by water uptake data), the composite membrane can hold water till 200
C in TG analysis. Therefore, the composite membranes lost lesser water than neat SPEEK membrane whereas they showed higher water uptake tendency. So, if composite membranes absorb more water and lose less water, then it suggests that water is strongly held by composite membrane and does not escape even up to 200
C. The extensive hydrogen bonding facilitated by POSS can be responsible for this behavior which is also a desirable property requirement for application of membranes in the fuel cells to maintain high values of proton conductivity at elevated temperatures. Fig. 1(a) and (b) depicts that addition of PPOSS and TSP POSS into SPEEK alters the thermal degradation pattern of neat SPEEK whereas incorporation of SPOSS into SPEEK did not affect the degradation behavior of SPEEK. The peak associated with loss of sulfonic acid group in the temperature range of 200e450
C is broad and bifurcated in case of PPOSS and TSP POSS based membranes, indicating a gradual removal of sulfonic acid groups over the whole temperature range. But, for neat SPEEK and SPOSS based membranes, the removal of sulfonic acid group occurs in single step indicating that the type of sulfonic acid groups in SPEEK and SPOSS behaves in similar way and giving the clue that incorporation of SPOSS might not tune the microphase separation of the composite membrane which is also evident from morphology where bigger agglomerates of SPOSS are observed suggesting that

polymer filler interaction is least when compared to PPOSS and TSP POSS based membranes. It can be proposed that TSP POSS or PPOSS creates certain physical or chemical interaction with sulfonic acid groups of SPEEK which alters the pattern of removal of sulfonic acid groups from SPEEK. It is speculated that TSP POSS and PPOSS might generate stronger affinity than SPOSS with SPEEK matrix which results into changed thermal behavior of TSP POSS and PPOSS based composite membranes. (ii) Dynamic mechanical analysis The storage modulus (E0 ) obtained from DMA measurement refers to the elastic response to the deformation and relating the ability of a material to store energy when an oscillatory load is applied to the membrane. Fig. 2(a) and Table 2 exhibit the temperature dependence of storage modulus of neat SPEEK, composite membranes and Nafion-112. All the SPEEK membranes showed higher value of storage modulus than Nafion-112 at all the measured temperatures indicating the presence of constrained regions in these membranes and hence are better in stiffness. The highest storage modulus was observed in case of PPOSS based composite membrane referring PPOSS as reinforcing filler. The variation in storage modulus can be attributed to anisotropic reinforcement due to thermal reorganization leading to increased mechanical strength with increasing temperature. Fig. 2(b) depicts the plot of tan d vs. temperature for neat SPEEK, composite membranes and Nafion-112. The peak temperature of each graph is considered as the glass transition temperature. It is noticed from Table 2 and Fig. 2(b) that SPEEK membranes containing POSS showed higher glass transition temperature as compared to neat SPEEK which could be due to the restricted mobility of polymer chain

Table 3 e Water uptake results of neat SPEEK and composite membranes at three different temperatures after 24 h of water immersion. Water uptake (%) at three different temperature with corresponding la values

Sample designation 30
C Neat SPEEK SP/SPOSS-2.0 SP/TSP POSS-2.0 SP/PPOSS-2.0

16.0  20.5  21.7  23.4 

0.3 0.5 0.4 0.5

a Hydration degree l ¼ n H2O/eSO3H.

l values 9.7 12.5 13.3 14.3

   

0.2 0.3 0.2 0.3

80
C 40.7 47.8 48.4 50.8

   

0.5 0.5 0.7 0.7

l values 24.8 29.2 29.5 31.0

   

0.3 0.3 0.4 0.4

100
C 60.4 80.3 82.1 85.7

   

0.8 0.8 1.0 1.0

l values 36.9  49.0  50.1  52.3 

0.4 0.4 0.5 0.5

12822

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

segmental motion of polymer chain might be restricted by inter or intra molecular hydrogen bonding whereas the bigger agglomerates generate heterogenous morphology leading to discontinuous structure and hence low value of storage modulus was recorded. On the other hand, for TSP POSS and PPOSS, the presence of relatively smaller particles creates more continuous and compact structure and so the transfer of energy is smoother resulting in high values of storage modulus. Therefore, it is possible that a material can show opposing trend of Tg and storage modulus. From the results of storage modulus and glass transition temperature, it can be concluded that SPEEK membrane containing TSP POSS showed an optimum combination of

Fig. 2 e (a) Plot of storage modulus vs. temperature (b) Plot of tan d vs. temperature for Nafion-112, neat SPEEK and SPEEK/POSS composite membranes.

segments in the presence of POSS. Further, it is observed that POSS functionality has pronounced effect on Tg of composite membranes because there is a gradual increase in Tg of composite membrane from PPOSS (177
C) to TSP POSS (196
C) and then to SPOSS (218
C). It can be proposed that the extent of hydrogen bonding is different in all three kinds of POSS which leads to the variation in Tg. The presence of sulfonic acid groups might generate extensive intra and intermolecular hydrogen bonding and therefore, the energy required for segmental motion increased, resulting in the highest Tg of SPOSS containing composite membrane. However, SPOSS based membranes showed the lowest storage modulus among all three types of composite membranes because Tg is a bulk property and concerned with the segmental motion of polymer chain irrespective of the sample’s physical state. But storage modulus is a material property and it depends on free volume and compactness of the sample. Any discontinuous structure can adversely affect the mechanical properties of the sample. For example, in case of SPOSS where the

Fig. 3 e TEM images of (a) SP/PPOSS-2.0 (b) SP/TSP POSS-2.0 and (c) SP/SPOSS-2.0.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

stiffness and glass transition temperature required for PEMFCs applications.

3.2.

Morphological characterization

The internal microstructure and the dispersion of POSS within the membranes were investigated by TEM and FESEM. Crystalline or amorphous region within the membranes was studied by XRD. AFM topographic images were captured to observe the surface roughness. (i) Transmission electron microscopy Incorporation of POSS to SPEEK matrix resulted in varying domain size of POSS depending on its functionality as shown in Fig. 3. The composite membranes based on PPOSS or/TSP POSS showed smaller particle size as compared to SPOSS based composite membrane. The sulfonic acid groups in SPOSS might generate strong intra molecular hydrogen bonding which hinder the finer dispersion of SPOSS into SPEEK matrix as compared to PPOSS and TSP POSS. The particle size was found to be in the range of 50e100 nm for PPOSS and TSP POSS composite membranes whereas bigger particles in the range of 200e300 nm were observed in SPOSS based composite membrane. It is well reported in literature that

12823

influence of dispersed phase on the properties of composite depends upon the distribution and size of particles [57]. As the size of filler particle decreases, the number of interaction sites between polymer and filler increases which in-turn affects several polymer properties such as water uptake, proton conductivity and the mechanical and thermo-mechanical properties. (ii) Field emission scanning electron microscopy The cross-sectional FESEM images of neat SPEEK and POSS based composite membranes are presented in Fig. 4. All the membranes show dense, smooth and defect free internal structure. The presence of POSS particle is evident from FESEM images and it is clear that particle size is dependent on the functionality of POSS. TSP POSS containing membranes exhibited smaller and uniformly dispersed particles ranging from 50 to 150 nm whereas agglomerates were observed for SPOSS particles in the range of 300e400 nm. Therefore, a similar kind of trend is observed from TEM and FESEM images regarding the particle size but the actual values of particle size were different in both the techniques which could be due to the technical difference in TEM and FESEM techniques. The influence of particle size and its distribution on water uptake and proton conductivity would be discussed in next sections.

Fig. 4 e FESEM cross-sectional image of neat SPEEK, SP/PPOSS-2.0, SP/TSP POSS-2.0, SP/SPOSS-2.0.

12824

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

Fig. 5 e AFM topographic images of (a) neat SPEEK (c) SP/PPOSS-2.0 (e) SP/TSP POSS-2.0 (g) SP/SPOSS-2.0 (in two dimensions) and (b) neat SPEEK (d) SP/PPOSS-2.0 (f) SP/TSP POSS-2.0 (h) SP/SPOSS-2.0 (in three dimensions).

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

(iii) Atomic force microscopy The topographic images (2D and 3D) of SPEEK and SPEEK based composite membranes were captured by using tapping mode AFM and are shown in Fig. 5. The surface roughness of composite membranes and compatibility between POSS and SPEEK was evaluated by topographic images. The relative surface roughness of different membranes is obtained by comparing the height in the z direction. Neat SPEEK membrane depicts smoother surface as compared to that of composite membranes prepared by using a variety of POSS. Further, it is observed that surface roughness increases successively moving from PPOSS (40 nm) to TSP POSS (60 nm) and further to SPOSS (120 nm) as shown in Fig. 5. The variation in surface roughness might be related to the particle size of POSS as shown in TEM and FESEM images. It can be proposed that PPOSS and TSP POSS are present in form of smaller particle ranging in size from 50 to 200 nm and finely distributed into SPEEK matrix, accounting for higher polymer filler interaction and hence have smoother surface. SPOSS based membrane was found to have the bigger particles w300e400 nm due to the cohesive interaction among sulfonic acid groups of SPOSS which did not allow the finer particle distribution of SPOSS into SPEEK matrix. As the particle size increases, surface roughness also increases which is evident by scale bar shown in the AFM image. SPOSS based membranes were found to have higher surface roughness as compared to TSP POSS and PPOSS based membranes and hence, AFM observation also supports TEM and FESEM results.

3.3.

Water uptake

An optimum amount of water absorption is one of the prime requirements for a PEM being used in a fuel cell. Water uptake comprises various processes such as adsorption, absorption, cluster formation and incorporation of water molecules into free space and the combination of other processes. It is

12825

proposed that proton travels along hydrogen bonded ionic channels in the cationic forms such as H3Oþ, H5 O2 þ and H9 O4 þ in water. In dry state, the membranes show low proton conductivity which arises due to the difficult segmental motion of polymeric chains whereas in hydrated state, the absorbed water molecules act as plasticizer increasing the mobility of chains augmenting the free ion concentration of polymer. Water uptake and l values of neat SPEEK and the composite membranes containing a variety of POSS are shown in Table 3. Water uptake experiments were carried out to mainly check the hydrolytic stability of the membranes at various temperatures. Water uptake for all the membranes increases with increasing temperature which could be due to faster movement of different mobile species such as protons, water molecules and increased segmental mobility of polymer chains which in-turn generate micro-voids in the membranes to accommodate more water molecules. Further, it is also noted that the composite membranes show high water uptake compared to neat SPEEK which could be due to hydrophilic nature of POSS which enable composite membranes to attract large number of water molecules. Among the composite membranes, SPOSS based membrane exhibits the lowest water uptake which is in contrast to the speculated trend where it was thought that SPOSS based membranes might absorb the highest amount of water due to higher hydrophilicity of sulfonic acid group over other functional groups such as silanol and PEG in TSP POSS and PPOSS respectively. However, the opposite trend was observed for SPOSS based membrane showing the lowest water uptake. It can be proposed that hydrophilicity of a functional group can only play its role when the functional group is sufficiently exposed to water molecules which depend on the surface area of the filler which is determined by the distribution and particle size of filler into polymer matrix. As smaller particle size corresponds to higher surface area and therefore, higher number of functional groups will be exposed to interact

Scheme 2 e Schematic drawing of membrane’s microstructure depicting the difference in extent and connectivity of water molecules.

12826

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

with water molecules. TEM and FESEM images showed that SPOSS was present in the form of bigger agglomerates (w400e500 nm) whereas TSP POSS and PPOSS particles were present in the range of 50e150 nm. So, inspite of being more hydrophilic, SPOSS based membrane absorbed the lowest amount of water among all three types of composite membranes. Scheme 2 shows effect of particle size and its distribution on the density and connectivity of water molecules which in-turn govern the values of proton conductivity.

3.4.

Proton conductivity

Proton conductivity of the membranes at 100
C with varying relative humidity from 40 to 100% is shown in Fig. 6. The proton conductivity of all the membranes increases with increasing relative humidity. Usually, proton conductivity is dependent on various factors such as microstructure, water

100

100

Proton conductivity (mS/cm)

uptake, size and connectivity of ionic domains into membrane. It is believed that proton transport requires water assisted pathways and as the amount of water increases in the membrane proton mobility also increases, resulting in the higher values of proton conductivity. All the membranes based on SPEEK show lower proton conductivity as compared to Nafion-112 but all the composite membranes show higher proton conductivity than neat SPEEK membrane. Furthermore, among the composite membranes, SPOSS based membrane exhibits the lowest proton conductivity (39.6 mS/cm) and TSP POSS (64.6 mS/ cm) containing membrane shows the highest proton conductivity [double in magnitude of neat SPEEK (31.3 mS/cm)]. It can be proposed that the type of functional group present on POSS alters the microstructure of the hybrid composite membranes. The change in microstructure affects the water uptake and connectivity of water molecules within

SP/SPOSS-2.0

Neat SPEEK

10

10

31.3 mS/cm

1

0.1

0.1

0.01

0.01 40

50

60

70

80

90

40

100

50

60

70

80

90

100

100

100

Proton conductivity (mS/cm)

39.6 mS/cm

1

SP/TSPPOSS-2.0

SP/PPOSS-2.0

10

10

64.6 mS/cm 46.7 mS/cm 1

1

0.1

0.1

0.01

0.01 40

50

60

70

80

90

40

100

50

60

Relative humidity (%)

70

80

90

100

Relative humidity (%)

Proton conductivity (mS/cm)

Nafion-112

100

138.6 mS/cm

10 40

50

60

70

80

90

100

Relative humidity (%)

Fig. 6 e Plot of proton conductivity vs. relative humidity at 100
C for neat SPEEK, composite membranes and Nafion-112.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

membrane thus varying the proton conductivity of hybrid membranes. Morphological studies of SP/SPOSS-2.0 revealed that polymerepolymer and fillerefiller interactions are predominant over the polymerefiller interactions so, SPOSS was dispersed in form of bigger agglomerates of size w400e500 nm whereas TSP POSS and PPOSS were present as smaller particles with size ranging from 30 to 150 nm. The presence of large agglomerates might hinder the sufficient exposure of sulfonic acid groups to water so this membrane did not absorb the optimum amount of water to make continuous water channels which may be responsible for the decrease in the proton mobility. For TSP POSS and PPOSS based membranes, fine distribution of particles permits appreciable interaction of water molecules with POSS and thereby creating well connected water channels which accounts for higher proton conductivity of these membranes. Though, there is not much significant difference in particle size and water uptake values between TSP POSS and PPOSS based composite membranes but the proton conductivity value is higher for TSP POSS membrane as compared to PPOSS membrane. It could be due to the open cage structure of TSP POSS which might allow easy entry for water molecules and created more connected water pathways than PPOSS membranes having closed cage structure of PPOSS. Among all the membranes based on SPEEK, TSP POSS based membrane showed the highest proton conductivity of 64.6 mS/cm which is more than double of neat SPEEK.

4.

Conclusions

Incorporation of POSS with different functionality into SPEEK matrix resulted into efficient membranes showing better properties than neat SPEEK membranes. By changing the functionality of POSS, it is possible to tune microstructure, water uptake and proton conductivity of hybrid composite membranes. Among all three types of POSS studied in this paper, TSP POSS based membranes showed the smallest particle size and uniform distribution which improved the polymer filler interaction giving high Tg and thermal stability. Smaller particle size of TSP POSS is also responsible for optimum and better connected water pathways providing smooth proton mobility thus resulting in the highest proton conductivity. Therefore, the hybrid composite membrane containing TSP POSS can be used as an efficient PEM for PEMFCs applications.

Acknowledgment The authors thank IIT Delhi and DAAD for financial support. DG is grateful to Prof. Martin Mo¨ller and Dr. Xiaomin Zhu at DWI, RWTH Aachen, Germany for giving the permission to work at DWI, RWTH Aachen, Germany. Special thanks to Dr. Rostislav Vinokur at DWI, RWTH Aachen, Germany for his kind support for proton conductivity measurement.

12827

references

[1] Geise GM, Lee HS, Miller DJ, Freeman BD, McGrath JE, Paul DR. Water purification by membranes: the role of polymer science. J Polym Sci Part B Polym Phys 2010;48:1685e718. [2] Chen DY, Wang SJ, Xiao M, Meng YZ. Synthesis and characterization of novel sulfonated poly(arylene thiether) ionomers for vanadium redox flow battery application. Energy Environ Sci 2010;3:622e8. [3] Bose S, Kuila T, Thi XLN, Kim NH, Lau KT, Lee JH. Polymer membranes for high temperature proton exchange membrane fuel cell: recent advances and challenges. Prog Polym Sci 2011;36(6):813e43. [4] Wang Y, Chen KS, Mishler J, Cho SC, Adroher XC. A review of polymer electrolyte membrane fuel cells: technology, applications, and needs on fundamental research. Appl Energy 2011;88(4):981e1007. [5] The fuel cell today industry review. Fulmar Colour Printing Co. Ltd 2011 ISSN: 1756-3186. [6] Pozio A, Cemmi A, Mura F, Masci A, Serra E, Silva RF. Longterm durability study of perfluoropolymer membranes in low humidification conditions. J Solid State Electrochem 2011;15:1209e16. [7] Xiao S, Zhang H, Li X, Mai Z. Investigation of the differences between the in situ open circuit voltage test and ex situ fenton test for PEM oxidation characterization. Int J Hydrogen Energy 2011;36:10934e9. [8] Kusoglu A, Santare MH, Karlsson AM. Aspects of fatigue failure mechanisms in polymer fuel cell membranes. J Polym Sci Part B Polym Phys 2011;49:1506e17. [9] Park CH, Lee CH, Guiver MD, Lee YM. Sulfonated hydrocarbon membranes for medium-temperature and lowhumidity proton exchange membrane fuel cells (PEMFCs). Prog Polym Sci 2011;36:1443e98. [10] Yang Y, Siu A, Peckham TJ, Holdcroft S. Structural and morphological features of acid bearing polymers for PEM fuel cells. Adv Polym Sci 2008;215:55e126. [11] Maier G, Meier-Haack J. Sulfonated polymers for fuel cell membranes. Adv Polym Sci 2008;216:1e62. [12] Roziere J, Jones DJ. Non-fluorinated polymer materials for proton exchange membrane fuel cells. Ann Rev Mater Res 2003;33:503e55. [13] Smitha B, Sridhar S, Khan AA. Synthesis and characterization of proton conducting polymer membranes for fuel cells. J Membr Sci 2003;225(1e2):63e76. [14] Smitha B, Sridhar S, Khan AA. Polyelectrolyte complexes of chitosan and poly(acrylic acid) as proton exchange membranes for fuel cells. Macromolecules 2004;37:2233e9. [15] Nolte R, Ledjeff K, Bauer M, Mu¨lhaupt R. Partially sulfonated poly(arylene ether sulfone) e a versatile proton conducting membrane material for modern energy conversion technologies. J Membr Sci 1993;83(2):211e20. [16] Mikhailenko SD, Wang K, Kaliaguine S, Xing P, Robertson GP, Guiver MD. Proton conducting membranes based on crosslinked sulfonated poly(ether ether ketone) (SPEEK). J Membr Sci 2004;233:93e9. [17] Mikhailenko SD, Robertson GP, Guiver MD, Kaliaguine S. Properties of PEMs based on cross-linked sulfonated poly(ether ether ketone). J Membr Sci 2006;285(1e2):306e16. [18] Hande VR, Rao S, Rath SK, Thakur A, Patri M. Crosslinking of sulphonated poly (ether ether ketone) using aromatic bis(hydroxymethyl) compound. J Membr Sci 2008;322(1):67e73. [19] Kerres VJ, Ullrich A, Meier F, Ha¨ring T. Synthesis and characterization of novel acidebase polymer blends for

12828

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

application in membrane fuel cells. Solid State Ionics 1999;125(1e4):243e9. Cui W, Kerres J, Eigenberger G. Development and characterization of ion- exchange polymer blend membranes. Sep Purif Technol 1998;14(1e3):145e54. Silva VS, Ruffmann B, Silvaa H, Gallego YA, Mendes A, Madeira LM, et al. Proton electrolyte membrane properties and direct methanol fuel cell performance: I. characterization of hybrid sulfonated poly(ether ether ketone)/zirconium oxide membranes. J Power Sources 2005;140(1):34e40. Nunes SP, Ruffmann B, Rikowski E, Vetter S, Richau K. Inorganic modification of proton conductive polymer membranes for direct methanol fuel cells. J Membr Sci 2002;203(1e2):215e25. Kim DS, Liu B, Guiver MD. Influence of silica content in sulfonated poly(arylene ether ether ketone ketone) (SPAEEKK) hybrid membranes on properties for fuel cell application. Polymer 2006;47(23):7871e80. Chang JH, Park JH, Park GG, Kim CS, Park OO. Protonconducting composite membranes derived from sulfonated hydrocarbon and inorganic materials. J Power Sources 2003;124(1):18e25. Alberti G, Casciola M, Palombari R. Inorgano-organic proton conducting membranes for fuel cells and sensors at medium temperatures. J Membr Sci 2000;172(1e2):233e9. Alberti G, Casciola M, Alessandro ED, Pica M. Preparation and proton conductivity of composite ionomeric membranes obtained from gels of amorphous zirconium phosphate sulfophenylenphosphonates in organic solvents. J Mater Chem 2004;14:1910e4. Xu T, Hou W, Shen X, Wu H, Li X, Wang J, et al. Sulfonated titania submicrospheres-doped sulfonated poly(ether ether ketone) hybrid membranes with enhanced proton conductivity and reduced methanol permeability. J Power Sources 2011;196(11):4934e42. Su YH, Liu YL, Sun YM, Lai JY, Wang DM, Gao Y, et al. Proton exchange membranes modified with sulfonated silica nanoparticles for direct methanol fuel cells. J Membr Sci 2007;296(1e2):21e8. Gaowen Z, Zhentao Z. Organic/inorganic composite membranes for application in DMFC. J Membr Sci 2005;261(1e2):107e13. Feher FJ, Wyndham KD, Baldwin RK, Soulivong D, Lichtenhan JD, Ziller JW. Methods for effecting monofunctionalization of (CH2CH2)8Si8O12. Chem Commun 1999:1289e90. Feher FJ, Wyndham KD, Soulivong D, Nguyen FJ. Syntheses of highly functionalized cube-octameric polyhedral oligosilsesquioxanes. Dalton Trans 1999:1491e8. Lichtenhan JD, Vu NQ, Carter JA, Gilman JW, Feher FJ. Silsesquioxaneesiloxane copolymers from polyhedral silsesquioxanes. Macromolecules 1993;26:2141e2. Lichtenhan JD, Otonari YA, Carr MJ. Linear hybrid polymer building blocks: methacrylate-functionalized polyhedral oligomeric silsesquioxane monomers and polymers. Macromolecules 1995;28:8435e7. Mantz RA, Jones PF, Chaffee KP, Lichtenhan JD, Gilman JW, Ismail IMK, et al. Thermolysis of polyheral oligomeric silsesquioxane (POSS) macromers and POSS-siloxane copolymer. Chem Mater 1996;8:1250e9. Haddad TS, Lichtenhan JD. Hybrid organiceinorganic thermoplastics: styryl based polyhedral oligomeric silsesquioxane polymers. Macromolecules 1996;29:7302e4. Huang JC, He CB, Xiao Y, Mya K, Dai J, Siow YP. Polyimide/ POSS nanocomposites: interfacial interaction, thermal properties and mechanical properties. Polymer 2003;44:4491e9.

[37] Choi JW, Yee AF, Laine RM. Organic/inorganic hybrid composites from cubic silsesquioxanes: epoxy resins of octa(dimethylsiloxyethylcyclohexylepoxide) silsesquioxane. Macromolecules 2003;36:5666e82. [38] Choi JW, Tamaki R, Kim SG, Laine RM. Organic/inorganic imide nanocomposites form aminophenylsilsesquioxanes. Chem Mater 2003;15:3365e75. [39] Tsai MH, Whang WT. Low dielectric polyimide/ poly(silsesquioxane)-like nanocomposite material. Polymer 2001;42:4197e207. [40] Wright ME, Schorzman DA, Feher FJ, Jin RZ. Synthesis and thermal curing of aryl-ethynyl-terminated coPOSS imide oligomers: newinorganic/organic hybrid resins. Chem Mater 2003;15:264e8. [41] Schwab JJ, Lichtenhan JD. Polyhedral oligomeric silsesquioxane (POSS) based polymers. Appl Organomet Chem 1998;12:707e13. [42] Li G, Wang L, Ni H, Pittman CU. Polyhedral oligomeric silsesquioxane (POSS) polymers and copolymers: review. J Inorg Organomet Polym 2001;11:123e54. [43] Gnanasekaran D, Madhavan K, Reddy BSR. Developments of polyhedral oligomeric silsesquioxanes (POSS), POSS nanocomposites and their applications: a review. J Sci Ind Res 2009;68:437e64. [44] Thompson CH, Merrington A, Carver PI, Keeley DL, Rousseau JL, Hucul D, et al. J App Polym Sci 2008;110: 958e74. [45] Decker B, Hartmann-Thompson C, Carver PI, Keinath SE, Santurri P. Multilayer sulfonated polyhedral oligosilsesquioxane (S-POSS)-sulfonated polyphenylsulfone (S-PPSU) composite proton exchange membranes. Chem Mater 2010;22(3):942e8. [46] Subianto S, Mistry MK, Choudhury NR, Dutta NK, Knott R. Composite polymer electrolyte containing ionic liquid and functionalized polyhedral oligomeric silsesquioxanes for anhydrous PEM applications. ACS Appl Mater Interface 2009;1:1173e82. [47] Misra R, Alidedeoglu AH, Jarrett WL, Morgan SE. Molecular miscibility and chain dynamics in POSS/polystyrene blends: control of POSS preferential dispersion states. Polymer 2009;50(13):2906e18. [48] Ni Y, Zheng SX, Nie KM. Morphology and thermal properties of inorganiceorganic hybrids involving epoxy resin and polyhedral oligomeric silsesquioxanes. Polymer 2004;45(16):5557e68. [49] Bishop MT, Karasz FE, Russo PS, Langley KH. Solubility and properties of a poly(aryl ether ketone) in strong acids. Macromolecules 1985;18(1):86e93. [50] Zaidi SMJ, Mikhailenko SD, Robertson GP, Guiver MD, Kaliaguine S. Proton conducting composite membranes from polyether ether ketone and heteropolyacids for fuel cell applications. J Membr Sci 2000;173:17e34. [51] Muthu Laxmi RTS, Choudhary V, Varma IK. Sulfonated poly (ether ether ketone): synthesis and characterization. J Mater Sci 2005;40:629e36. [52] Huang RYM, Shao P, Burns CM, Feng X. Sulfonation of poly (ether ether ketone) (PEEK): kinetic study and characterization. J App Polym Sci 2001;82:2651e60. [53] Gupta D, Choudhary V. Studies on novel heat treated sulfonated poly(ether ether ketone) [SPEEK]/diol membranes for fuel cell applications. Int J Hydrogen Energy 2011;36:8525e35. [54] Gupta D, Choudhary V. Sulfonated poly(ether ether ketone)/ ethylene glycol/polyhedral oligosilsesquioxane hybrid membranes for fuel cell applications. Int J Hydrogen Energy 2012;37:5979e91. [55] Xu K, Chanthad C, Gadinski MR, Hickner MA, Wang Q. Acid-functionalized polysilsesquixaneeNafion composite

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 2 8 1 7 e1 2 8 2 9

membranes with high proton conductivity and enhanced selectivity. App Mater Interface 2009;11: 2573e9. [56] Di Vona ML, Licoccia S, Knauth P. Organiceinorganic hybrid membranes based on sulfonated poly(aryl ether ketones):

12829

correlation between water uptake and electrical conductivity. Solid State Ionics 2008;179:1161e5. [57] Wu J, Haddad TS, Mather PT. Vertex group effects in entangled polystyrene- polyhedral oligosilsesquioxane (POSS) copolymers. Macromolecules 2009;42:1142e52.