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 x x x ( 2 0 1 5 ) 1 e8 Available online at www.sciencedirect.com ScienceDirect journal homepage: www.elsevier.com/locate/he Composite sPEEK-TPyP membranes development for portable applications  a, R. Pedicini a, I. Gatto a, E. Passalacqua a, A. Carbone a,*, A. Sacca A. Romeo b, L. Monsù Scolaro b, M.A. Castriciano c a Istituto di Tecnologie Avanzate per l'Energia “Nicola Giordano”, via S. Lucia Sopra Contesse 5, 98126 Messina, Italy Dipartimento di Scienze Chimiche, University of Messina Viale Ferdinando Stagno D'Alcontres n.31, 98166 Villaggio S. Agata, Messina, Italy c Istituto per lo Studio dei Materiali Nanostrutturati, c/o Dipartimento di Scienze Chimiche Viale Ferdinando Stagno D'Alcontres n.31, 98166 Villaggio S. Agata, Messina, Italy b article info abstract Article history: Composite membranes based on sulphonated Polyetheretherketone (sPEEK) and 5,10,15,20- Received 9 February 2015 tetra(4-pyridyl)porphyrin (TPyP) were developed for portable applications. A sulphonation Received in revised form degree of 65% and different loadings (0e5 wt%) of TPyP porphyrin were used. Spectroscopic 28 July 2015 studies (UVeVis and fluorescence emission), ionic exchange capacity, water uptake, Accepted 29 July 2015 structural and morphological analyses were carried out. A good interaction between the Available online xxx filler and the polymer was found, probably due to specific supramolecular interactions among the nitrogenous atoms present in the periphery of the porphyrin ring and sulphonic Keywords: groups of polymer. The membrane with the lowest loading (~1 wt%) of TPyP, shows slightly Sulphonated PEEK lower water uptake and unaltered l values than the reference membrane, resulting in quite TPyP unaltered proton conductivity also in anhydrous conditions. Electrochemical tests were PEFC performed in a PEFC 25 cm2 single cell at room temperature and anhydrous hydrogen/ Portable applications humidified air. A power density of 93 mW/cm2 was found in anhydrous conditions for sample with the lowest loading of TPyP. Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. Introduction Fuel cells are clean and highly efficient electrochemical systems for energy conversion, that are fed by hydrogen in advanced systems with the highest performance, or by alternative fuels (e.g. methanol or ethanol) with significant reduction in performance [1]. Among the existing typologies of fuel cells, the system that has attracted most interest is the Proton Exchange Membrane Fuel Cell (PEMFC) due to its wide range of power density (1  10 3e100 kW), simplicity of components and its friendly user operating conditions. A polymer electrolyte membrane fuel cell (PEMFC) is composed of a membrane able to conduct protons, between the anodic and cathodic compartments, separating two catalytic electrodes where the fuel oxidation and the comburent reduction take place. * Corresponding author. Tel.: þ39 090624273. E-mail address: alessandra.carbone@itae.cnr.it (A. Carbone). http://dx.doi.org/10.1016/j.ijhydene.2015.07.159 0360-3199/Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Carbone A, et al., Composite sPEEK-TPyP membranes development for portable applications, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.07.159 2 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 x x x ( 2 0 1 5 ) 1 e8 Due to the high energy density with respect to the traditional batteries, fuel cells are particularly suitable for the fabrication of portable systems, more efficient and less bulky than currently available systems. The Portable Fuel Cell systems tend to use either DMFC (direct methanol fuel cells) or PEFC (polymer electrolyte fuel cells fed by hydrogen) technology, even if the DMFC are nowadays the most diffused. However, several problems to be solved, such as catalyst poisoning during methanol oxidation reaction and methanol crossover through solid-polymer-electrolyte membrane, limit their commercialization [2]. The most used polymer electrolyte membrane is Nafion®, that has high proton conductivity, chemical and mechanical stability, but suffers a high methanol permeability, that drastically reduces the performance. To solve this problem, associated to a high production cost of this polymer, the research aims at developing low-cost and highly efficient proton conducting membranes to be used as polymer electrolytes for portable fuel cells fed with hydrogen and air under air-breathing conditions. The development of poly-aromatic based membranes plays a key role as an alternative to Nafion® membranes [3e7]. A wide literature exists on this class of polyelectrolyte but some problems should be overcome such as a low proton conductivity at reduced relative humidity, the mechanical stability and lifetime [8]. One of the most used polymer membrane is based on sulphonated polyetheretherketone (sPEEK) with a sulphonation degree in the range 40e70% [9e14]. Compared to the membranes reported in the state-of-the-art, the alternative membranes exhibit reduced cost (by a factor of about 40); better mechanical properties; improved stability to radical species; reduced hydrogen crossover; similar unit area resistance. In addition to the above mentioned characteristics, nowadays, the research is moving on the development of membranes containing nitrogenous groups that promotes specific acid-base interactions between functional aminosulphonic groups, by improving the proton conductivity [15e17]. In fact, the sulphonic acid groups interact with the nitrogenous base by forming hydrogen bridges, protonation of the nitrogen sites and polysalts formation. In this framework, to the best of our knowledge, only a report has been addressed to a class of membranes functionalized with porphyrins as nitrogenous sources [18]. Porphyrins are a class of versatile molecules and porphyrin-based architectures have diverse potential applications as biomimetic models, as functional materials for the transport of energy, charge, molecules, and ions [19] as well as hydrogen ion sieve structure [20e23]. Furthermore, it is widely reported in literature, that porphyrins, due to their chemical structure, are able, in particular experimental condition, to self-aggregate forming extended supramolecular network [24]. With this aim, the ability of sulphonated polyetheretherketone (sPEEK) membranes to specifically interact with 5,10,15,20-tetra(4-pyridyl)porphyrin (TPyP) was exploited and its proton path for the conduction mechanism for applications in portable devices was investigated. Experimental Polymer sulphonation A Polyetheretherketone (Victrex PF450) was functionalized in concentrated sulphuric acid according to a standardized procedure, reported elsewhere [25], in order to obtain a sulphonation degrees (SD) of 65%. Membrane preparation Membranes were prepared with a doctor-blade standardized procedure [25] Composite membranes were further prepared using the sulphonated polymer. The 5,10,15,20-tetrakis(4pyridyl)-21H,23H-porphine (TPyP) was purchased from Aldrich Chemical Co. and used without further purification. The porphyrin was added to the polymer in different weight percentages 0.77, 1.5 and 5wt% in DMAc solution. In order to better solubilize the porphyrin, the resulting mixture was stirred for 1 h before to be added to the polymeric solution. All the membranes were dried at 80  C for 3 h then detached from the glass by impregnation with H2O. A thermal treatment was carried out at 120  C for 16 h, An acid treatment was carried out at first in 1 M H2SO4 at 60  C and then in water at the same temperature, in order to purify the membranes of any residual solvent and activate groups for proton ion exchange for the coordination of water molecules. Membranes with a thickness ranging from 50 to 70 mm were obtained. Polymers and membranes X-ray analyses (XRD) The X-ray powder diffraction (XRD) analyses were performed by using a Philips X-ray automated diffractometer (model PW3710) with Cu Ka radiation source. The 2q Bragg angles were scanned between 5 and 100 2q. UVeVis and fluorescence UV/Vis spectra were obtained on a HewlettePackard mod. 8453 diode array spectrophotometer directly on the membranes. Fluorescence measurements were carried out on a Jasco mod.FP-750 spectrofluorimeter. SEMeEDX analyses A field emission Scanning Electron Microscope (Philips mod. XL30 S FEG) was used to investigate the cross-section morphology of the membranes. A gold coating was used to avoid sample charging and to permit electronic conduction. Samples for cross-section were prepared dipping the membrane in liquid Nitrogen and breaking the samples to have a perfect fracture. Ion exchange capacity (IEC) An acid-base titration was carried out to determine the membrane IECm (meqSO3H/mg) and is based on the Please cite this article in press as: Carbone A, et al., Composite sPEEK-TPyP membranes development for portable applications, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.07.159 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 x x x ( 2 0 1 5 ) 1 e8 neutralization of Hþ ions, belonging to the acid group eSO3H. The experimental procedure is described elsewhere [25] The IEC is calculated using the following formula: IECm ¼ (Vtit$[M])/mdry (1) where: To determine the interaction between SPEEK and TPyP, IEC of SPEEK polymer (IECp) within the composite membrane was determined considering the Polymer Fraction (PF) calculated from Eq. (2): (2) where IECm is the experimental IEC of the membrane. Water uptake, lambda and dimensional variation The ability to retain the water of the membrane (wu) is usually calculated from the difference in weight between the dried and the wet sample. The wet weight (mwet) is determined after immersion of the sample in distilled water at room temperature for 24 h, while for the dry weight (mdry), the sample is dried in a vacuum oven at 80  C for 2 h. The percentage of water absorbed is given by the following expression: wu ¼ (mwet mdry)/mdry 100 (3) The l value (expressed as moles H2O/moles-SO3H) was calculated through the water uptake and IEC values ratio, both expressed in moles: l ¼ (mol H2O)/(mol SO3H) (4) Through this parameter is possible to evaluate the capability of the SO3H groups in the composite membranes to coordinate water molecules. The dimensional variation (A%) was calculated in terms of geometrical are percentage using the following expression: A% ¼ (Awet Adry)/Adry 100 (5) Proton conductivity The conductivity was measured at 30  C and full humidification (100% RH) using a commercial cell (Bekktech). The measure is carried out in the longitudinal direction of the sample with a four-probe method and DC current, with a potentiostategalvanostat (AMEL mod.2049) and calculated using the formula: s ¼ L/RWT where: L ¼ 0.425 cm, constant distance between the two Pt electrodes; R ¼ resistance in U; W ¼ sample width in cm; T ¼ sample thickness in cm. All the measurements were carried out after an equilibration time of 1 h after reaching the operative conditions. After measuring the 100%RH the gas was turned to dry and 1 h was waited before to perform the measurement. Vtit ¼ titrant volume (ml); [M] ¼ titrant molarity; mdry ¼ dry mass of the sample (g); IECp ¼ IECm/PF 3 (6) Fuel cell tests Home-made electrodes were prepared by a spray technique described in the referenced paper [17,26] and coupled to the membranes to obtain MEAs. The same Pt loading (0.5 mg/cm2) in the catalytic layer was used for both anodes and cathodes and a 50% Pt/C (Alfa Aesar) was used as an electrocatalyst. A Sigracet-24BC (SGL group) was used as a gas diffusion layer. Membranes and electrodes were assembled by cold-pressing. Fuel cell tests, in terms of polarisation curves, were carried out in a commercial 25 cm2 single cell at 30  C, with fully humidified H2/air and dry H2/humidified air at 1 absolute bar. The gas fluxes were fixed at 1.5 and 2 times the stoichiometry at the current work for hydrogen and air, respectively. Electrochemical impedance spectroscopy (EIS) was performed by using a potentiostat/galvanostat (AUTOLAB PGSTAT30) equipped with a frequencies response analyzer (FRA module) and a 20 A BOOSTER. All impedance measurements were performed in the potentiostatic mode of fuel cell operation at a constant potential of 850 mV. The impedance spectra were obtained varying the frequency of the voltage perturbation signal from 0.1 Hz to 100 kHz, by using amplitude of 10 mV for the perturbing signal. The H2 cross-over of the membranes was measured by a linear sweep voltammetry method, in the same operative conditions of IeV curves, by varying the cell potential from 0 to 0.8 V with a scan rate of 4 mV/s. The current value at 0.4 V was used for cross-over calculation. Results and discussion In order to test the dispersion of porphyrin in the polymeric matrix, the membranes were first characterized in terms of XRD profile. In Fig. 1 is reported the comparison among the XRD patterns of pristine SPEEK membrane, composite membranes with different amount of tetra(4-pyridyl)porphine (TpyP, hereafter) and TPyP powder. The spectra show the typical amorphous profile of the sulphonated polymer centred at about 18 2q, that remains quite unaltered after the functionalization of the membrane with TPyP. The absence of TPyP peak is a clear indication that porphyrin is present in the membranes in its monomeric form or as small oligomers/ aggregates. Due to their structural features, i.e. the extended planarity and the presence in periphery of the ring of specific groups, the porphyrins are able to induce van der Waals, p e stacking Please cite this article in press as: Carbone A, et al., Composite sPEEK-TPyP membranes development for portable applications, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.07.159 4 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 x x x ( 2 0 1 5 ) 1 e8 Fig. 1 e XRD patterns of developed membranes compared to TPyP. and hydrophobic interactions, that allow, in appropriate experimental conditions, the formation of dimers, trimers, small oligomers or larger aggregates [27]. The aggregation phenomena for this class of molecules are accompanied by characteristic changes in their spectroscopic properties, i.e. bathochromic or hypsochromic shifts of the absorption and emission bands, Beer law's deviation, quenching of the luminescence. The photophysical properties of TPyP in its monomeric form have been reported in organic solvents [28,29] as well as the modulation of the protonation behaviour and of the self-aggregation [30,31]. To get better inside into the aggregation state of the TPyP porphyrin into membranes, spectroscopic investigation on the composites were performed. Figs. 2 and 3 show the UVeVis and fluorescence emission spectra of SPEEK-0.77% TPyP. The position of the absorbance as well as the emission bands remain almost unchanged for all the investigated samples. Due to high porphyrin Fig. 2 e Extinction spectra of SPEEK-0.77% TPyP (solid line) and, after immersion in 1 M NaOH aqueous solution (dashed line), T ¼ 298 K. Inset extinction value after cyclic dipping into acidic (1 M HCl) and alkaline solutions. concentrations it was not possible to detect the B-band typical for this chromophores and it was taken into account the position of the Q-bands (500e700 nm) which have a lower extinction coefficient with respect to the B-band. The extinction spectra of the membranes show the presence of 2 Qbands centred at 595 and 645 nm, respectively The fluorescence emission spectrum shows two bands centred at 662 and 718 nm. According to literature, these spectroscopic behaviour, especially the presence of 2 Q bands indicate a D4h geometry for the porphyrin ring, pointing out the occurring, into the membranes, of the porphyrin core protonation [30] Independently, from the thickness membranes and from the porphyrin loads, the chromophores embedded into membranes exhibit good stability and chemical resistance to thermal annealing and acidic treatment. In fact, the spectroscopic features before and after the treatments remain unchanged for all investigated samples. Furthermore, protonation and deprotonation of the porphyrin can be easily obtained by simple immersion of the membrane in an alkaline solution. As reported in Fig. 2, dipping the membranes into alkaline aqueous solutions the green colour of the samples turns instantaneously to red and its extinction spectrum is reported in Fig. 2. The presence of 4 Q-bands in the extinction spectrum centred at 518, 552, 590, 645 nm respectively, clearly indicate a D2h geometry for the porphyrin ring, pointing out the occurrence of the porphyrin core deprotonation. This evidence is supported by the presence of two emission bands centred at 650 and 716 nm in the fluorescence spectrum [30]. Porphyrin protonation can be restored and the systems switch fast and reversibly upon repetitive cycling by dipping into base and acidic solutions (inset of Fig. 2). The switching properties are extremely reproducible and the membranes remain stable for months. SEM analyses were performed to check the morphology of composite membrane compared to pristine SPEEK one (Fig. 4). In both cases a dense and homogeneous structure was highlighted, meaning that no agglomeration of TPyP occurred, even for membrane with the highest TPyP content (reported in Fig. 4b). Fig. 3 e Fluorescence emission spectrum of SPEEK-5% TPyP SPEEK-0.77% TPyP (solid line) and, after immersion in 1 M NaOH aqueous solution (dashed line), T ¼ 298 K. Please cite this article in press as: Carbone A, et al., Composite sPEEK-TPyP membranes development for portable applications, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.07.159 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 x x x ( 2 0 1 5 ) 1 e8 5 Fig. 4 e SEM analysis of a) SPEEK membrane, b) SPEEK-5%TPyP membrane. To verify the influence of TPyP on the proton exchange properties of SPEEK, the experimental (IECm) and the calculated IECp were compared and reported in Fig. 5. The IECm decreases with the increase of porphyrin amount, because the TPyP has not own exchangeable protons. In addition, the ability of SPEEK polymer to form interactions with nitrogenous groups is known and elsewhere reported [25]. Also with the TPyP introduction, specific interactions between sulphonic and nitrogenous groups are present, as highlighted from IECp data. This values were calculated by considering the nominal amount of TPyP introduced in the membranes and the obtained values are related to the SPEEK polymer in the composite membranes. As reported in Fig. 5, the IECp is still lower than pristine SPEEK membrane, indicating that interactions occurred and limit the proton exchange capacity. The behaviour of water uptake (w.u.) and dimensional variation (area %) is shown in Fig. 6, as a function of the TPyP content. In accordance to IEC behaviour, both w.u. and A% decrease by increasing filler loading, meaning that the TPyP interaction with the sulphonic groups of the PEEK reduces the swelling of the polymeric matrix. An interesting trend is highlighted in Fig. 7, where is reported the proton conductivity as a function of the lambda values. Because of the lambda values is Fig. 5 e IEC behaviour as a function of TPyP loading. Fig. 6 e Water uptake and area variation as a function of TPyP amount. generally related to the hydration of sulphonic groups of the polymer, the capability to maintain unaltered the properties to coordinate water in composite membranes is an important feature to be considered. As expected, the proton conductivity increases with the increase of l values and for samples SPEEK- Fig. 7 e Proton conductivity as a function of lambda values. Please cite this article in press as: Carbone A, et al., Composite sPEEK-TPyP membranes development for portable applications, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.07.159 6 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 x x x ( 2 0 1 5 ) 1 e8 Table 1 e Proton conductivity values at different operative conditions. Table 2 e Electrochemical data at 100%RH H2/air. Membrane OCV Rs (EIS) H2 cross-over PD @ 0.7 V V Ucm2 at 0.4 V mW/cm2 2 700 mV mA/cm SPEEK SPEEK-0.77%TPyP SPEEK-1.5%TPyP SPEEK-5%TPyP 0.985 1.000 0.999 1.000 Proton conductivity at 30  C, mS/cm Membrane SPEEK SPEEK-0.77%TPyP SPEEK-1.5%TPyP SPEEK-5%TPyP l, mol H2O/SO3H 100%RH 13 13 12 11 42 36 26 7 Dry <0.01 10 <0.01 <0.01 0.64 0.65 0.82 1.98 1.2$10 1.2$10 3.1$10 9.2$10 4 5 4 5 65 80 50 24 0.77% TPyP and pristine SPEEK recast the proton conductivity is similar due to the same l value (l ¼ 13). Moreover, only the SPEEK-0.77% TPyP membrane shows an appreciable value of proton conductivity in dry H2 condition (see Table 1), this behaviour could mean that this membrane is less sensitive to dehydration phenomena. Because the setup of the measurement is different from fuel cell, in which the membrane is constrained in the MEA, in this kind of measurement the sample is more available to hydration changes and can prompt respond to these changes. The SPEEK-0.77 TPyP sample is the only that maintains a suitable hydration to reach a proton conduction, due to its intrinsic properties. The other samples showed a drop in the proton conduction with a reported value < 0.01 mS/cm, under the instrument sensitivity limit, due to a rapid de-hydration. To verify the single cell performance useful for portable applications, the composite membranes, were tested in the operative conditions of low temperature and pressure and compared to recast SPEEK. In Fig. 8, the polarisation curves at 30  C, 1 ab bar and full hydrated gases are shown. Since the useful voltage range, for a real application, is between 0.6 and 0.7 V, the IeV curves are cut off at 200 mA/cm2. The performance of composite membranes are in accordance to the previous chemical-physical and ex-situ electrochemical data, in fact the performance decreases with the increase of TPyP amount. The SPEEK-0.77% TPyP shows slight higher performance than SPEEK recast membrane. All the investigated samples present good OCV values (Table 2), confirming the dense structure of the membranes. The increase of Rs values is due to TPyP introduction which interacts with sulphonic groups limiting the protonic transport. Despite the Rs value of sPEEK-0.77%TPyP is higher than the recast one, the performance increase is due to the lowest H2 cross-over, confirmed also from OCV data. The power density at 0.7 V is reported in Table 2, it is still evident the benefit of the small amount of TPyP (0.77%) in the polymer matrix, in fact a power density of 80 mW was reached against 65 mW of SPEEK. To simulate the real application in portable devices where gases are not humidified, operative conditions were changed in dry H2/100%RH air, supposing a hydrated cathode for the atmospheric humidity (air-breathing) and the water reaction formation. The polarisation curves are reported in Fig. 9. Also in this case the trend is similar to the full humidification. The performance slightly increased despite the dry condition of the anode. This behaviour could be probably caused by a not good water management in full humidification operative conditions, that affects the performance. Also in these operative conditions the OCV data remains good, the H2 cross-over is reduced of one order of magnitude, in particular for sPEEK-0.77%TPyP respect SPEEK membrane, and the power density is increased, with the highest value of 93 mW/cm2 at 0.7 V for membrane containing 0.77wt% TPyP (Table 3). Anyhow, the SPEEK-0.77% TPyP membrane represents the best compromise between chemicalephysical properties and Fig. 8 e IeV curves comparison at low T, P and 100%RH H2/ air. Fig. 9 e IeV curves comparison at low T, P and dry H2/100% RH air. Please cite this article in press as: Carbone A, et al., Composite sPEEK-TPyP membranes development for portable applications, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.07.159 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 x x x ( 2 0 1 5 ) 1 e8 Table 3 e Electrochemical data at dry H2/100%RH air. Membrane OCV Rs (EIS) H2 cross-over PD @ 0.7 V V Ucm2 at 0.4 V mW/cm2 2 700 mV mA/cm SPEEK SPEEK-0.77%TPyP SPEEK-1.5%TPyP SPEEK-5%TPyP 0.985 1.018 0.985 1.000 0.59 0.78 0.75 2.55 2.9$10 2.4$10 2.8$10 1.0$10 3 4 3 4 85 93 65 29 electrochemical performance and it could be considered a good candidate to be used in a stack for a real portable application. Conclusions Composite membranes based on sulphonated Polyetheretherketone (sPEEK) and 5,10,15,20-tetra(4-pyridyl) porphyrin (TPyP) were developed for portable applications. A sulphonation degree of 65% and different weight percentages (0e5%) of TPyP porphyrin were used for membranes realization. A complete chemical-physical and electrochemical characterization was carried out. The TPyP introduction does not modify the structural and morphology properties of the polymer while some changes occurs in the IEC, water uptake, dimensional variation and lambda values. This behaviour is attributed to the formation of interactions between the sulphonic groups of the polymer and the nitrogenous group of the TPyP. This interaction with a hydrogen sieve material such as TPyP, produces a not aggregated form of the TPyP but a good interaction with the polymeric matrix with a network formation. The proton conductivity of the membrane with the lowest loading (0.77wt% TPyP) does not change respect the pristine SPEEK reference, despite a reduction of water uptake and area % and is maintained in the same order of magnitude in dry conditions. In addition, during fuel cell operation at low temperature, pressure and dry hydrogen, this sample supplies the best power density. The properties of this membrane configuration renders it a good candidate for a real application in portable devices. references [1] Hoogers Gregor, editor. Fuel cell technology handbook. U.S.: CRC Press; 2003. [2] Meenakshi S, Bhat SD, Sahu AK, Sridhar P, Pitchumani S. Modified sulfonated poly(etheretherketone) based mixed matrix membranes for direct methanol fuel cells. Fuel Cells 2013;13(5):851e61. re J. Recent advances in the functionalisation [3] Jones DJ, Rozie of polybenzimidazole and polyetherketone for fuel cell applications. J Membr Sci 2001;185:41e58. [4] Alberti G, Casciola M, Massinelli L, Bauer B. Polymeric proton conducting membranes for medium temperature fuel cells (110e160 C). J Membr Sci 2001;185:73e81. [5] Kreuer KD. Hydrocarbon membranes. In: Vielstich W, Lamm A, Gasteiger H, editors. Handbook of fuel cell technology. London: Wiley & Sons; 2003. p. 420e35. 7 [6] Kreuer KD. On the development of proton conducting materials for technological applications. Solid State Ionics 1997;97:1e15. [7] Carbone A, Pedicini R, Portale G, Longo A, D'Ilario L, Passalacqua E. Sulphonated poly(ether ether ketone) membranes for fuel cell application: thermal and structural characterisation. J Power Sources 2006;1(163):18e26. [8] Sung KA, Kim W-K, Oh K-H, Choo M-J, Nam K-W, Park J-K. Proton exchange membranes based on hydrophilicehydrophobic multiblock copolymers using different hydrophobic oligomer. Int J Hydrogen Energy 2011;36:3956e64. [9] Peighambardoust SJ, Rowshanzamir S, Amjadi M. Review of the proton exchange membranes for fuel cell applications. Int J Hydrogen Energy 2010;35:9349e84. [10] Han SY, Park J, Kim D. Proton-conducting electrolyte membranes based on organosiloxane network/sulfonated poly(ether ether ketone) interpenetrating polymer networks embedding sulfonated mesoporous benzeneesilica. J Power Sources 2013;243:850e8. [11] Park J, Kim D. Sulfonated mesoporous benzene-silicaembedded sulfonated poly(ether ether ketone) membranes for enhanced proton conduction and anti-dehydration. Int J Hydrogen Energy 2014;39:1063e70. [12] Wang SJ, Zhang YF, Shu D, Tian SH, Mei DH, Xiao M, et al. Portable PEMFC stack using sulfonated poly(fluorenyl ether ketone) ionomer as membrane. Int J Hydrogen Energy 2012;37:4539e44. [13] Carbone A, Gatto I, Ohira A, Wu L, Passalacqua E. Influence of post-casting treatments on sulphonated polyetheretherketone composite membranes. J Power Sources 2010;195:6037e42.  A, Gatto I, Pedicini R, Passalacqua E. [14] Carbone A, Sacca Investigation on composite S-PEEK/H-BETA MEAs for medium temperature PEFC. Int J Hydrogen Energy 2008;33:3153e8. [15] Kerres J, Cui W. Acid-base polymer blends and their application in membrane processes, Patent US 6,300,381 B1, 2001. [16] Gao Y, Robertson GP, Guiver MD, Jian X, Mikhailenko SD, Kaliaguine S. Proton exchange membranes based on sulfonated poly(phthalazinone ether ketone)s/aminated polymer blends. Solid State Ionics 2005;176:409e15. [17] Zuo Zicheng, Fu Yongzhu, Manthiram Arumugam. Novel blend membranes based on acid-base interactions for fuel cells. Polymers 2012;4:1627e44. [18] Yu Fei, Zhou Zhen-Tao, Zhang Guo Ge, Modified sulfonated poly ether ether ketone (SPEEK) membrane applied to direct methanol fuel cell (DMFC) and preparation method of SPEEK membrane Patent CN 102020781 B, 2012. [19] Lin VS-Y, Di Magno SG, Therien MJ. Highly conjugated, acetylenyl bridged porphyrins: new model for lightharvesting antenna systems. Science 1994;264:1105e11. [20] Diskin-Posner Y, Goldberg I. From porphyrin sponges to porphyrin sieves: a unique crystalline lattice of aquazinc tetra(4-carboxyphenyl)porphyrin with nanosized channels. Chem Commun 1999:1961e2. [21] Diskin-Posner Y, Patra GK, Goldberg I. Supramolecular porphyrin-based materials. Assembly modes of [5,10,15,20tetrakis(4-hydroxyphenyl)porphyrinato]zinc with bipyridyl ligands. Cryst Eng Comm 2002;4(53):296e301. [22] Lin Kuan-Jiuh. SMTP-1: the first functionalized metalloporphyrin molecular sieves with large channels. Angew Chem Int Ed 1999;38(18):2730e2. [23] Goldberg I. Metalloporphyrin Molecular sieves. Chem Eur J 2000;6:3863e70. [24] Micali N, Villari V, Castriciano M, Romeo A, Scolaro L Monsù. From fractal to nanorod porphyrin J-aggregates. Please cite this article in press as: Carbone A, et al., Composite sPEEK-TPyP membranes development for portable applications, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.07.159 8 [25] [26] [27] [28] 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 x x x ( 2 0 1 5 ) 1 e8 Concentration-induced tuning of the aggregate size. J Phys Chem B 2006;110:8289e95.  A, Gatto I, Passalacqua E. Carbone A, Pedicini R, Sacca Composite S-PEEK membranes for medium temperature polymer electrolyte fuel cells. J Power Sources 2008;178:661e6.  A, Carbone A, Pedicini R, Urbani F, Gatto I, Sacca Passalacqua E. CO-tolerant electrodes developed with PhosphoMolybdic Acid for Polymer Electrolyte Fuel Cell (PEFCs) application. J Power Sources 2007;171:540e5. Fuhrhop JP. Porphyrin assemblies and their scaffolds. Langmuir 2014;30:1e12. Kalyanasundaram K. Photochemistry of water-soluble porphyrins: comparative study of isomeric tetrapyridyl- and tetrakis(N-methylpyridiniumyl)porphyrins. Inorg Chem 1984;23:2453e9. [29] Akins DL, Zhu HR, Guo C. Aggregation of tetraarylsubstituted porphyrins in homogeneous solution. J Phys Chem 1996;100:5420e5. [30] De Luca Giovanna, Romeo Andrea, Scolaro Luigi Monsù. Role of the counter-anions on acid-induced aggregation of isomeric Tetrakis(pyridyl)porphyrins in organic solvents. J Phys Chem B 2005;109(15):7149e58.  A, Donato MG, Micali N, [31] Castriciano M, Carbone A, Sacca Romeo A, et al. Optical and sensing features of TPPS4 Jaggregates embedded in nafion membranes: influence of casting solvents. J Mater Chem 2010;20:2882e6. Please cite this article in press as: Carbone A, et al., Composite sPEEK-TPyP membranes development for portable applications, International Journal of Hydrogen Energy (2015), http://dx.doi.org/10.1016/j.ijhydene.2015.07.159