- Email: [email protected]
Synthesis and proton conductivity of heteropolyacids loaded Y-zeolite as solid proton conductors for fuel cell applications
Microporous and Mesoporous Materials 91 (2006) 296–304 www.elsevier.com/locate/micromeso
Synthesis and proton conductivity of heteropolyacids loaded Y-zeolite as solid proton conductors for fuel cell applications Mohd. Irfan Ahmad a, S.M. Javaid Zaidi a
a,*
, S.U. Rahman a, Shakeel Ahmed
b
Department of Chemical Engineering, King Fahd University of Petroleum and Minerals, KFUPM, Box 1956, Dhahran 31261, Saudi Arabia b Research Institute, King Fahd University of Petroleum and Minerals, KFUPM, Dhahran 31261, Saudi Arabia Received 26 April 2005; received in revised form 25 October 2005; accepted 26 October 2005 Available online 7 February 2006
Abstract A novel solid proton conducting material has been prepared by loading various weight percentages of heteropolyacids (HPA) onto Y-zeolite. The synthesis conditions have been optimized to ensure complete loading of HPAs onto Y-zeolite structure. The proton conductivity of the prepared material enhanced with the loading of HPAs onto Y-zeolite and strongly affected by the presence of water, reaching to more than 20 times compared to dry conditions. The highest conductivity of the order of 102 S/cm was found at room temperature for fully hydrated solid proton conductors containing 50 wt.% HPAs. The prepared materials have been characterized by FT-IR, SEM and X-ray diffraction, which confirm the presence of heteropolyacids into Y-zeolite structure. Leaching study carried out on the powder samples, confirmed that the material leached out through the experiment was negligible and hence almost complete loading of HPAs into the Y-zeolite structures was achieved. The new material combines the high thermal and structural stability of Y-zeolite with outstanding conductivity of HPAs, which places them among one of the most promising solid proton conductors. 2005 Published by Elsevier Inc. Keywords: Proton conductivity; Heteropolyacids; Solid proton conductors; Y-zeolite; Impedance spectroscopy; Composite material
1. Introduction Solid proton conducting materials have grown in interest due to their potential applications in various electrochemical devices such as fuel cells, batteries, sensors and electrochromic display devices [1]. Although, a vast number of various solid electrolytes have already been identified but the development of chemically and thermally stable superionic conductors still remains one of the prime goals of research in solid state electrochemistry and material science [2]. Currently considerable efforts are being devoted to the synthesis and characterization of inorganic–organic hybrid materials, which form a new class of solids with properties, combining the high chemical and thermal stability of inorganic part with the high functionality and simple processing of organic polymers [2]. *
Corresponding author. Tel.: +966 3860 1242; fax: +966 3860 4234. E-mail address: [email protected] (S.M. Javaid Zaidi).
1387-1811/$ - see front matter 2005 Published by Elsevier Inc. doi:10.1016/j.micromeso.2005.10.029
In the present study, a novel solid proton conductor containing Y-zeolite and heteropolyacids has been synthesized and characterized. The prepared material combines the high thermal and structural stability of Y-zeolite with high conductivity of heteropolyacids. Zeolites are crystalline aluminosilicates. They are composed of cations (generally metal ions, but also ammonium ions or hydrogen ions) and an aluminosilicate anion framework. Zeolites are adsorbents for gases and liquids and they are cation exchange materials [3]. They have also been used in liquid phase acid catalytic reactions. Zeolites are reported to be conductors and that the conduction was assumed by migration of ions and not by electrons [4]. In addition Y-zeolite possesses excellent mechanical strength. All of these properties combined with high conductivity of heteropolyacid make it a promising material for the development of proton conducting material. Heteropolyacids are well known for their high proton conductivity at ambient as well as high temperatures. The
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
highest reported protonic conductivities in inorganic solids at temperatures near ambient are those of the heteropolyacids [5]. Heteropolyacids (HPAs) are one of the most attractive inorganic modifiers due to their crystalline form which has been demonstrated to be highly conductive and thermally stable. HPAs have different hydrated structures depending on the environment [6]. In the dehydrated phase in polar solvents the primary structure is called a keggin unit. The keggin unit consists of a central atom in a tetrahedral arrangement of oxygen atoms surrounded by 12 oxygen octahedra connected with tungsten or molybdenum. There are four types of oxygen atoms found in the keggin unit; the central oxygen atoms, two types of bridging oxygen atoms, and the terminal oxygen atoms. In the hydrated phase, water moieties bridge the HPA molecules by forming hydronium ions such as H5 Oþ 2 . The proton conduction mechanism for the HPAs has been extensively investigated [6]. The direct use of pure HPAs as solid electrolytes suffers from some drawbacks due to their high solubility in water and alcohols and strong influence of humidity on proton conductivity. In order to overcome these limitations, there have been several attempts to disperse HPA’s in polymers to make composite membranes for high temperature fuel cells [4,7]. However, high proton conductivity normally requires high relative humidity. In addition, HPAs are mobile and could leach out of the membranes. Heteropolyacids loaded Y-zeolite were shown to be effective shapeselective catalysts for alkylation and esterification reactions [9]. These types of solids can have strong acidity and hydrophilicity at elevated temperatures (e.g. higher than 100 C). Thus, they may prove useful as an effective additive to polymer electrolytes for applications in direct methanol fuel cell (DMFC) and in hydrogen proton exchange membrane fuel cells (PEMFC) to increase operating temperature to 150 C. It is interesting to note that the proton conductivity of the well dispersed HPA loaded Y-zeolite system can be quite promising at temperatures higher than 100 C [7,8], as it combines the high thermal and structural stability of Y-zeolite with high conductivity of HPAs. It is suggested that specific interactions between HPAs and Y-zeolite could have a significant influence on fuel cell performance at elevated temperatures. These phenomena have had relatively little attention until now. Therefore, it was concluded that it would be desirable to investigate the proton conductivity of the HPA loaded Y-zeolite solid powder. In this communication the preparation, conductivity study and characterization of heteropolyacids/Y-zeolite as solid proton conductors with different weight percentages of HPAs supported on Y-zeolite are described. 2. Experimental 2.1. Synthesis of heteropolyacids loaded Y-zeolite Analytical grade 99.5% tungstophosphoric acid (TPA) and molybdophosphoric acid (MPA) were used as received
297
from Fluka chemicals. The Y-zeolite CT-417 (H-form) was supplied by CATAL, UK. The method of Mukai et al. [9] of wet impregnation with some modifications was followed to load the various weight percentages of heteropolyacids onto Y-zeolite. In order to establish the procedure by which strong HPA and Y-zeolite interaction would be achieved, three procedures were adopted to impregnate HPAs on Y-zeolite. A predetermined amount of TPA was dissolved in distilled water and few drops of dilute HCl were added in order to avoid hydrolysis of TPA. Then required amount of Y-zeolite was added to make a suspension. The suspension was stirred and evaporated at 80 C until dryness. Then the solid was ground to fine particles and dried at 200 C for 6 h in air flowing oven. The second procedure involved overnight soaking of the suspension at room temperature before evaporating at 80 C following the same procedure mentioned above. In the third instance, the suspension was homogenized using ultrasonic gun in pulsating mode of operation for 30 min. After that the suspension was stirred and evaporated at 80 C following further drying at 200 C. The solids were then finely ground and collected for further characterization. The resulting materials were evaluated by washing the HPA supported Y-zeolite with hot water at 80 C in a beaker for 1 h and then analyzing the washing by atomic absorption spectroscopy for tungsten. The results of AAS analysis showed that small amount of the HPAs was leached out when samples were prepared by ultrasonic treatment. Therefore the procedure comprising ultrasonic treatment was followed to prepare other samples with different loadings of HPAs onto Y-zeolite. The details of all the samples prepared and their description are summarized in Table 1. 2.2. Leaching study of solid acids Leaching study was carried out on some selected samples in order to determine the amount of HPA leaching out from loaded Y-zeolite. It was carried out in a flow apparatus, which consisted of a simple U-tube, one end Table 1 Abbreviations used for various powder samples used in the work Sample designation
Description
Y T M TY MY TY-1 TY-2 TY1 TY2 MY-1 MY-2 MY1 MY2
Pure Y-zeolite Tungstophosphoric acid (TPA) Molybdophosphoric acid (MPA) Y-zeolite + TPA Y-zeolite + MPA Y-zeolite + 20% TPA Y-zeolite + 30% TPA Y-zeolite + 40% TPA Y-zeolite + 50% TPA Y-zeolite + 20% MPA Y-zeolite + 30% MPA Y-zeolite + 40% MPA Y-zeolite + 50% MPA
298
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
of which was connected to distill water supply, while the other end was connected to a conical flask to collect the leached solution. The prepared powder was kept in the tube and both ends of the tube were packed with glass wool. The U-tube was held in an oil bath, which was maintained 80 C throughout the experiment. The distilled water thus passed through the solid synthesized powders was collected and analyzed for molybdenum and tungsten metals, respectively with the help of atomic absorption spectrophotometer (Perkin–Elmer Analyst 100). 2.3. Proton conductivity measurement The proton conductivity of the powdered solid samples was measured by electrochemical impedance spectroscopy (EIS) technique. It was performed over the frequency range 0.1–1.2 · 106 Hz with 10 mV amplitude. A potentiostat (Model 283) and lock in amplifier (Model 5210) connected to a PC connected through GPIB cord, were used to measure the impedance data. The detailed procedure for conductivity measurements used in this work is described elsewhere [10]. Measured amounts of dry solids and water were mixed and the wet powder was immediately placed into a 1.73 cm inner diameter Teflon spacer which was then placed between two stainless steel electrodes of the conductivity cell and clamped therein. The design of conductivity cell is such that four clampers are equidistant which ensure homogeneous and equal pressure in all of the cases. Moreover equal amounts of each sample were used to ensure homogeneity in performing the experiments. Also, the use of Teflon spacer ensured almost zero conductivity because of the insulation properties of Teflon material. The sample weight was measured before and after the impedance test in order to detect any loss of water in the duration of the test. Since the duration of acquisition of a complex impedance spectrum did not exceed more then 10 min, the hydration degree did not change by more than 10–20%. 2.4. Characterization of solids The pore volume and the surface of the solid samples were measured by BET nitrogen adsorption technique using N2 at 77 K by Quanta Chrome model NOVA 1200 sorption analyzer. The FT-IR spectra were measured in transmittance mode on a Perkin–Elmer FC-16 FT-IR spectrometer. Spectra were taken by making pellet of different powder samples with potassium bromide (KBr). Around 4–5 mg of sample was mixed with approximately 200 mg of KBr to prepare the pellets. The spectrum for each pellet was taken with the above-mentioned spectrometer in the range 400– 4000 cm1. X-ray powder diffraction measurements were carried out on a JEOL JDX-3530 X-ray diffractometer instrument. Each sample was gently ground in an agar pestle and mortar. In finely powdered form, many grains come into orien-
tation and greatly improve the quality of diffraction pattern. The powder was packed into a sample holder having a diameter of 25 mm and depth of 3 mm. The surface of the packed sample was smoothed with a piece of flat glass. Cu broad focus tube at 40 kV and 40 mA was used with a divergence slit of 1 and scatter slit of 1. A curved graphic monochromator was used with a receiving slit of 0.2 mm. Scanning speed and interval of data collection was 0.01 and 2h/s, respectively. The diffraction patterns were recorded from 4 to 80 2h. Scanning electron micrographs (SEM) of pure heteropolyacids as well as that of synthesized solids were taken. Small amounts of the samples were spread on adhesive conductive copper tapes attached to a sample holder and examined with the JEOL (Model 5800LV SEM) scanning electron microscope with low vacuum capability. All of the images were taken in backscattered electron mode. 3. Results and discussion 3.1. Leaching study Leaching study was performed to confirm the retention of heteropolyacids loaded onto Y-zeolite and find leaching (if any) out of it. The amounts of tungsten and molybdenum leached out during the leaching experiments over the time period have been determined by atomic absorption spectrophotometer (AAS) technique. The amounts of tungsten and molybdenum (in mg) of the leaching water of different samples as a function of time (h) are given in Table 2. The results show that TPA is relatively strongly bound to the zeolite as compared with that of MPA. However, overall leaching of the metals in both the cases, i.e., either tungsten or molybdenum is appreciably small comparing the amount of HPAs loaded, i.e., 40% and 50%. The amounts of tungsten and molybdenum were found to be decreasing with increasing time intervals. The leaching studies were carried out on the high loaded heteropolyacids powder samples, i.e., 40 and 50 wt.% heteropolyacids, respectively since these samples possessed the highest proton conductivity. Results of leaching study showed strong interaction of heteropolyacids and Y-zeolite since the amounts leached in total were found to be negligibly small.
Table 2 Results of leaching study of powdered solids for tungsten and molybdenum No. of hours
Tungsten (mg)
Molybdenum (mg)
TY1
TY2
MY1
MY2
At start 1 2 3 4
29.31 10.8 7.14 5.04 4.83
32.19 16.8 10.14 7.89 7.26
130.5 26.1 6.6 2.22 2.07
140.1 30.9 8.7 2.94 2.82
Total
57.12
74.28
167.29
185.16
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
299
Fig. 1. Leaching plots of Y-zeolite loaded with 40 and 50 wt.% TPA and MPA.
The results of the leaching study are also plotted in Fig. 1 as CF (in mg/h) vs time (t) to calculate the total amounts of leached metal where C is the concentration of metal leached out in ppm (mg/l) and F is the flow rate of water through the leaching setup. As can be seen from Fig. 1, the amounts leached out through Y-zeolite are quite small and become almost negligible at 4 h. In fact amounts of leached metals were also calculated at time above 4 h but it was found almost negligible because of which the results are not reported here. Actual time span of the leaching studies was around 24 h but since almost negligible leaching after 4 h was observed hence the results upto significant limits are reported. The area under the curve between CF and t represents the total amount of metal leached in mg. Curves obtained were then fitted with a polynomial of third degree and total amounts of tungsten and molybdenum leached out are thus calculated. Once the total amounts are calculated then it becomes really easy to calculate the amounts leached out and consequently the amounts retained on the support which in our case is Y-zeolite. Significantly low amounts of 3.6 and 5 wt.% of initial amounts loaded were found to be leached out in case of solid proton conductors containing 40 and 50 wt.% heteropolyacids, respectively while 7.6 and 10 wt.% of initial amounts were found to be leached out in case of solid proton conductors containing 40 and 50 wt.% MPA, respectively. This study confirmed that the material leached out through the experiment was negligible and hence almost complete loading of HPAs onto the Y-zeolite structure was ensured. As discussed above, it can be clearly seen from Table 2 that the leached amounts become few mg
after 4 h of continuous flow through the solid powders which ultimately brings in the conclusion of almost complete loading of heteropolyacids onto Y-zeolite. This sounds interesting, because some applications of solid electrolytes involve the direct exposure of materials with liquid water for long periods of time such as in the case of direct methanol fuel cell applications. 3.2. Proton conductivity Proton conductivities of Y-zeolite, tungstophosphoric acid (TPA), molybdophosphoric acid (MPA) and Y-zeolite with different loadings of TPA and MPA were measured. Proton conductivity of Y-zeolite at ambient temperature was found to be 1.54 · 105 S/cm. Previous studies have reported the conductivities of Y-zeolite at ambient as well as with varying degree of hydration. At 100% RH and room temperature the conductivity of Y-zeolite has been reported to be 8 · 106 S/cm [4]. Heteropolyacids have been described among the solid acids possessing highest proton conductivities at room as well as high temperatures. The proton conductivity of 3.37 · 103 S/cm has been obtained in this work at ambient conditions for tungstophosphoric acid (TPA). The proton conductivity data for HPA loaded Y-zeolite with different weight percentages of TPA and MPA are plotted in Fig. 2. As can be seen from the figure, the proton conductivities of the powder samples increased with the loading of heteropolyacids. In TPA loaded Y-zeolite enhancement of conductivity is almost linear, while in MPA loaded Y-zeolite conductivity enhancement is not appreciable up to 20 wt.% but increased significantly
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
Proton conductivity, mS/cm
0.05
TY
10
MY
Proton conductivity, mS/cm
300
0.04
0.03
0.02
0.01
Y
TY1
TY2
1
0.1
0.01
0
5
10
15
20
25
30
35
40
wt. percent water 0 0
5
10
15
20
25
30
35
40
45
50
weight percentages of HPAs
Fig. 3. Proton conductivity variation as a function of water content for Y; TY1 and TY2.
afterwards. Proton conductivity for TPA loaded Y-zeolite is higher as compared to MPA loaded Y and it is obvious because TPA was found to be highly proton conductive as compared to MPA. Proton conductivity with 50 wt.% loading of heteropolyacids was found to be highest in both the cases, while no significant difference in conductivity was observed for 40 and 50 wt.% loading of MPA onto Y-zeolite. It is a well known fact that proton conductivity is a water assisted phenomena and sorbed water plays an important role in conduction, bridging particles of zeolites and assisting the ion hopping process. It influences interactions between cations and the negative framework and provides additional charge carriers (H+, OH). Significant increase in protonic conductivity was observed with increasing percentages of water with Y-zeolite. Since samples of Y-zeolite with 40% and 50% TPA and MPA showed highest proton conductivity, proton conductivity measurement with water variation was carried out on these samples. In presence of liquid water, proton conductivity increases by few orders of magnitude. It can be assumed then, that water dissociation increases due to the presence of surface silanols, which leads to the formation of a protonic space charge with increased charge-carrier concentration near the surface [4]. From the results of proton conductivity variation with water content in Y-zeolite for 40 and 50 wt.% of TPA and MPA given in Figs. 3 and 4, it is evident that proton conductivity increased significantly at 30 and 40 wt.% water contents. Conductivity measurements at water contents above 40 wt.% were also tried but made it difficult due to the problem of handling. This handling of solids was very difficult at higher water contents and it also made the measurements questionable. Nevertheless, with 40 wt.% water contents, still very encouraging results were obtained for both of the cases, i.e., TPA and MPA loaded Y-zeolites. It is evident from these graphs that up to 20 wt.% water there is no significant change in the conductivities, however it jumps to very high values at 30 wt.% water and excep-
Proton conductivity, mS/cm
Fig. 2. Effect of loading of TPA and MPA on the proton conductivity of HPA/Y-zeolite. 100
Y
MY1
MY2
10
1
0.1
0.01
0
5
10
15
20
25
30
35
40
wt. percent water
Fig. 4. Proton conductivity variation as a function of water content for Y; MY1 and MY2.
tionally high at 40 wt.% water for both the cases. It can be seen from Figs. 3 and 4 that conductivity of these solid powders is very sensitive to water adsorption and increases with the variation of water contents by two to three orders of magnitude. It changes for instance from 0.051 mS/cm for MPA loaded Y-zeolite at 10% water content to 11.0 mS/cm for 40 wt.% water content. As is known the proton conduction is a water assisted phenomenon, and water has a profound effect on proton conductivity. Higher water generates a more solvated species, which is needed for high proton conductivity. In all cases proton transfer is obviously associated with acid sites grafted within the pores. Hydration allows bridging acidic sites assisting ion hopping and providing additional charge carriers [2]. The heat of hydration for solid proton conducting hydrates (which can be taken as measure for the hydrogen bond interaction) strongly decreases with increasing hydration whereas the water-diffusion coefficient increases with progressive hydration. Therefore, protonic conductivity increases with increasing water content for all solid proton conducting hydrates [11]. The conductivity of the heteropolyacids loaded Y-zeolite of the order of 11 mS/cm (1.1 · 102 S/cm) is comparable with the commonly known solid powdered proton conducting zirconium phosphate, which has a conductivity of about 3 · 102 S/cm [11] and
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
the silica containing tungstophosphoric acid, which has a conductivity of the order of 102 S/cm [15].
BET surface area and pore volume measure measurements have been carried out before and after impregnating HPA on Y-zeolite. The results of BET surface area and pore volume measured by nitrogen adsorption for the powdered solid samples are given in Table 3. From this table the surface area for pure Y-zeolite was found to be 545 m2/g while its pore volume was 0.34 cm3/g. whereas the surface area and pore volume for the 50% TPA supported Y-zeolite (TY2) samples was 297 m2/g and 0.15 cm3/g, respectively. One can see from these results given in Table 3 that a marked reduction of surface area and total pore volume was observed. A reduction of about 45–47% in surface area and about 47–56% reduction in pore volume was found for TPA and MPA supported Yzeolites. This indicates that HPA (TPA and MPA) is blocking the pores of Y-zeolite and as a consequence of their surface area and pore volume are reduced, since the molecular dimensions of HPA do not permit the inclusion of HPA in to the supercage of Y-zeolite through a pore ˚ (Y-zeolite pore size). opening of 7.4 A The FT-IR spectra confirm the existence of TPA and MPA in the solid composite material. Fig. 5 shows the infrared spectra of the pure tungstophosphoric acid (TPA), Y-zeolite and TPA/Y-zeolite composite powders while Fig. 6 shows the IR spectra of the pure molybdo-
Table 3 Surface area and pore volume of the solid powdered samples Sample name
BET surface area (m2/g)
Pore volume (cm3/g)
Y (Y-zeolite) TY2 (Y-zeolite + 50% TPA) MY2 (Y-zeolite + 50% MPA)
545 297 289
0.34 0.15 0.18
% Transmittance
Y
600
T
TY1
700
800
900
Wavenumber (cm-1) Fig. 5. FT-IR spectra for Y; T and TY1 samples.
1000
Y
% Transmittance
3.3. Characterization of powdered solids
301
M
MY2
600
800
1000
Wavenumber
1200
1400
(cm-1)
Fig. 6. FT-IR spectra for Y; M and MY2 samples.
phosphoric acid (MPA), Y-zeolite and MPA/Y-zeolite composite powders. For pure TPA, six characteristic peaks of its keggin structure were observed at 1078, 980, 888, 788, 590 and 524 cm1, respectively [12]. In the IR spectrum of the solid composite solids, the characteristic peaks of the keggin anion were also observed. In addition, we have observed a small shift of (W–Ob–W) band from 888 cm1 in pure TPA to 896 cm1 in the composite powder with 50 wt.% TPA, while a shift of (W–Ob–W) band from 888 cm1 in pure to 898 cm1 in the composite powder with 40 wt.% TPA [13,14]. In case of molybdophosphoric acid loading onto Y-zeolite, six characteristic peaks of its keggin structure were observed at 1402, 1062, 956, 860, 594 and 454 cm1, respectively. In the IR spectrum of the solid composite solids with MPA, a small shift of (Mo–Ob–Mo) band from 956 cm1 in pure MPA to 960 cm1 in the composite solid powder with 40 and 50 wt.% MPA, respectively. The frequency shift of about 8 and 10 cm1 in case of TPA loading, while frequency shift of about 4 cm1 in case of MPA loading reveals that the keggin structure of TPA and MPA interacts with Y-zeolite mostly through corner-shared oxygen (Ob). So from the IR spectra of the solid composite powders, it is found that the keggin structure characteristic of the heteropoly anions (PX12 O3 40 ) is present in these solid powders which is strongly responsible for the high proton conductivity of heteropolyacids. X-ray powder diffraction (XRD) patterns of pure Y-zeolite, pure TPA and MPA and that of solid composite powders are shown in Figs. 7 and 8. Although the intensities change due to the influence of Y-zeolite, the characteristic diffraction peaks of pure TPA and Pure MPA were still observed in the patterns of solid composite powders containing 50 wt.% of TPA and MPA, respectively. This proves the existence of the keggin anions in the Y-zeolite matrix again, and confirms the observation of the infrared spectra findings. By looking at the XRD pattern of Fig. 7 one can see that most of the peaks are coinciding with each other but even then some of the major peaks of TPA on
302
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
CPS (a.u)
Y
T
TY2
0
10
20
30
40
2θ Fig. 7. XRD spectra for Y; T and TY2 samples.
Intensity (cps)
Y
M
MY2
20
30
2θ
40
↓
10
Fig. 8. XRD spectra for Y; M and MY2 samples.
zeolite are observed as marked by arrows. During impregnation the sample was dissolved and that affected the crystallinity of the TPA that was supported on Y-zeolite sample. The broad peaks in Fig. 7. Indicate the loss of crystallinity of TPA sample. However, this does not affect the conductivity of the samples. The XRD patterns of Y-zeolite containing 50% tungstophosphoric acid (TPA) and molybdophosphoric acid (MPA) show the amorphous states, but they are different from that of pure Y-zeolite whose most intense peak exists at about 15.8, while in case of TPA the most intense peak exists at about 7.6 [15]. Diffraction patterns of solid composite powder with 50 wt.% TPA shows its most intense peak at about 8.4 which shows the dominance of TPA on Y-zeolite. This above observation is also strengthened by the impedance spectroscopy of the solid composite powders in which solid composite powders are found to be highly proton conductive. Again, in case of MPA loading on Y-zeolite, the most intense peak for MPA exists at about 10.5, while with 50 wt.% MPA loaded in Y-zeolite, the most intense peak occurs at about 10.6. Again for the
case of MPA loading into Y-zeolite the particular characteristic of heteropolyacid dominate over Y-zeolite and hence confirms the findings of infrared spectra and impedance spectroscopy. For the SEM micrographs given in Fig. 9 the samples were coated with a thin layer of evaporated gold film for 30 s. Gold coating was performed to minimize charging effects and improve imaging in the SEM at a chamber pressure of 102 mbar. Microstructural examination and compositional analyses were carried out using scanning electron microscope fitted with an energy dispersive X-ray spectrometer. An accelerating voltage of 20 keV was used during analysis. Imaging was performed using both secondary and backscattered electron imaging modes. Fig. 9 shows the electron micrograph of pure Y-zeolite, pure TPA, pure MPA and Y-zeolite with 50 wt.% of TPA and MPA, respectively. Fig. 9(a) is the SEM micrograph of the pure Y-zeolite which suggests almost uniform crystal sizes with few large crystals [16]. The large particles could be aggregates of independent crystals of considerably smaller size. SEM micrograph of pure TPA in Fig. 9(b) suggests mixture of small crystals with few big crystals. The micrograph suggests that the large particles are observed possibly due to the polyoxometalate present in the heteropolyacids, which is responsible mainly for the keggin structure. Fig. 9(c) represents the SEM micrograph of the pure molybdophosphoric acid (MPA) in which an array of uniform small crystals is observed with few sparse areas. SEM micrograph of pure MPA suggests more uniformity as compared to pure TPA micrograph. SEM micrograph of Y-zeolite with 50 wt.% tungstophosphoric acid (TPA) (Fig. 9(d)) suggests almost uniform dispersion of HPA into the zeolite structure except very few cluster-like areas. These clusters could actually be aggregates of independent crystals of Y-zeolite and TPA which apparently do not affect proton conductivity. Definite evidence that the TPA is evenly dispersed throughout the thickness comes from the EDAX analysis. Concentrations of tungsten, silicon and aluminum along the layer thickness of the solid powder through EDAX analysis confirm the presence of HPAs onto zeolite structure. Tungsten is indicative of TPA whilst silicon and aluminum are indicative of Y-zeolite. With different weight percentages of heteropolyacids into Y-zeolites, the amount of major constituents like tungsten, molybdenum, silicon and aluminum varied accordingly which complement the FT-IR, XRD and SEM studies. Although, SEM image of Y-zeolite with 50 wt.% molybdophosphoric acid (MPA) (Fig. 9(e)) suggests a poor dispersion of MPA into Y-zeolite, but this sample shows highest conductivity among MPA loaded Y-zeolite. It can be justified because the proton conductivity in solid powders is mainly due to surface movement of the protons and the considerable presence of MPA in Y-zeolite is mainly responsible for the high proton conductivity of this solid composite powder. EDX study for this sample also confirmed poor dispersion of MPA into Y-zeolite.
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
303
Fig. 9. SEM for pure Y-zeolite (a); pure TPA (b); pure MPA (c); TY2 (d) and MY2 (e).
4. Conclusion In this work a novel solid inorganic proton conductor is developed based on the Y-zeolites loaded with heteropolyacids: namely tungstophosphoric (TPA) acid and molybdophosphoric acid (MPA). The preparation procedure has been optimized to ensure almost complete loading of HPAs onto Y-zeolite. Results of the leaching study confirmed that the material leached out through the experiment was less than 10% of initial amounts loaded; hence almost complete loading of HPA’s onto Y-zeolite structures was ensured. Proton conductivity of the prepared material was found to depend on the amount of heteropolyacids present and strongly affected by the presence of water in the sample. The highest proton conductivity of 11.0 mS/ cm was obtained for 50 wt.% MPA loaded Y-zeolite. Characterization of the prepared material by FT-IR, XRD and SEM confirmed the existence of keggin anions in Y-zeolite structure. The regularity of the change of the characteristics peaks in the infrared spectra was also investigated.
Since the conductivity in the presence of water was found to be exceptionally high, these novel solid proton conductors can be used in developing proton conductive membranes for PEMFC especially for direct methanol fuel cells. Acknowledgements We greatly appreciate the support provided by King Fahd University of Petroleum and Minerals (KFUPM) for this study. Authors wish to thank Mr. Abdur Rashid for XRD and SEM and Mr. Khursheed Alam for AAS analysis. References [1] S. Chandra, N. Lakshmi, Phys. Status Solidi (a) 186 (3) (2001) 383– 399. [2] S. Mikhailenko, D. Desplantier-Giscard, C. Danumah, S. Kaliaguine, Micropor. Messopor. Mater. 52 (2002) 29–37.
304
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
[3] P. Colomban, Chem. Solid State Mater. (2003) 581. [4] S.M.J. Zaidi, PhD Thesis, Laval University, Quebec, Canada, 2000. [5] K. Paulraj, S.J. Ethilton, D.P. Padiyan, Cryst. Res. Technol. 35 (1) (2000) 87–94. [6] Y.S. Kim, F. Wang, M. Hickner, T.A. Zawodzinski, J.E. McGrath, J. Membr. Sci. 212 (2003) 263–282. [7] S.M.J. Zaidi, S.D. Mikhailenko, G.P. Robertson, M.D. Guiver, S. Kaliaguine, J. Membr. Sci. 173 (2000) 17–34. [8] B. Tazi, O. Savadogo, Electrochim. Acta 45 (2000) 4329–4339. [9] S.R. Mukai, T. Masuda, I. Ogino, K. Hashimoto, Appl. Catal. A: Gen. 165 (1997) 219–226.
[10] S.M.J. Zaidi, S.D. Mikhailenko, S. Kaliaguine, J. Chem. Soc., Faraday Trans. 94 (11) (1998) 1613–1618. [11] K.D. Kreuer, T. Dippel, N.G. Hainovsky, J. Maier, Ber. Bunsenges. Phys. Chem. 96 (1992) 1736–1742. [12] L. Li, L. Xu, Y. Wang, Mater. Lett. 57 (2003) 1406–1410. [13] P. Staiti, S. Freni, S. Hocevar, J. Power Sources 79 (1999) 250–255. [14] A.S. Arico, V. Baglio, A.D. Blasi, V. Antonucci, Electrochem. Commun. 5 (2003) 862–866. [15] Q. Wu, S. Tao, H. Lin, G. Meng, Mater. Sci. Eng. B68 (2000) 161– 165. [16] V. Tricoli, F. Nannetti, Electrochim. Acta 48 (2003) 2625–2633.
Synthesis and proton conductivity of heteropolyacids loaded Y-zeolite as solid proton conductors for fuel cell applications Mohd. Irfan Ahmad a, S.M. Javaid Zaidi a
a,*
, S.U. Rahman a, Shakeel Ahmed
b
Department of Chemical Engineering, King Fahd University of Petroleum and Minerals, KFUPM, Box 1956, Dhahran 31261, Saudi Arabia b Research Institute, King Fahd University of Petroleum and Minerals, KFUPM, Dhahran 31261, Saudi Arabia Received 26 April 2005; received in revised form 25 October 2005; accepted 26 October 2005 Available online 7 February 2006
Abstract A novel solid proton conducting material has been prepared by loading various weight percentages of heteropolyacids (HPA) onto Y-zeolite. The synthesis conditions have been optimized to ensure complete loading of HPAs onto Y-zeolite structure. The proton conductivity of the prepared material enhanced with the loading of HPAs onto Y-zeolite and strongly affected by the presence of water, reaching to more than 20 times compared to dry conditions. The highest conductivity of the order of 102 S/cm was found at room temperature for fully hydrated solid proton conductors containing 50 wt.% HPAs. The prepared materials have been characterized by FT-IR, SEM and X-ray diffraction, which confirm the presence of heteropolyacids into Y-zeolite structure. Leaching study carried out on the powder samples, confirmed that the material leached out through the experiment was negligible and hence almost complete loading of HPAs into the Y-zeolite structures was achieved. The new material combines the high thermal and structural stability of Y-zeolite with outstanding conductivity of HPAs, which places them among one of the most promising solid proton conductors. 2005 Published by Elsevier Inc. Keywords: Proton conductivity; Heteropolyacids; Solid proton conductors; Y-zeolite; Impedance spectroscopy; Composite material
1. Introduction Solid proton conducting materials have grown in interest due to their potential applications in various electrochemical devices such as fuel cells, batteries, sensors and electrochromic display devices [1]. Although, a vast number of various solid electrolytes have already been identified but the development of chemically and thermally stable superionic conductors still remains one of the prime goals of research in solid state electrochemistry and material science [2]. Currently considerable efforts are being devoted to the synthesis and characterization of inorganic–organic hybrid materials, which form a new class of solids with properties, combining the high chemical and thermal stability of inorganic part with the high functionality and simple processing of organic polymers [2]. *
Corresponding author. Tel.: +966 3860 1242; fax: +966 3860 4234. E-mail address: [email protected] (S.M. Javaid Zaidi).
1387-1811/$ - see front matter 2005 Published by Elsevier Inc. doi:10.1016/j.micromeso.2005.10.029
In the present study, a novel solid proton conductor containing Y-zeolite and heteropolyacids has been synthesized and characterized. The prepared material combines the high thermal and structural stability of Y-zeolite with high conductivity of heteropolyacids. Zeolites are crystalline aluminosilicates. They are composed of cations (generally metal ions, but also ammonium ions or hydrogen ions) and an aluminosilicate anion framework. Zeolites are adsorbents for gases and liquids and they are cation exchange materials [3]. They have also been used in liquid phase acid catalytic reactions. Zeolites are reported to be conductors and that the conduction was assumed by migration of ions and not by electrons [4]. In addition Y-zeolite possesses excellent mechanical strength. All of these properties combined with high conductivity of heteropolyacid make it a promising material for the development of proton conducting material. Heteropolyacids are well known for their high proton conductivity at ambient as well as high temperatures. The
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
highest reported protonic conductivities in inorganic solids at temperatures near ambient are those of the heteropolyacids [5]. Heteropolyacids (HPAs) are one of the most attractive inorganic modifiers due to their crystalline form which has been demonstrated to be highly conductive and thermally stable. HPAs have different hydrated structures depending on the environment [6]. In the dehydrated phase in polar solvents the primary structure is called a keggin unit. The keggin unit consists of a central atom in a tetrahedral arrangement of oxygen atoms surrounded by 12 oxygen octahedra connected with tungsten or molybdenum. There are four types of oxygen atoms found in the keggin unit; the central oxygen atoms, two types of bridging oxygen atoms, and the terminal oxygen atoms. In the hydrated phase, water moieties bridge the HPA molecules by forming hydronium ions such as H5 Oþ 2 . The proton conduction mechanism for the HPAs has been extensively investigated [6]. The direct use of pure HPAs as solid electrolytes suffers from some drawbacks due to their high solubility in water and alcohols and strong influence of humidity on proton conductivity. In order to overcome these limitations, there have been several attempts to disperse HPA’s in polymers to make composite membranes for high temperature fuel cells [4,7]. However, high proton conductivity normally requires high relative humidity. In addition, HPAs are mobile and could leach out of the membranes. Heteropolyacids loaded Y-zeolite were shown to be effective shapeselective catalysts for alkylation and esterification reactions [9]. These types of solids can have strong acidity and hydrophilicity at elevated temperatures (e.g. higher than 100 C). Thus, they may prove useful as an effective additive to polymer electrolytes for applications in direct methanol fuel cell (DMFC) and in hydrogen proton exchange membrane fuel cells (PEMFC) to increase operating temperature to 150 C. It is interesting to note that the proton conductivity of the well dispersed HPA loaded Y-zeolite system can be quite promising at temperatures higher than 100 C [7,8], as it combines the high thermal and structural stability of Y-zeolite with high conductivity of HPAs. It is suggested that specific interactions between HPAs and Y-zeolite could have a significant influence on fuel cell performance at elevated temperatures. These phenomena have had relatively little attention until now. Therefore, it was concluded that it would be desirable to investigate the proton conductivity of the HPA loaded Y-zeolite solid powder. In this communication the preparation, conductivity study and characterization of heteropolyacids/Y-zeolite as solid proton conductors with different weight percentages of HPAs supported on Y-zeolite are described. 2. Experimental 2.1. Synthesis of heteropolyacids loaded Y-zeolite Analytical grade 99.5% tungstophosphoric acid (TPA) and molybdophosphoric acid (MPA) were used as received
297
from Fluka chemicals. The Y-zeolite CT-417 (H-form) was supplied by CATAL, UK. The method of Mukai et al. [9] of wet impregnation with some modifications was followed to load the various weight percentages of heteropolyacids onto Y-zeolite. In order to establish the procedure by which strong HPA and Y-zeolite interaction would be achieved, three procedures were adopted to impregnate HPAs on Y-zeolite. A predetermined amount of TPA was dissolved in distilled water and few drops of dilute HCl were added in order to avoid hydrolysis of TPA. Then required amount of Y-zeolite was added to make a suspension. The suspension was stirred and evaporated at 80 C until dryness. Then the solid was ground to fine particles and dried at 200 C for 6 h in air flowing oven. The second procedure involved overnight soaking of the suspension at room temperature before evaporating at 80 C following the same procedure mentioned above. In the third instance, the suspension was homogenized using ultrasonic gun in pulsating mode of operation for 30 min. After that the suspension was stirred and evaporated at 80 C following further drying at 200 C. The solids were then finely ground and collected for further characterization. The resulting materials were evaluated by washing the HPA supported Y-zeolite with hot water at 80 C in a beaker for 1 h and then analyzing the washing by atomic absorption spectroscopy for tungsten. The results of AAS analysis showed that small amount of the HPAs was leached out when samples were prepared by ultrasonic treatment. Therefore the procedure comprising ultrasonic treatment was followed to prepare other samples with different loadings of HPAs onto Y-zeolite. The details of all the samples prepared and their description are summarized in Table 1. 2.2. Leaching study of solid acids Leaching study was carried out on some selected samples in order to determine the amount of HPA leaching out from loaded Y-zeolite. It was carried out in a flow apparatus, which consisted of a simple U-tube, one end Table 1 Abbreviations used for various powder samples used in the work Sample designation
Description
Y T M TY MY TY-1 TY-2 TY1 TY2 MY-1 MY-2 MY1 MY2
Pure Y-zeolite Tungstophosphoric acid (TPA) Molybdophosphoric acid (MPA) Y-zeolite + TPA Y-zeolite + MPA Y-zeolite + 20% TPA Y-zeolite + 30% TPA Y-zeolite + 40% TPA Y-zeolite + 50% TPA Y-zeolite + 20% MPA Y-zeolite + 30% MPA Y-zeolite + 40% MPA Y-zeolite + 50% MPA
298
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
of which was connected to distill water supply, while the other end was connected to a conical flask to collect the leached solution. The prepared powder was kept in the tube and both ends of the tube were packed with glass wool. The U-tube was held in an oil bath, which was maintained 80 C throughout the experiment. The distilled water thus passed through the solid synthesized powders was collected and analyzed for molybdenum and tungsten metals, respectively with the help of atomic absorption spectrophotometer (Perkin–Elmer Analyst 100). 2.3. Proton conductivity measurement The proton conductivity of the powdered solid samples was measured by electrochemical impedance spectroscopy (EIS) technique. It was performed over the frequency range 0.1–1.2 · 106 Hz with 10 mV amplitude. A potentiostat (Model 283) and lock in amplifier (Model 5210) connected to a PC connected through GPIB cord, were used to measure the impedance data. The detailed procedure for conductivity measurements used in this work is described elsewhere [10]. Measured amounts of dry solids and water were mixed and the wet powder was immediately placed into a 1.73 cm inner diameter Teflon spacer which was then placed between two stainless steel electrodes of the conductivity cell and clamped therein. The design of conductivity cell is such that four clampers are equidistant which ensure homogeneous and equal pressure in all of the cases. Moreover equal amounts of each sample were used to ensure homogeneity in performing the experiments. Also, the use of Teflon spacer ensured almost zero conductivity because of the insulation properties of Teflon material. The sample weight was measured before and after the impedance test in order to detect any loss of water in the duration of the test. Since the duration of acquisition of a complex impedance spectrum did not exceed more then 10 min, the hydration degree did not change by more than 10–20%. 2.4. Characterization of solids The pore volume and the surface of the solid samples were measured by BET nitrogen adsorption technique using N2 at 77 K by Quanta Chrome model NOVA 1200 sorption analyzer. The FT-IR spectra were measured in transmittance mode on a Perkin–Elmer FC-16 FT-IR spectrometer. Spectra were taken by making pellet of different powder samples with potassium bromide (KBr). Around 4–5 mg of sample was mixed with approximately 200 mg of KBr to prepare the pellets. The spectrum for each pellet was taken with the above-mentioned spectrometer in the range 400– 4000 cm1. X-ray powder diffraction measurements were carried out on a JEOL JDX-3530 X-ray diffractometer instrument. Each sample was gently ground in an agar pestle and mortar. In finely powdered form, many grains come into orien-
tation and greatly improve the quality of diffraction pattern. The powder was packed into a sample holder having a diameter of 25 mm and depth of 3 mm. The surface of the packed sample was smoothed with a piece of flat glass. Cu broad focus tube at 40 kV and 40 mA was used with a divergence slit of 1 and scatter slit of 1. A curved graphic monochromator was used with a receiving slit of 0.2 mm. Scanning speed and interval of data collection was 0.01 and 2h/s, respectively. The diffraction patterns were recorded from 4 to 80 2h. Scanning electron micrographs (SEM) of pure heteropolyacids as well as that of synthesized solids were taken. Small amounts of the samples were spread on adhesive conductive copper tapes attached to a sample holder and examined with the JEOL (Model 5800LV SEM) scanning electron microscope with low vacuum capability. All of the images were taken in backscattered electron mode. 3. Results and discussion 3.1. Leaching study Leaching study was performed to confirm the retention of heteropolyacids loaded onto Y-zeolite and find leaching (if any) out of it. The amounts of tungsten and molybdenum leached out during the leaching experiments over the time period have been determined by atomic absorption spectrophotometer (AAS) technique. The amounts of tungsten and molybdenum (in mg) of the leaching water of different samples as a function of time (h) are given in Table 2. The results show that TPA is relatively strongly bound to the zeolite as compared with that of MPA. However, overall leaching of the metals in both the cases, i.e., either tungsten or molybdenum is appreciably small comparing the amount of HPAs loaded, i.e., 40% and 50%. The amounts of tungsten and molybdenum were found to be decreasing with increasing time intervals. The leaching studies were carried out on the high loaded heteropolyacids powder samples, i.e., 40 and 50 wt.% heteropolyacids, respectively since these samples possessed the highest proton conductivity. Results of leaching study showed strong interaction of heteropolyacids and Y-zeolite since the amounts leached in total were found to be negligibly small.
Table 2 Results of leaching study of powdered solids for tungsten and molybdenum No. of hours
Tungsten (mg)
Molybdenum (mg)
TY1
TY2
MY1
MY2
At start 1 2 3 4
29.31 10.8 7.14 5.04 4.83
32.19 16.8 10.14 7.89 7.26
130.5 26.1 6.6 2.22 2.07
140.1 30.9 8.7 2.94 2.82
Total
57.12
74.28
167.29
185.16
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
299
Fig. 1. Leaching plots of Y-zeolite loaded with 40 and 50 wt.% TPA and MPA.
The results of the leaching study are also plotted in Fig. 1 as CF (in mg/h) vs time (t) to calculate the total amounts of leached metal where C is the concentration of metal leached out in ppm (mg/l) and F is the flow rate of water through the leaching setup. As can be seen from Fig. 1, the amounts leached out through Y-zeolite are quite small and become almost negligible at 4 h. In fact amounts of leached metals were also calculated at time above 4 h but it was found almost negligible because of which the results are not reported here. Actual time span of the leaching studies was around 24 h but since almost negligible leaching after 4 h was observed hence the results upto significant limits are reported. The area under the curve between CF and t represents the total amount of metal leached in mg. Curves obtained were then fitted with a polynomial of third degree and total amounts of tungsten and molybdenum leached out are thus calculated. Once the total amounts are calculated then it becomes really easy to calculate the amounts leached out and consequently the amounts retained on the support which in our case is Y-zeolite. Significantly low amounts of 3.6 and 5 wt.% of initial amounts loaded were found to be leached out in case of solid proton conductors containing 40 and 50 wt.% heteropolyacids, respectively while 7.6 and 10 wt.% of initial amounts were found to be leached out in case of solid proton conductors containing 40 and 50 wt.% MPA, respectively. This study confirmed that the material leached out through the experiment was negligible and hence almost complete loading of HPAs onto the Y-zeolite structure was ensured. As discussed above, it can be clearly seen from Table 2 that the leached amounts become few mg
after 4 h of continuous flow through the solid powders which ultimately brings in the conclusion of almost complete loading of heteropolyacids onto Y-zeolite. This sounds interesting, because some applications of solid electrolytes involve the direct exposure of materials with liquid water for long periods of time such as in the case of direct methanol fuel cell applications. 3.2. Proton conductivity Proton conductivities of Y-zeolite, tungstophosphoric acid (TPA), molybdophosphoric acid (MPA) and Y-zeolite with different loadings of TPA and MPA were measured. Proton conductivity of Y-zeolite at ambient temperature was found to be 1.54 · 105 S/cm. Previous studies have reported the conductivities of Y-zeolite at ambient as well as with varying degree of hydration. At 100% RH and room temperature the conductivity of Y-zeolite has been reported to be 8 · 106 S/cm [4]. Heteropolyacids have been described among the solid acids possessing highest proton conductivities at room as well as high temperatures. The proton conductivity of 3.37 · 103 S/cm has been obtained in this work at ambient conditions for tungstophosphoric acid (TPA). The proton conductivity data for HPA loaded Y-zeolite with different weight percentages of TPA and MPA are plotted in Fig. 2. As can be seen from the figure, the proton conductivities of the powder samples increased with the loading of heteropolyacids. In TPA loaded Y-zeolite enhancement of conductivity is almost linear, while in MPA loaded Y-zeolite conductivity enhancement is not appreciable up to 20 wt.% but increased significantly
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
Proton conductivity, mS/cm
0.05
TY
10
MY
Proton conductivity, mS/cm
300
0.04
0.03
0.02
0.01
Y
TY1
TY2
1
0.1
0.01
0
5
10
15
20
25
30
35
40
wt. percent water 0 0
5
10
15
20
25
30
35
40
45
50
weight percentages of HPAs
Fig. 3. Proton conductivity variation as a function of water content for Y; TY1 and TY2.
afterwards. Proton conductivity for TPA loaded Y-zeolite is higher as compared to MPA loaded Y and it is obvious because TPA was found to be highly proton conductive as compared to MPA. Proton conductivity with 50 wt.% loading of heteropolyacids was found to be highest in both the cases, while no significant difference in conductivity was observed for 40 and 50 wt.% loading of MPA onto Y-zeolite. It is a well known fact that proton conductivity is a water assisted phenomena and sorbed water plays an important role in conduction, bridging particles of zeolites and assisting the ion hopping process. It influences interactions between cations and the negative framework and provides additional charge carriers (H+, OH). Significant increase in protonic conductivity was observed with increasing percentages of water with Y-zeolite. Since samples of Y-zeolite with 40% and 50% TPA and MPA showed highest proton conductivity, proton conductivity measurement with water variation was carried out on these samples. In presence of liquid water, proton conductivity increases by few orders of magnitude. It can be assumed then, that water dissociation increases due to the presence of surface silanols, which leads to the formation of a protonic space charge with increased charge-carrier concentration near the surface [4]. From the results of proton conductivity variation with water content in Y-zeolite for 40 and 50 wt.% of TPA and MPA given in Figs. 3 and 4, it is evident that proton conductivity increased significantly at 30 and 40 wt.% water contents. Conductivity measurements at water contents above 40 wt.% were also tried but made it difficult due to the problem of handling. This handling of solids was very difficult at higher water contents and it also made the measurements questionable. Nevertheless, with 40 wt.% water contents, still very encouraging results were obtained for both of the cases, i.e., TPA and MPA loaded Y-zeolites. It is evident from these graphs that up to 20 wt.% water there is no significant change in the conductivities, however it jumps to very high values at 30 wt.% water and excep-
Proton conductivity, mS/cm
Fig. 2. Effect of loading of TPA and MPA on the proton conductivity of HPA/Y-zeolite. 100
Y
MY1
MY2
10
1
0.1
0.01
0
5
10
15
20
25
30
35
40
wt. percent water
Fig. 4. Proton conductivity variation as a function of water content for Y; MY1 and MY2.
tionally high at 40 wt.% water for both the cases. It can be seen from Figs. 3 and 4 that conductivity of these solid powders is very sensitive to water adsorption and increases with the variation of water contents by two to three orders of magnitude. It changes for instance from 0.051 mS/cm for MPA loaded Y-zeolite at 10% water content to 11.0 mS/cm for 40 wt.% water content. As is known the proton conduction is a water assisted phenomenon, and water has a profound effect on proton conductivity. Higher water generates a more solvated species, which is needed for high proton conductivity. In all cases proton transfer is obviously associated with acid sites grafted within the pores. Hydration allows bridging acidic sites assisting ion hopping and providing additional charge carriers [2]. The heat of hydration for solid proton conducting hydrates (which can be taken as measure for the hydrogen bond interaction) strongly decreases with increasing hydration whereas the water-diffusion coefficient increases with progressive hydration. Therefore, protonic conductivity increases with increasing water content for all solid proton conducting hydrates [11]. The conductivity of the heteropolyacids loaded Y-zeolite of the order of 11 mS/cm (1.1 · 102 S/cm) is comparable with the commonly known solid powdered proton conducting zirconium phosphate, which has a conductivity of about 3 · 102 S/cm [11] and
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
the silica containing tungstophosphoric acid, which has a conductivity of the order of 102 S/cm [15].
BET surface area and pore volume measure measurements have been carried out before and after impregnating HPA on Y-zeolite. The results of BET surface area and pore volume measured by nitrogen adsorption for the powdered solid samples are given in Table 3. From this table the surface area for pure Y-zeolite was found to be 545 m2/g while its pore volume was 0.34 cm3/g. whereas the surface area and pore volume for the 50% TPA supported Y-zeolite (TY2) samples was 297 m2/g and 0.15 cm3/g, respectively. One can see from these results given in Table 3 that a marked reduction of surface area and total pore volume was observed. A reduction of about 45–47% in surface area and about 47–56% reduction in pore volume was found for TPA and MPA supported Yzeolites. This indicates that HPA (TPA and MPA) is blocking the pores of Y-zeolite and as a consequence of their surface area and pore volume are reduced, since the molecular dimensions of HPA do not permit the inclusion of HPA in to the supercage of Y-zeolite through a pore ˚ (Y-zeolite pore size). opening of 7.4 A The FT-IR spectra confirm the existence of TPA and MPA in the solid composite material. Fig. 5 shows the infrared spectra of the pure tungstophosphoric acid (TPA), Y-zeolite and TPA/Y-zeolite composite powders while Fig. 6 shows the IR spectra of the pure molybdo-
Table 3 Surface area and pore volume of the solid powdered samples Sample name
BET surface area (m2/g)
Pore volume (cm3/g)
Y (Y-zeolite) TY2 (Y-zeolite + 50% TPA) MY2 (Y-zeolite + 50% MPA)
545 297 289
0.34 0.15 0.18
% Transmittance
Y
600
T
TY1
700
800
900
Wavenumber (cm-1) Fig. 5. FT-IR spectra for Y; T and TY1 samples.
1000
Y
% Transmittance
3.3. Characterization of powdered solids
301
M
MY2
600
800
1000
Wavenumber
1200
1400
(cm-1)
Fig. 6. FT-IR spectra for Y; M and MY2 samples.
phosphoric acid (MPA), Y-zeolite and MPA/Y-zeolite composite powders. For pure TPA, six characteristic peaks of its keggin structure were observed at 1078, 980, 888, 788, 590 and 524 cm1, respectively [12]. In the IR spectrum of the solid composite solids, the characteristic peaks of the keggin anion were also observed. In addition, we have observed a small shift of (W–Ob–W) band from 888 cm1 in pure TPA to 896 cm1 in the composite powder with 50 wt.% TPA, while a shift of (W–Ob–W) band from 888 cm1 in pure to 898 cm1 in the composite powder with 40 wt.% TPA [13,14]. In case of molybdophosphoric acid loading onto Y-zeolite, six characteristic peaks of its keggin structure were observed at 1402, 1062, 956, 860, 594 and 454 cm1, respectively. In the IR spectrum of the solid composite solids with MPA, a small shift of (Mo–Ob–Mo) band from 956 cm1 in pure MPA to 960 cm1 in the composite solid powder with 40 and 50 wt.% MPA, respectively. The frequency shift of about 8 and 10 cm1 in case of TPA loading, while frequency shift of about 4 cm1 in case of MPA loading reveals that the keggin structure of TPA and MPA interacts with Y-zeolite mostly through corner-shared oxygen (Ob). So from the IR spectra of the solid composite powders, it is found that the keggin structure characteristic of the heteropoly anions (PX12 O3 40 ) is present in these solid powders which is strongly responsible for the high proton conductivity of heteropolyacids. X-ray powder diffraction (XRD) patterns of pure Y-zeolite, pure TPA and MPA and that of solid composite powders are shown in Figs. 7 and 8. Although the intensities change due to the influence of Y-zeolite, the characteristic diffraction peaks of pure TPA and Pure MPA were still observed in the patterns of solid composite powders containing 50 wt.% of TPA and MPA, respectively. This proves the existence of the keggin anions in the Y-zeolite matrix again, and confirms the observation of the infrared spectra findings. By looking at the XRD pattern of Fig. 7 one can see that most of the peaks are coinciding with each other but even then some of the major peaks of TPA on
302
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
CPS (a.u)
Y
T
TY2
0
10
20
30
40
2θ Fig. 7. XRD spectra for Y; T and TY2 samples.
Intensity (cps)
Y
M
MY2
20
30
2θ
40
↓
10
Fig. 8. XRD spectra for Y; M and MY2 samples.
zeolite are observed as marked by arrows. During impregnation the sample was dissolved and that affected the crystallinity of the TPA that was supported on Y-zeolite sample. The broad peaks in Fig. 7. Indicate the loss of crystallinity of TPA sample. However, this does not affect the conductivity of the samples. The XRD patterns of Y-zeolite containing 50% tungstophosphoric acid (TPA) and molybdophosphoric acid (MPA) show the amorphous states, but they are different from that of pure Y-zeolite whose most intense peak exists at about 15.8, while in case of TPA the most intense peak exists at about 7.6 [15]. Diffraction patterns of solid composite powder with 50 wt.% TPA shows its most intense peak at about 8.4 which shows the dominance of TPA on Y-zeolite. This above observation is also strengthened by the impedance spectroscopy of the solid composite powders in which solid composite powders are found to be highly proton conductive. Again, in case of MPA loading on Y-zeolite, the most intense peak for MPA exists at about 10.5, while with 50 wt.% MPA loaded in Y-zeolite, the most intense peak occurs at about 10.6. Again for the
case of MPA loading into Y-zeolite the particular characteristic of heteropolyacid dominate over Y-zeolite and hence confirms the findings of infrared spectra and impedance spectroscopy. For the SEM micrographs given in Fig. 9 the samples were coated with a thin layer of evaporated gold film for 30 s. Gold coating was performed to minimize charging effects and improve imaging in the SEM at a chamber pressure of 102 mbar. Microstructural examination and compositional analyses were carried out using scanning electron microscope fitted with an energy dispersive X-ray spectrometer. An accelerating voltage of 20 keV was used during analysis. Imaging was performed using both secondary and backscattered electron imaging modes. Fig. 9 shows the electron micrograph of pure Y-zeolite, pure TPA, pure MPA and Y-zeolite with 50 wt.% of TPA and MPA, respectively. Fig. 9(a) is the SEM micrograph of the pure Y-zeolite which suggests almost uniform crystal sizes with few large crystals [16]. The large particles could be aggregates of independent crystals of considerably smaller size. SEM micrograph of pure TPA in Fig. 9(b) suggests mixture of small crystals with few big crystals. The micrograph suggests that the large particles are observed possibly due to the polyoxometalate present in the heteropolyacids, which is responsible mainly for the keggin structure. Fig. 9(c) represents the SEM micrograph of the pure molybdophosphoric acid (MPA) in which an array of uniform small crystals is observed with few sparse areas. SEM micrograph of pure MPA suggests more uniformity as compared to pure TPA micrograph. SEM micrograph of Y-zeolite with 50 wt.% tungstophosphoric acid (TPA) (Fig. 9(d)) suggests almost uniform dispersion of HPA into the zeolite structure except very few cluster-like areas. These clusters could actually be aggregates of independent crystals of Y-zeolite and TPA which apparently do not affect proton conductivity. Definite evidence that the TPA is evenly dispersed throughout the thickness comes from the EDAX analysis. Concentrations of tungsten, silicon and aluminum along the layer thickness of the solid powder through EDAX analysis confirm the presence of HPAs onto zeolite structure. Tungsten is indicative of TPA whilst silicon and aluminum are indicative of Y-zeolite. With different weight percentages of heteropolyacids into Y-zeolites, the amount of major constituents like tungsten, molybdenum, silicon and aluminum varied accordingly which complement the FT-IR, XRD and SEM studies. Although, SEM image of Y-zeolite with 50 wt.% molybdophosphoric acid (MPA) (Fig. 9(e)) suggests a poor dispersion of MPA into Y-zeolite, but this sample shows highest conductivity among MPA loaded Y-zeolite. It can be justified because the proton conductivity in solid powders is mainly due to surface movement of the protons and the considerable presence of MPA in Y-zeolite is mainly responsible for the high proton conductivity of this solid composite powder. EDX study for this sample also confirmed poor dispersion of MPA into Y-zeolite.
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
303
Fig. 9. SEM for pure Y-zeolite (a); pure TPA (b); pure MPA (c); TY2 (d) and MY2 (e).
4. Conclusion In this work a novel solid inorganic proton conductor is developed based on the Y-zeolites loaded with heteropolyacids: namely tungstophosphoric (TPA) acid and molybdophosphoric acid (MPA). The preparation procedure has been optimized to ensure almost complete loading of HPAs onto Y-zeolite. Results of the leaching study confirmed that the material leached out through the experiment was less than 10% of initial amounts loaded; hence almost complete loading of HPA’s onto Y-zeolite structures was ensured. Proton conductivity of the prepared material was found to depend on the amount of heteropolyacids present and strongly affected by the presence of water in the sample. The highest proton conductivity of 11.0 mS/ cm was obtained for 50 wt.% MPA loaded Y-zeolite. Characterization of the prepared material by FT-IR, XRD and SEM confirmed the existence of keggin anions in Y-zeolite structure. The regularity of the change of the characteristics peaks in the infrared spectra was also investigated.
Since the conductivity in the presence of water was found to be exceptionally high, these novel solid proton conductors can be used in developing proton conductive membranes for PEMFC especially for direct methanol fuel cells. Acknowledgements We greatly appreciate the support provided by King Fahd University of Petroleum and Minerals (KFUPM) for this study. Authors wish to thank Mr. Abdur Rashid for XRD and SEM and Mr. Khursheed Alam for AAS analysis. References [1] S. Chandra, N. Lakshmi, Phys. Status Solidi (a) 186 (3) (2001) 383– 399. [2] S. Mikhailenko, D. Desplantier-Giscard, C. Danumah, S. Kaliaguine, Micropor. Messopor. Mater. 52 (2002) 29–37.
304
M.I. Ahmad et al. / Microporous and Mesoporous Materials 91 (2006) 296–304
[3] P. Colomban, Chem. Solid State Mater. (2003) 581. [4] S.M.J. Zaidi, PhD Thesis, Laval University, Quebec, Canada, 2000. [5] K. Paulraj, S.J. Ethilton, D.P. Padiyan, Cryst. Res. Technol. 35 (1) (2000) 87–94. [6] Y.S. Kim, F. Wang, M. Hickner, T.A. Zawodzinski, J.E. McGrath, J. Membr. Sci. 212 (2003) 263–282. [7] S.M.J. Zaidi, S.D. Mikhailenko, G.P. Robertson, M.D. Guiver, S. Kaliaguine, J. Membr. Sci. 173 (2000) 17–34. [8] B. Tazi, O. Savadogo, Electrochim. Acta 45 (2000) 4329–4339. [9] S.R. Mukai, T. Masuda, I. Ogino, K. Hashimoto, Appl. Catal. A: Gen. 165 (1997) 219–226.
[10] S.M.J. Zaidi, S.D. Mikhailenko, S. Kaliaguine, J. Chem. Soc., Faraday Trans. 94 (11) (1998) 1613–1618. [11] K.D. Kreuer, T. Dippel, N.G. Hainovsky, J. Maier, Ber. Bunsenges. Phys. Chem. 96 (1992) 1736–1742. [12] L. Li, L. Xu, Y. Wang, Mater. Lett. 57 (2003) 1406–1410. [13] P. Staiti, S. Freni, S. Hocevar, J. Power Sources 79 (1999) 250–255. [14] A.S. Arico, V. Baglio, A.D. Blasi, V. Antonucci, Electrochem. Commun. 5 (2003) 862–866. [15] Q. Wu, S. Tao, H. Lin, G. Meng, Mater. Sci. Eng. B68 (2000) 161– 165. [16] V. Tricoli, F. Nannetti, Electrochim. Acta 48 (2003) 2625–2633.