- Email: [email protected]
The influence of Ferric ion contamination on the solid polymer electrolyte water electrolysis performance
Accepted Manuscript Title: The influence of Ferric ion contamination on the solid polymer electrolyte water electrolysis performance Author: Xunying Wang Linsong Zhang Guangfu Li Geng Zhang Zhi-Gang Shao Baolian Yi PII: DOI: Reference:
S0013-4686(15)00186-3 http://dx.doi.org/doi:10.1016/j.electacta.2015.01.140 EA 24192
To appear in:
Electrochimica Acta
Received date: Revised date: Accepted date:
22-12-2014 24-1-2015 24-1-2015
Please cite this article as: Xunying Wang, Linsong Zhang, Guangfu Li, Geng Zhang, Zhi-Gang Shao, Baolian Yi, The influence of Ferric ion contamination on the solid polymer electrolyte water electrolysis performance, Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2015.01.140 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The influence of Ferric ion contamination on the solid polymer electrolyte water electrolysis performance Xunying Wang a,b, Linsong Zhang a,c, Guangfu Li a,b, Geng Zhang a,b, Zhi-Gang Shaoa, *, Baolian Yia
a
Fuel Cell System and Engineering Laboratory, Dalian Institute of Chemical Physics, Chinese Academy of
Sciences, 457 Zhongshan Road, Dalian 116023, PR China b
Graduate School of the Chinese Academy of Sciences, Beijing 100039, PR China c
Xingtai Polytechnic College, Xingtai 054035, PR China
* Corresponding Author: Telephone: +86 411 84379153; Fax: +86 411 84379185; E-mail: [email protected]
Highlights
The cathode possesses higher tolerance for the Fe3+ contamination than the anode.
Fe3+ are mostly reduced to Fe2+ rather than occur underpotential deposition.
Increased electrolysis voltage was mainly attributed to ohmic overpotential.
Voltage lags behind current for minutes in the multi-current-step test.
Poisoned electrolyser is mostly recovered by 0.5M H2SO4 solution treatment for 13h
1
Abstract: Fe3+ is a sort of common metal ion contaminant for the solid polymer electrolyte (SPE) water electrolyser. In this paper, the effect of Fe3+ on the performance of SPE water electrolyser has been investigated by both in-situ and ex-situ characterizations. The electron probe microanalysis and ultraviolet test results showed that Fe3+ could migrate from the anode to the cathode and mostly be reduced to Fe2+ in the cathode rather than occurred underpotential deposition as described in the previous report. The in-situ dynamic contamination test showed that the anode voltage increased sharply as soon as the Fe3+ was fed into the anode, while the cathode voltage kept constant until the contamination time was over 30 minutes, indicating the higher tolerance of the cathode than the anode for the Fe3+ contamination. The calculation results based on the electrochemistry impedance spectroscopy test results revealed that the striking increase of the electrolysis voltage was mainly attributed to the ohmic overpotential, which was due to the replacement of H+ by Fe3+ in the Nafion resin. Interestingly, the voltage lagged behind the current for several minutes in the multi-current-step test for the contaminated electrolyser, which phenomenon may be used for judging whether the SPE water electrolyser performance degradation is due to the metal ions contamination. Furthermore, recovery strategy has been developed, and it is found that the contaminated electrolyser could be mostly recovered by 0.5 M H2SO4 solution treatment for 13 h.
2
Key words: Solid polymer electrolyte water electrolysis; Ferric ion impurity; In-situ; Valence alternation
1. Introduction Solid polymer electrolyte (SPE) water electrolysis provides a promising method to produce hydrogen from renewable but intermittent energy sources [1-5]. Compared with the conventional alkaline water electrolyser, SPE water electrolyser possesses the advantages of higher safety, simplicity, greater energy efficiency, higher specific production capacity and lower maintenance [2-4, 6-8]. However, during water electrolysis process, the membrane electrode assembly is susceptible to cation impurities, including Na+, Ca2+, Cu2+, Ni2+ and Fe3+, originating from the fabrication of the MEA, the impurity of feedwater, the corrosion of water pipes and stack component materials, and other sources [9-13]. The cations may affect the electrolysis performance by exchanging with protons in the Nafion electrolyte and depositing at the cathode. The replacement of protons by metal ions significantly impacts the ionic conductivity of the Nafion electrolyte [14-17]. This is because that most of the metal ions possess higher affinities than protons for the sulfonic acid groups in Nafion, and that the mobility of the metal ions in Nafion electrolyte is quite lower than that of protons [18]. Moreover, the replacement of protons by metal ions impacts water transportation in Nafion electrolyte, for the much faster diffusion of H3O+ than that of the complex metal ion - water clusters [M-(H2O)m]n+ [14, 19]. 3
The exchanged metal ions, driven by the cell voltage and concentration difference, will transfer across the Nafion membrane to the cathode. Some metal ions, such as Cu2+, have positive reversible potential compared with the standard hydrogen electrode (SHE), thus bulk deposition of these metal ions can occur at the cathode of the water electrolyser. Some metal ions, e.g. Sn2+, possess negative reversible potential versus SHE, can occur underpotential deposition of monolayer on the surface of the catalyst [20]. Some other metal cations, such as Ca2+, have too negative Nernst potential and can not deposit on the cathode catalyst. But they could precipitate as hydroxides at the interface between the Nafion membrane and the cathode catalyst layer, thus decrease the conductivity of the MEA. The metal ions contamination is a slowly accumulative process during SPE water electrolysis. Therefore, it is necessary to take detail study on the dynamic process of the metal ions contamination to the electrolyser. However, there are limited reports about this study [10, 20-22]. Kötz et al.[20] discovered that RuO2 cathodes were insensitive to metal ions contamination for the underpotential deposition phenomenon did not occur on RuO2. Wei [10] et al. provided evidence for the accumulation of metal ions in the feed-water during electrolysis. The influence of Na+ on the SPE water electrolysis was studied in our previous work[21, 22]. It was found that 0.05 M Na+ caused serious performance degradation to the electrolyser, which mainly resulted from the cathodic overpotential increase for the reaction of 2H2O + 2e- = H2 +2OH- .
4
Fe3+ is a sort of common metal ion contaminant for the SPE water electrolyser. It can originate from the impurities of feed-water, the corrosion of water pipes and stack component materials. Understanding how and to what extent Fe3+ affects the SPE water electrolyser can help to choose proper feed-water purity standards and component materials with acceptable corrosion rates, develop analytic methods and recovery statics for the poisoned electrolyser. Herein, the effect of Fe3+ on the SPE water electrolyser performance was investigated by both in-situ contamination test and ex-situ characterization, the recovery strategy was also developed.
2. Experimental
2.1 Preparation of MEA The CCM was prepared by spraying 5 wt. % of Nafion solution (0.6 mg cm-2) and isopropanol onto each side of Nafion membrane (Dupont 115), followed by spraying catalyst layers. Iridium black (Johnson Matthey) and Pt/C (70 wt. %, Johnson Matthey) were used as the anode and cathode catalyst, respectively. Homogeneous ink consisting of catalyst (Iridium black or Pt/C), Nafion solution (5 wt. %, Du Pont) and isopropanol was sprayed onto the Nafion membrane with Nafion layers to form the catalyst layers. The catalyst loading of the CCM was about 2.0 mg cm−2 for iridium black and 1.0 mg cm−2 for Pt/C. Porous titanium (Pt plated, 0.7 mm in thickness) and wet-proof carbon paper (Toray, TGP-H-60) served as the anode and cathode diffusion layer, respectively. The MEA was fabricated by hot pressing the 5
CCM and the carbon paper together.
2.2 Single cell test and ferric contamination A home-made electrolysis test stand was employed to evaluate the performance of the CCM. The SPE water electrolyser performance was evaluated at atmosphere pressure and 80 oC. After the electrolyser operating galvanostatically at 500 mA cm−2 for 30 min with deionized water fed to the anode side at the flow rate of 10 mL min-1, 500 mL solution containing 0.005 mol L-1 Fe2(SO4)3 and 0.01 mol L-1 H2SO4 was continuously delivered to the anode side. The add of sulphuric acid was to avoid the hydrolysis of Fe3+. The water from the cathode was exported into a water tank containing 500 mL deionized water. The pH changes of the anode feed water and cathode water were measured by immersing the basic pH meter PB-10 (Sartorius AG) into the water tank. The single electrode potentials were measured against a saturation mercury electrode (SCE, E = 0.242 V vs. SHE) and finally calibrated against SHE. As schematically shown in Fig.1, the SCE was placed in 0.5 mol L-1 H2SO4 solution, then an ionic contact can be formed by immersing the piece of Nafion membrane extending from the MEA in 0.5 mol L-1 H2SO4 solution [21-24]. Electrochemistry impedance spectroscopy (EIS) was carried out at 1.45 V vs. SHE by Solartron 1287 Electrochemical Interface in conjunction with Solartron 1260 Frequency Response Analyzer [25].
6
2.2 Physical characterizations The iron element distribution on the cross-section of the contaminated CCM was investigated by electron probe microanalysis (EPMA, SHIMAZU EPMA-1600). The changes in weight percentage of iron element in the cathode water were determined by inductively coupled plasma (ICP, Optima, 2000DV) analysis. The valence of iron element in the cathode water was determined by phenanthroline spectrophotometry method [26] on a UV-VIS spectrolphotometer (SHIMADZU, UV-2550). All the chemicals used were of analytical grade. The test solution for determining the total iron content in cathode water was prepared by putting 10 mL cathode water into a 100 mL volumetric flask, then sequentially adding 50 mL deionized water, HCl solution to adjust the PH to 2, 2.5 mL ascorbic acid (20 g L-1) solution to reduce Fe3+ to Fe2+, 10 mL acetic acid-sodium acetate buffer solution (PH=5), 5 mL phenanthroline solution (2 g L-1), and finally diluting to the mark with deionized water. The test solution for determining the Fe2+ content in cathode water was prepared with the above procedure except that ascorbic acid solution was not used. The absorbance was measured at 510 nm and the Fe2+ concentration was read from the standard concentration curve. The Fe3+ concentration could be determined by difference.
3. Results and discussion Fig.2 showed the effects of 0.01 M Fe3+ on the electrolysis performance. As shown in Fig.2a, the anode voltage severely increased initially until reaching a maximum, 7
then slowly dropped and finally became steady. According to the previous reports [15, 27, 28], the mobility of Fe3+ in the Nafion membrane was between 2.6*10-9 – 7.1*10-9 m2 V-1 s-1, so it only needs no more than 5 s for Fe3+ to transfer from anode surface to cathode when the thickness of Nafion 115 membrane and anode catalyst layer is about 125 μm and 10 μm, respectively. What’s more, the ICP analysis (Fig.4) also showed that Fe element appeared in the cathode water when the electrolyser was contaminated for 5 min, which further verified that the Fe3+ reached the cathode catalyst layer. However, the cathode voltage was not influenced until when the contamination time was over 30 minutes, indicating that the cathode possessed higher tolerance for the Fe3+ contamination than the anode. It may be due to the much faster rate of proton reduction reaction on the Pt catalyst surface than that of the water oxidation reaction on the Ir and IrO2 catalysts. During the contamination, the PH changes of the feed water and the cathode water were also measured (Fig.2b). It can be seen that both the anode and the cathode PH decreased with contamination time. The initial sharp decrease of anode PH was due to the introduction of solution containing 0.005 M Fe2(SO4)3 and 0.01 M H2SO4. The EPMA test result at the cross-section of the contaminated CCM was shown in Fig.3. The distribution of Fe in the anode, membrane and cathode can be observed, demonstrating that Fe3+ migrated from the anode to the cathode during the contamination. There was lower content of Nafion in the catalyst layers than in the membrane, and Fe3+ mainly attached on the sulfonic acid groups in Nafion electrolyte, so the amount of Fe in the catalyst layers was smaller than that in the membrane. The 8
larger amount of Fe in the cathode than in the anode was due to the voltage difference between the anode and the cathode. Fig. 4 showed the change in weight percentage of Fex+ in the cathode water. It can be seen that the weight percentage of Fex+ increased with contamination time, indicating that Fex+ kept flowing out of the electrolyser, rather than being retained by the CCM. Furthermore, phenanthroline spectrophotometry was used to study the valence of Fex+ in the cathode water. Test result showed that 92% of the Fex+ in the cathode water was Fe2+, suggesting that most of the Fe3+ transferred to the cathode was reduced to Fe2+ rather than occurred underpotential deposition as described in the previous report [20]. The higher affinity of Fex+ than H+ to the sulfonic sites led to the replacement of H+ in Nafion electrolyte by Fex+ [18], which would ① reduce the effective triple-phase boundaries and thus increase the activation overpotential, ②
lower proton
concentration in the electrolyte and the proton conductivity of the electrolyte, therefore enhance the concentration overpotential. With the increase of the electrolysis voltage, the speed of Fe3+ transferring out of the anode enhanced, and the Fe3+ content in the anode catalyst layer decreased, thus lead to the decrease of the anode voltage instead (Fig.2a) [22]. In the anode, the protons produced by the water oxidation reaction partially entered the feed water due to the lack of the sulfonic sites, which decreased the PH of the feed water (Fig.2b). Whereas the decrease of the cathode water PH value was mainly caused by the hydrolysis of Fex+. We used EIS to further characterize the fresh and contaminated electrolyser, 9
respectively (Fig.5). The high frequency intercept with the real axis and the diameter of the medium frequency arc are the measures of the ohmic resistance (RΩ) and charge-transfer resistance (Rct), respectively. The EIS of the contaminated electrolyser exhibited a Warburg line at the low-frequency, indicating the diffusion resistance of the chemical species. The equivalent circuits to simulate the EIS data of the fresh and contaminated
electrolyser
were
presented
by
LRΩ(R1Q1)(RctQdl)
and
LRΩ(R1Q1)((RctZw)Qdl), respectively, where RΩ, Rct, Q, Zw and L are the ohmic resistance, charge-transfer resistance, capacitive element, Warburg element and inductor, respectively. The capacitive element is often presented by a constant phase element Q to model depressed semi-circle due to heterogeneities and rough surfaces. The Qdl represents double layer capacitance. The R1Q1 circuit is used to describe various phenomena, e.g. the diffusion of protons along the catalysts. The simulating results suggested that the RΩ values for the fresh and contaminated electrolyser were 0.235 and 1.34 Ω cm2, and the Rct values were 0.126 and 3.81 Ω cm2, respectively. It can be seen that both RΩ and Rct increased significantly after contamination. The increased ohmic overpotential induced by the Fe3+ contamination was about 548 mV at 500 mA cm-2. It accounted for 80% of the increased cell voltage (685 mV) when the electrolyser was contaminated for 320 min (Fig.2a), indicating that the ohmic overpotential was the primary reason for the cell performance loss. The substitution for the protons by Fe3+ decreased the conductivity of the Nafion electrolyte, and the relationship between the Fe3+ content (
) and the conductivity
(к, S m-1) of the membrane can be expressed by the following equations [24, 29]: 10
(1) (2) where F is Faraday constant, membrane,
and
is the concentration of the sulfonic sites in the
are the mobility of proton and Fe3+, respectively,
and
are the fractions of the sulfonic sites occupied by protons and Fe3+ in the Nafion membrane, respectively. For the water soaked Nafion 115 membrane, the equals 1.18*103 mol m-3, and its
is 8.78*10-8 m2 V-1 s-1 [30]. Previously reported
mobility of the Fe3+-form Nafion membrane was between 0.296 - 0.803 S m-1 [15, 27, 28], and the average value, 0.550 S m-1, is used here. Compared with the membrane resistance, the ohmic resistance caused by the other components can be ignored. So the conductivity of the contaminated Nafion membrane, estimated from the above EIS data, was about 1.09 S m-1. Accordingly, the Fe3+ composition in the membrane (
)
was about 94.3%. Subsequently, the decrease of the reversible electrode potential, , obtained according to the Nernst equation (Eq.3), was 87 mV when 94.3% of the protons in the electrolyte were exchanged, which accounted for about 13% of the increased cell voltage (685 mV). (3)
where
,
, R, T, F,
, and [H+] are the thermodynamic electrode potential,
standard cathode potential, gas constant, cell temperature, Faraday constant, partial pressure (atm) of H2 and molar concentration of protons, respectively. 11
Fig.6a showed the E-I curves of the SPE water electrolyser before and after contamination. Deionized water was used for this electrolysis test. The expected increase of electrolysis voltage was observed after contamination. Furthermore, the increase of anode potential was much smaller than that of cathode potential, indicating that Fe3+ affected the cathode more seriously than the anode, although the cathode had higher tolerance for the Fe3+ than the anode as shown in Fig.1. It was mainly due to the larger amount of Fex+ in the cathode than in the anode (Fig.3). The E-I curves were measured by multi-current-step method: the current density was increased by about 50 mA cm-2 each step, and each step lasted 5 min, as shown in Fig.6b. Interestingly, it was hard to reach the steady state after each step. The lag of voltage behind current was likely due to the outflow of the Fe3+ from the contaminated CCM during electrolysis. This phenomenon may be used for judging whether the SPE water electrolyser performance degradation is due to the metal ions contamination. In order to recover the electrolyser which had been contaminated by 0.005 M Fe2(SO4)3 for 320 min, 0.5 M H2SO4 solution was fed into the electrolyser for 13 h. The iron element distribution on the cross-section of the recovered CCM had been shown in Fig.3. It can be seen that the amount of Fe in the recovered CCM (No2) was significantly lower than that in the contaminated CCM (No1), demonstrating that 0.5 M H2SO4 treatment can effectively remove Fex+ from the CCM. However, it is impossible to remove Fex+ from the CCM completely for the equilibrium between H+ and Fex+. The E-I curves of the fresh electrolyser and the recovered electrolyser were 12
shown in Fig.7. The electrolysis performance obviously recovered after the H2SO4 treatment: the current density at 1.8 V recovered to 1850 mA cm-2, a 93% recovery relative to the 2000 mA cm-2 for the fresh electrolyser.
4. Conclusion The presence of Fe3+ impurity in the anode feed water could seriously degrade the SPE water electrolyser performance. However, the cathode possessed higher tolerance for the Fe3+ contamination than the anode, which may be due to the much faster rate of proton reduction reaction on the Pt catalyst surface than that of the water oxidation reaction on the Ir and IrO2 catalysts. The EIS test results revealed that both the ohmic resistance and charge transfer resistance increased after contamination, and further calculation revealed that the increased ohmic overpotential accounted for most of the increased electrolysis voltage. The EPMA and UV analysis results demonstrated that Fe3+ migrated from the anode to the cathode and mostly were reduced to Fe2+ in the cathode rather than occurred underpotential deposition as described in the previous report. Due to the outflow of the Fe3+ from the contaminated CCM during electrolysis when the deionized water was fed to the electrolyser, the voltage lagged behind the current for a few minutes in the multi-current-step test, which phenomenon may be used for judging whether the SPE water electrolyser performance degradation is due to the metal ions contamination. Finally, the Fex+ in the CCM was partly removed by 0.5 M H2SO4 solution treatment for 13 h, and the contaminated electrolyser was mostly recovered. 13
Acknowledgement We thank the National High Technology Research and Development Program of China (863 Program, 2013AA110201) and the National Natural Science Foundations of China (Nos. 21203191 and 21306190) for the financial support.
References [1] T.R. Cook, D.K. Dogutan, S.Y. Reece, Y. Surendranath, T.S. Teets, D.G. Nocera, Solar Energy Supply and Storage for the Legacy and Non legacy Worlds, Chem Rev, 110 (2010) 6474. [2] M. Carmo, D.L. Fritz, J. Merge, D. Stolten, A comprehensive review on PEM water electrolysis, Int J Hydrogen Energy, 38 (2013) 4901. [3] J. Mergel, M. Carmo, D. Fritz, Status on technologies for hydrogen production by water electrolysis, in: D. Stolten, V. Scherer (Eds.) Transition to renewable energy systems, Wiley-VCH, Weinheim, 2013, pp. 423. [4] A. Ursua, L.M. Gandia, P. Sanchis, Hydrogen Production From Water Electrolysis: Current Status and Future Trends, P IEEE, 100 (2012) 410. [5] B. Bensmann, R. Hanke-Rauschenbach, I.K.P. Arias, K. Sundmacher, Energetic evaluation of high pressure PEM electrolyzer systems for intermediate storage of renewable energies, Electrochim Acta, 110 (2013) 570. [6] H. Dau, C. Limberg, T. Reier, M. Risch, S. Roggan, P. Strasser, The Mechanism of Water Oxidation: From Electrolysis via Homogeneous to Biological Catalysis, Chemcatchem, 2 (2010) 724. 14
[7] J. Xu, M. Wang, G. Liu, J. Li, X. Wang, The physical-chemical properties and electrocatalytic performance of iridium oxide in oxygen evolution, Electrochim Acta, 56 (2011) 10223. [8] J. Cheng, H. Zhang, H. Ma, H. Zhong, Y. Zou, Study of carbon-supported IrO2 and RuO2 for use in the hydrogen evolution reaction in a solid polymer electrolyte electrolyzer, Electrochim Acta, 55 (2010) 1855. [9] P. Millet, F. Andolfatto, R. Durand, Design and performance of a solid polymer electrolyte water electrolyzer, Int J Hydrogen Energy, 21 (1996) 87. [10] G. Wei, Y. Wang, C. Huang, Q. Gao, Z. Wang, L. Xu, The stability of MEA in SPE water electrolysis for hydrogen production, Int J Hydrogen Energy, 35 (2010) 3951. [11] X. Cheng, Z. Shi, N. Glass, L. Zhang, J. Zhang, D. Song, Z.-S. Liu, H. Wang, J. Shen, A review of PEM hydrogen fuel cell contamination: Impacts, mechanisms, and mitigation, J Power Sources, 165 (2007) 739. [12] S. Sun, Z. Shao, H. Yu, G. Li, B. Yi, Investigations on degradation of the long-term proton exchange membrane water electrolysis stack, J Power Sources, 267 (2014) 515. [13] A. Pozio, R.F. Silva, M. De Francesco, L. Giorgi, Nafion degradation in PEFCs from end plate iron contamination, Electrochim Acta, 48 (2003) 1543. [14] M. Saito, K. Hayamizu, T. Okada, Temperature dependence of ion and water transport in perfluorinated ionomer membranes for fuel cells, J Phys Chem B, 109 (2005) 3112. 15
[15] K. Hongsirikarn, J.G. Goodwin Jr, S. Greenway, S. Creager, Effect of cations (Na+, Ca2+, Fe3+) on the conductivity of a Nafion membrane, J Power Sources, 195 (2010) 7213. [16] M.-J. Yang, K.-Y. Park, K.-B. Kim, H. Cho, H. Choi, J.-Y. Park, Degradation mechanisms and mitigation strategies of metal cations in recycled fuel for direct methanol fuel cell membrane electrode assembly, J Power Sources, 242 (2013) 646. [17] T. Okada, G. Xie, O. Gorseth, S. Kjelstrup, N. Nakamura, T. Arimura, Ion and water transport characteristics of Nafion membranes as electrolytes, Electrochim Acta, 43 (1998) 3741. [18] M.J. Kelly, G. Fafilek, J.O. Besenhard, H. Kronberger, G.E. Nauer, Contaminant absorption and conductivity in polymer electrolyte membranes, J Power Sources, 145 (2005) 249. [19] G. Suresh, A.K. Pandey, A. Goswami, Self-diffusion coefficients of water in Nafion-117 membrane with multivalent counterions, J Membrane Sci, 284 (2006) 193. [20] E.R. Kotz, S. Stucki, Ruthenium dioxide as a hydrogen-evolution cathode, J Appl Electrochem, 17 (1987) 1190. [21] L. Zhang, X. Jie, Z.-G. Shao, Z.-M. Zhou, G. Xiao, B. Yi, The influence of sodium ion on the solid polymer electrolyte water electrolysis, Int J Hydrogen Energy, 37 (2012) 1321. [22] L. Zhang, X. Jie, Z.-G. Shao, X. Wang, B. Yi, The dynamic-state effects of sodium ion contamination on the solid polymer electrolyte water electrolysis, J Power 16
Sources, 241 (2013) 341. [23] F.J.R. Varela, O. Savadogo, The effect of anode catalysts on the behavior of low temperature direct propane polymer electrolyte fuel cells (DPFC), J New Mat Electr Sys, 9 (2006) 127. [24] X. Jie, Z.-G. Shao, J. Hou, G. Sun, B. Yi, The influence of sodium ion as a potential fuel impurity on the direct methanol fuel cells, Electrochim Acta, 55 (2010) 4783. [25] G.F. Li, H.M. Yu, W. Song, M.L. Dou, Y.K. Li, Z.G. Shao, B.L. Yi, Preparation of IrO2 using hard-template method and its performance in a solid polymer electrolyte water electrolyser, ChemSusChem, 5 (2012) 858. [26] W.B. Fortune, M.G. Mellon, Determination of iron with o-phenanthroline: a spectrophotometric study, Ind. Eng. Chem. Anal. Edit. , 10 (1938) 60 [27] M. Doyle, M.E. Lewittes, M.G. Roelofs, S.A. Perusich, R.E. Lowrey, Relationship between ionic conductivity of perfluorinated ionomeric membranes and nonaqueous solvent properties, J Membrane Sci, 184 (2001) 257. [28] T. Okada, Y. Ayato, M. Yuasa, I. Sekine, The effect of impurity cations on the transport characteristics of perfluorosulfonated ionomer membranes, J Phys Chem B, 103 (1999) 3315. [29] T. Okada, H. Satou, M. Okuno, M. Yuasa, Ion and water transport characteristics of perfluorosulfonated ionomer membranes with H+ and alkali metal cations, J Phys Chem B, 106 (2002) 1267. [30] Product Information, DuPont Nafion PFSA Membranes, Product Information, 17
DuPont Nafion PFSA Membranes, (2009).
Fig.1 Schematic illustration for the single cell test stand of water electrolysis.
Fig.2 The effect of 0.01 mol L-1 Fe3+ on the (a) electrolyser voltage (E), anode voltage (Ea), cathode voltage (Ec), (b) PH of the anode water (PHa) and cathode water (PHc), at 80 oC, 500 mA cm-2 with flow rate of 10 ml min-1.
18
Fig.3 EPMA images about the Fe distribution at the cross-sections of the contaminated CCM (No.1) and recovered CCM (No.2).
19
Fig.4 The weight percentages of Fex+ in the cathode water in different contamination time.
Fig.5 The EIS patterns of the fresh and contaminated SPE water electrolyser measured at 1.45 V vs. SHE; the inset shows the enlarged EIS patterns of the fresh SPE water electrolyser, point – experimental data, line – fit data.
20
Fig.6 a) The E-I curves of the fresh (solid points) and contaminated (hollow points) SPE water electrolyser; b) the multi-current-step curves and the obtained E-t curves of the contaminated SPE water electrolyser at 80 oC. E - electrolyser voltage, Ea - anode voltage, Ec - cathode voltage
Fig.7 The E-I curves of the fresh electrolyser (solid points) and the recovered electrolyser (hollow points). E - electrolyser voltage, Ea - anode voltage, Ec - cathode voltage
21
S0013-4686(15)00186-3 http://dx.doi.org/doi:10.1016/j.electacta.2015.01.140 EA 24192
To appear in:
Electrochimica Acta
Received date: Revised date: Accepted date:
22-12-2014 24-1-2015 24-1-2015
Please cite this article as: Xunying Wang, Linsong Zhang, Guangfu Li, Geng Zhang, Zhi-Gang Shao, Baolian Yi, The influence of Ferric ion contamination on the solid polymer electrolyte water electrolysis performance, Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2015.01.140 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The influence of Ferric ion contamination on the solid polymer electrolyte water electrolysis performance Xunying Wang a,b, Linsong Zhang a,c, Guangfu Li a,b, Geng Zhang a,b, Zhi-Gang Shaoa, *, Baolian Yia
a
Fuel Cell System and Engineering Laboratory, Dalian Institute of Chemical Physics, Chinese Academy of
Sciences, 457 Zhongshan Road, Dalian 116023, PR China b
Graduate School of the Chinese Academy of Sciences, Beijing 100039, PR China c
Xingtai Polytechnic College, Xingtai 054035, PR China
* Corresponding Author: Telephone: +86 411 84379153; Fax: +86 411 84379185; E-mail: [email protected]
Highlights
The cathode possesses higher tolerance for the Fe3+ contamination than the anode.
Fe3+ are mostly reduced to Fe2+ rather than occur underpotential deposition.
Increased electrolysis voltage was mainly attributed to ohmic overpotential.
Voltage lags behind current for minutes in the multi-current-step test.
Poisoned electrolyser is mostly recovered by 0.5M H2SO4 solution treatment for 13h
1
Abstract: Fe3+ is a sort of common metal ion contaminant for the solid polymer electrolyte (SPE) water electrolyser. In this paper, the effect of Fe3+ on the performance of SPE water electrolyser has been investigated by both in-situ and ex-situ characterizations. The electron probe microanalysis and ultraviolet test results showed that Fe3+ could migrate from the anode to the cathode and mostly be reduced to Fe2+ in the cathode rather than occurred underpotential deposition as described in the previous report. The in-situ dynamic contamination test showed that the anode voltage increased sharply as soon as the Fe3+ was fed into the anode, while the cathode voltage kept constant until the contamination time was over 30 minutes, indicating the higher tolerance of the cathode than the anode for the Fe3+ contamination. The calculation results based on the electrochemistry impedance spectroscopy test results revealed that the striking increase of the electrolysis voltage was mainly attributed to the ohmic overpotential, which was due to the replacement of H+ by Fe3+ in the Nafion resin. Interestingly, the voltage lagged behind the current for several minutes in the multi-current-step test for the contaminated electrolyser, which phenomenon may be used for judging whether the SPE water electrolyser performance degradation is due to the metal ions contamination. Furthermore, recovery strategy has been developed, and it is found that the contaminated electrolyser could be mostly recovered by 0.5 M H2SO4 solution treatment for 13 h.
2
Key words: Solid polymer electrolyte water electrolysis; Ferric ion impurity; In-situ; Valence alternation
1. Introduction Solid polymer electrolyte (SPE) water electrolysis provides a promising method to produce hydrogen from renewable but intermittent energy sources [1-5]. Compared with the conventional alkaline water electrolyser, SPE water electrolyser possesses the advantages of higher safety, simplicity, greater energy efficiency, higher specific production capacity and lower maintenance [2-4, 6-8]. However, during water electrolysis process, the membrane electrode assembly is susceptible to cation impurities, including Na+, Ca2+, Cu2+, Ni2+ and Fe3+, originating from the fabrication of the MEA, the impurity of feedwater, the corrosion of water pipes and stack component materials, and other sources [9-13]. The cations may affect the electrolysis performance by exchanging with protons in the Nafion electrolyte and depositing at the cathode. The replacement of protons by metal ions significantly impacts the ionic conductivity of the Nafion electrolyte [14-17]. This is because that most of the metal ions possess higher affinities than protons for the sulfonic acid groups in Nafion, and that the mobility of the metal ions in Nafion electrolyte is quite lower than that of protons [18]. Moreover, the replacement of protons by metal ions impacts water transportation in Nafion electrolyte, for the much faster diffusion of H3O+ than that of the complex metal ion - water clusters [M-(H2O)m]n+ [14, 19]. 3
The exchanged metal ions, driven by the cell voltage and concentration difference, will transfer across the Nafion membrane to the cathode. Some metal ions, such as Cu2+, have positive reversible potential compared with the standard hydrogen electrode (SHE), thus bulk deposition of these metal ions can occur at the cathode of the water electrolyser. Some metal ions, e.g. Sn2+, possess negative reversible potential versus SHE, can occur underpotential deposition of monolayer on the surface of the catalyst [20]. Some other metal cations, such as Ca2+, have too negative Nernst potential and can not deposit on the cathode catalyst. But they could precipitate as hydroxides at the interface between the Nafion membrane and the cathode catalyst layer, thus decrease the conductivity of the MEA. The metal ions contamination is a slowly accumulative process during SPE water electrolysis. Therefore, it is necessary to take detail study on the dynamic process of the metal ions contamination to the electrolyser. However, there are limited reports about this study [10, 20-22]. Kötz et al.[20] discovered that RuO2 cathodes were insensitive to metal ions contamination for the underpotential deposition phenomenon did not occur on RuO2. Wei [10] et al. provided evidence for the accumulation of metal ions in the feed-water during electrolysis. The influence of Na+ on the SPE water electrolysis was studied in our previous work[21, 22]. It was found that 0.05 M Na+ caused serious performance degradation to the electrolyser, which mainly resulted from the cathodic overpotential increase for the reaction of 2H2O + 2e- = H2 +2OH- .
4
Fe3+ is a sort of common metal ion contaminant for the SPE water electrolyser. It can originate from the impurities of feed-water, the corrosion of water pipes and stack component materials. Understanding how and to what extent Fe3+ affects the SPE water electrolyser can help to choose proper feed-water purity standards and component materials with acceptable corrosion rates, develop analytic methods and recovery statics for the poisoned electrolyser. Herein, the effect of Fe3+ on the SPE water electrolyser performance was investigated by both in-situ contamination test and ex-situ characterization, the recovery strategy was also developed.
2. Experimental
2.1 Preparation of MEA The CCM was prepared by spraying 5 wt. % of Nafion solution (0.6 mg cm-2) and isopropanol onto each side of Nafion membrane (Dupont 115), followed by spraying catalyst layers. Iridium black (Johnson Matthey) and Pt/C (70 wt. %, Johnson Matthey) were used as the anode and cathode catalyst, respectively. Homogeneous ink consisting of catalyst (Iridium black or Pt/C), Nafion solution (5 wt. %, Du Pont) and isopropanol was sprayed onto the Nafion membrane with Nafion layers to form the catalyst layers. The catalyst loading of the CCM was about 2.0 mg cm−2 for iridium black and 1.0 mg cm−2 for Pt/C. Porous titanium (Pt plated, 0.7 mm in thickness) and wet-proof carbon paper (Toray, TGP-H-60) served as the anode and cathode diffusion layer, respectively. The MEA was fabricated by hot pressing the 5
CCM and the carbon paper together.
2.2 Single cell test and ferric contamination A home-made electrolysis test stand was employed to evaluate the performance of the CCM. The SPE water electrolyser performance was evaluated at atmosphere pressure and 80 oC. After the electrolyser operating galvanostatically at 500 mA cm−2 for 30 min with deionized water fed to the anode side at the flow rate of 10 mL min-1, 500 mL solution containing 0.005 mol L-1 Fe2(SO4)3 and 0.01 mol L-1 H2SO4 was continuously delivered to the anode side. The add of sulphuric acid was to avoid the hydrolysis of Fe3+. The water from the cathode was exported into a water tank containing 500 mL deionized water. The pH changes of the anode feed water and cathode water were measured by immersing the basic pH meter PB-10 (Sartorius AG) into the water tank. The single electrode potentials were measured against a saturation mercury electrode (SCE, E = 0.242 V vs. SHE) and finally calibrated against SHE. As schematically shown in Fig.1, the SCE was placed in 0.5 mol L-1 H2SO4 solution, then an ionic contact can be formed by immersing the piece of Nafion membrane extending from the MEA in 0.5 mol L-1 H2SO4 solution [21-24]. Electrochemistry impedance spectroscopy (EIS) was carried out at 1.45 V vs. SHE by Solartron 1287 Electrochemical Interface in conjunction with Solartron 1260 Frequency Response Analyzer [25].
6
2.2 Physical characterizations The iron element distribution on the cross-section of the contaminated CCM was investigated by electron probe microanalysis (EPMA, SHIMAZU EPMA-1600). The changes in weight percentage of iron element in the cathode water were determined by inductively coupled plasma (ICP, Optima, 2000DV) analysis. The valence of iron element in the cathode water was determined by phenanthroline spectrophotometry method [26] on a UV-VIS spectrolphotometer (SHIMADZU, UV-2550). All the chemicals used were of analytical grade. The test solution for determining the total iron content in cathode water was prepared by putting 10 mL cathode water into a 100 mL volumetric flask, then sequentially adding 50 mL deionized water, HCl solution to adjust the PH to 2, 2.5 mL ascorbic acid (20 g L-1) solution to reduce Fe3+ to Fe2+, 10 mL acetic acid-sodium acetate buffer solution (PH=5), 5 mL phenanthroline solution (2 g L-1), and finally diluting to the mark with deionized water. The test solution for determining the Fe2+ content in cathode water was prepared with the above procedure except that ascorbic acid solution was not used. The absorbance was measured at 510 nm and the Fe2+ concentration was read from the standard concentration curve. The Fe3+ concentration could be determined by difference.
3. Results and discussion Fig.2 showed the effects of 0.01 M Fe3+ on the electrolysis performance. As shown in Fig.2a, the anode voltage severely increased initially until reaching a maximum, 7
then slowly dropped and finally became steady. According to the previous reports [15, 27, 28], the mobility of Fe3+ in the Nafion membrane was between 2.6*10-9 – 7.1*10-9 m2 V-1 s-1, so it only needs no more than 5 s for Fe3+ to transfer from anode surface to cathode when the thickness of Nafion 115 membrane and anode catalyst layer is about 125 μm and 10 μm, respectively. What’s more, the ICP analysis (Fig.4) also showed that Fe element appeared in the cathode water when the electrolyser was contaminated for 5 min, which further verified that the Fe3+ reached the cathode catalyst layer. However, the cathode voltage was not influenced until when the contamination time was over 30 minutes, indicating that the cathode possessed higher tolerance for the Fe3+ contamination than the anode. It may be due to the much faster rate of proton reduction reaction on the Pt catalyst surface than that of the water oxidation reaction on the Ir and IrO2 catalysts. During the contamination, the PH changes of the feed water and the cathode water were also measured (Fig.2b). It can be seen that both the anode and the cathode PH decreased with contamination time. The initial sharp decrease of anode PH was due to the introduction of solution containing 0.005 M Fe2(SO4)3 and 0.01 M H2SO4. The EPMA test result at the cross-section of the contaminated CCM was shown in Fig.3. The distribution of Fe in the anode, membrane and cathode can be observed, demonstrating that Fe3+ migrated from the anode to the cathode during the contamination. There was lower content of Nafion in the catalyst layers than in the membrane, and Fe3+ mainly attached on the sulfonic acid groups in Nafion electrolyte, so the amount of Fe in the catalyst layers was smaller than that in the membrane. The 8
larger amount of Fe in the cathode than in the anode was due to the voltage difference between the anode and the cathode. Fig. 4 showed the change in weight percentage of Fex+ in the cathode water. It can be seen that the weight percentage of Fex+ increased with contamination time, indicating that Fex+ kept flowing out of the electrolyser, rather than being retained by the CCM. Furthermore, phenanthroline spectrophotometry was used to study the valence of Fex+ in the cathode water. Test result showed that 92% of the Fex+ in the cathode water was Fe2+, suggesting that most of the Fe3+ transferred to the cathode was reduced to Fe2+ rather than occurred underpotential deposition as described in the previous report [20]. The higher affinity of Fex+ than H+ to the sulfonic sites led to the replacement of H+ in Nafion electrolyte by Fex+ [18], which would ① reduce the effective triple-phase boundaries and thus increase the activation overpotential, ②
lower proton
concentration in the electrolyte and the proton conductivity of the electrolyte, therefore enhance the concentration overpotential. With the increase of the electrolysis voltage, the speed of Fe3+ transferring out of the anode enhanced, and the Fe3+ content in the anode catalyst layer decreased, thus lead to the decrease of the anode voltage instead (Fig.2a) [22]. In the anode, the protons produced by the water oxidation reaction partially entered the feed water due to the lack of the sulfonic sites, which decreased the PH of the feed water (Fig.2b). Whereas the decrease of the cathode water PH value was mainly caused by the hydrolysis of Fex+. We used EIS to further characterize the fresh and contaminated electrolyser, 9
respectively (Fig.5). The high frequency intercept with the real axis and the diameter of the medium frequency arc are the measures of the ohmic resistance (RΩ) and charge-transfer resistance (Rct), respectively. The EIS of the contaminated electrolyser exhibited a Warburg line at the low-frequency, indicating the diffusion resistance of the chemical species. The equivalent circuits to simulate the EIS data of the fresh and contaminated
electrolyser
were
presented
by
LRΩ(R1Q1)(RctQdl)
and
LRΩ(R1Q1)((RctZw)Qdl), respectively, where RΩ, Rct, Q, Zw and L are the ohmic resistance, charge-transfer resistance, capacitive element, Warburg element and inductor, respectively. The capacitive element is often presented by a constant phase element Q to model depressed semi-circle due to heterogeneities and rough surfaces. The Qdl represents double layer capacitance. The R1Q1 circuit is used to describe various phenomena, e.g. the diffusion of protons along the catalysts. The simulating results suggested that the RΩ values for the fresh and contaminated electrolyser were 0.235 and 1.34 Ω cm2, and the Rct values were 0.126 and 3.81 Ω cm2, respectively. It can be seen that both RΩ and Rct increased significantly after contamination. The increased ohmic overpotential induced by the Fe3+ contamination was about 548 mV at 500 mA cm-2. It accounted for 80% of the increased cell voltage (685 mV) when the electrolyser was contaminated for 320 min (Fig.2a), indicating that the ohmic overpotential was the primary reason for the cell performance loss. The substitution for the protons by Fe3+ decreased the conductivity of the Nafion electrolyte, and the relationship between the Fe3+ content (
) and the conductivity
(к, S m-1) of the membrane can be expressed by the following equations [24, 29]: 10
(1) (2) where F is Faraday constant, membrane,
and
is the concentration of the sulfonic sites in the
are the mobility of proton and Fe3+, respectively,
and
are the fractions of the sulfonic sites occupied by protons and Fe3+ in the Nafion membrane, respectively. For the water soaked Nafion 115 membrane, the equals 1.18*103 mol m-3, and its
is 8.78*10-8 m2 V-1 s-1 [30]. Previously reported
mobility of the Fe3+-form Nafion membrane was between 0.296 - 0.803 S m-1 [15, 27, 28], and the average value, 0.550 S m-1, is used here. Compared with the membrane resistance, the ohmic resistance caused by the other components can be ignored. So the conductivity of the contaminated Nafion membrane, estimated from the above EIS data, was about 1.09 S m-1. Accordingly, the Fe3+ composition in the membrane (
)
was about 94.3%. Subsequently, the decrease of the reversible electrode potential, , obtained according to the Nernst equation (Eq.3), was 87 mV when 94.3% of the protons in the electrolyte were exchanged, which accounted for about 13% of the increased cell voltage (685 mV). (3)
where
,
, R, T, F,
, and [H+] are the thermodynamic electrode potential,
standard cathode potential, gas constant, cell temperature, Faraday constant, partial pressure (atm) of H2 and molar concentration of protons, respectively. 11
Fig.6a showed the E-I curves of the SPE water electrolyser before and after contamination. Deionized water was used for this electrolysis test. The expected increase of electrolysis voltage was observed after contamination. Furthermore, the increase of anode potential was much smaller than that of cathode potential, indicating that Fe3+ affected the cathode more seriously than the anode, although the cathode had higher tolerance for the Fe3+ than the anode as shown in Fig.1. It was mainly due to the larger amount of Fex+ in the cathode than in the anode (Fig.3). The E-I curves were measured by multi-current-step method: the current density was increased by about 50 mA cm-2 each step, and each step lasted 5 min, as shown in Fig.6b. Interestingly, it was hard to reach the steady state after each step. The lag of voltage behind current was likely due to the outflow of the Fe3+ from the contaminated CCM during electrolysis. This phenomenon may be used for judging whether the SPE water electrolyser performance degradation is due to the metal ions contamination. In order to recover the electrolyser which had been contaminated by 0.005 M Fe2(SO4)3 for 320 min, 0.5 M H2SO4 solution was fed into the electrolyser for 13 h. The iron element distribution on the cross-section of the recovered CCM had been shown in Fig.3. It can be seen that the amount of Fe in the recovered CCM (No2) was significantly lower than that in the contaminated CCM (No1), demonstrating that 0.5 M H2SO4 treatment can effectively remove Fex+ from the CCM. However, it is impossible to remove Fex+ from the CCM completely for the equilibrium between H+ and Fex+. The E-I curves of the fresh electrolyser and the recovered electrolyser were 12
shown in Fig.7. The electrolysis performance obviously recovered after the H2SO4 treatment: the current density at 1.8 V recovered to 1850 mA cm-2, a 93% recovery relative to the 2000 mA cm-2 for the fresh electrolyser.
4. Conclusion The presence of Fe3+ impurity in the anode feed water could seriously degrade the SPE water electrolyser performance. However, the cathode possessed higher tolerance for the Fe3+ contamination than the anode, which may be due to the much faster rate of proton reduction reaction on the Pt catalyst surface than that of the water oxidation reaction on the Ir and IrO2 catalysts. The EIS test results revealed that both the ohmic resistance and charge transfer resistance increased after contamination, and further calculation revealed that the increased ohmic overpotential accounted for most of the increased electrolysis voltage. The EPMA and UV analysis results demonstrated that Fe3+ migrated from the anode to the cathode and mostly were reduced to Fe2+ in the cathode rather than occurred underpotential deposition as described in the previous report. Due to the outflow of the Fe3+ from the contaminated CCM during electrolysis when the deionized water was fed to the electrolyser, the voltage lagged behind the current for a few minutes in the multi-current-step test, which phenomenon may be used for judging whether the SPE water electrolyser performance degradation is due to the metal ions contamination. Finally, the Fex+ in the CCM was partly removed by 0.5 M H2SO4 solution treatment for 13 h, and the contaminated electrolyser was mostly recovered. 13
Acknowledgement We thank the National High Technology Research and Development Program of China (863 Program, 2013AA110201) and the National Natural Science Foundations of China (Nos. 21203191 and 21306190) for the financial support.
References [1] T.R. Cook, D.K. Dogutan, S.Y. Reece, Y. Surendranath, T.S. Teets, D.G. Nocera, Solar Energy Supply and Storage for the Legacy and Non legacy Worlds, Chem Rev, 110 (2010) 6474. [2] M. Carmo, D.L. Fritz, J. Merge, D. Stolten, A comprehensive review on PEM water electrolysis, Int J Hydrogen Energy, 38 (2013) 4901. [3] J. Mergel, M. Carmo, D. Fritz, Status on technologies for hydrogen production by water electrolysis, in: D. Stolten, V. Scherer (Eds.) Transition to renewable energy systems, Wiley-VCH, Weinheim, 2013, pp. 423. [4] A. Ursua, L.M. Gandia, P. Sanchis, Hydrogen Production From Water Electrolysis: Current Status and Future Trends, P IEEE, 100 (2012) 410. [5] B. Bensmann, R. Hanke-Rauschenbach, I.K.P. Arias, K. Sundmacher, Energetic evaluation of high pressure PEM electrolyzer systems for intermediate storage of renewable energies, Electrochim Acta, 110 (2013) 570. [6] H. Dau, C. Limberg, T. Reier, M. Risch, S. Roggan, P. Strasser, The Mechanism of Water Oxidation: From Electrolysis via Homogeneous to Biological Catalysis, Chemcatchem, 2 (2010) 724. 14
[7] J. Xu, M. Wang, G. Liu, J. Li, X. Wang, The physical-chemical properties and electrocatalytic performance of iridium oxide in oxygen evolution, Electrochim Acta, 56 (2011) 10223. [8] J. Cheng, H. Zhang, H. Ma, H. Zhong, Y. Zou, Study of carbon-supported IrO2 and RuO2 for use in the hydrogen evolution reaction in a solid polymer electrolyte electrolyzer, Electrochim Acta, 55 (2010) 1855. [9] P. Millet, F. Andolfatto, R. Durand, Design and performance of a solid polymer electrolyte water electrolyzer, Int J Hydrogen Energy, 21 (1996) 87. [10] G. Wei, Y. Wang, C. Huang, Q. Gao, Z. Wang, L. Xu, The stability of MEA in SPE water electrolysis for hydrogen production, Int J Hydrogen Energy, 35 (2010) 3951. [11] X. Cheng, Z. Shi, N. Glass, L. Zhang, J. Zhang, D. Song, Z.-S. Liu, H. Wang, J. Shen, A review of PEM hydrogen fuel cell contamination: Impacts, mechanisms, and mitigation, J Power Sources, 165 (2007) 739. [12] S. Sun, Z. Shao, H. Yu, G. Li, B. Yi, Investigations on degradation of the long-term proton exchange membrane water electrolysis stack, J Power Sources, 267 (2014) 515. [13] A. Pozio, R.F. Silva, M. De Francesco, L. Giorgi, Nafion degradation in PEFCs from end plate iron contamination, Electrochim Acta, 48 (2003) 1543. [14] M. Saito, K. Hayamizu, T. Okada, Temperature dependence of ion and water transport in perfluorinated ionomer membranes for fuel cells, J Phys Chem B, 109 (2005) 3112. 15
[15] K. Hongsirikarn, J.G. Goodwin Jr, S. Greenway, S. Creager, Effect of cations (Na+, Ca2+, Fe3+) on the conductivity of a Nafion membrane, J Power Sources, 195 (2010) 7213. [16] M.-J. Yang, K.-Y. Park, K.-B. Kim, H. Cho, H. Choi, J.-Y. Park, Degradation mechanisms and mitigation strategies of metal cations in recycled fuel for direct methanol fuel cell membrane electrode assembly, J Power Sources, 242 (2013) 646. [17] T. Okada, G. Xie, O. Gorseth, S. Kjelstrup, N. Nakamura, T. Arimura, Ion and water transport characteristics of Nafion membranes as electrolytes, Electrochim Acta, 43 (1998) 3741. [18] M.J. Kelly, G. Fafilek, J.O. Besenhard, H. Kronberger, G.E. Nauer, Contaminant absorption and conductivity in polymer electrolyte membranes, J Power Sources, 145 (2005) 249. [19] G. Suresh, A.K. Pandey, A. Goswami, Self-diffusion coefficients of water in Nafion-117 membrane with multivalent counterions, J Membrane Sci, 284 (2006) 193. [20] E.R. Kotz, S. Stucki, Ruthenium dioxide as a hydrogen-evolution cathode, J Appl Electrochem, 17 (1987) 1190. [21] L. Zhang, X. Jie, Z.-G. Shao, Z.-M. Zhou, G. Xiao, B. Yi, The influence of sodium ion on the solid polymer electrolyte water electrolysis, Int J Hydrogen Energy, 37 (2012) 1321. [22] L. Zhang, X. Jie, Z.-G. Shao, X. Wang, B. Yi, The dynamic-state effects of sodium ion contamination on the solid polymer electrolyte water electrolysis, J Power 16
Sources, 241 (2013) 341. [23] F.J.R. Varela, O. Savadogo, The effect of anode catalysts on the behavior of low temperature direct propane polymer electrolyte fuel cells (DPFC), J New Mat Electr Sys, 9 (2006) 127. [24] X. Jie, Z.-G. Shao, J. Hou, G. Sun, B. Yi, The influence of sodium ion as a potential fuel impurity on the direct methanol fuel cells, Electrochim Acta, 55 (2010) 4783. [25] G.F. Li, H.M. Yu, W. Song, M.L. Dou, Y.K. Li, Z.G. Shao, B.L. Yi, Preparation of IrO2 using hard-template method and its performance in a solid polymer electrolyte water electrolyser, ChemSusChem, 5 (2012) 858. [26] W.B. Fortune, M.G. Mellon, Determination of iron with o-phenanthroline: a spectrophotometric study, Ind. Eng. Chem. Anal. Edit. , 10 (1938) 60 [27] M. Doyle, M.E. Lewittes, M.G. Roelofs, S.A. Perusich, R.E. Lowrey, Relationship between ionic conductivity of perfluorinated ionomeric membranes and nonaqueous solvent properties, J Membrane Sci, 184 (2001) 257. [28] T. Okada, Y. Ayato, M. Yuasa, I. Sekine, The effect of impurity cations on the transport characteristics of perfluorosulfonated ionomer membranes, J Phys Chem B, 103 (1999) 3315. [29] T. Okada, H. Satou, M. Okuno, M. Yuasa, Ion and water transport characteristics of perfluorosulfonated ionomer membranes with H+ and alkali metal cations, J Phys Chem B, 106 (2002) 1267. [30] Product Information, DuPont Nafion PFSA Membranes, Product Information, 17
DuPont Nafion PFSA Membranes, (2009).
Fig.1 Schematic illustration for the single cell test stand of water electrolysis.
Fig.2 The effect of 0.01 mol L-1 Fe3+ on the (a) electrolyser voltage (E), anode voltage (Ea), cathode voltage (Ec), (b) PH of the anode water (PHa) and cathode water (PHc), at 80 oC, 500 mA cm-2 with flow rate of 10 ml min-1.
18
Fig.3 EPMA images about the Fe distribution at the cross-sections of the contaminated CCM (No.1) and recovered CCM (No.2).
19
Fig.4 The weight percentages of Fex+ in the cathode water in different contamination time.
Fig.5 The EIS patterns of the fresh and contaminated SPE water electrolyser measured at 1.45 V vs. SHE; the inset shows the enlarged EIS patterns of the fresh SPE water electrolyser, point – experimental data, line – fit data.
20
Fig.6 a) The E-I curves of the fresh (solid points) and contaminated (hollow points) SPE water electrolyser; b) the multi-current-step curves and the obtained E-t curves of the contaminated SPE water electrolyser at 80 oC. E - electrolyser voltage, Ea - anode voltage, Ec - cathode voltage
Fig.7 The E-I curves of the fresh electrolyser (solid points) and the recovered electrolyser (hollow points). E - electrolyser voltage, Ea - anode voltage, Ec - cathode voltage
21