Investigation on the durability of direct dimethyl ether fuel cell. Part II: Cathode degradation

Investigation on the durability of direct dimethyl ether fuel cell. Part II: Cathode degradation

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 1 6 6 9 5 e1 6 6 9 9 Available online at www.sciencedirect.co...

761KB Sizes 0 Downloads 10 Views

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 1 6 6 9 5 e1 6 6 9 9

Available online at www.sciencedirect.com

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

Investigation on the durability of direct dimethyl ether fuel cell. Part II: Cathode degradation Le-Hong Xing a,b,*, Ge-Ping Yin a, Chun-Yu Du a, Shu-Xin Cui b, Ming-Hui Zuo b, Hai-Peng Wang b a

State Key Laboratory of Urban Water Resource and Environment, School of Chemical Engineering and Technology, Harbin Institute of Technology, No. 92, West Da-Zhi Street, Harbin 150001, China b College of Chemistry and Chemical Engineering, Mudanjiang Normal University, Mudanjiang 157012, China

article info

abstract

Article history:

Cathodic Pt/C catalyst degradation for a direct dimethyl ether fuel cell (DDFC) has been

Received 12 December 2016

investigated after a 70.5 h galvostatic operation at 60  C under ambient pressure. The

Received in revised form

cathode electrochemical active surface (EAS) reduces from 349 to 267 cm2 mg1 Pt after

23 April 2017

durability test. Cathode EAS loss is associated with the growth of Pt catalyst particles. XRD

Accepted 14 May 2017

and TEM results show that the particle size of cathodic Pt catalyst increases from an

Available online 17 June 2017

original value of 3.0 nm to 5.4 nm. Furthermore, the anode failure accelerates dimethyl ether (DME) crossover, and consequently, poisoning of catalyst by absorbed DME and/or

Keywords:

intermediates aggravates the degradation of cathodic Pt/C catalyst. The effect of anode

Direct dimethyl ether fuel cell

degradation on the long-term performance of DDFC has been reported in companion

Durability test

article (Part I).

Cathodic catalyst degradation

© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Electrochemical active surface Dimethyl ether crossover

Introduction Recently, proton exchange membrane fuel cells (PEMFCs) [1e3], and direct methanol fuel cells (DMFCs) [4e6] have been investigated extensively for stationary and portable applications [7,8]. However, each of them has some shortcomings which limit their commercialization [9,10]. Dimethyl ether (DME) is known as a promising alternative fuel [11e13]. The transportation of DME is easier and safer than hydrogen because its physical properties are similar with liquefied petroleum gas (LPG) [14,15]. DME also has many advantages in comparison with methanol for its high energy density, low toxicity, and low fuel crossover [16e18]. Therefore, DME is a

new proper fuel for fuel cells and has been receiving increasing attention recently [19e22]. Up to now, most of the research has been focused on the activity of anodic catalyst and the short-term performance of direct dimethyl ether fuel cell (DDFC) [23,24]. The long-term performance and durability of the membrane electrode assembly (MEA) for DDFC have not been investigated in detail. DDFC commercialization demands stable operation for at least thousands of hours, which is difficult to achieve [25]. The electrocatalyst is one of the major components of MEAs, and their stability plays an important role in the long-term operation of fuel cells [26,27]. Pt/C catalyst failure results from the corrosion of carbon support [28], the Ostwald ripening (dissolution and redeposition) of Pt particles [29], the coalescence

* Corresponding author. State Key Laboratory of Urban Water Resource and Environment, School of Chemical Engineering and Technology, Harbin Institute of Technology, No. 92, West Da-Zhi Street, Harbin 150001, China. E-mail address: [email protected] (L.-H. Xing). http://dx.doi.org/10.1016/j.ijhydene.2017.05.134 0360-3199/© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

16696

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 1 6 6 9 5 e1 6 6 9 9

due to the high surface energy of small Pt nanoparticles [30] and poisoning of catalysts by accumulated intermediates from DME electro-oxidation or impurities [31]. The cathodic catalyst in DDFC operates in a serious corrosive environment, such as high water content, low pH value, high temperature, high oxygen concentration and high potentials (0.6e1.2 V) [32]. The stability of catalyst is an important factor which can strongly impact the long-tern performance of DDFC. In previous paper (part I) [33], we have reported the durability of anode catalyst for 70.5 h operation. The 70.5 h operation is composed of a 40.5 h operation beginning, and then cyclic voltammetry (CV) scans of the anode and cathode respectively, and another 30 h operation. In this paper (part II), we focus on exploring the durability of cathodic Pt/C catalyst for DDFC at the same condition. Electrochemical and physical characterizations are performed before and after the durability test, and their results are compared. Furthermore, the effect of DME crossover behavior on the cathodic catalyst deterioration is analyzed. The purpose of this study is to provide a better insight into the stability and changes of Pt/C catalysts under DDFC operating conditions and to investigate whether and how the anode degradation affects the performance of cathodic Pt/C catalyst by in situ CV and other techniques.

Experimental MEA fabrication The home-made MEA was fabricated by the gas diffusion layer-based method [33]. The working area of the MEA was 5 cm2. The anode and cathode catalyst were both 40 wt.% Pt/C. The Pt loading of anode and cathode was 3.2 mg cm2 and 2.5 mg cm2, respectively. The Nafion content was 20 wt.% in catalyst layers. The anode diffusion layers were wet-proofed Toray carbon papers (18 wt.% PTFE) coated with 1 mg cm2 Vulcan XC-72 carbon blacks and 18 wt.% PTFE. The cathode diffusion layers were wet-proofed Toray carbon papers (30 wt.% PTFE) coated with 1 mg cm2 Vulcan XC-72 carbon blacks and 30 wt.% PTFE. Nafion 115 polymer membranes (DuPont) were used to fabricate MEAs. The pretreated Nafion membranes sandwiched between anode and cathode and then the assemblies were hot pressed under a specific load of 100 kg cm2 for 1.5 min at 135  C.

During the test, the cathode was protected by humidified N2 at a flow rate of 200 mL min1. Simultaneously, the anode was fed with humidified H2 at a flow rate of 200 mL min1 as a dynamic hydrogen electrode (DHE). The CV sweep was performed at a scan rate of 0.01 V s1 between 0.05 and 1.2 V (vs. the anode). The cathodic EAS at different conditions could be determined through Eq. (1), where [Pt] denoted Pt loading (mg cm2) on the electrode, QH represented the charge (mC cm2) for hydrogen desorption, and 0.21 was the required charge (mC cm2) to oxidize a monolayer of H2 on bright Pt [34]. The similar CV tests were also employed to detect and eliminate the crossover DME and/or intermediates on cathodic Pt surface after the 40.5 h operation of the DDFC. ESAH ¼

QH ½Pt  0:21

(1)

DME crossover rate The DME crossover rate was estimated by the limiting current density at the cathode. During the test, a 1.5 mol L1 DME solution was fed to the anode at 3 mL min1. The linear sweep voltammetry (LSV) was performed at the cathode at a scan rate of 1 mV s1 with a protection of pure N2 at 200 mL min1. The anode was regarded as a DHE, because only the reduction of Hþ to H2 took place at the anode. After the electrochemical tests, the used MEA was peeled carefully and cut into small pieces for further physical analysis [35,36].

Physical measurements X-ray diffraction (XRD) XRD analysis was conducted on the catalysts scraped from the fresh and aged cathode to characterize the changes of the Pt particles. The XRD patterns were recorded by a Japan D/maxrB X-ray diffractometer, using a Cu Ka radiation source at a tube current of 100 mA and a tube voltage of 40 kV, with a scan rate of 5 min1.

Transmission electron microscopy (TEM) TEM for the catalyst samples was taken by a Hitachi H-7650 transmission electron microscope. Before taking the electron micrographs, the samples were finely ground and ultrasonically dispersed in isopropyl alcohol, and a drop of the resultant dispersion was deposited and dried on a standard copper grid coated with a polymer film.

Electrochemical measurements Single fuel cell tests The electrochemical tests of MEA were carried out by Fuel Cell Testing System (Arbin Instrument Corp.) using the single cell (Electrochemistry Corp.). All of the electrochemical measurements were carried out at 60  C. The 1.5 mol L1 DME solution was fed to the anode side with a flow rate of 3 mL min1. Pure oxygen was supplied to the cathode side with a flow rate of 200 mL min1 under ambient pressure.

Cyclic voltammetry Cyclic voltammetry (CV) scan was measured by a CHI 604B Electrochemical Analyzer (Shanghai Chen-Hua Instruments Corp. China) to evaluate the degradation of cathodic EAS.

Results and discussion After the durability test, the maximum power density of DDFC drops dramatically from an initial value of 37 mW cm2 to 12 mW cm2 [33]. Cyclic voltammograms in Fig. 1 present the decay of cathodic EAS. The hydrogen desorption region on Pt catalyst decreases after the life test. The calculated EAS decreases from 349 to 267 cm2 mg1 Pt after 70.5 h operation. The EAS loss may result from the deterioration of Pt metal and the worse contact between the catalyst and the electrolyte. The CV results indicate that the cathode ‘triple-phase boundaries’ where the electrolyte, oxygen, and electrically connected Pt/C catalyst contact together in cathode, are decreasing after the

16697

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 1 6 6 9 5 e1 6 6 9 9

40 a

mean=3.1nm

10

30

5

Co u n t / %

Current density / mA· cm

-2

15

0 -5 -10

20 10

-15 Initial Aged

-20

0

-25 0.0

0.2

0.4

0.6

0.8

1.0

2

1.2

3 4 Pt nanoparticle size / nm

Cathode potential / V vs. DHE Fig. 1 e Cyclic voltammograms of Pt at the initial and aged cathode, at a scan rate of 0.01 V s¡1, in a potential range between 0.05 and 1.2 V vs. DHE.

b

Count / %

30

mean=5.3nm

20

10

0

3

4 5 6 7 Pt nanoparticle size / nm

8

Fig. 4 e Histograms of particle size distributions of cathodic Pt catalyst before and after durability test: (a) initial Pt; (b) aged Pt.

Fig. 2 e XRD patterns of cathodic catalysts before and after durability test. life tests. The aging of cathodic catalyst would be one of the important reasons for the performance decay of DDFC cathode. XRD patterns of cathodic Pt/C catalyst measured before and after the life test are shown in Fig. 2. The first diffraction

peaks for cathode Pt/C at 2q about 26 can be attributed to the hexagonal graphite structures (0 0 2) of the XC-72 carbon black support. The Pt has a face-centered cubic (fcc) structure showing the major diffraction peaks of Pt (1 1 1), Pt (2 0 0), Pt (2 2 0), Pt (3 1 1), Pt (2 2 2), and so on. The Pt (2 2 0) peak is chosen to calculate the mean particle size of fresh and aged Pt

Fig. 3 e TEM images of cathodic Pt/C catalyst before and after durability test: (a) initial Pt/C; (b) aged Pt/C.

Current density / mA· cm

-2

16698

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 1 6 6 9 5 e1 6 6 9 9

15

adsorbed species

10 5 0 -5 -10

Cycle 1 Cycle 2 Cycle 6 Cycle 7

-15 -20 -25 0.0

0.2

0.4

0.6

0.8

1.0

1.2

Cathode potential / V vs. DHE Fig. 5 e Cyclic voltammograms of cathode after 40.5 h operation, at a scan rate of 0.01 V s¡1, in a potential range between 0.05 and 1.2 V vs. DHE.

Current density / mA· cm

-2

70 60

Initial Aged

50

of the adsorbed species gradually decreased. Simultaneously, the hydrogen desorption peak recovers. The CV profiles reach a steady state at the 6th cycle, and no further changes are observed from cycle 6 to cycle 7. It is indicated that the absorbed DME and/or intermediates have been fully oxidized. The poisoning of catalyst by adsorbed DME and/or intermediates species can debase the O2 reduction reaction on Pt catalyst surface, which is another cause of the cathode catalyst failure. The LSV curves for determining the DME crossover rate of fresh and faded cathode are shown in Fig. 6. At high cathode potential region, the DME electro-oxidation reaction is not dominated by the catalytic activity of the cathode, but governed by the DME permeation rate. The limiting current density of DME oxidation increases from about 41 mA cm2 to 48 mA cm2 after the durability test. The DME permeation rate gets higher after the durability test. It means that DME concentration at the interface between the anode and the membrane gets higher after the durability test, which caused by anode degradation. DME crossover is concomitant with performance loss of cathodic Pt due to the formation of mixed potential on cathodic catalyst. This mixed potential response [38] is also very deleterious and accelerates the sintering/degradation of cathodic Pt nanoparticles. Therefore, anode degradation is another factor which accelerates cathodic catalyst degradation.

40

Conclusions

30 20 10 0 0.30

0.35

0.40

0.45

0.50

0.55

0.60

Cathode potential / V vs. DHE Fig. 6 e Linear sweep voltammograms of DME permeating through the membranes at the cathode before and after durability test.

catalysts according to DebyeeScherrer formula because it is isolated from the diffraction peaks of carbon support [37]. The particle size of cathodic catalyst grows from an original value of 3.0 nm to 5.4 nm. The sintering rate of cathodic catalyst is evidently high because small Pt particles are unstable in harsh corrosion operation environment of cathode. Besides, DME crossover may aggravate cathodic Pt degradation. TEM images of the cathodic Pt/C catalyst before and after durability test are presented in Fig. 3 and the corresponding histograms are shown in Fig. 4. The dark black portions in Fig. 3 are Pt grains, and the gray portions are carbon support grains. The results show that Pt nanoparticles before durability test is small and even without agglomeration. After the durability test, Pt grains grow up evidently, which is consistent with the above XRD results. The growth of the Pt particles must be the main cause of the cathodic catalyst failure. Fig. 5 shows the CV scans of cathode after 40.5 h operation. The adsorbed DME and/or intermediates species which accumulate on cathodic Pt surface would be oxidized after the CV scans. With an increasing cycling number, the oxidation peak

The durability of cathodic Pt/C catalyst for DDFC has been investigated at a constant current density of 30 mA cm2 for 70.5 h discontinuous operation under ambient pressure and at a cell temperature of 60  C. The performance of cathodic Pt/C catalyst is decreased after the durability test. The cathodic EAS reduces from 349 to 267 cm2 mg1 Pt after 70.5 h operation. The particle size of cathodic Pt catalyst increased from an original value of 3.0 nm to 5.4 nm. The poisoning of catalyst by absorbed DME and/or intermediates is another cause of cathodic catalyst degradation. Furthermore, the anode decay accelerated the DME crossover and the severer poisoning of catalyst aggravated the degradation of the Pt/C cathodic catalyst in consequence.

Acknowledgment This work is financially supported by Young Creative Talent Training Project of Heilongjiang Province (UNPYSCT-2016108), Preliminary Investigation Project of Mudanjiang Normal University (GY201201), Doctoral Research Fund of Mudanjiang Normal University (MSB201208), Innovation Project of University Students (201610233031) and Natural Science Foundation of Heilongjiang Province (QC2014C009).

references

[1] Fletcher T, Thring R, Watkinson M. Int J Hydrogen Energy 2016;41:21503. [2] Ye LC, Gao Y, Zhu SY, Zheng JS, Li P, Zheng JP. Int J Hydrogen Energy 2017;42:7241.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 1 6 6 9 5 e1 6 6 9 9

[3] Authayanun S, Saebea D, Patcharavorachot Y, Assabumrungrat S, Arpornwichanop A. Int J Hydrogen Energy 2017;42:1977. [4] Wang ZB, Zuo PJ, Yin GP. Fuel Cells 2009;9:106. [5] Lai QZ, Yin GP, Wang ZB, Du CY, Zuo PJ, Cheng XQ. Fuel Cells 2008;8:399. [6] Zhao TS, Xu C, Chen R, Yang WW. Prog Energy Combust Sci 2009;35:275. [7] Shen SY, Zhao TS, Xu JB. Energy Environ Sci 2011;4:1428. [8] Xie F, Chen C, Meng H, Shen PK. Fuel Cells 2007;7:319. [9] Shao YY, Liu J, Wang Y, Lin YH. J Mater Chem 2009;19:46. [10] Liu P, Yin GP, Du CY. Electrochem Commun 2008;10:1471. [11] Heo P, Nagao M, Sano M, Hibino T. J Electrochem Soc 2008;155:B92. [12] Li Q, Wu G, Johnston CM, Zelenay P. Electrocatalysis 2014;5:310. [13] Li Q, Wen XD, Wu G, Chung HT, Gao R, Zelenay P. Angew Chem Int Ed 2015;127:7634. [14] Cai KD, Wang C, Pu WH, Gao Y, Mao ZQ. Int J Hydrogen Energy 2012;37:12605. [15] Semelsberger TA, Borup RL, Greene HL. J Power Sources 2006;156:497. [16] Kerangueven G, Coutanceau C, Sibert E, Leger JM, Lamy C. J Power Sources 2006;157:318. [17] Tong YJ, Lu LL, Zhang Y, Gao YZ, Yin GP, Osawa M, et al. J Phys Chem C 2007;111:18836. [18] Liu Y, Muraoka M, Mitsushima S, Ota KI, Kamiya I. Electrochim Acta 2007;52:5781. [19] Lu LL, Yin GP, Tong YJ, Zhang Y, Gao YZ, Osawa M, et al. J Electroanal Chem 2008;619:143. [20] Lu LL, Yin GP, Tong YJ, Zhang Y, Gao YZ, Osawa M, et al. J Electroanal Chem 2010;642:82.

16699

[21] Tadao H, Yasuyuki T, Hiroyuki T, Naoaki T, Susumu Y. Electr Eng Jpn 2005;150:19. [22] Zhang Y, Lu LL, Tong YJ, Osawa M, Ye S. Electrochim Acta 2008;53:6093. [23] Lu LL, Yin GP, Wang ZB, Cai KD, Gao YZ. Catal Commun 2009;10:971. [24] Yu JH, Choi HG, Cho SM. Electrochem Commun 2005;7:1385. [25] Yoo JH, Choi HG, Chung CH, Cho SM. J Power Sources 2006;163:103. [26] Jiang LH, Sun GQ, Wang SL, Wang GX, Xin Q, Zhou ZH, et al. Electrochem Commun 2005;7:663. [27] Schmittinger WG, Vahidi A. J Power Sources 2008;180:1. [28] Du CY, Chen M, Cao XY, Yin GP, Shi PF. Electrochem Commun 2009;11:496. [29] Zhang SS, Yuan XZ, Wang HJ, Merida W, Zhu H, Shen J, et al. Int J Hydrogen Energy 2009;34:338. [30] Zhang S, Shao YY, Yin GP, Lin YH. Angew Chem 2010;49:2211. [31] Baturina OA, Swider-Lyons KE. J Electrochem Soc 2009;156:B1423. [32] Wang ZB, Zuo PJ, Chu YY, Shao YY, Yin GP. Int J Hydrogen Energy 2009;34:4387. [33] Xing LH, Yin GP, Wang ZB, Zhang S, Gao YZ, Du CY. J Power Sources 2012;198:170. [34] Wu YN, Liao SJ, Su HN, Dang D, Liu B. Fuel Cells 2010;10:920. [35] Ren XM, Springer TE, Zawodzinski TA, Gottesfeld S. J Electrochem Soc 2000;147:466. [36] Xing LH, Gao YZ, Wang ZB, Du CY, Yin GP. Int J Hydrogen Energy 2011;36:11102. [37] H Zhang Z, Ge JJ, Ma L, Liao JH, Lu TH, Xing W. Fuel Cells 2009;09:114. [38] Hibino T, Kobayashi K. J Electrochem Soc 2014;5:H326.