Wire mesh current collectors for passive direct methanol fuel cells

Journal of Power Sources 272 (2014) 629e638

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Wire mesh current collectors for passive direct methanol fuel cells Naveen K. Shrivastava a, *, Shashikant B. Thombre a, Ramani V. Motghare b a b

Department of Mechanical Engineering, Visvesvaraya National Institute of Technology, Nagpur 440010, India Department of Chemistry, Visvesvaraya National Institute of Technology, Nagpur 440010, India

h i g h l i g h t s
Feasibility of stainless steel wire mesh as current collector (CC) in passive DMFC is investigated.
A novel single cell fixture is designed and fabricated.
Five different wire meshes are used as CC.
Wire mesh CC exhibited better fuel distribution at anode catalyst layer compared to conventional CC.
Identifies the wire mesh as promising material to be used as CC.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 23 July 2014 Received in revised form 29 August 2014 Accepted 2 September 2014 Available online 16 September 2014

This paper examines the feasibility of the stainless steel wire mesh as current collector in the passive direct methanol fuel cell (DMFCs-W). A novel single cell fixture is designed and fabricated. The cell performance is evaluated and compared with five different wire mesh current collectors. The supporting plates are optimized for every mesh. The performance of DMFCs-W is compared with the conventional passive DMFC which uses perforated metal plate as current collector (DMFC-P). The polarization tests and electrochemical impedance spectroscopy are performed to investigate the different aspects of the cell performance. The results reveal that the DMFCs-W yield better performance than the DMFC-P. Also, more uniform fuel distribution at catalyst layer and higher cell temperature is achieved with wire mesh current collectors. It is found that the wire mesh geometry has significant effect on the cell performance and the mesh made of relatively thick wires gives better cell performance. This study identifies the stainless steel wire mesh as promising material to be used as current collector and potential substitute to the perforated plate current collectors in the passive DMFC. © 2014 Elsevier B.V. All rights reserved.

Keywords: Passive direct methanol fuel cell Single cell Current collector Stainless steel wire mesh

1. Introduction The passive direct methanol fuel cells (DMFC) have emerged as potential source for powering portable electronic devices such as mobile, laptop, i-pod, etc. The passive DMFC does not have energy consuming auxiliary devices such as pumps, fans, blowers etc. and rely on passive means such as diffusion, natural convection and capillary action for supply/removal of the reactants/products [1,2]. Hence, the passive DMFCs have low parasitic power losses, high energy density, high efficiency, simpler and more compact structure and are considered as a great alternative to the rechargeable batteries [1e4]. The passive DMFC is primarily composed of methanol solution reservoir, membrane electrode assembly (MEA), anode and cathode

* Corresponding author. Tel.: þ91 712 2801535; fax: þ91 712 2223230. E-mail address: [email protected] (N.K. Shrivastava). http://dx.doi.org/10.1016/j.jpowsour.2014.09.010 0378-7753/© 2014 Elsevier B.V. All rights reserved.

current collector, anode and cathode end plate. The current collectors are one of the most important components of the passive DMFC. These are used to collect the current generated in the MEA and provide passage for transportation of the reactants (methanol and water on the anode and oxygen on the cathode) and products (carbon di-oxide on the anode and water on the cathode). The current collectors are expected to have good mechanical strength, high electrical conductivity, low thermal conductivity, high corrosion resistance, uniformly distributed transport area, light weight, low cost, easy machining and wide availability [1,2,4]. Typically, the 1 mme3 mm thick metal plate with the array of drilled holes (known as perforated current collector) or with rectangular parallel channels (known as parallel current collector) are used as current collectors [3e25]. Different materials such as stainless steel (SS) plate [5,9e11], gold coated SS plate [13], platinum coated SS plate [16], titanium nitride (TiN) coated SS plate [21], gold coated printed circuit board (PCB) plate [22] and graphite plate [19] are used as current collectors.

630

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638

Several studies related to the structural aspects of perforated plate (hole-array or parallel channels) current collectors have already been performed. The effect of non-uniform parallel channels on the performance of passive DMFC has been studied by Gholami et al. [11]. They showed that the current collector with non-uniform parallel channels is more effective in removing CO2 gas than the current collector with parallel channels. Yuan et al. [9] investigated the effects of structural diversity on the performance of a liquid-fed passive DMFC. They recommended the use of circular hole-array pattern with a lower open ratio at the anode but parallel-fence pattern with a higher open ratio at the cathode. A passive DMFC with its cathode current collector made of porous metal foam has been investigated experimentally by Chen & Zhao [26]. They showed that the passive DMFC having the porous current collector yielded much higher and more stable performance than the cell having the conventional perforated-plate current collector with high methanol concentration operation. Metal mesh provides an excellent alternate or substitute to the perforated metal plate current collectors. As compared to the

perforated metal plates; the metal meshes are lighter in weight, cheaper, involves low fabrication cost and offers high open ratio. The metal mesh can be categorized mainly into two types; expanded metal mesh and wire mesh. The expanded metal mesh is produced by simultaneously slitting and stretching a metal plate or sheet. In this process the cuts are expanded into diamond shaped holes. The expanded metal mesh has been used as current collector in many of the previous studies [27e37]. Platinum coated niobium expanded mesh, platinum coated stainless steel (SS) expanded metal mesh, 304 SS expanded mesh, SS expanded mesh with gold coating were used as current collector in several passive DMFC stacks and single cells. The expanded metal mesh can be characterized by opening dimensions, strand width, strand thickness and open ratio etc. Not much research efforts regarding structural optimization of the expanded metal mesh as current collector have been found. Guo and Faghri [36] compared the single cell performance with gold coated SS expanded mesh and platinum coated niobium expanded mesh as current collectors which was either put directly on the diffusion layers (DL) or hot-pressed on the diffusion

Current collectors in Passive DMFC

Perforated plate Perforate geometry

Circular holes [5, 9, 10, 11, 12, 17, 19]

Hexagonal holes [20]

Rectangula r parallel channels [9, 11]

Material

Pt coated [16, 20], Au coated [13], TiN plated [21] Stainless Steel plate

316L SS plate [5, 10, 12, 15, 17]

Graphite plate [19]

printed circuit board (PCB) plate covered by gold film [22], copper film [11] Ni–Cr alloy metal foam [26]

Thickness & open raƟo

Thickness : 1 mm [17, 10], 1.5 mm [16], 2 mm [23].

Open raƟo : 28.3% [9, 10, 12, 17], 33% [24], 38.5% [9, 10, 12, 17], 42% [11], 47.8% [16, 15], 54.1% [25], 58% [9], 63% [9], 74% [23].

Metal mesh Remark It is recommended to use hole-array paƩerns with lower open raƟos at the anode but parallel channels with higher open raƟos at the cathode [9].

Types Expanded metal mesh

Wire mesh Carbon dioxide removal is easier in non-uniform channels than in parallel one [11]. In parallel channels, water removal is easier and the cell has more performance stability [11]

Higher open raƟo facilitates the removal of the produced gas bubbles and residual water [12].

porous current collector yielded much beƩer performance than did the perforated-plate current collector [26].

Fig. 1. Current collectors in passive DMFCs.

Material

Pt coated niobium expanded metal mesh [27, 28, 29, 31, 32, 33, 34, 35]

Pt coated [37], Au coated [39, 40] Ni mesh

Stainless steel mesh [30, 41, 42]

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638

Fig. 2. Schematic of the passive DMFC with wire mesh current collector (DMFCs-W).

layer (DL). It was found that the cell with the platinum coated niobium expanded mesh which was hot pressed with DL showed the best performance. In order to improve the performance of a micro-direct methanol fuel cell; Zhang et al. [38] employed titanium nitride plated stainless steel mesh between the MEA and current collectors. It was found that the inclusion of stainless steel mesh is helpful in improving cell performance and efficiency.

631

The wire meshes are made up of thin wires; each wrap wire passes alternately over and under the successive weft wires. The woven wire mesh is a versatile material and offers a wide range of opening size, wire diameter and open ratio. Wire mesh is the only option if the required opening size has to be extremely small. Also, it is very flexible material and can be moulded into almost any shape. Therefore, the wire mesh can be considered as an attractive material to be used as current collector in the passive DMFC. Kim et al. [39,40] used gold coated nickel mesh as current collector in their study related to passive DMFC but did not disclose the structural details of the mesh. Shimizu et al. [41] used bare stainless steel (SUS) mesh and gold coated SUS mesh of 0.5 mm thickness and 20% open ratio as current collector in a study intended to design, fabricate and performance evaluation of a 36 cm2 passive air breathing DMFC. Zheng et al. [42] proposed a new structure of passive DMFC which consists of a stainless steel mesh directly welded onto the polymer electrolyte membrane (PEM), which was then sprayed with the catalyst ink. The fuel cell was operated successfully with a 2 mol L1 methanol concentration and produced a power output of 1.52 mW cm2. The literature summary on current collectors for passive DMFC is presented in Fig. 1. It can be seen that many research studies have been done on the structural diversity of metal plate current collectors and many recommendations related to current collector structure for the better cell performance have been made. The literature review also indicates that although, wire meshes were used as current collector in the few passive DMFCs, but no study has been found concerning the structural aspects of the wire mesh current collector in the passive DMFC. In the present work, the feasibility of using woven wire mesh as a current collector in the passive DMFC is assessed. For this, a novel single cell fixture of the passive DMFC that incorporates the wire mesh as current collector has been designed and fabricated. The cell performance is evaluated and compared with five different wire mesh current collectors. This contribution will be helpful in providing better insights and

Fig. 3. Stainless steel wire mesh.

632

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638

important information about the passive DMFC with wire mesh current collectors. 2. Experimental 2.1. Fabrication of membrane electrode assembly The MEA with an active area 5.0 cm  5.0 cm, was fabricated by hot pressing a Nafion®117 polymer electrolyte membrane between anode and cathode diffusion electrodes at 135  C, 8 MPa for 3 min. The Nafion® membrane was pretreated to remove the organic and inorganic contaminants before hot pressing. The pretreatment procedures included boiling the membrane in 3 wt.% H2O2 solution for 1 h, followed by washing in deionised (DI) water, then boiling in 0.5 M H2SO4 solution for 1 h and finally boiling in DI water for 1 h. The pretreated membrane was kept in DI water prior to the MEA fabrication. The carbon cloths (Avcarb 1071) with 20% polytetrafluoroethylene (PTFE) content were used as anode and cathode backing layer. The diffusion layers were made by coating a microporous layer on the anode and cathode backing layers with carbon loading of 1 mg cm2 (carbon powder, Vulcan XC-72) and PTFE content of 30%. The diffusion electrodes were made by coating the catalyst ink on the anode and cathode diffusion layer. The catalyst ink was prepared by dispersing an appropriate amount of the catalyst in a solution of DI water, isopropyl alcohol, and Nafion solution. 80 wt.% PteRu/C and 55 wt. % Pt/C were used as catalyst on anode and cathode side respectively. The catalyst loading was 4 mg cm2 on the anode and 2 mg cm2 on the cathode. 2.2. Single cell fixture The schematic diagram of the passive DMFC with wire mesh as current collector (DMFCs-W) is shown in Fig. 2. The single cell consists of fuel reservoir, MEA, current collectors, supporting plates, anode and cathode end plates and gaskets. The fuel reservoir with a volume of 25 ml was built in the anode fixture. The openings on the anode fixture were provided for fuel injection and CO2 exhaust. The anode fixture, cathode fixture and supporting plates were made of transparent acrylic plates. In this study, five stainless steel wire meshes with different structures were used as current collector and are shown in Fig. 3. The geometrical parameters of these wire meshes are listed in Table 1. The wire meshes are flexible in construction. Good contact between the MEA and wire mesh current collector is required for better cell performance. For this, supporting plates were used to press the wire mesh current collectors against the MEA, which is shown in Fig. 4. The supporting plates have square opening arrays uniformly distributed within the active area to facilitate fuel/ oxidant transport towards the electrodes. The square openings were separated by 2 mm thick successive ribs. The ribs were used to provide an even distribution of compressive force on the wire mesh current collector which ensures better contact between current collector and MEA. These ribs were evenly distributed over the active area along two mutually perpendicular directions. Table 1 Geometrical properties of wire meshes. Wire mesh

Mesh count

Wire diameter (mm)

Aperture size (mm)

Open ratio (%)

A B C D Ea

20 30 50 200 500

0.457 0.3150 0.1930 0.0610 0.0305

0.813 0.532 0.315 0.066 0.020

41.0 39.4 38.5 27.0 16.0

a

Wire mesh-E has been made of multiple mesh layers creating micron size pores.

Fig. 4. Photograph of the supporting plate SP6.

The number of ribs on the supporting plate may have a crucial impact on the fuel cell performance. A supporting plate with increased number of ribs may provide lower electrical contact resistance between the diffusion layer and the current collector and leads to better cell performance. On the other hand, upon increasing the number of ribs, the effective open ratio on the respective electrode side decreases, which may affect the reactant mass transport and can lead to fuel/oxidant starving condition at various locations in the catalyst layers. The number of ribs in the supporting plate has theses two contradictory effects associated with it, therefore it needs to be optimized. The different configurations of the supporting plate have been used in this connection. The geometrical details of the various supporting plates are listed in Table 2. The number of ribs (N) has been treated as “independent parameter” in the supporting plate design. The number of ribs N  N indicates N ribs in horizontal and N ribs in vertical direction and this facilitates (N þ 1)  (N þ 1) number of equal size square openings in the supporting plate. Other dependent parameters viz. opening size ((50  2N)/(N þ 1)  (50  2N)/(N þ 1) mm  mm) and open ratio ((50  2N)2/2500  100%) can be determined based on the number of ribs (N). As the number of ribs in the supporting plate increases, the open area decreases. The supporting plate with ‘N  N’ number of ribs (N ¼ 1, 2, 3…10) is denoted as ‘SPN’. For example the supporting plate with 6  6 ribs is denoted as SP6 (Fig. 4). Abbreviations have been used to denote various passive Table 2 Supporting plate configurations. Supporting plate

Number of ribs

Number of openings

Size of one opening (mm  mm)

Open ratio (%)

SP1 SP2 SP3 SP4 SP5 SP6 SP7 SP8 SP10

11 22 33 44 55 66 77 88 10  10

22 33 44 55 66 77 88 99 11  11

24.0  24.0 15.3  15.3 11.0  11.0 8.4  8.4 6.7  6.7 5.4  5.4 4.5  4.5 3.8  3.8 2.7  2.7

92.2 84.6 77.4 70.6 64.0 57.8 51.8 46.2 36.0

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638 Table 3 Abbreviations for passive DMFCs. DMFCs-W

DMFC-A DMFC-B DMFC-C DMFC-D DMFC-E DMFC-A-SPN (N ¼ 1, 2,…10) DMFC-P

Passive DMFCs with wire mesh current collector (represents the group of DMFC-A, DMFC-B, DMFC-C, DMFC-D and DMFC-E) Passive DMFC with wire mesh-A as current collector Passive DMFC with wire mesh-B as current collector Passive DMFC with wire mesh-C as current collector Passive DMFC with wire mesh-D as current collector Passive DMFC with wire mesh-E as current collector Passive DMFC with wire mesh-A as current collector and supporting plate configuration SPNa Conventional passive DMFC which uses perforated plate as current collector

Abbreviations: DMFC-B-SPN, DMFC-C-SPN, DMFC-D-SPN, DMFC-E-SPN (where N ¼ 1, 2…10) are used for a similar meaning. a For example; DMFC-A-SP4 represents the passive DMFC with wire mesh-A as current collector and supporting plate configuration SP4.

DMFCs which are listed in Table 3. The abbreviations denote the cells with various combinations of the wire mesh current collector and the supporting plate configuration. For example, the passive DMFC with “wire-mesh C” current collector and supporting plate configuration “SP6” is denoted as “DMFC-C-SP6”. A conventional passive DMFC (DMFC-P) which uses perforated stainless steel plate as current collector has also been used in this work to compare its performance with DMFCs-W. The schematic diagram of the DMFC-P is shown in Fig. 5. The anode and cathode current collectors were made of 3 mm thick 316L stainless steel plate with perforated hole-array (4.5 mm diameter, 64 Nos.) distributed within the active area, resulting in an open ratio of 40.7%. The anode end plate, the cathode end plate and the fuel reservoir were same as those for DMFCs-W.

633

value. The currentevoltage curves were recorded by using the DC electronic load (N3300A Mainframe & N3302A Module, Agilent Technologies). For each discharging current point along the IeV curve, a 60 s waiting time was used to obtain the stable voltage. Prior to the performance test, the new MEA was activated by filling the reservoir with 1 M methanol solution and allowing the cell to stand overnight and then running the cell at a constant load for 12 h. The cell temperature was measured by a thermocouple (Ktype), installed between anode current collector and MEA. All the experiments were carried out at a room temperature of 24e26  C and a relative humidity of 60e70%. Electrochemical impedance spectra of the passive DMFCs were measured under the potentiostatic mode (300 mV) using an electrochemical analysis instrument (Gamry Interface 1000) in a frequency range from 100 kHz to 0.01 Hz with 10 points per decade. The amplitude of the AC voltage was 10 mV. 3. Results and discussion 3.1. Supporting plate optimization of DMFC-C As mentioned earlier that the supporting plate configuration may affect the cell performance. Therefore, the supporting plate

2.3. Experimental setup and test conditions After filling the reservoir with methanol solution, the collection of the polarization data started when the cell voltage under the open circuit condition (open circuit voltage; OCV) reached a stable

Fig. 5. Schematic of the conventional passive DMFC with perforated plate current collector (DMFC-P).

Fig. 6. Supporting plate optimization of DMFC-C (a) IeV curve (b) IeP curve (Methanol feed concentration: 3 M).

634

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638

optimization has been performed for every mesh current collector. Fig. 6(a) & (b) shows the currentevoltage (IeV) and currentepower (IeP) characterization curve respectively for the different supporting plate configurations of the passive DMFC-C. The wire meshC is the current collector in DMFC-C. The experiments have been conducted with 3 M methanol feed concentration. The objective is to increase the number of ribs, which has been treated as an independent parameter with a constant thickness of 2 mm in all the supporting plates, and observe the effect on the cell performance. Eight supporting plate configurations namely SP1, SP2, SP3, SP4, SP5, SP6, SP7 and SP8 have been used in passive DMFC-C. Firstly, the cell with supporting plate configuration SP1 (denoted as DMFCC-SP1) is tested followed by supporting plates with successively increasing number of ribs employed in the cell in the order SP2, SP3…SP8. It can be seen from Fig. 6 that on varying the supporting plate configuration from SP1 to SP6 (increasing the number of ribs from 1  1 to 6  6); the cell performance increases. This indicates that on increasing the number of ribs on the supporting plate the electrical contact resistance between the MEA and current collector decreases which leads to better cell performance. Although, the open ratio is also decreasing on increasing the number of ribs, the contact resistance factor is dominant. On further increasing the number of ribs to 7  7 and 8  8 (by using SP7 and SP8) the cell performance decreases. This indicates that now the open ratio is producing the dominating effect, as its further decrement causes the increase in non-uniform distribution of fuel at catalyst layers which leads to insufficient fuel availability at the various locations on the catalyst layers. The optimum cell performance is obtained with SP6, which is having 6  6 number of ribs and open ratio of 57.8%. This represents the best compromise between the two phenomena, the contact resistance between MEA-diffusion layer and the fuel distribution uniformity at catalyst layers, which varies adversely with the change in supporting plate configuration. Fig. 7 compares the performance between DMFC-C with optimized supporting plate (SP6) and the DMFC-C without supporting plate (DMFC-C-SP6 vs DMFC-C-without SP). The passive DMFC without supporting plate is producing maximum power output of 1.724 mW cm2. Whereas, the cell with optimized supporting plate configuration (SP6; DMFC-C-SP6) is producing maximum power output of 2.264 mW cm2, which is 1.31 time the power produced by the cell without supporting plate. It is due to the better contact provided by the supporting plate between MEA and current collector. This justifies the significance of the supporting plate in the passive DMFC assembly, while using wire mesh current collector.

Fig. 7. Performance comparison between passive DMFC-C-SP6 and passive DMFC-C without supporting plate.

Interestingly, the four passive DMFCs-W (DMFC-A, B, C and D) gave optimum performance with the supporting plate SP6. 3.3. Supporting plate optimization of DMFC-E The cell performance with wire mesh-E (DMFC-E) is different and is shown in Fig. 9(a) and (b). The DMFC-E performance has been measured with following sequence of supporting plate: SP1 / SP2 / SP3 / SP4 / SP5 / SP6 / SP8 / SP10. It can be seen from Fig. 9 that the optimum cell performance is obtained with supporting plate SP5 (DMFC-E-SP5). Interestingly, the DMFC-E delivered almost same performance with SP5, SP6, and SP8. The cell performance decreases with the supporting plate SP10 but the decrement is insignificant, which can be clearly seen in Fig. 10. Fig. 10 depicts the voltage output of the various passive DMFCsW at a current density 14 mA cm2. The results are derived from the IeV characteristic curves obtained for the supporting plate optimization of passive DMFCs-W. It can be seen that the decreasing open ratio of the supporting plate does not have much impact on DMFC-E performance. This can be attributed to the physical characteristics of the wire mesh-E. The wire mesh-E has been made of multiple mesh layers, each of which has been closely woven by very thin wires creating micron size pores. These small pores enhance

3.2. Supporting plate optimization of DMFC-A, DMFC-B and DMFC-D Fig. 8 shows IeV and IeP curves for the supporting plate optimization of the passive DMFC with wire mesh-A (DMFC-A). It is important to mention here that, the optimum supporting plate configuration obtained for DMFC-C (i.e. SP6) has not been employed directly in the cell with other wire mesh current collectors (DMFC-A, B, D, E). Instead, the supporting plate optimization has been performed separately for every wire mesh current collectors. This has been done because it is felt that every wire mesh has different physical structure and therefore the optimum configuration of the supporting plate may be different for different wire meshes. Also, this makes the cell performance comparison between the wire mesh current collectors more reliable. Keeping the pattern of the results obtained for supporting plate optimization of the DMFC-C in mind and to reduce the number of experiments, the supporting plate optimization for DMFC-A, DMFC-B and DMFC-D has been performed with limited number of supporting plate configurations (SP5, SP6, SP7 and SP8). The sequence of supporting plates used in each cell was: SP5 / SP6 / SP7 / SP8.

Fig. 8. Supporting plate optimization of passive DMFC-A. (Methanol feed concentration: 3 M).

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638

635

much effect on the distribution of methanol solution at the anode catalyst layer and the cell performance remains almost same with supporting plates SP5, SP6, SP8, and SP10. Therefore it can be suggested that, when the optimum supporting plate configuration is not known, then the wire mesh-E can be used as current collector with the supporting plate having the higher number of ribs without worrying about the decrement in the cell performance. 3.4. Performance comparison between DMFCs-W and DMFC-P The performance of passive DMFCs-W (DMFC-A, B, C, D & E) and DMFC-P are shown together in Fig. 11(a) and (b). All the cell performances have been evaluated at 3 M methanol feed concentration. The DMFCs-W performance with optimized supporting plate configurations (as listed in Table 4) are used for the comparison. It is felt by the authors that, as the supporting plate optimization is over and now only the wire mesh type is of prime importance, there is no need to specify the supporting plate configuration explicitly in the abbreviations. So, the abbreviations DMFC-A, DMFC-B, DMFC-C, DMFC-D and DMFC-E replace DMFC-A-SP6, DMFC-B-SP6, DMFC-CSP6, DMFC-D-SP6 and DMFC-E-SP5 in further discussions. Fig. 11 clearly indicates that the passive DMFCs-W give better performance than that of the DMFC-P. It has been observed that the cell

Fig. 9. Supporting plate optimization for passive DMFC-E (a) IeV curve (b) IeP curve. (Methanol feed concentration: 3 M).

the capillary action of methanolewater solution in the in-plane direction of the mesh. Hence, the methanol solution easily reaches the mesh area under supporting plate ribs due to which the increasing number of ribs on the supporting plate does not have

Fig. 10. Performance variation of DMFCs-W with change in supporting plate configuration at current density: 14 mA cm2 (methanol feed concentration: 3 M).

Fig. 11. Performance comparison between DMFCs-W and DMFC-P (a) IeV curve (b) IeP curve. (Methanol feed concentration: 3 M).

636

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638

Table 4 Optimum supporting plate configurations for passive DMFCs-W. Passive DMFCs-W

Supporting plate configuration for optimum cell performance

DMFC-A DMFC-B DMFC-C DMFC-D DMFC-E

SP-6 SP-6 SP-6 SP-6 SP-5

compared to the perforated plate current collectors. Therefore, it is concluded that the wire mesh current collector is a potential material to be used as current collector and can be a substitute to the perforated plate current collector in the passive DMFC. It can also be seen in Fig. 11 that; DMFC-C, DMFC-D and DMFC-E give similar performance, DMFC-A and DMFC-B give better performance and DMFC-A gives the best performance amongst the DMFCs-W. The DMFCs-W performance can be summarized in the following order (from the best to the worst):

performance depends on the combined effect of several parameters such as reactant transport mechanism, uniformity of fuel distribution at catalyst layers, electrical resistances associated with the current collector (i.e. electrical contact resistance between MEA and current collector and in-plane electrical resistance of the current collector) and cell temperature etc. The better performance of the DMFC-W can be attributed to the fact that it exhibits more uniform fuel distribution at anode catalyst layer (ACL) and higher cell temperature; which is evident from Fig. 12. Fig. 12 shows cell temperature variation with time under open circuit condition; recorded immediately after injecting the methanol solution in the reservoir. It is interesting to note from the figure that DMFCs-W attains considerably higher temperature than the DMFC-P. This indicates that a better fuel distribution at ACL is achieved by DMFCs-W. A uniformly distributed methanol solution at ACL yields higher rate of methanol crossover (MCO) across the PEM. This leads to higher rate of exothermic reaction between the permeated methanol and oxygen at cathode catalyst layer (CCL) and results in higher cell temperature. A higher cell temperature promotes the electrochemical kinetics of methanol oxidation and oxygen reduction reaction and leads to better cell performance. It should be noted that in the conventional passive DMFC, the ribs of the perforated plate current collector comes in direct contact with the diffusion layer and a fuel vacant region is created under the current collector's rib [43,44]. Whereas in the DMFCs-W; the wavy surface of the wire mesh current collectors create enough space between current collector-supporting plate ribs and current collectorsdiffusion layer which allows mass transfer in the in-plane direction. It ensures better distribution of fuel on the diffusion layers and catalyst layers. The discussion made above indicates that the wire mesh current collectors produce more uniform fuel distribution over the catalyst layers and achieve higher cell temperature

DMFC  A > DMFC  B > DMFC  E > DMFC  D > DMFC  C

Fig. 12. Variation in cell operating temperature at OCV. (Methanol feed concentration: 3 M).

Fig. 13. Nyquist plots of passive DMFC impedance spectra. (Methanol feed concentration: 3 M).

It indicates that the physical characteristic of the wire mesh has crucial impact on the cell performance. It has been observed that the electrical resistances associated with wire mesh current collector and the uniformity of fuel distribution influenced by the wire mesh are playing important roles. The electrical resistance associated with wire mesh current collector includes electrical contact resistance between MEA and the current collector and in-plane electrical resistance of wire mesh. Fig. 13 shows the electrochemical impedance spectra (EIS) for the DMFC-A, DMFC-B, DMFC-C, DMFC-E and DMFC-P measured at 0.3 V with 3 M methanol feed concentration. The intersection of the first arc in the high frequency region on the real axis (X-axis) represents the total ohmic resistance of the cell [45]. The total ohmic resistance of the cell includes contact resistance between MEA and current collectors and ohmic resistance of cell components such as PEM, catalyst layers, diffusion layers (together as MEA) and current collectors. It is important to mention that the same MEA has been used in all the single cell experiments. Therefore, the variation in total ohmic resistance (indicated by EIS) is attributed to the contact resistance between MEA and current collector and the ohmic resistance of current collector. The wire mesh has a large number of wire-to-wire contacts and hence it generally suffers from higher in-plane electrical resistance compared to that of the perforated plate current collectors which is evident from Fig. 13. It can also be seen in Fig. 13 that among the DMFCs-W; the DMFC-A exhibits lowest, DMFC-B and DMFC-C exhibits moderate and DMFC-E exhibits the highest total ohmic resistance. However, wire mesh-A has highest, wire mesh-B and wire mesh-C has moderate and wire mesh-E has the lowest wire diameter (Table 1). It indicates that the total ohmic resistance of the cell is inversely proportional to the wire diameter of mesh used. The firmness with which wires of the mesh are in contact with each

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638

637

other (weaving strength) determines the in-plane electrical resistance of the wire mesh. Generally, the mesh made of thicker wires shows lower in-plane electrical resistance due to the better contact between the wires. Moreover, the mesh made with thicker wire has higher peaks, which penetrates the diffusion layer (carbon cloth) to a greater depth and makes better contact with it. It results in lower contact resistance between diffusion layer and wire mesh. Hence, it can be said that the mesh with thicker wires is expected to have lower overall electrical resistance. DMFC-A uses mesh made of thickest wires, exhibits lowest total ohmic resistance among the DMFCs-W and delivers the best cell performance, DMFC-B uses second thickest wire for mesh, exhibits second lowest total ohmic resistance among the DMFCs-W and delivers second best cell performance. DMFC-E exhibits the highest total ohmic resistance but gives a performance slightly better than DMFC-C which exhibits comparatively lower total ohmic resistance. The enhancement in the performance of DMFC-E is due to the better fuel distribution property shown by wire mesh-E. The wire mesh eE is made up of multiple mesh layers with very thin wires creating micron sized pores. The small pores enhance the capillary action of methanol water solution in the in-plane direction of the mesh and provide better fuel distribution at the ACL which leads to better cell performance. The DMFC-P however, exhibits lowest ohmic resistance but gives poorest cell performance. This is due to the inferior fuel distribution of perforated plate current collector as described earlier. 3.5. Performance comparison between DMFC-A and DMFC-P for different methanol feed concentration The performance comparison between DMFC-P and DMFC-A (with optimized supporting plate configurations, SP6 i.e. DMFCA-SP6) is shown in Fig. 14(a) & (b) for 1, 3 and 5 M methanol feed concentrations. The comparison of cell temperature variation at OCV is given in Fig. 15. It can be seen that, the performance of DMFC-A is better than DMFC-P for all methanol feed concentrations. The wire mesh-A as current collector in passive DMFC (DMFC-A) provides better methanol solution distribution at ACL compared to that provided by the perforated plate current collector of DMFC-P which leads to better cell performance. Moreover, the MCO increase due to the uniformly distributed methanol solution at ACL, which leads to higher cell temperature and enhances the cell performance. It is clearly indicated in Fig. 15 as DMFC-A achieves higher cell temperature with 3 M and 5 M methanol feed concentrations. However, the DMFC-P and DMFC-A temperature is almost same for 1 M methanol feed concentration. This is because the methanol crossover rate decreases with the decrease in methanol feed concentration and at 1 M methanol feed concentration the MCO is insignificant.

Fig. 14. Performance comparison between DMFC-P and DMFC-A (a) IeV curve (b) IeP curve. (Methanol feed concentration: 1 M, 3 M and 5 M).

4. Conclusions To investigate the feasibility of woven wire mesh as current collector in the passive DMFC; a novel single cell fixture is designed and fabricated. Five different wire meshes are used as current collector. The supporting plate optimization is performed for every mesh. Performance comparison between the DMFCs-W and DMFCP is presented. Based on the experimental results the following conclusions can be drawn: i. The supporting plate has been proved to be an important component of the DMFCs-W. ii. The performance of DMFCs-W varies with change in supporting plate configurations and the optimum cell performance can be achieved with the use of a unique supporting

Fig. 15. Variation in cell operating temperature at OCV for DMFC-P and DMFC-A. (Methanol feed concentration: 1 M, 3 M and 5 M).

638

N.K. Shrivastava et al. / Journal of Power Sources 272 (2014) 629e638

iii.

iv.

v. vi.

plate configuration. The unique supporting plate configuration may be different for different wire mesh current collectors. Since the wire mesh geometry determines the mass transfer pattern, cell electrical resistance etc. therefore, it has significant effect on the cell performance. It has been found that the mesh made of relatively thick wires give better cell performance. A mesh with thin wires and micro-sized openings (wire mesh-E; DMFC-E) exhibit excellent in-plane mass transport due to enhanced capillary action and maintain the optimum cell performance over a wide range of supporting plate configurations. When the optimum supporting plate configuration is not known then the wire mesh -E, can be used in a passive DMFC with supporting plates having excessively high number of ribs without worrying about the decrement in the cell performance. The DMFCs-W exhibit better fuel distribution at ACL and achieve higher cell temperature compared to those in DMFC-P. This study identifies the stainless steel wire mesh as promising material to be used as current collector in the passive DMFC and a potential substitute to the conventional perforated plate current collector.

References [1] S.K. Kamarudin, F. Achmad, W.R.W. Daud, Int. J. Hydrogen Energy 34 (2009) 6902e6916. [2] T.S. Zhao, R. Chen, W.W. Yang, C. Xu, J. Power Sources 191 (2009) 185e202. [3] M.Z.F. Kamaruddin, S.K. Kamarudin, W.R.W. Daud, M.S. Masdar, Renew. Sustain. Energy Rev. 24 (2013) 557e565. [4] S. Yousefi, M. Zohoor, Int. J. Hydrogen Energy 39 (2014) 5972e5980. [5] S. Yousefi, D.D. Ganji, Electrochim. Acta 85 (2012) 693e699. [6] S. Yousefi, M. Shakeri, K. Sedighi, Ionics 19 (2013) 1637e1647. [7] W.M. Yang, S.K. Chou, C. Shu, J. Power Sources 164 (2007) 549e554. [8] W. Yuan, J. Deng, Z. Zhang, X. Yang, Y. Tang, Renew. Energy 62 (2014) 640e648. [9] W. Yuan, Y. Tang, X. Yang, B. Liu, Z. Wan, Int. J. Hydrogen Energy 37 (2012) 9298e9313. [10] W. Yuan, Y. Tang, Z. Wan, M. Pan, Int. J. Hydrogen Energy 36 (2011) 2237e2249. [11] O. Gholami, S.J. Imen, M. Shakeri, Int. J. Hydrogen Energy 38 (2013) 3395e3400.

[12] Y. Tang, W. Yuan, M. Pan, B. Tang, Z. Li, Z. Wan, J. Power Sources 195 (2010) 5628e5636. [13] J. Cao, M. Chen, J. Chen, S. Wang, Z. Zou, Z. Li, D.L. Akins, H. Yang, Int. J. Hydrogen Energy 35 (2010) 4622e4629. [14] L. Feng, J. Zhang, W. Cai, Liangliang, W. Xing, C. Liu, J. Power Sources 196 (2011) 2750e2753. [15] Q.X. Wu, T.S. Zhao, R. Chen, W.W. Yang, Int. J. Hydrogen Energy 35 (2010) 10547e10555. [16] R. Chen, T.S. Zhao, J. Power Sources 167 (2007) 455e460. [17] W. Yuan, Y. Tang, Q. Wang, Z. Wan, Appl. Energy 88 (2011) 1671e1680. [18] J.G. Liu, T.S. Zhao, R. Chen, C.W. Wong, Electrochem. Commun. 7 (2005) 288e294. [19] Q.Z. Lai, G.P. Yin, J. Zhang, Z.B. Wang, K.D. Cai, P. Liu, J. Power Sources 175 (2008) 458e463. [20] Y.H. Chan, T.S. Zhao, R. Chen, C. Xu, J. Power Sources 178 (2008) 118e124. [21] Y. Zhu, J. Liang, C. Liu, T. Ma, L. Wang, J. Power Sources 193 (2009) 649e655. [22] V. Baglio, A. Stassi, F.V. Matera, A.D. Blasi, V. Antonucci, A.S. Arico, J. Power Sources 180 (2008) 797e802. [23] T. Tsujiguchi, M.A. Abdelkareem, T. Kudo, N. Nakagawa, T. Shimizu, M. Matsuda, J. Power Sources 195 (2010) 5975e5979. [24] T. Yuan, Z. Zou, M. Chen, Z. Li, B. Xia, H. Yang, J. Power Sources 192 (2009) 423e428. [25] Y.H. Chan, T.S. Zhao, R. Chen, C. Xu, J. Power Sources 176 (2008) 183e190. [26] R. Chen, T.S. Zhao, Electrochim. Acta 52 (2007) 4317e4324. [27] C. Xu, A. Faghri, X. Li, T. Ward, Int. J. Hydrogen Energy 35 (2010) 1769e1777. [28] G. Jewett, Z. Guo, A. Faghri, J. Power Sources 168 (2007) 434e446. [29] A. Faghri, Z. Guo, Appl. Therm. Eng. 28 (2008) 1614e1622. [30] J.J. Martin, W. Qian, H. Wang, V. Neburchilov, J. Zhang, D.P. Wilkinson, Z. Chang, J. Power Sources 164 (2007) 287e292. [31] G. Jewett, Z. Guo, A. Faghri, Int. J. Heat Mass Transf. 52 (2009) 4573e4583. [32] Z. Guo, A. Faghri, Int. Commun. Heat Mass Transf. 35 (2008) 225e239. [33] X. Li, A. Faghri, C. Xu, Int. J. Hydrogen Energy 35 (2010) 8690e8698. [34] C. Xu, A. Faghri, X. Li, Int. J. Hydrogen Energy 36 (2011) 8468e8477. [35] X. Li, A. Faghri, J. Power Sources 196 (2011) 6318e6324. [36] Z. Guo, A. Faghri, J. Power Sources 160 (2006) 1183e1194. [37] W.J. Kim, H.G. Choi, Y.K. Lee, J.D. Nam, S.M. Cho, C.H. Chung, J. Power Sources 163 (2006) 98e102. [38] B. Zhang, Y. Zhang, H. He, J. Li, Z. Yuan, C. Na, X. Liu, J. Power Sources 195 (2010) 7338e7348. [39] H. Kim, J. Power Sources 162 (2006) 1232e1235. [40] H. Kim, J. Oh, J. Kim, H. Chang, J. Power Sources 162 (2006) 497e501. [41] T. Shimizu, T. Momma, M. Mohamedi, T. Osaka, S. Sarangapani, J. Power Sources 137 (2004) 277e283. [42] W. Zhengn, A. Suominen, J. Kankaanranta, A. Tuominen, Chem. Eng. Sci. 76 (2012) 188e191. [43] C. Xu, A. Faghri, Int. J. Heat Mass Transf. 53 (2010) 1951e1966. [44] R. Chen, T.S. Zhao, W.W. Yang, C. Xu, J. Power Sources 175 (2008) 276e287. [45] Y.C. Park, S.H. Lee, S.K. Kim, S. Lim, D.H. Jung, K.B. Park, S.Y. Choi, J.H. Kim, D.H. Peck, Electrochim. Acta 56 (2011) 7602e7609.