Stack DMFC

Fuel cells can run on fuels other than hydrogen. In the direct methanol fuel cell (DMFC), a dilute methanol solution ( 3%) is fed directly into the anode, and a multistep process causes the liberation of protons and electrons together with conversion to water and carbon dioxide. Because no fuel processor is required, the system is conceptually vei"y attractive. However, the multistep process is understandably less rapid than the simpler hydrogen reaction, and this causes the direct methanol fuel cell stack to produce less power and to need more catalyst. 

Steady-state and Dynamic Operations of 3W DMFC stack... 

The transient response of DMFC is inherently slower and consequently the performance is worse than that of the hydrogen fuel cell, since the electrochemical oxidation kinetics of methanol are inherently slower due to intermediates formed during methanol oxidation [3]. Since the methanol solution should penetrate a diffusion layer toward the anode catalyst layer for oxidation, it is inevitable for the DMFC to experience the hi mass transport resistance. The carbon dioxide produced as the result of the oxidation reaction of methanol could also partly block the narrow flow path to be more difScult for the methanol to diflhise toward the catalyst. All these resistances and limitations can alter the cell characteristics and the power output when the cell is operated under variable load conditions. Especially when the DMFC stack is considered, the fluid dynamics inside the fuel cell stack is more complicated and so the transient stack performance could be more dependent of the variable load conditions. 

In this paper we report the effect of varying loads on a small size DMFC stack (10 cells with 9 cm active-area each). The transient responses of the stack voltage have been investigated upon variable current load conditions to obtain the information on the dynamic characteristics of the stack. Also, the transient responses of the stack current upon changing fuel flow rates have been monitored to obtain the optimal operating conditions for the staek. 

The DMFC used in this study was a 10-cell stack with 3cm X 3cm active area (total active... 

DMFC stack fabricated in this study, the minimal operating conditions are the flow rates of 3 ml/min and 2L/min for 2M methanol and air, respectively. 

TMs study has shown the dynamic behavior of a 5W DMFC stack when the current loads have changed by pulses and steps. In order to determine the optimum operating conditions of the stack, the dynamic behavior of the stack current has been studied under a constant voltage output of 3.8V, varying the flow rate of 2M methanol solution and air. For the stable operation of the 5W stack, the minimal fuel flow rates are found to be 3 ml/min and 2L/min for 2M methanol and air, respectively. 

In the United States, the Department of Defense (DOD) and the Department of Energy (DOE) promoted in 1992 the Defense Advanced Research Project Agency (DARPA) program to develop a DMFC for portable and mobile applications. Several institutions are involved (IFC, JPL, LANE, Giner, Inc.) and small stacks (up to 10 elementary cells) were built by IFC and JPL. The performances are quite encouraging, with power densities of 250 mW/cm at 0.5 V. More details are given in Section V.2. 

Figure 20. Performance characteristics of the International Fuel Cells 10-cell DMFC stack. (After Ref. 112 reproduced by permission of The Electrochemical Society, Inc.)...

The Jet Propulsion Laboratory and Giner Inc. have an on-going collaboration to develop electrochemical DMFC stacks. A 5-cell stack (with an active area of the electrode of 25 cm ) was designed and constructed for operation with unpressurized air. " The performance characteristics of the stack at two operating temperatures (60 and 90 °C) and two 1 M methanol flow rates (5 and 2 liter/min), are rather good 2 V at 250 mA/cm at 90 C. The variation in cell-to-cell performance was very small. Efforts are being made at several other laboratories (e.g., LANL, H-Power) to design, construct, and test DMFC stacks. 

In another study, Chen and Zhao [55] demonstrated that by using a Ni-Cr alloy metal foam as the cathode DL (and current collector), instead of a CFP or CC, the performance of a DMFC can be enhanced significantly due to the improvement of the mass transfer of oxygen and overall water removal on the cathode side. Fly and Brady [56] designed a fuel cell stack in which the distribution layers were made out of metal foams (open cell foams). In addition, more than one foam (with different porosity) could be sandwiched together in order to form a DL with variable porosity. 

The optimum amount of PTFE in the anode MPL depends on the operating conditions and design of the DMFC. Dohle et al. [176] used a 500 W DMFC stack and observed that anode MPLs with 13 wt% PTFE had the best performance. Peled et al. [177] designed anode MPLs that had around 20-40 wt% PTFE and determined that the layers with lowest PTFE content performed the best. 

Pt/Ru electrocatalysts are currently used in DMFC stacks of a few watts to a few kilowatts. The atomic ratio between Pt and Ru, the particle si2 e and the metal loading of carbon-supported anodes play a key role in their electrocatalytic behavior. Commercial electrocatalysts (e.g. from E-Tek) consist of 1 1 Pt/Ru catalysts dispersed on an electron-conducting substrate, for example carbon powder such as Vulcan XC72 (specific surface area of 200-250 m g ). However, fundamental studies carried out in our laboratory [13] showed that a 4 1 Pt/Ru ratio gives higher current and power densities (Figure 1.6). 

Most literature reports have addressed DM FC performance at the single cell level. More relevant for evaluating DMFCs as practical power sources is the performance obtained at the stack level, achieved under operating conditions appropriate for the complete power system to achieve acceptable energy conversion efficiency and with complete thermal and water balances. 

The results show that, at temperatures below 60 °C and an air feed stoichiometry below three, the cathode exhaust is fully saturated (nearly fully saturated at 60 °C) with water vapor and the exhaust remains saturated after passing through a condenser at a lower temperature. In order to maintain water balance, all of the liquid water and part of the water vapor in the cathode exhaust have to be recovered and returned to the anode side before the cathode exhaust is released to the atmosphere. Because of the low efficiency of a condenser operated with a small temperature gradient between the stack and the environment, a DMFC stack for portable power applications is preferably operated at a low air feed stoichiometry in order to maximize the efficiency of the balance of plant and thus the energy conversion efficiency for the complete DMFC power system. Thermal balance under given operating conditions was calculated here based on the demonstrated stack performance, mass balance and the amount of waste heat to be rejected. 

The amount of waste heat to be rejected from an operating stack constrains in a significant way the optimal DMFC stack operating conditions and therefore should be considered when designing the complete power system in order to minimize the size and weight and maximize the power and total energy conversion efficiency. 

As will be reported, as result of continuous efforts to optimize cell components and cell structure in developing DMFCs for portable power applications, we were able to demonstrate 30-cell DMFC stacks operated at 60 °C and fed with ambient air at 2-3 times faradaic stoichiometry, generating a power density of 320 W (active stack volume) during at least the initial week of testing in our laboratory. 

An overall DMFC stack energy conversion efficiency of 35% was achieved over a range of stack operating conditions of 0.46-0.57 V per cell. An extended life test over llOOh on a five-cell stack made of identical cell components and stack configurations was also performed. 

The newly assembled 30-cell stacks (one is shown in Figure 2.2) were immediately run in the DMFC mode at a fixed stack voltage at 60 °C with a 0.5 M methanol solution fed at the anode manifold and dry ambient air fed at the cathode manifold, without subjecting the stacks to any H2/air break-in conditions. The stacks gradually reached the reported levels of performance within a few hours and remained stable for at least the initial week of testing in our laboratory before they were sent to our partner for system integration. An extended life test over 1000 h on a five-cell stack built identically revealed stable stack performance. During the initial run, the... 

Figure 2.2 A photograph of the 30-cell DMFC stack shown together with a US Quarter for comparison of dimensions. More detailed stack properties and performance parameters for this current stack design are listed in Table 2.3.

Figure 2.3 Schematic diagram of the experimental setup for measuring amounts of liquid water and water vapor in the air cathode exhaust at a close point of cathode exit of an operating DMFC stack.

Figure 2.5 Amount of liquid water in the air cathode exhaust at cathode exit point of an operating DMFC stack as a function of stack operating temperature and air feed actual stoichiometry.

From the foregoing discussion, it is clear that, in a DMFC, the air cathode has to be operated under rather challenging conditions, that is, with a low air feed rate at nearly full water saturation. This type of operating conditions can easUy lead to cathode flooding and thus poor and unstable air cathode performance. To secure better air cathode performance, we have made great efforts to improve the ell cathode structure and cathode flow field design to facilitate uniform air distribution and easy water removal. The performance of our 30-cell DMFC stacks operated with dry air feed at low stoichiometry is reported in the following section. 

Three 30-cell DMFC stacks with 45 cm active electrode area were tested under various conditions to evaluate their performance for portable power applications. All three stacks showed nearly identical performance and the results from one stack that had gone through the most complete set of tests are presented here. 

Figure 2.9 Steady-state V-l curves for a 30-cell DMFC stack operated at various temperatures with a 0.5 M methanol solution feed at 125 ml min at the anode and with 0.76 atm dry air feed at 7.35 SLPM at the cathode. The steady state of the stack performance was verified by comparing the V—l curves with that of stack performance over 30 min for each given stack voltage listed in Table 2.2.

Table 2.3 DMFC stack properties and performance parameters when operated with dry air at 0.76 atm.

---