Overall and local effects of operating conditions in PEM fuel cells with dead-ended anode

https://doi.org/10.1016/j.ijhydene.2016.08.091Get rights and content

Highlights

  • Local current densities change very differently at different locations.

  • Impurities accumulate at the channel end first and progress gradually upstream.

  • Operating conditions have significant effects on water and nitrogen transfer.

  • The concentration gradient of impurities along the channel can be very steep.

Abstract

Nitrogen and water accumulations in fuel cells with dead-ended anode can cause severe cell performance decline and fluctuations. In this work, both overall and local effects of fuel cell operating parameters, i.e., cathode humidity, air stoichiometry, hydrogen pressure and operating current density, have been experimentally studied under galvanostatic mode. A purge at the anode is automatically triggered when the cell voltage has decreased by 0.1 V and the mean purge interval, defined as the average time between two purges, is recorded as a characteristic parameter. Local current densities are measured to study the local effects and detailed local characteristics of the fuel cell. The experimental results show that mean purge intervals decrease with cathode inlet humidity and operating current density, and increase with inlet hydrogen pressure and air stoichiometry. The experimental results also show that the local current densities change very differently at different locations and impurities first accumulate near the end of the anode channel and then gradually progress upstream.

Introduction

Proton exchange membrane fuel cells (PEMFCs) have received much attention because of their high efficiency and zero emission. However, their cost and durability are still the main challenges to their commercialization [1]. Commonly-used PEM fuel cell system is usually equipped with an anode recirculation system. In such a system, hydrogen is supplied in excess and the non-reacted hydrogen is recirculated back to the fuel cell inlet by a hydrogen recirculation blower or an ejector. Although such a system can have high fuel utilization, the recirculation devices increase system complexity and lead to higher cost, especially for portable applications. To simplify the anode supply system, the dead-ended anode (DEA) configuration has been used [2], [3], [4]. During DEA operation, dry hydrogen is supplied through a pressure regulator installed at the fuel cell inlet and a normally-closed solenoid valve is installed at the outlet. Without the recirculation devices, the fuel cell system cost, volume and weight are all reduced [5], [6], [7], [8], [9], [10], [11]. However, during DEA operation, impurities (liquid water and nitrogen gas) can transfer through the membrane from the cathode to the anode and their accumulation may block hydrogen gas from reaching the catalyst, leading to local hydrogen starvation and decrease in cell performance [12], [13], [14]. Furthermore, hydrogen starvation can result in corrosion of the carbon support and thus lead to irreversible cell degradations [15], [16], [17]. To prevent the performance decline and irreversible degradation during DEA operation, anode purge is commonly used. Anode purges are usually accomplished by opening the solenoid valve at the anode outlet periodically. During anode purges, the accumulated water and nitrogen are swept out of the anode channel by fresh dry hydrogen. The purge frequency depends on the rates of water and nitrogen accumulation in the anode and the allowable amount of decrease in cell output.

It is well known that operating conditions have significant effects on water and nitrogen accumulation in the anode, and in recent years, significant efforts have been devoted to this issue both at the single cell level [5], [8], [10], [14], [18] and at the stack level [11], [19]. Chen et al. [5] systematically investigated the operation characteristics of a single fuel cell with DEA under various operating parameters and their results showed that the cell performance decrease mainly resulted from water accumulation since pure oxygen was used in the cathode. Himanen et al. [10] studied the effect of hydrogen pressure and found that higher hydrogen pressure could reduce anode side flooding. Nikiforow et al. [8] investigated the impact of cathode humidity and their results showed that higher cathode humidity led to more water accumulation in the anode. Lee et al. [14] observed water accumulation in the anode gas channel under various operating parameters with a transparent single fuel cell. Their experimental results showed that water accumulation decreased as air stoichiometry increased. Siegel et al. [18] investigated water accumulation in the anode using neutron radiography technology and their results showed that water accumulation increased with operating current density.

The effects of operating conditions on fuel cell stacks with DEA have also been investigated. Dumercy et al. [11] investigated the effect of operating current density in a 3-cell stack with DEA and found that there was not apparent decrease in stack voltage when the operating current density was lower than 400 mA cm−2. Sasmito et al. [19] studied the effect of air stoichiometry on the purging process with a 24-cell PEM fuel cell stack at a current density of 300 cm2 and their results showed that nitrogen crossover from the cathode to the anode increased as air stoichiometry increased. Recently, Hu et al. [20] measured liquid water and nitrogen concentration in the anode exhaust of DEA fuel cell stacks and proposed optimized anode purge strategies.

During a DEA operation, water and nitrogen accumulation near the anode outlet can result in very uneven local performance [15], [21], [22]. Yu et al. [15] explored the behaviors of PEMFCs with a dead-ended anode by detecting current distribution and the local potentials and found that the local current density near the outlet decreased gradually during DEA operation. Manokaran et al. [22] measured the spatial-temporal evolution of the local current densities and their results showed that higher current density and thinner membrane led to more serious cell performance fluctuations. Abbou et al. [21], [23] investigated local degradation phenomena in a fuel cell during DEA operation, and their results showed that the areas near the outlet suffered more serious degradation compared with the areas near the inlet.

It is obvious from the above that previous studies mainly focused on the overall performance of fuel cells or stacks with DEA operations. Some limited studies on local current density are very general. However, the detailed effects of operating conditions on cell performance and its local characteristics with a DEA are still not fully understood. Hence, in this work, an experimental study is conducted to systematically investigate the effects of various key operation parameters, i.e., cathode humidity, air stoichiometry, hydrogen pressure and operating current density both at the overall and local levels in a DEA fuel cell. Besides, the evolutions of local current densities at different locations along the anode flow channel are carefully studied to understand how the impurities accumulate and their effects on both cell performance and local characteristics.

Section snippets

Experimental system

Fig. 1(a) shows a schematic of the experimental system. Hydrogen (purity > 99.99%) is supplied at the anode and air is supplied at the cathode. A fuel cell test station is used to control air flow rate and air relative humidity at cathode. Cell temperature and operating current density are controlled by the test station. At the anode, a pressure regulator is installed at the inlet to control the gas pressure in the anode channel. A normally-closed solenoid valve is installed at the outlet of

DEA operation

The experimental fuel cell is operated at a constant current density of 500 mA cm−2 in DEA operation and the cell voltage and current distribution are measured. Fig. 2(a) shows the typical time evolutions of fuel cell voltage and local current densities between two successive purge events. Local current distributions along the flow direction at different time are presented in Fig. 2(b). In Fig. 2, two anode purge events are conducted at 0 min and 33 min respectively. It can be seen that the

Conclusion

Effects of operating conditions on both overall and local performances of PEM fuel cells with dead-ended anode are systematically studied. Based on the experiments results, following conclusions can be made:

  • Operating conditions have significant effects on water and nitrogen transfer in PEM fuel cells with a dead-ended anode.

  • Local current densities change differently at different locations during impurity accumulations.

  • Impurities accumulate first at the end of the anode channel and progress

Acknowledgements

The financial supports from the National Science Foundation of China for Creative Research Groups (No. 51121092) are gratefully acknowledged.

References (30)

Cited by (0)

1

Fax: +86 29 8266 9033.

View full text