A Sensorless Methanol Concentration Controller Based on the Amplitude of Output Voltage for Direct Methanol Fuel Cells

There are steady interests in direct methanol fuel cells (DMFCs) as a clean energy technology for powering small and portable electronic devices because of several merits of DMFC systems such as high energy density, flexibility in power output from sub-watt to several hundred watts, and reliable operation, etc. Over the last decade, a number of worldwide efforts have been reported for improving the performance and durability of DMFCs. One of the key issues in DMFC systems is the methanol circulating unit that is required to reuse the exhaust methanol solution from DMFC stacks. Recirculation of the used methanol is accomplished by adjusting the concentration of the exhaust methanol solution to a pre-set level at a fuel mixing tank. In this fuel recirculation unit, electronic sensors are generally used to monitor and control the methanol concentration. However, the sensors are generally recognized to have many problems in terms of cost, size, durability and reliability. In order to solve these problems, there are resurgent research interests for the development of sensor-less control algorithms for DMFC systems. Several studies took advantages of methanol consumption equations or mathematical models to estimate the methanol concentration in the feed. Other studies presented the sensor-less controllers that used the fuel cell voltages and temperatures as the feedback parameters. The methods, however, suffer from the complexity in building database correlating methanol concentrations with other operating parameters such as output voltages, stack temperatures and methanol consumption rates and low degree of precision due to ever-changing performance of fuel cell electrodes.

We have found that there is a close relationship between the methanol concentration and the amplitude of voltage fluctuation of DMFC. The output voltage of DMFC experiences fluctuations during operation even under a constant electric load because of uneven reaction rates at the electrodes that are caused by mainly formation of CO₂ slugs at the anode and water flooding at the cathode. The CO₂ slugs and water flooding can clog the pores of the electrodes and retard the access of reactants to the catalytic sites, leading to a fluctuation of output voltage. Water flooding at the cathode is estimated to be more influential than the CO₂ slugs at the anode. In addition, the methanol concentration in the feed affects the methanol crossover rate from the anode to the cathode and thus influences water flooding at the cathode. Thereby the degree of voltage fluctuation is proportional to the methanol concentration.

This work presents a novel sensor-less control strategy for adjusting the methanol concentration in re-circulating direct methanol fuel cell (DMFC) systems by utilizing the amplitude of voltage fluctuation of DMFC as a feedback parameter. The effects of various operating conditions on the fluctuation amplitude are analyzed, and a new sensor-less algorithm is proposed based on the relationship between the methanol concentration in the feed and the amplitude of voltage fluctuation. The feasibility of the newly proposed algorithm is evaluated and compared with a control method using a methanol sensor in a continuous run of a 200 W-class DMFC system. It  is confirmed that the algorithm works very well and is able to control the methanol feed concentration within a small error bound. Moreover, this method enables steady operations of DMFC systems by minimizing the flooding phenomenon at the cathode. This sensor-less algorithm could be applied to any type of DMFC systems because it has an auto-tuning function.