Development of low cost polysulfone based anion exchange membranes and non-platinum oxygen reduction catalysts for fuel cell applications

Abstract

Proton exchange membrane fuel cells (PEMFCs) are currently based on high cost materials such as Nafion® membrane and Pt based catalysts. The high cost and limited abundance of noble metal hinder the commercialization of such fuel cells. For the future hydrogen economy, alkaline anion exchange membrane fuel cells (AAEMFCs) offer advantages of the potential use of non-Pt group metal catalysts, low-cost membranes (e.g. polysulfone based membranes) and cheaper bipolar plate (e.g. stainless steel). The research described in this thesis focused on the research and development of alternative anion conducting membranes and catalyst materials for fuel cells. Quaternary 1,4-diazabicyclo-[2.2.2]-octane (DABCO) polysulfone (QDPSU) was synthesized with different degree of substitution (DS) and characterized. The higher DS showed the better ionic conductivity; 0.015 S cm-1 for DS 58, 0.027 S cm-1 for DS 80 and 0.039 S cm-1 for DS 106 at 50 oC and 100 % relative humidity (RH). Based on the QDPSU, a thin PTFE-QDPSU composite membrane was prepared. Compared to the pristine QDPSU membrane, the composite membrane exhibited a better mechanical strength (32 MPa, maximum strength), less swelling and lower water uptake. The ionic conductivity of the composite membrane was 0.051 S cm-1 at 55 oC and 100 % RH. In fuel cell tests, power densities of 146 mW cm-2 and 103 mW cm-2 were achieved using oxygen and air, respectively. Severe degradation was found during preliminary experimental investigation on the KOH loaded polybenzimidazole (PBI) membrane including an ammonia smell came out the bottle and anode methanol solution turns yellow brown color in fuel cell tests. The QDPSU membrane was absorbed with phosphoric acid and tested in an intermediate temperature fuel cell. It was found that the higher DS, the higher membrane conductivity. When the DS reached 180 %, the QDPSU polymer cannot form ii a film with a suitable mechanical strength for the fuel cell application. A high power density of 400 mW cm-2 was achieved using DS106 of PA/QDPSU membrane at 150 oC and atmospheric pressure. Pd supported on carbons pre-treated in 5 % nitric acid, 0.07 M phosphoric acid, 0.2 M potassium hydroxide or 10 % hydrogen peroxide and evaluated in a half-cell. All Pd/C catalysts gave Tafel slopes close to 60 mV dec-1. Mass activities, measured at 0.025 V, for the 0.07 M H3PO4 and 0.2 M KOH treated carbon deposited with Pd were 6 mA mg-1Pd. Pre-treatments using 5 % HNO3 and 10 % H2O2 lead to an unfavorable effect on the morphology of Pd/C (metal particle agglomeration). Metal macrocycle based catalysts were examined for the ORR in alkaline media. FePc/KJB was found more active catalyst than the other metal macrocycles (CoPc, CoTMPP). The stability study in half-cell tests suggests that the FePc/KJB catalyst showed no degradation. The FePc/KJB was heat-treated under N2 atmosphere and 800 oC. The electrochemical behavior for ORR was characterized in half cell and single cell tests. The electron transfer number (n) of FePc/KJB-H8 was calculated to be 3.9 for ORR at -0.4 V. In AAEMFC, the peak power densities were 13.5 and 9.2 mW cm-2 for Pt/C and FePc/KJB-H8 under the same operating conditions, respectively. A direct methanol carbonate fuel cell using anion exchange materials and non-noble catalyst was demonstrated. The MEA performance using non-noble catalyst Acta 4020 was superior to the Pt/C based MEA. A maximum power density of 4.5 mW cm-2 was achieved at 50 oC using 6.0 M methanol and 2.0 M K2CO3. For the fuel cell stability study, the MEA exhibited a degradation rate of 2.52 μA cm-2 min-1.