A neural-network-like catalyst structure for the oxygen reduction reaction: carbon nanotube bridged hollow PtCo alloy nanoparticles in a MOF-like matrix for energy technologies

The rational design of a catalytic layer in a membrane-electrode assembly is the key to achieve high performances from proton exchange membrane fuel cells (PEMFCs). Herein, inspired by the neural-network structure of the brain, we constructed a bionic catalytic network for the oxygen reduction reaction (ORR), via setting up Pt-organic ligands-Co 2+ -organic ligands-Pt connections and then thermally transforming them into a metal-organic-framework (MOF)-like matrix in which hollow PtCo alloy nanoparticles (NPs) with an average particle size of 4.4 nm are bridged together by carbon nanotubes (PtCo@CNTs-MOF). The bionic catalytic network provides highly efficient linkages of various species-transport channels to active sites; as a result, an order of magnitude improvement is achieved in mass transfer efficiency as compared to the traditional Pt/C catalytic layer. Besides, the hollow PtCo alloy derived from Pt NPs shows a high initial mass activity of 852 mA mg Pt −1 @ 0.90 V and an undetectable decay in an accelerated aging test. Accordingly, a remarkable Pt utilization efficiency of 58 mg Pt kW −1 in the fuel cell cathode and 98 mg Pt kW −1 in both the anode and cathode was eventually achieved, respectively. The latter is almost 3 times higher than that of the traditional catalytic layer. Moreover, no decay was detected during continuous operation at 1 A cm −2 for 130 hours from the bionic catalytic network based fuel cell. This strategy offers a new concept for designing an ultra-low Pt loading yet highly active and durable catalytic layer for fuel cell applications and beyond. Inspired by the brain neural systems, an ideal multi-scale model for high level utilization of constitutional unit, a neural-network-like catalyst structure was produced with a remarkable Pt utilization efficiency of 98 mg Pt kW −1 in the fuel cell.