Covalent Organic Frameworks Based on Co-Porphyrin for Efficient O2 Reduction

In the electrochemical conversion (EC) process, the oxygen reduction reaction (ORR) is critical for energy storage and conversion. However, this reaction typically requires expensive platinum group metal (PGM) catalysts, severely limiting the commercialization of fuel cell vehicles. Currently, chemists have adopted several strategies to reduce the cost of catalysts, including metal-free carbon materials (MFCMs), single-atom electrocatalysts (SAECs), and metal alloys. Among these, single-atom electrocatalysts offer extremely high atomic utilization, catalytic activity, higher selectivity, and ease of integration into electrodes, presenting significant advantages. To design ideal SAECs, two factors need to be addressed: (1) control of appropriate coordination sites; (2) suitable material frameworks that can participate in effective electronic coupling with active sites.

Reticulated chemistry provides the opportunity to precisely adjust the distance between atomic sites using various ligands, which has a significant impact on electrodynamics. Covalent organic frameworks (COFs) possess high porosity and defined chemical structures, with broad applications in gas separation/storage, catalysis, and energy storage. The conjugated structure of COFs can enhance charge transport within the crystal and allow for the selection of suitable monomers as spacers to control the distance between atomic sites. Since 2015, redox-active porphyrin-based COFs have shone in various catalytic processes as carriers for SAECs, including CO₂ reduction reactions (CO₂RR), hydrogen evolution reactions, and oxygen evolution reactions.

Here, Professor Luis Echegoyen and colleagues from the University of Texas at El Paso reported two covalent organic frameworks (COFs) based on Co(II)-porphyrin/[2,1,3]-benzothiadiazole (BTD), Co@rhm-PorBTD and Co@sql-PorBTD, and utilized them for O₂ electrocatalytic reduction (ORR). These two cobalt-based SAECs exhibit excellent electrocatalytic efficiency for ORR. Due to its high electron affinity, BTD acts as a spacer group between the metal porphyrin components, effectively facilitating intermolecular charge transfer.

Figure 1. (a) Electrostatic potential (ESP) maps of various monomers; (b) Synthetic schemes of Co@rhm-PorBTD (left) and Co@sql-PorBTD (right).(Image source: ACS Nano)

As shown in Figure 1, by optimizing the amount of catalyst (acetic acid) and the types of mixed solvents, two covalent organic frameworks (COFs) based on Co(II)-porphyrin/[2,1,3]-benzothiadiazole (BTD), Co@rhm-PorBTD and Co@sql-PorBTD, were successfully synthesized, and the chemical structures of the materials were characterized by Fourier-transform infrared spectroscopy (FT-IR) and solid-state nuclear magnetic carbon spectroscopy (¹³CP-MAS).

Figure 2. (a, b) PXRD patterns and simulated structure diagrams of Co@rhm-PorBTD; (c, d) PXRD patterns and simulated structure diagrams of Co@sql-PorBTD; (e-f) HR-TEM images of Co@rhm-PorBTD; (g-h) HR-TEM images of Co@sql-PorBTD; (i) Elemental mapping images of Co@rhm-PorBTD and Co@sql-PorBTD; (j) Co 2p XPS spectrum of Co@rhm-PorBTD and (k) Co 2p XPS spectrum of Co@sql-PorBTD.(Image source: ACS Nano)

Powder X-ray diffraction (PXRD) patterns indicate that Co@rhm-PorBTD and Co@sql-PorBTD exhibit good crystallinity. High-resolution transmission electron microscopy (HR-TEM) images show the formation of ordered structures in both COFs. Energy dispersive X-ray spectroscopy (EDS) mapping analysis (Figure 2i) indicates a uniform distribution of carbon (C), nitrogen (N), sulfur (S), and cobalt (Co). X-ray photoelectron spectroscopy (XPS) results indicate that the coordination metal centers of the COFs are in the +2 state (Figure 2j, k).

Figure 3. (a) CV curves of Co@rhm-PorBTD and (b) Co@sql-PorBTD in saturated Ar (Ar) and O₂ conditions in 0.1 M KOH solution; (c) ORR polarization curves at an electrode rotation speed of 0 rpm and a scan rate of 10 mV s^–1; (d) ORR polarization curves at an electrode rotation speed of 1600 rpm and a scan rate of 10 mV s^–1; (e) Tafel plots of Co@rhm-PorBTD, Co@sql-PorBTD, and Pt/C (20%); (f) Comparison of the current density and half-wave potential of Co@rhm-PorBTD, Co@sql-PorBTD, and Pt/C (20%); (g) Comparison of ECSA, overpotential at half-wave potential (1.23-E1/2), specific activity (0.85 V vs RHE), and roughness; (h) Comparison of mass activities at different potentials.(Image source: ACS Nano)

Under O₂ conditions, reduction peaks for Co@rhm-PorBTD and Co@sql-PorBTD were observed (Figure 3a, b). The peak positions of Co@rhm-PorBTD show a slight positive shift (32 mV) compared to Co@sql-PorBTD. The reduction current of 20% Pt/C was 140μA, while the reduction currents of Co@rhm-PorBTD and Co@sql-PorBTD were 392 and 327μA, respectively, nearly twice that of the commercial electrode. LSV curves for both COFs were recorded under static and dynamic conditions (Figure 3c, d). The onset potentials of Co@rhm-PorBTD and Co@sql-PorBTD were approximately 0.89 and 0.85 V (vs RHE), with half-wave potentials of 0.83 and 0.75 V (vs RHE), respectively, while the values for Pt/C (20%) were slightly higher compared to those for COFs. The Tafel slopes of Co@rhm-PorBTD and Co@sql-PorBTD were 41 and 87 mV/dec, respectively (Figure 3e), much lower than that of 20% Pt/C (90 mv/dec), indicating that the COF-based SAC exhibits faster ORR kinetics than Pt/C (20%). The mass current density of Co@rhm-PorBTD was 38.7 mA/cm², slightly lower than that of Pt/C (20%) (41 mA/cm²), while Co@sql-PorBTD showed a current density of 15.9 mA/cm² (Figure 3f). A comprehensive comparison of the two synthesized COFs with commercial Pt/C (20%) was conducted (Figure 3g). At 0.9 V, the mass activities of Co@rhm-PorBTD and Co@sql-PorBTD were 458 A/g and 333 A/g, respectively, nearly 4.20 times and 1.7 times that of commercial Pt/C (20%) (Figure 3h), while at 0.85 V, the mass activities were 5.8 times and 1.3 times that of Pt/C, respectively. The SA values of Co@rhm-PorBTD (1.32 mA cm^–2) and Co@sql-PorBTD (0.324 mA cm^–2) were 10 times and 2.5 times higher than that of Pt/C (0.142 mA cm^–2). These activity values exceed the target values set by the U.S. Department of Energy for 2020.

Figure 4. (a) Schematic diagram of the homemade zinc-air battery; (b) Open circuit diagram of the zinc-air battery using Co@rhm-PorBTD and Co@sql-PorBTD as catalysts; (c) Discharge curves of the zinc-air battery using Co@rhm-PorBTD and Co@sql-PorBTD at a current density of 5 mA cm^-2; (d) Discharge and charge cycle curves; (e) Comparison of polarization and power density curves of zinc-air batteries using Pt/C (20%), Co@rhm-PorBTD, and Co@sql-PorBTD as cathodes; (f) Long-term discharge/charge cycling performance of zinc-air batteries containing Co@rhm-PorBTD, Co@sql-PorBTD, and Pt/C (20%) at a current density of 5 mA cm^-2.(Image source: ACS Nano)

COFs were coated on carbon cloth and assembled into batteries using zinc plates as anode (Figure 4a). The cathodes based on Co@rhm-PorBTD and Co@sql-PorBTD exhibited stable open-circuit voltages of 1.53 V and 1.48 V, respectively (Figure 4b). During constant current discharge at 5 mA/cm², no significant voltage drop was observed for both COFs over 25 hours (Figure 4c). The charge and discharge voltages of Co@rhm-PorBTD were 1.99 and 1.22 V, respectively, while those of Co@sql-PorBTD were 1.99 and 1.18 V, respectively (Figure 4d). The peak power densities of Co@rhm-PorBTD and Co@sql-PorBTD were 138 and 117 mW cm^–2, greater than that of commercial Pt/C (20%) (96 mW cm^–2) (Figure 4e). Long-term constant current charging/discharging cycles showed no significant performance loss over 1000 hours. In contrast, the performance of Pt/C (20%) significantly decreased after 100 cycles (Figure 4f), indicating that the COF-based air cathodes exhibit better stability than commercial Pt/C (20%).

Figure 5. (a) Non-in-situ and in-situ XANES characterizations of Co@rhm-PorBTD and (b) Co@sql-PorBTD; (c) Total density of states and predicted density of states in Co@rhm-PorBTD and (d) Co@sql-PorBTD; (e) Top view and (f) side view of optimized Co@rhm-PorBTD monolayer; (g) Top view and (h) side view of optimized Co@sql-PorBTD monolayer; (i) Free energy diagrams of the two COFs; (j) Proposed ORR catalytic scheme.(Image source: ACS Nano)