Per-active site design of oxygen evolution catalysts

The team studied metal-hydroxide organic frameworks (MHOFs), a tunable class of hybrid materials. Operando spectroscopy revealed degradation mechanisms and identified active lattice-oxygen sites. These insights enabled defect engineering via linker substitution, enhancing stability and OER activity. Furthermore, the team showed that Cl⁻ ions suppress RuO₂ dissolution in acidic OER significantly.

Focus Group: Electrocatalysts Based on or Derived from Crystalline Coordination Networks such as Metal-organic Frameworks and Related Materials

Prof. Yang Shao-Horn (Massachusetts Institute of Technology), Alumna Hans Fischer Senior Fellow I Lena Schröck, Haiting Yu (TUM), Doctoral Candidates
Hosts: Bandarenka, Roland A. Fischer (TUM)

Figure 1

Fig. 1, Structures of the MHOFs constructed with a) terephalate (L1) and b) azobenzene-4,4’-dicarboxylate (L4) linkers. c), d) show the evolution of the polarization curves of the L1 and L4 based MHOFs, respectively, as a function of cycle number. e) and f) show the change in Raman spectra of the L1 and L4-based MHOFs, respectively, as a function of applied potential. Images from Zheng et al. (2023).
source: 2023 American Chemical
Society, Chem. Mater. 2023, 35, 13,
5017–5031.

Investigation and engineering of metal-hydroxide organic frameworks during oxygen evolution

The electrochemical generation of all value-added molecules such as hydrogen and hydrocarbons involves the oxygen evolution reaction (OER). However, the sluggish kinetics of OER can often limit the overall efficiency of such processes. For example, in electrochemical water splitting, more than 60% of the voltage losses can be attributed to the energy needed for OER.

Metal-organic frameworks have garnered significant attention in recent years as OER catalysts. [1] Previous work in our group has shown that Ni-based metal-hydroxide organic frameworks (MHOFs), [2] a subset of MOFs, are significantly more tunable than their pure metal hydroxide analogues.

In order to understand the electrochemical stability of these materials during OER, we studied the stability of MHOFs as a function of the π-π stacking energy of the linkers (Fig. 1a,b). [3] [4] Operando Raman spectroscopy showed all MHOFs transformed to NiOOH_2-x-like phases in KOH-based electrolytes during OER (Fig. 1c, d).

However, scanning transmission electron microscopy revealed that a core-shell structure can form if the π-π stacking energy is high, whereby a shell of NiOOH_2-x-like phases forms around a MHOF core. Through local coordination structure measurements from X-ray absorption spectroscopy (XAS) of the cycled materials, we found that approximately the first two to three unit cells transform to NiOOH_2-x-like phase. Further electrochemical analysis coupled with XAS, Raman, and UV-vis spectroscopy revealed that the leaching of linkers was correlated with the Ni²⁺ to Ni³⁺ transition upon an oxidizing potential (Fig. 1e, f). Through this work, we were able to identify the fundamental mechanism in which linkers are lost in MOFs in electrochemical conditions.

We also studied the lattice oxygen exchange dynamics of MHOFs constructed with linkers with varying π-π stacking energies to further understand the OER mechanism using ¹⁸O-labeled H₂O [5]. An interesting discovery was an inverse relationship of MHOF stability with OER activity, while the MHOF stability was directly correlated with lattice oxygen exchange capacity. Further operando spectroscopy concluded that there are two potential sites for lattice oxygen exchange: 1) catalytic sites on MOOH_2-x (Fig. 2a) and 2) sites on the pristine MHOFs (Fig. 2b). Therefore, the relationships we uncovered can be explained by the higher π-π stacking energy of the linkers preserving more of the MHOF phase during OER, leading to lower OER activity but higher lattice oxygen exchange capacity (Fig. 2c), showing that -OH exchange on Ni²⁺ appears to be more facile than -O exchange on Ni³⁺.

The key findings in the above studies showed that linker loss is correlated with Ni²⁺ to Ni³⁺ transition and that the oxygen sites in MHOFs can be active for oxygen exchange. We therefore engineered MHOFs by introducing ferrocene-based linkers that create defects and generate three unsaturated Ni²⁺ sites per linker (Fig. 3a) [6]. The defect MHOF exhibited a 25% higher TOF than the pristine material, attributed to the defects creating more OER active sites per metal via the unsaturated sites. Operando Raman further showed that upon increasing the potential to 1.3 V_RHE, the pristine MHOF
lost all linkers (again aligning with the Ni²⁺/Ni³⁺ transition), whereas the defect MHOF remained stable up to 1.7 V_RHE (Fig. 3c). Finally, a water electrolyzer using the defect MHOF operated at 1 A cm^-1, 1.75 V_RHE for over 120 hours.

As a further extension, MHOFs with complete missing linkers were synthesized and tested for OER activity, transformation velocity, and phase. Initial results from this approach showed that cyclic voltammetry measurements indicate slightly higher activity for the defect material in 0.1 M KOH electrolyte when evaluated at 1.6 V_RHE. In 1 M KOH, however, the activities of all three MOFs appear to converge after 150 cycles, with no pronounced differences observed. These findings suggest that the introduced defects exert a more significant impact in less concentrated electrolytes, indicating that the pH is an important external stimulus in MOF transformation.

Figure 2

Fig. 2, a) Schematic demonstrating the lattice OER mechanism. b) Illustration of the lattice oxygen sites in MHOFs and MOOH2-x that are active and inactive for exchange. c) Schematic demonstrating why lattice oxygen exchange was observed to increase with more stable MHOFs and inversely correlate with OER activity. Reproduced from Reference 5.
source: 2025 American
Chemical Society

Toward understanding chloride-ion–induced enhancement of OER stability in RuO₂ catalysts

Currently, the large-scale deployment of PEM electrolyzers is primarily constrained by the scarcity of iridium (Ir). Ruthenium (Ru), being more earth-abundant, represents a promising alternative; however, it shows poor stability. Yet RuO₂-based dimensionally stable anodes (DSAs) can be operated at similar potentials as OER for decades in the chloro-alkaline industry. Inspired by these DSAs, we investigated the role of Cl^- ions in stabilizing RuO₂.

Using ICP-MS, EC-MS, and FTIR, we systematically investigated how Cl^- influences RuO₂ stability. We found that introducing small concentrations of Cl^- (e.g., 25 mM) significantly suppressed RuO₂ dissolution during acidic OER by a factor of six while maintaining Faradaic efficiencies above 95% to OER. Raising the Cl^- ion concentration to 100 mM further increased stability, with OER FE still exceeding 67%. The stabilizing trend was confirmed at elevated temperatures (60 °C), which boosted the dissolution rate but did not change the
relative dissolution rates as a function of Cl^-concentration. Furthermore, we constructed full electrolyzer cells to test chloride-induced stabilization at industrially relevant current densities and conditions. (Summary of results in Fig. 4)

This project is a joint effort between MIT and TUM within the framework of the TUM Institute for Advanced Study, in close collaboration with Daniel J. Zheng, Mikaela Görlin, Yuriy Román-Leshkov (all MIT) and Johannes Sterzinger (MIT/TUM). Other collaborators include Jan Rossmeisl’s group (University of Copenhagen) for theory, Yang Ha (UC Berkeley) for XAS, and Hubert Gasteiger’s group at TUM. Our results indicate that both the stability enhancement and the formation of Cl- containing surface species at high potentials exhibit strong pH dependence. These findings motivate future, detailed investigations into the specific surface intermediates and mechanistic pathways responsible for the improved stability of RuO₂ under OER conditions.

Figure 3

Fig. 3, Structures of the MHOFs constructed with a) terephalate (L1) and b) azobenzene-4,4’-dicarboxylate (L4) linkers. c), d) show the evolution of the polarization curves of the L1 and L4 based MHOFs, respectively, as a function of cycle number. e) and f) show the change in Raman spectra of the L1 and L4-based MHOFs, respectively, as a function of applied potential. Images from Zheng et al. (2023).

Figure 4

Fig. 4, Dissolution of RuO2 in 0.1 M HClO4 with varying concentrations of KCl/HCl added.

[1]
J. Du, F. Li, and L. Sun, “Metal–organic frameworks and their derivatives as electrocatalysts for the oxygen evolution reaction,” Chemical Society Reviews, vol. 50, no. 4, pp. 2663–2695, Jan. 2021, doi: 10.1039/d0cs01191f

[2]
S. Yuan et al., “Tunable metal hydroxide–organic frameworks for catalysing oxygen evolution,” Nature Materials, vol. 21, no. 6, pp. 673–680, Feb. 2022, doi: 10.1038/s41563-022-01199-0

[3]
X. Ma et al. (2023)

[4]
D. J. Zheng et al. (2025)

[5]
H. Xu et al. (2025)

[6]
J. Sterzinger, D.J. Zheng, H.H. Kristoffersen, H. Iriawan, J. Rossmeisl, A. Bandarenka, and Y. Shao-Horn, in preparation

Selected publications

D. J. Zheng et al., “Linker-Dependent stability of Metal-Hydroxide organic frameworks for oxygen evolution,” Chemistry of Materials, vol. 35, no. 13, pp. 5017–5031, Jun. 2023, doi: 10.1021/acs.chemmater.3c00316.
X. Ma et al., “Structure–Activity relationships in NI- Carboxylate-Type Metal–Organic frameworks’ metamorphosis for the oxygen evolution Reaction,” ACS Catalysis, vol. 13, no. 11, pp. 7587–7596, May 2023, doi: 10.1021/acscatal.3c00625.
X. Ma et al., “Tuning the Reconstruction of Metal–Organic Frameworks during the Oxygen Evolution Reaction,” ACS Catalysis, vol. 14, no. 21, pp. 15916–15926, Oct. 2024, doi: 10.1021/acscatal.4c03618.
D. J. Zheng et al., “Lattice oxygen exchange pathways in Nickel–Iron Metal–Organic Framework-Based Oxygen evolution electrocatalysts,” ACS Applied Materials & Interfaces, Dec. 2025, doi: 10.1021/acsami.5c12947.
H. Xu et al., “Stable Metal–Organic Electrocatalysts for Anion-Exchange membrane water electrolyzers by defect engineering,” Journal of the American Chemical Society, vol. 147, no. 33, pp. 29838–29851, Aug. 2025, doi: 10.1021/jacs.5c06156.
J. Sterzinger, D.J. Zheng, H.H. Kristoffersen, H. Iriawan, J. Rossmeisl, A. Bandarenka, and Y. Shao-Horn, “Stabilization of RuO2 with Cl- ion for Acidic Oxygen Evolution,” in preparation.

Full list of publications