The Role of Surface Oxophilicity in Copper-catalyzed Water Dissociation

Hydrogen evolution reaction; bimetallic catalysts; renewable energy; ... energy (HBE) over different monometallic surfaces in alkaline conditions via ...
3 downloads 0 Views 902KB Size
Download PDF
Subscriber access provided by ST FRANCIS XAVIER UNIV

Article

The Role of Surface Oxophilicity in Copper-catalyzed Water Dissociation Wesley W Luc, Zhao Jiang, Jingguang G Chen, and Feng Jiao ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b01710 • Publication Date (Web): 27 Aug 2018 Downloaded from http://pubs.acs.org on August 27, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

The Role of Surface Oxophilicity in Copper-catalyzed Water Dissociation Wesley Luc,1 Zhao Jiang,2,3 Jingguang G. Chen,*,3 and Feng Jiao*,1 [1] Center for Catalytic Science and Technology, Department of Chemical and Biomolecular Engineering, University of Delaware, Newark, DE 19716 USA [2] Xi’an Jiaotong University, Xi’an, Shaanxi, 710049 China [3] Department of Chemical Engineering, Columbia University, New York, NY 10027 USA *Corresponding authors: [email protected] & [email protected] Abstract Understanding water dissociation on non-precious metal surfaces is essential toward designing efficient catalysts for important chemical reactions such as water-splitting and carbon dioxide reduction. Hydrogen binding energy (HBE) has been proposed to be a good descriptor to predict hydrogen evolution reaction (HER) activity in alkaline electrolytes. In this work, we showed that although HBE values could predict the HER activity reasonably well, the oxophilicity of the catalyst surface also played a significant role in water dissociation. To elucidate the role of surface oxophilicity, a series of non-precious copper-based bimetallic materials were systematically studied for HER activity in alkaline conditions. By alloying copper with small amounts of oxophilic transition metals such as Ti, Co, or Ni, a significant enhancement in HER activity was achieved in comparison to pure copper. However, if the HBE was considered as the sole descriptor, the experimentally measured HER activity trend did not match the theoretical trend as predicted by density functional theory (DFT) calculations. Further studies combining both computational efforts and experimental investigation of metal oxide/hydroxide (MO/OH) clusters deposited on Cu surfaces showed that oxygen binding energies (i.e., the oxophilicity of the dopant metal) together with HBEs should be used as descriptors to predict HER activity in alkaline conditions due to the synergistic interactions between copper and the oxophilic metal. Keywords Hydrogen evolution reaction; bimetallic catalysts; renewable energy; electrocatalysis; copper

1 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Introduction Water dissociation is one of the most important chemical reactions because it provides a viable way to produce protons and electrons that are required for fuel and chemical productions at large scales. For example, the reduction of protons using electrons generated from water dissociation can produce hydrogen gas (H2) that can serve as an alternative carbon-free energy carrier in place of petroleum-based fuels.1-2 Alternatively, the protons and electrons can also be used to convert carbon dioxide (CO2) into valuable chemicals through CO2 electrolysis.3-4 Regardless of the targeted application, a good understanding of the key descriptor(s) for water dissociation is essential for the rational design of efficient catalysts. Recent studies, combing both density functional theory (DFT) and experimental efforts, have correlated the hydrogen evolution reaction (HER) activities in terms of exchange current density with hydrogen binding energy (HBE) over different monometallic surfaces in alkaline conditions via a volcano-type correlation with Pt residing nearing the apex.5-8 Stemming from these correlations, it has been suggested that the HBE can be tuned by alloying two different metals such that an optimal HBE can be obtained, and more importantly, this strategy provides a platform for the rational design of efficient catalysts with desired catalytic properties.9-12 Copper (Cu)-based materials are widely studied as potential catalysts for water-splitting and CO2 reduction.13-18 In the case of water-splitting, a catalyst surface that highly favors HER activity is desirable; however, in the case of CO2 reduction, the suppression of HER activity is necessary to maximize CO2 reduction Faradaic efficiency. As such, the ability to predict and control HER activity of Cu-based catalysts can enable one to design more efficient catalysts for both reactions. Recently, we reported a Cu-Ti bimetallic electrocatalyst that was able to reduce water to H2 under a mild overpotential at a rate orders of magnitude greater than that of pure Cu catalyst.19 DFT calculations showed that the combination of Cu and Ti created unique Cu-Cu-Ti hollow sites, exhibiting very similar HBE values to that of Pt. Although the findings clearly showed that HBE is a good descriptor for HER activity under alkaline conditions, the question whether HBE is the sole descriptor remains unclear and further investigation is required to answer this fundamental question. Herein, we synthesized a series of bulk copper-based bimetallics (Cu-M) using an arc melting process, where the HBEs of active sites were tuned using a range of metal dopants, such as Ti, Co, and Ni. The experimental results showed that the Cu-M bimetallics exhibited much better HER activities than pure Cu with a trend of Cu-Ti > Cu-Co > Cu-Ni > Cu. However, when the HBE was used as the sole descriptor, the observed experimental trend did not follow that as predicted by DFT calculations. Further studies combining both computational efforts and experimental investigation of metal oxide/hydroxide (MO/OH) clusters deposited on Cu surfaces suggested that the oxophilicity of the secondary metal can assist water dissociation through weak interactions with the oxygen atom of the water molecule. Therefore, although the HBE is still the dominant descriptor for HER in alkaline conditions, the oxygen binding energy (OBE) of the dopant metal should also be considered as a descriptor to characterize HER activity for Cu-M bimetallics. Additionally, the HER results on Cu-M bimetallics suggested that incorporating an highly oxophilic metal into Cu may not be a valid strategy to improve CO2 reduction performance on Cu-based catalysts due to the promotion of the undesired HER.

2 ACS Paragon Plus Environment

Page 2 of 13

Page 3 of 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Results Hydrogen Binding Energies on Cu-M Bimetallics In our previous work, it was shown that doping Cu with low concentrations of Ti created highly active sites for HER in alkaline conditions, possessing similar HBEs as Pt.19 To reveal the HER activity trend of Cu-M bimetallics, computational modeling were conducted to screen other first-row transition metals as dopants. A 3×3 Cu(111) surface with either one or two surface Cu atoms replaced with Ti, Co, or Ni atoms, corresponding to a surface concentration of 11.1% or 22.2%, respectively, was used as the model system as shown in Figure 1. The HBE was calculated for five types of adsorption sites (Tables 1 and S1): three Cu-Cu-M and two Cu-M-M hollow sites where M = Ti, Co, or Ni. For comparison, HBE values for pure Cu(111) and Pt(111) (Table S2) were also determined. In addition, a sub-layer substitution surface and a metal oxide cluster on Cu(111) surface were also investigated (Figures S1 and S2). Lastly, OBE values were also calculated with the same model surface (Tables 1, S1-S4). Although pure Cu is a poor HER catalyst as reflected by its relatively weak HBE values, computational predictions showed that doping a small amount of Ni, Co, or Ti can dramatically shift the HBE toward that of Pt. To facilitate the discussion, three HBE values and three OBE values (two for hollow Cu-Cu-M sites and one for hollow Cu-M-M sites) of each Cu-M system are shown in Table 1. The remaining calculated binding energies are tabulated in Table S2. Based on the calculations, Cu-Cu-M hollow sites exhibited HBE values very close to Pt’s, in which Cu-Ti(111) sites exhibited the closest HBE values while Cu-Co(111) sites exhibited slight stronger HBEs. On the contrary, Cu-M-M hollow sites bind hydrogen too strongly as reflected by its large negative HBE values. In the cases of the sub-layer substitution and metal oxide cluster on Cu(111) surfaces, the adsorption sites exhibited weak HBEs with similar values to those of pure Cu(111) surface, suggesting that these types of surfaces are not favorable for HER (Tables S3 and S4). Ideally, a 3×3 surface Cu unit cell with the middle Cu atom replaced with a dopant atom would be an ideal catalytic surface such that the surface would maximize the number of Cu-Cu-M sites without introducing undesired Cu-M-M sites. Based on the calculated HBEs, the predicted HER activity of Cu-M bimetallics and pure Cu should follow the trend of Cu-Ti > Cu-Ni > Cu-Co > Cu.

Figure 1. DFT models of a) Cu(111) surface with one Cu atom replaced with a M atom and b) Cu(111) surface with two Cu atoms replaced with two M atoms. Cu atoms are represented in orange. M = Ti, Co, or Ni atoms are represented in blue. 3 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table 1. Calculated HBEs and OBEs for Cu-M(111) and Cu(111) surfaces

[#]: Corresponding adsorption site on Figure 1. See Table S2 for binding energies over the remaining surface sites.

Experimental Validation using Model Catalysts To verify the computational predictions, experimental investigation began with the synthesis of bulk Cu-M bimetallic alloy ingots by melting desired amounts of dopant metals (Ti, Co, or Ni) with Cu to achieve the desired bulk atomic ratios of 5-10% via arc melting in vacuum. The as-synthesized ingots were then machined down to rotating disk electrode (RDE) disks, as shown in Figure S3. Figure 2 shows the powder X-ray diffraction (PXRD) patterns of the final state of each sample after arc melting, machining, and subsequent heating and quenching. In all cases, solid solution mixtures were achieved, as indicated by the clear FCC peaks and expected peak shifts due to alloy formation (Figure 2b and Table S6). The surface composition of each bimetallic was determined using X-ray photoelectron spectroscopy (XPS). In the case of Cu-Ti, a dopant atomic ratio of 5% was chosen such that that the surface compositions of all alloys were close to consistent at ~11%. The discrepancy between the surface and bulk compositions is likely due to the larger Ti atoms tending to aggregate toward the surface to minimize the overall lattice strain as consistent with our previous work.19-20 Furthermore, XPS analysis indicated that majority of the surface Cu atoms in all Cu-M materials were metallic Cu0, whereas majority of surface Ti, Co, and Ni atoms were in their oxidized Ti4+, Co2+, and Ni2+ states, respectively (Figure S4).21-22 We note that the exact surface nature of the bimetallic catalysts under HER conditions remains unclear due to lack of developed in-situ techniques that can resolve the valence state of surface atoms in presence of electrolyte under reaction conditions. The bulk compositions, measured with energy-dispersive X-ray spectroscopy (EDS), were consistent with the expect nominal values as reported in Table 2. The roughness factors (Rf) for all Cu-M surfaces were nearly identical as determined by atomic force microscopy (AFM). Typical AFM images and line scans of a Cu-M bimetallic surface are shown in Figure S5.

4 ACS Paragon Plus Environment

Page 4 of 13

Page 5 of 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 2. XRD characterization. a) The XRD patterns of Cu-M bimetallics and Cu and b) the enlarged region of Cu(111) diffraction peaks. Table 2. Bulk and surface compositions of Cu-M bimetallics and Cu

The as-synthesized Cu-M bimetallics were evaluated for HER activity in H2 saturated 0.1 M potassium hydroxide (KOH) using a three-electrode rotating disk electrode setup with the working electrode rotating at 1800 rpm. A high purity graphite rod was used as the counter electrode in all the experiments to avoid any potential contamination issues.23 Figure 3a shows the HER polarization curves of the current normalized to the geometric surface area versus the applied potential after iR-correction. In addition, Tafel slopes (b in η = b log|i| + a) were obtained from the polarization curves between -1 mA cm-2 and -8 mA cm-2 as shown in Figure 3b, while the exchange current densities at the reversible potential were determined by extrapolating the Tafel plots to 0 V vs the reversible hydrogen electrode (RHE). The polarization curves showed that alloying Cu with Ti, Co, or Ni dramatically improved the HER performance in comparison to pure Cu. It is clear that the Cu-Ti material exhibited the best performance in terms of current density versus applied potential as well as exchange current density. Based on experimental investigation (Figure 3a), the HER activity trend for the Cu-M bimetallics and pure Cu decreased in the following order Cu-Ti > Cu-Co > Cu-Ni > Cu, which deviated from the theoretical predictions when HBE was used as the sole descriptor. Replicates were also independently tested and similar HER performances were observed (Figure S6). The stability of the Cu-M bimetallics were also examine in long-term cycling experiments and post-reaction surface characterizations were performed to analyze possible changes in the surface composition. In the case of Cu-Co 5 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

and Cu-Ni, the catalytic activity was stable and post-reaction XPS analysis (Figures S7 and S8) showed similar surface compositions, indicating that the surfaces of Cu-Co and Cu-Ni bimetallics were preserved. Interestingly, in the case of Cu-Ti, degradation in HER activity was observed, and this was likely due to Ti surface enrichment as indicated by post-reaction XPS analysis. Most likely, under an energized environment and in the presence of electrolyte, additional Ti atoms migrates toward the surface to further reduce the overall lattice strain. Our recent work on Cu-Ti bimetallic electrocatalyst showed that the HER activity is strongly dependent on the surface concentration of the dopant metal; Ti surface concentrations >>10 at.% ultimately hurt the HER activity.19 These results suggest that there is a trade-off between achieving high HER activity and maintaining catalytic stability.

Figure 3. Electrochemical characterization of Cu-M bimetallic alloys. a) Typical HER activities for Cu-M bimetallics and Cu evaluated in H2 saturated 0.1 M KOH solution with a rotation speed of 1800 rpm at room temperature with a scan rate of 10 mV sec-1. b) Corresponding Tafel plots and (inset graph) exchange current densities.

The Role of Oxophilic Dopants The discrepancy in the HER-predicted activity trend strongly suggest that HBE is unlikely the sole descriptor for determining HER activity in alkaline conditions. Looking closely at the calculated OBEs for the Cu-M bimetallics and pure Cu, it is striking that the trend in catalytic improvement followed the trend in OBEs. Among the catalysts investigated, Cu-Ti and Cu had the highest and lowest HER activity with Cu-Ti and Cu sites bonding to oxygen the strongest and weakest, respectively. This suggests that oxophilicity could be an important factor for determining HER activity. However, as indicated by DFT calculations (Table 1), alloying Cu with Ti, Co, or Ni can change both the HBEs and OBEs, thus making it difficult to differentiate the effects of HBE and OBE on HER activity. In order to deconvolute the effects, a second system of Cu-based catalysts were also investigated. Metal oxide/hydroxide (MO/OH) clusters of the same dopant metals (TiO2, Co(OH)2, or Ni(OH)2) were chemically deposited on Cu such that the HBE can be pseudo-fixed while the HER activity can be studied as a sole function of 6 ACS Paragon Plus Environment

Page 6 of 13

Page 7 of 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

oxophilicity. XPS analysis confirmed the presence of the deposited TiO2, Co(OH)2, or Ni(OH)2 clusters on the Cu surface and the surface concentrations were kept between ~15-20 at.% (Figure S9). SEM imaging also confirmed the presence of these deposited clusters (Figure S10) with cluster sizes of ~20-100 nm. Figure 4 shows that the HER activity trend for the Cu/MO/OH catalysts decreased in the order Cu/TiO2 > Cu/Co(OH)2 > Cu/Ni(OH)2 > Cu. Replicates were also independently examined and similar HER performances were obtained (Figure S11). Since Cu/TiO2 had the highest activity, the loading of TiO2 was increased (>40%). However, a decrease in performance was observed (Figure S12), indicating that a delicate balance between Cu and TiO2 clusters is needed for enhancing HER activity. Most likely the increased amount of TiO2 at the surface blocked necessary Cu sites that is needed to form the key adsorbed hydrogen intermediate, consequentially leading to a decrease in HER activity. The stability of these Cu/MO/OH catalysts were also examined. In all three cases, degradation of activity was observed after long-term cycling experiments (Figure S13). Post-reaction XPS analysis showed a reduction of TiO2, Co(OH)2, and Ni(OH)2 on the Cu surfaces (Figures S13 and S14). Most likely there was detachment and/or dissolution of the MO/OH clusters under reaction conditions. Future studies employing in-situ inductively coupled plasma-mass spectrometer (ICP-MS) could provide additional insights on the detachment/dissolution rates of these Cu/MO/OH catalysts under HER conditions.24

Figure 4. Electrochemical characterization of Cu/MO/OH. a) Typical HER activities for Cu/MO/OH and Cu evaluated in H2 saturated 0.1 M KOH solution with a rotation speed of 1800 rpm at room temperature with a scan rate of 10 mV sec-1. b) Corresponding Tafel plots and (inset graph) exchange current densities. Experimental results clearly showed that the oxophilicity of the dopant metal and metal oxide/hydroxide played a critical role in facilitating water dissociation on Cu-based catalyst surfaces. Previous studies have indicated that the Volmer step, the water activation step to form the first adsorbed hydrogen (Hads), is likely the critical step for HER in alkaline conditions.6, 25-30 Because Cu has weak hydrogen binding sites, the adsorption of water molecule through the interaction of the proton group and eventual formation of Hads is rather difficult. However, with the addition of an oxophilic site such as Ti, the oxophilic site can help adsorb water to the surface through weak interactions with the oxygen atom of the hydroxyl group while a nearby Cu site can facilitate the formation of Hads, and thus the Volmer step. In a recent work, 7 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Subbaraman et al. clearly demonstrated that Ni(OH)2 deposited on Pt had the greatest improvement in HER activity, while more oxophilic metals (ex. Mn(OH)2) led to poisoning of active sites.27 However, in our Cu system, Ni(OH)2 had the least while TiO2 had greatest enhancement. We speculate the difference is likely due to the weak hydrogen binding strength of Cu in comparison to Pt; and therefore, a more oxophilic dopant (Ti) is needed to help facilitate the activation of water. In the case of pure Pt, Pt sites have near optimal HBEs; therefore, the formation of Hads is efficient without the need of an oxophilic metal to help facilitate the formation of Hads. Thus, Pt still had the highest HER activity compared to both Cu-M bimetallics and Cu/MO/OH catalysts even though it has the lowest OBE values (Figure S15).31 Stemming from these observations, we concluded that the HBE is still the dominant descriptor for HER in base, while the presence of an oxophilic metal or metal oxide/hydroxide could increase HER activity where the OBE can be used as a secondary descriptor to predict the level of enhancement. In relations to water activation, Fajin et al. presented a general Brønsted-EvansPolanyi (BEP) type relationship connecting the activation energy for the O–H bond scission of general RO–H (R = organic moiety) compounds.32 Future work applying similar approaches to estimate the trend in activation barriers for the dissociation of adsorbed water and O-H groups on Cu-M bimetallics surfaces, as well as the Ti3O6/Cu(111) and Ni3O3/Cu(111) interfaces, could provide additional insights for the design of HER catalysts. Understanding the role of oxophilic dopant in Cu-M bimetallics for water dissociation is also critical for the rational design of CO2 reduction catalysts. Cu-M bimetallics have been proposed as potential candidates for enhancing CO2 reduction activities, where the addition of a second metal can potentially alter the binding energies of key reaction intermediates on Cu surfaces.33 For example, it has been proposed that a Cu-M bimetallic surface that contains two different metal sites with one site having high oxygen affinity can stabilize the CHO* key reaction intermediate by simultaneously interacting with both the carbon and oxygen atoms. However, the results presented in this work showed that alloying Cu with oxophilic metals can also significantly enhance HER activity, the key side reaction that competes with CO2 reduction. This signifies that alloying Cu with a highly oxophilic metal may not be a suitable approach toward enhancing CO2 reduction selectivity. Other strategies maybe needed to promote CO2 reduction while simultaneously suppressing HER activity. However, future investigation under CO2 reduction conditions is needed to fully examine these Cu-M bimetallics and Cu/MO/OH catalysts for CO2 reduction. Conclusion In summary, we studied a series of Cu-M bimetallics and Cu/MO/OH catalysts to identify descriptors to guide the rational design of non-precious metal catalysts for water dissociation in alkaline conditions. As such, we showed that alloying Cu with low concentrations of oxophilic metals (Ti, Co, or Ni) or depositing oxophilic metal oxide/hydroxide clusters (TiO2, Co(OH)2, or Ni(OH)2) can dramatically improve the catalytic performance compared to pure Cu. The enhancement in the HER activity could be due to the synergistic interactions between Cu and the oxophilic component. Combining both DFT and experimental investigations, we concluded that HBE is the dominant descriptor for HER in base, while the OBE can be used as a secondary descriptor to describe the level of enhancement. Although this work primarily focused on bulk Cu-M bimetallic systems, this discovery can help the rational design of other non-precious bimetallic catalysts. 8 ACS Paragon Plus Environment

Page 8 of 13

Page 9 of 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Materials The Cu-M alloys (Figure S3a) were prepared by arc melting highly pure metal precursors with desired atomic ratios under argon atmosphere. All precursor metals were obtained from Alfa Aesar with purity of 99.99% or greater. The resulting bimetallic ingots were milled and machined down to standard RDE disk inserts with an outer diameter of 5.0 mm and length of 4.0 mm (Figure S3b). Subsequently, the as-machined disks were then sealed in a quartz tube (Quartz Scientific, Inc) under Ar atmosphere (Keen, Grade 5) (Figure S3c), heated to 950 oC in a furnace (MTI) for 24 hrs, and then rapidly quenched in an ice bath. The surface rust was removed using fine grade sand papers (400, 600, 1000, and 2000 Grit) and was finally polished with 0.5 µm alumina particles to a mirror finish. The disks were then sonicated and thoroughly washed to remove residual alumina particles. Finally, the disks were inserted into E4 Series RDE tips (Figure S3d). As for pure Cu, a highly pure copper rod with a diameter of 7.0 mm was obtained from Alfa Aesar and then machined down to standard RDE disk inserts. Similarly, the Cu RDE disks were sanded and polished to a mirror finish. The Cu/MO/OH catalysts were prepared by chemical deposition through the hydrolysis of metal ions. This procedure was found to be facile approach for depositing the desired MO/OH clusters. In general, a polished Cu RDE disk was immersed and equilibrated into a solution of 0.1 M metal chloride aqueous salt solution for about 0.5 or 12 hrs. Nickle(II) chloride and cobalt(II) chloride salts were obtained from Sigma Aldrich. As for the deposition of TiO2 clusters, a 0.01 M solution of titanium(IV) isopropoxide mixed with ethanol as the solution was used instead. For high loadings of TiO2, a 0.1 M solution was used. After deposition, the electrodes were washed thoroughly. Structural Characterization The crystal structure of the as-made Cu-M disks were examined using powder X-ray diffraction (PXRD) on a Bruker D8 Discovery diffractometer using a Cu Kα radiation source. Bulk compositions were determined using energy-dispersive X-ray spectroscopy (EDS) technique using a Zeiss Auriga-60 scanning electron microscopy (SEM) setup. EDS analysis was conducted at five different positions on each bimetallic electrode to determine uniformity of the bulk composition. The standard deviation of the measured composition was