Magnetic Field Assisted Oxygen Evolution Reaction: Beyond Spin Effects

Open AccessRenewablesPERSPECTIVES2 Aug 2024Magnetic Field Assisted Oxygen Evolution Reaction: Beyond Spin Effects Shi-Yi Lin and Jing Fu Shi-Yi Lin and Jing Fu *Corresponding author: E-mail Address: [email protected] Citation: Renewables. 2024;0:1–6https://doi.org/10.31635/renewables.024.202400067 SectionsAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The sluggish kinetics of oxygen evolution reaction (OER) have seriously obstructed the industrial application of water electrolysis and rechargeable metal-air batteries. Intrigued by the molecular orbital bonding of O2 in a spin triplet state, magnetic field-assisted strategies are widely explored to boost OER electroctalytic activities such as magnetohydrodynamic (MHD) effect, magnetoresistance effect, magnetothermal effect, and spin effect. Here, we investigated the impact of magnetic fields on ferromagnetic catalysts, demonstrating that magnetic fields enhance OER activity by aligning spins and improving mass transfer; then we proposed reasonable verification methods for spin alignments. Subsequently, the possibilities of both mass diffusion and surface conditions besides the spin effect were analyzed sufficiently. Finally, we highlighted the importance of distinguishing spin effects from MHD influences. Download figure Download PowerPoint Introduction As the vital half-reaction of water electrolysis and rechargeable metal-air batteries, oxygen evolution reaction (OER) with sluggish kinetics has profoundly obstructed their industrial application.1,2 Fundamentally, the reactant (H2O or OH−) in OER is in a spin-zero state while the product (O2) is in a spin triplet state, which demands the inevitable spin transition, leading to high energy barriers in the reaction process.3 A tremendous effort, including component and structural regulations, has been exerted to adjust spin-related electronic structures of the OER catalysts to accelerate the catalytic reaction kinetics. Recently, magnetic field-assisted OER has become an emerging direction, the enhancement of which has been observed in several aspects such as the magnetohydrodynamic (MHD) effect, magnetoresistance effect, magnetothermal effect, and spin effect.4,5 In the past decades, many efforts have been dedicated to exploring the impact of MHD effects on OER by improving diffusion kinetics.6–8 Driven by the Lorentz force perpendicular to both the directions of charge motion and magnetic field, the various forms of MHD-induced convections are thought of as extrinsic phenomena, which are not affected by a material's intrinsic spin configurations.9,10 During the OER process, the MHD flow has also been identified to assist the oxygen bubble detachment from the electrolyte surface.11 In recent years, some studies have shown the effects of magnetic fields on ferromagnetic catalysts in promoting OER activities.12 In view of the spin parallel alignment in each O2 molecule, aligning spins parallel by magnetic fields is considered a promising route for configuring *O neighbors, reducing their bonding energy barrier, and expediting the OER kinetics effectively.4,13 Due to the induced paramagnetism of *O, it is very low under typical magnetic fields (∼1 T) at room temperature (kBT/μB ∼ 450 T), the spin alignment is supposed to originate from interacting with the ferromagnetic substrates. On the other hand, a ferromagnet (FM) remains ferromagnetic even when its macroscopic magnetization appears zero. Only the domain wall density changes during magnetization processes in ferromagnetic materials and the spin effect is also thought to occur on the domain walls with progressively canted spins.14 The misaligned spins in domain walls are unfavorable to *O coupling, thus the reduction of domain walls under external fields is expected to benefit the reaction kinetics of OER. With the inherent complexity of these systems, it is compelling to explore alternative schemes for confirming the roles of spin alignment. Main Text Take a typical ferromagnetic material as an example, whose hysteresis loop is shown in Figure 1a.15,16 It is critical to note that a ferromagnetic material remains ferromagnetic even in a zero magnetic field.17 The apparent change in its magnetic moment is due to the orientation of magnetic domains. In the initial state when domains are randomly oriented, the apparent magnetization of the whole system is zero. At high fields, all magnetic moments point in the same direction, and the system shows the maximum magnetization, that is, saturation (Bs).18 This is the preferred spin configuration if the reaction of interest prefers spin parallel alignment (such as OER).19 When the magnetic field is reduced from saturation, more and more magnetic domains and magnetic domain walls are formed inside the material. The catalytic activities within each domain do not change, however, because a pair of reactants need to be atomically close to trigger the reaction and all spins are always parallel on this length scale within the domain (typical domain size is hundred nm and above).20 The difference occurs on the domain walls, where the neighboring spins are slightly canted relative to each other. Now the neighboring reaction sites would observe slightly misaligned spins (on the order of a few degrees), which could somewhat lower the catalytic activity.21 Since two close-by activity sites can only experience a slight spin angle change, the spin effect is expected to be relatively small. The spin canting angle and thickness of domain walls are intrinsic properties of the material, and the effect of the external field is mostly to change the population of the domain walls. In the initial state or at the coercivity (Hc) point, there exists a very high population of domain walls and the measured catalytic activity should show lower values. Figure 1 | (a) Hysteresis loop of a typical ferromagnetic material. The yellow dashed line illustrates the virgin magnetization curve that is highly reversible under small magnetization strengths. (b) Expected full sweep I–H curve if the spin effect is the only enhancement factor in the magnetic field. The right panel shows the spin configurations at the characteristic points. (c) Ideal LSV before applying, during applying, and after removing the magnetic field. Download figure Download PowerPoint Therefore, the ideal behavior of OER current density as a function of magnetic fields should be correlated to the schematic hysteresis loop (Figure 1a,b). Starting from the initial state, applying a magnetic field enhances the catalytic activity monotonically. Upon returning from saturation, the magnetization reverses at Hc. Hc is a point with no net magnetization and large amounts of domain walls, thus accounting for the lowest OER performance during the magnetization process. As the applied magnetic field is greater than Hc of the ferromagnetic catalysts, the internal magnetic domains begin to align in the same direction and the domain walls are eliminated. During this process, the adsorbed *O species mostly experience the same spin directions, which accelerates the generation of O2 and exhibits a large current density increment in the voltammograms. Once they reach saturation (Bs), domain walls no longer exist, thus further increases in magnetic fields no longer modify the spin states or catalytic activities, and the current density reaches the maximum and stops increasing. Nevertheless, quite a few reported articles showed that the current densities still kept increasing even when the external magnetic field strength had already far exceeded what is needed for saturation and attributed the current increase to spin effects.22–24 Instead, there are likely factors beyond spin effects. After removing the field, the spin configuration falls into the remanence state (Br); therefore, the catalytic activity only partially returns. Demagnetization (normally reached by gradually cycling fields toward zero) can restore the catalytic activity to the initial point. We recommend that a more solid proof would be to monitor the change of catalytic activity as a function of applied magnetic fields and observe response loops with the key factors of Hc, saturation, and hysteresis, all direct consequences of the material's domain structures. Accordingly, the expected linear sweep voltammograms (LSV) with or without a magnetic field are shown in Figure 1c, and some reported articles have provided similar plots. However, a full current intensity-magnetic filed intensity (I–H) curve, as shown in Figure 1b, is necessary to correlate the catalytic activities to the materials' intrinsic magnetization, while the MHD-induced magnetic field responses do not reach saturation and only become pronounced when the mass transfer is the rate-limiting factor. Compared to sweeping external magnetic fields without modifying the catalyst materials, directly modifying catalyst materials themselves leads to inevitable changes to extrinsic properties like effective areas, surface morphologies, stray field distributions, and so on, and makes the extraction of spin effects even more challenging. The other common type of spin ordering assumes the antiferromagnetic (AFM) configuration, where the neighboring spins align oppositely. Though domain structures do exist in an antiferromagnet, they do not directly respond to external magnetic fields. Thus, there should not be any current change in the LSV of AFM materials with magnetic fields, which is contrary to reported experimental results (AFM IrO2). This phenomenon further suggests that spin alignment is not the only reason for magnetic field-assisted OER.25 On the other hand, the question is, does the MHD effect only promote O2 bubbles escaping? Undoubtedly, magnetic convections act on both reactants and products. As schematically illustrated in Figure 2a, magnetic fields would bring less increment when there is higher reactant concentration in the electrolytes, which has been reported previously,25 and we also observed similar effects. OER in low KOH concentration is regarded as limited by the amount of OH− ions and their transfer (via a Grotthuss-like mechanism),26 while the mass transfer limitation can be improved in higher KOH concentrations resulting in a faster reaction. As a consequence, magnetic field-enhanced mass transfer tends to manifest more dramatically at lower OH− concentrations. In the literature there are also opposite observations that higher pH increases the magnetic field effect,12,27 and intrinsic spin activities are likely present in those studies, but caution needs to be paid to potential surface reconstructions. It is also often observed that the magnetic field enhancement decreases at very high potentials, and the CV curve also becomes linear at those potentials. This indicates the system could be dominated by certain Ohmic elements in the circuit; therefore, mass transfer or spin-related phenomena both become less relevant and cloud judgments. Figure 2 | (a) Schematic diagram of the LSV tested in different KOH concentrations (low, medium, and high). (b) Schematic diagram of local MHD effect: (i) a single domain particle generates stray magnetic fields around, (ii) a multiple domain particle tends to form closed magnetic flux outside, and (iii) a rough catalyst surface generates stray fields outside. Download figure Download PowerPoint We next take a closer look at the potential influence of the magnetic structures, which can be carefully engineered to promote MHD and mass transfer. For small magnetic nanoparticles, which tend to be single domain each, there can be magnetic fields and gradients around them. However, external magnetic fields can only rotate the particles' overall magnetization directions with negligible change to their spin configurations. The stray dipole fields quickly decay away from the particles and can generate very local MHD around them. Larger particles (typically > 100 nm) could host multiple domains and domain walls like bulk materials,28 with external fields able to align the domains and decrease the domain wall population. Moreover, as shown in Figure 2b, a rough catalyst surface can initiate stray magnetic fields causing a local MHD effect, further accelerating near-surface diffusion kinetics. Some research showed that CoFe2O4 particles (8 nm) with single domain had no OER catalytic enhancement in the magnetic field, but others found that when the size of single domain CoFe2O4 is enlarged (17–36 nm) under critical size (40 nm), the magnetic field enhancement increases accordingly.29,30 This stark difference may come from the local MHD effect, caused by exterior stray fields with larger spatial extension at increased particle size. The magnetic convections not only require magnetic fields but also enough action distance to influence the ion transport. The presence of a stray magnetic field may also greatly disturb the researchers' judgment of the spin effect. Another important consideration is that OER is often accompanied by surface reconstruction of transition metal compounds, which is difficult to avoid, and the reconstructed species are regarded as the true OER catalyst.31,32 Most of the reconstructed products, metal oxides and (oxy)hydroxides, are AFM (Figure 3a), therefore it was proposed that there could be spin-pinning effects between the surface oxyhydroxide layer and the underlying ferromagnetic materials.27,33 Such spin-pinning is analogous to exchange bias, which is an interfacial exchange coupling between an FM and an AFM (Figure 3b1).34 However, most of the reconstructed (oxy)hydroxides show paramagnetism at room temperature; due to their Néel temperatures (TN), they are only around tens of degrees K. According to the molecular-field theory, the limited exchange interactions cannot overcome the thermal fluctuations which leads to paramagnetism above TN with random spin alignments. When coupled with an FM, the exchange interactions from the interface can restore local spin ordering to some extent, even above TN (Figure 3b2). However, due to the intrinsically antiparallel molecular fields, the spins above the interface still prefer antiparallel alignments rather than the parallel alignments assumed in some earlier works.27,33 The antiparallel spins in AFM, even after the potential spin-pinning, are unfavorable to the formation of spin-triplet O2. Furthermore, though the spins at the interface between AFM and FM are ordered, the tiny actuating range from interfacial exchange interactions (on the order of nm35) can hardly lead to a significant spin enhancement of OER in a practical system, because the few spin-pinned layers are normally wrapped by more layers of surface reconstructed oxides and (oxy)hydroxides with random spins (exchange interactions away from the interface still not strong enough to withstand thermal fluctuations well above TN). Due to the low magnetic response of paramagnetism, it needs unachievably high magnetic fields to drive the random spins parallel. Thus, the external magnetic field strengths (∼1 T) adopted in the literature can only control the spin arrangement of FM, and accordingly, affect the AFM through the interfacial spin-pinning, but it is inappropriate to expect spin alignment all the way up to the exterior surface at room temperature (Figure 3b4). The overall consequence is that the spin-pinning effect is not enough to promote *O coupling on the actual reaction sites at the surface far from the interface. Figure 3 | Schematic diagrams of (a) change at FM surface with reconstruction, (b) spin configurations with or without magnetic field at different temperatures (only consider the case of single domains). Commonly encountered OER systems fall in the latter TN>T>TC case. Download figure Download PowerPoint On the contrary, stray magnetic fields from the underlying FM are long-range effects and can readily go through the thin reconstruction layers. It is possible that the origin of the OER enhancement in the (oxy)hydroxides/FM system is related to the local MHD effect, which works on transferring more OH- to the reaction surface. The change in reactant concentration, in turn, leads to further surface reconstruction and facilitates better OER. With further magnetic cycling, the continued reconstruction better catalyzes OER, which has been reported earlier.36 The instability of catalysts may influence the observation of the spin effect as well. It is worth pointing out that the surface reconstruction-related effect is most likely irreversible upon field removal, while the MHD effect would fall back to the remanence state. Conclusion In summary, it is of critical importance to bring to attention that when a magnetic field is applied, in addition to the expected spin effects, there also exist field-induced contributions such as magnetic convections and surface reconstructions, which are supposed to work synergistically. Simply observing OER efficiency change under magnetic fields is not enough to claim victory over spin effects. There are numerous uncertainties at the catalyst-electrolyte interface under the external magnetic field because the spin reconfiguration of reactants through exchange interactions or molecular orbital bonding normally requires atomic-scale close contact while maintaining a clean and spin-ordered surface proves difficult in common electrolyte cells. As a comparison, magnetic hydrodynamics has a longer-range effect and in general quite strong, especially in the high current, therefore, more mass-transfer limited scenarios tend to make it hard to isolate such contributions from those of the spin effects. A low current setting with the correct Tafel slope is the ideal region to seek intrinsic spin effects without much complication from MHD. More advanced in situ techniques could be applied to monitor the spin changes during OER processes. It is certainly possible to only look at the intrinsic thermodynamics governed region at very low reaction rates to unravel the spin effects using ultrahigh vacuum surface techniques such as spin-polarized scanning tunneling microscopy or spin-resolved photoemission, though for practical applications the higher current regions may be of more interest. We hope that our analysis offers more ways of clarification and guidance to future developments for magnetically assisted OER applications. Conflict of Interest There is no conflict of interest to report. Funding Information This work was supported by the National Natural Science Foundation of China (grant no. 52172226) and the Innovation Program of Shanghai Municipal Education Commission, China (grant no. 2023FGS3). References 1. Wang X.; Xi S.; Huang P.; Du Y.; Zhong H.; Wang Q.; Borgna A.; Zhang Y. W.; Wang Z.; Wang H.; Yu Z. G.; Lee W. S. V.; Xue J.Pivotal Role of Reversible NiO6 Geometric Conversion in Oxygen Evolution.Nature2022, 611, 702–708. Google Scholar 2. Bai L. C.; Hsu C. S.; Alexander D. T. L.; Chen H. M.; Hu X. L.Double-Atom Catalysts as a Molecular Platform for Heterogeneous Oxygen Evolution Electrocatalysis.Nat. Energy2021, 6, 1054–1066. Google Scholar 3. Torun E.; Fang C. M.; de Wijs G. A.; de Groot R. A.Role of Magnetism in Catalysis: RuO2 (110) Surface.J. Phys. Chem. C2013, 117, 6353–6357. Google Scholar 4. Ren X.; Wu T.; Sun Y.; Li Y.; Xian G.; Liu X.; Shen C.; Gracia J.; Gao H. J.; Yang H.; Xu Z. J.Spin-Polarized Oxygen Evolution Reaction Under Magnetic Field.Nat. Commun.2021, 12, 2608. Google Scholar 5. Zhou G.; Wang P.; Li H.; Hu B.; Sun Y.; Huang R.; Liu L.Spin-Sate Reconfiguration Induced by Alternating Magnetic Field for Efficient Oxygen Evolution Reaction.Nat. Commun.2021, 12, 4827. Google Scholar 6. Luo S.; Elouarzaki K.; Xu Z. J.Electrochemistry in Magnetic Fields.Angew. Chem. Int. Ed.2022, 61, e202203564. Google Scholar 7. Zhang C. Y.; Zhang C.; Sun G. W.; Pan J. L.; Gong L.; Sun G. Z.; Biendicho J. J.; Balcells L.; Fan X. L.; Morante J. R.; Zhou J. Y.; Cabot A.Spin Effect to Promote Reaction Kinetics and Overall Performance of Lithium-Sulfur Batteries under External Magnetic Field.Angew. Chem. Int. Ed.2022, 61, e202211570. Google Scholar 8. Wang H.; Wang K.; Zuo Y.; Wei M.; Pei P.; Zhang P.; Chen Z.; Shang N.Magnetoelectric Coupling for Metal–Air Batteries.Adv. Funct. Mater.2023, 33, 2210127. Google Scholar 9. Grant K. M.; Hemmert J. W.; White H. S.Magnetic Field-Controlled Microfluidic Transport.J. Am. Chem. Soc.2002, 124, 462–467. Google Scholar 10. Ragsdale S. R.; Grant K. M.; White H. S.Electrochemically Generated Magnetic Forces. Enhanced Transport of a Paramagnetic Redox Species in Large, Nonuniform Magnetic Fields.J. Am. Chem. Soc.1998, 120, 13461–13468. Google Scholar 11. Zhan S.; Huang Y.; Zhang W.; Li B.; Jiang M.; Wang Z.; Wang J.Experimental Investigation on Bubble Growth and Detachment Characteristics on Vertical Microelectrode Surface Under Electrode-Normal Magnetic Field in Water Electrolysis.Int. J. Hydrogen Energ.2021, 46, 36640–36651. Google Scholar 12. Garcés-Pineda A.; M.; J. Magnetic of Water in Google Scholar Gracia the Oxygen Chem. Chem. Google Scholar Ren X.; Wu T.; Gong Z.; Pan L.; J.; Yang H.; B.; A.; Gao H. J.; Xu Z. of in Water Google Scholar of Google Scholar R. Magnetic and Google Scholar T.; of the of a Google Scholar W. of the to Magnetic Google Scholar M.; in O2 on Google Scholar R.; A.; M.; M.; C.; G.; M.; C.; A.; A.; of Magnetic Google Scholar R.; J.; J. L.; K. J.; D. T.; R. J.; of at Google Scholar Zhang Z.; L.; Li T.; J.; X.; Xue Gao Magnetic Field Enhanced in Google Scholar Lin L.; R.; M.; Wang T.; Li J.; Xu Y.; Xu X.; Li M.; Du Y.; Wang J.; Wang S.; Jiang Wu W.; C.; Huang B.; Sun Z.; Liu J.; J.; Sun Spin Magnetic Effect of with Enhanced Oxygen Google Scholar Sun Z.; Lin L.; J.; Wang T.; Li J.; Li M.; Liu Y.; Li Y.; M.; Huang B.; Li H.; Sun the Spin of the Magnetic Effect of the Molecular Am. Chem. Google Scholar Zhang Y.; P.; Li S.; Sun J.; Wang W.; B.; Yang X.; Wang X.; Jiang Z.; Wu G.; Xu Field Assisted Oxygen Evolution Reaction of Chem. Google Scholar E.; and Transport of in Google Scholar Wu T.; Ren X.; Sun Y.; Sun S.; Xian G.; G. G.; C.; J. W.; A.; Wang J.; Shen C.; Yang H.; Gracia J.; Gao H. J.; Xu Z. Effect to on for Enhanced Water Commun.2021, 12, Google Scholar Li Q.; C. W.; S.; T.; T.; and Magnetic of Google Scholar J.; Ren X.; Chen R. R.; Sun Y.; Wu T.; S. J. H.; Xu Z. A Magnetic Field is a for Water Chem. Int. Google Scholar P.; Zhang Y.; Du Y.; B.; Wang W.; Wang X.; Zhou Y.; Xu the Oxygen Evolution of CoFe2O4 under a Magnetic Phys. Chem. Google Scholar Gao L.; X.; Wang Z.; C. Li Z.; S.; Zhang M.; Li J.; Hu Y.; Lin in for Enhanced Oxygen Google Scholar and Oxygen Evolution Catalysts or Google Scholar Chen R. R.; Chen G.; Ren X.; J.; S. J. H.; Xi S.; Wang X.; Xu Z. with a Surface for Water at Chem. Int. Google Scholar K.; S. and Magnetic Google Scholar V.; M.; A.; T.; of on an Google Scholar X.; Zhang Y.; Du Z.; Chen H.; Li S.; C.; Zhang W.; L.; W.; Wang J.; Zhang L.; Wang Q.; Huang C.; Field of in for Water Google Scholar evolution