1 
 

 

 

 

 

 

Magnetic enhancement of a high entropy spinel oxide 

electrocatalyst for rechargeable zinc-air batteries 

Ernst Heznz Hechter 

Dissertation 

submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the 

requirements for the degree of Master of Science (Chemistry) 

 

Supervised by  

Dr Dean Barrett 

Professor Kenneth Ozoemena 

Dr Aderemi Haruna 

 

 

June 2024 

 

 



2 
 

Declaration 

I declare that this Research Report is my own 

work.  It is being submitted for the Degree of 

Master of Science (Chemistry) at the University of 

the Witwatersrand, Johannesburg.  It has not been 

submitted before for any degree or examination at 

any other Institution. 

Signed: Ernst Heznz Hechter 

 

_______________________________________ 

  

On this 6th day of June in the year 2024 

 

 

 



3 
 

Abstract 

The exploration of high entropy materials (HEMs) as electrocatalyst materials has only recently 

begun to accelerate. Similarly, the effect of magnetic fields on the oxygen evolution and 

reduction reactions has recently begun to attract great interest. In this work nanoparticles of the 

high entropy oxide (CuCoFeMnNi)3O4 were synthesized and supported on Vulcan carbon for 

use as a bifunctional OER/ORR catalyst in a rechargeable zinc-air battery (RZAB). The 

products were characterized to confirm and investigate the solid solution high entropy phase, 

and the electrochemistry was investigated with and without an external magnetic field. The 

HEMs demonstrated moderate intrinsic electrochemical properties, with overpotentials and 

current densities comparable to commercial platinum on carbon catalysts even at low loadings. 

Here is reported the most significant magnetic enhancement in RZAB power profile in 

literature at the time of writing, as well as improved RZAB stability and areal energy. This 

work offers insight into the mechanism of magnetic enhancement in the case of high entropy 

materials, and pioneers the use of combined strategies to achieve stable, cost-efficient and 

effective bifunctional OER/ORR electrocatalysis. 

 

 

 

 

 

 

 



4 
 

 

 

 

 

Dedicated to my mother,  

to my friends,  

to Leonie,  

and our dreams for this world. 

 

 

 

 

 

 

 

 



5 
 

Acknowledgement 

I extend my heartfelt gratitude to my supervisors, Professor Kenneth Ozoemena, Dr Dean 

Barrett and Dr Aderemi Haruna for their unwavering guidance, invaluable insights, and 

continuous encouragement throughout the course of my research. Their expertise and 

commitment have been instrumental in shaping this project. 

I would also like to express my appreciation to the members of the CATMAT and MEET 

research groups. The collaborative environment fostered by these remarkable groups has 

provided diverse perspectives, expertise, and a supportive community that has enhanced the 

quality of my work. 

A special thanks goes to Wits University and the National Research Foundation for their 

financial support, which has been crucial in facilitating this research project 

(SRUG2204041774). Their investment in academic pursuits plays a pivotal role in advancing 

scientific knowledge and fostering innovation. 

I am indebted to the teams at Wuhan University of Technology, European Synchrotron 

Radiation Facility, Canadian Light Source, Council for Scientific and Industrial Research, 

University of South Africa, and Cornell University for their collaboration and assistance in 

conducting technically and materially demanding experiments. The access to their facilities 

and expertise has been invaluable in gathering essential data for this study. 

Additionally, I extend my gratitude to Maxwell Terban at Max Planck Institute for his expertise 

in the analysis and refinement of diffraction and total scattering PDF data. High entropy 

materials cannot be elucidated without help from such a master of the art. 

This journey would not have been possible without the support and encouragement from my 

family and friends. You listened to my ramblings, provided insights that made me think you’re 



6 
 

all secretly electrochemists, and helped me to keep going when the research seemed not to be 

going anywhere. 

In conclusion, I am grateful to everyone who has played a role in making this research a reality. 

Your support has been indispensable, and I am truly thankful for the collaborative spirit that 

has defined my MSc. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



7 
 

Contents 

Declaration .............................................................................................................................................. 2 

Abstract ................................................................................................................................................... 3 

Acknowledgement .................................................................................................................................. 5 

List of Figures .......................................................................................................................................... 9 

List of Tables .......................................................................................................................................... 10 

List of Terms and Symbols ..................................................................................................................... 11 

Research Output and Conferences ....................................................................................................... 12 

Chapter 1: Introduction ......................................................................................................................... 13 

1.1 Problem, Hypothesis and Aim ..................................................................................................... 13 

1.2 Background and Justification ...................................................................................................... 13 

Chapter 2: Literature Review ................................................................................................................ 22 

2.1 History ......................................................................................................................................... 22 

2.2 The Current Situation .................................................................................................................. 24 

2.2.1 The Limits of Lithium ............................................................................................................ 24 

2.2.2 Climate Concerns ................................................................................................................. 25 

2.2.3 Renewable Renaissance ....................................................................................................... 26 

2.2.4 Zinc for Net Zero? ................................................................................................................. 27 

2.3 Anode .......................................................................................................................................... 28 

2.4 Electrolyte ................................................................................................................................... 29 

2.5 Cathode ....................................................................................................................................... 30 

2.5.1 The Hidden Spintronics ........................................................................................................ 30 

2.5.2 Lorentz Force Transport ....................................................................................................... 32 

2.5.3 Magnetophoretic Transport ................................................................................................. 33 

2.5.4 Electron transfer in a Field ................................................................................................... 34 

2.5.5 Practical Advances in the Field ............................................................................................. 35 

2.6 Material design ........................................................................................................................... 38 

2.6.1 Bifunctionality ...................................................................................................................... 38 

2.6.2 High Entropy Materials ........................................................................................................ 39 

2.6.3 High Entropy Spinels ............................................................................................................ 41 

2.6.4 Bifunctional OER/ORR HEM Catalysts .................................................................................. 43 

2.6.5 (CuCoFeMnNi)3O4 ................................................................................................................. 44 

Chapter 3: Materials and Methods ....................................................................................................... 45 

3.1 Reagents ...................................................................................................................................... 45 



8 
 

3.2 Syntheses .................................................................................................................................... 45 

3.3 Characterisation .......................................................................................................................... 47 

3.3.1 Powder X-Ray Diffraction (PXRD) ......................................................................................... 47 

3.3.2 Total Scattering – Pair Distribution Function (PDF) analysis ................................................ 48 

3.3.2 Raman .................................................................................................................................. 50 

3.3.3 Microscopy ........................................................................................................................... 51 

3.3.4 X-ray Photoelectron Spectroscopy (XPS) .............................................................................. 52 

3.3.5 Electron Paramagnetic Resonance (EPR) ............................................................................. 53 

3.3.6 Ultraviolet and Visible light Spectroscopy (UV-VIS) ............................................................. 54 

3.3.7 Ultraviolet Photoelectron Spectroscopy (UPS) .................................................................... 55 

3.3.8 Thermogravimetry ............................................................................................................... 56 

3.3.9 Surface nitrogen adsorption/desorption ............................................................................. 56 

3.4 Electrochemistry ......................................................................................................................... 56 

3.4.1 Cathode testing .................................................................................................................... 56 

3.4.2 RZAB testing ......................................................................................................................... 61 

Chapter 4: Results and Discussion ........................................................................................................ 63 

4.1 Powder X-Ray Diffraction (PXRD) ................................................................................................ 63 

4.2 Pair Diffraction Function (PDF) .................................................................................................... 65 

 .......................................................................................................................................................... 67 

4.3 Raman Spectroscopy ................................................................................................................... 67 

4.4 Microscopy .................................................................................................................................. 70 

4.5 X-ray Photoelectron Spectroscopy (XPS) ..................................................................................... 73 

4.6 Electron paramagnetic resonance (EPR) ..................................................................................... 76 

4.7 Ultraviolet Photoelectron Spectroscopy (UPS) and UV-VIS Spectroscopy .................................. 78 

4.8 Thermogravimetric Analysis ........................................................................................................ 80 

4.9 Surface Area and Pore Analysis ................................................................................................... 81 

4.10 Three Electrode Electrochemistry ............................................................................................. 83 

4.11 Two-electrode tests on a zinc-air test cell ................................................................................. 90 

Chapter 5: Conclusion ........................................................................................................................... 97 

5.2 Recommendations ...................................................................................................................... 98 

References ........................................................................................................................................... 100 

Appendix ............................................................................................................................................. 109 

A1 Rietveld Refinement parameters ............................................................................................... 109 

A2 Simulated Crystal planes and D-spacings for HESO ................................................................... 111 

A3 Table of A1g Raman signals in partially inverted spinels............................................................. 114 

 



9 
 

 

List of Figures 

Figure 1: Diagram of a RZAB discharging, with permission from Haruna & Ozoemena, 20209 ............ 15 

Figure 2: Representation of the high entropy spinel oxide (CuCoFeMnNi)3O4 ..................................... 17 

Figure 3: Visual summary of the electron spin effect of magnetic enhancement ................................ 20 

Figure 4: The molecular orbital (MO) diagrams of species involved in the ORR/OER reactions. ......... 30 

Figure 5: Synthesis schemes of HESO and HESO/C ............................................................................... 46 

Figure 6: High resolution PXRD of the HESO and composites ............................................................... 63 

Figure 7: Rietveld Refinement of HESO ................................................................................................. 64 

Figure 8: HESO PDF data with cubic and triclinic refinements. ............................................................. 65 

Figure 9: PDF of the metal-oxygen and metal-metal bond distances ................................................... 65 

Figure 10: Comparison of PDF refinements of HESO  using various crystal systems ............................ 67 

Figure 11:  A1g vibration of an MO4 unit .............................................................................................. 68 

Figure 12:  a) Raman Spectra of lattice vibrations. b) Full Raman spectra of 10 and 20 % HESO/C ..... 69 

Figure 13: SEM micrographs of (a1) 20% HESO/C and (a2) Pristine HESO. (b) EDS elemental maps of 

each of the five metals. TEM micrographs of (c1) pristine HESO and (c2, c3) 20% HESO/C ................. 70 

Figure 14: TEM micrographs (a1) Pristine (CuCoFeMnNi)3O4 (a2) lattice fringes of the pristine HESO 

(a3) d spacings determined for the (111) plane (a4) d spacings determined for 202 plane (a5) 

Visualisation of the visible lattice planes. (a6) Micrograph of a particle at a different angle. (a7) FFT 

and (a8) d-spacing determined. (a9) 311 lattice plane (b1) 20% HESO/C (b2) lattice fringes of 

composite (b3) d spacings determined for 202 plane in composite (b4) visualistion of the HESO (202) 

plane ..................................................................................................................................................... 71 

Figure 15: O1s XPS spectra of a) pristine HESO and b) 20% HESO/C .................................................... 74 

Figure 16: Deconvoluted XPS spectra ................................................................................................... 75 

Figure 17: EPR spectra of the HESO and composites ............................................................................ 76 

Figure 18: Normalised EPR Spectra of the HESO and composites ........................................................ 77 

Figure 19: UPS spectra of HESO and composites .................................................................................. 78 

Figure 20: UV-VIS and Tauc plots of the HESO and composites ............................................................ 79 

Figure 21: TGA of 10 and 20% HESO/C under a) nitrogen and b) air .................................................... 80 

Figure 22: Pore size distribution (a) and BET adsorption-desorption curves (b) for Vulcan carbon, 

pristine HESO and 20% HESO/C composite .......................................................................................... 81 

Figure 23: Cyclic voltammetry of (a) pristine HESO, (b) 10% HESO/C, and (c) 20% HESO/C. Only the 5th 

cycles are shown. (d) ORR and (e) OER polarisation curve for the intrinsic catalyst at various loadings, 



10 
 

at 1600 RPM in 1 M KOH. (f) ORR and (g) OER polarisation curves for the 20% HESO/C in 1 M KOH at 

1600 RPM, with and without an applied field. (h) ORR polarisation curves for the 20% HESO/C in 0.1 

M KOH at 1600 RPM, with and without an external field applied. (i) Koutêcky-Levich plots for the 20% 

HESO/C compared to theoretical electron numbers.  CV of 20% HESO/C with and without the field, 

under nitrogen (j) and oxygen (k) atmospheres. .................................................................................. 84 

Figure 24: Tafel plots for 20% HESO/C with and without an applied field under a) ORR and b) OER 

conditions. EIS for 20% HESO/C at c) 0.85 V and d) 1.6 V ..................................................................... 88 

Figure 25: a) The assembled zinc-air battery. b) Power profiles of the standard 20% Pt/C + IrO2 

cathode and the 20% HESO/C cathode, with and without an applied field ......................................... 90 

Figure 27: Rate capability tests for RZABs with (a) and without (b) a magnetic field ........................... 94 

Figure 28: (a) deep discharge profiles of the standard and 20% HESO/C, 2 mA.cm-2 and 36 h cycles 

(b) Shallow discharge profiles of the standards and HESO/C 0.5 mA.cm-2 and 1 h cycles(c) ΔE of the 

20% HESO/C around 50 hours (d) ΔE of the 20% HESO/C around 200 hours of operation (e) EIS at 

the OCV of the HESO/C RZAB with and without a magnetic field (f) End of life EIS at the OCV of the 

HESO/C RZAB before and after deep discharge cycling for 150 hours ............................................... 96 

 

List of Tables 

Table 1: Summary of research in oxygen catalysis with magnetic fields ............................................... 37 

Table 2: A selection of recent publications on HEM bifunctional OER/ORR catalysts .......................... 43 

Table 3: UPS parameters for HESO and composites.............................................................................. 78 

Table 4: BET derived values. .................................................................................................................. 82 

Table 5: Comparison of ORR and OER magnetic enhancement with previous studies ........................ 89 

Table 6: Previous achievements in magnetic enhancement of RZABs .................................................. 91 

Table 7: EIS parameters of RZAB with 20% HESO/C cathode ................................................................ 93 

 

 

 



11 
 

List of Terms and Symbols 

BET Brunauer-Emmett-Teller, a model used for 

finding surface area and pore volume from 

N2 adsorption experiments 

 

EIS Electronic Impedance Spectroscopy 

EPR Electron Paramagnetic Resonance, also 

called ESR (Electron Spin Resonance) 

HEM High Entropy Material, any solid with a 

conformational entropy >1.5R 

HESO High Entropy Spinel Oxide 

OER Oxygen evolution reaction 

  

ORR Oxygen reduction reaction 

  

PDF Pair Distribution Function, a total scattering 

technique to find relative frequency of 

interatomic spacings 

RZAB Rechargeable Zinc-Air Battery 

SEM Scanning Electron Microscopy 

Spin-Orbit Coupling The interaction between the angular 

momentum of an electron around a nucleus 

and the electron’s own intrinsic angular 

momentum 

TEM Transmission Electron Microscopy 

TGA Thermo-gravimetric analysis 

UPS Ultra-violet Photoelectron Spectroscopy 

UV-Vis Ultra-violet and visible light spectroscopy 

Vulcan carbon Brand name for carbon black XC72 

XPS X-ray photoelectron Spectroscopy 

XRD X-ray Diffraction 

 



12 
 

Å Ångstrom, a distance of 10-10 m 

A Ampere, measure of electric current 

B or �⃗⃗�  Magnetic field 

CP Chronopotentiometry 

CV Cyclic Voltammetry 

ECSA Electrochemical Surface Area 

eV Electron-Volt, measure of energy equal to 

1.6*10-19 J 

LSV Linear Sweep Voltammetry 

T Tesla, measure of magnetic field strength 

V Volt, measure of electrochemical potential 

ΔE Difference in potential between an oxidation 

and reduction process, in this document 

referring to the difference between OER and 

ORR potentials 

ΔV The difference between charge and discharge 

potentials of a RZAB 

η Overpotential 

Ω Ohm, measure of electric resistance 

 

Research Output and Conferences 

Oral Presentation at CATSA 2022 

Flash Talk and Poster Presentation at ElectroChemSA 2023 

Oral Presentation at CATSA 2023 

Research paper submitted to Angewandte Chemie for publication 

 



13 
 

Chapter 1: Introduction 

1.1 Problem, Hypothesis and Aim 

Battery based energy storage requires extensive improvement in capacity, cost, and safety to 

meet the demands of the global renewable energy transition. Rechargeable Zinc-Air Batteries 

offer low cost, safe components, and high energy density, but are limited by the slow kinetics 

and high overpotentials of the oxygen reduction and evolution reactions at the cathode. By 

combining two recent approaches to electrocatalysis, namely high entropy spinel oxide ceramic 

nanoparticles and magnetic enhancement, a more efficient bifunctional catalysis approach 

could be demonstrated. The aim of this research is the synthesis and complete characterisation 

of (CuCoFeMnNi)3/5O4 high entropy spinel oxide (HESO), and the demonstration and 

investigation of the effect of a magnetic field on its catalytic activity. HESO synthesis by the 

Pechini method is untested, and full characterisation of this HESO is incomplete in the 

literature. 

1.2 Background and Justification 

Historically, the ever-increasing demand for energy has been met through the utilization of 

fossil fuels such as coal, oil, and natural gas. However, this pattern cannot continue indefinitely. 

While new fossil fuel reserves may yet be discovered, they are by their nature a limited 

resource, and supplies are likely to diminish, driving prices higher and ultimately restricting 

energy utilization.1  Additionally, fossil fuels must generally be transported between their point 

of extraction and point of use.2 These supply chains are fragile and prone to disruption, as has 

been seen in recent armed conflicts.2 

Furthermore, fossil fuel use has been firmly linked to the degradation of the environment and 

human health. Emission of toxic particulate matter, volatile organic compounds, and NOx and 

SOx gasses from vehicles, power plants and factories pose a serious threat, leading to an 



14 
 

estimated 1.3 million yearly excess deaths worldwide2,3. Combined with the direct effect of 

emissions, climate change resulting from the greenhouse effect of carbon dioxide, methane and 

other waste emissions threatens rising sea levels and meteorological disturbances that will 

render many areas uninhabitable.1 Faced with supply restrictions, loss of ecosystem services, 

public health crises and the looming threat of climate change, society has been forced to begin 

the transition towards sustainable energy sources.2 

The alternative to fossil fuel reliance is the adoption of renewable and nuclear energy. 

Photovoltaic and concentrated solar, wind and fission energy technologies are already viable, 

and are being adopted by commercial and government projects worldwide. However, energy 

production alone is not sufficient: storage for intermittent power sources and for vehicles, and 

the conversion of industries with high energy demands such as steel manufacture, remain 

challenges4. 

The current leading energy storage technology is the lithium-ion battery, which is used 

particularly in mobile devices and electric vehicles. However, this technology is faced by 

several limitations, including the high cost and accessibility of lithium and the requirement for 

toxic and expensive non-aqueous electrolytes, as well as the same supply chain vulnerabilities 

that face fossil fuels.5,6 These factors do not discredit the technology, but do slow adoption and 

limit grid-scale storage applications of lithium-ion batteries. Some researchers also believe that 

the technology is approaching a ceiling in terms of energy density.6,7 A potentially cheaper, 

safer, more environmentally friendly and energy dense solution may be found in metal-air 

batteries, shown schematically in figure 1.8 



15 
 

 

Figure 1: Diagram of a RZAB discharging, with permission from Haruna & Ozoemena, 20209 

These devices store energy in the potential between atmospheric oxygen and an elemental 

metal, such as aluminium, lithium, or, of specific interest in this study, zinc. A rechargeable 

zinc-air battery (RZAB) makes use of cheap and plentiful zinc metal, and has a safe, aqueous 

electrolyte. Since the oxygen is supplied by the atmosphere, there is no need for a bulky cathode 

compartment, giving a more compact battery. 

During discharge, oxygen from the atmosphere is reduced (oxygen reduction reaction, ORR) 

at the air cathode to produce aqueous hydroxide ions, while zinc is oxidized at the anode to 

produce Zn2+ ions, forming first aqueous zinc hydroxide and then solid zincate. These half 

reactions are coupled through an external circuit, where the electrons may do useful work. 

During charging, a voltage is applied to reduce zinc ions back to zinc metal, while hydroxide 

is oxidized to produce water and release gaseous oxygen back into the atmosphere (oxygen 

evolution reaction, OER).6,8,10 The appeal of zinc-air batteries is obvious: zinc is relatively 

plentiful, the aqueous alkaline electrolytes are benign, and the theoretical energy density of 

zinc-air batteries is up to five times greater than the lithium ion batteries, with costs per kWh 

of storage up to 40 times lower.6,10 The overall process is shown in reactions 10-13. 

There are two main barriers to the commercial use of zinc-air batteries: deterioration of the 

zinc anode, and sluggish ORR and OER kinetics at the cathode, requiring high overpotentials 

and noble metal catalysts to produce acceptable currents.6,8,10 Cathode catalyst design is 



16 
 

challenging, as they must be bifunctional: highly active toward both ORR and OER, processes 

that generally have different catalyst requirements.6,11 

The sluggish, four electron transfer of 

the oxygen evolution reaction (OER),12–

14 is shown in equations 1 through 415. 

ORR proceeds through one of two 

possible pathways: A four electron 

transfer involving a single catalytic 

active site shown in equations 5 through 

916, or more sluggish two two-electron 

transfers with production of peroxide as 

an intermediate. Currently, precious 

metal catalyst combinations such as 

ruthenium or iridium oxide and 

platinum-on-carbon are used to improve the rate of OER and ORR, but these catalysts are 

unstable and extremely expensive, hindering their use at scale. Development of economic, 

effective, and environmentally friendly catalysts is potentially the breakthrough needed in this 

area17,18. 

ORR/OER bifunctional catalysts for the air cathode should avoid expensive noble metals, be 

stable in oxygen and the alkaline electrolyte throughout the voltage range used and withstand 

at least hundreds of charge-discharge cycles. Most importantly, their catalytic activity should 

be comparable to or exceed the performance of the platinum group metal combination catalysts. 

These requirements are a challenge for traditional methods and materials.6,7,9,11 

* + OH- -> *OH + e-   (1) 

*OH + OH- -> *O + H2O + e-
  (2) 

*O + OH- -> *OOH + e-   (3) 

*OOH + OH- -> * + O2 +H2O + e- (4) 

* + O2 -> *O2    (5) 

*O2 + H2O + e- -> *OOH + OH-  (6) 

*OOH + e- -> *O + OH-   (7) 

*O + H2O + e- -> *OH + OH-  (8) 

*OH + e- -> OH- + *   (9) 

 

O2 + 2H2O + 2e- -> 2H2O2 + 2OH- (10) 

H2O2 + 2e- -> 2OH-   (11) 

 

Zn(s) + 4 OH-
(aq) ↔ Zn(OH)4

2-
(aq) +2e-   -1.25 V (10) 

Zn(OH)4
2-

(aq) ↔ ZnO(s) +2 OH-
(aq) + H2O(l) (11) 

O2(g) + 2 H2O(l) + 4 e- ↔ 4 OH-
(aq)

    0.40 V (12) 

Overall: 

2 Zn + O2 ↔ 2 ZnO    1.65 V (13) 

 



17 
 

A class of materials that have only recently been investigated as electrocatalysts, but that may 

have the complex properties required for OER/ORR bifunctionality, are high entropy oxides 

(HEOs). An example of a HEO crystal can be seen in figure 2. The concept of high entropy 

materials was first introduced by Cantor et. al. and Ye et. al., as materials with five or more 

metals in equimolar proportions, giving them a configurational entropy greater than 1.5R19,20. 

Though the original work was on alloys, the concept has been expanded to include MOFs, 

sulphides, and oxides, among others. Four properties distinguish HEOs from conventional 

oxides of fewer metals.  

 

Figure 2: Representation of the high entropy spinel oxide (CuCoFeMnNi)3O4 

The “high entropy effect” refers to the spontaneous formation of single-phase solid solutions 

at sufficiently high temperatures, due to the increased entropy in a solid solution. In these solid 

solutions, the cation positions of a crystal lattice, such as the spinel structure seen in this study, 

are randomly occupied by the metals. The “sluggish diffusion” effect is the extremely slow 

movement the ions in the HESO structure, as the randomly positioned metal ions create deep 

potential wells and steep energy barriers, effectively fixing the solid solution in a metastable 

state even at low temperatures and through a variety of conditions. The structure is maintained 



18 
 

by this kinetic barrier even when degradation, such as segregation of the metal ions into 

different regions, is thermodynamically favoured. In a RZAB, the bifunctional catalyst must 

retain its structure over a range of voltages and thousands of charge/discharge cycles, and so 

the stability granted by the slow diffusion effect is desirable. The “Lattice Distortion” effect 

arises from differences in sizes and metal-oxygen bond lengths between the ions, causing a 

great deal of strain, distortion and defects in the crystal structure. Strain and distortion modifies 

the electronic band structure of the HESO compared to similar, lower strain analogues, while 

defects such as oxygen vacancies provide additional active sites for electrocatalysis. Finally, 

the “Cocktail” effect is the synergistic effects of many metals in close proximity to one another, 

giving rise to unique active sites.13,21–24 For example, copper binds ORR reactants strongly, 

increasing the rate of reactant adsorption, while iron binds the products weakly, increasing the 

rate of product desorption. Cobalt and nickel, with intermediate binding energies, could provide 

pathways across the surface for adsorbed intermediates to move from strong to weak binding 

regions. This cocktail effect can be seen in other classes of materials, such as multiple metal 

single atom catalysts.25  

Another exotic area of research that has been seeing attention in electrocatalysis in recent years 

is the application of external magnetic fields to enhance the activity of a catalyst. Magnetic 

fields have several interesting effects on electrochemical systems. The simplest are the mass 

transport effects of the Lorentz and Kelvin forces. The Lorentz force arises from the interaction 

between a moving charge and an external magnetic field, with the force acting perpendicular 

to the direction of motion and applied field.26 This deflection leads to convection currents 

especially in areas where charged species typically have a preferred direction, such as near an 

electrode surface.27 Such convection currents improve mass transport and also ensure even 

dissolution and deposition, preventing dendrite formation on metal electrodes.26,28 The Kelvin 

force drives paramagnetic particles towards areas of higher magnetic field gradient, and 



19 
 

diamagnetic particles towards areas of lower gradient.26 This force finds application in 

preventing the accumulation of oxygen bubbles on the electrode surface during OER.28 

Magnetic fields also interact with electron spins of the catalyst. A principal factor in the 

sluggish kinetics of OER and ORR is that these reactions involve spin state changes: oxygen 

is triplet in its ground state, while hydroxide and water are both singlets, with no unpaired 

electrons. Changes from triplet to singlet or vice-versa are quantum mechanically forbidden, 

while singlet oxygen is significantly higher in energy than the triplet state. A catalyst is vital in 

facilitating a reaction pathway leading directly to and from triplet oxygen.29 It has been 

proposed that the alignment of electron spins in the catalyst in the presence of a magnetic field, 

as well as stabilisation of high spin electron states in comparison to low spin states,30 facilitates 

initial electron transfer processes between catalyst and reactant,29 and selects spins such that 

high energy intermediates or activated complexes of unfavourable spin state are avoided as is 

illustrated in figure 3.31 In contrast, Roy et. al. propose that the magnetic field enhances 

intersystem crossing of intermediate radical pairs from singlet to triplet state, preventing 

recombination and smoothing the way to complete the reaction to form triplet oxygen28. It is 

interesting that the facilitation of spin change through spin-orbit coupling is one of the reasons 

why heavy, precious metal catalysts like iridium, ruthenium or platinum are so active for the 

oxygen reaction: spin-orbit coupling allows nonradiative excitation and relaxation and hence 

spin changes while conserving angular momentum. Spin-orbit coupling becomes more 

significant as Z4, making it far more significant for the precious metals than for the 1st row 

transition elements that are the preferred catalysis candidates. The external magnetic field could 

therefore be seen as an artificial strengthening for spin orbit coupling, allowing the needed spin 

changes to occur in the lighter element catalysts. 

 



20 
 

Figure 3: Visual summary of the electron spin effect of magnetic enhancement 

 

Pauli exclusion principle does not allow parallel electrons to occupy the same orbital. In the 

randomly aligned spins of the catalyst, an antiparallel arrangement of spins on neighbouring 

metal centres will not allow reaction with triplet oxygen. 

Oxygen 

Metal ion 

Electron in HOMO 

The reaction may proceed in two ways. Triplet oxygen must adsorb to a site where adjacent metal centres 

happen to have HOMO electrons with parallel spin, which effectively reduces the number of active sites. 

Alternatively, oxygen that has been excited to the singlet state as shown above may react at the antiparallel 

site. Both options reduce the rate of the reaction, leading to slow kinetics and high overpotentials. 



21 
 

 

Though the exact mechanisms involved are debated, it is clear that through some combination 

of the Lorentz and Kelvin forces, spin polarisation, and intersystem crossing effects, the 

application of an external magnetic field can enhance electrocatalysis. 

In this work, we sought to combine the unique properties of high entropy material catalysts 

with the promise of magnetic enhancement to design an effective RZAB. 

 

 

 

 

 

The magnetic field provides the mechanism for spin changes in the catalyst, tending to align more 

electrons to form large magnetic domains where all metal centres have electrons of parallel spin. 

Reactions with triplet oxygen may then proceed rapidly and without the need for high energy 

intermediates like singlet oxygen. 



22 
 

Chapter 2: Literature Review 
 

2.1 History 

Since the dawn of civilisation, deterioration has been part of the human experience. Our food 

rots, our buildings crumble, and our tools corrode. Maybe the ancients saw this simply as the 

will of malicious gods, or some human failing keeping us short of perfection. As enlightenment 

slowly crept across different parts of the world, a deeper understanding became common. 

Entropy increases with the surety of statistics, and things tend towards their lower energy states. 

Water tends towards the oceans, buildings tend towards ruin, and the vast reserve of oxygen 

and moisture in our atmosphere makes metal tools tend towards rust. These are simple facts of 

nature, iron laws of the universe that cannot be violated. Humans, being human, of course learn 

how to harness such processes for our benefit. In the same way a waterwheel may be placed in 

the way of a stream, an electrical circuit may be placed between the atmosphere and a metal. 

This is the foundation of the zinc-air battery and its successor, the rechargeable zinc air battery 

(RZAB). 

Of course, the marvellous zinc-air battery did not emerge spontaneously. It was preceded by 

several different zinc-based batteries, starting with that invented by one of the fathers of the 

electrochemical field in his quest to win a bet. In 1799, Allesandro Volta published his “voltaic 

pile,” a device which could for the first time in history produce a continuous current. 

Unbeknownst to Volta himself, his device was driven by the reaction between zinc and water, 

catalysed by copper: 

Zn + 2H2O -> Zn2+
 + 2OH- + H2    0.76 V  (14) 

This reaction, which produced 0.76 Volts (named after the man himself), would quickly lead 

to a revolution in the natural sciences. His invention was immediately followed by the 

discovery of electrochemical water splitting and the isolation of the elements sodium, 



23 
 

potassium, calcium, strontium, and boron. A reliable source of electricity would find use in 

research, transport, communication, medicine, and soon every aspect of life. The large-scale 

scientific and commercial use of electricity had begun, and the world would never be the 

same.32 

Reaction 14, being the simple process of zinc corrosion, is not a fantastic source of energy, 

unless one is interested in capturing the hydrogen. In fact, in the modern RZABs that will be 

discussed in the following pages, this hydrogen evolving reaction is called a “parasitic 

reaction,” an unwanted redox process that causes loss in energy and irreversible corrosion of 

the anode.6,10,33  

Zinc based batteries would go through various iterations in the decades following the voltaic 

pile, including coupling to aqueous copper, nitric acid, and solid manganese oxide.32  It was 

only 90 years later that the role of the atmosphere in a zinc based battery was first 

investigated,32 though only as a “depolariser,” a source of oxygen to react with the hydrogen 

from reaction 1 and prevent its build-up in the cell.34 What seems to have gone largely 

unnoticed by these early chemists and inventors is the energy that might be supplied by the 

reduction of atmospheric oxygen. In 1932, the switch from acidic to alkaline electrolyte in the 

zinc batteries allowed the use of stable platinum on carbon as an oxygen reduction catalyst, 

and batteries for applications ranging from rail signals to communications to ocean navigation 

began to run on the familiar chemical reactions 10-13. In the 20th century, these reactions were 

seen as a one way street, with reuse requiring at least the replacement of the zinc anode. 32 

As the role of oxygen became better understood, the cathodes changed from carbon rods to thin 

layers of carbon to facilitate oxygen exchange, technology accelerated by advances in gas 

exchange membranes and fuel cells stemming from the cold war space race.32 By the 1980’s, 

zinc air batteries were sporting power densities in excess of 200 W.h.kg-1, allowing a ZAB 



24 
 

powered electric van to cross the Alps. There were more innovations, changes to the anode and 

cathode, innovative cell designs and novel electrolytes. The sky was the limit. However, the 

1980’s marked the beginning of the end for the zinc air battery. The “rocking chair” battery, 

that would become the omnipresent lithium-intercalation battery, was invented.35 

Lithium-ion intercalation batteries have obvious advantages. The battery is sealed, so the 

problems of electrolyte leakage and atmospheric interference, such as the dissolution of carbon 

dioxide in the electrode, are bypassed. It is relatively power-dense compared to early zinc-air 

batteries since the oxygen reduction and evolution reactions at the cathode are inherently 

sluggish. Most importantly, it was rechargeable, which was at best a distant dream for zinc-air 

batteries of the time. And so zinc, the anode that had helped spark the second industrial 

revolution, passed into obscurity, occasionally finding use in hearing aids and other low-power 

medical devices. 

2.2 The Current Situation 

2.2.1 The Limits of Lithium 

Lithium-ion batteries remain dominant in the rechargeable battery market, but several 

drawbacks have become obvious. Toxic organic electrolytes threatened explosive failure of 

devices and vehicles, and lithium itself is a rare, expensive and strategically important resource. 

Some progress was and still is being made in alternative intercalation ions such as sodium, 

potassium and even zinc, as well as improvements to lithium-ion battery electrolytes, but 

progress has been painstaking and there is usually no guarantee of commercial success. The 

writing is on the wall: alternatives to lithium intercalation, which has dominated the markets 

for 30 years, must be found. The urgency of this need for new energy storage technologies has 

been exacerbated by environmental and climate fears, and the push towards green energy. 



25 
 

2.2.2 Climate Concerns 

Climate change, much like the zinc air battery, is not a new idea. The earth is warm, while other 

terrestrial planets are cold and lifeless. The reason for this is our greenhouse effect, caused by 

gasses which absorb and scatter any infra-red radiation emitted from the earth’s surface and so 

cause a general heating up of the atmosphere. Any gas that absorbs the blackbody radiation 

emitted by the earth as it cools after the absorption of solar energy is called a greenhouse gas. 

The most important among these gasses include water, methane, and the notorious carbon 

dioxide. While water is the overwhelmingly most abundant greenhouse gas in our atmosphere, 

carbon dioxide plays an important role. CO2 absorbs radiation with wavelengths around 4.3 

and 14.9 μm, which are precisely the regions where atmospheric water does not absorb 

radiation36. CO2 therefore plays a far greater role in the greenhouse effect than its relatively 

low concentration would suggest. The increasing concentration of CO2 in the atmosphere is 

slowly choking away the wavelength windows through which radiation can escape the earth. 

The energy output from the sun is relatively constant, while the energy that can escape the earth 

is increasingly being captured. The result can be predicted by any student of thermodynamics: 

a heating of the globe, or global warming. This relatively simple conclusion was not well 

received, as it meant that many of the technologies that had propelled humanity to its current 

level of prosperity were becoming problematic with continued and increasing use. The burning 

of fossil fuels for electricity and transport releases millions of tons of carbon dioxide every day, 

along with other gasses like the NOx and SOx families that independently impact pulmonary 

health in nearby humans. The direct consequences of the warming phenomenon include rising 

sea levels, changing weather patterns and disruptions to ocean currents as vast reserves of 

freshwater flow from shrinking glaciers. Indirect consequences include earlier springtime, 

warmer weather, and billions of desperate climate refugees as areas of the world become 



26 
 

unsuitable for agriculture or even uninhabitable as temperatures soar.37,38 Disruption of the Gulf 

stream may actually cause northern Europe to become far colder in a twist of irony.39 

Vigorous efforts to suppress the idea of global warming or “climate change” were quite 

successful in the late 20th and early 21st centuries. However, even fossil fuel companies have 

begun to admit that our current energy sources must be replaced, if only because fossil fuels 

will become depleted in the not-too-distant future.40 

Nuclear fission power has unfortunately been hamstrung by public fear after the nuclear 

disasters at Chernobyl and the recent Fukushima disaster among others, and nuclear fusion is, 

as always, decades away. So, it is in renewable energy that we must place our hope. 

2.2.3 Renewable Renaissance 

Solar and wind energy have seen tremendous growth in research and commercialisation over 

the past decades, and today are cheaper per MWh than their fossil fuel alternatives even without 

considering the indirect health and environmental costs.41 The benefits are obvious: solar 

panels and wind turbines produce no greenhouse gasses or air pollution while operating, and 

harvest energy sources that are for our purposes inexhaustible. Solar or wind “farms” can be 

set up at any scale, and almost anywhere. The cost of entry to the renewable energy market is 

so low that domestic energy users can set up solar panels on their own roofs. 

However, all is not sunshine and blowing winds. As millions of people have said and written 

over the last 20 years, solar and wind power are intermittent, while power requirements are 

generally not. Additionally, the transport industries demand that energy be portable, a property 

conspicuously absent in 20-meter-tall wind turbines.  

Renewable energy needs to be stored, and while innovative technologies like flywheels and 

pumped water may play a part, the renewable energy transition requires battery storage on all 

scales, from devices to vehicles to electrical grids. Up to the time of writing, lithium 



27 
 

intercalation batteries have been the technology of choice, but the problems already stated as 

well as the fact that the lithium batteries are consistently the most expensive part of a renewable 

energy program limit their ability to meet exponentially growing demands.5,42,43 

2.2.4 Zinc for Net Zero? 

Zinc started the electrification of the world in 1799, starting us on the path towards climate 

disaster. However, perhaps this element may offer a way out of this crisis as well. Interest in 

Zinc-Air batteries was rekindled in the early 2010’s, as decades of investigation into oxygen 

catalysis for water splitting and fuel cells culminated in the concept of bifunctional oxygen 

catalysts: single materials that could catalyse not only the oxygen reduction reaction that allows 

the battery to discharge, but crucially also the hydroxide oxidation, or oxygen evolution 

reaction. This reaction takes place when an external potential strips electrons from water and 

hydroxide and forces them onto Zn2+ ions at the anode, regenerating the zinc metal that 

corroded away during discharge. This opened the door to rechargeable zinc-air batteries 

(RZABs) with a single catalyst and gas diffusion site, dramatically cutting down the bulk and 

weight of earlier, tentative RZAB designs that required separate catalyst layers for ORR and 

OER. A growing excitement emerged: here was the potential for a battery consisting of cheap 

and abundant materials, that forsook the dangerous electrolytes of intercalation batteries for a 

simple aqueous (or solid-state) electrolyte, and offered an energy density that could 

theoretically outstrip theorised ceiling for intercalation batteries by 400%.6,7 The fact that the 

cathode uses atmospheric oxygen is another obvious draw, as this slashes the bulk and weight 

of practical cells which do not need a cathode compartment to store ions or chemicals, as is the 

case in conventional batteries.  

All these advantages are well and good, but there is no denying that the road ahead for the 

RZAB is far from straight and easy. Every part of the RZAB has its associated problems and 

pitfalls, and despite the momentum in research there is no guarantee that the technology will 



28 
 

truly overtake intercalation batteries or other energy storage technologies. Let us address the 

factors that hold RZABs back from their destiny as the building blocks of the future. 

2.3 Anode 

The simplest anode is a metallic zinc plate, and this will be a useful starting point for exploring 

the challenges in anode design. The zinc anode undergoes a series of chemical and physical 

changes during the operation of the zinc air battery, summarised in reactions 2 and 3. During 

discharge, metallic zinc releases two electrons and so is oxidised to Zn2+ and enters solution as 

Zn(OH)2. Being in solution is not a favourite pastime of Zn2+, which parts from H2O and is 

deposited as solid ZnO. During charge, aqueous Zn2+ is once again reduced to metallic zinc, 

making room for zinc oxide to enter solution again and then be reduced as well. Theoretically, 

this process regenerates a pristine anode, ready for action as if nothing had happened. However, 

the reality at the electrode surface is hostile to such perfect reversibility. ZnO can deposit 

anywhere during discharge, and often it is deposited on top of active Zn, in several layers. Zn 

that is not in contact with the alkaline electrolyte cannot be oxidised. In this way, the formation 

of ZnO layers on top of the anode passivates some areas, reducing the available anode area and 

increasing internal resistance. A similar problem is that Zn can deposit anywhere on the anode 

during charging, and it has a tendency not to distribute perfectly evenly. It is possible, especially 

at high current density, for slight Zn protrusions called dendrites to form. These dendrites have 

a relatively large surface area compared to a flat plate, and so zinc can deposit on them at a 

faster than average rate. This allows the dendrites to grow at an increasing rate, sending blades 

of zinc stabbing away from the anode surface over successive charge-discharge cycles. Apart 

from the decreased capacity due to internal resistance, the dendrites can pierce battery 

components to cause physical damage, or even reach the cathode and cause a short circuit. 

The interface between zinc and electrolyte is also an area of concern. Recall reaction 14, where 

zinc reacts with water to produce zinc hydroxide and hydrogen gas. The aqueous electrolyte 



29 
 

has no shortage of water, and the precious metal catalysts like platinum that are commonly 

used at the cathode catalyse the hydrogen evolution half of the reaction: 

2H2O + 2e- -> 2OH- + H2      (15) 

This “parasitic” reaction does not provide a great deal of energy, since much of the chemical 

potential remains caught in the hydrogen gas. The hydrogen may react uselessly with oxygen 

at the cathode, or build up in the battery where it can increase resistance, damage cell 

components, or ignite and rob RZABs of their “does not catch fire or explode” marketing angle. 

2.4 Electrolyte 

The fact that RZABs are necessarily in contact with the atmosphere holds another challenge 

for the aqueous electrolyte. As was mentioned in the opening paragraphs, our atmosphere 

contains carbon dioxide with a concentration that is increasing over time and is especially high 

near the industrial areas and roads where RZABs may be used. Apart from being a greenhouse 

gas, carbon dioxide has the troubling property of dissolving in water and forming carbonic 

acid, by equation 16: 

H2O(l) + CO2(aq) -> H2CO3      (16) 

As CO2 partial pressure in the atmosphere increases, its concentration in water rises as well. In 

the environment, this causes a drop in pH in the ocean and other major water bodies, with 

potentially catastrophic consequences for marine ecosystems such as coral reefs.44 A lowering 

of pH is also problematic in RZABs, as the CO2 enters the electrolyte through the batteries gas 

exchange membranes and forms carbonic acid, which neutralises the alkaline electrolyte: 

KOH(aq) + H2CO3(aq) -> KHCO3(aq) + H2O(l)    (17) 

The resulting drop in [OH-] decreases the potential of equation 2, limiting the output voltage 

of the RZAB. It is a bitter irony that the increased concentration of carbon dioxide in the 



30 
 

atmosphere may handicap the very technology we need for the transition to a carbon neutral 

society.  

2.5 Cathode 

2.5.1 The Hidden Spintronics 

Blame for limitations in power density and charging power efficiency can be lain at the door 

of the OER and ORR reactions. Their crimes also include hampering water splitting and fuel 

cell technologies. The lethargy of these reactions has therefore been studied extensively, and a 

major culprit has only recently emerged: the spintronics of the reactions.45–47 The molecular 

orbital diagrams of the reactants in reaction 4 are shown in figure 4: 

 

 

 

 

 

 

 

It is immediately obvious that ORR/OER involve spin changes between triplet oxygen and 

singlet hydroxide. Such a spin state change is referred to as a “spin-forbidden” reaction, 

because in isolation such changes in spin state cannot occur. Large activation energies 

combined with spin-coupling allow progress to a transition state with a changed spin state. 

Such events are relatively rare, and even after attaining this transition state, collapse back into 

the reactant state instead of proceeding toward the product is possible. All these factors 

combine to make spin-forbidden reactions extremely slow compared to spin-allowed 

Figure 4: The molecular orbital (MO) diagrams of species involved in the ORR/OER reactions. 



31 
 

transitions, which leads to the need for expensive catalysts and high overpotentials.31,45,48 

Heavy, precious metal catalysts like platinum and iridium allow for very strong spin-orbit 

coupling, enabling the required spin changes. This is because spin-orbit coupling is 

proportional to Z4, and so the difference between abundant transition metals and the precious 

platinum group metals is vast49. Combined with the favourable adsorption energies, these 

metals have become the gold standard of the oxygen reactions: platinum for ORR, and iridium 

for OER. 

An important property of successful bifunctional catalysts is therefore the ability to select spins: 

that is, ensure that adsorbed reactants have the appropriate spin to react along spin allowed 

pathways. The oldest OER catalyst, Photosystem II in photosynthetic organisms, has been 

found to perform this spin selection using the magnetic active centre of the CaMn4O5 cofactor, 

which regulates the spins of reactants through exchange interactions, causing electrons to align 

with the unpaired electrons in the ferromagnetic enzyme active site.50 As is so often the case in 

chemistry, artificial systems may be enhanced using strategies fine-tuned by nature: indeed, 

transition metal oxides that are ferromagnetic or antiferromagnetic outperform those that are 

paramagnetic.31 The fields created by large magnetic domains compensate for individual 

atoms’ weak spin-orbit coupling, allowing spin changes. Electrolysers and metal air batteries 

may use applied voltage instead of light to provide the impetus, but the chemist’s reaction cells 

and a plant’s mesophyll cells are performing the same reaction.  

Ordered spins in the catalyst provide benefits on several levels. Chief among them are the 

thermodynamic effects of spin alignment: if reactants never need to go through high energy 

transition states like singlet oxygen (90 kJ.mol-1 higher energy than ground state triplet 

oxygen), the activation energy is lower and hence the reaction rate higher. Entropy also plays 

a role: an unpaired electron moving from a disordered paramagnetic material to having a fixed 

spin in a product causes a decrease in entropy, whereas an aligned electron moving to a product 



32 
 

causes no great entropy change.31 Finally, a magnetic field may inhibit reverse reactions by 

promoting intersystem crossing of photo or electrochemically generated radical pairs from 

singlet to triplet state, preventing their recombination.28 

The effects of a magnetic catalyst can be further enhanced by including an external magnetic 

field in the system. (Indeed, a great deal of evidence points towards plant growth being aided 

by external fields in the 100 mT range.51) There has been some controversy on the nature of 

the enhancement by an external field, with furious debate on whether the magnetic field purely 

enhances mass transport or whether there are effects on the level of electron transfer processes. 

The mass transport school of thought argues that the twin contributions of the Lorentz and 

Kelvin forces are sufficient to explain the enhanced activities. 

2.5.2 Lorentz Force Transport 

The Lorentz force will be familiar to any first year student of physics: it is simply that 

perpendicular force experienced by any charged particle moving through a magnetic field, with 

the magnitude 𝐹 = 𝑞(𝑣 ⃗⃗⃗   × �⃗� ) where F is the force, q is the particle’s charge, v is the particle’s 

velocity and B is the magnitude of the magnetic field. As implied by the cross product, the 

force acts perpendicularly to both the velocity and magnetic field direction, and is strongest 

when the velocity and field are perpendicular to each other. In any electrochemical system, we 

expect ions to be in motion, either towards or away from an electrode. An external magnetic 

field serves to exert an additional force on these ions, essentially stirring the solution. This is a 

potentially significant effect since mass transport considerations are of prime importance in 

catalysis and electrochemistry. By the Koutécky-Levich equation, the measured current 

obtained from an electrochemical system i depends on both the kinetic current ik, limited by 

the reaction kinetics, and by the mass transfer current iMT, according to the equation;    

1

𝑖
= 

1

𝑖𝑘
+ 

1

𝑖𝑀𝑇
. The mass transport current is limited by how quickly reactants approach the 



33 
 

electrode and products depart. The additional acceleration from the Lorentz force is believed 

to accelerate these processes for cases where the reactant and/or products are ionic, such as 

hydroxide or hydronium ions in the particular case of OER/ORR, as well as disrupting the 

Helmholtz plane near the electrode surface and decreasing concentration polarization 

effects.52,53 The introduction of additional diffusion currents using a magnetic field is known 

as magnetohydrodynamics. It should be noted that Lorentz effects on hydroxide and hydronium 

are unlikely, as these ions do not undergo generic mass transport under the effect of an electric 

potential. Rather, they propagate through the Grotthuss mechanism of protons “jumping” from 

oxygen ion to oxygen ion. While this may agitate the solution, the development of significant 

Lorentz force currents is not enabled. Indeed, studies by Garcés-Pineda et. al48 showed that 

OER enhancement under the magnetic field was unchanged under increasing electrolyte 

agitation, suggesting that mass transport is not a significant contribution. 

2.5.3 Magnetophoretic Transport 

The second mass transport effect is the Kelvin force or magnetophoretic effect, which is exerted 

on paramagnetic species such as molecular oxygen to draw them from regions of lesser to 

regions of greater magnetic field strength: 𝐹 =
1

2
𝜇0 𝑐𝜒∇𝐵2 where c is concentration of the 

species, χ is its magnetic susceptibility, and μ0 is the vacuum permeability. In the OER, the 

evolution of oxygen bubbles on the electrode surface at high current densities blocks catalyst 

sites, not allowing the electrolyte to contact the electrode. The Kelvin force serves to dislodge 

these bubbles of paramagnetic oxygen, freeing the surface.26,54 Both the Lorentz and Kelvin 

forces are significantly stronger near ferromagnetic catalyst particles.52,55 Indeed, mass 

transport effect are not seen when magnetic fields are applied to non-magnetic catalysts like 

IrO2.
48

 The forces exerted by the field alone, without the presence of magnetic particles to 

enhance it, are weak compared to the usual electric field and thermal effects on ions near an 

electrolyte.53  



34 
 

2.5.4 Electron transfer in a Field 

The mass transport effects alone are not sufficient to explain the enhancement seen in 

electrocatalysis in the presence of an external magnetic field. It has been found that kinetic 

currents may be increased, and charge transfer resistance decreased after the application of the 

field, which cannot be caused by simple mass transport improvements. Additionally, changing 

reaction conditions impacts the extent of the magnetic enhancement: Garcés-Pineda et. al48 

found that magnetic enhancement is more significant for oxygen evolution under alkaline 

conditions than neutral conditions, controlling for electrolyte concentration. They ascribe this 

pH dependence to a change in the rate limiting step, with a non-spin-dependent step being the 

limiting step under neutral conditions, while under alkaline conditions the spin-dependent O-

O bond formation step is rate-limiting. 

 A synergistic effect between ordered spins in the catalyst and the external field is likely, though 

the exact mechanism is unclear. It is possible that the external field simply enhances the effects 

of the magnetic catalyst, by aligning spins across the catalyst in the same direction so that the 

spin selection discussed earlier extends across the entire catalyst, rather than solely within small 

aligned domains. This is supported by temporary maintenance of enhanced activity of 

ferromagnetic catalysts after the external field is removed, at which point only the aligned spins 

of separate domains provide a magnetic field. It is also possible that the Zeeman effect, which 

increases the energy of electrons with their magnetic fields aligned antiparallel to the external 

field while decreasing the energy of electrons aligned parallel, facilitates electron transport: 

antiparallel electrons are donated more easily, while electrons can be more easily accepted into 

energy levels where they will be parallel to the field.    

 

 



35 
 

2.5.5 Practical Advances in the Field 

Let us follow this discussion of the theoretical considerations with a few practical 

demonstrations that have hit the journals in recent years. 

In those ancient days of 2016, Monzon et. al. examined increased ORR limiting current under 

an applied field of 360 mT. They found increases of 3%, 8% and 12% for zinc, cobalt and iron 

nanocrystal catalysts respectively. They ascribed this solely to magnetohydrodynamic effects, 

and the different responses of the metals to the extent of magnetisation of the nanocrystals.27 

In 2017, Elias & Hegde56 used an external magnetic field to enhance the hydrogen evolution 

reaction on nickel-tungsten alloy. They observed a 200 mV higher onset potential, which they 

also ascribed solely to magnetohydrodynamic effects and to the removal of hydrogen bubbles 

by the Kelvin force. They unfortunately did not yet consider the effects of the field on the OER 

in their full cell assembly. 

Closer to modernity, an explosion of investigation into the spintronics of the oxygen reactions 

started in 2019. Garcez-Pineda et. al.48 investigated nickel materials as catalysts under an 

external magnetic field. NiO, Raney Ni, Ni2Cr2FeOx, NiFe2Ox, FeNi4Ox, ZnFe2Ox, NiZnFe4Ox, 

and NiZnFeO were tested, and a significant increase in current density was observed for the 

more ferromagnetic members of the series, most notably NiZnFeOx. Li et. al.57 demonstrated 

50 mV earlier OER onset potential for a cobalt spinel oxide on carbon nanofibers after 

application of a 125 mT field perpendicular to the electric field, a result they ascribed to 

magnetohydrodynamics and changes to “electron energy states,” likely referring to low-spin to 

high-spin conversion as the magnetic field lowers the energy of the high spin state. Kiciński et. 

al. also extended the idea of spin selection into ORR applications for fuel cells, demonstrating 

a 30 mV higher onset potential under a 140 mT magnetic field.53 



36 
 

2020 saw a sharp decline in new research. Indeed, the only major research seems to be by 

Wesstson, Picker & Koper, who found no magnetic enhancement or OER or ORR on platinum 

nanoparticles.58 (They did find that the adsorption energy of hydrogen on platinum could be 

influenced by a magnetic field, which has interesting implications for fuel cell and water 

splitting applications). This decline was not due to a lack of interest, however. Several reviews 

were published investigating the state of spintronic and magnetic enhancement. 

Wang et. al.59 brought magnetic enhancement to the zinc-air battery in 2021, using cobalt 

nanodots electrospun onto carbon nanofibers and a 350 mT magnetic field. They observed a 35 

mV decrease in OER/ORR ΔE, with a 20 mV increase in E1/2 and 15 mV decrease in Ej=10. The 

RZAB they assembled showed a 60 mV reduced ΔE after the field was applied, as well as 

significantly greater stability than the same system without a magnetic field present. Their 

analysis acknowledges the role of spin selection in the significant improvement. 

Zhang et. al. proposed an interesting alternative to the operando magnetic field in 2022. Instead, 

they employed an external magnetic field during the annealing process, forming already 

magnetized FeCo2O4 nanoparticles.60  OER/ORR ΔE was reduced by 80 mV compared to the 

traditionally annealed material, and their RZAB showed an increase in peak power of 18 

mW.cm-2. In addition to the effects of spin selection, this approach alters nanoparticles’ active 

surfaces, giving an additional benefit as well as simplifying potential applications of magnetic 

enhancement: instead of manufacturing devices with bulky and expensive magnets attached 

(which would have to be further shielded to protect adjacent electronics), the cathode material 

itself may be magnetized prior to device assembly. 

In 2023, Zhang et al applied a 1 T magnetic field to their bifunctional FeCo2O4 spinel catalyst 

and rechargeable zinc-air battery. They observed a 40 mV earlier OER onset, 10 mW.cm-2 



37 
 

improved peak power, and a 10 Ω decrease in charge transfer resistance. As expected, mass 

transport resistance was unchanged.61 

Recent research into the use of magnetic fields in OER/ORR catalysis is summarised in table 

1: 

Table 1: Summary of research in oxygen catalysis with magnetic fields 

Catalyst Magnetic 

Field/ mT 

OER E10 /V ORR E1/2 /V ΔE/ mV Reference 

Co/CNF 0 1.650 0.800 850 [1]59 

350 1.635 0.820 820 

FeCo2O4 nanofibre 0 1.580 N/A N/A [2]61 

1000 1.540 N/A N/A 

FeCo2O4 0 1.58 0.78 800 [3]60 

Residual  1.52 0.80 720 

Ni1.5Co1.5O4/Ni foam 0 1.520 N/A N/A [4]62 

125 1.478 N/A N/A 

(Fe+N+S)/C 0 N/A 0.88 N/A [5]53 

140 N/A 0.91 N/A 

 

 

 



38 
 

As new research continues to flow and ever more novel approaches to and uses of spin selection 

and magnetic enhancement are discovered, this aspect of heterogenous catalysis will become 

increasingly important and useful in coming years. 

2.6 Material design 

2.6.1 Bifunctionality 

The recent investigation into spintronics and magnetic catalysts is a deviation from traditional 

catalysts, which include nonmagnetic platinum on carbon for ORR, and iridium or ruthenium 

oxide for OER. They are still commercially favoured in fuel cells and water splitting 

respectively, despite poor stability and extremely high cost.63 In RZABs they have an additional 

drawback in that their activity tends towards only one of the reactions: Pt/C is a poor OER 

catalyst, while IrO2 has little activity for ORR. Using them in a RZAB cathode therefore means 

fabricating a system involving both catalysts, increasing costs and complicating both the 

manufacture and operation of the cell. For RZABs to be truly viable, catalysts must be 

bifunctional, with a good activity towards both OER and ORR. This bifunctionality may be 

quantified by the ΔE, the difference in onset between the ORR and OER reactions. OER and 

ORR occur at different voltages and being favoured by materials with different structures and 

with different optimum adsorption energies of intermediates. An OER catalyst must adsorb 

OH- and readily desorb O2, while an ORR catalyst adsorbs O2 and is inhibited by strongly 

adsorbed OH- or H2O, and these opposing priorities are a clear problem for a catalyst that must 

do both.10,11 Despite this, there has been some success in the field of bifunctional catalysis 

using transition metal oxides.7,31,33,64 It should come as no surprise that a good fraction of these 

bifunctional catalysts also display magnetic properties. 

Gorlin et. al.11 used EXAFS studies on MnOx catalysts to show that the material underwent a 

structural reorganisation as the voltage was varied between the 0.7 V vs RHE applicable for 

ORR, and the 1.8 V conditions for OER. At 0.7 V a disordered Mn3
II,III,IIIO4 structure was 



39 
 

evident. This structure was oxidised to a layered birnessite structure where manganese is 

mainly in the III and IV oxidation state. While this restructuring is fascinating, such significant 

structural changes repeated every time the voltage is varied is not ideal for a commercial 

bifunctional catalyst, which must be stable over an extended period of use. Indeed, such 

restructuring over the wide operating voltage window limits the use of air cathodes with 

multiple catalysts: for example, the mixture Pt/C + IrO2 on the catalyst layer sees the platinum 

giving a high ORR activity while the iridium handles the OER. However, the platinum surface 

typically degrades at the OER potentials, while the IrO2 surface is slowly reduced over 

consecutive charge-discharge cycles.6 Significant structural changes between OER and ORR 

potentials is therefore more a drawback than a viable strategy for promoting bifunctionality. 

Other bifunctional catalysts may be synthesized in such a way that distinct active sites are 

created. For example, Cui et. al. showed that a nickel-iron nitride support with iron-platinum 

alloy nanoparticles showed good OER/ORR bifunctionality, as the nitride was highly active 

for OER while the alloy was more active for ORR.65 Of course such catalysts may require 

relatively complex syntheses, while not eliminating the use of rare metals. Spinel structured 

metal oxides such as Co3O4 are intrinsically bifunctional, though it is significantly more active 

towards OER than ORR. Activity towards both reactions have been enhanced by doping and 

the synthesis of bi- or tri- metallic spinel oxides such as NiCo2O4 or CuCo2O4, as the additional 

metals modify the valence band energetics and have synergistic effects, such as better 

optimising adsorption energies and preventing structural changes over large potential 

changes.6,66  

2.6.2 High Entropy Materials 

Careful synthesis, doping and rational design has been successful in reducing ΔE, however an 

alternative strategy has been gaining momentum in recent years. Instead of slightly altering the 

composition of a parent material, it is possible to create solid solutions of many metals in 



40 
 

equimolar proportions, creating single phased materials with properties that deviate wildly 

from the linear relations seen in conservative doping. Catalysis has entered the realm of high 

entropy materials. 

The concept of a high entropy material (HEM) was almost simultaneously described by Cantor 

et. al. and Yeh et. al. in 2004. They found that alloys containing at least 5 metals in equimolar 

proportions exhibited exciting features. Their high conformational entropy- greater than 1.5 R- 

favoured the formation of single phased solid solution as opposed to multiple phases seen when 

traditional materials are doped with too much enthusiasm. The high conformational entropy, as 

well as the heavily distorted lattice caused by atoms of different sizes coexisting in a single 

phase, also stabilised the materials, preventing degradation.20,67,68 The proximity of so many 

metal atoms leads to an additional effect, dubbed the “cocktail effect,” a rather nondescriptive 

catchall phrase for the synergies that arise from this situation. These effects could include 

modified energy levels, unique binding energies, magnetic or even superconducting 

properties.69 

Like many discoveries, high entropy alloys (HEAs) were not immediately recognised as the 

significant breakthrough they are. Part of the delay is perhaps due to the vast space that is 

opened by the HEAs. Traditional alloys consist of a single principal element with one or two 

additives or dopants, yet still thousands of traditional alloys find applications in the modern 

world. HEAs consist of at least 5 metals in near equimolar proportions, allowing for 

exponentially more combinations than traditional alloys. The first row of the D-block alone 

gives over 600 equimolar HEA possibilities, each with unpredictable properties or potential 

applications. In the decades since 2004, researchers have slowly discovered various 

applications of various compositions of high entropy alloy, including interesting structural, 

magnetic, electronic and superconducting properties. One of the more recent fields to 

investigate this potentially limitless space is that of catalysis: the high entropy and distorted 



41 
 

lattice effects enhance the stability of the catalyst under harsh and varied conditions, while the 

cocktail effect grants HEMs an astounding variety of useful properties, including tuneable 

adsorption strengths, valence band energies, and unique active site geometries.70  

One of the first forays into HEMs as catalysts was by Yao et. al. in 2018, when they 

experimented on the creation of HEA nanoparticles by carbothermal shock synthesis,71 

successfully synthesising a variety of alloys with up to 8 components. They demonstrated a 

good ammonia oxidation catalyst in PtPdRhRuCe, achieving 100% ammonia conversion with 

99% selectivity for NOx products. While their catalysts were platinum-group metal heavy, they 

successfully proved the concept of HEM catalysts and opened the door to subsequent 

investigations. 

2.6.3 High Entropy Spinels 

The discovery that the factors that give HEAs their unique properties are not unique to alloys 

has expanded the space for investigation even further. In 2015, Rost et. al. 72 demonstrated the 

formation of the “entropy stabilised oxide” (CuCoMgNiZn)Ox. They used energy dispersive 

spectroscopy in scanning transmission electron microscopy (STEM-EDS) to show the co-

occurrence of all 5 metals, and further confirmed the high entropy nature using extended x-ray 

absorption fine structures (EXAFS) synchrotron experiments. They found that the four 

transition metals had very similar cation-cation distances. These results combine to show that 

the rock-salt oxide was indeed a solid solution, with lattice positions randomly occupied by the 

metal cations. Furthermore, they found the single solution did not form when any of the 

precursor metal salts were removed- i.e. 4 component oxides did not form a solid solution, 

confirming the influence of the high entropy effect in an ionic solid. High entropy oxides, 

sulphides,73 phosphates,73 spinels,74 perovskites,75,76 and metal-organic frameworks77 are just a 

few of the examples that have been explored since this proof of concept. These more complex 

HEMs offer even greater tunability than HEAs. In the ionic HEMs, for example, we see 



42 
 

essentially two lattices: the anion lattice following a traditional crystal structure, and the cation 

lattice where the metals take up their random occupancy. While the cation sites have the same 

level of variability as the alloys, the anion lattice is also up for modification. An oxide can take 

a rock salt, spinel, or perovskite structure depending on the exact conditions, and each of these 

can be further modified by introducing oxygen defects and vacancies, or activating lattice 

oxygens.70 Feng et. al.78 used the same quinary HEO formula as Rost et. al., but synthesized a 

porous layered structure using a sacrificial graphene oxide template to produce a catalyst that 

was stable, highly active for benzyl alcohol oxidation, and with a highly controllable selectivity 

for the oxidation product: the product could be varied by simply altering reaction temperature 

while maintaining yields of over 90%. The nanosized HEO is rich in oxygen vacancies as 

analysed using XPS, with their density tuneable by adjusting the calcination temperature during 

synthesis. The propensity of the HEO to form surface oxygen defects exposes more metal ions 

as active sites and improves the reactants’ binding energies. 

Conductivity of HESO catalysts is a challenge, however. Spinels conduct electricity through 

the nearest-neighbour small polaron hopping mechanism. Electrons must “hop” from ion to ion 

through the lattice, at each step distorting the surrounding structure. In material science, a 

travelling distortion is called a phonon, and this moving electron accompanied by a phonon 

distortion is called a polaron. They found that polaron hopping is favourable only when moving 

between like atoms, as the energy gained in one site’s charge increases and surrounding 

distortion equals the energy lost from the next as the charge decreases and distortion relaxes. 

This process is therefore adiabatic. However, when jumping between unlike atoms, energy 

changes take place, resulting in such events having much higher activation energy. In higher 

order spinels such as high entropy spinels, paths of adjacent like atoms are rare, and so 

conductivity is lower.79,80 HESO electrocatalysts are therefore poorly conductive, a problem 

overcome by incorporation into an appropriately conductive support. 



43 
 

2.6.4 Bifunctional OER/ORR HEM Catalysts 

Characterisation of novel HEMs remained, and will always remain challenging, but the 

groundwork was laid for their use as technologically relevant catalysts. A new weapon against 

the endless demands of the OER and ORR was proven viable, and a staggering level of 

resources has since been thrown into HEM OER/ORR catalysis in general and RZABs in 

particular. Some of the significant papers on bifunctional HEM RZAB cathode catalysts are 

shown in table 2: 

Table 2: A selection of recent publications on HEM bifunctional OER/ORR catalysts 

Year Catalyst OER E10 

/V 

ORR 

E1/2 /V 

ΔE/ 

V 

Reference 

2023 AlNiCoFeCrMoV/Co-N-C 1.504 0.864 0.640 [81]81 

2022 PtPdAuAgCuIrRu/ 

(AlNiCoFeCrMoTi)3O4 

1.49 0.89 0.600 [82]82 

2023 CrMnFeCoNi 1.51 0.78 0.734 [83]83 

2020 AlNiCoRuMo 1.485 0.875 0.610 [84]84 

2020 AlFeCoNiCr HEA/HEO system 1.47 0.68 0.790 [85]85 

2023  Fe12Ni23Cr10Co30Mn25/CNT 1.55 0.85 0.700 [86]86 

2021 Mn70Ni7.5Cu7.5Co4.2V4.2Fe2Mo2Pd0.5P

t0.5Au0.5Ru0.5Ir0.5 

1.44 0.90 0.540 [87]87 

2021 (AlNiCoFeCr)3O4/Ag 1.51 0.74 0.770 [88]88 

 

A wider search will reveal hundreds of papers exploring the HEM space for electrocatalysis of 

OER or ORR alone. While researchers continue to explore novel HEMs, it may also be time to 

start applying other promising techniques to HEMs. In the same way a HEM’s many elements 

give rise to a cocktail effect of unpredictable synergies, combining HEMs with methods like 

stress, light, and electric or magnetic field enhancement could reveal breakthroughs in the field 

of oxygen catalysis. 



44 
 

2.6.5 (CuCoFeMnNi)3O4 

In 2019, Wang et. al. explored (CuCoFeMnNi)3O4 high entropy spinel nanoparticles as an 

oxygen evolution electrocatalyst.89 They were able to synthesize particles with diameters of 5 

nm using a solvothermal synthesis, avoiding the need for extreme temperatures to attain the 

high entropy effect. XPS was used to probe the oxidation states of the metal ions. While copper 

was found almost exclusively in the Cu2+ state, indicating occupation of tetrahedral voids only, 

the other 4 metals showed a random distribution between the octahedral and tetrahedral sites. 

The catalyst showed promising activity for OER, but had a low electrical conductivity and 

required support on multi-walled carbon nanotubes. The 5 metals were chosen for their similar 

size across oxidation states and ease of mixing in their binary and ternary systems.  

The metal combination used is more interesting than Wang et. al. give it credit for. For OER, 

iron and cobalt oxides show an optimum adsorption energy for reactants and products, while 

manganese oxides and nickel oxides show slightly stronger or weaker than ideal binding 

energies, respectively.90,91 Their combination is therefore promising, as they could form active 

sites that both strongly adsorb reactants and easily release products. For example, binary oxides 

of iron, cobalt and nickel have given good OER activity92. For ORR, copper binds oxygen most 

weakly (closest to the optimum binding energy where platinum and palladium are found) with 

Ni, Co and Fe binding oxygen progressively more strongly.92 This once again allows active 

sites where good reactant adsorption and good product desorption can be combined. Altogether, 

(CuCoFeMnNi)3O4, called HESO from here on, and the composite with conductive carbon, 

called HESO/C from here on, should be a single phased spinel exhibiting good reactant and 

product binding energies, a distorted lattice promoting active electrons for catalysis, richness 

in reactive oxygen defects, and ferromagnetic properties ready for enhancement by an external 

magnetic field. 

 



45 
 

Chapter 3: Materials and Methods 

3.1 Reagents 

>98% Purity cobalt II nitrate hexahydrate was obtained from ACE chemicals. >98% Purity 

copper II nitrate trihydrate and iron III nitrate nonahydrate were obtained from Sigma Aldrich 

>97% Purity manganese II nitrate tetrahydrate was obtained from Sigma Aldritch 

>99% Purity nickel II nitrate hexahydrate was obtained from Fluka Chemika 

>99.5% Purity zinc acetate was obtained from UNIVAR. 

5% Nafion 117 and 60 Wt. % polytetrafluoroethylene dispersed in water were obtained from 

Aldrich 

Anhydrous ethylene glycol with a 99.8% purity was obtained from Sigma Aldritch 

Potassium hydroxide electrolytes were made using >85% KOH pellets from ACE chemicals 

and water purified using a Purelab Flex device. 

Vulcan carbon and OLC were obtained from Sigma Aldrich. All reagents were used as received, 

without further purification. 

3.2 Syntheses 

The HESO was prepared by the Pechini method93: equimolar amounts of each nitrate salt were 

dissolved in water. The solution was heated to evaporate the water, giving a gel that was then 

pyrolyzed. The product was calcined at 500 °C to obtain (CuCoFeMnNi)3O4. This was then 

annealed in air at 750 ˚C for 12 hours. This annealed product will be referred to as HESO, for 

High Entropy Spinel Oxide. 

Dispersal on carbon was done as follows: Appropriate masses of carbon and HEO were 

dispersed separately in ethylene glycol. The HEO was added dropwise to the carbon and mixed 

under nitrogen. The mixture was then stirred and heated at 125 ˚C for 3 hours, still under 

nitrogen. The product was obtained by filtration and washed with distilled water. The synthesis 



46 
 

scheme is shown in figure 5. This supported catalyst will be referred to as HESO/C, for High 

Entropy Spinel Oxide on Vulcan Carbon 

 

Figure 5: Synthesis schemes of HESO and HESO/C 



47 
 

3.3 Characterisation 

3.3.1 Powder X-Ray Diffraction (PXRD) 

The powder x-ray diffraction (PXRD) data was collected at the Canadian Light Source 

Brockhouse lower energy wiggler beamline using a Huber diffractometer and a Mythen1k 

detector. The samples were placed in 0.54 mm diameter Kapton capillaries and were spun 

during the measurements to improve statistics. The beamline photon energy was 15.1328 eV  

// wavelength of 0.819308 Angstroms. Rietveld refinements were done using GSAS II and 

TOPAS V7 software.  

XRD is a characterisation method that takes advantage of the symmetries of solid, particularly 

crystalline, materials. X-rays of a specific wavelength are radiated onto a sample, where they 

diffract from the material according to Braggs law: 𝑛𝜆 = 2𝑑𝑠𝑖𝑛𝜃 where λ is the x-ray 

wavelength, d is the spacing between the atom layers, and θ is the angle between the incident 

x-rays and the atomic planes. Θ values where this Bragg condition is satisfied will result in 

intensity maxima of diffracted x-rays. By recording the angles where Braggs law are satisfied, 

the interplanar spacing ‘d’ can be found, and a full spectrum can give sufficient information to 

elucidate the crystal structure. In powder x-ray diffraction, the sample consists of many 

randomly oriented crystallites, and so information in a powder x-ray diffraction pattern is less 

powerful for elucidation. However, using other structural information, even a powder pattern 

can be used to determine accurate crystal lattice parameters. Crystal databases contain tens of 

thousands of PXRD patterns, and can be used to find possible matches to the pattern in 

question. For novel, defective, strained or otherwise modified materials that need their 

structures investigated, there is Rietveld refinement. This method is named after Hugo 

Rietveld, who in the 1960’s helped pioneer automation of structure elucidation. The idea is 

elegant: rather than generate a structure from XRD data, we generate theoretical XRD data 

from approximate structure and experimental parameters. Once a theoretical XRD pattern is 



48 
 

generated, the refinement process begins. Adjustments to parameters are made, and the change 

in the theoretical pattern tracked, with the goal of minimising the mean squares deviation from 

the experimental pattern. This is process that repeats until a minimum deviation is reached. 

There is great flexibility in which parameters that can be “refined” in this way, including but 

limited to any or all of the lattice parameters, crystallite size, background, microstrain, atomic 

positions, or occupancies. Of course, this process would be impractically tedious to do by hand, 

but can be realised using computation. Using punch cards and an Electrologica X1 computer, 

Rietfeld et. al. was able to compute the simultaneous refinement of 33 parameters of a neutron 

diffraction pattern, and modern computers can handle even more complex refinements.94 Of 

course, such a powerful process comes with great responsibility. The final refinement is 

sensitive to the accuracy of the initial assumptions, and it is very easy to obtain outrageously 

incorrect conclusions from an improper refinement.  

3.3.2 Total Scattering – Pair Distribution Function (PDF) analysis 

Total scattering data was collected at the ID11 beamline at the ESRF and used in conjunction 

with the XRD data to determine the pair distribution function (PDF) The X-ray wavelength 

used was 0.819308 nm. This technique gives a weighted distribution of inter-atomic distances. 

It therefore does not require long-range order as PXRD does, and importantly is not limited by 

peak broadening in small nanocrystalline materials.95  

Background subtraction, normalization, and Fourier transformation of the data was done using 

PDFgetX3 within xPDFsuite. The total scattering structure function S(Q) is obtained from the 

coherent scattering intensities Ic(Q), after removal of the self-scattering by, 

𝑆(𝑄) =
𝐼𝑐(𝑄)/𝑁 − ⟨𝑓(𝑄)2⟩ + ⟨𝑓(Q)⟩2

⟨𝑓(Q)⟩2
 



49 
 

Q is the magnitude of the scattering momentum transfer (Q = 4π sin(θ)/λ for elastic scattering, 

where λ is the wavelength, and 2θ is the scattering angle). ⟨f (Q)⟩ is the average atomic form 

factor over all atoms in the sample. The experimental PDF is obtained via truncated Fourier 

transformation, 

𝐺(𝑟) =  
2

𝜋
∫ 𝑄[𝑆(𝑄)  −  1] sin(𝑄𝑟) 𝑑𝑄

𝑄𝑚𝑎𝑥

𝑄𝑚𝑖𝑛

 

and is related to the density distribution by; 

𝐺(𝑟)  =  4𝜋𝑟 [𝜌(𝑟)  − 𝜌0𝛾0] 

where ρ0 is the average atomic number density and ρ(r) is the local atomic pair density, which 

is the average density of neighboring atoms at a distance r from an atom at the origin. γ0 is the 

characteristic function of the diffracting domains which equals 1 for bulk crystals, but has an 

r-dependence for nanosized domains. The PDFs were determined with 𝑄𝑚𝑎𝑥 = 25 Å-1. The 

instrumental resolution effects were determined by structure refinement to the reference CeO2 

sample resulting in values of 𝑄𝑑𝑎𝑚𝑝 = 0.02971 Å-1 and 𝑄𝑏𝑟𝑜𝑎𝑑 = 0.03443 Å-1, which were 

fixed for subsequent refinements. 

PDF refinements were performed using the program PDFgui. As in XRD, PDF refinement is a 

process of creating a model and calculating its theoretical PDF, and then comparing it to the 

observed PDF. The model is then altered to improve the correlation with the experimental PDF. 

The deviation between G(r)experimental and G(r)calculated is minimised to find a model that most 

closely matches reality, if the assumptions used to initiate refinement were correct. The refined 

parameters for each phase in the multiphase refinements included the lattice parameters by 

symmetry, phase specific scale factors, separate isotropic atomic displacement parameters for 

metal and oxygen atoms, an r-dependent peak sharpening term, δ1, to account for short-range 



50 
 

correlated motion, and a damping factor associated with finite spherical domain sizes, defined 

as 

𝛾(𝑟)𝑠𝑝ℎ𝑒𝑟𝑒   = [1 −
3𝑟

2𝑑
+

1

2
(
𝑟

𝑑
)
3

]𝐻(𝑑 − 𝑟) 

where 𝑑 is the domain diameter. 𝐻(𝑟) is a step function with value 1 for 𝑟 ≤ 𝑑 and 0 beyond. 

The PDF is simulated from a structure model by calculating the radial distribution function as 

𝑅(𝑟)   =
1

𝑁
∑

𝑓𝑖
∗𝑓𝑗

〈𝑓〉2
𝑖,𝐽≠𝑖

𝛿(𝑟 − 𝑟𝑖𝑗) 

where 𝑓𝑗 is the X-ray atomic form factor evaluated at 𝑄 =  0, and the delta functions are 

broadened by contributions from the atomic displacement parameters, 𝑄𝑚𝑎𝑥, 𝑄𝑏𝑟𝑜𝑎𝑑, and δ1. 

The overall equation, 

𝐺(𝑟) = (
𝑅(𝑟)

𝑟
− 4𝜋𝑟𝜌0) 𝛾(𝑟)𝑠𝑝ℎ𝑒𝑟𝑒 

then gives the PDF for the model.95  

3.3.2 Raman 

When radiation is scattered by an atom, molecule or structural motif in a crystal, the wavelength 

may change. Stokes-Raman scattering is the phenomenon where the scattered light has a lower 

wavelength than the incident radiation: the incident light excites an electron to a higher energy 

state, after which the electron returns to an intermediate energy state instead of the ground state. 

The change in energy of the scattered light is highly dependent on the electronic environment, 

and so can give chemical and structural information. Raman spectroscopy is done by 

illuminating a sample with a monochromatic laser, and measuring the energy and intensity of 

the scattered light. This is used to produce a graph of intensity vs Raman shift, or the deviation 

from the illuminating energy, in inverse centimetres. One common use of Raman spectroscopy, 



51 
 

for example, is to quantify the relative frequency of sp2 compared to sp3 hybridised carbon in 

a sample. This is important in carbon-supported catalysts. Sp2 hybridised or graphitic carbon 

forms conductive layers but is unreactive, and does not provide strong electronic metal-carbon 

interaction (EMSI). However, sp3 hybridised carbon atoms are associated with defects and 

dangling carbon bonds, able to interact more strongly with the catalyst.96 

Raman data was collected suing a 514.5 nm (green) line, Lexel Model 95 SHG argon ion laser, 

as is recommended for the study of metal-oxides and various inorganic systems. The short 

wavelength allows higher power than when using less energetic photons, improving Raman 

signal. The beam can also be more tightly focussed than longer wavelength lasers, increasing 

intensity at the sample and so increasing the Raman signal as well.97,98 

3.3.3 Microscopy 

Visible light microscopy reaches a limit when one wants to resolve detail finer than the 

wavelength of the light used. By accelerating electrons through a powerful potential, a beam 

of electrons with very short wavelengths may be produced. A beam of high energy electrons 

may be focussed in a narrower beam than is possible with visible light, giving electron 

microscopes far greater resolution. There are two general methods to electron microscopy: 

surface bombardment in scanning electron microscopy (SEM), and transmission in 

transmission electron microscopy (TEM). 

In SEM, electrons are focussed into a beam and scanned over the surface of the sample in a 

raster pattern. Data is collected at each point of the pattern, and includes scattered electrons, 

secondary electrons, and emitted x-rays. This data can be interpreted as simply pixels of 

varying brightness to give a greyscale image of the surface, or can be used to generate elemental 

maps using the characteristic x-rays emitted (known as energy dispersive spectroscopy or EDS) 



52 
 

In TEM, the electron beam passes through the sample and impacts a detector on the other side. 

The electrons interact with the atoms of the sample on their way through, altering their phase 

and intensity, and so an image of the sample is created on the detector. There are many different 

modes in which TEM can be used and many ways to generate the contrast in the image. In this 

research the focus was on identifying crystal grains and d spacing. 

SEM was done using a Zeiss Auriga Field Emissions SEM after mounting the samples on 

carbon tape and coating using an Epitech Carbon Coater. The software used was SmartSEM. 

EDS was done using a Zeiss Crossbeam 540 operated at 2 kV for imaging and 20 kV for EDS. 

Data was acquired using an Oxford Xmax detector with Aztec software optimised using copper 

reference. 

TEM was performed using a JEOL JEM 2100 at 200 kV. Samples were sonicated in ethanol 

before being dispersed on a carbon-coated copper grid 

3.3.4 X-ray Photoelectron Spectroscopy (XPS) 

X-ray photoelectron spectroscopy is based on the photoelectric effect. The photons bombard 

the sample, and have a chance of being absorbed by electrons therein. If the photon energy is 

sufficient to allow the electron to escape the atom, a free “photoelectron” is produced with a 

kinetic energy equal to the difference between the energy of the incident photon and the work 

function. By measuring the kinetic energy of the photoelectrons and subtracting the photon 

energy, a spectrum of work functions is made. This allows one to identify both the elements 

present in the sample as well as their oxidation states and chemical environments, as these 

modify the work function. The element, environment and oxidation state will determine the 

energy, and the relative abundance of the species will determine the area under the XPS peak. 

XPS was performed using an ESCAlab 250Xi instrument from Thermo. Monochromatic Al kα 

radiation (1486.7 eV) was used. X-ray power was 300 W and spot size was 900μm. Pass energy 



53 
 

was 100 eV for survey scans and 20 eV for high-resolution scans. Pressure in the system was 

less than 10 -8 mBar. Spectra were analysed using CASA-XPS and Origin software. Shirley 

background analysis was used. 

3.3.5 Electron Paramagnetic Resonance (EPR) 

Electron Paramagnetic resonance is used to probe the chemical environments and relative 

amounts of unpaired electrons in a sample. In oxides such as the spinel under investigation, the 

EPR intensity is closely correlated to the concentration O-, where the oxygen ion has an 

unpaired reactive electron. O- is of particular interest as an indicator of the surface defect 

concentration, which in turn indicates the abundance of active sites for electrocatalysis.99,100 

EPR is based on the Zeeman effect: a magnetic field applied to a sample causes the unpaired 

electrons to align either parallel or antiparallel to the field, which lowers or raises their energy, 

respectively. A radio-frequency photon can be absorbed by a low energy electron, temporarily 

flipping it to anti-parallel alignment. An EPR device works by applying a constant radio 

frequency to a sample and slowly varying the magnetic field while monitoring the absorption 

of radio photons, taking care that the sample not become saturated. When the magnetic field 

induces a Zeeman effect splitting equal to the photon energy, absorption will take place. The 

signal is always broad due to time-energy uncertainty, and so the derivative of the signal is 

used. The photon energy and magnetic field strength can trivially be used to find the electron 

g factor, using the equation hv = geB0μB where h is the Planck constant, v is the radio frequency, 

B0 is the magnetic field strength, and μB is the Bohr magneton. Any deviation from the free 

electron g factor of 2.0023 is caused by the chemical environment of the unpaired electron. 

Interpretation is similar to nuclear magnetic resonance experiments: a lower g value can be due 

to shielding, and a signal may be split by nearby spins such as other unpaired electrons or nuclei 

leading to doublets, triplets, and so on. Such splitting is referred to as the fine structure. The g-



54 
 

factor is in fact a tensor rather than a scalar, with x, y and z components to account f