Electrochemical oxidation and reduction of 5-hydroxymethylfurfural in alkaline water electrolyzers [original]

Electrochemical Oxidation and Reduction of

5-Hydroxymethylfurfural in Alkaline Water Electrolyzers

vorgelegt von

M. Sc.

Philipp Marcel Hauke

an der Fakultät II – Mathematik und Naturwissenschaften

der Technischen Universität Berlin

zur Erlangung des akademischen Grades

Doktor der Naturwissenschaften

Dr. rer. nat.

genehmigte Dissertation

Promotionsausschuss:

Vorsitzende: Prof. Dr. Maria Andrea Mroginski

Gutachter: Prof. Dr. Peter Strasser

Gutachter: Prof. Dr. Holger Dau (FU Berlin)

Tag der wissenschaftlichen Aussprache: 13. Juli 2023

Berlin 2023

Danksagung

Zunächst möchte ich mich ausdrücklich bei Herrn Prof. Peter Strasser bedanken, der es

ermöglichte, meine Fähigkeiten und Kenntnisse zu erweitern, indem er mich immer wieder vor

neue Herausforderungen stellte und mir half diese zu bewerkstelligen. Ebenso möchte ich mich

für die gute Zusammenarbeit bedanken. Über den gesamten Zeitraum meiner Promotion

ermöglichte er mir eine freie aber effektive Arbeitsweise.

Ebenfalls möchte ich mich bei Herrn Prof. Holger Dau für die Übernahme des 2. Gutachtens

bedanken.

Außerdem bedanke ich mich bei Frau Prof. Maria Andrea Mroginski für die Übernahme des

Vorsitzes.

Zusätzlich möchte ich mich bei dem gesamten Arbeitskreis für das nette Miteinander und

jegliche wissenschaftliche und nicht wissenschaftliche Hilfe, bedanken. Im Speziellen möchte

ich mich bei Sven, Trung, Beni, Thomas, Malte und Tim für die Zusammenarbeit bedanken.

Auch bei meinen ehemaligen Kommilitonen und Freunden Robert, Olli, Kai, Paul und Jannik

möchte ich mich für offene Ohren während des gesamten Studiums und auch für den

notwendigen Ausgleich neben dem Studium bedanken.

Am Ende gilt natürlich auch meiner Familie großer Dank, für die Motivation, Geduld und lieben

Worte während des gesamten Studiums.

Table of Contents

Table of Contents ...................................................................................................................... I

Zusammenfassung ................................................................................................................... V

Abstract ................................................................................................................................. VII

1 Introduction ...................................................................................................................... 1

1.1 Electrochemical Water Splitting ............................................................................................ 3

1.1.1 Alkaline Hydrogen Evolution Reaction (HER) ............................................................................... 5

1.1.2 Alkaline Oxygen Evolution Reaction (OER) .................................................................................. 6

1.2 Electrochemical conversion of 5-Hydroxymethylfurfural ..................................................... 8

1.2.1 Electrochemical Oxidation of HMF ................................................................................................ 9

1.2.2 Electrochemical Reduction of HMF .............................................................................................. 12

1.3 Motivation, Scope, and Outline ........................................................................................... 14

2 Theoretical Background ................................................................................................ 16

2.1 Fundamentals of Electrochemistry....................................................................................... 16

2.2 Thermodynamics of Electrochemistry ................................................................................. 18

2.3 Kinetics of Electrochemistry ................................................................................................ 23

2.3.1 Reaction Rate – Bond Strength – Relation .................................................................................... 27

2.4 Organic Electrochemistry .................................................................................................... 30

3 Experimental ................................................................................................................... 32

3.1 Synthesis .............................................................................................................................. 32

3.1.1 Microwave Assisted One-pot Synthesis of NiX(-CO32-)-LDH Powder ........................................ 32

3.1.2 Anion Exchange for NiX(-CO32-)-LDH Powder ........................................................................... 32

3.1.3 Microwave Assisted One-pot Synthesis of the As-prepared NiFe(-CO32-)-LDH@NF Electrodes 33

3.1.4 Anion Exchange of the As-prepared NiX(-CO32-)-LDH@NF Electrodes ..................................... 33

3.1.5 Precipitation Synthesis of Pure Metal Oxides (MOx; M= Cu, Ni, Fe, Co) .................................... 34

3.1.6 Precipitation Synthesis of Mixed Metal Oxides (CuO/MOx 10 mol%; M= Ni, Fe, Co) ............... 34

3.2 Characterization Methods .................................................................................................... 34

3.2.1 Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) .................................... 34

3.2.2 Powder X-Ray Diffraction and Thin Film X-Ray Diffraction (XRD) .......................................... 35

3.2.3 Scanning Electron Microscopy (SEM) .......................................................................................... 35

3.2.4 Transmission Electron Microscopy (TEM) ................................................................................... 35

3.2.5 X-Ray Photoelectron Spectroscopy (XPS) .................................................................................... 35

3.3 Setup Configurations ........................................................................................................... 36

3.3.1 Rotating Disk Electrode (RDE) Measurements for HMF Oxidation ............................................ 36

3.3.2 Differential Electrochemical Mass Spectrometry for HMF Oxidation ......................................... 36

3.3.3 Rotating Disk Electrode (RDE) Measurements for HMF Reduction ............................................ 37

3.3.4 Undivided Three-Electrode Cell Measurements for HMF Reduction ........................................... 38

3.3.5 Full-Cell measurements and Fabrication of the Applied Electrodes for HMF Oxidation ............. 38

3.3.6 Full-Cell Measurements and Fabrication of the Applied Electrodes for HMF Reduction ............ 39

3.3.7 Product Analysis for HMF Oxidation ........................................................................................... 40

3.3.8 Product Analysis for HMF Reduction ........................................................................................... 40

3.3.9 Calculation of ECSA, HMF Conversion, Product Yields, Faradaic efficiency, and STY ............. 41

4 Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance ................................ 42

4.1 Abstract ................................................................................................................................ 42

4.2 Introduction .......................................................................................................................... 43

4.3 Results and Discussion ........................................................................................................ 44

4.3.1 Morphology and Surface Composition of Tunable Anion-Interlaced Catalysts ........................... 44

4.3.2 Structure-Composition-Catalytic OER Reactivity Relations ........................................................ 47

4.3.3 Translating RDE to AEM Water Electrolyzer Cell Performance .................................................. 51

4.4 Conclusion ........................................................................................................................... 55

5 Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities ................... 56

5.1 Abstract ................................................................................................................................ 56

5.2 Introduction .......................................................................................................................... 57

5.3 Results & Discussion ........................................................................................................... 60

5.3.1 Synthesis and Structure of Anion-Modified -NiX-LDHs Supported on Ni Nanofoams ............. 60

5.3.2 Surface Redox Electrochemistry, Membrane and Cation Effects ................................................. 63

5.3.3 Continuous GDE Electrocatalysis and FDCA Product Space Time Yields .................................. 65

5.4 Conclusion ........................................................................................................................... 73

III

6 Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of 5-

Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics ........................................... 74

6.1 Abstract ................................................................................................................................ 74

6.2 Introduction .......................................................................................................................... 75

6.3 Results and Discussion ........................................................................................................ 77

6.3.1 Catalyst Synthesis ......................................................................................................................... 77

6.3.2 Characterizations ........................................................................................................................... 78

6.3.3 Morphology and Composition Transformations Under Reaction ................................................. 79

6.3.4 Catalytic Testing of Rotating Disk Electrodes .............................................................................. 81

6.3.5 Catalytic Testing of Stationary Foam Electrodes .......................................................................... 82

6.3.6 Alkaline Exchange Membrane HMF Electrolyzer Cell Tests ....................................................... 85

6.3.7 AEM HMF Electrolyzer Stability Tests ........................................................................................ 88

6.3.8 Mechanistic Discussion ................................................................................................................. 90

6.3.9 Conclusion .................................................................................................................................... 91

7 Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer ....................................................................... 92

7.1 Abstract ................................................................................................................................ 92

7.2 Introduction .......................................................................................................................... 93

7.3 Results and Discussion ........................................................................................................ 96

7.3.1 A Continuous Flow HMFOR Electrolyzer Cell ............................................................................ 96

7.3.2 Catalyst Layer Optimization for CO2 Reduction in a BPM Configuration ................................... 97

7.3.3 Paired CO2RR / HMFOR Electrocatalysis Using a Single BPM Electrolyzer Cell .................... 100

7.3.4 Paired BPM Electrolyzer Cell Performance ................................................................................ 101

7.4 Conclusion ......................................................................................................................... 106

8 Conclusion and Perspective ......................................................................................... 107

8.1 Perspective ......................................................................................................................... 109

9 References ..................................................................................................................... 111

10 Appendix ....................................................................................................................... 127

10.1 Supporting Information Chapter 4 ..................................................................................... 127

10.2 Supporting Information Chapter 5 ..................................................................................... 144

10.3 Supporting Information Chapter 6 ..................................................................................... 164

10.4 Supporting Information Chapter 7 ..................................................................................... 183

10.5 List of Figures .................................................................................................................... 202

10.6 List of Figures (Appendix) ................................................................................................. 206

10.7 List of Tables ..................................................................................................................... 217

10.8 List of Tables (Appendix) .................................................................................................. 217

10.9 List of Abbreviations ......................................................................................................... 219

10.10 List of Publications ............................................................................................................ 221

Zusammenfassung

Der durch Menschen verursachte fortschreitende Klimawandel und die in den letzten Jahren

nicht ausreichenden Maßnahmen diesem zu entgegnen, sind Gründe für eine aktuell

beschleunigte Energiewende. Allein durch den Ausbau von erneuerbaren Energiequellen wie

Windräder, Solarparks und Wasserturbinen wird diese jedoch nicht gelingen. Elektrolyseure,

die sowohl den Strom aus erneuerbaren Energien in Molekülen wie Wasserstoff speichern

können, als auch das Potenzial besitzen, erneuerbare Energie zu nutzen, um bisher

petrochemisch basierte großtechnische Prozesse zur Synthese von verschiedensten

Chemikalien zu ersetzen, werden dabei eine zentrale Rolle einnehmen. Insbesondere

nachhaltige Biomasse Moleküle wie 5-Hydroxymethylfurfural (HMF), die in der Lage sind, das

Gesamt-Zellpotential und somit den Energieverbrauch von Elektrolyseuren drastisch zu senken

und gleichzeitig wertvolle Produkte liefern, erweitern das Potenzial von Elektrolyseuren für die

Zukunft. Deswegen beschäftigt sich diese Arbeit, neben Scale-up Problemen in

elektrochemischen Zellen, mit der elektrochemischen Umsetzung von HMF in

Anionenaustauschmembran-Wasserelektrolyseuren. Im Speziellen wird die Oxidation des

HMF Moleküls durch Nickel-basierte geschichtete Doppelhydroxide katalysiert. Dabei konnte

mit dem hergestellten NiFe(-Cl-)-LDH@NF Katalysator 100% HMF vollends selektiv zum

gewünschten Produkt, 2,5-Furandicarbonsäure (FDCA) umgesetzt werden. Zusätzlich zeigte

der Katalysator eine stabile Performance über mehrere Stunden. Des Weiteren konnte gezeigt

werden, dass neben der Hydrierung von HMF, im stark alkalischen pH-Bereich, ebenfalls die

Hydrogenolyse möglich ist. Dafür wurden Kupferoxid basierte Mischmetalloxide verwendet.

Die, unter den Reaktionsbedingung ebenfalls teilweise reduzierten Kupferoxid basierten

Katalysatoren und insbesondere CuO/Fe2O3/CF zeigten hohe Selektivität gegenüber des

Hydrogenolyse Produkts 5-Methylfurfurylalkohol (MFA). Außerdem war es möglich eine

Korrelation zwischen der HMF Reduktionsproduktselektivität und der Wasserstoff Evolution

Reaktion Rate zu ermitteln. Abschließend konnten nicht nur die HMF Oxidationsreaktion

(HMFOR) und Reduktionsreaktion (HMFRR) in einer Elektrolysezelle kombiniert werden,

vielmehr gelang dies auch bei der Kombination von der HMFOR und der CO2

Reduktionsreaktion. Bei der Etablierung der HMFOR in einen CO2-Elektrolyseur wurde das

HMF System, unter Nutzung einer bipolaren Membran, zu einem kontinuierlichen, bei hohen

Stromdichten betriebenen System weiterzuentwickeln.

Somit liefert diese Arbeit grundlegende Einblicke in die Materialwissenschaft und die

Elektrokatalyse von HMFOR und HMFRR sowie in die Implementierung dieser Prozesse als

Anode und Kathode von Membranelektrolyseuren, sowohl in Kombination mit konventionellen

Gegenreaktionen wie OER und HER als auch in Kombination mit einer weniger

konventionellen Gegenreaktion wie der CO2RR. Zusammengenommen schlagen die

Ergebnisse dieser Dissertation eine Brücke zwischen Labor- und industriellen Ansätzen für die

elektrochemische Umwandlung von HMF und damit prinzipiell von Biomasse.

VII

Abstract

The progressive climate change caused by humans and the insufficient counteractions taken in

recent years are reasons for an urgent energy transition. However, this will not succeed solely

through expanding renewable energy sources such as wind turbines, solar parks, and water

turbines. Electrolyzers, which can store electricity from renewable energy in molecules such as

hydrogen, but also have the potential to use renewable energy to replace petrochemical-based

large-scale processes for synthesizing various chemicals, will play a central role. In particular,

sustainable biomass molecules such as 5-Hydroxymethylfurfural (HMF) can drastically reduce

the overall cell potential. Thus, the energy consumption of electrolyzers, while providing

valuable products, expands the potential of electrolyzers for the future. For this reason, in

addition to scale-up problems in electrochemical cells, this work deals with the electrochemical

conversion of HMF in anion exchange membrane water electrolyzers. Specifically, the

oxidation of HMF is catalyzed by nickel-based layered double hydroxides. Here, the NiFe(-Cl-

)-LDH@NF catalyst showed 100% HMF conversion and 100% selectivity towards the desired

product 2,5-Furandicarboxylic acid (FDCA), with high stability. Furthermore, it was shown

that, in addition to the hydrogenation of HMF, hydrogenolysis is also possible under strongly

alkaline conditions. The copper oxide-based mixed metal oxide catalysts are partially reduced

to high-active metal oxide-derived catalysts under the investigated reaction conditions. In

particular, CuO/Fe2O3/CF showed high selectivity towards the 5-Methylfurfurylalcohol (MFA)

hydrogenolysis product. Moreover, it was possible to establish a correlation between the HMF

reduction product selectivities and the hydrogen evolution reaction rate. Finally, not only the

HMF oxidation reaction (HMFOR) and reduction reaction (HMFRR) could be combined in one

electrolysis cell, but also the HMFOR and CO2 reduction reaction could be combined. In

establishing HMFOR in a CO2 electrolyzer, it was possible to develop the HMF system further,

using a bipolar membrane, into a continuous system operating at high current densities.

Thus, this work provides fundamental insights into the materials science and electrocatalysis of

HMFOR and HMFRR, as well as the implementation of these processes as anode and cathode

of membrane electrolyzer devices, combined with conventional counter reactions, such as the

OER and HER, as well as in combination with a less conventional counter reaction such as the

CO2RR. Together the results of this dissertation build a bridge between laboratory and industrial

approaches for the electrochemical conversion of HMF and, therefore, of biomass in principle.

Introduction

1 Introduction

The increasing global demand for energy and the fact that finite and environmentally harmful

fossil fuels are meeting this demand has led to environmentally friendly energy alternatives for

decades. Driven by the need for more sustainability, energy security, and independence, the

global investment in renewables like wind, solar, and hydropower rose for the fourth

consecutive year to 366 billion US$.1 As a result, the share of sustainable energy sources in the

global energy supply increased by 4% between 2009 and 2020 (Figure 1). In addition, the share

of renewables in electricity generation is more than a quarter worldwide and 41% for

Germany.1,2

Despite all this, electrical energy accounts for only a fraction (17%) of the worldwide total used

energy. The heat and transport sector represents the largest share, of which only 11.2% and

3.7% are covered by renewable energies (2019). Regardless of the sector, the percentage of

fossil fuels remains high. So in 2021, the increasing energy demand was again met mainly by

fossil fuels, leading to record emissions of environmentally harmful carbon dioxide.1

The decarbonization of energy and, thus, the reduction of CO2 emission was targeted in the

2015 Paris Agreement to limit the global temperature increase to 1.5°C, stop global warming,

and set the course for a future worth living on Earth.3 Even if the climate crisis cannot be solved

exclusively by renewable energies, as there are several other industrial processes with intensive

CO2 emissions, such as the production of cement, renewable energies remain one of the critical

factors whose development should be prioritized, along with, for example, carbon capture and

storage.4,5

Introduction

Figure 1: Share of renewable energy from the worldwide total energy and the global and German electrical energy.1,2

However, the energy transition from fossil fuels to renewable energy sources is not only

accompanied by installing solar plants and wind turbines. Instead, the adaption and expansion

of the energy infrastructure and short- and long-term energy storage and conversion play

comparably important roles and indicate the complexity of the energy transition. In more detail,

electrical energy storage and conversion technologies (besides non-electrical storage

technologies) ensure a balanced energy supply, frequency control, and energy reserves.

Introduction

The most common ones today are batteries, fuel cells, and electrolyzers. In contrast to batteries,

which can be used as a direct energy storage technology in cars or cell phones, fuel cells are

essential for applications that cannot yet be directly electrified, such as aircraft, ships, long-

distance transport, or seasonal energy storage. Therefore, electrolyzers ideally obtain energy

from renewable sources and store it in chemical molecules. These molecules (e-fuels) can then

be feed in reverse into a fuel cell, which uses the energy stored in the molecules to drive an

engine, for example. In fact, in addition to e-fuel production by electrolysis, many different

molecules and compounds can be produced directly or via so-called platform molecules. With

hydrogen, for example, not only molecularly stored energy can be made usable in a hydrogen

fuel cell, but hydrogen can also be used to produce synthesis gas, and thus ammonia or

methanol.6

Summarized, the process of electrolysis, powered by green energy from renewable sources, for

the sustainable production of e-fuels, platform molecules, and in particular, water electrolyzers,

will play a central role in the energy transition and combating climate change and is therefore

addressed in this work.

1.1 Electrochemical Water Splitting

Applying a sufficiently high cell potential to a water electrolyzer decomposes water into

hydrogen at the cathode and oxygen at the anode. Electrodes are classically surrounded by

liquid electrolytes to ensure charge transfer and balance and are conducted to an outer electrical

circuit and a power supply. Moreover, a diaphragm or a membrane separates the electrodes into

two chambers to avoid explosive mixing of hydrogen and oxygen.7 Today, the three most

common types of water electrolyzers are the Alkaline Water Electrolyzer (AWE), the Proton

Exchange Membrane Water Electrolyzer (PEMWE), and the Anion Exchange Membrane

Water Electrolyzer (AEMWE), demonstrated in Figure 2. They differ essentially in the choice

of electrolyte, with AWE and AEMWE using an alkaline electrolyte (mostly KOH) and

PEMWE using an acidic electrolyte, as well as in the choice of the separator, which is a porous

diaphragm in the case of AWE and a membrane in the case of PEMWE and AEMWE. The

combination of electrodes and membranes is referred to as a Membrane Electrode Assembly

(MEA). The preferred direct contact between electrodes and the membrane in MEAs minimizes

the distance between the electrodes and, thus, the voltage drop caused by ohmic losses.

Introduction

It ensures that the gas evolution occurs primarily on the back of the electrodes, which also

brings overpotential advantages. This further development is referred to as the zero-gap

configuration.8

Figure 2: Schematic illustration of an alkaline water electrolyzer (AWE), proton exchange membrane water electrolyzer

(PEMWE), and anion exchange membrane water electrolyzer (AEMWE).

The specifications of each electrolysis type result in a different selection of suitable catalysts.

In the PEMWE, for example, noble metal catalysts such as Pt are used at the cathode and IrO2

or RuO2 at the anode. Under alkaline conditions, on the other hand, transition metals, transition

metal alloys (AWE/AEMWE), and transition metal oxides (AEMWE) can be used as electrodes

(mostly nickel-based).9 Although AEMWE does not have the same technology stage as the

mature and commercialized AWE and PEMWE, it combines the advantages of these

technologies. The AEMWE enables the corrosion-free use of low-cost non-platinum group

metal (non-PGM) catalysts and flow field materials at differential pressure and high current

densities. Moreover, using an additional circulating electrolyte (e.g., KOH) reduces the ohmic

resistance and positively affects the reaction kinetics.10 Nevertheless, membrane stability, lack

of ionomers, and water management present challenges that keep AEMWE at a laboratory

stage.11

For a more fundamental understanding and improvements in catalytic activity and stability,

separate half-cell investigations of the cathodic Hydrogen Evolution Reaction (HER) and the

anodic Oxygen Evolution Reaction (OER) are still the focus of the AEMWE.

Introduction

1.1.1 Alkaline Hydrogen Evolution Reaction (HER)

In alkaline electrolytes, the kinetics of HER is much more inhibited than under acidic

conditions, especially if PGM catalysts are avoided.12,13 When comparing the HER reaction

equations in acidic and alkaline conditions (Figure 2), it is evident that under alkaline

conditions, water must first be dissociated in order to subsequently combine protons to form

hydrogen. The splitting of water at an active site of the catalyst (*) to OH- and an adsorbed

proton (Had) is called the Volmer step (Equation 1) and is usually the slowest and, thus, the

rate-determining step in the alkaline HER. The subsequent combination of Had and water, or

two Had, to form hydrogen is called Heyrovsky (Equation 2) and Tafel step (Equation 3),

respectively.

Volmer: ∗+H2O + e−→Had +OH− (Equation 1)

Heyrovsky: Had + H2O + e−→H2+OH− (Equation 2)

Tafel: Had + Had →H2 (Equation 3)

To balance the charge of the 2-electron process, the OH- ions formed in the Volmer and

Heyrovsky step are passed through the anion exchange membrane from the cathode to the

anode. That means, for AEMWE, a suitable HER catalyst needs to split water on the one hand

and recombine protons on the other hand.14-16

Even though the HER activity of pure metallic Ni is relatively low with respect to noble metal

Pt benchmark catalysts, in combination with other transition metals, supported on carbon, or as

mixed metal/metal oxide, Ni-based catalysts show promising activities towards alkaline

HER.17-19 Especially, the combination of Mo and Ni on a high surface area Ni foam substrate

by Zhang et al. show superior HER activity.20

In summary, high HER activity in alkaline medium for Ni-based catalysts can be attributed to

synergistic effects between the catalyst and the support or metal/metal oxide mixed phases.9

Introduction

1.1.2 Alkaline Oxygen Evolution Reaction (OER)

In contrast to the acidic oxygen evolution reaction, in which water provides oxygen, in the

alkaline oxygen evolution reaction, oxygen is obtained from hydroxide anions (OH-). Despite

this, with its sluggish kinetics, the alkaline OER remains the rate-determined reaction in

alkaline water splitting.11 To generate an oxygen molecule, four reaction steps involving 4 OH-

are needed (Equation 4-7).21

∗ + OH−→OHad +e− (Equation 4)

OHad + OH−→Oad +H2O+e− (Equation 5)

Oad + OH−→OOHad +e− (Equation 6)

OOHad + OH−→O2+H2O+e− (Equation 7)

In the first step, OH- is adsorbed at an active site (*) of the catalyst releasing one electron

(Equation 4, Figure 3 a), then the adsorbed OH (OHad) reacts with an additional OH- to form

Oad, H2O, and one electron (Equation 5, Figure 3 b). In the third step, Oad reacts with OH- to

form OOHad (Equation 6, Figure 3 c), which forms oxygen and water in the final fourth step

(Equation 7, Figure 3 d-e).

Introduction

Figure 3: Supposed alkaline oxygen evolution reaction (OER) mechanism on a metal substrate (a-e), with the change in

Gibbs free energy (ΔG) for every reaction step.

Thus, a total transfer of four electrons is needed for the OER. The search for suitable OER

catalysts is essential to minimize the kinetic and thermodynamic barriers of the OER reaction

steps, reduce the overall cell potential and thus ensure efficient hydrogen production in water

electrolyzers. The most promising non-noble transition metal OER catalysts today are Ni and

Co-based.9 In particular, the combination of Ni, Fe oxyhydroxides, or layered double

hydroxides (LDHs) have been intensively investigated in recent years due to promising

activities comparable to those of IrO2 benchmark catalysts.22-27 Even though NiFe-LDHs show

superior OER activity, for example, the leaching of Fe still poses problems.28

Based on the previous challenges, another option is to replace the OER with an alternative

reaction that can be operated at a lower overpotential and provide valuable products. The

electrochemical conversion of biomass meets these requirements with a sustainable approach.

Introduction

1.2 Electrochemical conversion of 5-Hydroxymethylfurfural

The electrochemical conversion of renewable, carbon-neutral, and abundant biomass resources

to produce high-value carbon-based products is attracting increasing attention. Not only can

they harness electricity from renewable energy sources and replace unfavorable counter

reactions such as OER, but the fact that previously only petrochemically manufactured products

could be produced in a climate-friendly manner demonstrates the future importance of the

electrochemical conversion of biomass. Lignocellulosic biomass, for example, after

biorefining, opens up for a variety of platform chemicals like polyols, carboxylic acids, amino

acids, aromatics, and furans.29 Very prominent candidates for the electrochemical conversion

of biomass are the oxidation of glycerol to, e.g., glyceraldehyde, glyceric acid, formic acid or

oxalic acid,30-33 the electrochemical valorization of levulinic acid to, e.g., valeric acid,

γ-valerolactone and further via Kolbe to octane or via Hofer-Moest to 1-butanol,34-37 and the

conversion of 5-Hydroxymethylfurfural (HMF) to, e.g., 2,5-Bis(hydroxymethyl)furan

(BHMF), 2,5-Dimethylfuran (DMF) and 2,5-Furandicarboxylic acid (FDCA). In particular, the

valorization of HMF via di- and polymerization, ring-opening, ring-expansion, hydrogenation,

hydrogenolysis, and oxidation leads to valuable products in the medicine, pharmaceutical,

agrochemical, food, plastics, and fuel industry.38-40 Figure 4 shows the reaction pathways of

HMF for the oxidation products (yellow) and the reduction products. The reduction of HMF

can be divided into di- and polymerization (black), ring-opening (dark blue), hydrogenation

(turquoise), and hydrogenolysis (blue). Due to their excellent utility, oxidation, hydrogenation,

and hydrogenolysis products are the focus of current scientific investigations. However, it is

essential to mention that in addition to controlled di- and polymerization, with increasing

alkalinity and HMF concentration, uncontrolled polymerization can also occur. The resulting

so-called humins are unreactive and unusable and represent a challenge for the long-term

continuous conversion of HMF.41,42

Introduction

Figure 4: 5-Hydroxymethylfurfural (HMF) oxidation and reduction reaction pathways. Poly- and Dimerization (black):

undesired side reaction of n HMF molecules and n Protons/electrons. Reductive Ring opening (dark blue): opening of the furan

ring with 6e- and 6H+. Hydrogenation (blue): conversion of HMF to BHMF with 2e- and 2H+. Hydrogenolysis (blue):

conversion of HMF to MF with 2e- and 2H+, conversion of BHMF or MF to MFA with 2e- and 2H+, and conversion of MFA

to DMF with 2e- and 2H+. Oxidation (yellow): conversion of HMF over HFCA or FDA to FFCA and FDCA (6e- and 6OH-).

Carbon atoms (black), Oxygen Atoms (red), and Hydrogen Atoms (white). This Figure is reproduced from Chapter 6.43

1.2.1 Electrochemical Oxidation of HMF

In principle, HMF can be described as a furan motive with a hydroxymethyl group at position

five and an aldehyde group at position two. The electrochemical oxidation of HMF leads to

four main products (Figure 4, yellow): 2,5-Furandialdehyde (FDA),

5-(Hydroxymethyl)-2-furancarboxylic acid (HFCA), 5-Formyl-2-furancarboxylic acid

(FFCA), and 2,5-Furandicarboxylic acid (FDCA). The highest demand among these four

products has FDCA because it can serve as a precursor for polyethylene furandicarboxylate

(PEF), a biologically sustainable alternative to the world's most widely used plastic molecule,

PET.44,45 Two hydroxide ions in combination with a 2-electron transfer are required for each

oxidation step. Although the oxidation of HMF to FDCA (6-electron transfer reaction) requires

two more electrons than OER, it is less thermodynamically and kinetically hindered and

therefore requires a lower overpotential.46,47 Mechanistically, HMF is first oxidized to either

FDA (oxidation of the alcohol group) or HFCA (oxidation of the aldehyde group) and then

further to FFCA and FDCA.

Introduction

Why FDA or HFCA is formed preferentially is not yet fully understood, making it difficult to

describe the exact mechanism. However, trends can be given for each product depending on

the reaction conditions. FDA is preferentially formed at a pH < 13, in areas of higher potential

and with noble metal catalysts such as Pt and Ru. Whereas the formation of HFCA is preferred

at pH ≥ 13, in areas of lower potential and with noble (Au and Pd) and non-noble metal

catalysts.48-53 It is important to say that these are only trends, and the actual HMF oxidation

reaction (HMFOR) mechanism is highly dependent on the reaction setup, catalyst structure, and

design.54 In particular, the specific catalyst design is probably the determining factor, as it

defines, among other things, the optimal potential window for each product, how strongly HMF

and OH- are adsorbed on the catalyst, and whether HMF is oxidized directly or indirectly.55-58

Indirect oxidation of HMF requires potential ranges in which the catalyst is first oxidized to a

higher valance state, and HMF or an intermediate is oxidized by the pre-oxidized catalyst. Thus,

the catalyst is reduced back to the initial stage. However, high potential ranges promote not

only the amount of indirect oxidation but also the production of FFCA and FDCA.59

Unfortunately, it is not yet evident to what extent the increased FFCA and FDCA production is

related to indirect oxidation or mainly caused by the increased charge per time. Especially the

production of FFCA is complex because the selective potential window is very narrow. Most

of the investigated catalysts show high selectivity for FDA and HFCA at low potentials, and

usually, HMF is directly oxidized to FDCA at higher potentials.52

There is a wide range of active catalysts for the electrochemical oxidation of HMF to FDCA,

including metals, alloys, oxides, hydroxides, oxyhydroxides, metal and covalent organic

frameworks (MOFs and COFs), phosphides, borides, nitrides, selenides, and sulfides. Besides

a few noble metal catalysts, the most frequently used catalysts are Ni-, Co-, and Cu-based.

However, Chadderdon et al.52 and Park et al.60 show that the complete oxidation to FDCA is

challenging with single noble metal catalysts, so they used Pd/Au noble metal alloys. Although

both were able to produce FDCA at potentials below 1 VRHE, only Chadderdon et al. achieved

a high FDCA yield and selectivity of over 80%. Further studies with non-noble metal catalysts

such as Ni, Co, and Cu show FDCA yields and selectivity close to 100%. Still, they should be

taken cautiously as these results were obtained in potential ranges where the metals are oxidized

to oxides or hydroxides.61-64 As a result, the active sites are no longer metallic, and catalysis is

likely driven mainly by an indirect mechanism.

Introduction

Moreover, it is likely that under these conditions (higher potential range and pH ≥ 13), the

oxides and hydroxides react further to oxyhydroxides, which are known to be highly active

electrooxidation catalysts. In many studies, the formation of Ni and Co oxyhydroxides (NiOOH

and CoOOH), starting from oxides or hydroxides, is observed and confirmed as active species

for the oxidation of HMF.65-71 Also originating from hydroxides and probably using

oxyhydroxides as active species but exhibiting particularly high HMF activities due to their

structural and synergistic effects are Ni, Cu, and Co LDHs.72-75 However, studies directly using

oxyhydroxides demonstrate partly lower activity towards HMF oxidation, suggesting a

beneficial effect of in-situ formed oxyhydroxides.54,59,76

Another group of catalysts for the electrochemical oxidation of HMF to FDCA is the

intercalation of Ni and Co into organic frameworks. MOFs and COFs show similar activities to

oxides and hydroxides but do not necessarily form oxyhydroxides as active sites. Rather, it is

believed that the metals in MOFs and COFs are oxidized to higher valence states, determining

the HMF activity.77-79 In despite of the high activity, the stability of MOFs and COFs in acidic

and alkaline environments is considered problematic.80

Likewise, high activities can be achieved by modifying Co and Ni with N, B, Se, S, and P.

However, the work of You et al., Barwe et al., and Zhang et al. confirms that the active sites

still have a high valence and oxyhydroxide character. Furthermore, a disappearance of S and P

could be observed by Zhang et al..81-84 Although the contribution of the modifications is not yet

fully understood, it is believed that a catalytic positive d-band shift can be achieved, and

heterojunction can lead to activity boosting oxygen vacancies.85-88

In summary, over the last 15 years, much research has been done in electrochemical HMF

oxidation, but still, there is a lack of fundamental understanding. Most recent work shows high

HMF conversion, FDCA yield, and selectivity exclusively at laboratory scale in a very narrow

potential range. Moreover, comparisons between studies are often difficult due to different

reaction conditions, such as HMF and KOH concentration, reaction time, and electrode surface

area. In particular, the choice of 1 M KOH could make a scale-up to an industrial approach

difficult as the formation of humins and the OER become more dominant.

Introduction

1.2.2 Electrochemical Reduction of HMF

In contrast to the oxidation of HMF, the reduction received far less attention over the past years.

In fact, the potential advantage that HMF oxidation has over OER cannot be created with the

HER competing HMF reduction reaction (HMFRR). Nevertheless, the electrochemical

reduction of biomass molecules as HMF represents a meaningful, sustainable opportunity for

future organic synthesis. The main products of the reduction process are either hydrogenation

(turquois) to 2,5-Bis(hydroxymethyl)furan (BHMF) or hydrogenolysis (blue) to

5-Methylfurfural (MF), 5-Methylfurfurylalcohol (MFA) and 2,5-Dimethylfuran (DMF)

(Figure 4). Hydrogenation describes the addition of H-atoms to the carbonyl oxygen, i.e., the

reaction of an aldehyde to an alcohol. And hydrogenolysis describes the reaction between H-

atoms and an aldehyde or alcohol group with water cleavage to CHx. For each reduction step,

regardless of whether hydrogenation or hydrogenolysis occurs, 2H+ and 2e- are required.

Although DMF would probably have the most significant benefit because it can be used as a

fuel, the most common product in research is BHMF, which can be used as a polyurethane foam

and polyester monomer.89 However, this is not by choice but due to the fact that hydrogenolysis,

especially of an alcohol group, is a major challenge in electrochemistry.90-92 For this reason, the

mechanism by which MFA and DMF are formed electrochemically has not yet been

conclusively clarified. However, several studies show that the reduction to MFA and DMF can

occur preferentially or exclusively under acidic conditions.91,93,94 An explanation can be given

by the high proton concentration in acidic electrolytes. The needed protons for HMF reduction

can thus be obtained directly from the electrolyte (Eley-Rideal mechanism). In an alkaline case,

the protons must first be formed by water splitting (Volmer step). Then the adsorbed protons

need to be used for HMF reduction (Langmuir-Hinshelwood mechanism) instead of forming

H2 (Tafel step). In summary, the catalyst must do both, provide protons and selectively reduce

HMF without forming hydrogen, which enormously increases the requirements of the catalyst

under alkaline conditions. For these reasons, among others, most studies on electrochemical

HMF reduction have been conducted in acidic to neutral electrolytes.

A study in strongly acidic electrolyte for various solid metal catalysts was investigated by Kwon

et al..95 It shows that metals such as Fe, Ni, Cu, and Pb more likely produce BHMF, while Pd,

Pt, Al, Zn, In, and Sb can produce 2,5-Dimethyl-2,3-dihydrofuran (DMDHF). In addition,

metals such as Co, Ag, Au, Cd, Sn, and Bi produce BHMF and DMDHF.

Introduction

This study's unique feature is the simultaneous reduction of functional groups and the furan

ring to produce DMDHF. As a direct comparison, the analysis of Lee et al. shows that under

neutral conditions, In, Cd, and Ag produce BHMF, Cu, Ni, Co, and Fe produce small amounts

of MFA, whereas W and Ti promote humin production.91 Moreover, Density Functional Theory

(DFT) calculations exhibit weak HMF binding for BHMF-producing metals, moderate HMF

binding for Cu, Ni, Co, and Fe, and strong HMF binding for W and Ti. Further studies under

neutral conditions based on Cu and Ag catalysts show a clear product preference favoring

BHMF.96-101 To date, the only selective electroreduction of HMF to DMF in strongly acidic

electrolyte has been demonstrated by Nilges et al. and Zhang et al. over metallic copper and

copper-nickel alloy catalysts.102,103 There are also studies showing reductive ring opening

(Figure 4, dark blue) over a Zn-based catalyst by Roylance et al. and controlled dimerization

of HMF to 5,5-Bis(hydroxymethyl)hydrofuroin (BHH) (Figure 4, black) over carbon-based

catalysts by Kloth et al..104,105

In summary, HMFRR research is still quite rudimentary. Even though selective hydrogenation

of HMF to BHMF and even hydrogenolysis to DMF was possible under strongly acidic

conditions, hydrogenolysis of HMF to MFA or DMF in alkaline electrolytes is considered to

be very unlikely.

Introduction

1.3 Motivation, Scope, and Outline

In order to stop climate change and at the same time meet the increasing energy demand, the

development of renewables is essential for the future. Therefore, it needs both the generation

and distribution of green energy and environmentally friendly energy storage and conversion

technologies. Electrolyzers, particularly water electrolyzers, will play a central role in energy

storage and conversion. To exploit the full potential of this technology, it is important to

generate high efficiencies combined with valuable products at both electrodes. Using biomass

such as HMF in water electrolyzers, particularly the replacement of the OER by the HMFOR,

not only lowers the required cell potential but also provides valuable products on the anode. In

addition to the cathodic green hydrogen production, HMFOR can be coupled with other

reactions, such as reductive organic synthesis (Figure 5) or CO2 reduction reaction (CO2RR).

Figure 5: Illustration summary of this work's motivation, idea, and scope. Using renewable energy sources for the

electrochemical conversion of HMF at the anode (blue) and cathode (orange) to valuable products.

Introduction

Up to now, current research on the electrochemical conversion of HMF has shown high

activities, but only on a laboratory scale. The scale-up to electrolyzer cells as well as the

industrial applicability, given by high and stable conversion rates at high current densities in a

continuous operation mode, are still lagging behind. The present thesis aims to explore,

characterize, and understand the electrochemical oxidation and reduction of HMF and to bridge

the gap between laboratory scale and an industrial approach and can be divided into four phases.

The first phase deals with the problems of scaling up a known OER catalyst system. Therefore,

the NiFe-LDH model catalysts are modified via anion exchange and tested on a laboratory scale

and in an AEMWE system.

The second phase describes the implementation of HMFOR in the established AEMWE system

from phase one. It is further crucial to create a highly active HMFOR system that delivers both

high HMF conversion and FDCA selectivity. At the same time, this high performance has to be

stable over a longer time.

The third phase focuses on investigating HMFRR in a highly alkaline environment and finally

on combining HMFOR and HMFRR in one AEMWE system. Contrary to the current opinion,

it should be shown that it is possible to carry out hydrogenolysis in an alkaline environment.

For this purpose, bimetallic copper-based oxides are used as catalysts, which should be able to

generate protons and selectively reduce HMF.

In the fourth and final phase, the focus is on coupling HMFOR with the CO2 reduction reaction.

In contrast to the previous phases, a bipolar membrane was used to exploit the potential of both

previously optimized reactions fully. Furthermore, continuous operation at high current

densities and high activities for the HMFOR/CO2RR electrolyzer is aimed.

In summary, besides a deeper understanding of the HMFOR and HMFRR, this work illustrates

the possibilities as well as the challenges of HMF in alkaline water electrolyzers.

Theoretical Background

2 Theoretical Background

Chapter 2 considers the theoretical background of electrochemistry by giving a deeper look into

the fundamentals, thermodynamics, and kinetics of electrochemistry (2.1, 2.2, 2.3). In addition,

subchapter 2.4 considers organic electrochemistry's historical and theoretical background. The

elaborations in this chapter are mainly based on the textbooks by Patrick T. Moseley, Thomas

F. Fuller, Allen J. Bard, Tom Smolinka, Franco Babir, Agata Godula-Jopek, Derek Pletcher,

Peter Atkins, Toshio Fuchigami and Ib Chorkendorff.7,106-116

2.1 Fundamentals of Electrochemistry

Electrochemical reactions can be described as heterogeneous catalytic processes with an

additional driving force, the electrode potential. The control of charge transferred to or from an

electrode allows electrochemistry to avoid, for example, high temperature or high pressure, in

contrast to classical heterogeneous catalysis. Fundamentally, an electrochemical system

consists of an anode, where oxidation occurs, and a cathode, where reduction occurs. Here, an

external circuit and the electrolyte provide the charge transfer between the electrodes. The

charge transfer from the bulk electrolyte to the electrode-electrolyte interface plays a significant

role in electrochemistry and is necessary to understand electrochemical reactions. Once

electrodes are polarized, ions rearrange within hundredths of a second to balance the charge,

forming the double layer (Figure 6). The first description of the double layer was given by H.

Helmholtz in 1853 and is called the outer Helmholtz plane (OHP).117 The OHP is defined by

solvated ions lined up at the electrode surface at a distance determined by the hydration shell.

The Helmholtz model was extended by Gouy and Chapman (1909-1913). In contrast to the

rigid structure of the Helmholtz layer, they established a diffuse double layer, which considers

the thermal motion of the ions. To consider the internal structure and the rigidity of the layer,

Otto Stern combined the outer Helmholtz plane and the diffuse double layer of Gouy and

Chapman in 1924.118 An extension to the Stern model was made in 1942 by David Grahame.119

He distinguished between solvated ions near and bare ions at the electrode-electrolyte interface

and thus introduced the inner Helmholtz plane (IHP).112,120 As shown in Figure 6, there is a

potential gradient over the electrical double layer which flattens out towards the bulk, where

the potential stays constant.

Theoretical Background

Because of the Helmholtz layer's capacitor character, the potential varies linearly (green area).

Whereas in the Gouy-Chapman plane (yellow), the potential varies exponentially due to the

Poisson-Boltzmann equation until it becomes constant in the bulk.

Figure 6: Scheme of the electrochemical double layer including the Helmholtz (inner (IHP) and outer Helmholtz plane

(OHP)), Grahame, and Stern models.

Although charge neutrality is given by moving ions and electrons, in practice, electrochemical

reactions such as electrolysis are usually studied by the movement of electrons from the anode

to the cathode via an external circuit. The number of electrons or charge Q that crosses the

external circuit can be measured and is referred to as the current i. In principle, the measured

current provides information on the chemical change of and at the electrode. Furthermore,

Faraday’s first law established a direct proportionality between the charge Q and the number of

converted mols n (Equation 8). Here, z is the number of electrons needed for the considered

reaction, and F is the Faraday constant (96485 C mol-1).

Q=∫idt

0=znF (Equation 8)

Theoretical Background

2.2 Thermodynamics of Electrochemistry

Controlling and understanding an electrochemical reaction presupposes controlling and

understanding the energy transfer within a defined system and, thus, compliance with the laws

of thermodynamics. By changing the electrochemical potential E, the energy of electrons either

increases for more negative potentials or decreases at more positive potentials. Thus, electrons

can reach energy levels high enough to move from the electrode to the solution (reductive

current) or low enough to move from the solution to the electrode (oxidative current).

Consequently, species A+ is reduced to B, or the other way around, species B is oxidized to A+.

Depending on the reaction, a specific amount of heat is released to or detracted from the reaction

environment. This heat is called enthalpy H and is negative for reactions where heat is released

(exothermic) or positive when heat is detracted (endothermic). For electrochemical reactions

under standard conditions (25°C and atmospheric pressure), Hess’s law describes the standard

reaction enthalpy as the difference in the enthalpy of formation from the products and the

reactants (Equation 9).

∆Hreac

0=∑∆Hp

0−∑∆Hr

0 (Equation 9)

From a purely thermodynamic point of view, the enthalpy is defined as the sum of the internal

energy U and the volumetric work pV of the system (Equation 10). It is described by the first

law of thermodynamics.

H=U+pV (Equation 10)

If the pressure is constant, the change in enthalpy can be expressed as the change in heat Δq

(Equation 11).

∆H=∆q (Equation 11)

Theoretical Background

Furthermore, the enthalpy can be expressed with the Gibbs-Helmholtz Equation (Equation 12).

∆H=∆G + T∆S (Equation 12)

Here, the change in enthalpy depends on the change of the Gibbs free energy ΔG, the

temperature T, and the entropy ΔS. As a descriptor for possible microstates of a system, entropy

gives information about the spontaneousness of a system’s change of states. The change in

entropy is described by the second law of thermodynamics and depends on the change in heat

and temperature (Equation 13).

∆S=∆𝑞

𝑇 (Equation 13)

In summary, the difference between enthalpy and entropy change (ΔG) provides information

on the spontaneousness of a system (Equation 12). If ΔG is negative, the process is exergonic

(spontaneous); if ΔG is positive, the process is endergonic (non-spontaneous). For an

electrochemical reaction, the standard Gibbs free energy is inversely proportional to the

electrical work Wel, equivalent to the product of transferred electrons, the Faraday constant, and

the standard potential E0 (Equation 14).

∆G0=−Wel =−zFE0 (Equation 14)

The standard potential can be defined as the maximum electrical work a galvanic reaction can

theoretically afford or the minimum electrical work that an electrolytic reaction theoretically

needs. A standard potential is tabulated for every reversible reaction at equilibrium, under

standard conditions, constant temperature, and pressure without net current flow. Besides, all

standard potentials are referred against a reference, universally against the theoretical standard

hydrogen electrode (SHE) or the normal hydrogen electrode (NHE).

Theoretical Background

The potentials of these two electrodes are very close to each other and differ marginally due to

the consideration of non-ideal effects in the more practical NHE. The NHE is provided with a

Pt electrode in a strongly acidic environment, a proton activity of one at 25°C, and a hydrogen

gas pressure of 1 bar. Furthermore, the standard potential of the hydrogen reaction is set as zero

at the SHE/NHE scale and serves as the reference point for all other standard potentials. One

of the fundamental electrochemical equations, the Nernst equation (Equation 15), expresses the

electrochemical half-cell potential of a specific reaction under consideration of the activity.

E=E0+𝑅𝑇

𝑧𝐹 ln ∏𝑎𝑜𝑥

𝑣𝑜𝑥

∏𝑎𝑟𝑒𝑑

𝑣𝑟𝑒𝑑 (Equation 15)

Here, R is the ideal gas constant (8.314 J K-1 mol-1), a is the activity of the oxidative and

reductive species, and v is the stoichiometry factor. Since the activity depends on the activity

factor fa and the concentration c, for highly diluted solutions where the activity factor is one,

the Nernst equation can also be expressed with concentration instead of the activity. Moreover,

at 25°C, the Nernst equation can be further simplified (Equation 16).

E=E0+0.059 𝑉

𝑧lg ∏𝑐𝑜𝑥

𝑣𝑜𝑥

∏𝑐𝑟𝑒𝑑

𝑣𝑟𝑒𝑑 (Equation 16)

Since the SHE defines, by definition, the potential at pH=0, possible pH shifts need to be

considered (Equation 17). However, for a practical approach, the reversible hydrogen electrode

(RHE) is used today.

ESHE =𝐸𝑆𝐻𝐸

0−0.059 V ∙ pH (Equation 17)

Theoretical Background

In electrochemistry, probably the most famous half-cell reactions are the hydrogen evolution

reaction/ hydrogen oxidation reaction (HER/HOR) and the oxygen evolution reaction/ oxygen

reduction reaction (OER/ORR). These reactions are given in Equation 18 and 19 in an acidic

milieu and correspond to the electrochemical water splitting (Equation 20), where hydrogen

and oxygen evolve, as well as the electrochemical combustion of hydrogen and oxygen in a

hydrogen fuel cell (Equation 21).

HER/HOR: 2H++ 2e−↔H2 E0=0 VSHE (Equation 18)

OER/ORR: 2H2O↔O2+4H++ 4e− E0=1.23 VSHE (Equation 19)

2H2O→H2+O2 E0=−1.23 VSHE (Equation 20)

H2+O2→2H2O E0=1.23 VSHE (Equation 21)

As can be seen from Equation 20, the splitting of water into hydrogen and oxygen is an

endothermic process (E0 < 0 → ΔG0 > 0 → ΔH0 > 0, Equation 14 and 12), which means that

the reaction needs the amount of energy corresponding to the formation enthalpy of water.

However, there are two standard reaction enthalpy values for water formation: higher (HHV)

and lower heating value (LHV). Depending on the aggregation state of the formed water, the

HHV of ΔH0R= 285.8 kJ mol-1 for liquid water or the LHV ΔH0R= 241.8 kJ mol-1 for water in

the vapor state is used. Accordingly, the difference between the HHV and the LHV describes

the enthalpy of vaporization ΔH0V= 44 kJ mol-1 and, thus, the energy required for the aggregate

state change.

A common way to evaluate an electrolyzer or a fuel cell is to calculate its efficiency. One way

to describe the efficiency of such an electrochemical system is the energy efficiency εE. For a

water electrolyzer, the εE is defined as the thermodynamic efficiency to produce hydrogen,

which can either be the HHV or LHV, divided by the real electrical energy used, expressed by

the Gibbs free energy (Equation 22).

εE=Energy output

Energy input =∆HHHV or LHV

∆G (Equation 22)

Theoretical Background

However, this approach to calculating energy efficiency can lead to εE values above 100% for

electrolyzers. A theoretical calculation with ΔHHHV= 285.8 kJ mol-1 and ΔG= 237.2 kJ mol-1

would result in an efficiency of εE= 120.5%.121 The problem is that the heat energy of

ΔH0V= 44 kJ mol-1, which would have to be expanded in addition to the electrical energy for

water splitting, is not considered, leading to unrealistic values. In the practical application of

electrolyzers (or fuel cells), the required energy is supplied by the heat generated when current

flows through the cell, and charges overcome the internal resistance. Besides the energy

efficiency, a voltage efficiency εV of an electrolyzer (Equation 23) can be calculated where it is

equally important to include ΔH0V in the theoretical thermodynamic voltage. By doing so, the

theoretical thermodynamic voltage is 1.48 V instead of the standard potential of 1.23 VSHE,

leading to the practical efficiency.

εV=Thermodynamic voltage

Operating voltage (Equation 23)

For a more reaction-related efficiency, the Faradaic efficiency FE can be calculated. Here, the

distribution of charge (electrons) to different reaction participants and, thus, products are

evaluated (Equation 24).

FE=current of target molecule

total current =𝑖𝑖

𝑖𝑡𝑜𝑡𝑎𝑙 (Equation 24)

Except for FE, it is impossible to run electrolyzers 100% efficiently. This is caused, among

others, by mass transport, charge transport, and kinetic limitations.

Theoretical Background

2.3 Kinetics of Electrochemistry

When considering the thermodynamics of an electrochemical reaction, i.e., the changes in the

energy states of reactants and products, an essential part of catalysis is neglected: the reaction

rate. Regardless of the type of catalysis and reaction, the reaction rate depends on the activation

energy barrier that must be overcome and defines the kinetics of the process. How a catalyst

influences the activation energy barrier and thus the kinetic of a general reaction between

reactant A and B to the product P is illustrated in Figure 7.

Figure 7: General activation energy diagram of the uncatalyzed and catalyzed reaction between precursors A and B to

product P.

In general, a catalyzed reaction can be described in four steps: the transport to (1) and interaction

(adsorption) of the reactants with the catalyst (2), the reaction of A and B to form P (3), and the

separation (desorption) of P (4). As can be seen, the catalyzed reaction (1-4) not only provides

a different reaction path but also has a much smaller activation energy barrier than the

uncatalyzed reaction (5). Nevertheless, it is important to say that the overall change in free

energy of the catalyzed and uncatalyzed reaction remains the same. In other words, catalysis

affects the kinetics and not the thermodynamics. For an electrochemical reaction, steps 2-3

occur via electron transfer from the electrode to the reactants (reduction) or vice versa

(oxidation). Thus, the rate of an electrochemical reaction depends strongly on the release or

acceptance of electrons.

Theoretical Background

For catalysts, the ability to release or accept electrons, in turn, is strongly dependent on the

energy level of the electrons and, therefore, on the orbital in which they are located. The

electronic structure of metals (most commonly used as catalysts in electrochemistry) is

described as bands due to the large orbital overlaps (Figure 8 a). Accordingly, the maximum

electron energy depends on the position of the bands and is called the Fermi level (EF). The

transition metal Fermi level dependency on the electrode potential is given in Figure 8 b-d. As

can be seen in Figure 8, at a negative electrode potential (b), the fermi level is higher than the

fermi level at equilibrium (c), and the electrode releases electrons (electrochemical reduction).

At a positive electrode potential (d) because of a lower fermi level, the electrode accepts

electrons (electrochemical oxidation).

Figure 8: Transition metal Fermi level dependency on the electrode potential.

Theoretical Background

Electron transfer processes are widely assumed to be first-order reactions and thus depend only

on the reaction rate constant (k) and the surface concentration of the reactive species at the

electrode (Equation 25 and 26).

roxidation =kAcred (Equation 25)

rreduction =kCc𝑜𝑥 (Equation 26)

The rate of electrochemical reactions is described by the current density j, which is the current

i divided by the electrode surface area. Accordingly, the reaction's net current density

(Equation 27) is given by the difference between Equation 25, 26, and the first Faradaic law

(Equation 8).

j=zF(kAcred −kCcox) (Equation 27)

Under consideration of the Bell-Evans-Polanyi principle, the rate constant can be written as a

function of the Gibbs free energy and, thus, of the electrode potential (Equation 28 and 29).

kox =k0,A exp(αAzF

RT E) (Equation 28)

kred =k0,C exp(−αCzF

RT E) (Equation 29)

Whereas k0 is the rate constant for the electron transfer at E= 0 versus a reference electrode,

and α is the transfer coefficient. In general, αA + αC = 1, and for common electrochemical

reactions on metal electrodes, αA= αC= 0.5. By merging the previous equations, Equation 30 is

obtained.

j=zF[k0,ox𝑐𝑟𝑒𝑑 exp(αAzF

RT E)−k0,redcox exp(−αCzF

RT E)] (Equation 30)

Theoretical Background

The net current is zero if an electrochemical reaction is considered at equilibrium (Ee). At this

point, the reaction rate is therefore called the exchange current density j0 (Equation 31).

j0=zFk0,Acred exp(αAzF

RT Ee)=zFk0,Ccox exp(−αCzF

RT Ee) (Equation 31)

By substituting the exchange current density and the definition of the overpotential η= E-Ee,

we obtain the Butler-Volmer equation (Equation 32).

j=j0[ exp(αAzF

RT η) − exp(−αCzF

RT η)] (Equation 32)

From Equation 32, it can now be derived that the measured current density is a function of the

exchange current density, the transfer coefficient, and the overpotential. Furthermore, the

Butler-Volmer equation can be simplified if the overpotential is made more positive, jA>>jC

(Equation 33), and the cathodic term becomes negligible or if the overpotential is made more

negative, jA<<jC (Equation 34) and the anodic term becomes negligible.

j=j0 exp(αAzF

RT η) (Equation 33)

j=−j0 exp(−αCzF

RT η) (Equation 34)

By applying the logarithm on Equation 33 and 34, the anodic (Equation 35) and cathodic

(Equation 36) Tafel equations are obtained.

logj=logj0+αAzF

2.3RTη (Equation 35)

logj=−logj0−αCzF

2.3RTη (Equation 36)

Theoretical Background

Using the Tafel equations by plotting the overpotential over log j, it is now possible to determine

j0, αA, and αC experimentally as intercepts and slopes.

2.3.1 Reaction Rate – Bond Strength – Relation

As mentioned in the previous section, a heterogeneously catalyzed reaction's rate (kinetics)

depends strongly on the interaction between the catalyst and the reactants. Therefore, it is

necessary to extend Figure 7 to include the Langmuir-Hinshelwood and Eley-Rideal

mechanisms (Figure 9). Both mechanisms are kinetic models describing the reaction of reactant

A (green) with reactant B (blue) at or near a catalyst surface (grey). However, in the Langmuir-

Hinshelwood mechanism (LHM), both reactants adsorb on the catalyst surface, whereas in the

Eley-Rideal mechanism (ERM), only reactant A adsorbs on the catalyst surface. Specifically,

for the LHM, reactants A and B adsorb on the catalyst surface (Figure 9 a-b), react and form

product P (Figure 9 b-c), and product P desorbs from the catalyst surface (Figure 9 c-d).

Consequently, the reaction rate is dependent on the reaction rate constant and the surface

coverage of A and B. For the ERM, reactant A adsorbs on the catalyst surface (Figure 9 e-f),

the adsorbed A reacts with B from the bulk to form product P (Figure 9 f-g), and P desorb from

the catalyst surface (Figure 9 g-h). Here, the reaction rate depends on the reaction rate constant,

the surface coverage of A, and the concentration of B.

Theoretical Background

Figure 9: Langmuir-Hinshelwood and Eley-Rideal reaction mechanisms.

Irrespective of the mechanism, the adsorption of the reactant(s) and the desorption of the

product play a decisive role for the reaction rate, which is always dependent on the slowest, i.e.,

the rate-determining step. In the early 20th century, Paul Sabatier postulated the reaction rate

dependence on the bond strength between reactant and catalyst.122 He characterized the

maximum reaction rate by a binding strength that is neither too weak nor too strong. This

relationship is now widely illustrated and referred to as a volcano curve (Figure 10). The top of

the volcano curve describes the optimal binding strength and, thus, the highest reaction rate

(green area). Lower binding strengths (blue area) are characterized by adsorption limitations,

which means that the reactants are not adsorbed properly at the catalyst (steps of Figure 9 a-b

and Figure 9 e-f are rate-determined). Higher binding strengths (red area) are characterized by

desorption limitations, which means that the reactants or products are not desorbed properly

(steps of Figure 9 c-d and Figure 9 g-h are rate-determined). Thus, Sabatier's principle provides

a guideline for practical heterogeneous catalysis by preventing sluggish kinetics through an

appropriate choice of catalyst.

Theoretical Background

Figure 10: Optimal binding relation based on the Sabatier principle illustrated in a volcano curve.

Theoretical Background

2.4 Organic Electrochemistry

Organic electrochemistry is, by definition, the combination of organic- and electrochemistry.

Even though organic electrochemistry is a relatively young concept compared to “classical”

electrochemistry, the history and the associated names of influential scientists are more closely

interlaced than one might think. At the beginning of the 19th century, the Italian physicist

Alessandro Volta invented the Voltaic pile. Immediately after that, the Russian Petrov

published the successful electrolysis of alcohols/aliphatic oils, and Grotthuss oxidized indigo

white electrochemically to indigo blue in Lithuania. Thirty years later, Michael Faraday

postulated Faraday’s law and discovered the formation of hydrocarbons from aqueous acetic

acid salt solutions via electrolysis. In 1849, the electrochemical oxidation of carboxylic acid to

the dimeric alkane and CO2 was investigated by Kolbe and has since been known as Kolbe

electrolysis. Although intensive research into organic electrochemistry continued, it was not

until the first half of the 20th century that significant advances were made, such as reducing

nitrobenzene to aniline or electroanalytical techniques such as polarography Heyrovský and

Tachi developed in the 1920s. After research in organic electrochemistry had come to a

standstill during the Second World War, Baizer helped it to a renaissance in 1964 with the

hydrodimerization of acrylonitrile, an important commercialized industrial process by

Monsanto for the production of adiponitrile.123,124

In organic electrochemistry, an organic starting molecule can be activated at the cathode by

electron transfer from the electrode to the molecule (reduction) or by electron transfer from the

molecule to the anode (oxidation). Electron transfer processes can occur via electron-bounded

transition complexes (inner-sphere mechanism) or without a transition complex (outer-sphere

mechanism). Furthermore, the spatial separation of reduction and oxidation is a unique feature

of organic electrochemistry and differs from “classic” organic chemistry. In addition to being

more environmentally friendly by avoiding hazardous reagents, organic electrochemistry offers

the additional advantage of potential-controlled reactions.125

In contrast to most inorganic electrochemical processes, reactions in organic electrochemistry

often involve transformations or reactions before and after the actual electrode reaction

(Figure 11).

Theoretical Background

Figure 11: General suggestion of an organic electrode reaction process.

Hence, not only the electrode reaction and thus the electrode material is decisive for the product

selectivity, but also which side reactions can take place, for example, in the electrolyte. Product

selectivity can be divided into chemoselectivity (at which functional group does the reaction

take place), regioselectivity (at which position of the molecule does the reaction take place),

and stereoselectivity (which spatial orientation do functional groups have after the reaction).

Control over chemoselectivity can be achieved, for example, via the potential, based on the

different redox potentials of the various functional groups. For the regioselectivity, on the other

hand, thermodynamics (stability of a reactive intermediate), kinetic control (reaction rate of an

intermediate), and selective adsorption of the reactant on the electrode, and thus the dipole

moment are decisive factors. In addition to steric effects, stereoselectivity can also be controlled

by the specific adsorption of reactive intermediate on the electrode.

However, in addition to the challenges posed by the above points, many different reaction

possibilities open up simultaneously, such as addition, insertion, substitution, elimination,

dimerization, or polymerization reactions. Thereby, the opportunities of organic chemistry are

combined with the advantages of electrochemistry and show the potential of organic

electrochemistry.

Experimental

3 Experimental

This chapter is reproduced with permission from Hauke, P., Klingenhof, M., Wang, X., de

Araújo, J. F. & Strasser, P. Efficient electrolysis of 5-hydroxymethylfurfural to the biopolymer-

precursor Furandicarboxylic acid in a zero-gap MEA-type electrolyzer. Cell Reports Physical

Science 2, 100650, doi:https://doi.org/10.1016/j.xcrp.2021.100650 (2021). And Chapter 6 from

Hauke, P., Merzdorf, T., Klingenhof, M. & Strasser, P. Hydrogenation versus hydrogenolysis

during alkaline electrochemical valorization of 5-hydroxymethylfurfural over oxide-derived

Cu-bimetallics. Nature Communications 14, 4708, doi:10.1038/s41467-023-40463-y (2023).43

3.1 Synthesis

3.1.1 Microwave Assisted One-pot Synthesis of NiX(-CO32-)-LDH Powder

An aqueous solution of Ni(OAc)2 4H2O (1200 µl, 0.6 M, Sigma-Aldrich, 99.98 % purity) was

mixed in a microwave tube with Fe(NO3)3 9H2O (240 µl, 0.6 M, Alfa Aesar, 98-101.0 %

crystalline), Co(NO3)2 9H2O (240 µl, 0.6 M, Strem Chemicals, 99.999 % purity), MnCl2

(240 µl, 0.6 M, Sigma-Aldrich, 99 % purity) or VCl3 (240-480 µl, 0.6/1.2 M, Alfa Aesar, 99 %

purity) and 6 ml N,N-DMF (Alfa Aesar, 99 %). The resulting solution was stirred overnight.

Additionally, 4 ml N,N-DMF, and 8 ml ultrapure H2O were added. The reaction mixture was

transferred into microwave assisted autoclave. The microwave treatment included a heating

step to 392 K, a holding step for 60 min, heating to 432 K, and holding for 90 min at 600 rpm.

After cooling down, the yellow-brown powder was collected by centrifuge, followed by

lyophilization.

3.1.2 Anion Exchange for NiX(-CO32-)-LDH Powder

Anion exchange for chloride was carried out by mixing NiX(-CO32-)-LDH with 200 ml of the

saturated salt solution (NaCl) and 67 µl of HCL (37.5 %). The reaction mixture was stirred

mechanically at RT, over 18 h at 600 rpm. The resulting product was collected with a

centrifuge, washed with Mili-Q and EtOH, followed by lyophilization.

Anion exchange for perchlorate was carried out by mixing NiFe(-Cl-)-LDH@NF with 200 ml

aqueous solution of 17.12 g NaClO4 and 25.7 µl of HCL (37.5 %). The reaction mixture was

Experimental

stirred mechanically at RT, over 16 h at 250 rpm. The resulting product was collected by

centrifuge, washed with Mili-Q and EtOH, followed by lyophilization.

3.1.3 Microwave Assisted One-pot Synthesis of the As-prepared NiFe(-CO32-)-

LDH@NF Electrodes

An aqueous solution of Ni(OAc)2 4H2O (1200 µl, 0.6 M, Sigma-Aldrich, 99.98 % purity) was

mixed in a microwave tube with Fe(NO3)3 9H2O (240 µl, 0.6 M, Alfa Aesar, 98-101.0 %

crystalline), Co(NO3)2 9H2O (240 µl, 0.6 M, Strem Chemicals, 99.999 % purity), MnCl2

(240 µl, 0.6 M, Sigma-Aldrich, 99 % purity) or VCl3 (240-480 µl, 0.6-1.2 M, Alfa Aesar, 99 %

purity) and 6 ml N,N-DMF (Alfa Aesar, 99 %) based on the powder NiFe layered double

hydroxide (LDH) synthesis by Dresp et al.126 The resulting solution was stirred overnight.

Additionally, 4 ml N,N-DMF, 8 ml ultrapure H2O, and cleaned (HCl, H2O, ETOH) Ni-foam (1

or 5 cm2, Alantum, 450 µm pore size, 420 g m-2 area density, 0.8 mm, and 1.5 mm thickness)

were added. The reaction mixture was transferred into the microwave-assisted autoclave. The

microwave treatment included a heating step to 392 K, which was kept for 60 min, heated to

432 K, and hold for 90 min (except noted otherwise) at 250 rpm. After cooling down the slightly

yellow Ni-foam was collected and was washed three times with H2O and ETOH and

subsequently dried with nitrogen.

3.1.4 Anion Exchange of the As-prepared NiX(-CO32-)-LDH@NF Electrodes

Anion exchange for chloride was carried out by mixing NiX(-CO32-)-LDH@NF with 200 ml of

the saturated salt solution (NaCl) and 67 µl of HCL (37.5 %). The reaction mixture was stirred

mechanically at RT, over 18 h at 250 rpm. The resulting product was collected, washed with

Mili-Q and EtOH followed by drying under nitrogen flow.

Anion exchange for perchlorate was carried out by mixing NiFe(-Cl-)-LDH@NF with 200 ml

aqueous solution of 17.12 g NaClO4 and 25.7 µl of HCL (37.5 %). The reaction mixture was

stirred mechanically at RT, over 16 h at 250 rpm. The resulting product was collected, washed

with Mili-Q and EtOH followed by drying under nitrogen flow.

Experimental

3.1.5 Precipitation Synthesis of Pure Metal Oxides (MOx; M= Cu, Ni, Fe, Co)

Preparation of the pure metal oxides (MOx; M= Cu, Ni, Fe, Co) was carried out by precipitation

method. 200 mg of powdered metal salt precursor (Cu(II)Cl2, Sigma-Aldrich 99%;

Ni(II)(NO3)2 6H2O, Roth 99%; Fe(II)Cl2 4H2O, Alfa Aesar 98%; Co(II)(NO3)2 6H2O, Acros

Organics 99%) was solved in 27 ml of water. NaOH (Sigma-Aldrich 99.99% trace metals basis,

3 M, 3 ml) was added dropwise to the mixture. The precipitate was collected by centrifuge and

was washed three times with H2O, ETOH, H2O and afterward dried by lyophilization. Then,

obtained powder was calcined at 300 °C in a muffle furnace (Carbolite) for 3 h with a heating

rate of 5 °C/min.

3.1.6 Precipitation Synthesis of Mixed Metal Oxides (CuO/MOx 10 mol%; M= Ni, Fe,

Co)

Preparation of the mixed metal oxides (CuO/MOx; M=Ni, Fe, Co) was carried out by

precipitation method. To achieve a molar ratio of 9/1 Cu/M, 116 mg of powdered Cu metal salt

precursor (Cu(II)Cl2, Sigma-Aldrich 99%) was mixed with 18-26 mg of second metal salt

precursor (Ni(II)(NO3)2 6H2O, Roth 99%; Fe(II)Cl2 4H2O, Alfa Aesar 98%; Co(II)(NO3)2

6H2O, Acros Organics 99%) and solved in 17 ml of water. NaOH (Sigma-Aldrich 99.99% trace

metals basis, 3 M, 3 ml) was added dropwise to the mixture. The precipitate was collected by

centrifuge and was washed three times with H2O, EtOH, H2O and afterward dried by

lyophilization. Then, obtained powder was calcined at 300 °C in a muffle furnace (Carbolite)

for 3 h with a heating rate of 5 °C/min.

3.2 Characterization Methods

3.2.1 Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES)

ICP-OES was carried out with a Varian 715-ES instrument. All samples were prepared by

diluting 1 mg of catalyst in Aqua Regia (1 ml) and ultrapure water (9 ml).

Experimental

3.2.2 Powder X-Ray Diffraction and Thin Film X-Ray Diffraction (XRD)

Powder X-ray diffraction was performed using a Bruker D8 Advance apparatus. A Cu Kα

(1.54 Ǻ) radiation source was used. The wide-angle measurements were taken in the range of

5°-80°.

Thin film X-ray diffraction was performed using a Bruker D8 Advance apparatus. A Cu Kα

(1.54 Ǻ) radiation source was used. The wide-angle measurements were taken in the range of

10°-80°.

3.2.3 Scanning Electron Microscopy (SEM)

Scanning electron microscope and Elemental Mapping (SEM-EDX) measurements were

carried out with Zeiss Gemini 982 and a Hitachi SU8020 instrument.

3.2.4 Transmission Electron Microscopy (TEM)

Transmission electron microscope (TEM) measurements were conducted using a Tecnai G2 20

s-Twin microscope, equipped with a LaB6-cathode and a GATAN MS794 P CCD-detector at

ZELMI Centrum, Technical University of Berlin. TEM samples were ultrasonicated in i-PrOH

and drop-dried on copper grids.

3.2.5 X-Ray Photoelectron Spectroscopy (XPS)

X-ray photoelectron spectroscopy (XPS) measurements were carried out with a

ThermoScientific K-Alpha+ X-ray photoelectron spectrometer (Group of Prof. Thomas, TU

Berlin) and at Fritz-Haber Institute (FHI) using non-monochromatized Al K ( eV) for

the NiFe-LDH powder sample and Mg Kα (1253.6 eV) excitation for the NiFe(-A-)-LDH@NF

samples. A hemispherical analyzer (Phoibos 150, SPECS) was used. The measured spectra

were analyzed using CasaXPS software. Binding energy (BE) was aligned by the C 1s spectra.

Experimental

3.3 Setup Configurations

3.3.1 Rotating Disk Electrode (RDE) Measurements for HMF Oxidation

The electrochemical measurements of the prepared catalysts were carried out in a conventional

three-electrode setup using a BioLogic Sp-200 potentiostat, Pt-mesh as the counter electrode,

and a reversible hydrogen electrode (RHE) as the reference electrode. The activities towards

OER and 5-HMF oxidation were investigated in 50 ml N2-saturated 0.1 M KOH with and

without 10 mM 5-HMF at room temperature (RT). Cyclic voltammetry (CV) techniques were

performed at scan rates between 5 mV s-1 and 50 mV s-1 and +1.1 V to +1.7 V. Due to the static

character of the working electrode (WE) holder, no rotation was applied (except noted

otherwise).

The same setup was used for impedance spectroscopy. NF blank, NiX(-CO32-)-LDH@NF

(X=Fe,Mn,Co,V), NiFe(-Cl-)-LDH@NF and NiFe(-ClO4-)-LDH@NF (Ageom.=2.5 cm2) were

investigated. Measurements were taken between 1 MHz and 10 mHz at different potentials

(+1.53 V, +1.58 V, +1.63 V) in 0.1 M KOH. Fits of the Nyquist plots and ECSA calculations

are based on the previous work of Dionigi et al.25

3.3.2 Differential Electrochemical Mass Spectrometry for HMF Oxidation

A new differential electrochemical mass spectrometry (DEMS) setup from LIQUIDLOOP

GmbH was used, employing two distinct electrochemical liquid/vacuum cell interfaces: the

primary high-vacuum mass spectrometer (MS) chamber (Pfeiffer PrismaPlus® QMG 220) and

a second chamber that served as an intermediate vacuum region (~10-3 mbar) between the MS

chamber and electrochemical cell. The Pfeiffer MS chamber was connected to the intermediate

chamber using a Ø6 mm open flange. A turbomolecular pump (Pfeiffer HiPace® 80) was

mounted directly in each stage and backed by a diaphragm pump and rotary vane pump. The

system design allows pressure reduction over the two stages from 10-3 mbar down to 10-6 mbar

in the MS chamber.

The Capillary DEMS flow cell (Figure S 48) allows working various types of electrode

geometries. This capability is useful when small, expensive electrolyte volumes are concerns,

such as isotope-labeled compounds.

Experimental

Thus, for all the isotope labeled OER experiments the droplet flow cell is used. The

electrochemical measurement is maintained under electrolyte convection conditions, and an

internal cell volume is maintained. The ability for visual inspection of the electrode and

capillary contact helps align position of the various inlet and outlet tubes. The formation of gas

bubble is controlled by flushing using an inlet tube in a direction of 45° with the planar surface

of the electrode. For the flow of the electrolyte stream, a hydrostatic pressure-controlled system

is employed to maintain constant the inlet flow at 10 µl s-1. The concentric larger-diameter

electrolyte outlet tube is placed 5 mm from the electrode surface. This outlet tube as a function

of promoting a withdrawal flow process and bringing the oxygen-concentrated electrolyte into

the capillary inlet. In capillary is the inner concentric tube to the outlet flow tube. The capillary

has Ø150 µm inner diameter and is placed as close as 500 µm from the electrode surface with

90° to the inlet tube. The setup is gastight and a relative pressure of 1.7 bar Argon is maintained

and controlled above the electrolyte by a high precision pressure gauge.

The counter electrode is meshed platinum and a home-made reversible hydrogen electrode,

RHE, made of a Pt wire that was constantly replenished with H2 gas, served as reference

electrode (RE). The reference electrode has a Ø1 mm inner diameter and is placed 5 mm from

the working electrode (NiFe(-Cl-)-LDH@NF, Ageom.=0.25 cm2). The liquid dissolved oxygen

is collected through the capillary at 5 µl s-1 and introduced to a flow splitter that directs the flow

to the disk-shaped liquid/vacuum chamber via four thin channels. In the disk-shaped

liquid/vacuum chamber, a PTFE membrane makes the interface between liquid and vacuum.

This compartment has an inward radial electrolyte flow pattern because the four thin channels

connect to the outer edges of the liquid/vacuum chamber. The stationary circulation of

electrolyte is achieved by a single exit hole in the center of the liquid/vacuum chamber.

3.3.3 Rotating Disk Electrode (RDE) Measurements for HMF Reduction

The electrochemical measurements of the prepared powder catalysts were carried out in a

conventional three-electrode setup using a BioLogic Sp-200 potentiostat, Pt-mesh as the

counter electrode, and a reversible hydrogen electrode (RHE) as the reference electrode. The

activities towards HER and 5-HMF reduction were investigated in 50 ml N2-saturated 0.1 M

KOH with and without 10 mM 5-HMF at room temperature (RT) at 2500 rpm.

Experimental

Cyclic voltammetry (CV) techniques were performed at scan rates between 5 mV s-1 and 50 mV

s-1 and 0 V to -0.6 V (if not noted otherwise). The catalyst powder was drop cast on a glassy

carbon disc electrode with an electrode area of 0.196 cm2 and a catalyst loading of 0.04 mg.

3.3.4 Undivided Three-Electrode Cell Measurements for HMF Reduction

The electrochemical measurements of the 1x1 cm prepared cathodes (1 mg cm-2) were carried

out in a conventional three-electrode setup using a BioLogic Sp-300 potentiostat, Pt-mesh as

the counter electrode, and a reversible hydrogen electrode (RHE) as the reference electrode.

For the preparation of the cathode, spray-coating technique was applied, aiming at a loading of

1 mg cm-2. For the preparation of the catalyst ink, 100 µL Mili-Q and 5 ml i-PrOH were added

to 100 mg catalyst and sonified subsequently, ionomer solution (3 wt% Nafion©-solution,

1100 g/mol, Sigma-Aldrichs) were added during the sonification resulting in a 1 wt%

dispersion of the catalytic active material. The activities towards HER and 5-HMF reduction

were investigated in 25 ml N2-saturated 0.1 M KOH with and without 10 mM 5-HMF at room

temperature (RT). Cyclic voltammetry (CV) techniques were performed at scan rates between

5 mV s-1 and 50 mV s-1 and 0 V to -0.8 V. Due to the static character of the working electrode

(WE) holder, no rotation was applied.

3.3.5 Full-Cell measurements and Fabrication of the Applied Electrodes for HMF

Oxidation

The MEA-type cell measurements were conducted using a commercial 5 cm2 electrolysis cell

(Dioxide Materials) equipped with linear serpentine flow fields made out of TiO2 (Anode and

Cathode). A detailed sketch of the applied cell is given in Figure 15. As anode blank Nickel

Foam (NF), as well as the prepared and modified NiX(-A-)-LDH electrodes, were used. For the

preparation of the cathode, spray-coating technique was applied, aiming at a loading of 0.1 mg

Pt cm-2 (Umicore, 48.46 wt% Pt at carbon, Elyst Pt50). For the preparation of the catalyst ink,

100 µl Mili-Q and 10.28 g i-PrOH were added to 120 mg Pt-C and sonified subsequently,

ionomer solution (3 wt% Nafion©-solution, 1100 g/mol, Sigma-Aldrichs) were added during

the sonification resulting in a 1 wt% dispersion of the catalytic active material.

Experimental

The measurements were conducted in 0.1 and 1 M KOH (99.99%, Sigma-Aldrich) respectively

LiOH(>98%, Sigma-Aldrich), NaOH (99.99%, Sigma-Aldrich), RbOH (99%, Alfa Aeser),

CsOH (99.95%, Acros Organics) with and without 10 mM 5-HMF and an electrolyte flow of

20 ml min- 1.

50 CVs from +1.0- +1.7 Vcell at a scan rate of 50 mV s-1 in 0.1 or 1 M KOH were carried out

for system activation. HMF electrolysis was performed at a constant potential of +1.56 Vcell

(except noted otherwise) over system-specific periods.

System stability tests were carried out under same conditions for 10h (5 cycles) in 0.1 M KOH

and 10 mM 5-HMF respectively for 5h (5 cycles) in 0.1 M KOH and 5 mM 5-HMF. Two

different types of 5-HMF injections were performed. First (10 h program), every two hours

(1 catalytic cycle) the anodic electrolyte was completely changed, and secondly (5 h program),

every hour (1 catalytic cycle) 5-HMF was injected into the existing electrolyte. Therefore, a

small amount of the electrolyte was taken to dilute the solid 5-HMF and subsequently, the

solution was given back in the electrolyte.

3.3.6 Full-Cell Measurements and Fabrication of the Applied Electrodes for HMF

Reduction

The MEA-type cell measurements were conducted using a commercial 5 cm2 electrolysis cell

(Dioxide Materials) equipped with linear serpentine flow fields made out of TiO2 (anode and

cathode). As anode blank Nickel Foam (NF), as well as the prepared and modified

NiFe(-Cl-)-LDH@NF electrode were used. For the preparation of the cathode, spray-coating

technique was applied, aiming at a loading of 1 mg cm-2. For the preparation of the catalyst ink,

100 µl Mili-Q and 5 ml i-PrOH were added to 100 mg catalyst and sonified subsequently,

ionomer solution (3 wt% Nafion©-solution, 1100 g/mol, Sigma-Aldrichs) were added during

the sonification resulting in a 1 wt% dispersion of the catalytic active material. The

measurements were conducted in 0.1 M KOH (99.99%, Sigma-Aldrich) with and without

10 mM 5-HMF and an electrolyte flow of 25 ml min- 1. 20 CVs from 0 to -0.6 Vcell at a scan

rate of 50 mV s-1 in 0.1 KOH were carried out for system activation. HMF electrolysis was

performed at a constant current density of -10, -20, and -30 mA cm-2 (except noted otherwise)

over 30 min, close to the theoretical charge needed for full HMF conversion to BHMF.

Experimental

Stability test was carried out under the same conditions at -20 mA cm-2 for 5 cycles in 0.1 M

KOH and 10 mM 5-HMF.

3.3.7 Product Analysis for HMF Oxidation

Product analysis was carried before, during, and after a constant applied potential. The samples

(500 µl) were taken by an automatic aliquot sampling device and analyzed by high performance

liquid chromatography (HPLC, Agilent 1200 series, Agilent Technologies Hi-Plex H column)

at a constant flow rate of 1 ml min-1 (0.01 M H2SO4) and detected by a Refractive Index Detector

(RID). Calibration curves and retention times of the different oxidation products of 5-HMF are

given in the supplementary information (Figure S 35).

3.3.8 Product Analysis for HMF Reduction

Product analysis was carried out before, during, and after a constant applied current density.

The samples (500 µl) were taken by an automatic aliquot sampling device and analyzed by high

performance liquid chromatography (HPLC, Agilent 1200 series, Agilent Technologies Zorbax

SB-C18 column) at a constant flow rate of 0.8 ml min-1 (H2O:C2H3N 95:5) and detected by a

Refractive Index Detector (RID).

Hydrogen was quantified by an Unisense© Hydrogen Sensor. The sensor was directly included

in the cathodic outlet line via a T-piece. Calibration was done in 0.1 M KOH saturated with

H2/Ar 5/95 gas mixture.

Experimental

3.3.9 Calculation of ECSA, HMF Conversion, Product Yields, Faradaic efficiency, and

STY

ECSA=CDL

CS (Equation 37)

XHMF =(1−nHMF,remain

nHMF,initial )∙100% (Equation 38)

YProduct =nProduct

nHMF,initial ∙100% (Equation 39)

FEProduct = nProduct

Qtotal

F∙z ∙100% (Equation 40)

STYFDCA =nFDCA

telectrolysis∙Ageom. (Equation 41)

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

4 Anion-Tuned Layered Double Hydroxide Anodes for

Anion Exchange Water Electrolyzers: From Catalyst

Screening to Single Cell Performance

This chapter is reprinted with permission from Klingenhof, M‡.; Hauke, P‡.; Kroschel, M.;

Wang, X.; Merzdorf, T.; Binninger, C.; Ngo Thanh, T.; Paul, B.; Teschner, D.; Schlögl, R.;

Strasser, P., Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Membrane

Water Electrolyzers: From Catalyst Screening to Single-Cell Performance. ACS Energy Letters

American Chemical Society.127 (Accepted manuscript version)

‡ M.K. and P.H. are contributed equally.

The study was designed and executed by P.H. and M.K.. P.H. performed the synthesis and

electrochemical evaluation in the 3 electrode setup. M.K. performed the ECSA evaluation in

the 3 electrode setup and the electrochemical evaluation in the two electrode setup. M.Kr.,

X.W., and T.M. assist in the characterization of the catalysts. C.B., T.N.T., and B.P. helped

with the setup design.

4.1 Abstract

Anion Exchange Membrane Water Electrolysis (AEMWE) is an attractive emerging green

hydrogen technology. Yet, the scaling of trends in activity of anode catalysts for the oxygen

evolution reaction (OER) from a liquid-electrolyte, three-electrode environment to the two-

electrode single cell format has remained poorly considered. Here, we critically investigate the

scalings of kinetic and catalytic properties of a family of highly active Ni-Foam (NF) supported,

anion (A-)-tuned NiFe(-A-)-OER catalysts. Trends in catalytic activity suggest impressive

improvements of up to 91x in three-electrode setups (3LC) compared to uncoated NF. While

we demonstrate the successful qualitative structure-performance tunability into 5 cm2 AEMWE

single cell, we also find serious limitations in the quantitative predictability of three-electrode

setups for single cell performance trends. Cell environments appear to equalize the cell

performances of designer catalysts, which has important ramifications for electrode

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

development. We succeed in analyzing and discussing some of these translation limitations in

terms of previously overlooked effects summarized in activity improvement factor f.

4.2 Introduction

Increasing global energy demand coupled with dwindling fossil raw materials and climate

change requires replacing conventional fossil fuels with sustainable ones based on renewable

energy. Hydrogen, due to its high gravimetric energy storage density and versatility as reactant

for industrial processes, is a favorable energy vector for future energy grids. Therefore, cost-

efficient and CO2 neutral hydrogen production is desirable.128,129 96 % of the hydrogen used

today is produced by steam methane reforming or coal gasification, both are greenhouse gas

(GHG) intensive processes.130,131 Water electrolyzers that split water into H2 and O2

(H2O → H2 + ½ O2) using surplus energy provided by renewable energy sources (RES)

represent promising near-zero emission alternatives.132,133

Currently, three different low-temperature water electrolyzer technologies are in different

stages of development. First, the proton exchange membrane water electrolysis (PEMWE), the

established liquid alkaline electrolyzers (AWE), and the emerging anion exchange membrane

water electrolysis AEMWE.134 PEM and AEL require expensive Ir electrode materials or

corrosive electrolytes (33 wt% KOH), respectively. Both appear unfavorable in the long term

integration into future RES-based energy grids.135 AEMWE combines the advantages of both

technologies. Local high pH allows for the application of non-noble, abundant and

cost-efficient materials, furthermore compact stack design is suitable for dynamic linkage with

fluctuating RES for production of green hydrogen.136 Despite broad research effort, current

densities greater than 500 mA cm-2 at reasonable potentials without use noble metals are still

rare, especially when applying low concentrated KOH solutions.131,137 Most previous studies

used either high concentrated KOH (1 M KOH), commercial noble metal-based catalyst or

both.138-140

A serious challenge for today’s AEMWE catalysts R&D is the correlation and translation from

liquid-electrolyte, three-electrode catalyst screening setups to realistic MEA-based

two-electrode single or multi-cell electrolyzer setups. And even though there are promising

laboratory scale catalysts, a reliable quantitative performance prediction to scaled-up water

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

electrolyzers of industrial design and power has remained challenging, unconsidered and in the

majority of studies unmentioned.136,141-145

Only few studies tackle the reasons for the performance differences between three-electrode

setups and two-electrode electrolyzer cell designs are often significant for various reasons.146

This work contributes to the difficulties in transferring a set of new OER catalyst from

three-electrode setup to realistic single electrolyzer cells.

In the present study, we address binder-free nanostructured NiFe(-CO32-)-LDH OER catalysts

prepared in-situ onto Nickel Foam (NF) substrate providing exceptionally active, synthetically

scalable, structurally tunable catalyst systems. After synthesis, a controlled exchange of the 2D

interlayer (-CO32-)- anions with Cl- and ClO4- is applied. We show that in-situ growth coupled

to the subsequent anion tuning results in the desired chemical and structural modifications. In-

situ impedance spectroscopy reveals structure-activity relations of the catalyst/support design

and gives clues to their favorable performance in liquid electrolyte test. Finally, we investigate

the scalability of these trends into realistic AEM water electrolyzer cells. We demonstrate the

qualitative translation of performance trends among the anode materials, yet note significant

losses in relative performance gains among the deployed NiFe(-A-) anode catalysts. We analyze

and discuss fundamental reasons for this limiting behavior.

4.3 Results and Discussion

4.3.1 Morphology and Surface Composition of Tunable Anion-Interlaced Catalysts

Powder NiFe(-CO32-)-LDH were prepared using a 30 ml autoclave and an autogenous pressure

synthesis was carried out for 90 minutes at 160 °C.147 Analog synthesis conditions were used

to prepare LDH on NF. Precleaned NF served as substrate for the synthesis of binder-free

catalyst overlayers.126 The prepared 3D catalyst structure will be referred to as

NiFe(-CO32-)-LDH@NF. The growth process on the NF creates a nanostructured porous

electrode monolith. In a second step, the interlaced anions located in the interlayers of the 2D

NiFe(-CO32-)-LDH surface were chemically exchanged into NiFe(-A-)-LDH@NF using a set

of different anions (A- = Cl- and ClO4-). To achieve this, the parent catalyst was exposed to

excess solutions of the target anion. Herein CO32- was exchanged with Cl- followed by

exchanging Cl- with ClO4- resulting in the corresponding binder-free NiFe(-A-)-LDH@NF.

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

To inspect the resultant catalyst morphologies, SEM images of NF (Figure S1 a-c),

NiFe(-CO32-)-LDH@NF (Figure 12 a, d, g), modified Cl- and ClO4- (Figure 12 b, e, h and

Figure 12 c, f, i) suggest the successful formation of Ni(oxy)hydroxide and Fe species as well

as (-A-) variants, though diffraction peaks of the NF background remained weak. Furthermore,

the surface lamella structure evolution of the catalytically active overlayer is revealed for

NiFe(-CO32-)-LDH@NF (Figure 12 a, d, g). The modifications via anion exchange (-A-)

roughens the surface accompanied with an increase of the surface area. We correlate these

transformations to a changing Ni/Fe ratio during the anion exchange.147 Lower magnification

images (Figure 12 a-f and Figure S1 d-f) give an overview about the macroscopic electrode

structures, indicating a homogenous coverage of NF with NiFe(-A-). The absence of binder

allows fast mass transport at the electrode-electrolyte interface, maximizes accessible surface

area and avoids I/C optimization. Also, the porous 3D structure of NF is retained after synthesis.

Based on the 3D morphology of the obtained catalytic anode structures, large catalyst-

electrolyte interfaces were achieved.

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

Figure 12: SEM, XPS and XRD investigations of NiFe(-CO32-)-LDH powder, Nickel Foam (NF) and NiFe(-A-)-LDH

grown on NF. a-i) SEM images of the prepared anodes, NiFe(-CO32-)-LDH@NF (red), NiFe(-Cl)-LDH@NF (blue),

NiFe(-ClO4)-LDH@NF (green). j) Comparison of the XPS Ni 2p regions of uncoated NF (black) and NiFe(-CO3)-LDH@NF

(red). k) Cl 2p XPS spectra of NiFe(-Cl-)-LDH at NF. l) XRD investigations of powder NiFe(-CO32-)-LDH including the shifts

of the (003) and (006) diffraction reflections during anion exchange and reversible structure change to NiFe(-Cl-)-LDH@NF

and NiFe(-CO32-)-LDH. The arrows indicate the shifts from CO32- to ClO4- (green) and the back shift to CO32- after aging in

KOH (red).

To verify the success of our binder-free autogenous pressure synthesis, the chemical state of

the catalyst surface was investigated using X-Ray Photoemission Spectroscopy (XPS). Data is

reported for the catalytically active parent NiFe(-CO32-)-LDH with (Figure 12) and without NF

support (Figure S2) as well as for the supported, anion-tuned NiFe(-A-)-LDH@NF variants

(Figure 12 k and Figure S3, Figure S4, Figure S5).

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

Clearly, the Ni 2p spectrum of the uncoated NF and NiFe(-CO32-)-LDH@NF differ

significantly (Figure 12 j). While both samples showed Ni(II) species, Ni(0) surface species

were only present in uncoated NF. The absence of Ni(0) species in NiFe(-CO32-)-LDH@NF

confirmed the complete coating of the NF with Ni-oxide species. The Ni 2p spectra of

NiFe(-CO32-)-LDH@NF corresponded to the XPS spectra of NiFe(-CO32-)-LDH powder and

reveals oxyhydroxide characteristics of the surface species, which was also confirmed for

NiFe(-A-)-LDH@NF (Figure S4 b, Figure S5 b). Fe 2p core level signals (Figure S3 c) further

confirmed the in-situ growth of NiFe(-CO32-)-LDH at NF. Weak Fe 2p signals resulted from

the overall very low Fe/Ni ratio.

Finally, Cl 2p spectra of NiFe(-A-)-LDH@NF confirm the chloride exchange (Figure 12k,

Figure S5 f). Ni 2p and Fe 2p signal intensity declined after (-A-)-modification. The O 1s region

showed a slight shoulder, arising from decreasing NiFe(-CO32-)-LDH layer thickness, revealing

the underlying oxidic NF. XRD diffraction patterns of NiFe(-CO32-)-LDH and

NiFe(-ClO4-)-LDH depicted in Figure 12 l show anion intercalation accompanied with

increasing interlayer distances and the transformation from NiFe(-ClO4-)-LDH back to NiFe(-

CO32-)-LDH after aging. Detailed synthesis-performance relations studying influences of

preparation parameters on the catalytic reactivity are shown in Figure S6- Figure S11 and

discussed in the Supplementary Discussion 1.

4.3.2 Structure-Composition-Catalytic OER Reactivity Relations

Electrochemical impedance spectroscopy (EIS) and modeling (Figure 13 and Figure S12) was

applied evaluating the faradaic capacitance and calculating the real electroactive surface area

(ECSA) of the anodes.148,149 The recently proposed and proven EIS-based faradaic capacitance

model took the presence of uncoated NiOx layers of the NF, as well as that of non-conductive

oxyhydroxide layers of anion-tuned NiFe(-A-)-LDH grown on NF into account.149 Figure S12 a

illustrates the electrical equivalent circuit (EEC) and analysis, which is described in SI.148

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

Figure 13: Electrochemical impedance spectroscopy (EIS) investigations determining the ECSA of the prepared

NiFe(-A-)-LDH@NF and NF in a three-electrode setup (3LC) and electrocatalytic OER reactivity and stability of

uncoated NF, compared to NiFe(-A-)-LDH@NF anodes. Nyquist plots of a) NF, b) NiFe(-CO32-)-LDH@NF,

c) NiFe(-Cl-)-LDH@NF and d) NiFe(-ClO4-)-LDH@NF recorded at 1.53 V, 1.58 and 1.63 V vs RHE. e) Calculated ECSA of

uncoated NF (black) and the prepared anodes, NiFe(-CO32-)-LDH@NF (red), NiFe(-Cl-)-LDH@NF (blue) and

NiFe(-ClO4-)-LDH@NF (green). f) iR-corrected linear sweep voltammetry (LSV) g) ECSA- and iR-corrected LSVs of

uncoated NF (black), NiFe(-CO32-)-LDH@NF (red), NiFe(-Cl-)-LDH@NF (blue) and NiFe(-ClO4-)-LDH@NF (green).

Stability investigations of h) uncoated NF, i) NiFe(-CO32-)-LDH@NF, j) NiFe(-Cl-)-LDH@NF and k) NiFe(-ClO4-)-LDH@NF

anodes. The measurements were conducted in 0.1 M KOH at RT using three-electrode setup.

The Nyquist spectra of NiFe(-A-)-anodes exhibit two arcs. The low frequency resistance (LFR)

arc proves electrode potential-dependency, the high frequency (HFR), in contrast, is

potential-independent (Figure 13 a-d). RΩ depends on the three-electrode liquid electrolyte

setup (3LC) and is used for iR-correction of the depicted LSVs and CVs (Figure 13 e and Figure

S12 b). While both semi-circles of NiFe(-CO32-)-LDH@NF are well separated, the arcs

associated with NiFe(-ClO4-)-LDH@NF and NiFe(-Cl-)-LDH@NF partially merge. LFR and

HFR of uncoated NF overlap. The potential dependence of the LFR confirmed its interfacial

charge transfer character, RCT. At +1.53 VRHE, RCT was dominant, yet dropped sharply at +1.58

VRHE and +1.63 V RHE (Figure 13 a-d and Figure S12 c). High voltages increase the rate of redox

reaction, thus decreasing the RCT.150 Still, the uncoated NF continued to suffer from large RCT,

more than an order of magnitude larger than for the NiFe(-A-)-LDH@NF anodes.

NiFe(-Cl-)-LDH@NF exhibited the smallest RCT confirming measured trends (Figure 13 c).

HFR remained independent of applied anode potentials for all materials suggesting specific

bulk and surface characteristics of the anodes. The impedance characteristics of

NiFe(-A-)-LDH@NF point to non-conductive coating layers, where the HFR can be used

evaluating faradaic pseudo-capacitance of the electrode and, from that, its real electrochemical

surface area (ECSA).148,149,151 When correlated with RCT, deeper insights into the origins of the

reported trends are possible. ECSAs plotted in Figure 13 e has the smallest ECSA for uncoated

NF, while NiFe(-CO32-)-LDH@NF and NiFe(-Cl-)-LDH@NF exhibit the largest ECSA, with

NiFe(-ClO4-)-LDH@NF being in-between over the entire potential range. ECSA values of

NiFe(-CO32-)-LDH@NF and NiFe(-Cl-)-LDH@NF declined with anodic potentials, resulting

from oxygen bubble blocking of fractions of the surface. Our impedance analysis agrees with

the trends in SEM surface morphology, where NiFe(-CO32-)-LDH@NF showed a roughening

of the electrolyte-electrode-interface. Despite comparable ECSA, NiFe(-CO32-)-LDH@NF is

less active for OER than NiFe(-Cl-)-LDH@NF (Figure 13 f, g). Replacing Cl- with ClO4-

decreases the ECSA leading to declining electrocatalytic performance.

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

Since the compositional data (Table S1) and the relatively flat (plateau-like) relation between

Fe mol% and OER activity152,153 in the Fe mol% range of interest, we conclude that there is no

significant dependence of the LDH OER reactivity on the Fe mol% of the anodes. Therefore,

replacing CO32- with A-, effectively modulates the reactivity of the active surface sites by a

targeted modulation of the interlayer chemistry. The interlayer chemistry, that is the nature of

the intercalation ions and their coulomb interaction with water and the adjacent “Brucite”-type

sheets controls the transition of from the non-catalytic -phase to the catalytically active

- LDH phase. The tuning of the interlayer chemistry by halogenate ions results in a larger

molar ratio of the catalytically more active -LDH phase compared to -phase, potentially

supported by varied crystallite size (larger with Cl-, smaller with ClO4-).25,26,154,155

The following section and Figure 13 f-k correlates the OER activities and the EIS data measured

in 3LC.

The recorded OER currents of the prepared anodes were compared to uncoated NF. A

three-electrode liquid electrolyte setup was applied. Figure 13 f-g depicts LSV and Figure 13

h-k CV stability measurements (Figure S13, discussed in Supplementary Discussion 2). The

LSV trends (Figure 13 f) reflect the improved performance of the anion tuned

NiFe(-CO32-)-LDH coating (red) compared to the uncoated NF (black). Subsequent adjustment

of the interlayer chemistry of NiFe(-CO32-)-LDH@NF using Cl- (blue) and ClO4- (green)

impacted the catalysis (Figure 13 f, g). Cl- improved the OER activity, while ClO4- decreased it

and is therefore not further part of the discussion. The anodic shift of the Ni redox wave further

confirms formation of NiFe-LDH at the surface of the NF.156 With the ECSA values, we

evaluated the intrinsic, ECSA-normalized OER LSVs (Figure 13 g) confirming the correlation

between ECSA and OER activity. Thus, ECSA can be used to explain the drastically increased

OER activities in the three-electrode setup applying NiFe(-A-)-LDH coating to uncoated NF.

Closer inspection of the voltammetric profiles in Figure 13 f and g revealed that

NiFe(-CO32-)-LDH@NF exhibited a defined redox wave at +1.4 VRHE associated with the

oxidation of Ni2+ to higher oxidation states (Ni3+/4+). We directly tracked this catalyst redox

process by operando optical analysis (change from metallic grey (Ni2+) to black (Ni3+/4+) in

Figure S14). Uncoated NF remained metallic grey. In order to exclude the influence of Ni and

Fe leaching on the catalytic activity, we performed ICP-OES of the electrolyte before and after

stability measurement with the most active catalyst NiFe(-Cl-)-LDH@NF (Table S2).

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

We detected a negligible amount of Ni (2.5 µg) and Fe (1.5 µg) from which we deduce a small

influence on the catalytic activity.

4.3.3 Translating RDE to AEM Water Electrolyzer Cell Performance

Overall goal of this study, like many other current studies, was tuning and enhancing the

performance of the AEMWE liquid electrolyte device guided by the performance gains

observed in three-electrode screenings. To check whether NiFe(-A-)-LDH@NF anodes OER

performance is actually transferable into practically relevant AEMWE anode environments, we

first conducted measurements for the three-electrode setup at 60 °C (Figure S15) for a more

realistic comparison. Then identical NiFe(-A-)-LDH coatings grown on NF were employed as

5 cm2 anodes in single AEM electrolyzer cells (AEM Cell). The electrolyzer cell setup is

provided in Figure S16 and an evaluation of different cathodic PTLs, temperatures, pH,

membranes and Pt cathode loadings are presented in Figure S17. The AEMWE polarization

curves are shown in Figure 14 a. It is very important for the discussion, that the prepared anodes

do not contain liquid solubilized binder or ionomer-based catalyst inks or powder catalyst

layers. This means that the analysis and discussion of the translation of three-electrode to cell

performances, in contrast to MEA and other catalyst-inkcoated design, can exclude the

consideration of catalyst ink parameters such as ionomer/catalyst ratios, catalyst loadings, film

porosity characteristics and other catalyst layer-related parameters impacting performance. The

ECSA and roughness factors of the cell anodes are consistent and identical to those in

three-electrode since synthesis parameters were kept constant. Thus, a comparative analysis of

how catalyst screening translates into cell performance is more possible and can be reduced to

fewer control parameters.

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

Table 1: Summarizing and comparing the 3LC with the

AEM cell at 60°C.

Figure 14: Activity tests and trends of NiFe(-A-)-LDH@NF

and uncoated NF measured in AEMWE single cell. (a)

Comparison of the polarization curves of NiFe(-A-)-LDH@NF

CO32- (red) and Cl- (blue) with uncoated NF (black), dashed

lines denote as iR-corrected polarization curves. (b)

comparative activity improvement factors 𝐟=𝐣𝐠𝐞𝐨𝐦.,𝐜𝐚𝐭 𝐚𝐭𝟏.𝟓𝟐𝐕

𝐣𝐠𝐞𝐨𝐦.,𝐍𝐅 𝐚𝐭𝟏.𝟓𝟐𝐕 of

anion tuning in three-electrode liquid cell (3LC) at 25°C, 60°C,

AEMWE single cell (AEM Cell) and blowup (same y-axis

labeling but different scale) of the AEM cell. Data for the AEM

cell are iR- and HER corrected (Figure S18 c). f was calculated

at the highest possible iR-corrected potential (EiR=1.52 VRHE)

measured in 3LC at 60°C (Figure S15).

The cell polarization curves in Figure 14 a qualitatively reproduce the three-electrode activity

trends: The NiFe(-CO32-)-LDH@NF anode showed superior electrocatalytic activities

compared to uncoated NF, NiFe(-Cl-)-LDH@NF offered some additional performance benefits

over NiFe(-CO32-)-LDH@NF. Thus, our data demonstrate that the structural and compositional

tuning of uncoated NF with LDH-catalyst, is principally scalable to single AEMWE cell level.

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

We note, however, that two-electrode electrolyzer cell tests did not yield the same quantitative

improvement factors as those found in three-electrode screenings, this is summarized in

Figure 14 b and Table 1.

The membrane cell environment appeared to equalize the LDH-catalysts, while significantly

raising uncoated NF performance. More specifically, we found that polarization curves of

uncoated NF anodes trail NiFe(-CO32-)-LDH@NF anodes by only about 20%, iR-corrected

polarization curves by about 40% and iR- and HER-overpotential corrected by 80% in OER

current density despite the sharply lower ECSA of uncoated NF. Calculation of the HER-

overpotential is explained in SI (Figure S18). Thus, despite identical ECSA of the applied

anodes, it was not possible to translate the 91x anode performance gain measured in three-

electrode setup to the single cell AEMWE design. The uncoated NF performed much better

than predicted. Furthermore, the calculated f depicted in Figure 14 b and Figure S19 a and b

clearly show the iR- and HER-contribution to the overall performance discrepancy between

three-electrode and AEMWE single cell activity, which is only 60%.

To further understand the serious performance discrepancies between three-electrodes and

electrolyzer cells in more detail, we re-investigated the cell anode environments including

impedance, electrocatalytic behavior and chemical states before and after electrolysis. We

uncovered component limitations of commercial alkaline electrolyzer setups that compromise

quantitative translations of OER anode catalyst performances.

In particular, Fe 2p core level photoemission of the initially Fe-free NF (Figure S20)

demonstrates the presence of Iron and NiOOH species at the surface cell testing (Figure S21).

The well-documented catalytic activation of Nickel OER catalysts by atomic Fe deposition156-

159 accompanied with anodic shift of the Ni redox wave fully accounts for the increased OER-

activity of the uncoated NF and is further reported with three-electrode measurements of the

NF after AEMWE single cell testing in SI (Figure S22). This partly explains the much smaller

OER activity differences between NF and NiFe(-A-)-LDH@NF anodes in the AEMWE cell

tests. As the level of Fe impurities in the alkaline electrolyte was identical in three-electrode

screening and electrolyzer cells (same commercial alkaline reagent used), we conclude that, Fe

impurities may originate from Fe corrosion of stainless-steel electrolyzer components. We use

EIS under OER conditions comparing NF with the NiFe(-A-)-LDH@NF anodes, all performed

in the AEMWE cell.

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

Exemplary Nyquist-Plots and EEC applied for the fitting are depicted in Figure S23 a and c. As

expected from our three-electrode screenings, NiFe(-CO32-)-LDH@NF showed lower RCT

(enhanced intrinsic OER activity) but increased RΩ resulting from the catalytic active LDH

overlayer (Figure S23 b and Supplementary Discussion 3).

Impedance data of the electrolyzer cell agrees well with the iR-corrected polarization curves of

Figure 14 a. Furthermore, the RCT and EIS-based ECSA of NF after AEMWE single cell testing

measured in three-electrode setup reveal decreased RCT but only insignificant increased ECSA

of the uncoated NF. This is why, in order to account for the equalization of the performance of

the anodes, we further propose OH- transport between cathode, membrane and anode as

bottleneck for translating catalyst three-electrode activity trends accurately into AEMWE. We

recall that optimal PEMWE performance requires perfect chemical and electrical contact

accompanied with ideal ionic and electronic pathways providing optimized catalyst

utilization.160 As OH- ions migrate from cathode to anode to replenish the local loss of OH- and

thereby regulate the local anode pH, only the anode surface closest to the solid membrane has

the strongest impact on OER. The large ECSA, previously determined as controlling in the

liquid electrolyte, three-electrode structure, turns out not decisive here, since the majority of

the rough nanostructured anode surface is not in direct contact with the membrane. In other

words, only a small fraction of the high ECSA designer anodes is actually involved in a high

pH OER catalysts. Both mechanisms, the Iron species incorporation as well as the OH- limited

local OER catalysts can account for the incomplete translation of anode performance from

screening to electrolyzer cell. Considering the identical liquid electrolytes (0.1 M KOH) in

three-electrode setup and AEMWE single cells, we are inclined to identify low, high local pH

as major reason for limited performance gains of NiFe(-A-)-LDH@NF anodes over NF, despite

their still significantly larger ECSA.

Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Water

Electrolyzers: From Catalyst Screening to Single Cell Performance

4.4 Conclusion

This study critically analyzed scaling behavior between three-electrode and single AEMWE

testing. A binder-free catalyst layer formed enables the comparison, since complicating catalyst

ink and layer parameters are absent. Combining NiFe-species with NF resulted in 91x

electrocatalytically OER increase in three-electrode setup compared to uncoated NF.

Impedance spectroscopy unraveled ECSA originating for favorable increased OER of

NiFe(-A-)-anodes. NF and NiFe(-A-)-LDH@NF anodes kept order of the OER activity when

applied to AEMWE single-cell-level. Despite identical ECSAs of the applied anodes, the same

quantitative improvement was not found for two-electrode setup as described for three-

electrode screenings. Strong improvement of NF is achieved via Fe incorporation and local pH

conditions. Solid electrolyte setups only offer an improvement of 1.9x iR- and HER-corrected.

Large ECSA, previously origin for OER activity in the three-electrode setup was not a control

parameter, because for cell performance presumably contact shortcomings between rough

nanostructured anode surface and membrane limit catalyst utilization. Our study confirms the

importance and usefulness of three-electrode catalyst screenings and development efforts for

AEMWEs. However, at the same time, unveils limitations associated with the quantitative

prediction and translation of screening performance results into AEMWE single cell or even

multi cell setups.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

5 Efficient Paired Electrolysis of

5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at

Industrial Current Densities

This chapter is reproduced from Hauke, P., Klingenhof, M., Wang, X., de Araújo, J. F. &

Strasser, P. Efficient electrolysis of 5-hydroxymethylfurfural to the biopolymer-precursor

furandicarboxylic acid in a zero-gap MEA-type electrolyzer. Cell Reports Physical Science 2,

100650, doi: https://doi.org/10.1016/j.xcrp.2021.100650 (2021). With permission of Cell

Reports Physical Science.155 (Accepted manuscript version)

The study was designed and executed by P.H.. M.K. support in the evaluation of the catalysts.

The SEM images were done by X.W.. J.F.A. designed the DEMS setup and supported the

DEMS measurements.

5.1 Abstract

Replacement of today’s established chemical production processes by “green” sustainable

alternatives has become a scientific and technological priority. The oxidative conversion of

5-Hydroxymethylfurfural (HMF) to the biopolymer component 2.5-Furandicarboxylic acid

(FDCA) is a promising emerging green process, in particular when performed on platinum

group metal-free catalysts and paired with green electrolytic hydrogen production at

significantly lower cell potentials. In the present contribution, we present a novel family of

active selective and stable interlayer anion-tuned NiX (X=Fe, Mn, Co, V) bimetallic layered

double hydroxide (LDH) catalysts for the selective oxidation of HMF to FDCA in continuously

operating, zero-gap MEA-type single electrolyzer cell at industrial current densities. We

discovered that tuning the structural interlayer distance of the catalyst -LDH phase by anion

exchange gives rise to previously unachieved catalytic performance for the anodic production

of the biopolymer building block. More specifically, a Cl- -exchanged NiFe LDH catalyst

showed previously unachieved 100% conversion XHMF, 100% faradaic efficiency FEFDCA,

100% yield YFDCA and 100% selectivity SFDCA.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

Operando Differential Electrochemical Mass Spectrometry (DEMS) analysis revealed the

electrode window of opportunity for the perfectly selective HMF conversion, where the

competing oxygen evolution process appears to be suppressed by surface intermediate

adsorption. The role of the catalyst dopants, their real surface areas, the stability of the catalytic

interface, and aspects of its favorable techno-economics are discussed. Paired with efficient

green hydrogen production at the cathode at industrially relevant current densities, the MEA-

type biomass electrolysis device yields previously unachieved product space time yields and

offers techno-economic cost advantages.

5.2 Introduction

Over the last decade, the worldwide production of plastic increased by dramatic 47%. In 2019

alone, 368 million metric tons of synthetic polymers with a significant carbon emission were

produced.161 For instance, given a carbon footprint of 1.538 kg CO2 per 1 kg Polyethylene

terephthalate (PET) (85% recycled content), this corresponds to 566 million metric tons CO2.162

Green sustainable catalytic process alternatives replacing high temperatures, pressures, fossil-

based feedstocks or toxic solvents of today’s established productions are therefore urgently

needed. An important industrial example of such a green alternative to PET is Poly-(ethylene

2.5-furandicarboxylate) (PEF) that is obtained by polymerization of 2.5-Furandicarboxylic acid

(FDCA), a thermally catalyzed oxidation product of the biomass-derived

5-Hydroxymethylfurfural (HMF) molecule. Even though the conventional thermal, aerobic

oxidation of HMF uses biomass as a starting material, this route continuous to involve elevated

temperatures (80-100°C), high pressures, and unsustainable noble metal-based catalysts.163-166

This is why ambient-temperature, renewable electricity-based electrochemical processes are an

attractive alternative. Thanks to mild operating conditions and wider availability of affordable

renewable energy due to power surpluses, electrochemistry meets a number of the above

requirements. Another possibly attractive application of the 5-HMF to FDCA electrochemical

oxidation reaction lies in its use at Power-to-X electrolyzer anodes replacing the energetically

demanding oxygen evolution reaction (OER), while boosting cell efficiency thanks to lower

required cell potentials. The overall half-cell reaction of the heterogeneously catalyzed

electrochemical 6-electron 5-HMF oxidation to FDCA reads

C6H6O3 + 6OH− → C6H4O5 + 4H2O+ 6e−, (Equation 42)

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

while the principle mechanistic reaction pathway using structural formula is shown in Figure

15. The 5-HMF oxidation to FDCA is a promising emerging industrial process,167,168 in

particular when paired with the half-cell reaction involving the generation of green electrolytic

hydrogen on platinum group metal-free catalysts according to

6 H2O +6e− → 3 H2+ 6OH− , (Equation 43)

resulting in the overall electrolyzer cell reaction

C6H6O3 + 2 H2O→ C6H4O5 + 3 H2 ΔG°=65.62 kJ mol−1. (Equation 44)

From this relation it is evident that the required Gibbs energy169 to generate one mol H2 amounts

to only 22 𝑘𝐽, less than the 237 𝑘𝐽 required when coupled to the oxygen evolution reaction

(OER), not counting the valorized FDCA production.

Figure 15: Electrochemical conversion of 5-Hydroxymethylfurfural over NiFe(-Cl-)-LDH@NF in a zero-gap flow cell

electrolyzer. Reaction pathways of the alkaline 5-HMF oxidation via 5-Hydroxyfuran-2-carboxylic acid (HFCA) or

Furan-2.5-dicarbaldehyde (FDA) to 5-Formyl-2-furancarboxylic acid (FFCA) and the desired final product

2.5-Furandicarboxylic acid (FDCA). Scheme of the zero-gap MEA-type electrolyzer with PTFE gaskets (white), Anodic and

Cathodic electrodes (ocher and black, respectively), and membrane (yellow). The serpentine flow fields are a direct part of the

current collectors.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

The electrochemical 6-electron HMF to FDCA process in itself results in an important industrial

chemical intermediate by valorizing a readily available sustainable precursor. Simultaneously,

it can serve as a practical alternative to conventional oxidative half-cell reactions, such as the

oxygen evolution reaction (E°= 1.23 VNHE) in the design of energy-efficient hydrogen

generating electrolytic cells and processes. Replacing the kinetically sluggish oxygen evolution

reaction (OER) 126,147 by the HMF/FDCA anode with its substantially lower standard potential

of 0.113 VNHE169 results in a “paired” electrochemical process with a standard Gibbs free energy

of 65.62 kJ mol-1 and a resulting, lower cell voltages, and thereby greatly improved the

electrolyzer total energy efficiency.

Nickel-based borides, nitrides, sulfides and phosphates as well as NiFe- Layered double

hydroxides have demonstrated moderate activity for the 5-HMF oxidation reaction (HMFOR),

however, these families of electrocatalysts have invariably resulted in limited space time yields

and efficiencies.64,70,73,81,83,84,170 Nanostructured catalyst designs, such as other Ni-based LDHs

and anion-doped LDH variants, on the other hand, have never been explored for the HMFOR

process.

In this work, we present a family of novel, bulk Ni foam (NF)-supported, “Cl-” anion-

exchanged NiFe layered double hydroxide catalysts, referred to as “NiFe(-Cl-)-LDH@NF”, for

the oxidation of 5-HMF to FDCA in a zero-gap continuous MEA-type electrolyzer cell

(Figure 15). Electrodes prepared using a facile binder- and noble metal-free approach

demonstrated unprecedented catalytic activity, selectivity, and stability. Operando DEMS

studies yielded the selective HMF oxidation potential windows. In direct comparison to noble

metal catalysts, our physicochemical characterization studies and electrochemical tests at

industrial currents densities reveal previously unachieved time-stable FDCA space time yields

(STY) that demonstrate the practical significance of the new catalysts, the described electrode

interface, and the associated electrolyzer design.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

5.3 Results & Discussion

5.3.1 Synthesis and Structure of Anion-Modified -NiX-LDHs Supported on Ni

Nanofoams

Anion (A)-exchanged - bimetallic NiX (X=transition metal) layered double hydroxides

(LDH), including NiFe-, NiMn-, NiCo-, NiV-LDH, were prepared as i) unsupported

nanostructured powder catalysts in their stable layered -form, as well as ii) catalytic coatings

directly grown on the surface of a high surface area Ni nanofoam (denoted

NiX(-A-)-LDH@NF). The synthesis was achieved using solvothermal, microwave-assisted

autoclave synthesis strategies,147 followed by a novel modification of the LDH interlayer

chemistry by replacing the default interlayer anions A=CO32- by A=Cl- and A=ClO4- anions.

The synthesis of each NiX LDH catalyst was individually optimized for catalytic performance

by systematic variation (not shown) metal precursor amounts and preparation conditions. The

formation of crystalline hydrotalcite-type LDH phases thereby served as primary selection

criterion, irrespective of the resulting Ni:X ratio in the final catalyst. Details of the different

synthesis steps are provided in the Methods section. The rationale of the anion exchange was i)

enhancement of the real surface area of the catalyst and ii) judicious tuning of the interlayer

Brucite layer distance which was assumed to affect the anion-cation exchange rate during the

- transition to the catalytic active catalyst phase.26 As direct element-specific characterization

of the in situ-grown LDH@NF catalyst turned out very challenging against the strong Ni and

Ni-oxide background of the bulk foam, structural and compositional analyses were carried out

on the catalyst nanopowders.

Inductively-coupled plasma optical emission spectroscopy (ICP-OES) provided the Ni:X metal

ratio and the total metal percentage in the powder catalysts (Table S 3). While the total metal

amount remained constant among the catalysts, the Ni:X ratio of the optimized crystalline

-phase increased from NiV to NiFe to NiCo to NiMn due to varying metal precipitation rates.

X-ray diffraction (Figure 16 a) showed the characteristic range of Bragg reflections associated

with the crystalline 2D NiX(-CO32-)-LDH structure, here represented by the archetypical

trigonal-rhombohedral LDH Hydrotalcite reflections (JCPDS: 00-014-0191) at 11.5° (003), 23°

(006), 35° (012), and 60° (110) for all NiX(-CO32-)-LDH materials.147,171-173

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

NiCo(-CO32-)-LDH showed a weak (003) reflection and varying morphology compared to

NiFe- NiMn- and NiV(-CO32-)-LDH evident by the Scanning electron micrographs (SEM) in

Figure 16 b)-e) and Figure S 24 a)-d). To learn more about the growth morphologies of the

nanoscale LDH catalyst on top of the NF, a time-resolved catalyst growth study was conducted

which demonstrated that controlled mono- and multi-layer growth of NiV(-CO32-)-LDH@NF

was possible by optimizing the catalyst precursor concentrations and reaction times

(Figure S 25).

The operando electrochemical surface areas (ECSA) of the bimetallic NiX(-CO32-)-LDH@NF

catalysts at operating cell conditions were evaluated using impedance measurements coupled

with faradaic capacitance circuit modeling.25 At 1.58 VRHE, the NF support displayed 1.90 cm2,

while NiFe(-CO32-)-LDH@NF, NiMn(-CO32-)-LDH@NF, NiCo(-CO32-)-LDH@NF and

NiV(-CO32-)-LDH@NF showed 187.83 cm2, 37.75 cm2, 50.73 cm2, 567.16 cm2, respectively.

The detailed ECSA trends (Equation 37) are provided in Table S 4 and Figure S 26.

Figure 16: Structural Characterization of bimetallic LDH catalysts: Characterization of NiX(-CO32-)-LDH nano

powders and directly grown NiX(-CO32-)-LDH on Ni nanofoam. a) XRD pattern of powdered NiFe(-CO32-)-LDH (red),

NiMn(-CO32-)-LDH (blue), NiCo(-CO32-)-LDH (green), and NiV(-CO32-)-LDH (purple). As a representative reference

rhombohedral Hydrotalcite, JCPDS: 00-014-0191 was added. b)-e) SEM images of NiFe(-CO32-)-LDH), NiMn(-CO32-)-LDH,

NiCo(-CO32-)-LDH and NiV(-CO32-)-LDH directly grown on Ni NF at 5000x magnification. The color code corresponds to a)

(clean NF is shown in Figure 17 c).

Inductively-coupled plasma optical emission spectroscopy (ICP-OES) provided the molar Ni

to Fe ratio and the total gravimetric metal percentage (wt%) of catalyst powders after

modification of the LDH interlayer chemistry (Table S 5).

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

Both the total metal molar amount and the resulting Ni ratio decreased after anion exchange

treatment which suggested a partial leaching of Ni. As a closer structural analysis around the

(003) reflection in Figure 17 a revealed, when going from carbonate anion- LDH to the

chloride- anion LDH and further to the perchlorate-anion LDH, an expansion of the interlayer

distance is clearly observed.147,171-173 A shift in the 003 reflection (Figure 17 b) of the carbonate,

chloride and perchloride doped NiFe(-A-)-LDH diffraction patterns suggested a varying, anion-

dependent interlayer distance between adjacent 2D Brucite layers. Hence, the scattering,

microscopic, and spectroscopic data suggest that controlled anion exchange enabled a tuning

of the atomic interlayer distances between adjacent Brucite sheets in the various LDH catalyst.

To assess the electrochemical surface area of the anion-exchanged NiFe LDH catalysts at

operando cell potentials a similar faradaic impedance analysis was conducted as above.25 The

experimental estimates of the ECSA at +1.58 VRHE of the NiFe(-Cl-)-LDH@NF and

NiFe(-ClO4-)-LDH-@NF revealed 139.04 cm2 and 45.97 cm2, respectively (Table S 6).

Figure 17: Structural Tuning of the Layered Double Hydroxide Catalysts: Characterization of the interlayer distance

and morphology of NiFe(-A-)-LDH and NiFe(-A-)-LDH@NF after anion exchange. a) Crystal structure and increasement

of the interlayer distance during the anion exchange steps with chloride and perchlorate. b) XRD pattern of powdered

NiFe-LDH (red), NiFe(-Cl-)-LDH (blue), NiFe(-ClO4 -)-LDH (green) visualizing the (003) reflection shift within the anion

exchange. c)-f) SEM images of the different catalysts at 5000x magnification. The color code corresponds to b). Additionally,

pure NF (black) was added as a reference.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

5.3.2 Surface Redox Electrochemistry, Membrane and Cation Effects

RDE surface voltammetry of (un)catalyzed Ni foams. The surface electrochemistry of

uncoated and catalyst-coated Ni foam (NF) electrodes was investigated using Rotating Disk

Electrodes (RDE) in HMF-free and HMF-containing KOH electrolytes. These Disk electrode

tests have preliminary screening character and were initially applied to obtain a sense for the

relative location and width of the electrode potential window of interest for the Niz+ and

HMF-related redox waves. Except for selected comparisons, the electrolyte pH was kept at

pH=13 to minimize parasitic side reactions between electrolyte and 5-HMF toward undesirable

formation of humins (Figure S 27).42,61 Figure S 28 reports RDE-type measurements of Ni foam

(NF), a Nickel disc (ND), and pure carbon paper reference (CP) as working electrodes (WE)

with (green, blue, purple) and without (red, black) 10 mM HMF reactant in 0.1 M KOH. The

voltammetric data suggested a coincidence of the Ni2+/3+ redox transition and the onset of the

electrochemical 5-HMF oxidation reaction (HMFOR) between +1.35 VRHE and +1.6 VRHE.59,81

At identical geometric surface area, the foam reference catalyst showed larger real surface area

which explains the apparent enhanced geometric redox currents. To demonstrate synthesis-

performance correlations that guided our catalyst optimization, Figure S 29 displays surface

voltammetry as well as catalytic reactivity of the NiV-LDH-catalyzed NF

(NiV(-CO32-)-LDH@NF) as a function of precursor concentration and catalyst growth time (see

catalyst in Figure S 25). For NiV-LDH, significantly improved stability, lower onset potential,

and higher currents within a specific HMFOR-selective region (see discussion further below)

between +1.4 VRHE - +1.6 VRHE were observed when employing increasing precursor

concentrations and reaction times at 160°C; these preparation conditions were adopted for all

subsequent full electrolyzer flow cell measurements using NiV(-CO32-)-LDH@NF and alike

procedures with their optimized synthesis conditions were applied to the other catalysts.

Preliminary electrolyzer cell tests. Industrially relevant electrocatalytic HMF oxidation

currents up to the 1 kA/m2 range and associated product space time yields of the NF-supported

anion-modified Ni-based LDH catalysts were recorded in a single flow cell electrolyzer

configuration consisting of LDH catalyst-coated NF substrates (CCS) as anode, an anionic

exchange membrane to ensure hydroxide migration/diffusion across the electrolyte, and a

state-of-art hydrogen gas evolution cathode to minimize kinetic barriers and activation losses.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

Pure 0.1 M KOH was chosen as circulating catholyte, while 5-HMF-containing 0.1 M KOH

served as circulating anolyte. This cell configuration allowed us to treat the uncompensated

ohmic electrolyte resistance-corrected cell voltage as an approximation of the kinetic anode

overpotential, assuming negligible cathodic overpotentials, mass transport overpotentials in the

electrode layers. Prior to electrolyzer cell tests employing the novel anion-doped LDH

catalyst-coated NF anodes, the baseline performance of the uncatalyzed Ni foam support was

investigated in pure KOH and varying 5-HMF/KOH concentrations (0 mM, 5 mM, 10 mM,

15 mM) at pH=13 (Figure S 30/Figure S 31). Current-potential polarization curves were

recorded during linear scans of the electrolyzer cell voltage from +1.0 Vcell to +1.7 Vcell. A

dependence of the onset potential of the 5-HMF oxidation process on the 5-HMF reactant bulk

concentration was evident. A detailed voltametric analysis of Figure S 30 shows that the

baseline Ni2+/3+ redox transition at +1.4 Vcell as well as the onset of molecular oxygen evolution

> +1.5 Vcell were still discernible at the lowest 5 mM 5-HMF concentration only. Above 5 mM

5-HMF, both redox processes remained fully masked by the HMFOR redox features until,

anodic of +1.6 Vcell, the oxygen evolution activity (OER) prevailed. We will provide further

experimental evidence below for this adsorptive masking effect by 5-HMF further below using

Differential Electrochemical Mass Spectrometry (DEMS). This finding calls for careful control

of the local 5-HMF concentration during long-term electrolysis and resulting decreasing local

interfacial 5-HMF concentrations in order to maintain the target of 100% selective HMF

oxidation. Clearly, the LDH@NF anodes discussed further below will lower required applied

electrode potential to achieve the desired catalytic reaction rate and selectivity. From the

preliminary electrolysis in Figure S 30, a constant electrolysis cell potential of +1.56 Vcell was

selected and applied in subsequent catalyzed NF-based 5-HMF electrolysis tests.

Membrane and cation selection using uncatalyzed Ni NF anodes. Prior to operation of the

HMF biomass electrolyzer using anion-doped LDH catalysts, a membrane and cation

optimization study was carried out with uncatalyzed Ni nanofoam (NF) as a reference anode

and a state of art low-PGM cathode. Six anion exchange membranes (AEMs) from different

commercial providers (Fumatech, IONOMR, and Dioxide Materials) were compared and

contrasted under identical conditions of LSVs and electrolysis at a constant current density over

2h (Figure S 32/Figure S 33 and Table S 7).

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

Figure S 32 illustrates the dramatic impact of the membrane type on the baseline biomass

oxidation, which resembles the membrane effects reported for AEM water electrolyzer.174 More

specifically, Figure S 32 reveals significant activity variations in the HMFOR potential region

of interest. Considerable current losses were observed when moving from the FAA-3-PK-130

(Fumatech) to a Dioxide Materials and an IONOMR membrane. FAA-3-PK-130 achieved the

best electrochemical performance confirmed by the largest electrochemical 5-HMF conversions

and FDCA yields (Table S 7) and was therefore used for further investigations. We suspect the

favorable HMFOR performance of FAA-3-PK-130 was caused by its network structure

(Figure S 34 a) which prevents membrane channel blockage by HMF and other reaction

intermediates or products during the electrolysis. Indeed, Figure S 34 b demonstrated

destructive deposits on the IONOMR HNN8 membrane after 2 h electrolysis. Consequently,

the current density, HMF conversion and FDCA yields decreased. Detailed information about

the specific HPLC analysis protocol used to determine the electrolysis figures of merit are

provided in the Method section and in Figure S 35. Finally, the influence of alkali metal cations

in the electrolyte on the HMF oxidation reactivity was evaluated. Figure S 36/Figure S 37

reports LSVs of HMF oxidation on uncatalyzed NF in different alkali metal hydroxide

electrolytes (LiOH, NaOH, KOH, RbOH, and CsOH). In the HMFOR region of interest, some

variations in catalytic activity were apparent that were previously attributed to variations in the

basicity of the cations.175 Based on its favorable performance in our preliminary cell tests,

potassium cations were considered in further investigations below.

5.3.3 Continuous GDE Electrocatalysis and FDCA Product Space Time Yields

NiX(-CO32-)-LDH@NF HMFOR anode catalysts. Using the optimized electrolyzer

components, the continuous electrolytic oxidation of 5-HMF to FDCA was examined using the

anion-exchanged - NiX(-CO32-)-LDH catalysts inside the HMFOR-relevant potential range of

+1.4 VRHE to +1.525 VRHE (equal to +1.4 Vcell to +1.56 Vcell) and at geometric current densities

of up to jgeom= 0.75 kA m-2.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

Electrolyzer cells equipped with anodes of directly grown NiFe-, NiMn-, NiCo- and

NiV(-CO32-)-LDH@NF as well as NiFe(-Cl-)-LDH@NF and NiFe(-ClO4-)-LDH@NF were

anodically polarized using LSV protocols (see Figure 18, Figure S 38-Figure S 41), followed

by 2 h constant potential electrolysis at cell voltage of E= +1.56 Vcell and concurrent

time-resolved HPLC-based product analysis (Table S 8/Table S 9).

All electrolysis tests were repeated at least twice and reproducibility of trends were confirmed.

Figure 18 a compares the NiX(-CO32-)-LDH@NF (X= Fe, Mn, Co, V) catalysts with the

uncoated NF reference revealing a strong catalytic enhancement of the catalyzed anode

materials in the HMFOR potential region of interest (green box). None of them displayed the

characteristic Ni redox wave near E= +1.4 Vcell at the used 5-HMF concentration, in agreement

with the results above (Figure S 30/Figure S 31). The NiMn and NiV catalysts exhibited a

slightly more cathodic onset potential of the faradaic 5-HMF oxidation current compared to the

NiFe and NiCo ones. Despite their more anodic onset potentials, however,

NiFe(-CO32-)-LDH@NF and NiV(-CO32-)-LDH@NF exhibited the largest current densities

past +1.55 Vcell (see green potential range of preferred selectivity in Figure 18 a). The significant

variation in the ECSA of the NiX catalysts was not reflected in the magnitude of the

experimental HMFOR current densities inside the FDCA selective potential range (green box).

The nature of the dopant metal X affected the electrochemical FDCA selectivity, SFDCA, the

faradaic FDCA efficiency, FEFDCA, as well as its yield, YFDCA, after 1 h and 2 h electrolysis at

+1.56 Vcell. This is displayed in Table S 8 and in the process performance radar plot in

Figure 18 c that documents the rising metrics from NiCo(-CO32-)- to NiMn(-CO32-)- and

NiFe(-CO32-)- to the best performing NiV(-CO32-)-LDH@NF. NiFe(-CO32-)-LDH@NF and

NiV(-CO32-)-LDH@NF displayed record-high 100% 5-HMF conversion, but differed in SFDCA

and FEFDCA. The NiV(-CO32-)-LDH@NF also displayed previously unachieved 75% FDCA

selectivity and an 98% FEFDCA.

Correlating the catalytic performance trends (Figure 18 c) to the catalyst composition

characteristics (Table S 3/Table S 4), we note that the significant differences in the catalytic

activity are by no means mirrored by the molar Ni:X ratios of the second metal X: The NiV

(1.5:1) and NiMn (23:1) LDH catalysts both show favorable performance, even though their Ni

to V/ Mn ratios vary strongly. A combined operando X-ray diffraction and DFT computational

study, performed on the LDH catalysts, helps rationalize this complex behavior.25

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

The present data suggest that Ni-V dual surface sites offer synergistic catalytic oxidation

reactivity, while dual Ni-Mn site exhibit negative synergy.25 This is why even low Mn at%

results in favorable reactivities, thereby maximizing the population of the more active Ni-Ni

dual surface sites. Also, Ni-Fe (3:1) LDH, ranging in the top three catalysts here, was found to

offer kinetic synergies through dual Ni-Fe surface sites, which is why a Ni:X ratio approaching

1:1 performs more active.

Increasing the Ni:Fe at% ratio to 1:1 could, according to the DFT findings, maximize the

number of active dual Ni-Fe sites; however, the synthesis of this ratio resulted in a low

crystalline, and hence lower active catalyst material. The crystalline NiCo LDH (7:1) ranged

lowest in performance, in line with the identified negative oxidation synergy between Ni and

Co dual surface sites.25

Performance metrics, such as FEFDCA and SFDCA, are typically not dependent on the active

catalyst surface area at controlled electrode potential conditions, which is why they reflect

rather intrinsic performance characteristics of the active surface sites. The real ECSA values of

the NiX catalysts showed some correlation with the experimental geometric faradaic current

densities at +1.56 Vcell and the conversion, X, (NiV ≥ NiFe > NiCo > NiMn). This suggests that

the electrolyzer anode operates in a reactive charge transfer-limited regime, possibly limited by

the surface coverage of reactive intermediates identified above (also see DEMS results below).

Based on the rising electrolyzer polarization curves within the HMFOR-selective potential

range (green box) in Figure 18, we exclude the presence of 5-HMF mass transport limitations

as origin for the weak correlation of the real ECSA with performance. In summary, among the

four NiX(-CO32-)-LDH@NF catalysts we can identify a catalytically superior, preferred set of

LDH catalysts including X=Fe,V,Mn, while the NiCo(-CO32-)-LDH@NF offered lower

performance metrics. The reference Ni NF performed poorly due to its initially largely metallic

Ni character.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

Figure 18: Single Electrolyzer Cell Performance of the new NiX(-A-)-LDH@NF catalysts: a)/b) Flow Electrolyzer Cell

based linear scan voltammograms (LSV) using NF and NiX(-A-)-LDH@NF (X=Fe,Mn,Co,V; A=CO32-,Cl-,ClO4-) anodes, C

supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used as catholyte

and 10 mM HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters:

flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell , scan rate 10 mV s-1, the highlighted green

box denotes the selective HMFOR potential region. c)/d) Catalytic performance overview of the set of NiX(-A-)-LDH@NF

(X=Fe,Mn,Co,V; A=CO32-,Cl-,ClO4-) catalysts using consistent color coding as in a) and b). 5-HMF Conversion, X, FDCA

yield, Y, FDCA selectivity, S, and Faradaic efficiency, FE, are given on a scale from 0-100%, while the current density at

+1.56 Vcell is plotted from 0-20 mA cm-2 in c) and 0-25 mA cm-2 in d).

Anion-doped NiFe(-A-)-LDH@NF HMFOR anode catalysts. To further improve the

electrochemical yield and faradaic efficiency of the 5-HMF to FDCA process, selected anion

exchange of CO32- with Cl- and ClO4- was carried out using the two most active

NiFe(-CO32-)-LDH@NF and NiV(-CO32-)-LDH@NF catalysts. For the anion-exchanged

NiV(-A-)-LDH@NF, the resulting FDCA yield (100%) and selectivity (73%) remain similar to

those of the carbonate version (Figure S 39/Figure S 40), which is why it was not further

investigated.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

By contrast, anion exchanged NiFe(-A-)-LDH@NF offered previously unachieved catalytic

HMFOR performance metrics in terms of XHMF, FEFDCA, YFDCA and SFDCA (Figure 18 b/d,

Figure S 41, Table S 9).

The NiFe(-Cl-)-LDH@NF (Figure 18 b, Table S 9) revealed a sharp catalytic performance boost

indicating a nearly perfect performance: With 100% XHMF, 100% FDCA selectivity SFDCA, and

100% faradaic efficiency FEFDCA, it is the most active material presented in this work

(Figure 18 d, Figure S 43) and unsurpassed by earlier work. Simply rationalizing this

outstanding performance by higher roughness contradicts the actual experimental ECSA trend

(Table S 6) and SEM morphology (Figure 17 c-f, Figure S 44). We attribute the favorable

intrinsic activity to the improved activation of the -LDH interlayer structure to form the

catalytically active -LDH phase (Figure 17), possibly supported by the slightly varied

crystallite size (larger with Cl-, smaller with ClO4-).176,177 Indeed, -NiFe(-CO32-)-LDH@NF

catalysts, when operated in Cl--containing seawater electrolytes under alkaline conditions,

exhibited enhanced - phase transformation kinetics and ultimately larger ratios of the

catalytically active -LDH structure.154

Figure 19 a and Figure S 45 illustrate the strong pH dependency of the HMFOR. By increasing

the pH from 13 to 14 it was possible to reduce the reaction time to full 5-HMF conversion

XHMF=100% and perfect FDCA selectivity SFDCA=100% by more than half to 45 min

(Figure S 46 and Table S 10). Despite this favorable reactivity, since formation of undesired

side products such as humins cannot be excluded at pH=14 at longer reaction times, our stability

tests were carried out in more moderate pH=13.

Differential Electrochemical Mass Spectrometry (DEMS). To pinpoint the HMFOR

reactivity range and provide direct evidence for the suppression of the OER process by the HMF

surface oxidation intermediates, we carried out real-time differential electrochemical mass

spectrometric (DEMS) tracking of molecular oxygen formation (at m/z= 32) as function of

applied electrode potential, using the most active NiFe(-Cl-)-LDH@NF catalyst in HMF-free

(black line) and HMF-containing (blue line) alkaline electrolyte (Figure 19 b). The experiments

demonstrate the suppression of oxygen production in presence of HMF (green potential region),

while OER was detectable starting +1.7 VRHE. From this result, we confirmed our preferred

potential range for selective HMFOR.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

We attribute a combination of strong adsorption of molecular 5-HMF onto the NiFe shared

catalytic surface sites25,26 as the origin for the suppression of the water oxidation.

Stability tests. To assess the time stability of the exceptional catalytic performance of the

NiFe(-Cl-)-LDH@NF catalyst for the HMFOR process, two distinct stability protocols were

conducted (Figure 19 c, Figure S 49). First, at 60 min intervals for a total of 5 hours, 5 mM

5-HMF aliquots were injected in the electrolysis reactor without changing the total electrolyte

volume. Figure S 49 shows a gradually decreasing current envelope after every injection due to

the decreasing reactant concentration in the cell concurrent to a rising product concentration.

This is why a separation of products at the cell outlet would be necessary to run such a protocol

efficiently. Second, at 2 h intervals over 10 hours, the electrolyte was replenished. After a series

of five cycles, 5-HMF conversion remained at favorable 99% and FDCA yields increased to

97% (Figure 19 c), which proved the time stability of the catalyst and electrodes using in the

flow cell electrolyzer.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

Figure 19: Electrolyzer Cell Performance and Stability: Evaluation of NiFe(-Cl-)-LDH@NF in the MEA cell setup and

DEMS. a) Variation of KOH electrolyte concentration with NiFe(-Cl-)-LDH@NF anodes, C supported Pt nanoparticles

(loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used as catholyte and 0.1 M KOH (dotted

line), 10 mM HMF in 0.1 M KOH (light blue) and 10 mM HMF in 1 M KOH (dark blue) used as anolyte. Catholyte and

anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials

1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1 b) Real-time mass spectrometric Linear Sweep Voltammograms (MSCVs) of the

NiFe(-Cl-)-LDH@NF electrocatalyst. The data was obtained in a custom-made DEMS flow cell setup in 0.1 M KOH and

10 mM HMF in 0.1 M KOH (flow rate: 10 µl s-1) at a scan rate 5 mV s-1 from 0- +2,0 VRDE. Plotted are signals of m/z= 32 of

16O 16O over the time resolved potential (Et). Corresponding Diagrams are given in the supporting information

(Figure S 47/Figure S 48). c) 10 h stability test with 5-HMF conversion and FDCA yield for NiFe(-Cl-)-LDH@NF in 0.1 M

KOH with 10 mM 5-HMF. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, at 1.56 Vcell.

Every 2 h HPLC samples were taken and the electrolyte was replaced.

Industrial anode FDCA Space Time Yields. To make the varying catalytic material systems

more comparable to each other in their performance, we calculated the resulting Space-Time-

Yields (STY) that describe the produced amount of FDCA per time and geometric surface area.

With NiFe(-Cl-)-LDH@NF we reached STYs of 0.0275 µmol s-1 cm-2 in 0.1 M KOH and

0.0733 µmol s-1 cm-2 in 1 M KOH respectively. A comparison with previous results

(Table S 11, Figure 20) impressively demonstrates the unprecedented performance of the new

NiFe(-Cl-)-LDH@NF catalysts even over existing NiFe-LDH systems.73

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

Our benchmarking considered the electrolyte concentration, as well, and confirmed the present

catalyst and electrode system as the most productive. We note that our electrolyzer is a paired

electrolyzer where molecular “green” hydrogen is stoichiometrically generated at the cathode.

We also note that, unlike the present catalyst system, selected previous reports of the HMFOR

reported strong competing OER oxygen evolution at cell potentials as low as +1.423 VRHE.76

Figure 20: Comparison between the FDCA product Space Time Yield achieved using the anion-tuned NiFe(-Cl-)-

LDH@NF catalyst deployed in paired H2-FDCA electrolyzer anodes (red data) and previous work. FDCA Space-Time-

Yield was plotted over the electrode potential in 0.1 M KOH (left, triangle) and 1 M KOH (right, diamonds).52,59-

61,63,64,70,71,73,76,79,81,84,167-170,178-181 Color code denotes a rising STY from blue to red. The references in the figure are kept

original.

To demonstrate the economic value and efficiency of the present paired electrolysis process, a

non-exhaustive technoeconomic estimate (see Supplementary Discussion 5) comparing the

present electrochemical process to an established high- temperature water process (HTW)

toward the commercially use analogous terephthalic acid revealed viable energy cost reductions

of a minimum of 2.5 billion US$ per year not factoring in CO2 emission pricing of additional

6-13 billion US$.

Efficient Paired Electrolysis of 5-Hydroxymethylfurfural (HMF) to the Biopolymer-

Precursor Furandicarboxylic acid (FDCA) at Industrial Current Densities

5.4 Conclusion

Concluding, we have demonstrated the superior catalytic reactivity, selectivity, and faradaic

efficiency of a novel family of anion-exchanged NiX(-CO32-)-LDH@NF catalysts, in particular

a set of NiFe(-Cl-)-LDH@NF catalysts, for the electrocatalytic paired conversion of the

biomass derived aqueous 5-HMF feeds to the biopolymer building block FDCA at the anode,

coupled to the evolution of green hydrogen at the cathode at industrially relevant current

densities. Focus of the present study was placed on the industrially relevant anodic process,

which replaced the conventional water oxidation reaction. Nickel-based NiX bimetallic layered

double hydroxides were grown directly on nickel nano foam substrates. Controlled anion-

exchange tuned the structural interlayer distance of the resting phase of the catalysts. Deployed

in flow electrolyzer cells, the NiFe(-CL-)-LDH@NF anodes achieved complete XHMF = 100%

ideal FEFDCA=100% and perfect FDCA selectivity of 100% over hours of operation at industrial

current densities.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

6 Hydrogenation versus Hydrogenolysis During

Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-

Bimetallics

Chapter 6 has been reproduced from Hauke, P., Merzdorf, T., Klingenhof, M. & Strasser, P.

Hydrogenation versus hydrogenolysis during alkaline electrochemical valorization of 5-

hydroxymethylfurfural over oxide-derived Cu-bimetallics. Nature Communications 14, 4708,

doi: https://doi.org/10.1038/s41467-023-40463-y (2023).43 (Accepted manuscript version)

The study was designed and executed by P.H.. All analysis tequnices (except SEM, TEM and

HR-STEM), electrochemical measurements and the synthesis of all catalysts was done by P.H.

The SEM images were done by T.M.. The TEM images were done by M.K.. The HR-STEM

analysis was done by Sören Selve.

6.1 Abstract

The electrochemical conversion of 5-Hydroxymethylfurfural, especially the reduction of

5-Hydroxymethylfurfural, is an attractive alternative production pathway for carbonaceous

green e-chemicals. We demonstrate the first successful reduction of 5-Hydroxymethylfurfural

to 5-Methylfurfurylalcohol under strongly alkaline reaction environments over oxide-derived

Cu bimetallic electrocatalysts. We investigate whether and how the surface catalysis of the MOx

phases tune the catalytic selectivity of oxide-derived Cu with respect to the

2-electron hydrogenation to 2.5-Bishydroxymethylfuran and the (2+2)-electron

hydrogenation/hydrogenolysis to 5-Methylfurfurylalcohol. We provide evidence for a kinetic

competition between the evolution of H2 and the 2-electron hydrogenolysis of

2.5-Bishydroxymethylfuran to 5-Methylfurfurylalcohol and discuss its mechanistic

implications. Finally, we demonstrate that the ability to conduct 5-Hydroxymethylfurfural

reduction to 5-Methylfurfurylalcohol in alkaline conditions over oxide-derived Cu/MOx Cu

foam electrodes enable an efficiently operating alkaline exchange membranes electrolyzer, in

which the cathodic 5-Hydroxymethylfurfural valorization is coupled to either alkaline oxygen

evolution anode or to oxidative 5-Hydroxymethylfurfural valorization.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

6.2 Introduction

The electrochemical conversion of biomass feedstock under near-ambient operation conditions

inside membrane electrolyzers using renewable energy is emerging as an attractive alternative

production pathway for carbonaceous green e-chemicals. In particular, the biomass compound

5-Hydroxymethylfurfural (HMF) - with its functional aldehyde and alcohol groups at positions

2 and 5 of the furan ring (Figure 21 a) - is a promising platform molecule for an electrochemical

oxidative and reductive valorization to bio-based polymer building blocks and fuel additives,

respectively. While the electrochemical oxidation of HMF is well documented,39,54,59,155 the

HMF reduction reaction (HMFRR) received far less attention over the past years. In fact, the

low kinetic barriers and low electrochemical overpotentials of the Hydrogen Evolution

Reaction (HER) make the competitive HMFRR process in an aqueous environment rather

unattractive. However, electrochemical reductive ring-opening, hydrogenation, and

hydrogenolysis of HMF represent attractive future electricity-based reaction process pathways

to agrochemicals, pharmaceuticals, bio-fuels, or polyesters.182 To date, the electrochemical

HMFRR has been largely reported over metallic Cu-based catalysts due to their unfavorable

competitive HER performance, coupled to their high chemical affinity to organic molecules. In

addition to the selective hydrogenation of HMF using noble metal catalysts, Roylance et al.

showed the reductive ring opening over a Zn catalyst, while Kloth et al. showed the

dimerization of HMF over carbon electrodes.96,98,104,105 Bimetallic HMF reduction

electrocatalysts, in particular bimetallic oxides in the form of their oxide-derived surface-

roughened catalyst analogs that evolve under reducing operando conditions, are essentially

unexplored. Prior work on the HMF reaction pathways seemingly established that the

electroreduction of HMF beyond the first 2e- hydrogenation product,

2,5-Bishydroxymethylfuran (BHMF), is not possible at high pH conditions. The rate of

subsequent hydrogenolysis of BHMF to 5-Methylfurfurylalcohol (MFA) or 2,5-Dimethylfuran

(DMF) sharply decreased with increasing pH.93,183 Only in strongly acidic electrolytes, work

by Nilges et al. and Zhang et al. yielded the formation of MFA and even DMF via 4 and 6e-

HMF hydrogenolysis.102,103

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

Figure 21: HMF reaction pathways and two-phase Cu based bimetallic oxide catalyst concept for enhance HMFRR. a)

Poly- and Dimerization (black): undesired side reaction of n HMF molecules and n Protons/electrons. Reductive Ring opening

(dark blue): opening of the furan ring with 6e- and 6H+. Hydrogenation (blue): conversion of HMF to BHMF with 2e- and 2H+.

Hydrogenolysis (blue): conversion of HMF to MF with 2e- and 2H+, conversion of BHMF or MF to MFA with 2e- and 2H+,

and conversion of MFA to DMF with 2e- and 2H+. Oxidation (yellow): conversion of HMF over HFCA or FDA to FFCA and

FDCA (6e- and 6OH-). Carbon atoms (black), oxygen atoms (red), and hydrogen atoms (white). a) Co-precipitated co-existing

Tenorite CuO/MOx (Hematite Fe2O3) nanoparticles at the nm-scale are hypothesized to offer enhanced catalytic HMFRR

reactivity after in-situ reduction to OD-Cu/MOx mixed phase catalysts. Adsorbed hydrogen atoms (Had) are given in white.

In this work, we show that, in contrast to the long-held view in literature, the valorization of

HMF to MFA at high pH via the 4-electron coupled hydrogenation/hydrogenolysis is actually

possible.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

We demonstrate that these conditions enable the use of CuO/MOx (M=first row transition

metals) bimetallic oxide electrocatalyst precursors, which – under the reductive reaction

conditions - transform into operating oxide-derived partially metallic (OD)-Cu/MOx catalysts.

In particular, the metal oxides NiO, Fe2O3, and Co3O4 were added to CuO to tune the resulting

HER activity of the two-phase system and by means of tuning the surface atomic Had coverage.

It is, to the best of our knowledge, the first report of electrocatalyst and electrochemical

interfaces that are able to catalyze the HMF reduction to MFA in alkaline conditions.

Mechanistically, we clarify whether and how the surface catalysis of the MOx phases tune the

catalytic selectivity of OD-Cu with respect to the 2-electron hydrogenation to BHMF and the

(2+2)-electron hydrogenation/hydrogenolysis to MFA. Finally, we design a foam-supported

CuO/MOx cathode and operate it inside an Alkaline Exchange Membranes (AEM) HMF

electrolyzer. Continuous reductive HMF valorization coupled with water oxidation and even

oxidative HMF valorization is demonstrated.

6.3 Results and Discussion

6.3.1 Catalyst Synthesis

Four crystalline mono-metallic powder oxides and three crystalline Cu-bimetallic oxide powder

electrocatalysts were prepared using a (co-)precipitation-calcination (air) protocol. The

calcined co-precipitated materials were deliberately designed as two-phase oxides, consisting

of the dominant Tenorite Copper(II) oxide phase mixed with the second MOx, M=Ni, Fe, Co,

oxide phase, henceforth referred to as CuO/MOx catalyst (see structural cartoon Figure 21 b).

The choice of a two-phase catalyst concept originated from the basic good HMFRR reactivity

of pure crystalline CuO that we intended to tune by the presence of a distinct second oxide

phase at nm scale proximity (rather than by forming a new mixed oxide phase) with varying

structure and chemisorption characteristics.184-187 Each two-phase catalyst synthesis was

individually developed and optimized such as to set the molar ratio of metal M to 10%, see

Table S 12. To achieve this, inductively coupled plasma optical emission spectroscopy

(ICP-OES) was used.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

The 10% molar ratio originated from a representative CuO/NiO structure-composition-

performance study (Figure S 50 and Supplementary Discussion 6) where the 10 mol% MOx

catalysts displayed the widest HMF-selective reduction activity window with a balanced HER

reactivity.

6.3.2 Characterizations

X-Ray Powder Diffraction (XRD) analysis of pure oxides revealed the characteristic Bragg

reflections of Tenorite CuO (C2/c; JCPDS: 01-089-2529, Figure 22 a) at 35.54°, 38.66°, 61.58°,

66.30° and 68.87°, Bunsenite NiO (Fm3

m; JCPDS: 00-002-1216, Figure 22 b) at 37.29°,

43.26°, 62.74°, 75.38°, and 79.86°, Hematite Fe2O3 (R3

c; JCPDS: 01-089-0599, Figure 22 c)

at 24.16°, 33.20°, 35.65°, 49.50°, 54.12°, 62.49° and 64.05, and for the Co3O4 Spinel (Fd3

m;

JCPDS: 00-043-1003, Figure 22 d) at 19.00°, 31.26°, 36.83°, 38.54°, 44.81°, 55.65°, 59.36°

and 65.22. Figure 22 e shows the gradual change in the XRD pattern of two-phase CuO/NiO

for increasing NiO content from 0 mol% (bottom) to 100 mol% NiO (top). Data indicate a

diffractive detection limit of NiO above 67 mol% that agrees with Figure 22 f, where exclusive

CuO Bragg reflections were detectable at the 10 mol% Ni level.

To further investigate the surface chemical composition and the valence states, X-Ray

Photoemission Spectroscopy (XPS) of pure CuO and the bimetallic metal oxide powders was

carried out. It provided further evidence of the presence of Cu(II)O, Ni(II)O, Fe(III)2O3, and

Co(II/III)3O4 (Figure 22 g-j and Figure S 51-Figure S 54). In more detail, Figure 22 g and

Figure S 51 reveal characteristic peaks for oxidic Cu at 953.8 eV (2p1/2) and 933.8 eV (2p3/2) as

well as strong Cu2+ satellites (sat.) peaks at 962.3 eV and 941.6 eV.188,189 In addition, the

characteristic peaks for NiO (872.6 eV and 855.3 eV), oxidic Fe (723.9 eV and 711.4 eV),

oxidic Co (795.2 eV and 779.7 eV) as well as Ni2+ (853.7 eV and 872.6 eV), Fe3+ (732.1 eV and

718.7 eV) and Co2+/3+ (795.6 eV, 780 eV, and 782.2 eV) are observed (Figure 22 h-j and

Figure S 52-Figure S 54).190-194 Besides, for the Ni 2p3/2 signal (855.3 eV), γ-NiOOH was

detected proportionally, which probably arose during the strongly alkaline synthesis and was

not further oxidized to NiO (Figure S 52 c).193,195,196

Brunauer-Emmett-Teller (BET) analysis was used to estimate the surface area of the mono- and

bimetallic oxide catalyst powders (Figure 22 k and Table S 13). CuO showed a relatively low

surface area yet twice the surface area of commercial Tenorite, CuO (Sigma-Aldrich).

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

The highest surface area was exposed by pure NiO followed by Co3O4 and Fe2O3. However,

this trend was not valid for the bimetallic powders, where CuO/NiO showed lower surface areas

than CuO/Fe2O3 and CuO/Co3O4.

6.3.3 Morphology and Composition Transformations Under Reaction

To inspect the CuO/MOx catalyst morphologies, in pure powder form Transmission Electron

Microscopy (TEM and HR-STEM) was used. Scanning Electron Microscopy-Energy

Dispersive X-Ray Analysis (SEM-EDX) was used to explore the morphology and composition

before and after electrolytic reaction (ae) of the corresponding catalyst thin films spray-coated

onto the CF. TEM and HR-STEM images did not reveal any distinct differences between the

catalyst morphologies or other visible differences between the various CuO/MOx catalysts

(Figure S 55 and Figure S 56). Individual oxide particle sizes ranged in the 5 to 100 nm range,

with particles forming linearly attached chains.

The SEM-EDX elemental analysis of the spray-coated CF-supported electrocatalyst films

showed the expected contributions of elements like O, Cu, Ni, Fe, and Co

(Figure S 57-Figure S 61). Contributions of C and F stemmed from the Nafion binder. Just like

the TEM analysis, the SEM images (Figure S 58-Figure S 61 a-d) confirmed the agglomeration

of induvial oval-shaped metal oxide nanoparticles. Although no significant morphological

changes were observed after electrolysis (Figure S 58-Figure S 61 a-d), a dramatic

compositional change was evident in all catalysts, more specifically a decreased molar O/Cu

ratio (Figure S 58-Figure S 61 g and h), indicating a partial reduction of the CuO catalyst phase

to metallic Cu. This justifies perceiving the CuO/MOx as two-phase precursor catalysts that

reductively transformed into a catalytically active oxide-derived OD-Cu/MOx catalysts. There

was no evidence that the MOx phases were reduced to a metallic state during catalytic reaction.

OD-Cu catalysts are well documented in the field of CO2 reduction.197,198 They offer a very

rough Cu surface characterized by many undercoordinated Cu adatoms which likely serve as

active sites for activation steps of reactant molecules.186,198,199

For the CuO/NiO/CF catalyst, needles were visible after electrolysis (Figure S 59 c and d).

KOH crystal formation was excluded by reference measurements in rinsed and ultrasonicated

electrodes. Similarly, judged by the SEM-EDX profiles, we did not see any Potassium salt

precipitation for any of the other catalysts (Figure S 59 h, Figure S 60 h, Figure S 61 h).

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

To convince ourselves of the formation of metallic OD-Cu phases under reactive electrode

potentials, thin film XRD studies were carried out on the as-prepared CuO precursor catalyst

(center and bottom Graph) and the resulting OD-Cu catalyst after catalytic HMFRR (top

profile) in Figure S 62 a and b. Evidently, the (-111), (111) and (200) reflections of Tenorite

CuO (orange) disappeared partially in favor of the emergence of (111) and (200) reflections of

metallic OD-Cu (red) confirming our SEM-EDX results.

Figure 22: XRD, XPS, and BET characterization of the CuO/MOx powder catalysts. a)-d) XRD patterns of pure powder

CuO (Tenorite), NiO (Bunsenite), Fe2O3 (Hematite), and Co3O4 (Spinel) XRD with references given in orange, red, purple,

blue, and inserted crystal structures. e) XRD patterns from pure powder CuO (black) over 10, 30, 67, and 80 mol% NiO to pure

NiO (red). f) XRD patterns of pure CuO (black), CuO/NiO (red), CuO/Fe2O3 (purple), and CuO/Co3O4 (blue). All CuO/MOx

are presented with 10mol% of the second metal oxide. CuO Tenorite reference is given in orange. g) Cu 2p XPS measurements

of pure CuO (orange). h) Ni 2p XPS measurements of CuO/NiO (red). i) Fe 2p XPS measurements of CuO/Fe2O3 (purple). j)

Co 2p XPS measurements of CuO/Co3O4 (blue). k) BET results of commercial CuO (orange), synthesized CuO (black), pure

NiO, Fe2O3, Co3O4 (light red, light purple, light blue), and mixed metal oxides CuO/NiO (red), CuO/Fe2O3 (purple),

CuO/Co3O4 (blue).

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

6.3.4 Catalytic Testing of Rotating Disk Electrodes

In order to investigate the catalytic HMF reduction reactivity (not HMFRR selectivity) of the

precursor CuO/MOx catalysts, we carried out electrochemical three-electrode RDE

measurements using both the pure individual metal oxide phases (Figure 23 a, b, d, e) and the

two-phase CuO/MOx catalysts (Figure 23 c, f) in presence (solid lines) and in absence (dashed

lines) of 10 mM HMF reactant. Figure 23 a-c report geometric area-corrected current densities,

whereas Figure 23 d-f show BET surface-corrected current densities, a popular intrinsic

measure of performance.

We first addressed the pure oxide catalysts, CuO, NiO, Fe2O3, and Co3O4 (Figure 23 a) to learn

more about their individual HMF reduction reactivity. Except for Fe2O3, the pure phase oxide

catalysts displayed significantly enhanced catalytic HMF reduction (HMFRR, solid) over the

hydrogen evolution reaction (HER, dashed). This was coupled to more anodic electrode

potentials at 10 mA cm-2 or likewise lower ηHMFRR than ηHER. The operating OD-Cu catalyst

(henceforth referred to by its precursor state CuO) by far outperformed other oxides

(Figure 23 a), which was even more obvious in the BET-normalized plots (Figure 23 d). Despite

evidence for the formation of OD-Cu and the existence of characteristic metal/metal oxides

couple of NiO, Fe2O3, and Co3O4 (Figure S 58-Figure S 62), no voltammetric redox waves were

obvious in base electrolyte without HMF.200-203 Notably, the present crystalline oxide precursor

catalysts showed substantially higher catalytic activities than previously reported metallic

catalysts,91,168 which we in part attribute to the higher surface roughness of the partially reduced

oxide catalysts.

We also compared our synthesized crystalline Tenorite CuO catalysts to a commercial CuO

material (Figure 23 b, e). The synthesized CuO catalyst showed higher geometric catalytic

HMFRR, but similar intrinsic HMFRR as the commercial CuO (Figure 23 b, e); its HER

performance was lower than the commercial one. This can be rationalized by distinct oxidic

surface states, as oxide surfaces generally display weaker H chemisorption and hence lower H

coverages. As the oxidic catalyst surface reduced to roughened metallic OD-Cu facets, the

rough undercoordinated Cu0 surface supported larger geometric HMF currents, yet the intrinsic

activity per site remained comparable to the reference.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

We now turn to the catalytic HMFRR RDE performance of the two-phase oxide catalysts. The

co-precipitated 10 mol% MOx phase was designed to enhance the HMFRR of the OD-Cu phase

by possible Hydrogen spillover. Figure 23 c indeed confirms that the secondary phase provided

the desired increase in HER activity (dashed lines) in the order CuO < CuO/Co3O4 < CuO/Fe2O3

< CuO/NiO. More importantly, the HMFRR reactivity of all CuO/MOx catalysts substantially

increased. The cathodic potential shifts at 10 mA cm-2 between the HER and the HMFRR

ranged at 200 mV, 270 mV, 37 mV, and 150 mV for CuO/Fe2O3, CuO, CuO/NiO, and

CuO/Co3O4, respectively. The absolute HMFRR performance at 10 mA cm-2 and at -0.3 VRHE

dropped in that same order. The CuO/Fe2O3 catalysts outperformed the reference CuO catalysts

and became one new catalyst of further interest. At increasing current densities and more

negative electrode potentials, CuO/NiO, and CuO/Co3O4 displayed cross-over points where the

current density in presence of HMF equalled that in absence (pure HER). Prior to the cross over

point, the catalysts showed electrode potential ranges of primary interest for device operation

with substantial HMFRR reactivity. Clearly, based on the vastly different BET surface areas

between CuO (22 m2/g) and the CuO/MOx (69 to 88 m2/g) the corrected reactivity trends were

CuO > CuO/Fe2O3 > CuO/NiO ~ CuO/Co3O4. We confirmed our hypothesis that a modulation

of the reactivity of CuO is possible by the presence of the second crystalline metal oxide phase

(Figure 22 f-j and Table S 12). We can also add that our (co-)precipitation-calcination (air)

synthesis protocol has slight catalytic advantages over the physical mixing of the individual

precipitated metal oxides (Figure S 63).

6.3.5 Catalytic Testing of Stationary Foam Electrodes

We carried out a stepwise scale-up from the RDE powder thin film level to the 5 cm2 spray-

coated CF-supported cathodes level, which were to be deployed in Alkaline exchange

membrane (AEM) flow electrolyzer cells. To achieve this, we prepared a 1 cm2 electrode by

spray-coating CuO/MOx powders onto a metallic Cu foam (CF) support (Figure S 62 b),

denoted CuO/MOx/CF. We tested the spray-coated film electrodes in an undivided three-

electrode cell configuration (UTEC) without rotation of the working electrode (Figure S 64).

The reactivity of the HMFRR rather than its chemical selectivity was of focus here.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

After a loading study (Figure S 65) to select suitable catalyst loadings, a catalytic HMFRR

activity screening with and without HMF (Figure 23 g and h) established the most favorable

current-potential operating regions for the design of an efficient AEM HMF electrolyzer.

HMFRR as well as baseline HER currents of the spray-coated electrodes increased substantially

compared to the powder thin film RDE study. The HMFRR activity increased in the order

CF < CuO/CF < CuO/Co3O4/CF ~ CuO/Fe2O3/CF < CuO/NiO/CF over the entire current range.

The current density differences between baseline HER and HMFRR, however, dropped

significantly to a value between 22 mV to 117 mV. Clearly, the background HER reactivity

sharply increased in the sprayed thin film format. This is due to the larger surface area of the

foam support and the thicker catalyst films. At the same time, however, Figure S 66 shows that

CF alone does not necessarily lead to a generally increased performance. Here it becomes clear

once again that the combined CuO/MOx metal oxides with or without CF support bring an

increase in activity.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

Figure 23: RDE and undivided three-electrode cell measurements of CuO/MOx and CuO/MOx/CF. a) RDE

measurements of pure CuO (black), NiO (red), Fe2O3 (purple), and Co3O4 (blue). b) RDE measurements of commercial CuO

(orange) and pure CuO (black). c) RDE measurements of mixed metal oxides CuO/NiO (red), CuO/Fe2O3 (purple), and

CuO/Co3O4 (blue). d)-f) BET corrected current densities of the corresponding plots from a)-c). All RDE LSV measurements

were taken between 0 VRHE to -0,6 VRHE at a scan rate of 10 mV s-1 in 0.1 M KOH with (solid line) and without (dashed line)

10 mM HMF at 2500 rpm with an electrode surface area of 0.19 cm2 and a catalyst loading of 0.04 mg. All measurements are

100% manual internal resistance (IR) corrected. g) undivided three-electrode cell (UTEC) measurements of CuO and CuO/MOx

on CF with CF (orange), CuO/CF (black), CuO/NiO/CF (red), CuO/Fe2O3/CF (purple), and CuO/Co3O4/CF (blue). h) blow up

of the UTEC measurements in g) with colored HMFRR selectivities areas of CF (orange), CuO/CF (black), CuO/NiO/CF (red),

CuO/Fe2O3/CF (purple), and CuO/Co3O4/CF (blue). All UTEC LSV measurements were taken between 0 VRHE to -0,8 VRHE at

a scan rate of 10mV s-1 in 0.1 M KOH with (solid line) and without (dashed line) 10 mM HMF without rotation and an electrode

area of 1 cm2 and a catalyst loading of 1 mg cm-2. All measurements are 100% manual internal resistance (IR) corrected.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

6.3.6 Alkaline Exchange Membrane HMF Electrolyzer Cell Tests

Based on our CuO/MOx/CF catalyst discovery and characterization studies above, we moved

to build the first 5 cm2 active area Alkaline Exchange Membrane electrolyzer for the

valorization of HMF in strongly alkaline pH on the cathode, coupled to the alkaline oxygen

evolution reaction on the anode using all PGM free catalysts (Figure S 67). As the cell operated,

we monitored the conversion of HMF (XHMF), the selectivity of products such as BHMF and

MFA (SBHMF and SMFA) as well as the Faradaic efficiencies (FE) of H2, BHMF, and MFA

(Figure 24 and Figure S 68). Other reaction products, such as Methylfurfural (MF) or

di/polymerized HMF are referred to as others.

Figure 24 a-e compare the faradaic product efficiencies of the uncoated CF support reference

compared to the CuO/MOx/CF electrodes at varying applied currents. Figure S 69 and

Table S 14 report the polarization behavior, the detailed time-resolved cell potentials during the

galvanostatic step protocol, as well as the corresponding H2 rates and performance parameters

for all catalysts.

For the pure metallic Cu foam (CF) support, high FEBHMF and FEothers were evident

(Figure 24 a). FEH2 and the absolute hydrogen production rate increase monotonically with

current density and cell potential (Figure 24 a and Figure S 69 a). The 2e- reduction of the

-COH aldehyde group appears to be fast and fairly selective on metallic Cu surfaces.

Note that no MFA was produced in Figure 24 a suggesting that the (2+2)e- sequential

hydrogenation/hydrogenolysis of HMF to MFA in high alkaline environment is unfavorable

(Figure 24 a and Figure S 62 b).

The CuO/CF reference electrode catalyzed both 2e- hydrogenation and some subsequent 2e-

hydrogenolysis to MFA in alkaline conditions (Figure 24 b, Figure S 69 b). The FEMFA values

exceeded FEBHMF at all current densities and peaked at 20 mA cm-2. Again, HER increases

monotonically and appears to compete with MFA production.

The CuO/NiO/CF cathode displayed somewhat lower FEMFA values in favor of FEHER, in

agreement with the well-documented HER reactivity of Ni oxides in alkaline conditions

(Figure 24 c, Figure S 69 c).19 The 20 mA cm-2 appears as narrow MFA selectivity sweet spot,

in line with the narrow selective operating regime of CuO/NiO/CF in Figure 23 h.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

We conclude that a high catalytic HER reactivity of the secondary MOx oxide is in direct

mechanistic competition to MFA production. We rationalize this as a competition for adsorbed

H atoms, Had needed in Langmuir Hinshelwood-type reaction pathways to H2 and MFA. We

further hypothesize that BHMF, on the other hand, is largely formed in direct proton reduction

according to an Eley-Rideal-type process. This is why BHMF production appears unaffected

by varying HER production. Interestingly, low FEothers values suggested an additional

proton/Had competition with di/polymeric HMF products with high proton demand.

The CuO/Fe2O3/CF cathode (Figure 24 d, Figure S 69 d) is the most interesting one. It displayed

not only the lowest cell potentials (highest energy efficiencies) (Figure S 69 d) but most

importantly, offered the most favorable FEMFA values at very low FEothers, FEH2, and FEBHMF,

particularly at 10 mA cm-2. CuO/Fe2O3/CF is an excellent MFA-producing

hydrogenation/hydrogenolysis catalyst over larger current ranges, as anticipated in Figure 23 h.

Again, FE values confirm the hydrogen competition between HER and MFA production, while

BHMF formation is unaffected. CuO/Fe2O3/CF proves that with rising HER, the

hydrogenation-hydrogenolysis-selectivity is developing in favor of hydrogenation.

The CuO/Co3O4/CF cathode (Figure 24 e, Figure S 69 e) displayed favorable FEMFA values up

to 20 mA cm-2, after which they dropped quickly as the FEH2 increased. This is in excellent

agreement with results from Figure 23 h, where HMFRR and HER polarizations cross around

at current. The secondary Co3O4 catalyst leads to high FEothers (Figure 24 e).

AEM HMF electrolyzer tests with a CuO/NiO/CF cathode were extended until complete HMF

conversion, while the time-resolve evolution of FEproducts and XHMF (Figure 24 f) was tracked in

the electrolyzer exit feed. Interestingly, MFA was selectively produced from HMF at a nearly

60% conversion during the first 15 minutes. At that point in time, MFA production through

hydrogenation/hydrogenolysis suddenly ceased, while the hydrogen evolution dominated the

electrolysis products at moderate BHMF production. In part, we attribute the sharply lower

HMF to MFA conversion to depleted local HMF concentration at the surface associated with

lower coverages of reactive HMF intermediates, such as BHMF. Sustained proton surface

reduction and high Had coverage render Volmer-Tafel HER the dominant process.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

We note that a compositional or structural change of the catalyst surface cannot be fully

excluded as an origin for the FE variations, yet the available catalyst stability data

(Figure S 57-Figure S 62) make this hypothesis unlikely. We also note that the HMFRR/HER

competition at the surface is in congruence to earlier own work on the competition of oxidative

HMF conversion and the oxidation oxygen evolution reaction.155 We further conclude from the

fact that no MFA was converted to 2,5-Dimethylfuran (DMF) nor BHMF was converted to

MFA after 15 minutes, that the electrocatalytic reduction of -CH2OH alcohol group is extremely

unfavorable in alkaline conditions.93,183 A partial deprotonation of -CH2OH at pH 13 to less

reactive alcoholate groups, -CH2O- (pKa ~ 12)204,205 is compounding the kinetic hindrance.

To display the impact of the secondary oxide phase on the competition between hydrogenolysis

and hydrogenation, in other words on the MFA/BHMF selectivity preference, we plotted the

HMF conversion vs the absolute SMFA/SBHMF ratio in Figure 24 g-i. As HMF conversions rise

to completion (XHMF =1 for CuO/Fe2O3/CF at 20 and 30 mA cm-2), the SMFA/SBHMF ratio

follows the FE trend. Hydrogenation prevails over hydrogenolysis at higher currents and

conversions. Only at 10 mA cm-2 for CuO/Fe2O3/CF and CuO/Co3O4/CF, where no or hardly

any hydrogen is formed, hydrogenolysis is preferred over hydrogenation.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

Figure 24: MEA-Flow-Cell performance measurements of CuO/MOx/CF electrodes. a)-e) Faradaic efficiencies for H2

(strong blue), BHMF (light blue), MFA (green), and other products (grey) of the different spray-coated CuO/MOx/CF catalysts.

f) Faradaic efficiencies and HMF conversion (yellow) are calculated for every 15 min time interval over 60 min using

CuO/NiO/CF as a catalyst. Product color code stays as in a)-e). g)-i) Scatter plot of the product selectivity preference. HMF

conversion over MFA/BHMF selectivity ratio, calculated by SMFA/SBHMF for all CuO/MOx catalysts on CF at different current

densities for 30 min. The color code stays the same as before. Cell reaction conditions: 0.1 M KOH with 10 mM HMF as

catholyte (100 ml, recycled), 0.1 M KOH as anolyte (100 ml, recycled), 5 cm2 electrode area, nickel foam (NF) as anode and a

flow rate of 25 ml min-1, at 10-30 mA cm-2 for 30 min. High frequency resistance results are between 0.7- 1.05 Ω (Figure S

67 b). Relative errors of 2-4% for FEproducts, 3-5% for FEH2, and 1-3% for XHMF resulted (Figure S 68).

6.3.7 AEM HMF Electrolyzer Stability Tests

We checked the galvanostatic stability of our AEM electrolyzers, for which we used the

favorable CuO/Fe2O3/CF electrode at 20 mA cm-2 (Figure 25 a) due to its favorable cell

parameters, such as overpotential, HMF conversion, Faradaic efficiencies and selectivities

(Figure 24 and Table S 14).

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

To validate CuO/Fe2O3/CF as a stable electrode, we tracked the conversion of HMF and the

selectivities of BHMF and MFA over 5 test protocol cycles of 2.5 h. Figure 25 a reveals 100%

HMF conversion at constant high selectivities.

In the last electrolyzer design step, we replace our standard Nickel foam (NF) anode with a

previously reported highly active NiFe(-Cl-)-LDH@NF material, an excellent OER catalyst and

HMF oxidation catalyst (Figure 25 b, HMF//KOH).127,155 We used the

CuO/Fe2O3/CF//NiFe(-Cl-)-LDH@NF cathode//anode AEM water electrolyzer (no HMF feed)

as a reference (KOH//KOH in Figure 25 b). Compared to the AEM water electrolyzer cell, the

CuO/Fe2O3/CF//(NiFe(-Cl-)-LDH@NF) cathode//anode design with HMF//HMF demonstrated

a 20% drop in cell input voltage at stable SBHMF, SMFA and stable SFDCA, the HMF oxidation

product 2,5-Furandicarboxylic acid (FDCA). A closed 100% HMF reduction product balance

remained elusive due to HMF polymerization (humin formation).

Figure 25: MEA-Flow-Cell stability measurements of CuO/Fe2O3/CF and combination of HMF reduction and

oxidation. a) Stability measurements of CuO/Fe2O3/CF over 5 cycles (2,5 h) at 20 mA cm-2. HMF conversion in yellow, BHMF

selectivity in turquoise, and MFA selectivity in green. b) cell potential and performance comparison between non-HMF

containing electrolytes on both sides, HMF at the cathode and HMF on anode and cathode, using CuO/Fe2O3/CF as the cathode

and NiFe(-Cl-)-LDH@NF as the anode. The cell potential is given in pink triangles, BHMF selectivity in turquoise, MFA

selectivity in green, and FDCA selectivity in light green. Cell reaction conditions: 5 cm2 electrode area, flow rate of 25 ml min-

1, at 20 mA cm-2 for 30 min.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

6.3.8 Mechanistic Discussion

Figure 1 a displays the rich electrochemical reactivity of the HMF molecule. The 2e-/2H+

reduction of a -CH2OH group (of HMF or BHMF) into a -CH3 group is associated with a C-O

bond breaking and, as such, is likely to require a surface-adsorbed reactive state of the HMF

intermediate and adsorbed Had in atomic proximity, following a Langmuir-Hinshelwood-type

reaction. By contrast, it appears feasible that the simpler 2e-/2H+ reduction of the aldehyde

group -COH to the -CH2OH group (no bond breaking, mere bond order reduction) is catalyzed

using protons from the double layer rather than Had, following the Eley-Rideal pathway. Indeed,

on CuO catalysts with their rough metallic OD-Cu surface state, moderate MFA yields were

detected. Undercoordinated Cu adatoms on the roughened Cu surface facets activate -COH

groups of (B)HMF molecules. The secondary MOx phase, remaining an oxidic surface state,

offers a balanced hydrogen chemisorption to Had, and, as such, may act as a source or pool of

reductive Had equivalents that diffusively spill over and aid in the stepwise reduction of

adsorbed HMF molecules and its derivatives (Figure 21 b). HER and MFA production compete

for Had.

Acidic electrolyte conditions suppress undesired electroless di- and polymerization of HMF

molecules, yet require noble and corrosion stable catalysts. Alkaline conditions favor humin

formation, however enable cost effective catalysts. Under the strongly alkaline conditions of

the present study (pH 13) some of the alcoholic -CH2OH protons (pKa ~ 12) are deprotonated

forming negatively charged alcoholate groups, -CH2O-, which are no longer available for

hydrogenolytic reduction to methyl groups of MFA. Hence, MFA formation becomes more

difficult under alkaline conditions.183 We suppose, if reactive BHMF intermediates, however,

remain adsorbed on the surface after the 2e-/2H+ aldehyde reduction, and are supplied with

additional reducing Had equivalents on the surface, this scenario favors the subsequent 2e-/2H+

hydrogenolysis to MFA.

Hydrogenation versus Hydrogenolysis During Electrochemical Valorization of

5-Hydroxymethylfurfural Over Oxide-Derived Cu-Bimetallics

6.3.9 Conclusion

We have explored the electrocatalytic reduction and valorization of HMF on novel noble metal-

free Cu-based two-phase oxide precursor catalysts. We have reported a new noble metal-free

oxide catalyst and the first electrochemical interface inside an HMF AEMWE that catalyzes the

electrochemical HMF reduction to MFA in strongly alkaline conditions. Of particular focus

was the competition between the 2e-/2H+ -COH hydrogenation to -CH2OH and its subsequent

2e-/2H+ hydrogenolysis to -CH3. We hypothesized that the nature of the secondary metal oxide

and it’s HER activity will tune the HMF reduction selectivity of CuO via its own HER

reactivity. That hypothesis that we validated. The use of alkaline electrolyzer conditions has

practical implications, as the use of cost-effective catalysts and the coupling of the HMF

reduction process to alternative anode reactions become possible. This was demonstrated by

building a paired HMF-HMF electrolyzer.

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

7 Paired Electrocatalytic Valorization of CO2 and

Hydroxymethylfurfural in a Noble Metal-Free

Bipolar Membrane Electrolyzer

Chapter 7 has been reproduced (adapted) with permission from Hauke, P., Brückner, S. &

Strasser, P. Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-free Bipolar Membrane Electrolyzer. ACS Sustainable Chemistry & Engineering, doi:

Society.206 (Accepted manuscript version)

The study was designed and executed by P.H. and S.B.. Measurements concerning HMF

oxidation were performed by P.H.. This includes the complete evolution of the anodic reaction

from the seprate up to the combined cell consideration. Measurements concerning CO2

reduction were performed by S.B..

7.1 Abstract

Direct electrocatalytic valorization of CO2 in low-temperature electrolyzers is emerging as a

new non-fossil, one-step process toward e-fuels and e-chemical, such as CO. However, faradic

and energy efficiencies have remained low due to the sluggish 4-electron oxidation of water

(OER) at the anode. Replacement of the OER with a thermodynamically and kinetically less

demanding reaction would raise efficiency and overall valorization. This manuscript

demonstrates the first full paired implementation of a noble metal free CO2 and HMF

valorization in a single cell at industrially relevant current densities. We stepwise design,

assemble, test and analyze the first complete paired low-temperature bipolar membrane (BPM)-

based Hydroxymethylfurfural oxidation (HMFOR) and CO2 electroreduction electrolyzer cell.

The electrolyzer couples a CO2-to-CO electrolyzer half-cell to an aqueous

HMF-to-2.5-Furandicarboxylic acid (FDCA) half-cell via a water dissociation membrane

operating in reverse bias. We investigate and compare the bipolar membrane voltage penalties

with the single pass reactant conversion advantages and estimate cell performance benefits due

to the more favorable thermodynamic and kinetic processes at the anode. We report successfully

suppressing undesired CO2 loss due to acid-base neutralization with generated alkalinity.

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

7.2 Introduction

Direct electrocatalytic CO2 reduction is an emerging "power-to-carbon chemicals" process

bridging the historically unconnected industrial sectors of power (free electrons) and chemicals

(molecular bonds). This process uses renewable power to valorize CO2 into various potentially

carbon-neutral e-chemicals and e-fuels, such as CO, ethylene, or alcohols, thereby closing

anthropogenic carbon and hydrogen cycles. However, CO2 electrolysis and CO2 electrolyzer

devices continue to be plagued by serious design challenges related to unfavorable

thermodynamics, sluggish chemical reaction kinetics, reactant depletion, and transport

limitations.207

State-of-the-art CO2 to CO electrolyzers operate at neutral-alkaline conditions deploying anion

exchange membranes (AEM) to coupled cathode and anode processes. The default anode

process is the 4-electron oxygen evolution reaction (OER),208-210 significantly raising the

required thermodynamic cell voltage. In addition, the OER in near-neutral pH conditions is

kinetically very sluggish. Platinum group metal catalysts such as Iridium are needed to achieve

industrial current densities at acceptable cell voltages.211 What is more, local alkalinity,

generated catalytically at the cathode, stoichiometrically consumes CO2 in a non-catalytic acid-

based reaction, which then crosses the AEM as carbonate, thereby lowering the CO2 single pass

conversion significantly below 100%. Other zero-gap AEM CO2 electrolyzer failure modes

include cathodic salt precipitation due to potassium crossover as well as cathode flooding.212-

214 Cation exchange membranes are designed to be conductive for protons. Still, they are also

conductive for other cations, which would increase the salt precipitation at the cathode with

high alkali anolytes, as alkali metal cations are responsible for the salt precipitation.

The use of bipolar membranes (BPM) operated in a reverse bias protolytic mode according to

H2O → H+ + OH-

can address some of the challenges.

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

BPMs prevent cathodic carbonate formation as they dissociate water, generate and transport

protons to the cathode and hydroxide ions to the anode. Protons at the cathode decrease the pH

and regenerate CO2 from carbonate, enabling a higher single pass and lowering the necessary

CO2 feed. Moreover, by avoiding acidification at the anode, the use of platinum group metal

(PGM)-free catalysts becomes possible.215-223

The use of alternative anode reactions generating useful e-chemicals in an alkaline environment

characterized by a more negative standard potential than water/oxygen can address some

additional CO2 electrolyzer challenges. The electrochemical oxidation of

5-Hydroxymethylfurfural (HMF) to the bio-polymer precursor 2.5-Furandicarboxylic acid

(FDCA) is an appropriate alternative to the OER.224 Earlier model studies demonstrated lower

cell potentials of the HMF oxidation reaction (HMFOR) compared to OER and high selectivity

of the value-added product FDCA in highly alkaline electrolytes.52,60,63,71,73,81 More recently,

we have demonstrated the successful integration of the HMFOR into a paired zero-gap

hydrogen generating electrolyzer.155 The logical next step is the direct coupling of the catalytic

CO2 reduction reaction (CO2RR) and the HMF oxidation reaction (HMFOR) valorization

processes. This work addressed this.

The present contribution demonstrates the first design, operation, and kinetic analysis of a

paired CO2 /HMF electrolyzer cell catalyzed by all-noble-metal-free catalysts and ionically

coupled by a bipolar membrane, operating in purely aqueous environments at high production

rates. We chose to integrate an alkaline PGM-free HMF oxidation electrode (see Figure 26 a)155

via a reverse BPM with an AEM CO2-to-CO electrolyzer single Ni metal atom catalysts, NiNC,

(Figure 26 b)225 in an optimized catalyst layer to create a novel paired electrolysis cell setup

shown in Figure 26 c.

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

Figure 26: Scheme of the paired electrochemical conversion of CO2 and HMF. Integration of the state-of-the-art AEM

HMFOR into a BPM CO2RR electrolyzer with two valued products. For the AEM setups standard potentials are given in an

alkaline environment (pH=14). Separated considered a) semi-batch HMF oxidation and b) continuous CO2 reduction

electrolyzers, c) combined to a continuous CO2/HMF electrolyzer.

More specifically, we demonstrate the step-by-step design of the overall CO2RR / HMFOR

electrolyzer, starting with improving each half-cell reaction and half-cell electrode separately.

To achieve this, we first substituted the state-of-art anion exchange membrane (AEM) with a

bipolar membrane (BPM), maintaining an alkaline environment at the anode. We also moved

from a semi-batch HMFOR setup (Figure S 71 a) to a continuous HMFOR (Figure S 71 b) setup

to optimize the anode and bridge the gap to industrial applications with industrially relevant

current densities and continuous operation modes. We then demonstrate the full benefit of a

BPM for the catalytic CO2 to CO reduction reaction thanks to the proton transport to the

cathode. In order to remain high selectivity towards CO, we optimized the catalyst layer in the

CO2RR cathode (Figure S 71 c) although we change the local environment at the interface

between the membrane and catalyst layer due to the proton transport to the cathode.

In the final step, we combine both optimized half-cell reaction systems into a high-performance

electrolyzer system operating at current densities of 200 mA cm-2 (Figure S 71 d).

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

We close by discussing the design advantages of the CO2RR / HMFOR electrolyzer and the

remaining challenges along with their solutions.

7.3 Results and Discussion

7.3.1 A Continuous Flow HMFOR Electrolyzer Cell

Prompted by the semi-batch results (Supplementary Discussion 7 and Figure S 72-Figure S 74),

we moved to a continuous HMFOR flow setup (Figure S 71 b and Figure 27 a), fed by

pre-mixed streams of HMF and supporting aqueous electrolyte to suppress the formation of

humins, even at high HMF concentrations in strongly alkaline environment. We reduced the

feed to 5 ml min-1 and tracked the single pass conversion (Equation 56) of HMF to FDCA

(SPFDCA, red). Initial electrolysis tests used 10 mM HMF / 1 M KOH and set current densities

at 50, 100 and 200 mA cm-2 (Figure 27 b, Table S 16): The experimental STY (Space-Time-

Yield, Equation 41) ranged between 7.4-7.9 µmol min-1 cm-2 with stable SFDCA (chemical

selectivity) of 80% with increasing SPFDCA (40 to 55%) and decreasing FEFDCA (Faradaic

efficiency towards FDCA, Equation 40) (74% to 20%). Figure 27 b evidences that a current

density of 200 mA cm-2, despite lower FEFDCA due to competing oxygen evolution (OER), did

not affect the SFDCA and STY and had a favorable impact on the HMF single pass conversion.

In order to suppress OER, thereby increasing the FEFDCA, we varied the HMF concentration up

to 20 mM at a constant current density of 200 mA cm-2 (Figure 27 c). This strongly increased

STY (by 32%) and FEFDCA (by 8%) yet lowered SPFDCA (-11%) and SFDCA (-26%) to

unacceptable levels. This is why we chose 10 mM HMF in subsequent experiments. We also

reduced the flow rate from 5 ml min-1 to 2 ml min-1 to increase the retention time of precursor

in the cell, which led to increased XHMF (conversion, Equation 38) and SPFDCA of 19% and 22%,

respectively (Figure 27 d). The stable selectivity of FDCA with rising conversion prompted us

to maintain 2 ml min-1 for subsequent experiments. In summary, we succeeded in designing a

selective and efficient HMFOR anode for coupling to a CO2 electroreduction electrode. The

HMFOR electrode was characterized by operating current densities of 200 mA cm-2 at a

2 ml min-1 feed of 10 mM HMF yielding unparalleled performance indicators of 90% XHMF,

75% SFDCA, 12% FEFDCA, 69% SPFDCA.

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

Figure 27: Continuous HMF oxidation operation condition optimization. a) MEA constellation and electrolyzer operation-

mode (II Continuous) overview. b) and c) Current density and concentration dependency of STYFDCA (grey), SFDCA (green),

FEFDCA (blue), and SPFDCA (red) over 1h electrolysis with a flow rate of 5 ml min-1 and current densities of 50-200 mA cm-2. d)

Bar plot comparing XHMF (yellow), SFDCA (green), FEFDCA (blue), and SPFDCA (red) at 5 ml min-1 and 2 ml min-1 flow rate over

1 h electrolysis at 200 mA cm-2. Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt

nanoparticles (loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 5-20 mM HMF in 1 M KOH used

as anolyte. Catholyte and anolyte volume of 50 ml. Standard deviation between 2-3%.

7.3.2 Catalyst Layer Optimization for CO2 Reduction in a BPM Configuration

State-of-the-art lab scale CO2-to-CO electrolyzer systems comprise a CO2 cathode paired with

the oxygen evolution reaction (OER) fed by a cycled anolyte (Figure S 75). The half-cell

reactions read

Cathode: CO2 + H2O + 2e- → CO + 2OH- E°= -0.94 VSHE (pH=14)

E°= -0.53 VSHE (pH=7)

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

Anode: 4OH- → O2 + 4e- + 2H2O E°= 0.40 VSHE (pH=14)

E°= 0.82 VSHE (pH=7)

As our next step toward the design of a full CO2 / HMF electrolyzer, we designed a new catalytic

CO2 reduction gas diffusion electrode (GDE) for use on the acidic side of a BPM (Figure 28 a)

and combined this with our PGM-free NiFe(-Cl-)-LDH@NF-based OER electrocatalyst on the

anode.226 Normally use of BPM requires the design of all acid-stable components on the

cathode, but we assume that we still create alkalinity at the cathode, which neutralizes the

protons. The salt precipitation can strengthen this assumption and indicate that the potassium

crossover participates in the electroneutrality of the cell. To this end, we synthesized and

utilized a selective Ni-imidazolium-derived Ni-NC single Ni metal atom catalyst. The catalyst

synthesis entails the leaching of excess Ni metal in concentrated mineral acids while the active

nitrogen-coordinated Ni single metal atom sites remain. This is why these catalysts and their

active sites can be operated in acidic conditions. Moreover, Ni-N4 single metal sites exhibit low

relative chemisorption energy compared to CO2, making Ni-N4 selective for CO2 reduction in

acidic conditions.225,227-233 A key challenge was the suppression of catalyst layer flooding,

which causes sharp declines in CO2RR selectivity. We applied various PTFE

(polytetrafluoroethylene) loadings to the CO2 gas diffusion cathode layer (GDL) to suppress

flooding and screened their performance over 2.5 h at 200 mA cm-2 (Figure 28 b). Increasing

PTFE loadings affected the cell performance due to rising layer hydrophobicity.

Hydrophobicity decreased the liquid layer thickness and improved CO2 permeation and, thus,

access to catalytic active centers. The 60wt% PTFE GDE (gas diffusion electrode) showed

promising activity, so we also screened a wider current density window to evaluate the

selectivity and load flexibility of the GDE at lower current densities (Figure 28 c). As the

selectivity remained sustainably above 90% FECO (Equation 60), we conclude good load

flexibility at a slightly larger cell potential compared to the reference AEM CO2 GDE setup.

Importantly, while the BPM voltage penalty ranged around 800 mV for pure water dissociation,

the penalty turned out less in our current GDE setup thanks to the pH shift at the OER anode

caused by the 1M KOH. We shift the pH from around 7.3 to 14, which reduced the effective

penalty to only 500 mV. Under our BPM cell conditions, we succeeded in reaching close to

100% FECO and > 30% SP (Equation 62) at 200 mA cm-2, which surpassed our own previously

reported benchmark performance of an AEM setup (Figure 28 d).

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

We highlight that this was achieved using a PGM-free Ni-based anode compared to IrO2

catalysts in the AEM setup. The much higher lambda value, the ratio of CO2 feed and total

consumed CO2 (Equation 63) at a lower CO2 inlet feed, and identical applied current density

proves that the proton channeling to the cathode is working well and that we are regenerating

CO2 continuously due to the neutralization. This raises the molar CO2 ratio and reduces the

molar CO ratio in the outlet stream. This regime makes it possible to reach higher currents,

higher single pass conversions, and higher atom efficiencies at identical CO2 inlet feed

compared to AEM configurations. To increase the accuracy of our product quantification, we

use a nitrogen bleed. As we connect the N2 after the cell, the N2 flow can not be influenced by

the cell (salt blockage membrane cross over, etc.). Hence the N2 can be used as flow calibration

to calculate the actual flow in the GC, which increase the accuracy of our results.

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

100

Figure 28: Performance of CO2 reduction GDE in a BPM configuration. a) Schematic of a BPM CO2 reduction

configuration. b) Key performance parameter as energy efficiency (purple), faradaic efficiency (blue), and single pass

conversion (red) of the different PTFE loadings. c) CO2 reduction performance in terms of cell potential and faradaic efficiency

at different current densities. 20 ml min-1 humidified CO2, 1M KOH with 2 ml min-1 flow rate and NiFe(-Cl-)-LDH@NF

anodes. d) Comparison to previously reported AEM configuration at a total current density of 200 mA cm-2 and 20 ml min-1

CO2 in the BPM configuration or 25 ml min-1 in the AEM configuration, respectively. Standard deviation between 2-5%.

7.3.3 Paired CO2RR / HMFOR Electrocatalysis Using a Single BPM Electrolyzer Cell

Finally, we integrated the new BPM design of the HMF oxidation anode (Figure 27 a) with the

BPM design of the CO2-to-CO reduction zero-gap GDE cathode (Figure 28 a) into a paired

electrolyzer device for stoichiometric concomitant HMFOR and CO2RR (Figure 29 a, setup in

supporting information Figure S 76). We analyzed the paired cell performance from 50 to

200 mA cm-2. With larger current densities, the CO selectivity remained stable at a favorable

level. At the same time, the HMF conversion and FEFDCA increased and decreased, respectively

(Figure 29 b, comparison between paired and OER electrolysis in Figure S 77).

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

101

SFDCA displayed a maximum of around 100 mA cm-2 before it dropped due to a noticeable

impact of the competing oxygen evolution reaction (gas bubbles) at 200 mA cm-2. As a

consequence of the dominant OER at high current densities, no significate changes in the cell

potential compared to the reaction without HMF (Figure 28 c) were visible. Then, we conducted

an electrolyzer stability test and followed two different protocols for the anode to address

possible parasitic HMF side reactions. In one stability test, we renewed 120 ml anolyte every

hour, and in another, we cooled the anolyte (600 ml) to 7 °C to slow down humin formation

(Figure 29 c). In both approaches, the anodic time-resolved performance descriptors (SFDCA,

FEFDCA, and SPFDCA) evolved favorably and almost identically, suggesting the suitability of

both approaches to minimize undesired side reactions (Figure 29 c and Table S 17). On the

cathode of the paired electrolyzer, the catalytic CO2 to CO reduction showed a moderate decline

in FECO, in SP but a much higher CO2 utilization (Figure 29 c, Figure S 78-Figure S 81 and

Supplementary Discussion 8). A rising inlet gas pressure from 0.2 bar to 1.2 bar in front of the

cell and an increasing N2 vol% in the GC (gas chromatograph) prompt us to safely conclude

that salt precipitation blockage due to potassium ions crossing the membrane to the cathode

started to occur during the stability test. The crossover led to potassium (bi-)carbonate

precipitation inside the cathode, raising the pressure and consuming CO2 via non-catalytic acid

base reactions. As only the pressure in front of the cell increases, we assume that the flow rate

decreases with increasing pressure. In the end, we can correlate the increasing cell potential

with an insufficient CO2 feed due to salt precipitation. Additionally, the potassium crossover

can reduce the proton transport to the cathode, influencing the neutralization and leading to a

more alkaline local pH at the cathode.

7.3.4 Paired BPM Electrolyzer Cell Performance

Our paired PGM-free catalytic HMFOR and CO2RR BPM electrolyzer demonstrated

comparable to slightly higher single pass conversions of 60% on the anode and 33% at the

cathode (Table S 18) which correspond to a higher CO2/CO ratio of 2 to 1 in the outlet stream

compared to the two AEM reference HMF/H2O or CO2/ H2O electrolyzers. Cell voltage savings

of 120 mV were realized, despite the competitive OER process causing moderate FEFDCA values

(Figure S 77). Adding cooled HMF/KOH feed to our electrolyzer lead to comparable

performance parameters (Figure 29 c and Table S 17).

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

102

Alternatively, 100 mM HMF feeds with regular replacement of aliquots immediately prior

before entering the electrolyzer cell enabled high HMF concentration paired electrolysis. In that

configuration, the electrolyzer cell potential could be lowered to 3.45 V, and FEFDCA was raised

to 49% (Table S 17). Clearly, high feed concentrations carry the penalty of lower single pass

conversions and FDCA selectivity in favor of HFCA. However, lower single pass FDCA yield

is less of a problem if the HFCA is recycled (Table S 17). Either way, it was not possible to

substantially reduce the cell potential with a higher HMF concentration and the resulting

increase in FE.

To highlight the economic advantages of our paired BPM CO2RR/HMFOR electrolyzer device,

we evaluated electrical energy efficiencies (EE) and CO/FDCA product valorizations ($ per

year) at a given applied current density for both “single target product” reference electrolyzers

and the paired “multi target product” electrolyzers at 10 mM and 100 mM feed HMF

concentrations (Figure 29 d). The energy efficiencies for the “single target product”

electrolyzers (EEi, i= CO or FDCA) are given by the standard reaction cell voltages (ΔE0)

according to Equation 64 (SI), the Faradaic efficiencies (FEi) and the measured cell voltage

ECell

EEi=𝛥E0∙FEi

Ecell . (Equation 45)

This definition related the minimum Gibbs reaction energy-based input power to the actual

electrical power provided to the cell. At 200 mA cm-2, for instance, the CO energy efficiency

(EECO) of the present reference CO2/H2O electrolyzer reads

EECO =1.34 V ∙ 0.97

3.8 V =34% . (Equation 46)

Similarly, the FDCA energy efficiency (EEFDCA) of the reference HMF/H2O electrolyzer at

200 mA cm-2 reads

EEFDCA =0.113 V ∙ 0.12

3.28 V =4% . (Equation 47)

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

103

Both values serve as references for our paired electrolyzer system and are represented in Figure

29 d. In the case of more than one desired target product at either electrode coupled to one and

the same counter electrode process, the total EE can be simply obtained by the sum of the

individual target product efficiencies according

∑EEi=EE1+EE2+⋯

i . (Equation 48)

In our case, however, we have one desired as well as one undesired product on either

electrolyzer electrode. Under a 200% cell assumption, one may still add up the individual

desired target product efficiencies according

∑EEi=EECO +EEFDCAi . (Equation 49)

However, the cathodic CO and the anodic FDCA production currents and corresponding

charges split among a total of four possible cell reaction pairs (CO2/HMF, CO2/H2O, and

HMF/CO2, HMF/H2O), each with distinct Gibbs energy and standard reaction cell voltage, ΔE0.

This is why the EECO and EEFDCA terms must each be split into two EE terms corresponding to

the reaction pairs CO2/H2O (EECOH2O), CO2/HMF (EECOHMF) and HMF/CO2 (EEFDCACO2),

HMF/H2O (EEFDCAH2O). Each of these terms must be weighted by the FEi values of the counter

reaction products, regardless of whether desired or not. The weight accounts for the charge

distribution among the possible cell reactions. The sum of all individual energy efficiencies

thus reads

∑EEi=

iFEO2 ∙EECOH2O +FEFDCA ∙EECOHMF +FECO ∙EEFDCACO2

+FEH2 ∙EEFDCAH2O. (Equation 50)

In the present case, the sum of the CO and FDCA energy efficiencies of the CO2/HMF

electrolyzer at 200 mA cm-2 amounts to

∑EEi=

i0.88∙1.34 V ∙ 0.97

3.65 V +0.12∙0.22 V ∙ 0.97

3.65 V +0.97∙0.22 V ∙ 0.12

3.65 V

+0.03∙0.11 V ∙ 0.12

3.65 V =33% . (Equation 51)

To exclude the penalty caused by the BPM (Emembrane corrected, Equation 65), the corresponding

0.8V were subtracted from Ecell. Here, exemplarily shown for the membrane corrected CO

energy efficiency (EECOcorrected) of the CO2/H2O electrolyzer at 200 mA cm-2 according to

Equation 46 and Equation 65 (SI)

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

104

EECOcorrected =1.34 V ∙ 0.97

(3.80 V−0.80 V) =43% . (Equation 52)

When considering the combined electrolyzer, it would also be possible to calculate an EE

averaged over the number of electrodes. The value calculated in Equation 51 would be divided

by 2. The resulting and comparably low EE of 16.5% is not surprising since, as shown in

Figure 29 d and Table S 19, the low EE of the anode is included in the averaging.

Our electrolyzer cell energy efficiency estimations in Figure 29 d (black circles) and

Table S 19 suggest that at comparable EE (CO2/H2O and CO2/HMF electrolyzer), the paired

electrolyzer doubled its product valorization (Equation 66) compared to either CO2 or HMF

reference electrolyzer by producing FDCA instead of oxygen at the anode. When increasing

the HMF concentration from 10 mM to 100 mM, it could result in a fourfold increase of

valorization. The capital costs of electrolyzer hardware and components are not considered. In

addition, separation costs were neglected because no adequate cost calculations could be found.

Note here that this is not a complete techno economic analysis. Instead, we consider the power

cost and the market price of the products to give a valorization estimate.

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

105

Figure 29: Combination of CO2RR with continuous HMFOR. a) Schematic of the combined CO2RR and HMFOR cell. b)

Performance parameter of the combined cell at different current densities. 20 ml min-1 humidified CO2, 1M KOH and 10 mM

HMF with 2 ml min-1 flowrate and NiFe(-Cl-)-LDH@NF anode at RT. Standard deviation between 2-3%. c) Comparison

between refreshing the anolyte (120 ml) for 5h and cooling the anolyte (600 ml) for 5h and the CO2 reduction of the initial

performance and performance after 5h with the data from Table S3. d) Economic evaluation showing the energy efficiency

(EE) for electrolyzers with BPM delay (yellow/orange), without BPM delay (black), and the product valorization in $ per m2

and year (FDCA in blue and CO in red) for the separated and combined electrolyzers, assuming an FDCA valorization between

375 and 1548 $ t-1.234,235

Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-Free Bipolar Membrane Electrolyzer

106

7.4 Conclusion

In conclusion, we stepwise designed, operated, and analyzed the first paired BPM-based

CO2/HMF electrolyzer cell with all PGM-free electrodes. We demonstrated the operation at

industrially relevant current densities with unprecedented high single pass conversions on both

electrodes and high FDCA (SFDCA) and CO (FECO) selectivities. Paired electrolysis offered

significantly enhanced valorization thanks to the HMF oxidation to FDCA but lower energy

efficiency. Further improvements of the paired electrolyzer will be possible if the competing

OER process at the anode can be suppressed at higher current density. This would require

further catalyst development, such as optimizing the indirect catalysis mechanism and thus

forming the catalytic active species at lower potentials. Future improvements at the CO2-to-CO

cathode include suppression of electrode flooding and water management to increase durability

and stability.

Conclusion and Perspective

107

8 Conclusion and Perspective

This work aimed to deepen the understanding of HMFOR and HMFRR. The focus was firstly

on synthesizing highly active and stable catalysts for the oxidation and reduction of HMF. As

well as broadening the horizons of these two reactions in terms of scaling. It was finally possible

to selectively oxidize and reduce HMF in discontinuous MEA AEMWEs. High HMF

conversion rates and high selectivity towards valuable products were obtained for both

reactions. These were maintained even with the combination of HMFOR and HMFRR in one

cell. In addition, HMFOR was established in a continuously operated CO2 electrolyzer.

Figure 30: Illustration of the investigated parts leading to efficient and universal usable alkaline HMF water

electrolyzer.

In Chapter 4, the scaling behavior of an established OER catalyst was critically addressed. To

investigate in detail the issues arising in the scaling from RDE to AEMWE, different

anion-modified NiFe-LDH-based catalysts were synthesized, characterized, and thoroughly

measured in both cell designs. Revealing an enormously increased OER activity after anion

exchange from NiFe(-CO32-)-LDH@NF to NiFe(-Cl-)-LDH@NF at RDE scale. Unfortunately,

this trend was not transferable to the AEMWE scale. In summary, this study strongly confirmed

that scaling up a highly active OER catalyst system from RDE to AEMWE scale is by no means

trivial. In this case, iron incorporation and the change in local pH at the anode could be

identified as crucial activity factors after upscaling.

Conclusion and Perspective

108

In following Chapter 5, the focus was on the investigation of HMFOR. This was preceded by

investigations of the optimal catalytic substrate, the initial HMF concentration, the membrane,

and the electrolyte cation. Nickel, 10 mM HMF concentration, the FAA-3-PK-130 membrane

from Fumatech, and KOH electrolyte were the most suitable. Further investigation of nickel-

based layered double hydroxides catalysts revealed NiFe(-CO32-)-LDH@NF and, even more,

the anion exchanged NiFe(-Cl-)-LDH@NF to be the most active HMFOR catalyst. For

NiFe(-Cl-)-LDH@NF, high stability could be demonstrated, and with DEMS investigations,

the optimal potential window for HMFOR could be determined. Moreover, DEMS

investigations confirmed the suppression of OER by HMFOR. Finally, the system developed

here was compared to previously published HMFOR systems based on the electrode area

normalized production rate (space-time yield). The comparison showed the highest space-time

yield observed for the here explored system.

The following 6th chapter investigated the HMFRR. In particular, the competing reactions of

hydrogenation to BHMF and hydrogenolysis to MFA in an alkaline membrane HMF

electrolyzer were investigated. It was shown that the CuO based catalysts enhanced with NiO,

Fe2O3, and Co3O4 exhibited widely varying selectivity for BHMF and MFA as well as different

HER activities. By characterizing the CuO based catalysts before and after electrolysis, it

became clear that partially metallic sites were formed in-situ at the catalyst. However, compared

to purely metallic catalysts, the oxide-derived materials showed significantly higher HMFRR

activity and MFA selectivity. Based on these results, it was concluded that HMFRR product

selectivity in alkaline electrolyte strongly correlates with the ability of the catalyst to provide

adsorbed protons and that Cu-based oxide-derived catalysts promote hydrogenolysis. Finally,

HMFRR could be efficiently combined with HMFOR in one cell. No activity loss occurred

compared to the separately considered HMFRR and HMFOR systems.

Chapter 7 shows the combination of the HMFOR with the CO2RR in a BPM water electrolyzer

(BPMWE) cell. In addition, the evolution from a discontinuous AEMWE HMFOR system

(Chapter 5) to a continuous BPMWE HMFOR system is shown. Initial HMF concentration and

electrolyte flow rate studies were performed to optimize HMFOR at high current densities. Due

to a low electrolyte flow rate, a significant increase in single pass conversion was achieved.

Conclusion and Perspective

109

Raising the initial HMF concentration was favorable only if the autocatalytic reaction to humins

could be avoided by cooling the electrolyte or mixing HMF and KOH directly before entering

the cell. Finally, despite lower Faradaic efficiency (compared to the system from Chapter 5), it

was possible to present an efficient HMFOR/CO2RR electrolyzer. However, it is important to

mention that the more dominant OER and the undesired reaction to humins remained

challenging with high current densities.

In summary, efficient systems were developed for HMFOR, HMFRR, and HMFOR coupled

AEMWEs. Besides fundamental insights into the oxidation and reduction of HMF, this work

also paves the way for the conversion of HMF on a larger scale.

8.1 Perspective

This thesis shows the importance of electrochemical oxidation and reduction of HMF, and thus

also of biomass, in the future. There is no question that especially the research on HMFOR at

laboratory scale shows very high HMF conversions and FDCA yields and selectivity. Faradaic

efficiencies and catalytic stabilities also show high potential for future use. Despite all this, a

completely elucidated mechanism for HMFOR is still missing. Furthermore, it is clear from

this thesis that the next step of research toward industrial applications will be challenging but

must be taken to see how feasible the concept of large-scale electrochemical oxidation of

biomass is. In particular, if it is intended to replace reactions such as OER in water or CO2

electrolyzers with HMFOR, the catalysts must remain selective and stable under industrial

conditions. Unfortunately, apart from this thesis, there is little to no evidence for it so far.

Since HMF reduction is still in its infancy compared to HMF oxidation, the focus in the

following years should continue to be on fundamental research. This mainly focuses on

elucidating the mechanism in acidic and alkaline environments. At the same time, product

selectivity should be increased, especially for the hydrogenolysis products, as these will find

high utilization. In particular, the production of DMF, which could be used as a potential bio

e-fuel, is practically impossible to synthesize at the current state of research.

Conclusion and Perspective

110

Finally, the handling of HMF under alkaline conditions is a major problem. Ideas and concepts

have to be developed that allow high HMF concentrations to be stored and converted in the

long term, even in an alkaline environment, without a large proportion reacting autocatalytically

to humins. Also necessary for an economically viable process is the cost reduction of HMF.

According to techno-economic calculations, purchasing HMF represents the largest cost factor.

Therefore, further research and development on the sustainable production of HMF is essential.

Finally, efficient and cost-effective methods for separating HMF products are needed. These

three points, handling, HMF costs, and separation of products, must therefore be advanced

independently of direct research on the electrochemical HMF conversion in order to establish

an overall profitable process.

References

111

9 References

1 REN21. Renewables 2022 Global Status Report, https://www.ren21.net/wp-

content/uploads/2019/05/GSR2022_Full_Report.pdf (2022).

2 Bürgerdialog. Energieerzeugung- und verbrauch, www.buergerdialog-

stromnetz.de/wissenswertes/energieerzeugung-und-verbrauch/ (2022).

3 United Nations. Paris Agreement,

https://unfccc.int/files/meetings/paris_nov_2015/application/pdf/paris_agreement_english_.pd

f (2015).

4 Haszeldine, R. S. Carbon Capture and Storage: How Green Can Black Be? Science 325, 1647-

1652, doi:10.1126/science.1172246 (2009).

5 Boot-Handford, M. E., Abanades, J. C., Anthony, E. J., Blunt, M. J., Brandani, S., Mac

Dowell, N., Fernández, J. R., Ferrari, M.-C., Gross, R., Hallett, J. P., Haszeldine, R. S.,

Heptonstall, P., Lyngfelt, A., Makuch, Z., Mangano, E., Porter, R. T. J., Pourkashanian, M.,

Rochelle, G. T., Shah, N., Yao, J. G. & Fennell, P. S. Carbon capture and storage update.

Energy & Environmental Science 7, 130-189, doi:10.1039/C3EE42350F (2014).

6 Lehner, F. & Hart, D. in Electrochemical Power Sources: Fundamentals, Systems, and

Applications, (eds Tom Smolinka & Jurgen Garche) Ch. 1, pp. 1-36 (Elsevier, 2022).

7 Smolinka, T., Ojong, E. T. & Garche, J. in Electrochemical Energy Storage for Renewable

Sources and Grid Balancing, (eds Patrick T. Moseley & Jürgen Garche) Ch. 8, pp. 103-128

(Elsevier, 2015).

8 Li, D., Motz, A. R., Bae, C., Fujimoto, C., Yang, G., Zhang, F.-Y., Ayers, K. E. & Kim, Y. S.

Durability of anion exchange membrane water electrolyzers. Energy & Environmental Science

14, 3393-3419, doi:10.1039/D0EE04086J (2021).

9 Miller, H. A., Bouzek, K., Hnat, J., Loos, S., Bernäcker, C. I., Weißgärber, T., Röntzsch, L. &

Meier-Haack, J. Green hydrogen from anion exchange membrane water electrolysis: a review

of recent developments in critical materials and operating conditions. Sustainable Energy &

Fuels 4, 2114-2133, doi:10.1039/C9SE01240K (2020).

10 Liu, J., Gerhardt, M. R., Li, D., Pak, M., Kang, Z., Alia, S. M., Bender, G., Kim, Y. S. &

Weber, A. Z. Mathematical Modeling of Hydroxide-Exchange-Membrane Water Electrolyzer.

ECS Meeting Abstracts MA2020-02, 2443, doi:10.1149/MA2020-02382443mtgabs (2020).

11 Ayers, K., Danilovic, N., Ouimet, R., Carmo, M., Pivovar, B. & Bornstein, M. Perspectives on

Low-Temperature Electrolysis and Potential for Renewable Hydrogen at Scale. Annual

Review of Chemical and Biomolecular Engineering 10, 219-239, doi:10.1146/annurev-

chembioeng-060718-030241 (2019).

12 Durst, J., Siebel, A., Simon, C., Hasché, F., Herranz, J. & Gasteiger, H. A. New insights into

the electrochemical hydrogen oxidation and evolution reaction mechanism. Energy &

Environmental Science 7, 2255-2260, doi:10.1039/C4EE00440J (2014).

13 Sheng, W., Gasteiger, H. A. & Shao-Horn, Y. Hydrogen Oxidation and Evolution Reaction

Kinetics on Platinum: Acid vs Alkaline Electrolytes. Journal of The Electrochemical Society

157, B1529, doi:10.1149/1.3483106 (2010).

14 Liu, E., Li, J., Jiao, L., Doan, H. T. T., Liu, Z., Zhao, Z., Huang, Y., Abraham, K. M.,

Mukerjee, S. & Jia, Q. Unifying the Hydrogen Evolution and Oxidation Reactions Kinetics in

Base by Identifying the Catalytic Roles of Hydroxyl-Water-Cation Adducts. Journal of the

American Chemical Society 141, 3232-3239, doi:10.1021/jacs.8b13228 (2019).

15 Subbaraman, R., Tripkovic, D., Chang, K.-C., Strmcnik, D., Paulikas, A. P., Hirunsit, P.,

Chan, M., Greeley, J., Stamenkovic, V. & Markovic, N. M. Trends in activity for the water

electrolyser reactions on 3d M(Ni,Co,Fe,Mn) hydr(oxy)oxide catalysts. Nature Materials 11,

550-557, doi:10.1038/nmat3313 (2012).

16 Subbaraman, R., Tripkovic, D., Strmcnik, D., Chang, K.-C., Uchimura, M., Paulikas, A. P.,

Stamenkovic, V. & Markovic, N. M. Enhancing Hydrogen Evolution Activity in Water

References

112

Splitting by Tailoring Li+-Ni(OH)2-Pt Interfaces. Science 334, 1256-1260,

doi:10.1126/science.1211934 (2011).

17 Danilovic, N., Subbaraman, R., Strmcnik, D., Chang, K.-C., Paulikas, A. P., Stamenkovic, V.

R. & Markovic, N. M. Enhancing the Alkaline Hydrogen Evolution Reaction Activity through

the Bifunctionality of Ni(OH)2/Metal Catalysts. Angewandte Chemie International Edition

51, 12495-12498, doi:https://doi.org/10.1002/anie.201204842 (2012).

18 Bates, M. K., Jia, Q., Ramaswamy, N., Allen, R. J. & Mukerjee, S. Composite Ni/NiO-Cr2O3

Catalyst for Alkaline Hydrogen Evolution Reaction. The Journal of Physical Chemistry C

119, 5467-5477, doi:10.1021/jp512311c (2015).

19 Gong, M., Zhou, W., Tsai, M.-C., Zhou, J., Guan, M., Lin, M.-C., Zhang, B., Hu, Y., Wang,

D.-Y., Yang, J., Pennycook, S. J., Hwang, B.-J. & Dai, H. Nanoscale nickel oxide/nickel

heterostructures for active hydrogen evolution electrocatalysis. Nature Communications 5,

4695, doi:10.1038/ncomms5695 (2014).

20 Zhang, J., Wang, T., Liu, P., Liao, Z., Liu, S., Zhuang, X., Chen, M., Zschech, E. & Feng, X.

Efficient hydrogen production on MoNi4 electrocatalysts with fast water dissociation kinetics.

Nature Communications 8, 15437, doi:10.1038/ncomms15437 (2017).

21 Hu, C., Zhang, L. & Gong, J. Recent progress made in the mechanism comprehension and

design of electrocatalysts for alkaline water splitting. Energy & Environmental Science 12,

2620-2645, doi:10.1039/C9EE01202H (2019).

22 Chen, J. Y. C., Dang, L., Liang, H., Bi, W., Gerken, J. B., Jin, S., Alp, E. E. & Stahl, S. S.

Operando Analysis of NiFe and Fe Oxyhydroxide Electrocatalysts for Water Oxidation:

Detection of Fe4+ by Mössbauer Spectroscopy. Journal of the American Chemical Society

137, 15090-15093, doi:10.1021/jacs.5b10699 (2015).

23 Trotochaud, L., Young, S. L., Ranney, J. K. & Boettcher, S. W. Nickel–Iron Oxyhydroxide

Oxygen-Evolution Electrocatalysts: The Role of Intentional and Incidental Iron Incorporation.

Journal of the American Chemical Society 136, 6744-6753, doi:10.1021/ja502379c (2014).

24 González-Flores, D., Klingan, K., Chernev, P., Loos, S., Mohammadi, M. R., Pasquini, C.,

Kubella, P., Zaharieva, I., Smith, R. D. L. & Dau, H. Nickel-iron catalysts for electrochemical

water oxidation – redox synergism investigated by in situ X-ray spectroscopy with millisecond

time resolution. Sustainable Energy & Fuels 2, 1986-1994, doi:10.1039/C8SE00114F (2018).

25 Dionigi, F., Zhu, J., Zeng, Z., Merzdorf, T., Sarodnik, H., Gliech, M., Pan, L., Li, W.-X.,

Greeley, J. & Strasser, P. Intrinsic Electrocatalytic Activity for Oxygen Evolution of

Crystalline 3d-Transition Metal Layered Double Hydroxides. Angewandte Chemie

International Edition 60, 14446-14457, doi:https://doi.org/10.1002/anie.202100631 (2021).

26 Dionigi, F., Zeng, Z., Sinev, I., Merzdorf, T., Deshpande, S., Lopez, M. B., Kunze, S.,

Zegkinoglou, I., Sarodnik, H., Fan, D., Bergmann, A., Drnec, J., Araujo, J. F. D., Gliech, M.,

Teschner, D., Zhu, J., Li, W.-X., Greeley, J., Cuenya, B. R. & Strasser, P. In-situ structure and

catalytic mechanism of NiFe and CoFe layered double hydroxides during oxygen evolution.

Nature Communications 11, doi:10.1038/s41467-020-16237-1 (2020).

27 Smith, R. D. L., Pasquini, C., Loos, S., Chernev, P., Klingan, K., Kubella, P., Mohammadi, M.

R., González-Flores, D. & Dau, H. Geometric distortions in nickel (oxy)hydroxide

electrocatalysts by redox inactive iron ions. Energy & Environmental Science 11, 2476-2485,

doi:10.1039/C8EE01063C (2018).

28 Speck, F. D., Dettelbach, K. E., Sherbo, R. S., Salvatore, D. A., Huang, A. & Berlinguette, C.

P. On the Electrolytic Stability of Iron-Nickel Oxides. Chem 2, 590-597,

doi:https://doi.org/10.1016/j.chempr.2017.03.006 (2017).

29 Lucas, F. W. S., Grim, R. G., Tacey, S. A., Downes, C. A., Hasse, J., Roman, A. M.,

Farberow, C. A., Schaidle, J. A. & Holewinski, A. Electrochemical Routes for the

Valorization of Biomass-Derived Feedstocks: From Chemistry to Application. ACS Energy

Letters, 1205-1270, doi:10.1021/acsenergylett.0c02692 (2021).

References

113

30 Kim, H. J., Lee, J., Green, S. K., Huber, G. W. & Kim, W. B. Selective Glycerol Oxidation by

Electrocatalytic Dehydrogenation. ChemSusChem 7, 1051-1056,

doi:https://doi.org/10.1002/cssc.201301218 (2014).

31 Costa Santos, J. B., Vieira, C., Crisafulli, R. & Linares, J. J. Promotional effect of auxiliary

metals Bi on Pt, Pd, and Ag on Au, for glycerol electrolysis. International Journal of

Hydrogen Energy 45, 25658-25671, doi:https://doi.org/10.1016/j.ijhydene.2019.11.225

(2020).

32 Houache, M. S. E., Hughes, K., Safari, R., Botton, G. A. & Baranova, E. A. Modification of

Nickel Surfaces by Bismuth: Effect on Electrochemical Activity and Selectivity toward

Glycerol. ACS Applied Materials & Interfaces 12, 15095-15107, doi:10.1021/acsami.9b22378

(2020).

33 Kwon, Y., Birdja, Y., Spanos, I., Rodriguez, P. & Koper, M. T. M. Highly Selective Electro-

Oxidation of Glycerol to Dihydroxyacetone on Platinum in the Presence of Bismuth. ACS

Catalysis 2, 759-764, doi:10.1021/cs200599g (2012).

34 Wu, H., Song, J., Xie, C., Hu, Y., Zhang, P., Yang, G. & Han, B. Surface engineering in PbS

via partial oxidation: towards an advanced electrocatalyst for reduction of levulinic acid to γ-

valerolactone. Chemical Science 10, 1754-1759, doi:10.1039/C8SC03161D (2019).

35 Qiu, Y., Xin, L., Chadderdon, D. J., Qi, J., Liang, C. & Li, W. Integrated electrocatalytic

processing of levulinic acid and formic acid to produce biofuel intermediate valeric acid.

Green Chemistry 16, 1305-1315, doi:10.1039/C3GC42254B (2014).

36 dos Santos, T. R., Nilges, P., Sauter, W., Harnisch, F. & Schröder, U. Electrochemistry for the

generation of renewable chemicals: electrochemical conversion of levulinic acid. RSC

Advances 5, 26634-26643, doi:10.1039/C4RA16303F (2015).

37 Nilges, P., dos Santos, T. R., Harnisch, F. & Schröder, U. Electrochemistry for biofuel

generation: Electrochemical conversion of levulinic acid to octane. Energy & Environmental

Science 5, 5231-5235, doi:10.1039/C1EE02685B (2012).

38 Sajid, M., Zhao, X. & Liu, D. Production of 2,5-furandicarboxylic acid (FDCA) from 5-

hydroxymethylfurfural (HMF): recent progress focusing on the chemical-catalytic routes.

Green Chemistry 20, 5427-5453, doi:10.1039/C8GC02680G (2018).

39 Yang, Y. & Mu, T. Electrochemical oxidation of biomass derived 5-hydroxymethylfurfural

(HMF): pathway, mechanism, catalysts and coupling reactions. Green Chemistry 23, 4228-

4254, doi:10.1039/D1GC00914A (2021).

40 Totaro, G., Sisti, L., Marchese, P., Colonna, M., Romano, A., gioia, c., Vannini, M. & Celli,

A. Current advances in the sustainable conversion of HMF into FDCA. ChemSusChem 15,

e202200501, doi:https://doi.org/10.1002/cssc.202200501 (2022).

41 Liu, S., Zhu, Y., Liao, Y., Wang, H., Liu, Q., Ma, L. & Wang, C. Advances in understanding

the humins: formation , prevention and application. Applications in Energy and Combustion

Science 10, 100062, doi:https://doi.org/10.1016/j.jaecs.2022.100062 (2022).

42 Motagamwala, A. H., Won, W., Sener, C., Alonso, D. M., Maravelias, C. T. & Dumesic, J. A.

Toward biomass-derived renewable plastics: Production of 2,5-furandicarboxylic acid from

fructose. Science Advances 4, eaap9722, doi:10.1126/sciadv.aap9722 (2018).

43 Hauke, P., Merzdorf, T., Klingenhof, M. & Strasser, P. Hydrogenation versus hydrogenolysis

during alkaline electrochemical valorization of 5-hydroxymethylfurfural over oxide-derived

Cu-bimetallics. Nature Communications 14, 4708, doi:10.1038/s41467-023-40463-y (2023).

44 Van Uytvanck, P. P., Haire, G., Marshall, P. J. & Dennis, J. S. Impact on the Polyester Value

Chain of Using p-Xylene Derived from Biomass. ACS Sustainable Chemistry & Engineering

5, 4119-4126, doi:10.1021/acssuschemeng.7b00105 (2017).

45 Jiang, L., Gonzalez-Diaz, A., Ling-Chin, J., Malik, A., Roskilly, A. P. & Smallbone, A. J. PEF

plastic synthesized from industrial carbon dioxide and biowaste. Nature Sustainability 3, 761-

767, doi:10.1038/s41893-020-0549-y (2020).

References

114

46 Simoska, O., Rhodes, Z., Weliwatte, S., Cabrera-Pardo, J., Gaffney, E., Lim, K. & Minteer, S.

Advances in Electrocatalytic Modification Strategies of 5-Hydroxymethylfurfural (HMF).

ChemSusChem 14, 1674-1686, doi:https://doi.org/10.1002/cssc.202100139.

47 Li, Y., Wei, X., Chen, L. & Shi, J. Electrocatalytic Hydrogen Production Trilogy. Angewandte

Chemie International Edition 60, 19550-19571, doi:https://doi.org/10.1002/anie.202009854

(2021).

48 Zhou, B., Dong, C.-L., Huang, Y.-C., Zhang, N., Wu, Y., Lu, Y., Yue, X., Xiao, Z., Zou, Y. &

Wang, S. Activity origin and alkalinity effect of electrocatalytic biomass oxidation on nickel

nitride. Journal of Energy Chemistry 61, 179-185,

doi:https://doi.org/10.1016/j.jechem.2021.02.026 (2021).

49 Ge, R., Wang, Y., Li, Z., Xu, M., Xu, S.-M., Zhou, H., Ji, K., Chen, F., Zhou, J. & Duan, H.

Selective Electrooxidation of Biomass-Derived Alcohols to Aldehydes in a Neutral Medium:

Promoted Water Dissociation over a Nickel-Oxide-Supported Ruthenium Single-Atom

Catalyst. Angewandte Chemie International Edition 61, e202200211,

doi:https://doi.org/10.1002/anie.202200211 (2022).

50 Kubota, S. R. & Choi, K.-S. Electrochemical Oxidation of 5-Hydroxymethylfurfural to 2,5-

Furandicarboxylic Acid (FDCA) in Acidic Media Enabling Spontaneous FDCA Separation.

ChemSusChem 11, 2138-2145, doi:10.1002/cssc.201800532 (2018).

51 Vuyyuru, K. R. & Strasser, P. Oxidation of biomass derived 5-hydroxymethylfurfural using

heterogeneous and electrochemical catalysis. Catalysis Today 195, 144-154,

doi:https://doi.org/10.1016/j.cattod.2012.05.008 (2012).

52 Chadderdon, D. J., Xin, L., Qi, J., Qiu, Y., Krishna, P., More, K. L. & Li, W. Electrocatalytic

oxidation of 5-hydroxymethylfurfural to 2,5-furandicarboxylic acid on supported Au and Pd

bimetallic nanoparticles. Green Chemistry 16, 3778-3786, doi:10.1039/C4GC00401A (2014).

53 Wang, T., Tao, L., Zhu, X., Chen, C., Chen, W., Du, S., Zhou, Y., Zhou, B., Wang, D., Xie,

C., Long, P., Li, W., Wang, Y., Chen, R., Zou, Y., Fu, X.-Z., Li, Y., Duan, X. & Wang, S.

Combined anodic and cathodic hydrogen production from aldehyde oxidation and hydrogen

evolution reaction. Nature Catalysis 5, 66-73, doi:10.1038/s41929-021-00721-y (2022).

54 Woo, J., Moon, B. C., Lee, U., Oh, H.-S., Chae, K. H., Jun, Y., Min, B. K. & Lee, D. K.

Collaborative Electrochemical Oxidation of the Alcohol and Aldehyde Groups of 5-

Hydroxymethylfurfural by NiOOH and Cu(OH)2 for Superior 2,5-Furandicarboxylic Acid

Production. ACS Catalysis 12, 4078-4091, doi:10.1021/acscatal.1c05341 (2022).

55 Holzhäuser, F. J., Janke, T., Öztas, F., Broicher, C. & Palkovits, R. Electrocatalytic Oxidation

of 5-Hydroxymethylfurfural into the Monomer 2,5-Furandicarboxylic Acid using

Mesostructured Nickel Oxide. Adv. Sustain. Syst. 4, 1900151,

doi:https://doi.org/10.1002/adsu.201900151 (2020).

56 Bender, M. T., Warburton, R. E., Hammes-Schiffer, S. & Choi, K.-S. Understanding

Hydrogen Atom and Hydride Transfer Processes during Electrochemical Alcohol and

Aldehyde Oxidation. ACS Catalysis 11, 15110-15124, doi:10.1021/acscatal.1c04163 (2021).

57 Bender, M. T. & Choi, K.-S. Electrochemical Oxidation of HMF via Hydrogen Atom Transfer

and Hydride Transfer on NiOOH and the Impact of NiOOH Composition on These

Mechanisms. ChemSusChem 15, 202200675, doi:https://doi.org/10.1002/cssc.202200675

(2022).

58 Yang, Y., Xu, D., Zhang, B., Xue, Z. & Mu, T. Substrate molecule adsorption energy: An

activity descriptor for electrochemical oxidation of 5-Hydroxymethylfurfural (HMF).

Chemical Engineering Journal 433, 133842, doi:https://doi.org/10.1016/j.cej.2021.133842

(2022).

59 Taitt, B. J., Nam, D.-H. & Choi, K.-S. A Comparative Study of Nickel, Cobalt, and Iron

Oxyhydroxide Anodes for the Electrochemical Oxidation of 5-Hydroxymethylfurfural to 2,5-

Furandicarboxylic Acid. ACS Catalysis 9, 660-670, doi:10.1021/acscatal.8b04003 (2019).

References

115

60 Park, M., Gu, M. & Kim, B.-S. Tailorable Electrocatalytic 5-Hydroxymethylfurfural

Oxidation and H2 Production: Architecture–Performance Relationship in Bifunctional

Multilayer Electrodes. ACS Nano 14, 6812-6822, doi:10.1021/acsnano.0c00581 (2020).

61 Nam, D.-H., Taitt, B. J. & Choi, K.-S. Copper-Based Catalytic Anodes To Produce 2,5-

Furandicarboxylic Acid, a Biomass-Derived Alternative to Terephthalic Acid. ACS Catalysis

8, 1197-1206, doi:10.1021/acscatal.7b03152 (2018).

62 Poerwoprajitno, A. R., Gloag, L., Watt, J., Cychy, S., Cheong, S., Kumar, P. V., Benedetti, T.

M., Deng, C., Wu, K.-H., Marjo, C. E., Huber, D. L., Muhler, M., Gooding, J. J., Schuhmann,

W., Wang, D.-W. & Tilley, R. D. Faceted Branched Nickel Nanoparticles with Tunable

Branch Length for High-Activity Electrocatalytic Oxidation of Biomass. Angewandte Chemie

International Edition 59, 15487-15491, doi:https://doi.org/10.1002/anie.202005489 (2020).

63 Lu, X., Wu, K.-H., Zhang, B., Chen, J., Li, F., Su, B.-J., Yan, P., Chen, J.-M. & Qi, W. Highly

Efficient Electro-reforming of 5-Hydroxymethylfurfural on Vertically Oriented Nickel

Nanosheet/Carbon Hybrid Catalysts: Structure–Function Relationships. 60, 14528-14535,

doi:https://doi.org/10.1002/anie.202102359 (2021).

64 Weidner, J., Barwe, S., Sliozberg, K., Piontek, S., Masa, J., Apfel, U.-P. & Schuhmann, W.

Cobalt–metalloid alloys for electrochemical oxidation of 5-hydroxymethylfurfural as an

alternative anode reaction in lieu of oxygen evolution during water splitting. Beilstein Journal

of Organic Chemistry 14, 1436-1445, doi:10.3762/bjoc.14.121 (2018).

65 Lu, Y., Liu, T., Huang, Y.-C., Zhou, L., Li, Y., Chen, W., Yang, L., Zhou, B., Wu, Y., Kong,

Z., Huang, Z., Li, Y., Dong, C.-L., Wang, S. & Zou, Y. Integrated Catalytic Sites for Highly

Efficient Electrochemical Oxidation of the Aldehyde and Hydroxyl Groups in 5-

Hydroxymethylfurfural. ACS Catalysis, 4242-4251, doi:10.1021/acscatal.2c00174 (2022).

66 Gou, W., Chen, Y., Zhong, Y., Xue, Q., Li, J. & Ma, Y. Phytate-coordinated nickel foam with

enriched NiOOH intermediates for 5-hydroxymethylfurfural electrooxidation. Chemical

Communications, doi:10.1039/D2CC02182J (2022).

67 Wang, C., Schüth, F., Bongard, H. & Yu, M. Highly ordered mesoporous Co3O4

electrocatalyst for efficient, selective and stable oxidation of 5-hydroxymethylfurfural to 2,5-

furandicarboxylic acid. ChemSusChem 14, 5199-5206,

doi:https://doi.org/10.1002/cssc.202002762.

68 Zhao, G., Hai, G., Zhou, P., Liu, Z., Zhang, Y., Peng, B., Xia, W., Huang, X. & Wang, G.

Electrochemical Oxidation of 5-Hydroxymethylfurfural on CeO2-Modified Co3O4 with

Regulated Intermediate Adsorption and Promoted Charge Transfer. Advanced Functional

Materials 33, 2213170, doi:https://doi.org/10.1002/adfm.202213170 (2023).

69 Grabowski, G., Lewkowski, J. & Skowroński, R. The electrochemical oxidation of 5-

hydroxymethylfurfural with the nickel oxide/hydroxide electrode. Electrochimica Acta 36,

1995, doi:https://doi.org/10.1016/0013-4686(91)85084-K (1991).

70 Gao, L., Bao, Y., Gan, S., Sun, Z., Song, Z., Han, D., Li, F. & Niu, L. Hierarchical Nickel–

Cobalt-Based Transition Metal Oxide Catalysts for the Electrochemical Conversion of

Biomass into Valuable Chemicals. ChemSusChem 11, 2547-2553,

doi:10.1002/cssc.201800695 (2018).

71 Zhong, Y., Ren, R., Qin, L., Wang, J., Peng, Y., Li, Q. & Fan, Y. Electrodeposition of hybrid

nanosheet-structured NiCo2O4 on carbon fiber paper as a non-noble electrocatalyst for

efficient electrooxidation of 5-hydroxymethylfurfural to 2,5-furandicarboxylic acid. New

Journal of Chemistry, doi:10.1039/D1NJ01489G (2021).

72 Gao, D., Han, F., Waterhouse, G. I. N., Li, Y. & Zhang, L. A highly efficient iron

phthalocyanine-intercalated CuFe-LDH catalyst for the selective oxidation of 5-

hydroxymethylfurfural to 5-formyl-2-furanic acid. Catalysis Communications 173, 106561,

doi:https://doi.org/10.1016/j.catcom.2022.106561 (2023).

73 Liu, W.-J., Dang, L., Xu, Z., Yu, H.-Q., Jin, S. & Huber, G. W. Electrochemical Oxidation of

5-Hydroxymethylfurfural with NiFe Layered Double Hydroxide (LDH) Nanosheet Catalysts.

ACS Catalysis 8, 5533-5541, doi:10.1021/acscatal.8b01017 (2018).

References

116

74 Musella, E., Gualandi, I., Scavetta, E., Rivalta, A., Venuti, E., Christian, M., Morandi, V.,

Mullaliu, A., Giorgetti, M. & Tonelli, D. Newly developed electrochemical synthesis of Co-

based layered double hydroxides: toward noble metal-free electro-catalysis. Journal of

Materials Chemistry A 7, 11241-11249, doi:10.1039/C8TA11812D (2019).

75 Seo, B., Woo, J., Kim, E., Cheong, S.-H., Lee, D. K. & Lee, H. Insight toward the role of Fe

in layered Ni(OH)2 for electrochemical oxidations of water and 5-hydroxymethylfurfural.

Catalysis Communications 170, 106501, doi:https://doi.org/10.1016/j.catcom.2022.106501

(2022).

76 Zhang, R., Jiang, S., Rao, Y., Chen, S., Yue, Q. & Kang, Y. Electrochemical biomass

upgrading on CoOOH nanosheets in a hybrid water electrolyzer. Green Chemistry,

doi:10.1039/D0GC04157B (2021).

77 Bai, X.-J., He, W.-X., Lu, X.-Y., Fu, Y. & Qi, W. Electrochemical oxidation of 5-

hydroxymethylfurfural on ternary metal–organic framework nanoarrays: enhancement from

electronic structure modulation. Journal of Materials Chemistry A 9, 14270-14275,

doi:10.1039/D1TA02464G (2021).

78 Deng, X., Li, M., Fan, Y., Wang, L., Fu, X.-Z. & Luo, J.-L. Constructing multifunctional

‘Nanoplatelet-on-Nanoarray’ electrocatalyst with unprecedented activity towards novel

selective organic oxidation reactions to boost hydrogen production. Applied Catalysis B:

Environmental 278, 119339, doi:https://doi.org/10.1016/j.apcatb.2020.119339 (2020).

79 Cai, M., Zhang, Y., Zhao, Y., Liu, Q., Li, Y. & Li, G. Two-dimensional metal–organic

framework nanosheets for highly efficient electrocatalytic biomass 5-(hydroxymethyl)furfural

(HMF) valorization. Journal of Materials Chemistry A 8, 20386-20392,

doi:10.1039/D0TA07793C (2020).

80 Nguyen, H. L. Metal–Organic Frameworks Can Photocatalytically Split Water—Why Not?

Advanced Materials 34, 2200465, doi:https://doi.org/10.1002/adma.202200465 (2022).

81 Barwe, S., Weidner, J., Cychy, S., Morales, D. M., Dieckhöfer, S., Hiltrop, D., Masa, J.,

Muhler, M. & Schuhmann, W. Electrocatalytic Oxidation of 5-(Hydroxymethyl)furfural Using

High-Surface-Area Nickel Boride. Angewandte Chemie International Edition 57, 11460-

11464, doi:10.1002/anie.201806298 (2018).

82 You, B., Jiang, N., Liu, X. & Sun, Y. Simultaneous H2 Generation and Biomass Upgrading in

Water by an Efficient Noble-Metal-Free Bifunctional Electrocatalyst. Angewandte Chemie

International Edition 55, 9913-9917, doi:https://doi.org/10.1002/anie.201603798 (2016).

83 You, B., Liu, X., Jiang, N. & Sun, Y. A General Strategy for Decoupled Hydrogen Production

from Water Splitting by Integrating Oxidative Biomass Valorization. Journal of the American

Chemical Society 138, 13639-13646, doi:10.1021/jacs.6b07127 (2016).

84 Zhang, N., Zou, Y., Tao, L., Chen, W., Zhou, L., Liu, Z., Zhou, B., Huang, G., Lin, H. &

Wang, S. Electrochemical Oxidation of 5-Hydroxymethylfurfural on Nickel Nitride/Carbon

Nanosheets: Reaction Pathway Determined by In Situ Sum Frequency Generation Vibrational

Spectroscopy. Angewandte Chemie International Edition 58, 15895-15903,

doi:https://doi.org/10.1002/anie.201908722 (2019).

85 Li, S., Sun, X., Yao, Z., Zhong, X., Cao, Y., Liang, Y., Wei, Z., Deng, S., Zhuang, G., Li, X.

& Wang, J. Biomass Valorization via Paired Electrosynthesis Over Vanadium Nitride-Based

Electrocatalysts. Advanced Functional Materials 29, 1904780,

doi:https://doi.org/10.1002/adfm.201904780 (2019).

86 Yang, C., Wang, C., Zhou, L., Duan, W., Song, Y., Zhang, F., Zhen, Y., Zhang, J., Bao, W.,

Lu, Y., Wang, D. & Fu, F. Refining d-band center in Ni0.85Se by Mo doping: A strategy for

boosting hydrogen generation via coupling electrocatalytic oxidation 5-

hydroxymethylfurfural. Chemical Engineering Journal 422, 130125,

doi:https://doi.org/10.1016/j.cej.2021.130125 (2021).

87 Huang, X., Song, J., Hua, M., Xie, Z., Liu, S., Wu, T., Yang, G. & Han, B. Enhancing the

electrocatalytic activity of CoO for the oxidation of 5-hydroxymethylfurfural by introducing

oxygen vacancies. Green Chemistry 22, 843-849, doi:10.1039/C9GC03698A (2020).

References

117

88 Li, S., Wang, S., Wang, Y., He, J., Li, K., Xu, Y., Wang, M., Zhao, S., Li, X., Zhong, X. &

Wang, J. Doped Mn Enhanced NiS Electrooxidation Performance of HMF into FDCA at

Industrial-Level Current Density. Advanced Functional Materials n/a, 2214488,

doi:https://doi.org/10.1002/adfm.202214488 (2023).

89 Moreau, C., Belgacem, M. N. & Gandini, A. Recent Catalytic Advances in the Chemistry of

Substituted Furans from Carbohydrates and in the Ensuing Polymers. Topics in Catalysis 27,

11-30, doi:10.1023/B:TOCA.0000013537.13540.0e (2004).

90 May, A. S., Watt, S. M. & Biddinger, E. J. Kinetics of furfural electrochemical hydrogenation

and hydrogenolysis in acidic media on copper. Reaction Chemistry & Engineering 6, 2075-

2086, doi:10.1039/D1RE00216C (2021).

91 Lee, D. K., Kubota, S. R., Janes, A. N., Bender, M. T., Woo, J., Schmidt, J. R. & Choi, K.-S.

The Impact of 5-Hydroxymethyl Furfural (HMF)-Metal Interactions on the Electrochemical

Reduction Pathways of HMF on Various Metal Electrodes. ChemSusChem 14, 4563-4572,

doi:https://doi.org/10.1002/cssc.202101037 (2021).

92 Chadderdon, X. H., Chadderdon, D. J., Matthiesen, J. E., Qiu, Y., Carraher, J. M., Tessonnier,

J.-P. & Li, W. Mechanisms of Furfural Reduction on Metal Electrodes: Distinguishing

Pathways for Selective Hydrogenation of Bioderived Oxygenates. Journal of the American

Chemical Society 139, 14120-14128, doi:10.1021/jacs.7b06331 (2017).

93 Yuan, X., Lee, K., Bender, M. T., Schmidt, J. R. & Choi, K.-S. Mechanistic Differences

between Electrochemical Hydrogenation and Hydrogenolysis of 5-Hydroxymethylfurfural and

Their pH Dependence. ChemSusChem 15, 202200952,

doi:https://doi.org/10.1002/cssc.202200952 (2022).

94 Li, Z., Kelkar, S., Lam, C. H., Luczek, K., Jackson, J. E., Miller, D. J. & Saffron, C. M.

Aqueous electrocatalytic hydrogenation of furfural using a sacrificial anode. Electrochimica

Acta 64, 87-93, doi:https://doi.org/10.1016/j.electacta.2011.12.105 (2012).

95 Kwon, Y., Birdja, Y. Y., Raoufmoghaddam, S. & Koper, M. T. M. Electrocatalytic

Hydrogenation of 5-Hydroxymethylfurfural in Acidic Solution. ChemSusChem 8, 1745-1751,

doi:10.1002/cssc.201500176 (2015).

96 Sanghez de Luna, G., Ho, P. H., Sacco, A., Hernández, S., Velasco-Vélez, J.-J., Ospitali, F.,

Paglianti, A., Albonetti, S., Fornasari, G. & Benito, P. AgCu Bimetallic Electrocatalysts for

the Reduction of Biomass-Derived Compounds. ACS Applied Materials & Interfaces 13,

23675-23688, doi:10.1021/acsami.1c02896 (2021).

97 Zhang, L., Zhang, F., Michel Jr., F. C. & Co, A. C. Efficient Electrochemical Hydrogenation

of 5-Hydroxymethylfurfural to 2,5-Bis(hydroxymethyl)furan on Ag-Displaced Nanotextured

Cu Catalysts. 6, 4739-4749, doi:https://doi.org/10.1002/celc.201900640 (2019).

98 Ji, K., Xu, M., Xu, S.-M., Wang, Y., Ge, R., Hu, X., Sun, X. & Duan, H. Electrocatalytic

Hydrogenation of 5-Hydroxymethylfurfural Promoted by a Ru1Cu Single-Atom Alloy

Catalyst. Angewandte Chemie International Edition 61, 202209849,

doi:https://doi.org/10.1002/anie.202209849 (2022).

99 Zhang, Z., Huang, K., Qiu, X., Ge, W., Yang, X., Zhu, Y., Lian, C., Liu, H., Jiang, H. & Li, C.

Operando Generated Copper-based Catalyst Enabling Efficient Electrosynthesis of 2,5-

Bis(hydroxymethyl)furan. Fundamental Research,

doi:https://doi.org/10.1016/j.fmre.2022.01.016 (2022).

100 Chadderdon, X. H., Chadderdon, D. J., Pfennig, T., Shanks, B. H. & Li, W. Paired

electrocatalytic hydrogenation and oxidation of 5-(hydroxymethyl)furfural for efficient

production of biomass-derived monomers. Green Chemistry 21, 6210-6219,

doi:10.1039/C9GC02264C (2019).

101 Liu, H., Lee, T.-H., Chen, Y., Cochran, E. W. & Li, W. Paired electrolysis of 5-

(hydroxymethyl)furfural in flow cells with a high-performance oxide-derived silver cathode.

Green Chemistry 23, 5056-5063, doi:10.1039/D1GC00988E (2021).

References

118

102 Nilges, P. & Schröder, U. Electrochemistry for biofuel generation: production of furans by

electrocatalytic hydrogenation of furfurals. Energy & Environmental Science 6, 2925-2931,

doi:10.1039/C3EE41857J (2013).

103 Zhang, Y.-R., Wang, B.-X., Qin, L., Li, Q. & Fan, Y.-M. A non-noble bimetallic alloy in the

highly selective electrochemical synthesis of the biofuel 2,5-dimethylfuran from 5-

hydroxymethylfurfural. Green Chemistry 21, 1108-1113, doi:10.1039/C8GC03689F (2019).

104 Roylance, J. J. & Choi, K.-S. Electrochemical reductive biomass conversion: direct

conversion of 5-hydroxymethylfurfural (HMF) to 2,5-hexanedione (HD) via reductive ring-

opening. Green Chemistry 18, 2956-2960, doi:10.1039/C6GC00533K (2016).

105 Kloth, R., Vasilyev, D. V., Mayrhofer, K. J. J. & Katsounaros, I. Electroreductive 5-

Hydroxymethylfurfural Dimerization on Carbon Electrodes. ChemSusChem 14, 5245-5253,

doi:https://doi.org/10.1002/cssc.202101575 (2021).

106 Fuller, T. F. & Harb, J. N. in Electrochemical Engeneering, Ch. 1, pp. 1-14 (Wiley, 2018).

107 Fuller, T. F. & Harb, J. N. in Electrochemical Engeneering, Ch. 3, pp. 41-62 (Wiley, 2018).

108 Bard, A. J., Faulkner, L. R. & White, H. S. in Electrochemical Methods: Fundamentals and

Applications, Ch. 1, pp. 1-60 (Wiley, 2022).

109 Barbir, F. in PEM Fuel Cells, Ch. 2, pp. 17-32 (Elsevier, 2013).

110 Millet, P. in Hydrogen Production: by Electrolysis, (ed A. Godula-Jopek) Ch. 2, pp. 33-62

(Wiley, 2015).

111 Pletscher, D. & Walsh, F. C. in Industrial Electrochemistry, Ch. 1, pp. 1-58 (Kluwer, 1990).

112 Atkins, P. W. & Paula, J. d. in Physikalische Chemie, Ch. 22, pp. 917-918 (Wiley, 2013).

113 Atkins, P. W. & Paula, J. d. in Physikalische Chemie, Ch. 2, pp. 45-96 (Wiley, 2013).

114 Atkins, P. W. & Paula, J. d. in Physikalische Chemie, Ch. 3, pp. 97-140 (Wiley, 2013).

115 Atobe, M. in Fundamental Principles of Organic Electrochemistry: Fundamental Aspects of

Electrochemistry Dealing with Organic Molecules, Ch. 1, pp. 1-10 (Wiley, 2014).

116 Chorkendorff, I. & Niemantsverdriet, J. W. in Concepts of Modern Catalysis and Kinetics, Ch.

6, pp. 215-266 (Wiley, 2003).

117 Helmholtz, H. Studien über electrische Grenzschichten. Annalen der Physik 243, 337-382,

doi:https://doi.org/10.1002/andp.18792430702 (1879).

118 Stern, O. ZUR THEORIE DER ELEKTROLYTISCHEN DOPPELSCHICHT. Zeitschrift für

Elektrochemie und angewandte physikalische Chemie 30, 508-516,

doi:https://doi.org/10.1002/bbpc.192400182 (1924).

119 Grahame, D. C. The electrical double layer and the theory of electrocapillarity. Chemical

reviews 41, 441-501 (1947).

120 Wu, J. Understanding the Electric Double-Layer Structure, Capacitance, and Charging

Dynamics. Chemical Reviews 122, 10821-10859, doi:10.1021/acs.chemrev.2c00097 (2022).

121 Harrison, K. W., Remick, R., Hoskin, A. & Martin, G. D. (Sponsor Org.: USDOE).

122 Sabatier, P. Hydrogénations et déshydrogénations par catalyse. Berichte der deutschen

chemischen Gesellschaft 44, 1984-2001, doi:https://doi.org/10.1002/cber.19110440303

(1911).

123 Fuchigami, T. in Fundamentals and Applications of Organic Electrochemistry, pp. I-XII

(Wiley, 2014).

124 Baizer, M. M. Electrolytic Reductive Coupling: I . Acrylonitrile. Journal of The

Electrochemical Society 111, 215, doi:10.1149/1.2426086 (1964).

125 Cardoso, D. S. P., Šljukić, B., Santos, D. M. F. & Sequeira, C. A. C. Organic Electrosynthesis:

From Laboratorial Practice to Industrial Applications. Organic Process Research &

Development 21, 1213-1226, doi:10.1021/acs.oprd.7b00004 (2017).

126 Dresp, S., Luo, F., Schmack, R., Kühl, S., Gliech, M. & Strasser, P. An efficient bifunctional

two-component catalyst for oxygen reduction and oxygen evolution in reversible fuel cells,

electrolyzers and rechargeable air electrodes. Energy & Environmental Science 9, 2020-2024,

doi:10.1039/C6EE01046F (2016).

References

119

127 Klingenhof, M., Hauke, P., Kroschel, M., Wang, X., Merzdorf, T., Binninger, C., Ngo Thanh,

T., Paul, B., Teschner, D., Schlögl, R. & Strasser, P. Anion-Tuned Layered Double Hydroxide

Anodes for Anion Exchange Membrane Water Electrolyzers: From Catalyst Screening to

Single-Cell Performance. ACS Energy Letters 7, 3415-3422,

doi:10.1021/acsenergylett.2c01820 (2022).

128 Ramachandran, R. & Menon, R. K. An Overview of Industrial Uses of Hydrogen.

International Journal of Hydrogen Energy 23, 593-598 (1998).

129 Ball, M. & Wietschel, M. The future of hydrogen – opportunities and challenges☆.

International Journal of Hydrogen Energy 34, 615-627, doi:10.1016/j.ijhydene.2008.11.014

(2009).

130 Kakoulaki, G., Kougias, I., Taylor, N., Dolci, F., Moya, J. & Jäger-Waldau, A. Green

hydrogen in Europe – A regional assessment: Substituting existing production with

electrolysis powered by renewables. Energy Conversion and Management 228, 113649,

doi:10.1016/j.enconman.2020.113649 (2021).

131 Kou, T., Wang, S. & Li, Y. Perspective on High-Rate Alkaline Water Splitting. ACS Materials

Letters 3, 224-234, doi:10.1021/acsmaterialslett.0c00536 (2021).

132 Noussan, M., Raimondi, P. P., Scita, R. & Hafner, M. The Role of Green and Blue Hydrogen

in the Energy Transition—A Technological and Geopolitical Perspective. Sustainability 13,

298, doi:10.3390/su13010298 (2020).

133 Burton, N. A., Padilla, R. V., Rose, A. & Habibullah, H. Increasing the efficiency of hydrogen

production from solar powered water electrolysis. Renewable and Sustainable Energy Reviews

135, 110255, doi:10.1016/j.rser.2020.110255 (2021).

134 Schalenbach, M. A Perspective on Low-Temperature Water Electrolysis – Challenges in

Alkaline and Acidic Technology. International Journal of Electrochemical Science, 1173-

1226, doi:10.20964/2018.02.26 (2018).

135 Latsuzbaia, R., Bisselink, R., Anastasopol, A., van der Meer, H., van Heck, R., Yagüe, M. S.,

Zijlstra, M., Roelands, M., Crockatt, M., Goetheer, E. & Giling, E. Continuous

electrochemical oxidation of biomass derived 5-(hydroxymethyl)furfural into 2,5-

furandicarboxylic acid. Journal of Applied Electrochemistry 48, 611-626,

doi:10.1007/s10800-018-1157-7 (2018).

136 Mukerjee, S., Yan, Y. & Xu, H. Hydrogen at Scale Using Low-Temperature Anion Exchange

Membrane Electrolyzers. Electrochem. Soc. Interface 30, 73 (2021).

137 Henkensmeier, D., Najibah, M., Harms, C., Zitka, J., Hnat, J. & Bouzek, K. Overview: State-

of-the Art Commercial Membranes for Anion Exchange Membrane Water Electrolysis. J. of

Echem. Energy Conversion and Storage 18, 024001, doi:10.1115/1.4047963] (2020).

138 Fortin, P., Khoza, T., Cao, X., Martinsen, S. Y., Oyarce Barnett, A. & Holdcroft, S. High-

performance alkaline water electrolysis using Aemion™ anion exchange membranes. Journal

of Power Sources 451, 227814, doi:10.1016/j.jpowsour.2020.227814 (2020).

139 Pushkareva, I. V., Pushkarev, A. S., Grigoriev, S. A., Modisha, P. & Bessarabov, D. G.

Comparative study of anion exchange membranes for low-cost water electrolysis.

International Journal of Hydrogen Energy 45, 26070-26079,

doi:10.1016/j.ijhydene.2019.11.011 (2020).

140 Motealleh, B., Liu, Z., Masel, R. I., Sculley, J. P., Richard Ni, Z. & Meroueh, L. Next-

generation anion exchange membrane water electrolyzers operating for commercially relevant

lifetimes. International Journal of Hydrogen Energy 46, 3379-3386,

doi:10.1016/j.ijhydene.2020.10.244 (2021).

141 Zhu, Y., Zhou, W., Zhong, Y., Bu, Y., Chen, X., Zhong, Q., Liu, M. & Shao, Z. A Perovskite

Nanorod as Bifunctional Electrocatalyst for Overall Water Splitting. Advanced Energy

Materials 7, doi:10.1002/aenm.201602122 (2017).

142 Zhao, J. W., Shi, Z. X., Li, C. F., Gu, L. F. & Li, G. R. Boosting the electrocatalytic

performance of NiFe layered double hydroxides for the oxygen evolution reaction by exposing

the highly active edge plane (012). Chem Sci 12, 650-659, doi:10.1039/d0sc04196c (2020).

References

120

143 Zhang, L., Lu, C., Ye, F., Pang, R., Liu, Y., Wu, Z., Shao, Z., Sun, Z. & Hu, L. Selenic Acid

Etching Assisted Vacancy Engineering for Designing Highly Active Electrocatalysts toward

the Oxygen Evolution Reaction. Adv Mater 33, e2007523, doi:10.1002/adma.202007523

(2021).

144 Wu, C., Zhang, X., Li, H., Xia, Z., Yu, S., Wang, S. & Sun, G. Iron-based binary metal-

organic framework nanorods as an efficient catalyst for the oxygen evolution reaction.

Chinese Journal of Catalysis 42, 637-647, doi:10.1016/s1872-2067(20)63686-5 (2021).

145 Gorlin, M., Ferreira de Araujo, J., Schmies, H., Bernsmeier, D., Dresp, S., Gliech, M., Jusys,

Z., Chernev, P., Kraehnert, R., Dau, H. & Strasser, P. Tracking Catalyst Redox States and

Reaction Dynamics in Ni-Fe Oxyhydroxide Oxygen Evolution Reaction Electrocatalysts: The

Role of Catalyst Support and Electrolyte pH. J Am Chem Soc 139, 2070-2082,

doi:10.1021/jacs.6b12250 (2017).

146 Xu, D., Stevens, M. B., Cosby, M. R., Oener, S. Z., Smith, A. M., Enman, L. J., Ayers, K. E.,

Capuano, C. B., Renner, J. N., Danilovic, N., Li, Y., Wang, H., Zhang, Q. & Boettcher, S. W.

Earth-Abundant Oxygen Electrocatalysts for Alkaline Anion-Exchange-Membrane Water

Electrolysis: Effects of Catalyst Conductivity and Comparison with Performance in Three-

Electrode Cells. ACS Catalysis 9, 7-15, doi:10.1021/acscatal.8b04001 (2018).

147 Klingenhof, M., Hauke, P., Brückner, S., Dresp, S., Wolf, E., Nong, H. N., Spöri, C.,

Merzdorf, T., Bernsmeier, D., Teschner, D., Schlögl, R. & Strasser, P. Modular Design of

Highly Active Unitized Reversible Fuel Cell Electrocatalysts. ACS Energy Letters 6, 177-183,

doi:10.1021/acsenergylett.0c02203 (2021).

148 Watzele, S. & Bandarenka, A. S. Quick Determination of Electroactive Surface Area of Some

Oxide Electrode Materials. Electroanalysis 28, 2394-2399, doi:10.1002/elan.201600178

(2016).

149 Dionigi, F., Zhu, J., Zeng, Z., Merzdorf, T., Sarodnik, H., Gliech, M., Pan, L., Li, W. X.,

Greeley, J. & Strasser, P. Intrinsic Electrocatalytic Activity for Oxygen Evolution of

Crystalline 3d-Transition Metal Layered Double Hydroxides. Angew Chem Int Ed Engl 60,

14446-14457, doi:10.1002/anie.202100631 (2021).

150 Vincent, I., Lee, E. C. & Kim, H. M. Comprehensive impedance investigation of low-cost

anion exchange membrane electrolysis for large-scale hydrogen production. Sci Rep 11, 293,

doi:10.1038/s41598-020-80683-6 (2021).

151 Watzele, S., Hauenstein, P., Liang, Y., Xue, S., Fichtner, J., Garlyyev, B., Scieszka, D.,

Claudel, F., Maillard, F. & Bandarenka, A. S. Determination of Electroactive Surface Area of

Ni-, Co-, Fe-, and Ir-Based Oxide Electrocatalysts. ACS Catalysis 9, 9222-9230,

doi:10.1021/acscatal.9b02006 (2019).

152 Görlin, M., Chernev, P., Ferreira de Araújo, J., Reier, T., Dresp, S., Paul, B., Krähnert, R.,

Dau, H. & Strasser, P. Oxygen Evolution Reaction Dynamics, Faradaic Charge Efficiency,

and the Active Metal Redox States of Ni–Fe Oxide Water Splitting Electrocatalysts. Journal

of the American Chemical Society 138, 5603-5614, doi:10.1021/jacs.6b00332 (2016).

153 Louie, M. W. & Bell, A. T. An Investigation of Thin-Film Ni–Fe Oxide Catalysts for the

Electrochemical Evolution of Oxygen. Journal of the American Chemical Society 135, 12329-

12337, doi:10.1021/ja405351s (2013).

154 Dresp, S., Dionigi, F., Klingenhof, M., Merzdorf, T., Schmies, H., Drnec, J., Poulain, A. &

Strasser, P. Molecular Understanding of the Impact of Saline Contaminants and Alkaline pH

on NiFe Layered Double Hydroxide Oxygen Evolution Catalysts. ACS Catal.,

https://doi.org/10.1021/acscatal.1021c00773 (2021).

155 Hauke, P., Klingenhof, M., Wang, X., de Araújo, J. F. & Strasser, P. Efficient electrolysis of

5-hydroxymethylfurfural to the biopolymer-precursor furandicarboxylic acid in a zero-gap

MEA-type electrolyzer. Cell Reports Physical Science 2, 100650,

doi:https://doi.org/10.1016/j.xcrp.2021.100650 (2021).

156 Deng, J., Nellist, M. R., Stevens, M. B., Dette, C., Wang, Y. & Boettcher, S. W. Morphology

Dynamics of Single-Layered Ni(OH)2/NiOOH Nanosheets and Subsequent Fe Incorporation

References

121

Studied by in Situ Electrochemical Atomic Force Microscopy. Nano Lett 17, 6922-6926,

doi:10.1021/acs.nanolett.7b03313 (2017).

157 Klaus, S., Cai, Y., Louie, M. W., Trotochaud, L. & Bell, A. T. Effects of Fe Electrolyte

Impurities on Ni(OH)2/NiOOH Structure and Oxygen Evolution Activity. The Journal of

Physical Chemistry C 119, 7243-7254, doi:10.1021/acs.jpcc.5b00105 (2015).

158 Son, Y. J., Kawashima, K., Wygant, B. R., Lam, C. H., Burrow, J. N., Celio, H., Dolocan, A.,

Ekerdt, J. G. & Mullins, C. B. Anodized Nickel Foam for Oxygen Evolution Reaction in Fe-

Free and Unpurified Alkaline Electrolytes at High Current Densities. ACS Nano,

doi:10.1021/acsnano.0c10788 (2021).

159 Trotochaud, L., Young, S. L., Ranney, J. K. & Boettcher, S. W. Nickel-iron oxyhydroxide

oxygen-evolution electrocatalysts: the role of intentional and incidental iron incorporation. J

Am Chem Soc 136, 6744-6753, doi:10.1021/ja502379c (2014).

160 Hegge, F., Lombeck, F., Cruz Ortiz, E., Bohn, L., von Holst, M., Kroschel, M., Hübner, J.,

Breitwieser, M., Strasser, P. & Vierrath, S. Efficient and Stable Low Iridium Loaded Anodes

for PEM Water Electrolysis Made Possible by Nanofiber Interlayers. ACS Applied Energy

Materials 3, 8276-8284, doi:10.1021/acsaem.0c00735 (2020).

161 PlasticsEurope. Plastics - the Facts 2020. (2020).

<https://www.plasticseurope.org/en/resources/publications/4312-plastics-facts-2020>.

162 Dormer, A., Finn, D. P., Ward, P. & Cullen, J. Carbon footprint analysis in plastics

manufacturing. Journal of Cleaner Production 51, 133-141,

doi:https://doi.org/10.1016/j.jclepro.2013.01.014 (2013).

163 Liu, B., Ren, Y. & Zhang, Z. Aerobic oxidation of 5-hydroxymethylfurfural into 2,5-

furandicarboxylic acid in water under mild conditions. Green Chemistry 17, 1610-1617,

doi:10.1039/C4GC02019G (2015).

164 Wan, X., Zhou, C., Chen, J., Deng, W., Zhang, Q., Yang, Y. & Wang, Y. Base-Free Aerobic

Oxidation of 5-Hydroxymethyl-furfural to 2,5-Furandicarboxylic Acid in Water Catalyzed by

Functionalized Carbon Nanotube-Supported Au–Pd Alloy Nanoparticles. ACS Catalysis 4,

2175-2185, doi:10.1021/cs5003096 (2014).

165 Wang, Y., Yu, K., Lei, D., Si, W., Feng, Y., Lou, L.-L. & Liu, S. Basicity-Tuned

Hydrotalcite-Supported Pd Catalysts for Aerobic Oxidation of 5-Hydroxymethyl-2-furfural

under Mild Conditions. ACS Sustainable Chemistry & Engineering 4, 4752-4761,

doi:10.1021/acssuschemeng.6b00965 (2016).

166 Lei, D., Yu, K., Li, M.-R., Wang, Y., Wang, Q., Liu, T., Liu, P., Lou, L.-L., Wang, G. & Liu,

S. Facet Effect of Single-Crystalline Pd Nanocrystals for Aerobic Oxidation of 5-

Hydroxymethyl-2-furfural. ACS Catalysis 7, 421-432, doi:10.1021/acscatal.6b02839 (2017).

167 Jiang, N., You, B., Boonstra, R., Terrero Rodriguez, I. M. & Sun, Y. Integrating

Electrocatalytic 5-Hydroxymethylfurfural Oxidation and Hydrogen Production via Co–P-

Derived Electrocatalysts. ACS Energy Letters 1, 386-390, doi:10.1021/acsenergylett.6b00214

(2016).

168 Kwon, Y., Schouten, K. J. P., van der Waal, J. C., de Jong, E. & Koper, M. T. M.

Electrocatalytic Conversion of Furanic Compounds. ACS Catalysis 6, 6704-6717,

doi:10.1021/acscatal.6b01861 (2016).

169 Xie, Y., Zhou, Z., Yang, N. & Zhao, G. An Overall Reaction Integrated with Highly Selective

Oxidation of 5-Hydroxymethylfurfural and Efficient Hydrogen Evolution. Advanced

Functional Materials 31, 2102886, doi:https://doi.org/10.1002/adfm.202102886.

170 Lu, Y., Dong, C.-L., Huang, Y.-C., Zou, Y., Liu, Y., Li, Y., Zhang, N., Chen, W., Zhou, L.,

Lin, H. & Wang, S. Hierarchically nanostructured NiO-Co<sub>3</sub>O<sub>4</sub> with

rich interface defects for the electro-oxidation of 5-hydroxymethylfurfural. SCIENCE CHINA

Chemistry 63, 980, doi:https://doi.org/10.1007/s11426-020-9749-8 (2020).

171 Le, K., Wang, Z., Wang, F., Wang, Q., Shao, Q., Murugadoss, V., Wu, S., Liu, W., Liu, J.,

Gao, Q. & Guo, Z. Sandwich-like NiCo layered double hydroxide/reduced graphene oxide

References

122

nanocomposite cathodes for high energy density asymmetric supercapacitors. Dalton

Transactions 48, 5193-5202, doi:10.1039/C9DT00615J (2019).

172 Zhao, J., Chen, J., Xu, S., Shao, M., Zhang, Q., Wei, F., Ma, J., Wei, M., Evans, D. G. &

Duan, X. Hierarchical NiMn Layered Double Hydroxide/Carbon Nanotubes Architecture with

Superb Energy Density for Flexible Supercapacitors. Advanced Functional Materials 24,

2938-2946, doi:https://doi.org/10.1002/adfm.201303638 (2014).

173 Wang, D., Li, Q., Han, C., Lu, Q., Xing, Z. & Yang, X. Atomic and electronic modulation of

self-supported nickel-vanadium layered double hydroxide to accelerate water splitting

kinetics. Nature Communications 10, 3899, doi:10.1038/s41467-019-11765-x (2019).

174 Henkensmeier, D., Najibah, M., Harms, C., Žitka, J., Hnát, J. & Bouzek, K. Overview: State-

of-the Art Commercial Membranes for Anion Exchange Membrane Water Electrolysis.

Journal of Electrochemical Energy Conversion and Storage 18, doi:10.1115/1.4047963

(2020).

175 Görlin, M., Halldin Stenlid, J., Koroidov, S., Wang, H.-Y., Börner, M., Shipilin, M., Kalinko,

A., Murzin, V., Safonova, O. V., Nachtegaal, M., Uheida, A., Dutta, J., Bauer, M., Nilsson, A.

& Diaz-Morales, O. Key activity descriptors of nickel-iron oxygen evolution electrocatalysts

in the presence of alkali metal cations. Nature Communications 11, 6181,

doi:10.1038/s41467-020-19729-2 (2020).

176 Hunter, B. M., Hieringer, W., Winkler, J. R., Gray, H. B. & Müller, A. M. Effect of interlayer

anions on [NiFe]-LDH nanosheet water oxidation activity. Energy & Environmental Science

9, 1734-1743, doi:10.1039/c6ee00377j (2016).

177 Song, F. & Hu, X. Exfoliation of layered double hydroxides for enhanced oxygen evolution

catalysis. Nature Communications 5, 4477, doi:10.1038/ncomms5477 (2014).

178 Deng, X., Kang, X., Li, M., Xiang, K., Wang, C., Guo, Z., Zhang, J., Fu, X.-Z. & Luo, J.-L.

Coupling efficient biomass upgrading with H2 production via bifunctional CuxS@NiCo-LDH

core–shell nanoarray electrocatalysts. Journal of Materials Chemistry A 8, 1138-1146,

doi:10.1039/C9TA06917H (2020).

179 Kong, F. & Wang, M. Preparation of Sulfur-Modulated Nickel/Carbon Composites from

Lignosulfonate for the Electrocatalytic Oxidation of 5-Hydroxymethylfurfural to 2,5-

Furandicarboxylic Acid. ACS Applied Energy Materials, doi:10.1021/acsaem.0c02418 (2021).

180 Zhang, J., Gong, W., Yin, H., Wang, D., Zhang, Y., Zhang, H., Wang, G. & Zhao, H. In-situ

Growth of Ultrathin Ni(OH)2 Nanosheets Catalyst for Electrocatalytic Oxidation Reactions.

ChemSusChem 14, 2935-2942, doi:https://doi.org/10.1002/cssc.202100811 (2021).

181 Song, X., Liu, X., Wang, H., Guo, Y. & Wang, Y. Improved Performance of Nickel Boride by

Phosphorus Doping as an Efficient Electrocatalyst for the Oxidation of 5-

Hydroxymethylfurfural to 2,5-Furandicarboxylic Acid. Industrial & Engineering Chemistry

Research 59, 17348-17356, doi:10.1021/acs.iecr.0c01312 (2020).

182 van Putten, R.-J., van der Waal, J. C., de Jong, E., Rasrendra, C. B., Heeres, H. J. & de Vries,

J. G. Hydroxymethylfurfural, A Versatile Platform Chemical Made from Renewable

Resources. Chemical Reviews 113, 1499-1597, doi:10.1021/cr300182k (2013).

183 Bender, M. T., Yuan, X., Goetz, M. K. & Choi, K.-S. Electrochemical Hydrogenation,

Hydrogenolysis, and Dehydrogenation for Reductive and Oxidative Biomass Upgrading Using

5-Hydroxymethylfurfural as a Model System. ACS Catalysis 12, 12349-12368,

doi:10.1021/acscatal.2c03606 (2022).

184 Mistry, H., Varela, A. S., Kühl, S., Strasser, P. & Cuenya, B. R. Nanostructured

electrocatalysts with tunable activity and selectivity. Nature Reviews Materials 1, 16009,

doi:10.1038/natrevmats.2016.9 (2016).

185 Tahira, A., Ibupoto, Z. H., Willander, M. & Nur, O. Advanced Co3O4–CuO nano-composite

based electrocatalyst for efficient hydrogen evolution reaction in alkaline media. International

Journal of Hydrogen Energy 44, 26148-26157,

doi:https://doi.org/10.1016/j.ijhydene.2019.08.120 (2019).

References

123

186 Wang, X., Klingan, K., Klingenhof, M., Möller, T., Ferreira de Araújo, J., Martens, I., Bagger,

A., Jiang, S., Rossmeisl, J., Dau, H. & Strasser, P. Morphology and mechanism of highly

selective Cu(II) oxide nanosheet catalysts for carbon dioxide electroreduction. Nature

Communications 12, 794, doi:10.1038/s41467-021-20961-7 (2021).

187 Elsayed, I., Jackson, M. A. & Hassan, E. B. Hydrogen-Free Catalytic Reduction of Biomass-

Derived 5-Hydroxymethylfurfural into 2,5-Bis(hydroxymethyl)furan Using Copper–Iron

Oxides Bimetallic Nanocatalyst. ACS Sustainable Chemistry & Engineering 8, 1774-1785,

doi:10.1021/acssuschemeng.9b05575 (2020).

188 Wang, X., Hou, X., Lee, H., Lu, L., Wu, X. & Sun, L. Copper Selenide–Derived Copper

Oxide Nanoplates as a Durable and Efficient Electrocatalyst for Oxygen Evolution Reaction.

Energy Technology 8, 2000142, doi:https://doi.org/10.1002/ente.202000142 (2020).

189 Wei, X., Li, Y., Chen, L. & Shi, J. Formic Acid Electro-Synthesis by Concurrent Cathodic

CO2 Reduction and Anodic CH3OH Oxidation. Angewandte Chemie International Edition 60,

3148-3155, doi:https://doi.org/10.1002/anie.202012066 (2021).

190 Kotta, A., Kim, E.-B., Ameen, S., Shin, H.-S. & Seo, H. K. Communication—Ultra-Small

NiO Nanoparticles Grown by Low-Temperature Process for Electrochemical Application.

Journal of The Electrochemical Society 167, 167517, doi:10.1149/1945-7111/abcf51 (2020).

191 Li, R., Hu, B., Yu, T., Chen, H., Wang, Y. & Song, S. Insights into Correlation among

Surface-Structure-Activity of Cobalt-Derived Pre-Catalyst for Oxygen Evolution Reaction.

Advanced Science 7, 1902830, doi:https://doi.org/10.1002/advs.201902830 (2020).

192 Cong, Y., Chen, M., Xu, T., Zhang, Y. & Wang, Q. Tantalum and aluminum co-doped iron

oxide as a robust photocatalyst for water oxidation. Applied Catalysis B: Environmental 147,

733-740, doi:https://doi.org/10.1016/j.apcatb.2013.10.009 (2014).

193 Biesinger, M. C., Lau, L. W. M., Gerson, A. R. & Smart, R. S. C. The role of the Auger

parameter in XPS studies of nickel metal, halides and oxides. Physical Chemistry Chemical

Physics 14, 2434-2442, doi:10.1039/C2CP22419D (2012).

194 Biesinger, M. C., Payne, B. P., Grosvenor, A. P., Lau, L. W. M., Gerson, A. R. & Smart, R. S.

C. Resolving surface chemical states in XPS analysis of first row transition metals, oxides and

hydroxides: Cr, Mn, Fe, Co and Ni. Applied Surface Science 257, 2717-2730,

doi:https://doi.org/10.1016/j.apsusc.2010.10.051 (2011).

195 Komath, M., Thomas, S., Cherian, K. A. & Ray, A. Preparation of gamma-nickel

oxyhydroxide through high temperature reaction of nickel with anhydrous sodium hydroxide.

Materials Chemistry and Physics 36, 190-193, doi:https://doi.org/10.1016/0254-

0584(93)90032-H (1993).

196 Liu, L., Zhou, Z. & Peng, C. Sonochemical intercalation synthesis of nano γ-nickel

oxyhydroxide: Structure and electrochemical properties. Electrochimica Acta 54, 434-441,

doi:https://doi.org/10.1016/j.electacta.2008.07.055 (2008).

197 Wang, X. L., Klingan, K., Klingenhof, M., Moller, T., de Araujo, J. F., Martens, I., Bagger,

A., Jiang, S., Rossmeisl, J., Dau, H. & Strasser, P. Morphology and mechanism of highly

selective Cu(II) oxide nanosheet catalysts for carbon dioxide electroreduction. Nature

Commun. 12, 794 (2021).

198 Möller, T., Scholten, F., Thanh, T. N., Sinev, I., Timoshenko, J., Wang, X., Jovanov, Z.,

Gliech, M., Roldan Cuenya, B., Varela, A. S. & Strasser, P. Electrocatalytic CO2 Reduction

on CuOx Nanocubes: Tracking the Evolution of Chemical State, Geometric Structure, and

Catalytic Selectivity using Operando Spectroscopy. Angewandte Chemie International Edition

59, 17974-17983, doi:https://doi.org/10.1002/anie.202007136 (2020).

199 Timoshenko, J. & Roldan Cuenya, B. In Situ/Operando Electrocatalyst Characterization by X-

ray Absorption Spectroscopy. Chemical Reviews 121, 882-961,

doi:10.1021/acs.chemrev.0c00396 (2021).

200 Ciesielczyk, F., Bartczak, P., Wieszczycka, K., Siwińska-Stefańska, K., Nowacka, M. &

Jesionowski, T. Adsorption of Ni(II) from model solutions using co-precipitated inorganic

oxides. Adsorption 19, 423-434, doi:10.1007/s10450-012-9464-5 (2013).

References

124

201 Xie, R.-C., Batchelor-McAuley, C., Rauwel, E., Rauwel, P. & Compton, R. G.

Electrochemical Characterisation of Co@Co(OH)2 Core-Shell Nanoparticles and their

Aggregation in Solution. ChemElectroChem 7, 4259-4268,

doi:https://doi.org/10.1002/celc.202001199 (2020).

202 Celante, V. G. & Freitas, M. B. J. G. Electrodeposition of copper from spent Li-ion batteries

by electrochemical quartz crystal microbalance and impedance spectroscopy techniques.

Journal of Applied Electrochemistry 40, 233-239, doi:10.1007/s10800-009-9996-x (2010).

203 Ning, J., Zheng, Y., Brown, B., Young, D. & Nešić, S. A Thermodynamic Model for the

Prediction of Mild Steel Corrosion Products in an Aqueous Hydrogen Sulfide Environment.

Corrosion 71, 945-960, doi:10.5006/1566 (2015).

204 Foo Wong, Y., Makahleh, A., Al Azzam, K. M., Yahaya, N., Saad, B. & Amrah Sulaiman, S.

Micellar electrokinetic chromatography method for the simultaneous determination of furanic

compounds in honey and vegetable oils. Talanta 97, 23-31,

doi:https://doi.org/10.1016/j.talanta.2012.03.056 (2012).

205 Rizelio, V. M., Gonzaga, L. V., da Silva Campelo Borges, G., Micke, G. A., Fett, R. & Costa,

A. C. O. Development of a fast MECK method for determination of 5-HMF in honey samples.

Food Chemistry 133, 1640-1645, doi:https://doi.org/10.1016/j.foodchem.2011.11.058 (2012).

206 Hauke, P., Brückner, S. & Strasser, P. Paired Electrocatalytic Valorization of CO2 and

Hydroxymethylfurfural in a Noble Metal-free Bipolar Membrane Electrolyzer. ACS

Sustainable Chemistry & Engineering, doi:10.1021/acssuschemeng.3c03144 (2023).

207 Bagger, A., Ju, W., Varela, A. S., Strasser, P. & Rossmeisl, J. Electrochemical CO2

Reduction: A Classification Problem. ChemPhysChem 18, 3266-3273,

doi:https://doi.org/10.1002/cphc.201700736 (2017).

208 Endrődi, B., Kecsenovity, E., Samu, A., Halmágyi, T., Rojas-Carbonell, S., Wang, L., Yan, Y.

& Janáky, C. High carbonate ion conductance of a robust PiperION membrane allows

industrial current density and conversion in a zero-gap carbon dioxide electrolyzer cell.

Energy & Environmental Science 13, 4098-4105, doi:10.1039/d0ee02589e (2020).

209 Samu, A. A., Kormanyos, A., Kecsenovity, E., Szilagyi, N., Endrodi, B. & Janaky, C.

Intermittent Operation of CO(2) Electrolyzers at Industrially Relevant Current Densities. ACS

Energy Lett 7, 1859-1861, doi:10.1021/acsenergylett.2c00923 (2022).

210 Dinh, C.-T., García de Arquer, F. P., Sinton, D. & Sargent, E. H. High Rate, Selective, and

Stable Electroreduction of CO2 to CO in Basic and Neutral Media. ACS Energy Letters 3,

2835-2840, doi:10.1021/acsenergylett.8b01734 (2018).

211 Vass, A., Kormanyos, A., Koszo, Z., Endrodi, B. & Janaky, C. Anode Catalysts in CO(2)

Electrolysis: Challenges and Untapped Opportunities. ACS Catal 12, 1037-1051,

doi:10.1021/acscatal.1c04978 (2022).

212 Baumgartner, L. M., Koopman, C. I., Forner-Cuenca, A. & Vermaas, D. A. When Flooding Is

Not Catastrophic─Woven Gas Diffusion Electrodes Enable Stable CO2 Electrolysis. ACS

Applied Energy Materials 5, 15125-15135, doi:10.1021/acsaem.2c02783 (2022).

213 Yang, K., Kas, R., Smith, W. A. & Burdyny, T. Role of the Carbon-Based Gas Diffusion

Layer on Flooding in a Gas Diffusion Electrode Cell for Electrochemical CO2 Reduction.

ACS Energy Letters 6, 33-40, doi:10.1021/acsenergylett.0c02184 (2021).

214 Sassenburg, M., Kelly, M., Subramanian, S., Smith, W. A. & Burdyny, T. Zero-Gap

Electrochemical CO2 Reduction Cells: Challenges and Operational Strategies for Prevention

of Salt Precipitation. ACS Energy Letters 8, 321-331, doi:10.1021/acsenergylett.2c01885

(2023).

215 Salvatore, D. A., Weekes, D. M., He, J., Dettelbach, K. E., Li, Y. C., Mallouk, T. E. &

Berlinguette, C. P. Electrolysis of Gaseous CO2 to CO in a Flow Cell with a Bipolar

Membrane. ACS Energy Letters 3, 149-154, doi:10.1021/acsenergylett.7b01017 (2017).

216 Weekes, D. M., Salvatore, D. A., Reyes, A., Huang, A. & Berlinguette, C. P. Electrolytic

CO(2) Reduction in a Flow Cell. Acc Chem Res 51, 910-918,

doi:10.1021/acs.accounts.8b00010 (2018).

References

125

217 Blommaert, M. A., Subramanian, S., Yang, K., Smith, W. A. & Vermaas, D. A. High Indirect

Energy Consumption in AEM-Based CO(2) Electrolyzers Demonstrates the Potential of

Bipolar Membranes. ACS Appl Mater Interfaces 14, 557-563, doi:10.1021/acsami.1c16513

(2022).

218 Vermaas, D. A. & Smith, W. A. Synergistic Electrochemical CO2 Reduction and Water

Oxidation with a Bipolar Membrane. ACS Energy Letters 1, 1143-1148,

doi:10.1021/acsenergylett.6b00557 (2016).

219 Xie, K., Miao, R. K., Ozden, A., Liu, S., Chen, Z., Dinh, C. T., Huang, J. E., Xu, Q., Gabardo,

C. M., Lee, G., Edwards, J. P., O'Brien, C. P., Boettcher, S. W., Sinton, D. & Sargent, E. H.

Bipolar membrane electrolyzers enable high single-pass CO(2) electroreduction to

multicarbon products. Nat Commun 13, 3609, doi:10.1038/s41467-022-31295-3 (2022).

220 Blommaert, M. A., Aili, D., Tufa, R. A., Li, Q., Smith, W. A. & Vermaas, D. A. Insights and

Challenges for Applying Bipolar Membranes in Advanced Electrochemical Energy Systems.

ACS Energy Lett 6, 2539-2548, doi:10.1021/acsenergylett.1c00618 (2021).

221 De Mot, B., Hereijgers, J., Daems, N. & Breugelmans, T. Insight in the behavior of bipolar

membrane equipped carbon dioxide electrolyzers at low electrolyte flowrates. Chemical

Engineering Journal 428, doi:10.1016/j.cej.2021.131170 (2022).

222 Pribyl-Kranewitter, B., Beard, A., Schuler, T., Diklić, N. & Schmidt, T. J. Investigation and

Optimisation of Operating Conditions for Low-Temperature CO2 Reduction to CO in a

Forward-Bias Bipolar-Membrane Electrolyser. Journal of The Electrochemical Society 168,

doi:10.1149/1945-7111/abf063 (2021).

223 Yan, Z., Hitt, J. L., Zeng, Z., Hickner, M. A. & Mallouk, T. E. Improving the efficiency of

CO(2) electrolysis by using a bipolar membrane with a weak-acid cation exchange layer. Nat

Chem 13, 33-40, doi:10.1038/s41557-020-00602-0 (2021).

224 Guo, L., Zhang, X., Gan, L., Pan, L., Shi, C., Huang, Z.-F., Zhang, X. & Zou, J.-J. Advances

in Selective Electrochemical Oxidation of 5-Hydroxymethylfurfural to Produce High-Value

Chemicals. Advanced Science 10, 2205540, doi:https://doi.org/10.1002/advs.202205540

(2022).

225 Brückner, S., Feng, Q., Ju, W., Galliani, D., Testolin, A., Klingenhof, M., Ott, S. & Strasser,

P. Design and Diagnosis of high-performance CO2-to-CO electrolyzer cells.

doi:10.26434/chemrxiv-2023-z6v6m (2023).

226 Klingenhof, M., Hauke, P., Kroschel, M., Wang, X., Merzdorf, T., Binninger, C., Ngo Thanh,

T., Paul, B., Teschner, D., Schlögl, R. & Strasser, P. Anion-Tuned Layered Double Hydroxide

Anodes for Anion Exchange Membrane Water Electrolyzers: From Catalyst Screening to

Single-Cell Performance. ACS Energy Letters, 3415-3422, doi:10.1021/acsenergylett.2c01820

(2022).

227 Li, C., Ju, W., Vijay, S., Timoshenko, J., Mou, K., Cullen, D. A., Yang, J., Wang, X.,

Pachfule, P., Bruckner, S., Jeon, H. S., Haase, F. T., Tsang, S. C., Rettenmaier, C., Chan, K.,

Cuenya, B. R., Thomas, A. & Strasser, P. Covalent Organic Framework (COF) Derived Ni-N-

C Catalysts for Electrochemical CO(2) Reduction: Unraveling Fundamental Kinetic and

Structural Parameters of the Active Sites. Angew Chem Int Ed Engl 61, e202114707,

doi:10.1002/anie.202114707 (2022).

228 Varela, A. S., Ju, W., Bagger, A., Franco, P., Rossmeisl, J. & Strasser, P. Electrochemical

Reduction of CO2 on Metal-Nitrogen-Doped Carbon Catalysts. ACS Catalysis 9, 7270-7284,

doi:10.1021/acscatal.9b01405 (2019).

229 Vijay, S., Ju, W., Brückner, S., Tsang, S.-C., Strasser, P. & Chan, K. Unified mechanistic

understanding of CO2 reduction to CO on transition metal and single atom catalysts. Nature

Catalysis 4, 1024-1031, doi:10.1038/s41929-021-00705-y (2021).

230 Bagger, A., Ju, W., Varela, A. S., Strasser, P. & Rossmeisl, J. Single site porphyrine-like

structures advantages over metals for selective electrochemical CO2 reduction. Catalysis

Today 288, 74-78, doi:10.1016/j.cattod.2017.02.028 (2017).

References

126

231 Bagger, A., Ju, W., Varela, A. S., Strasser, P. & Rossmeisl, J. Electrochemical CO2

Reduction: A Classification Problem ChemPhysChem 18 (2017).

232 Ju, W., Bagger, A., Hao, G.-P., Varela, A. S., Sinev, I., Bon, V., Cuenya, B. R., Kaskel, S.,

Rossmeisl, J. & Strasser, P. Understanding activity and selectivity of metal-nitrogen-doped

carbon catalysts for electrochemical reduction of CO 2. Nature communications 8, 944,

doi:10.1038/s41467-017-01035-z (2017).

233 Moller, T., Ju, W., Bagger, A., Wang, X. L., Luo, F., Thanh, T. N., Varela, A. S., Rossmeisl,

J. & Strasser, P. Efficient CO2 to CO electrolysis on solid Ni-N-C catalysts at industrial

current densities. Energy & Environmental Science 12, 640-647, doi:10.1039/c8ee02662a

(2019).

234 Patel, P., Schwartz, D., Wang, X., Lin, R., Ajao, O. & Seifitokaldani, A. Technoeconomic and

Life-Cycle Assessment for Electrocatalytic Production of Furandicarboxylic Acid. ACS

Sustainable Chemistry & Engineering, doi:10.1021/acssuschemeng.1c08602 (2022).

235 Jouny, M., Luc, W. & Jiao, F. General Techno-Economic Analysis of CO2 Electrolysis

Systems. Industrial & Engineering Chemistry Research 57, 2165-2177,

doi:10.1021/acs.iecr.7b03514 (2018).

236 Dionigi, F., Zeng, Z., Sinev, I., Merzdorf, T., Deshpande, S., Lopez, M. B., Kunze, S.,

Zegkinoglou, I., Sarodnik, H., Fan, D., Bergmann, A., Drnec, J., Araujo, J. F., Gliech, M.,

Teschner, D., Zhu, J., Li, W. X., Greeley, J., Cuenya, B. R. & Strasser, P. In-situ structure and

catalytic mechanism of NiFe and CoFe layered double hydroxides during oxygen evolution.

Nat Commun 11, 2522, doi:10.1038/s41467-020-16237-1 (2020).

237 Dunn, J. B. & Savage, P. E. Economic and environmental assessment of high-temperature

water as a medium for terephthalic acid synthesis. Green Chemistry 5, 649-655,

doi:10.1039/B305409H (2003).

238 GlobalPetrolPrices. Natural gas prices,

https://www.globalpetrolprices.com/natural_gas_prices/ (2021).

239 GlobalPetrolPrices. USA electricity prices,

https://www.globalpetrolprices.com/USA/electricity_prices/ (2021).

240 M.Garside. Production capacity of purified terephthalic acid worldwide in 2018 and 2023,

https://www.statista.com/statistics/1065886/global-purified-terephthalic-acid-production-

capacity/ (2019).

Appendix

127

10 Appendix

10.1 Supporting Information Chapter 4

This chapter is reprinted with permission from Klingenhof, M‡.; Hauke, P‡.; Kroschel, M.;

Wang, X.; Merzdorf, T.; Binninger, C.; Ngo Thanh, T.; Paul, B.; Teschner, D.; Schlögl, R.;

Strasser, P., Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Membrane

Water Electrolyzers: From Catalyst Screening to Single-Cell Performance. ACS Energy Letters

American Chemical Society.127

SEM investigations of NF, NiFe-LDH@NF and Cl- and ClO4--modified NiFe-LDH@NF

Figure S1: SEM images of NF (black framed); a)-c), NiFe(-CO32-)-LDH@NF (red framed); d), NiFe(-Cl-)-LDH@NF (blue

framed); e) and NiFe(-ClO4-)-LDH@NF (green framed); f).

Appendix

128

XPS investigations of NiFe powder, NF, NiFe-LDH@NF and NiFe-LDH/Cl-@NF

Figure S2: XPS measurements of NiFe(-CO32-)-LDH powder. a) Survey spectrum, b) Ni 2p region, c) Fe 2p region, d) O 1s

region, e) C 1s region.

Figure S3: NiFe(-CO32-)-LDH@NF XPS measurements. a) Survey spectrum, b) Ni 2p, c) Fe 2p, d) O 1s and e) C 1s regions

of the prepared electrode.

Appendix

129

Figure S4: NiFe(-Cl-)-LDH@NF XPS measurements. a) Survey spectrum, b) Ni 2p, c) Fe 2p, d) O 1s, e) C 1s.

Figure S5: NiFe(-ClO4-)-LDH@NF XPS measurements. a) Survey spectrum, b) Ni 2p, c) Fe 2p, d) O 1s, e) C 1s and f) Cl

2p regions of the prepared electrode.

Appendix

130

Supplementary Discussion 1: Results of the electrochemical RDE-Type investigation

In order to investigate the influence of the different used precursors on the synthesis and the

resulting electrochemical activity supporting the need for nickel salts within the synthesis,

experiments were carried out in which the composition of the reaction solution was varied with

regard to the metal salts used. The figures are related to the electrochemical investigations

depicted in Figure 13. As can be seen the presence of each used precursor is necessary to

provide the recorded high current densities. Despite the presence of Nickel hydroxide,

confirmed by XPS (Figure S10) after synthesis without any precursor and synthesis conducted

with only Fe-Prec. The electrocatalytic properties show significant decrease in activity,

compared to synthesis applying Ni and Fe precursors. Recorded LSVs of these electrode

materials are presented in the SI (Figure S6). Regardless of the use of nickel salts in the

synthesis, nickel oxy-hydroxide is formed on the surface of the nickel foams. XPS spectra of

nickel foam without any addition of metal salts shows clear nickel oxy-hydroxide signals in the

XPS spectrum (Figure S10b). The admission of exclusively iron salts leads not only to the

formation of nickel oxy-hydroxide but also to the formation of iron oxy-hydroxide surface

species (Figure S8). Regardless of the formation of the corresponding surface species, however,

there is a significant deterioration in the electrocatalytic activity of the resulting electrodes if

nickel salt is omitted as a precursor.

Appendix

131

Figure S6: The electro catalytic OER activity from NF (black), no precursor except blank Nickel Foam in the synthesis

(light purple) and Fe precursor only (ochre) to NiFe(-CO32-)-LDH@NF (red) for synthesis proof and catalytic-

performance-comparison. The CVs were investigated by using a Pt-mesh as counter electrode and a reversible hydrogen

electrode (RHE) as reference electrode in 50 ml N2-saturated 0.1 M KOH at room temperature.

Figure S7: RDE investigations of Nickel Foam and NiFe(-CO32-)-LDH@NF. Different reaction times were investigated (b)

blank NF, c) 30 min at 160 °C, d) 60 min at 160 °C and e) 90 min at 160 °C) to optimize the synthesis and achieve the most

active and stable material, a) summarizes the measurements in b) – e). Measured using a Pt-mesh as counter electrode and a

reversible hydrogen electrode (RHE) as reference electrode in 50 ml N2-saturated 0.1 M KOH at room temperature.

Appendix

132

Figure S8: XPS investigations of Nickel Foam treated only with Fe precursor (Fe(NO3)3*9H2O) in the synthesis. a)

Survey, b) Ni 2p, c) Fe 2p, d) O 1s and e) C 1s.

Figure S9: SEM images of Nickel Foam treated only with Fe salts during the synthesis. The scale bars are a) 100 µm, b)

1 µm and c) 1 µm. The formed layer appears to be quite dense and thick, typical Nickel foam characteristics are not visible

below the formed Iron layer at the surface. Due to the low contrast of the sample a low conductivity of the sample is expected.

Figure S10: XPS spectra of Nickel Foam treated with DMF, no additional metal salts were added in the synthesis. a)

Survey, b) Ni 2p, c) Fe 2p, d) O 1s, e) C 1s.

Appendix

133

Figure S11: SEM images of Nickel Foam treated without any metal salts during the synthesis. The scale bars are a) 100

µm, b) 1 µm and c) 1 µm. As expected, a very thin Nickel oxyhydroxide layer forms at the surface of the Nickle Foam. Nickel

Foam characteristics are still visible below the thin layer (b).

Figure S12: Impedance spectra recorded for different NiFe(-A-)-LDH@NF species in a three-electrode setup. a)

Exemplary impedance spectra of NiFe(-CO32-)-LDH@NF highlighting the different impedance regions (RΩ, RCT and Ra) with

inset: Schematic depiction of the used ESB. Resulting impedance values of b) RΩ, c) RCT and d) Ra in dependence of the applied

potential.

The ohmic resistance (RΩ) depicted in Figure S12 b results from setup specific characteristics

such as the distance between WE and reference electrode (RE) or the distance between WE and

CE. Therefore, RΩ does not trend with the measured electrochemical activity. RCT, depicted in

Figure S12 c clearly trends with the measured electrochemical activity towards OER

(Figure S13 and Figure 13). The activation of the electrode does not trend with the activity.

Appendix

134

Nevertheless, calculated Ra exhibits basically the same values for each material independent of

NiFe-LDH species, especially NiFe(-CO32-)-LDH@NF and NiFe(-Cl-)-LDH@NF. Ra is in

direct correlation with the constant phase element of the anode (CPEa), which is used for the

determination of the ECSA.

Table S1: ICP-OES results of powdered NiFe(-CO32-)-LDH, NiFe(-Cl-)-LDH, and NiFe(-ClO4-)-LDH. Based on the

identical NF synthesis protocol.

NiFe(-CO32-)-LDH

NiFe(-Cl-)-LDH

NiFe(-ClO4-)-LDH

Ni content [mg]

0.418

0.237

0.150

Fe content [mg]

0.134

0.098

0.095

Fe mol% [%]

Fe at% [%]

Supplementary Discussion 2: Stability investigations, three-electrode setup.

Beyond catalytic activity, the catalyst stability was examined by applying 1000 CVs between

+1.1 VRHE and +1.7 VRHE. NF, NiFe(-CO32-)-LDH@NF and NiFe(-ClO4-)-LDH@NF showed

insignificant activity losses, whereas the activity of NiFe(-Cl-)-LDH@NF declined within the

first 900 cycles (Figure 13h-k), yet remained the most active OER electrode. While the catalytic

performance of NiFe(-CO32-)-LDH@NF and NiFe(-ClO4-)-LDH@NF remained very stable,

the catalytic sweep voltammetry of NiFe(-Cl-)-LDH@NF converged to that of

NiFe(-CO32-)-LDH@NF (Figure S13), indicating a structural relaxation and back

transformation of NiFe(-Cl-)-LDH@NF to the catalytically active γ-phase

NiFe(-CO32-)-LDH@NF under operation conditions.176,236 This is in great agreement with the

XRD investigations shown in Figure 12. Further, a comparison of the uncompensated ohmic

RΩ evaluated at +1.53 VRHE, +1.58 VRHE and +1.63 VRHE revealed increasing values for

NiFe(-Cl-)-LDH@NF.

Appendix

135

Figure S13: Investigations of the electrocatalytic OER activity after 1000 cycles. The prepared and modified catalysts

(NiFe(-CO32-)-LDH@NF, NiFe(-A-)-LDH@NF) in 0.1 M KOH at RT in RDE scale experiment. LSV curves without iR

correction of the prepared electrodes a) NiFe(-CO32-)-LDH@NF, b) NiFe(-Cl-)-LDH@NF and c) NiFe(-ClO4-)-LDH@NF.

Figure S14: Color change of NF and NiFe(-CO32-)-LDH@NF during cyclovoltammetry a) NF during potential sweep does

not show change of the color, while b) NiFe(-CO32-)-LDH@NF during potential sweep indicate the oxidation of the catalyst

material (NiFe(-CO32-)-LDH) during cycling, changing from metallic gray to black (1-3) and back (4-6). Measured using a Pt-

mesh as counter electrode and a reversible hydrogen electrode (RHE) as reference electrode in 50 ml N2-saturated 0.1 M KOH

at room temperature.

Appendix

136

Table S2: ICP-OES results after stability protocol of NiFe(-Cl-)-LDH@NF.

Ni [g/l]

Fe [g/l]

Pure 0.1 M KOH

3.071∙10−6

1.593∙10−5

Electrolyte after

stability protocol

5.289∙10−5

4.457∙10−5

Deviation

+4.982∙10−5

+2.864∙10−5

Figure S15: Investigations of the electrocatalytic OER activity at 60°C. IR-corrected LSVs of NF (black), the prepared and

modified catalysts (NiFe(-CO32-)-LDH@NF (red) and NiFe(-Cl-)-LDH@NF (blue)) in 0.1 M KOH at RT (50% transparency)

and 60°C in RDE scale experiment.

Appendix

137

Cell

Figure S16: Applied single cell electrolyzer setup used for the AEMWE measurements. a) disassembled and b) assembled.

With 1: Endplate with electrolyte inlets; 2: Current collector, 3: Flow Field (Anode Titanium, Cathode: Carbon); 3a: Gaskets

and 4 MEA: Membrane Electrode Assembly.

Figure S17: Polarization performance of AEM cells evaluating PTL, T, pH, membranes, cathodes and cathode Pt-

loading. The following operation conditions and cell parameters were changed and modified to gain the best possible

performance with the applied system. a) Different types of cathodic PTLs (Sigracet) were loaded with 0.1 mg Pt cm-2 to choose

the best performing PTL, FAS-50 is applied as membrane. b) The temperature was adjusted to choose the best operation

temperature, here 38 BC with 0.1 mg Pt cm-2 was applied. c) Influence of the pH was measured by conducting a measurement

applying 0.1 M KOH and 0.0001 M KOH. d) FAA-3 based membranes were compared (FAS-50 and FAA-3-30). e) 0.1 mg Pt

cm-2 at 38 BC applied as cathode is compared to Nickel powdered on Nickel Foam. f) The influence of Pt loading (38 BC) is

investigated.

Appendix

138

Figure S18: a)-d) Representation of the uncorrected, iR-corrected, iR- and HER-overpotential-corrected polarization

curves, as well as the overpotentials of the different investigated anodes resulting from the HER. The measurements

correspond to the polarization curves shown in the Main and were carried out at 60°C using 0.1 M KOH. As an anode, the

different NiFe-based materials produced were measured and compared with uncoated NF. 0.1 mg Pt cm-2 spray coated at

Sigracet 38 BC was used as cathode.

The HER overpotentials and thus contributions to the total cell potential were calculated using

the Tafel slope for Pt-C measured in 0.1 M KOH reported by Zhu et al., as comparable

conditions were used here.141 In this study, a Tafel slope of 77 mV dec-1 is determined. Based

on these values, the HER contributions were determined for the measured current densities of

the manufactured materials, as well as uncoated NF. The graphs shown in Figure S18a-d)

contain the uncorrected polarization curves, the iR-corrected polarization curves, the iR and

HER corrected polarization curves, and the pure HER overpotentials. Due to the increasing

current densities, the HER contribution of the total cell potential is also increasing.

The polarization curves in Figure S18 were used to determine the activity improvement factors

depicted in the main and the following figure (Figure S19). In order to determine f, the current

density of the different prepared anodes as well as of uncoated NF was recorded at 1.52 V (iR

and HER corrected) and divided by the current density that uncoated NF has at this potential.

Therefore, f for uncoated NF is 1. iR and HER corrected values were used to exclude any

limitations arising from the SC setup.

Appendix

139

Figure S19: Activity improvement factors plotted over the applied potential a) without iR-correction and b) with iR and

HER correction polarization curves. f is calculated using 𝐟=𝐣𝐠𝐞𝐨𝐦.,𝐜𝐚𝐭 𝐚𝐭𝐂𝐞𝐥𝐥𝐕𝐨𝐥𝐭𝐚𝐠𝐞

𝐣𝐠𝐞𝐨𝐦.,𝐍𝐅 𝐚𝐭𝐂𝐞𝐥𝐥𝐕𝐨𝐥𝐭𝐚𝐠𝐞. f is directly derived from the plots depicted in

Figure S18.

As can be seen in Figure S19, significant differences between f derived from uncorrected

polarization curves and iR- and HER-corrected polarization curves occur. As expected, iR- and

HER-contributions increase with increasing potential and current densities. Therefore, f is less

pronounced in case of the the uncorrected polarization curves, lost with the iR- and HER-

corrected polarization curves. NiFe(-CO32-)-LDH@NF and NiFe(-Cl-)-LDH@NF constantly

show an f greater than 1, where f decreases slowly with increasing potential of the uncorrected

polarization curves, f from corrected polarization curves also decreases with increasing

potential. The course of f in this case is comparable and is probably within the error bar. f of

NiFe(-ClO4-)-LDH@NF increases with increasing potentials and increasing current densities

and only leads to an improvement over uncoated NF at 1.9 V cell potential uncorrected. It

appears, that the electrocatalytic activity is decreased with the second anion exchange.

However, the iR- and HER-contribution of the measured polarization curves is not sufficient to

explain the drastic loss of performance compared to three-electrode measurements. Main

contribution to improve f of NiFe(-A-)-LDH@NF over uncoated NF is the iR-correction. Thus,

NiFe(-A-)-LDH is accompanied with a non-beneficial increase in RΩ.

Appendix

140

Figure S20: NF XPS measurements of blank Nickel Foam before the exposure to OER relevant potential regions as well

as the unpurified or Iron containing electrolyte from the Greenlight test station. a) Survey spectrum, b) Ni2p region, c)

O1s region, d) C1s region.

Figure S21: XPS investigations of Nickel Foam after the exposure to OER relevant potential regions and unpurified

electrolyte resulting from stainless steel components used for the Greenlight electrolyzer. a) Survey spectrum highlighting

the Ni2p (b), Fe2p (c), C1s (d) and O1s (e) region.

Appendix

141

Ni 2p XPS spectra depicted in Figure S21b clearly proves Ni (oxy)hydroxide formation during

the exposure to the electrochemical protocol described in Table S1. Boettcher et al. have already

impressively described the incorporation of Fe impurities into Ni (oxy)hydroxide lattice

accompanied with a significant increase in OER activity.156

Figure S22: a) and b) Investigation of the electrocatalytic activity of NF before (black) and after (dark cyan) exposure

to AEMWE single cell testing. c) Nyquist-Plots of NF after AEMWE single cell testing. d) ECSA comparison of NF, NF after

WE and the prepared anodes, the ECSA was determined by EIS measurements The CVs were investigated by using a Pt-mesh

as counter electrode and a reversible hydrogen electrode (RHE) as reference electrode in 50 ml N2-saturated 0.1 M KOH at

RT.

Electrochemical testing of NF after AEMWE single cell testing confirm the significant increase

in OER activity (Figure S22). The depicted LSVs further proves the Ni (oxy)hydroxide and Fe

incorporation on the surface of the NF. Additionally, the anodic shift of the nickel redox wave

is another indication of Fe incorporation.

Appendix

142

Supplementary Discussion 3: Results of the impedance investigation of AEMWE single

cell PEIS.

Figure S23:a) Exemplary impedance spectra of NiFe(-CO32-)-LDH@NF recorded in a two electrode electrolyzer cell, b)

Results of RΩ, RCT and Ra applying the ESB schematically depicted in c).

A comparison between the three-electrode impedance spectra and single cell impedance spectra

clarifying a difference between both setups. While it is possible to clearly separate the areas

treated in the EEC within a three-electrode impedance measurement, strong overlaps appear in

the case of two-electron measurements. It is not possible to separate Ra and RCT and thus analyze

Anode and Cathode separated. This is due to the simultaneous resonance of the anode and

cathode. As a result, a separate consideration of anode and cathode cannot be performed.

Nevertheless, differences in the high-frequency range, measured in three-electrode setup, are

barely visible. Impedance spectra of implemented manufactured electrodes in full cells

therefore do not allow for detailed conclusions about individual properties of the applied

electrode materials, further stressing the need testing impedances of manufactured electrodes

in setups, reducing all other limitations except working electrode limitations to a minimum.

Impedance spectra comparing NiFe(-CO32-)-LDH@NF with (-A-)- show the same trend,

decreasing RCT with increasing current density.

Appendix

143

Figure S23b summarizes the calculated values of RΩ, RCT and Ra. Since the RΩ values are

comparable and almost the same only low influence of the setup is expected and proofs the

reproducible setup measurement conditions for each electrode. We note however that the LDH

coated and tuned electrodes show higher RΩ. Differences of RΩ can be attributed to an increase

of the introduction of an electrically poorly conductive layer consisting of NiFe-LDH species.

The RCT values are consistent with the measured electrochemical activities. The already

mentioned overlap of electrode activation and RCT prevents an interpretation of Ra, but the

calculated values nevertheless show about the expected trend with a higher activation resistance

for the electrodes coated with NiFe-LDH species.

Appendix

144

10.2 Supporting Information Chapter 5

This subchapter is reproduced from Hauke, P., Klingenhof, M., Wang, X., de Araújo, J. F. &

Strasser, P. Efficient electrolysis of 5-hydroxymethylfurfural to the biopolymer-precursor

furandicarboxylic acid in a zero-gap MEA-type electrolyzer. Cell Reports Physical Science 2,

100650, doi: https://doi.org/10.1016/j.xcrp.2021.100650 (2021). With permission of Cell

Reports Physical Science.155

Table S 3. ICP-OES results of the powdered NiX-LDH (X= Fe,Co,Mn,V) nanocatalysts. Comparison of the molar Ni:X

ratio and total wt% of metal for all NiX-LDH nanocatalysts. The standard deviation of the ICP-OES measured ppms are Ni:

1-5%, Fe: 5%, Mn: 3% Co: 19% and V: 3%.

Material

Ni:X molar ratio

Total wt% of metal

NiFe-(CO32-)- LDH

3:1

NiMn-(CO32-)- LDH

23:1

NiCo-(CO32-)- LDH

7:1

NiV-(CO32-)- LDH

1.5:1

Appendix

145

Figure S 24. Scanning Electron Microscopy images of a) NiFe(-CO32-)-LDH@NF, b) NiMn(-CO32-)-LDH@NF, c) NiCo(-

CO32-)-LDH@NF and d) NiV(-CO32-)-LDH@NF at 1000 x magnification.

Figure S 25. Scanning Electron Microscopy images of NiV(-CO32-)-LDH@NF at different synthesis parameters. a)-c)

show lower precursor concentration of 0.6 M and 30, 60, 90 min synthesis time (left to right) and d)-f) show higher precursor

concentration of 1.2 M and 30,60,90 min synthesis time (left to right) at 10000 x magnification.

Appendix

146

Table S 4. Electrochemical surface area of NiX-(CO32-)-LDH@NF (2.5 cm2). Results of the impedance measurements at

1.58 VRHE in 0.1 M KOH comparing the ECSA (cm2) of Ni nanofoam support with NiX-(CO32-)-LDH@NF (X=Fe,Mn,Co,V).

Data of NF and NiFe-(CO32-)-LDH@NF were obtained from a perspective published study of M.K. and P.H..

Figure S 26. Nyquist plots of NiX-LDH@NF. Exemplary Nyquist plots of Ageom.= 1.5 cm2 a) NiMn- (blue), b) NiCo- (green),

c) NiV(-CO32-)-LDH@NF (purple) in 0.1 M KOH at 1.53 VRHE, 1.58 VRHE and 1.63 VRHE for the electrochemical surface area

(ECSA) evaluation. Solid lines show fits of the raw data (dotted lines). d) the applied equivalent circuit to evaluate the measured

impedance spectra. Element R1 describes the resistance arising from the electrolyte and electrical connections, elements R4

and Q1 are the reaction charge transfer resistance (RCT), and Q2 and R5 are associated to be the response of changes in the

coverage of adsorbed species.

ERDE

[VRHE]

NiFe-(CO32-)-

LDH@NF

NiMn-

(CO32-)-

LDH@NF

NiCo-

(CO32-)-

LDH@NF

NiV-(CO32-)-

LDH@NF

1.58 V

1.906 cm2

187.838 cm2

37.756 cm2

50.733 cm2

567.160 cm2

Appendix

147

Table S 5. ICP-OES results of the powdered NiFe(-A-)-LDH (A= CO32-, Cl-, ClO4-) nanocatalysts. Comparison of the

molar Ni:Fe ratio and total wt% of metal for all NiX-LDH nanocatalysts. The standard deviation of the ICP-OES measured

ppms are Ni: 1-5%, Fe: 5%.

Table S 6. Electrochemical surface area after anion exchange. Results of the impedance measurements at 1.58 VRHE in

0.1 M KOH comparing the ECSA (cm2) of Ni nanofoam support with NiFe(-A-)-LDH@NF (A= CO32-, Cl-, ClO4-) (2.5 cm2).

Data were obtained from a perspective published study of M.K. and P.H..

Material

Ni:Fe molar ratio

Total wt% of metal

NiFe-(CO32-)-LDH

3:1

NiFe-Cl--LDH

2.3:1

NiFe-ClO4--LDH

1.5:1

ERDE

[VRHE]

NiFe-(CO32-)-

LDH@NF

NiFe(-Cl-)-

LDH@NF

NiFe(-ClO4-)-

LDH@NF

1.58 V

1.906 cm2

187.838 cm2

139.046 cm2

45.973 cm2

Appendix

148

Figure S 27. Evaluation of 5-HMF in 0.1 M KOH. Showing the decreasing 5-HMF concentration over time (5 h) in 0.1 M

KOH at room temperature (RT). Indicating a relevant autocatalytic reaction from 5-HMF with the electrolyte after 3h.

Figure S 28. RDE measurements of Ni nanofoam, Nickel Disc (ND), and Carbon Paper (CP). Working electrodes, WE,

geometric area of 0.196 cm2 with and without 10 mM 5-HMF in 0.1 M KOH. The LSVs (IR corrected) were taken between

1.0 VRHE-1.7 VRHE at a scan rate of 5mV s-1, no rotation was applied.

Appendix

149

Figure S 29. RDE measurements of NiV(-CO32-)-LDH@NF comparing the activity of different synthesis parameters (c,

t). Synthesis of NiV(-CO32-)-LDH@NF was varied by the holding time (30 min, 60 min, 90 min) and the Vanadium precursor

concentration using 0.6 M and higher concentrated 1.2 M solution corresponding to Figure S2. LSV measurements were taken

from 1VRHE to 2VRHE at a scan rate of 5mV s-1 in 0.1 M KOH and 10 mM 5-HMF. All measurements are internal resistance

(IR) corrected.

Figure S 30. 5-HMF concentration dependence. Flow Electrolyzer Cell based linear scan voltammograms (LSV) using pure

Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and FAS 50 membrane in 0.1 M KOH

used as catholyte and 0-15 mM HMF in 0.1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis

parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell - 1.7 Vcell,

scan rate 10 mV s-1.

Appendix

150

Figure S 31. 5-HMF concentration dependence (HFR corrected). Flow Electrolyzer Cell based linear scan voltammograms

(LSV) using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and FAS 50 membrane

in 0.1 M KOH used as catholyte and 0-15 mM HMF in 0.1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml.

Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell - 1.7 Vcell, scan rate 10 mV s-1.

All potential values are IR corrected for uncompensated ohmic HFR.

Figure S 32. Anion exchange membrane evaluation. Flow Electrolyzer Cell based linear scan voltammograms (LSV) using

pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and different anion exchange

membranes in 0.1 M KOH used as catholyte and 10 mM HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and anolyte

volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell - 1.7 Vcell,

scan rate 10 mV s-1.

Appendix

151

Figure S 33. Anion exchange membrane evaluation (HFR corrected). Flow Electrolyzer Cell based linear scan

voltammograms (LSV) using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and

different anion exchange membranes in 0.1 M KOH used as catholyte and 10 mM HMF (HMF) in 0.1 M KOH used as anolyte.

Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials

1.0 Vcell - 1.7 Vcell, scan rate 10 mV s-1.

Table S 7. HPLC results after 2 h of 5-HMF electrolysis at 150 mA cm-2 with NF (anode), Pt-C@CP (cathode) in 0.1 M

KOH with 10 mM 5-HMF (V= 100ml) and a flow rate of 20 ml/min for different membranes. The standard deviation for

all values is ±1%.

System

XHMF,2h

[%]

YHFCA,2h

[%]

YFFCA,2h

[%]

YFDCA,2h

[%]

SFDCA,2h

[%]

FAS30

7.5

31.5

FAS50

6.5

4.5

11.5

FAA-3-PK-

130

59.5

9.5

IONOMR

HNN8

26.5

Sustainion RT

Sustainion

X37

14.5

Appendix

152

Figure S 34. Images of two different used membranes. a) Images of the FAA-3-PK-130 Fumatech membrane. Showing the

web structure of it. b) IONOMR HNN 8 membrane after reaction as evidence for activity disparities.

Figure S 35. HPLC Calibration curves of 5-HMF its oxidation products. a)-e) are showing the HPLC response for 5-HMF

and its oxidation products at 1 mM (black), 5 mM (red), and 10 mM (blue) in 0.1 M KOH with H2SO4 (1 ml/min) as mobile

phase. It was not completely possible to avoid the conversion of c) FDA to a) 5-HMF and e) FFCA, which exclude 1 mM FDA

calibration.

Appendix

153

Figure S 36. Influence of the electrolyte cation. Flow Electrolyzer Cell based linear scan voltammograms (LSV) using pure

Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1

M XOH (X=Li,Na,K,Rb,Cs) used as catholyte and 10 mM HMF (HMF) in 0.1 M XOH used as anolyte. Catholyte and

anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2

Vcell, scan rate 10 mV s-1.

Figure S 37. Influence of the electrolyte cation (HFR corrected). Flow Electrolyzer Cell based linear scan voltammograms

(LSV) using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130

membrane in 0.1 M XOH (X=Li,Na,K,Rb,Cs) used as catholyte and 10 mM HMF (HMF) in 0.1 M XOH used as anolyte.

Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials

1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1.

Appendix

154

Figure S 38. Evaluation of a second metal in addition to Ni (HFR corrected). Flow Electrolyzer Cell based linear scan

voltammograms (LSV) using pure Ni foam (NF) and NiX(-CO32-)-LDHs (X=Fe,Mn,Co,V) anodes, C supported Pt

nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used as catholyte and 10 mM

HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml

min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1.

Table S 8. HPLC results after 1 and 2 h of 5-HMF electrolysis at 1.56 Vcell with NiX(-CO32-)-LDH@NF (X= Fe, Mn, Co

and V) (anode)// FAA-3-PK-130//Pt-C@CP (cathode) in 0.1 M KOH with 10 mM 5-HMF (V= 100ml) and a flow rate of

20 ml/min. The standard deviation for all values is around ±1%.

catalyst

XHMF

, 1h

[%]

XHMF

, 2h

[%]

YHFCA

, 1h

[%]

YHFCA,

[%]

YFFCA,

[%]

YFFCA

,2h

[%]

YFDCA,

[%]

YFDCA,

[%]

SFDCA,

[%]

SFDCA,

[%]

FEFDCA,

[%]

NF (150 mA cm-2)

59.5

9.5

NiFe(-CO32-)-

LDH@NF

3.5

1.5

88.5

NiMn(-CO32-)-

LDH@NF

13.5

14.5

12.5

NiCo(-CO32-)-

LDH@NF

NiV(-CO32-)-

LDH@NF

97.5

5.5

71.5

Appendix

155

Figure S 39. Investigation of the NiV(-A-)-LDH@NF interlayer chemistry. Flow Electrolyzer Cell based linear scan

voltammograms (LSV) using NiV(-CO32-)-LDH@NF and NiV(-Cl-)-LDH@NF anodes, C supported Pt nanoparticles (loading

of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used as catholyte and 10 mM HMF (HMF) in 0.1 M

KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end

LSV electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1.

Figure S 40. Investigation of the NiV(-A-)-LDH@NF interlayer chemistry (HFR corrected). Flow Electrolyzer Cell based

linear scan voltammograms (LSV) using NiV(-CO32-)-LDH@NF and NiV(-Cl-)-LDH@NF anodes, C supported Pt

nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used as catholyte and 10 mM

HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate

20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1.

Appendix

156

Figure S 41. Investigation of the interlayer chemistry of NiFe-(A-)-LDH@NF (HFR corrected). Flow Electrolyzer Cell

based linear scan voltammograms (LSV) using NF and NiFe(-A-)-LDH@NF (A=CO32-,Cl-,ClO4-) anodes, C supported Pt

nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used as catholyte and 10 mM

HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate

20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1.

Table S 9. HPLC results after 1 and 2 h of 5-HMF electrolysis at 1.56 Vcell with NiFe(-A-)-LDH@NF (A=CO32-, Cl- and

ClO4-)// FAA-3-PK-130//Pt-C@CP in 0.1 M KOH with 10 mM 5-HMF (V= 100ml) and a flow rate of 20 ml min-1. The

standard deviation for all values is around ±1%.

catalyst

XHMF

, 1h

[%]

XHMF

, 2h

[%]

YHFCA,

[%]

YHFCA,

[%]

YFFCA,

[%]

YFFCA

,2h

[%]

YFDCA,

[%]

YFDCA,

[%]

SFDCA,

[%]

SFDCA,

[%]

FEFDCA,

[%]

NF (150 mA cm-

59.5

9.5

NiFe(-CO32-)-

LDH@NF

3.5

1.5

88.5

NiFe(-Cl-)-

LDH@NF

100

3.5

NiFe(-ClO4-)-

LDH@NF

4.5

Appendix

157

Figure S 42. Color change within 2 h of electrolysis with NiFe(-Cl-)-LDH@NF, samples were taken every 10 min.

Figure S 43. Correlation of 5-HMF conversion, FDCA yield and current of NiFe(-Cl-)-LDH@NF. MEA-type cell NiFe(-

Cl-)-LDH@NF //FAA-3-PK-130// Pt-C@CP electrolysis with 10 mM 5-HMF in 0.1 M KOH (V= 100ml), a flow of 20 ml min-

1, at a at a constant potential of 1.56 V over 2 h and a Pt catalyst loading of 0.1 mg cm-2.

Appendix

158

Supplementary Discussion 4: Note S1. Evaluation of Figure S 43

Figure S 43 shows the temporal evolution of a complete (XHMF =1) oxidation of 5-HMF to

FDCA during a 2 h electrolysis at +1.56 Vcell.

The direct correlation between decreasing 5-HMF concentration, current density, and

accompanied by rising FDCA concentration es evident. Since the OER process no longer

competes with the HMFOR process at declining HMF concentrations the faradaic current

remains finite. Figure 19a confirms a current density of 6.852 mA cm-2 (34.26 mA) for

NiFe(-Cl-)-LDH@NF at +1.56 Vcell in the absence of 5-HMF (dotted line).

Figure S 44. Scanning Electron Microscopy images of NF and NiFe(-A-)-LDH@NF (A=CO32-b), Cl- c) and ClO4- d)) at

1000-2000x magnification. The color code corresponds to Figure S17.Correlation of 5-HMF conversion, FDCA yield and

current of NiFe(-Cl-)-LDH@NF. MEA-type cell NiFe(-Cl-)-LDH@NF //FAA-3-PK-130// Pt-C@CP electrolysis with 10 mM

5-HMF in 0.1 M KOH (V= 100ml), a flow of 20 ml/min, at a at a constant potential of 1.56 V over 2 h and a Pt catalyst loading

of 0.1 mg cm-2.

Appendix

159

Figure S 45. Evaluation of NiFe(-Cl-)-LDH@NF in different electrolytes. Flow Electrolyzer Cell based linear scan

voltammograms (LSV) using NiFe(-Cl-)-LDH@NF anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes

and a FAA-3-PK-130 membrane in 0.1 M and 1 M KOH used as catholyte and 10 mM HMF (HMF) in 0.1 M and 1 M KOH

used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV

electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1.

Figure S 46. Electrolysis of NiFe(-Cl-)-LDH@NF at a constant potential. Flow Electrolyzer Cell electrolysis using

NiFe(-Cl-)-LDH@NF anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane

in 1 M KOH used as catholyte and 10 mM HMF in 1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml.

Electrolysis parameters: flow rate 20 ml min-1, constant potential of 1.56 Vcell for 50 min.

Appendix

160

Table S 10. HPLC results after 45 min of 5-HMF electrolysis at 1.56 Vcell with NiFe(-Cl-)-LDH@NF// FAA-3-PK-

130//Pt-C@CP in 1 M KOH with 10 mM 5-HMF (V= 100ml) and a flow rate of 20 ml/min. The standard deviation for all

values is around ±1%.

Figure S 47. Time resolved mass spectrometric Linear Sweep Voltammograms of the NiFe(-Cl-)-LDH@NF

electrocatalyst. The data was obtained in a custom-made DEMS flow cell setup in a) 0.1 M KOH and b) 10 mM HMF in 0.1 M

KOH, flow rate 500 µl min-1, scan rate 5 mV s-1 from 0- +2,0 VRDE. Plotted are ERHE and the signals of m/z= 32 of 16O 16O

over the time.

Figure S 48. Schematic side view representation of the main flow path in the capillary flow cell. The liquid flow inlet

fresh electrolyte flow directed to catalysts surface and the concentric tube and capillary.

Time [min]

XHMF

[%]

YHFCA

[%]

YFFCA

[%]

YFDCA

[%]

93.5

2.5

100

Appendix

161

Figure S 49. Pulse Stability test of NiFe(-Cl-)-LDH@NF. Flow Electrolyzer Cell electrolysis using NiFe(-Cl-)-LDH@NF

anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used

as catholyte and 10 mM HMF (injected 5 times every 60 min) in 0.1 M KOH used as anolyte. Catholyte and anolyte volume

of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, constant potential of 1.56 Vcell for 60 min (5 cycles).

Appendix

162

Table S 11. Data collection of the latest published work about HMFOR. The references in the table are kept original

System

Conditions

V vs. RHE

XHMF, YFDCA, FE

STY [µmol s-1 cm-2]

Reference (Main)

CoB/NF//AEM//NF (1 cm2),

flow cell

1 M KOH, 10mM HMF,

1.45 V

100%, 94%, 99%

0.0269

NiOOH/FTO (4cm2), H cell

0.1M KOH, 5 mM HMF,

1.47 V

99.8%, 96%, 96%

0.0009

Cu foil (NCF) (4cm2), H cell

0.1M KOH, 5 mM HMF,

1.62 V

99%, 96.4%, 95.3%

0.0002

NixB/NF (1cm2), flow cell

1 M KOH, 10mM HMF,

1.45 V

100%, 98.5%, 99%

0.0547

PdAu/C (1mg/cm2) (5 cm2)*,

flow cell

0.1 M KOH, 0.02 M HMF,

0.9 V

100%, 83%

0.0230

Ni3N-C/NF (12 cm2), single

cell

1 M KOH, 10mM HMF,

1.45 V

100%, 98%, 99%

0.0544

NiFe-LDH/C (3cm2), H cell

1 M KOH, 10mM HMF,

1.33 V

98%, 98%, 98.6%

0.0177

NiCo2O4/NF (3cm2), H cell

0.1 M KOH, 10 mM HMF,

1.55 V

90%, 90%, 100%

0.0021

Co3O4/NF (2 cm2), H cell

1 M KOH, 10 mM HMF,

1.457 V

100%, 100%, 100%

0.0012

NiB-P/CP (1 cm2), divided

cell

0.1 M KOH, 10 mM HMF,

1.464 V

100%, 90.6%, 92.5%

0.0075

Co-P/CF (4 cm2), divided cell

1 M KOH, 50 mM HMF,

1.423 V

100%, 90%

0.0112

S-Ni/C-600 (0.5 cm2), H cell

1 M KOH, 10 mM HMF,

1.473 V

100%, 96%, 96%

0.0177

Ni2P NPA/NF (0.25 cm2), H

cell

1 M KOH, 10 mM HMF,

1.423 V

100%,100%, 100%

0.0444

(AuPd)7 (1 cm2)*

1 M KOH, 5 mM HMF,

0.82 V

49.3%, 11.1%, 72.8%

0.0003

Co-doped NiBDC-NF (1

cm2), H cell

0.1 M KOH, 10 mM HMF,

1.55 V

-, 99%, 78.8%

0.0055

CuS-NiCoOH LDH/CF (0.5

cm2)

1M KOH, 10 mM HMF,

1.32 V

100%, 99%, 99%

0.044

CoOOH-NF (0.25 cm2), H

Cell

1 M KOH, 10 mM HMF,

1.423 V

100%, 100%, 100%

0.0303

CoFe/NiFe (1cm2)

1 M KOH, 10 mM HMF,

1.4 V

100%, 99.8%, 99.8%

0.0277

NiCo2O4-CFP (1 cm2), H-cell

1 M KOH, 10 mM HMF,

1.43 V

98.4%, 94.3%, 89.6%

0.0137

Ni(OH)2@NF (1cm2), H-cell

1 M KOH, 10 mM HMF,

1.39 V

100%, 100%, 100%

0.0462

Ni(NS)@CP (3cm2), H-cell

0.1 M KOH, 5 mM HMF,

1.36 V

99.7%, 99.4%, 95.3%

0,0068

Appendix

163

Supplementary Discussion 5: Note S2. Technoeconomic of PTA vs FDCA production

Comparison of the energy requirements of the paired H2-FDCA electrolyzer and the

commercial high-temperature-water process of terephthalic acid (PTA).

Thermal energy input of the established thermal PTA process237: 300°C and 150 bar

correspond to a thermal energy input EPTA, FDCA = 4722 kWh tPTA-1 (17000 MJ tPTA-1)

Electrical energy input of the paired H2-FDCA electrolyzer during a single (ntest cycle=1) 2h

potentiostatic electrolysis test cycle: A 5 cm2 electrode area over 7200 seconds of test time at

an applied electrode cell potential +1.56 Vcell requires a total electric charge of Q= 528 C

yielding a FDCA space time yield of 0.0275 µmol s-1 cm-2. This corresponds to an electrical

energy input of 2.28 ∙ 10−4 𝑘𝑊ℎ . To produce 1 ton FDCA, 1475 kWh are necessary according

to the following estimate (Equation 45-47):

nFDCA =STY∙Acat ∙t=0.0275 µmol s−1 cm −2 ∙ 5 cm 2∙7200 s=990 µmol (Equation 53)

ntestcycle= 106g

990∙10−6mol∙156.09 gmol

⁄=6.47∙106 (Equation 54)

EHMF, FDCA =2.28 ∙10−4 kWh ∙ 6.47∙106=1475 kWh tFDCA-1 (Equation 55)

Assuming a natural gas price238 of 0.046 $ kWh-1 and an electricity price239 of 0.126 $ kWh-1

results in 217 $ tPTA-1 and 186 $ tFDCA-1, corresponding to 18.245 billion US$ energy costs for

PTA and 15.611 billion US$ for FDCA based on a production volume of 84 million metric tons

each240. The forecast increases in PTA production to 107 million metric tons until 2023 would

increase the energy cost to 23.241 billion US$ and 19.886 billion US$ respectively. Taking into

account a CO2-price of 25-50 $ tCO2-1 and 2.4 tCO2 tPTA-1 would increase the PTA price by

additional 6.420-12.840 billion US$ per year.44

Appendix

164

10.3 Supporting Information Chapter 6

This subchapter has been reproduced from Hauke, P., Merzdorf, T., Klingenhof, M. & Strasser,

P. Hydrogenation versus hydrogenolysis during alkaline electrochemical valorization of 5-

hydroxymethylfurfural over oxide-derived Cu-bimetallics. Nature Communications 14, 4708,

doi: https://doi.org/10.1038/s41467-023-40463-y (2023).43

Figure S 50: RDE measurements of CuO/NiO comparing the activity of different metal oxide molar ratios (mol%). The

Cu to Ni ratio was varied by different Ni precursor amounts. a) LSV measurements were taken between 0 VRHE to -0,8 VRHE

at a scan rate of 10 mV s-1 in 0.1 M KOH with and without 10 mM 5-HMF. All measurements are internal resistance (IR)

corrected. b) resulting overpotentials from a) for HMFRR (red) and HER (black) at 10 mA cm-2 for different NiO mol%.

Supplementary Discussion 6

To evaluate a preferred molar ratio of CuO and NiO, we tracked the HMF reduction reaction

(HMFRR) activity and the HER activity at 10 mA cm-2. The HER activity is assumed to be a

measure for Had coverage and chemisorption under reactive conditions. Chemisorbed Had

enables desired Langmuir-Hinshelwood-type multi-electron HMFRR, yet also leads to

competitive, undesired Volmer-Tafel Hydrogen Evolution. Based on this balance, we targeted

a 10 mol% metal Ni/Cu ratio in the two-phase catalysts due to its most favorable HMFRR

activity coupled to a balanced HER activity. In order to maintain comparability, the molar ratio

of 10% was kept for other catalysts. We are aware that, for the sake of the present study, the

optimum ratio may vary slightly for Co and Fe based bimetallic oxides. However, as shown in

Figure S 50b, the HER and HMFRR reactivity were similar over most of the molar range.

Appendix

165

Figure S 51: XPS measurements of CuO powder. a) Survey spectrum, b) Cu LMM Auger range c) O 1s region, d) Cu 2p

region. Measurements are given in black whereas Casa XPS fits are given in light blue for O and orange for Cu.

Appendix

166

Figure S 52: XPS measurements of CuO/NiO powder. a) Survey spectrum, b) O 1s region, c) Cu 2p region, d) Ni 2p region.

Measurements are given in black whereas Casa XPS fits are given in yellow for C, light blue for O, orange for Cu and red for

Ni.

Appendix

167

Figure S 53: XPS measurements of CuO/Fe2O3 powder. a) Survey spectrum, b) O 1s region, c) Cu 2p region, d) Fe 2p

region. Measurements are given in black whereas Casa XPS fits are given in yellow for C, light blue for O, orange for Cu and

purple for Fe.

Appendix

168

Figure S 54: XPS measurements of CuO/Co3O4 powder. a) Survey spectrum, b) O 1s region, c) Cu 2p region, d) Co 2p

region. Measurements are given in black whereas Casa XPS fits are given in yellow for C, light blue for O, orange for Cu and

blue for Co.

Appendix

169

Figure S 55: TEM images of the powder CuO/MOx catalysts. 1a)-c) showing CuO/NiO, 2a)-c) showing CuO/Fe2O3 and

3a)-c) showing CuO/Co3O4. A scale is provided in the right bottom corner.

Appendix

170

Figure S 56: HR-STEM images of the powder CuO and CuO/MOx catalysts. 1a) and b) showing CuO, 2a) and b) showing

CuO/NiO, 3a) and b) showing CuO/Fe2O3 and 4a) and b) showing CuO/Co3O4. A scale is provided in the right bottom corner.

Appendix

171

Figure S 57: SEM-EDX measurements of CF before and after electrolysis. a) and b) SEM images of CF before electrolysis

at 1000x and 5000x magnification. c) and d) SEM images of CF after electrolysis (ae) at 1000x and 5000x magnification. e)

SEM EDX images before electrolysis of CF with O in torques and Cu in orange. f) SEM EDX images after electrolysis of CF

with O in torques and Cu in orange. g) SEM EDX results before electrolysis of CF with O in torques, C in yellow and Cu in

orange. h) SEM EDX results after electrolysis of CF with O in torques, C in yellow and Cu in orange.

Appendix

172

Figure S 58: SEM-EDX measurements of CuO/CF before and after electrolysis. a) and b) SEM images of CuO/CF before

electrolysis at 1000x and 5000x magnification. c) and d) SEM images of CuO/CF after electrolysis (ae) at 1000x and 5000x

magnification. e) SEM EDX images before electrolysis of CuO/CF with O in torques and Cu in orange. f) SEM EDX images

after electrolysis of CuO/CF with O in torques and Cu in orange. g) SEM EDX results before electrolysis of CuO/CF with O

in torques, C in yellow, F in pink and Cu in orange. h) SEM EDX results after electrolysis of CuO/CF with O in torques, C in

yellow, F in pink and Cu in orange.

Appendix

173

Figure S 59: SEM-EDX measurements of CuO/NiO/CF before and after electrolysis. a) and b) SEM images of

CuO/NiO/CF before electrolysis at 1000x and 5000x magnification. c) and d) SEM images of CuO/NiO/CF after electrolysis

(ae) at 1000x and 5000x magnification. e) SEM EDX images before electrolysis of CuO/NiO/CF with O in torques, Ni in red

and Cu in orange. f) SEM EDX images after electrolysis of CuO/NiO/CF with O in torques, Ni in red and Cu in orange. g)

SEM EDX results before electrolysis of CuO/NiO/CF with O in torques, C in yellow, F in pink, Ni in red and Cu in orange. h)

SEM EDX results after electrolysis of CuO/NiO/CF with O in torques, C in yellow, F in pink, Ni in red and Cu in orange.

Appendix

174

Figure S 60: SEM-EDX measurements of CuO/Fe2O3/CF before and after electrolysis. a) and b) SEM images of

CuO/Fe2O3/CF before electrolysis at 1000x and 5000x magnification. c) and d) SEM images of CuO/Fe2O3/CF after electrolysis

(ae) at 1000x and 5000x magnification. e) SEM EDX images before electrolysis of CuO/Fe2O3/CF with O in torques, Fe in

purple and Cu in orange. f) SEM EDX images after electrolysis of CuO/Fe2O3/CF with O in torques, Fe in purple and Cu in

orange. g) SEM EDX results before electrolysis of CuO/Fe2O3/CF with O in torques, C in yellow, F in pink, Fe in purple and

Cu in orange. h) SEM EDX results after electrolysis of CuO/Fe2O3/CF with O in torques, C in yellow, F in pink, Fe in purple

and Cu in orange.

Appendix

175

Figure S 61: SEM-EDX measurements of CuO/Co3O4/CF before and after electrolysis. a) and b) SEM images of

CuO/Co3O4/CF before electrolysis at 1000x and 5000x magnification. c) and d) SEM images of CuO/Co3O4/CF after

electrolysis (ae) at 1000x and 5000x magnification. e) SEM EDX images before electrolysis of CuO/Co3O4/CF with O in

torques, Co in blue and Cu in orange. f) SEM EDX images after electrolysis of CuO/Fe2O3/CF with O in torques, Co in blue

and Cu in orange. g) SEM EDX results before electrolysis of CuO/Co3O4/CF with O in torques, C in yellow, F in pink, Co in

blue and Cu in orange. h) SEM EDX results after electrolysis of CuO/Co3O4/CF with O in torques, C in yellow, F in pink, Co

in blue purple and Cu in orange.

Appendix

176

Figure S 62: Thin film XRD of CuO on a GC disc electrode before and after electrolysis and of CF. a) XRD patterns of

powder CuO drop casted on a GC disk electrode before (dark red) and after electrolysis (ae) (light red) and the GC disk electrode

with Nafion binder as a background (grey), references are given in orange for CuO tenorite and red for metallic copper. The

transparent boxes mark the specific 2θ angles for CuO tenorite (orange) and metallic copper (red). b) XRD patterns of CF

(black) and the metallic copper reference (red).

Figure S 63: RDE three-electrode measurements and powder XRD characterization of CuO/MOx compared to

physically mixed CuO and MOx. a)-c) RDE measurements of the mixed metal oxides CuO/NiO (red), CuO/Fe2O3 (purple),

and CuO/Co3O4 (blue) compared to the physically mixed equivalents (green). All RDE LSV measurements were taken between

0 VRHE to -0.6 VRHE at a scan rate of 10 mV s-1 in 0.1 M KOH with (solid line) and without (dashed line) 10 mM HMF at

2500 rpm with an electrode surface area of 0.19 cm2 and a catalyst loading of 0.04 mg. All measurements are 100% manual

internal resistance (IR) corrected. d)-f) Powder XRD measurements of the mixed metal oxides CuO/NiO (red), CuO/Fe2O3

(purple), and CuO/Co3O4 (blue) compared to the physically mixed equivalents (green), CuO (black), and the CuO Tenorite

reference (orange).

Appendix

177

Figure S 64: Schematic drawing of the undivided three-electrode cell setup. Showing the configuration of the UTEC from

left to right, with Pt-mesh counter electrode (CE), 1 cm-2 CF based working electrode (WE) and reversible hydrogen electrode

(RHE) reference electrode (RE) and an electrolyte volume of 50 ml.

Figure S 65: Undivided three-electrode cell (UTEC) loading study of CuO/NiO/CF. a) Showing the overpotential at

10 mA cm-2 for different catalyst loadings without (black) and with (green) 10 mM HMF. b) Showing the overpotential at

10 mA cm-2 (mass corrected) for different catalyst loadings without (black) and with (green) 10 mM HMF. The loading of

0 mg cm-2 is out of the plotted range.

Appendix

178

Figure S 66: Activity comparison of CuO/MOx, CuO/CF, MOx/CF and CuO/MOx/CF. Comparing a) CuO/NiO (red),

CuO/CF (black), NiO/CF (light red) and CuO/NiO/CF (dark red), b) CuO/Fe2O3 (purple), CuO/CF (black), Fe2O3/CF (light

purple) and CuO/ Fe2O3/CF (shiny purple) and c) CuO/Co3O4 (blue), CuO/CF (black), Co3O4/CF (light blue) and CuO/

Co3O4/CF (dark blue). Reaction conditions are the same as in Figure 3 for RDE and UTEC measurements. All measurements

are 100% manual internal resistance (IR) corrected

Figure S 67: Electrochemical cell setup and the corresponding HFR results. a) 0.1 M KOH with 10 mM HMF as catholyte

(100 ml, recycled), 0.1 M KOH as anolyte (100 ml, recycled), 5 cm2 electrode area, CuO/MOx/CF as cathode (black), nickel

foam (NF) as anode (metallic grey), FAA-3-PK membrane (yellow), PTFE gaskets (white) and a flow rate of 25 ml min-1.

Based on the same design from earlier work.155 b) High frequency resistance results before and after constant current

electrolysis. Error bars with relative errors of 0.4-2%.

Appendix

179

Figure S 68: MEA-Flow-Cell performance measurements of CuO/MOx/CF electrodes (with error bars). a)-e) Faradaic

efficiencies for H2 (strong blue), BHMF (light blue), MFA (green) and other products (grey) of the different spray-coated

CuO/MOx/CF catalysts. f) Faradaic efficiencies and HMF conversion (yellow) are calculated for every 15 min time interval

over 60 min using CuO/NiO/CF as a catalyst. Product color code stays as in a)-e). g)-i) Scatter plot of the product selectivity

preference. HMF conversion over MFA/BHMF selectivity ratio, calculated by SMFA/SBHMF for all CuO/MOx catalysts on CF at

different current densities for 30 min. The color code stays the same as before. Cell reaction conditions: 0.1 M KOH with

10 mM HMF as catholyte (100 ml, recycled), 0.1 M KOH as anolyte (100 ml, recycled), 5 cm2 electrode area, nickel foam (NF)

as anode and a flow rate of 25 ml min-1, at 10-30 mA cm-2 for 30 min. Error bars with relative errors of 2-4% for FEproducts, 3-

5% for FEH2 and 1-3% for XHMF were added.

Appendix

180

Figure S 69: MEA-Flow-Cell activity measurements of CuO/MOx/CF electrodes. a)-e) Cell potential (red), current

densities (black) and H2 rate without (neon blue) and with HMF (neon green) of the spray-coated CuO/MOx/CF catalysts pure

CF (orange), CuO (black), NiO (red), Fe2O3 (purple) and Co3O4 (blue). Cell reaction conditions: 0.1 M KOH without 10 mM

HMF as anolyte (100 ml), and 0.1 M KOH with 10 mM HMF as catholyte (100 ml), 5 cm2 electrode area, nickel foam (NF) as

anode and a flow rate of 25 ml min-1, at 10-30 mA cm-2.

Figure S 70: Comparison with the literature presented in a radar plot. Performance and reaction parameters like the HMF

conversion (XHMF), selectivity towards BHMF (SBHMF) and MFA (SMFA), the potential at 10 mA cm-2 (E10mA/cm2), pH and initial

HMF concentration (cHMF) are compared between CuO/Fe2O3/CF and data from the literatur.96,99,101

Appendix

181

Table S 12: ICP-OES results of the bimetallic catalysts. The composition of powdered CuO/NiO, CuO/Fe2O3 and

CuO/Co3O4 are given in mol% and wt%.

Catalyst CuO/ MOx

M/Cu mol%

Cu wt%

CuO/NiO

10.26

54.32

CuO/Fe2O3

10.29

51.12

CuO/Co3O4

10.72

55.20

Table S 13: BET results in m2/g and in cm2/0.04 mg (RDE electrode catalyst loading).

Surface Area

CuO

com.

CuO

NiO

Fe2O3

Co3O4

CuO/NiO

10 mol%

CuO/Fe2O3

10 mol%

CuO/Co3O4

10 mol%

[m2/g]

11.06

21.78

183.89

95.98

114.79

68.54

84.20

87.97

[cm2/0.04mg]

0.04

0.08

0.73

0.38

0.45

0.27

0.33

0.35

Appendix

182

Table S 14: Performance parameters of all catalysts at different current densities over 30 min in the MEA flow cell.

Catalyst

[mA

cm-2]

XHMF

[%]

YBHMF

[%]

SBHMF

[%]

FEBHMF

[%]

YMFF

[%]

SMFF

[%]

FEMFF

[%]

YMFA

[%]

SMFA

[%]

FEMFA

[%]

rate

FEH2

[%]

0.85

16.73

35.76

CuO/CF

825

22.33

48.63

CuO/NiO/CF

9.9

18.61

50.01

CuO/Fe2O3/CF

0.43

16.33

32.33

CuO/Co3O4/CF

1.45

17.8

34.09

CuO/Fe2O3/CF

NiFe(-Cl-)-

LDH@NF

Appendix

183

10.4 Supporting Information Chapter 7

This subchapter has been reproduced (adapted) with permission from Hauke, P., Brückner, S.

& Strasser, P. Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a

Noble Metal-free Bipolar Membrane Electrolyzer. ACS Sustainable Chemistry & Engineering,

Chemical Society.206

SPFDCA =𝑛𝐹𝐷𝐶𝐴

𝑛𝐻𝑀𝐹,𝑖𝑛𝑖𝑡𝑖𝑎𝑙 ∙100% (Equation 56)

Full-Cell measurements and fabrication of the applied electrodes for CO2RR.

40 mg NiNC-IMI catalyst, 0-60 mg PTFE powder (Sigma Aldrich, 1µm particle size), 2 ml

ethanol were mixed and sonificated using sonifier horn for 15 mins. The prepared ink was

sprayed coated onto the commercial gas diffusion layer from Freudenberg (H14C7), on the

micro porous layer, at 80 °C.

The cathode GDE (5 cm2), membrane (Fumasep FBM BPM, Fuma Tech), and the anode

material (5 cm2, NiFe(-Cl-)-LDH electrodes) are layer by layer assembled in the electrolyzer. Both

cathode and anode are located and stabilized in PTFE gaskets (thickness: 120 or 800 microns,

respectively; window size: 5 cm2). After that, the cathode/membrane/anode layers are

compressed within the electrolyzer plates (serving as flow fields and electron conductor) to

ensure the reactor tightness, the electrode charge conductivity, and the ionic conductivity in the

MEA.

In the MEA cell testing protocol, no catholyte is applied. The humidified CO2 gas flow is driven

by a mass flow controller (MFC, Bronkhorst; flow rate: 20 ml min-1) into the cathode flow field

for the reaction. After the reaction, the product stream (out from the reactor) is purged through

the condenser and drier, then mixed with a N2 bleeding (5 ml min-1) for GC analysis. In the

anode chamber, 1M KOH solution is recycled at a 20 ml min-1 flow rate.

Appendix

184

All the CO2RR performance screening in this report is done at ambient temperature and

pressure. The anode and cathode are connected with a potentiostat (Bio-Logic SP150 with a

booster channel) to control the current densities. The reference electrode accompanies the anode

for a two-electrodes configuration, and potentio- and galvano- electrochemical impedance

spectra (PEIS and GEIS modules) are carried out to measure the cell-resistance. The CO2RR

performance is assessed at various current densities or at one constant current density. Each

current step is held for 15 min before moving on to the next current setting, and the gas stream

is injected in the GC sample loop at 14.5 min of each current step. If we use one constant current

density then we analyze the product stream every 15 min.

When driving the CO2RR in the flow cell, especially at large current densities or over longer

time, significant (bi-)carbonate precipitation should be considered, which cause obvious flow

rate depletion and feed blockage. Therefore, the N2 bleeding line implemented in our scheme

with a defined flow rate (5 ml min-1) serves as an internal standard flow rate.

Product analysis for CO2RR

A Shimadzu 2014 on-line GC is utilized for product quantification. The gas stream is separated

by the Hayesep Q + R columns and then analyzed by the TCD (Thermo Conductivity Detector)

and FID (Flame Ionization Detector). The TCD detects the volume percentage (%VOL) of the

H2 product, and the FID measures the CO after being methanized. In all tests, no liquid product

is found after the electrolysis.

At high current densities the alkaline environment reduces the CO2 outlet flow which would

cause big errors in the product analysis. We use a N2 bleeding after the reactor and in front of

the GC and use the N2 signal as in-line calibration. The actual total gas flow rate into the GC

sample loop could be calculated as Equation 57 and can be used to calculate FE, -jCO, etc.

V󰇗Total = V󰇗N2+V󰇗reactor =V󰇗N2

CN2 (Equation 57)

V󰇗Total: total stream flow rate in GC/ml s-1

V󰇗N2: defined N2 bleeding flow rate/ml s-1

Appendix

185

V󰇗reactor: reactor product stream flow rate/ml s-1

CN2: N2 concentration detected by GC/%VOL

Calculations of the production rate (Equation 58), partial current density (Equation 59), and

faradaic efficiency (Equation 60) are given below.

n󰇗Product =V󰇗Total × CProduct

A × VMOL (Equation 58)

n󰇗Product: geometric reaction rate of each product/ mol cm-2 s-1

CProduct: product concentration (volumetric ratio) from GC/ %VOL

A: geometric area of the electrode / cm2

VMOL: volume of gas per mole at atm / ml mol-

jProduct =n󰇗Product ×F×z (Equation 59)

jProduct: partial current density of each product / mA cm-2

F: faradaic constant / C mol-1

z: charge transfer per product molecule

FEProduct = jProduct

jTotal ×100% (Equation 60)

FEProduct: faradaic efficiency of each product / %

jTotal: total current density/ mA cm-2

Appendix

186

To gain more information about the CO2 consumption we calculate the carbon crossover

coefficient from Equation 61.

CCC=V󰇗H2+V󰇗CO2,in−V󰇗out

V󰇗CO+V󰇗H2 (Equation 61)

V󰇗H2: volumetric hydrogen flow out of the cell/ml s-1

V󰇗CO: volumetric carbon monoxide flow out of the cell/ml s-1

V󰇗CO2,in: volumetric carbon dioxide flow in the cell/ml s-1

V󰇗out: volumetric total outlet flow/ml s-1

The single pass conversion towards CO is calculate according Equation 62.

SP= 𝑉󰇗𝐶𝑂

𝑉󰇗𝐶𝑂2,𝑖𝑛 ×100 (Equation 62)

The Lambda value is calculated according:

stoich =𝐶𝑂2 𝑖𝑛

𝐶𝑂2 𝑐𝑜𝑛 = 𝑉󰇗𝐶𝑂2, 𝑖𝑛

(𝐶𝐶𝐶+1)∙𝑉󰇗 𝐶𝑂 + (𝐶𝐶𝐶) ∙ 𝑉

󰇗𝐻2 (Equation 63)

Standard potential and Valorization calculation:

ΔE0= Eanode

0 − Ecathode

0 (Equation 64)

𝐸𝑚𝑒𝑚𝑏𝑟𝑎𝑛𝑒 𝑐𝑜𝑟𝑟𝑒𝑐𝑡𝑒𝑑 = 𝐸𝑐𝑒𝑙𝑙 −0.8𝑉 (Equation 65)

ValorizationFDCA =MFDCA ∙ STYFDCA ∙toperation ∙(PriceFDCA −PriceHMF) (Equation 66)

Appendix

187

Figure S 71: Evolution of the continuous HMF and CO2 electrolyzer process. Evolution of the combined electrolyzer cell

starting with a semi-batch, electrolyte recycling HMF oxidation setup (a), over separated considered continuous HMF oxidation

(b) and CO2 reduction (c) to a combined continuous electrolyzer (d). The anode NiFe(-Cl-)-LDH@NF is given in porous grey,

Pt-C reference cathode in solid grey, the NiNC-IMI PTFE60% cathode in porous black and the bipolar membrane in yellow and

orange.

Supplementary Discussion 7

A Semi-Batch HMFOR electrolyzer cell. To design the catalytic reaction environment of an

active and selective paired HMFOR anode, we first explored the coupling of the HMF oxidation

catalysis with the hydrogen evolution reaction (HER) in BPM semi-batch environment

(Figure S 71 a and Figure S 72 a). The half-cell reaction processes of this cell read

Anode: C6H6O3 + 6OH- → C6H3O5 + 4H2O + 6e- E0= -0.71 VSHE (pH=14)

Cathode: 2H2O + 2e- → 2OH- + H2 E0= -0.83 VSHE (pH=14)

To this end, we grew a catalytically active anion-tuned layered double hydroxide

electrocatalysts, NiFe(-Cl-)-LDH, directly on a high surface-area Nickel foam (referred to as

NiFe(-Cl-)-LDH@NF) for use as anode, and we spray coated a high surface-area carbon-

supported Pt nanoparticle catalyst, Pt-C, onto carbon paper (Pt-C@CP) for use as cathode.

HMF/KOH feeds of the semi batch cell were recycled. The BPM (Fumasep FBM BPM, Fuma

Tech) was configured such that the cationic acid side faced the cathode and the anionic alkaline

side the anode. Polarization curves shown in Figure S 72 b evidenced that the use of the BPM

resulted in a significant positive cell voltage shift of ~0.8V required to maintain a potential

difference to split water molecules at the rate set by current or potential. BPM voltage loss

correction did not fully recover the reference polarization curve obtained when a AEM was

used (solid blue line to dashed blue line). We can attribute this to the high frequency resistance

(HFR) increase by 0.38 Ω associated with the use of the BPM. The cell exhibited a high HMF

Appendix

188

conversion (XHMF, yellow, Equation 38), FDCA Selectivity (SFDCA, green) and Faradaic

Efficiency (FEFDCA, blue, Equation 40) at 25 mA cm-2 in 10 mM HMF solution (Figure S 72c,

Table S 15, Figure S 73) which fully matched the AEM performance. These results suggested

the viability of BPMs for the HMFOR. FEFDCA and the FDCA space-time yield (STY, grey,

Equation 41) declined with increasing current densities (Figure S 72c, Table S 15, Figure S 73).

In order to exclude HMF mass transport limitations, we increased the HMF concentration to

20, 25 and 30 mM at a current density of 50 mA cm-2 (Figure S 72d, Table S 15, Figure S 74).

20 mM HMF pushed the FEFDCA to 84% and STY by 56%, at a constant XHMF of close to 80%.

A decline of the STY (-46%) and of SFDCA (-58%) occurred at 30 mM HMF concentration. We

note that at 30 mM HMF the sudden increase in HMF conversion at lower STY suggests the

competitive formation of unreactive, polymeric compounds, such as humins.41,42

Appendix

189

Figure S 72: Semi-Batch HMF oxidation operation condition optimization. a) MEA constellation and electrolyzer

operation-mode (I Semi-Batch) overview. b) Linear scan voltammograms (LSV) comparison of the AEM (FAA-3-PK-130)

and BPM (Fumasep FBM) using NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt nanoparticles (loading of 0.1 mg cm-2)

cathodes (5 cm2) in 1 M KOH used as catholyte and 10 mM HMF (HMF) in 1 M KOH used as anolyte. Catholyte and anolyte

volume of 100 ml (recycled). Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials

1.0 Vcell – 3 Vcell , scan rate 10 mV s-1. The black arrows show the 0.8 V voltage delay caused by the BPM. Green, yellow and

orange brackets show the different HMF reaction regions in AEM electrolysis. c) and d) current density and concentration

dependency of STYFDCA (grey), XHMF (yellow), SFDCA (green), FEFDCA (blue) over 1h electrolysis. Electrolysis parameters:

BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M

KOH used as catholyte and 10-30 mM HMF in 1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml (recycled),

flow rate 20 ml min-1 and current densities of 25-75 mA cm-2. Standard deviation between 2-3%.

Appendix

190

Figure S 73 shows increasing XHMF with increasing time and increasing current density. At the

same time, a slight increase in SFDCA at 25 and 50 mA cm-2 with longer reaction time can be

seen. At 75 mA cm-2, the positive trend for FDCA selectivity is even more pronounced. In

addition, the FEFDCA drops significantly with time at higher current densities, although a

constantly high FEFDCA could still be demonstrated for 25 mA cm-2.

Figure S 73: Time resolved current density study of a Semi-Batch Electrolyzer. Time resolved performance of a semi

batch electrolyzer is shown as function of the applied current density. Plotted are the conversion of XHMF (yellow), selectivity

SFDCA (green), and faradic efficiency FEFDCA (blue) in percentage (z-axis) over 1h electrolysis (x-axis) and the current density

in mA cm-2 (y-axis). Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt nanoparticles

(loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 10 mM HMF in 1 M KOH used as anolyte.

Catholyte and anolyte volume of 100 ml (recycled), flow rate 20 ml min-1 and current densities of 25-75 mA cm-2. Standard

deviation between 2-3%.

Appendix

191

Figure S 74 shows increasing XHMF with increasing time and increasing HMF concentration.

An exception is the highest HMF concentration (30 mM) which shows a comparatively high

and constant HMF conversion over time. At the same time, SFDCA decreases with increasing

HMF concentration, but remains roughly constant over time for the respective concentration.

In addition, within a concentration, the SFDCA drops significantly over time, but generally

increases with increasing HMF concentration.

Figure S 74: Time resolved concentration study of a Semi-Batch Electrolyzer. Time resolved performance of a semi batch

electrolyzer is shown as function of the initial HMF concentration. Plotted are the conversion XHMF (yellow), the selectivity

SFDCA (green), and the faradaic efficiency FEFDCA (blue) in percentage (z-axis) over 1h electrolysis (x-axis) and the initial HMF

concentration in mM (y-axis). Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt

nanoparticles (loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 10-30 mM HMF in 1 M KOH used

as anolyte. Catholyte and anolyte volume of 100 ml (recycled), flow rate 20 ml min-1 and current densities of 50 mA cm-2.

Standard deviation between 2-3%.

Appendix

192

Figure S 75: CO2 reduction with a BPM configuration setup. CO2 feed is connected to an MFC to control the flow rate and

humidified by a water bubbler before feed to the cell. After the cell liquid water will be collected by a condenser and after dried

through a column mol sieve. In front of the GC our nitrogen inline calibration will be added to the dry product stream. At the

anode we cycle 1M KOH

Figure S 76: CO2 reduction with a BPM configuration and combined continuously HMFOR setup. CO2 feed is connected

to an MFC to control the flow rate and humidified by a water bubbler before feeding to the cell. After the cell liquid water will

be collected by a condenser and after dried through a column mol sieve. In front of the GC our nitrogen inline calibration will

be added to the dry product stream. At the anode, we pump 1M KOH and 10mM HMF through the cell and collect the product

containing liquid to take HPLC samples.

Appendix

193

Figure S 77: Cell Potential comparison of CO2RR with and without HMFOR at different current densities. a) showing

LSVs of CO2RR without (black) and with (blue) HMFOR. b) Showing the cell voltage over time at current densities of 50, 100

and 200 mA cm-2 for CO2RR without (black) and with (blue) HMFOR. Reaction conditions: 20 ml min-1 humidified CO2, 1M

KOH and 10 mM HMF with 2 ml min-1 flowrate and NiFe(-Cl-)-LDH@NF anode at RT. Standard deviation between 2-3%.

Supplementary Discussion 8: Stability and Transport Analysis of the Paired CO2RR/

HMFOR Electrolyzer Dynamics

During our electrolyzer cell stability analysis, we noticed dynamic electrolysis events (see

supporting information Figure S 78 a), where the CO selectivity suddenly shot up momentarily

to unusually high values. We checked the FECO before and after the event and noticed a

declining CO selectivity after the event, yet a lower-than-expected rise in FEH2 (Figure S 78 b).

To diagnose and better understand the catalytic/mass transport events responsible for this

electrolyzer cell behavior, we evaluated the kinetic carbon crossover coefficient, CCC

(Equation 60).225 The CCC is measured as the ratio of the non-catalytic CO2 loss due to

carbonate formation relative to the catalytically generated alkalinity.

As a diagnostic tool, the CCC reveals the dominant ion transport's nature and indicates catalyst

layer mass transport limitations. In the present cell event, the CCC exhibited a sudden

discontinuity with a subsequent monotonic rise (Figure S 79). This CCC behavior suggested a

sudden flooding event in parts of the catalyst layer, where protons could no longer decompose

carbonate formed from CO2 and alkalinity. This caused increased (non-catalytic) CO2

utilization. Indeed, the experimental CO2/CO ratio suddenly decreased in the wake of flooding

(Figure S 80). As the inline N2 reference bleed is entering past the electrolyzer cell, it is not

affected, causing a larger than usual N2 GC signal.

Appendix

194

Our detailed analysis of the unfavorable dynamic events and the sequence of events leading to

the evolution of detrimental operating dynamic regimes of the paired electrolyzer is illustrated

in detail in Figure S 80 a-e: b) During the initial active and efficient dynamic regime of the

paired electrolyzer cell, protons had good access to the catalyst layer, decomposed (bi)carbonate

and thus regenerated gaseous CO2. Upon sudden flooding or salt blocking of portions of the

catalyst layer without direct access to the proton-supplying BPM, c) the proton supply was

disrupted, which caused consumption of CO2 due to sustained carbonate formation. As the

CO2/CO ratio did not change significantly during the event, we conclude that gases (CO, CO2,

and H2) were momentarily trapped inside the GDE/cell. d) At some point, a sudden release of

trapped gases led to the observed sudden GC-based spike in selectivity. e) Thereafter, a flooded

region spread, as suggested by the monotonically rising CCC value (Figure S 79). Flooding

events are further supported here by the absence of sudden cell voltage changes usually

associated with feed blockage (Figure S 81).

Figure S 78: Stability and degradation of paired CO2RR and HMFOR. a) Faradaic efficiency of CO2RR to CO over 5h.

b) Disproportional faradaic efficiency change over the first 2.5h

Appendix

195

Figure S 79: Carbon cross over coefficient (CCC) diagnosis. Changes in the CCC accurately identify the onset of detrimental

flow regime inside the catalyst layer, such as flooding. Shown are one one measurement without PTFE (brighter red,

corresponding to the measurement in Figure 4b) and one with 60wt% PTFE (darker red, corresponding to measurement in

Figure 4c). The CCC of the sample without PTFE is slowly rising with means the protons have good accessibility to the Anions

(OH-, CO32- and HCO3-) but if we take the selectivity into consideration than we assume that the catalyst layer is flooded with

would decrease the selectivity as shown. On the other site the sample with high PTFE loading provides good selectivity and an

initial comparable low CCC (Region 1). This means the electrode is not flooded but the protons have access to the Anions.

After 2.75h we have an event which change the CCC drastically (Region 2). As the selectivity remains high but the CCC after

3h is increasing. We assume that we flood/wet regions which are not connected to the membrane and would lead to no

accessibility of the protons.

Appendix

196

Figure S 80: Flooding mechanism analyzed using time-resolved CO2/CO ratio measurements during our stability test.

a) With the ratio of CO2 to CO over 5.25h we show that after 2.5h the ratio change which is a sign of non electrochemical

consumption The flooding processes splits in 4 phases. b) At the beginning, only the electrode surface near the membrane is

wet so the protons can neutralize the hydroxide and carbonate. c) After some time, regions that are not connected to the

membrane (blue stripe inside catalyst layer) become flooded. Proton transport in the flooded catalyst layer is suppressed, which

causes CO2 to be converted into carbonate in an acid base reaction. and as we have a disproportional change in FE with

increasing N2 proportion gas is trapped in the GDE. d) After reaching 2.5h the trapped gas will be released which increased

drastically the CO FE (Figure S7). e) The CO2/CO ratio is decreasing which means the reaction is taking place at areas which

are not connected to the membrane and so the protons can not neutralize the anions.

Appendix

197

Figure S 81: Potential change during an Electrolyzer stability test. Potential degradation during 5h stability test is about -

21.74 mV h-1.

Appendix

198

Table S 15: Performance results of the Semi-Batch HMFOR setup (Figure S 71 a) over 1 h electrolysis with standard

deviations between 1-3%. Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt nanoparticles

(loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 10-30 mM HMF in 1 M KOH used as anolyte.

Catholyte and anolyte volume of 100 ml (recycled), flow rate 20 ml min-1 and current densities of 25-75 mA cm-2. Standard

deviation between 2-3%.

Conditions

XHMF [%]

YHFCA [%]

YFDCA [%]

SFDCA [%]

FEFDCA [%]

25 mA cm-2

10mM

100

50 mA cm-2

10mM

75 mA cm-2

10mM

100

50 mA cm-2

20mM

50 mA cm-2

25mM

50 mA cm-2

30mM

Appendix

199

Table S 16: Performance results of the continuous HMFOR setup (Figure S 71 b) over 1 h electrolysis with standard

deviations between 1-3%. Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt nanoparticles

(loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 5-20 mM HMF in 1 M KOH used as anolyte.

Catholyte and anolyte volume of 50 ml, flow rates 5 and 2 ml min-1 and current densities of 50-200 mA cm-2. Standard deviation

between 2-3%.

Conditions

XHMF

[%]

YHFCA

[%]

YFDCA

[%]

SFDCA

[%]

FEFDCA

[%]

SPFDCA

[%]

50 mA cm-2 5mM

100

50 mA cm-2 10mM

100 mA cm-2 5mM

100

100 mA cm-2

10mM

200 mA cm-2 5mM

200 mA cm-2

10mM

200 mA cm-2

20mM

200 mA cm-2

10mM 2ml min-1

200 mA cm-2

20mM

2ml min-1

Appendix

200

Table S 17: Performance results of the combined continuous HMFOR and CO2RR setup (Figure S 71 d) electrolysis

with standard deviations between 2-4%. Electrolysis parameter: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2) with 1M KOH

and 10 mM HMF anolyte with 2 ml min-1 flowrate, NiNC-IMI PTFE60% cathodes with 20 ml min-1 humidified CO2.

Conditions

Ecell

[V]

XHMF

[%]

YHFCA

[%]

YFDCA

[%]

SFDCA

[%]

FEFDCA

[%]

SPFDCA

[%]

200 mA cm-2

10mM

3.53

200 mA cm-2

10mM cooled

3.53

200 mA cm-2

100mM

3.45

Table S 18: Performance results of the CO2RR setup with BPM (Figure S 71c) and combined continuous HMFOR and

CO2RR setup (Figure S 71d) electrolysis compared to an AEM reference225. Electrolysis parameter: BPM,

NiFe(-Cl-)-LDH@NF anodes (5 cm2) with 1M KOH with 20 ml min-1 and 1M KOH+10 mM HMF anolyte with 2 ml min-1

flowrate respectively, NiNC-IMI PTFE60% cathodes with 20 ml min-1 humidified CO2. Standard deviation between 2-3%.

Conditions

Ecell

[V]

FECO

[%]

SPCO

[%]

Lambda

[-]

[%]

200 mA cm-2

AEM OER225

3.14

97.92

27.42

1.89

41.79

200 mA cm-2

OER

3.65

93.18

32.61

2.97

55.57

200 mA cm-2

HMFOR

3.53

97.17

33.16

2.82

54.20

Appendix

201

Table S 19: Energy efficiencies of the different electrolyzer configurations with and without the BPM penalty. Standard

deviation between 2-3%.

Electrolyzer

EE [%]

EEBPM [%]

HMF/H2O

4.13

5.5

CO2/H2O

34.20

43.32

CO2/HMF

32.78

41.98

CO2/HMF 100mM HMF

25.41

29.96

Appendix

202

10.5 List of Figures

Figure 1: Share of renewable energy from the worldwide total energy and the global and German electrical

energy.1,2 ........................................................................................................................................................ 2

Figure 2: Schematic illustration of an alkaline water electrolyzer (AWE), proton exchange membrane water

electrolyzer (PEMWE), and anion exchange membrane water electrolyzer (AEMWE). .............................. 4

Figure 3: Supposed alkaline oxygen evolution reaction (OER) mechanism on a metal substrate (a-e), with the

change in Gibbs free energy (ΔG) for every reaction step. ........................................................................... 7

Figure 4: 5-Hydroxymethylfurfural (HMF) oxidation and reduction reaction pathways. Poly- and Dimerization

(black): undesired side reaction of n HMF molecules and n Protons/electrons. Reductive Ring opening

(dark blue): opening of the furan ring with 6e- and 6H+. Hydrogenation (blue): conversion of HMF to

BHMF with 2e- and 2H+. Hydrogenolysis (blue): conversion of HMF to MF with 2e- and 2H+, conversion

of BHMF or MF to MFA with 2e- and 2H+, and conversion of MFA to DMF with 2e- and 2H+. Oxidation

(yellow): conversion of HMF over HFCA or FDA to FFCA and FDCA (6e- and 6OH-). Carbon atoms

(black), Oxygen Atoms (red), and Hydrogen Atoms (white). This Figure is reproduced from Chapter 6.43 9

Figure 5: Illustration summary of this work's motivation, idea, and scope. Using renewable energy sources for

the electrochemical conversion of HMF at the anode (blue) and cathode (orange) to valuable products. .. 14

Figure 6: Scheme of the electrochemical double layer including the Helmholtz (inner (IHP) and outer Helmholtz

plane (OHP)), Grahame, and Stern models. ................................................................................................ 17

Figure 7: General activation energy diagram of the uncatalyzed and catalyzed reaction between precursors A and

B to product P. ............................................................................................................................................. 23

Figure 8: Transition metal Fermi level dependency on the electrode potential..................................................... 24

Figure 9: Langmuir-Hinshelwood and Eley-Rideal reaction mechanisms. ........................................................... 28

Figure 10: Optimal binding relation based on the Sabatier principle illustrated in a volcano curve. .................... 29

Figure 11: General suggestion of an organic electrode reaction process. ............................................................. 31

Figure 12: SEM, XPS and XRD investigations of NiFe(-CO32-)-LDH powder, Nickel Foam (NF) and NiFe(-A-)-

LDH grown on NF. a-i) SEM images of the prepared anodes, NiFe(-CO32-)-LDH@NF (red), NiFe(-Cl)-

LDH@NF (blue), NiFe(-ClO4)-LDH@NF (green). j) Comparison of the XPS Ni 2p regions of uncoated

NF (black) and NiFe(-CO3)-LDH@NF (red). k) Cl 2p XPS spectra of NiFe(-Cl-)-LDH at NF. l) XRD

investigations of powder NiFe(-CO32-)-LDH including the shifts of the (003) and (006) diffraction

reflections during anion exchange and reversible structure change to NiFe(-Cl-)-LDH@NF and NiFe(-

CO32-)-LDH. The arrows indicate the shifts from CO32- to ClO4- (green) and the back shift to CO32- after

aging in KOH (red). ..................................................................................................................................... 46

Figure 13: Electrochemical impedance spectroscopy (EIS) investigations determining the ECSA of the prepared

NiFe(-A-)-LDH@NF and NF in a three-electrode setup (3LC) and electrocatalytic OER reactivity and

stability of uncoated NF, compared to NiFe(-A-)-LDH@NF anodes. Nyquist plots of a) NF, b) NiFe(-

CO32-)-LDH@NF, c) NiFe(-Cl-)-LDH@NF and d) NiFe(-ClO4-)-LDH@NF recorded at 1.53 V, 1.58 and

1.63 V vs RHE. e) Calculated ECSA of uncoated NF (black) and the prepared anodes, NiFe(-CO32-)-

LDH@NF (red), NiFe(-Cl-)-LDH@NF (blue) and NiFe(-ClO4-)-LDH@NF (green). f) iR-corrected linear

Appendix

203

sweep voltammetry (LSV) g) ECSA- and iR-corrected LSVs of uncoated NF (black), NiFe(-CO32-)-

LDH@NF (red), NiFe(-Cl-)-LDH@NF (blue) and NiFe(-ClO4-)-LDH@NF (green). Stability investigations

of h) uncoated NF, i) NiFe(-CO32-)-LDH@NF, j) NiFe(-Cl-)-LDH@NF and k) NiFe(-ClO4-)-LDH@NF

anodes. The measurements were conducted in 0.1 M KOH at RT using three-electrode setup. ................. 49

Figure 14: Activity tests and trends of NiFe(-A-)-LDH@NF and uncoated NF measured in AEMWE single cell.

(a) Comparison of the polarization curves of NiFe(-A-)-LDH@NF CO32- (red) and Cl- (blue) with uncoated

NF (black), dashed lines denote as iR-corrected polarization curves. (b) comparative activity improvement

factors 𝐟=𝐣𝐠𝐞𝐨𝐦.,𝐜𝐚𝐭 𝐚𝐭𝟏.𝟓𝟐𝐕𝐣𝐠𝐞𝐨𝐦.,𝐍𝐅 𝐚𝐭𝟏.𝟓𝟐𝐕 of anion tuning in three-electrode liquid cell

(3LC) at 25°C, 60°C, AEMWE single cell (AEM Cell) and blowup (same y-axis labeling but different

scale) of the AEM cell. Data for the AEM cell are iR- and HER corrected (Figure S18 c). f was calculated

at the highest possible iR-corrected potential (EiR=1.52 VRHE) measured in 3LC at 60°C (Figure S15). .... 52

Figure 15: Electrochemical conversion of 5-Hydroxymethylfurfural over NiFe(-Cl-)-LDH@NF in a zero-gap

flow cell electrolyzer. Reaction pathways of the alkaline 5-HMF oxidation via 5-Hydroxyfuran-2-

carboxylic acid (HFCA) or Furan-2.5-dicarbaldehyde (FDA) to 5-Formyl-2-furancarboxylic acid (FFCA)

and the desired final product 2.5-Furandicarboxylic acid (FDCA). Scheme of the zero-gap MEA-type

electrolyzer with PTFE gaskets (white), Anodic and Cathodic electrodes (ocher and black, respectively),

and membrane (yellow). The serpentine flow fields are a direct part of the current collectors. .................. 58

Figure 16: Structural Characterization of bimetallic LDH catalysts: Characterization of NiX(-CO32-)-LDH nano

powders and directly grown NiX(-CO32-)-LDH on Ni nanofoam. a) XRD pattern of powdered NiFe(-CO32-

)-LDH (red), NiMn(-CO32-)-LDH (blue), NiCo(-CO32-)-LDH (green), and NiV(-CO32-)-LDH (purple). As a

representative reference rhombohedral Hydrotalcite, JCPDS: 00-014-0191 was added. b)-e) SEM images

of NiFe(-CO32-)-LDH), NiMn(-CO32-)-LDH, NiCo(-CO32-)-LDH and NiV(-CO32-)-LDH directly grown on

Ni NF at 5000x magnification. The color code corresponds to a) (clean NF is shown in Figure 17 c). ...... 61

Figure 17: Structural Tuning of the Layered Double Hydroxide Catalysts: Characterization of the interlayer

distance and morphology of NiFe(-A-)-LDH and NiFe(-A-)-LDH@NF after anion exchange. a) Crystal

structure and increasement of the interlayer distance during the anion exchange steps with chloride and

perchlorate. b) XRD pattern of powdered NiFe-LDH (red), NiFe(-Cl-)-LDH (blue), NiFe(-ClO4 -)-LDH

(green) visualizing the (003) reflection shift within the anion exchange. c)-f) SEM images of the different

catalysts at 5000x magnification. The color code corresponds to b). Additionally, pure NF (black) was

added as a reference. .................................................................................................................................... 62

Figure 18: Single Electrolyzer Cell Performance of the new NiX(-A-)-LDH@NF catalysts: a)/b) Flow

Electrolyzer Cell based linear scan voltammograms (LSV) using NF and NiX(-A-)-LDH@NF

(X=Fe,Mn,Co,V; A=CO32-,Cl-,ClO4-) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2)

cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used as catholyte and 10 mM HMF (HMF) in 0.1

M KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml

min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell , scan rate 10 mV s-1, the highlighted green

box denotes the selective HMFOR potential region. c)/d) Catalytic performance overview of the set of

NiX(-A-)-LDH@NF (X=Fe,Mn,Co,V; A=CO32-,Cl-,ClO4-) catalysts using consistent color coding as in a)

and b). 5-HMF Conversion, X, FDCA yield, Y, FDCA selectivity, S, and Faradaic efficiency, FE, are

Appendix

204

given on a scale from 0-100%, while the current density at +1.56 Vcell is plotted from 0-20 mA cm-2 in c)

and 0-25 mA cm-2 in d). ............................................................................................................................... 68

Figure 19: Electrolyzer Cell Performance and Stability: Evaluation of NiFe(-Cl-)-LDH@NF in the MEA cell

setup and DEMS. a) Variation of KOH electrolyte concentration with NiFe(-Cl-)-LDH@NF anodes, C

supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M

KOH used as catholyte and 0.1 M KOH (dotted line), 10 mM HMF in 0.1 M KOH (light blue) and 10 mM

HMF in 1 M KOH (dark blue) used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis

parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10

mV s-1 b) Real-time mass spectrometric Linear Sweep Voltammograms (MSCVs) of the NiFe(-Cl-)-

LDH@NF electrocatalyst. The data was obtained in a custom-made DEMS flow cell setup in 0.1 M KOH

and 10 mM HMF in 0.1 M KOH (flow rate: 10 µl s-1) at a scan rate 5 mV s-1 from 0- +2,0 VRDE. Plotted

are signals of m/z= 32 of 16O 16O over the time resolved potential (Et). Corresponding Diagrams are given

in the supporting information (Figure S 47/Figure S 48). c) 10 h stability test with 5-HMF conversion and

FDCA yield for NiFe(-Cl-)-LDH@NF in 0.1 M KOH with 10 mM 5-HMF. Catholyte and anolyte volume

of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, at 1.56 Vcell. Every 2 h HPLC samples were

taken and the electrolyte was replaced. ....................................................................................................... 71

Figure 20: Comparison between the FDCA product Space Time Yield achieved using the anion-tuned NiFe(-Cl-

)-LDH@NF catalyst deployed in paired H2-FDCA electrolyzer anodes (red data) and previous work.

FDCA Space-Time-Yield was plotted over the electrode potential in 0.1 M KOH (left, triangle) and 1 M

KOH (right, diamonds).52,59-61,63,64,70,71,73,76,79,81,84,167-170,178-181 Color code denotes a rising STY from blue to

red. The references in the figure are kept original. ...................................................................................... 72

Figure 21: HMF reaction pathways and two-phase Cu based bimetallic oxide catalyst concept for enhance

HMFRR. a) Poly- and Dimerization (black): undesired side reaction of n HMF molecules and n

Protons/electrons. Reductive Ring opening (dark blue): opening of the furan ring with 6e- and 6H+.

Hydrogenation (blue): conversion of HMF to BHMF with 2e- and 2H+. Hydrogenolysis (blue): conversion

of HMF to MF with 2e- and 2H+, conversion of BHMF or MF to MFA with 2e- and 2H+, and conversion of

MFA to DMF with 2e- and 2H+. Oxidation (yellow): conversion of HMF over HFCA or FDA to FFCA and

FDCA (6e- and 6OH-). Carbon atoms (black), oxygen atoms (red), and hydrogen atoms (white). a) Co-

precipitated co-existing Tenorite CuO/MOx (Hematite Fe2O3) nanoparticles at the nm-scale are

hypothesized to offer enhanced catalytic HMFRR reactivity after in-situ reduction to OD-Cu/MOx mixed

phase catalysts. Adsorbed hydrogen atoms (Had) are given in white. .......................................................... 76

Figure 22: XRD, XPS, and BET characterization of the CuO/MOx powder catalysts. a)-d) XRD patterns of pure

powder CuO (Tenorite), NiO (Bunsenite), Fe2O3 (Hematite), and Co3O4 (Spinel) XRD with references

given in orange, red, purple, blue, and inserted crystal structures. e) XRD patterns from pure powder CuO

(black) over 10, 30, 67, and 80 mol% NiO to pure NiO (red). f) XRD patterns of pure CuO (black),

CuO/NiO (red), CuO/Fe2O3 (purple), and CuO/Co3O4 (blue). All CuO/MOx are presented with 10mol% of

the second metal oxide. CuO Tenorite reference is given in orange. g) Cu 2p XPS measurements of pure

CuO (orange). h) Ni 2p XPS measurements of CuO/NiO (red). i) Fe 2p XPS measurements of CuO/Fe2O3

(purple). j) Co 2p XPS measurements of CuO/Co3O4 (blue). k) BET results of commercial CuO (orange),

Appendix

205

synthesized CuO (black), pure NiO, Fe2O3, Co3O4 (light red, light purple, light blue), and mixed metal

oxides CuO/NiO (red), CuO/Fe2O3 (purple), CuO/Co3O4 (blue). ............................................................... 80

Figure 23: RDE and undivided three-electrode cell measurements of CuO/MOx and CuO/MOx/CF. a) RDE

measurements of pure CuO (black), NiO (red), Fe2O3 (purple), and Co3O4 (blue). b) RDE measurements of

commercial CuO (orange) and pure CuO (black). c) RDE measurements of mixed metal oxides CuO/NiO

(red), CuO/Fe2O3 (purple), and CuO/Co3O4 (blue). d)-f) BET corrected current densities of the

corresponding plots from a)-c). All RDE LSV measurements were taken between 0 VRHE to -0,6 VRHE at a

scan rate of 10 mV s-1 in 0.1 M KOH with (solid line) and without (dashed line) 10 mM HMF at 2500 rpm

with an electrode surface area of 0.19 cm2 and a catalyst loading of 0.04 mg. All measurements are 100%

manual internal resistance (IR) corrected. g) undivided three-electrode cell (UTEC) measurements of CuO

and CuO/MOx on CF with CF (orange), CuO/CF (black), CuO/NiO/CF (red), CuO/Fe2O3/CF (purple), and

CuO/Co3O4/CF (blue). h) blow up of the UTEC measurements in g) with colored HMFRR selectivities

areas of CF (orange), CuO/CF (black), CuO/NiO/CF (red), CuO/Fe2O3/CF (purple), and CuO/Co3O4/CF

(blue). All UTEC LSV measurements were taken between 0 VRHE to -0,8 VRHE at a scan rate of 10mV s-1

in 0.1 M KOH with (solid line) and without (dashed line) 10 mM HMF without rotation and an electrode

area of 1 cm2 and a catalyst loading of 1 mg cm-2. All measurements are 100% manual internal resistance

(IR) corrected............................................................................................................................................... 84

Figure 24: MEA-Flow-Cell performance measurements of CuO/MOx/CF electrodes. a)-e) Faradaic efficiencies

for H2 (strong blue), BHMF (light blue), MFA (green), and other products (grey) of the different spray-

coated CuO/MOx/CF catalysts. f) Faradaic efficiencies and HMF conversion (yellow) are calculated for

every 15 min time interval over 60 min using CuO/NiO/CF as a catalyst. Product color code stays as in a)-

e). g)-i) Scatter plot of the product selectivity preference. HMF conversion over MFA/BHMF selectivity

ratio, calculated by SMFA/SBHMF for all CuO/MOx catalysts on CF at different current densities for 30 min.

The color code stays the same as before. Cell reaction conditions: 0.1 M KOH with 10 mM HMF as

catholyte (100 ml, recycled), 0.1 M KOH as anolyte (100 ml, recycled), 5 cm2 electrode area, nickel foam

(NF) as anode and a flow rate of 25 ml min-1, at 10-30 mA cm-2 for 30 min. High frequency resistance

results are between 0.7- 1.05 Ω (Figure S 67 b). Relative errors of 2-4% for FEproducts, 3-5% for FEH2, and

1-3% for XHMF resulted (Figure S 68). ......................................................................................................... 88

Figure 25: MEA-Flow-Cell stability measurements of CuO/Fe2O3/CF and combination of HMF reduction and

oxidation. a) Stability measurements of CuO/Fe2O3/CF over 5 cycles (2,5 h) at 20 mA cm-2. HMF

conversion in yellow, BHMF selectivity in turquoise, and MFA selectivity in green. b) cell potential and

performance comparison between non-HMF containing electrolytes on both sides, HMF at the cathode and

HMF on anode and cathode, using CuO/Fe2O3/CF as the cathode and NiFe(-Cl-)-LDH@NF as the anode.

The cell potential is given in pink triangles, BHMF selectivity in turquoise, MFA selectivity in green, and

FDCA selectivity in light green. Cell reaction conditions: 5 cm2 electrode area, flow rate of 25 ml min-1, at

20 mA cm-2 for 30 min. ............................................................................................................................... 89

Figure 26: Scheme of the paired electrochemical conversion of CO2 and HMF. Integration of the state-of-the-art

AEM HMFOR into a BPM CO2RR electrolyzer with two valued products. For the AEM setups standard

potentials are given in an alkaline environment (pH=14). Separated considered a) semi-batch HMF

Appendix

206

oxidation and b) continuous CO2 reduction electrolyzers, c) combined to a continuous CO2/HMF

electrolyzer. ................................................................................................................................................. 95

Figure 27: Continuous HMF oxidation operation condition optimization. a) MEA constellation and electrolyzer

operation-mode (II Continuous) overview. b) and c) Current density and concentration dependency of

STYFDCA (grey), SFDCA (green), FEFDCA (blue), and SPFDCA (red) over 1h electrolysis with a flow rate of 5 ml

min-1 and current densities of 50-200 mA cm-2. d) Bar plot comparing XHMF (yellow), SFDCA (green), FEFDCA

(blue), and SPFDCA (red) at 5 ml min-1 and 2 ml min-1 flow rate over 1 h electrolysis at 200 mA cm-2.

Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt nanoparticles

(loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 5-20 mM HMF in 1 M KOH

used as anolyte. Catholyte and anolyte volume of 50 ml. Standard deviation between 2-3%. .................... 97

Figure 28: Performance of CO2 reduction GDE in a BPM configuration. a) Schematic of a BPM CO2 reduction

configuration. b) Key performance parameter as energy efficiency (purple), faradaic efficiency (blue), and

single pass conversion (red) of the different PTFE loadings. c) CO2 reduction performance in terms of cell

potential and faradaic efficiency at different current densities. 20 ml min-1 humidified CO2, 1M KOH with

2 ml min-1 flow rate and NiFe(-Cl-)-LDH@NF anodes. d) Comparison to previously reported AEM

configuration at a total current density of 200 mA cm-2 and 20 ml min-1 CO2 in the BPM configuration or

25 ml min-1 in the AEM configuration, respectively. Standard deviation between 2-5%.......................... 100

Figure 29: Combination of CO2RR with continuous HMFOR. a) Schematic of the combined CO2RR and

HMFOR cell. b) Performance parameter of the combined cell at different current densities. 20 ml min-1

humidified CO2, 1M KOH and 10 mM HMF with 2 ml min-1 flowrate and NiFe(-Cl-)-LDH@NF anode at

RT. Standard deviation between 2-3%. c) Comparison between refreshing the anolyte (120 ml) for 5h and

cooling the anolyte (600 ml) for 5h and the CO2 reduction of the initial performance and performance after

5h with the data from Table S3. d) Economic evaluation showing the energy efficiency (EE) for

electrolyzers with BPM delay (yellow/orange), without BPM delay (black), and the product valorization in

$ per m2 and year (FDCA in blue and CO in red) for the separated and combined electrolyzers, assuming

an FDCA valorization between 375 and 1548 $ t-1.233,234 .......................................................................... 105

Figure 30: Illustration of the investigated parts leading to efficient and universal usable alkaline HMF water

electrolyzer. ............................................................................................................................................... 107

10.6 List of Figures (Appendix)

Figure S1: SEM images of NF (black framed); a)-c), NiFe(-CO32-)-LDH@NF (red framed); d), NiFe(-Cl-)-

LDH@NF (blue framed); e) and NiFe(-ClO4-)-LDH@NF (green framed); f). ......................................... 127

Figure S2: XPS measurements of NiFe(-CO32-)-LDH powder. a) Survey spectrum, b) Ni 2p region, c) Fe 2p

region, d) O 1s region, e) C 1s region. ....................................................................................................... 128

Figure S3: NiFe(-CO32-)-LDH@NF XPS measurements. a) Survey spectrum, b) Ni 2p, c) Fe 2p, d) O 1s and e) C

1s regions of the prepared electrode. ......................................................................................................... 128

Appendix

207

Figure S4: NiFe(-Cl-)-LDH@NF XPS measurements. a) Survey spectrum, b) Ni 2p, c) Fe 2p, d) O 1s, e) C 1s.

................................................................................................................................................................... 129

Figure S5: NiFe(-ClO4-)-LDH@NF XPS measurements. a) Survey spectrum, b) Ni 2p, c) Fe 2p, d) O 1s, e) C 1s

and f) Cl 2p regions of the prepared electrode. .......................................................................................... 129

Figure S6: The electro catalytic OER activity from NF (black), no precursor except blank Nickel Foam in the

synthesis (light purple) and Fe precursor only (ochre) to NiFe(-CO32-)-LDH@NF (red) for synthesis proof

and catalytic-performance-comparison. The CVs were investigated by using a Pt-mesh as counter

electrode and a reversible hydrogen electrode (RHE) as reference electrode in 50 ml N2-saturated 0.1 M

KOH at room temperature. ........................................................................................................................ 131

Figure S7: RDE investigations of Nickel Foam and NiFe(-CO32-)-LDH@NF. Different reaction times were

investigated (b) blank NF, c) 30 min at 160 °C, d) 60 min at 160 °C and e) 90 min at 160 °C) to optimize

the synthesis and achieve the most active and stable material, a) summarizes the measurements in b) – e).

Measured using a Pt-mesh as counter electrode and a reversible hydrogen electrode (RHE) as reference

electrode in 50 ml N2-saturated 0.1 M KOH at room temperature. ........................................................... 131

Figure S8: XPS investigations of Nickel Foam treated only with Fe precursor (Fe(NO3)3*9H2O) in the synthesis.

a) Survey, b) Ni 2p, c) Fe 2p, d) O 1s and e) C 1s. .................................................................................... 132

Figure S9: SEM images of Nickel Foam treated only with Fe salts during the synthesis. The scale bars are a) 100

µm, b) 1 µm and c) 1 µm. The formed layer appears to be quite dense and thick, typical Nickel foam

characteristics are not visible below the formed Iron layer at the surface. Due to the low contrast of the

sample a low conductivity of the sample is expected. ............................................................................... 132

Figure S10: XPS spectra of Nickel Foam treated with DMF, no additional metal salts were added in the

synthesis. a) Survey, b) Ni 2p, c) Fe 2p, d) O 1s, e) C 1s. ......................................................................... 132

Figure S11: SEM images of Nickel Foam treated without any metal salts during the synthesis. The scale bars are

a) 100 µm, b) 1 µm and c) 1 µm. As expected, a very thin Nickel oxyhydroxide layer forms at the surface

of the Nickle Foam. Nickel Foam characteristics are still visible below the thin layer (b). ...................... 133

Figure S12: Impedance spectra recorded for different NiFe(-A-)-LDH@NF species in a three-electrode setup. a)

Exemplary impedance spectra of NiFe(-CO32-)-LDH@NF highlighting the different impedance regions

(RΩ, RCT and Ra) with inset: Schematic depiction of the used ESB. Resulting impedance values of b) RΩ, c)

RCT and d) Ra in dependence of the applied potential. ............................................................................... 133

Figure S13: Investigations of the electrocatalytic OER activity after 1000 cycles. The prepared and modified

catalysts (NiFe(-CO32-)-LDH@NF, NiFe(-A-)-LDH@NF) in 0.1 M KOH at RT in RDE scale experiment.

LSV curves without iR correction of the prepared electrodes a) NiFe(-CO32-)-LDH@NF, b) NiFe(-Cl-)-

LDH@NF and c) NiFe(-ClO4-)-LDH@NF. .............................................................................................. 135

Figure S14: Color change of NF and NiFe(-CO32-)-LDH@NF during cyclovoltammetry a) NF during potential

sweep does not show change of the color, while b) NiFe(-CO32-)-LDH@NF during potential sweep

indicate the oxidation of the catalyst material (NiFe(-CO32-)-LDH) during cycling, changing from metallic

gray to black (1-3) and back (4-6). Measured using a Pt-mesh as counter electrode and a reversible

hydrogen electrode (RHE) as reference electrode in 50 ml N2-saturated 0.1 M KOH at room temperature.

................................................................................................................................................................... 135

Appendix

208

Figure S15: Investigations of the electrocatalytic OER activity at 60°C. IR-corrected LSVs of NF (black), the

prepared and modified catalysts (NiFe(-CO32-)-LDH@NF (red) and NiFe(-Cl-)-LDH@NF (blue)) in 0.1 M

KOH at RT (50% transparency) and 60°C in RDE scale experiment........................................................ 136

Figure S16: Applied single cell electrolyzer setup used for the AEMWE measurements. a) disassembled and b)

assembled. With 1: Endplate with electrolyte inlets; 2: Current collector, 3: Flow Field (Anode Titanium,

Cathode: Carbon); 3a: Gaskets and 4 MEA: Membrane Electrode Assembly. ......................................... 137

Figure S17: Polarization performance of AEM cells evaluating PTL, T, pH, membranes, cathodes and cathode

Pt-loading. The following operation conditions and cell parameters were changed and modified to gain the

best possible performance with the applied system. a) Different types of cathodic PTLs (Sigracet) were

loaded with 0.1 mg Pt cm-2 to choose the best performing PTL, FAS-50 is applied as membrane. b) The

temperature was adjusted to choose the best operation temperature, here 38 BC with 0.1 mg Pt cm-2 was

applied. c) Influence of the pH was measured by conducting a measurement applying 0.1 M KOH and

0.0001 M KOH. d) FAA-3 based membranes were compared (FAS-50 and FAA-3-30). e) 0.1 mg Pt cm-2

at 38 BC applied as cathode is compared to Nickel powdered on Nickel Foam. f) The influence of Pt

loading (38 BC) is investigated. ................................................................................................................ 137

Figure S18: a)-d) Representation of the uncorrected, iR-corrected, iR- and HER-overpotential-corrected

polarization curves, as well as the overpotentials of the different investigated anodes resulting from the

HER. The measurements correspond to the polarization curves shown in the Main and were carried out at

60°C using 0.1 M KOH. As an anode, the different NiFe-based materials produced were measured and

compared with uncoated NF. 0.1 mg Pt cm-2 spray coated at Sigracet 38 BC was used as cathode. ........ 138

Figure S19: Activity improvement factors plotted over the applied potential a) without iR-correction and b) with

iR and HER correction polarization curves. f is calculated using 𝐟=

𝐣𝐠𝐞𝐨𝐦.,𝐜𝐚𝐭 𝐚𝐭𝐂𝐞𝐥𝐥𝐕𝐨𝐥𝐭𝐚𝐠𝐞𝐣𝐠𝐞𝐨𝐦.,𝐍𝐅 𝐚𝐭𝐂𝐞𝐥𝐥𝐕𝐨𝐥𝐭𝐚𝐠𝐞. f is directly derived from the plots depicted in

Figure S18. ................................................................................................................................................ 139

Figure S20: NF XPS measurements of blank Nickel Foam before the exposure to OER relevant potential regions

as well as the unpurified or Iron containing electrolyte from the Greenlight test station. a) Survey

spectrum, b) Ni2p region, c) O1s region, d) C1s region. ........................................................................... 140

Figure S21: XPS investigations of Nickel Foam after the exposure to OER relevant potential regions and

unpurified electrolyte resulting from stainless steel components used for the Greenlight electrolyzer. a)

Survey spectrum highlighting the Ni2p (b), Fe2p (c), C1s (d) and O1s (e) region. ................................... 140

Figure S22: a) and b) Investigation of the electrocatalytic activity of NF before (black) and after (dark cyan)

exposure to AEMWE single cell testing. c) Nyquist-Plots of NF after AEMWE single cell testing. d)

ECSA comparison of NF, NF after WE and the prepared anodes, the ECSA was determined by EIS

measurements The CVs were investigated by using a Pt-mesh as counter electrode and a reversible

hydrogen electrode (RHE) as reference electrode in 50 ml N2-saturated 0.1 M KOH at RT. ................... 141

Figure S23:a) Exemplary impedance spectra of NiFe(-CO32-)-LDH@NF recorded in a two electrode electrolyzer

cell, b) Results of RΩ, RCT and Ra applying the ESB schematically depicted in c).................................... 142

Figure S 24. Scanning Electron Microscopy images of a) NiFe(-CO32-)-LDH@NF, b) NiMn(-CO32-)-LDH@NF,

c) NiCo(-CO32-)-LDH@NF and d) NiV(-CO32-)-LDH@NF at 1000 x magnification. ............................. 145

Appendix

209

Figure S 25. Scanning Electron Microscopy images of NiV(-CO32-)-LDH@NF at different synthesis parameters.

a)-c) show lower precursor concentration of 0.6 M and 30, 60, 90 min synthesis time (left to right) and d)-

f) show higher precursor concentration of 1.2 M and 30,60,90 min synthesis time (left to right) at 10000 x

magnification. ............................................................................................................................................ 145

Figure S 26. Nyquist plots of NiX-LDH@NF. Exemplary Nyquist plots of Ageom.= 1.5 cm2 a) NiMn- (blue), b)

NiCo- (green), c) NiV(-CO32-)-LDH@NF (purple) in 0.1 M KOH at 1.53 VRHE, 1.58 VRHE and 1.63 VRHE

for the electrochemical surface area (ECSA) evaluation. Solid lines show fits of the raw data (dotted lines).

d) the applied equivalent circuit to evaluate the measured impedance spectra. Element R1 describes the

resistance arising from the electrolyte and electrical connections, elements R4 and Q1 are the reaction

charge transfer resistance (RCT), and Q2 and R5 are associated to be the response of changes in the

coverage of adsorbed species. ................................................................................................................... 146

Figure S 27. Evaluation of 5-HMF in 0.1 M KOH. Showing the decreasing 5-HMF concentration over time (5 h)

in 0.1 M KOH at room temperature (RT). Indicating a relevant autocatalytic reaction from 5-HMF with the

electrolyte after 3h. .................................................................................................................................... 148

Figure S 28. RDE measurements of Ni nanofoam, Nickel Disc (ND), and Carbon Paper (CP). Working

electrodes, WE, geometric area of 0.196 cm2 with and without 10 mM 5-HMF in 0.1 M KOH. The LSVs

(IR corrected) were taken between 1.0 VRHE-1.7 VRHE at a scan rate of 5mV s-1, no rotation was applied.

................................................................................................................................................................... 148

Figure S 29. RDE measurements of NiV(-CO32-)-LDH@NF comparing the activity of different synthesis

parameters (c, t). Synthesis of NiV(-CO32-)-LDH@NF was varied by the holding time (30 min, 60 min,

90 min) and the Vanadium precursor concentration using 0.6 M and higher concentrated 1.2 M solution

corresponding to Figure S2. LSV measurements were taken from 1VRHE to 2VRHE at a scan rate of 5mV s-1

in 0.1 M KOH and 10 mM 5-HMF. All measurements are internal resistance (IR) corrected. ................. 149

Figure S 30. 5-HMF concentration dependence. Flow Electrolyzer Cell based linear scan voltammograms (LSV)

using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and FAS

50 membrane in 0.1 M KOH used as catholyte and 0-15 mM HMF in 0.1 M KOH used as anolyte.

Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV

electrode potentials 1.0 Vcell - 1.7 Vcell,...................................................................................................... 149

Figure S 31. 5-HMF concentration dependence (HFR corrected). Flow Electrolyzer Cell based linear scan

voltammograms (LSV) using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg

cm-2) cathodes and FAS 50 membrane in 0.1 M KOH used as catholyte and 0-15 mM HMF in 0.1 M KOH

used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1,

start and end LSV electrode potentials 1.0 Vcell - 1.7 Vcell, scan rate 10 mV s-1. All potential values are IR

corrected for uncompensated ohmic HFR. ................................................................................................ 150

Figure S 32. Anion exchange membrane evaluation. Flow Electrolyzer Cell based linear scan voltammograms

(LSV) using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes

and different anion exchange membranes in 0.1 M KOH used as catholyte and 10 mM HMF (HMF) in 0.1

M KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml

min-1, start and end LSV electrode potentials 1.0 Vcell - 1.7 Vcell, scan rate 10 mV s-1. ............................. 150

Appendix

210

Figure S 33. Anion exchange membrane evaluation (HFR corrected). Flow Electrolyzer Cell based linear scan

voltammograms (LSV) using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg

cm-2) cathodes and different anion exchange membranes in 0.1 M KOH used as catholyte and 10 mM

HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis

parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell - 1.7 Vcell, scan rate 10

mV s-1. ....................................................................................................................................................... 151

Figure S 34. Images of two different used membranes. a) Images of the FAA-3-PK-130 Fumatech membrane.

Showing the web structure of it. b) IONOMR HNN 8 membrane after reaction as evidence for activity

disparities. .................................................................................................................................................. 152

Figure S 35. HPLC Calibration curves of 5-HMF its oxidation products. a)-e) are showing the HPLC response

for 5-HMF and its oxidation products at 1 mM (black), 5 mM (red), and 10 mM (blue) in 0.1 M KOH with

H2SO4 (1 ml/min) as mobile phase. It was not completely possible to avoid the conversion of c) FDA to a)

5-HMF and e) FFCA, which exclude 1 mM FDA calibration. .................................................................. 152

Figure S 36. Influence of the electrolyte cation. Flow Electrolyzer Cell based linear scan voltammograms (LSV)

using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a

FAA-3-PK-130 membrane in 0.1 M XOH (X=Li,Na,K,Rb,Cs) used as catholyte and 10 mM HMF (HMF)

in 0.1 M XOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate

20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1. .................. 153

Figure S 37. Influence of the electrolyte cation (HFR corrected). Flow Electrolyzer Cell based linear scan

voltammograms (LSV) using pure Ni foam (NF) anodes, C supported Pt nanoparticles (loading of 0.1 mg

cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M XOH (X=Li,Na,K,Rb,Cs) used as catholyte and

10 mM HMF (HMF) in 0.1 M XOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis

parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10

mV s-1. ....................................................................................................................................................... 153

Figure S 38. Evaluation of a second metal in addition to Ni (HFR corrected). Flow Electrolyzer Cell based linear

scan voltammograms (LSV) using pure Ni foam (NF) and NiX(-CO32-)-LDHs (X=Fe,Mn,Co,V) anodes, C

supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M

KOH used as catholyte and 10 mM HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and anolyte

volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials

1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1. .................................................................................................... 154

Figure S 39. Investigation of the NiV(-A-)-LDH@NF interlayer chemistry. Flow Electrolyzer Cell based linear

scan voltammograms (LSV) using NiV(-CO32-)-LDH@NF and NiV(-Cl-)-LDH@NF anodes, C supported

Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M KOH used as

catholyte and 10 mM HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and anolyte volume of 100

ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2

Vcell, scan rate 10 mV s-1. ........................................................................................................................... 155

Figure S 40. Investigation of the NiV(-A-)-LDH@NF interlayer chemistry (HFR corrected). Flow Electrolyzer

Cell based linear scan voltammograms (LSV) using NiV(-CO32-)-LDH@NF and NiV(-Cl-)-LDH@NF

anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane

Appendix

211

in 0.1 M KOH used as catholyte and 10 mM HMF (HMF) in 0.1 M KOH used as anolyte. Catholyte and

anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode

potentials 1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1. .................................................................................... 155

Figure S 41. Investigation of the interlayer chemistry of NiFe-(A-)-LDH@NF (HFR corrected). Flow

Electrolyzer Cell based linear scan voltammograms (LSV) using NF and NiFe(-A-)-LDH@NF (A=CO32-

,Cl-,ClO4-) anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130

membrane in 0.1 M KOH used as catholyte and 10 mM HMF (HMF) in 0.1 M KOH used as anolyte.

Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, start and end LSV

electrode potentials 1.0 Vcell – 2.2 Vcell, scan rate 10 mV s-1. .................................................................... 156

Figure S 42. Color change within 2 h of electrolysis with NiFe(-Cl-)-LDH@NF, samples were taken every 10

min. ............................................................................................................................................................ 157

Figure S 43. Correlation of 5-HMF conversion, FDCA yield and current of NiFe(-Cl-)-LDH@NF. MEA-type

cell NiFe(-Cl-)-LDH@NF //FAA-3-PK-130// Pt-C@CP electrolysis with 10 mM 5-HMF in 0.1 M KOH

(V= 100ml), a flow of 20 ml min-1, at a at a constant potential of 1.56 V over 2 h and a Pt catalyst loading

of 0.1 mg cm-2. ........................................................................................................................................... 157

Figure S 44. Scanning Electron Microscopy images of NF and NiFe(-A-)-LDH@NF (A=CO32-b), Cl- c) and

ClO4- d)) at 1000-2000x magnification. The color code corresponds to Figure S17.Correlation of 5-HMF

conversion, FDCA yield and current of NiFe(-Cl-)-LDH@NF. MEA-type cell NiFe(-Cl-)-LDH@NF

//FAA-3-PK-130// Pt-C@CP electrolysis with 10 mM 5-HMF in 0.1 M KOH (V= 100ml), a flow of 20

ml/min, at a at a constant potential of 1.56 V over 2 h and a Pt catalyst loading of 0.1 mg cm-2. ............. 158

Figure S 45. Evaluation of NiFe(-Cl-)-LDH@NF in different electrolytes. Flow Electrolyzer Cell based linear

scan voltammograms (LSV) using NiFe(-Cl-)-LDH@NF anodes, C supported Pt nanoparticles (loading of

0.1 mg cm-2) cathodes and a FAA-3-PK-130 membrane in 0.1 M and 1 M KOH used as catholyte and 10

mM HMF (HMF) in 0.1 M and 1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml.

Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 2.2 Vcell,

scan rate 10 mV s-1. ................................................................................................................................... 159

Figure S 46. Electrolysis of NiFe(-Cl-)-LDH@NF at a constant potential. Flow Electrolyzer Cell electrolysis

using NiFe(-Cl-)-LDH@NF anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a

FAA-3-PK-130 membrane in 1 M KOH used as catholyte and 10 mM HMF in 1 M KOH used as anolyte.

Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml min-1, constant potential

of 1.56 Vcell for 50 min. ............................................................................................................................. 159

Figure S 47. Time resolved mass spectrometric Linear Sweep Voltammograms of the NiFe(-Cl-)-LDH@NF

electrocatalyst. The data was obtained in a custom-made DEMS flow cell setup in a) 0.1 M KOH and b)

10 mM HMF in 0.1 M KOH, flow rate 500 µl min-1, scan rate 5 mV s-1 from 0- +2,0 VRDE. Plotted are

ERHE and the signals of m/z= 32 of 16O 16O over the time. ........................................................................ 160

Figure S 48. Schematic side view representation of the main flow path in the capillary flow cell. The liquid flow

inlet fresh electrolyte flow directed to catalysts surface and the concentric tube and capillary. ............... 160

Figure S 49. Pulse Stability test of NiFe(-Cl-)-LDH@NF. Flow Electrolyzer Cell electrolysis using NiFe(-Cl-)-

LDH@NF anodes, C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes and a FAA-3-PK-130

Appendix

212

membrane in 0.1 M KOH used as catholyte and 10 mM HMF (injected 5 times every 60 min) in 0.1 M

KOH used as anolyte. Catholyte and anolyte volume of 100 ml. Electrolysis parameters: flow rate 20 ml

min-1, constant potential of 1.56 Vcell for 60 min (5 cycles). ..................................................................... 161

Figure S 50: RDE measurements of CuO/NiO comparing the activity of different metal oxide molar ratios

(mol%). The Cu to Ni ratio was varied by different Ni precursor amounts. a) LSV measurements were

taken between 0 VRHE to -0,8 VRHE at a scan rate of 10 mV s-1 in 0.1 M KOH with and without 10 mM 5-

HMF. All measurements are internal resistance (IR) corrected. b) resulting overpotentials from a) for

HMFRR (red) and HER (black) at 10 mA cm-2 for different NiO mol%. ................................................. 164

Figure S 51: XPS measurements of CuO powder. a) Survey spectrum, b) Cu LMM Auger range c) O 1s region,

d) Cu 2p region. Measurements are given in black whereas Casa XPS fits are given in light blue for O and

orange for Cu. ............................................................................................................................................ 165

Figure S 52: XPS measurements of CuO/NiO powder. a) Survey spectrum, b) O 1s region, c) Cu 2p region, d) Ni

2p region. Measurements are given in black whereas Casa XPS fits are given in yellow for C, light blue for

O, orange for Cu and red for Ni. ................................................................................................................ 166

Figure S 53: XPS measurements of CuO/Fe2O3 powder. a) Survey spectrum, b) O 1s region, c) Cu 2p region, d)

Fe 2p region. Measurements are given in black whereas Casa XPS fits are given in yellow for C, light blue

for O, orange for Cu and purple for Fe. ..................................................................................................... 167

Figure S 54: XPS measurements of CuO/Co3O4 powder. a) Survey spectrum, b) O 1s region, c) Cu 2p region, d)

Co 2p region. Measurements are given in black whereas Casa XPS fits are given in yellow for C, light blue

for O, orange for Cu and blue for Co. ........................................................................................................ 168

Figure S 55: TEM images of the powder CuO/MOx catalysts. 1a)-c) showing CuO/NiO, 2a)-c) showing

CuO/Fe2O3 and 3a)-c) showing CuO/Co3O4. A scale is provided in the right bottom corner. .................. 169

Figure S 56: HR-STEM images of the powder CuO and CuO/MOx catalysts. 1a) and b) showing CuO, 2a) and b)

showing CuO/NiO, 3a) and b) showing CuO/Fe2O3 and 4a) and b) showing CuO/Co3O4. A scale is

provided in the right bottom corner. .......................................................................................................... 170

Figure S 57: SEM-EDX measurements of CF before and after electrolysis. a) and b) SEM images of CF before

electrolysis at 1000x and 5000x magnification. c) and d) SEM images of CF after electrolysis (ae) at 1000x

and 5000x magnification. e) SEM EDX images before electrolysis of CF with O in torques and Cu in

orange. f) SEM EDX images after electrolysis of CF with O in torques and Cu in orange. g) SEM EDX

results before electrolysis of CF with O in torques, C in yellow and Cu in orange. h) SEM EDX results

after electrolysis of CF with O in torques, C in yellow and Cu in orange. ................................................ 171

Figure S 58: SEM-EDX measurements of CuO/CF before and after electrolysis. a) and b) SEM images of

CuO/CF before electrolysis at 1000x and 5000x magnification. c) and d) SEM images of CuO/CF after

electrolysis (ae) at 1000x and 5000x magnification. e) SEM EDX images before electrolysis of CuO/CF

with O in torques and Cu in orange. f) SEM EDX images after electrolysis of CuO/CF with O in torques

and Cu in orange. g) SEM EDX results before electrolysis of CuO/CF with O in torques, C in yellow, F in

pink and Cu in orange. h) SEM EDX results after electrolysis of CuO/CF with O in torques, C in yellow, F

in pink and Cu in orange............................................................................................................................ 172

Appendix

213

Figure S 59: SEM-EDX measurements of CuO/NiO/CF before and after electrolysis. a) and b) SEM images of

CuO/NiO/CF before electrolysis at 1000x and 5000x magnification. c) and d) SEM images of

CuO/NiO/CF after electrolysis (ae) at 1000x and 5000x magnification. e) SEM EDX images before

electrolysis of CuO/NiO/CF with O in torques, Ni in red and Cu in orange. f) SEM EDX images after

electrolysis of CuO/NiO/CF with O in torques, Ni in red and Cu in orange. g) SEM EDX results before

electrolysis of CuO/NiO/CF with O in torques, C in yellow, F in pink, Ni in red and Cu in orange. h) SEM

EDX results after electrolysis of CuO/NiO/CF with O in torques, C in yellow, F in pink, Ni in red and Cu

in orange. ................................................................................................................................................... 173

Figure S 60: SEM-EDX measurements of CuO/Fe2O3/CF before and after electrolysis. a) and b) SEM images of

CuO/Fe2O3/CF before electrolysis at 1000x and 5000x magnification. c) and d) SEM images of

CuO/Fe2O3/CF after electrolysis (ae) at 1000x and 5000x magnification. e) SEM EDX images before

electrolysis of CuO/Fe2O3/CF with O in torques, Fe in purple and Cu in orange. f) SEM EDX images after

electrolysis of CuO/Fe2O3/CF with O in torques, Fe in purple and Cu in orange. g) SEM EDX results

before electrolysis of CuO/Fe2O3/CF with O in torques, C in yellow, F in pink, Fe in purple and Cu in

orange. h) SEM EDX results after electrolysis of CuO/Fe2O3/CF with O in torques, C in yellow, F in pink,

Fe in purple and Cu in orange. .................................................................................................................. 174

Figure S 61: SEM-EDX measurements of CuO/Co3O4/CF before and after electrolysis. a) and b) SEM images of

CuO/Co3O4/CF before electrolysis at 1000x and 5000x magnification. c) and d) SEM images of

CuO/Co3O4/CF after electrolysis (ae) at 1000x and 5000x magnification. e) SEM EDX images before

electrolysis of CuO/Co3O4/CF with O in torques, Co in blue and Cu in orange. f) SEM EDX images after

electrolysis of CuO/Fe2O3/CF with O in torques, Co in blue and Cu in orange. g) SEM EDX results before

electrolysis of CuO/Co3O4/CF with O in torques, C in yellow, F in pink, Co in blue and Cu in orange. h)

SEM EDX results after electrolysis of CuO/Co3O4/CF with O in torques, C in yellow, F in pink, Co in blue

purple and Cu in orange............................................................................................................................. 175

Figure S 62: Thin film XRD of CuO on a GC disc electrode before and after electrolysis and of CF. a) XRD

patterns of powder CuO drop casted on a GC disk electrode before (dark red) and after electrolysis (ae)

(light red) and the GC disk electrode with Nafion binder as a background (grey), references are given in

orange for CuO tenorite and red for metallic copper. The transparent boxes mark the specific 2θ angles for

CuO tenorite (orange) and metallic copper (red). b) XRD patterns of CF (black) and the metallic copper

reference (red). .......................................................................................................................................... 176

Figure S 63: RDE three-electrode measurements and powder XRD characterization of CuO/MOx compared to

physically mixed CuO and MOx. a)-c) RDE measurements of the mixed metal oxides CuO/NiO (red),

CuO/Fe2O3 (purple), and CuO/Co3O4 (blue) compared to the physically mixed equivalents (green). All

RDE LSV measurements were taken between 0 VRHE to -0.6 VRHE at a scan rate of 10 mV s-1 in 0.1 M

KOH with (solid line) and without (dashed line) 10 mM HMF at 2500 rpm with an electrode surface area

of 0.19 cm2 and a catalyst loading of 0.04 mg. All measurements are 100% manual internal resistance (IR)

corrected. d)-f) Powder XRD measurements of the mixed metal oxides CuO/NiO (red), CuO/Fe2O3

(purple), and CuO/Co3O4 (blue) compared to the physically mixed equivalents (green), CuO (black), and

the CuO Tenorite reference (orange). ........................................................................................................ 176

Appendix

214

Figure S 64: Schematic drawing of the undivided three-electrode cell setup. Showing the configuration of the

UTEC from left to right, with Pt-mesh counter electrode (CE), 1 cm-2 CF based working electrode (WE)

and reversible hydrogen electrode (RHE) reference electrode (RE) and an electrolyte volume of 50 ml. 177

Figure S 65: Undivided three-electrode cell (UTEC) loading study of CuO/NiO/CF. a) Showing the

overpotential at 10 mA cm-2 for different catalyst loadings without (black) and with (green) 10 mM HMF.

b) Showing the overpotential at 10 mA cm-2 (mass corrected) for different catalyst loadings without (black)

and with (green) 10 mM HMF. The loading of 0 mg cm-2 is out of the plotted range. ............................. 177

Figure S 66: Activity comparison of CuO/MOx, CuO/CF, MOx/CF and CuO/MOx/CF. Comparing a) CuO/NiO

(red), CuO/CF (black), NiO/CF (light red) and CuO/NiO/CF (dark red), b) CuO/Fe2O3 (purple), CuO/CF

(black), Fe2O3/CF (light purple) and CuO/ Fe2O3/CF (shiny purple) and c) CuO/Co3O4 (blue), CuO/CF

(black), Co3O4/CF (light blue) and CuO/ Co3O4/CF (dark blue). Reaction conditions are the same as in

Figure 3 for RDE and UTEC measurements. All measurements are 100% manual internal resistance (IR)

corrected .................................................................................................................................................... 178

Figure S 67: Electrochemical cell setup and the corresponding HFR results. a) 0.1 M KOH with 10 mM HMF as

catholyte (100 ml, recycled), 0.1 M KOH as anolyte (100 ml, recycled), 5 cm2 electrode area,

CuO/MOx/CF as cathode (black), nickel foam (NF) as anode (metallic grey), FAA-3-PK membrane

(yellow), PTFE gaskets (white) and a flow rate of 25 ml min-1. Based on the same design from earlier

work.155 b) High frequency resistance results before and after constant current electrolysis. Error bars with

relative errors of 0.4-2%. ........................................................................................................................... 178

Figure S 68: MEA-Flow-Cell performance measurements of CuO/MOx/CF electrodes (with error bars). a)-e)

Faradaic efficiencies for H2 (strong blue), BHMF (light blue), MFA (green) and other products (grey) of

the different spray-coated CuO/MOx/CF catalysts. f) Faradaic efficiencies and HMF conversion (yellow)

are calculated for every 15 min time interval over 60 min using CuO/NiO/CF as a catalyst. Product color

code stays as in a)-e). g)-i) Scatter plot of the product selectivity preference. HMF conversion over

MFA/BHMF selectivity ratio, calculated by SMFA/SBHMF for all CuO/MOx catalysts on CF at different

current densities for 30 min. The color code stays the same as before. Cell reaction conditions: 0.1 M KOH

with 10 mM HMF as catholyte (100 ml, recycled), 0.1 M KOH as anolyte (100 ml, recycled), 5 cm2

electrode area, nickel foam (NF) as anode and a flow rate of 25 ml min-1, at 10-30 mA cm-2 for 30 min.

Error bars with relative errors of 2-4% for FEproducts, 3-5% for FEH2 and 1-3% for XHMF were added. ..... 179

Figure S 69: MEA-Flow-Cell activity measurements of CuO/MOx/CF electrodes. a)-e) Cell potential (red),

current densities (black) and H2 rate without (neon blue) and with HMF (neon green) of the spray-coated

CuO/MOx/CF catalysts pure CF (orange), CuO (black), NiO (red), Fe2O3 (purple) and Co3O4 (blue). Cell

reaction conditions: 0.1 M KOH without 10 mM HMF as anolyte (100 ml), and 0.1 M KOH with 10 mM

HMF as catholyte (100 ml), 5 cm2 electrode area, nickel foam (NF) as anode and a flow rate of 25 ml min-

1, at 10-30 mA cm-2.................................................................................................................................... 180

Figure S 70: Comparison with the literature presented in a radar plot. Performance and reaction parameters like

the HMF conversion (XHMF), selectivity towards BHMF (SBHMF) and MFA (SMFA), the potential at 10 mA

cm-2 (E10mA/cm2), pH and initial HMF concentration (cHMF) are compared between CuO/Fe2O3/CF and data

from the literatur.96,99,101 ............................................................................................................................ 180

Appendix

215

Figure S 71: Evolution of the continuous HMF and CO2 electrolyzer process. Evolution of the combined

electrolyzer cell starting with a semi-batch, electrolyte recycling HMF oxidation setup (a), over separated

considered continuous HMF oxidation (b) and CO2 reduction (c) to a combined continuous electrolyzer

(d). The anode NiFe(-Cl-)-LDH@NF is given in porous grey, Pt-C reference cathode in solid grey, the

NiNC-IMI PTFE60% cathode in porous black and the bipolar membrane in yellow and orange. ............... 187

Figure S 72: Semi-Batch HMF oxidation operation condition optimization. a) MEA constellation and

electrolyzer operation-mode (I Semi-Batch) overview. b) Linear scan voltammograms (LSV) comparison

of the AEM (FAA-3-PK-130) and BPM (Fumasep FBM) using NiFe(-Cl-)-LDH@NF anodes (5 cm2), C

supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and

10 mM HMF (HMF) in 1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml (recycled).

Electrolysis parameters: flow rate 20 ml min-1, start and end LSV electrode potentials 1.0 Vcell – 3 Vcell ,

scan rate 10 mV s-1. The black arrows show the 0.8 V voltage delay caused by the BPM. Green, yellow

and orange brackets show the different HMF reaction regions in AEM electrolysis. c) and d) current

density and concentration dependency of STYFDCA (grey), XHMF (yellow), SFDCA (green), FEFDCA (blue) over

1h electrolysis. Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt

nanoparticles (loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 10-30 mM

HMF in 1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml (recycled), flow rate 20 ml

min-1 and current densities of 25-75 mA cm-2. Standard deviation between 2-3%. ................................... 189

Figure S 73: Time resolved current density study of a Semi-Batch Electrolyzer. Time resolved performance of a

semi batch electrolyzer is shown as function of the applied current density. Plotted are the conversion of

XHMF (yellow), selectivity SFDCA (green), and faradic efficiency FEFDCA (blue) in percentage (z-axis) over

1h electrolysis (x-axis) and the current density in mA cm-2 (y-axis). Electrolysis parameters: BPM, NiFe(-

Cl-)-LDH@NF anodes (5 cm2), C supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes (5 cm2) in

1 M KOH used as catholyte and 10 mM HMF in 1 M KOH used as anolyte. Catholyte and anolyte volume

of 100 ml (recycled), flow rate 20 ml min-1 and current densities of 25-75 mA cm-2. Standard deviation

between 2-3%. ........................................................................................................................................... 190

Figure S 74: Time resolved concentration study of a Semi-Batch Electrolyzer. Time resolved performance of a

semi batch electrolyzer is shown as function of the initial HMF concentration. Plotted are the conversion

XHMF (yellow), the selectivity SFDCA (green), and the faradaic efficiency FEFDCA (blue) in percentage (z-

axis) over 1h electrolysis (x-axis) and the initial HMF concentration in mM (y-axis). Electrolysis

parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C supported Pt nanoparticles (loading of 0.1 mg

cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 10-30 mM HMF in 1 M KOH used as anolyte.

Catholyte and anolyte volume of 100 ml (recycled), flow rate 20 ml min-1 and current densities of 50 mA

cm-2. Standard deviation between 2-3%. ................................................................................................... 191

Figure S 75: CO2 reduction with a BPM configuration setup. CO2 feed is connected to an MFC to control the

flow rate and humidified by a water bubbler before feed to the cell. After the cell liquid water will be

collected by a condenser and after dried through a column mol sieve. In front of the GC our nitrogen inline

calibration will be added to the dry product stream. At the anode we cycle 1M KOH ............................. 192

Appendix

216

Figure S 76: CO2 reduction with a BPM configuration and combined continuously HMFOR setup. CO2 feed is

connected to an MFC to control the flow rate and humidified by a water bubbler before feeding to the cell.

After the cell liquid water will be collected by a condenser and after dried through a column mol sieve. In

front of the GC our nitrogen inline calibration will be added to the dry product stream. At the anode, we

pump 1M KOH and 10mM HMF through the cell and collect the product containing liquid to take HPLC

samples. ..................................................................................................................................................... 192

Figure S 77: Cell Potential comparison of CO2RR with and without HMFOR at different current densities. a)

showing LSVs of CO2RR without (black) and with (blue) HMFOR. b) Showing the cell voltage over time

at current densities of 50, 100 and 200 mA cm-2 for CO2RR without (black) and with (blue) HMFOR.

Reaction conditions: 20 ml min-1 humidified CO2, 1M KOH and 10 mM HMF with 2 ml min-1 flowrate

and NiFe(-Cl-)-LDH@NF anode at RT. Standard deviation between 2-3%. ............................................ 193

Figure S 78: Stability and degradation of paired CO2RR and HMFOR. a) Faradaic efficiency of CO2RR to CO

over 5h. b) Disproportional faradaic efficiency change over the first 2.5h ............................................... 194

Figure S 79: Carbon cross over coefficient (CCC) diagnosis. Changes in the CCC accurately identify the onset of

detrimental flow regime inside the catalyst layer, such as flooding. Shown are one one measurement

without PTFE (brighter red, corresponding to the measurement in Figure 4b) and one with 60wt% PTFE

(darker red, corresponding to measurement in Figure 4c). The CCC of the sample without PTFE is slowly

rising with means the protons have good accessibility to the Anions (OH-, CO32- and HCO3-) but if we take

the selectivity into consideration than we assume that the catalyst layer is flooded with would decrease the

selectivity as shown. On the other site the sample with high PTFE loading provides good selectivity and an

initial comparable low CCC (Region 1). This means the electrode is not flooded but the protons have

access to the Anions. After 2.75h we have an event which change the CCC drastically (Region 2). As the

selectivity remains high but the CCC after 3h is increasing. We assume that we flood/wet regions which

are not connected to the membrane and would lead to no accessibility of the protons. ............................ 195

Figure S 80: Flooding mechanism analyzed using time-resolved CO2/CO ratio measurements during our stability

test. a) With the ratio of CO2 to CO over 5.25h we show that after 2.5h the ratio change which is a sign of

non electrochemical consumption The flooding processes splits in 4 phases. b) At the beginning, only the

electrode surface near the membrane is wet so the protons can neutralize the hydroxide and carbonate. c)

After some time, regions that are not connected to the membrane (blue stripe inside catalyst layer) become

flooded. Proton transport in the flooded catalyst layer is suppressed, which causes CO2 to be converted into

carbonate in an acid base reaction. and as we have a disproportional change in FE with increasing N2

proportion gas is trapped in the GDE. d) After reaching 2.5h the trapped gas will be released which

increased drastically the CO FE (Figure S7). e) The CO2/CO ratio is decreasing which means the reaction

is taking place at areas which are not connected to the membrane and so the protons can not neutralize the

anions......................................................................................................................................................... 196

Figure S 81: Potential change during an Electrolyzer stability test. Potential degradation during 5h stability test is

about -21.74 mV h-1. .................................................................................................................................. 197

Appendix

217

10.7 List of Tables

Table 1: Summarizing and comparing the 3LC with the ...................................................................................... 52

10.8 List of Tables (Appendix)

Table S1: ICP-OES results of powdered NiFe(-CO32-)-LDH, NiFe(-Cl-)-LDH, and NiFe(-ClO4-)-LDH. Based on

the identical NF synthesis protocol. ........................................................................................................... 134

Table S2: ICP-OES results after stability protocol of NiFe(-Cl-)-LDH@NF. ................................................... 136

Table S 3. ICP-OES results of the powdered NiX-LDH (X= Fe,Co,Mn,V) nanocatalysts. Comparison of the

molar Ni:X ratio and total wt% of metal for all NiX-LDH nanocatalysts. The standard deviation of the

ICP-OES measured ppms are Ni: 1-5%, Fe: 5%, Mn: 3% Co: 19% and V: 3%. ..................................... 144

Table S 4. Electrochemical surface area of NiX-(CO32-)-LDH@NF (2.5 cm2). Results of the impedance

measurements at 1.58 VRHE in 0.1 M KOH comparing the ECSA (cm2) of Ni nanofoam support with NiX-

(CO32-)-LDH@NF (X=Fe,Mn,Co,V). Data of NF and NiFe-(CO32-)-LDH@NF were obtained from a

perspective published study of M.K. and P.H.. ......................................................................................... 146

Table S 5. ICP-OES results of the powdered NiFe(-A-)-LDH (A= CO32-, Cl-, ClO4-) nanocatalysts. Comparison

of the molar Ni:Fe ratio and total wt% of metal for all NiX-LDH nanocatalysts. The standard deviation of

the ICP-OES measured ppms are Ni: 1-5%, Fe: 5%. ................................................................................ 147

Table S 6. Electrochemical surface area after anion exchange. Results of the impedance measurements at

1.58 VRHE in 0.1 M KOH comparing the ECSA (cm2) of Ni nanofoam support with NiFe(-A-)-LDH@NF

(A= CO32-, Cl-, ClO4-) (2.5 cm2). Data were obtained from a perspective published study of M.K. and P.H..

................................................................................................................................................................... 147

Table S 7. HPLC results after 2 h of 5-HMF electrolysis at 150 mA cm-2 with NF (anode), Pt-C@CP (cathode) in

0.1 M KOH with 10 mM 5-HMF (V= 100ml) and a flow rate of 20 ml/min for different membranes. The

standard deviation for all values is ±1%. ................................................................................................... 151

Table S 8. HPLC results after 1 and 2 h of 5-HMF electrolysis at 1.56 Vcell with NiX(-CO32-)-LDH@NF (X= Fe,

Mn, Co and V) (anode)// FAA-3-PK-130//Pt-C@CP (cathode) in 0.1 M KOH with 10 mM 5-HMF (V=

100ml) and a flow rate of 20 ml/min. The standard deviation for all values is around ±1%. .................... 154

Table S 9. HPLC results after 1 and 2 h of 5-HMF electrolysis at 1.56 Vcell with NiFe(-A-)-LDH@NF (A=CO32-,

Cl- and ClO4-)// FAA-3-PK-130//Pt-C@CP in 0.1 M KOH with 10 mM 5-HMF (V= 100ml) and a flow

rate of 20 ml min-1. The standard deviation for all values is around ±1%. ................................................ 156

Table S 10. HPLC results after 45 min of 5-HMF electrolysis at 1.56 Vcell with NiFe(-Cl-)-LDH@NF// FAA-3-

PK-130//Pt-C@CP in 1 M KOH with 10 mM 5-HMF (V= 100ml) and a flow rate of 20 ml/min. The

standard deviation for all values is around ±1%. ....................................................................................... 160

Table S 11. Data collection of the latest published work about HMFOR. The references in the table are kept

original....................................................................................................................................................... 162

Appendix

218

Table S 12: ICP-OES results of the bimetallic catalysts. The composition of powdered CuO/NiO, CuO/Fe2O3

and CuO/Co3O4 are given in mol% and wt%. ........................................................................................... 181

Table S 13: BET results in m2/g and in cm2/0.04 mg (RDE electrode catalyst loading). .................................... 181

Table S 14: Performance parameters of all catalysts at different current densities over 30 min in the MEA flow

cell. ............................................................................................................................................................ 182

Table S 15: Performance results of the Semi-Batch HMFOR setup (Figure S 71 a) over 1 h electrolysis with

standard deviations between 1-3%. Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C

supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and

10-30 mM HMF in 1 M KOH used as anolyte. Catholyte and anolyte volume of 100 ml (recycled), flow

rate 20 ml min-1 and current densities of 25-75 mA cm-2. Standard deviation between 2-3%................... 198

Table S 16: Performance results of the continuous HMFOR setup (Figure S 71 b) over 1 h electrolysis with

standard deviations between 1-3%. Electrolysis parameters: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2), C

supported Pt nanoparticles (loading of 0.1 mg cm-2) cathodes (5 cm2) in 1 M KOH used as catholyte and 5-

20 mM HMF in 1 M KOH used as anolyte. Catholyte and anolyte volume of 50 ml, flow rates 5 and 2 ml

min-1 and current densities of 50-200 mA cm-2. Standard deviation between 2-3%. .................................. 199

Table S 17: Performance results of the combined continuous HMFOR and CO2RR setup (Figure S 71 d)

electrolysis with standard deviations between 2-4%. Electrolysis parameter: BPM, NiFe(-Cl-)-LDH@NF

anodes (5 cm2) with 1M KOH and 10 mM HMF anolyte with 2 ml min-1 flowrate, NiNC-IMI PTFE60%

cathodes with 20 ml min-1 humidified CO2. .............................................................................................. 200

Table S 18: Performance results of the CO2RR setup with BPM (Figure S 71c) and combined continuous

HMFOR and CO2RR setup (Figure S 71d) electrolysis compared to an AEM reference224. Electrolysis

parameter: BPM, NiFe(-Cl-)-LDH@NF anodes (5 cm2) with 1M KOH with 20 ml min-1 and 1M KOH+10

mM HMF anolyte with 2 ml min-1 flowrate respectively, NiNC-IMI PTFE60% cathodes with 20 ml min-1

humidified CO2. Standard deviation between 2-3%. ................................................................................. 200

Table S 19: Energy efficiencies of the different electrolyzer configurations with and without the BPM penalty.

Standard deviation between 2-3%. ............................................................................................................ 201

Appendix

219

10.9 List of Abbreviations

HMF

5-Hydroxymethylfurfural

ERM

Eley-Rideal mechanism

LDH

layered double hydroxides

N,N-DMF

N,N-Dimethylformamide

FDCA

2,5-Furandicarboxylic acid

EtOH

Ethanol

MFA

5-Methylfurfurylalcohol

ICP-OES

Inductively Coupled Plasma

Optical Emission Spectroscopy

HMFOR

5-Hydroxymethylfurfural

oxidation reaction

XRD

X-Ray Diffraction

Spectroscopy

HMFRR

5-Hydroxymethylfurfural

reduction reaction

SEM

Scanning Electron Microscopy

AWE

Alkaline Water Electrolyzer

TEM

Transition Electron

Microscopy

PEMWE

Proton Exchange Membrane

Water Electrolyzer

XPS

X-Ray Photoelectron

Spectroscopy

AEMWE

Anion Exchange Membrane

Water Electrolyzer

RDE

Rotating Disk Electrode

MEA

Membrane Electrode Assembly

Room Temperature

PGM

Platin Group Metal

RHE

Reversible Hydrogen Electrode

HER

Hydrogen Evolution Reaction

Cyclic Voltammetry

OER

Oxygen Evolution Reaction

Working Electrode

BHH

5,5-

Bis(hydroxymethyl)hydrofuroin

Counter Electrode

HFCA

5-(Hydroxymethyl)-2-

furancarboxylic acid

Reference Electrode

FDA

2,5-Furandialdehyde

DEMS

Differential Electrochemical

Mass Spectroscopy

FFCA

5-Formyl-2-furancarboxylic

acid

Mass Spectroscopy

Appendix

220

BHMF

2,5-Bis(hydroxymethyl)furan

i-PrOH

Propanol

5-Methylfurfural

HPLC

High Performance Liquid

Chromatography

DMF

2,5-Dimethylfuran

RID

Refraction Index Detector

PEF

Polyethylene

furandicarboxylate

ECSA

Electrochemical Surface Area

PET

Polyethylene terephthalate

Conversion

MOF

Metal Organic Framework

Yield

COF

Covalent Organic Framework

Faradaic Efficiency

DMDHF

2,5-dimethyl-2,3-dihydrofuran

STY

Space-Time-Yield

DFT

Density Functional Theory

3LC

Three Electrode Liquid Cell

CO2RR

CO2 reduction reaction

CCC

Carbon Crossover Coefficient

OHP

Outer Helmholtz Plane

BPM

Bipolar Membrane

IHP

Inner Helmholtz Plane

PTFE

Polytetrafluoroethylene

LHM

Langmuir-Hinshelwood

Mechanism

GDL

Gas Diffusion Layer

Appendix

221

10.10 List of Publications

Erstautorenschaft:

1. Anion-Tuned Layered Double Hydroxide Anodes for Anion Exchange Membrane

Water Electrolyzers: From Catalyst Screening to Single-Cell Performance.

Klingenhof, M., Hauke, P., Kroschel, M., Wang, X., Merzdorf, T., Binninger, C., Ngo

Thanh, T., Paul, B., Teschner, D., Schlögl, R. & Strasser, P. Anion-Tuned Layered

Double Hydroxide Anodes for Anion Exchange Membrane Water Electrolyzers: From

Catalyst Screening to Single-Cell Performance. ACS Energy Letters 7, 3415-3422, doi:

https://doi.org/10.1021/acsenergylett.2c01820 (2022).

Die Studie wurde von P.H. und M.K. konzipiert und durchgeführt. P.H. führte die

Synthese und die elektrochemische Bewertung mit dem 3-Elektroden-Setup durch.

M.K. führte die ECSA-Auswertung in der 3-Elektroden-Anordnung und die

elektrochemische Auswertung in der 2-Elektroden-Anordnung durch. M.Kr., X.W. und

T.M. helfen bei der Charakterisierung der Katalysatoren. C.B., T.N.T. und B.P. halfen

bei der Planung des Aufbaus.

2. Efficient paired electrolysis of 5-Hydroxy Methyl Furfural (HMF) to the biopolymer-

precursor Furan dicarboxylic acid (FDCA) at industrial current densities.

Hauke, P., Klingenhof, M., Wang, X., de Araújo, J. F. & Strasser, P. Efficient

electrolysis of 5-hydroxymethylfurfural to the biopolymer-precursor furandicarboxylic

acid in a zero-gap MEA-type electrolyzer. Cell Reports Physical Science 2, 100650,

doi: https://doi.org/10.1016/j.xcrp.2021.100650 (2021).

This article is licensed under a Creative Commons Attribution-NonCommercial-

NoDerivatives 4.0 International License (https://creativecommons.org/licenses/by-nc-

nd/4.0/).

Appendix

222

Die Studie wurde von P.H. konzipiert und durchgeführt. M.K. unterstützt bei der

Bewertung der Katalysatoren. Die SEM-Aufnahmen wurden von X.W. durchgeführt.

J.F.A. entwarf den DEMS-Aufbau und unterstützte die DEMS-Messungen.

3. Hydrogenation vs Hydrogenolysis during Electrochemical Valorization of 5-

Hydroxymethylfurfural over oxide-derived Cu-bimetallics in Alkaline Membrane

Electrolyzers.

Hauke, P., Merzdorf, T., Klingenhof, M. & Strasser, P. Hydrogenation versus

hydrogenolysis during alkaline electrochemical valorization of 5-

hydroxymethylfurfural over oxide-derived Cu-bimetallics. Nature Communications 14,

4708, doi: https://doi.org/10.1038/s41467-023-40463-y (2023).

This article is licensed under a Creative Commons Attribution 4.0 International License

(https://creativecommons.org/licenses/by/4.0/).

Die Studie wurde von P.H. konzipiert und durchgeführt. Alle Analysetechniken (außer

SEM, TEM und HR-STEM), elektrochemische Messungen und die Synthese aller

Katalysatoren wurden von P.H. durchgeführt. Die SEM-Bilder stammen von T.M..

Die TEM-Aufnahmen wurden von M.K. angefertigt. Die HR-STEM-Analyse wurde

von Sören Selve durchgeführt.

4. Paired Electrocatalytic Valorization of CO2 and Hydroxymethylfurfural in a Noble

Metal-free Bipolar Membrane Electrolyzer (under review, ACS Sustainable Chemistry

& Engineering).206

Hauke, P., Brückner, S. & Strasser, P. Paired Electrocatalytic Valorization of CO2 and

Hydroxymethylfurfural in a Noble Metal-free Bipolar Membrane Electrolyzer. ACS

Sustainable Chemistry & Engineering, doi:

https://doi.org/10.1021/acssuschemeng.3c03144 (2023).

Die Studie wurde von P.H. und S.B. konzipiert und durchgeführt. Die Messungen zur

HMF-Oxidation wurden von P.H. durchgeführt. Dies umfasst die vollständige

Entwicklung der anodischen Reaktion von der getrennten bis zur kombinierten

Zellbetrachtung. Die Messungen zur CO2-Reduktion wurden von S.B. durchgeführt.

Appendix

223

Co-Autorenschaft:

1. Efficient direct seawater electrolyzers using selective alkaline NiFe-LDH as OER

catalyst in asymmetric electrolyte feeds.

Dresp, S., Ngo Thanh, T., Klingenhof, M., Brückner, S., Hauke, P. & Strasser, P.

Efficient direct seawater electrolysers using selective alkaline NiFe-LDH as OER

catalyst in asymmetric electrolyte feeds. Energy & Environmental Science 13, 1725-

1729, doi: https://doi.org/10.1039/D0EE01125H (2020).

2. Modular design of highly active unitized reversible fuel cell electrocatalysts.

Klingenhof, M., Hauke, P., Brückner, S., Dresp, S., Wolf, E., Nong, H. N., Spöri, C.,

Merzdorf, T., Bernsmeier, D., Teschner, D., Schlögl, R. & Strasser, P. Modular Design

of Highly Active Unitized Reversible Fuel Cell Electrocatalysts. ACS Energy Letters 6,

177-183, doi: https://doi.org/10.1021/acsenergylett.0c02203 (2021).