The new generation of catalyst in
Electrocat-alytic Water Splitting
A study on doped carbon materials
Ing. Sander van Utrecht
TNO,
UV
A,
and
MSc Chemistry
Science for energy and sustainability
Literature Study
T
HE NEW GENERATION OF CATALYST IN
E
LECTROCATALY TIC
W
ATER
S
PLIT TING
A
STUDY ON DOPED CARBON MATERIALS
by
Ing. Sander van Utrecht
12000892
February 19, 2020
12 EC’s
01 November 2019 - 26 February 2020
Supervisor/Examiner: Examiner:
prof. dr. Bas de Bruin
dr. Chris Slootweg
Department of material solutions / TNO
A
BSTRACT
To complete the energy transition and move from fossil fuel sources, studies indicate that hydrogen produc-tion will be one of the cornerstone technologies. An expected electrolyzer capacity of ± 2600 GW would be needed to meet demands by 2050 in Europe (Extrapolated in Appendix A). [1] The most reliable technique to produce H2with the fluctuating renewable power sources is polymer electrolyte membrane (PEM).
The main difficulty with this system is the use of platinum group metals (PMG). While the first reaction hydro-gen evolution(HER) has received considerable improvements lowering the metal usage, the second reaction oxygen evolution (OER) has difficulties due to the large molecular rearrangement. Substituting the rare met-als for abundant and cheap alternatives is the only way to reach the goal of ± 2600 GW. During this literature review doped carbon materials where considered as the replacement, answering the question:
What would be a promising catalyst based on doped carbon materials regarding OER in electrolysis appli-cations, considering the published work starting from 2014?
In order to understand the potential of doped carbon the basic structural allotropes and properties of carbon are discussed. Going further into how doping the material can improve activity, including the formation of functional groups. For the most used dopend, nitrogen additional information is provided for example the most active species, namely pyrrolic and graphitic. [2] Synergic effects found in multi doped materials are shown to influence the activity. [3]
After discussing some more background in material synthesis and characterization, both physical and elec-trochemical, the catalyst requirements are discussed. The most important details are related to efficient mass transport true pores [4] and interference while multiple of the same depends are in close proximity. [5] Fol-lowed by a proposed standardization protocol allowing for easeare comparison between different catalytic material. [6]
When the state of the art materials are considered in general it was found that nitrogen doped materials showed the highest overpotentials, followed by hybrid materials and multi doped materials preforming best. Because the exact effect of the dopens are still unknown the precise effects will still need to be researched further. The main challenges being the complexity of multi doped systems and measuring insitu conditions for a better understanding of the chemical process. To conclude a combined effort between experimental and theoretical research will be needed to better understand the doped carbon materials. Combined with intelligent material design a new generation of electrolyzer catalyst could be within reach
C
ONTENTS
1 Introduction 1
2 Doped carbon materials 5
2.1 The shape of the carbon . . . 5
2.2 Adding dopants . . . 6
2.3 Synthesis . . . 9
2.4 Physicochemical characterization of doped carbons . . . 9
3 Electrochemical reaction 11
4 Requirement and tuning catalyst 13
5 Comparing parameters 15
6 State of the art 17
7 Future challenges 19
A Expected electrolyzer capacity 21
Bibliography 23
1
I
NTRODUCTION
For years civilization has depended on vast stored energy reserves in the form of fossil fuels. However, insights in global warming and social cost have driven the need for an energy transition. [7] During this transition, the goal is to move away from polluting and greenhouse gas-emitting energy sources. Unfortunately, the most promising renewable resources such as solar and wind energy do not have a stable output. [8, 9]
To be able to stabilize the grit, energy storage devices have to be implemented, for example, batteries and fuel cell systems. [10] Fuel cells use compounds like hydrogen (H2) as fuel, which is converted to water and energy in the form of electricity using an electrochemical reaction. Thus clean and green hydrogen sources are required, mostly found in electrolysis.
The most common electrolyzers are alkaline (AEL), polymer electrolyte membrane (PEM), and solid oxide (SOEC). Each with advantages and disadvantages. The most flexible of these three electrolyzers is the PEM cell, with the fastest response and cold start times. The system can be compared to a reverse fuel cell setup, as shown in figure 1.1. [8]
Figure 1.1: Schematic of membrane-based (PEM) electrolyzer converting water to hydrogen and oxygen using an external (electric)power supply. [8]
Combine the storage of energy with industrial applications already using H2, and an expected electrolysis
capacity of 335 GW will be required for the Netherlands and Germany by 2050 as predicted by a study from TenneT. [1] When extrapolated to the requirements of Europe, a need capacity of ± 2600 GW is expected (cal-culation included in appendix). Indicating that electrolysis will be one of the chief cornerstone technologies underpinning the establishment of the renewable energy-based, circular provision of fuels, chemicals, and materials in a forthcoming sustainable economy.
2 1. INTRODUCTION
With the current PEM electrolyzing technology, Platinum Group Metals (PMG) are used most commonly for the catalytic material, Pt in the hydrogen evolution reaction (HER; cathode) and Ir for the oxygen evolution reactions (OER; anode), shown in figure 1.1. [11, 12] To meet the expected demand, large quantities of the rare PMG would be required. However, other applications also require there use. [13] Much effort has gone into reducing the metal loadings of these catalysts. Whereas the HER reaction achieved significant progress, the challenge remains with the kinetically sluggish OER. [4] With besides electrolysis also Zn–Air Batteries using the OER, improved performance can lead to multiple technological improvements. [14]
Figure 1.2: The OER half-reaction in different acidities.[4]
The OER remains challenging because of the relatively large molecular rearrangement involved in the OER. [5] Depending on the conditions used during the OER, the reactions will proceed using different pathways. The main difference is found between alkaline and acid (/neutral) conditions, as shown in figure 1.2. [4, 15, 16]
Another aspect is what pathway is used by the catalyst to obtain the final product, primarily OER occurs by a four-electron pathway. Either as a direct 4e- reduction to H2O or a two-step reduction to a H2O2intermediate, then to H2O, with 2e-involved in each step. [17] While using the former 4e-reduction, two main mechanisms for water oxidation catalysis are known, the radical oxo-coupling mechanism (ROC) and the water nucle-ophilic attack mechanism (WNA). The later is considered the primary mechanism for alkaline conditions (as shown in table 1.1). Indications of the WNA are the presence of OH, O, and OOH groups on the surface of the catalyst. During the steps where water forms, the second and final reaction steps, have been observed to be barrierless. [5] Meaning the barriers are found in the first or third step, indicates that the rate-determining step is on one or the other.
OH–+* *OH + e– (1) OH–+*OH *O + H2O + e– (2) OH–+*O *OOH + e– (3) OH–+*OOH *+ O2+ H2O + e– (4)
3
Taking account of the problematic OER reaction and the limited supply of PMG, catalysts based on an abun-dant and affordable resource will have to be developed to meet the required electrolysis capacity. A promising material for this is N-doped carbon (NDC). These materials have proven to be useful as metal-free electrocat-alysts in several processes (e.g., Oxygen Reduction Reaction, H2O2 synthesis, etc.). [18]
NDCs show increased stability in oxidative conditions. [19, 20] Surface area, porosity, and active sites are tune able with a large verity of tools (e.g., doping, surface modifications). Material properties such as the metallic character (e.g., presenting delocalized sp2 electrons) [21].
These advantages concerning the NDC, further substantiated by the electronic character, with more "noble"-like properties. [22] As well as comparable activities to the best commercial precious-metal Ir/C OER catalyst in alkaline conditions. [23]
Thus regarding electrocatalyst development, NDCs are attractive as low-cost, efficient, and durable candi-dates [24], which can be seen by the vast quantity of research done on this topic. The purpose of this report is to get a clear overview of the current state of the art. Answering the question:
What would be a promising catalyst based on doped carbon materials regarding OER in electrolysis appli-cations, considering the published work starting from 2014?
Which will be answered by first elaborating on the background of the system focusing on the doped carbon materials and catalytic requirements. For the comparison of state of the art catalysts, some parameters will be given in comparing parameters, followed by the actual comparison. Finally, some of the future challenges will be discussed.
2
D
OPED CARBON MATERIALS
2.1. T
HE SHAPE OF THE CARBON
NDCs can start simple with only a carbon material and adding N doping. However, lots of different design choices are available to fine-tune the catalytic materials. The first consideration would be what type of car-bon. For this, the structure’s wich carbon forms are essential.
These different structures are called allotropes. Elemental carbon exists in two natural allotropes, diamond and graphite, which consist of extended networks of sp3- and sp2-hybridized carbon atoms, respectively. Starting from 1985, synthetic allotropes where found, and as technology advanced, more elaborate structures discovered (2.1). [21]
Figure 2.1: Different syntactic allotropes of carbon placed in a timeline. [21]
These structures build up in different dimensions. An example, zero-dimensional (0D) fullerenes, 1D carbon nanotubes, 2D graphene, and 3D graphite. [25] Their derivatives include quantum dots, nanofibres, nanorib-bons, nanospheres/capsules, and other nanostructured morphologies. [26]
Depending on how the material is buildup, the properties of hardness, thermal conductivity, lubrication be-havior, or electrical conductivity can be changed. For example, including a lot of sp2 hybridized carbons will allow for the conjugation of Pi molecular orbitals (MO), allowing electrons to move through the material, thus conducting electricity. [21]
6 2. DOPED CARBON MATERIALS
Biomass containing large amounts of proteins and amino acids is also considered a promising precursor for producing N-doped carbon materials, as obtaining these can be inexpensive and from waste streams. [27]
2.2. A
DDING DOPANTS
While bare carbon materials show low catalytic activity for OER. The properties can be significantly enhanced, which is usually done by either doping some heteroatoms or hybridization with other elements. [4] wich brings the next: doping the material with heteroatom dopants.
These consist of all elements which are not carbon or hydrogen. The dopant used mostly is N, but other elements such as O, B, P, and S can also improve performance. Combinations of these can further improve activity as well, by synergic effects. Where each dopant changes the charge distribution, in turn changing material and catalytic properties. [28]
To understand the function of dopants in materials, the classical doping effect for semiconductors and the functional groups have to be considered.
For the clasic semiconductor doping the conduction and valance band have to be cosiderd. These are a buildup of multiple energy levels stacked together. In figure 2.2, the conduction and valance band are repre-sented respectively as the blue and red cone. The valance band filled with electrons, whereas the conduction band is empty. The boundary between the filed and unfilled energy bands is called the Fermin level (repre-sented in yellow). Dopants thus move the Fermi level, based on the lack (P-doped) or excess (N-doped) of electrons in the dopant material, allows alterations in the electronic structure of the material.
Zhang et al. gave insight into alterations of the electronic structure for N doped systems. Namely, N-doping can reduce the bandgap between the highest occupied molecular orbital (HOMO) and the lowest-unoccupied molecular orbital (LUMO) by lifting the energy level of HOMO, which facilitates the electron transfer from the NDC’s to the adsorbed oxygen. [15] In this case, HOMO relates to the valance band and LUMO to the conduction band, in the model system of figure 2.2, we would consider this a blunting of the peaks allowing for easier moment between the two bands. But more important, the boundaries for each reaction step are changed, the nuclephile attacks depends on the location of the HOMO and LUMO level on both reactants.
Figure 2.2: Representation of how dopants alter the electronic structure of materials. [29]
Other properties result from functionalities on the dopant atoms itself. For example, (i) facilitating the O2 adsorption; (ii) increasing the number of active sites; and (iii) improving the surface hydrophilicity. [26]
These additional properties result from either direct charge transfer with an electron acceptor/donor (i.e., charge transfer doping) or through the introduction of defects (i.e., defective doping). [3] Charge transfer doping alters the electronic structure itself, whereas defective doping adds defects to the crystal structure moving energy levels of the material itself. The movement of energy levels is possible due to the way energy levels are related to chemical bonds with other atoms.
2.2. ADDING DOPANTS 7
There are two ways for the introduction of the dopants. The first method involves the in situ addition of the dopant. The second method uses some techniques to add dopants after the synthesis of the carbon material. The addition mechanism used for adding these dopants to the system is essential, because the location can determine the effectiveness of the dopants and how they participate in the desired reaction.
An example of some nitrogen dopant locations is given in figure 2.3.A, including some possible reactions in aqueous environments. [30] Lai at al described in a recent paper that graphitic N improves current density (a measure of rate). In contrast, pyridinic N lowers the onset potential (a measure of energy efficiency); the discussion of both these electrochemical properties is conducted in a later chapter. [2] This demonstrates the importance of the dopant location for specific properties.
Other experimental results also show the importance of pyridinic-N sites, showing that the ORR and OER activity observedly increased with an increase in pyridinic-N site density. [17] Further backed by the theoret-ical calculations done by Murdachaew et al., which produced volcano plots for the graphene and nanotubes NDC’s. [31] As shown in figure 2.3 B and C the pyrrolic and graphitic shows the most promising properties for the catalytic activity.
(a)
(b) Nano Tube
(c) Graphene
Figure 2.3: Some possible dopant locations for N atoms (A), with corresponding possible reactions in aqueous solutions. [30] Followed by volcano plots of oxygen absorbtion energy for diffrent dopant locations for nano tubes (B) and graphene (C). [31]
8 2. DOPED CARBON MATERIALS
Figure 2.4: An example of the introduction of other dopants to a carbon system, in this case, boron. [32]
Other dopants such as boron (as shown in figure 2.4) could also provide altered properties, increasing cat-alytic activity. [32] Acid, alcohol, and other oxygen-containing functional groups can also add properties. Most likely adding additional active sides to the catalytic material. [4] Multiple types of heteroatoms to a single material can provide another level of complexity, allowing for synergic interactions, further improving, and exceeding singly doped materials. [33]
However, as shown in figure 2.5 the placement of the different dopants (in this case, boron, and nitrogen) can have significant influences. As Hu et al. calculated for the adsorption energy of oxygen, based on the number of atoms between the N and B. As shown, the effect if increased absorption decreases over distance and skipping every other atom. [3]
Figure 2.5: The adsorption energy of oxygen, dependent on the number of atoms between N and B dopants for a carbon-based system. [3]
2.3. SYNTHESIS 9
The last doped material considered is hybrid systems, which consist of the combination of metal atoms and doped carbon materials. These are the most complicated option, with the disadvantage, some sort of metal element is stil used. The main advantage is that the best properties of both fields combine to increase catalytic activity.
2.3. S
YNTHESIS
know more of the catalyst background is known; the synthesis of the materials is considered. As there are more options available than can be analyzed within the scope of this project, only some introduction will be given for the more common practice methods. Such as molten salt templating /citeRN11, Facile electroless deposition /citeRN31, and chemical vapor deposition /citeRN16.
For molten salt templating, salt is mixed with the reactants, as shown in figure 2.6. Afterward, the mixture temperature is heated above the melting point of the salt in an inert atmosphere. Carbon structures form within the lattice of the salt. After cooling, the salt can be dissolved, and what remains is a pours carbon structure. [34, 35]
Figure 2.6: Schematic flow chart of reaction steps for the production of salt templated NDC. [35]
Deposition methods use, for example, vaporized or ionic reactants. Wich moves toward a target surface, either carbon material or a template material. The reactant is electrochemically or thermally reacted with the surface of the target, which allows for precise control in layer buildup of the material. When necessary, a leaching step removes the template material. Most of the time, a pyrolysis step is still required to lock the added compounds into place. [17, 35]
2.4. P
HYSICOCHEMICAL CHARACTERIZATION OF DOPED CARBONS
For the characterization of the materials, multiple analysis techniques can be applied. The most common consist of Raman spectroscopy and X-ray photoelectron spectroscopy (XPS). Raman spectroscopy allows studying the degree of graphitization of the electrocatalysts. While XPS allows investigating the surface com-position and the doping of carbon materials. Electron energy loss spectroscopy (EELS) has been used as an alternative technique to XPS to determine the surface composition of the doped carbon materials. Further-more, TEM is used to investigate the morphology, XRD, to study the crystalline (large angles) and the pore (small angles) structures. N2adsorption-desorption to investigate the pore structure and to determine the surface area of the N-doped ordered mesoporous carbons. [17, 18]
3
E
LECTROCHEMICAL REACTION
Besides physical and chemical characterization, the electrical functionality of the catalytic material also has to meet the expectation. The highly developed field of electrochemistry provides methods for this. Tech-niques such as voltammetry and amperometry provide information on how variables change and influence systems one’s electrical currents are passed through (as shown in figure 3.1). [36] Some of the more common benchmarking parameters are over potential, faradaic efficiency, and current density.
Figure 3.1: Schematic of basic electrochemical cell with influencing parameters. [36]
Over potential is related to the voltage efficiency, and is measured by comparing the measured potential at which the reaction is observed with the thermodynamic requirements based on the half-reactions.In regards to the theoretical thermodynamically needed potential is determind to be 1.23 V. [37] The faradaic efficiency describes how well electrons are used by the system to produce the desired product. Current density is used to determine reaction rates per are of catalyst ( or electrode), as the current is Coulomb per second, thus electrons per second.
Other relations such as the Koutecky-levich plot or the table plot can give insights in respectively electric current vs kinetic activity and overpotential vs. power density.[17, 28, 38]
4
R
EQUIREMENT AND TUNING CATALYST
While comparing catalysts it is essential to make some design choices. Such as is a bifunctional catalyst required (one that can do multiple reactions) or singly active catalyst, which is the predominant doping, etc. During this part of the report, some of the critical design aspects will be discussed.
The first design feature is pore size. As the O2molecules require a specific size to pass through pores, con-siderations have to be made for efficient mass transport within the electrode. [4] Where most reactions are found to occur as surface reactions or in the micropores. [34]
For the most commonly used doping, nitrogen, some more information is known for optimal utilization in catalytic material. For example, temperatures used during the synthesis process influence the N content. [34, 39, 40] As well that it was found that graphene and pyridinic nitrogen groups show the most activity toward the OER. [34, 41]
In theoretical solvated calculations, three systems where used to determine the effect of near N groups com-pared to single N groups (shown in figure 4.1). When comparing the carbon nano tube it was found that the doped N in close proximity showed higher barriers for the *OOH formation step, which is subsequently the rate-determining step. This indicates that the N groups should be well dispersed. [5]
However, there is still much unknown about the precise nature of the dopant function.
14 4. REQUIREMENT AND TUNING CATALYST
Figure 4.1: Show of the nano tube system used for solvation calculations (A), with a zoom of no N (B), one N (C), and two N in close proximity (D). [5]
5
C
OMPARING PARAMETERS
Catalyst can be vastly different but are made toward one goal: a highly active, selective, cheap, and stable catalyst. Determining which one performs the best can only be done accurately while using similar test con-ditions closely resembling real-world operating concon-ditions. To this end, McCrory et al. proposed a standard-ized benchmarking protocol for Electrocatalysts for the Oxygen Evolution Reaction. [6] This benchmarking protocol is used to compare the composition, surface area, catalytic activity, faradaic efficiency, and stability. To improve comparability, specific parameters are set, as shown in figure 5.1 , allowing for the best compar-isons between catalysts. The experiments are used to describe critical catalytic factors such as composition, surface area, catalytic activity, faradaic efficiency, and stability.
Figure 5.1: Flow chart with proposed standardized testing procedure for NDC materials. [6]
6
S
TATE OF THE ART
Some of the results found for recent publications on NDCs have been summarized in table 6.1. As shown, a lot of different materials have been researched with various applications (and reactions) in mind, such as fuel cells, batteries, capacitors (SCs), and electrolyzes. The doped carbon materials are well study only however only in the most resent years have researchers start focusing on the OER.
Table 6.1: Examples of uses for NDC from recently publiched articles.
Material Classification Reaction Source Selftemplated nitrogen-doped carbons Porous
carbon
ORR [40]
Hierarchically porous, N-doped carbon Porous carbon
ORR [41]
NDCN NDC Oxidation [38]
Polyaniline-derivated nitrogen-doped car-bon
Nanowire SCs [42]
(NH4)3PO4-carbon NDC ORR [43]
NC-1000 NDC SCs [30]
Bio mass based NDCs NDC [20] Heteroatom-Doped Graphitic Carbon Graphitic ORR [15] Graphitic carbon skeleton, Revieuw Graphitic [3] Boron doped graphene Graphene [32] Nitrogen-Enriched Nonporous Carbon Pyrrol-like
groups
SCs [44]
Nitrogen-doped graphene sheets Graphene SCs [45] Porous carbon networks codoped with
ni-trogen and phosphorus
Porous carbon
ORR, HER [46]
N-doped carbon derived from sheep bones Porous carbon
ORR [27]
18 6. STATE OF THE ART
To compare the most recent finding a table was compiled (table 6.2) where some of the features are summa-rized such as over potential, onset potential, and conditions. As a lot of papers do not provide all the date and similar test conditions comparing of the parameters is difficult.
However some clear observations stand out. first of the lowest overpotentials and thus activities were found for systems in which transition metals were still present (hybrid systems), the NDC was used in a support function in this case. However, the use of NDC has shown to be better than alternative materials such as Si-based doped catalysts. The most promising catalyst used multiple dopants showing even lower onset po-tentials compared to transition metals.
Table 6.2: Caption
Material Classification Reaction conditions Over po-tential (V at 10 mA cm−2) Onset po-tential Source Polycrystalline CoP/-CoP2 Transition metal
OER, HER Alkaline 0.25 [47]
CuCo2O4 nanosheets Transition metal
OER Alkaline 0.26 [48]
Bio mass based NDCs NDC [20]
N, F, P ternary doped macro-porous car-bon fibers (NFPC)
NDC OER, ORR Alkaline 1.3 [28]
Melamine/formaldehyde NDC NDC OER Alkaline 0.38 1.61 [39] Graphene/PDPS NDC NDC OER Alkaline 0.41 1.64 [39]
CNx catalysts NDC OER, ORR Acid 0.39 1.62 [17] Hydrated Man-ganese(II) Phosphate Transition metal OER Neutral 0.16 [49] N,O-dual doped graphene-CNT (NG-CNT) Graphene, CNT OER Alkaline, Acid [4]
Porous silicon and ni-trogen co-doped car-bon (SiNC) nanoma-terial Porous carbon OER, Carbon activation 0.65 [23]
7
F
UTURE CHALLENGES
Before the NDC’s can become the next cornerstone of the energy sector, some challenges remain. First of the fundamental understanding of how these catalysts work still needs to be understood more clearly. Synergic effects such as the corporation of multiple dopants to increase activity will have to be further investigated, as to know what each dopant does and optimize the functional catalytic sides. A combined effort between experimental and theoretical research will result in the best progress made.
This will only be possible by improving the way these reactions are analyzed. Considering the difficulties of observing the reactions and the complexity of the systems, this could be regarded as one of the biggest challenges at the moment. Smart solutions have to be found to do in situ measurements allowing to best possible understanding of the system, with additional information being gained by computational methods such as DFT calculations.
An interesting aspect to be further investigated will also be economics. What is the current business case for these materials? Where does this technology have to be before it becomes a viable option? As there is to the best of our knowledge, not much information available.
To conclude, doped carbon materials show great potential to replace PMG’s. However, headway has to be made with the precise determination of the doping effects and reaction mechanisms. Achievable with intel-ligent material design and theoretical studies. It could also be interesting to look more into economics and make a study on the viability of the business case.
A
E
XPECTED ELECTROLYZER CAPACITY
Using the [1] the expected electrolyzer capacity for the Netherlands and Germany in 2050 was found. This was converted to per capita by deviding it with the expected population by the United Nations1. the per capita for the Netherlands and Germany was avradged and using the same source for population of Europe converted to expected capacity for the continent.
Area Expected population 2050 Expected Electrolisis (Gw) Electrolisis per capita Netherlands 17 165 000 67.5 3.93242E-06
Germany 80 104 000 267.5 3.33941E-06 Sum (Netherlands and Germany) 335
Europe 710 486 000 2583.27 3.63591E-06 World 9 735 034 000 32509.26 3.33941E-06
Table A.1: Found and calculated value for population and expected electrolyzer capacity.
1United Nations, Department of Economic and Social Affairs, Population Division (2019). World Population Prospects 2019, custom data
acquired via website. seen on 10-12-19
B
IBLIOGRAPHY
[1] Gasunie and TenneT, Infrastructure outlook 2050, Infrastructure Outlook 2050 (2019).
[2] L. Lai, J. R. Potts, D. Zhan, L. Wang, C. K. Poh, C. Tang, H. Gong, Z. Shen, J. Lin, and R. S. Ruoff,
Ex-ploration of the active center structure of nitrogen-doped graphene-based catalysts for oxygen reduction reaction, Energy Environmental Science 5, 7936 (2012).
[3] C. Hu and L. Dai, Doping of carbon materials for metal-free electrocatalysis, Advanced Materials 31, 1804672 (2019).
[4] S. Chen, J. Duan, M. Jaroniec, and S.-Z. Qiao, Nitrogen and oxygen dual-doped carbon hydrogel film as a
substrate-free electrode for highly efficient oxygen evolution reaction, Advanced Materials 26, 2925 (2014).
[5] L. Partanen, G. Murdachaew, and K. Laasonen, Oxygen evolution reaction kinetic barriers on
nitrogen-doped carbon nanotubes, The Journal of Physical Chemistry C 122, 12892 (2018).
[6] C. C. L. McCrory, S. Jung, J. C. Peters, and T. F. Jaramillo, Benchmarking heterogeneous electrocatalysts for
the oxygen evolution reaction, Journal of the American Chemical Society 135, 16977 (2013).
[7] L. Brown, The great transition: Shifting from fossil fuels to solar and wind energy, The Humanist 76, 27 (2016).
[8] P.- Franc and P. Mateo, Hydrogen: the energy transition in the making! (Alternatives éditions, 2015). [9] De Bosatlas van de energie (Wolters-Noordhoff, Groningen, 2012).
[10] H. Leenaers, A. Noordhoff, and Lijn, De Bosatlas van de duurzaamheid (Noordhoff Uitgevers, Gronin-gen, 2019).
[11] T. Nakagawa, C. A. Beasley, and R. W. Murray, Efficient electro-oxidation of water near its reversible
po-tential by a mesoporous irox nanoparticle film, The Journal of Physical Chemistry C 113, 12958 (2009).
[12] Y. Zhao, E. A. Hernandez-Pagan, N. M. Vargas-Barbosa, J. L. Dysart, and T. E. Mallouk, A high yield
syn-thesis of ligand-free iridium oxide nanoparticles with high electrocatalytic activity, The Journal of Physical
Chemistry Letters 2, 402 (2011).
[13] N. Tian, Z.-Y. Zhou, and S.-G. Sun, Platinum metal catalysts of high-index surfaces: From single-crystal
planes to electrochemically shape-controlled nanoparticles, The Journal of Physical Chemistry C 112,
19801 (2008).
[14] Y. Guo, P. Yuan, J. Zhang, H. Xia, F. Cheng, M. Zhou, J. Li, Y. Qiao, S. Mu, and Q. Xu, Co2p–con double
active centers confined in n-doped carbon nanotube: Heterostructural engineering for trifunctional catal-ysis toward her, orr, oer, and zn–air batteries driven water splitting, Advanced Functional Materials 28,
1805641 (2018).
[15] J. Zhang and L. Dai, Heteroatom-doped graphitic carbon catalysts for efficient electrocatalysis of oxygen
reduction reaction, ACS Catalysis 5, 7244 (2015).
[16] R. Maji and J. Bhattacharjee, Synergistic view of magnetism, chemical activation, and oxygen reduction
reaction as well as oxygen evolution reaction catalysis of carbon-doped hexagonal boron nitride from first principles, The Journal of Physical Chemistry C 123, 16731 (2019).
[17] K. Mamtani, D. Jain, D. Dogu, V. Gustin, S. Gunduz, A. C. Co, and U. S. Ozkan, Insights into oxygen
reduction reaction (orr) and oxygen evolution reaction (oer) active sites for nitrogen-doped carbon nanos-tructures (cnx) in acidic media, Applied Catalysis B: Environmental 220, 88 (2018).
24 BIBLIOGRAPHY
[18] N. Daems, X. Sheng, I. F. J. Vankelecom, and P. P. Pescarmona, Metal-free doped carbon materials as
electrocatalysts for the oxygen reduction reaction, Journal of Materials Chemistry A 2, 4085 (2014).
[19] M. Soorholtz, R. J. White, T. Zimmermann, M.-M. Titirici, M. Antonietti, R. Palkovits, and F. Schüth,
Direct methane oxidation over pt-modified nitrogen-doped carbons, Chemical Communications 49, 240
(2013).
[20] Y. Men, M. Siebenbürger, X. Qiu, M. Antonietti, and J. Yuan, Low fractions of ionic liquid or poly(ionic
liq-uid) can activate polysaccharide biomass into shaped, flexible and fire-retardant porous carbons, Journal
of Materials Chemistry A 1, 11887 (2013).
[21] A. Hirsch, The era of carbon allotropes, Nature Materials 9, 868 (2010).
[22] M. Antonietti and M. Oschatz, The concept of “noble, heteroatom-doped carbons,” their directed synthesis
by electronic band control of carbonization, and applications in catalysis and energy materials, Advanced
Materials 30, 1706836 (2018).
[23] M. A. Ghausi, J. Xie, Q. Li, X. Wang, R. Yang, M. Wu, Y. Wang, and L. Dai, Co2 overall splitting by a
bifunctional metal-free electrocatalyst, Angewandte Chemie International Edition 57, 13135 (2018).
[24] J. Shui, M. Wang, F. Du, and L. Dai, N-doped carbon nanomaterials are durable catalysts for oxygen
reduction reaction in acidic fuel cells, Science Advances 1, e1400129 (2015).
[25] J. Zhang, Z. Xia, and L. Dai, Carbon-based electrocatalysts for advanced energy conversion and storage, Science Advances 1, e1500564 (2015).
[26] D.-W. Wang and D. Su, Heterogeneous nanocarbon materials for oxygen reduction reaction, Energy Envi-ronmental Science 7, 576 (2014).
[27] S. Li, R. Xu, H. Wang, D. J. L. Brett, S. Ji, B. G. Pollet, and R. Wang, Ultra-high surface area and mesoporous
n-doped carbon derived from sheep bones with high electrocatalytic performance toward the oxygen re-duction reaction, Journal of Solid State Electrochemistry 21, 2947 (2017).
[28] M. Wu, Y. Wang, Z. Wei, L. Wang, M. Zhuo, J. Zhang, X. Han, and J. Ma, Ternary doped porous carbon
nanofibers with excellent orr and oer performance for zinc–air batteries, Journal of Materials Chemistry
A 6, 10918 (2018).
[29] M. Pykal, P. Jureˇcka, F. Karlický, and M. Otyepka, Modelling of graphene functionalization, Physical Chemistry Chemical Physics 18, 6351 (2016).
[30] J. Biemolt, Boosting the supercapacitance of nitrogen-doped carbon by tuning surface functionalities, (2017).
[31] G. Murdachaew and K. Laasonen, Oxygen evolution reaction on nitrogen-doped defective carbon
nan-otubes and graphene, The Journal of Physical Chemistry C 122, 25882 (2018).
[32] S. Agnoli and M. Favaro, Doping graphene with boron: a review of synthesis methods, physicochemical
characterization, and emerging applications, Journal of Materials Chemistry A 4, 5002 (2016).
[33] J. Zhang, L. Qu, G. Shi, J. Liu, J. Chen, and L. Dai, N,p-codoped carbon networks as efficient metal-free
bifunctional catalysts for oxygen reduction and hydrogen evolution reactions, Angewandte Chemie
Inter-national Edition 55, 2230 (2016).
[34] J. Ding, S. Ji, H. Wang, H. Gai, F. Liu, V. Linkov, and R. Wang, Mesoporous nickel-sulfide/nickel/n-doped
carbon as her and oer bifunctional electrocatalyst for water electrolysis, International Journal of Hydrogen
Energy 44, 2832 (2019).
[35] T. Kimura, Molten salt synthesis of ceramic powders, Advances in Ceramics-Synthesis and Characteriza-tion, Processing and Specific Applications , 75 (2011).
[36] A. J. Bard, Electrochemical methods : fundamentals and applications, 2nd ed. (John Wiley Sons, Hobo-ken, NJ, 2001).
BIBLIOGRAPHY 25
[37] X. Zhang, Y. Chen, W. Zhang, and D. Yang, Coral-like hierarchical architecture self-assembled by cobalt
hexacyanoferrate nanocrystals and n-doped carbon nanoplatelets as efficient electrocatalyst for oxygen evolution reaction, Journal of Colloid and Interface Science 558, 190 (2020).
[38] Y.-Y. Yu, C. X. Guo, Y.-C. Yong, C. M. Li, and H. Song, Nitrogen doped carbon nanoparticles enhanced
extracellular electron transfer for high-performance microbial fuel cells anode, Chemosphere 140, 26
(2015).
[39] Y. Zhao, R. Nakamura, K. Kamiya, S. Nakanishi, and K. Hashimoto, Nitrogen-doped carbon
nanomateri-als as non-metal electrocatalysts for water oxidation, Nature Communications 4, 2390 (2013).
[40] D. Eisenberg, The evolution of hierarchical porosity in self-templated nitrogen-doped carbons and its
ef-fect on oxygen reduction electrocatalysis, (2016).
[41] D. Eisenberg, A rational synthesis of hierarchically porous, n-doped carbon from mg-based mofs:
Under-standing the link between nitrogen content and oxygen reduction electrocatalysis, (2016).
[42] Z. Zhao, Y. Xie, and L. Lu, Electrochemical performance of polyaniline-derivated nitrogen-doped carbon
nanowires, Electrochimica Acta 283, 1618 (2018).
[43] Q. Liu, Y. Zhou, S. Chen, Z. Wang, H. Hou, and F. Zhao, Cellulose-derived nitrogen and phosphorus
dual-doped carbon as high performance oxygen reduction catalyst in microbial fuel cell, Journal of Power
Sources 273, 1189 (2015).
[44] D. Hulicova-Jurcakova, M. Kodama, S. Shiraishi, H. Hatori, Z. H. Zhu, and G. Q. Lu, Nitrogen-enriched
nonporous carbon electrodes with extraordinary supercapacitance, Advanced Functional Materials 19,
1800 (2009).
[45] Y. Qiu, X. Zhang, and S. Yang, High performance supercapacitors based on highly conductive
nitrogen-doped graphene sheets, Physical Chemistry Chemical Physics 13, 12554 (2011).
[46] J. Zhang, L. Qu, G. Shi, J. Liu, J. Chen, and L. Dai, N,p-codoped carbon networks as efficient metal-free
bifunctional catalysts for oxygen reduction and hydrogen evolution reactions, Angewandte Chemie
Inter-national Edition 55, 2230 (2016).
[47] S. Li, G. Zhang, X. Tu, and J. Li, Polycrystalline cop/cop2 structures for efficient full water splitting, Chem-ElectroChem 5, 701 (2018).
[48] S. M. Pawar, B. S. Pawar, P. T. Babar, A. T. A. Ahmed, H. S. Chavan, Y. Jo, S. Cho, J. Kim, B. Hou, A. I. Inamdar, S. Cha, J. H. Kim, T. G. Kim, H. Kim, and H. Im, Nanoporous cuco2o4 nanosheets as a highly
efficient bifunctional electrode for supercapacitors and water oxidation catalysis, Applied Surface Science
470, 360 (2019).
[49] K. Jin, J. Park, J. Lee, K. D. Yang, G. K. Pradhan, U. Sim, D. Jeong, H. L. Jang, S. Park, D. Kim, N.-E. Sung, S. H. Kim, S. Han, and K. T. Nam, Hydrated manganese(ii) phosphate (mn3(po4)2·3h2o) as a water
