Document [original]

In Situ Studies of Pt Nanoparticles on Different Supports

for Corrosion Stable PEM Fuel Cell Cathodes

Vorgelegt von

M. Sc.

Henrike Schmies

geb. in Friesoythe

Von der Fakultät II – Mathematik und Naturwissenschaften

der Technischen Universität Berlin

zur Erlangung des akademischen Grades

Doktorin der Naturwissenschaften

- Dr. rer. nat. -

genehmigte Dissertation

Promotionsausschuss:

Vorsitzende: Prof. Dr. Marga Lensen

Gutachter: Prof. Dr. Peter Strasser

Gutachter: Prof. Dr. Michael Bron

Tag der wissenschaftlichen Aussprache: 18. Dezember 2018

Berlin 2019

Danksagung

Mein besonderer Dank gilt Herrn Prof. Peter Strasser für die Bereitstellung dieses interessanten

Promotionsthemas, die finanzielle Unterstützung und die persönliche Betreuung. Außerdem möchte ich

mich für das Vertrauen bedanken, meine Arbeit mit großer wissenschaftlicher Freiheit und

Selbstständigkeit ausführen zu dürfen.

Ich möchte mich bei Herrn Prof. Michael Bron bedanken, dass er sich bereit erklärt hat, meine Arbeit

als Gutachter zu bewerten. Mein Dank geht zudem an Prof. Marga Lensen für die Übernahme des

Prüfungsvorsitzes.

Des Weiteren möchte ich mich bei allen Kooperationspartnern für die Unterstützung mit

verschiedensten Messmethoden, die konstruktiven Diskussionen der Ergebnisse und die erfolgreiche

Zusammenarbeit bedanken. Mein besonderer Dank geht an Arno Bergmann, von dem ich unglaublich

viel lernen durfte und der die Zusammenarbeit sehr inspirierend und unkompliziert gemacht hat. Ich

möchte mich bei Jakub Drnec für seine zuverlässige Unterstützung bei zahlreichen Messzeiten am ESRF

bedanken. Auch danke ich ganz besonders Elisabeth, Fabio, Thomas, Sören und Malte für die produktive

Arbeit während auch manch schwerer (nächtlicher und zerstörerischer) Stunde an der ID31. Elisabeth

und Sebastian danke ich für die gemeinsame Synthese und Analyse so manch schwarzer Pulver und eine

erfolgreiche Zusammenarbeit. Ich möchte mich zudem herzlich bei Annette für die zahlreichen

Messungen sowie Bestellungen und die nette Gesellschaft innerhalb und außerhalb des Labors

bedanken. Außerdem möchte ich mich bei Annegret, Sabrina, Astrid, Carsten und dem ganzen Team

aus der Werkstatt für eine vielseitige Unterstützung bedanken. Frau Dr. Kühl danke ich für die

zahlreichen TEM Messungen und die Hilfe bei manchen nie endenden, bürokratischen

Bestellvorgängen. Ein großer Dank geht auch an die ganze Kickergemeinde für extrem spannende

Partien am Kickertisch. Dem ganzen Arbeitskreis danke ich für die immer großartige Stimmung, den

Spaß in den Kaffeepausen und so manch feuchtfröhliche Party.

Ich bedanke mich bei allen Mitbewohnern des Büros 304 für die produktive Arbeitsatmosphäre, aber

auch für willkommene Ablenkungen, die Pflege der Pflanzen und ein stets offenes Ohr bei Problemen

jeglicher Art. Ich danke Nhan und Fang für die Einführung und Vorführung asiatischer Kochkünste und

ich danke Vera für die unvergessliche Zeit, den Spaß während Konferenzen und Reisen und die

Unterstützung in jeder Lebenslage.

Ich möchte mich ganz herzlich bei Kathrin, Paulina, Fabio, Nhan und Vera für ihre Zeit bedanken, meine

Arbeit Korrektur zu lesen.

Zu guter Letzt geht mein großer Dank an meine Familie und Freunde. Vielen Dank für eure

Unterstützung, ihr seid die Besten!

III

Zusammenfassung

Wasserstoffbrennstoffzellen bieten als elektrochemische Energiewandler die Möglichkeit, Wasserstoff

aus erneuerbaren Resourcen als saubere und emissionsarme Energiequelle für elektrische Verbraucher,

wie Elektromotoren, zu nutzen. Um eine konkurrenzfähige Alternative zu heutigen

Verbrennungsmotoren in der Automobilindustrie darstellen zu können, müssen aber noch

verschiedenste Hindernisse überwunden werden. Im Blickfeld der Forschung sind dabei vor allem die

Katalysatoren, die größtenteils aus Platin bestehen und an der Sauerstoffreduktionsseite von

Niedertemperaturbrennstoffzellen zum Einsatz kommen. Neben der Maximierung der masse-basierten

Aktivität dieser Katalysatoren, ist vor allem die Langlebigkeit entscheidend, um leistungsfähige

Brennstoffzellen herzustellen. Dabei spielt der Katalysatorträger und sein chemisches und

elektrochemisches Korrosionsverhalten unter Brennstoffzellbedingungen eine entscheidende Rolle.

Hochoberflächige bis graphitisierte Kohlenstoffe sind dabei heute Stand der Technik mit ihren Vor- und

Nachteilen.

Ziel dieser Arbeit ist es, den Einfluss neuartiger oxidischer oder kohlenstoffbasierter Trägermaterialien

auf das elektrochemische und morphologische Stabilitätsverhalten der katalytischen Pt Nanopartikel zu

untersuchen. Solche Katalysatorträger, exemplarisch dargestellt an Metalloxiden und Heteroatom-

modifizierten Kohlenstoffen, wurden mit Pt Nanopartikeln dekoriert und umfangreich auf ihre

physikochemischen Eigenschaften hin untersucht. Dabei kamen verschiedenste in situ

Analysemethoden zum Einsatz, um die Katalysatorkomponenten im simulierten Brennstoffzellenbetrieb

zu evaluieren und mit den Ergebnissen zum fundamentalen Wissen über Degradationsprozesse

beizutragen. Es konnte gezeigt werden, dass Indium Zinn Oxid (ITO) als Trägermaterial unter

simulierten Betriebsbedingungen potentialbereich-abhängige Alterung zeigt. Den Platin Nanopartikeln

konnte eine exzellente strukturelle und morphologische Stabilität nachgewiesen werden. Jedoch

vergiftete Trägerdegradation die Platinoberfläche auf atomarer Ebene, was folglich zu erheblichen

Aktivitätsverlusten führte. Der Modifizierung von hochoberflächigen Kohlenstoffen mit Stickstoff

Heteroatomen konnte ein stabilisierender Effekt auf die Pt Nanopartikel zugewiesen werden. Der

modifizierte Kohlenstoff erwies sich als höchst korrosionsresistent und der Katalysator zeigte eine

außerordentliche morphologische und elektrochemische Stabilität in ausgedehnten

Alterungsprotokollen. Der stabilisierende Effekt konnte auf die Einführung von pyrrolischen

Stickstoffgruppen an der Oberfläche zurückgeführt werden. Des weiteren wurde die elektrochemische

Pt Oxidation auf verschiedenen Trägern untersucht um Rückschlüsse über mögliche

Degradationmechanismen auf Basis von Oxidationsvorgängen zu erhalten.

Diese Arbeit zeigt die Wichtigkeit der Untersuchung von Pt-basierten Brennstoffzellen-katalysatoren

unter simulierten Betriebsbedingungen und die Aufdeckung von Degradationsmechanismen und

stabilisierenden Trägereffekten. Eine detaillierte Kenntnis auf Grundlage von tiefgreifenden Studien ist

essentiell für die Weiterentwicklung effizienter Brennstoffzellen als Bestandteil einer emissionsärmeren

Gesellschaft.

Abstract

Hydrogen fuel cells offer the possibility as devices for electrochemical energy conversion to utilize

hydrogen from renewable sources as a clean and low-emission energy source for electrical motors.

However, to be able to represent a competitive alternative to todays internal combustion engines in the

automotive sector, several difficulties have to be overcome. Especially the catalysts, that are employed

at the cathode side for the oxygen reduction reaction (ORR) in low temperature fuel cells and mostly

consist of the scarce metal platinum, are studied extensively. Besides maximizing mass-based activities

of these catalysts, the durability is of tremendous importance for the design of efficient fuel cells.

Thereby, the catalyst support and its chemical and electrochemical corrosion behavior under operating

conditions plays a crucial role for the overall the stability. Nowadays, the use of graphitized and high

surface area carbons as supports offer advantages as well as disadvantages.

This thesis aims at understanding the influence of alternative oxidic or carbonaceous catalyst supports

on the electrochemical and morphological behavior of the catalytic Pt nanoparticles.

The use of alternative catalyst support materials is exemplified by the choice of metal oxides and

modified carbons, that were decorated with Pt nanoparticles and extensively studied with regard to their

physicochemical properties. Different in situ methods are employed to monitor the catalyst/support

components under simulated fuel cell operating conditions and to contribute to the scientific knowledge

on fundamental degradation processes. It could be shown that indium tin oxide (ITO) used as support

degrades depending on the applied potential range under simulated operating conditions. Pt

nanoparticles were found to have an excellent structural and morphological stability. However, support

degradation lead to a poisoning of the Pt surface on an atomic level and consequently to activity

deterioration. Modification of high surface area carbon with nitrogen heteroatoms resulted in an

increased corrosion resistance and was was further proven to have a stabilizing effect on the Pt

nanoparticles. The catalyst showed extraordinary morphological and electrochemical stability in long-

term stress tests. This was ascribed to surface modification in the form of introduction of pyrrolic N

groups as most abundant surface species. Furthermore, the electrochemical oxidation of Pt on different

supports was studied by various in situ methods to track the structural response of the Pt nanoparticles

and to draw conclusions about possible degradation pathways as a consequence of oxidation processes.

Together, this work illustrates the importance of investigating Pt-based fuel cell catalysts under

simulated working conditions and revealing degradation mechanisms and support-related stabilizing

effects. A detailed knowledge based on profound studies is essential for the ongoing development of

efficient fuel cells as part of a clean energy society.

 
VII 
 
Table of Contents 
 
Danksagung ............................................................................................................................... I 
Zusammenfassung ................................................................................................................. III 
Abstract .................................................................................................................................... V 
Introduction ....................................................................................................................... 1 
1. Hydrogen Fuel Cell ...................................................................................................... 3 
2. Degradation Mechanisms of supported, nanoscale Pt ORR Catalysts ........................ 7 
3. Metal Oxides as Supports for Pt Nanoparticles ......................................................... 11 
4. Carbon Heteroatom Modification and its Application............................................... 13 
5. Electrochemical Pt Oxidation .................................................................................... 15 
Motivation and Goals ...................................................................................................... 17 
Experimental Part ........................................................................................................... 19 
1. Synthesis Procedures ................................................................................................. 19 
1.1. Pt on Indium Tin Oxide .......................................................................................... 20 
1.2. Pt on Ruthenium Titanium Oxide .......................................................................... 20 
1.3. Pt on modified Vulcan............................................................................................ 21 
2. Physicochemical Characterization ............................................................................. 22 
2.1. X-ray Diffraction (XRD) ........................................................................................ 23 
2.2. Transmission Electron Microscopy (TEM) ........................................................... 23 
2.3. Inductively Coupled Plasma – Optical Emission Spectroscopy (ICP-OES).......... 24 
2.4. High-angle Annular Dark Field (HAADF) Scanning Transmission Electron 
Microscopy (STEM) and Energy-dispersive X-ray (EDX) Spectroscopy ........................... 24 
2.5. Elemental Analysis (EA) ........................................................................................ 24 
2.6. Nitrogen Physisorption by BET ............................................................................. 25 
2.7. Zeta Potential (ZP) ................................................................................................. 25 
2.8. High Temperature – Differential Electrochemical Mass Spectroscopy 
(HT-DEMS)….. .................................................................................................................... 25 
2.9. X-ray Photoelectron Spectroscopy (XPS) .............................................................. 26 
3. Electrochemical Characterization .............................................................................. 28 
3.1. Preparation ............................................................................................................. 28 
3.2. Activity and Stability Measurements ..................................................................... 29 
4. In situ Characterization .............................................................................................. 33 
4.1. Scanning Flow Cell Inductively Coupled Plasma Mass Spectroscopy 
(SFC ICP-MS)…… .............................................................................................................. 33 
4.2. In situ Electrochemical Cell and Setup .................................................................. 33 

 
VIII 
 
4.3. High Energy X-ray Diffraction (HE-XRD) ........................................................... 35 
4.4. Small Angle X-ray Scattering (SAXS) .................................................................. 37 
4.5. X-ray Absorption Spectroscopy (XAS) ................................................................. 40 
Unravelling Degradation Pathways of Oxide-Supported Pt Fuel Cell Nanocatalysts 
under In Situ Operating Conditions ..................................................................................... 41 
1. Introduction ............................................................................................................... 43 
2. Physicochemical Characterization ............................................................................ 43 
3. Electrochemical Characterization.............................................................................. 46 
4. XPS and STEM/EDX Results ................................................................................... 49 
5. In Situ HE X-ray Investigation .................................................................................. 51 
6. In Situ Pt, In and Sn Dissolution by SFC ICP-MS .................................................... 56 
7. Discussion ................................................................................................................. 58 
8. Conclusion ................................................................................................................. 63 
The Impact of Carbon Support Functionalization on the Electrochemical Stability 
of Pt Fuel Cell Catalysts ........................................................................................................ 65 
1. Introduction ............................................................................................................... 67 
2. Compositional and Surface Characterization ............................................................ 67 
3. Analysis on Carbon Surface Functionalization by XPS ............................................ 69 
4. Corrosion Behaviour by HT-DEMS.......................................................................... 72 
5. Pt Deposition and ORR Stability............................................................................... 73 
6. Conclusion and Summary ......................................................................................... 80 
On the Anisotropy of Pt Nanoparticles on Carbon- and Oxide-Support and Their 
Structural Response to Electrochemical Oxidation ............................................................ 81 
1. Introduction ............................................................................................................... 83 
2. Structure and Morphology......................................................................................... 83 
3. Electrochemical Characterization.............................................................................. 87 
4. In Situ Electrochemical Pt Oxidation ........................................................................ 89 
5. Conclusion ................................................................................................................. 99 
Summary and Outlook ................................................................................................. 101 
1. In Situ Stability Study of Pt Supported on Indium Tin Oxide................................. 101 
2. Carbon Heteroatom Modification in Pt/C ORR Catalysts ...................................... 103 
3. Electrochemical Oxidation of Pt on different Supports .......................................... 104 
4. Outlook .................................................................................................................... 106 
References ...................................................................................................................... 109 
Appendix ............................................................................................................................... 119 
A1 Supporting Information to Chapter 452 .................................................................... 119 

A2 Supporting Information to Chapter 571 .................................................................... 124

A3 Supporting Information to Chapter 6 ....................................................................... 131

List of Acronyms ................................................................................................................... 135

List of Chemicals .................................................................................................................. 137

List of Figures ....................................................................................................................... 138

List of Tables ......................................................................................................................... 143

List of Publications ............................................................................................................... 144

1. Introduction

The fact that between the years 2014 and 2016 the world’s carbon dioxide (CO2) emissions

seemed to remain on a stable level was a great achievement for todays society. However, in

2017 it was reported that CO2 emissions increased again by 2 %. This was mainly due to the

fact, that the observed decreasing emissions in 2017 from the US and Europe were unable to

offset steadily increasing emissions from India, China and the rest of the world, resulting in an

average global emission growth.1 Furthermore, it was observed that not only CO2 emissions

grew, but also the coal consumption increased in 2017 related to a rising need for energy.2

With a current average atmospheric CO2 concentration of 410.8 ppm (June 2018) this value is

as high as it ever was.3 Future trends are hard to predict due to relative large uncertainties in

estimating average CO2 emissions for different countries.4 CO2, as the main share of greenhouse

gas emissions and one of the biggest cause for global warming, mainly arises from the

combustion of fossil fuels in the industry and transportation sectors. Therefore, it has become

a major challenge to decouple a steadily increasing world gross product and world population

from the production of greenhouse gases.

Figure 1.1 Schematic illustration of the use of Hydrogen for energy conversion and storage. In here,

hydrogen is produced by water electrolysis powered by renewable energy sources. Hydrogen is stored

and distributed and can be used on demand in a fuel cell to generate electricity.

However, an increasing world population and energy demands of developing nations require

the expansion and utilization of clean energy sources. An expanded awareness for sustainability

of todays’ society combined with alternative and emission-neutral energy resources might be

Fuel Cell

Conversion

Electricity

Storage

Transportation

Distribution

H2O

Generation

Electrolysis

Electricity

Renewable

Energy Source

•Solar

•Wind

•Hydro

•Bio Mass Air

1. Introduction

the key in preventing a potentially upcoming environmental crisis. Renewable energies such as

wind, solar, hydro and bio mass represent a class of clean energy sources for implementation

in the energy grid to provide a sufficient part of the electricity demand. For example, Germany

has achieved a pioneering role in the last decade in generating electricity from renewable

sources with a contribution of electricity by renewable sources of 39 % in 2017, potentially

further increasing in future.5

However, renewable energy sources are intermittent with times when the supply does not meet

the demand, or when the supply surpasses the demand.6 Therefore, systems for both energy

storage and conversion are needed in order to overcome grit instabilities and ensure reliable

energy supply. One CO2 emission-neutral way to use excess power generated by renewable

energy sources is the electrolysis of water. The applied electricity enables an efficient splitting

of water into its components, oxygen (O2) and hydrogen (H2), see Figure 1.1. In this way,

energy generated by renewable resources can be stored chemically in the H2 molecule.

Hydrogen offers the possibility to distribute and transport energy which makes it available and

promising for multiple applications. It can be stored stationary (e.g. in metal hydrides, liquified

or compressed in tanks)7 and distributed for the use in industrial applications or the flexible

reconversion into water and electricity using a fuel cell device.

Especially for implementation in the transportation sector, low temperature proton exchange

membrane fuel cells (PEMFC) offer great potential due to their relative efficient fuel conversion

rates, reliability and emission-free character. Several difficulties in sufficient hydrogen storage

and transportation and establishing a hydrogen fuel grid as well as technologies for electrolyzers

and fuel cells that are comparable to state-of-the-art production/combustion methods are still in

the way of establishing a reliable hydrogen community.

1. Introduction

1.1. Hydrogen Fuel Cell

A fuel cell is a device that is able to efficiently convert chemical energy into electricity and

heat. The principle of a hydrogen fuel cell was first observed by W. Grove in 1839, who

demonstrated the generation of electricity by two separate Pt electrodes immersed in diluted

sulphuric acid in separated hydrogen and oxygen compartments.8 A single device was called a

“gas battery” and a multitude of “gas batteries”, connected in a row, was called a “gas chain”.

Grove used the generated electricity to produce hydrogen in an electrolyzer.9 The first real

application of a fuel cell was achieved by the NASA in their space program in 1960 where a

fuel cell was used to supply the Gemini and Apollo spacecrafts with electricity and drinking

water.

Hydrogen fuel cells are classified by the electrolyte used and by their operating temperature.

Proton exchange membrane fuel cells are typically run at 80 °C and use a Nafion® membrane

for proton conduction. Alkaline fuel cells (AFC) operate at around 100 °C and due to the lack

of proper anion conductive membranes, a liquid electrolyte containing potassium hydroxide is

commonly employed. Molten carbonate fuel cells (MCFC) are one example for fuel cells

operating at higher temperatures. In here, the electrolyte consists of liquid solution of lithium

and sodium carbonates that are soaked in a matrix, allowing operating temperatures up to

1000 °C. In a solid oxide fuel cell (SOFC), running at comparable temperatures to the MCFC,

conduction of oxygen anions is enabled by a yttria-stabilized zirconia (YSZ) type membrane.10

Figure 1.2 Schematic illustration of the general principle of a proton exchange membrane fuel cell

(PEMFC). At the anode the fuel (hydrogen) is oxidized to form protons and electrons. Protons travel

through the membrane to the cathode to form water by reacting with oxygen and electrons. The electrons

travel through an external electric circuit and can be used as power supply.

e-e-

Water

Excess H2

Fuel Input

Output

Air Input

Water/Heat Output

Anode

H22H++ 2e-

Cathode

O2+ 4H++ 4e-2H2O

PEM

Electric Circuit

1. Introduction

The development of PEMFC has increased in the past decades opening its way up to large-scale

applications especially in the transportation sector. Cars equipped with hydrogen fuel cells are

offered by different manufacturers such as Toyota, Honda or Hyundai.11 Fuel cell buses for

public transportation are already used in several cities around the world 12,13 and recently, a fuel

cell train was implemented into the rail network in northern Germany14.

The general principle of a PEMFC is schematically illustrated in Figure 1.2. A PEMFC

comprises of two electrodes that are separated by a proton conducting membrane. At the anode,

hydrogen as the fuel is supplied and oxidized to form protons and electrons, see Figure 1.2. The

protons can mitigate through the membrane to the cathode side. In a PEMFC, the membrane

consists of a polymer, a perfluorosulfonic acid (typically Nafion®) that, when hydrated, is

solely conductive for protons and provides the separation for the two electrodes. Hence, the

electrons are forced to reach the cathode via an external electric circuit that gives the

opportunity to use the electricity by an external device. At the cathode side, oxygen (typically

from air) is reduced by reacting with electrons and protons to form water. Catalysts employed

at both electrodes usually consist of carbon supported Pt nanoparticles, separated at each side

to gas flow fields by gas diffusion backings to provide sufficient gas concentrations at the

catalyst surface. Such individual fuel cells, called membrane electrode assemblies (MEAs), can

be aligned in rows to generate a stack. Stacks of different sizes and different power ouputs, in

which a single cell usually provides around 0.7 V, can be designed with respect to the desired

application.15

As mentioned before, PEMFC development and commercialization has made great progress

recently, still there are several challenges to overcome. Especially in fuel cell application for

transportation, storage and distribution of hydrogen becomes an additional issue as a sufficient

hydrogen infrastructure is yet not existent. Moreover, costs of fuel cell cars are still not able to

compete with cars equipped with state-of-the-art internal combustion engines. The reason for

that is the costs of a fuel cell stack for automotive application are strongly dependent on the

quantity produced. They were estimated to be 50 $/kWnet (for a 80 kWnet PEMFC system) when

manufactured at a volume of 100,000 units/yr in 2017. Costs are targeted to decrease to

40 $/kWnet by 2025 for such a manufacturing quantity as claimed by the US Department of

Energy.16 Major contributor to the fuel cell stack costs are not only the noble metal catalysts

employed at both electrodes but also other fuel cell components such as the bipolar plates, the

ionomer and the gas (re)circulation systems.16

1. Introduction

When operating a fuel cell, the theoretical available cell voltage (𝛥𝐸) for a single cell for the

hydrogen/air fuel cell reaction according to

𝐻2+ 1

2𝑂2 → 𝐻2𝑂

can be calculated by equation (1), with 𝑛 being the number of transferred electrons per H2

molecule, 𝐹 Faraday’s constant and 𝛥𝐺 the Gibbs free energy for the above reaction.

𝛥𝐸 = − 𝛥𝐺

𝑛𝐹

(1)

According to this equation, a single fuel cell gives a theoretical cell voltage (thermodynamic

reversible cell potential) of 1.23 V at 1 atm pressure and 25 °C. When a fuel cell is operated at

80 °C, this value reduces to 1.18 V due to the temperature dependency of the Gibbs free energy.

However, as mentioned above, single fuel cells nowadays run at around 0.7 V which means

that not 100 % of the chemical energy can be converted into electrical energy but partially into

heat. Efficiency losses observed in a running fuel cell and their origins will be briefly described

in the following paragraphs.

Activation losses are the largest contributor to a decreased efficiency and are originating from

sluggish kinetics of the oxygen reduction reaction (ORR) at the cathode. In comparison, the

hydrogen oxidation reaction (HOR) at the anode, is much faster and the total activation losses

are therefore almost entirely caused by the ORR. In contrast to the fast, two-step oxidation of

hydrogen at the anode, the ORR at the cathode involves four elementary reduction steps

including four single transfers of an electron and a proton. A possible ORR reaction pathway is

proposed as follows:17,18

𝑂2(𝑔)+ 4𝐻++4𝑒−+ ∗ → 𝑂𝑂𝐻∗+ 3𝐻+ +3𝑒−

𝑂𝑂𝐻∗+ 3𝐻+ +3𝑒−→ 𝑂∗+ 𝐻2𝑂(𝑙)+2𝐻++2𝑒−

𝑂∗+ 𝐻2𝑂(𝑙)+2𝐻++2𝑒−→ 𝑂𝐻∗+ 𝐻2𝑂(𝑙)+ 𝐻++ 𝑒−

𝑂𝐻∗+ 𝐻2𝑂(𝑙)+ 𝐻++ 𝑒−→2𝐻2𝑂(𝑙)

In this reaction equations, three different surface intermediates (𝑂𝑂𝐻∗,𝑂∗ 𝑎𝑛𝑑 𝑂𝐻∗) are

involved and adsorbed with different binding strength on the catalyst surface. The total ORR

rate strongly depends on the binding strength of the intermediate involved in the rate limiting

step, that is, the step with the highest energy barrier. The goal is to minimize energy barriers

for ad- and desorption and, thus, to facilitate each reaction step resulting in improved ORR

1. Introduction

reaction rates. Therfore, the right choice of catalyst (usually Pt and its alloys19), that enables an

optimized binding strength to those intermediates, is of tremendous importance.

Another cause for efficiency drops of fuel cells are Ohmic losses resulting from internal

resistances of fuel cell components including the membrane, electrodes, bipolar plates and the

interconnections. Fuel crossover can additionally cause an increase in overpotential due to fuel

that directly crosses from the anode through the membrane without being oxidized.

Furthermore, at high reaction rates, thus at high current densities, concentration losses can

occur. They result from mass transport limitations at both electrode sites where the

concentration of the gases reduces at high reaction rates. At the cathode the increased formation

of water at high current densities leading to flooding inhibits the transport of oxygen to the

catalyst surface which hampers the overall efficiency.

1. Introduction

1.2. Degradation Mechanisms of supported, nanoscale Pt ORR Catalysts

As mentioned above, Pt based materials are commonly employed at the PEM cathode as

catalysts for the ORR as they enable binding strengths of the intermediates close to the

optimum, making them excellent ORR catalysts. While maximizing the ORR activity by

engineering the Pt catalysts structure and morphology18,20-24, the stability behaviour of the

catalyst/support couple is of similar importance. In addition, when the catalyst/support

ensemble degrades during the lifetime of a fuel cell, severe performance losses are observed,

making this one of the biggest drawbacks in fuel cell application and large-scale

commercialization. On the way to competing with automotive internal combustion engines, it

was reported by the US Department of Energy (DoE) that the system durability still remains

the most unresolved issue and hence, has to be improved. On the other hand, system cost, energy

efficiency and specific power density are close to meeting the targets.25

In order to improve the overall durability, a detailed knowledge on fuel cell related degradation

phenomena has to be established. Therefore, the cathode catalyst lifetime behaviour has been

intensively studied in PEMFC research due to its major impact on the overall durability of a

running fuel cell. With an anticipated ten year lifetime of a fuel cell car (equals 5,000 operation

hours or 240,000 km driving distance) as proposed by the DoE16, it has become an important

challenge to investigate and understand the processes that contribute to activity deterioration in

PEMFC catalysts. Especially with the aim of continuously lowering Pt loadings, Pt degradation

is hardly acceptable as it makes the impact on performance loss even more fierce.

Several different degradation mechanisms, including Pt dissolution, detachment and

agglomeration as well as Ostwald ripening and carbon corrosion have been identified and

investigated in the past (see Figure 1.3) and will be described point by point in the following.

As this work focuses on degradation processes for monometallic Pt nanoparticles and the

support components, alloy PtX (X = transition metal) related degradation phenomena will not

be discussed.

1. Introduction

Figure 1.3 Schematic illustration of proposed degradation mechanisms for Pt nanoparticles supported

on carbon including Pt dissolution, Ostwald ripening, agglomeration, detachment and carbon corrosion.

Reprinted with permission from Ref. 26 from Beilstein Journal of Nanotechnology under the terms of

the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0).

Pt Dissolution and Ostwald Ripening

Pt dissolution and its role in catalyst degradation has been extensively studied in literature.27-34

The origin for Pt dissolution arises from the formation of Pt oxide and its thermodynamic

(in)stability. The oxidation of Pt is conversely discussed in literature due to the variety in

potential Pt oxidation states and Pt-oxide compounds (mostly PtO, PtO2 and PtO3). However,

the following reaction equations describe one potential acidic dissolution pathway, that is likely

to occur in a fuel cell environment (low pH and potentials close to 1 V).32 It involves the

formation of PtO and its acidic dissolution in the form of Pt2+:

𝑃𝑡+𝐻2𝑂→ 𝑃𝑡𝑂+ 2𝐻++ 2𝑒−

𝑃𝑡𝑂+ 2𝐻+→ 𝑃𝑡2++ 𝐻2𝑂

1. Introduction

Pt dissolution is believed to be more severe for smaller nanoparticles due to their higher surface

energy and can therefore dissolve at lower potentials compared to bulk Pt (Gibbs-Thomson

effect). Different studies observed a strong dependence of the Pt dissolution rate on the particle

size where the critical particle diameter was proposed to be 5 nm (particle size effect).35-37

Furthermore, it was reported that anodic Pt dissolution can be observed at potentials as low as

0.85 V, but more severe dissolution from oxide reduction was monitored at more positive

potentials corresponding to fuel cell start-up/shut-down procedures.38 As a consequence, the

dissolution of Pt from the particles can cause strong performance losses with time. It is

considered as a primary degradation mechanism because it can lead to other phenomena such

as Ostwald ripening.39 In this case, the Pt species that were dissolved from smaller Pt particles

redeposit on larger particles causing overall increase in particle size and surface area loss (3D

Ostwald ripening). Dissolved Pt can furthermore be reduced and deposit in the membrane upon

reaction with cross-over hydrogen from the anode.35,40

Carbon Corrosion

Another primary degradation mechanism is carbon corrosion originating from oxidation of the

carbon support.26,33,41-45 The carbon corrosion reaction in aqueous acidic environment can be

expressed as follows: 𝐶+2𝐻2𝑂 → 𝐶𝑂2+ 4𝐻++ 4𝑒−

The standard potential (E0(C/CO2)) for this reaction is 0.207 V.32 However, due to its slow

kinetics, it is not predominant in a fuel cell operating regime below 1 V. But, as during fuel cell

start-up/shut-down processes high potentials up to 1.5 V can be reached and as carbon corrosion

is believed to be catalyzed by Pt46,47, it can cause a severe loss of the catalyst’s structural

intigrety accompanied by performance deterioration. Changes in the hydrophilicity due to the

formation of oxygenated surface groups by carbon oxidation and concomitant changes in

flooding behaviour and mass transport limitations are other consequences of carbon corrosion.48

Particle Detachment and Agglomeration

Carbon corrosion can furthermore trigger other, secondary, degradation processes such as

detachment41,44,49 and agglomeration.37,50,51 Detachment of whole Pt particles from the support

can be caused by a weakened attachment of the particle to the support initiated by carbon

corrosion. Thus, even though the particles are still present in the cell, they are not connected to

the electrode resulting in both activity and surface area deterioration. Agglomeration or

coalescence of catalyst particles due to the migration and collision of single particles to larger

particles, which might be facilitated by (local) carbon corrosion weakening particle support

1. Introduction

interactions, is another cause for activity degradation. It can lead to a loss in overall particle

count and a decreased active surface area. Phenomena of agglomeration and coalescence were

found to be dependent on the Pt weight loading on the carbon support and on the type of high

surface area carbon used.43

In order to improve the catalysts overall durability, several synthetic and analytical approaches

can be followed. Therefore, the following two chapters (1.3 and 1.4) give a brief literature

overview on possibilities to reduce catalyst degradation by engineering of the support.

Replacement of the carbon support by a metal oxide or a modified carbon, its properties and

influence on the overall structural and electrochemical Pt stability is discussed. Additionally,

chapter 1.5 will a give a more detailed overview on Pt oxidation processes from literature and

its critical role for the development of stable ORR electrocatalysts.

1. Introduction

1.3. Metal Oxides as Supports for Pt Nanoparticles

Reproduced in part from Ref 52 (Adv. Energy Mat., 2018, 8 (4), 1701663) with permission from John Wiley and

Carbon corrosion, as explained before, can cause severe damage on the catalytic active

component accompanied by fuel cell performance and stability losses. One way to completely

prevent carbon corrosion, is the use of alternative, carbon-free catalyst supports. Intensive

research focused on the replacement of carbon as support in fuel cell catalysts.

Metal oxides have recently gained attention in this field because they offer a variety of features

required for a stable electrocatalyst for oxygen reduction.53,54 Essential properties for cathode

catalysts supports are a high surface area, high electrical conductivity, thermal stability, and

above all, their corrosion resistance. Therefore, several metal oxides have been investigated:

Popov and co-workers showed that Pt on TiO2 has a high stability toward ORR caused by the

high corrosion resistance and the strong metal support interaction (SMSI).55 This SMSI effect

was further analyzed by AlonsoVante and co-workers,56 they were able to show changes in the

local electron density of Pt when deposited on titanium oxide-carbon and tungsten oxide-carbon

support. By X-ray photoelectron spectroscopy (XPS) analysis they determined an increase in

the asymmetry of the Pt 4f peak which the authors attributed to either an alloy formation

between Pt and the support or a partial charge transfer from the support to Pt. Furthermore, a

molybdenum-doped Ti dioxide was found to be a stable electrocatalyst because of electronic

structure changes in the Pt upon synergistic interaction with the support material.57 A similar

phenomenon was observed for Pt/Ta0.3Ti0.7O2 where the electrochemical surface area (ECSA)

decreased less compared to Pt/C reference catalyst when exposed to accelerated load cycling

test designed to test the electrocatalyst stability.58,59 The catalyst exhibited good activity when

tested in an operating fuel cell and exceptional catalytic stability after 10,000 cycles of a

simulated start-up/shut-down stress test. The superior catalyst stability was explained with a

lower Pt dissolution rate compared to carbon supported Pt electrocatalyst and the SMSI between

Pt nanoparticles and Ta0.3Ti0.7O2.

Another class of stable carbon-free supports was found to be RuO2/TiO2 and RuO2/SiO2.60-63

These materials offer a conductivity and surface area comparable to mesoporous carbon, a

promising electrochemical stability against aggressive accelerated test protocols besides

improved membrane electrode assembly (MEA) performance. However, incorporation of Pt

nanoparticles with a well-defined size into the electrodes to decrease Ohmic and transport losses

has to be improved to make it a viable alternative in automotive application. Furthermore, Dou

1. Introduction

et al. reported on tin oxide as another candidate for carbon-free support in PEMFCs.64 The

material retained 27% of its initial ECSA while the Pt/C reference catalyst showed almost no

more surface area after 2,000 cycles.64 In another approach, Tsukatsune et al. found that their

Pt/SnO2 and Pt/Nb-SnO2 electrocatalysts suffer from low specifc surface area accompanied

with low specific activity.65

Cognard et al. investigated antimony-doped tin oxide supported Pt electrocatalyst as cathode

material for PEMFC.66 The harsh accelerated stress test (AST) conditions of potential jumps

between 1.0 and 1.5 V lead to depletion of Sb atoms from the support surface and a decreased

conductivity and hence to a loss in catalytic activity. Indium tin oxide (ITO) offers several

advantages in regard to the conditions that are present in the environment of fuel cell catalysts

at the cathode side such as low pH, high temperatures, and high potentials. Additionally, ITO

has the advantage of low electrical resistance. Pt/ITO catalyst systems have been tested as

alternative material to conventional Pt nanoparticles dispersed on high surface area carbon. Liu

et al. showed that a Pt/ITO ORR electrocatalyst offered activity as well as enhanced stability

during an AST.67-69 These Pt nanoparticles were deposited on the ITO via a galvanic

displacement of a Cu monolayer from a K2PtCl4 precursor. While the authors did not provide

evidence for a complete displacement or removal of contaminating Cu residues, their data

suggested that the enhanced ORR activity is caused by synergistic effects between surface Sn

and Pt.69 A recent XPS study by Wang et al. of Pt/ITO as electrocatalyst for PEMFC MEAs

revealed strong changes in the indium oxide component after operating conditions.70 These

changes resulted in poor conductivity of the support followed by large activity losses.

1. Introduction

1.4. Carbon Heteroatom Modification and its Application

Reprinted in part with permission from Ref 71 (Chem. Mat., 2018, 30 (20), 7287-7295). Copyright (2018) American

Chemical Society.

Instead of employing carbon-free materials as Pt support, one can also change the properties of

the high surface area carbons in order to make them more stable in a fuel cell environment.

Furthermore, as a result of improving the carbon corrosion resistance, other secondary

degradation processes might also be suppressed.

Porous carbon materials are the support material of choice for noble metal nanoparticle

catalysts. To make for a good catalyst support, the set of required properties range from high

surface area and porosity to sufficient conductivity, corrosion resistance and chemical stability.

Different carbons are usually classified by their BET surface area that is in the range of 200-

400 m2 g-1for Vulcan and CNTs, but can be as high as 1300 m2 g-1 for types of Ketjen black,

Black Pearls72 and hollow graphitic/carbon spheres (HGS/HCS)73,74. Among the large variety

of known porous carbon materials, quite a few carbons are known to meet at least some of the

criteria mentioned before and have therefore been studied extensively as supports.

The introduction of new chemical (surface) functionalities offers ample possibilities to tune a

carbonaceous material with respect to its electronic, structural, morphological, and hence

reactive properties and has emerged as a very popular method in the search for improved

catalyst supports.

Nitrogen is one prominent heteroatom used for introducing functional surface groups or for

modification of the carbon backbone. In applications for supercapacitors, it was found that

functionalization by pyrolysis of organic salts enhances both capacitance and cycle durability

of non-porous carbons.75,76 The efficiency towards electrochemical hydrogen peroxide

production was proven to be superior in the nitrogen-doped ordered mesoporous carbon CMK-

3 over a wide pH-range.72 Modified carbon-based materials were furthermore tested as metal-

free electrocatalysts towards the oxygen reduction reaction (ORR): nitrogen-doped multi-

walled carbon nanotubes (N-MWCNT) showed outstanding activity in alkaline medium.77,78

Functionalization of Vulcan carbon by coating with zeolitic imidazolate frameworks (ZIF) was

reported to enhance both ORR activity and stability.79 Much work was devoted to analyze and

understand the degree and nature of carbon modification/functionalization in various types of

carbons ranging from graphene to CNTs and carbon blacks.80-82 Ammonia treatment of pre-

oxidized carbons is one promising way of introducing N-functional groups in relative high

1. Introduction

concentrations on the carbon surface. While theoretical models have been employed to predict

the gasification rate and the creation of an internal porous network83, other studies were

focusing on acid-base properties of N-doped nanocarbons84,85, the thermal stability86 and the

translation of this method to a wider range of carbon-based materials87.

When modified carbons are used as support for nanoscale noble metal catalysts, a wide range

of application in heterogeneous catalysis such as hydrogenation reactions88, methanol

oxidation89 and selective oxidation90 opens up. Along with latest developments in the research

field of PEMFC, functionalized carbons are widely employed as electrocatalyst support to

stabilize the active component (mostly platinum) from altering under simulated working

conditions.91-96 When hollow carbon-spheres were doped with nitrogen by dopamine treatment

and Pt was deposited, an improved Pt distribution was observed which the authors ascribed to

interactions between the inserted nitrogen atoms and Pt as binding energy shifts from N 1s and

Pt 4f XP spectra were noticed. These interactions were further believed to cause the enhanced

ORR long-term stability.94 Pt supported on N-containing mesoporous carbon obtained from

ionic liquids showed higher electrochemical active surface area compared to a commercial Pt/C

catalyst with similar particle size.92 N-doped carbon shells around Pt on CNTs were obtained

by a subsequent polymerization of aniline around the Pt precursor impregnated on CNTs,

followed by reduction of the Pt precursor. The resulting composite showed enhanced stability

against any kind of Pt degradation which was explained by the stabilizing effect of the porous

carbon shells around the Pt.91 Pt on a polyaniline (PANI) based support was tested in an MEA

setup and showed first promising results for the application of Pt ORR catalysts on an

alternative carbon support.97 Another beneficial effect of surface modified carbons was recently

reported to be a superior distribution of the Nafion ionomer in the MEA.98 The enhanced fuel

cell performance was attributed to interactions between the N-modified carbon surface and the

sulfonic groups in the ionomer resulting in decreased local, that is, non Fickian mass transport

losses at high current density under hydrogen-air conditions.

1. Introduction

1.5. Electrochemical Pt Oxidation

As mentioned above, surface processes such as Pt oxidation have key influence on the catalysts

stability because the degradation due to Pt dissolution is directly linked to the formation of

PtOx-species and the consecutively reductive dissolution. Therefore, knowledge about the

origin of these surface processes is of great interest to understand the degradation behavior on

a fundamental level.

On Pt single crystal surfaces it was found that Pt oxidation occurs in the form of place exchange

of Pt and O atoms on the Pt(111) surface.99,100 By in situ surface X-ray diffraction (XRD) it was

shown that with increasing potential the oxide structure consists of a Pt-rich inner and an O-

rich outer layer which then transforms to a more homogenous oxide at above 1.1 V. By cycling

back to more negative potentials, the authors could proof that up to a critical oxide coverage

the process of place exchange is quasi-reversible because place-exchanged atoms are located

on top of their original positions. In another study by Feliu et al. no changes in the

voltammograms were observed by consecutive cycling up to 1.15 V evidencing reversible

ordering of Pt(111).101

The structural sensitivity of intermediate stages of Pt oxidation was investigated by surface-

enhanced Raman spectroscopy on Pt(111) and Pt(100). Therein, it was reported that place-

exchange and the formation of an amorphous 3D α-PtO2 proceed sequentially on the Pt(111)

whereas on Pt(100) both processes proceed simultaneously above 1.1 V.102 Studies on a

monolayer of Pt deposited on a Rh(111) single crystal showed the evolution of a Pt oxide

structure by electrochemical oxidation up to high potentials of 1.6 V while at potentials below

1 V chemisorbed oxygen-containing species at the Pt surface were proposed. The onset of Pt

oxide formation was found to be dependent on the kinetic stability of chemisorbed species

resulting from different surfaces (single crystal vs. nanoparticle). By combined computational

and experimental methods, Pt oxide in the form of Pt3O4 was in best agreement with both

methods.103

Furthermore, high-energy-resolution X-ray absorption showed that for small Pt nanoparticles

different adsorbed species on the Pt surface (hydrogen and O/OH, Pt oxide) can be

distinguished in certain potential regions.104 Onset of Pt oxidation was found at 0.96 V and with

that, somewhat lower than on single crystals, which can be explained by a stronger bond of the

Pt nanoparticle surface to oxygen. In another study by Adzic et al., a wide potential range up

to 2.6 V was investigated by in situ X-ray methods.105 According to the authors, O and OH

1. Introduction

adsorption at around 0.7 V and place exchange onset at 0.9 V was observed resulting in the

formation of an α-PtO2 which was dissolved at high potentials above 1.9 V. Additionally it was

found that Pt oxidation only occurs at the outer Pt layers while the inner, bulk part remains

metallic. In a work by Imai et al. a time-resolved X-ray diffraction and energy dispersive X-ray

absorption study on Pt nanoparticles also proposes the formation of an α-PtO2 by analyzing the

Pt-O bond length.106

A study on the potential dependent structural transformation of Pt on different carbon supports

in a potential window from 0.4 to 1.4 V showed that Pt nanoparticle surfaces after aging were

oxidized to Pt(2+)O at 1.4 V. Differences were observed for the onset of these oxide formations

and the reduction to the metallic state for the different supports from full reversibility to

irreversibility.107

Pt oxidation was also studied on nanoparticles supported on a metal oxide. For a ceria support

it was shown by XPS and TEM that the oxide support has an impact on the formation Pt oxide

as a composite of Pt oxide and Ce was formed, but it was reversibly reconverted to metallic

Pt.108 The authors attribute this phenomenon as evidence for strong interaction between CeO2

and Pt. Formation of Pt-O-Ce bonds stabilizes Pt from sintering under oxidative atmosphere.

While these studies were conducted at different temperatures and gas atmospheres and not in

real electrochemical environments, it still gives a first impression of Pt oxidation/reduction in

oxide-supported Pt catalysts.

2. Motivation and Goals

The overall objective of this work is to expand our current knowledge and understanding of

degradation mechanisms of supported Pt nanoparticle catalysts for PEMFC cathodes. Emphasis

is placed on a molecular-level understanding of the ORR activity and cycling stability of Pt

catalyst/support couples involving new support materials, in particular oxides and doped

carbons, as alternatives to traditional carbon blacks. Understanding is generated through

structure-composition-activity-stability relationships established using in situ analytics.

The fuel cell cathode environment with its time-variable, quite anodic potentials generates harsh

conditions for both the nano-scale catalysts and support components. Detailed knowledge about

the exact degradation mechanisms of new support materials is still unknown. This might be due

to the fact that, processes such as Pt dissolution, carbon corrosion, particle detachment and

coalescence or ripening are oftentimes coupled to each other. Therefore, the application of

several different spectroscopic and analytical (in situ) methods is the key to gain a

comprehensive knowledge. Simulated fuel cell environments and accelerated degradation

protocols will help to evaluate the catalysts suitability for fuel cell application on an early stage

of the development process.

To contribute to the general goals mentioned above, this work will focus on the following

projects:

 In the first part (chapter 4), an indium tin oxide (ITO) supported Pt electrocatalyst is

analyzed and evaluated for its suitability as durable fuel cell catalyst. Previous works

reported Pt/ITO as a promising catalyst system, but only little work had been done to

understand how the entire catalyst/support couple behaves directly under AST

conditions with special emphasis on Pt particle growth and agglomeration as well as

support degradation. Therefore, Pt/ITO is first intensively characterized by

physicochemical methods followed by the application of accelerated stress tests in fuel

cell relevant potential windows. Changes in Pt mass-based activity and electrochemicl

active surface area upon extended potential cycling are carefully followed. In a next

step, several different complementary methods are applied to track structural,

compositional and morphological changes in both Pt catalyst and metal oxide support

and to link those changes to trends in electrochemical activity and surface area of

Pt/ITO. With help of different in situ (mostly X-ray based) techniques, new insight on

the operating behaviour of this novel class of electrocatalyst is obtained.

2. Motivation and Goals

 In chapter 5 findings on the impact of carbon support modification by doping with

heteroatoms on its physicochemical properties and on the electrochemical Pt stability

are presented. A commercial, state-of-the-art high surface area carbon is modified in a

simple two step approach involving the acidic oxidation followed by amination at

elevated temperatures resulting in a family of heteroatom-doped carbons. The changes

in surface area, surface potential and bulk as well as surface composition upon this

treatment are reported. A special emphasis is put on the behaviour towards carbon

corrosion as it is highlighted to be a major contributor to PEMFC performance

deterioration. Furthermore, when used as Pt catalyst support, the influence of the

support heteroatom doping on the electrochemical long-term stability is investigated.

The application of in situ wide and small angle X-ray scattering methods contribute to

the understanding about the catalyst’s stability. The nature of heteroatom doping in form

of nitrogen incorporation in the carbon matrix is analyzed to determine the most

abundant surface species and to draw conclusion as to what extent the surface

modification influences the stability of the Pt nanoparticles.

 Chapter 6 addresses the electrochemical oxidation of supported Pt nanoparticles in a

detailed study applying different in situ methods. Three different catalyst supports

(indium tin oxide (ITO), mixed ruthenium titanium oxide (RTO) and carbon) are used

in order to determine its influence on the Pt oxidation behavior. The catalyst/support

systems are first intensively studied regarding their structure and morphology revealing

differences in particle shape and electronic structure depending on the different supports

and the different Pt synthesis routes. The changes of Pt nanoparticles upon

electrochemical oxidation up to 1.5 V are followed by in situ X-ray diffraction and

absorption. Rietveld refinement and peak fitting of the absorption spectra reveal the

potential-dependent structural responses of the Pt catalyst on the different supports.

Parameters such as Pt crystallite size, Pt lattice parameter, Pt scale factor as well as the

white line area and its components are analyzed as a function of applied oxidizing (and

reducing) potentials. Furthermore, in situ dissolution experiments show trends in Ptn+

dissolution to complement the findings from X-ray investigations and to link changes

in crystallite parameters and metallic and oxidized domains to the dissolution rate. With

that, support- and morphology-depending properties should help to understand the

nature of Pt oxidation at a fundamental level.

3. Experimental Part

Reproduced in part with permission from Ref 52 (Adv. Energy Mat., 2018, 8 (4), 1701663) with permission from

American Chemical Society.

The following sections describe all synthetic procedures including synthesis of the metal oxide

supports, the modification of the carbon support and the synthesis and deposition of Pt

nanoparticles performed in this work. Afterwards all methods for physicochemical

characterization as well as the procedures for electrochemical activity and stability

measurements will be described. In the end of this chapter, the experiemental details on the

performed in situ analytical and X-ray methods will be given including a brief explanation on

the theoretical background of the most frequently methods used in this work.

Detailed information on all chemicals used in this work, can be found in the List of Chemicals

on page 137.

3.1. Synthesis Procedures

In this work, two different families of catalyst systems were synthesized and analyzed, Table

3.1 gives an overview over the catalysts and in the corresponding chapters they are studied. The

first set of catalysts are metal oxide supported Pt nanoparticles that are analyzed in chapter 4

and 6. The second family of materials are Pt nanoparticles, that are supported on modified

carbon materials. The results of this study are shown in chapter 5. The synthesis of metal oxide

support, Pt nanoparticles and the process of carbon modification will be described in the

following.

Table 3.1 Overview over different materials used in this work with the corresponding chapters in which

their characterization is discussed.

Material

Chapter

Pt/ITO

4, 6

Pt/RTO

Pt/modified Vulcan

3. Experimental Part

3.1.1. Pt on Indium Tin Oxide

Indium Tin Oxide

In(Cl)3 (1.434 g) was dissolved in deionized (DI) water (2.16 mL) and the solution was stirred

for 10 min. Then SnCl4 (79.2 µL) was added to the solution under stirring for 10 min followed

by addition of ethanol (9 mL). Afterwards propylene epoxide (5.5 mL) was added drop-by-drop

to the alcoholic solution at room temperature under continuous stirring for 15 min. When the

solution turns milky, stirring was continued for another 10 s and put into an ice bath afterwards

for 10 min. The resulting gel was aged for 10 h and then subjected to a series of solvent

exchange in acetone for 10 times in 3 days. The acetone-loaded gels were introduced into a

supercritical dryer (SFT-100, Supercritical Fluid Technologies, Inc.), and acetone was removed

using supercritical CO2 extraction (45 °C and 275.8 bar). The dry gel was further calcined at

820 °C for 150 minutes in air.

Pt Deposition on ITO

Platinum nanoparticles were synthesized using a solvothermal synthesis route based on

previous work.109 In detail, Pt(acac)2 (0.6 mmol), 1,2-tetradecanediol (1.2 mmol), oleylamine

(300 µL), oleic acid (300 µL) and 100 mg ITO were dispersed in dibenzylether (50 mL). The

mixture was stirred under nitrogen atmosphere to remove excess of oxygen. Then, the

temperature was raised to 80 °C and held for 5 min to ensure complete dissolution of Pt

precursor and the reducing agent. Afterwards, the temperature was raised with the ramping of

1 °C min-1 to 165 °C and kept for 1h. The mixture was cooled down to room temperature and

ethanol (80 mL) was added, followed by sonication with an ultrasonic horn for 1 h and stirring

overnight. The supernatant was removed by centrifugation (7800 rpm, 15 min) and washed with

ethanol (40 mL) three times. The received catalyst was freeze dried overnight.

3.1.2. Pt on Ruthenium Titanium Oxide

Ruthenium Titanium Oxide

In a typical synthesis of Ruthenium Titanium Oxide (RTO)62,63, TiO2 powder first dispersed in

DI water (250 mL) in a sonification bath for 30 min. Afterwards, RuCl3·xH2O was added and

the mixture was stirred for another 30 min. The pH then was adjusted to 7 by adding

0.05 M KOH under continuous stirring. The resulting black powder was filtered and washed

several times with water, dried at 120 °C in air for 8 h and calcined at 450 °C in air.

3. Experimental Part

Pt Deposition on RTO

For the RTO support Pt was deposited via a wet chemical approach as reported elsewhere. 62,63

For a typical synthesis, RTO powder was added to a solution of H2PtCl6·6H2O in formic acid

and water (600 mL) . The mixture was sonicated for 30 min and then heated to 80°C for 2 h der

continuous stirring. Afterwards, the produced was filtered and washed several times with water

and dried at 60°C in air.

3.1.3. Pt on modified Vulcan

Carbon Modification

3 g Vulcan XC 72R (Cabot) were first treated in diluted 1 M HCl (300 mL) for 24 h at room

temperature (RT) to remove residual metal traces (referred to as HCl-Vulcan). Afterwards, the

carbon was intensively washed with miliQ until the filtrate was neutral and dried overnight at

90°C in air. Oxidation of the carbon was performed by treatment of the HCl-Vulcan (2 g) in

concentrated HNO3 (200 mL) at 90°C for 5 h and again washed with miliQ and dried overnight

at 90°C in air (referred to as O-Vulcan). For ammonolysis, O-Vulcan (ca. 300 mg) was placed

in an oven crucible and the tube furnace was first purged with nitrogen and afterwards heated

up (5 °C min-1) under constant NH3 flow (10 L h-1) to 400 or 800 °C. The temperature was kept

for 2 h and the sample was then naturally cooled down to RT under nitrogen atmosphere.

Samples are referred to as N-Vulcan 400°C and N-Vulcan 800°C, respectively.

Pt Deposition on modified Vulcan

To deposit the Pt nanoparticles on the different supports, a wet impregnation approach was

followed. Therefore, 150 mg of carbon support (Vulcan XC 72r, O-Vulcan, N-Vulcan 400°C

or N-Vulcan 800°C) were added to a solution of H2PtCl6·6H2O (99.6 mg) in a 1:1 mixture of

miliQ and iPrOH. The slurry was mixed using a horn sonifier (Branson, output 6 W) for 15 min

in an ice bath and afterwards freeze dried for 2 days. For reduction of the Pt precursor, the

powder was treated in hydrogen atmosphere (4 % H2 in Ar): After purging with nitrogen the

gas was changed to H2/Ar and the temperature was raised to 200 °C with a ramping of

2 °C min-1 and held for 2 h and afterwards cooled to room temperature under nitrogen

atmosphere.

3. Experimental Part

3.2. Physicochemical Characterization

The following sections will give an overview and experimental description of all

physicochemical characterization techniques used in this work, starting with essential methods

such as XRD, TEM and ICP-OES followed by more advanced spectroscopic and analytical

techniques with references to the corresponding result chapters.

Due to the variety of methods used in this work, the following Table 3.2 lists all physicochemial

characterization techniques. It gives references to the sub-sections with the experimental

descritptions and the corresponding chapters in which the results are presented and discussed.

Table 3.2 Overview of Methods for physicochemical characterization used in this work with reference

to the sections in which they are described. x indicates the the application of the method for the

corresponding result chapter.

Method

Exp. Section

Ch. 4 – Pt/ITO

Ch. 5 – Pt/N-C

Ch. 6 – Pt Ox

XRD

3.2.1

TEM

3.2.2

ICP-OES

3.2.3

HAADF STEM

3.2.4

Elemental Analysis

3.2.5

N2 Pysisorption

3.2.6

Zeta Potential

3.2.7

HT-DEMS

3.2.8

XPS

3.2.9

x (synchrotron)

x (laboratory)

In situ

SFC ICP-MS

3.4.1

HE-XRD

3.4.3

SAXS

3.4.4

XAS

3.4.5

3. Experimental Part

3.2.1. X-ray Diffraction (XRD)

XRD is a powerful technique to investigate crystalline materials on the bulk scale. It is possible

to analyze the sample with respect to its crystalline composition, crystallite size and lattice

parameter. The method is based on the elastic scattering of X-rays by atoms that are ordered in

lattice planes of a crystal structure.

Figure 3.1 Schematic illustration of the Bragg equation.

Figure 3.1 shows a schematic visualization of the Bragg equation that gives the dependence of

the scattering angle 𝜃, the incident X-ray wave 𝜆 and the distance between two lattice planes 𝑑,

while 𝑛 is an integer (equation (2)). Constructive interference is obtained if the scattered X-ray

waves are in phase (𝑛=1,2,…).

𝑛𝜆=2𝑑 𝑠𝑖𝑛𝜃

(2)

X-ray diffraction patterns were collected on a Bruker D8 Advance diffractometer (Bruker AXS)

in Bragg Brentano geometry using Cu Kα radiation (0.154 nm) and a position sensitive device

as detector. Profiles were recorded between 20 and 90 ° using a step size of 0.04 ° and 7 s time

at each step.

3.2.2. Transmission Electron Microscopy (TEM)

TEM images were obtained using a FEI Tecnai G2 20 S-TWIN equipped with a LaB6 cathode

operating with 200 kV acceleration voltage and a resolution limit of 0.24 nm. In order to prepare

a sample, a small amount of catalyst powder was dispersed in absolute ethanol using an

ultrasonic horn or a sonification bath and afterwards drop casted on a Cu grid (300 mesh with

lacey carbon film) and dried in air at 60 °C.

High resolution TEM (HRTEM) was performed using a FEI Titan 80-300 TEM electron

microscope with a Cs corrector for the objective lens (CEOS GmbH). The microscope was

operated at 300 kV.110

lattice plane

lattice plane d·sin(θ)

3. Experimental Part

3.2.3. Inductively Coupled Plasma – Optical Emission Spectroscopy

(ICP-OES)

Compositional analysis of the electrocatalysts to obtain the Pt weight loading on the supports

was performed using an ICP-OES (Varian 715-ES). 5-10 mg of catalyst powder were digested

in a 1:1:3-mixture of concentrated acids HNO3/H2SO4/HCl using a microwave protocol for

20 min at 180 °C and 18 bar. Standards with concentrations 1, 5 and 10 mg/L and Pt emission

lines at 203.646, 204.939, 212.863, 214.424, 217.468, 224.552 nm were used for quantitative

analysis.

3.2.4. High-angle Annular Dark Field (HAADF) Scanning

Transmission Electron Microscopy (STEM) and Energy-

dispersive X-ray (EDX) Spectroscopy

STEM was performed using a FEI Titan G2 80-200 electron microscope with a Cs-probe

corrector (CEOS GmbH) and a HAADF detector operated at 200 kV.111 To achieve Z-Contrast-

conditions, a probe semi-angle of 25 mrad and an inner collection semi-angle of the detector of

88 mrad were used. Compositional mappings were received by EDX spectroscopy using four

large-solid-angle symmetrical Si drift detectors. Pt L, Sn L and In L peaks were used for the

elemental mapping.

Results from STEM/EDX measurements on Pt/ITO are presented in chapter 4.

3.2.5. Elemental Analysis (EA)

EA for nitrogen, carbon and hydrogen content was performed using a Thermo FlashEA 1112

Organic Elemental Analyzer by dynamic flash combustion at 1020°C.

Bulk oxygen composition analysis by hot gas extraction was performed using a LECO TC-

300/EF-300 N/O analyzer. The carbons were thermally decomposed in a Ni/Sn/Pt-melt at ca.

3000 K in Helium atmosphere. For determination of oxygen content, the concentration of CO2

from oxidation of CO was measured using an IR cell.

Results from elemental analysis on (N-/O-)Vulcan are presented in chapter 5.

3. Experimental Part

3.2.6. Nitrogen Physisorption by BET

For nitrogen physisorption measurements, a defined amount of sample (around 20 mg) was

placed in glas tube and first degassed for at least 24 h at 140 °C. Adsorption isotherms were

recorded at a Quantachrome Autosorb-1-C at 78 K. The BET surface area was calculated using

a NLDFT equilibrium model assuming slit/cylindric pores.

Results from nitrogen physisorption measurements on (N-/O-)Vulcan are presented in

chapter 5.

3.2.7. Zeta Potential (ZP)

Measurement of the nanoparticles ZP is achieved by determining the velocity of particles

induced by an electric field. The velocity of particles between two electrodes is proportional to

the Zeta potential and can be obtained using a laser. Zeta potential was measured using a

Zetasizer Nano Z (Malvern Instruments) equipped with a He/Ne laser (633 nm, 4 mW). For

each measurement of the ZP, 5 mg of carbon were dispersed in 80 mL of miliQ and dispersed

using a sonification bath.

Results from Zeta potential measurements on (N-/O-)Vulcan are presented in chapter 5.

3.2.8. High Temperature – Differential Electrochemical Mass

Spectroscopy (HT-DEMS)

Gas diffusion electrodes (GDEs) were manufactured via a spray-coating technique to evaluate

carbon corrosion of carbon materials via HT-DEMS. For this purpose, inks consisting of the

various carbon materials (Vulcan, O-Vulcan, HCl-Vulcan, or N-Vulcan 400°C), a PTFE

dispersion and water were used. Firstly, the carbon material was dispersed with water by stirring

the mixture for 24 h. Afterwards, the obtained PTFE dispersion was added to the mixture until

5 wt% PTFE (referring to carbon material) was obtained. After sonification of the ink for

45 min, the electrode layer was applied onto gas diffusion layer (GDL) substrate (H23,

Freudenberg FCCT) via spray-coating technique with argon as spraying gas. The electrodes

were dried for 1 h at 130 °C. Subsequently, the GDEs were weighed and the carbon loading

was determined to be 2 mgC cm-2 by subtracting the mass of the empty GDLs. The prepared

GDEs were round and had a diameter of 2 cm (geometrical area 3.14 cm2).

In the DEMS setup an electrochemical half-cell is combined with a mass spectrometer (MS).112

Basically, the HT-DEMS setup allows electrochemical investigation of gas diffusion electrodes

(GDEs) under gas-phase conditions at elevated temperatures while monitoring products from

3. Experimental Part

the electrochemical reactions. The electrochemical cell was mantled by a heating jacket and the

temperature was controlled at 140 °C using a therocouple (tolerance of ± 1°C). With the

electrochemical cell it is possible to measure under gas flow. During the measurements, N2 gas

was supplied with a flow rate of 50 mL min-1. Connection between the cell and the MS was

achieved using a heated capillary allowing volatile products formed at the working electrode to

be transported to the MS. In this way, the products can be analyzed by measuring their specific

ion currents after being ionized.113 During electrochemical tests, the ion current of the MS signal

m/z=44 (CO2) was monitored. Preconditioning of GDE samples was achieved by cyclic

voltammetry (15 cycles) in N2 atmosphere at a scan rate of 10 mV s-1 from 0.06-1.05 V. To

evaluate the carbon corrosion, each sample was scanned with a pulse voltammetry technique.

The potential ranges used in the study are: 0.06 to 0.1 V, 0.10 to 0.20 V and 0.26 to 1.06 V with

300 s pulse every 0.01, 0.02 and 0.1 V, respectively. The time between pulses was 30 s at a

potential of 0.06 V. For evaluation, a mean value of the faradaic and ion current was formed

from the data obtained in the last 60 s of each step.

Results from HT-DEMS measurements on (N-/O-/HCl-)Vulcan are presented in chapter 5.

3.2.9. X-ray Photoelectron Spectroscopy (XPS)

Synchrotron-based X-ray photoelectron spectra were collected at the ISISS beamline of BESSY

II facility operated by Helmholtz Zentrum Berlin. For sample preparation of Pt/ITO (chapter

4), the catalyst ink was drop casted onto a 5 mm diameter glassy carbon disc and dried at 60 °C.

The catalyst was treated by the LP-AST electrochemical protocol. XPS measurements were

conducted at room temperature in ultra-high vacuum (UHV). Pt 4f depth profiling was accessed

by using kinetic energy of the photoelectrons of 210, 550 and 1200 eV.

Results from synchrotron-based XPS measurements on Pt/ITO are presented in chapter 4.

Laboratory-based XPS measurements were performed in an ultra-high vacuum (UHV) setup at

room temperature. Non-monochromatized Al Kα (1486.6 eV) excitation and a hemispherical

analyzer (Phoibos 150, SPECS) were used. Catalyst and support powders (chapter 5) were

dispersed in a 1:1 mixture of iPrOH and miliQ, horn sonified for 15 min and drop casted on a

clean titanium coated silicon wafer and dried afterwards. The binding energy (BE) scale was

calibrated by the standard Au4f7/2 and Cu2p3/2 procedure and the spectra were analyzed using

CasaXPS software. All spectra have been charge-corrected with respect to the main peak in the

C1s spectrum for adventitious carbon that was assigned to have a binding energy of 284.8 eV.114

To calculate the elemental composition of the modified carbons, theoretical cross sections from

3. Experimental Part

Yeh and Lindau115 were used. The N 1s region was fitted to identify the types of N species

present in the samples. All samples could be fitted with six peaks corresponding to pyridine,

pyrrolic, quaternary N, oxidized N, NO2- and NO3- groups in order of increasing binding energy.

Results from laboratory-based XPS measurements on (N-/O-)Vulcan and Pt/N-Vulcan are

presented in chapter 5.

3. Experimental Part

3.3. Electrochemical Characterization

3.3.1. Preparation

Electrochemical Setup: Rotating Disc Electrode

A commercial three electrode glas cell setup was applied for the electrochemical measurements

performed in this work. The cell consists of a Pt mesh (furled Pt 5x5 cm2) as counter electrode

and a mercury/mercury sulfate (MMS) reference electrode (Hg/Hg2SO4, Ametek, potential

+0.722 V vs. reversible hydrogen electrode (RHE), frequently calibrated against Pt/H2 system).

A Luggin Capillary connected the reference electrode with the main cell compartment. Diluted

0.1 M HClO4 was used as electrolyte for all electrochemical measurements. A Potentiostat (SP-

150 or SP-200, BioLogic Instruments) was used to control the potential of the working electrode

and a commercial AFMSRCE rotator (Pine Research) to control rotation.

Electrode and Ink preparation

Depending on the different supports used in this work, preparations of the ink for the

electrochemical measurements vary slightly from each other and are presented in Table 3.3.

Table 3.3 Details on Ink preparation and composition depending on the different supports used in this

work.

Metal Oxide

Modified Carbon

Chapter

4, 6

Mass of catalyst powder / mg

miliQ /mL

1.99

3.98

Isopropanol / mL

0.50

1.00

Nafion / μL

Time horn sonification / min

Pt loading / μgPt cm-2

25.0

12.5

Catalysts inks were prepared according to the table above by mixing the defined mass of

catalyst with the volumes of miliQ, isopropanol and Nafion, followed by horn sonification, see

Table 3.3. Glassy carbon (GC) electrodes (Pine Research) with diameter of 5 mm were polished

in two steps using alumina polishing solution (diameter 1.0 and 0.05 μm, Buehler) and cleaned

thoroughly for several times in miliQ, acetone and ethanol. 10 μL of catalyst ink were deposited

on the freshly polished, cleaned and dried GC electrode and dried in air at 60 °C for 7 min.

3. Experimental Part

3.3.2. Activity and Stability Measurements

Electrochemical activity and stability measurements were conducted in order as presented in

the following schematic illustration (Figure 3.2). In short, conditioning by cyclic voltammetry

(CV) was performed to clean the Pt surface and obtain a stable CV from which the Hupd-ECSA

was achieved followed by potentiostatic electrochemical impedance spectroscopy (PEIS) for

determination of the resistance. Afterwards, the ORR activity was determined by recording an

linear sweep voltammogram (LSV) and then the accelerated stress test (AST) was performed

followed by a final activity measurement. A schemtaic illustration of such a measurement

sequence is presented in Figure 3.2 and detailed information on each step are listed separately

below. All shown electrode potentials are referred to the RHE scale.

Figure 3.2 Schematic illustration of the sequence of an electrochemical activity and stability

measurement, including conditioning by cyclic voltammetry (CV), impedance measurement by

potentiostatic electrochemical impedance spectroscopy (PEIS), activity determination by linear sweep

voltammetry (LSV), accelerated stress test (AST) in two different potential windows (low potential (LP)

and high potential (HP)) and activity determination after the AST. CO stripping, as applied in chapter 4,

was performed directly before and after the AST.

3. Experimental Part

Conditioning by Cyclic Voltammetry (CV)

CVs were recorded between 0.05 and 1 V with a scan rate of 100 mV s-1 in nitrogen-saturated

electrolyte. For Pt on oxide supports (Chapter 3 and 5) 100 cycles were conducted, whereas for

carbon-based supports (Chapter 4) 50 cycles were applied. Determination of the

electrochemical active surface area (ECSA) based on Hydrogen underpotential deposition

(Hupd) was achieved by integration of both oxidative and reductive currents in the Hupd region

between 0.05-0.40 V while subtracting capacitive currents in this potential window. Assuming

a theoretical value of 𝑄𝐻

𝑡ℎ𝑒𝑜 =210 μC cm−2 for a one electron transfer process for the

desorption of hydrogen, the Hupd-ECSA can be calculated as follows:

𝐸𝐶𝑆𝐴=𝐴𝑔𝑒𝑜

𝑚𝑃𝑡 = 𝑄𝐻

𝑚𝑃𝑡 ∙𝑄𝐻

𝑡ℎ𝑒𝑜 𝑤𝑖𝑡ℎ 𝑄𝐻=1

𝑣∫ 𝐼 𝑑𝐸

𝐸2

𝐸1

(3)

In equation (3), 𝐴𝑔𝑒𝑜 is the catalyst’s geometric surface area and can be obtained by normalizing

the charge 𝑄𝐻 to the theoretical value for the one electron transfer 𝑄𝐻

𝑡ℎ𝑒𝑜 resulting in the Hupd-

ECSA. 𝑄𝐻 describes the charge for Hydrogen adsorption and desorption, that is obtained by

integrating the capacitive-corrected currents (𝐼) between potentials 𝐸1=0.05𝑉 and 𝐸2=0.4𝑉

and dividing this by the scan rate (𝑣). The geometrical surface area is then normalized to the Pt

mass on the elctrode (𝑚𝑃𝑡) resulting in the Hupd-ECSA.

Impedance by Potentiostatic Electrochemical Impedance Spectroscopy (PEIS)

For correction of the iR-drop, the potentials were iR-corrected by the high frequency resistance

RHF determined by potentiostatic electrochemical impedance spectroscopy (PEIS) performed at

0.5 V after the conditioning step in nitrogen-saturated electrolyte.

Activity by Linear Sweep Voltammetry (LSV)

LSVs were recorded in oxygen-saturated electrolyte between 0.05 and 1 V in anodic direction

with a scan rate of 5 mV s-1 and a rotation speed of 1600 rpm. Determination of kinetic currents

(𝑗𝑘) was done by applying the following equation (4):

𝑗= 1

𝑗𝑘+ 1

𝑗𝑑

(4)

Here, 𝑗 describes the current at 0.9 V whereas jd was determined by averaging the diffusion

limited current between 0.1-0.6 V. Mass based activities (𝑗𝑚 ) are obtained by normalization to

the Pt mass on the electrode (𝑚𝑃𝑡) according to equation (5):

𝑗𝑚 = 𝑗𝑘

𝑚𝑃𝑡

(5)

3. Experimental Part

Specific activities (𝑗𝑠𝑝𝑒𝑐 ) can be calculated by normalizing the kinetic current (𝑗𝑘) to the

geometrical surface area of the catalyst (𝐴𝑔𝑒𝑜 ) obtained from the CV in the following equation

(6).

𝑗𝑠𝑝𝑒𝑐 = 𝑗𝑘

𝐴𝑔𝑒𝑜

(6)

Accelerated Stress Test (AST)

In order to simulate fuel cell operating conditions to the electrocatalyst, ASTs were performed

in two different potential regimes in nitrogen-saturated electrolyte. In the low potential (LP)

regime, fuel cell lifetime conditions are simulated, which in RDE setup is achieved by cycling

between 0.6-0.95 V with a scan rate of 100 mV s-1. Start-up/shut-down conditions are simulated

by cycling between 1-1.5 V with a scan rate of 500 mV s-1 and are referred to as high potential

(HP) AST. Table 3.4 shows an overview of the different ASTs performed for the various

catalysts analyzed in this work.

Table 3.4 Overview of ASTs performed in this with work for Pt/ITO and Pt/(un)modified carbon with

respect to the applied cycle numbers.

Pt/ITO

Pt/(un)modified carbon

Chapter

Cycle number in LP-AST

5k, 10k and 30k

Cycle number in HP-AST

In order to determine the Hupd-ECSA before and after the AST protocols three CVs between

0.05-1 V with a scan rate of 100 mV s-1 were performed and the Hupd-ECSA was determined as

described above.

Parameters for both stability tests were adapted from the US Department of Energy (DoE) fuel

cell targets from the year 2016.116

CO Stripping

CO stripping was applied to the metal oxide supported catalysts (chapter 4 and 6) in order to

determine the CO-ECSA in addition to the Hupd-ECSA. In a CO stripping experiment, the

electrolyte is first saturated with CO to ensure complete CO coverage of the Pt surface. In a

next step, excess CO is removed from the electrolyte by saturating it with nitrogen. The CO is

stripped from the Pt surface by cycling between 0.05-1 V for 5 cycles with a scan rate of

50 mV s-1. CO-ECSA determination was achieved by subtracting the first two consecutive

3. Experimental Part

cycles of the stripping protocol and integration of the charge between the two cycles (𝑄𝐶𝑂).117

The measured 𝑄𝐶𝑂 values were normalized with respect to the theoretical value for a two

electron transfer 𝑄𝐶𝑂

𝑡ℎ𝑒𝑜 =420 μC cm−2 for CO as probe molecule. The CO-ECSA was

obtained as described for Hupd-ECSA above in Equation (3).

CVs for Pt Electrooxidation

For Pt electrooxiadtion as shown in chapter 6, each three CVs starting from 0.05-0.5 V to an

increasing upper potential limit of 1.4 V in steps of 0.1 V were measured under nitrogen-

saturated atmosphere with a scan rate of 100 mV s-1.

3. Experimental Part

3.4. In situ Characterization

In this work, various in situ methods were applied to analyze both catalyst and support

components with respect to their changes under accelerated stress test or Pt oxidation

conditions. These methods will be described in the following sections, beginning with the in

situ dissolution measurements followed by the in situ X-ray based setup and methods.

3.4.1. Scanning Flow Cell Inductively Coupled Plasma Mass

Spectroscopy (SFC ICP-MS)

For in situ SFC ICP-MS measurements, inks of Pt/RTO, Pt/ITO and Pt/C were prepared with a

solution of 25 µL 5 % Nafion and 4.975 mL ultrapure water (18.2 MΩcm, PureLab Plus

System, Elga). The ink was sonicated and subsequently 0.3 µL drops were deposited onto the

glassy carbon sheet (SIGRADUR® G, HTW) which was polished with 1 µm Al2O3 suspension

and cleaned with ultrapure water before. The spots of catalyst ink had a size of ca. 0.01 cm2 and

a final loading of approximately 1.6 µgPt cm-2 for Pt/ITO and Pt/C and 8.33 µgPt cm-2 for

Pt/RTO was obtained. In situ dissolution measurements were conducted using the previously

described SFC ICP-MS setup (NexION 300X, Perkin Elmer).118,119 Contact area between

working electrode and the cell was 035 cm2. A freshly calibrated saturated Ag/AgCl electrode

and graphite rod were used as reference and counter electrode, respectively. All shown

potentials are reported against RHE. Argon-saturated 0.1 M HClO4 was used as electrolyte. The

flow rate of the electrolyte in the SFC was approximately 170 µL min-1. 10 µg L-1 of Re (187Re

was monitored) was used as internal standard for Pt and 20 µg L-1 of 103Rh were used as internal

standards for In and Sn.

For the Pt, In and Sn dissolution as presented in Chapter 4, dissolution in µg µgPt-1 was

determined via integration of the ICP-MS signal data by the flow rate (normalized to Pt mass

loading on the working electrode). Ranges for the integration were taken from a constant signal

area before and after the analyzed peak.

Results from SFC ICP-MS measurements on Pt/C and Pt/ITO (Pt, In and Sn dissolution) are

presented in chapter 4 and on Pt/C, Pt/ITO and Pt/RTO (Pt dissolution) in chapter 6.

3.4.2. In situ Electrochemical Cell and Setup

In order to analyze morphological and structural changes of the electrocatalyst, in situ X-ray

measurements were performed. Such experiments were conducted at a synchrotron facility

where brilliant and high-flux radiation is produced by an electron accelerator as radiation

source. The fact that a wide spectrum from soft to hard X-rays can be generated with the

3. Experimental Part

opportunity to change the energy relatively easy at such a facility, makes it advantageous

compared to lab-based techniques. Figure 3.3 shows a schematic illustration of the setup as

applied for the in situ experiments performed at a synchrotron in this work.

Figure 3.3 Schematic illustration of the setup and electrochemical in situ cell as used at a synchrotron

facility, showing the incident and scattered X-rays, the 2D detector and the in situ transmission cell.

Distance between sample and detector is defined as working distance.

In such a setup, the sample is measured in transmission geometry, in which the incident X-ray

beam passes through the sample and is recorded together with the scattered X-rays at a 2D

detector. In the in situ cell, the electrochemical protocols are applied. Therefore, the catalyst

ink (5x10 µL) is drop-casted on a carbon paper sheet, dried and inserted into the electrolyte

compartment. A Pt wire acts as counter electrode and a 3M Ag/AgCl reference electrode

(freshly calibrated against Pt/H2 electrode) were used. This setup enables electrochemical

measurements in a three electrode configuration with a sufficient thickness of electrolyte at the

sample and accessibility of the catalyst while minimizing the X-ray attenuation/sacttering by

water molecules. Additionally, by fast changes of the working distance between sample and

detector, the setup at ID31 allows almost simultaneous measurements of high energy X-ray

diffraction (HE-XRD) and (anomalous) small angle X-ray scattering ((A)SAXS).

reference

electrode

working

electrode

counter

electrode

catalyst ink

2D detector

in situ cell

3. Experimental Part

3.4.3. High Energy X-ray Diffraction (HE-XRD)

Theoretical Background

X-ray diffraction at sufficiently higher energies than those used in lab-based assemblies is

referred to as high energy XRD (HE-XRD). Such measurementsb give rise to a much larger q,

and therefore 2θ, range of the diffraction pattern. It can only be performed at synchrotron

facilities due to the fact that the energy of the photons can be tuned with respect to the desired

properties. In this work, HE-XRD is applied to investigate the crystalline components of the

electrocatalyst under simulated fuel cell operating conditions. In this way, crystallite parameters

such as crystallite size, lattice constant and scale factor can be followed.

Analysis of the HE-XRD patterns includes not only a proper phase assignment, but also

Rietveld refinement. This method developed by Hugo M. Rietveld in the 1960s120,121 allows to

refine several different parameters of the analyzed material, such as crystallite size, weight

fraction (of different phases), lattice parameters, atomic positions and occupancies as well as

micro strain and scale factor (scattering intensity). Rietveld refinement applies a least square

approach and fits an entire profile of a diffraction pattern to a calculated profile. A major

advantage of this method is that patterns with several overlapping Bragg peaks can be

analyzed.122 Requirements for successful fitting procedures include high quality experimental

data and a large q-range as well as structure model(s) that make sense on a physical and

chemical level.

The goal is to minimize the residual function 𝑀 at 𝑖 steps of the pattern over all data points 𝑁

which can be expressed as:

𝑀= ∑𝑤𝑖

𝑁

𝑖=1 [𝐼𝑖(𝑜𝑏𝑠)− 𝐼𝑖(𝑐𝑎𝑙𝑐) ]2

(7)

In equation (7), 𝐼𝑖(𝑜𝑏𝑠) describes the intensity measured at point 𝑖 and 𝐼𝑖(𝑐𝑎𝑙𝑐) the

corresponding calculated intensity. 𝑤𝑖 is the weight of the diffraction pattern’s intensity and is

given by equation (8):

𝑤𝑖= [𝐼𝑖(𝑜𝑏𝑠) ]−1

(8)

The calculated intensity 𝐼𝑖(𝑐𝑎𝑙𝑐) is defined by the following equation (9):

𝐼𝑖(𝑐𝑎𝑙𝑐)=𝑠 ∑ 𝑚𝑘 𝐿𝑘 |𝐹𝑘|2 𝐺𝑖𝑘

𝑘2

𝑘=𝑘1

(9)

3. Experimental Part

This expression describes the summation over all reflections 𝑘1 to 𝑘2 for one phase containing

the scale factor 𝑠 that is depending on beam intensity, volume fraction and cell volume. 𝑚𝑘 is

the multiplicity factor for the 𝑘th reflection and 𝐿𝑘 the Lorentz-polarization factor. 𝐹𝑘

represents the structure factor and 𝐺𝑖𝑘 the so-called peak-shape function. A proper choice for

description of this function is often difficult and depends on the nature of the experimental

technique (X-ray or neutron diffraction).122 In case of X-rays, a mixed description of Lorentian

and Gaussian components, expressed in a Pseudo-Voigt function 𝑉, is commonly applied

(equation (10))123-125:

𝑉=𝜂𝐿+(1−𝜂)𝐺 𝑤𝑖𝑡ℎ (0≤𝜂≤1)

(10)

With the Gaussian component 𝐺 being the following equation (11):

𝐺= 𝐼0 exp(−𝑙𝑛2(2𝜃−2𝜃0

𝜔)2)

(11)

The Lorentzian contribution 𝐿 can be described by equation (12):

𝐿= 𝐼0 (1+(2𝜃−2𝜃0

𝜔)2)−𝑛 𝑤𝑖𝑡ℎ 𝑛=1;1.5;2

(12)

In these two components, 𝐼0 describes the peak intensity, 2𝜃0 the peak position in measures of

2𝜃 and 𝜔 includes the full width at half maximum (FWHM) in the following correlation

(equation (13)):

𝜔= 𝐹𝑊𝐻𝑀

(13)

A correct background estimation is also important in Rietveld refinement. The background can

either be determined in regions void of any Bragg reflections122 or, in another approach, the

background of the pattern can be included in the refinement model as a polynomial function.126

Experimental Description

In the presented work, the HE-XRD measurements were performed at beamline ID31 at the

European synchrotron radiation facility (ESRF) in Grenoble, France. Here, relatively high

energies (68-78 keV) were used resulting in transmission values for the sample including the

substrate and the electrolyte layer that are approaching one and therefore enabling high quality

data. A large area detector (Pilatus3X CdTe 2M, Dectris) was used to record the diffraction

images. The working distance between the sample and the detector was calibrated using a CeO2

standard (NIST SRM 674b). The diffraction patterns were corrected by the background.

Rietveld refinement was performed using the TOPAS® software package (Bruker).

3. Experimental Part

Results from HE-XRD and Rietfeld refinement of Pt and ITO are presented in chapter 4 and of

Pt on different carbon supports in chapter 5.

HE-XRD pattern of Pt/ITO, Pt/RTO and Pt/C (chapter 6) were integrated and the intensity plots

fitted using TOPAS®. Therein, the unit cell parameter was fitted and the occupancy of the

tetrahedral sites with O atoms refined. The size- and microstrain-induced broadening of the Pt

diffraction peaks were fitted. The size-induced broadening was fitted using the anisotropy

approach developed by Ectors et al.127,128 In the case of the HE-XRD pattern of Pt nanoparticles

on ITO, spherical nanoparticles were considered with all three main axes of the ellipsoids

identical (Rx=Ry=Rz). In the case of the Pt nanoparticles on RTO and on carbon, the size-

induced broadening was fitted with three independently-fitted principal axes of the ellipsoids

(Rx≠Ry≠Rz). The polydispersity of the nanoparticles representing the particle size distribution

was taken into account, too. The crystal properties of each catalyst were individually fitted in

each investigated state but a joint fit approach was applied to determine zero offset and sample

displacement.

Results from fitting of the HE-XRD patterns of Pt on ITO, RTO and carbon are presented in

chapter 6.

3.4.4. Small Angle X-ray Scattering (SAXS)

Theoretical Background

Small angle X-ray scattering is an useful technique to investigate materials on the nanoscale.129-

131 In this work, SAXS is applied to analyze the morphological stability of Pt with respect to

the size of the Pt nanoparticles during electrochemical degradation protocols in a simulated fuel

cell environment.

Figure 3.4 Schematic illustration of the elastic scattering as the basis of SAXS.

The information obtained from SAXS analysis are based on elastic scattering of X-rays at the

electron clouds of atoms and interference of the resulting scattered waves.132,133 The scattered

2θ

sample

3. Experimental Part

intensity is a function of the scattering vector 𝑞, that is, in contrast to the scattering angle 2θ,

independent on the used wavelength (Figure 3.4). The scattering vector can be described by the

incoming wave vector 𝑘0

󰇍



and the scattered one 𝑘1

󰇍



as follows in equation (14):

𝑞= 𝑘1

󰇍



− 𝑘0

󰇍



𝑤𝑖𝑡ℎ |𝑘

󰇍



|= 2𝜋

𝜆

(14)

The magnitude of the scattering vector can therefore be expressed according to equation (15):

|𝑞|=𝑞 = 4𝜋

𝜆sin (𝜃)

(15)

When incoming X-ray waves are scattered at the electron cloud of an atom, an outgoing X-ray

wave with the amplitude 𝐴1(𝑞) can be written as (equation (16)):

𝐴1(𝑞)= 𝐴0𝑏𝑒∫ 𝑛(𝑟)𝑒−𝑖𝑞𝑟𝑑𝑟

𝑉

(16)

In here, 𝑛(𝑟) describes the electron density distribution within one given atom and 𝑟 the

position of the electron from the atomic center. 𝐴0 represents the amplitude of the incoming

wave and 𝑏𝑒 the scattering length of a single electron. To obtain the amplitude of the scattered

wave, integration is performed over the volume of the atom.

When the amplitude is normalized to units of 𝐴0𝑏𝑒, the atomic scattering factor 𝑓(𝑞) is obtained

as follows in equation (17):

𝑓(𝑞)= ∫ 𝑛(𝑟)𝑒−𝑖𝑞𝑟𝑑𝑟

𝑉

(17)

Under the assumption, that a sample contains an ensemble of various atomic species 𝑖 with the

density distribution of the atomic centers 𝑛𝑖, the scattering length density distribution 𝜌(𝑟) can

be expressed in equation (18) as the sum of the various atomic species and the density

distribution of the centers:

𝜌(𝑟)= ∑𝑓𝑖𝑛𝑖(𝑟)

(18)

The scattered intensity 𝐼(𝑞), that describes the q-dependent pattern of the scattered X-rays, can

be written as (equation (19)):

𝐼(𝑞)= |𝐴1(𝑞)|2= |∫ 𝜌(𝑟)𝑒−𝑖𝑞𝑟𝑑𝑟

𝑉|2

(19)

When incoming X-rays are scattered at diluted and spherical particles (e.g. supported Pt

nanoparticles), it is assumed that the position of the individual particles are independent from

3. Experimental Part

each other. Therefore, the scattered intensity 𝐼𝑠𝑝ℎ𝑒𝑟𝑒 (𝑞) of such a system with particles of the

radius 𝑅 and the scattering length density distribution 𝜌 can be expressed as follows in

equation (20):

𝐼𝑠𝑝ℎ𝑒𝑟𝑒 (𝑞)= 𝜌2(4

3𝜋𝑅3)2[3(𝑠𝑖𝑛𝑞𝑅−𝑞𝑅𝑐𝑜𝑠𝑞𝑅)

(𝑞𝑅)3]2

(20)

The last term in this equation represents the squared particle form factor 𝑃(𝑞,𝑅) of a spherical

particle. In case all particle radii are of the same dimensions, hence the particle system is

monodisperse, periodic variations in 𝐼𝑠𝑝ℎ𝑒𝑟𝑒 (𝑞) are observed that smear out with increasing

polydispersity.

Anomalous SAXS is a method to measure element-specific scattering curves and can therefore

be applied to analyze bi- or trimetallic particles systems or, as used in this work, to isolate the

Pt-scattering from scattering of the support (particle) components. This variation to the

conventional SAXS method relies on the energy dependence of the atomic scattering factor,

exemplified in the following equation (21) for the atomic scattering factor of Pt132,134,135:

𝑓𝑃𝑡(𝐸)=𝑓𝑃𝑡

0+ 𝑓𝑃𝑡

′+ 𝑖𝑓𝑃𝑡

′′

(21)

In this equation, 𝑓𝑃𝑡

0 represents the energy independent scattering factor, 𝑓𝑃𝑡

′ the anomalous,

energy dependent scattering factor and 𝑖𝑓𝑃𝑡

′′ the imaginary part of the form factor. Therefore,

the scattering curves are recorded at an energy before the Pt absorption edge (𝐸1) and close to

the edge (𝐸2), while the scattering factor of the support components stays constant. The

difference of the two scattering intensities (𝐼(𝐸1)−𝐼(𝐸2)) can then lead to the isolated Pt-

scattering pattern.

Experimental Description

For the (A)SAXS experiments at ID31 at the ESRF a PerkinElmer DEXELA 2923 detector was

used and the working distance was calibrated using an Ag Behenate reference sample. In case

anomalous SAXS was performed, scattering patterns were recorded before (77 keV) and close

to the Pt K-edge (77.8 keV). Alignment for the energy of the beamline was achieved using the

Pt K-edge absorption spectrum. The scattering curves were corrected by the background and

normalized with respect to the transmission of the sample. Fitting of the (A)SAXS curves was

performed using the Sasfit software package.

Results from ASAXS measurements on Pt/ITO are presented in chapter 4 and from SAXS

measurements on Pt/(N-/O-)Vulcan in chapter 5.

3. Experimental Part

3.4.5. X-ray Absorption Spectroscopy (XAS)

In situ XAS experiments were carried out at the mySpot beamline of the BESSY II synchrotron

radiation facility operated by the Helmholtz-Zentrum Berlin (HZB). A Si(111) double crystal

monochromator was used and measurements were performed at the Pt L-edge. Alignment of

the X-ray energy for each spectrum was achieved by shifting the first derivate of the

simultaneously-measured X-ray absorption spectrum of the Pt reference (Pt foil) to 11654 eV.

Here, instead of 2D detector as used for the HE-XRD and SAXS experiments, a scintillation

counter was used as detector to monitor the fluorescence yield. The detector was covered with

Al foil in order to reduce background radiation.

For the in situ electrochemical measurements the transmission cell was used as shown in

Figure 3.3, but tilted by 45 ° with respect to the incoming beam while the fluorescence detector

was placed perpendicular to the incoming beam in vicinity of the sample.

Results from XAS measurements on Pt/C, Pt/ITO and Pt/RTO are presented in chapter 6.

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

4. Unravelling Degradation Pathways of Oxide-Supported Pt Fuel

Cell Nanocatalysts under In Situ Operating Conditions

Knowledge of degradation pathways of catalyst/support ensembles aids the development of

rational strategies to improve their stability. Here, this is exemplifed using indium tin oxide

(ITO)-supported Platinum nanoparticles as electrocatalysts at fuel cell (FC) cathodes under

degradation protocols to mimic operating conditions in two potential regimes. The evolution of

crystal structure, composition, crystallite and particle size is tracked by in situ X-ray techniques

(small and wide angle scattering), metal dissolution by in situ scanning ﬂow cell coupled with

mass spectrometry (SFC ICP-MS) and Pt surface morphology by advanced electron

microscopy. In a regular FC operation regime, Pt poisoning rather than Pt particle growth,

agglomeration, dissolution or detachment was found to be the likely origin of the observed

degradation and ORR activity losses. In the start-up regime degradation is actually suppressed

and only minor losses in catalytic activity are observed. The presented data thus highlight the

excellent nanoparticle stabilization and corrosion resistance of the ITO support, yet point to a

degradation pathway involving Pt surface modifcations by deposition of sub-monolayers of

support metal ions. The identifed degradation pathway of the Pt/oxide catalyst/support couple

contributes to our understanding of cathode electrocatalysts for polymer electrolyte fuel cells

(PEFC).

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

Chapter 4 and section Appendix A1 were reproduced from Ref 52 (Adv. Energy Mat., 2018, 8 (4), 1701663) with

Henrike Schmies, Arno Bergmann, Jakub Drnec, Guanxiong Wang, Detre Teschner, Stefanie

Kühl, Daniel J.S. Sandbeck, Serhiy Cherevko, Martin Gocyla, Meital Shviro, Marc Heggen,

Vijay Ramani, Rafal E. Dunin-Borkowski, Karl J.J. Mayrhofer, Peter Strasser, “Unravelling

Degradation Pathways of Oxide-Supported Pt Fuel Cell Nanocatalysts under In Situ Operating

Conditions”, Adv. Energy Mater. 2018, 8 (4), 1701663; doi: 10.1002/aenm.201701663

H.S. performed the experiments and analyzed the data, A.B. helped designing the experiments and analyzed the

synchrotron data; J.D. and A.B. helped performing and the synchrotron experiments; G.W. and V.R. synthesized

the support; D.T. performed XPS measurements and analysis; S.K. recorded TEM images; D.J.S.S., S.C. and

K.J.J.M. produced and analyzed the SFC ICP-MS data; M.G., M.S., M.H. and R.E.D.-B. performed and analyzed

the STEM EDX mappings; H.S., A.B. and P.S. wrote the manuscript; all authors contributed to the discussion.

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

4.1. Introduction

In this study, we report new molecular insight in the degradation mechanisms of Pt/ITO

electrocatalysts for oxygen reduction at fuel cell cathodes. What sets this study apart from

previous ones is the use of a number of different in situ electrochemical X-ray and scanning

ﬂow cell-inductively coupled plasma (ICP) techniques to track the change in the molecular

structure and composition of the catalysts during ORR catalysis. In situ high energy X-ray

diffraction (HE-XRD) and anomalous small angle X-ray scattering (ASAXS) were applied to

follow the structure and morphology of the Pt and ITO particles. In addition, scanning ﬂow cell

tests coupled to inductively coupled plasma mass spectroscopy (SFC ICP-MS), X-ray

photoelectron spectroscopy (XPS), high angle annular dark field scanning transmission electron

microscopy (HAADF-STEM) combined with energy-dispersive X-ray (EDX) spectroscopy

yielded a deeper understanding of the evolution of morphology and surface composition after

the ASTs. Together, our experimental study unravels critical parameters determining structural,

morphological, compositional, and, thus, electrochemical stability of Pt on ITO.

4.2. Physicochemical Characterization

Platinum nanoparticles supported on indium tin oxide have been synthesized using a

solvothermal route using oleylamine and oleic acid as surfactants and dibenzylether as the

solvent. The support material was introduced in the reaction mixture from the beginning to

enable Pt nanoparticle growth directly on the support. Slow heatin g rates of 1 °C min-1 were

applied to obtain monodisperse Pt nanoparticles. Figure 4.1 shows the XRD pattern of Pt on

ITO as well as the pure ITO.

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

Figure 4.1 X-ray diffraction patterns of the bare ITO support material (black, bottom) and Pt/ITO

catalyst (dark red, top) obtained using Cu Kα radiation. Solid lines denote pure In2O3 reference powder

diffraction pattern (PDF#00-006-0416). Vertical dashed lines denote reference patterns of fcc Pt

(PDF#00-004-0802).

All reﬂexes in the ITO pattern can be clearly assigned to an In2O3 phase (PDF#00-006-0416).

No additional Sn oxide phase was present indicating complete incorporation of Sn into the In2O3

lattice thus, forming In2–xSnxO3 crystallites. The Pt/ITO pattern shows broad Pt reﬂections for

pure face-centered cubic (fcc) Pt. To analyze Pt particle size and Pt particle distribution on the

support, transmission electron microscopy (TEM) was applied.

Figure 4.2 Morphology of Pt/ITO in the initial state: (a) showing overview HAADF-STEM image, (b)

TEM and (c) HR-TEM images.

20 30 40 50 60 70 80 90

2 / °

In2O3 PDF#00-006-0416

Intensity / a.u.

Platinum

(111)

(200)

(220)

(311)

(222)

Pt/ITO

ITO

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

Figure 4.2a,b and Figure 4.3a,b shows quite homogeneously distributed Pt nanoparticles on the

ITO support with an average size of 5.3 ± 0.7 nm (Figure 4.3c). Furthermore, the support

consists of mostly well faceted nanocrystals with a size >10 nm. These crystals are randomly

oriented and clearly distinguishable from the small Pt nanoparticles.

Figure 4.3 (a,b) TEM images of the as prepared Pt nanoparticles on the ITO support, after

electrochemical cycling for 5k times in d,e) lower potential region and in (g,h) the higher potential

region, as well as the corresponding histograms showing (c,f,i) the particle size distribution. Histograms

were obtained by measuring the diameter of at least 200 particles with errors obtained from standard

deviation of mean particle diameter.

The composition of the Pt/ITO as determined by inductively coupled plasma optical emission

spectroscopy (ICP-OES) revealed a weight loading of 22.0 wt% Pt on ITO for the initial state

(Figure 4.3a,b). This sample was used for low potential-AST (LP-AST) experiments. For all

high potential (HP)-AST experiments the sample had a weight loading of 29.9 wt% Pt on ITO.

Figure A1 (Appendix A1) shows TEM images and the corresponding histogram of the Pt

4 5 6 7 8 9 10

6.4±0.9 nm

counts

particle size / nm

4 5 6 7 8 9 10

80 5.3±0.7 nm

counts

particle size / nm

after LP-AST initial

a b c

def

4 5 6 7 8 9 10

60 5.2±0.6 nm

counts

particle size / nm

after HP-AST

ghi

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

particle size in the initial state. It can be clearly seen that Pt particle size and distribution as well

as the shape and size of the support crystallites was almost identical compared to the sample

with lower loading that was used for LP-AST.

4.3. Electrochemical Characterization

Figure 4.4a depicts the cyclic voltammograms (CVs) of the Pt/ITO catalyst in the initial state

and after 5k CV in the lower potential region from 0.6–0.95 V versus reversible hydrogen

electrode (RHE) (denoted as LP-AST). The ASTs were preceded and followed by CVs in the

activation regime (0.05–1.00 V).

Figure 4.4 Cyclic voltammograms of Pt/ITO catalyst before and after potential cycling in (a) lower and

scan rate of 100 mV s-1. b,d) LSVs of the particular states with the bar plots in the inlets representing

the mass activity (jm) evaluated at 0.9 V. LSVs were recorded in oxygen-saturated electrolyte from 0.05–

1 V with a scan rate of 5 mV s-1 and 1600 rpm. All electrode potentials have been corrected for iR drop.

It can be clearly seen that the CVs changed after the AST as the current in the Pt-O/-OH region

(0.6-0.8 V) drastically decreased. This suggests a lower number of oxidizable Pt sites on the

0.0 0.2 0.4 0.6 0.8 1.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

j / mA cm-2

E / V vs RHE

initial

after HP-AST

0.0 0.2 0.4 0.6 0.8 1.0

-1.0

-0.5

0.0

0.5

j / mA cm-2

E / V vs RHE

initial

after LP-AST

0.0 0.2 0.4 0.6 0.8 1.0

-6

-5

-4

-3

-2

-1

0.0

0.1

jm / A mg-1

j / mA cm-2

E / V vs RHE

initial

after LP-AST

- 54 %

jm-54%

jm-15%

Hupd

ECSA -8%

CO ECSA

-29%

Hupd ECSA

+35%

CO ECSA

-5%

a b

c d

0.0 0.2 0.4 0.6 0.8 1.0

-6

-5

-4

-3

-2

-1

0.0

0.1

jm / A mg-1

initial

after HP-AST

j / mA cm-2

E / V vs RHE

- 15 %

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

nanoparticles after LP-AST. By contrast, the charge transferred in the Hupd region (0.05-0.4 V)

decreased by only 8% as shown in Figure 4.5 representing a minor loss in the hydrogen-based

ECSA (Hupd-ECSA). The corresponding linear sweep voltammograms (LSV) in Figure 4.4b

revealed a loss of the platinum mass-based activity (jm) by 54% after LP-AST as evaluated at

0.9 V. To quantify the number of oxidizable Pt sites and to investigate their relation to the loss

in mass activity, CO stripping experiments were performed at selected stages of the

electrochemical stability test (Figure A3a).

Figure 4.5 Comparison of ECSAs based on the integration in the Hupd and the CO oxidation potential

range. CO-ECSAs were determined by integrating the CO oxidation peak area from the first cycle of

the CO stripping experiment after subtraction of the second cycle representing

the bare CO-free Pt surface. Hupd-ECSA was determined by subtracting the first from the second cycle

of the CO stripping experiment.

Figure 4.5 shows a comparison of ECSAs determined in the Hupd and CO oxidation region of

the CO stripping experiment. Note that the absolute values of CO-ECSAs as presented in

Figure 4.5 are larger compared to the Hupd based values due to a stronger adsorption of CO on

Pt as it is well known from literature for Pt-based systems.117,136-139 However, it can be seen that

the CO-ECSA decreased in the course of the LP-AST (-29%) measurement, whereas the

Hupd-ECSA was rather stable (-8%). This also becomes clear as the ratio (CO/Hupd) of both

ECSAs decreased as seen by Table A1. After the stability test in the higher potential range from

1.0-1.5 V (denoted as HP-AST) the cyclic voltammogram exhibited higher currents in the Hupd

region compared to the initial state representing an increase of the Hupd-ECSA by 35%

(Figure 4.4c). In contrast to the LP-AST, mass-based activity of Pt/ITO decreased by only 15%

+35%-5%-8%

before

after

HP-ASTLP-AST

Hupd

Hupd CO

ECSA / m2 gPt-1

-29%

before

after

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

(Figure 4.4d) and the CO-ECSA by only 5% during the HP-AST (Figure 4.5). Thus, the

catalytic activity of the Pt/ITO is signifcantly more stable in the HP-AST than in the LP-AST.

However, in both cases the catalytic activity followed rather the CO-ECSA than the

Hupd-ECSA.

To compare the Pt/ITO electrocatalyst to a commercial carbon supported material, a 20 wt%

Pt/C reference catalyst was applied to the same ASTs. Figure A2a,b shows CVs and LSVs

before and after the LP-AST. Pt/C exhibited higher stability in the lower potential regime than

Pt/ITO with a loss in mass activity of only 14% (see also Table A2). In contrast to that, the

carbon supported catalyst showed a decrease of 38% in ORR mass activity after the HP-AST

and an increase in Hupd-ECSA of 9% (Figure A2c,d) suggesting a lower electrochemical

stability than Pt/ITO in this potential regime. To explain the changes in electrochemical surface

area and catalytic activity toward the ORR, the evolution of the Pt particle size on ITO after the

ASTs was analyzed by TEM (Figure 4.3d-i). The size of the Pt nanoparticles increased only

slightly by around 1 nm from 5.3 ± 0.7 to 6.4 ± 0.9 nm after LP-AST (Figure 4.3d-e) which

could explain the small decrease observed in Hupd-ECSA. Additionally, the edges of ITO

nanocrystals appeared rather rounded compared to the faceted ITO crystallites of the initial state

and the Pt particles seem to be more agglomerated. After HP-AST (Figure 4.3g-i), the Pt

nanoparticles did not grow as they exhibited an average diameter of 5.2 ± 0.6 nm. Therefore,

the observed increase in Hupd-ECSA cannot be explained by the stable particle size distribution.

In contrast to LP-AST, the nanocrystals of the ITO support did not only change in shape after

HP-AST but also did not appear to be well separated anymore indicating coalescence to larger

particles.

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

4.4. XPS and STEM/EDX Results

To get deeper insight into the changes of the Pt surface, XPS was performed on the Pt/ITO

before and after the LP-AST and Figure 4.6 shows the Pt 4f as well as the In and Sn 3d spectra

of Pt/ITO. From the Pt 4f depth profile it can be seen that Pt in both samples is mostly metallic

with a binding energy of around 71.1 eV (Figure 4.6a,b).

Figure 4.6 X-ray photoelectron spectroscopy measurements. (a) and (b) are showing the Pt 4f depth

profiling accessed by the kinetic energy of the photoelectrons of 210, 550 and 1200 eV at the initial (a)

and the state after LP-AST (b) Dotted lines represent measured data and solid lines the fits and

component peaks. (c) and (d) are showing the 3d core levels for In and Sn, respectively, each at the

initial and the cycled state.

The initial state exhibited also a small fraction of PtOx species (contribution at around 72 eV)

which was significantly weaker at deeper probing depth of the photoelectrons and absent after

the LP-AST. This finding is in agreement with the high resolution TEM (HR-TEM)

micrographs in Figure 4.2c where a sub-nanometer amorphous shell around the metallic Pt

nanoparticles is visible in the as-prepared state. This shell can be explained with a thin layer of

Pt oxide around the nanoparticles that disappeared upon electrochemical cycling. The ITO was

a b

c d

456 454 452 450 448 446 444 442 440

Intensity / a.u.

Binding Energy / eV

initial

after LP-AST

In 3d

498 496 494 492 490 488 486 484

Intensity / a.u.

Binding Energy / eV

initial

after LP-AST

Sn 3d

78 76 74 72 70 68

KE: 1200 eV

KE: 550 eV

KE: 210 eV

Intensity / a.u.

Binding Energy / eV

Pt 4f

78 76 74 72 70 68

Pt 4f

Intensity / a.u.

Binding Energy / eV

KE: 210 eV

KE: 550 eV

KE: 1200 eV

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

assessed by In and Sn 3d core levels (see Figure 4.6c,d) and no difference was identified

between the two analyzed states.

Table 4.1 shows the quantitative results from the XP spectra as obtained at two different

positions of the sample. On the one hand the Pt/(In + Sn) ratio increased slightly indicating

support leaching and furthermore, the In/Sn ratio changed toward lower Sn concentration.

Table 4.1 Near-surface composition of Pt/ITO before and after LP-AST in the lower potential regime

as obtained from XPS measurements. The molar ratios Pt, In, and Sn were determined from the peak

areas of the Pt 4f as well as In and Sn 3d spectra recorded with a kinetic energy of the photoelectrons of

550 eV. The number in brackets denotes the composition at a second beam position of each sample.

Pt/(In+Sn)

Sn/In

initial

0.2 (0.21)

0.17 (0.17)

after LP-AST

0.25 (0.27)

0.14 (0.13)

To analyze the Pt surface and the elemental distribution in the catalyst in more detail, HAADF-

STEM images and EDX maps were acquired. In Figure 4.7a,b the STEM-EDX mappings of a

Pt nanoparticle deposited on ITO particles are shown before and after the LP-AST.

Figure 4.7 HAADF-STEM images and EDX mapping of (a) Pt/ITO initial and (b) after LP-AST.

Platinum is depicted in light blue, indium in green, and tin in red.

Pt can be clearly distinguished from the support and no clear difference between the two states

in terms of Pt particle size and elemental distribution is observed. In addition, In and Sn were

homogenously distributed in the ITO support and did not change drastically after potential

cycling.

3 nm

Sn PtIn

after LP-AST

3nm

initial

Sn PtIn

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

4.5. In Situ HE X-ray Investigation

We performed in situ high energy XRD and ASAXS to track and analyze the structural and

morphological evolution of Pt particles as well as of In2–xSnxO3 crystallites during the AST and,

thus, to gain deeper understanding about their degradation pathways. Figure 4.8a shows the

evolution of the in situ X-ray diffraction patterns during the LP-AST including the initial state.

Figure 4.8 In situ HE XRD measurements depicted as the evolution of diffraction patterns from the

initial state to the end of the electrochemical cycling for the LP-AST (a) and the HP-AST (b).

It can be clearly seen that the Pt reﬂexes did not decrease in intensity whereas the intensity of

the In2–xSnxO3 reﬂexes decreased drastically and thus, Pt/In2–xSnxO3 ratio increased. The results

from Rietveld refinement of the HE-XRD patterns are shown in Figure 4.9a,b. It can be seen

that the crystalline fraction of In2–xSnxO3 decreased from 79% to 61% during the LP-AST.

Figure 4.9 Results from in situ HE-XRD for LP-AST. Weight fractions of (a) crystalline phases and (b)

their crystallite size as determined by Rietveld refinement.

-1000 0 1000 2000 3000 4000 5000

In2O3

crystallite size / nm

cycle number

initial

-1000 0 1000 2000 3000 4000 5000

Xcryst. phase / wt%

In2O3

cycle number

initial

a b

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

The scale factor of the two crystalline phases (Figure 4.10a) confirms the constant absolute

amount of Pt and the loss of In2–xSnxO3 during LP-AST. Simultaneously, the Pt coherence

length grew only slightly from the initial value of 4.33 ± 0.27 nm to 4.59 ± 0.27 nm, whereas

the In2–xSnxO3 coherence length strongly increased from 13.89 ± 0.29 nm to 16.19 ± 0.46 nm.

Figure 4.10 Evolution of normalized scale factors for Pt and In2O3 from fits of HE-XRD patterns over

the cycle number for (a) LP-AST and (b) HP-AST.

To gain a deeper insight in the evolution of the Pt nanoparticles, Pt-specific scattering curves

were obtained from ASAXS measurements around the Pt K-edge. Figure 4.11a shows the

background-corrected scattering curves with a clear feature visible above 0.1 Å-1 resulting from

Pt nanoparticles. By fitting these scattering curves and assuming a Gaussian distribution of

spheres the particle size distribution (PSD) can be obtained (Figure 4.11b).

a b

-1000 0 1000 2000 3000 4000 5000

0.0

0.2

0.4

0.6

0.8

1.0

In2O3

- 58%

LP-AST

normalized scale factor

cycle number

initial

-1000 0 1000 2000 3000 4000 5000

0.0

0.2

0.4

0.6

0.8

1.0

In2O3

- 36%

HP-AST

- 19%

normalized scale factor

cycle number

initial

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

Figure 4.11 In situ ASAXS measurements for LP-AST: background subtracted, Pt element specific

scattering curves for selected cycle numbers (a), particle size distribution (PSD) as a function of rel.

intensity over the mean particle diameter for selected cycle numbers (b), mean Pt particle diameter (c)

and the evolution of monodispersity as a function of σ over the cycle number (d).

We identified a bimodal PSD with a high fraction (≈99%) of Pt particles with a size of around

4.5 nm and a small fraction (≈1%) of larger particles with a diameter of 9 nm (Figure 4.11c).

This finding shows that only a very slight fraction of the Pt particles was agglomerated. The

size of Pt nanoparticles stayed almost constant during the LP-AST but the agglomerated Pt

particles shrink by ≈1 nm. Additionally, we noticed a slight decrease in size of the Pt particles

between the as-prepared state and the start of the stability test. Furthermore, we identified that

the polydispersity σ of the Pt nanoparticles decreased slightly from 0.40 to 0.38 (Figure 4.11d).

Thus, the results achieved by ASAXS measurements support the findings from Rietveld

analysis and show the excellent morphological and structural stability of the Pt nanoparticles,

which contrasts the evolution of the ITO support. Furthermore, we state that no significant loss

in the absolute intensity of the Pt scattering/reﬂections (as seen by the scale factor) was obtained

although the ITO support dissolved or became amorphous indicating a negligible loss of Pt

nanoparticles.

-1000 0 1000 2000 3000 4000 5000

large particles

small particles

Dmean / nm

cycle number

initial

-1000 0 1000 2000 3000 4000 5000

0.38

0.39

0.40

nm

cycle number

initial

1 2 3 4 5 6 7 8 9 10 11 12

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Intensity / a.u.

Dmean / nm

initial

wet

0 CV

2500 CV

5000 CV

0.1

100

1000

10000

100000

1000000

q/ A-1

Intensity / a.u.

No. CV:

1000

2000

3000

4000

5000

a b

c d

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

Figure 4.12 Results from in situ HE-XRD measurements for HP-AST. Weight fractions of (a) crystalline

phases and (b) their crystallite size as determined by Rietveld refinement.

Figure 4.8b shows the evolution of diffraction patterns during HP-AST. It can be clearly seen

that the intensities of the ITO reﬂexes decreased less strongly than during the LP-AST. The

evolution of fraction of the crystalline phases as determined by Rietveld refinement is shown

in Figure 4.12a. It can be seen that the fraction of In2–xSnxO3 decreased by around 5% from

67% to 62%, whereas the Pt fraction simultaneously increased by the same extent. From the

initial dry state to the state directly before the stability test an initial drop of the In2–xSnxO3

fraction of 3% was observed which can be explained with a partial dissolution (or

amorphization) of the In2–xSnxO3 during the CVs in the activation regime. During the first

500 CVs of the HP-AST the In2–xSnxO3 fraction increased again. The scale factor of the two

crystalline phases (Figure 4.10b) confirms the loss of absolute amount of Pt and of In2–xSnxO3

during HP-AST. Throughout the whole HP-AST, the coherence lengths of both Pt and

In2-xSnxO3 remained constant with a size of 4.25 ± 0.19 nm and 12.94 ± 0.44 nm, respectively

(Figure 4.12b).

-1000 0 1000 2000 3000 4000 5000

Xcryst. phase / wt%

In2O3

cycle number

initial

-1000 0 1000 2000 3000 4000 5000

In2O3

crystallite size / nm

cycle number

initial

a b

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

Figure 4.13 In situ ASAXS measurements for HP-AST: background subtracted, Pt element specific

scattering curves for selected cycle numbers (a), particle size distribution (PSD) as a function of rel.

intensity over the mean particle diameter for selected cycle numbers (b), mean Pt particle diameter (c)

and the evolution of monodispersity as a function of σ over the cycle number (d).

Complementary to these in situ HE-XRD measurements, the results from ASAXS experiments

for the HP-AST are shown in Figure 4.13. They show the stability of the Pt nanoparticles during

the HP-AST as the mean particle size of both smaller and larger particle (agglomerates)

remained constant (Figure 4.13c) and the polydispersity σ increased slightly from 0.41 to 0.43

(Figure 4.13d).

Generally, we state that perfect agreement between particle, crystallite size, and structural

coherence length determination by TEM, (A)SAXS, and HE-XRD demands very high degree

of homogeneity in shape, size, and crystallinity, which was not given in this study.140

01000 2000 3000 4000 5000

Dmean / nm

cycle number

large particles

small particles

initial

1 2 3 4 5 6 7 8 9 10 11 12

0.0

0.2

0.4

0.6

0.8

1.0

1.2 initial

0 CV

2500 CV

5000 CV

after 5000 CV

Intensity / a.u.

Dmean / nm

0.1

100

1000

10000

100000

1000000

q/ A-1

Intensity / a.u.

No. CV:

1000

2000

3000

4000

5000

-1000 0 1000 2000 3000 4000 5000

0.41

0.42

0.43

nm

cycle number

initial

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

4.6. In Situ Pt, In and Sn Dissolution by SFC ICP-MS

To differentiate between amorphization and dissolution/detachment of the ITO support and to

track the dissolution of Pt, In, and Sn, experiments with an in situ SFC coupled to an ICP-MS

were performed. The LP- and HP-ASTs were applied to the Pt/ITO and to the Pt/C reference

catalyst. Figure 4.14 shows the results from in situ SCF ICP-MS experiments for the two

different potential regimes.

Figure 4.14 In situ scanning flow cell ICP-MS measurements. Depicted are the Sn, In and Pt dissolution

rates and the applied electrochemical protocols from the bottom to the top for (a) LP-AST and (b) HP-

AST. The respective dissolution rates in detected (det) metal in μg per volume electrolyte (μgdet L-1) are

plotted against the time. A Pt/C reference sample was measured and therefore, the Pt dissolution rate

was also normalized to the Pt mass loading on the working electrode (WE) (μgdet L-1 μgWE-1). The

electrochemical protocol was conducted as follows: Beginning with 100 CVs (activation regime) from

0.05-1 V, followed by potential cycling in the LP regime (0.6-0.95 V, 40 CVs, 100 mV s-1) or in the HP

regime (1.0-1.5 V, 200 CVs, 500 mV s-1) and followed by another 3 cycles from 0.05-1 V, all CVs were

recorded with a scan rate of 100mV/s. The first contacts between catalyst and electrolyte (cell contact)

are denoted with arrows. Breaks at the time axis of (a) and (b) have been implemented between 500 and

1700 s. Graphs with complete time axis can be found in Figure A5 and Figure A6.

It can be seen that a large dissolution peak arises for all three metals when the first contact

between the electrode and the acidic electrolyte under open circuit potential (OCP) conditions

was established (denoted as cell contact). During potential cycling in the activation regime Pt

dissolution was observed in case of the ITO and C support. The Pt dissolution was up to

0.5 µgdet L-1 µgWE-1 but limited to the first potential cycles (see Figure 4.14a) and no major

difference between Pt/ITO and Pt/C was observed. Furthermore, no Pt dissolution was observed

during potential cycling in the activation and the LP regime.

During potential cycling in the activation and in the LP regime a low but constant dissolution

of In and Sn was observed (0.05 and 0.1 µgdet L-1, respectively). Thus, the ITO dissolved

continuously over the whole course of the applied electrochemical protocol without any specific

potential dependence. In case of the HP regime (Figure 4.14b) a relatively broad Pt dissolution

-500 0 2000 2500

0.0

0.5

1.0

1.5

-500 0 2000 2500

0.0

0.2

E / V vs RHE

Pt / gdet l-1 gWE

-1

Pt/C

Pt/ITO

cell contact

In / gdet l-1

Sn / gdet l-1

time / sec

02000

0.0

0.5

1.0

02000

0.0

0.1

0.2

E / V vs RHE

Pt/C

Pt/ITO

Pt / gdet l-1gWE

-1

cell contact

In / gdet l-1

Sn / gdet l-1

time / sec

a b

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

peak can be observed for both carbon and oxide supports, but again, Pt dissolution was

significantly more pronounced for Pt/C than for Pt/ITO. We determined the amount of Pt

dissolution during the first cycles in the HP regime to be 1·10-3 µgdet µgWE-1 for Pt/C and

3·10-4 µgdet µgWE-1 for Pt/ITO (see Figure 4.15). Note that Pt dissolution becomes more

pronounced during the activation potential cycles after the potential cycling in the HP regime

compared to that in the LP regime. A likely explanation is the reductive dissolution of the

electrochemically formed PtOx species.38,141 Pt-specific dissolution behavior with a small

oxidative dissolution peak and a big and sharp peak in the reductive scan of the AST were

superimposed by a broad and decreasing dissolution signal due to the relatively high scanrate

of 500 mV s-1 in the HP-AST.

Figure 4.15 Results from in situ SFC ICP-MS of the integration of the peaks arising from Pt dissolution

over time of the measurement (in μgdet) detected by the ICP-MS per mass loading Pt on the working

electrode (in μgWE) for LP- and HP-AST.

In addition to the dissolution peaks of In and Sn upon first cell contact, a small Sn dissolution

peak was observed at the beginning of the activation regime which remained comparable in

intensity to Sn dissolution during the LP-AST. Slight In, and a more pronounced Sn dissolution

was also detected during the first cycles in the HP regime and, thus, the observed Pt loss was

likely caused by Pt nanoparticle detachment due to support dissolution. Data evidence that Sn

is more prone to dissolve during the HP-AST. Generally, we state that only 0.04% (0.10%) of

Pt and 1.89% (2.47%) of the ITO support was dissolved over the whole course of the applied

though shorter electrochemical protocol for the LP (HP) regime. For comparison, in case of the

Pt/C reference 0.09% (0.29%) of the total Pt mass was dissolved (see also Table A3).

0.0E+00

5.0E-04

1.0E-03

1.5E-03 HP-AST

2nd

peak

1st

peak

program

start

Pt/ITO

Pt/C

Pt / gdet gWE

-1

cell

contact

cell

contact

program

start

LP-AST

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

4.7. Discussion

Associated with the need of developing new and stable electrocatalysts for PEMFCs, an in situ

stability study of ITO- supported Pt nanoparticles has been conducted. We were able to

successfully deposit Pt nanoparticles with a uniform size on the crystalline support. The catalyst

was tested for electrochemical stability under simulated fuel cell conditions by applying two

different ASTs representing conventional operating (0.6-0.95 V) and start-up/shut-down

(1.0-1.5 V) conditions. By employing a wide range of in situ (X-ray) techniques as well as

advanced spectroscopy and electron microscopy, we aimed to determine fundamental

descriptors for electrochemical catalyst and support stability as well as the most prominent

degradation pathway.

Electrochemical versus Structural Stability

The investigation of electrocatalytic stability in acidic media applying two different protocols

has revealed that Pt/ITO shows higher electrochemical stability in the HP-AST than in the LP-

AST. In particular, ORR activity and CO-ECSA decreased strongly during the LP-AST, while

Hupd-ECSA stays rather stable, suggesting that the CO-ECSA is a better descriptor for the ORR

active site density. These findings show that the Pt surface after LP-AST remains accessible for

H adsorbates represented by a stable Hupd-ECSA but not for O/OH adsorbates which are

essential for the oxygen reduction reaction. TEM analysis shows almost no growth of Pt

nanoparticles and a slightly higher degree of Pt agglomeration. This slightly higher degree of

agglomeration of the Pt particles can explain the small decrease of the accessible surface area

as measured via Hupd but the CO-ECSA decreased to a higher extent. In situ X-ray investigations

again proved that Pt particle and crystallite size was stable over the LP-AST and in situ SFC

ICP-MS showed negligible Pt dissolution. The morphological and structural stability of the Pt

nanoparticles contrasts the strong loss in catalytic activity. Thus, the major changes in ITO

indicate a strong inﬂuence of the morphological stability of ITO on the catalytic activity of Pt.

In situ X-ray investigations revealed that the fraction of crystalline ITO strongly decreased and

the crystallite size increased. Hence, the ITO is partially dissolved/detached or loses

crystallinity during LP-AST. From in situ SFC ICP-MS measurements a small but continuous

dissolution of In and Sn was observed, but the observed loss from HE-XRD in ITO weight

fraction is much higher suggesting partial amorphization of ITO. Dissolved In/Sn ions are prone

to be (electro) chemically and isostructurally redeposited on the ITO crystallites which can

explain the determined growth via Ostwald ripening.39 Alternatively, the loss of smaller

crystallites could lead to a larger mean crystallite size.

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

We have to note that the loss of crystalline ITO is not directly linked to a loss of Pt

nanoparticles. Rietveld refinement shows that the absolute amount of Pt remained stable during

the LP-AST, whereas the absolute amount of crystalline ITO strongly decreased (as measured

by the scale factor extracted from fits of the HE-XRD pattern, see Figure 4.10a). These findings

suggest that ITO gets preferentially dissolved where locally no Pt is present and that the Pt

nanoparticles might be stabilizers of ITO. By acting as electron scavenger the Pt inhibits the

electrochemical reduction of the surface In/Sn ions and thus, reduces their dissolution into the

electrolyte. This function is similar to the role of an appropriate cocatalyst on photoanodes

acting as hole scavengers leading to enhanced photoelectrocatalytic stability.142 Thus, we can

conclude that while Pt is almost unaffected by the conditions present during LP-AST, the strong

changes of the ITO inﬂuence the stability of the electrocatalytic activity of the Pt nanoparticles.

The AST in the HP regime from 1.0-1.5 V revealed a significantly better stability of the catalytic

activity. TEM analysis and in situ X-ray results showed that ITO nanocrystals and Pt

nanoparticles did not grow and the composition remained stable compared to the LP-AST. This

agrees with a rather stable CO-ECSA and hence with a stable mass-based activity, which again

shows the dependency of these two descriptors. By contrast, the Pt surface area became larger

as it exhibited a higher Hupd-ECSA. This can be explained by an electrochemical roughening of

the Pt surface as elevated Pt dissolution was observed rom in situ SCF ICP-MS during the

HP-AST. Furthermore, the dissolved Pt ions could be redeposited on the cathode as sub-

nanometer clusters or single atom sites which we did not detect by electron microscopy and

HE-XRD. But these domains would contribute to the increase in the Hupd-ECSA and would

lower the dissolution rate with respect to the loss of crystalline Pt nanoparticles. The Pt

dissolution during HP-AST was accompanied by small In and more pronounced Sn dissolution

affecting the shape of the ITO crystallites as observed in TEM images. However, these changes

inﬂuenced the structure and morphology of the ITO less strongly than during the LP-AST and

without a tremendous impact on the catalytic performance. Additionally, the loss of crystalline

In2–xSnxO3 from in situ HE-XRD could rather be assigned to support leaching than to

amorphization as concluded for LP-AST.

The comparison of the electrochemical results revealed a higher stability for the Pt/C reference

in the LP-AST when compared to the Pt/ITO catalyst. Dissolution of Pt in Pt/C during the

LP-AST was very low and comparable to that of Pt/ITO, indicating similar morphological Pt

stability in Pt/C and Pt/ITO. After HP-AST the reference material revealed a lower stability

with a relatively big loss in catalytic activity. From in situ ICP-MS it can be seen that during

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

cycling in the range of 1.0-1.5 V Pt dissolution was much more pronounced in Pt/C than in

Pt/ITO. This can be explained with a stronger electronic interaction between the metal particles

and the oxide support impeding nanoparticle detachment or dissolution but could also be caused

by the differences in particle size and, thus, electrochemical surface area. In that respect, initial

Pt dissolution peaks at cell contact due to PtOx reduction at OCP conditions were again smaller

for ITO supported Pt. ITO might decrease PtOx formation resulting in lower peak intensity

compared to the Pt/C reference.

Catalyst/Support Degradation Pathways

From the above described electrochemical and structural behavior of Pt/ITO under operating

conditions we conclude that the support stability has a tremendous impact on the stability of the

catalytically active sites. We are able to exclude well-known catalyst degradation phenomena

caused by Pt particle growth, agglomeration, and dissolution/detachment and propose a

pathway based on strong inﬂuence of support degradation on the availability of Pt active sites.

In detail, we found an overall excellent structural and morphological stability of Pt but we

determined altering of the ITO support as a likely actuator for Pt (surface) poisoning and

resulting performance loss in the ASTs, especially in the LP-AST.

For the LP-AST, XPS data indicate support loss in the form of Sn leaching and continuous but

small support dissolution was also found by in situ ICP-MS. The dissolved metal ions could be

electrochemically redeposited not only preferentially on the ITO surface but also on the Pt

surface during the relative reductive conditions of cycling between 0.6 and 0.95 V. The

increasing Pt/(In+Sn) ratio from XPS after LP-AST is dominated by the preferential dissolution

(or detachment) of the support but does not exclude Pt surface modification.

Consequently, we aim to uncover the origin for this poisoning effect in the form of proposed

degradation pathways. Possible mechanisms for Pt surface poisoning could be (a) support metal

incorporation into Pt nanoparticles leading to alloy formation, (b) formation of an amorphous

metal (oxyhydr)oxide layer, or (c) redeposition of In and/or Sn ions on the Pt surface. Metal

diffusion into Pt leading to alloy formation preferentially due to Sn alloying into the Pt lattice

could also lower catalytic activity. In that respect, no other phase was observed during potential

cycling in the HE-XRD. Nevertheless, the evolution of Pt lattice constant based on Rietveld

refinement during the LP-AST was tracked, because Partial PtSn alloy formation should

increase the lattice constant (PtSn3, PDF#00-035-1360). Figure A7a shows the lattice constant

as a function of the cycle number in the LP-AST. Here, it can be seen that the lattice constant

decreases from beginning to end of cycling. Furthermore, the formation of a near-surface alloy

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

leading to a PtSn shell is unlikely because this would lead to a decrease of the Pt coherence

length. Thus, PtSn alloy formation is unlikely and can therefore be excluded for the major Pt

poisoning contributors.

Another interpretation of the combination of electrochemical and in situ results can be the

partial transformation of ITO to an amorphous In/SnOx(OH)y. This (oxyhydr)oxide formation

could then lead to a (partial) encapsulation in the form of a thin (mono)layer around Pt. This

theory would go in line with relatively strong loss of crystalline In2–xSnxO3 fraction from HE-

XRD and only little In and Sn dissolution from ICP-MS. Such a layer would be permeable for

protons but not for larger molecules like CO or OH and could thus, explain the different trends

in Hupd- and CO-ECSA. This phenomenon, also known as “decoration effect,” originates from

SMSI and is well known in literature for different metals (Rh, Pt, Pd, Ni, Co) supported on

metal oxides (TiO2, Fe3O4, WOx).143-152 In these studies, the encapsulation has been induced by

thermal/chemical reduction of the oxide support as well as electrochemical reduction at low

potentials in the case of Pt/WOx. In our case, the LP-AST in acidic electrolyte implies reductive

reaction conditions for the oxide support. Furthermore, in photocatalysis this effect is

intentionally induced to suppress ORR in hydrogen evolution catalysts by growing a proton-

and hydrogen-permeable Cr2O3 shell around an Rh catalyst. This shell is impermeable for O2

molecules inhibiting undesired ORR.153 We identified a marginal growth of Pt particles by less

than 1 nm in TEM micrographs after LP-AST (Figure 4.3), while the coherence length/domain

size (as obtained by HE-XRD) and particle size (as obtained by ASAXS) remained constant.

This finding would go in line with the formation of sub-nanometer (oxyhydr)oxide layer around

Pt but can also be within the error of particle size estimation from TEM due to higher degree of

agglomeration. However, we could not prove the formation of this layer by STEM images and

EDX spectroscopy even though a sub-monolayer coverage or adsorption might be hard to

detect. Also XP spectroscopy does not give the sensitivity to differentiate between the

chemically unchanged isostructurally grown In2–xSnxO3 crystallites and sub-monolayer

coverage of the support metal ions on the surface of the Pt nanoparticles. However, we expect

the ad-atoms to differ chemically from the Sn/In ions in the In2–xSnxO3.

It is known from literature that In and/or Sn metal adsorbates present as sub-monolayer on Pt

single crystal surfaces can strongly affect the CO oxidation behavior.154 Therein, adsorbed Sn

on various Pt surfaces lead to a growth of the oxidation peak at 0.53 V. In case of the Pt (111)

surface Sn coverage blocks the more anodic CO oxidation peak at 0.65 V, whereas it does not

affect the CO electrooxidation on Pt (100) and (110) surfaces. CO electrooxidation on the

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

Pt/ITO catalyst also shows a preferential poisoning of the high potential oxidation peak after

the LP-AST (Figure A3), and thus, our electrochemical data are in agreement with the surface

poisoning of the Pt nanoparticles with metal (Sn) ions. We note that the anodic shift of the CO

oxidation potential might be caused by the In adsorbates in addition to Sn. In another approach,

Pt supported on Nb- and Sb-doped SnO2 was used as catalyst for single cell stability tests and

it was found that Sn was dissolved at reducing potentials resulting in Pt poisoning by Sn

redeposition.155 Nakada et al. studied SnO2 supported Pt electrocatalyst and they found

decreased Pt oxidation/reduction currents due to stronger interactions of Sn with Pt after

electrochemical cycling.156 Additionally, under potential deposited (UPD) Sn on Pt was found

to change the Pt oxidation behavior when cycling in HClO4.157

In case of the HP-AST the dominant degradation mechanism is Pt dissolution from PtOx species

by reaching potentials as high as 1.5 V. Interestingly, ITO support dissolved as well in this

potential regime leading to support modification, but did not affect the electrochemical

performance drastically. We explain that with the slight but continuous dissolution of surface

Pt atoms under oxidative conditions, which continuously pours away potentially adsorbed In/Sn

atoms. This process prevents significant poisoning with In/Sn atoms and, thus, leads to a rather

stable mass-based catalytic activity. Furthermore, cathodic potentials as low as 0.6 V as in

LP-AST might be required to form metal (oxyhydr)oxides or Sn UPD layer on the Pt surface.

For the HP-AST regime we clearly established a superior structural stability of Pt nanoparticles

on ITO versus C supports. Some earlier work put forward strong electronic interactions between

metal particles and the oxide support impeding nanoparticle detachment.59,158 Recent DFT

calculations showed that the electronic inﬂuence of oxide support is limited to a few atom layers

of Pt.159 The Pt nanoparticles in this work are significantly larger and, thus, electronic

interactions may remain masked from spectroscopic identification although they are indeed

anchored more strongly to oxide supports.

Combining all these results, we propose that the activity loss during the LP-AST is caused by

a kind of Pt surface modification due to In and, more likely, Sn accumulation. During HP-AST

this degradation pathway is blocked because slight anodic dissolution of Pt keeps the active

sites accessible.

4. In Situ Stability Study of Pt Supported on Indium Tin Oxide

4.8. Conclusion

We have explored the stability and degradation of Pt nanoparticles supported on ITO during

the electrocatalytic reduction of molecular oxygen as well as during two distinct

degradation regimes corresponding to normal fuel cell operation and a start-up/shut-down

operation, respectively. The Pt/ITO electrocatalyst consisted of monodisperse, well-distributed

Pt particles with a size of around 5 nm and was investigated using a number of different in situ

techniques (i) to gain fundamental insight in the degradation mechanisms, (ii) to unravel

fundamental descriptors of structural and morphological stability, and (iii) to determine stability

limitations of this material system. Our combination of X-ray and electron microscopic

techniques has advanced our understanding of the fundamental processes associated with the

degradation behavior of Pt/ITO electrocatalysts.

Under catalytic ORR operation conditions (LP-AST) we have demonstrated that Pt

nanoparticles remain morphologically stable and degradation pathways due to Pt instability can

be excluded, whereas the ITO support suffers from cathodic corrosion coaffecting the catalytic

stability and mechanical/chemical attachment of the Pt nanoparticles. ITO crystallites partially

dissolve, become amorphous, and grow via an Ostwald ripening mechanism. Therefore, we

have strong indications that the significantly declined catalytic ORR activity after LP-AST is

attributed to partial In/Sn redeposition and concomitant surface decoration of active Pt particle.

Such oxidic Pt surface adlayers change the Pt redox chemistry and lower the density of specific

O-adsorption sites. An encapsulation of the Pt nanoparticles with an (oxyhydr)oxide layer could

equally account for this behavior, yet was not fully supported by the available data. By contrast,

during start-up/shut-down cycles (HP-AST), the catalyst/support couple showed excellent

stability, thanks to the improved structural and morphological durability of the oxide support

compared to the commercial Pt/C couple. We found evidence for slight continuous anodic Pt

dissolution which aided in suppressing Pt surface poisoning and thus stabilizing the available

active site density.

Overall, this study has unraveled the major degradation pathways of ITO-supported Pt

nanoparticle fuel cell catalysts over wide cathode potential ranges; our findings advance our

understanding of the fundamental aspects of support stability. They are also of practical value

for the development of improved fuel cell devices, as they offer design criteria and performance

limitations of oxide-based PEMFC cathode catalysts.

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

5. The Impact of Carbon Support Functionalization on the

Electrochemical Stability of Pt Fuel Cell Catalysts

Nitrogen-enriched porous carbons have been discussed as supports for Pt nanoparticle catalysts

deployed at cathode layers of proton exchange membrane fuel cells (PEMFC). Here, we present

an analysis of the chemical process of carbon surface modification using ammonolysis of pre-

oxidized carbon blacks, and correlate their chemical structure with their catalytic activity and

stability using in situ analytical techniques. Upon ammonolysis, the support materials were

characterized with respect to their elemental compositional, the physical surface area and the

surface zeta potential (ZP). The nature of the introduced N-functionalities was assessed by

X-ray photoelectron spectroscopy (XPS). At lower ammonolysis temperatures, pyrrolic-N were

invariably the most abundant surface species while at elevated treatment temperatures

pyridinic-N prevailed. The corrosion stability under electrochemical conditions was assessed

by in situ high temperature - differential electrochemical mass spectroscopy (HT-DEMS) in a

single gas diffusion layer (GDL) electrode; this test revealed exceptional improvements in

corrosion resistance for a specific type of nitrogen modification. Finally, Pt nanoparticles were

deposited on the modified supports. In situ X-ray scattering techniques (XRD and SAXS)

revealed the time evolution of the active Pt phase during accelerated electrochemical stress tests

(AST) in electrode potential ranges where the catalytic oxygen reduction reaction (ORR)

proceeds. Data suggest that abundance of pyrrolic nitrogen moieties lower carbon corrosion

and lead to superior catalyst stability compared to state-of-the-art Pt catalysts. Our study

suggests with specific materials science strategies how chemically tailored carbon supports

improve the performance of electrode layers in PEMFC devices.

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

Chapter 5 and section Appendix A2 were reprinted with permission from Ref 71 (Chem. Mat., 2018, 30 (20), 7287-

7295). Copyright (2018) American Chemical Society.

Henrike Schmies, Elisabeth Hornberger, Björn Anke, Tilman Jurzinsky, Hong Nhan Nong,

Fabio Dionigi, Stefanie Kühl, Jakub Drnec, Martin Lerch, Carsten Cremers, Peter Strasser,

“The Impact of Carbon Support Functionalization on the Electrochemical Stability of Pt Fuel

Cell Catalysts”, Chemistry of Materials 2018, 30 (20), 7287-7295; doi:

10.1021/acs.chemmater.8b03612

H.S. performed the experiments and analyzed the data, E.H. helped designing the experiments and discussing the

data; B.A. and M.L. performed the ammonolysis; T.J. and C.C. conducted the HT-DEMS measurements and

analysis; H.N.N. performed and analyzed the XPS measurements; F.D. and J.D. assisted in the synchrotron

experiments and data evaluation; S.K. recorded TEM images; H.S. and P.S. wrote the manuscript; all authors

assisted in discussing and writing the manuscript.

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

5.1. Introduction

In this work, we studied a family of nitrogen-doped carbons, prepared by ammonolysis, looked

at a range of their physicochemical properties and used the carbons as high surface area support

materials for catalytically active Pt nanoparticles. N-modified carbon-supported Pt particles,

with a high ration of pyrrolic nitrogen moieties, revealed exceptional catalytic performance

stabilities during accelerated stress tests. To learn about the origin and the mechanism of this

chemical stabilization, we utilized a range of in situ X-ray and mass spectrometric techniques.

These methods enabled us to pinpoint the underlying chemistry of the stabilization and to

exclude competing mechanisms. We close with a specific synthestic recommendation for more

corrosion stable fuel cell catalysts.

5.2. Compositional and Surface Characterization

The controlled modification of Vulcan XC72R powders was achieved by treatment in liquid

HCl, followed by subsequent oxidation in concentrated nitric acid (referrred to here as ”O-

Vulcan”), and then followed by ammonolysis in pure ammonia at two different temperatures

resulting in powder materials referred to as ”N-Vulcan 400°C” and ”N-Vulcan 800°C” (see

Figure 5.1).

Figure 5.1 Schematic illustration of the carbon modification procedure including oxidation step in

concentrated nitric acid resulting in O-Vulcan, and ammonolysis in pure NH3 at 400 and 800°C resulting

in N-Vulcan.

In order to analyze the degree of carbon modification, surface area and bulk compositional

analyses of the unmodified Vulcan, the oxidized Vulcan and the aminated Vulcans were

performed (Figure 5.2 and Table A4, Appendix A2). Elemental analysis revealed that highest

N content is achieved for N-Vulcan 400°C with 2.5 atomic (at)%, while for N-Vulcan 800°C it

was 1.5 at%, see Figure 5.2a. The oxidized Vulcan contained a small fraction of nitrogen,

presumably due to residues from the nitric acid treatment. The oxygen content was analyzed by

hot gas extraction and was highest for O-Vulcan with 12.6 at%, while a fraction of 3.0 and

0.9 at% is detected for N-Vulcan 400°C and N-Vulcan 800°C, respectively. This suggests that

the carbon modification through oxidation and ammonolysis involves the reduction of the

oxygenated surface groups to nitrogenated ones. Furthermore, it can be stated that this

modification approach leads to relative high concentrations of heteroatoms in the carbon but is

Vulcan XC 72r O-Vulcan N-Vulcan

Oxidation

Reflux in

conc. HNO3

Ammonolysis

5°C/min, NH3

2h at 400 and 800 °C

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

presumably dependent on the experimental conditions and the type of carbon used. Earlier

works applying comparable oxidation/amination routes resulted in slightly lower fractions of

heteroatoms.84-86,98

After the oxidation step, the BET surface area decreased from 300 m2 g-1 for Vulcan to

138 m2 g-1 for O-Vulcan (Figure 5.2b) which we ascribe to blocking of carbon pores. The value

stayed similar for N-Vulcan 400°C but increased to 252 m2 g-1 for N-Vulcan 800°C. The

increase of surface area with higher ammonolysis temperature could be attributed to ammonia

etching and clearance of previously blocked pores by the loss of oxygenated species (phenolic-

/ether-/carboxylic-groups84).

Figure 5.2 Surface and compositional analysis of the carbon materials: content of nitrogen (from

elemental analysis) and oxygen (from hot gas extraction) for the modified carbons (a), physical BET

surface area (b) and zeta potential (c) for modified carbons in comparison to the unmodified Vulcan.

Furthermore, the influence of carbon modification onto the electric surface zeta potential was

investigated. The zeta potential was positive (28 mV) for unmodified Vulcan and decreased to

a very negative value for O-Vulcan (-56 mV) and increased again after ammonolysis until it

reached 29 mV for N-Vulcan 800°C (Figure 5.2c). Herein, a negative surface potential can be

related to the presence of partially negatively charged O-groups on the surface, whereas the

reduction of the oxygenated surface groups into N-containing functionalities reflects a positive

surface potential. Accordingly, the observed variations in zeta potential confirm a successful

surface modification instead of bulk heteroatom incorporation. However, we do not expect any

drastic changes in conductivity upon carbon modification as shown in an earlier work by Arrigo

et al. on aminated CNFs.85

a b c

100

200

300

N-Vulcan

800°C

BET surface area / m2 g-1

Vulcan

O-Vulcan

N-Vulcan

400°C

-60

-40

-20

N-Vulcan

800°C

O-Vulcan

N-Vulcan

400°C

Vulcan

Zeta Potential / mV

N-Vulcan

800°C

N-Vulcan

400°C

Composition / at%

Nitrogen

Oxygen

O-Vulcan

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

5.3. Analysis on Carbon Surface Functionalization by XPS

To gain a deeper understanding of the chemical nature of the carbon functionalization, XPS

analysis of the aminated carbon support materials was performed. Four major N-functionalities

at distinct binding energies (BE) were identified by peak fitting and deconvolution (Figure 5.3a

and Table 5.1).

Table 5.1 Assignment of binding energy (BE) for different N-functionalities from individual

deconvolution of the N 1s XP spectra.

N-Species

BE / eV

Graphitic

402.8-402.9

Quaternary

401.2-401.4

Pyrrolic

400.0-400.3

Pyridinic

398.3-398.7

Graphitic N-functionalities were assigned to core level peaks around 402.8-402.9 eV, while

quaternary N functionalities, such as nitrogen atoms that substitute in-plane C atoms and carry

a partially positive charge or else substitute edge C atoms in six-membered rings and are

protonated, were assigned to a BE of 401.2-401.4 eV (see chemical structures in Figure 5.3b).

Peaks at 400.0-400.3 eV were lumped and assigned to surface species containing all possible

kinds of N-H bond motifs in five-membered rings, most prominently in pyrrolic N, where the

N contributes with two p-electrons to the π-system, but can also be ascribed to imides or

lactams. Pyridinic N, in which the N atom contributes with one p-electron to the π-system of a

six membered ring, are ascribed to peaks in the BE region of 398.3-398.7 eV, see also

Table 5.1.84-86,95 Figure 5.3a displays the N 1s spectra of the stepwise modified carbon

materials, including the individual peak deconvolutions for different types of N-moieties.

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

Figure 5.3 (a) XPS N 1s spectra and individual peak deconvolution for modified carbons and Pt/N-

Vulcan 400°C and (b) schematic illustrastion of different N-functionalities (green: graphitic N, blue:

quaternary N, yellow: pyrrolic N and red: pyridinic N) in graphene-like plane.

N-Vulcan 400°C shows the highest fraction of pyrrolic-type functionalities with around 40 %,

while for N-Vulcan 800°C pyridinic-type N-groups were most abundant species with 47 %

(Figure 5.4). The fraction of quaternary N decreased slightly with increased amination

temperature from ca. 15 % to 11 %, while the amount of graphitic N increased slightly from

6 % to 9 %. The total amount of surface nitrogen was highest for the N-Vulcan 400°C with

around 3.9 at% and decreases to 2.3 at% for the samples aminated at 800°C. This could be

explained by a temperature-dependent stability of N functionalities as well as by a lower local

ammonia partial pressure in the synthesis reactor, due to the dynamic nitrogen, hydrogen, and

ammonia equlibrium. Thus, pyridinic N was most stable at higher ammonolysis temperatures

in agreement with earlier ammonolysis studies84-86 and is genereally assumed to originate from

the decomposition of pyrrolic N by the release of HCN.86

A small amount of surface N was also found in O-Vulcan (0.7 at%) that is mostly ascribed to

residual nitrite/nitrate groups (NO2-/NO3-) from inclomplete removal of nitric acid.

Additionally, N-moieties at lower BE around 400-402 eV are present in O-Vulcan. The

occurrence of these N-functionalities in oxidized carbons is often observed in literature160-162

yet their origin is somewhat unclear. However, it is mostly believed to be formed by the

reduction of the nitric groups through the X-ray beam during the XPS measurement.85 The

nitrite/nitrate residues are not removed upon amination as they are also seen in the

deconvolution for the N-Vulcans but only contribute to the total N amount in a small extent.

a b

410 408 406 404 402 400 398 396

N-Vulcan 800°C

Pt/N-Vulcan 400°C

N-Vulcan 400°C

Normalized Intensity / a. u.

Binding energy / eV

exp. data

fit envelope

bkg

NO-

Graphitic

Quaternary

Pyrrolic

Pyridinic

N 1s

O-Vulcan

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

Figure 5.4 Fractions of the different N-moieties from deconvolution of the N 1s spectra in the modified

carbons and in Pt/N-Vulcan 400°C.

Deconvolution of the C 1s spectra (Figure A8) reveals a high majority of C-C/C-H species for

both unmodified and modified carbons. The amount of O-containing C-species changes due to

the different amounts of O found in the carbon surface (Table A4). Highest O-content on the

surface as determined by XPS is found for O-Vulcan with around 13 at%, which decreases with

increasing ammonolysis temperature to ca. 7 at% for N-Vulcan 800°C as similarly observed

from elemental analysis. Generally, a higher total fraction of N and O found by XPS analysis

points to the high degree of surface instead of bulk carbon modification.

O-Vulcan Pt/N-Vulcan

400°C

N-Vulcan

400°C

N-Vulcan

800°C

fraction / %

NO-

Graphitic

Quaternary

Pyrrolic

Pyridinic

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

5.4. Corrosion Behaviour by HT-DEMS

The carbon corrosion was quantitatively investigated by HT-DEMS performed at 140°C for the

unmodified and the N-modified carbon supports (Figure 5.5). The electrochemical current (j)

normalized to the initial mass of carbon, as depicted in Figure 5.5a, showed a similar trajectory

for unmodified and modified carbons up to 0.9 V. Past this potential, however, the unmodified

Vulcan displayed a sharp increase in corrosion current accompanied with a comparable strong

increase in mass ion current with m/z=44 for CO2 indicating carbon corrosion (Figure 5.5b).

Figure 5.5 High Temperature-DEMS measurements from 0.06-1.05 V at 140 °C for modified carbons

and the Vulcan reference carbon with respect to the resulting current j normalized to mass loading of

carbon (a) and the ion current for CO2 (m/z = 44) from the MS normalized to mass loading of carbon

(b). HCl-Vulcan represents a HCl-treated Vulcan (1M HCl, RT, 24h) in order to remove metal traces

from the unmodified Vulcan.

The lowest currents and tendency towards carbon corrosion were observed for N-

Vulcan 400°C, for which also no NO-species (m/z=30) could be detected. This suggests that

the high degree of surface functionalization is beneficial for a superior carbon stability. O-

Vulcan shows slightly higher currents and corrosion rates than the aminated carbon, but still

features a much higher stability than the unmodified carbon. When Vulcan carbon was

pretreated with diluted hydrochloric acid (HCl-Vulcan), it showed comparable currents and

corrosion to O-Vulcan. This indicates that residual metal traces in the unmodified Vulcan,

which are removed upon HCl-treatment, might catalyze carbon corrosion. On the other hand, it

confirms that the presence of relatively high concentrations of O for O-Vulcan (as seen from

elemental analysis and XPS) on the surface does not necessarily favor carbon corrosion and,

most importantly, that the N-functionalization prevents the carbon from corrosion.

0.0 0.2 0.4 0.6 0.8 1.0 1.2

0.0

0.5

1.0

1.5

j / mA mg-1

E / V vs RHE

N-Vulcan 400°C

O-Vulcan

HCl-Vulcan

Vulcan

0.0 0.2 0.4 0.6 0.8 1.0 1.2

0.0

0.1

0.2

0.3

ion current / nA mg-1

E / V vs RHE

m/z=44 (CO2)

N-Vulcan 400°C

O-Vulcan

HCl-Vulcan

Vulcan

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

5.5. Pt Deposition and ORR Stability

In a next step, Pt was deposited on the unmodified and modified carbon supports. Therefore, a

wet impregnation/reduction approach was applied resulting in a Pt mass loading of around

20 weight (wt)% for all samples (see Table A5). The reduction of the Pt precursor was

performed thermally in a tube furnace at 200 °C leading to crystalline particles with X-ray

diffraction patterns that could be clearly attributed to fcc Pt (Figure 5.6). All samples consist of

very small Pt particles and larger agglomerates, as seen from the TEM images (Figure 5.7a and

Figure A9). Here, the degree of agglomeration is linked to the BET surface area, as for the

Pt/Vulcan with the highest BET surface area the Pt nanoparticles seem to be distributed most

homogeneously.

Figure 5.6 X-ray powder diffraction patterns of the modified and unmodified carbon supports and the

respective Pt electrocatalysts. Vertical blue lines represent the reference pattern for fcc Pt (PDF#00-004-

0802).

The samples were tested as powder electrocatalysts for the oxygen reduction reaction (ORR)

during prolonged cycling tests in order to evaluate their long-term performance stability. To

this end, accelerated stress tests (AST) were performed including 5k, 10k and 30k cycles

between 0.6-0.95 V. The resulting mass activity values at 0.9 V (jmass), the electrochemical

active surface area (ECSA) and specific activity (jspec) values are all shown in Figure 5.8. All

samples showed comparable mass activities of around 0.2 A mgPt-1, while the values for the

20 30 40 50 60 70 80 90

Pt/N-Vulcan 800°C

Pt/N-Vulcan 400°C

Pt/O-Vulcan

N-Vulcan 800°C

N-Vulcan 400°C

O-Vulcan

Pt/Vulcan

Vulcan

Intensity / a.u.

2 / °

Pt fcc, #00-004-0802

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

ECSA range between 50-70 m2 gPt-1. Here, the highest ECSA is observed for the unmodified

Pt/Vulcan, while the lowest for the oxidized Pt/O-Vulcan. We note that there is a direct

correlation between the hydrogen underpotential deposition (Hupd) -derived catalytically active

Pt surface area (ECSA) and the nitrogen sorption-derived BET surface area of the carbon

support. The larger BET surface area of the unmodified carbon (cf. Figure 5.2b) favors a higher

particle dispersion and is thus conducive for a larger value of the ECSA. The ratio between the

Pt mass normalized catalytic ORR activity and the ECSA value is the specific ORR activity

and represent the reactivity of the catalyst normalized to the real surface area of the Pt

nanoparticles. The initial specific activity of Pt/N-Vulcan 400°C (Figure 5.7b) was consistently

the largest with a high level of statistical confidence. This indicates a higher intrinsic catalytic

ORR activity of these Pt particles, likely a compounded consequence of a slightly larger mean

diameter of the Pt particles and possibly interactions between Pt particles and the modified N-C

support.

Figure 5.7 TEM images (a) and ORR specific activity (jspec) for the four catalysts (b) for Pt/Vulcan,

Pt/O-Vulcan, Pt/N-Vulcan 400°C and Pt/N-Vulcan 800°C. See Figure A9 in the Appedix section A2 for

more TEM overview images.

When tested for durability using an AST, Pt/N-Vulcan 400°C catalysts showed a remarkable

stability with an exceptionally small loss of only around 10% in jmass, whereas Pt/Vulcan lost

more than 30% of its mass activity after 30k cycles under identical conditions (see

Figure 5.8a,g). Both Pt/O-Vulcan and Pt/N-Vulcan 800°C showed lower stability, in particular

ca. 40% losses in jmass. While a large number of ORR stability studies of carbon-supported Pt

catalysts in comparable potential windows were reported in literature to date34,43,92,163-170, there

is only one study that actually reported stability data up to 30k cycles and that study reported

losses in mass activity of 60%.170 Hence, to the best of our knowledge, the cycling stability

0.0

0.1

0.2

0.3

0.4

0.5

Pt/N-Vulcan 800°C

jspec / mA cm-2

Pt/Vulcan

Pt/O-Vulcan

Pt/N-Vulcan 400°C

Pt/N-Vulcan 400°CPt/Vulcan

Pt/O-Vulcan Pt/N-Vulcan 800°C

a b

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

presented here for Pt/N-Vulcan 400°C is very high on a Pt mass activity basis of a supported

Pt/C catalyst in a fuel cell relevant potential range.

Figure 5.8 Results from AST for 5k, 10k and 30k cycles from 0.6-0.95 V in Nitrogen-saturated 0.1 M

HClO4 for Pt/Vulcan, Pt/O-Vulcan, Pt/N-Vulcan 400°C and Pt/N-Vulcan 800°C as a function of mass

activity at 0.9 V jmass(a,d,g,j), ECSA (b,e,h,k) and specific activity jspec (c,f,i,l).

Considering the time changes of the real active electrochemical surface area (ECSA),

comparable drops of 20-30% for all four catalysts were observed (see Figure 5.8b,e,h,k). As a

result of this, trends in the specific, Pt surface area-normalized activity followed those of the Pt

mass-based ORR activities: Those of Pt/Vulcan and Pt/O-Vulcan decreased by around 20%,

0.00

0.05

0.10

0.15

0.20

0.25

-36.9%

-28.5%

30k

jmass / A mg-1

initial 5k 10k

-3.8%

30k

ECSA / m2 g-1

initial 5k 10k

- 5.0%

-15.2%

-20.3%

0.0

0.1

0.2

0.3

0.4

0.5

- 4.4%

- 14.9%

jspec / mA cm-2

- 2.6%

5k initial 10k 30k

0.00

0.05

0.10

0.15

0.20

0.25

- 10.1%

- 4.1%

30k

jmass / A mg-1

initial 5k 10k

- 9.1%

-20.3%

-15.2%

30k

ECSA / m2 g-1

initial 5k 10k

- 5.0%

-19.9%

-13.3%

30k

ECSA / m2 g-1

initial 5k 10k

-15.4%

0.00

0.05

0.10

0.15

0.20

0.25

-36.6%

-30.4%

30k

jmass / A mg-1

initial 5k 10k

-24.4%

-32.2%

-24.4%

30k

ECSA / m2 g-1

initial 5k 10k

-10.6%

0.00

0.05

0.10

0.15

0.20

0.25

-44.4%

-35.1%

30k

jmass / A mg-1

initial 5k 10k

-4.6%

a b

d e

g h

j k

Pt/Vulcan

Pt/O-Vulcan

Pt/N-Vulcan 400°C

Pt/N-Vulcan 800°C

0.0

0.1

0.2

0.3

0.4

0.5

- 20.9%

- 19.8%

- 10.7%

jspec / mA cm-2

initial 5k 10k 30k

0.0

0.1

0.2

0.3

0.4

0.5

-18.0%

-18.5%

jspec / mA cm-2

+6.7%

5k initial 10k 30k

0.0

0.1

0.2

0.3

0.4

0.5

+ 12.9%

+3.3%

jspec / mA cm-2

- 3.9%

5k initial 10k 30k

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

that of Pt/N-Vulcan 800°C decreased by 4%, while that of Pt/N-Vulcan 400°C actually showed

an increase of around 13% (Figure 5.8c,f,i,l). So we can conclude that support modifications

have a significant influence on the long-term electrochemical stability of the catalyst/support

couples, but do not directly cause an improvement of initial catalytic ORR activity. Amination

at an intermediate temperature resulted in the optimal overall electrochemical performance

stability.

In order to track the morphological and structural stability of the catalytically active Pt

nanoparticles in more detail, in situ high energy X-ray diffraction (HE-XRD) and small angle

X-ray scattering (SAXS) experiments during stability cycling were conducted. The diffraction

patterns for all four samples at selected stages of the stability test are presented in Figure A10.

Direct comparison between Pt/Vulcan and Pt/N-Vulcan 400°C is shwon in Figure 5.9.

Figure 5.9 In situ high energy X-ray diffraction patterns over 5k cycle of the AST for Pt/Vulcan (a),

and Pt/N-Vulcan 400°C (b). A full set of in situ HE-XRD patterns for all samples can be found in the

Appendix section A2, Figure A10.

During the stability test, the diffraction patterns did not significantly change, which is reflected

in a large stable crystallite size obtained from Rietveld refinement (see Figure A11). The

crystallite sizes for the different catalysts ranged from 3.7 nm for Pt/Vulcan to 4.8 nm for

4 6 8 10 12 14 16 18 20

No. CV

2500

5000

Intensity / a.u.

2 @ 68.5 keV / °

4 6 8 10 12 14 16 18 20

carbon from substrate

No. CV

2500

5000

Intensity / a.u.

2 @ 68.5 keV / °

Pt/Vulcan

Pt/N-Vulcan 400°C

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

Pt/N-Vulcan 400°C and showed negligible alterations or trends during the stress test. The direct

comparison between the in situ Pt crystallite size trajectories of the 400°C-aminated Vulcan

and the unmodified Vulcan support is presented in Figure 5.10. Interestingly, both the

electrochemically stable catalyst (Pt/N-Vulcan 400°C) and the electrochemically unstable

catalyst (Pt/Vulcan) showed little to no changes in crystallite size. This excludes particle growth

by Ostwald ripening as a major source of the observed ECSA and Pt mass activity losses. To

account for the comparable ECSA loss of all catalysts, we hypothesize that well-ordered

crystallites aggregate on their supports or get buried into the support bulk during the AST.

Figure 5.10 Crystallite sizes obtained from Rietveld Refinement of the in situ HE-XRD patterns over

5k cycles of the AST for Pt/N-Vulcan 400°C (a) and Pt/Vulcan (b). Inlets in both graphs showing the

mass activity up to 30k cycles between 0.6-0.95 V of the AST in nitrogen-saturated 0.1 M HClO4. A

full set evolution of crystallite size for all samples can be found in the Appendix section A2, Figure A11.

Similar observations can be made from the in situ SAXS results. Here, no major changes were

observed in the scattering profiles for all four samples (Figure 5.11 and Figure A12), coupled

to negligible changes in the mean particle diameter, as derived from the individual scattering

curves (Figure 5.11 and Figure A13). The initial particle sizes ranged from 2.8 nm for Pt/Vulcan

to 3.6 nm for Pt/O-Vulcan and Pt/N-Vulcan 400°C. Only minor changes were observed during

the AST, for Pt/Vulcan and O-Vulcan the mean particle size decreased by around 0.2 nm, while

a small increase by less than 0.2 nm after 5k cycles was observed for Pt/N-Vulcan 400°C, see

Figure 5.11d. We note that the observed variations were close to the experiemental error. Due

to the inhomogeneity in Pt particle distribution and size in all four samples, crystallite sizes

from Rietveld refinement and particle sizes from SAXS are not in good agreement to each other.

This is mainly due to the fact that SAXS fitting requires assumptions with respect to the particle

size distribution owing to the relative “feature-less” scattering curves for all four samples

(Figure A12).

Pt/N-Vulcan 400°C

01000 2000 3000 4000 5000

3.6

3.8

4.0

4.2

4.4

4.6

4.8

5.0

crystallite size / nm

cycle number

0.00

0.05

0.10

0.15

0.20

0.25

- 10.1%

- 4.1%

30k

jmass / A mg-1

initial 5k 10k

- 9.1%

Pt/Vulcan

0.00

0.05

0.10

0.15

0.20

0.25

-36.6%

-30.4%

30k

jmass / A mg-1

initial 5k 10k

-24.4%

01000 2000 3000 4000 5000

3.6

3.8

4.0

4.2

4.4

4.6

4.8

5.0

cycle number

crystallite size / nm

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

Figure 5.11 In situ small angle X-ray scattering curves and mean particle diameter from SAXS fitting

over 5k cycle of the AST for Pt/Vulcan (a,b) and Pt/N-Vulcan 400°C (c,d). A full set of in situ SAXS

curves and evolution of mean particle diameters for all samples can be found in the Appendix section

A2, Figure A12 and Figure A13.

The fact that there was no significant variation in the morphological stability of the Pt

nanoparticles among the catalysts suggests that the carbon supports were solely responsible for

the pronounced differences in long-term cycling stability (insets in Figure 5.10). While Pt/N-

Vulcan 400°C showed superior stability, Pt/Vulcan and Pt/O-Vulcan but also Pt/N-Vulcan

800°C degrade with activity losses up to 44 % (Figure 5.8).

To discuss the implications of our observations in terms of prevailing degradation mechanisms

we note that Pt/C catalyst degradation during potential cycling has been known to proceed via

a number of different pathways. Pt dissolution is most likely linked to Pt oxidation but starts

above 1.0 V29,32,38,99,104 and can therefore be excluded as a major contributor to the present

activity and ECSA losses. Agglomeration, detachment and Ostwald ripening39 are more likely

to cause degradation in the fuel cell lifetime regime below 1 V.26,32,51 In particular,

agglomoration of crystallites could account for the observed ECSA losses for all samples.

However, it would not account for the stark differences in the retention of the Pt mass activity,

with Pt/N-Vulcan 400°C displaying 90% retention. Therefore, other processes must contribute

to this exceptional catalyst stability.

0.01 0.1

100

1000

10000

100000 No. CV

2000

4000

5000

Intensity / a.u.

q / Å-1

0.01 0.1

100

1000

10000

No. CV

1000

3000

5000

Intensity / a.u.

q / Å-1

c d

a b

Pt/Vulcan

Pt/N-Vulcan 400°C

01000 2000 3000 4000 5000

2.6

2.8

3.0

3.2

3.4

3.6

3.8

particle diameter / nm

cycle number

Pt/Vulcan

01000 2000 3000 4000 5000

2.6

2.8

3.0

3.2

3.4

3.6

3.8

particle diameter / nm

cycle number

Pt/N-Vulcan 400°C

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

From HT-DEMS experiments, we found that the nitrogen functionalized N-Vulcan 400°C

showed the lowest carbon corrosion rates due to the highest chemical stability of the nitrogen-

doped carbon lattice. Considering that carbon corrosion is linked to particle migration,

diffusion, and detachment it appears only reasonable that the Pt/N-Vulcan 400°C showed the

highest electrochemical cycling stability. Furthermore, the catalyzing effect of the Pt

nanoparticles on the tendency towards carbon corrosion might be less pronounced in

Pt/N-Vulcan 400°C as a similar phenomenon was observed by Wang et al..47

In our XPS analysis, pyrrolic N functionalities were most prevalent on the surface of the stable

N-Vulcan 400°C catalyst, and their abundance did not change noticeably upon Pt deposition

(Figure 5.3a and Figure 5.4). Thus, the presence of surface pyrrolic N groups appears to have a

statistically significant stabilizing effect on the chemical structure of the carbon support and,

ultimately, on the electrochemical activity of supported Pt nanoparticles.

5. Carbon Heteroatom Modification in Pt/C ORR Catalysts

5.6. Conclusion and Summary

In this study, we prepared a family of nitrogen-functionalized carbon blacks and used them as

supports for Pt nanoparticles and employed these catalyst/support couples as powder

electrocatalysts for the catalytic electroreduction of molecular oxygen (ORR). Followed by

oxidation of the carbon surface, ammonia treatment at elevated temperatures lead to the

incorporation of atomic nitrogen into the chemical structure of the carbons. This was

accompanied by a concomitant reduction in the oxygen amount and concomitant increase in the

surface zeta potential. At modest ammonolysis temperatures of 400 °C, pyrrolic N moieties

were the most abundant surface N species. The resistance of the pyrrolic N-modified carbons

to carbon corrosion and CO2 formation was greatly improved in comparison to all other carbon

supports. Finally, in simulated fuel cell stability tests, platinized versions of the corrosion stable

carbons, such as Pt/N-Vulcan 400°C, again showed superior performance stabilities during

prolonged potential cycling, with only minor ORR activity losses of 10% compared to over

30% of reference catalysts. In contrast, pyridinic N moieties did not offer any beneficial stability

effects. Cross checks of the time evolution of the Pt particle size, using in situ techniques during

cycling, confirmed that differences in the stability of the active Pt particles can be excluded as

the origin for the observed stability difference among the catalysts. The stability benefits are a

direct consequences of the chemical behavior of the modified supports.

We conclude that the controlled introduction of chemical pyrrolic nitrogen into Vulcan carbons,

generated at intermediate ammonolysis temperatures, should be a pathway to fuel cell catalysts

with superior stability.

6. Electrochemical Oxidation of Pt on different Supports

6. On the Anisotropy of Pt Nanoparticles on Carbon- and Oxide-

Support and Their Structural Response to Electrochemical

Oxidation

Identifying the structural response of nanoparticle-support ensembles to the reaction conditions

are essential to determine their structure in the catalytically-active state as well as to unravel

possible degradation pathways. In this work, we investigate the (electronic) structure of carbon-

and oxide-supported Pt nanoparticles by in situ X-ray diffraction, absorption spectroscopy as

well as the Pt dissolution rate by in situ mass spectrometry during electrochemical oxidation.

We prepared ellipsoidal Pt nanoparticles by impregnation of carbon and titanium-based oxide

support as well as spherical Pt nanoparticles which incorporate interstitial oxygen atoms on an

indium-based oxide support by surfactant-assisted synthesis route. During electrochemical

oxidation we show that the oxide-supported Pt nanoparticles resist surface oxide formation and

Pt dissolution both phenomena independent of their morphology. The lattice of smaller Pt

nanoparticles exhibits a size-induced lattice contraction in the as-prepared state but it expands

reversibly during electrochemical oxidation. This expansion is suppressed for the O-containing

Pt nanoparticles with bulk-like relaxed lattice. We could correlate the formation of d-band

vacancies in the metallic Pt with the Pt lattice expansion whereas the formation of PtOx is

accompanied by a loss in structural coherence length. The PtOx formation is strongest for

platelet-like nanoparticles and we explain this with a higher fraction of exposed Pt(100) facets.

Of all investigated nanoparticle-support ensembles, the structural response of RuO2/TiO2-

supported Pt nanoparticles are most promising with respect to their morphological and

structural integrity under electrochemical reaction conditions.

6. Electrochemical Oxidation of Pt on different Supports

This chapter was prepared in collaboration with Arno Bergmann, Elisabeth Hornberger, Jakub

Drnec, Guanxiong Wang, Stefanie Kühl, Daniel J.S. Sandbeck, Serhiy Cherevko, Vijay

Ramani, Karl J.J. Mayrhofer and Peter Strasser

H.S. and A.B. performed the experiments and analyzed the data, E.H. and J.D. assited in the synchrotron

experiments and data evaluation; G.W. and V.R. synthesized the support; S.K. recorded TEM images; D.J.S.S.,

S.C. and K.J.J.M. produced and analyzed the SFC ICP-MS data; H.S., A.B. and P.S. wrote the manuscript; all

authors assisted in discussing and writing the manuscript.

6. Electrochemical Oxidation of Pt on different Supports

6.1. Introduction

In this work, we studied the structural response of Pt nanoparticles on different supports to the

electrochemical oxidation using different in situ X-ray techniques. We tracked the Pt

dissolution rate as well as their voltammetric profiles and monitored the crystallite properties

and the electronic structure during oxidation. Therefore, we aim to understand the differences

in electrochemical oxidation of Pt nanoparticles on different supports aiding the development

of improved durability in Pt fuel cell cathode materials.

6.2. Structure and Morphology

In order to investigate the differences in electrochemical oxidation of Pt nanoparticles on

various kinds of support, a commercial carbon supported Pt electrocatalyst was compared to

two oxide-supported materials; an Indium Tin Oxide (ITO) and RuO2-TiO2 (RTO).

Compositional analysis by ICP-OES revealed a Pt weight loading of 46.1 wt% for Pt/RTO, of

29.9 wt% for Pt/ITO and of 20 wt% for Pt/C (Table A6 in the Appendix section A3). Figure 6.1

shows the diffractograms and morphology by TEM of the three catalysts in the as-prepared

state.

The diffractograms for Pt/RTO, Pt/ITO and Pt/C in Figure 6.1a show broad reflexes

corresponding to fcc Pt phase. In the case of the ITO support, only one other phase additional

to fcc Pt is present that can be clearly assigned to In2O3 (see also Figure A14 for more detailed

information on XRD), meaning that Sn was incorporated into the Indiumoxide-lattice. In the

case of the RTO support, three additional crystalline support phases can be identified: an

anatase TiO2 as well as a rutile-type TiO2 and RuO2 as depicted in Figure A14b. Rietveld

refinement of the HE-XRD pattern in Figure A15 showed that the morphology of the

nanoparticles depends on the different Pt synthesis routes.

6. Electrochemical Oxidation of Pt on different Supports

Figure 6.1 Structure and morphology of Pt nanoparticles supported on RTO (blue), ITO (green) and

carbon (red) as determined by powder X-ray diffraction pattern (a) and transmission electron

microscopy images (b,c,d). The Pt particle sizes as depicted in the histograms in (e,f,g) were determined

from analyzing > 200 Pt particles along their shortest principal axes.

Impregnation of support as applied for RTO and C support leads to more ellipsoidal and

platelet-like nanoparticles, respectively, as shown by the crystallite diameters D that we

extracted from the principal axes of the ellipsoids. For Pt/C, Dx and Dy are ~2.4 nm and Dz is

~1.8 nm whereas for Pt/RTO Dx, Dy, and Dz are ~3.3 nm, 4.5 nm and 2.4 nm, respectively (see

also Table A7). The surfactant-assisted route yields in spherical Pt nanoparticles with a

structural coherence length of ~4.0 nm on the ITO. The morphology of the catalysts is depicted

in Figure 6.1b-d in TEM images. For all three supports, the Pt nanoparticles are homogenously

distributed on the supports (see also Figure A16 for overview images).

Pt/ITO

Pt/RTO

Pt/C

bcd

fe g

20 30 40 50 60 70 80 90

Pt fcc, PDF# 00-004-0802

Pt/RTO

Pt/C

2/ °

Intensity / a.u.

Pt/ITO

2 3 4 5 6

100

120

size / nm

counts

3.8 ± 0.6 nm

3 4 5 6 7

100

120

size / nm

counts

4.6 ± 0.7 nm

1 2 3 4

100

120

140

size / nm

counts

2.2 ± 0.5 nm

6. Electrochemical Oxidation of Pt on different Supports

Figure 6.2 Histograms of mean particle diameter for Pt supported on (a) ITO (green), (b) RTO (blue)

and (c) carbon (red) support. Top row shows the average particle diameter measured along their shortest

particle axis and bottom row along their longest particle axes.

By analyzing the Pt particle size along the shortest particle axes, ITO-supported nanoparticles

exhibit the largest average size with 4.6±0.7 nm, RTO-supported nanoparticles 3.8±0.6 nm and

carbon-supported nanoparticles the smallest size with 2.2±0.5 nm. To further quantify the

anisotropy of the Pt nanoparticles, the TEM particle sizes were also determined along the

longest particle axis. For Pt/RTO and Pt/C particle sizes estimated in this way are around 1.5

and 1.1 nm larger, proofing anisotropic ellipsoidal shape of these nanoparticles (see also

Figure 6.2). In the case of Pt/ITO, the Rietveld refinement estimated spherical nanoparticles

and the TEM nanoparticles size determined along the shortest and longest particle axes are

4.6±0.7 nm and 5.0±0.7 nm, respectively, and thus, in good agreement to each other. The

anisotropy of the Pt nanoparticles suggested by Rietveld refinement was additionally

investigated in a tilting study in TEM. Figure 6.3 shows a series of TEM micrographs for

Pt/RTO and Pt/C recorded under tilting angles between 0 and 30°. The tilting of the sample

plane with respect to the incident electron beam changes the apparent shape of individual Pt

nanoparticles showing their ellipsoidal shape.

2 3 4 5 6 7 8

100

120

counts

3.8 ± 0.6 nm

2 3 4 5 6 7 8

counts

size / nm

5.3 ± 0.8 nm

1 2 3 4 5 6

counts

size / nm

3.3 ± 0.9 nm

1 2 3 4 5 6

100

120

140

counts

2.2 ± 0.5 nm

3 4 5 6 7

100

120

counts

4.6 ± 0.7 nm

Pt/ITO Pt/RTO Pt/C

a b

3 4 5 6 7

counts

size / nm

5.0 ± 0.7 nm

6. Electrochemical Oxidation of Pt on different Supports

Figure 6.3 Morphology of Pt nanoparticles on (a) RTO (blue) and (b) carbon (red) support in a TEM

tilting study from 0-30°.

In addition to the morphology, we identified a correlation between structural coherence length

and the crystal lattice of the Pt nanoparticles. The Pt lattice parameter aPt as determined by

Rietveld refinement (Table A8) decreases from the Pt/ITO with aPt=3.930 Å to Pt/RTO with

aPt=3.920 Å to Pt/C with aPt=3.914 Å. The contraction of the Pt lattice is induced by decreasing

structural coherence length.171,172 We note that the lattice parameter of the nanoparticles can

also be influenced by various adsorbates. Additionally, the most expanded Pt lattice exhibits

interstitial oxygen on the tetrahedral sites (OTd) in Pt lattice as Rietveld refinement revealed an

occupancy of 0.15±0.02. In contrast, the occupancy of OTd for the Pt nanoparticles on RTO and

C was zero within the fit error (see also Table A8). Furthermore, any kind of metal incorporation

in the Pt nanoparticles from the oxide supports was not verifiable via Rietveld refinement.

Thus, we conclude that the impregnation route of RTO and C support lead to smaller ellipsoidal

and platelet-like Pt nanoparticles exhibiting a contracted crystal lattice due to the smaller

domain coherent size, whereas surfactant-assisted synthesis route leads to spherical Pt

nanoparticles with a bulk-like Pt lattice wherein interstitial O are present on the tetrahedral sites.

30°

Pt/C

0° 10°20°30°

Pt/RTO

0° 5° 15°

b30°

6. Electrochemical Oxidation of Pt on different Supports

6.3. Electrochemical Characterization

To identify differences in the electrochemical behavior towards Pt oxidation of the three

catalysts, a series of cyclic voltammograms (CV) in 0.1 M HClO4 were conducted.

Figure 6.4a-c shows CVs from 0.05 V with a stepwise increasing upper potential limit from

0.5-1.4 V after an initial activation step for each catalyst of 100 CVs between 0.05 and 1.0 V.

The typical Pt-Hupd and Pt-O/OH features are clearly visible for all three materials between 0.05

and 0.4 V as well as above 0.6 V, respectively.

Figure 6.4 Electrochemical characterization of Pt nanoparticles on RTO (blue), ITO (green) and carbon

(red) support by the evolution of cyclic voltammogram with increasing upper potential limit in N2-

saturated 0.1 M HClO4 from 0.5-1.4 V (a-c) and from CO stripping experiments (d).

By increasing the upper potential limit, the oxidative currents in the H-desorption region

between 0.05-0.4 V stay rather stable compared to the H-adsorption region of the reductive

scan. The H-desorption region of Pt/ITO shows the strongest growth with increasing upper

potential limit. Additionally, Pt-Oxidation peak(s) arising from around 0.6 V are stable, while

the corresponding reduction peak(s) steadily increase. These findings indicate an increasing

degree of Pt surface oxidation/reduction with higher upper potential limit. The larger currents

in the Hupd region can be explained with an increasing roughening of the nanoparticle surface

and with prolonged reduction of the Pt oxides under these dynamic conditions of the cyclic

voltammograms.

-0.5

0.0

0.5

i / mA cm-2

-0.5

0.0

0.5

i / mA cm-2

0.2 0.4 0.6 0.8 1.0 1.2 1.4

-2.0

-1.0

0.0

1.0

i / mA cm-2

E / V vs. RHE

0.2 0.4 0.6 0.8 1.0 1.2 1.40.0

1.0

2.0

3.0

4.0

i / mA cm-2

Pt/ITO

Pt/RTO

Pt/C

E / V vs. RHE

6. Electrochemical Oxidation of Pt on different Supports

Furthermore, the position of the Pt-O/OH reduction peak differs between the materials and

shifts cathodically with the increasing potential limit for Pt/C, anodically for Pt/ITO and stays

more or less constant for Pt/RTO. The reduction peak of Pt/C is convoluted by two individual

reduction features and this binary character is less pronounced for Pt/RTO. The shape of the

reduction peak does not strongly vary with increasing upper potential limit. The Pt-O/OH

reduction peak of Pt/ITO consists of a single reduction at low upper potential limit but a binary

shape developed with increasing upper potential limit similar to the reductive behavior

determined for Pt on the other two supports. These support-dependent variations in peak

positions indicate that not only the degree of Pt surface oxidation changes but also the surface

of the Pt nanoparticles itself, especially on the ITO, where the CVs alter with increasing

potential. Correlations could be drawn to a study by Chattot et al. where structural defects

originating from microstrain resulted in different CO oxidation profiles and peak positions.173

To further characterize the Pt surface on the different supports, the electrochemical active

surface area was determined from the Hupd-region and from CO stripping experiments.

Figure 6.4d shows the CO oxidation profiles and the carbon-support catalyst shows largest

oxidation peak corresponding to a high CO-ECSA of 67.5 m2 gPt-1. For Pt/ITO the CO-stripping

profile is rather broad and the overall current is lower resulting in an CO-ECSA of 25.6 m2 gPt-1.

CO electrooxidation on Pt/RTO caused a small and narrower peak at lower potentials compared

to Pt/C and Pt/RTO exhibits with 16.7 m2 gPt-1 the lowest CO-ECSA (see Table A9).

6. Electrochemical Oxidation of Pt on different Supports

6.4. In Situ Electrochemical Pt Oxidation

In situ High Energy X-ray Diffraction

To identify the structural response of the Pt nanoparticles on the different supports to

electrochemical oxidation, in situ high-energy diffraction (HE-XRD) patterns were recorded

during step-wise oxidation between 0.6-1.4 V and a reverse scan to 1.0 V to check the

reversibility (see Figure 6.5). Again, all three catalysts were first electrochemically-activated

by 100 CVs as described before.

Figure 6.5 In situ high-energy X-ray diffraction patterns of Pt nanoparticles supported on ITO (green, a),

RTO (blue, b) and carbon (red, c) on stepwise electrochemical oxidation of Pt (potential hold for 10 min

each).

Rietveld refinement of the HE-XRD patterns in Figure 6.5 allowed us to follow several

structural parameters of the crystalline Pt nanoparticles. Figure 6.6 shows the changes of Pt

lattice parameter ΔaPt as well as the Pt scale factor kPt which is directly correlated to the volume

of the crystalline Pt domains. The Pt lattice parameter aPt as shown in Figure 6.6a increases

almost linearly and reversibly with electrode potential above 0.9-1.0 V for the carbon- and

6. Electrochemical Oxidation of Pt on different Supports

RTO-supported Pt as a response to the electrochemical Pt oxidation; The Pt lattice expands to

a similar degree for Pt/RTO and Pt/C with aPt  0.3 pm at 1.4 V. The lattice of the ITO-

supported Pt nanoparticles expands significantly less strongly with aPt  0.07 pm at 1.4 V.

Figure 6.6 Structural response of crystalline Pt nanoparticles supported on ITO (green circles), RTO

(blue triangles) and carbon (red squares) on stepwise electrochemical oxidation (potential hold for 10

min each; reducing step to 1.0 V represented by open symbols) with respect to the change of Pt lattice

parameter (a), normalized change of the Pt scale factor (b) as determined by Rietveld refinement of in

situ high-energy X-ray diffraction pattern. The solid lines are shown to guide the eyes of the reader.

The scale factor kPt as extracted from Rietveld refinement is directly correlated to the absolute

Pt metal diffraction intensity and can thus be treated as a measure of absolute amount of metallic

Pt probed by the X-ray beam. Figure 6.6b shows the evolution of the normalized change in Pt

scale factor which decreased irreversibly above 1.0 V for all three catalysts. The normalized

changes follow the initial TEM particle size as the largest and ITO-supported nanoparticles

showed the least decrease in metallic Pt volume whereas the carbon-supported particles the

highest decrease.

In addition, we also tracked the evolution of the coherence length of the metallic Pt domains.

Despite the considerable fit error for the coherence length of the oxide-supported Pt particles,

we identified an irreversible decrease of the coherence length of the carbon-supported Pt

particles above 1.0 V as shown in Figure 6.7.

a b

6. Electrochemical Oxidation of Pt on different Supports

Figure 6.7 Evolution of the number of unit cells (2R/aPt) of the metallic Pt domains for Pt/C along the

x,y and z-axes. The solid lines are shown to guide the eyes of the reader.

Along the shortest axis of the platelet-like Pt particles (z-axis), the mean coherence length

decreased irreversibly by half a unit cell (2R/aPt) which corresponds to a structural change of

the first layer of Pt atoms on the (100) surface. The mean coherence length, perpendicular to

the shortest axis (x- and y-axis), shows no significant changes beyond the error of the fit with

a mean length of six unit cells, see Figure 6.7. This finding indicates an anisotropic structural

response of the Pt particles to the electrochemical oxidation. Furthermore, for all three catalysts,

no variation with electrode potential for the occupancy with interstitial O on the tetrahedral sites

was identified.

Thus, the oxidizing electrochemical conditions induce a structural response of the metallic Pt

domains of the nanoparticles with respect to their Pt lattice and to the volume of the metallic Pt

domains. The structural transformation of the Pt nanoparticles starts at ~0.9-1.0 V and is

irreversible between 1.0 and 1.4 V as primarily shown by the evolution of the scale factor of

the metallic Pt domains. The decrease in the metallic Pt volume follows the initial Pt coherent

domain size independent of the type of support material. This change in Pt coherent domain

size is induced by the formation of a structurally-different PtOx (shell).

In addition, the electrochemically-oxidizing conditions cause a reversible lattice expansion of

the metallic bulk of Pt nanoparticles on the different supports. This phenomenon contrasts the

correlation between Pt coherent domain size and lattice parameter because the formation of an

oxide shell on the smaller metallic Pt core should let the Pt lattice contract due to a particle size

effect.

6. Electrochemical Oxidation of Pt on different Supports

In situ X-ray Absorption Spectroscopy

In a next step, we recorded in situ X-ray absorption spectra on the Pt L3-edge at selected

electrode potentials to study the influence of the electrode potential and electrochemical

oxidation on the (electronic) structure of the Pt nanoparticles and obtain complementary

information. Figure 6.8 shows the normalized Pt L3-edge absorption spectra for Pt on the three

supports in potentials steps between 0.6 and 1.4 V.

Figure 6.8 Normalized in situ Pt L3–edge X-ray near-edge absorption spectra for Pt nanoparticles

supported on ITO (green, a), RTO (blue, b) and carbon (red, c) for potentials from 0.6 - 1.4 V of

electrochemical oxidation of Pt. The positions of the Pt L3 resonances applied in the fit procedure are

denoted.

The so-called white line (WL) between ~11564 and 11572 eV originates from 2p3/2 to 5d orbital

transitions and thus, its spectral weight is proportional to the total density of unoccupied 5d

states (d-band vacancies) in the Pt atoms. A higher WL area therefore indicates a higher density

of d-band vacancies associated with oxidized Pt. Figure 6.9a shows the integrated WL area for

Pt particle supported on ITO, RTO, and carbon. The WL area of all three catalysts increases

almost linearly with an onset potential between 0.8 and 1.0 V indicating electrochemical

oxidation of Pt and formation of additional d-band vacancies. The absolute change was found

0.5

1.0

1.5

0.5

1.0

1.5

11550 11560 11570 11580

0.5

1.0

1.5

Res3

Res2

Res1

Pt/ITO

1.4 V

1.2 V

1.0 V

0.8 V

0.6 V

Pt/RTO

norm. Fluorescence / -

1.4 V

1.2 V

1.0 V

0.8 V

0.6 V

Res1 Res2

Res3

Pt/C

C 1.4 V

1.2 V

1.0 V

0.8 V

0.6 V

Energy / eV

Res1 Res2

Res3

6. Electrochemical Oxidation of Pt on different Supports

to be strongest for the carbon-supported Pt particles and similarly strong for the oxide-supported

Pt particles (see Figure 6.9a) and qualitatively follows the initial Pt particle size.

Figure 6.9 Structural response of Pt nanoparticles supported on ITO (green circles), RTO (blue

triangles), and carbon (red squares) on stepwise electrochemical oxidation (potential hold for 10 min

each) with respect to the white line area (a) the integrated areas as the resonances 1 (Pt L3-Res 1) at

~11566 eV and Pt L3-Res 2 at ~11570 eV (b) and Pt L3-Res 3 at ~11580 eV (c) as determined by peak

fitting of in situ Pt L3–edge X-ray near-edge absorption spectra. The solid lines are shown to guide the

eye of the reader.

Following Merte et al.174 the Pt L3 XANES spectra are convoluted by at least two spectroscopic

features, the WL of the metallic Pt at ~11566 eV (Res1) and of a Ptn+ at ~11570 eV (Res2), see

also Figure 6.8. We tracked the spectral weight of these features with electrode potential by

fitting the corresponding XANES spectra to follow the increase of the d-band vacancies in the

metallic Pt and the formation of a Ptn+ (Figure 6.9b). Below 0.9 V the (electronic) structure of

the metallic Pt does not strongly vary for all three catalysts because the spectral weight of Res1

does not change. For Pt/C and Pt/RTO even the degree of Ptn+ remained constant as seen in the

spectral weight of Res2. Above ~0.9 V more d-band vacancies in the metallic Pt domains are

generated for all three catalysts as the spectral weight of Res1 increases. For the oxide-

supported Pt nanoparticles the spectral weight of Res1 increases linearly up to 1.4 V whereas

in the case of the carbon-supported Pt nanoparticles it grows only up to 1.1 V. Following the

spectral weight of Res2 in the XANES spectra, we identified that a significant fraction of the

carbon-supported Pt nanoparticles form Ptn+ species increases above 0.9 V. This Ptn+ formation

is strongly suppressed for the oxide-supported Pt nanoparticles as the spectral weight of Res2

remains almost constant. In contrast to carbon-supported Pt particles, no significant change in

the slope of Res1 can be seen. Thus, we conclude that there is a substantial difference in the

oxidation of the carbon- and oxide-supported Pt particles. A significant decrease in the volume

a b c

6. Electrochemical Oxidation of Pt on different Supports

of the metallic domains and in the coherence length of the carbon supported Pt particles above

1.1 V is observed accompanied by the formation of Ptn+ species and thus, a PtOx.

This restructuring can also be followed in the spectral weight of the first local maximum above

the WL at ~11580 eV (Res3), see Figure 6.8. We tracked this feature (see Figure 6.9c) in

addition to WL features typical for metallic Pt and the Ptn+ because it is typically well-

pronounced in the case of large Pt metallic domains such as Pt foils.172 Therefore, we assign

the Res3 feature to be a fingerprint of the metallic domains independent of the corresponding

d-band vacancies. We found that the spectral weight of Res3 is constant up to 1.1 V and

decreases at higher potentials for all three catalysts. The decrease is limited and significantly

stronger in the case of the carbon-supported Pt particles than for the oxide-supported particles

thus, being correlated to the initial Pt particle size. Together with the increasing spectral weight

of oxide-typical Res2, we conclude that the carbon-supported metallic Pt particle restructure

above 1.1 V leading to a strong but limited fraction of PtOx (shell). In contrast, the metallic

domains of the oxide-supported Pt particles restructure significantly less and negligible

spectroscopic contribution of features typical for PtOx can be seen showing a suppressed

structural response to the electrochemical oxidation.

We identified that the Pt lattice parameter and the spectral weight of the Pt L3 Res1 feature

exhibit a similar potential dependence and both are linked changes in the atomic and electronic

structure of the metallic Pt, respectively. In the case of the carbon-supported Pt particles the d-

band vacancy formation in the metallic Pt is superimposed by the restructuring to a PtOx leading

to a decreasing spectral weight above 1.1 V. However, we expect that more d-band vacancies

are formed in the remaining metallic Pt. Two processes can lead to an expansion of the Pt lattice.

On the one hand, the coverage of the Pt surface with O atoms as well as the place-exchange

mechanism occurring above 1.17 V on the Pt(111) can expel the surface atomic layers.100,175

The bulk of the Pt nanoparticles response to these process with an expansion of the lattice. On

the other hand, the formation of d-band vacancies in the metallic Pt decreases the electron

density and thus, weakens the metallic Pt-Pt bond leading to a longer Pt-Pt distance. The

identified reversibility of the lattice expansion between 1.0 and 1.4 V supports rather a

structural response to the electrode potential and the d-band vacancies initiated by a Pt-O

formation at ~0.9 V. A purely oxide-induced structural response is unlikely because the oxygen

coverage and the place-exchange process are not kinetically-limited leading to a potential-

dependent increase. Furthermore, we expect the PtOx shell (or an intermediate stage preceding

6. Electrochemical Oxidation of Pt on different Supports

the electrochemical reduction) yet to be present on the metallic Pt core (Figure 6.4) because the

PtOx get primarily reduced below 1.0 V.

The quantitative differences in the response to the electrochemical oxidation seen in the changes

of the Pt L3 features, e.g. slope of Res1, are caused by the different particle size leading to the

differences in the ECSA. The qualitative differences, however, can be partially explained with

the different morphology of the Pt nanoparticles. The rather platelet-like Pt nanoparticles on

the carbon support potentially exhibit a higher projected area of (100) facets to the electrolyte

than the rather spherical and ellipsoidal Pt particles on the oxide supports. Pt(100) facets are

known to oxidize differently than the (111) facets on which the place-exchange and the

formation of an amorphous 3D α-PtO2 proceeds sequentially at 1.1 V and above 1.3 V,

respectively. On Pt(100) both processes occur simultaneously above 1.1 V.102 In our case the

platelet-like Pt particles exhibit a strong structural response to the electrochemical oxidation

above 1.1 V as seen in the coherence length and as the oxide-typical WL feature grows. These

changes are indicative for the formation of an amorphous 3D α-PtO2 on the Pt particles. The

rather spherical/ellipsoidal Pt nanoparticles on the oxide support, which are thus characterized

by rather convex surfaces, exhibit more Pt sites similar to the Pt(111) or high-index surfaces.

We found that these particles response to the electrochemical oxidation (mainly in the metal

typical WL feature but also in the Pt lattice) but form negligible fraction of PtOx up to 1.4 V

and thus, formation of an amorphous 3D α-PtO2 on the Pt particles is strongly suppressed.

Interestingly, the contracted lattice of the ITO-supported Pt particles between 0.8 and 1.0 V is

accompanied with a reduced spectral weight of Res2. Thus, a lower fraction of oxidized (or

electron-deficient) Pt is accompanied by a contraction of the Pt lattice. This can be explained

with the higher electron density in the Pt d-band strengthening the metallic Pt-Pt bond.

6. Electrochemical Oxidation of Pt on different Supports

In situ Pt dissolution by SFC ICP-MS

The potential-dependent dissolution of Pt on the different supports was followed by in situ SFC

ICP-MS measurements in order to evaluate the stability towards Pt dissolution in comparison

to the above described responses of Pt towards electrochemical oxidation. Therefore, CVs

between 0.05 and 1.5 V after electrochemical activation of 100 CVs were conducted and the

resulting Pt dissolution profiles recorded, see Figure 6.10.

Figure 6.10 Pt dissolution profiles from in situ SFC ICP-MS measurement in gdet Pt per volume

electrolyte in l and electrochemical surface area (ECSA) in m2ECSA. (a) shows the whole electrochemical

measurement (100 CVs of activation with a scanrate 100 mV s-1 and 4 CVs of Pt Oxidation from

0.05-1.5 V with a scanrate 10 mV s-1) and the corresponding dissolution profiles for Pt on carbon (red),

ITO (green) and RTO (blue) support and (b) the four oxidation cycles from 0.05-1.5 V.

0.0

0.5

1.0

1.5

2000 2500 3000

0.0

0.1

0.2

0.3

E / V vs RHE

Pt dissolution /

gdet l-1 m-2

ECSA

time / s

0.0

0.5

1.0

1.5

0500 1000 1500 2000 2500 3000

0.0

0.1

0.2

0.3

E / V vs RHE

Pt/RTO Pt/ITO Pt/C

Pt dissolution /

gdet l-1 m-2

ECSA

time / s

cell contact

program start

6. Electrochemical Oxidation of Pt on different Supports

Dissolution peaks arise for all three materials for the first contact of the sample with the

electrolyte (denoted as cell contact) and at start of the electrochemical protocol (denoted as

program start), see Figure 6.10a. No Pt dissolution was observed during potential cycling in the

activation regime, but strong dissolution peaks arise during electrochemical oxidation up to

1.5 V and reduction to 0.05 V, Figure 6.10b.

Figure 6.11 shows the ECSA-normalized Pt dissolution of the initial anodic potential sweep to

1.5 V mimicking the electrochemical oxidation tracked by the in situ X-ray experiments. The

anodic Pt dissolution has an onset potential at 1.1-1.2 V and continuously increases up to 1.5 V.

The Hupd-ECSA-normalized Pt dissolution is strongest for the Pt/C, lowest for Pt/RTO and

Pt/ITO shows intermediate Pt dissolution rate with the highest background dissolution rate even

below 1.1 V.

Figure 6.11 Evolution of ECSA-normalized Pt dissolution rate for Pt nanoparticles supported on ITO

(green), RTO (blue), and carbon (red) during the initial anodic potential sweep from 0.05-1.5 V with a

scanrate of 10 mV s-1 as determined by SFC ICP-MS measurements.

The onset of Pt dissolution corresponds to the decrease in the Pt scale factor as well as in the

loss of spectral weight for Res1 and Res3. Therefore, the Ptn+ rate might be caused by

dissolution of Ptn+ species during place-exchange between surface-O and Pt and PtOx

formation. Thus, a fraction of the electrochemically formed Ptn+ species is dissolved into the

electrolyte. On the other hand, a constant anodic dissolution of Ptn+ could limit the growth of

an oxide shell around the nanoparticle or lead to the formation of a “stable” PtOx on the surface.

0.6 0.8 1.0 1.2 1.4

ITO RTO Carbon

Pt dissolution / mgdet l-1m-2

ECSA

E / V vs. RHE

6. Electrochemical Oxidation of Pt on different Supports

Additionally, one has to consider the difference between the dynamic in situ SFC ICP-MS

measurements and the quasi-stationary in situ X-ray experiments. Noteworthy, strong Pt

dissolution peaks arise in the reductive scan below 1.0 V with similar dissolution rate for Pt on

C and ITO and lowest dissolution for Pt/RTO, see Figure 6.12.

Figure 6.12 Evolution of ECSA-normalized Pt dissolution rate of electrochemically-oxidized Pt

nanoparticles supported on ITO (green), RTO (blue), and carbon (red) as determined by SFC ICP-MS

measurements during the cathodic potential sweep from 1.5-0.05 V with a scanrate of 10 mV s-1 after

initial anodic potential sweep up to 1.5 V.

Compared to anodic Pt dissolution, electrochemical reduction of the formed PtOx causes a Pt

dissolution which is one order of magnitude higher and thus, representing an important

degradation pathway of Pt-based electrocatalysts (Figure 6.12). The onset of the cathodic Pt

dissolution rate is below 1.0 V and might exhibit a binary profile especially for the Pt/C. It is

known that the PtO2 phase is reduced at lower electrode potentials than a lower PtOx.176,177

These differences furthermore suggest that PtO2 formation is suppressed for the oxide-

supported nanoparticles but more distinct for the carbon-supported.

0.2 0.4 0.6 0.8 1.0 1.2 1.4

0.00

0.05

0.10

0.15

0.20

0.25

ITO RTO Carbon

Pt dissolution / gdet l-1 m-2

ECSA

E / V vs. RHE

6. Electrochemical Oxidation of Pt on different Supports

6.5. Conclusion

In this study, we investigated the response of Pt nanoparticles on different supports towards

electrochemical oxidation in order to establish a more comprehensive understanding of stability

determining properties of fuel cell catalysts.

We found that the applied synthesis route determines the morphology of the Pt nanoparticles.

A surfactant assisted synthesis as used for the ITO support yields ~4 nm spherical nanoparticles

with interstitial O on tetrahedral (Td) sites of the Pt lattice and a bulk-like Pt lattice constant.

Impregnation of the support on the other hand leads to ellipsoidal/platelet-like nanoparticles in

smaller sizes without interstitial O on the tetrahedral sites and a size-induced contraction of the

Pt lattice. The anisotropic shape of Pt on carbon (more platelet-like) and on RTO (more

ellipsoidal) was confirmed from TEM micrographs.

Our in situ HE-XRD investigations revealed that the Pt lattice parameter of the nanoparticles

on carbon and RTO increases reversibly and linearly with potential above 1.0 V while only

small changes are observed for Pt/ITO. This could be due to place-exchange and/or buckling

of surface Pt atoms during the oxidation. From in situ X-ray absorption spectroscopy we show

that the Pt L3 WL intensity increases with potential above 1.0 V and peak fitting reveals that

the formation of Ptn+ species is much more pronounced for Pt/C but follows to a certain extent

the ECSA and particle size. Furthermore, in situ dissolution experiments show anodic Pt

dissolution that follows the trend as observed from the in situ X-ray investigations.

Overall, it can be stated that Pt on RTO shows the lowest tendency towards Ptn+ formation and

Pt dissolution that could point towards a higher total electrochemical stability of an RTO-

supported Pt electrocatalyst.

Overall, it can be stated that Pt on RTO shows the lowest tendency towards Ptn+ formation and

Pt dissolution that could point towards a higher total electrochemical stability of an RTO-

supported Pt electrocatalyst.

100

7. Summary and Outlook

101

7. Summary and Outlook

This work addressed degradation-related phenomena in supported Pt fuel cell catalysts with an

emphasis on the use of novel catalyst support materials. In light of the need for developing more

stable PEMFC cathode catalysts for an accelerated commercialization of fuel cells in the

transportation sector, it is highly important to improve the overall catalyst durability. This can

only be achieved by gaining a detailed knowledge on the catalyst and support properties under

(simulated) working conditions in order to design new and advanced electrocatalysts.

Therefore, a number of different supports were employed to disperse the catalytically active Pt

nanoparticles. A wide range of various analytical methods was applied to investigate the

catalysts and their components on a fundamental level and to draw structure-stability

correlations. The overall goal was to contribute to the understanding of PEMFC catalyst

degradation phenomena. This was engaged by using two different (mixed) metal oxides (ITO

and RTO) as Pt support and investigating the long term stability exemplified by accelerated

stress tests as well as stability determining properties exemplified by the behavior towards Pt

oxidation. Furthermore, a state-of-the-art carbon support was modified by introducing

heteroatoms and its influenece on the structural and morphological stability of the Pt

nanoparticles was analyzed.

7.1. In Situ Stability Study of Pt Supported on Indium Tin Oxide

In the first part of this thesis (chapter 4), indium tin oxide was employed as Pt catalyst support

in order to investigate its potential as carbon replacement in cathode catalysts. It was shown

that small Pt nanoparticles were successfully deposited on the oxide support resulting in a

narrow size distribution. Electrochemical degradation protocols in two potential regimes were

applied to mimic fuel cell operating conditions. In a regular fuel cell operation range (LP-AST),

a discrepancy between the stability of catalytic Pt mass activity and active surface area probed

by hydrogen underpotential deposition pointed towards a surface-induced Pt poisoning effect

responsible for overall stability losses. During simulated start/stop procedures, the Pt/ITO

catalyst proved superior stability when compared to a commercial Pt/C reference.

By applying in situ wide and small angle X-ray scattering, it could be shown that Pt

nanoparticles remained morphologically fully stable under operating conditions whereas the

support suffered from the reducing conditions, as the ITO crystallites partially dissolve and

grow via Ostwald ripening. With the help of in situ dissolution measurements by SFC ICP-MS

and STEM EDX mappings combined with the findings from XP spectroscopy, the loss in

7. Summary and Outlook

102

catalytic activity could be linked to Pt surface modification by most likely redeposited In/Sn

ions from the support. This modification led to a poisoning effect accompanied with a reduced

availability of specific O-adsorption sites on the Pt nanoparticles, Figure 7.1 schematically

illustrates these degradation mechanisms schematically.

Figure 7.1 Schematic illustration of degradation phenomena as found to occur in Pt/ITO electrocatalyst

during two different simulated degradation protocols: low potential AST and high potential AST

(denoted as LP-AST and HP-AST). Reproduced from Ref 52 (Adv. Energy Mat., 2018, 8 (4), 1701663)

During demanding high potential cycling (HP-AST), in contrast, the catalytic stability of oxide-

supported Pt nanoparticles was greatly enhanced compared to the reference carbon-supported

catalyst. This was due to a higher stability of the ITO support, as no ITO particle growth was

observed by in situ HE-XRD. Pt nanoparticles again showed a superior morphological stability

proven by TEM and ASAXS measurements. The small but continuous anodic Pt dissolution

during the HP-AST was proposed to prevent Pt surface poisoning and thus, contributing to

regaining ORR activity and stability.

By the application of complementary analytical techniques in a detailed study, several principle

degradation mechanisms could be excluded. This aided in identifying the cause for activity

deterioration as well as stabilization for Pt/ITO. Therefore, this study contributed to the

essential knowledge about fuel cell relevant degradation phenomena in novel oxide-supported

Pt cathode catalysts.

ITO

Ostwald

Ripening

Pt poisoning

Pt2+

dissolution

Sn2+/4+

In2+/3+

In/Sn dissolution

and redeposition

ITO

LP-AST

0.6 –0.95 V

HP-AST

1 –1.5 V

7. Summary and Outlook

103

7.2. Carbon Heteroatom Modification in Pt/C ORR Catalysts

In chapter 5, a study on the carbon support modification by heteroatom doping and its influence

on the long-term ORR stability of the Pt electrocatalyst was presented. Therefore, a family of

modified carbons from a two-step functionalization approach including acidic oxidation

followed by ammonolysis at elevated temperatures was investigated and used as Pt support.

Elemental analysis showed the introduction of O (up to 12.6 %) and N (up to 2.5 %) into the

carbon matrix yielding in O- and N-enriched carbons. Changes in the BET surface area and the

surface zeta potential upon oxidation and ammonolysis proved the successful modification of

the carbon surface. XP spectroscopy revealed differences in relative amounts of N-moieties on

the carbon surface depending the ammonolysis temperature; at 400 °C pyrrolic-N were found

to be the most abundant surface species while at 800 °C pyridinic-N were present in highest

fractions.

Differences in carbon corrosion behavior upon functionalization were analyzed by high

temperature DEMS in which the carbon aminated at 400 °C showed strongly improved

corrosion resistance with low CO2 production rates compared to the unmodified references and

the oxidized carbon. Furthermore, when used as Pt catalyst support, the influence of the support

heteroatom doping on the electrochemical long-term stability was investigated. By applying

ASTs up to 30k cycles to simulate fuel cell operating conditions to a row of Pt electrocatalysts

on the modified and reference supports, differences in catalytic stability were observed.

Figure 7.2 Schematic illustration of the process of support functionalization by ammonolysis at 400°C

and introduction of surface abundant pyrrolic-N. In the second step, Pt deposition on modified support

and the improved ORR long-term stability is illustrated. Reprinted with permission from Ref 71 (Chem.

Mat., 2018, doi: 10.1021/acs.chemmater.8b03612). Copyright (2018) American Chemical Society.

Pt/N-Vulcan

N-Vulcan

Vulcan

Pyrrolic N High Stability

0.0

0.1

0.2

30k

mass activity

initial 5k 10k

NH3

400°C

7. Summary and Outlook

104

Interestingly, it was found that the excellent stability of Pt/N-Vulcan 400°C was accompanied

with the highest specific ORR activity within the row of these catalysts, see also Figure 7.2.

The application of in situ wide and small angle X-ray scattering aided in excluding certain

degradation mechanisms. Over the first 5k cycles of the ASTs a stable particle and crystallite

sizes for all samples was observed, indicating that changes influencing the electrochemical Pt

stability appear in later stages of the AST. Correlating ORR performances and catalyst

properties, it was proposed that the outstanding stability and specific activity of

Pt/N-Vulcan 400°C was originating from a beneficial effect of the pyrrolic-N surface species

together with an enhanced corrosion resistance.

This study provided new valuable insights into the effect of carbon heteroatom modification

and the origin of the improved long-term ORR stability. Furthermore, it highlights the great

potential of a synthetic pathway for corrosion stabilized Pt supports in the field of PEM fuel

cells.

7.3. Electrochemical Oxidation of Pt on different Supports

The third study of this work (chapter 6) addressed the topic of electrochemical oxidation of Pt

as exemplified by using different supports and following the structural responses by in situ

X-ray methods and morphological changes due to Pt dissolution by in situ ICP-MS. Besides a

commercial carbon-supported Pt catalyst, two oxide-supported materials were used; an indium-

based (ITO) oxide and a mixed ruthenium-titaniumoxide (RTO) support. It could be shown that

the Pt nanoparticles exhibited different particle shapes depending on the applied synthesis route,

where impregnation led to more ellipsoidal Pt particles and surfactant-assisted synthesis

resulted in spherical particles.

Spherical particles were found on ITO support both by TEM analysis as well as by Rietveld

refinement. Additionally, it was shown that a small fraction of interstitial oxygen was

incorporated in the Pt lattice accompanied with an expansion of the Pt lattice. On RTO and

carbon support, the Pt particles exhibited a smaller size and showed ellipsoidal shape without

O incorporated in the Pt lattice.

7. Summary and Outlook

105

Figure 7.3 Schematic illustration of potential-dependent responses of Pt nanoparticles on carbon, RTO

and ITO support on electrochemical oxidation with respect to lattice expansion, formation of PtOx and

Ptn+ dissolution.

During electrochemical oxidation it could be proven by in situ HE-XRD measurements that

structural responses of the metallic Pt domains with respect to their Pt lattice, to the volume and

amount of the metallic Pt domain were induced. The smaller ellipsoidal particles showed initial

size-induced lattice contraction and a reversible lattice expansion upon oxidation to 1.4 V while

this was suppressed in the ITO-supported catalyst with already expanded lattice (see

Figure 7.3). Fitting of the WL-area of the in situ X-ray absorption spectra revealed that the

formation of PtOx species is favored in the carbon-supported nanoparticles due to an increased

tendency towards the formation of Ptn+. For Pt/ITO, a higher electron density in the d-band

strengthening the metallic Pt-Pt bonds resulted in a lower fraction of oxidized Pt. However, in

situ Pt dissolution measurements by SFC ICP-MS showed the highest dissolution rates for the

carbon supported Pt particles and lowest for Pt/RTO pointing towards a higher electrochemical

stability of the RTO-supported Pt nanoparticles.

The novel insights gained in this work can be of high importance for the determination of

catalytic active states in ORR catalysts as well as for the design of new and stable ORR

electrocatalyst. Besides the implementation of simulated degradation protocols as summarized

in the previous sections, the discovery of indications for possible degradation pathways as

exemplified by electrochemical Pt oxidation is another promising way for pioneering improved

catalysts materials.

7. Summary and Outlook

106

7.4. Outlook

With the overall goal towards an increased awareness for sustainability within the society, the

implementation of hydrogen fuel cells in the transportation sector can contribute to a green

energy future. However, the knowledge about a wide range of already known Pt-based catalysts,

that might be considered as fuel cell cathode materials has to be extended with respect to its

availability, activity and durability.

The stability of the electrocatalyst remains a major drawback in fuel cell commercialization.

Therefore, new materials have to be developed based on the knowledge of detailed evaluations

of former generations of catalysts. Thus, the insights gained in the presented studies on Pt/ITO,

Pt/RTO and Pt/N-Vulcan contributing to the overall understanding of cathode catalysts can lead

to the development of improved catalysts.

It was found that the support stability plays a key role in the overall catalyst durability. Hence,

as ITO seems to be suffering from the reducing conditions in ORR catalysis, it might be more

suitable as support for oxygen evolution reaction (OER) catalysts demanding stability at more

anodic potentials. Nevertheless, metal oxide supports represent a promising class of catalysts

supports, and, for example, doping of indium-oxide with other metals, could be a way to

stabilize the oxide and prevent Pt poisoning under cathodic working conditions. In this context,

in situ methods using X-ray radiation are promising tools to investigate more stable metal oxide

supported Pt catalysts. On the other hand, the superior stability of N-modified Vulcan offers

great possibilities for the use as support for highly active Pt-alloy ORR catalysts such as PtNi

or PtCo nanoparticles. The fast and facile two-step modification approach enables the

production of carbon supports with high fractions of stabilizing N-functionalities and the use

of such N-Vulcans as catalyst support for other electrochemical reactions.

Upscaling of established catalyst systems consisting of alternative supports with improved

stability and implementation in MEA setups would represent the ultimate goal. Catalyst

evaluation under MEA conditions would verify the “real” suitability of stabilizing catalysts

supports for an overall improved MEA durability.

107

108

8. References 
109 
 
8. References 
1  Jackson, R. B.;Le Quere, C.;Andrew, R. M.;Canadell, J. G.;Peters, G. P.;Roy, J.;Wu, L., 
Warning signs for stabilizing global CO2 emissions. Environ Res Lett 2017, 12 (11). 
2  Le Quéré, C.;Andrew, R. M.;Friedlingstein, P.;Sitch, S.;Pongratz, J.;Manning, A. 
C.;Korsbakken, J. I.;Peters, G. P.;Canadell, J. G.;Jackson, R. B.;Boden, T. A.;Tans, P. 
P.;Andrews, O. D.;Arora, V. K.;Bakker, D. C. E.;Barbero, L.;Becker, M.;Betts, R. A.;Bopp, 
L.;Chevallier, F.;Chini, L. P.;Ciais, P.;Cosca, C. E.;Cross, J.;Currie, K.;Gasser, T.;Harris, 
I.;Hauck, J.;Haverd, V.;Houghton, R. A.;Hunt, C. W.;Hurtt, G.;Ilyina, T.;Jain, A. K.;Kato, 
E.;Kautz, M.;Keeling, R. F.;Klein Goldewijk, K.;Körtzinger, A.;Landschützer, P.;Lefèvre, 
N.;Lenton, A.;Lienert, S.;Lima, I.;Lombardozzi, D.;Metzl, N.;Millero, F.;Monteiro, P. M. 
S.;Munro, D. R.;Nabel, J. E. M. S.;Nakaoka, S.-i.;Nojiri, Y.;Padin, X. A.;Peregon, A.;Pfeil, 
B.;Pierrot, D.;Poulter, B.;Rehder, G.;Reimer, J.;Rödenbeck, C.;Schwinger, J.;Séférian, 
R.;Skjelvan, I.;Stocker, B. D.;Tian, H.;Tilbrook, B.;Tubiello, F. N.;van der Laan-Luijkx, I. 
T.;van der Werf, G. R.;van Heuven, S.;Viovy, N.;Vuichard, N.;Walker, A. P.;Watson, A. 
J.;Wiltshire, A. J.;Zaehle, S.;Zhu, D., Global Carbon Budget 2017. Earth System Science Data 
2018, 10 (1), 405-448. 
3  https://www.esrl.noaa.gov/gmd/ccgg/trends/ (accessed 30.07.2018). 
4  Peters, G. P.;Le Quere, C.;Andrew, R. M.;Canadell, J. G.;Friedlingstein, P.;Ilyina, T.;Jackson, 
R. B.;Joos, F.;Korsbakken, J. I.;McKinley, G. A.;Sitch, S.;Tans, P., Towards real-time 
verification of CO2 emissions. Nat Clim Change 2017, 7 (12), 848-850. 
5  https://www.energy-charts.de/energy_pie.htm?year=2017 (accessed 01.08.2018). 
6  https://www.energy-charts.de/ren_share_de.htm?source=ren-share&period=monthly&year=all 
(accessed 24.09.2018). 
7  Pellow, M. A.;Emmott, C. J. M.;Barnhart, C. J.;Benson, S. M., Hydrogen or batteries for grid 
storage? A net energy analysis. Energy & Environmental Science 2015, 8 (7), 1938-1952. 
8  Grove, W. R., XXIV. On voltaic series and the combination of gases by platinum. The 
London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science 1839, 14 
(86-87), 127-130. 
9  Grove, W. R., On a Gaseous Voltaic Battery. Philos. Mag. 2012, 92 (31), 3753-3756. 
10  Larminie, J.;Dicks, A., Fuel Cell Systems Explained. John Wiley & Sons Ltd: West Sussex, 
2003. 
11  https://www.fueleconomy.gov/feg/fcv_sbs.shtml (accessed 09.08.2018). 
12  https://www.nrel.gov/docs/fy11osti/49342-1.pdf (accessed 18.09.2018). 
13  http://ec.europa.eu/regional_policy/en/projects/germany/hydrogen-fuel-cells-drive-clean-
quiet-buses-in-cologne-and-amsterdam (accessed 18.09.2018). 
14  https://www.alstom.com/press-releases-news/2018/9/world-premiere-alstoms-hydrogen-
trains-enter-passenger-service-lower (accessed 17.09.2018). 
15  Strasser, P., Fuel Cells. In Chemical Energy Storage, R. Schloegl, Ed. De Gruyter: 2012; pp 
163-184. 
16  https://www.hydrogen.energy.gov/pdfs/17007_fuel_cell_system_cost_2017.pdf (accessed 
01.08.2018). 
17  Stephens, I. E. L.;Bondarenko, A. S.;Gronbjerg, U.;Rossmeisl, J.;Chorkendorff, I., 
Understanding the electrocatalysis of oxygen reduction on platinum and its alloys. Energy & 
Environmental Science 2012, 5 (5), 6744-6762. 
18  Rossmeisl, J.;Karlberg, G. S.;Jaramillo, T.;Norskov, J. K., Steady state oxygen reduction and 
cyclic voltammetry. Faraday Discussions 2008, 140, 337-346. 
19  Stamenkovic, V.;Mun, B. S.;Mayrhofer, K. J. J.;Ross, P. N.;Markovic, N. M.;Rossmeisl, 
J.;Greeley, J.;Nørskov, J. K., Changing the Activity of Electrocatalysts for Oxygen Reduction 
by Tuning the Surface Electronic Structure. Angew. Chem. 2006, 118 (18), 2963-2967. 
20  Stamenkovic, V. R.;Fowler, B.;Mun, B. S.;Wang, G.;Ross, P. N.;Lucas, C. A.;Markovic, N. 
M., Improved Oxygen Reduction Activity on Pt3Ni(111) via Increased Surface Site 
Availability. Science 2007. 

8. References

110

21 Stamenkovic, V. R.;Mun, B. S.;Arenz, M.;Mayrhofer, K. J. J.;Lucas, C. A.;Wang, G. F.;Ross,

P. N.;Markovic, N. M., Trends in electrocatalysis on extended and nanoscale Pt-bimetallic

alloy surfaces. Nat. Mater. 2007, 6 (3), 241-247.

22 Gan, L.;Heggen, M.;Rudi, S.;Strasser, P., Core-Shell Compositional Fine Structures of

Dealloyed PtxNi1-x Nanoparticles and Their Impact on Oxygen Reduction Catalysis. Nano

Lett. 2012, 12 (10), 5423-5430.

23 Gan, L.;Cui, C. H.;Rudi, S.;Strasser, P., Core-Shell and Nanoporous Particle Architectures

and Their Effect on the Activity and Stability of Pt ORR Electrocatalysts. Top. Catal. 2014,

57 (1-4), 236-244.

24 Bligaard, T.;Norskov, J. K., Ligand effects in heterogeneous catalysis and electrochemistry.

Electrochimica Acta 2007, 52 (18), 5512-5516.

25 https://www.energy.gov/sites/prod/files/2017/11/f46/FCTT_Roadmap_Nov_2017_FINAL.pdf

(accessed 02.08.2018).

26 Meier, J. C.;Galeano, C.;Katsounaros, I.;Witte, J.;Bongard, H. J.;Topalov, A. A.;Baldizzone,

C.;Mezzavilla, S.;Schuth, F.;Mayrhofer, K. J. J., Design criteria for stable Pt/C fuel cell

catalysts. Beilstein J. Nanotechnol. 2014, 5, 44-67.

27 Keeley, G. P.;Cherevko, S.;Mayrhofer, K. J. J., The Stability Challenge on the Pathway to

Low and Ultra-Low Platinum Loading for Oxygen Reduction in Fuel Cells.

ChemElectroChem 2016, 3 (1), 51-54.

28 Rice, C. A.;Urchaga, P.;Pistono, A. O.;McFerrin, B. W.;McComb, B. T.;Hu, J., Platinum

Dissolution in Fuel Cell Electrodes: Enhanced Degradation from Surface Area Assessment in

Automotive Accelerated Stress Tests. Journal of The Electrochemical Society 2015, 162 (10),

F1175-F1180.

29 Topalov, A. A.;Katsounaros, I.;Auinger, M.;Cherevko, S.;Meier, J. C.;Klemm, S.

O.;Mayrhofer, K. J., Dissolution of platinum: limits for the deployment of electrochemical

energy conversion? Angew. Chem., Int. Ed. Engl. 2012, 51 (50), 12613-12615.

30 Xing, L. Y.;Hossain, M.;Tian, M.;Beauchemin, D.;Adjemian, K.;Jerkiewicz, G., Platinum

Electro-dissolution in Acidic Media upon Potential Cycling. Electrocatalysis 2014, 5 (1), 96-

112.

31 Matsumoto, M.;Miyazaki, T.;Imai, H., Oxygen-Enhanced Dissolution of Platinum in Acidic

Electrochemical Environments. J Phys Chem C 2011, 115 (22), 11163-11169.

32 Cherevko, S.;Kulyk, N.;Mayrhofer, K. J. J., Durability of platinum-based fuel cell

electrocatalysts: Dissolution of bulk and nanoscale platinum. Nano Energy 2016, 29, 275-298.

33 Borup, R.;Meyers, J.;Pivovar, B.;Kim, Y. S.;Mukundan, R.;Garland, N.;Myers, D.;Wilson,

M.;Garzon, F.;Wood, D.;Zelenay, P.;More, K.;Stroh, K.;Zawodzinski, T.;Boncella,

J.;McGrath, J. E.;Inaba, M.;Miyatake, K.;Hori, M.;Ota, K.;Ogumi, Z.;Miyata, S.;Nishikata,

A.;Siroma, Z.;Uchimoto, Y.;Yasuda, K.;Kimijima, K.-i.;Iwashita, N., Scientific Aspects of

Polymer Electrolyte Fuel Cell Durability and Degradation. Chem. Rev. 2007, 107 (10), 3904-

3951.

34 Pizzutilo, E.;Geiger, S.;Grote, J. P.;Mingers, A.;Mayrhofer, K. J. J.;Arenz, M.;Cherevko, S.,

On the Need of Improved Accelerated Degradation Protocols (ADPs): Examination of

Platinum Dissolution and Carbon Corrosion in Half-Cell Tests. J. Electrochem. Soc. 2016,

163 (14), F1510-F1514.

35 Holby, E. F.;Sheng, W.;Shao-Horn, Y.;Morgan, D., Pt nanoparticle stability in PEM fuel cells:

influence of particle size distribution and crossover hydrogen. Energy & Environmental

Science 2009, 2 (8), 865.

36 Darling, R. M.;Meyers, J. P., Kinetic model of platinum dissolution in PEMFCs. Journal of

the Electrochemical Society 2003, 150 (11), A1523-A1527.

37 Yano, H.;Watanabe, M.;Iiyama, A.;Uchida, H., Particle-size effect of Pt cathode catalysts on

durability in fuel cells. Nano Energy 2016.

38 Cherevko, S.;Keeley, G. P.;Geiger, S.;Zeradjanin, A. R.;Hodnik, N.;Kulyk, N.;Mayrhofer, K.

J. J., Dissolution of Platinum in the Operational Range of Fuel Cells. Chemelectrochem 2015,

2 (10), 1471-1478.

39 Ostwald, W., Studien über die Bildung und Umwandlung fester Körper. Z. Phys. Chem. 1897,

22, 289-330.

8. References

111

40 Shao-Horn, Y.;Ferreira, P. J.;O’, G. J. l.;Morgan, D.;Gasteiger, H.;Makharia, R., Coarsening

of Pt Nanoparticles in Proton Exchange Membrane Fuel Cells. ECS Transactions 2006, 1 (8),

185-195.

41 Speder, J.;Zana, A.;Spanos, I.;Kirkensgaard, J. J. K.;Mortensen, K.;Hanzlik, M.;Arenz, M.,

Comparative degradation study of carbon supported proton exchange membrane fuel cell

electrocatalysts – The influence of the platinum to carbon ratio on the degradation rate. J.

Power Sources 2014, 261, 14-22.

42 Riese, A.;Banham, D.;Ye, S.;Sun, X., Accelerated Stress Testing by Rotating Disk Electrode

for Carbon Corrosion in Fuel Cell Catalyst Supports. Journal of the Electrochemical Society

2015, 162 (7), F783-F788.

43 Castanheira, L.;Dubau, L.;Mermoux, M.;Berthome, G.;Caque, N.;Rossinot, E.;Chatenet,

M.;Maillard, F., Carbon Corrosion in Proton-Exchange Membrane Fuel Cells: From Model

Experiments to Real-Life Operation in Membrane Electrode Assemblies. ACS Cat. 2014, 4

(7), 2258-2267.

44 Shao-Horn, Y.;Sheng, W. C.;Chen, S.;Ferreira, P. J.;Holby, E. F.;Morgan, D., Instability of

supported platinum nanoparticles in low-temperature fuel cells. Top. Cat. 2007, 46 (3-4), 285-

305.

45 Antolini, E., Carbon supports for low-temperature fuel cell catalysts. Applied Catalysis B:

Environmental 2009, 88 (1-2), 1-24.

46 Willsau, J.;Heitbaum, J., The Influence of Pt-Activation on the Corrosion of Carbon in Gas-

Diffusion Electrodes - a Dems Study. Journal of Electroanalytical Chemistry 1984, 161 (1),

93-101.

47 Wang, X.;Li, W.;Chen, Z.;Waje, M.;Yan, Y., Durability investigation of carbon nanotube as

catalyst support for proton exchange membrane fuel cell. J. Power Sources 2006, 158 (1),

154-159.

48 Schulenburg, H.;Schwanitz, B.;Linse, N.;Scherer, G. G.;Wokaun, A.;Krbanjevic,

J.;Grothausmann, R.;Manke, I., 3D Imaging of Catalyst Support Corrosion in Polymer

Electrolyte Fuel Cells. The Journal of Physical Chemistry C 2011, 115 (29), 14236-14243.

49 Mayrhofer, K. J. J.;Meier, J. C.;Ashton, S. J.;Wiberg, G. K. H.;Kraus, F.;Hanzlik, M.;Arenz,

M., Fuel cell catalyst degradation on the nanoscale. Electrochemistry Communications 2008,

10 (8), 1144-1147.

50 Hasche, F.;Oezaslan, M.;Strasser, P., Activity, stability and degradation of multi walled

carbon nanotube (MWCNT) supported Pt fuel cell electrocatalysts. Physical Chemistry

Chemical Physics 2010, 12 (46), 15251-15258.

51 Shao-Horn, Y.;Sheng, W. C.;Chen, S.;Ferreira, P. J.;Holby, E. F.;Morgan, D., Instability of

supported platinum nanoparticles in low-temperature fuel cells. Top Catal 2007, 46 (3-4),

285-305.

52 Schmies, H.;Bergmann, A.;Drnec, J.;Wang, G.;Teschner, D.;Kühl, S.;Sandbeck, D. J.

S.;Cherevko, S.;Gocyla, M.;Shviro, M.;Heggen, M.;Ramani, V.;Dunin-Borkowski, R.

E.;Mayrhofer, K. J. J.;Strasser, P., Unravelling Degradation Pathways of Oxide-Supported Pt

Fuel Cell Nanocatalysts under In Situ Operating Conditions. Adv. Energy Mater. 2018, 8 (4),

1701663.

53 Shao, Y.;Liu, J.;Wang, Y.;Lin, Y., Novel catalyst support materials for PEMfuelcells: current

status and future prospects. J. Mater. Chem. 2009, 19 (1), 46-59.

54 Zheng, N. F.;Stucky, G. D., A general synthetic strategy for oxide-supported metal

nanoparticle catalysts. Journal of the American Chemical Society 2006, 128 (44), 14278-

14280.

55 Huang, S.-Y.;Ganesan, P.;Park, S.;Popov, B. N., Development of Titanium Dioxide-supported

Platinum catalyst with ultrahigh stability for PEMFC applications. JACS 2009.

56 Lewera, A.;Timperman, L.;Roguska, A.;Alonso-Vante, N., Metal–Support Interactions

between Nanosized Pt and Metal Oxides (WO3and TiO2) Studied Using X-ray Photoelectron

Spectroscopy. The Journal of Physical Chemistry C 2011, 115 (41), 20153-20159.

57 Ho, V. T.;Pan, C. J.;Rick, J.;Su, W. N.;Hwang, B. J., Nanostructured Ti(0.7)Mo(0.3)O2

support enhances electron transfer to Pt: high-performance catalyst for oxygen reduction

reaction. J Am Chem Soc 2011, 133 (30), 11716-11724.

8. References

112

58 Kumar, A.;Ramani, V., Ta0.3Ti0.7O2 Electrocatalyst Supports Exhibit Exceptional

Electrochemical Stability. Journal of the Electrochemical Society 2013, 160 (11), F1207-

F1215.

59 Kumar, A.;Ramani, V., Strong Metal–Support Interactions Enhance the Activity and

Durability of Platinum Supported on Tantalum-Modified Titanium Dioxide Electrocatalysts.

ACS Catalysis 2014, 4 (5), 1516-1525.

60 Kumar, A.;Ramani, V. K., RuO2–SiO2 mixed oxides as corrosion-resistant catalyst supports

for polymer electrolyte fuel cells. Applied Catalysis B: Environmental 2013, 138-139, 43-50.

61 Lo, C.-P.;Ramani, V., SiO2–RuO2: A Stable Electrocatalyst Support. ACS Applied Materials

& Interfaces 2012, 4 (11), 6109-6116.

62 Lo, C.-P.;Wang, G.;Kumar, A.;Ramani, V., TiO2–RuO2 electrocatalyst supports exhibit

exceptional electrochemical stability. Applied Catalysis B: Environmental 2013, 140-141,

133-140.

63 Parrondo, J.;Han, T.;Niangar, E.;Wang, C.;Dale, N.;Adjemian, K.;Ramani, V., Platinum

supported on titanium-ruthenium oxide is a remarkably stable electrocatayst for hydrogen fuel

cell vehicles. Proc Natl Acad Sci U S A 2014, 111 (1), 45-50.

64 Dou, M. L.;Hou, M.;Liang, D.;Lu, W. T.;Shao, Z. G.;Yi, B. L., SnO2 nanocluster supported

Pt catalyst with high stability for proton exchange membrane fuel cells. Electrochimica Acta

2013, 92, 468-473.

65 Tsukatsune, T.;Takabatake, Y.;Noda, Z.;Daio, T.;Zaitsu, A.;Lyth, S. M.;Hayashi, A.;Sasaki,

K., Platinum-Decorated Tin Oxide and Niobium-Doped Tin Oxide PEFC Electrocatalysts:

Oxygen Reduction Reaction Activity. Journal of the Electrochemical Society 2014, 161 (12),

F1208-F1213.

66 Cognard, G.;Ozouf, G.;Beauger, C.;Berthomé, G.;Riassetto, D.;Dubau, L.;Chattot,

R.;Chatenet, M.;Maillard, F., Benefits and limitations of Pt nanoparticles supported on highly

porous antimony-doped tin dioxide aerogel as alternative cathode material for proton-

exchange membrane fuel cells. Applied Catalysis B: Environmental 2017, 201, 381-390.

67 Liu, Y.;Mustain, W. E., High stability, high activity Pt/ITO oxygen reduction electrocatalysts.

J Am Chem Soc 2013, 135 (2), 530-533.

68 Zhao, S.;Wangstrom, A. E.;Liu, Y.;Rigdon, W. A.;Mustain, W. E., Stability and Activity of

Pt/ITO Electrocatalyst for Oxygen Reduction Reaction in Alkaline Media. Electrochim. Acta

2015, 157, 175-182.

69 Liu, Y.;Mustain, W. E., Stability limitations for Pt/Sn-In2O3 and Pt/In-SnO2 in acidic

electrochemical systems. Electrochimica Acta 2014, 115, 116-125.

70 Wang, G.;Bhattacharyya, K.;Parrondo, J.;Ramani, V., X-ray photoelectron spectroscopy study

of the degradation of Pt/ITO electrocatalyst in an operating polymer electrolyte fuel cell.

Chemical Engineering Science 2016, 154, 81-89.

71 Schmies, H.;Hornberger, E.;Anke, B.;Jurzinsky, T.;Nong, H. N.;Dionigi, F.;Kühl, S.;Drnec,

J.;Lerch, M.;Cremers, C.;Strasser, P., The Impact of Carbon Support Functionalization on the

Electrochemical Stability of Pt Fuel Cell Catalysts. Chemistry of Materials 2018.

72 Sun, Y. Y.;Sinev, I.;Ju, W.;Bergmann, A.;Dresp, S.;Kuhl, S.;Spori, C.;Schmies, H.;Wang,

H.;Bernsmeier, D.;Paul, B.;Schmack, R.;Kraehnert, R.;Cuenya, B. R.;Strasser, P., Efficient

Electrochemical Hydrogen Peroxide Production from Molecular Oxygen on Nitrogen-Doped

Mesoporous Carbon Catalysts. ACS Catalysis 2018, 8 (4), 2844-2856.

73 Mezzavilla, S.;Baldizzone, C.;Mayrhofer, K. J.;Schuth, F., General Method for the Synthesis

of Hollow Mesoporous Carbon Spheres with Tunable Textural Properties. ACS Appl Mater

Interfaces 2015, 7 (23), 12914-12922.

74 Galeano, C.;Meier, J. C.;Soorholtz, M.;Bongard, H.;Baldizzone, C.;Mayrhofer, K. J.

J.;Schüth, F., Nitrogen-Doped Hollow Carbon Spheres as a Support for Platinum-Based

Electrocatalysts. ACS Catalysis 2014, 4 (11), 3856-3868.

75 Xu, B.;Zheng, D.;Jia, M.;Cao, G.;Yang, Y., Nitrogen-doped porous carbon simply prepared

by pyrolyzing a nitrogen-containing organic salt for supercapacitors. Electrochim. Acta 2013,

98, 176-182.

8. References

113

76 Hulicova-Jurcakova, D.;Kodama, M.;Shiraishi, S.;Hatori, H.;Zhu, Z. H.;Lu, G. Q., Nitrogen-

Enriched Nonporous Carbon Electrodes with Extraordinary Supercapacitance. Adv. Funct.

Mater. 2009, 19 (11), 1800-1809.

77 Vikkisk, M.;Kruusenberg, I.;Joost, U.;Shulga, E.;Tammeveski, K., Electrocatalysis of oxygen

reduction on nitrogen-containing multi-walled carbon nanotube modified glassy carbon

electrodes. Electrochim. Acta 2013, 87, 709-716.

78 Zhao, A.;Masa, J.;Muhler, M.;Schuhmann, W.;Xia, W., N-doped carbon synthesized from N-

containing polymers as metal-free catalysts for the oxygen reduction under alkaline

conditions. Electrochim. Acta 2013, 98, 139-145.

79 Ni, B.;Ouyang, C.;Xu, X. B.;Zhuang, J.;Wang, X., Modifying Commercial Carbon with Trace

Amounts of ZIF to Prepare Derivatives with Superior ORR Activities. Adv. Mater. 2017, 29

(27), 1701354.

80 Toupin, M.;Belanger, D., Spontaneous functionalization of carbon black by reaction with 4-

nitrophenyldiazonium cations. Langmuir 2008, 24 (5), 1910-1917.

81 Lakhi, K. S.;Park, D. H.;Al-Bahily, K.;Cha, W.;Viswanathan, B.;Choy, J. H.;Vinu, A.,

Mesoporous carbon nitrides: synthesis, functionalization, and applications Chem. Soc. Rev.

2017, 46 (2), 560-560.

82 Punetha, V. D.;Rana, S.;Yoo, H. J.;Chaurasia, A.;McLeskey, J. T.;Ramasamy, M. S.;Sahoo,

N. G.;Cho, J. W., Functionalization of carbon nanomaterials for advanced polymer

nanocomposites: A comparison study between CNT and graphene. Prog. Polym. Sci. 2017,

67, 1-47.

83 Jaouen, F.;Dodelet, J. P., Non-noble electrocatalysts for O-2 reduction: How does heat

treatment affect their activity and structure? Part I. Model for carbon black gasification by

NH3: Parametric calibration and electrochemical validation. J. Phys. Chem. C 2007, 111 (16),

5963-5970.

84 Ortega, K. F.;Arrigo, R.;Frank, B.;Schlogl, R.;Trunschke, A., Acid-Base Properties of N-

Doped Carbon Nanotubes: A Combined Temperature-Programmed Desorption, X-ray

Photoelectron Spectroscopy, and 2-Propanol Reaction Investigation. Chem. Mat. 2016, 28

(19), 6826-6839.

85 Arrigo, R.;Havecker, M.;Wrabetz, S.;Blume, R.;Lerch, M.;McGregor, J.;Parrott, E. P.

J.;Zeitler, J. A.;Gladden, L. F.;Knop-Gericke, A.;Schlogl, R.;Su, D. S., Tuning the Acid/Base

Properties of Nanocarbons by Functionalization via Amination. J. Am. Chem. Soc. 2010, 132

(28), 9616-9630.

86 Arrigo, R.;Havecker, M.;Schlogl, R.;Su, D. S., Dynamic surface rearrangement and thermal

stability of nitrogen functional groups on carbon nanotubes. Chem. Commun. 2008, (40),

4891-4893.

87 Li, X. L.;Wang, H. L.;Robinson, J. T.;Sanchez, H.;Diankov, G.;Dai, H. J., Simultaneous

Nitrogen Doping and Reduction of Graphene Oxide. J. Am. Chem. Soc 2009, 131 (43), 15939-

15944.

88 Lepró, X.;Terrés, E.;Vega-Cantú, Y.;Rodríguez-Macías, F. J.;Muramatsu, H.;Kim, Y.

A.;Hayahsi, T.;Endo, M.;Torres R, M.;Terrones, M., Efficient anchorage of Pt clusters on N-

doped carbon nanotubes and their catalytic activity. Chem. Phys. Lett. 2008, 463 (1-3), 124-

129.

89 Du, H. Y.;Wang, C. H.;Hsu, H. C.;Chang, S. T.;Chen, U. S.;Yen, S. C.;Chen, L. C.;Shih, H.

C.;Chen, K. H., Controlled platinum nanoparticles uniformly dispersed on nitrogen-doped

carbon nanotubes for methanol oxidation. Diamond Relat. Mater. 2008, 17 (4-5), 535-541.

90 Chen, S.;Qi, P.;Chen, J.;Yuan, Y., Platinum nanoparticles supported on N-doped carbon

nanotubes for the selective oxidation of glycerol to glyceric acid in a base-free aqueous

solution. RSC Adv. 2015, 5 (40), 31566-31574.

91 Guo, L.;Jiang, W.-J.;Zhang, Y.;Hu, J.-S.;Wei, Z.-D.;Wan, L.-J., Embedding Pt Nanocrystals

in N-Doped Porous Carbon/Carbon Nanotubes toward Highly Stable Electrocatalysts for the

Oxygen Reduction Reaction. ACS Cat. 2015, 5 (5), 2903-2909.

92 Hasché, F.;Fellinger, T.-P.;Oezaslan, M.;Paraknowitsch, J. P.;Antonietti, M.;Strasser, P.,

Mesoporous Nitrogen Doped Carbon Supported Platinum PEM Fuel Cell Electrocatalyst

Made From Ionic Liquids. ChemCatChem 2012, 4 (4), 479-483.

8. References

114

93 Tuaev, X.;Paraknowitsch, J. P.;Illgen, R.;Thomas, A.;Strasser, P., Nitrogen-doped coatings on

carbon nanotubes and their stabilizing effect on Pt nanoparticles. Phys. Chem. Chem. Phys.

2012, 14 (18), 6444-6447.

94 Shi, Q. R.;Zhu, C. Z.;Engelhard, M. H.;Du, D.;Lin, Y. H., Highly uniform distribution of Pt

nanoparticles on N-doped hollow carbon spheres with enhanced durability for oxygen

reduction reaction. RSC Adv. 2017, 7 (11), 6303-6308.

95 Arrigo, R.;Schuster, M. E.;Xie, Z. L.;Yi, Y. M.;Wowsnick, G.;Sun, L. L.;Hermann, K.

E.;Friedrich, M.;Kast, P.;Havecker, M.;Knop-Gericke, A.;Schlogl, R., Nature of the N-Pd

Interaction in Nitrogen-Doped Carbon Nanotube Catalysts. ACS Cat. 2015, 5 (5), 2740-2753.

96 Shanmugam, S.;Sanetuntikul, J.;Momma, T.;Osaka, T., Enhanced Oxygen Reduction

Activities of Pt Supported on Nitrogen-Doped Carbon Nanocapsules. Electrochim. Acta 2014,

137, 41-48.

97 Peter, B.;Melke, J.;Muench, F.;Ensinger, W.;Roth, C., Stable platinum nanostructures on

nitrogen-doped carbon obtained by high-temperature synthesis for use in PEMFC. J. Appl.

Electrochem. 2014, 44 (5), 573-580.

98 Orfanidi, A.;Madkikar, P.;El-Sayed, H. A.;Harzer, G. S.;Kratky, T.;Gasteiger, H. A., The Key

to High Performance Low Pt Loaded Electrodes. J. Electrochem. Soc. 2017, 164 (4), F418-

F426.

99 Ruge, M.;Drnec, J.;Rahn, B.;Reikowski, F.;Harrington, D. A.;Carla, F.;Felici, R.;Stettner,

J.;Magnussen, O. M., Electrochemical Oxidation of Smooth and Nanoscale Rough Pt(111):

An In Situ Surface X-ray Scattering Study. J. Electrochem. Soc. 2017, 164 (9), H608-H614.

100 Drnec, J.;Ruge, M.;Reikowski, F.;Rahn, B.;Carlà, F.;Felici, R.;Stettner, J.;Magnussen, O.

M.;Harrington, D. A., Initial stages of Pt(111) electrooxidation: dynamic and structural studies

by surface X-ray diffraction. Electrochimica Acta 2017, 224, 220-227.

101 Gomez-Marin, A. M.;Feliu, J. M., Oxide growth dynamics at Pt(111) in absence of specific

adsorption: A mechanistic study. Electrochimica Acta 2013, 104, 367-377.

102 Huang, Y.-F.;Kooyman, P. J.;Koper, M. T. M., Intermediate stages of electrochemical

oxidation of single-crystalline platinum revealed by in situ Raman spectroscopy. Nat Commun

2016, 7, 12440.

103 Friebel, D.;Miller, D. J.;O'Grady, C. P.;Anniyev, T.;Bargar, J.;Bergmann, U.;Ogasawara,

H.;Wikfeldt, K. T.;Pettersson, L. G.;Nilsson, A., In situ X-ray probing reveals fingerprints of

surface platinum oxide. Phys Chem Chem Phys 2011, 13 (1), 262-266.

104 Merte, L. R.;Behafarid, F.;Miller, D. J.;Friebel, D.;Cho, S.;Mbuga, F.;Sokaras, D.;Alonso-

Mori, R.;Weng, T.-C.;Nordlund, D.;Nilsson, A.;Roldan Cuenya, B., Electrochemical

Oxidation of Size-Selected Pt Nanoparticles Studied Using in Situ High-Energy-Resolution X-

ray Absorption Spectroscopy. ACS Catal. 2012, 2 (11), 2371-2376.

105 Sasaki, K.;Marinkovic, N.;Isaacs, H. S.;Adzic, R. R., Synchrotron-Based In Situ

Characterization of Carbon-Supported Platinum and Platinum Mono layer Electrocatalysts.

ACS Catalysis 2016, 6 (1), 69-76.

106 Imai, H.;Izumi, K.;Matsumoto, M.;Kubo, Y.;Kato, K.;Imai, Y., In Situ and Real-Time

Monitoring of Oxide Growth in a Few Monolayers at Surfaces of Platinum Nanoparticles in

Aqueous Media. Journal of the American Chemical Society 2009, 131 (17), 6293-6300.

107 Nagasawa, K.;Takao, S.;Higashi, K.;Nagamatsu, S.;Samjeske, G.;Imaizumi, Y.;Sekizawa,

O.;Yamamoto, T.;Uruga, T.;Iwasawa, Y., Performance and durability of Pt/C cathode

catalysts with different kinds of carbons for polymer electrolyte fuel cells characterized by

electrochemical and in situ XAFS techniques. Phys Chem Chem Phys 2014, 16 (21), 10075-

10087.

108 Hatanaka, M.;Takahashi, N.;Takahashi, N.;Tanabe, T.;Nagai, Y.;Suda, A.;Shinjoh, H.,

Reversible changes in the Pt oxidation state and nanostructure on a ceria-based supported Pt.

Journal of Catalysis 2009, 266 (2), 182-190.

109 Rudi, S.;Gan, L.;Cui, C. H.;Gliech, M.;Strasser, P., Electrochemical Dealloying of Bimetallic

ORR Nanoparticle Catalysts at Constant Electrode Potentials. Journal of the Electrochemical

Society 2015, 162 (4), F403-F409.

110 Thust, A.;Barthel, J.;Tillmann, K., FEI Titan 80-300 TEM. Journal of large-scale research

facilities JLSRF 2016, 2.

8. References

115

111 Kovács, A.;Schierholz, R.;Tillmann, K., FEI Titan G2 80-200 CREWLEY. Journal of large-

scale research facilities JLSRF 2016, 2.

112 Niether, C.;Rau, M. S.;Cremers, C.;Jones, D. J.;Pinkwart, K.;Tübke, J., Development of a

novel experimental DEMS set-up for electrocatalyst characterization under working

conditions of high temperature polymer electrolyte fuel cells. Journal of Electroanalytical

Chemistry 2015, 747, 97-103.

113 Baltruschat, H., Differential electrochemical mass spectrometry. J. Am. Soc. Mass Spectrom.

2004, 15 (12), 1693-1706.

114 Biesinger, M. C.;Payne, B. P.;Grosvenor, A. P.;Lau, L. W. M.;Gerson, A. R.;Smart, R. S.,

Resolving surface chemical states in XPS analysis of first row transition metals, oxides and

hydroxides: Cr, Mn, Fe, Co and Ni. Appl. Surf. Sci. 2011, 257 (7), 2717-2730.

115 Yeh, J. J.;Lindau, I., Atomic Subshell Photoionization Cross-Sections and Asymmetry

Parameters - 1 Less-Than-or-Equal-to Z Less-Than-or-Equal-to 103. At. Data Nucl. Data

Tables 1985, 32 (1), 1-155.

116 https://www.energy.gov/sites/prod/files/2016/06/f32/fcto_myrdd_fuel_cells.pdf (accessed

17.07.2018).

117 Rudi, S.;Cui, C.;Gan, L.;Strasser, P., Comparative Study of the Electrocatalytically Active

Surface Areas (ECSAs) of Pt Alloy Nanoparticles Evaluated by Hupd and CO-stripping

voltammetry. Electrocatalysis 2014, 5 (4), 408-418.

118 Cherevko, S.;Zeradjanin, A. R.;Topalov, A. A.;Keeley, G. P.;Mayrhofer, K. J. J., Effect of

Temperature on Gold Dissolution in Acidic Media. Journal of the Electrochemical Society

2014, 161 (9), H501-H507.

119 Schuppert, A. K.;Topalov, A. A.;Katsounaros, I.;Klemm, S. O.;Mayrhofer, K. J. J., A

Scanning Flow Cell System for Fully Automated Screening of Electrocatalyst Materials.

Journal of the Electrochemical Society 2012, 159 (11), F670-F675.

120 Rietveld, H. M., A Profile Refinement Method for Nuclear and Magnetic Structures. Journal

of Applied Crystallography 1969, 2, 65.

121 Rietveld, H. M., Line profiles of neutron powder-diffraction peaks for structure refinement.

Acta Crystallography 1967, 22, 151.

122 Albinati, A.;Willis, B. T. M., The Rietveld method. International Tables for Crystallography

Vol. C 2006, 710-712.

123 Ahtee, M.;Unonius, L.;Nurmela, M.;Suortti, P., A Voigtian as Profile Shape Function in

Rietveld Refinement. Journal of Applied Crystallography 1984, 17 (Oct), 352-357.

124 Young, R. A.;Wiles, D. B., Profile Shape Functions in Rietveld Refinements. Journal of

Applied Crystallography 1982, 15 (Aug), 430-438.

125 Young, R. A.;Wiles, D. B., Profile Shape Functions in Rietveld Refinement. Journal of

Applied Crystallography 1982, 15, 430-438.

126 Will, G., Powder Diffraction: The Rietveld Method and the Two Stage Method to Determine

and Refine Crystal Structures from Powder Diffraction Data. Springer Science & Business

Media: 2006.

127 Ectors, D.;Goetz-Neunhoeffer, F.;Neubauer, J., Domain size anisotropy in the double-Voigt

approach: an extended model. J Appl Crystallogr 2015, 48, 1998-2001.

128 Ectors, D.;Goetz-Neunhoeffer, F.;Neubauer, J., A generalized geometric approach to

anisotropic peak broadening due to domain morphology. J Appl Crystallogr 2015, 48, 189-

194.

129 Li, T.;Senesi, A. J.;Lee, B., Small Angle X-ray Scattering for Nanoparticle Research. Chem

Rev 2016.

130 Brumberger, H.;Hagrman, D.;Goodisman, J.;Finkelstein, K. D., In situanomalous small-angle

X-ray scattering from metal particles in supported-metal catalysts. II. Results. Journal of

Applied Crystallography 2005, 38 (2), 324-332.

131 Schnablegger, H.;Singh, Y., The SAXS Guide - Getting acquainted with the principles. Anton

Paar GmbH: 2011.

132 Tuaev, X.;Strasser, P., Small angle X-ray scattering (SAXS) techniques for polymer

electrolyte membrane fuel cell characterization. In Polymer Electrolyte Membrane and Direct

Methanol Fuel Cell Technology - In Situ Characterization Techniques for Low Temperature

8. References

116

Fuel Cells, Christoph Hartnig & Christina Roth, Eds. Woodhead Publishing Limited: 2012;

Vol. 2, pp 87-119.

133 Guinier, A.;Fournet, G.;Walker, C. B.;Yudowitch, K. L., Small-Angle Scattering of X-Rays

Wiley: New York, 1955.

134 Brumberger, H.;Hagrman, D.;Goodisman, J.;Finkelstein, K. D., In situanomalous small-angle

X-ray scattering from metal particles in supported-metal catalysts. I. Theory. Journal of

Applied Crystallography 2005, 38 (1), 147-151.

135 Haubold, H. G.;Vad, T.;Waldofner, N.;Bonnemann, H., From Pt molecules to nanoparticles:

in-situ (anomalous) small-angle X-ray scattering studies. Journal of Applied Crystallography

2003, 36 (1), 617-620.

136 Cuesta, A.;Couto, A.;Rincón, A.;Pérez, M. C.;López-Cudero, A.;Gutiérrez, C., Potential

dependence of the saturation CO coverage of Pt electrodes: The origin of the pre-peak in CO-

stripping voltammograms. Part 3: Pt(poly). Journal of Electroanalytical Chemistry 2006, 586

(2), 184-195.

137 López-Cudero, A.;Cuesta, Á.;Gutiérrez, C., Potential dependence of the saturation CO

coverage of Pt electrodes: The origin of the pre-peak in CO-stripping voltammograms. Part 2:

Pt(100). Journal of Electroanalytical Chemistry 2006, 586 (2), 204-216.

138 Maillard, F.;Eikerling, M.;Cherstiouk, O. V.;Schreier, S.;Savinova, E.;Stimming, U., Size

effects on reactivity of Pt nanoparticles in CO monolayer oxidation: The role of surface

mobility. Faraday Discussions 2004, 125, 357-377.

139 van der Vliet, D. F.;Wang, C.;Li, D. G.;Paulikas, A. P.;Greeley, J.;Rankin, R. B.;Strmcnik,

D.;Tripkovic, D.;Markovic, N. M.;Stamenkovic, V. R., Unique Electrochemical Adsorption

Properties of Pt-Skin Surfaces. Angew Chem Int Edit 2012, 51 (13), 3139-3142.

140 Borchert, H.;Shevehenko, E. V.;Robert, A.;Mekis, I.;Kornowski, A.;Grubel, G.;Weller, H.,

Determination of nanocrystal sizes: A comparison of TEM, SAXS, and XRD studies of highly

monodisperse COPt3 particles. Langmuir 2005, 21 (5), 1931-1936.

141 Cherevko, S.;Zeradjanin, A. R.;Keeley, G. P.;Mayrhofer, K. J. J., A Comparative Study on

Gold and Platinum Dissolution in Acidic and Alkaline Media. Journal of the Electrochemical

Society 2014, 161 (12), H822-H830.

142 Wang, L.;Dionigi, F.;Nguyen, N. T.;Kirchgeorg, R.;Gliech, M.;Grigorescu, S.;Strasser,

P.;Schmuki, P., Tantalum Nitride Nanorod Arrays: Introducing Ni–Fe Layered Double

Hydroxides as a Cocatalyst Strongly Stabilizing Photoanodes in Water Splitting. Chemistry of

Materials 2015, 27 (7), 2360-2366.

143 Diebold, U., The surface science of titanium dioxide. Surface Science Reports 2003, 48 (5-8),

53-229.

144 Fu, Q.;Wagner, T., Interaction of nanostructured metal overlayers with oxide surfaces. Surface

Science Reports 2007, 62 (11), 431-498.

145 Fu, Q.;Wagner, T.;Olliges, S.;Carstanjen, H. D., Metal-oxide interfacial reactions:

Encapsulation of Pd on TiO2 (110). J Phys Chem B 2005, 109 (2), 944-951.

146 O'Shea, V. A.;Galvan, M. C.;Prats, A. E.;Campos-Martin, J. M.;Fierro, J. L., Direct evidence

of the SMSI decoration effect: the case of Co/TiO2 catalyst. Chem Commun (Camb) 2011, 47

(25), 7131-7133.

147 Yoshida, M.;Takanabe, K.;Maeda, K.;Ishikawa, A.;Kubota, J.;Sakata, Y.;Ikezawa, Y.;Domen,

K., Role and Function of Noble-Metal/Cr-Layer Core/Shell Structure Cocatalysts for

Photocatalytic Overall Water Splitting Studied by Model Electrodes. J Phys Chem C 2009,

113 (23), 10151-10157.

148 Bonanni, S.;At-Mansour, K.;Brune, H.;Harbich, W., Overcoming the Strong Metal−Support

Interaction State: CO Oxidation on TiO2(110)-Supported Pt Nanoclusters. ACS Catalysis

2011, 1 (4), 385-389.

149 Willinger, M. G.;Zhang, W.;Bondarchuk, O.;Shaikhutdinov, S.;Freund, H. J.;Schlogl, R., A

Case of Strong Metal-Support Interactions: Combining Advanced Microscopy and Model

Systems to Elucidate the Atomic Structure of Interfaces. Angew Chem Int Edit 2014, 53 (23),

5998-6001.

8. References

117

150 Shi, X. Y.;Zhang, W.;Zhang, C.;Zheng, W. T.;Chen, H.;Qi, J. G., Real-space observation of

strong metal-support interaction: state-of-the-art and what's the next. J Microsc-Oxford 2016,

262 (3), 203-215.

151 Spoeri, C.;Kwan, J. T.;Bonakdarpour, A.;Wilkinson, D.;Strasser, P., The Stability Challenges

of Oxygen Evolving Electrocatalysts: Towards a Common Fundamental Understanding and

Mitigation of Catalyst Degradation. Angew Chem Int Ed Engl 2016.

152 Micoud, F.;Maillard, F.;Bonnefont, A.;Job, N.;Chatenet, M., The role of the support in

CO(ads) monolayer electrooxidation on Pt nanoparticles: Pt/WO(x)vs. Pt/C. Phys Chem Chem

Phys 2010, 12 (5), 1182-1193.

153 Dionigi, F.;Vesborg, P. C. K.;Pedersen, T.;Hansen, O.;Dahl, S.;Xiong, A.;Maeda, K.;Domen,

K.;Chorkendorff, I., Suppression of the water splitting back reaction on GaN:ZnO

photocatalysts loaded with core/shell cocatalysts, investigated using a μ-reactor. Journal of

Catalysis 2012, 292, 26-31.

154 Rizo, R.;Lázaro, M. J.;Pastor, E.;Koper, M. T. M., Ethanol Oxidation on Sn-modified Pt

Single-Crystal Electrodes: New Mechanistic Insights from On-line Electrochemical Mass

Spectrometry. ChemElectroChem 2016.

155 Kakinuma, K.;Chino, Y.;Senoo, Y.;Uchida, M.;Kamino, T.;Uchida, H.;Deki, S.;Watanabe,

M., Characterization of Pt catalysts on Nb-doped and Sb-doped SnO2–δ support materials

with aggregated structure by rotating disk electrode and fuel cell measurements.

Electrochimica Acta 2013, 110, 316-324.

156 Nakada, M.;Ishihara, A.;Mitsushima, S.;Kamiya, N.;Ota, K., Effect of tin oxides on oxide

formation and reduction of platinum particles. Electrochem Solid St 2007, 10 (1), F1-F4.

157 Santos, M. C.;Bulhoes, L. O. S., The underpotential deposition of Sn on Pt in acid media.

Cyclic voltarnmetric and electrochemical quartz crystal microbalance studies. Electrochimica

Acta 2003, 48 (18), 2607-2614.

158 Oh, H.-S.;Nong, H. N.;Reier, T.;Bergmann, A.;Gliech, M.;Ferreira de Araújo, J.;Willinger,

E.;Schlögl, R.;Teschner, D.;Strasser, P., Electrochemical Catalyst–Support Effects and Their

Stabilizing Role for IrOxNanoparticle Catalysts during the Oxygen Evolution Reaction.

Journal of the American Chemical Society 2016, 138 (38), 12552-12563.

159 Vayssilov, G. N.;Lykhach, Y.;Migani, A.;Staudt, T.;Petrova, G. P.;Tsud, N.;Skala, T.;Bruix,

A.;Illas, F.;Prince, K. C.;Matolin, V.;Neyman, K. M.;Libuda, J., Support nanostructure boosts

oxygen transfer to catalytically active platinum nanoparticles. Nat. Mater. 2011, 10 (4), 310-

315.

160 Pietrzak, R.;Wachowska, H., The influence of oxidation with HNO3 on the surface

composition of high-sulphur coals: XPS study. Fuel Process. Technol. 2006, 87 (11), 1021-

1029.

161 Batich, C. D.;Donald, D. S., X-ray photoelectron spectroscopy of nitroso compounds: relative

ionicity of the closed and open forms. J. Am. Chem. Soc. 1984, 106 (10), 2758-2761.

162 Yang, C.-M.;Kim, D. Y.;Lee, Y. H., Formation of Densely Packed Single-Walled Carbon

Nanotube Assembly. Chem. Mater. 2005, 17, 6422-6429.

163 Beermann, V.;Gocyla, M.;Kuhl, S.;Padgett, E.;Schmies, H.;Goerlin, M.;Erini, N.;Shviro,

M.;Heggen, M.;Dunin-Borkowski, R. E.;Muller, D. A.;Strasser, P., Tuning the

Electrocatalytic Oxygen Reduction Reaction Activity and Stability of Shape-Controlled Pt-Ni

Nanoparticles by Thermal Annealing - Elucidating the Surface Atomic Structural and

Compositional Changes. J. Am. Chem. Soc. 2017, 139 (46), 16536-16547.

164 Aran-Ais, R. M.;Dionigi, F.;Merzdorf, T.;Gocyla, M.;Heggen, M.;Dunin-Borkowski, R.

E.;Gliech, M.;Solla-Gullon, J.;Herrero, E.;Feliu, J. M.;Strasser, P., Elemental Anisotropic

Growth and Atomic-Scale Structure of Shape-Controlled Octahedral Pt-Ni-Co Alloy

Nanocatalysts. Nano Lett. 2015, 15 (11), 7473-7480.

165 Tuaev, X.;Rudi, S.;Strasser, P., The impact of the morphology of the carbon support on the

activity and stability of nanoparticle fuel cell catalysts. Catal Sci Technol 2016, 6 (23), 8276-

8288.

166 Asset, T.;Job, N.;Busby, Y.;Crisci, A.;Martin, V.;Stergiopoulos, V.;Bonnaud, C.;Serov,

A.;Atanassov, P.;Chattot, R.;Dubau, L.;Maillard, F., Porous Hollow PtNi/C Electrocatalysts:

8. References

118

Carbon Support Considerations To Meet Performance and Stability Requirements. ACS Cat.

2018, 8 (2), 893-903.

167 Hasche, F.;Oezaslan, M.;Strasser, P., Activity, stability and degradation of multi walled

carbon nanotube (MWCNT) supported Pt fuel cell electrocatalysts. Phys. Chem. Chem. Phys.

2010, 12 (46), 15251-15258.

168 Antolini, E., Graphene as a new carbon support for low-temperature fuel cell catalysts. Appl.

Catal., B 2012, 123-124, 52-68.

169 Li, D.;Wang, C.;Strmcnik, D. S.;Tripkovic, D. V.;Sun, X.;Kang, Y.;Chi, M.;Snyder, J. D.;van

der Vliet, D.;Tsai, Y.;Stamenkovic, V. R.;Sun, S.;Markovic, N. M., Functional links between

Pt single crystal morphology and nanoparticles with different size and shape: the oxygen

reduction reaction case. Energy Environ. Sci. 2014, 7 (12), 4061-4069.

170 Beermann, V.;Gocyla, M.;Willinger, E.;Rudi, S.;Heggen, M.;Dunin-Borkowski, R.

E.;Willinger, M.-G.;Strasser, P., Rh-Doped Pt–Ni Octahedral Nanoparticles: Understanding

the Correlation between Elemental Distribution, Oxygen Reduction Reaction, and Shape

Stability. Nano Lett. 2016, 16 (3), 1719-1725.

171 Diehm, P. M.;Agoston, P.;Albe, K., Size-dependent lattice expansion in nanoparticles: reality

or anomaly? ChemPhysChem 2012, 13 (10), 2443-2454.

172 Lei, Y.;Jelic, J.;Nitsche, L. C.;Meyer, R.;Miller, J., Effect of Particle Size and Adsorbates on

the L3, L2 and L1 X-ray Absorption Near Edge Structure of Supported Pt Nanoparticles. Top.

Catal. 2011, 54 (5-7), 334-348.

173 Dubau, L.;Nelayah, J.;Moldovan, S.;Ersen, O.;Bordet, P.;Drnec, J.;Asset, T.;Chattot,

R.;Maillard, F., Defects do Catalysis: CO Monolayer Oxidation and Oxygen Reduction

Reaction on Hollow PtNi/C Nanoparticles. Acs Catalysis 2016, 6 (7), 4673-4684.

174 Merte, L. R.;Behafarid, F.;Miller, D. J.;Friebel, D.;Cho, S.;Mbuga, F.;Sokaras, D.;Alonso-

Mori, R.;Weng, T.-C.;Nordlund, D.;Nilsson, A.;Roldan Cuenya, B., Electrochemical

Oxidation of Size-Selected Pt Nanoparticles Studied Using in Situ High-Energy-Resolution X-

ray Absorption Spectroscopy. ACS Catalysis 2012, 2 (11), 2371-2376.

175 Ruge, M.;Drnec, J.;Rahn, B.;Reikowski, F.;Harrington, D. A.;Carla, F.;Felici, R.;Stettner,

J.;Magnussen, O. M., Electrochemical Oxidation of Smooth and Nanoscale Rough Pt(111):

An In Situ Surface X-ray Scattering Study. Journal of the Electrochemical Society 2017, 164

(9), H608-H614.

176 Angerste.H;Conway, B. E.;Sharp, W. B. A., Real Condition of Electrochemically Oxidized

Platinum Surfaces .1. Resolution of Component Processes. J Electroanal Chem 1973, 43 (1),

9-36.

177 Conway, B. E., Electrochemical Oxide Film Formation at Noble-Metals as a Surface-

Chemical Process. Prog Surf Sci 1995, 49 (4), 331-452.

Appendix

119

Appendix

A1 Supporting Information to Chapter 452

Figure A1 TEM images of as prepared Pt nanoparticles on the ITO support which was used for the HP-AST

experiments with a weight loading of 29.9 wt% (a,b) and the corresponding histogram showing particle size

distribution (c). Histograms were obtained by measuring the diameter of at least 200 particles with errors obtained

from standard deviation of mean particle diameter.

Figure A2 Cyclic voltammograms of Pt/C reference catalyst (weight loading 20 wt% Pt on carbon) before and

after potential cycling in lower (a) and higher (c) potential region. CVs were recorded in nitrogen saturated

electrolyte from 0.05-1 V with a scan rate of 100 mV s-1. Figure b and d are showing LSVs of the particular states

LSVs were recorded in oxygen saturated electrolyte from 0.05-1 V with a scan rate of 5 mV·s-1and 1600 rpm. All

electrode potentials have been corrected for iR drop.

3 4 5 6 7 8 9 10

100 5.3±0.6 nm

counts

particle size / nm

a b c

0.0 0.2 0.4 0.6 0.8 1.0

-6

-5

-4

-3

-2

-1

0 initial

after LP-AST

j / mA cm-2

E / V vs RHE

0.0 0.2 0.4 0.6 0.8 1.0

-2

-1

initial

after LP-AST

j / mA cm-2

E / V vs RHE

0.0 0.2 0.4 0.6 0.8 1.0

-3

-2

-1

initial

after HP-AST

j / mA cm-2

E / V vs RHE

0.0 0.2 0.4 0.6 0.8 1.0

-6

-5

-4

-3

-2

-1

0 initial

after HP-AST

E / V vs RHE

j / mA cm-2

jm-14%

Hupd ECSA

-11%

Hupd ECSA

+9%

jm-38%

a b

c d

Appendix

120

Table A1 Comparison of ECSAs based on the integration in the Hupd and the CO oxidation potential range. CO-

ECSAs were determined by integrating the CO oxidation peak area from the first cycle of the CO stripping

experiment after subtraction of the second cycle representing the bare CO-free surface. Hupd-ECSAs were

determined by subtracting the first from the second cycle of the CO stripping experiment.

ECSA / m2 gPt-1

LP-AST

HP-AST

before

after

before

after

25.6

18.4

28.9

27.4

Hupd

18.6

17.0

19.3

26.1

CO/Hupd

1.38

1.04

1.50

1.05

Figure A3 Electrochemical CO stripping experiments for Pt/ITO electrocatalyst. For the two different stability

tests in low (a) and high (b) potential regime three different points in the characterization protocol were chosen for

CO stripping: initial, after CV+LSV and after 5k CV+LSV.

Table A2 Comparison of jm and ECSA before and after LP- and HP-AST for Pt/C reference catalyst. jm was

determined at 0.9 V. ECSAs were determined by integrating the H-desorption and adsorption area between

0.05-0.4 V and subtracting the capacitive current.

LP-AST

HP-AST

initial

after

initial

after

jm/ A mgPt

-1

0.18

0.15

0.16

0.10

ECSA / m2 gPt

-1

61.1

54.0

61.4

67.0

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.5

1.0

1.5

j / mA cm-2

E / V vs RHE

initial

after CV+LSV

after 5kCV+LSV

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.5

1.0

1.5

2.0

2.5 initial

after CV+LSV

after 5k CV+LSV

j / mA cm-2

E / V vs RHE

a b

Appendix

121

Figure A4 In situ HE XRD measurements depicted as the evolution of diffraction patterns from the initial state to

the end of the electrochemical cycling for the LP-AST (a) and the HP-AST (b). The inlet in (a) is showing the Pt

(200) diffraction peak superimposed by an In2O3 peak (denoted with dashed vertical lines).

Figure A5 In situ scanning flow cell ICP-MS measurements. Depicted are the Sn, In and Pt dissolution rates and

the applied electrochemical protocols from the bottom to the top for LP-AST. The respective dissolution rates in

detected metal per volume electrolyte (μgdet l-1) are plotted against the time. A Pt/C reference sample was measured

and therefore, the Pt dissolution rate was also normalized to the Pt mass loading on the working electrode (WE)

(μgdet l-1μgWE-1). The electrochemical protocol was conducted as follows: Beginning with 100 CVs (activation

regime) from 0.05-1 V, followed by potential cycling in the LP regime (0.6-0.95 V, 40 CVs, 100 mV·s-1) and

followed by another 3 cycles from 0.05-1 V, all CVs were recorded with a scan rate of 100mV s-1. The first contacts

between catalyst and electrolyte (cell contact) are denoted with arrows.

a b

3 4 5 6 7

initial

5000 CV

2/ °

Intensity / a.u.

HP-AST

3 4 5 6 7

4.6 4.8

Pt (200)

Intensity / a.u.

In2O3

initial

5000 CV

2/ °

Intensity / a.u.

LP-AST

0500 1000 1500 2000

0.0

0.5

1.0

0500 1000 1500 2000

0.0

0.1

0.2

E / V vs RHE

Pt/C

Pt/ITO

Pt / gdet l-1gWE

-1

cell contact

In / gdet l-1

Sn / gdet l-1

time / sec

Appendix

122

Figure A6 In situ scanning flow cell ICP-MS measurements. Depicted are the Sn, In and Pt dissolution rates and

the applied electrochemical protocols from the bottom to the top for HP-AST. The respective dissolution rates in

detected metal per volume electrolyte (μgdet l-1) are plotted against the time. A Pt/C reference sample was measured

and therefore, the Pt dissolution rate was also normalized to the Pt mass loading on the working electrode (WE)

(μgdet l-1μgWE-1). The electrochemical protocol was conducted as follows: Beginning with 100 CVs (activation

regime) from 0.05-1 V, followed by potential cycling in the HP regime (1.0-1.5 V, 200 CVs, 500 mV·s-1) and

followed by another 3 cycles from 0.05-1 V, all CVs were recorded with a scan rate of 100mV s-1. The first contacts

between catalyst and electrolyte (cell contact) are denoted with arrows.

Table A3 Results from in situ SFC ICP-MS of the integration over time of the measurement of the peaks arising

from Pt dissolution in Pt/ITO and Pt/C reference and In and Sn from ITO support. Pt dissolution in μg is normalized

to Pt mass loading on the working electrode in μgPt and In and Sn dissolution in μg is normalized to In+Sn mass

loading on the working electrode in μg(In+Sn).Potential dissolution peaks are arising from cell contact, program start

and distinct dissolution peaks during the electrochemical cycling (1st and 2nd peak).Additionally, dissolution values

are listed each without (wo/) and with (w/) contribution of metal dissolution at cell contact.

LP-AST

HP-AST

Dissolved Metal

Pt in

Pt/ITO

Pt in

Pt/C

Pt in

Pt/ITO

Pt in

Pt/C

Unit

μg·μgPt

μg·μg(In+Sn)

μg·μgPt

μg·μg(In+Sn)

Cell Contact

3.22E-4

6.13E-4

5.55E-3

3.52E-4

2.33E-4

1.23E-3

1.04E-2

1.19E-3

Program Start

7.47E-5

4.66E-5

7.22E-6

7.74E-5

2.80E-4

1st Peak

3.38E-4

1.07E-3

3.00E-4

2nd Peak

4.32E-5

1.12E-4

Total wo/ Contact

5.96E-5

3.36E-4

9.61E-3

3.41E-3

7.37E-4

1.68E-3

9.12E-3

3.96E-3

Total w/ Contact

3.80E-4

9.49E-4

1.51E-2

3.77E-3

9.69E-4

2.91E-3

1.95E-2

5.15E-3

Total wo/

Contact %

0.01

0.03

0.96

0.34

0.07

0.17

0.91

0.40

Total w/

Contact %

0.04

0.09

1.51

0.38

0.10

0.29

1.95

0.52

-500 0 500 1000 1500 2000 2500

0.0

0.5

1.0

1.5

-500 0 500 1000 1500 2000 2500

0.0

0.2

E / V vs RHE

Pt / gdet l-1 gWE

-1

Pt/C

Pt/ITO

cell contact

In / gdet l-1

Sn / gdet l-1

time / sec

Appendix

123

Figure A7 Evolution of Pt lattice constant as extracted from Rietveld Refinement from in situ HE-XRD

measurements over the cycle number for (a) LP-AST and (b) HP-AST.

Appendix

124

A2 Supporting Information to Chapter 571

Table A4 Elemental bulk and surface composition in at% of unmodified Vulcan, O-Vulcan and N-Vulcan 400°C

and 800°C from elemental analysis, hot gas extraction and XPS.

[at%]

bulk from

elemental

analysis

bulk from

elemental

analysis

bulk from

elemental

analysis

bulk from

hot gas

extraction

bulk from

hot gas

extraction

surface

from

XPS

surface

from

XPS

surface

from

XPS

(substrate)

surface

from XPS

Vulcan

0.00

98.10

0.00

5.84

92.16

2.00

O-

Vulcan

0.19

85.58

0.38

0.268

12.6

0.73

12.61

86.15

0.51

N-

Vulcan

400°C

2.53

93.26

0.13

2.5

2.9

3.86

9.35

83.35

3.44

N-

Vulcan

800°C

1.53

95.08

0.02

1.8

0.85

2.26

7.3

87.38

3.07

Table A5 Pt weight loading (wt%) from ICP-OES for Pt/Vulcan, Pt/O-Vulcan, Pt/N-Vulcan 400°C and Pt/N-

Vulcan 800°C.

Pt loading

[wt%]

Pt/Vulcan

20.1

Pt/O-Vulcan

23.1

Pt/N-Vulcan 400°C

19.8

Pt/N-Vulcan 800°C

20.9

Appendix

125

Figure A8 XPS C 1s spectra and individual peak deconvolution for unmodified Vulcan (a), O-Vulcan (b), N-

Vulcan 400° (c), N-Vulcan 800°C (d) and for Pt/N-Vulcan 400°C (e).

296 294 292 290 288 286 284 282 280

0.0

0.1

0.2

0.3

0.4

O-Vulcan

Normalized intensity / a. u.

Binding energy / eV

experimental data

fit envelope

bkg

C-C, C-H

C-OH, C-O-

C=O

O-C=O

sat1

sat2

sat3

C 1s

296 294 292 290 288 286 284 282 280

0.0

0.1

0.2

0.3

0.4

N-Vulcan 800°C

Normalized intensity / a. u.

Binding energy / eV

experimental data

fit envelope

bkg

C-C, C-H

C-OH, C-O-

C=O

O-C=O

sat1

sat2

sat3

C 1s

296 294 292 290 288 286 284 282 280

0.0

0.1

0.2

0.3

0.4

Normalized intensity / a. u.

Binding energy / eV

experimental data

fit envelope

bkg

C-C, C-H

C-OH, C-O-

C=O

O-C=O

sat1

sat2

sat3

C 1s

N-Vulcan 400°C

296 294 292 290 288 286 284 282 280

0.0

0.1

0.2

0.3

0.4

Vulcan

Normalized intensity / a. u.

Binding energy / eV

experimental data

fit envelope

bkg

C-C, C-H

C-OH, C-O-

C=O

O-C=O

sat1

sat2

sat3

C 1s

296 294 292 290 288 286 284 282 280

0.0

0.1

0.2

0.3

0.4

Pt/N-Vulcan 400°C

Normalized intensity / a. u.

Binding energy / eV

experimental data

fit envelope

bkg

C-C, C-H

C-OH, C-O-

C=O

O-C=O

sat1

sat2

sat3

C 1s

Appendix

126

Figure A9 TEM images for Pt on unmodified Vulcan (a), O-Vulcan (b), N-Vulcan 400°C (c), and N-Vulcan 800°C

(d).

Pt/N-Vulcan 800°C

Pt/Vulcan

Pt/O-Vulcan

Pt/N-Vulcan 400°C

Appendix

127

Figure A10 In situ high energy X-ray diffraction patterns over 5k cycle of the AST for Pt/Vulcan (a), Pt/O-Vulcan

(b), Pt/N-Vulcan 400°C (c) and Pt/N-Vulcan 800°C (d).

4 6 8 10 12 14 16 18 20

***

No. CV

2500

5000

Intensity / a.u.

2 @ 78 keV / °

4 6 8 10 12 14 16 18 20

No. CV

2500

5000

Intensity / a.u.

2 @ 68.5 keV / °

4 6 8 10 12 14 16 18 20

No. CV

2500

5000

Intensity / a.u.

2 @ 68.5 keV / °

4 6 8 10 12 14 16 18 20

carbon from substrate

No. CV

2500

5000

Intensity / a.u.

2 @ 68.5 keV / °

Appendix

128

Figure A11 Crystallite size from Rietveld refinement of the in situ high energy X-ray diffraction patterns over 5k

cycle of the AST for Pt/Vulcan (a), Pt/O-Vulcan (b), Pt/N-Vulcan 400°C (c) and Pt/N-Vulcan 800°C (d).

a b

c d

01000 2000 3000 4000 5000

3.6

3.8

4.0

4.2

4.4

4.6

4.8

5.0

crystallite size / nm

cycle number

Pt/Vulcan

01000 2000 3000 4000 5000

3.6

3.8

4.0

4.2

4.4

4.6

4.8

5.0

cycle number

crystallite size / nm

Pt/O-Vulcan

01000 2000 3000 4000 5000

3.6

3.8

4.0

4.2

4.4

4.6

4.8

5.0

cycle number

crystallite size / nm

Pt/N-Vulcan 400°C

01000 2000 3000 4000 5000

3.6

3.8

4.0

4.2

4.4

4.6

4.8

5.0

cycle number

crystallite size / nm

Pt/N-Vulcan 800°C

Appendix

129

Figure A12 In situ small angle X-ray scattering curves over 5k cycle of the AST for Pt/Vulcan, Pt/O-Vulcan,

Pt/N-Vulcan 400°C and Pt/N-Vulcan 800°C (a,c,e,g) and scattering curves at start of the AST including the

individual fit curves (b,d,f,h).

0.01 0.1

100

1000

10000

100000 No. CV

2000

4000

5000

Intensity / a.u.

q / Å-1

0.01 0.1

100

1000

10000

100000 AST CV1

data

fit

Intensity / a.u.

q / Å-1

0.01 0.1

100

1000

10000

AST CV1

data

fit

Intensity / a.u.

q / Å-1

0.01 0.1

100

1000

No. CV

2000

4000

5000

Intensity / a.u.

q / Å-1

0.1

100

1000

AST CV1

data

fit

Intensity / a.u.

q / Å-1

0.01 0.1

100

1000

10000

100000 AST CV1

data

fit

Intensity / a.u.

q / Å-1

0.01 0.1

100

1000

10000

100000 No. CV

2000

4000

5000

Intensity / a.u.

q / Å-1

0.01 0.1

100

1000

10000

No. CV

1000

3000

5000

Intensity / a.u.

q / Å-1

c d

e f

g h

a b

Pt/Vulcan

Pt/O-Vulcan

Pt/N-Vulcan 400°C

Pt/N-Vulcan 800°C

Appendix

130

Figure A13 Mean particle diameter from fit of the in situ small angle X-ray scattering curves over 5k cycle of the

AST for Pt/Vulcan (a), Pt/O-Vulcan (b), Pt/N-Vulcan 400°C (c) and Pt/N-Vulcan 800°C (d).

01000 2000 3000 4000 5000

2.6

2.8

3.0

3.2

3.4

3.6

3.8

particle diameter / nm

cycle number

Pt/O-Vulcan

01000 2000 3000 4000 5000

2.6

2.8

3.0

3.2

3.4

3.6

3.8

particle diameter / nm

cycle number

Pt/Vulcan

01000 2000 3000 4000 5000

2.6

2.8

3.0

3.2

3.4

3.6

3.8

particle diameter / nm

cycle number

Pt/N-Vulcan 400°C

01000 2000 3000 4000 5000

2.6

2.8

3.0

3.2

3.4

3.6

3.8

particle diameter / nm

cycle number

Pt/N-Vulcan 800°C

a b

c d

Appendix

131

A3 Supporting Information to Chapter 6

Table A6 Pt weight loading (wt%) for Pt on ITO, RTO and C as determined by ICP-OES analysis.

Pt loading

[wt%]

Pt/C

20.0

Pt/ITO

29.9

Pt/RTO

46.1

Figure A14 X-ray diffractograms of (a) ITO and Pt/ITO with solid blue lines indicating In2O3 reference pattern

and (b) RTO and Pt/RTO with symbols indicating TiO2 anatase and rutile and RuO2. Vertical dashed lines denote

reference powder diffraction patterns of fcc Pt (PDF#00-004-0802). XRDs were obtained using Cu Kα radiation.

20 30 40 50 60 70 80 90

Pt/ITO

ITO

Intensity / a.u.

 / °

In2O3 #00-006-0416

20 30 40 50 60 70 80 90

+**

Intensity / a.u.

 / °

* TiO2 anatase #01-084-1286 + TiO2 rutile #01-076-1939 ~ RuO2 #01-088-0322

Pt/RTO

RTO

Appendix

132

Figure A15 High energy XRD patterns for Pt/RTO, Pt/ITO and Pt/C in the as-prepared state.

Table A7 Crystallite diameter (Dx, Dy and Dz) for ellipsoidal Pt particles on RTO and C support and for spherical

particles on ITO as determined by Rietveld refinement of HE-XRD.

crystallite diameter / nm

Pt/ITO

4.06±0.09

Pt/RTO

3.28±0.05

4.54±0.11

2.44±0.02

Pt/C

2.32±0.04

2.46±0.05

1.78±0.02

Figure A16 Morphology of Pt nanoparticles on a) ITO (green), b) RTO (blue) and c) carbon (red) support and

corresponding histograms showing mean particle diameter as determined by transmission electron microscopy.

The particle size histogram was determined from analyzing > 200 particles along their shortest particle axes.

Pt/C

Pt/RTO

Pt/ITO

a b c

1 2 3 4

100

120

140

size / nm

counts

2.2 ± 0.5 nm

3 4 5 6 7

100

120

size / nm

counts

4.6 ± 0.7 nm

2 3 4 5 6

100

120

size / nm

counts

3.8 ± 0.6 nm

Appendix

133

Table A8 Structural parameters of Pt on ITO, RTO and C: Otd-incorporation and lattice constant by Rietveld

refinement of HE-XRD.

OTd-incorporation

lattice parameter / Å

Pt/ITO

0.15±0.02

3.9301±E-4

Pt/RTO

0.000±0.016

3.9199±E-4

Pt/C

0.000±0.018

3.9136±E-4

Table A9 Comparison of ECSA values based on the integration in the Hupd region or the CO Oxidation peak area.

ECSA/ m2 g -1

Hupd

Pt/C

63.1

67.5

Pt/ITO

17.8

25.6

Pt/RTO

19.0

16.7

Figure A17 Evolution of mass-normalized Pt dissolution rate as determined by SFC ICP-MS measurements during

the initial anodic potential sweep from 0.05-1.5 V for Pt nanoparticles supported on ITO (green), RTO (blue), and

carbon (red).

Appendix

134

Figure A18 Evolution of mass-normalized Pt dissolution rate of electrochemically-oxidized Pt nanoparticles

supported on ITO (green), RTO (blue), and carbon (red) as determined by SFC ICP-MS measurements during the

cathodic potential sweep from 1.5-0.05V after initial anodic potential sweep up to 1.5V.

0.2 0.4 0.6 0.8 1.0 1.2 1.4

ITO RTO Carbon

Pt dissolution / gdet g-1

we l-1

E / V vs. RHE

List of Acronyms

135

List of Acronyms

Abbreviation Definition

AFC Alkaline fuel cell

ASAXS Anomalous small angle X-ray scattering

AST Accelerated stress test

at% atomic %

BE Binding energy

BET Brunauer Emmett Teller

CNF Carbon nanofibre

CNT Carbon nanotube

CV Cyclic Voltammetry

DFT Density functional theory

DI Deionized

DOE Department of energy

ECSA Electrochemical active surface area

EDX Energy dispersive X-ray spectroscopy

ESRF European Synchrotron Radiation Facility

fcc face centered cubic

FWHM Full width half maximum

GC Glassy carbon

GDE Gas diffusion electrode

GDL Gas diffusion layer

HAADF High-angle annular dark field

HE-XRD Hiigh energy X-ray diffraction

HOR Hydrogen oxidation reaction

HP High potential

HRTEM High resolution transmission electron mictroscopy

HT-DEMS High temperature-differential electrochemical mass spectroscopy

Hupd Hydrogen under potential deposition

HZB Helmholtz Zentrum Berlin

ICP-OES Inductively coupled plasma mass spectroscopy

ITO Indium tin oxide

LP Low potential

LSV Linear sweep voltammetry

MCFC Molten carbonate fuel cell

MEA Membrane electrode assembly

MMS Mercury/mercury sulfate reference electrode

MS Mass spectrometer

MWCNT Multi-walled carbon nanotube

OCP Open circuit potential

OER Oxygen evolution reaction

ORR Oxygen reduction reaction

PANI Polyaniline

PDF Powder diffraction file

PEFC Polymer electrolyte fuel cell

PEIS Potentiostatic electrochemical impedance spectroscopy

PEMFC Proton exchange membrane fuel cell

List of Acronyms

136

Abbreviation Definition

PSD Particle size distribution

PTFE Polytetrafluorethylene

RDE Rotating disc electrode

RHE Reversible hydrogen electrode

rpm Rounds per minute

RT Room temperature

RTO Ruthenium Titanium Oxide

SAXS Small angle X-ray scattering

SFC ICP-MS Scanning flow cell inductively coupled plasma-mass spectroscopy

SMSI Strong metal support interaction

SOFC Solid oxide fuel cell

STEM Scanning transmission electron mictroscopy

Td Tetrahedral

TEM Transmission electron mictroscopy

UHV Ultra-high vacuum

UPD Under potential deposited/deposition

WL White line

wt% weight %

XAS X-ray absorption spectroscopy

XPS X-ray photoelectron spectroscopy

XRD X-ray diffraction

YSZ Yttria-stabilized zirconia

ZIF Zeolitic imidazolate framework

ZP Zeta potential

yr year

List of Chemicals

137

List of Chemicals

Name

Abbreviation

Concentration /

Purity

Supplier

1,2-tetradecanediol

90 %

Sigma Aldrich

Acetone

100 %

VWR Chemicals

Ammonia

NH3

99.98 %

Air Liquide

Carbon monoxide

99.997 %

Air Liquide

Dibenzylether

Bn2O

98.0 %

Fluka

Ethanol

EtOH

100 %

VWR Chemicals

Formic Acid

∼98 %,

Fluka

Hexachloroplatinic(II) acid

hexahydrate

H2PtCl6·6H2O

Pt 37.5 % min

Alfa Aesar

Hydrochloric acid

HCl

37 %

VWR Chemicals

Hydrogen/Argon

H2/Ar

4% H2/Ar

Air Liquide

Indium(III) chloride

InCl3

Strem Chemicals

Isopropanol

iPrOH

100 %

VWR Chemicals

Nitric acid

HNO3

69 %

Merck

Nitrogen

99.999 %

Air Liquide

Nafion

5 wt%

Sigma Aldrich

Oxygen

99.998 %

Air Liquide

Oleic acid

90.00 %

Alfa Aesar

Oleylamine

70.0 %

Sigma Aldrich

Perchloric acid

HClO4

99.999 %

Sigma Aldrich

Polytetrafluorethylene

PTFE

60 wt%

ElectroChem Inc.

Propylene epoxide

Sigma Aldrich

Platinum(II) acetylacetonate

Pt(acac)2

Pt 48 % min.

Alfa Aesar

Ruthenium(III) chloride hydrate

RuCl3·xH2O

Ru 35-40 %

Acros Organics

Sulphuric acid

H2SO4

95 %

VWR Chemicals

Tin(IV) chloride

SnCl4

Acros Organics

Titanium dioxide powder

TiO2

Acros Organics

Ultrapure water

miliQ

16.8 MΩcm

List of Figures 
138 
 
List of Figures 
 
Figure 1.1 Schematic illustration of the use of Hydrogen for energy conversion and storage. In here, 
hydrogen is produced by water electrolysis powered by renewable energy sources. Hydrogen is stored 
and distributed and can be used on demand in a fuel cell to generate electricity. ................................... 1 
Figure 1.2 Schematic illustration of the general principle of a proton exchange membrane fuel cell 
(PEMFC). At the anode the fuel (hydrogen) is oxidized to form protons and electrons. Protons travel 
through the membrane to the cathode to form water by reacting with oxygen and electrons. The electrons 
travel through an external electric circuit and can be used as power supply. .......................................... 3 
Figure 1.3 Schematic illustration of proposed degradation mechanisms for Pt nanoparticles supported 
on carbon including Pt dissolution, Ostwald ripening, agglomeration, detachment and carbon corrosion. 
Reprinted with permission from Ref. 26 from Beilstein Journal of Nanotechnology under the terms of 
the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0). ................... 8 
Figure 3.1 Schematic illustration of the Bragg equation. ..................................................................... 23 
Figure  3.2  Schematic  illustration  of  the  sequence  of  an  electrochemical  activity  and  stability 
measurement,  including  conditioning  by  cyclic  voltammetry  (CV),  impedance  measurement  by 
potentiostatic electrochemical impedance spectroscopy (PEIS), activity determination by linear sweep 
voltammetry (LSV), accelerated stress test (AST) in two different potential windows (low potential (LP) 
and high potential (HP)) and activity determination after the AST. CO stripping, as applied in chapter 
4, was performed directly before and after the AST. ............................................................................ 29 
Figure 3.3 Schematic illustration of the setup and electrochemical in situ cell as used at a synchrotron 
facility, showing the incident and scattered X-rays, the 2D detector and the in situ transmission cell. 
Distance between sample and detector is defined as working distance. ................................................ 34 
Figure 3.4 Schematic illustration of the elastic scattering as the basis of SAXS. ................................ 37 
Figure  4.1 X-ray diffraction patterns of the bare ITO support material (black, bottom) and Pt/ITO 
catalyst (dark red, top) obtained using Cu Kα radiation. Solid lines denote pure In2O3 reference powder 
diffraction  pattern  (PDF#00-006-0416).  Vertical  dashed  lines  denote  reference  patterns  of  fcc  Pt 
(PDF#00-004-0802). ............................................................................................................................. 44 
Figure 4.2 Morphology of Pt/ITO in the initial state: (a) showing overview HAADF-STEM  image, (b) 
TEM and (c) HR-TEM images. ............................................................................................................. 44 
Figure  4.3  a,b)  TEM  images  of  the  as  prepared  Pt  nanoparticles  on  the  ITO  support,  after 
electrochemical cycling for 5k times in d,e) lower potential region and in g,h) the higher potential region, 
as well as the corresponding histograms showing c,f,i) the particle size distribution. Histograms were 
obtained by measuring the diameter of at least 200 particles with errors obtained from standard deviation 
of mean particle diameter. ..................................................................................................................... 45 
Figure 4.4 Cyclic voltammograms of Pt/ITO catalyst before and after potential cycling in (a) lower and 
(c) higher potential region. CVs were recorded in nitrogen saturated electrolyte from 0.05–1 V with a 
scan rate of 100 mV s-1. b,d) LSVs of the particular states with the bar plots in the inlets representing 
the mass activity (jm) evaluated at 0.9 V. LSVs were recorded in oxygen-saturated electrolyte from 0.05–
1 V with a scan rate of 5 mV s-1 and 1600 rpm. All electrode potentials have been corrected for iR drop.
 ............................................................................................................................................................... 46 
Figure 4.5 Comparison of ECSAs based on the integration in the Hupd and the CO oxidation potential 
range. CO-ECSAs were determined by integrating the CO oxidation peak area from the first cycle of 
the CO stripping experiment after subtraction of the second cycle representing the bare CO-free Pt 
surface. Hupd-ECSA was determined by subtracting the first from the second cycle of the CO stripping 
experiment. ............................................................................................................................................ 47 
Figure 4.6 X-ray photoelectron spectroscopy measurements. (a) and (b) are showing the Pt 4f depth 
profiling accessed by the kinetic energy of the photoelectrons of 210, 550 and 1200 eV  at the initial (a) 
and  the  state  after  LP-AST (b)  Dotted  lines  represent  measured  data  and  solid  lines  the  fits  and 

List of Figures 
139 
 
component peaks. (c) and (d) are showing the 3d core levels for In and Sn, respectively, each at the 
initial and the cycled state. .................................................................................................................... 49 
Figure  4.7 HAADF-STEM  images  and  EDX  mapping  of  a)  Pt/ITO  initial  and  b)  after  LP-AST. 
Platinum is depicted in light blue, indium in green, and tin in red........................................................ 50 
Figure 4.8 In situ HE XRD measurements depicted as the evolution of diffraction patterns from the 
initial state to the end of the electrochemical cycling for the LP-AST (a) and the HP-AST (b). .......... 51 
Figure 4.9 Results from in situ HE-XRD for LP-AST. Weight fractions of a) crystalline phases and b) 
their crystallite size as determined by Rietveld refinement................................................................... 51 
Figure 4.10 Evolution of normalized scale factors for Pt and In2O3 from fits of HE-XRD patterns over 
the cycle number for (a) LP-AST and (b) HP-AST. ............................................................................. 52 
Figure 4.11 In situ ASAXS measurements for LP-AST: background subtracted, Pt element specific 
scattering curves for selected cycle numbers (a), particle size distribution (PSD) as a function of rel. 
intensity over the mean particle diameter for selected cycle numbers (b), mean Pt particle diameter (c) 
and the evolution of monodispersity as a function of σ over the cycle number (d). ............................. 53 
Figure  4.12  Results  from  in  situ  HE-XRD  measurements  for  HP-AST.  Weight  fractions  of  (a) 
crystalline phases and (b) their crystallite size as determined by Rietveld refinement. ........................ 54 
Figure 4.13 In situ ASAXS measurements for HP-AST: background subtracted, Pt element specific 
scattering curves for selected cycle numbers (a), particle size distribution (PSD) as a function of rel. 
intensity over the mean particle diameter for selected cycle numbers (b), mean Pt particle diameter (c) 
and the evolution of monodispersity as a function of σ over the cycle number (d). ............................. 55 
Figure 4.14 In situ scanning flow cell ICP-MS measurements. Depicted are the Sn, In and Pt dissolution 
rates and the applied electrochemical protocols from the bottom to the top for (a) LP-AST and (b) HP-
AST. The respective dissolution rates in detected (det) metal in μg per volume electrolyte (μgdet L-1) are 
plotted against the time. A Pt/C reference sample was measured and therefore, the Pt dissolution rate 
was also normalized to the Pt mass loading on the working electrode (WE) (μgdet L-1  μgWE-1).  The 
electrochemical protocol was conducted as follows: Beginning with 100 CVs (activation regime) from 
0.05-1 V, followed by potential cycling in the LP regime (0.6-0.95 V, 40 CVs, 100 mV s-1) or in the HP 
regime (1.0-1.5 V, 200 CVs, 500 mV s-1) and followed by another 3 cycles from 0.05-1 V, all CVs were 
recorded with a scan rate of 100mV/s. The first contacts between catalyst and electrolyte (cell contact) 
are denoted with arrows. Breaks at the time axis of (a) and (b) have been implemented between 500 and 
1700 s. Graphs with complete time axis can be found in Figure A5 and Figure A6............................. 56 
Figure 4.15 Results from in situ SFC ICP-MS of the integration of the peaks arising from Pt dissolution 
over time of the measurement (in μgdet) detected by the ICP-MS per mass loading Pt on the working 
electrode (in μgWE) for LP- and HP-AST. ............................................................................................. 57 
Figure  5.1 Schematic illustration of the carbon modification procedure including oxidation step in 
concentrated nitric acid resulting in O-Vulcan, and ammonolysis in pure NH3 at 400 and 800°C resulting 
in N-Vulcan. .......................................................................................................................................... 67 
Figure  5.2  Surface  and  compositional  analysis  of  the  carbon  materials:  content  of  nitrogen  (from 
elemental analysis) and oxygen (from hot gas extraction) for the modified carbons (a), physical BET 
surface area (b) and zeta potential (c) for modified carbons in comparison to the unmodified Vulcan.68 
Figure 5.3 (a) XPS N 1s spectra and individual peak deconvolution for modified carbons and Pt/N-
Vulcan 400°C and (b) schematic illustrastion of different N-functionalities (green: graphitic N, blue: 
quaternary N, yellow: pyrrolic N and red: pyridinic N) in graphene-like plane. .................................. 70 
Figure 5.4 Fractions of the different N-moieties from deconvolution of the N 1s spectra in the modified 
carbons and in Pt/N-Vulcan 400°C. ...................................................................................................... 71 
Figure 5.5 High Temperature-DEMS measurements from 0.06-1.05 V at 140 °C for modified carbons 
and the Vulcan reference carbon with respect to the resulting current j normalized to mass loading of 
carbon (a) and the ion current for CO2 (m/z = 44) from the MS normalized to mass loading of carbon 
(b). HCl-Vulcan represents a HCl-treated Vulcan (1M HCl, RT, 24h) in order to remove metal traces 
from the unmodified Vulcan. ................................................................................................................ 72 

List of Figures 
140 
 
Figure 5.6 X-ray powder diffraction patterns of the modified and unmodified carbon supports and the 
respective Pt electrocatalysts. Vertical blue lines represent the reference pattern for fcc Pt (PDF#00-004-
0802). ..................................................................................................................................................... 73 
Figure 5.7 TEM images (a) and ORR specific activity (jspec) for the four catalysts (b) for Pt/Vulcan, 
Pt/O-Vulcan, Pt/N-Vulcan 400°C and Pt/N-Vulcan 800°C. See Figure A9 in the Appedix section A2 for 
more TEM overview images. ................................................................................................................ 74 
Figure 5.8 Results from AST for 5k, 10k and 30k cycles from 0.6-0.95 V in Nitrogen-saturated 0.1 M 
HClO4 for Pt/Vulcan, Pt/O-Vulcan, Pt/N-Vulcan 400°C and Pt/N-Vulcan 800°C as a function of mass 
activity at 0.9 V jmass(a,d,g,j), ECSA (b,e,h,k) and specific activity jspec (c,f,i,l). ................................... 75 
Figure 5.9 In situ high energy X-ray diffraction patterns over 5k cycle of the AST for Pt/Vulcan (a), 
and Pt/N-Vulcan 400°C (b). A full set of in situ HE-XRD patterns for all samples can be found in the 
Appendix section A2, Figure A10. ........................................................................................................ 76 
Figure 5.10 Crystallite sizes obtained from Rietveld Refinement of the in situ HE-XRD patterns over 
5k cycles of the AST for Pt/N-Vulcan 400°C (a) and Pt/Vulcan (b). Inlets in both graphs showing the 
mass activity up to 30k cycles between 0.6-0.95 V of the AST in nitrogen-saturated 0.1 M HClO4. A 
full set evolution of crystallite size for all samples can be found in the Appendix section A2, Figure A11.
 ............................................................................................................................................................... 77 
Figure 5.11 In situ small angle X-ray scattering curves and mean particle diameter from SAXS fitting 
over 5k cycle of the AST for Pt/Vulcan (a,b) and Pt/N-Vulcan 400°C (c,d). A full set of in situ SAXS 
curves and evolution of mean particle diameters for all samples can be found in the Appendix section 
A2, Figure A12 and Figure A13. ........................................................................................................... 78 
Figure 6.1 Structure and morphology of Pt nanoparticles supported on RTO (blue), ITO (green) and  
carbon  (red)  as  determined  by  powder  X-ray  diffraction  pattern  (a)  and  transmission  electron 
microscopy images (b,c,d). The Pt particle sizes as depicted in the histograms in (e,f,g) were determined 
from analyzing > 200 Pt particles along their shortest principal axes. .................................................. 84 
Figure 6.2 Histograms of mean particle diameter for Pt supported on (a) ITO (green), (b) RTO (blue) 
and  (c) carbon (red) support. Top row shows the average particle diameter measured along their shortest 
particle axis and bottom row along their longest particle axes. ............................................................. 85 
Figure 6.3 Morphology of Pt nanoparticles on (a) RTO (blue) and (b) carbon (red) support in a TEM 
tilting study from 0-30°. ........................................................................................................................ 86 
Figure 6.4 Electrochemical characterization of Pt nanoparticles on RTO (blue), ITO (green) and carbon 
(red) support by the evolution of cyclic voltammogram with increasing upper potential limit in N2-
saturated 0.1 M HClO4 from 0.5-1.4 V (a-c) and from CO stripping experiments (d). ......................... 87 
Figure 6.5 In situ high-energy X-ray diffraction patterns of Pt nanoparticles supported on ITO (green, a), 
RTO (blue, b) and carbon (red, c) on stepwise electrochemical oxidation of Pt (potential hold for 10 min 
each). ..................................................................................................................................................... 89 
Figure 6.6 Structural response of crystalline Pt nanoparticles supported on ITO (green circles), RTO 
(blue triangles) and carbon (red squares) on stepwise electrochemical oxidation (potential hold for 10 
min each; reducing step to 1.0 V represented by open symbols) with respect to the change of Pt lattice 
parameter (a), normalized change of the Pt scale factor (b) as determined by Rietveld refinement of in 
situ high-energy X-ray diffraction pattern. The solid lines are shown to guide the eyes of the reader. 90 
Figure 6.7 Evolution of the number of unit cells (2R/aPt) of the metallic Pt domains for Pt/C along the 
x,y and z-axes. The solid lines are shown to guide the eyes of the reader. ........................................... 91 
Figure  6.8 Normalized in situ  Pt L3–edge X-ray  near-edge  absorption  spectra  for  Pt nanoparticles 
supported on ITO (green, a), RTO (blue, b) and carbon (red, c) for potentials from 0.6 – 1.4 V of 
electrochemical oxidation of Pt. The positions of the Pt L3 resonances applied in the fit procedure are 
denoted. ................................................................................................................................................. 92 
Figure  6.9  Structural  response  of  Pt  nanoparticles  supported  on  ITO  (green  circles),  RTO  (blue 
triangles), and carbon (red squares) on stepwise electrochemical oxidation (potential hold for 10 min 
each) with respect to the white line area (a) the integrated areas as the resonances 1 (Pt L3-Res 1) at 
~11566 eV and Pt L3-Res 2 at ~11570 eV (b) and Pt L3-Res 3 at ~11580 eV (c) as determined by peak 

List of Figures 
141 
 
fitting of in situ Pt L3–edge X-ray near-edge absorption spectra. The solid lines are shown to guide the 
eye of the reader. ................................................................................................................................... 93 
Figure  6.10  Pt  dissolution  profiles  from  in  situ  SFC  ICP-MS  measurement  in  gdet  Pt  per  volume 
electrolyte in l and electrochemical surface area (ECSA) in m2ECSA. (a) shows the whole electrochemical 
measurement  (100  CVs of activation with a  scanrate 100 mV s-1  and 4  CVs  of  Pt  Oxidation from 
0.05-1.5 V with a scanrate 10 mV s-1) and the corresponding dissolution profiles for Pt on carbon (red), 
ITO (green) and RTO (blue) support and (b) the four oxidation cycles from 0.05-1.5 V. .................... 96 
Figure 6.11 Evolution of ECSA-normalized Pt dissolution rate for Pt nanoparticles supported on ITO 
(green), RTO (blue), and carbon (red) during the initial anodic potential sweep from 0.05-1.5 V with a 
scanrate of 10 mV s-1 as determined by SFC ICP-MS measurements. ................................................. 97 
Figure  6.12  Evolution  of  ECSA-normalized  Pt  dissolution  rate  of  electrochemically-oxidized  Pt 
nanoparticles supported on ITO (green), RTO (blue), and carbon (red) as determined by SFC ICP-MS 
measurements during the cathodic potential sweep from 1.5-0.05 V with a scanrate of 10 mV s-1 after 
initial anodic potential sweep up to 1.5 V. ............................................................................................ 98 
Figure 7.1 Schematic illustration of degradation phenomena as found to occur in Pt/ITO electrocatalyst 
during  two  different  simulated  degradation  protocols:  low  potential  AST  and  high  potential  AST 
(denoted as LP-AST and HP-AST). Reproduced from Ref 52 (Adv. Energy Mat., 2018, 8 (4), 1701663) 
with permission from John Wiley and Sons, Copyright 2018............................................................. 102 
Figure 7.2 Schematic illustration of the process of support functionalization by ammonolysis at 400°C 
and introduction of surface abundant pyrrolic-N. In the second step, Pt deposition on modified support 
and the improved ORR long-term stability is illustrated. Reprinted with permission from Ref71 (Chem. 
Mat., 2018, doi: 10.1021/acs.chemmater.8b03612). Copyright (2018) American Chemical Society. 103 
Figure 7.3 Schematic illustration of potential-dependent responses of Pt nanoparticles on carbon, RTO 
and ITO support on electrochemical oxidation with respect to lattice expansion, formation of PtOx and 
Ptn+ dissolution. ................................................................................................................................... 105 
 
Figure A1 TEM images of as prepared Pt nanoparticles on the ITO support which was used for the HP-
AST experiments with a weight loading of 29.9 wt% (a,b) and the corresponding histogram showing 
particle size distribution (c). Histograms were obtained by measuring the diameter of  at least 200 
particles with errors obtained from standard deviation of mean particle diameter. ............................ 119 
Figure A2 Cyclic voltammograms of Pt/C reference catalyst (weight loading 20 wt% Pt on carbon) 
before and after potential cycling in lower (a) and higher (c) potential region. CVs were recorded in 
nitrogen saturated electrolyte from 0.05-1 V with a scan rate of 100 mV s-1. Figure b and d are showing 
LSVs of the particular states LSVs were recorded in oxygen saturated electrolyte from 0.05-1 V with a 
scan rate of 5 mV·s-1and 1600 rpm. All electrode potentials have been corrected for iR drop. ......... 119 
Figure A3 Electrochemical CO stripping experiments for Pt/ITO electrocatalyst. For the two different 
stability tests in low (a) and high (b) potential regime three different points in the characterization 
protocol were chosen for CO stripping: initial, after CV+LSV and after 5k CV+LSV. ..................... 120 
Figure A4 In situ HE XRD measurements depicted as the evolution of diffraction patterns from the 
initial state to the end of the electrochemical cycling for the LP-AST (a) and the HP-AST (b). The inlet 
in (a) is showing the Pt (200) diffraction peak superimposed by an In2O3 peak (denoted with dashed 
vertical lines). ...................................................................................................................................... 121 
Figure A5 In situ scanning flow cell ICP-MS measurements. Depicted are the Sn, In and Pt dissolution 
rates and the applied electrochemical protocols from the bottom to the top for LP-AST. The respective 
dissolution rates in detected metal per volume electrolyte (μgdet l-1) are plotted against the time. A Pt/C 
reference sample was measured and therefore, the Pt dissolution rate was also normalized to the Pt mass 
loading on the working electrode (WE) (μgdet l-1μgWE-1). The electrochemical protocol was conducted as 
follows: Beginning with 100 CVs (activation regime) from 0.05-1 V, followed by potential cycling in 
the LP regime (0.6-0.95 V, 40 CVs, 100 mV·s-1) and followed by another 3 cycles from 0.05-1 V, all 
CVs were recorded with a scan rate of 100mV s-1. The first contacts between catalyst and electrolyte 
(cell contact) are denoted with arrows. ............................................................................................... 121 

List of Figures 
142 
 
Figure A6 In situ scanning flow cell ICP-MS measurements. Depicted are the Sn, In and Pt dissolution 
rates and the applied electrochemical protocols from the bottom to the top for HP-AST. The respective 
dissolution rates in detected metal per volume electrolyte (μgdet l-1) are plotted against the time. A Pt/C 
reference sample was measured and therefore, the Pt dissolution rate was also normalized to the Pt mass 
loading on the working electrode (WE) (μgdet l-1μgWE-1). The electrochemical protocol was conducted as 
follows: Beginning with 100 CVs (activation regime) from 0.05-1 V, followed by potential cycling in 
the HP regime (1.0-1.5 V, 200 CVs, 500 mV·s-1) and followed by another 3 cycles from 0.05-1 V, all 
CVs were recorded with a scan rate of 100mV s-1. The first contacts between catalyst and electrolyte 
(cell contact) are denoted with arrows. ................................................................................................ 122 
Figure A7 Evolution of Pt lattice constant as extracted from Rietveld Refinement from in situ HE-XRD 
measurements over the cycle number for (a) LP-AST and (b) HP-AST. ............................................ 123 
Figure A8 XPS C 1s spectra and individual peak deconvolution for unmodified Vulcan (a), O-Vulcan 
(b), N-Vulcan 400° (c), N-Vulcan 800°C (d) and for Pt/N-Vulcan 400°C (e). ................................... 125 
Figure A9 TEM images for Pt on unmodified Vulcan (a), O-Vulcan (b), N-Vulcan 400°C (c), and N-
Vulcan 800°C (d). ................................................................................................................................ 126 
Figure A10 In situ high energy X-ray diffraction patterns over 5k cycle of the AST for Pt/Vulcan (a), 
Pt/O-Vulcan (b), Pt/N-Vulcan 400°C (c) and Pt/N-Vulcan 800°C (d). ............................................... 127 
Figure  A11  Crystallite  size  from Rietveld refinement  of the  in  situ  high  energy  X-ray  diffraction 
patterns over 5k cycle of the AST for Pt/Vulcan (a), Pt/O-Vulcan (b), Pt/N-Vulcan 400°C (c) and Pt/N-
Vulcan 800°C (d). ................................................................................................................................ 128 
Figure A12 In situ small angle X-ray scattering curves over 5k cycle of the AST for Pt/Vulcan, Pt/O-
Vulcan, Pt/N-Vulcan 400°C and Pt/N-Vulcan 800°C (a,c,e,g) and scattering curves at start of the AST 
including the individual fit curves (b,d,f,h). ........................................................................................ 129 
Figure A13 Mean particle diameter from fit of the in situ small angle X-ray scattering curves over 5k 
cycle of the AST for Pt/Vulcan (a), Pt/O-Vulcan (b), Pt/N-Vulcan 400°C (c) and Pt/N-Vulcan 800°C 
(d). ....................................................................................................................................................... 130 
Figure A14 X-ray diffractograms of (a) ITO and Pt/ITO with solid blue lines indicating In2O3 reference 
pattern and (b) RTO and Pt/RTO with symbols indicating TiO2 anatase and rutile and RuO2. Vertical 
dashed lines denote reference powder diffraction patterns of fcc Pt (PDF#00-004-0802). XRDs were 
obtained using Cu Kα radiation. ........................................................................................................... 131 
Figure A15 High energy XRD patterns for Pt/RTO, Pt/ITO and Pt/C in the as-prepared state. ........ 132 
Figure A16 Morphology of Pt nanoparticles on a) ITO (green), b) RTO (blue) and  c) carbon (red) 
support and corresponding histograms showing mean particle diameter as determined by transmission 
electron microscopy. The particle size histogram was determined from analyzing > 200 particles along 
their shortest particle axes. .................................................................................................................. 132 
Figure  A17  Evolution  of  mass-normalized  Pt  dissolution  rate  as  determined  by  SFC  ICP-MS 
measurements during the initial anodic potential sweep from 0.05-1.5 V for Pt nanoparticles supported 
on ITO (green), RTO (blue), and carbon (red). ................................................................................... 133 
Figure  A18  Evolution  of  mass-normalized  Pt  dissolution  rate  of  electrochemically-oxidized  Pt 
nanoparticles supported on ITO (green), RTO (blue), and carbon (red) as determined by SFC ICP-MS 
measurements during the cathodic potential sweep from 1.5-0.05V after initial anodic potential sweep 
up to 1.5V. ........................................................................................................................................... 134 
 

List of Tables 
143 
 
List of Tables 
 
Table 3.1 Overview over different materials used in this work with the corresponding chapters in which 
their characterization is discussed. ........................................................................................................ 19 
Table 3.2 Overview of Methods for physicochemical characterization used in this work with reference 
to  the  sections  in  which  they  are  described.  x  indicates  the  the  application  of  the  method  for  the 
corresponding result chapter. ................................................................................................................ 22 
Table 3.3 Details on Ink preparation and composition depending on the different supports used in this 
work....................................................................................................................................................... 28 
Table 3.4 Overview of ASTs performed in this with work for Pt/ITO and Pt/(un)modified carbon with 
respect to the applied cycle numbers. .................................................................................................... 31 
Table 4.1 Near-surface composition of Pt/ITO before and after LP-AST in the lower potential regime 
as obtained from XPS measurements. The molar ratios Pt, In, and Sn were determined from the peak 
areas of the Pt 4f as well as In and Sn 3d spectra recorded with a kinetic energy of the photoelectrons of 
550 eV. The number in brackets denotes the composition at a second beam position of each sample. 50 
Table  5.1  Assignment  of  binding  energy  (BE)  for  different  N-functionalities  from  individual 
deconvolution of the N 1s XP spectra. .................................................................................................. 69 
 
Table A1 Comparison of ECSAs based on the integration in the Hupd and the CO oxidation potential 
range. CO-ECSAs were determined by integrating the CO oxidation peak area from the first cycle of 
the CO stripping experiment after subtraction of the second cycle representing the bare CO-free surface. 
Hupd-ECSAs  were  determined  by  subtracting  the  first  from  the  second  cycle  of  the  CO  stripping 
experiment. .......................................................................................................................................... 120 
Table A2 Comparison of jm and ECSA before and after LP- and HP-AST for Pt/C reference catalyst. jm 
was determined at 0.9 V.  ECSAs were determined by integrating the H-desorption and adsorption area 
between 0.05-0.4 V and subtracting the capacitive current................................................................. 120 
Table A3 Results from in situ SFC ICP-MS of the integration over time of the measurement of the peaks 
arising from Pt dissolution in Pt/ITO and Pt/C reference and In and Sn from ITO support. Pt dissolution 
in μg is normalized to Pt mass loading on the working electrode in μgPt and In and Sn dissolution in μg 
is normalized to In+Sn mass loading on the working electrode in μg(In+Sn).Potential dissolution peaks are 
arising from cell contact, program start and distinct dissolution peaks during the electrochemical cycling 
(1st  and  2nd  peak).Additionally,  dissolution  values  are  listed  each  without  (wo/)  and  with  (w/) 
contribution of metal dissolution at cell contact. ................................................................................ 122 
Table A4 Elemental bulk and surface composition in at% of unmodified Vulcan, O-Vulcan and N-
Vulcan 400°C and 800°C from elemental analysis, hot gas extraction and XPS................................ 124 
Table A5 Pt weight loading (wt%) from ICP-OES for Pt/Vulcan, Pt/O-Vulcan, Pt/N-Vulcan 400°C and 
Pt/N-Vulcan 800°C. ............................................................................................................................ 124 
Table A6 Pt weight loading (wt%) for Pt on ITO, RTO and C as determined by ICP-OES analysis. 131 
Table A7 Crystallite diameter (Dx, Dy and Dz) for ellipsoidal Pt particles on RTO and C support and for 
spherical particles on ITO as determined by Rietveld refinement of HE-XRD. ................................. 132 
Table A8 Structural parameters of Pt on ITO, RTO and C: Otd-incorporation and lattice constant by 
Rietveld refinement of HE-XRD. ........................................................................................................ 133 
Table A9 Comparison of ECSA values based on the integration in the Hupd region or the CO Oxidation 
peak area. ............................................................................................................................................. 133 
 

List of Publications

144

List of Publications

Publications as part of this thesis

Schmies, H.; Bergmann, A.; Drnec, J.; Wang, G.; Teschner, D.; Kühl, S.; Sandbeck, D. J. S.;

Cherevko, S.; Gocyla, M.; Shviro, M.; Heggen, M.; Ramani, V.; Dunin-Borkowski, R. E.;

Mayrhofer, K. J. J.; Strasser, P., Unravelling Degradation Pathways of Oxide-Supported Pt Fuel

Cell Nanocatalysts under In Situ Operating Conditions. Advanced Energy Materials 2018, 8

(4), 1701663.

Schmies, H.; Hornberger, E.; Anke, B.; Jurzinsky, T.; Nong, H. N.; Dionigi, F.; Kühl, S.;

Drnec, J.; Lerch, M.; Cremers, C.; Strasser, P., The Impact of Carbon Support Functionalization

on the Electrochemical Stability of Pt Fuel Cell Catalysts. Chemistry of Materials 2018, 30

(20), 7287-7295.

Other publications

Gorlin, M.; de Araujo, J. F.; 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. Journal of the American Chemical Society 2017, 139 (5),

2070-2082.

Beermann, V.; Gocyla, M.; Kuhl, S.; Padgett, E.; Schmies, H.; Goerlin, M.; Erini, N.; Shviro,

M.; Heggen, M.; Dunin-Borkowski, R. E.; Muller, D. A.; Strasser, P., Tuning the

Electrocatalytic Oxygen Reduction Reaction Activity and Stability of Shape-Controlled Pt-Ni

Nanoparticles by Thermal Annealing - Elucidating the Surface Atomic Structural and

Compositional Changes. Journal of the American Chemical Society 2017, 139 (46), 16536-

16547.

Sun, Y. Y.; Sinev, I.; Ju, W.; Bergmann, A.; Dresp, S.; Kuhl, S.; Spori, C.; Schmies, H.; Wang,

H.; Bernsmeier, D.; Paul, B.; Schmack, R.; Kraehnert, R.; Cuenya, B. R.; Strasser, P., Efficient

Electrochemical Hydrogen Peroxide Production from Molecular Oxygen on Nitrogen-Doped

Mesoporous Carbon Catalysts. ACS Catalysis 2018, 8 (4), 2844-2856.

Hornberger, E.; Bergmann, A.; Schmies, H.; Kühl, S.; Wang, G.; Drnec, J.; Sandbeck, D. J. S.;

Ramani, V.; Cherevko, S.; Mayrhofer, K. J. J.; Strasser, P., In Situ Stability Studies of Platinum

Nanoparticles Supported on Ruthenium−Titanium Mixed Oxide (RTO) for Fuel Cell Cathodes.

ACS Catalysis 2018, 8 (10), 9675-9683.