Design of a promoter to enhance the stability of catalysts for hydrocarbon reactions 1

Carbon binding energies for the on-surface hcp hollow sites, the octahedral sites for the first and second subsurface layer and for octahedral sites in the Ni bulk 54 Table 4.2 Influence

ACKNOWLEDGEMENTS

Firstly, I would like to express my sincere appreciation to my supervisor, Dr Mark Saeys, for his encouragement, insight, support and guidance throughout the course of

this research project He has been an invaluable help by providing technical guidance and support pertaining to my research work

I also wish to extend my sincere gratitude to Dr Armando Borgna from ICES for his supervision during my experimental studies in ICES I am also thankful to Dr Chen Luwei and other colleagues in ICES for their helpful discussions on my experimental

studies

I would sincerely like to thank my group members such as Sun Wenjie, Tan Kong Fei, Hong Won Keon, Chua Yong Ping Gavin, Fan Xuexiang, Zhuo Mingkun, Su Mingjuan, Ravi Kumar Tiwari, Shangguan Wangzuo and Dianna Otalvaro for their help, support

and encouragement throughout my research work

Finally, special thanks to my dear husband Ye Ming, for being there to support me as I

pursue my doctorate degree I am extremely grateful for his love, patience and especially his understanding, which have enabled my doctorate journey to be meaningful and successful

I

TABLE OF CONTENTS

Acknowledgements···I Table of contents··· II Summary ···VI Symbols and abbreviations ··· VIII List of tables ···XI List of figures··· XIII Publications ··· XVIII

Chapter 1 Introduction ··· 1

Chapter 2 First Principles Based Design of Metal Catalysts ··· 5

2.1 Introduction ··· 5

2.2 Models of Catalytic Surface ··· 6

2.3 Hydrogenation of Olefins and Aromatics ··· 8

2.4 Ammonia Synthesis··· 15

2.5 Steam Reforming ··· 19

2.6 Selective Catalytic Oxidation ··· 21

2.7 References ··· 24

Chapter 3 Computational Methods ··· 27

3.1 Quantum Chemistry: Theory and Methods ··· 27

3.1.1 Fundamentals··· 27

3.1.2 Density Functional Theory (DFT) ··· 28

3.1.3 Exchange-Correlation Functional ··· 30

3.1.4 Plane-Wave Basis Sets ··· 32

II

3.1.5 Pseudopotentials ··· 32

3.1.6 The Vienna Ab Initio Simulation Package··· 33

3.1.7 Nudged Elastic Band Method (NEB)··· 36

3.2 Computational Methodology ··· 38

3.3 References ··· 39

Chapter 4 First Principles Based Design of Ni Catalysts with Improved Coking

Resistance ··· 41

4.1 Introduction ··· 41

4.2 Computational Methods ··· 48

4.3 Thermodynamic Diagram for Chemisorbed Carbon on Ni Catalyst ··· 50

4.3.1 Chemisorption of On-surface Carbon on Ni(111)··· 50

4.3.2 Chemisorption of Subsurface Carbon on Ni(111)··· 51

4.3.3 Stability of Bulk Carbon in Ni Catalyst ··· 55

4.3.4 Distribution of Carbon Atoms between On-surface and Subsurface Sites ·· 56

4.3.5 Stability of Graphene Overlayer on Ni(111)··· 61

4.3.6 Chemisorption of Carbon at the Ni(211) steps··· 63

4.4 Kinetics of Formation of Subsurface and Bulk Carbon, and of Graphene Islands on a Ni Catalyst··· 65

4.4.1 Effect of the Unit Cell Size on the Diffusion Barrier of On-surface Carbon Atoms to the First Subsurface Layer··· 66

4.4.2 Effect of Coverage/Concentration on Kinetics of Carbon Diffusion··· 67

4.4.3 Kinetics of the Formation of Graphene Islands from On-surface Carbon Atoms ··· 77

4.5 Effect of Boron on the Stability of Carbon on a Ni Catalyst ··· 83

4.5.1 Thermodynamic Diagram for Boron Chemisorption on Ni(111) ··· 83

4.5.2 Chemisorbed Boron at the Ni(211) Steps··· 88

4.6 Summary ··· 90

4.7 References··· 92

III

Chapter 5 First Principles Study of the Effect of Carbon and Boron on the Activity

of a Ni Catalyst ··· 95

5.1 Introduction ··· 95

5.2 Computational Methods ··· 99

5.3 Effect of Subsurface Carbon and Boron on Methane Activation ··· 102

5.3.1 Stability of subsurface carbon and boron··· 102

5.3.2 Clean Ni(111) Surface··· 106

5.3.3 Ni(111) with Subsurface Carbon ··· 108

5.3.4 Ni(111) with Subsurface Boron··· 114

5.4 Effect of Carbon and Boron on Methane Activation at Step Sites ··· 115

5.4.1 Clean Ni(211) Surface··· 116

5.4.2 Ni(211) Surface with Step Sites Blocked by Carbon··· 117

5.4.3 Ni(211) Surface with Step Sites Blocked by Boron··· 120

5.5 Summary ··· 123

5.6 References ··· 125

Chapter 6 Effect of Boron on the Stability of Ni Catalysts during Steam Methane Reforming ··· 127

6.1 Introduction ··· 127

6.2 Catalyst Synthesis··· 129

6.3 Catalyst Characterization ··· 129

6.4 Catalyst Testing ··· 132

6.5 Results and Discussion··· 133

6.5.1 Catalyst Characterization ··· 133

6.5.2 Methane Steam Reforming··· 139

6.5 Summary ··· 147

6.5 References··· 148

IV

Chapter 7 Conclusions and Future Suggestions ··· 151

V

SUMMARY

Deactivation by carbon deposition is a common challenge in many catalytic processes involving hydrocarbons, such as steam reforming of methane and heavier hydrocarbons over Ni-based catalysts First principles Density Functional Theory (DFT) calculations were combined with experimental investigations to design Ni catalysts with improved stability

To develop a molecular level understanding of the coking mechanism on Ni catalysts, the stability of different forms of carbon that can exist on Ni catalyst and the kinetics of carbon diffusion were studied using first principles calculations Extended graphene islands were found to be the most stable form of carbon on a Ni catalyst, with a carbon binding energy of –760 kJ/mol However, the formation of graphene islands resembles a nucleation process and requires critical islands of about 15-20 carbon atoms Step sites are the preferred adsorption sites for carbon atoms and can act as nucleation sites for the formation of graphene islands On-surface carbon atoms are relatively unstable with binding energies of around –660 kJ/mol Subsurface octahedral sites are also more stable than on-surface sites, and subsurface carbon is expected to build up easily under typical steam reforming reaction conditions The presence of subsurface carbon significantly decreases the activity of Ni catalysts and the methane activation energy increases from 101 kJ/mol to 143 kJ/mol, when all the sites in the first subsurface layer are occupied by carbon atoms

VI

Calculations indicate that boron atoms preferentially bind at the step sites and at octahedral sites just below the surface Boron and carbon atoms hence show similar a relative binding preference, and boron is proposed to selectively block both step and subsurface sites In addition, subsurface boron atoms were found to restructure the Ni(111) surface and lower the methane dissociation barrier from 101 kJ/mol to 64 kJ/mol Hence, boron atoms are believed to enhance the catalyst stability and do not reduce the catalyst activity

In order to validate our DFT predictions, Ni catalysts promoted with 0.5 wt% and 1.0 wt% boron were synthesized, characterized and tested during steam methane reforming Experiments at 800 ºC and at a Gas Hourly Space Velocity (GHSV) of 330,000

cm3/hr·gcat demonstrate that promotion with 1.0 wt% boron not only reduces the activity loss from 21% to 6%, but also enhances the initial conversion from 56% to 61% At a higher GHSV of 660,000 cm3/hr·gcat, 1.0 wt% boron reduces the activity loss from 70% to 30% A Temperature Programmed Oxidation and Scanning Electron Microscopy study of the catalysts confirmed that boron assists in preventing carbon buildup The theoretical predictions were experimentally validated, showing that boron promotion enhances the catalysts stability during steam reforming

VII

SYMBOLS AND ABBREVIATIONS

Symbols

)

,

( R x

)

(r

xc

ε Exchange-correlation energy per particle of the uniform electron gas

E Total energy of the system

b

edge

ee

E Electron-electron repulsion energy

Egraphene Total energy per carbon atom for the graphene-covered surface

)]

(

[ 0 r

E ncl ρ Non-classical contribution in electron-electron repulsion energy

ne

E Nucleus-electron interaction energy

)]

(

[ 0

0 r

)]

(

[ r

E XC ρ Exchange-correlation functional

)]

(

[ 0 r

J ρ Coulomb integral in electron-electron repulsion energy

i

R Intermediate states in NEB

1

)]

(

[ r

T ρ Kinetic energy functional

)

,

( R x

)

(r

V ext External potential

VIII

Abbreviations

CPO Catalytic partial oxidation

DFT Density functional theory

FFT Fast fourier transformations

FLAPW Full-potential linearised augmented plane-wave-method

HREELS High-resolution electron energy loss spectroscopy

ICP-OES Inductively coupled plasma-optical emission spectrometry

LDA Local density approximation

MEP Minimum energy path

NEB Nudged elastic band

PDOS Projected density of states

PES Potential energy surface

PAW Projector-augmented-wave

PES Potential energy surface

RMM Residual minimization method

SEM Scanning electron microscopy

SHSV Gas hourly space velocity

SPARG Sulfur passivated reforming

TCD Thermal Conductivity Detector

TPD Temperature Programmed Desorption

TOF Turnover frequency

TPO Temperature programmed oxidation

VASP Vienna ab initio simulation package

XRD X-Ray diffraction

IX

XPS X-ray photoelectron spectroscopy

X

LIST OF TABLES

Table 4.1 Carbon binding energies for the on-surface hcp hollow sites,

the octahedral sites for the first and second subsurface layer and for octahedral sites in the Ni bulk

54

Table 4.2 Influence of the unit cell size on the activation energy and

reaction energy for carbon atom diffusion from the on-surface fcc hollow site to the octahedral site below

67

Table 4.3 Carbon diffusion barriers (kJ/mol) as a function of the carbon

concentrations in the first and the second subsurface layer

75

Table 4.4 Boron binding energies for the on-surface hcp hollow sites, the

octahedral sites for the first and second subsurface layer and for octahedral sites in the Ni bulk

87

Table 4.5 Boron binding energies for different configurations of four

boron atoms in a p(2x2) unit cell

87

Table 5.1 Structure and binding energy for different surface structures

Table 5.2 Methyl and hydrogen binding energies (kJ/mol) for the clean

Ni(111) surface, the Ni(111)-CSS surface with subsurface carbon, and the Ni(111)-BSS surface with subsurface boron

110

Table 5.3 Transition state geometries and barriers for methane activation

on Ni(111), Ni(111)-CSS and Ni(111)-BSS surfaces 113

Table 5.4 Methyl and hydrogen binding energies (kJ/mol) for the

Ni(211) surface, the Ni(211) surface with all step sites occupied by carbon, Ni(211)-Cstep,100%, by boron,

Ni(211)-Bstep,100%, and with half of the step sites occupied by boron, Ni(211)-Bstep,50%

119

Table 5.5 Transition state geometries and barriers for methane activation

on Ni(211), Ni(211)-Cstep,100%, Ni(211)-Bstep,50% and

Ni(211)-Bstep,100% surfaces

121

Table 6.1 Bulk composition (ICP-OES), surface composition (XPS),

particle size (XRD) and dispersion for calcined 15 wt % Ni/γ- 137

XI

Al2O3 catalysts promoted with boron

XII

LIST OF FIGURES

Figure 2.1 Three approaches and examples for modeling

chemisorption and reactivity on surfaces (Left) cluster approach, maleic anhydride on Pd; (center) embedding scheme: ammonia adsorption in a zeolite cage; (right) periodic slab model: maleic anhydride adsorption on Pd(111)

7

Figure 2.2 Representative kinetic Monte Carlo simulation snapshot for

ethene hydrogenation over Pd

13

Figure 2.3 Overview of the different reaction paths for benzene

hydrogenation The dominant reaction path is indicated in boldface The hydrogenation activation energies for every step along the dominant reaction path are indicated The energy values are given in kJ/mol

13

Figure 2.4 Calculated turnover frequencies for ammonia synthesis as a

function of the adsorption energy of nitrogen for various transition metals and alloys

16

Figure 2.5 The calculaterd potential energy diagram for NH3 synthesis

from N2 and H2 over Ru(0001) (dashed curve) and stepped Ru(0001) (solid curve)

16

Figure 2.6 Energies for the species on Ni(211) and Ni(111) All

energies are relative to CH4 and H2O in the gas phase and calculated using the results for the individual species

20

Figure 2.7 Conversion of n-butane as a function of time during steam

reforming in a 3% n-butane-7% hydrogen-3% water in helium mixture at a space velocity of 1.2h-1 The dashed curve shows the n-butane conversion for the Ni and the solid curve is for the Au/Ni supported catalyst

20

Figure 3.1 Typical flow-chart of VASP for the self consistent

determination of the Kohn-Sham ground state

35

XIII

Figure 3.2 Schematic illustration of the nudged elastic band method

Starting from an initially guessed reaction path (dashed line) the chain converges to the nearest minimum path on the PES (full line)

37

Figure 4.1 Carbon binding energies for chemisorption at the four high

symmetry sites of the Ni(111) surface as a function of coverage The symbols indicate the calculated binding energies

51

Figure 4.2 Binding energies for the on-surface hcp hollow sites (∆)

and the octahedral sites of the first (□) and second (x) subsurface layer of the Ni(111) surface as a function of the coverage and the concentration The symbols indicate calculated binding energies

54

Figure 4.3 Binding energies per carbon atom for selected

configurations of four carbon atoms distributed over the

on-surface and subon-surface sites of a p(2x2) unit cell as a function

of the carbon concentration in the first subsurface layer The distribution between the first and second subsurface layer (●), between the on-surface fcc and the first subsurface layer (■) and between the on-surface hcp and the first subsurface layer (▲) are presented

58

Figure 4.4 Energy diagram for different distributions of four carbon

atoms over the on-surface hollow and the subsurface

octahedral sites of a p(2x2) unit cell Starting from four

on-surface carbon atoms, the system lowers its total energy by filling subsurface octahedral sites Solid lines indicate the thermodynamically preferred pathways

59

Figure 4.5 Possible high symmetry adsorption modes for a graphene

overlayer on a Ni(111) surface Carbon atoms are located at (A) both the fcc and hcp threefold hollow sites; (B) atop and fcc threefold hollow sites; (C) atop and hcp threefold hollow sites; (D) two near atop sites

62

XIV

Figure 4.6 The geometry of carbon at fivefold coordinated site on

Ni(211) surface Carbon atoms are represented by smaller balls and Ni atoms are represented larger balls Left panel, 50% carbon step coverage and right panel, 100% carbon step coverage

64

Figure 4.7 Calculation procedure for carbon diffusion barriers as a

function of on-surface coverage and of subsurface carbon concentration

70

Figure 4.8 Carbon diffusion barriers as a function of carbon

concentration in the first subsurface layer for low (1/9 ML,

∆), average (4/9 ML, ●) and high (1.0 ML, ◇ ) surface coverages

71

Figure 4.9 Diffusion barrier as a function of surface coverage for a

subsurface carbon concentration of 0%

73

Figure 4.10 Relationship between diffusion barriers and reaction

energies for carbon diffusion from on-surface fcc sites to octahedral sites in the first subsurface layer

73

Figure 4.11 Model graphene structures used to determine the energy

cost for creating small size graphene islands: (A) a single line structure; (B) a double line structure The white circle indicate unsaturated carbon atoms of the graphene structure, termed edge atoms

78

Figure 4.12 Model used to calculate the stability of small graphene

islands on the Ni(111) surface Black and grey circles indicate saturated, internal graphene atoms with a carbon binding energy of –760 kJ/mol, white circles indicate unsaturated edge atoms with a carbon binding energy of –635 kJ/mol

79

Figure 4.13 The total energy of cluster as a function of the number of

atoms in the cluster

79

XV