Characterization of bio-hybrid interfaces under ambient conditions [original]

Characterization of bio-hybrid interfaces under

ambient conditions

vorgelegt von

M. Sc.

Mais Jamil A. Ahmad

ORCID: 0009-0000-0746-6063

an 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. Birgit Kanngießer

Gutachter: Prof. Dr. Norbert Esser

Gutachter: Prof. Dr. Stefan Krischok

Gutachter: Dr. Roland Hergenröder

Gutachter: PD Dr. Patrick Vogt

Tag der wissenschaftlichen Aussprache: 06. März 2024

Berlin 2024

The following research work has been carried out during the period of my stay at the

Leibniz-Institut für Analytische Wissenschaften -ISAS – e.V., Dortmund, and

Technical University Berlin, Germany.

To my dearest and beloved Father and Mother

ii

Acknowledgements

This research which was a very rewarding experience yet a challenging one at the same time. I

would like to seize the opportunity on this splendid occasion to express my sincere appreciation

to all those who assisted and supported me in my research work. My special thanks are to:

My scientific supervisors, Prof. Dr. Norbert Esser and Dr. Roland Hergenröder for

their supervising, invaluable scientific guidance, advices, generous time, support, and giving me

the opportunity to work in their groups in addition to facilitating and financing several missions for

me to attend and contribute in different regional and international workshops, seminars

and conferences in the field of my research.

Prof. Dr. Mousa Jafar for his support and valuable pointers on this thesis. He was always

ready to help me and was looking forward to any scientific discussion.

Dr. Ahmad Telfah for the fruitful discussions, support, ideas, encouragement and constant

assistance.

Dr. Sabine Alamé, for the constructive discussions on different topics regarding this work and

for teaching me how to work on experimental equipment and UHV chambers during my stay in

TU Belin. Dr. Astrid Jürgensen for her kind help, support and helpful discussions on various

aspects of this work. Mr. Robert Zielinski for the tremendous technical support given to me and

his continuous help and for working with me on preparing and fixing the UHV systems during my

stay in TU Berlin. Dr. Maram Naes for her help with the submission of all forms and for always

being ready to help and explain things to me.

Workshop in Leibniz-Institut für Analytische Wissenschaften -ISAS – e.V., Dortmund for the

immense amount of technical support given to me and for constructing several items/apparatus

during the course of this research.

Mr. Hannes Raschke for his assistance and discussions on the apparatus and

instrumentation utilized in my work and experimental measurements, and for the precious time

iii

he spent on identifying the cause of problems in the experimental equipment and repairing the

required components and instrumentation in the UHV-XPS machine.

Mr. Ravi Prakash for his valuable assistance in the experimental work I have carried out,

fruitful discussion on the subject of interest and for the nice and pleasant working atmosphere, as

well as being a good friend.

For proof reading of this thesis, I want to thank Prof. Dr. Nobert Esser and Dr. Roland

Hergenröder, Prof. Dr. Mousa Jafar, Dr. Ahmad Telfah and Dr. Astrid Jürgensen for English

proofreading.

I am extremely grateful to my siblings Sadiqa, Esraa and Amin for always being there, for

their reassurance, encouragement and for boosting my self-confidence whenever it was low.

I would like to express my deepest and most sincere gratitude to my beloved and kind Father

Jamil and Mother Najah for their continuous great support and the encouragement they offered

me, and for believing in me throughout the duration of my stay in Germany. Very special thanks

to my father Jamil, I owe a lot of what I am to him. He is the one who encouraged me to be myself

and not follow the crowd; the one who taught me to strive to be the best in any anything I

undertake, the one who always pushed me to free my mind and follow my biggest dreams, and

to whom I dedicate this thesis....

iv

Table of Contents

Zusamenfassung/Abstract

xix

1

Introduction and Motivation

1

2

Fundamentals of Gallium Nitride

8

2.1

Structural Properties …………………………………………………………

8

2.2

Band Gap and Electronic Properties ………………………………………

11

2.3

Band Bending at Wurtzite GaN Interface …………………………………

12

2.4

Common Substrates for GaN Layers ……………………………………..

15

2.5

Growth Methods of GaN Layers ...………………………………………...

17

2.5.1

Hydride Vapor Phase Epitaxy (HVPE) Growth Method …………..…

18

2.5.2

Plasma Induced Molecular Beam Epitaxy (PIMBE)Growth Method..

18

2.5.3

Metal Organic Chemical Vapor Deposition(MOCVD)Growth Method

19

3

Surface Sensitive Characterization Techniques

20

3.1

X-ray Photoelectron Spectroscopy (XPS) …………………………………

20

3.1.1

The Three-Stage Photoelectron Model ………………………………..

21

3.1.2

Energy Level Scheme of Photoemission Spectra ……………………

23

3.1.3

Core-level Photoemission Spectroscopy ……………………………...

24

3.1.4

VB Spectroscopy and Surface Band Bending ……………………......

25

3.2

XPS Spectrometer …………………………………………………………..

25

3.2.1

X-ray Apparatus and Related Accessories ……………………………

26

3.2.2

Near Ambient Pressure NAP- XPS Spectroscopy ……….……….….

28

3.2.2.1

SPECS PHOIBOS 100 Hemispherical Energy Analyzer ……..

28

3.2.2.2

Ion Sputtering Gun of SPECS Phoibos100 XPS ………………

31

3.3

Low Energy Electron Diffraction (LEED) ………………………………….

32

3.4

Atomic Force Microscopy (AFM) …………………………………………..

33

4

Experimental Details

35

4.1

Materials and Chemicals ……………………………………………………

35

4.1.1

Samples …………………………………………………………………....

35

4.1.2

Gas Supply and Sample Mounting ……………………………………...

36

4.1.3

Molybdenum Sample Holder ………………………………………….…

37

4.2

Ex-/In-Situ Cleaning and Etching of GaN Surface ……………………….

37

4.2.1

Ex-Situ Surface Cleaning ……………………………………………….

38

4.2.2

In-Situ Heating of c-sapphire/n-GaN(0001) Samples ………………..

38

v

4.2.3

In-Situ Ion Sputtering and Etching of GaN Surface …………………..

38

4.3

In-Situ Functionalization of GaN Surface ………………………………….

40

4.3.1

Functionalization of GaN Surface with H2O Molecules ……………...

40

4.3.2

Functionalization of GaN Surface with L-Cysteine ………………..….

42

4.4

Characterization Procedures and Measurements ………………………..

47

4.4.1

Sample Alignment/Positioning in Working Chamber ………………....

48

4.4.2

Atomic Force Microscopic (AFM) Imaging ………………………...…..

48

4.4.3

Low Electron Energy Diffraction (LEED) Imaging ………………..…...

49

4.4.4

X-ray Photoelectron Spectroscopic Measurements …….………........ 49

4.4.4.1

UHV-XPS Measurements ………………………………………..

49

4.4.4.2 NAP-XPS Measurements ………………………………............

50

5 Experimental Results and Discussion

51

+

5.1

AFM Images and LEED Micrographs of As-received and 𝐍𝐍

𝟐𝟐

-Sputtered

n-GaN(0001) Layers ……………………………………………………..…. 56

5.1.1

In-situ UHV LEED Micrographs ………………………………………...

56

5.1.2

Ex-situ AFM Images ……………………………………………………..

58

5.2

Experimental UHV-XPS Spectra of As-received and N2

+-Sputtered n-

GaN(0001) Layers ………………………………………………..………… 60

5.2.1

Survey Spectra of the Photoemission Lines ………………………….

60

5.2.2

UHV-XPS Surface Photoemission Core-Level Lines ………………..

63

5.3

NAP-XPS Core-Level and Valence Band Spectra of N2

+-Sputtered n-

GaN(0001) Surface Exposed to H2O Molecules ………………………… 68

5.3.1

NAP-XPS Surface Photoemission Core-Level Lines ……….………..

70

5.3.1.1

O 1s Photoemission Line ………………………………...……...

71

5.3.1.2

Ga 3d Photoemission Line ………………………………………

75

5.3.1.3

Ga 2p and Ga 3p Photoemission Lines ………………………..

78

5.3.2

UHV and NAP-XPS Valence Band Emission Spectra ……………….

81

5.3.2.1

Evaluation of the Band Bending at the n-GaN(0001) Surface

83

5.4

Pressure and Temperature Dependencies of the Band Bending at the

N2

+-Sputtered n-GaN(0001)/H2O Interface ……………………………….. 86

5.5

Exploration of Bonding and Interactions on n-GaN(0001) Surface

Coated with L-cysteine Monolayers by UHV-XPS …………………...….. 94

vi

+

5.5.1

AFM Images and LEED Micrographs of the 𝐍𝐍

𝟐𝟐

-Sputtered n-GaN

(0001)/L-cysteine Interface ……………………………………………... 97

+

5.5.2

UHV-XPS Spectra of Inner Core-Level Photoemission Lines of 𝐍𝐍

𝟐𝟐

-

Sputtered n-GaN(0001)/L-cysteine Interface ………………………… 99

6

Conclusions and Further Suggestions

110

Bibliography

113

Appendix A: Fitting Parameters of Core-Levels

124

Appendix B: Wet Cleaning Method

134

vii

List of Acronyms

All acronyms, symbols and abbreviations of the methods, techniques used in this work

and/or mentioned above in the text will be explained below.

DOS

Density of States

FWHM Full Width at Half Maximum

LEED

Low Energy Electron Diffraction

AFM Atomic Force Microscopy

IMFP Inelastic Mean Free Pass of Photoelectrons

TPP-2M Tanuma, Powell and Penn general 2nd modified equation for IMFP of

Photoelectrons

SPM Scanning Probe Microscope

MBE Molecular Beam Epitaxy

ML

Monolayer

SAMs Self-Assembling Monolayers

CL Core-Level

IR Infrared

KE Kinetic Energy

BE Binding Energy

PES Photoemission (Photoelectron) Spectroscopy

UPS Ultraviolet Photoelectron Spectroscopy

XPS X-ray Photoelectron Spectroscopy

ESCA Electron Spectroscopy for Chemical Analysis, commonly used synonym for XPS

AES Auger Electron Spectroscopy

RAS Reflectance Analyzer Spectroscopy

NAP Near Ambient Pressure

MOVPE Metal Organic Vapor Phase Epitaxy

MOCVD Metal Organic Chemical Vapor Deposition

HVPE Hydride Vapor Phase Epitaxy

EEA Electron Energy Analyzer

HAS Hemispherical Analyzer

HEEA Hemispherical Electron Energy Analyzer

HEMTs High Electron Mobility Transistors

CRR

Constant Retardation Ratio

viii

MCP

Microchannel Plates

FAT

Fixed Analyzer Transmission

SEM

Scanning Electron Microscopy

STM

Scanning Tunneling Microscopy

TEM

Transmission Electron Microscopy

RMS

Root Mean Square

UHV

Ultra High Vacuum

VB

Valence Band

VBM

Valence Band Maximum

CBM

Conduction Band Minimum

BB

Band Bending

SPV

Surface Photovoltage

XRD

X-ray Diffraction

eV

Electron Volt

GaN

Gallium Nitride

InN

Indium Nitride

RT

Room Temperature

AlN

Aluminum Nitride

PIMBE

Plasma-Induced Molecular Beam Epitaxy

SiC

Silicon Carbide

WZ

Wurtzite

LDA

Local/Nonlocal Density Approximation

CB

Conduction Band

NE

Normal Emission

SXE

Soft X-ray Emission

PICS

Proportional to Photo-Ionization Cross Section

HV

High Voltage

ix

c

List of symbols

Lattice Constant (Å)

a

Lattice Constant (Å)

Bandgap Energy (300K) (eV)

Eg

Root Mean Square (roughness) (nm)

RMS

EC Conduction Band Minimum (eV)

EV Valence Band Maximum (eV)

T Temperature (C)

Free-Electron Mass

𝒎𝒎𝟎𝟎

Dielectric Constant

𝜺𝜺

𝜺𝜺𝟎𝟎 Vacuum Permittivity

𝜺𝜺∞ High-frequency Permittivity

Μ Mobility of Electrons (cm2/Vs)

∆𝑬𝑬𝐂𝐂 Conduction Band Offset at an Interface (eV)

𝑬𝑬𝐁𝐁𝐁𝐁 Effective Midgap Energy (eV)

EF Fermi Level (eV)

VN Nitrogen Vacancies

𝐁𝐁SP Spontaneous Polarization

𝜻𝜻 Conduction-Band Referenced Fermi Level

𝑬𝑬𝐤𝐤 Kinetic Energy (eV)

𝚽𝚽 Work Function (J)

𝜶𝜶(𝒉𝒉𝒉𝒉) Absorption Coefficient

𝝀𝝀𝐦𝐦𝐦𝐦(𝑬𝑬𝐊𝐊) Electron Inelastic Mean Free Path

𝑬𝑬𝒗𝒗𝒗𝒗𝒗𝒗 Vacuum Energy Level (eV)

𝑵𝑵𝓵𝓵(𝑬𝑬) Partial Density of States

𝓵𝓵 Atomic Orbital

𝚽𝚽𝐁𝐁 Surface Barrier Height Potential

𝒒𝒒 Particle Charge

∆𝑽𝑽 Potential Difference

ΔEan Energy Resolution (eV)

𝐤𝐤 Calibration Constant (𝑅𝑅in 𝑅𝑅out)⁄2 𝑅𝑅0(𝑅𝑅out − 𝑅𝑅in) = 0.

𝒘𝒘𝐗𝐗𝐁𝐁𝐗𝐗 Spectrometer Instrumental Function

𝒘𝒘𝐬𝐬 Gaussian Function

x

𝒘𝒘𝐱𝐱 Gaussian Profile

𝐍𝐍𝐀𝐀 Avogadro Constant (6.022 × 10²³)

𝒆𝒆 Electron Charge (1.602*1019 As)

𝒁𝒁/𝒕𝒕 Sputter Rate (nm/s)

𝐣𝐣𝐦𝐦 Ion Current Density (A/m2)

𝒕𝒕 Time (s)

𝑴𝑴 Molar Mass (kg/mol)

𝝆𝝆𝒎𝒎 Density of The Target Material (kg/m3)

𝒑𝒑 Electron’s Momentum (eV·c−1)

𝒎𝒎𝐞𝐞 Electron’s Mass (9.1093837 × 10-31 kg)

𝝀𝝀𝐞𝐞 de Broglie Wavelength (nm)

𝒏𝒏 (Bulk) Electron Concentration (cm-3)

𝐃𝐃𝐦𝐦 Detector Pressure (mbar)

𝑻𝑻𝐀𝐀 Annealing Temperature (CO)

𝒕𝒕𝐀𝐀 Annealing Time (min)

𝑰𝑰𝐋𝐋 Laser Current (A)

𝑬𝑬𝐕𝐕𝐗𝐗 Surface Valence Band Edge (eV)

𝑽𝑽𝐛𝐛𝐛𝐛 Band Bending Voltage (eV)

𝑬𝑬𝐕𝐕𝐁𝐁 Valence Energy Band Edge of Bulk

𝑬𝑬𝐂𝐂𝐁𝐁 Conduction Energy Band Edge of Bulk

𝑬𝑬𝐅𝐅𝐁𝐁 Fermi Level of Bulk

𝑬𝑬𝐅𝐅𝐗𝐗 Fermi Level at Surface

𝒉𝒉 Planck’s constant (6.62607015 × 10-34 m2 kg / s)

∗

𝒎𝒎𝒆𝒆 Effective Mass

𝑵𝑵𝐜𝐜 Effective density of States

𝜸𝜸 , 𝜷𝜷 Varshni’s Fitting Parameters

𝑵𝑵𝐝𝐝 Donor Concentration

xi

List of Figures

Figure No

Page

Figure 2.1

WZ-GaN showing the a) coordination and diatomic close-packed

alternative planes of cations (Ga3+) and anions (N3−), after R. Bouveyron

[17] and b) conventions for Ga-faced and N-faced polarities along 〈0001〉

and 〈0001

�〉 directions, respectively, after O. Ambacher [40].

9

Figure 2.2

Typical XPS VB spectra of WZ-GaN a) N-polar GaN layer, b) In-/N-polar

InN layer and c) Ga-polar GaN layer taken under various preparation

and experimentation conditions [44, 121].

10

Figure 2.3

Energy band structure of WZ-GaN calculated by empirical pseudo

potential method [123]. 11

Figure 2.4

GaN material layers in wurtzite structure.

13

Figure 2.5

Tetrahedral bond configuration, individual dipole moments

spontaneous polarization for III-nitrides with wurtzite structure.

and

14

Figure 2.6

Schematic diagram of surface band bending and charge distribution of

Ga-face and N-face WZ-GaN

layer. The number of different charges

shown on the diagram is not to scale [130].

14

Figure 3.1

A three-stage model for the photoemission process [145]: (a) a

schematic picture of the three processes: 1) photoexcitation of electrons, 21

2) electron transport towards surface and 3) transmission through

surface into vacuum, and (b) a diagram showing these processes on

measured spectra [44].

Figure 3.2

(a) Schematic layout the angle Θ between the normal to sample’s surface

and the entrance to analyzing chamber (b) IMFP of electrons in various

materials with solid dots and stars are experimental data [44] and the

22

dashed line represents the theoretical function of Seah and Dench [154

].

Figure 3.3

(a) A diagram of energy states of atomic CLs and VB in a solid of work

function ΦS (= Φ), after Hüfner [144a] and 𝐸𝐸kin (= 𝐸𝐸k) of their electrons 24

produced by X-ray photons of energy ℏ𝜔𝜔 = ℎ𝑣𝑣, where ℏ =

ℎ⁄2𝜋𝜋 and 𝜔𝜔 = 2𝜋𝜋𝜋𝜋, and (b)

scheme of relevant energy terms used in

XPS spectra of solid surfaces, after Hofman [144b].

Figure 3.4

A photograph of NAP-XPS spectrometer used in present work showing

its main parts of SPECS Phoibos 100 he

mispherical energy analyzer

26

and X-ray apparatus.

Figure 3.5

Schematic diagram of the ellipsoidal crystal monochromator FOCUS 500

of the SPECS XR-500 microfocus X-ray assembly.

27

xii

Figure 3.6

A diagram [SPECS catalogues] of lens system, hemispherical capacitor

and voltage supplies of SPECS PHOIBOS 100 HEEA, with r0 = 100 mm

= R0 cited in text.

29

Figure 3.7

LEED setup with its main parts and working principle [43].

33

Figure 3.8

(a) A photograph of used AFM XE-100 SPM instrument and (b) a

schematic

diagram showing the operational principle of conventional

AFM apparatus based on a piezoelectric ceramic tube scanner.

34

Figure 4.1

The Mo holder annotated diagram shows the c-sapphire/n-GaN(0001)

Sample, spacer (Mo foil) fixed under the sample, main holder frame,

Fixer, PT100 sensor with connections, and the heat shield.

37

Figure 4.2

A plot of measured temperature 𝑇𝑇(℃) of GaN-Lcys-2 sample as a

function of the electric current 𝐼𝐼L(A) of the IRLH 150 laser heater up to

the annealing temperatures 𝑇𝑇A = 75 ℃, 100 ℃, and 130 ℃.

46

Figure 4.3

Temperature 𝑇𝑇(℃) of GaN-Lcys-3 sample versus electric current 𝐼𝐼

L

(A)

of IRLH 150 laser heater up to final annealing temperatures 𝑇𝑇A = 100 ℃,

130 ℃, and 150 ℃, each for 𝑇𝑇A equals to 30 minutes.

47

Figure 5.1

LEED patterns of the n-GaN(0001) surface of an as-received sample cut

by a fine laser beam. 57

Figure 5.2

LEED patterns of the n-GaN(0001) surface sputtered by 1-keV N

2

+ ions

for (a) 9 minutes and (b) 13 minutes. The LEED image shown in part (c)

is for 9-min, 1-keV N2

+-sputtered n-GaN(0001) surface after exposure to

H2O molecules at RT (23 °C) and an H2O-vapor pressure of 1 mbar.

57

Figure 5.3

Typical ex-situ AFM images measured on (2.5 × 2.5) µm2 spots on the

n-GaN(0001) surface (a) of an as-received sample and (b) after

sputtering the n-GaN(0001) surface with 1-keV N2

+ ions for 9 minutes.

59

Figure 5.4

AFM images measured on (2.5 × 2.5) µm2 spots on a 9 min, 1-keV N

2

+-

sputtered n-GaN(0001) surface exposed to 1-mbar H2O pressure at (a)

RT and (b) 60 oC and finally 160 oC.

59

Figure 5.5

Survey UHV-XPS spectra of the n-GaN(0001) surface taken at normal

emission (NE) with the monochromatized Al Kα1 X-ray line (1486.7 eV)

as excitation energy for a) as-received and b) 9-min, 1-keV N2

+-sputtered

n-

GaN(0001) layers. The photoemission lines are labeled with their

individual CL transitions.

61

Figure 5.6

The 20 eV pass energy UHV-XPS spectra of the C 1s CL transition for

the n-GaN(0001) surface of an as-received sample and for the 9-min, 1-

keV N2

+-sputtered n-GaN(0001) surface. The strong peak at the BE =

286.5 eV is related to C-C bonds and the weak peak at the BE around

290 eV can be allocated to C=O bonds [82c].

64

xiii

Figure 5.7

The 20 eV pass energy UHV-XPS spectra of the O 1s CL transition for

the n-GaN(0001) surface of an as-received sample and for the 9-min, 1-

keV N2

+-sputtered n-GaN(0001) surface. The fitted peaks corresponding

to the bonding states of existing oxides of gallium and carbon are shown

on the the measured spectra.

65

Figure 5.8

The 20 eV pass energy UHV-XPS spectra of the Ga 2p

3/2

CL transition

for the n-GaN(0001) surface sputtered with 9-min, 1-keV N2

+ ions. 66

Figure 5.9

The 20 eV pass energy UHV-XPS spectra of the Ga 3d

5/2,3/2

doublet CL

transition for the n-GaN(0001) surface after sputtering with 9-min, 1-keV

N2

+ ions.

66

Figure 5.10

The 20 eV pass energy UHV-XPS spectra of the N 1s CL transition for

the 9-min, 1-keV N2

+-sputtered n-GaN(0001) surface. The fitted main N

1s peaks and the overlapped Ga LMM sub-peaks are shown on the plot.

67

Figure 5.11

Measured NAP-XPS spectra for the O 1s CL transition of the N

2

+-

sputtered n-GaN(0001)/H2O interface of different samples exposed to 1-

mbar H2O molecules for a fresh sample at (a) 23 oC, for another new,

fresh sample at (b) 60 oC, and for a third fresh sample at (c) 160 oC.

71

Figure 5.12

Bond contributions to the O 1s line of the N

2

+-sputtered n-

GaN(0001)/H2O interface with varying H2O pressure at (a) 23 oC, (b) 60

oC, and (c) 160 oC, for different samples as explained in Figure 5.11. The

abbreviations UHV2 and UHV3 correspond to data taken at 23 °C in

vacuum for the N2

+-sputtered n-GaN(0001) surface before exposure to

H2O molecules and after the end of the NAP-XPS run on the sample,

respectively.

72

Figure 5.13

Bond contributions to the O 1s photoemission line of the N

2

+-sputtered n-

GaN(0001)/H2O interface of the same sample held at (a) 23 oC, (b) 60

oC, and (c) 160 oC in succession, while the NAP-XPS measurements

were taken at different H2O pressures.

73

Figure 5.14

(a) UHV-XPS spectrum for the Ga 3d CL of the N

2

+-sputtered n-

GaN(0001) surface, and the NAP-XPS spectra after injection of 1 mbar

of H2O molecules at varying sample temperatures of: (b) 23 oC, (c) 60

oC, and (d) 160 oC. Three different freshly N2

+sputtered c-sapphire/n-

GaN(0001) samples were used in different runs one at 23 oC, the second

at 60 oC, and the third at 160 oC. The insets in a) is a magnification of

the Ga-

metal and N 2s bond contributions, while the inset in (b) is

illustrative of the Ga-OH bonds at RT. It is also noted that intial Ga-oxides

exist on the surface even before functionalization of the surface with H2O

+

molecules, indicating that sputtering the n-GaN(0001) surface with N2

ions is not effective in this respect.

76

Figure 5.15

The XPS spectrum of the Ga 2p

3/2

peak of the N

2

+-sputtered n-

GaN(0001) surface in UHV and after exposure of three different freshly

N2

+

sputtered c-sapphire/n-GaN(0001) samples to H2O molecules at a

79

xiv

pressure of 0,02 mbar, one maintained at (a) 23 oC, the second one at

(b) 60 oC, and the last one at (c) 160 oC.

Figure 5.16

The O/N-ratio estimated from the quantitative analysis of the Ga 2p

3/2

photoemission line of the N2

+-sputtered n-GaN(0001)/H2

O interface of

different fresh samples as a function of (a) the H2O vapor pressure (𝑝𝑝)

at various sample temperatures and (b) the sample temperature (𝑇𝑇) at

various H2O vapor pressures.

80

Figure 5.17

The O/N-ratio versus H

2

O vapor pressure (𝑝𝑝), estimated from a

quantitative analysis of the (a) Ga 2p3/2 and (b) Ga 3d5/2 photoemission

lines of the N2

+-sputtered n-GaN(0001)/H2O interface at some

temperatures.

80

Figure 5.18

Data (open symbols) of (a) the UHV-XPS VB spectrum of the N

2

+-

sputtered n-GaN(0001) surface and (b) the NAP-XPS VB spectrum of

the N2

+-sputtered n-GaN(0001)/H2O interface. Solid lines (red) are linear

fits of the VB leading edge, extrapolated to meet the baseline (green) at

the VBM, below the Fermi level 𝐸𝐸F.

82

Figure 5.19

A schematic diagram for the evaluation of the band bending (BB) in the

vicinity of the n-GaN(0001) surface on the basis of Equations (5.1a) and

(5.1b) cited in the text. The symbols 𝐸𝐸FB, 𝐸𝐸g, and 𝐸𝐸CB denote the bulk

Fermi level, band gap energy, and CB edge, respectively. 𝐸𝐸VB and 𝐸𝐸VS

are the bulk and surface VB edges, respectively, while EGa3dB and EGa3dS

are the positions of the bulk Ga 3dB and surface Ga 3dS core levels,

respectively. The energy separation between the 𝐸𝐸CB edge and 𝐸𝐸FB of

bulk n-GaN(0001) is (𝐸𝐸CB − 𝐸𝐸FB), given in Equation (5.2)

in the text.

(𝐸𝐸VB − EGa3dB) is the energy difference between the 𝐸𝐸VB edge and the

Ga 3dB CL in bulk n-GaN(0001), which is taken as 17.5 eV, while (𝐸𝐸FS −

EGa3dS) is the energy separation between 𝐸𝐸FS and 𝐸𝐸Ga3dS at the sample

surface. The dependence of 𝐸𝐸g on 𝑇𝑇 can be found from Equation (5.3).

85

Figure 5.20

𝐸𝐸

g

(𝑇𝑇) of n-GaN(0001) as a function of 𝑇𝑇 on the basis of the Varshni

expression in Equation (5.3). The 𝐸𝐸g(𝑇𝑇)-

values at the temperatures

studied are shown (open symbols) with their Spline-fit curves obtained

using the Varshni fitting parameters given in [180]. The inset shows the

𝑇𝑇-dependency of the Fermi energy, 𝐸𝐸FB

, relative to the CB edge, i.e.

−(𝐸𝐸CB − 𝐸𝐸FB), w

ith the symbols being the values calculated from

Equation (5.2), and the solid line is a guide for the eye.

87

Figure 5.21

The band bending voltage 𝑉𝑉

bb

(open symbols) found from Equation

(5.1a) for the N2

+-sputtered n-GaN(0001) surface and the N2

+-sputtered

n-GaN(0001)/H2O interface of different fresh samples plotted as (a)

𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-p(mbar) at different temperatures and (b) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-T(oC)

at different H2O vapor pressures. The solid lines are guides for the eye.

90

Figure 5.22

The band bending 𝑉𝑉

bb

deduced from Equation (5.1b) for the N

2

+-

sputtered n-GaN(0001) surface and the N2

+-sputtered n-GaN(0001)/H2O

interface of different fresh samples plotted as (a) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-p(mbar) at

90

xv

a given temperature and (b) 𝑉𝑉

bb

(𝑒𝑒𝑉𝑉)-vs-T(oC) at various H

2

O vapor

pressures. The solid lines are guide for the eye.

Figure 5.23

The band bending 𝑉𝑉

bb

deduced from Equation (5.1b) for the N

2

+-

sputtered n-GaN(0001)/H2

O interface of the same sample used at all

temperatures and under the same pressures plotted as (a) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-

p(mbar) for various temperatures and (b) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-T(oC) under various

H2O vapor pressures. The solid lines are guidesfor the eye.

92

Figure 5.24

The band bending 𝑉𝑉

bb

deduced from Equation (5.1a) for the N

2

+-

sputtered n-GaN(0001)/H2

O interface of the same sample used at all

temperatures and under the same pressures plotted as (a) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-

p(mbar) at the given temperatures and (b) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-T(o

C) under

various H2O vapor pressures. The solid lines are guides for the eye.

92

Figure 5.25

LEED images taken with 120 eV electrons on the surface of (a) sputtered

by 1-keV N2

+ ions for 9 minutes and (b) the n-GaN(0001) layer coated

with L-cysteine at 105 oC for 40 minutes, and (c) the n-GaN(0001)/L-

cysteine sample after thermal annealing at 75 oC, then at 100 oC, and

finally at 130 oC with each temperature kept constant for 30 minutes.

98

Figure 5.26

An AFM image of an N

2

+-sputtered n-GaN(0001) layer coated with L-

cysteine at 105 oC for 40 minutes. 98

Figure 5.27

Experimental UHV-XPS spectra of the Ga 3s / S 2p photoemission lines

+

for the n-GaN(0001) surface (a) after sputtering with 9-min, 1-keV N2

ions and after depositing L-cysteine molecules at 105 oC for (b) 20

minutes and (c)

40 minutes with the samples being kept at room

temperature (RT) until the XPS measurements were collected. The

colored curves are the deconvoluted fits of the Ga 3s and S 2p CL lines

as labeled inside the plots.

100

Figure 5.28

Experimental UHV-XPS spectra of the N 1s photoemission lines for the

n-GaN(0001) surface (a) after sputtering with 9-min, 1-keV N2

+ ions and

after depositing L-cysteine molecules at 105 oC for (b) 20 minutes and

(c) 40 minutes, with the samples being kept at room temperature (RT)

until the XPS measurements were collected. The colored curves are the

deconvoluted fits of the overlapped Ga LMM sub-peaks and of the peaks

of the uncharged/charged amino and carboxylic acid groups as labeled

inside the plots.

102

Figure 5.29

Measured XPS photoemission lines of (a) C 1s and (b) O 1s core-level

transitions for the N2

+-sputtered n-GaN(0001)/L-

cysteine interface

coated with L-cysteine at 105 oC for 40 minutes, with the samples being

kept at RT until collecting the XPS measurements. The colored curves

are the deconvoluted fits of the respective sub-peaks as seen from the

plots.

103

Figure 5.30

Experimental UHV-XPS spectra of the Ga 3s / S 2p photoemission lines

for the N2

+

-sputtered n-GaN(0001)/L-cysteine interface obtained by

105

xvi

depositing L-cysteine at 105 oC for 40 minutes and after

heating/annealing the sample for 30 minutes at (a) 100 oC, (b) 130 oC,

and (c) 150 oC before collecting the XPS measurements. The colored

curves are the deconvoluted fits of the Ga 3s and S 2p CL lines as

labeled inside the plots.

Figure 5.31

Experimental UHV-XPS spectra of the N 1s photoemission line for the

N2

+-sputtered n-GaN(0001)/L-cysteine interface obtained by depositing

L-cysteine at 105 o

C for 40 minutes and after heating/annealing the

sample for 30 minutes at (a) 100 oC, (b) 130 oC, and (c) 150 oC before

collecting the XPS measurements. The colored cu

rves are the

deconvoluted fits of the N 1s line as labeled inside the plots.

106

Figure 5.32

Experimental UHV-XPS spectra of the C 1s photoemission line for the

N2

+-sputtered n-GaN(0001)/L-cysteine interface obtained by depositing

L-cysteine at 105 oC

for 40 minutes and after heating/annealing the

sample for 30 minutes at (a) 100 oC, (b) 130 oC, and (c) 150 oC before

collecting the XPS measurements.

The colored curves are the

deconvoluted fits of the C 1s line as labeled inside the plots.

107

Figure 5.33

Experimental UHV-XPS spectra of the O 1s photoemission line for the

N2

+-sputtered n-GaN(0001)/L-cysteine interface obtained by depositing

L-cysteine at 105 o

C for 40 minutes and after heating/annealing the

sample for 30 minutes at (a) 100 oC, (b) 130 oC, and (c) 150 oC before

collecting the XPS measurements. The colored curves are deconvoluted

fits of the O 1s line as labeled inside the plots.

108

xvii

List of Tables

Table No Page

Table 3.1

SPECS crystal monochromator key parameters for characteristics x-

27

rays of Al Kα1.

Table 4.1

Preparation of typical c-sapphire/n-GaN(0001) (SurfaceNet, Berlin)

39

samples Under various conditions.

Table 4.2

The voltage, emission current, power of x-ray source, and alignment

41

setting of the GaN surface of c-sapphire/n-GaN(

0001) samples

described in Table 4.1 for functionalization with H2O vapor of various

pressures at different sample temperatures. The symbol Dp designates

the pressure of

the detector (analyzer) chamber, whereas the scale

reading is the scale number of a water pump.

Table 4.3

Preparation and XPS setting details of the samples GaN-AsRec-2,

43

N2

+-sputtered GaN-Sput-2, GaN-AsRec-3, and N2

+-sputtered GaN-Sput-

3 before implementing L-cysteine functionalization on them.

Table 4.4

Sequential heating stages and annealing procedures carried out on

45

the L-cysteine functionalized GaN-Lcys-2 sample at temperatures 𝑇𝑇A =

75 ℃, 100 ℃, and 130 ℃, each for 𝑡𝑡A = 15, 30, and 60 minutes.

Table 4.5

Temperature 𝑇𝑇(℃) of L-cysteine functionalized GaN-Lcys-3 sample as

46

a function of current 𝐼𝐼(A) of IRLH 150 heater up to 𝑇𝑇A = 100 ℃, 130 ℃

and 150 ℃, for the same annealing time 𝑡𝑡A = 30 minutes.

xviii

Characterization of bio-hybrid interfaces under ambient conditions

Mais Jamil A. Ahmad

Abstract

The topography, morphology, and chemistry of the surface of HVPE-grown n-GaN(0001) layers

on c-sapphire/n-GaN(0001) samples were explored under various conditions before and after

+

sputtering with 1 keV nitrogen N2 ions, followed by the functionalization with deionized water

(H2O) and neutral L-cysteine (C3H7NO2S) molecules. Atomic force microscopy (AFM), low energy

electron diffraction (LEED), and X-ray photoelectron spectroscopy (XPS) in ultra-high vacuum

(UHV) and near-ambient-pressure (NAP) conditions were used for this investigation.

Monochromatic X-ray 𝐴𝐴𝐴𝐴 𝐾𝐾𝐾𝐾 radi ation (ℎ𝜋𝜋 = 1486.7𝑒𝑒𝑉𝑉) was the exc itat ion source for the XPS

experiments.

The results obtained yielded information on the modifications occurring at the surface of

the n-GaN(0001) layers upon N2

+-sputtering and on the contact formation and interfacial chemical

reactivity after exposure to H2O and L-cysteine, i.e., on the ensuing interactions and bonding

mechanisms between H2O and L-cysteine molecules and the N2

+-sputtered n-GaN(0001) surface.

This provided a clue on the suitability, stability, and functionality of GaN-OH and GaN-SH groups,

the primary links between the surface and inorganic, organic and biomolecules, and hence, on

the potentiality of using GaN-based surfaces for electronic devices and biosensors.

Experimental UHV-XPS spectra of C1s and O1s core-level (CL) transitions of the as-

grown n-GaN(0001) surface revealed adsorbed adventitious carbon (C) and oxygen (O2) species,

besides traces of Ga2O3 and Ga2Ox oxides. The surface AFM images and LEED patterns revealed

that N2

+-ion sputtering resulted in minimal surface damage and roughness, in addition to largely

reducing the amount of adsorbed carbon species. The NAP-XPS spectra of the Ga 2p3/2 and Ga

3d5/2 photoemission lines of the N2

+-sputtered n-GaN(0001)/H2O interface showed that the H2O

molecules adsorbed on the n-GaN(0001) surface likely dissociated into a hydroxyl ion (OH−) and

a proton (H+), which tend to combine with surface gallium (Ga+) dangling bonds and unbonded

nitrogen (N−) ions, respectively.

The N2

+-ion sputtering and H2O exposure modified the valence band (VB) structure of the

as-grown n-GaN(0001) surface. This is shown by the respective UHV and NAP-XPS VB

photoemission spectra, which displayed VB peaks related to the hybridized orbital states of a

xix

+-

sputtered n-GaN(0001) surface. Functionalization of the N2

+-sputtered n-GaN(0001) surface with

H2O molecules induced a noticeable modification to its chemistry and band bending (BB),

disclosed from the measured NAP-XPS O 1s and Ga 3d lines, and from the NAP-XPS VB spectra

of the ensuing N2

+-sputtered n-GaN(0001)/H2O interface. The formation of Ga-OH and N-H bonds,

and also Ga-O bonds that become prominent at high temperatures, are inferred from the NAP-

XPS peaks of the existing surface constituents. Such a compensation of dangling bonds and

surface states upon increasing the amount of H2O exposure typified itself as a continuous

decrease in the surface BB voltage 𝑉𝑉bb up to a pressure 𝑝𝑝 = 0.1 mbar, above which 𝑉𝑉bb leveled

off at a constant value, depending on the sample temperature 𝑇𝑇.

polar surface dominated by Ga, besides revealing a reduction in the band bending at the N2

The surface BB voltage 𝑉𝑉𝑏𝑏𝑏𝑏 was estimated by two methodologies. One approach used the

energy difference (𝐸𝐸FS − 𝐸𝐸VS) between the surface valence-band edge (𝐸𝐸VS) and the Fermi level

(𝐸𝐸FS), the VBM energy, by the linear extrapolation method. The other approach monitored the

measured peak energy of the Ga 3dS line of the N2

+-sputtered n-GaN(0001) surface and the N2

+-

sputtered n-GaN(0001)/H2O interface viz., (𝐸𝐸FS − 𝐸𝐸Ga3dS), combined with 17.5 eV, the energy

difference between VB edge and Ga 3dB CL of bulk GaN. The two approaches incorporated the

energy differences between the conduction band (CB) edge (𝐸𝐸CB) and the Fermi energy 𝐸𝐸𝐹𝐹𝐹𝐹 of

bulk GaN (𝐸𝐸CB − 𝐸𝐸FB) and between 𝐸𝐸CB and 𝐸𝐸VB, the band gap energy 𝐸𝐸g, both of which are

temperature dependent. This partially explains the observed decrease in 𝑉𝑉bb with increasing 𝑇𝑇.

+-

sputtered n-GaN(0001)/H2O interface, mainly at high H2O pressure, 𝑉𝑉bb was estimated from the

binding energy shifts of (𝐸𝐸FS − 𝐸𝐸Ga3dS), combined with (𝐸𝐸VB − 𝐸𝐸Ga3dB) = 17.5𝑒𝑒𝑉𝑉, and (𝐸𝐸FS − 𝐸𝐸VS),

the VBM energy. At a constant 𝑇𝑇, the decrease in 𝑉𝑉bb at the N2

+-sputtered n-GaN(0001)/H2O

interface with pressure 𝑝𝑝 below 0.1 mbar is ascribed to the compensation of surface dangling

bonds by dissociative H2O molecules. The constancy of 𝑉𝑉bb for 𝑝𝑝 > 0.1 mbar is understood in

terms of a saturation coverage of the n-GaN(0001) surface with H2O molecules. The reduction of

𝑝𝑝 > 0.1 𝑉𝑉bb with increasing 𝑇𝑇 can be partly accounted for by the decrease in 𝐸𝐸g(𝑇𝑇)=(𝐸𝐸CB − 𝐸𝐸VB)

and �𝐸𝐸CB − 𝐸𝐸FB(𝑇𝑇)� with 𝑇𝑇, though such a decrease cannot solely clarify the entire trend of 𝑉𝑉bb at

high 𝑇𝑇.

Due to the difficulty of finding a proper linear extrapolation for the VB spectra of the N2

The N2

+-sputtered n-GaN(0001) surface permits notable bonding of the sulfur-containing

amino-acid L-cysteine (L-CySH) molecules, as concluded from the respective UHV-XPS spectra.

Deposition of L-CySH on the n-GaN(0001) surface was made via sublimation of L-CySH powder

xx

at 105 oC for ≤ 40 min. The UHV-XPS spectra of the N2

+-sputtered GaN(0001)/L-cysteine interface

were taken at room temperature and after annealing at various temperatures (𝑇𝑇𝐴𝐴 = 75 − 150℃)

for annealing times (𝑡𝑡A) of 15, 30, and 60 minutes.

The UHV-XPS spectra of the N2

+-sputtered n-GaN(0001)/L-cysteine interface showed that

the thiol head group (SH group) of the L-cysteine molecule formed strong chemical bonding with

the surface metallic Ga. Such behavior was noted from the overlapped Ga 3s and S 2p XPS

photoemission lines, the intensity of which were enhanced with more deposited L-cysteine layers,

a feature that is characteristic for thiolate species. The bonding of carboxyl (COOH) and amino

(NH2) groups to the surface cannot be entirely excluded, as was noted from the respective C 1s,

O 1s, and N 1s XPS spectra. The strength and appearance of the doublet S 2p photoemission

lines diminished as 𝑇𝑇A increased, suggesting that most but not all L-cysteine layers re-evaporated

from the functionalized surface, with the SH-group of the lowermost L-cysteine layer being still

chemically bonded to the Ga atoms at the surface.

xxi

Charakterisierung von Bio-Hybrid Grenzflächen unter

Umgebungsbedingungen

Mais Jamil A. Ahmad

Zusammenfassung

Die Topographie, Morphologie und Chemie der Oberfläche von HVPE-gewachsenen n-

GaN(0001) Schichten auf c-Saphir/n-GaN(0001) Proben wurden vor und nach dem Sputtern mit

1 keV Stickstoffionen (N2

+) und der Funktionalisierung mit entionisiertem Wasser (H2O) und

neutralen L-Cystein Molekülen (C3H7NO2S) unter verschiedenen Bedingungen untersucht. Für

diese Untersuchung wurden die Rasterkraftmikroskopie (AFM), die

Niederenergieelektronenbeugung (LEED) und die Röntgenphotoelektronenspektroskopie (XPS)

im Ultrahochvakuum (UHV) und in einer niedrig-Druck Atmosphäre (Near-Ambient-Pressure,

NAP) angewendet. Die monochromatische Al-Kα Röntgenstrahlung (ℎ𝜋𝜋 = 1486.7𝑒𝑒𝑉𝑉) wurde bei

den XPS Experimenten als Anregungsquelle eingesetzt.

Die Ergebnisse lieferten Information über die durch das Sputtern verursachten Modifikationen an

der Oberfläche der n-GaN(0001) Schichten, über die Kontaktbildung und die chemische

Reaktivität an der Grenzfläche nach Einwirkung von H2O und L-Cystein, sowie zu die daraus

resultierenden Wechselwirkungen und Bindungsmechanismen zwischen H2O und L-Cystein

Molekülen und der gesputterten n-GaN(0001) Oberfläche. Dies lieferte einen Hinweis für die

Eignung, Stabilität und Funktionalität von GaN-OH und GaN-SH Gruppen, den primären

Verbindungen zwischen der Oberfläche und anorganischen organischen und Biomolekülen, und

daher auch die Möglichkeit, GaN-basierte Oberflächen für elektronische Geräte und Biosensoren

zu verwenden.

Experimentelle UHV-XPS Spektren der Kernschalenübergänge C 1s und O 1s der unbehandelten

n-GaN(0001) Oberfläche zeigten adsorbierte unspezifische Kohlenstoff-(C) und

Sauerstoffspezies (O2) sowie Spuren von Ga2O3 und Ga2Ox Oxiden. Die AFM Bilder und LEED

+

Muster der Oberfläche zeigten, dass das Sputtern mit 𝑁𝑁2 I onen n ur m inimale

Oberflächenschäden und Rauheiten erzeugte, und zusätzlich die Anzahl der adsorbierten

Kohlenstoffspezies deutlich verringerte. Die NAP-XPS Spektren der Ga 2p3/2 und Ga 3d5/2

Photoemissionslinien der gesputterten n-GaN(0001)/H2O Grenzfläche zeigten, dass auf der n-

GaN(0001) Oberfläche adsorbierte H2O Moleküle wahrscheinlich in ein Hydroxyl Ion (OH-) und

xxii

Proton (H+) dissoziierten, Ionen, die dazu neigen, sich mit ungesättigten Gallium (Ga+) Bindungen

bzw. ungebundenen Stickstoff (N−) Ionen an der Oberfläche zu verbinden.

+

Das Sputtern mit N2 Ionen und die Abdeckung der Oberfläche durch H2O veränderten die

Valenzbandstruktur (VB) der n-GaN(0001) Oberfläche, wie sie aus den jeweiligen UHV-und NAP-

XPS VB Photoemissionsspektren hervorgeht. Die VB Peaks im Zusammenhang mit

hybridisierten Orbitalzuständen zeigten eine polare, von Ga dominierte Oberfläche. Außerdem

+

zeigt sich eine Verringerung der Bandbiegung (BB) an der mit N2 gesputterten n-GaN(0001)

Oberfläche. Die Funktionalisierung der gesputterten n-GaN(0001) Oberfläche mit H2O Molekülen

führte zu einer merklichen Veränderung ihrer Chemie und Bandbiegung (BB), wie aus den

gemessenen NAP-XPS O 1s und Ga 3d Linien und den NAP-XPS VB Spektren der anschließend

gesputterten n-GaN(0001)/H2O Oberfläche hervorgeht. Die Bildung von Ga-OH und N-H

Bindungen sowie von Ga-O Bindungen, die bei hohen Temperaturen hervortreten, sind

Merkmale, die aus den NAP-XPS Peaks vorhandener Oberflächenbestandteile abgeleitet

werden. Eine solche Kompensation von freien Bindungen und Oberflächenzuständen bei

zunehmender H2O Exposition äußerte sich als kontinuierliche Abnahme der BB Spannung der

Oberfläche (𝑉𝑉bb) bis zu einem Druck 𝑝𝑝 = 0.1 mbar, oberhalb derer sie sich, abhängig von der

Probentemperatur (𝑇𝑇), auf einem konstanten Wert einpendelte.

Die BB Spannung der Oberfläche 𝑉𝑉𝑏𝑏𝑏𝑏 wurde mit zwei Methoden abgeschätzt. Ein Ansatz nutzte

die Energiedifferenz (𝐸𝐸FS − 𝐸𝐸VS) zwischen der Oberflächenvalenzbandkante (𝐸𝐸VS) und dem

Fermi-Niveau (𝐸𝐸FS), die VBM Energie, durch die lineare Extrapolationsmethode. Der andere

alternative Ansatz beruht auf der gemessenen Übergangsenergie der Ga 3dS Linie der

+

gesputterten n-GaN(0001) Oberfläche und der mit N2 gesputterten n-GaN(0001)/H2O

Grenzfläche, nämlich (𝐸𝐸FS − 𝐸𝐸Ga3dS), kombiniert mit 17.5 eV, der Energiedifferenz zwischen VB

Kante und Ga 3dB CL der GaN Masse. Die beiden Ansätze berücksichtigen die

Energieunterschiede zwischen der Kante des Leitungsbandes (CB) (𝐸𝐸CB) und der Fermi-Energie

𝐸𝐸FB des GaN Festkörpers (𝐸𝐸CB − 𝐸𝐸FB) sowie zwischen 𝐸𝐸CB und 𝐸𝐸VB, der Bandlückenenergie 𝐸𝐸g,

die beide temperaturabhängig sind. Die beobachtete Abnahme von 𝑉𝑉bb mit zunehmender

Temperatur wird teilweise dadurch erklärt.

Aufgrund der Schwierigkeit, eine exakte lineare Extrapolation für die VB Spektren der

gesputterten n-GaN(0001)/H2O Grenzfläche, hauptsächlich bei hohem H2O Druck, zu finden,

konnte 𝑉𝑉bb sinnvoll nur aus den Verschiebungen von (𝐸𝐸FS − 𝐸𝐸Ga3dS), in Kombination mit

(𝐸𝐸VB − 𝐸𝐸Ga3dB) = 17.5 𝑒𝑒𝑉𝑉, und (𝐸𝐸FS − 𝐸𝐸VS), der VBM Energie, geschätzt werden. Bei konstanter

xxiii

Temperatur wird die Abnahme von 𝑉𝑉𝑏𝑏𝑏𝑏 an der gesputterten n-GaN(0001)/H2O Grenzfläche bei

einem Druck 𝑝𝑝 unter 0,1 mbar auf die Kompensation von freien Bindungen an der Oberfläche

durch dissoziative H2O Moleküle zurückgeführt. Die Konstanz von 𝑉𝑉bb bei 𝑝𝑝 > 0.1 mbar wird als

Sättigungsbedeckung der n-GaN(0001) Oberfläche mit H2O Molekülen verstanden. Der

Rückgang von niedrig-Druck 𝑉𝑉bb mit steigender Temperatur lässt sich zum Teil durch die

Abnahme von 𝐸𝐸g(𝑇𝑇)=(𝐸𝐸CB − 𝐸𝐸VB) und �𝐸𝐸CB − 𝐸𝐸FB(𝑇𝑇)� mit steigender Temperatur erklären,

obwohl ein solcher Rückgang nicht allein den gesamten Trend von 𝑉𝑉bb bei hoher Temperatur

erklären kann.

Die gesputterte n-GaN(0001) Oberfläche ermöglicht eine bemerkenswerte Adhäsion von

Molekülen der schwefelhaltigen Aminosäure L-Cystein (L-CySH), wie aus den jeweiligen UHV-

XPS Spektren hervorgeht. Die Ablagerung von L-CySH auf der n-GaN(0001) Oberfläche erfolgte

durch Sublimation von L-CySH Pulver bei 105 °C für ≤ 40 Minuten. Die UHV-XPS Spektren

wurden an einer gesputterten GaN(0001)/L-Cystein Grenzfläche bei Raumtemperatur und nach

Tempern bei verschiedenen Temperaturen (𝑇𝑇A = 75 − 150℃) für Temperzeiten (𝑡𝑡A) von 15, 30

und 60 Minuten aufgenommen.

Die UHV-XPS Spektren der gesputterten n-GaN(0001)/L-Cystein Grenzfläche zeigten, dass die

Thiol Kopfgruppe (SH Gruppe) des L-Cystein Moleküls eine starke chemische Bindung mit den

nicht gesättigten Bindungen des metallischen Ga an der Oberfläche einging. Ein solches

Verhalten wurde anhand der überlappenden Ga 3s und S 2p XPS Photoemissionslinien

beobachtet, deren Intensität durch mehr abgelagerte L-Cystein Schichten verstärkt wurde, ein

Merkmal, das charakteristisch für Thiolatspezies ist. Der Bindung von Carboxyl-(COO-) und

Aminogruppen (NH2) an der Oberfläche kann nicht vollständig ausgeschlossen werden, wie aus

den jeweiligen C 1s, O 1s und N 1s XPS Spektren hervorgeht. Die Stärke und das

Erscheinungsbild der Dublett S 2p Photoemissionslinien nahmen mit zunehmender 𝑇𝑇𝐴𝐴 ab, was

darauf hindeutet, dass die meisten, aber nicht alle L-Cystein Schichten, wieder von der

funktionalisierten Oberfläche verdampften, wobei die SH Gruppe der untersten L-Cystein Schicht

immer noch chemisch an die Ga Atome der Oberfläche gebunden war.

xxiv

Scientific contribution in local and international conferences and workshops:

• Mais Ahmad, Roland Hergenröder, Norbert Esser, Near-ambient pressure xps applications,

SESAME-Germany Info Day, 21 April 2023 at DESY, Hamburg (Oral presentation).

• Mais Ahmad, Ravi Prakash, Hannes Raschke, Norbert Esser, Roland Hergenröder,

Characterization of functionalized GaN surfaces under ambient conditions by NAP-XPS,

18th European Conference on Applications of Surface and Interface Analysis, 15-20 Sep

2019, Dresden, Germany (poster presentation).

• Mais Ahmad, Ravi Prakash, Hannes Raschke, Norbert Esser, Roland Hergenröder, NAP-

XPS of Functionalized GaN Surface for Biomolecular Sensing, 5th Annual Ambient Pressure

X-ray Photoelectron Spectroscopy Workshop 2018, 11-14 Dec 2018, Berlin, Germany

(poster presentation).

• Mais Ahmad, Ravi Prakash, Hannes Raschke, Norbert Esser, Roland Hergenröder, A study

of morphology and chemical structure of GaN(0001) surfaces by NAP-XPS, LEED and AFM

techniques, 34th European Conference on Surface Science, 26-31 Aug 2018, Aarhus,

Denmark (poster presentation).

• Mais Ahmad, Jürgensen Astrid, Hannes Raschke, Norbert Esser, Roland Hergenröder,

Surface analysis of GaN by Near Ambient Pressure XPS N² Science Communication: The

joint event of the PhD networks Max Planck PhDnet, Helmholtz Juniors and Leibniz PhD

Network, 6-8 Nov 2017, Berlin, Germany (Oral presentation).

Publications:

1) Retrieval of optical constants of undoped amorphous selenium films from an analysis of

their normalincidence transmittance spectra using numeric PUMA method; Mousa M. Abdul-

Gader Jafar, Mahmoud H. Saleh, Mais Jamil A. Ahmad, Basim N. Bulos, and Tariq M. Al-

Daraghmeh: J. Materials Science: Materials in Electronics (2016) 27:3281-3291.

2) Evaluation of spectral dispersion of optical constants of a-Se films from their normal-

incidence transmittance spectra using Swanepoel algebraic envelope approach; Mahmoud

Hatem Saleh, Nidal M. Ershidat, Mais Jamil A. Ahmad, Basim N. Bulos, and Mousa M.

Abdul-Gader Jafar: Optical Review (2017) 24 (3):260-277.

3) Electrical transport mechanisms and photoconductionin undoped crystalline flash-

evaporated lead iodide thin films; Tariq M. Al-Daraghmeh, Mahmoud H. Saleh, Mais Jamil

A. Ahmad, Basim N. Bulos, Khawla M. Shehadeh and Mousa M. Abdul-Gader Jafar,

Journal of Electronic Materials (2017) 47 (3):1806-1818.

4) Evaluation of kinetic parameters and thermal stability of melt-quenched BixSe100-x alloys (x

≤ 7.5 at%) by non-isothermal thermogravimetric analysis; Mais Jamil A. Ahmad, Mousa M.

Abdul-Gader Jafar, Mahmoud H. Saleh, Ahmad Telfah, Khalil A. Ziq, and Roland

Hergenröder, Applied Microscopy (2017) 47 (3):110-120.

5) Identification of relaxation processes in pure polyethylene oxide (PEO) films by the dielectric

permittivity and electric modulus formalisms; Ahmad Telfah, Mousa M. Abdul-Gader Jafar,

xxv

Inshad Jum’h, Mais Jamil A. Ahmad, Jörg Lambert, Roland Hergenröder, Polymers for

Advanced Technologies (2018) 29 (7):1974 -1987.

6) Structural, stoichiometric and optical constants of crystalline undoped lead iodide films

prepared by the flash-evaporation method: Mousa M. Abdul-Gader Jafar, Mahmoud H.

Saleh, Tariq M. Al-Daraghmeh, Mais Jamil A. Ahmad, Maryam A. AbuEid, Nidal M.

Ershaidat & Basim N. Bulos, Applied Physics A (2019) 672

7) First-principles calculations of the high-pressure behavior, electronic, magnetic, and elastic

Properties of praseodymium pnictides: PrX (X = P, As and Bi); Ghellab, T., Baaziz, H.,

Charifi, Z., Latelli, H., Ahmad, M. J. A., Telfah, M., Alsaad, A., Telfah, A., Hergenröder, R.,

Sabirianov, R, Journal of Magnetism and Magnetic Materials (2021) 54.

8) DFT Investigation of Physical Properties of KCrZ (Z=S, Se, Te) Half

-

Heusler alloys; Telfah,

A., Sâad Essaou, S., Baaziz, H., Charifi, Z., Mohammad Alsaad, A., Ahmad, M. J. A.,

Hergenröder, R., Sabirianov, R., physica status solidi (b) (2021) 258.

9) Dielectric relaxation, XPS and structural studies of polyethylene oxide/iodine complex

composite films; Telfah, A., Al-Akhras, M., Al-Izzy, K. A., Ahmad, A. A., Ababneh, R.,

Ahmad, M. J. A., Tavares, C., Hergenröder, R., Polymer Bulletin (2021).

10) XPS, UV-Vis, XRD and PL spectroscopies for studying nikel nanoparticle positioning effect

on nanocomposite film proprerties: Inshad Jumh, Ahmad Telfah, Marwan S. Mousa, Mais

Jamil A. Ahmad, Carlos J. Tavares, Roland Hergenroder, Applied Polymer (2022)

11) XPS, UV-Vis and FTIR Spectroscopies for Studying Nickel Nanoparticles (NiNPs)

Positioning Effect on Properties of (PS-PANI/NiNPs) Nanocomposite Films: Mais Jamil A.

Ahmad, Ahmad Telfah, Qais M. Al-Bataineh, Carlos J. Tavares, Roland Hergenröder.

Polymers for Advanced Technologies (2023) 34.

12) Study of the band bending at n-GaN(0001)/H2O interfaces by near-ambient-pressure X-ray

photoelectron spectroscopy; Mais Jamil A. Ahmad, Ahmad Telfah, Ravi Prakash, Mousa

M. Abdul-Gader Jafar, Carlos J. Tavares, Norbert Esser, and Roland Hergenröder (Applied

Surface Science Advances) (Under Revision).

xxvi

1

Introduction and Motivation

Silicon and its composite compounds/alloys are still of scientific interest in various areas of basic

research and technological applications as metal-oxide-semiconductor-field-effect-transistors

(MOSFETs) utilized in many electronic devices, including chemical and biological sensors [1-6].

On the other hand, over the past four decades the group III-nitrides have emerged as well from

being the theme of general interest and basic research into real-world microelectronics as LEDs,

photovoltaics, and high-electron-mobility transistors (HEMTs) [7-14]. The group III-nitrides include

the binary compounds indium nitride (InN), aluminum nitride (AlN), gallium nitride (GaN), and their

III-nitride alloys. The group III-nitrides also got considerable attention in the development of

interfaces between pertinent electronics and chemical/biological materials [15-21].

The scientific and industrial ambitions for utilizing the group III-nitrides in the fabrication of

solid/liquid interfacing systems arises from their potentiality and viability in the field of easy-

handled, compact, cost-effective and reliable semiconductor-based biosensing systems [17-20].

The rich knowledge of properties of III-nitrides and great efforts to prepare clean/stoichiometric

and functionalized surfaces of them to detect physiological relevant biomolecules [20-26], render

them of favorable use not only in the construction of microelectronic devices but of biosensors.

A biosensor is an analytical device that combines a physiochemical detection device with

a biological receptor that binds and interacts with a target biological molecule (analyte) [15-19].

Such a biosensor should fulfill high specificity, robust and less affected by temperature and pH,

good biocompatibility, low cost, fast monitoring of obtained results and high sensitivity [26-28].

The workability of chemical and biosensors based on III-V nitrides [12,13,17-19,29] could be

further enhanced due to the industrial advances of modern semiconductor electronics and

computing systems. Silicon lacks long-term chemical and high-temperature stability due to its low

resistance to hostile chemicals and charge-carrier generation due to oxidation at the silicon/liquid

interfaces [17-19,30]. In biological ambient, Si has an intrinsic flaw since its surface is often

escorted by a native SiO2 layer, which it is hydrophilic and biocompatible, but permeable to ions

in physiological alkali solutions which diffuse freely through it, causing instability under electric

fields [30]. This makes fabricating a thermally stable, quality Si-based biosensor challenging and

1

more expensive for use in solutions of high ionic strength, as it deteriorates in chemically caustic

environments.

In view of aforementioned necessities for realizing a prototypical biosensing device, the

III-nitrides are considered to be talented future candidate materials in the area of developing

interfaces between electronic systems and biological bodies. Among the III-V, GaN evolves as a

particular potential competitor to biosensors and chemical/gas sensors based on silicon and

gallium arsenide (GaAs) [16-20,31-38]. This is because GaN has unique structural, mechanical,

thermal, electrical, and optical properties, as well as high optical transparency over a wide spectral

range due to its high bandgap energy 𝐸𝐸g (≅ 3.4 eV at 300 K) [22,23,39,40-42].

Hexagonal-structured (wurtzite) GaN has a high molecular bond energy (~ 9 eV/molcule).

Thus, it has extraordinary chemical and thermal stability up to 900 ℃, above which it decomposes

to metallic Ga and nitrogen gas N2 despite of its very high melting point (~ 2800 ℃) [23,42]. Also,

GaN is inert in harsh environments and insoluble in most aqueous solutions (acidic and basic);

however, in some cases these solutions result in spurious and undesired coatings on its surface

[42-46]. Additionally, GaN material is remarkably non-toxic, nonaggressive and has excellent

biocompatibility to many biomolecules [33,42]. It can be used under certain conditions in the

construction of room temperature (RT) photo-electrochemical and hydrogen fuel cells [47-49].

The surface of epitaxial GaN layers generally favors forming robust chemical bonds with

inorganic and organic molecules having good detection ability of induced structural change upon

interactions with in vitro and in vivo chemical and biological (thiol species) [18-20,33-36,47-49].

Such an aspect is of importance in initiating modification and functionalization of GaN surface to

facilitate chemical bonding with the optimal receptor from surrounding environment. The above-

cited expedient properties of GaN meet most of the criteria required for a virtuous biosensor.

There are up-to-date commercial techniques for growing low-cost, quality, and crystalline GaN

epitaxial layers on durable homo-and hetero-substrates [22-25,45,50-54]. This adds to the

advantages of utilizing GaN layers in hybrid bio-molecular detection systems.

The currently fabricated GaN layers still suffer from undesired bulk and surface structural flaws

and defective properties [23-25]. This results in creation of native defects, imperfections,

dislocations, charged dangling bonds, and unknown impurities, hence the formation of surface

charged states and bulk free electrons [42-45,55-57]. Further, unavoidable adventitious carbon

(C) and oxygen (O2) species on GaN surface are not desirable and cause problems when the

surface is to be capped with an inorganic, or organic, or a biological material [47-49]. Also, a thin

2

native overlayer of gallium oxide (GaxOy) is not desired as it retards adherence of inorganic or

biological molecules, thus impeding the formation of the required GaN/liquid interface [47,58-61].

Therefore, the surface of a GaN layer should be properly modified to facilitate its

functionalization with the sought after inorganic or organic molecules that can form covalent

linkages with specific biological species. GaN-based HEMT biosensing structures emerged to be

competitor to those based on gold layers, where a thiol-metal self-assembly-monolayer (SAM) is

required for biomolecule linking to complete surface functionalization [19,41,62-67]. Gold exhibits

good biocompatibility, relative inertness, hydrophobicity, comparable to GaN, and robust covalent

bonds to sulfur of thiol SH functional group or of certain amino-acids like L-cysteine.

There is yet a lack of an efficient route for wet chemical cleaning/etching or gas-ion etching

and sputtering to produce GaN surface free of contaminants and unwanted species. Etching and

sputtering procedures often result in damaged and wrinkled GaN surface and cannot aptly remove

undesired impurities and oxides [21-25,43-45,57,61,68-77]. This prohibits functionalization of

GaN surfaces for sensing chemical and biological species, as well as hindering the understanding

of interactions and bonding processes taking place on functionalized GaN surface [15-20,33-36].

The nature and viability of bonding mechanisms on functionalized GaN surface depend

on fabrication process of a GaN layer on an appropriate substrate. Bonding mechanisms are

influenced by cleaning procedure of GaN surface prior functionalization with inorganic and organic

species. To attain further insight into negatives and benefit from GaN prosperities, a methodology

that circumvents posed technical hitches is required. This is essential for improving quality and

structure of GaN surface and its susceptibility to accommodate inorganic and organic molecules.

As regards to fabrication methods of GaN films, simple, inexpensive and low temperature

sol-gel spin and dip coating methods were used to deposit thin GaN films on various substrates [

[78-80]. Improving fabrication procedures and using proper chemical precursors to get quality and

featured GaN films by such methods is yet indispensable. Alternatively, more advanced

techniques are used for fabricating epitaxial GaN films on unlike substrates as RF reactive

sputtering [81], plasma-induced molecular beam epitaxy (PIMBE), metal-organic chemical vapor

deposition (MOCVD), and hydride vapor phase epitaxy (HVPE) [22-25,45,50-54]. Film processing

by these techniques are highly elaborate, realized at high temperatures (800-1200 oC) under

rigorous preparation conditions, and use mixtures of inorganic and organic precursors, besides

harmful gases. High growth rates and thick GaN films (≥ 100 µm) with defects and dislocations

less than 107 cm– 2 and minimal carbon in bulk are attained by HVPE at low cost, compared to

3

PIMBE and MOCVD, which are optimum for indium or aluminum alloying and p-type doping of

GaN films [29,45].

The substrates for GaN films have to fulfill a variety of specifications such as good thermal

and lattice matches to GaN as well as they have to withstand high temperatures during fabrication

and post-growth heat treatment of GaN films [23,29,45]. The former requirement can be

circumvented via using GaN crystalline wafers as substrates to obtain homoepitaxial structures,

which is, however, still a critical and challenging issue [82].

Currently, substrates of dissimilar materials that have low thermal expansion and lattice

mismatch to GaN are frequently used. However, in case of GaN films grown by HVPE or MOCVD,

these heterogeneous substrates are limited to those unaffected by ammonia (NH3) and hydrogen

(H2) precursors above 1000 oC. This restricts the use of Si and GaAs substrates for GaN films

grown by these methods [23,29,54]. The PIMBE film deposition is proceeded under ultrahigh

vacuum (UHV) conditions at temperatures 250 oC lower than those used in HVPE and MOCVD

methods. Nevertheless, the heterogeneous substrates of PIMBE-grown GaN films should be

stable and slightly affected by encountered nitrogen radicals at temperatures above 800 oC.

The substrates of GaN films should also be atomically flat, cost-effective, and can be

produced in large sizes for device applications. Furthermore, the substrate for films of wurtzite

(WZ-) structured GaN(0001) is preferable to have a similar hexagonal structure. In WZ-

GaN(0001) semiconductor, the anion (N3−) and (Ga3+) cation bond is along the c-axis of

hexagonal structure or the (0001) direction, and a large spontaneous polarization readily exists.

In view of above suppositions, the hexagonal-structured silicon carbide (SiC) and sapphire

(Al2O3) are appropriate substrates for WZ-GaN(0001) films due to their adequate thermal and

chemical stability at high temperatures [11,16,22-25,42,50,53]. The c-sapphire of crystalline

(0001) orientation has good surface morphology, is inexpensive and optically transparent. These

aspects make Al2O3(0001) wafers favorable substrates for WZ-GaN(0001) films, despite of the

large thermal expansion mismatch between them and lattice mismatch (16%), compared to 3.5%

for SiC [29,31,50]. Yet, SiC wafers are not commonly used as substrates for GaN films because

their production is highly expensive and involves sophisticated fabrication procedures.

The main drawback of lattice and thermal-expansion mismatch between a GaN film and

a bare c-sapphire substrate is the creation of native structural defects, impurities and dislocations,

which are minimal in HVPE c-Al2O3/GaN films [45]. This results in large density of free electrons

4

in the GaN film, in addition to surface states. Diffusion of oxygen impurities is enhanced during

growth process above 800 oC into the GaN film from its c-sapphire substrate, thus a lot of free

charge carriers are created, and the production of unintentionally n-doped GaN(0001) films is

facilitated [22,23,83].

All above cited defective features affect and worsen the quality and structural morphology

of the surface of epitaxial WZ-GaN films laid on bare c-sapphire substrates, besides deteriorating

their optical and electrical properties. These would also lead to degradation of bonding processes

and interactions of the outermost surface of n-GaN(0001) films with incubated inorganic/organic

and biological molecules. The Interface between GaN surface and inorganic, or organic, or

biological molecules is a key issue in the development of devices based on chemo-/bio-molecular

recognition. A study of the feasibility of GaN for such applications is thus of importance. This can

be primarily realized via investigating GaN surface exposed to different modification schemes and

functionalization processes under various working conditions [16-20,32-36,47-49].

Various techniques were adopted to explore the desirable properties of GaN films for use

in GaN-based chemical and biological sensors. Electrical behavior of GaN surface capped with

molecular species was studied using HEMT structures [14-19,41]. Biosensing ability of GaN

surface was tested in conditions that mimic real world ambience by electrochemical impedance

spectroscopy [39,84-87], conductance measurements [63], and ellipsometry [88]. Structure and

morphology of GaN surface were probed by atomic force microscopy (AFM), low-energy electron

diffraction (LEED), and scanning and transmission electron microscopy (SEM and TEM) [24,25].

UHV X-ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy (AES) [89-91]

were used to study structure and chemistry of bare GaN surfaces [43-45,51,72,92] and GaN

surfaces functionalized with biological molecules [18-21,32-36,58,65].

The near-ambient-pressure NAP-XPS technique [93-97] are often utilized to investigate

solid/vapor (or liquid) interfaces under few-mbar pressures. NAP-XPS measurements were made

on GaN/water interfaces under different H2O vapor pressures (< 5 mbar) and various sample’s

temperatures (< 400 oC) [47-49,98-100]. These UHV-XPS and NAP-XPS studies on GaN surface

functionalized with H2O molecules or with organic molecules were primarily aimed to examine its

practicability to form physical/chemical linkages with inorganic, organic, and biological species.

In the present work, the AFM, LEED, UHV-XPS, and NAP-XPS techniques were exploited

to study the topography, morphology and chemistry of thermally-annealed and gas-ion sputtered

GaN surface of studied GaN(0001)/c-sapphire samples. This investigation has been carried out

5

in some detail since the numerous literature studies made on bare and modified GaN surfaces

yielded diverse and unresolved issues. Further, dissimilar and debated results were also reported

in studies conducted on GaN surface functionalized with H2O-vapor and with molecules of short-

chain amino acid peptides, in particular L-cysteine. In this work, a comprehensive study has been

also conducted on n-GaN(0001) surface functionalized with few monolayers of H2O and L-

cysteine molecules under different experimental conditions. The aim was to attain more

understanding of the nature of modifications as well as the ensuing bonding mechanisms taking

place at interfaces of both anticipated species of GaN/H2O and GaN/L-cysteine structures.

A crucial step prior functionalization of a GaN surface is a low-damage effective cleaning

or etching process. Etching and/or thermal treatment are needed to clean GaN surface from oxide

layers and native C and O2, besides other imperfections [24,25,45,61,101-105]. Literature is rich

in diverse results of experiments made on GaN surface exposed to wet etchants of harmful

chemicals that result in uncontrollable and adverse complexes and oxides on the etched surface

[43-45,68,70-72]. Aggregation of unknown species on GaN surface etched with wet chemical

makes the analysis of measured surface chemistry and morphology more elusive and

controversial.

Sputtering GaN surface by argon (Ar+) and hydrogen (H2

+) ions is destructive and results

in unrecoverable damage of GaN surface, deficient of its ingredients [45]. In contrast, sputtering

GaN surface with nitrogen (N2

+) ions is active with minimal decomposition, less damage, and low

roughness of N2

+-sputtered GaN surface [45,47]. Convenient cleaning and etching of GaN surface

lead to an appropriate surface ionic or molecular bridge favorable for chemical bonding with

specific inorganic molecules or biomolecules [15-19]. Treatment and modification of GaN surface

enhance its chemical stability, improved sensor selectivity, greater biocompatibility and stronger

bonding with implanted molecules [32-35,47-49].

Both covalent and non-covalent approaches are reported on molecular modification of

group III-nitrides surfaces. The diversity between the surface modifications approaches depend

on the type of used inorganic, organic, and thiol-like molecules, like the amino-acid L-cysteine.

The thiol molecule has the general chemical R−SH structure, which terminates with an SH

functional group that can bond to organic or biological molecules, with R being an alkyl or other

organic substituent. Covalent surface modifications require oxidized layer on the sample surface,

while other surface modifications favor molecular linkage to a cleaned or etched surface.

Organophosphonic acids and organosilanes were used for oxidized GaN surface and III-nitrides

6

and modifying GaN surface and GaN-based heterostructures for chemical or biomolecular

sensing [19,33-36,41,58]. An approach for modifying un-oxidized, clean GaN surface to form apt

linkage with organic and biological molecules is via attachment of hydrogen bridges or thiol-like

amine SH functional groups [18-23]. Covalent and non-covalent attachment can be achieved via

suitable peptide structures that recognize and biocompatible to GaN surface [19,32-36].

One purpose of this work is to scrutinize stability, morphology, chemical states and

+

interactions on the surface of n-GaN(0001)/c-sapphire samples exposed to N2 ion sputtering.

Another goal is to functionalize annealed or N2

+-sputtered GaN surface with H2O and sulfur

containing amino acid L-cysteine (CH7COOHS) molecules. This is to explore the nature of bonding

between functionalized GaN surface and inorganic and organic molecules under different

experimental conditions. This is also aimed to get further knowledge and insight on the efficacy,

viability and possibility of using functionalized GaN in gas or biomolecular sensing devices. To

achieve these goals, XPS was exploited to measure spectra of core-level (CL) excitations of

elements and of valence band (VB) on post-functionalized GaN surface. Additional

measurements were made using AFM (ex-situ) and UHV LEED (in-situ) to emphasize the

structure and morphology of the surface of studied samples.

This PhD thesis composed of six chapters which are organized as follows:

Chapter 1 briefs importance of utilizing wurtzite GaN in technological applications, in particular

those related to interfacing GaN layers with existing electronics in the realm of chemical sensors

and biosensors. Additionally, in this chapter, relevant literature review on GaN and GaN-based

electronic devices and the main purpose of this work are delivered.

Chapter 2 covers briefly the physical and chemical properties of GaN semiconductor in particular,

as well as a brief summary on the techniques exploited to prepare GaN films.

Chapter 3 summarizes the surface analysis characterization techniques used in this work.

Chapter 4 details experimental procedures adopted for in-situ dry ion-gas sputtering and etching

of studied HVPE-grown c-sapphire/GaN(0001) samples. The methods of functionalizing the GaN

surface with few H2O vapor and L-cysteine monolayers are also described in this chapter.

Chapter 5, the results of AFM, LEED, and UHV-XPS and NAP-XPS measurements made on

studied n-GaN(0001) films, as well as their analysis and discussion are presented.

Chapter 6 concludes the main results of this study, besides some important and motivated

suggestions for further work in the field of present investigation.

7

2

Fundamentals of Gallium Nitride

Gallium nitride (GaN) is a semiconducting compound material which has many unique physical

properties that categorize it among the group III-V as a particular prospective candidate for use

in inorganic/organic chemical/gas sensors and biosensors [21-32,44]. Moreover, GaN is non-toxic

to human health, insoluble in aqueous solutions, inertness, chemically, thermally and

mechanically stable, and biocompatible [62,63]. Scientific and technological interest in producing

and exploring GaN has been escalated because of advances of its processing that yield quality,

cost-effective heteroepitaxial films by various fabrication methods [16,28,35,45,62]. This interest

is also motivated because of the accessibility of a variety of sensitive and reliable characterization

techniques [106-116].

2.1 Structural Properties

Under normal ambient conditions, the thermodynamically stable structural phase of GaN is the α-

phase wurtzite (WZ) structure consisting of non-primitive hexagonal unit cells with 6 atoms of

each type WZ-GaN. The WZ-GaN can be pictured as two interpenetrating hexagonal closed-

packed (hcp) lattices, each with one type of atoms, offset along 𝑐𝑐-axis by 5𝑐𝑐/8, with the lattice

constants 𝑎𝑎 = 3.19 Å (edge length of basal hexagon) and 𝑐𝑐 = 5.19 Å (cell height) [15].

The planes and positions of the ions (atoms) in a unit cell of the hcp lattice structure of the

WZ-GaN semiconducting compound are stacked as usual in the order ABABAB as shown in

Figure 2.1(a) [15-17]. The tetrahedral covalent bonding between Ga and N atoms determines

predominantly the stability of the hcp structural phase of GaN compound, in addition to the partial

ionic contribution to the bond arising from the large differences in their electronegativities [44].

The anion-cation bond length (N3− − Ga3+) in the WZ-GaN and the strength along the (0001)-axis

determine its lattice constants and bandgap [16]. The molecular bond energy in the WZ-GaN is

fairly large (~ 9 eV/molecule) and this can be considered as a cause of its high chemical and

thermal stability at elevated temperatures [15].

8

a)

b)

Figure 2.1: WZ-GaN showing the a) coordination and diatomic close-packed alternative planes

of cations (Ga3+) and anions (N3−), after R. Bouveyron [17] and b) conventions for Ga-faced and

N-faced polarities along 〈0001〉 and 〈0001

�〉 directions, respectively, after O. Ambacher [40].

The wurtzite structure of GaN has polar axis (lack of inversion symmetry) with Ga-N bonds

in the 〈0001〉 direction, faced by nitrogen in the same direction and by the cation (Ga) in the

opposite direction. The WZ-GaN growth direction is normal to {0001} basal plane [16] and N and

Ga atoms are arranged in bilayers consisting of two hexagonal layers, each consists of N3− or

Ga3+ ions [15-17]. Thus, GaN itself shows either polar Ga-face or N-face on each surface in the

{0001} plane, with [0001] or [0001

�] polarity. The 〈0001〉 direction is a vector pointing from a Ga

atom to its nearest-neighbor N atom, as seen from Figure 2.1(b). Bulk properties depend on

whether GaN surface is faced by nitrogen or gallium atoms.

The Ga-polar (Ga-faced) GaN crystal does not mean Ga-terminated surface and N-polar

(N-faced) crystal does not mean N-terminated surface. A Ga-face surface may be covered by free

N ions and thus it is an N-terminated surface but the crystal itself is not N-faced, and vice versa.

The polarity of a GaN layer is a bulk property, while termination is a surface property, determined

by its structure, type of substrate, nearby ambient and type of surface ions (Ga3+ or N3−) [15].

The polarity of III-nitrides affects the surface topography of their epitaxial layers grown on

sapphire substrates [44]. The surface of Ga-polar GaN layer is relatively smoother and more

9

stable against chemical etching, compared to the surface features of N-polar GaN layers. As the

polarity determines the surface topography of III-nitride layers, their structural, optical and

electrical properties will also be affected. The feasibility of wet chemical etching is limited by

surface defects and impurities, physisorbed and chemisorbed adventitious species like carbon

(C) and oxygen (O2) on as-grown GaN layers. Wet etching-induced surface damage and inclusion

of spurious species from hydroxide precursors are also possible [44,117].

The polarity of the group III-nitride layers can be determined by the XPS technique using

the monochromatic Al Kα X-ray radiation of the energy ℎ𝜋𝜋 = 1486.7 eV as an excitation source

[44,118-122]. The experimental VB spectra of the surface of wurtzite III-nitride layers reveal two

detached peaks: A VB peak (𝑃𝑃I) is apparent at low binding-energy (BE) and is attributed to metallic

group III-polar layer, with the VB peak (𝑃𝑃II) of the relatively smaller intensity is seen at higher BE.

It is reported that when the XPS VB peak (𝑃𝑃I) at the low-energy side is on the contrary weaker

than the VB peak (𝑃𝑃II) at the higher BE, the III-nitride layer is (0001

�) N-polar [44,121].

The features of XPS VB spectra were reported to vary with surface contamination and

when the sample is exposed to different preparation and experimental conditions [44]. These

include growth technique, angle at which photoelectrons are collected, surface wet chemical

etching, oxidation, thermal annealing, ion-gas sputtering and functionalization with inorganic and

organic molecules. Figure 2.2 shows typical reported XPS VB spectra for metal-organic vapor

phase epitaxy (MOVPE) grown In-/N-and Ga-/N-polar layers taken under various experimental

conditions and features of their well-resolved experimental XPS VB peaks 𝑃𝑃I and 𝑃𝑃II [44,121].

Figure 2.2: Typical XPS VB spectra of WZ-GaN a) N-polar GaN layer, b) In-/N-polar InN layer

and c) Ga-polar GaN layer taken under various preparation and experimentation conditions

[44,121].

10

2.2 Band Gap and Electronic Properties

The band structure of WZ-GaN semiconductor is complicated and entangled, with features

that depend on the utilized theoretical energy band calculation method. First-principles electronic

structure calculations, local/nonlocal density approximation (LDA), and the empirical pseudo

potential method give reliable results. Figure 2.3 shows a diagram of GaN band structure

calculated by the empirical pseudo potential method [123]. The characteristics of GaN around

Brillouin zone (BZ) wavevector k = 0 (Γ point) yield important information about its optical

absorption behavior, governed by type of band-to-band (interband) electronic transitions between

filled energy states in VB and empty states in conduction band (CB).

Figure 2.3: Energy band structure of WZ-GaN calculated by empirical pseudo potential method

[123].

Transitions may occur between states in the vicinity of valance band maximum (VBM)

located at the highest band energy 𝐸𝐸v(𝐤𝐤 = 0 ) and empty states near conduction band minimum

(CBM) positioned at the lowest energy 𝐸𝐸c(𝐤𝐤 = 0 ). The excitation of electrons from VBM to CBM

is called direct interband electronic transition. The value of the bandgap energy 𝐸𝐸g (= 𝐸𝐸c − 𝐸𝐸v)

depends on the way it is being defined and its theoretical and experimental values reported in

literature are somewhat diverse. It depends on the used theoretical calculation approach and on

the experimental method exploited for its measurement. A commonly accepted value for direct

bandgap energy of WZ-GaN is that optically measured and lies between 3.35 eV and 3.45 eV, with

𝐸𝐸g(𝑇𝑇) increases slightly as temperature 𝑇𝑇 decreases [17,124]. Ideal pristine crystalline WZ-GaN

is thus optically transparent to up to the ultraviolet (UV) spectral edge. The WZ-GaN has other

11

advantageous electrical, dielectric and thermal properties that are related to its structure [15-17].

Its electron effective mass 𝑚𝑚𝑒𝑒

∗ is small (≅ 0.24 𝑚𝑚0 at Γ point, with 𝑚𝑚0 being the free-electron mass).

It has a static dielectric constant 𝜀𝜀⁄𝜀𝜀0 of 8.9 and a high-frequency (optical) dielectric constant

𝜀𝜀∞⁄𝜀𝜀0 of 5.2, where 𝜀𝜀0 is free-space permittivity and 𝜀𝜀∞ is the high-frequency permittivity. The

WZ-GaN has good thermal conductivity (1.3 W/cm2 K), large electron mobility 𝜇𝜇e (~ 2000 cm2/

V. s), and high drift saturation velocity v𝑒𝑒

sat (~ 2.5 x 107 cm/s) at 300 K.

2.3 Band Bending at Wurtzite GaN Interface

Knowledge of the band structure of WZ-GaN is useful to understand the nature of band bending

(BB) that depends on the location of Fermi level (𝐸𝐸F) with respect to CBM at 𝐸𝐸c(Γ) and to VBM at

𝐸𝐸v(Γ). The trend of BB of a GaN layer is governed by the distribution of its surface charged states

[44,121]. The BB can be pictured on the basis of the branch point energy 𝐸𝐸BP, which is an intrinsic

property of the semiconductor [125-129]. Computation of 𝐸𝐸BP was carried out for WZ-GaN and

other III-nitride semiconductors by various theoretical approaches [125,129]. Measurement of 𝐸𝐸BP

expt theor=was made on few semiconductors. For WZ-GaN, 𝐸𝐸BP = 2.4 eV agree with the theoretical 𝐸𝐸BP

theor expt

2.37 eV [128]. For comparison, these references reported 𝐸𝐸BP = 1.58 eV and 𝐸𝐸BP = 1.58-1.83

eV for WZ-InN semiconductor [129]. The 𝐸𝐸BP is the effective midgap energy or charge neutrality

level across the entire BZ region [128]. It tells that donor-like native defects are favorable if 𝐸𝐸F

-

level is lying below 𝐸𝐸BP, while acceptor-like defects may be formed if it is above 𝐸𝐸BP [44,129].

It is reported [125-129] that 𝐸𝐸BP is particularly above CBM for WZ-InN (𝐸𝐸g

= 0.71 eV) by

∆𝐸𝐸C = 𝐸𝐸C − 𝐸𝐸BP = − 0.87 eV, thus its 𝐸𝐸F is in the CB region, while for WZ-GaN (𝐸𝐸g

= 3.4 eV) the

𝐸𝐸BP level is below CBM by ∆𝐸𝐸C ≅ 1.0 eV, and therefore its 𝐸𝐸F is below CBM. The intrinsic charge

carrier concentration 𝑛𝑛i(≤ 300 K) is very small (< 10−10 cm−3) in undoped, pure and defect-free

WZ-GaN due to its large 𝐸𝐸g [18]. However, the as-grown undoped WZ-GaN layers are reported

to be unintentionally n-doped with electron density around 1017 cm−3

, the origin of which is

debatable [17]. It is often related to a donor-type (n-type) defect, probably due to the presence of

large number of nitrogen vacancies (VN) and other native defects in the crystal lattice.

As 𝐸𝐸F in WZ-GaN lies below CBM, which is above 𝐸𝐸BP-level by 1 eV (𝐸𝐸BP is above the

VBM by 2.4 eV), an upward BB is expected at its surface. In contrast, downward BB exists at the

surface of WZ-InN [44]. The origin of upward BB at the surface of a GaN layer can be related to

its crystal polarity and charge defect states present on its surface. The bonds between Ga and N

12

atoms are partially ionic and tetrahedral covalent, which leads to the creation of N-Ga double

layers along c-axis of the hexagonal structure Figure 2.4.

Figure 2.4: GaN material layers in wurtzite structure.

The lack of inversion center in WZ-GaN lattice leads to two crystal polarities either parallel

or antiparallel to the c-axis, which is the common growth direction [14]. The large difference

between electro negativities of atoms N(3.04) and Ga(1.81) [44] creates strong charge distribution

between bonded atoms, forming a dipole moment that is directed towards the Ga atom along the

bond [16]. In a perfect tetrahedron, the four dipole moments of the four bonds inclosing each atom

would cancel out. But, as the hcp lattice symmetry of WZ-GaN structure is low and lacks inversion

center, such tetrahedron becomes deformed in a manner that the center of positive charges is

displaced from that of negative charges. This results in spontaneous polarization 𝐁𝐁SP independent

of the sample’s stress state [39-41]. The direction of polarization is generally defined from N to

Ga, hence for Ga-face (Ga-polar) the spontaneous polarization is in the negative c-direction,

whereas the spontaneous polarization is opposite for N-face (N-polar).

The literature values of the spontaneous polarization bound sheet of the charge in the wurtzite

GaN range between – 0.034 C/m2 and – 0.029 C/m2 in the absence of external electric field

[17,39,40,41,62,130-133]. The used value of |𝐁𝐁SP| is 0.029 C/m2, corresponding to a negative

(positive) surface bound charge sheet of density 1.81 x 1013 charges/cm2 at the Ga-face (N-face).

However, this must be compensated to avoid a large internal polarization field in some

13

semiconducting III-nitride samples [134]. The tetrahedral bonds, dipole moments and over-all

spontaneous polarization of WZ-GaN are depicted in Figure 2.5.

Ga polarity

N polarity

Figure 2.5: Tetrahedral bond configuration, individual dipole moments and spontaneous

polarization for III-nitrides with wurtzite structure.

Surface charge compensation in WZ-GaN layers can be due to various types of charged

states that act as trapping centers such as N-vacancies related to defect states or pinning states.

The vector 𝐁𝐁SP points into/toward the surface of Ga-polar/N-polar WZ-GaN, as seen in Figure 2.6.

Figure 2.6: Schematic diagram of surface band bending and charge distribution of Ga-face and

N-face WZ-GaN layer. The number of different charges shown on the diagram is not to scale

[130].

14

The performance and stability of GaN-based devices would also be affected by surface

states that interact with adsorbates on the surface and alter its electronic structure. Naturally

occurring, spontaneous polarization in Ga-/N-polar WZ-GaN layers gives rise to positive/negative

surface polarization charges, which contribute significantly to surface band bending. A

combination of internal and external charge states usually screen polarization charges and

behave as trapping centers or pinning energy levels that lead to localization of 𝐸𝐸F below the CBM.

Therefore, the surface BB in both Ga-/N-polar WZ-GaN is upward Figure 2.6.

For n-type WZ-GaN, the internal charge screening is linked to bulk ionized donors or free

electrons, while external screening arises from surface states due to impurities, structural defects,

surface Ga-(or N-) termination, overlayer surface oxides, and adsorbates. The sum of electric

fields due to internal screening, external screening, and spontaneous polarization charges in WZ-

GaN give rise to a net polarization that determines the overall upward surface band bending [132].

The band bending at the surface of a WZ-GaN sample can be explored and modified by

post-growth cleaning/etching , thermal annealing, and functionalization with inorganic and/organic

molecules. Increase in upward BB at the surface of a WZ-GaN layer indicates an increase of the

internal screening of polarization charge. The corresponding density of surface net charges can

be found once BB near the surface region is known [5,130].

The type, polarity and doping of a semiconductor, and its surface cleanness, oxidation,

defects, and adsorbates, governed by growth conditions and post-growth surface treatment, play

a significant role in the pinning of its Fermi level 𝐸𝐸F above the VBM edge (EV) at different energies

𝜁𝜁 = 𝐸𝐸F− 𝐸𝐸V [131,132]. In case of WZ-GaN, whose 𝐸𝐸g = 3.4 eV, 𝐸𝐸F was reported to be pinned at

values of 𝜁𝜁 in the range 2.6 – 3.1 eV (0.3 – 0.8 eV below CBM edge) for Ga-polar samples and in

the range 1.8 – 2.0 eV for N-polar samples [94,126,130-132]. This implies that for small values of

𝜁𝜁, 𝐸𝐸𝐹𝐹 is close to VBM and empty surface states exist and surface electronic bands are largely

bent upward (if 𝜁𝜁 < 𝐸𝐸g), but when the surface states become filled (reduced), 𝐸𝐸F moves upward

and band bending is lessened. This was also reported for n-GaP(001) (𝐸𝐸g = 2.3 eV) as 𝐸𝐸F is

pinned at 0.2 eV (2.3 eV) above VBM edge for clean (sulfur-treated) samples, respectively [135].

2.4 Common Substrates for GaN Layers

The majority of III-nitrides growth is mostly conducted on heterosubstrates, which should have

thermal and structural properties that closely match those of overgrown layer, so reducing cracks

at its surface, besides internal defects and imperfections like lack of an atom, dislocations, and

15

stacking faults [17]. The heterosubstrate should withstand growth and preparation conditions of

the layer without being damaged or reacted with neither the layer nor its hosting impurities [40].

The heterosubstrate should also have long-term chemical and thermal stabilities for use in the

construction of sensing structures incorporating inorganic and organo-bio-molecular species.

Furthermore, the heterosubstrate material ought to be cost-effective and abundant in the market.

Crystalline sapphire (Al2O3), silicon carbide (SiC), silicon Si(111) and gallium arsenide

(GaAs) are popular substrates for growing epitaxial layers of the III-nitrides having WZ-structure.

A low lattice mismatch with the wurtzite-GaN (3.5%) and proper electrical and thermal conductivity

(4.9 W/cm. K) makes undoped crystalline SiC wafers a favorable substrate for growing GaN layers

[17,40,136]. Yet, the expensive production of high quality, single crystalline SiC wafers inhibits its

far reaching use.

Growing wurtzite III-nitride layers on substrates made from cubic Si(111) which has good

thermal conductivity (1.5 W/cm. K) is a challenging issue. Crystalline Si(111) has large thermal

and lattice mismatch (17%) with WZ-GaN and low bandgap energy (1.12 eV). Further, inevitable

insulating oxide (SiO2) overlayer readily forms especially under ambient atmosphere and are very

reactive with chemical precursors used in III-nitride growth techniques. This limits the use of

Si(111) substrates for growing GaN layers by hydride vapor phase epitaxy (HVPE) and metal-

organic chemical vapor deposition (MOCVD), where large concentrations of ammonia (NH3) and

hydrogen (H2) are used at high temperatures (> 1000 oC) [40]. Even in plasma-induced molecular

beam epitaxy (PIMBE), where growth temperatures are 250 oC lower than used in HVPE and

MOCVD, the substrate surfaces should be stable under the impact of nitrogen radicals at 800 oC.

The main aspects of these growth methods will be briefed in next section. Also, silicon

lacks long-term chemical stability due to its low resistance to hostile chemicals and to uncontrolled

generation of intrinsic charge carriers. This leads to oxidation of silicon/liquid interfaces, hence

limiting its usage in biosensing devices [23,29]. In biological ambience, Si has an intrinsic flaw as

its surface is always escorted by a native SiO2 layer, which is hydrophilic and biocompatible. Yet

it is permeable to ions prevalent in physiological alkali buffers that diffuse into SiO2 layer, causing

instability under electric fields [29,84].

The hexagonal crystalline (0001)-structure sapphire (c-Al2O3) is the most popular

heteroepitaxial substrate for layers of WZ-GaN. Production of Al2O3(0001) wafers of variable size

is cost-effective and economic, beside advantages like high crystallinity and optical transparency

over a broad spectral range. Further, c-sapphire wafers have good surface quality and high

16

stability at high temperatures and in harsh chemical environments. Nonetheless, there are few

drawbacks of using c-plane sapphire as a bare substrate for Wz-GaN layers. A large difference

in the lattice parameters and thermal expansion coefficients of c-sapphire and III-nitrides, causing

lattice mismatch of 14% for WZ-GaN [17,40]. This can be reduced by using a thin buffer layer

(e.g., thin AlN or AlGaN film) between the WZ-GaN layer and c-sapphire substrate [44,137].

The mismatch between a WZ-GaN layer and c-sapphire substrate leads to creation of

native defects, trapping centers and dislocations in GaN layer. This usually obscures the

investigation of GaN surface functionalized with organo-bio-molecules. Such imperfections and

dislocations are reported to be not large in HVPE-grown c-sapphire/WZ-GaN structures [40,82]

and can be reduced by proper post-growth surface modification and treatment.

The c-sapphire is an insulator with high electrical resistivity that bans the use of backside

contacts in vertical electrical devices. Thus, the c-sapphire/WZ-GaN structures can only be

integrated in lateral (coplanar) electronic devices, such as field-effect high-electron-mobility

transistors (HEMTs), where the c-sapphire/AlGaN/GaN is exploited for detecting and identifying

chemicals, gases and biological species [22-31]. The studied c-sapphire/n-GaN(0001) samples

were commercial HVPE-grown and only subjected afterwards to some cleaning/etching and

functionalization procedures. It is beneficial to brief in this chapter the general features and

advantages/drawbacks of growth methods and chemical precursors used to grow GaN layers,

knowledge of which gives an idea about their surface structure, impurities and defects.

2.5 Growth Methods of GaN Layers

The film deposition methods that are commonly used to grow epitaxial films of III-nitride materials

and III-nitride alloys include MOCVD, PIMBE and HVPE techniques, each of which has its own

merits and curbs. Detailed descriptions of such methods are found in [13,44]. The procedures of

film deposition start with volatile compounds in MOCVD and HVPE or evaporation beams in

PIMBE. The used reactants are transported towards a substrate, on which they are adsorbed,

reacted with each other to form the epitaxial film, with the byproducts being removed.

High-quality WZ-GaN epitaxial films deposited by MOCVD and HVPE on c-sapphire

substrates grow in the (0001) direction with smooth Ga-faced surfaces, while their MBE growth

occurs in the (0001

�) direction with more roughened N-faced surfaces [40].

17

2.5.1 Hydride Vapor Phase Epitaxy (HVPE) Growth Method

HVPE technique is the widely employed method for growing thick GaN epitaxial films because of

its high growth rates of up to a few tens of μm/h and its lack of carbon incorporation into the film

[40,82]. The growth rate in conventional chloride transport VPE of GaN is governed by the

pyrolysis process involving a mixture of gallium monochloride (GaCl) gas which is only stable at

temperatures above 600 oC, and ammonia gas NH3 diluted with a carrier gas [40,136,137].

The production of GaCl gas is obtained via reaction of hydrogen chloride (HCl) gas and

molten gallium metal, whose source is held at constant temperature between 850 oC and 900 oC.

The supply of GaCl gas is controlled by the temperature of the liquid gallium cell and by flow rates

of HCl gas and carrier gas, with a high conversion efficiency (~ 95%) of the reaction

2Ga(l) + 2HCl(g) ↔ 2GaCl(g)+ H2(g) (2.1)

The growth rate for VPE-deposited GaN film depends linearly on partial pressure of GaCl,

substrate temperature, desorption taking place from the substrate surface, formation of GaHx

molecules in the hydrogen ambient, and the flow rates of HCl, NH3, and carrier gas [40]. The

overall reaction for the formation of HVPE-GaN films is given by

GaCl(g) + NH3(g) ↔ GaN(s) + HCl(g) + H2(g) (2.2)

Under deposition rates near 900 μm/h, thick HVPE-GaN films can be grown on c-sapphire

substrates held at temperatures in the range 950 – 1100 oC [40]. Also, the GaCl3 gas can be used

as a Ga-source for growing thick HVPE-GaN films with slight structural defects and dislocations

(< 108 cm−2), as well as low carrier densities (< 1018 cm−3), compared to the heteroepitaxial thin

GaN films grown by MOCVD or MBE [40].

2.5.2 Plasma Induced Molecular Beam Epitaxy (PIMBE) Growth

Method

PIMBE is a multipurpose technique for growing thin epitaxial films of semiconductors, metals and

insulators. Thin films grown by MBE crystalize via reactions between thermal molecular or atomic

beams of their elements and a substrate surface held at a high temperature in UHV conditions.

The growth rate in PIMBE is determined by the flux of evaporated Ga atoms striking the substrate.

It is also governed by temperature and pressure of effusion cell, its orifice size, distance from

substrate, and by the sticking and accommodation coefficients on the substrate [40,136].

18

Compared to HVPE and MOCVD (discussed below), the film growth in PIMBE is generally

conducted at slow rate of pyrolysis of ammonia and at lower substrate temperatures, due to rapid

dissociation of nitrides at high temperatures under UHV conditions. The PIMBE growth process

is carried out in the molecular flow regime, where interaction between Ga atoms and activated

nitrogen ions. The use of low growth rates (< 1μm/h) under Ga-rich conditions results in quality,

smooth GaN epilayer with improved structural and electrical/optical properties [40]. The PIMBE

method and underlying growth modes and parameters, besides advantages and disadvantages

of PIMBE III-nitride epitaxial films and their devices,described in detail in [13,45].

2.5.3 Metal Organic Chemical Vapor Deposition (MOCVD) Growth

Method

MOCVD technique is complex and is still not well understood. The growth process of GaN films

by MOCVD entails the transport of gas phase organometallic precursors, hydrides (NxHy) as

nitrogen source, to a heated substrate on which the precursors are pyrolysed and the GaN film is

formed. The MOCVD technique has the advantages of large-area growth capability, good

coverage of film on substrate, and controlled epitaxial deposition process. Details on the MOCVD

method and used chemical precursors for growing III-nitride layers are given in [40,136].

19

3

Surface Sensitive Characterization Techniques

Highly accurate and sensitive analysis of the composition and chemistry of a substance surface

can be attained by X-ray photoelectron spectroscopy (XPS). The atomic force microscopy (AFM)

and low-energy electron diffraction (LEED) can be used to span the morphology of the surface to

obtain information on its atomic periodicity, symmetry of crystal lattice and roughness.

In this chapter, a description of XPS technique and underlying principle under ultrahigh

vacuum (UHV) and near-ambient-pressure (NAP) conditions is given. The AFM and LEED are

discussed in brief, in addition to the apparatus used for dry-ion sputtering and functionalizing a

solid surface with water (H2O) vapor and L-cysteine molecules.

3.1 X-ray Photoelectron Spectroscopy (XPS)

XPS is a non-destructive technique that gives evidence on the composition of surface elements,

chemical bonding among surface atoms, and surface band bending. UHV-XPS studies on solid

surfaces shed light on their quality and suitability for use in electronic devices. While NAP-XPS

measurements on a solid surface capped with inorganic or organic molecules are valuable for

exploring its capability to form surface chemical bonds that can adhere to species confronted in

chemical and biological sensors [64-69,84-95,113-116,138-141].

The XPS is based on the phenomenon of the photoelectric effect, where electrons are

emitted from atomic core levels and valence bands after X-ray irradiation [44,142-148]. The X-

rays used in XPS have large photon energies whose magnitude depends on the anode of

attached X-ray unit. For accurate XPS analysis, monochromatic X-ray lines with narrow width (<

1 eV) are desirable. The anode of X-ray unit generally emits multi-energy photons, which are

directed to a quartz crystal that diffracts them as a collimated beam of monochromatic photons

on the sample’s surface. Most XPS instruments utilize aluminum (Al) and magnesium (Mg)

anodes whose Al Kα and Mg Kα X-ray lines have photon energies of 1486.7 eV and 1254.6 eV,

respectively.

20

3.1.1 The Three-Stage Photoelectron Model

The features of photoemission spectra (PES) of a solid excited with monochromatic X-rays can

be understood by a phenomenological three-step model comprising of three independent

processes [44,142-148]. A schematic diagram of this model is depicted in Figure 3.1(a) [145],

while a more realistic description of these steps and photoemission spectrum is depicted in Figure

3.1(b) [44,144]. This three-step photoelectron model is briefed here to grasp features of

experimental atomic core levels (CLs) and valence band (VB) XPS spectra of solid surfaces.

(a)

(b)

Figure 3.1: A three-stage model for the photoemission process [145]: (a) a schematic picture of

the three processes: 1) photoexcitation of electrons, 2) electron transport towards surface and 3)

transmission through surface into vacuum, and (b) a diagram showing these processes on

measured spectra [44].

The first stepinvolves absorption of X-ray photons of energy ℎ𝜋𝜋 by electrons in initial states

of atomic core levels (CLs) and VB in a solid and subsequent photoexcitations to final energy

states. The second step is the transport of photoexcited electrons from the sample interior towards

its surface and the primary photoelectrons suffer energy losses via inelastic collisions with sample

atoms and create secondary electrons. The third step involves escape of photoelectrons,

secondary electrons and other photoexcitations from sample’s surface [145]. This occurs only if

photoelectrons with kinetic energy 𝐸𝐸kin surmount sample’s work function ΦS and being ejected

from surface into vacuum with residual kinetic energy 𝐸𝐸K [44,145]. The secondary electrons give

rise to the background signal recorded on experimental XPS spectra of detected photoelectrons.

The intensity 𝐼𝐼(𝐸𝐸K, ℎ𝜋𝜋) of all detected photoemissions is governed by the general function below

𝐼𝐼(𝐸𝐸K, ℎ𝜋𝜋) ∝ 𝑃𝑃(𝐸𝐸K, ℎ𝜋𝜋) 𝑇𝑇(𝐸𝐸K, ℎ𝜋𝜋) 𝐷𝐷(𝐸𝐸K, ℎ𝜋𝜋), (3.1)

21

where 𝑃𝑃(𝐸𝐸K, ℎ𝜋𝜋), 𝑇𝑇(𝐸𝐸K, ℎ𝜋𝜋), and 𝐷𝐷(𝐸𝐸K, ℎ𝜋𝜋) are cross sections of photoexcitation process, transport

of electrons to surface, and escape of photoelectrons, respectively [145]. 𝑃𝑃(𝐸𝐸K, ℎ𝜋𝜋) contains

information on electrons in initial states and photoelectron creation and annihilation, which depend

on atomic number of element, its orbital subshells, and on intensity and energy of incident photons

[44,145,149-152]. Bonding energies of electrons in initial states correspond to binding energies

of CL peaks of XPS spectra . 𝑇𝑇(𝐸𝐸K, ℎ𝜋𝜋) depends on the sample’s absorption coefficient 𝐾𝐾(ℎ𝜋𝜋) and

on electron inelastic mean free path (IMFP), labelled by 𝜆𝜆mp(𝐸𝐸K) [153,154].

In XPS measurements, the information depth depends on the angle Θ between the normal

to sample’s surface and the entrance to analyzing chamber as shown in the schematic Figure

3.2(a) [43]. Thus, XPS measurements at normal emission (NE) (Θ = 0o) give evidence on surface

states down to the maximum depth of atomic layers from which photoelectrons are emitted

(information depth), while higher angles enable more sensitive topmost surface measurements.

Electrons have very short IMFPs in material media due to inelastic collisions as photoelectrons

interact with electrons of atoms and molecules of the material, thus loose energy by inelastic

scattering, but in vacuum they can move over longer paths, depending on their kinetic energy

𝐸𝐸kin. The dependence of electron IMFP on 𝐸𝐸kin follows experimental and theoretical trends as in

Figure 3.2(b) [43,154]. Inside a solid, the electron IMFP relates to the information depth.

(a)

(b)

Figure 3.2: (a) Schematic layout for the angle Θ between the normal to sample’s surface and the

entrance to analyzing chamber (b) IMFP of electrons in various materials with solid dots and stars

are experimental data [44] and the dashed line represents the theoretical function of Seah and

Dench [154].

22

The information depth of an electron of kinetic energy 𝐸𝐸kin (or 𝐸𝐸K) can be estimated from

the IMFP-Vs-𝐸𝐸kin curve shown in Figure 3.2(b), from which IMFP is minimum for 𝐸𝐸K = 20 – 200

eV, with 𝜆𝜆mp(𝐸𝐸K) ≈ 0.1 – 0.7 nm [149-154]. Such values of 𝜆𝜆mp(𝐸𝐸K) match lattice constants of

several solids, so PES with such 𝐸𝐸K values reflect only surface electronic states. For Al Kα X-ray

line, information depth is less than 10 nm, so photoelectrons that suffer no inelastic scattering

yield information on electronic structure of thin surface layers. Larger information depths can be

attained by using higher X-ray energies. Over such a narrow 𝐸𝐸K-range, 𝐷𝐷(𝐸𝐸K, ℎ𝜋𝜋) and 𝑇𝑇(𝐸𝐸K, ℎ𝜋𝜋) ≈

𝐾𝐾(ℎ𝜋𝜋)𝜆𝜆mp(𝐸𝐸K) depend mildly on 𝐸𝐸K [44,145]. Photoelectrons must enter UHV space to ensure

minimal loss in their residual 𝐸𝐸K, which depends on the binding energy 𝐸𝐸B of CLs of their atoms.

Photoelectrons escaped from sample’s surface are focused on the entrance of energy analyzer

that electrically divert them to a counting system for further analysis Figure 3.2(a).

3.1.2 Energy Level Scheme of Photoemission Spectra

The distribution of electrons in CLs and VB of a solid excited by photons of energy ℎ𝑣𝑣 is presented

in terms of the number of photoelectrons 𝑁𝑁(𝐸𝐸) of energy 𝐸𝐸 per unit time per unit area as a function

of 𝐸𝐸K, relative to vacuum level 𝐸𝐸vac, as depicted in Figure 3.3(a) [144a]. To unveil energy levels

from which electrons are photo-ejected, 𝑁𝑁(𝐸𝐸) is displayed as a function of 𝐸𝐸B, which does not

depend on ℎ𝑣𝑣, but on respective CLs and VB states, and chemical environment of atoms. In PES

studies, 𝐸𝐸B is measured relative to the sample Fermi level 𝐸𝐸F (𝐸𝐸B = 0 at 𝐸𝐸F) below 𝐸𝐸vac by ΦS,

with EB being found from the relation

𝐸𝐸B = ℎ𝑣𝑣 − 𝐸𝐸K − ΦS (3.2)

Yet, CL and VB photoelectrons in a sample are recorded by the detector of an energy analyzer

electrically linked to the spectrometer. So, photoelectrons spend a part of their 𝐸𝐸K equals to ΦA −

ΦS to enter the vacuum above 𝐸𝐸F of analyzer of work function ΦA. ΦS depends on type of sample,

while ΦA is constant for a fixed spectrometer setup. Fermi levels of sample and analyzer are

aligned together if the sample is in electrical contact with the spectrometer body, as seen in Figure

3.3(b) [144b]. So, the final photoelectron kinetic energy measured by analyzer is EK = ℎ𝑣𝑣 − 𝐸𝐸B −

ΦS − (ΦA −ΦS)= ℎ𝑣𝑣 − 𝐸𝐸B − ΦA, or

EB = ℎ𝑣𝑣 − 𝐸𝐸K − ΦA (3.3)

Equation (3.3) is fundamental for XPS analysis as the sample work function cancels out. As ΦA

is a constant to be found and ℎ𝑣𝑣 is known, 𝐸𝐸K of a photoelectron determines its 𝐸𝐸B and vice versa.

23

In practice, ΦA in Equation (3.3) is eliminated by calibrating the spectrometer via setting

energy scale to zero at the Fermi edge of a reference metallic sample [142-144]; thus, 𝐸𝐸B = ℏ𝜔𝜔 −

𝐸𝐸K. This is solely valid for conducting samples as the 𝐸𝐸F level is the same for sample and analyzer.

However, for insulators or if charge is accumulated on the sample surface, 𝐸𝐸F is not well defined,

and energy scale of the whole XPS spectrum may shift. Hence, the binding energies of measured

XPS spectrum must be corrected using a well-known CL energy of a surface element.

(a)

(b)

Figure 3.3: (a) A diagram of energy states of atomic CLs and VB in a solid of work function ΦS

(= Φ), after Hüfner [144a] and 𝐸𝐸kin (= 𝐸𝐸k) of their electrons produced by X-ray photons of energy

ℏ𝜔𝜔 = ℎ𝑣𝑣, where ℏ = ℎ⁄2𝜋𝜋 and 𝜔𝜔 = 2𝜋𝜋𝜋𝜋, and (b) scheme of relevant energy terms used in XPS

spectra of solid surfaces, after Hofman [144b].

3.1.3 Core-level Photoemission Spectroscopy

Core-level shells of electrons with 𝐸𝐸B > 10 eV do not contribute to direct bonding between atoms

in a solid, but their binding energies (BEs) are related to electronic and chemical environment of

their atom. Analysis of XPS spectra of CL photoemission lines of a solid determines BEs, and

thus allowing identification of its elements and corresponding percentage compositions. Further,

accurate analysis of a CL spectrum and observed shifts of its emission lines yields information on

the chemical bonding configuration between atoms and on CL splitting due to spin-orbit coupling.

In addition, the accomplished results of such quantitative analysis can be readily exploited to

estimate the amount of band bending at the surface of film/layer under investigation arising from

dangling (free) bonds, charged and energy states, and adsorbates existing on the surface, as will

be described later.

24

XPS spectra of elements on a solid surface may contain, besides CL peaks, co-emitted

Auger-electron peaks and satellite peaks [44,145] that obscure quantitative analysis of XPS

spectra. The chemical environment of an atom on sample’s surface is dissimilar to its bulk

environs as surface atoms undergo electron charge redistribution due to unalike chemical bonds

and electro negativities. This causes re-arrangement of surface charges that induce surface band

bending (BB) and affect adsorption and linkage of contaminants and species on the surface.

3.1.4 VB Spectroscopy and Surface Band Bending

The VB photoemission of a sample reflects its VB density of states (DOS) described by a spectral

function 𝜌𝜌VB(𝐸𝐸K, ℎ𝑣𝑣) related to partial density of states 𝑁𝑁ℓ(𝐸𝐸) of atomic orbital ℓ as 𝜌𝜌VB(𝐸𝐸K, ℎ𝑣𝑣)=

∑ℓ 𝐶𝐶ℓ𝑁𝑁ℓ(𝐸𝐸), where 𝐸𝐸 = 𝐸𝐸K + ΦS − ℎ𝑣𝑣. 𝐶𝐶ℓ is proportional to photo-ionization cross section (PICS)

of electrons in orbital ℓ that has diverse values for different elements [146-152].

The XPS can be utilized to obtain sample’s VB PES whose analysis yields information on

surface energy states, BB, and 𝐸𝐸F pinning. Surface modification and BB depend on substance,

surface contaminants and adsorbed species, native or external. The energy difference η between

VB maximum (VBM) and 𝐸𝐸F can be found from a linear extrapolation of the VB leading edge to

the background base line [43,44,85,130-132]. This approach enables to eliminate the instrumental

broadening tail at low BE side. Photoemission from surface states takes place at BEs lower than

VBM but higher than 𝐸𝐸 F.

Values of η (= 𝐸𝐸 F− VBM) and bandgap energy 𝐸𝐸 g of a sample can be used to find the

surface barrier height potential ΦB(ΦB = 𝐸𝐸 g−η). A positive ΦB indicates an upward BB, while a

negative ΦB signifies downward BB [44]. The VB and CL XPS spectra of a sample can be

combined to determine its surface BB and net concentration of surface polarization charged

states.

3.2 XPS Spectrometer

Figure 3.4 depicts a photograph for an NAP-XPS spectrometer, whose main accessories required

for making UHV and NAP CL and VB XPS measurements are briefly described below. The

chambers of such XPS spectrometer are often connected to oil-free high-vacuum pumps, crossed

by absorbing filters and traps, to eliminate contamination of its interior and samples by

external spurious impurities, gas or oil vapors, dust and greases.

25

Figure 3.4: A photograph of NAP-XPS spectrometer used in present work showing its main parts

of SPECS Phoibos 100 hemispherical energy analyzer and X-ray apparatus.

3.2.1 X-ray Apparatus and Related Accessories

An XPS apparatus usually includes quartz crystal monochromator, anode X-ray source, focusing

capabilities, and water cooling unit. A high-voltage (HV) power supplies and controls voltages and

currents for the X-ray source. The X-ray radiation is produced by the anode, cooled via circulating

cold water during its bombardment by energetic electrons emitted from a heated filament under

UHV conditions (< 10– 7 mbar).

A monochromator with a single quartz crystal is often exploited to filter XPS radiation

emitted from the anode. This monochromatization process can be realized on the basis X-ray

diffraction from a crystalline material. Let a collimated beam of X-rays of wavelength 𝜆𝜆 to impinge

on a crystal of inter-planar lattice spacing 𝑑𝑑 at an angle of incidence 𝜃𝜃 Figure 3.5. Constructive

interference of elastically scattered X-ray radiation from the crystal satisfies the following formula

2 𝑑𝑑 cos 𝜃𝜃1 = 𝑛𝑛 𝜆𝜆 (3.4)

where 𝑛𝑛 is an integer specifying the order of interference of the diffracted peak and 𝜃𝜃1 = 90° − 𝜃𝜃.

In the SPECS 100 XPS system used in the present work, the X-ray apparatus was armed with a

SPECS XR-MF Al anode, a SPECS 500 microfocus incorporating ellipsoidal quartz single crystal

and SPECS CCX 60 water cooling unit.

26

Figure 3.5: Schematic diagram of the ellipsoidal crystal monochromator FOCUS 500 of the

SPECS XR-500 microfocus X-ray assembly.

The X-ray monochromator of SPECS Focus 500 utilizes θ1 = 30° with an average 𝑑𝑑-value

〈𝑑𝑑〉 = 4.255148 Å at 25 oC. For the principal order 𝑛𝑛 = 1 of strong diffracted Al K𝐾𝐾 X-ray line and

using Equation (3.4), 𝜆𝜆 = 0.851028 nm. For Al K𝛼𝛼1 X-rays, ℎ𝜋𝜋 = 1486.70 eV and using the formula

𝐸𝐸(eV) ≅ 1240 (eV .nm) (3.5)

𝜆𝜆(nm)

a 𝜆𝜆-value for the Al K𝛼𝛼1 X-ray line is 0.834 nm Table 3.1. This is not exactly matching the

calculated 𝜆𝜆 = 0.851028 nm. Thus, the angle of incidence θ1 on crystal must be set to satisfy

Bragg’s law and use the proper quartz-crystal lattice constants to get θ1 = 11.525°, which

corresponds to a Bragg’s angle θ of 78.474° for the monochromatic Al K𝛼𝛼1 X-ray line.

Table 3.1: SPECS crystal monochromator key parameters for characteristics X-rays of Al Kα1.

Description

Al K

α1

Energy (eV)

1486.70

λ (nm)

0.833956

Diffraction order

1

2θ1 (θ1 = 90 – θ)

23.05o

Bragg angle θ

78.4740o

27

3.2.2 Near Ambient Pressure NAP-XPS Spectroscopy

NAP-XPS is a technique for exploring chemistry and reforms of solid surfaces functionalized with

inorganic or organic molecules under atmospheric pressures of few mbar. NAP-XPS spectra of

vapor-solid, liquid-solid, and biological-solid interfaces yield information on bonding ensuing on

them, hence more insight on the performance of allied chemical and biological sensors

[115,116,138-141].

The NAP-XPS is based on a modification to UHV-XPS to work under NAP conditions and

to permit photoelectrons to enter the analyzing chamber via a miniature nozzle. To reduce

damage to the components of UHV-XPS setup, a series of differential pumping stages with

electrostatic lenses are fitted between nozzle and analyzing chamber. This reduces the pressure

back down to UHV of the analyzer space, limits the path of electrons in high pressure regions,

and directs them to detection system. The sample surface is placed close to the nozzle in the

NAP region, so permitting a large number of photoelectrons to enter nozzle to detector with

minimal loss in kinetic energy.

The design of NAP-XPS systems reduces sample preparation and pumping times, without

any load-lock kit between working and analyzer chambers. In SPECS 100 XPS system, NAP-

XPS measurements can readily be made in the working chamber due to the nozzle’s special

design that separates it from UHV analyzer space. This setup allows in-situ heating, ion-gas

sputtering, and water functionalization of sample’s surface, besides making LEED measurements

in the working chamber without affecting the components of UHV analyzer chamber.

3.2.2.1 SPECS PHOIBOS 100 Hemispherical Energy Analyzer

SPECS PHOIBOS 100 hemispherical electrostatic energy analyzer (HEEA) allows recording of

XPS spectra for charged particles with kinetic energies up to 3.5 keV in UHV (< 10−9 mbar). The

PHOIBOS 100 analyzer consists of an UHV housing, where the HEEA components are installed.

The components of the HEEA are a 180o hemispherical analyzer (HAS) of 100 mm radius for

measuring spectroscopic energy, an electrostatic system of pre-lens and a series of lenses for

receiving charged particles, and a special detection assembly. Since moving charges are

directionally diverted by a magnetic field, the HAS analyzer with lens system, detector unit and its

electronics are shielded from nearby external static and low-frequency magnetic fields. This is

done via surrounding them by two 1.5-mm thick layers of µ-metal and via constructing the

analyzer and lens system from non-magnetic materials. The HAS accessories include HAS 3500

28

plus power supplies, electrical and mechanical connectors and tubing, flanges, computer

software.

The photo-electrons first pass through an electron lens system and a tiny slit before

entering HAS. The electron lens system and sizes of entrance and exit slits affect energy spread

detected by analyzer. The pre-retarding-lens stages define the analysis area and acceptance

angle to give ultimate transmission of incoming electrons at the entrance slit. The particles will

then be focused by the voltage of the HAS capacitor on its output plane, coupled to a multi-

channel detection system that comprises of microchannel plates (MCP) and detector (3D-4040-

100 DLD). This enables simultaneous recording of an energy band of charged particles around

the nominal pass energy by the DLD electronics and interface of SPECS data acquisition

software. A diagram of lens system, hemispherical capacitor, electrons detection facilities and

voltage supplies is shown in Figure 3.6 [SPECS catalogues].

Figure 3.6: A diagram [SPECS catalogues] of lens system, hemispherical capacitor and voltage

supplies of SPECS PHOIBOS 100 HEEA, with r0 = 100 mm = R0 cited in text.

Pulse counting and amplification are attained by the MCP array of the SPECS 3D DLD

electronic system (x, y, t), whose in-vacuum read out unit consisting of two isolated and 90o-off

meander structured delay lines (DLD anode). The delay lines are positioned behind a chevron

MCP stack, biased relative to DLD anode, to amplify incoming electrons, with each hit position is

encoded by a fast data acquisition electronics controlled by a SpecsLab software.

29

The entrance-slit S1 and exit-slit S2 planes of PHIOBOS 100 HEEA are centered at radius

R0 = 100 mm = (Rin + Rout)/2. For an electrical field gradient, only particles with KEs in a certain

energy interval are able to pass through the full deflection angle from S1 to S2. Particles with

higher KE move further toward outer hemisphere (OH), whereas those with lower KE are focused

further inside on S2 plane toward inner hemisphere (IH). Particles enter HAS normal to S1 and

moving along central trajectory of radius 𝑅𝑅0 are focused at S2 plane with nominal pass 𝐸𝐸pass

𝐸𝐸pass = −𝑞𝑞 k ∆𝑉𝑉 (3.6)

where 𝑞𝑞 is the particle charge, ∆𝑉𝑉 is potential difference 𝑉𝑉out − 𝑉𝑉in across IH and OH shells, and

k is a calibration constant having the value (𝑅𝑅in 𝑅𝑅out)⁄2 𝑅𝑅0(𝑅𝑅out − 𝑅𝑅in) = 0.9375 Figure 3.6.

In PHIOBOS 100 systems utilize double entrance slits to restrict the width of the particle

beam along the central path and to minimize the angle of particle distribution at S1, hence a good

energy resolution ΔEan. If W1 and W2 are widths of S1 and S2 and α is analyzer acceptance angle,

then

W1 + W2 α2

∆𝐸𝐸an ≅ 𝐸𝐸pass � + � (3.7)

4𝑅𝑅0 4

The integral signal intensity 𝐼𝐼(𝐸𝐸k) at a certain kinetic energy 𝐸𝐸k of measured particles, the area

under peak of an XPS photoemission line with the background being subtracted, is given by

2

𝐸𝐸pass 𝐸𝐸pass

𝐼𝐼(𝐸𝐸k)= ∆𝐸𝐸anΩ0𝐴𝐴0 ~ (3.8)

𝐸𝐸k 𝐸𝐸k

where Ω0 and 𝐴𝐴0 are, respectively, the acceptance solid angle and area at S1 that are analyzer

constants. Knowing ΔEan is essential for interpreting CL and VB XPS spectra, whose acquisition

can be performed by HEEA in two modes: fixed retarding ratio (FRR) mode and fixed analyzer

transmission (FAT) mode. The FAT mode is used in XPS to get better energy reolution by

selecting low pass energy (𝐸𝐸pass) as ∆𝐸𝐸an in the FAT mode remains independent of 𝐸𝐸k across the

entire measured 𝐸𝐸k-range. The resolution ∆𝐸𝐸an links to the full width at half maximum (FWHM) of

spectrometer instrumental function 𝑤𝑤XPS. In SPECS XPS system with an Al X-ray source, 𝑤𝑤XPS =

(𝑤𝑤x2 + 𝑤𝑤s2)1/2

, where 𝑤𝑤s is a Gaussian function of the spectrometer and 𝑤𝑤x is a Gaussian profile

for Al Kα X-ray line of FWHM = 0.167 eV. Equation (3.7) works reasonably for HEEA in FAT mode

with ∆𝐸𝐸an

~ 𝑤𝑤s, and 𝐸𝐸pass and ∆𝐸𝐸an are adjustable constants, so 𝐼𝐼(𝐸𝐸K)~ 1/𝐸𝐸K for same ∆𝐸𝐸an,

2

regardless of 𝐸𝐸K. If 𝐸𝐸K is constant, 𝐼𝐼(𝐸𝐸K)~ 𝐸𝐸pass for the same peak.

30

If a photoelectron line taken by SPECS XPS system is fitted to a single Voigt profile, which

is a convolution of Lorentzian (L) profile of width 𝑤𝑤L and Gaussian (G) profile of width 𝑤𝑤G, the total

energy width ∆𝐸𝐸 of this line is ∆𝐸𝐸 = (𝑤𝑤x2 + 𝑤𝑤s2 + 𝑤𝑤G

2 + 𝑤𝑤L

2)1/2

. Thus, to get high energy resolution,

monochromatic X-rays is favored for excitation. In FAT mode, ∆𝐸𝐸an ≈ 𝑤𝑤s = S . 𝐸𝐸pass, with S can be

found from values of 𝑤𝑤G of a single line for different pass energies and curve-fit of 𝑤𝑤G-data to the

22

formula 𝑤𝑤G = 𝐶𝐶 + 𝑆𝑆2

. 𝐸𝐸pass, where 𝐶𝐶 is a constant independent of 𝐸𝐸pass.

3.2.2.2 Ion Sputtering Gun of SPECS Phoibos 100 XPS

In XPS measurements, one may need to sputter a solid surface by a beam of energetic ions of

inert or reactive gas. SPECS Phoibos 100 XPS systems install a SPECS IQE 12/38 sputter ion

gun. This ion gun consists of a base rotatable DN38CF flange, pumping ports, a cathode of yttrium

oxide (Y2O3)-coated iridium (Ir) ring filament, two lens system for adjusting spot size (0.2-1 mm),

and a power supply with integrated scan and deflection unit (SPECS PU-IQE 12/38).

The SPECS IQE 12/38 ion source is equipped with an UHV gas inlet system with leak

valve, a differential pumping system, and a Wien mass filter that can be retrofitted to the source.

The differential pumping allows a large pressure difference between UHV working chamber and

ion source, so it works at a maximum ion beam current. The IQE 12/38 ion gun uses Y2O3-coated

Ir ring filament that emits electrons to ionize gas ingredients. The ions are accelerated to KE in

the range 0.2-5 keV with ion beam of adjustable current density (1-4 mA/cm2). The Y2O3-coated

Ir-filament allows low emission temperatures, thus reducing ion beam contamination and allowing

ionizing reactive or inert gases. The filament is non-line-of-sight to the sample, so the ion beam

passes thru a Wien mass filter to remove impurities from ion beam, to reduce contamination.

In practice, the process of sputtering a material surface with gas ions (or atoms) is intricate

since the real physical and physiochemical interactions between the surface constituents and the

gas species are not aptly tracked [82a, 82b]. The process is governed by surface structure, type,

energy, intensity and incidence angle of gas atoms. Interpretation of the results of an analysis of

the structure of a gas ion/atom sputtered surface and of residual ingredients and by-products is

not understood. The sputter rate 𝑍𝑍/𝑡𝑡 (nm/s) and the number of the removed monolayers of a

sputtered surface, can be estimated if the primary ion current density jp(A/m2), sputtering time

𝑡𝑡(s) and sputter yield S (atom/ion) are known, using the expression [SPECS data sheets]

𝑍𝑍⁄𝑡𝑡 = 𝑀𝑀 𝑆𝑆 jp⁄𝜌𝜌mNA𝑒𝑒 (3.9)

31

where NA is Avogadro number, 𝑒𝑒 is electronic charge, 𝑀𝑀(kg/mol) and 𝜌𝜌m(kg/m3) are the molar

mass and density of the target material. The sputter yield depends on the primary energy 𝐸𝐸, type

of gas, and on surface atoms interacted with gas. For metals, 𝑍𝑍⁄𝑡𝑡 is found from experimental and

theoretical data/curves for sputter yield and inert/reactive gases. Formulas and curves for 𝑆𝑆(𝐸𝐸) of

compounds shelled with inert/reactive gas do not exist. For GaAs and few oxides, the 𝑆𝑆-values

range is 0.5-4 for 5-keV Ar-ions. For GaN layers shelled with 0.5-keV Ar ions at normal incidence,

𝑆𝑆 − E curves showed diverse sputter yields for N (𝑆𝑆 = 0.5) and Ga (𝑆𝑆 = 0.1) atoms.

3.3 Low Energy Electron Diffraction (LEED)

LEED is a surface analysis technique based on electron diffraction from a sample’s surface. LEED

measurements must be made in UHV (< 10−9 mbar) to minimize electron inelastic collisions with

air molecules as they have short IMFP in air, but can move over longer paths in UHV, depending

on their Ekin Figure 3.2(b). In the LEED apparatus, electrons emitted from a hot filament are

accelerated and focused on a sample’s surface by a system of electrical lenses. The electrons

diffracted from surface are collected by another series of hemispherical grids on a fluorescent

screen, where diffraction pattern is displayed in Figure 3.7.

LEED electrons typically have 𝐸𝐸kin (= 𝑝𝑝2/2𝑚𝑚e) between 20 eV and 200 eV, where 𝑝𝑝 and

𝑚𝑚e are electron’s momentum and mass. This low 𝐸𝐸kin corresponds to a de Broglie wavelength 𝜆𝜆e

(= ℎ/𝑝𝑝) = 0.1 – 0.3 nm, where ℎ is Planck’s constant. These 𝜆𝜆e values are comparable to the

atomic spacing in a crystal lattice, electron diffraction from lattice [43,44]. Inside solids, LEED

electrons have short IMFP over which electrons do not undergo collisions Figure 3.2(b). Only

electrons diffracted from topmost surface layers escape out and reach detector without inelastic

collisions. For LEED electrons with KE= 20-200 eV, IMFP is 0.4-0.6 nm, making LEED sensitive

to surface reconstruction and roughness with information depth covering at most 2-3 monolayers.

32

Figure 3.7: LEED setup with its main parts and working principle [43].

The LEED diffraction patterns reflect the symmetry of periodicity of scattering sites on the

sample’s surface, with the intensity of diffracted spots satisfies a two-dimensional Laue condition

[155,156]. Qualitative evidence on a surface is attained from the size, shape and brightness of

LEED diffracted spots. Large, weak and diffuse diffracted spots signify rough surface with poor

atomic periodicity, as for a polycrystalline sample, whose surface defects or adsorbates lead to

blurred LEED images, while no LEED image is visible for an oxidized surface. Yet, clean, smooth,

and well-ordered surfaces exhibit LEED images with bright and small diffracted spots [44].

3.4 Atomic Force Microscopy (AFM)

The AFM images surface topography at atomic level of samples and measures surface properties

with little preparation [43,157]. Figure 3.8(a) shows a photograph of the AFM model used in the

present study. In conventional AFM systems shown in Figure 3.8(b), the PSPD controls vertical

motion of a piezoelectric x-y-z-scanner and prevents crushing of cantilever tip. Yet, this AFM

system suffers from low speed of mechanically-driven cantilever, difficulty of getting all surface

parts, non-linearity and hysteresis. The cross coupling of scan axes in conventional AFM systems

forbids independent movement in x, y and z directions. This drawback is eliminated by physical

separation of Z-scanner from X-Y scanner to get independent vertical and horizontal movement

of the cantilever tip, as in the XE-series scanning probe microscopes (SPMs), where its motion is

scanned at a rate of 10–50 Hz for X-Y scanner and 10 kHz for Z-scanner without cross coupling.

This enables AFM measurements with high speed, minimal error, and slight non-linearity. The

XE-100 SPM was exploited to attain AFM images of the surface of studied GaN layers.

33

(a)

(b)

XE Head

Z Stage

Objective Lens

CCD Camera

Optical Microscope Fram

SPM Frame

XY Scanner

Figure 3.8: (a) A photograph of used AFM XE-100 SPM instrument and (b) a schematic diagram

showing the operational principle of conventional AFM apparatus based on a piezoelectric

ceramic tube scanner.

34

4

Experimental Details

Extensive investigation of HVPE-grown c-sapphire/n-GaN(0001) surface exposed to ion-gas

sputtering and water/L-cysteine functionalization helps to realize the behavior of n-GaN(0001)

layers when integrated in gas and biosensing devices. This aim was achieved by using UHV/NAP-

XPS technique to measure spectra of core-level (CL) transitions of constituents of GaN surface

and its valence band (VB), and by AFM and LEED techniques to obtain its surface microscopic

images and topography patterns. Modifying the GaN surface of c-sapphire/n-GaN(0001) samples

was attained by nitrogen-ion (N2

+) sputtering prior to functionalization process. The preparation

steps effected to expose the N2

+-sputtered n-GaN(0001) surface to water (H2O) and L-cysteine

molecules under different conditions are presented here as well. The procedures used to

characterize the N2

+-sputtered n-GaN(0001)/H2O and n-GaN(0001)/L-cysteine interfaces via

measuring the respective NAP-XPS CL/VB spectra at room temperature (RT) and higher, as well

as taking its AFM/LEED images are also discussed here.

Some of the studied c-sapphire/n-GaN(0001) samples are labelled hereafter according to

their N2

+-ion sputtering, H2O and L-cysteine functionalization procedures such as GaN-AsRec-X,

GaN-Sput-X, N2

+-sputtered GaN-H2O-X, and N2

+-sputtered GaN-LCys-X, where X = 1, 2, 3, 4, etc.

The GaN-AsRec-X denotes an as-received sample, whose n-GaN(0001) surface was not

exposed to in-situ treatment inside the XPS system. The GaN-Sput-X samples are those exposed

to in-situ N2

+-ion sputtering. The N2

+-sputtered GaN-H2O-X and N2

+-sputtered GaN-LCys-X

represent samples that were functionalized with H2O and L-cysteine molecules under various

experimental conditions after N2

+-ion sputtering such as the water-vapor pressure, the sample

temperature, etc., which will be discussed in detail below.

4.1 Materials and Chemicals

4.1.1 Samples

The as-supplied c-sapphire/n-GaN(0001) samples initially studied in the present work were of

different geometric thicknesses, sizes, and quality that were prepared by the hydride vapor phase

epitaxy (HVPE). A systematic investigation of different c-sapphire/n-GaN(0001) samples has

35

been conducted to see which type is the most appropriate for effective surface treatment and for

their feasibility to be functionalized with inorganic and organic molecules.

A few of the studied c-sapphire/n-GaN(0001) samples were supplied by Kyma company

(USA) with HVPE-grown n-GaN(0001) films having thickness of 5 µm and were unintentionally n-

type doped with RT bulk electron density 𝑛𝑛 = 5 x 1017cm−3

. These Kyma samples were large

sheets that did not fit on the stainless steel (SS) or molybdenum (Mo) compartment built in the

XPS system. Thus, they had to be cut into small pieces of 1x1-cm2 size at Dortmund Technical

University (TU) using a diamond knife so that they could be closely placed on a molybdenum (Mo)

holder. This ensured that the backside of the sapphire substrate of these c-sapphire/n-GaN(0001)

samples was thermally-anchored on the Mo-holder edges Figure 4.1.

The approximate 1x1 cm2 sizable Kyma c-sapphire/n-GaN(0001) samples cut with the

diamond knife collected surface carbon-like (charcoal) sediments from the water inlet fed during

cutting process. Such induced surface residuals were avoided when the large Kyma sheets were

cut to 1x1 cm2 in size by a laser machine at Bochum Technical University (TU), but still distressing.

The experimental results attained in the present work show that etching and cleaning of the n-

GaN(0001) surface by argon ions (Ar+) and by H*-plasma were not encouraging and unsuitable;

thus, omitted from this thesis. Sputtering with 1 keV N2

+-ions was appropriate for cleaning the

surface of studied n-GaN(0001) layers.

Purchasing other brands of 1x1 cm2 sizable c-sapphire/n-GaN(0001) samples avoid

cutting procedures made on Kyma samples. The 1x1-cm2 c-sapphire/n-GaN(0001) samples from

SurfaceNet company (GmbH, Berlin) were cost-effective, received swiftly, and convenient. Their

n-GaN(0001) layers were 430-µm thick, Ga-polar wurtzite (WZ-) hexagonal structure,

inadvertently doped with n = 5 x 1017 cm–3, and grown by the HVPE technique. Such HVPE-grown

n-GaN(0001) layers had a small threading dislocation density (~ 107cm−2), due to lattice

mismatch between the c-sapphire wafer and WZ n-GaN(0001) layer.

4.1.2 Gas Supply and Sample Mounting

In the XPS setup, several key components were integrated to facilitate our experiments. These

components included cylinders containing nitrogen (N2) and argon (Ar) gases, equipped with

vacuum-tight valves for sputtering the n-GaN(0001) surface within the ultra-high vacuum (UHV)

environment. Additionally, a Pt100 sensor was employed to continuously monitor and measure

the sample's temperature. Furthermore, we incorporated a vacuum-tight VAT sonic valve to

36

enable the precise injection of H2O molecules onto the sample surface. To enhance our

experimental capabilities, we also integrated a vacuum-tight SPECS OME 40 effusion cell, which

allowed us to deposit L-cysteine onto the n-GaN(0001) surface. For sample mounting and

temperature control, a molybdenum (Mo) sample holder was utilized and a set of thin Mo foils.

These components played a crucial role in achieving and maintaining the desired sample

temperature, which was carefully controlled using a SPECS IRLH 150 infrared laser unit.

4.1.3 Molybdenum Sample Holder

A Mo holder was designed and constructed to get proper housing for c-sapphire/n-GaN(0001)

samples. The PT100 sensor was anchored closely to the surface of n-GaN(0001) layer to attain

controlling and recording of its temperature. The thin Mo foils beneath the sapphire wafer is

heated by the IRLH 150 infrared laser. Figure 4.1 show photographs of the Mo holder and its

parts.

Figure 4.1: The Mo holder annotated diagram shows the c-sapphire/n-GaN(0001) sample, spacer

(Mo foil) fixed under the sample, main holder frame, fixer, PT100 sensor with connections, and

the heat shield.

4.2 Ex-/In-Situ Cleaning and Etching of GaN Surface

Cleaning and sputtering the n-GaN(0001) surface aid its functionalization with inorganic or organic

molecules to get n-GaN(0001) surface susceptible for the linkage of incubated chemical and

biological molecules. The surface of as-grown n-GaN(0001) layers is polluted with adventitious

carbon (C) and oxygen (O2) species during fabrication, and storing stages or during preparation.

Cleaning n-GaN(0001) surface can be achieved via ex-situ and/or in-situ handling procedures.

37

4.2.1 Ex-Situ Surface Cleaning

Physical cleaning of c-sapphire/n-GaN(0001) samples was made outside XPS machine (ex-situ

treatment) by sequentially 15-min immersion and sonication in acetone, methanol, and

deionized water, and flushed with Ar gas to reduce weakly-adhered surface impurities, described

in details in Appendix B.

4.2.2 In-Situ Heating of c-sapphire/n-GaN(0001) Samples

The c-sapphire/n-GaN(0001) sample placed on the Mo foil in the Mo home-constructed holder

Figure 4.1. The use of metal holders ensured good contact between the sample and the body of

the XPS instrument; thus, realizing electrical grounding of the sample to help energy calibration

of experimental XPS spectra. The holder design also allowed direct heating and permitted full

access for ion-gas sputtering, functionalization, and X-ray irradiation of the n-GaN(0001) surface.

The Mo holder and Mo foil were first cleaned in sequence by acetone, methanol, and

deionized water and dried at 100 oC. The sample was firmly affixed to the Mo foil-Mo holder

assembly and flushed with Ar gas before insertion in XPS (~ 10−8 mbar). Heating of n-GaN(0001)

samples was made using a water-cooled IRLH 150. To attain homogenous heating, the sample

was tightened on the Mo foil firmly on the edges of the rectangular-shaped opening of the Mo

holder, so the IR radiation from IRLH 150 source can be focused on the backside of the Mo foil.

Heating was required for samples exposed to H2O or L-cysteine molecules. The PT100 sensor

thermally anchored to the n-GaN(0001) surface monitors and measures its temperature within a

tolerance of ± 0.5 oC.

4.2.3 In-Situ Ion Sputtering and Etching of GaN Surface

Sputtering GaN surface with argon ions (Ar+) is an etching route used for reducing surface C and

native oxides [82a,82b]. The surface of n-GaN(0001) layers of this work exposed to a beam of

Ar+ ions of various kinetic energies (0.5-3 keV) for 5-40 min. The samples is mounted in the XPS

chamber so the n-GaN(0001) surface directly faced the Ar+ beam. Sputtering with 1-keV Ar+ ions

for < 10 minutes caused severe damage and deteriorated the surface, as unveiled from measured

AFM images; thus, Ar+-ion sputtering was not used further in this work.

Etching of the n-GaN(0001) surface of c-sapphire/n-GaN(0001) (SurfaceNet, Berlin)

samples by nitrogen gas (N2

+) ions was a proper route for removing a good fraction of its native

+

contaminants; thus, N2

+-ion sputtering route was regularly adopted during this work. Beams of N2

38

ions of various kinetic energies (0.5-3 keV) for 5 to 14 minutes were tried. The optimum dose and

kinetic energy of N2

+ ions suitable for sputtering n-GaN(0001) surface were a 10 mA beam and 1-

keV N2

+ ions for < 14 minutes. Bombarding the n-GaN(0001) surface with N2

+ ions was marginally

surface devastating [82b] and reduced its C species but not much of its oxides, as noted from its

O 1s and C 1s XPS photoemission lines. The corresponding LEED images were slightly weak

and blurred, implying that the N2

+-ion beam caused some surface roughness with little damage.

In-situ N2

+-sputtering of n-GaN(0001) surface by a collimated 10-mA beam of 1-keV N2

+ for

10 -14 minutes was conducted inside the UHV-XPS working chamber (≲ 10−9 mbar). Details of

procedures used for N2

+ sputtering of a typical sample, its alignment positions, and used X-ray

source setting for XPS measurements are given in Table 4.1.

Table 4.1: Preparation of typical c-sapphire/n-GaN(0001) (SurfaceNet, Berlin) samples under

various conditions.

µ-focus: X-ray setup for

UHV-/NAP-XPS

Sample position and

orientation (UHV-XPS)

Voltage

Emission current

14 kV

9 mA

θ(o)

ϕ(o)

x(mm)

0

0.620

y(mm)

-19.828

Power

126 W

z(mm)

9.904

Specification Sample GaN-AsRec-1

pressure for UHV-XPS

5.30 x 10−9 mbar

Specification

N

2

+-ion kinetic energy

Sputter time

Sample GaN-Sput-1 (1st sputtering)

1 keV

540 s

pressure for UHV-XPS

4.70 x 10−9 mbar

Specification

y-position thru sputter

N

2

+-ion kinetic energy

Sputter time

pressure for UHV-XPS

Sample GaN-Sput-2 (2nd sputtering)

-19.501 mm

1 keV

210 s

3.40 x 10−9 mbar

39

4.3 In-Situ Functionalization of GaN Surface

Two types of functionalization procedures were conducted on the HVPE-grown c-sapphire/n-

GaN(0001) samples: capping the n-GaN(0001) surface with few monolayers of H2O or L-cysteine

molecules. Exposure of n-GaN(0001) surface to H2O molecules proceeded in the working

chamber during NAP-XPS runs. The L-cysteine monolayers were adsorbed on n-GaN(0001)

surface in the preparation chamber. The aim was to investigate the stability, suitability, and

susceptibility of H2O/n-GaN(0001) and L-cysteine/n-GaN(0001) interfaces to accommodate and

detect inorganic and organic molecules in GaN-based gas sensors and biosensing devices.

4.3.1 Functionalization of GaN Surface with H2O Molecules

The surface of as-received and 1-keV N2

+-sputtered n-GaN(0001) layers was exposed to H2O

molecules at low pressures, keeping the sample at a fixed temperature. Various temperatures

were used (23, 40, 60, 80, 90, 100, 130, and 160 oC), at each NAP-XPS spectra of n-GaN(0001)

surface were collected under the H2O pressures 0.02, 0.05, 0.08, 0.1, 0.5, 0.8, and 1 mbar.

A stainless steel cylinder was purged in sequence several times with acetone, ethanol,

and deionized microfilter water. After that the cylinder was filled with deionized water of 200 mL

and the cylinder was frozen in liquid nitrogen. The frozen cylinder was connected to a turbo

molecular pump to evacuate it to 5 mbar and then defrosted the water. During this procedure all

gases in the cylinder will evaporate. This procedure was repeated 4 to 5 times. Then the deionized

water was gradually injected into the working XPS chamber through a tight-fitted sonic VAT valve

until the desired H2O-vapor pressure was reached.

Before H2O injection, the IRLH 150 laser used to heat up the sample to the desired

temperature, which was monitored and recorded by the Pt100 sensor anchored In thermal contact

to the GaN surface. After steadiness of chamber pressure and sample temperature,

measurements of the CL and VB NAP-XPS spectra of the n-GaN(0001)/H2O interface were

conducted at the selected constant sample temperature for all the above-cited H2O-vapor

pressures in succession.

Prolonged pumping of the chamber to UHV (10−9 mbar) was then carried out to remove

H2O vapor and gasses that might have been released from the XPS chamber and out of the n-

GaN(0001) surface, after which the UHV-XPS CL and VB measurements were made on the same

sample. The purpose of NAP-XPS measurements was to explore the reformation, surface

chemistry, stability, and interactions that took place on the ensuing n-GaN(0001)/H2O interface

40

under different experimental conditions (pressure and temperature). Table 4.2 illustrates the pre-

treatment and measurement conditions chosen for the samples functionalized with H2O vapor at

pressures ≥ 0.1 mbar. Similar CL and VB NAP-XPS measurements were also acquired at lower

pressures (0.02, 0.05, and 0.08 mbar) for the same temperatures.

Table 4.2: The voltage, emission current, power of X-ray source, and alignment setting of the

GaN surface of c-sapphire/n-GaN(0001) samples described in Table 4.1 for functionalization with

H2O vapor of various pressures at different sample temperatures. The symbol Dp designates the

pressure of the detector (analyzer) chamber, whereas the scale reading is the scale number of a

water pump.

µ-focus of X-

Voltage

Emission current

Power

ray setup

14 kV

9 mA

126 W

Specification

Sample GaN-H

2

O-1 (after 2nd sputtering)

Sample

temperature

pressure

(mbar)

0.1

0.5

0.8

1.0

0.1

again

UHV

Scale

7342

7370

7387

7392

7337

------

23 oC reading

D

p

(10−8

mbar)

1.54

5.36

8.25

9.55

1.78

0.533

Scale

7337

7369

7388

7395

7336

------

60 ℃

reading

D

p

(10−8

mbar)

1.7

5.57

7.70

10.1

1.85

1.71

160 ℃

Scale

reading

D

p

(10−8

mbar)

7333

1.75

7365

6.02

7385

8.09

7389

10.5

7335

1.95

------

0.795

+-

sputtered n-GaN(0001) surface using different new, fresh samples, with each being only used

once at a certain temperature. Sample heating was attained by the IRLH in UHV working chamber

to get the required temperature. After it was settled, H2O vapor was introduced to attain the

targeted pressure. Then, NAP-XPS measurements on the H2O/n-GaN(0001) interface were

made. At the same temperature and sample, more H2O vapor was added to reach the next

pressure, at which NAP-XPS data was then collected, and so on till the final H2O-vapor pressure.

The chamber was then pumped down overnight to UHV at the given temperature before acquiring

In addition NAP-XPS runs were also conducted with functionalization of 1-keV, 14 min N2

41

the UHV-XPS spectra of its n-GaN(0001) surface next day. The same procedures and NAP-XPS

measurements were made on a new, fresh sample that was maintained at another temperature.

4.3.2 Functionalization of GaN Surface with L-Cysteine

It has been observed that functionalization of the surface of n-GaN(0001) layers of the c-

sapphire/n-GaN(0001) (SurfaceNet, Berlin) samples with L-cysteine was effective when the n-

+

GaN(0001) surface was sputtered in prior with 1-keV N2 ions for sufficient times (9-14 min).

Experiments made in this part proceeded by leaving the as-supplied c-sapphire/n-GaN(0001)

sample in the UHV-XPS load-lock chamber overnight for degassing. The sample was then

installed in XPS working chamber to acquire UHV-XPS survey spectrum, core-level, and VB

spectra of its n-GaN(0001) surface.

As a comparison, the surface of the GaN layer of as-supplied c-sapphire/n-GaN(0001)

sample was not sputtered in prior with N2

+ ions and was directly installed to the XPS preparation

chamber for functionalization with L-cysteine. The crucible of the OME 40 effusion cell containing

the L-cysteine powder was heated gradually to 105 oC, at which it was kept for 40 min in ambient

of ~ 10−8 mbar. Afterwards, the sample was transferred to the UHV-XPS working chamber to

take the UHV-XPS survey spectrum of the non-sputtered n-GaN(0001) surface, as well as its CL

and VB XPS spectra. No traces of L-cysteine were observed in its respective XPS spectra.

Few of two as-supplied c-sapphire/n-GaN(0001) samples (GaN-AsRec-2 and GaN-

AsRec-3) were chosen, for functionalizing the surface of their n-GaN(0001) layers with L-cysteine

after bombarding with low-energetic N2

+ ions. LEED images and UHV-XPS spectra of the non-

sputtered n-GaN(0001) surface were first acquired, after which it was exposed to a collimated

beam of 1-keV N2

+ ions for 9 min. The resulting N2

+-sputtered n-GaN(0001) layers were re-labelled

as GaN-Sput-2 and GaN-Sput-3, on which LEED and UHV-XPS measurements were then

conducted on their N2

+-sputtered n-GaN(0001) surface. The preparation and setting details of

these samples for above-mentioned measurements are listed in Table 4.3.

The n-GaN(0001) surface of each of N2

+-sputtered GaN-Sput-2 and GaN-Sput-3 samples

were functionalized with L-cysteine in the XPS preparation chamber under various conditions, as

detailed below. The n-GaN(0001) surface deposition with L-cysteine molecules was achieved via

sublimation of L-cysteine powder stored in the crucible of a SPECS OME 40 effusion cell.

Deposition of L-cysteine on the n-GaN(0001) surface was commenced after gradual heating of

the crucible of the OME 40 effusion cell to a selected temperature between 95 oC and 105 oC.

42

Various thicknesses of L-cysteine deposited on the n-GaN(0001) surface were attained by varying

the crucible temperature and sublimation time, chosen in the range 20-55 minutes. Afterwards,

each L-cysteine functionalized sample was transferred to the XPS working chamber to take UHV

XPS spectra first for the sample held at RT. Next, the L-cysteine/n-GaN(0001) sample was heated

gradually to the required sample temperature in the range 75-150 oC, at each of which the sample

was annealed for a chosen time. After completing annealing of the sample, it was then left to cool

down to RT before taking UHV-XPS spectra of the n-GaN(0001)/L-cysteine interface.

Table 4.3: Preparation and XPS setting details of the samples GaN-AsRec-2, N2

+-sputtered GaN-

Sput-2, GaN-AsRec-3, and N2

+-sputtered GaN-Sput-3 before implementing L-cysteine

functionalization on them.

µ-focus: X-ray setup

Voltage

Emission current

Power

for UHV-XPS

14 kV

9 mA

126 W

Sample temperature

Room temperature (RT)

for XPS measurements

Specification

Sample GaN-AsRec-2

Sample GaN-AsRec-3

Working chamber

1.60 x 10−9 mbar

1.17 x 10−8 mbar

pressure for UHV-XPS

Specification

Sample GaN-Sput-2

Sample GaN-Sput-3

y-position thru sputter

-20.60 mm

-20.80 mm

𝑁𝑁

2

+-ion kinetic energy

1 keV

Sputter time

540 s

Working chamber

2.0 x 10−9 mbar

1.54 x 10−8 mbar

pressure for UHV-XPS

The first stage of the procedure adopted for the coating n-GaN(0001) surface of studied

c-sapphire/n-GaN(0001) samples with L-cysteine monolayers was carried out as follows. The

sample was aligned inside the XPS preparation chamber (𝑝𝑝 ~ 2× 10−9 mbar), so that the outer n-

GaN(0001) surface faced directly the opening of SPECS OME 40 effusion cell. Prior to L-cysteine

evaporating, the OME 40 crucible and the L-cysteine powder were cooled down to 20 ℃ via a

water-cooling system. The crucible was then heated up to reach a temperature sufficient for

sublimating L-cysteine but low enough to exclude thermal decomposition of L-cysteine molecules.

43

A compromise between the actual melting point and thermal decomposition temperature

of L-cysteine is not straightforward, depending on its purity. The most frequently reported melting

point of L-cysteine is between 170 oC to 240 oC. In this work, the temperature of the crucible with

its L-cysteine powder was raised gradually at 5 oC /min up to 105 °C, which was found to be a

proper temperature for sublimating L-cysteine molecules towards n-GaN(0001) surface. The L-

cysteine deposition lasted at 105 °C for 20 minutes on the GaN-Sput-2 sample, which was then

labelled as GaN-Lcys-2. The estimated thickness of its deposited L-cysteine was about 2.4 nm,

determined by analyzing the corresponding Ga2p CL spectra using the Quases-Tougaard (QT)

program (installed in the used XPS instrumentation).

For the GaN-Sput-3 sample, the L-cysteine deposition also proceeded at 105 ℃, but for a

deposition time of 40 minutes to increase the thickness of deposited L-cysteine. The resulting L-

cysteine functionalized sample was labelled as GaN-Lcys-3, for which the thickness of L-cysteine

on its n-GaN(0001) surface was around 4.2 nm, determined also by Quases-Tougaard analysis

of the corresponding CL Ga 2p XPS spectrum of its n-GaN(0001)/L-cysteine interface. The reason

behind depositing high L-cysteine monolayers was to infer from respective UHV-XPS spectra

what bonding mechanisms might have taken place among the L-cysteine monolayers and at the

n-GaN(0001)/L-cysteine interface. For a third N2

+-sputtered c-sapphire/n-GaN(0001) sample, the

L-cysteine heating was carried out at 95 ℃ for 50 minutes, but no L-cysteine traces were observed

from respective UHV-XPS spectra.

After each L-cysteine deposition process, the crucible was cooled down to 19 ℃, via a

water-cooled heat sink, and the L-cysteine functionalized c-sapphire/n-GaN(0001) sample was

left overnight in the preparation chamber under UHV (~ 10−9 mbar). Next, the L-cysteine

functionalized GaN sample was moved to the working chamber to take UHV-XPS spectra before

its heating and IR annealing. Once the chosen final annealing temperature 𝑇𝑇A was reached, the

sample was sustained at this temperature for the required annealing time, which was chosen to

be 15, 30, and 60 minutes at each 𝑇𝑇A. After completing sample annealing, the IRLH 150 laser

was switched off, and the sample was left to cool down to RT before taking UHV-XPS

measurements on the annealed L-cysteine functionalized n-GaN(0001) surface.

The IR heating and annealing response of the GaN-Lcys-2 sample that was functionalized

with L-cysteine at 105 oC for 20 minutes in the XPS preparation chamber will be discussed. The

IRLH 150 laser was used to heat it up constantly in the XPS working chamber from RT to a

maximum temperature 𝑇𝑇A =75 ℃, 100 ℃, and 130 ℃, at each of which the sample was annealed

44

for the times 𝑡𝑡A = 15, 30, and 60 min. Table 4.4 shows the increasing trend of the measured

sample temperature 𝑇𝑇(℃) as the current 𝐼𝐼L(A) through the IRLH 150 laser heater is increased to

reach the final annealing temperatures 75 ℃, 100 ℃, and 130 ℃.

Table 4.4: Sequential heating stages and annealing procedures carried out on the L-cysteine

functionalized GaN-Lcys-2 sample at temperatures 𝑇𝑇A = 75 ℃, 100 ℃, and 130 ℃, each for 𝑡𝑡A =

15, 30, and 60 minutes.

Specification

Sample GaN-Lcys-2

Time

𝑡𝑡

A

𝑇𝑇

A

𝐼𝐼

L

(A)

0

15 min

𝑇𝑇 (℃)

𝑝𝑝 (10−9)

mbar

24.9

1.45

𝐼𝐼

L

(A)

0

30 min

𝑇𝑇 (℃)

𝑝𝑝 (10−9)

mbar

24.2

1.27

𝐼𝐼

L

(A)

0

60 min

𝑇𝑇 (℃)

24.2

𝑝𝑝 (10−9)

mbar

1.09

1.7

31.1

1.46

1.7

30.2

1.29

1.7

28.6

1.09

1.8

34.5

1.47

1.8

33.4

1.29

2.7

75.2

1.2

1.9

75 ℃

2.0

40.6

43.5

1.49

1.50

2

2.2

41.7

50.1

1.30

1.33

2.1

48

1.52

2.7

75

1.45

2.3

55.5

1.55

2.5

64.8

1.65

2.7

75

1.79-2

100 ℃

0

1.7

2

2.4

2.8

3.05

3.2

24.1

29.4

44.2

62.6

81.0

92.5

100

1.08

1.10

1.14

1.23

1.33

1.64

0

2

2.5

3

3.2

24.6

42.3

66.4

93.6

100

1.19

1.18

1.2

1.35

1.55

0

3.2

24.3

101

1.13

1.22

130 ℃

0

2.5

3

3.5

3.9

24.6

61.2

82.4

105

131

1.04

1.05

1.06

1.11

1.65

1.6

2

2.5

3

3.5

3.9

25.4

37.9

50.2

84.4

106

131

1.02

1.03

1.04

1.07

1.31

0

2

2.5

3

3.9

3.92

24.7

44.3

64.1

82.5

125

130.4

1.04

1.05

1.14

1.22

A plot 𝑇𝑇(℃)-𝐼𝐼L(A) data of Table 4.4 is depicted in Figure 4.2, which shows good linearity

of heating via IRLH 150 laser. After finishing IR-heating/annealing cycle at each 𝑇𝑇A and time 𝑡𝑡A,

the UHV-XPS spectra were taken on n-GaN(0001)/L-cysteine interface of the GaN-Lcys-2 sample

to clarify the chemical modification and bonding mechanisms occurring at the n-GaN(0001)/L-

cysteine interface. To investigate the effect of high IR-annealing temperatures on chemical

45

interactions taking place on the n-GaN(0001)/L-cysteine interface, the GaN-LCys-3 sample was

annealed at 𝑇𝑇A = 100 ℃, 130 ℃, and 150 ℃, for the same time (𝑡𝑡A = 30 minutes), details of such

IR-heating/annealing procedures are cited in Table 4.5 and illustrated in Figure 4.3.

Table 4.5: Temperature 𝑇𝑇(℃) of L-cysteine functionalized GaN-Lcys-3 sample as a function of

current 𝐼𝐼(A) of IRLH 150 heater up to 𝑇𝑇A = 100 ℃, 130 ℃ and 150 ℃, for the same annealing time

𝑡𝑡A = 30 minutes.

Specification

Sample GaN-Lcys-3

Time 𝑡𝑡

A

𝑇𝑇

A

𝐼𝐼

L

(A)

𝟏𝟏𝟎𝟎𝟎𝟎 ℃

𝑇𝑇 (℃)

𝑝𝑝 (10−9)

mbar

𝐼𝐼

L

(A)

30 min

𝟏𝟏𝟏𝟏𝟎𝟎 ℃

𝑇𝑇 (℃)

𝑝𝑝 (10−9)

𝐼𝐼

L

(A)

mbar

𝟏𝟏𝟏𝟏𝟎𝟎 ℃

𝑇𝑇 (℃)

𝑝𝑝 (10−9)

mbar

0

2.0

3.0

3.2

3.4

3.5

24.7

37

62.5

84.5

92.4

100

5.04

5.16

5.80

5.96

6.09

0

4.9

26.4

5.05

0

154

5.52

3

3.5

3.92

4.15

4.2

4.27

24.7

68.6

82.9

111

120

125

130

4.07

4.10

4.12

4.37

4.61

4.83

5.01

: 15 min annealing-after heating to 75 oC

: 15 min annealing-after heating to 100 oC

125

: 30 min annealing-after heating to 100 oC

: 30 min annealing-after heating to 131 oC

: 60 min annealing-after heating to 130.4 oC

100

75

50

25

1.0 1.5 2.0 2.5 3.0 3.5 4.0

IRHL 150 Laser current (A)

Figure 4.2: A plot of measured temperature 𝑇𝑇(℃) of GaN-Lcys-2 sample as a function of the

electric current 𝐼𝐼L(A) of the IRLH 150 laser heater up to the annealing temperatures 𝑇𝑇A = 75 ℃,

100 ℃, and 130 ℃.

Sample temperature (

o

C)

46

Sample temperature (

o

C)

150

125

100

75

50

25

t

A

= 30 min

T

A

: 100

o

C

: 130

o

C

: 150

o

C

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

IRLH 150 Laser current (A)

Figure 4.3: Temperature 𝑇𝑇(℃) of GaN-Lcys-3 sample versus electric current 𝐼𝐼L(A) of IRLH 150

laser heater up to final annealing temperatures 𝑇𝑇A = 100 ℃, 130 ℃, and 150 ℃, each for 𝑇𝑇A equals

to 30 minutes.

4.4 Characterization Procedures and Measurements

The experiments made on the n-GaN(0001) surface were concerned with measurements of low

energy electron diffraction (LEED) and XPS spectra inside the UHV working chamber (in-situ

experiments), besides atomic force microscopic (AFM) imaging of the n-GaN(0001) surface taken

outside the XPS system (ex-situ experiment). The UHV and NAP-XPS measurements provided

us with XPS spectra of core-level (CL) transitions of the constituents existing on the surface of

the n-GaN(0001) layer and of respective valence band (VB). The XPS data facilitated probing

chemical bonding mechanisms and ensuing interactions at the surface. Further, the XPS spectra

can also be analyzed to determine surface chemical compositions before and after ion-gas

sputtering and after functionalizing the surface with H2O or L-cysteine molecules. The AFM

images and LEED patterns of the n-GaN(0001) surface of c-sapphire/n-GaN(0001) samples were

measured and utilized to reveal its roughness, topography, and structure.

47

4.4.1 Sample Alignment/Positioning in Working Chamber

The sample in the XPS working chamber was regularly mounted on a translational and rotatable

compartment held by movable stainless steel (SS) bar. This facilitated positioning and orientation

of the c-sapphire/n-GaN(0001) samples according to the type of investigation required, such as

subjecting the sample to heat treatment, or sputtering the surface of its GaN layer by dry gas ions,

or collecting LEED patterns and XPS spectra of its n-GaN(0001) surface.

To collect UHV or NAP-XPS spectra of c-sapphire/n-GaN(0001) sample, the sample was

adjusted to ensure apt irradiation of its GaN surface with the X-ray beam. It was properly oriented

and positioned in front of a conic nozzle affixed to the pre-lens system of the electron analyzer of

the UHV analyzing chamber. This was done to ensure maximum output yield on the detector at

the output exit plane of the analyzer. The optimal signal was attained by positioning and alignment

of the SS compartment in the XPS working chamber.

For in-situ LEED measurements, the samples position and orientation were adjusted to

face the electron gun so that a collimated beam of low-energy (≤ 160 eV) electrons hit its GaN

surface. The position coordinates and angular orientation of the samples mounted on the

compartment being installed inside the used XPS machine was regularly re-set to attain focused

LEED patterns. To achieve apt sputtering of the GaN surface by N2

+ ions only the y-coordinates

was often re-adjusted, with the other coordinates and orientation being kept as in Table 4.2.

4.4.2 Atomic Force Microscopic (AFM) Imaging

The surface roughness of GaN surface before and after sputtering with 1 keV N2

+ ions were ex-

situ studied by the AFM XE-scanning probe microscope (SPE) of PSIA corporation. The ex-situ

AFM imaging of GaN surface taken on different lateral spots of (2×2) µm2 in size showed a

relatively smoothed surface with compactly oriented structural topography, and surface

roughness of root mean square (RMS) < 1.5 nm. The roughness of studied GaN surface was

noted to increase slightly after sputtering the surface with 1 keV N2

+ ions. Measured AFM images

of the GaN surface of c-sapphire/n-GaN(0001) samples were noted to portray rare traces of

island-like structure and few surface dislocations, which is a characteristic feature of HVPE-grown

n-GaN(0001) layers on c-plane sapphire.

48

4.4.3 Low Electron Energy Diffraction (LEED) Imaging

The morphology and structure of GaN surface were studied by LEED patterns using SPECS

LEED apparatus installed in XPS working chamber under UHV conditions (≲ 2 x 10−9 mbar). A

collimated beam of <160 eV electrons was shot on the GaN surface, with the sample being

positioned as already described. The diffracted electron beam was scanned on the surface and

collected by the LEED detector and attached camera.

4.4.4 X-ray Photoelectron Spectroscopic Measurements

Two types of XPS experiments were conducted on studied c-sapphire/n-GaN(0001) samples.

+

One involved UHV-XPS measurements on as-received n-GaN(0001) surface, sputtered by N2

ions, and on N2

+-sputtered n-GaN(0001) surface coated with L-cysteine molecules. The other type

of XPS experiments was concerned with NAP-XPS measurements on N2

+-sputtered n-GaN(0001)

surface after exposure to low-pressure H2O vapor under different conditions as already described.

4.4.4.1 UHV-XPS Measurements

The c-sapphire/n-GaN(0001) sample affixed to the Mo holder was oriented and positioned as

previously discussed to attain the maximum possible signal from the photoelectrons ejected from

its GaN surface. The base pressure in the working chamber during UHV-XPS measurements was

around 10−9 mbar. The setting of the μ-focus X-ray apparatus power supply and electrical monitor

for the used Al-K𝛼𝛼 1486.71 eV X-ray photons exciting the GaN surface were 14 keV, 9 mA and

power 126 W. The measured UHV-XPS spectra of X-ray irradiated GaN surface was of CL

photoemission lines of several surface ingredients, viz. O 1s, C 1s, N 1s, Ga 2p, Ga 3p, and Ga

3d transitions, in addition to UHV-XPS spectra of its VB structure.

After finishing XPS measurements on as-received c-sapphire/n-GaN(0001) samples, their

n-GaN(0001) surface was subjected to two in-situ N2

+-ion sputtering procedures under UHV

conditions. The first was effected by a collimated beam of 1 keV N2

+-ions for 9 minutes and the

second for 3.5 minutes. Afterwards, UHV-XPS measurements were conducted on the N2

+-

sputtered n-GaN(0001) surface. UHV CL and VB XPS measurements were also carried out on

the surface of n-GaN(0001) layers that were functionalized with L-cysteine molecules under a

variety of preparation conditions as described above. The features and quantification details of all

of the obtained UHV CL/VB spectra are presented in chapter 5.

49

4.4.4.2 NAP-XPS Measurements

The NAP-XPS measurements performed on N2

+-sputtered n-GaN(0001) surface exposed to H2O

molecules inside the working chamber at various pressures between 0.02 mbar and 1 mbar at a

variety of sample temperatures (23 ℃− 160℃). H2O molecules were often injected into the XPS

working chamber once the base pressure was around 10−9 mbar.

First, for the sample at RT and after completing the injection of H2O at a lower vapor

pressure, NAP-XPS measurements were conducted on the resulting n-GaN(0001)/H2O interface.

With the sample being at RT, more H2O was injected to get higher settled pressures, and NAP-

XPS measurements were then repeated at each pressure. Later on, the working chamber was

pumped down for a long time to achieve UHV conditions before UHV-XPS measurements were

made on the n-GaN(0001) surface of the same sample. After keeping the same sample in UHV

for a few days, the above procedure of water injection and taking NAP-/UHV-XPS measurements

on the n-GaN(0001)/H2O interface of this sample was repeated but for higher temperature.

At a later stage of the work, NAP-XPS measurements were conducted on the GaN surface

of a c-sapphire/n-GaN(0001) sample maintained at a constant temperature in each run and

varying the H2O-vapor pressure in the range described above. The procedure was repeated for a

new fresh sample held at another temperature. The corresponding CL and VB NAP-XPS spectra

are discussed and quantified in chapter 5.

The XPS studies often involve measurements of the survey XPS spectra taken with high

pass energies (80 eV in present work) to collect peaks of CL transitions that correspond to all

surface constituents and bonding configurations over the whole energy range of the used X-ray

radiation emitted from the used source (0-1500 eV when the Al-K𝛼𝛼 (1486.7 eV) line is used). As

the energy resolution of XPS survey spectra is low, it is not appropriate for quantitative analysis

of individual CL XPS photoemission lines. The convoluted Lorentzian (natural line broadening

due to finite life time) and Gaussian (spectrometer resolution and sample inhomogeneities)

profiles combine together as a mixed GL-profile (Voigt profile) and affect the spectral width of

measured XPS photoemission lines. The signal of individual XPS photoemission lines were thus

collected in this work using a pass energy of 20 eV with a spectrometer resolution of about 0.6

eV. Analysis of these individual peaks of CL transitions were carried out using Voigt line-shape

or Lorentzian asymmetry LF line-shape functions, as seen in Appendix A.

50

5

Experimental Results and Discussion

Several studies were conducted on GaN films/layers under various preparation and experimental

conditions in effort to modify surface properties and morphology to adapt to different electrical,

chemical and biological applications [50a-c,63-69b,82b-86,163-168]. The reported results are not

all consistent and debatable views were proposed to interpret the response and chemistry of

modified GaN surface. This is due to dissimilar modification approaches and to the miscellaneous

functionalization ambient and procedures used to alter morphology and chemistry of GaN surface.

To bypass the adverse effect of by-products and residual species on GaN surface treated

by chemical solutions [44,69b,73-75,158-160] and to avoid negative impact of Ar+-ion sputtering

on n-GaN(0001) surface [82b,82c], the surface of as-received HVPE-grown n-GaN(0001) layers

studied in this work was only sputtered by the less destructive 1-keV N2

+ ions [82b]. Sputtering of

n-GaN(0001) surface by 1-keV N2

+-ions induced little surface rounghess and removed surface

adventitious carbon (C) species, but marginally reduced the initial surface Ga-oxides as inferred

from measured UHV-XPS spectra of C 1s and O 1s lines of N2

+-sputtered n-GaN(0001) surface.

From the observed results of literature studies [67,83-86,93-97,139-132,163-168], it

appears that oxygen and Ga-oxides, hydroxyl ions, surface states and adsorbates play a crucial

or adverse role in the stability, morphology, functionalization, and band bending of GaN surface.

Thus, a systematic study of UHV-/NAP-XPS CL lines and valence band (VB) spectra of studied

N2

+-sputtered n-GaN(0001) surface, N2

+-sputtered n-GaN(0001)/H2O and n-GaN(0001)/L-cysteine

interfaces was conducted under a variety of meticulous preparation and experimental conditions.

The surface of studied N2

+-sputtered n-GaN(0001) layers was exposed to H2O molecules

in sequence at the temperatures: 23 oC, 40 oC, 60 oC, 80 oC, 90 oC, 130 oC, and 160 oC, at each

the same sample was upheld throughout the NAP-XPS runs taken at the H2O-vapor pressures:

0.02 mbar, 0.05 mbar, 0.08 mbar, 0.1 mbar, 0.5 mbar, and 1 mbar. A series of individual NAP-

XPS runs had been also conducted at all these pressures on the N2

+-sputtered n-GaN(0001)/H2O

interface of a batch of different, fresh c-sapphire/N2

+-sputtered n-GaN(0001) samples, with each

sample being sustained at a different temperature from the above-cited temperatures. This

detailed collective procedure of functionalization of the N2

+-sputtered n-GaN(0001) surface with

51

H2O molecules under numerous H2O-vapor pressures and sample temperatures was not reported

in almost all literature studies.

+-

sputtered n-GaN(0001)/H2O interface unveil a notable effect of Ga-oxides on its chemistry. This

was inferred from the quantitative analysis of respective O 1s and Ga 2p3/2 photoemission lines,

which showed an enhancement in the O/N (O-Ga-OH/Ga-N ≅ O/N) ratio upon injection of H2O

molecules. Such an observation can be understood through presuming that H2O molecules

adsorbed on the N2

+-sputtered n-GaN(0001) surface dissociate into OH– and H+ species that bind

to dangling Ga+ and N– ions to form Ga-OH and N-H bonds [85, 89,93,168]. Another origin behind

increasing O/N ratio is the passivation of n-GaN(0001) surface by the chemically-inert Ga2O3

oxide that becomes notable at high annealing temperatures, at which the less stable (oxygen-

deficient) Ga2Ox molecules and the free Ga atoms oxidized to form thin G2O3-oxide layer

[50b,50c].

The results of obtained UHV-/NAP-XPS spectra of VB and photoemission lines of the N2

These arguments enable understanding of the observed variational trend of band bending

(BB) at the N2

+-sputtered n-GaN(0001) surface upon exposure to H2O molecules such that a

prominent decrease in the BB voltage 𝑉𝑉bb with increasing H2O-vapor pressure (𝑝𝑝) and sample

temperature (𝑇𝑇) was perceived. In this work, 𝑉𝑉bb was found by two methods. One includes finding

the surface valence band maxim (VBM) energy (𝐸𝐸F − 𝐸𝐸VS) by the linear extrapolation method,

where 𝐸𝐸F is the Fermi level and 𝐸𝐸VS is the surface valence band edge. The other exploites the

peak energy 𝐸𝐸Ga 3dS of the surface Ga 3d5/2 photoemission line, relative to 𝐸𝐸F, in conjunction with

the energy difference (𝐸𝐸VB − 𝐸𝐸Ga 3dB) between the Ga 3d level and VB level of bulk GaN, whose

optimum value 17.5 eV was adopted in this work. The symbols S and B refers to surface and bulk.

Both methods include bandgap energy 𝐸𝐸g(𝑇𝑇) and Fermi energy 𝐸𝐸F(𝑛𝑛, 𝑇𝑇), below conduction

band (CB) edge (𝐸𝐸C), and their variation with 𝑇𝑇, with 𝑛𝑛 the electron density of n-GaN(0001) layers.

The 𝑇𝑇-dependence of 𝐸𝐸g(𝑇𝑇) was found from the Varnishi equation and that of 𝐸𝐸F(𝑛𝑛, 𝑇𝑇) was found

from the formulas based on Boltzmann distribution function incorporating 𝑛𝑛 and 𝑁𝑁C(𝑇𝑇), the

effective density of states in CB. Similar BB methods were adopted in other studies but for a single

𝑝𝑝 and 𝑇𝑇 and for samples of dissimilar features [85,95,13-132,164-167,172,173]. The approach

based on surface VBM energy is not accurate enough as it is limited by the presence of surface

states and suffers from ambiguity in applying the linear extrapolation method for fluctuating XPS

data of VB leading edge at high 𝑝𝑝. The approach based on the features of Ga 3d level is more

subtle as the peak energy 𝐸𝐸Ga 3dS of surface Ga 3d5/2 line can be accurately determined.

52

Functionalizing N2

+-sputtered n-GaN(0001) surface with L-cysteine molecules was made

under UHV (𝑝𝑝 < 10−8 mbar) via sublimation of solid L-cysteine powder from a heated crucible of

an organic molecular effusion cell (SPECS OME 40 cell) installed inside XPS preparation

chamber. The used temperature of the OME-40 crucible was 105 oC, much below decomposition

temperature of L-cysteine molecules, which was constantly kept during the deposition times 20

and 40 minutes to get varying thickness of the deposited L-cysteine monolayers. The obtained

N2

+-sputtered n-GaN(0001)/L-cysteine samples were then heated gradually in succession in the

XPS working chamber to 25 oC, 75 oC, 100 oC, 130 oC and 150 oC, at each of which the samples

were annealed for 15 min, 30 min, and 1 h before collecting the room-temperature UHV-XPS

spectra of the thermally-annealed N2

+-sputtered n-GaN(0001)/L-cysteine interface.

There is no literature study being conducted on N2

+-sputtered n-GaN(0001) surface, on

which L-cysteine is deposited via the above-described route, but all were exposed to L-cysteine,

other peptides, proteins, and organic molecules dissolved in solutions under acid and/or alkaline

conditions [49,50a,51,57,64,66,68,69]. The UHV-XPS spectra of Ga 3s/S 2p and N 1s lines of

N2

+-sputtered n-GaN(0001)/L-cysteine interface show that dangling Ga ions bind to SH-group of

initial chemically adsorbed L-cysteine monolayer and to oxygen in the unstable Ga2Ox species.

The functional NH3 and NH2 groups form bonding configurations such as zwitterioinc and non-

zwitterioinc in the upper physisorbed L-cysteine monolayers, linked poorly among themselves.

Bonds between Ga atoms and sulfur atoms of thiol (-SH) functional group in the L-cysteine

molecule broke down for N2

+-sputtered n-GaN(0001)/L-cysteine samples subjected to prolonged

annealing at high temperatures (≥100 oC), and in turn oxidized to Ga2Ox oxide. The upper L-

cysteine monolayers appear to mask the chemisorbed Ga-S-H… bonding underneath, but upon

annealing these Ga-S-H…. bonds regained their original features, due to the removal of the

topmost physisorbed L-cysyeine molecules and/or to the breaking of their weak intermolecular

bonds, as well as to structural changes in these L-cysteine layers. The results of functionalzing

N2

+-sputtered n-GaN(0001) surface with L-cysteine molecules give a clue on the viability and

efficacy of n-GaN(0001) surface as a platform for binding organic molecules, hence on the

performance of biological sensing devices integrating GaN layers.

The detailed results of experiments made throughout the course of my PhD project are

presented, analyzed and discussed. The samples were commercial HVPE-grown wurtzite (WZ-)

GaN(0001) layers deposited on hexagonally-oriented crystalline sapphire (c-Al2O3) substrates.

The GaN(0001) layers were unintentionally n-doped with a room-temperature (RT) electron

53

concentration of 𝑛𝑛 = 5𝑥𝑥1017𝑐𝑐𝑚𝑚−3

. Some of the c-sapphire/n-GaN(0001) samples were large

sheets, whose WZ-GaN layers were 5-µm thick and had to be cut to nearly 1×1 cm2 in size to fit

in the sample holder of the X-ray photoelectron spectroscopy (XPS) instrument. For more

convenience and to bypass technical artifacts introduced by the cutting of such sheets, the

successive studies were made on c-sapphire/n-GaN(0001) samples of square wafers of 1×1 cm2

in size, whose 430 µm thick n-GaN(0001) layer also had the aforesaid electron concentration.

The effect of ex-situ wet physical and chemical cleaning and/or etching on the structure

and composition of the n-GaN(0001) surface of some of the as-received c-sapphire/n-GaN(0001)

samples was preliminary explored. Physical cleaning involved rinsing the sample in successive

ultrasonically agitated baths of pure acetone, methanol, and deionized water to remove surface

greases and loosely attached contaminant species, followed by drying them with a flow of Argon

(Ar) gas. The wet chemical etching has been reported to be essential for removing native oxide

monolayers from the n-GaN(0001) surface, on which a thin overlayer of gallium (Ga) native oxides

is often formed due to oxygen adsorbing from the moist ambient atmosphere [70-77]. However,

etching of the n-GaN(0001) surface with even diluted chemical solutions is not always

recommended because it usually alters the morphology and composition of the surface, as well

as it normally leads to the formation of pitting structures, residual impurities, and contaminants

arising from chemical reaction by-products remaining on the surface [72-76].

The properties of the n-GaN(0001) surface are highly sensitive to the etching time and

temperature and to the type and concentration of the etchants (inorganic acids and alkaline

solutions) used. Unwanted modifications and large structural roughness are induced on the n-

GaN(0001) surface upon etching with KOH or other alkali solutions or with HCl solutions [44,158-

160]. Thus, even the less harmful ex-situ treatments of the n-GaN(0001) surface of the as-

received c-sapphire/n-GaN(0001) samples of this work (wet physical cleaning and chemical

etching with diluted HCl solutions) were not performed. To obtain informative experimental

structural and compositional results, the ex-situ treatment of the as-received c-sapphire/n-

GaN(0001)samples was limited to flushing them thoroughly with pure argon (Ar) gas. AFM images

presented in Appendix B.

Other treatment procedures were endeavored to clean the surface of the n-GaN(0001)

layers of this work. In-situ sputtering with even low-energy Ar+ ions was found to induce severe

surface damage and island-like/aggregation regions, in agreement with other literature findings

Error! Bookmark not defined.[82]. This was inferred from the atomic force microscopy (AFM)

54

images and the ultrahigh vacuum (UHV) low energy electron diffraction (LEED) patterns of these

Ar+-sputtered n-GaN(0001) layers. On the other hand, in-situ sputtering of the surface of the n-

+

GaN(0001) layers with low energy (≤ 1 keV) N2 ions was more suitable, less damaging, and

yielding AFM images with fair surface roughness. Large AFM surface roughness (> 1 nm) and

+

unclear UHV LEED patterns with diffuse/blurred diffraction spots were observed upon using N2

+

ions of higher energy for prolonged sputtering of the surface of the n-GaN(0001) layers. Thus, N2

ions having an energy of 1 keV were used to sputter the surface of these layers for appropriate

time periods (9 – 13 minutes) to avoid surface damage and unwanted surface modifications. The

UHV LEED experiments were performed using electrons of low kinetic energy (100 – 200 eV) to

prevent blurring, damaging or destroying the structure of the n-GaN(0001) surface with high

energy electrons.

Thermal heating/annealing of some of the c-sapphire/n-GaN(0001) samples was carried

out inside the XPS machine (in-situ heat treatment) for the purpose of surface cleaning and

etching. The n-GaN(0001) surface of the thermally-annealed samples, however, became dirty

and blackened by the auxiliary carbon content of the heated parts of the XPS system and/or from

the resistive-heating unit itself. The results of the measurements made in this study on the

thermally-annealed n-GaN(0001) surface even if followed by N2

+-sputtering or vice versa were too

ambiguous to obtain information from them. Accordingly, cleaning by in-situ resistive-heating and

annealing of the c-sapphire/n-GaN(0001) samples was discarded, and measurements performed

on such samples will not be presented in this thesis.

The results of experiments performed on the c-sapphire/n-GaN(0001) samples studied

during my PhD research using AFM, UHV LEED, and UHV and near ambient pressure (NAP-

XPS) techniques are presented and discussed in this chapter. The outcomes of measurements

made on as-received c-sapphire/n-GaN(0001) samples after bombarding the surface of their n-

GaN(0001) layers by N2

+ ions are briefed first and scrutinized to reveal the modifications induced

on the structure and chemistry of n-GaN(0001) surface by N2

+-ion sputtering to assess its

suitability for functionalization with inorganic and organic species. The main part of this study is

allocated for the results of investigation of interactions taking place on the N2

+-sputtered n-

GaN(0001) surface of these samples after exposure to H2O and L-cysteine molecules under

various experimental conditions.

55

+-5.1 AFM Images and LEED Micrographs of As-received and 𝐍𝐍𝟐𝟐

Sputtered n-GaN(0001) layers

Illustrative AFM and LEED measurements performed on the n-GaN(0001) surface of as-received

c-sapphire/n-GaN(0001) samples and on their n-GaN(0001) surface after sputtering by N2

+ ions

are presented in this section. The measured LEED micrographs exemplify diffraction patterns

whose features depend on the cleanness and structure of the surface and on the energy of the

electrons used in the LEED experiment, which affect the appearance and intensity of the diffracted

spots. The AFM images reveal the smoothness and structure of the surface.

5.1.1 In-situ UHV LEED Micrographs

LEED micrographs of the n-GaN(0001) surface of the as-received c-sapphire/n-GaN(0001)

samples upon sputtering with a collimated beam of 1-keV N2

+ ions for 9 and 13 minutes, as well

as those of the N2

+-sputtered n-GaN(0001) surface after exposure to H2O molecules are presented

in this sub-section. Cutting the as-received c-sapphire/n-GaN(0001) large sheets to 1 × 1 cm2 in

size with a knife resulted in an unclean surface, with LEED images either disappearing or showing

weak and blurred diffraction spots. When these samples were cut by a fine laser beam, a clean

n-GaN(0001) surface was obtained and respective LEED patterns displayed structured, bright

diffraction spots. These samples were cut by a fine laser beam to get a clean n-GaN(0001) surface

as revealed from respective LEED patterns as seen from Figure 5.1.

It can be inferred from LEED patterns in Figure 5.1 that the atoms on the surface of n-

GaN(0001) layers of as-received samples are arranged in a 2D hexagonal structure and give rise

to bright diffraction spots, indicating that the examined surface is not much disordered,

contaminated and/or oxidized [43,44]. As the the kinetic energy of electrons striking the surface

increases, the 2D ring of diffracted sixsets becomes closer to the central spot (smaller angle of

diffraction due to decreasing electron wavelength), with the outer diffracted spots of less intensity

are propably belonging to higher diffraction orders. Similar LEED patterns were also measured

for the c-sapphire/n-GaN(0001) samples supplied with exact 1×1 cm2 size to bypass the efforts

of cutting large samples and to achieve direct mounting on the compartments installed in the

chambers of used XPS machine.

56

92 eV

150 eV

158 eV

Figure 5.1: LEED patterns of the n-GaN(0001) surface of an as-received sample cut by a fine

laser beam.

LEED images of n-GaN(0001) surface sputtered with 1-keV N2

+ ions for 9 and 13 minutes

are shown in Figures 5.2(a) and 5.2(b) using 150-eV electrons. The LEED images of 9-min, 1-

keV N2

+-sputtered n-GaN(0001) surface displays clear diffraction spots, but are disappeared for

13-min sputtering. Figure 5.2(c) depicts the LEED image of 9-min, 1-keV N2

+-sputtered n-

GaN(0001) surface when exposed to H2O molecules at room temperature (RT=23 °C) and 1-

mbar pressure, the LEED diffraction spots became faint and blurred due to water coverage of the

sample surface.

(a)

(b)

(c)

150 eV

+

Figure 5.2: LEED patterns of the n-GaN(0001) surface sputtered by 1-keV N2 ions for (a) 9

minutes and (b) 13 minutes. The LEED image shown in part (c) is for 9-min, 1-keV N2

+-sputtered

n-GaN(0001) surface after exposure to H2O molecules at RT (23 °C) and an H2O-vapor pressure

of 1 mbar.

57

Similar trend has been noticed when the surface of the n-GaN(0001) layer was coated

with fairly thick coverage of L-cysteine layers, after which no LEED spots were observed.

However, the spots reappeared upon heating the sample inside the UHV-XPS working chamber

to temperatures near 100 oC. This may be understood as follows: the deposited L-cysteine (layer)

molecules randomly mask the scattered atoms on the original uncoated surface, and evaporation

(desorption) of a fair part of L-cysteine molecules takes place upon heating around 100 oC.

5.1.2 Ex-situ AFM Images

Figures 5.3(a) and 5.3(b) depict, respectively, the AFM topography of micro-size spots (2.5 × 2.5

µm2) on the n-GaN(0001) surface of as-received HVPE-grown c-sapphire/n-GaN(0001) samples

and after sputtering by 1-keV N2

+ ions for 9 minutes. The topography of the surface of the as-

received n-GaN(0001) layer looks fairly smooth with few pits and a terrace-like undulated

structure and a root-mean-square (RMS) roughness of 0.644 nm. Sputtering of the GaN surface

by N2

+ ions of low energy (≤ 1 keV) was not much , yielded a clean surface, and creates extra

surface rougness RMS = 0.8 nm. Sputtering with higher energy N2

+ ions resulted in AFM images

with a large RMS roughness (> 5 nm) and unclear LEED images with diffuse and blurred

diffraction spots.

Figure 5.3(b) shows that the AFM image of the 1-keV N2

+-sputtered n-GaN(0001) surface

was slightly modified such that the terrace-like undulated surface structure exhibits a few bulging

protrusions with slightly larger RMS (0.8 nm), part of which can be related to uncertainties in the

background subtraction of the individual AFM images. These AFM images show no substantial

traces of cracks, or segregation (island-like) regions, or threading dislocations.

(a)

(b)

58

Figure 5.3: Typical ex-situ AFM images measured on (2.5 × 2.5) µm2 spots on the n-GaN(0001)

surface (a) of an as-received sample and (b) after sputtering the n-GaN(0001) surface with 1-keV

N2

+ ions for 9 minutes.

Also, ex-situ AFM images were taken on SurfaceNet c-sapphire/n-GaN(0001) samples of

alike electron concentration, but having exact 1×1 cm2 size and thicker n-GaN(0001) layer (430-

µm thick) and whose n-GaN(0001) surface was sputtered in prior with 9-min, 1-keV N2

+ ions and

then exposed to H2O molecules in the XPS working chamber under various conditions. Figure

5.4(a) depicts an AFM image (RMS ~ 0.8 nm) of the N2

+-sputtered n-GaN(0001) layer exposed to

1-mbar H2O pressure with the sample being held at RT. Figure 5.4(b) shows an AFM image (RMS

~ 0.806 nm) on an N2

+-sputtered n-GaN(0001) layer after exposure to 1-mbar H2O-vapor pressure,

with the sample was at room temperature, then at 60 oC, and finally at 160 oC, in each NAP-XPS

run. All AFM images were taken on 2.5 × 2.5 µm2 surface spots. These AFM images were taken

after collecting LEED patterns and NAP-XPS spectra of the n-GaN(0001) surface.

The AFM image of Figure 5.4(a) has better quality and exhibits terrace-like undulated

surface structure compared to the AFM image shown in Figure 5.4(b), which shows surface rich

in snail-like bulges. This may be related to modification of the n-GaN(0001) surface at high

temperatures, where surface chemical reactions, e.g. the formation of oxide compounds, and/or

removal of some of its constituents might have occurred.

(a)

(b)

Figure 5.4: AFM images measured on (2.5 × 2.5) µm2 spots on a 9 min, 1-keV N2

+-sputtered

GaN(0001) surface exposed to 1 mbar H2O pressure at (a) RT (b) RT, 60 oC and finally 160 oC.

59

AFM measurements have been made on the surface of N2

+-sputtered n-GaN(0001) layers

that were functionalized with L-cysteine molecules at various depositin and preparation

conditions. After the required UHV LEED and XPS measurements, the samples were taken

outside the XPS instrument with high care to obtain AFM images on their n-GaN(0001)/L-cysteine

interface. The acquired AFM images of the n-GaN(0001)/L-cysteine interface obtained under

different preparation conditions were rather identical with some modified features. They display

cascading terrace-like surface structure with elongated bulging protrusions portions with an RMS

of 0.729 nm on 2.5 × 2.5 µm2 surface spots.

5.2 Experimental UHV-XPS Spectra of As-received and 𝐍𝐍𝟐𝟐

+-Sputtered

n-GaN(0001) Layers

The UHV-XPS spectra of core-level (CL) transitions of the constituents of the n-GaN(0001)

surface of the as-received samples and of the N2

+-sputtered n-GaN(0001) surface are displayed,

analyzed, and discussed in this section.

5.2.1 Survey Spectra of the Photoemission Lines

A survey XPS spectrum of a surface is collected to display the number of detected photoelectrons

allied with the CL transitions of its constituents as a function of the binding energy (BE). Figures

5.5(a) and 5.5(b) depict survey UHV-XPS scans, taken with monochromatized Al Kα X-ray

excitation energy (1486.7 eV) in the normal-emission (NE) mode and 80 eV pass energy, for as-

received and 9-min, 1-keV N2

+-sputtered n-GaN(0001) layers in the range 0 –1200 eV.

Underlying the observed photoemission lines is a background signal, which increases at

high binding energies (BEs). It can be related to inelastic scattering of the emitted electrons during

their passage to the surface. The electrons from inner atomic shells with high BE experience a

higher probability of inelastic scattering than the near surface photoelectrons with lower BE. This

implies that the background contribution is generally high in the observed XPS spectra at the high

BE-side.

The detected photoemission CL lines for the surface of a bare n-GaN(0001) layer include

those originating from the Ga 3s, Ga 2p, Ga 3p, Ga 3d, and N 1s core-level transitions.

Furthermore, these UHV-XPS spectra disclose contributions from O 1s and C 1s CL transitions,

arising from native oxidation and adventitious C and O2 species, accumulated on the surface of

the n-GaN(0001) layers. The UHV-XPS survey spectra of the as-received n-GaN(0001) layers

60

exhibit higher intensity for the O 1s and C 1s photoemission peaks and a higher O/N ratio

compared to those observed for the 9-min, 1-keV N2

+-sputtered n-GaN(0001) surface. This entails

that reasonably prolonged sputtering of the n-GaN(0001) surface with a collimated beam of such

prolonged 1-keV N2

+ ions leads to partial cleaning from oxygen and its Ga-oxides, in addition to a

substantial reduction of the adventitious carbon species. The native O 1s photoemission line is

visible on the measured survey UHV-XPS spectra at a BE of 533 eV, while the main C 1s

photoemission line appears close to 285 eV.

Figure 5.5: Survey UHV-XPS spectra of the n-GaN(0001) surface taken at normal emission (NE)

with the monochromatized Al Kα1 X-ray line (1486.7 eV) as excitation energy for a) as-received

and b) 9-min, 1-keV N2

+-sputtered n-GaN(0001) layers. The photoemission lines are labeled with

their individual CL transitions.

61

The Ga 3s CL XPS photoemission line is not as informative as those arising from Ga 2p,

Ga 3p, and Ga 3d CL transitions; the XPS Ga 3p and Ga 3d photoemission lines provide more

information on bulk states deep below the surface due to lower BEs (higher photoelectron kinetic

energy). The spin-orbit doublet Ga 3p1/2, 3/2 lines form a broad peak around 107 eV that was

deconvoluted into doublet lines using CasaXPS software at the BEs of 106 eV and 109 eV. While,

the XPS spectrum of the more important and advantageous core-level Ga 3d3/2, 5/2 doublet line did

not exhibit distinguished spin-orbit coupling peaks. Most literature studies report a broad single

line peaked at a BE nearby 20 eV with an energy separation between these Ga 3d3/2, 5/2 doublet

lines around 0.5 eV [82b]. This is of the order of the energy resolution of the XPS instrumentation

used in the present work (~ 0.6 eV). As disclosed later, laborious CasaXPS analysis enables to

deconvolute the measured Ga 3d3/2, 5/2 doublet line of studied n-GaN(0001) layers into two broad

spin-orbit peaks with a 0.5 eV BE separation [85].

The Ga 2p photoemission line of the XPS spectrum of the n-GaN(0001) layers consists of

two doublet Ga 2p1/2, 3/2 lines with high BEs (low kinetic energy of photoelectrons), with the

respective peaks being well separated, due to the strong spin-orbit coupling, and located at the

BEs 1144.5 eV and 1117.5 eV, respectively [67,161,163]. Thus, the narrow, single Ga 2p3/2 line

of the Ga 2p doublet can be analyzed [85,93] to evaluate the ratio (O-Ga-OH/Ga-N ≅ O/N) due to

surface oxides existing on the uppermost surface parts.

The survey XPS spectra shown in Figures 5.5(a) and 5.5(b) display shallow peaks at the

low-BE side close to the Ga 3d photoemission line, and these are assigned to the valence band

(VB) structure of the n-GaN(0001) surface. This spectral part is highly important to be investigated

and analyzed: Combined with the Ga 3d CL energy, the band gap energy, and the Fermi energy

[162-168], the measured VB spectrum yields good information on the polarity of the GaN layer

investigated and on the electronic changes of the surface states and the induced charges

[121,122,130-132]. The low-BE VB exhibits stronger sensitivity to changes in the surface states

due to reforming of the GaN surface. Analysis of the Ga 3d/VB emissions can be used to obtain

evidence on the band bending at the n-GaN(0001) surface and on the chemical interactions and

bonding processes taking place on the surface upon functionalization with inorganic/organic

molecules.

The N 1s line is overlaid by the Ga LMM Auger lines, close to 400 eV, the BE of the

principal Ga-N bond, as seen in the XPS spectra taken with Al Kα excitation energy in the NE

mode. The CasaXPS analysis software can be used to separate overlapped Ga LMM Auger lines

62

from the N 1s Ga-N peak. The 397.8 eV BE of the N 1s and/or the 284.8 eV BE of the main C 1s

peak lines are occasionally adopted to calibrate the BE scale of measured XPS spectra to

compensate for the effect of charges accumulated on the X-ray irradiated GaN surface during the

measurements. Re-adjusting the BE scale of the XPS spectra of the n-GaN(0001) layers was

done using the measured Fermi-level edge of the molybdenum (Mo) foil affixed to the

spectrometer body to eliminate the effect of the spectrometer work function.

The low resolution of survey XPS spectra taken with 80 eV pass energy is not appropriate

for quantitative analysis of a CL line due to spectrometer-induced Gaussian broadening. The

convoluted Lorentzian (natural line broadening due to finite life time) and Gaussian (spectrometer

resolution and sample inhomogeneities) profiles combine together as a mixed GL-profile (Voigt

profile) and affect the spectral width of measured XPS lines. The signal of individual XPS

photoemission lines were thus taken using a pass energy of 20 eV and their analysis were carried

out using Voigt line-shape or Lorentzian asymmetry LF line-shape functions.

5.2.2 UHV-XPS Surface Photoemission Core-Level Lines

The photoemission lines of the O 1s, C 1s, N 1s, Ga 2p, Ga 3p, and Ga 3d CL transitions were

separated from the overall XPS signal of photoelectrons ejected from the X-ray irradiated surface

of the n-GaN(0001) layers. This was achieved via collecting the XPS spectra using a low-pass

energy of 20 eV, a small step width of data points in the range 0.1-0.5 eV, and a long-dwell time

at each data point (1-2 s). Using such parameters, the acquisition of NE XPS spectra of the

photoemission lines of interest took 6-12 hours, depending on the number of scans and on the

XPS measurements made on a particular sample. To acquire a feasible signal-to-noise ratio and

a smooth line shape, each photoemission line was accumulated about 5 – 7 times.

Most of the UHV-XPS spectra presented and discussed in this thesis are those of the

photoemission lines of CL transitions of interest for the surface of N2

+-sputtered n-GaN(0001)

layers, excited by the X-ray Al-𝐾𝐾𝛼𝛼 line (ℎ𝜋𝜋 = 1486.7𝑒𝑒𝑉𝑉). Figures 5.6 to 5.10 show typical 20-eV

pass energy XPS spectra for the surface C 1s, O 1s, Ga 2p, Ga 3p, Ga 3d, and N 1s CL

transitions. It was not possible to remove the surface oxides, but the carbon species

contaminating the surface of as-received n-GaN(0001) layers could be removed by sputtering the

surface for 9-min with 1-keV N2

+ ions, Figure 5.6.

63

Figure 5.6: The 20 eV pass energy UHV-XPS spectra of the C 1s CL transition for the n-

GaN(0001) surface of an as-received sample and for the 9-min, 1-keV N2

+-sputtered n-GaN(0001)

surface. The strong peak at the BE = 286.5 eV is related to C-C bonds and the weak peak at the

BE around 290 eV can be allocated to C=O bonds [82c].

The C 1s spectra of the samples are sometimes shifted in binding energy (BE) relative to

the standard C 1s spectra, which may result from built-in charges accumulated on the surface of

the non-conductive n-GaN(0001) layer. The measured UHV-XPS peaks of the other CL

transitions and bonding states that were shifted accordingly have been corrected by calibrating

the energy scale using the C 1s BE of adventitious carbon at 284.8 eV, which could be assigned

to C-C bonds or C-O bonding, while the weak signal seen in Figure 5.6 at the BE around 290 eV

might be linked to possible C=O bonding configuration [82c].

The UHV-XPS spectra of the O 1s CL transition for the surface of as-received and 9-min,

1-keV N2

+-sputtered n-GaN(0001) layers are broad and fluctuating at the extreme BE ends, Figure

5.7. The stable Ga2O3 oxide on the as-received n-GaN(0001) surface is seen to decrease slightly

upon N2

+ sputtering [82b], while the oxygen deficient Ga2Ox oxide (x < 3) seems to be enhanced

by bonding of oxygen to surface Ga species, a feature that is still to be comprehended. The

+

prominent effect of sputtering n-GaN(0001) surface with N2 ions was the large reduction of

adventitious carbon content as clearly seen from Figures 5.6 and 5.7.

64

Figure 5.7: The 20 eV pass energy UHV-XPS spectra of the O 1s CL transition for the n-

GaN(0001) surface of an as-received sample and for the 9-min, 1-keV N2

+-sputtered n-GaN(0001)

surface. The fitted peaks corresponding to the bonding states of existing oxides of gallium and

carbon are shown on the the measured spectra.

Due to weak spin-orbit coupling of the Ga 3p3/2,1/2 doublet, the UHV-XPS spectrum of the

surface of both as-received and 9-min, 1-keV N2

+-sputtered n-GaN(0001) layers displays a broad

and asymmetric peak. Deconvolution of this Ga 3p3/2,1/2 doublet peak yielded partially overlapped

peaks for the Ga 3p3/2 and Ga 3p1/2 CL transitions with the BEs 105.5 eV and 108.9 eV (FWHM =

2.5 eV), respectively [33,161]. Thus, no further discussion will be given thereafter on this broad,

asymmetric Ga 3p doublet peak as little reliable information could be attained from its analysis.

The Ga 2p1/2 and Ga 2p3/2 lines of the Ga 2p3/2,1/2 doublet in the XPS spectra of the surface

of n-GaN(0001) layers are narrow, symmetric, and well separated (by 27 eV). Analysis of these

Ga 2p1/2 and Ga 2p3/2 peaks yielded the BEs of 1144.6 eV and 1117.5 eV (FWHM of 1.5 eV),

respectively, assigned to Ga-N bond, besides subsidiary peaks for Ga-oxides (Ga2O3/Ga2Ox) and

Ga-Ga bonds, at the high and low energy sides, respectively, in agreement with other literature

results [50a,50b,50c,59,67,163]. Deconvolution of the Ga 2p3/2 line of the 9-min, 1-keV N2

+-

sputtered n-GaN(0001) surface yielded the main Ga-N and the subpeaks at BE ≈ 1118 eV (Ga-

oxide) and at BE ≈ 1116 eV (Ga-Ga), as illustrated in Figure 5.8.

It can be argued here, as in other studies, that hydroxyl group, oxygen and Ga-oxides play

a crucial role in stability, physical properties, chemical selectivity, and functionalization of the

interface of GaN and other III-V semiconductors [50a,50b,50c,59,67,93,163]. As described later,

the Ga 2p3/2 photoemission line, which is related to Ga 2p1/2 line, will be used to estimate the ratio

of O/N bonds on the surface of N2

+-sputtered and functionalized n-GaN(0001) layers.

65

Figure 5.8: The 20 eV pass energy UHV-XPS spectra of the Ga 2p3/2 CL transition for the n-

GaN(0001) surface sputtered with 9-min, 1-keV N2

+ ions.

Figure 5.9 depicts the broad CL Ga 3d5/2, 3/2 photoemission doublet line, whose Ga 3d3/2

and Ga 3d5/2 lines are partially overlaid with a binding energy difference of around 0.5 eV, close

to the resolution of the XPS spectrometer 0.6 eV. The BE of the measured XPS Ga 3d5/2 line has

been used in the calculation of the band bending on the n-GaN(0001) surface under various

preparation conditions, as will be detailed later. The Ga 3d5/2,3/2 doublet lines were deconvoluted

into four well-resolved components corresponding to Ga-Ox, Ga-N (Ga…H-N), N 2s, and metallic

Ga-Ga bonds [82,85].

Figure 5.9: The 20 eV pass energy UHV-XPS spectra of the Ga 3d5/2,3/2 doublet CL transition for

the n-GaN(0001) surface after sputtering with 9-min, 1-keV N2

+ ions.

66

The UHV-XPS N 1s spectra taken under similar experimental conditions are shown in

Figure 5.10. The measured photoelectron N 1s line was deconvoluted into seven prominent peaks

that were assigned to the N-X, Ga-N, and five overlapping Ga LMM lines, the BEs of all are in fair

agreement with those reported by others for GaN layers [82,85,161,169].

Figure 5.10: The 20 eV pass energy UHV-XPS spectra of the N 1s CL transition for the 9-min, 1-

keV N2

+-sputtered n-GaN(0001) surface. The fitted main N 1s peaks and the overlapped Ga LMM

sub-peaks are shown on the plot.

In principle, the area under a photoemission XPS curve of a particular element can be

reasonably calculated if this line is not superimposed by a line of another element or of another

emission process. Such superimposed photoemission lines were clearly seen for the XPS curve

of the N 1s CL transition when the monochromatized Al Kα1 (1486.7 eV) X-ray excitation energy

was used. The N 1s photoemission line (main Ga-N bond) is only partially resolved from five

overlapping Ga LMM Auger lines, as seen in Figure 5.10. This is important, if the N 1s spectrum

is needed for determining the Ga/N ratio, these overlapping N 1s photoemission and Ga LMM

Auger lines have to be accurately separated and determined. In the present study, the BE of

397.8 eV of the main N 1s peak was not adopted for BE calibration purposes, nor was it used to

determine the Ga/N ratio, because the peak overlapping causes significant uncertainty in the

determined CL binding energy and peak intensity.

67

A good piece of information has been learned from the experimental results of this part.

The surface n-GaN(0001) layers of as-received c-sapphire/n-GaN(0001) samples studied in this

work is contaminated by adventitious carbon (C) and gallium oxides. Such contaminantion was

revealed from the surface bonds of their UHV XPS spectra of C 1s, O 1s, Ga 2p3/2,1/2 doublet, and

Ga 3d5/2.3/2 doublet. The AFM images and LEED patterns of such surface support this finding. This

study was informative on the efficacy of N2

+-sputtering on the atomic structure, topography, and

chemistry of the n-GaN(0001) surface, which was not seriously damaged when low energy (1

keV) N2

+-ions were shooted on the surface within short time periods (< 13 min). Sputtering the

surface of n-GaN(0001) layers of as-received c-sapphire/n-GaN(0001) samples with 9-min, 1-keV

N2

+ ions was noted to remove partially the oxides and most of adventitious carbon species existing

on the surface. Such low-energy N2

+-sputtering was valuable for successful functionalization of

the N2

+-sputtered n-GaN(0001) surface with H2O and L-cysteine molecules (discussed next).

5.3 NAP-XPS Core-Level and Valence Band Spectra of 𝐍𝐍𝟐𝟐

+-Sputtered

n-GaN(0001) Surface Exposed to H2O Molecules

The electronic structure of the GaN surface under realistic conditions is still under debate and the

actual interactions occurring at the GaN surface functionalized with inorganic and organic

molecules are not yet fully understood [84-89,94,95]. Studying the n-GaN(0001) surface under

NAP conditions when exposed to inorganic or organic species provides evidence on the changes

occurring in its electronic structure and surface chemistry [163-168]. The knowledge of the

reactivity of H2O molecules with the n-GaN(0001) surface, for example, aids to unveil the resulting

modification in its chemical bonding and VB structure. Several literature reports alleged that H2O

molecules adsorbed on the surface of n-GaN(0001) layers decompose even at room temperature

into hydrogen (H+) and hydroxide (OH–) ions that are supposed to bind with surface N-and Ga-

sites to form Ga-OH and N-H bonds [166-168]. Probing the VB and the chemistry of the n-

GaN(0001)/H2O interface under various NAP conditions is an experimental route to get valuable

information on the Stability and functionality of the n-GaN(0001) surface to molecular species,

especially, the interaction processes and the band bending at the n-GaN(0001)/H2O interface

[88,95].

In addition to the surface polarity, the band bending (BB) at the surface of an n-GaN(0001)

layer depends on the position of the Fermi level (𝐸𝐸F) ,on the level of the dopants, and on the

surface electronic states and chemistry [44,130-132,170-173]. Surface cleanness and

adsorbates, induced by growth conditions, post-growth treatment, and functionalization, also play

68

a role in the formation of surface band bending. The BB at the n-GaN(0001) surface affects its

electrical and optical properties, and the performance and stability of GaN-based devices [174].

The naturally occurring spontaneous polarization of GaN gives rise to surface polarization of

bound charges that depend on the polarity of the GaN sample [130-132]. Internal ionized charges

and states on the GaN surface screen and compensate the polarization of bound charges and

act as trapping centers or energy levels that pin the Fermi level of the surface in the band gap,

depending on temperature, doping level, and polarity.

Technically modified XPS machines aid to get valuable quantitative information on the

chemistry and VB of a substance surface under both UHV and NAP conditions [115,116,144].

Studying the structure, composition, and bonding configurations on the surface of n-GaN(0001)

layers upon functionalization with inorganic and organic species may mimic real-world electronic

devices that integrate them. The VB NAP-XPS spectra of a functionalized n-GaN(0001) surface

help to explore the modification of its valence band structure and the NAP-XPS spectra of the CL

photoemission lines of its constituents provide their binding energies (BEs) [43,86,95]. One can

acquire from these measurements the energy separation between the Fermi level (𝐸𝐸F) and the

surface valence band edge (𝐸𝐸VS) and the peak BE of the surface Ga 3d photoemission line (Ga

3dS), relative to 𝐸𝐸FS. Henceforth, the symbols S and B are referred to surface and bulk,

respectively. Linking such energies with the Fermi energy, 𝐸𝐸FB (= 𝐸𝐸FS = 𝐸𝐸F), relative to conduc

tion band (CB) edge (𝐸𝐸CB), the band gap energy (𝐸𝐸g), and energy separation between the Ga 3dB

level (Ga 3dB) and the VB edge, 𝐸𝐸VB of bulk n-GaN(0001), one can evaluate the amount of band

bending at the surface, both in vacuum and after exposure to inorganic or organic molecular

layers [163-173].

The NAP-XPS CL spectra of the constituents of the N2

+-sputtered n-GaN(0001) surface

after exposure to H2O molecules under various pressures and sample temperatures measured

using the monochromatized X-ray Al Kα1 line (ℎ𝜋𝜋 = 1486.7 𝑒𝑒𝑉𝑉) and a 20 eV pass energy are

presented in the next sections. The measured CL photoemission lines were analyzed by a peak-

fitting program (CasaXPS) to determine their BEs, relative to 𝐸𝐸𝐹𝐹

. After a Shirley background

subtraction, a Voigt line function was used to fit the O 1s and Ga 2p CL peaks, while the Ga 3d

peak was fitted by an asymmetric Lorentzian function, LF(𝐾𝐾, 𝛽𝛽, 𝑤𝑤, 𝑚𝑚). When H2O molecules

adsorb onto the N2

+-sputtered n-GaN(0001) surface, the C 1s peak was unclear due to surface

coverage with H2O molecules; thus, it not possible to use the C 1s peak at 284.8 eV for BE

calibration.

69

The measured UHV and NAP VB XPS spectra of the N2

+-sputtered n-GaN(0001)/vacuum

and n-GaN(0001)/H2O interfaces are qualitatively discussed and quantitatively analyzed. The

energy separation between the surface VB edge (𝐸𝐸VS) and the Fermi level (𝐸𝐸FS), usually denoted

by VBM energy, was determined by the linear fit method, which is supposed to remove tails

introduced by instrumental broadening. This linear fit is achieved by extrapolating the VB leading

edge down to the baseline of the measured surface VB XPS spectra [171-176]. The surface VBM

energy was attained relative to 𝐸𝐸F, adjusted by the UHV-XPS VB spectrum of the Fermi edge of

molybdenum (Mo) metal, chosen for spectrometer calibration. The leading edge of the Mo VB

spectrum was analyzed by various methods [73,85,177]. In the XPS experiments of this work, a

kinetic energy of 1486.7 ± 0.4 𝑒𝑒𝑉𝑉 was found to center the XPS VB leading edge of the Mo sample.

Some NAP-XPS runs were completed successively on the same c-sapphire/N2

+-sputtered

n-GaN(0001) sample under all H2O-vapor pressures studied and at each of the adopted

temperatures without taking the sample outside the XPS working chamber. The H2O-vapor

pressures used were 0.02 mbar, 0.05 mbar, 0.08, 0.1 mbar, 0.5 mbar, and 1 mbar, while the

sample temperatures were 23 oC (RT), 40 oC, 60 oC, 80 oC, 90 oC, 130 oC, and 160 oC. Using the

same sample for all NAP-XPS runs at all temperatures and pressures raised questions about the

viability of the results. Therefore, a series of individual NAP-XPS runs were carried out on

different, fresh c-sapphire/N2

+-sputtered n-GaN(0001) samples. All samples were studied at all

above-mentioned H2O-vapor pressures, but a different temperature was chosen for each new,

fresh sample. Most of the results displayed and discussed below were those obtained on the latter

experimental sample preparations.

5.3.1 NAP-XPS Surface Photoemission Core-Level Lines

Selective NAP-XPS spectra of the O 1s, Ga 3d, and Ga 2p photoemission lines of the 9-min, 1-

keV N2

+-sputtered n-GaN(0001) surface after exposure to H2O molecules that were collected at

some of the studied H2O vapor pressures (𝑝𝑝) and sample temperatures (𝑇𝑇) are presented and

analyzed here. Only the final results of the analysis of the NAP-XPS spectra of these CL

photoemission lines that were taken at other pressures and temperatures are given. Also, some

of the results obtained on the same 9-min, 1-keV N2

+-sputtered n-GaN(0001) sample studied at

all adopted temperatures under the above-cited H2O vapor pressures are presented for

comparison.

70

5.3.1.1 O 1s Photoemission Line

+-

sputtered n-GaN(0001)/H2O interface measured at 1-mbar H2O vapor pressure at 23 oC (RT) for

a fresh sample, at 60 oC for another fresh sample, and at 160 oC for a third different fresh sample.

Each spectrum was deconvoluted into four well-resolved peaks associated with the bonding

contributions from the Ga2O3, Ga2Ox, Ga-OH, and H2O species. The fitted peak at 531.8 eV is

attributed to Ga2O3 bonds, due to oxidation of surface Ga atoms, that become more prominent at

high temperatures [174]. The peak at 530.6 eV is ascribed to Ga2Ox arising from a less covalent

form of oxygen, while the peak at 532.7 eV may be assigned to Ga-OH bonds, formed upon the

dissociation of H2O molecules into OH– and H+ ions at the GaN surface [163-168]. The number

of Ga-OH bonds are reduced at high temperatures. The peak at BE = 533.4 eV can be allotted to

physisorbed H2O molecules that became less significant at elevated temperatures due to re-

evaporation [178].

Figures 5.11(a-c) Illustrate the NAP-XPS spectra of the broad O 1s photoemission line of the N2

(a)

(b)

(c)

Figure 5.11: Measured NAP-XPS spectra for the O 1s CL transition of the N2

+-sputtered n-

GaN(0001)/H2O interface of different samples exposed to 1-mbar H2O molecules for a fresh

sample at (a) 23 oC, for another new, fresh sample at (b) 60 oC, and for a third fresh sample at

(c) 160 oC.

Figures 5.12 (a-c) portray the bonding contributions to the NAP-XPS O 1s photoemission

line of the N2

+-sputtered n-GaN(0001)/H2O interface at all H2O pressures for distinct, fresh

samples, each at a different 𝑇𝑇. No H2O molecules exist on bare N2

+-sputtered n-GaN(0001)

surface, but after H2O-vapor injection, some H2O molecules remain on the surface, as not all H2O

molecules adsorbed on the n-GaN(0001) surface dissociate [95]. At high 𝑇𝑇, some non-dissociative

71

adsorbed H2O molecules re-evaporate and some remain intact on the surface even when

pumping down the chamber back to UHV after the NAP-XPS run. It was not possible to determine

quantitatively how much H2O molecules re-evaporate or remain intact on the surface. Sputtering

+

n-GaN(0001) surface with 1-keV N2 ions was not effective in removing surface Ga2O3 and

oxygen-deficient Ga2Ox layers, formed on n-GaN(0001) surface exposed to air thru transportation

and storage or to traces of oxygen remaining in XPS chamber [178,179]. The plots show that

Ga2Ox species reduced more upon H2O injection at all 𝑝𝑝 and 𝑇𝑇, a feature that may be related to

the replacement of O in Ga2Ox bonds by OH− ions to form additional Ga-OH bonds.

(a)

(b)

(c)

Figure 5.12: Bond contributions to the O 1s line of the N2

+-sputtered n-GaN(0001)/H2O interface

with varying H2O pressure at (a) 23 oC, (b) 60 oC, and (c) 160 oC, for different samples as

explained in Figure 5.11. The abbreviations UHV2 and UHV3 correspond to data taken at 23 °C

in vacuum for the N2

+-sputtered n-GaN(0001) surface before exposure to H2O molecules and after

the end of the NAP-XPS run on the sample, respectively.

72

For comparison, Figures 5.13(a-c) show an analogous trend, at least at high H2O

pressures, of the bonding contributions to the O 1s photoemission line of the N2

+-sputtered n-

GaN(0001)/H2O interface for the same sample. The sample temperature was first held at 23 oC

while the UHV-XPS run was carried out. Then, the same sample was exposed to H2O molecules

at pressure 𝑝𝑝 = 0.02 mbar and the NAP-XPS spectra of the resulting N2

+-sputtered n-

GaN(0001)/H2O interface were taken at same temperature.

(a)

(b)

(c)

Figure 5.13: Bond contributions to the O 1s photoemission line of the N2

+-sputtered n-

GaN(0001)/H2O interface of the same sample held at (a) 23 oC, (b) 60 oC, and (c) 160 oC in

succession, while the NAP-XPS measurements were taken at different H2O pressures.

73

The results shown in the plots of Figure 5.13 were obtained with the same sample being

still held at 𝑇𝑇 = 23 oC and more, H2O molecules were injected into the XPS chamber to reach in

succession pressures of 0.05 mbar and 0.08 mbar, at each of which the surface NAP-XPS spectra

were collected. This procedure of NAP-XPS measurements was conducted on the surface of the

same sample that was heated to and maintained at 60 oC and 160 oC in sequence, at each of

which the NAP-XPS runs were repeated at the H2O-vapor pressures listed above. In this type of

NAP-XPS measurements, the sample was not removed from the XPS chamber; instead, it was

kept under vacuum after finishing each run until the required measurements on its N2

+-sputtered

n-GaN(0001)/H2O interface were completed.

It can be inferred from the varying trend of bonding configurations shown on the plot of

Figure 5.13 (a) that there were trivial traces of hydroxyl ions (OH–) and H2O molecules on the

surface of N2

+-sputtered n-GaN(0001) layer kept at 23 oC before being exposed to H2O molecules,

but there was some chemically inert (stoichiometric) Ga2O3 oxide and was rich in native oxygen-

deficient oxide (Ga2Ox) with open bonds. Upon injecting H2O molecules inside the chamber, some

of the H2O molecules adsorbed on the surface remain intact and the other appears to dissociate

into H+ and OH– species, which bind to free surface Ga ions to form Ga-OH bonds.

Interestly, the Ga2Ox species reduced drastically and the amount of the chemically inert

Ga2O3 oxide increased. It has been hypothetically argued that at 200 °C a reduction in the number

of OH− ions might occur due to dissociation of these hydroxide ions into chemisorbed oxygen

and, presumably, hydrogen [95]. Oxidation of the GaN surface seems to take place in the

presence of H2O vapor molecules in such a way that stronger Ga-O bonds are formed as Ga2O3

oxide, which become prominent at high temperatures, due to an excess of oxygen resulting from

possible breaking of OH− ions [160]. More probably, the results seem to indicate that upon H2O

exposure and annealing the Ga2Ox is fully removed and replaced by a Ga2O3 layer, which may

be terminated by Ga-OH at the surface. Below, it will be further elucidated that the exposure of

the N2

+-sputtered n-GaN(0001) surface to H2O molecules likely removes surface electronic states

(dangling bonds) and charged traps that were initially present on the sample surface, thus

reducing the apparent upward surface band bending [85,95]. This could be understood that the

oxygen open bonds interact with free surface Ga ions and converted to extra Ga2O3 thin layer.

Raising the temperature of the same sample, the initially formed bonding configurations that

passivate the surface remain almost unchaged, except upon annealing the sample at high

temperatures (around 160 oC) a notable increase in the Ga2O3 layer took place. Such

interpretation is consistent with the findings of the work made by Yamada et al. on GaN(0001)

74

layers [50b,50c]. As will be seen later, this can be presumed to be a possible cause for the

prominent decrease of band bending at N2

+-sputtered n-GaN(0001)/H2O interface upon sample

heating/annealing at high temperatures.

5.3.1.2 Ga 3d Photoemission Line

Figures 5.14(a-d) depict an UHV-XPS spectrum for the Ga 3d photoemission line of an n-

GaN(0001) surface exposed to 9-min sputtering by 1-keV N2

+-ions and the NAP-XPS spectra at

1-mbar H2O vapor pressure for the N2

+-sputtered n-GaN(0001)/H2O interface of three different

fresh samples. One fresh N2

+-sputtered c-sapphire/n-GaN(0001) sample was studied at 23 oC,

another sample at 60 oC, and a third one at 160 oC. The temperature of each fresh sample was

kept constant during the measurement of the NAP-XPS spectra of its surface at all H2O vapor

pressures studied.

XPS Ga 3d3/2, 5/2 spectrum has been reported to be deconvoluted into two main peaks

corresponding to Ga-O and Ga-N bonds, in addition to a broad peak positioned at approximately

17 eV, related to the photoemission of electrons from the N 2s state [82,85,160]. In practice, the

XPS Ga 3d line of the n-GaN(0001) surface is an overlapped Ga 3d3/2, 5/2 spin-orbit doublet with

the binding energies (BEs) for the (Ga-N)3/2 and (Ga-N)5/2 peaks differing by an amount of the

order of the instrumental broadening (0.4 − 0.6 𝑒𝑒𝑉𝑉) of the spectrometer. In the present work, each

NAP-XPS spectrum of the Ga 3d3/2, 5/2 doublet shown in Figures 5.14(a-d) has been de-convoluted

into several peaks, depending on the dominant surface bonding processes.

75

Figure 5.14: (a) UHV-XPS spectrum for the Ga 3d CL of the N2

+-sputtered n-GaN(0001) surface,

and the NAP-XPS spectra after injection of 1 mbar of H2O molecules at varying sample

temperatures of: (b) 23 oC, (c) 60 oC, and (d) 160 oC. Three different freshly N2

+sputtered c-

sapphire/n-GaN(0001) samples were used in different runs one at 23 oC, the second at 60 oC,

and the third at 160 oC. The insets in a) is a magnification of the Ga-metal and N 2s bond

contributions, while the inset in (b) is illustrative of the Ga-OH bonds at RT. It is also noted that

intial Ga-oxides exist on the surface even before functionalization of the surface with H2O

molecules, indicating that sputtering the n-GaN(0001) surface with N2

+ ions is not effective in this

respect.

Analysis of the measured Ga 3d3/2,5/2 doublet of the N2

+-sputtered n-GaN(0001) layers

yielded two peaks for the Ga-N bond at ca. 20 eV with a spin-orbit splitting of 0.4 eV. The relative

intensity (branching ratio) between the 3d3/2 and 3d5/2 states was found to be 2:3 (0.667),

76

consistent with the theoretical ratio of the degeneracies of the doublet components of the 3d

states. The Ga 3d3/2, 5/2 doublet component at the high BE side of the measured peak is associated

with a peak for the Ga-O bond.

The shape and features of the Ga 3d3/2, 5/2 doublet of the n-GaN(0001) surface after 9-min

1-keV N2

+-sputtering is indicative that the etching process was not very effective in removing

oxide-related species. The sputtering of the surface of the n-GaN(0001) layer with low-energetic

N2

+-ions does not cause extensive surface damage, but may yield a nitrogen-deficient n-

GaN(0001) surface, as breakage of Ga-N bonds would lead to additional Ga-O and Ga-Ga bonds

[82,85]. Figures 5.14(b-d) show that the XPS spectrum of the Ga 3d line of the N2

+-sputtered n-

GaN(0001) surface is modified upon exposure to H2O molecules such that new peaks appear

[85,168]. Deconvolution of the NAP-XPS spectra of the Ga 3d line of the N2

+-sputtered n-

GaN(0001)/H2O interface yields components that can be assigned to Ga-O, Ga-N, Ga-OH, and

metallic Ga-Ga bonds, in addition to the broad peak of the N 2s CL transition. It is difficult to obtain

solid evidence from these NAP-XPS spectra for the existence of surface Ga-H2O bonds as argued

elsewhere [178] or for the observation of an O 2s peak, as seen in the asymmetrically broadened

Ga 3d line at the high BE side (at ca. 23 eV) in the literature [95].

Few reports argued that H2O molecules adsorbed on GaN(0001) surface dissociate even

at room temperature to OH─ and H+ ions that bind to Ga and N dangling bonds to form Ga-OH

and H-N bonds, respectively [85,89,93,168]. The O-Ga-OH bonds form due to partial oxidation of

Ga atoms and no Ga-H or N-OH bonds, which are unlikely below 400 oC [85,168]. Such decrease

in dangling bonds results in a removal of surface states, thus reducing the apparent BB at the

surface [95,162]. In the context of above arguments, the Ga-OH peak grows as H2O pressure

increases and declines near 160 °C, as OH─ ions might break into oxygen and hydrogen [95,179],

which cannot be comprehended on physical grounds.

However, the conversion of the unstable surface Ga2Ox bonds to enhance formation of

thin layer of the chemically-inert Ga2O3 oxide upon water exposure, followed by thermal annealing

is far more probable. This reduction of Ga2Ox oxide and the enhancement of Ga2O3 oxide species,

besides the original surface Ga2O3 oxide overlayer, passivate the surface and thus are more

likekly to be responsible for the reduction of band bending at N2

+-sputtered n-GaN(0001)/H2O

interface. Compensation of (filling) the empty surface states at the bare N2

+-sputtered n-

GaN(0001) surface upon H2O exposure can be considered as another cause for the reduction in

the surface band bending.

77

The NAP-XPS Ga 3d spectra of the N2

+-sputtered n-GaN(0001)/H2O interface show that

the BE of the Ga-OH peak is about 1 eV greater than that of the Ga-N bond [82,168]. No peaks

can be assigned to Ga-H bonding in the Ga 3d photoemission line as its formation is unlikely and

would appear at ca. 1.7 eV below that of the Ga-N bond. Only minor changes are observed in the

shapes of the Ga 3d doublet and N 1s lines of the N2

+-sputtered n-GaN(0001)/H2O interface, as

the peaks resulting from the formation of Ga-OH or Ga-O bonds at the surface overlap with the

peak originating from the Ga-N bond of bulk GaN. Hydroxylation and oxidation of the n-GaN(0001)

surface may be significant at high temperatures [85]. The Ga-OH (O-Ga-OH) bonds remain intact

on the N2

+-sputtered n-GaN(0001) surface at 𝑇𝑇 < 100 oC but tend to decrease with pressure at

higher temperatures due to evaporation of H2O molecules.

5.3.1.3 Ga 2p and Ga 3p Photoemission Lines

The UHV-XPS photoemission peaks of the Ga 3p3/2 and Ga 3p1/2 doublet, due to p-state spin-

orbit splitting, of the N2

+-sputtered n-GaN(0001) surface and the N2

+-sputtered n-GaN(0001)/H2O

interface are overlapped, broad and asymmetric. The results of XPS experiments of this work

indicate that spin-orbit splitting energy is 27 eV for the Ga 2p1/2,3/2 line and about 3 eV for the Ga

3p1/2,3/2 line. Further, the NAP-XPS spectra of the overlapped Ga 3p3/2,1/2 photoemission lines of

the N2

+-sputtered n-GaN(0001)/H2O interface reveal that the effect of H2O molecules on their

features and shapes is minor. Hence, the amount of Ga-OH or Ga-O-OH bonds formed at the

surface cannot be determined accurately, and thus, in this work the Ga 3p3/2,1/2 doublet of the N2

+-

sputtered n-GaN(0001)/H2O interface is of little use for estimating the surface O/N-ratio under

various H2O functionalization conditions from the ensuing Ga-OH or Ga-O-OH bonds. The Ga

2p3/2,1/2 doublet photoemission lines are more useful for this purpose as these doublet components

are well-separated and narrow, with little asymmetry in their shoulders at their low and high BE

sides. Figure 5.15 shows the NAP-XPS spectrum of the Ga 2p3/2 component of the N2

+-sputtered

n-GaN(0001)/H2O interface for different fresh samples, one held at RT (23 oC), another at 60 oC,

and the third at 160 oC, all studied at the H2O vapor pressure of 0.02 mbar.

78

(a)

(b)

(c)

Figure 5.15: The XPS spectrum of the Ga 2p3/2 peak of the N2

+-sputtered n-GaN(0001) surface

in UHV and after exposure of three different freshly N2

+sputtered c-sapphire/n-GaN(0001)

samples to H2O molecules at a pressure of 0,02 mbar, one maintained at (a) 23 oC, the second

one at (b) 60 oC, and the last one at (c) 160 oC.

The NAP-XPS spectra of the Ga 2p3/2 photoemission line of the N2

+-sputtered n-

GaN(0001)/H2O interface collected at 0.02 mbar for the three different fresh samples held at the

different temperatures Figures 5.15(a-c) can be deconvoluted to the Ga-Ga, Ga-N, and Ga-O

(Ga-O-OH) peaks at the BEs of 1116.5 eV, 1118.2 eV, and 1118.9 eV, respectively, in agreement

with other findings [171]. Sputtering the n-GaN(0001) surface with 9-min, 1-keV N2

+ ions slightly

reduces the Ga-O peak due to partial removal of native oxides and the Ga-Ga peak, possibly due

to oxidation of free (unbound) Ga atoms upon exposure to atmosphere and H2O vapor. Analysis

of the NAP-XPS Ga 2p3/2 photoemission line reveals plausible Ga-N and Ga-O-OH bonding

contributions in good agreement with other studies on n-GaN(0001) wafers. Estimation of the

surface concentration of hydroxide ions (OH─) is not straightforward and challenging to achieve.

The ratio of O-Ga-OH to Ga-N bonds deduced from the analysis of the NAP-XPS Ga 2p3/2 line in

this work was close to the ratio of oxygen (hydroxide ions) to nitrogen, i.e. the O/N ratio, reported

by Hengshan Wang et al. [93]. This quantity is a measure for the concentration of surface OH─

species that increase with increasing H2O vapor pressure (𝑝𝑝) or, equivalently, with increasing

adsorption of H2O molecules onto the n-GaN(0001) surface at low sample temperatures (< 60

oC), as illustrated in Figures 5.16 (a-b).

79

(a)

(b)

Figure 5.16: The O/N-ratio estimated from the quantitative analysis of the Ga 2p3/2 photoemission

line of the N2

+-sputtered n-GaN(0001)/H2O interface of different fresh samples as a function of (a)

the H2O vapor pressure (𝑝𝑝) at various sample temperatures and (b) the sample temperature (𝑇𝑇)

at various H2O vapor pressures.

(a) (b)

Figure 5.17: The O/N-ratio versus H2O vapor pressure (𝑝𝑝), estimated from a quantitative analysis

of the (a) Ga 2p3/2 and (b) Ga 3d5/2 photoemission lines of the N2

+-sputtered n-GaN(0001)/H2O

interface at some temperatures.

These plots show that as the sample temperature is increased, the desorption of OH─ ions

from the n-GaN(0001) surface is probably enhanced, and OH─ ions undergo breakage into

80

oxygen and hydrogen at high temperatures near 160 oC, thus reducing the O/N ratio [93,95]. A

similar trend is also observed from the quantitative analysis of the NAP-XPS Ga 3d5/2

photoemission line of the same samples, as shown in Figures 5.17(a) and (b) for various sample

temperatures and H2O vapor pressures. The observed difference in the numerical values of the

O/N-ratio estimated from the analysis of the NAP-XPS Ga 3d5/2 photoemission line from those

estimated using the NAP-XPS Ga 2p3/2 line can be related to the difference in the measured

intensities of these photoemission lines and the greater surface sensitivity of the Ga 2p3/2 line

compared to the Ga 3d5/2 line.

5.3.2 UHV and NAP-XPS Valence Band Emission Spectra

Probing the valence band (VB) photoelectrons ejected from the surface of a GaN layer enables

access to information on its surface band bending (BB) under different experimental conditions

[43]. The literature results are not all consistent and diverse approaches are proposed to explain

the BB at the GaN surface [163-168]. To obtain further information on the features of the VB

structure and the band bending at the GaN surface, the UHV VB XPS spectra of the N2

+-sputtered

n-GaN(0001) surface and the NAP-XPS spectra of the N2

+-sputtered n-GaN(0001)/H2O interface

at different pressures (𝑝𝑝) and sample temperatures (𝑇𝑇) were measured. All measurements were

taken using a collimated beam of monochromatized Al Kα X-ray radiation.

The UHV-XPS VB spectrum of the N2

+-sputtered n-GaN(0001) surface unveil a VB width

of 9 eV and two peaks (PI and PII) at 4.8 eV and 9.2 eV with an intensity ratio PII/PI of 0.86 Figure

5.18(a). consistent with other findings [73,85]. Valence electrons of Ga and N link to the Ga-

4s4p3d and N-2s2p orbitals, so the PI and PII peaks are related to hybridized Ga 4s-N 2p and Ga

4p-N 2p states of s-and p-like orbitals [44,121,166,171]. The intensities of the PI and PII peaks

reveal the GaN polarity, where the PI-peak is larger than PII-peak for Ga-polar GaN, and the PII-

peak is larger than the PI-peak for N-polar GaN. The observed PII/PI intensity ratio of 0.86

indicates that the n-GaN(0001) surface studied here has Ga-face polarity, which remains after

N2

+-sputtering; metallic Ga atoms are released by the breakage of Ga-N bonds by ion shelling

[178] form oxide-related species at the surface. The shoulder seen on the low-energy side near

𝐸𝐸F lying below CB edge could be partially due to surface states (empty Ga dangling bonds), or to

impurities and adventitious speices on the surface, which are partially removed upon N2

+-

sputtering, hence less shoulder curvature. More lessening of the shoulder curvature is seen upon

exposure to H2O molecules, whose decomposed components largely compensate (fill) the empty

surface states.

81

(a)

(b)

Figure 5.18: Data (open symbols) of (a) the UHV-XPS VB spectrum of the N2

+-sputtered n-

GaN(0001) surface and (b) the NAP-XPS VB spectrum of the N2

+-sputtered n-GaN(0001)/H2O

interface. Solid lines (red) are linear fits of the VB leading edge, extrapolated to meet the baseline

(green) at the VBM, below the Fermi level 𝐸𝐸F.

The NAP-XPS spectrum of the VB of the N2

+-sputtered n-GaN(0001)/H2O interface at 𝑝𝑝 = 0.08

mbar and 𝑇𝑇 = 23 oC is shown in Figure 5.18(b). The shape of the VB structure of the 𝑁𝑁2

+-sputtered

n-GaN(0001) surface was not much modified upon exposure to H2O molecules at 𝑇𝑇 < 100℃ and

𝑝𝑝 < 0.1 𝑚𝑚𝑚𝑚𝑎𝑎𝑚𝑚, Figure 5.18(b). The shoulder seen around 2 eV in VB of the N2

+-sputtered n-

GaN(0001) relating to surface states and dangling bonds was reduced upon exposure to H2O

molecules, partially due to their termination by dissociative species of H2O molecules and more

probably due to surface passivation by Ga2O3 thin layer resulting from conversion of unstable

Ga2Ox molecules. At higher 𝑝𝑝, the NAP-XPS VB spectra become distorted and the 𝑃𝑃𝐼𝐼 and 𝑃𝑃𝐼𝐼𝐼𝐼 peak

intensities tend to decrease [85]. This can be linked to a reduction in the Ga and N suborbital p-

and s-states due to the bonding of OH– and H+ ions with surface Ga and N sites. At high 𝑝𝑝, non-

dissociated physisorbed H2O molecules rest on top of the first layer of chemisorbed dissociated

H2O molecules on the N2

+-sputtered n-GaN(0001) surface. Distortion of the NAP-XPS VB spectra

and diminishing of the 𝑃𝑃𝐼𝐼 and 𝑃𝑃𝐼𝐼𝐼𝐼 peaks became noticeable for the N2

+-sputtered n-GaN(0001)/H2O

interface at 𝑇𝑇 > 100℃, at which the amount of OH– ions on the surface is reduced due to re-

evaporation and/or decomposition to oxygen, and thus, formation of a Ga2O3 surface layers [85].

82

5.3.2.1 Evaluation of the Band Bending at the n-GaN(0001) Surface

The amount of band bending near the surface of an n-GaN(0001) layer can be evaluated by

various methodologies; each has its own drawbacks [130-132,171-174]. One of these approaches

is based on the measurement of the VBM energy, the difference between the valence band edge

𝐸𝐸VS and Fermi level 𝐸𝐸FS at the surface (𝐸𝐸FS − 𝐸𝐸VS), where the symbol S refers to surface. The

determination of (𝐸𝐸FS − 𝐸𝐸VS) is often hampered by the overlapping surface states [165-167], which

inhibit accurate values for the VBM energy, which is frequently determined by the linear

extrapolation of the valence band (VB) leading edge to the baseline of the experimental XPS VB

spectrum [43,44,85,176].

Theoretically, the VBM is identified as the energy at which the VB density of states (DOS)

goes to zero. The linear extrapolation method is supposed to reduce instrumental spectral

broadening. However, it still yields a result that differs from the real VBM value due to the

presence of surface states and due to the possible formation of surface photovoltage (SPV)

[85,95,165-167]. Besides, it is a less accurate method compared to the more detailed theoretical

calculations [175,176]. In addition, other hitches render the estimation of the VBM energy from

the experimental VB XPS spectra difficult to achieve accurately.

Another method depends on monitoring the peak position of the Ga 3dS photoemission

line in the XPS spectrum of the surface under study to measure the energy difference between

the valence band edge 𝐸𝐸VS and the Fermi level 𝐸𝐸FS at the surface that is, (𝐸𝐸FS − 𝐸𝐸VS) [130-

132,165-167]. This approach relies on the precise knowledge of the energy separation between

the Ga 3dB CL and the valence-band edge (𝐸𝐸VB) of bulk GaN that is, (𝐸𝐸VB − EGa3dB), where the

symbol B refers to bulk [164-166]. This bulk energy difference is equivalent to (𝐸𝐸VS − EGa3dS), the

separation between the edge, 𝐸𝐸VS of valence band (VB) and the Ga 3dS level at the surface of

the n-GaN(0001) layer.

For the bulk wurtzite n-GaN(0001), the energy difference (𝐸𝐸VB − EGa3dB) is a constant,

which has been reported recently to be in the range of 17.25 to 17.76 eV, with a proposed best

estimate of 17.5 eV, based on an accurate determination from the bulk-sensitive soft X-ray

emission (SXE) method with a nearly 50-nm information depth [164,165]. However, a variety of

other values in the range from 17.1 to 18.5 eV were reported in literature for the energy difference

(𝐸𝐸VB − EGa3dB) on n-GaN(0001) layers, and the value of 17.8 eV was adopted by some

researchers [130-132]. These discrepancies were connected to differences in the probing depth

of the applied methods [85,95,163-168]. By adopting (𝐸𝐸VB − EGa3dB) values less or larger than

83

17.5 eV, the estimation of the band bending at the surface of the n-GaN(0001) layers of c-

sapphire/n-GaN(0001) samples yields somewhat improper results [164-166].

The strategies discussed previously for analyzing the band bending at a semiconductor

surface can be exemplified in terms of the band bending voltage 𝑉𝑉bb, using the formulas cited

below in Equations (5.1a) and (5.1b). A schematic diagram for this evaluation is shown in Figure

5.19.

Vbb = 𝐸𝐸g(𝑇𝑇) − (𝐸𝐸CB − 𝐸𝐸FB) − (𝐸𝐸FS − 𝐸𝐸VS) (5.1a)

Vbb = 𝐸𝐸g(𝑇𝑇) − (𝐸𝐸CB − 𝐸𝐸FB)+ (𝐸𝐸VB − EGa3dB) − (𝐸𝐸FS − EGa3dS) (5.1b)

where 𝐸𝐸g(𝑇𝑇)=(𝐸𝐸CB − 𝐸𝐸VB) is the band gap energy of the bulk semiconductor at temperature 𝑇𝑇,

while (𝐸𝐸CB − 𝐸𝐸FB) is the energy separation from the Fermi level, 𝐸𝐸FB to the minimum conduction

band edge, 𝐸𝐸CB, of the bulk semiconductor, and (𝐸𝐸FS − 𝐸𝐸VS) is the energy separation from 𝐸𝐸FS to

the VB-edge, 𝐸𝐸VS at the surface, which can be determined from the measured XPS photoemission

spectra of the surface valence band.

The quantity (𝐸𝐸FS − EGa3dS) is the energy difference between the Ga 3dS CL and 𝐸𝐸FS at

the surface, which is the peak BE of the measured surface Ga 3dS photoemission line. It is worth

noting that a positive value of 𝑉𝑉bb indicates an upward band bending at the semiconductor’s

surface and that the 𝑉𝑉bb formulas of Equations (5.1a) and (5.1b) are valid when the potential

change in the range of the escape energy of the photoelectrons is sufficiently small. Application

of the approaches typified in Equations (5.1a) and (5.1b) are reliant on the determination of

𝐸𝐸g(𝑇𝑇)=(𝐸𝐸CB − 𝐸𝐸VB) and the position of 𝐸𝐸FB relative to 𝐸𝐸CB that is, (𝐸𝐸CB − 𝐸𝐸FB) of the bulk

semiconductor, both depend on temperature.

For an n-type semiconductor with a bulk electron concentration 𝑛𝑛, 𝐸𝐸F(𝑇𝑇) at temperature

𝑇𝑇 with respect to the conduction band (CB) edge, 𝐸𝐸𝐶𝐶𝐹𝐹 of the bulk semiconductor that is,

(𝐸𝐸CB − 𝐸𝐸FB) can be determined on the basis of the Boltzmann distribution of electrons in the

conduction band of the bulk semiconductor using the formula given below [165]:

𝑛𝑛

(𝐸𝐸CB − 𝐸𝐸FB)= − kB𝑇𝑇 ln � � (5.2a)

𝑁𝑁c(𝑇𝑇)

where kB is the Boltzmann constant and 𝑁𝑁c(𝑇𝑇) is the effective density of states in the conduction

band of the bulk semiconductor at temperature 𝑇𝑇 that is given by following formula:

3

2π𝑚𝑚𝑒𝑒

∗

kB𝑇𝑇 2

𝑁𝑁c(𝑇𝑇) = 2 � � (5.2b)

ℎ2

84

∗

where ℎ is the Planck’s constant and 𝑚𝑚𝑒𝑒 is the effective mass of an electron in the conduction

∗

band. For n-type GaN(0001), 𝑚𝑚𝑒𝑒 = 0.2 𝑚𝑚e [165], with 𝑚𝑚e the free-electron mass. Equation (5.2a)

shows that the 𝐸𝐸FB position relative to 𝐸𝐸CB edge depends on 𝑇𝑇 and on the semiconductor dopant

electron concentration 𝑛𝑛. The ratio [𝑛𝑛⁄𝑁𝑁c(𝑇𝑇)] expresses whether the Fermi level 𝐸𝐸FB(𝑇𝑇) is above

(𝑛𝑛 > 𝑁𝑁c(𝑇𝑇)) or below (𝑛𝑛 < 𝑁𝑁c(𝑇𝑇)) the 𝐸𝐸CB edge [164-167].

Figure 5.19: A schematic diagram for the evaluation of the band bending (BB) in the vicinity of

the n-GaN(0001) surface on the basis of Equations (5.1a) and (5.1b) cited in the text. The symbols

𝐸𝐸FB, 𝐸𝐸g, and 𝐸𝐸CB denote the bulk Fermi level, band gap energy, and CB edge, respectively. 𝐸𝐸VB

and 𝐸𝐸VS are the bulk and surface VB edges, respectively, while EGa3dB and EGa3dS are the positions

of the bulk Ga 3dB and surface Ga 3dS core levels, respectively. The energy separation between

the 𝐸𝐸CB edge and 𝐸𝐸FB of bulk n-GaN(0001) is (𝐸𝐸CB − 𝐸𝐸FB), given in Equation (5.2a) in the text.

(𝐸𝐸VB − EGa3dB) is the energy difference between the 𝐸𝐸VB edge and the Ga 3dB CL in bulk n-

GaN(0001), which is taken as 17.5 eV, while (𝐸𝐸FS − EGa3dS) is the energy separation between 𝐸𝐸FS

and 𝐸𝐸Ga3dS at the sample surface. The dependence of 𝐸𝐸g on 𝑇𝑇 can be found from Equation (5.3).

To determine the proper amount of band bending at a semiconductor surface in terms of

𝑉𝑉bb at sample temperature 𝑇𝑇, the values of 𝐸𝐸g(𝑇𝑇) and 𝐸𝐸FB(𝑇𝑇) at 𝑇𝑇 have to be inserted into

Equations (5.1a) and (5.1b). The latter can be achieved by the use of Equation (5.2), while the

evaluation of 𝐸𝐸g(𝑇𝑇) and its temperature dependence is physically quite intricate. A variety of

theoretical methods and models were proposed to describe 𝐸𝐸g(𝑇𝑇) in semiconductors [124,180],

where the most popular model is based on the widely-used Varshni empirical equation of the form

𝛾𝛾 𝑇𝑇2

𝐸𝐸g(𝑇𝑇)= 𝐸𝐸g(0) − (5.3)

(𝑇𝑇 + 𝛽𝛽)

85

where 𝐸𝐸g(0) is the band gap energy at absolute zero (𝑇𝑇 → 0 𝐾𝐾), while 𝛾𝛾 and 𝛽𝛽 are the Varshni

fitting parameters [180]. For n-GaN(0001), the reported values of 𝐸𝐸g(0) and the Varshni

parameters are diverse, scattered and controversial. This can be related to the dissimilar

techniques used to measure 𝐸𝐸g(𝑇𝑇) at temperature 𝑇𝑇, to the type of GaN-layer substrate, and to

the 𝑇𝑇-range chosen to fit the measured 𝐸𝐸g(𝑇𝑇) data to the adopted 𝐸𝐸g(𝑇𝑇) − 𝑇𝑇 formula. A set of

plausible values of 𝐸𝐸g(𝑇𝑇 → 0 𝐾𝐾) and the Varshni parameters for n-GaN(0001) films on c-sapphire

substrates in the range of 10 𝐾𝐾 < 𝑇𝑇 < 700 𝐾𝐾 are 𝐸𝐸𝑔𝑔(𝑇𝑇 → 0𝐾𝐾) ≅ 3.5 𝑒𝑒𝑉𝑉, 𝛾𝛾 = 0.84 𝑚𝑚𝑒𝑒𝑉𝑉⁄𝐾𝐾, and 𝛽𝛽 =

790 𝐾𝐾 [124].

The measured NAP-XPS spectra of the Ga 3d CL photoemission line and the VB of the

surface of the n-GaN(0001) layers of the c-sapphire/n-GaN(0001) samples sputtered by 9-min,

1-keV N2

+ ions and exposed to H2O molecules at various H2O vapor pressures (𝑝𝑝) and sample

temperatures (𝑇𝑇) are used to analyze the surface band bending. At each 𝑇𝑇 and 𝑝𝑝, the amount of

band bending at the n-GaN(0001)/H2O interface was found in terms of 𝑉𝑉bb using Equations (5.1a)

and (5.1b), with the position of 𝐸𝐸FB(𝑇𝑇) relative to 𝐸𝐸CB of bulk n-GaN(0001) that is, (𝐸𝐸CB − 𝐸𝐸FB) is

calculated from Equation (5.2a). The bulk band gap energy 𝐸𝐸g(𝑇𝑇)=(𝐸𝐸CB − 𝐸𝐸VB) was found from

the Varshni expression in Equation (5.3) using the above-cited values of 𝐸𝐸g(𝑇𝑇 → 0𝐾𝐾) and the

Varshni parameters 𝛾𝛾 and 𝛽𝛽. The results were correlated with the chemical composition of the n-

GaN(0001)/H2O interface.

5.4 Pressure and Temperature Dependencies of the Band Bending

at the 𝐍𝐍𝟐𝟐

+-Sputtered n-GaN(0001)/H2O Interface

In this section, the amounts of band bending at the surface of the N2

+-sputtered n-GaN(0001)

layers of the c-sapphire/n-GaN(0001) samples at the chosen H2O vapor pressures and sample

temperatures are determined using Equations (5.1a) and (5.1b). To achieve this goal, the

variation of 𝐸𝐸g(𝑇𝑇)=(𝐸𝐸CB − 𝐸𝐸VB) and of (𝐸𝐸CB − 𝐸𝐸FB) with temperature 𝑇𝑇 is taken into account.

𝐸𝐸g(𝑇𝑇) was found at the 𝑇𝑇-values studied from the Varshni empirical formula [180] given in

Equation (5.3) using the Varshni fitting parameters reported for the n-GaN(0001) layers on

sapphire substrates. For n-GaN(0001), 𝐸𝐸g(𝑇𝑇) was calculated using 𝐸𝐸g(𝑇𝑇 → 0𝐾𝐾) ≅ 3.5 eV and the

Varshni parameters 𝛾𝛾 = 0.84 𝑚𝑚𝑒𝑒𝑉𝑉⁄𝐾𝐾 and 𝛽𝛽 = 790 𝐾𝐾 [124]. Using this equation, 𝐸𝐸g(𝑇𝑇) values of

3.431 𝑒𝑒𝑉𝑉 at 296 𝐾𝐾 and 3.37 𝑒𝑒𝑉𝑉 at 433 𝐾𝐾 were obtained, a decrease of 0.06 𝑒𝑒𝑉𝑉 over the 𝑇𝑇-range

studied in this work, as seen from the 𝐸𝐸g(𝑇𝑇) − vs − 𝑇𝑇 plot in Figure 5.20.

86

Figure 5.20: 𝐸𝐸g(𝑇𝑇) of n-GaN(0001) as a function of 𝑇𝑇 on the basis of the Varshni expression in

Equation (5.3). The 𝐸𝐸g(𝑇𝑇)-values at the temperatures studied are shown (open symbols) with their

Spline-fit curves obtained using the Varshni fitting parameters given in [180]. The inset shows the

𝑇𝑇-dependency of the Fermi energy, 𝐸𝐸FB, relative to the CB edge, i.e. (𝐸𝐸CB − 𝐸𝐸FB), with the symbols

being the values calculated from Equation (5.2a), and the solid line is a guide for the eye.

For the n-GaN(0001) layers studied in this work, (𝐸𝐸CB − 𝐸𝐸FB) was calculated from Equation

∗

(5.2) using 𝑛𝑛 = 5𝑥𝑥1023m−3 and 𝑚𝑚𝑒𝑒 = 0.2 𝑚𝑚𝑒𝑒 [165]. At temperature 𝑇𝑇 = 296 𝐾𝐾 (23 oC), values of

𝑁𝑁𝑐𝑐 = 2.198𝑥𝑥1024 m−3 and (𝐸𝐸CB − 𝐸𝐸FB) = +0.0378 eV were obtained. This suggests that the 𝐸𝐸FB of

non-degenerate and low-doped n-GaN(0001) layers lies below the CB edge 𝐸𝐸CB of the bulk at all

temperatures studied in the present work, and it shifts downwards by 0.04 eV as 𝑇𝑇 increases to

433 K (= 160 oC), as illustrated in the inset to Figure 5.20. The calculated values of (𝐸𝐸CB − 𝐸𝐸FB)

and 𝐸𝐸g(𝑇𝑇) = (𝐸𝐸CB − 𝐸𝐸VB) were then inserted into Equations (5.1a) and (5.1b) to find the numerical

+-values of the band bending voltage 𝑉𝑉bb for the N2

+-sputtered n-GaN(0001) surface and for the N2

sputtered n-GaN(0001)/H2O interface at the temperatures 𝑇𝑇 and H2O vapor pressures 𝑝𝑝 studied

in the present work. For the calculation using Equation (5.1b), (𝐸𝐸VB − 𝐸𝐸Ga3dB)= 17.5 eV [165,166]

and the fitted values of the peak binding energy 𝐸𝐸Ga3dS, relative to 𝐸𝐸FS(𝐸𝐸FB), of the measured

surface Ga 3dS XPS photoemission line at various values of 𝑇𝑇 and 𝑝𝑝 were used. An improper

value of (𝐸𝐸VB − 𝐸𝐸Ga3dB) of bulk n-GaN(0001) and an inadequate determination of 𝐸𝐸Ga3dS for the

n-GaN(0001) surface often leads to an ambiguous shift in the calculated surface band bending

𝑉𝑉bb [130-132,163-168].

87

In application of Equation (5.1a), the values of the energy separation between the surface

valence band edge, 𝐸𝐸VS and the pinned Fermi level, 𝐸𝐸FS i.e. (𝐸𝐸FS − 𝐸𝐸VS) were determined from the

measured VB XPS spectra of the N2

+-sputtered n-GaN(0001) surface and the N2

+-sputtered n-

GaN(0001)/H2O interface by linear extrapolation of the VB leading edge to the baseline tail, Figure

5.18. This linear extrapolation method is supposed to reduce instrumental spectral broadening

[44,85,171]. However, the energy (𝐸𝐸FS − 𝐸𝐸VS) often called the VBM energy, found by the linear

extrapolation method differs from the real value [166]. The presence of electronic states due to

dangling bonds on the surface of the n-GaN(0001) layers causes the surface VB edge 𝐸𝐸VS to

bend upwards. As a result, the value of (𝐸𝐸FS − 𝐸𝐸VS) deduced from the measured XPS spectra of

the surface valence band by the linear extrapolation method is underestimated due to the

presence of these surface electronic states in addition to spectral broadening [165]. Thus, the 𝑉𝑉bb

obtained from Equation (5.1a) using the VBM energy (𝐸𝐸FS − 𝐸𝐸VS) found by linear extrapolation

differs from the actual value, and a correction to this VBM energy could sort out the problem [166].

Alternatively, a reasonable value of the VBM energy can be found by a less elaborate

approach that involves a subtraction of (𝐸𝐸VB − 𝐸𝐸Ga3dB) of the bulk semiconductor from the peak

energy of the measured surface Ga 3dS XPS photoemission line [130-132]. This approach uses

Equation (5.1b), and its plausibility is solely governed by the choice of (𝐸𝐸VB − 𝐸𝐸Ga3dB), which is

constant for the bulk GaN material [166]. A variety of diverse values of (𝐸𝐸VB − 𝐸𝐸Ga3dB) have been

reported in the literature, and most are higher than 17.5 eV. A true value of (𝐸𝐸VB − 𝐸𝐸Ga3dB) for the

bulk GaN is essential for the accurate evaluation of the band bending at the n-GaN(0001) surface,

and the value 17.5 eV is based on the experimental results of soft X-ray emission (SXE)

measurements, a bulk sensitive method with an information depth of around 50 nm [164,165]. As

such, it is a value which has been endorsed to be the best choice, and it has been adopted in this

work.

The band bending 𝑉𝑉bb at the N2

+-sputtered n-GaN(0001) surface was first estimated using

the models based on Equations (5.1a) and (5.1b). The value of the VBM energy (𝐸𝐸FS − 𝐸𝐸VS) found

by linear extrapolation of the measured VB XPS spectra of the n-GaN(0001) surface upon

exposure to 9-min, 1-keV N2

+-ions is 2.626 eV (ca. ±3%), as seen from Figure (5.18a). An upward

surface band bending is expected for Ga-polar n-GaN(0001) layers when the VBM energy (𝐸𝐸FS −

𝐸𝐸VS) is less than 𝐸𝐸g(𝑇𝑇), which is larger than 3.35 eV for all temperatures studied in this work Figure

(5.20).

The values of 𝑉𝑉bb at the surface of the N2

+-sputtered n-GaN(0001) sample at 𝑇𝑇 = 296𝐾𝐾

that were determined from Equations (5.1a) and (5.1b) are 0.766 𝑒𝑒𝑉𝑉 and 0.837 𝑒𝑒𝑉𝑉, respectively,

88

with both equations incorporating (𝐸𝐸CB − 𝐸𝐸FB) and 𝐸𝐸g(𝑇𝑇) = (𝐸𝐸CB − 𝐸𝐸VB) calculated at 𝑇𝑇 = 296K. In

Equation (5.1b), the values (𝐸𝐸VB − 𝐸𝐸Ga3dB)= 17.5 eV and 𝐸𝐸Ga3dS = 20.055 𝑒𝑒𝑉𝑉 were used, while

(𝐸𝐸FS − 𝐸𝐸VS) = 2.626 eV was used in Equation (5.1a). The value of 𝑉𝑉bb obtained from Equation

(5.1b) is 0.07 eV higher than the value of 𝑉𝑉bb found from Equation (5.1a). This may be due to an

inaccurate determination of the VBM energy (𝐸𝐸FS − 𝐸𝐸VS) by linear extrapolation [165,166]. Huang

et al. [166] found a VBM correction ∆𝐸𝐸𝑉𝑉 that varies with doping density 𝑁𝑁𝑑𝑑 of the n-GaN(0001)

layer. They developed a valence-band fitting procedure using a theoretical density of states and

a quadratic depletion-layer approximation for the surface potential that depends on the dopants

and the density of surface states [130-132] to model the band bending at the n-GaN(0001) surface

and to correct the deduced VBM energy.

Analyzing the ∆𝐸𝐸𝑉𝑉 values of Huang et al. [166] for various 𝑁𝑁𝑑𝑑 values of their Si-doped n-

Ga(0001) films gave a rough estimate of ∆𝐸𝐸V = −0.07 eV for the less doped n-GaN(0001) layers

studied in present work. Subtracting this ∆𝐸𝐸V value from (𝐸𝐸FS − 𝐸𝐸VS) = 2.626 eV, the 𝑉𝑉bb found

from Equation (5.1a) matches the value of 𝑉𝑉bb found from Equation (5.1b). This suggests that the

band bending model based on Equation (5.1b), where the VBM energy is obtained by subtracting

(𝐸𝐸VB − 𝐸𝐸Ga3dB)= 17.5 eV from the peak BE, 𝐸𝐸Ga3dS of the measured surface XPS Ga 3d line, is a

reasonable model for the estimation of the band bending 𝑉𝑉bb at the surface of the n-GaN(0001)

layers [82,95,179]. The values obtained for 𝑉𝑉bb of the N2

+-sputtered n-GaN(0001) surface of the

samples studied in the present work differs by ±0.3 eV from the values of 𝑉𝑉bb reported in the

literature for n-GaN(0001). Such a difference may be linked to different doping of the n-GaN(0001)

layers and to the surface photovoltage (SPV) at the n-GaN(0001) surface [165]. The diversity in

𝑉𝑉bb for n-GaN(0001) layers can also be related to the methodology used for calculating 𝑉𝑉bb and

to ex-/in-situ cleaning procedures of the n-GaN(0001) surface, as the type and number of

adsorbates and states largely affect band bending [44,85,130-132,165-174].

The calculated values of 𝐸𝐸g(𝑇𝑇) and (𝐸𝐸CB − 𝐸𝐸FB), together with the values of the measured

VBM energy (𝐸𝐸FS − 𝐸𝐸VS) were inserted in Equation (5.1a) to get the numerical values of 𝑉𝑉bb at the

𝑁𝑁2

+-sputtered n-GaN(0001)/H2O interface at the chosen H2O vapor pressures 𝑝𝑝 and sample

temperatures 𝑇𝑇. The NAP-XPS runs were made at all chosen values of 𝑝𝑝 on a fresh c-sapphire/n-

GaN(0001) sample held at a certain 𝑇𝑇. The 𝑉𝑉bb calculation was repeated for the results of the

NAP-XPS runs made at the same 𝑝𝑝-values on a different fresh sample, held at another 𝑇𝑇, and so

on. The outcome is illustrated in Figures (5.21a) and (5.21b), which depict, respectively, plots of

𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-p(mbar) and 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-T(oC) for these NAP-XPS runs.

89

(a)

(b)

Figure 5.21: The band bending voltage 𝑉𝑉bb (open symbols) found from Equation (5.1a) for the

N2

+-sputtered n-GaN(0001) surface and the N2

+-sputtered n-GaN(0001)/H2O interface of different

fresh samples plotted as (a) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-p(mbar) at different temperatures and (b) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-

T(oC) at different H2O vapor pressures. The solid lines are guides for the eye.

(a)

(b)

Figure 5.22: The band bending 𝑉𝑉bb deduced from Equation (5.1b) for the N2

+-sputtered n-

GaN(0001) surface and the N2

+-sputtered n-GaN(0001)/H2O interface of different fresh samples

plotted as (a) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-p(mbar) at a given temperature and (b) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-T(oC) at various H2O

vapor pressures. The solid lines are guide for the eye.

The 𝑉𝑉bb found from Equation (5.1a) for the 𝑁𝑁2+-sputtered n-GaN(0001)/H2O interface at

𝑇𝑇 = 296 K and 𝑝𝑝 = 0.02 mbar is 0.617 eV, which is less than 𝑉𝑉bb

= 0.766 eV obtained for the N2

+-

sputtered n-GaN(0001) surface. It deserves noting here that this is due to the termination and

removing of initial surface states and dangling bonds through Ga2O3 passivation layer, arising

90

from oxidation of dangling (free) Ga atoms and the unstable (oxygen-deficient) Ga2Ox species,

which is enhanced by annealing at high temperatures, upon H2O-vapor exposure; thus a

prominent decrease in BB at the N2

+-sputtered n-GaN(0001)/H2O interface [50b,50c].

The observed drop in 𝑉𝑉bb after injection of H2O molecules could also be linked to

dissociation of H2O molecules to OH− and H+ ions that react with surface dangling Ga+ and N−

ions to form Ga-OH and N-H bonds that terminate the initial surface states [85,95,165]. This

explains the decreasing trend of 𝑉𝑉bb as 𝑝𝑝 increases up to 𝑝𝑝 = 0.1 mbar, above which 𝑉𝑉bb levels

off. The constancy of 𝑉𝑉bb at high 𝑝𝑝 can be related to a saturation coverage of n-GaN(0001) surface

with such dissociated H2O molecules [95], or to the satiety of passivated Ga2O3 layer due to

removal of initially existing surface Ga2Ox layer and dangling Ga atoms.

On the other hand, the same calculated values of 𝐸𝐸g(𝑇𝑇) and (𝐸𝐸CB − 𝐸𝐸FB), combined with

(𝐸𝐸VB − 𝐸𝐸Ga3dB)=17.5 eV and the deduced peak energies 𝐸𝐸Ga3dS of the measured surface XPS

𝐺𝐺𝑎𝑎 3𝑑𝑑𝑆𝑆 photoemission line were inserted into Equation (5.1b) to find 𝑉𝑉bb for the samples

described in Figure (5.21) at the same 𝑇𝑇 and 𝑝𝑝. The results this calculation plotted as 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-

p(mbar) and 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-T(oC) are shown in Figures (5.22a) and (5.22b).

NAP-XPS runs were also conducted on the N2

+-sputtered n-GaN(0001)/H2O interface of

the same sample that was used at all chosen temperatures, with each 𝑇𝑇 being used at the same

values of 𝑝𝑝 ≥ 0.1 mbar. The calculated values of 𝑉𝑉bb from Equations (5.1b) and (5.1a) are shown

in the plots of Figures (5.23a) and (5.23b) and Figures (5.24 a) and (5.24 b), respectively.

(a)

(b)

91

Figure 5.23: The band bending 𝑉𝑉bb deduced from Equation (5.1b) for the N2

+-sputtered n-

GaN(0001)/H2O interface of the same sample used at all temperatures and under the same

pressures plotted as (a) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-p(mbar) for various temperatures and (b) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-T(oC)

under various H2O vapor pressures. The solid lines are guidesfor the eye.

(a)

(b)

Figure 5.24: The band bending 𝑉𝑉bb deduced from Equation (5.1a) for the N2

+-sputtered n-

GaN(0001)/H2O interface of the same sample used at all temperatures and under the same

pressures plotted as (a) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-p(mbar) at the given temperatures and (b) 𝑉𝑉bb(𝑒𝑒𝑉𝑉)-vs-T(oC)

under various H2O vapor pressures. The solid lines are guides for the eye.

+-Figures (5.23) and (5.24) show that the 𝑉𝑉bb obtained from Equation (5.1b) for the N2

sputtered n-GaN(0001)/H2O interface of the same sample has a trend with varying H2O vapor

pressure and sample temperature similar to that found for 𝑉𝑉bb calculated from Equation (5.1a).

The arguments proposed above to interpret the behavior of the 𝑉𝑉bb with 𝑇𝑇 and 𝑝𝑝 shown in Figures

+

5.22(a) and 5.22(b) are also valid for the plots of Figures 5.23 and 5.24. The sputtering with N2

ions reduces the adventitious carbon species, but it is not effective for removing oxides existing

on the Ga-polar n-GaN(0001) surface, and it might create extra N-vacancies and Ga dangling

bonds [130-135]. At a certain 𝑝𝑝, 𝑉𝑉bb at the N2

+-sputtered n-GaN(0001)/H2O interface was found to

decrease as 𝑇𝑇 increases. The decrease in 𝑉𝑉bb at a certain 𝑝𝑝 with increasing 𝑇𝑇 could be partly

attributed to the decrease in (𝐸𝐸CB − 𝐸𝐸FB) and 𝐸𝐸g(𝑇𝑇) of bulk GaN with increasing 𝑇𝑇 Figure 5.20.

However, the calculated variation (∆≤ 0.1 eV) of (𝐸𝐸CB − 𝐸𝐸FB) and 𝐸𝐸g(𝑇𝑇) with 𝑇𝑇 cannot be solely

responsible for the entire decrease in 𝑉𝑉bb with increasing 𝑇𝑇. Another probable cause for such

decrease of 𝑉𝑉bb with increasing sample temperature is the successive formation of passivated

92

layer of Ga2O3 oxide due to oxidation of Ga2Ox molecules that becomes more effective by

annealing.

In the case of the band bonding model based on Equation 5.1(a), part of 𝑉𝑉bb reduction

and uncorrelated behavior with increasing 𝑇𝑇 may be linked to the surface VBM energy (𝐸𝐸FS − 𝐸𝐸VS),

whose estimation by the linear extrapolation method incorporates significant uncertainty and is

not monotonous with varying 𝑇𝑇, depending on the smoothness of the measured VB XPS

spectrum. The accuracy of the linear fits of the VB leading edge and the baseline was limited by

the scattered VB data points in the measured VB NAP-XPS spectra, which is large at high sample

temperatures and H2O-vapor pressures. This results in noticeable diversity in the deduced VBM

energy, and thus in the respective surface band bending. Also, the spectral feature of VBM and

related spectrum near the surface is overlapped in the direction of the conduction band with that

of the surface states, leading to underestimation of the VBM energy [165-167].

In the case of the model based on Equation (5.1b), part of the decrease of 𝑉𝑉bb with

increasing 𝑇𝑇 might be assigned to the slight increase in the peak BE 𝐸𝐸Ga3dS of the surface Ga 3dS

photoemission line as 𝑇𝑇 increases, in addition to the variation of (𝐸𝐸CB − 𝐸𝐸FB) and 𝐸𝐸g(𝑇𝑇) with 𝑇𝑇.

The prominent decrease in 𝑉𝑉bb at 𝑇𝑇 = 160 ℃ cannot be entirely accounted for by these causes,

but surface gallium oxides (e.g. Ga2O3) is a likely cause as it is enhanced due to oxidation of the

initial Ga2Ox surface layer that is more prominent at high annealing temperatures [50b,50c],

leading to passivation of existing surface states [179] and not to improbable breakdown of OH−

bonds at high temperatures [95]. As pointed previously, determination of the peak energy BE

𝐸𝐸Ga3dS of the surface Ga 3dS photoemission line is accurate enough using the sophisticated

CassaXPS program and the value of energy difference (17.5 eV) between Ga 3d level and VB

edge of bulk GaN appears to be an optimum value [165-167]. Therefore, the model based on

Equation (5.1b) is more subtle and yields appropriate values of 𝑉𝑉bb compared with those found

from the model based on Equation (5.1a).

The behavior of 𝑉𝑉bb for the N2

+-sputtered n-GaN(0001)/H2O interface of different, fresh

samples held at different constant temperatures for 𝑝𝑝 ≥ 0.1 mbar, Figures 5.21 and 5.22 is similar

to the behavior deduced for the N2

+-sputtered n-GaN(0001)/H2O interface of the same sample

used at all temperatures, with each 𝑇𝑇 being fixed for all pressures 𝑝𝑝 ≥ 0.1 mbar, Figures 5.23 and

5.24. Yet, the method using different, fresh N2

+-sputtered n-GaN(0001) layers, each maintained

at a different temperature, to measure their surface NAP-XPS CL and VB spectra at various H2O

vapor pressures probably provides more informative results than those obtained by using the

same sample heated and sustained at all temperatures, each under the same pressures.

93

5.5 Exploration of Bonding and Interactions on n-GaN(0001) Surface

Coated with L-cysteine Monolayers by UHV-XPS

Probing the bonding interactions of biological molecules on the surface of semiconducting films

is important for evaluating the performance of biosensors and electronic devices integrating them

[1-6,19,181]. Among the organic molecules that form covalently bonded monolayers on a

semiconductor surface are the sulfur-containing amino acid L-cysteine and small peptides, the

basic building blocks of proteins and DNA.

The wide band gap III-nitride compounds are very good materials for this purpose as they

possess remarkable properties like living cell biocompatibility, and they can withstand severe

conditions [41,62,63,181]. Exploring the phenomena of the molecular adsorption of amino acids

onto the surface of III-nitride films aids in understanding the nature of the surface modifications

and interfacial bonding mechanisms occurring in ambient conditions, and the III-nitride-based

heterostructures linking amino acids reveal the direct relationship between the complex biological

world and digital electronics [41,62,72,76,82,181-183].

A bottom-up approach is to study complex biological systems by exploring the basic

processes occurring on surfaces functionalized with the amino acid L-cysteine that serves as a

model for the chemisorption of biofunctional molecules. The natural sulfur containing L-cysteine

is a potential candidate for preparing biological surfaces that can adapt their physical and

chemical properties to the external changing conditions that can occur in the bio-/organo-sensing

devices integrating these surfaces. This is because the thiol head of L-cysteine gives it the ability

to build self-assembled monolayers at specific surfaces via the formation of sulfur-metal bonds

with metal ions on the surface. Its amino (NH2) and carboxyl (COOH) functional groups provide

the possibility of a proton exchange with the environment or an internal proton transfer, resulting

in the L-cysteine chemical sates of neutral and zwitterionic molecules. The possibility of forming

such L-cysteine chemical states and the bonding of its molecules to solid surfaces were explored

by studying the physical and chemical properties of metal surfaces functionalized with L-cysteine

under various conditions.

Not much is known about the interactions on L-cysteine/semiconductor interfaces and

their stability in aqueous or biological media, only a few studies reported on functionalizing Si,

InAs, and GaAs with L-cysteine and peptides [50,59,68]. Hence, it is useful to realize the physics

and chemistry of L-cysteine/semiconductor interfaces to gain further insight into the performance

and the properties of the chemical/biological sensors incorporating them.

94

A good piece of information can be extracted from the characterization of the interfaces of

inorganic-organic structures by spectroscopic techniques [6,59,68], primarily when the surface of

the III-nitride layer is biocompatible and stable in physiological conditions. The organo-

functionalized layers of n-GaN(0001) are one example of the III-nitrides that forms a particular

interface whose featured properties render it a typical prospective model for hybrid organic-

inorganic heterostructures of biosensing devices operative in various conditions

[41,62,72,76,82,181-183]. The surface properties of n-GaN(0001) are not abundantly controllable

or explicable. The impact of its functionalization on device operation and interfacial bio-physical-

chemical interactions is crucial. So, the surface properties should be identified, as they influence

the performance, yield, and cost-effectiveness of the devices using them. Properties of the n-

GaN(0001) surface are governed by the fabrication and growth conditions, the surface-finishing

routes, and the structure and dangling bonds of the surface, which ultimately becomes the

interface for functionalized materials.

Various post-growth treatments are used for cleaning the surface of n-GaN(0001) layers,

but there is no standard and reproducible method of preparing a clean n-GaN(0001) surface prior

+

to characterization or use in electronic devices [76,182]. Sputtering with low-energetic (1-keV) N2

ions was found to cause minor damage and roughness of the surface of the n-GaN(0001) layers

studied in the present work, and it is effective in removing the adventitious surface carbon species,

but not the stabilized surface oxides. The surface modification induced by the passivation

processes, the features of the monolayers formed on the surface, and the ensuing bonding

mechanisms are reliant on the preparation history and cleanliness of the surface. This allows

exploration of the reactivity and selectivity of site-specific processes for amino acids containing

multiple functional groups, which yield unique hydrogen bonding among themselves due to the

amino and the carboxylic acid groups. Knowledge of the bonding interactions of amino acids on

the n-GaN(0001) surface helps to understand the biological processes in physicochemical

situations that mimic real biomolecular processes of living cells and various amino acids.

Some hitches are still faced when studying the adsorption behavior and physical-chemical

interactions of amino acids on the n-GaN(0001) surface [41,62-64,68]. The electronic structure

and chemistry of the n-GaN(0001) layers are still debated and the understanding of the physical

and chemical interaction mechanisms at the surface responsible for the n-GaN(0001) interfacial

organo molecular functionality is not complete. The diversity in the literature results may be

related to the use of differently-fabricated n-GaN(0001) layers and of dissimilar physical and

chemical cleaning procedures to remove unwanted spurious species from the surface. More

95

insight into these issues would be helpful to understand the ensuing surface contact formation

and chemical reactivity upon exposing and linking the n-GaN(0001) surface to molecules of L-

cysteine, the simple sulfur-containing amino acid that plays a primary role in complex biological

systems and living cell organisms. Such studies can be achieved with the aid of surface sensitive

spectroscopic and microscopic techniques, including the conventional UHV-XPS technique

[116,184].

Though n-GaN(0001) possesses unique properties and high living-cell biocompatibility,

little work has been conducted on n-GaN(0001)/L-cysteine interfaces, as getting amino acids and

peptides to bind reliably and firmly to the n-GaN(0001) surface is challenging, and the surface

had to be modified to become receptive to bind inorganic/organic molecules [51,64,68].

Functionalization of semiconductor surfaces with organo-/bio-molecules can be attained via

solution-growth [51,59,64] and thermal evaporation [185,186]; and various experimental

techniques are often used for characterization of these surfaces, including AFM, LEED and XPS.

These preparation methods and characterization systems were also used for metal surfaces

coated with such organo-/bio-molecules [187,188].

L-cysteine acts as a catalyzing site for certain peptides and links with a variety of human

disease species. Hence, using L-cysteine under various environmental conditions is valuable from

two viewpoints: fundamental research and physiological studies and clinical diagnosis. Some of

the above-cited literature studies on n-GaN(0001)/L-cysteine interfaces showed a noteworthy

catalytic response without the need for prior surface modification, and they provided evidence for

changes in the physical properties of the surface. In this study, AFM, LEED, and XPS were used

to detect L-cysteine molecules and sulfur-group attachment on the n-GaN(0001) surface, besides

exploring the induced chemical modification and structural stability of the ensuing n-GaN(0001)/L-

cysteine interfaces.

The influence of the temperature the crucible containing L-cysteine powder and of the time

of deposition of L-cysteine molecules onto the n-GaN(0001) surface and the post-deposition

annealing of the n-GaN(0001)/L-cysteine samples obtained on the bonding configurations taking

place on the surface was studied in detail. The focus was partially on the experimental procedures

adopted to attain a stable n-GaN(0001)/L-cysteine interface and on the modification of the n-

GaN(0001) surface for reliable binding to L-cysteine molecules. No ex-situ wet physical/chemical

cleaning was done on the as-received sapphire/n-GaN(0001) samples, except for flushing them

with argon gas. To proceed further, the sample was firmly fixed onto the Mo holder with a thin Mo

96

foil placed underneath the back surface of the sapphire wafer. The Mo-holder-sample assembly

was mounted onto a rotatable translational bar installed in the chambers of the SPECS XPS

instrument. In-situ heating of the c-sapphire/n-GaN(0001) sample was achieved via irradiating the

Mo foil back surface with a SPECS infrared laser heater (IRHL) source. As described in chapter

four, the PT100 sensor thermally anchored to the n-GaN(0001) surface was used to record its 𝑇𝑇

and its N2

+-ion sputtering was carried out for the reasons mentioned previously.

L-cysteine was sublimed onto the N2

+-sputtered n-GaN(0001) surface in UHV (𝑝𝑝 < 10−8

mbar) from the well controlled heated crucible of an organic molecular effusion cell (SPECS OME

40 cell) attached to the XPS preparation chamber. The temperature of the OME-40 crucible was

chosen to be between 90 oC and 105 oC, below the decomposition temperature of L-cysteine

molecules. Most of the L-cysteine deposition was carried out at 105 oC for deposition times

between 20 and 40 minutes, adopted to get varying L-cysteine monolayers. The n-GaN(0001)/L-

cysteine samples were heated in the XPS working chamber to temperatures of 25 oC, 75 oC, 100

oC, 130 oC and 150 oC. At each of these temperatures, the samples were annealed for 15 min,

30 min, and 1 h before collecting the UHV-XPS spectra of the n-GaN(0001)/L-cysteine interfaces

attained at each of these experimental conditions.

The structural features and chemistry of the N2

+-sputtered n-GaN(0001)/vacuum and n-

GaN(0001)/L-cysteine interfaces were investigated by measuring their ex-situ AFM images, in

addition to their in-situ UHV LEED micrographs and the XPS spectra of the CL transitions of their

constituents. The as-measured CL XPS spectra were analyzed with the CasaXPS software using

the GL30 fitting function, combined with Shirley background subtraction. The binding energy (BE)

shift due to instrumental effects was corrected relative to the Mo Fermi level, estimated from the

measured VB XPS spectra of the empty Mo holder. In particular, the measured UHV-XPS spectra

of annealed N2

+-sputtered n-GaN(0001)/L-cysteine samples yielded valuable information on the

chemical composition of the surface constituents and operative chemical states; thus, providing

a fingerprint on using n-GaN(0001) layers in biofunctionalization and biosensing devices.

5.5.1 AFM images and LEED Micrographs of the 𝐍𝐍𝟐𝟐

+-sputtered n-

GaN(0001)/L-cysteine Interface

The LEED micrographs were taken with 120 eV electrons on the surface of bare and N2

+-sputtered

n-GaN(00001) layers before and after depositing L-cysteine molecules at 105 oC for 40 minutes.

Figures 5.25(a) and 5.25(b) show that the LEED micrograph of the surface of the N2

+-sputtered

sample shows faint diffraction spots with a clear 1×1 microstructure. No diffraction spots are seen

97

on the n-GaN(0001)/L-cysteine interface, signifying that a thick L-cysteine layer adsorbed on the

surface is masking them. The diffraction spots reappear and form the 1×1 microstructure again

as depicted in Figure 5.25(c) after thermal annealing of the n-GaN(0001)/L-cysteine sample at 75

oC, then at 100 oC, and finally at 130 oC with each temperature kept constant for 30 minutes.

(a) (b) (c)

Figure 5.25: : LEED images taken with 120 eV electrons on the surface of (a) sputtered by 1-keV

N2

+ ions for 9 minutes and (b) the n-GaN(0001) layer coated with L-cysteine at 105 oC for 40

minutes, and (c) the n-GaN(0001)/L-cysteine sample after thermal annealing at 75 oC, then at

100 oC, and finally at 130 oC with each temperature kept constant for 30 minutes.

+-Figure 5.26 displays a typical AFM image of a 2.5 × 2.5 µm spot on the surface of the N2

sputtered n-GaN(0001) layer after deposition of L-cysteine molecules at 105 oC for 40 minutes. A

terrace-like undulated surface structure is seen to exhibit a few bulging protrusions with a small

RMS (~ 0.729 nm). However, not much can be inferred about the influence of L-cysteine depositin

from such AFM images.

Figure 5.26: An AFM image of an N2

+-sputtered n-GaN(0001) layer coated with L-cysteine at 105

oC for 40 minutes.

98

5.5.2 UHV-XPS Spectra of Inner Core-Level Photoemission Lines of

𝐍𝐍𝟐𝟐

+-sputtered n-GaN(0001)/L-cysteine Interface

The experimental UHV-XPS spectrum of the Ga 3s / S 2p photoemission line of the N2

+-sputtered

n-GaN(0001) surface is depicted in Figure 5.27(a) and the Ga 3s / S 2p lines of the N2

+-sputtered

n-GaN(0001)/L-cysteine interface after depositing L-cysteine molecules at 105 oC for the time

periods of 20 minutes and 40 minutes are shown in Figures 5.27(b) and 5.27(c), respectively.

Such samples were not heated or annealed after covering them with L-cysteine, but kept at RT

untill the XPS spectra were collected.

It is noted that significant changes in the features and intensity of the observed core-level

(CL) peaks of the constituents of the N2

+-sputtered n-GaN(0001) surface took place upon

functionalization with L-cysteine molecules, particularly for long deposition times. Adjacent to the

Ga 3s photoemission line positioned nearly at BE = 160.9 eV, there is an overlapped

photoemission peak that can be resolved with two binding energies (BEs) arising from the sulfur

presence and related to the spin-orbit splitting of the S 2p1/2,3/2 doublet and to the ensuing bonds

with a constituent of the functionalized surface. This suggests that the L-cysteine molecules are

chemisorbed to the N2

+-sputtered n-GaN(0001) surface through the sulfur atoms of the thiol (SH)

head group that probably make bonds with Ga dangling bonds Figures 5.27(b) and 5.27(c). The

thiolate species becoming more pronounced upon increasing the amount of the deposed L-

cysteine molecules, resulting in multi-overlayers, among which interactions are likely.

(a)

(b)

99

(c)

Figure 5.27: Experimental UHV-XPS spectra of the Ga 3s / S 2p photoemission lines for the n-

GaN(0001) surface (a) after sputtering with 9-min, 1-keV N2

+ ions and after depositing L-cysteine

molecules at 105 oC for (b) 20 minutes and (c) 40 minutes with the samples being kept at room

temperature (RT) until the XPS measurements were collected. The colored curves are the

deconvoluted fits of the Ga 3s and S 2p CL lines as labeled inside the plots.

Understanding of the origin of the bonding and the interactions responsible for the peaks

Figure 5.27 can be accomplished through proper analysis and interpretation of the features of the

associative N 1s, C 1s, and O 1s photoemission lines taken on the interface of the same samples.

This can be further comprehended with the knowledge of the molecular structure of the sulfur-

containing L-cysteine amino acid and of the nature of the chemical forms and interactions that

might existing [59,185-188].

As the amino acid L-cysteine (SH-CH2-CH(NH2)-COOH) has three coordination sites,

namely the carboxyl, the amino, and the side-chain thiol functional groups, it can be present in

three possible chemical forms, depending on the environmental conditions of the molecule

[6,189]. The most common forms are the neutral form, typical of isolated molecules in the gas

phase, and the zwitterionic form, typical for solid amino acid crystals and for molecules at poorly

reactive surfaces, where a proton is transferred from the carboxyl group (COOH) to the amino

acid group (NH2) of the same molecule, forming the charged COO− and NH3

+ groups. On surfaces

this zwitterionic form probably occurs via bonding among the amino and carboxylic acid groups

of L-cysteine molecules in monolayers linked poorly among themselves. The third form is the

anionic state, in which deprotonation occurred due to breakdown/formation of chemical bonds

upon interaction of the amino acid with the substrate, leaving a negatively charged molecule

through the H-deficient COO− group.

100

The type of bonding configurations L-cysteine molecules could form with their environment

depends largely on the nature and structure of the metal or semiconductor surface and on the

phase state of the L-cysteine source (solid, liquid, or gas) used to coat the surface [59,185-188].

In general, the zwitterionic amino acid (NH3

+) and carboxyl (COO−) groups could be oriented

towards or away from the surface and link directly to the surface and/or to the initial

monomolecular L-cysteine layer. The latter formed a strong bond (chemisorbed) via the thiol

group (-SH) to the intrinsic dangling bonds and the spurious impurity centers existing on the n-

GaN(0001) surface.

The S 2p doublet spectra of the N2

+-sputtered n-GaN(0001)/L-cysteine interface can be

resolved into two or four peaks depending on the surface coverage and the thickness of the L-

cysteine monolayers, hence on the amount of thiol (-SH) group heads as well as on the

cleanliness of the N2

+-sputtered n-GaN(0001) surface as seen from Figure 5.27. The two spin-

orbit S 2p3/2 and S 2p1/2 peak pairs giving the bonding with the Ga dangling bonds (S-Ga) were

fitted with a fixed FWHM of 1.2 eV and a fixed 2:1 intensity ratio at 163.1 eV and 164.3 eV, which

is characteristic for thiolate species; the reported diverse BE values of the S 2p doublet are related

to the type and nature of the investigated surface [59,187,188]. The other resolved peaks of the

S 2p doublet line as C-S bonds could be tentatively related to carbon atoms in the L-cysteine

monolayers [186,187] and/or to the linkage of the sulfur atom to the residual adventitious carbon

atoms remaining on the n-GaN(0001) surface after N2

+-sputtering. The nature of the peaks and

the corresponding BEs obtained from their fits to the measured Ga 3s / S 2p photoemission lines

are listed in Appendix A.

The distinction between the zwitterionic and non-zwitterionic forms can be inferred from

the N 1s XPS spectrum, since the NH2 and NH3

+ groups have different photoemission signatures.

+

The NH2 peak is observed at BE = 400 𝑒𝑒𝑉𝑉 for the neutral amino group, while the NH3 peak

appeared at BE ≥ 401 𝑒𝑒𝑉𝑉 [189]. The XPS N 1s line of the N2

+-sputtered n-GaN(0001) surface is

shown in Figure 5.28(a) and the N 1s line of the N2

+-sputtered n-GaN(0001)/L-cysteine interface

after depositing L-cysteine at 105 oC for 20 minutes and 40 minutes are shown in Figures 5.28(b)

and 5.28(c), respectively.

101

(a)

(b)

(c)

Figure 5.28: Experimental UHV-XPS spectra of the N 1s photoemission lines for the n-GaN(0001)

surface (a) after sputtering with 9-min, 1-keV N2

+ ions and after depositing L-cysteine molecules

at 105 oC for (b) 20 minutes and (c) 40 minutes, with the samples being kept at room temperature

(RT) until the XPS measurements were collected. The colored curves are the deconvoluted fits

of the overlapped Ga LMM sub-peaks and of the peaks of the uncharged/charged amino and

carboxylic acid groups as labeled inside the plots.

The N 1s XPS spectra of Figure 5.28 are obtained on the samples prepared under the

+

experimental conditions specified in Figure 5.27. The peaks associated with the NH2 and NH3

groups are fitted at BEs of 400 eV and 401.9 eV, respectively. It can be noticed that the

zwitterionic configuration involving the ensuing charged NH3

+ amino group profoundly appears for

102

a larger L-cysteine coverage of the N2

+-sputtered n-GaN(0001) surface. This implies that

interactions occur among the molecules of the upper L-cysteine layers and as a result of a proton

transfer from the carboxyl group to the amino group.

The viability of the above-discussed points of view is supported by the chemistry and

bonding features of deconvoluted UHV-XPS C 1s and O 1s photoemissions lines shown in

Figures 5.29(a) and 5.29(b), which were obtained for the same N2

+-sputtered n-GaN(0001)/L-

cysteine interface achieved by depositing thicker L-cysteine layers at 105 oC for 40 minutes.

(a)

(b)

Figure 5.29: Measured XPS photoemission lines of (a) C 1s and (b) O 1s core-level transitions

for the N2

+-sputtered n-GaN(0001)/L-cysteine interface coated with L-cysteine at 105 oC for 40

minutes, with the samples being kept at RT until collecting the XPS measurements. The colored

curves are the deconvoluted fits of the respective sub-peaks as seen from the plots.

The C 1s photoemission line of the N2

+-sputtered n-GaN(0001)/L-cysteine interface

without exposing it to heating/annealing is deconvoluted into three well resovable peaks at the

BEs of 285.35 eV, 286.6 eV, and 288.9 eV, which can be related to the three different carbon

atoms in the L-cysteine molecule, which can be labeled as CS, CN, and CO [185-187], respectively.

Functional groups bonded to the carbon atoms cause chemical shifts of the C 1s core level. It is

reported that the electronegativity of the C, N, O, and S atoms are 2.6, 3.0, 3.4, and 2.6,

respectively [185]. It is known that when the electronegativity of the bonding atom is large, the

binding energy shift also becomes large. On the basis of electronegativity, one can assign the

peaks at 285.5 eV, 286.5 eV, and 289 eV to CS, CN, and CO, respectively. This implies that the CS

peak can be related to the carbon bonding with the thiol group (-SH) that would gives rise to the

103

C-S bond due to residual carbon species remained intact on the surface. The CN peak results

from the C-N bonding with the amino (NH2) group and the CO peak corresponds to the C-O bond

in the carboxyl group. Diversity is reported in the literature on the values and origin of the binding

energies related to the peaks comprising the O 1s photoemission line of metallic and

semiconducting surfaces that are functionalized with L-cysteine [185,187]. Regarding the O 1s

photoemission line of the N2

+-sputtered n-GaN(0001)/L-cysteine interface that was not subjected

to excessive heating/annealing, the peaks corresponding to the BEs of 532 eV and 533 eV can

be cautiously assigned to the carboxyl group that exists in the ionic (COO–) and the protonated

(neutral) (COOH) molecular states, respectively.

Subjecting the N2

+-sputtered n-GaN(0001)/L-cysteine samples to gradual heating at

different temperatures near and above 100 oC, followed by annealing at these temperatures for

prolonged times reduced the intensity of the photoemission peaks of the S 2p1/2, 3/2 doublet, as

well as those of the uncharged/charged amino and carboxyl groups in the N 1s and the C 1s XPS

spectra, but such heating/annealing processes were found to enhance the surface oxidation

greatly. This probably due to a structural difference of the COOH group exists between multilayer

and monolayer L-cysteine and/or that re-evaporation of the upper L-cysteine layers had occurred

at elevated temperatures. Such behavior is illustrated in Figures 5.30 to 5.33, for the Ga 3s / S

2p, N 1s, C 1s, and O 1s photoemission lines of the N2

+-sputtered n-GaN(0001)/L-cysteine

interface after exposing the samples to prolonged heating/annealing at 100 oC, 130 oC, and 150

oC. At each of these temperatures, the sample was annealed for 30 minutes.

+-

sputtered n-GaN(0001) sample at elevated temperatures (≥ 100 ℃) results in breaking the

chemical bonding of Ga atoms with sulfur atoms of thiol (SH) functional group in L-cysteine

molecules, which in turn oxidized to form Ga2Ox oxide layer. Moreover, the upper L-cysteine

monolayers appear to mask the chemisorbed Ga-S-H bonding underneath. However, upon

annealing the sample at high temperatures for long times the chemical Ga-S-H… bonds regained

their original features before functionalization. This can be linked to the removal of the topmost

physisorbed L-cysteine molecules and/or to the breaking of their weak intermolecular bonds, as

well as to structural changes in these L-cysteine layers.

It can be argued that prolonged heating/annealing of the L-cysteine functionalized N2

104

(a)

(b)

(c)

+-

sputtered n-GaN(0001)/L-cysteine interface obtained by depositing L-cysteine at 105 oC for 40

minutes and after heating/annealing the sample for 30 minutes at (a) 100 oC, (b) 130 oC, and (c)

150 oC before collecting the XPS measurements. The colored curves are the deconvoluted fits of

the Ga 3s and S 2p CL lines as labeled inside the plots.

Figure 5.30: Experimental UHV-XPS spectra of the Ga 3s / S 2p photoemission lines for the N2

105

(a)

(b)

(c)

Figure 5.31: Experimental UHV-XPS spectra of the N 1s photoemission line for the N2

+-sputtered

n-GaN(0001)/L-cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes and

after heating/annealing the sample for 30 minutes at (a) 100 oC, (b) 130 oC, and (c) 150 oC before

collecting the XPS measurements. The colored curves are the deconvoluted fits of the N 1s line

as labeled inside the plots.

The above interpretation could be supported by the shape and features of N 1s line of the

N2

+-sputtered n-GaN(0001)/L-cysteine sample illustrated in Figures 5.28 and 5.31. The formation

of zwitterionic molecules became more pronounced with increasing coverage of the surface, as

seen from the intensity increase of the photoemission line of the NH3

+ group (BE = 402 eV) relative

to the one of the NH2 group (BE = 399.9 eV). Prolonged heating of the functionalized sample to

106

+

temperatures at 100 °C and higher decreased the intensity of the photoemission line of the −NH3

group relative to the one of the NH2 group, indicating a reduction in the concentration of zwitterions

on the surface. This suggests that the zwitterions play a role in the bonding process between the

L-cysteine molecules in the adsorbed layers.

(a)

(b)

(c)

Figure 5.32: Experimental UHV-XPS spectra of the C 1s photoemission line for the N2

+-sputtered

n-GaN(0001)/L-cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes and

after heating/annealing the sample for 30 minutes at (a) 100 oC, (b) 130 oC, and (c) 150 oC before

collecting the XPS measurements. The colored curves are the deconvoluted fits of the C 1s line

as labeled inside the plots.

107

(a)

(b)

(c)

Figure 5.33: Experimental UHV-XPS spectra of the O 1s photoemission line for the N2

+-sputtered

n-GaN(0001)/L-cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes and

after heating/annealing the sample for 30 minutes at (a) 100 oC, (b) 130 oC, and (c) 150 oC before

collecting the XPS measurements. The colored curves are deconvoluted fits of the O 1s line as

labeled inside the plots.

In summary, the results of this part demonstrate that the as-measured XPS spectra of the

N2

+-sputtered n-GaN(0001)/L-cysteine interfaces gave a clue on interaction and bonding

mechanisms taking place between L-cysteine molecules and the constituents of the N2

+-sputtered

n-GaN(0001) surface. At low sample temperatures, probable interaction between free Ga atoms

on the N2

+-sputtered n-GaN(0001) surface and sulfur of the thiol (-SH) functional group of the first

formed L-cysteine monolayer. Among the additional L-cysteine layers, covalent bonding between

108

other functional groups of the L-cysteine occurs. Heating the sample at elevated temperatures (≥

100 ℃) probably leads to breaking of Ga-S bonds and the resulting Ga atoms would be oxidized

giving rise to extra Ga2Ox oxide and to destruction of covant bonds between the NH2 and NH3+

functional groups. This opens the door for the use of c-sapphire/n-GaN(0001)/L-cysteine

interfaces in real world electronic biosensing devices. The N2

+-sputtered n-GaN(0001) surface

was found to be a workable substrate for functionalization with the simplest L-cysteine amino acid,

as its thiol (-SH) head steadfastly binds to Ga ions on such a modified n-GaN(0001) surface at

low temperatures. The formed zwitterions (NH3+ and COO– groups) may render the N2

+-sputtered

n-GaN(0001)/L-cysteine interface receptive to chemical bonding with external organic and

biological species; thus, such interfaces may function as an optical/electrical detecting probe for

them under normal operating ambient temperatures.

The preliminary analysis of a few XPS photoemission lines of the N2

+-sputtered n-

GaN(0001)/L-cysteine interface was informative, but detailed analysis of all its XPS CL lines and

VB spectra is required to obtain valuable and instructive information on the bonding mechanisms

involved. Extending this research to functionalizing the N2

+-sputtered n-GaN(0001) surface with

small peptides might help to gain further insight on the interactions and bonds of amino acid

molecules on n-GaN(0001) surfaces, and hence the performance of electronic biosensing devices

integrating them. Further, the use of other tactics to modify the n-GaN(0001) surface such as

thermal annealing, sputtering with other dry gas-ions (e.g., H* plasma), or capping it with specific

acids may be useful for reliable binding with the peptides of various biological analytes overlaying

it.

109

6

Conclusions and Further Suggestions

The structure and chemistry of the surface of HVPE-grown n-GaN(0001) layers of c-sapphire/n-

GaN(0001) samples were explored before and after sputtering with low-energy (1-keV) nitrogen

(N2

+) ions and functionalizing them with de-ionized water (H2O) and neutral L-cysteine (C3H7NO2S)

molecules under various experimental conditions. The atomic force microscopy (AFM), low

energy electron diffraction (LEED), ultra-high vacuum (UHV) X-ray photoelectron spectroscopy

(XPS), and near-ambient-pressure NAP-XPS, using X-ray 𝐴𝐴𝐴𝐴 𝐾𝐾𝐾𝐾 radiation (ℎ𝜋𝜋 = 1486.7𝑒𝑒𝑉𝑉) as the

excitation source, have been exploited to do such investigation.

The obtained results yielded a good piece of information on the modification taking place

at the surface of n-GaN(0001) layers upon N2

+-sputtering, as well as on the contact formation and

interfacial chemical reactivity after its functionalization with H2O and L-cysteine. Thus, on the

ensuing interactions and bonding mechanisms between H2O and L-cysteine molecules and the

N2

+-sputtered n-GaN(0001) surface. This gave us a glue on the suitability, stability and

functionality of GaN-OH and GaN-SH groups, the primary link to various inorganic and organo-

/bio molecules, and thus on the potentiality of using GaN semiconductor in electronic devices and

biosensors whose performance are greatly affected by the past history, pre-treatment and

functionalization of the used GaN surface.

The UHV-XPS spectra of C 1s and O 1s core-level (CL) transitions of the as-grown n-

GaN(0001) surface revealed adsorbed adventitious carbon (C) and oxygen (O2) species, besides

traces of Ga2O3 and Ga2Ox oxides. The measured AFM images and LEED patterns of the surface

of as-grown n-GaN(0001) layers show that in-situ N2

+-ion sputtering had minimal surface damage

and roughness, in addition to largely lessen its carbon species. The NAP-XPS spectra of CL

transitions of the constituents of N2

+-sputtered n-GaN(0001)/H2O interface exemplified that part of

the H2O molecules of injected water vapor were adsorbed on the n-GaN(0001) surface and likely

dissociated into carboxyl (OH−) and proton (H+) ions, which tend to combine with surface gallium

(Ga+) dangling bonds and (N−) ions, respectively. This was clear from a detailed de-convolution

of the measured NAP-XPS spectra of the surface Ga 2p3/2 and Ga 3d5/2 CL transitions.

110

The N2

+-ion sputtering and H2O capping modified the valence band (VB) structure of n-

GaN(0001) surface as typified from respective UHV and NAP-XPS photoemission spectra. These

spectra display VB structural peaks related to hybridized orbital states with a dominating Ga-polar

surface, besides revealing a reduction in the band bending at the N2

+-sputtered n-GaN(0001)

surface. Functionalization of N2

+-sputtered n-GaN(0001) surface with H2O molecules induced a

noticeable modification to its chemistry and band bending, as mostly seen from the measured

NAP-XPS O 1s and Ga 3d photoemission lines and from the NAP-XPS valence band spectra of

the ensuing N2

+-sputtered n-GaN(0001)/H2O interface. Exposure of N2

+-sputtered n-GaN(0001)

surface to H2O molecules and annealing at elevated results in a decrease in the deficient oxygen

Ga2Ox oxide thru conversion to the more stable Ga2O3 oxide.

Solid evidence for the dissociation of H2O molecules adsorbed on the n-GaN(0001)

surface into OH– and H+ ions and their bonding to surface Ga+ and N– dangling bonds are not well

established from all literature studies, and no rigorous confirmation of the occurrence of

dissociation of H2O molecules adsorbed at the surface of N2

+-sputtered n-GaN(0001) investigated

in the present work. The formation of Ga-OH and N-H bonds, besides Ga-O bonds that become

more prominent at high temperatures are features that may be inferred from the curve-fittings of

the NAP-XPS peaks of existing surface constituents. Such compensation of dangling bonds and

surface states upon increasing the amount of H2O exposure exemplified itself as a continuous

decrease in the surface band bending voltage 𝑉𝑉bb up to 𝑝𝑝 = 0.1 mbar, above which 𝑉𝑉bb levelled

off at a constant value, depending on the sample’s temperature 𝑇𝑇.

The amount of surface band bending has been explicated by calculating the Vbb using two

methodologies. One approach utilized the energy difference (𝐸𝐸FS − 𝐸𝐸VS) between the surface

valence-band edge and the Fermi level (𝐸𝐸F), which is commonly called the VBM energy, by the

linear extrapolation method. The other approach utilized measurements of the peak energy of the

Ga 3dS CL line, relative to 𝐸𝐸F, of the N2

+-sputtered n-GaN(0001) surface or of the N2

+-sputtered

n-GaN(0001)/H2O interface-viz., (𝐸𝐸FS − 𝐸𝐸Ga3dS), combined with a proper value for the energy

between the VB edge and Ga 3dB CL of the GaN bulk, which is taken in this study to be 17.5 eV.

Both approaches incorporated the energy differences between the conduction band (CB) edge

(𝐸𝐸CB) and the Fermi energy 𝐸𝐸FB of the bulk GaN, namely (𝐸𝐸CB − 𝐸𝐸FB) and between 𝐸𝐸CB and 𝐸𝐸VB,

namely the bandgap energy 𝐸𝐸g. Both of 𝐸𝐸FB and 𝐸𝐸g are temperature dependent, which explains

partially the observed decrease in 𝑉𝑉bb with increasing sample’s temperature.

111

Due to the modification of the valence band related to surface conditions and the difficulty

of realizing proper linear extrapolation of the VB spectra of N2

+-sputtered n-GaN(0001)/H2O

interface, particularly at high H2O pressure, it is more feasible to evaluate 𝑉𝑉bb by using the binding

energy shifts of the Ga 3dS5/2 line (𝐸𝐸FS − 𝐸𝐸Ga3dS) than of the surface VB-edge 𝐸𝐸VS, the (𝐸𝐸FS − 𝐸𝐸VS)

or the VBM energy. The plausibility and validity of the former approach depend on the proper

choice of (𝐸𝐸VB − 𝐸𝐸Ga3dB). At a constant 𝑇𝑇, the decrease in the amount of band bending at the 𝑁𝑁2

+-

sputtered n-GaN(0001)/H2O interface with 𝑝𝑝 below 0.1 mbar can be ascribed to the compensation

of surface dangling bonds by possible dissociative H2O molecules and to formation of extra Ga2O3

oxide. The constancy of 𝑉𝑉bb for 𝑝𝑝 > 0.1 mbar can be understood in terms of a saturation coverage

of the n-GaN(0001) surface with H2O molecules. The reduction of low-𝑝𝑝 𝑉𝑉𝑏𝑏𝑏𝑏 with increasing 𝑇𝑇 can

be partly accounted for by the decrease in 𝐸𝐸g(𝑇𝑇)=(𝐸𝐸CB − 𝐸𝐸VB) and �𝐸𝐸CB − 𝐸𝐸FB(𝑇𝑇)� with 𝑇𝑇, though

such a decrease cannot be solely responsible for the entire variation of 𝑉𝑉bb with 𝑇𝑇, especially at

high temperatures.

The N2

+-sputtered n-GaN(0001) surface admits notable adhesion of the molecules of the

sulfur-containing amino-acid L-cysteine (L-CySH) as confirmed from the respective UHV-XPS

spectra. Deposition of L-CySH on the n-GaN(0001) surface was carried out in UHV by sublimation

of L-CySH powder at 105 oC for time periods ≤ 40 min. The UHV-XPS spectra were taken on the

N2

+-sputtered GaN(0001)/L-cysteine interface at room temperature and after annealing at various

temperatures (75-150 oC) for the times 15 minutes, 30 minutes, and 60 minutes.

The UHV-XPS spectra of N2

+-sputtered n-GaN(0001)/L-cysteine interface showed that the

thiol head group (SH group) of the L-cysteine molecule formed strong chemical bonding with the

surface metallic Ga dangling bonds. Such behavior was noted from the overlapped Ga 3s / S 2p

XPS photoemission lines, the intensity of which were enhanced with more deposited L-cysteine

monolayers, a feature that is characteristics for thiolate species. The bonding mechanism of

carboxyl (COOH) and amino (NH2) groups cannot be entirely excluded as was noted from the

respective C 1s, O 1s, and N 1s XPS spectra. It has been noticed that the strength and

appearance of doublet S 2p photoemission lines diminished with increasing temperature, at which

the n-GaN(0001)/L-cysteine sample was annealed, suggesting that most but not all of the L-

cysteine monolayers were re-evaporated from upper functionalized surface, with the SH-group of

the lowermost L-cysteine monolayers were still bonded chemically to gallium dangling bonds, as

is illustrated in the schematic A1 and A2 figures of Appendix A.

112

Bibliography

1 M. J. Madou and S. Roy Morisson, Chemical Sensing with Solid State Devices. Academic Press,

Inc., Boston, London, 1989.

2 R. J. Hamers, Formation and Characterization of Organic Monolayers on Semiconductor Surfaces,

Annul. Rev. Anal. Chem., vol. 1, pp. 707-736, 2008.

3 F. Tao, S. L. Bernasek (eds), Functionalization of Semiconductor Surfaces, John Wiley & Sons, Inc.,

Hoboken, NJ, 2012.

4 E. H. Williams, Silicon, Silicon carbide, and Gallium Nitride Nanowire biosensors, PhD Thesis,

George Mason University, Fairfax, VA, USA, 2014.

5 D. C. Gleason-Rohrer, Measurement of the band bending and surface dipole at chemically

functionalized Si(111)/vacuum interfaces, PhD Thesis, California Institute of Technology, Pasadena,

California, USA, 2014.

6 F. R. Rahsepar, N. Moghimi, K. T. Leung, Surface-Mediated Hydrogen Bonding of Proteinogenic α-

Amino Acids on Silicon. Accounts of Chemical Research., vol. 49, pp. 942-951, 2016.

7 B. J. Baliga, Fundamentals of Power Semiconductor Devices, Springer International Publishing,

Switzerland, 2nd Edn. 2019.

8 Y. Xu, X. Hu, S. Kundu, A. Nag, N. Afsarimanesh, S. Sapra, S. C. Mukhopadhyay and T. Han, Silicon

based Sensors for Biomedical Applications: A Review, Sensors vol. 19, pp. 2908: 1-22, 2019.

9 H. Morkoç and S. N. Mohammad, “High-Luminosity Blue and Blue-Green Gallium Nitride Light-

Emitting Diodes,” Science, vol. 267, pp. 51-55, 1995.

10 S. N. Mohammad, A. A. Salvador, and H. Morkoç, “Emerging Gallium Nitride Based Devices,” Proc.

IEEE, vol. 83, No. 10, pp. 1306-1355, 1995.

11 S. J. Pearton and C. Kuo, “GaN and Related Materials for Device Applications,” Mater. Res. Soc.

Bull., vol. 22, No. 2, pp. 17-21, 1997.

12 M. S. Shur, GaN Based Transistors for High Power Applications, Solid-State Electronics, vol. 42,

No. 12, pp. 2131-2138, 1998.

13 S.-H. Wei, III-nitride semiconductors for photovoltaic application. NCPV and Solar Program Review

Meeting 2003, 1st Edn., 2003.

14 B. Gil (Ed.), III-Nitride Semiconductors and their Modern Devices. Oxford University Press, Oxford,

2013.

15 B. Gil (Auth.), Physics of Wurtzite Nitrides and Oxides: Passport to Devices, Springer International

Publishing, Switzerland, 2014.

16 Z. Mi and C. Jagadish (Eds.), III-Nitride Semiconductor Optoelectronics. Academic Press, Elsevier

Inc., London, Oxford, 2017.

17 R. Bouveyron, Development and understanding of III-N layers for the improvement of high power

transistors. PhD Thesis, Université Grenoble Alpes, 2018.

18 D. K. Saini, Gallium Nitride: Analysis of Physical Properties and Performance in High-Frequency

Power Electronic Circuits. MSc Thesis, Wright State University, 2015.

19 M. Stutzmann, G. Steinhoff, M. Eickhoff, O. Ambacher, C. E. Nebel, J. Schalwig, R. Neuberger, and

G. Müller, “GaN-based heterostructures for sensor applications”, Diamond and Related Materials,

vol. 11, No. 3-6, pp. 886-891, 2002.

20 S. Mehlhose, Biofunctionalization of GaN/AlGaN/GaN High Electron Mobility Transistors, PhD

Thesis, Heidelberg University, Germany, 2019.

113

21 C. Gergely, Nitride devices and their biofunctionalization for biosensing applications. Chapter 14,

pp. 505-518 in III-Nitride Semiconductors and their Modern Devices, edited by B. Gil, Oxford

University Press, Oxford, 2013.

22 F. Meierhofer, L. Krieg, and T. Voss, “GaN meets organic: technologies and devices based on

gallium-nitride/organic hybrid structures: Topical Review,” Semiconductor Science and Technology,

vol. 33, No. 8, pp. 083001: 1-30, 2018.

23 P. Bertani and W. Lu, “Hybrid Organic–Nitride Semiconductor Nanostructures for Biosensor

Applications,” chapter 13 in Functional Organic and Hybrid Nanostructured Materials: Fabrication,

Properties, and Applications, Edited by Q. Li. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim,

Germany, 2018.

24 Z. Oussama, B. Moussaab, and O. Amaria, The Role of Nitride Materials for Biological Applications

(Biosensors), Sch. Acad. J. Biosci., vol. 6, No 2, pp. 151-157, 2018.

25 S. J. Pearton, B. S. Kang, S. Kim, F. Ren, B. P. Gila, C. R. Abernathy, J. Lin, and S. N. G. Chu,

GaN-based diodes and transistors for chemical, gas, biological and pressure sensing: topical review,

J. Phys.: Condensed Matter, vol. 16, pp. R961–R994, 2004.

26 S. J. Pearton and F. Ren, Gallium Nitride-Based Gas, Chemical and Biomedical Sensors. IEEE

Instrumentation & Measurement Magazine, vol. 15, No. 1, pp. 16-21, 2012.

27 S. J. Pearton, F. Ren, and B. H. Chu. Functionalized GaN Based Transistors for Biosensing. In State

of the Art in Biosensors-Environmental and Medical Applications, T. Rinken (Editor), Chapter 11,

pp. 225-251, 2013.

28 Ι. N. Cimalla, AlGaN/GaN sensors for direct monitoring of fluids and bioreactions. PhD Thesis, TU

Ilmenau, 2011.

29 C. M. Foster, Fabrication and characterization of Gallium Nitride Biointerfaces. MSc Thesis, North

Carolina State University, Raleigh, North Carolina, 2013.

30 M. S. Makowski, Surface functionalization of gallium nitride for biosensor platform development.

PhD Thesis, Purdue University, 2013.

31 J. D. Howgate, GaN Heterostructures for Biosensing and Radiation Detection. PhD Thesis, TU

München, 2012.

32 S. Strite, M. E. Lin, and H. Morkoç, Progress and prospects for GaN and the III-V nitride

semiconductors, Thin Solid Films, vol. 231, pp. 197-210, 1993.

33 F. L. M. Khir, M. Myers, A. Podolska, T. M. Sanders, B. Murray, N. Brett, and P. Giacinta,

Synchrotron-based XPS studies of AlGaN and GaN surface chemistry and its relationship to ion

sensor behavior, Appl. Surface Sci., vol. 314, pp. 850-857, 2014.

34 B. A. K. Baur, Functionalization of Group III-Nitrides for Biosensor Applications. PhD Thesis, TU

München, 2011.

35 X. Wen, “Design, Fabrication, and Characterization of Field-Effect and Impedance Based

Biosensors,” PhD Dissertation, The Ohio State University, Columbus, OH, USA, 2011.

36 B. S. Kang, H. T. Wang, F. Ren, and S. J. Pearton, Electrical detection of biomaterials using

AlGaN/GaN high electron mobility transistors, Appl. Phys. Rev. vol. 104, p 031101: 1 – 11, 2008.

37 G. Steinhoff, Group III-Nitrides for Bio-and Electrochemical Sensors, PhD Thesis, TU München,

Germany, 2008.

38 L. E. Bain, S. A. Jewett, A. H. Mukund, S. M. Bedair, T. M. Paskova and A. Ivanisevic. Biomolecular

Gradients via Semiconductor Gradients: Characterization of Amino Acid Adsorption to InxGa1−xN

surfaces, ACS Appl. Mater. Interfaces vol. 5, pp. 7236-7243, 2013.

39 J. H. Edgar (Ed), Properties of Group III nitrides, INSPEC, London, 1994.

114

40 O. Ambacher, “Growth and applications of Group III—nitrides: Review article,” J. Phys. D: Appl.

Phys., vol. 31, pp. 2653-2710, 1998.

41 H. Morkoç, Handbook of Nitride Semiconductors and Devices: Materials Properties, Physics and

Growth, vol. 1, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2009.

42 Z. C. Feng (Ed), III-nitride: Semiconductor Materials, Imperial College Press, London, 2006.

43 S. Alamé, Surface Properties of Clean and Adsorbate-Terminated In(Ga)N layers, Diploma Thesis,

TU Berlin, 2012.

44 D. Skuridina, Structure and preparation of polar and semipolar InN surfaces, PhD Thesis, TU Berlin,

2014.

45 A. K. Winnerl, Electronic Processes at the GaN Surface. PhD Thesis, TU München, 2017.

46 V. Perumal, U. Hashim, Advances in biosensors: Principle, architecture and applications, J. Appl.

Biomedicine vol. 12, pp. 1-15, 2014.

47 N. Bhalla, P. Jolly, N. Formisano and P. Estrela, Introduction to biosensors, Essays in Biochemistry

vol. 60, pp. 1-8, 2016.

48 S. Alamé, Eigenschaften ultradünner Galliumnitridund-Oxidschichten über InGaN-Quantenfilmen für

sensorische Anwendungen (Properties of ultrathin gallium nitride and oxide films over InGaN

quantum wells for sensory applications), PhD Thesis, TU Berlin, 2019.

49 E. H. Williams, A. V. Davydov, V. P. Oleshko, K. L. Steffens, I. Levin, N. J. Lin, K. A. Bertness, A. K.

Manocchi, J. A. Schreifels, and M. V. Rao, Solution-based functionalization of gallium nitride

nanowires for protein sensor development,” Surface Science, vol. 627, pp. 23-28, 2014.

50a G. G. C. Arízaga, M. J. Oviedo, and O. E. C. López, “Electrical properties of polycrystalline GaN films

functionalized with cysteine and stabilization of GaN nanoparticles in aqueous media,” Colloids and

Surfaces B: Biointerfaces, vol. 98, p. 63-71, 2012.

50b T. Yamada, J. Ito, R. Asahara, K. Watanabe, M. Nozaki, S. Nakazawa, Y. Anda, M. Ishida, T. Ueda, A.

Yoshigoe, T. Hosoi, T. Shimura, and H. Watanabe, “Comprehensive study on initial thermal oxidation of

GaN(0001) surface and subsequent oxide growth in dry oxygen ambient,” J. Appl. Phys., vol. 121, p.

035303: 1-9, 2017.

50c T. Yamada, J. Ito, R. Asahara, K. Watanabe, M. Nozaki, T. Hosoi, T. Shimura, and H. Watanabe,

“Improved interface properties of GaN-based metal-oxide-semiconductor devices with thin Ga-oxide

interlayers,” Appl. Phys. Lett., vol. 110, pp. 261603: 1-5, 2017.

51 Y-T. Lai, A. Ganguly, Li-C. Chen, K-H. Chen, Direct voltammetric sensing of l-Cysteine at pristine

GaN nanowires electrode, Biosensors and bioelectronics, vol. 26, pp. 1688-1691, 2010.

52 A. Winnerl, J. A. Garrido, and M. Stutzmann, Electrochemical characterization of GaN surface

states, J. Appl. Phys. 122, 045302, 2017.

53 K. Hinrichs, M. Gensch, N. Esser, U. Schade, J. Rappich, S. Kröning, M. Portwich, and R. Volkmer,

“Analysis of biosensors by chemically specific optical techniques. Chemiluminescence-imaging and

infrared spectroscopic mapping ellipsometry,” Anal. Bioanalytical Chem., vol. 387, No. 5, pp. 1823-

1829, 2007.

54 M. Badea, L. Floroian, P. Restani, M. Moga, Simple Surface Functionalization Strategy for

Immunosensing Detection of Aflatoxin B1, Int. J. Electrochem. Sci. vol. 11, pp. 6719-6734, 2016.

55 B. S. Kang, F. Ren, L. Wang, C. Lofton, W. W. Tan, S. J. Pearton, A. Dabiran, A. Osinsky and P. P.

Chow, Electrical detection of immobilized proteins with ungated AlGaN/GaN high-electron-mobility

transistors, Appl. Phys. Rev. vol. 87, p 023508: 1 – 3, 2005.

56 J. Kochana, K. Starzec, M. Wieczorek, P. Knihnicki, M. Góra, A. Rokicińska, P. Kościelniak and P.

Kuśtrowski, Study on self-assembled monolayer of functionalized thiol on gold electrode forming

115

capacitive sensor for chromium(VI) determination, J. Solid State Electrochem. Vol. 23, pp 1463-

1472, 2019.

57 A. Ganguly, C-P. Chen, Y-T. Lai, Li-C. Chen, C-C. Kuo, C-W. Hsu, K-H. Chen, and L-C. Chen.

Functionalized GaN nanowire-based electrode for direct label-free voltammetric detection of DNA

hybridization, J. Mater. Chem. vol. 19, pp. 928-933, 2009.

58 I-H. Cho, D. H. Kim, and S. Park. Electrochemical biosensors: perspective on functional

nanomaterials for on-site analysis, Biomaterials Res. vol. 24, No. 6, pp. 1-12, 2020.

59 V. Lazarescu, M. Enache, M. Anastasescu, G. Dobrescu, C. Negrila, M. F. Lazarescu.

Electrochemical impedance spectroscopy investigations on the L-cysteine-thiolate self-assembled

monolayers formed at p-GaAs(100) electrodes, Electrochimica Acta vol. 131, pp. 42-51, 2014.

60 P. H. Holloway, G. E. McGuire, Handbook of compound semiconductors: Growth, Processing,

characterization, and Devices, Noyes Publications, New Jersey, USA, 1995.

61 E. Estephan, C. Larroque, F. J. G. Cuisinier, Z. Bálint, and C. Gergely, “Tailoring GaN

Semiconductor Surfaces with Biomolecules,” J. Phys. Chem. B, vol. 112, pp. 8799–8805, 2008.

62 V. M. Bermudez, “The fundamental surface science of Wurtzite gallium nitride,” Surface Science Reports,

vol. 72, pp. 147-315, 2017.

63 S. A. Jewett, M. S. Makowski, B. Andrews, M. J. Manfra, and A. Ivanisevic, “Gallium nitride is

biocompatible and non-toxic before and after functionalization with peptides: Brief communication,”

Acta Biomaterialia, vol. 8, pp. 728–733, 2012.

64 N. G. Berg, M. W. Nolan, T. Paskova, and A. Ivanisevic, Surface Characterization of Gallium Nitride

Modified with Peptides Before and After Exposure to Ionizing Radiation in Solution, Langmuir, vol.

30, No. 51, pp. 15477-15485, 2014.

65 H. Kim, P. E. Colavita, P. Paopraset, PadmaGopalan, T. F. Kuech, and R. J. Hamers. Grafting of

molecular layers to oxidized gallium nitride surfaces via phosphonic acid linkages, Surface Science,

vol. 602, pp. 2382-2388, 2008.

66 R. Stine, B. S. Simpkins, S. P. Mulvaney, L. J. Whitman, C. R. Tamanah, Formation of amine groups

on the surface of GaN: A method for direct biofunctionalization, Appl. Surf. Sci. vol. 256, pp. 4171-

4175, 2010.

67 S. J. Wilkins, M. Greenough, C. Arellano, T. Paskova and A. Ivanisevic, In Situ Chemical

Functionalization of Gallium Nitride with Phosphonic Acid Derivatives during Etching, Langmuir vol.

30, pp. 2038−2046, 2014.

68 M. S. Makowski, D. Y. Zemlyanov, J. A. Lindsey, J. C. Bernhard, E. M. Hagen, B. K. Chan, A. A.

Petersohn, M. R. Medow, L. E. Wendel, D. Chen, J. M. Canter, A. Ivanisevic. Covalent attachment

of a peptide to the surface of gallium nitride, Surf. Sci. vol. 605, pp. 1466-1475, 2011.

69a M. Hofstetter, J. Howgate, M. Schmid, S. Schoell, M. Sachsenhauser, D. Adigüzel, M. Stutzmann,

I. D. Sharp, and S. Thalhammer, “In vitro bio-functionality of gallium nitride sensors for radiation

biophysics,” Biochem. Biophys. Res. Commun. vol. 424, pp. 348-353, 2012.

69b X. Du, Z. Guo, Y. Meng, L. Zhao, X. Li, R. Feng, W. Zhao, and H. Zhong, “Effects of surface

properties of GaN semiconductors on cell behavior,” Heliyon, vol. 9, p e18150: 1-11, 2023.

70 K. M. Tracy, W. J. Mecouch, R. F. Davis and R. J. Nemanich, “Preparation and characterization of

atomically clean, stoichiometric surfaces of n- and p-type GaN(0001),” J. Appl. Phys., vol. 94, pp.

3163-3172, 2003.

71 S. W. King, J. P. Barnak, M. D. Bremser, K. M. Tracy, C. Ronning, R. F. Davis, and R. J. Nemanich,

“Cleaning of AlN and GaN surfaces,” J. Appl. Phys., vol. 84, pp. 5248-5260,1998.

72 B. L. Pearce, S. J. Wilkins, T. Paskova, and A. Ivanisevic. A review of in situ surface functionalization

of gallium nitride via beaker wet chemistry, J. Mater. Res., vol. 30, No. 19, pp. 2859-2870, 2015.

116

73 M. Mishra, T. C. S. Krishna, N. Aggarwal, and G. Gupta, “Surface chemistry and electronic structure

of nonpolar and polar GaN films,” Appl. Surf. Sci., vol. 345, pp. 440-447, 2015.

74 M. Mishra, T. C. Krishna, P. Rastogi, N. Aggarwal, A. K. S. Chauhan, L. Goswami, and G. Gupta,

"New approach to clean GaN surfaces," Materials Focus, vol. 3, No. 3, pp. 218-223, 2014.

75 M. Mishra, A. Gundimeda, S. Krishna, N. Aggarwal, B. Gahtori, N. Dilawar, V. V. Aggarwal, M. Singh,

R. Rakshit, and G. Gupta, "Wet chemical etching induced stress relaxed nanostructures on polar &

non-polar epitaxial GaN films," Phys. Chem. Chem. Phys., vol. 19, No. 13, pp. 8787-8801, 2017.

76 F. Machuca, Z. Liu, Y. Sun, P. Pianetta, W. E. Spicer, and R. F. W. Pease, "Simple method for

cleaning gallium nitride (0001)," J. Vac. Sci. Technol. A: Vacuum, Surfaces, and Films, vol. 20, No.

5, pp. 1784-1786, 2002.

77 L. L. Smith, S. W. King, R. J. Nemanich, and R. F. Davis, “Cleaning of GaN Surfaces,” J. Electron

Mater., vol. 25, pp. 805-810, 1996.

78 K. Sardar, R. Raju, and G. N. Subbanna, “Epitaxial GaN films deposited on sapphire substrates

prepared by the sol-gel Method,” Solid State communications, vol. 125, pp. 355-358, 2003.

79 C. Y. Fong, S. S. Ng, F. K. Yam, H. A. Hassan, and Z. Hassan, “An investigation of sol–gel spin

coating growth of wurtzite GaN thin film on 6H–SiC substrate,” J. Crystal Growth, vol. 413, pp. 1-4,

2015.

80 C. Y. Fong, S. S. Ng, N. M. Amin, F. K. Yam, and Z. Hassan, “Sol-gel-derived GaN thin films for

ultraviolet photodetection,” Microelectronics Inter., vol. 36, pp. 8-13, 2019.

81 C. G. Zhang, L. F. Bian, W. D. Chen, and C. C. Hsu, “Effect of growth conditions on the GaN thin

film by sputtering deposition,” J. Crystalline Growth, vol. 299, pp. 268-271, 2007.

82a A. Dolgov, D. Lopaev, C. J. Lee, E. Zoethout, V. Medvedev, O. Yakushev, and F. Bijkerk,

“Characterization of carbon contamination under ion and hot atom bombardment in a tin-plasma

extreme ultraviolet light source,” Appl. Surf. Sci., vol. 353, pp. 708-713, 2015.

82b A. N. Hattori, K. Endo, K. Hattori, and H. Daimon, “Surface treatments toward obtaining clean GaN

(0001) from commercial hydride vapor phase epitaxy and metal-organic chemical vapor deposition

substrates in ultrahigh vacuum,” Appl. Surf. Sci., vol. 256, No.14, pp. 4745-4756, 2010.

82c Ying-Huang Lai, Chuin-Tih Yeh, Jun-Min Hwang, Huey-Liang Hwang, Chien-Te Chen, and Wei-

Hsiu Hung, “Sputtering and Etching of GaN surface,” J. Phys. Chem. B, vol. 105, pp. 10029-10036,

2001.

83 F. A. Ponce, “Defects and Interfaces in GaN Epitaxy,” Mater. Res. Soc. Bull., vol. 22, No. 2, pp. 51-

57, 1997.

84 C. M. Foster, R. Collazo, Z. Sitar and A. Ivanisevic. Aqueous Stability of Ga-and N-Polar Gallium

Nitride, Langmuir vol. 29, pp. 216-220, 2013.

85 X. Zhang and S. Ptasinska, “Electronic and chemical structure of the H2O/GaN (0001) interface

under ambient conditions,” Scientific Reports, 24848: 1-6, 2016.

86 P. Lorenz, R. Gutt, T. Haensel, M. Himmerlich, J. A. Schaefer, and S. Krischok, “Interaction of GaN

(0001)-2×2 surfaces with H2O,” Phys. Status Solidi C, vol. 7, pp. 169-172, 2010.

87 A. V. Akimov, J. T. Muckerman, and V. O. Prezhdo, “Nonadiabatic Dynamics of Positive Charge

during Photocatalytic Water Splitting on GaN(10-10) Surface: Charge Localization Governs Splitting

Efficiency,” J. Am. Chem. Soc., vol. 135, pp. 8682-8691, 2013.

88 M. G. Kibria, S. Zhao, F. A. Chowdhury, Q. Wang, H. P. T. Nguyen, M. L. Trudeau, H. Guo,

and Z. Mi, “Tuning the surface Fermi level on p-type gallium nitride nanowires for efficient overall

water splitting,” Nature Comm., vol. 5, 3825, pp. 1-6, 2014.

117

89 J. Wang, L. S. Pedroza, A. Poissier, and M. V. Fernández-Serra, “Water Dissociation at the GaN

(1010) Surface: Structure, Dynamics and Surface Acidity,” J. Phys. Chem. C, vol. 116, No. 27, pp.

14382–14389, 2012.

90 X. Shen, Y. A. Small, J. Wang, P. B. Allen, M. V. Fernandez-Serra, M. S. Hybertsen, and J. T.

Muckerman, “Photocatalytic Water Oxidation at the GaN (1010)-Water Interface,” J. Phys. Chem.

C, vol. 114, pp. 13695-13704, 2010.

91 H. Ye, G. Chen, H. Niu, Y. Zhu, Li Shao, and Z. Qiao, “Structures and Mechanisms of Water

Adsorption on ZnO (0001) and GaN (0001) Surface,” J. Phys. Chem. C, vol. 117, pp. 15976-15983,

2013.

92 X. Shen, P. B. Allen, M. S. Hybertsen, and J. T. Muckerman, "Water adsorption on the GaN (101

�0)

nonpolar surface," J. Phys. Chem. C, vol. 113, No. 9, pp. 3365-3368, 2009.

93 H. Wang, H. Zhang, J. Liu, D. Xue, H. Liang, and X. Xia. Hydroxyl Group Adsorption on GaN (0001)

Surface: First Principles and XPS Studies, J. Electron. Mater. Vol. 48, pp. 2430-2437, 2019.

94 V. M. Bermudez, “Study of oxygen chemisorption on the GaN (0001)-(1×1) surface,” J. Appl. Phys.,

vol. 80, pp. 1190-1200, 1996.

95 V. M. Bermudez and J. P. Long, “Chemisorption of H2O on GaN(0001),” Surf. Sci., vol. 450, pp. 98-

105, 2000.

96 V. M. Bermudez, “Investigation of the initial chemisorption and reaction of fluorine (XeF2) with the

GaN(0001)-(1x1) surface,” Appl. Surf. Sci., vol. 119, pp. 147-159, 1997.

97 T. Ito, S. M. Forman, C. Cao, F. Li, C. R. Eddy, Jr., M. A. Mastro, R. T. Holm, R. L. Henry, K. L.

Hohn, and J. H. Edgar. Self-Assembled Monolayers of Alkylphosphonic Acid on GaN Substrates,

Langmuir vol. 24, pp. 6630-6635, 2008.

98 V. M. Bermudez, D. D. Koleske, A. E. Wickenden, “The dependence of the structure and electronic

properties of wurtzite GaN surfaces on the method of preparation,” Appl. Surf. Sci., vol. 126, pp. 69-

82, 1998.

99 H. Tang, Z. Q. Fang, S. Rolfe, J. A. Bardwell, and S. Raymond, "Growth kinetics and electronic

properties of unintentionally doped semi-insulating GaN on SiC and high-resistivity GaN on sapphire

grown by ammonia molecular-beam epitaxy," J. Appl. Phys., vol. 107, No. 10, pp. 103701: 1-12,

2010.

100 J. Ma, B. Garni, N. Perkins, W. L. O’Brien, T. F. Kuech, and M. G. Lagally, Photoemission

spectroscopy studies of the surface of GaN films grown by vapor phase Epitaxy, Citation: Appl.

Phys. Lett. 69, 3351-3353, 1996.

101 M Hartensveld, Optimization of Dry and Wet GaN Etching to Form High Aspect Ratio Nanowires,

MSc Thesis, Rochester Institute of Technology, Rochester, NY, USA, 2018.

102 S. J. Pearton, R. J. Shul and Fan Ren, A Review of Dry Etching of GaN and Related Materials, MRS

Internet Journal of Nitride Semiconductor Research, vol. 5, 11: 1-38, 2000.

103 Y-H. Yeh, K-M. Chen, Y-H. Wu, Y-C. Hsu, and W-I. Lee, "Hydrogen etching on the surface of GaN

for producing patterned structures," J. Crystal Growth, vol. 314, No. 1, pp. 9-12, 2011.

104 Y-H. Yeh, Y-C. Hsu, Y-H. Wu, K-M. Chen, and W-I. Lee, "Hydrogen etch of GaN and its application

to produce free-standing GaN thick films,” J. crystal Growth, vol. 333, pp. 16-19, 2011.

105 Y-H. Yeh, C-M. Chu, Y-H. Wu, Y-C. Hsu, T-Y. Yu, and W-I. Lee, "Use of hydrogen etching to remove

existing dislocations in GaN epitaxial layers," Semicond. Sci. Technol., vol. 30, No. 8, pp. 085002:

1-6, 2015.

106 P. Weightman, D. S. Martin, R. J. Cole, and T. Farrell, “Reflection anisotropy spectroscopy,” Reports

Progress Phys., vol. 68, pp. 1251-1341, 2005.

118

107 S. Bloom, G. Harbeke, E. Meier, and I. B. Ortenburger, “Band Structure and Reflectivity of GaN,”

Phys. Stat. Sol. (b), vol. 66, pp. 161-168, 1974.

108 R. Gargaliano, Characterization of L-cysteine thin films via photoemission spectroscopy. MSc

Thesis, University of South Florida, 2005.

109 D. S. Martin and P. Weightman, “Reflection anisotropy spectroscopy: a new probe of metal surfaces,”

Surface and Interface Analysis, vol. 31, pp. 915-926, 2001.

110 F. R. Rahsepar, L. Zhang, H. Farkhondeh and K. T. Leung, Biofunctionalization of Si(111)7x7 by

“Renewable” L-Cysteine Translational Layer, J. Am. Chem. Soc. vol. 136, pp. 16909-16918, 2014.

111 C. I. Wu, A. Kahn, N. Taskar, D. Dorman and D. Gallagher, GaN (0001)-(1x1) surfaces: Composition

and electronic properties, J. Appl. Phys. vol. 83, No. 8, pp. 4249-4252, 1998.

112 J. Ma, B. Gami, N. Perkins, W. L. O’Brien, T. F. Kuech, and M. G. Lagally, “Photoemission

spectroscopy studies of the surface of GaN films grown by vapor phase epitaxy,” Appl. Phys. Lett.,

vol. 69, pp. 3351-3353, 1996.

113 R. A. Beach, E. C. Piquette, and T. C. McGill, XPS Study of Oxygen Adsorption on (3x3)

Reconstructed MBE Grown GaN Surfaces, MRS Internet Journal of Nitride Semiconductor

Research, vol. 4S1, G6.26, pp. 648-652, 1999.

114 H. W. Jang, K. W. Ihm, T-H. Kang, J-H. Lee and J-L. Lee, Polarization-induced surface band

bendings of GaN films studied by synchrotron radiation photoemission spectroscopy, Phys. Status

Solid b vol. 240, No. 2, pp. 451-454, 2003.

115 A. Jürgensen, N. Heutz, H. Raschke, K. Merz, and R. Hergenröder, "Behavior of supported

palladium oxide nanoparticles under reaction conditions, studied with near ambient pressure

XPS," Anal. Chem., vol. 87, No. 15, pp. 7848-7856, 2015.

116 A. Jürgensen, N. Esser, and R. Hergenröder, “Near ambient pressure XPS with a conventional X-

ray source,” Surf. Inter. Anal., vol. 44, No. 8, pp. 1100-1103, 2012.

117 D. Li, M. Sumiya, S. Fuke, D. Yang, D. Que, Y. Suzuki and Y. Fukuda, Selective etching of GaN

polar surface in potassium hydroxide solution studied by x-ray photoelectron spectroscopy, J. Appl.

Phys. vol. 90, No. 8, 2001.

118 A. Winnerl, R. N. Pereira, and M. Stutzmann, Photo-induced changes of the surface band bending

in GaN: Influence of growth technique, doping and polarity, J. Appl. Phys. 121, 205307, 2017.

119 J. L. Rouviere, J. L. Weyher, M. S-E. and S. Porowski. Polarity determination for GaN films grown

on (0001) sapphire and high-pressure-grown GaN single crystals, Appl. Phys. Let. Vol. 73, No. 5,

pp. 668-670, 1998.

120 S. S. Kushvaha, M. S. Kumar, M. Maheshwari, A. K. Shukla, P. Pal, and K. K. Maurya, “Structural

and electronic properties of epitaxial GaN layer grown on sapphire (0001) using laser molecular

beam epitaxy,” Mater. Res. Express, vol. 1, No. 3, p. 035903: 1-9, 2014.

121 D. Skuridina, D. V. Dinh, B. Lacroix, P. Ruterana, M. Hoffmann, Z. Sitar, M. Pristovsek, M. Kneissl,

and P. Vogt, “Polarity determination of polar and semipolar (112

�2) InN and GaN layers by valence

band photoemission spectroscopy,” J. Appl. Phys., vol. 114, p. 173503, 2013.

122 T. D. Veal, P. D. King, P. H. Jefferson, L. F. J. Piper, C. F. McConville, H. Lu, W. J. Schaff, P. A.

Anderson, S. M. Durbin, D. Muto, H. Naoi and Y. Nanishi, In adlayers on c-plane InN surfaces: A

polarity-dependent study by x-ray photoemission spectroscopy, Phys. Rev. B vol. 76, pp. 075313:

1-8, 2007.

123 C. Hamaguchi, Basic Semiconductor Physics, 3rd edition, Springer-Verlag, Berlin, Germany, 2017.

124 K. B. Nam, J. Li, J. Y. Lin, and H. X. Jiang, Optical properties of AlN and GaN in elevated temperatures,

Appl. Phys. Lett. vol. 85, No. 16, pp. 3489-3491, 2004.

119

125 J. Tersoff, Schottky barriers and semiconductor band structures, Phys. Rev. B vol. 32, No. 10, pp.

6968-6971, 1985.

126 W. Mönch, Branch-point energies and the band-structure lineup at Schottky contacts and

heterostructures, J. Appl. Phys. vol. 109, pp. 113724: 1-10, 2011.

127 J. Tersoff, Theory of semiconductor heterojunctions: The role of quantum dipoles, Phys. Rev. B vol.

30, No. 8, pp. 4874-4877, 1984.

128 W. Mönch, Electronic Properties of Semiconductor Interfaces, Springer-Verlag, Berlin, 2004.

129 A. Schleife, F. Fuchs, C. Rödl, J. Furthmüller and F. Bechstedt, Branch-point energies and band

discontinuities of III-nitrides and III-/II-oxides from quasiparticle band-structure calculations, Appl.

Phys. Lett. vol. 94, pp. 012104: 1-3, 2009.

130 B. S. Eller, J. Yang, and R. J. Nemanich, “Polarization Effects of GaN and AlGaN: Polarization Bound

Charge, Band Bending, and Electronic Surface States,” J. Electron. Mater., vol. 43, pp. 4560-4568,

2014.

131 B. S. Eller, J. Yang, and R. J. Nemanich, “Surface band bending and interface alignment of plasma-

enhanced atomic layer deposited SiO2 on AlxGa1-xN,” J. Appl. Phys., vol. 122, pp. 125304: 1-6,

2017.

132 M. A. Foussekis, Investigation of Band Bending in n-and p-Type GaN, PhD Thesis, Virginia

Commonwealth University, Richmond, Virginia, USA, 2012.

133 I. Vurgaftman, J. R. Meyer, and L. R. Ram-Mohan, Band parameters for III–V compound

semiconductors and their alloys, J. Appl. Phys. vol. 89, No. 11, pp. 5815-5875, 2001 (and the

numerous references therein).

134 T. D. Veal, P. H. Jefferson, L. F. J. Piper, C. F. McConville, T. B. Joyce, P. R. Chalker, L. Considine,

Hai Lu, and W. J. Schaff, "Transition from electron accumulation to depletion at InGaN

surfaces," Appl. Phys. Lett., vol. 89, No. 20, pp. 202110: 1-3, 2006.

135 K. Z. Liu, M. Shimomura and Y. Fukuda, Band Bending of n-GaP(001) and p-InP(001) Surfaces with

and without Sulfur Treatment Studied by Photoemission (PES) and Inverse Photoemission

Spectroscopy (IPES), Adv. Mater. Res. vol. 222, pp. 56-61, 2011.

136 J. D. Hartman, Characterization of the Growth of Aluminum Nitride and Gallium Nitride Thin Films

on Hydrogen etched and/or Cleaned 6H-SiC(0001) Surfaces, PhD Thesis, North Carolina State

University, Raleigh, North Carolina, USA, 2000.

137 K. S. Kim, C. S. Oh, K. J. Lee, G. M. Yang, C-H. Hong, K. Y. Lim, H. J. Lee and A. Yoshikawa, Effect

of growth rate of a GaN buffer layer on the properties of GaN on a sapphire substrate, J. Appl. Phys.

vol. 85, pp. 8441-8444, 1999.

138 D. F. Ogletree, H. Bluhm, E. D. Hebenstreit, and M. Salmeron, Photoelectron spectroscopy under

ambient pressure and temperature conditions, Nuclear Instrum. Methods Phys. Res. A vol. 601, pp.

151-160, 2009.

139 D. I. Patel, T. Roychowdhury, V. Jain, D. Shah, T. G. Avval, S. Chatterjee, S. Bahr, P. Dietrich, M.

Meyer, A. Thiβen, and M. R. Linford, Introduction to near-ambient pressure x-ray photoelectron

spectroscopy characterization of various materials, Surface Sci. Spectra vol. 26, pp. 016801: 1-3,

2019.

140 M. Salmeron and R. Schlögl, “Ambient Pressure Photoelectron Spectroscopy: A new tool for the surface

science and nanotechnology,” http://escholarship.or/uc/item/6n75v256, 2008.

141 D. E. Starr, Z. Liu, M. Hävecker, A. Knop-Gericke, and H. Bluhm, “Investigation of solid/vapor interfaces

using ambient pressure X-ray photoelectron spectroscopy,” Chem. Soc. Rev., vol. 42, pp. 5833-5857,

2013.

142 J. F. Watts and J. Wolstenholme, An Introduction to Surface Analysis by XPS and AES, John Wiley &

Sons Ltd., New York, 2003.

120

143 F. Reinert and S. Hüfner, “Photoemission Spectroscopy-From Early days to Recent Applications,” New

Journal of Physics, vol. 7, pp. 97: 1-34, 2005.

144a S. Hüfner, Photoelectron Spectroscopy-Principles and Applications, Springer-Verlag, Berlin, 2003.

144b S. Hofman, Auger-and- X-ray Phptoelectron Spectroscopy in Materials Science: A user-Oriented Guide,

Springer-Verlag, Berlin, 2013.

145 S. Suga and A. Sekiyama, Photoelectron Spectroscopy, Springer-Verlag, Berlin Heidelberg, 2014.

146 P. Van der Heide, X-Ray Photoelectron Spectroscopy: An Introduction to Principles and Practices,

John Wiley & Sons, Inc., New Jersey, 2012.

147 D. Briggs and P. Seah, Practical surface analysis: by auger and x-ray photoelectron spectroscopy.

Wiley, 1983.

148 D. Vij, Handbook of applied solid state spectroscopy. Springer, 2010.

149 J. J. Yeh and I. Lindau, Atomic subshell photoionization cross sections and asymmetry parameters:

1≤ Z ≤ 103, Atomic Data and Nuclear Data Tables vol. 32, pp.1-155, 1985.

150 J. Yeh, Atomic calculation of photoionization cross-sections and asymmetry parameters. Gordon

and Breach Science Publishers, Langhorne, PE (USA), 1993.

151 M. B. Trzhaskovskaya, V. I. Nefedov and V. G. Yarzhemsky, Photoelectron Angular Distribution

Parameters for Elements Z = 1 to Z = 54 in the photoelectron energy range 100 – 5000 eV, Atomic

Data and Nuclear Data Tables vol. 77, pp. 97-159, 2001.

152 M. B. Trzhaskovskaya, V. I. Nefedov and V. G. Yarzhemsky, Photoelectron Angular Distribution

Parameters for Elements Z = 55 to Z = 100 in the photoelectron energy range 100 – 5000 eV, Atomic

Data and Nuclear Data Tables vol. 82, pp. 257-311, 2002.

153 S. Tanuma, C. J. Powell and D. R. Penn, Calculations of electron inelastic mean free paths. IX. Data

for 41 elemental solids over the 50 eV to 30 keV range, Surf. Interface Anal. vol. 43, pp. 689-713,

2011.

154 M. P. Seah and W. A. Dench, “Quantitative electron spectroscopy of surfaces: A standard data base

for electron inelastic mean free paths in solids,” Surf. Interface Anal. vol. 1, pp. 2-11, 1979.

155 J. Hudson, Surface science: an introduction. Wiley Interscience publication, Wiley, 1998.

156 G. Ertl and J. Küppers, Low energy electrons and surface chemistry. VCH, 1985.

157 S. Maghsoudy-Louyeh, M. Kropf and B. R. Tittmann, Review of Progress in Atomic Force

Microscopy: Review article, The Open Neuroimaging Journal vol. 12, pp. 86-104, 2018.

158 S. M. Widstrand, K. O. Magnusson, M. I. Larsson, L. S. O. Johansson, J. B. Gustafsson, E. Moons,

H. W. Yeom, H. Miki, M. Oshima, Preparation of stoichiometric GaN(0001)-1x1 studied with

spectromicroscopy, Surface Science 572, pp. 409-417, 2004.

159 S. M. Widstrand, K. O. Magnusson, L. S. O. Johansson, E. Moons, M. Gurnett, H. W. Yeom, H. Miki,

M. Oshima, Preparation of stoichiometric GaN(0001)-1x1: an XPS study, MRS Internet J. Nitride

Semicond. Res. 9, No. 4, pp. 1-7, 2004.

160 S. M. Widstrand, K. O. Magnusson, L. S. O. Johansson, E. Moons, M. Gurnett and M. Oshima,

Core-level photoemission From stoichiometric GaN(0001)-1x1, MRS Internet J. Nitride Semicond.

Res. 10, No. 1, pp. 1-9, 2005.

161 J. Hedman and N. Mårtnsson, Gallium Nitride studied by Electron Spectroscopy, Physica Scripta

22, pp. 1176-178, 1980.

162 D. C. Gleason-Rohrer, B. S. Brunschwig and N. S. Lewis, Measurement of the band bending and

surface dipole at chemically functionalized Si(111)/vacuum interfaces, J. Phys. Chem. C vol. 117,

pp. 18031-18042, 2013.

121

163 S. J. Wilkins, T. Paskova, C. L. Reynolds Jr., “Comparison of the Stability of Functionalized GaN

and GaP,” ChemPhysChem, vol. 16, pp. 1687−1694, 2015.

164 Y. Zhao, H. Gao, R. Huang, Z. Huang, F. Li, J. Feng, Q. Sun, A. Dingsun and H. Yang, Precise

determination of surface band bending in Ga-polar n-type GaN films by angular dependent X-ray

photoemission spectroscopy, Sci. Reports vol. 9, pp. 16969: 1-7, 2019.

165 Y. Imazeki, M. Sato, T. Takeda, M. Kobayashi, S. Yamamoto, I. Matsuda, J. Yoshinobu, M.

Sugiyama, and Y. Nakano, "Band bending of n-GaN under Ambient H2O vapor studied by x-ray

photoelectron spectroscopy," The Journal of Physical Chemistry C, vol. 125, no. 17, pp. 9011-9019,

2021.

166 Z. Huang, Y. Wu, Y. Zhao, L. Shi, R. Huang, F. Li, T. Liu, L. Xu, H. Gao, Y. Zhou, Q. Sun, S. Ding,

K. Xu and H. Yang, “Accurate Surface band bending determination on Ga-polar n-type GaN films

by fitting x-ray valence band photoemission spectrum, AIP Advances vol. 9, pp. 115106: 1-9, 2019.

167 M. Sato, Y. Imazeki, T. Takeda, M. Kobayashi, S. Yamamoto, I. Matsuda, J. Yoshinobu, Y. Nakano,

and M. Sugiyama, "Atomistic-level description of GaN/water interface by a combined spectroscopic

and first-principles computational approach,"The Journal of Physical Chemistry C, vol. 124, no. 23,

pp. 12466-12475, 2020.

168 S.-Y. Wu, L.-W. Lang, P.-Y. Cai, Y.-W. Chen, Y.-L. Lai, M.-W. Lin, Y.-J. Hsu, W.-I. Lee, J.-L. Kuo,

and M.-F. Luo, "Microscopic evidence for the dissociation of water molecules on cleaved GaN

(11

�00)," Physical Chemistry Chemical Physics, vol. 20, no. 2, pp. 1261-1266, 2018.

169 K. Idczak, P. Mazur, S. Zuber, and L. Markowski, An investigation of thin Zr films on 6H-SiC(0001)

and GaN(0001) surfaces by XPS, LEED, and STM, Appl. Phys. A 122-268, pp. 3-10, 2016.

170 V. Thakur and S. M. Shivaprasad, Electronic Structure of GaN nanowall network analyzed by XPS,

Appl. Surf. Sci. vol. 327, pp. 389-393, 2015.

171 W. R. L. Lambrecht, B. Segall, St Strite, G. Martin, A. Agarwal, H. Morkoç, and A. Rockett, X-ray

photoelectron spectroscopy and theory of the valence band and semicore Ga 3d states in

GaN, Phys. Rev. B, vol. 50, No. 19, pp. 14155-14160, 1994.

172 J. Long and V. Bermudez, Band bending and photoemission-induced surface photovoltages on

clean n-and p-GaN(0001) surfaces, Phys. Rev. B vol. 66, No. 12, p. 121308, 2002.

173 I. Bartoš, O. Romanyuk, J. Houdkova, P. Paskov, T. Paskova, and P. Jiricek, Electron band bending

of polar, semipolar, and non-polar GaN surfaces, J. Appl. Phys. vol. 119, no. 10, p. 105303, 2016.

174 P. Lorenz, T. Haensel, R. Gutt, R. Koch, J. A. Schaefer, and S. Krischok, Analysis of polar GaN

(0001) surfaces with photoelectron and high resolution electron energy loss spectroscopy, Phys.

Status Solidi b, vol. 247, no. 7, 1658-1661, 2010.

175 E. A. Kraut, R. W. Grant, J. R. Waldrop, and S. P. Kowalczyk (1983). Semiconductor core-level to

valence-band maximum binding-energy differences: Precise determination by x-ray photoelectron

spectroscopy. Phys. Rev. B 28, 1965-1977.

176 S. A. Chambers, T. Droubay, Tiffany C. Kaspar, and M. Gutowski (2004). Experimental

determination of valence band maxima for SrTiO3, TiO2, and SrO and the associated valence band

offsets with Si (001). J. Vac. Sci. & Technol. B: Microelectronics. Nanometer Structures, Processing,

Measurement, & Phenomena 22, 2205-2215.

177 M. G. Helander, M. T. Greiner, Z. B. Wang, and Z. H. Lu, Note: Binding energy scale calibration of

electron spectrometers for photoelectron spectroscopy using a single sample, Review Sci. Instrum.

vol. 82, no. 9, 2011.

178 R. Huang, F. Li, Y. Zhao, Y. Zhu, Y. Shen, X. Lu, Z. Huang, J. Liu, L. Zhang, S. Zhang, Z. Li, A.

Dingsun, and H. Yang, Ion Sputter Induced Interfacial Reaction in Prototypical Metal-GaN System,

Scientific Reports, vol. 8, no. 1, pp. 1-7, 2018.

122

179 Ł. Janicki, R. Korbutowicz, M. Rudziński, P. P. Michałowski, S. Złotnik, M. Grodzicki, S. Gorantla, J.

Serafińczuk, D. Hommel, and R. Kudrawiec, Thermal oxidation of [0001] GaN in water vapor

compared with dry and wet oxidation: Oxide properties and impact on GaN, Applied Surface

Science, vol. 598, p. 153872, 2022.

180 Y. P. Varshni, Temperature dependence of the energy gap in semiconductors, Physica, vol. 34, no.

1, pp. 149-154, 1967.

181 R. Kirste, N. Rohrbaugh, I. Bryan, Z. Bryan, R. Collazo, A. Ivanisevic, Electronic Biosensors Based on III-

Nitride Semiconductors. Ann. Rev. Anal. Chem. 8, pp. 8.1-8.21, 2015.

182 M. A. Mastro, O. M. Kryliouk, M. D. Reed, T. J. Anderson, A. Davydov, A. Shapiro, Thermal stability of

MOCVD and HVPE GaN Layers in H2, HCl, NH3, and N2, Physica Status Solidi (a) 188, no. 1, pp. 467-

471, 2001.

183 X. Zhang, S. Ptasinska, Dissociative Adsorption of Water on an H2O/GaAs (100) Interface: In Situ

Near-Ambient Pressure XPS Studies. J. Phys. Chem. C, 118, pp. 4259-4266, 2014.

184 F. Reinert, and S. Hüfner, Photoemission Spectroscopy-From Early days to Recent Applications. New

Journal of Physics 7, 97:1-34, 2005.

185 M. Honda, F. Matsui, H. Daimon, Chemisorption of L-cysteine on Au(111)/Si(111) and Si(111)√3 x √3-

Au surfaces. Surf. Review and Letters 13, 197-200, 2006.

186 J. Y. Huang, Y. S. Ning, K. S. Yong, Y. H. Cai, H. H. Tang, Y. X. Shao, S. F. Alshahateet, Y. M. Sun,

G. Q. Xu, Binding of Glycine and L-cysteine on Si(111)-7 x 7. Langmuir 23, 6218-6226, 2007.

187 K. Uvdal, P. Bodö, B. Liedberg, L-cysteine Adsorped on Gold and Copper: An X-Ray Photoelectron

Spectroscopy study. J. Colloid Interface Sci. 149, 162-173, 1992.

188 A-D. Filimon, P. Jacob, R. Hergenröder, A. Jürgensen, Study on the Reversible Changes of the

Surface Properties of an L-cysteine self-Assembled Monolayer on Gold As a function of pH. Langmuir 28,

8692-8699, 2012.

189 D.Costa,C-M. Pradier, F. Tielens, L. Savio, Adsorption and self-assembly of bio-organic molecules at

model surface: A route towards increased complexity. Surface Science Reports 70, 449-553, 2015.

123

Appendix A: Fitting Parameters of Core-Levels

The fit parameters for core-level (CL) XPS spectra of bonding configurations on the GaN surface

measured under different experimental conditions are listed in the tables below. The fits attained

by analyzing these spectra on the basis of line-profile models using a commercial CasaXPS

software (http://casaxps.com). It is shown, in each table, the type and numeral FWHM (full width

at half maximum) of the line-profile adopted in the fitting process, as the Voigt line-shape (mixed

Gaussian-Lorentzian GL(30)) function or the Lorentzian asymmetry LF(α, β, w, m) function,

convoluted with a Gaussian function of width m. The asymmetry tails being adjusted thru α and

β, restricted by the damping tail factor w. In each table, the CL electronic transitions and the peak

position, the peak binding energy (BE) (in eV), of unveiled surface bonds are also presented.

A1: Surface CL Photoemission Lines for as-received and 𝐍𝐍𝟐𝟐

+-sputtered

n-GaN(0001) Layers Recorded at Normal Emission (NE)

Table A1-1: The fit parameters of O 1s CL photoemission lines of the surface of the as-received

n-GaN(0001) layer recorded at NE and at 23 0C (sec. 5.2.2)

O 1s

Line-shape

profile

FWHM

(eV)

Peak

position (eV)

% Area

Residual

Ga2O3

GL(30)

1.5

532.983

61.89

1.448

Ga2OX

GL(30)

1.5

531.595

26.30

O-C

GL(30)

1.5

534.32

11.82

Table A1-2: The fit parameters of the O 1s, Ga 2p3/2, Ga 3d5/2,3/2 and N 1s CL lines recorded at

NE and at 23 oC for the N2

+-sputtered n-GaN(0001) surface (sec. 5.2.2)

O 1s

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga2O3

GL(30)

1.42

532.48

64.09

1.086

Ga2OX

GL(30)

1.42

530.363

35.91

Ga 2p

3/2

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga-N

LF(1.1,1.2,35,170)

1.57

1117.86

83.97

1.58

Ga-O

LF(1.1,1.2,35,170)

1.57

1118.63

15.04

Ga-Ga

LF(1.1,1.2,35,170)

1.57

1116.36

1

Ga 3d

5/2,3/2

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga-N5/2

LF(1.4,2,35,130)

1.06

19.9754

45.42

0.789

Ga-N3/2

LF(1.4,2,35,130)

1.06

20.4254

30.28

Ga-O5/2

LF(1.4,2,35,130)

1.06

20.7274

8.41

Ga-O3/2

LF(1.4,2,35,130)

1.06

21.1774

5.61

124

Ga metal5/2

LF(1.4,2,35,130)

1.06

18.6315

1.75

Ga metal3/2

LF(1.4,2,35,130)

1.06

19.0815

1.17

N 2s

GL(30)

2.76

17.1372

7.37

N 1s

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga LMM

GL(30)

1.65

396.112

0

1.25

Ga LMM

GL(30)

1.76

394.685

0

Ga LMM

GL(30)

2.7

392.621

0

Ga LMM

GL(30)

3.58

390.094

0

Ga LMM

GL(30)

2.66

400.702

0

Ga-N

GL(30)

1.13

397.627

87.89

N-X

GL(30)

1.13

398.651

12.11

A2: NAP-XPS CL Photoemission Lines of n-GaN(0001)/H2O Interface

Table A2-1: The fit parameters of O 1s CL line taken at NE and 23 oC, 60 oC, and 160 oC at

1 mbar for the n-GaN(0001)/H2O interface (sec. 5.3.1.1)

O 1s (23

oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga2O3

GL(30)

1.41

531.769

45.54

1.19

Ga2OX

GL(30)

1.42

530.652

4.51

Ga-OH

GL(30)

1.42

532.774

37.59

H2O

GL(30)

1.42

533.504

12.36

O 1s (60

oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga2O3

GL(30)

1.41

531.644

45.02

1.4

Ga2OX

GL(30)

1.41

530.527

4.73

Ga-OH

GL(30)

1.41

532.649

42.42

H2O

GL(30)

1.41

533.38

7.82

O 1s (160

oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga2O3

GL(30)

1.41

531.682

65.20

1.32

Ga2OX

GL(30)

1.41

530.565

5.87

Ga-OH

GL(30)

1.41

532.687

24.48

H2O

GL(30)

1.41

533.417

4.44

125

Table A2-2: The fit parameters of Ga 3d CL line taken at NE and 23 oC, 60 oC, and 160 oC at

1 mbar for the n-GaN(0001)/H2O interface (sec. 5.3.1.2)

Ga 3d (23

oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga-N5/2

LF(1.4,2,35,130)

1.05

19.995

41.20

0.849

Ga-N3/2

LF(1.4,2,35,130)

1.05

20.445

27.47

Ga-O5/2

LF(1.4,2,35,130)

1.05

20.747

12.23

Ga-O3/2

LF(1.4,2,35,130)

1.05

21.197

8.15

Ga metal5/2

LF(1.4,2,35,130)

1.05

18.6513

1.12

Ga metal3/2

LF(1.4,2,35,130)

1.05

19.101

0.74

N 2s

GL(30)

2.7

17.157

6.62

Ga-OH5/2

LF(1.4,2,35,130)

1.05

21.255

1.48

Ga-OH3/2

LF(1.4,2,35,130)

1.05

21.705

0.99

Ga 3d (60

oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga-N5/2

LF(1.4,2,35,130)

1.05

19.974

39.42

0.842

Ga-N3/2

LF(1.4,2,35,130)

1.05

20.424

26.27

Ga-O5/2

LF(1.4,2,35,130)

1.05

20.726

13.69

Ga-O3/2

LF(1.4,2,35,130)

1.05

21.176

9.13

Ga metal5/2

LF(1.4,2,35,130)

1.05

18.630

1.24

Ga metal3/2

LF(1.4,2,35,130)

1.05

19.0809

0.83

N 2s

GL(30)

2.7

17.1366

6.74

Ga-OH5/2

LF(1.4,2,35,130)

1.05

21.235

1.61

Ga-OH3/2

LF(1.4,2,35,130)

1.05

21.685

1.07

Ga 3d (160

oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga-N5/2

LF(1.4,2,35,130)

1.05

19.975

38.27

0.953

Ga-N3/2

LF(1.4,2,35,130)

1.05

20.425

25.51

Ga-O5/2

LF(1.4,2,35,130)

1.05

20.727

13.99

Ga-O3/2

LF(1.4,2,35,130)

1.05

21.177

9.32

Ga metal5/2

LF(1.4,2,35,130)

1.05

18.631

1.26

Ga metal3/2

LF(1.4,2,35,130)

1.05

19.081

0.84

N 2s

GL(30)

2.7

17.137

6.84

Ga-OH5/2

LF(1.4,2,35,130)

1.05

21.235

2.38

Ga-OH3/2

LF(1.4,2,35,130)

1.05

21.685

1.59

126

Table A2-3: The fit parameters of Ga 2p3/2 CL line taken at NE and 23 oC, 60 oC, and 160 oC at

0.02 mbar for the n-GaN(0001)/H2O interface (sec. 5.3.1.3)

Ga 2p

3/2

(23

oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga-N

LF(1.1,1.2,35,170)

1.56

1117.97

70.73

1.28

Ga-O

LF(1.1,1.2,35,170)

1.56

1118.74

27.68

Ga-Ga

LF(1.1,1.2,35,170)

1.56

1116.47

1.60

Ga 2p

3/2

(60

oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga-N

LF(1.1,1.2,35,170)

1.56

1118.04

53.59

1.28

Ga-O

LF(1.1,1.2,35,170)

1.56

1118.81

44.90

Ga-Ga

LF(1.1,1.2,35,170)

1.56

1116.54

1.51

Ga 2p

3/2

(160 oC)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga-N

LF(1.1,1.2,35,170)

1.56

1118.12

61.87

1.5

Ga-O

LF(1.1,1.2,35,170)

1.56

1118.89

36.46

Ga-Ga

LF(1.1,1.2,35,170)

156

1116.62

1.67

A3: UHV-XPS Surface CL Photoemission Lines for GaN(0001)/L-

cysteine

Table A3-1: The fit parameters of Ga 3s/S 2p CL lines taken at NE and 23 oC for the n-

GaN(0001)/L-cysteine interface after depositing L-cysteine molecules at 105 oC for 20 minutes

and 40 minutes (sec. 5.5.2)

Ga 3s/S 2p

(105 oC for 20

min)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga 3s

LF(0.6,0.6,25,280)

2.5

160.774

95.66

1.13

S 2p3/2(C-S)

LF(0.6,0.6,25,280)

1.2

163.121

2.89

S 2p1/2(C-S)

LF(0.6,0.6,25,280)

1.2

164.321

1.45

Ga 3s/S 2p

(105 oC for 40

min)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga 3s

LF(0.6,0.6,25,280)

2.5

160.847

73.43

1.08

S 2p3/2(C-H)

LF(0.6,0.6,25,280)

1.2

164.3

17.08

S 2p1/2(C-H)

LF(0.6,0.6,25,280)

1.2

165.5

8.54

S 2p3/2(C-S)

LF(0.6,0.6,25,280)

1.2

163.023

0.63

S 2p1/2(C-S)

LF(0.6,0.6,25,280)

1.2

164.224

0.32

127

Table A3-2: The fit parameters of N 1s CL lines taken at NE and 23 oC for the n-GaN(0001)/L-

cysteine interface after depositing L-cysteine molecules at 105 oC for 20 minutes and 40 minutes

(sec. 5.5.2)

N 1s

(105 oC for 20

min)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga LMM

GL(30)

1.65

396.12

0

1.26

Ga LMM

GL(30)

1.76

394.693

0

Ga LMM

GL(30)

2.7

392.629

0

Ga LMM

GL(30)

2.66

401.352

0

Ga LMM

GL(30)

4.43

390.233

0

Ga-N

GL(30)

1.13

397.635

82.93

N-X

GL(30)

1.13

398.659

13.11

NH2

GL(30)

1.13

399.95

3.97

N 1s

(105 oC for 40

min)

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga LMM

GL(30)

1.65

396.132

0

1.217

Ga LMM

GL(30)

1.76

394.705

0

Ga LMM

GL(30)

2.7

392.641

0

Ga LMM

GL(30)

2.66

401.364

0

Ga LMM

GL(30)

4.43

390.245

0

Ga-N

GL(30)

1.25

397.647

65.4

N-X

GL(30)

1.25

398.671

9.62

NH2

GL(30)

1.25

399.96

4.99

+

NH3

GL(30)

1.25

401.9

19.99

Table A3-3: The fit parameters of C 1s and O 1s CL lines taken at NE and 23 oC for the n-

GaN(0001)/L-cysteine interface after depositing L-cysteine molecules at 105 oC for 40 minutes

(sec. 5.5.2)

C 1s

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

C-N

GL(30)

1.41

286.687

47.89

1.34

C-S

GL(30)

1.41

285.356

24.42

C-O

GL(30)

1.41

288.952

27.70

O 1s

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga2O3/COO-

GL(30)

1.42

531.893

85.54

1.137

Ga2Ox

GL(30)

1.42

530.777

6.10

COOH

GL(30)

1.42

532.963

8.36

128

Table A3-4: The fit parameters of Ga 3s/S 2p CL lines taken at NE and 23 oC for the n-

GaN(0001)/L-cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes, and

after heating/annealing the sample for 30 minutes at 100 oC, 130 oC, and 150 oC (sec. 5.5.2)

Ga 3s/S 2p

at 100 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga 3s

LF(0.6,0.6,25,280)

2.5

160.946

91.09

1.234

S 2p3/2(C-H)

LF(0.6,0.6,25,280)

1.2

164.4

1.83

S 2p1/2(C-H)

LF(0.6,0.6,25,280)

1.2

165.6

0.91

S 2p3/2(C-S)

LF(0.6,0.6,25,280)

1.2

163.122

4.11

S 2p1/2(C-S)

LF(0.6,0.6,25,280)

1.2

164.322

2.06

Ga 3s/S 2p

at 130 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga 3s

LF(0.6,0.6,25,280)

2.5

160.99

93.14

1.17

S 2p3/2(C-S)

LF(0.6,0.6,25,280)

1.2

163.169

4.58

S 2p1/2(C-S)

LF(0.6,0.6,25,280)

1.2

164.369

2.29

Ga 3s/S 2p

at 150 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga 3s

LF(0.6,0.6,25,280)

2.5

161.05

93.85

1.182

S 2p3/2(C-S)

LF(0.6,0.6,25,280)

1.2

163.227

4.10

S 2p1/2(C-S)

LF(0.6,0.6,25,280)

1.2

164.427

2.05

Table A3-4: The fit parameters of N 1s CL lines taken at NE and 23 oC for the n-GaN(0001)/L-

cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes, and after

heating/annealing the sample for 30 minutes at 100 oC, 130 oC, and 150 oC (sec. 5.5.2)

N 1s

at 100 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga LMM

GL(30)

1.65

396.168

0

1.12

Ga LMM

GL(30)

1.76

394.741

0

Ga LMM

GL(30)

2.7

392.677

0

Ga LMM

GL(30)

2.66

401.4

0

Ga LMM

GL(30)

4.43

390.28

0

Ga-N

GL(30)

1.25

397.683

78.32

N-X

GL(30)

1.25

398.706

14.10

NH2

GL(30)

1.25

399.998

4.13

+

NH3

GL(30)

1.25

401.01

3.45

N 1s

at 130 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga LMM

GL(30)

1.65

396.197

0

1.10

Ga LMM

GL(30)

1.76

394.769

0

Ga LMM

GL(30)

2.7

392.709

0

129

Ga LMM

GL(30)

2.66

401.428

0

Ga LMM

GL(30)

4.43

390.309

0

Ga-N

GL(30)

1.25

397.711

81.41

N-X

GL(30)

1.25

398.735

14.48

NH2

GL(30)

1.25

400.026

4.11

N 1s

at 150 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga LMM

GL(30)

1.65

396.218

0

1.08

Ga LMM

GL(30)

1.76

394.791

0

Ga LMM

GL(30)

2.7

392.727

0

Ga LMM

GL(30)

2.66

401.45

0

Ga LMM

GL(30)

4.43

390.33

0

Ga-N

GL(30)

1.25

397.733

72.47

N-X

GL(30)

1.25

398.756

12.14

NH2

GL(30)

1.25

400.048

3.27

Table A3-5: The fit parameters of C 1s CL lines taken at NE and 23 oC for the n-GaN(0001)/L-

cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes, and after

heating/annealing the sample for 30 minutes at 100 oC, 130 oC, and 150 oC (sec. 5.5.2)

C 1s

at 100 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

C-N

GL(30)

1.43

286.674

24.74

1.01

C-S

GL(30)

1.43

285.343

57.45

C-O

GL(30)

1.43

288.939

11.17

C-X

GL(30)

1.43

289.874

6.64

C 1s

at 130 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

C-N

GL(30)

1.43

286.646

20.28

1.12

C-S

GL(30)

1.43

285.315

68.12

C-O

GL(30)

1.43

288.911

8.19

C-X

GL(30)

1.43

289.846

3.42

C 1s

at 150 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

C-N

GL(30)

1.43

286.663

17.06

1.15

C-S

GL(30)

1.43

285.332

72.97

C-O

GL(30)

1.43

288.928

5.40

C-X

GL(30)

1.43

289.863

4.56

130

Table A3-6: The fit parameters of O 1s CL lines taken at NE and 23 oC for the n-GaN(0001)/L-

cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes, and after

heating/annealing the sample for 30 minutes at 100 oC, 130 oC, and 150 oC (sec. 5.5.2)

O 1s

at 100 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga2O3/COO-

GL(30)

1.42

531.832

57.71

0.944

Ga2Ox

GL(30)

1.42

530.716

11.91

COOH

GL(30)

1.42

532.902

30.37

O 1s

at 130 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga

2

O

3

/COO-

GL(30)

1.42

531.832

51.83

1.10

Ga2Ox

GL(30)

1.42

530.715

16.42

COOH

GL(30)

1.42

532.901

31.75

O 1s

at 150 oC

Line-shape

profile

FWHM

(eV)

Peak

position

(eV)

% Area

Residual

Ga2O3/COO-

GL(30)

1.42

531.977

55.15

1.18

Ga2Ox

GL(30)

1.42

530.86

19.69

COOH

GL(30)

1.42

533.046

25.16

Table A3-7: The % Area of Ga 3s/S 2p CL lines taken at NE and 23 oC for the n-GaN(0001)/L-

cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes, and after

heating/annealing the sample for 30 minutes at 100 oC, 130 oC, and 150 oC (sec. 5.5.2)

% Area

Ga 3s

S 2p 3/2 (C-S)

S 2p 1/2 (C-S)

S 2p 3/2 (S-H)

S 2p 1/2 (S-H)

Sputtered

100 %

0

L-cys 105 oC

for 40 mins

73.43

0.63

0.32

17.08

8.54

heated 100

oC for 30

mins

91.09

4.11

2.06

1.83

0.91

heated 130

oC for 30

mins

93.14

4.58

2.29

0

heated 150

oC for 30

mins

93.85

4.10

2.05

0

131

Table A3-8: The % Area of NH3

+ and NH2 in N 1s CL lines taken at NE and 23 oC for the n-

GaN(0001)/L-cysteine interface obtained by depositing L-cysteine at 105 oC for 40 minutes, and

after heating/annealing the sample for 30 minutes at 100 oC, 130 oC, and 150 oC (sec. 5.5.2)

% Area

+

𝐍𝐍𝐍𝐍𝟏𝟏

NH2

Sputtered

0

L-cys 105 oC

for 40 mins

36.81

9.23

heated 100

oC for 30

mins

15.39

18.41

heated 130

oC for 30

mins

0

20.30

heated 150

oC for 30

mins

0

19.5

Table A3-9: The % Area of Ga 3s/S 2p CL lines taken at NE and 23 oC for the n-GaN(0001)/L-

cysteine interface obtained by depositing L-cysteine at 105 oC for 20 minutes, and after

heating/annealing the sample for 15, 30, 60 minutes at 75 oC, 100 oC, and 130 oC (sec. 5.5.2)

% Area

Ga 3s

S 2p 3/2 (C-S)

S 2p 1/2 (C-S)

Sputtered

100 %

0

L-cys 105 oC

for 20 mins

95.66

2.89

1.45

heated 75 oC

for 15 mins

95.60

2.93

1.47

heated 75 oC

for 30 mins

95.92

2.72

1.36

heated 75 oC

for 60 mins

95.65

2.90

1.45

heated 100 oC

for 15 mins

95.82

2.79

1.39

heated 100 oC

for 30 mins

96.39

2.47

1.23

heated 100 oC

for 60 mins

95.77

2.82

1.41

heated 130 oC

for 15 mins

95.84

2.77

1.39

heated 130 oC

for 30 mins

96.19

2.54

1.27

heated 130 oC

for 60 mins

96.09

2.61

1.3

132

Figure A1: A schematic picture of the L-cysteine molecule bonding structure to GaN.

Figure A2: A schematic picture of the L-cysteine molecule.

133

Appendix B: Wet Cleaning Method

a) Physical Cleaning step

This cleaning process includes rinsing the GaN samples into different baths:

i) The GaN samples will be rinsed in baths of acetone, then of methanol, with each sample-

soaked bath being ultrasonically agitated for sufficient periods of time around 5 mis. This

step is one approach for removing greases and atmospheric carbon (C) contaminants

residing on the surface of GaN samples via such organic solvents.

ii) The GaN samples will be rinsed in deionized water bath for enough time around 5 mins,

and then gently dried with pure nitrogen-gas (N2-) purge.

b) Chemical etching step

Chemical etching is sometimes required to remove native oxide (e.g., Ga2O3) monolayers

spontaneously formed on the surface of GaN samples during the above-described physical

cleaning process due to oxygen availability in the ambient environment. Etching alter both

the surface morphology and composition through the formation of pitting and rough

structures, which cannot be accomplished by only chemisorbing or physisorbing compounds

on the surface. Etching processes and hence the favorable ensuing properties of the GaN

surface are sensitive to the used etchant and its concentration, temperature at which etching

is being proceeded, and time of etching. A variety of inorganic acids can be used for etching

surfaces of GaN films and nanostructured wires like hydrochloric acid (HCl).

(a)

(b)

Figure B1: Typical ex-situ AFM images measured on (2.5 x 2.5) µm2 spots on the n-GaN(0001)

surface after wet cleaning meathod (a) RMS (1.723 nm) (b) RMS (1.926 nm)

134