3D multilayer graphene oxide thin film platform for

3D multilayer graphene oxide thin film platform for functional

devices

A thesis submitted for the degree of

Doctor of Philosophy

by

Yunyi Yang

Centre for Micro-Photonics

Faculty of Science, Engineering and Technology

Swinburne University of Technology

Melbourne, Australia

Principal supervisor: Professor Baohua Jia

Associate supervisors: Professor Minghui Hong, Dr. Han Lin and Dr. Jie Zhang

2019

2

i

Abstract

Small and big are antonyms. Investigate material down to atomic scale and

assembly as the building blocks for supporting the civilization are two pillars in the

arch of science. Carbon, regarded as the basis of all organic materials, shows an

enormous number of chemical structures due to the flexibility of its bonding that

offers the potential to bridge the two opposite ends (small and big) of material

science. Since the first discovery in 2004, graphene has been making a profound

impact that attracts tremendous efforts to explore the world beyond nanoscale and

discover the treasure at the atomic scale. Graphene, consisting of a monolayer sp2

hybridization carbon atoms arranged into a honeycomb lattice, has become a rising

star during the past decade due to its exceptional mechanical, thermal, optical, and

electrical properties. Due to its remarkable physicochemical characteristics,

graphene has been well explored in various applications of science and technology

including field-effect transistors, field emitters, biosensors, optical polarizers, and

transparent conducting electrodes.

Considerable efforts have been paid to produce high-quality graphene material.

Several thin film deposition techniques, such as chemical vapour deposition (CVD)

on metal (Cu and Ni) as well as epitaxial growth on silicon (Si) carbide and

ruthenium, have been developed to deposit graphene-based materials. However, the

costly and complex fabrication techniques will limit paths toward large-scale

fabrication. In addition, the sophisticated transfer process poses a hurdle for

integrating graphene with arbitrary surfaces. Therefore, the lack of an efficient

ii

approach to producing large-scale graphene sheets in large quantities with a cost-

effective methodology has been the major obstacle to exploiting most proposed

applications in the real-word. In comparison, graphene oxide (GO), prepared by

chemical oxidation of graphite and subsequent exfoliation in water and

reproduction, has been recognized as a promising precursor for scalable and low-

cost production of graphene-based materials and devices. From the structure

prospect, GO can be regarded as the graphene plane covalently decorated with

various oxygen functional groups (OFGs) either on the basal plane or at the edge.

Therefore, the reduction methodologies for GO are promising due to the advantages

of large-scale production of graphene-based materials, tractable processing, and the

controllable properties. The optical and electrical properties of reduced GO (rGO)

are close to pristine graphene. Moreover, the optical and electrical properties of GO

can be tailored by manipulating the relative content of the OFGs of GO during the

reduction process. Thus, the oxidization-reduction scheme to produce graphene-

based materials has been explored for numerous electronic and optoelectronic

applications including transparent conductive electrodes, transistors, field emitters,

photovoltaic devices, light-emitting devices, flexible electronic materials, electrical

sensors and energy storage devices, and so on.

For integrated optoelectronic applications, the thin film is a desired format. To

date, several methods have been explored for the preparation of GO thin films. The

approaches such as drop-casting, dip-coating, spraying, and spin-coating have been

reported to provide a one-step solution to coat GO film synthesis on flat surfaces.

However, such procedures will result in cracks, wrinkles and non-uniform

dispersions owing to the low accuracy of control of the process. These defects lead

to limited uniformity in film thickness, roughness, and quality, making the films

unsuitable for optical applications. Therefore, alternative GO preparation

approaches are highly required for high quality and precise control for optical or

photonic applications.

In the meantime, functional nanostructures, based on the dominant

semiconductor material and complementary metal–oxide–semiconductor (CMOS)

compatible technique, provide novel opportunities for exploring fascinating

performance and applications by developing designed three-dimensional (3D)

architecture with unique properties. With the integration of two-dimensional (2D)

iii

materials with well-defined functional nanostructures, the hybrid structures pave

the way for novel applications and enhanced performances in devices such as

photovoltaic cells, photodetectors, and integrated photonic devices. By utilizing the

large surface area and unique morphology of nanostructures, conformal coating of

ultrathin 2D materials on designed nanostructures plays the key role to improve

both the optic and electronic performances of hybrid structures. The existing GO

deposition methods can only coat on top of the nano-architecture rather achieve

conformal coating and the non-precisely controllable feature limits the optic and

electronic performance of the hybrid coated structure. Yet, realizing uniform

conformal coating of graphene-based 2D materials onto well-defined

nanostructures, for example, Si nanostructures, with an accurate control method is

highly desired for the future optoelectronic devices.

The aim of this thesis is to provide novel solutions to the challenges that lie in

the preparation of high-quality graphene-based film with accurate control and find

out a practical way to manufacture ultrathin and large scale graphene-based

platform for photonics application. Furthermore, the optical properties can be

dynamically tuned by laser-reduction that offers more potential for device design

and fabrication. By conformally coating the developed graphene-based material

onto Si nanostructures, hybrid photovoltaic and integrated photonic devices are

demonstrated and manufactured with unprecedented properties. The major findings

can be summarised as follows:

1. To overcome the challenge of large-scale preparation of precisely controlled

graphene-based material, the ultra-thin graphene-based platform for photonics

applications are developed through a solution-phase layer by layer (LBL)

deposition method. The method is able to one-step fabricate graphene-based

material on diverse substrates with arbitrary surface, shapes, and sizes without

transfer process. And the layer number can be controlled accurately down to a

monolayer (with a thickness: ~1 nm). The clear multilayer structure is achieved

with this method and well characterized by scanning electron microscopy (SEM)

and atomic force microscope (AFM) images that verify the LBL architecture.

Furthermore, this method is demonstrated for large scale fabrication of the films on

various substrates including centimetre-scale glass, 4-inch Si wafer, and A4 size

PET film. Due to the strong mechanical properties, the GO film can be well

iv

maintained on flexible substrates even with bending and twisting without noticeable

cracks and wrinkles. UV-Vis spectrometer and AFM are introduced to characterize

the optical property and film quality of the multilayer GO films. The achieved large-

scale GO films preserve their properties with an unprecedented low surface

roughness of ~2 nm, far transcending the most stringent optical standard.

2. In order to realize good integration with functional nanostructures and achieve

high-performance hybrid devices, the LBL method is further developed for GO

conformal coating by tuning the flake size and hydrophilicity of the GO suspension.

The well-defined conformal coating on Si nanostructures like Si nanowires

(SiNWs) has been demonstrated by manipulating the GO flake size and surface

chemistry through the LBL deposition method. We have systematically analyzed

the optical and electrical performance of the integrated GO-SiNWs hybrid

structures. Up to 20% reduction of reflection with broadband wavelength range and

enhanced passivation are achieved in GO-SiNWs hybrid structure compared with

SiNWs solely. By applying this method to Si solar cell, the overall improved

performance is observed. Up to 60% enhancement in carrier lifetime measurement

and a 20% improvement in external quantum efficiency (EQE) of the hybrid solar

cells are achieved. Our studies suggest that the LBL GO conformal coating provides

a precisely controllable and cost-effective novel material platform to enhance the

optical and electrical responses of photonics devices.

3. The prepared GO film can be converted to graphene by laser-reduction. The

laser-reduction can provide localized reduction for one-step fabrication of

functional devices. In addition, the effective parameters and bandgap of GO can be

in-situ manipulated by laser processing. Taking the advantages of the change in

optical property, a quick response (QR) code based on amplitude modulation and a

flat lens based on phase modulation are realized. Furthermore, the synthesized

graphene-based material is resilient to water. Thus we demonstrated an ultrathin

microfluidic flat lens that is able to focus optical energy with subwavelength

resolution in an aquatic environment. The graphene-based material is expected to

find broad applications in lab-on-chip, microfluidics, bio-medical optoelectronic

and integrated devices.

4. By incorporating graphene-based material with integrated photonics platform

based on the conformal coating method well-developed in this thesis, the improved

v

nonlinear performance of hybrid waveguides has been investigated. By taking the

advantages of both designed nanostructure and high nonlinearity of the GO, the

four-wave-mixing measurements in the hybrid waveguides integrated with thin GO

films lead to significant enhancement in the nonlinear conversion efficiency (CE)

of ~9.5 dB in a 1.5-cm-long waveguide with 2 layers of GO. The enhancement is

attributed to the high Kerr nonlinearity, low linear loss, and the strong mode overlap

of the GO films with the waveguide. The value of n2 that we extract from our

measurements agrees reasonably well with our previous Z-scan measurements of

thick (≥ 1 µm) films. We show theoretically that the enhancement in the conversion

efficiency through the integration of thin GO films can be as high as 20 dB in the

doped silica waveguides.

5. For further exploring the potential of the GO conformal coating and nonlinearity

onto Si photonic devices, two types of GO-Si hybrid waveguides have been

proposed. Simulations of GO-Si wire waveguide and GO-Si slot waveguide have

presented respectively. The maximum CE enhancement of over 30 dB can be

achieved in hybrid wire waveguide and over 35 dB in hybrid slot waveguide. With

the potential of photo-patterning to tune the nonlinearity of the GO films, these

hybrid integrated devices offer a powerful new way to construct high performance

nonlinear photonic devices, thus holding a great promise for future ultra-high-speed

all-optical information processing.

Despite its young age, the field of 2D materials has advanced significantly in

the last few years. This thesis not only develops a scalable and low-cost GO films

with unprecedented control on film thickness and roughness but also explores the

unique optical properties, which are crucial for understanding the fundamental

physics of 2D materials interactions with light. In the meantime, the flexibility of

GO material engineered by laser enables a versatile platform for wide range of

optoelectronic devices and functionalities that are inaccessible before. The

graphene-based film shows unique properties that are unavailable in conventional

materials, which opens up new avenues for various multidisciplinary applications

including non-invasive in-situ 3D biomedical imaging and sensing, all-optical

broadband photonic chips, photovoltaic, aerospace photonics, optical

microelectromechanical systems, and lab-on-chip devices.

vi

Acknowledgements

It is in the May of 2014, when I almost finished my master period study in Beijing

Institute of Technology in China, I got the opportunity to pursue a PhD study with

the supervision of Professor Baohua Jia at the Centre for Micro-Photonics (CMP).

Swinburne University of Technology (SUT). Although I nearly knew nothing at

that time, I still start this adventure (at least for myself) without hesitation. And I

am lucky enough to survive and move towards the completion of my PhD now

without any permanent head damage (another definition for PhD).

First of all, I would like to thank my principal supervisor, Professor Baohua Jia.

She offered me a great opportunity as a PhD candidate in Australia. With the almost

4 years PhD period, I am not alone on my journey. Baohua is always on my back

and gives help to me for both research and normal life. Moreover, she gives me the

freedom to do what I want and proposed, which makes me confident and

independent. When I face setbacks, she always offers me support and guidance

towards my goals. Here, I would like to say thank you again to my supervisor.

Many thanks go to my co-supervisor Dr. Han Lin, who has always been giving

valuable experience and skills. It is very helpful for helping me setup the system,

guide me in detail, which definitely benefits my PhD And I also would like to thank

my co-supervisor Dr. Jie Zhang. He kindly offers me his experience in chemistry,

material science, and devices fabrication. I would like to thank my co-supervisor

Professor Minghui Hong from the National University of Singapore. Although I

have never got a chance to meet him face to face, he still gives me his thoughts and

guidance to my projects as well as paper revision.

I would like to give the sincere thanks to my colleagues and friends Dr. Haoran

Ren, Dr. Qiming Zhang, Dr. Yaoyu Cao, Dr. Ye Chen, Dr. Xiaorui Zheng, Dr.

Boyuan Cai, Dr. Yinan Zhang and Professor Xiangping Li for their help and

support. Also, I would like to thank all of the members in LNI group for the kind

vii

help, especially for the students who are still on the journey, good luck! And also

the give the special thanks to the gumtree squad for their support.

I am grateful to the technical staff, Mr. Dan Kapsaskis, Dr. Tania Moein, Dr.

Xiaorui Zheng again, Dr. Xiaohan Yang, Dr. Xijun Li and so on. And I also want

to thank administration staff, Ms. Kellie Hamilton, Ms. Barbara Gillespie, Ms. Jia

Lou and Mr. Riaan Lourens.

Finally, I would like to thank my parents for their support during my long time

in school from a child to an adult. The most special thanks to my wife, who came

in my life ten years ago. It is her love to support me and encourage me to pursue

my dream. Their company is the courage and reason, which make me keep putting

one foot in front of the other on the journey far from the end.

Yunyi Yang

Melbourne, Australia, January 2018

viii

Declaration

I, Yunyi Yang, declare that this thesis entitled:

“3D multilayer graphene oxide thin film platform for functional devices”

is my own work and has not been submitted previously, in whole or in part, in

respect of any other academic award.

Yunyi Yang

Centre for Micro-Photonics

Faculty of Science, Engineering and Technology

Swinburne University of Technology

Melbourne, Australia

Date this day, 18th January 2018

ix

Contents

Abstract ............................................................................................................... i

Acknowledgements ............................................................................................. i

Declaration ...................................................................................................... viii

Contents ............................................................................................................ ix

List of Figures .................................................................................................. xii

List of Tables ................................................................................................... xxi

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

1.1 The rise of 2D materials ............................................................................. 1

1.2 Graphene-based 2D materials ..................................................................... 4

1.3 Objectives of the thesis ............................................................................... 8

1.4 Outline of the thesis.................................................................................... 9

2 Literature review .......................................................................................... 12

2.1 Introduction .............................................................................................. 12

2.2 Graphene-based thin-film preparation approaches .................................... 13

2.2.1 Graphene thin-film preparation approaches ...................................... 14

2.2.2 Graphene oxide film preparation approaches .................................... 18

2.2.3 Challenges and opportunities ............................................................ 21

2.3 Photoreduction methods for graphene oxide ............................................. 23

2.3.1 Review ............................................................................................... 23


2.4 Optical properties of graphene-based material .......................................... 26

2.4.1 Linear optical properties ................................................................... 26

2.4.2 Nonlinear optical properties .............................................................. 29


2.5 Graphene-based functional devices ........................................................... 33

2.5.1 Graphene-based linear photonic devices ........................................... 33

2.5.2 Graphene-based nonlinear photonic devices ...................................... 34


2.6 Conclusion ............................................................................................... 38

3 Ultra-thin graphene oxide film preparation ................................................ 39

3.1 Introduction .............................................................................................. 39

3.2 Layer by layer graphene oxide film preparation ........................................ 40

x

3.3 Multilayer structure characterization......................................................... 42

3.4 Optical properties and film quality characterization .................................. 44

3.5 Large scale fabrication on different substrates .......................................... 47

3.6 Conclusion ............................................................................................... 51

4 Graphene oxide 3D conformal coating on Si nanostructures ...................... 52

4.1 Introduction .............................................................................................. 52

4.2 3D conformal coating on Si nanowires ..................................................... 53

4.3 Graphene oxide wet chemical tunabilities ................................................. 58

4.3.1 Flake modification ............................................................................. 58

4.3.2 Surface processing ............................................................................. 59

4.4 Broadband anti-reflection of 3D GO conformal coating ............................ 61

4.4.1 Theoretical study and simulation ....................................................... 61

4.4.2 Experimental results .......................................................................... 66

4.5 Characterization of GO conformal coated Si solar cells ............................ 66

4.6 Conclusion ............................................................................................... 69

5 Ultra-thin graphene oxide platform for photonic devices ........................... 71

5.1 Introduction .............................................................................................. 71

5.2 Laser reduction of GO film ....................................................................... 73

5.3 Tunable metamaterial platform ................................................................. 75

5.3.1 Design of graphene based metamaterial ............................................ 76

5.3.2 Experimental realization of laser tunable graphene-based metamaterial

................................................................................................................... 79

5.4 Laser patterning characterization and the design of amplitude modulation

device ............................................................................................................ 82

5.5 Phase modulation device design for microfluidic bio lens ......................... 84

5.5.1 Design of the phased based microfluidic bio lens ............................... 84

5.5.2 Experimental realization of water-proof GO microfluidic lens ........... 86

5.6 Conclusion ............................................................................................... 88

6 Hybrid graphene oxide coated integrated photonics devices ...................... 90

6.1 Introduction .............................................................................................. 90

6.2 Nonlinear theory of GO coated integrated photonic devices ...................... 92

6.3 General nonlinear theory and simulation strategy ..................................... 94

6.4 Device fabrication and characterization .................................................... 96

6.5 Experiment results .................................................................................. 100

6.6 Results and discussion ............................................................................ 104

xi

6.7 Simulation of GO-Si hybrid waveguide .................................................. 105

6.7.1 Simulation results of hybrid GO-Si wire waveguide ......................... 107

6.7.2 Simulation results of hybrid GO-Si slot waveguide .......................... 109

6.8 Conclusion ............................................................................................. 112

7 Conclusions ................................................................................................. 113

7.1 Conclusions ............................................................................................ 113

7.2 Outlooks................................................................................................. 116

Bibliography ................................................................................................... 119

Publications .................................................................................................... 130

xii

List of Figures

Figure 1.1 2D material family. ............................................................................. 2

Figure 1.2 Formation of the van der Waals heterostructures with various 2D

materials. The mechanically assembled stacks are shown on top of the figure

and the large-scale growth by CVD or physical epitaxy is shown on the bottom

of this figure. ................................................................................................ 3

Figure 1.3 In graphene, carbon atoms are bonded together through the sp2

hybridization. The shiny and flexible graphene paper is prepared by controlled

restacking of graphene sheets ....................................................................... 4

Figure 1.4 Structures of graphene-based material. Graphene is a 2D building block

for carbon materials of all other dimensionalities. It can be formed into 0D

buckyballs, rolled into 1D nanotubes or layered to 3D graphite .................... 5

Figure 1.5 (a) Schematic illustration of the preparation of GO from graphite. (b)

Schematic illustration of the structural model of a GO sheet ......................... 6

Figure 2.1 Road map of current graphene deposition methods, which offer the

choices for potential applications based in terms of size, quality and price.65

................................................................................................................... 14

Figure 2.2 Mechanical exfoliation of graphene by scotch tape ............................ 15

Figure 2.3 Morphology of epitaxially grown graphene on Ru substrate .............. 15

Figure 2.4 Characterization of a CVD grown graphene film on a copper foil. (A)

SEM image of graphene on a Cu substrate. (B) Zoomed SEM image of a Cu

grain boundary and steps, two to three layers graphene flakes, and graphene

wrinkles. Inset is the TEM images of folded graphene edges (1L: one layer;

2L: two layers). Graphene films are transferred onto a Si substrate (C) and a

glass plate (D). ........................................................................................... 16

xiii

Figure 2.5 The schematic of preparation and PMMA-assisted transfer of monolayer

graphene. Cu foil is washed and baked firstly (a), and then graphene is

deposited on Cu by CVD (b). Secondly, the polymer is coated onto graphene

as a substrate (c) and removed Cu (d). Finally, the film is transferred to a SiO2

substrate (e) with removing the polymer layer (f). ...................................... 17

Figure 2.6 Schematic of the roll-to-roll production of graphene films on a copper

foil. The whole process includes attachment of polymer supports, copper

etching and dry transfer procedure on a target substrate. ............................. 18

Figure 2.7 (a) GO suspension prepared by a self-assembly method. (b) Schematic

of the spin coating process. (c) Optical microscope image of the as-prepared

GO thin film over a large area. Inset is the surface morphology of the GO layer.

................................................................................................................... 19

Figure 2.8 (a) The schematic of the filtration setup and the (b) vacuum filtration

process. (c) The filtrated GO thin films on the filtration membrane. (d) The

transferred GO thin film on the glass substrate. (e) The AFM images and (f)

the SEM image of the GO thin film ............................................................ 21

Figure 2.9 Free-standing graphene films show extremely high tensile strength. (a)

Cross-sectional SEM image of GO stacking in a film produced by filtration.

(b) Chemical reduced GO film shows a metallic-like shiny lustre ............... 23

Figure 2.10 Flash reduction process of GO. (a) Optical images of GO, flash

reduction of GO and RGO, from left to right. (b) Image of the arrays of

rGO/polystyrene interdigitated electrodes fabricated on a 1.5-inch diameter

GO/polystyrene thin film deposited on a Nylon filter paper. Inset: the close-up

view of one set of such electrode ................................................................ 24

Figure 2.11 Fabrication process and optical microscopy images of reduced and

patterned GO films. (a) Schematic and conceptual design of the GO

microcircuit. (b) - (d) Different designed patterns of GO microcircuits. (e)

Microscopy image of the university logo .................................................... 26

Figure 2.12 The dispersion relations of refractive indices (n) of rGO at different

laser powers have been measured by using ellipsometry ............................. 28

xiv

Figure 2.13 The dispersion relations of the extinction coefficient (K) of rGO at

different laser powers have been measured by using ellipsometry ............... 29

Figure 2.14 (a) Output fluence versus input fluence (scatters). Four different stages

(I, II, III and IV) are labelled. T: the linear transmittance (black dash line).

MD1 and MD2 represent the modulation depth in stages II and IV, respectively.

(b-e) Power dependent open aperture Z-scan results (scatters) fitted with

corresponding theory given in the reference (black solid lines). Inset:

schematic atomic structure of GO films in different stages. (f) Raman spectra

of GO and fully reduced GO films. Inset: a figure of laser-induced reduction

process ....................................................................................................... 30

Figure 2.15 (a) Schematic of the wavefront manipulation by the GO lens converting

the incident plane wave into a spherical wavefront. Inset: optical profile image

of the GO lens. Scale bar: 2 µm. Theoretical focal intensity distributions in the

lateral (b) and axial(c) directions. Experimental focal intensity distributions

along the lateral (d) and axial (e) directions ................................................ 33

Figure 2.16 Graphene-clad silicon photonic crystal nanostructures. (a) Scanning

electron micrograph (SEM) of the tuned photonic crystal cavity. (b) Measured

Raman scattering spectra of monolayer CVD-grown graphene on the photonic

crystal cavity membrane. (c) SEM of the suspended graphene–silicon

membrane. (d) Example measured graphene-clad cavity transmission. ....... 35

Figure 2.17 SEM image of the SGM resonator .................................................. 36

Figure 2.18 (a) Sketch band diagram of graphene. (b) Degenerate FWM energy

diagram. (c) SEM image of the cross-section of a SiN waveguide. (d)

Schematic of the gating scheme. (e) Optical microscope image of a set of

waveguides.121 ............................................................................................ 36

Figure 3.1 Schematics of the layer-by-layer process (a) and layered structure of a

5-layer GO film with the inset highlighting the molecular structure of the film

(b). ............................................................................................................. 40

Figure 3.2 Optical images of GO films on a glass substrate from 1 to 5 layers. ... 41

xv

Figure 3.3 GO multilayer film surface profile characterization. (a) Microscope

image of a 5-layer GO film. The surface is ultra-smooth under a microscope

(Nikon ECLIPSE MA100 microscope). (b) Scanning electron microscope

(RAITH150-TWO) image of 5-layer GO film. ........................................... 41

Figure 3.4 SEM image of a 5-layer GO multilayer structure, clearly showing the

layer-by-layer structure. ............................................................................. 42

Figure 3.5 AFM measured thickness profiles of a 5-layer GO multilayer film, with

corresponding AFM topographic image (inset). .......................................... 43

Figure 3.6 Thickness profiles of a PDDA layer, with corresponding AFM

topographic images (insets). ....................................................................... 43

Figure 3.7 Thickness profile of a GO monolayer, with corresponding AFM

topographic image (inset). .......................................................................... 44

Figure 3.8 Absorbance and roughness versus solution concentration at various

wavelengths of the 5-layer GO film. Phases I and II are separated by the grey

dash line. .................................................................................................... 45

Figure 3.9 Absorbance and roughness versus immersion time at various

wavelengths of the 5-layer GO film. ........................................................... 46

Figure 3.10 Broadband absorption spectra with increased layer number (1 to 5). 46

Figure 3.11 Thickness and roughness versus layer number from 1 to 10 layers... 47

Figure 3.12 (a) The optical image of a 5-layer GO film on a 4-inch silicon wafer.

(b) Corresponding Raman spectra of the Si wafer and the GO film coated Si

wafer. ......................................................................................................... 48

Figure 3.13 Thickness mapping of a 5-layer GO film on a Si wafer by an

ellipsometer................................................................................................ 48

Figure 3.14 A 72-mm-diameter curved (top left) acrylic lens (top right). The lens

with 10-layer (bottom left) and 20-layer (bottom right) GO coatings. ......... 49

Figure 3.15 A 5-layer GO Large-scale film integrated on flexible polyester (PET)

film (left) compared with pristine PET substrate (right). ............................. 50

xvi

Figure 3.16 A 5-layer GO film on a flexible transparent substrate with bending and

twisting. ..................................................................................................... 50

Figure 4.1 Schematic of 3D conformal coating process on Si nanostructures. ..... 54

Figure 4.2 SEM image of the Si nanowires......................................................... 54

Figure 4.3 SEM image of the conformal-coated Si nanowires. ........................... 55

Figure 4.4 SEM image (zoom in) of the morphology of the GO conformal coating

................................................................................................................... 55

Figure 4.5 SEM image of low (top) and high (bottom) aspect ratios of the .......... GO

coating. ...................................................................................................... 56

Figure 4.6 SEM image of the boundary of the GO coating. ................................ 56

Figure 4.7 SEM image of large scale with (left) and without (right) GO the coating.

Insets are zoomed in structures. .................................................................. 57

Figure 4.8 Raman spectra of the GO conformal coating on Si nanowires. ........... 57

Figure 4.9 AFM image of sonicated GO flake sizes with different sonication time

and corresponding flake profile. (a) AFM image of GO solution without

sonication (left) and flake morphology (right). (b) AFM image of GO solution

with 5-minute sonication (left) and flake morphology (right). (c) AFM image

of GO solution with 10 minutes sonication (left) and flake morphology (right).

................................................................................................................... 59

Figure 4.10 Contact angle of PDDA solution (left) and with surfactant (right).... 60

Figure 4.11 Contact angle of GO solution (left) and with surfactant (right). ....... 60

Figure 4.12 SEM images of the coating morphology with flake size control and

surface modification. (a) GO film floats on top of the SiNWs without any

process. (b) Partly conformal coating with reduced flake size. (c) Fully

conformal coating with reduced flake size and surface modification. .......... 60

Figure 4.13 (a) Schematic of the SiNW model. (b) Field distribution of single cell

of SiNW. .................................................................................................... 62

xvii

Figure 4.14 Reflectance for SiNWs with different heights (100-400 nm) with the

same diameter of 400 nm over a broadband wavelength (300 to 1100 nm). 62

Figure 4.15 Reflectance with different diameters (200-600 nm) with the same

height of 300 nm over a broadband wavelength (300 to 1100 nm). ............. 63

Figure 4.16 (a) Schematic of the hybrid GO-SiNW model. (b) Field distribution of

single cell of GO-SiNW model. .................................................................. 64

Figure 4.17 Reflectance with different heights (100-400 nm) for GO –SiNW hybrid

structures with the same aspect ratio (around 2) over a broadband wavelength

(300 to 1100 nm). ....................................................................................... 65

Figure 4.18 Reflectance spectra (300 to 1100 nm) of Si wafer (blue line), SiNWs

(black line) and GO-SiNWs hybrid structure (red line), respectively. ......... 66

Figure 4.19 Carrier lifetime mappings of bare SiNWs. ....................................... 67

Figure 4.20 Carrier lifetime mappings of GO-SiNW hybrid structures. .............. 68

Figure 4.21 Measured current density and voltage (J-V) curves of the GO coated

honey-combed solar cell (red line), compared with those of the bare solar cells

without GO coating (black line). Inset: optical image of GO coated honey-

combed solar cell. ...................................................................................... 68

Figure 4.22 Measured external quantum efficiency (EQE) of honey-combed Si solar

cell with (red line) and without (black line) GO coating.............................. 69

Figure 5.1 Schematic of femtosecond laser fabrication on GO film. ................... 73

Figure 5.2 Raman spectra of the GO film and GMLs.......................................... 74

Figure 5.3 Raman mapping of the GO film coated on cover glass....................... 74

Figure 5.4 Refractive index (n) and extinction coefficient (k) for GO and GML films

with broadband wavelength from 200 to 1600 nm. ..................................... 75

Figure 5.5 Schematic of the in-situ tunable graphene-based metamaterial. ......... 76

xviii

Figure 5.6 The simulation of real (a) and imaginary (b) parts of the permittivity of

GO metamaterial with different dielectric layer thickness. Simulated real (c)

and imaginary (d) parts of the permittivity of graphene-based metamaterial

with different dielectric spacing layer thickness. ......................................... 78

Figure 5.7 Optical microscopic images of laser reduced GO films with 8 reduction

levels. ......................................................................................................... 79

Figure 5.8 Refractive index (n) change versus (a) and extinction coefficient (k) (b)

versus different laser power for graphene metamaterial with broadband

wavelength from 200 to 1600 nm. .............................................................. 79

Figure 5.9 Bandgap tunning range from GO to rGO with laser power. ............... 80

Figure 5.10 Real and imaginary parts of the optical conductivity of GO and GML

film compared with a CVD grown graphene film, respectively. .................. 81

Figure 5.11 The fabricated lines with different laser power 5 µW (left) and 4 µW

(right) measured with an atomic force microscope (AFM). ......................... 82

Figure 5.12 Plot of line thickness and width versus laser power. ........................ 83

Figure 5.13 Microscopic image (top) and 3D surface profile (bottom) of a

Swinburne logo on an ultrathin GO film (30 nm) fabricated by the laser direct

writing method. .......................................................................................... 83

Figure 5.14 (a) Design of a QR code. (b) Laser-written QR code on an 18-nm thick

GO film coated on a cover glass. (c) Readout result of the QR code. .......... 84

Figure 5.15 Theoretical design of the flat lens (a) and the simulated focal spot (b).

................................................................................................................... 85

Figure 5.16 Schematic of the GO lens operating in a microfluidic device with a

biocompatible solution (a) and image of the experimental setup (b). ........... 86

Figure 5.17 (a) Microscope image (left) and focal spot of the laser inscribed flat

lens on an 18-nm thick GO film. (b) Microscope image (left) and focal spot of

the flat lens after working in microfluidic devices for one month (right). .... 87

xix

Figure 5.18 (a) The intensity distributions of the GO lens for theoretical flat lens,

laser inscribed flat lens and flat lens after immersion in a microfluidic device

for 1 month. (b) Topographic profile of GO lens (left) and cross section of the

thickness (right). ........................................................................................ 88

Figure 6.1 Schematic of GO coated silica planarized integrated waveguide. ....... 92

Figure 6.2 General process of simulation strategy. ............................................. 95

Figure 6.3 Lumerical MODE solution software simulation panel. ...................... 95

Figure 6.4 COMSOL software simulation panel. ................................................ 96

Figure 6.5 The MATLAB software simulation panel. ......................................... 96

Figure 6.6 Micrograph of the hybrid waveguide with two layers of GO.............. 97

Figure 6.7 Raman spectra of GO on the integrated chip. ..................................... 98

Figure 6.8 (a) Measured insertion loss of hybrid waveguides with different numbers

of GO layers. (b) The additional propagation loss of the hybrid waveguide with

different numbers of GO layers. ................................................................. 99

Figure 6.9 Experimental setup for testing FWM in the GO hybrid integrated

waveguide. EDFA: erbium-doped fiber amplifier. PC: polarization controller.

DUT: device under test. OSA: optical spectrum analyser. VOA: variable

optical attenuator. ..................................................................................... 100

Figure 6.10 FWM spectra of the integrated waveguide without GO and with 2 layers

of GO. ...................................................................................................... 101

Figure 6.11 Zoom in spectra of the generated idlers after FWM in the integrated

waveguide with 0 to 5 layers of GO. ......................................................... 101

Figure 6.12 Output powers of idler for various pump powers coupled to the

waveguide without GO and with 2 layers of GO. ...................................... 102

Figure 6.13 Net CE enhancements for various pump powers coupled to the

waveguide with 1 to 5 layers of GO. ......................................................... 103

xx

Figure 6.14 (a) Power variations of the output idler when the pump wavelength was

fixed at 1550 nm and the signal wavelength was detuned from -10 nm to 10

nm. (b) Output powers of idler for the waveguide with different coating lengths

of GO. ...................................................................................................... 104

Figure 6.15 (a) TE mode profile of the hybrid waveguide with 2 layers of GO at

the wavelength of 1550 nm. (b) The plot of η as a function of pump power.

................................................................................................................. 104

Figure 6.16 (a) − (b) η as a function of pump power, wavelength detuning, and GO

lengths, respectively. The dots represent the experimentally measured values

and the lines show the fit curve calculated based on Eqs. (6.2-6.6). WG:

waveguide. ............................................................................................... 105

Figure 6.17 Schematic of GO conformal coated Si wire waveguide.................. 106

Figure 6.18 Schematic of GO conformal coated Si slot waveguide. .................. 106

Figure 6.19 Mode distribution of GO coated Si wire waveguide. ...................... 107

Figure 6.20 Plot of simulated CE versus signal wavelength WLs....................... 108

Figure 6.21 The simulated CE versus pump power PP. ..................................... 108

Figure 6.22 The simulated CE versus waveguide length L. ............................... 109

Figure 6.23 The schematic of light-material interaction in a slot waveguide. .... 109

Figure 6.24 The simulated CE versus signal wavelength WLs. .......................... 110

Figure 6.25 The simulated CE versus pump power PP. ..................................... 111

Figure 6.26 The simulated CE versus waveguide length L ................................ 111

xxi

List of Tables

Table 2.1 Comparison of n2 of GO with different wavelength and thickness ...... 31

Chapter 1: Introduction

Yunyi Yang - January 2019 1

1 Introduction

In chapter 1, a background introduction to the subject, discussing the history and

future of two-dimensional (2D) materials, is presented in section 1.1. Section 1.2

introduces graphene-based materials including graphene and graphene oxide (GO).

Section 1.3 highlights the challenges in utilizing the functionality of graphene-

based 2D materials and the objectives of this thesis. At the end of the chapter, the

outline of the thesis is presented in section 1.4.

1.1 The rise of 2D materials

Graphene, first 2D material being successfully exfoliated from graphite in 2004 and

awarded the Nobel Prize in 2010, is regarded as one of the most attractive materials

of the last decade and the fascination for pursuing its properties and functionalities

has crossed over many fields in the scientific community.1,2 Following the

discovery of graphene, some efforts have been spread beyond graphene, for the

pursuit of atomically thin forms of other materials, such as boron nitride, black

phosphorene and transition metal dichalcogenides (TMDs) et al, as shown in Fig.

1.1. All these 2D materials offer endless possibilities for fundamental research, as

well as the demonstration of improved functionalities or even novel architectures

in both science and engineering fields.3



Figure 1.1 2D material family.4

Transition metal dichalcogenides (TMDs) are atomically thin semiconductors

of the type MX2, where a transition metal atom (M = Mo, W, Re et al.) is

sandwiched between two chalcogen atoms (X = S, Se, Te et al.).5 TMDs offer

something more than graphene: a bandgap, which makes them immediately suitable

candidates for semiconductor-based electronics and optoelectronics applications.

Additionally, the excitonic transitions in the ±K valleys (the local minimum and

maximum in the conduction and valence band, respectively) can be selectively

addressed with circularly polarized light, which has subsequently opened the

potential for the exploration of valleytronic devices, where the valley degree of

freedom is used to carry information.3,6

2D materials have enormous potential in terms of optoelectronics applications,

and in a broadband range of wavelengths — from the microwave range to the

visible. Graphene provides an excellent alternative to metal plasmonic due to the

combination of high intrinsic mobilities. Hexagonal boron nitride (h-BN) also

possesses many unique properties, for example, ideal dielectric; its in-plane

anisotropy renders it naturally hyperbolic, (the principal components of the

dielectric tensor have opposite signs), which could lead to hyperbolic phonon-

polaritons that do not suffer from losses. Although their lifetimes in multilayer h-

BN are comparable to those of optical phonons, the slow group velocity limits the

overall propagation length. Theoretical investigations indicate that other 2D

materials should also exhibit similar hyperbolic properties.3,7

The amazing and unique characteristics of 2D materials have stimulated a great

research effort to develop various applications and devices towards highly

integrated architecture. By simple stacking of multiple layers of a single material to



the integration of van der Waals heterostructures (Fig 1.2), novel functionalities like

diodes and field-effect transistors have triggered a lot of enthusiasm.3,4 Van der

Waals heterostructures are not limited to combinations of 2D materials. Recent

research shows that 2D materials can be combined with non-2D materials that

adhere primarily through non-covalent interactions. Interfacing 2D materials with

organic molecules and zero-dimensional (0D) quantum dots, one-dimensional (1D)

carbon nanotubes and bulk silicon (Si), Ge, III–V and II-VI three-dimensional (3D)

semiconductors can find applications in logic devices, photodetectors,

photovoltaics, and light-emitting devices.3,4,7,8 However, the production of high-

quality 2D material and assembly heterostructures over a large scale with

nanometre control, which is essentially for real-world applications, remains a

serious technological challenge.

Figure 1.2 Formation of the van der Waals heterostructures with various 2D

materials. The mechanically assembled stacks are shown on top of the figure and

the large-scale growth by CVD or physical epitaxy is shown on the bottom of this

figure.4



1.2 Graphene-based 2D materials

Graphene is named for a monolayer of carbon atoms depicted as a honeycomb

lattice or an assembled hexagonal mesh of carbon atoms. This kind of 2D material

often consists of sp2-bonded hybridized carbon atoms with exceptionally high

crystal and electronic quality. It is regarded as a novel ultra-thin material that has

emerged as a rapidly rising star in the material science and physics.2,9-11

Figure 1.3 In graphene, carbon atoms are bonded together through the sp2

hybridization. The shiny and flexible graphene paper is prepared by controlled

restacking of graphene sheets.10

Since the first discovery in 2004,1 graphene, regarded as the new class of 2D

carbon nanostructure, has been making a profound impact in various areas of

science and technology due to its remarkable physicochemical properties and

attracted tremendous attention from both the experimental and theoretical scientific

communities in recent years. Figure 1.4 presents the schematic of graphene-family

from 0D fullerenes, 1D carbon nanotube, and 2D graphene or layered to 3D

structure-graphite. Fig. 1.4 bottom right shows a crumbly substance that resembles

several layers of the weakly bonded graphene sheet. When graphene is wrapped

into rounded forms, it forms fullerenes. The family also includes honeycombed

cylinders known as carbon nanotubes (bottom row at centre) and soccer ball-shaped

molecules called fullerenes (bottom row at right), as well as various shapes that

combine the two forms.



Figure 1.4 Structures of graphene-based material. Graphene is a 2D building

block for carbon materials of all other dimensionalities. It can be formed into 0D

buckyballs, rolled into 1D nanotubes or layered to 3D graphite.2

The unique physicochemical properties suggest its great potential for providing

new approaches and critical improvements in many fields of science.11 Their unique

properties include a high specific surface area (theoretically 2630 m2/g for a single-

layer of graphene),2,12,13 extraordinary electronic properties and electron transport

capabilities such as giant carrier mobility,14-16 unprecedented pliability and

impermeability,17,18 strong mechanical strength12, chemical properties19,20,

excellent thermal13 and electrical conductivities21,22 and optical properties.23,24 The

rapid progress in the 2D material field resulted in raised interests in

multidisciplines.13,25-27 These unique properties have brought tremendous potential

applications in many technological fields such as nanoelectronics, sensors,

nanocomposites, batteries, supercapacitors and hydrogen storage.2

Numerous efforts have been paid to investigate the strategies to deposit

graphene-based materials such as chemical vapour deposition (CVD) on metal (Cu

and Ni) 28,29 as well as epitaxial growth on Cu30, silicon carbide31,32 and ruthenium33.

However, the costly and complex fabrication techniques limit paths toward large-

scale fabrication and its commercialization. In addition, the sophisticated transfer



process is required in current approaches that pose a hurdle for integrating graphene

with arbitrary surfaces. As a result, alternative methods that could produce

macroscale graphene sheets in large quantities with a cost-effective way are desired

but yet to be developed to boost the functionalities of the graphene family.34

Due to the fabrication limitations of graphene itself, GO, the oxygen-

functionalized and solution processable form of graphene, has been recognized as

a promising precursor for bulk production of graphene-like materials and devices.35-

40 This kind of material can be considered as a precursor for graphene synthesis by

either chemical or thermal reduction processes.

As early as 1859, graphite oxide was obtained by treating graphite with strong

oxidizers.41 Later, in 1957, Hummers and co-workers successfully developed an

efficient method for quick preparation of graphite oxide, which is still widely used

nowadays, known as the “Hummers’ method”.42 Since the carbon atom plane of

graphite oxide has been decorated by plenty of oxygen functional groups (OFGs)

which break up the extended 2D π-conjugation, the oxidized layers can be easily

exfoliated in water with the help of ultrasonication, forming a single layer of GO as

shown in Fig. 1.5.43 To date, the detailed structure of GO is still uncertain. The

reason is that the final structure varies with the synthesis method and the degree of

oxidation.37,44

Figure 1.5 (a) Schematic illustration of the preparation of GO from graphite. (b)

Schematic illustration of the structural model of a GO sheet.43



In the meantime, the OFGs in GO give rise to some remarkable properties. From

the structure prospect, GO can be regarded as the graphene plane covalently

decorated with various OFGs either on the basal plane or at the edge. This makes

the geometrical shape and composition of separate GO sheets different with each

other. The oxygenated groups in GO can strongly affect its electronic, mechanical,

and electrochemical properties. Hence they account for the differences between GO

and pristine graphene.45 The presence of these OFGs can also offer extra potential

advantages for using GO in numerous other applications. The polar OFGs of GO

render it strongly hydrophilic. This gives GO good dispersibility in many solvents,

particularly in water,13,46,47 which is accessible for processing and further

derivatization. The GO-stable dispersion can be subsequently deposited on various

substrates to prepare multifunctional thin films by solution-phase methods.

Recent research efforts have been focused on the preparation of GO thin films.

The approaches such as drop-casting, dip-coating, spraying and spin-coating are

explored to provide a one-step solution to coat GO materials onto planar surfaces48.

However, such procedures result in cracks, wrinkles and non-uniform dispersions

caused by the poor control of the process, which lead to unsatisfactory film

thickness, roughness and quality. Consequently, the optical properties of the

prepared GO film are degraded due to the uncontrollable morphology. Therefore,

alternative GO preparation approaches are eagerly required for high quality and

precise control.

Furthermore, the integration of 2D material with functional nanostructures such

as dielectric and metal materials provides novel opportunities for exploring

enhanced performance and broad applications by designed 3D architecture with

unique properties.49-53 For example, conformal coating of thin 2D materials on

designed nanostructures plays the key role to improve both the optic and electronic

performances of hybrid structures by utilizing the large surface area and unique

morphology nature of Si nanostructures.54,55 The well-coated hybrid structures pave

the way for novel applications and enhanced performances in devices such as

photovoltaic devices, photodetectors and integrated Si photonics.56-59 In many

optoelectronic device applications, conformal coating of functional nano-films is

required to achieve the desired functionalities. For example, in Si waveguide, the

high-quality conformal coating is the key to achieve desired mode distribution and



leading to low propagation loss. However, the existing GO deposition methods can

only coat on top of the architecture in particular when they are of a few nanometres

in size. The lack of precise control in resulted features limit the optic and electronic

performance of the hybrid coated structure. Therefore, the lack of an efficient

method to realise uniform conformal coating of graphene-based 2D materials onto

functional nanostructures with a low-cost and scalable method is the major obstacle

to exploiting potential applications of the hybrid devices.

Moreover, GO could be reduced by removing the OFGs and then the conjugated

structure could be partially recovered, which provides graphene-like properties.37

Enormous research efforts have been dedicated to the reduction of GO, as

evidenced by the vast body of related publications.37,60-62 Previous studies have

confirmed that residual functional groups and defects dramatically alter the

structure and properties of reduced GO with respect to pristine graphene63,64. But

the reduction methodologies for GO are still promising due to the advantages of

large-scale production of graphene-based materials, tractable processing, and the

modulation of electronic properties.37 In this regard, the reduction of GO and large-

scale fabrication capability is of great importance to the development of graphene-

based devices.25,37 Moreover, the optical and electrical properties of GO can be

tailored by manipulating the relative content of the OFGs of GO during the

reduction process, which enables numerous electronic and optoelectronic

applications including transparent conductive electrodes, field-effect transistors,

thin film transistors, field emitters, photovoltaic devices, light-emitting devices,

flexible electronic materials and electrical sensors.

1.3 Objectives of the thesis

Broadened the applications and precisely controlled preparation process have been

the major driving forces for the development of graphene-based material. However,

the lack of an efficient approach for producing a graphene-based platform in large

quantities and macroscale with a precise control methodology remains the major

challenge on its way to functionalization and real-life applications.

Firstly, the low productivity of mechanical exfoliation and sophisticated

conventional deposition processes of graphene are not suitable for large-scale

fabrication and commercialization. Moreover, the inevitable transfer process will



further damage the profile of the graphene sheet and it is difficult to align the

targeted area that becomes the obstacles for device fabrication.65 Secondly, the

solution-phase methods based on GO suspension fail to provide high quality and

uniform dispersion of the film due to the poor control process.11 These will severely

degrade the properties of the prepared GO film, especially for the photonic

applications. Furthermore, the existing GO deposition methods can only coat on top

of the nanostructures and the non-precisely controllable feature hinders the optic

and electronic performance of the hybrid coated structure. This will limit its

applications for novel hybrid devices.

The objectives of this thesis to provide breakthrough solutions for the above

challenges that lie across the way to manufacture large scale, ultrathin and high-

quality graphene-based materials film with accurate control of the quality, including

the film thickness, roughness and integratability, and find out a practical way to

realize it. The developed method could present high controllability and integrate

graphene-based material with the functionalized substrate without transfer process.

The process could be further manipulated to achieve conformal coating onto

designed nanostructures.

This thesis proposes a modified solution-based layer-by-layer (LBL) deposition

process using GO suspension with precise control of the flake size, concentration,

deposition time, and eventually could achieve a well-controlled film thickness

down to nanometric scale. Thus, a graphene-based platform is established for

photonic applications. By further developing the deposition environment and

process, the LBL method is able to conformally coating onto various nanostructures,

for example, the coating of Si nanostructures have been demonstrated in this thesis.

Optical devices based on the phase and amplitude modulations are enabled in this

graphene thin film platform combined with laser-patterning. At last, we further

expand this versatile platform for realizing novel hybrid photonic devices with

enhanced performance. To this end, photovoltaic devices and all-optical

communication devices have been demonstrated.

1.4 Outline of the thesis

Chapter 2 reviews the state-of-the-art film preparation methods for GO

materials as well as photonics applications of graphene-based materials. Firstly, the



thin-film preparation methods of graphene-based materials are reviewed focusing

on the complexity, scalability, integratability and controllability. Followed up are

the different reduction methods and fabrication techniques to achieve property

tuning and functional patterns in GO films. Moreover, optical properties including

the linear and nonlinear properties and the corresponding allocation of graphene-

based material have been reviewed. Challenges and opportunities are summarised

following each literature review section, bringing up the questions for the following

up chapters in this thesis.

Chapter 3 focuses on the study of ultra-thin GO film preparation method.

Firstly, the solution-based LBL approach has been developed for ultrathin GO film

platform. The characterization of the prepared film is presented. The large-scale

film preparation with different substrates is further demonstrated. In the end,

multilayer film structures and their optical properties are characterized.

Chapter 4 investigates the GO conformal coating onto Si nanostructures with

the LBL deposition method. Ultrathin GO film conformal coating on Si nanowires

(SiNWs) has been achieved by tuning the properties of the GO solution including

flake size and surface hydrophilicity status. Then the optical properties of this

coating are characterized, which help to reduce 20% reflection along with a

broadband wavelength range. Furthermore, the GO conformal coating is applied to

textured Si solar cell. Up to 60% enhancement in carrier lifetime measurement and

a 20% improvement in external quantum efficiency (EQE) of the hybrid solar cells

are achieved.

Chapter 5 realizes localized laser-reduction and patterning of LBL deposited

GO films. The optical properties and bandgap can be manipulated dynamically by

controlling the laser power. The changes of refractive index and surface

conductivity by femtosecond (fs) direct laser writing have been well studied.

Several photonic devices based on the amplitude and phase modulations by taking

advantages of both the laser fabrication technology and optical properties of GO

are demonstrated. At last, a microfluidic bio lens has been realized, which paves

the way for biological applications.

Chapter 6 proposes GO hybrid integrated photonic devices. We demonstrate

enhanced four-wave-mixing (FWM) in doped silica waveguides integrated with



GO. Owing to the strong mode overlap between the integrated waveguides and GO

films that have a high Kerr nonlinearity and low loss, the FWM efficiency of the

hybrid integrated waveguides can be significantly improved. The results show good

agreement with theory and achieve up to ~9.5 dB enhancement in the FWM

conversion efficiency for a 1.5 cm long waveguide integrated with 2 layers of GO.

Finally, the theoretical simulations show that the FWM efficiency in Si waveguides

(nanowire and slot waveguides) incorporating GO films can be enhanced by more

than 35 dB. This demonstrates the effectiveness of introducing GO films into

integrated photonic devices to enhance the performance of the nonlinear optical

process.

Chapter 7 summarises the research outcomes of this thesis. The outlook and

future work have also been discussed.

Chapter 2: Literature review


2 Literature review

2.1 Introduction

The rise of two-dimensional (2D) material, graphene-based material such as

graphene oxide (GO) and reduced graphene oxide (rGO), have attracted a great deal

of attention due to their unique physical and chemical properties. A number of

optoelectronic applications have been demonstrated with such materials. The high

demand in photonic applications calls for thin-film deposition of the material

towards actual device realization. High quality, large scale and cost-effectiveness

are the demand for commercialising 2D material platform for real-life applications.

Specifically, for photonics applications, the challenge is to maintain the accurate

control (down to nanometre scale) as well as to achieve scalability of large scale

fabrication at the same time.

Moreover, the optical properties of graphene-based 2D materials are

fundamental information for designing various optical components for different

optical applications. Compared with the broadly investigated electrical properties,

the optical properties of GO or rGO are relatively less explored. This specific nature

of GO offers us a chance to manipulate its optical properties by changing the content

of oxygen functional groups (OFGs) via the reduction process. Various methods

have been applied in the GO reduction field including thermal, chemical and photo

approaches. Furthermore, patterning during the reduction process is eagerly desired

for one-step device fabrication.



In this chapter, the thin-film preparation methods of graphene-based materials

have been reviewed firstly. Then the reduction methods for graphene-based

material are reviewed. The optical properties including linear and nonlinear

properties of GO film will be followed up. Specifically, the nonlinear optical

properties including nonlinear absorption and Kerr nonlinearity have also been

introduced, revealing the rich optical responses of the materials under high-intensity

laser illuminations. Moreover, based on the linear and nonlinear optical properties,

two potential applications on graphene-based material have been reviewed. Finally,

challenges and opportunities are summarised following each of the section

respectively, raising the scientific questions that will be addressed in this thesis.

2.2 Graphene-based thin-film preparation approaches

The research enthusiasm in 2D materials field has grown extraordinary during the

past decade.28 Nowadays, the core interest of 2D material has gradually shifted from

fundamental science to potential technological applications.65 Thin-film

preparation approaches of graphene-based materials, which are the key

prerequisites for the design and fabrication of the functional device, in particular

optoelectronics, are the necessary foundations for industry applications and

commercialization. Figure 2.1 shows the existing strategies for graphene film

deposition. A few key parameters are considered, including the size, quality, cost

and possible applications. Cost for mass production and film quality are

contradictory requirements in the graphene-based material preparation process.

Mechanical exfoliation can lead to a high quality but it is not suitable for mass

production. Liquid-based methods are cost-effective at mass production but have a

relatively low quality. To develop a film preparation method that could

simultaneously satisfy these requirements and offer a new avenue for graphene-

based material applications remain a key challenge. In this section, graphene film

deposition methods are reviewed first. The review of solution phase methods based

on GO suspension is followed up. Challenges and opportunities are discussed at the

end.



Figure 2.1 Road map of current graphene deposition methods, which offer the

choices for potential applications based in terms of size, quality and price.65

2.2.1 Graphene thin-film preparation approaches

Since the first isolated graphene was successfully prepared by mechanical

exfoliation of graphite,1,38,66,67 several approaches including epitaxial growth on

silicon carbide via silicon sublimation, 25,31-33,68 and chemical vapour deposition

(CVD) growth from carbon-based precursors on metal substrates such as Cu,

Ni28,29,37 have been developed.

The first straightforward approach to fabricate monolayer graphene is the

mechanical exfoliation method or scotch tape method shown in Fig. 2.2. The

method separates high purity graphite into very thin, few-layer graphene by using

a scotch-tape and repeatedly peeling flakes of graphite off the mesas.1,67 Due to its

low yield and long processing time, other deposition methods have been

investigated.



Figure 2.2 Mechanical exfoliation of graphene by scotch tape. 67

Epitaxial growth graphene could be prepared by annealing a SiC substrate at

around 1300 ºC under ultra-high vacuum, typically at pressures below 1×10-10 Torr.

Prior to that, the samples should be etched in hydrogen (1 bar, 1550ºC, 30 min) in

order to achieve a clean surface without polishing damage. Deposition of potassium

on the sample surface was carried out in ultra-high vacuum using a commercial

source (SAES Getters/USA. Inc). The morphology of epitaxially grown graphene

film on Ru substrate is shown in Fig. 2.3, the graphene flakes can be observed in

Fig 2.3 (a) and (b).33 However, the two major drawbacks of this method are the

high cost of the SiC wafers and the high temperatures (above 1000 ºC ) used, which

are not compatible with current Si electronics technology and needs extra transfer

process.65

Figure 2.3 Morphology of epitaxially grown graphene on Ru substrate.33

Chemical vapour deposition (CVD) deposition method has been used for

preparing graphene films. The developed graphene CVD growth process on copper



foils is introduced. The films grow directly on the surface by a surface catalysed

process. Graphene is grown on copper foils at temperatures up to 1000°C by CVD

of carbon using a mixture of methane and hydrogen. Fig. 2.4 (A) is the SEM image

of graphene on a copper foil with a growth time of 30 min. High-resolution SEM

image showing a Cu grain boundary and steps, two- and three-layer graphene

flakes, and graphene wrinkles in Fig. 2.4 (B). Inset of Fig. 2.4 (B) shows the TEM

images of different layers of graphene edges. Graphene films transferred onto a

SiO2/Si substrate is shown in Fig. 2.4 (C) and a glass plate is in Fig. 2.4 (D).29

Figure 2.4 Characterization of a CVD grown graphene film on a copper foil.29

(A) SEM image of graphene on a Cu substrate. (B) Zoomed SEM image of a Cu

grain boundary and steps, two to three layers graphene flakes, and graphene

wrinkles. Inset is the TEM images of folded graphene edges (1L: one layer; 2L:

two layers). Graphene films are transferred onto a Si substrate (C) and a glass

plate (D).

The schematic of preparation and PMMA-assisted transfer of graphene

monolayer is shown in Fig 2.5. PMMA is coated on the as-prepared samples

(graphene on copper) and immersed into dilute nitric acid for a couple of hours to

remove the copper, leaving a freely floating PMMA/graphene on the liquid surface.

Then the PMMA/graphene films are transferred onto SiO2 (glass) substrates and the

PMMA layer can be removed by using acetone.56



Figure 2.5 The schematic of preparation and PMMA-assisted transfer of

monolayer graphene.56 Cu foil is washed and baked firstly (a), and then graphene

is deposited on Cu by CVD (b). Secondly, the polymer is coated onto graphene as

a substrate (c) and removed Cu (d). Finally, the film is transferred to a SiO2

substrate (e) with removing the polymer layer (f).

Further development of CVD deposited graphene has realized roll-to-roll

graphene films for transparent electrodes shown in Fig. 2.6. Bae et al.69 report the

roll-to-roll production and wet-chemical doping of predominantly monolayer 30-

inch graphene films grown by CVD onto copper substrates. The films have sheet

resistances as low as ∼125 Ω square with 97.4% optical transmittance. The multiple

transfers and simple chemical doping of graphene films considerably enhance its

electrical and optical properties. Then the fabricated graphene electrodes have been

incorporated into a functional touch-screen panel device capable of sustaining high

strain. Although the CVD method has been well-studied for graphene deposition,

the transfer process that is as sophisticated as the stringent deposition requirement,

restriction to metal substrates, large energy consumption and the expensive process

will limit its further applications.



Figure 2.6 Schematic of the roll-to-roll production of graphene films on a copper

foil. The whole process includes attachment of polymer supports, copper etching

and dry transfer procedure on a target substrate.69

2.2.2 Graphene oxide film preparation approaches

Due to the low productivity and sophisticated facility requirements of graphene film

preparation, graphene oxide has attracted enormous attention in recent years. Being

regarded as the alternatives of graphene for easy preparation and cost-effective

production, GO, offers more viability of manipulation for tailored demands.

Different methods have been introduced to easily prepare GO films for bulk

production, such as spin-coating35,70, dip coating/drop casting71,72 and vacuum

filtration26,38,73.

Owing to the unique optical and electrical properties of monolayer and few-

layer graphene, it is highly desired to develop thin film preparation process with

monolayer controllability,34 in particular, for stringent photonic applications, which

require light manipulation on a nanometre scale. Under such a circumstance, the

requirements of the thin film are heavily focused on the high quality including

nanometer thickness and surface roughness of the graphene film. On the other hand,

for practical applications, the insufficient light modulation of monolayer graphene

has placed a high demand for designable multilayer graphene film and even layered

metamaterial or heterostructures to be developed.

The spin-coating method is one of the promising approaches, which is now

frequently used for the fabrication of GO-based devices and has proven for much

better control on defect density as well as overall thickness and roughness of the

GO film layer. The surface roughness of the spin-coating GO film is down to 100

nm. The thickness can be stably achieved from hundreds of nanometre to

micrometre scale.38,74



The standard spin-coating process is illustrated in Fig. 2.7. First, depositing a

small amount of a liquid GO suspension onto the centre of a selected substrate.

Then spinning the substrate at required speeds. The GO solution will be spread to a

large area. Moreover, the spin-coating method allows a full range of coverage

densities by fine-tuning of the spin-speed and applying pre-treatment to the

substrate surface. This method enables one to prepare continuous large-area films

with few nanometre scale wrinkling, as was found in other GO preparation

procedures.11 Figure 2.7 depicts the demonstration of the spin-coating process. The

GO solutions for film preparation (Fig. 2.7(a)) are synthesized by a self-assembly

method. High quality continuous GO thin films (See inset of Fig. 2.7(c)) with

controlled thicknesses could be obtained through a spin coating process (Fig.

2.7(b)) on a cover glass. Figure 2.7(c) presents the optical microscope image of the

as-prepared GO thin films over a large scale.74 For the spin-coating method, the

film thickness and homogeneity depend on the concentration of GO solution,

parameters of the spinning speed and spinning acceleration, and the number of spin-

coating cycles. Each parameter could affect the film quality and is hard to control

accurately in practical operation. All these lead to the poor controllability in film

thickness and homogeneity.

Figure 2.7 (a) GO suspension prepared by a self-assembly method. (b) Schematic

of the spin coating process. (c) Optical microscope image of the as-prepared GO

thin film over a large area. Inset is the surface morphology of the GO layer.74

Vacuum filtration has also been introduced to deposit uniform GO layers for

the fabrication of GO-based optical and electrical devices.26,72 This approach

involves filtering the mixed solution with GO flakes through a porous membrane,

such as an anodic aluminium oxide (AAO) membrane, a polyethylene terephthalate

(PET) membrane, or an anodic membrane filter. 11,26,38,73



Figure 2.8 presents the schematic process of the vacuum filtration process and

optical characterization of the prepared film. The schematic setup is shown in Fig.

2.8 (a). The vacuum filtration process is explained below. First, the filtration of a

GO suspension through an AAO or PET membrane. As the GO suspension passing

through the porous membrane, the GO flakes are filtered by the membrane, forming

high quality GO thin films with a desired thickness on top of the membrane, as

shown in Figs. 2.8 (b)-(c). Then the filtrated uniform GO thin films are transferred

from the membrane (Fig. 2.8 (c)) onto various desired substrates, such as the glass

slide (Fig. 2.8 (d)) or silicon, by either peeling off the film or dissolving the

membrane with the aid of aqueous solution such as water. The vacuum filtration

method could achieve the GO films thickness various from single layer to several

layers.38 The control over the film thickness can be achieved by simply tuning either

the concentration of the GO solution or the overall filtration volume.10,26,34,36,75

These two parameters could only offer very coarse control of the film thickness and

roughness, which cannot meet the monolayer control requirement of the prepared

film. Moreover, the vacuum filtration method needs the transfer process, which will

further damage the surface profile and result in defects such as wrinkles and cracks

on the film plane.



Figure 2.8 (a) The schematic of the filtration setup and the (b) vacuum filtration

process. (c) The filtrated GO thin films on the filtration membrane. (d) The

transferred GO thin film on the glass substrate. (e) The AFM images and (f) the

SEM image of the GO thin film.75

2.2.3 Challenges and opportunities

Although considerable attention and efforts have been devoted to the investigation

on the preparation methods for graphene-based materials since its first discovery,

challenges and obstacles still hinder the way to prepare high quality and large scale

graphene-based material and film with a monolayer control deposition process.

Firstly, for the existing graphene thin-film preparation methods, high-quality

pristine graphene can be prepared by the Scotch tape method for small-scale

production1. However, mechanical exfoliation of graphite leads to random sized

small flakes, which are challenging to achieve desired layer numbers. Epitaxial

growth on silicon carbide with silicon sublimation25,31,68 and chemical vapour

deposition (CVD) method on metal substrates have been demonstrated for



production of monolayer graphene28,29,37, but these methods place critical

requirements such as high temperature and high vacuum on the growth conditions

and substrates. More importantly, In addition, simply accumulating graphene layers

via the conventional film deposition methods leads to its bulk counterpart-graphite.

Thus, the unique atomic properties are no longer preserved. 76,77 These drawbacks

limit the large-scale production, functional device fabrication and industry

applications.

Secondly, materials are vulnerable and unstable in such an atomic scale, the

inevitable transfer process associated by other materials and etching or rinsing

process will further damage the ultra-thin profile such as cracks and wrinkles and

degrading properties of the material that is not acceptable in photonic

applications.1,65 Moreover, the current transfer process fails to provide localized and

precise transfer with the desired position and conformal coating onto designed

functional nanostructures.

Thirdly, GO is risen up due to the solution processable preparation and bulk

production that has attracted lots of interest. One step film preparation process

without transfer such as spin-coating, dip-coating and drop-casting make it a

promising candidate for photonics and electronics applications.11 However, the lack

of control in the film thickness and roughness will obstruct its future applications.11

The vacuum filtration method can offer limited control during the process based on

the solution volume, concentration and morphology of the membrane. Furthermore,

such a process is very slow and only applicable for small scale laboratory setting,

which greatly limits its scalability and productivity.38 In addition, the need for a

complex transfer process will also damage the material leading to defects such as

cracks and wrinkles shown in Figs. 2.8 (e) and (f). And the long preparation time

results in aggregation of the GO material that cannot keep the GO layers within

nanometre scale.

Last, all of the methods above failed to provide high-quality film deposition on

target substrates or structures with high accuracy of controllability. And the

graphene-based film could only float on top of the nanostructures when a hybrid

system is developed, which is unable to realize the conformal integration with

designed functions of the nanostructures.



Therefore, it is highly desirable to develop a game-changing film preparation

technique, which is simple, low cost, scalable, integratable without transfer and

with nanometre accuracy of controllability.65 However, such a technique is missing

and will be one of the major foci of this thesis.

2.3 Photoreduction methods for graphene oxide

2.3.1 Review

GO, in respect of mass-production and the capabilities in solution-processing,

generally prepared by chemical oxidation of graphite and subsequent exfoliation in

aqueous dispersion, are widely regarded as an effective route to produce graphene-

based materials. However, the OFGs on the graphene basal plane significantly

affects its properties, especially electronic and photonic properties. Therefore,

reduction or manipulation methods that are used to selectively remove OFGs

become critical. Figure 2.9 presents the SEM image of the GO stacking layers (a)

and an optical image of a chemically reduced GO film. Recently, in addition to the

conventional thermal and chemical reduction, photoreduction has emerged as an

appealing alternative by neither rely on high temperature nor toxic chemicals. As is

discussed in this section, the rapid development of photoreduction of GO is

classified into two major categories: photothermal reduction and photochemical

reduction.37

Figure 2.9 Free-standing graphene films show extremely high tensile strength. (a)

Cross-sectional SEM image of GO stacking in a film produced by filtration.36 (b)

Chemical reduced GO film shows a metallic-like shiny lustre.34



Considering the controllable and localized reduction of GO, photoreduction

approaches have emerged as promising alternatives to the conventional thermal and

chemical methods. Cote et al. reported flash reduction and patterning of GO.78 A

commercial Xenon lamp on a digital camera was used for the flash reduction of GO

and GO/polymer films under ambient conditions. The removal of OFGs could be

attributed to the photothermal effect because the camera flash provides mainly

visible (wavelength over 400 nm) light. Figure 2.10 shows the images of

photothermal reduction of a GO film by flashlight.

Figure 2.10 Flash reduction process of GO. (a) Optical images of GO, flash

reduction of GO and RGO, from left to right. (b) Image of the arrays of

rGO/polystyrene interdigitated electrodes fabricated on a 1.5-inch diameter

GO/polystyrene thin film deposited on a Nylon filter paper. Inset: the close-up

view of one set of such electrode.78

In addition to the photothermal approach, GO could be reduced via a

photochemical scheme as well. The photogenerated electron-hole pairs have been

verified to be the general mechanism for the removal of OFGs in this scheme,

leading to the restoration of sp2 domains. Kamat's et al. reported GO reduction by

UV irradiation with the help of photocatalyst TiO2.62 On the other hand, GO could

still be effectively reduced under UV irradiation without the help of any

photocatalysts. Matsumoto et al. presented the UV reduction of GO through a 500

W high-pressure Hg lamp where ultra-pure H2 or N2 flowed via a quartz cell.79 The



rGO sheets reduced by this method show a significant decrease of OFGs and

partially restored sp2 domains.

Laser reduction, proceed via a combination of both photochemical and

photothermal mechanisms, has been introduced for GO reduction due to the unique

advantages such as reliability, amenability, low cost, and flexible patterning.80 The

photoreaction types for various laser reductions could be deduced according to the

threshold effect corresponding to the wavelength.81 In principle, reduction by a laser

with a wavelength smaller than 390 nm, such as an excimer laser (248 nm), would

mainly undergo a photochemical process, for lasers with wavelengths larger than

390 nm, the photothermal effect is mainly accounted for the reduction of GO.37

Zhou et al. firstly presented a “direct laser writing” (DLW) technique for cutting

and reducing multilayer GO films, using a continuous wave diode laser (663 nm,

80 mW).82 In addition to continuous wave laser, ultrafast pulse laser has also been

used for GO reduction. Zhang et al. reported the direct imprinting of graphene

microcircuits by direct femtosecond laser writing induced reduction of GO under

ambient conditions with programmed patterns (Fig. 2.11).83 The height of laser

reduced area is lower than the rest of the GO film due to the removal of the OFGs.

The patterned GO was synchronously reduced and thus possessing good

conductivity for electrical applications. Zheng et al. reported a GO ultrathin (~200

nm) flat lens based on laser direct patterning and laser-induced reduction by a

femtosecond pulsed laser (100 fs pulse, 10 kHz, 800 nm).84 The surface morphology

and optical properties such as optical constants are manipulated by the laser

fabrication process.



Figure 2.11 Fabrication process and optical microscopy images of reduced and

patterned GO films. (a) Schematic and conceptual design of the GO microcircuit.

(b) - (d) Different designed patterns of GO microcircuits. (e) Microscopy image of

the university logo.83


Although tremendous effects have been devoted to the reduction of the GO film, it

is still challenging to achieve precisely controlled in situ reduction extent with the

patterned structure at the same time. For fine-tuning the optical properties such as

optical constants and bandgap of GO film, a dynamic controlled reduction process

is still required for the deep investigation to reveal the fundamental physical

changes.

2.4 Optical properties of graphene-based material

2.4.1 Linear optical properties

Intensive investigations have been focused on the electrical properties of graphene-

based materials including field effect, tunable electric conductivity and quantum

Hall effect.1,15,85 However, the studies of optical behaviours have been hindered

because of the challenges in high-quality film preparation, ultra-thin film thickness,

dielectric anisotropy, small flake size, random distribution of the GO flakes on the



substrate, and the limited optical characterization approaches. As the fundamental

optical parameter to identify the responses of light-materials interaction, the optical

constants including refractive index and extinction coefficient as a function of

broadband wavelength are key parameters for the photonic applications of GO film.

In addition, the reduction of GO films, as the selected strategy for manipulating the

GO film studied here, proves to generate complicated effects on the GO film, and

therefore result in different dispersion properties of the reduced GO films.

Therefore, in-depth investigations on the optical properties of GO and rGO

materials are yet to establish. Studies that focused on the optical characterization of

graphene-based films are meaningful for diverse photonic device, design

fabrication and applications.

Optical microscopy, regarded as a direct optical characterization method, is

used to measure the linear optical properties of GO and rGO films. The difference

between light reflected from the thin film and the substrate can be greatly enhanced

by a properly designed substrate under monochromatic illuminations. A

quantitative calculation of the complex refractive index of the thin sheet can be

obtained by the Fresnel’s equations. The effective refractive index (n) and

absorption coefficient (k) of GO films and thermally reduced rGO films by a

standard confocal microscopy were reported by Jung et al..86 A single layer of GO

can readily be identified with optical microscopy and the optical constants are

achieved by comparing the measured difference. In addition, the complex refractive

index of rGO increases compared to that of GO due to the thermal reduction.

However, it should be noted that such optical constants can only be measured at a

few discrete wavelengths without getting the overall dispersion relations.

The spectroscopic method was applied to obtain the refractive index of GO

sheets.87 The reflection spectrum of GO was compared with the spectrum

subtracting the background. Then, a contrast spectrum can be achieved and

analyzed by using the Fresnel’s equations. Other researcher performed terahertz

(THz) spectroscopy (time-domain) on thermally reduced GO. The refractive index

was measured in THz range with respect to the Drude free-electron model.88

Moreover, the demonstration of the manipulation of the refractive index of rGO can

be manipulated by controlling the reduction process was achieved.84



Spectroscopic ellipsometry is another major approach that is widely used to

determine the optical parameters such as film thickness and complex refractive

index by a nondestructive mechanism. Zheng et al. reported that optical constant of

refractive index (n) and extinction (k) of GO and laser reduced GO by spectroscopic

ellipsometry with a 200 nm GO film.84 The dispersions relations n of rGO at

different laser powers measured by ellipsometry are shown in Fig. 2.12 The

measured dispersion relations k of rGO at different laser powers are illustrated in

Fig. 2.13 The measured refractive index and the extinction coefficient of

single/multi-layer GO and thermally reduced GO by imaging spectroscopic

ellipsometry over the broadband wavelength range from 350-1000 nm was reported

by Jung et al..89 An increase in the refractive index (n) and extinction coefficient (k)

of GO after thermal treatment was observed. Shen et al.90 characterized the optical

responses of GO and chemically reduced GO with the standard spectroscopic

ellipsometry as well. By fitting with the Lorentz oscillator model, the bandgap

tuning through the reduction process of GO is investigated in details. The rGO

conductivity was found to increase owing to the sp3 to sp2 transition of carbon atoms.

Figure 2.12 The dispersion relations of refractive indices (n) of rGO at different

laser powers have been measured by using ellipsometry.84



Figure 2.13 The dispersion relations of the extinction coefficient (K) of rGO at

different laser powers have been measured by using ellipsometry.84

2.4.2 Nonlinear optical properties

Optical nonlinearity is an essential property especially for many potential

applications including all-optical communications, high-speed switching, optical

limiters and saturable absorbers.91 Materials with such nonlinear properties will

stimulate the applications of 2D material in integrated photonics field. Enthusiastic

attention has been attracted by the nonlinear optical properties of graphene-based

materials nowadays.92

Zheng et al.74 investigated the in situ nonlinear responses of the GO thin films

prepared from chemically synthesized GO solution by the spin-coating method.

This study covers the entire laser reduction process through the continuous increase

of the laser irradiance until the optical breakdown (Fig. 2.14). Four stages of

different nonlinear behaviours have been discovered with the increasing of laser

irradiance. Not only the tuning of the nonlinear absorption response but also a

switch sign of the nonlinear refractive index are observed during the transition

process from GO to rGO. Most importantly, the giant Kerr nonlinear activities are

observed to be three orders of magnitude larger than those in the previous reports,

which leads to a giant nonlinear figure of merit (FOM). Both the Kerr nonlinear

coefficient and FOM are essential for functional nonlinear photonic device design.

The above results have both explored the large nonlinear responses of GO thin film

during its reduction to rGO and proved the tunability of the nonlinear properties of

GO films for highly integrated nonlinear photonic devices on a thin film.



Figure 2.14 (a) Output fluence versus input fluence (scatters). Four different

stages (I, II, III and IV) are labelled. T: the linear transmittance (black dash line).

MD1 and MD2 represent the modulation depth in stages II and IV, respectively.

(b-e) Power dependent open aperture Z-scan results (scatters) fitted with

corresponding theory given in the reference (black solid lines). Inset: schematic

atomic structure of GO films in different stages. (f) Raman spectra of GO and

fully reduced GO films. Inset: a figure of laser-induced reduction process.74

The different nonlinear mechanisms of GO solutions in nanosecond and

picosecond regimes were reported by Liu et al..93 The findings show that two-

photon absorption (TPA) dominates the nonlinear absorption process of GO

solutions with picosecond pulses, while excited state absorptions (ESA) are

responsible in the case of nanosecond pulses. Jiang et al.94 presented the broadband

optical limiting (OL) behaviours of the GO film and laser-induced rGO for

femtosecond laser pulses at both 400 nm and 800 nm. The laser-induced rGO

showed enhancement of effective TPA coefficient by up to 19 times. Also, the

highly reduced GO film by chemical method displayed strong saturable absorption

(SA) behaviour. Liaros et al.95 reported the nonlinear optical responses of some

aqueous GO colloids under visible (532 nm) and IR (1064 nm), picosecond and

nanosecond laser excitations, which present large broadband OL, SA, and

negligible nonlinear refraction (NLR). And SA was found for the low incident

intensity, whereas reverse saturable absorption (RSA) displayed at the higher

intensity.



Table 2.1 Comparison of n2 of GO with different wavelength and thickness

GO74 GO75

Wavelength (nm) 800 1550

GO Film thickness ~2 μm ~1 μm

n2 (m2/W) 7.5 × 10 –14 4.5 × 10 –15

n2 normalized to

silicon nitrideb 3 × 105

1.8 × 104

bn2 for silicon nitride = 2.5 × 10 –19 m2/W 96.

The comparison of the n2 with different thicknesses at a various wavelength

(800 nm and 1550 nm) is shown in Table 2.1. The measured results of n2 show that

GO has a high Kerr nonlinearity with broadband wavelength especially the strong

nonlinear behaviour in the communication band, which is around 1550 nm. The

nonlinear property in 1550 nm wavelength enables GO be a promising candidature

for telecommunication and integrated photonics applications.


Although researchers have been exploring the linear and nonlinear optical

responses of GO and rGO for the past decade, a few topics are still unclear, but

critical for GO photonics.

For the linear properties of GO, firstly, few results have been reported on the

linear relations of optical constants for laser-induced rGO films so far, especially

when the film thickness is in a nanometre scale. Compared with the conventional

thermal or chemical reduction, laser reduction is preferred for flexible

manufacturing and localised manipulation of material properties. Thus, the linear

refractive index of the laser-induced rGO is supposed to be different from that of

thermally or chemically reduced ones owing to the different reduction mechanisms,

which lead to different resultant material morphology. Also, laser reduction of GO

has the advantage to precisely control the reduction extent and therefore control the



linear refractive index of rGO. As a result, the linear relations of optical constants

of the laser-induced rGO films at different reduction extents are eager to be

investigated not only from a fundamental research point of view but also for

practical applications. The technical challenge is the small sample size of the laser-

induced rGO films (hundreds of microns normally limited by the scanning stage

range), which is hard to measure by standard ellipsometry techniques (usually

requires millimetre sized samples).

Secondly, the linear relations of optical constants of GO or rGO have not been

modelled theoretically at a nanoscale thickness during the ellipsometry

characterization process. Although, the dispersion relations of GO or rGO have

been characterised by various fitting models such as Cauthy, Drude and B-spline,

the model is still needed to be modified when the thickness of the film is down to a

nanometre scale. Moreover, the unique layered structure of graphene-based

material should be taken into consideration rather than using the model for bulk

material.

For nonlinear properties, firstly, for integrated nonlinear photonic device

design, the third-order nonlinearity, in particular, the n2, is crucial for novel

functionalities including all-optical switching, fast optical communications and

signal regeneration. However, the research on nonlinearities of GO is heavily

limited to the nonlinear absorption such as OL or SA for modulating the light

intensity. Although experimental investigations have been reported on the Kerr

nonlinearity in GO suspensions or composites, which are not directly relevant to

the properties of a GO film and challenging to be used for actual integrated photonic

devices. Few results have been reported for Kerr nonlinearity of GO films

especially in the telecommunication band around 1550 nm.

Secondly, for the demand on increasing the optical nonlinearity of GO or GO-

based hybrids, the required large interaction area of light with the nonlinear media

such as waveguide and ring resonator can be significantly reduced with a larger

nonlinear coefficient, enabling a high level of the contact footprint. This is

particular critical for nonlinear materials with a significant linear loss, such as GO.

As a result, an ultrathin film with less optical absorptions and ultrahigh

nonlinearities is required to ensure sufficient signal transmission through the

devices. And the conformal coating technique, which enables the GO film tightly



attached to the 3D waveguide and thus enlarge the interaction area, is eagerly

desired for GO integrated nonlinear hybrid device.

2.5 Graphene-based functional devices

2.5.1 Graphene-based linear photonic devices

Benefitting from the unique optical properties, graphene-based materials have been

widely explored in photonics applications. By given the linear and nonlinear optical

properties of graphene-based materials such as GO and rGO, it is appropriate to

realize various ultrathin flat optical devices on graphene-based materials by

exploring the flexible patterning capabilities of laser fabrication technology as well

as the fully controlled optical properties during its laser-induced reduction process.

A GO flat lens (Fig. 2.15) is fabricated by the concentric ring with sub-micron

thickness fabrication on GO film prepared by a spray method (thickness: 200 nm)

using the mask-free DLW method to covert the GO into rGO through the

photoreduction process. Finally, a 3D focal spot can be achieved due to the

interferences of wavelets originated in the lens plane from different zones that is

confirmed by its cross-sectional plots in the lateral and axial directions both

theoretically and experimentally (Fig. 2.15). The ultra-thin flat lens presents far-

field 3D subwavelength focusing for a broad wavelength range from 400 to 1500

nm shown in Fig. 2.15 for theoretical ones (b, c) and experimental ones (d, e).

Figure 2.15 (a) Schematic of the wavefront manipulation by the GO lens

converting the incident plane wave into a spherical wavefront. Inset: optical

profile image of the GO lens. Scale bar: 2 µm. Theoretical focal intensity



distributions in the lateral (b) and axial(c) directions. Experimental focal intensity

distributions along the lateral (d) and axial (e) directions.84

Graphene-based materials have also been used for the light emitting

applications. Strong luminescence of GO can be realized by chemically bonding

fluorescence molecules on the GO frameworks. Lu et al. utilized the hybrid GO

materials with strong blue-emission centred at 400 nm through the GO surface

functionalization with aryl diazonium salts of 2-aminoanthreacene.97 Lee et al

investigated the polymer LED with a GO interlayer, which maximizes the hole-

electron recombinations within the emissive layer and enhances the efficiency of

the LED.98 Wang et al. fabricated the GO thin film as the anode interfacial layer in

quantum dot LED, acting as the electron blocking and hole-transporting layer.99

Moreover, GO has also been used as the electrode materials in LED. Bi et al

simultaneously reduced and patterned GO films into arbitrary shaped anodes for

micro-OLEDs via the mask-free laser writing method. Well-defined sizes, shapes

and uniform electroluminescence characteristics have been demonstrated, enabling

a wide range of applications of the micro-OLEDs.100

2.5.2 Graphene-based nonlinear photonic devices

All-optical integrated photonic devices offer competitive solutions to achieve on-

chip signal processing without the need for complex and inefficient optical-

electrical-optical (O-E-O) conversion.101 By directly processing signals in the

optical domain, these devices can reduce power consumption and speed up optical

telecommunication systems with the added benefits of providing a compact

footprint, high stability, mass-producibility, and significant potential to reduce

cost.96,102 Four-wave-mixing (FWM), as an important nonlinear optical process, has

been widely used to realize all-optical signal processing functions such as

wavelength conversion,103,104 optical logic gates,105,106 optical comb

generation,107,108 quantum entanglement,109,110 and more.111,112 Efficient FWM in

the telecommunications band has been demonstrated using silica fibre,113,114 III/V

semiconductor optical amplifiers (SOAs),115,116 integrated photonic devices based

on silicon103,111 and other complementary metal–oxide–semiconductor (CMOS)

compatible platforms,112,117 polymer composites,118,119 chalcogenide devices,120 and

others.58



The prominent nonlinear optical properties of graphene-based materials have

made them excellent candidates for nonlinear photonics devices. Gu et al. reported

regenerative oscillation and FWM in graphene optoelectronics (Fig. 2.16).58 They

have demonstrated, a hybrid graphene–silicon optical cavity for chip-scale

optoelectronics, with the exceptionally high third-order nonlinear response of

graphene, for the first time. It enables ultralow-power optical bistable switching,

self-induced regenerative oscillations and coherent FWM at femtojoule cavity

energies on a semiconductor chip platform.

Figure 2.16 Graphene-clad silicon photonic crystal nanostructures. (a) Scanning

electron micrograph (SEM) of the tuned photonic crystal cavity. (b) Measured

Raman scattering spectra of monolayer CVD-grown graphene on the photonic

crystal cavity membrane. (c) SEM of the suspended graphene–silicon membrane.

(d) Example measured graphene-clad cavity transmission.58

Ji et al. demonstrated the FWM enhancement in a compact silicon-graphene

microring (SGM) resonator with a radius of 10 μm (Fig. 2.17).59 The maximum

conversion efficiency of the FWM, defined as the ratio of the output idler to the

input signal, is 6.8 dB. A nonlinear propagation model in the microring resonator is

established to analyze the conversion efficiency.



Figure 2.17 SEM image of the SGM resonator.59

Alexander et al.121 reported a degenerate FWM experiment on a graphene

covered silicon nitride (SiN) waveguide shown in Fig. 2.18, which was gated

through a polymer electrolyte. The experiment presented that the nonlinear

conductivity of graphene has a sharp resonance as a function of signal-pump

detuning. And a broad asymmetric resonance shape in the vicinity of the absorption

edge has been observed. In the vicinity of the interband absorption edge, a peak

value of the waveguide nonlinear parameter of ≈ 6400 m−1W, corresponding to a

graphene nonlinear sheet conductivity around 4.3 × 10−19A m2V−3, is measured.

This research demonstrated that graphene can be integrated on a waveguide

platform and further applied as a building block for electrically tunable nonlinear

photonics.

Figure 2.18 (a) Sketch band diagram of graphene. (b) Degenerate FWM energy

diagram. (c) SEM image of the cross-section of a SiN waveguide. (d) Schematic

of the gating scheme. (e) Optical microscope image of a set of waveguides.121

The SA effect is another kind of nonlinear properties of graphene-based

material and has made them excellent candidates for the passive Q-switching and



mode-locking of lasers. Bao et al. reported the first graphene mode-locked fibre

laser. SA in graphene is observed due to the Pauli blocking of the electrons and

holes for the occupation of the energy levels in the conduction and valence bands

that are resonant with the incident photons.122 Sun et al used both single-layer

graphene and few-layer graphene flakes as saturable absorbers, to obtain passively

mode-locked erbium-doped fibre laser working at 1559 nm, with a 5.24 nm spectral

bandwidth and ~460 fs pulse duration, paving the way to graphene-based

photonics.123 Bonaccorso et al. reported the first GO mode-locked fibre laser in

2010, exhibiting the optical spectrum of ~743 fs pulse.124


Although photonic devices have been extensively studied and fabricated, there exist

major challenges as well as opportunities towards the practical applications.

For linear photonic devices, firstly, design and fabricating graphene-based

photonic devices in the nanometre scale or even in the atomic scale are still

challenging for existing strategies. Secondly, due to the OFGs of GO, the prepared

GO film is hydrophilic and therefore cannot remain its morphology and properties

in some extreme environment such as water and high-temperature atmosphere.

These drawbacks will severely affect the application area of graphene-based

photonic devices.

For nonlinear photonic devices, firstly, the transmission loss is a crucial factor

in nonlinear photonics area. Current graphene film suffers its high loss due to its

metallic-like nature. Material with high nonlinearity and low loss is in high demand

for the integrated photonic devices. GO with such properties is a promising

candidate for the integrated photonics field.

Secondly, in order to enhance the performance of hybrid structures with GO

coating, enlarged light-material interaction is required. With the integration of 2D

material and well-defined functional nanostructure, the hybrid structure paves the

way for novel applications and enhanced performances in devices such as

photovoltaic devices, photodetectors and Si photonics. By utilizing the large surface

area and unique morphology nature of Si nanostructure, conformal coating of thin

nano-layers of 2D material on designed Si nanostructures plays the key role to



improve the optic and electronic behaviours of the hybrid structures. Many thin film

deposition techniques have been investigated to coat graphene-based 2D materials

such as chemical vapour deposition (CVD) and epitaxial growth. However, the

sophisticated transfer process is the hurdle for conformal and desired coating. In

addition, using costly and complex fabrication techniques will also limit the paths

toward large-scale fabrication. The other approaches are the solution based methods

such as a spin and drop coating, which are reported to provide a one-step solution

to coat graphene-based material onto arbitrary structures. Since the solution based

materials will only coat on top of the architecture due to the hydrophobic nature of

the nanostructure surface. Therefore, realising uniform conformal coating of

graphene-based material onto well-defined nanostructures with a low-cost and

scalable method is greatly desired for future optoelectronic devices with a cost-

effective and scalable method.

2.6 Conclusion

In summary, this chapter reviews the state of the art graphene-based thin film

deposition approaches including graphene thin film deposition methods and GO

thin film deposition methods. The review of various photoreduction methods is

followed up, which introduced the existing technology for tuning the optical and

electronic properties of GO. Then the reported linear and nonlinear optical

properties of graphene-based material are presented that pave the way for its

applications. Graphene-based functional devices based on the linear and nonlinear

properties in section 2.4 are reviewed at the end of this chapter. By reviewing the

state of the art researches above, the major challenges in the respective fields have

been identified, which set a high benchmark for this thesis and ensure the research

presented is novel and significant.

Chapter 3: Ultra-thin graphene oxide film preparation


3 Ultra-thin graphene oxide

film preparation

In this chapter, a brief introduction to the subject, discussing the requirements of

the graphene-like two-dimensional (2D) material film for photonic applications are

presented in section 3.1. Section 3.2 introduces the layer by layer (LBL) method for

the preparation of GO film. Owing to the unique mechanism and process of LBL

method, a multilayered structure with nanometre layer thickness can be achieved.

In section 3.3, the achieved multilayer structures are characterised and studied

towards monolayer control of this method. In section 3.4, large scale GO films on

various substrates are demonstrated by this technique. The optical and film quality

characterization is shown in section 3.5 that answers the requirements in section

3.1. At the end of this chapter, a brief conclusion is presented in section 3.6.

3.1 Introduction

After the first successful exfoliation in 2004, monolayer graphene has promised

multiple functionalities in diverse fields for its unique electrons transferring

behavior1,2,14,23,125. However, in optical applications, monolayer graphene suffers

from limited performance due to its ultrathin nature (3.4Å), which is unable to

provide sufficient optical modulation77,126 specifically phase and amplitude

modulations. In contrast, high-quality multilayer graphene materials have

demonstrated the improved performance of functional optoelectronic devices due

to enhanced modulation depth124,127. However, simply accumulating graphene

layers via the conventional film deposition methods leads to its bulk counterpart-

graphite. Thus, the unique atomic properties are no longer preserved.76,77 As

discussed in Chapter 2, current approaches are difficult to deposit large-scale

graphene film and sophisticated transfer process would render ruptured 2D

structures with seriously degraded layer quality after transfer.128



As a result, the demand for pursuing large-scale, high-quality graphene-based

platform calls for the development of highly controllable multilayer graphene-based

films.

3.2 Layer by layer graphene oxide film preparation

To overcome these challenges, a low-cost, scalable process to fabricate high quality

GO ultrathin film separated by nanometer spacing layers without any transfer

process by precisely controlling the layer down to a monolayer (monolayer

thickness: ~1 nm) is developed. The achieved large-scale GO film preserves its

properties with an unprecedented surface roughness of ~2 nm, far transcending the

most stringent optical standard. The process started from the formation of GO

multilayers11,36,38,40,129,130 by using a layer-by-layer LBL electrostatic force

deposition method facilitated by nanometer spacing layers.

Figure 3.1 Schematics of the layer-by-layer process (a) and layered structure of a

5-layer GO film with the inset highlighting the molecular structure of the film (b).

As depicted in Fig. 3.1, GO is a graphene-based material decorated with

oxygen-functional groups (OFGs), which allow uniform dissolution of GO flakes

in water.34 The surface of GO is negatively charged, enabling their firm attachment

to any positively charged substrate by the electrostatic force. Here we use a

positively charged dielectric material polyelectrolyte

polydiallyldimethylammonium chloride (PDDA), which has low absorption from

visible to near-infrared wavelength suitable for optical devices, to attach to the

substrate first followed by the adsorption of GO flakes. By properly controlling



flake number, flake sizes and the concentration of the GO solution, a monolayer

GO film can be formed. Repeating the PDDA and GO deposition process,

multilayer GO films with precisely controlled layer number and layer thickness can

be achieved. Details of the method are presented below:

The high quality GO solution was synthesized by the chemical oxidation of

graphite through a modified Hummers method42 and sonicated vigorously by the

Branson Digital Sonifier34. The layer-by-layer process can be divided into four

steps: 1) immersing the substrate into a 2.0% (w/v) aqueous PDDA (Aldrich)

solution; 2) rinsing with a stream of deionized distilled water and drying by N2; 3)

immersing PDDA coated substrate into the aqueous GO solution; 4) washing the

sample with a stream of deionized water and drying by N2. After completing steps

1-4, a polycation/GO layer film was directly assembled on the desired substrate.

These steps could be repeated for constructing multilayer films and the process is

highly scalable131. In principle, the size of the film is only limited by the solution

container.

Figure 3.2 Optical images of GO films on a glass substrate from 1 to 5 layers.

Figure 3.3 GO multilayer film surface profile characterization. (a) Microscope

image of a 5-layer GO film. The surface is ultra-smooth under a microscope

(Nikon ECLIPSE MA100 microscope). (b) Scanning electron microscope

(RAITH150-TWO) image of 5-layer GO film.



Figure 3.2 shows the GO films from 1-5 layers on a glass substrate. From the

image, we can see the colour of the film is changed with the increasing of the layer

number, which means the optical properties such as absorption and refection can be

tuned by controlling deposited layers. The surface profile is characterized by an

optical microscope (Fig. 3.3a) and a SEM (Fig. 3.3b), which present the ultra-

smooth surface of the multilayer GO film. The images above illustrate the smooth

surface of the GO film without significant wrinkles and cracks. This can be

attributed to the LBL manner deposition process that enables monolayer control of

layered film structure.

3.3 Multilayer structure characterization

The morphology properties of the achieved multilayer GO film are characterized

by a scanning electron microscope (SEM) and atomic force microscope (AFM).

The SEM image of the resulted multilayer GO film is shown in Fig. 3.4, in which

five ultrathin layers can be clearly identified. The corresponding AFM image (Fig.

3.5) shows each GO-PDDA layer is approximately 3 nm thick, which results in an

overall thickness of 18 nm for a 5-layer structure, rendering each combined layer

thickness of 3.6 nm on average.

Figure 3.4 SEM image of a 5-layer GO multilayer structure, clearly showing the

layer-by-layer structure.



Figure 3.5 AFM measured thickness profiles of a 5-layer GO multilayer film, with

corresponding AFM topographic image (inset).

To achieve monolayer control of the GO film, there are two key factors to

consider: 1) the polymer layer has sufficient positive charges to ensure monolayer

GO attachment in each iteration. The thickness of a monolayer PDDA is confirmed

to be ~2 nm by AFM (Fig. 3.6). Here, the positively charged PDDA layer acts as

an intermediate layer, which could not only firmly connect the negatively charged

GO layer and the oxidized substrate surface, making the film smooth and stable but

also separate the adjacent monolayer GO films, maintain their 2D structure and

properties.

Figure 3.6 Thickness profiles of a PDDA layer, with corresponding AFM

topographic images (insets).

2) GO should be highly oxidized by controlling the oxidization time and

sufficiently exfoliated by vigorous sonication in solution. The resulted monolayer



GO flakes can be observed from the AFM images in Fig. 3.7. The opposing

electrostatic force between the same negatively charged GO layers can break the

van der Waals force between layered GO, ensuring the GO flakes to maintain as

monolayers.

Figure 3.7 Thickness profile of a GO monolayer, with corresponding AFM

topographic image (inset).

3.4 Optical properties and film quality characterization

In order to optimize the GO film fabrication process, including the GO flake size,

immersion time in each iteration and concentration of the GO solution, we used the

optical absorption spectra and surface roughness to characterize the quality of the

GO films (Figs. 3.8-3.9). The immersion time varies from 0.5 to 30 min and the

concentration of GO dispersion rises from 1 to 13 mg/ml. Figure 3.8 shows the

absorbance and roughness changes of a 5-layer film with different concentrations

of the GO solution (immersion time: 15 min). Two phases can be identified: Phase

I is when the GO concentration is ≤ 5 mg/ml, in which the absorbance and

roughness increase with GO concentration due to the incomplete covering of the

substrate by the GO flakes; Phase II is when the GO concentration is > 5 mg/ml, in

which both the absorbance and roughness become saturated and stable with

increased concentration, indicating that GO flakes have covered the entire polymer

layer and charge-neutralization has been achieved. Further increasing the

concentration does not have significant influences on the properties of the GO films

any more (absorption and roughness). To enable uniform distribution of the flakes,

the GO solution is vigorously sonicated every 5 cycles.



Figure 3.8 Absorbance and roughness versus solution concentration at various

wavelengths of the 5-layer GO film. Phases I and II are separated by the grey dash

line.

Figure 3.9 presents the absorption of a 5-layer GO film under various immersion

times at three different wavelengths and the corresponding roughness

(concentration: 5 mg/ml). The deposited GO film has an outstanding smooth

surface with a surface roughness of approximately 2 nm, which stems from the

monolayer GO flakes stitched on the positively charged dielectric polymer layer by

electrostatic force. The overall absorbance reflects the total amount of GO flakes

attached to the substrate. It is clear that the absorbance changes in a small range for

varied immersion times, indicating the majority (> 95%) of GO attachment was

achieved within 1 min. Further increasing the immersion time does not lead to

notable change due to the complete coverage of the substrate by the monolayer GO

film. Therefore, the optimized concentration to ensure a highly uniform film is at 5

mg/ml and the optimized immersion time is 1 min.



Figure 3.9 Absorbance and roughness versus immersion time at various

wavelengths of the 5-layer GO film.

Figure 3.10 Broadband absorption spectra with increased layer number (1 to 5).

Owing to the multilayer structure of the GO films, it is possible to tune the

optical properties by simply controlling the layer number. Figure 3.10 presents the

absorbance curve for different layer numbers from 1 to 5. As expected, the

absorbance is proportional to the layer number.



Figure 3.11 Thickness and roughness versus layer number from 1 to 10 layers.

The dependence of the film thickness versus the number of layers shows a clear

linear relationship (Fig. 3.11). The high surface quality and low roughness are

further confirmed by the AFM measurement, which shows a surface roughness of

2 nm independent on layer numbers (Fig. 3.11). The 2 nm surface roughness far

exceeds the optical standard and reaches λ/270 for visible light at 540 nm, which is

among the smoothest GO films demonstrated so far. The accurate control of the

film thickness by layer number as well as the ultra-smooth surface renders the GO

film a promising candidate for the implementation of optical and nano-photonic

applications.

3.5 Large scale fabrication on different substrates

Intensive researches have been focused on the methods for preparing high-quality

graphene-based material film towards large scale and high yield production for

years. And compatible with the currently available functionalized substrates is a

key challenge.



Figure 3.12 (a) The optical image of a 5-layer GO film on a 4-inch silicon wafer.

(b) Corresponding Raman spectra of the Si wafer and the GO film coated Si

wafer.

Given its strong mechanical robustness, large-scale integratability and transfer

free process, the LBL prepared GO ultrathin films have the potential to

revolutionize diverse photonic devices through integration with various materials

and arbitrary shaped surfaces. To demonstrate the capability of large-scale

fabrication by the LBL method, a 5-layer GO film was coated on a 4-inch silicon

(Si) wafer (Fig. 3.12a). The integration of GO film on the (Si) substrate is confirmed

by the Raman spectroscopic measurement (Fig. 3.12b) showing the representative

D (1345 cm-1) and G (1590 cm-1) peaks of GO. The D to G ratio down to 0.7 can

be achieved, indicating the low defect density of the GO layers.132,133

Figure 3.13 Thickness mapping of a 5-layer GO film on a Si wafer by an

ellipsometer.



The thickness mapping of the 5-layer GO film by an ellipsometer (Fig. 3.13)

demonstrates that the thickness is controlled at 18±1.5 nm over the entire wafer area

(4 inches in diameter). More importantly, the solution-based LBL process allows

the direct coating on substrates with arbitrary shapes without any transfer process.

Figure 3.14 A 72-mm-diameter curved (top left) acrylic lens (top right). The lens

with 10-layer (bottom left) and 20-layer (bottom right) GO coatings.

To demonstrate this flexibility, a 72-mm optical acrylic lens with a curved

surface is shown in Fig. 3.14 top, which is challenging to be coated uniformly by

conventional vacuum coating methods. By using our LBL deposition method, the

surface roughness is greatly restricted to be less than 2 nm even with a large number

of layers. As a result, uniform 10-layer (Fig. 3.14 bottom left) and 20-layer (Fig.

3.14 bottom right) GO multilayers were coated on the 72 mm curved optical acrylic

lens with a high transmittance, which has many potential applications in vision

correction and virtual reality industry.



Figure 3.15 A 5-layer GO Large-scale film integrated on flexible polyester (PET)

film (left) compared with pristine PET substrate (right).

Figure 3.16 A 5-layer GO film on a flexible transparent substrate with bending

and twisting.



To illustrate its mechanical flexibility and large-scale preparation, a 5-layer GO

thin film has been integrated on a 30 cm× 20 cm transparent flexible polyester

substrate (Fig. 3.15) via the LBL method. After various bending and twisting (Fig.

3.16), the GO film survived without any visible wrinkles and cracks, demonstrating

the excellent mechanical strength and flexibility. Here we successfully deposited

GO film onto various substrates including (glass, Si, acrylic and polyester),

representing its versatility and integratibility with different substrates.

3.6 Conclusion

In this chapter, a LBL deposition method has been employed and developed for

ultra-thin GO film preparation. The clear multilayer structure is achieved with this

method and well characterized by SEM and AFM images. Furthermore, this method

is applied for large scale fabrication of high-quality GO multilayer films on various

substrates including glass, PET and Si without any transfer process. Due to the

excellent mechanical properties, the GO film can be well maintained on the flexible

substrate after bending and twisting. UV-Vis spectrometer and AFM are introduced

to characterize the optical properties and film quality of the multilayer GO films.

The optical properties are also tuned by the layer numbers. Finally, we can precisely

control the layer down to a monolayer (monolayer thickness: ~1 nm) onto diverse

substrates. The achieved large-scale GO films preserve their properties with an

unprecedented surface roughness of ~2 nm, far transcending the most stringent

optical standard.

Chapter 4: Graphene oxide 3D conformal coating on Si nanostructures


4 Graphene oxide 3D

conformal coating on Si

nanostructures

In this chapter, an introduction for graphene oxide (GO) conformal coating is given

in section 4.1. Section 4.2 introduces the solution-based layer by layer method for

GO conformal coating onto the silicon (Si) nanostructures. Owing to its unique

properties of Si nanostructures, flake size and surface hydrophilicity are the key

parameters for the conformal coating. In section 4.3, the surface processing

methods are introduced to manipulate GO flake size in the solution and surface

hydrophilicity of the structure surface. In section 4.4, the optical properties of the

hybrid GO coated Si nanowires (SiNWs) hybrid structures are well studied

theoretically and experimentally. Section 4.5 presents the characterization of the

GO conformal coated Si solar cell. At the end of this chapter, a brief conclusion is

presented in section 4.6.

4.1 Introduction

Nanostructure, based on the dominant semiconductor material, provides novel

avenues for exploring fascinating performance and applications by developing



designed properties with precise three-dimensional (3D) architecture. Si, being

regarded as the dominant material in the semiconductor industry, have been

extensively applied in photovoltaic devices, photodetectors and integrated

photonics.134-136 However, the intrinsic characteristics of Si such as high refractive

index and strong two-photo absorption limit its further application in photovoltaic

and integrated photonic devices. Therefore, integrating the graphene-based function

layer to transform the Si platform to a multifunctional platform by overcoming its

drawbacks is eagerly desired. Conformally coated Si hybrid structures, which can

preserve the desired properties without changing its pre-designed profile, offers a

potential solution for the integration of thin nano-layers of graphene-based material

onto designed Si nanostructures. Furthermore, by utilizing the large surface area

and 3D morphology of Si nanostructure, the conformal coating can improve the

optic and electronic behaviours of the hybrid structure such as anti-reflection and

light harvesting efficiency. 54,55

As discussed in Chapter 2, conventional graphene deposition methods are

difficult to realize conformal coating due to the vacuum-based mechanism and

complex transfer process for each layer. Owing to the hydrophobic nature of the Si

nanostructure surface, the GO suspension can only stay on top of the surface rather

than dip into the gap between the nanostructures. This effect results in the GO film

floating on top of the structure. Thus, we demonstrate GO conformal coating onto

Si nanowires (SiNWs) and further expand it to the Si solar cell. A significant

improvement in light harvesting efficiency is achieved, which proves the potential

of conformal coating in photovoltaic applications.

4.2 3D conformal coating on Si nanowires

Si nanostructures, such as SiNWs, Si nanoholes, Si nanocones, nanopores, etc.

provide a low reflectivity strategy over a broad wavelength range in the solar

spectrum, making them promising substrates for solar cells due to their light-

harvesting capability compared with planar silicon wafers with an equivalent

volume. Conformal coating of such structures with 2D materials remains as a

challenge for high-performance photovoltaic devices. In this chapter, we introduce

a well-developed solution-based method which is able to in situ conformably coat

GO thin films onto SiNWs with monolayer controllability. By manipulating the GO



flake size and surface chemistry, a well-defined conformal coating on 3D Si

nanostructures can be achieved.

Figure 4.1 Schematic of 3D conformal coating process on Si nanostructures.

Based on the previous study in chapter 3, a well-developed solution based layer-

by-layer (LBL) method is capable to overcome this challenge in the conformal

coating on Si nanostructures. As shown in chapter 3, the LBL method is based on

electrostatic force to form a GO film. By exploiting the surface charge property of

GO and utilizing the electrostatic force, the LBL method is able to realize firm as

well as conformal attachment of GO material onto Si nanostructures. Figure 4.1

presents the schematic of the process of the GO film tightly coated onto the SiNWs

due to the electrostatic adhesion. This is verified by the corresponding scanning

electronic microscope (SEM) images of the SiNWs (Fig. 4.2) and GO conformal

coating (Fig. 4.3).

Figure 4.2 SEM image of the Si nanowires.



Figure 4.3 SEM image of the conformal-coated Si nanowires.

These two regions clearly show the GO conformal coating on the SiNWs. The

height of the SiNWs is 400 nm, the diameter is 200 nm and the pitch is 600 nm.

The thickness of the GO film is 20 nm. The overall GO film thickness can be

precisely controlled by the layer numbers, which can be repeated by the modified

LBL process. The firm contact of the GO conformal coating onto the SiNWs are

shown in the SEM image (Fig. 4.4).

Figure 4.4 SEM image (zoom in) of the morphology of the GO conformal coating

Surface profiles of GO conformal coating of low (top) and high (bottom) aspect

ratio of the SiNWs are shown in Fig. 4.5. From this image, it shows clearly that the

specific LBL method can coat GO film even on high aspect ratio structures. Figure

4.6 depicts the boundary of the GO conformal coating. The left parts show the

morphology details of the GO conformal coating, and the right part is the SiNWs.

The contrast is shown in Fig. 4.6 presents the profile change of the coating that will

modify the optical and electrical properties of the hybrid structures. All these results

show a great potential of integrating GO thin film with Si-based structures.



Figure 4.5 SEM image of low (top) and high (bottom) aspect ratios of the GO

coating.

Figure 4.6 SEM image of the boundary of the GO coating.

The LBL method also has the ability for large-scale coating. Fig. 4.7 shows the

large-scale coating for SiNWs, the left part is the conformally coated SiNWs and

the right part is bare SiNWs. The darker colour of the right part represents the GO



film on the SiNWs. Corresponding Raman characterization proves the successful

coating onto SiNWs. Raman spectra in Fig. 4.8 clearly shows the Si peak at 510

cm-1. The stronger D (1350 cm-1) and G (1590 cm-1) band signal and lower Si peak

appearing in Fig. 4.8 represent the GO is coating onto the Si nanostructures.

Figure 4.7 SEM image of large scale with (left) and without (right) GO the

coating. Insets are zoomed in structures.

Figure 4.8 Raman spectra of the GO conformal coating on Si nanowires.



4.3 Graphene oxide wet chemical tunability

During the wet chemical process, the flake size and surface hydrophilicity are the

two key factors for a successful coating. In this section, the tunablities of these two

parameters will be discussed.

4.3.1 Flake modification

In order to realize conformal coating, necessary treatments are introduced during

the film preparation process. Due to the excellent mechanical properties, the flake

size is a crucial parameter for GO conformal coating. The normal GO flake size is

in the order of micro-scale, which is much larger than the pitch of the

nanostructures. As a result, the GO flakes would float on top of the nanostructures.

In order to achieve conformal coating, the flake size must be decreased to hundreds

of nanometre, comparable to or even smaller than the size of Si nanostructures.

Figure 2 (a) presents the schematic of the flake size decreasing process. The GO

flakes are shattered into small pieces by introducing external mechanical force with

high frequency (i.e. ultrasonication). Here, we vigorously sonicate the GO solution

by a powerful sonifier (Branson digital sonifier). By controlling the sonication time,

the averaged flake size can be remarkably reduced from micro-scale to hundreds of

nanometres as demonstrated by the atomic force microscope (AFM) measurement

(right) in Figs. 4.9 (b)-(d). The original flake size is around 1 µm without sonication

(Fig. 4.9(b)). After 5-minute sonication, the flake size can be reduced to 500 nm

(Fig. 4.9(c)) and further increasing the sonication time to 10 minutes leads to flake

size as small as 100 nm (Fig. 4.9(d)), an order of magnitude decrease. Such a small

size is significantly below the pitch of the SiNWs and would be responsible for a

successful conformal coating onto SiNWs, but the approach still needs to be further

developed.



Figure 4.9 AFM image of sonicated GO flake sizes with different sonication time

and corresponding flake profile. (a) AFM image of GO solution without

sonication (left) and flake morphology (right). (b) AFM image of GO solution

with 5-minute sonication (left) and flake morphology (right). (c) AFM image of

GO solution with 10 minutes sonication (left) and flake morphology (right).

4.3.2 Surface processing

The silicon nanostructures fabrication will change the morphology of the surface to

hydrophobic. This is caused by the increased surface roughness. In order to

conformally coat GO film on the silicon, surface modification is needed. We used

triton as a surfactant and added it into GO and PDDA solution. Triton is a mild non-

ionic surfactant, which has a nonionized but polar structure at its hydrophilic end.

The non-ionic nature makes triton non-charged that will not affect the electrostatic

based LBL process. The huge decrease of the contact angle is shown in Fig. 4.10

before (left) and after (right) adding triton into GO solution. The measured contact

angle of PDDA solution decrease from 101.5º (hydrophobic) to 14º (hydrophilic).

The measured contact angle of GO solution (Fig. 4.11) decrease from 111º

(hydrophobic) to 23º (hydrophilic).



Figure 4.10 Contact angle of PDDA solution (left) and with surfactant (right).

Figure 4.11 Contact angle of GO solution (left) and with surfactant (right).

The decrease of the contact angle is attributed to the hydrophilic end due to the

effect of triton in solution. Small contact angle represents hydrophilic nature at the

interface and implies good wettability between silicon nanostructures. After surface

modification, the material solution can be infiltrated into the nanostructure easily

for conformal coating. After the modification for both flake size and surface

hydrophilicity, the Si nanostructures can be fully immersed in the environment

surrounded by GO solution leading to a well-defined conformal coating.

Figure 4.12 SEM images of the coating morphology with flake size control and

surface modification. (a) GO film floats on top of the SiNWs without any process.

(b) Partly conformal coating with reduced flake size. (c) Fully conformal coating

with reduced flake size and surface modification.

Figure 4.14 illustrates the three steps for pursuing a conformal coating. It starts

from large flake floating on top of the SiNWs, then the morphology changes due to

the reducing of the flake size and realize partly conformal coating. After reducing

14 ⁰

101.5⁰ PDDA solution

23⁰

111⁰ GO solution



flake size and surface hydrophilicity processing, we finally realized high quality

GO film conformal coating onto 3D SiNWs.

4.4 Broadband anti-reflection of 3D GO conformal coating

For the Si substrate, the high reflection will appear at the air-Si interface due to

its high refractive index. Si structures such as periodic or random arrays of Si

nanowires and nanopyramids have proven to reduce reflection over a broadband

wavelength. Recently, SiNWs have demonstrated unique optical properties by

designing the key parameters such as diameter, pitch and height, specifically. The

reflection of the SiNWs can be reduced by multiple scattering of light as well as

coupling by the gradually increasing index from the air to the substrate. And also,

the designed reflection over a broad range of wavelengths and structure filled with

air will minimize the overall refractive index mismatch that will help to further

reduce reflection. Moreover, the light can be confined in the structure that will help

to enhance the light harvesting efficiency due to the high refractive index of Si. As

a result, SiNWs provide a low reflectivity over a broad wavelength range in the

solar spectrum, making them promising light-weight flexible substrates for solar

cells due to their light-harvesting capability compared with planar Si wafers with

an equivalent volume. Benefitting from the achieved advantages of SiNWs, the

hybrid GO-SiNWs structure by GO conformal coating possesses good optical and

electrical property that will be studied in this section.

4.4.1 Theoretical study and simulation

In this section, a commercial Finite-Difference Time-Domain (FDTD) from

Lumerical is introduced to analysis the antireflection of the hybrid structure. Figure

4.13 (a) depicts the schematic of the simulation model. Single SiNW unit is used to

simulate the reflection and intensity distribution. For the reflectance simulation, a

transmission monitor is put above the plane wave source with a solar wavelength

from 300 nm to 1100 nm. For intensity field calculations, a two-dimensional

frequency-domain field monitor crosscutting the SiNW in its centre is used to

obtain the electric field intensity distribution. The intensity distribution in Fig. 4.13

(b) presents that the light is highly confined in the Si nanostructure.



Figure 4.13 (a) Schematic of the SiNW model. (b) Field distribution of single cell

of SiNW.

From Fig. 4.13, we can observe the strong light confinement in the single SiNW

that will reduce the light reflection at the interface due to the high refractive index

and designed structure of SiNW. The height of SiNW is 300 nm and diameter is

400 nm. To study this, we perform simulations of the reflectance of different SiNW

dimensions for the interested solar wavelength range from 300 nm to 1100 nm.

Figure 4.14 shows the simulated reflectance spectra with different SiNW height.

The overall reflectance is decreased with the height increasing (100-300 nm) and

achieves the lowest level at 300 nm height. The dips at different wavelength shown

in Fig. 4.14 are caused by the resonances for different dimensions of the SiNWs.

Figure 4.14 Reflectance for SiNWs with different heights (100-400 nm) with the

same diameter of 400 nm over a broadband wavelength (300 to 1100 nm).



Figure 4.15 shows the simulated reflectance spectra with different SiNW

diameter. The overall reflectance is decreased with the diameter increasing (200-

600 nm). The dips at different wavelength shown in Fig. 4.17 are caused by the

resonances for changing the dimension parameters of the SiNWs. The data in Fig.

4.14 and Fig. 4.15 show that the SiNW array reduces the reflectance of a Si surface

over the entire spectral range from 300-1100 nm. The optimized aspect ratio of the

SiNWs is 2.

Figure 4.15 Reflectance with different diameters (200-600 nm) with the same

height of 300 nm over a broadband wavelength (300 to 1100 nm).

The schematic of the simulation model for GO-SiNWs hybrid structure is

illustrated in Fig. 4.16(a). Figure 4.16(b) shows the simulation results for the

reduced reflection of GO conformal coating. After GO conformal coating on

SiNWs, the field distribution confined in the SiNW overlaps with the substrate

heavily thus introducing the strong coupling of light from the nanowire to the Si

substrate (Fig. 4.16 (b)).



Figure 4.16 (a) Schematic of the hybrid GO-SiNW model. (b) Field distribution of

single cell of GO-SiNW model.

Figure 4.17 illustrates the simulated reflectance spectra with different GO film

thickness. The overall reflectance is decreased with the GO thickness increasing

(20-200 nm). The dips at different wavelength caused by the resonances of SiNWs

in Fig. 4.17 are well maintained owing to the conformal morphology of the GO film

when GO thickness is below 100 nm. The profile of the hybrid structure is changed

when the thickness of GO film increases to above 100 nm that is one-third of the

diameter of the SiNW (~300 nm). The designed low reflection of SiNWs cannot be

preserved when the coating changes the overall morphology. This is also verified

by the simulation results shown in Fig. 4.17 that the dips of the reflectance spectra

are missing when GO thickness is over 100 nm. And the reflectance dip caused by

the resonance of SiNWs can be maintained when the GO film thickness is below

100 nm. In order to achieve the lowest reflection by taking the advantages of both

SiNWs and GO coating, the optimized GO thickness is ~100 nm.



Figure 4.17 Reflectance with different heights (100-400 nm) for GO –SiNW

hybrid structures with the same aspect ratio (around 2) over a broadband

wavelength (300 to 1100 nm).

The hybrid GO-SiNW structure will improve the light collecting efficiency for

the whole device. This effect is explained by the enhanced forward scattering of

light from the well-designed SiNWs due to the high optical mode density of the

high-index Si substrate as well as the GO conformal coating. The significant anti-

reflection effect could be attributed to the following reasons. First, the excellent

optical property of GO (refractive index is ~1.8 for a broadband wavelength in

section 3.5) will help to reduce the reflection at the Si-air interface by matching the

refractive index. The second reason is the ultra-low extinction coefficient of GO

(section 3.5) from 300 to 1100 nm results in little absorption of the conformal-

coated GO film thus enhance the light harvesting efficiency. The third reason is that

GO conformal coating is not only on top of the structure but also cover the side

walls, which will take advantage of the large surface area of the SiNWs. The

uniform core-shell structure will suppress the optical mode distribution in the

SiNWs and pushing the light leak to the substrate rather to the air side of the

interface. Last, the light absorption will be further improved owing to the

enhancement of the effective scattering area by the larger cross section of GO-

SiNWs hybrid structure compared with bare SiNWs. The above four reasons lead

to low reflection and GO-SiNWs hybrid structure. The demonstrated optical

properties of GO conformal coating will enable the hybrid GO-SiNWs structure to

become a promising candidate for photovoltaic optoelectronic devices.



4.4.2 Experimental results

As discussed in the last section, the GO coating could help to reduce the

reflection and improve the light collection based on the simulation results. Figure

4.18 presents the experimental results of Si wafer (blue line), SiNWs (black line)

and GO-SiNWs hybrid structure (red line) respectively. The GO thickness here is

~100 nm. The experimental results match well with the simulation. From Fig. 4.18,

it is found that more than 20% reduction in reflectance is achieved by a 100 nm GO

coating over a broadband wavelength from 300 nm to 1100 nm.

Figure 4.18 Reflectance spectra (300 to 1100 nm) of Si wafer (blue line), SiNWs

(black line) and GO-SiNWs hybrid structure (red line), respectively.

4.5 Characterization of GO conformal coated Si solar cells

As an application demonstration, we conformally coated the GO thin films on

textured Si solar cells to improve their conversion efficiency. Due to the low

reflection and enhanced light collection ability, textured Si structures have broad

applications in the photovoltaic field. However, textured Si structures here acting

as the light-trapping nanostructures will significantly enlarge the surface area of the

silicon substrates and induce additional surface trap states, leading to a high surface

recombination velocity and reduced PV performance. In addition, the extra

roughness and defects caused by the fabrication procedure at the surface will also

have negative effects on the electrical performance of the solar cell. Therefore, an

effective surface passivation layer is necessary to ensure minimal surface



recombination and long minority charge-carrier lifetime in a high-performance PV

device. We can achieve effective passivation by coating tens of layers of GO film

on silicon structures, leading to enhanced electrical performance in addition to the

antireflection benefit. The thickness of the GO films is ~60 nm, which is close to

the simulated optimized thickness (60 nm). The average lifetime will saturate when

the GO film thickness is over 60 nm. Therefore, 60 nm thickness is the optimized

parameter for minimized the material and time consumption. The characterized

carrier lifetime mappings of GO conformal coated honeycomb solar cell are

depicted in Fig. 4.19 and 4.20, respectively.

Figure 4.19 Carrier lifetime mappings of bare SiNWs.

The carrier lifetime mapping results of GO conformal coated honeycomb solar

cell is shown in Fig. 4.20. The calculated average carrier lifetime is improved from

8.9 µs (bare solar cell) to 14.2 µs (GO coated solar cell). A significant enhancement

about 60% has been proved with 60 nm thick GO conformal coating. The surface

recombination velocity is suppressed because the effective carrier lifetime is

significantly improved up to 60% caused by the ultrathin GO film. The suspending

bonds of the textured solar cell are saturated by conformal GO coating film. All

these contribute to the effective passivation against surface defect states on the

textured silicon.



Figure 4.20 Carrier lifetime mappings of GO-SiNW hybrid structures.

The measured current density and voltage (J-V) curves of the honey-combed Si

solar cell with (red line) and without (black line) GO coating are shown in Fig. 4.21.

After GO coating, the current density is increased from 25.9 mA/cm2 (bare solar

cell) to 30.7 mA/cm2 (GO coated solar cell) by GO resulted in antireflection effect.

It should be noted that the open circuit voltage of the solar cells also exhibits an

obvious increase from 586 mV to 596 mV. This is attributed to the effective

passivation effect of the GO films that leads to a suppressed surface recombination

and the saturation of the suspending bonds of Si surfaces caused by the conformal

GO coating film.

Figure 4.21 Measured current density and voltage (J-V) curves of the GO coated

honey-combed solar cell (red line), compared with those of the bare solar cells



without GO coating (black line). Inset: optical image of GO coated honey-combed

solar cell.

Figure 4.22 Measured external quantum efficiency (EQE) of honey-combed Si

solar cell with (red line) and without (black line) GO coating.

The EQE of the honey-combed textured solar cells after conformal coated with

the GO film, compared with the bare one is shown in Fig. 4.22. The efficiency of

the solar cells was increased to 12.4% from 10.1%, representing an enhancement

up to 23% over the entire band except for the short wavelength below 400 nm,

which is in the absorption band of GO. The significant EQE enhancement is due to

the antireflection effect of the GO coating.

4.6 Conclusion

In this Chapter, we have developed a solution based method which is able to in situ

conformably coat GO thin films onto 3D nanostructures, for example, SiNWs, with

monolayer controllability. As shown in Chapter 3, by exploiting the surface charge

property of GO, it can firmly attach onto the surface of nanostructures in a layer-

by-layer manner. By manipulating the GO flake size and surface chemistry, a well-

defined conformal coating on 3D Si nanostructures can be achieved. We have

comprehensively analysed the optical and electrical performance of the integrated

GO-SiNWs hybrid structures. Up to 20% reduction of reflection with the broadband

wavelength and enhanced passivation are characterized in GO-SiNWs hybrid

structure compared with solely SiNWs. By applying this method to Si solar cell, the

overall improved performance is observed. A 60% enhancement is achieved in



carrier lifetime measurement and a more than 20% improvement is achieved in EQE

of the hybrid solar cells. Our studies suggest that the LBL GO conformal coating

provides a novel material platform with a precisely controllable and cost-effective

approach to enhance optical and electrical response in nanoscale.

Chapter 5: Ultra-thin graphene oxide platform for photonic devices


5 Ultra-thin graphene oxide

platform for photonic

devices

5.1 Introduction

Graphene has promised multiple functionalities in diverse fields for the unique

electrons transferring behavior1,2,14,23,125 and outstanding optical properties137,

which are particularly advantageous for optoelectronics and energy conversion

applications134-136. The absence of the bandgap in graphene means that it absorbs

electromagnetic radiation over a very wide bandwidth, ranging from the ultraviolet

to the terahertz regime135. However, in optical applications, monolayer graphene

achieves limited performance due to its ultrathin nature (3.4Å) and limited optical

absorption of 2.3%137, which is unable to provide sufficient optical modulation77,126.

In contrast, the optical modulation can be enhanced by using graphene

metamaterials or graphene multilayers (GMLs), thus it has been theoretically

demonstrated broad applications of graphene-based metamaterial such as perfect

absorber138,139, photodetector140, directional light emission141, electro-absorption

modulator142 and tunable spin Hall effect143. However, it remains a significant



challenge in the experimental fabrication of graphene-based metamaterial, thus,

there have been only a few experimental demonstrations144,145.

Recently, it has been reported that experimental realization of graphene-based

metamaterial with a multilayer structure of alternating graphene and dielectric

layers using chemical vapour deposition (CVD) method and transfer of

graphene144,145. The quality of the metamaterial is sensitive to deposition condition

and becomes difficult to control when the number of layers increases due to the

sophisticated transfer process for each layer144,145. Therefore, to date, the maximum

number of periods of the graphene multilayer structure is 5 periods144. As a result,

it is desired to develop a technique that is able to fabricate high-quality graphene-

based material on substrates with arbitrary shape without transferring process. In

addition, it is preferred the technique is low-cost and capable of large scale

production, which will enable broad practical applications124,127.

In this chapter, the laser reduction method will be discussed in section 5.2. The

design for the demonstration of a graphene-based metamaterial consisting of a

multilayer structure of alternating monolayer graphene oxide (GO) /graphene and

a dielectric layer (2 nm thick) fabricated using a solution-based layer-by-layer

(LBL) method11,36,38,40,129,130 described in Chapter 3 will be presented in section

5.3.1. In addition, the properties of the metamaterial can be in-situ tuned by laser

processing, which modifies the bandgap of the GO/graphene layers. The effective

complex refractive indices and optical conductivities are measured using a spectral

ellipsometer. The optical conductivity of the graphene layers shows almost identical

results to CVD fabricated graphene12. The laser tuning process allows in-situ

fabrication of functional photonic devices37,84. The above studies are shown in

section 5.3.2. Laser patterning characterization and the fabricated amplitude

modulation devices based on the developed graphene-based ultrathin platform are

presented in section 5.4. The solution-based method allows applications of our

graphene-based metamaterial in a water immersion environment. Therefore, in

section 5.5 we design and fabricate a water immersion lens integrated onto a

microfluidic biophotonic device to demonstrate one unique application that takes

the advantages of laser processing and water resistance, which is challenging to

achieve for normal water soluble GO material. The lens is able to achieve stable

subwavelength focal resolution in water, which is valuable to lab-on-a-chip



biological devices. The graphene-based metamaterial opens a new experimental

platform for broad potential applications in on-chip integrated photonic, biomedical

and microfluidic devices as well as a perfect absorber and directional light

emission146. At the end of this chapter, a brief summary is given in section 5.6.

5.2 Laser reduction of GO film

Figure 5.1 Schematic of femtosecond laser fabrication on GO film.

Laser photo-reduction (Fig. 5.1) is able to convert GO to graphene-like

material37,83,84 and is an effective approach to adjust the optical and electrical

properties of GO. The laser reduced GO multilayers is defined as GMLs. Upon

laser-reduction, the oxygen functional groups (OFGs) are removed and the

following two processes occur simultaneously: 1) the permittivity of GO is

converted to that of graphene; 2) the layer spacing of GO is reduced as well as the

filling ratio of reduced GO film133. As a result, GO is converted to GMLs via laser

reduction process. The effective permittivity of the graphene multilayers (GMLs)

is effectively tuned by the layer-by-layer structure and the filling ratio is increased.



Figure 5.2 Raman spectra of the GO film and GMLs.

The GO and GML films are characterized by Raman spectroscopy (NT-MDT),

which shows significant changes in the D (1345 cm-1) and G (1590 cm-1) bands and

the emergence of 2D (2671 cm-1) band (Fig. 5.2).

Figure 5.3 Raman mapping of the GO film coated on cover glass.

The Raman mapping of G band image is shown in Fig. 5.3. The GO-based film

is coated onto a glass cover glass. The image clearly distinguishes the glass (left)

from the film (right).



Figure 5.4 Refractive index (n) and extinction coefficient (k) for GO and GML

films with broadband wavelength from 200 to 1600 nm.

The effective refractive index (n) and extinction coefficient (k) of a 5-layer GO

film and GMLs are measured using a spectral ellipsometer (Fig. 5.4). Within the

measured wavelength range, the GO exhibits a high refractive index around 2

before the reduction. After the reduction, the refractive index of the GMLs

significantly increases to ~ 2.6 at the near-infrared range. This dramatic change (∆n

≈ 0.5) in the refractive index is more than one or two orders of magnitude larger

than the conventional refractive materials147, offering a large dynamic tuning range

for potential phase modulations in photonic devices. On the other hand, the GO film

exhibits a low extinction coefficient (k) over the measured wavelength range as

compared with the GMLs due to the existence of the OFGs. An extinction

coefficient change of ∆k≈0.25 has been achieved over a broadband wavelength,

providing a mechanism to achieve efficient amplitude modulation.

5.3 Tunable metamaterial platform

In this section, we demonstrate a graphene-based metamaterial platform can be

constructed by the GMLs consisting of alternating monolayer GO or graphene and

the dielectric layer. The optical properties can be further tuned by laser and

multifunctions can be enabled.



5.3.1 Design of graphene-based metamaterial

Figure 5.5 Schematic of the in-situ tunable graphene-based metamaterial.

Figure 5.5 shows the structure of the graphene-based metamaterial, which consists

of periodically alternating nanometre-thin layers of GO or graphene (thickness 𝑡𝑡𝑔𝑔)

and a dielectric (thickness 𝑡𝑡𝑑𝑑). The two boundary states of the structures are shown

in Figs. 5.5a and b. The initial state of the metamaterial consists of alternating GO

layers and dielectric layers (Fig. 5.5a). In contrast, in the final state, the GO is

completely converted to graphene layers. The states in between are all accessible

and controlled by laser power, which controls the conversion extend of the material.

Since each of these layers is much thinner than the wavelength of light, the structure

is a uniform uniaxial metamaterial film with effective parameters.148 The effective

in-plane and out-of-plane permittivities are given by the long-wavelength limit of

the Bloch theory148, as shown below:

𝜀𝜀∥ = 𝑓𝑓𝜀𝜀𝑔𝑔 + (1− 𝑓𝑓)𝜀𝜀𝑑𝑑 (5.1)

𝜀𝜀⊥ = �𝑓𝑓𝜀𝜀𝑔𝑔−1 + (1− 𝑓𝑓)𝜀𝜀𝑑𝑑−1�−1

(5.2)

with 𝑓𝑓 = 𝑡𝑡𝑔𝑔/(𝑡𝑡𝑔𝑔 + 𝑡𝑡𝑑𝑑) the graphene filling fraction, 𝜀𝜀𝑑𝑑 the permittivity of the

dielectric, and 𝜀𝜀𝑔𝑔 the frequency-dependent, in-plane permittivity of a graphene

layer given by144

(5.3) 0

1gg

it

σεε ω

= +



where σ is the surface conductivity and 𝜀𝜀0 the permittivity of vacuum. The

theoretical conductivity of graphene is144:

0

20 0

2 2

2 2( ) tanh tanh2 4 4

( 2 ) 4log2 ( 2 ) (2 )

F F

B B

F F

F B

E Ek T k T

E Ei iE k T i

s w ws w

s w sp w p w g

+ −= +

+

− + − + +

(5.4)

where s0 equals to 2 / 4e , EF is the Fermi energy relative to the Dirac point and g

is the intraband scattering rate. T is the temperature, and kB is the Boltzmann

constant. In this expression, the first two terms correspond to the interband

transitions, while the third term is the Drude-like intraband conductivity.

From Eq. (5.4) it is showing that the optical conductivity of graphene material

can be effectively tuned by the Fermi energy level as well as the bandgap of the

material (Fig. 5.5). While most previous theoretical work has concentrated on using

high-mobility graphene that may be obtained from mechanically exfoliated or

epitaxially grown samples. We use GO in solution because it allows the application

of solution-based film synthesis technique. In addition, it can be directly coated

onto arbitrary surfaces without transferring requirements. More importantly, it has

been revealed that the as-synthesized GO undergoes insulator-semiconductor-semi-

metal transitions with reduction149 as the bandgap decreases. Thus, the tuning range

of the conversion from GO to graphene is much larger than doping in pristine

graphene144. Furthermore, the bandgap can be accurately controlled by varying the

degrees of reduction. In this way, we are able to tune the effective parameters of

the metamaterial by tuning the degrees of reduction. In the meantime, the two

boundary states are controlled by the filling ratio (f) and the permittivity of the

dielectric (εd).

Here we design the graphene metamaterial for applications in biophotonic

devices in a water environment. Thus, it is required the effective parameters are all

positive for the ease of device design and effective refractive indices should be

lowered from pristine graphene to minimize the reflection at the interface of

metamaterial and water. In addition, it is preferred optical modulation as strong as

possible to achieve high performance.



Figure 5.6 The simulation of real (a) and imaginary (b) parts of the permittivity of

GO metamaterial with different dielectric layer thickness. Simulated real (c) and

imaginary (d) parts of the permittivity of graphene-based metamaterial with

different dielectric spacing layer thickness.

The tuning range is shown in Figure 5.6. Filling ratio is defined as𝑓𝑓 = 𝑡𝑡𝐺𝐺/𝐺𝐺𝐺𝐺

𝑡𝑡𝐷𝐷+𝑡𝑡𝐺𝐺/𝐺𝐺𝐺𝐺,

tG, tGO and tD are the thickness of the GO layer, graphene layer and dielectric spacing

layer respectively. Here, tG = 3.4 Å and tGO = 8.1 Å are constants which indicates

tD is the key parameter for manipulation of filling ratio. As discussed previously,

filling ratio strongly affects the optical properties in the metamaterial structure.

Figure 5.6 illustrates the huge tuning range in the change of permittivity for both

GO and graphene metamaterial by controlling the thickness of dielectric spacing

layer. We have the balanced the thickness of the dielectric layer for ~2 nm, which

gives us good optical properties in the metamaterial structure.

By carefully designing our graphene-based metamaterial, we find a balance

between all requirements. We control the thickness of the graphene layers down to



nanometer accuracy while fixing the thickness of the dielectric layer at 𝑡𝑡𝑑𝑑 = 2 nm.

The dielectric layer has a refractive index around 1.5 and near zero absorption in

visible to near infrared region.

5.3.2 Experimental realization of laser tunable graphene-based

metamaterial

Figure 5.7 Optical microscopic images of laser reduced GO films with 8 reduction

levels.

Laser reduction is able to convert GO to graphene material37,83,84 and is an effective

approach to adjust the bandgap of GO. Upon laser-reduction, the OFGs are removed

and GO undergoes an insulator-semiconductor-semi-metal transition process149 due

to the decrease of the bandgap. The tuning is well-controlled by laser power as

shown in Fig. 5.7, in which 8 distinct reduction levels with the reduction power

from 3 µW to 10 µW (Coherent Libra, femtosecond laser, 800 nm, 10 KHz

repetition rate, 100 fs pulse width, focused by a 100×, 0.8NA objective lens) can be

visually identified in the microscopic images. The tuning range is decided

experimentally from identifiable contrast after reduction (3 µW) to the laser

ablation threshold (11 µW). The levels in between are all accessible continuously

by fine-tuning the laser power.

Figure 5.8 Refractive index (n) change versus (a) and extinction coefficient (k) (b)

versus different laser power for graphene metamaterial with broadband

wavelength from 200 to 1600 nm.



The effective refractive index (n) and extinction coefficient (k) of the

metamaterial before and after laser tuning are measured using a spectral

ellipsometer (Figs. 5.8a & b). Before laser tuning, the metamaterial exhibits a

moderate refractive index around 2 within the measured wavelength range. After

complete laser reduction (laser power equals to 10 µW), the refractive index of the

metamaterial significantly increases by about ∆n ≈ 0.5, which is more than one or

two orders of magnitude larger than the conventional refractive materials147,

offering a large dynamic tuning range for potential phase modulations in photonic

devices. On the other hand, the metamaterial before tuning exhibits a low extinction

coefficient (k) over the measured wavelength range. An extinction coefficient

change of ∆k≈0.25 has been achieved over a broadband wavelength (laser power

equals to 10 µW), providing a mechanism to achieve efficient amplitude

modulation. When the laser power increases over 11 µW, the GO film will be

ablated (Fig. 5.7).

Figure 5.9 Bandgap tunning range from GO to rGO with laser power.

The measured n and k values during the laser reduction process allow us to

calculate the bandgap of the GO/graphene layers, which gradually decreases from

2.1 eV of GO to 0. 1 eV after complete laser reduction (Fig. 5.9). The non-zero

bandgap is due to the existence of the small amount of defects in the reduced GO

material, which form nanoflakes of the graphene.



Figure 5.10 Real and imaginary parts of the optical conductivity of GO and GML

film compared with a CVD grown graphene film, respectively.

Furthermore, we study the optical conductivity (σ) of the GO or graphene layers

calculated by Eqs. (5.1-5.3). The deduced surface conductivity is extremely close

to the CVD grown graphene (Fig. 5.10).144 Therefore, it can be confirmed that the

properties of graphene have been maintained in the metamaterial. In the meantime,

the layer thickness of GO (8.1 Å) gradually decreases to graphene (3.4 Å), which

reduces the filling ratio of the metamaterial and results in an overall increase in the

effective permittivity and absorption.



5.4 Laser patterning characterization and the design of

amplitude modulation device

Figure 5.11 The fabricated lines with different laser power 5 µW (left) and 4 µW

(right) measured with an atomic force microscope (AFM).

As studied in the above sections, the laser is an effective tool to tuning the optical

properties of graphene-based material. To further quantify the laser fabricated

features and develop the laser fabrication process, the line thickness and width were

first correlated to the laser power for a 22-nm thick GO film (Fig. 5.11) measured

by AFM (NT-MDT). The colour gradient represents the variation in thickness

before and after laser fabrication. Both of that show a monotonic nonlinear

dependence with a saturation trend at high power as shown in Fig. 5.12. The

thickness and width are characterized by AFM. The error bars on both thickness

and linewidth are determined via resampling. The smallest feature size can be

fabricated is 300 nm, which is approaching the diffraction limit of the fabrication

laser.



Figure 5.12 Plot of line thickness and width versus laser power.

Pre-designed arbitrary patterns can be fabricated as shown in both the

microscopic and optical profiler images in Fig. 5.13. A Swinburne logo fabricated

by the laser direct writing on an ultrathin GO film is characterized by a microscopic

image (top) and 3D surface profile (bottom). The microscopic image is taken by a

Nikon N-STORM microscope. The surface profile is measured by an optical

profiler (Bruker ContourGT InMotion). The high colour contrast and resolution

prove the potential of graphene-based metamaterial for amplitude modulation

devices.

Figure 5.13 Microscopic image (top) and 3D surface profile (bottom) of a

Swinburne logo on an ultrathin GO film (30 nm) fabricated by the laser direct

writing method.



To demonstrate the versatility of the graphene-based metamaterial, amplitude

modulation based functional photonic devices was directly patterned on an ultrathin

18-nm thick film using the direct laser writing method. A macroscopic optical

image of a designed QR code pattern (Fig. 5.14 a) covering an entire cover glass

(22 mm × 22 mm) (Fig. 5.14b) is achieved, which can be read out as shown in Fig.

5.14c. Due to the significant change in absorption of graphene-based metamaterial

upon laser irradiation, strong amplitude modulation can be achieved by the ultrathin

film, leading to the direct inscription of information in functional information

encoding graphene-based metamaterial device. Combining with the capability of

attaching to any arbitrary substrate with any shapes, the graphene-based

metamaterial film provides a novel platform for optical data storage and encryption.

Figure 5.14 (a) Design of a QR code. (b) Laser-written QR code on an 18-nm

thick GO film coated on a cover glass. (c) Readout result of the QR code.

5.5 Phase modulation device design for microfluidic bio lens

The graphene-based metamaterials are resilient to water thanks to the solution-

based LBL synthesis method as detailed in Chapter 334. As a result, the graphene

metamaterial builds up a versatile platform for a lab-on-a-chip system for

microfluidics and bio-sensing/imaging applications. To demonstrate the unique

property and design flexibility, a water immersion lens in microfluidic device was

designed and fabricated in this section.

5.5.1 Design of the phased based microfluidic bio lens

For laser modulated GO region in this lens, the achievable phase modulation φ can

be expressed as 𝜑𝜑 = ∆𝑛𝑛 × |𝒌𝒌| × ∆𝑡𝑡, where ∆n is the refractive index difference

between the lens and the surrounding medium, k (|k|=2π/λ) is the wave vector of



light and ∆t is the thickness of the lens. The schematic of a well-designed

microfluidic device was fabricated as shown in Fig. 5.15.

Figure 5.15 Theoretical design of the flat lens (a) and the simulated focal spot (b).

For our GO lens design, the focusing is achieved through the interference of

light passing through the GO and graphene-metamaterial zones, which provide

phase and amplitude modulations simultaneously. In Fig. 5.15, the white area is GO

film and the black lines represent the laser reduced GMLs area. The phase

modulation between the adjacent GO (white area) and GML (laser reduced area,

black lines) zones can be expressed as:

∆𝜙𝜙 = 2πλ

. (𝑅𝑅2 − 𝑅𝑅1) + ∆𝜑𝜑 (5.5)

where Δφ is the phase modulation between GO and GMLs. Given the fact that the

phase modulation within one graphene-metamaterial zone is comparatively weak

due to the ultra-thin thickness of the GO film, it is physically sound to approximate

the phase modulation Δφ as a constant across each graphene-metamaterial zone. R

is the radius of each ring. R1 is the radius of first GO ring and R2 is the radius of the

first GML ring (Fig. 5.15). The difference between R1 and R2 is the width of the

laser reduced GML line. To guarantee a constructive interference between all the

GO zones and graphene-metamaterial zones, ∆ϕ should be fixed at 2 π ,

corresponding to constructive interference. As a result, we can get:

(𝑅𝑅𝑚𝑚 − 𝑓𝑓) = 𝑚𝑚𝑚𝑚 − 𝑚𝑚 . 2π∆𝜑𝜑

(5.6)



where f is the focal length of the GO lens, Therefore, by solving the Eq. (5.6) using

the relation R2m = a2

m + f2m, we can easily obtain that:

a= λ ⋅ f (2m− ∆ϕ /π ) (5.7)

During the laser-induced reduction process of GO, three physical properties

(film thickness, refractive index and extinction coefficient) are correlated and all

dependent on the reduction extent, which is eventually controlled by the laser

power. Therefore, for a given phase modulation, the corresponding amplitude

modulation is determined. As a result, both the amplitude and phase modulations

can be considered at the same time by only including the phase modulation factor

in the Eq. (5.7) for the lens design. The amplitude and phase modulations were

optimized according to the following two criteria. i): Maximizing the constructive

interference between the adjacent GO and graphene-metamaterial zones. ii):

Minimizing the destructive interference between the adjacent GO and graphene-

metamaterial zones. In this way, it is possible to achieve the best

focusing/interference condition for a given phase/amplitude modulation to

eventually optimize both the focal spot size and the focusing efficiency.

5.5.2 Experimental realization of water-proof GO microfluidic lens

Figure 5.16 Schematic of the GO lens operating in a microfluidic device with a

biocompatible solution (a) and image of the experimental setup (b).



The schematic of a well-designed microfluidic device was fabricated as shown in

Fig. 5.16(a). Then the experimental microfluidic device is presented in Fig. 5.16b.

The theoretical design of the flat lens is presented in Fig. 5.15 left and the simulated

focal spot is shown in Fig. 5.15 right. The optical image of the fabricated flat lens

is shown in Fig. 5.17(a) (left) with a measured focal spot in Fig. 5.17(b) (right). The

experiment curve (black line) agrees well with the theory one (blue line with dots

in Fig. 5.18(b), which can be attributed to the small roughness of the graphene-

based metamaterial and the precise control of the laser fabrication. By taking the

advantages of both phase and amplitude modulations, the full width at half

maximum (FWHM) is ~450 nm (0.6 λ), indicating almost diffraction-limited

performance has been achieved in an 18 nm GO film, which is among the thinnest

lens demonstrated, but with the best performance.

Figure 5.17 (a) Microscope image (left) and focal spot of the laser inscribed flat

lens on an 18-nm thick GO film. (b) Microscope image (left) and focal spot of the

flat lens after working in microfluidic devices for one month (right).

In order to quantify the durability of the graphene-based metamaterial in a water

environment, we performed the same measurement, including optical microscopic

imaging (Fig. 5.17(b) left) and intensity distribution (Fig. 5.17(b) right)

measurement, after one month immersing the lens in water. As illustrated in Fig.

5.18(a), the focal spot of the lens remains identical, demonstrating the water-proof

nature of our graphene-based metamaterial. By comparing the intensity distribution

curve before (black line) and after one month of immersion (red line) in a

microfluidic device (Fig. 5.18(a)), the two overlapped curves indicate that the



performance of flat lens has no degradation when working in the microfluidic

device. In addition, optical profiler images confirmed the surface profiles of the lens

are maintained without swelling (Fig. 5.18(b)). All these results verify that the

graphene-based metamaterial survives without any compromise of the morphology

or optical performance. Such a flat lens design demonstrates that the combined

phase and amplitude modulations in an ultrathin graphene-based metamaterial is

sufficient to achieve multiple functional diffractive optical devices in the aqueous

environment.

Figure 5.18 (a) The intensity distributions of the GO lens for theoretical flat lens,

laser inscribed flat lens and flat lens after immersion in a microfluidic device for 1

month. (b) Topographic profile of GO lens (left) and cross section of the thickness

(right).

5.6 Conclusion

In this chapter, we have realized a graphene-based metamaterial constructed by a

low-cost, scalable solution-based method for the first time. The effective

parameters of the metamaterial are controlled by the filling ratio and the permittivity

of the spacing material. In addition, the effective parameters can be in-situ and

localized tuned by laser processing. The photonic devices based on amplitude



modulation, a QR code device, and phase modulation, a flat lens, are demonstrated.

The graphene-based metamaterial is resilient to water. Thus we demonstrated a

high-performance lens fabricated in the graphene-based metamaterial that is able to

focus optical energy with subwavelength resolution in the aquatic environment. The

graphene-based metamaterial is expected to find broad applications in lab-on-chip,

microfluidics, bio-medical optoelectronic and integrated devices.

Chapter 6: Hybrid graphene oxide coated integrated photonics devices


6 Hybrid graphene oxide

coated integrated

photonics devices

6.1 Introduction

Without complicated optical-electrical-optical conversion process, directly

processing the signal in the optical domain is a key approach that can dramatically

decrease power consumption and speed up optical telecommunications. Four-wave-

mixing (FWM) is a fundamental process in nonlinear optics that has been widely

explored to realize all-optical signal processing functions such as wavelength

conversion, optical logic gates, optical sampling, and parametric

amplification.96,150,151 Realization of FWM based on photonic integrated circuits

(PICs) provides a competitive solution for on-chip nonlinear processing with a

compact footprint, high stability, mass-producibility, and excellent cost

performance.96,102

Unfortunately, although silicon (Si), the leading platform for integrated

photonic devices, has a high Kerr nonlinear coefficient (n2), it also suffers from

strong two-photon absorption (TPA) in the telecommunications band. TPA also



leads to parasitic free carrier effects, which seriously degrade the FWM conversion

efficiency (CE) and slows down the processing speed. The implementation of

hybrid integrated photonic devices with other materials having a high n2 and weak

TPA, such as graphene, opens a way towards improving the performance of the

nonlinear silicon photonic devices. Owing to its ease of preparation as well as the

tunability of its material properties, GO has become a highly promising member of

the graphene family.60 Previously,74,75 our group reported GO films with a giant

Kerr nonlinear response of 4 to 5 orders of magnitude higher than that of high-index

doped silica glass. Moreover, as compared with graphene, GO has much lower loss

and larger bandgap (2.4~3.1 eV) 152,153 which yields low TPA in the

telecommunications band. It also offers the better capability for large-scale

fabrication,27 critical for the practical implementation of high-performance

nonlinear photonic devices.

The TPA effect at near-infrared wavelengths poses a fundamental limitation to

the nonlinear optical performance of silicon photonic devices in the

telecommunications band.150 Hence, the quest for high-performance integrated

platforms for nonlinear optics has motivated the development of other CMOS

compatible platforms such as silicon nitride and high-index doped silica glass.96

Benefiting from extraordinarily low linear and nonlinear loss, high-index doped

silica glass has been a successful integrated platform for nonlinear photonic

devices. 9 109,110 Nevertheless, its relatively low Kerr nonlinearity as compared with

silicon and silicon nitride96 limits its performance for nonlinear optical processes.

The use of highly nonlinear materials that can be integrated on-chip such as GO,

could overcome these limitations. Schematics of GO coated silica waveguide is

shown in Fig. 6.1 and explored in this chapter.



Figure 6.1 Schematic of GO coated silica planarized integrated waveguide.

In this chapter, the nonlinear theory is introduced in section 6.2. Device

fabrication and characterization are shown in 6.3. In section 6.4, the experimental

results are presented and analysed. The discussion for the experimental results is

given in section 6.5. The simulation of GO-Si hybrid is presented in section 6.6. At

the end of this chapter, a brief summary is given in section 6.7.

6.2 Nonlinear theory of GO coated integrated photonic

devices

We used the theory in Refs. 154 and 155 to model and analyze the FWM performance

of the GO hybrid integrated waveguides. First, as in the common practice, we

assume that the χ(3) nonlinearity 156 for FWM is approximately equal to the Kerr

nonlinearity n2 – i.e.,

χ(3) (ωidler; +ωsignal, +ωpump, -ωpump) ≈ χ(3) (ω; ω, ω, -ω) = n2 (ω) (6.1)

where 2ωpump = ωidler + ωsignal. Note that although Eq. (6.1) is often considered for

FWM, it is only valid in the regime close to degeneracy where the three FWM

frequencies (pump, signal, idler) are close together compared with any variation in

n2 arising from the dispersion in n2. Since n2 is expected49 to vary significantly near

the material bandgap, there is no guarantee that Eq. (6.1) holds in cases where the

material bandgap is comparable to, or even smaller than, the photon energies being

employed. This particularly applies to graphene, for example. Further, Eq. (6.1)

also ignores the 4th rank tensor nature of χ(3). It is well known157,158 that even for



cubic materials such as silicon, the tensor elements for χ(3) are not all equal, resulting

in a variation in the nonlinear efficiency of up to 50% or more as a function of

orientation in silicon or germanium. While the doped silica waveguides studied here

are isotropic, it is not only possible but highly likely that the nonlinearity of the thin

GO films is indeed highly anisotropic.

This effect is in addition to any variations in the effective γ arising from

differences in the mode overlap with the GO films between the two polarizations.

In our experiments, we restricted the polarization to TE and so effectively we

measure the n2 that applies to this polarization. We note that previous measurements

in n2 in these films were performed in transmission in broad area films and so

correspond to TE polarization in our experiments. This topic (the anisotropy of

nonlinear processes in GO films) will be the subject of future work.

Considering the linear loss, the nonlinear loss induced by TPA, self-phase

modulation, and cross-phase modulations, the coupled differential equations for the

FWM processes can be expressed as 154,155

dAp(z)

dz = �-αp

2 -βTPA2Aeff

�Ap(z)�2�Ap(z)+jγp ��Ap(z)�2+2|As(z)|2+2|Ai(z)|2�Ap(z)

+j2γpAp*(z)As(z)Ai(z)exp(j∆βz), (6.2)

dAs(z)

dz = �-αs

2 -βTPAAeff

�Ap(z)�2�As(z)+jγs �|As(z)|2+2�Ap(z)�2+2|Ai(z)|2�As(z)

+jγsAi*(z)Ap

2(z)exp(-j∆βz), (6.3)

dAi(z)dz = �-

αi

2 -βTPAAeff

�Ap(z)�2�Ai(z)+jγi �|Ai(z)|2+2�Ap(z)�2+2|As(z)|2�Ai(z)

+jγiAs*(z)Ap

2(z)exp(-j∆βz). (6.4)

where Ap,s,i are the amplitudes of the pump, signal, and idler waves along the z-axis,

which we assume as the light propagation direction. αp,s,i are the linear loss factors.

βTPA is the TPA coefficient. Aeff is the effective mode area. h is the Planck constant.

c is the speed of light in vacuum. λp is the pump wavelength. Δβ = βs + βi - 2βp is

the linear phase mismatch,with βp,s,i denoting the propagation constants of the



pump, signal, and idler waves. γp,s,i are the nonlinear parameters calculated by way

of a full-vector finite-element-method based on the following equation:59

,)(

),(22

22

,,,,

∫∫∫∫=

dxdyS

dxdyyxnS

zD

zD

ispisp λ

πγ (6.5)

where Sz is the time-averaged Poynting vector, n2 is the Kerr coefficient, and D is

the integral domain including either propagating or evanescent optical fields. All

the pump, signal, and idler waves in Eqs. (6.2-6.4) are assumed to be polarized

along the TE axis. γp,s,i in Eq. (6.5) can be regarded as an effective nonlinear

parameter of the GO hybrid waveguides with different material areas weighted by

their n2.

By numerically solving Eqs. (6.2-6.4), we can obtain the FWM efficiency

defined as:

η (dB) = 10 × log10[|Ai(L)|2/|As(0)|2], (6.6)

where L is the length of the waveguide. Note that η is the ratio of the output power

of idler light to the input power of signal light, i.e., Pout, idler/Pin, signal.

6.3 General nonlinear simulation strategy

From the well-studied nonlinear theory in the previous section, dispersion D,

effective index neffective and nonlinear parameter γ are the key parameters in the

simulation process. The general simulation process is shown in Fig. 6.2. Here we

use Lumerical MODE solution software to calculate D and neffective. The software

panel is depicted in Fig. 6.3. COMSOL software is introduced to calculate nonlinear

parameter γ. COMSOL software simulation panel is shown in Fig. 6.4. In the end,

we use MATLAB to solve the Eqs. (6.1-6.3) with calculated D, neffective and γ and

finally get the conversion efficiency of FWM. The MATLAB software simulation

panel is shown in Fig. 6.5.



Figure 6.2 General process of simulation strategy.

Figure 6.3 Lumerical MODE solution software simulation panel.



Figure 6.4 COMSOL software simulation panel.

Figure 6.5 The MATLAB software simulation panel.

For nonlinear parameter γ, the combined nonlinear parameter is needed after

coating with GO film. The combined nonlinear parameter γ is defined in Eq. (6.5),

6.4 Device fabrication and characterization

Figure 6.1 presents a schematic illustration of the GO-coated integrated waveguides

made from high-index doped silica glass,96 with a cross section of 2 μm × 1.5 μm.

The integrated waveguide is surrounded by silica except that the upper cladding is

removed to enable coating of the waveguide with GO films. The films, with a

thickness of about 2 nm per layer, were introduced on top of the integrated



waveguide in order to introduce light-material interaction with the evanescent field

leaking from the integrated waveguide.

The Kerr coefficient of GO is on the order of 10-15~10-14 m2/W,74,75 which is

slightly lower than that of graphene (~10–13 m2/W),58,26,59 but still orders of

magnitude higher than that of high-index doped silica glass (~10–19 m2/W) and silica

(~10–20 m2/W).

The waveguides were fabricated via CMOS compatible processes. 159 First,

high-index doped silica glass films (n = ~1.60 at 1550 nm) were deposited using

standard plasma enhanced chemical vapour deposition (PECVD), then patterned

photo-lithographically and etched via reactive ion etching (RIE) to form

waveguides with exceptionally low surface roughness. After that, silica glass (n =

~1.44 at 1550 nm) was deposited via PECVD and the upper cladding of the

integrated waveguides was removed by chemical mechanical polishing (CMP).

Finally, the GO film was deposited on the surface of the planarized integrated

waveguide by the layer-by-layer method that yields precise control of the layer

number detailed in Chapter 3.160 As compared with graphene, GO can be processed

in solution,34 which makes it easy to grow in a large area with few defects, a critical

factor for the fabrication of large-scale integrated devices.

Figure 6.6 Micrograph of the hybrid waveguide with two layers of GO.

An optical image of the integrated waveguide incorporating two layers of GO

is shown in Fig. 6.6, which illustrates that the morphology is good, leading to a high

transmittance of the GO film on top of the integrated waveguide.



Figure 6.7 Raman spectra of GO on the integrated chip.

The integration of GO onto the integrated waveguide is confirmed by Raman

spectroscopic measurements that show the representative D (1345 cm-1) and G

(1590 cm-1) peaks of GO (Fig. 6.7).

The refractive index (n) and extinction coefficient (k) of the 5-layer GO film

measured by means of a spectral ellipsometer are shown in Fig. 5.4. For

comparison, the refractive index and extinction coefficient of graphene are also

shown. The GO film exhibits a high refractive index of ~2 in the

telecommunications band. On the other hand, due to the existence of oxygen

functional groups (OFGs), the GO film also exhibits an ultra-low extinction

coefficient in the telecommunications band, which leads to much lower material

absorption as compared with graphene. This property of GO could be of benefit for

FWM devices, where low loss is always desired for improved efficiency. It should

be noted that unlike graphene that has zero bandgap, GO has a distinct bandgap of

2.4 to 3.1 eV,152,153 which results in greatly reduced TPA in the telecommunications

band – a significant advantage.

To couple light into and out of the hybrid integrated waveguides, we employed

an 8-channel single-mode fibre (SMF) array for butt coupling. Since the

propagation loss of the doped silica waveguides was negligible for the lengths

studied here (contributing less than 0.1 dB), the total insertion loss for bare

(uncoated) waveguides was determined by the coupling loss, which we measured

as ~8 dB per facet. This can be reduced to ~1.5 dB/facet by using mode converter



between the SMF and the integrated waveguide modes. We chose the TE

polarization for the experiments because it supports in-plane interaction between

the evanescent field and the GO thin film, which is much stronger than out-of-plane

interaction due to the large optical anisotropy of 2D materials like graphene and

GO128.

Figure 6.8 (a) Measured insertion loss of hybrid waveguides with different

numbers of GO layers. (b) The additional propagation loss of the hybrid

waveguide with different numbers of GO layers.

Figure 6.8 (a) depicts the measured total insertion loss of the integrated

waveguides with different numbers of GO layers at a wavelength of 1550 nm. We

characterized four duplicate integrated waveguides with the same length of ~1.5

cm. The GO layers only affected the propagation (not coupling) loss, which is

shown in Fig. 6.8 (b) as a function of the number of layers. The overall propagation

loss of the GO hybrid integrated waveguides was on the order of a few dB/cm,

which is much lower than that of the graphene hybrid integrated waveguides161 and

confirms the low material absorption of GO in the telecommunications band. The

measured insertion loss did not show any significant variation with input power

even when the CW light power was increased up to ~33 dBm. Another interesting

phenomenon is that the rate of increase in propagation loss with layer number

increased (ie., became super-linear) at higher numbers of GO layers. This might be

attributed to interactions among the GO layers and an imperfect contact between

the multiple GO layers.



6.5 Experiment results

We used the experimental setup shown in Fig. 6.9 to perform FWM measurements

in the GO hybrid integrated waveguides. Two CW pump beams from tunable lasers

were separately amplified by two erbium-doped fiber amplifiers (EDFAs) and

served as the pump and signal sources, respectively. In each path, there was a

polarization controller (PC) to make sure that the input light was TE-polarized. The

pump and signals were combined by a 50:50 fibre coupler before being injected

into the GO hybrid integrated waveguide. The signal output from the waveguide

was sent to an optical spectrum analyser (OSA) with a variable optical attenuator

(VOA) inserted before the OSA to prevent the high-power output from damaging

it.

Figure 6.9 Experimental setup for testing FWM in the GO hybrid integrated

waveguide. EDFA: erbium-doped fiber amplifier. PC: polarization controller.

DUT: device under test. OSA: optical spectrum analyser. VOA: variable optical

attenuator.

Figure 6.10 shows the FWM experimental results. The FWM spectra of a 1.5-

cm-long integrated waveguide without GO and with 2 layers of GO are shown in

Fig. 6.10 For comparison, we kept the same pump power of ~30 dBm before the

input of the waveguide, which corresponded to ~22 dBm pump power coupled into

the waveguide. It can be seen that although the hybrid integrated waveguide had

additional propagation loss (~2.6 dB), it clearly shows enhanced idler output

powers as compared with the same waveguide without GO. The CE (defined as the

ratio of the output power of the idler to the output power of the signal, i.e., Pout,

idler/Pout, signal) of the integrated waveguide with and without GO were -47.1 dB and

-56.6 dB, respectively, corresponding to a CE enhancement of 9.5 dB for the hybrid

integrated waveguide. After excluding the additional propagation loss, the net CE

enhancement (defined as the improvement of the output power of the idler for the

same pump power coupled to the waveguide) is 6.9 dB.



Figure 6.10 FWM spectra of the integrated waveguide without GO and with 2

layers of GO.

For the integrated waveguide coated with 1 to 5 layers of GO, zoom-in spectra

of the generated idlers for the same pump power coupled to the waveguide (~22

dBm) are shown in Fig. 6.11. For the integrated waveguide coated with 1 and 2

layers of GO, there were positive net CE enhancements. When the number of GO

layers was over 2, the net CE enhancements were negative. This is mainly due to

the super-linear increase in the propagation loss for increased numbers of GO layers

noted above, and the trade-off between FWM enhancement and propagation loss.

Figure 6.11 Zoom in spectra of the generated idlers after FWM in the integrated

waveguide with 0 to 5 layers of GO.



The output powers for the idler for various pump powers coupled to the

waveguide without GO and with 2 layers of GO are shown in Fig. 6.12. One can

see that as the pump power increased, there is a nearly linear increase in idler power

with no obvious saturation for both conditions, which reflects the low nonlinear

absorption of both the high-index doped silica glass as well as the GO layers in the

telecommunications band.

Figure 6.12 Output powers of idler for various pump powers coupled to the

waveguide without GO and with 2 layers of GO.

The net CE enhancements for various pump powers coupled to the waveguide

with 1 to 5 layers of GO are shown in Fig. 6.13. There are positive net CE

enhancements for all the measured pump powers when the waveguide is coated

with either 1 or 2 layers of GO. We also note that there is a slight increase in CE

enhancement for high pump powers. This may be due to the fact that the material

properties of the GO films become slightly different at high powers.74



Figure 6.13 Net CE enhancements for various pump powers coupled to the

waveguide with 1 to 5 layers of GO.

The variation in idler power when the pump wavelength was fixed at 1550 nm

and the signal wavelength was detuned from −10 nm to 10 nm around the central

value, is shown in Fig. 6.14(a). There is obvious power degradation of the output

idler when the wavelength detuning is beyond ±5 nm. When the wavelength

detuning is beyond ±7.5 nm, we cannot observe any idler above the noise floor. The

output powers of idler light for a 1.5-cm-long integrated waveguide coated with

different lengths of GO are depicted in Fig. 6.14(b). We used three GO coating

lengths of ~0.5 cm, ~1.0 cm, and ~1.5 cm. For the waveguide coated with 2 layers

of GO, the output idler power increases with the GO length. Whereas for the

waveguide coated with 5 layers of GO, there is an opposite trend. This phenomenon

further confirms the trade-off between FWM enhancement and propagation loss in

these hybrid integrated waveguides.



Figure 6.14 (a) Power variations of the output idler when the pump wavelength

was fixed at 1550 nm and the signal wavelength was detuned from -10 nm to 10

nm. (b) Output powers of idler for the waveguide with different coating lengths of

GO.

6.6 Results and discussion

The simulated TE mode profile of the hybrid integrated waveguide with 2 layers of

GO is presented in Fig. 6.15(a). Figures 6.15(b) shows the measured and fit FWM

efficiency η for different pump powers. In the calculations, we used the pump power

coupled to the waveguide in our FWM experiment as the input pump power. Figures

6.16(a) and (b) show the measured and fit FWM efficiency η for wavelength

detuning, and increasing GO lengths. We also used the experimentally measured

values for the dispersion and loss of GO in the simulation.

Figure 6.15 (a) TE mode profile of the hybrid waveguide with 2 layers of GO at

the wavelength of 1550 nm. (b) The plot of η as a function of pump power.



Figure 6.16 (a) − (b) η as a function of pump power, wavelength detuning, and

GO lengths, respectively. The dots represent the experimentally measured values

and the lines show the fit curve calculated based on Eqs. (6.2-6.6). WG:

waveguide.

Finally, as we showed previously,74 GO films can be photo-reduced to provide

a continuously variable range of both n2 and the FOM, including even regimes

displaying both a negative n2 and negative nonlinear absorption (saturable

absorption). This opens up the very powerful possibility of patterning structures to

achieve quasi-phase matching and even using the negative n2 in normal dispersion

waveguides. Hence the potential and power of GO films for enhancing nonlinear

processes in waveguides and nanowires extend well beyond the enhancement in

overall nonlinear efficiency reported here.

6.7 Simulation of GO-Si hybrid waveguide

Si, the dominant material in the semiconductor industry, has been regarded as the

leading platform in integrated photonics field. In order to explore the potential of

the GO conformal coating and nonlinearity onto Si photonic devices, two kinds of

waveguides are introduced here, Si wire waveguide and slot waveguide. Figure 6.17

presents the GO conformal coating around the Si wire waveguide. The tightly

conformal coating will enable light-material interaction take place on both the top

surface and sidewalls of the wire waveguide. The large interaction area will have

the potential for enhancing the nonlinearity of the hybrid waveguide.



Figure 6.17 Schematic of GO conformal coated Si wire waveguide

Figure 6.18 illustrates the GO conformal-coated Si slot waveguide. Due to the

unique slot structure, the light field will highly be confined in the nanoscale gap

area between the two Si ribs and leads to the strong light material interaction with

filled GO film in such area.

Figure 6.18 Schematic of GO conformal coated Si slot waveguide.



6.7.1 Simulation results of hybrid GO-Si wire waveguide

Min

Figure 6.19 Mode distribution of GO coated Si wire waveguide.

Figure 6.19 presents the TE mode profile of the GO-Si wire waveguide (width:

600 nm and height: 300 nm). The light is highly confined in the centre of the

waveguide. When the GO film conformal coated onto the wire waveguide, the light

will interact with GO on both the top surface of the wire waveguide and sidewalls.

This interaction will cause the increasing of the nonlinearity of the hybrid

waveguide. The simulated CE versus signal wavelength WLs (1500-1600 nm) is

shown in Fig. 6.20. Four groups of Wwire × Hwire are selected: I - 600 nm × 300 nm,

II – 500 nm × 220 nm, III - 600 nm × 110 nm, and IV - 800 nm × 60 nm. The solid

and dashed curves correspond to the waveguides with and without GO,

respectively. The CE is calculated by resolving Eqs. (6.1-6.3) with MATLAB. The

waveguide dispersion is simulated by Lumerical MODE Solutions with our GO

experimental results. There is an obvious CE improvement for all the GO-Si hybrid

wire waveguides. The measured n2 of GO is from 0.25 × 10 –13 m2/W to 0.75 × 10

–13 m2/W in our experiment. Here, we use 0.25 × 10 –13 m2/W in our simulation.

Owing to the strong light-material interaction around wire waveguide region coated

with GO, up to ~ 32 dB improvement of CE can be obtained with the structure IV

(800 nm × 60 nm). Such the waveguide structure will enable strong light-material

interaction caused by more leakage from the waveguide to the GO film.



Figure 6.20 Plot of simulated CE versus signal wavelength WLs.

The simulated CE versus pump power PP, waveguide length L is shown in Fig.

6.21. At first, the CE is improved with the increase of the pump power. Then the

CE is saturated after the pump power is over 25 dBm. This effect is caused by the

increased TPA effect in the high pump power situation.

Figure 6.21 The simulated CE versus pump power PP.

The simulated CE versus waveguide length L is shown in Fig. 6.22. The CE is

saturated after the waveguide length is over 1 mm. This effect is caused by the

introduced propagation loss by the GO film. Although the area of light-material

interaction is still increasing with increased waveguide length, the additional loss



will have negative effects on the enhancement of the CE and leads to the saturation

in the Fig.6.22. In the current situation, shorter waveguide length is preferred to

achieve higher CE enhancement.

Figure 6.22 The simulated CE versus waveguide length L.

6.7.2 Simulation results of hybrid GO-Si slot waveguide

The simulation results for the Si-GO slot waveguides are shown in this section. We

chose three different values of Wslot: I - 50 nm, II - 100 nm, and III -150 nm. Figure

6.23 shows the schematic of light-material interaction in slot waveguide. Owing to

the strong light-material interaction in the slot region filled with GO film, the

nonlinear activity of the hybrid waveguide is significantly improved.

Figure 6.23 The schematic of light-material interaction in a slot waveguide.



The simulated CE versus signal wavelength WLs (1500-1600 nm) is shown in Fig.

6.24. Three groups of the width of the slot gap are selected: I - 50 nm, II – 100 nm

and III - 150 nm. The solid and dashed curves correspond to the waveguides with

and without GO, respectively. As the same simulation process presented in 6.3, the

CE is calculated by resolving Eqs. (6.1-6.3) with MATLAB. There is a huge CE

improvement for all the GO-Si hybrid slot waveguides. The measured n2 of GO is

from 0.25 × 10 –13 m2/W to 0.75 × 10 –13 m2/W in our experiment. Here, we use 0.25

× 10 –13 m2/W in our simulation.

Figure 6.24 The simulated CE versus signal wavelength WLs.

The simulated CE versus pump power PP is shown in Fig. 6.25. At first, the CE

is improved with the increase of the pump power. Then the CE starts to saturate

after the pump power is over 25 dBm. This effect is caused by the increased TPA

effect in the high pump power situation.



Figure 6.25 The simulated CE versus pump power PP.

The simulated CE versus waveguide length L is shown in Fig. 6.26. The CE is

saturated after the waveguide length is over 1 mm. In slot waveguide, the light is

propagating in the air between the two Si ribs and this phenomenon will cause

higher propagation loss compared with wire waveguide. The overall propagation

loss of the GO-slot hybrid waveguide will be further increased by the GO film filled

in the gap area. Although the area of light-material interaction is still increasing

with longer waveguide length, the additional loss will have negative effects to the

enhancement of the CE and leads to the saturation and even reduction in the Figure.

In the current situation, shorter waveguide length with low loss is preferred to

achieve higher CE enhancement.

Figure 6.26 The simulated CE versus waveguide length L



Owing to the strong light-material interaction around wire waveguide region

coated with GO, up to ~ 39 dB improvement of CE can be obtained with the

structure I (gap width 50 nm) in the simulation.

6.8 Conclusion

In this chapter, we perform FWM measurements in doped silica waveguides

integrated with thin GO films. We achieve a significant enhancement in the FWM

CE of ~9.5 dB in a 1.5-cm-long waveguide with 2 layers of GO. This results from

the high Kerr nonlinearity, low linear loss, and the strong mode overlap of the GO

films. The value of n2 that we extract from our measurements agrees reasonably

well with our previous Z-scan measurements of thick (≥ 1 µm) films. Further, we

show theoretically that the enhancement in the FWM efficiency through the

integration of thin GO films can be as high as 20 dB in the doped silica waveguides

and even higher (>35 dB) in silicon thin nanowire and slot waveguides. With the

potential for photo-patterning the nonlinearity of the GO films, these hybrid

integrated devices offer a powerful new way to implement high performance

nonlinear photonic devices, thus holding a great promise for future ultra-high-speed

all-optical information processing.

Chapter 7: Conclusions


7 Conclusions

7.1 Conclusions

This thesis investigates the fundamental linear and nonlinear optical properties of

the graphene-based materials, develops a low cost, solution-based layer-by-layer

(LBL) method with high controllability of the film quality and broads its

applications in nanophotonic devices, photovoltaic devices and hybrid integrated

photonics devices. A wide range of topics has been covered in not only theoretical

methods and simulations but also the experimental demonstrations. The physical

understanding of the optical properties and responses, the design and optimisation

of the graphene-based platform and well-studied photonics devices derived from

this platform are presented in the thesis as well. The major achievements can be

summarised as follows:

1. For the first time, the ultra-thin graphene-based film with high quality, scalability,

a well-controlled and low-cost synthesis method for photonics applications are

developed through a solution based LBL deposition method. By exploiting the

surface charge property of GO, it can firmly attach onto the surface of complex

three dimensional (3D) nanostructures in an LBL manner. The graphene multilayer

(GML) structure is achieved with this method. Its morphological properties are

characterized by scanning electronic microscope (SEM) and atomic force

microscope (AFM) and the optical properties are measured by an optical

microscope and spectrometer. Furthermore, this method is applied for large scale



fabrication on various substrates including glass, PET and silicon. Due to the

mechanical properties, the GO film can be well maintained on the flexible substrate

after bending and twisting. The optical properties can be tuned by layer numbers.

Finally, we can precisely control the layer number down to a monolayer (with a

thickness: ~1 nm) onto diverse substrates. The achieved large-scale GO film

preserves its properties with an unprecedented surface roughness of ~2 nm, far

transcending the most stringent optical standard, providing a versatile platform for

realizing various photonic applications.

2. By optimizing the flake sizes, deposition time and surface conditions of the LBL

deposition methods, GO thin films can be in situ conformably coated onto 3D

silicon nanowires (SiNWs) with monolayer controllability. We have

comprehensively analysed the optical and electrical performance of the integrated

GO-SiNWs hybrid structures. Up to 20% reduction of reflection in a broadband

wavelength and enhanced passivation are achieved in GO-SiNWs hybrid structure

compared with solely SiNWs. By applying this method to Si solar cell, the overall

improved performance is observed. A 60% enhancement in carrier lifetime was

measured and a more than 20% improvement is achieved in external quantum

efficiency of the hybrid solar cells. Our studies suggest that the LBL GO conformal

coating provides a novel material platform with a precisely controllable and cost-

effective approach to enhance both the optical and electrical responses in a

nanoscale of optoelectronic devices.

3. The graphene-based metamaterials have been constructed by LBL deposition

method for the first time. The effective parameters of the metamaterials are

controlled by the filling ratio and the permittivity of the spacing material. In

addition, the effective parameters can be in-situ tuned by laser processing in a

localized manner. Both significant amplitude and phase modulations can be

achieved and tuned by controlling the reduction laser power. An amplitude

modulation device of a quick response (QR) code was fabricated directly on the

graphene metamaterials by direct laser patterning, providing a swift way to inscribe

information on diverse substrates and surfaces. Unlike other water-soluble GO

films, the LBL graphene-based metamaterials are resilient to water. Thus we

demonstrated a high-performance lens fabricated in the graphene-based

metamaterial that is able to focus optical energy with subwavelength resolution and



remain operational in water even after immersion in water for a month. The

graphene-based metamaterial is expected to find broad applications in lab-on-chip,

microfluidics, bio-medical optoelectronic and integrated devices.

4. By incorporating graphene-based material with integrated photonics platform,

the hybrid waveguides have been investigated. The four-wave-mixing (FWM)

measurements in hybrid waveguides integrated with thin GO films are realized. A

significant enhancement in the FWM conversion efficiency (CE) of ~9.5 dB in a

1.5-cm-long waveguide with 2 layers of GO has been achieved. The enhancement

is attributed to the high Kerr nonlinearity, low linear loss, and the strong mode

overlap of the GO films with the waveguide. The value of n2 that we extract from

our measurements agrees reasonably well with our previous Z-scan measurements

of thick (≥ 1 µm) films. In addition, we show theoretically that the enhancement in

the FWM efficiency through the integration of thin GO films can be as high as 20

dB in the doped silica. With the potential for photo-patterning the nonlinearity of

the GO films, these hybrid integrated devices offer a powerful new avenue to

achieve high performance nonlinear photonic devices, thus holding a great promise

for future ultra-high-speed all-optical telecommunication and information

processing.

5. Two types of GO-Si hybrid waveguides have been proposed. Simulations of GO-

Si wire waveguide and GO-Si slot waveguide have been presented respectively.

The maximum CE enhancement of ~32 dB can be achieved in hybrid wire

waveguide and ~39 dB in hybrid slot waveguide.

In summary, innovative researches on the ultra-thin graphene-based platform

and its applications in nanophotonics, photovoltaic and integrated photonics

devices have been proposed and demonstrated in this thesis both theoretically and

experimentally. The well-controlled graphene multilayer films show unique

properties that are unavailable to conventional materials such as the tunable linear

and nonlinear optical properties, a versatile patterning capability and property

manipulation by laser-induced fabrication technology, the high nonlinearity,

surface functionalisation possibility, and the mechanical robustness and stress,

integration capability, which are highly demanded for the next generation

ultralightweight, highly efficient, highly integratable, and flexible photonics

systems. The recent findings in this thesis open up new avenues for various



multidisciplinary applications including non-invasive in-situ 3D biomedical

imaging and sensing, all-optical broadband photonic chips, photovoltaic, aerospace

photonics, optical microelectromechanical systems and lab-on-chip devices.

7.2 Outlooks

Although much progress has already been made in developing a graphene-based

ultra-thin platform for photonic applications, there are still many significant

directions yet to be further explored and exploited.

1. Optical applications. Firstly, based on the achieved phase and amplitude

modulations, on-chip flat lens and optical modulators for a specific area such as

aerospace and navigation area can be further investigated. Secondly, graphene

photodetector is one of the promising photonic devices. Unlike semiconductor

photodetectors, which have limited detecting spectral width, graphene-based

material can in principle be used for a wide spectral range from ultraviolet to

infrared and the high operating bandwidth makes it suitable for high-speed data

communications. Thirdly, for nonlinear optical devices, hybrid GO-Si integrated

photonic devices are of great value to explore. By taking advantage of the high Kerr

nonlinearity, the hybrid devices are believed to have enhanced nonlinear

performance in all-optical processing field. In addition, saturable absorber and

mode-locked laser can be realized through utilizing the saturable absorption of

graphene-based material. In the end, due to the anisotropy of the graphene material,

optical polarizers that are the crucial passive component in many optical systems

could be fabricated and integrated with on-chip devices.

2. Electrical applications. Firstly, transparent conductive coatings are widely used

in consumable electronic products such as touch screen, displays and light-emitting

diodes (LEDs) and require a low sheet resistance with high transmittance depending

on the specific application. Graphene-based material meets these requirements

would be a promising candidature for these applications. Secondly, due to its

controllable conductivity, graphene-based material could be applied to fabricate

transistors with high operating frequency and on/off ratio. It may become the

component for next-generation integrated circuit.



3. Energy generation and storage. Firstly, as we have demonstrated in section 4.5,

graphene-based material can act as a useful light trapping layer (such as the

antireflection coating and transparent electrode made by laser reduction) in solar

cell and other energy harvesting systems. Secondly, graphene-based material has

been currently used in batteries and supercapacitors. By utilizing its ultra-thin 2D

nature and large surface area, the novel sandwich could be fabricated that will help

to improve the power density of the current system.

4. Mechanical applications. Graphene-based material could be incorporated into

other material such as architecture and tissue engineering to enhance their

mechanical strength and elasticity. In addition, the mechanical sensor such as

cantilever and probe can be fabricated with the assist of laser fabrication.

5. Magnetical applications. There is relatively less research on the magnetical

application of graphene-based material. Due to the semimetal properties of

graphene-based material, magnetical applications are worth to be investigated.

6. Other applications. Graphene-based material is highly inert and can act as a

corrosion barrier to prevent water and oxygen diffusion. Based on research in this

thesis, graphene-based material can be grown directly on the arbitrary surface and

form a protective conformal layer onto complex surfaces. This technique can be

expended to conductive ink, protective shielding and gas barrier for extreme

conditions.

In conclusion, the LBL graphene-based film provides an exciting material

platform for various scientific disciplines including optics (O), electronics (E),

energy generation and storage (E), mechanics (M), magnetics (M) and other

applications (O) such as conformal protective coating (abbreviation for (OEM)2),

broadening the horizon towards the next generation flexible, all-graphene-based

thin-film technology. Finally, as we discussed in Chapter 3, the multilayer

graphene-based platform enables us to realize each application on individual layers

as the building blocks to assemble the multifunctional on-chip system. In addition,

we are certainly far from the end of the road. Studying bulk materials at the first,

then understand and manipulating down to the monolayers, this trend will still guide

us the way from ignorance to knowledge, from big to small. Our adventure is just



beginning. Despite its young age, the field of 2D materials has advanced

significantly in the last few years and the legend will continue.



Bibliography

1 Novoselov, K. S. et al. Electric field effect in atomically thin carbon films. Science 306, 666-669, (2004).

2 Geim, A. K. & Novoselov, K. S. The rise of graphene. Nat Mater 6, 183-191 (2007).

3 As thin as it gets. Nature Materials 16, 155, (2017).

4 Novoselov, K. S., Mishchenko, A., Carvalho, A. & Castro Neto, A. H. 2D materials and van der Waals heterostructures. Science 353, (2016).

5 Graphene is not alone. Nature Nanotechnology 7, 683, (2012).

6 Schaibley, J. R. et al. Valleytronics in 2D materials. Nature Reviews Materials 1, 16055, (2016).

7 Ashworth, C. 2D Materials: The thick and the thin. Nature Reviews Materials 3, 18019, (2018).

8 Jariwala, D., Marks, T. J. & Hersam, M. C. Mixed-dimensional van der Waals heterostructures. Nature Materials 16, 170, (2016).

9 Sykes, E. C. H. Graphene goes undercover. Nature Chemistry 1, 175, (2009).

10 Li, D. & Kaner, R. B. Graphene-Based Materials. Science 320, 1170-1171, (2008).

11 Chen, D., Feng, H. & Li, J. Graphene Oxide: Preparation, Functionalization, and Electrochemical Applications. Chemical Reviews 112, 6027-6053, (2012).

12 Stoller, M. D., Park, S., Zhu, Y., An, J. & Ruoff, R. S. Graphene-Based Ultracapacitors. Nano Letters 8, 3498-3502, (2008).

13 Park, S. & Ruoff, R. S. Chemical methods for the production of graphenes. Nature Nanotechnology 4, 217, (2009).

14 Novoselov, K. S. et al. Two-dimensional gas of massless Dirac fermions in graphene. Nature 438, 197, (2005).

15 Zhang, Y., Tan, Y.-W., Stormer, H. L. & Kim, P. Experimental observation of the quantum Hall effect and Berry's phase in graphene. Nature 438, 201, (2005).

16 Novoselov, K. S. et al. Room-Temperature Quantum Hall Effect in Graphene. Science 315, 1379-1379, (2007).

17 Bunch, J. S. et al. Impermeable Atomic Membranes from Graphene Sheets. Nano Letters 8, 2458-2462, (2008).



18 Lee, C., Wei, X., Kysar, J. W. & Hone, J. Measurement of the Elastic Properties and Intrinsic Strength of Monolayer Graphene. Science 321, 385-388, (2008).

19 Schedin, F. et al. Detection of individual gas molecules adsorbed on graphene. Nature Materials 6, 652, (2007).

20 Elias, D. C. et al. Control of Graphene's Properties by Reversible Hydrogenation: Evidence for Graphane. Science 323, 610-613, (2009).

21 Balandin, A. A. et al. Superior Thermal Conductivity of Single-Layer Graphene. Nano Letters 8, 902-907, (2008).

22 Bolotin, K. I. et al. Ultrahigh electron mobility in suspended graphene. Solid State Communications 146, 351-355, (2008).

23 Nair, R. R. et al. Fine Structure Constant Defines Visual Transparency of Graphene. Science 320, 1308-1308, (2008).

24 Pisana, S. et al. Breakdown of the adiabatic Born–Oppenheimer approximation in graphene. Nature Materials 6, 198, (2007).

25 Geim, A. K. Graphene: Status and Prospects. Science 324, 1530-1534, (2009).

26 Allen, M. J., Tung, V. C. & Kaner, R. B. Honeycomb Carbon: A Review of Graphene. Chemical Reviews 110, 132-145, (2010).

27 Eda, G. & Chhowalla, M. Chemically Derived Graphene Oxide: Towards Large-Area Thin-Film Electronics and Optoelectronics. Advanced Materials 22, 2392-2415, (2010).

28 Yazyev, O. V. & Chen, Y. P. Polycrystalline graphene and other two-dimensional materials. Nat Nano 9, 755-767, (2014).

29 Li, X. et al. Large-Area Synthesis of High-Quality and Uniform Graphene Films on Copper Foils. Science 324, 1312-1314, (2009).

30 Gao, L., Guest, J. R. & Guisinger, N. P. Epitaxial graphene on Cu(111). Nano Lett 10, 3512-3516, (2010).

31 Berger, C. et al. Electronic Confinement and Coherence in Patterned Epitaxial Graphene. Science 312, 1191-1196, (2006).

32 Ohta, T., Bostwick, A., Seyller, T., Horn, K. & Rotenberg, E. Controlling the Electronic Structure of Bilayer Graphene. Science 313, 951-954, (2006).

33 Sutter, P. W., Flege, J.-I. & Sutter, E. A. Epitaxial graphene on ruthenium. Nature Materials 7, 406, (2008).

34 Li, D., Muller, M. B., Gilje, S., Kaner, R. B. & Wallace, G. G. Processable aqueous dispersions of graphene nanosheets. Nat Nanotechnol 3, 101-105, (2008).



35 Becerril, H. A. et al. Evaluation of Solution-Processed Reduced Graphene Oxide Films as Transparent Conductors. ACS Nano 2, 463-470, (2008).

36 Dikin, D. A. et al. Preparation and characterization of graphene oxide paper. Nature 448, 457, (2007).

37 Zhang, Y.-L. et al. Photoreduction of Graphene Oxides: Methods, Properties, and Applications. Advanced Optical Materials 2, 10-28, (2014).

38 Eda, G., Fanchini, G. & Chhowalla, M. Large-area ultrathin films of reduced graphene oxide as a transparent and flexible electronic material. Nature Nanotechnology 3, 270, (2008).

39 Huang, X. et al. Graphene Oxide-Templated Synthesis of Ultrathin or Tadpole-Shaped Au Nanowires with Alternating hcp and fcc Domains. Advanced Materials 24, 979-983, (2012).

40 Dong, L., Yang, J., Chhowalla, M. & Loh, K. P. Synthesis and reduction of large sized graphene oxide sheets. Chemical Society Reviews, (2017).

41 B. C. Brodie, F. R. S. On the atomic weight of graphite. Phil. Trans. R. Soc. Lond 149, 249-259 (1859).

42 Hummers, W. S. & Offeman, R. E. Preparation of Graphitic Oxide. Journal of the American Chemical Society 80, 1339-1339, (1958).

43 He, H., Klinowski, J., Forster, M. & Lerf, A. A new structural model for graphite oxide. Chemical Physics Letters 287, 53-56, (1998).

44 Shin, H.-J. et al. Efficient Reduction of Graphite Oxide by Sodium Borohydride and Its Effect on Electrical Conductance. Advanced Functional Materials 19, 1987-1992, (2009).

45 Gómez-Navarro, C. et al. Atomic Structure of Reduced Graphene Oxide. Nano Letters 10, 1144-1148, (2010).

46 Paredes, J. I., Villar-Rodil, S., Martínez-Alonso, A. & Tascón, J. M. D. Graphene Oxide Dispersions in Organic Solvents. Langmuir 24, 10560-10564, (2008).

47 Compton Owen, C. & Nguyen SonBinh, T. Graphene Oxide, Highly Reduced Graphene Oxide, and Graphene: Versatile Building Blocks for Carbon-Based Materials. Small 6, 711-723, (2010).

48 Kim, F., Cote Laura, J. & Huang, J. Graphene Oxide: Surface Activity and Two-Dimensional Assembly. Advanced Materials 22, 1954-1958, (2010).

49 Kelzenberg, M. D. et al. Enhanced absorption and carrier collection in Si wire arrays for photovoltaic applications. Nature Materials 9, 239-244, (2010).



50 Yang, J. et al. Design and fabrication of broadband ultralow reflectivity black Si surfaces by laser micro/nanoprocessing. Light: Science & Applications 3, e185, (2014).

51 Zhang, B., Jie, J., Zhang, X., Ou, X. & Zhang, X. Large-Scale Fabrication of Silicon Nanowires for Solar Energy Applications. ACS Appl Mater Interfaces 9, 34527-34543, (2017).

52 Hu, Y. et al. Metal-catalyzed electroless etching of silicon in aerated HF/H2O vapor for facile fabrication of silicon nanostructures. Nano Lett 14, 4212-4219, (2014).

53 Zhang, X. et al. High-efficiency graphene/Si nanoarray Schottky junction solar cells via surface modification and graphene doping. Journal of Materials Chemistry A 1, 6593, (2013).

54 Gowda, S. R. et al. Conformal coating of thin polymer electrolyte layer on nanostructured electrode materials for three-dimensional battery applications. Nano Lett 11, 101-106, (2011).

55 Liu, H., Li, W., Shen, D., Zhao, D. & Wang, G. Graphitic Carbon Conformal Coating of Mesoporous TiO2 Hollow Spheres for High-Performance Lithium Ion Battery Anodes. J Am Chem Soc 137, 13161-13166, (2015).

56 Fang, Z. et al. Graphene-antenna sandwich photodetector. Nano Lett 12, 3808-3813, (2012).

57 Fan, G. et al. Graphene/silicon nanowire Schottky junction for enhanced light harvesting. ACS Appl Mater Interfaces 3, 721-725, (2011).

58 GuT et al. Regenerative oscillation and four-wave mixing in graphene optoelectronics. Nat Photon 6, 554-559, (2012).

59 Ji, M. et al. Enhanced parametric frequency conversion in a compact silicon-graphene microring resonator. Opt. Express 23, 18679-18685, (2015).

60 Gao, W., Alemany, L. B., Ci, L. & Ajayan, P. M. New insights into the structure and reduction of graphite oxide. Nature Chemistry 1, 403, (2009).

61 Stankovich, S. et al. Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite oxide. Carbon 45, 1558-1565, (2007).

62 Williams, G., Seger, B. & Kamat, P. V. TiO2-Graphene Nanocomposites. UV-Assisted Photocatalytic Reduction of Graphene Oxide. ACS Nano 2, 1487-1491, (2008).

63 Cheng, M. et al. Restoration of graphene from graphene oxide by defect repair. Carbon 50, 2581-2587, (2012).



64 Eigler, S., Dotzer, C. & Hirsch, A. Visualization of defect densities in reduced graphene oxide. Carbon 50, 3666-3673, (2012).

65 Novoselov, K. S. et al. A roadmap for graphene. Nature 490, 192-200, (2012).

66 Shuping, P. et al. Extrinsic Corrugation‐Assisted Mechanical Exfoliation of Monolayer Graphene. Advanced Materials 22, 5374-5377, (2010).

67 Chang, Y. M., Kim, H., Lee, J. H. & Song, Y.-W. Multilayered graphene efficiently formed by mechanical exfoliation for nonlinear saturable absorbers in fiber mode-locked lasers. Applied Physics Letters 97, (2010).

68 Emtsev, K. V. et al. Towards wafer-size graphene layers by atmospheric pressure graphitization of silicon carbide. Nature Materials 8, 203, (2009).

69 Bae, S. et al. Roll-to-roll production of 30-inch graphene films for transparent electrodes. Nat Nanotechnol 5, 574-578, (2010).

70 Robinson, J. T., Perkins, F. K., Snow, E. S., Wei, Z. & Sheehan, P. E. Reduced Graphene Oxide Molecular Sensors. Nano Letters 8, 3137-3140, (2008).

71 Kris, E. et al. Determination of the Local Chemical Structure of Graphene Oxide and Reduced Graphene Oxide. Advanced Materials 22, 4467-4472, (2010).

72 Wang, X., Zhi, L. & Müllen, K. Transparent, Conductive Graphene Electrodes for Dye-Sensitized Solar Cells. Nano Letters 8, 323-327, (2008).

73 Qiu, L., Liu, J. Z., Chang, S. L. Y., Wu, Y. & Li, D. Biomimetic superelastic graphene-based cellular monoliths. Nature Communications 3, 1241, (2012).

74 Zheng, X., Jia, B., Chen, X. & Gu, M. In situ third-order non-linear responses during laser reduction of graphene oxide thin films towards on-chip non-linear photonic devices. Adv Mater 26, 2699-2703, (2014).

75 Xu, X. et al. Observation of Third-order Nonlinearities in Graphene Oxide Film at Telecommunication Wavelengths. Scientific Reports 7, 9646, (2017).

76 Bruna, M. & Borini, S. Optical constants of graphene layers in the visible range. Applied Physics Letters 94, 031901, (2009).

77 Skulason, H. S., Gaskell, P. E. & Szkopek, T. Optical reflection and transmission properties of exfoliated graphite from a graphene monolayer to several hundred graphene layers. Nanotechnology 21, 295709, (2010).



78 Cote, L. J., Cruz-Silva, R. & Huang, J. Flash Reduction and Patterning of Graphite Oxide and Its Polymer Composite. Journal of the American Chemical Society 131, 11027-11032, (2009).

79 Matsumoto, Y. et al. Simple Photoreduction of Graphene Oxide Nanosheet under Mild Conditions. ACS Applied Materials & Interfaces 2, 3461-3466, (2010).

80 Zhang, Y.-L., Chen, Q.-D., Xia, H. & Sun, H.-B. Designable 3D nanofabrication by femtosecond laser direct writing. Nano Today 5, 435-448, (2010).

81 Smirnov, V. A. et al. Photoreduction of graphite oxide. High Energy Chemistry 45, 57-61, (2011).

82 Zhou, Y. et al. Microstructuring of graphene oxide nanosheets using direct laser writing. Adv Mater 22, 67-71, (2010).

83 Zhang, Y. et al. Direct imprinting of microcircuits on graphene oxides film by femtosecond laser reduction. Nano Today 5, 15-20, (2010).

84 Zheng, X. et al. Highly efficient and ultra-broadband graphene oxide ultrathin lenses with three-dimensional subwavelength focusing. Nat Commun 6, 8433, (2015).

85 Castro Neto, A. H., Guinea, F., Peres, N. M. R., Novoselov, K. S. & Geim, A. K. The electronic properties of graphene. Reviews of Modern Physics 81, 109-162, (2009).

86 Jung, I. et al. Simple Approach for High-Contrast Optical Imaging and Characterization of Graphene-Based Sheets. Nano Letters 7, 3569-3575, (2007).

87 Yang, H., Hu, H., Wang, Y. & Yu, T. Rapid and non-destructive identification of graphene oxide thickness using white light contrast spectroscopy. Carbon 52, 528-534, (2013).

88 Hong, J. T. et al. Terahertz conductivity of reduced graphene oxide films. Opt. Express 21, 7633-7640, (2013).

89 Jung, I. et al. Characterization of Thermally Reduced Graphene Oxide by Imaging Ellipsometry. The Journal of Physical Chemistry C 112, 8499-8506, (2008).

90 Shen, Y. et al. Optical investigation of reduced graphene oxide by spectroscopic ellipsometry and the band-gap tuning. Applied Physics Letters 99, 141911, (2011).

91 R., S. & B., T. Handbook of Nonlinear Optics. (CRC Press, 2003).

92 Loh, K. P., Bao, Q., Eda, G. & Chhowalla, M. Graphene oxide as a chemically tunable platform for optical applications. Nat Chem 2, 1015-1024 (2010).



93 Liu, Z. et al. Nonlinear optical properties of graphene oxide in nanosecond and picosecond regimes. Applied Physics Letters 94, 021902, (2009).

94 Jiang, X.-F., Polavarapu, L., Neo, S. T., Venkatesan, T. & Xu, Q.-H. Graphene Oxides as Tunable Broadband Nonlinear Optical Materials for Femtosecond Laser Pulses. The Journal of Physical Chemistry Letters 3, 785-790, (2012).

95 Liaros, N. et al. Nonlinear Optical Properties and Broadband Optical Power Limiting Action of Graphene Oxide Colloids. The Journal of Physical Chemistry C 117, 6842-6850, (2013).

96 Moss, D. J., Morandotti, R., Gaeta, A. L. & Lipson, M. New CMOS-compatible platforms based on silicon nitride and Hydex for nonlinear optics. Nature Photonics 7, 597, (2013).

97 Lu, Y. et al. Novel blue light emitting graphene oxide nanosheets fabricated by surface functionalization. Journal of Materials Chemistry 22, 2929-2934, (2012).

98 Lee, B. R. et al. Highly Efficient Polymer Light-Emitting Diodes Using Graphene Oxide as a Hole Transport Layer. ACS Nano 6, 2984-2991, (2012).

99 Wang, D.-Y. et al. Quantum Dot Light-Emitting Diode Using Solution-Processable Graphene Oxide as the Anode Interfacial Layer. The Journal of Physical Chemistry C 116, 10181-10185, (2012).

100 Bi, Y.-G. et al. Arbitrary Shape Designable Microscale Organic Light-Emitting Devices by Using Femtosecond Laser Reduced Graphene Oxide as a Patterned Electrode. ACS Photonics 1, 690-695, (2014).

101 Eggleton, B., Moss, D. & Radic, S. Optical Fiber Telecommunications V: Components and Subsystems. Chap 20, 759-828 (2008).

102 Salem, R. et al. Signal regeneration using low-power four-wave mixing on silicon chip. Nature Photonics 2, 35 (2008).

103 Fukuda, H. et al. Four-wave mixing in silicon wire waveguides. Opt. Express 13, 4629-4637, (2005).

104 Liu, Y. et al. Error-Free 320-Gb/s All-Optical Wavelength Conversion Using a Single Semiconductor Optical Amplifier. Journal of Lightwave Technology 25, 103-108 (2007).

105 Zhihong, L. & Guifang, L. Ultrahigh-speed reconfigurable logic gates based on four-wave mixing in a semiconductor optical amplifier. IEEE Photonics Technology Letters 18, 1341-1343, (2006).

106 Wang, J., Sun, Q. & Sun, J. All-optical 40 Gbit/s CSRZ-DPSK logic XOR gate and format conversion using four-wave mixing. Opt. Express 17, 12555-12563, (2009).



107 Del’Haye, P. et al. Optical frequency comb generation from a monolithic microresonator. Nature 450, 1214, (2007).

108 Pasquazi, A. et al. Micro-combs: A novel generation of optical sources. Physics Reports 729, 1-81, (2018).

109 Kues, M. et al. On-chip generation of high-dimensional entangled quantum states and their coherent control. Nature 546, 622, (2017).

110 Reimer, C. et al. Generation of multiphoton entangled quantum states by means of integrated frequency combs. Science 351, 1176-1180, (2016).

111 Foster, M. A. et al. Silicon-chip-based ultrafast optical oscilloscope. Nature 456, 81, (2008).

112 Pasquazi, A. et al. Sub-picosecond phase-sensitive optical pulse characterization on a chip. Nature Photonics 5, 618, (2011).

113 Inoue, K. & Toba, H. Wavelength conversion experiment using fiber four-wave mixing. IEEE Photonics Technology Letters 4, 69-72, (1992).

114 Zhang, A. & Demokan, M. S. Broadband wavelength converter based on four-wave mixing in a highly nonlinear photonic crystal fiber. Opt. Lett. 30, 2375-2377, (2005).

115 Ottavi, A. D. et al. Four-wave mixing in semiconductor optical amplifiers: a practical tool for wavelength conversion. IEEE Journal of Selected Topics in Quantum Electronics 3, 522-528, (1997).

116 Bilenca, A. et al. Broad-band wavelength conversion based on cross-gain modulation and four-wave mixing in InAs-InP quantum-dash semiconductor optical amplifiers operating at 1550 nm. IEEE Photonics Technology Letters 15, 563-565 (2003).

117 Levy, J. S. et al. CMOS-compatible multiple-wavelength oscillator for on-chip optical interconnects. Nature Photonics 4, 37, (2009).

118 Hochberg, M. et al. Terahertz all-optical modulation in a silicon–polymer hybrid system. Nature Materials 5, 703, (2006).

119 Koos, C. et al. All-optical high-speed signal processing with silicon–organic hybrid slot waveguides. Nature Photonics 3, 216, (2009).

120 Eggleton, B. J., Luther-Davies, B. & Richardson, K. Chalcogenide photonics. Nature Photonics 5, 141, (2011).

121 Alexander, K., Savostianova, N. A., Mikhailov, S. A., Kuyken, B. & Van Thourhout, D. Electrically Tunable Optical Nonlinearities in Graphene-Covered SiN Waveguides Characterized by Four-Wave Mixing. ACS Photonics 4, 3039-3044, (2017).



122 Bao, Q. et al. Atomic‐Layer Graphene as a Saturable Absorber for Ultrafast Pulsed Lasers. Advanced Functional Materials 19, 3077-3083, (2009).

123 Sun, Z. et al. Graphene Mode-Locked Ultrafast Laser. ACS Nano 4, 803-810, (2010).

124 Bonaccorso, F., Sun, Z., Hasan, T. & Ferrari, A. C. Graphene photonics and optoelectronics. Nat Photon 4, 611-622 (2010).

125 Fiori, G. et al. Electronics based on two-dimensional materials. Nat Nano 9, 768-779, (2014).

126 Kong, X.-T. et al. Graphene-Based Ultrathin Flat Lenses. ACS Photonics 2, 200-207, (2015).

127 Butler, S. Z. et al. Progress, Challenges, and Opportunities in Two-Dimensional Materials Beyond Graphene. ACS Nano 7, 2898-2926, (2013).

128 Lin, H. et al. Chalcogenide glass-on-graphene photonics. Nature Photonics 11, 798-805, (2017).

129 Ruoff, R. Graphene: Calling all chemists. Nat Nano 3, 10-11 (2008).

130 Dreyer, D. R., Park, S., Bielawski, C. W. & Ruoff, R. S. The chemistry of graphene oxide. Chemical Society Reviews 39, 228-240, (2010).

131 Kotov, N. A., Dékány, I. & Fendler, J. H. Ultrathin graphite oxide-polyelectrolyte composites prepared by self-assembly: Transition between conductive and non-conductive states. Advanced Materials 8, 637-641, (1996).

132 Han, D.-D. et al. Moisture-Responsive Graphene Paper Prepared by Self-Controlled Photoreduction. Advanced Materials 27, 332-338, (2015).

133 Zheng, X., Jia, B., Chen, X. & Gu, M. In Situ Third‐Order Non‐linear Responses During Laser Reduction of Graphene Oxide Thin Films Towards On‐Chip Non‐linear Photonic Devices. Advanced Materials 26, 2699-2703, (2014).

134 Bonaccorso, F. et al. Graphene, related two-dimensional crystals, and hybrid systems for energy conversion and storage. Science 347, 1246501 (2015).

135 Koppens, F. et al. Photodetectors based on graphene, other two-dimensional materials and hybrid systems. Nature Nanotechnology 9, 780-793 (2014).

136 Bernardi, M., Palummo, M. & Grossman, J. C. Extraordinary sunlight absorption and one nanometer thick photovoltaics using two-dimensional monolayer materials. Nano letters 13, 3664-3670 (2013).



137 Nair, R. R. et al. Fine structure constant defines visual transparency of graphene. Science 320, 1308-1308 (2008).

138 Xiang, Y. et al. Critical coupling with graphene-based hyperbolic metamaterials. Scientific reports 4, 5483 (2014).

139 Sreekanth, K. V. et al. A multiband perfect absorber based on hyperbolic metamaterials. Scientific reports 6 (2016).

140 Li, W. & Valentine, J. Metamaterial perfect absorber based hot electron photodetection. Nano letters 14, 3510-3514 (2014).

141 Ferrari, L., Smalley, J. S. T., Fainman, Y. & Liu, Z. Hyperbolic metamaterials for dispersion-assisted directional light emission. Nanoscale 9, 9034-9048 (2017).

142 Al Sayem, A., Mahdy, M. R. C., Jahangir, I. & Rahman, M. S. Ultrathin ultra-broadband electro-absorption modulator based on few-layer graphene based anisotropic metamaterial. Optics Communications 384, 50-58 (2017).

143 Gao, C. & Guo, B. Tunable the spin Hall effect of light with graphene metamaterial. Optik-International Journal for Light and Electron Optics (2017).

144 Chang, Y.-C. et al. Realization of mid-infrared graphene hyperbolic metamaterials. Nature communications 7, 10568 (2016).

145 Baitimirova, M. et al. Tuning of structural and optical properties of graphene/ZnO nanolaminates. The Journal of Physical Chemistry C 120, 23716-23725 (2016).

146 Zhang, B. Y. et al. Degenerately Hydrogen Doped Molybdenum Oxide Nanodisks for Ultrasensitive Plasmonic Biosensing. Advanced Functional Materials, 1706006-n/a, (2017).

147 Vázquez, R. M., Eaton, S. M., Ramponi, R., Cerullo, G. & Osellame, R. Fabrication of binary Fresnel lenses in PMMA by femtosecond laser surface ablation. Opt. Express 19, 11597-11604, (2011).

148 Fang, A., Koschny, T. & Soukoulis, C. M. Optical anisotropic metamaterials: Negative refraction and focusing. Physical Review B 79, 245127 (2009).

149 Eda, G., Mattevi, C., Yamaguchi, H., Kim, H. & Chhowalla, M. Insulator to semimetal transition in graphene oxide. The Journal of Physical Chemistry C 113, 15768-15771 (2009).

150 Leuthold, J., Koos, C. & Freude, W. Nonlinear silicon photonics. Nature Photonics 4, 535, (2010).

151 Cundiff, S. T. & Weiner, A. M. Optical arbitrary waveform generation. Nature Photonics 4, 760 (2010).



152 Luo, Z., Vora, P. M., Mele, E. J., Johnson, A. C. & Kikkawa, J. M. Photoluminescence and band gap modulation in graphene oxide. Applied physics letters 94, 111909 (2009).

153 Guo, L. et al. Bandgap Tailoring and Synchronous Microdevices Patterning of Graphene Oxides. The Journal of Physical Chemistry C 116, 3594-3599, (2012).

154 Lin, Q., Zhang, J., Fauchet, P. M. & Agrawal, G. P. Ultrabroadband parametric generation and wavelength conversion in silicon waveguides. Opt. Express 14, 4786-4799, (2006).

155 Liu, Q., Gao, S., Li, Z., Xie, Y. & He, S. Dispersion engineering of a silicon-nanocrystal-based slot waveguide for broadband wavelength conversion. Appl. Opt. 50, 1260-1265, (2011).

156 Moss, D., van Driel, H. M. & Sipe, J. E. Dispersion in the anisotropy of optical third-harmonic generation in silicon. Opt. Lett. 14, 57-59 (1989).

157 Moss, D., Ghahramani, E., Sipe, J. & Van Driel, H. Band-structure calculation of dispersion and anisotropy in χ→(3) for third-harmonic generation in Si, Ge, and GaAs. Physical Review B 41, 1542 (1990).

158 Moss, D., Van Driel, H. & Sipe, J. Third harmonic generation as a structural diagnostic of ion‐implanted amorphous and crystalline silicon. Applied physics letters 48, 1150-1152 (1986).

159 Xu, X. et al. Reconfigurable broadband microwave photonic intensity differentiator based on an integrated optical frequency comb source. Apl Photonics 2, 096104 (2017).

160 Yang, Y., Lin, H., Jia, B., Zhang, y. & Hong, M. in Frontiers in Optics 2017. FM3D.6 (Optical Society of America).

161 Li, H., Anugrah, Y., Koester, S. J. & Li, M. Optical absorption in graphene integrated on silicon waveguides. Applied Physics Letters 101, 111110, (2012).



Publications

Journal articles:

1. Yunyi Yang, Han Lin, Bao Yue Zhang, Yinan Zhang, Xiaorui Zheng, Aimin Yu,

Minghui Hong and Baohua Jia. In-situ tunable graphene-based metamaterial.

Submitted to Nature Materials.

2. Yunyi Yang, Jiayang Wu, Xingyuan Xu, Yao Liang, Sai T. Chu, Brent E. Little,

Roberto Morandotti, Baohua Jia and David J. Moss. Enhanced four-wave mixing

in waveguides integrated with graphene oxide. APL Photonics, revision submitted.

3. Yunyi Yang, Yinan Zhang, Jie Zhang, Han Lin, Minghui Hong and Baohua Jia.

Three-dimensional graphene oxide conformal coating on silicon nanostructures.

Ready to submit.

Conference papers:

1. Yunyi Yang et al. “Enhanced four-wave mixing in graphene oxide coated

waveguides”. CLEO, San Jose, May. 2018, Oral.

2. Yunyi Yang et al. “Four-wave mixing efficiency enhancement in silicon-

graphene oxide hybrid waveguides”. SPIE Nanophotonics, Melbourne, Dec. 2017,

Oral.

3. Yunyi Yang et al. “Broadband graphene oxide anti-reflection coating on silicon

nanostructures”. FiO, Washington D.C., Sep. 2017, Oral.

4. Yunyi Yang et al. “Enhanced four-wave mixing efficiency in silicon-graphene

oxide hybrid waveguide”. IPR, New Orleans, Jul. 2017, Oral.

5. Yunyi Yang et al. “Analysis of enhanced four-wave mixing in integrated silicon-

graphene oxide hybrid waveguides”. CLEO-PR, Aug. 2017, Oral.

6. Baohua Jia, Xiaorui Zheng, Han Lin, Yunyi Yang, Scott Fraser. “Graphene

oxide thin films for functional photonic devices”. FiO, Rochester, Oct. 2016, Oral.

javascript:void(0)

javascript:void(0)



7. Yunyi Yang, Han Lin, Baohua Jia. “Self-assembled graphene oxide film for 3D

coating on nanostructures”. IONS-KOALA, Melbourne, Nov. 2016, Oral.

8. Yunyi Yang, Han Lin, Baohua Jia. “Optical characters of self-assembled

graphene oxide film”. IONS-KOALA, Adelaide, Nov. 2014, Oral.

Documents

3D multilayer graphene oxide thin film platform for