By Muchun Liu B.Sc., Materials Science and Engineering

2D Nanochannels in Textured Graphene Films – Intercalated

Templating, Nanofluidic Transport and Controlled Release

By Muchun Liu

B.Sc., Materials Science and Engineering, Beihang University, 2012

M.E., Materials Engineering, Beihang University, 2015

A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

IN THE DEPARTMENT OF CHEMISTRY AT BROWN UNIVERSITY

Providence, Rhode Island

May 2020

© Copyright 2020 by Muchun Liu

III

This dissertation by Muchun Liu is accepted in its present form

by the Department of Chemistry as satisfying the

dissertation requirement for the degree of Doctor of Philosophy.

Date ______________ ________________________

Robert H. Hurt, Advisor

Recommended to the Graduate Council by

Date ______________ ________________________

Vicki L. Colvin, Reader

Date ______________ ________________________

Shouheng Sun, Reader

Approved by the Graduate Council

Date ______________ ________________________

Andrew G. Campbell, Dean of the Graduate School

IV

Vitae

Muchun Liu was born in 1990, China. She received a Bachelor of Science degree in

Materials Science and Engineering in 2012 and a Master of Engineering degree in

Materials Engineering in 2015. At Beihang University, she published 4 papers in peer-

reviewed journals and 3 patents. In the fall of 2015, she went on to the Department of

Chemistry at Brown University to continue her doctoral education. Under the tutelage of

Prof. Robert H. Hurt, she worked on assembly and applications of two-dimensional

nanochannels and published 9 papers in peer-reviewed journals.

V

Publications

1. M Liu, PJ Weston, RH Hurt. Controlling nanochannel orientation, length, and

width in graphene-based nanofluidic membranes. Submitted.

2. M Liu, RH Hurt. Controlled release from intercalated graphene oxide films: edge-

and basal-plane-specific kinetics. In preparation.

3. M Liu, L Qian, C Yu, G Xiao, RH Hurt. Stretching, bending and magnetic

properties of cobalt ferrite foldable films. In preparation.

4. EP Gray, CL Browning, CA Vaslet, KD Gion, A Green, M Liu, AB Kane, RH Hurt.

Chemical and colloidal dynamics of MnO2 nanosheets in biological media relevant

for nanosafety assessment. Small, 2020, 2000303.

5. CJ Castilho, D Li, M Liu, Y Liu, H Gao, RH Hurt. Mosquito bite prevention

through graphene barrier layers. Proc. Natl. Acad. Sci., 2019, 116, 18304-18309.

6. TM Valentin, AK Landauer, LC Morales, EM DuBois, S Shukla, M Liu, et al.

Alginate-graphene oxide hydrogels with enhanced ionic tunability and

chemomechanical stability for light-directed 3D printing. Carbon, 2019, 143, 447-

456.

7. M Liu, PY Chen, RH Hurt. Graphene inks as versatile templates for printing tiled

metal oxide crystalline films. Adv. Mater., 2018, 30, 1705080.

8. M Liu, CJ Castilho, RH Hurt. New material architectures through graphene

nanosheet assembly. Adv. Mater. Lett., 2018, 9, 843-850.

9. PY Chen, M Zhang, M Liu, IY Wong, RH Hurt. Ultrastretchable graphene-based

molecular barriers for chemical protection, detection, and actuation. ACS Nano,

2017, 12, 234-244.

VI

10. PY Chen, M Liu, Z Wang, RH Hurt, IY Wong. From flatland to spaceland: higher

dimensional patterning with two-dimensional materials. Adv. Mater., 2017, 29,

1605096.

11. Z Wang, YJ Zhang, M Liu, A Peterson, RH Hurt. Oxidation suppression during

hydrothermal phase reversion allows synthesis of monolayer semiconducting MoS2

in stable aqueous suspension. Nanoscale, 2017, 9, 5398-5403.

12. P Chen, M Liu, TM Valentin, Z Wang, RS Steinberg, J Sodhi, IY Wong, RH Hurt.

Hierarchical metal oxide topographies replicated from highly textured graphene

oxide by intercalation templating. ACS Nano, 2016, 10, 10869-10879.

13. M Liu, Y Zhao, S Gao, Y Wang, Y Duan, X Han, Q Dong. Mild solution synthesis

of graphene loaded with LiFePO4-C nanoplatelets for high performance lithium ion

batteries. New J. Chem., 2015, 39, 1094-1100.

14. M Liu, Y Duan, Y Wang, Y Zhao. Diazonium functionalization of graphene

nanosheets and impact response of aniline modified graphene/bismaleimide

nanocomposites. Mater. Des., 2014, 53, 466-474.

15. Y Wang, Y Zhao, J Yin, M Liu, Q Dong, Y Su. Synthesis and electrocatalytic

alcohol oxidation performance of Pd-Co bimetallic nanoparticles supported on

graphene. Int. J. Hydrog. Energy, 2014, 39, 1325-1335.

16. Q Dong, Y Zhao, X Han, Y Wang, M Liu, Y Li. Pd/Cu bimetallic nanoparticles

supported on graphene nanosheets: Facile synthesis and application as novel

electrocatalyst for ethanol oxidation in alkaline media. Int. J. Hydrog. Energy, 2014,

39, 14669-14679.

VII

Acknowledgements

First I would like to thank my advisor Prof. Robert Hurt. His patience, optimism and

generosity support me throughout my Ph.D. study. He encourages me to find the right

questions and incubate my own ideas. He actively presents me the juggling life of a modern

PI, where I learned to stretch my abilities in many aspects. He teaches me not only to

bounce back from failure but get used to it. I stay perfectly sane in the past five years thanks

to his calmness and subtle sense of humor. The only thing I would not take from him is to

choose to live without a cellphone in the 21st century.

I would also like to thank my committee members Prof. Vicki Colvin and Prof.

Shouheng Sun for their time and effort, and Prof. Sarah Delaney for her guidance and

kindness in GSLC. Like them, I met many great professors at Brown. They all are very

different people but shine in their own unique ways, every bit of it is helpful.

I would like to thank my collaborators and co-authors, especially Prof. Gang Xiao, Po-

yen Chen, Zhongying Wang, Evan Gray and Paula Weston. I would also like to thank my

labmates, present and former, Cintia Castilho, Mengke Zhang, Zachary Saleeba, Jonathan

Ström, Vidushi Shukla, Aidan Stone and everyone from the research group of Prof. Agnes

Kane, whose friendship and mentoring was essential for my research endeavors. I would

like to thank Dr. Indrek Külaots, who makes me feel like family, I wish nothing but the

best for him.

I would also like to thank my friends, near and far, especially Yue Hu, Yucheng Yuan,

Dr. Di Xia, Prof. Lei Zhou and Dr. Yan Wang. I love how we are always brutally honest

to each other and becoming stronger together.

VIII

Many thanks also go to my father, the first Ph.D. I know in my life, who did not get to

mention me in his dissertation acknowledgements; my mother, who is the sweetest person

and my best friend; my in-laws for their love and trust; and my husband – who sat through

my endless practice talks at many midnights – for his love, kindness and strong will.

Lastly, I would like to thank my difficult gradation time in current global pandemic,

which helps me understand how much I love my job.

IX

Abstract of “2D nanochannels in textured graphene films – intercalated templating,

nanofluidic transport and controlled release” by Muchun Liu, Ph. D., Brown University,

May 2020

There is great interest in exploiting van der Waals gaps in layered materials. Two-

dimensional (2D) nanosheets of graphene oxide (GO) films are of great potential due to

tunable interlayer spacing and high colloidal stability. Ionic or molecular substances with

physical compatibility are capable to be encapsulated, diffused or released within the

interlayer galleries. Therefore, the GO nanochannels can be viewed as nanofluidic channels

or confinement reaction vessels to template the synthesis of new nanosheet structures.

Moreover, the texturing technique enables 2D nanochannels with various spatial

topographies, which provides a new perspective to related applications. In this dissertation,

the intercalated templating, nanofluidic transport and controlled release of 2D

nanochannels in textured GO films are discussed.

The gallery spaces in multilayer GO can intercalate hydrated metal ions that assemble

into metal oxide films during thermal oxidation of the sacrificial graphene template. This

approach offers limited control of structure, however, and does not typically lead to 2D

atomic-scale growth of anisotropic platelet crystals, but rather arrays of simple particles

directionally sintered into porous sheets. In Chapter 2 we demonstrate a new graphene-

directed assembly route that yields fully-dense, space-filling films of tiled metal oxide

platelet crystals with tessellated structures. The method relies on colloidal engineering to

produce a printable “metallized graphene ink” with accurate control in metal loading, grain

size/porosity, composition and micro/nano-morphologies, and is capable to achieve higher

X

metal-carbon ratio than is achievable by intercalation methods. These tiled structures are

sufficiently robust to create free standing papers, complex microtextured films, 3D shapes,

and metal oxide replicas of natural biotextures.

A follow up research interest is in developing flexible devices for applications such as

electric displays, human-machine interfaces and biomedical devices. In fabrication,

ceramic components are particularly challenging due to extreme stiffness. In Chapter 3,

free-standing cobalt ferrite wrinkled films are obtained after removal of sacrificial GO

templates. The stretching, bending and magnetic properties of cobalt ferrite foldable films

are studied.

GO nanosheets are also known to spontaneously assemble into stacked planar

membranes with transport properties that are highly selective to molecular structure. Use

of conventional GO membranes in liquid-phase applications is often limited by low flux

values, due to intersheet nanochannel alignment perpendicular to the desired Z-directional

transport, which leads to circuitous fluid pathways that are orders of magnitude longer than

the membrane thickness. In Chapter 4 we demonstrate a new approach that uses

compressive instability in Zr-doped GO thin films to create wrinkle patterns that rotate

nanosheets to high angles. Capturing this structure in polymer matrices and thin sectioning

produces fully dense membranes with arrays of vertically aligned nanochannels. These

robust nanofluidic devices offer dramatic reduction in fluid path-length, while retaining the

high selectivity for water over non-polar molecules characteristic of GO interlayer

nanochannels.

The compressive wrinkling and crumpling of GO films can also be used to control the

release rates of molecular intercalants pre-loaded into GO gallery spaces. In Chapter 5,

XI

experimental studies on rhodamine B dye, used as a model, show diffusive release rates in

topography-related order. This type of fluidic-space manipulation should allow the

intelligent design of 2D-material-based technologies such as time-release drug eluting

coatings.

XII

Table of Contents

Vitae ……. ....................................................................................................................... IV

Publications ...................................................................................................................... V

Acknowledgements ....................................................................................................... VII

Abstract … ....................................................................................................................... IX

Chapter 1 Introduction of Two-Dimensional Materials and Nanochannels ............... 1

1.1 Introduction and historical perspective ................................................................. 1

1.2 Graphene oxide and 2D nanochannels .................................................................. 4

1.3 Assembly and applications of textured 2D nanochannels ..................................... 6

1.4 References ............................................................................................................. 9

Chapter 2 Textured 2D Nanochannels as Versatile Templates for Templating Tiled

Metal Oxide Crystalline Films ....................................................................................... 16

2.1 Introduction ......................................................................................................... 16

2.2 Results and discussion ......................................................................................... 18

2.3 Conclusions ......................................................................................................... 27

2.4 Materials and methods ........................................................................................ 28

2.5 Acknowledgements ............................................................................................. 32

2.6 References ........................................................................................................... 33

Chapter 3 Stretching, Bending and Magnetic Properties of Cobalt Ferrite Foldable

Films …… ........................................................................................................................ 38

3.1 Introduction ......................................................................................................... 38


3.3 Conclusions ......................................................................................................... 45

XIII


3.5 References ........................................................................................................... 48

Chapter 4 Controlling Nanochannel Orientation, Length, and Width in Graphene-

Based Nanofluidic Membranes ...................................................................................... 51

4.1 Introduction ......................................................................................................... 51


4.3 Conclusions ......................................................................................................... 61


4.5 References ........................................................................................................... 65

Chapter 5 Controlled Release from Intercalated Graphene Oxide Films: Edge- and

Basal-Plane-Specific Kinetics ......................................................................................... 70

5.1 Introduction ......................................................................................................... 70


5.3 Next steps ............................................................................................................ 78


5.5 References ........................................................................................................... 82

Chapter 6 Appendices..................................................................................................... 87

Appendix to Chapter 2 .................................................................................................. 87

Appendix to Chapter 3 ................................................................................................ 101

Appendix to Chapter 4 ................................................................................................ 105

XIV

List of Figures

Figure 1. 1. A brief history of ultrathin carbon materials. .................................................. 3

Figure 1. 2. Morphologies and structures of GO. ............................................................... 6

Figure 1. 3. Graphene nanosheet assemblies and their paper/fabric analogs. ..................... 7

Figure 2. 1. Textured metal oxide films from MGI. ......................................................... 20

Figure 2. 2. Effect of metal-carbon ratio on the micro- and nanostructures of Fe oxide

textured films fabricated from MGI. ................................................................................. 23

Figure 2. 3. Example applications of MGI in biotexture replication, paper-based 3D shape

creation, and printing. ....................................................................................................... 26

Figure 2. 4. Overview of assembly mechanisms and material structures fabricated from

MGI. .................................................................................................................................. 28

Figure 3. 1. Morphologies of CoFeFFs. ............................................................................ 41

Figure 3. 2. Stretch behavior and magnetic properties of CoFeFFs. ................................ 42

Figure 3. 3. Bend behavior and magnetic properties of CoFeFF. ..................................... 43

Figure 3. 4. Temperature dependence, anisotropy and mechanical fatigue on magnetic

behaviors of CoFeFF......................................................................................................... 44

Figure 4. 1. Schematic and fabrication of vertically aligned Zr-GO/epoxy membranes. . 54

Figure 4. 2. Morphologies of wrinkled Zr-GO films and VAGME during fabrication. ... 57

Figure 4. 3. Measurements of selective molecular transport through VAGME

nanochannels. .................................................................................................................... 59

XV

Figure 5. 1. Schematic and release behaviors of RhB and RhB/GO films in PBS solution.

........................................................................................................................................... 73

Figure 5. 2. Schematic of release pathways and surface morphologies of RhB intercalated

GO films............................................................................................................................ 74

Figure 5. 3. Release behaviors of RhB/GO textured films through different release

pathways. .......................................................................................................................... 76

Figure 5. 4. Calibration curve of concentration to absorption of RhB. ............................ 76

Figure 5. 5. Release behavior of RhB sample. .................................................................. 77

Figure 5. 6. Photos of edge-specific release of RhB/GO textured films in 28 days. ........ 77

Figure 5. 7. Dynamic XRD results of GO textured films swelling in PBS solutions. ...... 78

Figure 6. 1. Schematic of the fabrication process to generate textured GO. .................... 89

Figure 6. 2. FT-IR results of GO and Fe(III)-GO films. ................................................... 89

Figure 6. 3. Morphologies of GO and Fe(III)-GO nanosheets. ........................................ 90

Figure 6. 4. Experimental -potential of Fe(III)-Co(II) based MGI as a function of

([Fe(III)-Co(II)])/C ratio. .................................................................................................. 90

Figure 6. 5. Surface morphologies and crystal structures of metal oxide textured films

fabricated by various MGIs at colloidally stable loading. ................................................ 91

Figure 6. 6. Porosity determination for textured metal oxide films from SEM micrographs

by image analysis (ImageJ). ............................................................................................. 92

Figure 6. 7. Effect of metal-carbon ratio on surface morphologies of Fe oxide textured

films from Fe(III)-based MGIs, and crystal structures and nanostructures of tessellated Fe

oxide films (initial atomic Fe/C ~ 3/1). ............................................................................ 92

XVI

Figure 6. 8. Side views of Fe-based MGI deposits before annealing and Fe2O3 textured

films. ................................................................................................................................. 93

Figure 6. 9. Nanostructures of Fe2O3 films fabricated from simple casting of Fe(III) salts

on the external surfaces of reduced GO or HOPG. ........................................................... 94

Figure 6. 10. TGA curves of Fe-based MGI with metal-carbon ratio of 1/3, GO and

Fe(NO3)3 salts in air. ......................................................................................................... 94

Figure 6. 11. Effect of metal-carbon ratio on surface morphologies of Fe-Co oxide

textured films from Fe(III)/Co(II)-based MGIs, and morphologies of GO and Fe-Co

oxide textured films. ......................................................................................................... 95

Figure 6. 12. Nanostructure and crystal structure of CoFe2O4/Co2FeO4 textured film. ... 96

Figure 6. 13. Detailed example applications of MGI in biotexture replication, 3D shape

creation and printability. ................................................................................................... 97

Figure 6. 14. Fe2O3 textured structures fabricated in the absence of GO, or in the presence

of anionic polymer chains (poly(acrylic acid)) exhibit uncontrolled particle growth and

powdery coatings that lack mechanical integrity to form free-standing films. ................. 98

Figure 6. 15. Stretching behaviors of PDMS-fixed textured Fe2O3 film. ......................... 99

Figure 6. 16. Detailed fabrication process of CoFeFFs. ................................................. 101

Figure 6. 17. XRD spectrum of CoFeFF......................................................................... 102

Figure 6. 18. TGA curves of GO-Fe(III)/Co(II) films, GO and Fe(NO3)3/Co(NO3)2 salts in

air. ................................................................................................................................... 102

Figure 6. 19. Surface morphologies of GO wrinkled film and CoFeFF. ........................ 103

Figure 6. 20. Cross-section of CoFeFF. .......................................................................... 103

Figure 6. 21. SAED of CoFe2O4 single nanoplatelet. ..................................................... 103

XVII

Figure 6. 22. Morphologies of GO nanosheets. .............................................................. 104

Figure 6. 23. Morphologies of wrinkled films. ............................................................... 106

Figure 6. 24. Experimental ζ-potential of GO nanosheet suspensions with varying degrees

of ZrOCl2 addition, expressed as a function of [Zr]/C atomic ratio. .............................. 107

Figure 6. 25. Structural failures of VAGME appeared in developing microtome

sectioning technique........................................................................................................ 108

Figure 6. 26. Time-dependent behavior and properties of VAGMEs during exposure to

water vapor at 100 C. .................................................................................................... 109

Figure 6. 27. C1s XPS spectra for wrinkled Zr-GO films to 100 °C water vapor for

various times, in hrs ........................................................................................................ 110

XVIII

List of Tables

Table 4. 1 Water vapor fluxes measured through VAGME devices and conventional GO

films .................................................................................................................................. 61

1

Chapter 1 Introduction of Two-Dimensional Materials and Nanochannels

The content of this chapter contains modified contexts of published paper “Muchun Liu,

Cintia Juliana Castilho, Robert H. Hurt. Adv. Mater. Lett., 2018, 9, 843-850” and has been

reproduced here with permission. Copyright © 2018 IAAM-VBRI Press.

1.1 Introduction and historical perspective

In the past decade, two-dimensional (2D) materials are gaining considerable popularity

along with the rise of graphene. Distinctive from 0D and 1D nanomaterials, 2D materials

exhibit atomic thin and flexible structures with physiochemical functionalities. Modern

graphene research began in 2004 with the isolation and characterization of the monolayer

form.1 Much of graphene research has focused on understanding fundamental electronic

properties, developing improved synthesis methods, or exploring the technological

applications of this unique single-atom-thick sheet. An emerging subfield within graphene

research, however, does not view the monolayer nanosheet as an end product, but rather as

a precursor for new materials synthesis.2-4

The isolated graphene monolayer is a new development, but the materials derived from

graphene assembly are ultimately carbon materials - one of the oldest material classes.

Figure 1.1a shows charcoal sketches from the Chauvet Cave in southern France, where

analysis of the black deposits suggests an age of approx. 30,000 years.5-6 As 21st century

carbon scientists looking at the drawings, one is struck by three impressions. The first is

the beautiful artistry of these ancient Europeans, and the effects created by charcoal traces

continue to make this medium attractive to artists today. Secondly, if the charcoal was

2

taken from campfires or sites of human-initiated forest fires,5 this can reasonably be

regarded as an early use of synthetic carbon as a functional material - a dry-application

pigment. Finally, this artistic achievement has something in common with the work that

led to the 2010 Nobel Prize in Physics.1, 7 The ancient Europeans used a type of simple

mechanical exfoliation to achieve thin films of optically absorbing sp2-based carbon.

Today we might call such deposits “multilayer graphene”, especially in the case of pencil

traces, which derive from graphite with its very well-developed graphene layer structure.

A theme of this chapter is that all sp2-based (non-diamond-like) carbons consist of

graphene layers, even such common materials as charcoal produced by primitive methods

such as heating wood through accidentally incomplete combustion.

Modern attempts to exfoliate graphite into thin, flexible forms began well before 2004.

Starting in the mid-20th-century, graphite was exfoliated by formation of intercalation

compounds (typically graphite bisulfate formed by graphite treatment with concentrated

sulfuric acid and an oxidizing agent (e.g. H2O2, Br2, AsF5 or FeCl3) followed by thermal

shock to expel the intercalant and physically separate the layers.8-9 This rapid process leads

to massive Z-directional expansion that converts the thin graphite flakes into “worms”

(Figure 1.1b) of “exfoliated graphite” or “expanded graphite” (EG), which has an internal

structure consisting of thin graphite packets that remain connected at certain points. These

EG “worms” can be rolled or pressed to reconstitute graphite sheets that now contain

internal porosity separating ultrathin flakes and are thus flexible. This flexible graphite has

thermal, electrical, and chemical properties inherited from the graphite precursor, but is

soft and can be bent, rolled, or hand cut, and is used in a variety of industrial products

3

ranging from high-temperature, corrosion-resistant seals to heat spreaders used as backing

substrates in electronic devices.10

Figure 1. 1. A brief history of ultrathin carbon materials. (a) The charcoal-based painting in

Chauvet Cave, southern France, ~30,000 BP.5-6 Inset: close-up view of the carbon deposit.11 (b)

Morphology of expanded graphite made by intercalation and explosive thermal expansion, and

often used after subsequent compression to make flexible graphite products.12 (d) Scanning electron

micrograph from Geim and Novoselov showing a relatively large graphene crystal, whose faces

are clearly zigzag and armchair edges (see inset7). Reproduced from Liu, 2018.13

This thermal expansion of intercalated graphite is similar to some processes used

today to make multilayer graphene, or “graphene nanoplatelets”, in which the thin packets

in exfoliated graphite are more completely separated to make the distinct flakes or

nanoplatelets desired for compounding into composite materials. Then first single layer

graphene is demonstrated using tape-based mechanical exfoliation (Figure 1.1c), which

can provide micro-sized perfect nanosheets. To further scale up and mass produce defect-

4

free nanosheets, the development of bottom-up methods such as chemical vapor deposition

(CVD) is moving into high gear.

1.2 Graphene oxide and 2D nanochannels

Another route to ultrathin carbon forms passes through graphite oxide as an intermediate.

The term graphite oxide refers to the solid products of one of several protocols that use

intercalative oxidation with oxidants sufficiently powerful to attack the graphite basal plane.

14-19 Early graphite oxide synthesis dates back to the 1800’s, and for many years was the

subject of research as a bulk material.20-21 Graphite oxide decomposes on heating to release

gaseous products, which, in a manner similar to the intercalants described previously,

exfoliate the bulk material into ultrathin layer packets, some approaching monolayer

thickness. Figure 1.2a shows an early example of these “sehr dünnen Kohlenstoff -Folien”

(very thin carbon sheets) that today would be referred to as “reduced graphene oxide

nanosheets (rGO)”. Graphite oxide also undergoes near-spontaneous exfoliation in

aqueous media, without heating, to make GO with the full complement of oxygen-

containing groups originally formed through reaction in the expanded interlayer spaces of

the bulk graphite. The graphite-oxide route to ultrathin carbons (GO or rGO) has become

very popular due to the inherent scalability of this wet-chemical, natural-graphite-based

process. Indeed, most research on graphene assembly into new carbon architectures use

one of the bulk graphene-based materials (exfoliated graphite nanoplatelets or GO) rather

than forms made by CVD or the tape-based mechanical exfoliation used in early

5

fundamental studies. Modern GO nanosheets now can be mass produced in colloidal stable

suspensions by oxidation-sonication method (Figure 1.2b).

Angstrom-scale nanochannels in GO are promising in emerging technologies, for its

sub-nanosized nanochannels, high hydrophilicity and ease of scalable processing.22-23 GO

nanochannels – with interlayer spacing ~8 angstroms – are capable to selectively pass

molecules according to their physical size and chemical polarity (Figure 1.2c).24 When

stack GO nanosheets together, they automatically form a lamellar film where 2D

nanochannels are formed between each layer of nanosheets. This stacking is highly

reversible in water due to the absorption of water molecules and repulsive interactions

between ionized GO nanosheets. Once associated with long term water immersion, the

interlayer spacing will increase from 0.8 to 6 nm, resulting in failure of stability and

selectivity.25 Moreover, GO suspension can be deposited on various surfaces and form a

conformal coating, where the 2D nanochannels remain orderly arranged along with surface

topographies (Figure 1.2c).

6

Figure 1. 2. Morphologies and structures of GO. (a) Early electron microscope image of “sehr

dünnen Kohlenstoff-Folien” (very thin carbon sheets) produced by thermal exfoliation of graphite

oxide.26 (b) Modern AFM image of GO, showing a lateral size of 1-5 μm and thickness of ~ 1nm.27

(c) Schematic of planar and textured GO films, the interlayer spacing d is ~0.8 nm. 2D

nanochannels are orderly arranged in both structures. Reproduced from Liu 2018.13

1.3 Assembly and applications of textured 2D nanochannels

The rise of graphene in the 21st century is nevertheless a revolution for our field, because

for the first time we have access to isolated graphene layers to manipulate and use in

materials synthesis. The graphene layers in bulk carbons are the product of in situ organic

self-assembly, driven by chemical thermodynamics and the low-free energy of extended

conjugated structures. Our ability to control this assembly was limited, however, to the

selection of conditions (precursor, temperature, pressure) or the use of certain processing

tricks designed to improve graphene layer alignment (such as fiber spinning with discotic

flow alignment, hot stretching during polymer fiber carbonization, Z-directional

compression). Today the ready availability of isolated graphene sheets opens up a

completely new approach for creating designer carbon materials - an approach that

involves the manipulation of these pre-formed graphene.

Inspired by the new ability to assemble pre-fabricated graphene nanosheets, one of the

first questions that arises is what to make? Interestingly, this ex situ sheet assembly

approach has less in common with traditional carbonization approaches (where graphene

layers grow in situ) than it does with macroscopic paper- or fabric-based fabrication

methods common in everyday life. As such, researchers have been able to easily envision

microscopic versions of macroscopic objects such as multi-ply papers, sacks, crumpled

paper balls, wrinkled or textured films or coating, or complex origami/kirigami artwork

7

(Figure 1.3a-e). The creation of such nanosheet assemblies has become a significant

subfield in graphene research, and several recent reviews have covered the rapidly

expanding literature.3, 28-29 While some of the early work on graphene folding used pristine

graphene (from mechanical exfoliation or CVD), much of the recent work uses GO as the

nanosheet precursor for several reasons. First, it is easily processed as an aqueous

suspension, which under the right conditions (low-to-neutral pH, low ionic strength)

maintains the identify of individual nanosheets in their atomically thin and flexible form,

and prevents uncontrolled aggregation or premature sheet-stacking that destroy the

uniformity of the colloidal phase and the final product. Secondly, many applications of 3D

nanosheet assemblies will ultimately require significant quantities of nanosheet precursor,

and thus favor exfoliation-derived nanosheets that can more easily be produced at large

scale.

Figure 1. 3. Graphene nanosheet assemblies and their paper/fabric analogs. (a) Multilayer

graphene film and its analogy with book pages.30 (b) Wrinkled graphene film and its analogy with

corrugated cardboard.31 (c) Crumpled graphene film and its analogy with crumpled fabric.31 (d)

Graphene nanosacks as encapsulating agents and their analogy with a paper sack.32 (e) Graphene

aerogel structure and its analogy with a house of cards.33 Reproduced from Liu, 2018.13

8

The simplest of all nanosheet assemblies is the tiled film, which can be fabricated by

GO suspension casting or filtration, and even forms spontaneously when GO suspensions

are spilled and left to dry. If they are thick enough, GO deposits they can be removed from

underlying substrate to become free-standing GO papers34-35 or can be left in place as

ultrathin coatings or membranes on a backing support. Even this simplest architecture can

show emergent properties and advanced functionality. Spontaneous hydration swells GO

films and enlarges their interlayer spacing to create molecular sieve membranes that pass

water, but exclude solutes larger than about 0.9 nm in hydrated diameter.36-38 Restricting

hydration swelling and achieving precise control of the interlayer spacing is an active

research area, and covalent cross linking39-40 or external pressure41-42 have been used to

target particular separation challenges, including water desalination.43 Rather than

restricting swelling, an alternative goal can be interlayer space enlargement through

pillaring agents, and precise channel size control can be used to create tailored

ultrafiltration membranes. The simple tiled films can also serve as a starting point or

platform for more advanced structures that incorporate engineered wrinkling or crumpling.

28, 44 Periodic wrinkle textures,45 isotropic compression-induced crumpling46 or complex,

multi-length-scale fractal-like patterns have been created in GO films to enhance surface

area for catalysis, sensing applications or for stretchable barriers or devices.47-51

Furthermore, those 2D nanochannels that are deformed with the matrix are also promising

in permeation, drug eluting and templating studies. Manipulation of 2D nanochannels in

versatile GO architectures opens more opportunities in 2D reaction and fluidics within

spatial confined spaces.

9

1.4 References

1. Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang, Y.; Dubonos, S.

V.; Grigorieva, I. V.; Firsov, A. A., Electric Field Effect in Atomically Thin Carbon Films.

Science 2004, 306 (5696), 666.

2. Li, Z.; Wu, S.; Lv, W.; Shao, J.-J.; Kang, F.; Yang, Q.-H., Graphene Emerges as a

Versatile Template for Materials Preparation. Small 2016, 12 (20), 2674-2688.

3. Chen, P.-Y.; Liu, M.; Wang, Z.; Hurt, R. H.; Wong, I. Y., From Flatland to

Spaceland: Higher Dimensional Patterning with Two-Dimensional Materials. Adv. Mater.

2017, 29 (23), 1605096.

4. Quan, Q.; Lin, X.; Zhang, N.; Xu, Y.-J., Graphene and Its Derivatives as Versatile

Templates for Materials Synthesis and Functional Applications. Nanoscale 2017, 9 (7),

2398-2416.

5. Clottes, J. A., M., La grotte Chauvet: l'art des Origines. Seuil: Paris, 2001.

6. Valladas, H.; Clottes, J.; Geneste, J. M.; Garcia, M. A.; Arnold, M.; Cachier, H.;

Tisnérat-Laborde, N., Evolution of Prehistoric Cave Art. Nature 2001, 413 (6855), 479-

479.

7. Geim, A. K.; Novoselov, K. S., The Rise of Graphene. Nat. Mater. 2007, 6 (3),

183-191.

8. Kang, F.; Leng, Y.; Zhang, T.-Y., Influences of H2O2 on Synthesis of H2SO4-

GICs. J. Phys. Chem. Solids 1996, 57 (6), 889-892.

10

9. Kovtyukhova, N. I.; Wang, Y.; Berkdemir, A.; Cruz-Silva, R.; Terrones, M.; Crespi,

V. H.; Mallouk, T. E., Non-Oxidative Intercalation and Exfoliation of Graphite by

Brønsted Acids. Nat. Chem. 2014, 6 (11), 957-963.

10. Tang, K.; Ruan, J.; Huang, D.; Zhan, Q.; Xiao, W.; Li, H. In Application Study of

Flexible Graphite Grounding Electrode in Typical Tower Grounding Grid, 2016 IEEE

International Conference on High Voltage Engineering and Application (ICHVE), 19-22

Sept. 2016; 2016; pp 1-4.

11. Combier, J.; Jouve, G., Chauvet Cave's Art is not Aurignacian: A New Examination

of the Archaeological Evidence and Dating Procedures. Quartär 2012, 59.

12. Dowell, M. B.; Howard, R. A., Tensile and Cmpressive Properties of Flexible

Graphite Foils. Carbon 1986, 24 (3), 311-323.

13. Liu, M.; Castilho, C. J.; Hurt, R. H., New Material Architectures Through Graphene

Nanosheet Assembly. Adv. Mater. Lett. 2018, 9 (12), 843-850.

14. Brodie, B. C., XIII. On the Atomic Weight of Graphite. Royal Society: 1859; Vol.

149, p 249-259.

15. Staudenmaier, L., Verfahren zur Darstellung der Graphitsäure. Ber. Dtsch. Chem.

Ges. 1898, 31 (2), 1481-1487.

16. Hofmann, U.; König, E., Untersuchungen über Graphitoxyd. Z. Anorg. Allg. Chem.

1937, 234 (4), 311-336.

17. Hummers, W. S.; Offeman, R. E., Preparation of Graphitic Oxide. J. Am. Chem.

Soc. 1958, 80 (6), 1339-1339.

18. Hofmann, U.; Holst, R., Über die Säurenatur und die Methylierung von

Graphitoxyd. John Wiley & Sons, Ltd: 1939; Vol. 72, p 754-771.

11

19. Guo, F.; Silverberg, G.; Bowers, S.; Kim, S.-P.; Datta, D.; Shenoy, V.; Hurt, R. H.,

Graphene-based environmental barriers. Environ. Sci. Technol. 2012, 46 (14), 7717-7724.

20. Beckett, R. J.; Croft, R. C., The Structure of Graphite Oxide. J. Phys. Chem. 1952,

56 (8), 929-935.

21. Boehm, H. P.; Clauss, A.; Fischer, G.; Hofmann, U., Surface Properties of

Extremely Thin Graphite Lamellae. In Proceedings of the Fifth Conference on Carbon,

Pergamon: 1962; pp 73-80.

22. Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S., The chemistry of graphene

oxide. Chem. Soc. Rev. 2010, 39 (1), 228-240.

23. Dimiev, A. M.; Alemany, L. B.; Tour, J. M., Graphene oxide. Origin of acidity, its

instability in water, and a new dynamic structural model. ACS Nano 2013, 7 (1), 576-588.

24. Nair, R. R.; Wu, H. A.; Jayaram, P. N.; Grigorieva, I. V.; Geim, A. K., Unimpeded

permeation of water through helium-leak–tight graphene-based membranes. Science 2012,

335 (6067), 442-444.

25. Zheng, S.; Tu, Q.; Urban, J. J.; Li, S.; Mi, B., Swelling of graphene oxide

membranes in aqueous solution: characterization of interlayer spacing and insight into

water transport mechanisms. ACS Nano 2017, 11 (6), 6440-6450.

26. Boehm, H.-P., Graphene—How a Laboratory Curiosity Suddenly Became

Extremely Interesting. Angew Chem. Int. Ed. 2010, 49 (49), 9332-9335.

27. Liu, M.; Chen, P.-Y.; Hurt, R. H., Graphene Inks as Versatile Templates for

Printing Tiled Metal Oxide Crystalline Films. Adv. Mater. 2018, 30 (4), 1705080.

12

28. Deng, S.; Berry, V., Wrinkled, Rippled and Crumpled Graphene: An Overview of

Formation Mechanism, Electronic Properties, and Applications. Mater. Today 2016, 19 (4),

197-212.

29. Zhang, M.; Hou, C.; Halder, A.; Wang, H.; Chi, Q., Graphene Papers: Smart

Architecture and Specific Functionalization for Biomimetics, Electrocatalytic Sensing and

Energy Storage. Mater. Chem. Front. 2017, 1 (1), 37-60.

30. Huang, H.; Song, Z.; Wei, N.; Shi, L.; Mao, Y.; Ying, Y.; Sun, L.; Xu, Z.; Peng,

X., Ultrafast Viscous Water Flow through Nanostrand-Channelled Graphene Oxide

Membranes. Nat. Commun. 2013, 4 (1), 2979.

31. Chen, P.-Y.; Sodhi, J.; Qiu, Y.; Valentin, T. M.; Steinberg, R. S.; Wang, Z.; Hurt,

R. H.; Wong, I. Y., Multiscale Graphene Topographies Programmed by Sequential

Mechanical Deformation. Adv. Mater. 2016, 28 (18), 3564-3571.

32. Chen, Y.; Guo, F.; Jachak, A.; Kim, S.-P.; Datta, D.; Liu, J.; Kulaots, I.; Vaslet, C.;

Jang, H. D.; Huang, J.; Kane, A.; Shenoy, V. B.; Hurt, R. H., Aerosol Synthesis of Cargo-

Filled Graphene Nanosacks. Nano Lett. 2012, 12 (4), 1996-2002.

33. Qiu, L.; Liu, J. Z.; Chang, S. L. Y.; Wu, Y.; Li, D., Biomimetic Superelastic

Graphene-Based Cellular Monoliths. Nat. Commun. 2012, 3 (1), 1241.

34. Dikin, D. A.; Stankovich, S.; Zimney, E. J.; Piner, R. D.; Dommett, G. H. B.;

Evmenenko, G.; Nguyen, S. T.; Ruoff, R. S., Preparation and Characterization of Graphene

Oxide Paper. Nature 2007, 448 (7152), 457-460.

35. Chen, C.; Yang, Q.-H.; Yang, Y.; Lv, W.; Wen, Y.; Hou, P.-X.; Wang, M.; Cheng,

H.-M., Self-Assembled Free-Standing Graphite Oxide Membrane. Adv. Mater. 2009, 21

(29), 3007-3011.

13

36. Joshi, R. K.; Carbone, P.; Wang, F. C.; Kravets, V. G.; Su, Y.; Grigorieva, I. V.;

Wu, H. A.; Geim, A. K.; Nair, R. R., Precise and ultrafast molecular sieving through

graphene oxide membranes. Science 2014, 343 (6172), 752-754.

37. Abraham, J.; Vasu, K. S.; Williams, C. D.; Gopinadhan, K.; Su, Y.; Cherian, C. T.;

Dix, J.; Prestat, E.; Haigh, S. J.; Grigorieva, I. V.; Carbone, P.; Geim, A. K.; Nair, R. R.,

Tunable sieving of ions using graphene oxide membranes. Nat. Nanotechnol. 2017, 12 (6),

546-550.

38. Spitz Steinberg, R.; Cruz, M.; Mahfouz, N. G. A.; Qiu, Y.; Hurt, R. H., Breathable

vapor toxicant barriers based on multilayer graphene oxide. ACS Nano 2017, 11 (6), 5670-

5679.

39. Park, S.; Lee, K.-S.; Bozoklu, G.; Cai, W.; Nguyen, S. T.; Ruoff, R. S., Graphene

Oxide Papers Modified by Divalent Ions—Enhancing Mechanical Properties via Chemical

Cross-Linking. ACS Nano 2008, 2 (3), 572-578.

40. Chen, L.; Shi, G.; Shen, J.; Peng, B.; Zhang, B.; Wang, Y.; Bian, F.; Wang, J.; Li,

D.; Qian, Z.; Xu, G.; Liu, G.; Zeng, J.; Zhang, L.; Yang, Y.; Zhou, G.; Wu, M.; Jin, W.; Li,

J.; Fang, H., Ion Sieving in Graphene Oxide Membranes via Cationic Control of Interlayer

Spacing. Nature 2017, 550 (7676), 380-383.

41. Huang, H.; Mao, Y.; Ying, Y.; Liu, Y.; Sun, L.; Peng, X., Salt Concentration, pH

and Pressure Controlled Separation of Small Molecules through Lamellar Graphene Oxide

Membranes. ChemComm. 2013, 49 (53), 5963-5965.

42. Hung, W.-S.; An, Q.-F.; De Guzman, M.; Lin, H.-Y.; Huang, S.-H.; Liu, W.-R.;

Hu, C.-C.; Lee, K.-R.; Lai, J.-Y., Pressure-Assisted Self-Assembly Technique for

14

Fabricating Composite Membranes Consisting of Highly Ordered Selective Laminate

Layers of Amphiphilic Graphene Oxide. Carbon 2014, 68, 670-677.

43. Hegab, H. M.; Zou, L., Graphene Oxide-Assisted Membranes: Fabrication and

Potential Applications in Desalination and Water Purification. J. Membr. Sci. 2015, 484,

95-106.

44. Chen, P.-Y.; Liu, M.; Valentin, T. M.; Wang, Z.; Spitz Steinberg, R.; Sodhi, J.;

Wong, I. Y.; Hurt, R. H., Hierarchical Metal Oxide Topographies Replicated from Highly

Textured Graphene Oxide by Intercalation Templating. ACS Nano 2016, 10 (12), 10869-

10879.

45. Wang, Z.; Tonderys, D.; Leggett, S. E.; Williams, E. K.; Kiani, M. T.; Steinberg,

R. S.; Qiu, Y.; Wong, I. Y.; Hurt, R. H., Wrinkled, Wavelength-Tunable Graphene-Based

Surface Topographies for Directing Cell Alignment and Morphology. Carbon 2016, 97,

14-24.

46. Zang, J.; Ryu, S.; Pugno, N.; Wang, Q.; Tu, Q.; Buehler, M. J.; Zhao, X.,

Multifunctionality and Control of the Crumpling and Unfolding of Large-Area Graphene.

Nat. Mater. 2013, 12 (4), 321-325.

47. Kelley, S. O.; Mirkin, C. A.; Walt, D. R.; Ismagilov, R. F.; Toner, M.; Sargent, E.

H., Advancing the Speed, Sensitivity and Accuracy of Biomolecular Detection Using

Multi-Length-Scale Engineering. Nat. Nanotechnol. 2014, 9 (12), 969-980.

48. Goh, K.; Karahan, H. E.; Wei, L.; Bae, T.-H.; Fane, A. G.; Wang, R.; Chen, Y.,

Carbon nanomaterials for advancing separation membranes: A strategic perspective.

Carbon 2016, 109, 694-710.

15

49. Mistry, H.; Varela, A. S.; Kühl, S.; Strasser, P.; Cuenya, B. R., Nanostructured

Electrocatalysts with Tunable Activity and Selectivity. Nat. Rev. Mater. 2016, 1 (4), 16009.

50. Sun, Y.; Liu, N.; Cui, Y., Promises and Challenges of Nanomaterials for Lithium-

Based Rechargeable Batteries. Nat. Energy 2016, 1 (7), 16071.

51. Chen, P.-Y.; Zhang, M.; Liu, M.; Wong, I. Y.; Hurt, R. H., Ultrastretchable

Graphene-Based Molecular Barriers for Chemical Protection, Detection, and Actuation.

ACS Nano 2018, 12 (1), 234-244.

16

Chapter 2 Textured 2D Nanochannels as Versatile Templates for Templating

Tiled Metal Oxide Crystalline Films

The content of this chapter is a modified version of published paper “Muchun Liu, Po‐Yen

Chen, Robert H. Hurt. Adv. Mater., 2018, 30, 1705080.” and has been reproduced here

with permission. Copyright © 2017 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim.

2.1 Introduction

Research on 2D materials has focused on phases with layered crystal structures that can be

exfoliated from bulk lamellar materials or assembled through atomic-level growth along

preferred crystal axes.1-3 There is also interest in creating high-aspect-ratio plate-like

nanomaterials from non-lamellar materials through synthesis methods that direct growth

in two dimensions through templating. One approach uses the interlayer spaces in naturally

layered materials as reaction vessels to direct 2D growth through confinement.4-5 An

example is the intercalation of metal ions into multilayer GO followed by annealing and

oxidation to remove the template and produce metal oxide films.6-7 This process, however,

does not typically produce true atomic-scale growth of platelet crystals, but rather the

nucleation of simple equi-axed nanoparticles that assemble by confined sintering into

particle arrays that have the macroscopic form of porous sheets.7-8 The particle arrays have

the gross structure of nanosheets, but their 2D anisotropy lies not in their atomic

arrangements, but rather only in the preferred directions of the secondary particle-particle

sintered contacts. In some applications this interparticle porosity offers advantages,9 but

generally particle arrays lack the intrinsic anisotropic properties of platelet crystals and the

17

existing methods offer limited control over grain size, composition, porosity, and film

strength.4, 6-7, 10 We hypothesized that atomic-scale 2D growth is being prevented by

limitations on metal-carbon ratios achieved by the intercalation method -- the metal binding

sites in GO gallery spaces are too few to achieve the high metal loadings needed to replicate

the layered host with metal oxide at full density. Several studies report anisotropic growth

of metal oxides by non-intercalation routes, in which GO and metal salts are premixed in

solvent with ligands or cross-linking agents, and yield isolated micron-scale platelets either

within or liberated from the graphene template.11-12 The present goal to make fully dense

tiled-crystal metal oxide films that reproduce the micro- and macrostructure of a multilayer

graphene host. Achieving this goal would allow the many different wrinkled, folded and

patterned graphene assemblies demonstrated in the recent literature to be transcribed into

textured metal oxide films that would otherwise be too brittle to fabricate directly by

mechanical deformation methods.7, 13

To move from porous particle arrays to dense films requires an increase in metal-carbon

ratio beyond the natural limits set by intercalation phenomena.6-7 Solution-phase mixing of

nanosheets and metal cations allows wide ranging control of metal-carbon ratios, but

typically leads to colloidal instability and poor film quality at high metal ion concentration

due to salt-induced electrostatic screening and flocculation.14 Here we demonstrate a new

version of graphene templating, based on metal ion-GO nanosheet precursors engineered

for colloidal stability over a wide range of metal-carbon ratios. The “metallized graphene

ink” (MGI) precursors can be easily cast into high-quality planar or microtextured films,

which upon annealing achieve true 2D growth in the form of fully-dense, tiled films of Fe

and Co oxides. These films have the structure of mathematical tessellations, and are

18

sufficiently robust to serve as free standing films or to adopt and maintain complex out-of-

plane microtextures.

2.2 Results and discussion

Figure 2.1a and Figure 6.1 (Appendix to Chapter 2), show the example fabrication route,

in which metal precursor salts are dissolved in GO suspensions and cast onto thermally

responsive polymer substrates (polystyrene) to make conformal coatings, followed by

heating to trigger polymer contraction and compressive surface film texturing. The success

of this fabrication depends on colloidal stability of the mixed metal-nanosheet precursor

suspensions, so colloidal behavior was studied in detail. Figure 2.1b shows flocculation

behavior and electrophoretic mobility (plotted as -potential) over a wide range of metal-

carbon ratio (C content of GO is estimated based on its atomic C/O ~2.1). The Ag(I)-GO

system is observed to be stable and to experience significant -/- repulsion over the entire

concentration range. Most other metal cations are observed to flocculate at metal-carbon

ratios greater than about 1.5/1, which corresponds to their entry into the low-surface-charge

window ( > -15mV and < +15mV) where colloidal instability is expected (Figure 2.1b).

Interestingly, the Fe(III)-GO colloids also enter the instability window, but pass through

quickly as [Fe(III)] increases. That system undergoes a charge inversion at Fe/C atomic

ratios ~1/50 and remains above the observed +15mV stability threshold up to very high

[Fe(III)]. It is clear that Fe(III) complexation on GO sites overwhelms the native negative

charge on GO, and if the system can be induced to pass quickly through the charge

inversion region, a stable Fe(III)-GO colloid can be maintained for film processing in the

19

high Fe loading regime. According to the chemical structures and morphologies of GO and

Fe(III)-GO nanosheets (Figure 6.2, 6.3, Appendix to Chapter 2), Fe(III) ions are mostly

binding with carboxyl groups on GO, forming O=C–O–Fe complexes.15 Besides, after

adding Fe (III) ions, the height of GO nanosheets increased from ~1 to 2 nm (each GO

nanosheet attracts two layers of Fe(III) ions with hemispherical hydrated shells, the

diameter of hydrated Fe(III) ion is ~9 Å).16 The single nanosheet state enables the

subsequent uniform stacking of Fe(III)-GO nanosheets. The initial problems creating a

stable Co(II)-GO colloid could also be circumvented by addition of [Fe(III)], which then

shows a similar surface charge inversion (Figure 6.4, Appendix to Chapter 2).

The -potential behavior can be understood through a model of cation complexation and

electrostatic charge screening, as follows. Equilibrium constants for cation-GO

complexation K = [Mn+-GO*]/[Mn+][GO*] were estimated from published data on cation-

monocarboxylate complexes,17-18 where GO* represents an oxygen-containing binding site

on GO with unit negative charge, and used to estimate the population of surface bound

cations in the total pool of metal cations in the system. The resulting surface charge is then

the sum of the cation (+) and native GO (-) surface charges and can be converted to -

potential using the Gouy-Chapman equation (1)

𝜎𝑠 =2𝜀𝑘𝑇𝜅

𝑧𝑒sinh(

𝑧𝑒𝜁

2𝑘𝑇) (1)

to account for electrostatic screening (see Appendix of Chapter 2 for details).19-20 The

estimated zeta potentials are shown in Figure 2.1b for different metal ion-GO colloids for

comparison to the experimental data. The charge reversal is successfully predicted and

other data trends are in good agreement, indicating that cation complexation and charge

screening are the main factors determining colloidal stability in these systems.

20

We then used this colloidal stability theory to pursue our main goal of achieving 2D

growth and fully-dense oxide ultrathin films. Casting the Fe(III)-based MGI onto substrates

produces intercalated GO structures as shown by time-resolved XRD (Figure 2.1c).

Comparing the drying behavior of pure GO and Fe(III)-GO, the cation is seen to increase

the GO interlayer spacing (8.4 to 8.9 Å) and produce a secondary peak at ~17° with

calculated spacing of ~5.1Å (grey arrow), which likely reflects the presence of interlayer

Fe(III) structures. The final products of Fe oxide textured films and original GO textured

film are shown in Figure 2.1d. Other metal-ion/GO colloids are also tested and discussed

(Figure 6.5, Appendix to Chapter 2). The repulsive nanosheet colloids serve as the “MGIs”,

which we develop and exploit here as versatile precursors for creation of metal oxide

structures.

Figure 2. 1. Textured metal oxide films from MGI. a, Textured metal oxide films fabricated by

conformal coating and subsequent compression of Fe(III)-based MGI on thermally responsive

polymer substrates. Heating above glass transition temperature (Tg ~ 100 °C) triggers polymer

relaxation to the contracted state to produce buckle textures in the metal ion-GO composite film.

21

The film is calcined at 600 °C to oxidatively destroy the graphene and convert the metal ions to a

textured oxide film. After removing GO by air oxidation, free-standing metal oxide crumpled films

were obtained. b, Top: behavior of GO nanosheet dispersions in the presence of various soluble

metal salts (From left: 0.1mg ml-1 GO, 0.1mg ml-1 5 mM Fe(III)-, Co(II)- and Fe(III)/Co(II)-based

GO suspensions, respectively). Bottom: experimental and theoretical (model-predicted) -

potentials of 0.1mg ml-1 GO dispersions as function of metal cation concentration, [Mn+], showing

negative-to-positive charge inversion on addition of Fe(III) salts. c, Time-resolved X-ray

diffractograms tracking the appearance of lamellar structures in GO and Fe(III)-GO composite

films during drying. d, SEM images of a final Fe oxide textured film (false color) fabricated by

MGI templating, and a pure GO textured film for reference to show the replication ability. The

Fe(III)-based MGI is stable at GO concentrations up to 3 mg ml-1, and 0.65 mg ml-1 GO with Fe-C

ratio of 1:3 were used to prepare the films in panels d,e. Scale bar, 2 µm.

Figure 2.2a-b shows the effect of varying metal-carbon ratio on the structure of Fe oxide

films made through the Fe(III)-graphene ink. The polymer thermal contraction technique

was used to create microtextures that we hypothesized would be preserved in the Fe oxide

film after graphene template removal. Viewed at the microscale, all samples do indeed

replicate the characteristic GO crumple textures (2.2a). In contrast, the nanostructures (2.2b)

evolve dramatically as metal-carbon ratio increases. At low metal loading, we reproduce

the particle arrays seen in previous studies,6-7 but as metal loading increases the arrays lose

porosity (2.2c, 6.6, Appendix to Chapter 2) and then the equi-axed primary particles

transition to anisotropic platelet crystals that gradually fill 2D space. The close-packed tiled

structure is easily visible by TEM (2.2b), and visible in SEM only by close inspection as a

subtle snake-skin pattern (2.2a). At extremely high metal-carbon ratio (33/1), the excess

metal overwhelms the confining effect of graphene and leads to the appearance of a bulk

oxide phase with texture visible only on one surface (Figure 6.7a, Appendix to Chapter 2).

Figure 2.2d shows cross sectional micrographs of a textured Fe oxide film with initial

metal-carbon ratio of 3/1. The metal oxide body has fractured in a way that clearly reveals

an underlying multi-layered structure, in which individual tiled metal-oxide monolayers

are stacked (in planar regions) or nested (on ridges) to make the complete film. A range of

22

edge-on images (Figure 2.2d, 2.3d, 6.8, Appendix to Chapter 2) shows that the thickness

of these monolayers varies from about 20 - 75 nm, and generally increases with increasing

Fe-C ratio in the starting MGI deposit.

The crystal structures of the Fe oxide films were investigated further using HRTEM and

SAED. The tiled films consist primarily of α-phase Fe2O3 nanoplatelets whose faces are

(001) basal planes, and with a variety of side-plane structures (Figure 6.7b-d, Appendix to

Chapter 2). This mosaic texture is consistent with their independent nucleation and atomic-

scale growth followed in the later stages by side-plane interactions of the randomly oriented

platelets that fuse to fill 2D space. Individual Fe2O3 platelet crystals and related

nanoplatelet-composites have been reported previously by liquid phase growth or GO-

directed assembly.21-22 These examples of 2D growth required the addition of a molecular

ligand as structure directing agent, however, and in the absence of the ligand only Fe2O3

particles were formed.22 The current space-filling platelet crystals resemble mathematical

tessellation patterns - the tiling of a plane using one or more geometric shapes with no

overlap or gaps (Figure 2.2e). The anisotropic platelet crystals are a clear example of

graphene-directed 2D atomic assembly, and are fundamentally distinct from the particle

arrays reported previously,4-5 which represent a form of directional sintering of simple

primary particles without true crystal anisotropy.

23

Figure 2. 2. Effect of metal-carbon ratio on the micro- and nanostructures of Fe oxide

textured films fabricated from MGI. a, Metal oxide microstructures created by the thermal

compression texturing technique after MGI deposition. The polystyrene substrate undergoes

contraction to 20% of its original area, leading to highly crumpled MGI structures by out-of-plane

deformation followed by annealing and template removal. Note that the length scale of the

microtextural features is approximately independent of metal-carbon ratio. The initial metal-carbon

ratios of MGI are 1/333, 1/33, 1/3 and 3/1, respectively. Scale bar, 1 µm. b, Control of nanostructure,

which can be varied from sintered particle arrays at low metal loading to fully dense tessellated

platelet-crystal tiled films at high metal loading. Scale bar, 10 nm. c, Quantitative porosity of the

tessellated films by image analysis. d, Side view of textured Fe oxide films with initial metal-

carbon ratio of 3/1, showing multilayered structures. The inner layer exhibits similar intact tiled

pattern. Scale bar, 1 µm. Right: closed view of a piece of Fe oxide tessellation structure. Scale bar,

200 nm. e, Sketch of conversion of MGI to tessellated metal oxide film. The resulting oxide

nanoplatelet is α-Fe2O3 with basal surfaces that are primarily (001) planes.

24

We were interested whether 2D confinement was necessary to template these platelet

crystals, or whether one-sided contact with graphene would suffice by reducing the (001)

surface energy and inhibiting (001) facet growth by a thermodynamic effect. We carried

out an auxiliary experiment in which iron salts were deposited on the highly ordered

pyrolytic graphite (HOPG) basal plane followed by annealing and oxidation, but observed

only particles, and no platelet crystals (Figure 6.9, Appendix to Chapter 2). This supports

a kinetic interpretation of our main result, in which 2D confinement limits Z-directional

reactant mobility and Z-directional growth rates, leading to platelet crystals oriented

parallel to the confining graphene walls.

TGA experiments (Figure 6.10, Appendix to Chapter 2) were conducted to better

understand the sequence of events during calcination. These results show that the Fe(NO3)3

salt decomposes first (at the lowest temperature), and its decomposition is complete by 200

C. The rGO oxidation occurs much later, above 450 C under these conditions, and is only

slightly affected by the presence of the iron (Figure 6.10, Appendix to Chapter 2). The

comparison of these two reaction onset temperatures suggests that iron oxide structures

form first as the MGI sample is heated, and thus the platelets assemble and grow in the

presence of the intact rGO, which is only later removed as the oven temperature exceeds

450 C. We propose that this sequence, which occurs naturally upon heating, is important

for successful templating, as the main oxide assembly process must occur while the layered

host is still intact and able to direct the growth.

The ability to assemble platelet-crystal films is not limited to the Fe oxide system.

Although cobalt salts at high concentration lead to colloidal instability of GO (Figure 2.1b),

the addition of Fe(NO3)3 to make a mixed metal system restores colloidal stability. The

25

resulting mixed Co/Fe-oxide films exhibit similar micro- and nano-structures as the pure

iron system (Figure 6.11, Appendix to Chapter 2). The resulting oxide at Fe/Co atomic

ratio of 1.0 is a mixture of CoFe2O4/Co2FeO4, and XRD clearly demonstrates the existence

of the metastable Co2FeO4 phase (Figure 6.12, Appendix to Chapter 2). It is noteworthy

that we see similar tiled oxide structures in both the iron and cobalt-iron oxide systems.

We propose that the tiled structure is imposed on the oxide by the geometry of the graphene

gallery spaces and the stoichiometry (metal/carbon ratio), which directs the preferred

growth direction (X,Y) and sets the final film density. Being a geometric, not chemical

templating effect, we propose that the tiled structure is forced on the oxide films

irrespective of their specific chemistry, and irrespective of their intrinsic surface free

energies and normally preferred crystalline growth patterns that are normally associated

with the specific metal oxide phase.

The potential of MGIs to direct the growth of robust, 2D-structured metal oxide films

suggests their use in printing, patterning, or paper folding applications (Figure 2.3). First

we imagined that these robust metal oxide films could adopt a wide variety of

microtextures, not limited to simple modes of compressive wrinkling and crumpling.

Figure 2.3a and 6.13 (Appendix to Chapter 2) show a demonstration of the ability of the

Fe(III)-MGI system to reproduce complex micropatterns found in nature. Applying the

MGI coating to a rhododendron leaf followed by drying and calcining yields free-standing

Fe2O3 tiled films that accurately replicate the leaf microtextural features, including the leaf

stomata used for plant transpiration and photosynthesis (Figure 2.3a). Replicas can also be

prepared of the surface textures on human hair, whose Fe2O3 counterpart shows the fine

cuticle structures that control moisture migration and protect the innermost hair shaft.

26

Beyond biotexture replication, one can fabricate 3D bodies by manipulation of planar MGI-

based papers (2.3b). MGI coatings of thickness >300 nm are observed to be sufficiently

strong to be removed and manipulated to make metal oxide strands or loops (2.3c-d). The

final bodies have a lamellar tessellated structure that replicates the multilayer GO paper

(2.3d, 6.13e, Appendix to Chapter 2). Here GO is required as a structure-directing agent in

the initial paper-forming stage. In the absence of GO, or if we replace GO nanosheets with

anionic polymer chains, the pure metal ions participate in uncontrolled particle growth and

yield powdery, low-quality coatings (Figure 6.14, Appendix to Chapter 2). Finally, the ink-

like character of MGI co-suspensions enables various writing/patterning modes by the

pneumatic airbrush method or by an artist’s paintbrush (2.3e). Additional work is needed

to develop the MGI as inks for injet, offset, or other machine printing applications.

Figure 2. 3. Example applications of MGI in biotexture replication, paper-based 3D shape

creation, and printing. a,b: Demonstration of biotexture replication: a, Fe2O3 replica of the

complex surface of Rhododendron x 'Roseum Elegans' leaf, constructed by complex structured

metal oxide tessellation (bottom left). Scale bar, 10 µm. b, Fe2O3 replica of human hair cuticles.

Scale bar, 10 µm. Inset: top view of Fe2O3 hair replica. Scale bar, 20 µm. c,d: Paper-based 3D

shape creation: c, Structure of Fe2O3 strand made by MGI paper scrolling. Scale bar, 5 µm. d,

Photos of Fe2O3 replicas of custom-designed 3D shapes, constructed by planar structured metal

oxide tessellation (bottom middle). Scale bar, 1 cm. Bottom right: side view of Fe2O3 loop, showing

multilayered structures. Scale bar, 500 nm. Printability: e, Fe2O3 patterns fabricated from MGI

27

writing and airbrush painting on ceramic substrates. Scale bar, 1 cm. For detailed results and

comparison with original templates, see Figure 6.13, Appendix to Chapter 2.

2.3 Conclusions

In summary, true 2D growth of tiled metal oxide ultrathin films can be achieved using

graphene oxide gallery-space templating at very high metal-carbon ratio. The use of

specially engineered colloidal suspensions we refer to as “metallized graphene inks” allow

fabrication of high quality planar or microtextured films with simultaneous control of

micro- and nanostructures (Figure 2.4). The ink precursor allows the films to be cast or

printed into patterns or to be used for biotexture replication or 3D shapes creation. In

addition we anticipate that this unique approach to structure control in metal oxide films

can be utilized for enhancing total surface area and specific crystal facet area for oxide

catalyst supports,23-24 or for the establishment of high-curvature ridge/valley structures for

electric field enhancement or band gap modulation of interest in electrocatalysis.25 We

anticipate the crumpled microtexture of these films will provide some ability to

accommodate strain through conformational changes to improve crack resistance in these

brittle materials. To prove the concept, an as-prepared Fe2O3 textured film was slightly

embedded into PDMS and stretched to 125% of its original size (Figure 6.15, Appendix to

Chapter 2). The combination of mechanical robustness and full areal density make these

films of interest in crack-resistant barrier coatings, or other magnetic or photocatalytic

functional coatings on substrates. Other potential applications include the creation of

bioceramic materials with surfaces engineered for antibacterial activity, or modulation of

cell orientation or phenotype.26-28

28

Figure 2. 4. Overview of assembly mechanisms and material structures fabricated from MGI.

left, Biotexture replication: Fe(III)-based MGI cast onto an arbitrarily complex natural surface

microtexture and converted to free-standing metal oxide replica by simple annealing and oxidation.

middle, MGI structure and the process of atomic assembly. The GO gallery spaces guide local 2D

growth of metal oxide platelet crystals that are liberated by graphene thermal oxidation. right,

Fe(III)-based MGI cast on planar substrates to form composite papers that can be manipulated and

oxidized into free-standing 3D metal oxide bodies.

2.4 Materials and methods

Materials

Ethanol, iron(III) nitrate nonahydrate (Fe(NO3)3·9H2O), cobalt(II) nitrate hexahydrate

(Co(NO3)2·6H2O), nickel(II) chloride (NiCl2), silver nitrate (AgNO3), Titanium(IV)

chloride (TiCl4) in 0.1M hydrogen chloride (HCl), aluminum nitrate nonahydrate

(Al(NO3)3·9H2O), hydrazine hydrate (NH2NH2·H2O) and poly(acrylic acid) were

29

purchased from Sigma-Aldrich. Graphite powder (SP-1) was purchased from Bay Carbon

Inc.. HOPG was purchased from Bruker Nano Inc.. Thermally responsive polyethylene

heat shrink films were bought from Grafix. Polydimethylsiloxane (PDMS) is made from

SYLGARD® 184 silicone elastomer kit. Rhododendron x 'Roseum Elegans' leaves were

collected in Brown campus, Providence, RI. Human hair was collected from the first author.

All water was deionized (18.2 MΩ, milli-Q pore). All reagents were used as received

without further purification.

Fabrication of textured GO

GO nanosheets were prepared by a modified Hummers’ method, with lateral size ~1 µm,

thickness ~1 nm, and characterized in Figure 6.3, Appendix to Chapter 2. The GO

suspensions used in the colloidal and film formation experiments range in concentration

from 0.1 - 0.65 mg ml-1, with a C/O atomic ratio of approximately 2.1. Potential GO

impurities - such as N, S, Mn, K, Cl and P - were not detected by XPS.29 The polymer

shrink film was cut into 4 cm2 squares and washed with ethanol. Once dry, samples were

treated with air plasma in a Deiner Atto standard plasma system with a borosilicate glass

chamber and a 13.56 MHz, 0−50 W generator. The chamber pressure was pumped down

to and maintained at 0.13 mbar while being flushed with air for 5 min. Plasma was then

generated at 100% power (50 W) for 90s followed by slow venting of the chamber. Next,

150 μl of a GO suspension was drop-cast onto the substrates. Once dry, the planar GO

films were obtained, and the samples were placed and allowed to shrink in an oven at

140 °C for 30 min.

Preparation of chemically reduced GO

30

Hydrazine dilute solution (2 wt %) was prepared and stirred overnight. The GO films were

immersed fully in the hydrazine solution; the reduction was allowed to proceed at 80 °C

for at least 24 h, and a dark brown colored rGO film was produced. The rGO structures

were sequentially rinsed with water and dried in an oven at 70 °C.

Fabrication of metal oxide textured films (Fe2O3, Co3O4, NiO, TiO2, Ag and CoxFe(3-

x)O4 (x=1,2))

Suspensions of Fe(III)-, Co(II)-, Ni(II)-, Ag(I)- Ti(IV)-, and Fe(III)/Co(II)-based MGI

were prepared in different concentrations. Next, 150 μl of MGI was drop-cast onto plasma-

treated polymer shrink film. Once dry, the planar MGI films were obtained, and the

samples were placed and allowed to shrink in an oven at 140 °C for 30 min. The samples

were then calcined at 600 °C in a furnace for at least 2 h. The heating/cooling ramps were

set at 40 °C min−1.

Biotexture replication into Fe2O3 phase using Fe(III)-based MGI

Leaf textures: 150 μl of a 10 mM Fe(III)-based MGI was prepared and drop-cast on a clean

Rhododendron leaf (cut into 2×2 cm2). Once dry, the Fe(III)-based MGI coated leaf were

placed and stabilized in oven at 140 °C for 30 min. The Fe(III)-based MGI coated leaf was

then calcined at 600 °C in a furnace for at least 2 h. The heating/cooling ramps were set at

40 °C min-1. The as-prepared Fe2O3 replicas were carefully peeled off from the leaf residue

with high precision tweezers. Hair textures: An air plasma-treated hair was dip-coated in

10 mM Fe(III)-GO MGI, and then fixed at both ends inside a 5 ml alumina combustion

boat at both sides. Once dry, the suspending Fe(III)-based MGI coated hair was calcined

at 600 °C in a furnace for 1 h.

3D shapes creation of Fe2O3 from Fe(III)-based MGI

31

50 ml of 10 mM Fe(III)-based MGI was spread by bar coating on a clean Teflon plate

(10×10 cm2), then the sample was dried overnight and carefully lifted off as MGI paper.

These papers consist of GO nanosheets with intercalated Fe(III) ions. By shaping the MGI

papers, we prepared MGI strands, loops or other 3D objects, which were then calcined at

600 °C in a furnace for at least 2 h.

Printability of Fe(III)-based MGI

Writing: An artist’s paintbrush was dipped in a pool of 20 mM Fe(III)-based MGI and

applied to a ceramic sheet, which was then dried in air and calcined in air at 600 °C in a

furnace for 1 h. Airbrush painting: A selected stencil was placed on a piece of ceramic

substrate, then airbrush sprayed with a 20 mM Fe(III)-based MGI. The sample was dried

in air and calcined in air at 600 °C in a furnace for 1 h.

Fabrication of PDMS-fixed textured Fe2O3 films for stretching tests

The as-prepared Fe2O3 textured films (initial Fe/C ratio in MGI is 1/3) were infiltrated with

uncured PDMS mixture, the elastomer and curing agent were mixed at 10:1 ratio. After

degassing, the PDMS was cured at 80 °C for 2 hours. The textured Fe2O3 films were fixed

by PDMS and can be used for stretching test.

Characterization

The surface morphologies of the GO, Fe(III)-based MGI, metal oxide replicas (Fe2O3,

Co3O4, Ag, NiO, TiO2 and CoxFe(3-x)O4 (x=1,2)) were investigated using a field emission

scanning electron microscope (SEM) (LEO 1530 VP) operating at 20.0 kV for low-,

medium- and high-resolution imaging. Before the SEM imaging, all samples were coated

with a layer of AuPd (<1 nm). GO and Fe(III)-GO nanosheets were drop casted on Si

substrates and studied with SEM operating at 3.5 kV. Surface morphology and thickness

32

of GO and Fe(III)-GO nanosheets are also characterized by AFM (Asylum MFP-3D Origin)

operating in alternating contact mode. Chemical structures of GO and Fe(III)-GO films

were studied with a JASCO FT/IR-4100 Fourier Transform Infrared Spectroscopy (FTIR)

with the ATR accessory. Transmission electron microscopy (TEM) and selected area

electron diffraction (SAED) were performed using a JEOL 2100F TEM/STEM at an

acceleration voltage of 200 kV, equipped with an energy dispersive X-ray spectrometer

(EDS) for elemental analysis. All samples are suspended in ethanol for 30 mins of

sonication, then dropped on lacey carbon grids for observation. Thermogravimetric

analysis (TGA) is carried out on a METTLER TOLEDO TGA/DSC 1 STARe system, with

heating rate 10 C min-1, air rate 80 ml min-1, the sample masses are ~1 mg. The

compositions and phases of and as-prepared metal oxide products were identified by X-ray

diffraction spectrometry (XRD) on a Bruker AXS D8 Advance instrument with Cu Ka

radiation (λ = 1.5418 Å). The time-resolved XRD of suspension-based samples were

carried out by cyclic scans every 5 mins. Crystalline peaks for the Si substrate were

removed. Zeta-potential of MGI were evaluated on a dynamic light scattering (DLS)

technique, all Mn+-GO colloids were prepared in 20mM NaNO3 aqueous solution to

minimize the effect of different ionic strength on zeta potential. Stretching of PDMS-fixed

textured Fe2O3 films was done by uni/biaxial stretching on a paper-covered Cu substrate.

Photographs were taken by an EOS digital SLR and compact system camera Canon EOS

100D.

2.5 Acknowledgements

33

This work was supported by the US National Science Foundation INSPIRE Track 1 grant:

CBET-1344097. We acknowledge Dr. Ruben Spitz for the synthesis of graphene oxide.

2.6 References


183-191.

2. Halim, J.; Kota, S.; Lukatskaya, M. R.; Naguib, M.; Zhao, M.-Q.; Moon, E. J.;

Pitock, J.; Nanda, J.; May, S. J.; Gogotsi, Y.; Barsoum, M. W., Synthesis and

Characterization of 2D Molybdenum Carbide (MXene). Adv. Funct. Mater. 2016, 26 (18),

3118-3127.

3. Gao, J.; Li, L.; Tan, J.; Sun, H.; Li, B.; Idrobo, J. C.; Singh, C. V.; Lu, T.-M.;

Koratkar, N., Vertically Oriented Arrays of ReS2 Nanosheets for Electrochemical Energy

Storage and Electrocatalysis. Nano Lett. 2016, 16 (6), 3780-3787.

4. Boston, R.; Bell, A.; Ting, V. P.; Rhead, A. T.; Nakayama, T.; Faul, C. F. J.; Hall,

S. R., Graphene Oxide as a Template for a Complex Functional Oxide. CrystEngComm

2015, 17 (32), 6094-6097.

5. Li, Z.; Wu, S.; Lv, W.; Shao, J.-J.; Kang, F.; Yang, Q.-H., Graphene Emerges as a

Versatile Template for Materials Preparation. Small 2016, 12 (20), 2674-2688.

6. Gao, X.; Mazloumi, M.; Cheung, L.; Tang, X., Graphene Oxide Film as a Template

for the Creation of Three-Dimensional Lamellar Metal Oxides and Reduced Graphene

Oxide/Metal Oxide Hybrids. MRS Commun. 2014, 4 (4), 171-175.

34




10879.

8. Luo, Y.; Xu, X.; Zhang, Y.; Chen, C.-Y.; Zhou, L.; Yan, M.; Wei, Q.; Tian, X.;

Mai, L., Graphene Oxide Templated Growth and Superior Lithium Storage Performance

of Novel Hierarchical Co2V2O7 Nanosheets. ACS Appl. Mater. Interfaces 2016, 8 (4),

2812-2818.

9. Peng, L.; Xiong, P.; Ma, L.; Yuan, Y.; Zhu, Y.; Chen, D.; Luo, X.; Lu, J.; Amine,

K.; Yu, G., Holey Two-Dimensional Transition Metal Oxide Nanosheets for Efficient

Energy Storage. Nat. Commun. 2017, 8, 15139.

10. Garcia-Gallastegui, A.; Iruretagoyena, D.; Gouvea, V.; Mokhtar, M.; Asiri, A. M.;

Basahel, S. N.; Al-Thabaiti, S. A.; Alyoubi, A. O.; Chadwick, D.; Shaffer, M. S. P.,

Graphene Oxide as Support for Layered Double Hydroxides: Enhancing the CO2

Adsorption Capacity. Chem. Mater. 2012, 24 (23), 4531-4539.

11. Wang, X.; Tian, W.; Liu, D.; Zhi, C.; Bando, Y.; Golberg, D., Unusual Formation

of α-Fe2O3 Hexagonal Nanoplatelets in N-doped Sandwiched Graphene Chamber for

High-Performance Lithium-Ions Batteries. Nano Energy 2013, 2 (2), 257-267.

12. Huang, Z.; Zhou, A.; Wu, J.; Chen, Y.; Lan, X.; Bai, H.; Li, L., Bottom-Up

Preparation of Ultrathin 2D Aluminum Oxide Nanosheets by Duplicating Graphene Oxide.

Adv. Mater. 2016, 28 (8), 1703-1708.

13. Jong Bo, P.; Yong-Jin, K.; Seong-Min, K.; Je Min, Y.; Youngsoo, K.; Roman, G.;

Barbolina, I. I.; Sang Jin, K.; Sangmin, K.; Myung-Han, Y.; Sung-Pyo, C.; Konstantin, S.

35

N.; Byung Hee, H., Non-Destructive Electron Microscopy Imaging and Analysis of

Biological Samples with Graphene Coating. 2D Mater. 2016, 3 (4), 045004.

14. Sitko, R.; Turek, E.; Zawisza, B.; Malicka, E.; Talik, E.; Heimann, J.; Gagor, A.;

Feist, B.; Wrzalik, R., Adsorption of Divalent Metal Ions from Aqueous Solutions using

Graphene Oxide. Dalton Trans. 2013, 42 (16), 5682-5689.

15. Liu, R.; Zhu, X.; Chen, B., A New Insight of Graphene oxide-Fe(III) Complex

Photochemical Behaviors under Visible Light Irradiation. Sci. Rep. 2017, 7, 40711.

16. Kielland, J., Individual Activity Coefficients of Ions in Aqueous Solutions. J. Am.

Chem. Soc. 1937, 59 (9), 1675-1678.

17. Sillén, L. G.; Martell, A. E.; Bjerrum, J., Stability Constants of Metal-Ion

Complexes. Chemical Society: London, 1964.

18. Perrin, D. D.; Sillén, L. G., Stability Constants of Metal-Ion Complexes, Part B :

Organic Ligands. Pergamon Press: Oxford; New York, 1979.

19. Konkena, B.; Vasudevan, S., Understanding Aqueous Dispersibility of Graphene

Oxide and Reduced Graphene Oxide through pKa Measurements. J. Phys. Chem. Lett.

2012, 3 (7), 867-872.

20. Jimbo, T.; Higa, M.; Minoura, N.; Tanioka, A., Surface Characterization of

Poly(acrylonitrile) Membranes Graft-Polymerized with Ionic Monomers As Revealed by

ζ Potential Measurement. Macromolecules 1998, 31 (4), 1277-1284.

21. Peng, D.; Beysen, S.; Li, Q.; Sun, Y.; Yang, L., Hydrothermal Synthesis of

Monodisperse α-Fe2O3 Hexagonal Platelets. Particuology 2010, 8 (4), 386-389.

36

22. Zhu, X.; Zhu, Y.; Murali, S.; Stoller, M. D.; Ruoff, R. S., Nanostructured Reduced

Graphene Oxide/Fe2O3 Composite As a High-Performance Anode Material for Lithium

Ion Batteries. ACS Nano 2011, 5 (4), 3333-3338.

23. Liu, Y.; Liu, S.; He, D.; Li, N.; Ji, Y.; Zheng, Z.; Luo, F.; Liu, S.; Shi, Z.; Hu, C.,

Crystal Facets Make a Profound Difference in Polyoxometalate-Containing Metal–Organic

Frameworks as Catalysts for Biodiesel Production. J. Am. Chem. Soc. 2015, 137 (39),

12697-12703.

24. Xiao, X.; He, C.-T.; Zhao, S.; Li, J.; Lin, W.; Yuan, Z.; Zhang, Q.; Wang, S.; Dai,

L.; Yu, D., A General Approach to Cobalt-Based Homobimetallic Phosphide Ultrathin

Nanosheets for Highly Efficient Oxygen Evolution in Alkaline Media. Energy Environ.

Sci. 2017, 10 (4), 893-899.

25. Guo, Y.; Guo, W., Electronic and Field Emission Properties of Wrinkled Graphene.

J. Phys. Chem. C 2013, 117 (1), 692-696.

26. Naseri, N.; Solaymani, S.; Ghaderi, A.; Bramowicz, M.; Kulesza, S.; Talu, S.;

Pourreza, M.; Ghasemi, S., Microstructure, Morphology and Electrochemical Properties of

Co Nanoflake Water Oxidation Electrocatalyst at Micro- and Nanoscale. RSC Adv. 2017,

7 (21), 12923-12930.

27. Wang, Z.; Tonderys, D.; Leggett, S. E.; Williams, E. K.; Kiani, M. T.; Steinberg,

R. S.; Qiu, Y.; Wong, I. Y.; Hurt, R. H., Wrinkled, Wavelength-Tunable Graphene-Based

Surface Topographies for Directing Cell Alignment and Morphology. Carbon 2016, 97,

14-24.

37

28. Dizaj, S. M.; Lotfipour, F.; Barzegar-Jalali, M.; Zarrintan, M. H.; Adibkia, K.,

Antimicrobial Activity of the Metals and Metal Oxide Nanoparticles. Mater. Sci. Eng. C

2014, 44, 278-284.

29. Qiu, Y.; Moore, S.; Hurt, R.; Külaots, I., Influence of External Heating Rate on the

Structure and Porosity of Thermally Exfoliated Graphite Oxide. Carbon 2017, 111

(Supplement C), 651-657.

38

Chapter 3 Stretching, Bending and Magnetic Properties of Cobalt Ferrite Foldable

Films

3.1 Introduction

It is critical to develop next generation devices that adapt versatile environments and

conditions. To better assist correlative human activities, foldable and wearable design is

gaining rapidly growing popularity. Flexibility and stretchability are becoming vital factors

in future manufacturing market, including batteries, electronic displays, human-machine

interfaces and biomedical devices.1-3 Recent progress in achieving such devices mainly

falls into two approaches – compositional design and structural configuration.

Compositional design focus on replacing rigid components with flexible alternatives, such

as conductive polymer, carbon nanotubes and graphene.4-6 On the other hand, hard

components can also be structurally manipulated to adapt deformation by utilizing

thickness scale-down (Ag nanowires, nanoribbons), wavy structures (Si buckled film) and

open mesh geometries (patterned array of metallic interconnects).2, 7 However, many of

functional materials cannot be simply replaced or shaped due to extreme stiffness and

brittleness. Among those, ceramics are considered particularly challenging.

Ceramic materials such as magnetic nanoparticle (CoFe2O4), semiconductor (In2O3/ZnO,

SiC), insulator (Si3N4) and catalyst (Al2O3, Fe2O3) are essential in emerging devices in

terms of their outstanding magnetic, optical and electrical properties.3, 8 However, ceramics

often fracture at low strain of 0.1-0.2%, which makes it difficult to be stretched or

compressed.9-10 Therefore, ceramic components for engineered devices are mostly micro-

sized, soft-encapsulated and then integrated by stretchable interconnects which undertake

39

primary stress during deformation.11-13 It results in low materials loading, unease of

processing and potential interfacial failure.14-15

In order to enhance stability and flexibility of a whole system, improvements of ceramic

components are necessary. To achieve that, our group proposed a structural design of metal

oxide foldable film where metal oxide nanoplatelets can be densely assembled into a nano-

thick film with highly wrinkled textures. Thin film structure and prefabricated folds allow

metal oxides to be unfolded and refolded repeatably. This method is based on our previous

studies on hierarchical metal oxide topographies templated by GO nanosheets.16 By

colloidal engineering, metalized GO nanosheets can maintain colloidally stable – such as

Fe(III)-GO suspension – which facilitates a uniform stacking of 2D nanosheets onto

various surfaces. Within layered GO coating, metal ions are confined tightly inside 2D

galleries, then assembled into freestanding metal oxide replicas during calcination.

Versatile topographies can be obtained by harnessing surface instability of GO nanosheets,

including wrinkles, crumples and complex architectures.17 Metal loading and composition

can be tuned as stated previously.16 In this study, we picked cobalt ferrite (CoFe2O4) – one

of ferrimagnetic ceramics – as an example and prepared cobalt ferrite foldable film

(CoFeFF). It can be mechanically stretched and bended while maintaining stable magnetic

functionality.


Figure 6.16a in the Appendix to Chapter 3 shows the fabrication route of CoFeFF using

our reported method.16 In brief, a GO-Fe(III)/Co(II) mixed suspension is drop cast onto

40

thermally responsive polystyrene substrate. By physical fixation on both sides of coated

substrate, the compression can be limited along one direction to create uniaxial wrinkles.

The wrinkled GO-Fe(III)/Co(II) coating is then calcinated under 600 C to remove

graphene template and converted into free standing CoFeFF. As illustrated in Figure 6.16b,

the encapsulated metal ions are confined in 2D nanochannels and arranged along

continuous topographies, which subsequently achieves high quality replication of original

graphene textures. GO nanosheets are used as 2D vessels and sacrificial templates, but do

not exist in final products. The X-ray diffraction spectrometry (XRD) of CoFeFF shows

single phase of Co2FeO4 (Figure 6.17, Appendix to Chapter 3). Thermogravimetric

analysis (TGA) experiments (Figure 6.18, Appendix to Chapter 3) were also conducted to

illustrate the Co2FeO4 formation during calcination. Detailed morphologies of CoFeFFs

are shown in Figure 3.1. The free standing CoFeFFs exhibit dark brown color of Co2FeO4

(Figure 3.1a). Viewed at the microscale, CoFeFFs replicate the characteristic GO wrinkled

texture – continuous, out-of-plane undulations – with a sinuous wavelength of ~5 m

(Figure 3.1 and 6.19 in the Appendix to Chapter 3). Figure 3.1b and 6.20 in the Appendix

to Chapter 3 show the cross-section of CoFeFF, which contains ~3 layers with a single

layer thickness ~30 nm. The multiple-layered structures of CoFeFF are adopted from

graphene template. The surface of CoFeFF is constructed of tessellated nanoplatelets which

are ~150 nm in size, arranged in mathematically tessellation (Figure 3.1c). Unlike isotropic

nanoparticles, these nanoplatelets exhibit lamellar structure and smooth surface as results

of 2D crystal growth. From Figure 6.21, Appendix of Chapter 3 we can see the primarily

crystal growth direction of Co2FeO4 nanoplatelets is [011].

41

Figure 3. 1. Morphologies of CoFeFFs. a. Photo of a free-standing CoFeFF. Scale bar, 0.4 cm. b.

Side view of CoFeFF, exhibiting layered structure with single layer thickness ~30 nm. Scale bar,

200 nm. c. Top view of CoFeFF, with multiple layers exposed on edge (white arrows). Each layer

is constructed of tessellated nanoplatelets with lateral size ~150 nm. Scale bar, 200 nm. The primary

crystal orientation of nanoplatelets is [011].

The ultra-high stretchability of crumpled GO film was demonstrated in our previous

work.18 GO films were attached on a stretchable substrate and expandable to 1500% areal

strain. Similarly, CoFeFF s were attached on top of polydimethylsiloxane (PDMS) polymer

substrate and slowly stretched to strains of 50, 100, 150 and 200%. The stretch behaviors

and corresponding surface morphologies of CoFeFFs are shown in Figure 3.2a.

Surprisingly, the wavelengths of wrinkles in samples increased from 5 to 10, 15, 25 and 35

μm during stretching, corresponding to 100%, 200%, 400% and 600% as the real extension

ratios. The observed strains of CoFeFF s are greater than applied might be caused by

uneven stress distribution (Figure 3.2b). However, no obvious crack found at or below 100%

strain. Micro cracks initiated on the ridges at 150% strain, then propagated at 200%. To

test the magnetic behaviors, we measured original and stretched samples using a vibrating-

sample magnetometer (VSM). Magnetic performance of stretched samples is shown in

Figure 3.2c. At strains from 0 to 200%, the saturation magnetization (Ms) and remnant

42

magnetization (Mr) of CoFeFF along the x-axis (parallel to the wrinkle) are ~50 and 15

emu g-1, remaining stable under room temperature. After stretching, no weight loss is

measured. Therefore, CoFeFF shows high stretchability and stable magnetic behaviors.

Figure 3. 2. Stretch behavior and magnetic properties of CoFeFFs. a. Surface morphologies

of CoFeFF s at observed strains of 0-600%. Scale bar, 20 m (top row), 2 m (bottom row). b.

Schematic of stretch behavior of CoFeFF s. c. Magnetic properties of CoFeFFs at different strains.

To further investigate the flexibility of CoFeFFs, bend behaviors were studied. Similar

to stretching, all samples were placed on PDMS polymer substrate. Bent samples were

obtained by wrapping CoFeFF/PDMS around aluminum columns with radii of 10.0, 5.0

and 2.5 mm. Surface morphologies of CoFeFF with different bend radii were shown in

Figure 3.3a. Bending into different curves, CoFeFF shown insignificant change on surface

topographies. No prominent crack was observed during bending. As shown in Figure 3.3b,

bending of wrinkled films results in less configurational deformation which helps preserve

43

the structural integrity. The magnetic behaviors of bent CoFeFFs remain relatively stable

(Figure 3.3c). The Ms of CoFeFF is ~30 emu g-1 during bend process, where the difference

on bend radii is insignificant.

Figure 3. 3. Bend behavior and magnetic properties of CoFeFF. a. Surface morphologies of

CoFeFF under bend radii (none, 10.0, 5.0 and 2.5 mm). Scale bar, 20 m (top row), 2 m (bottom

row). b. Schematic of bend process of CoFeFF. c. Magnetic properties of CoFeFF at different bend

radii.

To better illustrate the magnetic behaviors of CoFeFFs, temperature dependence of

magnetization was studied (Figure 3.4a-b). Increasing temperature from 200 to 380 K, Ms

slightly decreases from 52 to 44 emu/g, which is consistent with previous report that the

Curie temperature Tc ~ 800K in Co2FeO4.19 However, the coercive field Hc is strongly

dependent on temperature. From 200 to 380 K, Hc decreased by 10 folds from 0.2680 to

44

0.0265 T. As shown in Figure 3.4a, both coercivity and remnant magnetization decreases

with increasing temperature, indicates the decreasing anisotropy along the x-axis with

increasing temperature due to the thermal energy. Moreover, the hysteresis loop along three

directions were measured, including x-axis (parallel to the wrinkle), y-axis (perpendicular

to the wrinkle) and z-axis (perpendicular to the film). In Figure 3.4c, MvsH curves along

three axis shown similar Ms. However, Mr measured along the x-axis is 20% larger than

that of y- and z-axis. In general, random distribution of the magnetocrystalline anisotropy

in each nanoplatelet leads to similar magnetization. The wavy topographies of wrinkled

films provide a smooth surface along x- comparing to y- and z-axis, which allows a smooth

and compact arrangement of nanoplatelets. Finally, a mechanical fatigue test was

conducted by stretching a CoFeFF to 150% strain over 100 times. Comparison of

magnetization is shown in Figure 3.4d, the Ms decreased to 73% of original after fatigue.

Figure 3. 4. Temperature dependence, anisotropy and mechanical fatigue on magnetic

behaviors of CoFeFF. a. The hysteresis loop of CoFeFF under different temperature, ranging

from 200 to 380 K. b. Temperature dependence on saturation magnetization of CoFeFF. c.

45

Magnetization of CoFeFF along x, y and z directions. d. Magnetization of CoFeFF before and after

stretching to 150% strain for 100 times.

3.3 Conclusions

In summary, we proposed and demonstrated a foldable CoFe2O4 film by in-situ

fabrication within graphene sacrificial templates. Resulting CoFeFFs achieved enhanced

stretchability (200% strain) and flexibility (2.5 mm bend radius) while maintaining stable

magnetic performance. This new technology is easy and efficient to obtain pre-folded,

flexible metal oxides thus can be extended to other hard materials. Furthermore, our work

is of interests in structural-functional ceramics, stretchable devices and other applications

for structural manipulation of intrinsically brittle materials.


Materials

Ethanol, iron(III) nitrate nonahydrate (Fe(NO3)3·9H2O), cobalt(II) nitrate hexahydrate

(Co(NO3)2·6H2O), anhydrous acetone and methylene chloride were purchased from

Sigma-Aldrich. GO was synthesized by a modified Hummer’s method.20 Thermally

responsive polyethylene heat shrink films were bought from Grafix. Polydimethylsiloxane

(PDMS) was made from a SYLGARD 184 silicone elastomer kit. All water was deionized

(18.2 MΩ, milli-Q pore). All reagents were used as received without further purification.

Fabrication of GO wrinkled film

GO nanosheets were prepared by a modified Hummers’ method, with lateral size ~1 μm,

thickness ~1 nm, and characterized in Figure 6.22 in the Appendix to Chapter 3. The GO

46

suspensions used in the colloidal and film formation experiments is 1.0 mg mL-1, with a

C/O atomic ratio of ~2.1. Potential GO impurities –such as N, S, Mn, K, Cl, and P – were

not detected by XPS.20 The polystyrene shrink film was cut into 1.54.0 cm2 rectangles

and washed with ethanol. Once dry, samples were treated with air plasma in a Deiner Atto

standard plasma system with a borosilicate glass chamber and a 13.56 MHz, 0-50 W

generator. The chamber pressure was pumped down to and maintained at 0.13 mbar while

being flushed with air for 5 min. Plasma was then generated at 100% power (50 W) for 15

min followed by slow venting of the chamber. 300 L GO 1.0 mg mL-1 suspension was

drop cast onto 1.54.0 cm2 polystyrene substrate. After 60 C oven drying, two sides of

polystyrene substrate were clapped. The GO coated polystyrene substrate was then put in

130 C oven for 6 min to achieve uniaxial contraction.

Fabrication of CoFeFF

1.0 mg mL-1 GO aqueous suspension was well mixed with 4.0 mM Fe(NO3)3 and 2.0 mM

Co(NO3)2 aqueous solutions. 300 L of the mixture was drop cast onto 1.54.0 cm2 plasma

treated polystyrene substrate. After 60 C oven drying, two sides of polystyrene substrate

were clapped. The GO coated polystyrene substrate was then put in 130 C oven for 6 min

to achieve uniaxial contraction. Then samples was immersed in methylene chloride to fully

dissolved polystyrene substrate and washed with acetone. Free-standing GO-Fe(III)/Co(II)

wrinkled films were carefully collected. Dried GO-Fe(III)/Co(II) wrinkled films were

calcinated in oven under 600 C for 2 hrs to obtain CoFeFFs.

Preparation of CoFeFFs attached on PDMS

PDMS was made from a SYLGARD 184 silicone elastomer kit. It is comprised of

base/curing agent to be mixed in a 10 (base) :1 (curing agent) ratio by weight for manual

47

mixing. 1.8 g base/curing agent mixture was pour in an aluminum dish under vacuum for

15 minutes, then put in a 60 C oven for 14 minutes. When the PDMS was partially cured,

a CoFeFF was gently placed on the surface of PDMS. The mixture was put back into 60

C oven for another 20 minutes to fix CoFeFF.

Stretch/bend tests of PDMS fixed CoFeFFs

For stretch tests, PDMS/CoFeFFs were stretched along folding direction (perpendicular to

wrinkles) to desired strains and clamped on a metal plate with calibration lines. Then the

plate was put into 130 C oven to completely cure PDMS polymer. Once cured, obtained

strains were well preserved and would not change. For bend test, PDMS/CoFeFFs were

wrapped around aluminum columns with desired radii, then put into 130 C oven to

completely cure PDMS polymer. Once cured, obtained deformation were well preserved

and would not change.

Characterization

The surface morphologies of GO and CoFeFFs were investigated using a field emission

SEM (LEO 1530 VP) operating at 10.0 kV for imaging. Before the SEM imaging, all

samples were coated with a layer of AuPd (<1 nm). Surface morphology and thickness of

GO nanosheets were also characterized by AFM (Asylum MFP-3D Origin) operating in

alternating contact mode. TEM and SAED were performed using a JEOL 2100F

TEM/STEM at an acceleration voltage of 200 kV, equipped with an energy dispersive X-

ray spectrometer for elemental analysis. All samples were suspended in ethanol for 30 min

of sonication, then dropped on lacey carbon grids for observation. TGA were carried out

on a METTLER TOLEDO TGA/DSC 1 STARe system, with heating rate of 10 °C min−1,

air rate 80 mL min−1, the sample masses were ~1 mg. The compositions and phases of as-

48

prepared metal oxide products were identified by XRD on a Bruker D8 Discovery 2D X-

ray Diffractometer with Cu Kα radiation (λ = 1.5418 Å). Magnetic measurements were

carried out using a physical property measurement system (PPMS) from Quantum

Design®.

3.5 References

1. Xu, S.; Zhang, Y.; Cho, J.; Lee, J.; Huang, X.; Jia, L.; Fan, J. A.; Su, Y.; Su, J.;

Zhang, H.; Cheng, H.; Lu, B.; Yu, C.; Chuang, C.; Kim, T.-i.; Song, T.; Shigeta, K.; Kang,

S.; Dagdeviren, C.; Petrov, I.; Braun, P. V.; Huang, Y.; Paik, U.; Rogers, J. A., Stretchable

Batteries with Self-Similar Serpentine Interconnects and Integrated Wireless Recharging

Systems. Nat. Commun. 2013, 4 (1), 1543.

2. Koo, J. H.; Kim, D. C.; Shim, H. J.; Kim, T.-H.; Kim, D.-H., Flexible and

Stretchable Smart Display: Materials, Fabrication, Device Design, and System Integration.

Adv. Funct. Mater. 2018, 28 (35), 1801834.

3. Jayathilaka, W. A. D. M.; Qi, K.; Qin, Y.; Chinnappan, A.; Serrano-García, W.;

Baskar, C.; Wang, H.; He, J.; Cui, S.; Thomas, S. W.; Ramakrishna, S., Significance of

Nanomaterials in Wearables: A Review on Wearable Actuators and Sensors. Adv. Mater.

2019, 31 (7), 1805921.

4. Mun, J.; Kang, J.; Zheng, Y.; Luo, S.; Wu, H.-C.; Matsuhisa, N.; Xu, J.; Wang, G.-

J. N.; Yun, Y.; Xue, G.; Tok, J. B. H.; Bao, Z., Conjugated Carbon Cyclic Nanorings as

Additives for Intrinsically Stretchable Semiconducting Polymers. Adv. Mater. 2019, 31

(42), 1903912.

49

5. Kim, T.; Cho, M.; Yu, K. J., Flexible and Stretchable Bio-Integrated Electronics

Based on Carbon Nanotube and Graphene. Materials 2018, 11 (7), 1163.

6. Jang, H.; Park, Y. J.; Chen, X.; Das, T.; Kim, M.-S.; Ahn, J.-H., Graphene-Based

Flexible and Stretchable Electronics. Adv. Mater. 2016, 28 (22), 4184-4202.

7. Kim, D.-H.; Rogers, J. A., Stretchable Electronics: Materials Strategies and

Devices. Adv. Mater. 2008, 20 (24), 4887-4892.

8. Rogers, J. A.; Someya, T.; Huang, Y., Materials and Mechanics for Stretchable

Electronics. Science 2010, 327 (5973), 1603.

9. Green, D. J., An Introduction to the Mechanical Properties of Ceramics. Cambridge

University Press: Cambridge, 1998.

10. Seshadri, S. G.; Chila, K. Y., Tensile Testing of Ceramics. J. Am. Ceram. Soc. 1987,

70 (10), C‐242-C‐244.

11. Sim, K.; Rao, Z.; Zou, Z.; Ershad, F.; Lei, J.; Thukral, A.; Chen, J.; Huang, Q.-A.;

Xiao, J.; Yu, C., Metal Oxide Semiconductor Nanomembrane–Based Soft Unnoticeable

Multifunctional Electronics for Wearable Human-Machine Interfaces. Sci. Adv. 2019, 5 (8),

eaav9653.

12. Koshi, T.; Iwase, E. In Stretchable Electronic Device with Repeat Self-Healing

Ability of Metal Wire, 2017 IEEE 30th International Conference on Micro Electro

Mechanical Systems (MEMS), 22-26 Jan. 2017; 2017; pp 262-265.

13. Brand, J. v. d.; Kok, M. d.; Sridhar, A.; Cauwe, M.; Verplancke, R.; Bossuyt, F.;

Baets, J. d.; Vanfleteren, J. In Flexible and Stretchable Electronics for Wearable

Healthcare, 2014 44th European Solid State Device Research Conference (ESSDERC),

22-26 Sept. 2014; 2014; pp 206-209.

50

14. Iwata, Y.; Iwase, E. In Stress-Free Stretchable Electronic Device using Folding

Deformation, 2017 IEEE 30th International Conference on Micro Electro Mechanical

Systems (MEMS), 22-26 Jan. 2017; 2017; pp 231-234.

15. Lee, S.; Song, Y.; Ko, Y.; Ko, Y.; Ko, J.; Kwon, C. H.; Huh, J.; Kim, S.-W.; Yeom,

B.; Cho, J., A Metal-Like Conductive Elastomer with a Hierarchical Wrinkled Structure.

Adv. Mater. 2020, 32 (7), 1906460.








ACS Nano 2018, 12 (1), 234-244.

19. Mathew, D. S.; Juang, R.-S., An Overview of the Structure and Magnetism of

Spinel Ferrite Nanoparticles and Their Synthesis in Microemulsions. Chem. Eng. 2007,

129 (1), 51-65.


Structure and Porosity of Thermally Exfoliated Graphite Oxide. Carbon 2017, 111, 651-

657.

51

Chapter 4 Controlling Nanochannel Orientation, Length, and Width in

Graphene-Based Nanofluidic Membranes

4.1 Introduction

2D sheet-like materials through molecular intercalation or pillaring can assemble to create

well-defined slit-shaped angstrom-scale interlayer channels that may be exploited in

emerging nanofluidic technologies, including molecule and ionic separations, hydraulic-

electric energy conversion and supercapacitor.1-5 Among the many materials with intrinsic

lamellar structures, GO has received the most attention, primarily as a selective membrane

with sub-nanometer channels formed by the pillaring action of oxygen-functional groups

on the nanosheet faces.6-8 GO van der Waals (vdW) films with interlayer spacings ~8

angstroms are capable of precise molecular sieving9 and can be readily fabricated over

large areas by a variety of water-based coating or printing processes.1, 10

Conventional GO filtration membranes consist of stacked micron-scale nanosheets that

align horizontally during evaporation/filtration assembly due to self-exclusion and

substrate templating, which in turn aligns the associated nanochannels perpendicular to the

desired flow direction through the membrane. The resulting fluid transport pathways are

highly torturous with total lengths scaling as L ~ tmembrane (L/d)nanosheet, where tmembrane is the

nominal membrane thickness and (L/d)nanosheet is the aspect ratio of the constituent

nanosheets.11-13 Typically monolayer nanosheets have very high (L/d)nanosheet values,

making permeate path lengths orders of magnitude greater than the membrane thickness,

and providing a serious throughput limitation to liquid phase applications such as

ultrafiltration and desalination.11-12, 14 Another important limitation of conventional GO

membranes is excess water absorption and swelling,7 which can increase interlayer spacing

52

from ~0.8 nm to as much as 6 nm, resulting in loss of molecular selectivity and mechanical

stability.15

One approach to overcome the flux limitation is to realign the nanosheets to create

vertical (Z-directional) transmembrane nanochannels with path-lengths ~ tmembrane. Several

methods have been demonstrated to create vertical graphene structures for applications

such as field emitters,16 capacitors,17 or edge-rich antibacterial surfaces.18 Copper-assisted

chemical vapor deposition has been used to directly grow vertical graphene,19 and GO

nanosheets have been vertically aligned through ice-growth-directed assembly20 and

magnetic field alignment.18 These methods successfully create vertical graphene

architectures, but not fully-dense, pore-free membranes in which transport occurs

exclusively through 2D interlayer nanochannels, which is a requirement for molecular

selectivity.

This work introduces a self-assembly method to create Vertically Aligned Graphene

Membranes (VAGMEs). The fabrication concept originates from recent studies of textural

control in graphene films, which exploit surface instability of planar graphene thin films

on soft substrates under compression to produce wrinkled, crumpled, and mixed-mode

surface deformation patterns.21-24 GO films are cast onto pre-stretched polystyrene

substrates (Figure 4.1a) below their glass transition temperature, then heated to release the

pre-strain and induce 2D isotropic crumpling in the stiff graphene coating. By physical

fixation of the substrate on two of the four sides, the compression can be limited to one

direction, creating unidirectional periodic wrinkle patterns(Figure 4.1b).21-22 The

individual wrinkles do not have smooth sinusoidal geometries, but rather nearly flat side

walls and sharp (high-curvature) ridge tips.21, 25 An ideal triangular zig-zag structure (see

53

Figure 4.1b) has a total film length in cross-section that is longer than the base substrate

length by a factor of 1/cos, where is the rise angle from the substrate plane (Figure 4.1

part i). This angle is directly related to the substrate relaxation ratio, (L/Linitial) by: = cos-

1(L/Linitial) and the transport path lengths across the membrane become t/sin, where t is

the length of each segment (Figure 4.1 part i). At high extents of compression (low L/Linitial

values) the side walls are forced to tilt at high rise angles to the substrate plane and thus

adopt a strong vertical component that forms the basis for the vertical nanochannel arrays.

Imbedding these zig-zag graphene structures in epoxy, with careful control of bubble

formation, creates fully dense structures that can be microtomed into thin membranes to

produce VAGME devices (Figure 4.1 part ii) whose structure and membrane performance

are described below.


Fabrication and morphology of vertically aligned Zr-GO membranes. Figure 4.1

shows the detailed VAGME fabrication process, which begins with conventional drying-

induced assembly of planar GO nanosheet vdW films on substrates to produce dense,

uniform arrays of nanochannels (Figure 4.1a). The nanochannels are then partially

reoriented by choosing pre-stretched polystyrene for the substrate, which undergoes

thermally activated shrinkage to compress the GO film and create 1D wrinkles (Figure

4.1b). We observed that neat GO films deform irregularly to produce chaotic

microstructures seen in cross-section (Figure 6.23a in the Appendix of Chapter 4). It might

be caused by weak binding between negatively charged GO nanosheets and substrate (after

54

air plasma treatment) where detachment takes place during thermal contraction. We

hypothesized that (+/-) electrostatic attraction may increase the interfacial interaction of

the vdW films with polystyrene and conducted a metal ion-doping method. It was found

that Zr doping using the water soluble ZrOCl2 at C/Zr atomic ratio ~ 22/1 produced more

ordered wrinkle textures (Figure 6.23b and 6.24 in the Appendix of Chapter 4), with a

interlayer spacing ~8.8 angstrom. Structural differences and colloidal stability of metal ion-

doped GO are discussed in Appendix of Chapter 4.

Figure 4. 1. Schematic and fabrication of vertically aligned Zr-GO/epoxy membranes. i.

Schematic of wrinkled films. Side view of planar and 1D wrinkled films. Figure illustrates that

wrinkling tilts the horizontal line into multiple near vertical line segments. ii. Fabrication of

vertically aligned Zr-GO/epoxy membranes. a, Drying-induced assembly of Zr-GO nanosheets

55

on pre-stretched polystyrene substrate (GO film thickness 1 μm). Inset: nanostructure of planar Zr-

GO films with horizontal alignment and tortuous flow pathways. Purple spheres show ZrO2+ cations

(unhydrated state for reference). Fully hydrated ZrO2+ diameter is ~1 nm, implying that ZrO2+ likely

exists in the interlayer spaces in a partially hydrated state complexed with O-containing groups on

GO. b, Wrinkled Zr-GO films are produced by thermally activated mechanical compression. c,

Wrinkled Zr-GO films are removed from the substrate and imbedded into epoxy resin. d, Multiple

cycles of microtome sectioning yields Zr-GO/epoxy composite membranes. e, Side view of

vertically aligned Zr-GO/epoxy membrane (VAGME) with the entrances to interlayer

nanochannels open at the top and bottom surface. This method transforms a single planar Zr-GO

film into hundreds of vertical film segments that each serve as one array of Z-directional

membrane-spanning nanochannels.

The wrinkled Zr-GO films are then removed from the relaxed substrate and imbedded

in an epoxy matrix (Figure 4.1c). Due to high viscosity and hydrophobicity, epoxy resin

cannot diffuse into GO gallery spaces14 (which would block the nanochannels) but

successfully impregnates the microstructural gaps around the wrinkle features and

mechanically stabilizes the structure. The wrinkled Zr-GO/epoxy composite is then

sectioned by microtome using heavy duty high profile stainless blades to remove the ridge

tips on the top and bottom of the zig-zag film, exposing nanochannel entrances and exits

(Figure 4.1d). A challenge was to develop a microtome sectioning technique that does not

introducing holes or cracks that would act as short circuits for molecular transport through

the membrane (see Figure 6.25 in the Appendix of Chapter 4 and associated discussion).

The final product – vertically aligned Zr-GO membrane (VAGME) – is shown in Figure

4.1e, where the original planar film has been realigned and sectioned into 100 per mm of

independent transmembrane film segments, each segment consisting of an array of GO

interlayer nanochannels.

Figure 4.2 presents detailed morphological characterization of the VAGME

microstructures at different stages of fabrication. Pre-stretched polystyrene rectangles of

11.0 3.0 mm2 thermally relaxes to 3.0 3.0 mm2. The unidirectional relaxation rates is

56

thus 3/11 = 0.272 corresponding to = 74 degree rise angles in the ideal zig-zag geometric

model (Figure 4.1 part i). Observed rise angles are in the range 60-80 degrees (Figure 4.2)

and the thickness of the films increases from 1 to about 100 μm. Figure 4.2a shows dark

brown-colored, wrinkled Zr-GO films imbedded in an epoxy cylinder, with is observed to

be free of pores or cracks under microscope observation. After thin sectioning, the

VAGME is a free-standing and mechanically robust film that reveals a pattern of parallel

dark strips in a transparent (epoxy) matrix. The top view of the wrinkled Zr-GO films

shows a corrugated texture with continuous 1D wrinkles of 3 mm in length, and ~20 μm in

wavelength (Figure 4.2b). Under SEM the VAGME shows a smooth surface with a pattern

of ~ 300 fine strips, each strip representing the terminus of a near-vertical Zr-GO film

segment of ~ 1 μm in width. The high magnification view of an individual Zr-GO strip

shows a tightly packed, multilayered structure, which appears identical to original planar

Zr-GO films. The side view of wrinkled Zr-GO films in matrix exhibits a zig-zag pattern

with a wrinkle amplitude ~50 μm (Figure 4.2c left). The final VAGME is 20 μm in

thickness after removing the ridge tops to access the GO nanochannels and convert the

original continues film to a set of near-vertical Zr-GO segments (Figure 4.2c right). After

deformation, the wrinkled Zr-GO is still a 2D film on the macroscale (L >> t), but at the

nanoscale the interlayer gallery spaces are preserved, and at the microscale the film

contains hundreds of zig-zag segments that give a vertical component to the nanochannel

orientation.

57

Figure 4. 2. Morphologies of wrinkled Zr-GO films and VAGME during fabrication. a,

Photographs of wrinkled Zr-GO films imbedded in epoxy matrix and, after thin sectioning, a free-

standing VAGME. Scale bar, 4 mm. b, Top views of wrinkled Zr-GO films and a VAGME.

Wrinkled Zr-GO films show corrugated textures. VAGME shows a strip pattern associated with

GO film segments whose edges intersect the top surface. Scale bar, 100 μm. High-magnification

view of one Zr-GO strip on the VAGME surface, showing a uniform, close-packed array of

nanosheet layers and interlayer nanochannels. Scale bar, 1 μm. c, Left, side view of wrinkled Zr-

GO/epoxy composite, exhibiting regular zig-zag patterns as desired. Right, side view of VAGME,

showing several individual Zr-GO vertical segments after ridge removal by microtome

(highlighted). Scale bar, 20 μm.

Selective molecular transport through VAGME nanochannels. A major challenge in

designing and fabricating vertically aligned nanochannel membranes is to ensure that

molecular transport occurs only through the nanochannels and cannot bypass the

nanochannels through membrane cracks or pores. Parallel transport through such

membrane defects would likely be rapid (due to their microscale dimensions) and non-

selective, thus negating the primary benefit of a nanochannel technology. Potential defects

in VAGMEs include trapped air bubbles in the epoxy, which become holes in thin sections,

58

cracks in the GO films, and delamination openings at the many epoxy-GO interfaces

occurring during curing or processing.

Microscopic inspection of our final VAGME membranes did not reveal visible flaws,

but we sought additional evidence that transport is limited to the GO interlayer channels.

A distinctive behavior of GO nanosheet films is their ability to pass water vapor rapidly,

while excluding all non-polar and most other polar molecules.3, 9, 26 We thus measured the

comparative rate of water vapor and hexane vapor permeation at the same partial pressure,

and for both VAGME films and epoxy-only controls.

Figure 4.3 shows the test apparatus27 which is heated to produce a range of vapor

pressures up to 1 bar at the normal boiling points of water (100 C) and hexane (68 C).

Magnetic stirring is introduced to reduce mass transfer resistance on the upstream side, and

the device was operated in a fume hood to provide convective flow over the VAGME

surface to reduce downstream mass transfer resistance. Water vapor transmission rates

through VAGME films show an approximately linear dependence on upstream H2O vapor

pressure and approach 10 mmol mm-1 hr-1 at 1 bar vapor pressure, while (pure) hexane

permeation through is at or below the detection limit (Figure 4.3b). Pure epoxy films (20

μm thick), used here as a negative control sample, show no measurable flux for either water

or hexane vapor (Figure 4.3c). The data clearly show that VAGME films exhibit the

characteristic high selectivity of GO nanochannels for water over hexane, and confirm the

absence of pores and interfacial cracks that would degrade this selectivity. Microscopic

inspection of the VAGME films after the vapor permeation experiments showed no cracks

or interfacial gaps, indicating thermal stability up to 100 C, unlike conventional GO films

which have a tendency to crack at temperatures above 40-60 °C.9, 27 Further, the width of

59

the nanochannel arrays that intersect the top VAGME surface were observed to be

unchanged by the permeation experiments, which suggests swelling-resistance , possibly

caused by the confining effect of the relatively rigid epoxy matrix.14

VAGMEs appear to be offer pure GO nanochannel transport in a platform that is

mechanically stable, thermally stable up to 100 C, and swelling resistant.

Figure 4. 3. Measurements of selective molecular transport through VAGME nanochannels.

a, Custom diffusion cell for temperature-dependent vapor permeation experiments. VAGME films

are adhered on the top of a copper sheet spanning a 4 mm diameter hole by epoxy adhesive, and

the plate is clamped and sealed on top of a stirred liquid/vapor cell. b, Measured vapor fluxes

through VAGME films for water and n-hexane. Flux values are normalized by the active Zr-GO

(nanochannel array) area. c, Control experiments for vapor permeation through epoxy-only

membranes of same (20 μm) thickness.

60

Table 4.1 gives comparisons of water vapor flux through VAGME vs. conventional GO

membranes. Data on conventional GO membrane are adopted from previous studies.9, 27

At 60 C, the water flux through VAGME is similar to that of conventional GO membranes

on a total area basis, even though only a fraction of the total membrane area is occupied by

nanochannel arrays in the current method. Renormalizing fluxes by active (GO) area

(determined by analysis of top surface images) shows a VAGME enhancement factor of ~

16, which might reflect the effect of tortuous pathways in conventional (horizontally

aligned) films. We tried further normalized the flux values by membrane thickness, which

differs significantly across the comparison cases. VAGME fluxes increase at higher

temperature and vapor pressure, but quantitative comparisons with conventional

membranes are not possible to their tendency to crack. On the contrary, VAGME maintains

functioning in 80 and 100 C water vapors during 12 hrs of testing and delivers enhanced

fluxes of 1.57103 and 3.49103 kg m-2 hr-1 μm respectively. Interestingly, our VAGME

fluxes at 100 C show a slow decline from ~ 4500 to 160 kg m-2 hr-1 μm after 24 hrs (Figure

6.26a in the Appendix of Chapter 4). We believe this is due to slow hydrothermal reduction

of GO which is known to reduce hydrophilicity and water permeability.28 XPS results

confirm the thermal reduction process (Figure 6.26-27 in the Appendix of Chapter 4). No

significant change in interlayer spacing was observed in XRD results due to the spacer

effect of intercalated metal ions.12 Detailed discussion is included in the Appendix of

Chapter 4.

61

Table 4. 1 Water vapor fluxes measured through VAGME devices and conventional

GO films

Membrane Flux (60 C)

kg m-2 hr-1

Flux (80 C)

kg m-2 hr-1

Flux (100 C)**

kg m-2 hr-1

Conventional

GO film 1.7-4.1*9, 27 (film cracking) (film cracking)

VAGME

(total area basis) 1.9 5.6 12.5

VAGME

(active area basis) 27.1 78.5 175

Normalized flux (60 C)

kg m-2 hr-1 μm


kg m-2 hr-1 μm


kg m-2 hr-1 μm

VAGME

(active area basis) 541 1.57103 3.49103

* The flux value from ref. 9 includes a calculated temperature correction based on vapor pressure

driving force. ** The 100 °C data is time-dependent and the table entries are average fluxes over 12 hrs.

4.3 Conclusions

In summary, the new field of 2D nanofluidics requires new methods to create well-

defined arrays of interlayer nanochannels with precise control of width, length, and

orientation. Here we applied recent techniques for compressive texturing of 2D thin films

to a new Zr-GO composite to direct the self-assembly of near vertically aligned

nanochannel arrays with uniform and controlled fluidic pathlengths. The self-assembled

structure can be effectively captured by epoxy imbedding and converted into mechanically

robust microscale membranes, where molecular transport is allowed only through Z-

directional, transmembrane GO interlayer channels of defined length. This approach

simultaneously addresses several well-known limitations in current GO nanofluidics

related to undesirable (horizontal) channel orientation, thermal instability, and

uncontrolled water swelling that degrades molecular selectivity. We anticipate future work

62

will focus on exploiting this approach to create devices for specific technological

applications, including stable selective molecular sieve membranes for liquid phase

separations such as ultrafiltration and reverse osmosis.


Materials

Ethanol, zirconyl chloride octahydrate (ZrOCl2·8H2O) and methylene chloride were

purchased from Sigma-Aldrich. Hexane was purchased from Fischer Scientific. Thermally

responsive polyethylene heat shrink films were purchased from Grafix, and Epofix from

Electron Microscopy Sciences. Copper sheets (0.3 mm thick) were purchased from

McMaster Carr. All water was deionized (18.2 MΩ, milli-Q pore). All reagents were used

as received without further purification.

Preparation of wrinkled Zr-GO films on polystyrene substrate.

GO nanosheets were prepared by a modified Hummers’ method, with lateral size ~1 μm,

thickness ~1 nm, with a C/O atomic ratio of ~ 2.1. Potential GO impurities N, S, Mn, K,

Cl, and P were not detected by XPS.29 The Zr-GO suspensions used in the colloidal and

film formation experiments contained 3.55 mg mL-1 GO and 8 mM ZrOCl2 aqueous

solution. The polymer shrink film was cut into 60.0 11.0 mm2 rectangles and washed

with ethanol. Once dry, samples were treated with air plasma in a Deiner Atto standard

plasma system with a borosilicate glass chamber and a 13.56 MHz, 0-50 W generator. The

chamber pressure was pumped down to and maintained at 0.13 mbar while being flushed

with air for 5 min. Plasma was then generated at 100% power (50 W) for 15 min followed

63

by slow venting of the chamber. Each polystyrene rectangle is masked with scotch tape

only to leave a 3.0 11.0 mm2 clean gap in the center, which is to limit subsequent drop

casting in selected area. Then 16.8 μL of Zr-GO suspension is drop cast on the gap and

form a 1 μm Zr-GO coating at 60 C. The protection tape was then removed and both

uncoated sides are clapped for 1D shrinking at 130 C for 9 min. Finally, 3.0 3.0 mm2

uniform wrinkled Zr-GO films are obtained in the center of the polystyrene substrate

without distortion.

Fabrication of wrinkled Zr-GO/epoxy composites and membrane sectioning.

5.00 g Epofix embedding resin 1232 R and 0.56 g Epofix hardener 1232 H were mixed in

an aluminum cup and under stirring for at least 3 min. Then the mixture was put into a

vacuum chamber to exhaust bubble for 20 min. ~20 μL mixture was dropped on the surface

of wrinkled Zr-GO films and followed by another 20 min of degassing, which facilitates

the impregnation of epoxy into micro gaps. The mixture was left in a desiccator for

hardening overnight. Once cured, wrinkled Zr-GO/epoxy composite was put upside down,

where the epoxy side was adhered to the surface of an epoxy pillar. Then the polystyrene

substrate was dissolved in methylene chloride for 15 min, please be aware that only the

polystyrene part was immersed. After removal of polystyrene substrate, another step of

Epofix impregnation was carried out on the exposed wrinkled Zr-GO films as mentioned

before (Figure 4.2a left). This procedure is to ensure Zr-GO films are always fixed on a

hard substrate during fabrication thus the structural integrity is maintained. Then, wrinkled

Zr-GO/epoxy composite was thin sectioned using an automated microtome, the cut

thickness is set to 20 μm. The active GO and total membrane areas are 0.9 and 12.6 mm2

respectively.

64

Vapor permeation experiments.

Water or n-hexane with a volume of 17 mL was pipetted into an open top glass vessel with

O-ring flange. Test membranes were adhered over a 4 mm diameter hole in a copper sheet

by Epofix, then hardened in a desiccator overnight. After curing, the copper plate was

clamped in a two-piece vessel and sealed with vacuum grease coated rubber rings. Then

the setup is put on a hot plate and heated to desired temperature under stirring at 200 rpm.

The temperature inside the vessel is calibrated according to different solvents before setting

up the hot plate. Water or n-hexane permeation was then measured as gravimetric loss after

4-6 hours and converted to average flux values.

Analysis of Zr-GO films exposed to 100 °C water vapor.

Planar Zr-GO films were prepared by drop casting 2 mL of Zr-GO suspension on a

hydrophobic Teflon substrate, followed by peeling to obtain a freestanding membrane. The

membrane is placed in a 130 C oven for 9 min to ensure consistent treatment with the

membranes subjected to the thermal compressive wrinkling process. The thermally treated

Zr-GO membranes were cut into multiple smaller pieces and placed above 1 bar boiling

water for various time. These treated Zr-GO films are then characterized by XRD and XPS.

Statistical Analysis.

All vapor tests were run in triplicates, and the standard errors for the analytical results were

used to generate and present error bars.

Characterization.

The surface morphologies of graphene structures were investigated using a field emission

scanning electron microscope (SEM) (LEO 1530 VP) operating at 10.0 kV for low-,

medium- and high-resolution imaging. Before the SEM imaging, all samples were coated

65

with a layer of AuPd (<1 nm). The phases of graphene films were identified by X-ray

diffraction spectrometry (XRD) on a Bruker AXS D8 Advance instrument with Cu Ka

radiation (λ = 1.5418 Å). Crystalline peaks for the Si substrate were removed. The chemical

compositions of Zr-GO films were identified by K-Alpha X-ray photoelectron

spectrometer (XPS). C1s peaks are fitted using XPSPEAK. Sectioning of Zr-GO/epoxy

composites is carried on a Leica RM2265 automated microtome, equipped with C.L.

Strukey, Inc. heavy duty high profile disposable microtome blades. The cut thickness is set

to 20 μm. Photographs were taken by an EOS digital SLR and compact system camera

Canon EOS 100D.

4.5 References

1. Wang, S.; Yang, L.; He, G.; Shi, B.; Li, Y.; Wu, H.; Zhang, R.; Nunes, S.; Jiang,

Z., Two-dimensional nanochannel membranes for molecular and ionic separations. Chem.

Soc. Rev. 2020, 49 (4), 1071-1089.

2. Kim, H. W.; Yoon, H. W.; Yoon, S.-M.; Yoo, B. M.; Ahn, B. K.; Cho, Y. H.; Shin,

H. J.; Yang, H.; Paik, U.; Kwon, S.; Choi, J.-Y.; Park, H. B., Selective gas transport through

few-layered graphene and graphene oxide membranes. Science 2013, 342 (6154), 91-95.

3. Joshi, R. K.; Carbone, P.; Wang, F. C.; Kravets, V. G.; Su, Y.; Grigorieva, I. V.;

Wu, H. A.; Geim, A. K.; Nair, R. R., Precise and ultrafast molecular sieving through

graphene oxide membranes. Science 2014, 343 (6172), 752-754.

4. Gao, J.; Feng, Y.; Guo, W.; Jiang, L., Nanofluidics in two-dimensional layered

materials: inspirations from nature. Chem. Soc. Rev. 2017, 46 (17), 5400-5424.

66

5. Shao, J.-J.; Raidongia, K.; Koltonow, A. R.; Huang, J., Self-assembled two-

dimensional nanofluidic proton channels with high thermal stability. Nat. Commun. 2015,

6 (1), 7602.

6. Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S., The Chemistry of Graphene

Oxide. Chem. Soc. Rev. 2010, 39 (1), 228-240.

7. Dimiev, A. M.; Alemany, L. B.; Tour, J. M., Graphene Oxide. Origin of Acidity,

Its Instability in Water, and a New Dynamic Structural Model. ACS Nano 2013, 7 (1), 576-

588.

8. Perreault, F.; Fonseca de Faria, A.; Elimelech, M., Environmental applications of

graphene-based nanomaterials. Chem. Soc. Rev. 2015, 44 (16), 5861-5896.

9. Nair, R. R.; Wu, H. A.; Jayaram, P. N.; Grigorieva, I. V.; Geim, A. K., Unimpeded

permeation of water through helium-leak–tight graphene-based membranes. Science 2012,

335 (6067), 442-444.

10. Goh, K.; Karahan, H. E.; Wei, L.; Bae, T.-H.; Fane, A. G.; Wang, R.; Chen, Y.,

Carbon nanomaterials for advancing separation membranes: A strategic perspective.

Carbon 2016, 109, 694-710.

11. Koltonow, A. R.; Huang, J., Two-dimensional nanofluidics. Science 2016, 351

(6280), 1395.

12. Canning, J.; Huyang, G.; Ma, M.; Beavis, A.; Bishop, D.; Cook, K.; McDonagh,

A.; Shi, D.; Peng, G.-D.; Crossley, M. J., Percolation Diffusion into Self-Assembled

Mesoporous Silica Microfibres. Nanomaterials 2014, 4 (1), 157-174.

13. Guo, F.; Silverberg, G.; Bowers, S.; Kim, S.-P.; Datta, D.; Shenoy, V.; Hurt, R. H.,

Graphene-based environmental barriers. Environ. Sci. Technol. 2012, 46 (14), 7717-7724.

67

14. Abraham, J.; Vasu, K. S.; Williams, C. D.; Gopinadhan, K.; Su, Y.; Cherian, C. T.;

Dix, J.; Prestat, E.; Haigh, S. J.; Grigorieva, I. V.; Carbone, P.; Geim, A. K.; Nair, R. R.,

Tunable sieving of ions using graphene oxide membranes. Nat. Nanotechnol. 2017, 12 (6),

546-550.

15. Zheng, S.; Tu, Q.; Urban, J. J.; Li, S.; Mi, B., Swelling of Graphene Oxide

Membranes in Aqueous Solution: Characterization of Interlayer Spacing and Insight into

Water Transport Mechanisms. ACS Nano 2017, 11 (6), 6440-6450.

16. Jiang, L.; Yang, T.; Liu, F.; Dong, J.; Yao, Z.; Shen, C.; Deng, S.; Xu, N.; Liu, Y.;

Gao, H.-J., Controlled synthesis of large-scale, uniform, vertically standing graphene for

high-performance field emitters. Adv. Mater. 2013, 25 (2), 250-255.

17. Cai, M.; Outlaw, R. A.; Butler, S. M.; Miller, J. R., A high density of vertically-

oriented graphenes for use in electric double layer capacitors. Carbon 2012, 50 (15), 5481-

5488.

18. Lu, X.; Feng, X.; Werber, J. R.; Chu, C.; Zucker, I.; Kim, J.-H.; Osuji, C. O.;

Elimelech, M., Enhanced Antibacterial Activity through the Controlled Alignment of

Graphene Oxide Nanosheets. Proc. Natl. Acad. Sci. 2017, 114 (46), E9793.

19. Ma, Y.; Jang, H.; Kim, S. J.; Pang, C.; Chae, H., Copper-assisted direct growth of

vertical graphene nanosheets on glass substrates by low-temperature plasma-enhanced

chemical vapour deposition process. Nanoscale Res. Lett. 2015, 10 (1), 308.

20. Zhang, P.; Li, J.; Lv, L.; Zhao, Y.; Qu, L., Vertically aligned graphene sheets

membrane for highly efficient solar thermal generation of clean water. ACS Nano 2017, 11

(5), 5087-5093.

68






2017, 29 (23), 1605096.




10879.



ACS Nano 2018, 12 (1), 234-244.

25. Wang, Z.; Tonderys, D.; Leggett, S. E.; Williams, E. K.; Kiani, M. T.; Spitz

Steinberg, R.; Qiu, Y.; Wong, I. Y.; Hurt, R. H., Wrinkled, Wavelength-Tunable Graphene-

Based Surface Topographies for Directing Cell Alignment and Morphology. Carbon 2016,

97, 14-24.

26. Zhou, K. G.; Vasu, K. S.; Cherian, C. T.; Neek-Amal, M.; Zhang, J. C.;

Ghorbanfekr-Kalashami, H.; Huang, K.; Marshall, O. P.; Kravets, V. G.; Abraham, J.; Su,

Y.; Grigorenko, A. N.; Pratt, A.; Geim, A. K.; Peeters, F. M.; Novoselov, K. S.; Nair, R.

R., Electrically controlled water permeation through graphene oxide membranes. Nature

2018, 559 (7713), 236-240.

69

27. Spitz Steinberg, R.; Cruz, M.; Mahfouz, N. G. A.; Qiu, Y.; Hurt, R. H., Breathable

vapor toxicant barriers based on multilayer graphene oxide. ACS Nano 2017, 11 (6), 5670-

5679.

28. Pei, S.; Cheng, H.-M., The reduction of graphene oxide. Carbon 2012, 50 (9), 3210-

3228.



657.

70

Chapter 5 Controlled Release from Intercalated Graphene Oxide Films: Edge-

and Basal-Plane-Specific Kinetics

5.1 Introduction

Substances encapsulation and release are important techniques in many applications,

including flavor, fertilizer, pesticide and drug deliveries.1-3 The matching of release and

demand is critical, which require the release to be slow or controlled.3-4 In a controlled

release system, substances are sustained in specific matrix and released into desired media

under physical or chemical control.1-2, 4 Different systems have been explored over years,

including polymer coated tablets,1, 5 composite gel6 and microsphere.7 In recent years,

discovery of 2D materials shed new light on this field due to their unique structures and

outstanding physicochemical properties.8-9 2D materials possess high specific surface area

and rich surface functionalities thus can be treated as carriers or hosts.10-12 Several methods

have been demonstrated to obtain controlled release vehicles using various 2D materials.

Covalent modification is widely discovered by chemically linking other molecules on the

surface of nanosheets, water-soluble polymers such as polyethylene glycol (PEG) and

polyethylenimine (PEI) can be readily attached on GO or MoS2 to enhance the

physiological stability and biocompatibility.13-14 Noncovalent functionalization can be

achieved through π-π stacking, hydrophobic bonding, or van der Waals interactions, such

as hematin-dextran conjugate attachment.15-16 2D nanosheets can also mitigate the burst

release using Layer-by-layer technique through π-π stacking, electrostatic or hydrogen

bonding interactions.17

71

However, current methods focus on covalent or noncovalent bonding which are in a

manner of point-to-point interaction with functional regions. The active sites/spaces on

basal-plane or edge that can form bonding with each molecule are with limited numbers,

which adversely affects the efficiency of substance loading and releasing.15, 18-19 Besides,

specific binding mechanism also limits the types of desired active materials and subsequent

applications.11-12 To fully utilize the ultrahigh specific surface area and unique layered

architectures, we pay attention to the confined space between stacked nanosheets – the

continuous 2D nanochannels – which can be viewed as empty capsules that are of potential

in pre-intercalation and controlled release.

GO nanochannels – with interlayer spacing ~8 angstroms – are natural matrices to

encapsulate molecules.20-21 Conventional GO films often stacks 2D nanosheets

horizontally by evaporation-induced assembly, building a tightly packed 2D film with

uniform nanochannels. However, the absorption of water molecules and repulsive

interactions between ionized GO nanosheets lead to inevitable swelling.20, 22 Once

associated with long term water immersion, the interlayer spacing will increase from 0.8

to 6 nm and widely open channels to media.23 As introduced in Chapter 2, metal ions-GO

suspensions can be colloidally engineered and co-deposited to obtain orderly intercalated

films. In dry state, intercalated ions are well confined inside of nanochannels, tightly

arranged along versatile topographies.21, 24 When the intercalated GO films are re-

immersed in water, the intercalants are likely to be released out due to expansion of

nanochannels. Therefore, we proposed and designed a release model from intercalated 2D

nanochannels where molecular intercalants are pre-loaded into GO gallery spaces.

Experimental studies on rhodamine B (RhB) dye, used as a model, are carried out to reveal

72

the release behaviors in phosphate-buffered saline (PBS) solution (pH 7.4). As introduced

in Chapter 2-4, the compressive wrinkling and crumpling of GO films can obtain 2D

gallery spaces with different topographical features, which is chosen as the main parameter

to study the diffusive release rates based on topography-related order.25-26 The edge- and

basal-plane-specific kinetics of planar, 1D wrinkled and 2D crumpled nanochannels are

tested and discussed. This type of fluidic-space manipulation should allow the intelligent

design of 2D-material-based technologies such as time-release drug eluting coatings.


As shown in Figure 5.1a, a pure RhB coated polymer substrate was immersed in

phosphate-buffered saline (PBS) solution, followed by an immediately release of pinkish

RhB dye molecules. UV absorption indicated that ~80% of RhB was released in the first

minute (characteristic peak of RhB at ~550 nm).27 The diffusion reached equilibrium in 30

min and maintained for 24 hrs. RhB contains a central planar section and planar groups, it

is 15 angstrom in length and 4.3 angstrom in thickness.28 Due to cationic center and

carboxylic group, RhB is water soluble and can be colloidally stable in GO suspension at

a mass ratio of 0.25 mg/mg (RhB/GO). In Figure 5.1b, the RhB/GO suspension was drop

cast on a fixed-area substrate to obtain intercalated GO films, where all RhB molecules are

physically confined into interlayer galleries. The resulting intercalated films represent

similar lamellar structures with GO. A RhB/GO coated polymer substrate was immersed

in PBS solution as shown in Figure 5.1c. The RhB within GO was slowly diffused into

73

solution, showing a faint trace of pinkish dye. UV absorption shows a gradual increasement

of RhB in solution through 48 hrs, which reached equilibrium in 24 hrs.

Figure 5. 1. Schematic and release behaviors of RhB and RhB/GO films in PBS solution. a,

Top, photo of a RhB coated polystyrene substrate immersed in PBS solution, releasing RhB

molecules rapidly into solution. Scale bar, 1cm. Bottom, UV absorption of pure RhB release

diagram ranging from 1 min to 24 hrs. The majority of RhB molecules was already released within

the first minute. b, Schematic of GO nanosheet and RhB molecule. Depositing a RhB/GO co-

suspension can give a intercalated planar film where RhB molecules are arranged within 2D

nanochannels. c, Top, photo of a RhB/GO planar film on polystyrene substrate immersed in PBS

solution, slowly releasing RhB molecules. Bottom, UV absorption of RhB released from RhB/GO

planar film ranging from 1 min to 48 hrs. Loaded RhB was gradually released from GO

nanochannels and reach equilibrium in ~24 hrs.

Figure 5.1 presents the basic phenomenon on diffusion of RhB from planar GO films.

To better understand the release rates and underlying pathways, we introduced three types

of nanochannels, each possesses two types of diffusion pathways. As shown in Figure 5.2,

planar, 1D wrinkled and 2D crumpled RhB/GO films are prepared using introduced pre-

intercalation method (See Materials and methods). RhB molecules are confined within

74

nanochannels along different topographies. GO nanochannels are constructed by

nanosheets, where gaps between those nanosheets tend to expand along X, Y and Z

directions during swelling. Figure 5.2a shows the edge specific pathways along planar, 1D

wrinkled and 2D crumpled nanochannels. In 1D wrinkled nanochannels, releasing from

edge includes being parallel (//) or perpendicular (⊥) to wrinkles. The diffusion of

intercalants through basal plane is shown in Figure 5.2b. The surface morphologies of

planar, 1D and 2D textured RhB/GO films are shown in Figure 5.2c, indicating different

nanochannels in a topographical manner.

Figure 5. 2. Schematic of release pathways and surface morphologies of RhB intercalated GO

films. RhB/GO textured films exhibit nanochannels with different topographies, each film contains

two types of diffusion pathways. a, Edge-specific release pathways for planar, 1D and 2D textured

RhB/GO films. b, basal plane-specific release pathways for planar, 1D and 2D textured RhB/GO

films. c, Surface morphologies of RhB/GO textured films. Scale bar, 20 μm.

Release isotherms of RhB/GO textured films under room temperature are obtained in

Figure 5.3. The concentration/absorption relationship of RhB is calibrated using standard

75

RhB samples under same conditions. The results are well fitted in a linear relationship as

shown in Figure 5.4. In the total release isotherms (Figure 5.3a), RhB molecules are

released though both basal plane and edge pathways. Planar RhB/GO films exhibit the

fastest release rate, reaching equilibrium in ~12 hrs. 2D crumpled RhB/GO films exhibit

the slowest release rate and reach equilibrium in ~48 hrs. The equilibrium concentrations

of all three RhB/GO films are 0.65 ppm, ~60% of total RhB loading (Figure 5.5).

Interestingly, the diffusion behaviors through only the basal plane pathways show similar

trends in Figure 5.3b, indicating basal plane pathways are dominating the release from

RhB/GO films. To further verify this phenomenon, isotherms of RhB/GO films through

only the edge pathways are monitored as shown in Figure 5.3c and 5.6. From the photos in

Figure 5.6 we can see, the release of edge-specific release on RhB/GO textured films is

considerably slower even though only one edge is exposed to aqueous media. In a period

of 28 days, all samples show faded pinkish color of dye molecules and expose brownish

color from GO matrix. A clear boundary of loaded/released region in RhB/GO film is

highlighted after 4 days of release.

Since RhB molecules are confined within dry GO matrix thus cannot diffuse until the

nanochannels are hydrated and expanded, the isotherms obtained might reflect the

cumulative results of hydrating and releasing rates. Therefore, the hydration rate of

nanochannels need to be addressed before we further discuss the release mechanism of

intercalated GO films. The time dependent XRD results of GO textured films are shown in

Figure 5.7, from which we can see the interlayer spacing of all GO samples increase from

8 to 10 angstrom in 5 min. After 6 min of swelling in PBS solution, GO peaks disappear in

all textured samples indicating disordered structures of nanochannels due to expansion.

76

The first data of isotherms is recorded after 5 min of immersion, therefore, the UV

absorption profiles can be used for analysis of release rates.

Figure 5. 3. Release behaviors of RhB/GO textured films through different release pathways.

a, Total release isotherms of different RhB/GO textured films, include basal-plane- and edge-

specific pathways. b, Basal-plane-specific release isotherms of different RhB/GO textured films. c,

Edge-specific release isotherms of different RhB/GO textured films.

Figure 5. 4. Calibration curve of concentration to absorption of RhB.

77

Figure 5. 5. Release behavior of RhB sample. RhB is drop cast on polystyrene substrate and

undergoes same treatments. Thermally compression shows no effect on release profile of RhB.

Figure 5. 6. Photos of edge-specific release of RhB/GO textured films in 28 days. Planar, 1D

and 2D RhB/GO films were prepared on polystyrene substrate, and hang vertically above the

surface of PBS solutions. Only one edge of textured films gets in contact with solution. RhB/GO

films underwent a slow release of pink-colored RhB and gradually exposed brownish color of GO.

A clear boundary of pink/brown was highlighted in planar RhB/GO film.

78

Figure 5. 7. Dynamic XRD results of GO textured films swelling in PBS solutions. The

interlayer spacing of GO textured films increased rapidly during immersion.

5.3 Next steps

RhB intercalated GO films with different topographies are prepared and characterized.

Release isotherms are measured based on basal plane- and edge- specific pathways.

Experimental data shows that diffusive release rates in the rank order: planar > 1D

wrinkled > 2D crumpled films. However, release rates through basal plane are higher than

that from edge pathways.

For next steps, we are going to establish a suitable model to describe the diffusion of

RhB molecules in 2D nanochannels and calculate the related diffusion coefficients. The

model is based on the lamellar structures of GO films, with dimensions of 1.5 cm 1.5 cm

370 nm. The effect of fluidic-space manipulation on diffusive kinetics will be addressed.

Detailed parameters and topographic-related coefficients will be illustrated during

modelling.

79


Materials

Ethanol and rhodamine B were purchased from Sigma-Aldrich. Gibco™ phosphate

buffered saline solution (1x, pH 7.4) was purchased from Thermo Fisher Thermo Fisher

Scientific. Thermally responsive polyethylene heat shrink films were bought from Grafix.

Polydimethylsiloxane (PDMS) was made from a SYLGARD 184 silicone elastomer kit.

All water was deionized (18.2 MΩ, milli-Q pore). All reagents were used as received

without further purification.

Fabrication of RhB/GO textured films

GO nanosheets were prepared by a modified Hummers’ method, with lateral size ≈1 μm,

thickness ≈1 nm. The GO suspensions used in the colloidal and film formation experiments

is 1 mg mL-1, with a C/O atomic ratio of ≈2.1. Potential GO impurities—such as N, S, Mn,

K, Cl, and P—were not detected by XPS.29 The polymer shrink film was cut into 1.5*1.5

cm2 squares and washed with ethanol. Once dry, samples were treated with air plasma in a

Deiner Atto standard plasma system with a borosilicate glass chamber and a 13.56 MHz,

0-50 W generator. The chamber pressure was pumped down to and maintained at 0.13

mbar while being flushed with air for 5 min. Plasma was then generated at 100% power

(50 W) for 15 min followed by slow venting of the chamber. An aqueous suspension of 1

mg mL-1 GO and 0.25 mg mL-1 RhB was prepared.

Planar RhB/GO films: Polystyrene rectangles with dimensions of 3.83.8 cm2 were

thermally contracted to 1.51.5 cm2 under 130 C to pre-obtain thermal stability. 150 L

80

of RhB-GO suspension was drop cast on the surface of the 1.51.5 cm2 polystyrene

substrate and dried in a 60 C oven to form a 370 nm thick RhB/GO planar film. Then the

RhB/GO planar film was stabilized in a 130 C oven for 6 min.

1D wrinkled RhB/GO films: Polystyrene rectangles with dimensions of 1.54.0 cm2 were

prepared. Each polystyrene rectangle was masked with scotch tape only to leave a 1.51.5

cm2 clean gap in the center, which was to limit subsequent drop casting in selected area.

Then 150 μL of RhB-GO suspension was drop cast on the gap and form a 370 nm thick

RhB/GO film at 60 °C. The protection tape was then removed and both uncoated sides are

clapped for 1D shrinking at 130 °C for 6 min. Finally, 1.51.5 cm2 uniform wrinkled

RhB/GO films were obtained in the center of the polystyrene substrate without distortion.

2D crumpled RhB/GO films: Polystyrene rectangles with dimensions of 1.51.5 cm2 were

prepared. 150 μL of RhB-GO suspension was drop cast on the rectangle and form a 370

nm thick RhB/GO film at 60 °C. Then samples were put into a 130 °C oven for 6 min and

thermally contracted to obtain 0.70.7 cm2 crumpled RhB/GO films.

Preparation of RhB samples

Planar RhB coated sample: Polystyrene rectangles with dimensions of 3.83.8 cm2 were

thermally contracted to 1.51.5 cm2 under 130 C to pre-obtain thermal stability. 150 L

of 0.25 mg mL-1 RhB aqueous suspension was drop cast on the surface of the 1.51.5 cm2

polystyrene substrate and dried in a 60 C oven. Then the RhB coated planar polystyrene

substrate was stabilized in a 130 C oven for 6 min.

2D crumpled RhB coated sample: 150 L of 0.25 mg mL-1 RhB aqueous suspension was

drop cast on the surface of the 1.51.5 cm2 polystyrene substrate and dried in a 60 C oven.

81

Then samples were put into a 130 °C oven for 6 min and thermally contracted to obtain

0.70.7 cm2 crumpled RhB coated substrates.

Release test of RhB and RhB/GO films – total release

RhB and RhB/GO films with different topographies were put into 30 ml of PBS solution

in 40 ml glass vials under mild shaking at 60 rpm. The tested solutions were carefully

sampled and tested by UV absorption in a period of 48 hrs.

Release test of RhB/GO films – basal plane specific release

PDMS was made from a SYLGARD 184 silicone elastomer kit. It is comprised of

base/curing agent to be mixed in a 10 (base) :1 (curing agent) ratio by weight for manual

mixing. 1.8 g base/curing agent mixture was pour in an aluminum dish under vacuum for


it was carefully poured along the edge of RhB/GO films to seal the edge specific pathways.

Then mixture was fully cured under room temperature overnight. RhB/GO films with

different topographies were put into 30 ml of PBS solution in 40 ml glass vials under mild

shaking at 60 rpm. The tested solutions were carefully sampled and tested by UV

absorption in a period of 48 hrs.

Release test of RhB/GO films – edge specific release

1.8 g of PDMS base/curing agent mixture was pour in an aluminum dish under vacuum for


it was carefully poured on the surface of RhB/GO films to seal the basal plane specific

pathways. Then mixture was fully cured under room temperature overnight. RhB/GO films

with different topographies were hang above 30 ml of PBS solution in 40 ml glass vials

under mild shaking at 60 rpm, where one exposed edge of RhB/GO films was in contact

82

with PBS solution. The tested solutions were carefully sampled and tested by UV

absorption in a period of 60 days.

Real-time XRD test of GO films swelling in PBS solution

Multiple planar, 1D wrinkled and 2D crumpled GO films are prepared. Each type of GO

films was submerged in PBS solutions for 0, 1, 2, 3, 4, 5, and 6 mins, then took out with

extra solution wiped. The XRD results of each sample were measured immediately in a

period of 2 min. After 6 mins, all three types of films show no peak, indicating

nanochannels are saturated and disordered.

Characterization

The surface morphologies of the GO, were investigated using a field emission SEM (LEO

1530 VP) operating at 10.0 kV for low, medium, and high resolution imaging. Before the

SEM imaging, all samples were coated with a layer of AuPd (<1 nm). Release of RhB was

measured using JASCO V-730 UV-Visible spectrophotometer. The interlayer spacings of

GO and RhB/GO films were identified by XRD on a Bruker AXS D8 Advance instrument

with Cu Kα radiation (λ = 1.5418 Å). Photographs were taken by an EOS digital SLR and

compact system camera Canon EOS 100D.

5.5 References

1. Shaviv, A.; Mikkelsen, R. L., Controlled-Release Fertilizers to Increase Efficiency

of Nutrient Use and Minimize Environmental Degradation - A Review. Fert. Res. 1993,

35 (1), 1-12.

83

2. Huang, X.; Brazel, C. S., On the Importance and Mechanisms of Burst Release in

Matrix-Controlled Drug Delivery Systems. J. Control. Release 2001, 73 (2), 121-136.

3. Madene, A.; Jacquot, M.; Scher, J.; Desobry, S., Flavour Encapsulation and

Controlled Release – A Review. Int. J. Food Sci. Technol. 2006, 41 (1), 1-21.

4. Azeem, B.; KuShaari, K.; Man, Z. B.; Basit, A.; Thanh, T. H., Review on Materials

& Methods to Produce Controlled Release Coated Urea Fertilizer. J. Control. Release 2014,

181, 11-21.

5. Barde, M.; Davis, M.; Rangari, S.; Mendis, H. C.; De La Fuente, L.; Auad, M. L.,

Development of Antimicrobial-Loaded Polyurethane Films for Drug-Eluting Catheters. J.

Appl. Polym. Sci. 2018, 135 (27), 46467.

6. Katouzian, I.; Jafari, S. M., Nano-Encapsulation as a Promising Approach for

Targeted Delivery and Controlled Release of Vitamins. Trends Food Sci. Technol. 2016,

53, 34-48.

7. He, Q.; Shi, J., Mesoporous Silica Nanoparticle Based Nano Drug Delivery

Systems: Synthesis, Controlled Drug Release and Delivery, Pharmacokinetics and

Biocompatibility. J. Mater. Chem. 2011, 21 (16), 5845-5855.

8. Sun, X.; Liu, Z.; Welsher, K.; Robinson, J. T.; Goodwin, A.; Zaric, S.; Dai, H.,

Nano-Graphene Oxide for Cellular Imaging and Drug Delivery. Nano Res. 2008, 1 (3),

203-212.


183-191.

84

10. Yuan, B.; Zhu, T.; Zhang, Z.; Jiang, Z.; Ma, Y., Self-Assembly of Multilayered

Functional Films Based on Graphene Oxide Sheets for Controlled Release. J. Mater. Chem.

2011, 21 (10), 3471-3476.

11. Li, B. L.; Li, R.; Zou, H. L.; Ariga, K.; Li, N. B.; Leong, D. T., Engineered

Functionalized 2D Nanoarchitectures for Stimuli-Responsive Drug Delivery. Mater. Horiz.

2020, 7 (2), 455-469.

12. Kurapati, R.; Kostarelos, K.; Prato, M.; Bianco, A., Biomedical Uses for 2D

Materials Beyond Graphene: Current Advances and Challenges Ahead. Adv. Mater. 2016,

28 (29), 6052-6074.

13. Zhang, L.; Lu, Z.; Zhao, Q.; Huang, J.; Shen, H.; Zhang, Z., Enhanced

Chemotherapy Efficacy by Sequential Delivery of siRNA and Anticancer Drugs Using

PEI-Grafted Graphene Oxide. Small 2011, 7 (4), 460-464.

14. Liu, T.; Wang, C.; Gu, X.; Gong, H.; Cheng, L.; Shi, X.; Feng, L.; Sun, B.; Liu, Z.,

Drug Delivery with PEGylated MoS2 Nano-sheets for Combined Photothermal and

Chemotherapy of Cancer. Adv. Mater. 2014, 26 (21), 3433-3440.

15. Liu, C.-C.; Zhao, J.-J.; Zhang, R.; Li, H.; Chen, B.; Zhang, L.-L.; Yang, H.,

Multifunctionalization of Graphene and Graphene Oxide for Controlled Release and

Targeted Delivery of Anticancer Drugs. Am. J. Transl. Res. 2017, 9 (12), 5197-5219.

16. Jin, R.; Ji, X.; Yang, Y.; Wang, H.; Cao, A., Self-Assembled Graphene–Dextran

Nanohybrid for Killing Drug-Resistant Cancer Cells. ACS Appl. Mater. Interfaces 2013, 5

(15), 7181-7189.

17. Tanum, J.; Heo, J.; Hong, J., Spontaneous Biomacromolecule Absorption and

Long-Term Release by Graphene Oxide. ACS Omega 2018, 3 (5), 5903-5909.

85

18. Machado, M.; Silva, G. A.; Bitoque, D. B.; Ferreira, J.; Pinto, L. A.; Morgado, J.;

Ferreira, Q., Self-Assembled Multilayer Films for Time-Controlled Ocular Drug Delivery.

ACS Appl. Bio Mater. 2019, 2 (10), 4173-4180.

19. Zhou, T.; Zhou, X.; Xing, D., Controlled Release of Doxorubicin from Graphene

Oxide Based Charge-Reversal Nanocarrier. Biomaterials 2014, 35 (13), 4185-4194.

20. Dimiev, A. M.; Alemany, L. B.; Tour, J. M., Graphene Oxide. Origin of Acidity,

Its Instability in Water, and a New Dynamic Structural Model. ACS Nano 2013, 7 (1), 576-

588.




10879.



23. Zheng, S.; Tu, Q.; Urban, J. J.; Li, S.; Mi, B., Swelling of Graphene Oxide

Membranes in Aqueous Solution: Characterization of Interlayer Spacing and Insight into

Water Transport Mechanisms. ACS Nano 2017, 11 (6), 6440-6450.





2017, 29 (23), 1605096.

86




27. Zhang, R.; Hummelgård, M.; Lv, G.; Olin, H., Real Time Monitoring of the Drug

Release of Rhodamine B on Graphene Oxide. Carbon 2011, 49 (4), 1126-1132.






657.

87

Chapter 6 Appendices

Appendix to Chapter 2

Description of the surface charge modelling

The surface composition of GO nanosheets, as well as complexation behaviors of different

cations on those surfaces is complicated. Here we provide a simplified model of Mn+-GO

colloids to explain the zeta potential data in Figure 2.1b in the main text. Although cation

complexation may occur on a variety of charged and polar sites,1 we assume in the model

that complexation occurs primarily occurs on fully charged acidic sites, including carboxyl

groups, and carbonyl associated and hydroxyl in the form of vinylogous acids. Equilibrium

stability constants for the Mn+-acetic acid system from the literature were used as estimates

for the Mn+-GO system. In detail, the stability constants for Fe(III)-, Al(III)-, Pb(II)-,

Co(II)-, Ni(II)- and Ag(I)-acetic acid complexes are logK = 4.29, 3.43, 2.70, 1.93, 2.12 and

0.73, respectively.2-3 The theoretical surface charge density on Mn+-GO nanosheets can be

calculated from the loading of ionized species on GO nanosheets as 4

𝜎 =𝑧𝑖𝑒𝑐𝑖𝑁𝐴

𝐴𝜌 (1)

Where zi is the valency of ith species, ci is the concentration of bound ith species, e =

1.6×10−19 Coulombs, NA is Avogadro’s constant, A is theoretical specific surface area of

GO (estimated as 1 578 m2 g-1 by combining theoretical surface area of graphene (2 630

m2 g-1) and atomic C/O ratio of GO ~2.1), is mass concentration of GO (0.1 mg ml-1). To

relate the concentration of Mn+-GO colloids with their zeta potentials, Gouy-Chapman

equation (2) was adopted for its suitability in describing a distribution of dissolved ions at

a charged surface (electrical double layers),4

88

𝜎𝑠 =2𝜀𝑘𝑇𝜅

𝑧𝑒sinh(

𝑧𝑒𝜁

2𝑘𝑇) (2)

where z is the valency of the counter-ions, is the solution permittivity (=r0, r = 78.5

is obtained on DLS), k the Boltzmann constant, T the temperature and κ the reciprocal of

Debye length (nm-1).5-6 The Debye length is given by the expression 0.304/√𝐼, where I is

the ionic strength defined as 1

2∑𝑧𝑖

2[xi]; xi is the molar concentration of the ith species, and

zi is valency. All Mn+-GO colloids were prepared in 20mM NaNO3 aqueous solution to

minimize the effect of different ionic strengths on zeta potential. Therefore, the Debye

lengths of all Mn+-GO colloids are calculated to lie between 0.5-1 nm.

First, by combining equation (1) and (2), with experimental zeta potential of 0.1 mg ml-

1 GO suspension (-48 mV), we can back-calculate the concentration of total acidic sites on

GO nanosheets (~0.03 mM). Then using the stability constants of Mn+-acetic acid and K =

[Mn+-GO*]/[Mn+][GO*] we can calculate the bound concentrations of ionized species in

each Mn+-GO colloid. Note the theoretical surface charge density of Mn+-GO nanosheets

now can be written as equation (3),

𝜎 =𝑧𝑖𝑒𝑐𝑖𝑁𝐴

𝐴𝜌=

𝑒𝑁𝐴(−1×𝐺𝑂𝑎𝑐𝑖𝑑𝑖𝑐𝑠𝑖𝑡𝑒𝑠+𝑛𝑖×[𝑀𝑛+−𝐺𝑂∗])

𝐴𝜌 (3)

where ni is effective charge of each Mn+-GO binding site, based on our observation of

colloidal behaviors of Mn+-GO suspensions, we proposed a multi-binding mode for Mn+-

GO nanosheets, where tri/divalent cations (Fe(III), Al(III), Co(II), Ni(II) and Pb(II))

interact with two complexation sites on GO nanosheets, while Ag(I) associates with only

one complexation site. That is, ni is 3/2 for Fe(III)-, Al(III)-; 2/2 for Pb(II)-, Co(II)-, Ni(II)-

and 1 for Ag(I)-GO nanosheets.

89

Finally, we combine equation (1) and (2) again, as well as the as-calculated ionized

species and then do a back calculation to obtain the theoretical zeta potentials of different

Mn+-GO colloids. The theoretical zeta potentials of different Mn+-GO colloids with

different concentrations can be found in Figure 2.1b, which are in good agreement with

experimental data.

Figure 6. 1. Schematic of the fabrication process to generate textured GO. The concentration

of stock GO suspension was 0.65 mg ml-1; the polymer shrink film was cut into 4 cm2 squares and

washed with ethanol. First, 150 μl of GO suspension was drop-cast onto plasma-treated substrates.

Once dry, the planar GO films were placed in an oven at 140 °C for 30 min to actuate shrinking

and film compression.

Figure 6. 2. FT-IR results of GO and Fe(III)-GO films. Comparing with GO, Fe(III)-GO shows

strong stretching vibrations at 1320 cm-1, corresponding to the formation of O=C-O-Fe complexes.

Besides, a new band at 685 cm-1 appeared in Fe(III)-GO, corresponding to the Fe-O groups.7-8

90

Figure 6. 3. Morphologies of GO and Fe(III)-GO nanosheets. a, AFM image and accompanying

height profile of GO and Fe(III)-GO nanosheets drop-cast from diluted suspension onto mica. GO

and Fe(III)-GO nanosheets both show lateral size of ∼1 μm, and thickness of ~1 and ~2 nm,

respectively. b, SEM images of GO and Fe(III)-GO nanosheets drop-cast from diluted suspension

onto Si substrates. Scale bar, 1 μm.

Figure 6. 4. Experimental -potential of Fe(III)-Co(II) based MGI as a function of ([Fe(III)-

Co(II)])/C ratio. -potentials of 0.1 mg ml-1 GO dispersions as function of metal cation

concentration (where [Fe(III)-Co(II)] = [Fe(III)] = [Co(II)]), the colloidal stability of Fe(III)/Co(II)-

based MGI can be achieved by > +15mV (+/+ repulsion due to surface charge inversion).

Various metal oxide textured films are shown in Figure 6.5, where one can see the Fe-

Co oxide film replicates the major features of crumpled GO films. Co3O4, NiO and TiO2

also exhibit crumpled features but with slightly different textures and in some cases

containing visible nanoparticles. We propose that these three colloidal unstable cases

neutralize or screen the negative charge on GO nanosheets and cause random restacking

91

and agglomeration, leading to imperfect film formation and texture development.9 Note

that although the bare Ti ion would be quadrivalent, Ti(IV) exists as [Ti(Cl5)-] in the

solution phase, and is thus not capable of flipping the native negative charge on GO

nanosheets. The Ag(I)-GO case is an exception in that it is colloidally stable (see Figure

2.1b, main text), but produces very low quality films. We believe the failure to replicate

textures in the Ag system is related to the formation of zerovalent Ag nanoparticles by GO-

induced reduction, followed by fusion and particle/droplet growth as reported in our

previous study.10

Figure 6. 5. Surface morphologies and crystal structures of metal oxide textured films

fabricated by various MGIs at colloidally stable loading. Metal oxide textured films are

assembled from MGIs at atomic M/C of 1/33, including CoFe2O4/Co2FeO4 (via Fe(III)/Co(II)-GO);

Co3O4 (via Co(II)-GO); NiO (via Ni(II)-GO), TiO2 (via Ti(IV) -GO) and Ag (via Ag(I)-GO). Scale

bar, 5 m.

92

Figure 6. 6. Porosity determination for textured metal oxide films from SEM micrographs by

image analysis (ImageJ). Morphologies of textured Fe2O3 film fabricated via Fe(III)-based MGI

at initial atomic Fe/C ~1/333. The contrast option was used to auto-calculate the area percentages

of solids and pores.

At very high metal-carbon ratio (33/1; Figure 6.7a), the bulk oxide phase appears (yellow

arrows) with a microtexture associated with the former site of the GO film found as an

imprint on its outer surface (white arrows), in the manner of impression fossils found in

natural sedimentary deposits displaying the textures of once-living plants or animals after

degradation (here combustion) of the organic components.11

Figure 6. 7. Effect of metal-carbon ratio on surface morphologies of Fe oxide textured films

from Fe(III)-based MGIs, and crystal structures and nanostructures of tessellated Fe oxide

films (initial atomic Fe/C ~ 3/1). a, Metal oxide microstructures created by the thermal

compression texturing technique after MGI deposition. The initial metal-carbon ratios of MGI are

1/333, 1/33, 1/17, 1/3, 3/1 and 33/1, respectively. Scale bar, 10 m (top), 2 m (middle), 1 m

(bottom). b, XRD spectrum of resulted Fe2O3 films. c, Surface morphology of tessellated Fe2O3

films. Scale bar, 100 nm. d, Crystal lattice of single nanoplate of Fe2O3 films. The fringe spacing

93

shown is 0.250 nm, corresponding to (110) plane of α-Fe2O3. Scale bar, 2 nm. e, SAED of Fe2O3

single nanoplatelet.

Figure 6.8 shows edge-on images of the films (MGI deposits with Fe/C ~1/3 both pre-

and post-annealing) clearly showing the multilayer structures. With increase in the metal

loading (initial Fe/C from 1/3 to 3/1), the thickness of the Fe oxide layers increases from ~

20 to 75 nm.

Figure 6. 8. Side views of Fe-based MGI deposits before annealing and Fe2O3 textured films.

left, Fe-based MGI depositions before annealing with atomic Fe/C ~1/3. Scale bar, 200 nm. middle,

Fe2O3 textured structures fabricated from Fe-based MGI with initial atomic Fe/C ~1/3. Scale bar,

200 nm. right, Fe2O3 textured structures fabricated from Fe-based MGI with initial atomic Fe/C

~3/1. Scale bar, 500 nm.

Figure 6.9 shows results of auxiliary experiments in which Fe(III) salts (without GO)

were deposited directly onto reduced GO or HOPG surfaces and calcined. The surface

films consist of particles rather than tessellated platelets, showing that two-sided

confinement in gallery spaces is necessary to achieve atomic-scale 2D growth, as reported

from the metallized graphene ink technique described in the main text.

94

Figure 6. 9. Nanostructures of Fe2O3 films fabricated from simple casting of Fe(III) salts on

the external surfaces of reduced GO or HOPG. a, Fe2O3 films formed on the external surface

of a reduced GO film. b, Fe2O3 films formed on the top surface of HOPG. 100mM Fe(III) salts

were cast on the carbon surfaces, followed by dehydration and annealing to remove the carbon-

based substrate. All scale bars, 200 nm.

Figure 6.10 shows results of thermal gravimetric analysis of GO, Fe-based MGI with

initial metal-carbon ratio of 1/3 and Fe(NO3)3 salts carried out in air at a heating rate of 10

C min-1. From the TGA curves we can see the decomposition of Fe(NO3)3 is complete at

~200 C with weight retention of 32.1%, which is very close to theoretical value

(FeO1.5/Fe(NO3)3 =33.1%). During the annealing of GO, the first weight loss at ~200 C is

caused by the thermal decomposition and deoxygenation GO, while the second weight loss

at ~450 C is rGO oxidation. In Fe-based MGI sample, the weight retention (22.1%) is also

close to theoretical value (FeO1.5/[Fe(NO3)3+3.3*C+1.57*O] =26.0%). Since GO is fully

oxidized over 450 C, the formation of Fe oxide must occur before the destruction of the

GO scaffold, which we propose is important for the assembly of platelet oxide structures.

After annealing, GO is completely burned out, and the final product of Fe-based MGI after

annealing (6.10 inset) shows the bright reddish color indicative of α-Fe2O3.

Figure 6. 10. TGA curves of Fe-based MGI with metal-carbon ratio of 1/3, GO and Fe(NO3)3

salts in air. The decomposition of Fe(NO3)3 to Fe2O3 is completed at ~200 C, while GO is fully

95

oxidized over ~450 C. inset: photo of final product of Fe-based MGI after annealing. Heating rate:

10 C min-1.

Figure 6. 11. Effect of metal-carbon ratio on surface morphologies of Fe-Co oxide textured

films from Fe(III)/Co(II)-based MGIs, and morphologies of GO and Fe-Co oxide textured

films. a, Metal oxide microstructures created by the thermal compression texturing technique after

MGI deposition. The initial metal-carbon ratios of MGI are 1/1/33, 1/1/17, 1/1/3 and 3/3/1,

respectively. Scale bar, 10 m (top), 2 m (middle), 1 m (bottom). b, Side view of multi-layered

GO film. Scale bar, 200 nm. c, Side view of multi-layered CoxFe3-xO4(x=1,2) (with initial atomic

Fe(III)/Co(II)/C = 1/1/3). Scale bar, 200 nm. d, Surface morphology of CoxFe3-xO4(x=1,2)

tessellation (with initial atomic Fe(III)/Co(II)/C = 3/3/1). Scale bar, 100 nm.

Although the lowest energy plane of CoFe2O4 is close-packed (111), the majority of

exhibiting crystal planes of Fe-Co oxide are (011) planes (Figure 6.12). The slight

distortion of crystal structure may be caused by the coexistence of both CoFe2O4 and

metastable Co2FeO4.

96

Figure 6. 12. Nanostructure and crystal structure of CoFe2O4/Co2FeO4 textured film. a, XRD

spectra of templated Fe-Co oxide films. b, SAED of CoFe2O4 single nanoplatelet. c-d, Co2FeO4

phase was unstable under the electron beam, while the CoFe2O4 phase remained stable. Scale bar,

100 nm.

97

Figure 6. 13. Detailed example applications of MGI in biotexture replication, 3D shape

creation and printability. a,b: Demonstration of biotexture replication: a, Topographic

surfaces of Rhododendron x 'Roseum Elegans' leaf (left) and Fe2O3 replica of the leaf (middle).

Scale bar, 10 µm. Detailed stoma structures of Fe2O3 replica (right). Scale bar, 200 nm. b, Human

hair (left), scale bar, 20 µm, and detailed cuticles (inset: scale bar, 10 µm). Fe2O3 hair replica with

side- and top-view (middle and inset). Scale bar, 20 µm. Detailed cuticle structures of Fe2O3 hair

replica (right). Scale bar, 10 µm. c-f: Paper-based 3D shape creation: c, Surface morphology of

GO strand made by GO paper scrolling (left). Scale bar, 5 µm. Photos of MGI strands and their

Fe2O3 replicas (right). Scale bar, 1 cm. d, Morphology of Fe2O3 strand replicas (left). Scale bar, 5

µm. Tessellation structure of Fe2O3 strand replica (right). Scale bar, 500 nm. e, Photos of MGI

loops and their Fe2O3 replicas (left and middle). Scale bar, 1 cm. Surface morphology of planar

structured metal oxide tessellation of Fe2O3 replica of loop (right). Scale bar, 300 nm. Printability:

f, MGI writing and airbrush painting on ceramic substrates, and their Fe2O3 patterns after annealing

and oxidation. Scale bar, 1 cm.

98

Figure 6. 14. Fe2O3 textured structures fabricated in the absence of GO, or in the presence of

anionic polymer chains (poly(acrylic acid)) exhibit uncontrolled particle growth and powdery

coatings that lack mechanical integrity to form free-standing films. a, Fe2O3 textured structures

fabricated from Fe(III) salts. b, Fe2O3 textured structures fabricated from Fe(III)-based PAA

suspensions. Scale bar, 2 m.

As shown in Figure 6.15, a trial on stretchability of Fe2O3 textured film is carried out.

Even though the crumpled morphologies theoretically render Fe2O3 films with promising

stretchability, the brittle nature of ceramic materials is unmodifiable. Our 2D crumpled

morphologies can only possibly provide a structural flexibility in designed (X,Y) directions

and contraction range (theoretically stretching percentage of as-prepared Fe2O3 textured

film is ~500%). Compression in Z direction or a certain degree of stress concentration on

films could result in crack and fracture. Only equally forcing on (X,Y) plane of crumpled

Fe2O3 film can allow the Fe2O3 textured film exhibits its geometrical flexibility. To reduce

the stress concentration and exhibit the structural stretchability, Fe2O3 textured film was

slightly embedded inside of an elastomer PDMS, then stretched conformally within the

elastomer along uni or biaxial directions. As shown in Figure 6.15, the textured Fe2O3 film

is capable to be stretched to 125% of its original size (limited by the stretchability of PDMS,

which is ~25%), and promising to be further stretched if embedded in an elastomer with

higher stretchability. No crack of Fe2O3 film is observed by optical microscope or naked

eye. This technique is promising for futuristic stretchable ceramic materials.

99

Figure 6. 15. Stretching behaviors of PDMS-fixed textured Fe2O3 film. Textured Fe2O3 film is

stretched in uni/biaxial directions on a paper-covered Cu substrate, it can be stretched to 125% of

its original size under biaxial stretching. No crack of Fe2O3 film is observed by optical microscope

or naked eye. Scale bar, 1 cm.

References

1. Dimiev, A. M.; Alemany, L. B.; Tour, J. M., Graphene oxide. Origin of acidity, its

instability in water, and a new dynamic structural model. ACS Nano 2013, 7 (1), 576-588.

2. Sillén, L. G.; Martell, A. E.; Bjerrum, J., Stability Constants of Metal-Ion

Complexes. Chemical Society: London, 1964.

3. Perrin, D. D.; Sillén, L. G., Stability Constants of Metal-Ion Complexes, Part B :

Organic Ligands. Pergamon Press: Oxford; New York, 1979.

4. Zhan, H.; Cervenka, J.; Prawer, S.; Garrett, D. J., Electrical Double Layer at

Various Electrode Potentials: A Modification by Vibration. J. Phys. Chem. C 2017, 121

(8), 4760-4764.

5. Jimbo, T.; Higa, M.; Minoura, N.; Tanioka, A., Surface Characterization of

Poly(acrylonitrile) Membranes Graft-Polymerized with Ionic Monomers As Revealed by

ζ Potential Measurement. Macromolecules 1998, 31 (4), 1277-1284.

6. Konkena, B.; Vasudevan, S., Understanding Aqueous Dispersibility of Graphene

Oxide and Reduced Graphene Oxide through pKa Measurements. J. Phys. Chem. Lett.

2012, 3 (7), 867-872.

100

7. Liu, R.; Zhu, X.; Chen, B., A New Insight of Graphene oxide-Fe(III) Complex

Photochemical Behaviors under Visible Light Irradiation. Sci. Rep. 2017, 7, 40711.

8. Si, Y.; Samulski, E. T., Synthesis of Water Soluble Graphene. Nano Lett. 2008, 8

(6), 1679-1682.

9. Luo, J.; Jang, H. D.; Sun, T.; Xiao, L.; He, Z.; Katsoulidis, A. P.; Kanatzidis, M.

G.; Gibson, J. M.; Huang, J., Compression and Aggregation-Resistant Particles of

Crumpled Soft Sheets. ACS Nano 2011, 5 (11), 8943-8949.




10879.

11. McNamara, M. E.; Briggs, D. E. G.; Orr, P. J.; Wedmann, S.; Noh, H.; Cao, H.,

Fossilized Biophotonic Nanostructures Reveal the Original Colors of 47-Million-Year-Old

Moths. PLoS Biol. 2011, 9 (11), e1001200.

101


Figure 6.16 shows results of thermal gravimetric analysis of GO-Fe(III)/Co(II) films, GO

and Fe(NO3)3/Co(NO3)2 salts carried out in air at a heating rate of 10 °C min-1. From the

TGA curves we can see the decomposition of Fe(NO3)3/Co(NO3)2 is complete at ~200 °C

with weight retention of 32.8%, which is very close to theoretical value

(CoFe2O4/[2*Fe(NO3)3+Co(NO3)2] =35.1%). During the annealing of GO, the first weight

loss at ~200 °C is caused by the thermal decomposition and deoxygenation GO, while the

second weight loss at ~450 °C is rGO oxidation. In GO-Fe(III)/Co(II) films, the weight

retention (25.1%) is also close to theoretical value (CoFe2O4/[2*Fe(NO3)3+Co(NO3)2+GO]

=20.1%). Since GO is fully oxidized over 500 °C, the formation of CoFe2O4 must occur

before the destruction of the GO scaffold, which enables the 2D assembly and replication

of wrinkled topographies.

Figure 6. 16. Detailed fabrication process of CoFeFFs. a. CoFeFFs fabricated by conformal

coating and uniaxial contraction of GO-Fe(III)/Co(II) suspension on polystyrene substrate. Heating

above glass transition temperature (Tg ~ 100 °C) triggers polymer relaxation to produce wrinkled

textures. The film is calcined at 600 °C to oxidatively remove graphene and convert metal ions to

a CoFeFF. b. Sketch of conversion of GO-Fe(III)/Co(II) suspension to tessellated CoFe2O4 film.

102

Figure 6. 17. XRD spectrum of CoFeFF.

Figure 6. 18. TGA curves of GO-Fe(III)/Co(II) films, GO and Fe(NO3)3/Co(NO3)2 salts in air.

The transformation of Fe(NO3)3/Co(NO3)2 salts to CoFe2O4 is completed at ~200 °C, while GO is

fully oxidized over 500 °C. Heating rate: 10 °C min-1.

103

Figure 6. 19. Surface morphologies of GO wrinkled film and CoFeFF. Scale bar, 20 μm.

Figure 6. 20. Cross-section of CoFeFF. From the side view, CoFeFF exhibits undulate structures,

with a film thickness of ~150 nm. Scale bar, 5 μm.

Figure 6. 21. SAED of CoFe2O4 single nanoplatelet. The resulting nanoplatelet with basal

surfaces that are primarily (011) planes.

104

Figure 6. 22. Morphologies of GO nanosheets. AFM image and accompanying height profile of

GO nanosheets drop-cast from diluted suspension onto plasma treated mica. GO nanosheets show

lateral size of ∼1 μm, and thickness of ~1 nm.

105


Figure 4.1 part i and Figure 6.23 shows idealized and actual structures of the VAGME

films in cross section. The top and side views of neat (Zr-free) wrinkled GO films exhibit

an irregularly wrinkling pattern instead of the desired regular arrangement of line segments

at same height (Figure 6.23a). Delamination voids are observed between the GO top films

and polystyrene substrates, and these may be responsible for the irregular wrinkle

geometries. Delamination is likely due to weak binding between the negatively charged

GO nanosheets and the air-plasma-treated polymer substrate. This phenomenon exists in

all neat GO samples regardless of the thickness (Figure 6.23a). The lack of structural

regularity makes it difficult to capture the full set of GO wrinkles in any one thin section,

resulting in a low yield of open nanochannels.

We attempted to achieve a more regular zig-zag pattern by creating stronger

film/substrate binding. Multivalent metal cations ZrO2+ and Fe3+ were doped into the films

by depositing GO from ZrOCl2 or Fe(NO3)3 solutions to introduce (+/-) electrostatic

attraction at the GO-substrate interface. To maintain film quality, the GO-metal

solution/suspensions were designed to flip the negative charge on GO nanosheets in

suspension to positive, with high enough zeta potential to create colloidal stability in the

+/+ electrostatic repulsion regime (discussed later). The side views of Zr- and Fe-GO

wrinkled films are shown in Figure 6.23b-c. All metal ion-doped GO wrinkled films show

no detachment voids and possess a more regular zig-zag pattern. Interestingly, the Zr-GO

wrinkled films (Figure 6.23b) exhibit higher ridge curvature (sharper corners) than the Fe-

GO films (Figure 6.23c), which facilitates the subsequent sectioning. The further

development of VAGMEs was therefore pursued using Zr-doped GO.

106

Figure 6. 23. Morphologies of wrinkled films. a, Side (cross-sectional) views of epoxy fixed-

neat GO wrinkled films. Left, 1 μm-thick GO films after compressive deformation and embedding

in epoxy. Scale bar, 20 μm. Right, 2 μm-thick GO films after compressive deformation and

embedding in epoxy. Scale bar, 50 μm. b, Sides views of metal ion-doped GO/epoxy composites.

1 μm-thick Zr-GO films after compressive deformation and embedding in epoxy. c, Sides views of

metal ion-doped GO/epoxy composites.1 μm-thick Fe-GO films after compressive deformation and

embedding in epoxy. Scale bar, 20 μm.

The Zr-GO films were made by drop casting from aqueous GO suspensions containing

the soluble Zr salt ZrOCl2. This “pre-stacking” method leads to films with ZrO2+ cations

intercalated between GO sheets. It is important for these GO/ZrOCl2 solution/suspensions

to be colloidal stable during preparation and casting. Suspensions with different Zr/C ratios

were prepared and the colloidal stability evaluated using zeta-potential measurements

(Figure 6.24). The pure GO aqueous suspension is colloidal stable due to (-/-) repulsion of

ionized nanosheets, reflected by a zeta potential of -48 mV at Zr/C = 0. Progressing

addition of ZrO2+ cations first reduces the magnitude of the negative charge then induces

a charge flip and eventually a second regime of colloidal stability at high positive zeta

107

potential (> +40 mV) for Zr/C ratios larger than 1/22. The (+/+) repulsion between Zr-GO

nanosheets maintains colloidally stable and produces high quality films during subsequent

drying induced assembly.

Figure 6. 24. Experimental ζ-potential of GO nanosheet suspensions with varying degrees of

ZrOCl2 addition, expressed as a function of [Zr]/C atomic ratio. GO dispersions are all at 0.1

mg mL-1 solids loading. Colloidal stability of ZrOCl2 doped GO can be achieved by ζ > +20mV

(+/+ repulsion due to surface charge inversion).

The microtome sectioning technique plays an important role in obtaining an intact

VAGME thin film. Several potential film flaws are shown in Figure 6.25. For very thin

sections (cut thickness < 10 μm), the 1 μm Zr-GO strips may fragment during sectioning

(Figure 6.25a). Use of very thick Zr-GO strips (> 2 μm) can lead to internal delamination

within the strips after sectioning (Figure 6.25b). In Figure 6.25c, multiple holes are

observed in the VAGME, which originated as trapped air bubbles in the epoxy caused by

insufficient degassing. Finally, full curing of the epoxy at 60 °C makes the matrix quite

stiff and can produce interfacial delamination between Zr-GO films and resin during

sectioning (Figure 6.25d). Observing and then avoiding these structural defects led to the

final protocol for VAGME fabrication (see Materials and methods in Chapter 4).

108

Figure 6. 25. Structural failures of VAGME appeared in developing microtome sectioning

technique. a, Zr-GO films break apart at cut thickness of 2 μm. Scale bar, 2 μm. b, Delamination

of nanosheets at Zr-GO film thickness of 2 μm. Scale bar, 10 μm. c, Holes on VAGME caused by

air bubbles after a 5-min degassing period. Scale bar, 100 μm. d, Delamination at the interfaces of

Zr-GO films and epoxy while epoxy is fully cured at 60 °C. Scale bar, 100 μm.

The water vapor flux through VAGME at 100 C is shown in Figure 6.26a, and indicates

a decline in water flux over time. We hypothesized this decline is due to the gradual thermal

reduction (deoxygenation) of GO which results in decrease of chemical polarity and

affinity for water.1-2 Interestingly, XRD results does not show a significant change of

interlayer spacing in Zr-GO films upon 100 C water vapor treatment. The interlayer

spacing remain at 8.83 angstrom (Figure 6.26b), which may be due to the presence of ZrO2+

ions that continue to act as pillars to set the spacing even while oxygen functional groups

are removed.3 XPS tests were therefore carried out to probe the chemical transformation of

GO nanosheets during thermal treatment. Since Zr-GO films contain ZrO2+ and absorbed

water molecules, the cumulative O/C ratio cannot be used to reliably monitor the oxygen

content of GO nanosheets. Instead we used the atomic ratio of oxidized carbon to total

carbon to track the degree of thermal reduction. From the trend in Figure 6.26c we can see

the oxygen content declines significantly after 12 hrs of exposure, which is consistent with

permeation results. The data are calculated from fitting profiles of C1s peaks as shown in

109

Figure 6.27. Therefore, thermal reduction of GO nanosheets results in a decline in

hydrophilicity, which further supports that molecules can only transport through Zr-GO

nanochannels on VAGME.

Figure 6. 26. Time-dependent behavior and properties of VAGMEs during exposure to water

vapor at 100 C. a, Time-resolved measurements of water vapor transmission fluxes at 100 C

over the course of 24 hrs exposure to the 100 C water vapor permeate. b, XRD spectra of Zr-GO

films after different treatments, including thermally activated compressive wrinkling and 100 °C

water vapor exposure for 6, 12, 18 and 24 hrs. c, O-C/C atomic ratios by XPS for wrinkled Zr-GO

films after 100 °C water vapor treatment for 0, 6, 12, 18 and 24 hrs for monitoring deoxygenation.

110

Figure 6. 27. C1s XPS spectra for wrinkled Zr-GO films to 100 °C water vapor for various

times, in hrs : (a) 0, (b) 6, (c) 12, (d) 18, (e) 24.

References

1. Pei, S.; Cheng, H.-M., The reduction of graphene oxide. Carbon 2012, 50 (9), 3210-

3228.






Documents

By Muchun Liu B.Sc., Materials Science and Engineering