University of Connecticut

DigitalCommons@UConn
Doctoral Dissertations University of Connecticut Graduate School

12-7-2015

Boron Nitride Nanosheets and Their Applications

in Polymeric Composites
Zhenhua Cui
unicuizhenhua@gmail.com

Follow this and additional works at: http://digitalcommons.uconn.edu/dissertations

Recommended Citation
Cui, Zhenhua, "Boron Nitride Nanosheets and Their Applications in Polymeric Composites" (2015). Doctoral Dissertations. 946.
http://digitalcommons.uconn.edu/dissertations/946
 Boron Nitride Nanosheets and Their Applications in Polymeric Composites
Zhenhua Cui, PhD
University of Connecticut, 2015

Boron nitride sheets, as the isoelectric analogue of graphene, have gradually attracted researchers’

attentions due to their excellent mechanical, electrical and thermal properties. Compared with

graphene, the exploration of potential applications of boron nitride sheets for polymeric

nanocomposites is limited due to the lack of practical and efficient preparation methods. Current

methods like solvent sonication and chemical vapor deposition are not ideal and difficult to scale

up for composites applications.

The work described in this thesis focuses on the solution of two key problems of fabricating boron

nitride/polymer nanocomposites. The first problem is the exfoliation of hexagonal boron nitride. In

this work, two approaches are developed and can be used for the preparation of boron nitride

nanosheets: chemical functionalization and direct exfoliation by interfacial trapping. By chemically

modifying hexagonal boron nitride, hydroxylated boron nitride nanosheets are prepared. Not only

are the van der Waals forces between sheets reduced, but functional groups are introduced on the

sheets increasing their dispersibility in water and polymer matrices. Further modifications are also

feasible based on the attached hydroxyl groups. Unfunctionalized pristine boron nitride sheets can

be directly prepared by the second approach: an interfacial trapping technique. Sheets are formed at

the interface of the phase-separated water/organic solvent mixture to minimize interfacial energy.

When prepared in different conditions, sheets stabilize the interface in two different forms: thin,

continuous films or emulsions. Mechanisms of the two exfoliation approaches are studied.

 
 
 Zhenhua Cui – University of Connecticut, 2015

The other problem is the preparation of BN sheets/polymer composites with improved properties.

To fully take advantage of each exfoliation method, melt mixing is employed for the functionalized

boron nitride nanosheets/polymer composites while several more complicated methods are used for

the direct use of exfoliated sheets. Those methods include solution casting, heat processing and in-

situ polymerization. Obtained nanocomposites show one or more improved properties, such as

mechanical properties, thermal stabilities and electrical properties. Potentially, these

nanocomposites may be used as fire resistant building materials, dielectric materials or

mechanically strong/lightweight materials.

 
 
 Boron Nitride Nanosheets and Their Applications in Polymeric Composites

Zhenhua Cui

B.S., Nankai University, 2009

A Dissertation

Submitted in Partial Fulfillment of the

Requirements for the Degree of

Doctor of Philosophy

at the

University of Connecticut

2015

i 
 
 Copyright by

Zhenhua Cui

2015

ii 
 
 APPROVAL PAGE

Doctor of Philosophy Dissertation

Boron Nitride Nanosheets and Their Applications in Polymeric Composites

Presented by
Zhenhua Cui, B.S.

Major Advisor ___________________________________________________________________

Douglas H. Adamson

Associate Advisor_________________________________________________________________
Richard Parnas

Associate Advisor_________________________________________________________________
Andrey V. Dobrynin

University of Connecticut
2015

iii 
 
 Acknowledgement
This work could not have been accomplished without the help and support of my advisor,

committee members, colleagues, friends and family. I especially want to thank my major

advisor, Professor Douglas Adamson. Thank you for teaching me on how to design and

conduct experiments and more importantly, how to think out-of-box to solve research

problems. And thank you for providing funding for my projects, which helped me to enjoy

the fun of doing research. I also want to thank my committee, made up of Professor Parnas,

Dobrynin, Nieh and Sun, all of whom have helped me a lot on my research. I also like to

thank Professor Cao for useful discussions about various aspects of dielectric materials,

Professor Schniepp at College of William & Mary for the AFM measurement. Thank you

to all the staff in IMS, without their help on so many instrumental techniques I won’t be

able to finish my work. I would also like to thank all of my collaborators, including Dr.

Andre Martinez for ATRP reactions, Rui Ma for dielectric measurements and valuable

discussions, Zhen Cao for computational work and Jingjing Liu for gas barrier properties

measurements. I want to say thank you to everyone in my research group for being great

labmates to work with, they are: AJ, Chetan, Steve, Kevin, Jennifer, Garrett, Martin,

Harish, Reihaneh, Drona, Elizabeth and Deepthi. I also need to thank all my undergraduate

students, Rui, Chris, Qui, Sheng and Kate who helped me on a lot of experiments that I

could not finish on my own. I would like to thank all of the friends I have made at UConn

for those good times we spent together these years. Finally I want to thank my family for

being supportive and encouraging all the time.

   

iv 
 
 Table of Content
Chapter 1 Introduction ........................................................ 1 
1.1 Hexagonal boron nitride .............................................................................................. 1 
1.1.1 Structure ................................................................................................................ 2 
1.1.2 Properties .............................................................................................................. 3 
1.2 Boron nitride nanosheets ............................................................................................. 3 
1.2.1 Methods of production of BNNS .......................................................................... 4 
1.2.1.1 Mechanical exfoliation (sonication) .............................................................. 4 
1.2.1.2 Chemical vapor deposition ............................................................................ 6 
1.2.2 Chemical functionalization of BNNS ................................................................... 7 
1.3 BNNS/polymer composites ......................................................................................... 9 
1.4 Characterization methods of BNNS .......................................................................... 11 
1.4.1 AFM .................................................................................................................... 11 
1.4.2 Raman ................................................................................................................. 11 
1.4.3 EM ...................................................................................................................... 12 

Chapter 2 Preparation and characterization of hydroxylated

hBN ................................................................................... 14 
2.1 Overview.................................................................................................................... 14 
2.2 Preparation of BNO ................................................................................................... 15 
2.3 Characterization of BNO ........................................................................................... 15 
2.3.1 Water dispersity .................................................................................................. 16 
2.3.2 Single layer sheets .............................................................................................. 16 
2.3.2.1 AFM ............................................................................................................. 16 
2.3.2.2 Raman spectroscopy .................................................................................... 19 
2.3.3 Sheet size and morphology ................................................................................. 20 
2.4 Reaction mechanism .................................................................................................. 26 
2.4.1 Reaction temperature .......................................................................................... 26 
2.4.2 Reaction time ...................................................................................................... 27 
2.4.3 Reaction atmosphere ........................................................................................... 29 
2.4.4 Functional groups on BNO ................................................................................. 30 
2.4.5 Reaction mechanism ........................................................................................... 32 
2.5 Reactivity of BNO ..................................................................................................... 34 

v 
 
 2.6 Conclusions................................................................................................................ 36 

Chapter 3 BNO/polymer and fBNO/polymer composites 38 

3.1 Overview.................................................................................................................... 38 
3.2 BNO/Polycarbonate nanocomposites ........................................................................ 39 
3.2.1 Preparation of BNO/PC nanocomposites ........................................................... 39 
3.2.2 Characterization of BNO/PC nanocomposites ................................................... 40 
3.2.3 Flammability test of BNO/PC composites ......................................................... 41 
3.3 PMMA functionalized BNO ...................................................................................... 43 
3.3.1 Introduction of polymer grafted BNO ................................................................ 44 
3.3.2 Reaction to grow PMMA.................................................................................... 44 
3.3.3 Characterization of PMMA grafted BNO........................................................... 46 
3.3.3.1 FTIR ............................................................................................................. 46 
3.3.3.2 FESEM ........................................................................................................ 48 
3.3.3.3 TGA ............................................................................................................. 51 
3.3.3.4 Gel permeation chromatography ................................................................. 53 
3.4 PMMA fBNO/PMMA composite.............................................................................. 54 
3.4.1 Composite preparation ........................................................................................ 54 
3.4.2 Dispersity of sheets in water ............................................................................... 55 
3.4.3 Mechanical properties ......................................................................................... 57 
3.5 Conclusions................................................................................................................ 61 

Chapter 4 Interfacial Exfoliation of BN ............................ 62 

4.1 Introduction................................................................................................................ 62 
4.2 Interfacial exfoliation................................................................................................. 62 
4.2.1 Sonication VS no sonication ............................................................................... 63 
4.2.2 Interfacial exfoliation with different solvents..................................................... 64 
4.2.3 Preparation of BN on glass slide ........................................................................ 65 
4.3 Molecular Dynamics Simulation ............................................................................... 66 
4.4 Characterization of BN network ................................................................................ 69 
4.4.1 Raman spectroscopy ........................................................................................... 69 
4.4.2 Morphology study ............................................................................................... 70 
4.5 Conclusions................................................................................................................ 74 

Chapter 5 BN/polymer thin films with improved dielectric

properties ........................................................................... 75 
vi 
 
 5.1 Introduction................................................................................................................ 75 
5.2 Solution casting method ............................................................................................ 75 
5.3 Heat processing .......................................................................................................... 77 
5.4 Characterization of BN/polymer film ........................................................................ 79 
5.4.1 Film structure ...................................................................................................... 79 
5.4.1.1 SEM ............................................................................................................. 79 
5.4.1.2 UV-Vis ......................................................................................................... 82 
5.4.2 Loading of BN in the thin film ........................................................................... 83 
5.4.3 Dielectric properties............................................................................................ 84 
5.4.3.1 Dielectric constant and loss ratio ................................................................. 85 
5.4.3.2 Breakdown strength ..................................................................................... 90 
5.4.4 Gas barrier properties ......................................................................................... 92 
5.5 Conclusions................................................................................................................ 92 

Chapter 6 BN/PS foam and sheets coated particles .......... 93 

6.1 Introduction................................................................................................................ 93 
6.2 Emulsions stabilized by BNNS ................................................................................. 93 
6.3 Water-in-oil and oil-in-water emulsions .................................................................... 96 
6.4 Polymerization of oil phase ....................................................................................... 99 
6.5 Characterization of composite foam ........................................................................ 101 
6.5.1 TGA .................................................................................................................. 101 
6.5.2 SEM .................................................................................................................. 102 
6.5.3 Flammability ..................................................................................................... 104 
6.6 Potential applications for oil absorption .................................................................. 105 
6.7 Conclusions.............................................................................................................. 106 

CHAPTER 7 Summary and future work......................... 108 

7.1 Summary .................................................................................................................. 108 
7.2 Future work .............................................................................................................. 109 

References ....................................................................... 110 

 

   

vii 
 
 List of Figures
Figure 1‐1. Lattice structure of graphite and hBN[8] .............................................................. 2 
Figure 1‐2. Photograph of dispersion of BN in isopropyl alcohol[32] ..................................... 5 
Figure 1‐3. SEM image of hBN domains grown on Cu surface by CVD (a‐c), A graph 
showing the size distributions of hBN domains as a function of annealing time(d) and  
SEM images showing that h‐BN domains grow into near‐fully covered and continuous film 
(e,f) [43] ................................................................................................................................... 7 
Scheme 1‐4. 1,6‐hexamethylene diisocyanate functionalization of OH‐BNNS via 
carbamate formation to yield isocyanate‐functionalized BNNS[42] ....................................... 9 
Figure 1‐5. Position of the Raman peak for different values of number of sheets.[68] ....... 12 
Figure 1‐6. SEM of small flake hBN (lefg) and large flake hBN(right). Scale bars are 1µm . 13 
Figure 2‐1. Unfunctionalized hBN in water (A) and functionalized and exfoliated hBN in 
water (B). ............................................................................................................................. 16 
Figure 2‐2. AFM images of BNO sheets (A, B and C) and height histogram of image C (D).
 ............................................................................................................................................. 19 
Figure 2‐3. Raman spectra of pristine hBN and BNO nanosheets ...................................... 20 
Figure 2‐4. FESEM images of: A) pristine hBN, B) BNO nanosheets from water suspension, 
C) BNO precipitate after washing with water. Scale bars are one micron .......................... 22 
Figure 2‐5. TEM images of (a, b) higher magnification and lower magnification of BNO 
sheets from water suspension, (c) BNO precipitate after addition of thermally treated 
sample to water with voids on the surface (marked by arrows), and (d) selected area 
(electron) diffraction pattern from (b) ................................................................................ 23 
Figure 2‐6. DLS result of BNO nanosheets water suspension ............................................. 24 
Figure 2‐7. TGA traces of oxidation at 800 °C, 900 °C and 1000 °C ..................................... 27 
Figure 2‐8. FTIR of oxidation product (BNO without any water treatment) of different 
reaction time, pristine hBN and boric acid for comparison ................................................ 28 
Figure 2‐9. TGA traces of hBN heated to 1000 °C in air, argon and nitrogen. .................... 29 
Figure 2‐10. Zeta potential (white Y axis) and DLS hydrodynamic radius (yellow Y axis) 
measurement of BNO/water suspension at different pH. .................................................. 31 
Figure 2‐11. XPS spectra of hydroxylated BNO with the boron peak showing both a B1s(N) 
and a B1s(O) component (left) and the nitrogen peak showing only a N1s(B) component 
(right). .................................................................................................................................. 32 
Scheme 2‐12. Mechanism of the formation of hydroxylated boron nitride (BNO) ............ 33 
Figure 2‐13. FTIR traces of hBN at different stages of preparation: (a) pristine hBN before 
heating, (b)  hBN after heating but before being placed in water, (c) BNO after washing 
with water, and (d) BNO that has been washed by water and reacted with phenyl 
isocyanate ............................................................................................................................ 34 
Figure 2‐14. BNO precipitate after reaction with phenyl isocyanate with the curling of the 
edges. The scale bar is one micron. ..................................................................................... 35 
Figure 2‐15. XRD of pristine hBN and isocyanate functionalized hBN with the 
functionalized material displaying a broader peak shifted to lower angle. ........................ 36 

viii 
 
Figure 3‐1. Picture of PC nanocomposites filled with small flake pristine hBN (top left), 
large flake pristine hBN (top right), small flake BNO (bottom left) and large flake BNO 
(bottom right) ...................................................................................................................... 40 
Figure 3‐2. X‐ray diffraction profiles of PC nanocomposites filled with small flake pristine 
hBN (blue), large flake pristine hBN (red), small flake BNO (green) and large flake BNO 
(purple) ................................................................................................................................ 41 
Figure 3‐3. Snapshots at different time of a video recorded shows BNO/PC strip self‐
extinguished after ignition. .................................................................................................. 42 
Figure 3‐4. Snapshots at different time of a video recorded shows neat PC strip dripped 
with fire ................................................................................................................................ 43 
Scheme 3‐5. Reaction route to prepare PMMA grafted BNO ............................................. 46 
Figure 3‐6. FTIR spectra of (a) BNO, (b) neat PMMA, (c) PMMA grafted BNO, (d) free 
polymer (filtrate) and (e) TGA residue of PMMA grafted BNO (residue). ........................... 47 
Figure 3‐7. SEM of BiBB functionalized BNNS (initiator fBNO) ........................................... 48 
Figure 3‐8. SEM of (a) BNO and PMMA grafted BNO at low (b) and high (c) magnification. 
The scale bars are one micron ............................................................................................. 50 
Figure 3‐9. SEM of the mixture of BNO and PMMA dissolved in THF and then precipitated 
out by methanol. The scale bar is one micron. ................................................................... 51 
Figure 3‐10. TGA of BNO, PMMA grafted BNO and neat PMMA[82] .................................... 52 
Figure 3‐11. SEM of TGA residue of PMMA grafted BNO ................................................... 52 
Table 3‐12. GPC results of neat PMMA as control group and free polymer from grafted 
BNO are shown below. ........................................................................................................ 53 
Figure 3‐13. GPC traces of (a) neat PMMA as control and (b) free polymer in the filtrate of 
product from ATRP reaction ................................................................................................ 54 
Figure 3‐14. Water suspension of BNO (left) and an equal mass of BNO that has been 
grafted with PMMA (right)[82] .............................................................................................. 56 
Figure 3‐15. Films of equal thickness of hBN filled PMMA (left) and PMMA grafted BNO 
filled PMMA (right)[82] .......................................................................................................... 57 
Table 3‐16. Thicknesses of three data points of each thin film sample in Figure 3‐15 ...... 57 
Table 3‐17. Tensile test result of neat PMMA, hBN filled composite and grafted BNO filled 
composites ........................................................................................................................... 58 
Figure 3‐18. Typical stress‐strain curve of neat PMMA, grafted BNO filled PMMA 
composite and pristine hBN filled PMMA composite. ........................................................ 60 
Figure 4‐1. (A) Heptane/water mixture with added BN following brief bath sonication (B) 
Water/heptane/BN initial sample and (C) Water/heptane/BN after sitting for 25 days with 
no sonication. ...................................................................................................................... 64 
Figure 4‐2. Pictures of exfoliated few‐layer BN sheets at interfaces of two water/organic 
solvent systems. The organic used are chloroform(left) and n‐heptane(right) .................. 65 
Figure 4‐3. Dried few‐layer BN film on glass slide ............................................................... 66 
Figure 4‐4. Number fraction distributions of atoms belonging to water molecules (dash‐
dot lines), heptane molecules (dot lines) and BN sheets (solid lines) along the z‐axis 
normal to the water/heptane interface for water/heptane mixtures containing single BN 
sheet (black lines), four BN sheets (red lines), and nine BN sheets (blue lines). Inserts 

ix 
 
show typical configurations of the BN layers at the water/heptane interface. Solvent 
shown is water, and heptane molecules are transparent. .................................................. 67 
Figure 4‐5. Variation of the potential of the mean force normalized by the BN sheet 
surface area along the z‐axis normal to the water/heptane interface. Insets show typical 
sheet configurations. Heptane phase is transparent. ......................................................... 69 
Figure 4‐6. Raman spectra of bulk hBN and few‐layer BN sheets network show changes in 
integrated intensity and the position of peak due to the exfoliation ................................. 70 
Figure 4‐7. Low‐magnification SEM image of BN sheets network ...................................... 71 
Figure 4‐8. TEM images of BN sheets network displaying few‐layer sheets stacks with 
offset and empty spaces ...................................................................................................... 72 
Figure 4‐9. Electron‐diffraction pattern revealing six‐fold symmetry of BN sheets ........... 73 
Figure 4‐10. Low magnification TEM of BN sheets network where an empty space is point 
out by a black arrow ............................................................................................................ 73 
Scheme 5‐1. Schematic illusion of preparation of few‐layer BN coated polymer film ....... 76 
Figure 5‐2. Photograph of PMMA film “coated” with BN ................................................... 77 
Figure 5‐3.  A picture showing BN coated PP film after heat processing. The BN sheets in 
left part of glass slide was erased prior to heat processing. The heat processed PP film 
clearly shows sheets incorporation did not occor in spots where glass air bubbles formed
 ............................................................................................................................................. 78 
Figure 5‐4. SEM of the few‐layer BN coated PMMA film, showing the same region before 
(a) and after (b) intense electron beam treatment and the same region of a neat PMMA 
film before (c) and after (d) intense electron beam treatment .......................................... 80 
Figure 5‐5. SEM of the cross section of few‐layer BN coated PMMA film at low (A) and 
high (B) magnification, showing the embedded BN sheets very close to the surface ........ 81 
Figure 5‐6. Optical transparency tests on neat PMMA film (blue), few‐layer BN film (red), 
and few‐layer BN coated PMMA film (green). .................................................................... 82 
Figure 5‐7. TGA of BN coated PP film and PP film. .............................................................. 84 
Figure 5‐8. Dielectric constant at room temperature as a function of frequency for 
unmodified PMMA film (black squares), BN‐PMMA film with dispersed BN sheets (red 
circles), and BN coated PMMA film (blue triangles). ........................................................... 86 
Figure 5‐9. The loss ratio as a function of frequency (curves have the same identity as in 
5‐7) ....................................................................................................................................... 87 
Figure 5‐10. Dielectric constant and loss tangents of PP and BN coated PP film. .............. 87 
Scheme 5‐11. Capacitors in series model used for calculation ........................................... 88 
Table 5‐12. Dielectric constant of PP and BN coated PP films. Each row shows dielectric 
constants of neat PP and BN coated PP obtained from two parts in one heat‐treated 
sample and the film thickness. ............................................................................................ 88 
d d 1
Figure 5‐13. The plot of  total  total  VS   and a linear fitting of the data .................... 89 
 PP  total  PP
Figure 5‐14. Weibull distribution of the breakdown field of BN coated PMMA, neat PMMA 
film and BN/PMMA composite film at room temperature. ................................................ 90 
Figure 5‐15. Breakdown field of neat PMMA, BN coated PMMA film and BN/PMMA 
composite film ..................................................................................................................... 91 

x 
 
Figure 6‐1. Snapshots at different time of a video recorded under optical microscope after 
water in n‐heptane spheres transferred on glass slide ....................................................... 95 
Figure 6‐2. Sphere size distribution of water‐in‐oil emulsion determined by acoustic 
measurements. .................................................................................................................... 96 
Figure 6‐3. Water‐in‐oil emulsion stabilized by large flake size hBN (left) and oil‐in‐water 
emulsion stabilized by small flake size hBN (right) ............................................................. 97 
Figure 6‐4. Optical microscopy image of small flake hBN stabilized oil‐in‐water emulsions 
(a) right after transferring on glass slide and (b) after about ten minutes ......................... 98 
Figure 6‐5. Photos of a vial of water after large flake hBN stabilized emulsion added into 
water (left route) and small flake hBN stabilized emulsion added into water (right route)
 ............................................................................................................................................. 99 
Figure 6‐6. SEM image of composite foam produced by polymerizing the continuous 
styrene phase. .................................................................................................................... 100 
Figure 6‐7. SEM image of composite particles polymerizing the inner monomer phase of 
small flake BNNS stabilized emulsion ................................................................................ 101 
Figure 6‐8. TGA trace of BNNS/PS composite foam .......................................................... 102 
Figure 6‐9. SEM of BNNS/PS composite foam with different magnifications ................... 103 
Figure 6‐10. Snapshots at different time of videos recorded after specimen (a, bulk PS, b, 
Styrofoam and c, BNNS/PS foam) are ignited ................................................................... 104 
Figure 6‐11. SEM of outer surface of composite foam after flammability test with char 
formed on it. ...................................................................................................................... 105 
Figure 6‐12. Mass change of the foam during three absorption‐burning (cleaning) cycles
 ........................................................................................................................................... 106 
 

xi 
 
 Chapter 1 Introduction

1.1 Hexagonal boron nitride

Hexagonal boron nitride (hBN), also called α-BN is a low density, inorganic ceramic

materials that has been widely used in applications from cosmetics to high temperature

lubricants.[1] The structure of hBN is very similar to the layered structure of graphite, but

unlike graphite, hBN does not occur naturally. Industrially, it is obtained by reacting boron

sources such as boron oxide or boric acid and nitrogen sources such as melamine, urea, or

ammonia at high temperatures (~900 °C) to obtain amorphous boron nitride, followed by

crystallization into the hexagonal phase of hBN at higher temperature under inert

atmosphere.

hBN has broad applications due to many unmatched properties. For example, hBN is used

as an insulating thermoconductive filler in composites because it is electrically insulating

but thermally very conductive.[2-5] Its Mohs hardness is only 1~2 and it is very “unctuous”.

This very low hardness is due to the layered structure and the weak van der Waal forces

that hold the sheets together, which leads to applications as a solid lubricant in rigorous

environments. Besides being used a lubricant, hBN has also served as a high-performance

oxidation-resistant releasing agent and coating due to its high thermal stability, chemical

inertness and oxidation resistance.[1] Last, but not least, due to hBN’s superb mechanical

strength, it’s frequently used in composite for reinforcement.[6, 7]

1 
 
1.1
1.1 Struccture
The lattice stru
ucture of hB
BN is very similar to that of grraphite (Figuure 1-1): saame

honeeycomb lattiice and very

y close cell parameters. They both consist of a 2-dimensioonal

netw
work of hexaagonal rings forming staccked sheets,, but in hBN
N alternating B and N atooms

subsstitute for C atoms. Thatt is why hBN

N is occasioonally referreed to as “whhite graphite” in

relev
vant literaturres. The diff
fference in colors is duee to the covaalent inter-laayer bondingg of

the boron
b and nitrogen
n atom
ms, which lo
ocalize the eelectrons com
mpared withh delocalizedd pi

bond
d electrons in
i black grap
phite. But th
he basal plannes of sheetts are layereed differentlyy in

the two
t materialls to form the 3D solids. In contrast to the AB sttacking withh offset betw
ween

layerrs in graphitte, AA’ staccking with alternative

a b oron and niitrogen atom
ms in successsive

layerrs are verticaally aligned in the stablee mode of hB

BN.

Figure 1-1.
1 Lattice structure
s of ggraphite andd hBN[8]

2 
 
1.1.2 Properties
Since the boron/nitrogen pair of atoms in hBN is isoelectric to a pair of carbon atoms in

graphite, they share a lot of typical properties; high thermal conductivity and excellent

mechanical strength.[1, 6, 7, 9] However, their electrical properties are significantly different.

While graphite, composed of only carbon, is known to be a very good electrical conductor,

the different electronegativities of boron and nitrogen in BN result in it being an insulator

with a large bandgap (ca. 6 eV)[10, 11], a dielectric constant of about 5[12] and a breakdown

voltage of 800 MV/m[13-15], suggesting a potential for electronic and energy storage

applications.[12, 16-20] hBN has also been shown to be a superior substrate for graphene-

based electrical devices.[13, 14, 21]

Other than that, hBN is more chemically inert and

temperature stable than graphite due to the strong and polar B-N bond.[1]

1.2 Boron nitride nanosheets

Since being isolated from graphite, graphene’s unique chemical, electronic and physical

properties rapidly became an important research field in materials science.[22-29] Similar to

graphene and other high aspect ratio stacked materials, in order to further explore hBN’s

extraordinary properties[30-33] hBN needs to be exfoliated into boron nitride nanosheets

(BNNS) and well dispersed in matrix. Thus the tendency for the nanosheets to aggregate or

restack needs to be suppressed. The state of research in BNNS is very like that of graphene

before methods of exfoliation and functionalization were well developed. An important

approach to graphite exfoliation for large-scale applications is to form graphite oxide (GO)

3 
 
to enable exfoliation and suspension of graphene in water.[34] This common route lessens

the sheets’ tendency to restack after exfoliation.[23] However, the oxidation conditions used

for graphite do not work with hBN, which necessitates different approaches for exfoliation.

The number of reports of BNNS is still small compared to those of graphene.[3, 6, 30, 35-37] A

review of the field published by Golberg[38] in 2010, when the work of this dissertation

started, attributed this to two main issues: no experimental evidence published with respect

to hBN nanosheet chemical modifications and the lack of effective methods to prepare

hBN nanosheets on a large scale.

1.2.1 Methods of production of BNNS

The exceptional thermal, mechanical, and electrical properties of high aspect ratio BNNS

make BN an attractive candidate for the design of multifunctional composite materials.

However, similar to graphene, BN’s propensity to restack after exfoliation, coupled with

its chemical inertness, creates obstacles for its applications to nanocomposites[38]. The

exfoliation of hBN has been studied through mechanical[32, 39] and chemical methods[16, 40-
43]
. Work published by using mechanical (sonication) methods and chemical methods

(includes chemical vapor deposition methods and chemical functionalization methods) to

exfoliate hBN will be discussed below.

1.2.1.1 Mechanical exfoliation (sonication)

4 
 
 Figu
ure 1-2. Phottograph of dispersion
d off BN in isoprropyl alcohool[32]

Worrk on hBN exfoliation

e have
h been reeported befoore, such as that by Kim
m and Kirklland

who sonicated hBN

h in N-m
methyl-2-pyrrrolidone (NM
MP) to prodduce nanoshheet samples for

ge study.[44] They
radiaation damag T reported finding tthin regions,, often less tthan five layyers

thick
k, with somee monolayer regions. As the study w
was focused oon the effectt of radiationn on

the hBN sheetss, no mentiion was maade of yielld or condition optimization, and no

funcctionalization
n was reported. Sonicatio
on in NMP w
was also repported by Waarner, who uused

the resulting
r hBN
N stacks as supports
s for studying M Se.[45] Zhi annd collaboraators
Mn doped ZnS

soniccated hBN in
i N, N-dim
methylformam
mide (DMF)) to producee hBN fillerss for compoosite

mateerials.[6] Thee material prroduced sho

owed multipple layers off hBN, withh the numberr of

sheeets in the staccks being a function

f of the
t speed at which the saample was ccentrifuged. The

funcctionality off the sheets, however, could not be determiined due too difficultiess in

obtaining FTIR spectra. In a study by Lin,[41] it iss shown thatt sonication in water coould

“unzzip” hBN to form sheets of hBN witth the edges functionalizzed by hydrooxyl groups and

again
n sheet dim
mensions beiing a functiion of centrrifugation coonditions. H
However, yieelds

weree not reporteed and a very

y long time of sonicatioon (eight houurs) was reqquired. Colem
man

5 
 
reported hBN sheets from sonication in isopropyl alcohol (IPA) and studied the effect of

solvent surface tensions (as shown in Figure 1-2).[32]

1.2.1.2 Chemical vapor deposition

Chemical vapor deposition (CVD) has also been used to produce pristine unfunctionalized

single layer BNNS.[46-48] BNNS with high purity and well controlled orientation can be

produced by CVD (as shown in Figure 1-3).[16, 43] Song reported the large area synthesis of

hBN films consisting of two to five atomic layers from CVD.[49] Choi and Kim produced

hBN sheets by CVD and used it as a substrate for graphene deposition.[50] Ajayan used

CVD to produce sheets with varying amounts of carbon, nitrogen and boron.[51] However,

high cost and limited product quantity remains a challenge for CVD. In order to fully

utilize the extraordinary properties of BNNS for large-scale applications, such as

nanofillers in composites, a more effective and more economically attractive method to

exfoliate hBN is required. Additionally, CVD grown thin BNNS films are brittle and hard

to transfer, so even for applications in which cost is not of high priority, the difficulties of

mixing BNNS with a polymer matrix are another limitation for CVD.

6 
 
 Fiigure 1-3. SE
EM image of
o hBN domaains grown oon Cu surfacce by CVD (a-c), A grapph

sh
howing the size
s distributtions of hBN
N domains ass a function of annealingg time(d) andd

SE
EM images showing
s thatt h-BN domaains grow innto near-fullyy covered annd continuouus

film
f (e,f) [43]

Otheer methods, not involviing sonicatiion or CVD

D, have inclluded using graphene aas a

temp
plate for carb
bon substitu m hBN[52] andd the splittinng of BN nannotubes to fo
ution to form form

ribbo t form grapphene ribbonns from carboon nanotubees[53]

ons, much liike what has been done to

are not
n discussed
d since they are out of th
he scope of th
this paper.

1.2
2.2 Chem
mical fun
nctionallization of BNN
NS
Chem
mical modiffication of BNNS
B by graafting polym
mers is an appproach to diirectly faciliitate

hBN
N exfoliation
n and polym
mer composittes. In orderr to attach ppolymer chaiins to hBN, the

sheeets must firstt be function

nalized, but due
d to boronn nitride’s chhemical inerttness, reportts of

N functionaliization[42, 54-556] are scarce. The exam

hBN mples we aree aware of aare the workk by

7 
 
Bhimanapati et al.[55] using a mixture of phosphoric and sulphuric acid and the work of

Sainsbury et al. by using oxygen radicals in sonicated NMP solutions to attach hydroxyl

groups to hBN, as shown in scheme 1-4.[42] While Bhimanapati did not report the growth

of polymers from their functionalized hBN, the work of Sainsbury described the

stoichiometric addition of peroxides to BNNS that enabled the addition of isocyanate

functional groups to the hBN. These reactive groups were then proposed to react with the

polymer matrix, specifically polyurethane. Methods for functionalizing hBN include the

method of Lin as previously described[41], as well as a method to form associations of hBN

nanosheets with alkyl amines by first using a ball mill to cleave the sheets and produce

defect sites to which an amine was associated.[57] Another reported approach involves the

use of hydrazine, hydrogen peroxide, nitric acid and sulfuric acid heated under pressure to

100 °C, followed by sonication, to produce 0.3 g/L suspensions of hBN.[58] One report

caught our attention, although not using hBN sheets, it was shown that at high

temperatures hBN nanotubes slowly form defect sites that can be used to break the tubes

into smaller segments to aid in solvent dispersion.[59] But the properties of functionalized

BNNS, especially the electrical properties, are affected due to the presence of functional

groups, which are acting as defects in electronic structure.

8 
 
 Scheme 1-4
4. 1,6-hexam
methylene diisocyanate ffunctionalizaation of OH--BNNS via

carb mation to yield isocyanatte-functionallized BNNS[42]

bamate form

1.3
3 BNNS
S/polym
mer com
mpositees
As discussed
d preeviously, a lo
ot of investig
gations in thhe literature w
were focusedd on BNNS and

theirr properties, like the earrlier stages in the studyy of graphenne. The reseearch efforts for

exfo
oliation and functionaliza
f ation have sttimulated annd enabled thhe exploitatioon of properrties

and applicationss of graphen mposite[24, 288, 29]. Similarrly, the reseaarch

ne in polymeeric nanocom

of BNNS/polym
B mer composites may bee expected by the intrroduction off BNNS intto a

poly
ymeric matriix. The exceeptional therrmal, mechaanical, and eelectrical prooperties of hhigh

aspeect ratio BNN

NS make th
hem valuable as novel nnanofiller caandidates foor the designn of

multtifunctional polymeric composite materials[99, 32, 40-42, 60, 61]

with outstandding

9 
 
mechanical[6, 62, 63], thermal[3, 19, 35] or electrically insulation properties, or as functional

materials for high-temperature stability.

Previous studies of BNNS composites [3, 6, 64] have started with hBN materials exfoliated to

various degrees by mechanical means without chemical functionalization. Zhi et al.

reported fabrication of PMMA/BNNS composites with unfunctionalized hBN

nanosheets.[6] Coleman et al. reported composite films of BN using a solution sonication

method involving 48 hours of sonication followed by filtering and adding the filter cake to

thermoplastic polyurethane dissolved in DMF. An improvement in modulus was reported

at 5 and 20% loadings at the cost of reduced strain at break.[32] Besides meeting the

challenge of BNNS/polymer composite’s preparation by mixing unfunctionalized BNNS

and polymer, as is commonly done in the literature, modification is another solution to

increase the interaction between the polymer matrix and nanofillers. So far, BNNS have

been modified through various interactions, including covalent or noncovalent bonding[41,

57]
as discussed previously.

Instead of being dispersed randomly, if the BNNS are dispersed in composites with a

defined structure; for example, a layered structure, it could be beneficial for some specific

composite applications like electronics for energy storage.[12, 16-20]

However, the

development of these applications has been hampered by the lack of a simple and cheap

method for preparation of BNNS/polymer composite without sheets aggregation or phase

separation.

10 
 
1.4 Characterization methods of BNNS
In addition to difficulties in the exfoliation of hBN, characterization of the morphology of

BNNS is complicated. Three methods are commonly used to identify single-layer BNNS:

atomic force microscopy (AFM)[40, 43, 48, 56, 65-67], Raman spectroscopy[42, 65, 68] and electron

microscope (EM)[9, 30, 31, 50, 69, 70]. AFM and Raman are suitable for the identification of

single layer nanosheets. Disadvantages of AFM and Raman compared with the electron

microscope (EM) is the small scan image size. In one image, AFM can only obtain

information from an about 5×5 micron area while EM can easily scan an area of square

millimeters with a depth of field on the order of tens of micron. Therefore EM is used to

study the overall morphology of few-layer sheets.

1.4.1 AFM
AFM provides a surface height profile by scanning the specimen surface using a cantilever

with a extremely sharp tip at its end. With this method, a cantilever tip scans across an area

of the sample while a laser beam measures subtle movements that translate into differences

in height. In this work, all measurements are taken in tapping mode. Tapping mode

oscillates the cantilever at a set frequency and any changes detected in the oscillation

height denote different interactions with the sample. The study of BNNS through AFM has

also revealed that observed height of a single sheet can vary from 0.6nm to 1nm depending

on the type of substrate used or the scan condition.

1.4.2 Raman

11 
 
Ram
man spectrosccopy provid
des the abilitty to identify
fy the presennce of monoolayer sheetss by

the blue N E2g peak att about 13677 cm-1. The bblue
b shift in the position of the charaacteristic BN

c −1 was rep
shiftt of up to 4 cm ported and ex
xplained by the hardeninng of the E2gg phonon duue to

Figure 1-5).[668]
ghtly shorterr B–N bond expected in isolated moonolayers (ass shown in F
a slig

Figure
F 1-5. Position of the
t Raman peak
p for diffeerent values of number oof sheets.[68]

1.4
4.3 EM
An electron
e miccroscope is a microscope that usess accelerated electrons as a sourcee of

illum
mination. Because the waavelength off an electronn beam can bbe much shorter than vissible

lightt, the electrron microsccope has a much highher resolvinng power thhan an opttical

ons producee the imagee of the saample after the

micrroscope. In SEM, scatttered electro

micrroscope deteects and an

nalyzes the scattered eelectrons. Inn TEM, eleectrons direectly

peneetrate throug
gh the samplee.

M focuses on
SEM n the samplle’s surface and its com
mposition w
while TEM pprovides dettails

abou
ut internal composition
c and structurre. TEM haas the abilityy of reachinng much higgher

resollution than SEM. SEM

M allows for large area of sample tto be analyzzed at one ttime

12 
 
wherreas with TE
EM only sm
mall area of sample can be analyzedd at a time. With their oown

advaantages, eacch of these two electro

on microscoopes is widdely used inn the studyy of

NS.[32, 41, 55, 71, 72] Figure 1-6 shows typical

BNN t SEM
M images of two grades of hBN usedd in

fference willl be discussed in details in

this study, smalll flake and large flake;; whose diff

follo
owing chapteers.

Figu
ure 1-6. SEM N(right). Scale bars are 1µm
M of small fllake hBN (leefg) and largge flake hBN

13 
 
 Chapter 2 Preparation and

characterization of hydroxylated hBN

2.1 Overview
For the study of graphene, the exfoliation and suspension of GO in water opened the door

to much of the current research in graphene materials.[34] To the best of our knowledge,

when this study started, there had been no report of the formation of hBN’s equivalent of

GO: hydroxylated sheets that enable the aqueous suspension of single sheets in high yield.

In this chapter the preparation of hydroxylated hBN (BNO) is discussed. To characterize

BNO and study the mechanism of the reaction, AFM, Raman spectroscopy, field emission

scanning electron microscopy (FESEM), TEM, thermo-gravimetric analysis (TGA),

Fourier transform infrared spectroscopy (FTIR), x-ray diffraction (XRD), X-ray

photoelectron spectroscopy (XPS) and dynamic light scattering (DLS) are used. The

reactivity of BNO is also demonstrated by one functionalization reaction with phenyl

isocyanate.

The procedure used to form GO does not work with hBN as hBN is very resistant to

oxidation.[73] However, it was discovered that heating hBN in air results in a mass gain as

oxygen is incorporated into the hBN lattice. Interestingly, following the heat treatment,

stirring the material in deionized water for several minutes results in the mixture

thickening. The reason appears to be the thermally treated material hydrolyzes and

exfoliates to form BNO. These sheets form a suspension without the need for sonication,

14 
 
although mild bath sonication is normally used to increase the rate at which the suspension

forms. Yields of water suspended sheets are up to ~65%, with the equilibrium of

precipitating from the aqueous suspension.

2.2 Preparation of BNO

Samples were prepared by the thermal treatment of hexagonal boron nitride in air. In a

typical experimental run, BN powder (10 g, 99.5%, Alfa Aesar, used as received) was

placed in a quartz tube with both ends opened. The furnace was preheated to 1,000 °C and

held at that temperature. After placing the quartz tube into tube furnace for one hour, the

tube was taken out from the furnace. After cooling, the material was washed with hot water

and dried. Approximately 700 mg of BNO could be collected from 1 g washed

hydroxylated BN powder after brief sonication and centrifugation.

The yield of the process is taken to be the amount of material suspended in water divided

by the amount of precipitate. While the precipitate was hydroxylated as well, it was not

exfoliated and thus did not suspend. A typical determination of yield: 15.2 mg of thermally

treated material was placed in water and briefly bath sonicated. The suspension was then

filtered, and 5.0 mg of material was recovered from the filter, giving a yield of 67%. The

mass of the precipitate was utilized rather than the exfoliated material as sheets were

retained in the filtered material.

2.3 Characterization of BNO

15 
 
2.3
3.1 Wateer disperrsity
Show
wn in Figurre 2-1(A) is a vial contaaining pristiine hBN in deionized w
water after bbath

soniccation. Figu
ure 2-1(B) shows BNO suspended in water aft
fter being soonicated for the

samee time. There is no appaarent water so

olubility witth pristine hB
BN, while thhe hydroxylaated

mateerial results in a cloudy

y suspension
n containingg a small am
mount of preecipitate on the

botto
om of the vial. This is
i clear eviidence that the water dispersity oof the BNO
O is

signiificantly imp
proved comp
pared with pristine hBN..

Figure 2-1. Unfunctionalized hBN in water

w (A) andd functionaliized and exffoliated hBN
N in

water (B).

2.3
3.2 Singlle layer sheets
It is important to determinee if single laayer sheets aare present iin BNO sincce exfoliatioon is

the primary
p objeective of thiss study. Therrefore AFM and Raman spectroscoppy are emplooyed

to fin
nd the evidence of mono
o layer sheetts.

2.3.2.1 AFM
M

16 
 
AFM measurements were conducted at room temperature using a NTEGRA Prima

Scanning Probe Laboratory (NT-MDT, Zelenograd, Russia). Scanning was carried out in

tapping mode. The AFM probes were ACTA silicon cantilevers (APPNANO, Santa Clara,

California) with a typical resonance frequency of f = 300 kHz, a radius of curvature of r <

10 nm, and a spring constant of k = 40 N/m.

Figure 2-2 shows AFM images of suspended BNO sheets drop cast on an HOPG surface at

different magnifications. Single layer sheets with lateral dimensions on the hundreds of

nanometers length scale were observed. A height histogram (Figure 2-2 D) shows two

distinct peaks. The large one with most pixels of the image is the substrate. Therefore, the

center of this peak is assigned as height 0. The height of the second peak is 0.69 nm,

representing the most often observed height of the sheets. The ratio of the areas under these

two peaks gives the ratio of substrate area vs. sheet area, about 15% of the substrate is

covered with sheets. In this histogram, the thickness of all sheets in image 2-2 C has been

averaged, giving higher precision and being more representative than a single height

profile. During the sample preparation, no centrifugation was employed to fractionate the

sheets for AFM imaging, and thus the image represents the entire population of the

suspended material. This is in stark contrast to other methods that rely on centrifugation to

separate the samples into fractions with different numbers of hBN layers.

17 
 
 A

B 

18 
 
 D 

Fig
gure 2-2. AF
FM images of
o BNO sheeets (A, B andd C) and heiight histograam of image C

(D).

2.3.2.2 Ram
man specttroscopy

Ram
man spectra of pristine hBN and BNO nanoosheets from
m water susspensions w
were

obtained by a Reenishaw Ran

nascope Systtem 2000 opperating at 514 nm. Figuure 2-3 indicates

19 
 
 ue shift of 2 cm-1 in thee E2g phonon
a blu n mode at 11369 cm-1 foor the BNO and changees in

integ
grated intenssity. This is consistent
c w report inn literature.[688]
with

Fig
gure 2-3. Raaman spectraa of pristine hhBN and BN
NO nanosheeets

2.3
3.3 Sheett size an
nd morp
phology
The sheet size and
a morpho
ology of BN
NO were invvestigated byy electron m
microscopy and

S.
DLS

FESEM measurrements werre performeed using a JJEOL JSM--6335F coldd cathode ffield

emisssion (12 kV
V) scanning
g electron microscope.
m Figure 2-4 shows FES
SEM imagess of

pristtine hBN, BN
NO nanosheeets from waater suspens ion and BN
NO precipitatte after washhing

with
h water.

20 
 
 A 

B 

21 
 
 C 

Figure 2-4. FESEM im

mages of: A) pristine hBN
N, B) BNO nnanosheets ffrom water

suspension,
s C)
C BNO preccipitate afterr washing w
with water. Sccale bars aree one micronn

M samples were
TEM w prepareed by drop casting the suspension with a low
w concentrattion,

geneerally at ~0
0.1 mg/mL, onto a carrbon grid aand viewed in both traansmission and

diffrraction modee on a FEI Tecnai T12

2 STEM. Figgure 2-5 shoows TEM images of B
BNO

sheeets from wateer suspensio

on at differen
nt magnificat
ations, BNO precipitate aafter additionn of

therm
mally treated
d sample to water
w and electron diffraaction patterrn.

22 
 
 Fig
gure 2-5. TE
EM images of
o (a, b) high
her magnificaation and low
wer magnificcation of BN
NO

sheets from water

w suspen
nsion, (c) BN
NO precipitaate after addiition of therm
mally treatedd

sample
s to waater with voiids on the su
urface (markked by arrow
ws), and (d) selected areaa

(eelectron) difffraction patttern from (b))

EM images (SE
EM and TEM
M) are in go
ood agreemeent with thee AFM anallysis, and shhow

osheets with lateral dimeensions of 10

nano 00-200 nm. F
Figure 2-4 ((a) shows priistine hBN w
with

23 
 
dimeensions of microns
m and smooth
s surfa mple (Figure 2-4
faces and edgges, while thhe BNO sam

(c)) clearly show

ws the rough
h edges and “divots”
“ on tthe surface ccorrespondinng to the size of

the single
s sheetss imaged by AFM in fig
gure 2-2. Thhe images off the sheets ((Figure 2-4 (b))

are consistent
c with
w those obtained
o by AFM, but with the high concentrration of sheets

leadiing to sheet overlap. Sim

milar results are also obttained by TE
EM, where rroughly circuular

thin spots are seeen decoratin

ng the previious (prior too heating) ssurface. Elecctron-diffracttion

patteerns of BNO
O reveal the typical
t six-fo
old symmetrry structure oof hBN (Figuure 2-5 d).

Dynamic light scattering

s (D
DLS) was utilized
u to deetermine thee hydrodynaamic radium
m of

sheeets in the clo

oudy suspen
nsions obtain
ned by usinng a NICOM
MP 300 Submicron Partticle

Sizer.

Figure
F 2-6. DLS
D result of
o BNO nanoosheets wateer suspensionn

24 
 
Result summary:

Mean diameter = 359.6 nm

Chi squared = 1.748

Stnd deviation = 285.5nm (79.4 %)

Baseline Adj. = 0.024%

Coeff. Of Var’n = 0.794

Mean Diff. Coeff. = 1.29E-008 cm2/s

Cumulative result:

25% of distribution < 144.7nm

50% of distribution < 245.8nm

75% of distribution < 418.5nm

90% of distribution < 678.0nm

99% of distribution < 1523.3nm

The results shown above (Figure 2-6) are in general agreement with the results obtained by

AFM, with DLS indicating a mean hydrodynamic radius of 360 nm, which is larger than

the 100 nm length scale obtained by AFM. This can be attributed to the difference of two

characterization methods and the ignorance of the disc shape of BNO during the

calculation of DLS result. Therefore, the parameters obtained from the calculation of size

by DLS are only estimates, and thus the AFM data is expected to be more reliable.

25 
 
2.4 Reaction mechanism

2.4.1 Reaction temperature

To further understand the reaction of hBN oxidation and to determine the optimal

conditions for oxidation, reactions at different temperatures were investigated. Thermal

gravimetric analysis (TGA) was carried out on a TA Instruments Q 500

Thermogravimetric Analyzer at a heating rate of 10 °C min-1 in a platinum pan. Shown

below (Figure 2-7) are three TGA traces run at constant temperature in air. As mentioned

previously, hBN is very thermally stable, and we find that at 800 °C, no significant

oxidation was observed. As the temperature raised, there is an increase of mass with time,

and the rate of oxidation increases with increasing temperature.

Based on these results, our standard oxidations were performed at 1000 °C for a typical 20

grams of hBN starting materials. At 1000 °C the several-hour reaction period provides the

possibility of better control for the extent of reaction which will be discussed in the

following paragraphs. The increase of weight plateaus after approximately 5 hours. This

may suggest the saturation of functionalization when all defects and edges are reacted,

which will also be discussed in the following paragraphs.

26 
 
 1000OC 

900OC 

800OC 

Figu
ure 2-7. TGA
A traces of oxidation at 8800 °C, 900 °C and 10000 °C

2.4
4.2 Reacction tim
me
Furth
her, the reacction was fo
ollowed by FTIR and sspectra from
m different ttime point w
were

comp
pared with boric
b acid (llowest specttrum in Figuure 2-8), whhich gives a clear indicaation

that new peaks are showin

ng up in thee product. F
FTIR transm
mission meassurements w
were

ducted on a Nicolet 56
cond 60 instrumeent coupled with SpecttraTech IR Plan 0044--003

micrroscope.

27 
 
 pBN 
10min 
20min 
1hr 
2hrs 

3hrs 

Boric 
acid 

Fiigure 2-8. FT
TIR of oxidaation producct (BNO withhout any watter treatmennt) of differennt

reaction tim
me, pristine hBN
h and borric acid for ccomparison

In orrder to show
w that our maaterial is not simply boriic acid from the decompposition of hB
BN,

the hydroxylated
h d BNO is rep
peatedly wasshed with hoot water to reemove boricc acid. This ddoes

not result
r in losss of the hydrroxyl peaks in the FTIR
R. Additionaally, boric accid melts at 171

°C and
a boils at 300
3 °C. As the
t oxidation
n is run openn to the air aat 1000 °C, the retentionn of

form
med boric accid is unlikeely and would be obseerved as a loss of masss in the TG
GA.

Additionally, FT
TIR spectra comparing boric acid with hydrooxylated hBN (Figure 22-8)

indiccates no sign
nificant form
mation of borric acid. Anoother possibiility, the form
mation of booron

triox
xide, B2O3, from
f boric accid is ruled out
o as no eviidence of it iis seen by A
AFM or FESE
EM,

and our heating rate is not conducive to its formatioon.[74] Even iif a trace am
mount of B2O3 is

form
med, it would
d be convertted to boric acid
a during treatment w
with hot wateer and wouldd be

disso
olved in the water.

28 
 
2.4
4.3 Reacction atm
mospherre
TGA
A studies ind
dicate that when
w heated in air at tem
mperatures att 1000 °C, hhBN gains m
mass

for a period of time, follow

wed by a leeveling off. In contrast,, when heatted in an arrgon

atmo
osphere at the
t same tem
mperature, no
n mass chhange is obsserved. Interrestingly, w
when

heateed in a nitro
ogen atmosp
phere at 1000
0 °C, the hB
BN losses m
mass (Figure 2-9). Whilee no

reacttion is seen in an argon

n atmospherre, a stable oxidation pproduct is foormed in airr. In

nitro
ogen, howev
ver, with no oxygen preesent, slow decomposition is obserrved. This is in

agreement with the finding

gs of others who have oobserved a loss of masss in a nitroogen

osphere[57] an
atmo n in air[75]. Unlike
nd mass gain U otherr studies how
wever, we obbserve a platteau

in th
he weight gaain with oxid
dation, apparrently causeed by the com
mplete oxidaation of regiions

susceptible to ox
xidation at th
hat temperatu
ure.

Figure 2--9. TGA tracces of hBN heated

h to 10 00 °C in air,, argon and nnitrogen.

29 
 
2.4.4 Functional groups on BNO
The hydrophilicity of BNO discussed in Chapter 2.3.1 implies possible functional groups

on BNO. In previous studies, other groups observed either only –BOH or both –BOH and

–NH groups.[41, 42]

To determine what functional groups are responsible for the

hydrophilicity, a zeta-potential measurement of BNO/water suspensions at pH from 1 to 11

was performed and results are plotted together with DLS measurements as shown in Figure

2-10. At pH lower than 6, BNO nanosheets start to aggregate in suspensions as the

hydrodynamic radius increases. Below pH 4, BNO precipitate out from suspensions,

resulting in clear water. The change of zeta potential is in correspondence with the

hydrodynamic size of DLS. The isoelectric point is at about pH 3, which suggests an acidic

surface. Therefore, –BOH group is dominating the surface charge if there is a small

amount of –NH group.

30 
 
 Figure 2-10. Zeeta potential (white Y ax
xis) and DLS
S hydrodynaamic radius ((yellow Y axxis)

measurement of BNO//water suspeension at diffferent pH.

To further
f study
y the functio
onal groups on BNO. X--ray photoellectron specctroscopy (X
XPS)

meassurements were
w perforrmed using a 595 Scaanning Augger Electronn Spectromeeter.

Evid
dence for thee formation of B-O rath
her than N-O
O bonds comes from reesults shownn in

Figu
ure 2-11. Thee XPS spectrrum of BNO
O identifies tthe B1s and N1s peaks. Both valuess are

in close agreem
ment with values
v reportted for prisstine hBN.[776] Due to tthe backgroound

gen content of the sample, which wee speculate tto result from
oxyg m trace amoounts of oxyygen

abso
orbed on the nanosheets,, we use fittted 1s peakss of differentt species to find the borron-

oxyg
gen bond. While
W boron atoms
a show both
b a B-N 1s peak at 191 eV and a B-O 1s peaak at

191.7 eV, nitrog

gen is found
d to be boun
nd only to booron atoms. The ratio oof B-OH to B
B-N

peak
k areas is app
proximately 1:10. In prisstine hBN, bboron and niitrogen atom
ms are bond oonly

31 
 
to eaach other, without
w any oxygen. Thee oxidation sensitivity oof boron sittes is consisstent

with m other publiished work.[42, 77]

h results from

Figure
F 2-11. XPS spectra of hydroxy
ylated BNO with the borron peak shoowing both a

B1s(N)
B and a B1s(O) com
mponent (lefft) and the niitrogen peakk showing onnly a N1s(B))

com
mponent (righht).

2.4
4.5 Reacction meechanism
m
Baseed on the fact that only –BOH grou
ups are deteected on BN
NO, the reacttion mechannism

can be
b proposed
d as follow: insertion of oxygen
o into the hBN latttice starts froom the edgees or

defects and cutss the hBN in

nto smaller flakes
f alongg the defectss. A similar result has bbeen

32 
 
obseerved in cheemical vapo n studies doone at the ssame tempeerature.[75] A
or infiltration After

wateer is introdu
uced, hydrox
xyl groups are formedd on boron atoms with oxygen atooms

inserrted. This prrocess tends to slow dow

wn when all defects andd edges are ssaturated, whhich

leadss to the plateau of weight change att 1000 °C affter about 5 hours in a ttypical reacttion.

Thiss mechanism
m is depicted in Scheme 2-12.
2

Scheme 2-12. Mechaniism of the fo

ormation of hhydroxylatedd boron nitriide (BNO)

Furth
her evidencee of the mecchanism of BNO
B sheet fformation coomes from F
FTIR. Figuree 2-

13 shows
s the FTIR spectraa obtained att different sstages of maaterial prepaaration. Prioor to

heatiing, the specctra of pristiine hBN exh

hibits only thhe characterristic peaks oof B-N in-pllane

70 cm-1 and B-N-B out of plane beending obserrved at 810 cm-1, as shoown
stretching at 137

ure 2-13(a).[778] The form

Figu mation of bo
oron–oxygenn bonds in tthe heated m
material (beffore

adding to water)) is indicated in the FTIIR spectrum

m 2-13(b) byy the B-O-H
H peak at ~3200

cm-1 and the sm

mall in-planee peak near 1200 cm-1.[442, 79] The ppeak at 12000 cm-1 has bbeen

assig
gned previou n coordinated with threee oxygen in aan extended lattice.[79] A
usly to boron After

hing with waater (Figure 2-13(c)), the peak near 1200 cm-1 ddisappears w
wash while the B-O
O-H

peak
k remains, with der at ~3400 cm-1 indicaating loss of the in-planee B-O bondss.[41]
w a should

33 
 
The reason for th d B-O-H peaak intensity sseen near 32200 cm-1 beffore washingg (2-
he decreased

13(b
b)), and afterr washing (2--13(c)) may be due to thhe dryness off the samplees, but storinng in

a vaccuum oven prior

p to anallysis did nott appear to hhave an effecct. Alternativvely, the losss of

peak
k area can allso be attribu
uted to the loss
l of smalll but very hhighly functiionalized shheets

durin
ng the washiing process.

Figu
ure 2-13. FT
TIR traces off hBN at diffferent stages of preparatiion: (a) pristtine hBN beffore

heating, (b) hB
BN after heaating but beffore being pllaced in wateer, (c) BNO after washinng

with water,, and (d) BN

NO that has been
b washedd by water annd reacted w
with phenyl

isocyanate

2.5
5 Reacttivity off BNO

34 
 
To demonstrate
d 1 ml phenyyl isocyanatee (99%, Acroos organics) and
the reactivitty of BNO, 10

3 drrops triethyll amine waas added to 20 mg preecipitated B

BNO (the laarger, not fu
fully

exfo
oliated sheets) after it had
h been waater washedd and complletely dried.. After reacting

undeer nitrogen at he reaction pproduct was rrepeatedly rinsed with T

a 100 °C forr one day, th THF

and dried under vacuum. FT

TIR of the prroduct is shoown in Figurre 2-13(d). T
The O-H streetch

peak
ks correspon
nd to hydro
oxyl groups completelyy disappeareed after the reaction siince

isocy
yanate is veery reactive towards hyd
droxyl groupps. From thhe FESEM image shownn in

Figu
ure 2-14, thee sheets app
pear to be pulled
p backk and foldedd as a resullt of the addded

funcctional group
ps. This is analogous to what is oobserved whhen graphitee is oxidizedd to

GO.[80]

Figu
ure 2-14. BN
NO precipitaate after reacction with phhenyl isocyaanate with thhe curling of the

edges. The scale

s bar is oone micron.

d-spacing betweeen sheets laayers are meeasured by X

X-ray diffracction (XRD)) with a Bruuker

AXS
S D2 Phaserr. A well-kno
own turbostatic structurre of graphitte has a paraallel layered but

35 
 
rand
domly stackeed structure with
w an interrlayer spacinng of 3.44 Å
Å, considerabbly greater tthan

ideall graphite (3.35 Å). Similarly, thee XRD specctrum of thee isocyanatee functionaliized

BNO
O is shown in Figure 2-1
15, where th
he XRD tracee of pristinee hBN is supperimposed w
with

the trace
t of isoccyanate funcctionalized BNO.
B After reaction witth the isocyaanate, the X
XRD

profi
files of the 002 reflection
n moves to a slightly low
wer angle of 26.41°, a llarger d spaccing

of 3..37 Å, and broadens com

mpared with pristine hBN
N (26.79°, 3..32 Å).

Figure 2-1
15. XRD of pristine
p hBN
N and isocyaanate functioonalized hBN
N with the

functio
onalized material display
ying a broadeer peak shiftted to lower angle.

2.6
6 Concllusions
In th
his chapter, the formatiion of the hBN
h analog of GO by heating in aair followedd by

treattment with hot

h water is described.
d Characterizat
C tion of the foormed BNO was perform
med

and a mechanism
m of this reaction is pro
oposed. Justt as the use of GO has allowed for the

36 
 
utilization of graphitic material in numerous applications, the availability of large

quantities of exfoliated, functionalized BNO is expected to allow for the incorporation of

BNO as a multifunctional nanofiller and other potential applications. With this method it

will be possible to fully utilize the thermal conductivity, chemical inertness and electrical

insulating properties of this promising material.

37 
 
 Chapter 3 BNO/polymer and

fBNO/polymer composites

3.1 Overview
The excellent mechanical and thermal properties of hBN have stimulated studies on their

polymeric composites.[81] However, another significant property, which sets hBN above

graphite for many applications, flame retardency has not drawn enough attention. Thus,

BNNS are very promising reinforcing nanofiller for engineering plastics as a flame

retardant. BNNS may be useful to achieve both improved mechanical properties and

thermal stability improvement with the preservation of the electrical insulation and optical

transparency of a polymer. The BN filled composites might be especially useful for

applications where it is important to have at least some of these properties such as building

materials (thermal stability and good mechanical properties).

The main goal of this chapter is discussing the use of BNO and polymer functionalized

BNO as fillers to develop polymeric composites with improved mechanical properties and

flame retardency. In order to obtain composites with desired properties, melt mixing and

solution mixing are employed. Chapter 3.2 discusses the BNO/polymer composites and

their flammability. Chapter 3.3 focuses on grafting polymer chains from BNO by atom

transfer radical polymerization (ATRP) and characterization. Chapter 3.4 includes

properties of composite filled with the polymer grafted BNO nanofillers are tested and

compared with neat polymer.

38 
 
3.2 BNO/Polycarbonate nanocomposites
Two grades of hBN with different flake sizes are used as starting materials to prepare the

BNO for BNO/PC composites. SEM images show their different lateral sizes in Figure 1-6.

Two BNO materials are prepared by the methods described in Section 2.2 from these two

grades of hBN respectively.

3.2.1 Preparation of BNO/PC nanocomposites

The BNO/PC nanocomposites were prepared by melt-blending 20 grams PC pellets with

0.2 gram BNO powders in a Brabender mixer. Two nanocomposites were prepared

separately from two BNOs (small flake size and large flake size). At 230 °C processing

temperature, the mixing time was set for 2 hours. The rotation rate selected was 40

rev/min. As control groups, two nanocomposites with untreated pristine small flake size

and large flake size hBN were prepared using the exact same procedure. A picture (Figure

3-1) of these composites is shown below.

39 
 
 Fig
gure 3-1. Piccture of PC nanocompos
n sites filled w
with small flaake pristine hhBN (top lefft),

larrge flake prisstine hBN (to

op right), sm
mall flake BN
NO (bottom left) and larrge flake BN
NO

(b
bottom right)
t)

3.2
2.2 Charracteriza
ation of BNO/P
PC nanocomposites
The dispersity of
o BNO in water was studied in previous chhapters. To investigate the

dispeersity of BN
NO in a polym
mer matrix, d spacing beetween indivvidual BN shheets in the ffour

nano
ocomposites were charracterized by
y XRD. Reesults are sshown in F
Figure 3-2. As

discu
ussed with Figure
F 2-15 in chapter 2,
2 unexfoliatted hBN exhhibits an inteense peak att 2θ

value of ~26.7°.. On the oth

her hand, thee XRD patteerns of nanoocomposites exhibit a brroad

characteristic peeak of PC matrix

m The loading of hBN orr BNO in eeach
at 2θθ = ~17°. T

nano
ocomposite and
a the processing proceedure is the ssame. The ddecrease of ppeak intensityy of

BN d spacing peeak is due to

o the exfoliaation and ranndom distribbution of nannosheets witthin

40 
 
the polymer
p maatrix. Interestingly, for small flakee size BN, hydroxyl grroups and hheat

treattment does not

n seem to change
c the dispersity
d siggnificantly. T
This is probaably becausee the

interraction betw
ween sheets is
i relatively
y weak, whicch leads to well dispersed fillers eeven

with
hout function
nalization an
nd exfoliation
n. In contrasst, large flakee unexfoliateed pristine hhBN

cann
not be disperrsed well durring processing. The hugge peak at 266.7° (red linne in Figure 33-2)

show
ws the largee amount of
o aggregateed, unexfoli ated BN fillers in thee composite. In

contrast, a therm
mally treated
d large flak
ke BNO/PC composite ((purple line in Figure 33-2)

doess not show an

n intense staacking peak.

Figu
ure 3-2. X-raay diffraction
n profiles off PC nanocom
mposites fillled with smaall flake prisstine

hBN
N (blue), large flake prisstine hBN (reed), small flaake BNO (ggreen) and laarge flake BN
NO

(purple)

3.2
2.3 Flam
mmability test off BNO/P
PC comp
posites

41 
 
Flam
me retardanccy is importaant in reduciing the impaact of fires oon people, pproperty andd the

environment. Th
he outstandiing high tem
mperature sttability of hhBN makes it a promissing

didate for prroducing maaterials with

cand h improved flame retarrdancy alongg with supeerior

mech
hanical prop
perties.

To study
s the flaame retarden
ncy of BNO
O/PC compoosites, materrials with higgher loadingg of

BNO
O were prepaared. Melt mixed
m 25 graams of PC annd 2.5 gram
ms of BNO (llarge flake ssize)

comp
posite was tested and compared with neat P
PC as conttrol group. Specimens are

proccessed by ho
ot pressing and
a cut into the same di mensions. F
Figure 3-3 demonstrates the

self--extinguishin
ng performaance of BNO
O/PC compoosites. Figuure 3-4 show
ws the hBN
N/PC

comp
posite dripped with fire and the lack
k of self-extinnguishing.

Figure 3-3. Snapshots

S at different tim
me of a videoo recorded shhows BNO/P
PC strip selff-

extinguiished after iggnition.

42 
 
 Fig
gure 3-4. Sn
napshots at different
d timee of a video recorded shows neat PC
C strip drippeed

with fire

3.3
3 PMM
MA funcctionalizzed BN
NO
43 
 
3.3.1 Introduction of polymer grafted BNO
As described in Chapter 1.3 a lot of previous studies of BNNS/polymer composites rely on

extended mechanical treatment (tip sonication) of the BN to produce exfoliated material at

very high dilution, and do not utilize chemical modification to aid in dispersion. Although

the chemistry of BNNS has correspondingly been developed to extend their applicability,

there are still many stubborn problems blocking the way toward the real applications of

BNNS composites, such as the lack of functional groups and low compatibility with

organic molecules. Chemically attaching polymer chains on BN seems to be a promising

solution.

In order to attach polymer chains to hBN, the sheets must first be functionalized and

followed by a polymerization reaction. One example we are aware of is the work of

Sainsbury[42], which described the stoichiometric addition of peroxides to BNNS that

enabled the addition of isocyanate functional groups to the hBN. These reactive groups

were then proposed to react with the polymer matrix, specifically polyurethane due to the

limited chemistry of isocyanate. This approach limits the choice of polymer and

furthermore, limits the choice of polymer matrices in which the filler can be used for.

Therefore, a new route towards the polymerization of a wide variety of monomers, leading

to a wide choice of polymer matrices is necessary.

3.3.2 Reaction to grow PMMA

44 
 
Our approach is fundamentally different than Sainsbury’s route[42]. We start with

hydroxylated hBN (BNO) produced at a large scale by the oxidation method discussed

previously. Then polymers can be grafted from the BNO sheets as outlined in Scheme 3-5.

Initiator is attached through the hydroxyl groups by first adding 4, 4’-Methylene di-p-

phenyl diisocyanate (MDI) in excess, resulting in isocyanate functionalized sheets. An

excess is used to avoid the difunctional MDI linking two sheets together. This is followed

by hydrolysis to form aniline groups. To minimize the possible formation of ureas, the

MDI functionalized BNO is added drop-wisely into excess water.

In a typical reaction, 800 mg BNO and 31 g MDI were added into 100 ml flask, 3 drops of

triethylamine was then added as catalyst. The mixture was heated to 120 °C to react

overnight. After the reaction was over, white powder was obtained by vacuum filtration,

rinsed three times with dichloromethane and dried under vacuum. 700 mg isocyanate

fBNO was dispersed into about 5 ml DMF then added drop-wise while stirring to

approximately 100 ml distilled water at room temperature. The mixture was degassed to

promote isocyanate hydrolysis and allowed to stir overnight. The white powder was

collected by vacuum filtration, rinsed several times with acetone and dried under vacuum.

The addition of α-bromoisobutyryl bromide (BiBB) then places an ATRP initiator on the

sheets. Polymerization of methyl methacrylate to form poly(methyl methacrylate)

(PMMA) from the attached initiator is accomplished. Reaction details of this grafting

reaction and the reaction to prepare neat PMMA as control have been described

45 
 
 where.[82] Th
elsew he polymer--functionalizzed sheets aare then waashed with tetrahydrofuuran

(THF
F) several tim
mes to remo
ove any poly
ymer not attaached to the ssheets.

Scheme
S 3-5. Reaction ro
oute to prepaare PMMA ggrafted BNO
O

3.3
3.3 Charracteriza
ation of PMMA
A grafted
d BNO

3.3.3.1 FTIR
R

IR trransmission measurements were con

nducted on a Nicolet 5660 instrumennt coupled w
with

SpecctraTech IR
R Plan 0044-003 microsscope. FTIR
R spectra off the grafteed material and

controls are show

wn in Figuree 3-6. The up
pper spectra,, Figure 3-6((a), shows thhe starting B
BNO

ublished woork,[54] havinng the charaacteristic peeaks

mateerial, consistent with prreviously pu

correesponding to
o B-N in-plaane stretching m-1, B-N-B oout of plane at 810 cm-1 and
g at 1370 cm

the B-O-H
B peak c -1. Figure 3-6(b) is thhe spectrum of neat PMM
k at ~3200 cm MA, shownn for

comp
parison, witth a strong peak
p 00 cm-1 from
at ~170 m the carbonnyl group annd weaker peeaks

arou m-1 and 3000

und 1150 cm 0 cm-1 assign
ned to C-O- C single boond stretchinng vibration and

w PMMA chains attacched after waashing with THF to rem

C-H stretching. The BNO with move

unattached chaiins is show

wn in Figu
ure 3-6(c). The spectrrum is dom
minated by the

N peak at 1370 cm-1, bu

characteristic BN rbonyl peak near 1700 ccm-1
ut also cont ains the carb
46 
 
obseerved in the neat PMM d washings with THF result in noo change in the
MA. Repeated

specctrum. In thee filtrate, sho

own in Figurre 3-6(d), noo BN peak iss found and the spectrum
m is

nearly identical to
t pure polymer from the control reaaction. Figurre 3-6(e) shoows the mateerial

afterr heating to 600

6 °C. The spectrum is identical too that of the sstarting BNO
O. Additionaally,

the FTIR
F revealss the reappeaarance of the B-O-H peaak at ~3200 cm-1. This ppeak is not sseen

in th
he FTIR of th
he grafted BNO,
B but is visible
v in thee BNO priorr to grafting, supportingg the

contention that the

t function
nalization reaction occurrs through tthe hydroxyll groups on the

BNO
O.

Fig
gure 3-6. FT
TIR spectra of
o (a) BNO, (b)
( neat PMM
MA, (c) PM
MMA graftedd BNO, (d) frree

polymeer (filtrate) an
nd (e) TGA residue of P
PMMA graftted BNO (residue).
47 
 
FTIR
R of other reeaction interm n be found inn a previouslly publishedd paper. [82]
mediates can

3.3.3.2 FESE
EM

valently attaached to the BNO sheetss is also suppported by SEM

The presence off PMMA cov

ges. FESEM
imag M measuremeents were performed ussing a JEOL
L JSM-63355F cold cathhode

field
d emission (1
12 kV) scann
ning electron
n microscopee.

The functionalizzation of the sheets with

h ATRP initiiator makes no observabble differencce in

the sheet
s morpho
ology as sho
ow in Figure 3-7.

Fig
gure 3-7. SEM of BiBB functionaliz ed BNNS (innitiator fBN
NO)

48 
 
 MMA as seen in
Afteer the polymeerization reaaction, the grrafted BNO sheets are ccoated by PM

SEM
M images (F NO sheets aappear to hhave a contiinuous polymer
Figure 3-8). Grafted BN

coatiing with sm
mooth edges and
a lumpy surface
s struccture in conttrast to the ssharp edges and

flat surface
s of th
he starting BN
NO.

a 
a

b 
b

49 
 
 c 
c

Figure 3-8. SEM of (aa) BNO and

d PMMA graafted BNO att low (b) andd high (c)

magn
nification. Th
he scale barss are one micron

In order to conffirm the mo

orphology ch
hanges are nnot simply ddue to physsically adsorrbed

poly
ymer, 50/50 by mass mixture
m of BNO
B and neeat PMMA in THF is precipitatedd by

meth
hanol. As sh
hown in Figure 3-9, thee BNO morpphology is ssimilar to thhe starting B
BNO

(Figu
ure 3-8(a)), suggesting that polym
mer does nott adsorb onn the sheet ssurface withhout

covaalent bond. The sharp edges and smooth surrfaces in Fiigure 3-9 aare significantly

diffeerent with the morpholog

gy of PMMA
A grafted BN
NO in Figuree 3-8(c).

50 
 
 Figure 3-9
9. SEM of the mixture off BNO and P
PMMA dissoolved in THF
F and then

precipitated
d out by meth
hanol. The sscale bar is oone micron.

3.3.3.3 TGA
A

mer grafted tto the sheets, TGA was ccarried out oon a
In orrder to deterrmine the maass of polym

TA Instruments
I Q 500 Therrmogravimetric Analyzeer at a heatinng rate of 10 °C min-1 in a

platiinum pan. Reesults are sh

hown in Figu
ure 3-10. As the base BN
NO is stable in air up to 800

°C, any
a weight loss
l below that
t temperaature is due tto the degraadation of PM
MMA. We ffind

that 53% by masss of the grafted and wasshed materiaal is PMMA

A. For compaarison, Figurre 3-

10 also shows th
hat no mass loss
l is obserrved with BN
NO and puree PMMA dissplays compplete

masss loss.

51 
 
 Figurre 3-10. TGA
A of BNO, PMMA
P MA[82]
graftted BNO andd neat PMM

The TGA residu

ue of PMMA
A grafted BN
NO was alsso studied byy SEM. Figuure 3-11 shoows

nano
osheets with the typical morphology
y of BNO, w
which also cconfirms thee degradationn of

all PMMA
P chain
ns after heatiing up the materials
m to 800 °C.

Figure
F 3-11. SEM of TG
GA residue oof PMMA grrafted BNO

52 
 
3.3.3.4 Gel permeation chromatography

During the wash of PMMA grafted materials, unattached polymer is dissolved in THF and

analyzed by gel permeation chromatography (GPC) to estimate the molecular weight of the

PMMA chains attached. Gel permeation chromatography (GPC) was performed using a

Waters 717 plus autosampler with THF as eluent, a Waters 1515 isocratic HPLC pump, a

Jordi flash gel DVB column, and a Varian 380-LC ELS detector. Spectra indicate a

molecular weight of 174k and polydispersity of 2.52 (Table 3-12). As a control, PMMA is

prepared by the ATRP of MMA using the same procedure but without the sheets, giving a

MW of 199 K and a PDI of 1.15. The high polydispersity in the BNO system compared

with the control PMMA reflects the effect of growing from nanosheets versus from

solution, and similar result has been observed when growing polymer chains from

graphene nanosheets.[83]

Table 3-12. GPC results of neat PMMA as control group and free polymer from grafted

BNO are shown below.

Retention Time (min) Mn Mw Polydispersity

Neat PMMA 16.28 1.99 × 105 2.30× 105 1.16

Free polymer 15.16 1.74 × 105 4.38× 105 2.52

53 
 
 Fig
gure 3-13. GPC
G traces off (a) neat PM
MMA as conntrol and (b) free polymeer in the filtraate

of productt from ATRP

P reaction

3.4
4 PMM
MA fBNO
O/PMM
MA com
mpositee

3.4
4.1 Comp
posite preparat
p tion
54 
 
To prepare the PMMA composites filled with PMMA grafted BNO, 0.05g PMMA grafted

BNO was added to THF followed by bath sonication for 15 min. 0.95 g PMMA (molecular

weight 540k) was then added and stirred for 3 days, allowed to sit overnight to let large

particles to settle down. At last, the solution was cast onto a glass plate. The obtained films

are dried at room temperature for two days and at 90 °C for two days. The pBN filled

PMMA composites are prepared by the same procedure but using 0.025 g pBN and 0.975 g

PMMA. The weight percentage of grafted BNO in PMMA is doubled to maintain the same

weight percentage of the portion of BN sheets. Neat PMMA films are prepared without

any BN.

3.4.2 Dispersity of sheets in water

The attachment of polymer chains to the sheets is also seen to result in differences in the

suspendablity of the sheets in liquid and their dispersibility in a polymer matrix as

compared to controls. It has been shown in a previous chapter (2.3.1) that BNO forms a

stable suspension in water, while hBN precipitates.[54] In Figure 3-14, two samples are

compared in terms of their ability to form stable suspensions in water. The control on the

left is an aqueous suspension of BNO that had been mixed with PMMA in THF,

precipitated in methanol, dried, then bath sonicated in water. On the right is an equal mass

of BNO that is grafted with PMMA and sonicated in the same manner the control sample

in water. Simply mixing with PMMA does not change the hydrophilicity of BNO; after

sitting overnight, the sheets are still suspended in water with the PMMA particles

precipitated out. Conversely, the grafted BNO is hydrophobic and settles out of

suspension. The clear solution implies the PMMA is chemically attached to BNO.

55 
 
 Fiigure 3-14. Water
W suspension of BN
NO (left) andd an equal maass of BNO that has beeen

w PMMA ((right)[82]
grafted with

The improved dispersion off the nanofilller in a polyymer matrix is shown vissually in Figgure

3-15
5. BNO with
hout polymeer grafting and
a PMMA grafted BN
NO are added to PMMA
A in

THF
F and cast on
nto glass slid
des. The agg
gregated inhoomogeneouss composite on the left iis in

sharp
p contrast to
o the uniform
m dispersion of the grafteed BNO on tthe right.

56 
 
 Figure 3-15. Films of equ
ual thicknesss of hBN fillled PMMA ((left) and PM
MMA graftedd

BNO filleed PMMA ((right)[82]

Thicckness measu
urement of films
f in Figure 3-15 weere done to m
make sure thhe differencce is

not due
d to different thickneesses. The th
hin film sam
mple thickneess was deteermined usinng a

thick
kness gauge (Model LE1
1000-2, MeaasureItAll) ass the averagee of three daata points.

Table
T 3-16. Thicknesses
T m sample in Figure 3-155
s of three datta points of eeach thin film

com
mposites T
Thickness(mm
m)

grafted BNO
B /PMMA
A 0.03
3695 0.04505 0.033130

hBN
N/PMMA 0.03
3255 0.03935 0.033280

3.4
4.3 Mech
hanical propert
p ties

57 
 
To demonstrate the improved dispersability of the PMMA grafted BNO leading to superior

mechanical properties in composites, PMMA composites with 1% (w/w) loadings of

PMMA grafted BNO are prepared. Two control samples are also prepared using the same

procedure; one with neat PMMA without any filler and the other with 0.5% (w/w)

unfunctionalized pristine hBN, maintaining the same weight percentage of BN as in the

grafted BNO/PMMA composite. Tensile tests are then performed on a computer interfaced

Instron-1011 with a 500 N load cell. The rate of crosshead motion was 1 mm/min with the

data acquisition rate of 10 points per second. Results are shown in Table 3-17 and figure 3-

18. Compared with blank PMMA, the grafted BNO filled samples show improvements of

Young’s modulus, elongation and toughness. The effect of ungrafted hBN on the

properties of the PMMA is consistent with results found by others[32] in which the addition

of hBN to thermoplastic polyurethane leads to an increase in modulus with a

corresponding decrease in extension at break.

Table 3-17. Tensile test result of neat PMMA, hBN filled composite and grafted BNO

filled composites

Toughness Young’s modulus

Elongation (%)
(MJ/m3) (MPa)

Neat PMMA 8.0±3.6 739.3±47.9 24.2±10.8

hBN filled 6.7±2.5 793.6±65.6 18.9±6.7

Grafted BNO filled 13.5±4.0 802.8±66.7 34.4±8.3

The improved modulus of grafted BNO filled composite can be explained by three effects.

First, this is a result of the high modulus along the sheet direction of hBN.[84] The force

58 
 
required to deform hBN sheets that are not perpendicular to the stretching direction

contributes to the overall force needed to stretch the composite. That the modulus benefits

from the nature of hBN, even when not exfoliated or well dispersed, is known from studies

of graphene nanocomposites.[4] Such fillers result in increased Young’s modulus even

while reducing the elongation and toughness of the materials.[85] The positive effect on

Young’s modulus is diminished in the control hBN composites compared to the well-

dispersed gBN sheets with grafted polymer chains, which can be explained by the second

effect. When aggregation of hBN fillers happens in the composite, as shown in Figure 3-

15, stacked sheets may slide over each other, acting as lubricants and preventing further

increase of modulus. Last but not least, the third effect can be attributed to the interactions

between fillers and the polymer matrix. The attached chains are expected to entangle with

the chains in matrix and increase the interaction of the nanofillers with the matrix,

improving dispersion and more importantly, strengthening the filler/matrix interface. The

positive effect only exists in the grafted BNO filled composites. Therefore, the mechanism

of strengthening is a combination of the high modulus of BN, well dispersed fillers and

increased strength of the filler/matrix interface. Similar results and conclusions have been

reported in other filler systems.[42]

59 
 
 Figure
F 3-18.. Typical streess-strain cu PMMA, graafted BNO filled PMMA
urve of neat P A

compositee and pristin

ne hBN filledd PMMA coomposite.

The improvemen
nt in elongaation can be explained bby a similar mechanism. The improoved

dispeersity of graafted BNO nanosheets an

nd the interaaction of the PMMA maatrix and graafted

sheeets leads to larger deform

mation prior to ultimate failure, as sshown in Tabble 3-17, whhere

we observe
o greaater elongatio
on for the PM
MMA/grafteed sheets com
mposites com
mpared withh the

PMM
MA/hBN co
omposites. The
T grafted filler increeases elongaation while the hBN fi
filler

actuaally reducess elongation

n. This is primarily
p duue to the sstacked sheeets and clearly

demonstrates a fundamental
f difference arising
a from grafted polyymer chains on the sheetts.

Finaally, the toughness of th

he composittes is calcullated as thee work per unit volumee of

ples, which is the total area under the stress/sstrain curve.. As shown in Table 3-17,
samp

graftted BNO enh

hances PMM
MA considerrably while hhBN reducedd the toughnness of PMM
MA.

60 
 
3.5 Conclusions
In this chapter, two methods to prepare BNO/polymer nanocomposites are discussed. The

first method produces composites by direct melt processing BNO/polymer mixture shows

improved dispersity of BNO in nanocomposites compared with hBN. The flammability of

obtained BNO/PC nanocomposites is improved. The second method was achieved by

grafting PMMA chains from BNO by addition polymerization. Significant increases in

Young’s modulus and elongation are observed relative to neat polymer and

unfunctionalized hBN filler does not provide the same improvement. Those two

approaches are easily scalable, and open the door for the application of this new 2D

nanofillers.

61 
 
 Chapter 4 Interfacial Exfoliation of BN

4.1 Introduction
As pointed out in previous chapters, BNNS with extraordinary thin-layered structure

would be beneficial for applications for electronics.[12, 16-20]

Chemical functionalization

methods for this application of BNNS are limited because the electrical properties of

BNNS may be affected due to the presence of functional groups. The desirable
[16,
unfunctionalized and exfoliated BNNS can be produced by chemical vapor deposition
43]
with high purity and controlled orientation, but the high cost and limited product

quantity remains a challenge. In order to fully utilize the extraordinary properties of

BNNS, an effective and cheap method to produce pristine BN sheets and their composites

is required.

In this chapter, we describe a simple but effective method to produce continuous and

uniform few-layer hBN sheet networks utilizing an interfacial trapping technique. The

morphology and structure of the few-layer sheets network were analyzed by several

characterization techniques.

4.2 Interfacial exfoliation

It was noticed that when placed into a phase-separated mixture of water and organic

solvent, hBN stays at the interface stably, similar to graphite exfoliation.[22] This can be

explained by the incapability of hBN with either of these two phases and the system is also

62 
 
stabilized by the effect of minimizing the high-energy interface of two immiscible liquids.

As a result, hBN’s tendency of spreading and occupying the whole interfacial area

becomes the main driving force of exfoliating hBN to few-layer sheets and sliding BN

sheets along the interface.

The typical procedure of preparing few-layer BN sheets films is described as follows: 2mg

of pristine hBN (from UK Abrasives, Inc.) was first put into a 20 ml scintillation vial. 5ml

organic solvent (Heptane as in Figure 4-1(A, B and C)) and 5ml DI water were then added

and the system was bath sonicated for about 15 seconds to break up large particles. Tip

sonication was then used for 15 minutes at 40% power amplitude. After the sonication, the

vial sat until both water layer and organic layer became clear to allow the interfacial

trapping and self-assembly to stabilize at the interface. After that, once the vial is shaken, a

continuous white film can be observed on the wall of glass vial, or a glass slide put into the

vial. After the glass slide was coated by BN sheets, it was taken out and dried immediately

at 70 ºC. To maintain the integrity of the sheet network, any contact of the surface with BN

sheets should be avoided.

4.2.1 Sonication VS no sonication

This spreading process is spontaneous, and if left alone for enough time, the BN will

spread at the interface with no sonication or mechanical mixing. This is illustrated in

Figure 4-1(A), (B) and (C), where the initial mixture is shown in Figure 4-1(B). The BN

sheets climb the hydrophilic walls of the glass vial in both Figure 4-1(A) and (C), but the

process is much faster with external energy input. Sonication has been proven to be a

63 
 
conv
venient and effective
e waay to shorten the self-exffoliation timee to ~30 minnute (Figure 4-1

(A)) for a typicaal water/orgaanic solvent system in a 20 ml scinttillation vial compared w

with

seveeral days witthout sonicattion (Figure 4-1 (C)). As shown in F

Figure 4-1 (C), the BN w
will

clim
mb even without sonicatio
on, but the process
p is veery slow. In a typical expperiment, w
water

and heptane aree used alon

ng with brieef bath soniication to sppeed up thee process. T
This

behaavior is analo or graphite iin previous sstudies.[22, 86]]

ogous to thatt observed fo

Figu
ure 4-1. (A) Heptane/waater mixture with added BN followinng brief bathh sonication (B)

Waater/heptane//BN initial saample and (C

C) Water/hepptane/BN affter sitting foor 25 days w
with

no
o sonicationn.

4.2
2.2 Interrfacial ex
xfoliatio
on with differen
nt solven
nts
One interesting phenomena
p is that this self-exfoliatioon happens on the wall of the glass vial

ve the liquid
abov d, which sug
ggests the in
nterface alsoo present at a hydrophilic glass surfface

with
h a thin layerr of water/m
moisture on glass
g above tthe liquid. T
This thin layeer of water iis in

contact with thee organic sollvent, so thaat exfoliatedd BN sheets also coat thhat part of gglass

and form the sam

me few-layerr thin, as well as a continnuous and unniform film at the interfa
face.

Figu
ure 4-2(left) shows a scintillation vial with chhloroform ((HPLC gradde, from Fissher

64 
 
Chem
mical), wateer and BN after
a sonicattion, in whicch the wholle interface outside orgaanic

layerr was coated

d with few-laayer BN sheeets. If an org
rganic solvennt with a dennsity lower tthan

wateer, e.g. heptaane (anhydrous, from Alfa

A Aesar) iis used, BN
N sheets “climb” up agaainst

grav
vity onto thee glass wall since the heptane
h vapoor filled the atmospheree in the viall, as

show
wn in Figuree 4-2 (right)). It was also
o observed that BN aggglomerates aand precipitates

throu
ugh water layer
l after the
t heptane is evaporatted. This suuggests thatt the interfaacial

trapp
ping is not siimply due to
o the density
y difference.

gure 4-2. Picctures of exffoliated few-layer BN shheets at interffaces of twoo water/organnic

Fig

solvent systems.
s Thee organic useed are chlorooform(left) aand n-heptanne(right)

4.2
2.3 Preparation of BN on
o glass slide
Few-layer BN sh
heets on a glass
g slide caan be preparred by puttinng a glass sliide into the vial

and shaking the vial with th

he cap closed
d. After clim
mbing, the slide is removved and drieed at

70 °C
C, resulting in a dry BN film as show
wn in Figuree 4-3.

65 
 
 Figure 4-3.
4 Dried feew-layer BN
N film on glaass slide

4.3
3 Moleccular Dynamic
D cs Simu
ulation
Thiss work is do
one by Prof. Andrey V. Dobrynin’ss group

To understand
u the
t mechansim of hBN
N sheets intterfacal spreeading, all aatom molecuular

dynaamics simullations of BN
B sheets at
a the interrface betweeen water annd heptane are

perfo
omed. Figurre 4-4 show
ws the distrib
bution functtion of the atoms belonnging to waater,

heptane and BN
N sheets obttained from
m molecular dynamics ssimulations of the systeems

containing a single sheet, fo

our sheets, and
a nine sheeets of BN. F
For multi-shheet simulatiions

we observe
o the aggregation of the BN sheets
s and fformation off a BN sheett network att the

interrface betweeen water and

d heptane (ssee insets inn Figure 4-44). Formatioon of the muulti-

sheeet aggregatess is indicateed by the ap

ppearance off the multiplle peaks in the distribution

66 
 
funcction seen in Figure 4-4. Location off the origin oof the z-axiss is set to thee location off the

poin
nt where the number fracction of atom
ms belongingg to water m
molecules ouutside the reggion

occu
upied by the BN sheets decreases
d y half from itts bulk valuee, wat =0.5. It is interesting
by

to no
ote that the single BN sheet residees close to tthe water/heeptane interfface, but in the

heptane phase, while

w sheetss belonging to multi-sheeet aggregattes penetratee into the w
water

phasse in order to
o allow accom
mmodation of the assem
mbly in the innterface regiion.

Fig
gure 4-4. Nu
umber fractio
on distributions of atoms belonging to water moolecules (dassh-

dot
d lines), heeptane moleccules (dot lin
nes) and BN
N sheets (solid lines) alonng the z-axiss

no
ormal to the water/heptan
ne interface for water/heeptane mixtuures containiing single BN
N

sh
heet (black lines), four BN
B sheets (reed lines), andd nine BN shheets (blue llines). Insertts

show
w typical con
nfigurations of the BN laayers at the w
water/heptanne interface. Solvent shoown

is wateer, and heptaane moleculees are transpparent.

67 
 
In order to establish the affinity of BN sheets to the water/heptane interface, the potential

of the mean force along the z-axis normal to the water/heptane interface was calculated

(Figure 4-5) by using the Weighted Histogram Analysis Method. The minimum of the

potential is located in the heptane phase, ~6 Å away from the interface, indicating the

preference of BN for heptane rather than water. The increase of the potential in the heptane

phase is not as steep as that observed in the water phase, an indication that heptane is a

better solvent for BN than is water, although BN is insoluble in both of these two solvents.

The magnitude of the potential in the plateau regime is on the order of 1.8 mN/m. This free

energy change is lower than the corresponding 2.2 mN/m change of the free energy

obtained for graphene displacement from an interface using similar calculations.[22, 86]

Using this value, we estimate the work required to displace BN sheets with 100 X 100 nm

lateral dimensions from the water/heptane interface into the heptane phase to be 4420 kBT.

This energy is sufficiently strong to localize the sheets at the water/heptane interface.

68 
 
 Figure
F 4-5. Variation
V off the potentiaal of the meaan force norm
malized by thhe BN sheett

su
urface area allong the z-ax
xis normal to
o the water/hheptane interrface. Insetss show typicaal

sheet co
onfigurationss. Heptane pphase is transsparent.

4.4
4 Chara
acteriza
ation off BN neetwork

4.4
4.1 Raman specttroscopy
y
The presence off single layerr sheet in th
he network w
was confirm
med by Ramaan spectrosccopy

on a Renishaw
w Ranascop
pe System 2000 operaating at 5114 nm. Thhe monolayyer’s

c -1 was observed (Figuure 4-6). This is consisteent with repports

characteristic peeak at 1369 cm

in litterature.[68]

69 
 
Figu
ure 4-6. Ram
man spectra of
o bulk hBN
N and few-layyer BN sheetts network sshow changees in

inteegrated inten
nsity and the position of ppeak due to the exfoliatiion

4.4
4.2 Morp
phology study
To characterize
c the morpho
ology of the few-layer B
BN film, thee film was ttransferred tto a

SEM
M stub and TEM
T grid by o method ddescribed ellsewhere[22]. Figure 4-7 is a
y a floating off

repreesentative lo
ow magnifiication SEM
M image, w
which showss a continuous BN sheets

netw
work compossed of paralleel packed flaat sheets with
th lateral sizee of 1 to 4 m
microns.

70 
 
 Figu
ure 4-7. Low
w-magnificaation SEM im
mage of BN sheets netwoork

As SEM
S on of sheetss morphologyy at low maagnification, TEM was uused
providees informatio

to fu
urther undersstand the mo
orphology off the BN film
m, especiallyy at high maggnification. The

film is observed
d to be comp
posed of few
w empty spacces containinng no BN shheets as poinnted

out in
i Figure 4--8(a) and few
w-layer sheeets overlappping as show
wn in Figuree 4-8(b). It aalso

need
ds to be poiinted out thaat, even tho
ough the sheeets are staccked, offset is observedd in

betw
ween them, which imp
plies that sh
heets sprea d and slidee over eachh other at the

nterface. Thiss mobility is lost after thhe sample waas dried.

wateer/organic in

71 
 
 a 

b 

Figure 4-8. TE
EM images of
o BN sheetss network diisplaying few
w-layer sheeets stacks witth

offset and empty sspaces

Elecctron diffracttion patternss in Figure 4-9 reveal thee typical six-fold symmeetry structurre of

BN sheets.
s

72 
 
 Figure 4-9. Electron-diffraction patttern revealinng six-fold ssymmetry off BN sheets

The area of em
mpty spaces is roughly 10% of thee whole areea, which w
was obtainedd by

analy
yzing a low magnificatio
on TEM imaage. (Figure 4-10)

Figure 4-10. Low magniffication TEM

M of BN sheeets network where an em
mpty space iis

point ou
ut by a blackk arrow
73 
 
4.5 Conclusions
In this chapter, we have demonstrated the formation of a uniform and continuous BNNS

film by interfacial trapping technique. Compared to chemical methods, this exfoliation

method does not involve any covalent bond or non-covalent bond modification. Therefore,

this simple and inexpensive deposition technique is applicable to utilize the excellent

electrical properties of pristine BN sheets without any functional groups acting as defects.

Furthermore, the morphology of the sheets network was also characterized. The unique,

highly oriented sheets network is comparable with CVD but can be produced with much

lower cost and better scalability. This work paves the way for BN sheets’ application in

electronics.

74 
 
 Chapter 5 BN/polymer thin films with

improved dielectric properties

5.1 Introduction
In chapter 4, a simple and efficient method to exfoliate hBN into BN sheets was described.

To incorporate the sheet network with its structure unaffected, a subsequent processing

method was developed. However, without functional groups on BNNS, BNNS’s limited

solubility and difficulties in manipulation with polymer matrix have imposed challenges

for composites preparation. In this chapter two processing methods are used to

incorporated sheets network into polymer film with unique orientated BNNS interlayer

configuration. Obtained nanocomposites displayed remarkably improved dielectric

properties, providing a new route of polymer nanocomposites for dielectric applications.

5.2 Solution casting method

Since the mobility of the BN sheet network is lessened after drying, we expected that when

being incorporated into polymer matrix, the morphology of sheets network would remain

unaffected. To cast polymer solution, the BN film on glass substrate needs to be dried

thoroughly. Aqueous solution of polymer was first tried but ended up tearing apart the

secured network and agglomerating exfoliated sheets. Alternatively, organic solvent

tetrahydrofuran (THF) was found to be effective and inert with respect to the BN film

structure. The different result of using water and THF is presumably attributed to both

75 
 
denssity and surrface tension
n. To prepaare a BN c oated PMM
MA film, 0.005g/ml PMM
MA

(app
prox. Mw: 54
40,000 from
m Scientific Polymer
P Prooducts, Inc.)) in THF sollution was ddrop

cast on a glass slide with a BN sheett network. A

After THF eevaporated, the BN coaated

MA film wass first dried at 50 °C oveernight then ccould be peeeled off from
PMM m the glass sslide

with
h caution. After
A drying under vacu
uum overniight, the peeeled film w
with BN sheets

incorporated in it was obtaiined. After this

t proceduure (illustrateed in Schem
me 5-1 (a)), BN

sheeets were fou mer film aftter peeling ooff. But thee sheets are not
und to “coatt” the polym

coatiing outside the polymerr, they actuaally have intterlayer strucctures as shoown in Scheeme

5-1 (a).
( Details of the structture of BN coated
c PMM
MA film cann be found inn chapter 5.4.1.

Figu
ure 5-2 show
ws a picture of the peeled off few-laayer BN coaated PMMA film on a gglass

platee.

Scheme
S 5-1. Schematic illusion
i of prreparation o f few-layer B
BN coated ppolymer film
m

76 
 
 Figure 5-2.. Photograph
h of PMMA film “coatedd” with BN

5.3
3 Heat process
p sing
To further
f simp
plify this procedure an
nd to expannd the selecction of polymers, a m
melt

proccessing proceedure was developed.

d This
T is impoortant for tw
wo reasons; the first onne is

becaause solution
n casting method
m invollves organicc solvents aand the com
mplexity of this

proccedure can be decreased by using a solvent-free

s method. Thhe second reason is solution

castiing limits the selection of

o polymers.. Only polym
mers solublee in specific organic solvvent

can be
b used. Pollypropylene (PP), which
h has low sollubility in coommon orgaanic solventss, is

the most-used
m diielectric film
m material in
n industrial ccapacitors. H
However, thhe low dielecctric

consstant of PP (2.2)
( leaves a lot of spacce to improvve the perforrmance of itts devices. M
Melt

proccessing was successfully

y applied to coat a PP ffilm. In a tyypical experiiment, PP fiilms

with
h different th
hickness werre prepared by
b hot press ing at 155°C
C. The thickkness of the thin

filmss below 0.5 mm is prefe

ferred for thrree reasons. First, the laarge mass off thick PP fiilms

may damage th
he BN sheets network. Second, iff they are ttoo thick, im
mprovementt of

77 
 
dieleectric constaant is diminiished due to
o the large amount of bbulk materiaal. Last but not

leastt, for practical reason, veery thin thick

kness is usedd in commerrcial film cappacitors.

Befo
ore using thee BN sheets on glass plaate, half of thhe slide wass cleaned byy wiping off any

sheeets with papeer tissue and

d acetone to provide a hheat-treated nneat PP film
m as control.. PP

(Isottactic, melt index

i 20 g/1
10 min from
m Sigma-Aldrrich Co. LL
LC) thin film
ms with different

thick
kness were prepared
p by hot pressing
g. Then the gglass plate w
was slowly hheated to 1766°C,

hot pressed
p PP thin
t film waas put on wiith cautions.. A good coontact of PP and glass sslide

overr the whole area is cruccial for the process.

p Airr bubbles beeing trappedd in betweenn, as

show
wn in Figuree 5-3, should
d be avoided
d because it was observved that BN sheets on sppots

wherre air bubblles formed were

w not em
mbedded intoo polymer. W
When the fi
film was coooled

down after anneealing for ~3

3 minutes, th
he BN netw
work was em
mbedded intoo the PP mattrix.

The neat PP arrea without BN “coatin

ng” was useed as a conntrol for dielectric consstant

meassurement because the thermal historry may affectt the dielectrric constant of the polym
mer.

ure 5-3. A picture

Figu p show
wing BN coatted PP film aafter heat prrocessing. Thhe BN sheets in

leeft part of glaass slide wass erased prio

or to heat proocessing. Thhe heat proceessed PP film
m

cleearly shows sheets

s incorp
poration did not occor inn spots wheree glass air buubbles form
med
78 
 
Since Bi-Oriented Polypropylene (BOPP), a polypropylene (PP) film is oriented in two

directions (machine and transverse directions) is the dielectric material within high-

performance and low-loss radio frequency capacitors. Effort has also been made to

incoorperate the exfoliated BN network into commercial BOPP dielectric films. It was

found that this processing method did not work with BOPP. This is probably due to the

surface coating/treatment of commercial BOPP films, which is usually for better sealability

and barrier properties. Trace amounts of calcium were found on the surface of commercial

BOPP films we used, suggesting the surface coating/treatment.

5.4 Characterization of BN/polymer film

5.4.1 Film structure

5.4.1.1 SEM

As portrayed in the side view of BN coated PMMA film in Scheme 5-1, although very

close to the surface, the BN “coating” interlayer was found to be actually embedded in the

polymer matrix. This structure was characterized by viewing the bottom side of the BN

coated PMMA film, the side that contacted the glass plate, under SEM. Blurry outlines of

flat BN sheets covered by polymer can be seen under SEM (Figure 5-4(a)). This

morphology is similar to Figure 4-7, BN sheets prior to polymer solution casting. To

confirm there is a thin layer of polymer on top of BN sheets, a high intensity electron beam

was zoomed and focused on a small area in the middle of the surface shown in Figure 5-

4(a). After exposure to high-energy electrons for less than a minute, the heat produced

79 
 
degrradation of a layer of po
olymer, reveaaling the BN
N sheets embbedded at thhe curling eddges

(Figu
ure 5-4(b)). When neat PMMA is exposed to intense elecctron beam, a square shhape

poly
ymer surface was burnt off
o without showing any sheet-like eedges, as shoown in Figurre 5-

4(c, d).

Figure
F 5-4. SEM
S of the few-layer
f BN
N coated PM
MMA film, sshowing the same regionn

before (a) and

d after (b) in
ntense electro
on beam treaatment and tthe same reggion of a neaat

PMM
MA film befo
ore (c) and after
a (d) intennse electron beam treatm
ment

The cross sectio

ons of BN coated PMM
MA films w
were also stuudied by SE
EM. Figure 5-5

show
ws the crosss section at different maagnification.. Embeddedd BNNS cann be found vvery

80 
 
closee to the surfa
face of the fillm. And the orientation of the sheetss is parallel to the thin fi
film.

Thesse images su
upport the strructure proposed and illuustrated in S
Scheme 5-1 ((a).

Fig
gure 5-5. SEM
M of the cro
oss section of few-layer B
BN coated P
PMMA film at low (A) aand

high
h (B) maagnification, showing thee embedded BN sheets vvery close to the surface
81 
 
5.4.1.2 UV-V
Vis

The difference of
o transparen
ncy between
n BN sheets film and BN
N coated PM
MMA film coould

be id
dentified witth the naked
d eye by com
mparing Figuure 5-2 and 4-3. Accordding to the U
UV-

Vis spectra in Figure

F 5-6, after
a incorporating BN sheets into polymer, thhe transpareency

imprroved by 30%
% compared
d with films of
o BN sheetss. The opticaal absorptionn below 400 nm

indiccates the edg

ge of the op g of BN ssheets.[87] Thhe low transsparency of BN
ptical band gap

sheeets films was attributed to the surfaace scatterinng owing to the surfacee roughness. As

poly
ymer layers were
w introdu
uced on both
h sides of thhe BN sheetss, scattering from the roough

BN sheets is diiminished. The

T presencee of polymeer on both sides of BN
N sheets is aalso

supp
ported by th
he observatio
on that BN sheets couldd not be wiiped from thhe film withhout

damaging the su
urface.

Figure
F 5-6. Optical
O transparency testts on neat PM
MMA film (bblue), few-laayer BN film
m

(red), and
d few-layer BN
B coated P
PMMA film (green).

82 
 
The roughness of the sheets network can be explained by the structure of sheets after

drying. Since the sheets are only a single atom thick, gravity does not have a significant

effect on the BN sheet network. Therefore, during the rapid drying process, solvent

interface disappeared and the sheet network supported its own structure since all sheets are

connected together. This support comes from those sheets contacting the glass slide,

probably at the edges of this network. The thin empty space formed between sheets and

substrate was then refilled by polymer solution, creating BN incorporated PMMA with this

unique interlayer structure.

5.4.2 Loading of BN in the thin film

Loading of filler is one of the essential parameters for composite materials.

Thermogravimetric analysis (TGA) was performed to determine the loading of BN in

sheets coated polymer films. Due to the remarkable thermal stability of BNNS

(decomposition temperatures up to 800 °C in air[43, 54, 65]), any remaining residue above the

polymer degradation temperature can be attributed to the few layer BN sheets (as shown in

Figure 5-7). But the loading could not be simply read from this measurement. It was found

that the mass of BN sheets in a practical-sized sample for TGA is much lower than the

detection limit of the instrument. Trials of measuring the mass of the glass slide before and

after BN sheets climbing did not work either. However, an estimated maximum possible

loading was calculated by assuming five-layer sheets with the lateral size equals to the film

size composed the BN interlayer, which leads to a loading slightly lower than 0.01%

(w/w). Since there are empty spaces in the sheet network, this 100% covered model has the

83 
 
maximum loadin
ng for a giv
ven number of layers. E
Even this esttimated maxximum posssible

loading is exceedingly smalller than any of the reporrted BN/polyymer compoosite systemss.[17-
19, 88,, 89]

Figure 5-7.
5 TGA of BN coated P
PP film and PP film.

Due to the special structure of

o the BN co
oated thin fillms, a controol with equaal loading off BN

but randomly
r disspersed in th
he compositee is needed. W
While the exxact loadingg of BN was still

unclear, to prepaare such a sample the BN

B coated P
PMMA film was first dissolved in T
THF

and followed by
y recasting on
o a glass slide
s with thhe same areea with BN coated film
m, as

show
wn in Scheme 5-1(b). This proceedure removves the orientation off BN interlaayer

morp
phology and
d makes a PM
MMA film with
w BN randdomly disperrsed within iit.

5.4
4.3 Dieleectric prropertiess

84 
 
A significant improvement of properties from BN filler whose loading is lower than 0.01%

(w/w) is not expected if the special morphology of BN sheets is not retained. Dielectric

properties of BN coated polymer films were investigated in terms of dielectric constant,

loss ratio and breakdown field.

5.4.3.1 Dielectric constant and loss ratio

Dielectric spectroscopies were obtained using an Agilent 4284A Precision LCR meter

averaging 5 measurements of the capacitance and dissipation over a frequency range of

100 Hz to 1 MHz. The sample thickness was determined using a thickness gauge (Model

LE1000-2, MeasureItAll) as the average of three measurements. We observed a significant

increase in the dielectric constant of few-layer BN sheet coated PMMA compared to neat

PMMA films, as shown in Figure 5-8. The loss ratio of coated materials is not affected by

the BN sheets (Figure 5-9).

The increase in dielectric constant with such a low loading can be attributed to the

presence of the continuous overlapping BN sheets that constitute the thin BN layer. Recent

studies of a BN-graphene nanocapacitor observed a significant increase (from 4.9 to 7.9) of

dielectric constant of BN sheets with number of layers from 16 to 2.[12]

Dramatically, even slight increase of dielectric constant was not observed (Figure 5-8) with

the control sample. In this composite film, the configuration of BN sheets is disrupted and

randomly distributed in the polymer matrix without orientation, compared to BN coated

PMMA film. Therefore, this effect of enhancing dielectric constant must be related to

85 
 
exfo
oliated nanossheets and th
he structure of
o the interlaayer in the ffilm, not the compositionn of

BN filler. Given
n the extremeely low volu
ume fractionn of BN sheeets in the coomposite film
m, it

ot an unantiicipated resu
is no ult that the dielectric
d coonstant of thhe compositte film overllaps

with
h neat PMMA
A.

Figure 5-8. Dielectric constant

c at ro
oom temperaature as a funnction of freequency for

unm
modified PM
MMA film (b
black squaress), BN-PMM
MA film withh dispersed B
BN sheets (rred

circles), and
a BN coatted PMMA ffilm (blue triiangles).

86 
 
 Fig
gure 5-9. Th
he loss ratio as
a a function
n of frequenccy (curves haave the samee identity as in

5-7)

Dielectric constaant and loss were also in

nvestigated on heat processed PP fi
film coated w
with

BN layer.
l Impro
ovements of the dielectriic constant bbetween 10 tto 40% weree achieved w
with

the loss
l ratio un
naffected com
mpared to neeat PP filmss treated by the exact saame method but

with
hout BN sheeets (Figure 5-10).

Figu
ure 5-10. Dieelectric consstant and loss tangents off PP and BN
N coated PP film.

87 
 
To normalize
n the effect of BN
B sheets on
n PP films w
with differennt thickness, a capacitorss-in-

seriees model (Sccheme 5-11) was applied

d.

Scheme
S 5-11
1. Capacitorss in series moodel used foor calculationn

Tab
ble 5-12. Dieelectric consstant of PP and BN coateed PP films. Each row shhows dielecttric

consstants of neaat PP and BN

N coated PP obtained froom two partss in one heat-treated sam
mple

and th
he film thickkness.

εtotal at 1kHz εpp at 1kHz T

Thickness [m
mm]

3.64
3 2.66 0.31

3.64
3 3.08 0.36

3.79
3 3.29 0.38

3.86
3 3.52 0.39

3.70
3 3.00 0.41

88 
 
Calcculations of effective cap
pacitors in series
s modell for BN coated PP film
m were donee by

the following
f eq
quations:

1 1 1
 
Ctottal CPP Caff

dtotal=dPP+daffff

A
C   0
d

dtotal dPP daff dtotal  daff daff

   
total  PP  aff  PP
P  aff

dtotal dtotal 1 daff

  daff 
 PP total  PP  aff

The data in Tablle 5-11 is fittted into this equation andd plotted in Figure 5-13

d total d total 1
Figure 5-13.
5 The plo
ot of  VS and a lineear fitting off the data
 PP  total  PP
P

A reesult of daff=0.2386mm, εaff= 4.157 was

w obtainedd. Thereforee the BN afffected layer was

foun
nd to have ~2
24 microns thickness
t witth a dielectriic constant oof 4.2.

89 
 
5.4.3.2 Brea
akdown strength
s

Breaakdown stren
ngth measurements of th
hree sampless (BN coatedd PMMA, neeat PMMA ffilm

and BN/PMMA composite film)

f were performed
p ussing a linearr voltage ram
mp generatedd by

a ressistor capacitor (RC) circuit. When the

t first breaakdown evennt occurs, thhe power suppply

is sh
hut off throu
ugh an interlo
ock input by
y a silicon c ontrolled recctifier (SCR
R) circuit, whhich

uses the breakd

down-induced ground-risse voltage ccapacitively coupled to the gate off an

R. The break
SCR kdown voltaage of the saample is reaad from a peeak-holding voltmeter. The

breakdown voltaage data is sh

hown below, in Figure 55-14.

Figure
F BN coated P
5-14. Weibull disstribution of the breakdoown field of B PMMA, neatt

PMMA film and

d BN/PMMA
A compositee film at room
m temperatuure.

90 
 
Breaakdown stren
ngth of three samples are
a plotted inn Figure 5-114. Compareed to PMMA
A, a

~48 % increase of breakdow

wn field wass observed, w
which statess the fact thaat the BN shheet

interrlayer existss as an exceellent insulaating barrierr against spaace charge conduction and

curreent leakage. In addition

n to the fact that hBN ittself exhibitss high breakkdown strenngth,

this improvemeent can also

o be attribu
uted to the highly orieented and ccontinuous BN

interrlayer structture, which is perpend

dicular to thhe directionn of electroonic field. The

multtilayer polym
mer dielectrric has alreeady been pproved to bbe beneficiaal for dielecctric

ngth,[90-92] siince the inteerfaces perp

stren pendicular too the externnal field exxist as traps for

electtron injection, where thee space charg

ges and imppurity ions caan accumulaate. Our metthod

consstructed a sim
milar config
guration with
h 2D materiial interlayerr laterally diistributed allong

the film
f directio
on, which ach
hieved increeased breakddown field aas well. Simiilar to dielecctric

consstant, the maajority of im

mprovement on
o breakdow
wn strength of BN coateed polymer ffilm

did not
n survive in BN/PMMA compositee film (Figurre 5-15).

Figu
ure 5-15. Breeakdown fieeld of neat PM
MMA, BN ccoated PMM
MA film and BN/PMMA

comp
posite film

91 
 
5.4.4 Gas barrier properties
The BN coating, in addition to improving the electrical properties of the polymer film, also

improves the polymer’s barrier properties. Without the BN coating, PMMA obtained by

drop casting from a THF solution allows the permeation of gas at the rate of 2.35 x 10-14

cm3 cm / cm2 s Pa. The presence of the BN layer, even though less than four sheets thick

and with approximately 10 % uncovered area, results in a decrease of the gas permeation to

2.85 x 10-14 cm3 cm / cm2 s Pa. This is nearly a 21 % change with a loading of less than

0.01 %. The presence of a near continuous coating of the polymer with overlapping BN

sheets seems the only plausible explanation for such a large effect from such a small

loading.

5.5 Conclusions
In this chapter, we demonstrated the preparation of a BN sheet network incorporated

polymer matrix with the morphology unaffected by the solution casting and heat

processing procedure. The BN/polymer films’ unique properties, including extremely low

loading, exfoliated and highly ordered structure improve both dielectric constant,

breakdown strength and gas barrier properties of the BN sheets coated films. The film

materials with these improved properties are promising for the applications as dielectrics in

electronics and energy storage. This coupled exfoliation and solution/heat incorporation

method also paves the way for preparation and application of other 2d materials.

92 
 
 Chapter 6 BN/PS foam and sheets coated

particles

6.1 Introduction
In chapter 5, we discussed that hBN can be exfoliated at the interface of the phase

separated mixture of water and organic solvent. To obtain a continuous and uniform

exfoliated sheets network, the amount of starting material, hBN, should be carefully

determined by experiment. While not enough hBN results in more empty spaces in the

sheets network; excess hBN may lead to either stacked sheets or the formation of excess

sheets stabilized interfaces. Through experimentation we have found that stable emulsions

were produced when excess hBN is present in the system. This is similar to the

stabilization of emulsions by pristine graphene sheets demonstrated previously.[86] One

phase, the continuous phase is outside the emulsions with the other phase occurring inside

the emulsions. To take advantages of this emulsion structure, polymerization can take

place when a monomer used as the oil in the organic phase. Once the reaction is finished,

nanocomposites with different unique structures were prepared and characterized. One

potential application of the obtained nanocomposites foam was demonstrated.

6.2 Emulsions stabilized by BNNS

BNNS stabilized emulsions can be prepared by following the same procedure of preparing

exfoliated BN “climbing” but using 0.5 g hBN (large flake size or small flake size) in a

93 
 
mixture of 10 ml water and 10 ml heptane. Due to the viscosity of the mixture increasing

significantly when using excess hBN, tip sonication does not emulsify the mixture

thoroughly. Instead, a blender (Model 33BL79) was used to emulsify and mix the mixture

for 1 minute. A water-in-organic emulsion stabilized by BN sheets can be obtained if large

flake hBN is used, in which heptane forms the continuous phase. BNNS skins stabilized

spherical water droplets. The emulsion was transferred onto a glass slide by pipettes and

studied by optical microscope. Snapshots of the video taken under the microscope are

shown in Figure 6-1. Since heptane evaporates very fast at room temperature, the video

shows spheres swell, expanded and eventually bursting. After water was released from

inside, the folded and wrinkled sheets dried on glass slide. The number on each image

denotes the elapsed time. The fact that spheres are able to expand and change the shape

implies the spheres are stabilized by BNNS instead of aggregated BN particles.

94 
 
 Figu
ure 6-1. Snaapshots at diffferent time of a video reecorded undder optical m
microscope affter

water in n-heptane sp
pheres transfferred on glaass slide

The size of the dispersed aq

queous phase spheres w
was determinned with a D
DT-100 acouustic

specctrometer fro
om Dispersiion Technollogy Inc. T
The attenuatiion spectra were analyyzed

using
g Dispersion
n Technolog f polydispeerse emulsioons.[93] The ddistribution w
gy software for was

obtained from ultrasound

u atttenuation sp ween 1 and 100
pectra in thhe frequencyy range betw

MHzz. From thee acoustic measurement

m result show
wn in Figurre 6-2, the ssize of aqueeous

spheeres ranges from

f 200 μm
m to 500 μm.. The averagge size is 321 μm while the median size

95 
 
is 32
20 μm. The fact
f that meaan size and median sizee are very cloose, which ccan also be sseen

from
m Figure 6-2,, suggests th
hat the size iss symmetriccally distribuuted. This ressult is generrally

conssistent with optical imag

ges (Figure 6-1). Coalesscence may happen durring the trannsfer

proccess to the gllass slide duee to the sheaaring of the eemulsions byy using pipetttes.

Figure
F 6-2. Sphere
S size distribution
d of
o water-in-ooil emulsionn determinedd by acousticc

measurements
m s.

6.3
3 Waterr-in-oill and oiil-in-waater em
mulsionss

96 
 
Wheen small flak
ke size hBN instead of the
t large hBN
N is used, itt results in thhe formationn of

oil-in
n-water emu
ulsions. Figu
ure 6-3 show
ws a clear topp layer of hheptane on thhe top of waater-

in-oiil emulsion on
o the left and
a a cloudy
y water layerr at the bottoom in oil-in--water emulssion

on th
he right. Thiis is due to the density of water beiing higher th
than the dennsity of heptaane.

The small amou

unt of phase separated laayer is causeed by coalesccence whichh is unavoidaable

due to
t the vibrattion during moving
m the emulsified
e saamples.

ure 6-3. Watter-in-oil em

Figu mulsion stabillized by largge flake size hBN (left) aand oil-in-water

emulsion
n stabilized by small flaake size hBN
N (right)

It waas also noticced that the sphere size of small flaake size hBN
N emulsion iis much smaaller

than the millimeeter size sph

heres of larg
ge flake sizee hBN emullsions (Figurre 6-4 (a)). The

reaso
on why the sphere
s size is different might
m be relaated to the diifferent flakee size. Sincee the

vapo
or pressure of
o water (2.34 kPa at 20°°C) is much lower than tthat of heptaane (5.33 kP
Pa at

20.0 °C), the continuous phase of oil-in

n-water emullsions does not evaporaate as fast as the

97 
 
wateer-in-oil emu
ulsions. But given long enough timee, small patcches of sheeets formed fr
from

burst spheres can

n also be fou
und (as show
wn in Figure 6-4 (b)).

Figure 6-4. Optical microscopy im

mage of smalll flake hBN stabilized oil-in-water

emulsions (a)
( right afterr transferring
g on glass sllide and (b) aafter about tten minutes

98 
 
The determinatiion of oil-in
n-water or water-in-oil emulsions is not onlyy based on the

obseervation of phase
p separaation. Both emulsions
e w
were added ddrop wise innto water. Laarge

flakee hBN emulsions burst immediately

i y and form ann oil layer fl
floating on thhe top as shoown

in Figure
F 6-5 (lleft arrow). Small flakee hBN emullsions are sstill very staable after beeing

addeed into waterr and oil sph

heres disperseed all over w
water (right aarrow in Figgure 6-5).

Fig
gure 6-5. Pho
otos of a vial of water affter large flakke hBN stabbilized emulssion added innto

waater (left routte) and small flake hBN stabilized em

mulsion addded into wateer (right routte)

6.4
4 Polym
merizatiion of oil
o phasse
Sincce the emulsiions are stab
ble for week
ks if they aree not open tto air, it is ppossible to caarry

out reactions in
n the emulsiion. When using
u an orrganic monoomer as thee oil phase, the

emulsion templaates a solid composite with

w BNNS at the surfaace of pores or particless. A

foam
m structure composite (F
Figure 6-6) can be obtainned from thee water-in-oil emulsion aafter

poly
ymerization. After polym
merization off the continuuous oil phasse, the water-filled poress are

coveered by BNN
NS. When water
w in the pore
p is removved during a drying process, compoosite

99 
 
foam
m was obtain
ned. In a typiical experim
ment, 2.5 gram
ms of hBN ((large flake)), 50 ml styreene,

5 mll divinyl ben

nzene (DVB
B) and DI waater are addded to a jar iin the order of hBN, waater,

styreene, then DV
VB followed
d bu bath son
nicating the jar for 30 seconds to brreak up any big

partiicles. After bath

b sonicattion, 0.53 grram benzoyll peroxide (B
BPO) is addded followedd by

stirriing to dissollve the initiaator in mono

omer. The m
mixture is thhen emulsifieed by a blennder

(Model 33BL79) for 1 minu

ute. The jar was
w then plaaced in an ovven at ∼70 °°C for 24 h. The

comp
posite samplle is removeed from the jar and driedd at 80 °C unntil the weighht is constannt.

Figure 6-6. SEM

S image of
o compositee foam produuced by polyymerizing thhe continuouus

sttyrene phasee.

A paarticle structture compossite (Figure 6-6) is prodduced by thee same proccedure but fr
from

oil-in
n-water emu
ulsion stabiliized by smalll flake hBN. SEM of thee BNNS coaated polystyrrene

(PS)) particles is shown in Fiigure 6-7.

100 
 
 Fig
gure 6-7. SEM image of composite particles
p polyymerizing thhe inner monnomer phasee of

sm
mall flake BN
NNS stabilizzed emulsionn

6.5
5 Chara
acteriza
ation off compoosite fooam

6.5
5.1 TGA
A
The composition
n of the com
mposite foam was analyzeed by TGA, which was carried out oon a

TA Instruments
I Q 500 Therrmogravimetric Analyzeer at a heatinng rate of 10 °C min-1 in a

platiinum pan. The

T weight loss
l at aroun
nd 150 °C can be attributed to thee water trappped

de the close--cell spheress. The main weight losss at 300 °C is attributedd to the therrmal
insid

degrradation of polystyrene.
p The
T loading of BNNS inn the compossite is ~5%.

101 
 
 Figure 6-8. TGA tracce of BNNS
S/PS compossite foam

6.5
5.2 SEM
M
Figu
ure 6-6 show
ws a SEM im
mage of the morphology
m y of composiite foam. SE
EM images w
with

high
her magnificaations shown
n in Figure 6-9
6 clearly ssupport that the BNNS oonly exist att the

interrface of the spheres

s and no BNNS caan be seen inn the continuuous polymeer matrix (w
white

arrow
w in Figure 6-9(c)).

102 
 
 Figure 6--9. SEM of BNNS/PS co
omposite foaam with diffferent magniifications

103 
 
6.5
5.3 Flam
mmability
The flammability with bulk PS and
y of compossite foam maaterials was sstudied and compared w

poly
ystyrene foam
m. Specimen
ns with dimeensions of 5ccm×5cm×1cm were preppared by cuttting

from
m larger sam
mples. Figu
ure 6-10 (cc) shows thhe self-extinnguishing pperformancee of

BNN
NS/PS comp
posite foam. Figure 6-10 (a,b) show
ws both bulkk PS and Styyrofoam do not

self--extinguish and
a Styrofoaam dripped with
w fire.

Fig
gure 6-10. Sn
napshots at different
d tim
me of videos rrecorded after specimenn (a, bulk PS,, b,

Styro
ofoam and c, BNNS/PS ffoam) are ignnited

The residue of BNNS/PS

B foam
f after th
he flammabiility test waas observed under SEM
M. A

continuous and compact fraamework wh

hich could rresist furtherr degradatioon was foundd at

the surface
s of th
he residue, in
ndicating thaat condensedd chars on thhe BNNS prrevented the fire

from
m developing
g. As a resultt, the overalll structure off foam was m
maintained vvery well.

104 
 
 Fiigure 6-11. SEM
S of outeer surface of composite ffoam after flammability test with chaar

formed
f on it..

6.6
6 Poten
ntial ap
pplicattions fo
or oil a
absorpttion
An oil spill is the release of a liquid
d crude oil or refined fuel into thhe environm
ment,

espeecially marin
ne areas. Cleeanup and reecovery from
m an oil spilll is difficultt and traditioonal

wayss of cleaning
g oil spills, such
s as solid
difying and sskimming, aare expensivee and slow. The

BNN
NS/PS comp
posite foam might be a potential maaterial that ccan be used for oil cleaanup

due to
t its foam structure
s and
d flame retardency.

To teest the abilitty of using th

his material for oil cleannup, a 50/500 water/organnic mixture w
was

prep
pared. Both heptane
h and
d waste pum
mp oil were uused as the organic phaase and the test

resullts are very similar. In a typical exp

periment, 0.55 gram of coomposite foaam was put into

the mixture
m and
d sat for abo
out 30 secon
nds. The com
mposite foam
m floated onn the top off the

mixtture due to its

i low density. After taaking the foaam from thee mixture, thhe weight off the

foam
m was measu
ured to deterrmine the mass
m of absorrbed oil. Aftter burning tthe foam with a

105 
 
torch
h, a thin layeer of black char
c was form
med outsidee the foam, w
which was thhen removedd by

blow
wing with co
ompressed aiir. The masss of the cleaaned foam iss measured aagain. Figuree 6-

12 sh
hows the maass change in
n a three-cyccle absorb-clleanup test. The mass off absorbed ooil is

abou
ut 100 mg in
n each of thee three cycles while the w
weight loss of solid com
mposite mateerial

is ~7
70 mg in eacch cleanup. And
A after th
hree cycles, 660% of origginal materiaal is left. Duue to

this foam materrials can bee easily prep

pared and rrelatively chheap, this m
might lead tto a

soluttion of oil sp
pills.

gure 6-12. Mass

Fig M change of the foam during threee absorptionn-burning (clleaning) cyclles

6.7
7 Concllusions
In th
his chapter, we
w have dem
monstrated th
he formationn of BNNS sstabilized em
mulsions andd the

abiliity of polym when a monoomer styrenee is used as oil.

merizing the oil phase off emulsion w

Com
mposites in forms of fo
oam and paarticles can be prepareed and theirr structures are

characterized by
y SEM. Add he flammabi lity of compposite foam was tested and
ditionally, th

106 
 
flame retardency of the foam is dramatically improved through the formation of chars on

BNNS. Potential applications of these foam materials were explored.

107 
 
 CHAPTER 7 Summary and future work

7.1 Summary
Two methods for the production of exfoliated BNNS/polymer composites from hBN have

been presented. The first method focused on the chemical oxidation modification of hBN

to allow for exfoliation and dispersion of the sheets (BNO) in both water and polymer

melt. BNO dispersed in polycarbonate improves the flame retardency of the polymer. The

oxidation temperature and time was shown to play a significant role in reaction. The

reaction mechanism was systematically studied and supported by further functionalization

reactions. Based on this functionalization, grafting polymer chains from BNO is confirmed

by FTIR and SEM. When polymer modified BNO is used as nanofiller, reinforcement of

the polymeric composites is obtained.

The second method to produce BNNS/polymer composites was developed based on the

interfacial trapping and exfoliating of hBN in a phase separated mixture of organic and

water. Characterization of the exfoliated BNNS network revealed the presence of

monolayers and the morphology of the sheet network. This sheet network was incorporated

into a polymer matrix by polymer processing and polymerization. By solution casting or

melt processing, polymer thin films with BNNS network embedded in them may be

suitable for electronics due to their improved dielectric properties. Composite foams

prepared by in-situ polymerization shows excellent flame retardency.

108 
 
7.2 Future work
Currently the high temperature oxidation method can be easily scaled up, which opens the

door to a number of promising chemistries with BNNS. One simple route towards PMMA

was demonstrated but other interesting properties should be explored by modifying BNNS

with other functional groups, small molecules or polymers.

The interfacial exfoliation method at the organic/water interface can be further optimized

for scaling up. And composites with more complicated structures, for example, multi-layer

structures, can be achieved and hopefully their unique and desired properties will be

discovered. For composites with foam or particle structures, further processing instead of

using as is may lead to potential thermal, electrical applications.

109 
 
 References
[1] M. Engler, C. Lesniak, R. Damasch, B. Ruisinger, J. Eichler, Cfi‐Ceram. Forum Int.
2007, 84, E49.
[2] G.‐W. Lee, M. Park, J. Kim, J. I. Lee, H. G. Yoon, Composites Part A: Applied
Science and Manufacturing 2006, 37, 727.
[3] W.‐L. Song, P. Wang, L. Cao, A. Anderson, M. J. Meziani, A. J. Farr, Y.‐P. Sun,
Angew Chem Int Edit 2012, 51, 6498.
[4] K. Kim, M. Kim, Y. Hwang, J. Kim, Ceramics International 2014, 40, 2047.
[5] Z. Lin, A. McNamara, Y. Liu, K.‐s. Moon, C.‐P. Wong, Composites Science and
Technology 2014, 90, 123.
[6] C. Zhi, Y. Bando, C. Tang, H. Kuwahara, D. Golberg, Adv Mater 2009, 21, 2889.
[7] N. Ayrilmis, T. Dundar, A. Kaymakci, F. Ozdemir, J. H. Kwon, Polym. Composite.
2014, 35, 194.
[8] N. G. A. Earnshaw, Chemistry of the Elements, Elsevier, 1984.
[9] Y. Lin, J. W. Connell, Nanoscale 2012, 4, 6908.
[10] K. Watanabe, T. Taniguchi, H. Kanda, Nat Mater 2004, 3, 404.
[11] A. Zunger, A. Katzir, A. Halperin, Phys Rev B 1976, 13, 5560.
[12] G. Shi, Y. Hanlumyuang, Z. Liu, Y. Gong, W. Gao, B. Li, J. Kono, J. Lou, R. Vajtai, P.
Sharma, P. M. Ajayan, Nano lett 2014, 14, 1739.
[13] C. R. Dean, A. F. Young, I. Meric, C. Lee, L. Wang, S. Sorgenfrei, K. Watanabe, T.
Taniguchi, P. Kim, K. L. Shepard, J. Hone, Nature Nanotech 2010, 5, 722.
[14] A. G. F. Garcia, M. Neumann, F. Amet, J. R. Williams, K. Watanabe, T. Taniguchi,
D. Goldhaber‐Gordon, Nano lett 2012, 12, 4449.
[15] A. F. Young, C. R. Dean, I. Meric, S. Sorgenfrei, H. Ren, K. Watanabe, T.
Taniguchi, J. Hone, K. L. Shepard, P. Kim, Phys Rev B 2012, 85, 235458.
[16] K. K. Kim, A. Hsu, X. Jia, S. M. Kim, Y. Shi, M. Dresselhaus, T. Palacios, J. Kong,
ACS Nano 2012, 6, 8583.
[17] Q. Li, K. Han, M. R. Gadinski, G. Zhang, Q. Wang, Adv Mater 2014, 26, 6244.
[18] S. Takahashi, Y. Imai, A. Kan, Y. Hotta, H. Ogawa, J Alloys Compd 2014, 615,
141.
[19] X. Wang, A. Pakdel, J. Zhang, Q. Weng, T. Zhai, C. Zhi, D. Golberg, Y. Bando,
Nanoscale Res Lett 2012, 7, 1.
[20] C. Zhang, L. Fu, S. Zhao, Y. Zhou, H. Peng, Z. Liu, Adv Mater 2014, 26, 1776.
[21] P. Sutter, R. Cortes, J. Lahiri, E. Sutter, Nano lett 2012, 12, 4869.
[22] S. J. Woltornist, A. J. Oyer, J.‐M. Y. Carrillo, A. V. Dobrynin, D. H. Adamson, ACS
Nano 2013, 7, 7062.
[23] Y. W. Zhu, S. Murali, W. W. Cai, X. S. Li, J. W. Suk, J. R. Potts, R. S. Ruoff, Adv
Mater 2010, 22, 3906.
[24] C. N. R. Rao, A. K. Sood, K. S. Subrahmanyam, A. Govindaraj, Angewandte
Chemie‐International Edition 2009, 48, 7752.
[25] J. M. Englert, C. Dotzer, G. Yang, M. Schmid, C. Papp, J. M. Gottfried, H.‐P. Steinr
ück, E. Spiecker, F. Hauke, A. Hirsch, Nat Chem 2011, 3, 279.

110 
 
[26] A. K. Geim, K. S. Novoselov, Nat Mater 2007, 6, 183.
[27] Y. Hernandez, V. Nicolosi, M. Lotya, F. M. Blighe, Z. Sun, S. De, I. T. McGovern, B.
Holland, M. Byrne, Y. K. Gun'Ko, J. J. Boland, P. Niraj, G. Duesberg, S. Krishnamurthy, R.
Goodhue, J. Hutchison, V. Scardaci, A. C. Ferrari, J. N. Coleman, Nat Nano 2008, 3, 563.
[28] S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas, E. J. Zimney, E. A.
Stach, R. D. Piner, S. T. Nguyen, R. S. Ruoff, Nature 2006, 442, 282.
[29] A. K. Geim, Science 2009, 324, 1530.
[30] M. Xu, T. Liang, M. Shi, H. Chen, Chem Rev 2013, 113, 3766.
[31] C. N. R. Rao, A. Nag, European Journal of Inorganic Chemistry 2010, 2010,
4244.
[32] J. N. Coleman, M. Lotya, A. O’Neill, S. D. Bergin, P. J. King, U. Khan, K. Young, A.
Gaucher, S. De, R. J. Smith, I. V. Shvets, S. K. Arora, G. Stanton, H.‐Y. Kim, K. Lee, G. T.
Kim, G. S. Duesberg, T. Hallam, J. J. Boland, J. J. Wang, J. F. Donegan, J. C. Grunlan, G.
Moriarty, A. Shmeliov, R. J. Nicholls, J. M. Perkins, E. M. Grieveson, K. Theuwissen, D.
W. McComb, P. D. Nellist, V. Nicolosi, Science 2011, 331, 568.
[33] M. Sajjad, W. M. Jadwisienczak, P. Feng, Nanoscale 2014, 6, 4577.
[34] S. Park, J. An, R. D. Piner, I. Jung, D. Yang, A. Velamakanni, S. T. Nguyen, R. S.
Ruoff, Chem. Mater. 2008, 20, 6592.
[35] C. Y. Zhi, Y. B. Xu, Y. Bando, D. Golberg, Acs Nano 2011, 5, 6571.
[36] S. Kemaloglu, G. Ozkoc, A. Aytac, Polym. Composite. 2010, 31, 1398.
[37] T.‐L. Li, S. L.‐C. Hsu, J Appl. Polym. Sci. 2011, 121, 916.
[38] D. Golberg, Y. Bando, Y. Huang, T. Terao, M. Mitome, C. C. Tang, C. Y. Zhi, Acs
Nano 2010, 4, 2979.
[39] A. Pierret, J. Loayza, B. Berini, A. Betz, B. Plaçais, F. Ducastelle, J. Barjon, A.
Loiseau, Phys Rev B 2014, 89, 035414.
[40] X. Li, X. Hao, M. Zhao, Y. Wu, J. Yang, Y. Tian, G. Qian, Adv Mater 2013, 25, 2200.
[41] Y. Lin, T. V. Williams, T.‐B. Xu, W. Cao, H. E. Elsayed‐Ali, J. W. Connell, J Phys
Chem C 2011, 115, 2679.
[42] T. Sainsbury, A. Satti, P. May, Z. Wang, I. McGovern, Y. K. Gun’ko, J. Coleman, J
Am Chem Soc 2012, 134, 18758.
[43] L. Wang, B. Wu, J. Chen, H. Liu, P. Hu, Y. Liu, Adv Mater 2014, 26, 1559.
[44] J. S. Kim, K. B. Borisenko, V. Nicolosi, A. I. Kirkland, Acs Nano 2011, 5, 3977.
[45] A. W. Yimin, I. K. Angus, S. Franziska, P. Kyriakos, P. Y. Neil, G. A. D. Briggs, H.
W. Jamie, Nanotechnology 2011, 22, 195603.
[46] Y. Shi, C. Hamsen, X. Jia, K. K. Kim, A. Reina, M. Hofmann, A. L. Hsu, K. Zhang, H.
Li, Z.‐Y. Juang, M. S. Dresselhaus, L.‐J. Li, J. Kong, Nano lett 2010, 10, 4134.
[47] H. Sachdev, F. Müller, S. Hüfner, Angew Chem Int Edit 2011, 50, 3701.
[48] A. Ismach, H. Chou, D. A. Ferrer, Y. Wu, S. McDonnell, H. C. Floresca, A.
Covacevich, C. Pope, R. Piner, M. J. Kim, R. M. Wallace, L. Colombo, R. S. Ruoff, Acs
Nano 2012, 6, 6378.
[49] L. Song, L. Ci, H. Lu, P. B. Sorokin, C. Jin, J. Ni, A. G. Kvashnin, D. G. Kvashnin, J.
Lou, B. I. Yakobson, P. M. Ajayan, Nano lett 2010, 10, 3209.
[50] K. H. Lee, H.‐J. Shin, J. Lee, I.‐y. Lee, G.‐H. Kim, J.‐Y. Choi, S.‐W. Kim, Nano lett
2012, 12, 714.
[51] L. Ci, L. Song, C. Jin, D. Jariwala, D. Wu, Y. Li, A. Srivastava, Z. F. Wang, K. Storr,
L. Balicas, F. Liu, P. M. Ajayan, Nat Mater 2010, 9, 430.
111 
 
[52] W.‐Q. Han, H.‐G. Yu, Z. Liu, Appl Phys Lett 2011, 98, 203112.
[53] K. J. Erickson, A. L. Gibb, A. Sinitskii, M. Rousseas, N. Alem, J. M. Tour, A. K. Zettl,
Nano lett 2011, 11, 3221.
[54] Z. Cui, A. J. Oyer, A. J. Glover, H. C. Schniepp, D. H. Adamson, Small 2014, 10,
2352.
[55] G. R. Bhimanapati, D. Kozuch, J. A. Robinson, Nanoscale 2014, 6, 11671.
[56] M. Du, X. Li, A. Wang, Y. Wu, X. Hao, M. Zhao, Angew Chem Int Edit 2014, 53,
3645.
[57] Y. Lin, T. V. Williams, W. Cao, H. E. Elsayed‐Ali, J. W. Connell, J Phys Chem C
2010, 114, 17434.
[58] A. S. Nazarov, V. N. Demin, E. D. Grayfer, A. I. Bulavchenko, A. T. Arymbaeva, H.‐
J. Shin, J.‐Y. Choi, V. E. Fedorov, Chem‐Asian J 2012, 7, 554.
[59] C. Zhi, N. Hanagata, Y. Bando, D. Golberg, Chemistry‐an Asian Journal 2011, 6,
2530.
[60] T. Brugger, H. Ma, M. Iannuzzi, S. Berner, A. Winkler, J. Hutter, J. Osterwalder, T.
Greber, Angew Chem Int Edit 2010, 49, 6120.
[61] K. S. Novoselov, D. Jiang, F. Schedin, T. J. Booth, V. V. Khotkevich, S. V. Morozov,
A. K. Geim, Proceedings of the National Academy of Sciences of the United States of
America 2005, 102, 10451.
[62] X. Wang, C. Zhi, L. Li, H. Zeng, C. Li, M. Mitome, D. Golberg, Y. Bando, Adv Mater
2011, 23, 4072.
[63] M. A. Rafiee, T. N. Narayanan, D. P. Hashim, N. Sakhavand, R. Shahsavari, R.
Vajtai, P. M. Ajayan, Advanced Functional Materials 2013, 23, 5624.
[64] U. Khan, P. May, A. O'Neill, A. P. Bell, E. Boussac, A. Martin, J. Semple, J. N.
Coleman, Nanoscale 2013, 5, 581.
[65] L. H. Li, J. Cervenka, K. Watanabe, T. Taniguchi, Y. Chen, ACS Nano 2014, 8,
1457.
[66] A. Nag, K. Raidongia, K. P. S. S. Hembram, R. Datta, U. V. Waghmare, C. N. R. Rao,
Acs Nano 2010, 4, 1539.
[67] A. Pakdel, C. Zhi, Y. Bando, T. Nakayama, D. Golberg, Acs Nano 2011, 5, 6507.
[68] R. V. Gorbachev, I. Riaz, R. R. Nair, R. Jalil, L. Britnell, B. D. Belle, E. W. Hill, K. S.
Novoselov, K. Watanabe, T. Taniguchi, A. K. Geim, P. Blake, Small 2011, 7, 465.
[69] K. Raidongia, A. Gomathi, C. N. R. Rao, Israel Journal of Chemistry 2010, 50,
399.
[70] Q. Tang, Z. Zhou, Prog Mater Sci 2013, 58, 1244.
[71] H. Zeng, C. Zhi, Z. Zhang, X. Wei, X. Wang, W. Guo, Y. Bando, D. Golberg, Nano
lett 2010, 10, 5049.
[72] C. H. Jin, F. Lin, K. Suenaga, S. Iijima, Phys. Rev. Lett. 2009, 102.
[73] Y. Zhao, X. Wu, J. Yang, X. C. Zeng, Phys. Chem. Chem. Phys. 2012, 14, 5545.
[74] S. Kocakuşak, K. Akçay, T. Ayok, H. J. Koöroǧlu, M. Koral, Ö. T. Savaşçi, R. Tolun,
Chem. Eng. Process. 1996, 35, 311.
[75] N. Jacobson, S. Farmer, A. Moore, H. Sayir, J Am. Ceram. Soc. 1999, 82, 393.
[76] O. Dugne, S. Prouhet, A. Guette, R. Naslain, R. Fourmeaux, K. Hssein, J. Sevely, C.
Guimon, D. Gonbeau, G. Pfister‐Guillouzo, J. Phys. Colloques 1989, 50, C5.
[77] W. Lei, D. Portehault, R. Dimova, M. Antonietti, J Am Chem Soc 2011, 133, 7121.
[78] R. Geick, C. H. Perry, G. Rupprecht, Phys. Rev. 1966, 146, 543.
112 
 
[79] S. Anderson, R. L. Bohon, D. D. Kimpton, J Am. Ceram. Soc. 1955, 38, 370.
[80] E. D. Grayfer, A. S. Nazarov, V. G. Makotchenko, S.‐J. Kim, V. E. Fedorov, J Mater.
Chem. 2011, 21, 3410.
[81] C. C. Teng, C. C. M. Ma, K. C. Chiou, T. M. Lee, Y. F. Shih, Materials Chemistry and
Physics 2011, 126, 722.
[82] Z. Cui, A. P. Martinez, D. H. Adamson, Nanoscale 2015, 7, 10193.
[83] M. Fang, K. Wang, H. Lu, Y. Yang, S. Nutt, J Mater. Chem. 2010, 20, 1982.
[84] V. L. Solozhenko, T. Peun, Journal of Physics and Chemistry of Solids 1997, 58,
1321.
[85] J. R. Potts, D. R. Dreyer, C. W. Bielawski, R. S. Ruoff, Polymer 2011, 52, 5.
[86] S. J. Woltornist, J.‐M. Y. Carrillo, T. O. Xu, A. V. Dobrynin, D. H. Adamson,
Macromolecules 2015, 48, 687.
[87] J.‐X. Deng, X.‐K. Zhang, Q. Yao, X.‐Y. Wang, G.‐H. Chen, D.‐Y. He, Chinese Physics
B 2009, 18, 4013.
[88] X. Huang, C. Zhi, P. Jiang, D. Golberg, Y. Bando, T. Tanaka, Advanced Functional
Materials 2013, 23, 1824.
[89] J. Gu, Q. Zhang, J. Dang, C. Xie, Polymers for Advanced Technologies 2012, 23,
1025.
[90] Z. Zhou, M. Mackey, K. Yin, L. Zhu, D. Schuele, L. Flandin, E. Baer, J Appl. Polym.
Sci. 2014, 131, 39877.
[91] M. Matt, H. Anne, B. Eric, F. Lionel, A. W. Mason, S. S. James, Journal of Physics
D: Applied Physics 2009, 42, 175304.
[92] Z. Zhou, M. Mackey, J. Carr, L. Zhu, L. Flandin, E. Baer, Journal of Polymer
Science Part B: Polymer Physics 2012, 50, 993.
[93] A. S. Dukhin, (Eds: P. J. Goetz, A. S. Dukhin), Elsevier, Amsterdam ; 2010.
 

113