Nanomaterials Stabilized Pickering Emulsions and Their

Applications in Catalysis

Chenhui Han

B. Eng.; M. A. Sci.

Submitted in fulfilment of the requirement for the degree of

Doctor of Philosophy

School of Chemistry, Physics and Mechanical Engineering

Science and Engineering Faculty

Queensland University of Technology

2019

i

Keywords

Pickering emulsions; Graphitic carbon nitride; Dynamic observation; Emulsion microreactors;

Hydrogenation; Palladium nanoparticles; Phase-switchable; Organic-inorganic hybrids;

Dynamic adsorption/desorption.

ii

Abstract

Pickering emulsions are emulsions stabilized by solid particles located at surfaces/interfaces

of liquid droplets that have promising applications in the food, paint, pharmaceutical,

cosmetic, material, and oil industries. In 2017, our group reported graphitic carbon nitride (g-

C3N4), a two-dimensional metal-free semiconductor material, acts as an effective stabilizer

for Pickering emulsions. Following this research direction, this thesis aims to investigate the

properties and stabilizing mechanism of g-C3N4 stabilized Pickering emulsions and explore

its application as a microreactor in catalysis. Besides, this thesis also reports a kind of phase-

switchable emulsion which is highly desired in emulsion-based catalysis.

The first part of the thesis focuses on the direct observation of g-C3N4 stabilized Pickering

emulsions. For most of Pickering emulsions, direct observation of the movement of the

stabilizer can be challenging due to the small size of those particles. Normally, cryoelectron

microscopy is used to understand these types of emulsion systems, but the freezing process

can also damage original morphologies. From this point of view, the g-C3N4-stabilized

emulsions can represent an excellent platform for the in situ study of emulsifying behavior.

Owing to its large lateral size and blue, stable fluorescence, the locations and motions of the

g-C3N4 stabilizer can be in situ monitored by light microscopy, fluorescence microscopy, and

confocal microscopy. Based on the observed results, two stabilizing configurations of the g-

C3N4 particles, with respect to the emulsion droplets under static conditions, were illustrated.

In addition, this research also demonstrates the capability to manipulate emulsion droplets

and investigate their response to external forces. Moreover, real-time observations were

performed to reveal the movement of g-C3N4 particles and emulsion droplets and study their

adsorption kinetics toward each other. Finally, the π-π interaction between the stabilizer and

iii

aromatic liquid phase (e.g., toluene) is considered and studied as an influencing factor on

emulsifying behavior.

In the second part of the thesis, the g-C3N4-stabilized emulsion was used as a microreactor to

conduct a hydrogen evolution-hydrogenation tandem reaction. Direct hydrogenation of

alkenes is a basic transformation in organic chemistry which is vanishing from simple

practice because of the need for pressurized hydrogen. Ammonia borane (AB) has emerged

as a hydrogen source due to its safety and high hydrogen content. However, in conventional

systems, the hydrogen liberated from the high-cost AB cannot be fully utilized. Herein, we

developed an emulsion-based catalytic system in which alkenes are hydrogenated in the oil

phase with hydrogen originating from AB in the water phase, catalyzed by the Pd

nanoparticles at the interfaces. In such a system, benefiting from the large biphasic interface

at where hydrogen transfer occurs as well as the hydrogen storage ability of Pickering

emulsions, 100% of hydrogen utilization efficiency was realized. This has obvious

advantages for more economical hydrogen utilization over conventional systems. The

emulsion microreactors can be applied to a range of alkene substrates, with conversion rates

achieving >95% by a simple modification. This case illustrates a good example of the

application of Pickering emulsions to catalysis.

In emulsion-based catalytic systems, phase-switchable emulsions are highly desirable due to

their advantage of consecutive operation. Therefore, the third part of the thesis relates to this

kind of emulsion. We develop an emulsion stabilized by organic-inorganic hybrid in which

the phase inversion between oil-in-water and water-in-oil type can be mechanically triggered

over many cycles. The hybrid was prepared by combining an oil-soluble molecule, stearic

acid, with water-dispersible Al2O3 nanofibers via chemisorption at the oil-water interface.

The hybrid exhibits reversible switching between hydrophilic and lipophilic states, by

mechanically manipulating (aging-sonicating operation) the adsorption-desorption volume of

iv

stearic acid attached to the Al2O3 nanofibers. As a result, the emulsions stabilized by the

hybrid can reversibly transform between oil-in-water and water-in-oil type. A detailed phase

inversion mechanism is demonstrated based on a series of interfacial chemistry

measurements and fluorescence-assisted in situ observation. As a bonus, 3D organic-

inorganic solid foams can be readily prepared based on the emulsion system, which

demonstrates potential applications for remediation of oil-spills in the environment.

i

List of Publications

Publications in this thesis:

1. Chenhui Han, Qianling Cui, Peng Meng, Eric R. Waclawik, Hengquan Yang, and

Jingsan Xu*. Direct Observation of Carbon Nitride-Stabilized Pickering Emulsions.

Langmuir 34.34 (2018): 10135-10143.

2. Chenhui Han, Peng Meng, Eric R. Waclawik, Chao Zhang, Xin-Hao Li, Hengquan

Yang, Markus Antonietti, and Jingsan Xu*. Palladium/Graphitic Carbon Nitride (g-

C3N4) Stabilized Emulsion Microreactor as a Store for Hydrogen from Ammonia

Borane for Use in Alkene Hydrogenation. Angewandte Chemie International Edition

57.45 (2018): 14857-14861.

Manuscripts in this thesis:

1. Chenhui Han, Eric R. Waclawik, Xiaofei Yang, * Peng Meng, Hengquan Yang,* Ziqi

Sun and Jingsan Xu*. Mechanically switchable emulsions. Manuscript submitted to

The Journal of Physical Chemistry Letters.

Other publications not in this thesis:

1. Qian Cao, Qianling Cui,* Yu Yang, Jingsan Xu,* Chenhui Han, and Lidong Li*.

Graphitic Carbon Nitride as a Distinct Solid Stabilizer for Emulsion Polymerization.

Chemistry-A European Journal 24.9 (2018): 2286-2291.

2. Jingwen Sun, Ravindra Phatake, Adi Azoulay, Guiming Peng, Chenhui Han, Jesús

Barrio, Jingsan Xu, Xin Wang, and Menny Shalom. Covalent Functionalization of

Carbon Nitride Frameworks through Cross-Coupling Reaction. Chemistry-A

European Journal 24.56 (2018): 14921-14927.

ii

Statement of Original Authorship

The work contained in this thesis has not been previously submitted to meet requirements for

an award at this or any other higher educational institution. To the best of my knowledge and

belief, the thesis contains no material previously published or written by another person

except where due reference is made.

Signature:

Date : ________________

QUT Verified Signature

i

Acknowledgments

First and foremost, I would like to express my sincere gratitude to my principal supervisor,

Dr. Jingsan Xu, for his expert and far-sighted academic guidance throughout my Ph.D. He

not only shows me how to do research in a professional way but also tells me how to balance

work and life. He is my supervisor but more like a friend. It was my fortune to complete my

Ph.D. under his supervision.

I would also express my gratitude to my associate supervisor, Assoc. Prof. Eric Waclawik,

for his valuable suggestion and guidance towards the completion of my publication and

candidature.

I would also thank the technicians from Science and Engineering Faculty (SEF) and Central

Analytic Research Facility (CARF) of QUT for their kindly technical support: Mr. Peter

Hegarty, Mrs. Leonora Nebwby, Dr. Jamie Riches, Dr. Sanjleena Singh, Dr. Josh Lipton-

Duffin, and Ms. Elizabeth Graham.

My gratefulness also goes to my dear lab mates: Peng Meng and Siliang Liu for their

cooperation and help on my research and for all the good times we had. Many thanks to all

my friends around QUT and throughout my life.

Thanks to QUT for the scholarship, this sponsorship offers me the opportunity to do

innovative research and complete my Ph.D. candidature.

Finally, I would give my deepest gratitude to my family. My wife and parents always support

me in my studies and work. Without their trust and support, I would never come here to do

research and complete my candidature. No words could ever express my appreciation to them.

ii

Table of Contents

Keywords .................................................................................................................................... i

Abstract ...................................................................................................................................... ii

List of Publications ..................................................................................................................... i

Statement of Original Authorship .............................................................................................. ii

Acknowledgments....................................................................................................................... i

Table of Contents ....................................................................................................................... ii

Introductory Remarks ............................................................................................................... iv

Chapter 1: Introduction and literature review ............................................................................ 1

1.1 Introduction ................................................................................................................. 1

1.2 Literature review ......................................................................................................... 4

1.3 Research problems and objectives ............................................................................ 22

Chapter 2: Direct Observation of Carbon Nitride-Stabilized Pickering Emulsions ................ 30

2.1 Introductory remarks ...................................................................................................... 32

2.2 Article 1 .......................................................................................................................... 33

Chapter 3: Application of g-C3N4 Stabilized Emulsion in Catalysis ....................................... 66

3.1 Introductory remarks ...................................................................................................... 68

3.2 Article 2 .......................................................................................................................... 69

Chapter 4: Mechanically Switchable Emulsions ..................................................................... 97

4.1 Introductory remarks ...................................................................................................... 99

4.2 Article 3 ........................................................................................................................ 100

Chapter 5: Conclusions & Future work ................................................................................. 126

iii

List of Abbreviations

LM Light Microscopy

FM Fluorescence Microscopy

g-C3N4 Graphitic Carbon Nitride

GC Gas Chromatography

NPs Nanoparticles

HUE Hydrogen Utilization Efficiency

SEM Scanning Electron Microscopy

TEM Transmission Electron Microscopy

XPS X-ray Photoelectron Spectroscopy

XRD X-ray Diffraction

SA Stearic acid

CCA Coumarin 343 X carboxylic acid

AB Ammonia borane

iv

Introductory Remarks

Graphitic carbon nitride (g-C3N4) as a metal-free semiconductor has been extensively studied

in multiple areas including energy conversion, optoelectronics, and catalysis. However, little

attention has been paid to the performance of g-C3N4 as an amphiphile. In 2017, our group

reported that g-C3N4 can act as the stabilizer for Pickering emulsions. Based on that work, by

taking advantage of its fluorescent nature, this thesis investigates the stabilizing configuration

and observed the movement of g-C3N4 at oil-water interfaces using fluorescence microscopy.

In addition, the response of emulsion drops to external forces and the effect of π-π interaction

between the conjugated structure of the oil and g-C3N4 on emulsion stability were studied.

Developing on results given in the first part of the thesis, the second part of the thesis focuses

on the application of the g-C3N4 stabilized Pickering emulsions. Pristine g-C3N4 was

modified by Pd nanoparticle deposition and used as catalysts to catalyze a tandem reaction in

the emulsion system. The prepared Pd/g-C3N4 catalysts show good emulsifying ability to

stabilize Pickering emulsions at the biphasic interfaces. Therefore, they are able to catalyze

ammonia borane hydrolysis in the water phase and alkene hydrogenation in the oil phase at

the same time, with mass transfer (hydrogen) happening at the interface. In this system, the

hydrogen utilization efficiency (HUE) can reach unity due to the in situ generation and

consumption of hydrogen.

The third part of the thesis investigated mechanically switchable emulsions in which the

phase inversion can be mechanically triggered between oil-in-water and water-in-oil type for

many cycles. The stabilizer used in this emulsion is an organic-inorganic hybrid which was

prepared by combining an oil-soluble molecule, stearic acid, with water-dispersible Al2O3

nanofibers via chemisorption at the oil-water interface. Therefore, the amphiphilicity of the

hybrid can be reversibly switched by adsorption-desorption manipulating. The detailed phase

v

inversion mechanism is demonstrated based on a series of interfacial chemistry

measurements and fluorescence-assisted in situ observation.

This thesis is a collection of published and submitted manuscripts of the author. Therefore,

the format mainly follows the style of specific journals with minor adjustments. Repetition

and redundancy in some parts of the thesis are unavoidable due to the close relationship

between the thesis and published papers.

1

Chapter 1: Introduction and literature review

1.1 Introduction

Emulsions are dispersing systems consisting of two immiscible liquids, in which oil/organic

solvents or water is dispersed into another phase to form micrometer-sized droplets. This

system is thermodynamically unstable due to the high surface energy resulting from the large

oil/water interface, and therefore the addition of a third phase, the emulsifier, is required to

obtain stable emulsions. Surfactants and some surface-active polymers such as proteins and

polysaccharides are widely used as emulsifiers to stabilize traditional emulsions. At the

beginning of the last century, Ramsden reported that a membrane consisting of finely divided

solid particles was also able to confine air bubbles in water or oil drops in water.[1] After that,

Pickering investigated this phenomenon carefully and noted that those particles wetted more

by water than by oil tended to stabilize oil-in-water emulsions by residing at the interface.[2]

Hence, the emulsions stabilized by solid particles were named “Pickering emulsions”.

Compared with traditional emulsions stabilized by surfactants, Pickering emulsions are

advantageous in the following three aspects: (1) less emulsifier is needed, therefore saving

the cost; (2) solid particles are less harmful to human bodies and the environment relative to

surfactants; (3) the formed emulsions are very stable and insensitive to the pH, salinity,

temperature, and oil composition of the system. Hence, Pickering emulsions have significant

application-value in many areas, such as the food, paint, pharmaceutical, cosmetic, material,

and oil industries.

Unlike traditional emulsions in which surfactants significantly decrease the interface tension

between oil and water phases and therefore make one phase dispersed into the other in the

form of micro drops, solid particles stabilize emulsions by adsorbing at the oil-water interface

2

and providing steric hindrance between dispersed drops to prevent coalescence. For those

particles having intermediate wettability, the adsorption free energy at the oil-water interface

could be very large, reaching more than 1000 kT, which means such particles are able to be

strongly held by the interface and therefore this kind of adsorption is believed irreversible.

Benefiting from this irreversible adsorption, Pickering emulsions are normally much more

stable than those surfactants stabilized ones (some Pickering emulsions are stable against

coalescence for more than three years).

Besides, for those particles too hydrophilic, or too hydrophobic, the adsorption free energy

falls rapidly to < 10 kT, resulting in an unstable adsorption and therefore those particles can

be easily moved into the bulk water phase or oil phase, respectively. In that case, the particles

are unable to stabilize an emulsion solely due to the inappropriate wettability. However, the

wettability of those particles can be altered through chemical method or physical adsorption

to reach an intermediate level so that becomes suitable for emulsion stabilization. For

example, essentially-hydrophilic silica particles can be modified through a controlled reaction

with dichlorodimethylsilane in the vapor phase to obtain a series of fumed silica particles

with different hydrophobicity that are excellent Pickering emulsion stabilizers.

More examples were reported on altering wettability by physical adsorption of particles such

as silica, alumina, clay, carbon, etc. with cationic, anionic, or polymeric surfactants.[3-9]

Cationic surfactants can adsorb on negatively charged particles, and this exposes their

hydrophobic tail outwards, therefore rendering a change on the wettability of particles, by

which form some essentially poor stabilizers can stabilize Pickering emulsions with the

addition of suitable surfactants. In addition to the adsorption, the added surfactants can also

play a role in decreasing the interface tension between oil and water phases, resulting in a

more stable emulsion. Binks et.al. conducted a detailed investigation of the synergistic

interaction between silica nanoparticles and cationic surfactants.[10] They believe slight

3

flocculation of particles can contribute to the stability of emulsions and such flocculation

results from the hydrophobic chain-chain interaction of surfactants.

During the recent decade, a new application of Pickering emulsions has been developing,

benefiting from its superior stability and the large interface between the immiscible two

phases. The stable droplets of Pickering emulsions are excellent micro rectors for many

organic reactions and the large interface is beneficial to the mass transfer and energy transfer

of these reactions between oil and water phases. Therefore, the Pickering emulsion system is

advantageous to its homogeneous or heterogeneous counterparts in terms of energy-saving,

product separation, and reactants compatibility.

Noticeably, Hengquan Yang et al. reported a series of continuous-flow reaction systems

based on modified SiO2-stabilized Pickering emulsions, in which homogeneous catalysts are

immobilized in dispersed droplets and the reactants can freely flow through the continuous

phase, with the catalytic process occurring at the interface. This strategy effectively avoids

the loss of homogeneous catalysts during the reaction process and therefore significantly

increased the catalysis efficiency of per-unit catalysts. Such an increase is especially valuable

for some costly homogeneous catalysts such as biological enzymes. Furthermore, the

Pickering-emulsion-based continuous flow reactor is superior to the corresponding batch

reactor in terms of the reaction rate and selectivity control. In the traditional batch reactors,

normally, certain reaction time is applied to the reaction in order to reach a decent conversion

rate. So a part of the products which is generated first has to stay in the reactor until the end

of the reaction. This drawback of the batch reactors could result in a waste of reaction time

and overreaction of products, thus decreases the reaction rate and product selectivity. In

contrast, the Pickering-emulsion-based continuous flow reactor can effectively avoid the two

drawbacks.

4

Besides the above-mentioned advantages that a Pickering-emulsion-based reactor holds

against its batch counterpart, a more intriguing strategy is using the whole emulsion system

as a platform to conduct two reactions at the same time in the dispersed phase and continuous

phase respectively, and make the mass transfer happen at the interface. For some specific

reactions, the product of one reaction is also the reactant of the other one but the two

reactions are unable to be conducted in one pot because of mutual destruction or interference.

In that case, the Pickering emulsions can play an ideal platform for those reactions.

In summary, although the Pickering emulsions have been extensively studied and well

documented since its discovery more than a century ago, investigating the stabilizing

mechanism and interface properties in detail is still of great importance for application in

various fields. Furthermore, the application of Pickering emulsions as a reactor for chemical

reactions shows great potential for practice use and could become a promising research field

in the near future.

1.2 Literature review

1.2.1 Basic knowledge of Pickering emulsions

1.2.1.1 The stable mechanism and emulsion type

It has now well understood that finely divided solid particles with intermediate wettability are

able to stabilize Pickering emulsions by being captured at the interface to prevent coalescence

of dispersed droplets. The wettability of particles is normally quantified by the contact angle.

Figure 1.1 schematically defines the contact angle. For hydrophilic particles, more surface is

wetted by the water phase and the contact angle is less than 90o while for hydrophobic

particles, more surface is wetted by oil phase and the contact angle is larger than 90o.

5

Figure 1.1. The definition of contact angle.

Normally, intermediately wettable particles refer to those particles with contact angles

between 50 and 130o. Such particles are good emulsifiers being able to stabilize Pickering

emulsions through adsorption at the interface. This kind of adsorption is normally believed to

be irreversible due to the large free energy of adsorption for these particles. The following

equation shows the relationship between the free energy (-∆intG) required to remove a particle

from the oil-water interface and the contact angle (θow) of a small spherical particle at a

planar interface:

-∆intG = πr2γow(1±cos θow)2 (1)

in this equation, r is the radius of the particle and γow is the interface tension between the two

phases; the effect of gravity of a very small solid particle is negligible and therefore can be

ignored. The sign inside the bracket is negative for moving a particle from the interface into

the water phase and is positive for moving into the oil phase. To a given particle (the radius

typically less than 1 μm) and a certain oil-water interface (e.g. toluene-water interface, γow =

36 mN/m), the particle is most strongly held at the interface if the θow is exactly 90o (half of

the particle wetted by water and the other half by oil), with -∆intG exceeded 2000 kT. Either

side of 90o, -∆intG decreases rapidly with the change of contact angle and drops to only

several kT when θow is between 0o and 20o (very hydrophilic) or between 160o and 180o (very

hydrophobic). So particles at the interface with a very low or very high contact angle are

6

much easier to be moved to the water phase or oil phase. Therefore, neither too hydrophilic

nor too hydrophobic particles are able to stabilize Pickering emulsions.

Figure 1.2. The relationship between hydrophobicity of solid particles and the corresponding

stabilized Pickering emulsion type: (left) oil-in-water emulsion (o/w) stabilized by slightly

hydrophilic particles; (right) water-in-oil emulsion (w/o) stabilized by slightly hydrophobic

particles.

For those slightly hydrophilic particles, e.g. metal oxides, their adsorption on oil-water

interface could render the interface curving towards the oil phase because the majority of the

particle surface is wetted by water, by which the oil-in-water (o/w) emulsions are stabilized.

In contrast to this, those slightly hydrophobic particles tend to stabilize water-in-oil (w/o)

emulsions (Figure 1.2). For instance, a typical water-wet particle, untreated silica, should

stabilize o/w emulsion while the oil-wet particles such as carbon black tends to stabilize w/o

emulsions. Binks et al. published a review paper in 2003, presenting some important data on

the relationship between contact angles and corresponding emulsion types by summarizing a

series of frequently-used solid emulsifiers. [11] This data shows good correspondence with the

established theory. However, the particle wettability could be changed by adsorption of

another surface-active species that results in emulsion phase inversion in some cases. A

detailed discussion on this phenomenon is given later.

7

1.2.1.2 Influencing factors to Pickering emulsions

The behavior and properties of Pickering emulsions could be influenced by factors such as

the particle concentration, particle location, electrolyte concentration, oil-water ratio, and oil

type. The stability and rheology of Pickering emulsions highly depend on the above factors.

Furthermore, a catastrophic or transitional phase inversion (the emulsion type changes from

o/w to w/o or vice versa) can also result from changes of such factors

1. Effect of particle concentration

The initial particle concentration mainly affects the average emulsion drop size and stability

of emulsions. For given emulsification conditions, assuming that all the particles are

monodispersed and they stabilize drops by adsorbing at the oil-water interface forming a

hexagonally close-packed monolayer, the higher the particle concentration, the larger surface

area they can cover and therefore the smaller average drop size obtained.

Figure 1.3. Median drop diameter as a function of particle concentration and the ratio of the

total number of particles available to the number required to provide a monolayer around all

drops. Taken from reference [11].

8

Binks et al. systematically studied the relationship between particle concentration and the

median drop diameter using partially hydrophobized monodisperse spherical silica particles,

25 nm in diameter, as emulsifier. As shown in Figure 1.3, they found that the average drop

size decreases with the increase of particle concentration and tends to remain constant when

the concentration is high. This is similar to the emulsions that stabilized by a surfactant, in

which the average drop size reaches the minimum at the critical micelle concentration (CMC)

and levels off after that. Besides, the ratio of the total number of particles (nt) to the number

of particles required to adsorbed at the interface (na) to stabilize drops significantly increased

after the average drop size tends to constant. This result indicates that the excess particles are

unable to adsorb at the interface. Instead, they stay in the phase in which they are initially

dispersed in.

As an additional bonus, the increased particle concentration could also result in an increase in

the viscosity of emulsions and the emulsion stability against creaming by gelation of the

continuous phase, therefore endowing the emulsions with an extremely long life. This effect

is particularly obvious for clay-stabilized Pickering emulsion systems, in which the excess

clay not only increases the viscosity of the continuous phase but also forms a three-

dimensional network surrounding the drops through the interaction between particles. This

form provides additional stability to the emulsion systems because the oil drops were

captured or immobilized in an array of clay platelets in water.

2. Effect of electrolyte

Almost all the solid particles in their aqueous dispersion are charged positively or negatively.

The addition of salt can reduce the surface potentials of particles and result in coagulation

between particles to form flocs. Binks et al. investigated the effect of LaCl3 concentration on

silica stabilized Pickering emulsions.[12] As shown in Figure 1.4, they found that at very low

9

LaCl3 concentration, the emulsion stability against both creaming (open points) and

coalescence (filled points) are very poor; with the increase of the concentration, the stability

also increases and the most stable emulsions are obtained when the electrolyte concentration

between 2-5 mM. Beyond this concentration, the stability drops again. They ascribe the

improvement of stability to the weakly flocculation of particles resulted from the addition of

LaCl3. Small flocs are able to enhance the emulsion stability because the moderately

increased particle radius (namely the flocs radius) could result in an increase in the adsorption

energy. However, destabilization is observed at high salt concentration due to the oversize

flocs which are not favorable for emulsion stabilization.

Figure 1.4. Stability to creaming (open points) and coalescence (filled points) as a function

of LaCl3 concentration at pH 10. Taken from reference [12].

Similar results were also reported by other researchers. Among them, Dejun Sun et al. studied

a kind of inorganic two-dimensional material, layered double hydroxides (LDH), stabilized

Pickering emulsions and revealed the dependence between emulsion stability and NaCl

concentration in the water phase.[13] They found that the surface potential of LDH sheets is

very high at low NaCl concentration. Therefore, it is difficult for these LDH sheets to adsorb

10

at the oil-water interface to form an interfacial film. However, the surface potential could be

significantly reduced by moderately increasing NaCl concentration, which is favorable for the

adsorption of LDH sheets and emulsion stability. When the NaCl concentration exceeds 1M,

the LDH colloidal coagulates and precipitates quickly, resulting in unstable emulsions.

Apart from various salts, surfactants can also interact with dispersed particles in a similar

way to electrolytes. For example, positively charged particles in the water phase attract

anionic surfactants while negatively charged particles interact with cationic surfactants.

Flocculation could happen between particles as a result of the interaction between the

hydrophobic tails of surfactants adsorbed on the particle surface. Furthermore, through the

absorption of surfactant, the wettability of the particle surface could be altered to some extent.

These changes are normally beneficial to the stability of emulsions and therefore the majority

of commercialized emulsions contain both solid particles and surfactants.

The behavior of Pickering emulsions including the viscosity and rheology is also affected by

other factors such as the pH of the water phase, the initial location of particles (dispersed first

in oil or water phase), and the type of oil (polar or non-polar oil), etc. These factors will not

be discussed in detail in this thesis.

2. 1.2.1.3 The phase inversion behavior of Pickering emulsions.

An o/w Pickering emulsion could become w/o type or vice versa by altering the oil-water

ratio or the wettability of solid particles. The former situation, phase inversion resulted from

oil-water ratio, is known as catastrophic phase inversion while the later one is known as

transitional phase inversion. Binks et al. investigated a series of Pickering emulsions

stabilized by silica particles possessing different amount of SiOH at their surfaces by which

the wettability of these particles are varied. With more SiOH group at the surface, the particle

is more hydrophilic; while with less SiOH, the particle is hydrophobic.

11

1. Catastrophic phase inversion

In the first set of experiments, they found that phase inversion can be achieved without

altering the particle wettability and particle concentration; simply by changing the oil-water

ratio. They showed an example in which the w/o emulsions stabilized by 2 wt.% of silica

particles (50% SiOH) inverted to o/w type upon the increase of dispersed phase volume

fraction (ϕw). The phase inversion happened around ϕw = 0.7 and this ratio seems fixed

irrespective of the way of the ϕw changing, i.e. increasing or decreasing ϕw. Such a no

hysteresis phase inversion behavior for Pickering emulsions is quite different from that of in

surfactant stabilized systems in which the hysteresis can be as much as 0.3 in volume fraction.

Furthermore, they found that the most stable emulsions against creaming (for o/w type) or

sedimentation (for w/o type) can be obtained near the phase inversion point where the

viscosity of emulsions reach maximum.

2. Transitional phase inversion

The transitional phase inversion is induced by the progressive altering of the average

wettability of solid particles. The first way of achieving this is by directly using a series of

particles with different wettability, e.g. using silica with different amount of SiOH on the

surface. Binks et al. conducted such an experiment and found that the initially o/w emulsions

stabilized by hydrophilic silica gradually inversed into w/o type with the increase of particles’

hydrophobicity (given by decrease of SiOH in the table, the less the SiOH, the stronger the

hydrophobicity).[14] Moreover, the median diameter of drops goes through a minimum and

the stability a maximum near the phase inversion point. They ascribe the high stability to the

high attachment energy of particles because the emulsions near the inversion point are

stabilized by particles with intermediate wettability, Eq. (1).

12

Figure 1.5. (left) Conductivity and type of emulsions (ϕw = 0.5) containing 2 wt.%

hydrophobic silica particles as a function of added hydrophilic particles. (right) Conductivity

and type of emulsions (ϕw = 0.5) containing 2 wt.% hydrophilic silica particles as a function

of added hydrophobic particles. Taken from reference [15].

Another method of achieving transitional inversion is by stabilizing emulsions using a

mixture of hydrophobic particles and hydrophilic particles in order to change the average

wettability. There are three forms to achieve this goal: (a) introducing hydrophobic particles

into o/w emulsions stabilized by hydrophilic particles; (b) introducing hydrophilic particles

into w/o emulsions stabilized by hydrophobic particles; (c) varying the ratio of one particle

type at constant total particle concentration. Figure 1.5 shows data of the emulsion type

(indicated by the conductivity of emulsions) as a function of the second type particles adding

amount.

It is clear that transitional phase inversion occurs for both initially w/o (left) and o/w

emulsions (right). Although these emulsions are very stable to coalescence during the

inversion process, the stability to gravity-induced separation, namely creaming for o/w

emulsions and sedimentation for w/o emulsions, go through a minimum approaching the

inversion conditions, which is companied by the fact that the median drop sizes reaching a

13

maximum. This is exactly opposite to the transitional inversion behavior of using single type

particles.

1.2.2 Graphitic carbon nitride (g-C3N4) stabilized Pickering emulsions

Most of the Pickering emulsion stabilizers are spherical particles such as silica and barium

sulfate as these particles are easily monodispersed and therefore are ideal models for the

establishment of theory related to Pickering emulsions.[16] However, apart from spherical

particles, some two-dimensional materials such as graphene oxide [17, 18] and layered double

hydroxide (LDH) [13, 19] also showed good performance to stabilizing Pickering emulsions. In

2017, our group reported an organic two-dimensional material, graphitic carbon nitride (g-

C3N4), to be used as an emulsifier. [20]

Graphitic carbon nitride is a metal-free polymeric semiconductor that has been widely used in

fields including catalysis,[21, 22] energy conversion,[23-25] and optoelectronics.[26, 27] Benefiting

from its unique structure (Figure 1.6 a), g-C3N4 is able to act as an amphiphile. The

conjugated basal plane is hydrophobic, while hydrophilicity results from the edge functional

groups. Hexane-in-water emulsions can be readily obtained by manual shaking of a mixture

of hexane and water in the presence of g-C3N4. The average drop sizes of the obtained

Pickering emulsions decreased with the increase of the g-C3N4 concentration, spanning from

around 90 μm when 1mg/ml of g-C3N4 dispersion was used to 140 μm when 0.083 ml/ml of

g-C3N4 dispersion was used. Furthermore, by coloring the water phase in advance, an oil-in-

water emulsion type was identified because the whole emulsion phase presented the same

color after emulsification (Figure 1.6 b), which is in line with the result obtained through

drop testing. Moreover, a Tyndall effect was observed in the bottom aqueous phase,

indicating the presence of residual g-C3N4 nanoparticles (Figure 1.6 c). Besides, g-C3N4 also

showed generality to be used as an emulsifier. Four different oils including chloroform,

14

toluene, dichloromethane, and n-heptane were tested and the corresponding Pickering

emulsions are shown in Figure 1.6 d. The ability of g-C3N4 to emulsify chloroform and

dichloromethane was lower than to emulsify the other two oils, probably due to the relatively

high polarizabilities and dielectric constants of chloroform and dichloromethane.

Figure 1.6. (a) The framework of g-C3N4; (b) hexane-in-water Pickering emulsions stabilized

by g-C3N4 and corresponding colored ones. From left to right: the original one, colored by

methyl orange, and colored by red MX-5B. (c) Tyndall effect of the lower colloidal phase

and emulsion phase. (d) Pickering emulsions formed between water and different oils. From

left to right: chloroform, toluene, dichloromethane, and n-heptane. Taken from reference [20].

As g-C3N4 itself is an organic semiconductor which is widely used in multiple areas, it may

also widen the application of Pickering emulsions in terms of catalysis and energy conversion.

The g-C3N4 sheets captured at interfaces may promote the interaction between the two phases,

such as intriguing an interface reaction. Furthermore, as an amphiphile, g-C3N4 is also able to

disperse hydrophobic solids such as graphite and carbon nanotubes (CNTs) in the aqueous

phase. Such a function is similar to the behavior of surfactants rather than traditional

Pickering emulsion stabilizers, which is also an interesting point and deserves an in-depth

study.

15

1.2.3 The application of Pickering emulsions in chemical reactions

During the past decade, the application of Pickering emulsions has been extended to catalysis.

For some well-designed solid particle stabilizers, their wettability is adjustable upon the

addition of acid or base and therefore the emulsions stabilized by them are able to be

switched from o/w to w/o or vice versa. Based on this property, such an emulsion system can

be used for fast catalyst recycling or product separation. In addition to this, benefiting from

the large biphasic interface existed in Pickering emulsions, they are also ideal alternatives for

biphasic catalysis reactions.

1. The application of Pickering emulsions for solid catalysts recycling

Switchable Pickering emulsions are normally prepared by surface-modified solid particles

whose wettability can be altered by the introduction of other species such as acid/base,

CO2/N2, etc.[28-32] Hengquan Yang et al.[31, 32] reported a Pickering emulsion stabilized by

silica microsphere with their surface modified by octyl or triamine (see Figure 1.7 a). Such a

structure endows the silica microsphere responsiveness to acid or base. Under acidic

conditions, the surface is more hydrophilic due to the protonation effect while the wettability

could be switched to more hydrophobic under basic conditions as a result of deprotonation.

Figure 1.7. (a) The structure of an octyl-triamine modified silica microsphere. (b) The

strategy used for catalyst separation and recycling. Taken from reference[31].

16

These silica microspheres are further deposited with palladium nanoparticles and used as

catalysts to catalyzing the hydrogenation of styrene. As shown in Figure 1.7 b, before

reaction, an o/w emulsion can be readily generated under acidic conditions and styrene is

dissolved in the oil phase. The catalysts are evenly held at the interface, where styrene is

hydrogenated. After reaction, by adding a small amount of base, the wettability of silica

microsphere is changed to hydrophobic and is, therefore, more suitable to stabilize a w/o

emulsion. So the bulk oil phase (the products are in the oil phase) is released and can be

easily removed. The resulted water phase including solid emulsifier/catalyst can be directly

used in the next reaction cycle. This strategy realizes the in situ separation of products and

recycling of catalysts, which decreases the loss of catalyst during the traditional recycling

process.

2. The application of Pickering emulsions in stirring-free biphasic reaction systems

For achieving the goal of fast mass transfer in an organic-aqueous biphasic reaction, normally

vigorous stirring is necessary. However, such vigorous stirring is not only a high energy-

consuming process but also causes fragmentation of catalyst particles. Therefore, Pickering

emulsions are ideal alternatives for biphasic reaction systems due to the large static interfaces

created in an emulsion system.

As an example, Hengquan Yang et al. conducted a nitrobenzene reduction reaction in a

Pickering emulsion system stabilized by palladium nanoparticles deposited on silica

microspheres.[33] In this system, oil-soluble nitrobenzene was reduced to aniline in the

presence of sodium borohydride (NaBH4) in the water phase catalyzed by palladium

nanoparticles at the interface. With the increase of emulsifier used, smaller average drop size

was obtained and more interface was created which is beneficial to the formation rate of

products. Under the optimized reaction conditions, the formation rate in the Pickering

17

emulsion system is comparable to that of a typical biphasic system stirred at 2200 rpm. In

addition to this, they further extended this system to the reaction of allylic alcohol

epoxidation and obtained positive results.

A similar strategy was employed by Wen-Juan Zhou et al. [34] who reported a solvent-free

acetalization of immiscible long-chain fatty aldehydes with ethylene glycol conducted in

emulsion systems. More noticeably, some biphasic reactions catalyzed by enzymes that are

very valuable and sensitive, e. g. Candida Antarctica lipase B (CALB) catalyzed hydrolysis

of racemic esters, which are also able to be conducted by such a strategy and therefore

effectively avoid the loss of enzymes. [35]

3. Pickering emulsions application in multistep reactions.

It is well known that the living systems are able to process multistep reactions in a one-pot

fashion but this is a formidable challenge in actual production. One main problem is the

mutual destruction of incompatible reagents such as acid and base, oxidants and reductants

and so on. The best method to tackle this problem is to compartmentalize the incompatible

reagents into different parts, avoiding the direct contact of them, and meanwhile, let other

reagents freely diffuse in the system and contact with the two incompatible reagents

respectively. For this purpose, Pickering emulsions are an ideal platform for this kind of

reaction. Figure 1.8 schematically illustrates such a strategy used by Hengquan Yang et al. [36]

who conducted a series of cascade reactions in Pickering emulsions. They realized the

coexistence of two incompatible reagents through confining them into the dispersed phases of

two emulsions respectively and then laminating the two emulsions together.

18

Figure 1.8. The strategy used for compartmentalizing the incompatible reagents in the

Pickering emulsions system. Taken from reference [36].

HCl and NaOH were first used as model reagents to show the feasibility of the strategy. They

first prepared two Pickering emulsions (w/o type) stabilized by a kind of

methyltrimethoxysilane modified silica particles, with the water phase containing HCl and

NaOH, respectively (Figure 1.9 a1 and a2). For better visualization, the emulsions were

stained by a trace amount of Congo Red which presents blue color in acidic conditions and

red color in basic conditions. Then they obtained a mixed emulsion by laminating the two

emulsions together. Such an emulsion is stable for more than 144 h which is long enough for

most chemical reactions (Figure 1.9 a3). In contrast, immediately mutual destruction

happened when directly mixed the two non-emulsified mixtures (Figure 1.9 a4).

Figure 1.9. Appearance of the layered Pickering emulsions with an indicator of Congo Red:

(a1) 0.01 M of HCl in water phase; (a2) 0.01 M of NaOH in the water phase; (a3) mixing a1

20

Figure 1.10. Schematic illustration of the Pickering-emulsion-based continuous flow reactor

for organic-aqueous biphasic catalysis reactions. Taken from reference [37].

As shown in Figure 1.10, in such a reactor, the water-soluble catalyst is confined inside of

the drops (dispersed phase) while oil-soluble reactants are carried by the continuous phase

flowing through the space between drops, with catalytic reaction happening at the interface.

The generated product which is carried by the continuous phase is continuously discharged

with the outflow.

This flow Pickering emulsion system (FPE) is advantageous for avoiding the loss of catalysts

during reaction. The authors investigated the retention fraction of catalysts after reaction

using highly water-soluble reagents as models. The results show that the retention fractions of

all the tested reagents, H2SO4, heteropolyacid (H3PW12O40, HPA), tetraethylenepentamine

(TEPA), trisodium salt of tri-(m-sulfonphenyl)phosphine (TPPTS) and Candida antarctica

enzyme (CALB, lipase) exceed 93% after 120 h continuous flow. This merit is especially

valuable to those reactions catalyzed by homogeneous catalysts in which the recycling of

catalysts is difficult, but this strategy immensely prolongs the service life of catalysts and

enhances the catalysis efficiency (mole of products produced on per mole of catalysts per

21

hour). Three interface reactions were chosen as models and conducted in the flow Pickering

emulsion system. As shown in Figure 1.11, all the three reactions were effectively catalyzed

by corresponding homogeneous catalysts and superior reaction stability and durability were

achieved, with no activity fluctuation and degradation happening during 1000-2000 hours of

reaction.

Figure 1.11. Catalytic reactions conducted in flow Pickering emulsion systems: (a) H2SO4-

catalyzed addition reactions for the synthesis of tetrahydropyranyl ethers, (b) HPA-catalyzed

ring-opening reactions of epoxides with anilines, (c) CALB enzyme-catalyzed hydrolysis

kinetic resolution of racemic acetates. Taken from reference. [37]

The same strategy was also used for HPA-catalyzed cyclization of citronellal, [38] CALB-

catalyzed enantioselective trans-esterification and amidation reactions, CuI-catalyzed

azide−alkyne cycloadditions, [39] and asymmetric ring-opening reactions of epoxides over a

22

liquid−solid hybrid catalyst.[40] In comparison to the traditional method, all of these reactions

were proven more effective when conducted in a flow Pickering emulsion system.

1.3 Research problems and objectives

As discussed above, Pickering emulsions have been extensively studied for decades and most

of their macroscopic behavior, such as the effect of particle concentration, type of oils,

electrolyte and phase inversion has been well known so far. However, on the microscale, due

to the small size of solid stabilizer and the limitation of observation method, it is difficult to

reveal the configuration and evolution of solid particles at oil-water interfaces, which are an

important complement to the emulsion knowledge-base and meaningful for expanding the

application of Pickering emulsions in the catalysis area.

Our group reported a Pickering emulsion stabilized by g-C3N4 in 2017.[20] In comparison to

the traditional solid stabilizer such as SiO2 and BaSO4 (normally < 1μm in diameter), g-C3N4

is larger in morphology, 20-50 μm in diameter, and more importantly, it is fluorescent. The

two merits provide us with an ideal model to study the microscopic properties of Pickering

emulsions with the help of fluorescence microscopy.

Herein, the first part of the thesis aims to provide more detail of emulsion stabilization and

therefore enrich the knowledge-base of Pickering emulsions. The main objectives of this part

are listed as follows:

1. Investigate the stabilization mechanism and stabilizing configurations of g-C3N4

stabilized Pickering emulsions.

2. Observe the evolution and motion of g-C3N4 nanosheets on the oil-water interface.

3. Investigate the response of emulsion drops to external forces.

23

4. Shed light on the adsorption kinetics between g-C3N4 nanosheets and oil-water

interface.

5. Study the effect of π-π interaction between conjugated structure of oil and g-C3N4 on

emulsion stability.

Apart from the utilization of g-C3N4 in stabilizing Pickering emulsions, more commonly, it is

widely used as support materials for metal nanoparticles to prepare various catalysts,

benefiting from its two-dimensional structure. Combining the two characteristics, it is

possible to locate the catalyst exactly on the biphasic interface and therefore provide an ideal

platform to conduct tandem reactions, one in the oil phase and the other in the water phase. In

this case, the product of reactions conducted in the water phase is also the reactant of the

coupled reaction in the oil phase where the final product is obtained. During this process, the

numerous emulsion droplets act as microreactors for the oil-phase-reaction and mass transfer

between the oil and water phases happens at the interface.

Therefore, to explore the new application of g-C3N4 stabilized Pickering emulsion in the area

of catalysis, the objectives of the second part of the thesis are listed as follows:

1. Synthesize a g-C3N4 supported Pd catalyst which is active to both alkene

hydrogenation in the oil phase and hydrolysis of ammonia borane in the water phase.

2. Characterize the prepared g-C3N4-based catalysts and evaluate the effect of Pd

nanoparticle deposition on the emulsions stabilization.

3. Conduct the tandem reactions, alkene hydrogenation and ammonia borane hydrolysis,

in g-C3N4-based catalysts stabilized Pickering emulsions and compare the hydrogen

utilization efficiency under various reaction conditions.

4. Conduct a kinetic study on the reaction to reveal the hydrogen transfer process.

24

5. Evaluate the recyclability of catalysts and the applicability of the system to other

substrates.

The third part of the thesis focuses on designing emulsions that are able to phase invert many

times, by reversible switching of the amphiphilicity of stabilizer. As introduced in section

1.2.1, the emulsion types highly depend on the hydrophobicity of the stabilizer which is

normally quantified by the indicator of contact angle. Those moderately hydrophilic particles

(contact angle slightly smaller than 90o) tend to stabilize an oil-in-water (o/w) emulsion while

those moderately hydrophobic particles (contact angle slightly larger than 90o) are more

likely to stabilize water-in-oil (w/o) emulsions. Therefore, if the contact angle of the

stabilizer can be dynamically altered at around 90o, the emulsion types could also be changed

correspondingly. To achieve this goal, organic-inorganic hybrids are good candidates because

their hydrophobicity could be adjusted by controlling the ratio of the two components. With

this idea, the main objectives of the third part of the thesis are listed as follows:

1. Synthesize and characterize Al2O3 nanofibers that can easily adsorb stearic acid on

their surface.

2. Use the synthesized Al2O3 nanofibers, together with stearic acid, to stabilize

emulsions and try to trigger multiple phase inversion by manipulating the adsorption-

desorption of stearic acid.

3. Screen the switchable concentration range of stearic acid.

4. Investigate the phase inversion mechanism.

5. Try to find other applications of the organic-inorganic hybrids obtained from the

emulsion framework.

25

Reference

[1] W. Ramsden, Separation of solids in the surface-layers of solutions and 'suspensions'

(observations on surface-membranes, bubbles, emulsions, and mechanical coagulation).--

preliminary account, Proc. R. Soc. London, Ser. A, 72 (1903) 156-164.

[2] S.U. Pickering, Emulsions, J. Chem. Soc., 91 (1907) 2001-2021.

[3] A. Gelot, W. Friesen, H.J.C. Hamza, Emulsification of oil and water in the presence of

finely divided solids and surface-active agents, Colloids Surf., 12 (1984) 271-303.

[4] V. Menon, D. T. Wasan, Particle-fluid interactions with application to solid-stabilized

emulsions part I. The effect of asphaltene adsorption, Colloids Surf., 19 (1986) 89-105.

[5] H. Hassander, B. Johansson, B. Törnell, The mechanism of emulsion stabilization by

small silica (Ludox) particles, Colloids Surf., 40 (1989) 93-105.

[6] G. Lagaly, M. Reese, S.J.A.C.S. Abend, Smectites as colloidal stabilizers of emulsions: I.

Preparation and properties of emulsions with smectites and nonionic surfactants, Appl. Clay

Sci., 14 (1999) 83-103.

[7] K.-L. Gosa, V. Uricanu, Emulsions stabilized with PEO-PPO-PEO block copolymers and

silica, Colloids Surf. A, 197 (2002) 257-269.

[8] A.P. Sullivan, P.K.J.I. Kilpatrick, The effects of inorganic solid particles on water and

crude oil emulsion stability, Ind. Eng. Chem. Res., 41 (2002) 3389-3404.

[9] J. Legrand, M. Chamerois, F. Placin, J. Poirier, J. Bibette, F.J.L. Leal-Calderon, Solid

colloidal particles inducing coalescence in bitumen-in-water emulsions, Langmuir, 21 (2005)

64-70.

[10] B.P. Binks, J.A. Rodrigues, W.J. Frith, Synergistic interaction in emulsions stabilized by

a mixture of silica nanoparticles and cationic surfactant, Langmuir, 23 (2007) 3626-3636.

[11] R. Aveyard, B.P. Binks, J.H. Clint, Emulsions stabilized solely by colloidal particles,

Adv. Colloid Interface Sci., 100 (2003) 503-546.

26

[12] B.P. Binks, S.O. Lumsdon, Stability of oil-in-water emulsions stabilized by silica

particles, PCCP, 1 (1999) 3007-3016.

[13] F. Yang, Q. Niu, Q. Lan, D. Sun, Effect of dispersion pH on the formation and stability

of Pickering emulsions stabilized by layered double hydroxides particles, J. Colloid Interface

Sci., 306 (2007) 285-295.

[14] B.P. Binks, S.O. Lumsdon, Influence of particle wettability on the type and stability of

surfactant-free emulsions, Langmuir, 16 (2000) 8622-8631.

[15] B.P. Binks, S.O. Lumsdon, Transitional phase inversion of solid-stabilized emulsions

using particle mixtures, Langmuir, 16 (2000) 3748-3756.

[16] B.P. Binks, J.H. Clint, Solid wettability from surface energy components: relevance to

Pickering emulsions, Langmuir, 18 (2002) 1270-1273.

[17] J. Kim, L.J. Cote, F. Kim, W. Yuan, K.R. Shull, J. Huang, Graphene oxide sheets at

interfaces, J. Am. Chem. Soc., 132 (2010) 8180-8186.

[18] Y. He, F. Wu, X. Sun, R. Li, Y. Guo, C. Li, L. Zhang, F. Xing, W. Wang, J. Gao,

Factors that affect Pickering emulsions stabilized by graphene oxide, ACS Appl. Mater.

Interfaces, 5 (2013) 4843-4855.

[19] F. Yang, S. Liu, J. Xu, Q. Lan, F. Wei, D. Sun, Pickering emulsions stabilized solely by

layered double hydroxide particles: the effect of salt on emulsion formation and stability, J.

Colloid Interface Sci., 302 (2006) 159-169.

[20] J. Xu, M. Antonietti, The performance of nanoparticulate graphitic carbon nitride as an

amphiphile, J. Am. Chem. Soc., 139 (2017) 6026-6029.

[21] Y. Wang, X. Wang, M. Antonietti, Polymeric graphitic carbon nitride as a

heterogeneous organocatalyst: from photochemistry to multipurpose catalysis to sustainable

chemistry, Angew. Chem. Int. Ed., 51 (2012) 68-89.

27

[22] Y. Zheng, L. Lin, B. Wang, X. Wang, Graphitic carbon nitride polymers toward

sustainable photoredox catalysis, Angew. Chem. Int. Ed., 54 (2015) 12868-12884.

[23] Y. Zheng, Y. Jiao, Y. Zhu, L.H. Li, Y. Han, Y. Chen, A. Du, M. Jaroniec, S.Z. Qiao,

Hydrogen evolution by a metal-free electrocatalyst, Nat. Commun., 5 (2014) 3783.

[24] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen, M.

Antonietti, A metal-free polymeric photocatalyst for hydrogen production from water under

visible light, Nat. Mater., 8 (2009) 76.

[25] Y. Zheng, J. Liu, J. Liang, M. Jaroniec, S.Z. Qiao, Graphitic carbon nitride materials:

controllable synthesis and applications in fuel cells and photocatalysis, Energy Environ. Sci.,

5 (2012) 6717-6731.

[26] J. Xu, T.J.K. Brenner, L. Chabanne, D. Neher, M. Antonietti, M. Shalom, Liquid-based

growth of polymeric carbon nitride layers and their use in a mesostructured polymer solar cell

with Voc exceeding 1 V, J. Am. Chem. Soc., 136 (2014) 13486-13489.

[27] H. Guo, J. Zhang, L. Ma, J.L. Chavez, L. Yin, H. Gao, Z. Tang, W. Chen, A Non‐rare‐

Earth ions self‐activated white emitting phosphor under single excitation, Adv. Funct. Mater.,

25 (2015) 6833-6838.

[28] P. Liu, W. Lu, W.J. Wang, B.G. Li, S. Zhu, Highly CO2/N2-switchable zwitterionic

surfactant for Pickering emulsions at ambient temperature, Langmuir, 30 (2014) 10248-

10255.

[29] C. Liang, Q. Liu, Z. Xu, Surfactant-free switchable emulsions using CO2-responsive

particles, ACS Appl. Mater. Interfaces, 6 (2014) 6898-6904.

[30] Y. Zhang, S. Guo, X. Ren, X. Liu, Y. Fang, CO2 and redox dual responsive Pickering

emulsion, Langmuir, 33 (2017) 12973-12981.

28

[31] H. Yang, T. Zhou, W. Zhang, A strategy for separating and recycling solid catalysts

based on the pH-triggered Pickering-emulsion inversion, Angew. Chem. Int. Ed., 52 (2013)

7455-7459.

[32] Y. Yu, L. Fu, F. Zhang, T. Zhou, H. Yang, Pickering-emulsion inversion strategy for

separating and recycling nanoparticle catalysts, Chemphyschem, 15 (2014) 841-848.

[33] W. Zhang, L. Fu, H. Yang, Micrometer‐scale mixing with Pickering emulsions: biphasic

reactions without stirring, ChemSusChem, 7 (2014) 391-396.

[34] W.J. Zhou, L. Fang, Z. Fan, B. Albela, L. Bonneviot, F. De Campo, M. Pera-Titus, J.M.

Clacens, Tunable catalysts for solvent-free biphasic systems: Pickering interfacial catalysts

over amphiphilic silica nanoparticles, J. Am. Chem. Soc., 136 (2014) 4869-4872.

[35] L. Wei, M. Zhang, X. Zhang, H. Xin, H. Yang, Pickering emulsion as an efficient

platform for enzymatic reactions without stirring, ACS Sustain. Chem. Eng., 4 (2016) 6838-

6843.

[36] H. Yang, L. Fu, L. Wei, J. Liang, B.P. Binks, Compartmentalization of incompatible

reagents within Pickering emulsion droplets for one-pot cascade reactions, J. Am. Chem. Soc.,

137 (2015) 1362-1371.

[37] M. Zhang, L. Wei, H. Chen, Z. Du, B.P. Binks, H. Yang, Compartmentalized droplets

for continuous flow liquid-liquid interface catalysis, J. Am. Chem. Soc., 138 (2016) 10173-

10183.

[38] H. Chen, H. Zou, Y. Hao, H. Yang, Flow Pickering emulsion interfaces enhance

catalysis efficiency and selectivity for cyclization of citronellal, ChemSusChem, 10 (2017)

1989-1995.

[39] M. Zhang, R. Ettelaie, T. Yan, S. Zhang, F. Cheng, B.P. Binks, H. Yang, Ionic liquid

droplet microreactor for catalysis reactions not at equilibrium, J. Am. Chem. Soc., 139 (2017)

17387-17396.

29

[40] X. Zhang, Y. Hou, R. Ettelaie, R. Guan, M. Zhang, Y. Zhang, H. Yang, Pickering

emulsion-derived liquid-solid hybrid catalyst for bridging homogeneous and heterogeneous

catalysis, J. Am. Chem. Soc., 141 (2019) 5220-5230.

30

Chapter 2: Direct Observation of Carbon Nitride-Stabilized

Pickering Emulsions

This chapter is made up of the published journal article below:

Chenhui Han, Qianling Cui, Peng Meng, Eric R. Waclawik, Hengquan Yang, and Jingsan

Xu*. Direct Observation of Carbon Nitride-Stabilized Pickering Emulsions. Langmuir 34.34

(2018): 10135-10143.

31

Statement of Contribution of Co-Authors

The authors listed below have certified that:

1. they meet the criteria for authorship in that they have participated in the conception,

execution, or interpretation, of at least that part of the publication in their field of expertise;

2. they take public responsibility for their part of the publication, except for the responsible

author who accepts overall responsibility for the publication;

3. there are no other authors of the publication according to these criteria;

4. potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or

publisher of journals or other publications, and (c) the head of the responsible academic

unit, and

5. they agree to the use of the publication in the student’s thesis and its publication on the

QUT’s ePrints site consistent with any limitations set by publisher requirements.

In the case of this chapter:

Chenhui Han, Qianling Cui, Peng Meng, Eric R. Waclawik, Hengquan Yang, and Jingsan

Xu*. Direct Observation of Carbon Nitride-Stabilized Pickering Emulsions. Langmuir 34.34

(2018): 10135-10143.

Contributor Statement of contribution

Chenhui Han Conceived the idea of the project; conducted the

experiments; analyzed the data; wrote the

manuscript. Signature:

Dr. Qianling Cui Discussed the results and revised the manuscript.

Peng Meng Provided the TEM and SEM data.

A/Prof. Eric R. Waclawik Discussed the results and revised the manuscript.

Prof. Hengquan Yang Provided the interfacial tension data.

Dr. Jingsan Xu Conceived the idea of the project; analyzed the

data; revised the manuscript.

Principal Supervisor Confirmation

I have sighted email or other correspondence from all co-authors confirming their certifying

authorship.

Jingsan Xu ______________

Name Signature Date

QUT Verified Signature

32

2.1 Introductory remarks

In a previous work of our group, g-C3N4 was proven to be a good emulsifier for stabilizing

Pickering emulsions. As an extension of that work, the direct observation of the movement

and configuration of g-C3N4 at the oil-water interface was conducted in this work using light

microscopy and fluorescence microscopy, benefiting from the large lateral size and blue,

stable fluorescence of g-C3N4. Based on the obtained results, we illustrate two stabilizing

configurations of the g-C3N4 particles with respect to the emulsion droplets under static

conditions. In addition to this, the adsorption kinetics between g-C3N4 particles and emulsion

droplets and the response of those droplets to external forces were also investigated. Finally,

the π-π interaction between g-C3N4 and aromatic oil phases such as toluene and p-xylene is

considered and studied as an influencing factor on emulsifying behavior. This work provides

significant insights into the interfacial properties of g-C3N4 and enriches the knowledge of the

Pickering emulsion stabilizing mechanism.

33

2.2 Article 1

Direct Observation of Carbon Nitride-Stabilized Pickering Emulsions

Chenhui Han,† Qianling Cui,§ Peng Meng,† Eric R. Waclawik,† Hengquan Yang‡ and Jingsan

Xu*†

† School of Chemistry, Physics and Mechanical Engineering, Queensland University of

Technology, Brisbane, QLD 4001, Australia

§ State Key Laboratory for Advanced Metals and Materials, School of Materials Science and

Engineering, University of Science and Technology Beijing, Beijing 100083, China

‡ School of Chemistry and Chemical Engineering, Shanxi University, Wucheng Road 92,

Taiyuan 030006, China

ABSTRACT: Pickering emulsions are emulsions stabilized by solid particles located at

surfaces/interfaces of liquid droplets that have promising applications for drug delivery and in

nanomaterials synthesis. Direct observation of Pickering emulsions can be challenging.

Normally, cryoelectron microscopy needs to be used to better understand these types of

emulsion systems, but cryofreezing these emulsions may cause them to lose their original

morphologies. In this work, we demonstrate that graphitic carbon nitride (g-C3N4) can

stabilize oil-in-water (o/w) emulsions, with hexane illustrated as a typical oil phase. The g-

C3N4-stabilized emulsions can act as an excellent platform for in situ study of emulsifying

behavior from the mechanical point of view. Owing to its large lateral size and blue, stable

fluorescence, the locations and motions of the g-C3N4 stabilizer can be finely in situ

monitored by light microscopy, fluorescence microscopy, and confocal microscopy.

Accordingly, we illustrate two stabilizing configurations of the g-C3N4 particles with respect

to the emulsion droplets under static conditions. Further, we demonstrate the capability to

manipulate emulsion droplets and investigate their response to external forces. We perform

34

real-time observations of the g-C3N4 particles and the emulsion droplets that move in the

continuous phase and study their adsorption kinetics toward each other. Finally, the π-π

interaction between the stabilizer and aromatic liquid phase (e.g., toluene) is considered and

studied as an influencing factor on emulsifying behavior.

1. INTRODUCTION

Emulsions have been widely explored and used in industrial processes and in daily

consumables, including the petroleum, food, pharmaceutical, and cosmetic industries.

Traditionally, emulsions are stabilized by molecular surfactants that can be dissolved in either

water or oil. Concerns regarding the widespread use of surfactants are that they are difficult

to recover, and their intensive use may impact the environment with the possibility of

inducing cell damage or causing tissue irritation to living bodies. Pioneering work conducted

by Ramsden and Pickering revealed that emulsions can also be stabilized by solid particles,

rather than surfactants or polymers.[1, 2] Pickering emulsions usually feature superior stability

and low toxicity and hold many potential applications, such as nanocomposite synthesis via

Pickering emulsion polymerization,[3, 4] interfacial reactions for improved catalytic

efficiency,[5-7] as well as potential drug-delivery systems or in biosensing.[8] Among various

solid compounds used as Pickering emulsifiers (e.g., silica,[9, 10] Laponite clays,[11-13]

graphene oxide (GO) sheets,[14, 15] and metal oxides[16-19]), GO can be regarded as a two-

dimensional, colloidal stabilizer and used in emulsion polymerization for generating

polymer-GO nanocomposites.[14, 15] SiO2 nanoparticles are the most popular Pickering-type

stabilizer due to their high homogeneity, well-controlled particle size, and surface wettability.

As such, SiO2 nanoparticles have frequently been deployed to produce Pickering emulsions

for different applications, and also as a model agent to study the stabilization mechanism of

Pickering emulsions more generally.[9] Normally, the SiO2 stabilizer is small compared to the

35

emulsion droplets (∼100 nm vs ∼100 μm), where the emulsion stabilization mechanism is

considered to involve steric hindrance against droplet-droplet coalescence, provided by SiO2

aggregates at the interface.[20] However, direct observation and proof of this stabilization

mechanism is still lacking because visualization of the nanosized SiO2-stabilized emulsions

requires freeze-fracture electron microscopy, which precludes in situ monitoring of the solid

emulsifiers, the emulsion droplets, and their movements. Moreover, the emulsions may lose

their original morphology upon freezing, so information obtained by this process is limited.

Therefore, it is desirable to find a material platform to directly observe Pickering emulsion

systems, to verify and investigate the stabilization configuration. The ideal material platform

would allow the detection and recording of the dynamic changes of emulsifiers and emulsion

droplets and thereby assist in understanding the nonequilibrium stability of Pickering

emulsions.

Over the past decade, the unique, metal-free semiconductor g-C3N4 has been extensively

studied in many areas, including energy conversion and storage,[21, 22] optoelectronics,[23-25]

and catalysis.[22, 26, 27] g-C3N4 possesses a layered structure with each layer consisting of tri-s-

triazine units bridged by nitrogen atoms.[28] In fact, g-C3N4 is also rich in primary/secondary

amines attached to the conjugated framework. In addition to applications in the fields

mentioned above, the physicochemical properties of g-C3N4 in liquid phases are beginning to

attract increasing attention from researchers. For instance, g-C3N4 can be readily processed

into few-layered, aqueous dispersed nanoparticles via liquid-assisted exfoliation, which leads

to blue-shifted photoluminescence.[29, 30] Zhang and co-workers reported that after

hydrolyzing g-C3N4 in base solution, the as-obtained nanofiber dispersion can form a

hydrogel network triggered by CO2, which is presumably covalently bonded by connecting

the amine groups from different g-C3N4 frameworks.[31] Very recently, our team has found

that g-C3N4 can behave as a surface-active material at the liquid-liquid, water-solid, and

36

water-air interfaces.[32] In particular, g-C3N4 can act as a straightforward stabilizer for

Pickering emulsions, generated by simply shaking a mixture of g-C3N4 aqueous dispersion

along with a hydrophobic organic solvent.[32]

In this context, we consider that g-C3N4-stabilized emulsions can be an excellent platform for

direct observation and mechanistic study of Pickering-type systems. First, the as-used g-C3N4

possesses a sheetlike morphology with lateral sizes up to tens of micrometers and hence they

can be readily observed by optical microscopy, in an ambient environment. Second, g-C3N4 is

intrinsically photoluminescent, a property resulting from its conjugated framework that

causes it to emit blue light upon UV irradiation.[33] This character endows fluorescence

microscopy (FM) a viable technique to more precisely and clearly locate the g-C3N4 and

meanwhile exclude the disturbance of possible solid impurities. Moreover, different from

conventional SiO2 colloidal nanoparticles, g-C3N4 typically possesses a nonuniform, layered

morphology, so the study of its emulsifying properties can significantly supplement the

understanding of Pickering-type stabilization. In this work, we employed g-C3N4 as a model

Pickering emulsifier and studied its stabilization configurations by in situ observation of the

emulsions through light microscopy (LM) and fluorescence microscopy (FM). Furthermore,

the movements of the g-C3N4 particles and the emulsion droplets were detected in real-time,

providing solid evidence for understanding how the emulsions dynamically stabilized. We

also found that the emulsifying function can be different, depending on the type of organic

phase, aromatic solvents in particular, presumably owing to the π-π interactions between the

conjugated framework of g-C3N4.

2. EXPERIMENTAL SECTION

2.1. Materials.

37

Cyanuric acid (Sigma-Aldrich), melamine (Sigma-Aldrich), toluene (AR, Fisher Scientific),

hexane fraction (Ajax Finechem), p-xylene (Sigma-Aldrich), and decane (Sigma-Aldrich)

were used as received without further purification. Ultrapure water (18.2 MΩ cm, Synergy

UV Water Purification System, Merck Millipore) was used for all of the emulsion preparation.

2.2. Synthesis of g-C3N4.

g-C3N4 was synthesized using the cyanuric acid-melamine supramolecular precursor

according to a previous report.[34] Typically, a 1:1 molar ratio of cyanuric acid and melamine

were mixed in water and shaken for 24 h at room temperature. Then, the obtained milky

suspension was centrifuged and washed with water three times and dried at 50 °C under

vacuum. Afterward, the precursor powder was annealed in capped crucibles at 550 °C for 4 h

at a heating rate of 2.3 °C/min in nitrogen atmosphere. After cooling down, the yellow g-

C3N4 powder was collected.

2.3. Preparation of Pickering Emulsions.

g-C3N4 aqueous suspensions of different concentrations were prepared by sonicating g-C3N4

solid in ultrapure water for 5 min. The suspensions were then mixed with an organic solvent

(e.g., hexane, 1/1 mL) in a vial (5 mL) and sealed by a cap with a Teflon gasket. The liquid

mixture was shaken by hand for about 30 s to produce Pickering emulsions. The quoted

concentrations of g-C3N4 represent concentrations in the dispersed aqueous phase. The o/w

emulsions with toluene, decane, and p-xylene as the oil phase were prepared by a similar

method.

2.4. Interfacial Tension Measurement.

The dynamic water-hexane interfacial tension was measured by the pendant drop method

(OCA 25, Dataphysics, Germany). A pendant drop of g-C3N4 aqueous dispersion (0.05 and

38

0.25 mg/mL) was formed at the end of a needle immersed in the hexane oil phase. The aging

of the interface was monitored by measuring the interfacial tension of a drop over 50 min.

2.5. Observation of Pickering Emulsions.

The microscope (LM, FM, and confocal) images and the video of the manipulating process

on droplets were obtained through observing a water-diluted emulsion placed in the middle of

two pieces of glass slides separated by a distance of 1 mm. During the manipulating process,

a thin copper wire (0.1 mm in diameter) was used as a probe to touch and manipulate

emulsion droplets. A Nikon Eclipse Ni light microscope, a Nikon Eclipse Ti inverted

fluorescence microscope, and a Nikon A1R Confocal microscope were used to image the

emulsions. The stability of the emulsions has been evaluated by monitoring the size variation

of at least 300 droplets against aging in a bulk sample (1 mL of emulsion).

2.6. Contact Angle Measurement.

The contact angle was measured on a g-C3N4 disk (diameter, 13 mm; thickness, ca. 1 mm)

made from approximately 1 g of g-C3N4 particles. The powder was compressed into a disk in

a steel die with a pressure of 6 × 108 N/m2. The sessile water droplet method was performed

on an FTA200 Contact Angle Analyser. The picture was taken by a camera and analyzed

using Fta32 Video 2.0 software. Nine independent measurements were performed and the

mean values presented.

3. RESULTS AND DISCUSSION

3.1. Static Observation of Emulsions.

The g-C3N4 emulsifier was synthesized using the approach of heating a cyanuric acid-

melamine supramolecular precursor, resulting in aggregates with a thickness of 15-30 nm and

lateral sizes of up to ∼10 μm (Figure S1).[34] The UV-vis absorption and photoluminescence

39

spectra of g-C3N4 bulk powder and aqueous dispersion were measured and are recorded in

Figure S1d, e. Typically, Pickering emulsions were generated by manually shaking a mixture

of hexane and g-C3N4 aqueous dispersion in a vial (Figure S2). The emulsions were then

transferred to a homemade setup (Figure S3) for direct LM and FM observation.

We noted that oil-in-water (o/w) emulsions were formed under a quite low g-C3N4

concentration (0.25 mg/mL), with the oil droplets having sizes of 50-150 μm. The formation

of o/w type of emulsions agreed with the contact angle measurements, which illustrates that

g-C3N4 was predominantly hydrophilic and had a contact angle of ca. 58° (Figure S4).

Figure 1a shows that the average droplet size of the freshly prepared emulsions decreased

from 125 to 75 μm when the g-C3N4 concentration increased from 0.05 to 2 mg/mL, agreeing

with the normal property of Pickering emulsions.[35] The stability of the emulsions was

examined by monitoring the droplet size variation against different aging times in an ambient

environment. We observed that the Pickering emulsion stabilized by 0.05 mg/mL g-C3N4

started to collapse after two weeks, while the emulsions remained stable for at least three

months when the concentration of g-C3N4 reached 0.25 mg/mL or higher (Figure 1b).

Further, the emulsifying efficiency (= (1 −remaining g-C3N4 concentration in water

original g-C3N4 concentration in water ) × 100%）of g-

C3N4 was determined to be above 90% for all concentrations (Figure 1c), suggesting the

strong capability of g-C3N4 for emulsification.

The dynamic hexane-water interfacial tension with g-C3N4 dispersed in the water phase was

monitored by the pendant drop method. As shown in Figure 1d, the hexane-water interfacial

tension was measured over 50 min. As a reference, the pure water-hexane interfacial tension

was measured and determined to be 47.5 mN/m, which is in agreement with previous

reports.[36] The presence of g-C3N4 lowered the interfacial tension to 32 and 24 mN/m for

concentrations of 0.05 and 0.25 mg/mL, respectively, indicating the surface activity of g-

40

C3N4. Besides, the interfacial tension did not reach equilibrium even after 50 min, suggesting

slow partitioning of the g-C3N4 particles to the hexane-water interfaces, probably owing to

their large hydrodynamic sizes.[37]

Figure 1. Hexane-in-water emulsion stabilized by g-C3N4: (a) droplet size of different g-

C3N4 concentrations, (b) average droplet size variation against aging time under different g-

C3N4 concentrations, and (c) emulsifying efficiency under different g-C3N4 concentrations. (d)

Dynamic hexane-water interfacial tension, with 0.05 and 0.25 mg/mL g-C3N4 dispersed in

water, respectively.

It can be seen in the LM and FM pictures (Figure 2a,b) that the surfaces of droplets were

rarely completely covered by g-C3N4 particles when the concentration of g-C3N4 in the

dispersing phase was relatively low. Instead, the g-C3N4 particles aggregated at the droplet

interfaces, while only a small amount of g-C3N4 was nonuniformly adhered to the droplet

surface. This observation was confirmed by threedimensional (3D) confocal imaging (Figure

41

2c,d). The results clearly show that emulsion droplets were not necessarily fully covered by

the g-C3N4 to gain stability; rather, islands of g-C3N4 particles at droplet junctions provided

the steric barrier required to prevent coalescence between neighboring droplets, resulting in

exceedingly high emulsifying efficiency (Figure 1c). This phenomenon significantly differs

from the commonly proposed stabilization geometry for Pickering emulsions that a dense (or

sparse) particle layer must evenly form around the dispersed droplet to act as the barrier to

coalescence.[38-40]

Figure 2. Hexane-in-water emulsions stabilized by 0.25 mg/mL g-C3N4: (a) LM and (b)

corresponding FM images. Blue color indicates the fluorescence emitted by g-C3N4, denoting

the exact position of the g-C3N4 particles. Images of the emulsion taken by a confocal

microscope: (c) normal black-white picture and (d) the corresponding 3D confocal picture,

where the colors indicate the height.

42

To exclude the possible effect of surface-active small molecules being associated with the g-

C3N4, two control experiments were carried out: (1) the pristine g-C3N4 powder was

thoroughly washed with acetone and ethanol and then dried in a vacuum oven. The obtained

g-C3N4 was then dispersed in water to form a suspension. We found that this pretreated g-

C3N4 retained its emulsifying capacity (Figure S5a). (2) The pristine g-C3N4 was first

dispersed in water by sonication and then the g-C3N4 powder was fully removed by

centrifugation and filtering. The aqueous supernate did not show any activity for

emulsification (Figure S5b). These results confirmed that g-C3N4 can behave as an effective

emulsifier.

Vignati et al. reported silica colloid (ca. 500 nm)-stabilized emulsions with particle coverages

as low as 5% and attributed the stabilization mechanism to a kinetic effect, i.e., the Brownian

motion and the associated redistribution of nanoparticles.[41] For the case of g-C3N4

aggregates described here, however, the surfactant size of several tens of micrometers makes

it unlikely that a Brownian motion-based stabilization mechanism occurred. Besides, we did

not observe Brownian-type motion during LM imaging. Therefore, we consider the high

stability of g-C3N4 Pickering emulsions to originate from two effects: first, the hexane-water

interfacial tension was lowered owing to the presence of g-C3N4; second, and more

importantly, a steric barrier was provided by the interfacial adsorption of the solid particles

such that the adjacent droplets would not coalesce over a long time course. We consider that

these effects should be suitable for a range of Pickering emulsions, including those stabilized

by silica, clays, and a variety of metal oxides.

In another set of experiments, the concentration of g-C3N4 in the dispersion was increased to

2 mg/mL and the obtained emulsion was observed. As can be seen in Figure 3a, the droplets’

surfaces had a “powdery” appearance due to the adhesion of a large amount of g-C3N4. The

visualization of g-C3N4 coverage could be further improved by FM, which clearly imaged the

43

full coverage of the emulsion droplets by g-C3N4 particles (Figure 3b), as indicated by the

blue emission. Interestingly, the excess g-C3N4 preferred to be located in the conjoining

regions of droplets, forming networks that bridged the droplets and contributed to emulsion

stability.[42] The emulsion was further observed by confocal microscopy, and the 3D confocal

image clearly exhibits the spatial distribution of the g-C3N4 particles (Figure 3c).

Figure 3. (a) LM and (b) the corresponding FM image of emulsions prepared by 2 mg/mL

aqueous g-C3N4 dispersion and hexane. (c) Threedimensional confocal image of the emulsion

covered by g-C3N4. The colors indicate the height of the g-C3N4. (d) Schematic illustration of

two configurations of g-C3N4-stabilized emulsions.

It is worth noting that the Pickering emulsions stabilized by g-C3N4 (0.25 mg/mL or higher)

all showed remarkable stability. On the basis of these results, we propose a stabilization

mechanism for Pickering emulsions under static conditions, with the g-C3N4 emulsifier

present in two configurations. As described by the scheme in Figure 3d, when the g-C3N4

concentration is relatively low, the oil droplets require only partial coverage of g-C3N4 to

44

inhibit coalescence, with the solid particles located at the interfaces of the droplets (scheme i).

With increased g-C3N4 concentration, more g-C3N4 particles adsorb to the droplet surfaces

and eventually achieve full coverage (scheme ii).

3.2. Manipulation and Dynamic Observation of Emulsions.

Figure 4. (a-c) LM images of the emulsion droplets manipulated by a copper wire. The

emulsion was prepared by 0.25 mg/mL g-C3N4 dispersion and hexane. (d) LM image of g-

C3N4 aggregated between two droplet interfaces.

Real-time observation can be used to uncover the behavior and stabilization mechanism of

Pickering emulsions; however, due to the invisibility of the normal emulsifier particles

(mainly due to their small size and nonfluorescence) under a light microscope, this remains a

great challenge. In the present work, we were able to conduct dynamic monitoring of the g-

C3N4-stabilized emulsion system, taking advantage of the relatively large size of the g-C3N4

sheets. In the first set of experiments, we used a thin copper wire (diameter, 100 μm) to

manipulate oil droplets under LM. As illustrated in Figure 4a, we used the probe to carefully

45

touch one of the droplets (with a diameter of ∼240 μm). Interestingly, the droplet did not

break after contacting the wire but attached to the wire instead. Next, a significant

deformation occurred to this droplet when we tried to pull away the wire. Noting that the

opposite side of this droplet adhered to several smaller droplets, this ability to physically

deform the shape indicates that the droplet possesses certain elasticity (Figure 4b). By

pulling the tip further away, the smaller droplets detached from their original sites and moved

with the wire, as an aggregate, together with the controlled large droplet, as depicted in

Figure 4c. These results demonstrate that the emulsion droplets are stable even against a

strong, physical external force. Moreover, it can be deduced that the g-C3N4 emulsifier

aggregates located at droplet junctions were connected and shared by the neighboring

droplets. This was further confirmed by examining a high-resolution LM image focused on

the interfacial g-C3N4 particles (Figure 4d). Similar results were reported by Binks and co-

workers using silica colloids as the emulsifier, which formed a monolayer at the junction and

bridged the emulsion droplets, while a water-in-oil emulsion was generated in their case.[39]

In a separate set of experiments, we monitored the motions of oil droplets in real-time, as

well as the associated particle emulsifier upon minor disturbance. It can be seen in Figure 5a

that one droplet was initially (0 s) floating in the continuous phase along with several g-C3N4

aggregates in the vicinity (indicated by the arrows). During the droplet’s random motion, a

collision occurred and g-C3N4 sheets were subsequently adsorbed to the droplet (3 s, Figure

5a). Afterward, they moved together until being anchored to a cluster of droplets (11 and 28 s,

Figure 5a). These pictures clearly exhibit how the g-C3N4 emulsifier was captured by the

moving droplet. Alternatively, the emulsion droplets might remain still, but rather the g-C3N4

particles in the continuous phase floated around and were trapped onto the droplet surface

(Figure 5b). For either pathway, the adsorption is an irreversible process owing to the

associated decrease in surface energy.[38] The in situ observation of the adsorption process

46

(more scenarios are shown in Figures S6 and S7 in the Supporting Information) for both

conditions is helpful to understand the emulsifying process, which involves the interaction

and stabilization of multiple phases (oil, water, and solids). Modeling can be carried out in

the future to further investigate the formation process of Pickering emulsions based on the

presented phenomena.

Figure 5. Continuous picturing of emulsion droplets and g-C3N4 particles using LM: (a) one

droplet moving, (b) g-C3N4 sheets moving, and (c) one droplet sliding accompanied by g-

C3N4 shifting on another droplet’s surface. The emulsion was made by shaking 0.25 mg/mL

g-C3N4 dispersion and hexane. All scale bars represent 100 μm.

In another experiment (Figure 5c), we found that the droplet (A, as noted) inside the

continuous phase appeared to slide around the larger droplet to which it was attached (B),

supported by the interfacial g-C3N4 particles, as indicated by white arrows. This motion

should be expected to result from the vibration of the continuous phase (induced by external

disturbance), and the motion ceased after droplet A was linked to other droplets (D, and then

47

C). Focussing on the g-C3N4 particles on the surface of the neighboring droplet (C), these

particles simultaneously shifted toward A during the motion of droplet (A) and were

eventually pinned to the interface, acting as the steric barrier between A and C (red arrows).

Here, the spontaneous movement of g-C3N4 accompanying the movement of the nearby

droplet reveals the kinetic nature of stabilization of Pickering emulsions, that is to say, the

solid particle aggregates spontaneously shift their locations to the droplet junctions, playing a

key role in the prevention of coalescence. The movement is probably induced by capillary

force. It is worth highlighting that more events were captured and are displayed in Figure S8.

These results correspond to the above results that g-C3N4 particles are preferably located at

the interfaces of emulsion droplets to realize high stability (Figure1).

3.3. Interaction between g-C3N4 Emulsifier and Oil Phase.

Differing from conventional inorganic emulsifiers (e.g., SiO2 and Laponite clays), g-C3N4

possesses a π-conjugated structure that could strongly interact with certain aromatic organic

solvents such as toluene. As illustrated by the LM and FM images (Figure 6a-c) of the

emulsions, the droplet coating by g-C3N4 was quite low when using hexane as the oil phase,

with g-C3N4 aggregating mainly at droplet interfaces. When the toluene solvent was used

instead, most of the oil droplets appear to be fully covered by g-C3N4 particles (Figure 6d-f).

It is worth noting that the g-C3N4 concentration was the same (0.5 mg/mL) and o/w

emulsions were generated for both cases. We ascribe the higher level of g-C3N4 coverage per

droplet with toluene to the π-π interaction between the conjugated structure of g-C3N4 and the

aromatic solvent. This attractive interaction is likely to enhance the affinity of the g-C3N4

particles to the organic solvent. This type of interaction has previously been demonstrated by

Antonietti and Wang in g-C3N4-catalyzed benzene transformations.[43, 44] Where no such π-π

attraction force was present, as that between g-C3N4 and hexane, a much lower surface

coverage resulted. Recalling that toluene has a more similar polarity to g-C3N4 compared to

48

hexane (which has very low polarity), polarity also contributes to the possible overall

interaction between toluene molecules and g-C3N4 particles. Similar results were observed

when comparing decane-water and xylene-water emulsions, as shown in Figure S9.

Figure 6. LM, FM, and overlapped images of emulsions prepared using 0.5 mg/mL g-C3N4

aqueous dispersion with (a-c) hexane and (d-f) toluene. Blue color indicates the fluorescence

from g-C3N4, which denotes the position of g-C3N4 particles. Digital photographs of

emulsions made from shaking the mixture of (g) toluene-water and (h) hexane-water. For

each mixture: left, g-C3N4 initially dispersed in water; middle, g-C3N4 initially dispersed in

organic phase; right, g-C3N4 initially dispersed in organic phase and stirred at 1500 rpm for

20 min with water. Then, the mixture was shaken to make emulsions.

49

In addition, it is worth mentioning that when the g-C3N4 concentration was at a very low

level (e.g., 0.25 mg/mL), both the hexane and toluene droplets were barely covered by the

particles (Figure S10 a-c), with clusters of g-C3N4 particles primarily lying at the droplet

interfaces. When the loading of g-C3N4 in the solution reached a high level (2 mg/mL), the g-

C3N4 sheets were able to assemble into large aggregates over the entire droplet surface for

both hexane and toluene (Figure S10 d-f), in addition to staying at the interfacial areas. The

interactions between g-C3N4 and toluene may also be reflected by another phenomenon.

When changing the initial dispersion medium of g-C3N4 from water to toluene, an emulsion

was not able to be formed anymore by just shaking the mixture (Figure 6g, middle), and the

g-C3N4 particles tended to adsorb onto the bottle wall. We consider that this is because the

preformed g-C3N4/toluene interface generated from π-π interaction and similar polarities of

these two species could not be displaced by the g-C3N4/water interface if the shearing force

was not strong enough. In this case, interface replacement could be expected to occur under

elevated shear force.[44] After vigorous stirring (1500 rpm for 20 min), a toluene-water

emulsion was readily produced by shaking (Figure 6g, right). In contrast to this behavior, for

the hexane-water system, emulsions were easily formed regardless of the order in which g-

C3N4 was dispersed, in water or hexane first (Figure 6h).

The interaction between emulsion droplets and the g-C3N4 emulsifier might affect the

geometry and therefore the assembly of solid particles on the droplet surface. Therefore, we

examined the g-C3N4 particles that were obtained by naturally drying the emulsions. As

shown in Figure 7a,b, the g-C3N4 obtained from the hexane-water emulsion dried into a

gallimaufry of aggregates. In comparison, the toluene-water system resulted in an

interconnected, sheetlike g-C3N4 form. We believe that this difference can also be assigned to

the π-π interaction between toluene and g-C3N4, as discussed above, and therefore, the g-

C3N4 particles assembled and a smooth texture was maintained after evaporation of the

50

liquids. Since the interactions between the solid stabilizers and the liquids on emulsification

can be complicated, further work to understand this behavior is warranted.

Figure 7. Scanning electron microscopy images of the g-C3N4 powders obtained by naturally

drying (a, b) hexane-water emulsion and (c, d) toluene-water emulsion.

4. CONCLUSIONS

In summary, we have investigated the stabilizing mechanism of g-C3N4-stabilized Pickering

emulsions by in situ microscopy observations. Benefitting from the large sizes and

photoluminescence of g-C3N4 particles, their positions and movements could be directly

observed using a light/fluorescence microscope in ambient environment. Under static

conditions, two stable configurations of the g-C3N4 particles with respect to the emulsion

droplets were revealed: (i) g-C3N4 particles located at interfaces of adjoining oil droplets and

(ii) g-C3N4 particles fully covering oil droplets. Further, we carried out real-time monitoring

of the g-C3N4 particles and the associated emulsion droplets, recording their movements and

adsorption kinetics toward each other. These results directly demonstrated the dynamic

51

stabilization of Pickering emulsions and may explain their superior stability compared to

normal emulsions. Moreover, the oil droplets were stable enough to be physically

manipulated and probed, permitting a study of their response to external forces. Emulsion

droplets readily deformed upon stress but did not break apart. Finally, the possible effect of

the interaction between stabilizer and organic phase on the emulsifying behavior was

examined. By comparing emulsions formed in solvents that could form π-π interactions with

the g-C3N4 stabilizer to those that cannot, this work provides significant insights into the

interfacial (liquid-liquid) properties of g-C3N4 and the stabilizing mechanism of Pickering

emulsions.

■ ASSOCIATED CONTENT

*S Supporting Information

The Supporting Information is available free of charge on the ACS Publications website at

DOI: 10.1021/acs.langmuir.8b02347.

LM and FM pictures, homemade setup, contact angle measurement, and TEM images (PDF)

■ AUTHOR INFORMATION

Corresponding Author

*E-mail: Jingsan.Xu@qut.edu.au.

Notes

The authors declare no competing financial interest.

52

■ ACKNOWLEDGMENTS

J.X. is grateful to the Discovery Early Career Researcher Award (DECRA) by the Australian

Research Council (DE160101488). Science and Engineering Faculty and Central Analytical

Research Facility (CARF) at QUT are greatly acknowledged for technical access.

References

[1] W. Ramsden, Separation of solids in the surface-layers of solutions and 'suspensions'

(observations on surface-membranes, bubbles, emulsions, and mechanical coagulation).--

preliminary account, Proc. R. Soc. London, Ser. A, 72 (1903) 156-164.

[2] S.U. Pickering, Emulsions, J. Chem. Soc., 91 (1907) 2001-2021.

[3] A. Schrade, K. Landfester, U. Ziener, Pickering-type stabilized nanoparticles by

heterophase polymerization, Chem. Soc. Rev., 42 (2013) 6823-6839.

[4] Q. Cao, Q. Cui, Y. Yang, J. Xu, C. Han, L. Li, Graphitic carbon nitride as a distinct solid

stabilizer for emulsion polymerization, Chem. Eur. J., 24 (2018) 2286-2291.

[5] J. Huang, F. Cheng, B.P. Binks, H. Yang, pH-responsive gas-water-solid interface for

multiphase catalysis, J. Am. Chem. Soc., 137 (2015) 15015-15025.

[6] L. Wei, M. Zhang, X. Zhang, H. Xin, H. Yang, Pickering emulsion as an efficient

platform for enzymatic reactions without stirring, ACS Sustainable Chem. Eng., 4 (2016)

6838-6843.

[7] H. Chen, H. Zou, Y. Hao, H. Yang, Flow Pickering emulsion interfaces enhance catalysis

efficiency and selectivity for cyclization of citronellal, ChemSusChem, 10 (2017) 1989-1995.

[8] J. Wu, G.H. Ma, Recent studies of Pickering emulsions: particles make the difference,

Small, 12 (2016) 4633-4648.

[9] B.P. Binks, S.O. Lumsdon, Influence of particle wettability on the type and stability of

surfactant-free emulsions, Langmuir, 16 (2000) 8622-8631.

53

[10] P.J. Colver, C.A.L. Colard, S.A.F. Bon, Multilayered nanocomposite polymer colloids

using emulsion polymerization stabilized by solid particles, J. Am. Chem. Soc., 130 (2008)

16850-16851.

[11] B.l.m. Brunier, N. Sheibat-Othman, M. Chniguir, Y. Chevalier, E. Bourgeat-Lami,

Investigation of four different laponite clays as stabilizers in Pickering emulsion

polymerization, Langmuir, 32 (2016) 6046-6057.

[12] N. Negrete-Herrera, J.-L. Putaux, L. David, E. Bourgeat-Lami, Polymer/laponite

composite colloids through emulsion polymerization:  influence of the clay modification level

on particle morphology, Macromolecules, 39 (2006) 9177-9184.

[13] R.F.A. Teixeira, H.S. McKenzie, A.A. Boyd, S.A.F. Bon, Pickering emulsion

polymerization using laponite clay as stabilizer to prepare armored “soft” polymer latexes,

Macromolecules, 44 (2011) 7415-7422.

[14] J. Kim, L.J. Cote, F. Kim, W. Yuan, K.R. Shull, J. Huang, Graphene oxide sheets at

interfaces, J. Am. Chem. Soc., 132 (2010) 8180-8186.

[15] S.C. Thickett, P.B. Zetterlund, Preparation of composite materials by using graphene

oxide as a surfactant in Ab initio emulsion polymerization systems, ACS Macro Lett., 2 (2013)

630-634.

[16] S. Stiller, H. Gers-Barlag, M. Lergenmueller, F. Pflücker, J. Schulz, K.P. Wittern, R.

Daniels, Investigation of the stability in emulsions stabilized with different surface modified

titanium dioxides, Colloids Surf. A, 232 (2004) 261-267.

[17] J.H. Chen, C.-Y. Cheng, W.-Y. Chiu, C.-F. Lee, N.-Y. Liang, Synthesis of

ZnO/polystyrene composites particles by Pickering emulsion polymerization, Eur. Polym. J.,

44 (2008) 3271-3279.

[18] T. Chen, P.J. Colver, S.A.F. Bon, Organic-inorganic hybrid hollow spheres prepared

from TiO2-stabilized Pickering emulsion polymerization, Adv. Mater., 19 (2007) 2286-2289.

54

[19] Z. Xu, A. Xia, C. Wang, W. Yang, S. Fu, Synthesis of raspberry-like magnetic

polystyrene microspheres, Mater. Chem. Phys., 103 (2007) 494-499.

[20] D.E. Tambe, M.M. Sharma, The effect of colloidal particles on fluid-fluid interfacial

properties and emulsion stability, Adv. Colloid Interface Sci., 52 (1994) 1-63.

[21] Y. Zheng, L. Lin, B. Wang, X. Wang, Graphitic carbon nitride polymers toward

sustainable photoredox catalysis, Angew. Chem. Int. Ed., 54 (2015) 12868-12884.

[22] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen, M.

Antonietti, A metal-free polymeric photocatalyst for hydrogen production from water under

visible light, Nat. Mater., 8 (2009) 76.

[23] H. Guo, J. Zhang, L. Ma, J.L. Chavez, L. Yin, H. Gao, Z. Tang, W. Chen, A Non-rare-

Earth ions self-activated white emitting phosphor under single excitation, Adv. Funct. Mater.,

25 (2015) 6833-6838.

[24] S.K. Lai, C. Xie, K.S. Teng, Y. Li, F. Tan, F. Yan, S.P. Lau, Polymeric carbon nitride

nanosheets/graphene hybrid phototransistors with high responsivity, Adv. Opt. Mater., 4

(2016) 555-561.

[25] J. Xu, T.J.K. Brenner, L. Chabanne, D. Neher, M. Antonietti, M. Shalom, Liquid-based

growth of polymeric carbon nitride layers and their use in a mesostructured polymer solar cell

with Voc exceeding 1 V, J. Am. Chem. Soc., 136 (2014) 13486-13489.

[26] W.-J. Ong, L.-L. Tan, Y.H. Ng, S.-T. Yong, S.-P. Chai, Graphitic carbon nitride (g-

C3N4)-based photocatalysts for artificial photosynthesis and environmental remediation: Are

we a step closer to achieving sustainability?, Chem. Rev., 116 (2016) 7159-7329.

[27] X.-H. Li, X. Wang, M. Antonietti, Solvent-free and metal-free oxidation of toluene using

O2 and g-C3N4 with nanopores: nanostructure boosts the catalytic selectivity, ACS Catal., 2

(2012) 2082-2086.

55

[28] A. Thomas, A. Fischer, F. Goettmann, M. Antonietti, J.-O. Müller, R. Schlögl, J.M.

Carlsson, Graphitic carbon nitride materials: variation of structure and morphology and their

use as metal-free catalysts, J. Mater. Chem., 18 (2008) 4893-4908.

[29] X. Zhang, X. Xie, H. Wang, J. Zhang, B. Pan, Y. Xie, Enhanced photoresponsive

ultrathin graphitic-phase C3N4 nanosheets for bioimaging, J. Am. Chem. Soc., 135 (2013) 18-

21.

[30] Q. Cui, J. Xu, X. Wang, L. Li, M. Antonietti, M. Shalom, Phenyl-modified carbon

nitride quantum dots with distinct photoluminescence behavior, Angew. Chem. Int. Ed., 55

(2016) 3672-3676.

[31] Y. Zhang, Z. Zhou, Y. Shen, Q. Zhou, J. Wang, A. Liu, S. Liu, Y. Zhang, Reversible

assembly of graphitic carbon nitride 3D network for highly selective dyes absorption and

regeneration, ACS Nano, 10 (2016) 9036-9043.

[32] J. Xu, M. Antonietti, The performance of nanoparticulate graphitic carbon nitride as an

amphiphile, J. Am. Chem. Soc., 139 (2017) 6026-6029.

[33] M. Groenewolt, M. Antonietti, Synthesis of g-C3N4 nanoparticles in mesoporous silica

host matrices, Adv. Mater., 17 (2005) 1789-1792.

[34] M. Shalom, S. Inal, C. Fettkenhauer, D. Neher, M. Antonietti, Improving carbon nitride

photocatalysis by supramolecular preorganization of monomers, J. Am. Chem. Soc., 135

(2013) 7118-7121.

[35] R. Aveyard, B.P. Binks, J.H. Clint, Emulsions stabilized solely by colloidal particles,

Adv. Colloid Interface Sci., 100 (2003) 503-546.

[36] J. Saien, A. Rezvani Pour, S. Asadabadi, Interfacial tension of the n-hexane-water

system under the influence of magnetite nanoparticles and sodium dodecyl sulfate assembly

at different temperatures, J. Chem. Eng. Data, 59 (2014) 1835-1842.

56

[37] J. Zhang, R. Pelton, Poly(N-isopropylacrylamide) microgels at the air-water interface,

Langmuir, 15 (1999) 8032-8036.

[38] Y. Chevalier, M.-A. Bolzinger, Emulsions stabilized with solid nanoparticles: Pickering

emulsions, Colloids Surf. A, 439 (2013) 23-34.

[39] T.S. Horozov, B.P. Binks, Particle-stabilized emulsions: a bilayer or a bridging

monolayer?, Angew. Chem. Int. Ed., 45 (2006) 773-776.

[40] B.P. Binks, M. Kirkland, Interfacial structure of solid-stabilised emulsions studied by

scanning electron microscopy, PCCP, 4 (2002) 3727-3733.

[41] E. Vignati, R. Piazza, T.P. Lockhart, Pickering emulsions:  interfacial tension, colloidal

layer morphology, and trapped-particle motion, Langmuir, 19 (2003) 6650-6656.

[42] Velikov, K. P.; Velev, O. D. Colloid stability; Wiley-VCH Verlag GmbH & Co. KGaA,

(2010) p277.

[43] X. Chen, J. Zhang, X. Fu, M. Antonietti, X. Wang, Fe-g-C3N4-catalyzed oxidation of

benzene to phenol using hydrogen peroxide and visible light, J. Am. Chem. Soc., 131 (2009)

11658-11659.

[44] G. Chen, D. Tao, An experimental study of stability of oil-water emulsion, Fuel Process.

Technol., 86 (2005) 499-508.

57

Supporting Information

Direct Observation of Carbon Nitride Stabilized Pickering

Emulsions

Chenhui Han,† Qianling Cui,§ Peng Meng,† Eric R. Waclawik,† Hengquan Yang‡ and Jingsan

Xu*†

† School of Chemistry, Physics and Mechanical Engineering, Queensland University of

Technology, Brisbane, QLD 4001, Australia

§ State Key Laboratory for Advanced Metals and Materials, School of Materials Science

and Engineering, University of Science and Technology Beijing, Beijing 100083, China

‡ School of Chemistry and Chemical Engineering, Shanxi University, Wucheng Road 92,

Taiyuan 030006, China

*Email: Jingsan.Xu@qut.edu.au

58

Figure S1. (a,b) TEM images and (c) SEM image of the g-C3N4 aggregates. (d) UV-vis

spectra and (e) photoluminescence spectra of g-C3N4 bulk powder and water dispersion.

Figure S2. (a) The supposed structure of g-C3N4. (b) from left to right: picture of g-C3N4

powder, aqueous dispersion, and g-C3N4 stabilized emulsion, and (c) corresponding

fluorescence photographs under 365 nm UV light.

59

Figure S3. Home-made setup for in-situ observing Pickering emulsions under microscopes.

Figure S4. Contact angle measurement of 9 independent batches.

60

Figure S5. (a) Hexane-in-water emulsion prepared by pre-treated g-C3N4. The pristine g-

C3N4 powder was thoroughly washed by acetone and ethanol to remove any possible small

molecules, which was then dried in a vacuum oven. Then the obtained g-C3N4 was dispersed

in water to form a suspension. (b) The pristine g-C3N4 was first dispersed in water by

sonication, then the g-C3N4 powder was fully removed by multiple centrifugation (6000 rpm

for 10 min) and filtering cycles. The aqueous supernate did not show any activity for

emulsification. (c) Blank sample as a reference (pure water and hexane after shaking and

standing for 30 min).

61

Figure S6. Continuous picturing of emulsion droplets and g-C3N4 particles using LM at three

locations (A-C): droplets moved and captured g-C3N4 particles. The emulsion was made by

shaking 0.25 mg/ml g-C3N4 dispersion and hexane. All scale bars represent 100 µm.

62

Figure S7. Continuous picturing of emulsion droplets and g-C3N4 particles using LM at three

locations (A-C): g-C3N4 particles moved and were captured by droplets. The emulsion was

made by shaking 0.25 mg/ml g-C3N4 aqueous dispersion and hexane. All scale bars represent

100 µm.

63

Figure S8. Continuous picturing of emulsion droplets sliding along the other droplets at

different locations. The emulsion was made by shaking 0.25 mg/ml g-C3N4 aqueous

dispersion and hexane. All scale bars represent 100 µm.

64

Figure S9. (a, c) LM and (b, d) corresponding FM images of Pickering emulsions prepared

by 0.5 mg/ml g-C3N4 dispersion with decane (a, b) and p-xylene (c, d). Blue color indicates

the fluorescence emitted by g-C3N4 which denotes the position of g-C3N4 particles.

65

Figure S10. (a, d) LM, (b, e) FM and (c, f) overlapped images of Pickering emulsions

prepared by g-C3N4 aqueous dispersion with toluene: (a-c) 0.25 mg/ml g-C3N4 and (d-f) 2

mg/ml g-C3N4.

66

Chapter 3: Application of g-C3N4 Stabilized Emulsion in Catalysis

This chapter is made up of the published journal article below:

Chenhui Han, Peng Meng, Eric R. Waclawik, Chao Zhang, Xin-Hao Li, Hengquan Yang,

Markus Antonietti, and Jingsan Xu*. Palladium/Graphitic Carbon Nitride (g-C3N4) Stabilized

Emulsion Microreactor as a Store for Hydrogen from Ammonia Borane for Use in Alkene

Hydrogenation. Angewandte Chemie International Edition 57.45 (2018): 14857-14861.

68

3.1 Introductory remarks

In this work, a tandem reaction was conducted in a Pickering emulsion-based microreactor,

with Pd/g-C3N4 acting as stabilizer. The Pd/g-C3N4 catalyst at the oil-water interface catalyze

both hydrolysis of ammonia borane (AB) in the water phase and hydrogenation of alkenes in

the oil phase, with hydrogen transfer happening at biphasic interface. In such a system, the

hydrogen utilization efficiency can reach unity, with all the liberated hydrogen from AB

consumed by the following hydrogenation reaction. This is benefiting mainly from the large

interface between two phases and the hydrogen storage capability of Pickering emulsions.

Such an approach is advantageous for more economical hydrogen utilization over

conventional systems. Furthermore, the emulsion microreactor can be applied to a range of

alkene substrates, with the conversion rates achieving >95% by a simple modification. This

study expands the application of Pickering emulsions and opens a new avenue for developing

more economical and efficient hydrogenation processes for the production of a wide range of

valuable chemicals.

69

3.2 Article 2

Palladium/Graphitic Carbon Nitride (g-C3N4) Stabilized Emulsion

Microreactor as a Store for Hydrogen from Ammonia Borane for Use in

Alkene Hydrogenation

Chenhui Han,a Peng Meng,a Eric R. Waclawik,a Chao Zhang,b Xin-Hao Li,c Hengquan

Yang,d Markus Antonietti,e and Jingsan Xu*a

a. School of Chemistry, Physics and Mechanical Engineering Queensland University of

Technology, Brisbane, QLD 4001 (Australia)

b. College of Materials Science and Engineering, Donghua University, Shanghai, 201620

(China)

c. School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai,

200240 (China)

d. School of Chemistry and Chemical Engineering, Institute of Molecular Science, Shanxi

University, Taiyuan, 030006 (China)

e. Max Planck Institute of Colloids and Interfaces, Potsdam, 14476 (Germany)

Abstract: Direct hydrogenation of C=C double bonds is a basic transformation in organic

chemistry which is vanishing from simple practice because of the need for pressurized

hydrogen. Ammonia borane (AB) has emerged as a hydrogen source through its safety and

high hydrogen content. However, in conventional systems, the hydrogen liberated from the

high-cost AB cannot be fully utilized. Herein, we develop a novel Pd/g-C3N4 stabilized

Pickering emulsion microreactor, in which alkenes are hydrogenated in the oil phase with

hydrogen originating from AB in the water phase, catalyzed by the Pd nanoparticles at the

interfaces. This approach is advantageous for more economical hydrogen utilization over

conventional systems. The emulsion microreactor can be applied to a range of alkene

substrates, with the conversion rates achieving >95% by a simple modification.

70

Hydrogenation of C=C double bonds is one of the most important transformations in organic

chemistry and is widely applied in pharmaceutical and fine chemical industries.[1-3] This

transformation can be realized by direct hydrogenation with gaseous H2,[4, 5] requiring

elevated hydrogen pressure, safety measures and specialist reactors.[6, 7] Alternatively, the

transformation can be achieved by transfer hydrogenation, which refers to the transfer of

hydrogen to the target molecule from a liquid hydrogen donor, such as an alcohol.[8, 9] This

approach is environmentally more benign and is receiving increasing attention.[9-11] Recently,

ammonia borane (NH3BH3, AB) has proven to be an excellent hydrogen source owing to its

stability, aqueous solubility and high hydrogen content (19.4 wt%).[12-18] It is convenient and

useful for the efficient hydrogenation of C=C, C=N, and N=O in the presence of Pd catalysts,

where the Pd catalyzes both the AB hydrolysis and the hydrogenation.[19, 20] One potential

drawback of using AB is that the cost is still quite high.[21, 22] Another drawback that needs

redressing is that the H2 produced from AB partly escapes from the reaction system due to

local H2 overproduction, resulting in a waste of the reagent. In this context, the full use of AB

for organic transformations is a significant task.

Pickering emulsions are emulsions stabilized by solid particles rather than surfactant

molecules, in which oil/organic solvent or water is dispersed into the other phase in the form

of micrometer-sized droplets.[23-25] Taking advantage of the unique organic-solid-water

morphology, Pickering emulsions can convert traditional biphasic reactions to a

macroscopically homogeneous emulsion phase, with the incompatible reactants confined in

either oil or the aqueous phase.[26-29] This approach exposes an abundant reactive interface

surface area and is free of extra solvents or surfactants/transfer agents often required to make

the reaction system potentially miscible. In a recent study, it has been demonstrated that

graphitic carbon nitride (g-C3N4) can serve as a low-cost Pickering stabilizer.[30] Unlike the

commonly used SiO2 emulsifier, no surface modification of g-C3N4 is required. In addition,

71

g-C3N4 is a well-known support for the immobilization of metal nanoparticles and molecular

catalysts.[31-33] It also has excellent stability across a broad range of conditions and

solvents.[34]

Herein, Pd nanoparticles (NPs) were loaded onto g-C3N4 sheets by simple photo-deposition to

endow the material specific catalytic activity while keeping the emulsifying capability. The

obtained Pd/g-C3N4 was then employed to stabilize Pickering emulsion for conducting an

efficient alkene hydrogenation in the oil phase, where the hydrogen was generated in situ

from AB hydrolysis in water, with Pd NPs acting as a bifunctional catalyst located at the

interfaces. It is shown that the efficiency of hydrogen utilization in this Pickering emulsion

system can reach unity, that is, the added AB is used without loss for the hydrogenation

occurring in the other phase. Kinetic insights and in situ observation show that the emulsion

is capable of temporally storing hydrogen that is not directly consumed, and the stored

hydrogen reacts with the alkene afterward until depleted.

Figure 1a shows a TEM image of 2 wt% Pd/g-C3N4. Pd NPs have a size distribution centered

at 2~6 nm and were uniformly dispersed on the g-C3N4 support. HRTEM imaging (Figure 1b)

of a single Pd nanoparticle measured a fringe spacing of 0.225 nm, which corresponds to the

(111) planes of face-centered cubic palladium. Measurements of the g-C3N4 sheet supports

reveal that they were typically weakly ordered. Elemental line-scans across two different

locations show the distribution of the Pd element, which corresponds with the dark-field

TEM image (Figure 1c). The chemical nature of the supported Pd NPs was studied by X-ray

photoelectron spectroscopy (XPS). Fitted XPS spectra (Figure 1d) of Pd 3d3/2 and 3d5/2

orbitals are strong evidence that the Pd NPs were predominantly in metallic state (Pd0), with

the presence of a small proportion of PdO and PdO2.[35, 36] Importantly, the Pd/g-C3N4

composite can act as an efficient stabilizer for Pickering emulsions, an effect originating from

74

As shown in Figure 2c, experiments with 0.5 wt% Pd loading (Figure S2a) show relatively

low activity, and 0.31 mmol of ethylbenzene was obtained after 18 h reaction time. For the

nanocomposites with 2 and 5 wt% Pd loading (TEM images shown in Figure S2b), the

reactions proceeded much more quickly, and 0.32 mmol of ethylbenzene was already reached

in the first 8 h. Then the reactions slowed down owing to the consumption of AB, and 0.35

mmol of ethylbenzene was eventually generated. In this regard, the H2 released from AB (3

equiv. of AB) fully reacted with styrene; in other words, the hydrogen utilization efficiency

(HUE), defined as HUE = [(mol of ethylbenzene)/3 × (mol of AB)] × 100%, reached unity (>

99%). Control experiments showed that no ethylbenzene was detected when only g-C3N4 was

used, verifying that Pd provides the catalytic sites for the coupled reactions.

Next, a series of reference and blind experiments were conducted for better understanding the

reactions in the Pickering emulsion microreactors. As shown in Figure 2d, when AB was

replaced by 0.35 mmol of gaseous H2, a very low amount of ethylbenzene and HUE were

observed, due to slow mass transfer from gas to the liquid phase (Figure 2d, entry 2).

Additionally, when AB was used but no emulsification was applied, 0.11 mmol of

ethylbenzene and a HUE of 33% was obtained, resulting from the limited macro-interface

between the two immiscible phases (Figure 2d, entry 3). Applying stirring (500 rpm) led to a

slightly increased interface area and thus the catalytic performance slightly improved (Figure

2d, entry 4). In another experiment, ethanol was added to generate a miscible, homogenous

solvent for the reactions. In this condition, the catalytic performances were still low,

achieving only 0.15 mmol of ethylbenzene and a HUE of 45% (Figure 2d, entry 5). These

comparison studies confirmed the efficacy of a combined dehydrogenation-hydrogenation in

the emulsion microreactor. This statement can also be verified by using water-soluble alkenes

(allylamine and 2,5-dihydrofuran) for more efficient hydrogenation in Pickering emulsions

than in normal aqueous solutions (Figure S3).

76

consumption of AB and the accumulation of hydrogen. Moreover, the gap between the

kinetic profiles became narrower when the temperature was raised to 50oC, with the reaction

rate of dehydrogenation and hydrogenation becoming very close. These results indicate that

in this emulsion-based biphasic reaction, the majority of the hydrogen from AB is consumed

by the subsequent styrene hydrogenation, catalyzed by the interfacial Pd sites via direct

hydrogen transfer [Eq. (S1) in the Supporting Information]. The abundant reactive and

catalytic sites at the droplet surfaces are highly beneficial to the coupled reactions.

Nevertheless, a small amount of locally overproduced hydrogen is not available to

immediately react with styrene, which form gaseous H2 and then react with the styrene

molecules [Eq. (S2) and (S3)].

Based on the above discussion and considering the overall HUE reached unity at the end of

the reactions, we propose that a small fraction of the hydrogen produced at the beginning of

the reaction which is not instantly consumed by the slower hydrogenation reaction did not

escape, but conserved in the system, that is, within the Pickering hybrid system which is

effectively acting as a hydrogen buffer. It is well documented that Pd is able to chemosorb a

large number of hydrogen.[37-39] These chemisorbed hydrogen are highly active and can react

with styrene afterward. The second hydrogen storage mechanism is hydrogen microbubbles,

which—due to the viscosity and roughness of the Pickering system—can stay trapped at the

g-C3N4 particle surface in the continuous aqueous phase (Figure S5) as well as in the

dispersed organic droplets. As shown in Figure S6, micro H2 bubbles were formed in the

styrene droplets after the injection of AB, and the bubbles grew a bit larger in the next 10 min.

Interestingly, the bubbles shrink afterward and then completely vanished, owing to the

reaction with styrene. The hydrogen buffer effect of the Pd/g-C3N4 stabilized Pickering

emulsion is schematically shown in Figure 3d.

78

which was expected due to the reduced catalyst amounts. More importantly, the HUE

concurrently reduced to 54% and 27%, respectively, even when the reaction was prolonged to

45 h (Figure S7). These results strongly imply that in these conditions, the liberated H2

escaped from the emulsion system and did not participate further in the hydrogenation, most

likely resulting from insufficient Pd adsorption sites and emulsion stabilizer (Figure S8).

Moreover, the hydrogen storage capability can be taken advantage of to improving the

transformation rate of styrene, and meanwhile maintaining unity HUE. As shown in Figure

4b, when the AB loading was increased to 8 and 12 mg, 0.7 and 1.05 mmol of ethylbenzene

were obtained, respectively. It is worth pointing out that the gap between the kinetic profiles

of hydrogen evolution and ethylbenzene generation became larger at higher AB amounts,

which is reasonable as more hydrogen was released. Correspondingly, the profiles almost

totally matched when the AB input was down to 2 mg.

Furthermore, the recyclability of the Pd/g-C3N4 catalyst in the current Pickering emulsion

microreactor was evaluated. After simply washing, the nanomaterials can be reused for the

next-cycle emulsification and catalytic reaction (Figure S9). As shown in Figure 4c, high

catalytic activity toward the hydrogen evolution-hydrogenation reaction was well maintained

up to 10 cycles, with no decline of HUE and ethylbenzene production. TEM observation and

XPS measurement demonstrate that after 10 cycles, the Pd NPs grew a bit larger, while their

chemical state remained unchanged (Figure S10). Turnover frequency (TOF) [(mol of

ethylbenzene/mol of Pd) per hour] was determined and only showed a slight decrease upon

each catalysis cycle (Figure S11), probably because of the slow growth of Pd NPs, indicating

the superior reusability of the Pd/g-C3N4 catalyst.

Finally, the conversion rate of alkene can be substantially promoted by simply modifying the

Pickering emulsion microreactor. Briefly, cyclohexane was used as a solvent to dissolve 0.35

mmol of styrene and then emulsified with Pd/g-C3N4 aqueous dispersion, followed by the

79

addition of AB (0.117 mmol) as described above (see Supporting Information for details). In

this situation, the reagents can fully react with each other and resulted in a final conversion of

97% (Figure 4d, entry 1). More importantly, this modification enabled us to largely extend

the range of the alkene substrates, as some of the educts cannot be directly emulsified. As

shown in entry 2-5 (Figure 4d), four substrates with very different types of C=C double

bonds were chosen for hydrogenation, and a high conversion rate and HUE (95-99%) were

achieved.

In summary, we have presented a novel Pd/g-C3N4 stabilized Pickering emulsion

microreactor to couple hydrogen generation from AB with hydrogenation of alkenes at the

oil-water interfaces of the droplets. The hydrogen utilization efficiency can reach unity in

such an emulsion system, mainly due to the localized production, storage, and consumption

of hydrogen on the Pd NPs. Detailed analysis and observations revealed that the Pickering

emulsion can effectively store hydrogen via atomic adsorption in the Pd nanoparticles and

microbubble formation within the liquid phases. The stored hydrogen then reacted with

alkene molecules until complete consumption. This hydrogen storage capability obviously

played a key role in the efficient use of hydrogen. The Pd/g-C3N4 catalyst illustrated excellent

reusability and turned out to be rather generally applicable to various alkene molecules.

Importantly, the conversion rate of alkene can reach > 95% by simply modifying the

Pickering emulsion microreactors. This work opens a new avenue for developing more

economical and efficient hydrogenation processes for the production of a wide range of

valuable chemicals.

80

Acknowledgments

J.X. is grateful to the Discovery Early Career Researcher Award (DECRA) by Australian

Research Council (DE160101488). Science and Engineering Faculty and Central Analytical

Research Facility (CARF) at QUT are greatly acknowledged for technical assistance.

Conflict of interest

The authors declare no conflict of interest.

Keywords: emulsion microreactors, graphitic carbon nitrides, hydrogenation, palladium

nanoparticles

Reference

[1] L. Cerveny, Catalytic hydrogenation, Vol. 27, Elsevier, Amsterdam, 1986.

[2] E.N. Jacobsen, A. Pfaltz, H. Yamamoto, Comprehensive asymmetric catalysis:

supplement 1, Vol. 1, Springer Science & Business Media, Cham, 2003.

[3] B. Chen, U. Dingerdissen, J.G.E. Krauter, H.G.J. Lansink Rotgerink, K. Möbus, D.J.

Ostgard, P. Panster, T.H. Riermeier, S. Seebald, T. Tacke, H. Trauthwein, New developments

in hydrogenation catalysis particularly in synthesis of fine and intermediate chemicals, Appl.

Catal. A, 280 (2005) 17-46.

[4] T. Sato, W. Mori, C. Kato, E. Yanaoka, T. Kuribayashi, R. Ohtera, Y. Shiraishi, Novel

microporous rhodium(II) carboxylate polymer complexes containing metalloporphyrin:

syntheses and catalytic performances in hydrogenation of olefins, J. Catal., 232 (2005) 186-

198.

81

[5] X. Ma, T. Jiang, B. Han, J. Zhang, S. Miao, K. Ding, G. An, Y. Xie, Y. Zhou, A. Zhu,

Palladium nanoparticles in polyethylene glycols: Efficient and recyclable catalyst system for

hydrogenation of olefins, Catal. Commun., 9 (2008) 70-74.

[6] S.J. Roseblade, A. Pfaltz, Iridium-catalyzed asymmetric hydrogenation of olefins, Acc.

Chem. Res., 40 (2007) 1402-1411.

[7] F.R.S. Pederzolli, S.I. Wolke, R.G. da Rosa, Recyclable rhodium-Cp′-heterogenized

catalysts for hydrogenation of olefins, J. Catal., 360 (2018) 201-212.

[8] C.H. Hao, X.N. Guo, Y.T. Pan, S. Chen, Z.F. Jiao, H. Yang, X.Y. Guo, Visible-light-

driven selective photocatalytic hydrogenation of cinnamaldehyde over Au/SiC catalysts, J.

Am. Chem. Soc., 138 (2016) 9361-9364.

[9] D. Wang, D. Astruc, The golden age of transfer hydrogenation, Chem. Rev., 115 (2015)

6621-6686.

[10] R.A. Johnstone, A.H. Wilby, I.D. Entwistle, Heterogeneous catalytic transfer

hydrogenation and its relation to other methods for reduction of organic compounds, Chem.

Rev., 85 (1985) 129-170.

[11] G. Brieger, T.J. Nestrick, Catalytic transfer hydrogenation, Chem. Rev., 74 (1974) 567-

580.

[12] Z. Huang, T. Autrey, Boron–nitrogen–hydrogen (BNH) compounds: recent

developments in hydrogen storage, applications in hydrogenation and catalysis, and new

syntheses, Energy Environ. Sci., 5 (2012) 9257.

[13] J.-M. Yan, X.-B. Zhang, T. Akita, M. Haruta, Q. Xu, One-step seeding growth of

magnetically recyclable Au@Co core-shell nanoparticles: highly efficient catalyst for

hydrolytic dehydrogenation of ammonia borane, J. Am. Chem. Soc., 132 (2010) 5326-5327.

82

[14] N.-Z. Shang, C. Feng, S.-T. Gao, C. Wang, Ag/Pd nanoparticles supported on amine-

functionalized metal-organic framework for catalytic hydrolysis of ammonia borane, Int. J.

Hydrog. Energy, 41 (2016) 944-950.

[15] H. Dai, J. Su, K. Hu, W. Luo, G. Cheng, Pd nanoparticles supported on MIL-101 as

high-performance catalysts for catalytic hydrolysis of ammonia borane, Int. J. Hydrog.

Energy, 39 (2014) 4947-4953.

[16] W. Chen, J. Ji, X. Duan, G. Qian, P. Li, X. Zhou, D. Chen, W. Yuan, Unique reactivity

in Pt/CNT catalyzed hydrolytic dehydrogenation of ammonia borane, Chem. Commun., 50

(2014) 2142-2144.

[17] Z. Li, T. He, D. Matsumura, S. Miao, A. Wu, L. Liu, G. Wu, P. Chen, Atomically

dispersed Pt on the surface of Ni particles: synthesis and catalytic function in hydrogen

generation from aqueous ammonia–borane, ACS Catal., 7 (2017) 6762-6769.

[18] T.-J. Zhao, Y.-N. Zhang, K.-X. Wang, J. Su, X. Wei, X.-H. Li, General transfer

hydrogenation by activating ammonia-borane over cobalt nanoparticles, RSC Advances, 5

(2015) 102736-102740.

[19] C.E. Hartmann, V. Jurcik, O. Songis, C.S. Cazin, Tandem ammonia borane

dehydrogenation/alkene hydrogenation mediated by [Pd(NHC)(PR3)] (NHC = N-heterocyclic

carbene) catalysts, Chem Commun (Camb), 49 (2013) 1005-1007.

[20] L.-H. Gong, Y.-Y. Cai, X.-H. Li, Y.-N. Zhang, J. Su, J.-S. Chen, Room-temperature

transfer hydrogenation and fast separation of unsaturated compounds over heterogeneous

catalysts in an aqueous solution of formic acid, Green Chemistry, 16 (2014) 3746-3751.

[21] B. Peng, J. Chen, Ammonia borane as an efficient and lightweight hydrogen storage

medium, Energy Environ. Sci., 1 (2008) 479-483.

[22] The price of ammonia borane complex for technical grade 90% is ca. $150 per 10 g and

for 97% is $390 per 10 g, according to the Sigma-Aldrich website.

83

[23] R. Aveyard, B.P. Binks, J.H. Clint, Emulsions stabilized solely by colloidal particles,

Adv. Colloid Interface Sci., 100 (2003) 503-546.

[24] J. Xu, A. Ma, T. Liu, C. Lu, D. Wang, H. Xu, Janus-like Pickering emulsions and their

controllable coalescence, Chem. Commun., 49 (2013) 10871-10873.

[25] P. Facal Marina, J. Xu, X. Wu, H. Xu, Thinking outside the box: placing hydrophilic

particles in an oil phase for the formation and stabilization of Pickering emulsions, Chem.

Sci., 9 (2018) 4821-4829.

[26] W. Zhang, L. Fu, H. Yang, Micrometer‐scale mixing with pickering emulsions: biphasic

reactions without stirring, ChemSusChem, 7 (2014) 391-396.

[27] H. Chen, H. Zou, Y. Hao, H. Yang, Flow Pickering emulsion interfaces enhance

catalysis efficiency and selectivity for cyclization of citronellal, ChemSusChem, 10 (2017)

1989-1995.

[28] H. Yang, L. Fu, L. Wei, J. Liang, B.P. Binks, Compartmentalization of incompatible

reagents within Pickering emulsion droplets for one-pot cascade reactions, J. Am. Chem. Soc.,

137 (2015) 1362-1371.

[29] A. Ma, J. Xu, L. Yu, X. Zhang, D. Wang, H. Xu, Adsorbed emulsion droplets: capping

agents for in situ heterogeneous engineering of particle surfaces, Chem. Commun., 49 (2013)

11563-11565.

[30] J. Xu, M. Antonietti, The performance of nanoparticulate graphitic carbon nitride as an

amphiphile, J. Am. Chem. Soc., 139 (2017) 6026-6029.

[31] X. Wang, X. Chen, A. Thomas, X. Fu, M. Antonietti, Metal‐containing carbon nitride

compounds: a new functional organic-metal hybrid material, Adv. Mater., 21 (2009) 1609-

1612.

[32] J. Sun, J. Zhang, M. Zhang, M. Antonietti, X. Fu, X. Wang, Bioinspired hollow

semiconductor nanospheres as photosynthetic nanoparticles, Nat. Commun., 3 (2012) 1139.

84

[33] X. Chen, J. Zhang, X. Fu, M. Antonietti, X. Wang, Fe-g-C3N4-catalyzed oxidation of

benzene to phenol using hydrogen peroxide and visible light, J. Am. Chem. Soc., 131 (2009)

11658-11659.

[34] Z. Zhou, Y. Zhang, Y. Shen, S. Liu, Y. Zhang, Molecular engineering of polymeric

carbon nitride: advancing applications from photocatalysis to biosensing and more, Chem.

Soc. Rev., 47 (2018) 2298-2321.

[35] M.C. Militello, S.J. Simko, Elemental palladium by XPS, Surf. Sci. Spectra, 3 (1994)

387-394.

[36] K.S. Kim, A. Gossmann, N. Winograd, X-ray photoelectron spectroscopic studies of

palladium oxides and the palladium-oxygen electrode, Anal. Chem., 46 (1974) 197-200.

[37] L. McKeehan, The crystal structures of the system palladium-hydrogen, Phys. Rev., 21

(1923) 334.

[38] F. A. Lewis, The palladium/hydrogen system, Academic Press, London and New York,

1967.

[39] F. Manchester, A. San-Martin, J. Pitre, The H-Pd (hydrogen-palladium) system, J. phase

equilib., 15 (1994) 62-83.

85

Supporting Information

Palladium/Graphitic Carbon Nitride (g-C3N4) Stabilized

Emulsion Microreactor as a Store for Hydrogen from Ammonia

Borane for Use in Alkene Hydrogenation

Chenhui Han,a Peng Meng,a Eric R. Waclawik,a Chao Zhang,b Xin-Hao Li,c Hengquan

Yang,d Markus Antonietti,e and Jingsan Xu*a

a. School of Chemistry, Physics and Mechanical Engineering Queensland University of

Technology, Brisbane, QLD 4001 (Australia)

b. College of Materials Science and Engineering, Donghua University, Shanghai, 201620

(China)

c. School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai,

200240 (China)

d. School of Chemistry and Chemical Engineering, Institute of Molecular Science, Shanxi

University, Taiyuan, 030006 (China)

e. Max Planck Institute of Colloids and Interfaces, Potsdam, 14476 (Germany)

86

Experimental

1. Materials:

Cyanuric acid (98%, Sigma-Aldrich), Melamine (99%, Sigma-Aldrich), Potassium

hexachloropalladate (K2PdCl6, 99%, Sigma-Aldrich), Styrene (99%, SigmaAldrich),

Ethylbenzene (99.8%, Sigma-Aldrich), Methanol (99.9%, Sigma-Aldrich), ammonia borane

(AB, 90%, Sigma-Aldrich), Toluene (AR, Fisher Scientific), Cyclohexane (99.5%, Sigma-

Aldrich), 1-Dodecene (99%, Sigma-Aldrich), Bicyclo[2.2.1]hept-2-2ene (99%, Sigma-

Aldrich), Trans-Stilbene (96%, Sigma-Aldrich), Cinnamyl alcohol (98%, Sigma-Aldrich). All

chemicals were used as received without further purification.

2. Synthesis of g-C3N4:

Graphitic carbon nitride (g-C3N4) was synthesized using the cyanuric acid-melamine

supramolecular precursor according to previous reports. Typically, 1:1 molar ratio of

cyanuric acid and melamine were mixed in water and shaken for 24 hours at room

temperature. Then the obtained milky suspension was centrifuged and dried at 50 oC under

vacuum. Afterward, the precursor powder was annealed in capped crucibles at 550 oC for 4 h

at a heating rate of 2.3 oC/min in a nitrogen atmosphere. After cooling down the yellow g-

C3N4 powder was collected.

3. Synthesis of Pd/g-C3N4 catalysts:

Pd/g-C3N4 catalysts with different Pd loading were prepared by an in situ photodeposition

method. 50 mg of g-C3N4 was pre-dispersed in 20 ml of deionized water by bath sonication

and then the desired amount of K2PdCl6 aqueous solution (0.02 M) was added and stirred for

87

10 min. Afterward, 30 ml of methanol was added into the beaker and the mixture was then

irradiated under a 50 W UV-light LED for two hours. Finally, the obtained brown solid was

washed three times with water and one time with ethanol and dried at 60 oC under vacuum.

4. Materials characterization:

The morphology of prepared catalysts was studied using a JEOL 2100 transmission electron

microscopy (TEM) equipped with a Gatan Orius SC1000 CCD camera. The accelerating

voltage of TEM was 200 kV. X-ray photoelectron spectroscopy (XPS) analysis was

performed with a Kratos Axis Ultra photoelectron spectrometer using mono Al Kα (1486.6

eV) X-ray. Nikon Eclipse Ni light microscope and Nikon A1R confocal microscope were

used to image the emulsions. Products were analyzed by an Agilent Gas Chromatograph

(Agilent HP6890) equipped with an FID detector and HP-5 column. The conductivity of the

water phase was monitored by a conductivity meter (DDS-11A, Rex Company, Shanghai,

China).

5. Catalysis activity test:

In a typical procedure, 2 mg of 2 wt% Pd/g-C3N4 was first dispersed in 1 ml of deionized

water in a 5 ml glass vial, then 1 ml of styrene was added and the mixture was emulsified by

manually shaking for around 1 min. Afterward, 0.5 ml of 8 mg/ml AB (4 mg) aqueous

solution was injected into the above vial to initiate the reactions. Then the vial was sealed and

put into a constant temperature oven (30, 40 or 50 oC) for 18 h. After the reaction, the

emulsion was filtered to remove the catalysts and meanwhile the organic phase was separated

from the water phase. Then the organic phase was analyzed by gas chromatograph (Agilent

HP6890) to determine the amount of ethylbenzene. The produced hydrogen was determined

according to the amount of the decomposed AB (NH3BH3 + 2H2O → NH4BO2 + 3H2), which

88

can be quantified by measuring the conductivity of the water phase. Upon AB decomposition,

NH4 + and BO2- are produced and the conductivity of the final solution increases linearly (see

the standard curve below).

Calibration curve: Solution conductivity vs. decomposed AB to determine the produced

hydrogen.

6. Recycling of the catalysts:

To avoid the loss of catalysts during the regeneration process, the reaction vials were directly

centrifuged at 6000 rpm to demulsify and remove the liquid phase, then the used catalysts

were washed three times using acetone and one time using ethanol. Finally, the washed

catalyst along with the vial was dried in vacuum (60 oC) and then used for the next cycle.

7. Emulsion modification for >95% conversion: Styrene and other substrates dissolved

in hexane as the oil phase

89

(1) Styrene, 1-dodecene and bicyclo[2.2.1]hept-2-2ene:

2 mg of 2 wt% Pd/g-C3N4 was first dispersed in 1 ml deionized water in a 5 ml glass vial,

then 0.35 mmol of different substrates were dissolved in 1 ml of cyclohexane as oil phase and

added into the reaction vial. The mixture was emulsified by manually shaking for around 1

min. Afterward, 0.5 ml of 8 mg/ml ammonia borane (4mg) aqueous solution was injected into

the above vial. Finally, the vial was sealed and put into a constant temperature oven (30, 40

and 50 oC) for 18 h. After the reaction, the emulsion was filtered to remove catalysts and

obtain the oil phase and water phase respectively. The organic phase was analyzed by gas

chromatograph (Agilent GC HP6890) to determine the amount of the product.

(2) Cinnamyl alcohol:

The procedure was the same as the above three substrates except for the solvent (cyclohexane)

which was replaced by the same amount of toluene.

(3) Trans-Stilbene:

This substrate was hydrogenated under 80 oC and all the other reaction conditions were the

same as that of cinnamyl alcohol.

8. Hydrogenation of water-soluble allylamine and 2,5-dihydrofuran in aqueous solution

and Pickering emulsions.

In aqueous solution: 2 mg of 2 wt% Pd/g-C3N4 and 0.35 mmol of water-soluble substrates

(Allylamine and 2, 5-dihydrofuran, respectively) were first added into 1 ml of ultrapure water

in a 5 ml glass vial, then 0.5 ml of 8 mg/ml AB solution (4 mg) was injected into the vial, no

any oil phase was used in this case. Finally, the vial was put into a constant temperature oven

(40 oC) for 18 h and then the products were analyzed by GC.

90

In Pickering emulsion: 2 mg of 2 wt% Pd/g-C3N4 and 0.35 mmol of water-soluble substrates

(Allylamine and 2, 5-dihydrofuran, respectively) were first added into 1 ml of ultrapure water

in a 5 ml glass vial, then 1 ml of cyclohexane was added and the mixture was emulsified by

manually shaking for around 1 min. Afterward, 0.5 ml of 8 mg/ml AB solution (4 mg) was

injected into the vial. Finally, the vial was put into a constant temperature oven (40 oC) for 18

h then the products were analyzed by GC.

91

Figure S1. Light microscopy image of 2 wt% Pd/C3N4 stabilized oil-in-water Pickering

emulsions. The water phase was colored by 0.5 mg of methylene blue in order to identify the

emulsion type.

Figure S2. TEM images of (a) 0.5 wt% Pd/g-C3N4 and (b) 5 wt% Pd/g-C3N4

92

Figure S3. Yield and HUE of hydrogenation of allylamine and 2,5-dihydrofuran in aqueous

solution and in the Pickering emulsion system.

Figure S4. Reaction-time dependent HUE of styrene hydrogenation under 30, 40 and 50 oC

in the Pickering emulsion system.

94

Figure S7. Kinetic study of dehydrogenation and hydrogenation with different amount of 2

wt% Pd/g-C3N4 at 40 oC. Brown lines 0.5 mg, red lines 1 mg, and blue lines 1.5 mg. Dashed

lines illustrate the hydrogen amount and solid lines illustrate the ethylbenzene amount.

Figure S8. Confocal images of Pickering emulsions stabilized by different amount of 2 wt%

Pd/g-C3N4.

0.5 mg 1 mg

2 mg 4 mg

96

Figure S11. TOF of the 2 wt% Pd/g-C3N4 catalyst for 10 cycles of styrene hydrogenation at

40 oC in a Pickering emulsion microreactor.

Chemical equations of styrene hydrogenation using AB as the hydrogen source

97

Chapter 4: Mechanically Switchable Emulsions

This chapter is made up of the manuscript below:

Chenhui Han, Eric R. Waclawik, Xiaofei Yang, * Peng Meng, Hengquan Yang,* Ziqi Sun and

Jingsan Xu*. Mechanically switchable emulsions. (Manuscript submitted to The Journal of

Physical Chemistry Letters).

99

4.1 Introductory remarks

This chapter includes one article submitted to The Journal of Physical Chemistry Letters.

In this work, we reported an emulsion stabilized by an organic-inorganic hybrid that in situ

formed by combining an oil-soluble molecule, stearic acid, with water-dispersible Al2O3

nanofibers via chemisorption at the oil-water interface. By manipulating the adsorption-

desorption volume of stearic acid, the amphiphilicity of the hybrid could be reversibly

switched, which resulted in a reversibly transform of emulsion types between oil-in-water

(o/w) and water-in-oil (w/o). The detailed phase inversion mechanism is demonstrated based

on a series of interfacial chemistry measurements and fluorescence-assisted in situ

observation. Furthermore, organic-inorganic 3D solid foams were easily obtained based on

the emulsion template, which showed potential applications for oil adsorption at the water

surface. This work highlights a new phase-switchable emulsion that may raise both

fundamental research interest and industrial application values.

100

4.2 Article 3

Mechanically Switchable Emulsions

Chenhui Han,a Eric R. Waclawik,a Xiaofei Yang,*b Peng Meng,a Hengquan Yang,*c Ziqi

Suna and Jingsan Xu,*a

a. School of Chemistry, Physics and Mechanical Engineering, Queensland University of

Technology, Brisbane, QLD 4001, Australia.

b. College of Science, Institute of Materials Physics and Chemistry, Nanjing Forestry

University, Nanjing 210037, China.

c. School of Chemistry and Chemical Engineering, Shanxi University, Taiyuan 030006, China.

Abstract: Phase-switchable emulsions are of great significance due to their potential uses in

green catalysis, emulsion polymerization, cleaning, and degreasing, etc. Here, we report on

switchable emulsions in which the phase inversion between oil-in-water and water-in-oil type

can be mechanically triggered for many cycles. The emulsions are stabilized by a surface-

active, organic-inorganic hybrid by combining an oil-soluble molecule, stearic acid, with

water-dispersible alumina nanofibers via chemisorption. The long carbon-chain of stearic

acid functions as the hydrophobic tail of the stabilizer, while the inorganic nanofibers can act

as the hydrophilic head. A novel emulsion phase inversion mechanism is demonstrated based

on a series of interfacial chemistry measurements and fluorescence-assisted in situ

observation, i.e. the hybrid exhibits reversible switching between hydrophilic and lipophilic

states, by mechanically manipulating (aging-sonicating operation) the adsorption-desorption

volume of stearic acid attached to the Al2O3 nanofibers. As a result, the emulsions stabilized

by this organic-inorganic hybrid can reversibly transform between oil-in-water and water-in-

oil type. As a bonus, 3D organic-inorganic solid foams can be readily prepared based on the

101

emulsion system, which demonstrates potential applications for remediation of oil-spills in

the environment.

Introduction

Emulsions stabilized by molecular surfactants have been extensively studied due to their wide

application in the fields of pharmaceutical, cosmetic, oil industry, and material synthesis.[1] In

certain circumstances, emulsions with switchable phases are desired, such as

emulsion/microemulsion polymerization, cleaning and degreasing, and consecutive catalytic

processes.[2] Although phase-switchable emulsions have been developed by using responsive

surfactants, based on structural changes of the stabilizers upon external stimuli such as

acids/bases, oxidants/reductants, the use of chemical stimuli often result in product

contamination and environmental and economic costs.[3-5] Besides, yet some emulsions are

reported to be responsive to benign gases,[2, 6, 7] these emulsions can only transform between

emulsified and demulsified states, rather than between different emulsion types (i.e. oil-in-

water and water-in-oil). A similar situation occurs in emulsions that are responsive to

physical stimuli such as temperature, light, and magnetism.[8]

In the past years, colloidal particles stabilized emulsions (namely Pickering emulsions) have

been of interest owing to advantages of ease of recycling, excellent stability, and low

toxicity.[9-11] The stability and type of these emulsions are largely determined by the

wettability of the particles: a good emulsifier should possess intermediate wettability, where

more hydrophilic particles stabilize oil-in-water (o/w) emulsions and more

hydrophobic/lipophilic particles stabilize water-in-oil (w/o) emulsions.[12-14] Inorganic

nanoparticles in combination with surfactants can generate phase-inversion emulsions by

continuously increasing the concentration of surfactant or salt.[15, 16] However, since the

concentration changing is irreversible, the number of times phase transformation can be

102

achieved with these systems is limited to only two-cycles (o/w to w/o and back-to o/w type),

which limits the consecutive operation of emulsion-based catalytic system.

Herein, we developed a hybrid stabilized emulsion in which reversible phase inversion

between o/w and w/o type can be realized for multiple cycles, simply by aging-sonicating

operations. Briefly, stearic acid (SA) dissolved in oil and Al2O3 nanofibers dispersed in water

are combined to form a hybrid via chemisorption at the oil-water interfaces, with the long

carbon chain in SA performing the role as the hydrophobic tail and the Al2O3 nanofiber

acting as the hydrophilic head. The amphiphilicity of the hybrid, which is determined by the

content ratio of SA and Al2O3, undergoes reversible switching between more hydrophilic and

more hydrophobic via mechanically managing the adsorption-desorption of SA molecules on

the Al2O3 surface. As a result, emulsions stabilized by the hybrid can reversibly transform

between o/w and w/o type upon adsorption/desorption of SA.

Results and discussion

The Al2O3 nanofibers were synthesized based on a previously reported method with minor

modifications.[17] The transmission electron microscopy (TEM) images illustrate that the

fibers have a diameter of around 5 nm and a length in the range of 1-5 µm (Figure 1a,b). The

X-ray diffraction (XRD) pattern suggests that the Al2O3 sample was of the γ phase (Figure

1c). The Al2O3 nanofibers were dispersible in water and remained stable against

sedimentation owing to strong electrostatic repulsion (zeta potential +38 mV). X-ray

photoelectron spectroscopy (XPS) measurement demonstrates the binding energy of Al 2p at

74.2 eV and the peak of O 1s can be deconvoluted into one at 531.1 eV and the other at 532.4

eV, which could be assigned to the lattice oxygen in Al2O3 and the surface adsorbed

hydroxide group, respectively.[18] Fourier-transform infrared (FTIR) spectra indicate strong

chemical interaction between SA and Al2O3. As shown in Figure 1e, the Al2O3 sample does

not show any obvious absorption bands except the one at 3450 cm-1, corresponding to the -

103

OH stretching vibration in adsorbed water. For the SA-Al2O3 hybrid, bands at 2911 and 2844

cm-1 are assigned to the antisymmetric and symmetric stretching vibration of the C-H bond

from SA, respectively, indicating attachment of SA onto the Al2O3 surface. Interestingly, the

stretching vibration of C=O at 1694 cm-1 completely vanishes after adsorption to the Al2O3

surface, owing to the disintegration of -COOH group from SA.[19] Instead, a new peak at

1560 cm-1 appears, which is assigned to the C=O stretching vibration in the as-formed

carboxylate (Figure 1f). These observations indicate that chemisorption between SA

molecules and Al2O3 nanofibers occurs, generating an amphiphilic hybrid at the liquid

interface. The organic-inorganic chemical interaction is found to have a significant effect on

emulsification, which is discussed below.

Figure 1. (a,b) TEM images, (c) XRD pattern, blue bars showing the standard diffraction

pattern of γ-Al2O3; (d) XPS spectra of Al2O3 nanofibers. (e) FTIR spectra of Al2O3, SA and

SA-Al2O3 hybrid. (f) Schematic illustration of the SA-Al2O3 structure with SA in oil

adsorbed on Al2O3 in water.

104

The ability to lower the interfacial tension between water and hydrophobic solvents

determines the surface activity of a substance, and high surface activity is favorable to the

generation of emulsions. In the present case, interfacial tension measurement illustrates that

Al2O3 nanofibers did not possess surface activity (Figure S1). Also, SA alone showed weak

surface activity (Figure 2a), owing to its structural characteristics of a carbon chain

connected to a carboxylic acid group. By dispersing Al2O3 in water and dissolving SA in

hexane together, higher surface activity was achieved, especially when the concentration of

SA (denoted as n[SA]) was greater than 0.1 mM. For instance, the interfacial tension was

41.1 mN/m when solely SA was added in hexane (8.2 mM), while it decreased to 34.6 mN/m

in the presence of additional Al2O3 in water (0.33 wt%). These results demonstrate the

synergistic effect of SA coupled Al2O3 for enhanced surface activity (Figure 1f). This effect

also applies to a range of solvents including alkanes and arenes (Figure 2b), demonstrating

high versatility.

Accordingly, water-oil emulsions can be readily stabilized by the SA-Al2O3 hybrid, using

hexane as a typical oil phase. As a control, Al2O3 nanofibers alone were too hydrophilic to

stabilize emulsions even upon strong shearing force. In the absence of Al2O3, on the other

hand, no fine emulsion was generated until quite high n[SA] was used (0.35 mM), and the

emulsions exhibited low stability (Figure S2). However, by combining Al2O3 with a very

small amount of additional SA in oil (0.01 mM), an o/w emulsion was readily generated by

shaking the water-oil mixture (Figure 2c, left). Increasing n[SA] up to 1 mM improved the

emulsifying efficiency and the obtained emulsions remained stable against coalescence for at

least a year (Figure S3). These results demonstrate the high effectiveness of the hybrid

emulsifier, benefiting from the improved surface activity and the steric effect of the

nanofibers.[20] Noticeably, when n[SA] was increased high enough (35 and 70 mM), the

emulsions converted to w/o type (Figure 2c, right), as the hybrid emulsifier turning more

105

hydrophobic due to more adsorption of SA molecules onto the Al2O3. Similar results were

also observed for emulsions stabilized by some other surfactant-nanoparticle mixtures.[21-23]

Figure 2. (a) Water-hexane interfacial tension with different SA concentrations in hexane

without and with 0.33 wt% Al2O3 in water. (b) Interfacial tensions between water and

different organic solvents under different conditions: pure water and pure solvents, pure

water and n[SA] = 0.35 mM in solvents, and m[Al2O3] = 0.33 wt% in water and n[SA] = 0.35

mM in solvents. (c) The water-hexane emulsion phases with the increase of n[SA]; m[Al2O3]

= 0.33 wt% in water.

We observed that the obtained emulsions crossed a transitional state when n[SA] was

prepared with intermediate levels (Figure 2c, middle): they can reversibly transfer between

o/w and w/o phase by simple manipulation. As illustrated in Figure 3a, an o/w emulsion was

initially generated by shaking the water/hexane mixture when 0.33 wt% Al2O3 and 7 mM SA

were used. After 20 min standing, this emulsion unprecedentedly transformed to w/o type

106

upon further shaking, as shown in Figure 3b. Note that the water was dyed blue and hexane

was dyed red for better light microscopy picturing. Quick sonication broke-up the w/o

emulsion and followed by shaking, the emulsion was changed back to o/w type, illustrating

the reversibility of the phase inversion. This reversible, dynamically-controlled, emulsion

phase inversion should originate from the switchable amphiphilicity of the SA-Al2O3 hybrid.

Figure 3. Reversible water-hexane (1:1 volume ratio) emulsion phase inversion between (a)

o/w type and (b) w/o type by aging-shaking and sonicating-shaking processing. Water was

dyed with methylene blue and hexane was dyed with Oil Red O for light microscopy imaging.

(c) Water-hexane emulsion phase diagram with different n[SA] in hexane and m[Al2O3] in

water. (d) Water-solvent emulsion phase diagram with different n[SA]; m[Al2O3] = 0.33 wt%.

Screening experiments show that the mechanically switchable emulsion can be realized in

wide concentration windows of SA and Al2O3. As shown in Figure 3c, phase-switchable

emulsions were obtained with the concentration of Al2O3 (denoted as m[Al2O3]) in the range

107

of 0.1 to 1.3 wt%, as long as n[SA] was set an accordingly appropriate value. For example,

when m[Al2O3] was 0.2 wt%, n[SA] between 3.5 and 7.1 mM was suitable for achieving the

switchable emulsions. Fixed-type emulsions (either o/w or w/o type) will otherwise be

formed when n[SA] was below or above the critical concentrations, respectively. In another

set of experiments, different organic solvents were used to test the versatility of the

switchable emulsions stabilized by the SA-Al2O3 hybrid. As shown in Figure 3d, in addition

to hexane, alkanes including octane and decane can also be used as the oil phase for making

switchable emulsions. It is worth noting that less SA was needed for the alkanes with longer

carbon chains, probably because of the structural similarity between them. The range of

organic solvents can extend to arenes such as toluene, p-xylene, and ethylbenzene. Higher

concentrations of SA were needed in these cases, which should be owing to the structural

difference between the aromatic rings and the carbon chains.

Figure 4. (a) Water contact angle of the SA-Al2O3 hybrid with different concentrations of SA;

m[Al2O3] = 0.33 wt%. (b) Dynamic interfacial tension measurement with n[SA] = 0.35 mM

in hexane and with m[Al2O3] = 0, 0.1 and 0.33 wt% in water.

The wettability of a solid emulsifier plays a critical role in determining the emulsion phases.

To monitor the wettability of the SA-Al2O3 hybrids, the water contact angle was measured

over varied contents of SA and Al2O3. As shown in Figure 4a, the pristine Al2O3 exhibits a

108

contact angle of 21.8o, well below 90o, confirming its high hydrophilicity as mentioned above.

After hybridizing with SA (4.2 mM), the contact angle increased to 73o, reflecting the hybrid

turning less hydrophilic. The hybrid became fully hydrophobic when n[SA] was above 14

mM, with the contact angle reaching 117o and higher. It is worth mentioning that when n[SA]

= 7 mM, the hybrid resulted in switchable emulsions as aforementioned, and the contact

angle reached accordingly a critical value (very close to 90o). Under this condition, the as-

stabilized emulsion type (o/w or w/o) became very sensitive and could vary upon vibration of

the solution environment.

To better understand the switchable emulsion phase inversion, the adsorption behavior of SA

and Al2O3 were carefully investigated. As aforementioned, an o/w emulsion was formed at

first when the ratio of n[SA]/m[Al2O3] lied in the transitional range (Figure 3c,d). To convert

this o/w emulsion to w/o type, tens-of-minutes aging (Step 1, Figure 6) was required,

followed by shaking. By contrast, we noticed that longer aging time was needed when

m[Al2O3] was reduced from 0.33 to 0.1 wt%. In this context, the dynamic water-hexane

interfacial tension with SA in hexane (7 mM) and with or without Al2O3 in water was

monitored. It can be seen the interfacial tension quickly reached equilibrium (42.6 mN/m)

within 100 s when only SA was used (Figure 4b). However, the interfacial tension decreased

much more slowly when additional Al2O3 (0.33 wt%) was added in water. We can see that

the interfacial tension started to get stabilized (36.3 mN/m) after 10 min and it even took a

longer time (20 min) when m[Al2O3] was down to 0.1 wt%. This phenomenon should be

assigned to the relatively slow kinetics of SA molecules adsorbing to the Al2O3 nanofibers

and thereafter reaching adsorption equilibrium at the liquid interface;[24] this process became

slower when the concentration of Al2O3 was lower.

109

Figure 5. (a) Confocal images of a single droplet stabilized by fluorescence-labelled-SA and

pristine Al2O3 at different aging times and (b) corresponding quantitative fluorescence

intensity. (c) 3D images of a single droplet stabilized by fluorescent Al2O3 nanofibers and

pristine SA at different aging times and (d) quantitative fluorescence intensity at 0 min and

60 min.

In addition, the adsorption process between SA and Al2O3 was in situ monitored by

fluorescence-labeling experiments. Coumarin 343 X carboxylic acid (denoted as CCA,

Figure S4), a blue-fluorescent molecule that resembles the structure of SA, is used to trace

the location and movement of SA molecules and Al2O3 nanofibers. Firstly, we detected the

diffusing of CCA onto the oil-water interface. SA was dissolved in toluene and labeled by

CCA (2×10-3 mM in SA solution), and pristine Al2O3 nanofibers were dispersed in water.

110

Immediately after emulsification, one emulsion droplet was taken out and pictured using a

confocal microscope after different time intervals. As shown in Figure 5a and b, the surface

fluorescence intensity of the droplet gradually increased with aging time, indicating that more

and more molecules moved to the droplet surfaces and adsorbed onto the Al2O3 nanofibers.

Secondly, the CCA-labelled-Al2O3 nanofibers (Figure S5) were used to trace the movement

of Al2O3 nanofibers toward the interface through observing an o/w emulsion stabilized by the

fluorescent Al2O3 and pristine SA. As shown in Figure 5c, the fluorescence intensity at the

emulsion droplet surface increased with aging time, indicating an increasing amount of Al2O3.

The fluorescence intensity was also quantitatively measured and shown in Figure 5d,

revealing that the Al2O3 nanofibers tended to accumulate at the oil-water interface and form

surface-active hybrids upon adsorbing SA molecules.

Figure 6. Proposed mechanism of the switchable emulsion phase inversion between o/w and

w/o type induced by tuneable amphiphilicity of the hybrid surfactant.

Based on the above measurements and analysis, we propose a mechanism for better

understanding the reversible emulsion phase transition resulted from the switchable

amphiphilicity of the SA-Al2O3 hybrid. In the first stage, an o/w emulsion is stabilized by the

SA-Al2O3 hybrid (Figure 6). Upon aging, the remaining SA molecules in the oil and Al2O3

111

nanofibers shift to the surface of emulsion droplets, until adsorption equilibrium is reached

(Step 1, Figure 6). Hereby, the as-formed SA-Al2O3 hybrid possesses a critical

amphiphilicity, with a contact angle very close to 90o. External disturbance, such as shaking,

can induce swing of the oil-water interface and hence more of the Al2O3 surface is covered by

the SA molecules, changing the wettability of the hybrid from more hydrophilic to more

hydrophobic. Therefore, shaking again results in phase inversion and a w/o emulsion is

obtained (Step 2 and 3, Figure 6). In the next step, this w/o emulsion is broken by sonication.

Meanwhile, the ultrasonic energy could also result in the desorption of SA from the Al2O3

surface,[25] which can be confirmed by sonicating the CCA-labelled-Al2O3 nanofibers. As

shown in Figure S6, the fluorescence intensity significantly decreased after sonication,

indicating the detachment of adsorbed carboxylic acid from the nanofibers. This detachment,

which could be only a minor portion, can change the hybrid back to be more hydrophilic.

Therefore, an o/w emulsion is formed again by subsequent shaking (Step 4, Figure 6). This

controllable emulsion phase transition, which results from the dynamically tunable

hydrophobic/hydrophilic ratio of the SA-Al2O3 hybrid relies on the reversible adsorption-

desorption process.

Due to the strong interactions between the “hard” Al2O3 nanofibers and the “soft” SA

molecules at the liquid interfaces, the as-stabilized emulsions possess certain mechanical

strength. As shown in Figure S7, the drying process of the emulsion was monitored in real-

time under the microscope, exhibiting flexibility and robustness of the droplet surfaces,

induced by the SA-Al2O3 hybrid. Hence, the emulsions can serve as a skeleton for the

preparation of three-dimensional bulk foams. The hybrid foams were simply prepared by

centrifuging the w/o emulsions and drying the collections in a vacuum oven. The bulk foams

show a 3D network with the Al2O3 nanofibers closely connected together by SA molecules

(Figure S8), originating from the 3D structure of the emulsions. Compared with the pristine

112

Al2O3, the foams were highly hydrophobic with a water contact angle up to 134o (Figure 7a,

b). It had a density lower than water and showed no adsorption capacity of water (Figure S9).

Benefiting from this, the hybrid foam was able to easily adsorb the oil film floating on the

water surface. Figure 7c illustrates the fast absorption of decane, with 73 mg of decane

completed absorbed by a piece of the SA-Al2O3 foam in 6 s. Such superior adsorption

performance of the hybrid foam occurs with a range of organic liquids, including alkanes, p-

xylene, silicone oil and peanut oil (Figure 7d), reaching up to 10-15 times its own weight.

The low cost, easy processing and environmental-friendly endow the SA-Al2O3 foam high

potential for practical oil-water separation.

Figure 7. Water contact angle of (a) pure Al2O3 and (b) the hybrid foam. (c) Images of the

foam sucking up decane (dyed with Oil Red A) on water surface. (d) Adsorption capacity of

the hybrid foam for different organic liquids.

113

Conclusions

In summary, we developed a mechanically switchable emulsion stabilized by organic-

inorganic hybrids, in which multiple-times phase inversion can be achieved by simple aging-

sonicating operations, without using any chemical additives. The long carbon chains in SA

function as the hydrophobic tail while the Al2O3 nanofibers act as the hydrophilic head. The

amphiphilicity of the hybrid, which is determined by the content ratio of SA and Al2O3,

exhibits reversible switching between more hydrophilic and more hydrophobic by

dynamically managing the adsorption amount of SA on Al2O3. Consequently, emulsions

stabilized by the hybrid can reversibly transform between o/w and w/o type upon adsorption-

desorption manipulation. As an added benefit, 3D solid foams can be facilely derived from

the emulsions, which show high adsorption capacity and thus potential applications in the

remediation of oil-spills in the environment, owing to the low cost, easy processing, and

environmental-friendly.

Experimental section

The experimental section is implemented in the Supporting Information.

Acknowledgments

The authors are grateful to Australian Research Council and the Start-Up Fund from Nanjing

Forestry University.

References

[1] R. Aveyard, B.P. Binks, J.H. Clint, Emulsions stabilized solely by colloidal particles, Adv.

Colloid Interface Sci., 100 (2003) 503-546.

[2] Y. Liu, P.G. Jessop, M. Cunningham, C.A. Eckert, C.L. Liotta, Switchable surfactants,

Science, 313 (2006) 958.

114

[3] C. Liang, Q. Liu, Z. Xu, Surfactant-free switchable emulsions using CO2-responsive

particles, ACS Appl. Mater. Interfaces, 6 (2014) 6898-6904.

[4] J. Li, H.D.H. Stöver, Doubly pH-responsive Pickering emulsion, Langmuir, 24 (2008)

13237-13240.

[5] Q. Lan, C. Liu, F. Yang, S. Liu, J. Xu, D. Sun, Synthesis of bilayer oleic acid-coated

Fe3O4 nanoparticles and their application in pH-responsive Pickering emulsions, J. Colloid

Interface Sci., 310 (2007) 260-269.

[6] C.I. Fowler, C.M. Muchemu, R.E. Miller, L. Phan, C. O’Neill, P.G. Jessop, M.F.

Cunningham, Emulsion polymerization of styrene and methyl methacrylate using cationic

switchable surfactants, Macromolecules, 44 (2011) 2501-2509.

[7] C. Liang, J.R. Harjani, T. Robert, E. Rogel, D. Kuehne, C. Ovalles, V. Sampath, P.G.

Jessop, Use of CO2-triggered switchable surfactants for the stabilization of oil-in-water

emulsions, Energy & Fuels, 26 (2012) 488-494.

[8] P. Brown, C.P. Butts, J. Eastoe, Stimuli-responsive surfactants, Soft Matter, 9 (2013)

2365-2374.

[9] Q. Cao, Q. Cui, Y. Yang, J. Xu, C. Han, L. Li, Graphitic carbon nitride as a distinct solid

stabilizer for emulsion polymerization, Chem.: Eur. J., 24 (2018) 2286-2291.

[10] J. Xu, M. Antonietti, The performance of nanoparticulate graphitic carbon nitride as an

amphiphile, J. Am. Chem. Soc., 139 (2017) 6026-6029.

[11] J. Xu, X. Li, X. Wu, W. Wang, R. Fan, X. Liu, H. Xu, Hierarchical CuO colloidosomes

and their structure enhanced photothermal catalytic activity, J. Phys. Chem. C, 120 (2016)

12666-12672.

[12] B.P. Binks, S.O. Lumsdon, Influence of particle wettability on the type and stability of

surfactant-free emulsions, Langmuir, 16 (2000) 8622-8631.

115

[13] J. Kim, L.J. Cote, F. Kim, W. Yuan, K.R. Shull, J. Huang, Graphene oxide sheets at

interfaces, J. Am. Chem. Soc., 132 (2010) 8180-8186.

[14] J. Xu, A. Ma, T. Liu, C. Lu, D. Wang, H. Xu, Janus-like Pickering emulsions and their

controllable coalescence, Chem. Commun., 49 (2013) 10871-10873.

[15] B.P. Binks, J.A. Rodrigues, Double inversion of emulsions by using nanoparticles and a

di-chain surfactant, Angew. Chem. Int. Ed. Engl., 46 (2007) 5389-5392.

[16] J. Zhang, L. Li, J. Wang, H. Sun, J. Xu, D. Sun, Double inversion of emulsions induced

by salt concentration, Langmuir, 28 (2012) 6769-6775.

[17] J. Yang, R.L. Frost, Synthesis and characterization of boehmite nanofibers, Int. J. Inorg.

Chem., 2008 (2008) 1-4.

[18] R.-C. Fang, Q.-Q. Sun, P. Zhou, W. Yang, P.-F. Wang, D.W. Zhang, High-performance

bilayer flexible resistive random access memory based on low-temperature thermal atomic

layer deposition, Nanoscale Res. Lett., 8 (2013) 92.

[19] L.J. Bellamy, The infrared spectra of complex molecules, Wiley, New York, 1975.

[20] E. Vignati, R. Piazza, T.P. Lockhart, Pickering emulsions:  interfacial tension, colloidal

layer morphology, and trapped-particle motion, Langmuir, 19 (2003) 6650-6656.

[21] Z.G. Cui, K.Z. Shi, Y.Z. Cui, B.P. Binks, Double phase inversion of emulsions

stabilized by a mixture of CaCO3 nanoparticles and sodium dodecyl sulfate, Colloids Surf. A,

329 (2008) 67-74.

[22] D.J. Miller, T. Henning, W. Grünbein, Phase inversion of W/O emulsions by adding

hydrophilic surfactant - a technique for making cosmetics products, Colloids Surf. A, 183-

185 (2001) 681-688.

[23] J. Wang, F. Yang, C. Li, S. Liu, D. Sun, Double phase inversion of emulsions containing

layered double hydroxide particles induced by adsorption of sodium dodecyl sulfate,

Langmuir, 24 (2008) 10054-10061.

116

[24] Z. Hu, T. Patten, R. Pelton, E.D. Cranston, Synergistic stabilization of emulsions and

emulsion gels with water-soluble polymers and cellulose nanocrystals, ACS Sustainable

Chem. Eng., 3 (2015) 1023-1031.

[25] M. Breitbach, D. Bathen, H. Schmidt-Traub, Effect of ultrasound on adsorption and

desorption processes, Ind. Eng. Chem.Res., 42 (2003) 5635-5646.

117

Supporting Information

Mechanically Switchable Emulsions

Chenhui Han,a Eric R. Waclawik,a Xiaofei Yang,*b Peng Meng,a Hengquan Yang,*c Ziqi

Suna and Jingsan Xu,*a

a. School of Chemistry, Physics and Mechanical Engineering, Queensland University of

Technology, Brisbane, QLD 4001, Australia.

b. College of Science, Institute of Materials Physics and Chemistry, Nanjing Forestry

University, Nanjing 210037, China.

c. School of Chemistry and Chemical Engineering, Shanxi University, Taiyuan 030006, China.

118

Materials

Aluminum nitrate nonahydrate (≥ 98%, Sigma-Aldrich), ammonia solution (30% in water,

Chem-Supply), stearic acid (95%, Sigma-Aldrich), Tergitol (Type 15-S-9, Sigma-Aldrich),

hexane fraction (≥97%, Ajax Finechem), cyclohexane (99.5%, Sigma-Aldrich), octane (≥

99%, Sigma-Aldrich), decane (≥98%, Sigma-Aldrich), toluene (AR, Fisher Scientific),

ethylbenzene (99.8%, Sigma-Aldrich), p-Xylene (99%, Sigma-Aldrich), methylene blue

(Sigma-Aldrich), Oil Red O (Sigma-Aldrich). All chemicals were used as received without

further purification. Ultrapure water was used in all experiments (18.2 MΩ·cm, Synergy UV

Water Purification System, Merck Millipore).

Synthesis of Al2O3 nanofibers

Briefly, 15 g of Al(NO3)3·9H2O was dissolved in 25 mL of ultrapure water (the pH of the

solution was around 2.3). Under vigorously stirring, ammonia aqueous solution (mass

fraction 10%) was added dropwise to the above solution until the pH reached 5, and kept

stirring for one more hour. Afterward, the produced precipitate was separated by

centrifugation, transferred into a stainless steel autoclave and heated at 170 oC for 48 hours.

After cooling down, the resulting solid was washed with ultrapure water twice. Finally, the

mixture was heated at 450 oC for 5 hours to obtain the final products.

Preparation of emulsions

Al2O3 aqueous suspensions with various concentrations (expressed as wt% in water phase)

were prepared by sonicating Al2O3 nanofiber in ultrapure water for 5 minutes. Stearic acid

(SA) was dissolved in hexane to obtain solutions with different concentrations of SA

(expressed as mM in the oil phase). Next, the Al2O3 aqueous suspension and the SA solution

(1.5 mL/1.5 mL) were mixed in a vial. The vial was then sealed and shaken by hand for 30

119

sec to produce emulsions. Other emulsions with different oil phases were prepared by a

similar method. In some cases, for better visualization, methylene blue and Oil Red O were

used to color the water phase and oil phase, respectively.

Fluorescence-labelling experiment

Fluorescence-labeled stearic acid solution:

The stearic acid solution was fluorescence-labeled by adding a small amount of Coumarin

343 X carboxylic acid (denoted as CCA). Toluene was used as a solvent and the

concentration of CCA in the solution was 2×10-3 mM.

Fluorescence-labeled Al2O3 nanofibers:

Pristine Al2O3 nanofibers were dispersed in CCA solution (2×10-2 mM in ethanol) and stirred

for 5 hours to reach adsorption equilibrium. After centrifuging, washing with ethanol and

drying, the CCA-labelled-Al2O3 nanofibers were collected.

Emulsions observation

The light microscope (LM) images were obtained through observing a water-diluted

emulsion placed in the middle of two pieces of glass slides separated by a distance of 1mm.

The emulsion type was identified by observing an emulsion in which the water phase and oil

phase were colored by methylene blue and Oil Red O, respectively.

Fluorescent-assisted in situ observation of the adsorption of carboxylic acid and evolution of

Al2O3 nanofibers at interface was conducted on a confocal laser scanning microscopy. The

sample preparation process is the same as that for light microscopy picturing.

120

Contact angle and interfacial tension measurement

The sessile drop method was used to measure the contact angle of the SA-Al2O3 hybrid. For

the measurement, 1 g of SA-Al2O3 hybrid was obtained by centrifuging the corresponding

emulsions and drying the sediment overnight. Then the SA-Al2O3 hybrid was filled in a steel

die and compressed under a pressure of 6 × 108 N/m2 to obtain a disk (diameter, 13mm;

thickness, ca. 1mm). The contact angle measurement was performed on this disk using

FTA200 Contact Angle Analyser. Briefly, a drop of water was first dripped on the disk, and

then the droplet shape was captured by the camera and analyzed through software to give

contact angle.

The pendant drop method was used for interfacial tension measurement. Prior to the

measurement, an oil phase (3 mL, with or without SA) was first filled in a quartz cuvette, and

a pendant drop of water (with or without Al2O3) was formed on the end of a stainless steel

needle immersed in the oil phase. Then the images of the drop were taken automatically at a

certain frequency and analyzed through software to give a series of interfacial tension values

during the aging process.

Adsorption capacity measurement of the SA-Al2O3 hybrid foam

A 50 mL beaker with a small amount of water was placed on an analytical balance; then a

piece of SA-Al2O3 hybrid foam was placed on the water surface and its weight was recorded.

Oil phase (colored by Oil Red O for easily observing) was dropped onto the water surface

drop by drop and adsorbed by the hybrid foam until saturated. The weight of the adsorbed oil

was used to evaluate the adsorption capacity of the hybrid foam according to the following

equation.

Weight gain (%) = 𝑡ℎ𝑒 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑎𝑏𝑠𝑜𝑟𝑏𝑒𝑑 𝑜𝑖𝑙

𝑡ℎ𝑒 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 ℎ𝑦𝑏𝑟𝑖𝑑 𝑓𝑜𝑎𝑚 × 100%

121

Characterization

The morphology of prepared Al2O3 nanofiber was studied using a JEOL 2100 transmission

electron microscopy (TEM) equipped with a Gatan Orius SC1000 CCD camera and the

accelerating voltage of TEM was 200 kV. X-ray photoelectron spectroscopy (XPS) analysis

was performed with a Kratos Axis Ultra photoelectron spectrometer using mono Al Kα

(1486.6 eV) X-ray. The X-ray diffraction (XRD) patterns of the Al2O3 nanofiber were

obtained on a Philips PANalytical X’Pert PRO diffractometer using Cu Kα radiation (λ =

1.5418 Å) working at 40 mA and 40 kV. The diffraction data were collected from 10° to 90°

with a resolution of 0.01° (2θ). The SA-Al2O3 hybrid foam was observed using a ZEISS

Sigma scanning electron microscopy (SEM) at accelerating voltages of 5 kV. The infrared

spectrum (IR) was collected between the wavenumber of 400 to 4000 on a BRUKER Alpha

infrared spectrometer. The interfacial tension and contact angle measurement were performed

on an FTA200 Contact Angle Analyser. Nikon Eclipse Ni light microscope and Nikon A1R

confocal laser scanning microscopy were used to image the emulsions.

123

Figure S4. (a) The structure and (b) absorption and emission spectra of the used fluorescence

molecule.

Figure S5. Digital photos of (a) pristine Al2O3 fibers and (b) CCA-labelled-Al2O3 nanofibers

excited under 365 nm UV light.

124

Figure S6. (a) The confocal image of two kinds of Al2O3 fibers: (i) the original CCA-

labelled-Al2O3 nanofibers; and (ii) after sonicating for 2 min in the oil phase. (b), (c)

corresponding quantitative fluorescence intensity under different angles of view.

Figure S7. LM images of the drying process of the emulsion monitored in real-time.

125

Figure S8. SEM (a) and TEM (b) images of the hybrid foam.

Figure S9. A series of digital photos showing the hydrophobic property of the hybrid foam.

When pressing the hybrid foam into water, it rebounded immediately.

126

Chapter 5: Conclusions & Future work

Conclusions

This work has contributed to the knowledge of two-dimensional materials-stabilized

Pickering emulsions and its application in catalysis. The behavior of the stabilizer at the

interface was well illustrated by in situ observation. Based on the g-C3N4 stabilized emulsion

system, a tandem reaction was conducted and high hydrogen utilization efficiency was

reached. Controllable multiple phase inversion was also realized in a SA-Al2O3 hybrid-

stabilized Pickering emulsion and the phase inversion mechanism was carefully investigated

and illustrated.

In Chapter 2, the stabilizing mechanism of g-C3N4-stabilized Pickering emulsions was

investigated by in situ microscopy observations. Benefitting from the large sizes and

photoluminescence of g-C3N4 particles, their positions and movements could be directly

observed using a light/fluorescence microscope in an ambient environment. Under static

conditions, two stabilizing configurations of the g-C3N4 particles were revealed: (i) g-C3N4

particles located at interfaces of adjoining oil droplets and (ii) g-C3N4 particles fully covering

oil droplets. Further, we carried out real-time monitoring of the g-C3N4 particles and the

associated emulsion droplets, recording their movements and adsorption kinetics toward each

other. These results directly demonstrated the dynamic stabilization of Pickering emulsions

and may explain their superior stability compared to normal emulsions. Moreover, the oil

droplets were stable enough to be physically manipulated, permitting a study of their

response to external forces. Emulsion droplets readily deformed upon stress but did not break

apart. Finally, the possible effect of the interaction between stabilizer and organic phase on

the emulsifying behavior was examined. By comparing emulsions formed in solvents that

could form π−π interactions with the g-C3N4 stabilizer to those that cannot, this work

127

provides significant insights into the interfacial (liquid−liquid) properties of g-C3N4 and the

stabilizing mechanism of Pickering emulsions.

In Chapter 3, Pd/g-C3N4 stabilized Pickering emulsion was used as a microreactor to couple

hydrogen generation from ammonia borane (AB) with hydrogenation of alkenes at the oil-

water interfaces of the droplets. The hydrogen utilization efficiency can reach unity in such

an emulsion system, mainly due to the localized production, storage, and consumption of

hydrogen on the Pd NPs. Detailed analysis and observations revealed that the Pickering

emulsion can effectively store hydrogen via adsorption in the Pd nanoparticles and

microbubble formation within the liquid phases. The stored hydrogen then reacted with

alkene molecules until complete consumption. This hydrogen storage capability played a key

role in the efficient use of hydrogen. The Pd/g-C3N4 catalyst had excellent reusability and

turned out to be rather generally applicable to various alkene molecules. Importantly, the

conversion rate of alkene can reach >95% by simply modifying the Pickering emulsion

microreactors. This work opens a new avenue for developing more economical and efficient

hydrogenation processes for the production of a wide range of valuable chemicals.

In Chapter 4, a novel phase-switchable emulsion was developed in which the phase inversion

between oil-in-water and water-in-oil type can be mechanically triggered for many cycles.

The emulsions are stabilized by an organic-inorganic hybrid formed at the interface through

combining stearic acid (SA) molecules in oil and Al2O3 nanofibers in water via

chemisorption. The amphiphilicity of the hybrid, which is determined by the content ratio of

SA and Al2O3, exhibits reversible switching between more hydrophilic and more

hydrophobic by dynamically manipulating the adsorption amount of SA on Al2O3.

Consequently, emulsions stabilized by such a hybrid can reversibly transform between o/w

and w/o type upon adsorption-desorption manipulation, without the use of any external

stimulus. Detailed phase inversion mechanism is demonstrated based on a series of interfacial

128

chemistry measurements and fluorescence-assisted in situ observation. Furthermore, organic-

inorganic 3D foams can be facilely derived from the emulsion systems, which demonstrate

high potential for applications in the remediation of oil-spills in the environment.

In summary, by using g-C3N4 as a model stabilizer, the movements of stabilizers at interfaces

are well illustrated through in situ observation. The g-C3N4-stabilized emulsion is proven to

be effective microreactors for tandem reactions. Besides, a mechanically phase-switchable

emulsion is developed through dynamically altering the wettability of the stabilizer. The

work presented in this thesis enriched the knowledge-base of Pickering emulsions and

promoted the understanding of the emulsifying and stabilizing process of Pickering

emulsions stabilized by other stabilizers. On the other side, this work also explored the

interfacial properties of g-C3N4 that are chronically ignored by researchers. Therefore, the

application of g-C3N4 could be further extended. Especially in catalysis, the g-C3N4-

stabilized-emulsion-based reaction system provides an efficient way of organic synthesis.

The phase-switchable-emulsion-based reaction systems may further elevate the efficiency

owing to the advantages of easy product separation and consecutive operation in such

systems. The results of this thesis contribute to the practical application of Pickering-

emulsion-based reaction systems.

Future Work

Pickering emulsions were discovered more than one hundred years ago and have been

extensively studied since. Many research papers have focused mainly on the intrinsic

properties of Pickering emulsions. Less attention was paid to developing new applications of

Pickering emulsions, although it has shown application potential in many areas such as the

food, paint, pharmaceutical, material, and oil industries. Furthermore, some functional

129

Pickering emulsions need to be developed to realize the controllable release of drugs and the

synthesis of new materials. Therefore, exploiting new applications and developing functional

Pickering emulsions are two promising research directions.

Based on the outcomes of this thesis and relevant published works, future work is proposed

as follows:

1. Pristine g-C3N4 can only stabilize oil-in-water emulsions due to its intrinsic

hydrophilicity, which limits the application of g-C3N4-based Pickering emulsions.

Therefore, the synthesis of modified g-C3N4 is desired. For example, adding more

hydrophobic groups onto its framework in order to alter its amphiphilicity. The

application of g-C3N4-based Pickering emulsions would be further widened if the

formed emulsion types were controllable or adjustable. I have tried to graft alkyl

chains onto g-C3N4 sheets in order to obtain different types of emulsions. This project

is still developing and is expected to be completed in the future.

2. Pickering emulsion reaction systems are advantageous to traditional batch reactors for

some interfacial reactions and tandem reactions due to the large biphasic interface

created in such a system. Among various solid stabilizers, g-C3N4 is one of the few

candidates which is a semiconductor and can be excited by visible light. This property

provides an opportunity to conduct a photocatalytic oxidation reaction and a

photocatalytic reduction reaction at the same time in the emulsion system, with the

photo-induced electrons consumed by the reduction reaction happening in oil or water

phase and the photogenerated holes utilized by the oxidation reaction in the other

phase. Such a design not only increases the quantum efficiency but also enhanced the

reaction rate due to the effective separation of photo-induced electron-hole pairs. We

have made some progress in this project and will conduct an in-depth study in the

future.

130

3. Based on the outcomes in chapter 4, emulsion types can be controllably switched

using SA-Al2O3 hybrids as a stabilizer. This property is highly demanded in the

emulsion-based catalytic systems due to the advantage of product separation and

consecutive operation. Therefore, developing more stable and controllable emulsion

systems that can work well under harsh catalytic conditions is meaningful for both

fundamental research and industrial application. Related work may be conducted in

the future.