1
Pickering Emulsions Stabilized by Cellulose
Nanocrystals Grafted with Thermo-responsive Polymer
Brushes
Justin O. Zoppe,
1
Richard A. Venditti,
1
and Orlando J. Rojas
1,2,
*
1
Department of Forest Biomaterials, North Carolina State University, Campus Box 8005, Raleigh, NC
27697-8005, USA
2
Faculty of Chemistry and Materials Sciences, Department of Forest Products Technology, Aalto
University, P.O. Box 16300, FI-00076, Aalto, Finland
*Corresponding author: E-mail: ojrojas@ncsu.edu, Phone: +1-919-513 7494, Fax: +1-919-515 6302
ABSTRACT
Cellulose nanocrystals (CNCs) from ramie fibers are studied as stabilizers of oil-in-water emulsions.
The phase behavior of heptane and water systems are studied and emulsions stabilized by CNCs are
analyzed by using drop sizing (light scattering) and optical, scanning and freeze-fracture electron
microscopies. Water-continuous Pickering emulsions are produced with cellulose nanocrystals (0.05 to
0.5 wt %) grafted with thermo-responsive poly(NIPAM) brushes (poly(NIPAM)-g-CNCs). They are
observed to be stable during the time of observation of four months. In contrast, unmodified CNCs are
unable to stabilize heptane-in-water emulsions. After emulsification poly(NIPAM)-g-CNCs are
observed to form aligned, layered structures at the oil-water interface. The emulsions stabilized by
2
poly(NIPAM)-g-CNCs break after heating at a temperature above the LCST of poly(NIPAM), which is
taken as indication of the temperature-responsiveness of the brushes installed in the particles and thus
the responsiveness of the Pickering emulsions. This phenomenon is further elucidated via rheological
measurements, in which viscosities of the Pickering emulsions increase on approach of the low critical
solution temperature of poly(NIPAM). The effect of temperature can be counterbalanced with the
addition of salt which is explained by the reduction of electrostatic and steric interactions of
poly(NIPAM)-g-CNCs at the oil-water interface.
KEYWORDS. Pickering emulsions, oil-in-water emulsions, Cellulose nanoparticles; cellulose
nanocrystals, thermo-responsive emulsions, poly(N-isopropylacrylamide), LCST, grafts, Surface-
Initiated Single-Electron Transfer Living Radical Polymerization.
INTRODUCTION
The irreversible adsorption of solid particles at the oil-water interface has been known for over a
century [1]. Such phenomenon is critical in the stabilization of surfactant-free emulsions, in the so-
called Pickering emulsions [2], which have recently found use in applications spanning cosmetics,
biomedical, and food products [3]. The high stability of emulsions stabilized by colloidal particles is
derived from the energy barrier required to remove the particles from the interface in order to facilitate
droplet coalesce. Pickering emulsions have a number of advantages over conventional surfactant-
stabilized emulsions in that they can reduce tissue irritation and their viscosity can be easily adjusted by
solid content and/or solid type [4]. Nanoparticles modified with surface-active polymers have been
shown to be very efficient emulsifiers with nanoparticle concentrations as low as 0.04 wt % [5]. In
addition, the colloidal assembly of solid particles within Pickering emulsions can be used as templates
for advanced materials such as Janus colloids [6], composite microcapsules [7-9] or microspheres [10].
In particular, recent efforts focused on the development of environmentally friendly systems have led to
bio-based materials such as renewable nanocomposite foams [11] and bio-inorganic microcapsules [12]
from Pickering emulsion templates.
3
The development of renewable biomaterials for advanced applications has been recently gained
momentum in the research community [13, 14]. Cellulose, in particular, is an attractive material source
due to its availability, biodegradability, renewability, and amenability for surface modification.
Cellulose fibers, containing crystalline and amorphous regions, can be subjected to strong acid
hydrolysis yielding rod-like crystals having even more unique surface, optical and mechanical
properties [15, 16]. Cellulose nanocrystals (CNCs) or nanocrystalline cellulose (NCC) produced by
sulfuric acid hydrolysis yields aqueous suspensions electrostatically-stabilized by sulfate ester groups
installed on the surface which promotes uniform aqueous dispersions [17], and the ability to self-
assemble into chiral nematic liquid crystals [18]. The utility of nanocellulosic nanoparticles as stabilizer
of Pickering emulsion is expected to be advantageous over some inorganic nanoparticles when
biocompatibility, degradability, density and cost issues are considered.
Due the hydrophilicity of cellulose surfaces, surface chemical modification is needed in order to
impart an appropriate balance of surface energies or amphiphilicity, especially when stabilizing water-
in-oil (w/o) emulsions [19]. In the past, microcrystalline cellulose (MCC) has been shown to stabilize
oil-in-water (o/w) emulsions through network formation around oil droplets without surface
modification [20, 21]. The long-term stabilization of Pickering emulsions and foams with
microparticles from hydrophobic cellulose has been shown by Wege, et al. [22] Ougiya et al. first
studied the use of unmodified bacterial cellulose to stabilize o/w emulsions in 1997 [23]. In addition, a
number of patents have been filed involving the use of fibrillated cellulose as emulsion stabilizers [24,
25]. Furthermore, recent efforts have focused on the preparation of w/o emulsions by using silylated
micro- or nanofibrils from wood pulp and bacterial cellulose [11, 26-29]. The utility of bacterial
cellulose nanocrystals as a Pickering emulsion stabilizer has been recently demonstrated by
Kalashnikova et al. [30]
As previously noted, Pickering emulsifiers irreversibly adsorb at the oil-water interface and require a
much higher energy for desorption as compared to conventional surfactants. The use of responsive
polymer grafts that react to changes in their environment, such as light, heat, ionic strength, and pH [31-
33] can be used as a means to control Pickering emulsion stability [34]. Thermo-responsive polymers
4
may have an upper or a lower critical solution temperature (UCST and LCST), depending on entropic
changes and phase behaviors associated with their molecular structure. A number of researchers have
demonstrated thermo-responsive Pickering emulsions mostly by the use of (2-(dimethylamino)ethyl
methacrylate)-carrying silica and polystyrene latex nanoparticles [34, 35]. Poly(N-isopropylacrylamide)
(poly(NIPAM)), one of the most studied thermo-responsive polymers, has an LCST in aqueous solution
ranging between 30 to 35 °C [36]. Above the LCST poly(NIPAM) phase-separates due to thermal-
driven chain dehydration. Since the LCST of poly(NIPAM) is near the physiological temperature of ca.
37 °C, it has been extensively used in applications involving controlled drug release [37]. In addition,
poly(NIPAM) has been shown to be surface-active due to the presence of amide and isopropyl
functional groups [38-40]. The surface activity of particles containing poly(NIPAM) functionalities
have been confirmed at the air-water interface [41, 42]. Pickering emulsions stabilized by particles
carrying copolymers of NIPAM, methacrylic acid [43, 44], and acrylic acid [45] have been
demonstrated. Most recently, silica [46] and polystyrene [47] particles modified by poly(NIPAM) have
been developed as thermo-responsive Pickering emulsion stabilizers.
We have reported on the synthesis of poly(NIPAM) brushes grafted from CNCs via surface-initiated
single-electron transfer living radical polymerization (SI-SET-LRP) [48] and their surface interaction
forces [49]. The thermo-responsive behavior of poly(NIPAM)-g-CNCs and response to ionic strength
was demonstrated. The aim of this study was to demonstrate the use of cellulose nanocrystals grafted
with thermo-responsive functionalities as stabilizers of o/w Pickering emulsions. The effect of solids
content, temperature and ionic strength on emulsion stability and droplet size was investigated. The
self-assembly of poly(NIPAM)-g-CNCs at the oil-water interface was visualized via Freeze-Fracture
Electron Microscopy (FFEM) and Scanning Electron Microscopy (SEM).
EXPERIMENTAL
Materials. Pure ramie fibers were obtained from Stucken Melchers GmbH & Co. (Germany). 2-
bromoisobutyryl bromide (BriB), 2-dimethylaminopyridine (DMAP), tetrahydrofuran (THF, 99%), N-
isopropylacryalamide (NIPAM), copper (I) bromide and N,N,N′,N′′,N′′-pentamethyldiethylenetriamine
5
(PMDETA), were all obtained from Sigma-Aldrich. Sulfuric acid (95%), acetic acid (glacial),
triethylamine (TEA, 99.5%), acetone (99%), ethanol (95%), methanol (99%), heptane (99%), sodium
hydroxide pellets, and sodium chlorite were all purchased from Fisher Scientific. All solvents were
dried over molecular sieves (3 Å, 4-8 mesh beads, Sigma-Aldrich) for 48 h before use.
Cellulose nanocrystals and surface functionalization. Poly(NIPAM) chains were grafted from
cellulose nanocrystals (CNCs) by Surface-Initiated Single-Electron Transfer Living Radical
Polymerization (SI-SET-LRP). Details about this procedure and main properties of the obtained
poly(NIPAM)-g-CNCs can be found in our earlier reports [48, 49]. The surface charge of unmodified
ramie CNCs was determined to be 0.30 e/nm
2
by conductometric titration. The CNCs utilized in this
study were carrying poly(NIPAM) grafts polymerized from their surfaces with a molar ratio between
initiator [Br] and anhydroglucose units [AGU] in CNCs ([Br]:[AGU]) of 5:3 and a number average
molecular weight (M
n
) of 12,170 g/mol.
Water Contact Angle. The initial water contact angle (WCA) of spin-coated films of both
unmodified and poly(NIPAM) grafted CNCs (poly(NIPAM)-g-CNCs) [49] were determined using a
Phoenix 300 contact angle analyzer (SEO Co. Ltd, Lathes, South Korea). Contact angles were
measured at room temperature and immediately after being placed in an oven at 80 °C for 1 hour.
Preparation of Pickering Emulsions. All emulsions (total volume 10 mL) were prepared with a
water-to-oil ratio (WOR) of one containing given concentrations of poly(NIPAM)-g-CNCs in the
aqueous phase and emulsified with an Ultraturrax T18 basic homogenizer (IKA, Wilmington, NC USA)
at 6,000 rpm for 60 seconds at 25 °C. Volume fractions of the organic and aqueous phases were noted
30 minutes and 4 days after emulsification at 25 °C. The stability index of Pickering emulsions was
determined by the change in volume fraction of emulsified phase from 30 minutes after preparation to 4
days. The type of emulsion was determined by the drop test. A drop of emulsion was added to neat
water and neat heptane and their dispersion ability was observed. An oil-in-water emulsion disperses
readily in water, while a water-in-oil emulsion disperses readily in heptane (respective continuous
phases).
