Accepted Manuscript Title: Preparation and Stabilization of D-limonene Pickering Emulsions by Cellulose Nanocrystals Author: Chunxia Wen Qipeng Yuan Hao Liang Frank Vriesekoop PII: DOI: Reference:
S0144-8617(14)00627-4 http://dx.doi.org/doi:10.1016/j.carbpol.2014.06.051 CARP 9016
To appear in: Received date: Revised date: Accepted date:
4-7-2013 15-9-2013 19-6-2014
Please cite this article as: Wen, C., Yuan, Q., Liang, H., and Vriesekoop, F.,Preparation and Stabilization of D-limonene Pickering Emulsions by Cellulose Nanocrystals, Carbohydrate Polymers (2014), http://dx.doi.org/10.1016/j.carbpol.2014.06.051 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Preparation and Stabilization of D-limonene Pickering Emulsions by Cellulose Nanocrystals
1 2 3
Chunxia Wena, Qipeng Yuana, Hao Lianga,*, Frank Vriesekoopb
4
a
6 7
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical
b
cr
Technology, Beijing 100029, People’s Republic of China
Department of Food Science, Harper Adams University, Newport, Shropshire, TF10 8NB, England
us
8
ip t
5
9
an
10 11
M
12 13
ed
14 15
pt
16
Ac ce
17 18 19 20 21 22
*
Corresponding author. Tel.: +86-10-6443-1557; Fax: +86-10-6443-7610.
E-mail address: [email protected].
1
Page 1 of 25
23 24
Abstract The aim of this study was to investigate D-limonene Pickering emulsion stabilized by cellulose
26
nanocrystals (CNCs) and factors that may affect its properties. CNCs were prepared by ammonium
27
persulfate hydrolysis of corncob cellulose, and D-limonene Pickering emulsions were generated by
28
ultrasonic homogenizing method. The morphology and size of the prepared emulsions with different
29
CNCs concentrations were studied by optical microscopy and laser light diffraction. In addition,
30
factors that may affect the stability of emulsions such as ionic concentration, pH and temperature
31
were also studied. As indicated by the experiment data, when temperature rose, the stability to of
32
emulsions would be increased, and the stability of emulsions was reduced with low pH or high salt
33
concentration due to electrostatic screening of the negatively charged CNC particles. In conclusion,
34
high stability of D-limonene Pickering emulsions could be obtained by CNCs.
M
an
us
cr
ip t
25
35
39 40 41 42 43
ed
38
pt
37
Key words: D-limonene; Cellulose nanocrystals; Pickering emulsions; Stability
Ac ce
36
44 45 46 47
2
Page 2 of 25
48
1. Introduction D-limonene (4-isopropenyl-1-methylcyclohexene) is a major flavor component of citrus fruits. It
50
is reported that D-limonene has bactericide, antioxidant, chemo-preventative and therapeutic
51
activities, so it is widely used in cosmetics, foods, and other consumer products (Li & Chiang, 2012).
52
However, since D-limonene is insoluble in water and easy to oxidative degradation, it is difficult to
53
keep the lemon-like flavor especially in the areas such as water-rich phases and liquid-solid
54
interfaces. Therefore, it is very important and necessary to utilize a delivery system to protect D-
55
limonene from chemical degradation and improve its water-solubility. Recently, much attention has
56
been paid to emulsion technology to solve the above problems.
us
cr
ip t
49
Emulsions are heterogeneous systems consisting of droplets of a liquid dispersed in another non-
58
miscible or partly miscible liquid, and stabilized by surfactants, surface-active polymers, solid
59
particles or natural polymers such as proteins and polysaccharides (Chen et al., 2011; Bouyer et al.,
60
2012.). However, the classical synthetic emulsifiers get more and more limited with increasing legal
61
and consumer requirements such as non-toxicity, biocompatibility and high ecological acceptability
62
(Marku et al., 2012; Zoppe, Venditti, & Rojas, 2012). Pickering emulsions are emulsions of any type,
63
either oil-in water(o/w), water-in-oil (w/o), or multiple, stabilized by solid particles instead of
64
surfactants ( Aveyard, Binks, & Clint, 2003; Binks, 2002; Binks, Horozov, 2006). Because of the
65
highly enhanced stabilization against coalescence and Oswald ripening, Pickering emulsions could
66
conserve the droplets under high concentration of dispersed phase. There has been increasing interest
67
in renewable, environment-friendly biomaterials recent years, and it is a good choice to introduce
68
these materials into solid particles preparation to stabilize Pickering emulsions. Several
69
investigations have reported to use cellulose to stabilize oil in water emulsions (Kalashnikova et al.,
70
2011b, a, 2011). Cellulose nanocrystals (CNCs) are regarded as ideal biomaterial due to its nice
71
properties, such as low density, chemical tunability, environmental sustainability, and anticipated
72
low cost (Leung et al., 2011).
Ac ce
pt
ed
M
an
57
3
Page 3 of 25
In this study, CNCs were prepared from corncob cellulose hydrolyzed by ammonium persulfate.
74
Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD) and Thermogravimetric
75
analysis (TGA) techniques were used to investigate cellulose structure, crystallinity index and
76
thermal stability respectively. Then D-limonene Pickering emulsions were successful prepared by
77
sonicating D-limonene and CNCs aqueous dispersion, and their properties were studied by laser
78
particle size analyzer and microscope. Furthermore, effects of different environmental factors on the
79
stability of Pickering emulsion were analyzed in detail.
cr
ip t
73
81
us
80
2. Materials and Methods 2.1. Materials.
83
Corncob cellulose was supplied by Fustian Pharmaceutical co., ltd (Shan Dong province, China).
84
Ammonium persulfate (APS), sodium chloride and sodium hydroxide were all of reagent grade and
85
purchased from Beijing Chemicals (Beijing, China). D-limonene was obtained from Shanghai
86
Meishuang Company (Shanghai, China).
ed
M
an
82
2.2. Methods
88
2.2.1. Cellulose nanocrystals preparation
89
Two gram of corncob cellulose (CC) was added to 100 ml of 1 M APS solution, and the mixture
90
was heated to 60 for 16 h(Leunget al., 2011), and then a suspension of CNCs was obtained. The
91
suspension of CNCs was washed by repeated centrifugations at 5000 rpm for 10 min until the
92
solution conductivity was ≈ 5 μS cm-1 (pH4), close to that of deionized water.
Ac ce
pt
87
93
2.2.2. X-ray diffraction (XRD)
94
The X-ray diffractograms of CC and CNCs were obtained at room temperature within a 2θ ranging
95
from 5° to 80° and a scan rate of 0.02◦ min−1. The equipment used was a diffractometer Shimadzu
96
LabX XRD-6000, operating at a power of 40 kV with a current of 30 mA and Cu Kɑ radiation
97
(1.5406Å). Before performing the XRD, two samples were dried at 80 for 16 h in a vacuum drying
4
Page 4 of 25
98
oven. The crystallinity index (CRI) of CNCs was then estimated using the empirical method, as
99
shown in the follow Eq. CRI=
101
materials, respectively (Sheltami et al., 2012).
ip t
I 002 Iam 100% , where I002 and Iam are the peak intensities of crystalline and amorphous I 002
100
2.2.3. Fourier transfer-infra red (FTIR) spectroscopy
103
A Shimadzu IR Prestige-21 Infrared spectrophotometer was used to obtain spectra for the CC and
104
CNCs. The KBr disk (ultrathin pellets) method was used in taking the IR spectra. Samples were
105
ground and mixed with KBr (sample/KBr ratio, 1/100) to prepare pastilles. The experiments were
106
carried out in the range of 399-4000 cm−1 with a resolution of 2 cm−1 and a total of 32 scans for each
107
sample.
an
us
cr
102
2.2.4. Size distribution and Zeta-Potential Measurements
109
All experiments were carried out in dilute CNC suspension (0.2 w/w %) where CNC particles
110
randomly oriented. The dynamic light scattering (DLS, Zeta sizer Nano ZS, Malvern instrument) was
111
used to evaluate the size and zeta potential of CNC suspensions as functions of ionic concentration
112
and pH. This method provided indirect information for the nanoparticle size of CNCs. Samples were
113
measured in deionized water. CNCs particle sizes were measured after 5 min equilibration at 25 °C
114
and results are reported as the average of 3 measurements. The intensity average cellulose
115
nanocrystals diameter, and polydispersity of each sample was obtained from the Cumulant analysis
116
of each sample’s correlation function. The distribution of sizes was obtained using the CONTIN
117
analysis of each samples correlation function.
Ac ce
pt
ed
M
108
118
The zeta potential CNC particles in ionic solutions were measured over a range of NaCl
119
concentration (0–100 mM) at 25. The influence of pH (4.2–7.8) of CNC suspension was also
120
investigated. The CNC dispersion was mixed with appropriate quantities of NaOH solutions to
121
achieve a required pH. All suspensions were prepared in deionized water. Data are representative of
5
Page 5 of 25
122
at least three experiments varied by less than 5% in all the cases studied. 2.2.5. Thermogravimetric analysis (TGA)
124
Thermal stabilities of CC and CNC were evaluated by thermogravimetric analysis (TGA). The
125
analysis conditions were: a nitrogen atmosphere with flow 30 mLmin−1, heating rate of 5 ℃ min−1,
126
temperature range from 25 ℃ to 1000 ℃, sample mass between 5 and 7 mg and aluminum pans.
ip t
123
2.3. Preparation and characterization of Pickering emulsions
128
2.3.1. Preparation of D-limonene Pickering emulsions by CNCs
129
CNCs suspensions with different content of CNCs were prepared by ultrasonic dispersion method.
130
Practically, different amounts of CNCs powder (5, 10 and 20 mg) were added to 10 g deionized
131
water respectively and sonicated at 0.4 kW power level (3s sonication, 2s standby) for 3min. D-
132
limonene Pickering emulsions were prepared using D-limonene and CNCs aqueous suspension at the
133
required concentrations without further dilution to match an oil/water ratio of 10/90 (Kalashnikova et
134
al., 2011a). Practically, 0.5 g of D-limonene was added to 4.5 g of CNCs aqueous suspension in a
135
plastic vial and sonicated at 0.4 kW power level (3s soniacation, 2s standby) for 1minute.
ed
M
an
us
cr
127
2.3.2. Morphology of D-limonene Pickering emulsion
137
Emulsion droplets were visualized using optical micrograph. The CNCs-stabilized Pickering
138
emulsions were captured by an Olympus BX 51 optical microscope fitted with a digital camera
139
(Olympus, DP50). Emulsion droplets were placed directly onto a glass microscope slide and viewed
140
under 4-40×magnification.
Ac ce
pt
136
141
2.3.3. Droplet diameters measure
142
Droplet diameters were measured by laser light diffraction using a Malvern 2000 granulometer
143
apparatus equipped with a He-Ne laser (Malvern Instruments,U.K.) with Fraunhofer diffraction.
144
2.3.4 Zeta potential measurements of emulsions
145
The zeta potential of CNCs stabilized emulsion was measured using a particle electrophoresis
146
instrument (Zetasizer Nano ZS, Malvern Instruments Ltd., Malvern, and Worcestershire, UK).
6
Page 6 of 25
Immediately prior to measurements, the samples were diluted to a concentration of approximately
148
0.05% w/w using deionized water to avoid multiple scattering effects. The zeta-potential was
149
reported as the average and standard deviation of measurements made on three freshly prepared
150
samples, with three readings taken per sample.
151
2.3.5. Test of emulsion stability to creaming.
152
The stability to creaming of the generated emulsions was checked by centrifugation for 10 min at
153
4000 rpm and the thickness of the creaming layer was measured with a digital caliper and converted
154
to volume (Kalashnikova et al., 2011a). Different factors that may influence the stability of
155
emulsions to creaming were also studied. The emulsions were prepared as described above using 0.5
156
g D-limonene and 4.5 g CNCs suspensions with various concentrations, i.e. 0.05%, 0.10%, 0.20%
157
w/w CNCs. These emulsions were then used to different pH and temperature and NaCl solutions for
158
the stability tests to creaming. For the thermal stability test, the diluted emulsions were incubated in a
159
water bath for 30 min at 20, 30, 40, 50, 60 and 70℃. The pH stability test was performed by
160
adjusting pH of the emulsions to 4.2, 4.8, 5.5, 6.2, 7.0 and 7.8 using 0.1 M NaOH solution. For the
161
stability tests to ionic strength, the emulsions were diluted using NaCl solutions to a final
162
concentration of 0-100 mM NaCl. After each treatment, the treated emulsions were centrifuged and
163
the thickness of the creaming layer was measured with a digital caliper and converted to volume.
165
cr
us
an
M
ed
pt
Ac ce
164
ip t
147
3. Results and discussion
166
3.1 Preparation and characterization of cellulose nanocrystals
167
CNCs were obtained from CC with a novel one-step procedure using APS. The yield of CNCs,
168
under the condition of reaction to 16 h, with respect to the initial amount of dried CC was 50%. APS
169
reaction leads to stable aqueous suspensions of CNCs which are negatively charged, thus, are very
170
stable (Boluk et al., 2011). During the APS oxidation process, carboxylic acid groups presented. The
171
reaction conditions used led to homogeneous and stable aqueous suspensions. The size distribution
7
Page 7 of 25
of CNCs by intensity was showed in Fig.1, and mean size of CNCs was 363nm, which showed that
173
CNCs were obtained by APS hydrolysis. The X-ray diffraction patterns of CC and CNCs were
174
shown in Fig.2. In CC diffractogram, there was a predominance of type I cellulose, verified by the
175
presence of peaks at 2θ=16.4°, 22.5°, and 34°, and the amorphous background was characterized by
176
the low diffracted intensity at around 18◦. The CNC diffractogram exhibited the most intense peak
177
(002) with a shoulder (101) and one lower peak (004). Such features resembled the diffraction
178
pattern of cellulose I (Borysiak and Garbarczyk, 2003; Flauzino Neto et al., 2013) and confirmed the
179
integrity of the material during the course of treatment with APS. The crystallinity index (CrI) of CC
180
and CNCs were 63.6% and 67.2%. From FTIR of CC and CNCs (Fig.3), it can be found that CC and
181
CNCs all showed adsorption bands at 3500 cm-1, which was the signal of alcoholic hydroxyl groups.
182
The IR spectra of CNCs displayed a signal at 1735cm−1, which is attributed to C=O. The presence of
183
signals at 1429, 1161, 1110, and 897 cm−1indicated that the CNCs were primarily in the form of
184
cellulose I, and the absence of OH stretching and out-of-plane bending at 3240 and 750 cm-1
185
characteristic of cellulose Iα confirmed the cellulose was Iβ structure（Jiang, Han, & Hsieh, 2013
186
Leung et al., 2011), indicating no significant changes of the cellulose structure and mercerization did
187
not occur(Leunget al., 2011). Thermal decomposition parameters were determined from the TG,
188
DTG curves as described below at a heating rate of 5 ℃/min. The TG and DTG curves of CC and
189
CNCs powder are shown in Fig.4A and Fig.4B respectively. For two samples, the weight loss
190
profiles exhibited essentially two events. The first was related to the evaporation of water-adsorbed
191
materials. Here a small weight loss was found in the range of 30-140 ℃ . The second event
192
corresponded to the process of cellulose degradation. In the present study, the decomposition of the
193
CC showed a main stage, indicating that the composition of material was main cellulose and lignin.
194
the DTG curve of CC showed a main peak, which accounted for the pyrolysis of cellulose (about 299
195
℃). But CNCs has an increased thermal stability. The thermal degradation of CNCs shows an
Ac ce
pt
ed
M
an
us
cr
ip t
172
8
Page 8 of 25
196
increased value 323℃ than CC of 299℃. As the result, nanocellulose obtained from corncob
197
cellulose has higher thermal stability than CC. The small amount of hemicelluloses and lignin
198
existing in CC were removed by the oxidation of APS, and the stability of CNCs with high
199
crystalline might be improved. 3.2. Emulsion Preparation.
201
The aqueous phase, for all tested emulsions consisted of CNCs dispersed in water at the required
202
concentration without further treatment. D-limonene was added afterward. The biphasic system was
203
then sonicated, resulting in a stable oil-water emulsion. Fig.5 shows the optical microscopy image of
204
D-limonene Pickering emulsion. Emulsions prepared with 0.05 w/w % CNCs yielded large droplets
205
(>5 μm), as shown in Fig.5A, compared with those made with higher concentrations, as 0.2 w/w %
206
in Fig.5B. Fig.6 showed droplets of emulsions under different CNCs concentrations. Larger emulsion
207
droplets were much less stable under low shear by mixing compared with the smaller droplets
208
formed at higher nanoparticle concentrations. In the case of 0.05 w/w % CNCs, the volume mean
209
particle diameters were about 6.9 μm. When the concentration of CNCs was up to 0.2 w/w%, the
210
volume mean particle diameters were 4.2 μm. It was obvious that the concentration of CNCs had a
211
significant impact on the particle size of emulsions.
pt
ed
M
an
us
cr
ip t
200
3.3. Stability of emulsions to creaming under different temperature ionic, concentration and pH.
213
Pickering emulsions were formulated at a 10/90 ratio (o/w), with 0.2w/w % suspension of CNCs
214
in the water phase. And then the prepared emulsions were used to test emulsion to creaming stability
215
under different factors, i.e. temperature, pH and ionic concentration. Temperature is an important
216
factor that could influence the stability of Pickering emulsions. Ionic strength and pH also play
217
important roles in regulating the electrostatic interactions that may occur between adjacent
218
nanoparticles at the oil–water interface (Dyab, Binks, & Fletcher, 2012; Zoppe, Venditti, & Rojas,
219
2012). Therefore, temperature, salt concentration, pH and tests were then carried out to specify
220
stability characteristics.
Ac ce
212
9
Page 9 of 25
The influence of temperature on the behaviour of CNCs-stabilized emulsions was explored by
222
applying samples under different temperature from 20-70 ℃ as shown in Fig.7. Results showed that
223
the emulsions were heated from 20-70 ℃, the stability of emulsions to creaming increased. It could
224
be explained by that CNCs were irreversibly adsorbed to cover emulsions droplets and formed 2 D-
225
network, when temperature raised, CNCs tended to give a stronger structure, as a result, the inter-
226
particle distances were decreased and the associative forces among the nanocrystals were enhanced
227
(Tzoumaki et al., 2009a, b, 2011). Thus, the stability of emulsions to creaming was enhanced.
228
Besides, it was also indicated that the increase in the stability of emulsions became higher with
229
increasing CNCs concentration. That was because under the condition of low CNCs concentration,
230
the inter-particle of CNCs had weak associative forces, so CNCs formed a weak droplet network.
231
When CNCs concentration increased, the particle could form a stronger network, as a result, the
232
stability of emulsions increased.
M
an
us
cr
ip t
221
Effects of pH and ionic concentration on stabilizing emulsions were examined by measuring the
234
zeta-potential of the droplets. The dependence of zeta-potential of the droplets in 0.2w/w % CNCs
235
stabilized emulsion on the ionic environment (NaCl concentration) were similar to that of the 0.2
236
wt% CNCs suspension, and the results are showed in Fig.8 A and Fig.8 C. Aqueous solutions with 0,
237
20, 40, 60, 80 and 100 mM NaCl concentrations were added to aqueous dispersions of CNCs and
238
emulsions. As the salt concentration increased, there was a greater tendency for CNCs to aggregate,
239
since the stabilizing repulsive forces were weakened by the increase in electrolyte concentration
240
(Tzoumaki et al., 2011). When NaCl concentrations reached 100mM, a small amount of CNCs
241
aggregated in the CNCs suspension, and zeta potential of CNCs suspension and emulsion entered the
242
domain of low-charged surfaces below 20mV. It was observed that the absolute value of the zeta-
243
potential of the emulsion droplets across the whole ranges of NaCl concentration remained relatively
244
lower than those of the CNCs particles. The low net charge on the droplets could be explained in
245
terms of the relatively low surface load of CNCs and/or because of ionic impurities in the samples
Ac ce
pt
ed
233
10
Page 10 of 25
that adsorbed to the surface of the oil droplets and changed their electrical charge (Surh, Decker, &
247
McClements, 2006). The magnitude of zeta-potential of the emulsion droplets decreased with
248
increasing the concentration of NaCl from 46.3 mV with no added salt to16.7 mV at 100 mM NaCl
249
due to the electrostatic screening of the charged (-COO-) groups by the counter-ions (Na+). It is
250
conventionally recognized that the larger the absolute magnitude of zeta-potential, the greater is the
251
electrostatic repulsion between droplets, and therefore the better the stability to their surface charge.
252
Similar responses to those obtained by varying the ionic strength were also noted by changing the pH.
253
As can be seen in Fig.8B and Fig.8D, the trends of zeta-potential of the droplets in 0.2w/w % CNCs
254
stabilized emulsion on different pH environment were similar to that of the 0.2wt% CNCs
255
suspension, and the zeta potential of CNCs suspension and emulsion were both increased when pH
256
increasing. The 0.20 w/w % CNCs-stabilized emulsion volume was increased with increasing of pH
257
value from 4.2 to 7.8. The magnitude of zeta-potential increased when 0.1 M NaOH was added, and
258
the zeta-potential of the emulsion droplets and CNCs remained negative at all pH values, possibly
259
because of the negatively charged -COO- groups on CNC particles. The CNCs has no cationic
260
groups and therefore did not become positively charged at any pH. An increase in pH from 4.2 to 7.8
261
caused an increase in the magnitude of negative charge on the droplets from -32.9 to -46.3 mV,
262
which was probably due to deprotonation of some of the protonated carboxyl groups (COOH→
263
COO- + H+) (Winuprasith & Suphantharika, 2013). It is conventionally recognized that increasing of
264
the surface charge can improve the stability of emulsions because of the increasing repulsive forces
265
between droplets against flocculation and coalescence. In our study, when the pH of the system was
266
increased from 4.2 to 7.8, the surface charge decreased from -42.9mV to -54.5 mV. So the larger the
267
absolute magnitude of zeta-potential, the greater is the electrostatic repulsion between droplets, and
268
therefore the better the stability (Dickinson, 2009).
Ac ce
pt
ed
M
an
us
cr
ip t
246
269
4. Conclusion
270
In this study, cellulose nanocrystals were prepared from corncob cellulose by one step with 11
Page 11 of 25
ammonium persulfate hydrolysis. The degree of cellulose crystallinity increased during ammonium
272
persulfate hydrolysis, which could be explained by the degradation of the amorphous cellulose
273
during reaction process. Besides, the hydrolysis conditions used led to obtain stable aqueous
274
suspensions of CNC which are negatively charged, due to the presence of carboxyl groups. D-
275
limonene Pickering emulsions were prepared by CNCs at high sonication power and long time with
276
0.2% (w/w) CNCs exhibited satisfactory size and high stability to creaming. Furthermore, the
277
stability of the CNCs-stabilized emulsions was decreased at low pH or high salt concentration due to
278
the electrostatic screening effect. In addition, the stability of emulsion increased when temperature
279
increased from 20 ℃ to 70 ℃, during the thermal treatment. The possible reason may be that CNCs
280
were irreversibly adsorbed to cover emulsions droplets and formed 2D-network. Thus, CNCs from
281
corncob cellulose are useful stabilizer for preparing D-limonene Pickering emulsion.
an
us
cr
ip t
271
Acknowlegment
ed
283
M
282
This study was supported by the National Natural Science Foundation of China (Grant No.
285
20806005) and the National High Technology Research and Development Program of China (863
286
Program, Grant No. 2012AA021403).
288
Ac ce
287
pt
284
289 290 291 292 12
Page 12 of 25
293
References
294
Aveyard, R., Binks, B. P., & Clint, J. H. (2003). Emulsions stabilised solely by colloidal particles.
297 298 299
Binks, B. P. (2002). Particles as surfactants—similarities and differences. Current Opinion in Colloid
ip t
296
Advances in Colloid and Interface Science, 100, 503-546.
& Interface Science, 7(1), 21-41.
Binks, B. P., & Horozov, T. S. (2006). Colloidal particles at liquid interfaces. Cambridge University
cr
295
In Press.
Boluk, Y., Lahiji, R., Zhao, L., & McDermott, M. T. (2011). Suspension viscosities and shape
301
parameter of cellulose nanocrystals (CNC). Colloids and Surfaces A: Physicochemical and
302
Engineering Aspects, 377(1), 297-303.
an
us
300
Borysiak, S., & Garbarczyk, J. (2003). Applying the WAXS method to estimate the supermolecular
304
structure of cellulose fibres after mercerisation. Fibres and Textiles in Eastern Europe, 11(5),
305
104-106.
ed
M
303
Bouyer, E., Mekhloufi, G., Rosilio, V., Grossiord, J. L., & Agnely, F. (2012). Proteins,
307
polysaccharides, and their complexes used as stabilizers for emulsions: Alternatives to synthetic
308
surfactants in the pharmaceutical field?. International Journal of Pharmaceutics.
Ac ce
pt
306
309
Chen, J., Vogel, R., Werner, S., Heinrich, G., Clausse, D., & Dutschk, V. (2011). Influence of the
310
particle type on the rheological behavior of Pickering emulsions. Colloids and Surfaces A:
311
Physicochemical and Engineering Aspects, 382(1), 238-245.
312 313 314 315
Dickinson, E. (2009). Hydrocolloids as emulsifiers and emulsion stabilizers. Food Hydrocolloids, 23(6), 1473-1482. Dyab, A. K., Binks, B. P., & Fletcher, P. D. (2012). Destabilisation of Pickering emulsions using pH. Colloids and Surfaces A: Physicochemical and Engineering Aspects.
13
Page 13 of 25
316
Flauzino Neto, W. P., Silvério, H. A., Dantas, N. O., & Pasquini, D. (2013). Extraction and
317
characterization of cellulose nanocrystals from agro-industrial residue–Soy hulls. Industrial
318
Crops and Products, 42, 480-488.
322 323 324
ip t
Kalashnikova, I., Bizot, H., Cathala, B., & Capron, I. (2011a). Modulation of cellulose nanocrystals
cr
321
assembly of cellulose nanofibrils into highly crystalline fibrous materials. RSC Adv.
amphiphilic properties to stabilize oil/water interface. Biomacromolecules, 13(1), 267-275.
us
320
Jiang, F., Han, S., & Hsieh, Y. L. (2013). Controlled defibrillation of rice straw cellulose and self-
Kalashnikova, I., Bizot, H., Cathala, B., & Capron, I. (2011b). New Pickering emulsions stabilized by bacterial cellulose nanocrystals. Langmuir, 27(12), 7471-7479.
an
319
Leung, A. C., Hrapovic, S., Lam, E., Liu, Y., Male, K. B., Mahmoud, K. A., & Luong, J. H. (2011).
326
Characteristics and Properties of Carboxylated Cellulose Nanocrystals Prepared from a Novel
327
One-Step Procedure. Small, 7(3), 302-305.
ed
M
325
Li, P. H., & Chiang, B. H. (2012). Process optimization and stability of D-limonene-in-water
329
nanoemulsions prepared by ultrasonic emulsification using response surface methodology.
330
Ultrasonics sonochemistry, 19(1), 192-197.
Ac ce
pt
328
331
Marku, D., Wahlgren, M., Rayner, M., Sjöö, M., & Timgren, A. (2012). Characterization of starch
332
Pickering emulsions for potential applications in topical formulations. International Journal of
333
Pharmaceutics.
334
Sheltami, R. M., Abdullah, I., Ahmad, I., Dufresne, A., & Kargarzadeh, H. (2012). Extraction of
335
cellulose nanocrystals from mengkuang leaves (Pandanus tectorius). Carbohydrate Polymers,
336
88(2), 772-779.
337 338
Surh, J., Decker, E. A., & McClements, D. J. (2006). Properties and stability of oil-in-water emulsions stabilized by fish gelatin. Food Hydrocolloids, 20(5), 596-606.
14
Page 14 of 25
339 340 341 342
Tzoumaki, M. V., Moschakis, T., & Biliaderis, C. G. (2009). Metastability of nematic gels made of aqueous chitin nanocrystal dispersions. Biomacromolecules, 11(1), 175-181. Tzoumaki, M. V., Moschakis, T., Kiosseoglou, V., & Biliaderis, C. G. (2011). Oil-in-water emulsions stabilized by chitin nanocrystal particles. Food hydrocolloids, 25(6), 1521-1529. Winuprasith, T., & Suphantharika, M. (2013). Microfibrillated cellulose from mangosteen (Garcinia
344
mangostana L.) rind: Preparation, characterization, and evaluation as an emulsion stabilizer.
345
Food Hydrocolloids.
cr
ip t
343
Zoppe, J. O., Venditti, R. A., & Rojas, O. J. (2012). Pickering emulsions stabilized by cellulose
347
nanocrystals grafted with thermo-responsive polymer brushes. Journal of colloid and interface
348
science, 369(1), 202-209.
an
us
346
349
M
350
354 355 356 357 358
pt
353
Ac ce
352
ed
351
359 360 361 362 15
Page 15 of 25
363 364
Figure Captions
365
Fig.1. Size distribution of CNCS by volume and CNCs suspension (0.2wt %) after treatment with 1
367
M APS.
ip t
366
368
Fig.2. X-ray diffraction patterns for CC and CNCS.
cr
369 370
Fig.3. FTIR spectra of CC and CNCs.
372 373
Fig.4. (A) TG of CC and CNCs; (B) DTG of CC and CNCs.
an
374
us
371
Fig.5. Optical microscopy images of D-limonene Pickering emulsions prepared with 0.05% (A) and
376
0.20% (B) CNCs (oil to water ratio of 1:9).
377 378
M
375
Fig.6. Emulsion droplet with different CNCs concentration.
ed
379
Fig.7. Effect of temperature on volume of emulsion with (A) 0.2w/w% CNCs and (B) 0.1w/w%
381
CNCs.
pt
380
382
Fig.8. Effect of the ionic strength (with NaCl) and pH (with NaOH) on CNCs suspension and D-
384
limonene Pickering emulsions zeta potential and emulsion volume with increasing ionic
385
concentration and pH.
386 387 388 389 390 391 392 393 394 395 396
Ac ce
383
16
Page 16 of 25
20
ip t
10
cr
Intensity (%)
15
0 300
600
900
1200
1500
1800
2100
2400
2700
3000
an
0
us
5
Size (d.nm)
M
397 398 399
Ac ce
pt
ed
Fig 1
17
Page 17 of 25
400 401 402
22.5
2000
403 404
408
1000
ip t
407
CNC
16.4
cr
406
1500
Intensity count
405
34 500
410
0
CC 0
10
412
20
an
411 30
40
50
60
70
80
90
Scattering angel (degrees of 2 theta)
M
413 414
pt
ed
Fig 2
Ac ce
415
us
409
18
Page 18 of 25
415 110
416 100
417 90
418
423 424 425 426
ip t
60
CNC
50
cr
422
1735
70
40 30 20 10 4000
3500
3000
2500
2000
us
421
CC
1500
1000
an
420
80
Transmitance (a.u.)
419
500
0
-1
Wavenumbers (cm )
427
M
428 429
Fig 3
Ac ce
pt
ed
430
19
Page 19 of 25
CC CNC
A
100
100
80
442 443
40
cr
441
60
60
40
20
20
0
444
0 0
200
400
600
0.002
B
449
455 456 457 458
ed
-0.004
-0.006
-0.008
0.000
-0.005
pt
Derivated weight of CC (mg/min)
454
-0.002
Ac ce
453
0.000
-0.010
-0.010
299
-0.012
0
200
-0.015
323 400
600
800
Derivated weight of CNC (mg/min)
452
CC CNC
M
447
451
1000
Temperature (℃)
446
450
800
an
445
448
us
440
ip t
80
Weight loss of CNC (%)
431 432 433 434 435 436 437 438 439
.
Weight loss of CC (%)
430
1000
Temperature (℃)
Fig 4
459 460
20
Page 20 of 25
460 461 462
A A
20μm
20μm
BB
463 464
ip t
465 466
cr
467 468
us
469 470 471
an
Fig 5
Ac ce
pt
ed
M
472
21
Page 21 of 25
472 473
0.1% CNCs 0.2% CNCs 0.05% CNCs
16
ip t
14
cr
10 8 6
us
Volume (%)
12
2 0 -2
0
2
4
6
8
an
4
10
12
14
16
18
20
M
Droplet diameter (µm) 474 475
ed pt Ac ce
476
Fig 6
22
Page 22 of 25
476 0.1% CNCs 0.2% CNCs
477
A
478
483 484 485 486
B
335
330 20
30
40
50
60
Temperature (℃ )
488
Fig 7
489
M
490
pt
494
Ac ce
493
ed
491 492
an
487
70
ip t
482
340
cr
481
345
us
480
Volume of emulsions (μL)
479
350
23
Page 23 of 25
494 495
CNCs Pickering emulsion
-38
499 500
-42
-25 -30 -35 -40
-44 -46 -48 -50 -52 -54
-45
501
-56
-50 0
20
502
40
60
80
100
4.0
500 480
514 515
400
-35
380 360 340
-45
6.5
7.0
7.5
us
M
-30
ed
Zeta potential (mV)
420
8.0
zeta potential of emulsion volume of emulsion 460
-42
440 420
-44
400
-46
380 -48
360 340
-50
320 -52
300 280
-54
320
-50 0
20
40
pt
513
440
60
80
100
260 -56 4.0
Ionic concentration (mmol/L)
Ac ce
512
-25
-40
6.0
Volume of emulsions (μL）
511
460
Volume of emulsions (µL)
510
-20
Zeta potential of emulsions (mV)
D
-15
506 C
5.5
an
zeta potential volume of emulsions
505
509
5.0
pH
504
508
4.5
Ionic concentration (mmol/L)
503
507
cr
498
-40
B
-20
Zeta potential (mV)
497
-15
ip t
A Zeta potential (mV)
496
Pickering emulsion CNCs
-10
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
pH
Fig 8
24
Page 24 of 25
515
Highlights
516
> Cellulose nanocrystals were successfully prepared by ammonium persulfate hydrolysis.>
518
D-liomonene Pickering emulsions prepared by cellulose nanocrystals exhibited high
519
stability.> Electrostatic interactions and temperature had obvious effect on emulsion
520
stability to creaming.
cr
ip t
517
Ac ce
pt
ed
M
an
us
521
25
Page 25 of 25
S0144-8617(14)00627-4 http://dx.doi.org/doi:10.1016/j.carbpol.2014.06.051 CARP 9016
To appear in: Received date: Revised date: Accepted date:
4-7-2013 15-9-2013 19-6-2014
Please cite this article as: Wen, C., Yuan, Q., Liang, H., and Vriesekoop, F.,Preparation and Stabilization of D-limonene Pickering Emulsions by Cellulose Nanocrystals, Carbohydrate Polymers (2014), http://dx.doi.org/10.1016/j.carbpol.2014.06.051 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Preparation and Stabilization of D-limonene Pickering Emulsions by Cellulose Nanocrystals
1 2 3
Chunxia Wena, Qipeng Yuana, Hao Lianga,*, Frank Vriesekoopb
4
a
6 7
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical
b
cr
Technology, Beijing 100029, People’s Republic of China
Department of Food Science, Harper Adams University, Newport, Shropshire, TF10 8NB, England
us
8
ip t
5
9
an
10 11
M
12 13
ed
14 15
pt
16
Ac ce
17 18 19 20 21 22
*
Corresponding author. Tel.: +86-10-6443-1557; Fax: +86-10-6443-7610.
E-mail address: [email protected].
1
Page 1 of 25
23 24
Abstract The aim of this study was to investigate D-limonene Pickering emulsion stabilized by cellulose
26
nanocrystals (CNCs) and factors that may affect its properties. CNCs were prepared by ammonium
27
persulfate hydrolysis of corncob cellulose, and D-limonene Pickering emulsions were generated by
28
ultrasonic homogenizing method. The morphology and size of the prepared emulsions with different
29
CNCs concentrations were studied by optical microscopy and laser light diffraction. In addition,
30
factors that may affect the stability of emulsions such as ionic concentration, pH and temperature
31
were also studied. As indicated by the experiment data, when temperature rose, the stability to of
32
emulsions would be increased, and the stability of emulsions was reduced with low pH or high salt
33
concentration due to electrostatic screening of the negatively charged CNC particles. In conclusion,
34
high stability of D-limonene Pickering emulsions could be obtained by CNCs.
M
an
us
cr
ip t
25
35
39 40 41 42 43
ed
38
pt
37
Key words: D-limonene; Cellulose nanocrystals; Pickering emulsions; Stability
Ac ce
36
44 45 46 47
2
Page 2 of 25
48
1. Introduction D-limonene (4-isopropenyl-1-methylcyclohexene) is a major flavor component of citrus fruits. It
50
is reported that D-limonene has bactericide, antioxidant, chemo-preventative and therapeutic
51
activities, so it is widely used in cosmetics, foods, and other consumer products (Li & Chiang, 2012).
52
However, since D-limonene is insoluble in water and easy to oxidative degradation, it is difficult to
53
keep the lemon-like flavor especially in the areas such as water-rich phases and liquid-solid
54
interfaces. Therefore, it is very important and necessary to utilize a delivery system to protect D-
55
limonene from chemical degradation and improve its water-solubility. Recently, much attention has
56
been paid to emulsion technology to solve the above problems.
us
cr
ip t
49
Emulsions are heterogeneous systems consisting of droplets of a liquid dispersed in another non-
58
miscible or partly miscible liquid, and stabilized by surfactants, surface-active polymers, solid
59
particles or natural polymers such as proteins and polysaccharides (Chen et al., 2011; Bouyer et al.,
60
2012.). However, the classical synthetic emulsifiers get more and more limited with increasing legal
61
and consumer requirements such as non-toxicity, biocompatibility and high ecological acceptability
62
(Marku et al., 2012; Zoppe, Venditti, & Rojas, 2012). Pickering emulsions are emulsions of any type,
63
either oil-in water(o/w), water-in-oil (w/o), or multiple, stabilized by solid particles instead of
64
surfactants ( Aveyard, Binks, & Clint, 2003; Binks, 2002; Binks, Horozov, 2006). Because of the
65
highly enhanced stabilization against coalescence and Oswald ripening, Pickering emulsions could
66
conserve the droplets under high concentration of dispersed phase. There has been increasing interest
67
in renewable, environment-friendly biomaterials recent years, and it is a good choice to introduce
68
these materials into solid particles preparation to stabilize Pickering emulsions. Several
69
investigations have reported to use cellulose to stabilize oil in water emulsions (Kalashnikova et al.,
70
2011b, a, 2011). Cellulose nanocrystals (CNCs) are regarded as ideal biomaterial due to its nice
71
properties, such as low density, chemical tunability, environmental sustainability, and anticipated
72
low cost (Leung et al., 2011).
Ac ce
pt
ed
M
an
57
3
Page 3 of 25
In this study, CNCs were prepared from corncob cellulose hydrolyzed by ammonium persulfate.
74
Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD) and Thermogravimetric
75
analysis (TGA) techniques were used to investigate cellulose structure, crystallinity index and
76
thermal stability respectively. Then D-limonene Pickering emulsions were successful prepared by
77
sonicating D-limonene and CNCs aqueous dispersion, and their properties were studied by laser
78
particle size analyzer and microscope. Furthermore, effects of different environmental factors on the
79
stability of Pickering emulsion were analyzed in detail.
cr
ip t
73
81
us
80
2. Materials and Methods 2.1. Materials.
83
Corncob cellulose was supplied by Fustian Pharmaceutical co., ltd (Shan Dong province, China).
84
Ammonium persulfate (APS), sodium chloride and sodium hydroxide were all of reagent grade and
85
purchased from Beijing Chemicals (Beijing, China). D-limonene was obtained from Shanghai
86
Meishuang Company (Shanghai, China).
ed
M
an
82
2.2. Methods
88
2.2.1. Cellulose nanocrystals preparation
89
Two gram of corncob cellulose (CC) was added to 100 ml of 1 M APS solution, and the mixture
90
was heated to 60 for 16 h(Leunget al., 2011), and then a suspension of CNCs was obtained. The
91
suspension of CNCs was washed by repeated centrifugations at 5000 rpm for 10 min until the
92
solution conductivity was ≈ 5 μS cm-1 (pH4), close to that of deionized water.
Ac ce
pt
87
93
2.2.2. X-ray diffraction (XRD)
94
The X-ray diffractograms of CC and CNCs were obtained at room temperature within a 2θ ranging
95
from 5° to 80° and a scan rate of 0.02◦ min−1. The equipment used was a diffractometer Shimadzu
96
LabX XRD-6000, operating at a power of 40 kV with a current of 30 mA and Cu Kɑ radiation
97
(1.5406Å). Before performing the XRD, two samples were dried at 80 for 16 h in a vacuum drying
4
Page 4 of 25
98
oven. The crystallinity index (CRI) of CNCs was then estimated using the empirical method, as
99
shown in the follow Eq. CRI=
101
materials, respectively (Sheltami et al., 2012).
ip t
I 002 Iam 100% , where I002 and Iam are the peak intensities of crystalline and amorphous I 002
100
2.2.3. Fourier transfer-infra red (FTIR) spectroscopy
103
A Shimadzu IR Prestige-21 Infrared spectrophotometer was used to obtain spectra for the CC and
104
CNCs. The KBr disk (ultrathin pellets) method was used in taking the IR spectra. Samples were
105
ground and mixed with KBr (sample/KBr ratio, 1/100) to prepare pastilles. The experiments were
106
carried out in the range of 399-4000 cm−1 with a resolution of 2 cm−1 and a total of 32 scans for each
107
sample.
an
us
cr
102
2.2.4. Size distribution and Zeta-Potential Measurements
109
All experiments were carried out in dilute CNC suspension (0.2 w/w %) where CNC particles
110
randomly oriented. The dynamic light scattering (DLS, Zeta sizer Nano ZS, Malvern instrument) was
111
used to evaluate the size and zeta potential of CNC suspensions as functions of ionic concentration
112
and pH. This method provided indirect information for the nanoparticle size of CNCs. Samples were
113
measured in deionized water. CNCs particle sizes were measured after 5 min equilibration at 25 °C
114
and results are reported as the average of 3 measurements. The intensity average cellulose
115
nanocrystals diameter, and polydispersity of each sample was obtained from the Cumulant analysis
116
of each sample’s correlation function. The distribution of sizes was obtained using the CONTIN
117
analysis of each samples correlation function.
Ac ce
pt
ed
M
108
118
The zeta potential CNC particles in ionic solutions were measured over a range of NaCl
119
concentration (0–100 mM) at 25. The influence of pH (4.2–7.8) of CNC suspension was also
120
investigated. The CNC dispersion was mixed with appropriate quantities of NaOH solutions to
121
achieve a required pH. All suspensions were prepared in deionized water. Data are representative of
5
Page 5 of 25
122
at least three experiments varied by less than 5% in all the cases studied. 2.2.5. Thermogravimetric analysis (TGA)
124
Thermal stabilities of CC and CNC were evaluated by thermogravimetric analysis (TGA). The
125
analysis conditions were: a nitrogen atmosphere with flow 30 mLmin−1, heating rate of 5 ℃ min−1,
126
temperature range from 25 ℃ to 1000 ℃, sample mass between 5 and 7 mg and aluminum pans.
ip t
123
2.3. Preparation and characterization of Pickering emulsions
128
2.3.1. Preparation of D-limonene Pickering emulsions by CNCs
129
CNCs suspensions with different content of CNCs were prepared by ultrasonic dispersion method.
130
Practically, different amounts of CNCs powder (5, 10 and 20 mg) were added to 10 g deionized
131
water respectively and sonicated at 0.4 kW power level (3s sonication, 2s standby) for 3min. D-
132
limonene Pickering emulsions were prepared using D-limonene and CNCs aqueous suspension at the
133
required concentrations without further dilution to match an oil/water ratio of 10/90 (Kalashnikova et
134
al., 2011a). Practically, 0.5 g of D-limonene was added to 4.5 g of CNCs aqueous suspension in a
135
plastic vial and sonicated at 0.4 kW power level (3s soniacation, 2s standby) for 1minute.
ed
M
an
us
cr
127
2.3.2. Morphology of D-limonene Pickering emulsion
137
Emulsion droplets were visualized using optical micrograph. The CNCs-stabilized Pickering
138
emulsions were captured by an Olympus BX 51 optical microscope fitted with a digital camera
139
(Olympus, DP50). Emulsion droplets were placed directly onto a glass microscope slide and viewed
140
under 4-40×magnification.
Ac ce
pt
136
141
2.3.3. Droplet diameters measure
142
Droplet diameters were measured by laser light diffraction using a Malvern 2000 granulometer
143
apparatus equipped with a He-Ne laser (Malvern Instruments,U.K.) with Fraunhofer diffraction.
144
2.3.4 Zeta potential measurements of emulsions
145
The zeta potential of CNCs stabilized emulsion was measured using a particle electrophoresis
146
instrument (Zetasizer Nano ZS, Malvern Instruments Ltd., Malvern, and Worcestershire, UK).
6
Page 6 of 25
Immediately prior to measurements, the samples were diluted to a concentration of approximately
148
0.05% w/w using deionized water to avoid multiple scattering effects. The zeta-potential was
149
reported as the average and standard deviation of measurements made on three freshly prepared
150
samples, with three readings taken per sample.
151
2.3.5. Test of emulsion stability to creaming.
152
The stability to creaming of the generated emulsions was checked by centrifugation for 10 min at
153
4000 rpm and the thickness of the creaming layer was measured with a digital caliper and converted
154
to volume (Kalashnikova et al., 2011a). Different factors that may influence the stability of
155
emulsions to creaming were also studied. The emulsions were prepared as described above using 0.5
156
g D-limonene and 4.5 g CNCs suspensions with various concentrations, i.e. 0.05%, 0.10%, 0.20%
157
w/w CNCs. These emulsions were then used to different pH and temperature and NaCl solutions for
158
the stability tests to creaming. For the thermal stability test, the diluted emulsions were incubated in a
159
water bath for 30 min at 20, 30, 40, 50, 60 and 70℃. The pH stability test was performed by
160
adjusting pH of the emulsions to 4.2, 4.8, 5.5, 6.2, 7.0 and 7.8 using 0.1 M NaOH solution. For the
161
stability tests to ionic strength, the emulsions were diluted using NaCl solutions to a final
162
concentration of 0-100 mM NaCl. After each treatment, the treated emulsions were centrifuged and
163
the thickness of the creaming layer was measured with a digital caliper and converted to volume.
165
cr
us
an
M
ed
pt
Ac ce
164
ip t
147
3. Results and discussion
166
3.1 Preparation and characterization of cellulose nanocrystals
167
CNCs were obtained from CC with a novel one-step procedure using APS. The yield of CNCs,
168
under the condition of reaction to 16 h, with respect to the initial amount of dried CC was 50%. APS
169
reaction leads to stable aqueous suspensions of CNCs which are negatively charged, thus, are very
170
stable (Boluk et al., 2011). During the APS oxidation process, carboxylic acid groups presented. The
171
reaction conditions used led to homogeneous and stable aqueous suspensions. The size distribution
7
Page 7 of 25
of CNCs by intensity was showed in Fig.1, and mean size of CNCs was 363nm, which showed that
173
CNCs were obtained by APS hydrolysis. The X-ray diffraction patterns of CC and CNCs were
174
shown in Fig.2. In CC diffractogram, there was a predominance of type I cellulose, verified by the
175
presence of peaks at 2θ=16.4°, 22.5°, and 34°, and the amorphous background was characterized by
176
the low diffracted intensity at around 18◦. The CNC diffractogram exhibited the most intense peak
177
(002) with a shoulder (101) and one lower peak (004). Such features resembled the diffraction
178
pattern of cellulose I (Borysiak and Garbarczyk, 2003; Flauzino Neto et al., 2013) and confirmed the
179
integrity of the material during the course of treatment with APS. The crystallinity index (CrI) of CC
180
and CNCs were 63.6% and 67.2%. From FTIR of CC and CNCs (Fig.3), it can be found that CC and
181
CNCs all showed adsorption bands at 3500 cm-1, which was the signal of alcoholic hydroxyl groups.
182
The IR spectra of CNCs displayed a signal at 1735cm−1, which is attributed to C=O. The presence of
183
signals at 1429, 1161, 1110, and 897 cm−1indicated that the CNCs were primarily in the form of
184
cellulose I, and the absence of OH stretching and out-of-plane bending at 3240 and 750 cm-1
185
characteristic of cellulose Iα confirmed the cellulose was Iβ structure（Jiang, Han, & Hsieh, 2013
186
Leung et al., 2011), indicating no significant changes of the cellulose structure and mercerization did
187
not occur(Leunget al., 2011). Thermal decomposition parameters were determined from the TG,
188
DTG curves as described below at a heating rate of 5 ℃/min. The TG and DTG curves of CC and
189
CNCs powder are shown in Fig.4A and Fig.4B respectively. For two samples, the weight loss
190
profiles exhibited essentially two events. The first was related to the evaporation of water-adsorbed
191
materials. Here a small weight loss was found in the range of 30-140 ℃ . The second event
192
corresponded to the process of cellulose degradation. In the present study, the decomposition of the
193
CC showed a main stage, indicating that the composition of material was main cellulose and lignin.
194
the DTG curve of CC showed a main peak, which accounted for the pyrolysis of cellulose (about 299
195
℃). But CNCs has an increased thermal stability. The thermal degradation of CNCs shows an
Ac ce
pt
ed
M
an
us
cr
ip t
172
8
Page 8 of 25
196
increased value 323℃ than CC of 299℃. As the result, nanocellulose obtained from corncob
197
cellulose has higher thermal stability than CC. The small amount of hemicelluloses and lignin
198
existing in CC were removed by the oxidation of APS, and the stability of CNCs with high
199
crystalline might be improved. 3.2. Emulsion Preparation.
201
The aqueous phase, for all tested emulsions consisted of CNCs dispersed in water at the required
202
concentration without further treatment. D-limonene was added afterward. The biphasic system was
203
then sonicated, resulting in a stable oil-water emulsion. Fig.5 shows the optical microscopy image of
204
D-limonene Pickering emulsion. Emulsions prepared with 0.05 w/w % CNCs yielded large droplets
205
(>5 μm), as shown in Fig.5A, compared with those made with higher concentrations, as 0.2 w/w %
206
in Fig.5B. Fig.6 showed droplets of emulsions under different CNCs concentrations. Larger emulsion
207
droplets were much less stable under low shear by mixing compared with the smaller droplets
208
formed at higher nanoparticle concentrations. In the case of 0.05 w/w % CNCs, the volume mean
209
particle diameters were about 6.9 μm. When the concentration of CNCs was up to 0.2 w/w%, the
210
volume mean particle diameters were 4.2 μm. It was obvious that the concentration of CNCs had a
211
significant impact on the particle size of emulsions.
pt
ed
M
an
us
cr
ip t
200
3.3. Stability of emulsions to creaming under different temperature ionic, concentration and pH.
213
Pickering emulsions were formulated at a 10/90 ratio (o/w), with 0.2w/w % suspension of CNCs
214
in the water phase. And then the prepared emulsions were used to test emulsion to creaming stability
215
under different factors, i.e. temperature, pH and ionic concentration. Temperature is an important
216
factor that could influence the stability of Pickering emulsions. Ionic strength and pH also play
217
important roles in regulating the electrostatic interactions that may occur between adjacent
218
nanoparticles at the oil–water interface (Dyab, Binks, & Fletcher, 2012; Zoppe, Venditti, & Rojas,
219
2012). Therefore, temperature, salt concentration, pH and tests were then carried out to specify
220
stability characteristics.
Ac ce
212
9
Page 9 of 25
The influence of temperature on the behaviour of CNCs-stabilized emulsions was explored by
222
applying samples under different temperature from 20-70 ℃ as shown in Fig.7. Results showed that
223
the emulsions were heated from 20-70 ℃, the stability of emulsions to creaming increased. It could
224
be explained by that CNCs were irreversibly adsorbed to cover emulsions droplets and formed 2 D-
225
network, when temperature raised, CNCs tended to give a stronger structure, as a result, the inter-
226
particle distances were decreased and the associative forces among the nanocrystals were enhanced
227
(Tzoumaki et al., 2009a, b, 2011). Thus, the stability of emulsions to creaming was enhanced.
228
Besides, it was also indicated that the increase in the stability of emulsions became higher with
229
increasing CNCs concentration. That was because under the condition of low CNCs concentration,
230
the inter-particle of CNCs had weak associative forces, so CNCs formed a weak droplet network.
231
When CNCs concentration increased, the particle could form a stronger network, as a result, the
232
stability of emulsions increased.
M
an
us
cr
ip t
221
Effects of pH and ionic concentration on stabilizing emulsions were examined by measuring the
234
zeta-potential of the droplets. The dependence of zeta-potential of the droplets in 0.2w/w % CNCs
235
stabilized emulsion on the ionic environment (NaCl concentration) were similar to that of the 0.2
236
wt% CNCs suspension, and the results are showed in Fig.8 A and Fig.8 C. Aqueous solutions with 0,
237
20, 40, 60, 80 and 100 mM NaCl concentrations were added to aqueous dispersions of CNCs and
238
emulsions. As the salt concentration increased, there was a greater tendency for CNCs to aggregate,
239
since the stabilizing repulsive forces were weakened by the increase in electrolyte concentration
240
(Tzoumaki et al., 2011). When NaCl concentrations reached 100mM, a small amount of CNCs
241
aggregated in the CNCs suspension, and zeta potential of CNCs suspension and emulsion entered the
242
domain of low-charged surfaces below 20mV. It was observed that the absolute value of the zeta-
243
potential of the emulsion droplets across the whole ranges of NaCl concentration remained relatively
244
lower than those of the CNCs particles. The low net charge on the droplets could be explained in
245
terms of the relatively low surface load of CNCs and/or because of ionic impurities in the samples
Ac ce
pt
ed
233
10
Page 10 of 25
that adsorbed to the surface of the oil droplets and changed their electrical charge (Surh, Decker, &
247
McClements, 2006). The magnitude of zeta-potential of the emulsion droplets decreased with
248
increasing the concentration of NaCl from 46.3 mV with no added salt to16.7 mV at 100 mM NaCl
249
due to the electrostatic screening of the charged (-COO-) groups by the counter-ions (Na+). It is
250
conventionally recognized that the larger the absolute magnitude of zeta-potential, the greater is the
251
electrostatic repulsion between droplets, and therefore the better the stability to their surface charge.
252
Similar responses to those obtained by varying the ionic strength were also noted by changing the pH.
253
As can be seen in Fig.8B and Fig.8D, the trends of zeta-potential of the droplets in 0.2w/w % CNCs
254
stabilized emulsion on different pH environment were similar to that of the 0.2wt% CNCs
255
suspension, and the zeta potential of CNCs suspension and emulsion were both increased when pH
256
increasing. The 0.20 w/w % CNCs-stabilized emulsion volume was increased with increasing of pH
257
value from 4.2 to 7.8. The magnitude of zeta-potential increased when 0.1 M NaOH was added, and
258
the zeta-potential of the emulsion droplets and CNCs remained negative at all pH values, possibly
259
because of the negatively charged -COO- groups on CNC particles. The CNCs has no cationic
260
groups and therefore did not become positively charged at any pH. An increase in pH from 4.2 to 7.8
261
caused an increase in the magnitude of negative charge on the droplets from -32.9 to -46.3 mV,
262
which was probably due to deprotonation of some of the protonated carboxyl groups (COOH→
263
COO- + H+) (Winuprasith & Suphantharika, 2013). It is conventionally recognized that increasing of
264
the surface charge can improve the stability of emulsions because of the increasing repulsive forces
265
between droplets against flocculation and coalescence. In our study, when the pH of the system was
266
increased from 4.2 to 7.8, the surface charge decreased from -42.9mV to -54.5 mV. So the larger the
267
absolute magnitude of zeta-potential, the greater is the electrostatic repulsion between droplets, and
268
therefore the better the stability (Dickinson, 2009).
Ac ce
pt
ed
M
an
us
cr
ip t
246
269
4. Conclusion
270
In this study, cellulose nanocrystals were prepared from corncob cellulose by one step with 11
Page 11 of 25
ammonium persulfate hydrolysis. The degree of cellulose crystallinity increased during ammonium
272
persulfate hydrolysis, which could be explained by the degradation of the amorphous cellulose
273
during reaction process. Besides, the hydrolysis conditions used led to obtain stable aqueous
274
suspensions of CNC which are negatively charged, due to the presence of carboxyl groups. D-
275
limonene Pickering emulsions were prepared by CNCs at high sonication power and long time with
276
0.2% (w/w) CNCs exhibited satisfactory size and high stability to creaming. Furthermore, the
277
stability of the CNCs-stabilized emulsions was decreased at low pH or high salt concentration due to
278
the electrostatic screening effect. In addition, the stability of emulsion increased when temperature
279
increased from 20 ℃ to 70 ℃, during the thermal treatment. The possible reason may be that CNCs
280
were irreversibly adsorbed to cover emulsions droplets and formed 2D-network. Thus, CNCs from
281
corncob cellulose are useful stabilizer for preparing D-limonene Pickering emulsion.
an
us
cr
ip t
271
Acknowlegment
ed
283
M
282
This study was supported by the National Natural Science Foundation of China (Grant No.
285
20806005) and the National High Technology Research and Development Program of China (863
286
Program, Grant No. 2012AA021403).
288
Ac ce
287
pt
284
289 290 291 292 12
Page 12 of 25
293
References
294
Aveyard, R., Binks, B. P., & Clint, J. H. (2003). Emulsions stabilised solely by colloidal particles.
297 298 299
Binks, B. P. (2002). Particles as surfactants—similarities and differences. Current Opinion in Colloid
ip t
296
Advances in Colloid and Interface Science, 100, 503-546.
& Interface Science, 7(1), 21-41.
Binks, B. P., & Horozov, T. S. (2006). Colloidal particles at liquid interfaces. Cambridge University
cr
295
In Press.
Boluk, Y., Lahiji, R., Zhao, L., & McDermott, M. T. (2011). Suspension viscosities and shape
301
parameter of cellulose nanocrystals (CNC). Colloids and Surfaces A: Physicochemical and
302
Engineering Aspects, 377(1), 297-303.
an
us
300
Borysiak, S., & Garbarczyk, J. (2003). Applying the WAXS method to estimate the supermolecular
304
structure of cellulose fibres after mercerisation. Fibres and Textiles in Eastern Europe, 11(5),
305
104-106.
ed
M
303
Bouyer, E., Mekhloufi, G., Rosilio, V., Grossiord, J. L., & Agnely, F. (2012). Proteins,
307
polysaccharides, and their complexes used as stabilizers for emulsions: Alternatives to synthetic
308
surfactants in the pharmaceutical field?. International Journal of Pharmaceutics.
Ac ce
pt
306
309
Chen, J., Vogel, R., Werner, S., Heinrich, G., Clausse, D., & Dutschk, V. (2011). Influence of the
310
particle type on the rheological behavior of Pickering emulsions. Colloids and Surfaces A:
311
Physicochemical and Engineering Aspects, 382(1), 238-245.
312 313 314 315
Dickinson, E. (2009). Hydrocolloids as emulsifiers and emulsion stabilizers. Food Hydrocolloids, 23(6), 1473-1482. Dyab, A. K., Binks, B. P., & Fletcher, P. D. (2012). Destabilisation of Pickering emulsions using pH. Colloids and Surfaces A: Physicochemical and Engineering Aspects.
13
Page 13 of 25
316
Flauzino Neto, W. P., Silvério, H. A., Dantas, N. O., & Pasquini, D. (2013). Extraction and
317
characterization of cellulose nanocrystals from agro-industrial residue–Soy hulls. Industrial
318
Crops and Products, 42, 480-488.
322 323 324
ip t
Kalashnikova, I., Bizot, H., Cathala, B., & Capron, I. (2011a). Modulation of cellulose nanocrystals
cr
321
assembly of cellulose nanofibrils into highly crystalline fibrous materials. RSC Adv.
amphiphilic properties to stabilize oil/water interface. Biomacromolecules, 13(1), 267-275.
us
320
Jiang, F., Han, S., & Hsieh, Y. L. (2013). Controlled defibrillation of rice straw cellulose and self-
Kalashnikova, I., Bizot, H., Cathala, B., & Capron, I. (2011b). New Pickering emulsions stabilized by bacterial cellulose nanocrystals. Langmuir, 27(12), 7471-7479.
an
319
Leung, A. C., Hrapovic, S., Lam, E., Liu, Y., Male, K. B., Mahmoud, K. A., & Luong, J. H. (2011).
326
Characteristics and Properties of Carboxylated Cellulose Nanocrystals Prepared from a Novel
327
One-Step Procedure. Small, 7(3), 302-305.
ed
M
325
Li, P. H., & Chiang, B. H. (2012). Process optimization and stability of D-limonene-in-water
329
nanoemulsions prepared by ultrasonic emulsification using response surface methodology.
330
Ultrasonics sonochemistry, 19(1), 192-197.
Ac ce
pt
328
331
Marku, D., Wahlgren, M., Rayner, M., Sjöö, M., & Timgren, A. (2012). Characterization of starch
332
Pickering emulsions for potential applications in topical formulations. International Journal of
333
Pharmaceutics.
334
Sheltami, R. M., Abdullah, I., Ahmad, I., Dufresne, A., & Kargarzadeh, H. (2012). Extraction of
335
cellulose nanocrystals from mengkuang leaves (Pandanus tectorius). Carbohydrate Polymers,
336
88(2), 772-779.
337 338
Surh, J., Decker, E. A., & McClements, D. J. (2006). Properties and stability of oil-in-water emulsions stabilized by fish gelatin. Food Hydrocolloids, 20(5), 596-606.
14
Page 14 of 25
339 340 341 342
Tzoumaki, M. V., Moschakis, T., & Biliaderis, C. G. (2009). Metastability of nematic gels made of aqueous chitin nanocrystal dispersions. Biomacromolecules, 11(1), 175-181. Tzoumaki, M. V., Moschakis, T., Kiosseoglou, V., & Biliaderis, C. G. (2011). Oil-in-water emulsions stabilized by chitin nanocrystal particles. Food hydrocolloids, 25(6), 1521-1529. Winuprasith, T., & Suphantharika, M. (2013). Microfibrillated cellulose from mangosteen (Garcinia
344
mangostana L.) rind: Preparation, characterization, and evaluation as an emulsion stabilizer.
345
Food Hydrocolloids.
cr
ip t
343
Zoppe, J. O., Venditti, R. A., & Rojas, O. J. (2012). Pickering emulsions stabilized by cellulose
347
nanocrystals grafted with thermo-responsive polymer brushes. Journal of colloid and interface
348
science, 369(1), 202-209.
an
us
346
349
M
350
354 355 356 357 358
pt
353
Ac ce
352
ed
351
359 360 361 362 15
Page 15 of 25
363 364
Figure Captions
365
Fig.1. Size distribution of CNCS by volume and CNCs suspension (0.2wt %) after treatment with 1
367
M APS.
ip t
366
368
Fig.2. X-ray diffraction patterns for CC and CNCS.
cr
369 370
Fig.3. FTIR spectra of CC and CNCs.
372 373
Fig.4. (A) TG of CC and CNCs; (B) DTG of CC and CNCs.
an
374
us
371
Fig.5. Optical microscopy images of D-limonene Pickering emulsions prepared with 0.05% (A) and
376
0.20% (B) CNCs (oil to water ratio of 1:9).
377 378
M
375
Fig.6. Emulsion droplet with different CNCs concentration.
ed
379
Fig.7. Effect of temperature on volume of emulsion with (A) 0.2w/w% CNCs and (B) 0.1w/w%
381
CNCs.
pt
380
382
Fig.8. Effect of the ionic strength (with NaCl) and pH (with NaOH) on CNCs suspension and D-
384
limonene Pickering emulsions zeta potential and emulsion volume with increasing ionic
385
concentration and pH.
386 387 388 389 390 391 392 393 394 395 396
Ac ce
383
16
Page 16 of 25
20
ip t
10
cr
Intensity (%)
15
0 300
600
900
1200
1500
1800
2100
2400
2700
3000
an
0
us
5
Size (d.nm)
M
397 398 399
Ac ce
pt
ed
Fig 1
17
Page 17 of 25
400 401 402
22.5
2000
403 404
408
1000
ip t
407
CNC
16.4
cr
406
1500
Intensity count
405
34 500
410
0
CC 0
10
412
20
an
411 30
40
50
60
70
80
90
Scattering angel (degrees of 2 theta)
M
413 414
pt
ed
Fig 2
Ac ce
415
us
409
18
Page 18 of 25
415 110
416 100
417 90
418
423 424 425 426
ip t
60
CNC
50
cr
422
1735
70
40 30 20 10 4000
3500
3000
2500
2000
us
421
CC
1500
1000
an
420
80
Transmitance (a.u.)
419
500
0
-1
Wavenumbers (cm )
427
M
428 429
Fig 3
Ac ce
pt
ed
430
19
Page 19 of 25
CC CNC
A
100
100
80
442 443
40
cr
441
60
60
40
20
20
0
444
0 0
200
400
600
0.002
B
449
455 456 457 458
ed
-0.004
-0.006
-0.008
0.000
-0.005
pt
Derivated weight of CC (mg/min)
454
-0.002
Ac ce
453
0.000
-0.010
-0.010
299
-0.012
0
200
-0.015
323 400
600
800
Derivated weight of CNC (mg/min)
452
CC CNC
M
447
451
1000
Temperature (℃)
446
450
800
an
445
448
us
440
ip t
80
Weight loss of CNC (%)
431 432 433 434 435 436 437 438 439
.
Weight loss of CC (%)
430
1000
Temperature (℃)
Fig 4
459 460
20
Page 20 of 25
460 461 462
A A
20μm
20μm
BB
463 464
ip t
465 466
cr
467 468
us
469 470 471
an
Fig 5
Ac ce
pt
ed
M
472
21
Page 21 of 25
472 473
0.1% CNCs 0.2% CNCs 0.05% CNCs
16
ip t
14
cr
10 8 6
us
Volume (%)
12
2 0 -2
0
2
4
6
8
an
4
10
12
14
16
18
20
M
Droplet diameter (µm) 474 475
ed pt Ac ce
476
Fig 6
22
Page 22 of 25
476 0.1% CNCs 0.2% CNCs
477
A
478
483 484 485 486
B
335
330 20
30
40
50
60
Temperature (℃ )
488
Fig 7
489
M
490
pt
494
Ac ce
493
ed
491 492
an
487
70
ip t
482
340
cr
481
345
us
480
Volume of emulsions (μL)
479
350
23
Page 23 of 25
494 495
CNCs Pickering emulsion
-38
499 500
-42
-25 -30 -35 -40
-44 -46 -48 -50 -52 -54
-45
501
-56
-50 0
20
502
40
60
80
100
4.0
500 480
514 515
400
-35
380 360 340
-45
6.5
7.0
7.5
us
M
-30
ed
Zeta potential (mV)
420
8.0
zeta potential of emulsion volume of emulsion 460
-42
440 420
-44
400
-46
380 -48
360 340
-50
320 -52
300 280
-54
320
-50 0
20
40
pt
513
440
60
80
100
260 -56 4.0
Ionic concentration (mmol/L)
Ac ce
512
-25
-40
6.0
Volume of emulsions (μL）
511
460
Volume of emulsions (µL)
510
-20
Zeta potential of emulsions (mV)
D
-15
506 C
5.5
an
zeta potential volume of emulsions
505
509
5.0
pH
504
508
4.5
Ionic concentration (mmol/L)
503
507
cr
498
-40
B
-20
Zeta potential (mV)
497
-15
ip t
A Zeta potential (mV)
496
Pickering emulsion CNCs
-10
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
pH
Fig 8
24
Page 24 of 25
515
Highlights
516
> Cellulose nanocrystals were successfully prepared by ammonium persulfate hydrolysis.>
518
D-liomonene Pickering emulsions prepared by cellulose nanocrystals exhibited high
519
stability.> Electrostatic interactions and temperature had obvious effect on emulsion
520
stability to creaming.
cr
ip t
517
Ac ce
pt
ed
M
an
us
521
25
Page 25 of 25
