- Email: [email protected]
Emulsifying properties of succinylated arabinoxylan-protein gum produced from corn ethanol residuals
Accepted Manuscript Emulsifying Properties of Succinylated Arabinoxylan-Protein Gum Produced from Corn Ethanol Residuals Zhouyang Xiang, Troy Runge PII:
S0268-005X(15)30028-X
DOI:
10.1016/j.foodhyd.2015.07.018
Reference:
FOOHYD 3074
To appear in:
Food Hydrocolloids
Received Date: 25 April 2015 Revised Date:
15 July 2015
Accepted Date: 17 July 2015
Please cite this article as: Xiang, Z., Runge, T., Emulsifying Properties of Succinylated ArabinoxylanProtein Gum Produced from Corn Ethanol Residuals, Food Hydrocolloids (2015), doi: 10.1016/ j.foodhyd.2015.07.018. 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.
ACCEPTED MANUSCRIPT
Particle size (nm) 2200
RI PT
Arabinoxylan-Protein Gum (APG) from Distillers’ Grains (DG)
AC C
EP
TE D
M AN U
SC
1800 1400
pH=3.5 pH=5.5 pH=6.5
1000 600 0
0.04
0.08 0.12 Degree of Substitution
ACCEPTED MANUSCRIPT
Emulsifying Properties of Succinylated Arabinoxylan-Protein Gum
2
Produced from Corn Ethanol Residuals
3
Zhouyang Xiang and Troy Runge*
4
Department of Biological Systems Engineering, University of Wisconsin-Madison, Madison, WI
5
53706, United States
6
*Corresponding author:
7
Abstract
SC
RI PT
1
M AN U
Email: [email protected]; Tel: 1-608-890-3143
This study investigated the possibilities of making valuable products from corn ethanol
9
byproducts and providing the beverage industries more variety of high quality emulsifiers other
10
than gum arabic. An arabinoxylan-protein gum (APG) was extracted from distillers’ grains (DG),
11
a low-value corn ethanol byproduct, and modified through acylation with succinic anhydride. The
12
effects of pH and degree of substitution (DS) on the emulsifying properties of succinylated APG,
13
referred to as SAPG, were investigated. Emulsion particle size and stability of APG and gum
14
arabic were comparable at pH 3.5-6.5. Succinylation could enhance the emulsifying properties of
15
APG. Compared to gum arabic, at pH < 5, SAPG emulsions had larger particle size but
16
comparable stability, whereas at pH > 5, SAPG had much smaller particle size and better stability
17
than gum arabic. The results suggested that SAPG, compared to gum arabic, could be a
18
comparable emulsifier at low pH values and a better emulsifier at neutral pH values.
19
Keywords: arabinoxylan-protein gum; degree of substitution; distillers’ grains; gum arabic;
20
oil-in-water emulsions; particle size; emulsion stability; viscosity
AC C
EP
TE D
8
1
ACCEPTED MANUSCRIPT
21
1. Introduction Oil-in-water emulsions are common for many beverages with the oil droplets dispersed in the
23
beverage’s aqueous phase serving primarily as flavor oils (Chanamai & McClements, 2002; Given,
24
2009; Qian, Decker, Xiao, & McClements, 2011). Some non-flavor oils, such as omega-3 fatty
25
acids, are also sometimes included to provide special nutrients (Given, 2009). Different from other
26
food emulsions, beverage emulsions are very dilute (~0.2 wt% oil) and should have a great
27
stability for a long period of time (Given, 2009; Yadav, Johnston, Hotchkiss, & Hicks, 2007). To
28
disperse oil droplets in the aqueous phase, emulsifiers are needed to both facilitate the emulsion
29
formation and stabilize the emulsion preventing the droplets from aggregating (Walstra, 1993).
M AN U
SC
RI PT
22
The emulsifiers used in beverage emulsions are primarily amphiphilic polysaccharides due to
31
their excellent emulsion stabilizing ability under changing environmental conditions (Chanamai &
32
McClements, 2002; Garti, 1999; Qian et al., 2011). Gum arabic is the most commonly used in
33
beverages due to its bio-based feature, food safety, high water solubility, low solution viscosity and
34
great emulsion stability (Chanamai & Mcclements, 2002; Yadav et al., 2007). Gum arabic is a
35
polysaccharides-protein complex with arabinogalactan as the main polysaccharides units; during
36
emulsifying process, the hydrophobic protein portions anchor into the oil droplets, while the
37
hydrophilic polysaccharides portions stick out into the aqueous phase, preventing the droplets
38
from aggregation by steric or electrostatic effects (Chanamai & Mcclements, 2002; Islam, Phillips,
39
Sljivo, Snowden, & Williams, 1997; Jayme, Dunstan, & Gee, 1999; Randall, Phillips, & Williams,
40
1988; Walstra, 1993; Xiang, Anthony, Tobimatsu, & Runge, 2014a). Gum arabic is an effective
41
emulsifier for beverage; however, the plants (Acacia senegal) producing gum arabic are primarily
42
grown in the Sahelian region of Africa, and thus its supply is limited and price is sensitive to
43
climate changes and is comparatively high to other food ingredients (Yadav et al., 2007).
AC C
EP
TE D
30
2
ACCEPTED MANUSCRIPT
As the demand for beverages and thus beverage emulsifiers increases, many studies have been
45
conducted to find substitutes for gum arabic. Modified starch is common replacement beverage
46
emulsifier for gum arabic (Chanamai & McClements, 2002). Hydrophobic or non-polar groups are
47
introduced onto starch through its abundant hydroxyls imparting it with both hydrophobic and
48
hydrophilic portions similar to gum arabic; one of the examples is octenyl succinic acid (OSA)
49
starch (Caldwell & Wurzburg, 1952; Chanamai & McClements, 2002; Viswanathan, 1999). Gum
50
arabic has also been modified with octenyl succinic acid to increase the amount of its hydrophobic
51
portion and reduce its usage level (Ward, 2002). Globular proteins, such as whey protein, are able
52
to create oil droplets with size smaller than gum arabic during homogenization; however their
53
stability is vulnerable to environmental condition changes (pH, temperature, etc.) (Chanamai &
54
McClements, 2002; McClements, 1999; Qian et al., 2011). More recently, an arabinoxylan-protein
55
gum (APG) extracted from corn ethanol byproducts was found with promising emulsifying
56
properties (Xiang et al., 2014a; Yadav, Johnston, & Hicks, 2009; Yadav et al., 2007).
TE D
M AN U
SC
RI PT
44
The U.S. produced approximately 56 billion liters of corn ethanol in 2014 yielding huge
58
amount of fiber-protein rich byproducts (e.g. yielding ~40 million metric tons of distillers’ grains)
59
(RFA, 2015). Wet mill and dry grind are the two major processes in corn ethanol production
60
industry in the US, between which dry grind is the most predominant process (Bothast &
61
Schlicher, 2005; RFA, 2015). Many of the corn ethanol byproducts, such as corn fiber and
62
distillers’ grains (DG), from wet mill and dry grind processes respectively, are sold as animal
63
feed at a very low price (Xiang & Runge, 2014). Therefore, effectively utilizing the proteins and
64
polysaccharides in corn ethanol byproducts to make more valuable products could potentially
65
increase corn ethanol profitability and sustainability (Xiang & Runge, 2014; Xiang et al., 2014a;
66
Xiang, Watson, Tobimatsu, & Runge, 2014b).
AC C
EP
57
3
ACCEPTED MANUSCRIPT
Recently, an arabinoxylan-protein gum (APG) was extracted by alkali and/or H2O2 from corn
68
fibers (Yadav et al., 2007, 2009) or distillers’ grains (Xiang et al., 2014a). This gum is also an
69
amphiphilic polysaccharide, similar to gum arabic (an arabinogalactan-protein complex). The
70
structural differences between APG and gum arabic might be reflected by their chemical
71
compositions as shown in Table 1. Other than the amphiphilic polysaccharide feature, the
72
comparable molecular weight of APG with gum arabic (Table 1) also suggests the potential
73
application of APG as beverage emulsifiers (Dickinson, Galazka, & Anderson, 1991).
74
Arabinoxylan is a heterogeneous polymer and a major type of hemicelluloses found in corn
75
(Gáspár, Kalman, & Reczey, 2007; Xiang et al., 2014b), with the backbone of
76
β-(1-4)-D-xylopyranose being substituted primarily of α-L-arabinofuranose and slightly of
77
D-galactopyranose
78
corn, some of the hydroxyproline-rich glycoproteins are believed to link to the arabinoxylan
79
through covalent bonds that were stable in alkali or H2O2 (Saulnier, Marot, Chanliaud, & Thibault,
80
1995; Xiang et al., 2014a; Yadav et al., 2007). Thus, some protein could be co-extracted with
81
arabinoxylan as an arabinoxylan-protein complex and not be removed by the physical purification
82
process (Xiang et al., 2014a).
M AN U
SC
RI PT
67
EP
TE D
and D-glucuronic acid (Ebringerová & Heinze, 2000; Xiang et al., 2014b). In
The corn fiber APG has been shown to have better emulsifying properties than regular or
84
modified gum arabic (Yadav et al., 2007, 2009). The emulsifying properties of the DG APG were
85
only comparable to gum arabic (Xiang et al., 2014a), and thus could be further improved through
86
chemical modifications. Gum arabic or starch is typically modified with octenyl succinic acid to
87
increase the amount of its hydrophobic portion (Ward, 2002). However, as shown in Table 1, the
88
DG APG has protein content more than 10% (Xiang et al., 2014a), much higher than that of gum
89
arabic (~1%) and corn fiber APG (1-5%) (Yadav et al., 2007, 2009). Additionally, gum arabic has
AC C
83
4
ACCEPTED MANUSCRIPT
10-20% uronic acid (Islam, 1997; Randall et al., 1988) compared to ~5% in DG APG (Xiang et al.,
91
2014a). Thus, it is hypothesized that chemically modifying the DG APG hydrophilic arabinoxylan
92
portion with more steric or electrostatic effects might be an effective way to improve its
93
emulsifying properties.
RI PT
90
Acylation of polysaccharides with acid anhydrides is an effectively way to impart
95
polysaccharides with carboxylic acid groups. High degree of substitution (DS) of acylation of
96
polysaccharides with acid anhydrides is usually conducted in N,N-dimethyl formamide
97
(DMF)/LiCl solutions with or without catalysts (Ebringerová et al., 2005; Lindblad & Albertsson,
98
2005; Sun, Sun, & Zhang, 2001). More recently, ionic liquids have also been used as an effective
99
solvent (Hansen & Plackett, 2011; Peng, Ren, Zhong, & Sun, 2011a; Peng, Ren, & Sun, 2010;
100
Ren, Peng, Feng, & Sun, 2013). Additionally, low DS acylations with acid anhydride were
101
reported on starch and hemicelluloses by using alkaline solution as solvents (Betancur-Ancona,
102
Garcia-Cervera, Canizares-Hernandez, & Chel-Guerrero, 2002; Jeon, Viswanathan, & Gross,
103
1999; Sun, Sun, & Bing, 2002). Avoiding the use of the organic solvents or ionic liquid would
104
both economically and environmentally benefit the application of modified DG gum in food
105
industries.
EP
TE D
M AN U
SC
94
Therefore, to find suitable substitutes for gum arabic to relieve its limited supply, and to
107
improve the value from the corn ethanol byproducts, we investigated acylating the
108
arabinoxylan-protein gum (APG) from DG by succinic anhydride. The effects of DS and pH on the
109
emulsifying properties of modified DG APG were quantified and compared to GA.
AC C
106
110
5
ACCEPTED MANUSCRIPT
111
2. Materials and methods
112
2.1. Fractionation of APG Distillers’ grains (DG) samples were obtained from Didion Milling Inc. (Cambria, WI,
114
USA). The arabinoxylan-protein gum (APG) was extracted and purified according to the
115
procedure described previously (Xiang et al., 2014a). The DG was pre-extracted by acetone for
116
12h in a Soxhlet extractor to remove fats and organic impurities and then was reacted in 3%
117
(w/v) NaOH solution at 50 °C at 1:10 solid to liquid ratio (w/v) for 3h. The separated
118
alkali-soluble part was adjusted to pH 5.5, purified with bentonite clays to remove soluble
119
proteins, and slowly poured into three volume of 95% ethanol with constant stirring. The
120
precipitated solid was washed several times with 70% ethanol, freeze-dried, and ground to give a
121
powder form of APG.
M AN U
SC
RI PT
113
The extracted APG consisted of (based on the oven-dried weight) 19% arabinan, 28% xylan,
123
5% galactan, 6% glucan, 2% mannan, 5% uronic acid, and 12% crude protein (Table 1). Number
124
average molecular weight (Mn) of the APG was 62 kDa, weight average molecular weight (Mw)
125
was 450 kDa, and the polydispersity index (Mw/Mn) was 7.2 (Table 1). The compositions and
126
molecular weights of the APG were consistent with our previous studies, which provide a
127
detailed characterization (Xiang et al., 2014a, b).
128
2.2. Succinylation of APG
AC C
EP
TE D
122
129
APG was acylated by succinic anhydride (SA) based on the procedure modified from Sun et
130
al. (2002). The unmodified APG was dissolved in 1% (w/v) NaOH solution to make a 5% (w/v)
131
solution. Different amounts of SA (99.9%, Chem-impex Int’l Inc., IL, USA) leading to
132
SA/arabinoxylose unit (AXU) mole ratios of 0.25, 0.33, 0.50 and 0.75 were added. The
133
abbreviation AXU represents arabinoxylan mono sugar unit, which has a molar mass of 100
6
ACCEPTED MANUSCRIPT
g/mol assuming the arabinoxylan consists of only arabinose and xylose since the content of other
135
minor units is very small (Xiang et al., 2014a, b). The mixture was adjusted to pH 8.5-9.0, and
136
maintained at 30 °C with stirring for 1h. The succinic anhydride modified DG
137
arabinoxylan-protein gum (SAPG) was precipitated by pouring the mixture into three volumes of
138
95% ethanol and was washed twice with 70% ethanol solution and once with acetone in order to
139
remove the unreacted SA.
140
2.3. Characterization of succinylated APG
SC
RI PT
134
The degree of substitution (DS) of a SAPG sample was defined as the mole of SA reacted
142
per mole of AXU. The samples were coded as SAPGn, where the number n denotes the DS.
143
The DS of SAPG was determined by measuring the amount of carboxylic acid group using a
144
procedure modified from Stojanović, Jeremic, Jovanovic, & Lechner (2005). A weighed SAPG
145
sample of about 0.125 g was completely dissolved in 5 ml 0.2 M NaOH solution, to which 75 ml
146
DI water was added. The excess NaOH was titrated with standard 0.05 M H2SO4 until a 3.75 pH
147
was achieved. A blank sample was also titrated. The mole of COOH in the SAX sample was
148
calculated as:
TE D
nCOOH = (Vb − V ) ⋅ 2 ⋅ cH 2SO4
EP
149
M AN U
141
where Vb is the volume (ml) of H2SO4 used for the titration of the blank; V is the volume (ml) of
151
H2SO4 used for the titration of the sample; cH2SO4 is the mole concentration (mol/ml) of H2SO4
152
used for the titration. Using the measured moles of COOH, the DS can be calculated as:
153
AC C
150
DS =
132 × nCOOH ms − 100 × nCOOH
154
where 132 is the molar mass (g/mol) of an AXU; 100 is the increase in molar mass (g/mol) of
155
an AXU for each SA substituted; ms is the oven dry mass (g) of the SAPG sample used for
156
titration; nCOOH is the mole of COOH. 7
ACCEPTED MANUSCRIPT
157
13
C NMR spectra of APG and SAPG samples were acquired on a Bruker Biospin (Billerica,
MA, USA) AVANCE 700 MHz spectrometer fitted with a cryogenically-cooled 5-mm TXI
159
gradient probe with inverse geometry (proton coils closest to the sample). Finely ball-milled
160
samples (~50 mg) were swelled in D2O (1 mL). The spectra were recorded after 8000 normal
161
scans and processed using Topspin 3.1 software (Bruker, Billerica, MA, USA).
RI PT
158
Fourier transform infrared (FT-IR) spectra were obtained using a total of 10 scans using a
163
Spectrum 100 FT-IR spectrometer (Perkin Elmer, Waltham, MA, USA) equipped with a
164
universal ATR sampling accessory.
165
2.4. Emulsion preparation and emulsifying properties
M AN U
SC
162
APG, SAPG or a gum arabic sample (MP Biomedicals, Santa Ana, CA, USA) were dissolved
167
(5 wt%) in the aqueous solution by using a sonicator for 30 min. The gum solution was then diluted
168
50 times into a 0.1 wt% gum solution and adjusted to pH = 3.5, 4.5, 5.5 or 6.5 by 5 wt% or 0.5 wt%
169
citric acid. The gum solutions were homogenized with 0.5 wt% orange oil (Now Foods,
170
Bloomingdale, IL, USA) at a high shear (20000 rpm) in a Waring 700s Blender for 2 min and 2
171
times. The solutions were allowed to sit 5 min between the 2 blendings.
TE D
166
The surface-area-average diameter (droplet size, d32) and zeta potential of the emulsions were
173
measured right after preparations by a particle size analyzer (90Plus, Brookhaven, Long Island,
174
NY, USA) according to the procedure described by Sun & Gunasekaran (2009). Cream stability
175
test was conducted over a range of 8 days according to the procedure described by
176
Keowmaneechai & McClements (2002). A creaming index (CI) from this procedure was recorded
177
each day, which was calculated and as:
AC C
EP
172
178 179
where HT is the total emulsion height and Hs is the serum layer height. 8
ACCEPTED MANUSCRIPT
The rheological properties of the gum solutions were investigated. The dynamic (absolute)
181
viscosity of the gum solutions was determined at 25 °C by using a rotational viscometer
182
(Fungilab Adv Series, New York, USA) at rotational speeds of 1, 2.5, 5, 10, 20, 50, 100 rpm.
183
Due to the viscometer’s limited capacity, samples may not be measured at all selected speeds.
184
3. Results and discussions
185
3.1. Characterization of succinylated APG
SC
RI PT
180
Acylation of arabinoxylan with succinic anhydride (SA) would couple carboxylic acids to
187
arabinoxylose units (AXU) through ester bonds (Figure 1). To reach a high degree of substitution
188
(DS) up to 1.0, the reaction should be conducted in DMF/LiCl with catalyst (Sun et al., 2001).
189
However, this process might not be economical and might introduce toxicity for the final food
190
grade products. Reaction of arabinoxylan with SA under mild alkaline condition (pH = 8.5-9.0)
191
have been reported to be capable of controlling the DS below 0.2 (Sun et al., 2002), which was a
192
more suitable modification method for this study.
TE D
193
M AN U
186
The succinylated DG arabinoxylan-protein gum (SAPG) was first characterized by FTIR The peak at ~1735 cm-1 of SAPG corresponded to the C=O stretching of the ester
(Figure 2).
195
bond or the carboxylic acid group, and the peak at ~1160 cm-1 should be related the C-O
196
stretching of the ester bond (Sun et al., 2001, 2002). The peak at ~~1580 cm-1 could include the
197
N-H stretching of protein (Jagadeesh et al., 2001; Xiang et al., 2014b) and the asymmetric stretch
198
of the carboxylates (Cabaniss, Reddy, & Rajulu, 1998; Peng et al., 2011a). Additionally, the
199
absence of signal between 1840 cm−1 and 1760 cm−1 indicated that the SAPG samples were free
200
of detectable amount of unreacted SA. The
201
consistent with the FTIR spectra. The peaks at 181.8, 178.0, and 176.3 ppm correspond to the
AC C
EP
194
13
C NMR spectra (Figure 3) of the SAPG were in
9
ACCEPTED MANUSCRIPT
202
carbons in the carbonyl of ester bonds and carboxylic acids confirming successful succinylation
203
on the APG to SAPG. The degree of substitution (DS) of the SAPG samples was varied using different SA/AXU
205
mole ratio for the reaction (Figure 4). The overall range of DS was between 0.02-0.12, which is
206
consistent with the previous study (Sun et al., 2002). The DS increased as SA/AXU mole ratio
207
increased, but when SA/AXU mole ratio went beyond 0.5, adding more SA did not have a
208
positive response for increasing DS. Similar phenomenon was found when acylating starch with
209
dodecenyl succinic anhydride in alkaline solution (Jeon et al., 1999). This may be due to
210
insufficient mixing of SA and APG or insufficient reaction time for high extents of acylation
211
under high SA concentration (Jeon et al., 1999; Sun et al., 2001). The low DS of the SAPG
212
samples agreed with the low intensity of the carbonyl peaks from the NMR spectrum (Figure 3).
213
The increasing intensities of the FTIR spectra peaks at ~1160 cm-1, ~1580 cm-1 and ~1735 cm-1
214
also agreed with the increasing DS of the analyzed SAPG samples (Figure 2).
215
3.2. Particle size and zeta potential
TE D
M AN U
SC
RI PT
204
Particle size (surface-area-average diameter d32) is a key measure of the quality of an
217
emulsion formed. During oil and water homogenization, three processes occur simultaneously:
218
(1) large oil droplets are torn into smaller droplets; (2) the hydrophobic portion of the emulsifiers
219
is absorbed onto the droplet surface; (3) oil droplets re-encounter each other, while the
220
hydrophilic portion of the emulsifiers, which sticks out into the aqueous phase, prevents the
221
droplets from aggregation through steric or electrostatic effects (Jayme et al., 1999; Walstra,
222
1993). Thus, oil droplet or particle size reflects both of how efficient the emulsifier is absorbed
223
onto the oil droplet surface and prevents the droplet from aggregation. Zeta potential reflects the
AC C
EP
216
10
ACCEPTED MANUSCRIPT
224
electrostatic repulsion between the adjacent droplet electrical double-layer and thus determines
225
the electrostatic contribution to the stabilization of the emulsions (Jayme et al., 1999). The particle size (d32) and zeta potential of oil-in-water emulsions stabilized by gum arabic,
227
APG and SAPG at different pH (3.5-6.5) was measured, shown in Figure 5. The particle size
228
measurements of gum arabic emulsions were constant at ~1400 nm among different pH (Figure
229
5a), which was consistent with previous studies (Qian et al., 2011). Emulsions stabilized by APG
230
also demonstrated constant particle size over different pH, and the values were similar to those of
231
gum arabic. These results suggested that APG had similar performance and likely would be
232
useful in similar applications such as in assisting the formation of beverage oil-in-water
233
emulsions, typically done under pH values ranging between 3.5 and 6.5.
M AN U
SC
RI PT
226
The emulsions stabilized by SAPG with a low DS (0.02) also showed relatively stable
235
particle size over different pH, but at a higher value of ~1800 nm, compared to gum arabic and
236
APG (Figure 5a). However, the particle size of emulsions stabilized by SAPG with a high DS
237
(0.09) decreased significantly with increased pH, from ~1800 nm at pH = 3.5 and 4.5 to ~1200
238
nm at pH = 5.5 and ~800 nm at pH = 6.5. This phenomenon might be explained by the zeta
239
potential changes of the emulsions (Figure 5b). Compared to APG and SAPG0.02 emulsions at
240
high pH (5.5 and 6.5), SAPG0.09 emulsion demonstrated significantly higher zeta potential
241
indicating higher electrostatic repulsions between oil droplets. Gum arabic emulsion had high
242
zeta potential at pH 6.5 too, but its particle size was constant at either high or low pH. The reason
243
is that steric effects might be more prevalent than electrostatic repulsions in stabilizing the gum
244
arabic emulsion (Jayme et al., 1999; Randall et al., 1988). However, for SAPG0.09 sample, as
245
the zeta potential increased significantly and the particle size decreased significantly at pH = 5.5
246
and 6.5, it was clear that the electrostatic repulsion contributed significantly in stabilizing the
AC C
EP
TE D
234
11
ACCEPTED MANUSCRIPT
SAPG0.09 emulsion. The electrostatic repulsion starting to take major effects at pH = 5.5 was
248
likely due to that the succinic acid had pKa1 = 4.2 and pKa2 = 5.4, and thus at pH = 5.5, the
249
carboxylic acid groups on the SAPG sample were greatly charged. It suggests that at pH > 5.5,
250
the carboxylic acid group introduced by succinylation increased the charges of APG and the
251
electrostatic repulsion, thus preventing oil droplets from aggregating.
RI PT
247
Since at high DS, SAPG had better performance in stabilizing the emulsions, the effects of
253
DS on the particle size of the SAPG emulsions were also plotted in Figure 6. At pH = 3.5, as DS
254
increased, the particle size slightly increased from ~1500 nm to ~1800 nm. While at pH = 5.5
255
and 6.5, the particle size increased slightly at DS = 0.02, and then decreased significantly at DS =
256
0.09. Continuing to increase the DS to 0.12 had limited effects on the particle size at pH = 5.5
257
and 6.5. It clearly indicated that at pH = 3.5 the carboxylic groups of SAPG had very limited
258
charges due to their relatively high pKa and thus did not contribute to the electrostatic repulsion
259
between oil droplets. When the DS increased beyond 0.09, it was postulated that adding more
260
carboxylic groups was not effective in stabilizing the emulsions.
TE D
M AN U
SC
252
In summary, at a pH range between 3.5 and 6.5, APG had similar performance in forming
262
emulsions compared to gum arabic. Succinylation imparted APG with better emulsion forming
263
abilities at high DS and pH > 5, but reduced its emulsion forming abilities at low DS and pH < 5.
264
3.3. Emulsion stability
AC C
EP
261
265
Emulsion stability is critical in many applications such as beverage emulsions, since many
266
of these products are require to be stored for months or even years (Given, 2009; Yadav et al.,
267
2007). Gum arabic is excellent in its emulsion stability but not in creating an initial small particle
268
size. Some globular proteins are able to make emulsions with very small particle size, but very
269
unstable (Chanamai & McClements, 2002; McClements, 1999; Qian et al., 2011).
12
ACCEPTED MANUSCRIPT
The cream index (CI) reflects the emulsion’s stability to creaming. The CI of the emulsions
271
was recorded over a period of 7 days for the gum arabic, APG and SAPG stabilized oil-in-water
272
emulsions at pH = 3.5 and 5.5 (Figure 7). At pH = 3.5, graphs of CI of SAPG0.09 and SAPG0.12
273
emulsions overlapped with the gum arabic emulsion, while the graphs of APG and SAPG0.02
274
fell below gum arabic. It shows that at low pH, APG emulsions had better stability than those of
275
gum arabic, and even though SAPG emulsions had larger initial particle size (Figure 6), and their
276
stability was still comparable to gum arabic. This phenomenon could possibly be explained by
277
that, even though larger particle size should induce faster creaming (Keowmaneechai and
278
McClements, 2002; McClements, 1999), the oil droplets with absorbed SAPG may have high
279
density due to the succinylation on APG and thus reduce the creaming rate (Sun et al., 2009). At
280
pH = 5.5, APG and gum arabic emulsions demonstrated very similar stability, while all the
281
SAPG sample, including SAPG0.02, which had the initial highest particle size among the all, had
282
stability better than gum arabic and APG. This could also possibly be explained by the higher
283
density of the oil droplets with absorbed SAPG compared to that with APG (Sun et al., 2009).
284
On the other hand, at high pH, it was possible that the carboxylic groups on SAPG were greatly
285
charged and the electrostatic repulsion between oil droplets was increased, hindering the
286
flocculation and thus slowing down the creaming (McClements, 1999). Additionally, the higher
287
molecular weight of APG than gum arabic (Table 1) may contribute to the APG or SAPG
288
emulsion stabilities, since larger protein-containing macromolecules might be more responsible
289
to the emulsion stability (Dickinson et al., 1991).
290
3.4. Viscosity of gum solutions
AC C
EP
TE D
M AN U
SC
RI PT
270
291
Viscosity is another characteristic that should be considered in evaluating beverage
292
emulsifiers. Beverage emulsifiers might require low solution viscosity and gum arabic is well
13
ACCEPTED MANUSCRIPT
293
known for its low solution viscosities even at high concentration (Chanamai & McClements,
294
2002; Yadav et al., 2007). In this study, the dynamic (absolute) viscosity of the gum solution was evaluated at 5 wt%,
296
2.5 wt% and 1.25 wt% concentrations, which was much higher than the concentration typically
297
used in the emulsions (0.1 wt%). These higher concentrations were selected to accentuate the
298
differences between the gums. Figure 8 shows the dynamic (absolute) viscosity changes with
299
shear rates by varying the rotational speed. At 5 wt% gum solution (Figure 8a), APG
300
demonstrated non-Newtonian fluid behavior, showing an apparent shear thinning phenomenon
301
(shear rate) (Nishiwaki-Akine and Watanabe, 2014). SAPG0.12 sample behaved as a Newtonian
302
fluid and had viscosity (~70 mPa·s) lower than that of APG (~90 mPa·s at high shear rate). At
303
2.5 wt% gum solution (Figure 8b), both APG and SAPG sample behaved as a Newtonian fluid
304
and showed a large drop in viscosity to ~16 and ~14 mPa·s, respectively. At 1.25 wt% gum
305
solution (data not shown in the graph), the viscosity of APG and SAPG were ~6 mPa·s. The
306
non-Newtonian behavior of the APG solution was similar to the native wheat straw
307
hemicelluloses (Peng, Ren, Zhong, Cao, & Sun, 2011b). The SAPG showing lower viscosity
308
than the APG was in consistent with previous studies that succinylated cellulose and
309
carboxymethyl hemicelluloses showed lower viscosity than their native forms (Li, Jin, Liu, Sun,
310
Zhang, & Kennedy, 2009; Peng et al., 2011b), which was probably due the weakening of the
311
intermolecular interaction between polysaccharide chains (Peng et al., 2011b). The significant
312
changes of viscosity with solution concentration for APG and SAPG indicated strong
313
hydrogen-bonding between their polymer chains or high entanglement at high concentrations
314
(Peng et al., 2011b).
AC C
EP
TE D
M AN U
SC
RI PT
295
14
ACCEPTED MANUSCRIPT
The dynamic viscosity of gum arabic solution was constant at ~5 mPa·s among different
316
studied concentrations, which was much lower than that of the APG and SAPG solutions at high
317
concentration (Figure 8), but approximately the same at concentrations < 1 wt%. These results
318
suggested that the solution viscosity of SAPG was successfully reduced compared to APG
319
making it more suitable to be used as a beverage emulsifier. Even though the solution viscosity
320
of SAPG was still higher than gum arabic at high concentration, at low usage level (below 1 wt%)
321
of emulsifiers, which is most likely the case for making beverages (Yadav et al., 2007), it might
322
not be necessary to worry about the viscosity since APGS, APG and gum arabic are all showing
323
similar values. At high concentrations, APG and SAPG could also be used as either a food
324
thickener or an emulsifier. For some food products that are required to be thick and be a
325
non-Newtonian fluid such as salad dressing (Tabilo-Munizaga and Barbosa-Ca´novas, 2005),
326
APG might be a more promising food additive than SAPG.
327
4. Conclusions
TE D
M AN U
SC
RI PT
315
Reacting the arabinoxylan-protein gum (APG) extracted from distillers’ grains (DG) with
329
succinic anhydride (SA) in mild alkaline solution, a succinylated APG (SAPG) was able to be
330
produced without using organic solvents while allowing control of the degree of substitution (DS)
331
to levels below 0.2. Between pH = 3.5 and 6.5, the unmodified APG was capable of creating
332
dilute oil-in-water emulsions with particle size and stability comparable to gum arabic, with even
333
better stability at pH = 3.5. SAPG emulsifying properties were more susceptible to pH. When pH
334
< 5, SAPG created oil-in-water emulsions with larger particle size but comparable stability;
335
however, at pH > 5, SAPG created emulsions with much smaller particle size and better stability,
336
compared to APG and gum arabic. Increasing the DS could further enhance the SAPG
337
emulsifying properties, but beyond DS = 0.09, the DS effects were not significant. It may be
AC C
EP
328
15
ACCEPTED MANUSCRIPT
postulated that the carboxylic acid groups introduced by the succinylation had pKa value of
339
about 5, and thus at high pH, they were greatly charged providing highly improved electrostatic
340
repulsions between oil droplets. This suggests that unmodified and succinylated APG can
341
potentially substitute gum arabic as an emulsifier for producing beverage oil-in-water emulsions;
342
while succinylated APG (SAPG), especially at high pH, was possibly a much better emulsifier
343
than gum arabic.
RI PT
338
The solution viscosity of APG was greatly reduced by succinylation, making SAPG a more
345
suitable beverage emulsifier than APG from a viscosity standpoint. However, the non-Newtonian
346
behavior of concentrated APG solution could possibly make it a food thickener especially for the
347
food products required to be a non-Newtonian fluid.
348 349
Acknowledgements
M AN U
SC
344
This work was supported by U.S. Department of Agriculture, under contract USDA Critical
351
Agricultural Material Grant (2013-38202-20400). The authors gratefully acknowledge Dr.
352
Sundaram Gunasekaran’s assistance with FTIR, particle size and zeta potential analyses, and
353
Didion Milling Inc. for materials and valuable discussions.
354
EP
TE D
350
Abbreviations
356
APG, arabinoxylan-protein gum; AXU, arabinoxylose unit; CI, creaming index; DG, distillers’
357
grains;
358
arabinoxylan-protein gum
AC C
355
DS,
degree
of
substitution;
SA,
359 360
16
succinic
anhydride;
SAPG,
succinylated
ACCEPTED MANUSCRIPT
361
References
362
Betancur-Ancona, D., Garcia-Cervera, E., Canizares-Hernandez, E., & Chel-Guerrero, L. (2002). Chemical
364
succinylation. Starch-Starke, 54, 540-546.
366 367
of
Jack
Bean
starch
by
ethanol. Applied Microbiology and Biotechnology, 67, 19-25.
Cabaniss, S. E., Leenheer, J. A., & McVey, I. F. (1998). Aqueous infrared carboxylate Aliphatic
di-acids.
Spectrochimica
369
Biomolecular Spectroscopy, 54, 449-458.
Acta
Part
a-Molecular
and
M AN U
absorbances:
371
ensiformis)
Bothast, R., & Schlicher, M. (2005). Biotechnological processes for conversion of corn into
368
370
(Canavalia
SC
365
modification
RI PT
363
Caldwell, C. G., & Wurzburg, O. B. (1953). Polysaccharide derivatives of substituted dicarboxylic acids. Patent: US 2661349 A, USA.
372
Chanamai, R., & McClements, D. J. (2002). Comparison of gum arabic, modified starch, and
373
whey protein isolate as emulsifiers: Influence of pH, CaCl2 and temperature. Journal of
374
Food Science, 67, 120-125.
Dickinson, E., Galazka, V. B., & Anderson, D. M. W. (1991). Emulsifying behavior of
376
gum-arabic. Part 2: Effect of the gum molecular-weight on the emulsion droplet-size
377
distribution. Carbohydrate Polymers, 14, 385-392. Ebringerova, A., & Heinze, T. (2000). Xylan and xylan derivatives‒biopolymers with valuable
EP
378
TE D
375
properties,
380
properties. Macromolecular Rapid Communications, 21, 542-556.
381 382 383 384 385 386
1.
Naturally
occurring
xylans
structures,
procedures
and
AC C
379
Ebringerova, A., Hromadkova, Z., & Heinze, T. (2005). Hemicellulose. In: T. Heinze (Ed.), Polysaccharides I: Structure, Characterization and Use (pp. 1-67). Springer.
Garti, N. (1999). Hydrocolloids as emulsifying agents for oil-in-water emulsions. Journal of Dispersion Science and Technology, 20, 327-355.
Gáspár, M., Kalman, G., & Reczey, K. (2007). Corn fiber as a raw material for hemicellulose and ethanol production. Process Biochemistry, 42, 1135-1139.
17
ACCEPTED MANUSCRIPT
388 389
Given, P. S. (2009). Encapsulation of flavors in emulsions for beverages. Current Opinion in Colloid & Interface Science, 14, 43-47. Hansen, N. M. L., & Plackett, D. (2011). Synthesis and characterization of birch wood xylan
390
succinoylated
391
2010-2020.
in
1-n-butyl-3-methylimidazolium
chloride. Polymer
Chemistry, 2,
RI PT
387
Islam, A. M., Phillips, G. O., Sljivo, A., Snowden, M. J., & Williams, P. A. (1997). A review of
393
recent developments on the regulatory, structural and functional aspects of gum
394
arabic. Food Hydrocolloids, 11, 493-505.
SC
392
Jagadeesh, D., Reddy, D. J. P., & Rajulu, A. V. (2011). Preparation and properties of
396
biodegradable films from wheat protein isolate. Journal of Polymers and the
397
Environment, 19, 248-253.
398 399 400 401
M AN U
395
Jayme, M. L., Dunstan, D. E., & Gee, M. L. (1999). Zeta potentials of gum arabic stabilised oil in water emulsions. Food Hydrocolloids, 13, 459-465.
Jeon, Y-S., Viswanathan, A., & Gross, R. A. (1999). Studies of starch esterification: Reactions with alkenylsuccinates in aqueous slurry systems. Starch-Starke, 51, 90-93. Keowmaneechai, E., & McClements, D. J. (2002). Influence of EDTA and citrate on
403
physicochemical properties of whey protein-stabilized oil-in-water emulsions containing
404
CaCl2. Journal of Agricultural and Food Chemistry, 50, 7145-7153.
TE D
402
Li, W. Y., Jin, A. X., Liu, C. F., Sun, R. C., Zhang, A. P., & Kennedy, J. F. (2009).
406
Homogeneous modification of cellulose with succinic anhydride in ionic liquid using
407
4-dimethylaminopyridine as a catalyst. Carbohydrate Polymers, 78, 389-395.
AC C
EP
405
408
Lindblad, M. S., & Albertsson, A. C. (2005). Chemical modification of hemicelluloses and gums.
409
In S. Dumitriu (Ed.), Polysaccharides: Structural Diversity and Functional Versatility.
410 411 412
New York: Marcel Dekker.
McClements, D. J. (1999). Food Emulsions: Principles, Practice and Techniques. Boca Raton, FL: CRC Press.
18
ACCEPTED MANUSCRIPT
413
Peng, X. W., Ren, J. L., & Sun, R. C. (2010). Homogeneous esterification of xylan-rich
414
hemicelluloses
415
3519-3524.
with
maleic
anhydride
in
ionic
liquid. Biomacromolecules, 11,
Peng, X. W., Ren, J. L., Zhong, L. X., & Sun, R. C. (2011a). Homogeneous synthesis of
417
hemicellulosic succinates with high degree of substitution in ionic liquid. Carbohydrate
418
Polymers, 86, 1768-1774.
RI PT
416
Peng, X. W., Ren, J. L., Zhong, L. X., Cao, X. F., & Sun, R. C. (2011b). Microwave-induced
420
synthesis of carboxymethyl hemicelluloses and their rheological properties. Journal of
421
Agricultural and Food Chemistry, 59, 570-576.
SC
419
Qian, C., Decker, E. A., Xiao, H., & McClements, D. J. (2011). Comparison of biopolymer
423
emulsifier performance in formation and stabilization of orange oil-in-water
424
emulsions. Journal of the American Oil Chemists Society, 88, 47-55.
M AN U
422
425
Randall, R. C., Phillips, G. O., & Williams, P. A. (1988). The role of the proteinaceous
426
component on the emulsifying properties of gum arabic. Food Hydrocolloids, 2, 131-140.
427
Ren, J. L., Peng, X. W., Feng, P., & Sun, R. C. 2013. A new strategy for acid anhydrides-modified xylans in ionic liquids. Fibers and Polymers, 14, 16-21.
TE D
428 429
Renewable Fuel Association (RFA) (2015) 2015 Ethanol Industry Outlook: Going Global.
430
Saulnier, L., Marot, C., Chanliaud, E., & Thibault, J. F. (1995). Cell-wall polysaccharide interactions in maize bran. Carbohydrate Polymers, 26, 279-287.
EP
431
Stojanovic, Z., Jeremic, K., Jovanovic, S., & Lechner, M. D. (2005). A comparison of some
433
methods for the determination of the degree of substitution of carboxymethyl
434
AC C
432
starch. Starch-Starke, 57, 79-83.
435
Sun, C. H., & Gunasekaran, S. (2009). Effects of protein concentration and oil-phase volume
436
fraction on the stability and rheology of menhaden oil-in-water emulsions stabilized by
437
whey protein isolate with xanthan gum. Food Hydrocolloids, 23, 165-174.
438
Sun, R. C., Sun, X. F., & Bing, X. (2002). Succinoylation of wheat straw hemicelluloses with a
439
low degree of substitution in aqueous systems. Journal of Applied Polymer Science, 83,
440
757-766. 19
ACCEPTED MANUSCRIPT
442 443 444
Sun, R. C., Sun, X. F., & Zhang, F. Y. (2001). Succinoylation of wheat straw hemicelluloses in N,N-dimethylformamide/lithium chloride systems. Polymer International, 50, 803-811. Tabilo-Munizaga, G., & Barbosa-Canovas, G. V. (2005). Rheology for the food industry. Journal of Food Engineering, 67, 147-156.
RI PT
441
445
Viswanathan, A. (1999). Effect of degree of substitution of octenyl succinate starch on the
446
emulsification activity on different oil phases. Journal of Environmental Polymer
447
Degradation, 7, 191-196.
449
Walstra, P. (1993). Principles of emulsion formation. Chemical Engineering Science, 48,
SC
448
333-349.
Ward, F. M. (2002). Modified hydrocolloids with enhanced emulsifying properties. In P. A.
451
Williams, & G. O. Philips (Eds.), Gums and Stabilizers for the Food Industry (Vol. 11).
452
United Kingdom: Royal Society ofChemistry.
453 454
M AN U
450
Xiang, Z., & Runge, T. (2014). Co-production of feed and furfural from dried distillers' grains to improve corn ethanol profitability. Industrial Crops and Products, 55, 207-216. Xiang, Z., Anthony, R., Tobimatsu, Y., & Runge, T. (2014a). Emulsifying properties of an
456
arabinoxylan-protein gum from distillers' grains and the co-production of animal
457
feed. Cellulose, 21, 3623-3635.
TE D
455
Xiang, Z., Watson, J., Tobimatsu, Y., & Runge, T. (2014b). Film-forming polymers from
459
distillers' grains: structural and material properties. Industrial Crops and Products, 59,
460
282-289.
462 463
Yadav, M. P., Johnston, D. B., & Hicks, K. B. (2009). Corn fiber gum: New structure/function
AC C
461
EP
458
relationships for this potential beverage flavor stabilizer. Food Hydrocolloids, 23, 1488-1493.
464
Yadav, M.P., Johnston, D.B., Hotchkiss, A.T., Hicks, K.B. 2007. Corn fiber gum: A potential
465
gum arabic replacer for beverage flavor emulsification. Food Hydrocolloids, 21,
466
1022-1030.
467
20
ACCEPTED MANUSCRIPT
1
Table Caption
2
Table 1. Composition (% based on oven-dried weight) and molecular weight of gum arabic, corn
4
fiber APG and DG APG.
AC C
EP
TE D
M AN U
SC
RI PT
3
1
ACCEPTED MANUSCRIPT
Table 1. Composition and
DG APG b
Corn fiber APG
Gum arabic
Arabinan (%)
19
~30
20-50
Galactan (%)
5
~6
Glucan (%)
6
~1
Xylan (%)
28
~42
Mannan (%)
2
-
Uronic acid (%)
5
Protein (%)
12
Mw (kDa) c
450
Mn ((kDa)
62
Mw/Mn
7.2
RI PT 25-50 -
~4
10-20
1-5
0.5-2
300-450
300-900
-
-
1.3-2.0
-
Yadav et al., 2007, 2009
Islam, 1997; Randall, 1988
EP
Note: athe compositions might be recalculated to represent the percent of oven-dried weight of the gums from the data provided by the references; bthe data for DG APG were from the composition and molecular weight for the specific batch of this study, and were consistent with the references provided in the table; cMw is the weight average molecular weight, Mn is the number average molecular weight, Mw/Mn is the polydispersity index.
AC C
2 3 4 5 6
Xiang et al., 2014a, b
TE D
References
SC
molecular weight
a
M AN U
1
1
ACCEPTED MANUSCRIPT
1
Figure Captions
2
Figure 1. Succinylation of arabinoxylan.
4
Figure 2. FT-IR spectra of (a) APG; (b) SAPG with DS = 0.02; (c) SAPG with DS = 0.09.
5
Figure 3. 13C NMR spectra of (a) APG and (b) SAPG with DS = 0.12.
6
Figure 4. Degree of substitution (DS) of SAPG vs the amount of succinic anhydride (SA) added to
7
the succinylation reaction.
8
Figure 5. (a) Particle size and (b) zeta potential of emulsions stabilized by gum arabic (GA),
9
APG, SAPG0.02 and SAPG0.09 at pH 3.5 to 6.5 (e.g. SAPG0.02 means SAPG with DS = 0.02).
10
Figure 6. The effects of degree of substitution (DS) on the particle size of emulsions stabilized
11
by SAPG at pH 3.5, 5.5 and 6.5.
12
Figure 7. Cream index over a period of 7 days for emulsions stabilized by gum arabic, APG and
13
SAPG of different DS at (a) pH = 3.5 and (b) pH = 5.5 (e.g. SAPG0.02 means SAPG with DS =
14
0.02).
15
Figure 8. Dynamic viscosity of (a) 5 wt%, (b) 2.5 wt% and (c) 1.25 wt% solutions of gum
16
arabic, APG and SAPG with DS = 0.12 (Note: due to the viscometer’s limited capacity, at 1.25
17
wt% gum solution, the viscosity could only be measured at 100 rpm).
AC C
EP
TE D
M AN U
SC
RI PT
3
1
H
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
-
ACCEPTED MANUSCRIPT
RI PT SC
b c
M AN U
80
a
60
TE D
~1735 cm-1
EP
40 20 0 4000
~1580 cm-1 ~1160 cm
-1
AC C
% Transmittance
100
3500
3000
2500
2000
1500
1000
500
Wavenumber (cm-1)
SC M AN U
178.0 176.3
TE D
181.8
AC C
EP
b
{
AXU
a 200
150
100
50
0
ppm
ACCEPTED MANUSCRIPT
Degree of Substitution
RI PT
0.14
SC
0.12
M AN U
0.10 0.08
TE D
0.06
AC C
EP
0.04 0.02 0.00 0.2
0.4
0.6 0.8 SA added mol/mol AXU
a
Particle Size (nm)
2500
ACCEPTED MANUSCRIPT
GA APG SAPG0.02 SAPG0.09
2000
RI PT
1500
SC
1000
M AN U
500 0
b
pH=4.5
pH=5.5
pH=6.5
pH=5.5
pH=6.5
TE D
pH=3.5
Zeta Potential (mv)
-70
-50 -40
EP
GA APG SAPG0.02 SAPG0.09
AC C
-60
-30 -20 -10 0 pH=3.5
pH=4.5
ACCEPTED MANUSCRIPT
Particle size (nm) 2200
RI PT
2000
SC
1800
M AN U
1600
TE D
1400 pH=3.5 pH=5.5 pH=6.5
AC C
1000
EP
1200
800
600 0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
Degree of Substitution
a
Cream Index ACCEPTED MANUSCRIPT
100% 80%
RI PT
60%
GA
40%
SC
APG
0%
b
1
2
Cream Index
4
SAPG0.09 SAPG0.12
5
6 7 Days
EP
100%
3
SAPG0.02
TE D
0
M AN U
20%
AC C
80% 60%
GA
40%
APG SAPG0.02
20%
SAPG0.09 SAPG0.12
0% 0
1
2
3
4
5
6
Days
7
ACCEPTED MANUSCRIPT
c
mPa·s
160
mPa·s
10 APG SAPG0.12 GA
140 120
8 6
80
5
60
4 3
40
M AN U
2
20
RI PT
7
100
1
0 10
20
30
40
50
rpm
TE D
0
b
APG SAPG0.12 GA
9
SC
a
mPa·s
30
APG SAPG0.12 GA
EP
25
AC C
20 15 10 5 0 0
20
40
60
80
100
rpm
0
80
90
100
110
120
rpm
ACCEPTED MANUSCRIPT
Highlights 1. The arabinoxylan-protein gum (APG) from distillers’ grains (DG) was modified.
RI PT
2. APG was acylated by succinic anhydride with a max degree of substitution of 0.12. 3. Succinylated APG showed much better emulsifying properties than gum arabic at pH > 5.
AC C
EP
TE D
M AN U
SC
4. Succinylation eliminated the non-Newtonian fluid behavior of the APG solution.
S0268-005X(15)30028-X
DOI:
10.1016/j.foodhyd.2015.07.018
Reference:
FOOHYD 3074
To appear in:
Food Hydrocolloids
Received Date: 25 April 2015 Revised Date:
15 July 2015
Accepted Date: 17 July 2015
Please cite this article as: Xiang, Z., Runge, T., Emulsifying Properties of Succinylated ArabinoxylanProtein Gum Produced from Corn Ethanol Residuals, Food Hydrocolloids (2015), doi: 10.1016/ j.foodhyd.2015.07.018. 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.
ACCEPTED MANUSCRIPT
Particle size (nm) 2200
RI PT
Arabinoxylan-Protein Gum (APG) from Distillers’ Grains (DG)
AC C
EP
TE D
M AN U
SC
1800 1400
pH=3.5 pH=5.5 pH=6.5
1000 600 0
0.04
0.08 0.12 Degree of Substitution
ACCEPTED MANUSCRIPT
Emulsifying Properties of Succinylated Arabinoxylan-Protein Gum
2
Produced from Corn Ethanol Residuals
3
Zhouyang Xiang and Troy Runge*
4
Department of Biological Systems Engineering, University of Wisconsin-Madison, Madison, WI
5
53706, United States
6
*Corresponding author:
7
Abstract
SC
RI PT
1
M AN U
Email: [email protected]; Tel: 1-608-890-3143
This study investigated the possibilities of making valuable products from corn ethanol
9
byproducts and providing the beverage industries more variety of high quality emulsifiers other
10
than gum arabic. An arabinoxylan-protein gum (APG) was extracted from distillers’ grains (DG),
11
a low-value corn ethanol byproduct, and modified through acylation with succinic anhydride. The
12
effects of pH and degree of substitution (DS) on the emulsifying properties of succinylated APG,
13
referred to as SAPG, were investigated. Emulsion particle size and stability of APG and gum
14
arabic were comparable at pH 3.5-6.5. Succinylation could enhance the emulsifying properties of
15
APG. Compared to gum arabic, at pH < 5, SAPG emulsions had larger particle size but
16
comparable stability, whereas at pH > 5, SAPG had much smaller particle size and better stability
17
than gum arabic. The results suggested that SAPG, compared to gum arabic, could be a
18
comparable emulsifier at low pH values and a better emulsifier at neutral pH values.
19
Keywords: arabinoxylan-protein gum; degree of substitution; distillers’ grains; gum arabic;
20
oil-in-water emulsions; particle size; emulsion stability; viscosity
AC C
EP
TE D
8
1
ACCEPTED MANUSCRIPT
21
1. Introduction Oil-in-water emulsions are common for many beverages with the oil droplets dispersed in the
23
beverage’s aqueous phase serving primarily as flavor oils (Chanamai & McClements, 2002; Given,
24
2009; Qian, Decker, Xiao, & McClements, 2011). Some non-flavor oils, such as omega-3 fatty
25
acids, are also sometimes included to provide special nutrients (Given, 2009). Different from other
26
food emulsions, beverage emulsions are very dilute (~0.2 wt% oil) and should have a great
27
stability for a long period of time (Given, 2009; Yadav, Johnston, Hotchkiss, & Hicks, 2007). To
28
disperse oil droplets in the aqueous phase, emulsifiers are needed to both facilitate the emulsion
29
formation and stabilize the emulsion preventing the droplets from aggregating (Walstra, 1993).
M AN U
SC
RI PT
22
The emulsifiers used in beverage emulsions are primarily amphiphilic polysaccharides due to
31
their excellent emulsion stabilizing ability under changing environmental conditions (Chanamai &
32
McClements, 2002; Garti, 1999; Qian et al., 2011). Gum arabic is the most commonly used in
33
beverages due to its bio-based feature, food safety, high water solubility, low solution viscosity and
34
great emulsion stability (Chanamai & Mcclements, 2002; Yadav et al., 2007). Gum arabic is a
35
polysaccharides-protein complex with arabinogalactan as the main polysaccharides units; during
36
emulsifying process, the hydrophobic protein portions anchor into the oil droplets, while the
37
hydrophilic polysaccharides portions stick out into the aqueous phase, preventing the droplets
38
from aggregation by steric or electrostatic effects (Chanamai & Mcclements, 2002; Islam, Phillips,
39
Sljivo, Snowden, & Williams, 1997; Jayme, Dunstan, & Gee, 1999; Randall, Phillips, & Williams,
40
1988; Walstra, 1993; Xiang, Anthony, Tobimatsu, & Runge, 2014a). Gum arabic is an effective
41
emulsifier for beverage; however, the plants (Acacia senegal) producing gum arabic are primarily
42
grown in the Sahelian region of Africa, and thus its supply is limited and price is sensitive to
43
climate changes and is comparatively high to other food ingredients (Yadav et al., 2007).
AC C
EP
TE D
30
2
ACCEPTED MANUSCRIPT
As the demand for beverages and thus beverage emulsifiers increases, many studies have been
45
conducted to find substitutes for gum arabic. Modified starch is common replacement beverage
46
emulsifier for gum arabic (Chanamai & McClements, 2002). Hydrophobic or non-polar groups are
47
introduced onto starch through its abundant hydroxyls imparting it with both hydrophobic and
48
hydrophilic portions similar to gum arabic; one of the examples is octenyl succinic acid (OSA)
49
starch (Caldwell & Wurzburg, 1952; Chanamai & McClements, 2002; Viswanathan, 1999). Gum
50
arabic has also been modified with octenyl succinic acid to increase the amount of its hydrophobic
51
portion and reduce its usage level (Ward, 2002). Globular proteins, such as whey protein, are able
52
to create oil droplets with size smaller than gum arabic during homogenization; however their
53
stability is vulnerable to environmental condition changes (pH, temperature, etc.) (Chanamai &
54
McClements, 2002; McClements, 1999; Qian et al., 2011). More recently, an arabinoxylan-protein
55
gum (APG) extracted from corn ethanol byproducts was found with promising emulsifying
56
properties (Xiang et al., 2014a; Yadav, Johnston, & Hicks, 2009; Yadav et al., 2007).
TE D
M AN U
SC
RI PT
44
The U.S. produced approximately 56 billion liters of corn ethanol in 2014 yielding huge
58
amount of fiber-protein rich byproducts (e.g. yielding ~40 million metric tons of distillers’ grains)
59
(RFA, 2015). Wet mill and dry grind are the two major processes in corn ethanol production
60
industry in the US, between which dry grind is the most predominant process (Bothast &
61
Schlicher, 2005; RFA, 2015). Many of the corn ethanol byproducts, such as corn fiber and
62
distillers’ grains (DG), from wet mill and dry grind processes respectively, are sold as animal
63
feed at a very low price (Xiang & Runge, 2014). Therefore, effectively utilizing the proteins and
64
polysaccharides in corn ethanol byproducts to make more valuable products could potentially
65
increase corn ethanol profitability and sustainability (Xiang & Runge, 2014; Xiang et al., 2014a;
66
Xiang, Watson, Tobimatsu, & Runge, 2014b).
AC C
EP
57
3
ACCEPTED MANUSCRIPT
Recently, an arabinoxylan-protein gum (APG) was extracted by alkali and/or H2O2 from corn
68
fibers (Yadav et al., 2007, 2009) or distillers’ grains (Xiang et al., 2014a). This gum is also an
69
amphiphilic polysaccharide, similar to gum arabic (an arabinogalactan-protein complex). The
70
structural differences between APG and gum arabic might be reflected by their chemical
71
compositions as shown in Table 1. Other than the amphiphilic polysaccharide feature, the
72
comparable molecular weight of APG with gum arabic (Table 1) also suggests the potential
73
application of APG as beverage emulsifiers (Dickinson, Galazka, & Anderson, 1991).
74
Arabinoxylan is a heterogeneous polymer and a major type of hemicelluloses found in corn
75
(Gáspár, Kalman, & Reczey, 2007; Xiang et al., 2014b), with the backbone of
76
β-(1-4)-D-xylopyranose being substituted primarily of α-L-arabinofuranose and slightly of
77
D-galactopyranose
78
corn, some of the hydroxyproline-rich glycoproteins are believed to link to the arabinoxylan
79
through covalent bonds that were stable in alkali or H2O2 (Saulnier, Marot, Chanliaud, & Thibault,
80
1995; Xiang et al., 2014a; Yadav et al., 2007). Thus, some protein could be co-extracted with
81
arabinoxylan as an arabinoxylan-protein complex and not be removed by the physical purification
82
process (Xiang et al., 2014a).
M AN U
SC
RI PT
67
EP
TE D
and D-glucuronic acid (Ebringerová & Heinze, 2000; Xiang et al., 2014b). In
The corn fiber APG has been shown to have better emulsifying properties than regular or
84
modified gum arabic (Yadav et al., 2007, 2009). The emulsifying properties of the DG APG were
85
only comparable to gum arabic (Xiang et al., 2014a), and thus could be further improved through
86
chemical modifications. Gum arabic or starch is typically modified with octenyl succinic acid to
87
increase the amount of its hydrophobic portion (Ward, 2002). However, as shown in Table 1, the
88
DG APG has protein content more than 10% (Xiang et al., 2014a), much higher than that of gum
89
arabic (~1%) and corn fiber APG (1-5%) (Yadav et al., 2007, 2009). Additionally, gum arabic has
AC C
83
4
ACCEPTED MANUSCRIPT
10-20% uronic acid (Islam, 1997; Randall et al., 1988) compared to ~5% in DG APG (Xiang et al.,
91
2014a). Thus, it is hypothesized that chemically modifying the DG APG hydrophilic arabinoxylan
92
portion with more steric or electrostatic effects might be an effective way to improve its
93
emulsifying properties.
RI PT
90
Acylation of polysaccharides with acid anhydrides is an effectively way to impart
95
polysaccharides with carboxylic acid groups. High degree of substitution (DS) of acylation of
96
polysaccharides with acid anhydrides is usually conducted in N,N-dimethyl formamide
97
(DMF)/LiCl solutions with or without catalysts (Ebringerová et al., 2005; Lindblad & Albertsson,
98
2005; Sun, Sun, & Zhang, 2001). More recently, ionic liquids have also been used as an effective
99
solvent (Hansen & Plackett, 2011; Peng, Ren, Zhong, & Sun, 2011a; Peng, Ren, & Sun, 2010;
100
Ren, Peng, Feng, & Sun, 2013). Additionally, low DS acylations with acid anhydride were
101
reported on starch and hemicelluloses by using alkaline solution as solvents (Betancur-Ancona,
102
Garcia-Cervera, Canizares-Hernandez, & Chel-Guerrero, 2002; Jeon, Viswanathan, & Gross,
103
1999; Sun, Sun, & Bing, 2002). Avoiding the use of the organic solvents or ionic liquid would
104
both economically and environmentally benefit the application of modified DG gum in food
105
industries.
EP
TE D
M AN U
SC
94
Therefore, to find suitable substitutes for gum arabic to relieve its limited supply, and to
107
improve the value from the corn ethanol byproducts, we investigated acylating the
108
arabinoxylan-protein gum (APG) from DG by succinic anhydride. The effects of DS and pH on the
109
emulsifying properties of modified DG APG were quantified and compared to GA.
AC C
106
110
5
ACCEPTED MANUSCRIPT
111
2. Materials and methods
112
2.1. Fractionation of APG Distillers’ grains (DG) samples were obtained from Didion Milling Inc. (Cambria, WI,
114
USA). The arabinoxylan-protein gum (APG) was extracted and purified according to the
115
procedure described previously (Xiang et al., 2014a). The DG was pre-extracted by acetone for
116
12h in a Soxhlet extractor to remove fats and organic impurities and then was reacted in 3%
117
(w/v) NaOH solution at 50 °C at 1:10 solid to liquid ratio (w/v) for 3h. The separated
118
alkali-soluble part was adjusted to pH 5.5, purified with bentonite clays to remove soluble
119
proteins, and slowly poured into three volume of 95% ethanol with constant stirring. The
120
precipitated solid was washed several times with 70% ethanol, freeze-dried, and ground to give a
121
powder form of APG.
M AN U
SC
RI PT
113
The extracted APG consisted of (based on the oven-dried weight) 19% arabinan, 28% xylan,
123
5% galactan, 6% glucan, 2% mannan, 5% uronic acid, and 12% crude protein (Table 1). Number
124
average molecular weight (Mn) of the APG was 62 kDa, weight average molecular weight (Mw)
125
was 450 kDa, and the polydispersity index (Mw/Mn) was 7.2 (Table 1). The compositions and
126
molecular weights of the APG were consistent with our previous studies, which provide a
127
detailed characterization (Xiang et al., 2014a, b).
128
2.2. Succinylation of APG
AC C
EP
TE D
122
129
APG was acylated by succinic anhydride (SA) based on the procedure modified from Sun et
130
al. (2002). The unmodified APG was dissolved in 1% (w/v) NaOH solution to make a 5% (w/v)
131
solution. Different amounts of SA (99.9%, Chem-impex Int’l Inc., IL, USA) leading to
132
SA/arabinoxylose unit (AXU) mole ratios of 0.25, 0.33, 0.50 and 0.75 were added. The
133
abbreviation AXU represents arabinoxylan mono sugar unit, which has a molar mass of 100
6
ACCEPTED MANUSCRIPT
g/mol assuming the arabinoxylan consists of only arabinose and xylose since the content of other
135
minor units is very small (Xiang et al., 2014a, b). The mixture was adjusted to pH 8.5-9.0, and
136
maintained at 30 °C with stirring for 1h. The succinic anhydride modified DG
137
arabinoxylan-protein gum (SAPG) was precipitated by pouring the mixture into three volumes of
138
95% ethanol and was washed twice with 70% ethanol solution and once with acetone in order to
139
remove the unreacted SA.
140
2.3. Characterization of succinylated APG
SC
RI PT
134
The degree of substitution (DS) of a SAPG sample was defined as the mole of SA reacted
142
per mole of AXU. The samples were coded as SAPGn, where the number n denotes the DS.
143
The DS of SAPG was determined by measuring the amount of carboxylic acid group using a
144
procedure modified from Stojanović, Jeremic, Jovanovic, & Lechner (2005). A weighed SAPG
145
sample of about 0.125 g was completely dissolved in 5 ml 0.2 M NaOH solution, to which 75 ml
146
DI water was added. The excess NaOH was titrated with standard 0.05 M H2SO4 until a 3.75 pH
147
was achieved. A blank sample was also titrated. The mole of COOH in the SAX sample was
148
calculated as:
TE D
nCOOH = (Vb − V ) ⋅ 2 ⋅ cH 2SO4
EP
149
M AN U
141
where Vb is the volume (ml) of H2SO4 used for the titration of the blank; V is the volume (ml) of
151
H2SO4 used for the titration of the sample; cH2SO4 is the mole concentration (mol/ml) of H2SO4
152
used for the titration. Using the measured moles of COOH, the DS can be calculated as:
153
AC C
150
DS =
132 × nCOOH ms − 100 × nCOOH
154
where 132 is the molar mass (g/mol) of an AXU; 100 is the increase in molar mass (g/mol) of
155
an AXU for each SA substituted; ms is the oven dry mass (g) of the SAPG sample used for
156
titration; nCOOH is the mole of COOH. 7
ACCEPTED MANUSCRIPT
157
13
C NMR spectra of APG and SAPG samples were acquired on a Bruker Biospin (Billerica,
MA, USA) AVANCE 700 MHz spectrometer fitted with a cryogenically-cooled 5-mm TXI
159
gradient probe with inverse geometry (proton coils closest to the sample). Finely ball-milled
160
samples (~50 mg) were swelled in D2O (1 mL). The spectra were recorded after 8000 normal
161
scans and processed using Topspin 3.1 software (Bruker, Billerica, MA, USA).
RI PT
158
Fourier transform infrared (FT-IR) spectra were obtained using a total of 10 scans using a
163
Spectrum 100 FT-IR spectrometer (Perkin Elmer, Waltham, MA, USA) equipped with a
164
universal ATR sampling accessory.
165
2.4. Emulsion preparation and emulsifying properties
M AN U
SC
162
APG, SAPG or a gum arabic sample (MP Biomedicals, Santa Ana, CA, USA) were dissolved
167
(5 wt%) in the aqueous solution by using a sonicator for 30 min. The gum solution was then diluted
168
50 times into a 0.1 wt% gum solution and adjusted to pH = 3.5, 4.5, 5.5 or 6.5 by 5 wt% or 0.5 wt%
169
citric acid. The gum solutions were homogenized with 0.5 wt% orange oil (Now Foods,
170
Bloomingdale, IL, USA) at a high shear (20000 rpm) in a Waring 700s Blender for 2 min and 2
171
times. The solutions were allowed to sit 5 min between the 2 blendings.
TE D
166
The surface-area-average diameter (droplet size, d32) and zeta potential of the emulsions were
173
measured right after preparations by a particle size analyzer (90Plus, Brookhaven, Long Island,
174
NY, USA) according to the procedure described by Sun & Gunasekaran (2009). Cream stability
175
test was conducted over a range of 8 days according to the procedure described by
176
Keowmaneechai & McClements (2002). A creaming index (CI) from this procedure was recorded
177
each day, which was calculated and as:
AC C
EP
172
178 179
where HT is the total emulsion height and Hs is the serum layer height. 8
ACCEPTED MANUSCRIPT
The rheological properties of the gum solutions were investigated. The dynamic (absolute)
181
viscosity of the gum solutions was determined at 25 °C by using a rotational viscometer
182
(Fungilab Adv Series, New York, USA) at rotational speeds of 1, 2.5, 5, 10, 20, 50, 100 rpm.
183
Due to the viscometer’s limited capacity, samples may not be measured at all selected speeds.
184
3. Results and discussions
185
3.1. Characterization of succinylated APG
SC
RI PT
180
Acylation of arabinoxylan with succinic anhydride (SA) would couple carboxylic acids to
187
arabinoxylose units (AXU) through ester bonds (Figure 1). To reach a high degree of substitution
188
(DS) up to 1.0, the reaction should be conducted in DMF/LiCl with catalyst (Sun et al., 2001).
189
However, this process might not be economical and might introduce toxicity for the final food
190
grade products. Reaction of arabinoxylan with SA under mild alkaline condition (pH = 8.5-9.0)
191
have been reported to be capable of controlling the DS below 0.2 (Sun et al., 2002), which was a
192
more suitable modification method for this study.
TE D
193
M AN U
186
The succinylated DG arabinoxylan-protein gum (SAPG) was first characterized by FTIR The peak at ~1735 cm-1 of SAPG corresponded to the C=O stretching of the ester
(Figure 2).
195
bond or the carboxylic acid group, and the peak at ~1160 cm-1 should be related the C-O
196
stretching of the ester bond (Sun et al., 2001, 2002). The peak at ~~1580 cm-1 could include the
197
N-H stretching of protein (Jagadeesh et al., 2001; Xiang et al., 2014b) and the asymmetric stretch
198
of the carboxylates (Cabaniss, Reddy, & Rajulu, 1998; Peng et al., 2011a). Additionally, the
199
absence of signal between 1840 cm−1 and 1760 cm−1 indicated that the SAPG samples were free
200
of detectable amount of unreacted SA. The
201
consistent with the FTIR spectra. The peaks at 181.8, 178.0, and 176.3 ppm correspond to the
AC C
EP
194
13
C NMR spectra (Figure 3) of the SAPG were in
9
ACCEPTED MANUSCRIPT
202
carbons in the carbonyl of ester bonds and carboxylic acids confirming successful succinylation
203
on the APG to SAPG. The degree of substitution (DS) of the SAPG samples was varied using different SA/AXU
205
mole ratio for the reaction (Figure 4). The overall range of DS was between 0.02-0.12, which is
206
consistent with the previous study (Sun et al., 2002). The DS increased as SA/AXU mole ratio
207
increased, but when SA/AXU mole ratio went beyond 0.5, adding more SA did not have a
208
positive response for increasing DS. Similar phenomenon was found when acylating starch with
209
dodecenyl succinic anhydride in alkaline solution (Jeon et al., 1999). This may be due to
210
insufficient mixing of SA and APG or insufficient reaction time for high extents of acylation
211
under high SA concentration (Jeon et al., 1999; Sun et al., 2001). The low DS of the SAPG
212
samples agreed with the low intensity of the carbonyl peaks from the NMR spectrum (Figure 3).
213
The increasing intensities of the FTIR spectra peaks at ~1160 cm-1, ~1580 cm-1 and ~1735 cm-1
214
also agreed with the increasing DS of the analyzed SAPG samples (Figure 2).
215
3.2. Particle size and zeta potential
TE D
M AN U
SC
RI PT
204
Particle size (surface-area-average diameter d32) is a key measure of the quality of an
217
emulsion formed. During oil and water homogenization, three processes occur simultaneously:
218
(1) large oil droplets are torn into smaller droplets; (2) the hydrophobic portion of the emulsifiers
219
is absorbed onto the droplet surface; (3) oil droplets re-encounter each other, while the
220
hydrophilic portion of the emulsifiers, which sticks out into the aqueous phase, prevents the
221
droplets from aggregation through steric or electrostatic effects (Jayme et al., 1999; Walstra,
222
1993). Thus, oil droplet or particle size reflects both of how efficient the emulsifier is absorbed
223
onto the oil droplet surface and prevents the droplet from aggregation. Zeta potential reflects the
AC C
EP
216
10
ACCEPTED MANUSCRIPT
224
electrostatic repulsion between the adjacent droplet electrical double-layer and thus determines
225
the electrostatic contribution to the stabilization of the emulsions (Jayme et al., 1999). The particle size (d32) and zeta potential of oil-in-water emulsions stabilized by gum arabic,
227
APG and SAPG at different pH (3.5-6.5) was measured, shown in Figure 5. The particle size
228
measurements of gum arabic emulsions were constant at ~1400 nm among different pH (Figure
229
5a), which was consistent with previous studies (Qian et al., 2011). Emulsions stabilized by APG
230
also demonstrated constant particle size over different pH, and the values were similar to those of
231
gum arabic. These results suggested that APG had similar performance and likely would be
232
useful in similar applications such as in assisting the formation of beverage oil-in-water
233
emulsions, typically done under pH values ranging between 3.5 and 6.5.
M AN U
SC
RI PT
226
The emulsions stabilized by SAPG with a low DS (0.02) also showed relatively stable
235
particle size over different pH, but at a higher value of ~1800 nm, compared to gum arabic and
236
APG (Figure 5a). However, the particle size of emulsions stabilized by SAPG with a high DS
237
(0.09) decreased significantly with increased pH, from ~1800 nm at pH = 3.5 and 4.5 to ~1200
238
nm at pH = 5.5 and ~800 nm at pH = 6.5. This phenomenon might be explained by the zeta
239
potential changes of the emulsions (Figure 5b). Compared to APG and SAPG0.02 emulsions at
240
high pH (5.5 and 6.5), SAPG0.09 emulsion demonstrated significantly higher zeta potential
241
indicating higher electrostatic repulsions between oil droplets. Gum arabic emulsion had high
242
zeta potential at pH 6.5 too, but its particle size was constant at either high or low pH. The reason
243
is that steric effects might be more prevalent than electrostatic repulsions in stabilizing the gum
244
arabic emulsion (Jayme et al., 1999; Randall et al., 1988). However, for SAPG0.09 sample, as
245
the zeta potential increased significantly and the particle size decreased significantly at pH = 5.5
246
and 6.5, it was clear that the electrostatic repulsion contributed significantly in stabilizing the
AC C
EP
TE D
234
11
ACCEPTED MANUSCRIPT
SAPG0.09 emulsion. The electrostatic repulsion starting to take major effects at pH = 5.5 was
248
likely due to that the succinic acid had pKa1 = 4.2 and pKa2 = 5.4, and thus at pH = 5.5, the
249
carboxylic acid groups on the SAPG sample were greatly charged. It suggests that at pH > 5.5,
250
the carboxylic acid group introduced by succinylation increased the charges of APG and the
251
electrostatic repulsion, thus preventing oil droplets from aggregating.
RI PT
247
Since at high DS, SAPG had better performance in stabilizing the emulsions, the effects of
253
DS on the particle size of the SAPG emulsions were also plotted in Figure 6. At pH = 3.5, as DS
254
increased, the particle size slightly increased from ~1500 nm to ~1800 nm. While at pH = 5.5
255
and 6.5, the particle size increased slightly at DS = 0.02, and then decreased significantly at DS =
256
0.09. Continuing to increase the DS to 0.12 had limited effects on the particle size at pH = 5.5
257
and 6.5. It clearly indicated that at pH = 3.5 the carboxylic groups of SAPG had very limited
258
charges due to their relatively high pKa and thus did not contribute to the electrostatic repulsion
259
between oil droplets. When the DS increased beyond 0.09, it was postulated that adding more
260
carboxylic groups was not effective in stabilizing the emulsions.
TE D
M AN U
SC
252
In summary, at a pH range between 3.5 and 6.5, APG had similar performance in forming
262
emulsions compared to gum arabic. Succinylation imparted APG with better emulsion forming
263
abilities at high DS and pH > 5, but reduced its emulsion forming abilities at low DS and pH < 5.
264
3.3. Emulsion stability
AC C
EP
261
265
Emulsion stability is critical in many applications such as beverage emulsions, since many
266
of these products are require to be stored for months or even years (Given, 2009; Yadav et al.,
267
2007). Gum arabic is excellent in its emulsion stability but not in creating an initial small particle
268
size. Some globular proteins are able to make emulsions with very small particle size, but very
269
unstable (Chanamai & McClements, 2002; McClements, 1999; Qian et al., 2011).
12
ACCEPTED MANUSCRIPT
The cream index (CI) reflects the emulsion’s stability to creaming. The CI of the emulsions
271
was recorded over a period of 7 days for the gum arabic, APG and SAPG stabilized oil-in-water
272
emulsions at pH = 3.5 and 5.5 (Figure 7). At pH = 3.5, graphs of CI of SAPG0.09 and SAPG0.12
273
emulsions overlapped with the gum arabic emulsion, while the graphs of APG and SAPG0.02
274
fell below gum arabic. It shows that at low pH, APG emulsions had better stability than those of
275
gum arabic, and even though SAPG emulsions had larger initial particle size (Figure 6), and their
276
stability was still comparable to gum arabic. This phenomenon could possibly be explained by
277
that, even though larger particle size should induce faster creaming (Keowmaneechai and
278
McClements, 2002; McClements, 1999), the oil droplets with absorbed SAPG may have high
279
density due to the succinylation on APG and thus reduce the creaming rate (Sun et al., 2009). At
280
pH = 5.5, APG and gum arabic emulsions demonstrated very similar stability, while all the
281
SAPG sample, including SAPG0.02, which had the initial highest particle size among the all, had
282
stability better than gum arabic and APG. This could also possibly be explained by the higher
283
density of the oil droplets with absorbed SAPG compared to that with APG (Sun et al., 2009).
284
On the other hand, at high pH, it was possible that the carboxylic groups on SAPG were greatly
285
charged and the electrostatic repulsion between oil droplets was increased, hindering the
286
flocculation and thus slowing down the creaming (McClements, 1999). Additionally, the higher
287
molecular weight of APG than gum arabic (Table 1) may contribute to the APG or SAPG
288
emulsion stabilities, since larger protein-containing macromolecules might be more responsible
289
to the emulsion stability (Dickinson et al., 1991).
290
3.4. Viscosity of gum solutions
AC C
EP
TE D
M AN U
SC
RI PT
270
291
Viscosity is another characteristic that should be considered in evaluating beverage
292
emulsifiers. Beverage emulsifiers might require low solution viscosity and gum arabic is well
13
ACCEPTED MANUSCRIPT
293
known for its low solution viscosities even at high concentration (Chanamai & McClements,
294
2002; Yadav et al., 2007). In this study, the dynamic (absolute) viscosity of the gum solution was evaluated at 5 wt%,
296
2.5 wt% and 1.25 wt% concentrations, which was much higher than the concentration typically
297
used in the emulsions (0.1 wt%). These higher concentrations were selected to accentuate the
298
differences between the gums. Figure 8 shows the dynamic (absolute) viscosity changes with
299
shear rates by varying the rotational speed. At 5 wt% gum solution (Figure 8a), APG
300
demonstrated non-Newtonian fluid behavior, showing an apparent shear thinning phenomenon
301
(shear rate) (Nishiwaki-Akine and Watanabe, 2014). SAPG0.12 sample behaved as a Newtonian
302
fluid and had viscosity (~70 mPa·s) lower than that of APG (~90 mPa·s at high shear rate). At
303
2.5 wt% gum solution (Figure 8b), both APG and SAPG sample behaved as a Newtonian fluid
304
and showed a large drop in viscosity to ~16 and ~14 mPa·s, respectively. At 1.25 wt% gum
305
solution (data not shown in the graph), the viscosity of APG and SAPG were ~6 mPa·s. The
306
non-Newtonian behavior of the APG solution was similar to the native wheat straw
307
hemicelluloses (Peng, Ren, Zhong, Cao, & Sun, 2011b). The SAPG showing lower viscosity
308
than the APG was in consistent with previous studies that succinylated cellulose and
309
carboxymethyl hemicelluloses showed lower viscosity than their native forms (Li, Jin, Liu, Sun,
310
Zhang, & Kennedy, 2009; Peng et al., 2011b), which was probably due the weakening of the
311
intermolecular interaction between polysaccharide chains (Peng et al., 2011b). The significant
312
changes of viscosity with solution concentration for APG and SAPG indicated strong
313
hydrogen-bonding between their polymer chains or high entanglement at high concentrations
314
(Peng et al., 2011b).
AC C
EP
TE D
M AN U
SC
RI PT
295
14
ACCEPTED MANUSCRIPT
The dynamic viscosity of gum arabic solution was constant at ~5 mPa·s among different
316
studied concentrations, which was much lower than that of the APG and SAPG solutions at high
317
concentration (Figure 8), but approximately the same at concentrations < 1 wt%. These results
318
suggested that the solution viscosity of SAPG was successfully reduced compared to APG
319
making it more suitable to be used as a beverage emulsifier. Even though the solution viscosity
320
of SAPG was still higher than gum arabic at high concentration, at low usage level (below 1 wt%)
321
of emulsifiers, which is most likely the case for making beverages (Yadav et al., 2007), it might
322
not be necessary to worry about the viscosity since APGS, APG and gum arabic are all showing
323
similar values. At high concentrations, APG and SAPG could also be used as either a food
324
thickener or an emulsifier. For some food products that are required to be thick and be a
325
non-Newtonian fluid such as salad dressing (Tabilo-Munizaga and Barbosa-Ca´novas, 2005),
326
APG might be a more promising food additive than SAPG.
327
4. Conclusions
TE D
M AN U
SC
RI PT
315
Reacting the arabinoxylan-protein gum (APG) extracted from distillers’ grains (DG) with
329
succinic anhydride (SA) in mild alkaline solution, a succinylated APG (SAPG) was able to be
330
produced without using organic solvents while allowing control of the degree of substitution (DS)
331
to levels below 0.2. Between pH = 3.5 and 6.5, the unmodified APG was capable of creating
332
dilute oil-in-water emulsions with particle size and stability comparable to gum arabic, with even
333
better stability at pH = 3.5. SAPG emulsifying properties were more susceptible to pH. When pH
334
< 5, SAPG created oil-in-water emulsions with larger particle size but comparable stability;
335
however, at pH > 5, SAPG created emulsions with much smaller particle size and better stability,
336
compared to APG and gum arabic. Increasing the DS could further enhance the SAPG
337
emulsifying properties, but beyond DS = 0.09, the DS effects were not significant. It may be
AC C
EP
328
15
ACCEPTED MANUSCRIPT
postulated that the carboxylic acid groups introduced by the succinylation had pKa value of
339
about 5, and thus at high pH, they were greatly charged providing highly improved electrostatic
340
repulsions between oil droplets. This suggests that unmodified and succinylated APG can
341
potentially substitute gum arabic as an emulsifier for producing beverage oil-in-water emulsions;
342
while succinylated APG (SAPG), especially at high pH, was possibly a much better emulsifier
343
than gum arabic.
RI PT
338
The solution viscosity of APG was greatly reduced by succinylation, making SAPG a more
345
suitable beverage emulsifier than APG from a viscosity standpoint. However, the non-Newtonian
346
behavior of concentrated APG solution could possibly make it a food thickener especially for the
347
food products required to be a non-Newtonian fluid.
348 349
Acknowledgements
M AN U
SC
344
This work was supported by U.S. Department of Agriculture, under contract USDA Critical
351
Agricultural Material Grant (2013-38202-20400). The authors gratefully acknowledge Dr.
352
Sundaram Gunasekaran’s assistance with FTIR, particle size and zeta potential analyses, and
353
Didion Milling Inc. for materials and valuable discussions.
354
EP
TE D
350
Abbreviations
356
APG, arabinoxylan-protein gum; AXU, arabinoxylose unit; CI, creaming index; DG, distillers’
357
grains;
358
arabinoxylan-protein gum
AC C
355
DS,
degree
of
substitution;
SA,
359 360
16
succinic
anhydride;
SAPG,
succinylated
ACCEPTED MANUSCRIPT
361
References
362
Betancur-Ancona, D., Garcia-Cervera, E., Canizares-Hernandez, E., & Chel-Guerrero, L. (2002). Chemical
364
succinylation. Starch-Starke, 54, 540-546.
366 367
of
Jack
Bean
starch
by
ethanol. Applied Microbiology and Biotechnology, 67, 19-25.
Cabaniss, S. E., Leenheer, J. A., & McVey, I. F. (1998). Aqueous infrared carboxylate Aliphatic
di-acids.
Spectrochimica
369
Biomolecular Spectroscopy, 54, 449-458.
Acta
Part
a-Molecular
and
M AN U
absorbances:
371
ensiformis)
Bothast, R., & Schlicher, M. (2005). Biotechnological processes for conversion of corn into
368
370
(Canavalia
SC
365
modification
RI PT
363
Caldwell, C. G., & Wurzburg, O. B. (1953). Polysaccharide derivatives of substituted dicarboxylic acids. Patent: US 2661349 A, USA.
372
Chanamai, R., & McClements, D. J. (2002). Comparison of gum arabic, modified starch, and
373
whey protein isolate as emulsifiers: Influence of pH, CaCl2 and temperature. Journal of
374
Food Science, 67, 120-125.
Dickinson, E., Galazka, V. B., & Anderson, D. M. W. (1991). Emulsifying behavior of
376
gum-arabic. Part 2: Effect of the gum molecular-weight on the emulsion droplet-size
377
distribution. Carbohydrate Polymers, 14, 385-392. Ebringerova, A., & Heinze, T. (2000). Xylan and xylan derivatives‒biopolymers with valuable
EP
378
TE D
375
properties,
380
properties. Macromolecular Rapid Communications, 21, 542-556.
381 382 383 384 385 386
1.
Naturally
occurring
xylans
structures,
procedures
and
AC C
379
Ebringerova, A., Hromadkova, Z., & Heinze, T. (2005). Hemicellulose. In: T. Heinze (Ed.), Polysaccharides I: Structure, Characterization and Use (pp. 1-67). Springer.
Garti, N. (1999). Hydrocolloids as emulsifying agents for oil-in-water emulsions. Journal of Dispersion Science and Technology, 20, 327-355.
Gáspár, M., Kalman, G., & Reczey, K. (2007). Corn fiber as a raw material for hemicellulose and ethanol production. Process Biochemistry, 42, 1135-1139.
17
ACCEPTED MANUSCRIPT
388 389
Given, P. S. (2009). Encapsulation of flavors in emulsions for beverages. Current Opinion in Colloid & Interface Science, 14, 43-47. Hansen, N. M. L., & Plackett, D. (2011). Synthesis and characterization of birch wood xylan
390
succinoylated
391
2010-2020.
in
1-n-butyl-3-methylimidazolium
chloride. Polymer
Chemistry, 2,
RI PT
387
Islam, A. M., Phillips, G. O., Sljivo, A., Snowden, M. J., & Williams, P. A. (1997). A review of
393
recent developments on the regulatory, structural and functional aspects of gum
394
arabic. Food Hydrocolloids, 11, 493-505.
SC
392
Jagadeesh, D., Reddy, D. J. P., & Rajulu, A. V. (2011). Preparation and properties of
396
biodegradable films from wheat protein isolate. Journal of Polymers and the
397
Environment, 19, 248-253.
398 399 400 401
M AN U
395
Jayme, M. L., Dunstan, D. E., & Gee, M. L. (1999). Zeta potentials of gum arabic stabilised oil in water emulsions. Food Hydrocolloids, 13, 459-465.
Jeon, Y-S., Viswanathan, A., & Gross, R. A. (1999). Studies of starch esterification: Reactions with alkenylsuccinates in aqueous slurry systems. Starch-Starke, 51, 90-93. Keowmaneechai, E., & McClements, D. J. (2002). Influence of EDTA and citrate on
403
physicochemical properties of whey protein-stabilized oil-in-water emulsions containing
404
CaCl2. Journal of Agricultural and Food Chemistry, 50, 7145-7153.
TE D
402
Li, W. Y., Jin, A. X., Liu, C. F., Sun, R. C., Zhang, A. P., & Kennedy, J. F. (2009).
406
Homogeneous modification of cellulose with succinic anhydride in ionic liquid using
407
4-dimethylaminopyridine as a catalyst. Carbohydrate Polymers, 78, 389-395.
AC C
EP
405
408
Lindblad, M. S., & Albertsson, A. C. (2005). Chemical modification of hemicelluloses and gums.
409
In S. Dumitriu (Ed.), Polysaccharides: Structural Diversity and Functional Versatility.
410 411 412
New York: Marcel Dekker.
McClements, D. J. (1999). Food Emulsions: Principles, Practice and Techniques. Boca Raton, FL: CRC Press.
18
ACCEPTED MANUSCRIPT
413
Peng, X. W., Ren, J. L., & Sun, R. C. (2010). Homogeneous esterification of xylan-rich
414
hemicelluloses
415
3519-3524.
with
maleic
anhydride
in
ionic
liquid. Biomacromolecules, 11,
Peng, X. W., Ren, J. L., Zhong, L. X., & Sun, R. C. (2011a). Homogeneous synthesis of
417
hemicellulosic succinates with high degree of substitution in ionic liquid. Carbohydrate
418
Polymers, 86, 1768-1774.
RI PT
416
Peng, X. W., Ren, J. L., Zhong, L. X., Cao, X. F., & Sun, R. C. (2011b). Microwave-induced
420
synthesis of carboxymethyl hemicelluloses and their rheological properties. Journal of
421
Agricultural and Food Chemistry, 59, 570-576.
SC
419
Qian, C., Decker, E. A., Xiao, H., & McClements, D. J. (2011). Comparison of biopolymer
423
emulsifier performance in formation and stabilization of orange oil-in-water
424
emulsions. Journal of the American Oil Chemists Society, 88, 47-55.
M AN U
422
425
Randall, R. C., Phillips, G. O., & Williams, P. A. (1988). The role of the proteinaceous
426
component on the emulsifying properties of gum arabic. Food Hydrocolloids, 2, 131-140.
427
Ren, J. L., Peng, X. W., Feng, P., & Sun, R. C. 2013. A new strategy for acid anhydrides-modified xylans in ionic liquids. Fibers and Polymers, 14, 16-21.
TE D
428 429
Renewable Fuel Association (RFA) (2015) 2015 Ethanol Industry Outlook: Going Global.
430
Saulnier, L., Marot, C., Chanliaud, E., & Thibault, J. F. (1995). Cell-wall polysaccharide interactions in maize bran. Carbohydrate Polymers, 26, 279-287.
EP
431
Stojanovic, Z., Jeremic, K., Jovanovic, S., & Lechner, M. D. (2005). A comparison of some
433
methods for the determination of the degree of substitution of carboxymethyl
434
AC C
432
starch. Starch-Starke, 57, 79-83.
435
Sun, C. H., & Gunasekaran, S. (2009). Effects of protein concentration and oil-phase volume
436
fraction on the stability and rheology of menhaden oil-in-water emulsions stabilized by
437
whey protein isolate with xanthan gum. Food Hydrocolloids, 23, 165-174.
438
Sun, R. C., Sun, X. F., & Bing, X. (2002). Succinoylation of wheat straw hemicelluloses with a
439
low degree of substitution in aqueous systems. Journal of Applied Polymer Science, 83,
440
757-766. 19
ACCEPTED MANUSCRIPT
442 443 444
Sun, R. C., Sun, X. F., & Zhang, F. Y. (2001). Succinoylation of wheat straw hemicelluloses in N,N-dimethylformamide/lithium chloride systems. Polymer International, 50, 803-811. Tabilo-Munizaga, G., & Barbosa-Canovas, G. V. (2005). Rheology for the food industry. Journal of Food Engineering, 67, 147-156.
RI PT
441
445
Viswanathan, A. (1999). Effect of degree of substitution of octenyl succinate starch on the
446
emulsification activity on different oil phases. Journal of Environmental Polymer
447
Degradation, 7, 191-196.
449
Walstra, P. (1993). Principles of emulsion formation. Chemical Engineering Science, 48,
SC
448
333-349.
Ward, F. M. (2002). Modified hydrocolloids with enhanced emulsifying properties. In P. A.
451
Williams, & G. O. Philips (Eds.), Gums and Stabilizers for the Food Industry (Vol. 11).
452
United Kingdom: Royal Society ofChemistry.
453 454
M AN U
450
Xiang, Z., & Runge, T. (2014). Co-production of feed and furfural from dried distillers' grains to improve corn ethanol profitability. Industrial Crops and Products, 55, 207-216. Xiang, Z., Anthony, R., Tobimatsu, Y., & Runge, T. (2014a). Emulsifying properties of an
456
arabinoxylan-protein gum from distillers' grains and the co-production of animal
457
feed. Cellulose, 21, 3623-3635.
TE D
455
Xiang, Z., Watson, J., Tobimatsu, Y., & Runge, T. (2014b). Film-forming polymers from
459
distillers' grains: structural and material properties. Industrial Crops and Products, 59,
460
282-289.
462 463
Yadav, M. P., Johnston, D. B., & Hicks, K. B. (2009). Corn fiber gum: New structure/function
AC C
461
EP
458
relationships for this potential beverage flavor stabilizer. Food Hydrocolloids, 23, 1488-1493.
464
Yadav, M.P., Johnston, D.B., Hotchkiss, A.T., Hicks, K.B. 2007. Corn fiber gum: A potential
465
gum arabic replacer for beverage flavor emulsification. Food Hydrocolloids, 21,
466
1022-1030.
467
20
ACCEPTED MANUSCRIPT
1
Table Caption
2
Table 1. Composition (% based on oven-dried weight) and molecular weight of gum arabic, corn
4
fiber APG and DG APG.
AC C
EP
TE D
M AN U
SC
RI PT
3
1
ACCEPTED MANUSCRIPT
Table 1. Composition and
DG APG b
Corn fiber APG
Gum arabic
Arabinan (%)
19
~30
20-50
Galactan (%)
5
~6
Glucan (%)
6
~1
Xylan (%)
28
~42
Mannan (%)
2
-
Uronic acid (%)
5
Protein (%)
12
Mw (kDa) c
450
Mn ((kDa)
62
Mw/Mn
7.2
RI PT 25-50 -
~4
10-20
1-5
0.5-2
300-450
300-900
-
-
1.3-2.0
-
Yadav et al., 2007, 2009
Islam, 1997; Randall, 1988
EP
Note: athe compositions might be recalculated to represent the percent of oven-dried weight of the gums from the data provided by the references; bthe data for DG APG were from the composition and molecular weight for the specific batch of this study, and were consistent with the references provided in the table; cMw is the weight average molecular weight, Mn is the number average molecular weight, Mw/Mn is the polydispersity index.
AC C
2 3 4 5 6
Xiang et al., 2014a, b
TE D
References
SC
molecular weight
a
M AN U
1
1
ACCEPTED MANUSCRIPT
1
Figure Captions
2
Figure 1. Succinylation of arabinoxylan.
4
Figure 2. FT-IR spectra of (a) APG; (b) SAPG with DS = 0.02; (c) SAPG with DS = 0.09.
5
Figure 3. 13C NMR spectra of (a) APG and (b) SAPG with DS = 0.12.
6
Figure 4. Degree of substitution (DS) of SAPG vs the amount of succinic anhydride (SA) added to
7
the succinylation reaction.
8
Figure 5. (a) Particle size and (b) zeta potential of emulsions stabilized by gum arabic (GA),
9
APG, SAPG0.02 and SAPG0.09 at pH 3.5 to 6.5 (e.g. SAPG0.02 means SAPG with DS = 0.02).
10
Figure 6. The effects of degree of substitution (DS) on the particle size of emulsions stabilized
11
by SAPG at pH 3.5, 5.5 and 6.5.
12
Figure 7. Cream index over a period of 7 days for emulsions stabilized by gum arabic, APG and
13
SAPG of different DS at (a) pH = 3.5 and (b) pH = 5.5 (e.g. SAPG0.02 means SAPG with DS =
14
0.02).
15
Figure 8. Dynamic viscosity of (a) 5 wt%, (b) 2.5 wt% and (c) 1.25 wt% solutions of gum
16
arabic, APG and SAPG with DS = 0.12 (Note: due to the viscometer’s limited capacity, at 1.25
17
wt% gum solution, the viscosity could only be measured at 100 rpm).
AC C
EP
TE D
M AN U
SC
RI PT
3
1
H
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
-
ACCEPTED MANUSCRIPT
RI PT SC
b c
M AN U
80
a
60
TE D
~1735 cm-1
EP
40 20 0 4000
~1580 cm-1 ~1160 cm
-1
AC C
% Transmittance
100
3500
3000
2500
2000
1500
1000
500
Wavenumber (cm-1)
SC M AN U
178.0 176.3
TE D
181.8
AC C
EP
b
{
AXU
a 200
150
100
50
0
ppm
ACCEPTED MANUSCRIPT
Degree of Substitution
RI PT
0.14
SC
0.12
M AN U
0.10 0.08
TE D
0.06
AC C
EP
0.04 0.02 0.00 0.2
0.4
0.6 0.8 SA added mol/mol AXU
a
Particle Size (nm)
2500
ACCEPTED MANUSCRIPT
GA APG SAPG0.02 SAPG0.09
2000
RI PT
1500
SC
1000
M AN U
500 0
b
pH=4.5
pH=5.5
pH=6.5
pH=5.5
pH=6.5
TE D
pH=3.5
Zeta Potential (mv)
-70
-50 -40
EP
GA APG SAPG0.02 SAPG0.09
AC C
-60
-30 -20 -10 0 pH=3.5
pH=4.5
ACCEPTED MANUSCRIPT
Particle size (nm) 2200
RI PT
2000
SC
1800
M AN U
1600
TE D
1400 pH=3.5 pH=5.5 pH=6.5
AC C
1000
EP
1200
800
600 0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
Degree of Substitution
a
Cream Index ACCEPTED MANUSCRIPT
100% 80%
RI PT
60%
GA
40%
SC
APG
0%
b
1
2
Cream Index
4
SAPG0.09 SAPG0.12
5
6 7 Days
EP
100%
3
SAPG0.02
TE D
0
M AN U
20%
AC C
80% 60%
GA
40%
APG SAPG0.02
20%
SAPG0.09 SAPG0.12
0% 0
1
2
3
4
5
6
Days
7
ACCEPTED MANUSCRIPT
c
mPa·s
160
mPa·s
10 APG SAPG0.12 GA
140 120
8 6
80
5
60
4 3
40
M AN U
2
20
RI PT
7
100
1
0 10
20
30
40
50
rpm
TE D
0
b
APG SAPG0.12 GA
9
SC
a
mPa·s
30
APG SAPG0.12 GA
EP
25
AC C
20 15 10 5 0 0
20
40
60
80
100
rpm
0
80
90
100
110
120
rpm
ACCEPTED MANUSCRIPT
Highlights 1. The arabinoxylan-protein gum (APG) from distillers’ grains (DG) was modified.
RI PT
2. APG was acylated by succinic anhydride with a max degree of substitution of 0.12. 3. Succinylated APG showed much better emulsifying properties than gum arabic at pH > 5.
AC C
EP
TE D
M AN U
SC
4. Succinylation eliminated the non-Newtonian fluid behavior of the APG solution.