Food Hydrocolloids 103 (2020) 105684

Contents lists available at ScienceDirect

Food Hydrocolloids
journal homepage: http://www.elsevier.com/locate/foodhyd

The radiation assisted-Maillard reaction comprehensively improves the

freeze-thaw stability of soy protein-stabilized oil-in-water emulsions
Yuying Wang , Anqi Zhang , Xibo Wang *, Ning Xu **, Lianzhou Jiang
College of Food Science, Northeast Agricultural University, Harbin, 150030, China

A R T I C L E I N F O A B S T R A C T

Keywords: For the first time, the irradiation technology combined with Maillard reaction were used to improve the freeze -
Soy protein isolate thaw stability of soybean protein emulsion. The freeze-thaw stability of emulsions prepared with soy protein
Irradiation isolate (SPI), soy protein isolate and maltose mixture (SPI þ M) or glycosylated soy protein isolate with maltose
Maillard reaction
formed by irradiation of 7.5 kGy and 12.5 kGy (named SPI-M7.5 and SPI-M12.5, respectively) was compared.
Emulsion
Freeze-thaw stability
Fourier transform infrared spectroscopy, fluorescence spectroscopy and ultraviolet spectroscopy confirmed the
change of the structure of soy protein isolate, indicating that maltose was covalently linked to soybean protein
isolate. Scanning electron microscopy showed that the modified protein particles were more loose, uniform in
size, and significantly reduced in molecular aggregation than untreated protein. The freeze-thaw stability of SPI,
SPI þ M, and SPI-M7.5 and SPI-M12.5 emulsions was evaluated. It was found that after three freeze-thaw cycles,
the creaming index, oiling off, particle size, coalescence degree and flocculation degree of emulsions prepared
with irradiated SPI samples were all lower than those prepared with SPI and SPI þ M. Zeta potential, laser
confocal and optical microscopy showed that the emulsion was still in a relatively stable state, and the SPI-M7.5
emulsion after freeze-thaw treatment was more stable than the SPI-M12.5 emulsion.

1. Introduction between the free amino group of the amino acid side chain of the protein
molecule and the carbonyl group at the reducing end of the saccharide
Due to its amphiphilic nature, soy protein isolate can be adsorbed at molecule, which belongs to the category of chemical modification,
the oil-water interface, providing electrostatic repulsion and steric namely the protein-saccharide grafting polymerization. It not only re­
hindrance between oil droplets to stabilize the emulsion, so it is often tains the surface activity of protein but also has the hydrophilic property
used as an emulsifier (Diftis, Biliaderis, & Kiosseoglou, 2005; Puppo, of saccharide, which is one of the ideal methods for protein modification
Beaumal, & Chapleau, 2008). However, protein-stabilized emulsions are (Dickinson, 1999). Irradiation is an emerging cold treatment method,
susceptible to instability under the influence of external conditions such which can ensure the original flavor of food to the maximum extent, and
as pH, ionic strength, and temperature (Dickinson, 2008; Pongsa­ has the characteristics of low energy consumption, high efficiency, and
watmanit, Harnsilawat, & Mcclements, 2006). When a little damage to nutrients (Kuan, Bhat, Patras, & Karim, 2013). In 2003,
protein-stabilized emulsion is freeze-thawed, a series of physicochem­ the Codex Alimentarius Commission stipulated that the irradiation dose
ical changes occur due to the formation of ice crystals. Sharp ice crystals could be could be more than 10 kGy under the condition of ensuring the
will penetrate into the droplets to destroy the interface film, and the integrity, functionality and safety of the food structure, which provided
protein will adsorb onto the ice crystals at the same time, resulting in the guarantee for the application of irradiation technology in the food
insufficient protein adsorption. Further, when the emulsion is thawed, industry (Farkas & Moha �csi-Farkas, 2011). In addition to its application
flocculation, coalescence, unstable oil release, and even complete sep­ in food sterilization and preservation, irradiation can also change the
aration of the oil phase and the water phase may occur (Gu, Decker, & structure of protein molecules and thus change their functional prop­
Mcclements, 2007; Magnusson, Christer Ros� en, & Nilsson, 2011). erties. Research has shown that irradiation changed the conformation of
The glycosylation reaction is mainly based on the carbonyl reaction protein molecules and improved the solubility, emulsifying properties

* Corresponding author. College of Food Science, Northeast Agricultural University, Harbin, 150030, Heilongjiang, China.
** Corresponding author. College of Food Science, Northeast Agricultural University, Harbin, 150030, Heilongjiang, China.
E-mail addresses: [email protected] (Y. Wang), [email protected] (A. Zhang), [email protected] (X. Wang), [email protected] (N. Xu),
[email protected] (L. Jiang).

https://doi.org/10.1016/j.foodhyd.2020.105684
Received 24 July 2019; Received in revised form 10 January 2020; Accepted 16 January 2020
Available online 20 January 2020
0268-005X/© 2020 Elsevier Ltd. All rights reserved.
Y. Wang et al. Food Hydrocolloids 103 (2020) 105684

and foam stability of soybean protein isolate (Pednekar, Das, Rajalaksh, 7.5 kGy, 10 kGy, 12.5 kGy), and the optimal irradiation dose was 7.5
& Sharma, 2010). After irradiation, changes in the structure of peanut kGy. In order to prove that the higher dose would affect the freeze-thaw
protein led to changes in its properties, the protein molecular structure stability of the emulsion, the experimental group with the irradiation
was unfolded, hydrophobic groups were exposed, and various functional dose of 12.5 kGy was added (Wang et al., 2019). The irradiated reaction
properties were improved. However, when the irradiation dose was solution was freeze-dried to prepare the lyophilized powder. The
further increased, the antigenicity of allergenic protein decreased (Luo, lyophilized powder obtained was the glycosylated soy protein isolate
Hu, Gao, Li, Wu, & Yang et al., 2013). It could be seen that the irradi­ formed by irradiation treatment at 7.5 and 12.5 kGy, which were named
ation mainly caused a series of reactions such as deamination, decar­ as SPI-M7.5 and SPI-M12.5, respectively.
boxylation, amino acid oxidation, cleavage or reconstitution of disulfide
bonds, degradation or cross-linking of peptide chains. The advanced 2.4. Degree of grafting
structure of protein molecules changed, which influenced on the func­
tional characteristics of protein molecules (Kuan et al., 2013; Wihodo & The free amino group was determined by the OPA method (Jiang,
Moraru, 2013). Wang, Wu, & Wang, 2013; Vigo, Malec, Gomez, & Llosa, 1992). After
There are currently few studies on the combination of radiation and dissolving 40 mg of o-phthalaldehyde (OPA) in 1 mL of methanol, the
glycosylation modifications. Due to the influence of irradiation on pro­ 25 mL of 0.1 mol/L sodium tetraborate, 2.5 mL of 20% (w/v) SDS and
tein structure and properties, it has been gradually developed into a 100 μL β-mercaptoethanol were added, respectively, and then the mixed
modification method (Kuan, Bhat, Patrasn, & Karim, 2013) and glyco­ solution was diluted to 50 ml with distilled water, which was OPA re­
sylation modification is also one of the commonly used methods for agent. During the measurement, 4 mL of OPA reagent was transferred to
protein modification (Zhang et al., 2017). There are few studies on the a test tube, and 200 μL of 5 mg/mL sample solution was injected, mixed
combination of irradiation and glycosylation modification. In this in a vortex mixer, and reacted in a water bath at 35 � C for 2 min. With
experiment, irradiation combined with Maillard reaction was used to distilled water as the blank, the absorbance was read at 340 nm. The
improve the freeze-thaw stability of soybean protein isolate. The struc­ grafting degree (DG) was calculated as follows:
ture and properties of the modified product were analyzed by Fourier
transform infrared spectroscopy, fluorescence spectroscopy, ultraviolet DG ¼
A1 A2
� 100%
spectroscopy, scanning electron microscopy, emulsification index, oiling A1
off, particle size, flocculation and coalescence, zeta potential, optical
microscopy and laser confocal microscopy. The study will enrich and Where, A1 is the absorbance of the sample before the graft reaction; A2 is
develop the protein modification system, providing a theoretical basis the absorbance of the sample after the graft reaction.
for the application of comprehensive modification technology in food
processing. 2.5. Browning degree

2. Materials and methods A 2 mg/ml sample solution was prepared, and water was used as a
blank control. The absorbance was measured at a wavelength of 420 nm,
2.1. Materials and the degree of browning of soybean glycoprotein was expressed by
absorbance (Kato & Fujimaki, 1968).
Low temperature defatted soy flour (Soybean variety was Nannong
86–4), Shandong Yuwang Company; maltose, Beijing Chemical Reagent 2.6. Fourier transform infrared spectrum (FTIR)
Factory; Jiu San soybean oil, commercially available; all other reagents
were analytically pure. An appropriate amount of sample was mixed with a certain amount
of potassium bromide, ground into powder, and pressed into thin slices
2.2. Preparation of soy protein isolate under the pressure of 20 MPa. Resolution was set at 4 cm 1, and 32 times
of scanning were performed. Then FTIR-8400S Fourier transform
100 g of defatted soy flour was dissolved in 1000 ml of deionized infrared spectrometer was used for full-band (4000-400 cm 1) scanning
water by the magnetic stirring (600 rpm, 2 h, 25 � C) and mixed well. The (Du, Huang, Wang, & Xiao, 2018).
pH of the mixed solution was adjusted to 8.5 with 2 mol/L NaOH, and
the mixed solution was then centrifuged (3170�g, 20 min, 4 � C). The
supernatant was adjusted to pH 4.5 with 2 mol/L HCl, and allowed to 2.7. Fluorescence spectroscopy
stand overnight at 4 � C. The supernatant was discarded and the pellet
was centrifuged (3170�g, 20 min, 4 � C), and washed three times with The sample was dissolved in 0.01 mol/L pH 7.0 sodium phosphate
deionized water. After the precipitate was redissolved, the pH was buffer solution and prepared into a sample solution with a protein
adjusted to 7.0 with 2 mol/L NaOH, and then lyophilized to obtain the concentration of 0.5% (w/v). The excitation wavelength was set at 347
soy protein isolate, and the protein mass fraction was determined to be nm and the scanning range of emission wavelength was 375–550 nm
90.23% by Kjeldahl method (Yu et al., 2018). (Morales & Jim�enez-P�erez, 2001). The sample solution was then diluted
20 times to 0.025% (w/v). The excitation wavelength was set to 290 nm,
2.3. Preparation of irradiated glycosylated soybean protein isolate and the scanning range of emission wavelength was 300–400 nm. Both
excited and emitted slits were 5.0 nm, and the scanning speed was 240
Soy protein isolate (SPI) and maltose (M) were dissolved in 0.01 nm/min (Taboada, Barbosa, Castro, Guti� errez-Pichel, & Mosquera,
mol/L pH 7.0 sodium phosphate buffer at a ratio of 4:1, and stirred at 25 2007).
�
C for 4 h to prepare a solution with a protein concentration of 4%. The
0.02% (w/v) sodium azide was added to the mixture solution to prevent 2.8. UV–visible spectroscopy
microbial growth and to prevent spoilage. The mixture solution was
hydrated overnight in a refrigerator at 4 � C. After the next day, it was The sample was dissolved in 0.01 mol/L pH 7.0 sodium phosphate
taken out and stirred until the liquid temperature reached 25 � C. 100 mL buffer solution to prepare a sample solution with a protein concentration
of the mixture solution was placed in an irradiation apparatus of 60Coγ of 0.02% (w/v), and then subjected to ultraviolet spectrum scanning, the
radiation source to carry out an irradiation treatment. We selected wavelength range was 190–500 nm, and the scanning rate was 100 nm/
irradiation doses according to our previous research (2.5 kGy, 5 kGy, min. The resolution is 0.2 nm (Liang & Tang, 2013).

2
Y. Wang et al. Food Hydrocolloids 103 (2020) 105684

2.9. Scanning electron microscopy

D4; 3 D4; 3SDS
FDð%Þ ¼ � 100
D4; 3SDS
After the appropriate amount of sample powder was uniformly
adhered to the observation table, gold was sprayed on the surface of the D4; 3SDSf t D4; 3SDS
protein sample by an ion sputter coater. The field emission scanning CDð%Þ ¼
D4; 3SDS
� 100
electron microscopy was used for observation, and the acceleration
voltage was 5 kV. where D4,3 and D4,3SDS were the volume-weighted mean diameter of
emulsion diluted in sodium phosphate buffer without and with SDS,
2.10. Preparation of emulsions respectively and D4,3SDSf-t was the volume-weighted mean diameter of
freeze-thawed emulsions previously diluted in sodium phosphate buffer
Samples were dissolved in 0.01 mol/L pH 7.0 sodium phosphate with SDS (Zhu, Zhang, Lin, & Tang, 2017).
buffer solution and 10% soybean oil was added, which was placed in a
high-speed homogenizer and treated for 3 min at 11000 rpm to form 2.14. Zeta potential
crude emulsion. The crude emulsion was then emulsified once at 60 MPa
using a high-pressure homogenizer to obtain the microemulsion. The The sample emulsion was dispersed in deionized water at a mass
0.02% sodium azide was added to prevent microbial contamination. fraction of 0.005%. The zeta potential of the sample emulsion was
measured at room temperature using a Zetasizer Nano Z-type potenti­
2.11. Creaming index ometer. The wavelength and scattering angle were fixed at 632 nm and
90� , respectively, and each sample was measured in parallel three times.
The freshly prepared sample emulsion was transferred to a test tube,
stored in a refrigerator at 20 � C for 22 h, and then thawed in a 20 � C 2.15. Optical microscopy
water bath for 2 h. The freeze-thaw cycle was repeated 3 times (Pala­
zolo, Sobral, & Wagner, 2011). After freeze-thaw treatment, the height After shaking the sample emulsion to be tested, 8 μL of the emulsion
of the transparent or turbid serum layer and the total height of emulsions was placed on a glass slide, covered with a cover glass, placed in a mi­
were measured with a ruler. The freeze-thaw stability of emulsions was croscope observation area, and the eyepiece was adjusted to 400 times
expressed by the creaming index (CI). The creaming index (CI) was for observation.
calculated as (Surh, Decker, & Mcclements, 2006):
Hs 2.16. Laser confocal microscopy
CI ¼ � 100%
Ht
1 mL sample emulsion sample was added to 40 μL of Nile Blue and
Where, Hs is the height of whey layer (cm); Ht is the total height (cm) of Nile Red staining solution for 30 min, and 10 μL of the stained sample
the emulsion. emulsion was placed on a clean glass slide, covered with a cover glass
and sealed with glycerin. Laser confocal scanning was performed at an
excitation wavelength of 488 nm, and an oil mirror was used for image
2.12. Oiling off
acquisition.
1000 g of soybean oil and 0.015 g of Sudan III were mixed well to
form Sudan III oil solution. The emulsion and Sudan III oil solution were 2.17. Data processing
accurately placed in a centrifuge tube at 4:1, and centrifuged at
16000�g for 20 min and then measured at 508 nm with soybean oil as All experiments were repeated three times, and the experimental
the blank (Palanuwech, Potineni, Roberts, & Coupland, 2003). The data were analyzed by statistical analysis and analysis of variance using
oiling off was calculated as: IBM SPSS Statistics 20 software. The data were expressed as mean �
standard deviation. The Origin 8 software was used for mapping, and the
Φ¼
m0 � ða 1Þ
� 100% peak spectrum of the amide I band was analyzed by Peak Fit 4.12 soft­
me ϕd ware. The different letters in the chart indicated significant difference (P
< 0.05), and the same letters indicated no significant difference (P >
Where: m0—Sudan III oil solution mass; me—emulsion quality; a—the 0.05).
ratio of the absorbance of Sudan III oil solution to the absorbance of
Sudan III oil solution after centrifugation; φd—the mass fraction of the 3. Results and discussion
oil phase in the emulsion.
3.1. Grafting degree and browning index analysis of modified products
2.13. Particle size, flocculation degree and coalescence degree
The essence of Maillard reaction between protein and saccharide was
The emulsion to be tested was divided into two equal parts and the reaction between the free amino group in the amino acid side chain
diluted by 0.01 mol/L pH 7.0 sodium phosphate buffer and 1% (w/v) of protein molecules (mainly the ε-amino group on the lysine side chain)
SDS for 5 times respectively. Approximately 1000 mL of ultrapure water and the reduced carbonyl group of reducing saccharide molecules (Kato,
was poured into the beaker. The pump of the sample processing unit was Sasaki, Furuta, & Kobayashi, 1990). The Maillard reaction could be
turned on and the pump speed was adjusted to 2000 RPM. The emulsion further determined by the content of the free amino group, so deter­
was then dispersed in deionized water to achieve a shading degree of mining the degree of grafting in the reaction system could reflect the
10%. The laser particle size analyzer (British Malvern Instrument Co., change in the free amino group of the soy protein isolate and understand
Ltd., Worcestershire, UK) was used to determine the particle size dis­ the modification of the amino acid (Vhangani & Van Wyk, 2013). It
tribution, surface-weighted mean diameter (D3,2) and volume-weighted could be seen from Fig. 1 that the graft of soybean protein isolate and
mean diameter (D4,3). Parameters: the absorption index of particles maltose (SPI-M7.5) formed by irradiation at a dose of 7.5 kGy had a
(0.001), particle refractive index (1.460), dispersant refractive index grafting degree of 33.76% and a degree of browning of 0.24. The graft of
(1.330). The flocculation degree (FD%) and coalescence degree (CD%) soybean protein isolate and maltose (SPI-M12.5) formed by irradiation at
were calculated as: a dose of 12.5 kGy had a grafting degree of 27.71% and a degree of

3
Y. Wang et al. Food Hydrocolloids 103 (2020) 105684

B DG
35 0.40
OA
A 0.35
30

0.30
25 a
Degree of grafting(%)

Browning index
0.25
20
0.20
15
0.15

10
0.10

5 0.05

0 0.00
SPI-M7.5 SPI-M12.5
Samples

Fig. 1. Grafting degree and browning index of SPI-M7.5 and SPI-M12.5. Different
letters indicate significant differences between different samples, P < 0.05.

browning of 0.26. It indicated that irradiation could promote the Mail­

lard reaction between soy protein isolate and maltose, and consume
some amino acids. When the degree of grafting changed, it was often
accompanied by browning. At the later stage of the Maillard reaction,
advanced stage products such as melanoidin were formed, causing an
increase in the degree of browning. The serious browning could
adversely affect the sensory color of certain products, so browning was
also an important indicator of the degree of Maillard reaction and
product quality. As can be seen from the figure, the degree of browning
of SPI-M7.5 and SPI-M12.5 was relatively low. Although the degree of
browning of SPI-M12.5 was slightly increased, the difference was not
significant. Therefore, appropriate irradiation dose was beneficial to
both the occurrence of Maillard reaction and the degree of Maillard
reaction.

3.2. Structural characterization of modified products

Infrared spectroscopy was an important method for the qualitative

analysis of organic structures. A typical feature was the increase in the
number of hydroxyl groups. On the infrared spectrum, there was a wide
stretching vibration peak at a wavenumber of 3700–3200 cm 1, and
strong vibration occurs at a wavenumber of 1260–1000 cm 1 (Oliver,
Kher, Mcnaughton, & Augustin, 2009; Turner et al., 2002). It could be
seen from Fig. 2(a) that both SPI-M7.5 and SPI-M12.5 had broadening
absorption peaks at 3700-3200 cm 1, indicating that the number of
hydroxyl groups increased after the maillard reaction occurred, causing
vibration absorption. At the same time, the absorption peak of the re­
action product at 1260-1000 cm 1 was significantly enhanced, indi­
cating an increase in the number of C–N. And the absorption vibration of
SPI-M7.5 and SPI-M12.5 was stronger than SPI, and SPI-M7.5 has stronger
absorption vibration than SPI-M12.5. FTIR could reflect the change of
peptide chain structure, which affected the conformation stability of
protein molecules. It could be seen from Table 1 that, compared with
SPI, the content of α-helix was reduced, and the content of β-turn and
random coil were increased. The secondary structure of SPI-M7.5
changed the most, indicating that protein changed from ordered to
disordered state and molecular flexibility increased.
The occurrence of Maillard reaction could be assessed by fluores­
cence spectroscopy. Its typical characteristic fluorescence spectrum was Fig. 2. Fourier transform infrared (a), fluorescence (b) (c) and ultraviolet (d)
that the excitation wavelength was 320–370 nm, and the emission spectra of SPI, SPI þ M, SPI-M7.5, and SPI-M12.5 (The excitation wavelength of
wavelength had the maximum fluorescence intensity in the range of figure b was set to 347 nm; the excitation wavelength of figure c was set to 290
420–440 nm (Liu, Zhao, Zhao, Ren, & Yang, 2012). It could be seen from nm; figure d scan wavelength range was 190–500 nm).
Fig. 2(b) that the excitation wavelength was 347 nm for scanning, SPI
and SPI þ M had maximum fluorescence intensity at 445 nm, while

4
Y. Wang et al. Food Hydrocolloids 103 (2020) 105684

Table 1 groups, thereby increasing the contact probability of free amino and
Secondary structure contents of SPI, SPI þ M, SPI-M7.5, and SPI-M12.5. maltose carbonyl groups, and accelerating the Maillard reaction of
Sample α-helix/% β-sheet/% β-turn/% random coil/% protein and saccharide.
c a a Protein spatial structure reflected the aggregation state of protein
SPI 18.87 � 0.21 43.14 � 0.25 17.51 � 0.26 20.48 � 0.56a
SPI þ M 18.79 � 0.18c 43.96 � 0.13b 17.70 � 0.32a 19.55 � 0.58a molecules and played an important role in protein function. Fig. 3
SPI-M7.5 15.77 � 0.06a 43.85 � 0.21b 18.74 � 0.14b 21.64 � 0.10b showed the scanning electron micrograph of SPI, SPI þ M, and SPI-M7.5
SPI-M12.5 16.11 � 0.17b 44.02 � 0.41b 18.39 � 0.02c 21.48 � 0.02c and SPI-M12.5 lyophilized powder at 250 � and 8000 � magnification. It
Note: Different lowercase letters in the same column indicated significant dif­ could be seen from Fig. 3 that the unmodified SPI particles had compact
ferences (P < 0.05). structure, sharp edges and different sizes. When sugar molecules were
added, the small molecules of sugar tightly surrounded the protein
SPI-M7.5 and SPI-M12.5 had maximum fluorescence intensity at 425 nm molecule to make the structure more compact. The irradiated SPI-M7.5
and 430 nm, respectively. The λmax (Absorption wavelength with and SPI-M12.5 particles were more loose, fuzzy and even uniform. After
maximum absorption peak) shifted to the short wavelength direction, a 8000 times magnification, it was found that SPI-M7.5 and SPI-M12.5 were
blue shift occurred, and the fluorescence intensity of the reaction loose and porous, and the evacuation was uniform and network-like. It
product was higher than SPI, indicating that the Maillard reaction might be because the irradiation promoted the protein structure
occurred. Tryptophan mainly affected the endogenous fluorescence of unfolding (Kuan et al., 2013; Wihodo & Moraru, 2013), the access of the
SPI and was very sensitive to environmental changes, so it reflected the sugar molecule caused the surface of the protein molecule to be covered
conformational changes of proteins faster (Matiacevich & Buera, 2006). with the thick layer of sugar coating, the original rigid structure of the
It could be seen from Fig. 2(c) that when the excitation wavelength was protein molecule disappeared, and the degree of aggregation between
290 nm, the λmax was shifted to the long-wavelength direction of the proteins was significantly reduced.
SPI-M7.5 and SPI-M12.5 compared with SPI, and different degrees of red
shift occurred, indicating that the isolated soy protein and the maltose
underwent a deep Maillard reaction with a decrease in fluorescence 3.3. Creaming index and oiling off analysis of modified products
intensity. It might be that the irradiation caused the protein structure to
open, exposing more chromophores to fluorescence quenching and When the emulsion was frozen, crystallization occurred in the oil and
reducing the fluorescence intensity. water phases. After thawing, the droplets gathered together, and insta­
The ultraviolet absorption spectrum of proteins was mainly due to bility phenomena such as droplet coalescence, oil phase separation and
the absorption of ultraviolet light by tryptophan and tyrosine residue sedimentation occurred (Ghosh & Coupland, 2008). Freeze-thaw sta­
side chain groups, and the conformational changes of proteins could be bility of emulsion referred to the stability of emulsion during
inferred from the differences in the absorption of ultraviolet spectrum of freeze-thaw process, which was an important property of protein
proteins (Wang, Sun, Pu, & Wei, 2017). It could be seen from Fig. 2(d) emulsion. The creaming index was one of the important indexes of
that the SPI and SPI þ M ultraviolet spectra almost overlapped, showing freeze-thaw stability, the lower the creaming index, the more stable
two characteristic absorption peaks in the range of 220–240 nm and freeze-thaw stability (Xiaodan et al., 2018). It could be seen from Fig. 4
260–280 nm. They were the absorption peaks of peptide bonds and the (a) that the creaming index of both SPI and SPI þ M emulsions increased
absorption peaks of phenylalanine, tryptophan and tyrosine, respec­ sharply after the freeze-thaw cycles, indicating that the simple addition
tively, indicating that simple mixed sugar did not change the structure of of sugar did not improve the freeze-thaw stability of the emulsion. The
the protein (Yu, Zhao, Hu, Zeng, & Bai, 2012). The absorption peaks of SPI-M7.5 and SPI-M12.5 emulsions had relatively stable creaming index
SPI-M7.5 and SPI-M12.5 at 260–280 nm almost disappeared, indicating after undergoing freeze-thaw cycles, showing good freeze-thaw stabil­
that new compounds might be formed, and soy protein isolate and ity. After the emulsion freeze-thaw treatment, the separation of the oil
maltose were already in the state of bonding. The addition of the sugar phase and the water phase occurred due to the difference of density
chain enlarged the spatial structure of the reaction product, resulting in between the fat globule and the continuous phase, and the stability of
weak absorption of tryptophan and tyrosine, and no obvious charac­ the emulsion was characterized by oiling off (Oliver et al., 2009). Ac­
teristic absorption peak. At the same time, the absorption peak at cording to Fig. 4(b), after freeze-thaw cycles, the oiling off of SPI and
220–240 nm was shifted to around 200 nm. The absorption peak was SPI þ M emulsion was significantly increased. Although SPI þ M
blue-shifted, and the absorption intensity was weakened. Irradiation decreased compared with SPI, the overall improvement was not signif­
made the protein structure unfold, increasing the content of free amino icant. The emulsions prepared with irradiated SPI samples had lower
oiling off, but SPI-M7.5 emulsion was more stable than SPI-M12.5

Fig. 3. Scanning electron microscopy (SEM) images of SPI (a) (e), SPI þ M (b) (f), SPI-M7.5 (c) (g) and SPI-M12.5 (d) (h). (a–d) at magnification of 250 � , (e–h) at
magnification of 8.00k � .

5
Y. Wang et al. Food Hydrocolloids 103 (2020) 105684

Cycle1 Cycle1
(a) 80 Cycle2 (b) 40 Cycle2
Cycle3 Cycle3
Dc
Dc
Db Cc Cc
Cb
60 30
Da Db
Ca
Cb
Creaming index(%)

Bc

Oiling off(%)
40
Bb
Ba 20
Ab Ac
Aa
Da
Ca Bc
20 10 Ac Bb
Ab Ba
Aa

0 0
SPI SPI+M SPI-M7.5 SPI-M12.5 SPI SPI+M SPI-M7.5 SPI-M12.5
Samples Samples

Fig. 4. Creaming index (a) and oiling off (b) of SPI, SPI þ M, SPI-M7.5 and SPI-M12.5 emulsions after freeze-thaw treatment. Different uppercase letters indicated
significant differences between different samples at the same cycle number and different lowercase letters indicated significant differences between the different
cycles of the same sample, P < 0.05.

emulsion. It was possible that appropriate irradiation improved emul­

Table 2
sification, thereby reducing interfacial tension, reducing collision be­
Volume-weighted mean diameter (D4,3), Surface-weighted mean diameter
tween oil droplets, and stabilizing the emulsion (Song et al., 2009).
(D3,2), flocculation degree, and coalescence degree of SPI, SPI þ M, SPI-M7.5, and
SPI-M12.5 emulsions.
3.4. Particle size and stability analysis of modified products Freeze- Samples Volume- Surface- Flocculation Coalescence
thaw weighted weighted degree FD/% degree CD/%
The particle size of the emulsion was the key factor affecting the cycles mean mean
stability of the emulsion. Generally speaking, the smaller the average diameter diameter
D4,3/μm D3,2/μm
droplet size of the protein emulsion, the more stable the emulsion
(McClements, 2007). As shown in Table 2, the D4,3 of the initial emulsion Cycle 0 SPI 3.2 � 0.5a, 2.4 � 0.2a, 20.7 � 0.7d,A –
A A
SPI þ M, SPI-M7.5 and SPI-M12.5 increased slightly due to the addition of SPI þ M 14.6 � 0.1c,A –
a, a,
SPI- 3.4 � 0.4 2.4 � 0.6 10.7 � 0.0a,A –
maltose molecules. With the increase of freeze-thaw cycles, the particle M7.5 A A
12.1 � 0.7b,A –
size of SPI and SPI þ M emulsion increased significantly, and the addi­ SPI- 4.8 � 0.0b, 2.3 � 0.1a,
tion of maltose did not slow down the increase of particle size of SPI þ M M12.5 A A

emulsion. While the emulsion prepared by SPI-M7.5 and SPI-M12.5 5.8 � 0.2c, 2.3 � 0.3a,
A A
experienced a small increase in particle size after the freeze-thaw cycles,
Cycle 1 SPI 28.8 � 18.6 � 34.0 � 0.6c,B 82.8 � 0.3d,A
and the stability of SPI-M7.5 emulsion was better than that of SPI-M12.5 SPI þ M 0.6d,B 0.2c,B 33.9 � 0.2c,B 81.9 � 0.1c,A
emulsion. Compared with the SPI control group, the volume-weighted of SPI- 16.8 � 16.8 � 20.1 � 0.3a,B 27.3 � 0.6a,A
SPI-M7.5 emulsion decreased by 19.3 μm, 21.2 μm and 43.4 μm after M7.5 0.9c,B 0.8b,B 23.6 � 0.3b,B 28.5 � 0.2b,A
three times freeze-thaw cycles, which was consistent with the results of SPI- 9.5 � 0.2a, 2.9 � 0.3a,
B B
M12.5
the creaming index. The D4,3 was more sensitive to the process of 11.8 � 3.1 � 0.3a,
emulsion droplet aggregation than the D3,2, so the change of D4,3 was 0.2b,B B

used to evaluate the degree of flocculation and coalescence of the Cycle 2 SPI 32.2 � 25.8 � 70.0 � 0.8c,C 100.7 � 0.6d,
emulsion (Fioramonti, Arzeni, Pilosof, Rubiolo, & Santiago, 2015; SPI þ M 0.8d,C 0.3c,C 68.4 � 1.0c,C B

SPI- 28.5 � 20.0 � 28.8 � 0.0a,C 97.7 � 0.8c,B

Zhang et al., 2017). The FD% was obtained both for unfrozen and
M7.5 0.0c,C 0.4b,C 30.2 � 0.8b,C 39.4 � 0.6a,B
freeze-thawed emulsions while CD% was obtained only for SPI- 11.0 � 2.6 � 0.1a, 41.4 � 0.1b,B
freeze-thawed emulsions. As can be seen from Table 2, the initial SPI and M12.5 0.3a,C B

SPI-M emulsions showed the FD of 20.7% and 14.6%, respectively, while 12.7 � 3.3 � 0.6a,
the SPI-M7.5 and SPI-M12.5 emulsions only had the FD of 10.7% and 0.5b,C B

Cycle 3 SPI 54.4 � 49.3 � 97.0 � 0.6d,D 131.8 � 0.2d,

12.1%. With the increase of the number of freeze-thaw cycles, the degree SPI þ M 0.5d,D 0.5c,D 94.1 � 0.7c,D C

of flocculation and the degree of coalescence increased continuously. SPI- 51.5 � 29.7 � 38.5 � 0.2a,D 130.4 � 0.3c,
After three freeze-thaw cycles, SPI-M7.5 and SPI-M12.5 emulsions still M7.5 1.0c,D 0.4b,D 39.1 � 0.2b,D C

maintained good freeze-thaw stability relative to the control. It might be SPI- 11.0 � 2.7 � 0.2a, 41.6 � 0.4a,C
M12.5 0.5a,C B
43.8 � 0.5b,C
because the irradiation made the tight structure of SPI unfold, the
13.2 � 3.4 � 0.5a,
flexibility was increased, the emulsifying property was improved, the 1.0b,C B

thick adsorption layer could be formed, and the spatial barrier was
Note: a-d represents the significant differences between same freeze-thaw cycles
formed between the droplets, thereby weakening the droplet
of the different samples; A-D represents the significant differences between
flocculation.
different freeze-thaw cycles of the same samples (p < 0.05).

3.5. Potential and microstructure analysis of modified products emulsions, had a similar microstructure. After the freezing-thawing
cycles, SPI and SPI þ M emulsion gradually appeared large and round
Fig. 5 showed the optical microstructures of SPI, SPI þ M, SPI-M7.5 oil droplets and irregular clumps, showing poor freezing-thawing
and SPI-M12.5 initial emulsions and emulsions after three cycles of characteristics. When the emulsion was frozen, the water phase
freeze-thawing. It could be seen that the four initial, unfrozen,

6
Y. Wang et al. Food Hydrocolloids 103 (2020) 105684

Fig. 5. Optical microstructure of SPI, SPI þ M, SPI-M7.5 and SPI-M12.5 emulsions before and after freeze-thaw treatment. The scale bar was 50 μm in length.
Magnification was 400 times.

formed ice crystals puncturing membrane, which caused the fat crystals freeze-thaw stability of emulsions (Kuan, Bhat, & Karim, 2011).
to penetrate each other between the oil droplets, and the oil droplets Laser confocal microscopy could visually represent the particle size,
gathered after thawing (Aoki, Decker, & Mcclements, 2005). The irra­ dispersion and stability of the emulsion. Fig. 6 showed the laser confocal
diated SPI-M7.5 and SPI-M12.5 emulsions produced only a small amount microstructure of SPI, SPI þ M, SPI-M7.5 and SPI-M12.5 emulsions before
of oil droplets, probably because irradiation promoted the soy protein and after freezing and thawing. The initial emulsions droplet distribu­
molecular structure to unfold and increased the Maillard reaction rate tion of the four samples were uniform and the microstructure were not
with maltose. The flexibility of the molecules was increased, and more significantly different. After the freezing-thawing cycles, the oil droplets
hydrophobic groups were exposed at the same time, which enhanced the were observed in SPI and SPI þ M protein emulsions. It might be due to
adsorption of the oil-water interface, so that the protein molecules could the poor solubility of untreated SPI and SPI þ M, and the protein fraction
rapidly develop at the interface to form the solid viscoelastic layer, and could not quickly adsorb to the oil-water interface, and could not form
the interfacial tension was also rapidly reduced, thereby increasing the the thick protective film (Palazolo et al., 2011). Irradiation treatment

SPI SPI+M SPI-M7.5 SPI-M12.5

Cycle 0

Cycle 3

Fig. 6. Laser confocal microstructure before and after freeze-thaw of SPI, SPI þ M, SPI-M7.5 and SPI-M12.5 emulsions. The scale bar was 50 μm in length. The
excitation wavelength was set to 488 nm.

7
Y. Wang et al. Food Hydrocolloids 103 (2020) 105684

made the soy protein isolate form the complex with maltose. The Cycle 0 Cycle 1 Cycle 2 Cycle 3
emulsion particles of SPI-M7.5 and SPI-M12.5 complex were evenly 0

dispersed, and the oil droplets were tightly surrounded by protein. It

-5
might be because the irradiation treatment improved the emulsification
properties of the protein, accelerated the Maillard reaction with the
-10
sugar, and rapidly formed the interface film on the surface of the oil
droplet in the oil-water mixed system (Song, Kim, Choe, Jung, Kim, &
-15
Kim et al., 2009), effectively suppressing coalescence and flocculation

ZETA/mV
between oil droplets at low temperature state, thereby improving the -20 Cd Cd
freeze-thaw stability of the emulsion. Cc Cc
The potential was an important index for judging the stability of the -25
emulsion. The smaller the dispersed particles, the higher the absolute Cb Bb
Ca Ca Bc
value of the potential, the greater the repulsive force between the par­ -30 Ab Bb Ac
ticles, and the more stable the system. Moreover, the emulsion with the Aa
Ba Aa Aa
SPI
absolute value of zeta potential exceeding 30 mV was stable, indicating -35 SPI+M
that there was a large electrostatic repulsion between particles (Mar­ SPI-M7.5
kiewicz, Mrozik, & Rezwan, 2013; Vallar, Houivet, El Fallah, Kervadec, -40
SPI+M12.5
& Haussonne, 1999). In Fig. 7, the initial emulsions prepared from SPI,
SPI þ M, SPI-M7.5, and SPI-M12.5 had potentials of 26.5 mV, 26.6 mV, Fig. 7. Zeta potential of SPI, SPI þ M, SPI-M7.5 and SPI-M12.5 emulsions before
30.2 mV, and 29.4 mV, respectively. As the number of freeze-thaw and after freeze-thaw treatment. Different uppercase letters indicated signifi­
cycles increased, the absolute potential value of the emulsion system cant differences between different samples at the same cycle number and
with SPI and SPI þ M mixture decreased gradually, showing instability. different lowercase letters indicated significant differences between the
The absolute values of the irradiated SPI-M7.5 and SPI-M12.5 emulsions different cycles of the same sample, P < 0.05.
remained stable, and the SPI-M7.5 emulsion exhibited better freeze-thaw
stability than the SPI-M12.5 emulsion. The microstructure analysis of the Conceptualization, Writing - review & editing. Ning Xu: Resources.
emulsion also showed that the soy protein emulsion obtained by Mail­ Lianzhou Jiang: Project administration, Funding acquisition.
lard reaction promoted by irradiation had good freeze-thaw stability.
Acknowledgements
4. Conclusion
This paper was founded by the National Key R&D Program of China
In this paper, the soy protein isolate was modified by irradiation (2018YFD0400600), Heilongjiang Province Key S&T Program
technology and glycosylation technology to improve its freeze-thaw (2019ZX08B01) and National Soybean Industrial Technology System of
stability. Fourier transform infrared spectroscopy, fluorescence spec­ China (CARS-04-PS28). The authors are grateful to the editors and
troscopy and ultraviolet spectroscopy confirmed the change of the anonymous reviewers for their valuable comments and suggestions.
structure of soy protein isolate, which proved that the soy protein isolate
reacted with maltose. The freeze-thaw stability of SPI, SPI þ M, SPI-M7.5 References
and SPI-M12.5 emulsions was evaluated by creaming index, oiling off,
Aoki, T., Decker, E. A., & Mcclements, D. J. (2005). Influence of environmental stresses
degree of flocculation, degree of coalescence, particle size, and Zeta
on stability of o/w emulsions containing droplets stabilized by multilayered
potential. Compared with SPI and SPI þ M, the emulsion prepared by membranes produced by a layer-by-layer electrostatic deposition technique. Food
irradiating SPI-M after 3 freeze-thaw cycles had lower creaming index Hydrocolloids, 19, 209–220.
Dickinson, E. (1999). Adsorbed protein layers at fluid interfaces: Interactions, structure
and oiling off, and the increase of particle size, flocculation degree and
and surface rheology. Colloids and Surfaces B, 15, 161–176.
coalescence degree was smaller. Zeta potential, laser confocal and op­ Dickinson, E. (2008). Interfacial structure and stability of food emulsions as affected by
tical microscopy indicated that the emulsion was still relatively stable. protein-polysaccharide interactions. Soft Matter, 4, 932–942.
Scanning electron microscopy studies showed that the modified protein Diftis, N. G., Biliaderis, C. G., & Kiosseoglou, V. D. (2005). Rheological properties and
stability of model salad dressing emulsions prepared with a dry-heated soybean
particles were more loose, uniform in size, and significantly reduced in protein isolate-dextran mixture. Food Hydrocolloids, 19, 1025–1031.
molecular aggregation. The emulsions prepared with irradiated SPI Du, Y. L., Huang, G. Q., Wang, H. O., & Xiao, J. X. (2018). Effect of high coacervation
samples had good freeze-thaw stability, and the SPI-M7.5 after freeze- temperature on the physicochemical properties of resultant microcapsules through
induction of maillard reaction between soybean protein and chitosan. Journal of Food
thaw treatment was more stable than the SPI-M12.5 emulsion. This Engineering, 234, 91–97.
study provided new ideas and references for the modification of soy Farkas, J., & Moh� acsi-Farkas, C. (2011). History and future of food irradiation. Trends in
protein isolate and the application of irradiation technology in the food Food Science & Technology, 22, 0–126.
Fioramonti, S. A., Arzeni, C., Pilosof, A. M. R., Rubiolo, A. C., & Santiago, L. G. (2015).
industry, which could promote the practical application of irradiation Influence of freezing temperature and maltodextrin concentration on stability of
technology and Maillard reaction in SPI modification. linseed oil-in-water multilayer emulsions. Journal of Food Engineering, 156, 31–38.
Ghosh, S., & Coupland, J. N. (2008). Factors affecting the freeze-thaw stability of
emulsions. Food Hydrocolloids, 22, 105–111.
Declaration of competing interest
Gu, Y. S., Decker, E. A., & Mcclements, D. J. (2007). Application of multi-component
biopolymer layers to improve the freeze-thaw stability of oil-in-water emulsions:
No conflict of interest exits in the submission of this manuscript, and β-lactoglobulin-ι-carrageenan-gelatin. Journal of Food Engineering, 80, 1246–1254.
Jiang, Z., Wang, L., Wu, W., & Wang, Y. (2013). Biological activities and physicochemical
manuscript is approved by all authors for publication. I would like to
properties of maillard reaction products in sugar-bovine casein peptide model
declare on behalf of my co-authors that the work described was original systems. Food Chemistry, 141, 3837–3845.
research that has not been published previously, and not under Kato, H., & Fujimaki, M. (1968). Formation of n-substituted pyrrole-2-aldehydes in the
consideration for publication elsewhere, in whole or in part. All the browning reaction between d-xylose and amino compounds. 33, 445–449.
Kato, A., Sasaki, Y., Furuta, R., & Kobayashi, K. (1990). Functional protein-
authors listed have approved the manuscript that is enclosed. polysaccharide conjugate prepared by controlled dry-heating of ovalbumin-dextran
mixtures. Agricultural & Biological Chemistry, 54, 107–112.
CRediT authorship contribution statement Kuan, Y. H., Bhat, R., & Karim, A. A. (2011). Emulsifying and foaming properties of
ultraviolet-irradiated egg white protein and sodium caseinate. Journal of Agricultural
and Food Chemistry, 59, 4111–4118.
Yuying Wang: Investigation, Formal analysis, Writing - original
draft. Anqi Zhang: Investigation. Xibo Wang: Supervision,

8
Y. Wang et al. Food Hydrocolloids 103 (2020) 105684

Kuan, Y. H., Bhat, R., Patras, A., & Karim, A. A. (2013). Radiation processing of food Surh, J., Decker, E. A., & Mcclements, D. J. (2006). Influence of ph and pectin type on
proteins-a review on the recent developments. Trends in Food Science & Technology, properties and stability of sodium-caseinate stabilized oil-in-water emulsions. Food
30, 105–120. Hydrocolloids, 20, 607–618.
Liang, H. N., & Tang, C. H. (2013). Emulsifying and interfacial properties of vicilins: Role Taboada, P., Barbosa, S., Castro, E., Guti�errez-Pichel, M., & Mosquera, V. (2007). Effect
of conformational flexibility at quaternary and/or tertiary levels. Journal of of solvation on the structure conformation of human serum albumin in aqueous-
Agricultural and Food Chemistry, 61, 11140–11150. alcohol mixed solvents. Chemical Physics, 340, 59–68.
Liu, Y., Zhao, G., Zhao, M., Ren, J., & Yang, B. (2012). Improvement of functional Turner, J. A., Sivasundaram, L. R., Ottenhof, M. A., Farhat, I. A., Linforth, R. S. T., &
properties of peanut protein isolate by conjugation with dextran through maillard Taylor, A. J. (2002). Monitoring chemical and physical changes during thermal
reaction. Food Chemistry, 131, 901–906. flavor generation. Journal of Agricultural and Food Chemistry, 50, 5406–5411.
Luo, C., Hu, C., Gao, J., Li, X., Wu, Z., Yang, A., et al. (2013). A potential practical Vallar, S., Houivet, D., El Fallah, J., Kervadec, D., & Haussonne, J. M. (1999). Oxide
approach to reduce ara h 6 allergenicity by gamma irradiation. Food Chemistry, 136, slurries stability and powders dispersion: Optimization with zeta potential and
1141–1147. rheological measurements. Journal of the European Ceramic Society, 19, 1017–1021.
Magnusson, E., Ros� en, C., & Nilsson, L. (2011). Freeze-thaw stability of mayonnaise type Vhangani, L. N., & Van Wyk, J. (2013). Antioxidant activity of maillard reaction products
oil-in-water emulsions. Food Hydrocolloids, 25, 707–715. (mrps) derived from fructose-lysine and ribose-lysine model systems. Food Chemistry,
Markiewicz, M., Mrozik, W., & Rezwan, K. (2013). Changes in zeta potential of 137, 92–98.
imidazolium ionic liquids modified minerals-implications for determining Vigo, M. S., Malec, L. S., Gomez, R. G., & Llosa, R. A. (1992). Spectrophotometric assay
mechanism of adsorption. Chemosphere, 90, 706–712. using o-phthaldialdehyde for determination of reactive lysine in dairy products. Food
Matiacevich, S. B., & Buera, M. P. (2006). A critical evaluation of fluorescence as a Chemistry, 44, 363–365.
potential marker for the maillard reaction. Food Chemistry, 95, 423–430. Wang, K. Q., Sun, D. W., Pu, H. B., & Wei, Q. Y. (2017). Principles and applications of
McClements, D. J. (2007). Critical review of techniques and methodologies for spectroscopic techniques for evaluating food protein conformational changes: A
characterization of emulsion stability. Critical Reviews in Food Science and Nutrition, review. Trends in Food Science & Technology, 67, 207–219.
47, 611–649. Wang, Y. Y., Zhang, A. Q., Wang, L., Zhou, G. W., Xu, N., & Wang, X. B. (2019).
Morales, F. J., & Jim�enez-P�erez, S. (2001). Free radical scavenging capacity of maillard Improving freeze-thaw properties of soybean protein emulsion by irradiation-
reaction products as related to colour and fluorescence. Food Chemistry, 72, 119–125. assisted grafting. Shipinkexue/Food Science. https://doi.org/10.7506/spkx1002-
Oliver, C. M., Kher, A., Mcnaughton, D., & Augustin, M. A. (2009). Use of ftir and mass 6630-20190527-325.
spectrometry for characterization of glycated caseins. Journal of Dairy Research, 76, Wihodo, M., & Moraru, C. I. (2013). Physical and chemical methods used to enhance the
105–110. structure and mechanical properties of protein films: A review. Journal of Food
Palanuwech, J., Potineni, R., Roberts, R. F., & Coupland, J. N. (2003). A method to Engineering, 114, 292–302.
determine free fat in emulsions. Food Hydrocolloids, 17, 55–62. Xiaodan, Z., Chonghui, Y., Mujun, L., Huanyu, Z., Xiufang, X., & Guoping, Y. (2018).
Palazolo, G. G., Sobral, P. A., & Wagner, J. R. (2011). Freeze-thaw stability of oil-in- Improvement of freeze-thaw stability of oil-in-water emulsions prepared with
water emulsions prepared with native and thermally-denatured soybean isolates. modified soy protein isolates. Lebensmittel-Wissenschaft und -Technologie- Food Science
Food Hydrocolloids, 25, 398–409. and Technology, 102, 122–130.
Pednekar, M., Das, A. K., Rajalakshmi, V., & Sharma, A. (2010). Radiation processing and Yu, J., Wang, G., Wang, X., Xu, Y., Chen, S., & Wang, X. (2018). Improving the freeze-
functional properties of soybean (glycine max). Radiation Physics and Chemistry, 79, thaw stability of soy protein emulsions via combing limited hydrolysis and maillard-
490–494. induced glycation. LWT-Food Science and Technology, 91, 63–69.
Pongsawatmanit, R., Harnsilawat, T., & Mcclements, D. J. (2006). Influence of alginate, Yu, X., Zhao, M., Hu, J., Zeng, S., & Bai, X. (2012). Correspondence analysis of
ph and ultrasound treatment on palm oil-in-water emulsions stabilized by antioxidant activity and uv-vis absorbance of maillard reaction products as related to
β-lactoglobulin. Colloids & Surfaces A Physicochemical & Engineering Aspects, 287, reactants. Lebensmittel-Wissenschaft und -Technologie- Food Science and Technology, 46,
59–67. 0–9.
Puppo, M. C., Beaumal, V., & Chapleau, N. (2008). Physicochemical and rheological Zhang, Z., Wang, X., Yu, J., Chen, S., Ge, H., & Jiang, L. (2017). Freeze-thaw stability of
properties of soybean protein emulsions processed with a combined temperature/ oil-in-water emulsions stabilized by soy protein isolate-dextran conjugates.
high-pressure treatment. Food Hydrocolloids, 22, 1079–1089. Lebensmittel-Wissenschaft und -Technologie- Food Science and Technology, 78, 241–249.
Song, H. P., Kim, B., Choe, J. H., Jung, S., Kim, K. S., Kim, D. H., et al. (2009). Zhu, X. F., Zhang, N., Lin, W. F., & Tang, C. H. (2017). Freeze-thaw stability of pickering
Improvement of foaming ability of egg white product by irradiation and its emulsions stabilized by soy and whey protein particles. Food Hydrocolloids, 69,
application. Radiation Physics and Chemistry, 78, 217–221. 173–184.