Received: 15 August 2019 

|   Revised: 7 January 2020 

|  Accepted: 5 February 2020

DOI: 10.1002/cche.10269

RESEARCH ARTICLE

Development of wall material for the microencapsulation of

natural vanilla extract by spray drying

Israel Oswaldo Ocampo-Salinas1   | Carlos Alberto Gómez-Aldapa1   |

Javier Castro-Rosas1   | Enaim Aída Vargas-León2   | Fabiola Araceli Guzmán-Ortiz1,3   |
Netzahualcoyol Calcáneo-Martínez1   | Reyna Nallely Falfán-Cortés1,3

1
Universidad Autónoma del Estado de
Hidalgo (UAEH), ICBI. Car, Hidalgo,
Abstract
México Background and objectives: The aim of this study was to develop a wall material
2
Tecnológico Nacional de México-Celaya, using rice starch (Oryza sativa) to optimize the microencapsulation of vanilla ex-
Celaya, Guanajuato, México
tract by spray drying. Native starch rice was hydrolyzed, succinated, extruded, and
3
Catedráticos, CONACYT. Car, Hidalgo,
characterized.
México
Findings: The peak viscosity of modified starches was reduced from 3,875  cP to
Correspondence <100  cP by effect of the modification step. Native starch was gelatinized by ex-
Reyna Nallely Falfan-Cortes, Catedrática
CONACYT, Universidad Autónoma del
trusion according to gelatinization enthalpy (∆H J/g). Natural vanilla extract was
Estado de Hidalgo 42183, Mineral de la encapsulated, and the most optimal conditions for the drying process were air inlet
Reforma, Hidalgo, México. temperature of 125°C, 14.0% of total solids, and 86.2% of encapsulation efficiency.
Email: nfalfan@hotmail.com
The scanning electron microscopy evidenced the formation of microcapsules; how-
ever, the confocal laser scanning microscopy showed no interaction between the wall
material and the vanilla extract. This could be related to the zeta potential values of
the vanilla extract and the wall material.
Conclusions: Modified rice starch is a potential wall material for the encapsulation
of different bioactive compounds in the food sector.
Significance and novelty: Chemical and physical modifications such as hydrolysis,
succinylation, and extrusion in native rice starch offer adequate properties to achieve
86.2% encapsulation efficiency of natural vanilla extract using spray drying.

KEYWORDS
microencapsulation, modified starch, rice, spray drying, vanilla

1  |   IN T RO D U C T ION the predominant flavor constituent of these extracts, contrib-

uting with a third of the whole flavor (De la Vega, Cervantes,
Vanilla is a widely accepted flavoring agent around the world Alvarado, Romero & Hegel, 2016; Sinha, Sharma & Sharma,
used in the preparation of a number of food products, phar- 2008). However, natural vanilla extracts are a complex mix-
maceuticals, and cosmetics. Genuine natural vanilla extracts ture of volatile and delicate substances of which over 170
are obtained from the percolation of Vanilla planifolia pods components have been identified (Sostaric, Boyce, & Spickett,
in an aqueous ethanol solution. This is important because 2000). Nonetheless, unlike artificial preparations, this ingre-
nowadays, an increasing number of customers demand natural dient is expensive and, just as any flavor, is susceptible to
flavorings due to health interest without sacrificing quality. oxidation and sensitive to heat- and light-induced reactions
The vanillin molecule (4-hydroxy-3-methoxybenzaldehyde) is (Milanovic et al., 2010; Wang, Yuan, & Yue, 2015). Given the

Cereal Chemistry. 2020;97:555–565. wileyonlinelibrary.com/journal/cche © 2020 Cereals & Grains Association     555 |
 |
556      OCAMPO-SALINAS et al.

above, flavor principles can be lost or degraded by processing material for microencapsulation by spray drying. Interestingly,
and storage. In addition, packaging materials, interaction with such microcapsules were, in some cases, formed for starch
other food compounds, and flavor–flavor reactions might re- aggregates. Beirão, Duarte, Moldão, and Beirão (2011), and
duce the intensity of the aroma or produce off-flavors. Such Surojanametakul et al. (2005) found good microcapsule ap-
changes in flavor could result in a lower quality and reduce the pearance in rice starch aggregates prepared by spray drying
acceptability of food products in consequence. The microen- together with binding agents. Trindade and Grosso (2000)
capsulation of flavor ingredients is widely useful to limit such obtained microcapsules of ascorbic acid with a high amount
undesirable actions and keep flavoring properties during food of recovery in rice starch aggregates with gelatine. Recently,
storage, providing protection, and controlling flavor release Márquez, Galicia, Márquez, Ruiz, and Quintero (2018) micro-
(Baranauskienė, Bylaitė, Žukauskaitė, & Venskutonis, 2007; encapsulated orange essential oil by spray drying without the
Madene, Jacquot, Scher, & Desobry, 2006). formation of aggregates as result of prior modification; they
To carry out the encapsulation, an active principle is cov- reported that the encapsulation efficiency was higher using
ered with a wall material which protects the active principle modified rice starch (acetylated starch) by extrusion than native
from adverse conditions that cause its degradation; for flavors, one (98.69 and 36.76%, respectively). Based on the above and
it is encapsulated to prevent evaporation and reaction within due to the importance of natural vanilla extract and the need to
a food. Different methods of encapsulation have been used generate new materials for encapsulation of flavor substances,
in the encapsulation of flavors, including spray, spray chill- the aim of this study was to develop a wall material using rice
ing or spray cooling, extrusion, freeze drying, coacervation, starch (Oryza sativa) to optimize the encapsulation of natural
and molecular inclusion (Madene et al., 2006). Spray drying vanilla extract (Vanilla planifolia) by spray drying.
is a fast and cost-effective drying method that can produce
dry spherical powder particles (Sosnik & Seremeta, 2015).
An ideal spray-drying carrier must have a high degree of sol- 2  |  M ATERIAL S AND M ETHOD S
ubility, limited solution viscosity at 35%–45% solid content,
good emulsifying properties, good drying properties, nonhy- 2.1  | Materials
groscopic character, bland taste, nonreactivity, and low cost
(Murúa, Beristain, & Martínez, 2009). In addition to being Rice (Oryza sativa) was purchased from a local supermar-
nonallergenic, GRAS, and affordable, starch is one of the most ket in Pachuca, Hidalgo, Mexico. Vanilla pods (Vanilla
abundant natural carbohydrate polymers. However, some of planifolia) were provided by the National Council of Walnut
its drawbacks include tendency to retrograde, low shear resis- Producers in Papantla de Olarte, Veracruz, Mexico.
tance, lack of hydrophobicity, and low solubility (Ashogbon
& Akintayo, 2014; Sweedman, Tizzotti, Schäfer, & Gilbert,
2013). There is an increasing interest in using native and modi- 2.2  |  Vanilla extract
fied starch to encapsulate food ingredients as flavors; physical,
chemical, and enzymatic modifications have been reported in To obtain the vanilla extract, 100 g of pods was cut into sec-
starches from different sources (Zhu, 2017). Modified starches tions measuring 0.5  cm approximately, added to an amber
of different sources have been studied as wall materials for glass container with 1 L ethanol–water (60:40 v/v) solution,
the encapsulation of vanillin and vanilla extracts by spray and macerated for seven days at a controlled temperature of
drying. Among them are oxidized corn and amaranth starch 25°C. The mash was filtered afterward with filter paper. The
(Chattopadhyaya, Singhal, & Kulkarni, 1998) as well as com- ethanol in the vanillin extract was removed with the aid of
mercial products as Hi Cap 100 and Capsul (from National a rotary evaporator (BÜCHI R-205) to obtain an aqueous
Starch and Chemical) starches (Rodríguez, Wilderjans, Sosa, extract under the following conditions: temperature 55°C,
& Bernik, 2013); still, reports show low values of encapsu- 130 rpm, and pressure of 175 mbar.
lation efficiency (≈60%). With respect to vanillin retention
using different sources of starches as wall material to obtain
microcapsules by spray drying, Tari, Annapure, Singhal, and 2.3  |  Alkaline extraction of rice starch
Kulkarni (2003) obtained a lower recovery of vanillin from rice
starch aggregates due to a higher amylose content (39%); they For the extraction of rice starch, the techniques reported by
considered that the branched structure of amylopectin together Wang and Wang (2004, 2001) were used with some modi-
with its high molecular mass could be responsible for a high fications. Five hundred g of rice was added to a solution of
entrapment of vanillin. The amylase/amylopectin ratio likely 0.1% NaOH at a ratio of 1:2 (rice/solution), and the mixture
plays an important role in encapsulation efficiency and stability was left at room temperature for 18 hr. After the resting time,
of core material (Hoyos, Bello, Alvarez, & Garcia, 2018). Rice three washings were carried out to eliminate excess NaOH.
starch and its derivatives have been previously used as wall The rice was ground in an Oster blender at maximum speed
OCAMPO-SALINAS et al.
|
     557

for 1  min, and the suspension obtained was sieved through 2.7  |  Pasting property
a 125-μm mesh. The supernatant was neutralized with HCL
(5N), centrifuged at 7,500  g for 15  min, and washed four The viscosity profiles of the native and modified starches
times. The starch obtained was dried at 40°C for 24 hr, ground were measured with a Rapid Visco Analyzer (RVA model
dry, sieved through a 100-µm mesh, weighed, and stored in 3C, Newport Scientific, Warriewood, Australia). The heating
a plastic container. The percentage of amylose (%) in starch and cooling cycles in the RVA were programmed as follows:
was determined using a K-amyl 04/06 Amylose/Amylopectin Samples were held at 50°C for 1  min, heated to 92°C at a
assay kit (Biotechnology Megazyme International, Ireland). heating rate of 5.6°C/min, held at 92°C for 5 min, cooled to
50°C at a cooling rate of 5.6°C/min, and held at a final tem-
perature of 50°C for 2 min.
2.4  |  Acid hydrolysis of rice starch

The first chemical modification was acid hydrolysis in native 2.8  |  Thermal properties
rice starch, carried out according to Vargas-León et al. (2019). In
a 500-mL beaker, 100 g (dry basis) native starch was mixed with A differential scanning calorimeter (Model 822E, Mettler
300 ml of 1.2 N hydrochloric acid. The solution was kept under Toledo, Greifensee, Zurich, Switzerland) was used to meas-
stirring for 2 hr and afterward adjusted to pH 7 with a solution ure gelatinization enthalpy (ΔH) and the onset (To), peak
of 5 N sodium hydroxide. The suspension obtained was centri- (Tp), and ending (Te) temperatures of the starches to assess
fuged at 7,500 g for 15 min, and three washes were performed to the thermal properties. Samples (4  mg) were placed in a
eliminate acid residues. The starch was dried in an oven at 40°C standard 40-µL aluminum crucible and added distilled water
for 24 hr, ground, and sieved through a 250-μm mesh. at a 1:4 (starch:water) ratio. They were hermetically sealed
and evaluated at a scanning interval from 25 to 95°C and
5°C/min heating rate (Fonseca-Florido et al., 2017).
2.5  |  Succinylation of rice starch and
degree of substitution
2.9  |  Vanilla extract
The second chemical modification after acid hydrolysis in the microencapsulation process
starch was succinylation. Hydrolyzed starch (125  g, d.b.) and
475 ml water were mixed in a beaker and constantly stirred for A mixture of extruded starch and vanilla extract was pre-
8 hr. Afterward, 3.75 ml 2-Octen-1-ylsuccinic anhydride were pared according to the percentage of total solids of the ex-
added to the mixture (pH 8). After 8 hr, the sample was neutral- perimental design (Table 1) and was considered the sum of
ized with HCL (0.2 N) solution. The starch was washed with dis- the total solids contained in the modified starch and vanilla
tilled water, and samples were dried in a convection oven (Series extract. The mixture was completely homogenized once with
9000, Thermolyne, Dubuque, IA) at 45°C for 24 hr, milled with a T25 IKA homogenizer (dispersing element 18G-ST) at
a hammer mill (model 200, Pulvex, Mexico City, Mexico), and 12,000 rpm for 4 min. The microencapsulation process was
sieved through a 125-μm mesh (Tyler, USA). Succinylation and carried out using a Buchi spray dryer B-191 (Labortechnik
degree of substitution (DS) were carried out according to the AG laboratory, Flawil, Switzerland). The inlet temperature
methodology described by Bhosale and Singhal (2006). was adjusted according to the experimental design, while
the outlet temperature varied proportionally with the air
inlet temperature (Table 1). The liquid flow rate was 4 ml/
2.6  |  Thermoplastic extrusion of rice starch min, the diameter of the nozzle was 0.5 mm, and air pressure
was set at 3.0 bar. The sample was collected after the drying
The sample of succinyl starch (300 g) was adjusted at 20% process in polyethylene bags and stored in a glass desiccator
moisture with water distillate. The extrusion process was protected from light, was obtained the total recovery after
carried out in a single-screw laboratory extruder 19/25 drying (%) (Table 1).
(Brabender). Barrel temperatures were 70, 90, 110, and
130°C at the feeding, transition, and high-pressure ex-
trusion zones, respectively. The compression ratio of the 2.10  |  Encapsulation efficiency
screw was 2:1, and a 4.0  mm diameter die nozzle was
used. Extruded starches were dried in a convection oven The total and superficial content of vanillin was measured for
at 45°C for 2 hr, milled with a hammer mill (model 200, encapsulation efficiency (EE%). The total amount of vanillin
Pulvex, Mexico City, Mexico), and sieved through a 125- was quantified using method 984.13 (AOAC 2005) with some
μm mesh. modifications. Vanillin microcapsules (0.2 g) were weighed
 |
558      OCAMPO-SALINAS et al.

T A B L E 1   Central composite rotatable design model with two independent variables and dependent response (total recovery after drying, %)

Independent variable Dependent variable  

Assay Inlet temperature (°C) Total solids (%) Total recovery after drying (%) Outlet temperature (°C)
1 134 10.7 40.1 92
2 120 12.5 24.0 75
3 120 12.5 23.2 75
4 120 15 20.0 75
5 120 12.5 20.0 75
6 105 10.7 28.9 72
7 105 14.2 22.5 72
8 120 12.5 27.2 75
9 134 14.2 33.8 90
10 120 12.5 28.8 75
11 120 10 44.0 75
12 120 12.5 25.6 75
13 140 12.5 20.0 94
14 100 12.5 24.0 70
Note: Inlet air temperature (100–140°C) and total solids in the mixture (10%–15%). The outlet temperature varied proportionally with the air inlet temperature.

and dissolved in 5  ml water (stock solution); then, 200  μL (c) mixture of wall material and extract with 14.27% total sol-
of the stock solution was taken and adjusted to 10  ml with ids (premix to dry). All readings were performed in triplicate.
distilled water (neutral solution). Subsequently, alkaline so-
lutions were prepared for each sample: Equal parts (200 μL)
of stock solution and 0.1 N NaOH were mixed and gauged to 2.12  |  Water activity and moisture
10  ml with distilled water. The absorbances of the alkaline
solutions were obtained at 348  nm in a spectrophotometer Water activity was assessed by weighing 2  g of capsules
using the neutral solutions as adjustment target. In addition, a and placing them in Aqualab LITE containers (Washington,
standard curve with vanillin and water was also developed to USA) at 25°C. Moisture content was determined according
calculate vanillin concentration. The amount of surface van- to Hernández et al. (2018).
illin was quantified following the methodology described by
Rodríguez, Wilderjans, Sosa, and Bernik (2013) with some
modifications: The microcapsules (0.1  g) were placed in a 2.13  |  Scanning electron microscope
30-mL beaker, and 10 ml ethanol was added. The beaker was
then sealed perfectly with parafilm paper and stirred on a The modified starch and microcapsules were observed in a
grill with a magnetic stirrer at 125 rpm for 90 min. The sam- scanning electron microscope (SEM) (model JEOL JSM-6300
ple was finally filtered with Whatman qualitative filter paper Akishima, Tokyo, Japan). The dried samples were placed in a
No. 6, and the supernatant was measured at a wavelength sample holder covered with double-sided carbon electrocon-
of 277 nm. The formula below, proposed by Tonon, Pedro, ductive tape and then coated with a layer of gold (0.5–1 nm
Grosso, and Hubinger (2012), was used to calculate EE: thick) for 3 min. They were observed at 15 A and 20 kV.
Encapsulation efficiency(EE%)

=
Total vanillin − Surface vanillin
× 100 2.14  |  Confocal laser scanning microscopy
Total vanillin
The microcapsules obtained from optimal drying conditions
2.11  |  Determination of zeta potential were dispersed in fluorescence free immersion oil and ob-
served under a multiphotonic confocal microscope (LSM 710
The zeta potential was determined in a Zetasizer Nano ana- NL0, Carl Zeiss, Germany) with a 63X objective. The laser
lyzer (Malvern, Worcestershire, UK). Deionized water (pH excitation wavelengths were 405, 488, 561, and 633 nm. The
4.0) was used as dispersant material to read samples of (a) fluorescent signals of microcapsules were compared against
modified rice starch (wall material), (b) vanilla extract, and pattern spectra of individual materials: vanilla extract and
OCAMPO-SALINAS et al.
|
     559

wall material (modified starch). Detection channels of va- lower viscosity. Finally, there was no maximum viscosity
nilla extract and wall material were assessed using ZEN 2010 peak for extruded starch because the disintegration of starch
(Carl Zeiss) and four images of two different samples. The structure is the result of high shearing, pressure, and tempera-
images were saved in RGB format with a frame of 512 × 512 ture conditions during extrusion (Chang & Lii, 1992). One of
pixels. the most important characteristics to consider when using a
polymer as wall material in encapsulation by spray drying is
the presence of limited solution viscosity at 35%–45% solid
2.15  |  Experimental design and content (Murúa et al., 2009). Therefore, the results obtained
data analyses were favorable after each modification.

All experiments using modified starches were performed in

triplicate, and differences between means of treatment were 3.2  |  Thermal properties
determined with ANOVA and Tukey's test (p  <  .05) using
Statistica version 7.0. The microencapsulation process was Table 2 shows the thermal properties of the native and
carried out using a central composite rotatable design model modified rice starches. Native starch showed initial, peak,
composed of two factors: inlet air temperature (100–140°C) and ending gelatinization temperatures of 71.94, 77.92,
and total solids in the mixture (10%–15%) (Table 1). The data and 83.41°C, respectively, and gelatinization enthalpy
were then analyzed using the response surface methodology of 12.82  J/g. These results were similar to those reported
(RSM) with Design-Expert version 7.1.5. The experimen- by Wang and Wang (2001) for rice starch with 21.8% of
tal data were adjusted to a second-order polynomial model amylose. The rice starch obtained in this work presented
shown in the equation: 22.4 ± 0.120% of amylose. Contrastingly, hydrolyzed starch
exhibited increased gelatinization temperatures. This can be
Yi = b0 + b1 X1 + b2 X2 + b12 X1 X2 + b11 X12 + b22 X22 attributed to the fact that the surface of the starch granule
and the amorphous regions of amylopectin (α 1–6 linkage)
where y is the response variable, X1 represents the total solids in are more rapidly affected during acid hydrolysis at tempera-
the mixture, and X2 corresponds to the temperature (°C) of the tures below the gelatinization temperature. Consequently,
inlet in the spray-drying process. The significance of the mod- the results indicate that a higher percentage of amorphous
els was tested using variance analysis (F test). The optimization structure was hydrolyzed in acid-thinned rice starch, lead-
of the microencapsulation process was performed using RSM ing to an increase in relative crystallinity (Wang & Wang,
in which the selected response was EE%. 2001). Succinylated rice starch showed decreased To and
Tp of gelatinization with respect to native and hydrolyzed
starch. The effect can be explained as a weakening of inter-
3  |  R E S U LTS A N D D IS C U S S ION nal hydrogen bonding by the hydrophobic alkenyl groups,
helping starch swelling at relatively low temperatures, and
3.1  |  Pasting property hence gradually decreasing the enthalpy of all the succi-
nylated starch. Similar results were reported by Bao, Xing,
Viscosity profiles were obtained for native, hydrolyzed, suc- Phillips, and Corke (2003). Finally, the phenomenon of ge-
cinylated, and extruded rice starch (Figure 1). For native rice, latinization was absent from extruded rice starch due to the
the maximum viscosity peak was 3,875 cP. This result was disintegration of the granule during the extrusion process
similar to that reported by Giuberti, Marti, Fortunati, and (Figure 2). Rice starch is a polymeric mixture of essentially
Gallo (2017). The maximum viscosity peak for the hydro- linear amylose and highly branched amylopectin. These two
lyzed starch was 1,152 cP. When compared with the native polymers are organized into a semi-crystalline structure in
starch, viscosity was reduced by 50%; acid-thinned starch ex- the starch granule (Jenkins & Donald, 1995). After the mod-
hibited a reduced viscosity because of the depolymerization ifications and with the values obtained from the DSC tech-
of starch molecules (Ai & Jane, 2015). For succinylated rice nique, significant changes in mainly extruded starch were
starch, a decrement was observed as compared against hydro- evidenced. The absence of gelatinization temperatures and
lyzed starch with maximum viscosity peak of 88 cP. These enthalpy of gelatinization indicate a change in the crystal-
results can be attributed to the effect of the succinylation pro- line and amorphous part of the extruded starch as a function
cess because during the reaction the pH 8 was maintained of the extrusion process. These characteristics of the wall
for 8 hr; this as results to OSA groups acted by first attack- material (extruded starch) make it possible to form a net-
ing the surface and some pores formed, this allows starch work during spray drying with the available amylose and
granules to absorb water, swell rapidly, and therefore have amylopectin polymers that allow for core encapsulation.
 560     
| OCAMPO-SALINAS et al.

95 F I G U R E 1   Viscosity profiles for

4,000 native, hydrolyzed, succinylated, and
90 Temperature
Native starch extruded rice starch
Hydrolyzed starch 3,500
85 Succinylated starch
Extruded starch
3,000
Temperature (ºC)

80

Viscosity (cP)
2,500
75

2,000
70

65 1,500

60 1,000

55 500

50 0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Time (min)

T A B L E 2   Thermal properties of native and modified rice starches

Onset temperature (To) Peak temperature (Tp) Ending temperature (Te) Gelatinization
Starch (°C) (°C) (°C) enthalpy ∆H (J/g)
Native 71.9 ± 0.06 a 77.9 ± 0.00 a 83.4 ± 0.35 ac 12.8 ± 0.11 ab
Hydrolyzed 73.2 ± 0.13 b 79.6 ± 0.00 b 85.9 ± 0.15 b 11.0 ± 0.44 ab
Succinylated 69.2 ± 0.12 c 76.8 ± 0.11 c 82.8 ± 0.27 ac 8.9 ± 0.61 c
Extruded – – – –
Note: Means ± standard deviations of three replicates. Means with different letters in the same column are significantly different (p < .05).

(2006) for succinylated rice starch without previous hy-

drolysis. The authors suggest that succinylated starches
with DS lower than 0.04 do not present changes in the crys-
tal structure. However, although the DS was not greater
after the hydrolysis process, the viscosity profiles and the
thermal properties were significantly modified in the suc-
cinate starches. The FDA states that the maximum level of
OSA treatment allowed is 3% (DS 0.02). The starch ob-
tained has a high potential for application in the food sector.
Recently, Fonseca-Florido et al. (2018) reported emulsify-
ing properties of waxy cassava starch. The authors found
that the emulsification index values were acid treated/suc-
cinated starch> native starch, which could affect the EE%.
However, there are no reports on the relationship between
DS and encapsulation efficiency. The former is a key pa-
rameter for both future research and experiments based on
the degree of substitution of modified starch (succinate)
F I G U R E 2   Thermograms for (a) native, (b) hydrolyzed, (c)
and encapsulation efficiency.
succinylated, and (d) extruded rice starches

3.3  |  Degree of substitution for succinylated 3.4  |  Encapsulation efficiency

rice starch
The surface response graph for the EE% variable
Succinylated rice starch exhibited 0.01 ± 0.005 DS, a simi- (Figure 3a) shows that higher solid concentration
lar result to that reported by Song, He, Ruan, and Chen (14.27%) and drying temperature (134.14°C) lead to
OCAMPO-SALINAS et al.
|
     561

higher EE (79.67%). The maximum efficiency found 3.6  |  Determination of zeta potential
in this work was greater than the one reported by
Chattopadhyaya, Singhal, and Kulkarni (1998), who The zeta potential values of vanilla extract, wall material
developed oxidized corn and amaranth starches as wall (modified starch), and mixture of extract and wall material
materials, obtaining 60.89 and 58.61 EE%, respectively, were −14.00 ± 0.43, −5.01 ± 0.42, and −5.87 ± 0.34 mV, re-
for vanillin after spray-drying process. Rodríguez et al. spectively. According to Rafiee, Barzegar, Sahari, Maherani,
(2013) reported ≈64 EE% using Hi Cap 100 and Capsul and Rafiee (2017), phenolic molecules increase the negative
commercial starches from National Starch and Chemical. charge on surface particles. The vanilla extract presented a neg-
Finally, ANOVA of EE% showed a significant regression ative charge, considering its quantity of phenolic compounds
model with a value of R2 ≥ 0.77, a variation coefficient (Sinha, Verma, & Sharma, 2007). Modified starch dispersions
of 8.55, and p-value < .0001. The lack of adjustment was were similar to the zeta potential observed by Joye, Nelis,
significant (<0.0001). The estimated coefficients of the and McClements (2015). With respect to optimized formula-
adjusted model are shown in Table 3; both independent tion (mixture of extract and wall material), a polysaccharide
variables in linear term (b1 and b2) showed a significant with enough concentration could entrap particles (McClement,
effect on EE (<0.0001). 2000) and keep the dominant zeta potential of modified starch.

3.5  |  Optimization of spray-drying process 3.7  |  Water activity and moisture

The spray-drying process was optimized for industrial ap- The water activity of the microcapsules obtained by opti-
plications and improved performance (inlet air temperature mal drying conditions was 0.306  ±  0.005, while values of
and total solids) in the production of microencapsulation with 5.74 ± 0.7 were obtained for moisture. Low values of water
vanillin. Figure 3b shows the area corresponding to the best activity and residual moisture content (<4%–5% w/w) are
drying conditions. The optimal drying conditions predicted prerequisites for the commercial production of spray-dried
in the software were an entry temperature in the dryer of powders with good handling characteristics, such as high
125.11°C with 14.27% total solids and 76.25% EE for vanil- fluidity, low tackiness, and agglomeration (Barbosa, Ortega,
lin. Microencapsulation showed EE of 86.23% with experi- Juliano & Yan, 2005). The result obtained could be favorable
mentally optimal drying conditions. for the stability of the vanilla extract during storage.

(a) (b)
Overlay plot
134.14 3

Encapsulati 76.252
X1 14.27
127.07 X2 125.11

120.00 18

112.93

105.86 3 3
10.73 11.62 12.50 13.38 14.27

X1: A: Total solids (%)

X2: B: Temperature (°C)

F I G U R E 3   (a) Surface response graph for encapsulation efficiency (EE%) in relation to temperature (°C) of dry and total solids (%) in starch
mixture and vanilla extract and (b) area corresponding to the best drying conditions for vanillin encapsulation through spray drying [Colour figure
can be viewed at wileyonlinelibrary.com]
 |
562      OCAMPO-SALINAS et al.

T A B L E 3   Estimated coefficients of the adjusted model and level of significance for encapsulation efficiency (%)

Intercept Linear Interaction Quadratic

Response b0 b1 b2 b12 b11 b22

Encapsulation 68.54 7.56 (<0.0001) 6.39 (<0.0001) –1.5 (0.4705) –0.80 (0.4967) –7.23 (<0.0001)
efficiency (%)
Note: Model in which b1 = total solids (%); b2 = dryer inlet temperature (°C).

(a) (b) (c)

(d) (e) (f)

F I G U R E 4   Micrographs of modified rice starches: (a) native, (b) hydrolyzed, (c) succinylated, and d) extruded. Microcapsules of vanillin
with modified rice starch at (e) 1000× and (f) 3000×

3.8  |  Scanning electron microscopy deform and contract (Alamilla et al., 2005). In consequence,
said anomalies on the surface could be desirable in order to
Figure 4a shows the characteristic images of native rice improve rehydration of the powders (Guadarrama-Lezama
starch, previously described by other authors (Vandapuente et al., 2012).
& Delcour, 2004; Jane et al., 1994): small size (2–7  µm)
and polyhedral and irregular shape, like sharp stones. No
changes were observed on the surface of native rice starch 3.9  |  Confocal laser scanning microscopy
and hydrolyzed starch (Figure 4b). However, there was a
certain loss of sharpness and some granules showed slightly Figure 5 shows the 3D reconstruction images from opti-
rough surfaces (white arrows) after succinylation reactions mized process for vanilla extract encapsulation. The im-
(Figure 4c), matching with the report by Song, Chen, Ruan, ages were acquired by confocal laser scanning microscopy
He, and Xu (2006). After the extrusion process (Figure 4d), (CLSM) and are similar to those obtained by SEM. The
the starch granules completely lost their individual shape, green channel corresponds to modified starch, (b) while the
forming a mass of porous laminar structures due to the ex- red one represents vanilla extract distribution along micro-
pansion suffered by melted starch (Yan & Zhengbiao, 2010). capsules (a) in a matrix-type conformation of small aggre-
Microcapsules were mostly spherical with an estimated gates. Interestingly, colocalization signals represented by
size under 20  µm (Figure 4e and f); the surface appeared the overlapping of both channels (Liu, Ma, Xue, & Shi,
deformed with concavities and dents without presence of 2011), in yellow, were absent, likely because there was no
breakage or fractures. This description fits with the charac- solid interaction between the two components. Even though
teristics of those microcapsules produced by spray drying at the vanilla extract is water soluble, the modified starch
low temperatures, as in this work. The relatively small size only reached a 30.50% WSI after extrusion. In addition,
and appearance can be attributed to shrinkage incidence be- according to the z-potential analysis, the charges of the va-
cause water diffusion takes longer, allowing for structures to nilla extract and aqueous starch could cause the repulsion
OCAMPO-SALINAS et al.
|
     563

F I G U R E 5   Micrographs (3D
(a) (b) (c)
reconstruction) showing the distribution
of components along the microcapsules
obtained with optimal conditions: (a)
vanilla extract (red channel), (b) wall
material (green channel), (c) overlapped
red and green channel, (d) clipping of the
overlapped image. Scale: 10 µm [Colour
figure can be viewed at wileyonlinelibrary.
com]
(d)

between their particles, albeit at a different magnitude, and

limit their interaction. It is important to point that the het- Enaim Aída Vargas-León  https://orcid.
erogeneous spread of very tiny red bodies mainly located org/0000-0003-3440-5981
in surface cavities (c and d) explains the relative higher Fabiola Araceli Guzmán-Ortiz  https://orcid.
result of encapsulation efficiency in this work. org/0000-0003-1333-3624
Netzahualcoyol Calcáneo-Martínez  https://orcid.
org/0000-0002-6568-8454
4  |   CO NC LU SION S Reyna Nallely Falfán-Cortés  https://orcid.
org/0000-0003-2767-9896
The rice starch modified by hydrolysis, succinylation, and
thermoplastic extrusion, presented good characteristics in R E F E R E NC E S
profiles of viscosity, thermal properties, and degree sub- AOAC (2005). Methods of Analysis, 18th ed. Washington, D.C.:
stitution. Therefore, it constitutes an excellent wall ma- Association of Official Analytical Chemist.
terial for vanilla encapsulation with an efficiency higher Ashogbon, A. O., & Akintayo, E. T. (2014). Recent trend in the phys-
ical and chemical modification of starches from different botanical
than 80%. The spray-drying process to encapsulate natural
sources: A review. Starch-Stärke, 66(1–2), 41–57.
vanilla was optimized; the best conditions were 125°C and
Bao, J., Xing, J., Phillips, D. L., & Corke, H. (2003). Physical prop-
14.2% total solids. The confocal laser scanning microscopy erties of octenyl succinic anhydride modified rice, wheat, and po-
showed no interaction between the wall material and the tato starches. Journal of Agricultural and Food Chemistry, 51(8),
vanilla extract. In conclusion, chemically and physically 2283–2287. https://doi.org/10.1021/jf020​371u
modified rice starch is potentially an ideal wall material Baranauskienė, R., Bylaitė, E., Žukauskaitė, J., & Venskutonis, R. P.
for the encapsulation of different bioactive compounds in (2007). Flavor retention of peppermint (Menthapiperita L.) essen-
the food sector. tial oil spray-dried in modified starches during encapsulation and
storage. Journal of Agricultural and Food Chemistry, 55(8), 3027–
3036. https://doi.org/10.1021/jf062​508c
ACKNOWLEDGMENTS
Barbosa, C. G. V., Ortega, R. E., Juliano, P., & Yan, H. (2005). Food
The authors thank Universidad Autónoma del Estado de powders: physical properties, processing, and functionality (Vol.
Hidalgo and CONACYT for their support to this work. 86, pp. 71–75). New York: Kluwer Academic/Plenum Publishers.
Beirão, D. C. S., Duarte, C., Moldão, M. M., & Beirão, D. C. M. L.
ORCID (2011). Physical characterization of rice starch spherical aggregates
Israel Oswaldo Ocampo-Salinas  https://orcid. produced by spray-drying. Journal of Food Engineering, 104(1),
org/0000-0002-5507-4889 36–42. https://doi.org/10.1016/j.jfood​eng.2010.11.024
Bhosale, R., & Singhal, R. (2006). Process optimization for the synthesis
Carlos Alberto Gómez-Aldapa  https://orcid.
of octenylsuccinyl derivative of waxy corn and amaranth starches.
org/0000-0002-4723-3093 Carbohydrate Polymers, 66(4), 521–527. https://doi.org/10.1016/j.
Javier Castro-Rosas  https://orcid.org/0000-0002-7823-8954 carbp​ol.2006.04.007
 |
564      OCAMPO-SALINAS et al.

Chang, Y., & Lii, C. (1992). Preparation of starch phosphates by ex- Milanovic, J., Manojlovic, V., Levic, S., Rajic, N., Nedovic, V., &
trusion. Journal of Food Science, 57(1), 203–205. https://doi. Bugarski, B. (2010). Microencapsulation of flavors in carnauba
org/10.1111/j.1365-2621.1992.tb054​56.x wax. Sensors, 10(1), 901–912. https://doi.org/10.3390/s1001​00901
Chattopadhyaya, S., Singhal, R. S., Kulkarni, P. R. (1998). Oxidised Murúa, P. B., Beristain, G. C. I., & Martínez, B. F. (2009). Preparation
starch as gum arabic substitute for encapsulation of flavours. of starch derivatives using reactive extrusion and evaluation of mod-
Carbohydrate Polymers, 37(2), 143–144. https://doi.org/10.1016/ ified starches as shell materials for encapsulation of flavoring agents
S0144-8617(98)00054-X by spray drying. Journal of Food Engineering, 91(3), 380–386.
Fonseca-Florido, H. A., Hernández-Ávilab, J., Rodríguez-Hernández, https://doi.org/10.1016/j.jfood​eng.2008.09.035
A. I., Castro-Rosas, J., Acevedo-Sandoval, O. A., Chavarria- Rafiee, Z., Barzegar, M., Sahari, M. A., & Maherani, B. (2017).
Hernández, N., & Gómez-Aldapa, C. A. (2017). Thermal, rheolog- Nanoliposomal carriers for improvement the bioavailability of
ical, and mechanical properties of normal corn and potato starch high–valued phenolic compounds of pistachio green hull extract.
blends. International Journal of Food Properties, 20(3), 611–622. Food Chemistry, 220, 115–122. https://doi.org/10.1016/j.foodc​
https://doi.org/10.1080/10942​912.2016.1171779 hem.2016.09.207
Fonseca-Florido, H. A., Vázquez-García, H. G., Méndez-Montealvo, Rodríguez, S. D., Wilderjans, T. F., Sosa, N., & Bernik, D. L. (2013).
G., Basilio-Cortés, U. A., Navarro-Cortés, R., Rodríguez-Marín, Image texture analysis and gas sensor array studies applied to va-
M. L., … Gómez-Aldapa, C. A. (2018). Effect of acid hydrolysis nilla encapsulation by octenyl succinic anhydride starches. Journal
and OSA esterification of waxy cassava starch on emulsifying prop- of Food Research, 2(2), 36. https://doi.org/10.5539/jfr.v2n2p36
erties in Pickering-type emulsions. LWT, 91, 258–264. https://doi. Romero de la Vega, G., Salgado Cervantes, M. A., Garcia Alvarado,
org/10.1016/j.lwt.2018.01.057 M. A., Romero-Martínez, A., & Hegel, P. E. (2016). Fractionation
Giuberti, G., Marti, A., Fortunati, P., & Gallo, A. (2017). Gluten free of vanilla oleoresin by supercritical CO2 technology. Journal
rice cookies with resistant starch ingredients from modified waxy of Supercritical Fluids, 108, 79–88. https://doi.org/10.1016/j.
rice starches: Nutritional aspects and textural characteristics. supflu.2015.10.022
Journal of Cereal Science, 76, 157–164. https://doi.org/10.1016/j. Sinha, A. K., Sharma, U. K., & Sharma, N. (2008). A comprehensive re-
jcs.2017.06.008 view on vanilla flavor: Extraction, isolation and quantification of van-
Hernández-López, Z., Rangel-Vargas, E., Castro-Rosas, J., Gómez- illin and others constituents. International Journal of Food Science
Aldapa, C. A., Cadena-Ramírez, A., Acevedo-Sandoval, O. A., and Nutrition, 59(4), 299–326. https://doi.org/10.1080/09687​63070​
… Falfán-Cortés, R. N. (2018). Optimization of a spray-drying 1539350
process for the production of maximally viable microencapsu- Sinha, A. K., Verma, S. C., & Sharma, U. K. (2007). Development
lated Lactobacillus pentosus using a mixture of starch-pulque and validation of an RP-HPLC method for quantitative determina-
as wall material. LWT, 95, 216–222. https://doi.org/10.1016/j. tion of vanillin and related phenolic compounds in Vanilla plani-
lwt.2018.04.075 folia. Journal of Separation Science, 30(1), 15–20. https://doi.
Hoyos, L. J. D., Bello, P. L. A., Alvarez, R. J., & Garcia, H. S. org/10.1002/jssc.20060​0193
(2018). Microencapsulation using starch as wall material: A re- Song, X. Y., Chen, Q. H., Ruan, H., He, G. Q., & Xu, Q. (2006).
view. Food Reviews International, 34(2), 148–161. https://doi. Synthesis and paste properties of octenyl succinic anhydride modi-
org/10.1080/87559​129.2016.1261298 fied early Indica rice starch. Journal of Zhejiang University Science
Jenkins, P. J., & Donald, A. M. (1995). The influence of am- B., 7(10), 800–805. https://doi.org/10.1021/jf020​371u
ylose on starch granule structure. International Journal of Sosnik, A., & Seremeta, K. P. (2015). Advantages and challenges of the
Biological Macromolecules, 17(6), 315–321. https://doi. spray-drying technology for the production of pure drug particles and
org/10.1016/0141-8130(96)81838-1 drug-loaded polymeric carriers. Advances in Colloid and Interface
Joye, I. J., Nelis, V. A., & McClements, D. J. (2015). Gliadin-based Science, 223, 40–54. https://doi.org/10.1016/j.cis.2015.05.003
nanoparticles: Stabilization by post-production polysaccharide Sostaric, T., Boyce, M. C. & Spickett, E. E. (2000). Analysis of the
coating. Food Hydrocolloid, 43, 236–242. https://doi.org/10.1016/j. volatile components in vanilla extracts and flavorings by solid-phase
foodh​yd.2014.05.021 microextraction and gas chromatography. Journal of Agriculture
Liu, K. L., Widjaja, E., Huang, Y., Ng, X. W., Loo, S. C. J., Boey, F. Y. and Food Chemistry, 48(12), 5802–5807. https://doi.org/10.1021/
C., & Venkatraman, S. S. (2011). A new insight for an old system: jf000515
Protein-PEG colocalization in relation to protein release from PCL/ Surojanametakul, V., Tungtrakul, P., Varannanond, W., Supasri, R.,
PEG blends. Molecular Pharmaceutics, 8(6), 2173–2182. https:// Boonbumroung, S., & Themtakul, K. (2005). Properties of spray-
doi.org/10.1021/mp200​513b dried rice starch microcapsule. Kasetsart Journal (Natural Science),
Madene, A., Jacquot, M., Scher, J., & Desobry, S. (2006). Flavour 39, 730–738.
encapsulation and controlled release–a review. International Sweedman, M. C., Tizzotti, M. J., Schäfer, C., & Gilbert, R. G. (2013).
Journal of Food Science & Technology, 41(1), 1–21. https://doi. Structure and physicochemical properties of octenyl succinic anhy-
org/10.1111/j.1365-2621.2005.00980.x dride modified starches: A review. Carbohydrate Polymers, 92(1),
Márquez, G. M., Galicia, G. T., Márquez, M. R., Ruiz, G. M., & 905–920. https://doi.org/10.1016/j.carbp​ol.2012.09.040
Quintero-Ramos, A. (2018). Spray-dried microencapsulation of Tari, T. A., Annapure, U. S., Singhal, R. S., & Kulkarni, P. R. (2003).
orange essential oil using modified rice starch as wall material. Starch-based spherical aggregates: Screening of small granule sized
Journal of Food Processing and Preservation, 42(2), e13428. starches for entrapment of a model flavouring compound, vanillin.
https://doi.org/10.1111/jfpp.13428 Carbohydrate Polymers, 53(1), 45–51. https://doi.org/10.1016/
McClements, D. J. (2000). Comments on viscosity enhancement and de- S0144-8617(02)00293-X
pletion flocculation by polysaccharides. Food Hydrocolloids, 14(2), Tonon, R. V., Pedro, R. B., Grosso, C. R., & Hubinger, M. D. (2012).
173–177. https://doi.org/10.1016/S0268-005X(99)00065-X Microencapsulation of flaxseed oil by spray drying: Effect of oil
OCAMPO-SALINAS et al.
|
     565

load and type of wall material. Drying Technology, 30(13), 1491– 53(11), 570–576. https://doi.org/10.1002/1521-379X(20011​
2150. https://doi.org/10.1080/07373​937.2012.696227 1)53:11<570:AID-STAR5​70>3.0.CO;2-S
Trindade, M. A., & Grosso, C. R. F. (2000). The stability of ascor- Wang, X., Yuan, Y., & Yue, T. (2015). The application of starch-based
bic acid microencapsulated in granules of rice starch and in gum ingredients in flavor encapsulation. Starch-Starke, 67(3–4), 225–
Arabic. Journal of Microencapsulation, 17(2), 169–176. https://doi. 236. https://doi.org/10.1002/star.20140​0163
org/10.1080/02652​04002​88409 Zhu, F. (2017). Encapsulation and delivery of food ingredients using
Vargas-León, E. A., Falfan-Cortes, R. N., Navarro-Cortez, R. O., starch based systems. Food Chemistry, 229, 542–552. https://doi.
Hernández-Ávila, J., Castro-Rosas, J., & Gómez-Aldapa, C. A. org/10.1016/j.foodc​hem.2017.02.101
(2019). Double chemical modification in rice starch: Acid hydro-
lysis optimization process and phosphating. CyTA-Journal of Food,
17(1), 632–639. https://doi.org/10.1080/19476​337.2019.1624623 How to cite this article: Ocampo-Salinas IO, Gómez-
Wang, L., Wang, Y. J. (2004). Rice starch isolation by neutral prote- Aldapa CA, Castro-Rosas J, et al. Development of wall
ase and high-intensity ultrasound. Journal of Cereal Science, 39(2), material for the microencapsulation of natural vanilla
291–296. https://doi.org/10.1016/j.jcs.2003.11.002
extract by spray drying. Cereal Chem. 2020;97:555–
Wang, L., & Wang, Y. (2001). Comparison of protease digestion at
neutral pH with alkaline steeping method for rice starch isola-
565.
tion. Cereal Chemistry, 78(6), 690–692. https://doi.org/10.1094/ https://doi.org/10.1002/cche.10269
CCHEM.2001.78.6.690
Wang, L., & Wang, Y. (2001). Structures and physicochemical proper-
ties of acid-thinned corn, potato and rice starches. Starch-Starke.,