Preparation and characterization of octenyl succinylated normal and waxy starches of maize as encapsulating agents for anthocyanins by spray-drying

Abstract

The encapsulation by spray drying of maize anthocyanins was evaluated using two types of wall materials, consisting of normal and waxy maize starch, which were esterified with octenyl succinic anhydride. The X-ray diffraction analysis revealed that SWMS possessed a completely amorphous, while SNMS had a crystalline structure. SNMS showed peaks at 2θ = 13.1°, 19.8° and 22.4°. The results revealed that SNMS and SWMS had almost the same encapsulation productivity (EP); SNMS showed the best performance because its EP was higher (95%) than in SWMS (90%). The stability of microcapsules produced with SNMS showed the highest anthocyanin retention after storage in the water activity (a_w) range of 0.11–0.94 at 40 °C.

Introduction

The attractive color of anthocyanins and their properties for disease prevention make them suitable for applications in food matrices. While synthetic colorants are widely used as food additives, anthocyanins have a limited applicability as food additives since they are more sensitive to factors such as temperature, exposure to light and UV radiation, pH, and oxygen among others (Li et al. 2013). In order to overcome the influence of these factors, one approach consists of encapsulating anthocyanins, which leads to their stabilization. Spray drying is the most common method for encapsulating anthocyanins. It consists in spraying a feeding fluid into a gaseous drying medium, as a result the feeding fluid is converted into a dried particulate form (Cal and Sollohub 2010). This technology allows encapsulating compounds with the aid of an encapsulating agent that acts as a physical barrier against the influence of the oxygen and small molecules. In addition, the encapsulating agent inhibits the chemical and enzymatic degradation (Wang et al. 2009).

There is a limited number of encapsulating agents used in the encapsulation of anthocyanins in the spray drying process, including starches, maltodextrins, proteins, gums and mixtures of them. Maltodextrins constitute a good option for the microencapsulation of anthocyanins given that the microencapsulation efficiency is increased because of their high solubility, compared to the native starches (Tonon et al. 2010). However, no emulsifying properties are presented in the hydrolyzed starches. The esterification with lipophilic octenyl succinic anhydride (OSA) equips the starch granule with amphiphilic properties (Borrmann et al. 2013). The OSA starch is one of the most important encapsulating agents used for microencapsulation of hydrophobic compounds such as oils and flavoring agents (Murúa-Pagola et al. 2009; Wang et al. 2009). Anthocyanins are hydrophilic colorants and recently, a few articles report the use of a commercial OSA starch as encapsulating agent for anthocyanins (da Silva et al. 2013; Villacrez et al. 2014).These works report high encapsulation efficiencies for the microencapsulation of anthocyanins, further the work of Villacrez reports a higher storage stability in the anthocyanin microparticles produced with an OSA starch derived from waxy maize starch (Hi-Cap^TM), than those maltodextrin microparticles produced with DE 20.

With respect to the encapsulation of anthocyanins by spray-drying, up to best knowledge it seems that there is no report about the use of an OSA starch derived from normal maize starch for encapsulating anthocyanins. In this way, the objectives of the present work are: (1) to prepare and characterize the succinylated normal and waxy maize starches as encapsulating agents for purple maize anthocyanins using spray-drying; (2) to characterize the productivity and efficiency of the encapsulated purple maize anthocyanins; (3) to investigate the influence of the kind of starch in the water sorption isotherms as well as the storage stabilities at water activities (a_w) ranging from 0.11 to 0.94 at 40 °C.

Materials and methods

Raw materials

Grains of purple maize (Zea mays L. cv Cónico) were harvested (San Juan Ixtenco, Tlaxcala, México) at about 13% moisture in january 2011, and stored at 4 °C before use. Normal maize starch (containing approximately 27% of apparent amylose) and waxy maize starch (~ 1% of amylose) were obtained from CPIngredients (San Juan del Rio, Querétaro, México). Solvents and KCl were obtained from J. T. Baker (Edo. de México, México); 2-Octen-1-ylsuccinic anhydride (Cat. No. 416487, 97%) was purchased from Sigma-Aldrich Chemical Co. (Toluca de Lerdo, México).

Anthocyanins extract preparation

Purple maize grains were equilibrated to 16% moisture content. Then the bran, which is purple in color, was mechanically removed using a Strong-Scott barley pearler. After that, 12.5 g of bran was added to 150 mL of a mixture of methanol, acetic acid and water (10:1:9) in an Erlenmeyer flask. Then, the mixture was sonicated in an ultrasonic bath Branson 2510 Marshall Scientific (Hampton, NH, USA) for 15 min and stirred during 1.45 h in a horizontal shaker MaxQ6000 ThermoFisher^TM (Hopkinton, MA, USA) at 110 rpm (this procedure was done twice). Finally, the solids were separated by filtration using filter paper Whatman #4 by mild vacuum suction using a Buchner funnel the dissolvent (methanol, acetic acid and water) was evaporated by rotary evaporator at 35 °C. The anthocyanin extract was kept at − 20 °C until use.

The quantification of total anthocyanin content (TAC) was determined by the pH-differential method (Giusti and Wrolstad 2001) using a spectrophotometer Lambda 25 UV VIS PerkinElmer (Miami, FL, USA) at 510 and 700 nm. The TAC was calculated on the basis of the content of cyanidin-3-glucoside, using the following formula:

$$\text{Anthocyanin}\text{pigment}\left(\text{cyanidin-3-glucoside equivalents},\text{mg/L}\right)=\text{A}\times \text{MW}\times \text{DF}\times {10}^3/\mathrm{\epsilon }\times 1,$$ 1

where A = (A_520nm–A_700nm)pH1.0 − (A_520nm–A_700nm)pH4.5; MW (molecular weight) = 449.2 g/mol for cyanidin-3-glucoside; DF = dilution factor; 1 path length in cm; ε = 26 900 molar extinction coefficient, in L × mol⁻¹×cm⁻¹, for cyanidin-3-glucoside; and 10³ = factor for conversion from g to mg.

Preparation of starch derivatives

Normal and waxy maize starches were hydrolyzed with hydrochloric acid. Starch slurry was prepared by dispersing 400 g of starch in 1 L of 0.384 N HCl. Both starches were esterified with OSA according to the methodology proposed by Murúa-Pagola et al. (2009). Starches were esterified by adding 4 mL of OSA/100 g of starch (d.b.) drop-wise over 2 h at room temperature (25 ± 5 °C). After that, the reaction proceeded for another 4 h by maintaining the pH between 8.5 and 9.0, using an aqueous 1 N NaOH solution. After the esterification process, the pH was adjusted to 7.0 ± 0.20 using a solution of 5% (v/v) of aqueous HCl. The resulting slurry was centrifuged for 10 min at 5000×g. Finally, the modified starches were washed with distilled water and then dried to approximately 7% moisture content, for 24 h in a convection oven at 45 °C. The powder was ground and sieved (mesh 100 of 149 µm opening size). The succinylated normal maize starch (SNMS) and succinylated waxy maize starch (SWMS) were processed in a single screw extruder (CINVESTAV IPN, Mexico), with screw compression ratio of 3:1, internal barrel diameter of 20 mm and L/D of 20. The used temperature profiles were (± 5) were 75, 150, and 180 °C for the feeding, transition, and high-pressure zones respectively, while the screw speed was 80 rpm, feed rate was 70 g/min. And die diameter was 4.0 mm.

XRD analysis

The X-ray diffraction patterns (XRD) of native and succinylated starches, and those from the microcapsules of MSNMS and MSWMS were obtained on a Rigaku Dmax2100 (Rigaku Denki Co. Ltd, Japan) diffraction instrument operating at 30 kV with Cu-K∝ radiation wavelenght of λ = 1.5444 Å. The diffraction angle was from 4° to 60° on a 2θ scale with a stepwise as at the increment of 0.03⁰. The relative crystallinity of starch granules was calculated as the ratio of the crystalline area (A_c) and the total area under the major diffraction peaks (Nuwamanya et al. 2010), the software used to analyze the spectrum was OriginPro 8 (OriginLab Corporation, MA, USA) through the Wizard of Baseline and Peaks function.

Degree of substitution

The degree of substitution (DS) was determined according to the method of Jeon et al. (1999). One sample (1 g) from each SNMS and SWMS starches was dissolved in 10 mL of absolute dimethyl sulfoxide for 10 min at 70 °C. Once the dispersion was cooled, about to 5–6 drops of phenolphthalein were used as indicator. Then, the solution was titrated with a standard solution of 0.05 M sodium hydroxide up to a pale pink color was obtained. The DS was calculated with the equation of Song et al. (2006) using the molecular weight of OSA:

$$DS=\left(0.162\times \left(A\times N\right)/W\right)/\left(1-\left(0.101\times \left(A\times N\right)/W\right)\right),$$ 2

where A is the titration volume of NaOH solution (mL), N is the normality of NaOH solution, and W is the dried basis weight (209 g) of octenyl succinyl group minus the molecular weight of the hydrogen atom. Three repetitions were performed for each analysis, so the mean value was reported.

Water solubility index and water absorption index

Water solubility index (WSI) and water absorption index (WAI) were determined according to the method described by Anderson et al. (1970). The sample (2.5 g on d. b.) was vigorously mixed in 30 mL of water in a 50 mL centrifuge tube. After that, the mixture was incubated in a water bath at 30 °C for 30 min, and then it was centrifuged at 3000×g for 15 min. The resulting supernatant was collected in a pre-weighed Petri dish and dried into an oven overnight at 105 °C, the amount of solids in the dried supernatant as a percentage of the total dried solids in the original 2.5 g sample is an indicator of the WSI (see formula 3). The gel was weighed and the WAI defined as the grams of gel per gram of solids was calculated (see formula 4).

$$\text{WSI}=\left(\text{weight of solids in supernatant/weight of dry solids}\right)\times 100,$$ 3

$$\text{WAI}=\left(\text{weight of sediment/weight of dry solids}\right)\times 100,$$ 4

Preparation of microcapsules by spray-drying

For the encapsulation of anthocyanins, 20 mL of resuspended extract of anthocyanins (0.94 mg/mL of cyanidin-3-glucoside) together with 20 g of SNMS or SWMS were dispersed individually with the aid of a plastic spoon in 80 mL of distilled water at 40 °C. The mixture was homogenized at 14,000 rpm during 10 min (Ultra Turrax T-25-SI, USA).

The encapsulation was carried out by spray-drying (LabPlant SD-Basic, Huddersfield, UK), with the following drying conditions: inlet air temperature of 170 ± 5 °C; outlet air temperature of 80 ± 10 °C; nozzle diameter of 0.5 mm; and liquid flow rate of 10 mL/min. Equipment’s air flow was set at 70 m³/h.

Morphology

The external morphology of the microparticles was observed by SEM (XL30 ESEM, EDAX, Philips, Holland), using an acceleration voltage of 10 kV. For observing the internal morphology of the microcapsules, the samples were glued between two adhesion tapes after which the two tapes were separated to fracture some of the capsules.

Moisture content, hygroscopicity and water activity (a_w)

The moisture content was determined according to the AACC method No. 44-19 (AACC 1995). Briefly, 2 g of the powder formulations were placed in aluminium pans, dried at 135 °C for 90 min, and the residual moisture content was expressed as percentage of the initial weight. Hygroscopicity was determined according to the method proposed by Nijdam and Langrish (2006). Samples of powder (1 g on d. b.) of each SNMS and SWMS starches were placed at 25 °C in a container with saturated KCl solution (84.26% of relative humidity). After 1 week, these samples were weighed and the hygroscopicity was expressed as g of adsorbed moisture per 100 g dried solids (g/100 g). Measurements of a_w in the anthocyanin microcapsules were determined using an Aqualab equipment (Models Series 3TE, Decagon Devices, Pullman, WA, USA).

Determination of the total and superficial anthocyanins

In order to completely dissolve the microcapsules, 100 mg of them were weighed and mixed with 2 mL of methanol: acetic acid: water (10:1:9). This dispersion was agitated using a vortex (1 min) and then put in an ultrasonic bath for 20 min (it was performed twice). For the quantification of the superficial anthocyanins: 500 mg of microcapsules were treated with 5 mL of isopropanol. The dispersion was agitated on a vortex at room temperature during 1 min and then filtered (0.45 µm Millipore filter). Samples were centrifuged at 5000×g at 4 °C during 10 min and then they were decanted. The moisture content of microcapsules was considered for the anthocyanins quantification. The encapsulation productivity (EP) and the encapsulation efficiency (EE) were calculated according to the following formula:

$$EP=\left(TA{C}_e/TA{C}_t\right)\times 100,$$ 5

$$EE=\left(1-SAC/TA{C}_e\right)\times 100,$$ 6

where TAC_e is the total anthocyanins content obtained experimentally (mg/g powder), and TAC_t is the total anthocyanins content theoretically calculated (mg/g powder) and SAC is the superficial anthocyanins content (mg/g powder).

Storage of microcapsules

For testing the storage stability, samples of microcapsules (1 g) were put into dishes. The dishes were placed into sealed flasks, each containing one of the following saturated solutions: LiCl, CH₃CO₂K, KCl, K₂CO₃, Mg(NO₃)₂, NaCl, KCl, BaCl. The relative humiditiy values ranged from 0.11 to 0.94 a_w. These samples were stored at 40 °C during 30 days in a dark room. The percent of anthocyanin at the end of this period was calculated according to the equation

$$AR=\left(TA{C}_{30}/TA{C}_e\right)\times 100,$$ 7

where TAC₃₀ is the total anthocyanins content after 30 days of storage, and TAC_e is the total anthocyanins content experimentally measured.

Sorption isotherms and inflection point

The Peleg model available in the literature (Peleg 1993) was evaluated by determining the best fit to the experimental data

$$m=A·a_w^B+C·a_w^D,$$ 8

where the m and a_w are equilibrium moisture content and water activity, respectively. The other symbols are the model constants. The fitting of Peleg model was determined by coefficient of determination (R²) and the mean relative percentage deviation modulus (E), which is defined as follows

$$E=\left(100/N\right)\sum _{i=1}^N\left|\left(M_i-M_{\mathit{pi}}\right)/M_i\right|,$$ 9

where M_i is the experimental value, M_pi is the predicted value, and N is the number of experimental data.

The critical relative (RHC), at which the glass transition occurs in microcapsules at constant temperature (40 °C), was determined by the method of Yuan et al. (2011) with modifications. Yuan et al. (2011) determined the RHC as the 2nd-derivative of a smooth cubic spline fit on the isotherm. The modification consits in calculating the 2nd-derivative of the resulting equation of the Peleg model, which leads us to an equation for calculating the inflexion point in Wolfram Mathematica.

Statistical analysis

Experiments were performed in triplicate. Sigma-Stat Version 3.01 (Systat 2004) was used to conduct an analysis of variance (ANOVA) to determinate differences between the means of the treatments. The means of the treatments were considered significantly different at P ≤ 0.05 using pairwise multiple comparison procedures (Tukey Test).

Results and discussion

Crystalline structure by XRD from starches and microcapsules

The diffractograms of native, hydrolyzed, succinylated starches and those of their corresponding microcapsules are shown in Fig. 1, Normal and waxy, maize starches showed peaks of high intensity at 2θ = 15°, 17°, 18° and 23°, which was an A-type diffraction pattern, characteristic of cereal starches (Meng et al. 2014).The acid hydrolyzed, normal (HNMS) and waxy (HWMS) maize starches; showed similar X-ray patterns as their native starch counterpart. Both HNMS and HWMS presented slightly sharper peaks at 2θ = 20°, and 2θ = 26.5°. Wang and Wang (2001) suggested that cleavage of starch chains in the amorphous regions allowed the reordering of the chain segments, which produced a more crystalline structure with a sharper X-ray pattern. The relative crystallinity in the NMS was 15%, while for the WMS was 28%. On the other hand, the relative crystallinity of HNMS was 19%, while for HWMS was 26.8%. These results suggested that acid hydrolysis affected the amorphous structure of the starches (Miao et al. 2011). Succinylated waxy maize starch was completely amorphous, however, succinylated normal maize starch after extrusion, showed a V-type pattern at angles of 2θ = 13.1°, 19.8° and a low intensity peak at 22.4°. Meng et al. (2014) attributed this phenomenon to the V-helical complexes between the starch and the palmitic acid, which involved the destruction of the double amylopectin helices through heating and high pressure homogenization, so that the palmitic acid can form an helical inclusion complex with the amylose molecules and V-helical complexes between the starch and the palmitic acid (in this work OSA). Normal potato starches modified by extrusion presented some peaks of crystallinity, though the waxy potato starches had a completely amorphous structure (O’Brien and Wang 2009).

Fig. 1.

X ray diffraction of NMS: normal maize starch; HNMS: hydrolyzed normal maize starch; SNMS: succinylated normal maize starch; MSNMS: microcapsules of succinylated normal maize starch; WMS: waxy maize starch; HWMS: hydrolyzed waxy maize starch; SWMS: succinylated waxy maize starch; MSWMS: microcapsules of succinylated waxy maize starch

Microcapsules of succinylated normal maize starches (MSNMS) as well as their respective SNMS shell material showed peaks at 2θ = 12.97°, 17.05° and 19.84° owing to the residual crystallinity present in the modified starches. However, diffractograms of MSWMS microcapsules from succinylated waxy maize starch showed a completely amorphous structure. Similar results were observed in microcapsules of passion fruit (passiflora) produced with commercial OSA starch, Borrmann et al. (2013) reported that the structure of such microcapsules is amorphous as well as MSWMS. Amorphous samples tend to be very hygroscopic.

Degree of substitution (DS), water solubility index (WSI), water absorption index (WAI) and crystallinity

The DS obtained for both SNMS and SWMS was higher for modified waxy maize starch than for normal maize starch (Table 1). This behavior was attributed to the content of amylopectin in starches. Thus, waxy maize starch swells more than normal maize starch during the early stage of succinylation, which resulted in products with higher DS (Luo and Shi 2012; Phillips et al. 2000).

Table 1.

Degree of substitution (DS), crystallinity, water solubility index (WSI) and water absorption index (WAI) of starches

Starches	DS	Crystallinity	WSI	WAI*
HNMS		19 ± 0.02b	1.00 ± 0.01e	1.06 ± 0.02c
SNMS	0.064b	7 ± 0.01c	57.1 ± 1.05c	5.31 ± 0.09a
MSNMS		4 ± 0.01d	33.46 ± 0.12d	N.D.
HWMS		26.8 ± 0.05a	0.27 ± 0.04f	2.05 ± 0.09b
SWMS	0.096a	0	74.13 ± 0.21b	1.56 ± 0.40b
MSWMS		0	96.04 ± 0.17a	N.D.

*HNMS: hydrolyzed normal maize starch, SNMS: succinylated normal maize starch, MSNMS: microcapsules of succinylated normal maize starch, HWMS: hydrolyzed waxy maize starch, SWMS: succinylated waxy maize starch, MSWMS: microcapsules of succinylated waxy maize starch. Results are mean ± standard error of mean value (n = 3). The mean values in the same column that have no common superscript are significantly different (P < 0.05)

Table 1, shows the WSI and WAI of starches, as well as of the microcapsules prepared with them. The solubility was higher in SWMS than in SNMS. According to these results, the degree of crystallinity may have an influence on both the WSI and WAI of succinylated starches after esterification and extrusion process, WSI was higher in MSWMS than in MSNMS, because MSWMS is more amorphous. Starch swelling is mainly due to the amylopectin content (Diop et al. 2011), thus in highly depolymerized starches, such as in SWMS, the crystallinity is lost and a lower WAI is observed than in SNMS. Therefore, SWMS was unable to retain water inside its structure, as a result of the extrusion process.

Morphology

In general, the microcapsules produced with succinylated normal and waxy maize starches have spherical shape with dented surfaces. This phenomenon was associated to a rapid drying rate which caused the solidification of the microsphere wall and dent smoothing could not occur (Su et al. 2008). The observed microsphere prepared with SNMS (see Fig. 2), shows pressure of a vacuole inside. The vacuole is created in the center of the microparticle during solvent evaporation; may be due to the great partial pressure generated inside (Porras-Saavedra et al. 2015). The active core was evenly distributed throughout the wall material matrix (Ré 1998). The anthocyanins microcapsules showed a particle size ranging from 1 to 23 µm.

Fig. 2.

SEM images of external and internal morphology of MSNMS (a, b) and MSWMS (c, d)

Physicochemical properties of microcapsules

Moisture content, water activity (a_w) and hygroscopicity are shown in Table 2. The moisture content in both MSNMS and MSWMS treatments was similar, while the value of a_w were higher in MSWMS than in MSNMS. The hygroscopicity obtained in this work is higher than that reported for microspheres of açai produced with maltodextrins (13.62 g/100 g) prepared under similar conditions, at 170 °C inlet temperature and 94 °C outlet temperature (Tonon et al. 2008). On the other hand, the hygroscopicity observed in this work is lower than the hygroscopicity values of 28.30 g/100 g and 24.60 g/100 g in betacyanin amaranthus powders produced with phosphorylated waxy maize starch and maize starch, respectively (Cai and Corke 2000).

Table 2.

Physicochemical characteristics of microcapsules

Analysis	MSNMS	MSWMS*
Moisture (%)	4.34 ± 0.03a	4.62 ± 0.02b
H	15.26 ± 0.28a	17.01 ± 0.22b
a w	0.19 ± 0.00a	0.26 ± 0.00b
EP (%)	95.29 ± 2.3a	90.26 ± 1.73b
EE (%)	97.37 ± 0.01a	96.54 ± 0.05a
TAC e	0.895 ± 0.06a	0.847 ± 0.02a

*MSNMS: microcapsules of succinylated normal maize starch, MSWMS: microcapsules of waxy maize starch. Results show the mean value ± standard error from 3 samples. The mean values in the same file that have no common superscript are significantly different (P < 0.05). H: hygroscopicity expressed in mg H₂O/g of powder

Total anthocyanins content expressed as mg of cyanidine-3-glucoside/g of microcapsules

Evaluation of microencapsulated purple maize anthocyanins

The quality of the microencapsulation is indicated by encapsulation productivity (EP), encapsulation efficiency (EE) and anthocyanin retention (AR) after storage (see Table 2). In the present work, EP was higher in MSNMS than in MSWMS. Furthermore, the value of EP obtained in this work was higher than the values reported for Myrciaria jaboticaba extract for several mixtures of Capsul™ and Maltodextrin as encapsulating agents (Silva et al. 2013). They reported a EP value of 80.11% in microparticles composed from Myrciaria jaboticaba dried at 180 °C (inlet temperature) y 79 °C (outlet temperature). García-Tejeda et al. (2015), using the same methodology on acetylated normal maize starch, it was obtained a value of 90.88% for EP, which is lower than the value obtained in the present work for the MSNMS microcapsules. On the other hand, for EE and TAC, there were no significant differences (P < 0.5) between MSNMS and MSWMS.

The agglomeration and caking of powders during storage are deleterious phenomena (Selvamuthukumaran and Khanum 2014). In order to investigate the effect of a_w in the anthocyanin (AR) on the microcapsules prepared with SNMS and SWMS, samples of both type of starches were stored during 30 days under different a_w (0.11–0.94), at 40 °C (Table 3). After the storage, it was observed that AR was higher in the microcapsules produced with SNMS than with SWMS. In general, it was observed that the higher the value of RH, the lower the value of AR. The free-flowing powder was observed at relative humidities below 53% for MSWMS and 75% for MSNMS, and above of those mentioned RH, there was observed agglomeration.

Table 3.

Anthocyanin retention values in microcapsules produced with succinylated starches after 30 days storage at 40 °C

a w	MSNMS (%)	MSWMS (%)*
0.11	99.16 ± 0.46a	63.08 ± 0.02a
0.22	87.25 ± 0.28c	62.78 ± 0.35a
0.33	81.27 ± 0.13d	53.48 ± 0.36c
0.42	80.76 ± 0.13d	49.00 ± 0.48d
0.52	76.49 ± 0.23e	48.66 ± 0.16d
0.75	69.54 ± 0.19f	43.33 ± 0.31e
0.84	56.91 ± 0.59 g	13.11 ± 0.40b
0.94	20.27 ± 0.21b	6.67 ± 0.07b

*MSNMS: microcapsules of succinylated normal maize starch, MSWMS: microcapsules of waxy maize starch. Results are mean ± standard error of mean value (n = 3). The mean values in the same column that have no common superscript are significantly different (P < 0.05)

The main cause of agglomeration is water-induced plasticization of the particle surface (Selvamuthukumaran and Khanum 2014). These physical changes are explained by the glass transition concept, the glass transition (T_g) is defined as a second-order phase change temperature at which a solid “glass” is transformed to a liquid-like “rubber” (Goula et al. 2008).

In order to determine the T_g from the sorption properties of the powders, an isotherm curve was fitted with the Peleg model from the experimental data, and the inflection point in the curve was determined. The resulting isotherm curves corresponding to experimental adsorption data for microparticles are shown in Fig. 3. The standard deviation for the equilibrium moisture content (g H₂O/100 of powder) of each experimental point was between 0.0003 and 0.001. The error associated to the Peleg model was of 5.96 and 4.9% for MSNMS and MSWMS, respectively. The equation for the Peleg models corresponding to MSNMS and MSWMS are the following

$$m=37.64a_w^{16.83}+16.13a_w^{0.67}$$ 10

$$m=50.12a_w^{9.55}+10.41a_w^{0.48}$$ 11

Fig. 3.

Water adsorption isotherms of anthocyanins microcapsules SNMS (filled square) and SWMS (open square) prepared at 40 °C. The solid and dashed lines showed the fitting data to Peleg model

The RHc is the inflection point in the isotherm curve evaluated at 40 °C, which is calculated from the second derivative of the Peleg model. The inflection points for MSNMS and MSWMS were 0.61 and 0.44, respectively (Fig. 3). The lower value corresponding to MSWMS is related to a higher degradation of anthocyanins.

At lower water activities, the isotherm showed a flat slope, but at the transition point where T_g occured, the isotherm had an inflection point. Above the T_g, the isotherm showed a higher water sorption (Yuan et al. 2011). After 30 days storage, MSNMS remained as free flowing powders at 0.52 a_w, however MSWMS became sticky powder. Microcapsules lose their structure at a_w = 0.94, which agreed with earlier findings (Liang et al. 2013), which reported a collapsed structure at 97% RH. These results are intrinsically related to the residual crystallinity in OSA normal derivative starch, which could favor the reinforcing of the microcapsule’s shell.

Conclusion

Both the succinylated normal and waxy maize starches showed good properties for encapsulating purple maize anthocyanins. The stability studies showed that succinylated normal starch was a better encapsulating agent than the succinylated waxy maize starch, owing to its higher encapsulation productivity and retention of anthocyanins after 30 days’ storage, at different water activities. Thus, the succinylated normal maize starches act as a good wall material for anthocyanins encapsulation by spray-drying.

Abbreviations

NMS

Normal maize starch

WMS

Waxy maize starch

HNMS

Hydrolyzed normal maize starch

HWMS

Hydrolyzed waxy maize starch

SNMS

Succinylated normal maize starch

SWMS

Succinylated waxy maize starch

MSNMS

Microcapsules of succinylated normal starch

MSWMS

Microcapsules of succinylated waxy starch

TAC

Total anthocyanins content

SAC

Superficial anthocyanins content

AR

Anthocyanin retention

EP

Encapsulation productivity

RHC

Critical relative humidity

Compliance with ethical standards

Conflict of interest

The authors have declared no conflict of interest.