Development and optimization of spray‐dried functional oil microcapsules: Oxidation stability and release kinetics

Abstract This study aimed to optimize the microencapsulation method for a functional oil using high amylose corn starch (HACS) and assessed its structure and antioxidant capacity. The results showed that the optimal microencapsulation condition is achieved by using 28.5% of functional oil, 15.75% of HACS, and 57.86% of proportion of monoglyceride in emulsifier with 94.86% microencapsulation efficiency. Scanning electron microscopy and particle size measurement showed that the functional oil microcapsules were uniform size, smooth surface, spherical shape, and without cracks in the wall of the capsules. In vitro oil release of microencapsulates results showed that microencapsulated functional oil containing HACS has a better sustained release effect. The microcapsules containing HACS exhibited a lower lipid oxidation rate during storage. In conclusion, microencapsulation of HACS as wall material improved the stability of functional oil and this formulation of microcapsules was satisfactorily applied in powdered food for diabetic patients.

Although functional oil has many health benefits, there are safety problems associated with functional oil rich in ω-3 PUFA. The main drawback of functional oils is the rapid oxidation of PUFAs, which can lead to the formation of toxic substances (Barroso, Pierucci, Freitas, Torres, & Rocha-Leao, 2014;Binsi et al., 2017).
Microcapsule technology is a mature industrial technology, and it has been widely used in food, medicine, and other fields through interdisciplinary research in recent years (Gharsallaoui, Roudaut, Chambin, Voilley, & Saurel, 2007). Among the various techniques used for microencapsulation, due to its simple operation, small footprint, and low cost, spray drying is the most widely used in the food industry (Tan, Chan, & Heng, 2005). The common wall materials are divided into three major categories: carbohydrates, proteins, and hydrophilic colloids (Ahn et al., 2008;Krishnan, Kshirsagar, & Singhal, 2005). Glucose index in diabetic patients fluctuates greatly when they intake carbohydrates, so the choice of wall materials is particularly critical in the embedding of functional oils. The Food and Drug Administration (FDA) indicated that high amylose corn starch (HACS) can reduce the risk of type 2 diabetes mellitus. Several studies reported that HACS intake significantly reduces the weight gain and lipid profiles of mice fed with high-fat diet (Lee, Yoo, & Lee, 2012;Shimotoyodome, Suzuki, Fukuoka, Tokimitsu, & Hase, 2010). Previous researches confirmed that the application of microcapsule technology has positive effects on the relief of obesity or diabetes (Arifin et al., 2011;Li, Wu, & Dou, 2019;Mooranian, Negrulj, & Al-Salami, 2016;Schneider et al., 2005). Hence, the functional oil microcapsules used HACS as wall materials may be highly advisable to control the condition of diabetic patients (Kieffer et al., 2016;Maki et al., 2012).
In addition to preventing oxidation, microencapsulation can also control the release of core materials ingredients (Koupantsis, Pavlidou, & Paraskevopoulou, 2016). In general, carbohydrates and proteins are not conducive to controlled release applications when used as wall materials (Cho, Shim, & Park, 2003). However, the content of resistant starch in HACS is relatively high, and the stability of microcapsules would be greatly increased when HACS was used as the wall material. Therefore, the purpose of the current study was to optimize the conditions for preparing functional oil microcapsules with HACS during spray drying. Besides, the structural, oxidative stabilization and sustained release effects of the microcapsules were also evaluated.

| Materials
Chia seed oil was provided by Aarhus Kars Oil Co., Ltd; peanut oil was provided by Shandong Jinsheng Grain and Oil Group Co., Ltd; and olive oil, maltodextrin, sodium caseinate, and HACS were purchased in local markets. All chemical reagents (analytical grade) were purchased from Sinopharm Chemical Reagent Co., Ltd.

| Preparation of functional oil
Chia seed oil, peanut oil, and olive oil were used as raw materials, and the functional oil was prepared according to the ratio of Chia seed oil:peanut oil:olive oil = 1:8:1. The ratio of saturated fatty acids (SFA), monounsaturated fatty acids (MUFA) and PUFA in functional oils was 1:2:2, and the relative content ratio of ω-6 PUFA to ω-3 PUFA was 5:1. The samples were stored at 4°C until used.

| Preparation of emulsions
The wall material which includes the maltodextrin, sodium caseinate, and HACS was dissolved in water (65°C). The functional oil was emulsified at 65°C. Functional oil and wall materials were emulsified for 5 min at 8,950 g.

| Spray drying of emulsions
After a stable emulsion was obtained by three times of 30 MPa homogenization (Changzhou Homogenizer Machinery Corporation, Ltd., GJB 8-20), spray drying was employed using a small-scale spray dryer (Buchi, model 290). The operating conditions were as follows: Inlet temperature was 185°C, outlet temperature was about 90°C, and feed rate was 5 ml/min.

| Microcapsule formulation optimization
Design-Expert 8.0.6.1 software was used to design the experiments. Response surface methodology (RSM) by a 3-factor-3-level Box-Behnken design was used to investigate the variation of MEE with respect to formula parameters including HACS, functional oil content, and percentage of monoglyceride in emulsifier. The variables and their concentration ranges are as follows: core material (functional oil) content (X 1 ) from 20% to 40% (w/w), proportion of HACS in wall polymers (X 2 ) from 10% to 30% (w/w), and proportion of monoglyceride in emulsifier (X 3 ) from 50% to 70%. The actual variable was coded to facilitate multiple regression analysis (

| Determination of bulk density
Accurately weighed 5 g of functional oil microcapsule samples, loaded them into graduated cylinders, measured and calculated their volume, and calculated the mass per unit volume of microcapsules.

| Fourier transform-infrared spectroscopic analysis (FT-IR)
The FT-IR was used to analyze functional oil, HACS, maltodextrin, sodium caseinate, and functional oil microcapsules.

Solid sample analysis
Accurately weigh 1 mg of microcapsule samples. Potassium bromide mass ratio = 1:100. After mixing and grinding in a mortar, the tablets were compressed and scanned in the range of 400-4,000 cm −1 .

Liquid sample analysis
A small amount of functional oil samples was pipetted by a dropper and uniformly applied to the prepared potassium bromide tablet for infrared analysis, and scanned in the range of 400-4,000 cm −1 .

| Particle size measurement
The functional oil microcapsule samples were uniformly dispersed in distilled water at room temperature. The particle size distribution of the functional oil microcapsules was determined by BT-2001 laser particle size analyzer, and data sampling analysis was performed.

| Morphological study
The surface and internal morphology of functional oil microcapsules was observed using a scanning electron microscope (SEM). Attach one side of the double-sided tape to the sample stage, then stick small amount of microcapsule samples on it, and spray the gold (thickness 100 μm). The structure is observed and photographed at an acceleration voltage of 10 kV.

| Storage stability of microcapsules
The functional oil and microcapsules were placed in 62°C ± 2°C oven for accelerated oxidation experiments. The change in peroxide value was measured every 24 hr during the heating period of 7 days.

| Optimization of microcapsule sample by RSM
The RSM was used to evaluate the impact of multiple parameters on response variables. Based on the coding levels of the three independent variables (Table 1), a three-factor BBD was used to obtain 17 simplified experimental set ( Table 2) The optimal formula for the preparation of functional oil microcapsules was 28.5% of functional oil, 15.75% of HACS, and 57.86% of proportion of monoglyceride in emulsifier. The MEE of functional oil microcapsules was theoretically 95.01% under these conditions.
Repeated experiments showed that the average MEE of functional oil microcapsules was actually 94.86%, which was close to the theoretical optimal value and good repeatability. Furthermore, the models explain 96.3% of the variability in the responses.

| Scanning electron microscopy (SEM)
As shown in Figure 2, SEM was used to observe the morphology of microcapsules under optimized conditions. MEFO at the optimized condition (94.86% in MEE) showed approximately spherical, exhibited smooth and free of pores, the cyst wall is relatively complete, compact and without apparent fissures or cracks. Moreover, the shape of MEFO appeared to be slightly rough and concave on the surface due to instability during the drying process. Teixeira et al.
reported that depressions and roughness of the surface are more prevalent in small particles than in larger ones, which indicated that the microcapsules expanded into a round shape and then wall mate-

| Moisture content and bulk density
Moisture content of MEFO is a parameter closely related to oil oxidation. The average moisture content of MEFO was 1.88% (Table 4).
The reason for the low moisture content in MEFO was mainly the

| FT-IR spectra of microcapsules
FT-IR spectroscopy analysis was used to explore the interaction of wall materials and functional oil in MEFO (Figure 3). At the 3,600-3,300 cm −1 , since maltodextrin and HACS a large amount of OH.
When applied to MEFO, this characteristic peak also exists obviously. In the region from 1,800 to 1,400 cm −1 , the characteristic peak may be caused by the stretching vibration peak caused by C=C.
The characteristic peaks of functional oil appeared at 2,990 cm −1 ,

| Powder X-ray diffraction
As shown in Figure 4, the microcapsule samples were amorphous; however, the HACS has a semicrystalline structure. The cause of this phenomenon may be that the oil interacted with the HACS hydrophobic group, resulting in the disappearance of the crystal structure. The highly ordered state of the molecules caused sharp and clear peaks in the crystalline material, while the disordered display of amorphous  molecules was responsible for the appearance of diffuse and large peaks in amorphous materials (Yu, 2001). The amorphous structure in the microcapsule promotes high dispersibility in water, which is one of the important properties of its application in the food industry.

| Particle size measurement
The particle size measurement showed that the minimum range of microcapsules appeared was 0.29-0.36 μm, and the particle size distributions of D10, D50, and D90 were 0.795, 3.871, and 14.45 μm, respectively. The average particle size was 6.064 ± 0.38 μm, and the smaller or larger diameter particles were less. The average particle size in the current study is lower than that observed in other oil microcapsules studies. Pu, Bankston, and Sathivel (2011) used sodium caseinate as the wall material and flaxseed oil as the core material to obtain microcapsules with an average diameter of 25.1 μm. Reineccius (1989) found that particles within 100 mm increased the solubility of the powder, so smaller particle sizes were important for microcapsules.

| Effect of SGF and SIF on oil release
In order to evaluate the release pattern of microcapsules in vitro, simulated digestion was used to test the stability and release be- Although strong acidic environment had a certain threat to the protein in microcapsules, the high viscosity of HACS and emulsifier made the wall materials of microcapsules more intact and prevented the release of core materials. When digested in small intestine for 100-250 min, the core material of functional oil microcapsules was released in large quantities due to the decomposition of trypsin and the acid-alkaline environment. At the end of 250 min, the total release of microcapsules with different formulations was similar, and more than 80% of functional oil was released.

| Storage stability analysis
In order to ascertain the antioxidant stability of microcapsules, the change of peroxide value (POV) in the outer free oil of and inner oil of MEFO during the accelerated storage of the oven at 62 ± 2°C was observed. As shown in Figure 6, initial POV of functional oil was 4.0 meq/kg and initial POV of functional oil after microencapsulation was 4.9 meq/kg. This result indicated that the high temperatures during spray drying caused the oil to oxidize to some extent before getting microencapsulated. However, the oil oxidation was greatly retarded by microencapsulation during stored. The functional oil was similar to the oxidative trend of MEFO at first, but the POV content increased significantly after 2 days of stored. At the end of the accelerated oxidation experiment, the peroxide values of functional oil and microcapsules increased to 42.5 and 10.3 meq/kg, respectively. These results indicated that the MEFOS has superior oxidative stability, which prevented diffusion of functional oil and against oxygen from contact with the core material. Previous studies found the similar results in microencapsulation. Asensio et al. (2017) reported that the composition of wall material is critical in antioxidation. Kolanowski, Jaworska, Weiβbrodt, and Kunz (2007) found that microencapsulated fish oil oxidized rapidly in the presence of air, which was improved after microencapsulation.

| CON CLUS ION
The purpose of the current study was to demonstrate the feasibility of HACS as wall material for the production of functional oil microcapsules by spray drying and optimize the microencapsulation condition. The optimal microencapsulation condition obtained through RSM was 28.5% of functional oil, 15.75% of HACS, and 57.86% of proportion of monoglyceride in emulsifier. The MEE of functional oil microcapsules was 94.86% under these conditions. The results of SEM and particle size measurement showed that the addition of HACS did not affect the uniformity and surface morphology of the particles. FT-IR spectroscopy analysis of the microcapsules con- incorporation of HACS in functional oil microcapsules in spray drying may be advocated, the process of encapsulation itself not only improved the oxidative stability of the oils, but also enhance the nutritional value of microcapsules. The use of this microencapsulation design may result in increased utilization of functional oil in food industry, which may help improve the health of diabetic patients.

CO N FLI C T O F I NTE R E S T
All authors declared that they have no personal or financial conflicts of interest.