Optimization for spray drying process parameters of nutritionally rich honey powder using response surface methodology

Abstract

The purpose of the present work was to study the effect of inlet temperature (160–180°C), feed rate (0.08–0.13 ml/s), concentration of gum Arabic (35–45%), aonla extract (6–8%), and basil extract (6–8%) on the product properties (bulk density, hygroscopicity, total phenolic content (TPC), antioxidant activity (AOA), and vitamin C content) of spray-dried nutritionally rich honey powder using response surface methodology. Higher inlet air temperatures led to lower bulk density and hygroscopicity, whereas addition of aonla and basil extracts led to higher TPC, AOA, and vitamin C content which were encapsulated by gum Arabic. Statistical analysis showed that independent variables significantly affected all the responses (p < 0.0001). Perturbation and 3D surface plots were drawn for each of the responses from the mathematical models. Second-order polynomial models with high R² (0.97–0.99) values were constructed for each powder physicochemical properties namely bulk density, hygroscopicity, TPC, AOA, and vitamin C content. Desirable nutritionally rich honey powder was obtained at inlet temperature of 170°C, 0.11 ml/s feed rate, 45% gum Arabic, 8% aonla extract, and 6% basil extract.

Keywords:

• honey
• spray drying
• response surface methodology
• antioxidant activity
• vitamin C content

Public Interest Statement

Honey is a very important energy food and sweetener. The difficulty in handling due to high viscosity of honey restricts its use in food formulations. So, it is preferable to be used in powder form and spray drying is considered as a good drying approach. Thus, in the present study, optimization of spray drying operating conditions for the production of spray-dried honey powder was done. The results can be helpful for food industries looking for the production of nutritionally rich honey powder with high retention of total phenolic content and vitamin C content. The results indicate that spray-dried nutritionally rich honey powder can be used in yoghurts, drinks, sauces, edible coatings, as well as dietary supplements and therapy-supporting preparations of bakery and meat industries to enhance oxidative stability.

Competing Interests

The authors declare no competing interest.

1. Introduction

Honey has been used since ancient times mainly for its sweetness which is contributed by monosaccharides—fructose and glucose (60–85%)—and possesses antioxidant activity (AOA) (Nayik & Nanda, 2015). It contains more than 180 substances including amino acids, vitamins, minerals, enzymes, organic acids, and phenol compounds (Nayik, Dar, & Nanda, 2015). However, industrial application of honey is restricted due to its highly viscous nature which causes difficulties in trade and handling. These problems may be overcome by converting liquid honey into powder form, resulting in an increase in its stability and ease in handling of the product (Samborska, Langa, Kamińska-Dwórznicka, & Witrowa-Rajchert, 2015; Suhag & Nanda, 2015). Spray drying is the widely used technique in food industries for commercial production of food powders due to low-cost and high-quality end product. During spray drying, evaporation occurs at a higher rate due to the increased surface area of liquid feed during atomization. Spray-dried powders owe good reconstitutional properties, longer shelf life, low water activity, and are suitable for transport and storage (Chegini, Khazaei, Ghobadian, & Goudarzi, 2008; Pang, Yusoff, & Gimbun, 2014). However, the production of sugar-rich powders by spray drying may present some problems such as stickiness and hygroscopicity, mainly contributed by low molecular weight sugars having low glass transition temperature. Drying carriers such as gum Arabic, maltodextrin, starches, proteins, and methyl cellulose are commonly employed to overcome the problem of stickiness (Shi, Fang, & Bhandari, 2013; Tonon, Brabet, & Hubinger, 2008). Gum Arabic is a hydrocolloid with polysaccharide chains containing small fractions of proteins which encapsulate sensitive components due to their good emulsifying capacity and low viscosity in aqueous solution. It has been found that it contributes to the retention of some food properties, such as nutrients, color, and flavor, during spray drying and storage (Igual, Ramires, Mosquera, & Martínez-Navarrete, 2014; Mosquera, Moraga, & Martínez-Navarrete, 2012). Appropriate drying compositions should be designed in order to improve encapsulation efficiency, reduce stickiness, and diminish compounds’ degradation. In an earlier study, the authors noted an increase in AOA, total phenolic content (TPC), and vitamin C content of honey due to addition of aonla and basil extracts after spray drying (Suhag & Nanda, 2015). Drying conditions are of vital importance, and variables such as inlet temperature, feed flow rate, and feed concentration are important factors that have to be controlled in a spray drying process. Experimental design, response surfaces, and multiple response analysis (so-called desirability approach) have been employed to evaluate and optimize several variables involved in the spray drying with a minimum number of experiments (Cortés-Rojas, Souza, & Oliveira, 2015; Prasad et al., 2011). Therefore, the aim of this study was to use response surface methodology (RSM) to optimize the inlet temperatures, feed rate, and concentration of carrier for the spray drying encapsulation of the honey powder in terms of bulk density, hygroscopicity, TPC, AOA, and vitamin C content.

2. Materials and methods

2.1. Materials

Samples of Helianthus annuus honey were collected with the help of local bee-keepers. The botanical origin of the samples of honey was based on the pollen spectrum (45% and above), which was the ratio of the frequency of each pollen type in honey (Louveaux, Maurizio, & Vorwohl, 1978). The following terms were used for frequency classes: predominant pollen (>45% of pollen grains counted), secondary pollen (16–45%), important minor pollen (3–15%), and minor pollen (<3%).

The chemicals Folin–Ciocalteau reagent, gallic acid (GA), and sodium carbonate were purchased from Loba Chemie Pvt. Ltd., Mumbai; 2, 2-diphenyl picryl hydrazyl (DPPH) was purchased from Fluka Goldie, Mumbai; acetone and methanol (HPLC grade) were purchased from Ranbaxy, New Delhi; and gum Arabic was from Loba Chemie Pvt. Ltd., Mumbai. Aonla (Neelam variety) was purchased from PAU (Punjab Agriculture University, Ludhiana) and basil leaves (holy basil) were purchased from a local farmer.

2.2. Honey blend preparation

Based on the results of preliminary experiments, honey was blended with water in the ratio 1:3.5 to avoid clogging caused by high viscous nature of honey. Gum Arabic was added at three different concentrations viz. 35, 40, and 45% of honey (w/w) as described in Table 2; (6–8%) each of aonla and basil extracts in liquid form was added in the feed mixture.

2.3. Spray drying

The solutions were fed into a tall-type laboratory-scale spray dryer (S.M. Scientech, Calcutta, India) with co-current arrangement and a pneumatic nozzle (two-fluid nozzle) atomizer having nozzle tip diameter of 0.7 mm and a 1.5-mm diameter nozzle screw cap was used. The range of droplet size of pneumatic nozzle atomizer was 5–300 μm. Feed was metered into the dryer by means of a peristaltic pump. The spray dryer can be operated at inlet temperature ranged from 160 to 180°C and feed flow rate from 0.08 to 0.13 ml/s. In all the experiments, outlet temperature and blower speed were kept at 90°C and 2,000 rpm, respectively. Nutritionally rich honey powder samples were collected in the glass bottle at the base of the cyclone and stored in airtight containers in desiccators containing silica gel for a week for further analysis (Suhag & Nanda, 2015).

2.4 Moisture, pH, % acidity, and HMF content of aonla, basil, and honey

The samples of honey were analyzed according to the AOAC Official Methods of Analysis (2000) methods for moisture, pH, % acidity, and hydroxymethylfurfural content, whereas moisture, pH, % acidity of aonla and basil were analyzed by methods as described by Rangana (2000). All results are expressed as the average of three replications.

2.5. Bulk density

Bulk density (g/ml) was measured by gently pouring the known mass of sample into an empty graduated cylinder and gently tapping 20–25 times and recording the volume (Goula, Adamopoulos, & Kazakis, 2004). All results are expressed as the average of three replications.

2.6. Hygroscopicity

One gram sample of powder was placed at 25°C in a container with NaCl saturated solution (75% RH). After one week, sample was weighed and hygroscopicity was expressed as g of adsorbed moisture per 100 g dry solids (g/100 g) as the average of three replications (Cai & Corke, 2000).

2.7. Total phenolic content

Briefly, 250 mg of sample was mixed with 10 ml of 60% acetone and the mixture was stirred for 30 min at 30°C. Then, 60 μl of supernatant, 300 μl of Folin–Ciocalteau reagent, and 750 μl of 20% sodium carbonate in water were added in 4.75 ml of water. After 30 min, absorbance was measured at 760 nm using UV–vis spectrophotometer (Hach DR 6000, Germany) with methanol as the reference. GA (0–100 mg/l) was used to produce a standard calibration curve. The TPC was expressed in mg of GA equivalents (mg GAE/100 g of spray-dried powder) as the average of three replications (Liu, Cui, & Zhao, 2008).

2.8. Antioxidant activity

Two hundred and fifty milligrams of sample were mixed with 10 ml of 60% acetone and the mixture was stirred for 30 min at 30°C. Two milliliters of extract were mixed with 2-ml methanolic solution containing 1 mM DPPH. The mixture was shaken vigorously and then left to stand for 30 min in the dark. The absorbance was measured at 517 nm using UV–vis spectrophotometer (Hach DR 6000, Germany). The absorbance of control was obtained by replacing the sample with methanol (Luo, Zhao, Yang, Shen, & Rao, 2009). DPPH radical scavenging activity of the sample was calculated as follows and results are expressed as the average of three replications.$$\text{DPPH}\text{radical}\text{scavenging}\text{activity}\left(\%\right)=\frac{\text{absorbance}\text{of}\text{control}-\text{absorbance}\text{of}\text{sample}}{\text{absorbance}\text{of}\text{control}}\times 100$$

2.9. Vitamin C content

Honey powder (1 mg) was treated with 20 ml of 0.4% oxalic acid at room temperature for 5 min and filtered through Whatman No. 4 filter paper. The filtrate (1 ml) was mixed with 9-ml 2,6-dichlorophenol-indophenol dye and absorbance was read within 15 min at 520 nm using UV–vis spectrophotometer (Hach DR 6000, Germany) against a blank. Vitamin C was calculated on the basis of the calibration curves of vitamin C, and was expressed as mg/100 g of vitamin C as the average of three replications (Egoville, Sullivan, Kozempel, & Jones, 1998).

2.10. Experiment design

For this study, RSM based on the multivariate nonlinear model was used to evaluate the effect of five process independent variables on five response variables mainly related to the profitability of the process and the quality of the powder (Montgomery, 2001). This method was applied using Design-Expert version 8.0.7.1 (Statease Inc., Minneapolis, USA) to identify the optimum levels of the five variables, i.e. temperature (°C), feed rate (ml/s), concentration of gum Arabic (%), aonla extract (%), and basil extract (%), regarding five responses: bulk density, hygroscopicity, TPC, AOA, and vitamin C of spray-dried honey powder. A five-factor and three-level Box–Behnken design (Tekindal, Bayrak, Ozkaya, & Genc, 2012) consisting of 46 experimental runs including six replicates at center point was employed for this purpose. The order of experiments was fully randomized. These data were analyzed by multiple regressions using the least squares method.

Results were adjusted to a second-order polynomial model according to the following equation:$$Y_k={\mathit{\beta }}_{k0}+\sum _{i=1}^n{\mathit{\beta }}_{ki}{x}_i+\sum _{i=1}^n{\mathit{\beta }}_{kii}{x}_i^2+\sum _{i=1}^{n-1}\sum _{j=i+1}^n{\mathit{\beta }}_{kij}{x}_i{x}_j$$

where Y_k = response variable; Y₁ = bulk density (g/ml); Y₂ = hygroscopicity(%); Y₃ = TPC (mg GAE/100 g); Y₄ = AOA (%); and Y₅ = Vitamin C (mg/100 g); x_i represents the coded independent variables (x₁ = temperature of inlet air, x₂ = feed flow rate, x₃ = concentration of gum Arabic, x₄ = aonla extract, and x₅ = basil extract); where β_k0 was the value of the fitted response at the center point of the design, i.e. point (0,0,0), β_ki, β_kii, and β_kij were the linear, quadratic, and cross-product regression coefficients, respectively. The test of statistical significance was performed on the total error criteria, with a confidence level of 95%.

The significant terms in the model were found by analysis of variance (ANOVA) for each response. The goodness of the fit of the final reduced models to the experimental data was evaluated from R², adj-R², Pred R², and Fisher’s F-tests. The mathematical model is reliable with an R² value closer to 1. The regression of coefficients was then used to make statistical calculation to generate three-dimensional plots from the regression model. These graphical presentations were employed since they permit the visualization of the responses’ behavior as affected by the variables investigated. The goal of this optimization procedure was to obtain processing conditions which give higher TPC, AOA, vitamin C content, low bulk density, and hygroscopicity.

Correlation coefficient between TPC and AOA of spray-dried honey powder was calculated by correlation graphics using software Statistica 7.0.

3. Results and discussion

The moisture content, pH, and % acidity of honey, aonla, and basil extract were found to be 18.8%, 82.5%, 82.3%, 3.9, 3.1, 5.3, 0.4, 2.5, and 1.9%, respectively. The HMF content of honey was observed to be 5.49 mg/100 g, whereas it was not detected in aonla and basil extracts (Table 1). The experimental results obtained for each response variable are shown in Table 2. The final reduced models relating each response variable with the independent variables are shown in Table 3. The adequacy of the response surface equation was checked by the comparison of experimental and predicted values (Table 4). For any of the terms in the model, a large regression coefficient and a small p-value would indicate a more significant effect on the respective response variables. ANOVA showed that the resulting quadratic model adequately represented the experimental data with coefficients of multiple determinations (R²) of 0.98, 0.98, 0.99, 0.99, and 0.99 for the responses of bulk density, hygroscopicity, TPC, AOA, and vitamin C, respectively.

Table 1. Physicochemical analyses of honey, aonla, and basil extracts

Properties	Honey	Aonla extract	Basil extract
Moisture (%)	18.8 ± 1.06	82.5 ± 0.72	82.3 ± 0.89
pH	3.9 ± 0.33	3.1 ± 0.16	5.3 ± 0.24
Acidity (%)	0.4 ± 0.03	2.5 ± 0.17	1.9 ± 0.14
HMF (mg/100 g)	5.49 ± 0.28	ND	ND
Vitamin C (mg/100 g)	ND	449 ± 4.05	7.3 ± 0.35
Antioxidant activity (%DPPH)	61.89 ± 1.08	69.94 ± 1.48	65.36 ± 0.63

Notes: HMF—hydroxy methy furfural(mg/100 g) and TSS—total soluble solids(˚B). Values are mean of triplicate determinations. ND—Not detected.

Table 2. Effect of independent variables on variables of responses (bulk density (BD), hygroscopicity, TPC, AOA, and vitamin C of developed honey powder)

S. No.	Inlet temp (°C)x1	Feed rate (ml/s)x2	Gum arabic (%)x3	Aonla extract (%)x4	Basil Extract (%)x5	BD (g/ml)	Hygroscopicity (%)	TPC (mg GAE/100 g)	AOA (%)	Vitamin C (mg/100 g)
1	160	0.08	40	7	7	0.865	28.77	79.41	61.39	88.58
2	180	0.08	40	7	7	0.617	25.45	75.58	56.16	83.55
3	160	0.13	40	7	7	0.893	28.99	79.41	61.31	88.58
4	180	0.13	40	7	7	0.611	25.91	75.46	56.27	83.59
5	170	0.11	35	6	7	0.793	28.21	76.41	58.53	85.36
6	170	0.11	45	6	7	0.662	25.68	78.47	60.41	87.03
7	170	0.11	35	8	7	0.837	28.19	76.88	58.75	85.79
8	170	0.11	45	8	7	0.626	25.41	78.94	60.88	87.26
9	170	0.08	40	7	6	0.734	27.25	77.39	59.37	86.28
10	170	0.13	40	7	6	0.721	27.76	77.31	59.31	86.36
11	170	0.08	40	7	8	0.728	27.31	77.63	59.75	86.76
12	170	0.13	40	7	8	0.731	27.92	77.58	59.13	86.68
13	160	0.11	35	7	7	0.914	29.11	79.16	60.73	88.19
14	180	0.11	35	7	7	0.626	26.03	74.77	55.93	82.56
15	160	0.11	45	7	7	0.818	27.61	80.62	62.64	88.81
16	180	0.11	45	7	7	0.521	24.03	76.93	57.55	84.31
17	170	0.11	40	6	6	0.742	27.65	76.93	59.26	86.19
18	170	0.11	40	8	6	0.754	27.51	77.19	59.57	86.61
19	170	0.11	40	6	8	0.762	27.59	77.81	59.88	86.95
20	170	0.11	40	8	8	0.769	27.45	77.34	59.12	86.07
21	170	0.08	35	7	7	0.813	28.03	76.57	58.46	85.66
22	170	0.13	35	7	7	0.782	28.31	76.65	58.61	85.58
23	170	0.08	45	7	7	0.653	25.57	78.83	60.58	87.19
24	170	0.13	45	7	7	0.634	26.24	78.74	60.63	87.14
25	160	0.11	40	6	7	0.876	28.69	79.35	61.24	88.32
26	180	0.11	40	6	7	0.592	25.85	75.31	56.34	83.47
27	160	0.11	40	8	7	0.882	28.61	79.63	61.52	88.74
28	180	0.11	40	8	7	0.587	25.67	75.72	56.82	83.74
29	170	0.11	35	7	6	0.772	28.28	76.47	58.38	85.46
30	170	0.11	45	7	6	0.683	26.16	78.65	60.52	87.08
31	170	0.11	35	7	8	0.825	28.15	76.72	58.68	85.71
32	170	0.11	45	7	8	0.619	25.43	79.32	60.93	87.38
33	160	0.11	40	7	6	0.855	28.56	79.23	61.19	88.46
34	180	0.11	40	7	6	0.574	25.72	75.52	56.51	83.41
35	160	0.11	40	7	8	0.846	28.49	79.55	61.46	88.63
36	180	0.11	40	7	8	0.562	25.56	75.64	56.72	83.65
37	170	0.08	40	6	7	0.748	27.38	77.34	59.12	86.07
38	170	0.13	40	6	7	0.769	27.87	77.12	59.08	86.01
39	170	0.08	40	8	7	0.756	27.17	77.73	59.63	86.89
40	170	0.13	40	8	7	0.739	27.81	77.68	59.67	86.83
41	170	0.11	40	7	7	0.719	26.88	77.48	59.53	86.55
42	170	0.11	40	7	7	0.716	26.74	77.42	59.71	86.51
43	170	0.11	40	7	7	0.713	26.68	77.53	59.47	86.42
44	170	0.11	40	7	7	0.708	26.54	77.59	59.43	86.47
45	170	0.11	40	7	7	0.704	26.47	77.61	59.55	86.38
46	170	0.11	40	7	7	0.701	26.38	77.46	59.59	86.33

Table 3. Significant regression coefficients of second-order polynomial Equation (1) for the responses

Coefficients	Bulk density	Hygroscopicity	TPC	AOA	Vitamin C
β0	0.710^**	26.615^**	77.515^**	59.546^**	86.443^**
β1	−0.141^**	−1.538^**	−1.964^**	−2.448^**	−2.501^**
β2	–	0.242^**	–	–	–
β3	−0.0716^**	−1.136^**	1.054^**	1.004^**	0.743^**
β4	–	–	0.148^**	0.131^**	0.158^**
β5	–	–	0.181	–	0.123
β12	–	–	–	–	–
β13	–	–	0.175^*	–	0.282^**
β14	–	–	–	–	–
β15	–	–	–	–	–
β23	–	–	–	–	–
β24	–	–	–	–	–
β25	–	–	–	–	–
β34	−0.02^*	–	–	–	–
β35	−0.029^*	–	–	–	–
β45	–	–	−0.182^*	−0.267^**	−0.325^**
${\mathit{\beta }}_1^2$	–	–	–	−0.542^**	−0.400^**
${\mathit{\beta }}_2^2$	0.017^**	0.517^**	–	−0.151^*	–
${\mathit{\beta }}_3^2$	–	–	0.274^**	0.160^*	–
${\mathit{\beta }}_4^2$	0.025^**	0.433^**	–	–	–
${\mathit{\beta }}_5^2$	–	0.437^**	–	–	–
R^2	0.98	0.98	0.99	0.99	0.99
Adjusted R^2	0.96	0.97	0.98	0.98	0.98
Pred R^2	0.92	0.94	0.97	0.97	0.97
Adeq. Precision	35.19	38.55	56.35	57.19	54.18
Lack of fit	0.015	0.424	0.036	0.069	0.033

* Significant at p < 0.05.

** Significant at p < 0.0001.

Table 4. Experimental and predicted values of responses of nutritionally rich honey powder

Response	Experimental result	Predicted result
Bulk density (g/ml)	0.619	0.663
Hygroscopicity (%)	25.43	25.95
Total phenolic content (mg GAE/100 g)	79.32	79.61
Antioxidant activity (%)	60.93	61.54
Vitamin C content (mg/100 g)	88.46	87.87

3.1. Bulk density

Bulk density is important for packaging and shipping operations. The bulk density includes the volume of the materials and all the pores, closed or open, to the surrounding atmosphere, and is generally used to characterize the final product obtained after drying. Linear terms for both variables like inlet temperature and concentration of gum Arabic (presented in Table 3) had a strong significant influence on powder bulk density (p < 0.0001). The coefficient of determination (R²) of the fitted model was 0.98. This value indicates that the model obtained from results explains 98% of variation of the observed data (Table 3). A decrease in bulk density was observed with increase in inlet temperature and the decrease in gum Arabic concentration (Figure 1(a)). The reason behind decrease in bulk density of the powder samples may be the higher evaporation rates, resulting in powders with hollow and porous structures. Similar effect of the inlet air temperature on bulk density was also observed by Julio, Rodriguez, Tonon, and Hubinger (2014) for spray-dried blue shark skin protein hydrolysate. Moreover, the lowest hygroscopicity values were obtained at higher gum Arabic concentrations which might be related to their high degree of agglomeration and structural collapse, resulting in subsequent decrease in volume of the powder particles. A similar trend was reported by Fazaeli, Emam-Djomeh, Kalbasi Ashtari, and Omid (2012) for spray drying of black mulberry juice.

Figure 1. Response surface plots for the (a) bulk density (b) hygroscopicity (c) TPC (d) AOA, and (e) vitamin C as a function of inlet temperature and concentration of gum Arabic.

3.2. Hygroscopicity

Honey contains high content of low molecular weight sugars which makes it hygroscopic and hence makes the drying process of honey quite difficult. The effect of process variables on hygroscopicity of the honey powder is shown in Figures 1(b) and 2. Inlet temperature and feed flow rate influenced the hygroscopicity of the powder samples. The hygroscopicity of honey powder decreased (p ˂ 0.0001) with increase in inlet temperature which may be attributed to greater temperature gradient between the atomized feed and the drying air, resulting in a higher rate of heat transfer for water evaporation. This result is in line with the previous works done by Muzaffar and Kumar (2015) on tamarind pulp powder and Santana, Kurozawa, de Oliveira, and Park (2013) on spray-dried pequi powder. According to Figure 1(b), gum Arabic concentration also showered significant positive effect on hygroscopicity. With increase in gum Arabic concentration, decrease in hygroscopicity was observed which might be the result of less hygroscopic nature of this encapsulating agent. Similar results were demonstrated by Igual et al. (2014) and Kurozawa, Morassi, Vanzo, Park, and Hubinger (2009) for spray drying of lulo pulp and chicken meat protein hydrolysate, respectively.

Figure 2. Response surface plot for the hygroscopicity as a function of inlet temperature and feed rate.

3.3. Total phenolic content

The regression analysis report (Table 3) showed that inlet temperature had a negative linear while gum Arabic concentration, aonla extract, and basil extract had positive linear effects on TPC of honey powder (p < 0.05). It was observed from the response surface plot (Figure 1(c)) that the higher inlet temperature led to significant reduction of phenolic compounds because the components of honey powder polyphenols which are responsible for the AOA of honey get easily oxidized; their molecular structure gets altered; and also gets thermally degraded during the spray drying process. Quek, Chok, and Swedlund (2007) reported that the spray drying of watermelon juice at over 165°C led to inferior products due to greater loss of bioactive compounds, and spray drying at temperatures of 150°C was not suitable for phenolic compounds of lulo pulp powder (Igual et al., 2014). Figures 1(c) and 3 show the significant interactions of independent variables, namely: gum Arabic and inlet temperature and aonla extract and basil extract on the TPC (p < 0.05). Retention of phenolic content of honey powder was increased with addition of aonla and basil extracts which indicated that TPCs were well encapsulated by gum Arabic. This can be explained by the fact that gum Arabic possesses good emulsifying capacity, thus efficiently able to encapsulate sensitive components against oxidation. Murugesan and Orsat (2011) found that the gum Arabic application could provide better encapsulation efficiency, size, stability, and oxidation resistance for spray-dried elderberry juice powder.

Figure 3. Response surface plot for the TPC as a function of aonla extract and basil extract.

3.4. Antioxidant activity

Surface responses (Figures 1(d) and 4(a)) show the effect of independent variables on AOA of the honey powder. The optimization study illustrated that inlet temperature adversely affected the AOA of honey powder in all the experiments, depicting that AOA of the honey powder was significantly decreased with increase in inlet temperature due to molecular disintegration of phenolic compounds. Kha, Nguyen, & Roach, 2010 reported that increased significant loss of AOA was observed from 0.14 to 0.08 mmol TE/g of powder with increase in inlet temperature from 120 to 200°C.

Figure 4. Response surface plots for the (a) AOA and (b) vitamin C as a function of inlet temperature and concentration of aonla extract.

A significant positive influence of gum Arabic, aonla, and basil extract addition on AOA responses can be clearly seen in the graphics of honey powder samples (p < 0.0001). Gum Arabic microcapsules promoted a gradual increase in phenolics content and AOA of honey powder due to their high viscosity, resulting in an increase in the time of droplet formation and internal mixing during spray drying. Similar results were reported by Murali, Kar, Mohapatra, and Kalia (2014) for black carrot juice powder encapsulation, where gum Arabic retained maximum AOA activity. The interaction between aonla and basil extracts had a significant positive linear effect on AOA (p < 0.0001) (Table 3). The reason might be the addition of aonla and basil extracts which are rich sources of natural antioxidants. Pearson’s statistical correlation analysis was used to establish a positive correlation between the TPC and AOA honey powder in Figure 5 (r = 0.973). Kha et al. (2010) reported that a high correlation between total carotenoid content and the total AOA was found in spray-dried gac powder.

Figure 5. Relation between TPC and AOA measured by correlation graphics.

3.5. Vitamin C content

Vitamin C content was not detected in honey; however, in the present study, it was found that vitamin C content of honey powder increased with the incorporation of aonla extract. The regression analysis report (Table 3) showed that inlet temperature had a negative linear effect, while gum Arabic concentration and aonla extract had positive linear effects on vitamin C content of honey powder (p < 0.0001). It was observed from the response surface plot (Figure. 1(e)) that increase in inlet air temperature caused a reduction in vitamin C of honey powder due to oxidation and hydrolysis, while gum Arabic increased retention of vitamin C content which can be accredited to the better emulsifying properties of gum Arabic. Similar reports of loss of vitamin C with increase in inlet temperature were reported by Patil, Chauhan, and Singh (2014) during the production of spray-dried guava powder. Santana et al. (2013) revealed that gum Arabic microcapsules acted as good encapsulates for retention of vitamin C content of spray-dried pequi powder. Vitamin C of honey powder increased due to incorporation of aonla extract in the present study (Figure. 4(b)). Agarwal and Chopra (2004) reported that aonla is the richest source of vitamin C content.

3.6. Optimization

Spray-dried nutritionally rich honey powder could be considered an optimum product if the criteria applied to achieve the optimization resulted in the highest TPC, AOA, and vitamin C contents as well as the lowest bulk density and hygroscopicity. Multiple response optimization suggested that the optimal conditions for producing the best spray-dried honey powder were attained at an inlet air temperature of 170°C, feed rate of 0.11 ml/s, 45% gum Arabic, 8% for aonla extract, and 6% for basil extract. Table 4 shows the comparison between experimental and predicted results of selected properties of the spray-dried product generated. Validation step showed a good agreement between the predicted and experimental data. Thus, response surface optimization suitably predicted the optimum conditions.

4. Conclusion

The optimization of the spray drying conditions for the honey powder was successfully executed using the Box–Behnken of the RSM. Honey powder with the retention of vitamin C and AOA could be produced by spray drying at 170°C temperature with 0.11 ml/s feed rate and by the addition of aonla (Emblica officinalis) extract, basil (Ocimum Sanctum) extract, and gum Arabic with low bulk density and hygroscopicity. Strong positive correlation among TPC and AOA of nutritionally rich honey powder developed was also confirmed (r = 0.973). The results obtained in this work showed the importance of the simultaneous investigation of processing and feed composition variables during the development of optimized dried honey powder.

Acknowledgments

The authors are very much grateful to local bee-keepers of Punjab for providing sunflower honey samples and also to the Department of Vegetable Science, Punjab Agricultural University, Ludhiana for providing aonla variety.

Additional information

Funding

Funding. The authors received no direct funding for this research.

Notes on contributors

Yogita Suhag

Yogita Suhag is a research scholar in the Department of Food Engineering and Technology, SLIET, Longowal. She completed her master’s in Food Engineering and Technology. Her area of interest is drying technology, fruit and vegetable processing, and nutraceuticals. She has published five research papers in international journals.

Vikas Nanda

Vikas Nanda is a professor, in Department of Food Engineering and Technology, SLIET, Longowal, Punjab, India. He earned MSc and PhD degrees in Food Technology, from Punjab Agricultural University, Ludhiana and postgraduate course in product development and handling of analytical tools from University of Jerusalem, Israel. Through the various international conferences at the Netherlands, Argentina, Greece, France, Czech Republic, and Thailand, he extended his knowledge in value addition of cheese, evaluating physical and rheological properties of food products. He is also the vice chairman of International Honey Commission. He has more than 25 international research papers to his credit and also published one book.