1 Introduction

Mango (Mangifera indica L.) is one of the dominant tropical fruits with worldwide acceptability and account for more than half of the total global tropical fruit production [1]. With production area of 2.26 million acres, mango is considered as the third largest agricultural commodity in India [2]. Postharvest processing of mango generates a large quantity of byproducts including peels and seeds which raises many environmental as well as economical concerns [3]. An increasing interest has been sparked in the scientific community for the valorization of mango byproducts based on its therapeutic properties and health-enhancing effects [4]. Mango peels account for 15–25% of the total weight of fruit and considered a critical source of dietary fiber, polyphenols, carotenoids, vitamin C, etc. [5].

Dietary fiber imparts several health benefits and is widely utilized as a functional food ingredient in processed foods [6]. Dietary fiber is classified into soluble (SDF) and insoluble dietary fiber (IDF), where, IDF constitutes lignin, chitosan, cellulose, hemicellulose, and SDF fraction consists of mucilage, pectic substances, gums, and soluble hemicelluloses [7]. IDF has bulking action in gastrointestinal tract, whereas, SDF absorbs water forming a viscous, gelatinous matter which can be fermented by digestive bacteria. The extraction of dietary fiber from varied sources is an expensive and tedious process [8]. The type of extraction method and processing condition may affect the functional properties of dietary fiber due to microstructural as well as compositional changes [9, 10]. Several conventional methods including hot water extraction, ethanolic extraction, and alkaline extraction have been reported for the extraction of dietary fiber. However, it may lead to changes in structure of the dietary fiber which may result in loss of nutritional qualities as well as cause secondary pollution [11]. This increased the interest in scientific community for search of mild and ecofriendly extraction techniques to replace traditional chemical methods. Several non-conventional techniques including enzymatic, ultrasound, and ultrasound-assisted enzymatic extraction (UAEE) have been used individually or in combination with other technique to enhance the extraction yield [12, 13] and improve the functional properties of dietary fiber. The main force behind ultrasonic-assisted extraction is acoustic cavitation, which involves the creation, enlargement, vibration, and forceful collapse of gas bubbles within the solvent. The collapse of bubbles generates localized, vigorous agitation on a small scale which involves the solvent to penetrate the matrix more effectively affecting the integrity of cell walls. Consequently, this process enhances the extraction solvent’s ability to release intracellular content and facilitates improved mass transfer mechanisms [14].

Among these, the enzymatic method is an eco-friendly method reported to possess a higher extraction rate [15, 16]. Apart from this, ultrasonic technology has been used in the recent years due to its eco-friendly nature, high extraction yield, and reduction in treatment time [17]. A limited number of studies have been carried out on synergistic effect of enzymatic and ultrasound treatment on the extraction of dietary fiber from mango peels. Moreover, information on mass transfer mechanism of the extraction process using different kinetic models is also scarce. Kinetic studies using different models may help broaden the industrial applicability of ultrasound technique for extraction of dietary fiber from different bioresources [18, 19].

In view of the above, the present investigation endeavors optimization of ultrasound-assisted extraction (UAE) of soluble and insoluble fractions of dietary fiber using response surface methodology as well as to study the effect of sequential treatment of ultrasound and enzymes in enhancement of extraction yield. A suitable kinetic model of SDF and IDF extraction from mango peels was established, and the kinetic parameters were calculated based on an established model to provide theoretical reference for improvement of dietary fiber extraction. Along with this, investigation of three important thermodynamic parameters (∆G, ∆H, and ∆S) and structural elucidation of dietary fiber fractions (SDF and IDF) were also carried out.

2 Materials and methods

2.1 Materials

Mangifera indica peels were procured from the local fruit processor of Longowal, Punjab, India, and sun-dried for 3–4 days followed by grinding to fine powder (200 μm). It was then stored in an air tight container at 4 ± 1 °C for further analysis. Enzymes including heat stable α-amylase (300 IU/g), protease (2.4 IU/g), and amyloglucosidase (300 IU/g) were purchase from Sigma-Aldrich. All the chemicals of analytical grade were utilized in this study.

2.2 Ultrasound-assisted extraction of dietary fiber

Mango peel’s dietary fiber was extracted using ultrasound technique with some modifications [8]. An adequate amount of sample (1 g) was weighed, and phosphate buffer (0.08 M, 50 mL, pH 6.0) was added to it. Ultrasound probe (Sinaptech NexTgen Lab 500) was used to sonicate the mixture at 25 °C using 40% amplitude for 9 min.

The mixture was centrifuged (5000 g, 15 min) for separation of SDF and IDF fractions. The residual biomass was separated, rinsed using distilled water and then dried to obtain IDF. Ethanol (four-fold volume) was added to the supernatant, and solution was left to precipitate (2 h) soluble dietary fiber. The collected residue was washed using 100% ethanol and dried. Total dietary fiber was calculated as the sum of SDF and IDF.

2.3 Ultrasound-assisted enzymatic extraction of dietary fiber

Ultrasonication technique assisted with enzymes was used to extract dietary fiber from mango peels [8]. The experiment was carried out using definite amount of sample (1 g) in phosphate buffer (0.08 M, 50 mL, pH 6.0). Ultrasonic probe (Sinaptech NexTgen Lab 500) was used for sonication of the sample. The mixture was then treated with alpha-amylase (40 μL, pH 6.0) and placed in boiling water bath (60 °C, 15 min). Subsequent treatment included addition of protease (100 μL, pH 7.5) followed by addition of amyloglucosidase (200 μL, pH 4.5) assisted with incubation (60 °C, 30 min) after each treatment. After this, SDF and IDF fractions were obtained by using a similar procedure as followed after ultrasonication treatment in Sect. 2.2.

2.4 Experimental design

Single factor experiment was firstly carried out to determine constant parameter conditions for the extraction of dietary fiber using UAE. Liquid-to-solid ratio (30:1, 40:1, 50:1, 60:1), extraction temperature (25, 35, 45, 55 °C), ultrasound amplitude (20, 30, 40, 50, 60%), and time (3, 6, 9, 12 min) were varied each at a time by keeping other variables constant as performed in our previous studies [10]. Based on the preliminary experiments, process parameters range was selected for further experimentation. Optimization of soluble and insoluble dietary fiber was carried out) with Box–Behnken design (BBD) of response surface methodology (RSM). Four independent variables including time (A), temperature (B), liquid-to-solid ratio (C), and amplitude (D) were used to optimize the extraction process, and soluble and insoluble dietary fiber yields were kept as response for the independent variables which has been indicated in Table 1. The experiment was designed and statistical analysis was performed using Design Expert 13.0 software. The adequacy of the model was checked by the values of R2 and adjusted R2. Analysis of variance (ANOVA) was used for testing the significance of the model. F test was used to identify the significant terms (linear, quadratic, and interactive) affecting the responses. Quadratic polynomial equation was used to correlate the independent variable with the response variable:

$$Y= {\beta }_{0} + \Sigma\; {\beta }_{ixi} + \Sigma\; {\beta }_{iixi} + \Sigma \;{\beta }_{ijxixj}$$

where, β0, βixi, βiixi, and βijxixj indicate the intercept, linear, squared, and interaction terms, respectively. xi and xj represent the coded levels of independent variables and Y stands for the predicted response value.

Table 1 Box–Behnken design matrix for extraction of SDF and IDF from mango peels by using ultrasound assisted extraction
Full size table

2.5 Kinetic study models

Kinetic study of solid–liquid extraction is associated with mass transfer mechanisms, where chemical forces tie up the solid matrices and solutes which are then conveyed to solvent phase by dissolution [20]. Study of extraction kinetics using suitable models may help understand the involved chemical reactions and mass transfer mechanism. The study included applicability of suitable kinetic models (Elovich’s, hyperbolic, pseudo first-order and pseudo second-order models) for understanding mass transfer reaction (Table 2). These models were used to describe the kinetic behavior of dietary fiber (soluble and insoluble) extracted using UAE.

Table 2 Description of the models for the linear, non-linear, and slope-intercept forms
Full size table

2.5.1 Elovich’s model

This model depicts that increase in extraction yield has a linear relationship with accession in the extraction rate in a liquid–solid system [21]. Hence, below, Eq. (1) illustrates the Elovich’s linearized form, whereas Eq. (2) depicts the differential equation.

$$q={E}_{0}+ {E}_{1} ln t$$
(1)
$$dy/dt= \beta \times exp (-\;a\overline{y })$$
(2)

In this equation, β represents the initial extraction rate, such that when \(\overline{y }\) advances to (\(\overline{y }\) →0), (d \(\overline{y }\)/dt) → β. So, a plot of \(\overline{y }\) versus ln t gives a straight-line graph with and E0 and E1 as the intercept and gradient, respectively.

2.5.2 Hyperbolic model

The given Eq. (3) presents the non-linear form of this model.

$$\overline{y }=\frac{{C}_{1}t}{1+{C}_{2}t}$$
(3)

In this equation, \(\overline{y }\) represents the extraction yield, maximum yield constant (min−1) and the initial extraction rate (min−1) correspond to C2 and C1, respectively, and t is the extraction time (min). Therefore, Eq. (4) gives out the linear expression of Eq. (3).

$$\frac{1}{y}=\frac{1}{{C}_{1}} \times \frac{1}{t}+ \frac{{C}_{2}}{{C}_{1}}$$
(4)

As per Eq. (4), a plot of 1/y against 1/t accounts for C2/C1 and 1/C1 as gradient and intercept, respectively.

2.5.3 Lagergren pseudo first-order model

Lagergren model was used to due to its simplicity and good fit but might not represent the true nature of extraction kinetics. The Lagergren’s equation (linear form) of pseudo first-order model can be explained as:

$$\log\;\left(y_e-y_t\right)=\log\;y_e-\frac{k_1t}{2.303}$$
(5)

In which k1 (min−1) is the rate constant of the overall extraction process. Where ye and yt are the extraction yield at equilibrium and at time t, respectively. Therefore, extraction time t versus plot of log (ye − yt) produces a line with a slope = k1/2.303.

2.5.4 Pseudo second-order model

Pseudo second-order model is used to determine the dissolving rate of solute raw material into the solvent. Equation (5) represents the expression of this model.

$$d{C}_{s}/dt={ K ({C}_{s}- {C}_{t})}^{n}$$
(6)

where, concentration of the extract yield at time is represented by Cs; the fraction of extract yield after time t is given by Ct, extraction rate constant is K, and the order of reaction is n, which is two (2). When Eq. (7) is linearized, a non-linear form of the expression is obtained (Eq. 8).

$$C =\frac{C{s}^{2}Kt}{1+CsKt}$$
(7)
$$\frac{t}{Ct}=\frac{1}{KC{s}^{2}}+ \frac{1}{Cs}t$$
(8)

As a result, the extraction rate constant K can be determined graphically, when equation (8) is expressed in the form of y = mx + c. Thus, a plot of t/Cs against t produced 1/KCs2 and 1/Cs, as intercept and slope, respectively.

2.6 Thermodynamic parameters

The conditions of the experiment were considered and positioned with the thermodynamic parameters used in this study. Therefore, the thermodynamics second law (Eq. 9), describes Gibbs’ free energy (ΔG). Therefore, with the use of obtained K values, ΔG can be obtained, with R given as 8.314 Jmol−1K−1, i.e., the molar gas constant, T (K) is the absolute temperature while Eqs. (10, 11, 12 and 13) were used to determine other thermodynamic parameters such as ΔS and ΔH.

$$\Delta G = -RTln K$$
(9)
$${\text{ln}} \;K= \frac{\Delta G}{-RT}$$
(10)
$${\text{ln}} \;K = - \frac{\Delta H}{RT}+ \frac{\Delta S}{R}$$
(11)
$${\text{ln}} \;K= -\frac{\Delta G}{RT}= -\frac{\Delta H}{RT} + \frac{\Delta S}{R}$$
(12)
$$\frac{\Delta S}{R}=- \frac{\Delta G+ \Delta H}{RT}$$
(13)

2.7 Morphological and microstructural characterization

2.7.1 Scanning electron microscopy

The soluble and insoluble dietary fiber samples after gold plating were observed with scanning electron microscope (JSM, 7610 F plus, JEOL, Japan) using an accelerating voltage of 5 kV at 250 × , 1500 × , and 10,000 × magnification [22].

2.7.2 Fourier transform infrared spectroscopy

The soluble and insoluble dietary fiber samples were analyzed for the identification of functional groups using FTIR instrument (Perken Elmer Spectrum, RX-I, USA) with a resolution of 4 cm−1 in the range 400 to 4000 cm−1 [23].

3 Results and discussion

Optimization of process parameters for ultrasound-assisted extraction of SDF and IDF from mango peels was carried out using response surface methodology and effect of UAEE treatment on extraction yield of dietary fiber was studied. Along with this, extraction kinetics and thermodynamic parameters were evaluated, and structural elucidation of dietary fiber components (SDF and IDF) was carried out.

3.1 Optimization of UAE process

The evaluation of various extraction parameters (liquid-to-solid ratio 40:1–60:1), temperature (20–30 °C), ultrasound amplitude (30–50%), and treatment time (6–12 min) selected on basis of single factor experiment was carried out to study their effect on extraction yield of dietary fiber (soluble and insoluble fractions). The interactive effect of process variables was determined using BBD. The BBD and the responses expressed as soluble dietary fiber and insoluble dietary fiber (%, w/w) have been indicated in Table 1. All the process parameters showed significant effect (p < 0.05) on the extraction yield of soluble and insoluble dietary fiber. The model exhibits adequate fit with R2 value of 0.98 and 0.94 for soluble and insoluble dietary fiber, respectively. The statistical significance of the model has been demonstrated (F test) in Table 3. The significant (p < 0.0001) regression model of ANOVA with nonsignificant lack of fit (p > 0.05) further confirmed the validity of model. The extraction yield of dietary fiber could be explained by the following quadratic equation:

$$\begin{array}{l}\mathrm{SDF}=23.18-0.291667\times\mathrm A+1.51667\times\mathrm B+0.3\\\;\;\;\;\;\;\;\;\times\mathrm C-0.391667\times\mathrm D+0.575\times\mathrm{AB}+0.55\times\mathrm{AC}+1.25\\\;\;\;\;\;\;\;\;\times\mathrm{AD}+0.425\times\mathrm{BC}+1.95\times\mathrm{BD}-1.375\times\mathrm{CD}-3.23167\\\;\;\;\;\;\;\;\;\times\text{A}^2-2.79417\times\text{B}^2-2.69417\times\text{C}^2-2.23167\times\text{D}^2\end{array}$$
$$\begin{array}{l}\mathrm{IDF}=47.9-0.266667\times\mathrm A+0.416667\times\mathrm B-0.3\\\;\;\;\;\;\;\;\times\mathrm C-0.45\times\mathrm D-0.75\times\mathrm{AB}+1.72765\text{e}-16\times\mathrm{AC}+1.15\\\;\;\;\;\;\;\;\times\mathrm{AD}-0.475\times\mathrm{BC}-0.625\times\mathrm{BD}-0.525\times\mathrm{CD}-1.3625\\\;\;\;\;\;\;\;\times\text{A}^2-0.8125\times\text{B}^2-1.2625\times\text{C}^2-1.5875\times\text{D}^2\end{array}$$

where, SDF and IDF is the yield (%) of soluble and insoluble dietary fiber, respectively. A, B, C, and D are the coded variables and represent time, temperature, liquid-to-solid ratio, and amplitude, respectively.

Table 3 ANOVA table for ultrasound assisted extraction of dietary fiber (SDF and IDF) from mango peels
Full size table

3.2 Effect of process variable on soluble and insoluble dietary fiber yield

The 3-D response surface graphs (Figs. 1 and 2) illustrating the interactive effect of two process parameters on the responses by keeping other parameters at zero level are obtained and discussed as below.

Fig. 1
figure 1

Interactive effect of a temperature and time; b liquid-to-solid ratio and time; c amplitude and time; d temperature and liquid-to-solid ratio; e temperature and amplitude; f liquid-to-solid ratio and amplitude) on extraction yield of SDF obtained using ultrasound assisted extraction

Full size image
Fig. 2
figure 2

Interactive effect of a temperature and time; b liquid-to-solid ratio and time; c amplitude and time; d temperature and liquid to solid ratio; e temperature and amplitude; f liquid-to-solid ratio and amplitude) on extraction yield of IDF obtained using ultrasound assisted extraction

Full size image

3.2.1 Effect of liquid to solid ratio

As shown in Figs. 1 and 2, liquid-to-solid ratio had a significant effect (p < 0.05) on the extraction yield of soluble and insoluble dietary fiber. It was observed that as the liquid to solid ratio was increased from 40:1 to 50:1, extraction yield of soluble and insoluble dietary fiber was found to increase. However, further increase in liquid-to-solid ratio led to a gradual decrease in the extraction yield. Higher extraction yield with increase in liquid-to-solid ratio can be attributed to the larger volume of solvent, which offer a greater surface area for interaction with solutes. Consequently, this facilitates the dissolution of solutes, resulting in a higher yield. Moreover, powerful jet streams are created in the liquid media on collapse of ultrasound-induced cavitation bubbles which induces transport mechanism, and therefore, diffusion-limiting barrier around the solid matrix is lowered [24]. Higher liquid-to-solid ratio is also associated with lower viscosity and increased dissolution of soluble fibers in the solvent [25]. The insignificant change in the extraction yield of SDF and IDF after a certain point might be caused at higher volumes as the threshold limit of the solvent is attained beyond which further dissolution of the solutes cannot takes place [26]. Similar effect of increase in liquid-to-solid ratio was observed in previous studies on soya bean residues, where optimum liquid-to-solid ratio for the extraction of dietary fiber was found to be 50:1 [27].

3.2.2 Effect of temperature

In the study, it was observed that as the temperature increased from 20 to 25 °C, an increase in the extraction yield of soluble and insoluble dietary fiber fractions was observed, but after a certain point the yield declined gradually. Temperature is one of the prime factors affecting the extraction of bioactive compounds as mass transfer and diffusion of fibers is enhanced with increase in temperature. The temperature at which the extraction is to be performed is greatly influenced by the type of polysaccharide to be recovered as well as the matrix characteristics [28]. The sensitivity of the material to higher temperature may be considered a major reason for the selection of ambient temperature conditions of extraction. The results were in coherence with the previous studies on Canadian horseweed, where, 25 °C was selected as the ambient temperature for ultrasound-assisted extraction of dietary fiber [29]. It should be noted that beyond the optimal values, an increase in temperature can result in decline of dietary fiber which may occur as a result of degradation of already extracted polysaccharides which can break due to exposure to high temperature [30]. The results were in coherence with a recent study on oat cultivar 2000, where ultrasound-assisted extraction of dietary fiber indicated highest extraction yield at 26.9 °C [31].

3.2.3 Effect of ultrasound amplitude

The extraction yield of dietary fiber, both soluble and insoluble fractions increased significantly (p < 0.05) upon increase in ultrasonic amplitude from 30 to 40%. However, as the amplitude was further increased, a decline in the extraction yield was observed. A higher ultrasound amplitude accelerates the development and simultaneous collapse of cavitation bubbles which generates high speed jet and shock waves into the solvent. This may enhance solvent penetration into plant cells and release of polysaccharides by disruption of plant cell walls [32]. As the amplitude is increased beyond threshold level, a drop in extraction yield may be noticed as lesser energy get transmitted due to increase in concentration of bubble volume around probe. Moreover, collapse of bubbles in an asymmetric way may result in reduced mass transfer rate and a subsequent decrease in yield. Moreover, degradation of extracted polysaccharide at higher amplitude may also be responsible for reduction in dietary fiber yield [33]. In previous studies on Citrus limetta peels, 37% amplitude was found optimum for the extraction of pectin [34]. In another study on yellow dragon fruit peels, 40% amplitude was selected for ultrasound-assisted extraction of dietary fiber [35]. Similarly, 40% amplitude was also used in a previous study on ultrasound-assisted extraction of cellulose fibers from rice straw [36].

3.2.4 Effect of treatment time

The extraction yield of soluble and insoluble dietary fiber was found to increase by increasing sonication time from 6 to 9 min. However, with further increase in time, a gradual decrease in extraction yield was observed in both cases. As the sonication time is increased, wettability of plant material is also increased which may lead to enlargement of cell wall pores, thus accelerating the infusion of solvent in the plant matrix. This may boost release of polysaccharide into the solvent and causing a significant increase in dietary fiber yield [37]. However, a decline in polysaccharide yield after a certain point may be explained by the fact that an equilibrium is attained after sometime between solvent and solid matrix after which no dissolution of solute occurs [11]. Moreover, at higher treatment time, excessive ultrasound may cause degradation of the extracted polysaccharide [38]. In previous studies, treatment time of 10 min was taken as the optimum time for extraction of insoluble dietary fiber from soya bean residues [27]. Similarly, in another study on passion fruit peels, maximum polysaccharide yield was observed after 10 min of sonication time [39]. Sonication time of 10 min was also found optimum in previous study on ultrasound-assisted extraction of dietary fiber from culinary banana bract [40].

3.3 Optimum extraction conditions

The results of the optimization of various process parameters (liquid-to-solid ratio, temperature, ultrasound amplitude, and treatment time) revealed that the highest yield of SDF and IDF was found at 26 °C, 38.6% ultrasound amplitude using 49.8:1 liquid-to-solid ratio after treatment time of 8.6 min. At these conditions, the predicted value of SDF and IDF fractions of dietary fiber was found to be 23.2% and 48.07%, respectively, and it matched well with experimental value of 22.9% (SDF) and 47.9% (IDF).

3.4 Ultrasound-assisted enzymatic extraction of dietary fiber

The sequential treatment of ultrasound (under the optimized conditions) and enzymes to obtain the maximum extraction yield of dietary fiber from mango peels was carried out. The enzymatic treatment of peels after sonication treatment led to an increase in the extraction yield of dietary fiber from 71 to 73.5%. This increase in dietary fiber yield may be attributed to disintegration of cell wall by ultrasonication which resulted in more favorable conditions for enzymes to act. The non-fiber components (lipids, proteins, phenolic acids) also get reduced in the final extract due to degradative action of enzymes. The effect of enzymatic hydrolysis increases the porosity of substrate, and thus, solvent gets easily penetrated through the plant matrix which enhances the extraction yield [41]. The results were in coherence with the previous reported studies on extraction of dietary fiber from mango peels where the maximum yield was found to be 72.5% [6].

3.5 Impact of temperature and extraction time on the extraction rate of SDF and IDF

The optimized values obtained using response surface methodology was used to further study impact of varying temperature (20 °C, 25 °C, 30 °C, 35 °C, 40 °C, 45 °C) on soluble and insoluble dietary fiber fractions at different time intervals (3, 6, 9, 12, 15, 18 min) as indicated in Fig. 3. A positive effect of temperature (20–25 °C) on the extraction yield of dietary fiber was observed up to 9 min of treatment time. However, with further increase in temperature over extraction time, a significant drop in the extraction yield was observed. However, the drop in extraction yield of IDF was less pronounced than SDF at elevated temperature range over time. A forward reaction is favored over time (up to 9 min) as demonstrated at lesser temperature with more solutes diffusing into liquid phase [42]. The higher extraction yield of fiber components (soluble and insoluble) at lesser temperature and shorter time duration may be attributed to the high efficiency of ultrasound-assisted extraction process. The advantages of this technique over classical procedures support the importance and great potential of ultrasonication technique at industrial scale for extraction of functional fibers from agro-industrial byproducts [27].

Fig. 3
figure 3

Effect of extraction time and temperature on a SDF and b IDF yield obtained from mango peels using UAEE

Full size image

3.6 Kinetic analysis

The parameters of the model and the associated R2 value calculated in the work have been indicated in Table 4. Graphs of the various models were generated by using Eqs. (1), (2), (3), and (4) for hyperbolic, Elovich’s, pseudo first-order, and pseudo second-order models, respectively. For Elovich’s model, the values of E1 and E0 were found to display an inconsistent pattern in case of both dietary fiber fractions. This was reflected by the average correlation coefficient (R2) value of the model in case of SDF (0.63) and IDF (0.50) extraction yield over time. The hyperbolic model’s C1 and C2 value was found to show an irregular pattern of trend for SDF and IDF yield with increase in temperature. This implied that according to hyperbolic model, the extraction rate at the initial stage and at optimum level was not linearly correlated to the temperature rise. The pseudo first-order model also failed to provide a realistic estimate of kinetic parameter, ye. However, the average correlation coefficient (R2) in case of pseudo first-order model for SDF (0.77) and IDF (0.59) fraction was found to be higher than Elovich’s and hyperbolic model. The correlation coefficients (R2) in case of pseudo second-order model were found to be close to 1 for both SDF and IDF yield over time at different temperature ranges. The value of pseudo second-order rate constant (K) was found to increase linearly with increase in temperature which indicated the endothermic nature of the extraction process. Hence, pseudo second-order model was found to be more suitable than pseudo first-order, hyperbolic, and Elovich’s models for describing extraction kinetics of dietary fiber from mango peels as per the results obtained for the kinetically studied analyzed data. The regression results obtained from the experimental data also indicated that the UAE kinetics have a strong dependency on the extraction conditions and the applicability of different kinetic models. The kinetic pathway in this study was similar to the adsorption kinetics of arabinoxylan and pectin obtained from cactus pear peels and adsorption kinetics of pectin-rich fruit wastes reported in previous studies [43, 44]. Similarly, ultrasound-assisted extraction of antioxidants from pomegranate peels also followed second-order kinetic model [45].

Table 4 Model’s parameters for the extraction of dietary fiber (SDF and IDF) from mango peels by using ultrasound assisted extraction
Full size table

3.7 Thermodynamic parameters

According to Eq. (12), the plot of ln K vs. 1/T gives a straight line with ∆H/R and ∆S/R as slope and intercept, respectively (Fig. 4). The R2 value greater than 0.92 confirmed satisfactory values of ∆H and ∆S. The estimated thermodynamic parameters according to Van’t Hoff’s equation for SDF and IDF have been presented in Table 5. The negative ∆G values calculated for SDF and IDF was found negative and increased with increase in temperature signifying a spontaneous and feasible extraction process. The positive values of ∆H indicated endothermic nature of the extraction process, whereas, the positive ∆S value indicate higher degree of randomness and that the process is irreversible [46]. The degree of spontaneity increases with increase in temperature for the extraction process. These observations indicated extraction process may be enhanced by increase in temperature [43].

Fig. 4
figure 4

Plot of Ln K against 1/T for thermodynamic parameters determination for SDF and IDF fractions of dietary fiber extracted from mango peels by using ultrasound-assisted extraction

Full size image
Table 5 Thermodynamic parameters for extraction of dietary fiber (SDF and IDF) from mango peels by using ultrasound-assisted extraction
Full size table

3.8 Microstructure analysis

Scanning electron microscopy analysis was carried out to investigate the microstructure of SDF and IDF samples extracted using ultrasound-assisted enzymatic method which have been presented in Fig. 5. SDF sample was characterized with a wrinkled surface with a block shape and compact texture. The surface of SDF presented a dense structure which was covered with many massive material and spherical particles, most of which were protein wedges and starch granules [47]. IDF sample was found to possess a honeycomb or cavernous structure which may be due to combined effect of ultrasound and enzymes treatment. The IDF had more wave drapes with looser and enlarged surface area and large pores. Ultrasonication may cause the dietary fiber to be more readily hydrolyzed by enhancement of its accessible surface area and destruction of cellulose-hemicellulose-lignin matrix [48]. The combined effect of ultrasound and enzymes is responsible for loosening the structure of mango peels which increase the surface area and may affect several functional properties of dietary fiber such as adsorption capacities of water, oil, and glucose [49]. Similar microstructure of SDF and IDF was observed in previous studies on Rubus chingii fruits [50] and orange byproducts [51].

Fig. 5
figure 5

Surface morphology of a SDF (× 1500), (a’) SDF (× 250), b IDF (× 1500), (b’) IDF (× 250) extracted from mango peels by using UAEE

Full size image

3.9 FTIR analysis

The spectroscopic features of SDF and IDF samples obtained using ultrasound-assisted enzymatic extraction was analyzed using FTIR which has been shown in Fig. 6. The broad band area in the range of 3700–3000 cm−1 indicated the combination of hydrogen and hydroxyl groups in hemicellulose and cellulose. However, the intensity of the peak in this area was less in case of IDF sample probably due to destruction of hydrogen bonds of dietary fiber during ultrasonication treatment [52]. This may facilitate the formation of hydrogen bonds between water and dietary fiber leading to improvement in its water holding capacity. The peak observed in the region 2920–2930 cm−1 may be attributed to the C-H stretching of methylene group characteristic of pectin [53]. The characteristic peak near 1740 cm−1 indicated C-O stretching of hemicelluloses whereas, the peaks near 1620 cm−1 and 1430 cm−1 were attributed to the presence of aromatic hydrocarbons of lignin and CH2 stretching of cellulose, respectively [54]. All the peaks confirmed the presence of major components of dietary fiber in SDF and IDF fractions.

Fig. 6
figure 6

FTIR spectrum of a SDF and b IDF extracted from mango peels by using UAEE

Full size image

4 Conclusion

In the present investigation, ultrasound-assisted enzymatic extraction of dietary fiber from mango peels was carried out to achieve the maximum extraction yield (73.5%) The optimal conditions during ultrasound treatment including liquid to solid ratio (49.8:1), temperature (26 °C), ultrasound amplitude (38.6%), and treatment time (8.6 min) were used and combined with enzymatic treatment to further enhance extraction yield of SDF and IDF. Among all the studied kinetic models, pseudosecond-order model presented a good fit to the experimental assays with R2 value of 0.95 and 0.99 for SDF and IDF extraction, respectively. Thermodynamic parameters, including ∆G, ∆H, and ∆S indicated spontaneous, endothermic, and irreversible nature of the process. SEM analysis indicated that UAEE technique could positively influence the functional properties of dietary fiber fractions, whereas, FTIR spectrographs confirmed the expected groups present in dietary fiber. The study indicated that UAEE technique can be a potential and effective green technology for the extraction of dietary fiber from mango peels. This may not only act as a sustainable valorization strategy for food industries but also generate economy by development of functional food products from the extracted dietary fiber. However, further investigation can be carried out on the optimization and kinetic studies not only based on the yield but also the functional properties of extracted dietary fiber and to establish correlation of kinetic parameters with ultrasound power. Moreover, complete understanding of UAE technology is essential for its application in commercialization of food products.