Papaya by-products for providing stability and antioxidant activity to oil in water emulsions

Abstract

The consumption of food with health benefits is growing today worldwide. This study was designed in order to incorporate papaya dietary fibre concentrates (DFCs) from peel and pulp dehydrated with the use of microwave (MW), or convection with hot air (CV) in oil-in- water emulsions. Results of studies indicated that Pulp DFC produced more stability to creaming (18 weeks) than Peel DFC (6 weeks). It was found that peel DFCs exerted up to 30% reduction in lipid peroxidation in comparison to the reference system during storage. Rheological analysis showed a similar behaviour when emulsions were mixed with pulp DFCs either dehydrated by MW or CV, while the dressing with peel DFCs had a much lower consistency than the former. The analysis of the emulsions micro-structure showed a polydisperse system of oil droplets and fiber structures trapping oil. Finally, emulsions with pulp DFCs showed a better consumer´s acceptance. These results also suggested that the use of DFCs may have high industrial potential in contributing to dietary fibre enrichment through technological intervention of emulsion formulation by papaya pulp and peel, increasing antioxidant property, consistency and stability during storage.

Electronic supplementary material

The online version of this article (10.1007/s13197-020-04679-w) contains supplementary material, which is available to authorized users.

Introduction

Many of the foods which are currently sold have an emulsion structure. That is the case of mayonnaise, margarine, salad dressings, snack sauces, etc. These products have in common the use of water and oil as their main ingredients. The most common products are the oil-in-water (O/W) emulsions, which consist of oil droplets dispersed and stabilised in an aqueous phase (Raikos and Ranawana 2017). Different types of additives are used in order to stabilise the emulsions, for example, emulsifiers, texture modifiers, and antioxidants. An emulsifier is adsorbed on the surface of the droplets as a means to prevent the aggregation, it reduces the interfacial tension and facilitates the disruption of emulsion droplets during homogenisation step, allowing to produce emulsions with smaller droplet sizes (McClements 2007). The emulsifiers commonly used are proteins, polysaccharides, phospholipids, small molecule surfactants and solid particles. Texture modifiers are substances that increase the viscosity of the continuous phase, slowing down the movement of droplets and avoiding their aggregation and coalescence; while antioxidants preclude lipids oxidation (McClements and Decker 2018).

Nowadays, the interest in natural food products, processed without artificial additives, has become widespread. Therefore, the food industry research has focused on natural compounds with technological functionality, health benefits and low environmental impact. In the food emulsions field, research is aiming to find new texture modifiers (Qiu et al. 2015a), compounds with antioxidant activity (Qiu et al. 2015b), emulsifying compounds and/or multifunctional ingredients (Chung et al. 2014). Recent studies reported on fruits by-products, demonstrated that their dietary fibre has great potential to be used as multifunctional ingredients in food formulations (Mangal et al. 2015).

Papaya (Carica papaya L.) is a fruit widely cultivated in the world, especially in tropical and subtropical regions. By-products with high potential to be exploited coming from peel or pulp, are generated during papaya industrialisation. Nieto-Calvache et al. (2016, 2018) studied the production of dietary fibre concentrates from papaya by-products with an important content of phenolic and carotenoids compounds as well as ascorbic acid, which remained associated with the dietary fibre and conferred antioxidant properties to these concentrates. Different authors reported the antioxidant benefits of dietary fibre isolated from plant tissues (Idrovo Encalada et al. 2019; Basanta et al. 2018; Saura-Calixto 1998), but to our knowledge studies concerning the addition of these dietary fibre concentrates in food formulations are scarce. Therefore, the objective of this work was to evaluate the use of dietary fibre concentrates (DFCs) from papaya pulp or peel, obtained by microwave dehydration (MW) or hot air convection (CV), on the stability of O/W emulsions and to study the rheological, microstructural and sensorial hedonic characteristics of selected formulations.

Material and methods

Materials

The ingredients used in the preparation of emulsions, were food grade products obtained from different suppliers: xanthan gum (Cargill, Buenos Aires, Argentina), sunflower oil (Molinos Rio de la Plata, Buenos Aires, Argentina), tween 80 (Alkest, Manuchar, Mexico), potassium sorbate (Parafarm, Buenos Aires, Argentina), sodium chloride (Dos Anclas, Buenos Aires, Argentina) and sucrose (Ledesma, Buenos Aires, Argentina). Citric acid, potassium iodide and sodium thiosulphate were analytical grade (BioPack, Buenos Aires, Argentina). Milli-Q™ deionized water was used in emulsion formulation and in the preparation of the solutions for the different analysis.

Dietary fibre concentrates

Each dietary fibre concentrate (DFC) was produced according to the methodology previously reported (Nieto-Calvache et al. 2018). Briefly, fresh tissue pulp or peels from papaya were treated with ethanol (96 mL/100 mL) for 20 min at 20 °C, using an ethanol/sample ratio of 2.9 mL/g. Subsequently, a drying step was carried out with microwave assistance (MW) or hot air convection (CV). MW drying was performed with an Ethos Plus microwave equipment (Milestone, Italy) working at a maximum power of 450 W at 40 °C. For CV drying, a forced convection equipment (FAC, Argentina, model SRBCO 4040) was used, with an air velocity of 2.0 m/s at a temperature of 40 °C. In both processes, drying was conducted until constant weight was achieved, measuring the water activity (a_w) in order to corroborate that it reached values below 0.6 to assure product stability (Muggeridge and Clay 2001). According to DFC source, pulp or peel, and the drying process applied, MW or CV, different drying times were needed. For DFCs dehydrated by MW: pulp DFC-MW was dried for 180 min achieving a moisture content of 8.7 ± 0.1% (wet basis, wb) and a_w of 0.47 ± 0.02 and peel DFC-MW was dried for 90 min, achieving 7.1 ± 0.1% (wb) and a_w of 0.41 ± 0.02. For DFCs dehydrated by CV, pulp DFC-CV was dried for 440 min achieving 6.6 ± 0.3% (wb) of moisture and a_w of 0.40 ± 0.01 while peel DFC-CV was dried for 410 min achieving moisture content of 6.8 ± 0.3% (wb) and a_w of 0.36 ± 0.02 (Nieto-Calvache et al. 2016, 2018).

Food emulsions production

A first screening was performed in order to assay the different DFCs in O/W emulsion and to select the best formulations. Systems containing increasing amount of DFCs and decreasing amount of xanthan gum (XG), as shown in Table 1, were analysed. Additionally, a reference emulsion was prepared only with XG (0.5%), taking into account that it had been determined in previous assays that this concentration allows the stability of the emulsion during storage (data not shown) and it is generally used in commercial preparations (Dickinson 2003, CAA 2019).

Table 1.

Peroxide value in dressings supplemented with DFCs and the reference emulsion at the end of storage

DFC (g)1	XG (g)1	Pulp DFC-MW	Pulp DFC-CV	Peel DFC-MW	Peel DFC-CV	Reference
0	0.5	A,B30 ± 1a	D30 ± 1a	D30 ± 1a	A30 ± 1a	A30 ± 1a
2	0.2	A32 ± 2b	A,D28 ± 1a,d	A27 ± 1a,c	A32 ± 1b	A30 ± 1b,c,d
3	0.1	A,C30.5 ± 0.8c,d	A,C25 ± 2b	B,C23 ± 1a,b	B,D22.2 ± 0.7a	A30 ± 1d
5	0.0	B,C27.6 ± 0.5c,d	B,C24.3 ± 0.7b	C21.0 ± 0.8a	C,D21 ± 1a	A30 ± 1d

Peroxide value measured as milliequivalents of oxygen per kilogram of sample (meq/kg)

DFC-MW, dietary fibre concentrates dehydrated by microwaves; DFC-CV, dietary fiber concentrates dehydrated by hot air convection. Different lowercase letters mean significant differences (p < 0.05).in the row Different capital letters mean significant differences (p < 0.05) in the column

¹Grams of DFCs and/or XG added into 41.88 g of water in the aqueous phase of the emulsion

The aqueous phase of the emulsions was prepared with 41.88 g of water, where XG and/or DFCs were dispersed to be hydrated for 12 h with agitation at 18 °C in accordance with the levels shown in Table 1. Thereafter, 11.63 g of citric acid solution (5 g/100 mL), 0.05 g of potassium sorbate, 0.15 g of sodium chloride and 3.4 g of sucrose were added and the systems were homogenised. On the other hand, oily phase was prepared with 38.53 g of sunflower oil and 0.56 g of tween 80. The emulsions were prepared mixing the oily phase into the aqueous phase with a Basic Ultra-Turrax T25 Homogeniser, equipped with a stirring tip S25N-25F (IKA Works Inc., Wilmington, NC). A homogenisation initial step at a rate of 13,500 rpm for one minute, followed by emulsification step at 24,000 rpm for three minutes, were used (Zalazar et al. 2016b). All emulsions were distributed in graduated tubes and stored at 8 °C for 18 weeks.

The three formulations with the best stability and antioxidant activity were chosen to be analysed in terms of rheology, microstructure and sensory acceptability and new systems without DFCs, were formulated for comparison purposes. For selected emulsion containing pulp DFCs, one system was formulated without XG, and for selected emulsion containing peel DFC, another system was prepared with 0.1 g of XG.

Methods

Determination of peroxide value

Peroxide value (PV) was determined in the oil phase of emulsions at the end of storage. For this determination, emulsions were frozen for one day and defrosted at room temperature in order to break the emulsion. Subsequently, the emulsions were centrifuged at 9000 rpm for 20 min to separate the oil phase (McClements 2007).

PV was measured by the method AOCS Cd 8–53 (AOCS 2003). Briefly, 2 g of sample were homogenised with 30 mL of acetic acid:chloroform in a volumetric proportion of 3:2 (v/v), then 0.5 mL of saturated potassium iodide solution were added and the mixture was occasionally shaken for 1 min. Finally, thirty millilitres of water were added. The mixture was titrated with 0.01 N standard solution of sodium thiosulphate. The results were expressed as milli-equivalent of peroxide/kg oil phase.

Creaming stability and droplet size analysis

The formation of a serum layer was considered visual indication of the occurrence of creaming in the food emulsions. For this assay, samples (10 mL) of each emulsion were stored at 8 °C in graduated tubes of 15 mL of volume. The volume of serum separated at the bottom of the tubes was measured after 18 weeks, and the percentage ratio of the volume of serum separated to the initial volume of the emulsion was calculated.

Droplet size of each emulsion was measured during the storage by static light scattering with a Mastersizer 2000 Hydro 2000MU as dispersion unit (Malvern Instruments Ltd, UK.). The pump speed was set at 1800 rpm. The average drop size was characterised by the diameter d_(3,2) defined by:

$$d_{(3,2)}=\sum _{}{}_i{n}_i{d}_i^3/\sum {}_i{n}_i{d}_i^2$$

where n_i is the number of droplets of diameter d_i.

The drop size was measured after the dressing production, considering the “time 0” at 24 h. Measurement was repeated at one week intervals for 18 weeks.

Rheological characterisation

Rheological characterisation of the emulsions was performed by means of oscillatory assays and flow assays, using a rheometer (Paar Physica MCR 300, Anton Paar GMBH, Germany) as stated by Fissore et al. (2012). For oscillatory assays parallel plate (PP 30) geometry (30 mm-diameter) was used and sample size of ≈ 1 mL were applied. While for recording the flow curves, cone and plate (CP 75–2) geometry with sample size of ≈ 5 mL were used. Both rheological methods were briefly summarised below. Each emulsion was measured at least in duplicate.

Oscillatory assays

After performing the amplitude sweep at a fixed frequency of 1 Hz and 25 °C to determine the linear viscoelasticity range (LVR) for each emulsion, a frequency sweep was carried out at constant stress, obtaining the mechanical dynamic spectra in terms of storage (G’) and loss (G’’) moduli variations with angular frequency (w; rad/s).

Flow assays

Flow assays were developed at a constant temperature of 25 °C in the 0.01—100 s⁻¹ shear rate ($\dot{\gamma }$) range. Experimental data were adjusted to the Power law equation (Ostwald model).

$$\tau =k{·\dot{\gamma }}^n$$

where τ represents the shear stress, k is the consistency index, and n is the flow index. A Newtonian behaviour is inferred when n = 1, whereas a pseudoplastic flow is determined for n < 1. In addition, the apparent viscosity values for a shear rate of 50 s⁻¹ (η₅₀) were used to compare the different formulations responses.

Microscopic characterisation

The structure of the emulsions was analysed within 24 h of storage, in line with the work of Zalazar et al. (2016a). Briefly, a portion of each emulsion was mixed with 10 μL of 0.1 g/100 mL ethanolic solution of Nile Red and then, the sample was analysed microscopically. This colorant is used to stain the oil phase (Auty 2013). The microstructure of the dressings was observed using a confocal scanning laser microscope (FV 300, Olympus, UK), fitted with a He–Ne laser (543 nm). A 60X APO PLAN lens and a 20X magnification were used. The samples were excited at 488 nm. The emission was recorded at 10 nm above the excitation wavelength. The images were taken and processed using Viewer FV10-ASW version 4.1 software, UK.

Sensory evaluation

A sensory analysis test was conducted to evaluate the acceptance and/or preference of the dressings. The dressings were prepared by adding to the basic formulation 0.15% of garlic powder and 0.05% of onion powder as flavour condiments.

The sensory test was achieved with the participation of seventy-six volunteers untrained panellists, who were frequent consumers of dressings, following the recommendations of Lawless (2013). The age of the panellists was between 21 and 55 years old. They were instructed to rinse their mouths with water and eat a cracker between samples to avoid carryover effects (Genevois et al., 2018). A 9-point structured hedonic scale ranging from 9 for a qualification “ I like extremely” to 1 for “I dislike extremely”, was used to evaluate the overall acceptability, texture, flavour and colour (Stone et al. 2012).

Statistical analysis

Statistical analysis of the results was performed through ANOVA for a level of significance (α) of 0.05 followed by Tukey´s post-test to identify significant differences among samples. Statistical and regression analyses were accomplished using the GraphPad Prism version 5.00 for Windows (GraphPad Software, San Diego California USA). Pearson product moment correlations, between each pair of variables were carried out using the Statgraphics Centurion XV program (V 2.15.06, 2007, Warrenton, VA, USA).

Results and discussion

Antioxidant activity evaluated by peroxide value

Lipid oxidation is an important food deteriorative reaction, altering shelf life, quality attributes and nutritional value. In addition, phenolic compounds have been determined as primary antioxidants, which complex free radicals either delaying or inhibiting the initiation stages or interrupting the propagation stages of lipid oxidation (Shahidi and Ambigaipalan 2015).

The PV observed at the end of storage for emulsions are summarised in Table 1. In general, there was a decrease in the PV when the DFCs were added and the greatest reduction, up to 30%, in relation to the reference system without DFCs, was observed when 5 g peel DFCs were incorporated. Alternatively, it was found for emulsions that contained pulp DFCs, lower PV when the DFC was dehydrated by CV, whereas for those with peel DFCs, the PV did not show significant differences between both drying processes neither between 3 and 5 g DFC addition. Nieto-Calvache et al. (2018) reported that the antioxidant activity of papaya DFCs might be mainly attributed to carotenoids and phenolic compounds, which were affected differently conforming to the drying method applied showing different concentrations depending on tissue type, pulp or peel. In addition they also reported that DFCs from pulp was ≈ 60% more water soluble than that from peel. This fact allows to infer that part of the total phenolic compounds were bound to insoluble peel DFCs affecting their effectiveness in the emulsion and therefore, differences between emulsions with peel DFCs dried by CV and MW could not be detected in this case (Nieto-Calvache et al. 2018; Spaggiari et al. 2020). Nevertheless, it was possible to verify that the addition of peel DFCs, in general, allowed a better conservation of the properties of the oil, exerting a greater protection against the oxidation of the lipids in the emulsion.

Moreover, a Pearson correlation analysis between the PV of the samples containing 5 g of DFCs, herein assayed, with antioxidant properties previously reported by Nieto-Calvache et al. (2018), showed significant and negative correlation coefficients (CC) being //CC = − 0.7765, P = 0.003 with DPPH; CC = − 0.7493, P = 0.005 with FRAP; CC = − 0.7516, P = 0.0048 with phenolic compounds; CC = − 0.7535, P = 0.0047 with ascorbic acid and CC = − 0.5822, P = 0.047 with carotenoids compounds. The coefficients obtained, were far from unit which represents that the correlations are not completely linear; nevertheless, P value < 0.05 shows that they are significant. The negative sign in the coefficients explains that lower PV can be obtained in the emulsions when DFCs with higher antioxidant properties and compounds were added.

Creaming stability

Different mechanisms can generate food emulsions instability. Among them, gravitational separation (creaming/sedimentation), flocculation, coalescence and Ostwald maturation can be found. Gravitational separation is the process in which the droplets move, for instance upward (creaming), because they have a lower density than the surrounding liquid. Flocculation is the process in which two or more drops adhere to form an aggregate that preserves the integrity of each drop. In coalescence, two or more droplets join together to form a single, larger droplet; while in Ostwald maturation, the largest droplets grow at the expense of the smallest droplets (McClements 2007). To evaluate the occurrence of this instability, droplet sizes were registered and the average diameter d_(3,2) for the tested systems, are shown in Fig. 1. In general, all dressings containing DFCs as well as the reference system (0.5 g XG without DFCs) presented diameters ranging from 1 to 2.5 µm, along 18 weeks storage, values that are in the range of 100 nm to 100 µm, reported by McClements and Decker (2018) for a conventional emulsion. Nevertheless, a remarkable increase in droplet size was observed in the week 7 (Fig. 1c), for the emulsion with 5 g of peel DFC-CV, probably due to instabilisation of drops, which consequently, caused the breaking of the emulsion. On the other hand, the emulsion with 5 g pulp DFC-MW—0 g XG (Fig. 1c), presented the highest drop diameter, however, it remained constant over time. In the study of the hydration properties (Nieto-Calvache et al. 2018), it had been demonstrated that the pulp DFCs-MW had almost 50% more swelling capacity than the pulp DFCs-CV and a much greater value than that from peel. It must be pointed out that the presence of dietary fibre in emulsions interferes with the determination of drop diameter (Chung et al. 2014) since the technique is designed to measure the size of all particles in suspension. Hence, it can be inferred that, in this case, the higher value in the droplet diameter might have been influenced by the greater size of the hydrated and swelled fibre.

Fig. 1.

Droplet size d_(3,2) in µm of dressings along storage. a: 2 g of DFC—0.2 g of XG, b: 3 g of DFC—0.1 g XG, c with 5 g of DFC—0 g of XG

Visual observation was one of the techniques used for monitoring the stability of the emulsions. As expected, reference system (0 g DFCs 0.5 g XG) did not present visual instabilisation along 18 weeks storage. It must be highlighted that none of the emulsions supplemented with pulp DFCs presented visual, instabilisation even though XG was absent in some formulations. Contrarily, most emulsions containing DFCs from peel showed creaming. For those emulsions, the percentage (in volume) of separated serum was registered at the end of 18 weeks storage. As it can be observed in Fig. 2, this percentage increased with XG reduction and peel DFC addition. The highest values were registered in systems containing 5 g of peel DFC CV—0 g of XG, which is in accordance with the detected change in the average diameter, d_(3,2) (Fig. 1c). From Fig. 2 it can be concluded, that among the emulsions that presented visual creaming, those added with peel DFC-MW, produced a better protective effect against creaming for all fibre addition levels. The highest pectin content (11 vs 7.5 g of uronic acids/100 g of DFC) and the greatest water holding capacity (27 vs 23.3 g/100 g) of the peel DFCs dehydrated by microwaves (Nieto-Calvache et al. 2018), could help to explain the better stabilising emulsion capacity of the peel DFCs-MW, since pectin has been found to have both emulsifying and stabilising properties for food emulsions (Ngouémazong et al. 2015). It has been proposed that pectin in emulsions, acts as a stabiliser through two mechanisms, on the one hand, due to its steric and/or electrostatic properties attributed to the side chains of rhamnogalacturonan-I, which help to keep the drops of oil separated and, on the other hand, due to its ability to increase the viscosity of the continuous phase by forming gel networks that prevent the mobility of the drops of oil and the creaming. Regarding DFCs from pulp, the capacity for inhibiting visual creaming in the emulsions might also be related to the high hydration properties of these fractions, which could be due to their hydrophilic compounds, such as pectin (Nieto-Calvache et al. 2018). Its thickening effect is similar to the one exerted by XG which is, in general the polysaccharide used in the production of commercial dressings (Dickinson, 2003). The results herein reported, contribute to the literature concerning the use of different types of dietary fibre for stabilising food emulsions, such as meat emulsions (Ağar et al. 2016), ice cream (Rodríguez et al. 2006) and salad dressings (Tseng and Zhao 2013). This technological functionality also represents a great potential for papaya processors as well as the emulsion industry since it might allow profiting papaya by-products as alternative stabilising compounds. In addition, these products make an important nutritional contribution to the formulation, through their dietary fibre content.

Fig. 2.

Percentage (in volume) of serum separated after 18 weeks of storage at 8 °C, of dressings containing peel DFCs. 2 g DFC—0.2 g XG; 3 g DFC—0.1 g XG; 5 g DFC—0 g XG

Emulsion selection

Considering that the emulsions containing 5 g pulp DFCs were stable from the creaming and the lipid oxidation point of view, and that the emulsion with 3 g peel DFC-MW presented better antioxidant activity and a low percentage of separated serum (less than 1%, Fig. 2), it was decided to select these three emulsions for rheological and microscopic characterisation as well as sensory assay. For comparing purposes with the systems containing 5 g pulp DFCs and 0 g XG, one system without DFC (0 g DFC and 0 g XG) was performed. Whereas for comparing with the system containing 3 g peel DFC and 0.1 g XG, one system containing 0 g DFC and 0.1 g XG was prepared. Thereby, possible differences observed in the emulsion rheology, microscopic and sensory analysis can be attributed to DFC addition.

Rheological characterisation: flow assay

Results regarding the study of the flow behaviour of emulsions are summarised in Table 2 and the fitted model plots as well as experimental data are also provided in supplementary data (Fig. S1). It was found that the data fit the power law model appropriately, obtaining R² values ≥ 0.9832. All emulsions supplemented with either DFCs or XG showed pseudoplastic behaviour (n < 1, Table 2), while the systems without DFC and without XG, showed Newtonian flow (n = 1, Table 2). The addition of pulp DFCs produced emulsions with the lowest flow indexes (n) which means a less dependence between the stress (τ) and the shear rate ($\dot{\gamma }$). The addition of pulp DFC-MW rendered emulsions with a higher (P < 0.05) consistency index (k) than those with pulp DFC-CV. Since the values “n” from the other emulsions were significantly different (P < 0.05), the corresponding consistency indexes are not in the same units and therefore it is not correct to compare them. Then, the apparent viscosity at shear rate of 50 s⁻¹ (η₅₀), was chosen for comparison purposes. The apparent viscosity, η₅₀, of the emulsions with pulp DFCs were higher (P < 0.05) than the values found for those with peel DFC-MW and for that without DFCs. Probably, the good hydration properties of the pulp DFC-MW (Nieto-Calvache et al. 2016) determined its high consistency and optimised the functionality of this fibre over the aqueous phase of the emulsion, contributing to dressing stability. Chung et al. (2014) studied the effect of dietary fibre and proteins on the stability and microstructure of food emulsions reduced in calories and obtained similar results in relation to the stability and effect of fibre on the viscosity of the aqueous phase of the emulsion.

Table 2.

Effect of the addition of papaya DFCs and XG on the flow behaviour of dressings modelled through the power law

Dressings	η50 (Pa.s)	k (Pa.sn)	n	R2
5 g pulp DFC-MW—0 g XG	3.885 ± 0.007A	74.2 ± 0.9A	0.245 ± 0.003A	0.9892
5 g pulp DFC-CV—0 g XG	3.14 ± 0.02B	57 ± 4B	0.26 ± 0.02A	0.9832
3 g peel DFC-MW -0.1 g XG	0.560 ± 0.008C	5.27 ± 0.05C	0.4277 ± 0.0007B	0.9980
0 g DFC—0.1 g XG	0.0472 ± 0.0003D	0.199 ± 0.003C	0.6304 ± 0.003C	0.9999
0 g DFC—0 g XG	0.00499 ± 0.00007E	0.0031 ± 0.0003C	1.085 ± 0.008D	0.9890

Different capital letters in a column mean significant differences (P < 0.05)

Rheological characterisation: oscillatory studies

The amplitude sweep test carried out between 0.1 and 1000 Pa at a fixed frequency of 1 Hz, allowed to establish the LVR for each emulsion and to subsequently perform the frequency sweep (Franco et al. 1995). Figure 3 shows the dynamic spectra (G’, G’’) of the emulsions as a function of angular frequency, comparing the formulations with 5 g of pulp DFCs—0 g GX, with the emulsion with 0 g DFCs—0 g GX (Fig. 3a), and the dressing with 3 g peel DFC—0.1 g GX, with the emulsion with 0 g DFCs—0.1 g GX (Fig. 3b). As can be observed, in both cases DFCs addition increased G’ and G’’.

Fig. 3.

Mechanical spectra of the dressings. a: 5 g pulp DFC—0 g XG: x G’ DFC-MW, □ G’ DFC-CV, ○ G’’ DFC-MW, ∆ G’’ DFC-CV. 0 g DFC-0 g XG: “black square” G’, “diamond” G’’. b: 3 g peel DFC-MW—0.1 g XG: “filled triangle” G’, “plus” G’’. 0 g DFC—0.1 g XG: “cross mark” G’, “diamond” G’’

The emulsions with added pulp DFCs (Fig. 3a), produced a similar spectra, independently of the drying process applied (CV or MW), G’ and G’’, in both cases, showed a great difference in their magnitudes. The storage modulus (G’) for these two dressings produced values between 2000 and 2500 Pa, while the loss modulus (G’’) oscillated between 400 and 700 Pa. This type of spectra is characteristic of foods such as commercial mayonnaise (Gallegos et al. 1992) and salad dressings (Franco et al. 1995) reported in the literature, where the module G’ is always presented above G’’. Besides, the systems without DFCs (0 g DFCs—0 g XG), presented storage modulus (G’) that ranged between 0.01 and 0.1 Pa and a loss modulus, G’’ ranging between 0.01 and 1 Pa. As expected, polysaccharides from pulp DFCs conferred elastic behaviour when hydrated into the aqueous phase of the emulsion. As for the emulsion with peel DFC (Fig. 3b), the results showed a similar tendency to those with pulp, with G’ always above G’’, being G’ between 2 and 3 order of magnitude higher when comparing to systems without DFCs (0 g DFC—0.1 g XG); since G’ ranged between 0.1 and 4 Pa, while G’’ value was between 0.02 and 2 Pa. Although the values for this last system were higher than those for the emulsion without XG, certainly because of the XG presence in the aqueous phase, it must be highlighted that the addition of peel DFC conferred the system a more elastic behaviour (Fig. 3b) due to the presence of fibre polysaccharides in that concentration. The highest consistency and elastic component of emulsion observed through flow and dynamic studies when fibre is added, delayed the occurrence of different mechanisms that generate food emulsions instability, keeping the size of dispersed droplet constant (Fig. 1).

Microscopic characterisation

Through the confocal microscopy analysis, it was possible to verify the state of the oily phase in the emulsions after 24 h of storage (Fig. 4). It is important to remark that the emulsion formulated with pulp DFC-MW presented the lowest proportion of coalesced drops (Fig. 4a) trend that could be associated with the highest consistency and viscosity of this system (Table 2). In this dressing, oily spots of irregular shapes were observed probably due to the trapping of the oil in the fibre. In Fig. 4a, c, the drops of oil, showed a spherical shape. The emulsion without DFC and with 0.1 g of XG (Fig. 4d), showed a low quantity of coalesced drops. The emulsion that had neither XG nor DFC (Fig. 4e), presented the largest number of coalesced drops and other droplets in the process of aggregation, being this behaviour expected since this emulsion did not have a stabilising hydrocolloid. It can be concluded that when DFC was added (Fig. 4a–c) some coalescence could be observed at microscopic scale, although the creaming was not visually detected as previously discussed.

Fig. 4.

Confocal micrographs (10X magnification) of dressings after 24 h of storage. a: 5 g pulp DFC-MW—0 g XG, b: 5 g pulp DFC-CV—0 g GX, c: 3 g peel DFC-MW—0 g XG, d: 0 g DFC—0.1 g XG, e: 0 g DFC—0 g XG

Sensory analysis

Finally, the acceptance of the three dressings with DFCs was analysed through sensory analysis, the information is provided as supplementary data (Fig. S2). Results show that both dressings formulated with pulp DFCs had a good acceptance, obtaining grades between 7 and 8 in a nine points scale, for the evaluated attributes: global acceptance, texture, flavour and colour, without significant differences (P < 0.05) in the results. Therefore, drying mechanism on pulp DFCs did not produce differences in the scores of sensory attributes of the emulsions where DFCs were incorporated. In addition, many panellists stated that they felt a slight sensation of lumps in the dressings added with pulp DFC-CV, while in those added with pulp DFC-MW; none of the panellists expressed this perception. Possibly, the better hydration properties and the higher pectin content of pulp DFC-MW, and consequently the higher consistency of the corresponding system (Table 2), could be, in part, the cause of the panellists perception of a smoother dressing without lumps, even when particles with higher diameter (d_3,2) were detected in this system (Figs. 1 and 4a).

The dressings with peel DFC had a lower acceptance (P < 0.05) in terms of the analysed attributes, rendering 4.49, 4.97, 4.16 and 5.82 in a nine point scale, for global acceptance, texture, flavour and colour respectively. It was also detected a common perception of a lower consistency than the other two dressings. Many of the panellists suggested that this last emulsion could be used as a salad dressing, while the formulations with pulp DFC had characteristics of the dressings or sauces which are used with snacks. These results are in accordance with the greater consistency previously discussed (Table 2, Fig. 3).

Conclusion

Dietary fibre concentrates from papaya pulp and peel, obtained by microwave dehydration or hot air convection drying, were used as ingredients in the formulation of dressing emulsions. In terms of the stabilising effect of DFCs, it could be verified that the addition of pulp DFCs produced stability against creaming for, at least, 18 weeks of storage in the absence of xanthan gum. Besides, the dressings with peel DFCs, were the ones that presented the best stability from the point of view of the oxidation which was evaluated in relation to the formation of peroxides at the end of the storage period. In addition, this antioxidant effect for a 3 or 5 g DFC addition was observed for both drying mechanisms used.

Through the rheological and sensorial study, it was verified that the dressings with pulp DFCs dried by different mechanisms (CV or MW), presented similar sensorial characteristics and dynamic spectra, and had a similar acceptance. By contrast, for the dressing that contained peel DFC in the formulation, it presented lower acceptance values in the sensorial analysis, but was not rejected by the consumers.

Finally, it can be concluded that DFCs produced from the papaya pulp and peel, are materials which show great potential to be used as natural ingredients in food formulation, due to their ability to trap water and modify the viscosity of aqueous systems, providing also antioxidant activity. Moreover, they are a source of functional and nutritional compounds due to its dietary fibre content.

Electronic supplementary material

Below is the link to the electronic supplementary material.