Physical-chemical characterization of yellow mombin ( Spondias mombin L . ) foam-mat drying at different temperatures

Caracterização físico-química de cajá (Spondias mombin L.) seco por camada de espuma em diferentes temperaturas R E S U M O Objetivou-se realizar a secagem da polpa dos frutos de cajá em camada de espuma, verificando a cinética e modelagem matemática do processo, e caracterizar o produto obtido quanto às características físicas e químicas, sendo estas comparadas à polpa in natura. A secagem foi realizada em camada de espuma, com auxílio do agente espumante emustab, nas temperaturas de 50, 60, 70 e 80 °C. Os dados da secagem foram analisados e ajustados por quatro modelos matemáticos (Wang e Singh, Verma, Page e Midilli). Foram determinados o coeficiente de difusão efetiva e energia de ativação. Na polpa in natura e desidratada foram analisados o teor de acidez titulável, pH, teor de sólidos solúveis e teor de vitamina C, assim como a solubilidade em água do pó de cajá e análise de variância da cor (coordenadas de L*, a* e b*). O modelo Wang e Singh indicou melhor ajuste para a temperatura de 50 °C e o modelo de Midilli demonstrou o melhor ajuste matemático para as temperaturas de 60, 70 e 80 °C. Os valores do coeficiente de difusão líquida e energia de ativação foram proporcionais à temperatura de secagem. A secagem da espuma a 60 °C indicou garantia da qualidade da polpa do cajá para acidez titulável, pH e coloração da coordenada de a*.


Introduction
Yellow mombin (Spondias mombin L.) belongs to the Anacardiaceae family and produces fruits that are grown in tropical areas of America, being known as 'taperebá' , 'cajámirim' or 'cajazinho' .Yellow mombin has the form of an ovoid drupe, between 3 and 5 cm in length, thin yellowish-colored peel and bittersweet taste (Tiburski et al., 2011).Yellow mombin fruits are used in the form of frozen pulps, juices, jellies, ice creams and jams (Janick & Paull, 2008).
Drying of agricultural products contributes to their conservation, reduces weight and volume occupied during industrialization and commercialization, protects against degradation reactions, and makes them available in any period of the year.In addition, it contributes to the concentration of nutrients because water is removed from the product (Fernandes et al., 2014).
Foam-mat drying is appreciated in foods sensitive to heat, viscous and with high sugar levels, such as fruits and fruit juices.Since this process occurs at lower temperatures of dehydration and for shorter time -because the surface area is larger in comparison to the drying with conventional bed -there is greater preservation of nutritional characteristics (Baptestini et al., 2015).According to Kadam et al. (2009), foam-mat drying contributes to the conservation of nutrients in tomato pulp.
Yellow mombin pulp processing by foam-mat drying can promote the preservation of nutrients in the fruit.Given the above, this study aimed to perform foam-mat drying of yellow mombin pulp, verifying the kinetics and mathematical modeling of the process, and characterize the obtained product with respect to physical and chemical characteristics, compared with the fresh pulp.

Material and Methods
Yellow mombin pulp was obtained by pulping fruits from a natural vegetation in the region of Montes Claros de Goiás,Goiás,Brazil (16º 06' 20" S and 51º 17' 11" W), manually harvested from December 2015 to February 2016, and transported in thermal boxes to the Laboratory of Fruits and Vegetables of the Federal Institute of Goiás -Campus of Rio Verde.
The foam was formed using Emustab ® , purchased at the local market, at 5% proportion (m/m).Yellow mombin (300 g) was added and subjected to agitation in an Arno ® food mixer for 20 min.Then, the samples were placed in triplicate on aluminum trays (5 x 9 x 20 cm) and taken to the oven at different temperatures.
Samples were dried in forced-air oven at temperatures of 50, 60, 70 and 80 ºC.According to Fernandes et al. (2014), 80 °C is the limit temperature for food dehydration, and lower temperatures are tested aiming at preserving nutrients at the end of the process.The samples were weighed until constant weight at regular intervals (30 min).The drying process was analyzed using the experimental data presented in Table 1.The models were applied to the experimental drying data through nonlinear regressions using the computer program Statistica, version 7.0.
The representativeness of the models was evaluated based on the coefficient of determination (R 2 ) and magnitude of the mean squared deviation (MSD) (Eq.5).Net diffusion was evaluated using the flat plate model, with approximation of eight terms (Eq.6) (Brooker et al., 1992).The model was fitted to the experimental data of yellow mombin pulp foam-mat drying, considering the superficial area and volume, according to the following expression: where: RX -moisture ratio of the product, dimensionless; n t -number of terms; D -effective diffusion coefficient, m²s -1 ; T -time, h; S -surface area of the product, m 2 ; and, V -volume of the product, m 3 , obtained by: where: W -width, m; L -length, m; and, T -thickness, m.
The surface area of the yellow mombin pulp foam was calculated by measuring length, width and thickness using a digital caliper (Leetools -150 mm, 0.01-mm precision).Then, the area was obtained by multiplying the length (L) by the width (W) of the tray used in the analysis, according to the Eq.8: where: S -tray surface area, m 2 ; L -length, m; and, ( R. Bras.Eng.Agríc.Ambiental, v.22, n.6, p.430-435, 2018.
The relationship between drying air temperature elevation and effective diffusion coefficient (D) was described by the Arrhenius equation (Eq.9).evaluated characteristics were interpreted firstly by grouping the cultivars with dry pulp, followed by comparison of all cultivars.

Results and Discussion
The experimental curves of yellow mombin pulp drying for the temperature range from 50 to 80 ºC are presented in Figure 1.The shortest drying time occurred at temperature of 80 ºC.Higher temperature reduces the drying time because of the greater amount of heat transferred from the air to the material, a fact also observed by Baptestini et al. (2015) in soursop foam-mat drying and by Furtado et al. (2010) in red mombin pulp drying.
Different mathematical models were fitted to the data obtained during the drying, to represent the phenomenon (Table 1).The experimental data of yellow mombin pulp drying kinetics and the coefficients of determination (R 2 ) and mean squared deviations for the temperatures of 50, 60, 70 and 80 ºC are presented in Table 2.
The lower the MSD, the better the fit to the experimental data.Experiments with fruit pulp use this parameter to selection the mathematical models (Feitosa et al., 2017).Among the models tested, Midilli showed the best fit, except for the temperature of 50 °C, in which the best fit was found using the Wang & Singh model, which showed the highest coefficient of determination (R 2 ) and lowest mean squared deviations (MSD).
After drying, the samples were removed from the trays and slightly agitated, manually, to obtain the yellow mombin pulp powder.Powder samples were diluted in distilled water at 1:2 proportion for physical-chemical characterization.Titratable acidity was determined by titrating the filtered juice, through a Unifil filter (150-mm mesh), with NaOH solution (0.01N), and the results were expressed in % of citric acid according to the method nº 942.15 of AOAC (2010).pH was measured using a digital benchtop pH meter (Luca -210 P, MS Tecnopon) according to the method nº 981.12 of AOAC (2010).In the quantification of soluble solids content, expressed in °Brix, the reading was taken in juice filtered in refractometer (ATAGO PR-101), according to the method n° 932.12 of AOAC (2010).Ascorbic acid contents were determined by oxidation-reduction volumetric analysis, with titration of yellow mombin samples in 2,6-dichlorophenolindophenol sodium (DCIP), according to the method n° 967.21 of AOAC (2010).Yellow mombin foam solubility in water was calculated by the difference of weight and expressed on dry basis according to the method described by Cano-Chauca et al. (2005).
Colorimetric evaluation was carried out in HunterLab ColorQuest II spectrophotometer at the Laboratory of Postharvest of Plant Products of the Federal Institute of Goiás, Campus of Rio Verde, measuring the values of luminosity index L* and chroma indices a* and b*.The results were expressed in L*, a* and b* values in the following range: L* (luminosity or brightness), from black (0) to white (100), a* from green (-60) to red (+60), and b* from blue (-60) to yellow (+60).
The results were subjected to analysis of variance (ANOVA) by F test, followed by means analysis by Tukey test at 0.05 probability level and regression analysis at 0.05 probability level, using the statistical program Sisvar (Ferreira, 2010)   that this mathematical model was the one that best described myrtle pulp foam-mat drying.For carambola drying, Santos et al. (2010) observed that the Page model showed the best fit to the experimental data.
Effective diffusion coefficients for yellow mombin pulp foam-mat drying are shown in Table 3.The values increased with the elevation in drying air temperature, probably because water removal is faster at higher temperatures.
These data agree with the results of Alves et al. (2016), who studied foam-mat drying of avocado pulp at the same temperatures from 50 to 80 ºC.This increasing trend occurs because the diffusivity represents the speed at which water moves from the inside to the surface of the material, being vaporized (Goneli et al., 2014).
According to Kayacier & Singh (2004), activation energy decreases with the reduction in the water content of the product during its drying.Thus, the activation energy found is an average of the values along the process.The average activation energy was 54.983 kJ mol -1 , and the activation energy for agricultural products varies from 12.7 to 110 kJ mol -1 according to Zogzas et al. (1996), which coincides with the value found in the present study.
The titratable acidity of yellow mombin powder dried at 50 °C differed from the other treatments (Figure 2A).For the treatment at 50 °C, the yellow mombin powder showed acidity of 1.2 g 100g -1 of citric acid, while in the other treatments the values ranged from 0.6 and 0.8 g 100g -1 of citric acid.In general, acidity levels do not exceed 1.5 to 2.0% (Franco, 1998).
The pH values of yellow mombin powder dried at different temperatures showed significant differences, as shown in Figure 2B.Elevation of temperature caused a decline in pH values, except at the temperature of 60 °C, as reported by Abbasi & Azizpour (2015), who found reduction in pH values with the increment of temperature in foam-mat drying of sour cherry pulp.Falade & Okocha (2010) reported similar results for plantain powder using the same method as the present study.
Foam-mat drying caused increment in the contents of soluble solids when compared with the yellow mombin pulp.This characteristic is due to the foaming agent and the dehydration process, which cause concentration of solid substances (Figure 2C).The soluble solids content showed highest value at 60 °C, which contributed to claiming that the greatest water elimination occurred in this situation, confirming the high pH values and low contents of titratable acids.
The treatment at 80 ºC showed lower concentration of soluble solids, which may indicate that at higher temperature pulp drying is faster and may lead to higher moisture content at the end of the process (Figure 2C).The increase in soluble solids content was reported for the foam-mat drying of genipap and 'araçá-boi' fruits, using the additive Emustab ® (Pinto, 2009;Soares, 2009).

E.
Titratable Acidity (g 100g -1 citric acid) Vitamin C (g 100g -1 ascorbic acid) Figure 2D shows the vitamin C content of yellow mombin pulp in the different treatments.There were differences and the highest content of ascorbic acid was found at 80 °C, due to the concentration of substances present in the pulp.
Similar results were reported by Silva et al. (2008) in the dehydration of tamarind pulp by the foam-mat drying method.Drying was carried out at the same temperatures (50, 60, 70 and 80 ºC) as the present study and the same variations occurred; vitamin C decreased and increased when all dried pulps were analyzed, with highest value at the temperature of 80 ºC.
The longer time for the treatments at 50, 60 and 70 °C may have caused vitamin C exposure to moisture, light and oxygen, contributing to its degradation, which is due to the sensitivity to heat of the ascorbic acid molecule (Fernandes et al., 2014).Ascorbic acid content decreased in the drying of potato (Mehta et al., 2007) and onion (Kadam et al., 2009) subjected to different treatments of temperature.
According to the solubility in water, as temperature increased there was a reduction in the solubility of yellow mombin powder produced by foam-mat drying (Figure 2E).For the lowest temperature (50 °C), the solubility of the powder is higher because it still contains water molecules in its composition.According to Nadeem et al. (2011), the more stable the drying process, the more bubbles will form in the foam mat, which contributes to porosity and increment in powder's solubility in water.The coefficient of determination indicated good fit of the model (R² = 0.9397).
Significant differences occurred in the luminosity parameters between the fresh pulp and the pulp dried at different temperatures (Table 4).The results obtained as temperatures increased demonstrated that the drying process led to a darker pulp.Batista et al. (2014) found that the drying of banana samples at various temperatures caused a decline in luminosity, as occurred for yellow mombin pulp samples.Fruits subjected to drying processes become darker due to the non-enzymatic browning and possible losses of pigments (Travaglini et al., 2002).
The characteristic of green (negative values) to red (positive values) color, represented by the coordinate a*, demonstrates that at higher temperatures there is an increment in a* values, intensifying the red color.The lowest effect of temperatures, compared with the fresh pulp, occurred for the lowest temperatures (50 and 60 ºC) and, consequently, there was no significant difference between them.This fact was also reported by Medeiros (2007) in the foam-mat drying of mangaba, in which the values of red intensity were proportional to the increase of temperature.Such difference may result from the formation of compounds with color close to brownish-red (Reis et al., 2006).
For the coordinate b*, varying between yellow and blue, there were variations; however, the lowest temperatures (50 and 60 ºC) did not differ significantly (p ≥ 0.05), tending to cause a yellow color.At temperature of 80 ºC, the color was the closest one to that of the fresh pulp.

Conclusions
Figure 1.Experimental moisture contents of yellow mombin (Spondias mombin L.) subjected to different drying temperatures Net effective diffusion coefficient (m² s -1 ) and activation energy (kJ mol -1 ) Figure 2. Mean and standard deviation of titratable acidity (A), pH (B), soluble solids content (C), vitamin C content (D) and solubility in water (E) of yellow mombin (Spondias mombin.L) dried at different temperatures

Table 1 .
Mathematical models evaluated to describe the drying process . The

Table 2 .
Values of coefficient of determination (R 2

Table 4 .
Color evaluation of samples of yellow mombin (Spondias mombin L.) pulp, fresh and powdered, after foam-mat drying *Means followed by lowercase letters differ significantly by Tukey test 0.05 probability level