The study used the Malayan dwarf coconut variety from the Colombian Pacific coast, and the CP+FAC was obtained through SD according to the methodology described in Lucas et al., [17]. The product was packaged with N₂ and atmospheric air, using multilayered bags (Alico S.A.), laminated film (PET) thickness: 12 (m, aluminum foil: 8 (m, polyethylene sealant layer: 100 µm, grammage of 136.4 g/m2, water vapor barrier (<5 cc/m² x 24 h x atm) and O2 (<5 cc/m² x 24 h x atm). Storage was carried out in climatic chambers conditioned to 65% relative humidity, temperatures: 15, 25, and 35 °C and times: 0, 30, 60, 90, 120, 150, and 180 days.

Vitamin C extraction was conducted according to the methodology described by Silva et al., [18], and vitamins D₃ and E according to the methodology proposed by Ruiz-Ruiz et al., [19], modified by inclusion treatment through ultrasound (40 kHz) during 20 min, for the samples treated with hexane. Vitamin C quantification was performed through high-resolution liquid chromatography (H-RLC) (Shimatzu Prominence 20A), using a column in reverse phase (C18 - 5 µm, 4.6 mm x 250 mm), diode array detector, mobile phase: KH₂PO₄ 0-02 M pH = 3.00 (Ortho-Phosphoric Acid 85%), flow: 1 mL/min, furnace temperature of 35 ºC, 244-nm wavelength and injection volume of 5 µL. Vitamins E and D₃ were quantified using a column in reverse phase (C18 - 5 μm, 4.6 mm x 250 mm), diode array detector, mobile phase: acetonitrile/methanol/water (45.3/51.2/3.5), flow: 1 mL/min, furnace temperature of 40 ºC, and wavelengths of 285 and 265 nm, respectively [20].

Total phenol (TP) content was determined according to the methodology described by Zorić et al., [14], (mg Gallic acid/g CP+FAC); The DPPH free radical scavenging activity (mg of trolox/g CP+FAC) and ABTS (mg of trolox/g CP+FAC), was determined according to the extraction methodology proposed by Sridhar and Linton-Charles, [21] and quantification according to Casagrande et al., [22].

The evolution of the FAC content, total phenols, and antioxidant activity in the CP+FAC was quantified in terms of retention percentage (Eq. 1) and modeling with zero - first - and second-order kinetics, described in Eq. 2, 3, and 4, respectively, where C ₀ and C _t are the values of the quality attribute evaluated at time 0 and t, with k being the degradation constant [14,23,24].

Additionally, the study evaluated the effect of temperature from the Arrhenius equation (Eq. 5), determining the activation energy (E_a) and the pre-exponential or frequency (A) factor, with T being the absolute temperature (K) and R the universal constant of gases (8.314*10^-3 kJ/K* mol) [14, 23, 25]; where E_a was calculated from the slope of the graphic Ln (k) vs. 1/T.

The evaluation of the stability of the CP+FAC properties during storage was conducted by applying a design with sixth-* seventh-factor factorial arrangement with two independent variables: treatment and time, where treatment is defined as the combinations (temperature - packaging). The samples were stored in multilayered bags (Alico S.A.) and packaged in N₂ atmosphere and with atmospheric air (Amb). The treatments applied were the following: 15 ºC-N₂, 15 ºC-Amb, 25 ºC-N₂, 25 ºC-Amb, 35 ºC-N₂, and 35 ºC-Amb, and the storage times (t_S): 0, 30, 60, 90, 120, 150, and 180 days. The dependent variables evaluated were: retention percentage of vitamins and antioxidants (%R): %R-VC, %R-VD₃, %R-VE, %R-TP, %R-DPPH, and %R-ABTS.

Analysis of the results was performed by using Statgraphics XVI.I software, with analysis of variance (ANOVA) with multiple range test, 95% confidence interval, and the Tukey’s test of honest significant difference (HSD). The linear model generalized to relate the response to the independent variables was the following (Eq. 6):

The properties of the CP+FAC obtained through SD were as follows: humidity: 1.68±0.15%; a_w: 0.17±0.02; solubility: 58.40±2.06%; Hygroscopicity: 8.36±0.55%; color: (L*: 79.45±0.90, a*: 1.50±0.13, b*: 9.50±0.45); wettability: 263.00±19.80 s, Peroxide index: 2.43±1.28 meqH₂O₂/kg fat; distribution of average particle size: (D₁₀: 1.70±0.05 µm, D₅₀: 8.46±2.09 µm, D₉₀: 78.18±24.30 µm).

The proximal composition of the CP+FAC was: fat: 30.54±0.90%; protein: 4.07±0.49; total dietary fiber: 23.87±1.61%; ashes: 2.33±0.03%, and humidity: 1.68±0.48%, and the amounts of vitamins and antioxidants at time zero (0 days) were: vitamin C (VC): 2.11(0.10 mg/g of CP+FAC; vitamin D₃ (VD₃): 223.76(12.51 mg/g CP+FAC; vitamin E (VE): 0.20(0.01 IU/g CP+FAC; total phenols (TP): 29.7(7.6 mg Gallic acid/g CP+FAC, DPPH: 0.46(0.004 mg trolox/g CP+FAC and ABTS: 0.54(0.002 mg trolox/g CP+FAC.

The ANOVA had statistically significant differences (p <0.05) in all the dependent variables, with respect to t_S, treatment (Temperature-Packaging), and the Treatment-t_S interaction; in general, during 180 days of storage under the conditions described, significant losses occurred in all the treatments, both in vitamins as in antioxidants; with greater %R-Vitamins and Antioxidants at lower T_A, and in treatments where the CP+FAC was packaged with N₂ (Tables 1 and 2).

Table 1

Source: The Authors.

Table 2

Source: The Authors

This losses are associated with multiple factors, related with storage conditions, like pH, T_A, and t_S, presence of catalysis of metal ions, concentration of salt, sugar, oxygen, enzymes, light, increased a_w, as well as X_w, thermal treatment applied to the product (Table 1) [2,16,20,24,26-29].

Stability studies are limited and contradictory; degradation of VC, VD₃, and VE during storage could be related to the reactivity with the oxygen present in the samples packaged with atmospheric air and to the presence of ferrous ions, which - in part - explains the greatest loss of VC when compared with that packaged with N₂, where, by it being an inert, colorless, odorless, and insipid gas, it does not react chemically with other substances and has low solubility. It is used as oxygen substitute and displaces it in the packaging head space to avoid the development of aerobic microorganisms and oxidation problems [16,20,24].

Hymavathi and Khader, [30], Chávez-Servín et al., [4], Liu et al., [31] and Bosch et al., [32], reported significant losses of L-ascorbic acid, which increased with increased t_S and T_A in mango powder, powdered milk for infants, powdered tomato, and fruit-based foods for children, respectively; oxidation of L-ascorbic acid starts with the formation of dehydroascorbic acid, which can undergo a series of reactions with amino acids that lead to the active formation of pigments.

Udomkun et al., [1] and Prodyut et al., [33], with dry chunks of papaya and pineapple, respectively, reported that degradation of ascorbic acid (AA) accelerated exponentially during the first three months of storage in different types of packaging, which is associated with increased a_w, as well as X_w, given that free water can act as a solvent for reagents and as a catalyst. Hemery et al., [34], reported losses of VD₃ and VE during storage of fortified soy oil, when the oils were exposed to natural light; the factors with greatest influence on the retention of vitamin D₃ were content of α-tocopherol and the interaction between time and type of oil.

Syamila et al., [16], found instability when evaluating the effect during storage of temperature, oxygen, and light on the degradation of β-carotene, lutein, and α-tocopherol in spray dried spinach juice powder, where the reduction of oxygen levels through vacuum packing and exclusion of light intensity below 800 lux did not improve the stability of the micronutrients. This made it possible for the residual oxygen in the powder to be sufficient to cause the degradation of the nutrients under study.

Romero-Braquehais [24], evaluated the behavior of VD and VE under storage in liquid milk at different temperatures (23 - 30 and 37 °C), demonstrating the effect of temperature on the instability of vitamin D, with 37 °C being where the highest loss occurs during the first three months of storage.

This behavior was similar to that found in the present work. It is known that vitamin D is susceptible to an alkaline pH range, light and heat, lipid oxidation, and to small amounts of transition metals.

Retention percentages of the antioxidant capacity %R-TP and %R-DPPH degraded linearly throughout the storage period, except for %R-ABTS that accelerated during the last 30 days of storage (Table 2), which is associated with T_A, t_S, pH, as well as with exposure to oxygen, light, changes in X_w, a_w, RH, formation and accumulation of melanoidins, along with Maillard reaction products and type of packaging material, where the availability of oxygen for oxidation reactions as a result of the different gas permeability of the packaging materials, the oxidative enzymes, like polyphenoloxidase, or the chemical oxidation of phenolic compounds [1,2,7,14,28,29,35-37].

Losses of vitamins and of antioxidant capacity followed different kinetic models of degradation, where VC, DPPD, and ABTS had zero-order reactions, while TP were of first order and VD₃ and VE of second order for both the CP+FAC packaged with atmospheric air and with N₂ (Table 3).

Table 3

k: rate constant; -(-) the minus sign indicates FAC degradation; C_o: Initial concentration; R²: determination coefficient.

In VC, the correlation coefficients (R²), for all temperatures and types of packaging had - in general - values (0.9; the values of k show the degradation rate, which are higher for that packaged with ambient air than with N₂ (higher values of k, lower stability) (Table 3). Similar behavior was reported by Romero-Braquehais [24], in liquid milk. While Shinwari and Rao [29], reported that ascorbic acid in berry marmalades and jams is more sensitive during storage after first-order degradation kinetics, where the storage temperature has a significant effect.

The effect of storage temperature on the degradation of VC determined via the Arrhenius equation, under zero- and first-order kinetic models, yielded that the activation energy (E_a) for the first-order kinetic model was 44.679 kJ/mol, and for the zero-order model it was 24.584 kJ/mol, in ambient packaging, while values of CP+FAC packaged with N₂ were lower at 27.317 and 21.743 kJ/mol, for each kinetic model respectively; while the correlation coefficients (R²), were higher for the zero-order kinetic model, in both types of packaging (Ambient - 0.993; N₂ - 0.978) (Table 4). Similar results were reported by Romero-Braquehais (2008) [24] in liquid milk regarding the zero-order kinetic model and R²= 0.99, but lower regarding E_a= 13.91 Kcal/mol. The difference in E_a with respect to those obtained in the different studies may be due to the different temperature ranges and to the different environmental conditions, such as oxygen, humidity, pH, and - of course - by the composition of the matrix [38].

Table 4

E_a: activation energy (kJ/mol); A: pre-exponential factor; R²: determination coefficient.

The E_a can be expressed as the energy barrier the molecules need to cross to be able to react. The proportion of molecules capable of doing so increases with temperature; which is why the effect of temperature on the rates and the physical significance of A represents the rate constant at which all the molecules have enough energy to react (E_a= 0) [24]. The activation energy indicates, somehow, how fast the reaction rate is. In this case, the activation energy for the zero-order kinetics was inferior to that obtained by the first-order kinetics, indicating that VC degradation is higher in zero-order kinetics than in first-order kinetics. Romero-Braquehais [24], Bosch et al., [32], evaluated the degradation kinetics of ascorbic acid (AA) in liquid milk and in fruit-based foods for children during storage, reporting a first-order kinetics, where the values registered for k, obtained for the first-order kinetics were 0.05415, 0.2199, and 0.7469 at 25, 37, and 50 °C, respectively, thus, confirming the degradation of AA dependent on temperature and the value of the E_a was 20.11±0.33 kcal/mol, proving that the highest kinetic degradation of the FAC is obtained with lower E_a.

The degradation kinetics of VD₃ had second-order reactions, where k values, which show the rate of degradation, were very similar, both for that ambient packaged and that packaged with N₂ (Table 3). The correlation coefficients (R²), for all temperatures and types of packaging, generally had relatively valid values (0.67 and (0.84, values similar to those reported in liquid milk by Romero-Braquehais [24], showing first-order kinetics and with scarce data available in the literature, it is not possible to ensure if the losses of vitamin D follow a zero-, first-, or second-order kinetic equation, thereby, it would be necessary to increase the t_S or the T_A to have a clearer vision. Regarding the effect of temperature on VD₃ degradation, the first-order kinetics presented R² (0.86 for that packaged with atmospheric air (EA) and that packaged with N₂ (EN₂), while those of zero order had R² = 0.646 and 0.853, respectively. The E_a for the zero-order reaction were 3.498 (EA) and 4.069 (EN₂) kJ/mol, lower than those of first order at 8.505 (EA) and 8.757 (EN₂) kJ/mol, suggesting higher dependence on temperature in the rate of VD₃ degradation (Table 4).

Vitamin E had a second-order kinetic degradation model, for all temperatures and packaging conditions, with similar behavior for liposoluble vitamins, the R² were (0.9; rate constants (k) were higher for the samples packaged with atmospheric air than with N₂ (Table 3), which indicates a slower degradation of the VE packaged with N₂. The effect of storage temperature on the degradation of VE, for the zero-order and first-order kinetics, showed R² (0.75 and (0.99 for EA and EN₂, respectively, while the E_a was lower for the zero-order reactions and for that packaged with N₂ (Table 4).

Romero-Braquehais [24], reported for liquid milk first-order degradation kinetics of VE, and E_a similar to those found in this work. Syamila et al., [16], in powdered spinach, reported first-order kinetics for α-tocopherol, showing a gradual decrease of the content through the storage time. The high rate of degradation of nutrients with low water activity suggests that the reaction was of self-oxidative nature, and it could be argued it was due to the presence of residual oxygen in the packaging. It is known that the dissolved residual oxygen is responsible for the deterioration of β-carotene and α-tocopherol during long periods of storage, even if the packaging head space is purged with N₂.

With respect to the degradation of antioxidants; TP had first-order degradation kinetics for all temperatures and packaging conditions, where k were higher for EA (more susceptible to degradation), but the higher R² for EN₂ ((0.9) (Table 3). The effect of storage temperature on TP degradation determined through the Arrhenius equation, under the zero-order and first-order kinetic models, yielded that the activation energy for the first-order kinetic model and for the EA, was higher than for those of zero order and EN₂ (Table 4), and the R² were (0.95. The aforementioned agrees with that reported by Shinwari and Rao [29], where the TP diminish during storage in jams and marmalades at high storage temperatures, bearing in mind the composition; where pectin stabilizes highly reactive compounds, like anthocyanins, vitamins, and antioxidants through hydrogen or the hydrophobic union. The magnitude of these interactions is based on the type and concentration of pectin. Where pectins with a low degree of esterification minimize the loss of bioactive compounds and retention is better in higher concentrations of pectin. Likewise, the stability of the functional compounds varies with the type and variety of the fruit, given that they differ in their composition (sugar, pectin, minerals, etc.).

The loss of antioxidant capacity, determined for ABTS and DPPH, presented zero-order degradation kinetics, with the same behavior in the values of k and R², although in ABTS, the values of R² ranged between 0.61 and 0.69; while DPPH (0.9; better describing the behavior of loss of antioxidant capacity in these (Table 3). While the E_a and R² of the effect of the temperatures under the zero-order and first-order kinetic models in the DPPH were higher for the EA, but the R² was lower than the EN₂, presenting the contrary effect in ABTS in the E_a, but similar R² and (0.82 (Table 4).

Kim et al., [39], reported that the degradation of the total phenolic content and the antioxidant activity (DPPH) in kiwi (Actinidia arguta) puré, stored at various temperatures between 5 and 45 °C during 72 h, followed a first-order kinetic model and indicated a strong dependence on the temperature of the phenol content and the antioxidant activity, thus, the degradation of the phenolic compounds and the antioxidant activity occurs faster with increased temperature and storage time, with values of E_a of 28.15 and 29.07 kJ/mol, respectively.

Zorić et al., [14], studied the phenolic degradation kinetics of cherry powders and found a first-order reaction, represented by a more progressive loss at higher storage temperatures and it was confirmed by kinetic parameters, where the rate constant (k) varied between 0.02 and 0.067 and the activation energy (E_a) between 10.956 and 31.59 kJ/mol and the R² were between 0.64 and 0.99, finding that the highest phenolic degradation in powders takes place with lower E_a; a similar behavior was found in the present work.

Henríquez et al., [35], with apple peel powder and using two packaging materials (high-density polyethylene, HDPE, and metalized high-barrier films, MHBF), observed that phenols diminished during the storage period in both packaging materials tested, independent of the storage conditions (temperature and RH) used. This was confirmed by the kinetic parameters of the phenolic degradation throughout the storage time, where in all cases, the regression analysis suggests a zero-order kinetics, with a relatively lower value of E_a in HDPE compared with MHBF (3.038±84.85 cal/mol K and 3.220±157.03 cal/mol K, respectively) and where the highest k values (constants for phenolic degradation) in the first package could suggest that the degradation of phenols may be associated with the product’s humidity gain. Daza et al., [37], with cagaita (Eugenia dysenterica) powder, reported that degradation of phenols exhibited first-order kinetic behavior. In effect, the loss of attributes of food quality takes place in higher humidity conditions, when there is greater mobility of the reagents and dissolution of organic acids.

Udomkun et al., [1], in dry papaya, evaluated the effects through type of packaging (aluminum and polyethylene films, ALP, and polyamide/polyethylene, PA/PE, films) and through t_S over the antioxidant capacity and the total phenol content, reporting that the quality parameters followed zero-order reactions, except for ascorbic acid (AA), which was described better by using a first-order quality model parameter and where the R² were >0.95; and the values of k, which show that the rate changes were more drastic for the samples stored in PA/PE packages compared with those in ALP bags, meaning that greater degradation was produced in dry papaya when packaged in PA/PE.