Water Diffusivity and Quality Attributes of Fresh and Partially Osmodehydrated Cactus Pear (Opuntia Ficus Indica) Subjected to Air-Dehydration

Abstract

Water diffusivity, vitamin C degradation, and color change were assessed in two batches of cactus pear that were dried; one by regular air-drying; and the other one by applying partial osmodehydration followed by air-drying. The drying was done with a convection oven at 40, 50, 60, and 70°C. The pretreatment was performed by immersing the samples for 3 hours in a 40°Brix sucrose solution at 40°C. An analysis of variance (ANOVA) with a confidence level of 95% (p < 0.05) revealed that water diffusivity was mainly affected by the temperature and the pretreatment. Both the vitamin C loss and the color change behaved as a first-order reaction; the set temperature used was found to be the most important factor in the process.

Keywords:

• Water diffusivity
• Mass transfer
• Cactus pear
• Vitamin C degradation
• Color change

INTRODUCTION

Conventional air-drying is one of the most widely used techniques in food dehydration. This method improves the stability and extends the shelf life of the product.[1] Dehydration also brings about substantial reduction in weight and volume; this reduces packing, storage, and transportation costs. However, the removal of water may considerably reduce the quality of these products. The most common quality defects of dehydrated foods are: slow or incomplete rehydration, loss of typical juiciness of fresh foods, as well as unfavorable changes in color and flavor. The drying time and the temperature applied are the most important factors in the process.[2] The use of low temperatures gives the product better stability, but reduces its quality, since the exposure times have to be longer. Thus, the use of an osmotic treatment before conventional drying as a means of enhancing the texture, taste, nutrient retention and color stability of several products[3,4] has received considerable attention recently. Osmotic treatment consist in partially removing water from the product by immersing it in aqueous solutions of sugar, salt or spices.[5–7] During this treatment, a significant amount of water is removed and simultaneously infusion of desirable solutes takes place.[5–7] However, when the process is concluded, further drying such as convective or freeze-drying, is still necessary to reduce moisture to a minimum. This step improves the stability of the product.

Although Lahsasni et al.[8] have recently proposed convective solar drying and, similarly, Moreno-Castillo et al.[9] have suggested osmotic dehydration for that purpose, there is not published information concerning the use of an osmotic treatment followed by convective drying applied to cactus pear. Therefore, this work was undertaken to investigate this aspect further, and the main objective was to evaluate the effect of the temperature on the drying kinetics, ascorbic acid degradation and color change of fresh and osmotically pretreated cactus pear subjected to air-dehydration. Drying kinetics was described using a diffusional model adapted to slab geometry, whereas first-order kinetics was proposed to describe the rate of vitamin C loss and the rate of color change.

MATERIALS AND METHODS

Experiments with Fresh versus Pretreated Samples

Cactus pear fruits (variety Opuntia ficus indica) with an average weight between 100 and 110 g and similar ripening (considering total soluble solids, °Brix, and peel color) were purchased at a local market. Commercial sucrose was used in experiments. The fruits were carefully washed with water to get rid of the spines, then hand peeled and cut into cylindrical slices 5-mm thick and approximately 49 mm in diameter. Two similar batches were formed with the fruit in order to perform the two different experiments.

For the conventional drying process, an air flow was applied on the two circular faces of the slices. This was performed with a convection oven, using 4 different temperatures (40, 50, 60, and 70°C), and with an air velocity of 1 m/s. The drying process was monitored periodically weighing the samples with an electronic balance (± 0.0001 g), and finally stopped when the weight reached a constant value.

The pre-treatment of the other batch was done by placing the samples into a beaker filled with a 40° Brix sugar solution. Then, the beaker was immersed into a water-bath maintained at a constant temperature of 40°C. Three-hour immersion periods under static conditions and a product/solution ratio of 1/15 were used in these experiments. When the pre-treatment was complete the samples were subjected to the air-drying process described above.

A factorial experimental design was used, all tests were done in triplicate, and a total of 24 runs were carried out (Table 1). Samples were taken out of the drier at regular intervals to mesure ascorbic acid content and color change. The former was determined by direct iodine titration (0.1 N) of suitable dilutions of the product in sulphuric acid and starch solutions.[10] In the case of color, the L*, a*, and b* parameters (Hunter values) were measured before and during the drying by means of a colorimeter (AccuProbe HH06^TM, Accuracy Microsensors, Inc. Pittsford, New York).

Table 1 Values of the water diffusivity (D _W ) and reaction rate constants (k ₁ ) and (k ₂ ) determined through regression method for each experimental condition

Run no.	T (°C)	Treatment	Eq. (1)		Eq. (3)			Eq. (5)
			D W *10^−10(m^2/s)	r^2	Co (mg/100 g sample)	k 1(min^−1)	r^2	k 2(min^−1)	ΔEmax	r^2
1	40	FS	1.75	0.993	25	7.49 × 10^−4	0.957	0.0035	13.70	0.986
2	50	FS	2.89	0.988	23.43	1.34 × 10^−3	0.890	0.0006	38.93	0.959
3	60	FS	4.57	0.997	22.01	1.15 × 10^−3	0.864	0.0025	19.80	0.931
4	70	FS	4.71	0.990	25	1.49 × 10^−3	0.936	0.0177^a	7.93	0.928
5	40	PTS	1.58	0.996	25	2.89 × 10^−4	0.935	0.0025	6.73	0.845
6	50	PTS	2.11	0.991	25.3	1.02 × 10^−3	0.942	0.0026	11.69	0.913
7	60	PTS	3.01	0.993	26	1.73 × 10^−3	0.917	0.0051	13.21	0.979
8	70	PTS	4.99	0.997	25	2.33 × 10^−3	0.961	0.0086	10.76	0.956
9	40	FS	1.75	0.994	25	6.68 × 10^−4	0.809	0.0031	7.45	0.873
10	50	FS	2.68	0.987	25	6.93 × 10^−4	0.851	0.0003	99.82	0.967
11	60	FS	4.20	0.997	22.6	1.29 × 10^−3	0.911	0.0093^a	11.80	0.968
12	70	FS	4.92	0.993	25	2.01 × 10^−3	0.924	0.0023	15.01	0.937
13	40	PTS	1.58	0.983	25	4.46 × 10^−4	0.943	0.0056	8.961	0.958
14	50	PTS	2.80	0.997	25	8.15 × 10^−4	0.958	0.0050	6.84	0.896
15	60	PTS	2.88	0.998	26	2.50 × 10^−3a	0.943	0.0039	13.12	0.919
16	70	PTS	4.37	0.999	25	1.99 × 10^−3	0.959	0.0103	11.20	0.930
17	40	FS	1.51	0.996	25	9.57 × 10^−4	0.943	0.0029	16.02	0.886
18	50	FS	2.81	0.989	24.3	1.49 × 10^−3	0.973	0.0042	17.96	0.979
19	60	FS	4.40	0.994	21.2	1.29 × 10^−3	0.974	0.0032	14.01	0.972
20	70	FS	5.32	0.994	21.5	1.93 × 10^−3	0.976	0.0066	15.26	0.933
21	40	PTS	1.67	0.994	25	4.70 × 10^−4	0.900	0.0034	12.58	0.984
22	50	PTS	2.57	0.998	25	7.49 × 10^−4	0.883	0.0057	15.10	0.995
23	60	PTS	3.31	0.997	24.5	1.73 × 10^−3	0.945	0.0034	12.04	0.990
24	70	PTS	4.20	0.996	21	1.23 × 10^−3a	0.965	0.0088	12.53	0.984
FS: Fresh samples; PTS = Pretreated samples; and ^aOutlier values.

Moisture Diffusivity Calculation

The slices were considered as infinite slabs and the analytical solution of Fick's second law for unsteady state diffusion was fitted to the drying curves by a non-linear regression procedure, taking the first 10 terms of the series[11]:

(1)

where: X(t) and X_o represent, respectively, moisture content (g water/g dry matter) at time tand at initial time; D_w is water effective diffusivity (m²/s); Lrepresents the half thickness of the slice (m); and tis the process time (s). In order to apply Eq. (1), it was assumed that a) the mass transfer takes place only in one direction; b) the initial moisture content of the product is uniform; c) the slice surface has already reached moisture equilibrium; d) D_w is constant; e) slice shrinkage is negligible; and f) the values of the equilibrium moisture content are relatively small compared to X(t) or X_o .

Kinetics of Ascorbic Acid Degradation

To describe nutrient degradation in thermally processed foods, as well as during food storage periods, a first-order kinetic has been generally used[12]:

(2)

where Cis the instantaneous nutrient concentration (mg/100g sample); k ₁is the temperature dependent degradation rate constant (min⁻¹); and the exponent mis the reaction order. For isothermal conditions and in the case of a first-order reaction (m  =  1), Eq. (2)can be analytically integrated to obtain a degradation function:

(3)

C_o is the initial ascorbic acid concentration and tis the drying time. Eq. (3)has been used to assess vitamin C degradation in several processes, such as the storage of orange juice[13] and dehydrated guava,[14] as well as the dehydration of pineapple.[15]

Kinetics of Color Change

The overall color change (ΔE) was measured with Hunter's values L*, a*, and b* using Eq. (4):

(4)

where L _o *, a _o *, and b _o * denote the color parameters before the drying process; and L*, a*, and b* correspond to the ones obtained afterwards. The experimental data were fitted using a first-order kinetics model[16] to estimate color change as a function of time, as follows:

(5)

where ΔE _maxis the largest color change reached in the experimental condition; and k ₂is a first-order constant (min⁻¹) representing the rate of the food's color change.

Statistical Analyses

The experimental data were fitted using a polynomial model to predict the values of the dependent variable Y(D_W, k₁and k₂ ):

(6)

where β _i (i= 0, 1 … 5) represents the linear regression coefficients of the model; and T and Tr, are the coded values of the independent variables (temperature, pretreatment). The analysis of variance (ANOVA) was performed with a confidence level of 95% (p < 0.05) using the software Modde 7.0.

RESULTS AND DISCUSSIONS

Drying Rate

The moisture content versus drying rate are shown in Fig. 1a(F samples) and 1b (PT samples), respectively. The drying process took place in the falling rate period, since the drying rate dropped steadily as the moisture content decreased. A similar phenomenon was observed in the dehydration of papaya,[3] pear,[4] and bell pepper[17]; it has been attributed to the colloidal and hydrophilic characteristics of biological materials. In this case, the water molecules are strongly bound and the presence of free-water is insignificant.[17] Actually, diffusivity is the physical mechanism governing the moisture transfer inside foods during the falling rate period. It is a complex mechanism that involves movement of water, in both liquid and vapor phase, and it is often characterized by a coefficient named effective diffusivity (D_W ). Figure 1(a, b) shows that, in general, higher temperatures brought about increments in the drying rate, and that this varied linearly with the moisture content.

Figure 1Moisture content versus drying rates of cactus pear. (a) Fresh samples; and (b) pretreated samples.

Drying Kinetics

Figure 2a(F samples) and 2b (PT samples) presents the variation of the moisture content as a function of time. At the beginning of the process, the moisture content quickly dropped, but this variation slowed down as the drying progressed. The temperature clearly affected the water loss rate, since higher temperatures noticeably reduced the corresponding drying times. The shortest drying times occurred at 60 and 70°C.

Figure 2Effect of the temperature on the drying curves of cactus pear. (a) Fresh samples; and (b) pretreated samples.

The fresh samples registered an initial moisture content of 5.45 ± 0.39 g of water/g of dry matter, while the pretreated samples had 2.78 ± 0.19 g of water/g of dry matter, which corresponds to 84.5% and 73.5%, respectively; this means that the pretreatment caused an humidity decrease of about 11%.

The simulated curves obtained with Fick's equation [Eq. (1)] are also shown in Fig. 2. When calculating the values for water diffusivity (D_w ), the corresponding values for the determination coefficients (r²) were all greater than 0.98; they are listed on Table 1. The water effective diffusivity varied from 1.51 × 10⁻¹⁰to 5.32 × 10⁻¹⁰m²/s, within the range of temperature (40–70°C) covered in this study. This range of variation is similar to that reported for the dehydration of bell pepper[17]; it is also similar to values reported by Maroulis et al. [18] in their compilation about water diffusivity in biological products.

To evaluate the effects of the experimental conditions variations on the D_W -values, a multiple linear regression analysis was performed using Eq. (6)and a confidence level of 95%. The regression coefficients of the model are shown in Table 2. D_W was influenced by the temperature (T), the pretreatment (Tr), and by the T* Trand T ²* Trinteractions. Table 2also shows that, according to the lowest p(t)values, the temperature and the pre-treatment were the most important process variables affecting the water diffusivity. The following equations are the transfer mass models used to describe water diffusivity during prickly pear drying (both with R² = 0.958):

(7)

(8)

Table 2 Regression coefficients of the polynomial model [Eq. (6)] to evaluate the effect of temperature and pretreatment on the D _W , k ₁ , and k ₂ values (p < 0.05). The bold character indicates that the corresponding parameter had a significant effect on the properties evaluated

Regression coefficients of the linear model [Eq. (6)]
Response	β0	β1	β2	β3	β4	β5
Water diffusivity DW	3.1846	1.5634	−0.4397	0.01181	−0.16955	0.30918
p(t)	1.3528e-17	1.583e-13	0.0002317	0.9312	0.049377	0.034215
Constant k1	0.0012326	0.00069515	3.5249e-5	6.2385e-5	0.0002264	−4.093e-5
p(t)	5.2961e-11	1.99345e-8	0.666138	0.588246	0.00421982	0.72175
Constant k2	0.00300479	0.0015675	0.0009973	0.0021769	0.0008175	0.00035437
p(t)	4.019e-5	0.0033	0.0817755	0.0109016	0.091217	0.645604
p(t): t-student probability.

The graphs in Fig. 3apresent water diffusivity as a function of temperature. It is clear that D_w increased as the temperature increased; the values recorded for water diffusivity at 70°C were about 2.5 times greater than the ones obtained at 40°C. Figure 3aalso shows that the D_w values of the PT samples were slightly smaller than those of the FS samples; a layer of sucrose might have formed on the surface of the pretreated slices, and this in turn might have reduced the amount of water given off by the fruit.

Figure 3Water diffusivity versus air-drying temperature. (a) Polynomial model; and (b) Arrhenius' model.

The relationship between diffusivity and temperature has also been represented with Arrhenius' model by many authors[17,19,20]:

(9)

where D_o is the pre-exponential factor (m²/s); Eais the activation energy (kJ/mol); Ris the ideal gas constant (kJ/mol K); and T is the absolute temperature of the air (K). Applying Eq. (9)to the dehydration of cactus pear, the variation of D_W as a function of the temperature would be:

(10)

(11)

Eqs. (10)and (11)allow to determine the activation energy, which was 33.56 kJ/mol for the fresh samples, and 29.45 kJ/mol for the pretreated samples. This was due to initial difference in moisture content (84.5% and 73.5%), which was discussed above. These activation energy values are consistent with the ones obtained for the dehydration of olive cake, [20] and dehydration of yerba mate leaves,[21] but they are about 2 times smaller than the values reported for the dehydration of bell pepper.[17]

By comparing the graphs in Fig. 3a 3b, it becomes clear that in both cases, D_W varied as a function of the temperature in the same way. This indicates that the mathematical models proposed in this paper (Eq. 7and 8)are as precise as the Arrhenius model. These graphs also show how the D_w values (predicted and experimental) were influenced by the temperature and pre-treatment.

Ascorbic Acid Degradation

Ascorbic acid is an essential ingredient in people's daily diet, and it is found in a wide variety of foods, mainly in fruits and vegetables. Ascorbic acid is also a natural antioxidant used in foodstuff formulations in order to prevent browning, discoloring and also to enhance shelf-life.[22] It is know to be thermolabile and its degradation depends on several factors such as temperature, water content, pH, storage time and metal ions.[15] It has been indicated by several authors[14] that one of the possible indicators for the effectiveness of the process is ascorbic acid degradation. It is generally observed that, if vitamin C is well retained, the other nutrients are also well retained.[14] Thus, the present work pays especial attention to the degradation undergone by the vitamin C during the processes studied.

The curves characterizing ascorbic acid loss, for fresh and pretreated samples, can be seen in Fig. 4aand 4b, respectively. The initial ascorbic acid content of the fruits used in the experiences varied between 21 and 26 mg/100 g sample (Table 1). Figure 4also shows the agreement between experimental data and predicted values using Eq. (3). The values for k ₁and r²obtained are also given in Table 1. The proposed model fitted the experimental data with r²values above 0.80. This confirms that the ascorbic acid degradation is a first order reaction. It can be seen that according to reaction rate constant (k ₁), vitamin C loss increased as the temperature was raised. Indeed, this nutrient is highly sensitive to heat. Actually, the retention of vitamin C was found to be maximal when the dry temperature was fixed between 40 and 50°C, while a greater degradation of the nutrient was observed during the drying at 70°C. Similar results were reported by Ramallo and Mascheroni[15] in the case of the pineapple dehydration.

Figure 4Effect of the temperature on the ascorbic acid loss in cactus pear. (a) Fresh samples; and (b) pretreated samples.

The coefficients of the polynomial model [Eq. (6)] corresponding to the statistical analysis are also shown in Table 2. From these results, it is possible to notice that k ₁was affected mainly by the temperature (T) and by the T* Trinteraction. The pretreatment (Tr) did not significantly affected the degradation constant (k ₁).

Experimental and predicted rate constants [with Eqs. (12, 13); R² = 0.875) for both, fresh and pretreated samples, were plotted versus the drying temperature in Fig. 5.

(12)

(13)

Figure 5Variation of the constant k ₁with the air-drying temperature.

It can be noted that the magnitude of k ₁drastically increased as the temperature was raised. Eq. (12)applied to the F samples permits to estimate a degradation level about 2 times grater at 70°C than that obtained at 40°C. Furthermore, this difference became much important in the case of the PT samples: the values of k ₁, obtained with Eq. (13), were at least 6 times larger for 70°C than those corresponding to 40°C. The pretreatment, evidently, was not favorable to the retention of the vitamin at high drying temperature (60–70°C). However, the lowest k ₁-values, corresponding to the best vitamin C retention, were obtained in pretreated samples dried at 40 and 50 °C.

The activation energy necessary for the vitamin C degradation was calculated using Eq. (14)and (15):

(14)

(15)

and the values obtained were 22.57 kJ/mol (F samples) and 51.51 kJ/mol (PT samples). The difference in activation energy is probably due to the sucrose coat formed on the surface of the pretreated slices. This range of variation in activation energy is similar to that reported for high pressurized and thermally pasteurized orange juice[13] and during storage study of citrus juice concentrates.[23]

Color Change

The change in color during thermal processing of foods may occur by any one or more of the following mechanisms: pigments degradation, ascorbic acid oxidation, browning, non-enzymatic reaction, the Maillard reaction, or combinations of the above mechanisms.[24]

The color of natural cactus pear can be described as green-clear. The L_o *, a_o *, b_o * values were 60.43 ± 2.76, −4.73 ± 1.23, and 19.67 ± 2.63, respectively. In this study, it was observed that all the three parameters were found to determine changes. As the drying progressed, L* decreased, while a* and b* increased (results are not shown). This behavior was also observed by Mayor and Sereno[25] during osmotic dehydration of pumpkin. The mechanisms of these changes are not clear, they can be due to enzymatic and non-enzimatic browning reactions. The decrease in L*and increase in *aand b*values was attributed to occurrence of Maillard reactions during drying of banana.[26] The decrease of L* ties with the browning of foods as it has been reported by several authors.[27] Concerning the increase in a*and b*; this caused the samples to become more yellow-red.

Preliminary results showed that L*, a*, and b* values can be used to monitor the color change in cactus pear, the graphs describing the overall color change are shown in Figure 6a(F samples) and 6b (PT samples). Only the mean values were plotted. The simulated curves obtained with a first-order kinetic model represented by Eq. (5)are also given in Fig. 6. The values of the constants k ₂and ΔE_max, as well as those obtained for r² through the experimental data adjustment, are shown in Table 1. A statistical analysis based on the total color change (ΔE) was carried out. The procedure and the model [Eq. (6)] were the same of those used for the D_w and k ₁analysis. In Table 2, it can be observed that there was a significant linear and quadratic effect of the temperature on the values of k ₂. Minima values of ΔE are observed at 40°C, while the maxima ones occurred at 70°C.

Figure 6Effect of the temperature on the total color change of cactus pear. (a) Fresh samples; (b) pretreated samples.

To predict the k ₂variation, the following models were used (both with R² = 0.709):

(16)

(17)

The behavior predicted by models (Eqs. 16, 17)in Figure 7shown that the magnitude of constant color degradation should increase proportionally to the drying temperature. The pretreatment did not influence k ₂ in a significant way, even though the values of this constant tended to be greater in the case of the PT samples. This difference seems to be a result of solids uptake by the samples during the pretreatment.

(18)

(19)

Figure 1Variation of the constant k ₂with the air-drying temperature.

The activation energy needed to color change in the cactus pear was calculated using Eq. (18)and (19); the values obtained were 12.93 kJ/mol for the F samples, and 22.57 kJ/mol for the PT samples. These values are very close to those found in the assessment of the color change in peach and plums subjected to thermal treatment over a 56–80°C temperature range.[16]

CONCLUSIONS

The drying kinetics was efficiently described by Fick's second law of diffusion. During the dehydration, only the falling rate period was present. This confirmed that diffusivity controls the movement of moisture into the cactus pear. The water effective diffusivity (D_w ) varied from 1.51 × 10⁻¹⁰to 5.32 × 10⁻¹⁰m²/s as a function of the temperature range (40–70°C); both the polynomial model and the Arrhenius model satisfactorily described this variation. The vitamin C loss and the color change, at all four drying temperatures, behaved as a first-order reaction, and the degradation constants k₁ and k₂ were affected mainly by the temperature. According to the constant k₁ , the best vitamin C retention was obtained in the pretreated samples dried at 40 and 50°C, while k₂ indicated that the osmotic pretreatment brings about the greater color change in the samples during drying.

ACKNOWLEDGMENTS

The authors wish to acknowledge the financial support of the Fundación Produce San Luis Potosí, A.C. (Project Research 2004), and we would also like to thank O. Beltrán-Guevara for the revision of the English version.