Sample preparation

Mexican plum ecotypes (‘beetroot’ and ‘yellow”) (from the Coscomatepec, Veracruz, Mexico production area( were chosen for this study. Fruits were selected according to the color of the peel when they were cut: ‘beetroot’ ecotype showed 30 % green and 70 % red, and ‘yellow’ ecotype showed 75 % yellow. The plums were washed and cut with a stainless-steel cutter into 2.5 cm wide and 0.5 cm thick slices. The slices were cut parallel to the axial axis of the fruit. All tests were carried out in triplicate.

Freeze-drying conditions

The slices were frozen (Tc) at (20 °C, during 180 min, in an ultra-freezer (fast freezing) and -5 °C, during 2,880 min (slow freezing), in a domestic freezer (DAEWOO Mod FR-62R, Korea). The frozen samples were dehydrated during 360 min, in a laboratory freeze dryer (LABCONCO 2.5 L-Mod 742002, USA), with space for 12 flasks, which are attached to the valves of the drying chamber with stainless steel adapters and a condensed moisture collector. The drying kinetics were carried out with 0.05 and 0.1 mbar vacuum pressure (Pr) at -85 °C in the condenser. The samples were removed at specific periods, in order to evaluate moisture, water activity, and color changes.

Physical and chemical determinations

The moisture content was determined in 1 g of sample, in a Moisture Analyzer with Halogen Lamp (MB35 HALOGEN, OHAUS, Switzerland), at 65 °C; the results were expressed in terms of dry basis (g H₂O g s s^-1). The water activity (a_w) of the plum slices during freeze-drying was determined at 25±1 °C using a water activity meter (Aqualab, series 3 TE, DECAGON; Washington, USA).

The optical properties of fresh and lyophilized plums were analyzed based on their reflectance, using a colorimeter (HunterLab, model MiniScan XE plus; Hunter Associates Laboratory Retson, VA, USA). The colorimeter was calibrated using black and white standards. Parameters L^* (luminosity) and chromaticity a^* (green-red) and b^* (blue-yellow) were measured using the Hunter scale. Based on these parameters, Hue (hue angle), Chroma (color saturation), and ΔE (total color difference) were calculated. According to ^{Guiné and Barroca (2012)}, the mathematical expressions for these parameters are as follows:

where: L^* , a^* , b^* are the values of the lyophilized plum sample in different times, and L0*,a0*,b0* are the color parameters of the fresh sample.

Degree of rehydration

Between 0.5 and 0.6 g of dehydrated product were submerged in 150 mL of distilled water, at 10 and 30 °C, for 60 min. Sample rehydration was evaluated at 2, 4, 6, 10, 15, 20, 25, 30, 35, 40, 50, and 60 min (^{Reyes et al., 2008}), recording their weight, after removing the liquid from the surface with absorbent paper. Rehydration was also evaluated in whole milk, taking into account that dehydrated plums may be mixed with cereals. Whole and pasteurized milk were obtained in a local supermarket. The equation used to calculate rehydration was:

where: m1 is the dry sample mass (g) and m2 is the rehydrated slice mass (g) (^{Zhang et al., 2015}).

Pectin content

Pectin was extracted from the plum slices using the method described by ^{Kliemann et al. (2009)}, with some modifications. One-hundred mL of distilled water were added to 5±0.01 g of sample crushed in mortar; citric acid was added to achieve a 2.5 pH; finally, the sample was heated for 1 h at 90 °C, with continuous stirring. The heated extract was filtered through a cheesecloth. An equal volume of ethanol was added to the filtered sample and let stand for 2 h. The flocculated pectin was separated by filtration and was dehydrated in an oven, until its constant weight was obtained. The pectin yield was obtained using the following equation:

where: Y_pec is the yield of pectin extract (in percentage), P is the amount of pectin extract in g, and B_i is the initial amount of sample in g.

Experimental design and statistical analysis

The experimental design was factorial, with two factors and two levels: freezing temperature (-5 and -20 °C) and pressure (0.05 and 0.1 mbar); it was completely random and included three repetitions. The experimental unit was a sample of each plum ecotype. The results are presented as means of the variables and their standard deviation. The results were analyzed using ANOVA and were considered significant when p≤0.05. All the analysis were carried out with Minitab Realese 6.

Moisture change curves

The initial moisture of beetroot and yellow ecotypes was 78 and 85 %, respectively. The drying curves showed that freeze-drying pressure had a significant effect (p≤0.05) in both ecotypes. At the end of the process, moisture was lower or equal to 0.0206 g H₂O g s s^-1, with 0.1 mbar. Moisture values were statistically lower than 0.05 mbar (Figure 1).

Figure 1 Moisture change in plum slices (Spondias purpurea L.): A) beetroot and B) yellow, during freeze-drying. 

The slices of beetroot plums frozen at -5 °C and lyophilized with 0.05 and 0.1 mbar showed the greatest loss of moisture, at 60 and 120 minutes, respectively. In slices frozen at -20 °C and lyophilized at 0.1 and 0.05 mbar, the moisture decreased in the first 60 min and 180 min, respectively; nevertheless, the moisture in all treatments stabilized at 300 min (Figure 1A).

Moisture decreased in yellow plum (Figure 1B) after 60 min, under all conditions evaluated; however, with 0.1 mbar and freezing at -5 °C, moisture stabilized at 240 min; finally, at 300 min, it stabilized in the other treatments. In all the treatments, there were two acceleration and deceleration periods of moisture loss. ^{Pei et al. (2014)} noted that the acceleration period may be attributed to the greater sublimation area during the first stage of drying, and that it is gradually reduced as the process continues, because it increases the mass-transfer resistance. The effect of freezing temperature in the final moisture of the slices was not significant. However, slow freezing caused rapid sublimation of yellow plums, probably because the ice crystals were larger compared to fast freezing. These results are similar to those reported by ^{Reyes et al. (2011)} and ^{Ceballos et al. (2012)} in apple and soursop slices, respectively. These authors found that small pores generated narrow roads with greater resistance to water flow.

Water activity change curves

The a_w of the beetroot plum slices (Figure 2A), lyophilized with both pressures, decreased significantly from 180 min, and was lower when the tissue was frozen at -20 °C. In contrast, the aw of the yellow plum decreased (Dunnett’s test p=0.05) after 120 min, with pressures and freezing at -5 ° C; also, with freezing at -20 °C, the decrease started from 60 and 180 min, with the lower and higher pressure, respectively (Figure 2B).

Figure 2 Water activity of plum slices (Spondias purpurea L.): A) beetroot and B) yellow, during freeze-drying. 

All plum slices were lower than 0.5 a_w from 300 min. This indicated that microbial growth is unlikely. The a_w balance values in lyophilized beetroot plum slices ranged between 0.275 and 0.381, and those of yellow plum, between 0.249 and 0.275. In this way, a_w balance decreased significantly as the freezing temperature fell for both plum types (p=0.05). In the case of drugs, during slow freezing, as freezing advances, the solute concentrates and it forms a glassy (almost waterproof) layer on the material’s surface; this obstructs the flow of steam water through the pores (^{Kasper and Friess, 2011}). ^{Marques et al. (2007)} observed that the decrease of a_w in the freeze-drying of acerola depended on the freezing process, which produces different porous structures.

Analysis of color

Beetroot fresh plum showed 16.24 L^* and this value did not change in lyophilized samples, after freezing at -5 °C. In contrast, freezing at -20 °C increased luminosity (Table 1). L^* is defined as the flux of light in a given direction, and depends on the number and orientation of the surfaces that reflect light. Therefore, the freezing speed affects the luminosity of lyophilized samples. Fast freezing generates small randomly-positioned crystals; therefore, the luminosity of quickly frozen products is higher than that of slowly frozen products (^{Ceballos et al., 2012}).

The Chroma and Hue angle values of lyophilized beetroot plum did not show significant differences regarding fresh slices. This indicates that anthocyanins -pigments related to the red color of fruits- are stable. ^{Beaudry et al. (2004)} and ^{Franceschinis et al. (2015)} lyophilized cranberries and cherries, and observed that those berries still possessed an intense red shade, because their anthocyanins were preserved, while thermal treatments brown them and dull their color. The overall color difference (ΔE) in the slices was significant (p≤0.05) with regard to the beetroot plum unprocessed slices (Figure 3). Since the freeze-drying increased the luminosity of the sample, the value of ΔE was affected.

Figure 3 Color of beetroot plum (Spondias purpurea L.) slices: A) fresh and B) lyophilized. 

L^* of lyophilized yellow plum increased significantly in comparison with fresh fruit; lower freezing temperature produced a greater increase (Table 2). These results are similar to those obtained with carrots and asparagus (^{Reyes et al., 2008}; ^{Jokic et al., 2009}).

Chroma increased with freeze-drying. This shows that carotenoids were concentrated in the tissue. Results are similar to those observed in mashed mango, in which β-carotenes increased with freezedrying, compared to samples dried with convective heat transfer (^{Caparino et al., 2012}). The initial Hue angle value in yellow plum was 101.41°; this is found in the yellow/green region (90-180° Hue). However, the shade of the slices decreased (Hue angle (90°). This was due to the negative-to-positive variation of the a^* parameter, possibly as a result of the enzymatic browning caused by mechanical damage when the fruit was cut.

ΔE showed significant differences (p<0.05) between yellow plum treatments. This was due to the high luminosity of the samples and a slight shift of Hue towards the red shades (Figure 4).

Figure 4 Color of yellow plum (Spondias purpurea L.) slices: A) fresh and B) lyophilized. 

Pectin content

There were no significant differences (p>0.05) in pectin content (y_pec) between treatments: the freezing temperature or vacuum pressure did not affect the polysaccharide content of the samples (Table 3).

High pressures can compress the raw material and modify its surface; as a result, the following actions take place: the chemicals reactions are balanced, while the solvent penetration is accelerated, and its diffusion in the food is sped up. The results were similar to those reported by ^{Guo et al. (2012)} and ^{Naghshineh et al. (2013)} who applied high vacuum to lime and orange peels. The use of high vacuum to obtain ingredients from medicinal herbs and the food matrix increases the extraction of bioactive compounds (^{Xi, 2009}; ^{Huang et al., 2013}).

Rehydration of lyophilized Mexican plum

The rehydration ratio (RR) measures the structural damage and the reconstitution properties of fruits and vegetables that have undergone a drying process. The RR of lyophilized beetroot plum was not affected by the conditions of the process (Table 4).

Beetroot plum slices kept the structure required to recover their original size during rehydration. The rehydrating medium seems to have spread properly in the uniform and porous structure, as well as in the cell walls, without damaging the slice. The RR was greater in water than in milk, probably as a result of the higher viscosity of the latter, although it can be a source of solids whose nutritional importance can be transported to the rehydrated product (^{Marín et al., 2006}). However, the RR values in milk (2.42-2.94 g g^-1) of the beetroot plum slices were greater than those reported by ^{Shih et al. (2008)} in strawberry slices (less than 2 g g^-1).

The lyophilized slices of yellow plum shed off material shortly after rehydration starts. In the samples that were frozen at -5 °C, the rehydration took 6 min, and in the ones frozen at -20 °C, 15 min. This difference is the result of long-term freezing, which creates bigger pores and weakens the structure. These results are similar to those obtained in the rehydration of lyophilized avocado (^{Arriola-Guevara et al., 2006}). The freezing temperature had a significant effect on RR in water at 30 °C, and freezing at -5 °C increased the value of RR. Similarly, the RR of the yellow plum was lower in milk (less than 3 g g^-1). In the case of fruits combined with cereals or used as snacks, RR should be low.