- Email: [email protected]
Micronutrient flow to the osmotic solution during grapefruit osmotic dehydration
Journal of Food Engineering 74 (2006) 299–307 www.elsevier.com/locate/jfoodeng
Micronutrient flow to the osmotic solution during grapefruit osmotic dehydration R. Peiro´ a, V.M.C. Dias b, M.M. Camacho a, N. Martı´nez-Navarrete
a,*
a
b
Department of Food Technology, Universidad Polite´cnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain Escola Superior de Biotecnologia, Universidade Cato´lica Portuguesa, Rua Dr. Anto´nio Bernardino de Almeida, 4200-072 Porto, Portugal Received 27 July 2004; accepted 7 March 2005 Available online 26 April 2005
Abstract The purpose of this work was to quantify the flow of soluble micronutrients, such as acids, minerals and pectins, from the grapefruit to the osmotic solution (OS) used to dehydrate the fruit when recycling it in successive osmotic dehydration (OD) operations, without reconcentrating. OD was carried out for 3 h at 30 °C with an OS:fruit rate 5:1, using a 55 °Brix sucrose solution. Soluble solids (°Brix), water activity (aw), pH, ascorbic (AA), citric acid (CA), galacturonic acid and major minerals were measured in the fruit and in the OS as a function of the number of uses (up to 8). Electrical conductivity (EC) and viscosity (l) of the OS were also analysed. The characteristics of the obtained dehydrated grapefruit and the observed recovery of the quantified micronutrient loss by the fruit in the OS allows us to propose the reuse of the OS as a good way of contributing to the economic and ambient profitability of the OD operation. Moreover, its final use as ingredient in new food formulations may be proposed. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Osmotic solution; Ascorbic acid; Citric acid; Minerals; Pectin; Electrical conductivity; Viscosity
1. Introduction Fruits are an important part of the human diet as, besides contributing with some nutrients, they contain a series of non-nutritive substances, called phytochemical. These bioactive compounds, although having neither a classically defined nutritional function nor being considered essential for the human health, can have a significant impact on prevention of some diseases. Phytochemicals have multiple biological effects, including antioxidant activity (Liu et al., 2002; Prior & Guohua, 2000; Wolfe & Liu, 2003), antimutagenic (Wargovich, 2000) antibacterial and angioprotective properties (Venant, Borrel, Mallet, & Van Neste, 1989). This group of compounds mainly includes phenolic compounds,
*
Corresponding author. Tel.: +34 96 387 9362; fax: +34 96 387 7956. E-mail address: [email protected] (N. Martı´nez-Navarrete).
0260-8774/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jfoodeng.2005.03.022
lignans, monoterpenes, etc. Their intake has been associated with a reduced risk of coronary heart disease and strokes (Hertog, Hollman, & Van de Putte, 1993). On the other hand, also micronutrients present in fruits, such as vitamins, acids, fiber, etc. also contribute to their well known beneficial properties. Nevertheless, in general, fruits are products with a relatively short post-harvest life-span and recent food habits (easy and quick to eat food) have promoted a decrease in the consumption of fresh fruit, which has been replaced by juices, dairy products with added fresh or processed fruits, preserves, confectioneries, etc. (Torreggiani & Bertolo, 2001). To this end, many technologies have been used to make the food system formulation more suitable. One of the possible techniques for fruit processing is osmotic dehydration (OD) with sugar solutions, working at middle temperatures to preserve product flavour and other sensory properties.
300
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
During osmotic dehydration of fruit in an osmotic solution (OS), a two-way mass transfer is established: water and some natural soluble substances (sugars, vitamins, pigments, organic acids, mineral salts, etc.) flow out of the fruit into the OS, and in the opposite direction, soluble solutes may be transferred from the solution to the fruit. Due to the kinetics of the operation it may be used to obtain products of reduced but still relatively high moisture content, classified as intermediate moisture products. The operation may be used as an intermediate step in other processes such as drying (Fito et al., 2001) and freezing (Chiralt et al., 2001) or to produce minimally processed fruit (Martı´nez-Monzo´, Martı´nez-Navarrete, Chiralt, & Fito, 2001). Moreover, the application of a vacuum pulse for a short period of time at the beginning of the process (pulsed vacuum osmotic dehydration, PVOD) may have beneficial effects on the transfer kinetics leading to shorter processing times (Fito & Chiralt, 2000). Mass transfer rates during OD operations depend on factors such as temperature, type and concentration of osmotic medium, type, size and geometry of the sample, sample to solution ratio and degree of agitation of the solution (Taiwo, Angersbach, Ade-Omowaye, & Knorr, 2001). This method has received considerable attention due to the low energy requirements (Taiwo et al., 2001) and fruit quality improvement (Heng, Guilbert, & Cup, 1990; Panagiotou, Karathanos, & Maroulis, 1998), compared to alternative process schemes. As fruits are not submitted to high temperatures, sensory attribute changes, such as colour, aroma flavour and texture are minimised (Fito, Andre´s, Pastor, & Chiralt, 1995; Heng et al., 1990; Raoult-Wack, 1994). Nevertheless, a loss of vitamins, polysaccharides and minerals, that flow from the fruit to the osmotic solution, has been observed (Co´rdoba, Garcı´a-Martı´nez, Martı´nezNavarrete, Camacho, & Martı´nez-Monzo´, 2003; Garcı´a-Martı´nez, Martı´nez-Monzo´, Camacho, & Martı´nez-Navarrete, 2002). Moreover, in comparison with other traditional drying treatments, OD does not affect the food structure because water elimination does not involve phase changes (Forni et al., 1987; Giangiacomo, Torreggiani, Erba, & Messina, 1994; Pinnavaia, DallaRosa, & Lerici, 1988). Despite the above mentioned advantages, one limitation for large scale production using the fruit OD process is the management of the osmotic solution, commonly sucrose syrup, which, if not used, has to be processed as waste thus increasing the cost of the osmodehydrated products. Different alternatives may be pointed out to solve this problem. One is to submit the used OS to a reconcentration step. Nevertheless, the installation and use of this stage will also increase the cost of the dehydration operation. Another possibility is to reuse or recycle it in successive dehydration cycles without any treatment, which will be limited by its
dilution related to the dehydration level of the obtained fruit and also to microbiological aspects. Finally, it is possible to use the OS as an ingredient to be added in some product formulations. To this end, and taking into account that, as commented on above, during fruit OD a loss of valuable natural substances such as vitamins, acids, minerals, etc. occurs, a previous reuse of the OS for some cycles without any reconcentration treatment, will be desirable. The objective of this work was to study some compositional (water, soluble solids, ascorbic acid, citric acid, major minerals and pectin) and physicochemical (water activity, pH, electrical conductivity and viscosity) changes associated with osmodehydration of grapefruit, both in the fruit and in the sucrose solution used as the osmotic agent.
2. Materials and methods 2.1. Raw matter Fresh grapefruit (Citrus paradise) of red coloured variety (Star Ruby), purchased in local markets in Valencia (Spain), were used and stored under chilling conditions (8 °C) till the moment of use (maximum 24 h). They were selected on the basis of a similar ripening degree (between 11 and 13 °Brix) and uniform visual quality. The fruits were manually peeled and cut, perpendicularly to the fruit axis, into uniform half-slices with a knife (1 cm thickness). The osmotic solution was prepared using commercial sugar mixed with heated (30 °C) distilled water until total dissolution to form a 55 °Brix syrup. 2.2. Osmotic dehydration process The grapefruit samples were immersed in a plastic tank filled with 55 °Brix syrup for 3 h at 30 °C. For the first 10 min a vacuum pulse (50 mbar) was applied. Afterwards, osmodehydration was conducted at atmospheric pressure in a temperature-controlled water bath, where the syrup was continually stirred with a Heidolph RZR2102 stirrer at 200 rpm. A plastic screen was placed on the basket to keep the fruit pieces totally immersed in OS and separated from stirrer. This dehydration time was chosen on the basis of the results obtained in previous osmotic dehydration kinetics studies to obtain grapefruit with about 22 °Brix (Moraga, 2004). The syrup to fruit ratio was 5:1. Successive OD cycles, up to a maximum of eight, were carried out using the same OS for all the cycles, without any treatment, and renewing the fruit after each OD cycle. Three days were used to complete the experience (two or three OD cycles by day) and OS was kept at 8 °C from one day to another.
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
2.3. Analysis OS was analysed after 0, 1, 2, 3, 4, 6, and 8 OD cycles as to soluble solids (°Brix) at 20 °C (ATAGO NAR-3T refractometer, Japan), water activity (aw; Decagon model CX-2 dew point hygrometer), pH (Crison micro pH 2001 pH-meter), electrical conductivity (K; Crison 522 conductimeter), ascorbic acid (AA) (AOAC 985.33, 1997) and citric acid (CA) (AOAC 985.33, 1997). To calculate citric acid content, AA presence was taken into account (Eq. (1)), where m, v and M are the mass (g), valence and molecular weight of CA and AA and V and N the volume and normality of the NaOH used for titration. V NaOH N NaOH ¼
mAC mAA þ M AC M AA vAC vAA
301
fuged (Selecta model S40) for 10 min at 12 000 rpm, and the floating liquid was used to follow the above mentioned AOAC method. xw was determined according to the AOAC method 950.46 (1997). All other analyses were measured following the same methods and equipment described for OS analysis. Statistical analyses to obtain means, standard deviations of means, analysis of variance (ANOVA), including the study of homogeneous groups established throughout the Least Significant Difference (LSD) test, and stepwise multiple regression analysis were performed using the Statgraphics Plus 4.0 program.
3. Results and discussion ð1Þ
Besides viscosity (Physica Rheolab MC1 rheometer at 25 °C, shear rate sweep from 0 to 150 s1 in 90 s), pectin content, and major mineral content of OS were also determined, although these were only analysed in cycles 1, 3, 5 and 8. Galacturonic acid quantification was the method used to determine the amount of pectin present in the samples (Kitner & Van Buren, 1982). To this end, samples were ethanol extracted by centrifugation and the precipitate that contains the alcohol-insoluble solids was extracted with sulphuric acid and distilled water (Ahmed & Labavitch, 1977). Determination of pectin content of the extract was carried out by the m-hydroxydiphenyl method (Kitner & Van Buren, 1982) reading the absorbance of the samples at 520 nm (Cecil 1020 spectrophotometer). Minerals were analysed according to the method described by De la Fuente, Montes, Guerrero, and Jua´rez (2003), consisting in a mineralisation of samples by dry ashing in a muffle furnace and a posterior dilution of the ashes with concentrated hydrochloric acid and distilled water. Calcium and magnesium were determined by atomic absorption spectrometry using a multi-element (Ca–Mg–Zn) hollow-cathode lamp. Potassium and sodium were analysed by atomic emission spectrometry. A 3110 atomic absorption spectrometer (Perkin-Elmer) in an air-acetylene flame was used for the analysis. Phosphorus was measured using a UV–visible spectrophotometer, Cecil 1020. Three mineralisation of each sample were carried out and each one was analysed in triplicate. Fresh grapefruit used for each OD cycle as well as the dehydrated one were analysed as to water content (xw), °Brix, aw, pH, ascorbic acid, citric acid, pectin content and the same mineral content described for OS. In all cases, fruit samples were previously homogenised with an Ultra-Turrax T25. For the ascorbic acid content determination, the obtained pulp was previously centri-
3.1. Water content, °Brix and water activity Table 1 shows xw, °Brix and aw of fresh grapefruit used for each OD cycle and of grapefruit obtained after each dehydration process. The ANOVA carried out with data of fresh fruit used for each OD cycle showed significant differences (a < 0.05) in xw and °Brix, which ranged between 85.2–87.1 g water/100 g fruit and 11.9– 13.5 g soluble solids/100 g fruit. Nevertheless, this compositional variability, that seems to be normal in this kind of raw material, was not reflected by significant differences in the aw of pieces (a > 0.05), which oscillated between 0.989 and 0.992. After each dehydration cycle, under the conditions pointed out in the methodology, Table 1 °Brix, water activity (aw) and water content (xw) (g water/100 g sample) of fresh (FG) and osmodehydrated (ODG) grapefruit used and obtained for each osmotic dehydration cycle (1–8) Sample
°Brix
aw
xw
FG1 ODG1
11.9 ± 0.1 23 ± 0
0.9915 ± 0.0007 0.981 ± 0.005
86.4 ± 0.1 75.5 ± 0.7
FG2 ODG2
12.2 ± 0.1 23.1 ± 0.1
0.9890 ± 0.0014 0.9785 ± 0.0007
85.2 ± 0.1 74.20 ± 0.02
FG3 ODG3
11.9 ± 0.1 24.3 ± 0.1
0.989 ± 0.003 0.976 ± 0.002
86.58 ± 0.05 75.8 ± 0.6
FG4 ODG4
13.25 ± 0.07 21.4 ± 0.1
0.9890 ± 0.0014 0.9855 ± 0.0007
86.440 ± 0.002 78.1 ± 0.1
FG5 ODG5
13.5 ± 0.1 22.6 ± 0.1
–
–
FG6 ODG6
12.4 ± 0.1 20 ± 0
0.9885 ± 0.0007 0.983 ± 0.002
FG7 ODG7
12.3 ± 0.1 22.5 ± 0.1
–
FG8 ODG8
12.7 ± 0.1 21.7 ± 0.1
0.989 ± 0.002 0.9800 ± 0.0014
86.8 ± 0.1 78 ± 1
Mean FG Mean ODG
12.5 ± 0.6 22.3 ± 1.3
0.9895 ± 0.0012 0.981 ± 0.003
86.4 ± 0.6 77 ± 2
Mean values also appear in the last row.
87.09 ± 0.05 80.1 ± 0.2 –
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
60
0.98
50
0.97
40
0.96
30
0.95
tively. Nevertheless, aw was not significantly affected by the reuse of OS. From this point of view, it can be considered possible to reuse the OS for eight cycles, under the conditions of this study, to obtain grapefruit with the same stability (aw). This is important in order to make the OD process more profitable, as the reuse of the OS without any reconcentration treatment is possible maintaining the same OD time.
aw
ºBrix
302
0.94
20 ºBrix aw Predicted aw
10 0 0
1
2
3
4 5 Cycles
6
0.93 0.92 7
3.2. Citric acid, ascorbic acid and pH
8
Fig. 1. Experimental mean values and standard deviation of °Brix and aw of the osmotic solution when reused for different osmotic dehydration cycles. Dotted lines reproduce the model fitted to experimental data (Table 5). Continuous line corresponds to aw values predicted by Norrish equation.
samples reached approximately the expected value of 22 °Brix (Moraga, 2004). Fig. 1 shows the decrease in soluble solid content and the increase in water activity observed in the OS after each OD cycle. As expected, a progressive syrup dilution was observed due to the water transferred from grapefruit to the solution. After the 8th OD cycle studied, the solution became diluted from an initial value 55.1 ± 0.1 to 41.9 ± 0.1 °Brix. Consequently aw increased from 0.920 ± 0.005 to 0.960 ± 0.005. Water activity of OS was compared with aw predicted for sucrose solutions with the same °Brix reached by OS after each OD cycle. To this end, Norrish equation (Eq. (2); Martı´nez-Navarrete, Andre´s, Chiralt, & Fito, 2000) was used, assuming that sucrose was the only soluble solute present in the osmotic solution. aw ¼ xw expð6.47x2s Þ
ð2Þ
where xw and xs are the molar fraction of water and soluble solids (°Brix), respectively. Fig. 1 shows the obtained predicted values. As can be observed, experimental values and predicted curve differed more with cycles. The lowest values of predicted aw could be related with the contribution of other solutes, different from sugars, to the °Brix of the OS and with a weaker interaction with water. As concerns to dehydrated fruit (Table 1), if the progressive dilution of OS observed when reusing it had a significant effect on the dehydration kinetics, a different dehydration level in the fruit processed in each OD cycle could be expected. The statistic study carried out using an ANOVA, taking the cycle factor into account, confirmed significant differences (a < 0.05) in xw and °Brix but only allows us to consider two homogeneous groups of samples: one corresponding to cycles 1, 2, 3, with mean values of xw and °Brix of 75.1 and 23.6, respectively, and another one for the rest of the samples with mean values of xw and °Brix of 78.7 and 21.6, respec-
Table 2 shows citric, ascorbic and galacturonic acid of different fresh and dehydrated grapefruit pieces used and obtained after each dehydration process. A certain variability in these parameters was observed in the fresh fruit again, with citric and ascorbic acid content ranging between 941–1167 mgcitric acid/100 gfruit and 10–14.5 mgascorbic acid/100 gfruit, respectively. pH of fresh
Table 2 Citric acid (CA), ascorbic acid (AA) and galacturonic acid (AGU) of fresh (FG) and osmodehydrated (ODG) grapefruit used and obtained for each osmotic dehydration cycle (1–8) and percentage of loss of each component Sample
Citric acid
FG1a ODG1a ODG1b Loss1c
1035 ± 5 757 ± 14 865 ± 16 27 ± 1
12 ± 2 7.4 ± 0.2 8.5 ± 0.2 6±2
4453 ± 192 3756 ± 164 4870 ± 164 16 ± 4
FG2a ODG2a ODG2b Loss2c
1050 ± 11 834 ± 21 929 ± 23 21 ± 2
14.5 ± 0.2 9.8 ± 0.2 10.9 ± 0.2 6±1
5267 ± 41 4130 ± 196 5253 ± 196 22 ± 4
FG3a ODG3a ODG3b Loss3c
941 ± 8 642 ± 3 728 ± 3 31.7 ± 0.4
14 ± 1 9±1 10 ± 1 9±1
4558 ± 149 3706 ± 206 4656 ± 206 19 ± 3
FG4a ODG4a ODG4b Loss4c
1167 ± 5 830.5 ± 0.3 947.2 ± 0.3 29 ± 0.3
14 ± 1 7.4 ± 0.3 8.4 ± 0.3 11 ± 2
4667 ± 173 4115 ± 376 5032 ± 376 12 ± 3
FG6a ODG6a ODG6b Loss6c
1032 ± 11 846 ± 28 938 ± 31 18 ± 2
10 ± 1 7±1 8±1 12 ± 3
4207 ± 11 2964 ± 191 4236 ± 191 30 ± 3
FG8a ODG8a ODG8b Loss8c
986 ± 4 780 ± 9 917 ± 10 21.4 ± 0.2
10 ± 0 6.1 ± 0.2 7.2 ± 0.2 25 ± 1
4086 ± 21 3559 ± 243 5120 ± 243 13 ± 4
1035 ± 76 782 ± 76 887 ± 83 25 ± 5
11 ± 4 8±1 9±1 12 ± 7
4540 ± 417 3705 ± 429 4861 ± 429 18 ± 7
Mean Mean Mean Mean
FGa ODGa ODGb Lossc
Ascorbic acid
Mean values also appear in the last row. a mg component/100 g fresh fruit. b mg component/100 g osmodehydrated fruit. c mg component loss/100 mg initial component.
AGU
500
5
CA AGU AA
400
4
300
3
200
2
100
1
0
mg AA/ 100 g OS
mg CA or AGU/ 100 g OS
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
0 0
1
2
3
4 5 Cycles
6
7
8
9
Fig. 2. Mean values and standard deviation of ascorbic acid (AA), citric acid (CA) and galacturonic acid (AGU) of the osmotic solution when reused for different osmotic dehydration cycles.
grapefruit used in this study varied between 2.94 and 3.34, being 3.19 ± 0.16 its mean value. A loss of acids was detected in grapefruit due to dehydration process. This was reflected in lower acid content in the dehydrated fruit, although its pH (3.16 ± 0.05) was not significant different from pH of fresh fruit. Consequently, a progressive AA and CA enrichment of OS was observed (Fig. 2), changing from 1.30 ± 0.04 in cycle 1 up to 4.33 ± 0.04 mg AA/100 g OS in cycle 8 and from 72 ± 4 to 414 ± 8 mg CA/100 g OS. On the other hand, the greatest pH variation in the syrup took place during the first usage, changing from 6.7 ± 0.2 to 3.50 ± 0.01. By the third use the syrup reached the same pH as the fruit and remained at about the same value (pH = 3.2) for all subsequent cycles. As the most abundant acid in grapefruit is citric, which is a weak acid, the observed change in its concentration does not affect the pH. Similar behaviour has been observed by ValdezFragoso, Welti-Chanes, and Giroux (1998) who studied changes in 60 °Brix OS when used successively in apple dehydration after a reconcentration process and also by Garcı´a-Martı´nez et al. (2002) when analysing changes of a 55 °Brix OS in successive kiwifruit dehydration cycles for three OS:fruit ratios. These authors point out the interest of the observed pH decrease that could contribute inhibiting the growth of microorganisms when reusing the OS in successive OD cycles. The presence of acids in the solution could also have a beneficial influence on the inhibition of enzymatic browning (Cheftel & Cheftel, 1986; Valdez-Fragoso et al., 1998). To quantify the loss of acids in the grapefruit due to the osmotic dehydration of the fruit, the amount of acids in dehydrated samples was also expressed by 100 g of fresh fruit (Table 2). An ANOVA was again carried out to study differences in the amount of acid loss due to the reuse of OS. Loss of AA showed no significant differences (a > 0.05) associated to this factor. In the case of CA, these differences were significant (a < 0.05) but as no clear tendencies were observed associated to the number of cycles, the variability may be attributed to small differences in controlled OD
303
variables from one cycle to another. In this way, a mean value of acid loss in the grapefruit during its osmotic dehydration may be calculated, these being 12 ± 7 mg of AA/100 mg of AA present in the fresh fruit and 25 ± 5 mg CA/100 mg of CA present in the fresh fruit. 3.3. Mineral content Major minerals of grapefruit (Mataix, 1998) were analysed before and after dehydration process (Table 3). As no significant differences (a < 0.05) were obtained for the different fresh fruit pieces used in each OD cycle, the mean values for each mineral also appear in Table 3, which agree with data reported by Mataix (1998): K (190 mg/100 g), Ca (14 mg/100 g), P (14 mg/100 g) and Mg (10 mg/100 g). As can be observed, and as is typical in the fruits, the potassium was the most abundant mineral, present in about 10 times as great a quantity as others such as magnesium or calcium. Sodium was present in a very low quantity. In all the cycles, a loss in minerals content was observed when OD fruit was analysed. Fig. 3 shows the minerals recovered in the OS over the course of the eight OD cycles and Table 3 shows fruit losses. In this case, results have again been referred to the corresponding fresh sample to compare differences. No significant differences (a > 0.05) were obtained in the mineral loss by the fruit due to the reuse of OS. The mean values of mineral content in the dehydrated fruit appear in Table 3, together with the percentage of each mineral lost (referred to the amount present in the fresh fruit) during OD. These values ranged from 28% to 59%. 3.4. Electrical conductivity of OS The electrical conductivity (EC) of the syrup depends on the sugar concentration, temperature, chemical composition of water used in its preparation and on the type and concentration of some soluble components that may be present coming from the fruit together with the water loss during the dehydration process. The distilled water used to prepare OS had an EC = 4.13 ± 0.02 lS/cm which changed to 12.38 ± 0.05 lS/cm when sucrose was added to obtain a 55 °Brix osmotic solution. Compositional changes of OS during the OD process promote an EC increase over the course of the OD cycles (Fig. 4). This increase in EC from the first use of the syrup could be attributed to the observed progressive enrichment of mineral salts and organic acids coming from the grapefruit. Nevertheless, the greater mobility of the system, due to the lower viscosity of the OS associated to its dilution, commented on later, could also contribute to this increase. In order to separate both effects, a sucrose solution with the °Brix of that obtained after each OD cycle was prepared and the EC was measured (Fig. 4). The obtained values were much lower than that corresponding to the OS used in grapefruit
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
304
Table 3 Mineral content of fresh (FG) and osmodehydrated (ODG) grapefruit used and obtained for each osmotic dehydration cycle (1–8) and percentage of loss of each mineral Sample
Ca
FG1a ODG1a ODG1b Loss1c
14.7 ± 0.4 10 ± 1 12 ± 2 29 ± 7
11.8 ± 0.3 8±2 9±2 33 ± 11
206 ± 4 66 ± 13 76 ± 15 68 ± 7
14.9 ± 0.2 7±1 8±1 51 ± 7
3.5 ± 0.2 2±1 2±1 52 ± 21
FG3a ODG3a ODG3b Loss3c
14.5 ± 0.3 11 ± 1 13 ± 1 23 ± 6
11.4 ± 0.2 9±1 10.42 ± 0.03 19 ± 4
194 ± 5 100 ± 24 114 ± 27 48 ± 13
14.63 ± 0.04 11 ± 1 12 ± 1 28 ± 9
4.3 ± 0.2 1.71 ± 0.02 1.94 ± 0.03 60 ± 1
FG5a ODG5a ODG5b Loss5c
14.7 ± 0.3 11 ± 1 12 ± 1 28 ± 4
12.1 ± 0.2 7±1 8±0 35 ± 9
190 ± 2 85 ± 25 96 ± 28 50 ± 7
14.7 ± 0.1 8.1 ± 0.2 9.2 ± 0.3 41 ± 2
3.80 ± 0.14 1.7 ± 0.2 1.9 ± 0.2 53 ± 3
FG8a ODG8a ODG8b Loss8c
14.3 ± 0.3 10 ± 1 12 ± 1 28 ± 7
12.8 ± 0.7 9.0 ± 0.5 10.6 ± 0.5 30 ± 1
195 ± 6 70 ± 2 82 ± 2 64 ± 1
14.62 ± 0.11 8.2 ± 0.3 9.6 ± 0.4 41 ± 3
4.0 ± 0.3 1.21 ± 0.03 1.42 ± 0.03 70 ± 1
14.5 ± 0.3 10.6 ± 0.4 12.1 ± 0.4 28 ± 5
12.0 ± 0.6 8±1 10 ± 1 29 ± 9
196 ± 7 80 ± 16 92 ± 17 57 ± 10
14.71 ± 0.13 9±2 10 ± 2 40 ± 9
3.9 ± 0.3 1.7 ± 0.3 1.8 ± 0.3 59 ± 10
Mean Mean Mean Mean
FGa ODGa ODGb Lossc
Mg
K
P
Na
Mean values also appear in the last row. a mg mineral/100 g fresh fruit. b mg mineral/100 g osmodehydrated fruit. c mg mineral loss/100 mg initial mineral.
50
12 Ca
Mg
P
K
Na 40 mg / 100 g OS
mg / 100g OS
10 8 6 4
30 20 10
2
0
0 0
1
2
3
4
5
6
7
Cycles
8
0
1
2
3
4 5 Cycles
6
7
8
Fig. 3. Mean values and standard deviation of different minerals analysed in the osmotic solution when reused for different osmotic dehydration cycles.
dehydration, which confirms the fruit solutes outflow and its main influence on EC value. In order to determine the solutes that may be responsible for this EC increase, a stepwise multiple regression analysis procedure was carried out considering AA, CA and different mineral content analysed in OS. In a first step, the analysis was carried out without considering interactions between compounds. The best equations (R2 > 98.9) were obtained when AA, CA, Mg and P were correlated with EC, although different combinations of these compounds lead to different equations
(Table 4). The close fit in all cases allows us to propose these equations as a tool for knowing the mineral content and acid content of the OS from just a simple electrical conductivity measurement. As an example, the EC value of cycle 8 (430 lS/cm) was used to predict values of 442 mg CA/100 g OS (Eq. (3)), 0.939 mg Mg/100 g OS (Eqs. (3) and (4), Table 4), 9.5 mg P/100 g OS (Eqs. (3)–(5), Table 4) and 4.3 mg AA/100 g OS (Eqs. (3)–(6), Table 4). The agreement of these predicted values and the experimental data can be observed comparing them with Figs. 2 and 3.
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307 20
sucrose
OS
400
17
µ (mPa.s)
EC ( µ S/cm)
500
305
300 200 100
14 11 OS sucrose
8
0 0
1
2
3
4
5
6
7
5
8
0
Cycles
Table 4 Best equations obtained from the stepwise multiple regression analysis carried out to correlate electrical conductivity (EC) of OS with analysed compounds (mg/100 g) (CA = citric acid, AA = ascorbic acid, Mg = magnesium, P = phosphorous) EC ðlS=cmÞ ¼ a CA þ b AA þ c Mg þ d P Equation
a
b
c
d
R2
(3) (4) (5) (6)
0.9708 0.7096 0.5489 –
– – – 29.6986
– 123.152 100.407 171.433
– – 9.9703 12.9281
98.9425 99.1851 99.3944 99.4584
R2: determination coefficient.
2
3
4 5 Cycles
6
7
8
Fig. 5. Mean values and standard deviation of viscosity (l) of the osmotic solution when reused for different osmotic dehydration cycles. Dotted line reproduces the model fitted to experimental data (Table 5). Viscosity of a sucrose solution with the same °Brix as osmotic solution at different osmotic dehydration cycles also appear in the figure.
20 y = 0,0004x 2 - 0,1354x + 18,291 R2 = 0,8847
15
µ (mPa.s)
Fig. 4. Mean values and standard deviation of electrical conductivity (EC) of the osmotic solution when reused for different osmotic dehydration cycles. Dotted line reproduces the model fitted to experimental data (Table 5). Electrical conductivity of a sucrose solution with the same °Brix as osmotic solution at different osmotic dehydration cycles also appears in the figure.
1
10 5 0 0
50
100
150
200
mg AGU/ 100 g OS
Fig. 6. Osmotic solution viscosity related to galacturonic acid content. Correlation equation between variables.
3.5. Pectin content and viscosity The galacturonic acid, related to the pectin content, was analysed in fresh and dehydrated grapefruit and in the OS (Table 2 and Fig. 2). A loss of galacturonic acid of grapefruit during osmotic dehydration was confirmed, which will be related with the loss of some soluble pectin that forms part of the fiber of the fruit. As observed for CA, significant differences (a < 0.05) both in AGU of different pieces of fresh fruit and in AGU loss by samples at different OD cycles were observed. Nevertheless, no clear tendencies associated to a different loss due to the reuse of the OS were observed. So, a certain variability of the OD operation could contribute to explaining the slight differences observed in AGU loss. The mean value of fresh and dehydrated samples appears in Table 2. The mean loss calculated corresponds to 18% of the initial galacturonic acid present in the sample (Table 2). At the same time, a progressive increase of the pectin content was observed in OS (Fig. 2). On the other hand, the viscosity changes in OS were analysed. Rheograms obtained with OS from each OD cycle showed a linear relationship between shear stress (r) and shear rate (_c), which points to the syrupÕs Newtonian behaviour. Viscosity (l) was calculated from the
curve slope. A decrease in l was observed due to OS dilution (Fig. 5). Nevertheless, when these results were compared with the viscosity value of sucrose solutions with the same °Brix as osmotic solution at different OD cycles (Pancoast & Junk, 1980), higher viscosity values of the reused OS were observed. The pectin content detected in the OS may be responsible for the observed differences as this component contributes to an increase in the viscosity. A second grade polynomic equation was found which closely correlates l with galacturonic acid presence (Fig. 6). 3.6. Correlation between OS physicochemical properties and the number of syrup uses As some of the measured properties in the OS, such as EC and l, have been related with some of its soluble components, it seems to be useful for predicting the value of these properties for a determined number of uses of the OS. For °Brix, aw, EC and l, empirical equations were found to correlate experimental values with the number of syrup uses. Table 5 shows the obtained parameters in each case and Figs. 1, 4 and 5 the predicted curves. As can be observed, a second grade polynomic equation
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
306
Table 5 Parameters of the model (y = ax2 + bx + c) fitted to predict changes in some osmotic solution properties as a function of number of dehydration cycles (x) Property
a
b
c
R2
°Brix aw EC (lS/cm) l (mPa s)
0.0998 0.0005 – 0.1931
2.4116 0.0089 51.679 3.1306
55.096 0.9186 5.7671 20.402
0.9982 0.9801 0.9949 0.9698
despite the contribution to the economic and ambient profitability of the OD operation. In this sense no problems related to colour or microbial counts of OS have been detected in a similar study carried out with kiwifruit (Garcı´a-Martı´nez et al., 2002). Good equations have been found not only to correlate physicochemical properties of the OS with the number of reuses but also with its composition.
R2: determination coefficient.
Acknowledgments Table 6 Parameters of the model (y = ax2 + bx + c) fitted to aw, EC and l vs. °Brix of osmotic solution (x) Property
a
b
c
R2
aw EC (lS/cm) l (mPa s)
– 1.5418 1.4963
0.0032 180.65 157.78
1.0947 5294.3 4175.5
0.9757 0.9927 0.9986
R2: determination coefficient.
fitted closely in all cases except for EC which was a linear model. These equations could be useful for determining the number of OD cycles that can be carried out depending on the desired final properties of the OS to be used, for example, in certain food formulations. On the other hand, practical equations were found to correlate aw, EC and l with OS °Brix for all the studied cycles. The results obtained showed a relationship in each case that was closely fitted mathematically to a second grade polynomic equation except for aw that was a linear model (Table 6). Despite the good correlation obtained in the case of EC and l it has to be remembered that, as commented on above, the increase in its value is related to the presence of other soluble solutes in the OS, apart from sugar, coming from the fruit. So the equations are only useful for predicting the concentration of the reused osmotic solution. Similar relationships have been described for OS reused in kiwifruit dehydration (Garcı´a-Martı´nez et al., 2002).
4. Conclusions Losses of ascorbic acid, citric acid, minerals and galacturonic acid associated to osmotic dehydration of grapefruit have been detected. Nevertheless, these micronutrients flow to the OS. On the other hand, OS dilution associated to the grapefruit OD under the conditions of this study allow the OS to be reused for at least eight cycles without significant changes in the water activity of the obtained fruit. From this point of view, the reuse of the OS becomes an advantage, as it can be used as an ingredient in new food formulations, with the presence of natural compounds coming from fruit,
The authors thank the Ministerio de Ciencia y Tecnologı´a and to the Fondo Europeo de Desarrollo Regional (FEDER) for the financial support throughout the project AGL2002-01793.
References Ahmed, A. E. R., & Labavitch, J. M. (1977). A simplified method for accurate determination of cell wall uronide content. Journal Food Biochemistry, 1, 361–365. AOAC (1997). Official methods of analysis (16th ed.). Washington, DC: Association of Official Analytical Chemist. Cheftel, J. C., & Cheftel, H. (1986). Introduction a` la biochimie et a` la technologie des aliments. Technique et Documentation Lavoisier, 1, 333–363. Chiralt, A., Martı´nez-Navarrete, N., Martı´nez-Monzo´, J., Talens, P., Moraga, G., Ayala, A., et al. (2001). Changes in mechanical properties throughout osmotic processes. Cryoprotectant effect. Journal of Food Engineering, 49, 129–135. Co´rdoba, A., Garcı´a-Martı´nez, E., Martı´nez-Navarrete, N., Camacho, M. M., & Martı´nez-Monzo´, J. (2003). Enriquecimiento de la disolucio´n osmo´tica en a´cido asco´rbico durante ciclos sucesivos de deshidratacio´n de kiwi. In P. Fito, A. Chiralt, A. Andre´s, & N. Martı´nez-Navarrete (Eds.), Series de Ciencia e Ingenierı´a de Alimentos. Investigacio´n del postgrado del IAD-DTA (pp. 103–121). Valencia: Editorial de la Universidad Polite´cnica de Valencia. De la Fuente, M. A., Montes, F., Guerrero, G., & Jua´rez, M. (2003). Total and soluble contents of calcium, magnesium, phosphorus and zinc in yoghurts. Food Chemistry, 80, 573–578. Fito, P., Andre´s, A., Pastor, R., & Chiralt, A. (1995). Mass transfer behaviour and structure changes during osmotic dehydration (VOD) of foods. In A. Lenart & P. P. Lewicki (Eds.), Osmotic dehydration of fruits and vegetables (pp. 5–19). Warsaw: Warsaw Agricultural University Press. Fito, P., & Chiralt, A. (2000). Vacuum impregnation of plants tissues. In S. M. Alzamora, M. S. Tapia, & A. Lo´pez-Malo (Eds.), Design of minimal processing technologies for fruit and vegetables (pp. 189–205). Maryland: Aspen Publishers. Fito, P., Chiralt, A., Barat, J., Andre´s, A., Martı´nez-Monzo´, J., & Martı´nez-Navarrete, N. (2001). Vacuum impregnation for development of new dehydrated products. Journal of Food Engineering, 49, 297–302. Forni, E., Torreggiani, D., Crivelli, G., Mastrelli, A., Bertolo, G., & Santelli, F. (1987). Influence of osmosis time on the quality of dehydrofrosen kiwifruit. Acta Horticulturae, 282, 425–433. Garcı´a-Martı´nez, E., Martı´nez-Monzo´, J., Camacho, M. M., & Martı´nez-Navarrete, N. (2002). Characterisation of reused osmotic solution as ingredient in new product formulation. Food Research International, 35, 307–313.
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307 Giangiacomo, R., Torreggiani, D., Erba, M. J., & Messina, G. (1994). Use of osmodehydrofrozen fruit cubes in yoghurt. Italian Journal of Food Science, 6(3), 345–350. Heng, K., Guilbert, S., & Cup, J. (1990). Osmotic dehydration of papaya: Influence of process variables on the product quality. Science des Aliments, 10, 831–848. Hertog, M. G. L., Hollman, P. C. H., & Van de Putte, B. (1993). Content of potentially anticarcinogenic flavonoids of tea infusions, wines, and fruit juices. Journal of Agricultural and Food Chemistry, 41(8), 1242–1246. Kitner, P. K., III, & Van Buren, J. P. (1982). Carbohydrate interference and its correction in pectin analysis using the mhydroxydiphenil method. Journal Food Science, 47, 756–759, 764. Liu, Ming, Li, Xin-Qi, Weber, C., Lee, Chang-Yong, Brown, J., & Liu, Rui-Hai (2002). Antioxidant and antiproliferative activities of raspberries. Journal of Agricultural and Food Chemistry, 50(10), 2926–2930. Martı´nez-Monzo´, J., Martı´nez-Navarrete, N., Chiralt, A., & Fito, P. (2001). Combined vacuum impregnation and osmotic dehydration in fruit cryoprotecction. In P. Fito, A. Chiralt, J. M. Barat, W. E. L. Spiess, & D. Behsnilian (Eds.), Osmotic dehydration and vacuum impregnation. Application in food industries (pp. 61–78). Lancaster: Technomic Publishing Company. Martı´nez-Navarrete, N., Andre´s, A., Chiralt, A., & Fito, P. (2000). Termodina´mica y cine´tica de sistemas alimento-entorno (pp. 116– 132). Valencia: Servicio de Publicaciones de la Universidad Polite´cnica de Valencia. Mataix. (1998). Tabla de composicion de alimentos. Ed. Universidad de Granada. Moraga, M. J. (2004). Pomelo mı´nimamente procesado por deshidratacio´n osmo´tica. Cambios en las propiedades fı´sicas durante el almacenamiento. PhD thesis, Universidad Polite´cnica de Valencia.
307
Panagiotou, N. M., Karathanos, V. T., & Maroulis, Z. B. (1998). Mass transfer modelling of the osmotic dehydration of some fruits. International Journal of Food Science and Technology, 33, 267–284. Pancoast, H. M., & Junk, W. R. (1980). Handbook of sugars (p. 46). Westport: Avi Publishing Company. Pinnavaia, G., Dalla-Rosa, N. I., & Lerici, C. R. (1988). Dehydrofreezing of fruit using direct osmosis as concentration process. Acta Alimentaria Polonica, 14, 51–57. Prior, R. L., & Guohua, Cao. (2000). Antioxidant phytochemicals in fruits and vegetables: Diet and health implications. HortScience, 35(4), 588–592. Raoult-Wack, A. L. (1994). Recent advances in osmotic dehydration of foods. Trends in Food Science and Technology, 5, 255–260. Taiwo, K. A., Angersbach, A., Ade-Omowaye, B. I. O., & Knorr, D. (2001). Effects of pre-treatments on the diffusion kinetics and some quality parameters of osmotically dehydrated apple slices. Journal Agriculture Food Chemistry, 49, 2804–2811. Torreggiani, D., & Bertolo, G. (2001). Osmotic pre-treatments in fruit processing: Chemical, physical and structural effects. Journal of Food Engineering, 49, 247–253. Valdez-Fragoso, A., Welti-Chanes, J., & Giroux, F. (1998). Properties of sucrose solution reused in osmotic dehydration of apples. Drying Technology, 6(7), 1429–1445. Venant, A., Borrel, S., Mallet, J., & Van Neste, E. (1989). Gel permeation chromatography as a method for the simultaneous clean up of organochlorine, organophosphate and polychlorinated biphenyl residues in fat extracts. Analusis, 17, 64–68. Wargovich, M. J. (2000). Anticancer properties of fruits and vegetables. HortScience, 35(4), 573–575. Wolfe, K. L., & Liu, Rui-Hai (2003). Apple peels as a value-added food ingredient. Journal of Agricultural and Food Chemistry, 51(6), 1676–1683.
Micronutrient flow to the osmotic solution during grapefruit osmotic dehydration R. Peiro´ a, V.M.C. Dias b, M.M. Camacho a, N. Martı´nez-Navarrete
a,*
a
b
Department of Food Technology, Universidad Polite´cnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain Escola Superior de Biotecnologia, Universidade Cato´lica Portuguesa, Rua Dr. Anto´nio Bernardino de Almeida, 4200-072 Porto, Portugal Received 27 July 2004; accepted 7 March 2005 Available online 26 April 2005
Abstract The purpose of this work was to quantify the flow of soluble micronutrients, such as acids, minerals and pectins, from the grapefruit to the osmotic solution (OS) used to dehydrate the fruit when recycling it in successive osmotic dehydration (OD) operations, without reconcentrating. OD was carried out for 3 h at 30 °C with an OS:fruit rate 5:1, using a 55 °Brix sucrose solution. Soluble solids (°Brix), water activity (aw), pH, ascorbic (AA), citric acid (CA), galacturonic acid and major minerals were measured in the fruit and in the OS as a function of the number of uses (up to 8). Electrical conductivity (EC) and viscosity (l) of the OS were also analysed. The characteristics of the obtained dehydrated grapefruit and the observed recovery of the quantified micronutrient loss by the fruit in the OS allows us to propose the reuse of the OS as a good way of contributing to the economic and ambient profitability of the OD operation. Moreover, its final use as ingredient in new food formulations may be proposed. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Osmotic solution; Ascorbic acid; Citric acid; Minerals; Pectin; Electrical conductivity; Viscosity
1. Introduction Fruits are an important part of the human diet as, besides contributing with some nutrients, they contain a series of non-nutritive substances, called phytochemical. These bioactive compounds, although having neither a classically defined nutritional function nor being considered essential for the human health, can have a significant impact on prevention of some diseases. Phytochemicals have multiple biological effects, including antioxidant activity (Liu et al., 2002; Prior & Guohua, 2000; Wolfe & Liu, 2003), antimutagenic (Wargovich, 2000) antibacterial and angioprotective properties (Venant, Borrel, Mallet, & Van Neste, 1989). This group of compounds mainly includes phenolic compounds,
*
Corresponding author. Tel.: +34 96 387 9362; fax: +34 96 387 7956. E-mail address: [email protected] (N. Martı´nez-Navarrete).
0260-8774/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jfoodeng.2005.03.022
lignans, monoterpenes, etc. Their intake has been associated with a reduced risk of coronary heart disease and strokes (Hertog, Hollman, & Van de Putte, 1993). On the other hand, also micronutrients present in fruits, such as vitamins, acids, fiber, etc. also contribute to their well known beneficial properties. Nevertheless, in general, fruits are products with a relatively short post-harvest life-span and recent food habits (easy and quick to eat food) have promoted a decrease in the consumption of fresh fruit, which has been replaced by juices, dairy products with added fresh or processed fruits, preserves, confectioneries, etc. (Torreggiani & Bertolo, 2001). To this end, many technologies have been used to make the food system formulation more suitable. One of the possible techniques for fruit processing is osmotic dehydration (OD) with sugar solutions, working at middle temperatures to preserve product flavour and other sensory properties.
300
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
During osmotic dehydration of fruit in an osmotic solution (OS), a two-way mass transfer is established: water and some natural soluble substances (sugars, vitamins, pigments, organic acids, mineral salts, etc.) flow out of the fruit into the OS, and in the opposite direction, soluble solutes may be transferred from the solution to the fruit. Due to the kinetics of the operation it may be used to obtain products of reduced but still relatively high moisture content, classified as intermediate moisture products. The operation may be used as an intermediate step in other processes such as drying (Fito et al., 2001) and freezing (Chiralt et al., 2001) or to produce minimally processed fruit (Martı´nez-Monzo´, Martı´nez-Navarrete, Chiralt, & Fito, 2001). Moreover, the application of a vacuum pulse for a short period of time at the beginning of the process (pulsed vacuum osmotic dehydration, PVOD) may have beneficial effects on the transfer kinetics leading to shorter processing times (Fito & Chiralt, 2000). Mass transfer rates during OD operations depend on factors such as temperature, type and concentration of osmotic medium, type, size and geometry of the sample, sample to solution ratio and degree of agitation of the solution (Taiwo, Angersbach, Ade-Omowaye, & Knorr, 2001). This method has received considerable attention due to the low energy requirements (Taiwo et al., 2001) and fruit quality improvement (Heng, Guilbert, & Cup, 1990; Panagiotou, Karathanos, & Maroulis, 1998), compared to alternative process schemes. As fruits are not submitted to high temperatures, sensory attribute changes, such as colour, aroma flavour and texture are minimised (Fito, Andre´s, Pastor, & Chiralt, 1995; Heng et al., 1990; Raoult-Wack, 1994). Nevertheless, a loss of vitamins, polysaccharides and minerals, that flow from the fruit to the osmotic solution, has been observed (Co´rdoba, Garcı´a-Martı´nez, Martı´nezNavarrete, Camacho, & Martı´nez-Monzo´, 2003; Garcı´a-Martı´nez, Martı´nez-Monzo´, Camacho, & Martı´nez-Navarrete, 2002). Moreover, in comparison with other traditional drying treatments, OD does not affect the food structure because water elimination does not involve phase changes (Forni et al., 1987; Giangiacomo, Torreggiani, Erba, & Messina, 1994; Pinnavaia, DallaRosa, & Lerici, 1988). Despite the above mentioned advantages, one limitation for large scale production using the fruit OD process is the management of the osmotic solution, commonly sucrose syrup, which, if not used, has to be processed as waste thus increasing the cost of the osmodehydrated products. Different alternatives may be pointed out to solve this problem. One is to submit the used OS to a reconcentration step. Nevertheless, the installation and use of this stage will also increase the cost of the dehydration operation. Another possibility is to reuse or recycle it in successive dehydration cycles without any treatment, which will be limited by its
dilution related to the dehydration level of the obtained fruit and also to microbiological aspects. Finally, it is possible to use the OS as an ingredient to be added in some product formulations. To this end, and taking into account that, as commented on above, during fruit OD a loss of valuable natural substances such as vitamins, acids, minerals, etc. occurs, a previous reuse of the OS for some cycles without any reconcentration treatment, will be desirable. The objective of this work was to study some compositional (water, soluble solids, ascorbic acid, citric acid, major minerals and pectin) and physicochemical (water activity, pH, electrical conductivity and viscosity) changes associated with osmodehydration of grapefruit, both in the fruit and in the sucrose solution used as the osmotic agent.
2. Materials and methods 2.1. Raw matter Fresh grapefruit (Citrus paradise) of red coloured variety (Star Ruby), purchased in local markets in Valencia (Spain), were used and stored under chilling conditions (8 °C) till the moment of use (maximum 24 h). They were selected on the basis of a similar ripening degree (between 11 and 13 °Brix) and uniform visual quality. The fruits were manually peeled and cut, perpendicularly to the fruit axis, into uniform half-slices with a knife (1 cm thickness). The osmotic solution was prepared using commercial sugar mixed with heated (30 °C) distilled water until total dissolution to form a 55 °Brix syrup. 2.2. Osmotic dehydration process The grapefruit samples were immersed in a plastic tank filled with 55 °Brix syrup for 3 h at 30 °C. For the first 10 min a vacuum pulse (50 mbar) was applied. Afterwards, osmodehydration was conducted at atmospheric pressure in a temperature-controlled water bath, where the syrup was continually stirred with a Heidolph RZR2102 stirrer at 200 rpm. A plastic screen was placed on the basket to keep the fruit pieces totally immersed in OS and separated from stirrer. This dehydration time was chosen on the basis of the results obtained in previous osmotic dehydration kinetics studies to obtain grapefruit with about 22 °Brix (Moraga, 2004). The syrup to fruit ratio was 5:1. Successive OD cycles, up to a maximum of eight, were carried out using the same OS for all the cycles, without any treatment, and renewing the fruit after each OD cycle. Three days were used to complete the experience (two or three OD cycles by day) and OS was kept at 8 °C from one day to another.
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
2.3. Analysis OS was analysed after 0, 1, 2, 3, 4, 6, and 8 OD cycles as to soluble solids (°Brix) at 20 °C (ATAGO NAR-3T refractometer, Japan), water activity (aw; Decagon model CX-2 dew point hygrometer), pH (Crison micro pH 2001 pH-meter), electrical conductivity (K; Crison 522 conductimeter), ascorbic acid (AA) (AOAC 985.33, 1997) and citric acid (CA) (AOAC 985.33, 1997). To calculate citric acid content, AA presence was taken into account (Eq. (1)), where m, v and M are the mass (g), valence and molecular weight of CA and AA and V and N the volume and normality of the NaOH used for titration. V NaOH N NaOH ¼
mAC mAA þ M AC M AA vAC vAA
301
fuged (Selecta model S40) for 10 min at 12 000 rpm, and the floating liquid was used to follow the above mentioned AOAC method. xw was determined according to the AOAC method 950.46 (1997). All other analyses were measured following the same methods and equipment described for OS analysis. Statistical analyses to obtain means, standard deviations of means, analysis of variance (ANOVA), including the study of homogeneous groups established throughout the Least Significant Difference (LSD) test, and stepwise multiple regression analysis were performed using the Statgraphics Plus 4.0 program.
3. Results and discussion ð1Þ
Besides viscosity (Physica Rheolab MC1 rheometer at 25 °C, shear rate sweep from 0 to 150 s1 in 90 s), pectin content, and major mineral content of OS were also determined, although these were only analysed in cycles 1, 3, 5 and 8. Galacturonic acid quantification was the method used to determine the amount of pectin present in the samples (Kitner & Van Buren, 1982). To this end, samples were ethanol extracted by centrifugation and the precipitate that contains the alcohol-insoluble solids was extracted with sulphuric acid and distilled water (Ahmed & Labavitch, 1977). Determination of pectin content of the extract was carried out by the m-hydroxydiphenyl method (Kitner & Van Buren, 1982) reading the absorbance of the samples at 520 nm (Cecil 1020 spectrophotometer). Minerals were analysed according to the method described by De la Fuente, Montes, Guerrero, and Jua´rez (2003), consisting in a mineralisation of samples by dry ashing in a muffle furnace and a posterior dilution of the ashes with concentrated hydrochloric acid and distilled water. Calcium and magnesium were determined by atomic absorption spectrometry using a multi-element (Ca–Mg–Zn) hollow-cathode lamp. Potassium and sodium were analysed by atomic emission spectrometry. A 3110 atomic absorption spectrometer (Perkin-Elmer) in an air-acetylene flame was used for the analysis. Phosphorus was measured using a UV–visible spectrophotometer, Cecil 1020. Three mineralisation of each sample were carried out and each one was analysed in triplicate. Fresh grapefruit used for each OD cycle as well as the dehydrated one were analysed as to water content (xw), °Brix, aw, pH, ascorbic acid, citric acid, pectin content and the same mineral content described for OS. In all cases, fruit samples were previously homogenised with an Ultra-Turrax T25. For the ascorbic acid content determination, the obtained pulp was previously centri-
3.1. Water content, °Brix and water activity Table 1 shows xw, °Brix and aw of fresh grapefruit used for each OD cycle and of grapefruit obtained after each dehydration process. The ANOVA carried out with data of fresh fruit used for each OD cycle showed significant differences (a < 0.05) in xw and °Brix, which ranged between 85.2–87.1 g water/100 g fruit and 11.9– 13.5 g soluble solids/100 g fruit. Nevertheless, this compositional variability, that seems to be normal in this kind of raw material, was not reflected by significant differences in the aw of pieces (a > 0.05), which oscillated between 0.989 and 0.992. After each dehydration cycle, under the conditions pointed out in the methodology, Table 1 °Brix, water activity (aw) and water content (xw) (g water/100 g sample) of fresh (FG) and osmodehydrated (ODG) grapefruit used and obtained for each osmotic dehydration cycle (1–8) Sample
°Brix
aw
xw
FG1 ODG1
11.9 ± 0.1 23 ± 0
0.9915 ± 0.0007 0.981 ± 0.005
86.4 ± 0.1 75.5 ± 0.7
FG2 ODG2
12.2 ± 0.1 23.1 ± 0.1
0.9890 ± 0.0014 0.9785 ± 0.0007
85.2 ± 0.1 74.20 ± 0.02
FG3 ODG3
11.9 ± 0.1 24.3 ± 0.1
0.989 ± 0.003 0.976 ± 0.002
86.58 ± 0.05 75.8 ± 0.6
FG4 ODG4
13.25 ± 0.07 21.4 ± 0.1
0.9890 ± 0.0014 0.9855 ± 0.0007
86.440 ± 0.002 78.1 ± 0.1
FG5 ODG5
13.5 ± 0.1 22.6 ± 0.1
–
–
FG6 ODG6
12.4 ± 0.1 20 ± 0
0.9885 ± 0.0007 0.983 ± 0.002
FG7 ODG7
12.3 ± 0.1 22.5 ± 0.1
–
FG8 ODG8
12.7 ± 0.1 21.7 ± 0.1
0.989 ± 0.002 0.9800 ± 0.0014
86.8 ± 0.1 78 ± 1
Mean FG Mean ODG
12.5 ± 0.6 22.3 ± 1.3
0.9895 ± 0.0012 0.981 ± 0.003
86.4 ± 0.6 77 ± 2
Mean values also appear in the last row.
87.09 ± 0.05 80.1 ± 0.2 –
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
60
0.98
50
0.97
40
0.96
30
0.95
tively. Nevertheless, aw was not significantly affected by the reuse of OS. From this point of view, it can be considered possible to reuse the OS for eight cycles, under the conditions of this study, to obtain grapefruit with the same stability (aw). This is important in order to make the OD process more profitable, as the reuse of the OS without any reconcentration treatment is possible maintaining the same OD time.
aw
ºBrix
302
0.94
20 ºBrix aw Predicted aw
10 0 0
1
2
3
4 5 Cycles
6
0.93 0.92 7
3.2. Citric acid, ascorbic acid and pH
8
Fig. 1. Experimental mean values and standard deviation of °Brix and aw of the osmotic solution when reused for different osmotic dehydration cycles. Dotted lines reproduce the model fitted to experimental data (Table 5). Continuous line corresponds to aw values predicted by Norrish equation.
samples reached approximately the expected value of 22 °Brix (Moraga, 2004). Fig. 1 shows the decrease in soluble solid content and the increase in water activity observed in the OS after each OD cycle. As expected, a progressive syrup dilution was observed due to the water transferred from grapefruit to the solution. After the 8th OD cycle studied, the solution became diluted from an initial value 55.1 ± 0.1 to 41.9 ± 0.1 °Brix. Consequently aw increased from 0.920 ± 0.005 to 0.960 ± 0.005. Water activity of OS was compared with aw predicted for sucrose solutions with the same °Brix reached by OS after each OD cycle. To this end, Norrish equation (Eq. (2); Martı´nez-Navarrete, Andre´s, Chiralt, & Fito, 2000) was used, assuming that sucrose was the only soluble solute present in the osmotic solution. aw ¼ xw expð6.47x2s Þ
ð2Þ
where xw and xs are the molar fraction of water and soluble solids (°Brix), respectively. Fig. 1 shows the obtained predicted values. As can be observed, experimental values and predicted curve differed more with cycles. The lowest values of predicted aw could be related with the contribution of other solutes, different from sugars, to the °Brix of the OS and with a weaker interaction with water. As concerns to dehydrated fruit (Table 1), if the progressive dilution of OS observed when reusing it had a significant effect on the dehydration kinetics, a different dehydration level in the fruit processed in each OD cycle could be expected. The statistic study carried out using an ANOVA, taking the cycle factor into account, confirmed significant differences (a < 0.05) in xw and °Brix but only allows us to consider two homogeneous groups of samples: one corresponding to cycles 1, 2, 3, with mean values of xw and °Brix of 75.1 and 23.6, respectively, and another one for the rest of the samples with mean values of xw and °Brix of 78.7 and 21.6, respec-
Table 2 shows citric, ascorbic and galacturonic acid of different fresh and dehydrated grapefruit pieces used and obtained after each dehydration process. A certain variability in these parameters was observed in the fresh fruit again, with citric and ascorbic acid content ranging between 941–1167 mgcitric acid/100 gfruit and 10–14.5 mgascorbic acid/100 gfruit, respectively. pH of fresh
Table 2 Citric acid (CA), ascorbic acid (AA) and galacturonic acid (AGU) of fresh (FG) and osmodehydrated (ODG) grapefruit used and obtained for each osmotic dehydration cycle (1–8) and percentage of loss of each component Sample
Citric acid
FG1a ODG1a ODG1b Loss1c
1035 ± 5 757 ± 14 865 ± 16 27 ± 1
12 ± 2 7.4 ± 0.2 8.5 ± 0.2 6±2
4453 ± 192 3756 ± 164 4870 ± 164 16 ± 4
FG2a ODG2a ODG2b Loss2c
1050 ± 11 834 ± 21 929 ± 23 21 ± 2
14.5 ± 0.2 9.8 ± 0.2 10.9 ± 0.2 6±1
5267 ± 41 4130 ± 196 5253 ± 196 22 ± 4
FG3a ODG3a ODG3b Loss3c
941 ± 8 642 ± 3 728 ± 3 31.7 ± 0.4
14 ± 1 9±1 10 ± 1 9±1
4558 ± 149 3706 ± 206 4656 ± 206 19 ± 3
FG4a ODG4a ODG4b Loss4c
1167 ± 5 830.5 ± 0.3 947.2 ± 0.3 29 ± 0.3
14 ± 1 7.4 ± 0.3 8.4 ± 0.3 11 ± 2
4667 ± 173 4115 ± 376 5032 ± 376 12 ± 3
FG6a ODG6a ODG6b Loss6c
1032 ± 11 846 ± 28 938 ± 31 18 ± 2
10 ± 1 7±1 8±1 12 ± 3
4207 ± 11 2964 ± 191 4236 ± 191 30 ± 3
FG8a ODG8a ODG8b Loss8c
986 ± 4 780 ± 9 917 ± 10 21.4 ± 0.2
10 ± 0 6.1 ± 0.2 7.2 ± 0.2 25 ± 1
4086 ± 21 3559 ± 243 5120 ± 243 13 ± 4
1035 ± 76 782 ± 76 887 ± 83 25 ± 5
11 ± 4 8±1 9±1 12 ± 7
4540 ± 417 3705 ± 429 4861 ± 429 18 ± 7
Mean Mean Mean Mean
FGa ODGa ODGb Lossc
Ascorbic acid
Mean values also appear in the last row. a mg component/100 g fresh fruit. b mg component/100 g osmodehydrated fruit. c mg component loss/100 mg initial component.
AGU
500
5
CA AGU AA
400
4
300
3
200
2
100
1
0
mg AA/ 100 g OS
mg CA or AGU/ 100 g OS
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
0 0
1
2
3
4 5 Cycles
6
7
8
9
Fig. 2. Mean values and standard deviation of ascorbic acid (AA), citric acid (CA) and galacturonic acid (AGU) of the osmotic solution when reused for different osmotic dehydration cycles.
grapefruit used in this study varied between 2.94 and 3.34, being 3.19 ± 0.16 its mean value. A loss of acids was detected in grapefruit due to dehydration process. This was reflected in lower acid content in the dehydrated fruit, although its pH (3.16 ± 0.05) was not significant different from pH of fresh fruit. Consequently, a progressive AA and CA enrichment of OS was observed (Fig. 2), changing from 1.30 ± 0.04 in cycle 1 up to 4.33 ± 0.04 mg AA/100 g OS in cycle 8 and from 72 ± 4 to 414 ± 8 mg CA/100 g OS. On the other hand, the greatest pH variation in the syrup took place during the first usage, changing from 6.7 ± 0.2 to 3.50 ± 0.01. By the third use the syrup reached the same pH as the fruit and remained at about the same value (pH = 3.2) for all subsequent cycles. As the most abundant acid in grapefruit is citric, which is a weak acid, the observed change in its concentration does not affect the pH. Similar behaviour has been observed by ValdezFragoso, Welti-Chanes, and Giroux (1998) who studied changes in 60 °Brix OS when used successively in apple dehydration after a reconcentration process and also by Garcı´a-Martı´nez et al. (2002) when analysing changes of a 55 °Brix OS in successive kiwifruit dehydration cycles for three OS:fruit ratios. These authors point out the interest of the observed pH decrease that could contribute inhibiting the growth of microorganisms when reusing the OS in successive OD cycles. The presence of acids in the solution could also have a beneficial influence on the inhibition of enzymatic browning (Cheftel & Cheftel, 1986; Valdez-Fragoso et al., 1998). To quantify the loss of acids in the grapefruit due to the osmotic dehydration of the fruit, the amount of acids in dehydrated samples was also expressed by 100 g of fresh fruit (Table 2). An ANOVA was again carried out to study differences in the amount of acid loss due to the reuse of OS. Loss of AA showed no significant differences (a > 0.05) associated to this factor. In the case of CA, these differences were significant (a < 0.05) but as no clear tendencies were observed associated to the number of cycles, the variability may be attributed to small differences in controlled OD
303
variables from one cycle to another. In this way, a mean value of acid loss in the grapefruit during its osmotic dehydration may be calculated, these being 12 ± 7 mg of AA/100 mg of AA present in the fresh fruit and 25 ± 5 mg CA/100 mg of CA present in the fresh fruit. 3.3. Mineral content Major minerals of grapefruit (Mataix, 1998) were analysed before and after dehydration process (Table 3). As no significant differences (a < 0.05) were obtained for the different fresh fruit pieces used in each OD cycle, the mean values for each mineral also appear in Table 3, which agree with data reported by Mataix (1998): K (190 mg/100 g), Ca (14 mg/100 g), P (14 mg/100 g) and Mg (10 mg/100 g). As can be observed, and as is typical in the fruits, the potassium was the most abundant mineral, present in about 10 times as great a quantity as others such as magnesium or calcium. Sodium was present in a very low quantity. In all the cycles, a loss in minerals content was observed when OD fruit was analysed. Fig. 3 shows the minerals recovered in the OS over the course of the eight OD cycles and Table 3 shows fruit losses. In this case, results have again been referred to the corresponding fresh sample to compare differences. No significant differences (a > 0.05) were obtained in the mineral loss by the fruit due to the reuse of OS. The mean values of mineral content in the dehydrated fruit appear in Table 3, together with the percentage of each mineral lost (referred to the amount present in the fresh fruit) during OD. These values ranged from 28% to 59%. 3.4. Electrical conductivity of OS The electrical conductivity (EC) of the syrup depends on the sugar concentration, temperature, chemical composition of water used in its preparation and on the type and concentration of some soluble components that may be present coming from the fruit together with the water loss during the dehydration process. The distilled water used to prepare OS had an EC = 4.13 ± 0.02 lS/cm which changed to 12.38 ± 0.05 lS/cm when sucrose was added to obtain a 55 °Brix osmotic solution. Compositional changes of OS during the OD process promote an EC increase over the course of the OD cycles (Fig. 4). This increase in EC from the first use of the syrup could be attributed to the observed progressive enrichment of mineral salts and organic acids coming from the grapefruit. Nevertheless, the greater mobility of the system, due to the lower viscosity of the OS associated to its dilution, commented on later, could also contribute to this increase. In order to separate both effects, a sucrose solution with the °Brix of that obtained after each OD cycle was prepared and the EC was measured (Fig. 4). The obtained values were much lower than that corresponding to the OS used in grapefruit
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
304
Table 3 Mineral content of fresh (FG) and osmodehydrated (ODG) grapefruit used and obtained for each osmotic dehydration cycle (1–8) and percentage of loss of each mineral Sample
Ca
FG1a ODG1a ODG1b Loss1c
14.7 ± 0.4 10 ± 1 12 ± 2 29 ± 7
11.8 ± 0.3 8±2 9±2 33 ± 11
206 ± 4 66 ± 13 76 ± 15 68 ± 7
14.9 ± 0.2 7±1 8±1 51 ± 7
3.5 ± 0.2 2±1 2±1 52 ± 21
FG3a ODG3a ODG3b Loss3c
14.5 ± 0.3 11 ± 1 13 ± 1 23 ± 6
11.4 ± 0.2 9±1 10.42 ± 0.03 19 ± 4
194 ± 5 100 ± 24 114 ± 27 48 ± 13
14.63 ± 0.04 11 ± 1 12 ± 1 28 ± 9
4.3 ± 0.2 1.71 ± 0.02 1.94 ± 0.03 60 ± 1
FG5a ODG5a ODG5b Loss5c
14.7 ± 0.3 11 ± 1 12 ± 1 28 ± 4
12.1 ± 0.2 7±1 8±0 35 ± 9
190 ± 2 85 ± 25 96 ± 28 50 ± 7
14.7 ± 0.1 8.1 ± 0.2 9.2 ± 0.3 41 ± 2
3.80 ± 0.14 1.7 ± 0.2 1.9 ± 0.2 53 ± 3
FG8a ODG8a ODG8b Loss8c
14.3 ± 0.3 10 ± 1 12 ± 1 28 ± 7
12.8 ± 0.7 9.0 ± 0.5 10.6 ± 0.5 30 ± 1
195 ± 6 70 ± 2 82 ± 2 64 ± 1
14.62 ± 0.11 8.2 ± 0.3 9.6 ± 0.4 41 ± 3
4.0 ± 0.3 1.21 ± 0.03 1.42 ± 0.03 70 ± 1
14.5 ± 0.3 10.6 ± 0.4 12.1 ± 0.4 28 ± 5
12.0 ± 0.6 8±1 10 ± 1 29 ± 9
196 ± 7 80 ± 16 92 ± 17 57 ± 10
14.71 ± 0.13 9±2 10 ± 2 40 ± 9
3.9 ± 0.3 1.7 ± 0.3 1.8 ± 0.3 59 ± 10
Mean Mean Mean Mean
FGa ODGa ODGb Lossc
Mg
K
P
Na
Mean values also appear in the last row. a mg mineral/100 g fresh fruit. b mg mineral/100 g osmodehydrated fruit. c mg mineral loss/100 mg initial mineral.
50
12 Ca
Mg
P
K
Na 40 mg / 100 g OS
mg / 100g OS
10 8 6 4
30 20 10
2
0
0 0
1
2
3
4
5
6
7
Cycles
8
0
1
2
3
4 5 Cycles
6
7
8
Fig. 3. Mean values and standard deviation of different minerals analysed in the osmotic solution when reused for different osmotic dehydration cycles.
dehydration, which confirms the fruit solutes outflow and its main influence on EC value. In order to determine the solutes that may be responsible for this EC increase, a stepwise multiple regression analysis procedure was carried out considering AA, CA and different mineral content analysed in OS. In a first step, the analysis was carried out without considering interactions between compounds. The best equations (R2 > 98.9) were obtained when AA, CA, Mg and P were correlated with EC, although different combinations of these compounds lead to different equations
(Table 4). The close fit in all cases allows us to propose these equations as a tool for knowing the mineral content and acid content of the OS from just a simple electrical conductivity measurement. As an example, the EC value of cycle 8 (430 lS/cm) was used to predict values of 442 mg CA/100 g OS (Eq. (3)), 0.939 mg Mg/100 g OS (Eqs. (3) and (4), Table 4), 9.5 mg P/100 g OS (Eqs. (3)–(5), Table 4) and 4.3 mg AA/100 g OS (Eqs. (3)–(6), Table 4). The agreement of these predicted values and the experimental data can be observed comparing them with Figs. 2 and 3.
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307 20
sucrose
OS
400
17
µ (mPa.s)
EC ( µ S/cm)
500
305
300 200 100
14 11 OS sucrose
8
0 0
1
2
3
4
5
6
7
5
8
0
Cycles
Table 4 Best equations obtained from the stepwise multiple regression analysis carried out to correlate electrical conductivity (EC) of OS with analysed compounds (mg/100 g) (CA = citric acid, AA = ascorbic acid, Mg = magnesium, P = phosphorous) EC ðlS=cmÞ ¼ a CA þ b AA þ c Mg þ d P Equation
a
b
c
d
R2
(3) (4) (5) (6)
0.9708 0.7096 0.5489 –
– – – 29.6986
– 123.152 100.407 171.433
– – 9.9703 12.9281
98.9425 99.1851 99.3944 99.4584
R2: determination coefficient.
2
3
4 5 Cycles
6
7
8
Fig. 5. Mean values and standard deviation of viscosity (l) of the osmotic solution when reused for different osmotic dehydration cycles. Dotted line reproduces the model fitted to experimental data (Table 5). Viscosity of a sucrose solution with the same °Brix as osmotic solution at different osmotic dehydration cycles also appear in the figure.
20 y = 0,0004x 2 - 0,1354x + 18,291 R2 = 0,8847
15
µ (mPa.s)
Fig. 4. Mean values and standard deviation of electrical conductivity (EC) of the osmotic solution when reused for different osmotic dehydration cycles. Dotted line reproduces the model fitted to experimental data (Table 5). Electrical conductivity of a sucrose solution with the same °Brix as osmotic solution at different osmotic dehydration cycles also appears in the figure.
1
10 5 0 0
50
100
150
200
mg AGU/ 100 g OS
Fig. 6. Osmotic solution viscosity related to galacturonic acid content. Correlation equation between variables.
3.5. Pectin content and viscosity The galacturonic acid, related to the pectin content, was analysed in fresh and dehydrated grapefruit and in the OS (Table 2 and Fig. 2). A loss of galacturonic acid of grapefruit during osmotic dehydration was confirmed, which will be related with the loss of some soluble pectin that forms part of the fiber of the fruit. As observed for CA, significant differences (a < 0.05) both in AGU of different pieces of fresh fruit and in AGU loss by samples at different OD cycles were observed. Nevertheless, no clear tendencies associated to a different loss due to the reuse of the OS were observed. So, a certain variability of the OD operation could contribute to explaining the slight differences observed in AGU loss. The mean value of fresh and dehydrated samples appears in Table 2. The mean loss calculated corresponds to 18% of the initial galacturonic acid present in the sample (Table 2). At the same time, a progressive increase of the pectin content was observed in OS (Fig. 2). On the other hand, the viscosity changes in OS were analysed. Rheograms obtained with OS from each OD cycle showed a linear relationship between shear stress (r) and shear rate (_c), which points to the syrupÕs Newtonian behaviour. Viscosity (l) was calculated from the
curve slope. A decrease in l was observed due to OS dilution (Fig. 5). Nevertheless, when these results were compared with the viscosity value of sucrose solutions with the same °Brix as osmotic solution at different OD cycles (Pancoast & Junk, 1980), higher viscosity values of the reused OS were observed. The pectin content detected in the OS may be responsible for the observed differences as this component contributes to an increase in the viscosity. A second grade polynomic equation was found which closely correlates l with galacturonic acid presence (Fig. 6). 3.6. Correlation between OS physicochemical properties and the number of syrup uses As some of the measured properties in the OS, such as EC and l, have been related with some of its soluble components, it seems to be useful for predicting the value of these properties for a determined number of uses of the OS. For °Brix, aw, EC and l, empirical equations were found to correlate experimental values with the number of syrup uses. Table 5 shows the obtained parameters in each case and Figs. 1, 4 and 5 the predicted curves. As can be observed, a second grade polynomic equation
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307
306
Table 5 Parameters of the model (y = ax2 + bx + c) fitted to predict changes in some osmotic solution properties as a function of number of dehydration cycles (x) Property
a
b
c
R2
°Brix aw EC (lS/cm) l (mPa s)
0.0998 0.0005 – 0.1931
2.4116 0.0089 51.679 3.1306
55.096 0.9186 5.7671 20.402
0.9982 0.9801 0.9949 0.9698
despite the contribution to the economic and ambient profitability of the OD operation. In this sense no problems related to colour or microbial counts of OS have been detected in a similar study carried out with kiwifruit (Garcı´a-Martı´nez et al., 2002). Good equations have been found not only to correlate physicochemical properties of the OS with the number of reuses but also with its composition.
R2: determination coefficient.
Acknowledgments Table 6 Parameters of the model (y = ax2 + bx + c) fitted to aw, EC and l vs. °Brix of osmotic solution (x) Property
a
b
c
R2
aw EC (lS/cm) l (mPa s)
– 1.5418 1.4963
0.0032 180.65 157.78
1.0947 5294.3 4175.5
0.9757 0.9927 0.9986
R2: determination coefficient.
fitted closely in all cases except for EC which was a linear model. These equations could be useful for determining the number of OD cycles that can be carried out depending on the desired final properties of the OS to be used, for example, in certain food formulations. On the other hand, practical equations were found to correlate aw, EC and l with OS °Brix for all the studied cycles. The results obtained showed a relationship in each case that was closely fitted mathematically to a second grade polynomic equation except for aw that was a linear model (Table 6). Despite the good correlation obtained in the case of EC and l it has to be remembered that, as commented on above, the increase in its value is related to the presence of other soluble solutes in the OS, apart from sugar, coming from the fruit. So the equations are only useful for predicting the concentration of the reused osmotic solution. Similar relationships have been described for OS reused in kiwifruit dehydration (Garcı´a-Martı´nez et al., 2002).
4. Conclusions Losses of ascorbic acid, citric acid, minerals and galacturonic acid associated to osmotic dehydration of grapefruit have been detected. Nevertheless, these micronutrients flow to the OS. On the other hand, OS dilution associated to the grapefruit OD under the conditions of this study allow the OS to be reused for at least eight cycles without significant changes in the water activity of the obtained fruit. From this point of view, the reuse of the OS becomes an advantage, as it can be used as an ingredient in new food formulations, with the presence of natural compounds coming from fruit,
The authors thank the Ministerio de Ciencia y Tecnologı´a and to the Fondo Europeo de Desarrollo Regional (FEDER) for the financial support throughout the project AGL2002-01793.
References Ahmed, A. E. R., & Labavitch, J. M. (1977). A simplified method for accurate determination of cell wall uronide content. Journal Food Biochemistry, 1, 361–365. AOAC (1997). Official methods of analysis (16th ed.). Washington, DC: Association of Official Analytical Chemist. Cheftel, J. C., & Cheftel, H. (1986). Introduction a` la biochimie et a` la technologie des aliments. Technique et Documentation Lavoisier, 1, 333–363. Chiralt, A., Martı´nez-Navarrete, N., Martı´nez-Monzo´, J., Talens, P., Moraga, G., Ayala, A., et al. (2001). Changes in mechanical properties throughout osmotic processes. Cryoprotectant effect. Journal of Food Engineering, 49, 129–135. Co´rdoba, A., Garcı´a-Martı´nez, E., Martı´nez-Navarrete, N., Camacho, M. M., & Martı´nez-Monzo´, J. (2003). Enriquecimiento de la disolucio´n osmo´tica en a´cido asco´rbico durante ciclos sucesivos de deshidratacio´n de kiwi. In P. Fito, A. Chiralt, A. Andre´s, & N. Martı´nez-Navarrete (Eds.), Series de Ciencia e Ingenierı´a de Alimentos. Investigacio´n del postgrado del IAD-DTA (pp. 103–121). Valencia: Editorial de la Universidad Polite´cnica de Valencia. De la Fuente, M. A., Montes, F., Guerrero, G., & Jua´rez, M. (2003). Total and soluble contents of calcium, magnesium, phosphorus and zinc in yoghurts. Food Chemistry, 80, 573–578. Fito, P., Andre´s, A., Pastor, R., & Chiralt, A. (1995). Mass transfer behaviour and structure changes during osmotic dehydration (VOD) of foods. In A. Lenart & P. P. Lewicki (Eds.), Osmotic dehydration of fruits and vegetables (pp. 5–19). Warsaw: Warsaw Agricultural University Press. Fito, P., & Chiralt, A. (2000). Vacuum impregnation of plants tissues. In S. M. Alzamora, M. S. Tapia, & A. Lo´pez-Malo (Eds.), Design of minimal processing technologies for fruit and vegetables (pp. 189–205). Maryland: Aspen Publishers. Fito, P., Chiralt, A., Barat, J., Andre´s, A., Martı´nez-Monzo´, J., & Martı´nez-Navarrete, N. (2001). Vacuum impregnation for development of new dehydrated products. Journal of Food Engineering, 49, 297–302. Forni, E., Torreggiani, D., Crivelli, G., Mastrelli, A., Bertolo, G., & Santelli, F. (1987). Influence of osmosis time on the quality of dehydrofrosen kiwifruit. Acta Horticulturae, 282, 425–433. Garcı´a-Martı´nez, E., Martı´nez-Monzo´, J., Camacho, M. M., & Martı´nez-Navarrete, N. (2002). Characterisation of reused osmotic solution as ingredient in new product formulation. Food Research International, 35, 307–313.
R. Peiro´ et al. / Journal of Food Engineering 74 (2006) 299–307 Giangiacomo, R., Torreggiani, D., Erba, M. J., & Messina, G. (1994). Use of osmodehydrofrozen fruit cubes in yoghurt. Italian Journal of Food Science, 6(3), 345–350. Heng, K., Guilbert, S., & Cup, J. (1990). Osmotic dehydration of papaya: Influence of process variables on the product quality. Science des Aliments, 10, 831–848. Hertog, M. G. L., Hollman, P. C. H., & Van de Putte, B. (1993). Content of potentially anticarcinogenic flavonoids of tea infusions, wines, and fruit juices. Journal of Agricultural and Food Chemistry, 41(8), 1242–1246. Kitner, P. K., III, & Van Buren, J. P. (1982). Carbohydrate interference and its correction in pectin analysis using the mhydroxydiphenil method. Journal Food Science, 47, 756–759, 764. Liu, Ming, Li, Xin-Qi, Weber, C., Lee, Chang-Yong, Brown, J., & Liu, Rui-Hai (2002). Antioxidant and antiproliferative activities of raspberries. Journal of Agricultural and Food Chemistry, 50(10), 2926–2930. Martı´nez-Monzo´, J., Martı´nez-Navarrete, N., Chiralt, A., & Fito, P. (2001). Combined vacuum impregnation and osmotic dehydration in fruit cryoprotecction. In P. Fito, A. Chiralt, J. M. Barat, W. E. L. Spiess, & D. Behsnilian (Eds.), Osmotic dehydration and vacuum impregnation. Application in food industries (pp. 61–78). Lancaster: Technomic Publishing Company. Martı´nez-Navarrete, N., Andre´s, A., Chiralt, A., & Fito, P. (2000). Termodina´mica y cine´tica de sistemas alimento-entorno (pp. 116– 132). Valencia: Servicio de Publicaciones de la Universidad Polite´cnica de Valencia. Mataix. (1998). Tabla de composicion de alimentos. Ed. Universidad de Granada. Moraga, M. J. (2004). Pomelo mı´nimamente procesado por deshidratacio´n osmo´tica. Cambios en las propiedades fı´sicas durante el almacenamiento. PhD thesis, Universidad Polite´cnica de Valencia.
307
Panagiotou, N. M., Karathanos, V. T., & Maroulis, Z. B. (1998). Mass transfer modelling of the osmotic dehydration of some fruits. International Journal of Food Science and Technology, 33, 267–284. Pancoast, H. M., & Junk, W. R. (1980). Handbook of sugars (p. 46). Westport: Avi Publishing Company. Pinnavaia, G., Dalla-Rosa, N. I., & Lerici, C. R. (1988). Dehydrofreezing of fruit using direct osmosis as concentration process. Acta Alimentaria Polonica, 14, 51–57. Prior, R. L., & Guohua, Cao. (2000). Antioxidant phytochemicals in fruits and vegetables: Diet and health implications. HortScience, 35(4), 588–592. Raoult-Wack, A. L. (1994). Recent advances in osmotic dehydration of foods. Trends in Food Science and Technology, 5, 255–260. Taiwo, K. A., Angersbach, A., Ade-Omowaye, B. I. O., & Knorr, D. (2001). Effects of pre-treatments on the diffusion kinetics and some quality parameters of osmotically dehydrated apple slices. Journal Agriculture Food Chemistry, 49, 2804–2811. Torreggiani, D., & Bertolo, G. (2001). Osmotic pre-treatments in fruit processing: Chemical, physical and structural effects. Journal of Food Engineering, 49, 247–253. Valdez-Fragoso, A., Welti-Chanes, J., & Giroux, F. (1998). Properties of sucrose solution reused in osmotic dehydration of apples. Drying Technology, 6(7), 1429–1445. Venant, A., Borrel, S., Mallet, J., & Van Neste, E. (1989). Gel permeation chromatography as a method for the simultaneous clean up of organochlorine, organophosphate and polychlorinated biphenyl residues in fat extracts. Analusis, 17, 64–68. Wargovich, M. J. (2000). Anticancer properties of fruits and vegetables. HortScience, 35(4), 573–575. Wolfe, K. L., & Liu, Rui-Hai (2003). Apple peels as a value-added food ingredient. Journal of Agricultural and Food Chemistry, 51(6), 1676–1683.