Journal of Food Engineering 80 (2007) 914921 www.elsevier.

com/locate/jfoodeng

A membrane-based process for the clarication and the concentration of the cactus pear juice
A. Cassano
a

a,*

, C. Conidi a, R. Timpone b, M. DAvella b, E. Drioli

Institute on Membrane Technology, ITM-CNR, c/o University of Calabria, via P. Bucci, cubo 17/C, I-87030 Rende (Cosenza), Italy b Citrech Snc, Projects, Applications Consulting for Food and Citrus Industry, Viale Regina Margherita, 61 I-98121 Messina, Italy Received 31 May 2006; received in revised form 8 August 2006; accepted 9 August 2006 Available online 6 October 2006

Abstract Cactus pear fruit is a food of nutraceutical and functional importance. Unfortunately, the low acidity and the high soluble solids content make the pulp of this fruit a very attractive media for growth of microorganisms requiring a thermal treatment (115.5 C or higher) to obtain a good control of the microbial invasion. A relatively long thermal treatment (100 C for 20 min) can produce a good microbiological stability; however the nal product is characterised by an unattractive hay taste and does not resemble the original fresh juice due to changes in colour and avour. The aim of this study was to study the potentiality of a membrane-based process for the clarication and the concentration of the cactus pear fruit juice. The fresh juice, with a total soluble solids (TSS) content of about 11 Brix, was previously claried by an ultraltration (UF) step, on laboratory scale, according to the batch concentration mode. The claried juice was then concentrated by osmotic distillation (OD) up to a TSS content of 61 Brix at 28 C. An initial evaporation ux of 1.16 kg/m2 h was obtained using a calcium chloride dehydrate solution at 60 w/w% as stripper. The juice quality was analysed in terms of total antioxidant activity (TAA), ascorbic, citric and glutamic acid, betalains and viscosity in order to evaluate the eects of the membrane processes on the quality and composition of the juice. 2006 Elsevier Ltd. All rights reserved.
Keywords: Cactus pear juice; Ultraltration; Osmotic distillation; Betalains; Total antioxidant activity

1. Introduction Cactus pears (Opuntia cus-indica) originated in tropical America (Russel & Felker, 1987) and can be found in subtropical and tropical areas (Barbera, 1995). Fruits are eshy berries, varying in shape, size and colour containing hard seeds. They have a great interest because of their nutritional and antioxidant properties mainly due to the presence of ascorbic acid, bers and aminoacids (Stintzing, Schieber, & carle, 2001). In particular the level of ascorbic acid can reach values of 40 mg/ 100 g (Rodriguez, Orphee, Macias, Generoso, & Gomes Garcia, 1996); Mineral salts such as calcium and phospho*

Corresponding author. Tel.: +39 0984 492067; fax: +39 0984 402103. E-mail address: a.cassano@itm.cnr.it (A. Cassano).

rous are present at levels of 15.432.8 mg/100 g and 12.8 27.6 mg/100 g, respectively. Potassium is another important source of minerals contained at levels of about 217 mg/100 g (Sepulveda & Saenz, 1990). Betalains are pigments of cactus fruits with high molar extintion coecients and with a colouring power competitive to synthetic colorants. They must be preserved during processing since they are responsible for the nal attractive colour of the product. The cactus pear fruit is classied in the low-acid group (pH > 4.5) requiring a thermal treatment of 115.5 C, or greater, to obtain a good control of microorganisms. The pH, low acidity and high soluble solids content make cactus-pear pulp susceptible to microbial attack, thus limiting its storage life (Joubert, 1993). In this juice, several lengthy thermal treatments (100 C for 20 min) permit to produce

0260-8774/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jfoodeng.2006.08.005

A. Cassano et al. / Journal of Food Engineering 80 (2007) 914921

915

Nomenclature Jw T aw x TSS Q evaporation ux (kg h1 m2) temperature (C) water activity solute mass fraction (w/w%) total soluble solids (Brix) ow rate (l/h) Daw l water activity dierence dynamic viscosity (Pa s)

Greek symbols DP transmembrane pressure (bar)

Subscripts b brine j juice p permeate r retentate

bottled juice with a good stability; however the nal product did not resemble the original fresh juice due to changes in colour and avour other than an unattractive hay taste (Carrandi, 1995). In order to preserve the juice Espinosa et al. (1973) added lemon juice reducing its pH to 4.0 and then the blend was submitted to a mild thermal treatment (80 C for 20 min): however the juice preserved in these conditions presented acetic fermentation and the juice could not preserved for a long time. The increasing market demand for this fruit has lead researchers to develop processed items with increased shelf-life able to retain as much as possible the peculiarity of fresh fruit, as well as colour, aroma, nutritional value and structural characteristics. Another possibility related to cactus pear juices was concentrated juice production. The lower water activity of the concentrated juice relative to the natural juice is a clear protection against the growth of microorganisms and can extend the shelf-life of the juice. In addition, package, storage and costs are remarkably reduced. Concentrated juices with 6367 Brix were obtained by using an Alfa-laval centrifuge vacuum evaporator, at approximately 40 C. The stability of the juice against microorganisms growth was good, but the sensorial analyses found the acceptability was only 5.0 (19 point scale). This unsatisfactory ranking was due to damage to the colour and the herbaceous aroma that appeared after the concentration process (Saenz, Sepulveda, Araya, & Calvo, 1993). In addition colour changes were observed during thermal treatment in pasteurised and concentrated juices of green cactus pear. Alternative methods of concentration, such as freeze concentration, require less energy, but they are expensive and limited in the degree of achieved concentration (Koseoglu, Lawhon, & Lusas, 1990). The basic properties of membrane operations make them ideal in the production of fruit juices with high quality, natural fresh taste and additive-free. They are generally athermal and do not involve phase changes or chemical additives; besides, they are simple in concept and operation and characterised by low energy consumption. Juice clarication, stabilisation, depectinization and concentration are typical steps where membrane processes as microltration (MF), ultraltration (UF), nanoltration

(NF) and reverse osmosis (RO) have been successfully uti lised (Alvarez, Riera, Alvarez, & Coca, 1998; Gokmen, Borneman, & Nijhuis, 1998; Todisco, Tallarico, & Drioli, 1998). In particular, UF and MF represent a valid alternative to the use of traditional ning agents (gelatine, bentonite and silica sol) and lter aids which cannot be reused and cause pollution problems due to their disposal (Eykamp, 1995). In these processes the juice is separated into a brous concentrated pulp and a claried fraction free of spoilage microorganisms. Membrane concentration processes such as reverse osmosis (RO), membrane distillation (MD) and osmotic distillation (OD) are able to overcome some of the problems associated with thermal evaporation. RO has achieved some commercial success in the fruit juice concentration: it presents the advantages of a lower thermal damage to the product, reduction in energy consumption and lower capital equipment costs. However the nal concentration of juices is limited to about 25 30 Brix due to the high osmotic pressure of the feed at those levels. So it could be employed as a preliminary step in a two-stage concentration (Barbe, Bartley, Jacobs, & Johnson, 1998). MD, in comparison with pressure-driven membrane processes, permits to reach higher concentrations of dissolved solids (6065 Brix); however, the requirement to heat the feed stream to maintain the water vapour pressure gradient can be the cause of a signicant loss of organic volatiles. Concentration by OD is an emerging membrane technique based on the use of an hydrophobic microporous membrane to separate two liquid phases that dier greatly in terms of solute concentration (Alves, Koroknai, Bela Bako, & Coelhoso, 2004; Alves & Coelhoso, 2006; Cassano et al., 2003; Hogan, Canning, Peterson, Johnson, & Michaels, 1998; Jiao, Cassano, & Drioli, 2004; Lebfevre, 1988; Shaw et al., 2001; Vaillant et al., 2001). The hydrophobic nature of the membrane prevents penetration of the pores by aqueous solutions, creating air gaps within the membrane. The dierence in solute concentration, and consequently in water activity between the two sides of the membrane, induces a vapour pressure dierence causing a water vapour transfer across the pores from

916

A. Cassano et al. / Journal of Food Engineering 80 (2007) 914921

high-vapour pressure phase to the low one. Since the driving force is not a hydraulic pressure dierence, concentrations higher than 60 Brix can be achieved. Moreover, OD is not limited by the osmotic pressure of the feed and can be carried out at low temperatures, thus improving preservation of volatile compounds. The aim of this study was to evaluate, on laboratory scale, the potential of UF and OD processes for clarifying and concentrating fresh cactus pear juice. The performance of the membrane-based process was evaluated on the basis of the quality of the products through the analytical measurement of the parameters which characterise the high nutritional and organoleptic properties of the fresh juice. 2. Materials and methods

pumps, was used to maintain the temperature of the feed juice constant. The juice was claried according to a batch concentration procedure (the permeate is collected separately and retentate is recycled to the feed tank) at a transmembrane pressure of 2.15 bar, an axial retentate ow rate of 18 l/h and a temperature of 25 2 C. The UF system was operated to clarify the juice up to a volume reduction factor (VRF, dened as the ratio between the initial feed volume and the volume of the resulting retentate) of 2.92. The UF process produced two fractions: a claried and sterile juice (permeate) and a brous concentrated pulp one (retentate). The former was submitted to a concentration step performed by OD. 2.3. OD experimental set-up

2.1. Juice extraction O. cus-indica (L.) Mill. fruits cv. Gialla (orangeyellow) collected in Sicily were purchased from a local market. Fruits (104 kg) were manually washed in water and then peeled by hand, with a knife, obtaining 49 kg of peeled fruits (yield 47.1%). Seeds and mesocarp bres were removed with a squeezer and then washed with water. The raw puree (40.7 kg) and the cleaning solution (8.7 kg) were pooled. An acid pectinase (Natuzym CCL, Add Food Service GmbH, Ascheberg, Germany) was added in quantity of 1.8 g/kg and the puree was kept for 4 h at room temperature in plastic tanks without modication of the pH. The puree was nally ltered through a 200 lm lter before the clarication step. This method gave an average juice yield of 100 w/w% starting from the peeled fruits. 2.2. Clarication procedure Cactus pear juice was claried by using a laboratory pilot unit equipped with two hollow bre membrane modules (Koch/Romicon HF 25-60-PM10) supplied by KochGlitsch Italia S.r.l. (Milan, Italy). Characteristics of the membranes are reported in Table 1. The equipment consists of a feed tank, two feed pressure pumps, four manometers located at the inlet (Pin) and outlet (Pout) of each membrane module and two magnetic ow meters for the measure of the retentate (Qr) and permeate (Qp) ow rate. A tube and shell heat exchanger, placed after the feed
Table 1 Characteristics of Koch ultraltration membrane module Membrane type Conguration Membrane polymer Nominal molecular weight cut-o Membrane area Approximate number of bers Fiber inside diameter pH operating range Maximum operating temperature Maximum inlet pressure Koch/Romicon HF 25-60-PM 10 Hollow bre Polysulphone 10 kDa 2.3 m2 460 1.5 mm 1.513 40 C (at pH 6) 2.1 bar (at 25 C)

The concentration process of the claried cactus pear juice was carried out by using an OD laboratory bench plant supplied by Hoechst-Celanese Corporation (Wiesbaden, Germany) equipped with a Liqui-Cell Extra-Flow 2.5 800 membrane contactor (Hoechst-Celanese Corporation, Wiesbaden, Germany) (Table 2). The juice, with an initial concentration of 11 Brix, was pumped through the shell side of the membrane module, while in the bre lumens (tube side) owed a 60 w/w% calcium chloride dihydrate (Fluka Chemie GmbH, Buchs, Switzerland) solution, in a countercurrent mode, using two independent magnetic drive gear pumps with variable motor velocity. Both solutions were recirculated back to their reservoirs, after passing through the contactor, at a temperature of 28 2 C. The initial weight of the stripping solution (generally 8 kg) was two times higher compared to that of the juice, in order to prevent a signicant dilution with consequent decreasing of the driving force during the process. Inlet and outlet pressures for both tube side and shell side streams were registered by pressure gauges in order to control the pressure dierentials between the two sides of the membrane. OD system was generally operated with
Table 2 Data sheet of Liqui-Cel Extra-Flow 2.5 8 in membrane contactor Fibre characteristics Fibre type External diameter Internal diameter Length Cartridge operating limits Maximum transmembrane dierential pressure Maximum operating temperature range Cartridge characteristics Cartridge dimensions (DxL) Eective surface area Eective area/volume Fiber potting material Celgard microporous polypropylene hollow bre 300 lm 220 lm 0.16 m 4.2 kg/cm2 (60 psi) 40 C (104 F)

8 28 cm (2.5 8 in) 1.4 m2 (15.2 ft2) 29.3 cm2/cm3 Polyethylene

A. Cassano et al. / Journal of Food Engineering 80 (2007) 914921

917

a slightly higher pressure on the shell side (0.59 bar) of the module than the lumen side (0.11 bar) in order to avoid the leakage of the brine strip into the product. The ow rates of extracted water, at various points during the concentration process, were calculated by measuring the weight loss of the juice over time by a balance (Gibertini Elettronica, Milan, Italy) placed under the juice tank. Flow rates normalised by the surface area of the membrane (1.4 m2) gave the evaporation ux (Jw) values. In Fig. 1 the scheme of the concentrated juice production, starting from cactus pear fruits, is reported. 2.4. Juice analyses Juice samples (permeates and retentates) coming from the UF and OD experiments were collected and stored at 20C and then submitted to analytical procedures. The Total Antioxidant Activity (TAA) was determined applying an improved version of the ABTS radical cation decolourisation assay (Rice-Evans & Miller, 1994) in which
Cactus pear fruits

Washing Peeling Peel (55 kg) Peeled fruits (49 kg)

Scrap

Squeezing

the radical monocation of the 2,2 0 -azino-bis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS+) is generated by oxidation of ABTS with potassium persulphate before the addition of the antioxidant (Re et al., 1999). The decolourisation of the blue/green (ABTS+) chromofore is measured as the percentage inhibition of absorbance at 734 nm and calculated relative to the reactivity of Trolox (6-hydroxy2,5,7,8-tetramethylchroman-2-carboxylic acid) as standard under the same conditions. The (ABTS+) solution was diluted with PBS (phosphate buered saline), pH 7.4, to an absorbance of 0.70 (0.02) at 734 nm and equilibrated at 30 C. After addition of 1.0 ml of diluted (ABTS+) solution to 10 ll of sample the absorbance reading was taken exactly 1 min after initial mixing and up to 6 min. Samples were analysed at three dierent dilutions, within the linearity range of the assay. Results were expressed as Trolox Equivalents Antioxidant Capacity (TEAC). Retentate samples coming from OD were diluted to the same Total Soluble Solids (TSS) concentration of the fresh juice (13 Brix) before the analysis, in order to allow the direct comparison between the dierent values. ABTS, potassium persulphate and Trolox were obtained from Sigma-Aldrich (Milan, Italy). Quantication of betalains was carried out in triplicate in deionised water without pH adaptation applying the molar extinction coecients of betanin (e = 60,000 l/mol cm in H2O; k = 538 nm; MW = 550 g/mol) and indicaxanthin (e = 48,000 l/mol cm in H2O; k = 480 nm; MW = 308 g/mol). The juice was diluted with deionised water to obtain absorption values of 0.8 6 A 6 1.0. The betalain content (BC), expressed as mg/l, was calculated by using the following equation: BC A DF MW 1000 eL

Washing

Puree (40.7 kg)

Recovered solution (8.7 kg) Mixing (49.4 kg)

Incubation with CCL (1.8 g/l, room temperature, 4h)

Sieving (200 m)

Cactus pear juice

ULTRAFILTRATION Permeate (32.5 kg) Concentrated juice (5.8 kg) Retentate (16.9 kg)

OSMOTIC DISTILLATION

Fig. 1. Scheme of the production of claried and concentrated cactus pear juice.

where A is the absorption at 538 and 480 nm for betacyanins and betaxanthins, respectively; DF is the dilution factor and L the pathlength of the 1-cm cuvette. For MW and e, the molecular weights and extinction coecients of the representative compounds betanin and indicaxanthin have to be considered, respectively (Cay & Corke, 1999). Spectrophotometric measurements were performed by an UVVisible Recording spectrophotometer (UV-160 A, Shimadzu Scientic Instruments, Inc., Japan) at 30 C. Ascorbic, glutamic and citric acid were determined by HPLC by using an HPLC-system (Agilent 1100 Series, USA) equipped with an UV detector. The following conditions were used: an Alltima C18 HP5U column, 5lm, 250x4.6 mm (Alltech Associates, Inc., Deereld, IL) with a mobile phase of H3PO4 0.05 M, ux = 0.7 ml/min, T = 25C, pressure = 85 bar, k 205 nm. The external standard method was applied. For each organic acid a calibration curve with three standard concentrations was constructed. TSS measurements were carried out using hand refractometers (Atago Co., Ltd., Tokyo, Japan) with scale range of 032, 2862 and 5890 Brix.

918

A. Cassano et al. / Journal of Food Engineering 80 (2007) 914921

Viscosity was measured using a RFS III viscometer (Rheometric Scientic, USA). pH was measured by an Orion Expandable ion analyzer EA 920 pH meter (Allometrics, Inc. LA, USA) with automatic temperature compensation. The water activity of the cactus pear juice (awj) was estimated by correlating the data published in the literature for glucose solutions at various solute concentrations and at a temperature of 25 C (Peng, Chow, & Chan, 2000). The water activity of the brine solution (awb) was estimated by correlating the data published in the literature for CaCl2 solutions at various solute concentrations and at a temperature of 25 C (Courel, 1999). 3. Results and discussion 3.1. Osmotic distillation

0.7

0.6

0.5

a w

0.4

0.3

0.2

0.1 0 50 100 150 200 250 300

Time (min)

1.4

70 60
Jw TSS

1.2

50 40

Jw ( kg/m2 h)

Fig. 2 shows experimental results concerning an OD process in which the claried cactus pear juice, with a TSS concentration of 11 Brix (aw = 0.985), was concentrated up to 61.4 Brix (aw = 0.828). The juice and the brine were pumped through the shell and the tube side of the membrane module, respectively, at a ow rate of 33 l/h; both solutions were recirculated back to their reservoirs, after passing through the contactor, at a temperature of 28 2 C. Transmembrane pressure was xed at 0.48 bar. The driving force for water transport is sustained by the water activity dierence between the juice and the brine (Daw = awj awb) (Fig. 2a). At rst the brine concentration was 60 w/w% giving rise to an evaporation ux of about 1.16 kg/m2 h. In the range 0175 min the decrease of the evaporation ux had a behavior very similar to those observed for the dilution of the stripping solution while the juice viscosity remained almost constant (Fig. 2b and c). In this range a 48% reduction of the evaporation ux, with respect to the initial value, was observed. In the tighter range 175255 min, starting from a TSS concentration of 33 Brix, a more rapid decline of the evaporation ux (55%) was observed. In this range the ux decay is more attributable to the exponential increasing of the juice viscosity. The break-even point of the viscosity curve with respect to concentration corresponds to this TSS value. On the other hand, at TSS concentrations lower than 33 Brix, the evaporation ux decay can be mainly attributed to the dilution of the stripper. These observations conrm data reported in literature (Cassano, Jiao, & Drioli, 2004; Courel, Dornier, Herry, Rios, & Reynes, 2000; Vaillant et al., 2001) concerning the concentration of passion fruit, kiwifruit and sucrose solution by OD: at low TSS of the feed the ux decay is more attributable to the dilution of the stripper; at higher TSS concentration values it mainly depends on juice viscosity and, consequently, on juice concentration and temperature. Previous experiments, in which the ow rates of the stripping solution were modied in the range 1833 l/h, showed an increasing of the evaporation ux from 0.56

0.8 30 0.6 20 0.4 10 0 300

0.2 0 50 100 150 200 250

Time (min)

65

180

60

xb j

160 140 120 100

55

x b (w/w%)

50 80 45 60 40 40 20 35 0 50 100 150 200 250 0 300

Time (min)

Fig. 2. Concentration of claried cactus pear juice by osmotic distillation. Time course of: (a) water activity dierence between juice and brine; (b) evaporation ux and TSS; (c) brine concentration and cactus pear fruit juice viscosity (Operating conditions: DP = 0.48 bar; T = 28 2 C; Qj = 33 l/h; Qb = 33 l/h).

to 1.16 kg/m2 h (more than 50%) in the same experimental conditions (T = 28 2C; DP = 0.48 bar; Qj = 33 l/h). This result can be attributed to a stronger shear stress along the condensation side of the membrane when the circulation velocity of the salt solution is increased. In Fig. 3 the eect of the total soluble solids content on the juice viscosity is shown: as can be seen the juice viscosity increases exponentially with the increasing of the total soluble solids.

j (mPa s)

TSS (Brix)

1.0

A. Cassano et al. / Journal of Food Engineering 80 (2007) 914921

180 160 140 120
8 10

919

TAA (mM Trolox)

0 10 20 30 40 50 60 70

j (mPa s)

100 80 60 40 20 0

0 10 20 30 40 50 60 70

TSS (Brix)

TSS (Brix)

Fig. 3. Eect of TSS on the cactus pear juice viscosity (T = 25 C).

Fig. 4. Total antioxidant activity of cactus pear fruit juice vs. TSS during the OD process. Table 4 Spectrophotometric quantication of betaxanthins and betacyanins in samples of cactus pear juice coming from UF/OD treatment Sample Feed UF Retentate UF Permeate UF Retentate OD Retentate OD Retentate OD Retentate OD Bx (SD) (mg/l) 57.1 61.6 53.4 52.7 53.1 53.7 52.5 (0.7) (0.7) (3.4) (0.2) (1.9) (0.4) (0.7) Bc (SD) (mg/l) 19.9 32.8 12.7 11.3 11.0 12.2 12.4 (0.7) (0.4) (0.2) (0.6) (0.1) (0.1) (0.2) Bx (%) 74.2 65.3 80.8 82.3 82.8 81.5 80.9 Bc (%) 25.8 34.7 19.2 17.7 17.2 18.5 19.1

3.2. Analytical evaluations In Table 3 the evaluation of TSS, TAA and ascorbic, citric and glutamic acid in samples coming from the UF/OD treatment is shown. Only a 4% loss in the TAA was found in the UF permeate with respect to the fresh juice. During the concentration by OD the TAA of the juice remained almost constant independently by the achieved level of the TSS (Fig. 4). A 5% rejection of the UF membrane towards the ascorbic acid and citric acid was measured. The reduction of these components in the nal retentate of the OD process was only 3.5 and 5%, respectively, with respect to the claried juice. For the glutamic acid the UF membrane showed a 8% rejection: the concentration of this component in the retentate of the OD process remains constant independently by the achieved level of the concentration. Betalains concentration are economically important with respect to their potential use as colorants. In Table 4 the measurement of their concentration in the samples coming from the UF/OD process is reported. In the fresh juice submitted to the UF process 57.1 and 19.9 mg/l of betaxanthins and betacyanins were found, respectively. These values correspond to those measured by Stintzing et al. (2001) in the juice of the O. cus-indica cv. Gialla (53.7 for betaxanthins and 19.0 for betacyanins). The rejection of the UF membrane towards betaxanthins and betacyanins were 6.5% and 36%, respectively. During the OD

1 2 3 4

Bx = betaxanthins, Bc = betacyanins, Means of triplicate analyses, SD = standard deviation.

process the levels of betalains and the ratio betacyanins/ betaxanthins (1:4) remained constant independently by the concentration of the achieved TSS level. In Table 5 the mass balance of the UF process is reported. The balance is quite equilibrate for most analysed parameters: losses can be ascribed to experimental errors in their determination and to the interactions of the juice compounds with the UF membrane which are responsible of the fouling phenomenon. On the basis of the results obtained on laboratory scale it is possible to suggest an integrated membrane process for producing cactus pear juice concentrate with high nutritional value. In this process the UF and OD steps represent

Table 3 Analytical evaluations on samples of cactus pear juice coming from UF/OD treatment Sample Feed UF Retentate UF Permeate UF Retentate OD Retentate OD Retentate OD Retentate OD TSS (Brix) 13.4 14.1 13.0 25.6 38.8 45.0 58.0 TAA (mM Trolox) 5 0.6 4.9 1.1 4.8 0.8 4.7 0.4 4.5 0.9 4.4 0.7 4.6 0.3 Ascorbic acid (mg/L) 39.3 43.0 37.3 35.0 30.0 36.0 36.0 Citric acid (mg/L) 416.0 427.4 395.0 370.0 365.0 370.0 375.0 Glutammic acid (g/L) 2.06 1.95 2.05 2.05 2.10 2.08 2.05

1 2 3 4

920 Table 5 Mass balance of the UF process Feed Volume TSS Ascorbic acid Citric acid Glutamic acid Betaxanthins Betacyanins TAA

A. Cassano et al. / Journal of Food Engineering 80 (2007) 914921

Permeate 32.5 l 4225.0 g 1212.2 mg 12837.5 mg 66.6 g 1735.5 mg 412.7 mg 156.0 mmol Trolox 65.8% 63.8% 62.4% 62.4% 65.5% 61.5% 42.0% 63.1%

Retentate 16.9 l 2394.6 g 726.7 mg 7223.0 mg 32.955 g 1041.0 mg 554.32 mg 82.8 mmol Trolox 34.2% 36.2% 37.4% 35.1% 32.4% 36.9% 56.4% 33.5%

Balance 100.0% 100.0% 99.8% 97.5% 97.9% 98.4% 98.4% 96.6%

49.4 l 6619.6 g 1941.4 mg 20550.4 mg 101.7 g 2820.7 mg 983.0 mg 247.0 mmol Trolox

a valid alternative to the traditional clarication and concentration procedures carried out by using ning materials (gelatine, bentonite, silica sol) and thermal evaporation, respectively. The residual brous phase coming from the UF process (retentate) could be submitted to a stabilising treatment (pasteurisation, ohmic heating, high pressures) and successively added, together with the water, to the nal OD concentrate for the preparation of bres enriched beverages (Cassano et al., 2004). Besides, the nal retentate of the OD process is a good source of antioxidants and it can be used in foods and nutritional supplement formulations. Finally, the high betalain concentration achieved in the OD retentate (227 mg/l for betaxanthins and 54 mg/l for betacyanins) makes it a promising future source for colouring foodstu. 4. Conclusions Cactus juices combine an attractive appearance with nutritionally valuable compounds. The clarication and concentration of the cactus pear juice by using UF and OD techniques represent a valid approach to process the juice at low temperatures preserving the organoleptic, nutritional and sensorial characteristics of the fresh fruit. Analytical results of the UF process conrm the possibility to recover the most part of ascorbic, glutamic and citric acid in the claried juice. The rejection of the UF membrane towards betaxanthins was lower than the rejection measured for betacyanins. Only a 4% loss in the TAA was found in the UF permeate with respect to the fresh juice. In the OD process the claried juice with a TSS content of about 11 Brix was concentrated up to 61 Brix. An initial evaporation ux of 1.16 kg/m2 h was obtained using a calcium chloride dehydrate solution at 60 w/w% as stripper. All the analytical parameters determined in the claried juice remain constant during the OD process. In particular the nal retentate of the OD process is a good source of antioxidants and it can be used in foods and nutritional supplement formulations. The high betalain concentration makes it a promising future source for colouring foodstu. Finally, the OD retentate can be added to the water and to

the UF retentate for the preparation of bres enriched beverages. Based on the above the UF/OD integrated process permits to preserve the nutraceutical and functional importance of the cactus pear fruit juice with respect to the traditional clarication and concentration procedures. References
Alvarez, S., Riera, F. A., Alvarez, R., & Coca, J. (1998). Permeation of apple aroma compounds in reverse osmosis. Separation and Purication Technology, 14, 209220. Alves, V. D., & Coelhoso, I. M. (2006). Orange juice concentration by osmotic evaporation and membrane distillation: a comparative study. Journal of Food Engineering, 74, 125133. Alves, V. D., Koroknai, B., Bela-Bako, K., & Coelhoso, I. M. (2004). Using membrane contactors for fruit juice concentration. Desalination, 162, 263270. Barbe, A. M., Bartley, J. P., Jacobs, A. L., & Johnson, R. A. (1998). Retention of volatile organic avour/fragrance components in the concentration of liquid foods by osmotic distillation. Journal Membrane Science, 145, 6775. Barbera, G. (1995). History, economic and agro-ecological importance. In Barbera, G., Inglese, P., & Pimienta-Barrios, P. (Eds.), Agro-ecology cultivation and uses of cactus pear (Vol. 132, pp. 111). FAO plant production and protection paper. Carrandi, L. (1995). Efecto de conservantes en la estabilidad de jugo de tuna pasteurizado. Memoria de Titulo. Ing. Agronomo, Facultad de Ciencias Agrarias y Forestales, Universidad de Chile (p. 61). Cassano, A., Drioli, E., Galaverna, G., Marchelli, R., Di Silvestro, G., & Cagnasso, P. (2003). Clarication and concentration of citrus and carrot juices by integrated membrane processes. Journal of Food Engineering, 57(2), 153163. Cassano, A., Jiao, B., & Drioli, E. (2004). Production of concentrated kiwifruit juice by integrated membrane processes. Food Research International, 37(2), 139148. Cay, Y. Z., & Corke, H. (1999). Amaranthus betacyanin pigments applied in model food systems. Journal of Food Science, 64, 869873. Courel, M. (1999). Mass transfer study in osmotic evaporation: application to fruit juice concentration, Ph.D. Thesis, University of Montpellier II, France. Courel, M., Dornier, M., Herry, J. M., Rios, G. M., & Reynes, M. (2000). Eect of operating conditions on water transport during the concentration of sucrose solutions by osmotic distillation. Journal of Membrane Science, 170, 281289. Espinosa, J., Borrocal, R., Jara, M., Zorrilla, C., Zanabria, C., & Medina, J. (1973). Quelques proprietes et essais preliminaires de conservation des fruits et du jus de gue de barbarie (Opuntia cus indica). Fruits, 28, 285289. Eykamp, W. (1995). Microltration and ultraltration. In R. D. Noble & S. A. Stern (Eds.), Membrane Separation Technology Principles and Applications (pp. 37). Amsterdam: Elsevier.

A. Cassano et al. / Journal of Food Engineering 80 (2007) 914921 Gokmen, V., Borneman, Z., & Nijhuis, H. H. (1998). Improved ultraltration for colour reduction and stabilization of apple juice. Journal of Food Science, 63(3), 504507. Hogan, P. A., Canning, R. P., Peterson, P. A., Johnson, R. A., & Michaels, A. S. (1998). A new option: osmotic distillation. Chemical engineering progress(July), 4961. Jiao, B., Cassano, A., & Drioli, E. (2004). Recent advances on membrane processes for the concentration of fruit juices: a review. Journal of Food Engineering, 63(3), 303324. Joubert, E. (1993). Processing of the fruit of ve prickly pear cultivars grown in South Africa. International Journal of Food Science and Technology, 28, 377387. Koseoglu, S. S., Lawhon, J. T., & Lusas, E. W. (1990). Use of membranes in citrus juice processing. Food Technology, 44(12), 9097. Lebfevre, M. S. M. (1988). Method of performing osmotic distillation. Patent no. US 4 781 837. Peng, C., Chow, A. H. L., & Chan, C. K. (2000). Hygroscopic study of glucose, citric acid and sorbitol using an electrodynamic balance: comparison with UNIFAC predictions. Aerosol Science and Technology, 35, 753758. Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., & RiceEvans, C. A. (1999). Antioxidant activity applying and improved ABTS radical cation decolorization assay. Free Radical Biology and Medicine, 26, 12311237. Rice-Evans, C. A., & Miller, N. J. (1994). Total antioxidant status in plasma and body uids. Methods in Enzimology, 234, 279298.

921

Rodriguez, S., Orphee, C., Macias, S., Generoso, S., & Gomes Garcia, L. (1996). Tuna: proprietades sico-quimicas de dos variedades. La Alimentacion Latinoamericana, 210, 3437. Russel, C. H., & Felker, P. (1987). The prickly pears (Opuntia spp Cactaceae) A source of human and animal food in semiarid regions. Economic Botany, 41, 433445. Saenz, C., Sepulveda, E., Araya, E., & Calvo, C. (1993). Colour changes in concentrated juices of prickly pear Opuntia cus indica during storage at dierent temperatures. Lebensmittel-Wissens chaft und Technologie, 26, 417421. Sepulveda, E., & Saenz, C. (1990). Chemical and physical characteristics of prickly pear (Opuntia cus indica) pulp. Revista de Agroquimica Y Tecnologia de Alimentos, 30, 551555. Shaw, P. E., Lebrun, M., Dornier, M., Ducamp, M. N., Courel, M., & Reynes, M. (2001). Evaluation of concentrated orange and passionfruit juices prepared by osmotic distillation. Lebensmittel-Wissenschaft und Technologie, 34, 6065. Stintzing, F. C., Schieber, A., & carle, R. (2001). Phytochemical and nutritional signicance of Cactus pear. Food Research and Technology International, 212, 396407. Todisco, S., Tallarico, P., & Drioli, E. (1998). Modelling and analysis of ultraltration eects on the quality of freshly squeezed orange juice. Italian Food & Beverage Technology, XII(May), 38. Vaillant, F., Jeanton, E., Dornier, M., O Brien, G. M., Reynes, M., & Decloux, M. (2001). Concentration of passion fruit juice on an industrial pilot scale using osmotic evaporation. Journal of Food Engineering, 47, 195202.