Accepted Manuscript

Concentration of cranberry juice by osmotic distillation process

C. Zambra, J. Romero, L. Pino, A. Saavedra, J. Sanchez

PII: S0260-8774(14)00309-4

DOI: http://dx.doi.org/10.1016/j.jfoodeng.2014.07.009

Reference: JFOE 7868

To appear in: Journal of Food Engineering

Received Date: 9 April 2014

Revised Date: 10 July 2014

Accepted Date: 12 July 2014

Please cite this article as: Zambra, C., Romero, J., Pino, L., Saavedra, A., Sanchez, J., Concentration of cranberry

juice by osmotic distillation process, Journal of Food Engineering (2014), doi: http://dx.doi.org/10.1016/j.jfoodeng.

2014.07.009

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Concentration of cranberry juice by osmotic

distillation process C. Zambraa,b*, J. Romeroc, L. Pinoc, A. Saavedrac, J. Sanchezd

a) University Arturo Prat, Av. Arturo Prat 2120, Iquique, Chile. Zip code: 1100000. Email:

carlos.zambra.s@gmail.com. Phone: 56-9-65214386.

b) Center of Studies of Processed Food (CEAP), Av. San Miguel 3425, Talca, Chile. Zip code: 3460000.

Email: carlos.zambra.s@gmail.com. Phone: 56-9-65214386.

c) Laboratory of Membrane Separation Processes (LabProSeM), Department of Chemical Engineering,

University of Santiago de Chile (USACH), Av. Lib. Bdo. O’Higgins 3363, Estación Central, Chile. Zip code:

8320000. Email: julio.romero@usach.cl.

d) Institut Européen des Membranes, Université de Montpellier 2, cc 047, 2 Place Bataillon, 34095,

Montpellier Cedex 5, France. Email : Jose.Sanchez-Marcano@univ-montp2.fr

*Corresponding author: Center of Studies of Processed Food (CEAP), Av. San Miguel 3425, Talca, Chile.

Zip code: 3460000. email: carlos.zambra.s@gmail.com. Phone: 56-9-65214386

2

Abstract

An Osmotic Distillation (OD) system was implemented to concentrate cranberry

juice at laboratory scale. The experimental setup allows the circulation of cranberry juice

(Vaccinium macrocarpon Ait.) and an osmotic agent with flow rates varying from 0.5 to 1.5

L min-1 at temperatures between 30 and 40ºC. The osmotic agent selected in this study was

a concentrated CaCl2(aq) solution with concentrations ranged from 30 to 50% w/w.

The transmembrane flux of water vapor was ranged between 0.25 and 1.21 L h-1 m-

2. The comparative low content of TSS in cranberry juice allows obtaining fast water

removal from 500 mL of juice, achieving concentrations from 8.6 to 48 ºBrix in 18 min.

The total phenolic content was preserved after concentration.

A mass transfer model was developed to explain the concentration kinetics of the

juice. The solution simulations allow obtaining a maximum deviation of 32% between

experimental and simulated values of transmembrane water flux.

Keywords: Osmotic distillation, cranberry juice, concentration, modeling, hydrophobic

porous membranes.

3

1. Introduction

In order to optimize the storage conditions and transportation costs, fruit juices are

generally concentrated by vacuum evaporation. However due to thermal effects, this

process involves changes in the organoleptic properties of products. The traditional

evaporation process allows obtaining concentrations up to 45–71 ºBrix, while the

increasing of viscosity is limited by the rising of the temperature while the process takes

place. The fouling reduces the heat transfer rate, that’s why the surfaces of evaporation

must be cleaned regularly to insure the efficient working of the evaporator. The selection of

an adequate system must be considered if these devices are apt to concentrate fruit juices at

high concentration levels and low costs (Cissé et al., 2011; Hernandez et al, 1995).

Osmotic Distillation (OD) is an interesting alternative process for the concentration

of thermosensible solutions. This process has been used for concentrating liquid foods,

such as milk, fruit and vegetable juices, instant tea and coffee because it works under room

pressure and temperature, preserving the nutritional characteristics of foods (Hogan et al,

1998). Moreover, using concentrated brine as extracting phase allows decreasing the loss of

aromatic compounds from the juice to be concentrated to the brine keeping then a good

level of organoleptic properties (Vaillant et al., 2001; Cissé et al., 2005). This isothermal

concentration method can be applied to fruit and vegetable juices whose properties may be

altered by thermal treatments. Thus, a product with nutraceutical properties is an ideal

candidate to be concentrated with this technique (Romero et al., 2003b).

The concentration of cranberry juice (Vaccinium macrocarpon Ait.) of OD is

theoretically and experimentally analyzed in this study. The high commercial value of this

product could justify the implementation of a system of re-concentrating the brine used as

4

an extracting solution to receive the evaporated water from the juice. The low evaporation

flux and the regeneration of the diluted brine represent the main technical and economic

barriers of this process for concentrating the majority of liquid foods. These particular

conditions have motivated this work in order to analyze the performance of the OD process

for the concentration of cranberries juice, identifying the main operation variables. In our

knowledge it is the first work published concerning the application of osmotic distillation

for the concentration of the cranberry juice.

2. Concentration of cranberry juice by means of osmotic distillation

a. Principle of the osmotic distillation process

Osmotic distillation is a membrane process applied to concentrate solutions under

isothermal conditions. In this process, an aqueous solution can be concentrated by an

osmotic gradient using an aqueous osmotic agent with low water activity (i.e. concentrated

brine). Fig. 1 shows an outline describing the principle of the process where a macroporous

and hydrophobic membrane separates both solutions. In this figure, three regions may be

identified in the proximities of the membrane: (I) the boundary layer of the feeding solution

to be treated; (II) the membrane pore filled with gas; (III) the boundary layer of the osmotic

agent. The OD is an evaporative process where simultaneous mass and heat transfer is

observed with its respective concentration and temperature profiles. The temperature

profile can be explained by a temperature polarization phenomenon, which involves a latent

heat transfer through the membrane. This latent heat transfer decreases the temperature at

the evaporation interface and increases the temperature at the condensation interface. The

interfaces formed by the liquid phases and the gas retained in the pores are considered in

thermodynamic equilibrium. Thus, taking into account the volatile condition of the

5

components to be transferred, the driving force generated by the mass transfer through the

porous medium is the difference in vapor pressure between both interfaces. The solutions to

be concentrated in the majority of the studies on OD (Ali et al., 2003; Alves and Coelhoso,

2006; Courel et al., 2001; Romero et al., 2003a; Petrotos and Lazarides, 2001; Cassano and

Drioli, 2007; Cassano et al., 2007; Jiao et al., 2004; Bélafi-Bakó et al., 2006; Koroknai et a

al., 2006) contain a low concentration of non-volatile solutes from moderate to high

molecular weight (carbohydrates, polysaccharides, carbolic acid salts and proteins), which

have limited stability at high temperatures and pressures. In a previous work, Romero and

coworkers (Romero et al., 2003a) analyzed the effect of the boundary layers on the

concentration and temperature polarization phenomena, developing an algorithm that solves

the equations attached to the simultaneous mass and heat transfer for flat sheet membrane

modules.

b. Properties of cranberry juice.

Cranberry, Vaccinium macrocarpon Ait., is a specie that grows wild in United State

of America. This country is the leading world producer followed by Canada and Chile

(Buzeta, 1997). The interest of consumers is based on its low calories content, high

vitamins content, minerals and fiber percentage (Buzeta, 1997). The use of concentrates of

this juice, in liquid and pill form has become massive because of its medicinal properties,

such as being anti-microbial, anti-carcinogenic, analgesic and anti-inflammatory (Wu et al.,

2008). These properties are attributed to the main nutrients in cranberry represented by

phenolic compounds (Wu et al., 2009). Cranberries are rich in phenolic antioxidants with

redox properties that allow them to act as hydrogen donors and singlet oxygen quenchers

(Lin et al., 2005; Kwon et al., 2007). Recent studies have shown anthocyanins,

proanthocyanidins, and phenolics from cranberries are active components in molecular

6

mechanism behind various health benefits of cranberries (Lin et al., 2005; Guo et al., 2007;

Apostolidis et al., 2008; Wu et al., 2009).

This study used diluted samples of cranberry juice concentrate made by

OceanSpray®. The concentrate is prepared from depectinized, filtered juice derived from

properly matured, cleaned cranberries. It is concentrated under low temperatures and

vacuum, and the essence fraction is returned (OceanSpray, 2013). The main components of

this product are shown in detail on Table 1.

3. Experimental procedure

The cranberry juice concentrated in these experiments was obtained from

OceanSpray®. Commercial concentrated juice was used in experiments in order to obtain

constant quality, concentration of sugars and phenolic content for all samples to be

concentrated by OD. Concentrated juice with initial 50º Brix was diluted up to º8.6 Brix.

This concentration was chosen because it was verified as an average value of sugar

concentration in fresh juice in previous tests.

The experimental device used in this study is constituted by a system with two

independent circuits for the circulation of the solutions: one for the solution to be treated

(water or cranberry juice) and the other for the concentrated brine used as osmotic agent.

Both circuits have peristaltic pumps connected to graded vessels where the volume

variations of the solutions are measured as a function of time. The vessels are submerged in

a thermostated bath in order to maintain a constant temperature (30 or 40 ºC). The circuits

are connected to a hollow fiber membrane contactor. The outline of this arrangement is

presented in Fig. 2. One module was used in all experiments: a hollow fiber minimodule

7

Celgard Liquicel™ 1.7×5.5 (7400 fibers, 0.58m2 of surface contact area). The

characteristics of this module are presented in Table 2.

In this work, the solution to concentrate was cranberry juice and the osmotic agent

was concentrated brine. The juice circulates on the shell side of the hollow fiber contactor

in order to obtain the best mass transfer conditions and facilitate cleaning of the module.

250mL of juice was concentrated using 1000 mL of concentrated brine, which was

circulated into the lumen side as osmotic agent. The circulation configuration was also

chosen in order to facilitate the cleaning procedure of the module considering the

characteristics of both solutions. The brine is an aqueous solution of calcium chloride with

concentrations of 30%, 40% and 50% p/v. In this way, the solution volume change of each

vessel was monitored as a function of time. The conductivity of the juice was quantified

after every experiment in order to verify no leaks from the receiving brine through the

membrane pores.

Each experimental run was stopped when the volume of cranberries juice was

reduced from 250 to 50 mL. These experimental runs were achieved in operation times

ranged from 18 to 84 min depending on the operating conditions. The mass transfer flux

was quantified in function of the change in the brine level. At the end of the runs the

membrane was cleaned with an ethanol solution 30% (v/v) and then water. CaCl2 was used

as solute in the osmotic agent since this produces a decrease of water activity greater than

other salts, which allows an increment of the driving force for the mass transfer (Romero et

al., 2003b). The experimental design considered 3 variables: the concentration of the

concentrated brine, the circulation rates of the solution to be treated and the process

temperature. Each variable was evaluated in two levels taking into consideration a central

8

point, except for the temperature case. The operation conditions are summed up on Table

3.

The cranberries juice treated in the concentration experiments was previously

filtered using borosilicate microfiber filters MFS GC50 of 47 mm with a 99.9% efficiency

to remove particles with a size of 0.3 μm. In addition to the rates of evaporated volume, the

concentration of phenolic compounds was determined in function of the time during the

treatment. The phenolic compounds content in the cranberry juice would not have to be

modified by this type of processing. However, it was necessary to verify the concentration

of these compounds in the juice because they are considered to be the active compounds for

the health (Lin et al., 2005; Guo et al., 2007; Apostolidis et al., 2008; Wu et al., 2009;

Koroknai et al., 2008). In this way, samples (200 μL) of the juice were extracted every 7

min to quantify the polyphenols content using the modified Folin–Ciocalteu method

(Zoecklein et al., 2001) 1mL of Folin–Ciocalteu reactive was added to the samples in a

vessel, after 2 min 0.8mL of Na2CO3 (7.5%) was added. Then it is kept at 30ºC for 90 min.

Finally, the absorbance is measured at 765 nm.

Total monomeric anthocyanin pigment (ACN) contents in diluted and concentrated

cranberry juice samples were determined by using the pH differential method (Giusti and

Wrolstad 2001). Absorbance was measured at 510 and 700 nm. ACN was calculated using

cyanidin-3-glucoside coefficients (molar extinction coefficient of 26900 L cm-1 mol-1 and

molecular weight of 449.2 g mol-1).

9

4. Modeling and simulation of mass transfer

4.1. Mass transfer equations

The fundamentals of the osmotic evaporation process were explained on Section 2.1

and shown in Fig. 1. In this work, a resistances-in-series model has been used to explain the

transport of water through the membrane. Thus, the water transfer through the membrane

can be represented by the sequential transfer across (1) the boundary layer of the feeding

solution to be treated; (2) the membrane pore filled with gas; (3) the boundary layer of the

osmotic agent. Between the steps (1)-(2) and (2)-(3) there are two gas-liquid interfaces

established at the pore mouths where evaporation and condensation take place. Two

equilibrium conditions could be supposed at these locations in order to describe the

evaporation and condensation of water on the feed solution and receiving brine,

respectively.

From this description, the water transmembrane flux, J (mol m-2 s-1), can be

quantified of a set of transfer equation describing the phenomena at the proximities of the

membrane.

The water transmembrane flux through the boundary layer of the fruit juice (step I

in Fig.1) was estimated by linear relationship of the mass transfer coefficient and the

concentration gradient of water across this layer.

( )( ) ( )1 1 F Fw b 1J k x x= − (1)

where xbF and x1

F are the molar fraction of water in the bulk of the feed solution (fruit juice)

and in the interface with the membrane, respectively. k(1) is the coefficient of mass transfer

through feed boundary layer (in mol m-2 s-1). Meanwhile, the mass transfer coefficient was

calculated from the Sherwood number correlation reported by Valdés and coworkers

10

(Valdés et al., 2009) as a function of the Graetz number in the shellside of the membrane

contactor:

Sh(1) = 0.09Gz1/3 (2)

At the fruit juice-membrane interface the evaporation of water was described by the

following equation:

PwF = a

wF P

wF* = x

1Fγ

1F P

wF*

(3)

where PwF is the vapor pressure of water at the interface, aw

F and γ1F are the water activity

and activity coefficient at the feed solution-membrane interface, respectively. PwF*

represents the water saturation pressure. In this work, the water activity, awF, was calculated

by means of UNIFAC method (Achard et al., 1992) considering the cranberry juice (feed

phase) as an ideal solution of sucrose with the same TSS content of the original juice.

On the other hand, the evaporated water is transferred through the membranes

pores. This transport can be described of molecular diffusion in the pores filled with gas.

Thus, the water transmembrane flux in this second step was calculated from equation 4,

which describes the transport of water vapor through stagnant air.

*( )

*( )( ) ln( )

E E2 w air w w

w F Fw w

D P a PPJ

RT P a P

ετδ

− −=−

(4)

where P is the total pressure, aw water activity at interfacial condition in brine (E) and feed

(F) phases, respectively. Pw* represents water saturation pressure at interfacial condition for

brine (E) and feed (F), R is the universal constant of gasses, Dw–air the diffusion coefficient

of water in the air; ε, τ, δ represent membrane porosity, tortuosity and thickness respectively.

After diffusion through the membrane pores, water condensates on the membrane-

brine interface. This vapor-liquid equilibrium can be described by equation 5:

11

PwE = a

wE P

wE* = x

2Eγ

2E P

wE*

(5)

where PwE is the vapor pressure of water at the interface, aw

E and γ2E are the water activity

and activity coefficient at the membrane-brine interface, respectively. PwE* represents the

water saturation pressure. If temperature is constant through the membrane PwE* can be

considered equal to PwF*. However, the evaporation-condensation process through the

membrane involves the latent heat transfer of water. Thus, temperature in the feed solution

and brine could be slightly different, but the heat transfer was not considered in

calculations. In this study, the water activity, awE, was estimated of the modified ASOG

method reported by Correa and coworkers (1997) for concentrated solutions.

Finally, the third step of mass transfer in Fig. 1 represents the water transfer through

the boundary layer of the receiving brine. The mass transfer flux was estimated by equation

(6).

( )( ) ( )3 3 E Ew 2 bJ k x x= − (6)

where F1x and E

2x are the molar fractions of the water coming in and going out the pores,

respectively. In this work, brine circulates into the lumen side and the mass transfer

coefficient in this boundary layer, k(3), was estimated from Sherwood number correlation of

Notter under laminar flow conditions (Gabelman and Hwang, 1999):

( ) /( ) ( ) . ( ). (Re )1 33 3 0 88 3Sh 0 0149 Sc= (7)

This equation is applicable when Sc>100.

In a short period of time, it could be considered that the water transfer takes place

under steady state conditions. In this way, the concentration of juice and brine was

considered constant for a brief period of time and the mas transfer flow through each step

12

must be considered identical ( ( ) ( ) ( )1 2 3w O D a v w IDJ A J A J A= = ). AOD and AID represent the

external and internal surface area available for mass transfer in the hollow fibers. The mean

value of the contact surface area through the membrane, Aav, was estimated as the average

value between the inside and outside diameter of the fibers.

The water transfer model defined by equations 1-7 shows the sequential steps

through each local transport layer at the proximities of the membrane. This approach was

applied as a pseudo-steady state model. Thus, after a calculation in a specific time a water

balance was done on the feed and receiving solution in order to recalculate the volumes and

concentrations of water in the juice and the brine to repeat the calculation at the new

condition.

4.2 Simulation of the mass transfer in OD

The water transfer model described by phenomenological equations 1, 4 and 6,

includes the equilibrium conditions at the juice-membrane and membrane-brine interfaces

(equations 3 and 5) as well as the water balance in the feed solution (juice) and in the

receiving brine. This system of equations was solved at steady state conditions for each

time step, and then the effect of the operating variables was assessed.

The solution of this model based on a resistances-in-series structure, was found by

means of an iterative calculation for each time step. The iteration cycle was accelerated

using the Regula Falsi method, which has been reported in previous studies (Valdés et al.,

2009; Hasanoglu et al., 2012).

Both transport equations and thermodynamic methods were implemented in a

Matlab® 7.1 script. The equation set was solved by means of the iterative calculation of

13

transmembrane water flux, J (mol m-2 s-1), as a function of unknown concentration of water

at the interfaces in contact with the membrane, x1F and x2

E in Fig. 1. After an instantaneous

calculation of the water flux, mass balance allows recalculating concentrations of water,

obtaining the concentration dynamic.

5. Results and discussions

5.1 Influence of the operation variables in the concentration of cranberries juice.

The experimental runs were planned according to a 2x3x3 experimental design

where the average transmembrane water flux was measured as a function the flow rate of

solutions, the concentration of the concentrated brine and temperature. These results are

shown in table 4, which shows the experimental runs, average transmembrane flux, ºBrix

and the necessary time to reach 50 mL of cranberries juice and the last column shows the

time required to increase the TSS of the concentrated juice from 8.6 to close to 40 ºBrix in

each experiment.

The maximum value of transmembrane flux observed in these experiments was

equal to 1.21 L h-1 m-2 when the solutions were circulated with a flow rate of 1.5 L min-1

and the concentration of the brine and temperature were 50% p/v and 40ºC, respectively.

Under these conditions the time required to reduce the volume of the juice from 250 mL to

50 mL was 18 min. The initial juice/brine volumetric ratio was 1:4 for all experiments. This

volumetric ratio between solutions involves a slower dilution of the brine maintaining a low

value of its water activity and a relatively high driving force of the concentration process.

Thus, the water removal during the OD runs was maintained almost constant during the

time of runs.

14

5.2. Phenomenological mass transfer model

Figure 3 shows the experimental values of juice volume reduction (points) as a

function of the time. In these figures, lines represent the simulation results obtained with

the water transfer model described in sections 4.1 and 4.2. From these figures, it could be

observed a constant concentration rate during the experiment times. There was a good

agreement between experimental values and prediction obtained from simulations for lower

concentration of the brine (30%p/v), but this prediction was less accurate when the

concentration of the brine increased. Moreover, best agreement was observed in the results

reported in figure 3, which were measured with solutions circulating with a flow rate equal

to 1.5 L min-1. The better or worse prediction capacity of the mass transfer model could

depend on accuracy of the water activity prediction methods and the prediction capacity of

the mass transfer coefficients at the proximities of the membrane. Experiments with higher

concentration of the brine and faster concentration rates could be more difficult to describe,

especially in this case where the estimate of water activity in the cranberry juice was done

by assimilation with an aqueous solution of sucrose. Lower accuracy in the calculation of

the water activity gradient, which is the driving force of this process, can negatively impact

the prediction capacity of the transfer model.

On the other side, the increasing of viscosity during the concentration of the juice

could generate a significant modification of the mass transfer coefficient in the juice

boundary layer. Thus, this modification of the transport properties during the process could

explain the decreasing of agreement between experiments and theoretical predictions,

especially for those experiments with higher water transfer rates. This phenomenon was

already reported in previous studies with kiwi juice (Cassano and Drioli, 2006).

15

5.3. Concentration of total phenolic compounds and anthocyanins.

During the concentration experiments four samples for different experiments were

collected in order to determine the concentration change of total phenolic compounds.

These concentration experiments showed the same concentration ratio, decreasing the

volume of the juice in 3.7 times. Table 5 shows the initial and final total phenolic

concentration of these juice samples, which are reported in mg of Gallic Acid Equivalent

per liter (mg GAE/L) obtained by means of the Folin-Ciocalteu method.

From the results reported in table 5, it is possible to observe that the mean

increasing factor of total phenolic concentration matches with the concentration factor 3.7.

Thus, the osmotic evaporation process does not involve a significant degradation of the

phenolic content under the operating conditions applied in this study.

In these experiments, the TTS content of the juice increased in 28ºBrix, which are

mainly sugars that could interfere on the quantification of total phenolic compounds, since

sugars can affect the oxidation of the Folin-Ciocalteu method. Nevertheless, the

concentration level of these experiments and the relatively low concentration of TSS of the

cranberry juice allow considering the quantification of phenolic compounds as acceptable.

The juice used in this work shows a concentration of phenolic compounds significantly

lower than the values reported in literature for fresh cranberry juice (Cotê et al., 2011). The

cranberry juice used in OD experiments was reconstituted from the commercial concentrate

supplied by OceanSpray®. In this way, the preservation of the phenolic content after OD

processing was based on the original value of concentration of phenolic compounds in this

commercial juice.

On the other hand, the concentration of anthocyanins was also assessed. The initial

concentration of anthocyanins in the diluted juice was 0.03825±0.0032 mg L-1. This

16

concentration is significantly lower than the anthocyanin content reported for fresh

cranberries in a previous work (Vvedenskaya and Vorsa, 2004). In fact the content of the

anthocyanins has been estimated between 31.5 and 47.4% of the total flavonoids,

depending of the variety. The juice used in this study was reconstituted from a commercial

sample, which could be strongly affected by the previous concentration processing and the

storage time.

Samples concentrated by means of OD have shown a mean concentration of

anthocyanins of 0.1587±0.0224 mg L-1. This concentration represents 4.2 times the

concentration of the diluted juice and it is slightly higher than the mean concentration factor

3.7 of these experiments. Thus, anthocyanin content of the diluted juice does not seem to be

remarkably affected by the OD process. These results encourage future tests with fresh fruit

in order to verify the preservation of these compounds in the concentrated juice with a

higher concentration of anthocyanins where the quantification method could be more

accurate.

6. Conclusions

An osmotic distillation process was implemented using a hollow fiber membrane

module to concentrate cranberry juice; meanwhile a concentrated CaCl2 solution was used

as extracting phase. Transmembrane water flux was quantified as a function the flow rate of

solutions, the concentration of the concentrated brine and temperature. The maximum value

of transmembrane water flux obtained from these experiments was equal to 1.21 L h-1 m-2

when the flow rate of solutions was 1.5 L min-1 and the concentration of the brine and

temperature were 50% p/v and 40ºC, respectively. Under these conditions the time required

to reduce the volume of the juice from 250 mL to 50 mL was 18 min. This fast

17

concentration rate can be explained by the composition and moderated concentration of

TSS in the cranberry juice, which allows preserving high water transfer rate during the

osmotic distillation process. A mass transfer model has been proposed to predict the water

transfer through the membrane, obtaining an estimation of this value with a maximum

deviation of 32%. In this way, the driving force of this process represented by the activity

difference between the cranberry juice and the receiving brine shows a significant effect on

the concentration performance. Thus, the water activity of solutions (juice and brine)

should be correctly estimated in order to obtain an accurate prediction of the water transfer

and the concentration rate.

Finally, it has been demonstrated that the concentration of cranberry juice by

osmotic distillation does not affect the content of phenolic compounds and specifically of

anthocyanins. This result is promising to encourage the concentration of other berry juices

by means of this technique.

Acknowledgements

The authors acknowledge the support of the Chilean National Commission for

Scientific & Technological Research (CONICYT) through the Project FONDECYT

11110097 and the Franco-Chilean collaboration Program ECOS CONICYT (Project

C10E05).

18

Nomenclature

A Activity (dimensionless)

A Surface area for mass transfer (m2)

Dw-air Diffusion coefficient of water in air (m2 s-1)

J Mass transfer flux (mol m-2 s-1)

Gz Graetz number (dimensionless)

K Mass transfer coefficient (mol m-2 s-1)

P Pressure (Pa)

Pw Partial pressure of water (Pa)

Pw* Partial pressure of pure water (Pa)

R Universal constant of gases (8.314 J mol-1 K-1)

Re Reynolds number (dimensionless)

Sc Schmidt number (dimensionless)

Sh Sherwood number (dimensionless)

T Temperature (K)

X Molar fraction (dimensionless)

Greek letters

Γ Activity coefficient (dimensionless)

Ε Porosity (dimensionless)

Δ Membrane thickness (m)

Τ Membrane tortuosity (dimensionless)

Subscripts

1 Refers to the feed solution-membrane interface

19

2 Refers to the membrane-brine interface

Av Refers to average diameter of the fibers

B Refers to the bulk of the phase

ID Refers to internal diameter of the fibers (lumen)

OD Refers to outside diameter of the fibers (shell)

W Refers to water

Superscripts

(1) Refers to feed solution boundary layer

(2) Refers to membrane pores

(3) Refers to receiving phase (brine) boundary layer

E Refers to extractant phase (brine)

F Refers to feed solution

20

References

1) Achard, C., Gros, J. B., & Dussap, C. G. (1992). Prédiction de l'activité de l'eau, des

températures d'ébullition et de congélation de solutions aqueuses de sucres par un

modèle UNIFAC. Cahier Scientifique, 109-IAA, 93-101.

2) Ali, F., Dornier, M.,Duquenoy, A., & Reynes, M. (2003). Evaluating transfers of aroma

compounds during the concentration of sucrose solutions by osmotic distillation in a

batch-type pilot plant. Journal of Food Engineering, 60, 1–8.

3) Alves, V., & Coelhoso, I. (2006). Orange juice concentration by osmotic evaporation

and membrane distillation: a comparative study. Journal of Food Engineering, 74, 125–

133.

4) Apostolidis, E., Kwon, Y.I., & Shetty, K. (2008). Inhibition of Listeria monocytogenes

by oregano, cranberry and sodium lactate combination in broth and cooked ground beef

systems and likely mode of action through proline metabolism. International Journal of

Food Microbiology, 128, 317–324.

5) Bélafi-Bakó, K., Koroknai, B. (2006). Enhanced flux in fruit juice concentration:

coupled operation of osmotic evaporation and membrane distillation. Journal of

Membrane Science, 269, 187-193.

6) Buzeta, A. (1997). Chile: Berries para el 2000. (1st ed., pp. 15-19). Fundación Chile.

Departamento Agroindustrial Santiago, Chile.

7) Cassano, A., & Drioli, E. (2007). Concentration of clarified kiwifruit juice by osmotic

distillation. Journal of Food Engineering, 79, 1397-1404.

8) Cassano, A., Conidi, C., Timpone, R., D’Avella, M., & Drioli, E. (2007). A membrane-

based process for the clarification and the concentration of the cactus pear juice. Journal

of Food Engineering, 80, 914–921.

21

9) Cissé, M., Vaillant, F., Bouquet, S., Pallet, D., Lutin, F., Reynes, M., Dornier, M.

(2011). Athermal concentration by osmotic evaporation of roselle extract, apple and

grape juices and impact on quality. Innovative Food Science and Emerging

Technologies, 12, 352–360.

10) Cissé, M., Vaillant, F., Perez, A., Dornier, M., & Reynes, M. (2005). The quality of

orange juice processed by coupling crossflow microfiltration and osmotic evaporation.

International Journal of Food Science & Technology, 40(1), 105−116.

11) Correa, A., Comesaña, J. F., Correa, J. M., & Sereno, A. M. (1997). Measurement and

prediction of water activity in electrolyte solutions by a modified ASOG group

contribution method. Fluid Phase Equilibria, 129, 267-283.

12) Cotê, J., Caillet, S., Doyon, G., Dussault, D., Salmieri, S., Lorenzo, G., Sylvain, J.-F., &

Lacroix, M. (2011). Effects of juice processing on cranberry antioxidant properties.

Food Research International, 44, 2907-2914.

13) Courel, M., Tronel-Peyroz, E., Rios, G., Dornier, M., & M. Reynes. (2001). The

problem of membrane characterization for the process of osmotic distillation.

Desalination, 140, 15–25.

14) Gabelman, A., & Hwang, S.-T. (1999). Hollow fiber membrane contactors. Journal

Membrane Science, 159, 61–106.

15) Giusti, M.M, Wrolstad, R.E. (2001). Unit F1.2: Anthocyanins.Characterization and

measurement with UVvisible spectroscopy. In R.E. Wrolstad, (Ed.). Current Protocols

in Food Analytical Chemistry (pp. 1-13). New York: John Wiley & Sons.

16) Guo, M., Perez, C., Wei, Y., Rapoza, E., Su, G., Bou-Abdallah, F., & Chasteen, N.D.

(2007). Iron binding properties of plant phenolics and cranberry's bio effects. Dalton

Transactions, 43, 4951–4961.

22

17) Hasanoglu A., Rebolledo F., Plaza A., Torres A., & Romero J. (2012). Effect of the

operating variables on the extraction and recovery of aroma compounds in an osmotic

distillation process coupled to a vacuum membrane distillation system. Journal of Food

Engineering, 111, 632–641.

18) Hernandez, E., Chen, C.S., Johnson, J., & Carter, R.D. (1995). Viscosity changes in

orange juice after ultrafiltration and evaporation. Journal of Food Engineering, 25, 387–

396.

19) Hogan, P., Canning, R., Peterson, P., Johnson, R., & Michaels, A. (1998). A new option:

osmotic distillation. Chemical Engineering Progress, 94, 49–61.

20) 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, 303–324.

21) Koroknai, B., Csanádi, Zs., Gubicza, L., Bélafi-Bakó, K. (2008). Preservation of

antioxidant capacity and flux enhancement in concentration of red fruit juices by

membrane processes. Desalination, 228, 295-301.

22) Koroknai, B., Gubicza, L., Bélafi-Bakó, K. (2006). Coupled membrane process applied

for fruit juice concentration. Chemical Papers, 60, 399-403.

23) Kwon, Y.I., Apostolidis, E., Labbe, R.G., & Shetty, K. (2007). Inhibition of

Staphylococcus aureus by phenolic phytochemicals of selected clonal herbs species of

Lamiaceae family and likely mode of action through proline oxidation. Food

Biotechnology, 21, 71–89.

24) Lin, Y.T., Labbe, R.G., & Shetty, K. (2005). Inhibition of Vibrio parahaemolyticus in

seafood systems using oregano and cranberry phytochemical synergies and lactic acid.

Innovative Food Science and Emerging Technologies, 6, 453–458.

23

25) OceanSpray. Cranberries Ingredient Technology Group (2013). One Ocean Spray Drive,

Lakeville, Middleboro, Massachusetts 02349, USA. Web page:

http://www.oceansprayitg.com/downloads/concentrate/nutritionals/UPC_94040-

94041.pdf. Last access, February 19 of 2013.

26) Petrotos, K., & Lazarides, H. (2001). Osmotic concentration of liquid foods. Journal of

Food Engineering, 49, 201–206.

27) Romero, J., Rios, G.M., Sanchez, J., & Saavedra, A. (2003a). Analysis of boundary

layer and solute transport in osmotic evaporation. AIChE Journal, 49, 2783–2793.

28) Romero, J., Rios, G.M., Sanchez, J., Bocquet, S. & Saavedra, A. (2003b). Modelling of

heat and mass transfer in osmotic evaporation process. AIChE Journal, 49, 300–308.

29) Thanedgunbaworn, R., Jiraratananon, R., & Nguyen, M.H. (2007). Shell-side mass

transfer of hollow fibre modules in osmotic distillation process. Journal of Membrane

Science, 290, 105–113.

30) 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(3), 195−202.

31) Valdés, H., Romero, J., Saavedra, A., Plaza, A, & Bubnovich, V. (2009). Concentration

of noni juice by means of osmotic distillation. Journal of Membrane Science, 330, 205-

213.

32) Vvedenskaya, I.O., & Vorsa, N. (2004). Flavonoid composition over fruit development and

maturation in American cranberry, Vaccinium macrocarpon Ait. Plant Science, 167, 1043-

1054.

33) Wu V., Qiu X., G. de los Reyes B., Lin C., & Pan Y. (2009). Application of cranberry

concentrate (Vaccinium macrocarpon) to control Escherichia coli O157:H7 in ground beef

24

and its antimicrobial mechanism related to the down regulated slp, hdeA and cfa. Food

Microbiology, 26, 32-38.

34) Wu, V.C.H, Qiu, X.J., Bushway, A., & Harper, L. (2008). Antibacterial effects of

American cranberry (Caccinium macrocarpon) concentrate on foodborne pathogens. Food

Science and Technology, 41, 1834-1841.

35) Zoecklein, B. Fugelsang, K. Gump, B. Nury, F (2001). “Análisis y producción de vino” 1st

Ed., Aspen Publisher, Inc., Zaragoza, Spain.

25

Table 1. Composition of the concentrated cranberry juice (50ºBrix) (OceanSpray, 2013)

Unit per 100 g

Moisture g 49.24

Calories kcal 198,00

Protein g 0.27

Total fat g <0.25

Carbohydrates g 49.30

Sugars g 22.02

Minerals

Calcium mg 39

Phosphorus mg 28

Sodium mg 14

Vitamins

Vitamin C mg 58

Niacin mg 0.57

Thiamin mg 0.04

26

Table 2. Characteristics of hollow fiber contactor modules used in this work.

Property (unit) Value or characteristic Module type Minimodule Liquicel® 1.7x5.5 Membrane material Polypropylene Number of fibers 7400 Porosity (%) 40 ID of fiber (m) 2.2*10-4 OD of fiber (m) 3.0*10-4 Length of fiber (m) 0.120 ID of shell (m) 0.043 Contact surface area (m2) 0.580

27

Table 3. Experimental design: variables and its levels.

Type of the feed solution

Cbrine (% p/v)

Ffeed (L min-1)

Tfeed (ºC)

Fbrine (L min-1)

Tbrine (ºC)

water and diluted cranberry juice

30-40-50 0.5-1.0-1.5 30 and 40 0.5-1.0-1.5 30 and 40

28

Table 4. Resume of results for each experimental runs.

Run Ffeed =Fbrine

(L min-1)

Cbrine (%p/v)

T (°C)

Average Jw (L h-1 m-2)

Initial °BRIX

Final °BRIX

t (min)

1 0.5

30

30

0.25

8.6 ±0.3

40.0 84

2 1.0 0.31 38.0 66

3 1.5 0.37 38.4 57

4 0.5

40

0.45 37.0 45

5 1.0 0.53 38.0 39

6 1.5 0.63 32.0 33

7 0.5

50

0.74 25.0 27

8 1.0 0.86 44.0 24

9 1.5 1.17 42.0 18

10 0.5

30

40

0.30 34.0 66

11 1.0 0.40 36.0 51

12 1.5 0.37 36.0 42

13 0.5

40

0.48 28.0 41

14 1.0 0.61 44.0 33

15 1.5 0.69 27.0 30

16 0.5

50

0.62 47.0 33

17 1.0 0.99 40.0 21

18 1.5 1.21 48.0 18

29

Table 5. Concentration of total phenolic compounds in juice samples before and after concentration by osmotic distillation process.

Sample Initial total phenolic concentration mg

GAE/L

Final total phenolic concentration mg

GAE/L 1 2.345 7.703 2 2.220 7.810 3 2.304 9.280 4 2.400 9.524

30

Figure Caption

Figure 1. Outline of the transport phenomena with the concentration and temperature

profiles trough the membrane.

Figure 2. Osmotic distillation setup used in this work with Celgard liquicel minimodule.

Figure 3. Comparison between experimental and simulated juice volume decreases in the

osmotic evaporation at 40 ºC with different circulation velocity and brine concentration.

FIGUURRE 1

31

32

FIGURE 2

Exxper

F

rime

IGU

enta

UR

al

RE 33

Simmulaationn

33

34

Highlights

-Osmotic distillation (OD) is applied to concentrate cranberry juice.

-The anthocyanins compounds of cranberry juice are concentrated after process

(OD).

-Experimental results are compared with simulation values of mass transfer obtained

by means of mathematical modeling.