Golden delicious ) Desorption isotherms of fresh and osmotically dehydrated apples (

45

Desorption isotherms of fresh and osmotically dehydrated apples (Golden delicious)

S. Bellagha^1*a, A. Sahli^1**, M. Ben Zid¹ and A. Farhat^2**

1Institut National Agronomique de Tunisie, 43 Av. Charles Nicolle, Tunis, Tunisie 2Centre de Recherche et des Technologies de l’Energie, B.P. 95 Hammam-Lif, Tunisie

*Groupe de Recherche en Génie des Procédés Agroalimentaires, 02/UR/11-02.

**Laboratoire de Maîtrise des Technologies de l’Energie, ‘LMTE’

Abstract – The each year larger apples national production, as well as, the higher demand of the food industry for such pre-treated dried fruits, has conducted researchers to better investigate this field. Osmotic dehydration is a partial dehydration through an osmosis process which involves immersing fruits for a given period of time in a hypertonic solution, here a sugar solution. Sugar impregnation allows an inhibition of polyphenoloxydase and prevents the loss of volatile compounds during the dehydration process. It is also used as a pretreatment process to improve the sensory quality of dried products. This work is divided in two main parts. First, an experimental determination of the desorption isotherms of fresh and osmotically dehydrated apples at different sugar concentrations (0 %, 30 % and 40 % w/w) was conducted. A static gravimetric method, based on the use of 9 saturated salts solutions was used to determine the sorption isotherms of fresh apple and apple subjected to osmo-dehydration processes in sucrose syrup solution at 30 % and 40 % concentration and at three different temperatures: 30, 40 and 50

°C. The fresh apple isotherms showed a type III sigmoid shape, with a decrease of the equilibrium moisture content with increasing temperature at constant relative humidity. However, in the water activity below 0.6, temperature effect seems to be negligible. Several models were adjusted to the experimental sorption data and the GAB equation gave the best fit. Sorption isotherms curves of osmotically dehydrated apples showed the same shape and for a constant water activity the equilibrium water content decreased with increasing sugar concentration. Thus, for 40 %, sucrose solution, the equilibrium water content is lower for the same equilibrium RH. Finally, Mathematical prediction of the experimental isotherms of sucrose treated apples may be satisfactorily done using the GAB model.

Key-words: Apples - Sorption isotherms - Osmotic dehydration.

1. INTRODUCTION

Sorption isotherms consist on following the evolution of product equilibrium water content as a function of water activity at constant temperature. Sorption isotherms reveal information about sorption mechanisms and interactions of food biopolymers with water.

Desorption isotherms are of great importance in the design of a food dehydration process, and in the determination of the drying time.

Fairly often, the drying process of food materials must be accomplished with the determination of their water sorption isotherms with the intention of predicting the end- point of drying. Moreover, modeling sorption isotherms facilitates the estimation of the amount of bound water to specific polar sites in dried foods and at monolayer moisture a product should be secure [1].

abelsih@gnet.tn ; sahli_inat_tn@yahoo.fr ; abdelhamid.farhat@inrst.rnrst.tn

Numerous mathematical equations can be found in the literature describing water sorption isotherms of food products [2-4]. Each of the models proposed, empirical, semi-empirical or theoretical has had different success in fitting equilibrium moisture content evolution of a given food at a different range of water activity.

Hence, the GAB model is generally reported to better reproduce the sorption isotherms data for the majority of food product up to a water activity of about 0.9 [5-7].

Osmotic dehydration is characterized by a fairly complex mass transfer process. The transport of water out of the tissue is completed by counter courant diffusion on the solute ending up to a depressing effect on the product water activity.

Very few studies were found on the sorption behavior of osmotically treated products and moreover on their mathematical representation [5].

Hence, this work aims in a first part to experimentally determine the desorption isotherms of fresh apples at three different temperatures, in an attempt to characterize the raw materials, then to correlate these data with mathematical models and in a second part to study the influence of sugar concentration of osmotically treated apples on water sorption behavior and evaluate the accuracy of the GAB model when used in osmotic dehydration cases.

2. MATERIALS AND METHOD 2.1 Samples preparation

Apples, Golden delicious variety, with an average moisture content of 86.2 % wet basis and Brix of 11.5 % were used as a raw material for all experiments.

The fruits were washed, hand peeled, cored and cut into cubic shapes (10×10×10 mm) in the stem-base direction by the mean of a cutting device. Samples were kept in a tinfoil hindering dehydration.

Sucrose solutions were prepared by dissolving the sugar in distilled water on a weight to weight basis.

Apple cubes were impregnated then in the osmotic solution (30 % and 40 % concentration) for 4 hours at 22 °C. The solid-to-solution mass ratio was maintained at 1:25 allowing unchanging driving force.

Once osmotic dehydration was accomplished, samples were strained and blotted smoothly.

2.2 Sorption isotherms 2.2.1 Experiments

The sorption isotherms of fresh and osmotically dehydrated apples at 30, 40 and 50°C were determined using a static gravimetric method. Samples were put in open weighing bottles and stored in air-sealed glass jars until maintaining equilibrium relative humidity with saturated salt solutions { NaOH, MgCl2, K2 CO3, Mg(NO3)2, SrCl2, NaCl, KCl, BaCl2, LiCl} [1].

The glass jars are kept in an oven at a constant temperature. Once equilibrium is reached (after 21 days), the equilibrium moisture content of samples was measured gravimetrically by oven drying for 24 hours at 105 °C.

2.2.2 Modeling sorption isotherm curves

The isotherm equations selected for modeling sorption isotherm curves are shown in Table 1.

Table 1: Equations of sorption isotherms models

Model Equation and parameters*

BET ^{X =}

(

) (

)

GAB ^{X =}

(

) (

)

Oswin n

w w

a 1 a a

X ⎟⎟

⎠

⎜⎜ ⎞

⎝

⎛

× −

=

*Xm monolayer moisture content (dry basis), wa water activity, CBET, C, K, a, n model parameters

2.2.3 Data analysis

The estimation of equation parameters was executed by the mean of a curve fitting system for Windows Curve Expert 1.3.

For curve fits, error is assessed using the standard error (S) and correlation coefficient (r) which are defined as follows:

( )

param data

. exp n

1 i

eqcali 2 eqi

n n

X X S

data . exp

−

−

=

∑

(1)

( )

( )

∑

∑

=

=

−

−

−

=

data . exp data . exp

n 1 i

2 eqi eq n

1 i

eqcali 2 eqi

X X

X X 1

r (2)

Where X_eqcali is the calculated value of equilibrium moisture content from the tested model, X_eqi is the experimental value of equilibrium moisture content, n_param is the number of parameters of the particular model and n_exp.datais the number of experimental points.

As the quality of the data model increases, the standard error approaches zero and the correlation coefficient r will approach unity.

3. RESULTS AND DISCUSSION 3.1 Sorption isotherm

The sorption isotherms of fresh apples plotted in Fig. 1 show a type II sigmoid shape. In the activity water above 0.6 and at constant relative humidity, an increase of temperature brings about relative humidity decrease. As said by Kaya et al. (2007) this can be explained by the excitation states of molecules which ,due to temperature augmentation, increase their distance apart and hence attractive forces between them decrease. Subsequently increasing temperature lessens the water sorption degree. This finding is consistent with the results of various researchers [8-10]. As well, for constant value of moisture content rising temperature leads to water content increase.

Fig. 1: Experimental desorption isotherms of fresh apples at three different temperatures 3.2 Modeling sorption isotherms

According to their correlation coefficient r and standard error of estimate S, the models equations were compared (Table 2). From the gathered information it can be made out the goodness of fit of each of the models used (Fig. 2; Fig. 3 and Fig. 4). In addition, the GAB model fit which shows sigmoid shaped sorption isotherms appears to be the most adequate one (Table 2, Fig. 2).

Fig. 2: Desorption isotherms of fresh apples fitted with GAB model at different temperatures

Fig. 3: Desorption isotherms of fresh apples fitted with BET model at different temperatures

The monolayer moisture content value of the GAB model decreases with increasing temperature. This was because increased temperatures raise activation energy of a number of water molecules to high levels that facilitate their taking apart from the sorption sites [11].

Fig. 4: Desorption isotherms of fresh apples fitted with Oswin model at different temperatures 3.3 Osmotically treated apples desorption isotherms

The shape of the isotherms of osmotically treated apples at different sugar syrup concentrations (Fig. 5) are characteristic of high sugar foods, which sorbs small amounts of water at low water activities.

Sharp increase in water content at high values of a_w is due to sugar, amorphous sugars being known to sorbs more water than the crystalline form.

Product composition, i.e. sugars in this case, is of great influence on the water sorption behavior. Here, at water activities lower than 0.8, the equilibrium moisture content of the 30 % sugar treated apples was higher than the 40 % treated ones.

Table 2: Parameters estimation, r and S of BET, GAB and Oswin equations fitted to sportion isotherms of fresh apples

Model Temperature

30 °C 40 °C 50 °C

BET X_m 0.4860 0.1509 0.1008

CBET 0.2192 8.5745 1.39 10¹⁰

S 0.1881 0.0715 0.0301

r 0.9848 0.9868 0.9926

GAB X_m 1.2034 0.1610 0.1073

C 0.0832 5.7584 -2.5 10¹⁰

K 0.9640 0.9915 0.9981

S 0.1990 0.0765 0.0293

r 0.9849 0.9871 0.0041

Oswin a 0.1874 0.2581 0.2205

n 1.2956 0.7833 0.6419

S 0.1930 0.0438 0.0452

r 0.9640 0.9857 0.9834

Fig. 5: Experimental and GAB modeling desorption isotherms of apples treated with sucrose solutions at 0 %, 30 % and 40 % concentration However, the inverse effect was observed at higher aw and the apples adsorbed more water at higher sugar concentration, as predicted by the GAB model. This crossing of the isotherms (Fig. 5), has often been described as a result of temperature effect on sorption isotherms of high sugar content foods. Abdullah [12], found similar results for date paste from three different dates cultivars containing different level of sugar.

Results of desorption isotherms of osmotically dehydrated food products at different sugar content are not available in the literature. Hence further investigation on this field for a better and more complete interpretation of the experimental results is under study at the present time.

GAB model was also used to predict the experimental sorption data of sugar treated apples and it gave a very good fitting as it can be seen in Fig. 5 and from the r and the

S values in Table 3.

GAB parameters have been determined and the monolayer moisture content varies from 0.7753 to 0.0879 kg/kg D.B., values close to those of other high sugar foods reported in the literature [5, 12, 13].

Table 3: GAB Model parameters for the desorption isotherms of sucrose treated apples at 30 % and 40 % sugar concentration and at 40 °C

Sugar concentration ( w/w )

0 % 30 % 40 %

Xm 0.1610 0.7753 0.0879

C 0.9915 0.5854 42.645⁰

K 0.9915 0.5854 0.9803

S 0.0765 0.0397 0.0289

r 0.9871 0.9907 0.9923

4. CONCLUSION

The investigation of sorption isotherms of fresh apples, Golden delicious variety, at three temperatures (30, 40 and 50 °C) unveiled sigmoid shape type III typical for most fruits. Equilibrium moisture content decreases with increasing temperature at constant equilibrium relative humidity and increases with increasing equilibrium relative humidity at constant temperature. Among the sorption models selected to fit sorption isotherms the GAB equation seems to be the most appropriate.

Desorption isotherms of osmotically treated apples at 30 % and 40 % sugar solution have been experimentally determined. Typical shape of high sugar product was found.

A higher equilibrium water content was measured for a same water activity when the sugar content is lower. GAB model gave a satisfactorily prediction for all isotherms.

NOMENCLATURE aw : Water activity

BET: Brunauer, Emmet and Teller

GAB: Guggenheim, Anderson and de Boer S: Standard error

r: Correlation coefficient

Xm Xm: monolayer water content, kg/kg C, K, a , n: parameters of the models tested

REFERENCES

[1] S.S. Sablani, S. Kasapis and M.S. Rahman, ‘Evaluating Water Activity and Glass Transition Concepts for Food Stability’, Journal of Food Engineering, Vol. 78, N°1, pp. 266 - 271, 2007.

[2] G.V. Barbosa-Canova and H. Vega-Mercado, ‘Physical, Chemical and Microbiological Characteristics of Dehydrated Foods’, In: G.V. Barbosa-Canova and H. Vega-Mercado (Eds.), Dehydration of Foods, pp. 29 - 99, N.Y. Chapman and Hall, 1996.

[3] G.H. Crapiste and E. Rotstein, ‘Prediction of Sorptional Equilibrium Data for Starch- Containing Foodstuffs’, Journal of Food Science, Vol. 47, pp. 1501 - 1507, 1982.

[4] N. Wang and J.G. Brennan, ‘Moisture Sorption Isotherms Characteristics of Potatoes at Four Temperatures’, Journal of Food Engineering, Vol. 14, N°4, pp. 269 - 287, 1991.

[5] F. Prothon and L.M. Ahrné, ‘Application of the Guggenheim, Anderson and De Boer Model to Correlate Water Activity and Moisture Content During Osmotic Dehydration of Apples’, Journal of Food Engineering, Vol. 61, N°3, pp. 467 - 470, 2004.

[6] E.O. Timmermann, J. Chirife and H.A. Iglesias, ‘Water Sorption Isotherms of Foods and Foodstuffs: BET or GAB Parameters?’, Journal of Food Engineering, Vol. 48, N°1, pp. 19 - 31, 2001.

[7] E. Tsami, M.K. Krokida and A.E. Drouzas, ‘Effect of Drying Method on the Sorption Characteristics of Model Fruit Powders’, Journal of Food Engineering, Vol. 38, N°4, pp. 381 - 392, 1999.

[8] A. Kaya, O. Aydin and C. Demirtas, ‘Drying Kinetics of Red Delicious Apple’, Biosystems Engineering, Vol. 96, N°4, pp. 517 - 524, 2007.

[9] F. Kaymak-Ertekin and A. Gedik, ‘Sorption Isotherms and Isosteric Heat of Sorption for Grapes, Apricots, Apples and Potatoes’, Lebensm.-Wiss.u.-Technol, Vol. 37, pp. 429 - 438, 2004.

[10] S. Timoumi, F. Zagrouba, D. Mihoubi and M.M. Tlilli, ’Experimental Study and Modelling of Water Sorption Desorption Isotherms on Two Agriculture Products: Apple and Carrot’, Journal de physique IV, Vol. 122, pp. 235 – 240, 2004.

[11] S. Lahsasni, M. Kouhila and M. Mahrouz, ‘Adsorption-Desorption Isotherms and Heat of Sorption of Prickly Pear fruit (Opuntia ficus indica)’, Energy Conversion and Management, Vol. 45, N°2, pp. 249 - 261, 2004.

[12] M.A. Alhamdan and B.H. Hassan, ‘Water Sorption Isotherms of Date Pastes as Influenced by Date Cultivar and Storage Temperature’, Journal of Food Engineering, Vol. 39, N°3, pp.

301 - 306, 1999.

[13] G.D. Saravacos, D.A. Tsiourvas and E. Tsami, ‘Effect of Temperature on the Water Adsorption Isotherms of Sultana Raisins’, Journal of Food Science, Vol. 51, N°2, pp. 381 - 387, 1986.