jfpe_437 826..

843

MODELING WATER ADSORPTION ISOTHERMS OF PINHÃO

(ARAUCARIA ANGUSTIFOLIA SEEDS) FLOUR AND
THERMODYNAMIC ANALYSIS OF THE ADSORPTION PROCESS

FLORENCIA CLADERA-OLIVERA1,2,3,
LIGIA DAMASCENO FERREIRA MARCZAK1,
CACIANO PELAYO ZAPATA NOREÑA2 and ANA CAROLINA PETTERMANN2
1
Department of Chemical Engineering
Federal University of Rio Grande do Sul (UFRGS)
Rua Luis Englert, s/n.Campus Central 90040-000
2
Institute of Food Science and Technology
Federal University of Rio Grande do Sul (ICTA-UFRGS)
Av. Bento Gonçalves 9500, Campus do Vale 91501-970
Porto Alegre, Rio Grande do Sul, Brazil

Accepted for Publication June 3, 2009

ABSTRACT

The seeds of Araucaria angustifolia, commonly known as pinhão, are

widely consumed in both Southern and Southeastern Brazil because of their
high nutritious value. The literature on the technological aspects of this seed
is still very scarce. Moisture adsorption isotherms of pinhão flour were deter-
mined at 10, 20, 30 and 40C using the gravimetric method. Results show that
temperature has little effect on the adsorption process and the Chirife model
was found to best represent the experimental data. The isosteric heat of
sorption (differential enthalpy) was calculated by using the moisture adsorp-
tion isotherm and decreased as moisture content increased, becoming almost
constant at 0.3 kg water/kg dry solids. The enthalpy–entropy compensation
theory was applied to adsorption isotherms and the isokinetic temperature for
pinhão flour was calculated by plotting the differential enthalpy versus differ-
ential entropy. It was found that the adsorption process investigated was
enthalpy controlled.

PRACTICAL APPLICATIONS

Moisture isotherms are important for several applications in food tech-

nology (they can be used to predict potential changes in food stability, for
3
Corresponding author. TEL: +55-51-3308-6249; FAX: +55-51-3308-7048; EMAIL: fcladera@
yahoo.com

Journal of Food Process Engineering 34 (2011) 826–843. All Rights Reserved.

826 © 2009 Wiley Periodicals, Inc.
DOI: 10.1111/j.1745-4530.2009.00437.x
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 827

storing method determination, packaging and ingredients selection, design and

optimization of drying equipment, among others). This work presents the
isotherms of a new product (pinhão flour), that do not exist in the market yet
and with very scarce literature. As pinhão is a seasonal product, the study of
new products using pinhão as raw material is also important.

INTRODUCTION

Brazilian Pine (Araucaria angustifolia) belongs to the Araucariaceae

family and is the most economically important native conifer species in Brazil
(Zandavalli et al. 2004). It is found in both Argentina and Brazil, stretching
from the southern states of Paraná, Santa Catarina and Rio Grande do Sul to
northern Argentina. People in the South of Brazil usually consume the seeds of
this tree, known as pinhão after being cooked and peeled or they are used as
raw flour for regional dishes. Pinhão is considered a source of starch, dietary
fiber, magnesium and copper, and its intake produces a low glycemic index
(Cordenunsi et al. 2004). Pinhão is a seasonal product, which is produced
from April to August. As the seeds have a high moisture content, they can be
easily contaminated by mushrooms during the stockpiling, jeopardizing its
commercialization. Nutritional and technological aspects of pinhão are
scarcely found in scientific literature. We have previously reported the desorp-
tion isotherms of raw pinhão (Cladera-Olivera et al. 2008); however, there are
no reports about moisture sorption models for pinhão flour.
The sorption isotherms describe the relationship between Aw (equilibrium
relative humidity) and the equilibrium moisture content (EMC) of a given food
at constant temperature; these represent an extremely valuable tool for food
scientists and technologists as it can be used to predict potential changes in
food stability, for storing method determination, packaging and ingredients
selection, design and optimization of drying equipment. Several mathematical
equations can be found in literature describing water sorption isotherms and
nine of these equations are used in this work.
Thermodynamics is one of the approaches used in order to understand the
properties of water and calculate energy requirements associated with heat and
mass transfer in biological systems. Some thermodynamic functions employed
in the analysis of sorption behavior in biological systems include the DH,
differential enthalpy, Ds and enthalpy–entropy compensation, which are cal-
culated from sorption isotherms. The DH is the total energy required to transfer
water molecules from vapor state into a solid surface or vice-versa. It is useful,
for example, in predictive drying models and in the design of drying equip-
ment (Fasina 2006). The Dh is DH minus the latent heat of vaporization for
water and is a measure of the water-solid binding strength (McMinn and
828 F. CLADERA-OLIVERA ET AL.

Magee 2003). This is a vital property, which is employed for both design work
and qualitative understanding of the water state on the food surface (Tolaba
et al. 2004). The Ds of a material is proportional to the number of available
sorption sites at a specific energy level (Madamba et al. 1996). Gibbs DG may
be indicative of the affinity of sorbents for water and provides a criterion to
whether water sorption occurs as a spontaneous process or not (Apostolopou-
los and Gilbert 1990; Telis et al. 2000). Enthalpy–entropy compensation
theory (or isokinetic theory) is used to evaluate physical and chemical phe-
nomena such as sorption reactions. The theory states, in order to minimize DG
changes caused by these phenomena, that: (1) the compensation (by changing
Dh or Ds) arises from the nature of the interaction between the solute and
solvent causing the reaction; and (2) the relationship between the enthalpy and
entropy for a specific reaction is linear (Labuza 1980; Madamba et al. 1996).
The present work aims to: (1) obtain adsorption isotherms of pinhão flour
at selected temperatures by using mathematical models described in the lit-
erature; (2) determine some thermodynamic functions (differential enthalpy
and entropy); and (3) evaluate the application of the enthalpy–entropy com-
pensation theory. As information on pinhão flour has never been placed in
literature, the proximate composition of this product was also determined.

MATERIALS AND METHODS

Experimental Procedure

Materials. The chemicals were from Quimex (São Paulo, Brazil), Lab-
synth (Diadema, São Paulo, Brazil), Nuclear (Diadema) and Vetec (Duque de
Caxias, Rio de Janeiro, Brazil).
The Araucaria angustifolia seeds used in this study were purchased at a
local market (Porto Alegre, Rio Grande do Sul, Brazil). The seeds were
cleaned with abundant water, dried at ambient conditions for 24–48 h, selected
and stored at -20C in polythene bags.

Flour Production. First, the seeds of pinhão were defrozen at the

ambient condition. Then, the seeds were peeled manually with a knife, the
internal coat was removed and the seeds were cut into 5-mm-thick slices
approximately. After that, the slices were dried in a convective bin dryer
(designed in the Institute of Food Science and Technology of Federal Univer-
sity of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil) at 70C for
10 h and finally milled to flour in a variable speed rotor mill (Pulverisette 14
– Fritsch, Idar-Oberstein, Germany) to pass a Tyler N° 35 mesh screen
(425 mm).
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 829

Proximate Composition. Proximate composition (moisture, protein, fat,

ash, crude fiber, starch), total acidity and pH of the pinhão flour were deter-
mined. The total protein content was determined by the Kjeldahl method
according to the method 979.09 of AOAC (1990), using a Tecator digestor
(Perstorp Analytical Co., Höganäs, Sweden). The conversion factor used was
% nitrogen ¥ 5.75 (Cladera-Olivera et al. 2008). Ash was determined by incin-
eration in a muffle furnace (Temco, Desoto, IA) at 550C (method 923.03 of
AOAC 1990). Crude fiber was determined as ash after acidic and basic diges-
tion according to the Method 962.09 of AOAC (1990). Lipids were extracted
in a Soxhlet extractor (Tecnal, Piracicaba, São Paulo, Brazil) with petroleum
ether (6 h) without acid digestion and were determined gravimetrically. Starch
was determined by direct acid hydrolysis and by determining the amount of
glucose in filtrate with Lane-Eynon volumetric method (Pearson 1975). Mois-
ture content was calculated through the weight loss of a 10 g sample after
heated in a conventional oven at 105C (Biomatic 303, Porto Alegre, Rio
Grande do Sul, Brazil) until attaining constant weight. Aw (at 25C) was deter-
mined in a water activity instrument (AquaLab 3TE-Decagon, Pullman, WA).
For pH analysis, 10 g of the sample were mixed with 75 mL of water at 25C
and the pH was measured after 1 h with a digital pH-meter (Quimis Q-400M,
São Paulo, Brazil). Total acidity was determined with 0.1 N sodium hydroxide
titration, by using phenolphthalein as indicator, according to the Method
942.15 (AOAC 1990) and expressed as mL of NaOH 0.1 N/g. Analyses were
carried out in triplicate and results were expressed as average value ⫾ standard
deviation in percentage and in w.b.

Adsorption Isotherms. Adsorption isotherms were determined through

the employment of the static gravimetric method recommended by the COST
90 Project (Speiss and Wolf 1983) using 10 saturated salt solutions (lithium
chloride, potassium acetate, magnesium chloride, potassium carbonate, potas-
sium nitrite, sodium chloride, potassium chloride, barium chloride and copper
sulfate) with relative humidities ranging from 11 to 98% (Young 1967). For
experimental points with relative humidities less than 30%, flour samples were
first dried in a vacuum oven (Quimis Q819V, São Paulo, Brazil) to a level
containing 3% water contents (w.b.) to ensure adsorption process. Samples of
about 5 g of pinhão flour were placed in small plastic cups and then stored
inside sealed jars, having their respective solution for each different relative
humidity. The jars were placed in an air-circulating, thermally-insulated,
temperature-controlled equipment (BOD TE 381 Tecnal) maintained at the
specified temperature within ⫾ 0.3C, until equilibrium had been reached
(between 30 and 40 days). After the first 15 days, samples were weighed every
4 days in order to verify if equilibrium had been reached in analytical balance.
With the purpose of preventing microbial spoilage of samples, crystalline
830 F. CLADERA-OLIVERA ET AL.

TABLE 1.
SORPTION ISOTHERM MODELS USED IN THIS WORK

Name of the equation Equation

Aw 1 (C − 1) Aw
BET (Brunauer et al. 1938) = +
(1 − Aw ) ⋅ X Xm ⋅C Xm ⋅C
X m ⋅ C ⋅ K ⋅ Aw
GAB (Van den Berg 1985) X=
( 1 − K ⋅ Aw ) × (1 − K ⋅ Aw + C ⋅ K ⋅ Aw )

Aw = exp ⎛ − B ⎞
A
Halsey (Halsey 1948)
⎝ X ⎠
Peleg (Peleg 1993) X = k1 ⋅ Awn1 + k2 ⋅ Awn1
⎛ Aw ⎞
Oswin (Oswin 1946) X = A⋅⎜
⎝ 1 − Aw ⎟⎠
Chung–Pfost (Chung and Pfost 1967) ln(Aw) = -A · exp(-B · X)
Henderson (Henderson 1952) (l - Aw) = exp(A · XB)
Chirife (Chirife and Iglesias 1978) X = exp(A + B · ln[C - lnAw])
Smith (Smith 1947) X = A + (B · log[l - Aw])

A, B, C, K, k1, k2, n1, n2 and k are parameters of the equations.

BET, Brunauer, Emmett and Teller; GAB, Guggenheim–Anderson–de Boer.

tymol was placed in the jars where high water activities occurred (Aw > 0.7)
(Wolf et al. 1985). After the equilibrium had been reached, moisture content
was determined as described earlier.
Data Analysis

Isotherm Models. The experimental data for the adsorption isotherms

obtained was fitted to nine moisture sorption isotherm models presented in
Table 1 using the nonlinear regression module of Statistica 5.0 software (Stat-
soft, Tulsa, OK). Regressions were repeated with various initial estimated
values both above and below those that had been calculated to confirm whether
or not convergence was reliable (Peleg 1993). For Brunauer, Emmett and
Teller (BET) model only values of Aw < 0.5 were used. The goodness of fit of
the different models was evaluated with the coefficient of determination (r2)
and the MRD. MRD has been widely adopted throughout the literature to
evaluate the accuracy of the fit of isotherm models and is defined by:

100 N X ei − X pi
MRD = ∑ X
N i =1
(1)
ei

where Xei and Xpi are, respectively, the measured and predicted EMC (kg
water/kg dry solids) and N is the number of data points. The lower the values
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 831

of MRD the better fit of the model and MRD values below 10% generally
indicate an adequate fit for practical purposes (Lomauro et al. 1985).
Statistical differences between the isotherms were tested using Tukey’s
studentized range test for variable temperature.

Calculation of the Isosteric Heat (Differential Enthalpy) of

Adsorption and the Ds. The differential enthalpy and entropy of sorption can
be determined from moisture sorption data by using the following equation
(Fasina et al. 1997, 1999):

Δh Δ s
( ln Aw )X = + (2)
RT R

where Aw represents the water activity, Dh the differential enthalpy of sorption

(J/mol), Ds the differential entropy (J/mol·K), R the universal gas constant
(J/mol·K), X the EMC (kg water (kg/dry solids) and T the absolute temperature
(K).
The DH (J/mol) can be obtained from the following equation:

ΔH = Δ h + ΔH vap (3)

where DHvap is the latent heat of vaporization of free water (J/mol).

The differential enthalpy of sorption (Dh) was calculated with basis on
Eq. (2), where ln(Aw) was plotted versus 1/T for a certain moisture content
thus determining the slope of the curve, which is equal to –Dh/R; Ds is
obtained from the linear coefficient (Ds/R) of the straight (McMinn and
Magee 2003). The model that best describes the experimental sorption data
for pinhão flour was used to determine the Aw value for each moisture
content.

Enthalpy–Entropy Compensation Theory (Isokinetic Theory).

Ferro-Fontan et al. (1982) suggested the existence of a linear relationship
between enthalpy and entropy for water sorption in some foods and this is
confirmed by several other authors (Madamba et al. 1996; Telis et al. 2000;
McMinn and Magee 2003; Fasina 2006). The isokinetic temperature, evalu-
ated from the slope of the straight, represents the temperature at which all
reactions in series proceed at the same rate (Heyrovsky 1970). The compen-
sation theory proposes a linear relationship between Dh and Ds (Leffler and
Grunwald 1963; Telis et al. 2000; McMinn et al. 2005):
832 F. CLADERA-OLIVERA ET AL.

Δ h = Tβ ⋅ Δ s + ΔGβ (4)

where Tb (K) is the isokinetic temperature and DGb (J/mol) is the DG at Tb and
these values were obtained by linear regression. From a thermodynamic view-
point, the free energy (DG) can be used as an indicative of the sorbent affinity
for water, hence providing a criterion as to whether water sorption is a spon-
taneous (DG < 0) or nonspontaneous (DG > 0) process (Apostolopoulos and
Gilbert 1990).
As proposed by Krug et al. (1976a,b) a statistical analysis test can be used
to corroborate the compensation theory, by comparing Tb with the harmonic
mean temperature, defined as:

n
Thm = n
(5)
∑ (1 T )
1

where n is the total number of isotherms used.

The compensation theory only applies if Tb ⫽ Thm. An approximate (1-a)
100% confidence interval for Tb may be calculated from the following equa-
tion (Berinstain et al. 1996; Telis et al. 2000):

Tβ = Tβ ± tm −2,α 2 Var (Tβ ) (6)

where

Tβ =
∑ ( Δ h − Δ h ) ( Δ s − Δ s)
(7)
∑ ( Δ s − Δ s)
2

and

∑ ( Δ h − ΔGβ − Tβ ⋅ Δ s)
2

Var (Tβ ) = (8)

(m − 2)⋅ ∑ ( Δ S − Δ S )
2

where m is the number of (Dh, Ds) data pairs, Δh is the average enthalpy, Δs
is the average entropy and ΔGβ is the average DG.
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 833

RESULTS AND DISCUSSION

Proximate Composition
In order to prepare 1 kg of pinhão flour, 2.8 kg of pinhão (with coats)
were necessary as roughly 30% of the seed weight is represented by its coats
and its moisture content is very high (about 50% w.b.).
Moisture represents 5.83 ⫾ 0.02 % (w.b.) of the pinhão flour and the Aw
was 0.305 ⫾ 0.030 (at 25C). Starch is the main component with 63.9 ⫾ 0.5 %.
Other components included: protein, 5.29 ⫾ 0.06%; crude fiber, 1.54 ⫾
0.28%; lipid 2.35 ⫾ 0.05%; and ash 2.52 ⫾ 0.10%. These results show that
pinhão flour is a good source of complex carbohydrates, just like raw pinhão
(Cladera-Olivera et al. 2008). The acidity was found to be 3.02 ⫾ 0.07 mL of
0.1N NaOH and pH was equal to 5.7 ⫾ 0.1.

Adsorption Isotherms
Figure 1 shows the experimental values of EMC of pinhão flour as a
function of Aw at different temperatures. The experimental data represent the
mean of three replications. The standard deviation for the EMC (kg water/kg
dry solids) of each experimental point was within the range of 0.0003–0.008.
As expected, the EMC decreased as Aw decreased at constant temperature. The
adsorption isotherms of pinhão flour showed the typical sigmoid shape of Type
II, according to the BET classification. This is in agreement with the results
obtained for semolina and farina (Erbas et al. 2005), potato (McMinn and
Magee 2003; Kaymak-Ertekin and Gedik 2004), starch powders (Al-Muhtaseb
et al. 2004) and for many other products. The main component of pinhão flour
is starch (63.9% d.b.), which consists of both crystalline and amorphous
regions. The starch sorption isotherm is attributable to hydrogen-bonding of
water molecules to the hydrogen-bonding of water molecules to the available
hydroxyl groups of the substrate, i.e., those in the amorphous regions and on
the surfaces of the crystallites (Urquhart 1959). The crystalline regions typi-
cally exhibit resistance to solvent penetration. Hence, water affect is very small
and the mobility of the amorphous regions is restricted. However, as the Aw
increases, the sorbed moisture causes a subsequent swelling of the biopolymer,
the degree of crystallinity decreases, and there is an increase availability of the
polar groups to the water molecules and, finally, the swelled polysaccharide
forms a solution (Al-Muhtaseb et al. 2004).
The analysis of Fig. 1 also indicates that, in the range of the temperature
used (between 10 and 40C), the adsorption isotherms of pinhão flour were
close to each other at different temperatures, as obtained for raw pinhão
(Cladera-Olivera et al. 2008). However, statistical differences between iso-
therms were tested using Tukey’s Studentized range test and difference
834 F. CLADERA-OLIVERA ET AL.

0.7

0.6
Equilibrium moisture content (kg water (kg ds)-1)

10 C
20 C
0,5 30 C
40 C
Chirife 10 C
0.4 Chirife 20 C
Chirife 30 C
Chirife 40 C
0.3

0.2

0.1

0.0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Water activity (aw)

FIG. 1. EXPERIMENTAL VALUES OF EQUILIBRIUM MOISTURE CONTENT (DRY BASIS)

AS A FUNCTION OF WATER ACTIVITY FOR PINHÃO FLOUR AT
DIFFERENT TEMPERATURES

TABLE 2.
TUKEY’S STUDENTIZED RANGE TEST FOR VARIABLE TEMPERATURE

Temperature Lower confidence Difference Upper confidence Significant

comparison limit between means limit at 0.05

10–20 -0.009404 0.005929 0.021262

10–30 0.008297 0.023795 0.039294 *
10–40 0.027309 0.027309 0.057975 *
20–30 0.002367 0.017866 0.033365 *
20–40 0.021380 0.036713 0.052045 *
30–40 0.003348 0.018846 0.034345 *

(P < 0.05) was obtained between all temperatures except between 10 and 20C
(Table 2). As pinhão flour is rich in starch and protein, these components affect
the sorption data considerably, because protein and starch are macromolecules
rich in polar sites such as –OH and –H groups, which behave as active (polar)
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 835

TABLE 3.
ESTIMATED VALUES OF COEFFICIENTS AND MRD OBTAINED FOR SELECTED
SORPTION MODELS APPLIED TO EXPERIMENTAL ADSORPTION DATA FOR
PINHÃO FLOUR

Model Constant Temperature (C) Average

10 20 30 40

BET (with experimental Xm 0.0605 0.0604 0.0558 0.0525

points at Aw < 0.5) C 201.34 38.19 63.93 25.05
r2 0.9650 0.9722 0.9516 0.9632
MRD (%) 3.45 3.43 3.13 4.68 3.67
GAB Xm 0.0660 0.0604 0.0577 0.0517
C 241.3 70.5 89.8 40.2
K 0.8254 0.8768 0.8447 0.8978
r2 0.9991 0.9961 0.9894 0.9962
MRD (%) 1.27 4.31 4.13 4.43 3.54
Chirife A -2.34507 -2.40885 -2.51639 -2.57765
B -0.46009 -0.51916 -0.58317 -0.61225
C -0.00237 0.00156 0.01105 0.02152
r2 0.9998 0.9998 0.9946 0.9988
MRD (%) 0.99 2.18 6.44 3.77 3.34

BET, Brunauer, Emmett and Teller; GAB, Guggenheim–Anderson–de Boer.

sorptions centres (Erbas et al. 2005). For semolina, a product rich in both,
starch and protein, Erbas et al. (2005) found a significant effect of temperature
(between 20 and 60C) on sorption behavior; however, Hebrard et al. (2003)
reported that changes in temperature (between 25 and 45C) seemed to have no
effect on the isotherms of this product.
Table 3 shows the fitting constants of the selected models, the coefficient
determination and the MRD. Chirife model best describes the experimental
data throughout the range of temperature studied as obtained for raw pinhão
(Cladera-Olivera et al. 2008), followed by the Guggenheim–Anderson–de
Boer (GAB) model. The Chirife model gives MRD values ranging from 0.99
to 6.44% (average value 3.34%). Other empirical models (Halsey and Peleg)
also give good fits with average MRD of 3.59 and 4.14, respectively (result
not shown).
Treatments of sorption data according to the BET and GAB equations
(theoretical models) allow the evaluation of monolayer moisture content
values (Xm) of foods. This is an important parameter in food storage and
deterioration. The values of monolayer moisture content of pinhão flour esti-
mated with BET and GAB models can also be observed in Table 3 and are in
the acceptable range for food products. Labuza (1984) indicated 10% (d.b.) as
the maximum monolayer moisture content for foods. The monolayer values
836 F. CLADERA-OLIVERA ET AL.

8000

7000
Differential enthalpy (J mol-1)

6000

5000

4000

3000

2000

1000

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
-1
Moisture content (kg water kg ds)

FIG. 2. DIFFERENTIAL ENTHALPY OF ADSORPTION OF PINHÃO FLOUR AS A

FUNCTION OF EQUILIBRIUM MOISTURE CONTENT

decreased as temperature increased calculated by both equations. Similar

results were found by several authors (McLaughlin and Magee 1998; Ayranci
and Duman 2005; Erbas et al. 2005; Cassini et al. 2006). The Xm values
obtained using BET and GAB models were approximately equal to each other.
It has been observed by several authors that the monolayer moisture content by
BET is lower than the GAB value (Timmermann et al. 2001).

Isosteric Heat of Sorption (Differential Enthalpy) and Ds

The differential enthalpies of adsorption of pinhão flour were calculated
by applying Eq. (2). The Chirife model, which best describes the experimental
adsorption data, has previously been used to determine Aw values for each
EMC. Figure 2 shows the results obtained, where the differential enthalpy of
adsorption was plotted against the EMC. The analysis of this figure reveals that
the differential enthalpy has a strong dependence on moisture content at low
EMCs, as the energy required for sorption increases at low EMCs. Isosteric
heats of sorption are high at low moisture contents (<0.15 kg water/kg ds),
decrease sharply as the moisture content increases and become almost constant
at 0.3 kg water/kg ds. Similar results were obtained for desorption of raw
pinhão (Cladera-Olivera et al. 2008).
The Ds values for sorption at given moisture content were also calculated
by linear regression using Eq. (2). It was assumed that, at a specific moisture
content, Dh and Ds did not vary with temperature (Aguerre et al. 1986). These
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 837

20
18
Differential entropy (J mol-1 K-1 )

16
14
12
10
8
6
4
2
0
-2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Moisture content (kg water kg-1 ds)

FIG. 3. DIFFERENTIAL ENTROPY OF ADSORPTION OF PINHÃO FLOUR AS A FUNCTION

OF EQUILIBRIUM MOISTURE CONTENT

results are shown in Fig. 3, which presents the Ds as a function of the moisture
content. This figure indicates that the Ds also displays a strong dependence on
moisture content at low EMCs and increases sharply at moisture content below
0.15 kg water/kg ds.
Enthalpy–Entropy Compensation Theory
Figure 4 shows the differential enthalpy (Dh) versus Ds. As can be noted,
a linear relation between these variables was obtained (with a coefficient
determination r2 = 0.993), indicating that compensation exists. The parameters
Tb and DGb (Eq. 4) were calculated from the data by means of linear regres-
sion. The isokinetic temperature (Tb ) is the temperature at which all sorption
reactions will take place at the same rate. A value of 398 ⫾ 18 K (within a 95%
confidence interval) for adsorption of pinhão flour was found. For desorption
of raw pinhão Cladera-Olivera et al. (2008) found a value of 421 ⫾ 17 K. For
adsorption isotherms of starch materials, McMinn et al. (2005) found values
between 366.8 and 466.8 K. For quinoa grains, Tolaba et al. (2004) found a
value of 361.0 K, for garlic, Madamba et al. (1996) found a value of
348 ⫾ 9 K, for oatmeal biscuit and oat flakes, McMinn et al. (2007) found
values of 430.9 ⫾ 1.88 and 443.4 ⫾ 0.91 K, respectively. Values between 458
and 475 K were obtained for adsorption and desorption of persimmon skin and
pulp (Telis et al. 2000). The harmonic mean temperature (Eq. 5) was calcu-
lated and a value of 297.7 K was found. This value is significantly different
from Tb, confirming the suitability of the isokinetic theory.
838 F. CLADERA-OLIVERA ET AL.

9000

8000

7000
Differential entalpy (J mol )
-1

6000

5000 y = 397,83x + 338,74

2
r = 0,9932
4000

3000

2000

1000

-1000
-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
-1 -1
Differential entropy (J mol K )

FIG. 4. DIFFERENTIAL ENTHALPY AS A FUNCTION OF DIFFERENTIAL ENTROPY FOR

PINHÃO FLOUR

According to Leffler (1955), if Tb > Thm the process is enthalpy driven,

whereas if Tb < Thm the process is entropy controlled. As the former condition
is satisfied in this study, adsorption mechanism can be considered to be
enthalpy-driven. McMinn et al. (2005), Aguerre et al. (1986) and Berinstain
et al. (1996) also reported an enthalpy-controlled mechanism in the sorption of
starch materials and Cladera-Olivera et al. (2008) for desorption of raw
pinhão.
The DG can be used as an indicative of the affinity of the sorbent for
water, and further provide a criterion as to whether water sorption is a spon-
taneous or nonspontaneous process depending on the sign of the DG value
(Apostolopoulos and Gilbert 1990). In this work, the value of DG obtained
(339 J/mol) suggests that adsorption process is nonspontaneous (DG > 0). This
suggests that pinhão flour does not adsorb moisture from the ambient sponta-
neously. The lower the moisture adsorbed from the environment, the less the
likelihood of degradation and this is desirable to keep the product stable. For
adsorption isotherms of oatmeal biscuit and oat flakes McMinn et al. (2007)
also suggest a nonspontaneous process, as they found values of 280 and
325 J/mol, respectively. However, for adsorption isotherms of starch materials,
McMinn et al. (2005) suggests that the sorption process in most of the starchy
materials examined were spontaneous. The authors found values for DG
between –1525 and 69 J/mol. Berinstain et al. (1996) also reported spontane-
ous sorption isotherm for starch materials.
Concluding, within the temperature range studied (10 to 40C) and Aw
range of 0.11 to 0.98, the Chirife model was found to best represent the
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 839

experimental data. The differential enthalpy of adsorption was calculated and

found to decrease as the moisture content increased; and it was also found that
this variable became almost constant as moisture content of 0.3 kg water/kg ds
was approached; the Ds behaved similarly. Plots of differential enthalpy versus
Ds satisfied the enthalpy–entropy compensation theory (isokinetic theory).
Results suggested that adsorption processes in pinhão flour are enthalpy-
controlled (isokinetic temperature > harmonic mean temperature) and not
spontaneous (DG > 0).

NOMENCLATURE

a confidence interval
A, B, C isotherm models parameters
Aw water activity
DH total heat of sorption (J/mol)
DHvap latent heat of vaporization of free water (J/mol)
Dh isosteric heat of sorption or differential enthalpy (J/mol)
Δh average differential enthalpy (J/mol)
Ds differential entropy (J/mol·K)
Δs average differential entropy (J/mol·K)
DG free energy (J/mol)
DGb free energy at Tb (J/mol)
d.b. dry basis
K, k, k1, k2 isotherm models parameters
m number of (Dh, DS) data pairs
MRD mean relative deviation modulus
N number of data points
n number of isotherms
n1, n2 Peleg model parameters
R universal gas constant (8.314 J/mol·K)
r2 coefficient determination
t Student’s t
T temperature (K)
Tb isokinetic temperature (K)
Thm harmonic mean temperature (K)
w.b. wet basis
X equilibrium moisture content (kg water/kg dry solids)
Xm monolayer moisture content (kg water/kg dry solids)
Xe experimental equilibrium moisture content (kg water/kg dry
solids)
Xp predicted equilibrium moisture content (kg water/kg dry solids)
840 F. CLADERA-OLIVERA ET AL.

ACKNOWLEDGMENTS

The authors gratefully acknowledge the financial support of “Conselho

Nacional de Desenvolvimento Científico e Tecnológico” (CNPq) and
“Fundação de Apoio à Pesquisa do Rio Grande do Sul” (FAPERGS) from
Brazil.

REFERENCES

AGUERRE, R.J., SUAREZ, C. and VIOLLAZ, P.E. 1986. Enthalpy-entropy

compensation in sorption phenomena: Application to the prediction of
the effect of temperature on food isotherms. J. Food Sci. 51, 1547–
1549.
AL-MUHTASEB, A.H., MCMINN, W.A.M. and MAGEE, T.R.A. 2004.
Water sorption isotherms of starch powders Part 1: Mathematical descrip-
tion of experimental data. J. Food Eng. 61, 297–307.
AOAC. 1990. Official Methods of Analysis, 16th Ed., Association of Official
Analytical Chemists, Washington, DC.
APOSTOLOPOULOS, D. and GILBERT, S. 1990. Water sorption of coffee
solubles by frontal inverse gas chromatography: Thermodynamic consid-
erations. J. Food Sci. 55, 475–477.
AYRANCI, E. and DUMAN, O. 2005. Moisture sorption isotherms of cowpea
(Vigna unguiculata L. Walp) and its protein isolate at 10, 20 and 30C.
J. Food Eng. 70, 83–91.
BERINSTAIN, C.I., GARCIA, H.S. and AZUARA, E. 1996. Enthalpy-
entropy compensation in food vapor adsorption. J. Food Eng. 30, 405–
415.
BRUNAUER, S., EMMETT, P.H. and TELLER, E. 1938. Adsorption of gases
in multimolecular layer. J. Am. Chem. Soc. 60, 309–319.
CASSINI, A.S., MARCZAK, L.D.F. and NOREÑA, C.P.Z. 2006. Water
adsorption isotherms of texturized soy protein. J. Food Eng. 77, 194–199.
CHIRIFE, J. and IGLESIAS, H.A. 1978. Equations for fitting water sorption
isotherms of foods. 1. A review. J. Food Technol. 13, 159–174.
CHUNG, D.S. and PFOST, H.B. 1967. Adsorption and desorption of
water vapor by cereal grains and their products. Trans. ASAE 10, 549–
551.
CLADERA-OLIVERA, F., PETTERMANN, A.C., NOREÑA, C.P.Z.,
WADA, K. and MARCZAK, L.D.F. 2008. Thermodynamic properties of
moisture desorption of raw pinhão (Araucaria angustifolia seeds). Int. J.
Food Sci. Technol. 43, 900–907.
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 841

CORDENUNSI, B.R., MENEZES, E.W., GENOVESE, M.I., DOLLI, C.,

SOUZA, A.G. and LAJOLO, F.M. 2004. Chemical composition and
glycemic index of Brazilian pine (Araucaria angustifolia) seeds. J. Agric.
Food Chem. 52, 3412–3416.
ERBAS, M., ERTUGAY, M.F. and CERTEL, M. 2005. Moisture adsorption
behaviour of semolina and farina. J. Food Eng. 69, 191–198.
FASINA, O.O. 2006. Thermodynamic properties of sweetpotato. J. Food Eng.
75, 149–155.
FASINA, O.O., AJIBOLA, O.O. and TYLER, R.T. 1999. Thermodynamics of
moisture sorption in winged bean seed and gari. J. Food Proc. Eng. 22,
405–418.
FASINA, O.O., SOKHANSANJ, S. and TYLER, R.T. 1997. Thermodynamics
of moisture sorption in alfalfa pellets. Drying Technol. 15, 1553–1570.
FERRO-FONTAN, C., CHIRIFE, J., SANCHO, E. and IGLESIAS, H.A.
1982. Analysis of a model for water sorption phenomena in foods. J.
Food Sci. 55, 475–477.
HALSEY, G. 1948. Physical adsorption on non-uniform surfaces. J. Chem.
Phys. 16, 931–937.
HEBRARD, A., OULAHNA, D., GALET, L., CUQ, B., ABECASSIS, J. and
FAGES, J. 2003. Hydration properties of durum wheat semolina: Influ-
ence of particle size and temperature. Powder Technol. 130, 211–218.
HENDERSON, S.M. 1952. A basic concept of equilibrium moisture. Agric.
Eng. 33, 29–32.
HEYROVSKY, J. 1970. Determination of isokinetic temperature. Nature 227,
66–67.
KAYMAK-ERTEKIN, F. and GEDIK, A. 2004. Sorption isotherms and isos-
teric heat of sorption for grapes, apricots, apples and potatoes. Lebensm.-
Wiss. Technol. 37, 429–438.
KRUG, R.R., HUNTER, W.G. and GRIEGER, R.A. 1976a. Enthalpy-entropy
compensation 1. Some fundamental statistical problems associated
with the Vant’Hoff and Arrhenius data. J. Phys. Chem. 80(21), 2335–
2342.
KRUG, R.R., HUNTER, W.G. and GRIEGER, R.A. 1976b. Enthalpy-entropy
compensation 2. Separation of the chemical from the statistical effect. J.
Phys. Chem. 80(21), 2335–2342.
LABUZA, T.P. 1980. Enthalpy-entropy compensation in food reactions. Food
Technol. 34, 67–77.
LABUZA, T.P. 1984. Moisture Sorption: Practical Aspects of Isotherm Mea-
surement and Use. American Association of Cereal Chemists, St. Paul,
MN.
LEFFLER, J.E. 1955. The enthalpy-entropy relationship and its implications
for organic chemistry. J. Org. Chem. 20, 1202–1231.
842 F. CLADERA-OLIVERA ET AL.

LEFFLER, J.E. and GRUNWALD, E. 1963. Rates and Equilibria of Organic

Reactions. Wiley, New York, NY.
LOMAURO, C.J., BAKSHI, A.S. and LABUZA, T.P. 1985. Evaluation of
food moisture sorption isotherm equations. Part I: Fruit, vegetable and
meat products. Lebensm.-Wiss. Technol. 18, 111–117.
MADAMBA, P.S., DRISCOLL, R.H. and BUCKLE, K.A. 1996. Enthalpy-
entropy compensation models for sorption and browning of garlic. J.
Food Eng. 28, 109–119.
MCLAUGHLIN, C.P. and MAGEE, T.R.A. 1998. The determination of sorp-
tion isotherm and the isosteric heats of sorption for potatoes. J. Food Eng.
35, 267–280.
MCMINN, W.A.M. and MAGEE, T.R.A. 2003. Thermodynamic properties of
moisture sorption of potato. J. Food Eng. 60, 157–165.
MCMINN, W.A.M., AL-MUHTASEB, A.H. and MAGGE, T.R.A. 2005.
Enthalpy-entropy compensation in sorption phenomena of starch materi-
als. Food Res. Int. 35, 505–510.
MCMINN, W.A.M., MCKEE, D.J. and MAGEE, R.A. 2007. Moisture adsorp-
tion behaviour of oatmeal biscuit and oat flakes. J. Food Eng. 79, 481–
493.
OSWIN, C.R. 1946. The kinetics of package life III. The isotherm. J. Chem.
Ind. 65, 419–421.
PEARSON, D. 1975. Laboratory Techniques in Food Analysis. Butterworth &
Co. Ltd., London, U.K.
PELEG, M. 1993. Assessment of a semi empirical four parameter general
model for sigmoid moisture sorption isotherms. J. Food Proc. Eng. 16,
21–37.
SMITH, S.E. 1947. The sorption of water vapour by high polymers. J. Am.
Chem. Soc. 69, 646.
SPEISS, W.E.L. and WOLF, W.F. 1983. The results of the COST 90 project on
water activiti. In Physical Properties of Foods (R. Jowitt, ed.) pp. 65–91,
Applied Science Publishers, London, U.K.
TELIS, V.R.N., GABAS, A.L., MENEGALLI, F.C. and TELIS-ROMERO, J.
2000. Water sorption thermodynamic properties applied to persimmon
skin and pulp. Thermochim. Acta 343, 49–56.
TIMMERMANN, E.O., CHIRIFE, A. and IGLESIAS, H.A. 2001. Water
sorption isotherms of foods and foodstuffs: BET or GAB parmeters? J.
Food Eng. 48, 19–31.
TOLABA, M.P., PELTZER, M., ENRIQUEZ, N. and POLLIO, M.L. 2004.
Grain sorption equilibria of quinoa grains. J. Food Eng. 61, 365–371.
URQUHART, A.R. 1959. Sorption of water by cellulose and starch. In Recent
Advances in the Chemistry of Cellulose and Starch (J. Honeymand, ed.)
Heywood & Company, London, U.K.
 ADSORPTION ISOTHERMS OF PINHÃO FLOUR 843

VAN DEN BERG, C. 1985. Development of B.E.T. like models for sorption of
water of foods, therory and relevance. In Properties of Water in Foods (D.
Simatos and J.L. Multon, eds.) pp. 119–135, Martinus Nijhoft, Dordrecht,
the Netherlands.
WOLF, W., SPIESS, W.E.L. and JUNG, G. 1985. Standardization of iso-
therms measurements (COST-PROJECCT 90 and 90 bis). In Properties
of Water in Foods in Relation to Quality and Stability (D. Stimatos and
J.L. Multon, eds.) pp. 661–679, Martinus Nijhoff Publishers, Dordrecht,
the Netherlands.
YOUNG, J.F. 1967. Humidity control in the laboratory using salt solutions –
a review. J. Appl. Chem. 17, 241–245.
ZANDAVALLI, R.B., DILLENBURG, L.R. and DE SOUZA, P.V.D. 2004.
Growth response of Araucaria angustifolia (Araucariaceae) to inocula-
tion with the mycorrhizal fungus Glomus clarum. Appl. Soil Ecol. 24,
245–255.