Professional Documents
Culture Documents
Modeling Water Adsorption Isotherms of P PDF
Modeling Water Adsorption Isotherms of P PDF
843
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
ABSTRACT
PRACTICAL APPLICATIONS
INTRODUCTION
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.
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.
TABLE 1.
SORPTION ISOTHERM MODELS USED IN THIS WORK
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])
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
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.
Δh Δ s
( ln Aw )X = + (2)
RT R
ΔH = Δ h + ΔH vap (3)
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
Tβ =
∑ ( Δ h − Δ h ) ( Δ s − Δ s)
(7)
∑ ( Δ s − Δ s)
2
and
∑ ( Δ h − ΔGβ − Tβ ⋅ 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
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)
TABLE 2.
TUKEY’S STUDENTIZED RANGE TEST FOR VARIABLE TEMPERATURE
(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
10 20 30 40
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)
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)
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
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 )
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
REFERENCES
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.