Estimation of the diffusion coefficient of water in

henna leaves by adjustment of drying curves

Khamsa SGHAIER Mohamed BAGANE

Unité de recherché, Thermodynamique Appliquée Unité de recherché, Thermodynamique Appliquée
Université de Gabès, Ecole Nationale d’Ingénieurs Université de Gabès, Ecole Nationale d’Ingénieurs
Gabes, Tunisia Gabes, Tunisia
[email protected] [email protected]
Roman PECZALSKI
Laboratoire d’Automatique et Génie des Procèdes (LAGEP)
UMR CNRS 5007, UCB Lyon 1/ESCPE Lyon, Bât. 308 G, 43, Bd. 11 Nov. 1918, 69621 Villeurbanne Cedex
Lyon, France

Abstract—This electronic document is a “live” template and Dynamic Steam Sorption (DVS) has been used in
already defines the components of your paper Simulation and numerous research works. Buckton and Darcy (1995) [2]
optimization of the drying process may require the calculation of studied the water sorption properties of amorphous and
the water diffusivity of the product. In this work we have crystalline α-lactose monohydrate spray dried lactose mixtures.
described the estimation of the coefficients of diffusion of the Mackin et al. (2002) [3] used the DVS technique to quantify
moisture in the henna leaves from the drying curves obtained the amount of amorphous material in pharmaceutical solids.
with DVS device. Two different methods can be used to evaluate
this coefficient (analytical method and numerical method). In this FIG. 1 illustrates the diagram of this apparatus
application note, we discussed the method based on the analytical
solution of a Fick diffusion model which allowed the
determination of the diffusion coefficient by simple linear
regression on the experimental data. The results obtained showed
that the diffusivity is not strongly dependent on the water content
of the product. On the other hand, it increases well with the
increase in the temperature of the product.

Keywords— Henna leaves; Drying; Sorption isotherm; Water

diffusivity.

I. INTRODUCTION (HEADING 1)
The aromatic and medicinal plants, which are produced in
many countries in the industrial world under very varied forms,
are an inexhaustible spring of molecules. Molecules stemming
from these plants are often likened to active ingredients
possessing specific properties which confer them an only Figure 1. Principle of the DVS. [MANUEL DVS ET, Université Lyon 1,
character. [1]. LAGEP. SMS (Surface Measurement Systems)]

The aims of this work were to provide experimental This device consists of a microbalance that can measure
sorption isotherm data of henna leaves at various temperatures mass variations of the order of 5 μg. This microbalance allows
and relative humidity to estimate the water diffusivity of henna the differential weighing of two crucibles (sample and
leaves at 4 levels of temperatures (30, 40, 50 and 60°C). reference port) in a temperature-controlled chamber.
The sample is subjected to a stream of nitrogen charged
II. MATERIALS AND METHODS water vapor, the percentage of which is varied at the analysis
temperature. We have applied a complete desorption-
A. DVS device adsorption cycle by varying the relative humidity of the gas
The apparatus we used is a dynamic steam sorption (DVS) stream by 15% by weight from 90% to 0% and then from 0%
apparatus. This is a gravimetric measurement where the mass to 90%. The apparatus is computer-controlled, allowing
of the sample, subjected to given ambient humidity and continuous recording of temperature, gas moisture and sample
temperature conditions, is recorded in time to equilibrium mass during the experiments. [4]
(constant mass).
B. Sample preparation For a small diffusion time, equation (2) becomes:
Henna leaves natural coloring agent (Lawsonia inermis)
X 4D
was used in the experiments. All henna was supplied by the  1  2 t  tcr (3)
same producer in order to maximize the reproducibility of the X cr L
results. These leaves were harvested by hand and stored in For a long diffusion time equation (2) becomes:
aluminum foil in the refrigerator at -5 ° C. Before drying the
henna is cut into slices in the form of a rectangle of very small X 8 D 2 (t  tcr
dimensions. ln  ln 2  )(4)
Xcr  4 L2

C. Method The plot of the curve

X
as a function of the square root
The estimation of the diffusivity of the water from the X cr
drying curve has been described in numerous publications (t  t cr ) X (t  t cr )
of silk of ln as a function of makes it
including Hassini. L et al in 2006 [5]. Two methods of data L2 X cr L2
processing have been applied. The first was based on the possible to determine the diffusion from the slope of the linear
linearized analytical solutions of Fick's equation, with the part of the curves. The water content Xcr and the time tcr
assumption of constant diffusivity. The main advantage of this correspond to the beginning of the linear part of the curve for
method is its simplicity. The second method was based on the the period considered.
numerical solution of a complete model combining the solid
transfer equation with the moisture transfer equation. In our study, we exploited the beginning of the drying
curve, that is, the period of 'short time'.
In this work, the simplified analytical method was used.
Indeed, L. Hassini et al adopted the following assumptions: For each relative humidity step, X / Xcr is plotted as a
function of the square root of t as shown in FIG. 3. From this
 Unidimensional transfer curve, the diffusion coefficients are calculated and are given in
 Constant diffusivity Table 1 below.

 Sample plan
With these assumptions, the Fick diffusion equation is
written as follows.

X  2 X
 2 (1)
t y

With
X: water content of the product
t: Time
D: water diffusivity of the product Figure 3. The relative water content as a function of the
square root of time for a 90% moisture steps and a temperature
Y: spatial coordinate of 50°C.
With the hypotheses adopted and by averaging over the
thickness of the sample, the solution of this equation for III. RESULTS AND DISCUSSION
desorption is:
FIG. 2 shows the mass variations of product in sorption and
exp 
 2 D  desorption in the henna leaves at different stages of the relative
X 8 1
 2n1
2


 2 
t (2)
 21
X ini  2 n0 2 humidity ranging from 90% to 0% and from 0% to 90% by
 4 L  15% plateau for three levels of Temperatures. Relative gap on
m  mréf
With the ordinate is defined as : where mref is the mass of the
mréf
X : Average water content of the sample. dry product.
X ini :: Initial water content

L : Thickness of sample
This equation can be linearized for two distinct periods. At
the beginning of diffusion processes (low diffusion time) and at
the end of the diffusion process (long diffusion time).
 Figure2. Kinetics of sorption and of desorption in the henna
leaves at 30, 40 and 50°C obtained in the device DVS.
Table 1 presents the values of the water diffusivity at a
given temperature for different water contents of the product
and at a given level for different temperatures. The indicated
moisture content is the one of the product at the beginning of to
mitigate of considered humidity.
TABLE I. DIFFUSION COEFFICIENTS FOR HENNA LEAVES
Effect of temperature Effect of moisture content
(HR 60%) (T 60°C)
T (°C) D (m2/s) HR (%) D (m2/s)

90
30 0.0510-10 0.11810-10
Xin =1.81
75
40 0.08610-10 Xin =0.48 0.30710-10

60
50 0.15610-10 Xin =0.17 0.37610-10

45
60 0.37610-10 Xin =0.09 0.4110-10

The data presented above show that the diffusivity

increases with increasing temperature. On the other hand, it
varies little with the water content of the product.
These values were comparable to a few others reported in
the literature: Drying green beans 2,641 to 5,711 m2 / s in the
ambient temperature range of 50 - 70 ° C [6], Drying gombo
4.27 to 1.3 in the temperature range of 50 - 70 ° C [7], drying
of mint at 2.15 to 2.61 m2 / s in the relative humidity range
between 31 and 90% [8].

CONCLUSION
The paper describes a method for determining the diffusion
constants of water in henna leaves. The proposed method is
based on the use of an apparatus of the DVS type and the
treatment of the drying curves obtained for different humidities
and ambient temperatures. This method allowed determining
the diffusivity values for different water contents and product
temperature.
 The results show that the diffusivity is mainly dependent d’alumines : influence des différents paramètres tels l’humidité ou le
on the drying temperature. dopage en magnésie. Thèse Génie de Procédés et Environnement.
Institut National Polytechnique de Toulouse, France.
[5] Hassini. L ; Azzouz. S; Peczalski . R; Belghith. A,. (2006). Estimation
REFERENCES of potato moisture diffusivity from convective drying kinetics with
[1] Cháfer, M., González-Martinez, C.,Chiralt, A., Fito, P., 2003. correction for shrinkage. Journal of Food Engineering 79, 47–56.
Microstructure and vacuum impregnation response of citrus peels. Food [6] I. Doymaz, ‘Thin- Layer Drying Behaviour of Mint Leaves (Mentha
Research International. 36, 35-41 Spicata L.)’, Journal of Food Engineering, Vol. 74, pp. 370 – 375, 2006.
[2] Buckton G and Darcy P (1995). The use of gravimetric studies to assess [7] F. Kaymak-Ertekin, ‘ Drying and Rehydrating Kinetics of Green and
the degree of crystallinity of predominantly crystalline poweders. Red Peppers’, Journal of Food Science, Vol. 67, N°1, pp. 168 - 175,
International Journal of Pharmaceutics 123:265-271. 2002.
[3] Mackin L, Zanon R, Park JM, Foster K, Opalenik H, and Demonte M [8] A. Aghfir, S. Akkad, M. Rhazi, C.S.E. Kane et M. Kouhila.
(2002). Quantificaton of low levels (<10%) of amorphous content in Détermination du coefficient de diffusion et de l’énergie d’activation de
micronized active batches using dynamic vapor sorption and isothermal la menthe lors d’un séchage conductif en régime continu. Revue des
microcalorimetry. International Journal of Pharmaceutics 231:227-236. Energies Renouvelables Vol. 11 N°3 (2008) 385 – 394
[4] Christelle TISSERAND. (2007). Etude par chromatographie gazeuse
inverse des propriétés de surface de deux formes christallographiques