International Journal of Thermal Sciences 79 (2014) 103e110

Contents lists available at ScienceDirect

International Journal of Thermal Sciences

journal homepage: www.elsevier.com/locate/ijts

Infrared drying of water based varnish coated on elastomer substrate

Nadine Allanic a, *, Pascal Le Bideau b, Patrick Glouannec b, Alain Bourmaud b
a
b

LUNAM Universit, IUT de Nantes, CNRS, GEPEA, UMR 6144, BP 539, 44475 Carquefou Cedex, France
Laboratoire dIngnierie des Matriaux de Bretagne, Universit Europenne de Bretagne, BP 92116, 56321 Lorient Cedex, France

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 4 August 2012
Received in revised form
17 September 2013
Accepted 14 January 2014
Available online 14 February 2014

This study deals with the drying and curing of polyurethane water based varnish by infrared radiation.
The varnish is thinly coated on rectangular elastomer substrates. After characterizing their main thermophysical properties, the curing rate is linked to the thermal behavior of varnish. A laboratory setup is
developed to retrieve mass and temperature evolutions. First, drying experiments with a constant
infrared radiation inferior to 20 kW m
2 are performed. The thermal and hydric behaviors of the product
are analyzed in term of drying time and heating rate. Then, several experiments with modulated infrared
radiation are carried out. The impact of drying conditions on curing rate of varnish is then discussed.
2014 Elsevier Masson SAS. All rights reserved.

Keywords:
Experiment
Nanoindentation
Calorimetric analysis
Numerical model
Inverse method
Crosslinking

1. Introduction
In industry, rubber proles designed for the automotive sector
are produced on extrusion lines. During its manufacturing, the
extruded product successively undergoes many thermal operations. One of them consists in drying and reticulating a thin lm of
varnish applied on the prole surface. The reticulated varnish
protects the prole of its environment (light, humidity), increases
its water tightness, accentuates the product brilliance or increases
the abrasion protection.
The topical environmental challenges encourage industries to
use water based varnishes instead of organic based varnish [1,2].
Thus, the drying step needs a larger amount of energy. A classical
solution is to add in the industrial process a supercial radiative
heat source, such as infrared technologies [3,4]. Previous works
[5,6] have shown the difculty to dry aqueous polymer solution or
dispersion without forming an impermeable skin at the surface of
the solution. The use of supercial heating, in case of thin polymer
lm drying such as varnishes, needs a good control of energy inputs
to avoid material deterioration.
This work deals with the infrared heating of a water based
varnish containing polyurethane, coated on an elastomer substrate.
Due to several physical phenomena involved in the drying (evaporation, shrinkage, crystallization, crosslinking.) [7,8], the product
* Corresponding author. Tel.: 33 (0) 2 28 09 20 26; fax: 33 (0) 2 28 09 20 17.
E-mail address: nadine.allanic@univ-nantes.fr (N. Allanic).
1290-0729/$ e see front matter 2014 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ijthermalsci.2014.01.013

is very sensitive to energy inputs. Thus, to understand well the

products behavior, an experimental laboratory setup, enabling to
obtain the drying kinetics of thin coat during drying under infrared
radiation, is involved. First, the inuence of level of infrared radiation on the heat and mass transfers during the drying is studied. In
parallel, a comparison of drying with short-wave and mediumwave infrared radiations is performed. After examining the results, an experimental investigation which consists in modulating
the infrared irradiation to dry the product at a constant level of
temperature is presented. The nal state of crosslinking is linked
with the mechanical properties of the varnish by a nanoindentation
method, which conrms the possibility to improve the quality of
the nal product by the control of the infrared inputs.
2. Material and techniques
2.1. Experimental dryer and procedure
The experimental dryer enables to perform experiments with
infrared heating combined with natural convection (Fig. 1). The
infrared emitters are placed at the upper side of the chamber. Experiments were performed separately with short-wave and
medium-wave infrared emitters having a low thermal inertia. The
nominal powers of the lamps are respectively of 2 kW for shortwave infrared lamp and 1.5 kW for medium-wave infrared lamp.
Each lamp is tted out with a parabolic reector to obtain a homogenous radiation on the product surface. A power controller unit

104

N. Allanic et al. / International Journal of Thermal Sciences 79 (2014) 103e110

Fig. 1. Experimental setup.

operating in syncopated wave train controls the electrical power

supplied to the emitter [9]. This unit is controlled by an input
voltage (0e10 V). The infrared radiation received by the product is
measured using a sensor developed at the laboratory for both kinds
of emitters [10].
The laboratory dryer was designed in order to measure the
evolutions of mass and temperatures. Thus, the substrate is placed
on a weighting system, located in the chamber. Several precautions
are necessary to limit disturbances of mass measurements during
the drying stage. Two glass plates are located between the emitters
and the sample to avoid convection phenomena involved by
emitters fan inside the chamber. It is also important to avoid the
increase of air temperature around the sample during drying. Thus,
the chamber is a closed enclosure, whose walls are cooled down
with a water circulation. The sample is surrounded with a plate
cooled down with the same water circulation.
The sample, schematized in Fig. 1, is constituted of an elastomer
plate of dimensions 100  50  4 mm. A spray gun is used to apply a
thin lm of varnish, corresponding to a thickness of 80 mm, at the
upper surface of the substrate. In order to record the temperature
evolutions of the substrate, three K-type thermocouples of diameter 200 mm are placed inside the elastomer: at the surface (T1),
2 mm-deep (T2) and 4 mm-deep (T3). An optical pyrometer measures the surface temperature (Tsurf) of the varnish. The emissivity
is assumed to be constant and equal to 0.93. The measurement of
air temperature is also performed by a K-type thermocouple
inserted in a cylindrical radiative shield. All sensors are connected
to an acquisition system.
Despite the signicant precautions taken to avoid it, a drift of
the weighting caused by the increase of the air temperature inside
the chamber is observed. Thus, before each experiment, a no-load
test (substrate without varnish and without thermocouples) is
performed to correct it. Then, at each level of infrared radiation, an
experiment is performed with a rst instrumented coated substrate
to measure the temperature evolutions. A second experiment is
performed using a coated substrate without thermocouples to

avoid disturbances on the mass measurement. Using this conguration, the precision on mass measurement is estimated to 30 mg.

2.2. Material description

A water-based black varnish containing polyurethane is studied.
A thermogravimetric analysis (TGA), performed at 2  C/min,
showed that water is not the only volatile constituent of the
varnish [11]. Several weight losses were observed before 100  C,
corresponding to an evaporation of solvent. The density of its
dried and reticulated base, measured with a pycnometer, is
rvd 1120 kg m
3. The varnish density is expressed by Ref. [12]:

rv rvd

1W
1 jW

(1)

with W the moisture content in dry basis.

Fig. 2. Calorimetric analysis of dried varnish from 20  C to 240  C at heating rate of

20 K/min.

N. Allanic et al. / International Journal of Thermal Sciences 79 (2014) 103e110

To characterize the curing phenomena, a differential scanning

calorimetry analysis (DSC) was investigated on samples of varnish
dried three hours at 80  C. Two thermal cycles were carried out in a
temperature range of 20  Ce240  C at a heating rate of 20  C/min.
Product can be assumed to be not cured during the rst temperature cycle and cured during the second one. Fig. 2 presents the
obtained heat ows. Below 120  C, the observed peaks correspond
to evaporation of residual solvent. The comparison of the rst and
second increases of temperature shows several exothermic peaks
after 120  C, corresponding to curing. Moreover, the peak centered
around 115  C is present at the rst and second temperature cycle,
which is characteristic of the crystallization of one of the constituents of the varnish.
Therefore, the crosslinking reaction was studied by means of a
second calorimetric analysis, performed on varnish with its
initial moisture content. The varnish was heated from the
ambient temperature to 180  C according to a temperature increase of 10  C/min. Then, the temperature remains constant
during 10 min. Two temperature cycles were performed. During
the rst cycle, the varnish is dried and reticulated. Fig. 3 presents
a zoom of this analysis after the evaporation stage of water and
solvents. The difference of heat ow levels between both temperature cycles is mainly due to the reticulation reaction. After
10 min with a constant temperature of 180  C, the heat ows
become equivalent. The varnish is thus totally reticulated. This
analysis shows that the kinetic of the reticulation may be
deduced from the temperature evolution of the product during
drying.
The isotherm desorption (water activity of a product) of the
product was measured at a temperature of 60  C. Varnish was
introduced in a dish of 30 mm and placed into a climatic room,
which humidity and temperature were regulated [4]. Fig. 4 shows
that the evolution of water activity decreases for low moisture
content. Indeed, the critical moisture content is located between
WC 0.1e0.15 kg kg
1. The isotherm desorption was expressed
using a GAB model [13e15]:

Wm CKaw
1
Kaw 1
Kaw CKaw

(2)

with the tted parameters Wm 0.0152, C 0.6139 and K 0.9887.

The studied elastomer contains about 28 wt% EPDM gum, 39 wt
% reinforcing agent (carbon black, chalk.) and about 33 wt% of
other components (oil, sulfur, activators, accelerators, antioxidant.). Its properties were previously determined [9] and are
given in Table 1.

105

Fig. 4. Isotherm desorption (water activity) of varnish measured at 60  C.

2.3. Nanoindentation measurements

Nanoindentation tests were investigated, involving the contact of an indenter on the samples surface and its penetration to
a specied load or depth. Using the load versus penetration
depth responses, effective elastic modulus and hardness were
calculated [16,17]. Indentation tests were performed with a
commercial nanoindentation system (Nanoindenter XP, MTS
Nano Instruments) at room temperature (23  1  C) with a
continuous stiffness measurement (CSM) technique. In this
technique, an oscillating force at controlled frequency (70 Hz)
and amplitude (3 nm) was superimposed onto a nominal
applied force. After the indenter made contact with the surface,
it was driven into the material with a constant strain rate,
0.05 s
1 to a depth of 5000 nm; the load was held at its
maximum value for 60 s; nally, the indenter was withdrawn
from the surface with the same rate as loading until 10% of the
maximum load was reached. The modulus values have been
obtained from the unloading segment according to the work of
Oliver and Pharr. [18].
2.4. Mathematical modeling
Model is briey presented here since it was previously described
[4,23]. Mass transfers are unidirectional and varnish temperature is
homogenous. Inside the product, only liquid diffusion is considered. The evaporation (Fm) is located at the products surface and
involves a linear shrinkage (j). By assuming a local thermodynamical equilibrium [19,20], the drying rate Fm is usually given by
Refs. [21,22]:

Fm km






Pt Mv
Pt
Pv T
ln
Pt
aw Pvsat T
RT

(3)

Governing equations are expressed in a dimensionless form and

in a xed base linked to the dried product (0 < x < ed) to ease
numerical resolution. Thus, the mass balance is written [4]:

vW *
v
*
vt *
vx

D* W; T

vW *

!
(4)

2

*
vx
1 jW0 W *

with D* D/D0, W* W/W0, t* tD0/e2d, and x* x/ed.

Table 1
Thermophysical properties of elastomer substrate.

Fig. 3. Calorimetric analysis of varnish.

l (W m
1 K
1)

r (kg m
3)

Cp (J kg
1 K
1)

0.32

0.73 T 1215

2.73 T 1420

106

N. Allanic et al. / International Journal of Thermal Sciences 79 (2014) 103e110

The mass diffusion coefcient is expressed as follows [23,24]:

ERTa

DW; T D1 D2 e

e
W
b

(5)

with D1, D2, Ea and b constant values.

The parameter D1 was introduced in the expression in order to
avoid the divergence of numerical solving when the moisture
content tends to zero (end of the drying).
The mass transfer boundary condition at the varnish surface is
expressed by:

x* 1 Fm 1 JW2
r0v D* D0

W0 vW

(6)

ed vx

The assumption of impermeable substrate results in the

following boundary condition:

vW *

x* 0

vx

(7)

The heat balance for the varnish is given by:

 D0 vT *

rv cp;v e

e2d vt *





*4
hc Ta*
T * s Tw

T *4

1
f
Tl
T0 s

av EIR
Fm Lv
(8)
with T* T
T0/Tl
T0, Ta* Ta
T0 =Tl
T0 and
* T
T =T
T .
Tw
w
0 l
0
Tl corresponds to a limit temperature and s is the conductive
heat transfer at the substrate/varnish interface. T0 is the initial
temperature of the varnish.
An efcient heat capacity of dried varnish cp,vd is deduced from
calorimetric analysis and introduced into the model (Fig. 2). Thus,
the heat capacity of varnish cp,v is expressed by:

cpv cp;vd

1
W
cp;w
1W
1W

(9)

Due to its radiative properties [11], the infrared radiation av EIR is

assumed absorbed at the varnish surface. Thus, the transfer of energy inside the substrate is mainly due to conductive transfer. The
experimental temperature (T3) is introduced as a boundary condition inside the model for the bottom face. However, numerical
studies conducted in parallel have shown that this temperature can
be numerically determined by assuming that the substrate is a
semi-innite medium.
The initial conditions are:

t* 0

W* 1

T* 0

3.1. Preliminary drying experiments

In this experimental investigation, samples were exposed to
infrared radiation until their temperature reached the limit temperature Tl 180  C. Considering the calorimetric analysis (Figs. 2
and 3), the varnish is partially cured at this temperature. Fig. 5
presents the experimental drying kinetics obtained with a short
infrared irradiation of Esw1 10.2 kW m
2. In order to analyze the
impact of varnish presence on the thermal response, the measured
temperature of the substrate without varnish (noted no varnish
on Fig. 5) for the same solicitation (Esw1) is also reported. Moreover,
the evolution of air temperature inside the chamber, which progressively increases to reach 36  C at the end of experiment, is also
given.
During the rst 20 s, the temperature of product remains constant, which corresponds to the period before starting the emitters.
Ten seconds after, the gradient of temperature inside the product,
varnished or not, reaches 25  C. For the substrate without varnish,
the elastomer surface temperature, measured with thermocouple
T1 and optical pyrometer (Tsurf), are similar. The temperature difference stays lower than 1 K. In the case of varnished substrate, due
to the energy used for water evaporation, the surface temperature
remains relatively constant while mean moisture content begins to
considerably decrease. This step corresponds to a water activity
product close to 1 (Fig. 4). During this step, both temperatures Tsurf
and T1 are similar. Then, one observes a difference less than 1 K
which progressively increases and reaches 10 K at 250 s. This result
shows that the surface temperature is certainly underestimated.
Indeed, in this work, the emissivity of the product is assumed to be
constant. In fact, it decreases all along the drying due to water
evaporation which has a consequence on measured temperature by
optical pyrometer. In another way, it is possible that the thermocouple T1 receives a part of infrared irradiation and thus gives an
overestimated temperature. In the elastomer, the product temperature at the middle position (T2) and on the bottom face (T3) is
progressively lower for the varnished substrate compared to not
varnished substrate. At the end of the experiment, the gaps are
respectively of 5  C for T2 and 7  C for T3. After 70 s, the mean
moisture content of varnish begins to slowly decrease, which corresponds to the third stage of drying, when activity is lower. Thus,
the surface temperature increases to reach a value close to the case
of substrate heated without varnish.
Thus, the experimental curves are used to determine the global
absorptivity of varnish towards emitters infrared radiation. At the
beginning of heating, the substrate can be considered as a semi-

(10)

A control volume method is used to spatially integrate the differential equations. All the equations were simultaneously solved
by means of a predictorecorrector method with a high order. The
coupling and the nonlinearity of equations were solved by a
Newton method. Forty control volumes were considered in the
varnish (only for mass balance) and twenty control volumes in the
elastomer substrate. The time step was variable to ensure the
convergence of calculations.
3. Results and discussion
Several congurations of drying have been tested and are now
discussed. A mass of 415  35 mg is applied on substrate, which
corresponds to a thickness of 80  7 mm.

Fig. 5. Evolution of temperature obtained for substrate with and without varnish for
short-wave infrared with Esw1 10 kW m
2.

N. Allanic et al. / International Journal of Thermal Sciences 79 (2014) 103e110

107

innite medium, with constant thermophysical properties. We

assume that convective and long wave radiative exchanges are
negligible comparable to incident infrared radiation. Knowing
substrate absorptivity (as 0.57) [9], varnish absorptivity is estimated using the temperature evolutions of substrate with and
without varnish thanks to the following relation:

av as

v =dt
dTsurf
s =dt
dTsurf

(11)

Mean absorptivity av calculated during the period t 5 s and

t 20 s is av 0.64. A similar analysis carried out on dried product
gives an identical value.
3.2. Curing with short-wave and medium-wave infrared radiation
The inuence of the spectral domain of the emitters (short-wave
and medium-wave) is now studied. Indeed, it avoids moving
the emission spectrum to medium-wave infrared wavelengths.
Experiments were performed with infrared irradiation of
Esw1 10.2 kW m
2 and Esw2 15.4 kW m
2 corresponding
respectively to 40% and 60% of control voltage. For medium-wave
infrared
lamps,
the
infrared
radiations
tested
are
Emw1 10 kW m
2 and Emw2 13 kW m
2, corresponding
respectively of 70% and 100% of control voltage.
Fig. 6 presents the evolutions of product temperature (6a and
6b), mean moisture content and drying rate (6c), obtained for the
four tests previously described. Experiments were stopped when
the product temperature reached 180  C to avoid its deterioration.
First, the comparison of evolution of short-wave infrared radiation
(6a) and medium-wave infrared irradiation (6b) shows a quicker
setting period for short-waves emitters. For each test, the three
steps in heating previously described are also observed. For the
same infrared irradiation, even if the varnish temperature increases
slightly quicker with short-wave infrared than with medium-wave
infrared, the behavior of varnish is globally the same. When the
drying rate decreases, differences begin to appear between the two
drying kinetics. At 350 s, when the drying can be considered as
over, the moisture content is W 0.78 kg kg
1 for Esw1 and
W 0.98 kg kg
1 for Emw1. Esw1 10.2 kW m
2 is slightly higher
than Emw1 10 kW m
2. Thus, the surface temperature increases
slightly higher in the rst case. At the end of drying, Tsurf is equals to
180  C with Esw1 and only 175  C with Emw1. The consequence of
this temperature increase could be a lower nal moisture content.
However, the precision on mass measurements was estimated to
Dm 30 mg, which involves an uncertainty on moisture content
lower than DX 0.25 kg kg
1. Thus, it is difcult to interpret further
the difference observed between both mass evolutions at the end of
drying. Considering the drying rate, it clearly appears that the level
of radiation is the essential element in short-wave and mediumwave infrared. Indeed, the higher drying rate is obtained with the
infrared irradiation Esw2.
3.3. Drying experiments with a temperature set point
Drying experiments are performed by regulating the surface
temperature at different set points. The maximum of infrared
irradiation is limited at 10.2 kW m
2 and is controlled using a
proportional corrector. The aim is to obtain samples with different
nal moisture content to be analyzed by nanoindentation method.
Table 2 gives the condition of experiments performed for three
levels of bound temperature (130  C, 150  C and 180  C). Fig. 7a
shows the incident irradiation variations during the heating step
and the corresponding evolutions of surface temperature. We
notice the thermal responses are similar during the rst 150 s (i.e.

Fig. 6. Evolutions of product temperatures and infrared irradiation in the case of shortwave infrared emitters (a), evolutions of product temperatures and infrared irradiation
in the case of medium-wave infrared emitters (b), evolution of moisture content and
drying rate obtained with short-wave and medium-wave infrared emitters (c).

the infrared irradiation set point equals to 10.2 kW m
2). Mainly

due to the initial varnish mass, which is not exactly the same
(Table 2), the mean moisture content evolutions differed during
this rst stage (Fig. 7b). Finally, the samples have a nal moisture
content varying from 0.4 kg kg
1 to 0.75 kg kg
1.
During the varnish crosslinking, the mechanical properties, like
elastic modulus or hardness, are supposed to be increasing. In order
Table 2
Results of drying experiments with a temperature set point.
Test Tsp Initial varnish
( C) mass (mg)

Final moisture
content (kg kg
1)

Mean modulus from

unload (MPa)

RMS
(MPa)

1
2
3

0.75
0.6
0.4

39.4
32.7
24.6

3.5
2.8
2.6

180 451
150 425
130 435

108

N. Allanic et al. / International Journal of Thermal Sciences 79 (2014) 103e110

Table 3
Simulated results of diffusion coefcients estimated parameters.
D1 (m2 s
1)

D2 (m2 s
1)

Ea (kJ mol
1) b

Research intervals 10
17e10
13 5  10
7e5  10
5 20e40

Estimated values
10
14
10
5
25.065

0.05e6
4.65

From these measurements, we could conclude that the crosslinking is incomplete for the tests performed at 130  C and 150  C.
Indeed, the corresponding moduli are inferior to those obtained
with the full crosslinking (180  C).
3.4. Comparison of simulated and experimental results
An equivalent Cp taking into account crystallization peak is
deduced of calorimetric analysis (Fig. 2) and introduced into the

Fig. 7. Drying results for three bound temperature (130  C, 150  C and 180  C) e
surface temperature and infrared irradiation (a), moisture content and drying rate (b).

to characterize its state of curing after the drying and reticulation

tests, we measured the elastic modulus of thin varnish lm. The
load-penetration curves are presented in Fig. 8 and Table 3 presents
the different modulus values and their root mean squares. The
measured stiffnesses are included between 24.6 and 39.4 MPa.
A good agreement is found between our Youngs modulus
values and literature. For example, Ferencz et al. [25] evidenced an
EPDM Youngs modulus from 30 to 48 MPa by using AFM nanoindentation. In another study, Yusoh and Song [26] studied the
mechanical behavior of polyurethane elastomer/organoclay composites; they evidenced some stiffness values closed to 14 MPa for
the virgin elastomer. As expected, our values could present an
overestimate due to the varnish presence.

Fig. 8. Load-penetration curves resulting of nanoindentation tests.

Fig. 9. Confrontation of experimental and simulated results for three infrared irradiations (a) Esw1 10.2 kW m
2, (b) Esw2 15.4 kW m
2, (c) Esw3 20.3 kW m
2.

N. Allanic et al. / International Journal of Thermal Sciences 79 (2014) 103e110

model. For the simulations; the limit temperature Tl 180  C is

considered. In this study, the values of D1 and D2 are xed. At the
opposite, the parameters Ea and a are estimated by inverse
method. The minimization is performed on mass and surface
temperature evolutions, obtained for low infrared irradiation
[23]. Thus, an experiment with short-wave infrared irradiation
Esw1 10.2 kW m
2 is used for the estimation procedure. Tests
are performed for several values of D1 and D2 and boundary
values of parameters. The retained set of parameters is given in
Table 3. Fig. 9 compares experimental results (noted exp) and
simulated results (noted sim) for three short-wave infrared irradiations Esw1 10.2 kW m
2, Esw2 15.4 kW m
2 and
Esw3 20.3 kW m
2. For each case, the gap between simulated
and experimental moisture content remains lower than
0.25 kg kg
1. This value corresponds to the uncertainty on mass
measurement. In parallel, we can note that the mean squares
between experimental and simulated surface temperature of
varnish are respectively 2.6  C for Esw1, 8  C for Esw2 and 14  C for
Esw3. These gaps are mainly due to an overestimation of simulated temperature during the second stage of drying, when the
drying rate is highest and activity close to one. These results
show the necessity to improve the determination of water activity. Moreover, it can be interesting to introduce reticulation
energy before estimating mass diffusion coefcient. Despite of
this fact, the nal substrate temperatures are well described in
the three cases: this denotes a good description of thermal and
radiative properties of materials.
4. Conclusion
In this work, the behavior of water based varnish exposed to
infrared radiation was analyzed with an adapted instrumentation
of temperature sensors inside the substrate and the measurement
of weight loss in continuous. The experimental investigation
showed the inuence of infrared irradiation on drying rate and
heating rate. The analysis performed on varnish showed the necessity to control the infrared irradiation and to perform experiments with low infrared irradiation with well-known mechanisms
to improve numerical parameters determination. The study has
also enabled to determine the methodology to link the state of
reticulation with the varnish mechanical properties. In the
following steps, the experimental and numerical tools could be
used to dene quality criteria and determine optimal temperature
and moisture content evolutions to obtain a xed nal reticulated
stage.
References
[1] R. Dhib, Infrared drying: from process modelling to advanced process control,
Dry. Technol. 25 (2007) 97e105.
[2] Z. Ul-Islam, R. Dhib, Y. Dahman, Modeling of infrared drying of polymer solutions, Chem. Prod. Process Model. 5 (2010) 1. Article 21.
[3] P. Dufour, Control engineering in drying technology: review and trends, Dry.
Technol. 24 (2006) 889e904.
[4] N. Allanic, P. Salagnac, P. Glouannec, Convective and radiant drying of a
polymer aqueous solution, Heat Mass Transfer 43 (10) (2007) 1087e1095.
[5] A. Overbeek, Polymer heterogeneity in waterborne coatings, J. Coat. Technol.
Res. 7 (1) (2010) 1e21.
[6] I. Ludwig, W. Schabel, P. Ferlin, J.C. Castaing, M. Kind, Drying, lm formation
and open time of aqueous polymer dispersions, Eur. Phys. J. Spec. Top. 166
(2009) 39e43.
[7] J.Z. Lai, Y.C. Chang, J.T. Yeh, K.N. Chen, Single component self-curable aqueousbased PU system with new aziridinyl curing agent, J. Appl. Polym. Sci. 91
(2004) 1997e2007.
[8] J.Z. Lai, P.J. Chen, J.T. Yeh, K.N. Chen, A cross self-curing system for an aqueousbased PU hybrid, J. Appl. Polym. Sci. 97 (2005) 550e558.
[9] P. Le Bideau, J.P. Ploteau, P. Dutournie, P. Glouannec, Experimental and
modelling study of supercial elastomer vulcanization by short wave infrared
radiation, Int. J. Therm. Sci. 48 (2009) 573e582.

109

[10] J.P. Ploteau, P. Glouannec, H. Nol, Thermoelectric ux-meters for infrared

radiation measurements, Appl. Therm. Eng. 27 (2e3) (2007) 674e681.
[11] N. Allanic, P. Salagnac, P. Glouannec, J.P. Ploteau, Experimental study of the
drying and curing of water based varnishes by infrared radiation, in:
A.S. Mujumdar (Ed.), 16th International Drying Symposium, Hyderabad, India,
9e12 November 2008, 2008, pp. 1219e1224.
[12] S. Khallou, C. Almeida-Rivera, P. Bongers, A fundamental approach and its
experimental validation to simulate density as a function of moisture content
during drying processes, J. Food Eng. (2010) 177e187.
[13] J. Frias, J.C. Olivieira, K. Schittkowski, Modelling and parameter identication
of a maltodextrin DE 12 drying process in a convection oven, Appl. Math.
Model. 25 (2001) 449e462.
[14] S. Basu, U.S. Shivhare, A.S. Mujumdar, Models for sorption isotherms for foods:
a review, Dry. Technol. 24 (2006) 917e930.
[15] A. Belghit, A. Bennis, Experimental analysis of the drying kinetics of cork,
Energy Convers. Manage. 50 (2008) 618e625.
[16] X. Li, B. Bhushan, A review of nanoindentation continuous stiffness
measurement technique and its applications, Mater. Charact. 48 (1)
(2002) 11e36.
[17] I.B. Topcu, T. Bilir, Experimental investigation of drying shrinkage cracking of
composite mortars incorporating crushed tile ne aggregate, Mater. Des. 31
(9) (2010) 4088e4097.
[18] W.C. Oliver, G.M. Pharr, An improved technique for determining hardness and
elastic modulus using load and displacement sensing indentation experiments, J. Mater. Res. 7 (1992) 1564e1583.
[19] A. Erriguible, P. Bernada, F. Couture, M.A. Roques, Simulation of vaccum drying
by coupling models, Chem. Eng. Process. (2007) 1274e1285.
[20] A.S. Mujumdar, Handbook of Industrial Drying, third ed., Taylor & Francis,
Philadelphia, 2007.
[21] F.P. Incropera, D.P. De Witt, Fundamentals of Heat and Mass Transfer, fourth
ed., John Wiley & Sons Inc., 1996.
[22] R.B. Bird, W.E. Stewart, E.N. Lightfoot, Transport Phenomena, second ed., John
Wiley & Sons, Inc., 2002.
[23] N. Allanic, P. Salagnac, P. Glouannec, B. Guerrier, Estimation of an effective
water diffusion coefcient during infrared-convective drying of a polymer
solution, AIChE J. 55 (9) (2009) 2345e2355.
[24] P. Navarri, J. Andrieu, High-intensity infrared drying study. II. Case of thin
coated lms, Chem. Eng. Process. 32 (1993) 319e325.
[25] R. Ferencz, J. Sanchez, B. Blmich, W. Herrmann, AFM nanoindentation to
determine Youngs modulus for different EPDM elastomers, Polym. Test. 31
(3) (2012) 425e432.
[26] K. Yusoh, J. Jin, M. Song, Subsurface mechanical properties of polyurethane/
organoclay nanocomposite thin lms studied by nanoindentation, Prog. Org.
Coatings 67 (2) (2010) 220e224.

Nomenclature

Symbols
aw: water activity
b: water diffusion coefcient parameters
C: water activity parameter
Cp: specic heat capacity, J kg
1 K
1
D: water diffusion coefcient, m2 s
1
D1, D2: water diffusion coefcient parameters, m2 s
1
hc: heat convective exchange coefcient, W m
2 K
1
e: thickness, m
E: infrared radiation, W m
2
Ea: activation energy, J mol
1
Fm: evaporation rate, kg m2 s
1
K: water activity parameter
km: mass transfer coefcient, m s
1
Lv: latent heat of vaporization, J kg
1
Mv: molecular weight, kg mol
1
Pt: atmospheric pressure, Pa
Pv: water vapor pressure, Pa
R: gas constant, J mol
1 K
1
T: temperature, K
T1, T2, T3: thermocouple measures, K
t: time, s
W: moisture content in dried basis, kg
1 kg
1
Wm: water activity parameter
Greek letters

a: infrared absorptivity
: product emissivity
x: spatial coordinate, m
: heat ow, W m
2
l: thermal conductivity, W m
1 K
1
r: density, kg m
3

110

j: linear shrinkage coefcient

s: StefaneBoltzmann constant, W m
2 K
4
Subscripts
a: ambient
d: dried product
IR: infrared
l: limit

N. Allanic et al. / International Journal of Thermal Sciences 79 (2014) 103e110

s: substrate
sat: saturated
sp: set point
surf: surface
sw: short waves
mw: medium waves
0: initial value
*: dimensionless
v: varnish