Water Evaporation from Porous Media by Dynamic Vapor Sorption

Page 1

Seediscussions,stats,andauthorprofilesforthispublicationat:http://www.researchgate.net/publication/270875748

WaterEvaporationfromPorousMediabyDynamicVaporSorption

ARTICLEinCOLLOIDSANDSURFACESAPHYSICOCHEMICALANDENGINEERINGASPECTS·SEPTEMBER2015

ImpactFactor:2.75·DOI:10.1016/j.colsurfa.2014.11.011

READS

97

5AUTHORS,INCLUDING:

DarioDonnarumma

UniversitédeMontpellier

8PUBLICATIONS0CITATIONS

SEEPROFILE

GiovannaTomaiuolo

UniversityofNaplesFedericoII

32PUBLICATIONS185CITATIONS

SEEPROFILE

SergioCaserta

UniversityofNaplesFedericoII

39PUBLICATIONS326CITATIONS

SEEPROFILE

YonasGizaw

Procter&Gamble

15PUBLICATIONS48CITATIONS

SEEPROFILE

Availablefrom:DarioDonnarumma

Retrievedon:13October2015

Page 2

C

W

DYa

Nb

Ic

h

•

•

•

•

a

ARRAA

KSDAMWT

1

bo

h0

ARTICLE IN PRESSG ModelOLSUA-19527; No. of Pages 6

Colloids and Surfaces A: Physicochem. Eng. Aspects xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Colloids and Surfaces A: Physicochemical andEngineering Aspects

journa l h om epage: www.elsev ier .com/ locate /co lsur fa

ater evaporation from porous media by Dynamic Vapor Sorption

ario Donnarummab,∗, Giovanna Tomaiuoloa, Sergio Casertaa,b,onas Gizawc, Stefano Guidoa,b

Dipartimento di Ingegneria Chimica dei Materiali e della Produzione Industriale, Università degli studi di Napoli Federico II, P.zzle Tecchio 80, 80125apoli, ItalyConsorzio Interuniversitario Nazionale per la Scienza e Tecnologia dei Materiali (INSTM), UdR INSTM Napoli Federico II, P.le Tecchio, 80, 80125 Naples,

talyThe Procter & Gamble Co., Winton Hill Business Center, 6210 Center Hill Avenue, Cincinnati, OH 45224, United States

i g h l i g h t s

A cotton fabric was used as a model tostudy water evaporation from porousmedia.The experimental technique is basedon the Dynamic Vapor Sorption.The effects of salts and surfactants onevaporation were investigated.Salts and surfactants influence onlyone of the evaporation phases (CDP).

g r a p h i c a l a b s t r a c t

r t i c l e i n f o

rticle history:eceived 31 July 2014eceived in revised form 6 November 2014ccepted 7 November 2014vailable online xxx

eywords:

a b s t r a c t

Evaporation from porous media is a subject of growing interest in view of its industrial relevance andof the different spatial scales involved in the problem. A common, yet not fully elucidated feature ofevaporation from porous media is the presence of a constant drying rate phase (CDP), which is followedby a falling drying rate phase (FDP). Here, we use a cotton fabric as a model porous medium and study theeffects of salts and surfactants on CDP and FDP of aqueous solutions by using the Dynamic Vapor Sorption(DVS) technique. The cotton samples were first characterized by adsorption/desorption isotherms. Then

urfactantryingdsorptionoistureater hardness

ransport phenomena

the drying process of fabrics soaked in distilled water and in aqueous solutions at different hardness wasinvestigated. The same procedure was used to study the drying of a dilute surfactant aqueous solution.All these experiments show that the presence of salts and surfactants in water affects the CDP, but not theFDP. At variance with the bulk behavior, an increase of the CDP drying rate was found for the surfactantsolution and for aqueous salt solutions up to some hardness value.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

Please cite this article in press as: D. Donnarumma, et al., Water evapSurf. A: Physicochem. Eng. Aspects (2014), http://dx.doi.org/10.1016/j

Drying of porous media is a fundamental operation in a num-er of applications, including agriculture, soil technology, recoveryf volatile hydrocarbons from oil reservoirs, cosmetics, building

∗ Corresponding author. Tel.: +33 467143860.E-mail address: [email protected] (D. Donnarumma).

ttp://dx.doi.org/10.1016/j.colsurfa.2014.11.011927-7757/© 2014 Elsevier B.V. All rights reserved.

restoration and material processing, such as the production of food,wood, paper, textiles, pharmaceuticals and washing powders [1].Several mechanisms have been hypothesized to describe drying atthe pore scale: (i) vapor diffusion, which is modeled by Fick’s lawequation with a diffusivity dependent on pore tortuosity [2]; (ii)

oration from porous media by Dynamic Vapor Sorption, Colloids.colsurfa.2014.11.011

capillary flow, due to pressure differences depending on meniscicurvature (Laplace equation) [3], and (iii) gas transport, elicitedby pressure gradients according to Darcy’s law [4,5]. Another pos-sible mechanism which has recently attracted much attention is

Page 3

IN PRESSG ModelC

2 s A: Physicochem. Eng. Aspects xxx (2014) xxx–xxx

llcmgmmhb

osmricdPteIfstimmtafpr

pcettaiffs[iscoii

2

2

2

iv[o

ww

Table 1Salts of calcium and magnesium content in pure water.

Sample name Hardness[ppm] Salts content[mg/250 ml]

Moderately hard water (MW) 255 702.99 CaCl2·2H2O324.01 MgCl2·6H2O

Hard water (HW) 420 1157.78 CaCl2·2H2O533.66 MgCl ·6H O

ARTICLEOLSUA-19527; No. of Pages 6

D. Donnarumma et al. / Colloids and Surface

iquid flow in the films developing along the corners of rectangu-ar cross-section pores, which acts to increase the drying rate asompared to circular cross-section capillaries [6–8]. Some recentodels based on a phenomenological pore-network approach sug-

est that the diffusion based Capillary and Peclet numbers are theain dimensionless parameters of the drying process in porousedia [9]. While a complete analysis would require to include mass,

eat and momentum transfer mechanisms, simplified models haveeen proposed for specific industrial drying application [10].

Going from the pore scale to the macroscopic behavior, the ratef water transport in fabrics depends on environmental conditions,uch as air temperature, air velocity and pressure gradient and onaterial properties, such as density, thickness, heat conduction

esistance and knitting patterns. Three periods can be commonlydentified in the drying process in porous media [11,12], and espe-ially in cotton textiles [13,14]: preheating period (PP), constantrying period (CDP) and falling drying period (FDP). During theP, thermal energy is transferred from air to fabric heating uphe fabric surface. In the CDP, the fabric absorbs more thermalnergy, thus the moisture content on the fabric surface evaporates.n these conditions, a simple conductive and convective heat trans-er process between hot air stream and the moisture on the fabricurface is present, eliciting a water reduction directly proportionalo time. Water moves from the core to the surface of fabric by cap-llary action to maintain a balance of the change of the gradient of

oisture content. While the moisture content reaches the criticaloisture content (�k) the drying rate behavior becomes non-linear

ill all the moisture evaporates, determining the end of the CDPnd the begin of the FDP. The FDP is apparently controlled by dif-usion, which is usually slower than the convective heat transferrocess present in the CDP. Thus, in FDP, a decrease of the moistureeduction rate is observed.

In light of the complexity of the problem, it is difficult to obtainredictions of rate of drying. In particular, it is not clear which is theontrolling mechanism of the drying process. It is likely that differ-nt regimes can become dominant depending on the pore size andhermodynamic properties, as well as on transport parameters. Inhis work, we focus on the characterization of the drying process of

cotton fabric sample by using DVS technique. The main observables moisture content sorption as a function of relative humidity (RH),or cotton fabric soaked in water of different hardness and in sur-actant solutions. In particular, the influence of linear alkylbenzeneulphonic acid (HLAS), a surfactant of particular industrial interest15,16], on drying rate of cotton fabrics, is analyzed. Although thenfluence of surfactants on interfacial phenomena has been widelytudied [16,17], as well as the effect of surfactants on the drying pro-ess of droplets [18–21] and films [22], showing that in the presencef surfactants the evaporation rate increases, very few reports existn the literature about the impact of surfactants on evaporation raten porous media.

. Materials and methods

.1. Materials

.1.1. Cotton fabric samplesCotton is the most common textile material, frequently used

n clothes making; it is light, soft, and has a high water absorptionalue that contributes to the high wearing comfort of cotton clothes23,24]. In general, cellulosic materials show a complex mechanism

Please cite this article in press as: D. Donnarumma, et al., Water evapSurf. A: Physicochem. Eng. Aspects (2014), http://dx.doi.org/10.1016/

f moisture sorption [25,26].The fabrics textile used in the experiments is 100% white cotton

ith a simple thin structure, taken out from a cloth; its bone dryeight is ca. 10 mg and have a porosity value equal to 0.7; the tissue

2 2

Very hard water (VW) 540 1433.45 CaCl2·2H2O660.73 MgCl2·6H2O

is knitted in plain wave arrangement with a number of yarn equalto 170 dtex and have a specific weight of 70 g/m2.

2.1.2. Hard water solutionsWater solutions with different hardness have been tested. We

started with purified water (PW) for technical uses (Grade 3, conf.ISO 3696), which is commonly used in ironing to prevent the for-mation of calcium aggregates and in electrical batteries. Watersolutions at different hardness have been prepared by dissolvingsalts of calcium and magnesium hydrates in PW at different con-centrations, as summarized in Table 1.

The water solutions have been characterized in terms of pH andconductivity (Table 2), to understand the possible contribution ofelectrolytic parameters to the drying process.

Values reported in Table 2 were measured using apH/conductivity-meter (Oakton PC2700) equipped with pHelectrode and conductivity/temperature probe; the device wascalibrated before each measurement performing a two point pHcalibration with buffer solutions at 4.01 and 10.01 pH and a onepoint conductivity calibration with a standard solution of knownconductivity (1413 �S).

2.1.3. Surfactant solutionsProcter & Gamble (Italian Research Center, Pomezia, Italy) sup-

ply a commercial sample of the linear alkylbenzene sulphonic acid(HLAS), an anionic surfactant of industrial grade purity; this sam-ple was used without any further purification at a concentration of20,000 ppm in water (or 20 ml/L or 2 wt%). Between 25 and 40 ◦CHLAS solutions have a Critical Micellar Concentration in water ofabout 0.05 wt%. At these temperatures, only the isotropic micellarL1 phase is present, while the lamellar L� phase is present only athighest concentration (>28 wt%) [27,28].

2.2. Methods

2.2.1. Dynamic Vapor Sorption analysisDVS is a well-established technique for the study of the interac-

tion of water molecules with porous media [29,30] and of moisturesorption in general [31,32]. This technique has high reproducibilityand provides accurate isotherms over a wide RH range [33]. DVS hasbeen successfully used in the study of water sorption in cellulosicmaterials [34,35].

In a DVS experiment, a sample is exposed to a series of stepchanges in RH, monitoring mass changes as a function of time. Sam-ple mass must reach gravimetric equilibrium at each step beforeproceeding to the next humidity level. Then, the equilibrium massvalues at each RH step are used to generate the isotherm. Isothermsare typically divided into two components: sorption for increasinghumidity steps and desorption for decreasing humidity steps. Sorp-tion can be further divided into adsorption (sorbate located on thesurface) and absorption (sorbate penetrates the bulk) [36].

oration from porous media by Dynamic Vapor Sorption, Colloidsj.colsurfa.2014.11.011

Sorption isotherms were recorded using a Q5000SA DVS ana-lyzer (TA Instruments) equipped with an autosampler and amicrobalance with a maximum weight capacity of 100 mg andcapable to detect mass differences <0.1 �g. Before each test,

Page 4

ARTICLE IN PRESSG ModelCOLSUA-19527; No. of Pages 6

D. Donnarumma et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects xxx (2014) xxx–xxx 3

Table 2Chemical properties of saline solutions.

Sample name pH±0.002 Temperature±0.5 [◦C] Conductivity±1% [�S/m] Potential difference±0.05% [mV]

PW 5.52 27 5300 94.25366 74.56860 775541 74.8

du

s1ninttc

t2arTtiwaaDsa

uCtf

R

s

2

rtti1tmiu

2

(itpwm

Table 3BET parameters and regression coefficients for moisture adsorption of cotton fabricsat different temperatures.

Temperature BET parameters

[◦C] mo CB R2

25 0.01837 7.152 0.9998

MW 5.86 27

HW 5.81 27

VW 5.83 27

eliquescent salts (sodium bromide and potassium chloride) weresed to verify the humidity value.

The cotton fabric samples (ca. 4 mg) were placed onto a cleanedample pan (a semi-spherical metal-coated quartz crucible, volume80 �l) which was carefully hooked onto a hang down wire con-ected to a microbalance in a climate controlled chamber, located

n a thermostatically controlled cabinet. A constant flow of dryitrogen gas (100 ml min−1) is mixed into a stream of nitrogen con-aining water vapor at saturation; the mixed stream passes throughhe measurement chamber to maintain a set RH point, automati-ally controlled with the Q5000SA equipment.

The instrument maintained the sample at a constant RH untilhe rate of mass change (dm/dt) was less than 0.005% min−1 over a0 min period [37]. Every 5 s the change in weight was recordednd saved. The sample mass readings from the microbalanceeveal the vapor adsorption/desorption behavior of the material.emperature and humidity values were very stable (RH ± 0.1%,emperature ± 0.1 ◦C) during the whole measure. The running time,sotherm temperature, target RH, actual RH and sample weight

ere recorded throughout the isotherm run, thanks to humiditynd temperature probes located in close proximity to the samplend providing direct measurements. At the end of each test, theVS instrument automatically tares the sample mass, so that all

ubsequent moisture content values can be accurately calculateds a mass percentage.

From the data acquired, the drying curves in terms of Resid-al Moisture Content (RMC) over time were calculated, as done byarter et al. [23,24]. The RMC is expressed as the ratio betweenhe weight of the solution (Ww) and the dry weight (Wd) of cottonabrics, expressed in percentage, as in Eq. (1):

MC = Ww − Wd

Wd× 100 (1)

Two kinds of tests were carried out using the DVS apparatus:orption test and drying test.

.2.2. Determination of sorption equilibrium isothermsThe schedule for the adsorption/desorption test was set to

ecord the sample mass at each of the following RH steps of adsorp-ion at 0, 10, 20, 30, 40, 50, 60, 70, 80 and 90% RH and at a constantemperature of 25 ± 0.1 ◦C, 40 ± 0.1 ◦C and 60 ± 0.1 ◦C (adsorptionsotherm) and in reverse sequence at 90, 80, 70, 60, 50, 40, 30, 20,0, 0% RH for the desorption isotherm. The RH schedule was ini-iated once the rate of change of the initial drying curve rate of

ass change reached the set value. The sample is initially weighedn the atmosphere in the laboratory and consequently contains annknown amount of moisture.

.2.3. Drying testsDuring all the drying test both temperature (40 ± 0.1 ◦C) and RH

0%) was set to be constant. The cotton fabric samples were placednside the pan and soaked with 0.025 ml (25 mg) of solution. Then

Please cite this article in press as: D. Donnarumma, et al., Water evapSurf. A: Physicochem. Eng. Aspects (2014), http://dx.doi.org/10.1016/j

he environmental controlled chamber was closed and the tem-erature rose to the constant value in less than 5 min. The test endshen the water inside the sample weight reaches the equilibriumoisture content value.

40 0.01830 6.733 0.998360 0.01602 6.687 0.9960

3. Results and discussion

3.1. Sorption isotherms

The sorption curves for cotton fabric samples at 25 ◦C, 40 ◦C and60 ◦C are shown in Fig. 1. For each temperature value a typicalhysteresis curve (Fig. 1a) is observed, in agreement with sorptioncurves of other mature natural fibers [33–35].

At each temperature the shape of the absorption curve is a typeII-isotherm that describes the adsorption on macro-porous andnon-porous adsorbents with strong adsorbate–adsorbent inter-actions [38–40], according to IUPAC classification [41]. Moistureadsorption decreases with temperature. In Fig. 1 a distinct hystere-sis between sorption and desorption cycle can be noticed at eachtemperature, indicating changes in the interaction between waterand the cotton fiber surface [26].

DVS-generated isotherms were then fitted by the device soft-ware (TA Universal Analysis) with two theoretical equations:the two-parameter Brunauer, Emmett and Teller (BET) equation[42–44] and the three parameter Guggenheim Anderson and DeBoer (GAB) equation [45–47].

The BET equation is reported as follows:

m = moCBaw

[(1 − aw)(1 − aw + CBaw)](2)

where m is the moisture content, aw is the corresponding wateractivity, mo is the moisture content value when the adsorbent sur-face area is covered with a complete monolayer, CB is the energyconstant.

In Table 3 BET parameters and regression coefficients corre-sponding to the data showed in Fig. 1 were reported and a good fitbetween the BET model and the experimental data can be evincedlooking at the value of the correlation coefficient R2.

Since the validity of the BET equation stay between 5% and 30%RH [48], GAB equation, that is valid between 5% and 80% RH [45],has been also used:

m = moCGKaw

[(1 − Kaw)(1 − Kaw + CGKaw)](3)

where CG is the Guggenheim constant and K is the correction factorwhich is typically smaller then 1.

As it can be observed, the fit with the GAB model is almost com-plete, based on the value of the correlation coefficient R2 betweenexperimental and theoretical values (Table 4).

oration from porous media by Dynamic Vapor Sorption, Colloids.colsurfa.2014.11.011

As can be noticed in Fig. 1 and by looking at the mo valuesin Tables 3 and 4, both the moisture content and the monolayervalues of absorbed moisture on the surface of the cotton fab-rics decrease while temperature increases. This effect has been

Page 5

ARTICLE IN PRESSG ModelCOLSUA-19527; No. of Pages 6

4 D. Donnarumma et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects xxx (2014) xxx–xxx

Fig. 1. Adsorption/desorption test: graphs show the equilibrium moisture content (%) as

adsorption isotherms were fitted with BET (gray line) and GAB (bold black line) model.

Time[min]

0 10 20 30 40 50

Moi

stur

e C

onte

nt [%

]

0

50

100

150

200

250

Time[min]17 19 21 23

Moi

stur

e C

onte

nt [%

]

80

90

100

110

PW (DR : 6.6 7)MW (DR: 6. 88)HW (DR:7.19)VW (DR: 6.71)

F(o

piT

3

awraped

TGa

ig. 2. Drying test: graphs show the moisture content (%) as a function of timemin) at 40 ◦C for cotton fabrics soaked in hard water solutions. In the inset, slopesf different curves are highlighted.

reviously observed in cotton jersey at 25 ◦C and 35 ◦C [49] andn carrot chips at higher temperatures [50]. The CG and K values inable 4 are confirmed by the study of Bhouri et al. [49].

.2. Water solutions at different hardness

The cotton fabric sample was soaked with ca. 25 mg of solutionnd the amount of moisture was determined by measuring the neteight (i.e. the weight during the drying process minus the dry fab-

ics weight) as a function of time by a thermo balance device (DVS),

Please cite this article in press as: D. Donnarumma, et al., Water evapSurf. A: Physicochem. Eng. Aspects (2014), http://dx.doi.org/10.1016/

s described in Section 2.2.3. Solution of water at different hardness,repared as described in Section 2.1.2, were tested two times forach sample, and the drying curves show a complete overlap in theuplicate experiments (Fig. 2).

able 4AB parameters and regression coefficients for moisture adsorption of cotton fabricst different temperatures.

Temperature GAB parameters

[◦C] mo CG K R2

25 0.03117 6.422 0.5216 0.999640 0.02778 7.423 0.5319 1.000060 0.02537 6.592 0.5457 0.9999

a function of the RH (%) at 25 ◦C (a), 40 ◦C (b) and 60 ◦C (c) for cotton fabrics. All the

All curves show the typical trend that characterizes the dryingof porous media, the constant drying period (from 0 to ∼35 min)and falling drying period (for time >35 min) being easy to iden-tify. Results show that all hard water solutions dry faster than purewater. This is a counterintuitive result if we consider the boiling-point elevation that occurs when a salt is dissolved in a liquid.However, the drying process should be considered more as a trans-port phenomenon than a thermodynamic one, due to the transportat the gas/liquid interface. In Table 5 the drying rate for each solu-tion in the constant rate period is reported, showing the effect of saltconcentration on water evaporation. The percent deviation of eachdrying rate value from the pure water (PW) value is also reported.

The highest drying rate value is related to the HW sample. Anonlinear dependence between water hardness and drying rateincrease can be identified. About the sensitive decreasing betweenHW and VW, this dependence cannot be related to the salts solu-bility concentration because the water hardness for both HW andVW is 3–4 orders of magnitude below the solubility concentrationboth for calcium and magnesium chloride. It should be pointedout that a water hardness value of 300 ppm is a considerably highconcentration in many applications.

3.3. Surfactant and pure water

It has been recently reported that water hardness seems to havea role also in modifying the surface tension of dilute surfactantsolutions [51,52]. In the following, experiments on HLAS watersolutions (concentration: 20,000 ppm) are described. The dryingrate value of pure water (PW) was taken as the reference valuein order to evaluate the drying rate increase associated with thepresence of surfactants, as shown in Fig. 4.

In the constant rate period the drying rate value of the surfactant

oration from porous media by Dynamic Vapor Sorption, Colloidsj.colsurfa.2014.11.011

solution is up to 10% higher than the value for pure water. If welook at the drying rate as a function of the moisture content, asshown in the inset of Fig. 3 (where the absolute value of the drying

Table 5Hard water solutions effect on the drying rate in a cotton fabric.

Sample name Hardness Drying rate Deviation[ppm] [%/min] [%]

PW 0 6.67 0MW 255 6.88 +3HW 420 7.19 +8VW 530 6.71 0

Page 6

ARTICLE IN PRESSG ModelCOLSUA-19527; No. of Pages 6

D. Donnarumma et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects xxx (2014) xxx–xxx 5

Time [m in]

0 10 20 30 40 50

Moi

stur

e C

onte

nt [%

]

0

50

100

150

200

250

Moisture Content [ %]

0 40 80 12 0 16 0 20 0 24 0D

ryin

g R

ate

[%/m

in]

0

2

4

6

8

PWPS

Fig. 3. Drying curves of pure water solution (black dots) and HLAS solution (whitedots). In the inset the drying rate change in function of the moisture content isplotted.

Table 6Surfactant and hard water solution effect on the drying rate of a cotton fabric.

Sample Composition Drying rate DeviationHLAS [ppm] [%/min] [%]

PW 0 6.67 0

rd

f

3

tdipttim(cti

e

bopac

wocbd

Moi sture Con ten t [%]

0 50 100 15 0 200 250

Dry

ing

Rat

e [%

/min

]

0

2

4

6

8

40

PWPSMWHWVW

Dry

ing

Rat

e [%

/min

]

6,6

6,9

7,2

7,5

Falli ngRat e

Consta nt Rate

Fig. 4. In the graph the drying rate change in function of the moisture content isshown. An evaluation of the critical moisture content (40%) is represented by thevertical axis. In the inset, error bars for each curve were reported.

Time[mi n]

0 5 10 15 20 25

Moi

stur

e C

onte

nt [%

]

0

10

20

30

40

Pure Water Pure Water+HLAS Med ium Hard Water Hard Water Very Hard Water

Time [mi n]0 5 10 15 20 25

Moi

stur

e C

onte

nt [l

og(%

)]

0,01

0,1

1

10

PWPSMWHWVW

Fig. 5. In the graph the drying rate change in function of the moisture content is

PS 20,000 7.13 +7

ate is shown), both the constant rate and the falling rate are easilyistinguishable.

In Table 6 the (absolute) drying rates in the constant rate periodor both solutions are reported.

.4. Critical moisture content determination

The drying process in cotton fabrics, as discussed before, showswo main regimes identified by the change in drying rate of therying curve. The two periods of a fabric drying cycle have been

dentified as CDP and FDP in terms of the drying rate. After thereheating, there is a CDP followed by the FDP. The �k is the mois-ure content at the beginning of the FDP, i.e. the value separatinghe two periods. The values of �k were identified by plotting thenstantaneous drying rate, given by the ratio between the change in

oisture content and the change in time of two consecutive pointssome data averaging was done to reduce noise), versus moistureontent. �k can be approximately identified as the moisture con-ent at the start of the plateau region of the drying rate as shownn Fig. 4.

Error bars in Fig. 4 represents the standard deviation of 3 differ-nt experiments with MW, which is about 0.2.

The so determined value of �k for fabric samples is evaluated toe around 40% for all the water solutions. A possible interpretationf this result is that �k depends on the microstructure of the fabricsore network. A further support to this hypothesis is that by shiftingll drying curves so that they have 40% as the initial point, all dataollapse on the same curve in the FDP, as reported in Fig. 5.

In the inset in Fig. 5, the same data were reported in a semi-logay to show that a consistent deviation of the curves will happen

nly below 1% of moisture content, when the salt and surfactantoncentrations become relevant. Once more Fig. 5 shows that the

Please cite this article in press as: D. Donnarumma, et al., Water evapSurf. A: Physicochem. Eng. Aspects (2014), http://dx.doi.org/10.1016/j

ehavior in the FRP is the same for all the solutions because itepends on the fabrics structure rather than solution properties.

shown. An evaluation of the critical moisture content (40%) is represented by thevertical axis.

4. Conclusions

The present work shows that DVS is a useful tool to characterizeevaporation from cotton fabrics by nitrogen adsorption/desorptionisotherms and to investigate the drying process in such a porousmedium. To investigate the effect of water hardness on the dry-ing process, the drying rates of some water solutions at differenthardness in a cotton sample were compared with the drying rate ofpure water. Using the same procedure, the drying rate of a dilutedHLAS aqueous solution has been evaluated. The results of this workshows that the drying rate in a porous medium is affected both bywater hardness and surfactant in the concentration range investi-gated. In fact, an increment of 7% and 8% in drying rate was foundfor HLAS and HW solutions, respectively. This effect could be due toan increase of surface area due to the possible interactions betweenthe OH groups on the cotton fibers surface and the ions or the

oration from porous media by Dynamic Vapor Sorption, Colloids.colsurfa.2014.11.011

ionic groups of the surfactant molecules. This drying rate incre-ment is only found in the constant drying rate region, whereas inthe falling drying phase all the drying curves collapse on the same

Page 7

ING ModelC

6 s A: Ph

mt(

A

PaCu

R

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

ARTICLEOLSUA-19527; No. of Pages 6

D. Donnarumma et al. / Colloids and Surface

aster-curve, showing no effect of the solution properties. The lat-er effect could be explained by the dominant role played by fabricsporous) microstructure in the falling rate region.

cknowledgments

The authors thank Sergio Barbarino and Vincenzo Guida fromrocter & Gamble for supporting this work. Part of the experimentalctivity was performed at the Thermal-Lab, Pomezia R&D Researchenter of Procter & Gamble. This work has been done under thembrella of COST Actions MP1106 and CM1101.

eferences

[1] N. Arsalan, S.S. Palayangoda, D.J. Burnett, J.J. Buiting, Q.P. Nguyen, Surfaceenergy characterization of carbonate rocks, Colloid Surf. A: Physicochem. Eng.Aspects 436 (2013) 139–147.

[2] S. Webb, K. Pruess, The use of Fick’s law for modeling trace gas diffusion inporous media, Transp. Porous Media 51 (2003) 327–341.

[3] S. Amico, C. Lekakou, Mathematical modelling of capillary micro-flow throughwoven fabrics, Compos. Part A: Appl. Sci. Manuf. 31 (2000) 1331–1344.

[4] H. Darcy, Les fontaines publiques de la ville de Dijon: exposition et application,Victor Dalmont, 1856.

[5] M. Fireman, Permeability measurements on disturbed soil samples, Soil Sci. 58(1944) 337–354.

[6] M. Prat, On the influence of pore shape, contact angle and film flows on dryingof capillary porous media, Int. J. Heat Mass Transfer 50 (2007) 1455–1468.

[7] F. Chauvet, P. Duru, S. Geoffroy, M. Prat, Three periods of drying of a singlesquare capillary tube, Phys. Rev. Lett. 103 (2009).

[8] D.B. Jazia, L. Vonna, S. Knopf, G. Schrodj, H. Nouali, B. Lebeau, Y. Holl, H. Haidara,Absorption of water/ethanol microdroplets into model porous networks, Col-loid Surf. A: Physicochem. Eng. Aspects 436 (2013) 363–370.

[9] A.G. Yiotis, A.K. Stubos, A.G. Boudouvis, I.N. Tsimpanogiannis, Y.C. Yortsos, Pore-network modeling of isothermal drying in porous media, Transp. Porous Media58 (2005) 63–86.

10] H.S. Lee, W.W. Carr, H.W. Beckham, J. Leisen, A model of through-air drying oftufted textile materials, Int. J. Heat Mass Transfer 45 (2002) 357–366.

11] S.J. Kowalski, Thermomechanics of Drying Processes, Springer, 2003.12] S. Kowalski, G. Musielak, J. Banaszak, Experimental validation of the heat and

mass transfer model for convective drying, Drying Technol. 25 (2007) 107–121.13] R.W.L. Ip, E.I.C. Wan, The New Use of Diffusion Theories for the Design of Heat

Setting Process in Fabric Drying, 2012.14] R.W. Ip, E.I.C. Wan, Development of nonlinear analytical models for the study

of heat/mass transfer properties in the drying of porous type fabric, in: Defectand Diffusion Forum, Trans Tech Publ, 2012, pp. 593–598.

15] V. Guida, Thermodynamics and kinetics of vesicles formation processes, Adv.Colloid Interface Sci. 161 (2010) 77–88.

16] A. Pommella, S. Caserta, V. Guida, S. Guido, Shear-induced deformation of sur-factant multilamellar vesicles, Phys. Rev. Lett. 108 (2012) 138301.

17] A. Pommella, S. Caserta, S. Guido, Dynamic flow behaviour of surfactant vesiclesunder shear flow: role of a multilamellar microstructure, Soft Matter 9 (2013)7545–7552.

18] T. Still, P.J. Yunker, A.G. Yodh, Surfactant-induced Marangoni eddiesalter the coffee-rings of evaporating colloidal drops, Langmuir 28 (2012)4984–4988.

19] P.J. Yunker, T. Still, M.A. Lohr, A. Yodh, Suppression of the coffee-ring effect by shape-dependent capillary interactions, Nature 476 (2011)308–311.

20] I. Aranberri, B. Binks, J. Clint, P. Fletcher, Evaporation rates of water from con-centrated oil-in-water emulsions, Langmuir 20 (2004) 2069–2074.

21] M.D. Doganci, B.U. Sesli, H.Y. Erbil, Diffusion-controlled evaporation of sodiumdodecyl sulfate solution drops placed on a hydrophobic substrate, J. ColloidInterface Sci. 362 (2011) 524–531.

22] P. Alexandridis, S.Z. Munshi, Z. Gu, Evaporation of water from structured sur-factant solutions, Ind. Eng. Chem. Res. 50 (2010) 580–589.

Please cite this article in press as: D. Donnarumma, et al., Water evapSurf. A: Physicochem. Eng. Aspects (2014), http://dx.doi.org/10.1016/

23] D.L. Carter, D.O. Shah, The role of surface tension on the residual water contentof fabrics, J. Surfact. Deterg. 8 (2005) 91–94.

24] D.L. Carter, M.C. Draper, R.N. Peterson, D.O. Shah, Importance of dynamic sur-face tension to the residual water content of fabrics, Langmuir 21 (2005)10106–10111.

[

PRESSysicochem. Eng. Aspects xxx (2014) xxx–xxx

25] S. Okubayashi, U.J. Griesser, T. Bechtold, A kinetic study of moisture sorptionand desorption on lyocell fibers, Carbohydr. Polym. 58 (2004) 293–299.

26] S. Okubayashi, U. Griesser, T. Bechtold, Water accessibilities of man-made cel-lulosic fibers – effects of fiber characteristics, Cellulose 12 (2005) 403–410.

27] C. Richards, G.J. Tiddy, S. Casey, Lateral phase separation gives multiple lamellarphases in a binary surfactant/water system: the phase behavior of sodium alkylbenzene sulfonate/water mixtures, Langmuir 23 (2007) 467–474.

28] J.A. Stewart, A. Saiani, A. Bayly, G.J.T. Tiddy, The phase behaviour of lyotropicliquid crystals in linear alkylbenzene sulphonate (LAS) systems, Colloid Surf. A:Physicochem. Eng. Aspects 338 (2009) 155–161.

29] V. Guida, L. Scelsi, F. Zonfrilli, Humidity mass transfer analysis in packed powderdetergents, in: Proceedings of the 2012 COMSOL Conference, Milan, 2012.

30] M. Brundu, V. Guida, Modeling the chemical decomposition of sodium carbon-ate peroxyhydrate, in: Submitted to the COMSOL Conference, 2012.

31] N. Markova, E. Sparr, L. Wadso, On application of an isothermal sorptionmicrocalorimeter, Thermochim. Acta 374 (2001) 93–104.

32] P.G. Maher, Y.H. Roos, M.A. Fenelon, Physicochemical properties of spray driednanoemulsions with varying final water and sugar contents, J. Food Eng. 126(2014) 113–119.

33] C.A. Hill, A. Norton, G. Newman, The water vapor sorption behavior of naturalfibers, J. Appl. Polym. Sci. 112 (2009) 1524–1537.

34] Y. Xie, C.A. Hill, Z. Jalaludin, S.F. Curling, R.D. Anandjiwala, A.J. Norton, G. New-man, The dynamic water vapour sorption behaviour of natural fibres and kineticanalysis using the parallel exponential kinetics model, J. Mater. Sci. 46 (2011)479–489.

35] Ö. Ceylan, L. Van Landuyt, F. Meulewaeter, K. De Clerck, Moisture sorption indeveloping cotton fibers, Cellulose 19 (2012) 1517–1526.

36] G. Morozov, V. Breus, S. Nekludov, I. Breus, Sorption of volatile organic com-pounds and their mixtures on montmorillonite at different humidity, ColloidSurf. A: Physicochem. Eng. Aspects 454 (2014) 159–171.

37] C.A.S. Hill, Y.-J. Xie, The water vapour sorption kinetics of Sitka spruce at differ-ent temperatures analysed using the parallel exponential kinetics model, in:The Future of Quality Control for Wood & Wood Products, The Final Conferenceof COST Action E53, Edinburgh, 2010.

38] C. Sangwichien, G. Aranovich, M. Donohue, Density functional theory pre-dictions of adsorption isotherms with hysteresis loops, Colloid Surf. A:Physicochem. Eng. Aspects 206 (2002) 313–320.

39] O. Carmody, R. Frost, Y. Xi, S. Kokot, Surface characterisation of selected sorbentmaterials for common hydrocarbon fuels, Surf. Sci. 601 (2007) 2066–2076.

40] S.E. Smith, The sorption of water vapor by high polymers, J. Am. Chem. Soc. 69(1947) 646–651.

41] J. Rouquerol, D. Avnir, C. Fairbridge, D. Everett, J. Haynes, N. Pernicone, J. Ram-say, K. Sing, K. Unger, Recommendations for the characterization of poroussolids (Technical Report), Pure Appl. Chem. 66 (1994) 1739–1758.

42] S. Brunauer, P.H. Emmett, E. Teller, Adsorption of gases in multimolecular lay-ers, J. Am. Chem. Soc. 60 (1938) 309–319.

43] M. Caurie, The unimolecular character of the classical Brunauer, Emmett andTeller adsorption equation and moisture adsorption, Int. J. Food Sci. Technol.40 (2005) 283–293.

44] M. Pudipeddi, T.D. Sokoloski, S.P. Duddu, J.T. Carstensen, Quantitative charac-terization of adsorption isotherms using isothermal microcalorimetry, J. Pharm.Sci. 85 (1996) 381–386.

45] E.O. Timmermann, Multilayer sorption parameters: BET or GAB values? ColloidSurf. A: Physicochem. Eng. Aspects 220 (2003) 235–260.

46] R.B. Anderson, Modifications of the Brunauer, Emmett and Teller equation1, J.Am. Chem. Soc. 68 (1946) 686–691.

47] R.B. Anderson, W.K. Hall, Modifications of the Brunauer, Emmett and TellerEquation II1, J. Am. Chem. Soc. 70 (1948) 1727–1734.

48] K.S. Sing, Reporting physisorption data for gas/solid systems with special ref-erence to the determination of surface area and porosity (Recommendations1984), Pure Appl. Chem. 57 (1985) 603–619.

49] N. Bhouri, S. Bennasrallah, P. Perre, Influence of geometrical structure onsorption isotherms of Jersey and yarns made of cotton at two temperatures,Microporous Mesoporous Mater. 163 (2012) 76–84.

50] L.p. Fan, M. Zhang, G.n. Xiao, J.c. Sun, Q. Tao, The optimization of vacuum fryingto dehydrate carrot chips, Int. J. Food Sci. Technol. 40 (2005) 911–919.

51] V. Fainerman, S. Lylyk, E. Aksenenko, J. Petkov, J. Yorke, R. Miller, Surface ten-sion isotherms, adsorption dynamics and dilational visco-elasticity of sodiumdodecyl sulphate solutions, Colloid Surf. A: Physicochem. Eng. Aspects 354

oration from porous media by Dynamic Vapor Sorption, Colloidsj.colsurfa.2014.11.011

(2010) 8–15.52] V. Ulaganathan, M. Krzan, M. Lotfi, S. Dukhin, V. Kovalchuk, A. Javadi, D. Gunes,

C. Gehin-Delval, K. Malysa, R. Miller, Influence of �-lactoglobulin and its sur-factant mixtures on velocity of the rising bubbles, Colloid Surf. A: Physicochem.Eng. Aspects 460 (2014) 361–368.