Ultrasonics Sonochemistry 17 (2010) 318–325

Contents lists available at ScienceDirect

Ultrasonics Sonochemistry

journal homepage: www.elsevier .com/locate /u l tsonch

Study on ultrasonically assisted emulsification and recovery of copper(II) fromwastewater using an emulsion liquid membrane process

Mahdi Chiha a,*, Oualid Hamdaoui a, Fatiha Ahmedchekkat b, Christian Pétrier c

a Laboratory of Environmental Engineering, Department of Process Engineering, Faculty of Engineering, University of Annaba, P.O. Box 12, 23000 Annaba, Algeriab LARMACS, Faculty of Engineering, University of Skikda, P.O. Box 26, 21000 Skikda, Algeriac LEPMI, Université Joseph Fourier, 38402 Saint Martin d’Hères Cedex, France

a r t i c l e i n f o a b s t r a c t

Article history:Received 19 May 2009Received in revised form 30 August 2009Accepted 2 September 2009Available online 8 September 2009

Keywords:WastewaterCopper(II)Emulsion liquid membraneUltrasoundD2EHPA

1350-4177/$ - see front matter � 2009 Elsevier B.V. Adoi:10.1016/j.ultsonch.2009.09.001

* Corresponding author. Tel.: +213 790681263; faxE-mail address: chiha_m_f@yahoo.fr (M. Chiha).

The aim of this work was to study the emulsification assisted by ultrasonic probe (22.5 kHz) and inves-tigate the removal of copper(II) ions from aqueous solution using water-in-oil-in-water (W/O/W)emulsion liquid membrane process (ELM). The membrane was prepared by dissolving the extractantbis(2-ethylhexyl)phosphoric acid (D2EHPA) and the hydrophobic surfactant sorbitan monooleate (Span80) in hexane (diluent). The internal phase consisted of an aqueous solution of sulfuric acid. Effects ofoperating parameters such as emulsification time, ultrasonic power, probe position, stirring speed, carrier(D2EHPA) and surfactant (Span 80) concentrations volume ratios of organic phase to internal stripingphase and of external aqueous phase to membrane (W/O) phase, internal phase concentration and choiceof diluent on the membrane stability were studied. With ultrasound, the W/O emulsion lifetime weremuch higher than those reported previously by mechanical agitation. The effect of carrier and Cu(II) ini-tial concentration on the extraction kinetics was also investigated. Nearly all of the Cu(II) ions present inthe continuous phase was extracted within a few minutes. Additionally, the influence of H2SO4 concen-tration on the stripping efficiency was examined.

� 2009 Elsevier B.V. All rights reserved.

1. Introduction

Water contamination by heavy metals is a serious environmen-tal problem, which has been extensively discussed. One of thepromising techniques for the separation of heavy metal ions fromwastewater is the emulsion liquid membrane (ELM) technique,which is first described by Li [1] and other scientists [2]. Emulsionliquid membrane separation process constitutes an emerging tech-nology with a wide variety of applications, such as the removal,recovery, and purification of many heavy metal ions from dilutesolutions of industrial interest.

In the ELM process, a simultaneous extraction and stripping areaccomplished through a large surface area (1000–3000 m2/m3) of aliquid membrane prepared with a minimum quantity of selectiveextractant [3]. Multiple emulsion systems are generally preparedusing a two-step process. In the W1/O/W2 multiple emulsion type,primary water-in-oil (W1/O) emulsion is prepared in a first stepusing a hydrophobic surfactant. The W1/O emulsion is then dis-persed in a second aqueous phase W2 (Fig. 1). The size of the dis-persed drops is important for determining the stability of the

ll rights reserved.

: +213 38876560.

liquid membrane, efficiency of extraction and evaluating the inter-facial contact area. It is also important to obtain information aboutthe swelling, breakage and coalescence of dispersed drops.

Unfortunately, ELM process has not found widespread usage inindustrial application. The slow industrialization of this technologyis mainly due to the stability problems (i.e., the tendency of anemulsion to remain dispersed and resist coalescence) associatedwith the emulsion and their tendency to undergo swelling [4].The main factors affecting emulsion stability encompass mem-brane formulation, technique of emulsion preparation, and thecondition under which the emulsion is contacted with an externalphase. Energy must be supplied to produce such meta-stable mix-tures. Energy may be provided through various means includingmechanical agitation (stirrer, colloid mill, mixer, valve homoge-nizer) and ultrasound generation. In water/oil system, the processof emulsification assisted by ultrasound initiates when the cavita-tion threshold is attained. Ultrasound can provide an excess energyfor new interface formation; hence it is possible to obtain emul-sions even in few quantities of surfactants (emulsifiers).

The sono-emulsification process is dependent upon a range ofparameters, notably the acoustic power, location of the energy dis-sipating source and chemical composition of the system; beyond acertain range of optimum conditions, ultrasound may cause the in-verse effect (coagulation and precipitation phenomenon) [5].

Internal aqueous phase Membrane phase

Emulsification External aqueous phase

Splitting (demulsification)

Extract

Rafinate

Settling Extraction (Permeation)

Fig. 1. The emulsion liquid membrane process.

M. Chiha et al. / Ultrasonics Sonochemistry 17 (2010) 318–325 319

Ultrasound has been proved to be suitable for emulsification[6]. Ultrasound emulsification was reported for the first time byWood and Loomis [7]. With ultrasound, for example, the size ofW/O emulsions (Sauter diameter d32) is much smaller than thosegiven by mechanical agitation under the same conditions, as con-sequence it increase the stability [8].

Removal of copper ions from aqueous solution using ELMs hasbeen an area of interest ever since ELMs were invented. Mostinvestigators [9–12] preferred studying the effect of chemical com-position of the system using chelating extractants, such as LIX 63,LIX 64 N, LIX 65 and SME 529, because of their immiscibility withwater. Economic evaluation showed that ELM extraction of copperusing LIX 64 N turns out to be 40% cheaper than solvent extraction[13].

Little work has been reported in the literature for the utilizationof ultrasound in all the steps of ELM process (emulsification,extraction and demulsification). Chakravarti et al. [14] have inves-tigated the extraction of copper by ELMs using mechanical emulsi-fication. Also, Sengupta et al. [15] preferred using mechanicalagitation (12000 rpm) to prepare W/O emulsion for copper extrac-tion with LIX 984 N-C as carrier. Valenzuela et al. [16,17] havestudied the removal of copper ions from a wastewater by a liquidemulsion membrane method using an Ultraturrax Janke Kunkelultrasonic agitator in first step preparation of W/O emulsion andan Elma Ultrasonic bath in second step was used to acceleratethe phase separation process by enhancing the coalescence of thesmall droplets. Behrend et al. [18,19] have investigated vegetableoil-in-water emulsion stability using ultrasonic emulsification be-cause the power density sufficiently characterizes the process con-ditions in terms of droplet disruption. Juang and Lin [20] havestudied the efficiency of emulsification with ultrasonic probe(20 kHz) and the possibility of ultrasound demulsification. Theyhave indicated that emulsification can be realized with the supplyof a suitable form of energy. All the oil-in-water (O/W) emulsionswere prepared by the ultrasonic method (Ultrasonic 42, Branson)by Zha et al. [21]. By understanding the relationship (ultrasound –

emulsification), industrial processes involving ultrasound can beoptimized for scale up.

In the present work, the preparation of W/O emulsion assistedby ultrasound and the extraction of copper from aqueous mediainto ELMs using D2EHPA as carrier are reported. The influence ofoperational conditions on the prepared W/O emulsion as well ason the extraction efficiency by ELM was investigated.

2. Experimental section

2.1. Reagents

D2EHPA (bis(2-ethylhexyl) phosphoric acid) was analyticalgrade product (Aldrich) and was used as received. The non-ionicsurfactant Span 80 (sorbitan monooleate, Aldrich) was used as anemulsifier. Diluents (hexane, heptane and dodecane) were ob-tained from Fluka. Copper(II) solutions were prepared by dissolv-ing requisite amount of copper sulfate (CuSO4�5H2O, Prolabo) inbidistilled water. Analytical pure sulfuric acid, obtained fromMerck, was employed for the preparation of the aqueous internalphase.

2.2. Apparatus and measurement

The 22.5 kHz ultrasonic wave was emitted from a titanium horn(tip diameter 3 mm) connected to a commercial supply (Microson2000 XL). Acoustic power dissipated in the reactor was estimatedusing standard calorimetric method [22,23]. The used experimentalset-up was shown in Fig. 2. The determination of Cu(II) concentra-tion was carried out using atomic absorption spectrophotometer(AAS, Shimadzu A. A-6601 F, Atomic Absorption Flame EmissionSpectrophotometer) at 325 nm. Each experiment was performedtwice at least and the mean values were presented. The reproduc-ibility of the concentrations measurements was within 3%.

Cooling water outlet

Cooling water inlet

To generator

Transducer

Ultrasound Probe

Vessel

Tip

Fig. 2. Experimental set-up for ultrasound-assisted preparation of the W/O emulsion.

320 M. Chiha et al. / Ultrasonics Sonochemistry 17 (2010) 318–325

2.3. Experimental procedures

The ELM used in this work was water-in-oil-in-water (W/O/W)emulsion formed by dissolving the extractant (D2EHPA) and thesurfactant (Span 80) in diluents (hexane, heptane or dodecane).Internal aqueous phase was a solution of H2SO4. Water-in-oil (W/O) emulsion was made by slowly adding the internal phase tothe organic membrane phase upon intensive emulsification withthe ultrasonic probe.

A certain volume of the obtained stable emulsion (20 mL) wasdispersed in the feed phase (100 mL of aqueous solution). Theextraction runs were performed in a glass vessel of 61 mm diame-ter using a mechanical agitator (Junke & Kunkel RW20) with four-paddle impeller of 20 mm diameter. All experiments were carriedout at constant temperature (25 ± 1 �C) using water jacket aroundthe vessel. The concentration of copper in the aqueous externalphase was determined by AAS and in the internal phase was deter-mined by mass balance.

0

5

10

15

20

25

30

0 2 4 6 8 10 12Emulsification time (min)

Bre

akag

e (%

)

Fig. 3. Effect of emulsification time on the W/O emulsions stability (experimentalconditions: emulsion volume: 20 mL; external phase (pure water) volume: 100 mL;volume ratio of internal phase to organic phase: 1; emulsification time: 1–10 min;ultrasound power: 20 W, stirring speed: 200 rpm; concentration of Span 80: 4% w/w; carrier concentration: 20% w/w; volume ratio of W/O emulsions to externalphase: 0.2; internal phase concentration (H2SO4): 0.3 N; diluent: hexane; contacttime: 18 min; distance of the tip horn from the bottom of vessel: 20 mm).

3. Results and discussion

3.1. Emulsion stability

Several parameters can affect the stability of W/O emulsionsuch as the ultrasound power, emulsification time, probe position,stirring speed, surfactant and carrier concentrations, volume ratiosof organic phase to internal striping phase and of external aqueousphase to membrane (W/O) phase, internal phase concentration andchoice of diluent were studied. The produced W/O emulsionsshould be employed for the removal of copper(II) from aqueoussolutions.

The stability of emulsions was investigated by using the break-up, e, defined by the following equation:

e ¼ V s

V int� 100

The emulsion breakage represents the ratio in percentage of thevolume of internal phase leaked into the external phase by split-ting (VS) to the initial volume of the internal phase (Vint). The vol-ume VS is calculated by the mass balance.

V s ¼ Vext10�pH0 � 10�pH

10�pH � CintHþ

Where Vext is the initial volume of external phase, CintHþ the initial

concentration of H+ in the internal phase, pH0 the initial pH of theexternal phase and pH is the external phase pH being in contactwith the emulsion after a certain time of stirring.

3.1.1. Effect of emulsification timeOne of the most important parameter for ELM processes is the

emulsification time. Fig. 3 shows the effect of emulsification timeon the emulsion stability using a surfactant concentration of 4%(w/w) in the membrane phase (oil phase). For emulsification timeshigher than 3 min, it has been observed that the breakage percent-age gradually increased because of the high internal shearing lead-ing to a very high number of small droplets, which is conducive totheir diffusion into external phase (bidistilled water). For loweremulsification times (<3 min), the relatively higher breakage per-centage values are due to the large size of inner droplet that causes

0

10

20

30

40

50

60

70

0 100 200 300 400 500 600Stirring speed (rpm)

Bre

akag

e (%

)

Fig. 5. Effect of stirring speed on the W/O emulsions stability (experimentalconditions: emulsion volume: 20 mL; external phase (pure water) volume: 100 mL;volume ratio of internal phase to organic phase: 1; emulsification time: 3 min;ultrasound power: 20 W, stirring speed: 100–500 rpm; concentration of Span 80:4% w/w; carrier concentration: 20% w/w; volume ratio of W/O emulsions toexternal phase: 0.2; internal phase concentration (H2SO4): 0.3 N; diluent: hexane;contact time: 18 min; distance of the tip horn from the bottom of vessel: 20 mm).

0

2

4

6

8

10

12

14

16

0 0.2 0.4 0.6 0.8 1 1.2 1.4

Sulfuric acid concentration (N)

Bre

akag

e (%

)

Fig. 6. Effect of internal aqueous phase (H2SO4) concentration on the W/Oemulsions stability (experimental conditions: emulsion volume: 20 mL; externalphase (pure water) volume: 100 mL; volume ratio of internal phase to organicphase: 1; emulsification time: 3 min; ultrasound power: 20 W, stirring speed:200 rpm; concentration of Span 80: 4% w/w; carrier concentration: 20% w/w;volume ratio of W/O emulsions to external phase: 0.2; internal phase concentration(H2SO4): 0.1–1.2 N; diluent: hexane; contact time: 18 min; distance of the tip hornfrom the bottom of vessel: 20 mm).

M. Chiha et al. / Ultrasonics Sonochemistry 17 (2010) 318–325 321

coalescence phenomenon. The lower breakage is obtained for anemulsification time of 3 min.

3.1.2. Effect of ultrasonic powerThe emulsification efficiencies at different ultrasound powers

are shown in Fig. 4. The emulsion breakage profile shows theimportant effect of ultrasound power on emulsion stability. Forlow power values (5–15 W), the sound field is insufficient to givethe necessary energy for a good dispersion of aqueous droplets inthe membrane phase. The percentage of breakage decreaseswith the increase of the power of ultrasound. 20 W is the optimumvalue. Beyond this value, higher ultrasonic power produces higherspecific surface area of W/O emulsions, but the phenomenon ofcoalescence is more significant. Freitas et al. [24] have found thesame behavior and showed that oil droplets of 5–10 lm diameterscould be observed in emulsions processed at an ultrasonic powerof 25 W, and practically no droplets were microscopically visibleat 32 W, i.e. at full power.

3.1.3. Effect of stirring speedStirring speed directly influences membrane stability behavior.

Since globule size gets affected by stirring, higher agitation speedslead to the formation of smaller sized globules, thereby increasingthe interfacial area between continuous phase and the membranephase, and accelerate the mass transfer for extraction. Also, higherstirring speed may bring about higher breakage percentage. There-fore, it is very important to select a suitable stirring speed and keepstable mixing conditions during the process in order to maintainadequate membrane stability and minimize the emulsion swelling.According to Fig. 5, for lower stirring speeds (100 and 150 rpm con-ditions), the breakage percentage is low because the size of theemulsion globules increases and the interfacial area available formass transfer decreases. Increasing the stirring speed above a crit-ical value (200 rpm) affects the stability of the emulsion and in-creases the osmotic swelling of the membrane and makes theemulsion unstable. Thus, 200 rpm was the best stirring speed toensure a good stability of the W/O emulsion and to enhance theinterfacial area available for mass transfer for the subsequentexperiments to fulfill the demand of well dispersing and lowbreakage.

3.1.4. Effect of internal phase concentrationThe effect of internal phase concentration (H2SO4) on the ELM

stability is illustrated in Fig. 6. When the sulfuric acid concentra-

0

2

4

6

8

10

12

14

16

18

0 5 10 15 20 25 30 35 40

Ultrasonic power (W)

Bre

akag

e (%

)

Fig. 4. Effect of ultrasonic power on the W/O emulsions stability (experimentalconditions: emulsion volume: 20 mL; external phase (pure water) volume: 100 mL;volume ratio of internal phase to organic phase: 1; emulsification time: 3 min;ultrasound power: 10–35 W, stirring speed: 200 rpm; concentration of Span 80:4% w/w; carrier concentration: 20% w/w; volume ratio of W/O emulsions toexternal phase: 0.2; internal phase concentration (H2SO4): 0.3 N; diluent: hexane;contact time: 18 min; distance of the tip horn from the bottom of vessel: 20 mm).

tion increased from 0.1 to 0.3 N, the percentage of breakage de-creased. A concentration of 0.1 N leads to high breakagepercentage because the ionic strength difference between internaland external phases is not sufficient conducting in high emulsionbreakage. Concentration of the internal phase giving the weakestbreakage (0.3 N) was chosen in this study. Beyond this concentra-tion the emulsion stability slightly decreased. The slight decreaseof the W/O emulsion stability may be due to the reaction of sulfuricacid with Span 80, which results in a partial loss of its surfactantproperties [30].

3.1.5. Effect of extractant concentrationThe effect of D2EHPA (carrier) concentration on the W/O emul-

sion stability is shown in Fig. 7. This figure shows that when thecarrier concentration in the membrane phase increased from 5%to 20% (w/w), the breakage percentage decreased. This behavioris due to the interfacial properties of the D2EHPA that leads to agood stability of the emulsion. A very high content of carrier inthe membrane (higher than 20% (w/w)) does not result in a benefitdue to the increase in viscosity, which leads to larger globules.Swelling phenomenon was observed when carrier concentration

0

2

4

6

8

10

12

14

16

18

20

0 10 20 30 40 50

Extractant concentration (% w/w)

Bre

akag

e (%

)

Fig. 7. Effect of carrier concentration on the W/O emulsions stability (experimentalconditions: emulsion volume: 20 mL; external phase (pure water) volume: 100 mL,volume ratio of internal phase to organic phase: 1; emulsification time: 3 min;ultrasound power: 20 W, stirring speed: 200 rpm; concentration of Span 80: 4% w/w; carrier concentration: 5–40% w/w; volume ratio of W/O emulsions to externalphase: 0.2; internal phase concentration (H2SO4): 0.3 N; diluent: hexane; contacttime: 18 min; distance of the tip horn from the bottom of vessel: 20 mm).

0

5

10

15

20

25

30

35

40

0 0.5 1 1.5 2 2.5Vint /Vorg

Bre

akag

e (%

)

Fig. 9. Effect of the volume ratio of internal phase to organic phase on the W/Oemulsions stability (Experimental conditions: emulsion volume: 20 mL; externalphase (pure water) volume: 100 mL; volume ratio of internal phase to organicphase: 0.5–2; emulsification time: 3 min; ultrasound power: 20 W, stirring speed:200 rpm; concentration of Span 80: 4% w/w; carrier concentration: 20% w/w;volume ratio of W/O emulsions to external phase: 0.2; internal phase concentration(H2SO4): 0.3 N; diluent: hexane; contact time: 18 min; distance of the tip horn fromthe bottom of vessel: 20 mm).

322 M. Chiha et al. / Ultrasonics Sonochemistry 17 (2010) 318–325

is 40% because D2EHPA is an anionic surfactant in nature, thestability of W/O emulsion in its presence would decrease due towater transport [20]. The optimum concentration of carrier was20% (w/w).

3.1.6. Effect of surfactant concentrationSurfactant concentration has an important bearing on the sta-

bility of the emulsion. Fig. 8 shows the effect of surfactant concen-tration on the W/O emulsion breakage after stirring at 200 rpm.The percentage of breakage was less than 10% for Span 80 concen-trations above 4% (w/w), indicating that the membrane was suffi-ciently stable, while the breakage percentage increases with thedecrease in surfactant concentration (less than 4% (w/w)). BelowSpan 80 concentration of 2% (w/w), no emulsion was formed owingto a lack of the surfactant adsorbing the organic/aqueous interface.The concentration of surfactant chosen in this work to produce effi-ciently stabilized membrane was 4% (w/w).

3.1.7. Effect of volume ratio of aqueous internal phase to organic phase(Vint/Vorg)

Another important experimental parameter is the appropriatevolume ratio of inner aqueous solution (the stripping phase) to or-

0

10

20

30

40

50

60

0 1 2 3 4 5 6 7 8 9Surfactant concentration (% w/w)

Bea

kage

(%

)

Fig. 8. Effect of surfactant concentration on the W/O emulsions stability (exper-imental conditions: emulsion volume: 20 mL; external phase (pure water) volume:100 mL; volume ratio of internal phase to organic phase: 1; emulsification time:3 min; ultrasound power: 20 W, stirring speed: 200 rpm; concentration of Span 80:2–8% w/w; carrier concentration: 20% w/w; volume ratio of W/O emulsions toexternal phase: 0.2; internal phase concentration (H2SO4): 0.3 N; diluent: hexane;contact time: 18 min; distance of the tip horn from the bottom of vessel: 20 mm).

ganic solution in the emulsion phase. Fig. 9 shows the experimen-tal results of the membrane stability as a function of this ratio inthe range 0.5–2. It can be seen that the breakage percentage isdependent of the Vint/Vorg ratio. When the ratio increased from0.5 to 0.8, the stability increased slightly, because the ejection ofthe internal phase is easier when its proposition in the emulsionis higher. These results may be explained on the basis that increas-ing the internal phase volume in this interval makes the emulsionmore stable. This quantity is insufficient for mass transfer in thesecond step. Hence, in order to obtain a uniform and homogeneousdistribution of the internal phase droplets in the membrane spaceand to avoid the influence of H2SO4 on the emulsion stability, theoptimum ratio of the internal aqueous phase to the organic phasewas taken to be 1. Beyond the ratio of 1.2, further increase in thevolume of stripping solution leads to a decrease in membrane sta-bility, because when the volume of the stripping phase increases,the thickness of film in droplets thin off.

3.1.8. Effect of volume ratio of water-in-oil (W/O) emulsions toaqueous external phase

This parameter expresses the treatment ratio of the wholeemulsion liquid membrane process. Its variation cannot be ran-dom, because it is necessary to guarantee that the acid quantitypresent in the internal phase is high enough to react with the sol-ute present in the external phase. Experiments were conducted bytaking into account the parameters already optimized and using avolume ratio of the organic phase to the aqueous internal phase of1. The volume ratio of emulsion to external phase varied between0.05 and 1. As shown in Fig. 10, the increase of volume ratio ofemulsion to aqueous external phase (Vem/Vext) leads to an increaseof the breakage percentage. The increase of volume ratio of themembrane phase to the external phase beyond 0.4 led to an in-crease of the emulsion coagulation. Indeed, with increasing thevolume ratio, the swelling phenomenon becomes remarkable, fast,and accompanied by an embrittlement following a more signifi-cant coalescence of the internal droplets which grow. This behaviorinvolves an increase of the emulsion breakage. Therefore, it seemsthat the optimum volume ratio of emulsion to the external aque-ous phase is 0.2.

3.1.9. Choice of diluentAlthough experiments have been performed with aromatic dil-

uents [25], aliphatic diluents are generally preferred because of the

0

1

2

3

4

5

6

7

8

0 50 100 150 200 250 300 350

Time (s)

Bre

akag

e (%

)

Dodecane

Heptane

Hexane

Fig. 11. Effect of diluents nature on the W/O emulsions stability (experimentalconditions: emulsion volume: 20 mL; external phase (pure water) volume: 100 mL;volume ratio of internal phase to organic phase: 1; emulsification time: 3 min;ultrasound power: 20 W, stirring speed: 200 rpm; concentration of Span 80: 4% w/w; carrier concentration: 20% w/w; volume ratio of W/O emulsions to externalphase: 0.2; internal phase concentration (H2SO4): 0.3 N; diluent: hexane, heptaneand dodecane; contact time: 18 min; distance of the tip horn from the bottom ofvessel: 20 mm).

0

1

2

3

4

5

6

7

8

9

10

20 22 24 26 28 30 32 34 36

Distance of the tip of probe from the botom of vessel (mm)

Bre

akag

e (%

)

Fig. 12. Effect of the tip of horn position on the W/O emulsions stability(experimental conditions: emulsion volume: 20 mL; external phase (pure water)volume: 100 mL; volume ratio of internal phase to organic phase: 1; emulsificationtime: 3 min; ultrasound power: 20 W, stirring speed: 200 rpm; concentration ofSpan 80: 4% w/w; carrier concentration: 20% w/w; volume ratio of W/O emulsionsto external phase: 0.2; internal phase concentration (H2SO4): 0.3 N; diluent:hexane; contact time: 18 min; distance of the tip horn from the bottom of vessel:20–35 mm).

0

10

20

30

40

50

60

70

80

90

0 0.2 0.4 0.6 0.8 1 1.2Vem /Vext

Bre

akag

e (%

)

Fig. 10. Effect of the volume ratio of the emulsion to the external phase on the W/Oemulsions stability (experimental conditions: volume ratio of internal phase toorganic phase: 1; emulsification time: 3 min; ultrasound power: 20 W, stirringspeed: 200 rpm; concentration of Span 80: 4% w/w; carrier concentration: 20% w/w; volume ratio of W/O emulsions to external phase: 0.05–1; internal phaseconcentration (H2SO4): 0.3 N; diluent: hexane; contact time: 18 min; distance ofthe tip horn from the bottom of vessel: 20 mm).

M. Chiha et al. / Ultrasonics Sonochemistry 17 (2010) 318–325 323

insolubility in water and better emulsion stability. The mass trans-fer is in many cases diffusion controlled, and therefore the viscosityof the diluent is an important parameter. As shown in Fig. 11, suit-able diluent for a good stability of the W/O emulsions is hexanewith viscosity of 0.294 cP, because hexane is less viscous than hep-tane and dodecane (0.386 and 1.34 cP, respectively). This is inagreement with the ultrasound theory; cavities are more readilyformed when using a solvent with a high vapor pressure and lowviscosity [26]. Because viscosity is a measure of resistance to shear,it is more difficult to produce cavitation in a viscous liquid [27].

3.1.10. Effect of the probe positionFig. 12 shows the effect of probe position in the emulsifying

vessel on the emulsion stability. It is well known that the ultra-sonic energy produces an alternating adiabatic compression andrefraction of the liquid being irradiated. In the refraction part ofultrasonic wave, micro-bubbles form due to sufficiently large neg-ative pressures. These bubbles contain gas and vaporized liquidand can be either stable about their mean size for many cycles ortransient as they grow to certain size and violently collapse duringthe compression part of the wave, i.e. imploding cavitation bubbles

cause intensive shock waves in the surrounding liquid and the for-mation of liquid jets of high velocity. According to these phenom-ena, in this work, the position of the tip of horn ultrasound (Fig. 1)from the bottom of vessel affects the emulsion stability. The dis-tance of this tip from vessel bottom (20 mm) that conducted tothe lower breakage is selected because this distance coincide tocavitation rich region. When the distance of probe from the bottomis short, a good stability of emulsion was obtained. In this case, thereal microstreaming phenomenon is obtained by the cavitationrich region and the presence of turbulent regime due the reflectionof ultrasound wave.

3.2. Copper extraction into ELM

Extraction of copper(II) by ELM process is governed by severalparameters and poses a challenging problem in the field of hydro-metallurgy. The equations given below exhibit extraction (1) andstripping (2) reactions of copper occurring in ELM technique [28]

Cu2þaq þ 2H2R2org ¡ CuR2 � 2HRorg þ 2Hþaq ð1Þ

CuR2 � 2HRorg þ 2Hþint ¡ Cu2þint þ 2H2R2org ð2Þ

where, (H2R2) represents the D2EHPA dimmer and the molecularstructure of the complex CuR2�2HRorg is shown in Fig. 13.

Facilitated transport of behavior through ELM, being permeatedfrom the external into the internal aqueous phase, consisted of foursteps: (1) diffusion of copper through the external aqueous bound-ary film, (2) distribution of copper between the external aqueousand liquid membrane phase, (3) diffusion of copper through themembrane phase (Fig. 13), and (4) stripping of copper, accompa-nied by reaction (2), into the internal aqueous phase.

The performance of the extraction depends on the nature ofemulsion (previously optimized) and various operational parame-ters as detailed below.

3.2.1. Effect of copper concentration in the external aqueous phaseThe effect of copper ion concentration (50, 100, 150 and

200 ppm) in the feed phase on the extraction efficiency was inves-tigated. The extraction was performed at pH 2 using a strippingacid concentration of 0.3 N. The obtained results are presented inFig. 14. It is observed that the percentage of extraction decreasedwhen the initial content of copper increased. At low copper con-centrations (50–100 ppm), high extraction efficiency is obtained

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20

Time (min)

Ext

ract

ion

effi

cien

cy (

%)

50ppm

100ppm

150ppm

200ppm

Fig. 14. Effect of external phase concentration on the extraction efficiency andextraction kinetics (experimental conditions: emulsion volume: 20 mL; externalphase volume: 100 mL; volume ratio of internal phase to organic phase: 1;emulsification time: 3 min; ultrasound power: 20 W, stirring speed: 200 rpm;concentration of Span 80: 4% w/w; carrier concentration: 20% w/w; volume ratio ofW/O emulsions to external phase: 0.2; internal phase concentration (H2SO4): 0.3 N;diluent: hexane; contact time: 18 min; distance of the tip horn from the bottom ofvessel: 20 mm).

0

10

20

30

40

50

60

70

80

90

100

5 10 15 20 25 30

D2EHPA (%)

Ext

ract

ion

effi

cien

cy (

%)

Fig. 15. Effect of carrier concentration on the extraction efficiency (experimentalconditions: emulsion volume: 20 mL; external phase volume: 100 mL; volume ratioof internal phase to organic phase: 1; emulsification time: 3 min; ultrasoundpower: 20 W, stirring speed: 200 rpm; concentration of Span 80: 4% w/w; carrierconcentration: 5–30% w/w; volume ratio of W/O emulsions to external phase: 0.2;internal phase concentration (H2SO4): 0.3 N; diluent: hexane; contact time: 18 min;Distance of the tip horn from the bottom of vessel: 20 mm).

10

20

30

40

50

60

70

80

90

100

Ext

ract

ion

effi

cien

cy (

%)

O HO OC8H17 H17C8O

P P

O OC8H117 O H17C8O

Cu

O O

CH3

Fig. 13. Molecular structure of the complex formed between coppper(II) andD2EHPA dimmer in organic phase.

324 M. Chiha et al. / Ultrasonics Sonochemistry 17 (2010) 318–325

(95%). But for 200 ppm solutions, copper extraction reached about74%. This is due the saturation of the internal droplets of the emul-sion that is attained more rapidly for high concentrations in theexternal phase. When the solute concentration is high, the coppercomplex must diffuse through the membrane phase to the internalphase. This suggests that the mass transfer resistance in the emul-sion globule is important. In the extraction kinetic process, thecontact time between W/O emulsion and feed aqueous phase hasa significant influence on the extraction of copper. According toFig. 14, it was observed that the kinetics of the process is very fast.The level of extraction increased with the increase in contact time.For copper concentrations of 50 and 100 ppm, the removal per-centage is good and exceeded 95% after 18 min.

0100 200 250 300 400 500 600 800 1000 1200

Stirring speed (rpm)

Fig. 16. Effect of Stirring speed on the extraction efficiency (experimental condi-tions: emulsion volume: 20 mL; external phase volume: 100 mL; volume ratio ofinternal phase to organic phase: 1; emulsification time: 3 min; ultrasound power:20 W, stirring speed: 100–1200 rpm; concentration of Span 80: 4% w/w; carrierconcentration: 20% w/w; volume ratio of W/O emulsions to external phase: 0.2;internal phase concentration (H2SO4): 0.3 N; diluent: hexane; contact time: 18 min;distance of the tip horn from the bottom of vessel: 20 mm).

3.2.2. Effect of carrier concentrationFig. 15 shows the effect of D2EHPA concentration on the ELM

extraction of copper. It was found that extraction increased withincreasing carrier concentration. This is because increase inD2EHPA concentration will increase the mass transfer. On theother hand, an increase in carrier concentration in the membranephase increases the viscosity of this membrane, which limits theextraction rate of the metal ions. Above a certain value (20%

(w/w)), the permeation efficiency decreases by increasing the car-rier concentration because the extractant acts as a thinner for themembrane phase (i.e. the increasing concentration of carrier pro-motes the permeation swelling, which dilutes the aqueous receiv-ing phase and decreases the efficiency of the process). Additionally,the stability of emulsion decreases by increasing the carrier con-centration over a certain limit. The best value of D2EHPA concen-tration was found to be 20% (w/w).

3.2.3. Effect of stirring speedThe stirring speed has a profound influence on the extraction

behavior. The effect of this parameter on the extraction of copperions by ELM is shown in Fig. 16. By increasing the stirring speedfrom 100 to 200 rpm, the shear force, which acts on the emulsionglobules, increases and this makes the globules smaller resultingin an enhancement of the mass transfer and extraction efficiency.Increasing the stirring speed in the range 200–500 rpm not onlydecreases the extraction efficiency slightly, but also affects the sta-bility of the emulsion and makes the emulsion unstable. When thelevel of stirring speed is increased above 500 rpm, the extractionefficiency significantly decreased because high stirring speedproduced coalescence and finally rupture of globules, making the

0

10

20

30

40

50

60

70

80

90

100

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 1 1.2 1.4

Sulfuric acid concentration (N)

Ext

ract

ion

effi

cien

cy (

%)

Fig. 17. Effect of H2SO4 concentration in the internal phase on Cu(II) stripping.(experimental conditions: emulsion volume: 20 mL; external phase volume:100 mL; volume ratio of internal phase to organic phase: 1; emulsification time:3 min; ultrasound power: 20 W, stirring speed: 200 rpm; concentration of Span 80:4% w/w; carrier concentration: 20% w/w; volume ratio of W/O emulsions toexternal phase: 0.2; internal phase concentration (H2SO4): 0.1–1.4 N; diluent:hexane; contact time: 18 min; distance of the tip horn from the bottom of vessel:20 mm).

M. Chiha et al. / Ultrasonics Sonochemistry 17 (2010) 318–325 325

primary emulsion unstable. A stirring speed of 200 rpm was cho-sen in this study.

3.2.4. Effect of H2SO4 concentration in the internal phase on Cu(II)stripping

The capacity of the emulsion to extract solute is limited by theinternal phase stripping acid concentration. The influence of theinternal phase reagent concentration on the extraction efficiencyis shown in Fig. 17. It was found that the extraction increased whenthe concentration of sulfuric acid in the internal phase increasedfrom 0.1 to 0.3 N. However, further increases in the concentrationreduce the level of copper extraction. This is a result of the increasein the ionic strength difference between internal and externalphases that resulted in an increase in swelling, which conductedin greater amounts of water to permeate through the membranecausing the internal phase droplets to swell and coalesce and final-ly breakdown of globules. On the other hand, high acid content instripping solution would cause a faster degradation of bothextractant and surfactant compounds. Internal phase concentra-tion of 0.3 N was chosen in the present study.

4. Conclusions

Sonication can be used to produce stable W/O emulsion in ELMprocess. This is possible because cavitational collapse at or near theinterface disrupts it and impels jets of one liquid into the other toform the emulsion. Optimum sonication parameters giving stableemulsion are an ultrasonic power of 20 W and a distance of20 mm of the probe from the bottom of emulsification cell. Theconcentrations of carrier and surfactant leading to stable emul-sions were 20% and 4% w/w, respectively. The concentration of sul-furic acid which conducts to the lower breakage was 0.3 N. Thevolume ratio of organic phase to aqueous internal phase giving astable emulsion was selected to be 1. The effect of the volume ratioof water-in-oil (W/O) emulsions to aqueous external phase on thestability was 0.2. The stirring speed which ensures a good stabilityof the emulsion and an acceptable size of the globules ensuring fastremoval kinetics from aqueous external solution was 200 rpm.Hexane has a low viscosity compared to heptane and dodecane,which facilitates ultrasonic cavitation and ensures a good emulsifi-cation. The W/O emulsion lifetime produced by ultrasonication ismuch higher than that prepared previously by mechanical agita-tion compared with previous studies [29]. The exact process ofdroplet disruption due to ultrasound, especially as a result of

cavitation, is not yet fully understood. D2EHPA can be used to sep-arate copper ions from aqueous solutions in an ELM process with95% of extraction efficiency. An extractant concentration of 20%(w/w) in the membrane phase and a concentration of H2SO4 of0.3 N in the internal solution were found to be optimum conditionsfor obtaining a higher separation factor. It was found that 200 rpmcan be recommended as the best stirring speed.

References

[1] N.N. Li, Separating hydrocarbons with liquid membranes, U.S. Patent 3, 1968,pp. 410–794.

[2] G. Sznejer, A. Marmur, Cadmium removal from aqueous solution by anemulsion liquid membrane: the effect of resistance to mass transfer at theouter oil–water interface, Collo Surf. A: Physicochem. Eng. Asp. 151 (1999) 77–83.

[3] G. Schulz, Separation techniques with supported liquid membranes,Desalination 68 (1988) 191–202.

[4] J. Yang, R. Pal, Effects of aqueous-phase acidity and salinity on isotonic swellingof W/O/W emulsion liquid membranes under agitation conditions, J. Membr.Sci. 244 (2004) 193–203.

[5] J.D. Wadhawan, F. Marken, R.G. Compton, Biphasic sonoelectrosynthesis. Areview, Pure Appl. Chem. 73 (2001) 1947–1955.

[6] D. Ensminger, Ultrasonic: the Low- and High-Intensity Applications, MarcelDekker, New York, 1973.

[7] W. Wood, A.L. Loomis, The physical and biological effects of high frequencysound waves of great intensity, Philos. Mag. J. Sci. 44 (22) (1927), pp. 2E–2F3.

[8] B. Abismail, J.P. Conselier, A.M. Wilhelm, H. Delmas, C. Gourdon, Emulsificationby ultrasound: drops size distribution and stability, Ultrason. Sonochem. 6(1999) 75–83.

[9] T. Kitagawa, Y. Nishikawa, J.W. Frankenfeld, N.N. Li, Waster water treatment byliquid membrane process, Environ. Sci. Technol. 11 (1977) 602–605.

[10] B.A. Mikucki, K. Osseo-Asare, The liquid surfactant membrane process: effectof the membrane process: effect of the emulsifier type on copper extraction byLIX65 N-LIX 63 mixtures, Hydrometallurgy 16 (1986) 209–229.

[11] T.P. Martin, G.A. Davies, The extraction of copper from dilute aqueoussolutions using a liquid membrane process, Hydrometallurgy 2 (1977) 315–334.

[12] W. Volkel, W. Halwachs, K. Schugerl, Copper extraction by means of a liquidsurfactant membrane process, J. Membr. Sci. 6 (1980) 19–31.

[13] J.W. Frankenfeld, R.P. Cahn, N.N. Li, Extraction of copper by liquid membranes,Sep. Sci. Technol. 16 (4) (1981) 385–402.

[14] A.K. Chakravarti, S.B. Chowdhury, D.C. Mukherjee, Liquid membrane multipleemulsion process of separation of copper(II) from waste waters, Colloids Surf.A: Physicochem. Eng. Asp. 166 (2000) 7–25.

[15] B. Sengupta, R. Sengupta, N. Subrahmanyam, Copper extraction into emulsionliquid membranes using LIX 984N-C, Hydrometallurgy 81 (2006) 67–73.

[16] F. Valenzuela, C. Fonesca, C. Basualto, O. Correa, C. Tapia, J. Sapag, Removal ofcopper ions from a waste mine water by a liquid emulsion membrane method,Miner. Eng. 18 (2005) 33–40.

[17] F. Valenzuela, J. Cabrera, C. Basualto, J.S. Hagar, Kinetics of copper removalfrom acidic mine drainage by a liquid emulsion membrane, Miner. Eng. 18(2005) 1224–1232.

[18] O. Behrend, K. Ax, H. Schubert, Influence of continuous phase viscosity onemulsification by ultrasound, Ultrason. Sonochem. 7 (2000) 77–85.

[19] O. Behrend, H. Schubert, Influence of ultrasonic pressure and gas content oncontinuous ultrasound emulsification, Ultrason. Sonochem. 8 (2001) 271–276.

[20] R.S. Juang, K.H. Lin, Ultrasound-assisted production of W/O emulsions in liquidsurfactant membrane process, Colloids Surf. A: Physicochem. Eng. Asp. 238(2004) 43–49.

[21] F.F. Zha, A.G. Fane, C.J.D. Fell, Instability mechanisms of supported liquidmembranes in phenol transport process, J. Membr. Sci. 107 (1995) 59–74.

[22] S. Koda, T. Kimura, H. Kondo, H. Mitone, A standard method to calibratesonochemical efficiency of an individual reaction system, Ultrason. Sonochem.10 (2003) 149–156.

[23] T.J. Mason, J.P. Lorimer, D.M. Bates, Quantifying sonochemistry: casting somelight on a ‘black art’, Ultrasonics 30 (1992) 40–42.

[24] S. Freitas, G. Hielscher, H.P. Merkle, B. Gander, Continuous contact-andcontaminant-free ultrasonic emulsification- a useful tool for pharmaceuticaldevelopment and production, Ultrason. Sonochem 13 (2006) 76–85.

[25] J. Draxler, R. Marr, Emulsion liquid membranes, Part I: phenomenon andindustrial application, Chem. Eng. Process. 20 (1986) 319–329.

[26] L.H. Tompson, L.K. Doraiswamy, Sonochemistry: science and engineering, Ind.Eng. Chem. Res. 38 (1999) 1215–1249.

[27] T.J. Mason, Sonochemistry, Oxford University, 1999, New York.[28] E.A. Fouad, Zinc and copper separation through an emulsion liquid membrane

containing di-(2-ethylhexyl)phosphoric acid as a carrier, Chem. Eng. Technol.31 (3) (2008) 370–376.

[29] M. Chiha, M.H. Samar, O. Hamdaoui, Extraction of chromium(VI) from acidaqueous solution by a liquid surfactant membrane (LSM), Desalination 194(2006) 69–80.

[30] R. Sabry, A. Hafez, M. Khedr, A. El-Hassanin, Removal of lead by an emulsionliquid membrane Part I, Desalination 212 (2007) 165–175.