Transport of Ag+ through tri-n-dodecylamine supported liquid membranes

Transport of Ag+ through tri-n-dodecylamine supported liquid membranes

Journal of Membrane Science 389 (2012) 287–293 Contents lists available at SciVerse ScienceDirect Journal of Membrane Science journal homepage: www...

702KB Sizes 0 Downloads 14 Views

Journal of Membrane Science 389 (2012) 287–293

Contents lists available at SciVerse ScienceDirect

Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci

Transport of Ag+ through tri-n-dodecylamine supported liquid membranes Saeed ur Rehman a,∗ , Gul Akhtar a , M. Ashraf Chaudry a , Khurshid Ali a , Najeeb Ullah b a b

Institute of Chemical Sciences, University of Peshawar, Pakistan University of Engineering and Technology Peshawar, Peshawar 25120, Khyber Pakhtunkhwa, Pakistan

a r t i c l e

i n f o

Article history: Received 15 May 2011 Received in revised form 25 October 2011 Accepted 28 October 2011 Available online 3 November 2011 Keywords: Silver SLM Membrane TDDA

a b s t r a c t Silver (I) has been transported from feed in to strip solution via tri-n-dodecylamine-cyclohexanepolypropylene supported liquid membrane (SLM). The transport of Ag+ has been found to be dependent on different parameters such as concentration of H+ in feed solution, tri-n-dodecylamine (TDDA) concentration in membrane phase, stripping phase composition and membrane thickness. The optimized conditions obtained for Ag+ transport are: 0.75 mol/dm3 of HNO3 in feed solution, 0.788 M of TDDA in cyclohexane in membrane phase and 1.0 mol/dm3 of NH3(aq) in stripping phase. The stoichiometry of the complex was calculated from the flux data of Ag+ transport across the membrane. The complex responsible for the transport of Ag+ has been investigated to be (LH)·Ag(NO3 )2 . Recovery of more than 98% of silver has been observed from feed solution at optimized conditions. Stability of this SLM system has been studied and found stable for 120 h. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Silver is one of the most ancient metals used in jewellery and arts. The various types of alloys used in dentistry contain silver, gold and copper with a trace amount of palladium and platinum. Silver has been used in the field of communication, aerospace, chemical industry, electroplating and photographic materials. Furthermore, silver has been reported to be used in catalysts, mirrors, cloud seeding, disinfection of water and some medicines [1–4]. It is estimated that about 12% of the world’s silver resources are used in the production of light sensitive devices. Silver can enter the environment through industrial water because it is often present as an impurity in Cu, Zn, As, and Sb ores. Being precious and toxic, silver must be recovered and separated from industrial effluents [5–7]. Different techniques have been used for the extraction of silver such as solvent extraction [8–10], adsorption [11,12], cloud point extraction [13], emulsion liquid membrane (ELM) [14,15], bulk liquid membrane (BLM) [16,17]. Traditional solvent extraction and other conventional methods may not be appropriate for large scale processes due to their major shortcomings such as time consuming, requirements of costly solvents and equipments and generation of toxic sludge [18,19]. Facilitated transport of cations, especially of metal ions using supported liquid membrane is one of the useful techniques in separation science [20,21]. Supported liquid membrane consists of hydrophobic polymeric microporous thin sheet which is

∗ Corresponding author. Tel.: +92 91 5641658; fax: +92 91 9216687. E-mail addresses: [email protected] (S.u. Rehman), gul [email protected] (G. Akhtar). 0376-7388/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2011.10.040

impregnated with hydrophobic organic carrier. In this method the extraction from donor phase and stripping in receiving takes place simultaneously [19]. There is a limited study for extraction of Ag(I) by SLM. Gherrou et al. [22] have studied the transport of Ag(I) from acidic thiourea solution via SLM containing D2EHPA as Ag(I) ion carrier and it has been found that Ag(I) species predominate at very low thiourea concentration (10−5 to 10−4 M). Shamsipur et al. [23] have reported the selective transport of Ag(I) through SLM composed of aza-thioether crown containing 1,10-phenanthroline as a specific ion carrier. It has been observed that the presence of thiosulfate as a metal ion acceptor in the strip solution, enhance the transport of Ag ion and almost all the silver was recovered after 3 h. Selective transport of Ag(I) and Hg(II) through two microporous supported membrane loaded with a mixed N/O/S donor macrocycle aza-thioether crown containing 1,10-phenanthroline sub unit and terathia has been reported by Shamsipur et al. [24]. Amiri et al. [25] have investigated selective transport of silver ion through supported liquid membrane using calix [4] pyrroles as a suitable ion carrier. Furthermore, it has been reported that various parameters such as carrier concentration in the membrane phase, thiosulfate concentration in strip phase, picric acid concentration in the feed phase, stirring speed and the type of solvent affect the transport of silver ion. Earlier workers have successfully extracted Ag(I) through SLM using triethanolamine as a carrier and KCN as a stripping agent [26]. However, the reported method has limitations since KCN in stripping phase is not environmental friendly and its disposal is one of the serious issues. In the present work we report a new supported liquid membrane system containing tri-n-dodeclyamine as a carrier dissolved in cyclohexane and immobilized in a thin polypropylene film. The

288

S.u. Rehman et al. / Journal of Membrane Science 389 (2012) 287–293

(LH)n.Ag(NO3)n+1

Ag+

If Ag stands for the distribution coefficient of Ag+ for distribution between the membrane and aqueous phase then,

[Ag(NH3)2]+

Ag =

NH3(aq)

nH+

[(LH)n · Ag(NO3 )n+1 ]org

(5)

[Ag+ ]aq

and Eq. (4) can be written as: (n+1)NO3-

H2O + (n+1)NO3-

nL

KAg =

+

Fig. 1. Schematic diagram of Ag transport.

Ag [NO− 3 ]aq

n+1

n

· [H+ ]aq [L]org n

(6)

and on re-arranging Eq. (6), we obtain NH3(aq) in stripping phase has been found to play significant role in the transport of Ag+ . The various process parameters such as acid concentration in feed solution, TDDA concentration in membrane phase, NH3(aq) concentration in strip phase etc. were first optimized for Ag+ transport. The optimized SLM was then used for recovery of Ag+ from silver plating and photographic waste solution. The stoichiometry of the chemical reactions as well as mechanism of Ag+ transport has also been evaluated. 2. Theory In the present study the supported liquid membrane is made of tri-n-dodecylamine in cyclohexane supported in microporous polymeric polypropylene film. Being hydrophilic in nature Ag+ from feed phase cannot enter the liquid membrane organic phase directly and hence the direct transport of silver ions through the membrane is not possible. Tri-n-dodecylamine may be indicated as L. The TDDA that is basic carrier and its pKa value is 11.3 [27]. The TDDA at the feed membrane interface may be protonated to LH+ in acidic medium as nitrogen atom of TDDA has lone pair of electrons. It is assumed that in the presence of HNO3 in feed solution the silver nitrate is converted to [Ag(NO3 )n+1 ]n− [26]. AgNO3

aq

+ nNO3

aq



 [Ag(NO3 )n+1 ]aq n−

(1)

Ag = KAg · [NO3 − ]aq

org

(2)

where the subscript org is the organic phase and aq is the aqueous phase. To show the contribution of H+ and NO3 − , Eq. (2) may be represented as: Agaq + (n + 1)(NO3 )aq − + nLorg + nHaq +  (LH)n · Ag(NO3 )n+1

org (3)

Ag+

The visualized transport mechanism of is shown in Fig. 1, in which H+ , NO3 − and Ag+ move in the same direction towards stripping phase. The complex formed as per Eq. (3) diffuses from feed membrane interface into liquid membrane phase and then it diffuses to membrane–strip interface. At membrane–strip interface the complex dissociates due to NH3(aq) in stripping phase and form [Ag(NH3 )2 ]+ [28]. The free carrier (L) diffuses back to feed membrane interface, and net result is the transport of Ag+ from feed to strip phase. According to Wilke-Chang [29], the diffusion coefficient of the backward transported free carrier (L) molecules should be much greater as compared to the diffusion coefficient of the forward transported complex (LH)n ·Ag(NO3 )n+1 . Due to this reason the concentration of TDDA at feed membrane interface will always be higher as compared to complex. The equilibrium constant of Eq. (3) can be written as under. KAg =

[(LH)n · Ag(NO3 )n+1 ]org +

[Ag ]aq · [NO3 − ]aq

n+1

n

· [H+ ]aq · [L]org n

(4)

n

· [H+ ]aq · [L]org n

(7)

In the above equations the brackets indicate the concentration. According to the Fick’s law, the rate of diffusion of a solute dN/dt (where N is amount of substance and t stands for time), across an area A is known as diffusion flux and given by the symbol J. J=

1 dN A dt

(8)

J is proportional to the concentration gradient dc/dx. Since the concentration gradient is negative in the direction of flux i.e. −dc/dx, thus J = −D

dc dx

(9)

where J is the flux of species under transport and dc is the difference in concentration of that species across a very small segment of the membrane thickness dx and D is the diffusion coefficient. It is supposed that the x-axis is perpendicular to the face boundaries of the membrane. Then, at the feed side boundary of the membrane x = 0, at the strip side boundary of the membrane x = , and Cfm and Csm are the membrane phase concentrations of Ag+ at feed and strip side respectively. Assuming a linear concentration gradient for Ag+ within the membrane one can write: J∝

The species LH+ and [Ag(NO3 )n+1 ]n− then react at the feed membrane interface and form the complex as shown below. n− nLH+  (LH)n · Ag(NO3 )n+1 org + [Ag(NO3 )n+1 ]aq

n+1

Cfm − Csm 

(10)

Hence Cfm − Csm dc =− dx 

(11)

Eq. (9) then becomes, J=D

C

fm

− Csm 



(12)

where, as mentioned earlier, D is the diffusion coefficient. Since distribution coefficient of Ag+ at the membrane surface in the feed side and strip side can be given by f and s respectively.i.e. f =

Cfm Cf

and

s =

Csm Cs

(13)

where Cf and Cs represent bulk feed and strip concentration of Ag+ respectively. From relation (13) we obtain Cfm = f · Cf

and

Csm = s · Cs

(14)

and with the help of Eq. (14), Eq. (12) becomes J=D

f · Cf − s · Cs 

(15)

As there is no extraction from the strip to the membrane phase, so s → 0 and as a result, s Cs ≈ 0 and Eq. (15) transforms to J=

D · f · Cf 

(16)

S.u. Rehman et al. / Journal of Membrane Science 389 (2012) 287–293

Here in our case f = Ag as given by Eq. (7), thus Eq. (16) becomes D · KAg · [NO3 − ]aq

J=

n+1

n

· [H+ ]aq · [L]org n Cf



(17)

According to the Wilke-Chang relation [30]. D=

k · T 

k · T · KAg · [NO3 − ]aq

n+1

n

· [H+ ]aq · [L]org n · Cf

·

(19)

Since k and KAg is constant equal to Y, a new constant. Therefore, J=

Y · T · [NO3 − ]aq

n+1

n

· [H+ ]aq · [L]org n · Cf

·

(20)

However, if Cf does not change rapidly, then for a short interval of time, Cf may be taken nearly as constant. Taking log of Eq. (20) log J =

log Y + log T + (n + 1) log[NO3 − ]aq + n log[H + ]aq + n log[L]org + log Cf 

(21)

As Y, T, and the thickness  of the membrane is constant, and then Eq. (21) becomes

log J =

Table 1 Parameters of microporous polymeric support. Membrane

Thickness (␮m)

Porosity (%)

Material

Celgard 2500 Celgard 2502 Celgard 4400

25 50 175

45 45 68

Polypropylene Polypropylene Polypropylene

(18)

where T is the absolute temperature,  is viscosity and k is constant and so J=

289

constant + (n + 1) log[(NO3 )

n−

3.2. Membranes Three types of Celgard polypropylene hydrophobic membranes 2500, 2502 and 4400 were used as a solid support for liquid membrane. The specifications of membranes are given in Table 1. The supported liquid membranes were prepared by soaking the support in a predetermined concentration of TDDA in cyclohexane for 24 h. The membrane was taken out from the organic solution and allowed to drain off for 5 min to remove excess amount of carrier and diluent.

3.3. Apparatus/instruments The atomic absorption spectrometer Perkin Elmer model 400 was used for determination of metal ions concentration in feed and strip solutions. pH meter Metrohm model 827 was used for measurement of pH. Brookfield viscometer/rheometer LVDV-III was used for viscosity measurement of TDDA in cyclohexane. The sodium and potassium metal ions were analyzed with industrial flame photometer JENWAY, model PFP7 UK.

3.4. Membrane cell

]aq + n log[H+ ]aq + n log[L]org + log Cf 

(22) Eq. (22) can be used to determine the number of H+ associated with L in the form LH+ . This can be done in different ways, one similar way is to keep [NO3 − ], [L],  and Cf constant and plotting log J versus log [H+ ], the slope of the curve will give the “n” value for H+ . Similarly by keeping [NO3 − ], [H+ ],  and Cf constant, the slope of log J versus log [L] will give number of moles of tri-ndodecylamine taking part in complexation.

All the metal ion permeation experiments were performed in a two compartment cell as shown in our previous study [30]. Each compartment of membrane cell had a volume capacity of 250 cm3 . The effective membrane area was 23.75 cm2 . The solution in both compartments of the cell was stirred at 1500 rpm by synchronous motors at a temperature of 25 ◦ C ± 1 ◦ C. The stirring speed was optimized for similar type of carrier group [30] and cell, hence all study was performed at this stirring speed (1500 rpm).

3.5. Permeation study 3. Experimental 3.1. Chemicals and reagents The silver nitrate (Scharlau 99.99%) in HNO3 (Fisher Scientific 69–72%) was used as feed in given concentration. The liquid membrane composed of TDDA (Merck ≥95%) as a metal ion carrier in cyclohexane (Panreac 99.5%) as a diluent to get various composition of liquid membrane phase. The stripping solutions were made of NH3(aq) (35% BDH) in given concentration. All experiments were carried out in double distilled deionized water. All other chemicals were used of analytical grade. Flux =

The feed and strip solutions, 250 cm3 each, were added to both compartments of the cell simultaneously. Both compartments of the cell were then covered and agitated continuously to avoid concentration polarization at membrane faces. Samples of 1.0 cm3 from feed and strip solution were drawn after regular time intervals and analyzed for metal ion concentration. The experimental conditions for most of the experiments were: initial Ag+ concentration in feed solution was 7.42 × 10−4 mol/dm3 , the diluent was cyclohexane and the membrane was celgard 2500 (except for membrane thickness study on transport of Ag+ ). The flux was calculated using relation [30]:

concentration change of Ag+ ion(mol/dm3 ) × solution volume in feed or strip (dm3 ) effective membrane area (m2 ) × t

(23)

and permeability was calculated using following formula [31]: P=

[ln (initial feed concentration/feed concentration at time(t))] × volume in feed phase effective membrane area × t where t indicates time interval in seconds.

(24)

290

S.u. Rehman et al. / Journal of Membrane Science 389 (2012) 287–293 120

6 0.158M TDDA

100

5

Viscosity(cP)

0.315M TDDA

80

0.630M TDDA 0.788M TDDA

60

0.946M TDDA 1.103M TDDA

40

4 3 2 1

1.261M TDDA

20

0 0

0 0

50

100

150

200

250

300

350

0.2

4. Results and discussion 4.1. Effect of TDDA concentration To study the effect of concentration of TDDA on transport of Ag+ various concentration of TDDA ranging from 0.158 M to 1.261 M were used. During this study the concentration of AgNO3 was kept at 7.42 × 10−4 mol/dm3 in 0.75 mol/dm3 of HNO3 in feed solution and concentration of NH3(aq) was fixed at 1.0 mol/dm3 in strip solution. Fig. 2 indicates the effect of carrier concentration on extraction of Ag+ in stripping phase with respect to time. It is clear from this figure that extraction of Ag+ increases from 46.49% to 98.65% as the concentration of TDDA increases from 0.158 M to 0.788 M in membrane phase. The extraction of Ag+ decreases beyond 0.788 M of TDDA. Similar trend of increase and then decrease for permeability (p) and flux (J) of Ag+ is also observed for various concentration of TDDA as shown in Fig. 3. As expected and as per Eq. (22), that as concentration of TDDA increases in membrane phase, the flux of Ag+ also increases as more numbers of TDDA interact with H+ to form LH+ , which on further reaction with anions ([Ag(NO3 )n+1 ]n− ) enhance the formation of complex. The increase in metal ion transport with increasing carrier concentration is available in literature for similar type of carrier [31,32]. The decrease in transport of Ag+ beyond 0.788 M of TDDA could be explained by an increase in the viscosity of the liquid membrane phase. As the concentration of TDDA in cyclohexane increases, viscosity of liquid membrane also increases. The viscosity of various composition of TDDA is shown in Fig. 4. The diffusion coefficient of

Permeability (p)

J (10

5.2

7

4.2

6

3.2

5 4

2.2

3

1.2

2

p (10-6 m/s)

mol/m .s)

2 -10

6.2

Flux (J)

0.2

1 0

-0.8 0

0.2

0.4

0.6

1

1.2

the solute across the liquid membrane is shown by the following Stokes–Einstein equation [33]: D=

8

0.8

Fig. 4. Viscosity of TDDA–cyclohexane solutions.

Fig. 2. Effect of TDDA concentration on recovery of Ag+ in stripping phase with time (HNO3 concentration in feed solution = 0.75 mol/dm3 , TDDA concentration in membrane phase = 0.158 M–1.261 M, NH3(aq) concentration in stripping solution = 1.0 mol/dm3 ).

9

0.6

[TDDA] mol/dm 3

Time (min)

10

0.4

400

0.8

1

1.2

1.4

TDDA (mol/dm 3)

KT 6r

(25)

where T is the absolute temperature, K is the Boltzmann constant, r is the ionic radius of solute and  is the viscosity of the organic phase in cP, equilibrated with the aqueous phase. Since viscosity is inversely proportional to diffusity, that is why diffusion coefficient of complex (LH)n ·Ag(NO3 )n+1 in membrane phase decreases, and results in decrease in extraction of Ag+ . The decrease in metal ion transport with increasing viscosity of liquid membrane phase has already been observed in our previous study for Mn(II), W(VI) and Pd(II) transport [30,32,34]. Hence 0.788 M of TDDA was considered to be the optimum concentration for extraction of Ag+ for subsequent study to optimize various parameters. The number (n) of TDDA(L) involved in complex of (LH)n ·Ag(NO3 )n+1 can be determined by plotting log [TDDA] versus log J as shown in Fig. 5. The slope of the graph calculated 1.0023, this means that one molecule of TDDA is involved in the complex formation. 4.2. Effect of HNO3 concentration HNO3 plays a significant role in extraction of Ag+ , because it provide H+ and NO3 − for the formation of LH+ and [Ag(NO3 )n+1 ]n− respectively. During this study the concentration of HNO3 in feed solution was varied from 0.25 mol/dm3 to 1.25 mol/dm3 , while concentration of TDDA in membrane phase and that of NH3(aq) in stripping phase was fixed at 0.788 M and 1.0 mol/dm3 respectively. Fig. 6 shows that the flux of Ag+ increases from 4.682 × 10−10 mol/m2 s to 8.901 × 10−10 mol/m2 s with increase in the concentration of HNO3 from 0.25 mol/dm3 to 0.75 mol/dm3 and becomes maximum at 0.75 mol/dm3 of HNO3 . This behavior is in accordance with Eq. (22) where flux is directly related to [H+ ]. Further increase in concentration of HNO3 beyond 0.75 mol/dm3 results in decrease in the flux of Ag+ . This can be attributed to the formation of Hn Ag(NO3 )n+1 instead of (LH)n ·Ag(NO3 )n+1 due to large number of H+ and NO3 − , and Eq. (1) is inhibited in forward direction. Thus 0.75 mol/dm3 of HNO3 in feed solution was considered log [TDDA] -0.9

-0.8

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-8.6 -0.1 -8.7 0

0.1

-8.8 y = 1.0023x - 8.7182 R = 0.9978

-8.9 -9 -9.1

log J η

[Ag] (%age)

0.473M TDDA

-9.2 -9.3 -9.4 -9.5 -9.6

Fig. 3. Flux (J) and permeability (p) variation at various concentrations of TDDA in membrane phase (Same operating conditions as given in Fig. 2).

Fig. 5. Plot of log [TDDA] versus log J (same operating conditions as given in Fig. 2).

S.u. Rehman et al. / Journal of Membrane Science 389 (2012) 287–293 10

Table 2 Effect of membrane thickness on flux of Ag+ .

9 8 J (10-10 mol/m 2.s)

291

7

Membranes

Flux of Ag+ (×10−10 mol/m2 s)

6

2500 2502 4400

8.901 8.053 7.231

5 4 3

Conditions: (HNO3 concentration in feed solution = 0.75 mol/dm3 , TDDA concentration in membrane phase = 0.788 M, NH3(aq) concentration in stripping solution = 1.0 mol/dm3 , time = 4.0 h).

2 1 0 0

0.2

0.4

0.6

0.8

1

1.2

1.4

HNO 3 (mol/dm3 )

Fig. 6. Effect of HNO3 concentration in feed solution on flux (J) of Ag+ (HNO3 concentration in feed solution = 0.25–1.25 mol/dm3 , TDDA concentration in membrane phase = 0.788 M, NH3(aq) concentration in stripping solution = 1.0 mol/dm3 , time = 4.0 h).

log [HNO3] -8.5 -0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

-8.7

y = 1.0013x - 8.9088 R = 0.9986

-9.1

log Jη

-8.9

absence of strippant no transport of Ag+ was observed even after conducting the experiment for long time (8 h). This indicates that strippant plays a vital role in transport of Ag+ . It is seen from Fig. 9 that as the concentration of NH3(aq) increases the permeability of Ag+ also increases and becomes maximum at 1.0 mol/dm3 of NH3(aq) . The extraction phenomena with NH3(aq) can be explained by the following two reasons. Firstly, the OH− of NH3(aq) can react with the H+ of the complex (LH)n ·Ag(NO3 )n+1 which was formed as according to Eq. (3) and it results in the decomposition of complex in the reverse way as shown below. (LH)n · Ag(NO3 )n+1 + OH− → nL + [Ag(NO3 )n+1 ]n− + H2 O

-9.3

(26)

Secondly the free NH3 of NH3(aq) can react with [Ag(NO3 )n+1 ]n− and can form soluble diammine–argentate complex ion [28,35].

-9.5 -9.7

[Ag(NO3 )n+1 ]

Fig. 7. Plot of log [HNO3 ] versus log J (same operating conditions as given in Fig. 6).

as the optimum concentration for transport of Ag+ for this SLM system. Eq. (22) was used to determine the number of H+ in the proposed complex of (LH)n ·Ag(NO3 )n+1 by plotting log [HNO3 ] versus log J as shown in Fig. 7. The slope calculated 1.0013, indicates that one mole of H+ is involved in complex formation. To study the transport of H+ , pH of the feed solution was measured after regular interval of times as shown in Fig. 8. Increase in pH of the feed solution was observed, this shows that protons of feed solution are being utilized by carrier (TDDA) molecules to form LH+ species at feed membrane interface. An experiment was conducted to show the transport of H+ in absence of Ag(NO3 )aq in feed solution at optimum experimental conditions. No change in pH of feed and strip solution was observed, indicates no transfer of H+ towards the stripping phase. 4.3. Effect of stripping phase concentration NH3(aq) was used as strippant with different concentration ranging from 0.25 mol/dm3 to 1.5 mol/dm3 whereas the concentration of HNO3 was kept at 0.75 mol/dm3 in feed solution and TDDA concentration at 0.788 M in membrane phase. It was found that in

n−

+

+ 2NH3  Ag(NH3 )2 + (n + 1)NO3 −

Kf = 2.5 × 107

(27)

To investigate the transport of H+ towards strip solution, the pH of the strip solution was measured. The pH of the strip solution at the start of the experiment at optimum conditions was 10.55, while after 4 h the pH decreased to 9.08. This decrease in pH may be due to that OH− of NH3(aq) are consumed in the neutralization of H+ of the complex (LH)n ·Ag (NO3 )n+1 as per Eq. (26). 4.4. Effect of Ag+ concentration in feed solution To investigate the effect of Ag+ concentration on transport of Ag+ , concentration range of 3.71 × 10−4 mol/dm3 to 22.26 × 10−4 of Ag+ was used in feed solution. During this study the concentration of HNO3 in feed solution was adjusted at 0.75 mol/dm3 , NH3(aq) in stripping solution at 1.0 mol/dm3 and TDDA in membrane phase at 0.788 M. As shown in Fig. 10, that flux of Ag+ increases from 4. 463 × 10−10 mol/m2 s to 26.788 × 10−10 mol/m2 s as the concentration of Ag+ increases from 3.71× 10−4 mol/dm3 to 22.26× 10−4 mol/dm3 in feed solution. This behavior is in accordance with Eq. (22), where flux is directly proportional to feed concentration (Cf ). No metal loading behavior for carrier up to the concentration of 22.26 × 10−4 mol/dm3 of Ag+ in feed solution was observed. This phenomenon is unlike our earlier observation [30] with Triethanolamine SLM, where transport of Mn(VII) was limited due to carrier loading effect. 6

2.5

5 p (10 m/s)

2

pH

1.5 1

4 3 2 1

0.5

0

0 0

50

100

150

200

250

300

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

3

Time (min)

Fig. 8. pH variation in the feed solution (HNO3 concentration in feed solution = 0.75 mol/dm3 , TDDA concentration in membrane phase = 0.788 M, NH3(aq) concentration in stripping solution = 1.0 mol/dm3 ).

(NH 3(aq) mol/dm )

Fig. 9. Effect of NH3(aq) concentration on permeability of Ag+ (HNO3 concentration in feed solution = 0.5 mol/dm3 , TDDA concentration in membrane phase = 0.788 M, NH3(aq) concentration in stripping solution = 0.25–1.5 mol/dm3 , time = 4.0 h).

292

S.u. Rehman et al. / Journal of Membrane Science 389 (2012) 287–293

Table 3 Composition of silver plating and photographic waste solution. Sample

Metal ions

[M] in feed solution (×10−4 mol/dm3 )

[M] in strip solution (×10−4 mol/dm3 )

Silver plating

Ag Cr Zn Cu Ag K Mg Na Fe

4.339 Nil Nil Nil 2.887 23.75 1.22 21.65 3.345

4.335 Nil Nil Nil 2.885 Nil Nil Nil Nil

Photographic waste

30

mol/m . s)

20

J (10

25

10

15

5 0 0

5

10

15 [Ag ]x10

20

25

mol/dm

Fig. 10. Effect of Ag+ concentration in feed solution on its flux (initial Ag+ concentration in feed = 3.71 × 10−4 mol/dm3 to 22.26 × 10−4 mol/dm3 , HNO3 concentration in feed solution = 0.75 mol/dm3 , TDDA concentration in membrane phase = 0.788 M, time = 4.0 h).

reasons are: shear induced emulsion formation in membrane phase, Pore blocking of membrane by precipitation of carrier complex, solubility of carrier or solvent in aqueous phase, osmotic pressure and wettability of support pores by aqueous phase [38]. To show the long term stability of this SLM, it was impregnated once in 0.788 M of TDDA in cyclohexane and was used continuously for 5 runs. Each run was of 4 h duration and time between successive run was of 24 h. Fig. 11 shows that permeability is consistent for continuous operation of 120 h and recovery of Ag+ for each run was more than 98%. The study replicated twice with relative standard deviation ± 1%. This study indicates that this SLM configuration is quite stable and can be scaled up at industrial level. 6. Recovery of silver from silver plating and photographic waste solution

10 9

J (10

mol/m .s)

8 7 6 5 4 3 2 1 0 0

1

2

3

4

5

6

No. of runs

Fig. 11. Stability of SLM, number of runs versus permeability (HNO3 concentration in feed solution = 0.75 mol/dm3 , TDDA concentration in membrane phase = 0.788 M, NH3(aq) concentration in stripping solution = 1.0 mol/dm3 ).

The tri-n-dodecylamine SLM has the most effective transport ability for Ag+ . To show practical utilization of this SLM system, it was used for recovery of Ag+ from silver plating and photographic waste solution. The conditions used for extraction of Ag+ were: 0.75 mol/dm3 of HNO3 in feed solution, 0.788 M of TDDA in membrane phase and 1.0 mol/dm3 of NH3(aq) in stripping solution. Table 3 shows almost complete recovery of Ag+ from silver plating waste solution. To confirm further application of this technique, the optimized SLM was used for recovery of Ag+ from photographic waste solution. The results indicate that only silver is transported which show the selectivity and efficiency of this method for silver recovery. The composition of silver plating and photographic waste solution is shown in Table 3.

4.5. Effect of membrane thickness on transport of Ag+ 7. Conclusions To investigate the effect of membrane thickness on transport of Ag+ , three microporous hydrophobic membranes of celgard 2500, 2502 and 4400 were used. The specifications of the membranes are given in Table 1. It can be seen from Table 2 that the flux of Ag+ decreases as thickness of membrane film increases. The celgard 4400 membrane with the highest thickness shows the lowest flux of 7.231 × 10−10 mol/m2 s. These results are in accordance with the proposed theoretical model shown in Eq. (22), where the flux is inversely proportional to membrane thickness. This decrease in Ag+ transport with membrane thickness is comparable to those reported earlier for supported liquid membranes [22,36,37]. 5. Stability of SLM The supported liquid membrane is one of the promising techniques for separation and recovery of metal ions; nevertheless their applications were limited for two decades due to instability problems. The main degradation or instability

(1) Ag+ can be transported through tri-n-dodecylamine flat sheet supported liquid membrane. (2) The transport of Ag+ was found to be co-ions coupling transport mechanism, as H+ , NO3 − and Ag+ move in same direction. (3) One mole of TDDA and one mole of H+ associate with one mole of Ag+ forming a complex (LH)·Ag(NO3 )2 responsible for transport of Ag+ . (4) The optimum conditions found for this SLM system are: 0.75 mol/dm3 of HNO3 in feed solution, 0.788 M of TDDA in membrane phase and 1.0 mol/dm3 of NH3(aq) in stripping phase. (5) This SLM system was found quite stable for 120 h. (6) Almost all Ag+ have been removed when this SLM system was applied to silver plating and photographic waste solution, which concludes that this technique can be used for recovery of Ag+ from industrial wastes effluent.

S.u. Rehman et al. / Journal of Membrane Science 389 (2012) 287–293

Nomenclature TDDA, L D2EHPA SLM LM ELM BLM [M] Cf Cs Cfm Csm D J p n A t T K r KAg 

tri-n-dodecylamine bis-(2-ethylhexyl) phosphoric acid supported liquid membrane liquid membrane emulsion liquid membrane bulk liquid membrane metal ion concentration concentration of Ag+ in the bulk feed (mol/dm3 ) concentration of Ag+ in the stripping solution (mol/dm3 ) concentration of Ag+ in the membrane phase on feed side of membrane concentration of Ag+ in the membrane phase on strip side of membrane diffusion coefficient (m2 s−1 ) flux (mol/m2 s) permeability (m/s) number of moles of TOA or H+ associated in the complex area of the membrane time (s) absolute temperature (K) boltzmann constant ionic radius of solute equilibrium constant of silver membrane thickness

Greek symbols viscosity (cp)  Ag distribution coefficient of silver distribution coefficient of metal ions into membrane f from feed solution s distribution coefficient of metal ions into membrane from strip solution Subscripts aq. aqueous organic org. m membrane f feed s strip

References [1] P.N.J. Dennis, Photodetectors, Plenum Press, New York, 1986. [2] X. Wang, L. Zhang, C. Ma, R. Song, H. Hou, D. Li, Enrichment and separation of silver from waste solutions by metal ion imprinted membrane, Hydrometallurgy 100 (2009) 82–86. [3] S. Song, C. Ji, M. Wang, C. Wang, C. Sun, R. Qu, C. Wang, H. Chen, Adsorption of silver (I) from aqueous solution by chelating resins with 3-aminopyridine and hydrophilic spacer arms: equilibrium kinetic, thermodynamic and mechanism studies, J. Chem. Eng. Data 56 (2011) 1001–1008. [4] R.J. White, An historical overview of the use of silver in wound management, Br. J. Nurs. 10 (2) (2001) 3–8. [5] E. Main (Ed.), Metals and their Compounds in Environment, VCH, New York, 1991. [6] D.E. Kimbrough, P.W. Wong, J.K. Biscoe, A critical review of photographic and radiographic silver recycling, J. Solid Waste Technol. Manage. 23 (1996) 197–207. [7] F.Z.E. Aamrani, A. Kumar, L. Beyer, A. Florido, A.M. Sastre, Mechanistic study of active transport of silver (I) using sulfur containing novel carriers across a liquid membrane, J. Membr. Sci. 152 (1999) 263. [8] K.C. Sole, T.L. Ferguson, J.B. Hiskey, Solvent extraction of silver by cyanex 272, cyanex 302 and cyanex 301, Solvent Extr. Ion Exch. 12 (1994) 1033. [9] D. Sevdic, H. Meider, D. Parkovic, Solvent extraction of silver and mercury by O-dibutyl phenylamino-phenylmethanethio phosphonate from chloride solutions, Solvent Extr. Ion Exch. 8 (1990) 643.

293

[10] V. Stankovic, L. Outarra, F. Zonnevijlle, C. Comninellis, Solvent extraction of silver from nitric acid solutions by calix [4] arene amide derivatives, Sep. Purif. Technol. 61 (3) (2008) 366–374. [11] M.A. El-Ghaffar, M.H. Mohamed, K.Z. Elwakeel, Adsorption of silver (I) on synthetic chelating polymer derived from 3-amino-1,2,4-triazole-5-thiol and glutaraldehyde, Chem. Eng. J. 151 (1-3) (2009) 30–38. [12] M. Tuzen, M. Soylak, Column solid phase extraction of nickel and silver in environmental samples prior to their flame atomic absorption spectrometric determinations, J. Hazard. Mater. 164 (2–3) (2009) 1428–1432. [13] M. Ghaedi, A. Shokrollahi, K. Niknam, E. Niknam, A. Najibi, M. Soylak, Cloud point extraction and flame atomic absorption spectrometric determination of cadmium (II), lead (II), palladium (II) and silver (I) in environmental samples, J. Hazard. Mater. 168 (2009) 1022–1027. [14] S.C. Lee, Y.S. Mok, W.K. Lee, Application of emulsion liquid membrane technology to the extraction of silver, J. Ind. Eng. Chem. 2 (1996) 130–136. [15] N. Othman, H. Mat, M. Goto, Separation of silver from photographic wastes by emulsion liquid membrane system, J. Membr. Sci. 282 (1–2) (2006) 171–177. [16] G.H. Rounaghi, M.S. Kazemi, H. Sadeghian, Transport of silver ion through bulk liquid membrane using macrocyclic and acyclic ligands as carriers in organic solvents, J. Incl. Phenom. Macrocycl. Chem. 60 (2008) 79–83. [17] K. Farhadi, S. Bahar, R. Maleki, Separation study of silver (I) ion through a bulk liquid membrane containing meloxican, J. Braz. Chem. Soc. 18 (3) (2007) 595–600. [18] N.M. Kocherginsky, Q. Yang, Big carrousel mechanism of copper removal from ammonical wastewater through supported liquid membrane, Sep. Purif. Technol. 54 (2007) 104–116. [19] J.D. Gyves, E.R. San-Miguel, Metal ion separations by supported liquid membranes, Ind. Eng. Chem. Res. 38 (1999) 2182–2202. [20] N. Bukhari, M.A. Chaudry, M. Mazhar, Triethanolamine-cyclohexanone supported liquid membranes study for extraction and removal of nickel ions from nickel plating wastes, J. Membr. Sci. 283 (2006) 82–189. [21] B. Swain, J. Jeong, J.C. Lee, G.H. Lee, Separation of Co (II) and Li (I) by supported liquid membrane using cyanex 272 as mobile carrier, J. Membr. Sci. 297 (2007) 253–261. [22] A. Gherrou, H. Kerdjoudj, R. Molinari, E. Drioli, Removal of silver and copper ions from acidic thiourea solutions with a supported liquid membrane containing D2EHPA as carrier, Sep. Purif. Technol. 28 (2002) 235–244. [23] M. Shamsipur, S.Y. Kazemi, G. Azimi, S.S. Madaeni, V. Lippolis, A. Garau, F. Isaia, Selective transport of silver ion through a supported liquid membrane using some mixed aza-thioether crowns containing a 1,10-phenanthroline sub-unit as specific ion carriers, J. Membr. Sci. 215 (2003) 87–93. [24] M. Shamsipur, O.R. Hashemi, V. Lippolis, A supported liquid membrane system for simultaneous separation of silver (I) and mercury (II) from dilute feed solutions, J. Membr. Sci. 282 (2006) 322–327. [25] A.A. Amiri, A. Safavi, A.R. Hasaninejad, H. Shrghi, M. Shamsipur, Highly selective transport of silver ion through a supported liquid membrane using calix [4] pyrroles as suitable ion carriers, J. Membr. Sci. 325 (2008) 295–300. [26] M.A. Chaudry, N. Bukhari, M. Mazhar, Coupled transport of Ag (I) ions through triethanolamine-cyclohexanone-based supported liquid membranes, J. Membr. Sci. 320 (2008) 93–100. [27] Z. Pawelka, T.Z. Huyskens, Alkyl substituent effect on the polarity of phenols–tri-n-alkylamine complexes, Can. J. Chem. 81 (2003) 1012–1018. [28] G. Svehla, Vogel’s Qualitative Inorganic Analysis, seventh ed., Pearson Education Publisher, 2007, pp. 83–84. [29] S.A. Ansari, P.K. Mohapatra, V.K. Manchanda, Recovery of actinides and lanthanides from high level waste using hollow- fiber supported liquid membrane with TODGA as the carrier, Ind. Eng. Chem. Res. 48 (2009) 8605–8612. [30] S. Rehman, G. Akhtar, M.A. Chaudry, N. Bukhari, Najebullah, N. Ali, Mn (VII) ions transport by triethanolamine cyclohexanone based supported liquid membrane and recovery of Mn (II) ions from discharged zinc carbon dry battery cell, J. Membr. Sci. 366 (2011) 125–131. [31] K. Chakrabarty, P. Saha, A.K. Ghoshal, Simultaneous separation of mercury and lignosulfonate from aqueous solution using supported liquid membrane, J. Membr. Sci. 346 (2010) 37–44. [32] M.A. Chaudry, M.T. Malik, M. Ahmed, Supported liquid membrane extraction of W(VI) ions using tri-n-octylamine as carrier, J. Radioanal. Nucl. Chem. 150 (2) (1991) 383–396. [33] S. Panja, R. Ruhela, S.K. Misra, J.N. Sharma, S.C. Tripathi, A. Dakshinamoorthy, Facilitated transport of Am(III) through a flat-sheet supported liquid membrane. (FSSLM) containing tetra(2-ethyl hexyl) diglycolamide (TEHDGA) as carrier, J. Membr. Sci. 325 (2008) 158–165. [34] M.A. Chaudry, N.U. Islam, N.U. Rehman, Extraction of Pd (II) ions using tri-noctylamine xylene base supported liquid membranes, J. Radioanal. Nucl. Chem. 218 (1) (1997) 53–60. [35] D. Gary, Christan, Analytical Chemistry, sixth ed., John Wiley & Sons, 2004, pp. 295–296. [36] B. Swain, J. Jeong, J.C. Lee, G.H. Lee, Extraction of Co (II) by supported liquid membrane and solvent extraction using Cyanex 272 as an extractant: a comparison study, J. Membr. Sci. 288 (2007) 139–148. [37] S. Sriram, P.K. Mohapatra, A.K. Pandey, V.K. Manchanda, L.P. Badheka, Facilitated transport of americium(III) from nitric acid media using dimethyldibutyltetradecyl-1,3-malonamide, J. Membr. Sci. 177 (2000) 163–175. [38] N.M. Kocherginsky, Q. Yang, L. Seelam, Recent advances in supported liquid membrane technology, Sep. Purif. Technol. 53 (2007) 171–177.