Hydrometallurgy 99 (2009) 25–32

Contents lists available at ScienceDirect

Hydrometallurgy
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / h yd r o m e t

Extraction of zinc and copper–zinc mixtures from ammoniacal solutions into

emulsion liquid membranes using LIX 84I®
Bina Sengupta, Mathurkumar S. Bhakhar, Ranjan Sengupta ⁎
Chemical Engineering Department, The Maharaja Sayajirao University of Baroda, Vadodara, 390 001, India

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

Article history: Emulsion liquid membranes with LIX 84I as extractant were used to extract zinc and copper–zinc mixtures
Received 16 March 2009 from ammoniacal media. The effects of various emulsion parameters and process parameters on zinc
Received in revised form 29 May 2009 extraction were investigated. The optimal pH was 8.1. The stripping rates were found to be slower than the
Accepted 29 May 2009
loading rates in ELMs. The extraction and stripping data using LIX 84I were obtained for Zn and Cu–Zn
Available online 12 June 2009
mixtures to discern the influence of individual steps on ELM extraction. A synergistic effect was observed for
loading and stripping zinc in the presence of copper. Zinc stripped faster than copper when the zinc
Keywords:
Emulsion liquid membrane concentration was higher or comparable with copper. Separation factors for Cu/Zn were found to be in the
Copper range of 1.1–1.4. When Cu–Zn mixtures were extracted in ELMs, it was observed that the whole extraction
Zinc process was governed by the loading of copper and zinc in the membrane phase and other considerations
Lix84I such as stripping, diffusion etc. were completely masked by the rapid loading rates of copper in the ELMs. The
Ammonia separation factor βCu/Zn was found to be a maximum of 3.5 indicating the possibility of selective separation of
Extraction copper from Cu–Zn mixtures using ELMs.
Stripping © 2009 Elsevier B.V. All rights reserved.
Separation factor

1. Introduction Out of the various techniques available, such as precipitation,

Extraction of zinc and copper from acidic leach liquors using
acidic/chelating extractants is well established and industrially
exploited for decades. Ammoniacal leaching is also practiced for
recovery of metals such as nickel, copper, zinc etc. from oxide and
sulfide ores. It is particularly an attractive proposition to use
ammoniacal leaching when ore bodies are associated with large
amounts of acid consuming constituents such as calcite and magnesia
(Gupta and Mukherjee, 1990). Ammonia leaching processes have
found application for the recovery of copper from its wide ranging
secondary resources such as flotation tailings, metal scrap, brass scrap,
furnace drosses etc. (Flett, 1979). In recent years there is considerable
interest to recover metals such as copper, zinc, nickel, lead etc. from
the sludge of printed circuit board industry using ammoniacal
leaching to meet the toxicity characteristic leaching procedure
regulation limit. Copper and zinc appear simultaneously in the
ammoniacal wastewaters of printed circuit board manufacturing
units. The copper concentrations in these waters usually range
between 300 and 3000 mg/L while zinc concentrations range
between 300 and 1700 mg/L (Yang et al., 2006; Chen and Huang,
2007).
⁎ Corresponding author. Fax: +91 0265 2423898.
E-mail addresses: bina_msu@yahoo.co.in (B. Sengupta), ranjan_msu@yahoo.co.in
(R. Sengupta).
adsorption and solvent extraction, for the removal of metals from
ammoniacal leach solutions, solvent extraction appears to be the most
attractive. This is because it not only offers distinct advantages of ease
and flexibility of operations, ability to handle wide range of
concentrations and control over the selectivity of the separations,
but also because it is a clean technique with the absence of sludge. In
spite of these advantages, solvent extraction becomes uneconomical
for treating leach solutions having low metal concentrations due to
the large inventory of extractant, solvent and strip reagent, solvent
losses, solvent carry over etc. These drawbacks can be circumvented
using the emulsion liquid membrane technique (ELMs).
Since its invention over forty years ago, emulsion liquid mem-
branes and their variant supported liquid membranes have been used
to separate a variety of solutes from aqueous as well as organic
streams. Alguacil (2001) studied the permeation of copper from
ammoniacal–ammonium carbonate solutions using supported liquid
membranes containing LIX 973N as extractant. Gameiro et al. (2007)
reported copper extraction from ammoniacal media using ELMs with
LIX 54 as extractant and in our earlier studies (Sengupta et al., 2007)
copper was extracted from ammoniacal solutions using LIX 84I as
extractant. These investigations found that the rate of extraction was
very fast, reaching quantitative removal in 2 min of contact between
feed and emulsion. It was concluded that the loading of copper in the
membrane phase governed the rate of extraction, whilst other
physico-chemical factors such as stripping acid concentration, emul-
sion drop size distribution etc. became important only when

0304-386X/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.hydromet.2009.05.021
26 B. Sengupta et al. / Hydrometallurgy 99 (2009) 25–32

Table 1 as the membrane material. Span 80 (Sorbitan mono-oleate), supplied

Range of parametric variations studied for zinc extraction (ELMs). by S.D.Fine Chemicals, was the emulsifier. The carrier, kerosene and
Variants pH ϕ Cc (% v/v) Cio (M) TR Ceo (mg/L) the emulsifier were used directly as received from the manufacturer.
pH 7, 8.1, 8.6 0.5 10 1.5 1:6 1000 All other chemicals used were of AR grade.
ϕ 8.1 0.5, 0.64 10 1.5 1:6 1000
Cc 8.1 0.5 5, 10 1.5 1:6 1000
Cio 8.1 0.5 10 0.5, 1, 1.5 1:6 1000 3. Experimental
TR 8.1 0.5 10 1.5 1:6, 1:10, 1:15 1000
Ceo 8.1 0.5 10 1.5 1:6 280,490, Ammoniacal feed solution for zinc extraction was prepared by
1000,2000
dissolving zinc sulfate and ammonium sulfate (50 g/L) in deionised
Stirring speed (N) = 160 rpm, Surfactant surfactant concentration(Wsurf) = 3% wt of oil water. The feed pH was adjusted to the desired value using
phase.
ammonium hydroxide. The water in oil emulsions were prepared
and characterized using techniques discussed in our earlier paper
(Sengupta et al., 2006). FT-IR spectra of LIX 84I complexes with Cu, Zn
membrane loading approached saturation. Recently, Kumbasar and
and Cu + Zn were acquired using a Thermo Nicolet 6700 FT-IR
Kasap (2009) selectively separated nickel from cobalt in ammoniacal
spectrometer.
solutions using ELMs with 8-hydroxyquinoline as carrier and obtained
Experiments were carried out at 30 ± 2 °C in a four-baffled batch
almost 96% recovery of nickel from nickel–cobalt mixtures.
reactor equipped with a six bladed turbine. The feed was agitated at
The extraction of zinc from ammoniacal solutions has received
160 rpm to obtain a fine dispersion of the ELM in the feed. Requisite
much less attention when compared with copper. Ritcey and Lucas
amount of the emulsion was added to the feed phase to achieve the
(1975) reported the extraction of zinc along with copper, nickel and
required treat ratio (ratio of volume of emulsion to feed). Samples
cobalt from ammoniacal solutions using Kelex 100. Przeszlakowski
from the feed were withdrawn at definite time intervals and analyzed
and Wydra (1982) extracted copper, nickel, zinc and cobalt from
for the metal content using AAS (Chemito AA 203).
ammoniacal solutions using Hostarex DK-16, a commercial β-
On completion of extraction, the emulsion was separated from feed
diketone, and observed that the extraction followed the order
phase and demulsified. Extraction data were plotted as ratio of
Cu N Co N Ni N Zn. Later, Rao et al. (1990, 1992) made a more detailed
instantaneous Zn concentration to initial Zn concentration (Ce/Ceo)
study of zinc extraction from ammoniacal solutions using Hostarex
versus time. The range of the process parameters studied, inclusive of
DK-16 and observed that zinc extraction was influenced by the
the optimal pH, is listed in Table 1. To assess the stripping behavior of
ammonium sulfate concentration in the aqueous phase and increased
Zn and Cu from the membrane phase of ELMs, experiments were
with pH up to 7.5 and then declined. This behavior was attributed to
designed to load the metals in the oil phase containing LIX 84I and
the formation of the non-extractable zinc ammine complex at higher
subsequently strip them using 1.5 M H2SO4.
pH. Similarly, Hoh et al. (1982; Hoh and Chuang, 1983) observed that
the extraction of zinc with LIX 34 in the pH range 7 to 11 approached a
maximum at pH 7.8. Their results indicated that the extraction was 4. Results and discussion
exothermic while the stripping was endothermic.
Yang et al. (2006) reported the permeation of copper, zinc, The extraction of zinc by LIX 84I can be represented by the
cadmium and nickel through supported liquid membranes impreg- equilibrium
nated with LIX 54 as well as LIX 84. They reported separation factors of
2þ þ
Cu/Zn as 21.25 and 1.21 with LIX 54 and LIX 84 respectively at flow ½Zn A þ ½2HRO ⇆½ZnR2 O þ ½2H A ð1Þ
rates of 140 mL/min. Thereby indicating superior selective perfor-
mance of LIX 54 for extracting copper over zinc in comparison with LIX where HR denotes the extractant and subscripts A and O denote the
84 from ammoniacal solutions. However, the extraction of zinc from aqueous and organic phases respectively. The above equation
ammoniacal solutions using ELMs has not been reported so far, hence indicates that the pH of the aqueous phase should decline as
there is scope to study the efficacy of ELMs. extraction proceeds since two protons are released into the aqueous
In the present work, LIX 84I as extractant was investigated. The phase for every mole of zinc extracted in the organic phase. Hence, the
influence of variables such as pH, feed concentration, internal phase pH of the aqueous solution becomes a primary parameter controlling
volume fraction of emulsion, treat ratio etc. on the rate of extraction of the reaction.
zinc were studied. It is commonly accepted that β-diketone reagents Experimental studies with LIX 34 by Hoh et al. (1982), Hostarex
(LIX 54) are better suited than hydroxyoximes (LIX 84I) as metal DK16 by Rao et al. (1992) and LIX 54 by Alguacil and Cobo (1998)
extractants from ammoniacal media because they are not susceptible show that %Zn extraction from ammoniacal media increases up to pH
to potential ammonia co extraction. Nevertheless, hydroxyoximes such
as LIX 84I have a much wider range of applicability and generally
provide a greater trans membrane flux. The current investigation not
only complements our earlier work on the efficacy of LIX 84I for copper
extraction from ammoniacal solutions into ELMs (Sengupta et al.,
2007), but also becomes the foundation for simultaneous extraction of
copper and zinc from ammoniacal solutions. The possibility of the
selective extraction of these metals from their mixtures in ammoniacal
solutions was also explored.

2. Materials and reagents

The extractant LIX 84 I (Cognis Inc.) 2-hydroxy-5-nonyl-aceto-

phenone oxime in a high flash point diluent was used as the carrier.
Kerosene (Indian Oil Corp. Ltd.) having a boiling range of 152 °C–
271 °C containing n-paraffins (27.1%), naphthenes (55.9%), aromatics
(16%) and olefins (1%) having a density (ρ15) of 821.3 kg/m3 was used Fig. 1. Effect of feed pH on extraction of zinc (conditions for Figs. 1–6 in Table 1).
 B. Sengupta et al. / Hydrometallurgy 99 (2009) 25–32
Fig. 2. Effect of internal phase volume fraction (ϕ) on extraction of zinc.
7.8–8.1 and thereafter declines. The decline with increasing pH can
readily be explained by Eq. (2) (Kordosky, 1979).
2þ
½ZnðNH3 Þ4 A þ ½2RHO ⇆½ZnR2 O þ ½2NH3 A þ ½2NH4 A :
The ammonia in the solution functions as a coordinating ligand
that competes favorably with the organic extractant for the metal ion.
Stoichiometric excess of ammonia in the aqueous phase leads to the
formation of a zinc ammine complex with the metal ion in the
aqueous phase, resulting in a decrease in the extent of extraction.
In an ELM system, these effects get compounded since at higher pH
even free ammonia is transported and trapped within the membrane.
Lee and Chan (1990) have shown that free ammonia transport occurs
at aqueous phase pH N 11. They observed that at pH 9.5 the free
ammonia content in the aqueous solution was 50% while at pH 8 it was
less than 10%. It appears that ammoniacal solutions at pH ~ 8 are most
appropriate for extraction of zinc. Under such conditions the metal ion
would be largely free to complex with the extractant and free
ammonia concentration would be reasonably low for ammonia to get
transported into ELMs and deplete the stripping acid concentration. In
view of this, the effect of pH on extent of extraction was first
established.
4.1. Effect of pH
The rates of extraction were established at three pH values 7, 8.1
and 8.6 and the results are presented in Fig. 1. It was observed that at
pH 8.1, the extent of extraction was substantially greater in
comparison with the other pH values. At pH 8.6, only 20% of the
initial Zn was extracted. This behavior is likely due to the formation of
the zinc ammine complex in the ammoniacal solution leaving small
amount of free zinc to complex with the extractant and get extracted.
At pH 7 about 30% of Zn was extracted in a minute and thereafter no
Fig. 3. Effect of carrier concentration (Cc) on extraction of zinc.
27
Fig. 4. Effect of internal phase acid concentration (Cio) on extraction of zinc.
further extraction took place. This behavior could be attributed to the
pH equilibrium of Zn-LIX 84I systems (Redbook, 2007).
4.2. Effect of emulsion parameters
ð2Þ
The emulsion parameters, namely internal phase volume fraction
(ϕ), extractant concentration in the membrane phase (Cc), initial
stripping acid concentration (Cio) and surfactant concentration
(Wsurf) in the membrane are known to influence the rates of
extraction in ELM systems. In earlier work Sengupta et al. (2006,
2007) showed that emulsions formulated with 3% wt surfactant based
on the membrane phase were adequate for membrane stability. Hence
all emulsions were formulated with 3% wt. surfactant concentration.
The initial feed pH was maintained at 8.1. All other process parameters
are listed in Table 1.
4.2.1. Effect of ϕ
A change in the internal phase volume fraction ϕ changes the
emulsion morphology, influences the rates of extraction as well as the
stability of the emulsion. All sensible properties of the emulsion
change with changes in ϕ. An increase in ϕ leads to an increase in the
amount of stripping phase in the emulsion thereby increasing its
extraction capacity. It also leads to a decrease in the diffusional
resistance of the metal-oxime due to the formation of thinner
membranes.
The effect of ϕ is shown in Fig. 2. Contrary to expectation of rapid
diffusion and faster extraction with thinner membranes, it is seen that
an increase in ϕ from 0.5 to 0.64 leads to a substantial decrease in the
amount extracted. The lower rate of extraction stems from a number
of physico-chemical changes with increase in ϕ; predominantly
among them are the decrease in extractant concentration present in
the system for a given emulsion hold-up and the increase in internal
drop sizes. The Sauter mean diameter d32 for emulsions with ϕ = 0.5
were found to be 2.72 µm while that for ϕ = 0.64 were 10.74 µm. This
large increase in d32 contributes to a decrease in internal phase
Fig. 5. Effect of treat ratio on extraction of zinc.
28 B. Sengupta et al. / Hydrometallurgy 99 (2009) 25–32
Fig. 6. Effect of initial feed concentration (Ceo) on extraction of zinc.
stripping areas, which is not adequately compensated by the
reduction in membrane thickness.
4.2.2. Effect of carrier concentration Cc
The effect of carrier concentration is shown in Fig. 3 and as
expected, increasing the extractant concentration from 5% to 10%(v/v)
gives a 20% enhancement in the amount extracted. It is noteworthy
that under identical conditions in the case of copper extraction from
ammoniacal media it was found that LIX 84I concentration in the
membrane phase in the range of 2% to 10% (v/v) did not influence the
rate of extraction of copper (Sengupta et al., 2007).
4.2.3. Effect of internal phase acid concentration Cio
An increase in the internal phase stripping acid concentration
increases the rate and extent of extraction as shown in Fig. 4. When
Cio was 0.5 M, the rate of extraction was minimum while at
Cio = 1.5 M the rate of extraction was maximum. At Cio = 1 M the
rate was intermediate to the other two conditions, but rates are
relatively closer to the case where Cio was 0.5 M.
4.3. Effect of process parameters
The process parameters affecting the rates of extraction are the pH,
treat ratio of emulsion to feed and the initial feed concentration. Since
the optimal pH for zinc removal was already established as pH 8.1 the
extractions were carried out at this pH. The other emulsion
parameters are listed in Table 1.
4.3.1. Effect of treat ratio
The treat ratio in an ELM extraction is the ratio of emulsion phase
to feed phase. It is identical to solvent to feed ratio in conventional
liquid extraction. Increasing the treat ratio generally leads to an
increase in the capacity and the rate of extraction. This is because
there is an increase in emulsion volume in the system and corre-
Fig. 7. Zinc extraction, stripping patterns in ELMs.
Fig. 8. Stripping rate of zinc within ELMs.
spondingly an increase in the carrier, strippant amount and also
overall surface area for mass transfer.
Fig. 5 shows the effect of treat ratio on the extraction of zinc from
ammoniacal solutions. As the treat ratio increased there was an
increase in the extent of extraction as well as the extraction rates, but
the increase was sharp when treat ratio increased from 1:15 to 1:10 in
comparison to the increase in treat ratio from 1:10 to 1:6. This pattern
could be attributed to a possible increase in globule size distribution
due to increased hold-up of the emulsion. Sengupta et al. (2006) have
observed a strong decline in extent and rates of extraction when treat
ratios were increased from 1:6 to 1:4 due to increased globule size
distribution at larger emulsion hold-ups. Formation of larger sized
globules not only reduces the external surface areas but also increases
the effective diffusive path lengths within the globule that results in a
decline in the extraction rates.
In the present investigation treat ratios of 1:15, 1:10 and 1:6 were
used resulting in stripping acid to zinc molar ratios of 3.27, 4.9 and 8.18
respectively. These values indicate a substantial increase in extraction
capacity when treat ratio was increased from 1:10 to 1:6 but at the
same time, due to the increase in emulsion hold-up, the globule size
distribution also tended to shift towards larger sizes resulting in a
decrease in rates. The combined effect culminates in the behavior
observed in Fig. 5.
4.3.2. Effect of initial zinc concentration
The effect of initial [Zn] concentration in the feed on the rate of zinc
extraction were investigated using emulsions having ϕ = 0.5, Wsurf = 3%
wt of membrane phase and Cio = 1.5 M. The initial pH and treat ratio
were maintained at 8.1 and 1:6 respectively. Extraction results are
shown in Fig. 6, which is a plot of change in zinc concentration in
aqueous phase with time. The pattern of zinc loading in ELMs along with
Fig. 9. Loading and stripping patterns of zinc.
 B. Sengupta et al. / Hydrometallurgy 99 (2009) 25–32 29

Table 2
Simultaneous loading, stripping patterns of Cu–Zn.

Initial feed concentration (mg/L) Amount loaded in oil phase (mg/L) Amount stripped from loaded oil phase (mg/L) Separation factor β(Cu/Zn)
Cu Zn Cu Zn Cu Zn
a 1705 230 1705 166 1411.9 50 1.4
b 257 1500 257 1374 140.7 1037.1 1.1
c 1000 820 1000 740 671 689.9 1.1

a quantitative assessment of the amount of zinc stripped in the internal Hence, to get a quantitative assessment of the individual steps
stripping phase of the emulsion, after a 180 s contact between the feed involved in the ELM processe, experiments were carried out to
and the ELMs, for initial feed concentration variations is shown in Fig. 7. independently study the loading and stripping of zinc using LIX 84I by
It was observed that as the initial feed concentration increased, the contacting equal volumes (50 mL) of kerosene containing 10% v/v of the
extent of zinc extraction into ELMs also increased. When zinc loading extractant with aqueous ammoniacal solution containing 1000 mg/L Zn
was low most of the zinc extracted in the ELMs got stripped in the at an initial pH 8.1. The pH change and amount of zinc loaded in the oil
internal phase of the ELMs. However, at high zinc loadings the amount phase were recorded with time for a period of 180 s corresponding to
of zinc stripped in the internal phase of the ELMs did not increase time of contact in an ELM system. Subsequently, zinc was stripped by
substantially, hence most of the zinc extracted by the ELMs was contacting equal volumes of loaded oil phase with an aqueous strip
retained in the membrane phase. The low rate of zinc stripping could solution containing 1.5 M H2SO4. The amount stripped at definite time
be attributed to the slow stripping kinetics as well as the diffusional intervals was recorded. Loading and stripping results with time are
effects that play an important role in further slowing down the shown in Fig. 9. Initial rates of loading were steeper than the rates of
stripping rates. High values of Ceo lead to greater zinc loadings in the stripping indicating that the stripping kinetics were slower than the
ELMs causing quick saturation of the peripheral internal phase loading kinetics.
droplets in the emulsion necessitating deeper penetration of the
Zn–oxime complex within the emulsion globules to get stripped. Thus
when Ceo = 2000 mg/L the net zinc uptake by the emulsion was
825 mg, out of which, zinc retained in the membrane phase of the
emulsion was as high as 468 mg. Whereas, when Ceo was 1000 mg/L,
the net Zn uptake by the emulsion was 380.5 mg out of which zinc
accumulated in the membrane phase was only 85.3 mg.
To assess the extent of zinc stripped during the time span of the
experimental runs, a number of experiments were carried out with
zinc concentration Ceo = 1000 mg/L at the (above) specified operat-
ing and emulsion conditions. These experiments were stopped at time
intervals of 60,120,150 and 180 s. The emulsions were quickly
separated from the feed and demulsified. The internal phase was
analyzed for zinc. Although such data is not exact, due to the time
delay of 2–3 min for emulsion separation from feed, during which the
stripping reaction continues, it does provide a representative idea of
the amount stripped with time. The results of these experiments show
a linear relationship (Fig. 8) indicating a constant rate of zinc stripping
under the experimental conditions.

4.4. Extraction and stripping of copper and zinc using LIX 84I

4.4.1. Zinc extraction and stripping

ELMs are complex systems with extraction and stripping phenom-
ena encountered simultaneously at the external feed–membrane
interface and the membrane–strippant interface within the emulsion
globules. Both these phenomena are equilibrium limited and proceed
at finite rates. Further the physico-chemical features of the emulsion
influence the transport rate of the metal from the feed–membrane
interface to the membrane–strippant interface. If the equilibrium
conditions are adequately taken care of then any one of the three
kinetic steps could be the rate controlling one.
If the loading kinetics is rate controlling then it is expected that the
metal retention in the membrane phase would be negligible.
However, if the stripping was the rate-limiting step then metal
retention in the membrane phase would build up with time along
with a decline in the loading rates since the extractant would not be
free to bind the metal. Qualitatively, almost a similar situation would
exist if diffusion was the rate-limiting step. It is difficult to discern the
exact mechanism in ELM extractions because of the short contact
times and the change in pH at the feed–membrane interface with
progress of the extraction, as well as the inter-connectedness of the
various mechanisms. Fig. 10. Stripping patterns of copper, zinc mixtures.
30 B. Sengupta et al. / Hydrometallurgy 99 (2009) 25–32

Fig. 11. IR spectra of LIX 84I metal complexes.

 B. Sengupta et al. / Hydrometallurgy 99 (2009) 25–32
The stripping rates have a considerable influence on the overall rates
of extraction in ELM systems, particularly at high feed concentrations
leading to large membrane loadings. Even in the case of copper
extraction from ammoniacal solutions, it was observed that while
loading was instantaneous, stripping was relatively slow and was also
influenced by the carrier concentration. The decline in extraction rates at
high copper loadings in ELM systems were attributed to the slow
stripping rates by Sengupta et al. (2007).
4.4.2. Simultaneous extraction and stripping of copper and zinc
Prior to simultaneous extraction of copper and zinc in ELM systems,
the nature of simultaneous stripping of copper and zinc from the oil
phase was investigated by loading the metals from a mixture of their
aqueous ammoniacal solutions at an initial pH of 8.1. The loaded oil
phases were then stripped with 1.5 M H2SO4 by contacting equal
volumes of the metal loaded oil and the stripping phases for a time span
of 180 s. Table 2 lists the initial conditions, loading of the metals in the oil
phase, amount stripped and the separation factors defined as
ðCu = ZnÞoil
β Cu = Zn = :
ðCu = ZnÞfeed
It was found that zinc loading in the oil phase substantially
increased in the presence of copper as seen in Table 2, indicating a
synergistic influence. Furthermore, it was observed that zinc gets
stripped more rapidly in comparison with copper when the zinc
concentration was comparable to or more than the copper concentra-
tion as seen in Fig. 10(a) and (c). Even the extent of zinc stripping is
more in the presence of copper. However, the initial rate of zinc
stripping in absence of copper was very rapid but quickly settled to a
constant value and the overall extent of stripping was also lower as
seen in Fig. 10. When the copper concentration was greater than zinc
in the oil phase, copper stripped faster than zinc as seen in Fig. 10(b).
The plateau region observed during stripping in Fig. 10(c) is
inexplicable, but could be due to competitive mechanism of
stripping–loading–stripping in acidic media. However, conclusive
analysis is currently beyond the scope of this work and calls for more
detailed investigation.
The IR spectra of the LIX 84I complex with copper, zinc and
simultaneously with copper–zinc mixtures are shown in Fig. 11. The
band in the region 3500–3200 cm− 1 appears in the spectra of all the
complexes and is representative of the characteristic stretching
vibration of the intermolecular hydrogen bond in the oxime (Yang
et al., 2006). All spectra have strong peaks around 3200–2800 cm− 1
that are characteristic peaks of chelated OH groups (Nakamoto, 1963).
The small peaks around 1025 cm− 1 could be attributed to the
presence of ammonia in the extracted solutions (Alguacil et al., 2001).
Note should be taken of the small peak around 800–700 cm− 1
appearing in the spectra. These bands are attributed to the oxime–
Fig. 12. Simultaneous extraction of Cu–Zn into ELMs: effect of Cio.
31
Fig. 13. Simultaneous extraction of Cu–Zn into ELMs: effect of Ceo.
metal complexes (Yang et al., 2006). Classically for coordination
compounds, peaks appearing in the region 850–650 cm− 1 are
attributed to rocking frequencies. After extensive studies on ammine
complexes, Svatos et al. (1957) found rocking frequency was most
sensitive to the central metal ion. Fig. 11 shows the augmentation of
the peak observed at 757 cm− 1 with the metal in the order Cu–
Zn N Zn N Cu which also corresponds to the ease of stripping observed
in this investigation.
4.5. Simultaneous extraction of copper and zinc into ELMs
Simultaneous extraction of zinc and copper from ammoniacal
solutions was investigated using ELMs having ϕ = 0.5, Cc = 10% v/v
LIX 84I in membrane phase, Wsurf = 3% wt membrane phase with
treat ratio 1:6. The desired feed solutions were prepared by dissolving
requisite amounts of copper sulfate, zinc sulfate and ammonium
sulfate (50 g/L) in deionised water and the pH was adjusted to 8.1
using ammonium hydroxide. The effects of the strip acid concentra-
tion and the initial concentrations of zinc and copper in the feed on
the rates of extraction were investigated.
4.5.1. Effect of stripping acid concentration
Internal phase acid concentrations were varied from 0.5 M to 1.5 M.
The initial feed concentrations of zinc and copper, though not exactly
same, were comparable. The zinc concentration ranged from 880 to
1037 mg/L and the copper concentration ranged from 990 to
1032 mg/L. Extraction results are shown in Fig. 12. It is clear that
the rate of extraction of both zinc and copper increased with stripping
acid concentration. However, the rates of copper extraction were
much larger than that observed for zinc. Moreover, while copper was
almost quantitatively extracted in 180 s into ELMs having Cio of 1 M
and 1.5 M, even in the best case only 50% Zn was extracted in the same
time with emulsions having Cio 1.5 M.
Comparing Fig. 4 with Fig. 12 the effect of Cio on zinc extraction in
the presence of copper is identified. There was a significant decline in
the extent and rates of zinc extraction in the presence of copper. This
behavior is contrary to the loading and stripping behavior of copper
and zinc shown in Table 2 and Fig. 10(a). During the simultaneous
loading and stripping of copper and zinc it was observed that zinc
loads to a greater extent in the presence of copper and also strips
faster and to a greater extent in the presence of copper. This pattern
should also be true with ELMs. However, it is noteworthy that in ELMs,
loading and stripping occur simultaneously and the large difference in
rates observed in Fig. 11 could only be attributed to the loading
behavior of the metal and the affinity of the oxime for copper. The
higher loading capacity and loading rate of the oxime for copper
masks the slow stripping kinetics as well as the diffusional effects
caused by the various physico-chemical factors.
32 B. Sengupta et al. / Hydrometallurgy 99 (2009) 25–32
Table 3
Extraction and stripping of Cu–Zn in ELMs.
Initial feed concentration Net amount in feed Amount loaded in ELM
(mg/L) (mg) (mg)
Cu Zn Cu Zn Cu
a 1705 230 1023 138 1000.6
b 257 1500 154.2 900 143.6
c 1000 820 600 492 571.3
a
At 120 s.
4.5.2. Effect of initial feed compositions
Three variations of the initial feed concentrations were explored
ranging from feeds where copper concentration was in large excess in
comparison with zinc, when copper and zinc concentrations were
comparable and when zinc concentration was much larger than
copper. Furthermore, the total metal concentrations in the feeds were
also comparable ranging from 1757 mg/L to 1935 mg/L. The ELMs
used had Cio 1.5 M all other conditions remained identical.
Extraction results are shown in Fig. 13. In all three cases copper
extraction was much faster than zinc and almost quantitative removal
of copper took place within 180 s while a maximum of 55% Zn removal
took place in the same time duration and that too when its
concentration in feed was quite low. Table 3 details the quantitative
aspects of the Cu–Zn separation using ELMs. It is distinctly observed
that ELMs with LIX 84I show a preference for the selective removal of
copper from Cu–Zn mixtures. All other considerations including
stripping kinetics of copper and zinc along with other physico-
chemical features such as diffusion of the oxime complex within the
emulsion globule etc are completely masked by the affinity of the
oxime for copper. The separation factor βCu/Zn obtained at time
duration of 120 s, corresponding to onset of maximum copper
removal, ranged from 1.8 to 3.5 (Table 3).
It was found that ELMs loaded with LIX 84I had the efficacy of
removing zinc as well as mixtures of copper and zinc from aqueous
ammoniacal media. Furthermore, they can also be used for selective
removal of copper from Cu–Zn mixtures in ammoniacal media.
5. Conclusions
The significant conclusions drawn from the work are the following:
i. pH 8.1 was optimal for the removal of zinc as well as copper
from ammoniacal solutions using ELMs.
ii. The emulsion parameters ϕ, Cio and Cc considerably influence
the zinc extraction rates. An increase in ϕ had a deleterious
effect on the metal extractions; optimal values of emulsion
parameters were Cio = 1.5 M, Cc = 10% LIX 84I and ϕ = 0.5.
iii. An increase in the feed [Zn] concentrations resulted in larger
zinc uptake by the ELMs. However, zinc stripping was slow,
hence there was considerable amount of zinc retained in the
membrane phase, particularly when zinc uptake in the ELMs
was large. The zinc stripping rate was found to be constant in
the time span of the experiments.
iv. Loading–stripping experiments with zinc indicate that strip-
ping is slower in comparison to loading.
v. Simultaneous extraction and stripping experiments indicate
that zinc loads in the oil phase to a larger extent in the presence
of copper. Zinc strips faster than copper from the oil phase
when the zinc concentration is comparable or more than
copper.
vi. Simultaneous extraction of copper and zinc using ELMs
indicates that the overall copper extraction rates are much
larger than zinc. Loading of copper dictates the separation.
ELMs with LIX 84I show preference for selective removal of
Amount stripped in internal phase of ELM Separation factora β(Cu/Zn)
(mg)
Zn Cu Zn
82.5 626.5 82.5 3.5
405 138 345 1.8
228 197.1 146 2.7
copper from zinc with separation factor β(Cu/Zn) = 3.5. These
ELMs could be used for treating the ammoniacal wastewaters
obtained from PCB manufacturing units containing Cu–Zn
mixtures.
Acknowledgements
We would like to thank AICTE, R&D Scheme F.No.8021/RID/
NPROJ/R&D-2/2002-03 for support to carry out this work, Indian Oil
Corporation Limited for supply of kerosene and Cognis Inc. (Ireland)
for supply of samples of LIX reagents.
References
Alguacil, F.J., 2001. Mechanistic study of active transport of copper(II) from ammoniacal/
ammonium carbonate medium using LIX 973 N as a carrier across a liquid membrane.
Hydrometallurgy 61, 177–183.
Alguacil, F.J., Cobo, A., 1998. Extraction of zinc from ammoniacal/ammonium sulphate
solutions by LIX 54. J. Chem. Tech. Biotechnol. 71, 162–166.
Alguacil, F.J., Alonso, M., Lopez, F.A., 2001. Influence of ammonium salts on solvent
extraction of nickel using LIX 54. J. Chem. Eng. Jpn. 34, 83–86.
Chen, J.H., Huang, C.E., 2007. Selective separation of Cu and Zn in the citric acid leachate
of industrial printed wiring board sludge by D2EHPA — modified Amberlite XAD-4
resin. Ind. Eng. Chem. Res. 46, 7231–7238.
Cognis Corporation, MCT Redbook, 2007. Solvent Extraction Reagents and Applications.
Flett, D.S., 1979. Solvent extraction in scrap and waste processing. J. Chem. Tech.
Biotechnol. 29, 258–264.
Gameiro, M.L.F., Bento, P., Ismael, M.R.C., Reis, M.T.A., Carvalho, J.M.R., 2007. Extraction
of copper from ammoniacal medium by emulsion liquid membranes using LIX 54. J.
Membr. Sci. 293 (1–2), 151–160.
Gupta, C.K., Mukherjee, T.K., 1990. Hydrometallurgy in Extraction Processes, vol I. CRC
Press, Boca Raton, pp. 127–165.
Hoh, Y.C., Chuang, W.S., 1983. Influence of temperature on the extraction of zinc by LIX
34. Hydrometallurgy 10, 123–128.
Hoh, Y.C., Chou, N.P., Wang, W.K., 1982. Extraction of zinc by LIX 34. Ind. Eng. Chem.
Process Des. Dev. 21, 12–15.
Kordosky, G.A., 1979. The Chemistry of Metals Recovery using LIX Reagents, Mineral
Industrial Group. Henkel Corporation.
Kumbasar, R.A., Kasap, S., 2009. Selective separations of nickel and cobalt in
ammoniacal solutions by emulsion type liquid membranes using 8-hydroxyquino-
line (8-HQ) as mobile carrier. Hydrometallurgy 95, 121–126.
Lee, C.J., Chan, C.C., 1990. Extraction of ammonia from a dilute aqueous solution by
emulsion liquid membranes. 1. Experimental studies in a batch system. Ind. Eng.
Chem. Res. 29, 96–100.
Nakamoto, K., 1963. Infrared Spectra of Inorganic and Coordination Compounds. John
Wiley and Sons, NewYork.
Przeszlakowski, S., Wydra, H., 1982. Extraction of nickel, cobalt and other metals (Cu,
Zn, Fe(III)) with a commercial β-diketone extractant. Hydrometallurgy 8, 49–64.
Rao, K.S., Sarma, P.V.R.B., Jena, P.K., 1990. Effect of ammonium salts on the extraction of
zinc using Hostarex DK-16 as solvent. Erzmetall 43, 366–369.
Rao, K.S., Sahoo, P.K., Jena, P.K., 1992. Extractions of zinc from ammoniacal solutions by
Hostarex DK-16. Hydrometallurgy 31, 91–100.
Ritcey, G.M., Lucas, B.H., 1975. Extraction and separation of copper, nickel, zinc and
cobalt from ammoniacal solution using Kelex 100. C.I.M. Bull. 68 (754), 105–113.
Svatos, G.F., Sweeny, D.M., Mizushima, S., Curran, C., Quagliano, J.V., 1957. Infra-red
absorption spectra of inorganic coordination complexes. XII. The characteristic NH3
deformation vibrations of solid inorganic complexes. J. Am. Chem. Soc. 79,
3313–3315.
Sengupta, B., Sengupta, R., Subrahmanyam, N., 2006. Process intensification of copper
extraction using emulsion liquid membranes: experimental search for optimal
conditions. Hydrometallurgy 84 (1–2), 43–53.
Sengupta, B., Bhakhar, M.S., Sengupta, R., 2007. Extraction of copper from ammoniacal
solutions into emulsion liquid membranes using LIX 84I. Hydrometallurgy 89 (3–
4), 311–318.
Yang, Q., Jiang, J., Chung, T.S., Kocherginsky, N.M., 2006. Experimental and computa-
tional studies of membrane extraction of Cu(II). AIChE J. 52, 3266–3277.