Separation and Purification Technology 167 (2016) 37–44

Contents lists available at ScienceDirect

Separation and Purification Technology

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

Development of an efficient process for the recovery of zinc and iron as

value added products from the waste chloride solution
Manish Kumar Sinha a,b, Swati Pramanik a, Sushanta Kumar Sahu a,⇑, Lal Bahadur Prasad b,
Manis Kumar Jha a, Banshi Dhar Pandey a
a
Metal Extraction and Forming Division, CSIR-National Metallurgical Laboratory, Jamshedpur 831007, India
b
Department of Chemistry, Banaras Hindu University, Varanasi 221005, India

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

Article history: The present paper deals with the recovery of the Zn(II) and Fe(II)/Fe(III) from the waste chloride solution
Received 26 March 2016 of steel industry using solvent extraction and precipitation-stripping process for the synthesis of high
Received in revised form 29 April 2016 purity value added products. The extraction of zinc and iron was performed using Cyanex 923. It was
Accepted 30 April 2016
observed that Cyanex 923 selectively extracted Zn(II) and Fe(III) from the acidic chloride solution leaving
Available online 30 April 2016
behind Fe(II) in the raffinate. The loaded organic was stripped efficiently using oxalic acid to precipitate
Zn(II) as zinc oxalate and Fe(III) was transferred to the aqueous phase as highly soluble ferric oxalate
Keywords:
solution while simultaneously regenerating the Cyanex 923. The zinc oxalate was calcined to obtain
Solvent extraction
Precipitation-stripping
the high purity zinc oxide (99.9% pure). Further the ferric oxalate solution was reduced by using Fe-
Cyanex 923 powder to get ferrous oxalate, which on calcination yielded pure hematite. The oxalates and oxides were
Zinc/ferrous oxalate characterized by chemical analysis, XRD and SEM studies to establish the purity and morphology of the
Metal oxides as-synthesised products.
Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction ondary resource for Zn and Fe if processed suitably. Although,

Prior to hot dip galvanizing, steel surface is generally washed or
cleaned in order to dissolve the rust products (iron oxides into iron
chloride) by dipping it into a hydrochloric acid bath of certain con-
centration (
20%). The same acid bath is used continuously till its
acidity falls to an unacceptable range (<5%) for further pickling
process. Sometimes the same pickling bath is also used periodically
for the removal of depleted zinc layers from the old or used gal-
vanised steel. The zinc contamination also comes from the hooks
and zigs used for hanging the steel during pickling process. Conse-
quently, in the course of pickling process, large concentration of Zn
and Fe (mainly Fe(II)) builds up in the solution. Based on the avail-
able reports by various investigators [1–11], the average composi-
tion of spent hydrochloric acid pickling solution can be represented
as: 10–250 g/L (0.27–6.8 M) HCl, 5–120 g/L (0.076–1.8 M) Zn and
5–200 g/L (0.09–3.6 M) Fe(T). Owing to its acidic nature and high
metal content the pickling solution is considered as hazardous
for environment and is not endorsed for discarding as such before
an appropriate pre-treatment. At the same time such a spent pick-
ling solution has a significant potential to be considered as sec-
⇑ Corresponding author.
E-mail address: sushanta_sk@yahoo.com (S.K. Sahu).
many processes are available for the treatment of this types of
wastes such as spray roasting, electrodialysis, solvent extraction,
and precipitation. But most commonly used technique for the pro-
cessing of spent pickling solution is the precipitation-
neutralization process with lime. This process generates huge
quantity of sludge which is either disposed as landfill or stored
for further processing. The loss of metal values during the precip-
itation makes this process inappropriate in terms of economic
and environmental advantages. Whilst the applications of other
processes (spray roasting, electrodialysis, etc.) are associated with
some disadvantages like energy inefficiency and production of
impure metals values. In recent years various techniques involved
in the processing of spent pickling solution have been extensively
reviewed [8,12]. However, the use of solvent extraction (SX)
method is often recommended over other methods due to its high
selectivity and versatility, and ability to produce highly pure prod-
ucts as well as handling of large amount of solutions [13].
For the recovery of Zn(II), Fe(II), Fe(III) and acid from the spent
pickle liquor various extractants such as D2EHPA, DBBP, TBP,
Cyanex 921, Cyanex 923, Cyanex 301, Cyanex 302, Cyanex 272,
Alamine 336, Alamine 308, Aliquat 336, TEHA, and TiOA
[1,9,10,13–21], have been comprehensively applied. Above
reagents extract Zn(II) and Fe(III) simultaneously from the waste

http://dx.doi.org/10.1016/j.seppur.2016.04.049
1383-5866/Ó 2016 Elsevier B.V. All rights reserved.
38 M.K. Sinha et al. / Separation and Purification Technology 167 (2016) 37–44
pickle liquor. Zn(II) and Fe(III) present in the organic phase are sub-
sequently separated in the stripping stage.
In the solvent extraction, stripping of loaded metal from the
organic phase is conventionally carried out with acid/alkali. From
the stripped liquor metal values are then recovered by precipita
tion/crystallization/electrowinning. However, in precipitation-
stripping method stripping and precipitation of loaded metal ions
takes place with the help of a stripant containing functional groups
which form insoluble salt in the stripped aqueous phase. The pre-
cipitates so produced upon decomposition in hydrogen or oxygen
atmosphere, are converted to metal/metal oxide powder [22]. For
the technological applications of such products, the particles must
have uniform shape and close size distribution, low impurity con-
tent, etc. Control of particle size is inherently difficult by the usual
aqueous precipitation processes [23,24]. It is however, rather
easier to control the particle size by varying the precipitation con-
ditions during the precipitation-stripping step [25]. As such the
precipitation-stripping process has been predominantly used and
reported in the production of rare earth oxalates/carbonates
[24,26–30], while fewer reports are available on the production
of base metal salts/oxides during the recovery of these metal ions
[25,31]. The use of oxalic acid as the precipitation-stripping agent
is often preferred due to its ability to form stable chelates with dif-
ferent metal ions generating precursors with a high degree of
homogeneity. The availability of oxalic acid at low cost as well
makes the production economical. The basic principle or require-
ment of oxalic acid precipitation-stripping in multi-metal system
depends upon the stability constant of the particular metal oxalate
and acidity of the solution.
Among several solvent extraction reagents, the use of Cyanex
923 as an extractant for the extractive-recovery of Zn(II) and Fe
(III) has been reported by various investigators [8,14,15,21,32–
35]. As such Cyanex 923 in spite of being a stable reagent in highly
acidic conditions and showing good extraction efficiency, is not
recommended for use in the recovery of Zn(II) and Fe(III) from
waste pickle liquor due to its poor performance during stripping
with hot/cold water, HCl or NaOH solutions [8,14,15]. In view of
the above, the present investigation has been carried out with
the purpose of extracting zinc and iron from the waste pickle
liquor using Cyanex 923 as an extractant in kerosene and separat-
ing them by the precipitation-stripping method. This could conve-
niently be achieved by selective stripping and precipitation of zinc
simultaneously from the loaded organic with oxalic acid. Although
Fe(III) is also stripped with oxalic acid, but it would remain in the
soluble form. While the precipitated zinc oxalate can easily be sep-
arated from iron(III) oxalate by filtration, iron(III) oxalate in the fil-
trate is precipitated as iron(II) oxalate after reduction with iron
powder. Subsequently zinc oxalate and iron(II) oxalate have been
roasted to produce high pure zinc oxide and hematite, respectively.
The products were characterized by chemical analysis, XRD, TG-
DTA and SEM-EDS studies.
2. Experimental
2.1. Reagents and materials
Analytical grade reagents such as zinc chloride (ZnCl2, 99%), fer-
rous chloride (FeCl24H2O, 98%), ferric chloride (FeCl3, 98%) and
hydrochloric acid (37%) were used to prepare the model solution
simulating the average composition of the spent pickle liquor hav-
ing composition- HCl: 2 M, Zn: 0.76 M, Fe(II): 0.36 M and Fe(III):
0.05 M. Oxalic acid (H2C2O42H2O) was used for preparing the
stripping solution. All these chemicals/reagents were obtained
from Merck, India. Cyanex 923 which is mixture of trialkyl phos-
phine oxides (alkyl: n-octyl and n-hexyl; Mol. Wt.- 348), was
kindly provided by Cytec Inc., Canada and used without any further
purification or pre-equilibration with water. Commercial grade
kerosene (Indian Oil Ltd.) was used as the diluent.
2.2. Procedure-solvent extraction and material preparation
Solvent extraction experiments were carried out by shaking
equal volumes of the model spent pickle liquor and Cyanex 923
of known concentration in a separating funnel for 30 min, which
was found to be sufficient to reach equilibrium. After the phase
separation the aqueous and organic phases were separated. Acid
concentration in the aqueous phase was determined by potentio-
metric titration with 1 N NaOH solution, whereas iron was anal-
ysed by dichromate method using BDS (barium salt of diphenyl
ammine sulfonate) as indicator. The concentration of zinc in the
aqueous phase was volumetrically analysed using EDTA in the
presence of EBT (Erichrome Black-T) indicator. Precipitation-
stripping of Zn(II) was performed after the solvent extraction of
Zn(II) and Fe(III) from the aqueous phase. The stripping of loaded
Zn(II) and Fe(III) were accomplished by contacting the Cyanex
923 phase with oxalic acid of a known concentration for 30 min.
The oxalic acid solution was added dropwise under constant stir-
ring. The stirring was controlled in such a way that a full dispersion
of the solution mixture could be attained to avoid the agglomera-
tion of the particles. After mixing the mixture was centrifuged to
separate the liquid phases (aqueous ferric oxalate solution and
stripped organic solution) from the precipitate (zinc oxalate). The
stripped ferric oxalate solution obtained was reduced with iron
scraps to form the ferrous oxalate precipitate. After complete reac-
tion, both the products viz. oxalates of zinc and ferrous were fil-
tered by using vacuum filtration then washed with distilled
water and stirred in 2-propanol solution for 15 min to remove
any organic contaminants. Finally, the washed oxalates were cal-
cined in ceramic crucible for producing zinc oxide and iron oxide
powders.
2.3. Characterization of the synthesised products
The oxalate and oxide particles obtained were characterized
using various techniques. X-ray powder diffraction (XRD) patterns
(Bruker D8-discover) were obtained with Cu Ka (k = 1.54178 Å)
radiation ranging from 2h = 10–90° at a scanning rate of 2°/min
to identify the phases present. The operation voltage and current
were kept at 40 kV and 40 mA, respectively. The thermal behaviour
of the oxalate powders was investigated in air atmosphere employ-
ing thermo-gravimetric (TG) and differential thermal analysis
(DTA) (LINSIES GmbH, Germany). The morphologies and particle
sizes were also analysed after sputtering with a conductive layer
of silver, using a FEI NOVA Nano SEM-430 scanning electron micro-
scope (SEM) with an EDAX attachment for X-ray microanalysis.
3. Results and discussion
3.1. Extraction and separation of zinc and iron from the solution
3.1.1. Solvent extraction of Zn(II) and Fe(III)
In order to optimize the conditions, solvent extraction of Zn(II),
Fe(II and III) and HCl from a model waste pickle solution containing
0.76 M Zn, 0.36 M Fe(II), 0.05 M Fe(III) and 2 M HCl was carried out
with various concentrations of Cyanex 923 (0.25–1.76 M) in kero-
sene as diluent, and the results are depicted in Fig. 1. It is clear
from the plots that the extraction of different metals followed
the order Fe(III) > Zn(II) > HCl. In the entire range of concentration
of Cyanex 923 investigated, Fe(III) was completely extracted into to
the organic phase, whereas Fe(II) extraction was negligible. The
 M.K. Sinha et al. / Separation and Purification Technology 167 (2016) 37–44 39

100 0.9

90 0.8
80 Zn
0.7

Cumulative loading (M)

70 HCl
0.6
% Extraction

60 Fe(III)
Zn 0.5 Fe(II)
50 HCl
0.4
40 Fe(II)
Fe(III) 0.3
30

20 0.2

10 0.1

0 0
0 0.5 1 1.5 2 0 1 2 3 4 5
Cyanex 923 (M) Number of contacts

Fig. 1. Effect of extractant concentration on the solvent extraction of zinc and iron. Fig. 2. Loading capacity of Cyanex 923. Aq. phase: 0.76 M Zn, 0.36 M Fe(II), 0.05 M
Aq. phase: 0.76 M Zn, 0.36 M Fe(II), 0.05 M Fe(III) and 2 M HCl, Org. phase: 0.25– Fe(III) and 2 M HCl, Org. phase: 1.5 M Cyanex 923, time: 30 min, O/A = 1:1.
1.76 M Cyanex 923 in kerosene. time: 30 min, O/A = 1:1.

extraction of Fe(III) using the neutral extractants occurs predomi- 0.8

nantly through the formation of HFeCl42L species under the stud-
ied range of chloride concentration [33,36]. Negligible extraction of 0.7
Fe(II) may be due to the fact that in the aqueous phase Fe(II)
mainly exists as Fe2+ and FeCl+ which are not extracted by any sol-
0.6
vating extractants [9,17,37]. Extraction of Zn(II) from model waste
pickle liquor increased from 9 to 86% with the increase in Cyanex
0.5
923 concentration from 0.25–1.76 M. Under the highly acidic con-
[Zn]org, M

ditions acido-chlorocomplexes of Zn such as HZnCl3 and H2ZnCl4

0.4
are predominant formed in the aqueous phase [6]. These species
of Zn(II) are readily extracted by Cyanex 923 forming HZnCl32L/
H2ZnCl42L (L = Cyanex 923) complexes in the organic phase, which 0.3
O/A = 1
was confirmed from the plot of log DZn versus log [Cyanex 923]
having slope value of two (Figure not included). The formation of 0.2
similar zinc complex during the extraction with Cyanex 923 has
also been reported [34]. Along with the extraction of Fe(III) and 0.1
Zn(II), a maximum of 17% of HCl was also extracted with Cyanex
923. Small extraction of HCl by Cyanex 923 during extraction of 0
Zn(II) and Fe(III) has also been reported [14]. Further experiments 0 0.2 0.4 0.6 0.8
were performed using 1.5 M Cyanex 923 as above this concentra- [Zn]aq, M
tion separation of phases takes longer time due to the formation
of dense phases. Fig. 3. McCabe Thiele plot for the extraction of Zn(II). Aq. phase: 0.76 M Zn, 0.36 M
Fe(II), 0.05 M Fe(III) and 2 M HCl, Org. phase: 1.5 M Cyanex 923.
Loading capacity was determined by repeatedly contacting
1.5 M Cyanex 923 with the fresh model pickle solution. In three
repeated contacts organic phase got saturated with 0.76 M Zn(II),
0.16 M Fe(III) and 0.41 M HCl (Fig. 2). On further contacting the be easily used for the synthesis of monohydrated ferrous sulphate
organic phase with the model solution neither the concentration salt in the presence of excess sulphuric acid or hydrated ferrous
of these metal ions increased in the organic phase nor the acid. chloride salt by evaporation/crystallization process.
McCabe Thiele plot was drawn using the data obtained from the
loading capacity experiment to determine the number of 3.1.2. Precipitation-stripping of the loaded Cyanex 923 with oxalic acid
counter-current stages required for complete extraction of Zn(II). Ferric ions in oxalic acid form water soluble ferric oxalate com-
It was observed that three counter-current stages are enough for plex, whereas Zn(II) ions readily precipitate in oxalic acid forming
the complete extraction of Zn(II) from the model solution (Fig. 3). highly insoluble zinc(II) oxalate complex (Ksp (zinc oxalate) =
A laboratory scale mixer settler unit was used for the extraction 1.38  109) [38]. Thus oxalic acid can be used as stripping
of Zn(II) and Fe(III) from a model waste pickle solution containing agent for simultaneous stripping as well as separation of Zn(II)
0.76 M Zn, 0.36 M Fe(II), 0.05 M Fe(III) and 2 M HCl using 1.5 M and iron(III) from the organic phase. The stripping of Fe(III) and
Cyanex 923 in kerosene as extractant. The flow rates of aqueous precipitation-stripping of Zn(II) with oxalic acid can be expressed
and organic phases were maintained at 8.1 L/h. At the end of the as:
three stages of extraction 0.76 M Zn(II) and 0.16 M Fe(III) were 2HFeCl4  2L þ 3H2 C2 O4 ! Fe2 ðC2 O4 Þ3 þ 4L þ 8HCl ð1Þ
loaded into the organic phase. The raffinate obtained after the
HZnCl3  2L þ H2 C2 O4 ! ZnC2 O4 # þ2L þ 3HCl ð2Þ
counter current extraction contained mainly Fe(II), which could
40 M.K. Sinha et al. / Separation and Purification Technology 167 (2016) 37–44

Stripping of loaded Cyanex 923 containing 0.76 M Zn and 16000

0.16 M Fe(III) was carried out with various concentration of oxalic
acid (0.5 M to 2 M) at a phase ratio of O/A = 1. Due to limited sol- 14000 Zinc Oxalate dihydrate
ubility of oxalic acid in water (1.13 M at 25 °C), stripping experi-
(PDF: 00-0001-0295)
ments with oxalic acid concentration above 1 M was conducted at 12000
35 °C. As stated above after stripping, the Fe(III) remained in the
aqueous solution whereas Zn(II) precipitated as zinc oxalate. It 10000
was found that stripping of both Fe(III) and Zn(II) increased with

Counts
increase in oxalic acid concentration in the aqueous phase. When 8000
the oxalic acid concentration was varied in the range 0.5–2 M,
almost 48% to quantitative stripping of Fe(III), and 10–50% precip- 6000
itation stripping of Zn(II) was observed (Fig. 4). Iron in the zinc
oxalate precipitate and zinc in the iron solution (stripped liquor)
4000
was not detected. Due to the limited solubility of oxalic acid in
water at room temperature further stripping experiments were
2000
carried out at different phase ratios (A/O) with 1 M oxalic acid
solution. The phase ratios (A/O) were varied from 1 to 5. It was
0
found that at a phase ratio A/O = 3, both Fe(III) and Zn(II) were 10 30 50 70 90
stripped quantitatively and generating organic phase suitable 2-Theta
for reuse. The regenerated Cyanex 923 was tested for 10
extraction-stripping cycles and the extraction efficiency of regen- Fig. 5. XRD patterns of zinc oxalate particles.
erated Cyanex 923 was found to be nearly unaltered in compar-
ison to the freshly prepared organic phase. After complete
Kk
stripping of Fe(III) and Zn(II) with 1 M oxalic acid solution at D¼ ð3Þ
the phase ratio (A/O) of 3, zinc oxalate precipitate produced in b cos h
the aqueous phase was separated by filtration. The precipitated
where K is a constant, k is wavelength of the X-ray (0.154 nm), b is
zinc oxalate and the iron containing oxalate solutions were trea-
the full width at half maximum (FWHM) of the most intense X-ray
ted separately to produce value added products of zinc and iron,
diffraction line and h is the corresponding Bragg angle and D is par-
respectively.
ticle diameter size. The average crystallite size of the zinc oxalate
was estimated to be
38 nm.
3.2. Preparation of zinc oxide from zinc oxalate and its
TG-DTA curves were recorded on heating the above synthesised
characterization
zinc oxalate up to 600 °C at the heating rate of 10 °C/min in air.
Fig. 6 shows that zinc oxalate decomposes via two weight loss pro-
The XRD of zinc oxalate as shown in Fig. 5 confirmed the syn-
cesses in three steps viz. dehydration to form anhydrous oxalate
thesis of zinc oxalate dihydrate (ZnC2O42H2O) during the process
and decomposition of oxalate to oxide. The first weight loss occurs
of precipitation-stripping. The XRD pattern was identical to the
in the temperature range 115–170 °C with the removal of water
reported data in JCPDS card (JCPDS 001-0295) for cubic zinc oxa-
molecules, leading to the formation of anhydrous zinc oxalate. Fur-
late dihydrate without presence of any other impurities/phases.
ther, the second weight loss region (370–413 °C) accompanied
The sharp and strong peaks imply that particle growth takes place
with the decomposition of the anhydrous oxalate and release of
with high crystallinity. The average crystal size was calculated by
CO2 and CO molecules, is attributed synthesis of zinc oxide. The
the Scherrer’s formula [39]:
total weight loss of
58% is found to be very close to the theoret-
ical value (
57.1%) for the conversion of zinc oxalate dihydrate to

100 60 100 -70

Fe ZnC2O 4.2H 2O
50 -60
Zn 90
% Precipitation yield (Zn)

80 TG

DTA -50
% Stripping (Fe)

40
80
60
% wt. loss

-40
µV/mg

30
70
40
-30
20
60
20 -20
10
50 -10
0 0
0 0.5 1 1.5 2
Oxalic acid, M 40 0
50 150 250 350 450 550
Fig. 4. Effect of oxalic acid concentration on the precipitation-stripping of Zn(II) Temperature (oC)
and iron. Org. phase: 1.5 M Cyanex 923 loaded with 0.76 M Zn(II), 0.16 M Fe(III) and
0.41 M HCl. Aq. Phase: 0.5–2 M oxalic acid, time: 30 min, O/A = 1:1. Fig. 6. TG-DTA curves of zinc oxalate particles.
 M.K. Sinha et al. / Separation and Purification Technology 167 (2016) 37–44 41

zinc oxide. The decomposition of oxalate precursor is also signified
by two characteristic endothermic peaks in DTA curve. The first
peak at
160 °C corresponds to the dehydration and the second
peak at
400 °C implies for the conversion of oxalate into oxide.
The decomposition of zinc oxalate into zinc oxide can be repre-
sented as:
ZnC2 O4  2H2 O ! ZnC2 O4 þ 2H2 O
ZnC2 O4 ! ZnO þ CO2 þ CO
Thus the TG-DTA analysis confirms the formation of dihydrate
zinc oxalate and its conversion to the zinc oxide phase. Accord-
ingly, the zinc oxalate as a precursor was calcined at 600 °C for
1 h to convert it completely into zinc oxide which was subse-
quently characterized by XRD and SEM studies. The XRD patterns
of the ZnO (Fig. 7) show that all the diffraction peaks can be
indexed for the hexagonal wurtzite structure of ZnO (JCPDS 36-
1451) with the lattice parameters of a = 3.249 and c = 5.206, with-
out any identifiable impurities. The absence of any impurities and
oxalate phase confirms the complete calcination of the product.
The average crystallite size of the ZnO calculated from the Scher-
2000
1800
rer’s equation was found to be 19 nm. The decrease in crystallite
size from oxalate to oxide may be attributed to the evolution of
CO and CO2 during the thermal decomposition process.
Morphology and microstructure of the zinc oxalate and zinc
oxide were evaluated with SEM as shown in Fig. 8a and b. It is
apparent from Fig. 8a that zinc oxalate is composed of mainly pris-
matic particles of 5 lm size range. The crystallite size calculated
from the XRD is considerably smaller than the particle size
observed by SEM suggesting that the final particles are polycrys-
ð4Þ
talline in nature and composed of several crystals, which are
ð5Þ deemed to have the same orientation within the particles. EDX
(energy dispersive X-ray) analysis has further confirmed the purity
of produced zinc oxalate as all the peaks correspond to only Zn, C
and O. The additional peaks of Ag in the EDX spectrum originates
from the silver coating which was used for the preparation of the
sample for SEM observation. On the other hand the SEM image
(Fig. 8b) of the zinc oxide produced by calcining zinc oxalate at
600 °C reveals that ZnO morphology mainly consists of needle like
particles along with some globular shaped particles; these particles
essentially have an irregular size range from about 0.5 (spherical)
to 2 lm (needle). The EDX (inset of Fig. 8b) showed the typical
spectra of pure ZnO phase consists of only Zn and O peaks. The
chemical analysis of ZnO product indicated the presence of

80.2% of Zn with 0.05% Fe as impurity. Thus EDS and chemical
analysis confirmed that the ZnO particles produced in the process
have a purity of 99.9%.
3.3. Preparation of iron oxide powder from iron(III) oxalate solution

1600 Zinc Oxide Ferric oxalate solution produced during stripping of the organic
(PDF: 00-036-1451) phase can be used for the preparation of ferric oxalate crystals by
1400 evaporation and crystallization process. However, the process is
1200 difficult and may lead to some decomposition of oxalate.
Counts

Preparation of iron oxide from ferric oxalate solution by solution

1000 decomposition at high temperature yields mixed iron oxide
agglomerates [40].
800 Ferric oxalate is highly stable in solution, while ferrous oxalate
600 in oxalic acid is highly insoluble (Ksp (Ferrous Oxalate) = 2.1  107) and
easily gets precipitated. Iron powder may be considered as the
400 most suitable reducing agent for this purpose since no external
ions will be introduced into the reaction medium. Fe scrap in the
200
form of powder not only reduces the ferric ion but also increases
0 the concentration of iron in the system. The primary reaction
10 30 50 70 90 involved in the reduction of ferric oxalate with Fe scrap-powder
2-Theta can be expressed as [41]:

Fig. 7. XRD patterns of zinc oxide particles. 2Fe2 ðC2 O4 Þ3 þ Fe  scrap ! 3FeðC2 O4 Þ # ð6Þ

Fig. 8. SEM images of the (a) zinc oxalate and (b) zinc oxide particles.
42 M.K. Sinha et al. / Separation and Purification Technology 167 (2016) 37–44

Therefore, ferric oxalate present in the solution was converted
to ferrous oxalate by adding stoichiometric amount of Fe scrap
powder (<200 lm) while stirring it at 50 °C for 60 min. After reac-
tion a bright yellow precipitate of ferrous oxalate was filtered and
washed as discussed in the experimental section.
The XRD pattern (Fig. 9) of the ferrous oxalate so produced
matches well with the diffraction peaks indexed for the mineral
Humboldtine (monoclinic ferrous oxalate dihydrate); FeC2O42H2O
(JCPDS 023-0293) having lattice parameters a = 9.921, b = 5.556
and c = 9.707. The absence of impurity phases suggests the ferrous
oxalate to be sufficiently pure. The sharp and strong diffraction
peaks indicate highly crystalline nature of the ferrous oxalate pro-
duced. The average crystallite size of ferrous oxalate was also cal-
culated from the Scherrer’s equation and was found to be about
72 nm.
TG-DTA curves (Fig. 10) for the iron oxalate (FeC2O42H2O) were
also obtained by conducting the thermal analysis of the sample up
1800
1600
Ferrous Oxalate Dihydrate
1400 (PDF: 00-023-0293)
1200
Counts
to 600 °C at a heating rate of 10 °C/min in air atmosphere. The
results obtained are in good agreement with those reported earlier
[42–45]. The TG curve of ferrous oxalate shows a continuous loss of
mass in the temperature range 160–265 °C and further increase in
the temperature had negligible effect on the weight loss. The total
weight loss of
57% corresponds to the formation of Fe2O3 from
the Fe(II)-oxalate dihydrate phase. The weight loss is accompanied
by an endothermic peak at 222 °C in the DTA curve which corre-
sponds to the dehydration/decomposition step. The single step
weight loss suggests that dehydration and decomposition of fer-
rous oxalate occurs simultaneously in the air atmosphere. There-
fore, the decomposition of the ferrous oxalate to iron oxide can
be represented by the following oxidation reaction [46]:
4FeC2 O4  2H2 O þ O2 ! 2Fe2 O3 þ 4CO þ 4CO2 þ 8H2 O ð7Þ
Ferrous oxalate dihydrate was calcined at 400 °C for 4 h to con-
vert it into iron oxide. Powder X-ray diffraction studies confirmed
the formation of rhombohedral hematite viz. a-Fe2O3 phase (JCPDS
01-087-1165) with a lattice parameter of a = 5.035 and c = 13.749
(Fig. 11). No additional peaks were found in the product which can
ascribe for the presence of any identifiable impurities. The appear-
ance of stronger and sharper peaks approves the good crystallinity
of the iron oxide. The average crystallite size calculated to be

46 nm using the Scherrer’s equation. Again the decrease in size
of the iron oxides can be ascribed to the decomposition of the Fe
(II) oxalate dehydrate while evolving CO and CO2 during the calci-
nations process.
The size and morphology of the oxalate/oxide were also deter-

1000
mined using SEM-EDS studies. The SEM image of ferrous oxalate
800 (Fig. 12a) shows that the product is aggregate of cubical columnar
particles. EDS of the ferrous oxalate had shown the presence of Fe,
600 C and O without any impurity elements. The SEM image of a-Fe2O3
particles is shown in Fig. 12b. The figure clearly shows that the
400 sample is composed of mainly small columnar bar like particles
with a few big cubical shaped particles. The average particle size
200 of hematite ranges from 5 lm to 10 lm. EDS (inset of Fig. 12b)
clearly shows that a-Fe2O3 produced is quite pure. EDS quantifica-
0
10 20 30 40 50 60 tion confirmed the presence of Fe and O with the expected ratio of
2:3 and chemical analysis confirms the high purity (>99.9%) of the
2-Theta
as prepared hematite with
69.8% of Fe.
Fig. 9. XRD patterns of ferrous oxalate particles.

110 40
600
FeC2O4.2H2O
100 Hematite
20 (PDF: 04-006-6579)
TG 500
90 DTA 0
400
% Wt. loss

80
µV/mg

Counts

-20
70
300
-40
60
200
-60
50

40 -80 100
100 300 500 10 30 50 70 90
Temperature (oC) 2-Theta
Fig. 10. TG-DTA curves of ferrous oxalate particles. Fig. 11. XRD patterns of iron oxide particles.
 M.K. Sinha et al. / Separation and Purification Technology 167 (2016) 37–44
Fig. 12. SEM images of the (a) ferrous oxalate and (b) iron oxide particles.
4. Conclusions
Spent iron pickle liquor can be considered as both a hazardous
waste and a potential valuable resource. In the present investiga-
tion the high pure zinc oxide and hematite particles were synthe-
sised by combined process of solvent extraction-precipitation
stripping and aqueous reduction. Oxalic acid was used for the
precipitation stripping of zinc oxalate and iron oxalate from the
loaded Cyanex 923 containing 0.76 M Zn (II) and 0.16 M Fe(III).
At a phase ratio (A/O) of 3 both zinc and iron were quantitatively
stripped off which simultaneously regenerated Cyanex 923. The
obtained zinc oxalate precipitate and ferric oxalate solution were
further treated to produce zinc oxide and hematite particles of
nano-size range as the value added products. To produce high
value iron oxide powder, ferric oxalate solution was reduced to fer-
rous oxalate by adding stoichiometric amount of Fe scrap-powder.
The oxalates of zinc and iron(II) were calcined using the informa-
tion gathered from the TG-DTA analysis. To know the purity and
morphology oxalate and oxide of both the metals were well char-
acterized by XRD, SEM and chemical analysis. The current process
development can thus be useful for exploiting the secondary
resource like spent pickle solution to produce the value added
products.
Acknowledgements
The authors are thankful to the Director, CSIR-National Metal-
lurgical Laboratory, Jamshedpur, India for giving permission to
publish the paper. The authors would like to thank Mr. S. Mandal
and Mr. B. Sivakumar for their assistance in TG-DTA analysis.
References
[1] M.B. Mansur, S.D.F. Rocha, F.S. Magalhaes, J.S. Benedetto, Selective extraction
of zinc (II) over iron (II) from spent hydrochloric acid pickling effluents by
liquid – liquid extraction, J. Hazard. Mater. 150 (2008) 669–678.
[2] J. Laso, V. Garc, E. Bringas, A.M. Urtiaga, I. Ortiz, Selective recovery of zinc over
iron from spent pickling wastes by different membrane-based solvent
extraction process configurations, Ind. Eng. Chem. Res. (2015), http://dx.doi.
org/10.1021/acs.iecr.5b00099.
[3] J. Carrillo-Abad, M. García-Gabaldón, V. Pérez-Herranz, Treatment of spent
pickling baths coming from hot dip galvanizing by means of an
electrochemical membrane reactor, Desalination 343 (2014) 38–47.
[4] M. García-Gabaldón, J. Carrillo-Abad, E. Ortega-Navarro, V. Pérez-Herranz,
Electrochemical study of a simulated spent pickling solution, Int. J.
Electrochem. Sci. 6 (2011) 506–519.
[5] J. Carrillo-Abad, M. García-Gabaldón, E. Ortega, V. Pérez-Herranz,
Electrochemical recovery of zinc from the spent pickling baths coming from
the hot dip galvanizing industry: potentiostatic operation, Sep. Purif. Technol.
81 (2011) 200–207.
43
[6] M. Regel-Rosocka, M. Wisniewski, Selective removal of zinc(II) from spent
pickling solutions in the presence of iron ions with phosphonium ionic liquid
Cyphos IL 101, Hydrometallurgy 110 (1–4) (2011) 85–90.
[7] U. Kerney, Treatment of spent pickling acids from hot dip galvanizing, Resour.
Conserv. Recycl. 10 (1994) 145–151.
[8] M. Regel-rosocka, A review on methods of regeneration of spent pickling
solutions from steel processing, J. Hazard. Mater. 177 (1–3) (2010) 57–69.
[9] M. Rozenblat, M. Regel-rosocka, J. Szymanowski, Metal removal from spent
pickling solutions of high zinc (II) concentration, Chem. Eng. 38 (2004) 121–
129.
[10] H. Samaniego, M. San Roman, I. Ortiz, Modelling of the extraction and back-
extraction equilibria of zinc from spent pickling solutions, Sep. Sci. Technol. 41
(4) (2006) 757–769.
[11] G. Csicsovszki, T. Kekesi, T.I. Torok, Selective recovery of Zn and Fe from spent
pickling solutions by the combination of anion exchange and membrane
electrowinning techniques, Hydrometallurgy 77 (2005) 19–28.
[12] A. Devi, A. Singhal, R. Gupta, A review on spent pickling liquor, Int. J. Environ.
Sci. 4 (3) (2013) 284–295.
[13] M.K. Sinha, S.K. Sahu, P. Meshram, B.D. Pandey, Solvent extraction and
separation of zinc and iron from spent pickle liquor, Hydrometallurgy 147–
148 (2014) 103–111.
[14] M. Regel, A.M. Sastre, J. Szymanowski, Recovery of Zinc(II) from HCl spent
pickling solutions by solvent extraction, Environ. Sci. Technol. 35 (2001) 630–
635.
[15] M. Regel-rosocka, A.M. Sastre, J. Szymanowski, Zinc(II) extraction from
hydrochloric acid solutions with basic and solvating extractants, in: XVII-th
ARS SEPARATORIA – Borówno, 2002 (Poland).
[16] M. Regel-rosocka, J. Stelmaszyk, A.M. Sastre, J. Szymanowski, Zinc (II)
extraction-stripping with tributyl phosphate, in: XVIII-th ARS SEPARATORIA
– Zloty Potok, Poland, 2003, pp. 93–95.
[17] R. Cierpiszewski, I. Miesia, M. Regel-rosocka, A.M. Sastre, J. Szymanowski,
Removal of Zinc (II) from spent hydrochloric acid solutions from zinc hot
galvanizing plants, Ind. Eng. Chem. Res. 41 (2002) 598–603.
[18] A. Grzeszczyk, M. Regel-rosocka, Extraction of zinc (II), iron (II) and iron (III)
from chloride media with dibutylbutylphosphonate, Hydrometallurgy 86
(2007) 72–79.
[19] S.J. Cook, J.M. Perera, G.W. Stevens, S.E. Kentish, The screening of extractants
for the separation of Zn(II) from Australian hot-dip galvanizing effluent, Sep.
Sci. Technol. 46 (13) (2011) 2066–2074.
[20] M. Bartkowska, M. Regel-Rosocka, J. Szymanowski, Extraction of zinc(II), iron
(III) and iron(II) with binary mixtures containing tributyl phosphate and di(2-
ethylhexyl)phosphoric acid or Cyanex 302, Physicochem. Probl. Miner.
Process. 36 (2002) 217–224.
[21] A. Agrawal, S. Kumari, K.K. Sahu, Studies on solvent extraction of iron (III) as a
step for conversion of a waste effluent to a value added product, J. Environ.
Manage. 92 (2011) 3105–3111.
[22] S. Nishimura, Production of metallic oxide and metal powders by
crystallization stripping applying solvent extraction, Min. Pro. Ext. Met. Rev.
21 (2008) 143–165.
[23] J.H. Yoon, F.M. Doyle, Precipitation of rare earths powders from aqueous
solutions and emulsions, in: F.M. Doyle, S. Raghavan, P. Somasundara, G.W.
Warren (Eds.), Innovations in Materials Processing Using Aqueous, Colloid and
Surface Chemistry, TMS, Warrendale, PA, 1989, pp. 195–211.
[24] N.E. El-Hefny, Y.A. El-Nadi, H.F. Aly, Synthesis and characterization of
neodymium phosphate powder resulted from neodymium loaded-HDEHP
organic solutions by a stripping-precipitation process using a Lewis type cell,
Sep. Purif. Technol. 78 (2011) 330–335.
[25] D.X. Zhang, H. Xu, Y.Z. Liao, H.S. Li, X.J. Yang, Synthesis and characterisation of
nano-composite copper oxalate powders by a surfactant-free stripping–
precipitation process, Powder Technol. 189 (3) (2009) 404–408.
[26] E. Combes, C. Sella, D. Bauer, J.L. Sabot, A study of the formation of acid and
neutral yttrium oxalate powders from precipitation-stripping reaction in a
Lewis-type cell, Hydrometallurgy 46 (1997) 137–148.
44 M.K. Sinha et al. / Separation and Purification Technology 167 (2016) 37–44
[27] M. Iglesias, E. Anticó, V. Salvadó, A. Masana, M. Valiente, Effect of Y(III)
distribution between aqueous nitrate and organic D2EHPA solutions on the Y
(III) precipitation stripping using oxalic acid, Solvent Extr. Ion Exch. 17 (2)
(1999) 277–300.
[28] Y. Konishi, Y. Noda, S. Asai, Precipitation stripping of yttrium oxalate powders
from yttrium-loaded carboxylate solutions with aqueous oxalic acid solutions,
Ind. Eng. Chem. Res. 37 (1998) 2093–2098.
[29] Y. Konishi, Y. Noda, Precipitation stripping of rare-earth carbonate powders
from rare-earth-loaded carboxylate solutions using carbon dioxide and water,
Ind. Eng. Chem. Res. 40 (2001) 1793–1797.
[30] P.M. Smith, High-purity rare earth oxides produced via precipitation stripping,
Metall. Mater. Trans. B 38 (5) (2007) 763–768.
[31] S. Zielinski, M. Buca, M. Famulski, Precipitation – stripping processes for heavy
metals, Hydrometallurgy 48 (1998) 253–263.
[32] A. Agrawal, K.K. Sahu, Treatment of chloride waste pickle liquor by solvent
extraction for the recovery of iron, Miner. Process. Extr. Metall. Rev. 31 (2010)
121–134.
[33] J. Saji, T. Prasada Rao, C.S.P. Iyer, M.L.P. Reddy, Extraction of iron (III) from
acidic chloride solutions by Cyanex 923, Hydrometallurgy 49 (1998) 289–296.
[34] J.F. Alguacil, S. Martinez, Solvent extraction of Zn(II) by Cyanex 923 and its
application to a solid-supported liquid membrane system, J. Chem. Technol.
Biotechnol. 76 (2001) 298–302.
[35] R.K. Mishra, P.C. Rout, K. Sarangi, K.C. Nathsarma, A comparative study on
extraction of Fe(III) from chloride leach liquor using TBP, Cyanex 921 and
Cyanex 923, Hydrometallurgy 104 (2010) 298–303.
[36] B. Gupta, A. Deep, P. Malik, S.N. Tandon, Extraction and separation of some 3d
transition metal ions using Cyanex 923, Solvent Extr. Ion Exch. 20 (2002) 81–
96.
[37] I. Miesiac, J. Szymanowski, Separation of Zinc(II) from hydrochloric acid
solutions in a double lewis cell, Solvent Extr. Ion Exch. 22 (2004) 243–265.
[38] H.L. Clever, M.E. Derrick, S.A. Johnson, The solubility of some sparingly soluble
salts of zinc and cadmium in water and in aqueous electrolyte solutions, J.
Phys. Chem. Ref. Data 21 (1992) 941–1004.
[39] B.D. Cullity, Elements of X-ray Diffraction, second ed., Addison-wesley,
Reading, MA, 1978.
[40] T.V. Barinova, I.P. Borovinskaya, Solution-combustion synthesis of nanosized
iron oxide from ferric oxalate, Int. J. Self Propag. High Temp. Synth. 21 (2012)
1–6.
[41] Y. Yang, X. Wang, M. Wang, H. Wang, P. Xian, Hydrometallurgy recovery of
iron from red mud by selective leach with oxalic acid, Hydrometallurgy 157
(2015) 239–245.
[42] D. Dollimore, D.L. Griffiths, D. Nicholson, The thermal decomposition of
oxalates. Part II. Thermogravimetric analysis of various oxalates in air and in
nitrogen, J. Chem. Soc. 2617–2623 (1963).
[43] R.A. Brown, S.C. Bevan, The thermal decomposition of ferrous oxalate
dihydrate, J. Inorg. Nu¢l. Chem. 28 (1966) 387–391.
[44] D. Broadbent, D. Dollimore, J. Dollimore, The thermal decomposition of
oxalates. Part IX. The thermal decomposition of the oxalate complexes of iron,
J. Chem. Soc. A (1967) 451–454.
[45] E.D. Macklen, Influence of atmosphere on the thermal decomposition of
ferrous oxalate dihydrate, J. Inorg. Nucl. Chem. 29 (1967) 1229–1234.
[46] A. Angermann, J. Topfer, Synthesis of magnetite nanoparticles by thermal
decomposition of ferrous oxalate dihydrate, J. Mater. Sci. 43 (2008) 5123–
5130.