Reductive leaching of manganese and zinc from spent alkaline and zinc–carbon
batteries in acidic media
, T. Kukrer
, F. Ferella
, A. Akcil
, F. Veglio
, M. Kitis
Department of Environmental Engineering, Suleyman Demirel University, 32260 Isparta, Turkey
Department of Chemistry, Chemical Engineering and Materials, University of L'Aquila, 67040 Monteluco di Roio, L'Aquila, Italy
Department of Mining Engineering, Mineral Processing Division, Suleyman Demirel University, 32260 Isparta, Turkey
Received 20 November 2008
Received in revised form 19 January 2009
Accepted 19 January 2009
Available online 29 January 2009
The main purpose of this work was to investigate the effectiveness of oxalic acid as a reductant for the
leaching of manganese and zinc from spent alkaline and zinc–carbon batteries in sulfuric acid or hydrochloric
acid media. Three different types of battery powders were tested: zinc–carbon, alkaline and mixed (50%
zinc–carbon, 50% alkaline). Kinetic tests were initially conducted with the mixed battery powder. Leaching
experiments were carried out according to 2
full factorial design, and regression equations for the leaching
of Mn and Zn were determined. Washing of the powders (neutral leaching) was effective on the removal of
potassium and chloride. Increasing solid/liquid ratio from 1/5 to 1/10 in neutral leaching did not
signiﬁcantly change potassium and chloride removal. A leach duration of 3 h was found to be generally
sufﬁcient for the equilibrium to be reached for both Zn and Mn. Oxalic acid concentration had strongest
negative effect on Zn leaching in both sulfuric and hydrochloric acid media, whereas the concentrations of
sulfuric and hydrochloric acids exhibited strongest positive effect for both Mn extraction yield (MnEY) and
Zn extraction yield (ZnEY). In the range of tested conditions, pulp density had no important effect on MnEY
and ZnEY for both acids. Temperature had negative effect for both MnEY and ZnEY in sulfuric acid solution;
however, such effect was less pronounced in hydrochloric acid solution. For the sulfuric acid solution, 91%
MnEY and 112% ZnEY were achieved at 45 °C after 3 h of leaching by 10% pulp density, −30% oxalic acid (30%
less than the stoichiometric requirement), +30% H
. For the hydrochloric acid solution, about 86% MnEY
and 95% ZnEY were obtained at 20% pulp density, −30% oxalic acid, +30% HCl, at 45 °C after 3 h of leaching.
© 2009 Elsevier B.V. All rights reserved.
The zinc–carbon and alkaline batteries are non-rechargeable
batteries (primary cells), which are designed to be fully discharged
only once, and then discarded (Rayovac Corp., 1999). Usually, they run
out rapidly and are thrown away; they are a special residue because
of the mercury, zinc, manganese, and other heavy metals which they
contain (Bartolozzi, 1990). This presents a major environmental and
health threat (Kierkegaard, 2007).
Due to increasing environmental concerns and raw material
consumptions, worldwide stringent regulations are being imposed
for spent batteries. All these regulations and environmental issues
result in a driving force for the collection of spent batteries and
recovery and further reuse of metals. In Turkey, the regulation on the
Control of Spent Battery and Accumulators was published on August
2004. To adjust the regulation to those of European Union (EU),
such regulation was modiﬁed on March 2005 (Turkish Ministry of
Environment and Forestry, 2005). According to this regulation, the use
of spent batteries in a way which may harm the environment and their
direct or indirect discharge to the environment are all prohibited.
Legal applications about collection, transportation, recovery, and
disposal of waste batteries and accumulators are well deﬁned in this
At present, the collection, treatment and recycling of waste
batteries in Europe are patchy. For example, only six of the 25 EU
countries have collection systems for spent batteries. Belgium collects
59% of all spent batteries, Sweden 55%, Austria 44%, Germany 39%, the
Netherlands 32%, and France 16%. Almost half of all portable batteries
sold in the EU-15 Member States in 2002 were eventually incinerated
or disposed of in landﬁlls (Meller, 2006; Kierkegaard, 2007). In Turkey,
while there has been no major recovery applications for spent
batteries yet, a total of 13 companies have already received accu-
mulator recovery licenses. The spent batteries are generally not
separately collected in Turkey, thus, they are landﬁlled in municipal
landﬁlls with other solids wastes. For example, only about 2% of spent
batteries have been collected in Turkey (Erdem, 2007). In the long
term, landﬁlling of spent batteries is not an effective disposal method
mainly due to limited storage capacities, hazardous waste issues,
increasing landﬁll costs, and the increasing need for metal recoveries.
In the last two decades, several processes have been studied and
developed to recycle batteries and recover metals such as Zn, Mn, Fe,
Hydrometallurgy 97 (2009) 73–79
⁎Corresponding author. Tel.: +90 246 2111289; fax: +90 246 237 0859.
E-mail address: email@example.com (M. Kitis).
0304-386X/$ –see front matter © 2009 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/hydromet
Li, Co, and Ni as a result of new strict regulations in many countries
around the world (Hurd et al., 1993; Fröhlic and Sewing, 1995; De
Souza et al., 2001; Bernardes et al., 2003, 2004; Li and Xi, 2005; De
Michelis et al., 2007; Ferella et al., 2008). Most of the battery contents
can be technically recovered by means of mechanical and chemical
treatment. The recovered materials can be reused in battery produc-
tion or for other purposes. The physical recovery processes for spent
batteries usually consist the following sequential steps: sorting,
magnetic separation, dismantling, and grinding. These kinds of pre-
treatment steps are required in order to improve metal dissolution
rates in the aqueous phase (Salgado et al., 2003). Pyrometallurgical
processes are widely used for the recovery of metals from spent
batteries. However, the gas emissions due to incineration at high
temperatures (above 600 °C) and high energy demands are the major
drawbacks of pyrometallurgical processes. Hydrometallurgical pro-
cesses on the other hand have some inherent advantages including
relatively simpler operation, less energy demand, and no gas
emissions. However, some pre-treatment steps are required to
improve metal dissolution rates in the aqueous phase, such as battery
classiﬁcation, dismantling, magnetic separation, and leaching (Sal-
gado et al., 2003; Karakaya et al., 2007). Pre-treatment steps have
become more common through improved technologies; however,
these technologies require extra investment and operational costs.
De Souza et al. (2001) described an acidic leaching process without
reducing agent and they found MnEY as 30%. At similar conditions,
Salgado et al. (2003) found MnEY as 40%. The investigations with
reducing agents showed better results for leaching of Mn. Sahoo et al.
(2001) found that 98.4% of Mn was extracted from low-grade
manganese ore with 30.6 g/L oxalic acid, 0.543 M sulfuric acid
concentration at 85 °C in 105 min. De Michelis et al. (2007)
investigated similar medium for recovery of zinc and manganese
from alkaline and zinc–carbon spent batteries. They found that the
highest manganese (70%) and zinc (100%) extractions were obtained
in the following reductive acid leaching conditions: 20% pulp density,
1.8 M sulfuric acid, 59.4 g/L oxalic acid, 5 h of leaching at 80 °C.
Avraamides et al. (2006) proposed a method for the leaching of spent
zinc–carbon battery scraps by sulfur dioxide. It was shown that over
90% zinc and manganese could be leached in 20–30 min at 30 °C using
0.1–1.0 M sulfuric acid in the presence of sulfur dioxide.
The main objective of this study was to investigate the recovery of
zinc and manganese from spent I. group batteries (alkaline and zinc–
carbon) on lab-scale employing hydrometallurgical leaching by oxalic
acid as the reductant in sulfuric acid or hydrochloric acid media. Oxalic
acid was selected due to its high efﬁciency in the reductive dissolution
of Mn (IV). Reductive acidic leaching tests with H
or HCl were
designed according to the full factorial design (4 factors, 2 levels) with
4 central point tests (Montgomery,1991). The four factors tested were
pulp density, acid concentration (H
or HCl), oxalic acid concen-
tration, and temperature. The following reactions were considered for
the dissolution of zinc and manganese oxides from battery powders.
The required stoichiometric dosages of H
, HCl and oxalic acid
were calculated in deference to these reactions. Zinc oxide and
manganese oxide can be fully leached by sulfuric and hydrochloric
acid media according to the Eqs. (1)–(4) (Sahoo et al., 2001; Li and Xi,
2005; De Michelis et al., 2007).
However, the dissolution of Mn
is partial due to the
formation of MnO
according to Eqs. (5)–(8) (Sahoo et al., 2001; Li
and Xi, 2005; De Michelis et al., 2007).
Therefore, to improve the leaching efﬁciency of MnO
, a reducing
agent like oxalic acid is required. As shown in Eqs. (9)–(10), it is
possible to leach MnO
with the use of oxalic acid (Sahoo et al., 2001;
Li and Xi, 2005; De Michelis et al., 2007).
2. Materials and methods
2.1. Pretreatment of alkaline and zinc–carbon batteries
Spent alkaline and zinc–carbon batteries were collected from
Suleyman Demirel University Campus in the City of Isparta, Turkey
and a volunteer foundation in the City of Istanbul, Turkey. Most of
these spent batteries were D size with about 10% of A size. Spent
alkaline and zinc–carbon batteries were ﬁrst separated and then
manually dismantled. Dismantling products such as plastic ﬁlms,
ferrous scraps and paper pieces were discarded. After dismantling, all
battery components were separated and weighed. Table 1 shows the
percent mass of each component in spent batteries.
The powders, which were about 40–64% of the total weight of
dismantled batteries, were dried for 24 h at 105 °C. The powder
samples were crushed using a jaw crusher (Fritsch) and then ground
using a ball mill (Fritsch) for particle size reduction. Afterwards the
powder samples were manually sieved to obtain particles with size
b425 μm. All leaching tests were carried out using this particle size
fraction. Both zinc–carbon and alkaline battery powders were washed
with distilled and deionized water (DDW) at different solid/liquid
ratios (1:5 and 1:10) by mixing at 200 rpm, at 80 °C, for 1 h. By testing
different solid/liquid ratios, optimum washing conditions (neutral
leaching) were determined. Solution pH values were measured after
neutral leaching of each sample.
The distribution of different components (% mass) in spent alkaline and zinc–carbon
Components Zinc–carbon Alkaline
Battery A Battery B Battery C Battery D
Size D Size A Size D Size A Size D Size D
Paper 0.96 3.21 1.69 5.80 2.24 1.15
Steel 18.30 35.79 24.06 30.48 23.76 11.66
Plastic 3.08 2.66 2.20 2.87 2.11 1.27
Carbon rod 6.12 5.85 5.20 6.86 4.54 0.00
Powder 64.46 46.33 52.91 49.19 52.47 40.76
Moisture 5.80 4.93 13.61 1.50 13.91 2.71
Grey paste ––––– 41.81
Loss during dismantling 1.28 1.24 0.34 3.31 0.97 0.64
Batteries A, B, C, and D represent different brands.
74 E. Sayilgan et al. / Hydrometallurgy 97 (2009) 73–79
2.2. Powder characterization and neutral leaching
Both original (unwashed) and washed zinc–carbon,alkaline, mixed
(50% zinc–carbon, 50% alkaline) battery powders were analyzed by X-
ray ﬂuorescence (XRF) (Spectro Xepos). Table 2 shows the elemental
composition of each powder. Leaching tests with mixed alkaline and
zinc–carbon battery powders were also conducted to simulate full-
scale applications since it is challenging to separate such batteries in
full-scale recovery applications.
2.3. Acidic leaching tests
factorial design with replicated central point tests was chosen
for conducting the leaching tests where the factors were pulp density,
oxalic acid concentration, acid (H
or HCl) concentration, and
temperature. The factors and levels investigated are given in Table 3.
Acidic leaching tests were performed in 250 mL high-density
polyethylene ﬂasks (solution volume 100 mL) which were mixed at
200 rpm in a temperature-controlled water bath (ST402, Nuve).
Required quantities of acid (H
or HCl) and oxalic acid to be dosed
to the ﬂasks were calculated according to the 2
full factorial design.
The ﬂasks were loosely covered with caps to allow carbon dioxide
release which is formed during the chemical reactions. Kinetic tests
were initially conducted to determine the impacts of reaction time on
the leaching efﬁciency. Tested leach durations were 0.5, 1, 2, 3, 4, and
5 h. After the kinetic tests, further leach tests were conducted. During
these leach tests, 1.5 mL of leach liquor sample was withdrawn from
ﬂasks after 1, 3 and 5 h to determine the concentrations of Mn and Zn.
Before the analysis for Mn and Zn, such liquor samples were
centrifuged at 10,000 rpm for 5 min (MiniSpin Plus, Eppendorf).
Each sample was diluted by 1:10 using nitric acid solution (pH~2) to
avoid the precipitation of metals and then samples were kept at 4 °C in
the fridge until analysis. Zn and Mn concentrations were determined
by inductively coupled plasma (ICP-OES) (DV2100, Perkin Elmer).
Solution pHs of the leach samples were also measured using a pH-
meter (6250, Jenco). The solid residues remaining after centrifugation
(containing mainly graphite and non-dissolved compounds) were
dried in the oven (FN 500, Nuve) at 105 °C for 24 h, which was found
to be sufﬁcient duration to achieve constant weight. The dried solid
residues were then weighed to determine the mass losses during
A 98% sulfuric acid and a 37% hydrochloric acid solution (Merck)
were used as the stock solutions for all experiments. Oxalic acid
O) (Merck) was used as the reducing agent. DDW was
used for stock solution preparations and dilutions.
3. Results and discussion
3.1. Characterization of spent batteries and powders
The percent mass distribution of different components in spent
batteries is shown in Table 1. The powders obtained from zinc–carbon
batteries ranged between 49 and 64% of the total mass depending on
different brands. For the alkaline battery, powder content was about
40%. Similarly, Rabah et al. (1999) found that the powder contents of
dry battery cells were nearly 40%. Steel contents of all batteries ranged
between 12 and 36%. Grey paste constituted about 42% of the alkaline
battery, while no grey paste was present in zinc–carbon batteries.
Grey paste in alkaline batteries mainly contains electrolyte (KOH).
The XRF analysis indicated that manganese and zinc were the
predominant metals in the powders of alkaline and zinc–carbon
batteries. Table 2 compares the quantitative metal contents of the
spent alkaline, zinc–carbon and mixed (50% zinc–carbon, 50% alka-
line) battery powders. The results for the mixed powders showed Zn
and Mn contents were about 10 and 40% of the total powder mass,
respectively, consistent with the literature (De Souza et al., 2001; De
Souza and Tenorio, 2004; Veloso et al., 2005; Avraamides et al., 2006;
De Michelis et al., 2007). Zinc–carbon batteries have zinc as anode,
manganese dioxide as cathode, and ammonium chloride and/or zinc
chloride as electrolyte. Alkaline batteries only differ from zinc–carbon
batteries by the use of potassium hydroxide as electrolyte (Rayovac
Corp., 1999). Washing of the powders (neutral leaching) was very
effective for the removal of potassium and chloride. This ﬁnding was
valid for both alkaline and zinc–carbon battery powders and the mixed
powder. In terms of the impact of solid/liquid ratio on neutral leaching
effectiveness, it was found in the preliminary tests that increasing
solid/liquid ratio from 1/5 to 1/10 did not signiﬁcantly change
potassium and chloride removals. Therefore, all acid leaching tests
were carried out after employing neutral leaching with 1/5 solid/
liquid ratio. MW2 powder (Table 2) was used in all acid leaching tests.
3.2. Kinetic experiments
Kinetic leaching experiments were conducted with the mixed
battery powder to evaluate the effectiveness of H
and HCl as the
acid media. Furthermore, different reducing agents (oxalic acid,
ascorbic acid, citric acid) were also tested in both of the acidic
media. For all kinetic tests, pulp density, mixing rate and leaching
temperature were kept constant at 15%, 200 rpm and 60 °C,
respectively. Required stoichiometric quantities of acids and reducing
agents calculated from Eqs. (1)–(10) were dosed for all kinetic
leaching tests. All kinetic tests were performed using completely
mixed batch reactors (CMBR) (250 mL ﬂasks) and the CMBRs were
sampled at leach durations of 0, 0.5, 1, 2, 3, 4, and 5 h.
Fig. 1 shows the leaching kinetic results for different acid and
reducing agent types. Extraction yields higher than 100% are mainly
due to inherent analytical measurement errors. Percent coefﬁcient of
variations for the triplicate measurements of Zn and Mn concentra-
tions were generally less than ±6%. The results indicated that ZnEY
Characterization of spent alkaline and zinc–carbon battery powders by XRF (% mass).
Sample Al (%) Si (%) K (%) Cr (%) Mn (%) Fe (%) Zn (%) Cl (%) Ti (%)
AU1 0.22 0.06 5.56 b0.0033 48.66 0.05 2.35 0.03 0.51
ZCU1 0.44 1.35 0.15 b0.0 028 26.60 1.58 13.24 4.26 0.01
MU1 0.28 0.73 1.98 0.003 39.31 0.82 8.50 1.97 0.25
AW2 b0.10 0.06 1.48 b0.0 033 53.01 0.05 2.56 0.0 09 0.49
ZCW2 0.51 1.48 0.13 0.003 30.27 1.77 12.16 1.13 0.008
MW2 0.34 0.91 0.22 b0.0029 41.90 0.88 9.63 0.34 0.25
MW3 0.33 0.82 0.83 b0.0030 42.31 0.91 8.34 0.61 0.26
AU1: Unwashed alkaline battery powder.
ZCU1: Unwashed zinc–carbon battery powder.
MU1: Unwashed mixed (50% zinc–carbon, 50% alkaline) battery powder.
AW2: Washed alkaline battery powder (solid/liquid ratio: 1/10).
ZCW2: Washed zinc–carbon battery powder (solid/liquid ratio: 1/10).
MW2: Washed mixed (50% zinc–carbon, 50% alkaline) battery powder (solid/liquid
MW3: Washed mixed (50% zinc–carbon, 50% alkaline) battery powder (solid/liquid
Factors and levels investigated in leaching tests (acid: H
Code Factor (variable) Level
A Pulp density (%) 10 15 20
B Oxalic acid (reducing agent) concentration
(0%: stoichiometric requirement)
C Acid (0%: stoichiometric requirement) −30%
D Temperature (°C) 45 60 75
30% less than the stoichiometric requirement.
30% more than the stoichiometric requirement.
75E. Sayilgan et al. / Hydrometallurgy 97 (2009) 73–79
and MnEY did not generally increase after 1 h and 3 h of leaching time,
respectively. In other words, as a general trend, a leach duration of 3 h
was found to be sufﬁcient for the leaching equilibrium tobe reached for
both Zn and Mn. Thus, the leach duration was chosen as 3 h for the
following leaching tests. The best extraction yields were generally
achieved by the use of ascorbic acid and HCl. ZnEY and MnEY were
about 95–100% and 65–95%, respectively, when ascorbic acid was used
as the reducing agent in both of the acidic media. Similar results were
found for the citric acid in kinetic tests. Both the MnEY and ZnEY
decreased in excess oxalic acid with H
and HCl acid solutions;
practically 70–80% of the manganese was leached with H
while zinc was leached only around 20–30%. This result is expected due
to the formation of zinc-oxalate precipitates. Similarly, precipitation of
Zn with oxalic acid was also observed in synthetic model solutions.
This ﬁnding is also consistent with our previous work (De Michelis et
al., 2007), in which both manganese and zinc precipitation increased
with oxalic acid concentration. In that work, when oxalic acid
concentration was 100 g/L, a complete precipitation of zinc (98%)
was achieved together with a partial precipitation (38%) of manga-
nese. This difference is due to different solubility of the two oxalates:
the solubility product (Ksp at 25 °C) of MnC
is 1.70× 10
, while it
is 1.38× 10
(De Michelis et al., 2007). Thus, manganese
oxalate is more soluble than zinc oxalate and precipitation phenomena
especially for zinc might take place also during battery powder
The results overall revealed that the both acids (H
were very effective in leaching of Mn and Zn from spent battery
powders, consistent with the literature (El-Nadi et al., 2007). Similar
leaching effectiveness was found for both acid types. Furthermore, the
results indicated that while manganese leaching was also not affected
by the type of acid as for the Zn, it was strongly dependent on the type
of reducing agent. Therefore, the selection of reducing agent is critical
in full-scale applications for the recovery of Zn and Mn from spent
alkaline and zinc–carbon batteries.
Fig. 1. Leaching kinetics of Zn and Mn from mixed battery powder using H
HCl (B). OA: oxalic acid, AA: ascorbic acid, CA: citric acid.
Leaching test results with H
(A) and HCl (B).
B (oxalic acid)
D (temperature, °C) MnEY
110 −30 −30 45 77.5 85.4
220 −30 −30 45 68.4 70.8
310 30 −30 45 67.2 7.1
420 30 −30 45 68.9 9.0
510 −30 30 45 91.2 112.5
620 −30 30 45 82.0 104.5
7 10 30 30 45 80.3 8.0
8 20 30 30 45 80.5 8.0
910 −30 −30 75 35.3 28.1
10 20 −30 −30 75 45.8 27.0
11 10 30 −30 75 50.1 5.8
12 20 30 −30 75 61.5 7.7
13 10 −30 30 75 80.8 101.7
14 20 −30 30 75 82.9 97.3
15 10 30 30 75 77.8 9.5
16 20 30 30 75 69.8 8.2
17 15 0 0 60 77.8 32.3
18 15 0 0 60 76.6 31.3
19 15 0 0 60 76.9 31.5
20 15 0 0 60 76.2 32.3
21 5 0 0 60 7 1.4 11.3
22 25 0 0 60 81.3 25.1
23 15 −60 0 60 25.9 53.0
24 15 60 0 60 4 9.2 3.7
25 15 0 −60 60 36.4 7.3
26 15 0 60 60 98.5 32.8
27 15 0 0 30 58.2 24.9
28 15 0 0 90 92.4 91.9
B (oxalic acid)
D (temperature, °C) MnEY
110 −30 −30 45 66.1 50.4
220 −30 −30 45 69.1 58.4
310 30 −30 45 63.8 8.4
420 30 −30 45 65.7 10.5
510 −30 30 45 84.0 86.9
620 −30 30 45 85.6 95.0
7 10 30 30 45 79.9 15.2
8 20 30 30 45 68.7 14.4
910 −30 −30 75 66.7 37.2
10 20 −30 −30 75 57.6 52.0
11 10 30 −30 75 66.3 3.2
12 20 30 −30 75 52.1 7.2
13 10 −30 30 75 80.8 88.6
14 20 −30 30 75 79.3 95.2
15 10 30 30 75 74.2 4.8
16 20 30 30 75 81.5 5.5
17 15 0 0 60 82.9 43.8
18 15 0 0 60 81.8 45.2
19 15 0 0 60 82.8 49.9
20 15 0 0 60 83.8 49.0
21 5 0 0 60 88.7 16.2
22 25 0 0 60 64.0 58.2
23 15 −60 0 60 55.2 92.4
24 15 60 0 60 55.8 4.9
25 15 0 −60 60 48.4 7.0
26 15 0 60 6 0 91. 0 57. 5
27 15 0 0 30 81.4 48.6
28 15 0 0 90 92.9 57.6
Leach duration: 3 h.
Percent oxalic acid and sulfuric acid dosages with respect to their stoichiometric
requirement. (−) indicates less than stoichiometric requirement, (+) indicates more
than stoichiometric requirement.
Percent oxalic acid and hydrochloric acid dosages with respect to their
stoichiometric requirement. (−) indicates less than stoichiometric requirement,
(+) indicates more than stoichiometric requirement.
76 E. Sayilgan et al. / Hydrometallurgy 97 (2009) 73–79
3.3. Leach experiments
Table 4 shows the leaching test results obtained by H
based on the full factorial design. Tests were numbered in conformity
with Yates algorithm (Montgomery, 1991). Extraction efﬁciency of
about 91% Mn and 112% Zn was achieved under the following
conditions: pulp density 10%, oxalic acid dosage less than 30% of its
dosage more than 30% of its stoichiometry, and
temperature 45 °C (Test No. 5 in Table 4A). Similarly, extraction of Mn
and Zn with HCl was 84% and 87%, respectively (Test No. 5 in Table 4B).
Extraction efﬁciencies larger than 100% are mainly due to inherent
errors in experiments and analytical measurements. Reaction tem-
peratures ranging from 30 to 90 °C had signiﬁcant effects on the
dissolution of zinc with H
, consistent with the ﬁndings of Abdel-
Fig. 2 shows the main and interaction effects on the MnEY and
ZnEY values obtained by H
leaching. The estimation of the main
and interaction effects was performed by ANOVA using Yates method.
The main effect of the pulp density (A) did not inﬂuence the MnEY
whereas it had a small and negative inﬂuence on ZnEY. This ﬁnding
suggests that no mixing limitations were present in the leaching
experimental system. The main effect of the oxalic acid concentration
(B) practically did not enhance the MnEY, whereas it had a very strong
negative effect on the ZnEY. The presence of oxalic acid permits the
chemical reduction and then the solubility of MnO
; but at the same
time can produce a precipitation of Mn-oxalate. In terms of impacts on
ZnEY, the presence of oxalic acid produces chemical precipitation as
Zn-oxalate (De Michelis et al., 2007). A suitable compromise should be
found to improve both MnEYand ZnEY. Overall, the highest Mn and Zn
extractions were obtained at test no. 5 with sulfuric acid (Table 4A).
The main effect of the sulfuric acid concentration (C) was found to
be positive for both MnEY and ZnEY. On the other hand, the main
effect of the temperature (D) was negative for both MnEY and ZnEY.
This result may be due to the fact that there is a complex network of
chemical reactions in the leaching process including chemical
precipitation of Mn and Zn oxalate. While metal dissolution rates
may increase with increasing temperatures, chemical precipitation
rates also increase with the temperature. Naik et al. (2000) pointed
out that reductive leaching of MnO
was exothermic and they carried
out leaching tests at room temperature. El-Nadi et al. (2007) indicated
that a slight increase in Zn and Mn leaching was obtained when the
temperature was increased from 25 to 50 °C. However, no further
improvement in leaching effectiveness was found at higher
The most important interactions were the effects of BC (negative)
and CD (positive) for MnEY whereas for ZnEY the important
interactions were the effects of BC (negative) and BD (positive). The
interaction effects indicate that the mechanisms present in this
leaching system are complex and that it is necessary to include them
in the response surface methodology (RSM). Interactions effects on
MnEY and ZnEY can be determined by suitable empirical models
Further leach tests were added to the initial full factorial design
and conducted (tests 21–28, Table 4). Using the data of all leach tests,
the regression equations for MnEY and ZnEY were developed and
main interaction coefﬁcients were tested for signiﬁcance at 95%
conﬁdence level. The generic empirical model can be described by the
Y: the percentage of the metals (Zn, Mn) extracted; a: empirical
model coefﬁcients; X
: dimensionless coded factors for pulp
density, oxalic acid concentration, sulfuric or hydrochloric acid
concentration, and temperature, respectively. The relations between
the coded and actual values are given as:
The experimental results were evaluated utilizing Minitab 15
Statistical Software. Multiple regression analysis of the experimental
data provided the following equations for MnEY and ZnEY with H
(Eqs. (16) and (17)). These empirical equations are suitable to predict
the leaching behavior in the range of the studied experimental
conditions for process optimization purposes.
The comparisons among the experimental extraction yields of Mn
and the estimated ones by regression models are shown in the scatter
diagram (Fig. 3). The results suggested that there is a good agreement
between predicted and measured MnEY. Regarding Zn and Mn
dissolution by oxalic acid and HCl, Fig. 4 shows the main factors and
interaction effects. The percentage of pulp density (A) had little effect
in the investigated ranges. Therefore, it may be suggested that acidic
leaching may be carried out with 20% pulp density in full-scale
Fig. 2. Main and interaction effects on the MnEY (A) and ZnEY (B) values obtained by
leaching with oxalic acid.
77E. Sayilgan et al. / Hydrometallurgy 97 (2009) 73–79
applications to prevent the excessive usage of water. Oxalic acid
concentration (B) had a negative effect (−5%) on the dissolution of
manganese. This is probably due to more precipitation of Mn-oxalate
with the excess oxalic acid addition. Similarly, De Michelis et al.
(2007) found that a partial precipitation of Mn (38%) was observed
after adding 100 g/L oxalic acid. The effect of oxalic acid on the
dissolution of Zn was extremely negative (−60%). As discussed
before, this is due to the formation of Zn-oxalate precipitates.
HCl concentration (factor C) had positive effects both for MnEY and
ZnEY in the investigated conditions. The tested HCl concentrations all
provided high leaching yields for zinc. The main effect of temperature
(factor D) did not inﬂuence MnEY and ZnEY. The interaction effect of
oxalic acid and HCl (BC) was negative (−20%) since the negative
effect of the oxalic acid overcame the positive effect of HCl.
Regression analysis was performed to obtain a quadratic response
surface model and the following equations were obtained for HCl.
Regression equations and coefﬁcients (R
) were evaluated to test
the ﬁt of the model. The value of the determination coefﬁcients
indicated that the variation of 90–95% for extraction of Mn and Zn
with HCl was attributed to the independent variables and only 5–10%
of the total variations were not explained by the model. The scatter
diagram of the experimental extraction yields and the calculated
extraction yields of Mn by HCl is shown in Fig. 5. The predicted values
of both MnEY and ZnEY obtained using Eqs. (18) and (19) were
comparable to the experimental values, indicating that the models for
HCl are also applicable for predicting leaching effectiveness.
Neutral leaching of the powders with distilled and deionized water
was effective in removing potassium and chloride. Increasing solid/
liquid ratio from 1/5 to 1/10 in neutral leaching did not signiﬁcantly
change potassium and chloride removals. A leach duration of 3 h was
found to be generally sufﬁcient for the maximum extraction of both Zn
and Mn. Pulp density did not exhibit any signiﬁcant effect on both
MnEY and ZnEY in the investigated range of operating conditions for
both acids. It is suggested that acidic leaching of battery powders may
be carried out with 20% pulp density in full-scale applications to
prevent the excessive usage of water. Temperature had a negative
effect on both MnEY and ZnEY in sulfuric acid solution; however, such
effect was less pronounced in hydrochloric acid solution.
For the sulfuric acid solution, 91% MnEY and 112% ZnEY were
achieved at 45 °C after 3 h of leaching at 10% pulp density, −30% oxalic
acid, +30% H
. For the hydrochloric acid solution, 86% MnEY and
95% ZnEY were obtained at 20% pulp density, −30% oxalic acid, +30%
HCl, at 45 °C after 3 h of leaching. RSM analysis indicated that both
and HCl had comparable effects on MnEY and ZnEY, suggesting
that both of these acids are effective in leaching of Zn and Mn from
spent zinc–carbon and alkaline batteries. MnEY and ZnEY were
strongly inﬂuenced by the oxalic acid concentration. Excessive oxalic
acid dosages result in the formation of Mn- or Zn-oxalate precipitates,
which further reduces leaching effectiveness. This is important
especially for Zn since excessive oxalic acid dosages may signiﬁcantly
decrease ZnEY even to b10% levels. Therefore, the selection of the
Fig. 3. Scatter diagram of the experimental MnEY versus the calculated MnEY (H
and oxalic acid, 3 h of leach duration).
Fig. 4. Main and interaction effects on the MnEY (A) and ZnEY (B) values obtained by
HCl leaching with oxalic acid.
Fig. 5. Scatter diagram of the experimental MnEY versus the calculated MnEY (HCl and
oxalic acid, 3 h of leach duration).
78 E. Sayilgan et al. / Hydrometallurgy 97 (2009) 73–79
dosage of oxalic acid is very critical if this acid is to be used in full-scale
Obtained empirical regression models for both acids were
successful in predicting MnEY and ZnEY values, suggesting that they
may be practically used in leaching applications of spent alkaline
and zinc–carbon batteries. The efﬁciency of other reducing agents
(ascorbic acid, citric acid) on the reductive leaching of Mn and Zn from
spent alkaline and zinc–carbon battery powders will be evaluated in
This work was supported by research grants from the Scientiﬁc and
Technical Research Council of Turkey (TUBITAK) (project no. 108Y018)
and Research Projects Funding Unit of the Suleyman Demirel
University (project no. BAP 1489-D-07). The authors would like to
thank Fabiola Ferrante for conducting the XRF analysis, Ida De Michelis
for her technical assistance during ANOVA analysis.
Abdel-Aal, E.A., 2000. Kinetics of sulfuric acid leaching of low-grade zinc silicate ore.
Hydrometallurgy 55, 247–254.
Avraamides, J., Senanayake, G., Clegg, R., 2006. Sulfur dioxide leaching of spent zinc–
carbon-battery scrap. J. Power Sources 159, 1488–1493.
Bartolozzi, M., 1990. The recovery of metals from spent alkaline-manganese batteries: a
review of patent literature. Resour. Conserv. Recycl. 4, 233–240.
Bernardes, A.M., Espinosa, D.C.R., Tenorio, J.A.S., 2003. Collection and recycling of
portable batteries: a worldwide overview compared to the Brazilian situation.
J. Power Sources 124, 586–592.
Bernardes, A.M., Espinosa, D.C.R., Tenorio, J.A.S., 20 04. Recycling of batteries: a review of
current processes and technologies. J. Power Sources 130, 291–298.
De Michelis, I., Ferella, F., Karakaya, E., Beolchini, F., Veglio, F., 2007. Recovery of zinc and
manganese from alkaline and zinc–carbon spent batteries. J. Power Sources 172,
De Souza, C.C.B.M., Tenorio, J.A.S., 2004. Simultaneous recovery of zinc and manganese
dioxide from household alkaline batteries through hydrometallurgical processing.
J. Power Sources 136, 191–196.
De Souza, C.C.B.M., Oliveira, D.C., Tenorio, J.A.S., 2001. Characterisation of used alkaline
batteries powder and analysis of zinc recovery by acid leaching. J. Power Sources
El-Nadi, Y.A., Daoud, J.A., Aly, H.F., 2007. Leaching and separation of zinc from the black
paste of spent MnO2–Zn dry cell batteries. J. Hazard. Mater. 143, 328–334.
Erdem, M., 2007. Directive of the control of spent battery and accumulators, Ministry of
Environment and Forestry.1. APAK (Controls of Waste Batteries and Accumulators)
National Symposium, Istanbul, Turkey, November 21, 2007 (in Turkish).
Ferella, F., De Michelis, I., Veglio, F., 2008. Process for the recycling of alkaline and zinc–
carbon spent batteries. J. Power Sources 183 (2), 805–811.
Fröhlic, S., Sewing, D., 1995. The Batenus process for recycling mixed battery waste.
J. Power Sources 57, 27–30.
Hurd, D.J., Muchnick, D.M., Schedler, M.F., 1993. Recycling of dry cell batteries. Pollution
Technology (Review Number 213). Noyes Data Corporation.
Karakaya, E., Kukrer,T., Ferella, F., Michelis, I., Beolchini, F., Veglio, F., Akcil, A.U., Kitis, M.,
2007. Recovery of zinc and manganese from spent alkaline and zinc–carbon
batteries. VI. International Congress Valorisation and Recycling of Industrial Waste,
L'Aquila, Italy, June 27–29, 2007.
Kierkegaard, S., 2007. EU Battery Directive, charging up the batteries: squeezing more
capacity and power into the new EU Battery Directive. Comput. Law Secur. Rep. 23,
Li, Y., Xi, G., 2005. The dissolution mechanism of cathodic active materials of spent Zn–
Mn batteries in HCl. J. Hazard. Mater. B127, 244–248.
Meller, P., 2006. EU approves radical battery recycling . Available from bhttp://www.
Montgomery, D.C., 1991. Design and Analysis of Experiments, 3rd ed. Wiley, New York.
Naik, P.K., Sukla, L.B., Das, S.C., 2000. Aqueous SO
leaching studies on Nishikhal
manganese ore through factorial experiment. Hydrometallurgy 54, 217–228.
Rabah, M.A., Barakat, M.A., Mahrous, Y.Sh., 1999. Recovering metal values hydro-
metallurgically from spent dry battery cells. General Hydrometallurgy, Research
Summary, JOM 41–43.
Rayovac Corp., 1999. Material safety data sheet. Available from http://rayovac.com
Salgado, A.L., Veloso, A.M.O., Pereira, D.D., Gontijo, G.S., Salum, A., Mansur, M.B., 2003.
Recovery of zinc and manganese from spent alkaline batteries by liquid–liquid
extraction with Cyanex 272. J. Power Sources 115, 367–373.
Sahoo, R.N., Naik, P.K., Das, S.C., 2001. Leaching of manganese from low-grade manganese
ore using oxalic acid as reductant in sulfuric acid solution. Hydrometallurgy 62,
Turkish Ministry of Environment and Forestry, 2005. Directive of the Control of Spent
Battery and Accumulators (in Turkish).
Veloso, L.R.S., Rodrigues, L.E.O.C., Ferreira, D.A., Magalhaes, F.S., Mansur, M.B., 2005.
Development of a hydrometallurgical route for the recoveryof zinc and manganese
from spent alkaline batteries. J. Power Sources 152, 295–302.
79E. Sayilgan et al. / Hydrometallurgy 97 (2009) 73–79