Magnetic separation of iron and heavy metals from water

J.D. Navratil* and M.T. Shing Tsair**

* Clemson University, Environmental Engineering and Science, 342 Computer Court, Anderson,
SC 29625-6510 USA
** Department of Life Sciences and Chemical Technology, Ngee Ann Polytechnic, 535 Clementi Road
Singapore 599489
Abstract A magnetic separation device is being developed for removal of iron and heavy metals from water.
The device consists of a column of supported magnetite surrounded by a movable permanent magnet. The
mineral magnetite, or synthetically prepared iron ferrite (FeO.Fe
2
O
3
), is typically supported on various
materials to permit adequate water passage through the column. In the presence of an external magnetic
eld, enhanced capacity was observed in using supported magnetite for removal of actinides and heavy
metals from wastewater. The enhanced capacity is primarily due to magnetic ltration of colloidal and
nanoscale particles along with some complex and ion exchange sorption mechanisms. This paper will review
some previous work on the use of magnetite for wastewater treatment and discuss the development and
potential of the magnetic nanoscale ltration/sorption process for water treatment. Recent research results
are also presented on preliminary experimental studies using the process with water samples containing
iron.
Keywords Adsorption; ltration; iron; iron ferrite; magnetic separation; water treatment
Introduction
Drinking water quality is being increasingly compromised because of outdated water treat-
ment plants and older distribution piping systems. Impurities can add objectionable taste
and odors. Iron is one of the major impurities that is commonly found in many sources
of water. Iron deposited in the distribution system may promote the growth of micro-
organisms leading to high turbidity in drinking water.
The European Committee (EEC, 1980) standard recommends that the iron concentration
in drinking water should be less than 0.2mg/L. In the United Status, the guideline level of
iron in drinking water is less than 0.05 mg/L (VEWIN, 1993). The lower level of iron is
desirable in order to avoid problems and increased cost in the maintenance of water distri-
bution systems. Iron related problems have become a significant water quality issue to
water supply authorities in Adelaide, Melbourne, the Gold Coast of Queensland and in the
urban regions north and south of Sydney (OConnor, 1995). It should be recognized that the
problem is not unique to Australia with, for example, roughly 40 per cent of public water
supplies in the United States exceeding recommended levels of iron (Salem, 1995). In
recent years there has also been interest in iron removal from well water in the New Nile
Valley in Egypt (Aly, 2000).
This paper briefly reviews a new technique, magnetic filtration/sorption, as a potential
field device for removing iron from water. The proposed method has the potential to reduce
contamination problems inexpensively. Recent studies have demonstrated the magnetic
enhanced removal of radioactive and heavy metals from water (Boyd et al., 1986; Kochen
and Navratil, 1997; Navratil et al., 1995 and Cotton and Navratil, 1997). Several suitable
support materials including glass wool and sand were tested for mixing with magnetite to
achieve satisfactory water flow characteristics while retaining the magnetic field enhanced
sorption properties of magnetite (Cotton and Navratil, 1997). This paper will review
W
a
t
e
r

S
c
i
e
n
c
e

a
n
d

T
e
c
h
n
o
l
o
g
y

V
o
l

4
7

N
o

1

p
p

2
9

3
2

I
W
A
P
u
b
l
i
s
h
i
n
g

2
0
0
2
29
selected previous work on the use of magnetite for wastewater treatment and discuss the
development and potential of the magnetic filtration/sorption process for water and waste-
water treatment. Recent preliminary research results will also be presented that have
focused on experimental studies with water samples containing iron.
Review of past developments
Magnetite or iron ferrite has been used to separate a wide variety of substances such as dis-
solved metal species, particulate matter, organic and biological materials (Rubio et al.,
1996). Magnetic separation and high gradient magnetic separation have been used exten-
sively in the processing of minerals and more recently water treatment and environmental
application (Davis, 1997).
Magnetite has the ability to remove heavy metals as reported by Kochen and Navratil
(1997) who used a magnetic polymer resin for the removal of actinides and other heavy
metals from contaminated water. A polyamine epichlorohydrine polymer was synthesized
containing magnetite on the surface of the resin beads. In the absence of an external mag-
netic field, the iron ferrite readily adsorbed numerous metal species. In the presence of an
external magnetic field, it was observed that there was an increased capacity for the
removal of plutonium-239 and americium-241. These results imply that adsorption proper-
ties of magnetite can be used in conjunction with a magnetic field to remove both metal ions
and metal colloids and nanoparticles. Most mechanical filtration systems are only slightly
effective at removing colloids, so this magnetic enhanced process could offer a substantial
improvement.
Other investigators have also reported heavy metal removal with magnetite both in and
out of the magnetic field. In 1969, Petkovic and Milonjic (1995) reported the adsorption of
cesium on magnetite between the pH of 7.6 and 10.4. It was observed that cesium adsorp-
tion increased with pH although the mechanism was not presented. Peak adsorption was
reported at a pH of 10.4. In 1970, Milonjic and Ruvarac also reported the adsorption of
cesium (I), cobalt (II) and cerium (II) by magnetite at a pH of 2.2 by batch experiments.
Cobalt adsorption was reported at less than 5% w/v although actual data was not enough to
verify the results. In addition, work of Boyd et al. (1986) reported the iron ferrite removal in
the pH range of 913. Boyd et al. (1986) also reported the effect of dissolved chemical
species on ferrite formation including the use of commercially available magnetite.
In addition, Krumm (1989) described another application of carrier magnetic separation
(CMS) for the purification of wastewater containing phosphate, metals, and heavy metal
and pigment particles with magnetite and a magnetic carrier. The removal of lead and cop-
per ions from diluted solution by sorption into clinoptilolite, together with magnetite, as
reported by Feng et al. (2000) showed that the process was very rapid, effective and yielded
clear solutions with low residual concentrations of heavy metals. In their experimentation,
concentration of clinoptilolite fines were used as heavy metal adsorbent, while low dosages
of magnetite fines were used as magnetic carrier materials. In addition, hydrolyzed Fe
+3
was used as a coagulant.
Materials and methodology
Feed solutions of concentration 10
4
, 10
5
and 10
6
mole/litre (mol/L) were prepared using
iron ammonium sulfate. Feed solutions were sent to columns up flow using a Master Flex
console drive tubing pump fitted with two Master Flex pumping head. Flow rates were var-
ied 0.1 ml/min to 6 ml/min to determine the adsorption kinetics, i.e. the residence time
requirement. Three columns (of each type) were run in conjunction with a blank, with and
without 0.5 tesla magnetic field. The columns used were made of plastic droppers 15.7 cm
long with a diameter of 0.3 cm and a plastic filter funnel of diameter 5.3 cm. The magnetite
J
.
D
.

N
a
v
r
a
t
i
l

a
n
d

M
.
T
.

S
h
i
n
g

T
s
a
i
r
30
and wool added to the columns were of mass percentage 80% and 20% respectively. Thus a
mass of 0.8 g of magnetite and 0.2 g of wool were used throughout the experimentation.
The volume of feed varied as iron solution was continuously passed through all thee
columns until 100% breakthrough took place. To determine breakthrough capacities, frac-
tions of the samples were collected using test tubes of 4 mL volume and transferred to a 25
ml volumetric flask containing buffering reagents. Flow rate and pH were measured at each
interval and samples were analyzed using a spectrophotometer. Sample blanks and period-
ically calibration checks ensured sample accuracy and consistency.
The iron concentration was determined colorimetrically using a Spectronic 20 Genesys
spectrophotometer equipped with single cell holder, 5 mm disposable cuvets. Ferrous iron
feed solution was made from reagent grade iron ammonium sulfate. Hydroxylamine
hydrochloride, sodium acetate, ortho phenanthroline and sulphuric acid were reagent grade
materials. Prior to use, hydroxylamine hydrochloride and sodium acetate were made by
diluting to stock solutions of approximately 5% w/v, and stored in plastic bottles until
required. 0.001 mol/l of sulphuric acid was prepared and standardized according to stan-
dard method. 0.1% w/v of ortho phenanthroline iron indication solution from Fisher
Scientific USA was used prior to each experiment. Unused solution was refrigerated. All
dilutions were made using distilled, deionization water.
Results and discussion
Table 1 shows the summary results of the preliminary column experiments. In comparing
the duplicate data in Table 1, the breakthrough capacities are in good agreement (510%
error). Generally, the breakthrough capacity decreases with increasing iron concentration
in the feed and with increasing flow rate, as expected.
The presence of a magnetic field does play a part in the adsorption of iron onto mag-
netite. As shown in Table 1, for a concentration of 10
5
mol/l, the column surrounded by a
0.5 tesla magnetic field resulted in a higher iron breakthrough capacity. From this result, in
agreement with reported results, we can conclude that magnetic field does affect the
removal of colloidal ferric hydroxide particles along with the adsorption of iron from
solution.
When sea sand was added as part of the supporting material for the column, the rate of
removal of iron was also affected. As we can see in Table 1 for the same concentration and
flow rate, the breakthrough capacities were higher. Sea sand does not only reduce channel-
ing but also improve the surface area in contact between magnetite and the colloids present.
Thus we can also conclude that sea sand is a useful support material that permits adequate
water flow through the column of magnetite.
J
.
D
.

N
a
v
r
a
t
i
l

a
n
d

M
.
T
.

S
h
i
n
g

T
s
a
i
r
31
Table 1 Column breakthrough capacity tests
Feed iron(II) Column type (mol/l) Flow rate 10% 50% Magnetic
concentration (ml/min) breakthrough breakthrough eld
(mol/l) vol feed (ml) vol feed (ml)
2.75 10
4
Magnetite and glass wool 0.8 < 12 ml 15 Yes
2.75 10
4
Magnetite and glass wool 0.8 12 20 Yes
2.75 10
4
Magnetite and glass wool 2 4 14 Yes
2.75 10
4
Magnetite and glass wool 2 10 20 Yes
2.75 10
4
Sea sand, wool and magnetite 0.7 15 20 Yes
2.75 10
4
Sea sand, wool and magnetite 0.7 <12 18 Yes
2.75 10
5
Magnetite and glass wool 0.75 28 40 No
2.75 10
5
Magnetite and glass wool 0.75 24 76 Yes
Conclusions
This work provided a quantitative framework for assessing the utility of applied magnetic
fields for creating an attractive magnetic force between the adsorbent (magnetite) and the
adsorbate (iron). The use of magnetite adsorbent material in a fixed bed model surrounded
by a high gradient magnetic field was shown to be efficienct for the removal of iron from
wastewater. The above experimentation proved that iron could be adsorbed by magnetite
provided with the correct type of support material and flow rate.
References
Apelblat, A. (1993). Solubilitys of organic salts of Magnesium, Calcium and Iron in Water, Journal of
Chemical Thermodynamics, 25(12), 14431445.
Aly, H. (2000). Egyptian Atomic Energy Authority, Cairo Egypt, private communication.
Boyd, T.E., Cusick, M.J. and Navratil, J.D. (1986). Chapter 6: Ferrite Use in Science and Technology,
Recent Developments in Separation Science Volume (III), CRC Press, inc, Boca Ration, Fl.
Cotton, G.B. and Navratil, J.D. (1997). Proceedings of Waste Management Conference.
Davis, J. (1997). Removal Iron and manganese from nature water Diversity Water Technology, Incm
Chargin Falls, OH, Oct 1997, 4547.
EEC (1980). Official journal of the European Communities 30 August, 23: Official Directive
L229/11L229/23.
Ellis, D., Bouchard, C. and Lantagne, G. (2000). Removal of iron and manganese from groundwater by oxi-
dation and filtration, Desalination, 130, 255264.
Feng, D., Aldrich, C. and Tan, H. (2000). Remoal of Heavy metals by carrier magnetic Separation of adsorp-
tive particulates, Department of chemical engineering m University of Stellbosch, Hydrometallurgy, 56,
359368.
Kochen, R.L. and Navratil, J.D. (1997). Removal of Radioactive Materials and Heavy metals from Water
Using Magnetic Resin, United States Patent 5,595666.
Milonjic, S.K. and Ruvarac, A. (1970). Adsorption of Cs
+
, Co
2+
and Ce
3+
From Acid Aqueous Solutions on
Natural Magnetite, Bulletin of the Boris Kidric Institute of Nuclear Sciences, Vol. 21, Chemistry, No. 3
P/462.
Navratil, J.D., Kochen R.L. and Ritter, J.A. (1995). Magnetic Swing Adsorption Process, Proceedings of
Waste Management Symposia 1995, Tucson, AZ.
OConnor, J.T. (1995). Iron and Manganese; Water Quality and Treatment A Handbook of Public Water
Supplies, Chapt 11, pp. 378396.
Petkovic, D.M. and Milonjic, S.K. (1969). Adsorption of Cesium From Basic Water Solutions on Natural
Magnetite, Bulletin of the Boris Kidric Institute of Nuclear Sciences, Vol. 20, Chemistry, No. 3 P/449.
Rubio, J., Schneider, L.A.H. and Allaga, W. (1996). New Processes for heavy metals separation from waste
water streams in the mining industry, Proc. Clean Technology for Mining Industry, Santiago, pp. 8598.
Salem, F.B. (1995). Determination of Iron in Water Samples, Reviews in Analytical Chemistry, 14(1),
5973.
VEWIN Aanbevelingen (1993). The Netherlands.
J
.
D
.

N
a
v
r
a
t
i
l

a
n
d

M
.
T
.

S
h
i
n
g

T
s
a
i
r
32