Resources, Conservation and Recycling 87 (2014) 109–116

Contents lists available at ScienceDirect

Resources, Conservation and Recycling

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

Review

Environment-friendly technology for recovering nonferrous metals

from e-waste: Eddy current separation
Ruan Jujun a,b , Qian Yiming a , Xu Zhenming b,∗
a
Jiangsu Key Laboratory of Environmental Material and Environmental Engineering, School of Environmental Science and Engineering, Yangzhou
University, 196 West Huayang Road, Yangzhou, Jiangsu 225127, China
b
School of Environmental Science and Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, People’s Republic of China

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

Article history: The current generation pattern of e-waste consisted of dead electronic and electrical equipments poses
Received 20 November 2013 one of the world’s greatest pollution problem due to the lack of appropriate recovery technology. Crude
Received in revised form 16 February 2014 recovery methods of resource materials (aluminum, zinc, copper, lead, gold) from e-waste caused seri-
Accepted 28 March 2014
ous pollution in China in the past years. Thus, environment-friendly technologies have been the pressing
demand in e-waste recovering. Eddy current separation (ECS) was advised as the preferable technology
Keywords:
for recovering nonferrous metals from e-waste. However, just a few reports focused on the application of
e-Waste
ECS in e-waste recovering. This paper introduced the information about ECS including the models of eddy
Nonferrous metals
Recovery
current force and movement behavior of nonferrous metallic particle in the separation process. Mean-
Eddy current separation while, the developing process of eddy current separator was summarized. New industrial applications of
ECS in e-waste (waste toner cartridges and refrigerator cabinets) recovering were also presented. Finally,
for improving separation rate of ECS in industrial application of e-waste recovering, some suggestions
were proposed related to crushing process, separator design, and separator operation. The aim of this
paper is to demonstrate the effectiveness of ECS technology as practical and available tool for recovering
non-ferrous metals from e-waste which is now being ignored.
© 2014 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
2. Eddy current force and eddy current separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
2.1. Eddy current force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
2.2. Eddy current separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
3. Efforts for improving separation rate of ECS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
3.1. Influencing of particle characteristics on separation rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
3.2. Influence of operation conditions on separation rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
3.3. Development of eddy current separator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
4. New industrial application of ECS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
5. Measures for improving separation rate of ECS in its new industrial application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
6. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

∗ Corresponding author. Tel.: +86 21 54747495; fax: +86 21 54747495.

E-mail addresses: zmxu@sjtu.edu.cn, ruanjujun@yzu.edu.cn (X. Zhenming).

http://dx.doi.org/10.1016/j.resconrec.2014.03.017
0921-3449/© 2014 Elsevier B.V. All rights reserved.
110 R. Jujun et al. / Resources, Conservation and Recycling 87 (2014) 109–116
Nomenclatures
bn Fourier coefficient
Bm magnetic flux density of the magnetic drum surface
(T)
Bp induced magnetic flux density in the particle (T)
Br (B, Ba ) magnetic flux density of the field (T)
Fr repulsive force between the particle and the magnet
(N)
g acceleration of gravity force (m/s2 )
G gravity force of the aluminum flake (N)
J (j) induced eddy current in particle (flakes) (A)
k pairs of the magnets placed in the magnetic drum
l height of the rectangle/triangle flake (m)
L circumference of the triangle flake (m)
r radius distance between the particle and the center
of magnetic drum (m)
r radial distance between the particle and the induc-
ing magnet (m)
R (Rdrum ) radius of the magnetic drum (m)
Sm per magnet side area which facing the flake (m3 )
Sp (d) maximal cross area of the flake in horizontal (m2 )
t time cost for the magnetic field rotation (s)
T thickness of the flake (m)
v (˝) feeding speed of the particle (flake) (m/s)
v relative linear velocity between the flake and mag-
netic drum (m/s)
V volume of the particle (rectangle/triangle flake)
(m3 )
w width of the rectangle/triangle flake (m)
˝ resistance of circular/rectangle/triangle coil
˛0 angle of the coordinate in the cylindrical coordinate
system
 (, ) conductivity of the flake (S/m)
ı (s) oriental (shape) factor of the flake in eddy current
separation
εi induced emf in the circular/rectangle/triangle coil
(V)
r relative magnetic permeability of iron (H/m)
0 magnetic permeability of vacuum (H/m)
ωm (ωdrum , ω) rotation velocity of the magnetic drum (rad/s)
Фp induced magnetic flux in the particle (Wb)
Фm magnetic flux of the permanent magnet (Wb)
Ф variation of magnetic flux in the particle (Wb/s)
w( ) width of pole
1. Introduction
Quantities of e-waste are generating resulted from the use
of electronic and electrical products. Computer accessories and
mobile telephones are disproportionately abundant because of
their short lifespan. The current global production of e-waste is
estimated to be 20–25 million tons per year and about 95% useful
materials were recovered (Robinson, 2009). About 2.5 million tons
e-waste appeared in Chinese mainland including self-generated
and imported from developed countries per year (Ongondo et al.,
2011; Stone, 2009; Widmera et al., 2005). Fig. 1 shows samples of
nonferrous metals found in e-waste. Waste PCB contains nearly 28%
metals including copper, zinc, and other nonferrous metals (Li and
Xu, 2010). Waste toner cartridge has 11.7% aluminum (Ruan et al.,
2011). Waste refrigerator cabinet includes about 8.9% copper and
aluminum (Ruan and Xu, 2011a,b). Additionally, purity of metals
in e-waste is higher than that of rich-content minerals (Laner and
Rechberger, 2007; Ilgin and Gupta, 2010a,b). Thus, recovering e-
waste can bring great economic benefits.
Environmental sound technologies of recovering e-waste are
developing challenges today. In the early stage, crude technologies
(acid-washing or open incineration) were employed (Ruan and Xu,
2012a,b; Ilgin and Gupta, 2010a,b) and resulted to serious environ-
mental pollution by the hazardous materials contained in e-waste
(Duan et al., 2011; Huo et al., 2007; Leung et al., 2008). Then, for the
sake of environmental protection and clean recovery of nonmetals
from e-waste, it was proposed a procedure including crushing pro-
cess and psychical separations of screen, shape sorting, jigging,
magnetic separation, air current separation, corona-electrostatic
separation, and eddy current separation (ECS) (Zhou and Xu, 2012;
Cui and Forssberg, 2003). However, each technology has special
limitation. Screen and shape sorting cannot separate the parti-
cles that have similar size and shape. Jigging brings waste water
in separation process. Magnetic separation can only separate fer-
rous metals. Air current separation demands particles having great
density difference when being in similar size. Corona-electrostatic
separation and ECS are the preferable technologies for recovering
nonferrous metals from e-waste. Corona-electrostatic is skilled in
separating nonferrous metallic particles (NMP) less than 1 mm in
size (Wu et al., 2008). ECS is adept in separating NMPs ranged from
2 to 50 mm in size. ECS may be the fittest technology for recovering
nonferrous metals from large-scale (coarse crushing is enough for
liberating the materials) e-waste (Zhang et al., 2002; Benaboua and
Georgesa, 2008).
ECS is an environment-friendly technology for separating non-
ferrous metals from solid waste. In separation process, eddy current
is induced in nonferrous metal when meeting variable magnetic
field. Interaction between eddy current and magnetic field changes
the trajectory of nonferrous metal as well as separates them from
others. No waste water, air pollution, and solid waste are gener-
ated in the separation process. Unfortunately, the public paid less
attention on this environment-friendly technology.
This paper discussed ECS technology from the models of eddy
current force (ECF), models of particle movement behavior, devel-
oping of separator, and new industrial application standpoints for
e-waste recovering. Furthermore, suggestions for improving sepa-
ration rate of ECS are presented.
2. Eddy current force and eddy current separation
2.1. Eddy current force
ECS is a physical method for separating nonferrous metals from
inert materials. ECF is the cause of ECS. In general, eddy current sep-
arator is comprised of magnetic drum. The magnetic drum always
consists of magnetic poles placed in N–S–N (see Fig. 2). A changing
magnetic field will be induced by the rotation of magnetic drum
(Peterson, 2003; Rolicz, 2009). There are two recognitions about
the generation of ECF. Recognition (1): eddy current will appear in
NMP when it experiences the changing magnetic field; a repulsive
force will be generated between magnetic field and NMP possess-
ing eddy current; this repulsive force is called ECF. Recognition (2):
direction of the magnetic field changes constantly because of the
running of magnetic drum; consequently, the direction of eddy cur-
rent induced in NMP also changes continuously resulting in the
generation of a new magnetic field in NMP; direction of the new
magnetic field is also changed constantly; the two magnetic fields
have the same directions and repulse each other; this repulsion is
called ECF.
Because of being the most important influencing factor of ECS,
many models for computing ECF were constructed. The models of
ECF established from recognition (1) including:
 R. Jujun et al. / Resources, Conservation and Recycling 87 (2014) 109–116 111

Fig. 1. Nonferrous metals in e-waste: (a) waste PCBs, (b) waste toner cartridges, (c) waste refrigerator cabinet.

(i) Models of radial and tangential components of ECF (Rem et al., magnetic poles can be used to guide the manufacture of eddy
1997; Zhang and Rem, 1999): current separator. Parameters of rotation speeds of magnetic
drum and particle can guide the operation of ECS.
2
sV (kωdrum + ˝)
Ft = 2 2
Ba2 (1) (ii) The modified models of radial and tangential components of
0 w 1 + (kω
drum + ˝)  ECF (Lungu and Schlett, 2001):
2
2
sV (kωdrum + ˝)  2 Fn = s0 (ω − ˝)R2 Ft (3)
Ft = 2 2
Ba2 (2)
0 w 1 + (kω  2
R  T
drum + ˝) 
Ft = (4)
The models indicated that ECF was determined by the param- R
eters of separator operation, particle characteristics, and 
separator design. The force increased with the increasing of
T= r × fdV (5)
particle volume, intensity of magnetic field, and decreased V
with the increasing of width of magnetic pole. Additionally,
particle shape, pairs of magnetic poles, rotation speed of mag- f = j × BdV (6)
netic drum, and angular velocity of particle influenced the
magnitude of ECF. Parameters of particle volume and shape These models introduced the parameter of radius distance
can guide crushing process for liberating nonferrous metals. between the particle and the center of magnetic drum as well
Parameters of intensity of magnetic field, pairs and width of as adding the parameter of eddy current induced in particle.

Fig. 2. NMP was separated from nonmetal because of eddy current force in ECS.
112 R. Jujun et al. / Resources, Conservation and Recycling 87 (2014) 109–116

(iii) Models for computing the moment of ECF (Maraspin et al., Movement behavior of NMP in ECS is divided into three stages:
2004): (1) entering magnetic field, (2) detaching from conveyor belt sur-
  face, (3) exiting from magnetic field (see Fig. 3).
(k + 1)B̂2 V R() At the beginning of stage (1), eddy current force increases as the
Fm = (7)
0 r I() particle moving close to magnetic drum, and the force is divided
into vertical component and horizontal component. Horizontal
B̂2 V component is counteracted by the friction force of conveyor belt
Tm = − I (8) and no relative motion happens between NMP and belt. As the
0 ()
increasing eddy current force, vertical component will be greater
 = 0 (kωm + ˝)d2 (9) than the gravity force (G) of the flake, and NMP will have a vertical-
upward acceleration and move upward. Meanwhile, eddy current
Besides the parameters contained in models (i) and (ii), maxi-
force will decrease along with the upward movement of NMP due
mal cross area of particle in horizontal was introduced in model
to the decline of magnetic flux of magnetic drum. When the ver-
(iii). This parameter could guide the crushing process more
tical component is equal to gravity force, NMP will be suspended
specifically and predicted that flake shape brought positive
and keep constant radial distance to the axis (O) of separator. The
influence on ECF.
position of the flake is called point (x0 , y0 ). Point (x0 , y0 ) can be
considered as the detachment point of NMP from separator sur-
The models of ECF established from recognition (2) (Ruan and
face. Meanwhile, point (x1 , y1 ) is supposed as the symmetry point
Xu, 2011a,b):
of (x0 , y0 ) at the right of Y-axis (Fig. 3). The movement of NMP
Model for circle flake:
from point (x0 , y0 ) to point (x1 , y1 ) can be considered as rectilin-
Br k(ωm R − v)VSp Bm Sm 1 ear motion. At this rectilinear movement, magnetic fluxes of NMP
Fr C = (10)
16
3 R3 (sec˛0 − 1)2 and magnetic drum is considered to be parallel, horizontal compo-
nent of eddy current force can be neglected, and repulsive force can
Models for rectangle flake were:
be supposed as equal to its gravity force. When NMP passes over
Br k(ωm R − v)VSp Bm Sm ıR 1 point (x1 , y1 ), vertical component of eddy current force will be less
Fr R =
16
2 R3 (sec˛0 − 1)2 than gravity force. Horizontal component of repulsive force can no
(11) longer be neglected since the directions of the two magnetic fluxes
W
ıR = will no longer been parallel. Horizontal component will accelerate
2(L + W )
NMP in horizontal direction until it passes through the boundary
Models for triangle flake were: of the magnetic field. Furthermore, the movement of NMP in verti-
Br k(ωm R − v)VSp Bm Sm ıT 1 cal direction is controlled by gravity force and vertical component
Fr T = of ECS. We suppose point (x2 , y2 ) as the exiting position of NMP
16
2 R3 (sec˛0 − 1)2 (12) from the magnetic field. As passing through point (x2 , y2 ), NMP
L
ıT = only subjects to gravity force and the movement can be considered
C as horizontal projectile motion.
The parameters, contained in the above models of ECF, could be Due to no response to magnetic field, movement behavior of
used to guide crushing, separator operation, and separator design. plastic in ECS is considered as horizontal projectile motion.
Compared the models established from recognitions (1) and (2), the The trajectory models of NMP were given as Eq. (13). New influ-
models established from recognition (2) paid more attention to the encing factors of magnetic field boundary and collection position
influencing of shape of particle to ECF, and were more specific and were introduced in the trajectory models.
accurate. Precisely, the application range was limited and smaller ⎧ 

than the models established from recognition (1). ⎪
⎪ −
 v 2 v 2
⎪
⎪
1
y = − (g − ay )
2(x x1 )
+ − + y1
⎪
⎪
⎪
⎪
2 ax ax ax
2.2. Eddy current separation ⎪
⎪
⎪
⎪ xb2 + yb2 = (R + )2
⎪
⎪
ECF causes ECS. By neglecting air friction, NMP in ECS are subject ⎪
⎪
⎨ g sin a1
to ECF and gravity force. Due to the absence of electrical conduc- ax =
2 (13)
tivity, nonmetals are only subjects to gravity force. Thus, NMP are ⎪
⎪
separated from nonmetals due to the different movement behav- ⎪
⎪ g cos ˛1
⎪ ay =
⎪
iors in ECS. The sketch map of the movement behavior of NMP and ⎪
⎪
2
⎪
⎪
nonmetal are presented in Fig. 2. ⎪
⎪ x = Rtg˛
⎪
⎪
1 0
ECS also can be used to separate different nonferrous metals ⎪
⎪
as long as they have different electrical conductivities. Different ⎩ y1 = R
cos ˛0
conductivities of nonferrous metals cause different magnitudes of
ECF in order to bring distinguished movement behaviors of NMPs.
For improving separation rate of ECS, the movement behavior of 3. Efforts for improving separation rate of ECS
NMP was studied and the trajectory models were constructed. The
models contained abundant influencing factors of ECS and could be Experiments of ECS were performed to study the influence of
used to predict separation results. particle characteristics and operation conditions (convenient to be
Based on the models of (i), (ii), (ii) above, movement behavior of controlled in ECS and crushing) on separation rate.
NMP in ECS was simulated by computer software (Maraspin et al.,
2004; Zhang and Forssberg, 1999). 3.1. Influencing of particle characteristics on separation rate
Based on the models of ECF constructed from cognition (2),
trajectory models of NMP in different position inside and outside ECS of aluminum, copper, zinc, lead, and PVC sheets showed
the magnetic field (Fig. 3) of ECS were constructed (Ruan and Xu, metal sheets that of differing in the ratio of electrical conductivity to
2012a,b). density could be separated successfully by the condition of having
 R. Jujun et al. / Resources, Conservation and Recycling 87 (2014) 109–116
Fig. 3. Analysis of the forces and movement behaviors of NMP and nonmetal in and out of the magnetic field.
the similar size and shape (Zhang and Forssberg, 1999). The metal
sheets that differ in shape could be separated as long as having sim-
ilar size and ratio of electrical conductivity to density. The metal
sheets that differ in size could also be separated if they have simi-
lar shape and ratio of electrical conductivity to density. The metal
sheets that differ in size, shape, and ratio of electrical conductivity
to density could not be separated effectively.
Particle size was of critical consideration. Particle size below
2 mm was not effectively responsive to the separator regardless
of how the operating parameters were adjusted. Additionally, ECS
was unable to recover nonferrous metallic foils. Particle’s shape
was one of the most influential variables. Irregular shape of par-
ticle minimized ECF. ECF of plate particle was greater than those
of ball-shaped particles and irregular shapes (Zhang et al., 2002).
Additionally, moisture was another important influential factor.
Wet fed would decrease the separation rate of ECS. Small metal par-
ticles would stick to the feed conveyor belt (Rahman and Bakker,
2013).
3.2. Influence of operation conditions on separation rate
Feeding the particles with slight wet to traditionally eddy cur-
rent separator could separate NMPs in the range from 2 to 6 mm
from waste streams (Köhnlechner et al., 2002). Rotor speeds should
be relatively low so as to keep ECF being used only to break the wet
bond between NMPs and belt surface.
Feeding system and feed speed significantly influenced sep-
aration rate. Multilayer feed stream could seriously deteriorate
the separation process. NMPs were shielded from approaching
the effective magnetic field by nonmetallic particles. Nonmetallic
particles could flip away together with NMPs. Apparently, a mono-
layer feed stream could ensure effective separation (Zhang et al.,
2002).
Results of orthogonal experiment of ECS showed that differ-
ence between feeding speed and rotation speed of magnetic field
(ωR − v) was the critical influencing factor on separation rate. Feed-
ing speed (v) was general factor, and the collection position (H) was
subordinate factor (Ruan and Xu, 2012a,b).
113
A splitter with appropriate position involved in eddy current
separator during ECS was of great importance. Splitter could sepa-
rate NMP timely when it deflected from nonmetals caused by ECF
(Schlett and Lungu, 2002).
3.3. Development of eddy current separator
The traditional eddy current separator was horizontal magnetic
drum separator (see Fig. 4(a)). For improving the separation rate,
vertical drum eddy current separator (Fig. 4(b)) was developed to
separate small conductive nonferrous particles whose dimensions
were within 2–8 mm (Lungu and Schlett, 2001; Lungu, 2005). Wet
eddy current separator (Meier-Staude et al., 2002) was developed
to separate the NMP whose particle size was from 2 to 5 mm.
The sketch was given in Fig. 4(c), where: T, magnetic drum with
4 poles; I, dielectric chamber for liquid having two collectors C1 and
C2; A, feeding system. The particles falling from the feeder in to the
liquid are subjected to both Magnus force and translational force.
The gravity and buoyancy force act upon the particles as well. The
Magnus and the translational forces will make the trajectory of the
particles deviate from the vertical. If the weight of the particles is
higher than that of the other forces, the particles will continue to
fall into the liquid being collected in the collector C2. The light par-
ticles continue to rotate and move to the left. The translational force
will act afterwards, but the Magnus force will modify its direction
having as an effect the removal of the particles from the magnetic
drum. The effect of the Magnus force will rise up the particles. In
this way, under the influence of the translational force, the light
particles will be moved to the left, falling finally in the C1 collector.
4. New industrial application of ECS
ECS is efficient skilled in separating NMPs whose size ranged
from 5 to 10 mm. The materials contained in large-scale e-waste,
such as waste computers, refrigerator cabinets, toner cartridges,
can be completely liberated when crushed into 5–10 mm. Thus, ECS
may be the most suitable technology for separating NMPs from the
crushed large-scale e-waste.
114 R. Jujun et al. / Resources, Conservation and Recycling 87 (2014) 109–116

Fig. 4. (a) Horizontal drum ECS; (b) the sketch of vertical drum eddy current separator; (c) the sketch of the wet eddy current separator.

The lines of recovering waste toner cartridges and refrigerator closed crushing, activated carbon adsorption tower, magnetic sep-
cabinets had been constructed (Ruan et al., 2011; Ruan and Xu, aration, air current separation, and ECS (Fig. 5(b)). Crushing was
2011a,b) (Fig. 5). Waste toner cartridges recovering line was com- performed to liberate the materials. Activated carbon adsorption
prised of coarse crushing, magnetic separation, ECS, and bag-type tower was used to collect CFC-11 emitted from polyurethane foam
dust collector (Fig. 5(a)). Waste toner cartridge was comprised of in the crushing process. Air current separation was designed to
steels, toners, aluminum, and plastics (Table 1). Coarse crushing collect the crushed polyurethane foam powder. Magnetic separa-
was employed to liberate the materials of waste toner cartridges. tion was employed to separate steels from the crushed materials.
Magnetic separation was used to separate steels from the crushed ECS was adopted to separate aluminum and copper particles from
materials. Bag-type dust collector was adopted to collect toners plastic. In experiment, 50 waste refrigerators cabinets (1547.7 kg)
during the crushing process of toner cartridges. ECS was employed with different sizes and types were fed into the automated produc-
as crucial process for separating aluminum from crushed materi- tion line. The comprised materials were liberated from the cabinets
als. In experiment, about 500 kg various waste toner cartridges with in shearing process, and the released CFC-11 gas was adsorbed
different sizes (5 cm × 5 cm × 18 cm to 10 cm × 12 cm × 40 cm) and by the activated carbon adsorption tower. Air current separation
types were fed into the recovery line. Aluminum were crushed into collected 196.5 kg polyurethane foam. 828.5 kg ferrous, 347.7 kg
flake granule with sizes concentrating on 10 mm × 10 mm × 1 mm plastic, 97.7 kg aluminum, and 39.6 kg copper were obtained by the
to 15 mm × 15 mm × 2 mm, and then separated from other parti- processes of magnetic/ECS. Recovery rate of cabinet reached 97.6%
cles by ECS. The recovery rates of steel (magnet), toner, aluminum, by the recovery line.
and plastic were 98.4%, 95%, 97.5%, and 98.8%, respectively. Traditional eddy current separators were employed in the two
Waste refrigerator cabinet was comprised of steels, plastics, alu- lines. Traditional eddy current separator always has high separa-
minum, copper, polyurethane foam, and CFC-11 (see Table 1). Thus, tion rate (above 95%) in its applications for mineral processing
the recovering line of waste refrigerator cabinets was comprised of and municipal solid waste treatment (Cui and Forssberg, 2003).
Unfortunately, separation rates of ECS in the two recovering lines
were less than 85%, far below the standard separation rate of 95%.
Table 1
Comprised materials of waste toner cartridge and refrigerator cabinet. The reasons of this gap may be: (1) crushing process of e-waste
produces complex particles’ shapes (Fig. 5(a)); besides flake, large
Comprised Waste toner Waste refrigerator
proportion of cuboid, triangular, and spherical particles, are pre-
materials cartridge (wt.%) cabinet (wt.%)
sented in the liberated materials. The complex shape decrease the
Steels/iron 39.3 53.5 effects of ECF so that the separation rate is decreased; (2) nonfer-
Plastics 34.6 22.4
Aluminum 11.7 6.3
rous metals in e-waste are firmly tight combined to inert materials,
Copper – 2.5 greater degree of crushing is therefore needed to liberate the non-
Toner 7.5 – ferrous materials completely; The size of NMPs in crushed e-waste
Magnet 6.9 – are smaller than in mineral and municipal solid waste; Small size
CFC-11 – 2.7
of particle decreases the separation rate; (3) kinds of nonferrous
Polyurethane foam – 12.6
metals that are present in e-waste (Fig. 5(b)), ECS is not an efficient
Total 100 100 tool to separated different kinds of nonferrous metals as outlined
 R. Jujun et al. / Resources, Conservation and Recycling 87 (2014) 109–116 115

Fig. 5. (a) Crushed materials of waste toner cartridges after magnetic separation and toner removed, (b) the flowcharts and the production line of recovering waste toner
cartridges, (c) the flowcharts and the production line of recovering of waste refrigerator cabinets.
116 R. Jujun et al. / Resources, Conservation and Recycling 87 (2014) 109–116
Fig. 6. Approaches to improve separation rate of ECS.
by the low separation rate between copper and aluminum in the
recovering line of waste refrigerator cabinets.
5. Measures for improving separation rate of ECS in its new
industrial application
Approaches to improve separation rate of ECS were concluded
in Fig. 6. Separation rate of ECS will increase with the increasing
volume, cross area, and conductivity of particle. Flake shape particle
always gets better separation rate than irregular shape particle. The
increasing of intensity of magnetic field, number of magnetic poles,
and radius of magnetic drum will improve separation rate. High
rotation speed of magnetic field contributes to high separation rate.
However, high feeding speed has negative effect on separation rate.
In order to improve the separation rate of ECS in its industrial
application for e-waste recovery, crushing should be controlled to
keep NMP at great size (volume) and two-dimension area as long
as the materials can be completely liberated. The shape of sieve
mesh should have a circle-like design shape. On the premise of
saving cost, eddy current separator should employ high intensity
magnets, more pairs of magnetic poles, and small radius of mag-
netic drum. Rotation speed of magnetic drum should be increased.
Meanwhile, feeding speed should keep be as low as possible. New
eddy current separator, such as the innovative two-magnetic-drum
separator, is the pressing demand for better application of ECS in
e-waste recovering growing industry.
6. Conclusion
The models of eddy current force and movement behavior
of NMP in ECS were summarized in this paper. Meanwhile, the
developing process of eddy current separator and its industrial
applications in e-waste (waste toner cartridges and refrigerator
cabinets) recovering were also presented. For improving separa-
tion rate of ECS in industrial application of e-waste recovering,
some suggestions were made related to crushing process, separator
design, and separator operation. The aim of this paper is to demon-
strate the effectiveness of ECS technology as practical and available
tool for recovering non-ferrous metals from e-waste which is now
being ignored.
Acknowledgments
This work was supported by the National Natural Science
Foundation of China (51308488), the Natural Science Foundation
of Jiangsu province (BK20130449), the Science and Technology
Cooperation Fund Program of Yangzhou City-Yangzhou University
(2012038-15) and Development Program of China (863 program
2012AA063206). The authors are grateful to the reviewers who
help us improve the paper by many pertinent comments and sug-
gestions.
References
Benaboua A, Georgesa S. Permanent magnet modeling for dynamic applications. J
Magn Magn Mater 2008;320(6):830–5.
Cui J, Forssberg E. Mechanical recycling of waste electric and electronic equipment:
a review. J Hazard Mater 2003;99(3):243–63.
Duan H, Li J, Liu Y, Yamazaki N, Jiang W. Characterization and inventory of PCDD/Fs
and PBDD/Fs emissions from the incineration of waste printed circuit board.
Environ Sci Technol 2011;45(15):6322–8.
Huo X, Peng L, Xu X, Zheng L, Qiu B, Qi Z, et al. Elevated blood lead levels of children
in Guiyu, an electronic waste recycling town in China. Environ Health Perspect
2007;115(7):1113–7.
Ilgin M, Gupta S. Environmentally conscious manufacturing and product recovery:
a review of the state of the art. J Environ Manage 2010a;91(3):563–91.
Ilgin M, Gupta S. Comparison of economic benefits of sensor embedded products
and conventional products in a multi-product disassembly line. Comput Ind Eng
2010b;59(4):748–63.
Köhnlechner R, Schlett Z, Lungu M, Caizer C. A new wet Eddy-current separator.
Resour Conserv Recy 2002;37:55–60.
Laner D, Rechberger H. Treatment of cooling appliances: interrelations between
environmental protection, resource conservation, and recovery rates. Resour
Conserv Recy 2007;52(1):136–55.
Leung A, Duzgoren-Aydin N, Cheung K, Wong M. Heavy metals concentrations of sur-
face dust from e-waste recycling and its human health implications in southeast
China. Environ Sci Technol 2008;42(7):2674–80.
Li J, Xu Z. Environmental friendly automatic line for recovering metal from waste
printed circuit boards. Environ Sci Technol 2010;44(4):1418–23.
Lungu M, Schlett Z. Vertical drum eddy-current separator with permanent magnets.
Int J Miner Process 2001;63:207–16.
Lungu M. Separation of small nonferrous particles using an angular rotary
drum eddy-current separator with permanent magnets. Int J Miner Process
2005;78:22–30.
Maraspin F, Bevilacqua P, Rem P. Modeling the throw of metals and nonmetals in
eddy current separations. Int J Miner Process 2004;73(1):1–11.
Meier-Staude R, Schlett Z, Lungu M. A new possibility in eddy-current separation.
Miner Eng 2002;15:287–91.
Ongondo F, Williams I, Cherrett T. How are WEEE doing? A global review of the man-
agement of electrical and electronic wastes. Waste Manage 2011;31(4):714–30.
Peterson W. Numerical solution of eddy current problems in ferromagnetic
bodies travelling in a transverse magnetic field. Int J Numer Methods Eng
2003;58(12):1749–64.
Rahman M, Bakker M. Sensor-based control in eddy current separation of incinerator
bottom ash. Waste Manage 2013;33:1418–24.
Rem P, Leest P, van den Akker A. A model for eddy current separation. Int J Miner
Process 1997;49(3–4):193–200.
Robinson B. E-waste: an assessment of global production and environmental
impacts. Sci Total Environ 2009;408(2):183–91.
Rolicz P. Eddy currents generated in a system of two cylindrical conduc-
tors by a transverse alternating magnetic field. Electr Power Syst Res
2009;79(2):295–300.
Ruan J, Li J, Xu Z. An environmental friendly recovery production line of waste toner
cartridges. J Hazard Mater 2011;185(2–3):696–702.
Ruan J, Xu Z. A new model of repulsive force in eddy current separation for recovering
waste toner cartridges. J Hazard Mater 2011a;192(1):307–13.
Ruan J, Xu Z. Environmental friendly automated line for recovering the cabinet of
waste refrigerator. Waste Manage 2011b;31(11):2319–26.
Ruan J, Xu Z. Approaches to improve separation efficiency of eddy current separa-
tion for recovering aluminum from waste toner cartridges. Environ Sci Technol
2012a;46(11):6214–21.
Ruan J, Xu Z. Risks in the physical recovery system of waste refrigerator cabinets
and the controlling measure. Environ Sci Technol 2012b;46(24):13386–92.
Schlett Z, Lungu M. Eddy-current separator with inclined magnetic disc. Miner Eng
2002;15(5):365–7.
Stone R. Confronting a toxic blowback from the electronics trade. Science
2009;325:1055.
Widmera R, Oswald-Krapf H, Sinha-Khetriwal D, Schnellmann M, Boni H. Global
perspectives on e-waste. Environ Impact Assess Rev 2005;25:436–58.
Wu J, Li J, Xu Z. Electrostatic separation for recovering metals and nonmetals from
waste printed circuit board: problems and improvements. Environ Sci Technol
2008;42(14):5272–6.
Zhang S, Forssberg E. Separation mechanisms and criteria of a rotating eddy-current
separator operation. Resour Conserv Recy 1999;25:215–32.
Zhang S, Forssberg E, Arvidson B, Moss W. Aluminum recovery from electronic scrap
by High-Force eddy-current separator. Resour Conserv Recy 2002;37:55–60.
Zhang S, Rem P. Particle trajectory simulation of two-drum eddy current separators.
Resour Conserv Recy 1999;26:71–90.
Zhou L, Xu Z. Response to waste electrical and electronic equipments in China: leg-
islation, recycling system, and advanced integrated process. Environ Sci Technol
2012;46(9):4713–24.