Aspects of the recovery of nonferrous metals from solid residues by magnetic separation.

Nemes, Toderita ; Petrescu, Valentin ; Deac, Cristian 等

Abstract: The paper presents some considerations on the process of separating nonferrous metals from solid residues, using the method of magnetic separation in specially designed separators. This method is based on the induction, in the nonferrous particles, of eddy currents generated by a variable magnetic field originating in a rotating barrel. The efficiency of the separation process was evaluated for different metals and process parameters.

Key words: nonferrous metals, magnetic separation, eddy currents

1. INTRODUCTION

The recovery of nonferrous metals and their separation with the help of eddy currents is based on the phenomenon of inducing electric currents by a variable magnetic field, currents that exercise attraction forces on the metallic particles.

The most frequently employed systems are HRD-type separators with eddy currents, with horizontal rotating barrel, where the active part is a rotating barrel covered with strings of permanent magnets of alternating polarity, placed parallel to the barrel axis. The separator's feeding is done by means of a conveyer band that brings the particles to the barrel, where they are accelerated at different rates, according to electrical conductivity. There can appear, however, problems with separating conductive particles of nonferrous metals with sizes less than 5 mm from other non-conductive particles. This inconvenience can be solved by using a dynamic separator with eddy currents and with sloped barrel (IDECS) (Lungu & Schlett, 1997; Nemes & Petrescu, 2004).

The IDECS separator consists of a rotating barrel covered with an alloy of NdFeB and with permanent magnets alternatively oriented N-S and S-N, fastened directly on the shaft of an electric motor.

The barrel's vertical position allows the mounting of magnets with lengths of only 4 ... 6 mm and the separation of nonferrous metallic particles from mixtures with low conductivities.

The particles to be separated are brought in the magnetic field by the conveyer belt on a trajectory that is sloped both in the horizontal plane and in the vertical plane and they fall on the surface of the rotating barrel. On the particles there acts the combined effect of the deviation caused by the collision with the barrel and of the electrodynamic forces generated by the variable magnetic field, so that the strongly conductive particles are deviated at greater distances than the less conductive ones and thus they can be separated. The vertical sloping of the barrel allows a stronger deviation of strongly conductive particles after the exit from the separation area.

2. THEORETICAL PRINCIPLES

At the IDECS separator, as in the case of the other separators with eddy currents, the variable magnetic field generated by the rotating barrel induces eddy currents in the conductive nonferrous metallic particles moving inside the barrel. According to Faraday's law, eddy currents are induced as consequence of the magnetic field's high variation frequency, and they propel the particle by means of the Lorentz forces. The particle's trajectory is given by the size and orientation of the magnetic field, but also by the own rotation and translation motion.

When the particle's dimensions are small compared to the width d of a barrel pole, the variations of the field induced in the particle are smooth and the force F and the torque T can be expressed function of the field's gradient and the magnetic moment M of the particle (Lindley & Rowson, 1997).

The torque T sets the particle in a rotation motion in the same direction as the one of the magnetic field and in the opposite direction from the rotor. As a result, the forces acting on the particle get smaller and the particle closes in on the rotor.

The particle's final deviation depends on the tangential force [F.sub.t], on the torque T but also on the deviation produced by the impact with the barrel's surface.

The separation process is strongly influenced by the separation factor [sigma]/[rho]. The values of this factor for the main nonferrous metals Al, Cu, Zn, Pb are given in table 1 (Rezlescu & Barbu 1989):

3. EXPERIMENTAL RESEARCHES

3.1 The IDECS separator

Figure 1 shows the principle scheme of an IDECS separator made of following main parts (Lungu, 2005):

--the barrel 1 is made of weakly magnetic steel and covered with permanent magnets, alternately oriented N-S and S-N;

--the electric motor 2 ensures a speed that can be adjusted between 0 and 4500 rpm;

--the motor shaft 3, on which the barrel is fastened with a slope [[alpha].sub.2] from the horizontal

The material to be separated is brought in the active area of the magnetic field by means of a conveyer belt that provides for the particles a trajectory sloped by [[alpha].sub.1] from the horizontal.

The values of [[alpha].sub.1] and [[alpha].sub.2] are determined by successive tests for each type of residue, so that weakly conductive particles have a minimal deviation ([d.sub.1]) from the vertical plane and the strongly conductive particles have a maximal deviation ([d.sub.2]).

The strongly conductive particles (with high separation factor) that reach the active area of the magnetic field, fall, under the influence of the electromagnetic force Ft, of the torque T and the impact with the surface of the rotating barrel, in compartment III of the collecting device, while the weakly conductive particles fall in compartment I. Compartment II will contain an intermediary product containing particles of both types. This is then resubjected to separation.

[FIGURE 1 OMITTED]

The angle [beta] between horizontal and the particle's incident trajectory (figure 2) on the barrel surface and the barrel's rotation speed n (rpm) are the main functional parameters of the separator.

3.2 Materials subjected to the magnetic separation For the experimental determinations, two groups of nonferrous residues were used:

--type I residues--mixture of Zn-Cu with 75% Zn particles of 3 ... 6 mm in size and irregular shapes and Cu wires with the length of 5 ... 7 mm and diameters of 2.2 mm (25%);

--type II residues--mixture of Pb-Al with 70% Pb particles of 3 ... 5 mm in size and irregular shapes and Al wires with the length of 3 ... 5 mm and diameters of 1,5. ... 2 mm (30%).

3.3 Experimental results

In order to quantify the efficiency of the separation process, experiments were made for various values of the engine speed (n = 3000; 3500; 400 and 4500 rpm) and of the angle [beta] (15[degrees]; 30[degrees]; 45; 60[degrees]).

After the separation, the quantities of materials collected in the compartments I and III were weighed and the following parameters were calculated:

--G (%) the ratio between the mass of a metal's fraction in the mixture--the whole quantity collected in a compartment--and the residue's overall mass;

--R (%) the ratio between the mass of a metal's fraction in the mixture and the mass of the same metal's fraction in the feed.

The results of the experiments for the two types of residues are presented in tables 2, 3 and 4.

[FIGURE 2 OMITTED]

4. CONCLUSIONS

Following the analysis of results, it can be concluded that:

--for a given value of angle [beta], the maximal separation degree is obtained at an intermediate value of engine speed, n=4000 rpm, since at higher speeds conductive particles are repelled stronger, hit weakly conductive particles and fall into compartment II;

--at the same barrel speed, the separation degree is maximal for an incidence angle [beta] = 30[degrees];

--through an adequate positioning of [d.sub.1] and [d.sub.2], the product collected in compartment II might contain particles in strict proportion to that of the fed residue, and thus it is possible to pass it again through the separator, without changing the separator's parameters.

5. REFERENCES

Lindley, K.S., Rowson, N.A. (1997) Charging of Particles Prior to Electrostatic Separation, Mangnetic and Electrical Separation, vol. 8, no. 2, p. 101-113, ISSN 1055-6915

Lungu, M., Schlett, Z (1997) Electrical Separation of Metals from Wastes of Printed Circuits, Journal of Mineral Resource Engineering, Vol, 6, no. 2, p.89-95.

Lungu, M. (2005) Methods for Separating Recyclable Materials (in Romanian), Publishing House of the Western University, Timisoara.

Nemes, T., Petrescu V. (2004) Materials Technology (in Romanian). Publishing House of the Lucian Blaga University, Sibiu.

Rezlescu, N., Barbu, E.B. (1989) Aplications of magnetic separation of materials, Publishing House of the Academy, Bucharest.

Table 1 Values of the separation factor

Material Al Cu Zn Pb

[sigma]/[rho] x [10.sup.3] 13.1 6.6 4.3 0.4
 [[m.sup.3]/[OHM]kg]

Table 2. G (%) and R (%) for Cu collected in compartment I

[eta]([min.sup.-1])
[beta] ([degrees]) 3000 3500 4000 4500

15 37,2 42,5 57,3 45,2
 30,6 36,3 46,4 37,8
30 55,4 64,2 78,8 66,5
 45,2 58,1 65,4 58,3
45 26,4 28,5 37,5 28,8
 20,6 24,2 28,8 23,5
60 23,2 26,7 30,2 24,7
 17,8 21,4 26,3 20,8

Table 3 G (%) and R (%) for Zn collected in compartment III

[eta]([min.sup.-1])
[beta] ([degrees]) 3000 3500 4000 4500

15 80,2 82,6 88,4 84,2
 76,5 78,9 80,9 80,6
30 89,3 92,8 94,2 93,1
 84,2 86,7 88,4 86,8
45 78,4 80,5 86,3 82,1
 71,2 75,2 78,8 78,8
60 72,5 78,4 81,2 80,3
 68,8 71,3 75,4 73,7

Table 4 G (%) and R (%) for Pb collected in compartment I

[eta]([min.sup.-1])
[beta] ([degrees]) 3000 3500 4000 4500

15 48,7 53,4 62,5 58,8
 42,4 46,2 51,7 51,2
30 62,8 73,2 82,6 79,5
 56,4 60,5 71,4 68,3
45 35,4 38,7 52,6 51,2
 26,6 30,1 48,8 46,3
60 22,8 29,4 38,4 36,4
 17,5 21,2 32,6 30,2