Electrostatic Separation for Recycling Conductors, Semiconductors

Article pubs.acs.org/est

Electrostatic Separation for Recycling Conductors, Semiconductors, and Nonconductors from Electronic Waste Mianqiang Xue, Guoqing Yan, Jia Li, and Zhenming Xu* School of Environmental Science and Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, People’s Republic of China S Supporting Information *

ABSTRACT: Electrostatic separation has been widely used to separate conductors and nonconductors for recycling e-waste. However, the components of e-waste are complex, which can be classified as conductors, semiconductors, and nonconductors according to their conducting properties. In this work, we made a novel attempt to recover the mixtures containing conductors (copper), semiconductors (extrinsic silicon), and nonconductors (woven glass reinforced resin) by electrostatic separation. The results of binary mixtures separation show that the separation of conductor and nonconductor, semiconductor and nonconductor need a higher voltage level while the separation of conductor and semiconductor needs a higher roll speed. Furthermore, the semiconductor separation efficiency is more sensitive to the high voltage level and the roll speed than the conductor separation efficiency. An integrated process was proposed for the multiple mixtures separation. The separation efficiency of conductors and semiconductors can reach 82.5% and 88%, respectively. This study contributes to the efficient recycling of valuable resources from e-waste.

1. INTRODUCTION The global amounts of e-waste increase significantly in these years and continue rising at a faster rate.1−3 Because of high content of valuable resources and absent facilities, primitive techniques tend to be used for recovering metals in the economically underdeveloped areas.4,5 Without much concern for adverse impacts on both labor safety and environment, these activities will threaten the public health and ecosystem seriously.6−9 As a result of this challenge presented by e-waste, developing nonpolluting and efficient recycling technologies is of considerable significance to enhance the comprehensive utilization of resources and to promote cyclic economy development.10 Separation is a vital component of recycling, particularly for mechanical treatment methods. Electrostatic separation is a promising technology providing many advantages, such as simplicity, low energy consumption, and no wastewater discharge.11 Charge mechanisms and particle movement behavior have been well-investigated.12−14 Moreover, researchers have done a lot of work on the optimization of the separating process, since its efficiency is influenced by electrical, mechanical, material, and environmental factors.15−17 At present, electrostatic separation already has been proved to be efficient for recycling conductors and nonconductors18,19 and has been applied in industrial scale with remarkable economic and environmental benefits.20,21 However, the compositions of e-waste are complex. They can be classified as conductors, semiconductors, and nonconductors © 2012 American Chemical Society

according to their conducting properties. Semiconductor materials, the foundation of modern electronics, are increasingly employed in a wide variety of situations. Unfortunately, their purification from minerals is a complicated process with high pollution. In general, silicon, germanium, and gallium arsenide are the most commonly used ones. Their conductivity is intermediate between that of a nonconductor and a conductor. This unique property implies the potential for sorting the conductors, semiconductors, and nonconductors by electrostatic separation. In this context, we make a novel attempt to cope with mixtures containing conductors (copper), semiconductors (extrinsic silicon), and nonconductors (woven glass reinforced resin) using an electrostatic separator. In this work, we try to establish an efficient and costless process to recycle the conductors, semiconductors, and nonconductors from e-waste. First, the separation of binary mixtures (conductor and semiconductor mixed particles; semiconductor and nonconductor mixed particles; conductor and nonconductor mixed particles) was studied. The influencing factors of the separation process including particle size, high voltage level, and roll speed were investigated. Second, charging mechanism and force analysis were conducted to guide the application of the recycling process. Third, on the Received: Revised: Accepted: Published: 10556

May 7, August August August

2012 26, 2012 27, 2012 27, 2012

dx.doi.org/10.1021/es301830v | Environ. Sci. Technol. 2012, 46, 10556−10563

Environmental Science & Technology

Article

Figure 1. Preparation of granular materials: (a) silicon; (b) woven glass reinforced resin.

angular position of electrostatic electrode, α2 = 75° (Supporting Information, Figure S1). 2.3. Method. First, the binary mixtures separation tests were carried out. Effects of size grade, high voltage level, and roll speed on separation efficiency were investigated. The size grade was divided into five levels: 1 # (−0.125 + 0.045 mm), 2 # (−0.3 + 0.125 mm), 3 # (−0.45 + 0.3 mm), 4 # (−0.6 + 0.45 mm), 5 # (−0.8 + 0.6 mm). Figure S2 of the Supporting Information presents the sample of particles with different size grade. Parameter settings were presented in Table 1. Second,

basis of the analytical results, an integrated process was proposed for the multiple mixtures (conductor, semiconductor, and nonconductor mixed particles) separation. This work enriches the application of electrostatic separation. The multiple mixtures separating methodology can be applied to other separation processes in the recycling industry.

2. EXPERIMENTAL SECTION 2.1. Materials. Binary mixtures and multiple mixtures are involved in the present work. For binary mixtures, it refers to the following categories: (i) copper and woven glass reinforced resin mixed particles; (ii) silicon and woven glass reinforced resin mixed particles; (iii) copper and silicon mixed particles. For multiple mixtures, it consists of copper, silicon, and woven glass reinforced resin. The samples were synthetic. The materials were obtained from woven glass reinforced resin board and silicon wafer, as shown in Figure 1. The characteristics of the silicon wafer that was used for integrated circuits production were illustrated in Table S1 of the Supporting Information. After crushing, the scraps were sieved by a screen machine. 2.2. Apparatus. A roll-type corona electrostatic separator was used for separation tests in the present study. As shown in Figure S1 of the Supporting Information, it has a grounded rotating roll electrode and two active electrodes (wire-type corona electrode and electrostatic electrode) connected to a DC high voltage supply (DHV-50 kV/20 mA). A high intensity electric field is generated between the grounded and active electrodes. The particulate mixture is fed onto the electric feeder that ensures a monolayer of material on the surface of the rotating roll. Particles with different electrical conductivity have different movement behavior. The products were recovered in several collecting tanks. Electrostatic separation is a typical multifactorial process. The output of the process could be controlled by regulating input variables. In addition to be adjusted easily, high voltage level (U) and roll speed (n) have significant effects on the separation efficiency.16 Moreover, particle size can result in some problems about robustness of the separation process. In fact, the crushed wastes contain granules with different sizes in industrial application. Hence, high voltage level, roll speed, and size grade were selected as variables for the separation tests. The other parameters were fixed: radial position of corona electrode, s1 = 70 mm; angular position of corona electrode, α1 = 25°; radial position of electrostatic electrode, s2 = 90 mm;

Table 1. Parameter Settings of the Binary Mixtures Separation Processa group 1 2 3

separation of C and N

separation of S and N

G = 1, 2, 3, 4, 5; U = G = 1, 2, 3, 4, 5; U = 20 ; n = 50 20 ; n = 50 U = 15, 20, 25, 30, U = 15, 20, 25, 30, 35; G = 3; n = 50 35; G = 3; n = 50 n = 35, 50, 65, 80, n = 35, 50, 65, 80, 95; G = 3; U = 20 95; G = 3; U = 20

separation of C and S G = 1, 2, 3, 4, 5; U = 5 ; n = 80 U = 2.5, 5, 7.5, 10, 12.5; G = 3; n = 80 n = 50, 60, 70, 80, 90; G = 3; U = 5

a

C: conductors; N: nonconductors; S: semiconductors; G: size grade; U (kV): high voltage level; n (rpm): roll speed.

on the basis of the above studies on binary mixtures separation, the multiple mixtures separation tests were conducted by employing a two-stage and a multiple-stage separation process. The weight of each fraction was measured by an electronic balance with resolution 0.1 g. All separation tests were conducted at a relative humidity of 40−50% and a temperature of 20−25 °C.

3. RESULTS AND DISCUSSION 3.1. Separation of Conductors and Nonconductors. The tests were conducted with 20 g of copper and 75 g of woven glass reinforced resin mixed particles. As shown in Figure 2, the separation efficiency of small particles (grade 1) and large particles (grade 5) are low with more middlings. On the one hand, small particles have large specific area. Particles can adhere to the active electrodes, which results in low electrical field strength. Moreover, some copper particles may be enfolded by woven glass reinforced resin particles. In that case, the electrical conductivity of the copper particle is weakened. On the other hand, the gravity force has a greater influence on particles with larger diameter. Large nonconductive particles detach from the roll electrode early. 10557



Article

Figure 2. Separation of copper and woven glass reinforced resin: the mass of product as a function of size grade (a), high voltage level (b), and roll speed (c); copper separation efficiency as a function of size grade (d), high voltage level (e), and roll speed (f).

reinforced resin mixed particles separation (shown in Figure 3). More specifically, satisfying separation results were achieved when the particle size ranged from 0.125 to 0.45 mm. The mass of middlings is greater at a lower voltage level and a higher roll speed. Remarkably, the silicon separation efficiency is more sensitive to high voltage level and roll speed than the copper separation efficiency. By taking high voltage level as an example, when the high voltage level increases from 15 to 35 kV, the silicon separation efficiency increases from 76% to 98% while the copper separation efficiency increases from 87% to 98%. This phenomenon can be explained by the following reasons. One is that the electrical conductivity of copper is better than that of silicon. The difference between copper and woven glass reinforced resin are more significant than that between silicon and woven glass reinforced resin. The other is that the density of copper is much greater than that of silicon. For copper, the gravity force will play a bigger role in the movement behavior during the separation process. 3.3. Separation of Conductors and Semiconductors. The tests were conducted with 20 g of copper and 5 g of silicon mixed particles. As shown in Figure 4, the mass of silicon

Large conductive particles fall into the collecting tanks without far enough horizontal distance. The mass of middlings decreases with the increase of high voltage level, accompanied by an increase of copper recovering efficiency. Obviously, high voltage level leads to high electrical field strength. The electric force acting on the particle is enhanced. Hence, the separation results are better at a higher voltage level. However, it should be noted that too high voltage level will cause spark discharge. The mass of middlings increases gradually with increasing of the roll speed. Simultaneously, the mass of copper also increases. Copper separation efficiency increases from 89.5% to 98.5%. The reason is that centrifugal force increases if the roll electrode runs faster. Some woven glass reinforced resin particles fall into the middlings tanks and copper particles move far horizontally. 3.2. Separation of Semiconductors and Nonconductors. The tests were conducted with 5 g of silicon and 75 g of woven glass reinforced resin mixed particles. As a whole, the trend of silicon and woven glass reinforced resin mixed particles separation was the same as that of copper and woven glass 10558



Article

Figure 3. Separation of silicon and woven glass reinforced resin: the mass of product as a function of size grade (a), high voltage level (b), and roll speed (c); silicon separation efficiency as a function of size grade (d), high voltage level (e), and roll speed (f).

Charging mechanism is very important for better understanding the electrostatic separation process. The charging properties of conductors, semiconductors, and nonconductors are quite different. Figure 5 presents the corona charge and induction charge models in the separation process. On the one hand, the air around the corona electrode is ionized, and the electron current flows to the grounded roll electrode. All the particles are negatively charged by the ion bombardment when they pass through the ionized space. The conductors discharge rapidly to the roll electrode. However, the nonconductors hardly discharge due to the broad forbidden band, and the semiconductors release partial negative charge. On the other hand, the particles are polarized during the transit time. The charge can distribute uniformly on the surface of the conductors within a short time, and the negative charge can release to the roll electrode rapidly. Nevertheless, due to the distinctiveness of conducting properties, the semiconductors induce less charge. For nonconductors, the charge distributes and releases slowly, and they are not affected by the electrostatic induction.

decreases with the increase of size grade. That is also the case for copper. Separation efficiency of copper and silicon reach a maximum when the particle size is −0.125 + 0.045 mm. It indicates that small particle size is appropriate for the separation of copper and silicon. An increasing high voltage level results in a dramatic increase of middlings. In contrast, the mass of copper and silicon both decrease. The copper separation efficiency decreases from 64% to 19%. Thus, a low voltage level is needed to separate the copper and silicon mixed particles. When the roll speed increases from 50 to 90 rpm, the copper separation efficiency increases from 34% to 77%. The mass of silicon increases by 51.7%. Meanwhile, the mass of middlings decreases by 66%. It is interesting that the mass of copper increases significantly while the mass of silicon increases gradually when the roll speed increases at a high level. Hence, there is not much room for improving the separation efficiency of silicon with a high level roll speed. 3.4. Charging Mechanism in the Electrostatic Separation Process. According to the above analytical results, we find that the electrostatic separation is a complicated process associated with material, electrical, and mechanical factors. 10559



Article

Figure 4. Separation of copper and silicon: the mass of product as a function of size grade (a), high voltage level (b), and roll speed (c); copper separation efficiency as a function of size grade (d), high voltage level (e), and roll speed (f).

3.5. Force Analysis in the Electrostatic Separation Process. During the separation process, particles may be subject to the electric field force (Fe), gravity force (Fg), air drag force (Fd), electric image force (Fi), and centrifugal force (Fc) as shown in Figure 6. From the viewpoint of mechanics, the trajectories of particles depend on the resultant force. Fg, Fd, Fi, and Fc are applying to the woven glass reinforced resin. The particles are detaching from the rotating roll electrode when the radial forces achieve equilibrium: Fg sin ϕ + Fi = Fc

angle ranging from 0° to 90°. The equilibrium equation of the critical point can be expressed as follows: Fg sin ϕ + Fi − Fc − Fe = 0

(2)

where Fc = 0.832QE,22 and E is the electric field strength. When the voltage is set at a low level, the main charging mode is the induction charge. The copper particle is subject to a high electric field force and departures with a large angle. The tangential velocities of the particles are the same because the particles rotate with the roll electrode before detachment. Nevertheless, the horizontal component of velocity is high for particles with large detachment angle. Hence, the copper and silicon are thrown from the roll electrode, and they are collected by different collecting tanks. The electric field force is proportional to the electric field strength. The centrifugal force is proportional to the roll speed. Hence, the separation process can be controlled by regulating the input variables including the high voltage level and roll speed. 3.6. Separation of Conductors, Semiconductors, and Nonconductors. To separate the multiple mixtures containing

(1)

where Fg = mg, Fi = (Q2)/(4πε0h2), Fc = mRω2, m is mass, g is gravitational acceleration, Q is charge, ε0 is vacuum dielectric constant, h is the thickness of flat particle, R is the radius of the roll electrode, and ω is angular velocity. When the voltage is set at a high level, the concentration of electron current is high. If Fg + Fi > Fc when φ = −90°, then the woven glass reinforced resin particle rotates with the roll electrode and at last are cleared by a brush. Experimental results show that copper and silicon particles do parabolic movement with a detachment 10560



Article

Figure 5. Charge model of nonconductor (N), semiconductor (S), and conductor (C).

Figure 6. Schematic diagram of force analysis.

Figure 7. Flowchart of electrostatic separation of copper, silicon, and woven glass reinforced resin by a two-stage process.

copper (20 g), silicon (5 g), and woven glass reinforced resin (75 g) with size −0.45 + 0.3 mm, experiments were performed through a two-stage separation process (Figure 7) on the basis of the above mechanism analysis and force analysis. In the first stage, the parameters of electrostatic separator were set as U = 30 kV and n = 50 rpm. Woven glass reinforced resin (73.7 g)

was separated from the mixed particles. In the second stage, the parameters of the electrostatic separator were set as U = 5 kV and n = 80 rpm. This stage aims to separate copper from silicon. The recovery of copper and silicon were 11.9 and 3.7 g, respectively. The middlings (1.8 g of product B1 and 8.9 g of 10561



■

product B2) were delivered back to the electrostatic separator-1 for further separation with new feedstock. To improve the efficiency of the multiple mixtures separation, we proposed a multiple-stage separation process (Figure 8) given the fact that there was still a good deal of

Article

ASSOCIATED CONTENT

S Supporting Information *

Table showing the characteristics of silicon wafer and figures showing the schematic diagram of roll-type corona electrostatic separator and the sample of particles with different size grades. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +86 21 54747495; fax:+86 21 54747495; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National High Technology Research and Development Program of China (863 program 2009AA06Z318; 2012AA063206) and the National Natural Science Foundation of China (21077071; 51008192).

■

REFERENCES

(1) Robinson, B. T. E-waste: An assessment of global production and environmental impacts. Sci. Total Environ. 2009, 408, 183−191. (2) Widmer, R.; Oswald-Krapf, H.; Sinha-Khetriwal, D.; Schnellmann, M.; Boni, H. Global perspectives on e-waste. Environ. Impact Assess. Rev. 2005, 25, 436−458. (3) Babu, B. R.; Parande, A. K.; Basha, C. A. Electrical and electronic waste: a global environmental problem. Waste Manage. Res. 2007, 25, 307−318. (4) Nnorom, I. C.; Osibanjo, O. Overview of electronic waste (ewaste) management practices and legislations, and their poor applications in the developing countries. Resour., Conserv. Recycl. 2008, 52, 843−858. (5) Wong, M. H.; Wu, S. C.; Deng, W. J.; Yu, X. Z.; Luo, Q.; Leung, A. O. W.; Wong, C. S. C.; Luksemburg, W. J.; Wong, A. S. Export of toxic chemicalsA review of the case of uncontrolled electronic-waste recycling. Environ. Pollut. 2007, 149, 131−140. (6) Stone, R. Confronting a toxic blowback from the electronics trade. Science 2009, 325, 1055. (7) Ogunseitan, O. A.; Schoenung, J. M.; Saphores, J. M.; Shapiro, A. A. The electronics revolution: from e-wonderland to e-wasteland. Science 2009, 326, 670−671. (8) Fu, J. J.; Zhou, Q. F.; Liu, J. M.; Liu, W.; Wang, T.; Zhang, Q. H.; Jiang, G. B. High levels of heavy metals in rice (Oryza sativa L.) from a typical E-waste recycling area in southeast China and its potential risk to human health. Chemosphere 2008, 71, 1269−1275. (9) Leung, A. O. W.; Duzgoren-Aydin, N.; Cheung, K. C.; Wong, M. Heavy metals concentrations of surface dust from e-waste recycling and its human health implications in southeast China. Environ. Sci. Technol. 2008, 42, 2674−2680. (10) Betts, K. Producing usable materials from e-waste. Environ. Sci. Technol. 2008, 42, 6782−6783. (11) Wei, J.; Realff, M. J. Design and optimization of free-fall electrostatic separators for plastics recycling. AIChE J. 2003, 49, 3138− 3149. (12) Dascalescu, L.; Morart, R.; Luga, A.; Samuila, A.; Neamtu, V.; Suarasan, L. Charging of particulates in the corona field of roll-type electro-separators. J. Phys. D: Appl. Phys. 1994, 27, 1242−1251. (13) Lindley, K. S.; Rowson, N. A. Charging mechanisms for particles prior to electrostatic separation. Magn. Electr. Sep. 1997, 8, 101−113. (14) Lu, H. Z.; Li, J.; Guo, J.; Xu, Z. M. Movement behavior in electrostatic separation: Recycling of metal materials from waste printed circuit board. J. Mater. Process. Technol. 2008, 197, 101−108. (15) Medles, K.; Tilmatine, A.; Miloua, F.; Bendaoud, A.; Younes, M.; Rahli, M.; Dascalescu, L. Set point identification and robustness

Figure 8. Flowchart of electrostatic separation of copper, silicon, and woven glass reinforced resin by a multiple-stage process.

product B2 after the two-stage separation process. Experiments were conducted using multiple mixtures containing copper (20 g), silicon (5 g), and woven glass reinforced resin (75 g) with size −0.45 + 0.3 mm. In the first stage, the parameters of the electrostatic separator were set as U = 30 kV and n = 50 rpm. In the second and the third stages, the parameters were set as U = 5 kV and n = 80 rpm. The recoveries of copper and silicon were 16.5 and 4.4 g, respectively. The middlings (1.6 g of product B1 and 3.9 g of product B3) were delivered back to the electrostatic separator-1 for further separation. Compared with the twostage separation process, the separation efficiency of copper increased from 59.5% to 82.5%. For silicon, separation efficiency increased from 74% to 88%. The grade of copper and silicon can reach 90% and 95%, respectively. The multiple mixtures containing conductors, semiconductors, and nonconductors can be recycled successfully by the proposed separation process. Remarkably, the process can be expected to have low energy consumption, and there is not any pollution medium. It is of significance from the aspects of economic benefits, pollution mitigation, and waste recycling: (i) conductors could be further purified by an environmentally friendly vacuum metallurgy technology to get pure metal;23 (ii) semiconductors that are separated from the mixtures can be reused as silicon source instead of minerals; (iii) nonconductors that are obtained from e-waste can be recycled as raw material of wood plastic composite24 or as additive of the modified asphalt.25 10562



Article

testing of electrostatic separation processes. IEEE Trans. Ind. Appl. 2007, 43, 718−625. (16) Wu, J.; Li., J.; Xu, Z. M. Optimization of key factors of the electrostatic separation for crushed PCB wastes using roll-type separator. J. Hazard. Mater. 2008, 154, 161−167. (17) Zhang, S.; Forssberg, E. Optimization of electro-dynamic separation for metals recovery from electronic scrap. Resour., Conserv. Recycl. 1998, 22, 143−162. (18) Dascalescu, L.; Morar, R.; Iuga, A.; Samuila, A.; Neamtu, V. Electrostatic separation of insulating and conductive particles from granular mixes. Part. Sci. Technol. 1998, 16, 25−42. (19) Li, J.; Xu, Z. M.; Zhou, Y. H. Application of corona discharge and electrostatic force to separate metals and nonmetals from crushed particles of waste printed circuit boards. J. Electrost. 2007, 65, 233− 238. (20) Li, J.; Xu, Z. M. Environmental friendly automatic Line for recovering metal from waste printed circuit boards. Environ. Sci. Technol. 2010, 44, 1418−1423. (21) Xue, M. Q.; Yang, Y. C.; Ruan, J. J.; Xu, Z. M. Assessment of noise and heavy metals (Cr, Cu, Cd, Pb) in the ambience of the production line for recycling waste printed circuit boards. Environ. Sci. Technol. 2012, 46, 494−499. (22) Félici, N. J. Forces et charges de petits objets en contact avec une électrode affectée d’un champ électrique. Rev. Gen. Electr. 1996, 75, 1145−1160. (23) Zhan, L.; Xu, Z. M. Application of vacuum metallurgy to separate pure metal from mixed metallic particles of crushed waste printed circuit board scraps. Environ. Sci. Technol. 2008, 42, 7676− 7681. (24) Guo, J.; Tang, Y.; Xu, Z. M. Wood plastic composite produced by nonmetals from pulverized waste printed circuit boards. Environ. Sci. Technol. 2010, 44, 463−468. (25) Guo, J. Y.; Guo, J.; Wang, S. F.; Xu, Z. M. Asphalt modified with nonmetals separated from pulverized waste printed circuit boards. Environ. Sci. Technol. 2009, 43, 503−508.

10563


Electrostatic Separation for Recycling Conductors, Semiconductors

Recommend Documents