Triboelectrostatic separation of covering plastics in chopped waste electric wire.
Nowadays, the generation of waste electric wire such as power transmission cable, control cable, transport cable, communication cable used in the electric power, electronics, and communication industries is constantly increasing because of industrial development and short life span of these equipments [1, 2]. Generally, electric wire consists of metal conductors such as copper or aluminum and polymer insulators such as polyvinyl chloride (PVC) and cross-linked polyethylene (XLPE).
Figure 1 shows a flowchart of the waste electric wire recycling process. After shredding and granulating electric cable, copper, and aluminum in coarse size are separated using an air classifier and shaking table and then particulate metals are recovered by induction electrostatic separation, but the covering plastics are accumulated in the fields or mostly land-filled [3, 4]. Hence, these cause economic loss and environmental contamination. Especially, PVC in covering plastics pollutes the environment by generating hazardous hydrogen chloride gas and dioxins containing chlorine. Also, it decreases recycling ratio of plastics by forming compounds or deteriorating the nature of other materials even if a small quantity of PVC is present in the plastics [5-8].
The landfill and incineration of waste plastics have been restricted by law and the recycling of waste plastics has been promoted by the enforcement of the extended producer responsibility (EPR) system of Korea since the year 2003 . Therefore, the development of separation technique for covering plastics in waste electric wire is of great importance from the environmental standpoint, particularly in Korea that occupies limited land with a large population.
In general, physical separation techniques that can recycle mixed plastics are classified as electrostatic separation, dry, and wet gravity separation, froth flotation, near infrared ray (NIR), and color sorting [10-12]. Hu and Calo  separated heavy group (PVC, PET, PS, PC) from light group (HDPE, LDPE, PP) using water as the fluidization medium. Dodbiba et al.  recovered polyethylene terephthalate (PET) from PET/PP (polypropylene) or PET/PE in PET bottle by a(the) combination of sink-float separation and flotation technique. Also, PVC (95-100% recovery) from a PVC/PET mixture of variable composition using a NaOH solution was separated by Drelich . Marques and Tenorio  reported that PVC of 99.3% purity was separated from PET, using reagents such as calcium lignin sulfonate and methyl isobutyl carbinol (MIBC) . Wet gravity separation and froth flotation of waste plastics are considered costly compared to dry separation. In a wet separation process, mixed PVC, PET, and rubber raise more difficulties because of similar specific gravity and flotation agents may cause water disposal problems [12, 15].
[FIGURE 1 OMITTED]
In separation techniques by NIR spectroscopy, an array multichannel detector made of InGaAs semiconductor was introduced to measure some hundred NIR spectra per second [17, 18]. The potentiality of a reliable distinction between 5 major plastics (PE, PET, PP, PS, and PVC) in household garbage was reported by Th. Huth-Fehre . Wienke et al.  obtained a median sorting purity of better than 98% for nonblack plastics. At this time, more than 75 samples per second can be identified by a combination of InGaAs diode array and neural network. At least 80% of plastic material and nonplastic material in municipal solids were recognized by Van den Broeka . However, this method is a difficult task because of the close similarities between the materials and it needs a further reduction of shadow contributions, and stabilization of sensor or light source to obtain reproducible measurements [19-21]. Also, a color sorting technique has difficulty to separate particles of mixed plastics having similar properties such as color [6, 22].
The electrostatic methods to separate mixed materials include corona discharge, electrostatic induction, and triboelectrostatic separation. Corona discharge and electrostatic induction can separate a mixture of conductor and nonconductor (metal/insulator), whereas the triboelectrostatic method has an advantage of separating different types of materials [23, 24]. Tribo-charging phenomenon is utilized in numerous technical applications such as electro-photography, electrostatic copy and printing techniques, electrostatic filtration, precipitation, and coloring. Also, this separation has been used for processing valuable minerals such as coal and fly ash [6, 25].
Various previous works on triboelectrostatic separation of plastics regarding tribo-charger types, charge density measurement, and process variables have been described in the literature [26-30]. Pearse and Hickey  and Yanar and Kwetkus  measured the charge density of plastics with a nickel and copper cyclone, using Faraday cage in consideration of air velocity and relative humidity. Fujita et al.  and Higashiyama and Asano  measured the charge density of plastics with vibrating feeder material. Inculet et al.  separated PVC, Nylon, PE and acrylic, using a fluidized bed and rotating tube. Matsushita and Sometani  separated PVC, PE, PP, and PS using rotary blades and confirmed that separation efficiency increases with increasing mixing time, rotation speed and with decreasing relative humidity. They have reported that the separation efficiency increases with decreasing relative humidity and with decreasing air velocity and electric field [15.1-15.6].
Tribo-charging occurs when particles are charged with opposite polarity by particle-particle and particle-surface charge mechanisms because of their work function or triboelectric series [23, 31]. Table 1 shows the tribo-electric series of common plastics which represent the degree of work function of materials . Figure 2 shows the process of tribo-charging and separation of materials. When two materials are brought into contact or collision, charge transfer can occur between them until their Fermi levels equalize by the work function difference between the two materials. Then, material with high work function and low work function are charged positively and negatively, respectively: negatively charged material is moved toward positive electrode and vice versa [33, 34]. Figure 2(a) shows the triboelectrification of particle-particle, and Fig. 2(b) shows that of particle-charger surface. Triboelectrostatic separation can improve separation efficiency according to the development of charging material and tribo-charger. Triboelectrostatic separation is much cheaper and the separation efficiency is much better than that using the above mentioned classical separation methods [25, 31, 33].
[FIGURE 2 OMITTED]
This study aims at recovering PVC from mixed covering plastics in chopped electric wire using a pipeline-fluidized bed tribo-charger. Hence, we designed a bench scale triboelectrostatic separator unit and investigated several factors such as air velocity, mixture ratio, relative humidity on the charge polarity and charge density of PVC/XLPE. Also, the optimal electrode potential and splitter position affecting the separation efficiency in a triboelectrostatic separator has been standardized. A triboelectrostatic separator unit with a capacity of 350 kg/h has been set up for scale-up tests. Trial runs have been conducted to optimize separation technique, including improving charging property.
Virgin XLPE and PVC materials 4 mm in average diameter were obtained from a petrochemical plant and used in charging property tests. Covering plastic feedstocks of -8 + 1 mm size were obtained from Econics, a recycling company. The covering plastic samples were a residue remaining after removing metals using an air classifier, shaking table and induction electrostatic separator. The covering plastics contained 9.45% XLPE and 90.54% PVC. The samples were shredded using a cutting mill (pulverisette 19, Fritsch GmbH, Germany) and then sieved into a fraction of -4.0 + 1.6 mm. The plastic particles charged with a cutting mill neutralized the initial charges on each particle by means of ionized air produced with a discharger (Kasuga Denki, Japan).
Eight kinds of materials, that is polytetrafluoroethylene (PTFE), PVC, PP, high-density polyethylene (HDPE), copper (Cu), PET, acrylonitrile butadiene styrene (ABS), and polymethyl methacrylate (PMMA) were used as a material for tribo-charger in charging material selection tests.
Figure 3 shows a schematic diagram of a triboelectrostatic separator unit used in this work. It consists of feeding zone (6 7 8 9), charging zone (1), and separation zone (2 3 4 5). The air-conditioning system supplies necessary air for fluidizing and moving covering plastic samples at a certain relative humidity and temperature. Charged particles in charging zone are separated under the influence of the electric field between the electrodes that are connected to a high-voltage power supply ([+ or -]30 kV).
[FIGURE 3 OMITTED]
The charge density was calculated by counting positive and negative charges that were collected in a Faraday cup and was determined by charge to mass ratio. A Faraday cage (Model KQ-1400, Kasuga Dencki Inc., Japan) was used in the range [+ or -]1 to [+ or -] 9999 nC. The weight of particles was measured with an electronic balance (Sartorius BP 2100s).
Charging Property Tests Using Virgin Samples
In studying charging property using virgin samples, several factors affecting the charging efficiency of particles such as charging material, relative humidity, velocity, and charging mechanism were studied.
Particularly, in charging material selection tests, a single 20 g charge of PVC or XLPE was fed into a pipeline charger made of PTFE, PVC, PP, HDPE, Cu, PET, ABS, or PMMA in the constant conditions of 30% relative humidity and 12 m/s air velocity. Then their charge polarity and charge density were measured with a Faraday cage. Hence, PP material that generates high charging efficiency with opposite polarity for PVC and XLPE was selected as the material for the charger. Tests were repeated at least three times and the average charge density was chosen.
The relative humidity was controlled from 20 to 70% with a dehumidifier (Model AD0502XA, Whirlpool, USA) at 12 m/s air velocity with the PP charger.
In charging tests on the effect of air velocity and charging mechanism, individual XLPE and PVC or a 1:1 mixture of XLPE and PVC were fed into the PP charger at 30% relative humidity. To improve the separation efficiency of the mixed plastics and to gain accurate density value of individual product, a voltage of over 30 kV was applied in the charge density measurement of mixed samples.
[FIGURE 4 OMITTED]
Separation Tests With Covering Plastic Samples
Figure 4 shows a flowsheet for the separation of mixed covering plastics. Particles crushed in the optimum size of -4.0 + 1.6 mm with a cutting mill and sieves were fed into a pipeline charger with air. Charged particles were deflected under the influence of electric field, split and collected in a storage compartment. Therefore, a function of electrode potential for separation of covering plastics was tested out with a PP charger at 12 m/s air velocity and 30% relative humidity. At this time, electrode potential was adjusted from 10 kV to 30 kV.
Also, change tests on splitter position were performed at 25 kV electrode potential in the same condition aforementioned. The splitter positions were moved by 2 cm toward positive electrode or negative electrode from the center. The impurity in one type of the plastic product was handpicked with naked eyes and the weights of both the remaining product and impurity were measured to determine the grade of the product.
RESULTS AND DISCUSSION
Charging Material Selection
First, Fig. 5 shows the results of the charging material selection tests on XLPE and PVC plastics. XLPE and PVC were charged in the tribo-charger made of various charging materials at 30% relative humidity and 12 m/s air velocity. Then, the charge polarity and charge density were measured with the Faraday cage. The results indicated that the XLPE was charged positively and the PVC negatively when PP was used as the charging material. However, both of them were charged with the same polarity in all the other cases. This is attributed to the work function of PP laid between PVC and XLPE in the tribo-electric series . Both XLPE and PVC are charged positively with the charging material of PTFE because their work function is lower than that of PTFE, whereas they are charged negatively with PET, Cu, ABS and PMMA. On the other hand, the difference in charge density between XLPE and PVC is the maximum with PP: the charge density of XLPE was +3.6 nC/g and that of PVC -7.5 nC/g and so the charge density difference totaled 11.1 nC/g. Therefore, we selected the charger made of PP for further tests to separate XLPE and PVC.
[FIGURE 5 OMITTED]
Figure 6 shows the effect of relative humidity on the charge density of XLPE and PVC in tribo-charging tests using a PP charger. These tests were carried out in the relative humidity range 20-70% at a constant air velocity of 10 m/s. The charge densities of XLPE and PVC were decreased as the relative humidity increased in the opposite direction. The effect of relative humidity on the charging and discharging behavior of the plastics can be explained by the formation of water films onto the plastic surface. Nemeth  reported that water molecules in the atmosphere can adsorb onto charged plastic surface and form a layer and then its charge is discharged. The change of charge density by relative humidity is a result of disturbing surface polarization between particles by water. In other literature, the polarity of particles can be changed to opposite charge. Nonetheless, we found that relative humidity affected greatly the charge density of XLPE and PVC without changing its polarity. As shown in the figure the separation efficiency was decreased as the relative humidity increased, probably due to the discharge of the electron through the moisture layer attached on the surface. Therefore, it can be concluded that relative humidity is a very important factor for tribo-charging and the separation of mixed plastics in triboelectrostatic separation.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
Air Velocity and Charging Mechanism
Figure 7 shows the effect of air velocity on tribo-charging of XLPE particle-PP charger surface, PVC particle-PP charger surface, and a 1:1 mixture of XLPE particle-PVC particle at a constant 30% relative humidity. As shown in Fig. 7, all the magnitudes of the charge density were increased with increasing air velocity. An increase in air velocity in the fluidized bed may cause an increase in the impact force and frequency between the particles and their contacting surfaces [23, 36].
The charge densities of a 1:1 mixture of XLPE and PVC by contacting particle to particle were +9.8 and -13.3 nC/g, respectively at 12 m/s air velocity. On the other hand, those densities of individual PVC and XLPE by contacting particle to charger were only +3.5 and -7.5 nC/g, respectively. In light of the results, the tribo-charging of both particle-particle and particle-surface in the case of a XLPE and PVC mixture can occur at the same time. To better understand tribo-charging mechanisms as shown in Fig. 2, the results indicate that the dual charging mechanism is more effective than that of particle-surface alone.
Effect of Electrode Potential on Separation Efficiency (Covering Plastics)
The covering plastics charged in the tribo-charger are separated by splitter position and the electric field of electrodes. Hence, we investigated the effect of electrode potential and splitter position on PVC recovery from mixed covering plastics in chopped electric wire. As previously shown in charging property tests, XLPE and PVC are charged positively and negatively, respectively. If a charged particle is approximated into the electric field which has been formed between the electrodes, electrostatic force is interacted by Coulomb's law . Therefore, charged XLPE and PVC will be deflected toward negative and positive electrode, respectively.
[FIGURE 8 OMITTED]
Figure 8 shows PVC grade and PVC recovery as a function of electrode potential. The tests were carried out with the PP charger at 12 m/s air velocity and 30% relative humidity. The PVC grade and PVC recovery were increased as the electrode potential increased. For example, At 10 kV electrode potential, the PVC grade was 93.1% and its recovery 82.5%. It appears that a electrode potential of 10 kV is not strong enough to pull the charged particles toward the electrodes. 99.50% PVC grade and 98.05% PVC recovery were successfully obtained at 25 kV. The results show that the PVC grade and PVC recovery are increased with increasing electrode potential since the charged particle and electrode potential seem to directly related to the separation efficiency. If the charge density of a particle is high, it can be deflected although electrode potential is relatively low. For the purpose of improving the separation efficiency, electrode potential has to increase in order that a particle with low charge density can be deflected.
A Function of Splitter Position
Figure 9 shows PVC grade and PVC recovery as a function of splitter position. These tests were carried out at 12 m/s air velocity, 30% relative humidity and 25 kV electrode potential using the PP charger. Also, "-" and "+" cm sign in the splitter position in Fig. 9 signify moving direction from the center to the negative electrode and the positive electrode, respectively. As shown in Fig. 9, the PVC grade were increased as the splitter position was moved to the positive electrode from the center, and PVC recovery increased as the splitter position was moved to the negative electrode. The falling position of particles can vary, depending on charging factors such as triboelectric series, mixture ratio, air velocity and relative humidity under a definite gravity force, drag force and electrostatic force [23, 27]. PVC particles which have high negative charge density are strongly deflected to the positive electrode but some PVC particles which have low or neutral charge density fall freely or to the opposite side.
Also, in case of XLPE, some XLPE particles are not deflected to negative electrode and behave similarly to PVC. Such behavior of particles deteriorates the separation efficiency for PVC and XLPE. A PVC grade of 99.50% and recovery of 98.05% were obtained at the splitter position 2 cm from the center to the positive electrode, which seems to be the optimum position. However, A PVC purity of 99.9% could be obtained at the splitter position between +4 cm and the positive electrode although the PVC recovery was decreased considerably.
Relationship Between Charge Density and Separation Efficiency
Figure 10 shows the effect of charge density on the separation efficiency of PVC from a mixture of XLPE and PVC in the triboelectrostatic separation unit. A net-charge density of separation products is measured with the Faraday cage. The results indicate that the PVC grade and PVC recovery are increased as the net-charge density rises gradually. The particle which has high charge density can be easily deflected toward the electrode although its potential is low. Therefore, we confirmed that the optimum charge density of materials is most important parameter in triboelectrostatic separation.
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
On the basis of the systematic approach presented in this work, a scale-up triboelectrostatic separator unit has been set up on the plant site of Econics and trial runs are being performed. This apparatus has a capacity of 350 kg/h. Until now, PVC grade and PVC recovery were about 98.50 and 98.40%, respectively.
Continued economic operations are expected since the triboelectrostatic separator is much cheaper and the separation efficiency is much better than conventional separation methods. However, the charge efficiency and separation efficiency of plastics in electrostatic separation are sensitive to the change of relative humidity. Controlling relative humidity below 20% in a plant is not easy because of the limited capacity of the dehumidifier. Generally, charging and separating particles are sufficient at 30-40% relative humidity, but more effective at a relative humidity lower than that. Also, the plant equipment can be tolerable to a maximum relative humidity of 40%.
For future plan, actual charging technique to optimize charge density and selective charging will be developed. The optimal electrode potential and splitter position affecting separation efficiency in the scale-up triboelectrostatic separator will be determined.
In this study, we have designed a bench scale triboelectrostatic separator for the separation of covering plastics in waste electric wire, and studied the effects of several factors on the charge polarity and charge density of plastics in the tirbo-charger and the effects of electrode potential and splitter position on separation efficiency. In charger material selection tests, the charge polarity and charge density of XLPE and PVC were very effective with the tirbo-charger made of PP with the decrease in relative humidity and the increase in air velocity. The dominant charging mechanism in the tribo-charger was the charging of a combination of particle-particle and particle-charger surface rather than that of particle-charger surface.
The PVC grade and its recovery considerably depend on electrode potential and splitter position. A PVC grade of 99.50% and a recovery of 98.05% are achieved at 25 kV and the splitter position +2 cm from the center. A high purity PVC of over 99.9% could be obtained at the splitter position between +4 cm and the positive electrode at the expense of reduced recovery. In scale-up tests with a triboelectrostatic separator unit, a 98.50% PVC grade and 98.40% PVC recovery were successfully achieved.
1. J. Marcher, Wire J. Int., 17, 106 (1984).
2. S. Zhang and E. Forssberg, Resour. Conserv. Recy., 21, 247 (1997).
3. R. Koehnlechner, WAI 71st Annual Convention, Atlanta, GA, USA (2001).
4. C.C. White, J. Wagenblast, and M.T. Shaw, Polym. Eng. Sci., 40, 863 (2000).
5. H.S. Jeon, C.H. Park, B.G. Kim, and J.K. Park, J. Korean. Inst. Resour. Recy., 15, 28 (2006).
6. V. Gente, F.L. Marca, F. Lucci, and P. Massacci, Waste Manage., 23, 951 (2003).
7. J.C. Arnold and B. Maund, Polym. Eng. Sci., 39, 1234 (1999).
8. M.Y. Wey, L.J. Yu, and S.I. Jou, J. Hazard. Mater., 60, 259 (1998).
9. T. Lindhquist, International Workshop on Extended Producer Responsibility (EPR), Seoul, Korea, June 27 (2003).
10. R.H. Yoon, Processing International Symposium on Establishment of Resource Recycle Society, Seoul, Korea, October 1-2 (2002).
11. APC Durables Recycling Workshop III. Arlington, VA, USA, November 9 (1999).
12. R.D. Pascoe and B. O'Connell, Miner. Eng., 16, 1205 (2003).
13. X. Hu and J.M. Calo, AlChE. J., 52, 1333 (2006).
14. G. Dodbiba, N. Haruki, A. Shibayama, T. Miyazaki, and T. Fujita, Int. J. Miner. Process., 65, 11 (2002).
15. Drelich, T. Paynte, J.H. Kim, and J.D. Miller, Polym. Eng. Sci., 38, 1378 (1998).
16. G.A. Marques and J.A.S. Tenorio, Waste. Manag., 20, 265 (2000).
17. Th. Huth-Fehre, R. Feldhoff, Th. Kantimm, L. Quick, F. Winter, K. Cammann, W. van den Broek, D. Wienke, W. Melsen, and L. Buydens, J. Mol. Struct., 348, 143 (1995).
18. D. Wienke, W. van den Broek, W. Melsen, L. Buydens, R. Feldhoff, K. Kantimm, T. Huth-Fehre, L. Quick, F. Winter, and K. Cammann, Anal. Chim. Acta., 317, 1 (1995).
19. W.H.A.M. van den Broeka, D. Wienkeb, W.J. Melssenb, and L.M.C. Buydens, Anal. Chim. Acta., 361, 161 (1998).
20. P. Tatzer, M. Wolf, and T. Panner, Real-Time Imaging, 11, 99 (2005).
21. E.N. Lewis and I.W. Levin, JMSA, 1, 35 (1995).
22. W. Jing and M.J. Realff, AlChE. J., 49, 3138 (2003).
23. E.G. Kelly and D.J. Sottiswood, Miner. Eng., 2, 33 (1989).
24. Y. Higashiyama and K. Asano, Particul. Sci. Technol., 16, 77 (1998).
25. T.X. Li, H. Ban, J.C. Hower, and K. Saito. J. Electrostat., 47, 133 (1999).
26. M.J. Pearse and T.J. Hickey, Resour. Recov. Conserv., 3, 179 (1978).
27. D.K. Yanar and B.A. Kwetkus, J. Electrostat., 35, 257 (1995).
28. T. Fujita, Y. Kamiya, N. Shimizu, and T. Tanaka, Processings of 3th International Symposium on East Asian Recycling Technology, November 21-24 (1995).
29. I.I. Inculet, G.S.P. Castle, and J.D. Brown, Part. Sci. Technol., 16, 91 (1998).
30. Y. Matsushita, N. Mori, and T. Sometani, Electr. Eng. Jap., 127, 33 (1999).
31. G.L. Hearn and J.R. Ballard, Resour. Conserv. Recy., 44, 91 (2005).
32. A.F. Diaza, J. Electrostat., 62, 277 (2004).
33. M. Lungu, Miner. Eng., 17, 69 (2004).
34. E.G. Kelly and D.J. Sottiswood, Miner. Eng., 2, 193 (1989).
35. E. Nemeth, V. Albrecht, and G. Schubert. J. Electrostat., 58, 3 (2002).
Chul-Hyun Park, (1,2) Ho-Seok Jeon, (2) Bong-Gyoo Cho, (2) Jai-Koo Park (1)
(1) Department of Geoenvironmental System Engineering, Haengdangdong, Hanyang University, Seongdonggu, Seoul, 133-791 Republic of Korea
(2) Korea Institute of Geoscience and Mineral Resources, Gajeongdong 30, Yuseonggu, Daejeon, 305-350 Republic of Korea
Correspondence to: H.S. Jeon; e-mail: firstname.lastname@example.org
Contract grant sponsor: Resource Recycling R & D Center, Ministry of Science and Technology, Korea.
TABLE 1. Triboelectric series of plastics. Materials Series Polytetrafluoroethylene (PTFE) More negative Polyvinyl chloride (PVC) Polypropylene (PP) Low density Polyethylene (LDPE) High density Polyethylene (HDPE) Polyethylene terephthalate (PET) Rubber Polystyrene (PS) Polycarbonate (PC) Acrylonitrile butadience styrene (ABS) Polymethyl methacrylate (PMMA) Polyurethane More positive
|Printer friendly Cite/link Email Feedback|
|Author:||Park, Chul-Hyun; Jeon, Ho-Seok; Cho, Bong-Gyoo; Park, Jai-Koo|
|Publication:||Polymer Engineering and Science|
|Article Type:||Technical report|
|Date:||Dec 1, 2007|
|Previous Article:||A short review on rubber/clay nanocomposites with emphasis on mechanical properties.|
|Next Article:||Molecular, rheological, and crystalline properties of low-density polyethylene in blown film extrusion.|