Article pubs.acs.org/est
Environmental Friendly Crush-Magnetic Separation Technology for Recycling Metal-Plated Plastics from End-of-Life Vehicles Mianqiang Xue, 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: Metal-plated plastics (MPP), which are important from the standpoint of aesthetics or even performance, are increasingly employed in a wide variety of situations in the automotive industry. Serious environmental problems will be caused if they are not treated appropriately. Therefore, recycling of MPP is an important subject not only for resource recycling but also for environmental protection. This work represents a novel attempt to deal with the MPP. A selfdesigned hammer crusher was used to liberate coatings from the plastic substrate. The size distribution of particles was analyzed and described by the Rosin-Rammler function model. The optimum retaining time of materials in the crusher is 3 min. By this time, the liberation rate of the materials can reach 87.3%. When the density of the suspension is 31 250 g/m3, the performance of liberation is the best. Two-step magnetic separation was adopted to avoid excessive crushing and to guarantee the quality of products. Concerning both the separation efficiency and grade of products, the optimum rotational speed of the magnetic separator is 50−70 rpm. On the basis of the above studies about the liberating and separating behavior of the materials, a continuous recycling system (the technology of crush-magnetic separation) is developed. This recycling system provides a feasible method for recycling MPP efficiently, economically, and environmentally. tendency of increased amount of plated polyamides.9 And the ingredients of the coatings are generally copper, nickel, and chromium.10 On the one hand, MPP is a rich resource due to its high-value plastics and metals. On the other hand, serious environmental problems will be caused if they are not treated appropriately.11 Therefore, it is urgent to develop technologies for recycling MPP not only for resource recycling, but also for environmental protection. At present, chemical and electrochemical processes are the common recycling methods for MPP.11 But they have some problems: (a) the process is complicated; (b) a huge amount of chemical reagent is consumed; (c) large quantities of waste gas and waste acid liquid are produced; and (d) the plastics are somewhat destroyed. Development of mechanical recycling technology is being recognized because it may be applicable both environmentally and technically.12 So a mechanical method was tried to deal with the MPP in this study. The mechanical recycling process starts with the liberation of coatings from the plastic substrate which directly influences the purity of the products and the subsequently recovery rate.12,13
1. INTRODUCTION The automotive industry is one of the largest industries in the world today.1 It has been estimated that there will be 2 to 3.5 billion light vehicles worldwide by the middle of this century.2 However, due to the ever increasing production of vehicles, the automotive industry is facing significant challenges: resource consumption and waste generation during component production and release of hazardous substances when the car becomes an end-of-life vehicle (ELV).3−6 As various resources are rapidly being depleted, recycling and recovery of end-of-life products are considered as one of the most important methods to promote sustainable development. Unfortunately, only 3% or 4% of plastics contained in ELVs are really recycled because of the economic or technological limits of the recycling process.7 In the automotive industry, MPP are increasingly employed in a wide variety of situations.8 For instance, door handles, letterings, grills, and manufacturers’ badges are all made of plastics and covered with metals. For this reason, a huge amount of waste MPP has been generated during component electroplating and production. In fact, the presence of coatings is important from the standpoint of aesthetics or even performance. Furthermore, a plated piece of plastic is easy to give it any shape by molding it.9 Currently, more than 90% of all plated plastic parts are made of either acrylonitrilebutadiene-styrene (ABS) or ABS/polycarbonate, with the © 2012 American Chemical Society
Received: Revised: Accepted: Published: 2661
August 18, 2011 February 1, 2012 February 2, 2012 February 2, 2012 dx.doi.org/10.1021/es202886a | Environ. Sci. Technol. 2012, 46, 2661−2667
Environmental Science & Technology
Article
Figure 1. The technology proposed for recycling of WPP.
Figure 2. The structure and comminuting model of the hammer crusher.
0.16 wt %, respectively. The MPP were supplied by a local electroplate factory. The dimension of the materials was 180 ×150 ×80 mm, as shown in Figure 1. 2.2.1. Methods. Materials Crushing. The substrate of the MPP used in the automotive industry has high hardness and strength. Besides, many of the parts have great threedimensional scale with lower thickness. Coatings could be easily stripped when the MPP are subjected to shear force and impulsive force. Consequently, a self-designed hammer crusher was employed in this study. The structure and comminuting model of the crusher were shown in Figure 2. The materials were crushed in four ways: (a) impact crushing by the rotating hammer with high speed; (b) grinding by the rotating hammers and the lining plate; (c) collision between particles and lining plate; (d) interparticle collision. Figure 2 shows the force analysis of the particles. According to a previous study,18 the equation for momentum is as follows:
In order to achieve liberation of coatings, crushing was applied in this study. Magnetic separation is a clean technology for separating ferromagnetic materials from nonferromagnetic materials: there is not any pollution medium; lower energy is consumed in the separating process. For this reason, magnetic separation is widely used in the recycling and mining industries.14−17 The substrate of MPP was plastics, on which was electroplated with Cu, Ni, and Cr. The coatings are influenced by the magnetic field. Therefore, magnetic separation was used to separate coatings from plastics. In this study, a new technology was developed with the aim of recycling plastics and metals from MPP, one of the typical parts of ELVs, as shown in Figure 1. The technology contained crushing and two-step magnetic separation. Key technological parameters were determined in the processes of materials liberating and separating. This technology points out a nonpolluting, economic, and efficient way for recycling resources from MPP.
Fc =
2. EXPERIMENTAL SECTION 2.1. Materials. The MPP used in this study were typical of that used in the automotive industry for interior plastic parts, door handles, lettering, grills, and badges. The substrate was ABS, on which was electroplated with Cu, Ni, and Cr. The content of ABS, Cu, Ni, and Cr were 85.72, 12.73, 1.39, and
mn2r 2 91.28 × d
(1)
Where m is the mass of MPP, n is the rotational speed of the hammer, r is radius of the rotor, and d is the minimum dimension of the plastic. The materials were crushed when the energy that they obtained was higher than their impact strength. 2662
dx.doi.org/10.1021/es202886a | Environ. Sci. Technol. 2012, 46, 2661−2667
Environmental Science & Technology
Article
2.2.2. Materials Sieving. In order to study the size distribution and liberation characteristics of the MPP, comminuted materials were sieved by a screen machine. Sieve analysis is accomplished by passing a known weight of sample through finer sieves and weighting the amount collected on each sieve. According to the preliminary test, the materials can be liberated with a diameter of 5 mm. So the size grade is divided into seven levels: 1 # (+5 mm), 2 # (−5 + 3.2 mm), 3 # (−3.2 + 2.5 mm), 4 # (−2.5 + 1.6 mm), 5 # (−1.6 + 0.8 mm), 6 # (−0.8 + 0.45 mm), and 7 # (−0.45 mm). 2.2.3. Materials Separating. Magnetic separation was employed to sort the coatings from the plastics in this study. The magnetic force was determined by the following equation: Fmag =
m χBgradB μo
mm). The mass fraction of grade 1 was 45.48%. While the mass fractions of grade 6 (−0.8 + 0.45 mm) and grade 7 (−0.45 mm) were 4.81 and 4.86%, respectively. When the materials were crushed for 2 min, the grade 2 and grade 3 particles increased dramatically. The mass fraction of particles in grade 2 and grade 3 contributed to 23.69 and 21.68%, respectively. For 3 min, most of the particles were in grade 2. Mass fraction of particles in grade 1 decreased to 20.10%. From the fourth minute to the sixth minute, samples showed a significant proportion of small grade particles. The mass fraction of particles in grade 4, grade 5, grade 6, and grade 7 were increased, but the increment was not obvious. In the early stage of shredding, the rate of breakage of coarse particles was high. But with the time going it slowed down. Temperature control is a big problem in the comminuting process. One solution is to reduce as far as possible the mean resident time of materials in the crusher. Moreover, there is an enrichment process after comminuting. Hence, it is suggested that the optimum comminuting time is 3 min. The average particle size of the materials was determined by the following equation:
(2)
where m is the mass of particle, kg; μo is vacuum permeability, N/A2; χ is the magnetic susceptibility of the particle, m3/kg; B is the magnetic flux density, T; gradB is the gradient of magnetic flux density, T/cm. Figure 3 shows the diagram of
D=
∑in= 1 di × mi ∑in= 1 mi
(3)
where di is the average particles size of each grade, mm; mi is particle mass of each grade. After crushing for 3 min, the average particle size of the materials was decreased to 4.17 mm. The size distribution function (model) is the most accurate method to describe the materials size distribution in the comminuting studies. Rosin-Rammler, Gaudin-Schuhmann, and the normal distribution function are the three most commonly used functions.19 In this study, the Rosin-Rammler function, which is applicable to many crushers and millers is used to simulate the size distribution of MPP particles that is crushed for 3 min. The function is defined as follows: w DP = 100exp( −bDPn)
Figure 3. The schematic diagram of magnetic separator.
(4)
where wDp is the cumulative oversize ratio, %; b is a constant related to crushing; n is evenness index; Dp is particle size. n (1.1607) and b (0.1711) were obtained by the least-squares method according to the size distribution of particles comminuted for 3 min. So the size distribution function can be expressed as wDP = 100 exp(−0.1711DP1.1607). The correlation coefficient is 0.9603, which demonstrated that the model is effective. The size distribution model can be used to compute the characteristic parameters of particles. 3.1.2. Liberation Mechanism. Liberation is the objective of materials comminuting in the recycling process, which directly influences the quality of recycling streams. Figure S1 in the Supporting Information (SI) illustrates the liberation characteristics of MPP. It is seen that the shapes of the coatings are lamellate and angular while the shapes of the plastics are mostly globular and granular. As shown in Figure S1 of the SI, the materials were completely liberated with particle size decreasing to 1.6 mm. Figure 5 presents the liberation models of MPP. Crushing is a complicated process with size change of materials: (a) plastics were crushed, and the production was dominated by middle size particles. With the process of crushing, middle size particles were gradually smashed into small particles; (b) plastics were crushed from the material surface, and small particles were constantly stripped. The combination of the two models constitutes the crushing process in practice.
magnetic separator. Nonferromagnetic particles are affected by the gravity force and centrifugal force while ferromagnetic particles are subjected to gravity force, centrifugal force, and magnetic force. For ferromagnetic particles, they adhered to the surface of the belt if the magnetic force acting on them is higher than the opposing gravity force and centrifugal force. And then they were collected by the coatings collector. Meanwhile, nonferromagnetic particles fell into the plastic collector. When coatings are separated from the plastics, middlings were enriched simultaneously, which can be delivered back to the crusher. In this case, excessive crushing could be avoided to some degree. Since magnetic separation is a technology with low energy consumption, multilevel magnetic separation was attempted to guarantee the quality of products and enhance the stability of the process.
3. RESULTS AND DISCUSSION 3.1.1. Materials Liberation Part. Size Distribution. The determination of size distribution of materials is critical for the crushing process control. After crushing, the samples were sieved and their size distributions were determined. Figure 4 illustrates the size distribution of the MPP scraps that were crushed for different times. When the materials were crushed for 1 min, the scrap was dominated by coarse particles (+4 2663
dx.doi.org/10.1021/es202886a | Environ. Sci. Technol. 2012, 46, 2661−2667
Environmental Science & Technology
Article
Figure 4. The size distribution of the comminuted MPP.
Figure 5. Liberation models of MPP.
2664
dx.doi.org/10.1021/es202886a | Environ. Sci. Technol. 2012, 46, 2661−2667
Environmental Science & Technology
Article
Figure 6. Curves of liberation rate via time (a) and load (b).
Figure 7. The effect of rotational speed on separation efficiency and grade: (a) plastics; (b) coatings.
the loading of the impact crusher was 750 g. The damage was mainly due to impact crushing and grinding. The density of the suspension (DS) being impacted by the hammers was introduced in this study. DS is the mass of materials divided by the volume of the crusher chamber. From the above, we come to the conclusion that the optimal DS is 31 250 g/m3. This parameter can be an important reference for the charging rate in the continuously recycling process. 3.2.1. Materials Separation Part. Determination of the Steps. This experiment was carried out with a rotational speed of 60 rpm. Figure S2 of the SI presents the effect of steps on the separation efficiency. The price of the separated materials depends heavily on its purity. To guarantee the grade of plastics to 97%, the separation efficiency of plastics and coatings were only 74 and 83% after level-one magnetic separation. However, the separation efficiency of plastics and coatings can reach 94 and 96% after level-two magnetic separation. In addition, the increase of separation efficiency becomes slower and slower after level-two magnetic separation as shown in Figure S2 of the SI. Hence, two-step magnetic separation was adopted concerning both the energy consumption and the quality of products. 3.2.2. Determination of the Rotational Speed. After crushing for 3 min, the scraps were separated by the magnetic separator. The experiments were conducted in 1000 g scale. Figure 7 presents the effect of rotational speed on separation efficiency and grade. With the increase of rotational speed, the
Figure 6a illustrates the liberation rate of the materials as a function of comminuting time. It is observed that the liberation rate increased noticeably from the first minute to the third minute. When the materials were crushed for 3 min, the liberation rate of the materials can reach 87.3%. This certifies that the impact crushing is an effective method for the liberation of the MPP. However, it is also observed that the third minute is a critical point. After the third minute, the trend of the increase of the liberation rate becomes slow. Consequently, an enrichment process (magnetic separation) was employed from both an economic as well as environmental point of view. The liberated materials (coatings and plastics) were separated from the mixed scrap. And the unliberated materials could be delivered back to the impact crusher by a screw conveyor for further liberation. In order to further understand the liberation mechanism of the materials, 500, 750, 1000, 1250, and 1500 g MPP were crushed successively for 3 min. Figure 6b illustrates the liberation rate of materials under different loadings. On the one hand, a significant decline of the liberation rate was observed when the loadings increased. When the crusher was full of materials, the damage was mainly due to extruding in the crushing process. This resulted in a lower liberation rate as shown in Figure 6b. On the other hand, the effect of materials liberation was also not ideal if too few MPP were fed. In this case, the interaction between hammers and materials was inadequate. It is clear that the liberation rate was highest when 2665
dx.doi.org/10.1021/es202886a | Environ. Sci. Technol. 2012, 46, 2661−2667
Environmental Science & Technology
Article
Figure 8. The flowchart for processing MPP.
Table 1. Economic Evaluation of the Recycling System in Industrial Scale cost ($/t) service life(year)
power (KW)
coatings
plastics
salary
raw materials
electric power
depreciation cost
maintenance cost
profit ($/t)
10
60
2484.5
1987.0
10.8
1187.5
15
3.2
0.8
417.9
3.3.2. Technical Analysis. To avoid high temperature inside the crusher, the residence time of the materials were controlled to 2−3 min. Further, the circulating water system was fixed on the hammer crusher. Then the temperature of the materials in the recycling stream was controlled to 60 °C. However, dust could be produced during the recycling process. They can be collected by a bag-type dust collector fixed into the system. Taking the liberation characteristics of the materials and energy consumption in the whole recycling process into account, the size of screen hole was designed to 4 mm. According to the size distribution model, 59% of materials passed though the screen and enriched by the magnetic separator. For magnetic separation, the optimum rotational speed was 50−70 rpm. Concerning its economical advantages, two-step magnetic separation was employed in the recycling system to guarantee the purity of the separated products. The performance of the recycling system was evaluated by the purity of products and the recycling rate. The grade of plastics and coatings are 97 and 84%, respectively. The recycling rate of plastics and coatings are 79 and 82%, respectively. 3.3.3. Economic Assessment. Taking a recycling system (a production capacity of 0.6 t/h) as a case study, the power of the crusher and the magnetic separator were 40 and 10 kW, respectively. Table 1 presents the results of the economic evaluation of the recycling system. The cost of the equipment is $38 819. The cost of electric power applying to industry is $0.15/kWh in Shanghai, P. R. China. Four workers are needed and the salary of one worker is $1.62/h. There are 250 working days in a year. The maintenance cost is estimated as 25% of the depreciation cost. The profit calculation is detailed in the SI. As shown in Table 1, the profit is $417.9/t. Consequently, this is an economically feasible method to deal with the MPP. The scenarios in other countries can be easily obtained according to the economic analysis method applied in this study. In summary, the technology of crush-magnetic separation was developed to settle the problem of MPP. On the one hand, resources could be recovered efficiently. On the other hand, secondary pollution (which may be caused by reducing gas and acid liquid generated in the acid-washing process) can be avoided. Therefore, this new process provides a nonpolluting, economic, and efficient way for recycling resources from MPP.
separation efficiency of plastics increased while the grade of plastics decreased. When the rotational speed was 20 rpm, the grade of plastics can reach 99.2%. However, the separation efficiency was only 60%. When the rotational speed increased to 100 rpm, the separation efficiency of plastics was 98% but the grade was only 85%. Compared with the separating characteristics of plastics, an opposite trend was observed for coatings. This is because that the centrifugal force (Fc = mv2r‑1) is proportional to the square of the velocity while the magnetic force (Fmag = mχμo−1BgradB) and the gravity force (Fg = mg) are not related to the rotational speed. If the magnetic separator was working with a high speed, then quite a few coatings and middlings fell off the belt in advance. In contrast, if the magnetic separator was working with a low speed, then the discrimination between middlings and plastics was insufficient. For one thing, the separation efficiency of plastics and grade of coatings could be improved by increasing the rotational speed. For the other thing, the rotational speed must be reduced to control the separation efficiency of coatings and grade of plastics. Therefore, the optimum rotational speed is 50−70 rpm concerning both the separation efficiency and the grade of products. The rotational speed of the level-one magnetic separator can be set as 70 rpm to guarantee the grade of coatings. For a level-two magnetic separator, the rotational speed can be set as 50 rpm to ensure the grade of plastics. The rotational speed of each step could be adjusted, which enables the magnetic separation to be a flexible way to treat WPP. Another interesting phenomenon was observed in this study. The separation efficiency and grade of plastics were more sensitive to the rotational speed of the magnetic separator. The possible reason for this result was that the characteristic of the middlings was close to that of plastics. 3.3.1. Novel Process for Recycling of MPP. The Continuous Recycling System. On the basis of the above studies about liberating and separating behavior of the materials, a continuous recycling system (the technology of crush-magnetic separation) is developed, as presented in Figure 8. First, the MPP were crushed and the coatings were stripped from the substrate. Then the coatings were separated from the liberated plastics by two-step magnetic separation. Finally, the middlings that were not liberated completely were delivered back to the crusher for further liberation. The liberated plastics can be reutilized by granulating and the coatings can be used as medium alloy. It is important to notice that there is no wastewater generated in the recycling system. Therefore, MPP can be disposed by this process without negative effect to the environment.
■
ASSOCIATED CONTENT
S Supporting Information *
Figure showing the liberation characteristics of MPP; Figure showing the effect of steps on the separation efficiency; Figure 2666
dx.doi.org/10.1021/es202886a | Environ. Sci. Technol. 2012, 46, 2661−2667
Environmental Science & Technology
Article
showing the average annual value of crushed ABS; description of the profit calculation. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Tel:+86 21 54747495; fax:+86 21 54747495; e-mail: zmxu@ sjtu.edu.cn. Notes
The authors declare no competing financial interest.
■
REFERENCES
(1) Orsato, R. J.; Wells, P. The automobile industry & sustainability. J. Clean Prod. 2007, 15, 989−993. (2) Mcauley, J. W. Global sustainability and key needs in future automotive design. Environ. Sci. Technol. 2003, 37, 5414−5416. (3) Xiang, W.; Ming, C. Implementing extended producer responsibility: Vehicle remanufacturing in China. J. Clean Prod. 2011, 19, 680−686. (4) Vermeulen, I.; Vancaneghem, J.; Block, C.; Baeyens, J.; Vandercasteele, C. Automotive shredder residue (ASR): Reviewing its production from end-of-life vehicles (ELVs) and its recycling, energy or chemicals’ valorization. J. Hazard. Mater. 2011, 190, 8−27. (5) Maclean, H. L.; Lave, L. B. Life cycle assessment of automobile/ fuel option. Environ. Sci. Technol. 2003, 37, 5445−5452. (6) Park, C. H.; Jeon, H. S.; Yu, H. S.; Han, O. H.; Park, J. K. Application of electrostatic separation to the recycling of plastic wastes: Separation of PVC, PET, and ABS. Environ. Sci. Technol. 2008, 42, 249−255. (7) Froelich, D.; Maris, E.; Haoues, N.; Chemineau, L.; Renard, H.; Abraham, F.; Lassartesses, R. State of the art of plastic sorting and recycling: Feedback to vehicle design to vehicle design. Miner. Eng. 2007, 20, 902−912. (8) Shipway, P. H.; Bromley, J. P. D.; Weston, D. P. Removal of coatings from polymer substrates by solid particle blasting to enhance reuse or recycling. Wear 2007, 263, 309−317. (9) Middeke, H. J. Plating on plastics Part I history, application of metal plated plastics, kinds of plastics. Electroplat. Fin. 2005, 1, 35−39. (10) Nagashima, T.; Akiyama, H.; Namihira, T. Recycle of metalplating on plastics by pulse arc discharges. Plasma Sci. 2007, 8, 1471− 1476. (11) Qiu, Z. M.; Liu, Z. W. Research progress on the domestic recycling method of ABS plastic electro-plating. Jangxi Energy 2010, 3, 1−3 (in Chinese). (12) Zhang, S.; Forssberg, E. Mechanical separation-oriented characterization of electronic scrap. Resour. Conserv. Recycl. 1997, 21, 247−269. (13) Eswaraiah, C.; Kavitha, T.; Vidyasagar, S.; Narayanan, S. S. Classification of metals and plastics from printed circuit boards (PCB) using air classifier. Chem. Eng. Process. 2008, 47, 565−576. (14) Ruan, J. J.; Li, J.; Xu, Z. M. An environmental friendly recovery production line of waste toner cartridges. J. Hazard. Mater. 2011, 185, 696−702. (15) Huang, K.; Li, J.; Xu, Z. M. A novel process for recovering valuable metals from waste nickel-cadmium batteries. Environ. Sci. Technol. 2009, 43, 8974−8978. (16) Watson, J. H. P.; Younas, I. Superconducting discs as permanent magnets for magnetic separation. Mater. Sci. Eng. B-Adv. Funct. SolidState Mater. 1998, 53, 220−224. (17) Augusto, P. A.; Augusto, P.; Grange, T. G. Magnetic classification. Miner. Eng. 2002, 15, 35−43. (18) Lu, H. Z.; Li, J.; Guo, J.; Xu, Z. M. Pulverization characteristics and pulverizing of waste printed circuit boards based on resource utilization. J. Shanghai Jiaotong Univ. 2007, 04, 551−556 (in Chinese). (19) Duan, C. L.; Zhao, Y. M.; He, J. F.; Zhou, N. X. Research on liberation mechanism of the impact crushing waste printed circuit board. Adv. Mater. Res. 2010, 113−116, 730−734. 2667
dx.doi.org/10.1021/es202886a | Environ. Sci. Technol. 2012, 46, 2661−2667
