Shredding and liberation characteristics of refrigerators and small appliances

Waste Management xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Waste Management journal homepage: www.elsevier.com/locate/wasman

Shredding and liberation characteristics of refrigerators and small appliances Joonheon Lee a, Kihong Kim a, Heechan Cho a,⇑, Jeonghoon Ok a, Sookyung Kim b a b

Department of Energy Systems Engineering, Seoul National University, 1, Gwanak-ro, Gwanak-gu, 08826 Seoul, Republic of Korea Korea Institute of Geoscience and Mineral Resources, 124, Gwahak-ro, Yuseong-gu, 34132 Daejeon, Republic of Korea

a r t i c l e

i n f o

Article history: Received 18 July 2016 Revised 21 October 2016 Accepted 21 October 2016 Available online xxxx Keywords: Shredding Liberation Waste electrical and electronic equipment Refrigerator Small appliance

a b s t r a c t Mechanical disaggregation, or shredding, is an important part of the recycling process. Occurring at the beginning of the processing sequence, it significantly affects the efficiency of downstream processing stages. This study examines the size reduction and liberation characteristics of the single-stage shredding of household appliances to improve the efficiency and quality of the recycling process. Several disposed appliances, including 75 L refrigerators and five major categories of small appliances (vacuum cleaners, videocassette recorders (VCRs), electric rice cookers, fans, and electric heaters), were shredded using a high-speed vertical shredder under varying discharge clearance conditions. The fragments were analyzed according to size, composition, and degree of liberation. It was found that single-stage crushing with the high-speed vertical shredder was sufficient to produce fragments at an appropriate size and with a high degree of liberation. Based on the experimental results, an optimal shredding and separation scheme for the process is proposed. Ó 2016 Published by Elsevier Ltd.

1. Introduction In many countries, WEEE (waste of electrical and electronic equipment) management schemes have been established to promote the recycling of all types of electrical goods. WEEE is collected by various means and delivered to treatment plants. The treatment process at these plants generally involves: (1) manually disassembling parts and removing harmful components from appliances; (2) disintegrating them into fine fragments; and (3) recovering marketable secondary raw materials through physical separation. However, there are still several problems in WEEE recycling associated with heterogeneous components, product uncertainty, low recovery and purity of the products, economical value, and environmental pollution. For these reasons, various studies have been conducted to improve the efficiency and economics for recycling various types of WEEE, such as cathode ray tube (CRT) (Tian et al., 2016; Yoshida et al., 2016) and printed circuit boards (PCBs) (Habib et al., 2016; Huang et al., 2009; Wang and Xu, 2015). In addition, the recovery and process problem has been studied. For example, Oguchi et al. (2012) studied various metals’ distribution and flow in WEEE by using typical conditions in Japan and also discussed whether pre-separation could helpful ⇑ Corresponding author. E-mail address: [email protected] (H. Cho).

for metal recovery from small digital products; however, problems remained regarding amount and the differences between equipment. Considering these problems, recent studies tend to concentrate more on total processes and value evaluation. For example, an indepth review study about electronic recycling technology, including the potentiality of risk and economical value (Kaya, 2016), the new idea of ‘‘Control-Alt-Delete,” for dealing with the electronic waste problem in various regions (Li et al., 2015), the design of an electronic waste (e-waste) recycling process and examination for specific cases (lithium-ion batteries and PCBs) (Li et al., 2016b), the development of an eco-friendly integrated mobile recycling plant (Zeng et al., 2015), a dynamic sustainable supply model to examine the uncertainties of resources from WEEE (Gu et al., 2016), a quantitative measurement model for recyclability that can be a guideline for e-waste management (Zeng and Li, 2016), and WEEE treatment cost evaluation (Li et al., 2016a) have been conducted and suggested. Additionally, the themes of pollution and environment and public health have been dealt with (Cayumil et al., 2016; Wu et al., 2016). This study focuses on the mechanical disintegration process, which is important unit operation at recycling plants, because this operation takes place at the beginning of the processing sequence and has a big impact on the efficiency of the downstream processing stages. The primary goal of disintegration is to acquire a high

http://dx.doi.org/10.1016/j.wasman.2016.10.030 0956-053X/Ó 2016 Published by Elsevier Ltd.

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

2

J. Lee et al. / Waste Management xxx (2016) xxx–xxx

degree of liberation for the separating processes. This process is one of the main aspects of primary energy consumption in the material processing flow. Thus, there is a definite need to increase energy efficiency by applying adequate stress so that a high degree of liberation can be acquired at the coarsest size without overdisintegration. This rationale also applies to the processing of natural ores, in which valuable minerals are separated from rocks. Since more than 50% of the total energy consumption in mineral processing is in the comminution process, extensive research has been undertaken in the fields of liberation and comminution characterization, and the modeling of mineral resources (Fandrich et al., 1997; Gay, 2004; King, 1994; King and Schneider, 1998). However, few studies have been conducted on waste (Castro et al., 2005; Chao et al., 2011; Cui and Forssberg, 2007; Zhang and Forssberg, 1999). The primary reason for the lack of research is the difficulty involved in performing the same type of analysis due to the extreme heterogeneity of waste materials. For instance, unlike natural ores, waste products are not a cut and dry system with only wanted and unwanted components, but rather a mixture of numerous materials, such as plastics, metals, ceramics, and many other organic and inorganic components. In turn, shredding produces fragments composed of many materials, and categorizing these fragments in terms of size and composition is not an easy task. Moreover, these materials exhibit very different mechanical behaviors under stress. Natural ores, mostly being brittle, can easily be broken into pieces under simple compression or impact stress. On the other hand, waste products contain ductile materials that deform plastically before fracturing, which results in shape alteration without breakage, or even in size enlargement, after passing through shredding devices. Therefore, it is not possible for a single machine to efficiently reduce the size of all components. In turn, the shredding of wastes is typically conducted in multiple stages, employing a series of shearing and impacting shredders. The first stage of shredding is often conducted in a low-speed, high-torque shear shredder that shreds the feed material in a range of 150–250 mm. The secondary shredder further reduces particle size, usually using high-speed, low-torque hammer mills. In most of the waste electrical and electronic equipment (WEEE) recycling plants in Korea, shredders and impact mills are combined to create a four-stage shredding process. However, a recent analysis (Kim et al., 2014) revealed that this four-stage process is energy inefficient and potentially redundant as the last stage does not improve the degree of liberation but instead grinds up already liberated material even more. This indicates that increased efficiencies can be achieved through the effective utilization of available technology. Recently, new types of vertical hammer-style shredders have been developed so that the stages of size reduction can be achieved in a single machine. These devices use intense impact and shear force to treat almost all categories of electronic and electrical wastes, from small devices to large machines, and achieve outstanding selective disaggregation. However, detailed assessments of the effectiveness of these devices for breaking down and liberating home appliance components have not yet been conducted. The work reported here details investigations on shredding and liberation characteristics of mid-sized refrigerators and five representative categories of small appliances using a high-speed vertical shredder. Tests were conducted at various discharge clearance settings. The resulting products were analyzed in terms of size, composition, and liberation characteristics. The aim of the study is: (1) to examine the effectiveness of breaking down entire refrigerators and small appliances into small pieces in a single pass by using a high-speed vertical shredder; and (2) to find the optimum shredding conditions for different requirements and study the separation characteristics in the subsequent process.

2. Experimental procedure 2.1. Shredding device Fig. 1 illustrates the high-speed vertical shredder (Kubota KE100, 75 kW–110 kW) employed in this study. It consists of a chamber containing a rotor assembly with attached hammers of bars and gear-shaped grinders. The high-speed rotating bars located in the top of the rotor perform the primary shredding operation by coarsely crushing the bulky materials. The grinders in the lower section of the rotor shred materials into smaller sizes, carrying out the secondary and tertiary shredding. The required product sizes can be obtained by controlling the outlet clearance, which is done by adjusting the choke rings placed between the shell and the discharge ring. There are several similar models available in the market, such as BMH grinder mills (Danieli Centro Recycling) or V100 – V1000 models (Industrial Shredder). 2.2. Sample A medium-sized refrigerator (75 L) and five categories of small appliances (vacuum cleaners, videocassette recorders (VCRs), electric rice cookers, fans, and electric heaters) were chosen for this study. For each type, four to nine units were collected from recycling facilities. For refrigerator samples, the compressor, refrigerant gas, internal shelf, PCBs, and packing rubber were manually removed from the units prior to shredding in order to simulate a typical recycling facility’s process. In contrast, the samples of the small appliances were tested without any pre-removal of parts. The average dimensions and weights of the appliances are listed in Table 1. 2.3. Shredding and analysis Fragment sizes are the most important factor for determining the efficiency of separation as well as the degree of liberation. Therefore, the shredding tests were conducted at different settings of the discharge clearance (20–50 mm) to obtain shredded products with different degrees of size reduction. For refrigerators, four units were tested and fed one unit at a time into the shredder. For small appliances, three units were shredded simultaneously. After completion of shredding, all fragments were collected and analyzed in terms of size and composition. For size analysis, two steps of size fractionation were applied due to the large amount of shredded fragments. First, the shredded fragments were preclassified into the various size fractions using a large vibrating screen. These size fractions, however, contained fragments with significantly different shapes (acicular, flake, plate, granular, irregular etc.), which would result in a completely different size distribution depending on the method of measurement; thus it was necessary to define the size specially. In this study, as visual examination was used to observe the fragments, the longest dimension was used to define the size. The second step of size fractionation was conducted by measuring each fragment with a ruler and grouping them into <10 mm, 10–20 mm, 20–40 mm, 40–80 mm, 80–160 mm, 160–320 mm, and >320 mm categories. At the same time, each fragment was examined for composition. The major portion of each fragment was comprised of one type of material, including iron, plastic, urethane, aluminum, copper, PCB, electric wire, sponge, and rubber. Fragments comprised of two or more materials were classified according to material combinations. About 15–30 groups of fragments with two or more materials were identified. The <10 mm fragments were not categorized due to their presence in only a small amount and to the difficulty in visually determining their composition.

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

3

J. Lee et al. / Waste Management xxx (2016) xxx–xxx

Fig. 1. Schematic diagram of the high-speed vertical shredder (KUBOTA Environmental Service Co., Ltd., 2016).

Table 1 Average weights and dimensions of four 75 L refrigerators and nine small appliances and maximum possible units shredded per hour used in the KE-100 shredding experiment.

75 L refrigerator Vacuum cleaner VCR Electric rice cooker Fan Electric heater

Max. unit/h

Width (mm)

Length (mm)

Height (mm)

Weight (kg)

30 108 72 120 180 120

484 ± 31 298 ± 43 438 ± 4 291 ± 4 378 ± 21 216 ± 19

475 ± 6 377 ± 35 234 ± 13 388 ± 7 284 ± 106 198 ± 14

819 ± 19 340 ± 125 97 ± 2 289 ± 4 618 ± 171 380 ± 29

14.7 ± 0.6 5.1 ± 1.9 4.7 ± 0.8 4.0 ± 0.3 2.9 ± 1.0 1.9 ± 0.4

Once each fragment was analyzed for its compositional characteristics, liberation analyses were conducted in terms of the degree of liberation achieved at various settings of discharge clearance. The degree of liberation was calculated as the percentage of the fragments composed of one material containing no other materials in relation to the total weight of the produced fragments. The discharge clearance settings for achieving at least 95% liberation were identified for each type of appliance. 3. Results and discussion 3.1. Composition analysis results The material compositions of all appliance fragments are shown in Table 2. Of the fragments of the 75 L refrigerators larger than 10 mm, the vast majority of the mass was made up of iron (60%), plastic (17%), and urethane (9%). There was a minimal amount (
0%) of fragments with two or more materials for all discharge clearance settings. As mentioned previously, the compressor, PCBs, internal shelves, and other parts were removed prior to shredding, leaving only the cabinet, doors, and insulation. In consequence, no components of complex compositions remained that would produce fragments of multi-materials after shredding. Material composition was more complex for small appliances. While iron and plastic still made up the majority of the mass (65–80%), their proportions varied between the types of appliance, reflecting the different applications in which these items are used. The mass of other materials, including aluminum, copper, sponge,

rubber, etc., ranged between 10% and 20%. Also, less than 10% of the mass was fragments of mixed composition. This is understandable as the small appliances were shredded without removing any parts. Overall, the material categories were similar among appliances, indicating that small appliances can be processed together for material recovery. The locked fragments composed of multi-materials can be further liberated by further shredding. However, the additional cost of shredding may negatively impact the potential profit from producing higher quality products in the subsequent processes. The shredding process is then performed to produce an economic degree of liberation and the separation process is designed to produce a concentrate consisting mainly of one material with an accepted degree of impurities. 3.2. Fragment size distribution The fragment size distributions of the 75 L refrigerator are shown in Fig. 2 for various discharge settings. In most cases, the 20–40 mm and 40–80 mm size ranges are the two main ranges accounting for more than 70% of the fragments. In the case of the 50 mm discharge clearance, the dominant mode of the size distribution occurred at 40–80 mm. As the discharge setting was reduced, the amount of 40–80 mm fragments decreased while more 20–40 mm fragments were produced. Under the 20 mm discharge clearance condition, the 20–40 mm range fragments accounted for the vast majority of fragments, making up 56% of the total.

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

4

J. Lee et al. / Waste Management xxx (2016) xxx–xxx

Table 2 Average composition (wt.%) of waste refrigerators and small appliances obtained by visual examination of fragments after shredding.

Iron Plastic Urethane Aluminum Copper PCB E. wire Sponge Rubber Mixture Ceramic Magnet Others <10 mm

75 L Refrigerator

Vacuum cleaner

VCR

E. rice cooker

Fan

Electric heater

60.26 16.75 9.29 0.47 0.42 0.00 0.22 0.21 0.00 0.00 0.00 0.00 0.00 12.4

18.46 60.66 0.00 0.86 2.31 0.14 4.98 1.17 1.55 3.38 0.00 0.07 0.12 6.31

56.06 16.87 0.00 3.70 1.75 8.32 0.92 0.00 0.28 4.09 0.01 0.24 0.15 7.60

26.04 40.07 0.00 10.89 3.47 2.58 2.41 0.33 0.68 3.82 0.00 0.21 0.43 9.19

32.92 39.91 0.00 2.64 4.90 0.49 2.55 0.02 0.09 5.88 0.01 0.15 0.81 9.64

50.70 27.86 0.00 2.30 0.19 0.12 1.78 0.00 0.47 1.89 3.26 0.04 0.51 10.86

Discharge clearance 20 mm Discharge clearance 30 mm Discharge clearance 40 mm Discharge clearance 50 mm

60

wt.%

40

20

0 <10

10-20

20-40

40-80

80-160

>160

n

Fragment size (mm)

Y ¼ 1  exp½ðx=bÞ 

Fig. 2. Size distribution of shredded 75 L refrigerator fragments under 50 mm, 40 mm, 30 mm, and 20 mm discharge clearances.

Similar trends were observed for small appliances. A typical result is shown in Fig. 3, using electric heaters as an example. Most fragments were concentrated in the 20–40 mm and 40–80 mm size ranges. As the discharge clearance was reduced from 50 mm to 20 mm, fragments of 20–40 mm increased and of 40–80 mm decreased. Overall, the particle size of the small appliances was coarser than that of the refrigerators. This again can be ascribed

50 Discharge clearance 20 mm Discharge clearance 35 mm Discharge clearance 50 mm 40

wt.%

30

20

10

0 <10

10-20

20-40

40-80

to the fact that small appliances were shredded without removing the internal parts. As a result, they contained a wide variety of materials with different mechanical properties, some of which were tougher than others. The compositional analysis (Table 2) shows that the small appliances contained copper and aluminum and the refrigerators did not. These materials are not strong, but they are not easy to break due to their high toughness. This may explain the higher portions of large fragments (>80 mm) that continued to exist for the small appliances even when the discharge clearance setting was decreased from 50 mm to 20 mm. It would be more convenient to characterize the size distributions if they could be represented in a functional form. It was found that the cumulative fragment size distribution of all the products followed the Rosin-Rammler equation, as shown in Eq. (1) (Rosin and Rammler, 1933; Vesilind, 1980), which states that:

80-160

>160

Fragment size (mm) Fig. 3. Size distribution of shredded electric heater fragments under 50 mm, 35 mm, and 20 mm discharge clearances.

ð1Þ

where x is the particle size; Y is the cumulative passing of each particle size in percentage; n is the distribution modulus; and b is the size modulus. The values of b and n obtained by regression for each shredding experiment are shown in Table 3. All coefficients of determination were higher than 0.949, showing that the data fit the Rosin-Rammler distribution well. The value b represents the 63.2% passing size (i.e., 63.2% of the particles are smaller than b). It is like an average and a larger b defines a coarser particle size of the product. For solid materials to fracture, a stress high enough to exceed the fracture strength of the material is required. From the view point of fractures, materials can be divided into two main types: brittle and ductile. Brittle materials fracture without appreciable deformation, whereas ductile materials deform plastically before fracturing. The compressive force on the brittle material can cause the material to disintegrate into smaller pieces, giving a finer product (lower b values). On the other hand, the shear force is more effective for causing the ductile material to fracture, but results in tearing into pieces without generating fine particles. The product size of ductile materials would then be coarser with large b values. At a given discharge clearance, the b values are smaller for refrigerators than small appliances. In other words, the shredding of refrigerators produced fragments in smaller sizes compared to small appliances. In all cases, the b values decreased as the discharge clearance was reduced. The rate of change with the discharge clearance differed somewhat among appliance types, which may result from the differences in material composition of each unit. However, it confirms that the discharge clearance can be used as a primary operating variable to control product sizes. The parameter n represents the broadness of the size distribution. The higher the n-value, the narrower the size distribution.

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

5

J. Lee et al. / Waste Management xxx (2016) xxx–xxx Table 3 Size modulus b and distribution modulus n of the Rosin–Rammler distribution. 50 mm clearance

40 mm clearance

35 mm clearance

30 mm clearance

20 mm clearance

Size modulus b 75 L Refrigerator Vacuum cleaner VCR Electric rice cooker Fan Electric heater

45.56 60.22 47.37 49.60 65.37 65.57

41.16 – – – – –

– 52.94 48.06 44.41 51.76 58.56

36.70 – – – – –

30.81 47.05 40.40 40.95 45.83 50.46

Distribution modulus n 75 L Refrigerator Vacuum cleaner VCR Electric rice cooker Fan Electric heater

1.64 1.79 1.92 1.58 1.34 1.35

1.66 – – – – –

– 1.75 1.64 1.62 1.30 1.25

1.70 – – – – –

1.76 1.77 1.88 1.75 1.43 1.15

Electric heaters showed the widest fragment size distribution, and VCRs showed the narrowest distribution. The change in the n-value with the discharge clearance varied among the appliances. For refrigerators, rice cookers, and fans, the value of n increased. Conversely, the value decreased for VCRs and heaters while it remained the same for vacuum cleaners. If the unit is homogeneous in terms of composition, the n values should remain the same, as is often observed in grinding natural minerals, and is referred to as self-similar particle size distribution (Kapur, 1972). The appliances are composed of many materials of different mechanical properties. When they are under stress, some materials are broken further to produce more fines, while others may just deform without breakage. This would result in broadening the size distribution, and the appliances showing an increase in n values may be presumed to be relatively more heterogeneous than other appliances showing the opposite trend. For example, electric rice cookers showing a definite trend of an increasing n value with decreasing discharge clearance decreases contained approximately an equal amount of iron and nonferrous metal (aluminum and copper). In contrast, for electric heaters showing an opposite trend, iron made up the majority of metal and the nonferrous metal portion was very small (<3%). On the other hand, the difference in the mechanical properties among the appliance constituents may cause an uneven size distribution after shredding, which can be utilized for material separation. Fig. 4 compares variations in the size distributions of some materials by clearance settings, using rice cookers and fans as an example. It can be seen that the evolution of the size distribution of a material with clearance setting shows a different pattern depending on the type of appliance. In the case of iron in the rice cooker, the size distribution changed significantly when the clearance setting decreased from 50 mm to 35 mm, but moved into a finer size range when the clearance setting was decreased to 20 mm. For the fan, the change in the size distribution of iron occurred between 50 mm and 35 mm, but not between 35 mm and 20 mm clearance settings. For plastics and copper, the size distribution did not change much with clearance setting for either the rice cookers or the fans. This may be due to the fact that these materials are more ductile than iron, and thus have a tendency to deform rather than break. However, there is a significant overlap in the size distribution of iron, plastic, and copper materials, which makes it difficult to separate these materials by size. An exceptional case is electric wires, which remained in the +40 mm range even at the smallest clearance setting. At the 50 mm clearance setting, it may be possible to separate the electric wires by size, as the size of the electric wires was in the +80 mm range, while the other materials were in the 80 mm range.

Table 4 shows the material composition of the size fractions of fragments after shredding at three different discharge clearance settings. There were some differences in the material compositions among the size fractions, but no size fractions were dominantly composed of one type of material to allow material separation by size only with the exception of electric wires. Additionally, there was no marked difference in material distributions among the small appliances, as the main materials were iron and copper. Therefore, there seems to be no clear beneficial effect of the separate treatment of one type of unit in overall processing efficiency. 3.3. Liberation characteristics and fragments with two or more components The degree of liberation was calculated as the weight percentage of the fragments that were composed of one material in relation to the total fragments; the results are shown in Table 5. For the 75 L refrigerator, which had a relatively simple composition since many parts had been removed prior to shredding, all fragments were found to be liberated, i.e., composed of one material even at the largest discharge clearance setting (50 mm). Surprisingly, high degrees of liberation (more than 95%) were also obtained for small appliances. This was not expected since they were shredded without pre-removal of any parts. However, the degree of liberation did not increase much for all types of small appliances as the discharge clearance further decreased from 50 mm to 20 mm, which is most likely due to the fact that they contained many parts made up of a variety of connected materials. Further liberation may be achieved with additional intensive shredding, but it would be more practical to shred to an optimal size with an acceptable degree of materials locked to contaminants since size reduction is by far the greatest energy consuming operation in the recycling process. To obtain more than a 95% degree of liberation, a 50 mm discharge clearance was sufficient for vacuum cleaners, electric rice cookers, and electric heaters. For fans and VCRs, a smaller discharge clearance setting (20 mm and 35 mm, respectively) is required for the same degree of liberation since they showed the lowest degree of liberation among all small appliances. The number of fragment types composed of two or more materials (locked particles) is shown in Table 6 for different appliances and discharge clearances. VCRs had the highest number of locked particles, while electric heaters had the least. As the discharge clearance decreased from 50 mm to 20 mm, the number of fourmaterial locked particles decreased from eight to four, whereas that of three-material and two-material locked particles increased by 12 and 5, respectively indicating that they were liberated into those composed of fewer materials.

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

6

J. Lee et al. / Waste Management xxx (2016) xxx–xxx 80

80

Iron (Electric rice cooker)

Iron (Fan) 60

wt.%

wt.%

60

40

20

40

20

0

0 <10

10-20

20-40

40-80

80-160

>160

<10

Fragment size (mm)

10-20

>160

Plastic (Fan) 50

40

40

wt.%

50

30

30

20

20

10

10 0

0 <10

10-20

20-40

40-80

80-160

>160

<10

Fragment size (mm)

10-20

20-40

40-80

80-160

>160

Fragment size (mm)

50

50

Copper (Electric rice cooker)

Copper (Fan)

40

40

30

30

wt.%

wt.%

80-160

Fragment size (mm)

Plastic (Electric rice cooker)

20 10

20 10

0

0 <10

10-20

20-40

40-80

80-160

>160

<10

Fragment size (mm)

10-20

20-40

40-80

80-160

>160

Fragment size (mm)

60

60

Electric wire (Electric rice cooker)

Electric wire (Fan)

50

50

40

40

wt.%

wt.%

40-80

60

60

wt.%

20-40

30

30

20

20

10

10

0

0 <10

10-20

20-40

40-80

80-160

>160

Fragment size (mm)

<10

10-20

20-40

40-80

80-160

>160

Fragment size (mm) Discharge clearance 50 mm Discharge clearance 35 mm Discharge clearance 20 mm

Fig. 4. Size distribution of individual material fragments of electric rice cookers and fans from the KE-100 shredder by discharge clearance.

In the case of small appliances, the degree of liberation was lower than for refrigerators. However, the degree of liberation did not change much with a decrease in the discharge clearance, with the exception of VCRs and fans. Therefore, a 50-mm clearance setting would be sufficient for vacuum cleaners, electric rice cookers, and electric heaters. For VCRs and fans, changing the discharge clearance from 50 mm to 35 mm produced 3.9% and 2.7% increases

in the degree of liberation, respectively. However, this would result in minimal additional recovery of certain materials, which may not be sufficient to justify finer shredding considering the energy cost and reduction in throughput. Table 7 shows the degree of liberation for four major materials achieved at three different discharge clearance settings. It can be seen that iron and plastics were liberated very well (more than

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

7

J. Lee et al. / Waste Management xxx (2016) xxx–xxx Table 4 Material composition of the size fractions of fragments after shredding at three different discharge clearance settings.

Discharge settings

(a) Vacuum Cleaner 50 mm

Size (mm)

wt.% of size fraction

Material composition of the size fractions, wt.% Iron

Plastic

Alum.

Copper

PCB

E. wire

10  20 20  40 40  80 80  160 +160

4.7 22.4 55.8 10.0 2.1

8.3 18.1 23.7 6.2 12.4

65.6 68.5 68.8 59.1 0.0

0.8 0.9 0.9 0.4 0.0

15.1 3.8 1.0 2.1 0.0

1.0 0.3 0.2 0.0 0.0

0.0 0.6 0.3 10.7 73.9

35 mm

10  20 20  40 40  80 80  160 +160

6.0 32.2 45.7 7.9 1.8

10.9 33.7 17.9 10.5 14.1

63.1 54.1 73.3 48.1 0.0

1.7 2.5 0.1 0.3 0.0

6.5 1.6 1.3 1.7 0.0

0.2 0.1 0.0 0.0 0.0

7.0 2.1 4.0 23.7 71.1

20 mm

10  20 20  40 40  80 80  160 +160

7.5 42.2 35.7 5.6 1.7

12.1 22.1 12.7 19.2 15.4

61.7 67.3 74.2 23.9 0.0

2.0 0.8 1.0 0.0 0.0

11.8 2.7 1.9 4.7 0.0

0.3 0.1 0.2 0.0 0.0

3.1 1.5 4.2 44.8 66.2

18.5

60.7

0.9

2.3

0.1

5.0

Composition of the feed, wt% (b) VCR 50 mm

10  20 20  40 40  80 80  160 +160

7.3 30.4 53.0 2.6 0.5

22.3 45.4 73.8 45.6 100.0

37.6 20.5 13.6 37.1 0.0

5.3 7.2 1.9 0.0 0.0

3.1 0.4 0.1 1.1 0.0

18.6 17.6 5.1 0.0 0.0

0.0 0.2 0.1 4.3 0.0

35 mm

10  20 20  40 40  80 80  160 +160

7.8 31.9 44.7 6.5 0.2

18.0 54.2 83.6 78.8 61.7

31.8 20.9 7.9 2.7 0.0

2.8 4.8 3.5 1.2 0.0

6.2 2.2 0.4 1.4 0.0

34.8 12.7 2.8 0.0 0.0

0.3 0.4 0.7 7.5 38.3

20 mm

10  20 20  40 40  80 80  160 +160

12.5 45.1 32.2 2.3 0.4

15.6 52.9 73.0 47.6 83.6

44.7 22.8 14.1 9.2 0.0

4.4 5.8 2.8 0.0 0.0

9.8 2.7 1.9 13.6 0.0

18.3 9.8 2.7 0.0 0.0

0.4 0.6 2.2 20.8 16.4

56.1

16.9

3.7

1.8

8.3

0.9

Composition of the feed, wt% (c) Electric rice cooker 50 mm

10  20 20  40 40  80 80  160 +160

10.6 28.7 42.8 7.0 2.5

8.0 13.4 32.2 15.4 25.7

63.2 55.5 40.5 56.1 4.7

5.8 10.8 17.1 3.8 0.0

12.2 6.0 2.0 6.3 0.0

5.1 4.6 2.5 1.7 0.0

0.2 0.5 0.5 16.5 69.6

35 mm

10  20 20  40 40  80 80  160 +160

10.2 34.4 40.0 5.0 0.4

6.1 18.6 40.3 26.3 0.0

63.0 54.2 30.7 29.8 0.0

4.7 12.3 18.1 5.3 0.0

10.1 4.2 1.6 6.5 0.0

5.5 3.0 2.4 3.3 0.0

0.5 0.3 1.3 15.4 58.8

20 mm

10  20 20  40 40  80 80  160 +160

11.5 41.9 34.0 2.7 0.6

14.4 41.8 40.5 15.7 0.0

57.9 39.9 37.9 39.2 0.0

8.2 9.4 12.5 0.0 0.0

7.1 2.5 1.5 9.6 0.0

4.6 2.0 1.7 0.0 0.0

0.9 0.5 1.7 30.7 90.7

26.0

40.1

10.9

3.5

2.6

2.4

Composition of the feed, wt% (d) Fan 50 mm

10  20 20  40 40  80 80  160 +160

11.9 23.4 35.6 16.3 5.7

10.4 21.2 53.0 69.0 36.8

69.4 59.8 30.8 14.6 0.0

2.2 2.4 0.5 0.0 0.0

11.1 7.4 2.4 2.7 0.0

0.7 1.3 0.8 0.0 0.0

0.2 0.8 0.9 8.9 10.2

35 mm

10  20 20  40 40  80 80  160 +160

11.6 34.0 28.0 13.3 2.0

23.1 30.9 26.7 37.1 37.4

51.2 53.7 61.8 25.2 14.6

3.6 4.3 1.0 0.0 0.0

16.0 7.2 4.4 4.6 0.0

0.4 0.2 0.8 0.0 0.0

0.2 0.6 2.4 5.7 48.1

20 mm

10  20 20  40 40  80 80  160 +160

12.8 39.1 28.8 8.2 0.5

11.6 35.4 45.3 69.1 0.0

66.1 48.4 37.7 10.2 0.0

7.0 7.0 3.9 0.0 0.0

10.0 4.2 2.9 5.2 0.0

0.9 0.5 0.5 0.0 0.0

0.2 0.5 1.9 14.7 100.0

32.9

39.9

2.6

4.9

0.5

2.6

Composition of the feed, wt%

(continued on next page)

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

8

J. Lee et al. / Waste Management xxx (2016) xxx–xxx

Table 4 (continued) (e) Electric heater 50 mm

10  20 20  40 40  80 80  160 +160

6.1 30.0 35.6 14.3 5.8

9.2 58.0 64.1 76.4 100.0

51.0 35.3 26.1 6.1 0.0

1.9 2.1 6.2 0.0 0.0

1.6 0.2 0.1 0.1 0.0

0.0 0.4 0.0 0.0 0.0

0.3 0.4 1.2 9.6 0.0

35 mm

10  20 20  40 40  80 80  160 +160

8.4 34.5 28.6 12.9 5.0

23.7 45.5 49.0 77.3 90.9

44.8 45.7 40.6 7.8 0.0

0.0 4.6 6.1 0.0 0.0

2.4 0.1 0.2 0.1 0.0

0.1 0.1 0.0 0.0 0.0

0.3 0.6 1.6 8.9 0.0

20 mm

10  20 20  40 40  80 80  160 +160

11.3 39.1 20.9 10.7 4.3

14.4 57.6 58.5 80.4 80.4

41.0 35.5 38.1 10.6 0.0

0.4 1.5 0.1 0.0 0.0

0.3 0.0 0.0 0.0 0.0

0.6 0.3 0.0 0.0 0.0

0.2 0.9 2.0 7.4 0.0

50.7

27.9

2.3

0.2

0.1

1.8

Composition of the feed, wt%

Table 5 Shredding time, degree of liberation for three small appliances, and shredding energy and ratio of liberation difference and shredding energy difference for a small appliance. 50 mm clearance

35 mm clearance

20 mm clearance

Vacuum cleaner

Shredding time (s) Degree of liberation (%) Shredding energy (kWh/unit) DLiberation/DEnergy (% lib./kWh)

100 95.8 0.69

110 96.7 0.76

160 97.3 1.11

VCR

Shredding time (s) Degree of liberation (%) Shredding energy (kWh/unit) DLiberation/DEnergy (% lib./kWh)

150 94.0 1.04

Electric rice cooker

Shredding time (s) Degree of liberation (%) Shredding energy (kWh/unit) DLiberation/DEnergy (% lib./kWh)

90 95.1 0.63

Fan

Shredding time (s) Degree of liberation (%) Shredding energy (kWh/unit) DLiberation/DEnergy ((% lib./kWh)

60 91.8 0.42

Electric heater

Shredding time (s) Degree of liberation (%) Shredding energy (kWh/unit) DLiberation/DEnergy (% lib./kWh)

90 97.7 0.63

12.9

1.7 180 97.9 1.25

220 95.9 1.53 7.1

18.6 100 96.1 0.69

210 97.3 1.46

16.7

1.6 70 94.5 0.49

80 96.0 0.56

38.6

21.4 110 98.2 0.76

115 98.4 0.80

3.8

5.0

Table 6 Number of fragment types with two or more components obtained for different discharge clearances for each small appliance. 50 mm clearance

35 mm clearance

20 mm clearance

Number of components

2

3

4

2

3

4

2

3

4

Vacuum cleaner VCR Electric rice cooker Fan Electric heater

13 15 18 7 8

6 7 7 4 4

1 2 4 0 1

16 16 14 12 13

7 11 8 2 3

1 3 4 0 0

15 19 17 13 9

4 12 8 6 3

1 1 2 0 0

Total

61

28

8

71

31

8

73

33

4

90% liberation at all discharge clearances), whereas the degree of liberation for aluminum and copper was not as high, but there is more room for improvement by decreasing the discharge clearance. For VCRs, when the discharge clearance changed from 50 mm to 35 mm, the degree of liberation increased: 2.9% for iron, 4.6% for plastics, 3.9% for aluminum, and 55.7% for copper. This would result in the additional recovery of 74.8 g of iron, 35.8 g of plastics, 6.6 g of aluminum, and 47.1 g of copper per VCR unit if the separation was perfect. The additional revenue from selling

these materials at the prices of $150/t of iron, $0.18/kg of plastics, $1.32/kg of aluminum, and $3.52/kg of copper is $0.19. As will be discussed in the next section, the decrease in the discharge clearance from 50 mm to 35 mm requires an additional shredding time of 30 s. The additional energy consumption is 0.625 kWh, as the shredder is powered by a 75-kW motor. The additional energy cost is then estimated to be $0.075 if the electricity cost is $0.12/kWh. Therefore, the shredding of VCRs at 35-mm discharge clearance is justifiable in the economic sense. Similar calculations show that

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

9

J. Lee et al. / Waste Management xxx (2016) xxx–xxx Table 7 Degree of liberation of four major components from three small appliances. Discharge clearance

Iron

Plastic

Aluminum

Copper

Vacuum cleaner

50 mm 35 mm 20 mm

92.4 94.1 95.8

98.2 98.4 98.4

84.8 83.0 81.4

72.6 82.5 91.2

VCR

50 mm 35 mm 20 mm

96.1 99.0 97.7

92.3 96.9 95.5

89.5 93.4 86.2

20.2 75.9 83.9

Electric rice cooker

50 mm 35 mm 20 mm

92.3 96.0 97.4

97.0 97.0 97.8

91.8 94.5 95.7

92.5 88.9 90.2

Fan

50 mm 35 mm 20 mm

92.4 91.2 95.8

93.1 98.9 97.9

43.0 82.8 86.2

84.8 76.7 89.7

Electric heater

50 mm 35 mm 20 mm

99.1 98.9 98.6

96.5 97.2 98.1

93.5 93.3 77.7

62.2 77.0 39.6

economic gain can be realized for fans with a net profit of $0.03 per unit when the discharge clearance is reduced from 50 mm to 20 mm and for vacuum cleaners with a net profit of $0.02 per unit when the discharge clearance is reduced from 50 mm to 35 mm. For the other types of appliances, no economic benefit is estimated to be derived from decreasing the discharge clearance from 50 mm to the lower settings. However, these estimates are rough approximates and can vary according to the commodity prices and the cost of electricity.

3.4. Time required to complete shredding of a single unit Since the shredding process requires a large amount of energy (75 kW), time is an important economic factor and, therefore, the operating time should be considered when designing a shredding process or establishing optimal operating conditions. The KEseries high-speed vertical shredder is fed from the top and outputs the shredded fragments downwards. The time required to complete the pass was measured for each shredding condition and is shown in Table 5. For the 20 mm, 30 mm, and 40 mm discharge clearance cases, the elapsed time was similar at around two and a half minutes, whereas for the case of the 50 mm discharge clearance, the time elapse was reduced to approximately two minutes for the 75 L refrigerator. The general trend shows that shredding times increased as discharge clearances decreased. The maximum average shredding time was three minutes for three VCRs, and the minimum average shredding time was one minute for three fans. In particular, process time increased drastically when discharge clearance decreased from 35 mm to 20 mm. If at least a 95% degree of liberation is to be achieved in the least amount of shredding time, up to 30 refrigerators or 135 small appliances can be shredded per hour. It is expected that the number of appliances processed per unit time could increase under a continuous feeding system. Table 5 shows the degree of liberation and energy consumption versus shredding time. Additionally, the percent increase of liberation per additional energy input is shown. It can be seen that the energy return on liberation was very small when the discharge clearance was reduced from 35 mm to 20 mm compared with from 50 mm to 35 mm for vacuum cleaners and electric rice cookers. In contrast, it was very low for electric heaters even when the discharge clearance was reduced from 50 mm to 35 mm. For fans, the energy return on liberation was still high when the discharge clearance was reduced from 35 mm to 20 mm. This again indicates that the optimum clearance is 35 mm for vacuum cleaners and VCRs, 20 mm for fans, and 50 mm for electric heaters.

3.5. Optimal shredding conditions The optimal shredding conditions for subsequent separation processes aim to maximize the degree of liberation while minimizing the operating time. Currently, the separation process in recycling facilities is optimized to achieve fragments with an average size of 35 mm. Therefore, in the case of current recycling facility separation processes (Kim et al., 2014), 20-mm discharge clearances, which mostly produce fragments between 20 and 40 mm, seem most suitable. Even if there are a number of 10–20-mmand 40–80-mm-sized fragments, it is unlikely that there will be problems given that current separating processes classify fragments into three groups (small, medium, and large sizes) with vibrating screens. When the separation operates without fractionating the shredded products into size groups, a 30 mm discharge clearance is appropriate, as this size shreds fragments in the 20–40 mm and 40–80 mm ranges. When using a highly efficient separation process, such as those allowing color or near-infrared sorting, a viable option is to operate the shredder using a 50 mm discharge clearance, since most of the shredded fragments are between 40 and 80 mm in size. Moreover, under this condition, further economic benefits can be expected, as the average operating time is reduced by about 30 s in comparison to the 20, 30, and 40 mm discharge clearance conditions. Recently, a new method for the optical sorting of lightweight metals with a sorting efficiency of 85% was developed (Koyanaka and Kobayashi, 2010, 2011). It was claimed that the developed system could be built for less than $40,000 and handle 200 kg/h of waste with a power consumption of less than 1 kW. As mentioned previously, the problem of recycling WEEE is lower purity. If high-grade products are required, then good liberation is essential. If the liberation is obtained with a relatively

Table 8 Capacity and motor output of the present facility’s equipment and a high-speed vertical shredder (Kim et al., 2014; KUBOTA Environmental Service Co., Ltd., 2016). Equipment

Capacity (t/h)

Motor (kW)

Present facility

Stage Stage Stage Stage

7.6 7.6 3.6 1.8

90  2 EA 180 55 75

High-speed vertical shredder

KE-100

2

75–110

1 2 3 4

(Vertical shredder) (Hammer mill) (Vertical shredder) (Hammer mill)

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

10

J. Lee et al. / Waste Management xxx (2016) xxx–xxx

the recycling plants in Korea (Kim et al., 2014). Therefore, it is necessary to consider the capacity of each machine and its energy consumption. For example, when comparing four-step recycling facilities with three high-speed vertical shredders from the KE100 series, as outlined in Table 8, the capacity is approximately 2 t/h using a 75–110 kW motor. If the motor power is utilized for

coarse size, then not only is energy saved, but any subsequent separation stages become easier and cheaper to operate. It was found that the single-stage crushing performance of 75 L refrigerators and five small appliances by a high-speed vertical shredder was sufficient for 95% liberation with a product size comparable to the four-stage shredding system currently employed at

Unit :

wt.%

Fan

Shredding (20 mm)

Iron

33.5

Iron

33.9

Rubber

0.2

Plastic

39.0

Magnet

0.2

Alumi.

4.8

Vinyl

0.2

Plastic

0.6

Copper

4.2

Dust

10.6

Alumi.

0.2

E. wire

2.4

Mixture

Co p p er

0.3

PCB

0.4

PCB

0.2

Magnet

0.2

Dust

1.1

Mixture Total (% Iron

Dust

Total

3.0

4.1 100

Mag.

Magnetic

39.1

Separation

85.7)

Non-Mag.

-10 mm

9.5

Plastic 60.9

Copper

+80 mm

Screening

0.4

E. wire

1.8

Mixture

0.1

Total

10-80

0.8

3.1

48.3

mm

Con.

Eddy Current Separation Non-Con.

43.5

Iron

0.4

Plastic

0.3

Alumi. PCB

Rubber

0.2

Vinyl

0.2

Total

0.4

Float

Total

Gravity

(% Alumi.

Separation (1.0) Sink

Float

4.0 0.1 4.8 83.3)

43.1

Gravity

Sink

Separation (1.25)

Plastic

37.3

(% Plastic

100)

Copper

3.5

Alumi.

0.6

E. wire

0.6

PCB Mixture Total (% copper

0.1 1.0 5.8 60.3)

Fig. 5. Separation process for a fan.

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

11

J. Lee et al. / Waste Management xxx (2016) xxx–xxx

shredding, the energy consumption of the four-stage shredding of refrigerators is approximately 63 kWh/t, whereas that of the onestep process with the shredder investigated in this study is approximately 46 kWh/t. Therefore, the energy use could be cut by more than 25% if the current four-stage shredding process was replaced with this one-stage process. 3.6. Optimum process route Based on the experimental results of this study, we propose a route for the optimal processing of small appliances. In this process, the feed is shredded at an optimum discharge clearance setting (50 mm for vacuum cleaners, electric rice cookers, and electric heaters; 35 mm for VCRs; 20 mm for fans) for 95% liberation. The first step of the separation process was magnetic separation for iron and mixtures containing iron. After magnetic separation, fragments less than 10 mm and greater than 80 mm were screened out with vibrating sieves so that they would not be problematic for subsequent processes; fragments between 10 and 80 mm were retained for subsequent steps. Next, eddy current separation was employed for classifying nonferrous metals and their mixtures and non-conductive materials. The final step was separating the remaining fragments by gravity in two stages using water in the first stage and a salt solution (e.g. calcium nitrate solution) in the second stage (Pascoe, 2000). In order to assess the proposed processing route, separation tests were conducted using laboratory equipment. The separation results in terms of product yield and composition are shown in Fig. 5, using fans as an example. In magnetic separation using a drum magnetic separator (Hyoung Chang Magnetic Co., Ltd.), it was observed that all fragments containing iron would flow into the magnetic stream. As a consequence, the iron content of the magnetic product was about 85% due to the presence of ironcontaining metals. The +80 mm fraction removed after screening was mainly comprised of plastic and electric wires, which could easily be separated with a vibrating screen if further separation was desired. Eddy current separation relies on the magnetic force between the applied magnetic field and the induced magnetic field in the conductive material. Since iron, copper, and aluminum are conductive, they are segregated from the nonconductive material when subjected to eddy current separation. However, a high density of iron and copper limits its movement under magnetic forces and hence the eddy current separation permits a highly selective separation for aluminum. As such, the eddy current separation using a nonferrous metal separator (Hyoung Chang Magnetic Co., Ltd.)

produced an aluminum concentrate with a grade of 83%. The final separation focuses on the recovery of plastic materials since the market is well established for plastic recycling. Gravity separation was employed since plastic is sufficiently lighter than other inorganic materials. Heavy liquid separation was used to simulate gravity recovery. It can be seen that the separation at 1.25 specific gravity produced the float product of 100% plastic. The main constituent of the sink product was copper, which was not recovered from the eddy current separation. In Table 9, the yield and purity of the separated products are summarized for other small appliances when the same separation test was conducted. The yields of the products obtained after successive separations are different among the small appliances due to varied compositions. The purity of each product varied for the same reason and was also due to the different degrees of liberation obtained for each appliance. The iron content of the magnetic products ranged from 75% to 95%. The aluminum content of the products from eddy current separation showed the largest variations: around 60% for vacuum cleaners and VCRs, and more than 90% for rice cookers and heaters. The gravity separation resulted in the perfect separation of plastics with a purity of 100% for all appliances. Overall, the highest purity products were obtained for heaters, with more than 94% for all categories, since it has the highest degree of liberation and the lowest number of locked fragments. However, the separation results were generally not satisfactory. The separation efficiency was mainly limited by two factors: (1) the liberation characteristics of feed particles and (2) the performance of the separation technology applied. Since the feed was in the state of a high degree of liberation (>95%), the separation inefficiency was the result of incomplete separation. For example, it can be seen from Fig. 5 that the product streams from the magnetic and the eddy current separation contained a variety of misplaced materials. In magnetic separation, particles tend to agglomerate due to the strong magnetic force and nonmagnetic particles tag along with magnetic particles. In eddy current separation, impurities contain iron and copper, as these materials are conductive. The purity of the products could be improved by adjusting the operating condition in such a way that the selectivity of the separation process is high. However, this can be accompanied by the loss of recovery. Therefore, the separation process is always confronted with a trade-off between purity and recovery. One way to improve the purity and recovery is to employ secondary treatments in which the products are re-separated to remove the impurities and the rejects are recycled back to the feed.

Table 9 Yield and purity of each separation step for every small appliance. Separation step

Target material

Yield (%)

Purity (%)

Vacuum cleaner

Magnetic separation Eddy current separation Gravity separation

Iron Aluminum Plastic

23.4 1.6 94.3

79.5 60.0 100.0

VCR

Magnetic separation Eddy current separation Gravity separation

Iron Aluminum Plastic

70.3 21.4 59.3

87.2 55.6 100.0

Electric rice cooker

Magnetic separation Eddy current separation Gravity separation

Iron Aluminum Plastic

27.2 18.8 83.6

74.3 92.7 100.0

Fan

Magnetic separation Eddy current separation Gravity separation

Iron Aluminum Plastic

39.1 9.9 86.5

85.7 83.3 100.0

Electric heater

Magnetic separation Eddy current separation Gravity separation

Iron Aluminum Plastic

60.7 8.2 85.9

94.6 100.0 100.0

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

12

J. Lee et al. / Waste Management xxx (2016) xxx–xxx

4. Conclusions

Acknowledgements

In this study, the shredding and liberation characteristics of household appliances were examined when being passed through a high-speed vertical shredder at various discharge clearance settings. It was found that the high-speed vertical shredders managed to achieve a similar fragment size distribution to the four-stage shredding system currently employed at most recycling centers in Korea. The composition analyses obtained by visual examination of all fragments revealed that the refrigerators contained 60% iron, 17% plastic, and 9% urethane. For the small appliances, the vast majority of the mass was made up of plastic and metals (iron, aluminum, and copper), with well over 60% of each appliance being made up of iron and plastic. The fragments of the 75 L refrigerator were mainly in the range of < or = 80 mm (accounting for more than 95%). In contrast, a significant portion of fragments larger than 80 mm was detected for the small appliances due to a more complex mix of materials as the whole unit was shredded without removing any parts. The degree of liberation obtained by the single-stage shredding was very high, reaching more than 90% for mid-sized refrigerators and small appliances. To obtain more than a 95% degree of liberation, a 50-mm discharge clearance was sufficient for vacuum cleaners, electric rice cookers, and electric heaters. For fans and VCRs, a smaller discharge clearance setting (20 mm and 35 mm, respectively) is required for the same degree of liberation, since they showed the lowest degree of liberation among all small appliances. However, there were some differences in the degree of liberation among the material types. Iron and plastics were liberated very well (more than 90% liberation at all discharge clearances), whereas aluminum and copper required finer shredding to reach 90% liberation. A rough estimate of the balance between the shredding cost and the additional revenue showed that the optimal discharge clearance settings were 50 mm for electric rice cookers and electric heaters, 35 mm for vacuum cleaners and VCRs, and 20 mm for fans. In these conditions, the energy requirements per unit were estimated to be 0.76 kWh for vacuum cleaners, 1.25 kWh for VCRs, 0.64 kWh for electric rice cookers, 0.56 kWh for fans, and 0.63 kWh for electric heaters. Based on the experimental results of this study, a method for the optimal processing of small appliances was proposed, which consists of magnetic separation, screening, eddy current separation, and gravity separation. However, the separation results were generally not as good as the expected results due to misplaced materials, especially for the magnetic and eddy current separation. Since the feed was in the state of a high degree of liberation (>95% liberation), the separation inefficiency was the result of incomplete separation. Therefore, optimization of the processing should focus on the separation step. Under each optimal crushing condition, up to 30 refrigerators or 135 small appliances could be shredded per hour. This amount has the potential to increase when a continuous feeding process is applied. Therefore, considering the crushing performance, capacity, and energy consumption together, a sufficiently high degree of improvement in performance was obtained to encourage substituting the current four-stage shredding process with this one-stage process. The potential energy savings were estimated to be 25% when switching to the one-stage shredding process. This study’s results can significantly help with the application of a single-stage crushing and subsequent separation process by showing the results from a high-speed vertical shredder or other similar shredding machine to achieve desired fragment characteristics under real circumstances.

This work was supported by the R&D Center for Valuable Recycling (Global-Top Environmental Technology Development Program) funded by the Ministry of Environment (Project No. GT12-C-01-330-0). We thank Sambo, Kubota, and Hyoung Chang Magnetic Co., Ltd. for their valuable support and cooperation in this investigation.

References Castro, M.B., Remmerswaal, J.A.M., Brezet, J.C., van Schaik, A., Reuter, M.A., 2005. A simulation model of the comminution–liberation of recycling streams: relationships between product design and the liberation of materials during recycling. Int. J. Miner. Process. 75, 255–281. Cayumil, R., Khanna, R., Rajarao, R., Ikram-ul-Haq, M., Mukherjee, P.S., Sahajwalla, V., 2016. Environmental impact of processing electronic waste – key issues and challenges. In: Mihai, F.-C. (Ed.), E-waste in Transition – From Pollution to Resource. InTech, Croatia, pp. 9–36. Chao, G., Hui, W., Wei, L., Jiangang, F., Xin, Y., 2011. Liberation characteristic and physical separation of printed circuit board (PCB). Waste Manage. 31, 2161– 2166. Cui, J., Forssberg, E., 2007. Characterization of shredded television scrap and implications for materials recovery. Waste Manage. 27, 415–424. Fandrich, R.G., Bearman, R.A., Boland, J., Lim, W., 1997. Mineral liberation by particle bed breakage. Miner. Eng. 10, 175–187. Gay, S.L., 2004. A liberation model for comminution based on probability theory. Miner. Eng. 17, 525–534. Gu, Y., Wu, Y., Xu, M., Mu, X., Zuo, T., 2016. Waste electrical and electronic equipment (WEEE) recycling for a sustainable resource supply in the electronics industry in China. J. Clean. Prod. 127, 331–338. Habib, M., Sohaib, Q., Ullah, A., Habib, U., Khan, A., 2016. Mechanical strength evaluation of composites made from waste printed circuit boards (PCBs). Particul. Sci. Technol. 34, 301–307. Huang, K., Guo, J., Xu, Z.J., 2009. Recycling of waste printed circuit boards: a review of current technologies and treatment status in China. Hazard Mater. 164, 399– 408. Kapur, P.C., 1972. Self-preserving size spectra of comminuted particles. Chem. Eng. Sci. 27, 425–431. Kaya, M., 2016. Recovery of metals and nonmetals from electronic waste by physical and chemical recycling processes. Waste Manage. Kim, K., Cho, H., Jeong, J., Kim, S., 2014. Size, shape, composition and separation analysis of products from waste refrigerator recycling plants in South Korea. Mater. Trans. 55, 198–206. King, R.P., 1994. Comminution and liberation of minerals. Miner. Eng. 7, 129–140. King, R.P., Schneider, C.L., 1998. Mineral liberation and the batch comminution equation. Miner. Eng. 11, 1143–1160. Koyanaka, S., Kobayashi, K., 2010. Automatic sorting of lightweight metal scrap by sensing apparent density and three-dimensional shape. Resour. Conserv. Recycl. 54 (9), 571–578. Koyanaka, S., Kobayashi, K., 2011. Incorporation of neural network analysis into a technique for automatically sorting lightweight metal scrap generated by ELV shredder facilities. Resour. Conserv. Recycl. 55 (5), 515–523. KUBOTA Environmental Service Co., Ltd., 2016. Models and Specifications (accessed October 19, 2016). Li, J., Zeng, X., Chen, M., Ogunseitan, O.A., Stevels, A., 2015. ‘‘Control-Alt-Delete”: rebooting solutions for the e-waste problem. Environ. Sci. Technol. 49, 7095– 7108. Li, J., Dong, Q., Liu, L., Song, Q., 2016a. Measuring treatment costs of typical waste electrical and electronic equipment: a pre-research for Chinese policy making. Waste Manage. Li, J., He, X., Zeng, X., 2016b. Designing and examining e-waste recycling process: methodology and case studies. Environ. Technol., 1–9 Oguchi, M., Sakanakura, H., Terazono, A., Takigami, H., 2012. Fate of metals contained in waste electrical and electronic equipment in a municipal waste treatment process. Waste Manage. 32, 96–103. Pascoe, R.D., 2000. Sorting of Waste Plastics for Recycling. Rapra Technology Ltd., Shrewsbury, UK, p. 15. Rosin, P., Rammler, E., 1933. Laws governing the fineness of powdered coal. J. Inst. Fuel 7, 29–36. Tian, X., Yin, X., gong, Y., Wu, Y., Tan, Z., Xu, P., 2016. Characterization recovery potentiality and evaluation on recycling major metals from waste cathode-ray tube phosphor powder by using sulphuric acid leaching. J. Clean. Prod. 135, 1210–1217. Vesilind, P.A., 1980. The Rosin-Rammler particle size distribution. Resour. Recov. Conserv. 5, 275–277. Wang, J., Xu, Z., 2015. Disposing and recycling waste printed circuit boards: disconnecting, resource recovery, and pollution control. Environ. Sci. Technol. 49, 721–733. Wu, N., Yang, Y., Song, Y., Gao, M., Huang, H., 2016. Regional distribution and human health effects of Persistent Organic Pollutants (POPs) in Zhejiang Province. In:

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030

J. Lee et al. / Waste Management xxx (2016) xxx–xxx Mihai (Ed.), 2016, E-waste in Transition – From Pollution to Resource. InTech, Croatia, pp. 37–64. Yoshida, A., Terazono, A., Ballesteros Jr., F.C., Nguyen, D.-Q., Sukandar, S., Kojima, M., Sakata, S., 2016. E-waste recycling processes in Indonesia, the Philippines, and Vietnam: a case study of cathode ray tube TVs and monitors. Resour. Conserv. Recy. 106, 48–58.

13

Zeng, X., Song, Q., Li, J., Yuan, W., Duan, H., Liu, L., 2015. Solving e-waste problem using an integrated mobile recycling plant. J. Clean. Prod. 90, 55–59. Zeng, X., Li, J., 2016. Measuring the recyclability of e-waste: an innovative method and its implications. J. Clean. Prod. 131, 156–162. Zhang, S.L., Forssberg, E., 1999. Intelligent liberation and classification of electronic scrap. Powder Technol. 105, 295–301.

Please cite this article in press as: Lee, J., et al. Shredding and liberation characteristics of refrigerators and small appliances. Waste Management (2016), http://dx.doi.org/10.1016/j.wasman.2016.10.030