Patent Grant
8056728
This disclosure relates to material separations, including recycling plastics from streams of waste plastics and other materials.
The recovery of plastics from waste streams such as durable goods is a considerable challenge due the presence of several types of plastics as well as non-plastics contaminants. High purity flakes recovered from a recycling process can be extruded and pelletized to make a high quality product, but such extrusion can be difficult and the quality of the product can be poor unless certain types of contaminants are reduced to very low levels prior to extrusion.
Such problematic contaminants include, among others, non-plastics such as wood, particle board, paper, cardboard, rubber, textiles, metallic coatings, wires and circuit boards. Because such materials do not properly melt during the recycling process, they can quickly block melt filtration equipment used during the extrusion step even when these contaminant materials are present in small amounts. Non-melt particles smaller than the size of screens used for melt filtration can pass through the melt filtration equipment and end up as cosmetic and/or mechanical defects in products. In addition, cellulose-based contaminants (i.e., wood, particle board, cardboard and paper) can start degrading at extrusion temperatures typical for many thermoplastics. Therefore dark specks of degraded material can disintegrate from the original contaminant particles and pass through the melt filtration equipment. In addition, some plastics contain coatings, such as metallic paint or metal plating, that can cause similar difficulties in the extrusion step and in the products derived from extrusion of recycled plastics.
Some thermoplastics and thermosets found in durable goods are rigid foam materials known as structural foam. In some cases, this structural foam may contain flame retardants, glass fibers, residual blowing agents or other additives that could reduce the mechanical properties or marketability of the final product. In addition, the porous structure of these structural foams means that their density is lower than the density of the actual solid portion of the material.
In order to create flakes of a size suitable for extrusion from durable goods, the recycling process can include at least one and typically two or more size reduction steps. These steps can include the use of shredders, hammer mills, rotary grinders, granulators, or various other size reduction processes known in the art.
For at least some of the size reduction steps, the mixture containing plastics from durable goods also typically contains ferrous metals as well as slightly magnetic metals, such as alloys containing nickel, cobalt, molybdenum, vanadium, chromium or titanium. During the size reduction process, the high pressures and/or rapid cutter speeds tends to embed small fragments of these magnetic materials into some of the non-magnetic materials in the mixture. Small fragments worn from the size reduction equipment itself can also embed in these same non-magnetic materials.
Plastic recycling processes also typically include a wide array of conveying equipment, bins and chutes, many of which are made of steel. Over time, the steel abrades from equipment in contact with the plastics-rich mixture, and this abraded magnetic material can therefore be present in the mixture in trace amounts. Screw conveyors are one form of conveying equipment where this principle may occur, but there are other locations throughout the process as well.
Recycling processes can use high strength magnetic rods to prevent the accidental introduction of metals, such as small metal screws, into the extruder. Arrays of such magnets are, for example, sold by vendors such as Eriez Magnetics (Erie, Pa.) as grate magnets. Because there are very few such magnetic particles, the magnetic rods can easily capture the particles. The rods may be cleaned relatively infrequently and still perform their function of capturing the stray magnetic particles.
Another process sometimes included in the recycling process for durable goods is density separation. One method for separating plastics by density at densities greater than 1.0 is to use slurries of particulates, such as titanium oxide, magnetite or ferrosilicon. The particulate slurries can have a well-defined particle size distribution for improved separation.
In the following, additional methods, systems and devices are described for the selective removal of contaminant materials from plastic flakes.
Methods, systems, and devices are described for separating mixtures of solid materials based on magnetic and/or frictional differences between the materials.
In some aspects, a method includes incorporating a ferromagnetic particulate material into a mixture of solid materials and, after the incorporating step, separating the mixture of solid materials into at least two groups based on the magnetic and frictional properties of the different components of the mixture of solid materials. The ferromagnetic particulate material can differentially incorporate into or adhere to different components of the mixture of solid materials. The different components of the mixture have different frictional properties and, after the incorporating step, the different components have different magnetic properties.
In some aspects, a method includes incorporating a ferromagnetic particulate material into a mixture of solid materials, separating at least a portion of the mixture of solid materials into at least two groups based on the density of the different components of the mixture of solid materials after the incorporation of the ferromagnetic particulate material, and separating at least a portion of the mixture of solid materials into at least two groups based on the magnetic properties of the different components of the mixture of solid materials after the incorporation of the ferromagnetic particulate material.
In some aspects, a method includes supplying a stream of a mixture of solid materials to a surface of a rotating member and collecting a plurality of subgroups of the mixture of solid material from the surface of the rotating member. The rotating member rotates about an axis that is not perpendicular to the direction of gravity. The surface of the rotating member has a downward slope that is not perpendicular to the axis of rotation. Accordingly, each collected subgroup is defined by the amount of movement of the different components of the mixture of solid materials in a direction perpendicular to the direction of the downward slope due to the rotation of the rotating member.
In some aspects, a method includes supplying a stream of a mixture of solid materials to a stationary inclined surface and collecting a plurality of subgroups of the mixture of solid materials based on different trajectories of the different components of the mixture of solid materials as the materials fall off a lower edge of the stationary inclined surface. The mixture of solid materials includes different components having different frictional properties such that each component of the mixture of solid materials contacts the stationary inclined surface and travels downward along the surface due, at least in part, to gravitational force. The different trajectories are based on the frictional properties of the different components of the mixture of solid materials.
This application describes methods, systems, and devices for separating materials based on magnetic and/or frictional differences between the materials. In some embodiments, the differences in both magnetism and frictional properties can be exploited in a single step and/or with a single system or device. In some embodiments, these methods, systems, and devices can be used in combination with other processes for the separation of materials. For example, the described methods, systems, and devices can be used in a recycling plant for the recovery of plastics from goods containing plastics. These processes can be used at one or more locations in the overall process of separating plastic material from other materials in a recycling plant.
The methods, systems, and devices described herein can separate materials based on the magnetic differences between materials. In some embodiments, magnetic materials are differentially incorporated into or attached to different types of materials in a way similar to the way particles are incorporated in elevated density separation procedures (e.g., as described in U.S. Pat. No. 7,111,738, which is hereby incorporated by reference). For example, a small amount of a magnetic material (e.g., magnetite or ferrosilicon) can be incorporated into or adhered to the materials to be separated by immersing the materials in a dense aqueous suspension of a magnetic particles (e.g. magnetite). The suspension can have a well-defined particle size distribution.
Based on the different porosities of the different materials, the particles in the suspension differentially incorporate into or adhere to the different materials to be separated. For example, magnetic particles larger than around 5 microns do not tend to stick strongly to plastic flakes, especially after the flakes are rinsed. This is because typical plastic fracture surfaces tend to have voids that are smaller than 5 microns. Larger particles therefore have no place to stick to the flake when the flakes are rinsed, and/or subjected to mechanical action after the elevated density separation.
In contrast to most plastic flakes, contaminants in a plastic recycle stream such as wood, particle board, fiberboard, paper, cardboard, cloth, thermoplastic foams and thermoset foams tend to be very porous. These contaminants, which can be woven and non woven fibrous materials, can absorb particles in pores that are much larger than 5 microns. Therefore, fibrous materials and foam materials tend to capture a larger mass of particles than solid plastics. In addition, foam materials and fibrous materials tend to have rough surfaces on which large numbers of particles can stick due to the recesses in the surface. These foam materials and fibrous materials therefore readily absorb magnetic particles used in the elevated density separation as well as some small magnetic metal fragments abraded from size reduction and conveying equipment and/or other metal particles present in the recycle stream.
Rubber particles are sometimes porous as well. A fraction of the rubber found in streams of durable goods waste is foamed rubber that can readily capture small particles in its interior. Other rubber that is not foamed can also capture small particles when sufficient pressure is applied to the magnetic particles to press them into the soft rubber surface. Rubber particles therefore readily absorb magnetic particles used in the elevated density separation as well as some small magnetic metal fragments abraded from size reduction and conveying equipment.
The voids in the exposed interiors of structural foam materials can capture magnetic particles in much the same way as wood or foamed rubber. These structural foam materials therefore readily absorb magnetic particles used in the elevated density separation as well as some small magnetic metal fragments abraded from size reduction and conveying equipment.
Plastics with residual adhesive on the surface can also preferentially adhere to magnetic particles. Once the adhesive is contacted by the magnetic particles, the particles stick strongly to the plastic surface. Plastics with residual adhesive therefore readily adhere to magnetic particles used in the elevated density separation as well as some small magnetic metal fragments abraded from size reduction and conveying equipment.
Fine magnetic particles stick to all particles in the mixture to some extent, especially in the absence of rinsing or when rinsing is insufficient to remove the fine magnetic particles. Copper is non-magnetic, for example, but small amounts of magnetic particles can adhere to copper wires or thin copper pieces, making the copper slightly magnetic. Copper wires are often very thin, so they have a very large surface area to mass ratio. Because the magnetic coating is on the surface, these particles can be readily attracted to magnets. This same principle applies to thin pieces of slightly magnetic aluminum or aluminum coated films.
As described above, magnetic particles are captured by some of the materials in the plastic-rich stream after size reduction, mechanical conveying operations or elevated density separation processes using magnetic particulates. If the amount of magnetic material in some of the materials is sufficient for them to be attracted to a magnet, and if there is some differential magnetism between some of the contaminant materials (i.e., wood, rubber and foamed plastic) and the target materials (i.e., non-foamed plastic), it should be possible to reduce the amount of unwanted contaminants in the plastic flake products by including magnets somewhere after size reduction and/or the elevated density separation.
The methods, systems, and devices described herein can be arranged to allow for different magnetic separation sensitivities. The sensitivity of a particular method, system, or device can depend on a number of factors, which can include, for example, the strength and field depth of the magnet used, the velocities of the components of the mixture of materials relative to the magnet, and the arrangement and velocity of the different parts of the apparatus used. As an example of how sensitivities can be adjusted, the amount of magnetic particles adhering to surfaces of components of the mixture. Simple calculations suggest that the amount of magnetic particles (e.g., 10 micron diameter magnetite particles) is approximately 1 percent by volume if present as a complete monolayer on the surface of a smooth 4 mm diameter particle that is 2 mm thick. Magnetic particles might only cover less than about 10 percent of the surface of some materials to be separated from the mixture of materials. Accordingly, some of these materials might have between 0.01 percent and 0.1 percent by volume of magnetic particles. Magnetic particles sticking to plastic surfaces, however, might be closer to 1-5 microns in diameter, and the fractional coverage might be lower than 1 percent due to fewer 1-5 micron particles in the dense aqueous suspension. The amount of magnetic particles for 4 mm plastic particles would therefore be less than about 0.001 percent by volume. Accordingly, the sensitivity of some methods, systems, or devices described herein can allow for the separation of components of the mixture of materials having greater than between about 0.001 volume percent and about 0.1 volume percent magnetic material. The magnetic particles can include both magnetic particles incorporated into or adhered to the components of the mixture of materials and other contaminants present in the mixture of materials. For example, a less sensitive separation process could remove components of a mixture of materials having about 0.1 volume percent or more magnetic material, a more sensitive process could remove components having about 0.01 volume percent or more of magnetic material, and a still more sensitive process could remove components having about 0.001 volume percent or more of magnetic material. The methods, systems, and devices described herein, for example, can be tuned to a particular sensitivity based on the expected composition of the mixture of materials to be separated.
The amount of magnetic particles on the surface of component particles can also be controlled by the particle size distribution of the magnetic particles in the slurry, how well the magnetic particles are rinsed from the particles and how much the particles are subjected to mechanical action that might further embed the magnetic particles into the particles in the mixture or abrade magnetic particles from the particles in the mixture. The sensitivity of the magnetic separation can, for example, be selected based on the amount of residual magnetic particles on the materials. In some embodiments, the magnetic separation process can be tuned based on how well the particles are rinsed off after the density separation.
The methods, systems, and devices described herein can also separate materials based on the differences in the frictional behavior of different materials. For example, an amount of friction against a surface for a piece of material can be based on the intrinsic surface and bulk properties as well as on the contact area against the surface. The frictional forces acting on each of the different components of the mixture of materials against the surface, however, are also based on the dynamic frictional properties of the components as they tumble or bounce along the surface. For example, plastic material can experience lower amounts of frictional forces in dynamic processes than contaminant components in a mixture of materials. The amount of frictional force exerted on the different components of the mixture of materials depends on the properties of the apparatus used and the operation of that apparatus. The different methods, systems, and devices can be designed to allow for different frictional separation sensitivities depending on the desired purity of the resulting separated groups of materials and the expected composition of the mixture of materials to be separated. In some embodiments, the sensitivity can be based on an amount of movement of the different components due to frictional forces in comparison to an average amount of movement for the components of the desired final product.
Wood is a contaminant that has a high effective frictional force against surfaces because it tends to lie flat on surfaces and because the surface area tends to be high relative to the mass of the wood particles.
Rubber tends to have a high frictional force against surfaces because it tends to be softer and tackier than thermoplastics and can therefore grab surfaces more effectively than thermoplastics.
Fabric and textiles also tend to have a high frictional coefficient against surfaces. Fabrics tend to be flexible, so they can lie flat on surfaces and therefore have a greater contact area than thermoplastics, which are rigid unless stress is applied to force them to conform to surfaces. Fabrics also tend to have a very high surface to mass ratio, and can be easily dragged against the force of gravity.
Foam is soft and also has an extremely high surface to mass ratio. It can therefore be easily dragged against the force of gravity.
Similarly, fine particles have a higher surface to mass ratio, and therefore a high friction to mass ratio. Fine particles can therefore be dragged against the force of gravity.
Wires can also have a greater frictional force relative to plastics. Wires typically exist in long and flexible pieces (and can be very thin), thus much of their surface may be in contact with other surfaces (e.g., conveyance equipment). Because wires also are very dense relative to plastics and some of the other contaminants, they do not tend to bounce as much. Therefore, wires are often in greater total frictional contact with surfaces, even though the actual frictional forces of wires against a surface may not be any greater than plastics. Note also that the wire can be a slightly magnetic material in some cases.
Coated wires and cables have different frictional properties than plastic materials and thus can be separated from plastic materials based on friction. The jacketing on wires and cables is often either rubber or plasticized PVC, thus coated wires and cables often have high friction. Note also that the metal conductor can be slightly magnetic.
Circuit boards can also end up as thin fragments after size reduction, and these thin fragments have different frictional properties than plastic materials and thus can be separated from plastic materials based on friction. Circuit boards also can contain ferromagnetic material.
Accordingly, in the following, methods of using magnetic and frictional differences to enable the selective removal of contaminant materials from plastic flakes are described.
A recycling plant for the recovery of plastics from durable goods typically includes a number of process steps. For example, U.S. Patent Application Publication No. 2006/0001187, which is hereby incorporated by reference, describes various sequences of various process steps for the removal of contaminants from plastics recycle streams. The methods, systems, and devices described herein can be used in sequence with or in substitution for the various process steps described in U.S. Patent Application Publication No. 2006/0001187. These sequences of processes apply to both streams derived from durable goods and to streams of packaging materials, bottles or other mixtures rich in plastics. The process can include the use of one or more size reduction steps performed on a plastics-rich mixture from durable goods. The feed mixture can be shredded material from which some metal has been removed. The durable goods themselves can be size reduced two or more times prior to extrusion.
A mixture rich in plastic material can be processed through size reduction equipment one or more times. The size reduction steps can include rotary grinding, hammermill size reducing, shredding, granulating, or any other size reduction processes known by those skilled in the art. The plastic-rich mixture can also contain rubber, wood and other non-plastics. The flakes can range in size from around 1 mm to around 50 mm, although the process works best when the particles are between about 2 mm and about 10 mm. Size reduction, in some embodiments, can precede the elevated density separation process. In other embodiments, size reduction can also follow the density separation process to create a final flake size between about 2 mm and about 10 mm. In other embodiments, a density separation process is not used.
The mixture rich in plastic flakes can be processed through an elevated density separation in which a magnetic material (e.g., magnetite) is used to create a dense aqueous suspension, for example as described in U.S. Pat. No. 7,111,738, which is hereby incorporated by reference. The elevated density separation may be carried out in any of the types of density separation equipment. For example, hydrocyclones can efficiently separate materials of different densities based on the high centrifugal forces present in the liquid slurry swirling inside a cyclone. Within hydrocyclones, pressures on the order of 0.5 to 2 bar above atmospheric pressure, which can cause magnetic materials to enter porous materials and/or embed in softer materials even better than under ambient conditions are used.
An appropriate rinsing step can be used after the elevated density separation. The rinsing step may include, for example, using small water jets that are designed to rinse the majority of the magnetic particles off the materials in the plastic-rich flake mixture. The rinsing system should be sufficient to rinse magnetic material off plastics, but not so intense that it scours the magnetic particles off of materials in which the magnetic particles embed preferentially (e.g., rubber).
The mixture can also be dried in a controlled manner after the elevated density separation. Flake materials tend to adhere to surfaces if the surfaces are overly damp or wet, and this can result in poor separation performance for some of the processes described herein.
Mechanical conveyance processes can be used after the elevated density separation. Mechanical conveyance processes can apply pressure to flake-rich mixtures and therefore “press” magnetic material onto material surfaces so that it can preferentially embed in wood, rubber and other soft materials. The same mechanical action tends to “scour” plastic surfaces and scrape magnetic particles off of the plastic surface to therefore further improve the separation process.
Magnetic particles can also be incorporated into different materials without the use of the elevated density separation process. In some embodiments, suspended magnetic particles are incorporated or adhered to the materials for separation without the use of a density separation process. In other embodiments, magnetic particles can be incorporated using other methods. For example, magnetic material can be incorporated into the material during the size reduction processes. The incorporation of the magnetic material can then be used to separate materials based on the magnetic properties of the materials. In other embodiments, a combination of magnetic and frictional properties of the materials can be exploited to separate the materials without the prior incorporation of magnetic materials.
Magnets and/or separators can be used to exploit differences in frictional properties of materials. In some embodiments, processes that exploit differences in both magnetism and frictional properties can be combined in a single step. These various processes using magnetic and/or frictional differences can be used at one or more locations in the overall process of purifying mixtures including plastics.
Magnets can be used after at least some of the size reduction steps and after the elevated density separation. Magnets can be employed advantageously at a variety of process locations, and their use in multiple locations is desirable in order to extract the maximum amount of undesirable materials. High strength magnets can enable the removal of many of the contaminants from the plastic-rich mixtures. High strength magnets include rare earth magnets such as neodymium-iron-boron magnets and samarium-cobalt magnets, or high strength electromagnets. In some embodiments, high strength magnets are used because the magnetic attraction of most of the materials involved in the separations described here are too small when employing lower strength magnets such as ceramic magnets. High strength magnets can be employed in different geometries and configurations and with different field strengths at the various locations throughout the process. Magnets with slightly higher magnetic field strengths can be appropriate for slightly larger particles because the magnetic components of the particles may be embedded in the surface and the surface to mass ratio is lower for the large particles. The magnetic force (in comparison with forces such as gravity) is therefore lower for larger particles. Lower magnetic field strengths can be more useful when attempting to remove high surface area to mass materials such as foam and magnetic dust. Higher strength magnetic field strength magnets, however, can be used for foam and magnetic dust because the highly magnetic materials can strongly adhere to the magnet and limit the magnetic attraction of larger materials lower magnetism to mass ratios. Super high field strength magnets can also be used. In some embodiments, the process employs multiple magnets with magnetic strengths progressively increasing from the feed end towards the product end.
The amount of magnetic material incorporated into the plastics containing mixture can be present at levels between 0.01% and 50%, or more typically at levels between 0.1% to 10%. Removal of the slightly magnetic material present in the mixture at these levels with fixed rod grate magnets, such as those sold by Eriez Magnetics (Erie, Pa.), requires frequent cleaning of the rods, which leads to significant downtime and/or incomplete separation of the magnetic material. Accordingly, the instantly described methods, systems, and devices use high strength magnets that are built into rolling drums or pulleys or are configured in some other way to enable their continuous use. In addition to the need for continuous use, rotation provides a benefit of utilizing some frictional differences that tend to further enhance the separability of the more magnetic materials from the less magnetic materials. High strength drum or roller magnets of the type used in the currently described process may be purchased from a number of suppliers. One such supplier is Eriez Magnetics (Erie, Pa.).
Referring to
The drum 14 may operate in either a forward or reverse mode, as shown in
The placement of the feeder 12 relative to the drum 14 can also impact the different flows formed. For example,
In the reverse mode, as shown in
A system with high strength magnet rolls can include pulleys made of high strength magnets that rotate beneath a thin belt. Such a separator is shown schematically in
A system with multiple stages of high strength magnet drums or rollers may be employed. For example, feed mixture 22 can be fed from above and will drop past multiple separation stages. Nearly all of the slightly magnetic material can be removed with the multiple separation stages, and the stages can be configured such that very little target material is lost with the stream containing the slightly magnetic contaminants.
Other types of continuous high strength magnet processes can be in-line belt or cross belt magnets constructed with rare earth magnets or electromagnets.
Separators which exploit differences in frictional properties can also enable the efficient removal of certain types of contaminants found in plastic-rich streams recovered from durable goods. These methods, systems, and devices can include the use of rotating rollers, disks or other three dimensional objects.
These friction-based separators can be fed using standard feeding devices that provide a uniform feed. Such devices can include vibratory feeders, for example, although other devices such as screw conveyors or even controlled-flow chutes from hoppers can be used.
One separator configuration for contaminant removal based on friction is called the perpendicular roller separator (“PRS”). Referring to
The PRS may operate in either a forward or reverse mode, as shown in
One particular implementation of a system with counter-rotating cylinders is the roller sizer as manufactured by American International Manufacturing (Woodland, Calif.). The roller sizer, or roll sorter, is tilted at an angle of approximately 25 degrees with respect to horizontal and is fed from the upper end. The rollers of the roller sizer can be made of stainless steel, nickel plated steel, or a polymer (e.g., polycarbonate, high density polyethylene, or acrylonitrile butadiene styrene). The rollers can have a diameter of between 10 cm and 15 cm. The roller can rotate at a rate of 60-600 rpm (e.g., between about 100 and 300 rpm). The feed materials 22 slide down the sorter and are separated based on different frictional characteristics and particle thicknesses. Tuning parameters to control the sort include roller speed and the gap dimension at both lower and upper ends of the roll sorter.
The roll sorter can be operated in two particular modes. A mode which is referred to herein as the “thick gap” mode is shown in
Another operating strategy for the roll sorter is a so-called “zero gap” mode. In this mode, shown in
Because of the different frictional characteristics of rubber and plastic, differences in how fast each material type slides down an inclined chute can be exploited. As shown in
Another simple device that can be used for friction-based separation is a single drum 242 tilted at an angle and fed from the upper end of the drum. Such a device is shown in
The material used for the separating surface can influence the separation performance. In particular, it may be advantageous to use materials for which the difference in dynamic friction coefficients between contaminant particles and plastic flakes is largest. Separating surfaces may be made of metals, such as carbon steel or stainless steel to simplify construction. They can also be made out of other metals, such as aluminum or brass, or they can be made of other non-metallic materials such as plastic, rubber, wood or ceramic. Surfaces can also of course be coated with paints or other surface coatings that may be of particular usefulness for aiding the separation.
The texture of the separating surface can also influence the separation performance. Textured surfaces may enhance the frictional drag on materials, and in some cases may enhance the ability of particles to tumble along the friction surface. Such textures can include, but are not limited to, directional grooves cut into the surface, mesh screen material, or perforated plate with holes smaller than the particles.
The temperature of the separating surface and of plastic-rich mixture is also important to control in order to best exploit frictional differences. At lower temperatures, materials such as rubber become more rigid, and are therefore less distinguishable from plastics in terms of frictional properties. Thus, it can be advantageous to perform separations at temperatures greater than 20 degrees Celsius.
One issue of concern in frictional separators is static charge on the particles to be separated and possibly on the surfaces themselves. Charges on the various particles and surfaces can influence how particles move and can even cause them to stick to charged surfaces. Conductive surfaces are able to dissipate charges. Thus using grounded metal materials is often desirable. Antistatic devices such as ion air blowers or air-knives can also be employed on particles and/or the frictional surface to dissipate charge.
Contaminants can also be removed by exploiting differences in both magnetic and frictional properties in a single device. Such a device are referred to as magnetofrictional separators.
One potential magnetofrictional separation device is a magnetic drum similar to those described for purely magnetic separation, but now additionally utilizing frictional properties. Such a drum 14 can operate in either a forward or reverse mode, as shown in
One potential device for magnetofrictional separation is based on a rotating disk 216 with a magnet underneath 212, as shown in
It is also possible to conceive of various other devices employing frictional effects in combination with magnets. These could include almost any of the rotating frictional separators but with magnets and other slight modifications.
It is also possible to employ sequences of steps of magnetic and/or frictional separation. These steps may be used in different orders depending on the exact performance of the particular separation on the particular material.
It is useful to employ both frictional and magnetic separations because some types of contaminants separate better with one method than with the other. As an example, some rubber particles tend to be easier to separate using magnets, whereas others are easier to separate based on their frictional properties.
The described methods, systems and devices can create a higher purity plastic flake mixture by enabling the removal of wood, particle board, fiberboard, cardboard, paper, film, rubber, structural foam plastic, foam, fabric, circuit boards, copper wire and metal fragments. These methods, systems, and devices can also be useful for removing painted plastics and plastics with metal coatings. In addition to contamination of the final product, the surface coatings may interfere with separation processes which rely on surface characteristics of the material. By removing such coated material, some of the plastic-plastic separation processes work more effectively and can therefore provide products that are higher in purity in terms of the plastic composition.
These methods, systems, and devices can also allow for the preferential removal of plastic flakes containing adhesives because magnetic particles stick to the adhesive. Since adhesives are often an undesirable contaminant in the final product, it is useful to remove them prior to extrusion.
An additional useful function of the magnetic separators is that they can effectively remove residual fine magnetic particles from the system. A few such magnetic particles stick on plastic surfaces after the elevated density separation, but many of these particles fall off the surfaces after abrasion in mechanical conveyance devices such as screw conveyors and pneumatic conveyors. Such particles can result in dust that coats conveyance and separation equipment. The removal of this dust by high strength magnets functions can reduce dust levels and maintain equipment at a higher level of cleanliness.
The use of high performance magnets is also effective at improving our ability to extrude the product. In addition to non-melt or poorly melting contaminants, such as wood and rubber, the melt filtration screens used in the extrusion step can also be clogged by residual magnetic particles from the elevated density separation. Since these magnetic particles can be up to around 50 or 100 microns in diameter, the presence of only a few can quickly clog melt filtration screens that typically have holes of approximately 100 microns or less. Removal of more of these residual magnetite particles prevents screen clogging and allows the extrusion to be carried out with a finer mesh screen that in turn produces plastic pellets with smaller and fewer particulate contaminants that might affect the cosmetic properties of the product.
The magnetic separators can also significantly reduce the amount of magnetic material in the final product. Some customers, especially those who commonly purchase recycled plastics, are sensitive to the presence of metal in the plastic pellets. In some cases, these customers test for metal contamination by placing a high strength magnet in a container of the pellets. Batches of product may be rejected if pellets stick to the magnet in excess of some amount specified by the customer. Because the currently described methods, systems, and devices can utilize high strength magnets, it is less likely for magnetic material to get into the product pellets. This results in a product that is more desirable to a wider range of customers.
Two or more product streams can be recovered from each separation process or a series of separation processes as a whole. Each product from the different separation processes often contains two or more types of plastics and small amounts of non-plastics. Such a product might therefore require further purification steps, as described in U.S. Patent Application Publication No. 2006/0001187. These purification steps typically include processes relying on a narrow surface to mass distribution which are preceded by surface to mass control operations. After purification of the plastics by type (and also sometimes grade), the material can be melt compounded. The flake to be melt compounded can be blended prior to extrusion in order to improve product uniformity. The melt compounding step can employ melt filtration equipment to remove most of the non-melt contaminants.
In embodiments that utilize magnets to separate magnetic materials from non-magnetic materials, the magnetic forces can be overcome due to gravity and/or centripetal forces when in operation to separate the magnetic materials from the apparatus (e.g., drum, disk, cone, roller).
The following examples demonstrate the effectiveness of high strength magnets, friction or magnets in combination with friction for the removal of contaminants from plastic-rich mixtures.
A plastic-rich flake mixture derived from shredded televisions was processed through an elevated density separation process which used a magnetite slurry for the density separation medium. The less dense product from the elevated density separation was further processed using a triboelectrostatic separator to create a high purity plastic product.
This high purity plastic product flake mixture also contained small amounts of wood, rubber and metal-coated plastic flakes. These contaminants were each present in amounts less than 1% by mass.
A hand-held high strength rare earth magnet was placed approximately 1 cm above this plastic-rich flake sample. Material that adhered to the rare earth magnet included 16% wood and 7% rubber. The remaining material consisted of plastic with embedded magnetic particulates (10%), plastic with a surface coated with fine metal or magnetite (29%) and foamed plastic with “trapped” fine metal or magnetite (37%).
No metal fragments imbedded in the wood fibers was observed, but the wood pieces were dark in color. It is therefore likely that the cause of its magnetization is the magnetite powder coating the wood surface.
Small metal fragments were embedded in the surface of the rubber particles that were removed by the hand-held high strength rare earth magnet. It is postulated that metal fragments might embed based on the relatively soft rubber surface allowing penetration of these small and rigid particles.
A plastic-rich flake mixture derived from shredded refrigerators was processed through an elevated density separation process which used a magnetite slurry for the density separation medium.
A fixed high strength rare earth magnet was placed in a stream of the material while it was being fed to another plastic-plastic separation process. The flow rate of plastic-rich flake material was approximately 750 kg/hr. After 60 seconds (corresponding to 10-20 kg of flake), 26 grams of material were recovered on the rare earth magnet (approximately 0.2% of the feed). The composition of this recovered material included 73% plastic with metallic coating (paint or plating), 11% plastic with a sticky surface (residual foam and/or adhesive), 8% rubber or wood, and 8% magnetic dust. The magnetic dust was primarily magnetite.
A hand-held rare earth magnet was used to obtain magnetite-rich dust from a location in the process after the elevated density separation.
A hand-held high strength rare earth magnet was passed over a sample of plastic-rich flake material derived from electronics waste. The flakes were approximately 8 mm or less in size. A very small portion of the sample was picked up by the magnet. The pieces that were picked up consisted mainly of plastics with embedded metal chips, stainless steel, and fluff.
Several pieces of plastic, wood and rubber were selected from the same sample of plastic-rich flake material derived from electronics waste. The pieces of wood, rubber, and plastic selected from the sample pile were pieces that were not initially attracted to the magnet.
The plastic, wood, and rubber pieces were combined with approximately 20 g of the magnetic dust in a small plastic bag. The bag was shaken vigorously for about 1 minute. It was then placed on the floor and stepped on once. The test pieces and dust from the plastic bag were then poured onto a paper sheet. The paper sheet was then shaken to allow the magnetite-rich dust settle to the bottom of the pile and to help singulate the pieces. The test pieces were then dropped from a height of approximately 8 cm onto a hard surface to liberate any residual magnetic dust. The plastic flake bounced the hardest and became dust free, while the rubber bounced much less and the wood barely bounced at all.
A hand-held rare earth magnet was then placed close to test pieces and the pieces were separated into those that stick to the magnet (M1) from those that do not stick (N1). The separated pieces were then counted.
The non-magnetic N1 pieces were then combined with a magnetite dust that was finer than that used in the first mixing step described above. The sequence of shaking the sample in the bag, stepping on the sample bag, shaking the sample on a piece of paper and dropping the pieces was then repeated on the N1 pieces with the finer magnetite.
A hand-held rare earth magnet was then placed close to treated N1 test pieces and the pieces were separated into those that stick to the magnet (M2) from those that do not stick (N2). The separated pieces were then counted.
The data for this trial are summarized in Table 1 and
The results of this simple trial indicate that a high strength magnet can be effective at removing both rubber and wood from plastics after the mixture has been combined with magnetic dust such as magnetite. The trials also demonstrate that there are some critical considerations for optimizing this sort of process. The three keys to making this viable are magnetite particle size, the mechanics of mixing and the liberation of dust.
Finer magnetite dust added to the mixture enabled the high strength magnet to remove nearly all of the contaminants (i.e., rubber and wood) that remained after the first pass with coarser magnetite.
Apply a fairly large force to the mixture (e.g., stepping on the bag) ensures that magnetite dust is embedded into the wood and rubber. In industrial practice, such a force could be imparted using a screw conveyor at high loading, for example.
In order to have different magnetic properties for the different types of materials, it was important to liberate the magnetite dust from the particles. In this example, dropping the material approximately 8 cm was sufficient for liberating much of the dust clinging to the surface.
The starting material for this example was a plastics-rich mixture recovered from Waste Electronics and Electronics Equipment (WEEE). The mixture had been sorted by density at 1.0 and at approximately 1.1. The elevated density sort was accomplished using a hydrocyclone with magnetite as a dense medium.
The material was processed over a rare earth drum magnet. The nominal magnetism of the magnet was 7000 gauss. The diameter of the drum was 24 inches (60 cm). The non-magnetic material from the first pass was processed over the magnet a second time.
Table 2 shows the yield and compositions of the feed, non-magnetic product and magnetic byproducts. The magnetic byproducts are enriched in rubber, wood, foam/fluff and fines (i.e., smaller than 3.2 mm).
The starting material for this example was a plastics-rich mixture recovered from Waste Electronics and Electronics Equipment (WEEE). The mixture had been sorted by density at 1.0 and at approximately 1.1. The elevated density sort was accomplished using a hydrocyclone with magnetite as a dense medium.
The material described in Example 4 was separately processed over a rare earth roller magnet as shown schematically in
Table 3 shows the yield and compositions of the feed, non-magnetic product and magnetic byproducts. The magnetic byproducts are enriched in rubber, wood, foam/fluff and fines (i.e., smaller than 3.2 mm).
The starting material for this trial was a 50/50 mixture (by weight) of black plastic and rubber. The particles had been picked by hand from automotive shredder residue (ASR) and had been granulated to under approximately 25 mm.
A perpendicular roller separator was used in “reverse mode” (RPRS), as shown in
During the processing, pieces of the mixture containing fibrous material were easily taken backwards towards the rubber-rich stream. A significant fraction of rubber was in the “reverse” stream. The reverse stream also contained a few pieces of thin plastic that likely had significant frictional surface contact with the roller. Many thicker pieces of rubber hit the roller and bounced high in the air in seemingly random directions. Some of this rubber bounced out of the separator and spilled onto the floor. Table 4 shows the yields and approximate compositions of the plastic-rich product, the rubber-rich reject, and the material that spilled out of the RPRS.
The feed material for this trial was a mixture of plastic flakes derived from refrigerators with rubber particles. The size of the plastic and rubber particles was approximately 8 mm or less.
Rubber tended to concentrate in the forward stream, as expected, whereas the rubber content of the reverse stream was reduced.
The forward perpendicular roller separator (FPRS) was tested using a plastic-rich flake mixture recovered from automotive shredder residue (ASR). The mixture, which contained particles smaller than about 8 mm, contained about 0.1% or less rubber.
After the first pass through the FPRS, the plastic-enriched product was depleted in rubber while the reject contained a small but noticeable fraction of rubber. This reject was further processed through the system, and then processed that reject through the system in a third pass. At each stage, almost no rubber was in the product but a gradual enrichment of rubber in the reject. This result suggests the potential effectiveness of using multiple PRS units in series.
The feed material used in this example consisted of particles smaller than about 8 mm in size. The mixture was derived from waste electronics and electronics equipment (WEEE) from a European source.
The mixture was processed in “reverse mode” (RPRS), applying multiple stages to create plastic-rich and rubber-rich products. The process strategy is shown in
The compositions of the various streams are given in Table 5. The reject streams are greatly enriched in rubber and fines. In addition, components (e.g. foam, wood and fabric) that were not observed in the feed were found in the contaminant-enriched reject stream.
A material derived from waste electronics and electronics equipment (WEEE) was processed using a roll sorter in “thick gap” mode. The diameter of the rollers was four inches and the length was thirty inches. The gap at the upper end of this sorter was approximately four millimeters, and the gap at the lower end was approximately six millimeters. The rollers rotated at approximately 150 revolutions per minute. The roll sorter was fed at approximately 220 kg/hr.
Table 6 shows the yield and compositions of the feed and product streams, as well as an intermediate “medium” stream that would likely be re-run to maintain a reasonably high yield. These results clearly show that rubber concentrates in the thicker streams and is depleted in the thinner streams.
A material derived from waste electronics and electronics equipment (WEEE) was processed using a roll sorter in “zero gap” mode. The diameter of the rollers was four inches and the length was thirty inches. The gaps at the upper and lower ends of this sorter were both approximately one millimeter. The rollers rotated at approximately 600 revolutions per minute. The roll sorter was fed at approximately 220 kg/hr.
Table 7 shows the yield and compositions of the feed and product streams. These results clearly show that rubber concentrates in the byproduct stream that has bounced out to the sides and is depleted in the product stream exiting at the end.
A material derived from waste electronics and electronics equipment (WEEE) was processed by allowing it to slide down a steel plate that was two feet in length. The plate was at an angle of 25 degrees with respect to horizontal. The material was sorted as shown in
A material derived from waste electronics and electronics equipment (WEEE) was processed by allowing it to slide down surface that was four feet in length. The plate was at an angle of 25 degrees with respect to horizontal. The material was sorted as shown in
Various materials with different frictional properties were used as the slide surface to determine the effects on sorting. The surface materials used were metal, wood (pine), plastic (polycarbonate) and rubber.
Table 9 shows the results for the different surface materials. The friction in the rubber surface slide was so high that no separation was achieved so it is not presented in Table 9. The composition of the feed for each test varied slightly, so it is difficult to define which surface is best. The polycarbonate surface was best at concentrating rubber in the byproduct stream, whereas metal was the worst material (of the three in the table) for this purpose.
A material derived from waste electronics and electronics equipment (WEEE) was processed using a single roll frictional sorter as shown in
A mixture of flakes recovered from WEEE was processed over a forward rotating magneto-frictional sorter such as and the one shown in
Table 11 shows the overall yields, compositions and fractions of total rubber in the feed reporting to the product and reject streams. These results show a good concentration of rubber in the reject streams. Note that the amount of rubber removed decreases for each pass, yet remains significant even for the fifth pass.
A mixture of flakes recovered from automotive shredder residue (ASR) was processed over a forward rotating magneto-frictional sorter such as described in Example 15 and as shown in
Table 12 shows the overall yields, compositions and fractions of total rubber in the feed reporting to the product and reject streams. These results show a good concentration of rubber in the reject streams. Note that the amount of rubber removed decreases for each pass, especially after the second pass, yet remains significant even for the fifth pass.
A mixture of flakes recovered from WEEE was processed once over a rare earth magnet as shown in
Table 13 shows the overall yields, compositions and fractions of total rubber in the feed reporting to the product and reject streams. These results show a good concentration of rubber in the reject streams. Note that the amount of rubber removed decreases for each pass, especially after the second pass, yet remains significant even for the fifth pass.
This application claims the benefit of Provisional Application No. 61/041,089, filed Mar. 31, 2008, and Provisional Application No. 61/044,156, filed Apr. 11, 2008, which are incorporated by reference herein.
| Number | Name | Date | Kind |
|---|---|---|---|
| 861782 | Wait | Jul 1907 | A |
| 1318003 | Wheeler et al. | Oct 1919 | A |
| 1414170 | Bethke et al. | Apr 1922 | A |
| 3929627 | Frangiskos et al. | Dec 1975 | A |
| 4225426 | Price et al. | Sep 1980 | A |
| 4839033 | Sano et al. | Jun 1989 | A |
| 5161694 | Yoon et al. | Nov 1992 | A |
| 5303826 | Buzga | Apr 1994 | A |
| 6382427 | Nakhei-Nejad | May 2002 | B1 |
| 6467629 | Exner | Oct 2002 | B1 |
| 7063213 | Allen et al. | Jun 2006 | B2 |
| 7111738 | Allen | Sep 2006 | B2 |
| 20060001187 | Allen et al. | Jan 2006 | A1 |
| 20090188842 | Spangenberger et al. | Jul 2009 | A1 |
| Number | Date | Country |
|---|---|---|
| 119442 | Sep 1984 | EP |
| 2411635 | Jul 1979 | FR |
| WO2006106234 | Oct 2006 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20100078362 A1 | Apr 2010 | US |
| Number | Date | Country | |
|---|---|---|---|
| 61041089 | Mar 2008 | US | |
| 61044156 | Apr 2008 | US |
