Description:
BACKGROUND OF INVENTION
The invention relates to a magnetic separation method and a magnetic separator, and more particularly to the separating of materials of different magnet susceptibility.
One method of magnetic separation involves passing a fluid containing the higher susceptibility material and lower susceptibility material to be separated through a canister containing a matrix of ferromagnetic material such as steel balls, steel wool or tacks subject to a magnetic field. The higher susceptibility materials adhere to the magnetic collection sites and the low susceptibility materials pass through the canister. Periodically the flow of fluid to be processed may be halted and a flushing operation initiated simultaneously with a deenergization of the magnetic field to remove the higher susceptibility material from the canister. Interest in separating small particles such as colloidal or subcolloidal particles, and in separating materials having low magnetic susceptibility, including diamagnetic and paramagnetic substances is increasing. With this increase comes the demand for magnetic separators having intense magnetic fields, and maximum effective utilization thereof to make magnetic separation techniques technically and economically efficient for separation of materials of minute size and low magnetic susceptibility in large flow volume processes.
SUMMARY OF INVENTION
It is therefore an object of this invention to provide a magnetic separation method and a magnetic separator capable of quickly, efficiently, and effectively separating particles of differing magnetic properties with particle sizes as small as subcolloidal.
It is a further object of this invention to provide a magnetic separator having a high-intensity magnetic field in its separation chamber to provide therein high magnetic field gradients at a multiplicity of collection sites.
It is a further object of this invention to provide a magnetic separation method and a magnetic separator utilizing cryogenic or superconducting electromagnet coils.
It is a further object of this invention to provide a magnetic separator having a highly efficient magnetic circuit and uniform flow characteristics in the separation chamber.
The invention may be accomplished by a magnetic separator including an electromagnetic coil positioned within a recess in a ferromagnetic return frame. Centered within the coil is an enclosure containing a magnetic matrix of ferromagnetic materials such as steel balls, wool, or tacks. Fluid to be processed flows into the enclosure through an inlet means and out through an outlet means. With the coil energized, a high intensity, axial magnetic field establishes high field gradients at a multiplicity of collection sites within the matrix at which the high susceptibility materials collect, while lower susceptibility or nonmagnetic materials pass through the enclosure. Maximum utilization of the available ampere turns is aided by maximizing the amount of ferromagnetic material in the magnetic circuit contributing to the magnetic field in the matrix. For example, the return frame includes a first portion covering the area enclosed by the coil and a second portion adjacent the other side of the coil and covering the area enclosed by the coil. The return frame may include a third portion extending about the external periphery of the coil between the first and second portions of the frame.
In preferred embodiments the inlet means may include an enlarged section whose cross section area increases toward the enclosure and a ferromagnetic plug spaced from the surface of the enlarged section to provide a peripheral channel between the section and the plug. The outlet means may also be similarly constructed and the area within the peripheral channel in the enclosure is approximately the same as the area of the enclosure outside the channel whereby more uniform axial flow to and through the enclosure is effected.
DISCLOSURE OF PREFERRED EMBODIMENT
Other objects, features, and advantages will occur from the following description of a preferred embodiment and the accompanying drawings, in which:
FIG. 1 is a cross-sectional, diagrammatic view of a cylindrically symmetrical magnetic circuit configuration according to this invention with a magnet matrix having a great number of collection sites.
FIG. 2 is a diagram of one collection site in the magnetic matrix shown in FIG. 1 and the field gradient thereat.
FIG. 3 is a schematic diagram of the flow system of a magnetic separator using the magnet circuit of FIG. 1 according to this invention.
FIG. 4 is a timing chart for the flow system of FIG. 3.
FIG. 5 is a sectional diagram of a conventional coil usable in the magnetic circuit of FIG. 1.
FIG. 6 is a sectional diagram of a cryogenic or a superconducting coil with refrigeration chamber usable in the magnetic circuit of FIG. 1.
FIG. 7 is a schematic sectional diagram of a coil showing regionalization of the conductors.
FIG. 8 is a sectional diagram of the arrangement of the superconductors and normal conductors in the conductors of the outer region of the coil of FIG. 7.
FIG. 9 is a sectional diagram of the arrangement of the superconductors and normal conductors in the conductors of the inner region of the coil of FIG. 7.
The magnetic circuit configuration according to this invention may include a cylindrically symmetrical soft iron return frame 10 having a coil 12 located in a central recess 14. The chamber 16 formed centrally of coil 12 in recess 14 may be lined with a canister 18 which extends through inlet 20 and outlet 22 to external connections. Inlet 20 and outlet 22 are provided with widened inner ports 24, 26 having conical shaped surfaces 28, 30. Located within ports 24, 26 are conical iron plugs 32, 34 attached to surfaces 28, 30 by spacers 36, 38, respectively, which establish peripheral channels 40, 42 between those plugs and surfaces 28, 30. The fluid to be processed flows in inlet 20 through matrix 52 in canister 18 and out outlet 22. Plugs 32, 34 and surfaces 28, 30 are shown having conical shapes, but this is not necessary: they may have shapes resembling pyramids, hemispheres, cylinders or they may be asymmetrical and each of different shapes; further, their surfaces may be irregular. The return frame 10 and coil 12 may also deviate from symmetry, cylindrical or otherwise.
Preferably, channels 40, 42 are uniform and enclose an area of cross-sectional flow through chamber 16 between points 44, 46, and 48, 50, respectively, that is equal to approximately one half the total cross-sectional flow area of chamber 16. The volume of chamber 16 fed by channel 40 and emptied by channel 42 is evenly divided by the peripheral channels 40, 42 whereby more uniform axial flow is achieved.
The more intense magnetic field throughout chamber 16 and the more uniform flow produced by the arrangement of FIG. 1 enables the use of a more dense or finer matrix 52 in chamber 16 whereby more effective utilization is made of the magnetic field volume provided in chamber 16. The magnetic utilization factor, i.e. the ratio of ampere turns or magnetomotive force present at the chamber to the total ampere turns provided, is very high with the structure of FIG. 1. The low leakage flux and low return path magnetomotive force drop enabled by the magnetic circuit of this invention has realized a magnetic utilization factor greater than 0.9: more than 90 percent of the magnetomotive force generated by the coil appears across the magnetized volume at the canister. The uniform axial flow contributes to the effective use of a very dense matrix and a fine matrix provides a very high number of collection sites for magnetic particles. Further, the high-intensity axial magnetic field available throughout chamber 16 produces high field gradients at those sites to attract magnetic particles.
Such sites as discussed supra are depicted in FIG. 2 where collection sites 60, 62 are shown as north N and south S magnetic poles, respectively, in a field 64 which is concentrated in their vicinity to provide a high field gradient. A body placed in a magnetic field can become magnetized whereby a magnetic dipole moment is induced in it: magnetic poles are induced at the ends of the body aligned with the magnetizing field. The magnitude of the dipole moment is a function of the magnetic properties and geometry of the body, and also of the intensity of the applied magnetic field. In a uniform field the force on each pole is the same and there is no net force on the body. In a field gradient the force exerted on the pole at the higher field is greater than that exerted on the other pole and there is a net force on the body. Such bodies as would be present in the fluid flow through chamber 16 are shown adhered to sites 60, 62; particles 66, and moving to those sites, particles 68. Lower susceptibility particles 70, those of such low susceptibility that they are nearly unaffected by field 64, move freely past sites 60, 62. For the smallest and most weakly magnetic particles viscous drag will limit their motion through the fluid to a collection site. Such particles will be retained in the matrix if they impinge directly on a collection site. The magnetic separator of this invention is particularly well suited for operation in this mode because of its high magnetic and hydraulic efficiency which permit uniform flow distribution throughout a great multiplicity of small, high-gradient collection sites in large volumes of intense magnetic field.
The magnetic circuit configuration of this invention, such as embodied in the device of FIG. 1 contributes to the high-efficiency, high-intensity magnetic field in chamber 16 containing canister 18 and matrix 52. One section 10' of return frame 10 adjacent one side of coil 12 covers the coil 12 and the area enclosed by it, the other section 10" adjacent the other side of the coil 12 covers the coil 12 and the area enclosed thereby. In this manner the field applied at the matrix 52 is optimized both as to uniformity and high intensity. A third section 10'" extending about the outer periphery of coil 12 may also be used to increase the utilization of the available ampere turns and reduce leakage flux which is a consideration both as a safety factor and to further improve efficiency. In FIG. 1 section 10'" is shown coextensive with section 10' and 10" but the relationship may as well reversed, i.e. sections 10' and 10" may extend beyond coil 12 to the outer surface of section 10'".
The arrangement of FIG. 1 may be used in a flow system such as depicted in FIG. 3 including a feed tank 80, two pumps 82, 84, four directional valves 86, 88, 90, 92, with four drive units 94, 96, 98, 100, a throttling valve 102, a magnet power supply 104, and a timing control 108 for passing raw feed slurry to be separated through canister 18 to separate the higher susceptibility particles or fraction(s) from the lower susceptibility fraction(s). The two outputs are, as a convenience, referred to respectively as the magnetics and the nonmagnetics. In addition, there may be a third output, referred to as the middlings, whose magnetic susceptibility is between that of the higher and lower susceptibility particles. The lower susceptibility particles pass through the matrix, the higher susceptibility particles adhere to the collection sites and the middlings are loosely attached to the sites. A flushing of the matrix with the field on produces middlings, while flushing the canister with the field off, produces the high-susceptibility particles. Any one or more of these three separate outputs may be a "product" desired for a particular purpose. If coil 12 is a conventional water cooled coil its cooling may be performed by the system.
In the first part of the operation cycle, or feed period, the nonmagnetics may be produced at pipe 106. The coil 12 is energized by the power supply 104. Raw feed is moved by pump 82 from tank 80 up pipe 110 to chamber 112 between pistons 114, 116 of valve 86. From there the feed is pumped through pipe 118 to canister 18. The speed of pump 82 is adjusted to obtain a flow-velocity in the canister that is neither so great that the magnetic particles 66 are stripped from collection sites 60, 62, nor so low that settling of the slurry occurs. Out of canister 18 the processed slurry enters chamber 120 between pistons 122, 124 of valve 88 and out pipe 106. During this part of the cycle, water from pipe 126 is drawn by pump 84 and delivered through chamber 128 between pistons 130, 132 of valve 90 through pipe 134 to cool coil 12 and then exits from pipe 136. And water from pipe 126 also reaches chamber 138, beyond piston 132 of valve 90, through throttling valve 102. Pipe 150 connects chamber 138 to chamber 148, beyond piston 124 of valve 88. Since chamber 148 communicates only with pipe 150 during this period, there is no flow of rinse water to the canister 18.
In the second part of the cycle, or rinse period, timing control 108 operates units 94, 96 to change the state of valves 86, 88. As a result, in valve 86, chamber 112 now connects pipe 110 to pipe 140; and chamber 142, between pistons 116 and 144, connects pipe 118 to pipe 146. In valve 88 chamber 120 communicates only with pipe 106 while chamber 148 connects pipe 150 to canister 18. Feed now flows from tank 80 through pump 82, pipe 110, chamber 112, and pipe 140 back to tank 80. In this way the fluid is kept moving to prevent settling. Water flows through valve 102, chamber 138, pipe 150, and chamber 148 to canister 18 where it back flushes middlings into pipe 118. The coil 12 is still energized and excessive flow velocity in the canister 18 is prevented by the throttling valve 102. The magnetics are thus retained by the matrix. From pipe 118 the middlings flow through chamber 142 of valve 86, pipe 146, chamber 152, between pistons 154 and 156 of valve 92, and out pipe 158. Cooling water continues to flow through coil 12 as previously.
In the third part of the cycle, or flush period, timing control 108 operates units 98 and 100 and coil 12 is deenergized. As a result, in valve 92 chamber 152 now connects pipe 146 to pipe 160 and in valve 90 chamber 128 connects the output of pump 84 to line 150. The matrix in canister 18, no longer subject to the magnetic field of coil 12, is now back flushed by water under high pressure from pump 84 through chamber 128 of valve 90, through pipe 150 and chamber 148 of valve 88. The freed magnetic particles are now driven through pipe 118, chamber 142 of valve 86, pipe 146, chamber 152 of valve 92 and out pipe 160. No cooling water is supplied to deenergized coil 12 and pump 82 continues to recirculate the feed in tank 80. At the end of this part of the cycle the first part of the cycle or feed period begins again.
The relationship of the feed, rinse, and flush periods of the operation cycle is shown in conjunction with the coil energization time in FIG. 4. Typically in a 20-minute cycle, the field is on the first 17 minutes, line 162, the feed period lasts for the first 15 minutes, line 164, the rinse period occupies the next 2 minutes, line 166, and the flush period occupies the last 3 minutes, line 168.
Coil 12 may be a conventional water cooled coil 12', FIG. 5, having a toroidal body 170 with a hollow center 172 for receiving a canister. The body is formed of a plurality of double-wound layers 174, 176, 178, 180 of hollow conductor 182 having longitudinal channels 184 therein for receiving coolant. Each such double layer begins with an inlet member 186 to be connected to a source of electrical energy and to a source of coolant to pass through channel 184 of conductors 182, and is wound inwardly in the first layer 188 to the I.D. then lapped over and wound outwardly in the second layer 190 to the O.D. finally terminating in outlet member 192. Successive double layers are generally connected in series to a single source of electrical energy and hydraulically in parallel to a single source of coolant, but numerous alternate hookups are possible. Insulation 194 is provided between conductors.
Coil 12 may also be a cryogenic or a superconducting coil 12", FIG. 6. In systems, such as shown in FIG. 3, wherein cyclical operation of the magnet is contemplated, cryogenic magnetics may be preferred to superconductor types because of their greater efficiency in such operations. In the context of this patent a cryogenic coil is one which utilizes conductor materials which exhibit a large reduction in resistivity when cooled to low temperatures. A notable example of such a material would be aluminum of 99.9999 percent purity which when cooled from room temperature to 4.2° Kelvin exhibits a reduction in resistance of 10,000 to 1. Thus, the electric power required to operate a device utilizing such a "cryogenic conductor" would also be reduced by a factor of 10,000 to 1. Other high-purity materials may be used. Operating temperatures may be as high as 100° Kelvin. Coil 12" may include conductors 200 of high-purity aluminum, niobium tin alloy, or niobium titanium alloy separated by insulation 202 and maintained at 4.2° K, or other "cryogenic" temperature, in a refrigeration unit 204. Unit 204 may include an insulating support 206 for coil 12" mounted in helium vessel 208 which receives liquid helium or other "cryogenic" coolant through neck 210 integral with vessel 208. Vents 209 may be provided in support 206 to permit flow of coolant beneath the coil 12". Suspended from neck 210 is vacuum vessel 212; a radiation shield 214 may be positioned between vessels 208, 212. Electrical connection to coil 12" is made through helium boiloff cooled and properly insulated leads to reduce parasitic heat transfer. Helium boiloff is recovered from neck 210 which may be provided with cooling coils 216 containing a suitable coolant for utilizing the still quite low temperature of the helium boiloff to reduce the temperature of shield 214 with lower cooling coils 218 to further reduce heat transfer. The same double wound layer construction used in FIG. 5 may be used to construct the superconducting magnet of FIG. 6.
The cryogenic or superconducting magnet represents an extremely significant improvement in the performance and also in the processing economy of the magnetic separation device. Superconductors are materials which when cooled to near absolute zero exhibit a transition from a normal resistive state to a superconducting state characterized by zero resistivity. This means that there is zero or very little power generated by a current flowing in a superconductor. Thus, superconducting windings can maintain a magnetic field for an indefinite period of time without requiring any power. It is this fact which allows the production of very large volumes of very high magnetic fields and field gradients in working magnetic separation systems having low operating costs.
The systems described utilize superconductors whose normal to superconducting transition temperature is below 20° Kelvin and therefore for convenience are operated in a bath of liquid helium at approximately 4.2° Kelvin. Significant improvement may be obtained in some separation processes by operating at field strengths in excess of 15 Testla. The device described is capable of separating magnetic particles of subcolloidal size having magnetic susceptibility of less than 10 - 5 cg.
Coil 12'" may consist of two regions 230, 232, FIG. 7. The most economic utilization of materials can be achieved by proper selection of superconducting alloys, stabilizing material, current density and mechanical design for each region. In general, the outer region 230, conductors 234, use copper stabilized niobium titanium alloys and the inner 232 (higher field) region, conductors 236, use niobium tin alloys stabilized with either copper or high-purity aluminum. Stabilization may be achieved in the conductors 234 of outer region 230 by placing low resistivity normal conductors 201 in intimate electrical and thermal contact with the superconductors 203, FIG. 8, and may be achieved in the conductors 236 of inner region 232 by placing low resistivity normal conductors 201 in intimate electrical and thermal contact with superconductors 203, FIG. 9. Thus, if a normal region is established in the superconductor 203 the current simply transfers into the stabilizing conductors 201 and "shunts" around the normal region. The cross-sectional area and heat-transfer surface of the high-conductivity normal conductors 201 is selected so that the composite conductor temperature does not exceed the superconductor transition temperature under the above condition and the normal region reverts to the superconducting state. The superconductors 203 FIG. 8 and 9, may be twisted or otherwise positionally transposed along the direction of the current flow to reduce parasitic heating associated with time changing magnetic fields and a phenomena generally referred to as "flux jumping". The regionalization technique described in connection with FIG. 7 in relation to superconducting coils is also beneficial to use with cryogenic and conventional coils.
Other embodiments will occur to those skilled in the art and are within the following claims: