US3460672A - Method and apparatus for separating particles of different sizes - Google Patents

Description

Aug. 12, 1969 P. IMRIS 3,460,672

METHOD AND APPARATUS FOR SEPAHATING PARTICLES OF DIFFERENT SIZES Filed Sept. 6. 1967 INVEN' TOR.

PAUL [MR/S A I forn eys United States Patent 3,460,672 METHOD AND APPARATUS FOR SEPARATING PARTICLES OF DIFFERENT SIZES Paul Imris, 507 Pittsburgh, St., Springdale, Pa. 15144 Filed Sept. 6, 1967, Ser. No. 665,871 Int. Cl. B03c 7/00, 1/00 US. Cl. 209-1 6 Claims ABSTRACT OF THE DISCLOSURE This patent discloses a method and apparatus for separating particles of different sizes, and in particular, to a method and apparatus useful for size-fractionating powdered material having particles ranging in diameter from 0.01 to 1000 microns. The method works rapidly and accurately and may be practiced with inexpensive equipment. It is suitable not only for size analysis of small samples but also for the production of quantities of powdered material falling within a particular size range. According to the invention, separation is obtained by forcing an input powder through a bed of particles of ferromagnetic material, such as iron powder, while the particles of ferromagnetic mtaerial are caused to vibrate in a magnetic field.

BACKGROUND OF THE INVENTION Field of the invention This invention relates to a method for separating according to size the particles of a powder material, and to apparatus for use in the practice of such method.

Description of the prior art lar size for use, and then giving the particles of other size some further treatment (agglomeration or comminution) to bring their size into the desired range. Lately, however, there has been an increasing amount of research done with respect to powder materials that are too fine to be classified with the conventional wire-mesh series. That is to say, these are powder materials with particles having the size of about 0.1 to 100 microns diameter. A wide range of materials fall into this category, among which may be mentioned abrasives, alumina, biologicals, carbon, cements, ceramics, chemicals, clays, coal fines, drugs, dusts, dies, explosives, flours, fly ashes, glass, gypum, insecticides, latex, mica, various ores, pesticides, powdered metals, propellents, pulps, quartz, resins, sands, silicates, starches, sulphates, tungsten metal, and uranium oxides.

Various methods have been proposed to analyze or classify powdered materials of the sub-sieve range, but the known methods have drawbacks, chiefly as respects the cost of the equipment required, the skill required of the operator, the speed of the operation, and the failure of many methods to produce quickly and in substantial quantity a classified product.

There is, for example, the method of elutriating the particular material in a suitable gas, such as air, at a controlled velocity, using either one chamber through which there is passed gas at ditferent and increasing velocities, or a plurality of chambers of diflerent diameter arranged in cascade, so that the particles are passed successively from larger-diameter chambers into smallerdiameter ones. The process yields a number of fractions of different particle size, but the equipment is rather expensive and the process is rather slow.

Another method for analysis of the sample sub-sieve sized particles is a centrifugal line-start method. It is useful only with a small sample and it suffers the known disadvantages of methods involving mixing of the powder with water.

Another method is sedimentation. With really small particles, such as about 1 micron or less, the particles tend to become colloidally suspended in the liquid medium; rates of sedimentation theoretically obtainable with particles of, for example, 0.1 micron are about 1 inch in 30 days, and the natural convection currents in the liquid, acting together with the mutual repulsion of the particles as a result of the electric charge (zeta potential) that they exhibit when placed in water or other suitable medium, prevents any appreciable settling. Even with particles slightly larger, there are the problems that the adding of water or other liquid and its subsequent removal constitute added operations, as compared with methods that operate upon a dry powder, so that the expense of a sedimentation technique is necessarily greater. Moreover, some agglomeration tends to occur whenever finely divided materials are added to a liquid for a sedimentation determination, and this leads to faulty results.

Other methods that might be mentioned include: (1) electrolyte voltage-drop-change analysis, (2) flow-ultramicroscope techniques, (3) small-angle X-ray scattering analysis, (4) double-layer capacitance, and (.5) micro scopic examination. All of these have one or more of the drawbacks that (1) the powder is mixed with liquid, (2) the process does not yield a sized fraction at all, (3) the process yields sized fractions, .but they are so small that the method would never be used to separate a substantial amount of material into sized fractions, (4) expensive equipment is required, (5) the method .is slow and (6) the method requires a highly skilled operator and is thus impractical for routine use on relatively low-cost materials.

SUMMARY OF THE INVENTION In accordance with the present invention, a particular material is classified into sized fractions by causing it to be pressed against a quantity of finely divided iron powder or other ferromagnetic material which is suspended and vibrated by the action of a reversing magnetic field. At first, the layer of iron powder is substantially flat, so that the average spacing between its particles is at a minimum, and only the smallest of the particles in the input powder may pass through. The sample is pressed against the ferromagnetic material, however, and as time passes, the layer of ferromagnetic material is bowed downward, at first very slightly and then later somewhat. more markedly, with the result that the average spacing between its particles is very gradually increased. This permits the passage of, at first, very slightly larger particles, and then, particles still larger. In short, the ferromagnetic material acts as a micro-mesh sieve that does not bind and is adjustable as to its screen size. Thus, the invention affords a method that can be used for the determination of particle-size distribution in powder materials having particles ranging in size from 0.01 to 1000 microns, provided that the powder material is not itself ferromagnetic. The method operates on the powder in the dry state; it is both accurate enough for use in laboratory analysis and fast-acting enough to be useful for industrial separation of powder materials into fractions of different sub-sieve particle sizes; and it is technically simple and uses economical apparatus, calling for relatively little skill on the part of the operator.

BRIEF DESCRIPTION OF THE DRAWING A complete understanding of the invention may be obtained from the foregoing and following description thereof, taken together with the accompanying drawing, the sole figure of which is a schematic diagram of apparatus in accordance with the invention, for use in practicing the method of the invention.

DESCRIPTION OF THE PREFERRED EMBODIMENTS As shown in the figure, apparatus for practicing the present invention comprises a sample holder 2, which may be of glass or other suitable material; an electromagnetic core 4 having poles 6; a transformer 8, the secondary winding 10 of which is connected through leads 12 to a coil 14 wrapped about the core 4; a piston 16 received snugly within the sample holder 2 and connected by a shaft 18 to a weight 20 or other suitable means for exerting pressure on the piston 16; and a holder 22 for receiving the particles 24 that pass through the apparatus.

The apparatus of my invention may also have certain additional features shown in the figure. The sample holder 2 may be provided with a scale 26. The secondary winding 10 of the transformer 8 may be provided with a plurality of taps 28, in order that different potentials may be applied by the leads 12 to the coil 14, so as to vary the intensity of the magnetic field exerted between the poles 6 of the core 4. The weight 20 may be connected mechanically, as indicated at 30, to the stylus 32 of a strip-chart recorder 34 having a scale 36, in order that, as a run progresses, the stylus 32 may generate a trace 38 or 40 on the chart 42 of the strip-chart recorder 34. In certain instances, it will be desirable to use a sample holder 22 provided with a stopper 44, with a layer of iron powder 46 being provided in the sample holder 22 before the beginning of a run. It will be seen that the sample holder 2 and the container 22, as shown in the figure, have the same shape; while this is helpful, it is not essential.

The apparatus described above may be operated as follows. Into the sample holder 2 (which is provided with a stopper, not shown) there is poured a quantity of finely divided ferromagnetic material, such as iron powder having a particle size of 10-50 microns in diameter. Cobalt or nickel powder may also be used. The particle size of the ferromagnetic material should be approximately the same as that of the input powder being analyzed or classified. In the figure, the apparatus of my invention is shown in the middle of a run, the ferromagnetic material added to the sample holder 2 being shown at 50. An input powder to be analyzed or classified is indicated at 52, and as shown, this comprises relatively small particles 54 and larger particles 56.

In operation, power is supplied through lines 58 to the primary winding 60 of the transformer 8, so that power is supplied through the leads 12 to the coil 14 surrounding the core 4. Sufficient current is supplied to the coil 14 so that there is created, between the poles 6, a magnetic field having an intensity of about 4000 gausses. The magnetic field alternates with a frequency of from 10 cycles to 1000 cycles per second; for most purposes, ordinary 60-eycle alternating current is satisfactory, but it may prove advantageous to use lower frequencies when larger particles are being analyzed or classified, and higher frequencies when smaller particles are being analyzed or classified.

The sample holder 2 is then lowered between the poles 6 of the core 4, and the ferromagnetic material rises from the bottom of the sample holder 2 and arranges itself in a bed as at 50 extending between the poles 6. The piston 16 is lowered into the sample holder 2, and the weight 20 is caused to press the piston 16, driving the input powder 52 against the ferromagnetic material 50.

At first, with the ferromagnetic material 50 in the positlon Shown in full lines in the figure, the average spacing between the particles thereof is at a minimum, and only the finest particles 24 pass therethrough. Later, however, as only the increasingly larger particles 56 are left, the force exerted by the piston 16 causes the ferromagnetic material 50 to bulge outward and downward, as indicated by the dotted line 62, so that the spacing between the particles of the ferromagnetic material 50 becomes greater and particles of increasing size are passed.

At the same time, if the strip-chart recorder 34 is used, there is drawn a trace such as the trace 38.

If, during a run, the particles 24 passing through the ferromagnetic material 50 are gathered in a container 22 that has been provided with a stopper 44 and a layer 46 of ferromagnetic material, and the container 22 is subsequently used as the sample holder 2 in a second run, there will be obtained a second trace 40 differing somewhat in shape from the trace 38. If the procedure is repeated again, the trace 40 is reproduced. The trace 40 ditfers from the trace 38 because the resistance encountered by the piston 16 is different; the trace 38 is made when the piston 16 operates upon a homogeneous material, and the trace 40 is made when the piston 16 operates upon a material that is not homogeneous, but rather has the smaller particles at the bottom and increasingly larger particles as one comes nearer to the top.

The trace 4-0 corresponds to a conventional sedimentation curve, i.e., a graph in which there is plotted along the horizontal axis time in minutes and along the vertical axis the weight percentage of the sample that has been deposited as sediment. It will be understood that the trace 40 corresponds in shape to the known sedimentation curves, provided that it is replotted in such manner that both the scale 36 and the time scale 64 on the chart 42 are reversed. In other words, something very much like a conventional sedimentation curve is obtained if the trace 40 is drawn on translucent paper, and is then looked at through the rear of the paper with the paper held upside-down. In this way, there can be seen a curve having the typical shape of a sedimentation curve; it rises at first rather sharply, and then levels off asymptotically to a peak value corresponding to the total weight of the sample. It needs to be kept in mind, however, that the time scale is relative, the total time required for a run being from shorter than the time required to conduct a sedimentation analysis upon the same particle sample, if indeed a sedimentation analysis can be conducted at all (none can be, if the particles are of colloidal size). Moreover, it will not be possible to arrive at a cumulative weight distribution curve from the sedimentation curve so obtained without first obtaining some additional information; the sizes of the largest and smallest particles in the sample must be determined by other means, such as by microscopic examinaiton. Then it will be possible, by using a calculation of the kind described by S. Oden in vol. 36 of the Proceedings of the Royal Society of Edinburgh, page 219 (1916) to arrive at a proper cumulative curve.

The use of the invention for the analysis of a sample of uranium dioxide is described in the following example.

EXAMPLE I A homogeneous sample of uranium dioxide powder material weighing about 250 grams was passed twice through apparatus as described above. Iron powder was used as the ferromagnetic sieve material, and the magnetic field between the poles was about 5000 gausses. The cross-sectional area of the sample holder was about 10 square centimeters, and the weight 20 was about 3 kilograms. The equipment was powered with -volt, 60-cycle alternating current. From the second run, there was ob tained a trace 40, from which and from a knowledge of the size of the largest and smallest particles in the sample, there was calculated the particle-size distribution of the sample. The results are presented in the following table, which also presents for comparison the results obtained from a conventional sedimentation analysis of the same material.

TABLE.--DISTRIBUTION OF PARTICLES (WT. PERCENT) Sedimentation method U02, Wt. This invention Particle size, microns percent U wt. percent The foregoing results show good agreement between the data obtained with the method of the present invention and the data obtained from a time-consuming conventional sedimentation analysis. It is not unreasonable to believe that the method of this invention is more accurate than the sedimentation method in its determination of the quantities of the smallest particles. The particles of 1 to 12 microns are near-colloidal in size and will tend to settle, under ideal conditions, at a very slow rate. Slight currents in the medium, such as convection currents, can easily delay the settling of such small particles materially, with the result that the sedimentation method will give data indicating a higher proportion of very fine material than is actually present. In fact, the data from the sedimentation method would be still more inaccurate if it were not for the fact that there is some agglomeration of the finer particles, so that settling is accelerated.

It will be apparent to those skilled in the art that the use of the invention is not limited to analysis. The invention may also be used for the production of substantial quantities, such as 500 grams or more, of powder material having particles within a particular size range, starting with input powder in batches of, for example, 2 kilograms or greater.

My invention is susceptible of numerous modifications of equivalents, as will be apparent to those skilled in the art. For example, diiferent means may be provided for exerting pressure on the input powder to cause it to be forced through the ferromagnetic sieve material. For example, the sample holder may be provided with a plug, and relative movement between the sample holder and the means generating the magnetic field may be brought about in various ways. The sample holder may remain fixed and the field be moved, or vice versa. Other ferro: magnetic materials such as cobalt powder, nickel powder, iron oxides, barium ferrite and strontium ferrite may be used, but iron powder is preferred, because its magnetic permeability is higher (about 4000) and its cost lower. In general, however, it is preferred to use a ferromagnetic material of high magnetic permeability, i.e., at least 400, as otherwise the expense of generating an adequate magnetic field tends to become too great. The particle-size distribution of the ferromagnetic material should correspond to that of the non-ferromagnetic material to be analyzed or classified. The strength of the magnetic field may be varied according to requirements, but in most instances a field of about 2000 to 10,000 gausses will prove satisfactory. The quantity of ferromagnetic sieve material used may vary from about 1 to 20% of the weight of the input powder batch to be processed. Enough must be used in order that, considering the strength of the magnetic field, the input powder batch will have its weight adequately 6 supported. With a field of 5000 gausses, about 7% of the input powder batch weight proves adequate in most cases, when iron powder is used as the ferromagnetic material. With a weaker field or a ferromagnetic powder of lower permeability, more will be required.

I claim as my invention:

1. A method of sifting a quantity of sold non-ferromagnetic material containing a plurality of particles of different dimensions so as to pass particles of increasing size as the sitting progresses, said method comprising pressing said material While in a dry state into contact with a layer of particles of ferromagnetic material having an average dimension of about the same size as said plurality of particles of non-ferromagnetic material, said particles of non-ferromagnetic material and said particles of ferromagnetic material each have an average dimension of about .01 to 1000 microns, said layer being maintained in a reversing magnetic field of such strength as to support said ferromagnetic material and the weight of said solid non-ferromagnetic material, and collecting the particles passing through said layer.

2. A method as defined in claim 1, further characterized in that said ferromagnetic material has a magnetic permeability of at least 400.

3. Apparatus for sifting a quantity of solid non-ferromagnetic material containing a plurality of particles of diiferent dimensions so as to pass particles of increasing size as the sifting progresses, said apparatus comprising, in combination, means for holding said quantity of material, means for generating a reversing magnetic field within which there is maintained a layer of ferromagnetic material contained within said means for holding said quantity of non-ferromagnetic material, said reversing magnetic field being of such strength as to support said ferromagnetic material and the weight of said solid nonmagnetic material has and means for causing said quantity of non-ferromagnetic material to be pressed against said layer of ferromagnetic material.

4. Apparatus as defined in claim 3, further characterized in that said means for generating a reversing magnetic field produces a field of from 2000 to 10,000 gausses that reverses at about 10 to 1000 cycles per second.

5. Apparatus as defined in claim 4, further characterized in that said means for causing Said quantity of nonferromagnetic material to be pressed against said layer of ferromagnetic material comprises a piston fitting snugly with said means for holding said quantity of material, and a weight loading said piston.

6. Apparatus as defined in claim 5, further characterized in that said apparatus also includes a strip-chart recorder having a stylus for generating a. trace on a chart on said recorder, and means operatively connecting said stylus and said weight.

References Cited UNITED STATES PATENTS 890,527 6/1908 Nutter 209-262 3,133,876 5/1964 Klass 209-1 3,181,370 5/1965 Dietent 209-237 X FOREIGN PATENTS 816,974 7/ 1959 Great Britain.

HARRY B. THORNTON, Primary Examiner ROBERT HALPER, Assistant Examiner US. Cl. X.R. 209-223, 233

Images