US 2724056 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
1955 J. SLEPIAN 2,724,056
IONIC CENTRIFUGE Filed June 19, 1942 3 Sheets-Sheet l I I I I I 'l l l l M I WITNESSES: v INVENTOR 022 WZ'W JOJcj0/7 Sh p/Q27.
BY MM ATTOR NEY Nov. 15, 1955 J. SLEPIAN IONIC CENTRIFUGE 5 Sheets-Sheet 2 Filed June 19, 1942 INVENTOR Joseph 5/6 2/0/1.
WITNESSES: owfiw BY WM W ATTORNEY Nov. 15, 1955 sLEPlAN 2,724,056
IONIC CENTRIFUGE Filed June 19, 1942 3 Sheets-Sheet 3 WITNESSES: I INVENTOR W 253M- J066jD/7 5/ 0/02 ATTORN EY United States Patent- IONIC CENTR-IFUGE Joseph Slepian, Pittsburgh, Pa., assignor to Westinghouse Electric Corporation, East Pittsburgh, Pa., 2 corporation of Pennsylvania Application June 19, 1942, Serial No. 447,679
14 Claims. (Cl. 250-413) My invention relates to electric discharge apparatus, and has particular relation to apparatus for separating the isotopes of substances. I
In accordance with the teachings of the prior art of which I am aware, isotopes have been separated by diffusion and by mechanical centrifuge and electromagnetic methods. For any of the methods, the rate of enrichment is roughly proportionalto stances, such .as :uranium, in the isotope separation of V which .1 am particularly interested, law is of the order of 1% of W. For othersubstances, the relationship between AW and lcT is of the same order. Therefore, the enrich- "men-t factor per unit operation for diffusion separated methods is of the order of less than e and the rate of enrichment is small. Because the enrichment rate is small diffusion operations must be repeated an excessively "large number of times to effectively separate isotopes and, therefore, the diffusion apparatus is large and cumbersome.
In the mechanical centrifuge, with the highest speeds attainable, W .is approximately equal to kT, so that for uranium, [the enrichment per stage isxof the order of e and again a very large number of :stages is necessary.
Electromagnetically, isotopes are separated with a mass spectrograph or with a magnetron. In either case, W can be as high as 10,000 151. For substances, such as uranium, AW is, therefore, of the order of 100 kT, the enrichment factor is of the order of c 9, and the rate of enrichment is large.
In separating isotopes with .a mass spectrograph, ions of the isotopes are projected into electric and magnetic fields. The masses of the ions of the different isotopes are differout and, under the proper conditions, thetrajectories of the ions of differentisotopes, as they move :under the influence 10f forces exerted by the fields, are slightly different. Separate electrodes are positioned in such manner as to receive the ions moving in the different trajectories.
.It is essential ,for the operation of the 'mass spectrograph that ions of different isotopes move in distinguish ably separate trajectories. This necessity imposes a limitation on the apparatus which is undesirable fromithe standpoint of isotope separation. To maintain the trajectories separate, the ion current must be so small that the effect of space charge and collision between ions (on the ion current is negligible. Experiments with the mass spectrograph have revealed that the ion current should be of theorder of at most a few :milliamperes for proper resolution of the trajectories in .a vessel of ordinary dimensions. For such currents'the deposition of thesepa rated material is at a :minute rate. For somewhat larger 2,724,056 Patented Nov. 15 1955 current, there is a substantial loss in enrichment which far outweighs the measured current in its effect on the yield.
In amagnetron, a source of the isotope ions is disposed at the center of a conducting cylinder. A radial electric field is impressed between the source and the cylinder, and a magnetic field perpendicular to the electric field isapplied. Under the proper conditions, the electric and magnetic fields may be so adjusted that the trajectories of the ions of a lighter isotope return toward the source, while the trajectories of the ions of a heavier isotope ir'npinge on the cylinder. In this case again, the separation of the trajectories imposes the condition that the ion current be so small that space-charge effects and collision between ions are negligible. Experiments with the magnetron have revealed that here again separation is accomplished only with a small ion current. At ion currents less than approximately one-half milliampere, the separation is satisfactory.
if the current is increased to a magnitude somewhat greater than one-half milliampere, for example, to a few milliamperes, a large loss in enrichment which far outweighsthe effect on the yield of increasing the current occurs.
In the construction and the operation .of both the mass spectrograph and the magnetron, high precision is required. The difference in the masses of the different isotopes of a substance is small, and the difference in the effects of electric and magnetic fields on the ions of different isotopes is correspondingly small. Small variations in the fields under which the ions move or in the spacing of the collecting electrodes may, therefore, entirely vit'iate the results.
It is accordingly an object of my invention to provide apparatus for separating isotopes to obtain substantial quantities of the separated substances.
Another object of my invention is to provide a method of separating isotopes that shall have a high rate of en richment and that shall yield substantial quantities of the separated substances.
A more specific object of my invention is to provide electromagnetic apparatus for separating isotopes that shall operate with substantial ion currents.
Another specific object of my invention is to provide electromagnetic apparatus for separating isotopes, the construction and operation of which shall not require high precision.
An ancillary object of my invention is .to provide a source of isotope ions.
A more specific ancillary object of my invention is to provide a source of ions of uranium isotopes.
More generally stated, it is an object of my invention to provide apparatus for separating isotopes that shall have the high rate of enrichment of the electromagnetic method of separation and that shall operate with substantial .masses of isotope ions.
Broadly stated, it is an object of my invention to provide apparatus for separating ions of different masses.
In accordance with my invention, I provide an ionic centrifuge, that is, an isotope separator in which the separation is dependent on the difference in the effects of centrifugal .forces of electric and magnetic origin on the ions of different isotopes. My isotope separator is of the electromagnetic type and has a relatively high enrichment factor. It comprises .a source of isotope ions disposed along the axis of a conducting cylinder and external radial electric and axial magnetic fields cooperating to separate the ions. However, the operation of my separator does not depend on the detailed resolution of trajectories of different isotope ions and neither space charge nor collision between ions need be suppressed. On the contrary, in accordance with my 'invencurrent is, therefore, substantial.
ions are substantially circles.
rent, it reaches a certain magnitude.
" tion, the space charge modifies the electric field impressed on the ions'in such manner as to produce the desired centrifugal effects.
In the practice of my invention, the ion current is so i large that the'resultant of the space-charge field and the externally impressed electric field exerts a force on the ions wihin any region which is greater than the electric equivalent of the force exerted by the magnetic field for one of the isotopes and less than the electric equivalent for another and this over a considerable portion of space instead of over a very limited region as in the magnetron. Under the action of the resultant electric field and the magnetic field, the ions of the last-mentioned isotopes move in, closely wound spirals, that is, spirals of small pitch, about the source of ions as a center. The ions of the lighter of the two isotopes preponderantly move in spirals which first recede from the source and then return toward the source; the ions of the heavier of the isotopes 'preponderantly move in spirals which recede from the source until they impinge on the cylinder.
Since the space-charge effect contributes towards the operation of my separator, it need not be suppressed and the ionic current may be large. The quantity of material enriched in an isotope which is deposited by the Since the spirals in which the ions move are of small pitch, the paths of the The angles between the paths of colliding ions are at the points of collision,
therefore, preponderantly small, and the collisions thus 7 do not materially increase the random energy, kT, of the ions. Their only effect is to impart a common circumferential velocity to the colliding ions. The centrifugal force exerted on an ion is directly proportional to the product of its mass and the square of its circumferential velocity. equalize the circumferential velocity, the movement of Since the effect of the collisions is to the ions under the action of the centrifugal force produced by coaction of the electric and magnetic fields is to a large extent dependent only on the mass of the ions,
and ions which have collided and have not substantially the same circumferential velocity lend themselves to separation in accordance with their different masses. Rather than obstruct the separation process as they do in prior art apparatus, collisions in apparatus, according to my invention, actually expedite the separation.
My invention arises from the realization that in an electromagnetic system the enrichment does not decrease indefinitely as theionic current is increased. I have found that as the ion current is increased the enrichment decreases at a high rate until, for a certain value of car- As the current is increased further, the enrichment does not decrease substantially below the magnitude. reaches the substantially constant value, the loss in enrichrnent far outweighs the effect of the increased current, and the net effect is a large loss in the rateof separation of 'the material. After the constant enrichment value is reached, any increase in the ion current constitutes a net gain in the rate of separation, and for large enough ion currents, it may counterbalance the loss in enrichment.
To obtain a measure of the accomplishment of a separator, let M be the mass of the material which the separator is capable of treating in a single operation and E its isotopic enrichment, that is, the ratio of the masses of the isotopes in the treated substances after treatment to the ratio of the masses of the isotopes before treatment. Compare the separator with a unit separator having the same enrichment E, but capable of treating the unit mass in a single operation. M unit separators are required to produce a mass of enriched material, such as V is produced bythe separator under observation. The
total enrichment produced by M unit separators operatmg successively on the same material is E. Therefore, the accomplishment of the separator under observation Until the enrichment.
4. V can be measured by the quantity E or the log of the quantity, that is, M log E. e
In the mass spectrographs and magnetrons provided in accordance with the teachings of the prior art, M is maintained at a small value so that E is large, but M log E is relatively small. As M is increased, log E decreases at a higher rate, and the result is that M log E decreases. The electromagnetic separator, in accordance with my invention in its broader aspects, is based on the concept that if M is increased beyond a certain value, log Eremains substantially unchanged, and a further increase'in M results in an increase in M log E.
The novel features that I consider characteristic of my invention are set forth with particularity in the appended claims- The invention itself, however, both as to its organization and its method of operation, together with additional obiects and advantages thereof, Will best be understood from the following description of a specific embodiment when read in connection with the accompanying drawings, in which:
Figure l is a diagrammatic view showing an embodiment of my invention;
Fig. 2 is a section taken along the line 11-1); of Fig. '1;
Fig. 3 is a graph illustrating the operation of the embodiment of my invention shown in Fig. 1;
Fig. 4 is a diagrammatic view showing a modification of my invention; and
Fig. 5 is a graph illustrating the operation of the apparatus shown in Fig. 4.
The apparatus shown in Figs. 1 and 2 comprises a substantially vacuum-tight container 7 having a circularly tight to the top plate 10 in a position such that its axis is coextensive with the axis of the glass cylinder 8. The can has oppositely disposed openings 12 and 13, respectively,
in its side wall. A short tube 14 is welded vacuum-tight to the opening 13. The tube 14 is connectedto a vacuum pump system (not shown) through a flexible hose 15' which engages the end of the tube.
Isotope ions are derived from an are 17 produced between anode and cathode electrodes 19 and 21, respectively, disposed at the center of the container 7. The cathode 2 1 is composed of the material, the isotopes of which are to be se arated. The anode 19 may be composed of another suitable material. Since I am primarily interes'ed in separating the isotopes of uranium, I provide a cathode of uranium metal or of a suitable uranium alloy or compound. I have found that a satisfactory arc is produced between a cathode 21 of uranium metal and an anode 19 of carbon.
When the apparatus is in operation, the are 17 is maintained between the anode and the cathode by repeatedly connecting and disconnecting the two electrodes. To accomplish this object, the cathode 21 is moved in and out of engagement with the anode 19 by the cooperation of a motor operated cam 22 and an arm 23. At one end, the arm is urged into engagement with the cam by a spring 24 which extends between the arm and the rim of the top plate 10. At the other end, the arm 23 is pivoted to a rod 25 which carries the cathode 21. The rod 25 passes through an opening in a plug 26 which closes the lower end of the can 11. The opening in the plug 26 is just large enough to permit the rod 25 to slide freely, but to prevent substantial lateral movement. I
At the point where the arm 23 enters the evacuated space, a vacuum-tight joint is provided between the arm and the boundary of the space. The arm passes into the space through a rubber'plug 27 which is clamped tightly to its side. The plug is mounted centrally on a circular bracket 28 which is sealed to one end of a Sylphon 29. The other end of the Sylphon is sealed'in the remaining opening 12 of the can 11. The arm 23 and the rod 25 may be properly positionedv by rotating a screw 30' which nmgose passes through the rim of the bracket 28 and engages the can 11. n p l n The stability of the ar'c,that is, the length of time for which it will .burn after it is struck, can be greatlyfincreased by continually supplying a small stream of arc stabilizing gas, preferably oxygen or air, to the neighborhood of the arc. Since the arc stabilizing gas impairs in some degree the subsequent isotope separating effect, I adjust its magnitude to as small a value as to give a reasonable stability to the arc. The gas is supplied through a perforation 33 in the anode 19. The anode is mounted in 'ametal tube 34 which is sealed v'acum tight through the base 9. At the point where it passes through the base, the tube .34 is constricted and bent at right angles. The open end of tube 34 is engaged by a flexible hose 35 through which the arc stabilizing gas is transmitted. Near the mouth of tube 34, a wire 36 is clamped tightly between "the walls of the hose, providing a. constricted opening through which the gas may leak slowly to the space between the electrodes 19 and 21.
The ions in the are 17 are subjected to an accelerating radial electric field which is impressed between a pair of coextensive conducting cylinders 39 and the cathode 21. Under the action of the field, the ions are projected radially through a circumferential .slot 41 between the cylinders 39. On passing through the slot 41, the ions are subjected to a second accelerating radial electric field which is impressed between a second set of coextensive conducting cylinders 43 and the first set. Under the action of the accelerating fields, the ions move through a second circumferential slot 45 between the cylinders 43 into a ring-shaped region 46 bounded by the latter cylinders, an external conducting cylinder 47 having a height equal to the distance between the outer ends of the latter coextensive cylinders 43, and a pair of ring-shaped conducting side plates 49 extending between the edges of the coextensive cylinders and the edges of the outer cylinder.
The ion streams produced bythe electric field are also subjected to a magnetic field, the lines of force of which are parallel tothe axis of the cylinders 7, '39, 43 and 47. The latter field is produced by an elecromagnet :51 having horizontal cylindrical poles 53 extending above and below the base 9 and top 10 of .theevacua'ted container 7. The poles 53 are provided with suitable exciting windings 55 which are energized from the source 56 or any other suitable direct-current source, and a magnetizable yoke 57, extends between them. The top 10 and base 9 of the container 7 and the end plates 49 should be composed of a material which permits the passage of the magnetic lines of force from the poles 53 through the ring-shaped space 46 bounded by the coextensive cylinders 43, the external cylinder 47, and the side plates 49.
p The ionic current produced within the ring space 46 is of substantial magnitude, and the space-charge effect produced by it so modifies the external electric field impressed on the ions that the resultant electric 'fieldgi's at each point within the ring space greater than the electric equivalent of the magnetic field for a heavier isotope and less than the electric equivalent for a lighter isotope. The ions of the latter isotopes move in closely wound spirals, 'the ions of the lighter isotope first receding from thesource 17 and then returning to it, the side plates 49, or the coextensive cylinders 39 or 43, and the ions of the heavier isotope receding from the source until they impinge on the bounding cylinder 47. The apparatus may be operated for a substantial time interval. Then the conductors bounding the ring space 46 may be re moved from the container 7, and the substance enriched in one or the other of the isotopes may be removed from the Walls of the outer cylinder 47, the coextensive cylinders 39m 43, or the side plates 49 to which it has adhered. r v The-operation of the separator shown in Figs. 1 and 2 is illustrated graphically in Fig. 3. -Inthis view, the cletci'omagnetic fields impressed 'on an ion are plotted as a function of the radial position of the ion. The ion has a certain initial kinetic energy which is of a random character. The total additional kinetic energy imparted to .a singly ionized ion by the electric field at any point is Vs, where V is the potential of the electric field at the point, and e is the charge on an electron (that is, the charge on the ion). The magnetic field causes some of the kinetic energy of the ion which is partly initial, and partly imparted by the electric field, to be converted to energyof rotation about the central axis, thus counteracting the effect of the electric field in producing energy of pure radial motion and reducing the energy of purely radial motion. The magnitude of the counteracting effect of the magnetic field is ii 22 Hr where m is the mass of the ion, H the magnetic field, and r the radial distance of the point under consideration. The effect of the magnetic field may be regarded as produced by an equivalent electric field, the potential of which is -HW volts 8 m which acts to drive the ion inwardly toward the center.
' In Fig. 3, the field potential V, whether it be of elec trict or magnetic origin, is plotted vertically, and the radial distance r is plotted horizontally. The light curve 59 represents the externally impressed electric field, such as would exist with the electrode arrangement of Fig. l, and in the absence of space charge, as a function of the radial distance. The field is equal numerically to the external potential which is impressed and rises at a moderate rate from zero at the source 17 until it reaches a substantial magnitude at the radial distance ri correspending to the position of the external coextensive cylinders 43. Between the radial distance r1 and the position To corresponding to the external cylinder, the external field does not rise substantially. In the absence of space charge, the light curve 59 would represent the actual potential impressed on the ions. The lower medium weight curve 61 represents the electric potential equivalent of the action of the magnetic field on the ions of the isotope which, to simplify the explanation, I shall assume to be the heaviest of the treated elements. The curve 61 is a parabola. The upper medium weight curve 63 represents the electric potential equivalent of the magnetic field for a lighter isotope.
I shall first consider the situation which arises in prior art electromagnetic separators in which space-charge and collision effects are substantially absent. For such a separator the curve 59 represents the actual potential of the electric field in which the ions move. If the ions start with negligible initial velocity, the curve 59 also represents the total kinetic energy which the ions have at any radius r. For values of r for which the electric potential exceeds the electric potential equivalent of .the magnetic field, the ions have a finite energy of radial motion, and can move away from the center. At the radial distance corresponding to the point of intersection of the curve of electric potential and the potential equivalent of the magnetic field, the radial velocity of the ions is Zero, if they have made no collisions previously as I have assumed. The ions, therefore, predominantly do not move beyond the radial distance corresponding to the point of intersection of the curves, but return toward the are 17. Thus the light ions are predominantly turned back at the radial distance corresponding to the intersection 64 of the curve 59 with the curve 63, while the heavy ions go on, and are not turned back until they reach the radial distance corresponding to the intersection 66 of the curve 59 with the curve 61. If the outer collecting electrode is placed at a distance corresponding to the radial distancesbeuranium, it is of the order of one percent of the radius of the separator. Thus high precision is required in the construction of an electromagnetic separator in which the ions move in a space-charge-free field, and in such a separator the electric and magnetic fields must be adjusted precisely to their proper values and maintained uniformly at the proper values. Moreover, the current of positive ions used in this separator must be kept small enough so that with the degree of neutralization by electrons present, space-charge effects will not appreciably alter the electric field, since any such alteration of the electric field will shift the positions of the intersections 64 and 66 of curve 59 with curves 63 and 61. The current of positive ions must also be kept small, so as to avoid collision effects between ions. Such collisions will profoundly change the trajectories of the ions, and completely change the radii at which the trajectories turn from the outward motion back toward the inward motion. This condition arises because the colliding ions will in the space-charge-free region have large radial velocities, one outwardly, and the other inwardly.
I have found experimentally that space-charge and collision effects become disturbing for small current of positive ions. For example, in a magnetron with a diameter of 16 inches for the outer electrode and with a' uranium ion current of only a milliampere or two, the space charge caused many of the heavy isotope ions to turn back at a radial distance substantially smaller than the theoretical radial distance corresponding to the intersection points of the field curves.
In the ionic centrifuge which embodies the broad aspects of my invention, space charge is permitted to develop to such a degree that the resultant electric po tential takes the form indicated by curve 65, that is, lies very close to the curves 61 and 63. An ion starting from rest at the are 17 of an ionic centrifuge will have a total kinetic energy given by curve 65. at any radial distance. The circumferential velocity component of the energy will be given by curve 63 or 61, depending on whether it is alight or heavy isotope ion. The radial energy of the ion which determines the radial velocity component is given by the difference between the total energy (curve 65) and the circumferential energy (curves 61 and 63). Since the total energy curve 65 lies near the circumferential curves 61 and 63, the difference is small and the radial velocity is small compared .to the circumferential velocity. The ions, therefore, move over large circumferential arcs while they are moving a short distance radially, and at any radius the paths of the ions of both isotopes will be closely wound spirals. The point of intersection 68 of. curve 65 with curve 63 corresponds to the radial position in the centrifuge at which V the ion of the light isotope will stop spiraling outwardly, 'and will begin to spiral inwardly. The point of interof any; collision will spiral out until they reach the radial position corresponding to the former inter ection point.
In general, an ion will leave the center with a finite initial kinetic energy. In discussing prior art' electromagnetic separators, the initial energy can be neglected, as it is small compared to. the energy which is imparted to the ions by the external electrical field. In the ionic centrifuge, however, the initial kinetic energy may constitutea substantial portion of thetotal energy and cannot be neglected.
If the initial energy is taken into consideration, Fig. 3 does not precisely represent the actual conditions in'an ionic centrifuge. The total kinetic energy of the ions is represented not by the curve 65, but by a plurality of curves displaced upwardly from curve by distances corresponding to the initial kinetic energy of the ions. The intersection points of the displaced curves 65 with curve 63 are shifted to the right, and the turning points of the light isotope ions are at different radial distances. The turning point of the spiral path of a light isotope ion with one initial energy occurs at a radius which is larger than that for a light isotope ion with a smaller initial energy. Since the distribution of initial energies of the ions is Maxwellian, the distribution of the turning points of the light ions is Maxwellian; that is, the radii at which the light isotope ions turn back are distributed about amean radius exponentially, roughly in accordance with the relationship r, where .e is the natural logarithm base and F is the distance of any radius from the mean radius. Similarly, radii at which the heavy isotope ions turn' back are distributed about a mean radius in accordance with the Maxwellian relationship. However, the latter mean radius is larger than the former. In accordance with my invention, the outer collector should be placed between the mean turning radii of the light and heavy ions, so as to achieve a maximum separating effect of the isotopes being collected.
In a separator in accordance with my invention as contrasted to prior art electromagnetic separators, there is frequent collision between ions, but the collisions do not appreciably impair the effectiveness of the apparatus in separating the ions. This condition arises because at any radius, the ions are moving circumferentially with nearly the same velocity, while the radial velocities, although differing in magnitude and direction, are small. 'fi'hen a collision occurs between two ions, their common circumferential velocityis preserved, and their radial velocities are exchanged so that the paths of the ions taking part in the collision are shifted to other paths. Each of the new paths corresponds to the path of an ion similar to the colliding ion with a different initial energy. As a result of collisions, some ions will be shifted to paths corresponding to higherinitial energies, and some to paths corresponding to lower initial. energies. On thewhole, the totality of paths of the ionswill not be changed substantially by collisions, and, therefore, collisions will not materially impair the separation..
In terms of its charge e, its mass m, the magnetic field H to which it is subjected and its radial position r, the energy of circumferential motion of the ions is given approximately by the expression In terms of its mass and its angular velocity w about the axis of the container 7,.the same energyis given bythe expression l/2mr w Equating the two relationships, it follows that; V
w m n and is for each isotope constant at all radii. The rotation of the ions isv thus the same as the rotation of a rigid body, and no disturbances arise from the circumferential motion by rcason of the slip of the ions of one isotope at any radius relative to the ions at an adjacent radius. A small slip disturbance docs arise because the ions of different isotopes move at different angular velocities relative to each other. v I
The motion of the ions in the space 46 is similar to that of the'molecules of a gas in a mechanical centrifuge.
The mass as a whole rotates like a rigid body, but the centrifugal 'forces superimposed on the rigid rotation cause a preferential diffusion of the heavier molecules to the outer radii of the centrifuge, and a diffusion of the lighter molecules to the inner radii. 'It is for this reason that Icall my device anionic centrifuge-meaning thereby that the ions undergo motions similar to the motions of the molecules of a gas in a mechanical centrifuge. While my invention is of importance because of its application to the separation of isotopes, it is not limited in this respect. Apparatus embodying the "broad concepts of my invention may 'be applied to the separation of molecules of any type as, for example, molecules of different materials.
The operation of my ionic centrifuge depends on an ionic space-charge effect of proper magnitude and distribution. To explain how the space-charge effect is developed, 1 shall first consider the oversimplified case in which the positive ions are the only charged particles .in the centrifuge. The positive charge carried by the positive ions in transit from the source to the collecting electrodes forms a positive space charge whosefeifect upon the electric field and the potential acting on the ions can be calculated by applying the usual electrostatic theory. Qualitatively, the effect of the space-charge -is to reduce the potential in the space to a smaller value than the externally impressed potential which is given by curve 59 in Fig. 3. The space-charge effect thus brings the space potential down from values corresponding to the curve 59 towards values corresponding to curves 63 and 61 :(that is, towards curve 65) as the current ofpositive ions available from the source is increased. At any point at which the potential represented by curves 63 and 61 is equaled by the net space potential, ions are turned back, those with smaller initial kinetic energy being turned back if the net field is represented by a curve which is tangent to curve 63, and those with larger initial kinetic energies being turned back for field represented by a curve (such as 65) which penetrates the region 'of curves 63 and 61 more deeply.
The turning back of the ions lessens the space charge at larger radial distances than those at which the ions are turned back, and increases the space charge at smaller radial distances. The decrease of the space charge beyond the turning back radial distances has the effect of preventing the space potential from sinking substantially below values corresponding to the region between the curves 63 and 61 at larger radial distances, and of causing the space potential to approach values corresponding to this region at smaller radial distances. Thus, as the n'umber of available ions is increased at the source, the net potential approaches values corresponding to the region between curves 63 and 61 throughout most of the space 46. (The net potential represented by the curve which results when curve 59 is modified by space-charge effect-- does in fact fall slightly below that represented by'curve :61 :for extreme radial distances. This condition arises because there are some heavy isotope ions in the space which have suflicient initial kinetic energy to be carried beyond radial distances corresponding to the intersection of curve 65 and curve 61 against the net field.)
The foregoing analysis is based on the assumption that only positive ions are present in the space 46. Under such circumstances, the space char ge effect would be large even for small ion currents, because the positive ions move slowly. The space potential would 'then have values represented by points deep in the region of curves 63 and 61, and substantially all the ions set free at the source would be turned back there, and only the few ions which have large initial velocities would penetrate to larger radial distances, Thus, ,forthe .structure pictured in Fig. 1, with a sixteen-inch diameter outer collector 47 and with a magnetic field of 8000 gauss and an electric potential of 1200 volts, only a few m'icroampere's of positive ions would pass to the outer electrode in the absence "to an extent.
In fact, there is space-charge neutralization within the region 46. Neutralizing effects arise from electrons which are abstracted by positive ion bombardment from the metallic surfaces 49 which bound the ionic centrifuge axially with respect to the magnetic field. The number of electrons per ion emitted from the surfaces depends upon the nature of the surface, and the energy with which the ions strike the surface and is ordinarily small. 'The electron emission can be increased by coating the surface with alkali or alkaline earth metals or their oxides, or by insulating surfaces and impressing negative potentials on them. However, only a small electron emission is needed to neutralize the space charge of a relatively large ion current, because the electrons remain in the radial plane where they are emitted. After leaving the surfaces 49, the electrons are able to move freely axially in the direction of the magnetic field, but are prevented fro rr' moving rapidly in a radial direction. The magnetic field converts any kinetic energy of radial motion which "the electrons may acquire as they leave the: surfaces 49 into kinetic energy of circumferential motion.
The apparatus shown in Fig. 4 differs from that shown in Fig. 1 in several minor respects (certain parts, such as the magnet 51, are in Fig. 4, omitted for the purpose of clarity).
The cam electrode drive in the latter modification is replaced by an automatic electrode drive in the former. The are electrodes 19 and 21 are supplied from the source 31 through the exciting coil of a relay 74. As long as there is an are between the electrodes 19 and '21, the relay is energized and pivots arm 23 downward against the action of a spring 72 to maintain the electrodes separate. When the arc is interrupted, current flow through the coil of the relay ceases, and the arm 23 pivots downward so that the electrodes 19 and '21 are engaged and close the circuit. The relay is now "reenergized, the electrodes are separated, and "an are between them is initiated.
The conducting side plates 49 bounding the ring space 46 in the modification shown in Fig. l are replaced by a plurality of separate rings 71 of diameters which 'progressively increase from the shells 43 to the shell 47 in the Fig. 4 modification. The rings are supported from insulating brackets 73 secured to the base 9 and the top plate, and are maintained at potentials which become gradually more negative as their distance from the are 17 increases. The external fields impressed by the potentials applied to the rings 73 thus approach the electric equivalents of the magnetic field for the isotopes more closely than in the Fig. 1 embodiment, and the desired resultant field is produced more effectively by the space charge.
By varying "potentials impressed upon the rings 71, space-charge neutralization occurring under each ring may be controlled, and thus the depth to which the resultant space potential corresponds to the values in the region of curves 63 and 61 of Fig. 3 is controlled. If the potential of a ring 71 is increased in magnitude, the electron emission from it per ion striking it is increased, and the space-charge neutralization in the space opposite it "is increased. The space potential in the space opposite it rises then above the values in the region between the curves 63 and 61, and fewer ions are turned back in the space. If, the potential of a ring 71 is reduced in magnitude, the electron emission from it is reduced, the space-charge neutralization opposite it is reduced, the space potential opposite it sinks to values which correspond to "points deep in the region between curves 63 and 61, and more ions are turned back in the space opposite it. v
In practice, the potential of each of the rings 71 is decreased from a large magnitude potential until further small reductions in potential magnitudes begin to affect the ion collection at the outer collector 47. At thispoint, the space potential begins to assume values corresponding to the region between curves 63 and 61. The potential magnitudes of each of the rings is then lowered by small amounts until the collection at the outer electrode is reduced by an amount corresponding to the turning back of all the light ions.
In Fig. 5, the operation of the Fig. 4 modification is illustrated graphically. The medium curves 61 and 63 again represent the electric equivalents of the magnetic field forthe heaviest isotope and a lighter isotope. The light curve 75 represents the external electric field impressed on the ions. The heavy curve 77 represents the resultant electric field attained by combining the spacecharge field and the external electric field.
Although I have shown and described certain specific embodiments of my invention, 1 am fully aware that many'modifications thereof are possible. My invention, therefore, is not to be restricted except insofar as is necessitated by the prior art and by the spirit of the appended claims.
I claim as my invention;
.1. For use in separating the isotopes of an element, the combination comprising a source of ions of the isotopes of said element, means for subjecting said ions to an electric field radial about said source, means for subjecting said ions to a magnetic field substantially perpendicular to said electric field, a substantially cylindrical collector for said ions with said source on its axis, the magnitudes of said electric and magnetic fields being so related to the massof said isotopes that the ions of one or" said isotopes'predominantly initially move away from said source and finally return in the direction of said source without reaching said collector and the ions of another of said isotopes predominantly move away from said source and reach said collector, said electricfield being modified substantially by the space-charge effect of said ions.
2. For use in separating the isotopes of an element, .the combination comprising asource of ions of the isotopes of said element, means for subjecting said ions to an external radial electric field having substantially circular symmetry about said source, means for subjecting said ions to a magnetic field at right angles to said electric field, a collector ,for said isotopes also having circular symmetry? about said source, the resultant of the field of the space charge of said ions and said external electric field being a field which, over a substantial range of the distances from said source, is, at each point, not much different from the electric field equivalent of the magnetic field for one of said isotopes.
3. An electric discharge device comprsing a source of positiveions of a material having more than one isotope, and collecting means for said positive charges consisting of a hollowring of conducting material the bases of which are made up of insulated laminations of conductive material.
4. An electromagnetic isotope separator comprising a cylindrical enclosure having a central axis, a source of ions having more than one isotope positioned substantially on said axis and means for adjusting the mean value with respect to time of the electric potential within said cylinder at a plurality of radii thereof.
5. An electromagnetic isotope separator, with ion source means and voltage applying means acting to attract ions from said source to a collector electrode and the magnitudes of each said means being suflicient so that said separator may be operated beyond the minimum in M log E, Where M is the mass of the isotopic material treated, and E is the enrichment factor;
6. An eiectromagnetic isotope separator, comprising a vacuum space traversed by a magnetic field and bounded by metallic surfaces certain of which are perpendicular to said magnetic field and have high electron emissive properties.
vacuum space in which ions of the isotopes move, and
means for adjusting at will a space charge neutralizing electron emission into the space. a
8. The method of operating an electromagnetic isotope separator, having an ion source means and a voltage applying means subjecting said ions to an .electric field, which comprises making the flow of ions from said source and the magnitude of saidvoltage of the values which are necessary and sufiicient so that said'separator operates beyond the minimum in its M log E characteristic where M is the mass of isotopic material undergoing treatment in unit time and E is the enrichment factor for said material.
9. An electric discharge device comprising a source of positive gaseous ions positioned in the axis of a cylindrical chamber the side walls of which constitute a collecting electrode for said ions, at least one end wall for said chamber being made up of insulated annular conductors, means for inducing a magnetic field in said chamber parallel to said axis, and connections for impressing predetermined potentials on said annular conductors relative to said source.
10. An electric discharge device comprising a source of positive ions of a material having more than one isotope, collecting means for said positive charges consisting of a hollow ring of conducting materialhaving said source at its center, means for impressing a negative potential on said ring relative to said source, and partitions which are made up of insulated laminations of conductive material across the respective end faces of said ring.
11. An electromagnetic isotope separator comprising a cylindrical enclosure having a central axis, a source'of ions having more than one isotope positioned substantially on said axis, means for producing a radial electric field in the space between said source and said enclosure, and means for adjusting the mean value with respect to time of the electric potential within said enclosure at a plurality of radii thereof.
12. An electromagnetic isotope separator comprising a cylindricalenclosure having a central axis, a source of ions having more than one isotope positioned substanelectric field between said source and the walls of said enclosure, and means for adjusting the mean value with respect to time of the electric potential of said electric 'field at a plurality of radii thereof.
14. An electromagnetic isotope separator comprising a cylindrical enclosure having a central axis, a source of ions having more than one isotope positioned substantially on said axis, means for'producing an aperiodic radial electric field between said source and the walls of said enclosure, means for producing a magnetic field substantially parallel to said axis within said enclosure, and means for adjusting the mean value with respect to time of the electrical potential of said electric field at a plurality of radii thereof.
References Cited in the file of this patent UNITED STATES PATENTS King Apr. 18, 18 82 Snook May 17, 1927 (Other references on following page) 13 UNITED STATES PATENTS Slepian Oct. 11, 1927 Lawrence Feb. 20, 1934 Muller Dec. 4, 1934 Hollrnann Mar. 28, 1939 5 Kuhn et a1 Oct. 22, 1940 Bleakney Nov. 12, 1940 Jonas Jan. 21, 1941 14 2,252,508 Hoff Aug. 12, 1941 2,258,149 Schutze Oct. 7, 1941 2,261,569 Schutze Nov. 4, 1941 OTHER REFERENCES Physical Review, vol. XI, No. 4, pages 316325. Oliphant et a1.: Proceedings Royal Society of London (1934), V146A. Pages 922929.