IL36461A - Method and apparatus for separating isotopes - Google Patents

Description

36461 /2 «· s ■·¾"·«!·· at » St «» e»»ss ·»«:·» ·»»«· as =;·-» a_«- - Method and apparatus for separating isotopes JERSEY NUCLEAR-AVCO ISOTOPBS, Ci -54722 DKT- AERL 107 The present invention relates to a method of and apparatus for separating isotopes of a material containing a first isotope and a second isotope.

The separation of quantities of uranium into two parts of which one is enriched in the fissionable isotope while the other part is correspondingly depleted in U23 is an important process for nuclear applications. The only separation method now used on an industrial scale is the gaseous diffusion process * Many alternative methods have been suggested such asr for example, electromagnetic separation, centrifugal separation, and separation by thermal diffusion, but none of these -are known to exhibit substantial advantages over the method of separation by gaseous diffusion.

The present invention is an improvement of yet another isotopic separation process shown and described only broadly in the United States Patent No, 3*1 4-3» 087, issued May 6. 1969, to Jean Robieux and Jean-Michel Auclair to which reference is made, TJie isotope shift which pertains to the spectra of elements or molecular compounds is an effect important to the Robieux et al patent as well as the present invention. Particular lines in the spectrum of a particular isotope of a given element may have wavelengths element. Similarly, if the elements from which a molecular compound is constituted are isotopically pure, the spectral examination of compounds containing different isotopes will generally show slight shifts in the wavelengths of particular lines in the spectrum of the compound.

The isotope shift is present in both the emission and absorption of light. In k emission, the effect has been used to measure the isotopic composition of samples of several elements. In the case of uranium, for example, o a suitable line is one which has the wavelength 5027 . kA. This line is centered at 0 7 . 29 ^ for the fissionable isotope U2^, and at 5027 . 398A for U2^8. The corresponding o isotope shift in this case is 0.103A. Other lines also exhibit isotope shifts. In absorption, a gas composed of a mixture of isotopes may be illuminated by light containing a narrow range of wavelengths at or near the natural wavelength of one particular isotope. For example, a sample of uranium vapor containing atoms of U2^ and U2^® may be illuminated with a beam of light in which the wavelengths pre- o sent are confined to a narrow band of 0. 01A centered at an average wavelength of 5027 . 295A . The light will be strongly absorbed by U^3 atoms, but only weakly absorbed by U^39 atoms. Atoms absorbing, light quanta beoome excited, so that the effect of the illumination is to make a gas mixture composed of excited and unexcited (ground state) atoms in which the population of excited atoms will be enriched in U2^ ^ and the population of ground state atoms will be correspondingly depleted in U235.

The aforementioned Robieux et al Patent in accordance with isotope shift contemplates the illumination by two light sources of a gas composed of an uranium molecular compound such as uranium hexafluoride , UP^. The bandwidth of the first light source is narrow and its wavelength is so chosen that it preferentially excites those UF^ molecules which contain an atom of U23i>. The second light source functions to ionize only those molecules in the mixture which are in excited states, leaving ground state moleoules unaffected. Since the population of excited molecules produced by the first light source is enriched in u23S, it follows that the population of molecular ions produced by the second light source is enriched in U2^. The molecular ions produced in the previously described manner may then be separated from the neutral molecules by use of electric or magnetic fields.

According to the invention there is provided a method of separating isotopes of a material containing a first isotope and a second isotope comprising the steps of: (a) producing a vapor comprising atoms of said material; (b) subjecting said vapor to a first radiation emitted by a first radiation source at a predetermined frequency that excites the atoms of said first isotope but does not substantially excite the atoms of the second isotope; (c) subjecting said vapor to a second radiation emitted by a second radiation source at a predetermined frequency which ionizes the excited atoms of the first isotope but does not substantially ionize the atoms of the second isotope; and (d) separating said ionized atoms from the non-ionized atoms.

We also provide an apparatus for separating isotopes of a material containing a first isotope and a second isotope comprising: (a) scalable container means adapted to be connected to a vacuum pump, said container means having an opening concentric about an optical axis for admitting radiation to the interior of said container means; (b) means disposed within said container means for providing an atomic beam of said material normal to and passing through said optical axis, said beam having a dimension parallel to said optical axis that is long compared to its dimension normal to said optical axis; (c) first collector plate means disposed within said container means for receiving said atomic beam; (d) second collector plate means disposed within said container means adjacent said beam; (e) a first radiation source for producing a first beam of radiation at a predetermined frequency that excites the atoms of said first isotope but does not substantially excite the atoms of said second isotope; (f ) a second radiation source for producing a second beam of radiation at a predetermined frequency which ionizes the excited atoms of said first isotope but does not substantially ionize the atoms of said second isotope; (g) means for directing said beams of radiation into said container along said optical axis; and (h) means for causing said ionized atoms to im Whereas the Robieux et al patent teaches the provision of the working medium in the form of a molecular compound a feature of the present invention is the provision of the working medium in the form of atoms, such as uranium atoms where isotopes of uranium are to be separated. The provision of uranium atoms rather than molecular compounds containing uranium is advantageous for several reasons.

One such reason concerns the availability of lasers suitable for use as light sources for the primary illumination. The spectrum of uranium atoms exhibits many examples of the isotope shift effect in the visible part of the spectrum whereas for the most part identifiable isotope shifts in molecular compounds of uranium are found in the infra-red region of the spectrum. Lasers are desirable as light sources in this application by virtue of the large intensities available, but the narrow bandwidth and exact wavelength requirements for the first light source will require, in general, that a finely tunable laser be used. Presently available dye lasers having bandwidths as small as o 0.01A may be tuned across the whole visible spectrum. "Dial-a-line" lasers which are manufactured and sold by the Avco Everett Research Laboratory, Everett, Massachusetts may be used as sources of both exciting and ionizing photons.

Should minor modification of this laser system be necessary in order to obtain the necessary narrow bandwidth and intensity, this may be accomplished by using a conventional etalon filter to obtain a narrow bandwidth and by configuring the laser in a master oscillator power amplifier mode of operation wherein one laser is used to control another to source. Such lasers may be used for the selective excitation of atoms in accordance with the teaching of the present invention. In particular, the spectrum of the uranium atom exhibits at least one line which lends itself to the present invention. This is the above-mentioned line o at 5027. l\.A . This wavelength is achievable by the aforemen- o tioned dye laser. Furthermore, the bandwidth of about 0.01A achievable by the dye laser is much smaller than the isotope shift of 0.103A between U23 and U2^8.., From the preceding, it will clearly be seen that practical selective excitation of u235 is possible in accordance with the present invention.

However, so far as is known, no tunable laser with the required narrow bandwidth is available that operates in the infra-red region of the spectrum. Accordingly, it is not apparent that a suitable laser is now or ever was available for use in accordance with the teaching of the Robieux et al patent.

A second advantage associated with the utilization of uranium atoms in carrying out the present invention is the relative freedom it permits in the choice of the wavelength of the second or ionizing laser. This freedom may be illustrated by supposing that the first or exciting laser o is tuned to the aforementioned wavelength 5027. l+A. The energy of a photon at this wavelength is about 2.5 eV. The ionization potential of uranium is about I4. eV. Thus, the energy required to ionize a uranium atom excited by the o absorption of a photon at 5027.ij-A is somewhat greater than I.5 ©V. Hence, photons from the second or ionizing laser must have an energy in excess of about 1.5 e in order to be they should not have an energy in excess of l eV in order to avoid the direct ionization of ground state uranium atoms. The corresponding allowable wavelength range extends o o from about 3100A to about 8000A. In contrast to the wide range of choices that are available when uranium atoms are used in carrying out the present invention, the use of molecular compounds selectively excited in the infra-red in accordance with the Robieux et al patent requires the ionizing radiation to lie in a narrow band in the ultraviolet. Within the wide range of wavelengths made available in the method and apparatus to be described hereinafter, it is only necessary to insure that the wavelengths chosen for the second light source does not correspond to a resonance line in uranium. This can be accomplished by referring to a standard table of the optical spectrum of uranium.

A third advantage associated with the utilization of uranium atoms in carrying out the present invention is that the above-mentioned wide range of wavelengths that can be used in the ionizing laser can permit two different types o of exploitation. Firstly, for the example of the 027. l\A line mentioned above, it may be noted that this wavelength is itself a permissible choice for the second laser. That is, the absorption of a second photon from the first laser by a single excited atom of 11^35 can achieve the desired ionization from the excited state. In this event, the second light source may be dispensed with entirely. Secondly the wide range of wavelengths suitable for the second laser may permit the phenomenon of autoionization to be used to advantage. In the phenomenon of autoionization the rob b of ch certain well-defined wavelengths. The use of such a wavelength can increase the efficiency of a separation plant or it can reduce the optimum size of such a plant. In contrast to this, it is to be noted that the possibility of dispense ing with the second light source or of us¾ng the phenomenon of autoionization is quite small, if not non-existent, where a molecular compound of uranium is used as the working medium in accordance with the teaching of the Robieux et al patent.

A fourth advantage associated with the use of atoms in accordance with the present invention instead of molecular compounds is that such use of atoms avoids entirely a class of problems associated with chemical or photochemical reactions. For instance, the illumination of UF^ with ultraviolet light can lead to ionization, dissociation, or both. If only dissociation takes place, separation by electromagnetic means can not be achieved. Further, if both dissociation and ionization take place, then only one component can be ionized but it is not clear that the correct component will be ionized. Such problems do not arise when atoms are used in accordance with the present invention.

In carrying out the present invention, the working medium is preferably provided in the form of a colli-mated atomic beam. Such a beam can, for example, be conveniently formed in atomic uranium by using an oven in which a uranium vapor is formed and which is connected by a collimator to an evacuated chamber. The atoms in such a beam are essentially all collected on a cooled plate facing collecting plates are disposed in the evacuated chamber containing the aforementioned collection plate at suitable locations not facing the collimator, U^3£ atoms once ionized in accordance with the invention can be deflected out of the beam by electromagnetic means, for example, and then be collected on such other collecting plates.

In carrying out the present invention, the operation of the separation system may be in part continuous and in part repetitively pulsed as follows : the oven and collimator operate best in a continuous mode whereby a beam of uranium atoms emerges steadily from the collimator.

However, on account of the short natural lifetimes before spontaneous decay back to the ground state of the excited uranium atoms, it is desirable to operate the lasers in repetitive high power pulses. The natural lifetime of a o uranium atom excited by a photon at 5027.l+A is about 55 nanoseconds (ns ). Thus, the efficiency of the process will be improved if the excitation and ionization steps are completed in a time less than 55 ns. Fortunately, this requirement fits well with typical characteristics of lasers useful in the process. The two lasers must be arranged to illuminate the same volume and the firing of the two lasers must be nearly simultaneous, The repetition rate is determined by the consideration that essentially all the uranium emerging from the collimator should be subjected to at least one pair of light pulses from the lasers. The interval between pulses should, therefore, be approximately equal to the time taken by a typical atom of uranium in the beam emerging from the collimator to cross the region subject to Among other things, the present apparatus can be so arranged as to permit the achievement of large, efficient separation of uranium in a single stage, and in any event a separation substantially greater and more efficient than taught by the above-ment oned Robieux et al patent. The considerations associated with the achievement of large efficient separations may be better understood by reference to an ideal process in which: a) the beam of uranium emerging from the collimator consists wholly of neutral atoms in the ground state; b) the process of extracting the ions from the atoms is free from losses associated with mixing, charge exchange, etc; c) every exciting photon is absorbed by a ground state U235 atom, and every ionizing photon is absorbed by an excited TJ235 atom; d) the laser pulses excite and ionize every atom but no U238 atom.

When all these conditions are satisfied, the separation achieved is perfect in that each emerging stream is iso-topically pure; furthermore, the process is ideally efficient in that just one exciting photon and one ionizing photon is required to separate one atom of U235e The present description teaches means and apparatus for approaching at least some of the ideal conditions outlined above.

Condition (a) is essential to the achievement of a good separation since if some of the uranium emerging from the collimator is excited or ionized before illumination, illumination can consist solely of XJ235 will clearly be impaired. In general, excitation is not expected to be a problem at typical densities, collision frequencies, and dimensions. But since temperatures higher than about 2000°C. are required in the oven, some degree of ionization of the emerging uranium beam, possibly on the order of a few per cent, may occur as a result of thermal effects in accordance with the well-known Saha equation, and also as a result of contact ionization. Specific means to reduce the number of ions emerging from the collimator are discussed below in connection with the description of the preferred embodiment.

Condition^ (b ) is likewise essential to the achievement of a good separation. In the ideal process, the situation after illumination by the two lasers is such that a perfect separation has been achieved in the limited sense that all the TJ23 is ionized and none of the U238 i 3 ionized.

But at that stage, no physical separation has been achieved. Physical separation may be achieved by the application of suitable electric or magnetic forces to the partially ionized gas - the ions being U23i> and the neutrals being Such a force is felt directly only by the ions, or in other words, only by the The acceleration produced by this force gives to the ions (u235 ) a certain velocity relative to the neutrals (TJ238 ) . if this relative velocity is sustained, the ultimate result is the desired physical separation between TJ23 and U238.

Two types of collisions can interfere with the effective accomplishment of this separation, namely, charge exchange and kinetic collisions. In charge exchange col interaction, it is the U235 that is neutral while the U238 is ionized. However, in charge exchange collisions, the exchange of momentum is negligible, so that any relative velocity between the U 35 ancL U 38 is not altered.

In carrying out the present invention, the electromagnetic force is preferably applied to the partially ionized gas in a manner such that the acceleration phase is accomplished without appreciable degradation through charge exchange collisions. This means that the force is applied as a short pulse; the displacement of the ions during the pulse is so short that no appreciable charge exchange can occur. Upon application of the force in this manner, the U235 ions will indeed acquire a velocity relative to the U23$ neutrals. The force is turned off after a sufficient acceleration has been accomplished. Since, as explained above, there is no appreciable momentum transfer in charge exchange collisions, such collisions occurring after the acceleration pulse cannot degrade the degree of separation finally achieved, but merely affect the ratios of the ions to the neutrals in the streams of U23S and These ratios have no importance in determining either the degree of separation achieved, or the efficiency of the overall process .

The second type of collision which can interfere with effective separation of the isotopes is momentum transfer collisions, also known as gas kinetic collisions. If the U235 is given a velocity relative to the such collisions will eventually destroy such relative velocity and consequently make separation (or at least large degrees Accordingly, in order to reduce the effect of gas kinetic collisions, the beam of atomic uranium is preferably provided in a form such that the dimension of the beam in one direction at an angle to the direction of its mean motion is short. The direction associated with this short dimension is referred to as the extraction direction. More precisely, for large separations, the thickness of the atomic beam in the extraction direction should be not greater than about one gas kinetic mean free path. The exact choice of this dimension depends inter alia, on the desired degree of separation, the density of the uranium vapor, the gas kinetic cross-section for uranium and other factors.

Condition (c) pertains rather to the efficient use of laser output power than to the degree of separation. If the lasers are to be used efficiently, they must illuminate enough gas to absorb their entire output. More precisely, the product of the density of U23i? atoms, and the length of the vapor in the direction of illumination must have a sufficiently great magnitude. The actual magnitude is dependent upon the cross -section for absorbing exciting arid' ionizing photons. In practice, of these two cros3-sections , the cross-section for a photoionization is by far the smaller; therefore, the above-mentioned product is determined by the cross-section for photoionization. Although this cross-section is not precisely known, it is certainly so small that considerable lengths of uranium vapor will be required. In this connection, the phenomenon of autoionization may be important, since this phenomenon, if present, permits the section is unusually large.

The direction of the laser light should be perpendicular to the direction of mean motion of the atomic beam, But as explained above, condition (b ) requires that the dimension of the atomic beam in one direction (the extraction direction) at an angle to the direction of mean motion of the atomic beam be short. On the other hand, conditions (c) requires that another dimension of the atomic beam (the illumination direction) perpendicular to the direction of mean motion of the atomic beam be long. Therefore, it is preferred that the cross -section of the atomic beam in a plane perpendicular to the direction of mean motion of the atomic beam should be long and thin in shape. The direction of mean motion, the extraction direction, and the illumination direction should be more or less perpendicular. Apparatus for achieving this objective, and typical dimensions are discussed below in connection with the preferred embodiment.

The final consideration relating to the degree and efficiency of separation is condition (d). While the exciting light will be much more strongly absorbed by the U23i than by the U238, there will nevertheless be at least some absorption by The ionizing light will ionize all excited atoms, without distinction as to isotope. Thus, the absorption of exciting light by the TJ238 will decrease the degree of separation achieved, and decrease the efficiency of utilization of the light. In addition, condition (d) requires that there be at least as many photons in each illuminating pulse as there are TJ235 atoms in the column of If the column of U235 vapor has a length in the direction of illumination on the order of the absorption length for an ionizing photon, it will, in the same direction, have a length equal to many absorption lengths for an exciting photon. The absorption length for absorption of an exciting photon in the column of U238 is longer than the corresponding length of the absorption of an exciting photon in the column of TJ235 by the ratio of the respective cross-sections, but shorter in the ratio of the respective densities of U238 and TJ23^. In some cases this absorption length may imply a significant absorption of the exciting light by the U 38, with consequent degradation of the overall performance. Accordingly, a variety of optional means for overcoming this problem are discussed below in connection with the preferred embodiment. These all involve means for introducing the exciting light into the uranium vapor in ways such that such exciting light does not traverse the full length of the vapor column before arriving at the point where it interacts. Of course, in all cases, the two illuminations must coincide in both space and time in the interaction region.

In order that the invention may be fully understood, it will now be described with reference to the accompanying drawings, in which: Figure 1 is a perspective view of an apparatus including a series of modules by which the method in accordance with the invention may be carried out; Figure 2 is a side elevation of a typical module shown in Figure 1; and Figure 3 s a diagrammatic representation of the electrode configuration shown in Figure 2.

As previously mentioned, it is desirable to produce an atomic beam having a long thin cross-section. Whil the thickness of the beam may be on the order of centimeters, for example, the total length may be on the order of hundreds of meters or even more, and is only limited in principle by the laser powers available, and by optical dif fraction effects. In order to simplify the construction of such an apparatus, it may be constructed as shown in Figure 1 of a number of substantially similar modules 1 disposed along and enclosing an optical axis. Each module may be, for example, on the order of 1 meter in length, while trans verse dimensions may be on the order of a few centimeters. The modules are preferably essentially interchangeable.

As shown in Figure 1 the modules 1 are connected to each other by lengths of tubing 2, each such length of tubing carrying a vacuum valve 3° The vacuum valves 3 are so constructed that at least when they are open, an uninter rupted path is provided for the laser beams down the entire length of the column of modules. Each module 1 is vacuum tight, and is connected by a vacuum port Ij. to a vacuum system (not shown) designed to maintain by standard means a base pressure on the order of about 10"^ mm. of mercury in all parts of the apparatus. Each module also requires the provision of cooling and electric power for the ovens and other functions more fully described hereinafter. Electrical power is supplied to all the modules by a conventional power source 5 through leads 6 and 7. Cooling may be conventionally supplied by water from a suitable source (not shown) carried through pipes 8.

Figure 1 also shows two lasers 9 and 10 at one end of the column. In this embodiment, the lasers illuminate substantially the same columnar volume extending through all the modules. This may, for example, be achieved as shown in Figure 1 as follows : the output beam 11 from laser 9 and the output beam 12 from laser 10 both impinge upon the same area of a dichromic mirror 13. The properties of the dichromic mirror 13 are such that the beam 11 is substantially transmitted and the beam 12 is substantially reflected by the said dichromic mirror 13- When the angles are suitably adjusted, the desired illumination of the same columnar volume by both laser beams can be obtained. Thus the beam II4. leaving the dichromic mirror 13 contains both the exciting and the ionizing laser beams. It is not significant whether laser 9 or laser 10 is the exciting or ionizing laser or vice versa. It should also be understood that the single lasers shown, 9 and 10, may represent arrays of such lasers, suitably aligned and synchronized.

The laser light enters the column of modules and the vacuum system through the window 1$ which may be composed, for example, of optical grade quartz. A long evacuated pipe 16 is provided between the window and the point of entry of the first module in order to avoid possible contamination of the window surface by, for example, uranium vapor. For purposes of convenience and clarity, such electronic and other apparatus as may be needed to control the operation of the entire assembly is not shown. Figure 2 shows a cross-section of a module in a plane perpendicular to the direction of the laser beam ll . Most of beam of about 1 meter. The walls of the module are composed of a good vacuum material such as, for example, stainless steel, and are shown in sections labelled 17 through 23. Sections 21, 22 and 23 generally enclose the oven and its supply of uranium; sections 17 through 20 generally enclose the region of interaction with the laser beams, the region of application of the electromagnetic separation system, and the storage arrangements for the separated isotopes. One section, such as, for example, section 18 is made removable to provide access to the interior of the module for isotope removal, replenishment of the feed material, repair and maintenance and the like. End plates completing the enclosure and connected to the pipes 2 of Figure 1 are not shown.

Cooling pipes 8 are carried on the various walls of the module to maintain the module at a moderate temperature. Each module in the assembled condition is vacuum tight and is provided with a connection I. to a vacuum pumping system (not shown) designed to maintain the proper vacuum within the modules.

In the lower part of the module are disposed a uranium oven 2I . and collimating means 28. These are comprised of heating elements 25, a crucible 26 containing liquid uranium metal 27, a collimator 28 and radiation baffles 29 and 30 suitably spaced one from another. The crucible 26 and the collimator 28 are composed of a suitable refractory material such as, for example, graphite. The temperature at the exposed surface of the uranium may be about 2300°C. in which case the density of uranium atoms in The design of the collimator is such that it defines a slightly diverging beam of uranium vapor. One purpose of having the extent of the liquid uranium surface extend outside the area from which a straight line might be drawn right through the collimator is the following: contact ionization of uranium atoms is more probable on graphite surfaces than on liquid uranium surfaces. Thus, the arrangement shown reduces the fraction of ions in the emergent beam of uranium. The purpose of the baffles 29 and 30 is to reduce the loss of heat from the oven region. The vertical arrangement of the oven assembly ensures that uranium atoms which strike the collimator (as opposed to those that escape through the collimator) will condense there into liquid form (the melting point of uranium is about 11 0°G. ) and flow back under gravity into the main pool of liquid uranium 27. Note that the collimator 28 is maintained at a temperature lower than the liquid uranium pool 27 but above the melting point of uranium.

The atomic beam emerging from the collimator into the upper section of the module substantially fills the region 31 · The portion 32 of region 31 illuminated by the lasers is shown in Figure 2 as circular; it need not necessarily have this shape, but it should be such that all the atoms in the beam must pass through the illuminated region. At the same time portion 32 should not extend far beyond the edges of the atomic beam. The repetition rate of the lasers preferably corresponds to the characteristic time taken by a uranium atom to cross the illuminated portion 32 j 10 kiloHerz may be taken as typical. around the illuminated portion 32 as shown. These plates may be composed of stainless steel or a suitable refractory material. The collecting plates are supported by the walls of the module, but the mechanical connections (which may be conventional) are arranged and adapted to leave the plates electrically insulated from the walls as well as from each other. The plates are maintained at a temperature sufficiently low that uranium atoms striking the plates condense in solid form. If necessary.,* cooling can be supplied to the plates to accomplish this purpose. In the absence of the extraction system, most of the uranium would impinge on plates 33 and 3 · The extraction system as described below functions to drive TJ235 onto plate 36, leaving mostly U238 to impinge upon plates 33 and 35· Plate 3k- will not collect much uranium, but is helpful in maintaining symmetry of the electric fields during the extraction phase.

The extraction arrangement illustrated in Figures 2 and 3 operates on the principle of the low density crossed-field magnetohydrodynamic accelerator. The detailed manner of operation of this device will be described after its component parts have been indicated. The entire module rests on a support plate 37» and is almost surrounded by a piece of magnet iron 38 containing pole pieces 39, 1+0 and I4.I. Each pole piece has a ~¾T*£'fcabTy energized coil (shown, but not numbered). In the region 32 illuminated by the lasers the direction of the magnetic field is substantially from pole piece 39 to pole piece I.O. This magnetic field is essential to the working of the crossed-field accelerator.

In the region of the collimator 28 , the magnetic that is, it is at an angle to the mean direction of motion of the atomic beam in the collimator. The purpose of the pole piece I.I is to produce a magnetic field at or about the collimator 28 that inhibits the flow of any ionized uranium out of the collimator.

Typical magnetic field strengths may be between 100 and 1000 gauss.

Further components of the crossed-f ield magneto-hydrodynamic accelerator are best shown in Figure 3· Elec-trodes -4.2a, b, and c and l3a, b, and c are provided at each end of each column illuminated by the laser, that is, about 1 meter apart in each module. A power supply ijlj. is connected to the oppositely disposed electrodes -4.2a, b, and c and -4.3a, b, and c to provide a controlled electric current through the column of partially ionized uranium vapor in portion 32. The power supply I4I1. is arranged and adapted in any suitable manner to maintain electrodes 12a, b, and c respectively at controlled different voltages, The aforementioned electrodes may be composed of a suitable material such as, for example, graphite. The subdivision of the electrodes is helpful in controlling the operation of the accelerator and the number of such subdivisions may be greater than or less than three. The electrodes must be disposed so as not to intercept the laser beams; they can be displaced slightly from the ends of the column in a direction parallel to the magnetic field. When the electrodes are subdivided as shown, the potentials on adjacent electrodes at each end of the column are maintained in such a way as to provide an electric field which will drive the It is convenient to consider that the power supply 1+ provides a pulse of constant current, although in practice other arrangements may be acceptable. If the current, and the current density remain constant, the magnitude of the applied force per unit volume also remains constant. Therefore the ionized component of the uranium is subjected to a constant acceleration. If, for example, the current is 1 ampere, the cross section of the column is 1 square centimeter, the magnetic field is 1000 gauss, and the number density of ions (which are IJ235>) is lO^-2 per cubic centimeter, the resulting acceleration is approximately 2.5 x ion centimeters per second per second. If this acceleration is sustained for 1 microsecond, the final velocity of the U2 ions in the direction of the plate 36 is 2. x 10^ centimeters per second. Since this speed is substantially larger than speeds characteristic of the atomic beam which are on the order of 3 x 10^- centimeters per second, a laser acceleration may be sufficient. The displacement of the U23 during the acceleration phase is 1.25 millimeters, a distance significantly smaller than likely values for the charge exchange means free path. Thus, these sample values are consistent with the feature of the present invention mentioned above, namely that the extraction velocity should be acquired by the U235 ions in such a manner that their displacements during the acceleration phase should be substantially less than a charge exchange mean free path.

The power supply JL-JL4. must provide the required current against the sum of two voltages. The first of these is the small voltage associated with the electrical re modulus is 1 meter long. The second voltage is the back e.m.f. induced by the acceleration of the ions. This voltage is initially zero, but rises to a final value which may be about 250 volts when the ion speed is 2.5 x 10^ centimeters per second.

According to a further feature of the present invention, reversal of the direction of the magnetic field and of the current flow in alternate modules prevents the build up of unnecessarily large voltages along the length of the column of modules.

The actual acceleration of the ions is accomplished by an electric field in the direction of the collection plate 36. This electric field is induced by the current flow and the magnetic field in accordance with the principles of cross-field magnetohydrodynamic accelerators operating in a regime where the electron cyclotron frequency greatly exceeds the collision frequency (^θ^θ»1). The magnitude of this electric field for the parameters discussed above is approximately 60 volts per centimeter and this field governs the relative potentials of the electrodes at each end of the column when these electrodes are subdivided. The greater the degree of subdivision, the greater will be the reduction in losses due to inhomogeneities at the ends of the column. Electrical insulation of the four collecting plates from each other also helps to maintain the correct electric field distribution.

When the acceleration phase is complete, the power supply I .I . is open-circuited so that no further forces are applied externally to the ions. Thereafter, as indicated towards the collecting plate 36 regardless of the occurrence of charge exchange collisions.

The overall operation of the separation system as described above will now be discussed. The beam of atomic uranium emerges continuously from the oven via the collimator. The lasers operate in simultaneous pulses of perhaps 10 nanoseconds duration. Immediately following the laser pulses, the power supply for the crossed-field magneto-hydrodynamic accelerator provides a pulse of perhaps 1 ampere for 1 microsecond and then the power supply is open-circuited. This procedure is repeated at a rate such that all the uranium emerging from the collimator is illuminated; an appropriate frequency being as previously noted, about 10 kiloHerz. Eventually, all the natural uranium in the oven will have been evaporated and at this time the module may be removed from the column and opened. The collecting plates carrying enriched and depleted uranium are removed, a fresh supply of raw material and fresh collecting plates are introduced, and the module is closed and returned to service.

Although the preferred embodiment has been discussed above in conjunction with Figures 1-3, many variations and modifications are possible, some of which are indicated below.

The considerable length of the system suggested in Figure 1 stems from the necessity of providing a long path through the uranium vapor if the ionizing photons are to be substantially absorbed. However, the length of the system may be reduced by making the laser beams pass not once, but coraplished by using mirrors, For example, a mirror placed at the opposite end of the column of modules (using another long evacuated pipe similar to the pipe 16 in Figure 1) will cause the laser beams to traverse the column twice and additional optical components can provide still further tra-versals. Since the absorption of the ionizing photons depends on the total path length through the vapor, the length of the column of modules can be reduced in rough proportion to the number of traversals. A further advantage is that in this way a smaller system may be built which (apart from losses due to the additional optical components) is just as efficient as a larger system.

In some cases it may also be desired to modify the optical system to take account of the problem mentioned above of absorption of some exciting radiation by U238.

This problem can be reduced by supplying the exciting radiation (but not the ionizing radiation) separately to each of several groups of modules, or even separately to each module. In either case, additional optical components and synchronizing arrangements will be required. Further, there is no requirement that both photon beams must be supplied to the uranium vapor from the same direction,, If several exciting lasers are used for any reason, it will be convenient to utilize one low power laser with suitable feedback frequency control to serve as a frequency standard for all the other exciting lasers. In this mode, the controlling laser operates as a high precision oscillator, and all other lasers operate as amplifiers.

It should also be understood that it is alterna instead of towards plate 3& by reversing either the current direction or the magnetic field direction but not both.

This would require some modification of the geometry such that plate 35 would not intercept the neutral atomic beam, but may be found useful in the event that difficulties are experienced with contamination of plate 36 as a result of U 38 tails peeling off of plates 3 and 33= A modification of the crossed-field magneto-hydrodynamic accelerator can be obtained using a basic module similar to that shown and described in connection with Figure 2 with the exception of the magnet and electrode structure „ If the directions of the current and magnetic field are interchanged, the resultant force is unchanged in direction but reversed in sense. Thus, an accelerator is possible in which the magnetic field is directed parallel to the laser beam, and a current is caused to flow in the vapor in the direction connecting the collecting plates 3i+ and 35 of Figure 2. Such a magnetic field could be provided for example by enclosing each module in a large solenoid. The electrodes in this case may extend along the length of the vapor column in each module, and be disposed on the sides of the column closest to plates 3k and 33» subdivision of the electrodes again being helpful. As compared with the case illustrated, an accelerator of this type may require a power supply providing a higher current at lower voltage for the same effect.

There are advantages and disadvantages to both systems.

It might be possible to dispense entirely with the crossed-field magne ohydrodynamic accelerator. This liberated from the U23i> atoms by the ionizing radiation have (or are given) energies on the order of a few electron volts the gas composed of IJ235 ions and electrons will expand at a speed greater than the thermal speed in the atomic beam by the square root of the ratio of the electron energy to the mean energy of the atoms in the beam. This expansion which is driven by electrical forces tends to remove the from the atomic beam so that it may be collected separately from the U236. However, consideration of charge exchange effects suggests that this separation technique will only be efficient at densities lower than those discussed above. The use of a magnetic field to restrict the directions in which such expansion is possible will be helpful.

Finally, there are also several possible methods of producing the atomic vapor beam. The radiant oven heater could be replaced by electron beam or R.P. (radio frequency) induction heating of the crucible and its charge. Alternatively, the crucible problem can be avoided entirely by utilizing either an electron beam or an intense laser beam to heat the surface of the metal and thereby stimulate direct surface evaporation without the necessity for raising the temperature of the entire body of material. Further more, it should be understood that the methods and apparatus described herein can also be utilized for the separation of other uranium isotopes than from a mixture of uranium isotopes by slight modification of the wavelength of the exciting radiation.

Claims (1)

1. > DKT. AERL 107 1, A method of separating isotopes of a material containing a first isotope and a second isotope comprising the steps of : (a) producing a vapor comprising atoms of said material; (b) subjecting said vapor to a first radiation emitted by a first radiation source at a predetermined frequency that excites the atoms of said first isotope but does not substantially excite the atoms of the second isotope; (c) subjecting said vapor to a second radiation emitted by a second radiation source at a predetermined frequency which ionizes the excited atoms of the first isotope but does not substantially ionize the atoms of the second isotope; and (d) separating said ionized atoms from the non-ionized atomsu 2e The method as defined in claim 1, wherein said vapor comprises an atomic beam 3o The combination as defined in claim 1, wherein said vapor is a monatomic gas of said material* i|<, The method as defined in claim 3» wherein said monatomic gas is provided with a first dimension in the direction of emission of said second radiation that is long and a second dimension normal thereto that is small compared to said first dimension. S The method as defined in claim Ij., wherein the radiation emitted by said first radiation source is in the visible spectrum. e The method as defined in claim $, wherein the wavelength of said second radiation substantially corresponds to a transition between the excited state and an autoioniza-tion level of the excited atoms of said first isotope 0 7o The method as defined in claim l\.9 wherein the separation of said ionized atoms comprises: (a) providing a magnetic field at an angle to the direction of said second dimension, and (b) providing an electric field at an angle both to the direction of said magnetic field and to the direction of said second dimension. 8. The method as defined in claim 7> wherein said electric field is pulsed, each pulse having a duration less than the mean free time for a charge exchange collision in said vapor. 9o The method as defined in claim I., wherein said ionized atoms are given an energy of at least about one electron volt to cause them to expand at a speed greater than the thermal speed of unionized atoms in said vapor by about the square root of the ratio of the electron energy to the mean energy in the atoms in said vapor whereby said ionized atoms separate from said unionized atoms. 10. The method as defined in claim I., and additionally including the step of providing a magnetic field at the source of said vapor at an angle to the direction of flow of said vapor. llo A method of separating isotopes of a material containing a first isotope and a second isotope comprising the steps of: (a) producing a raonatomic vapor of said material; (b) forming said vapor into a long, narrow beam generally rectangular in cross -section, the width dimension of said beam normal to its flow direction being large compared to its thickness direction also normal to said flow direction and said width dimension; (c) subjecting said vapor in its width direction to a first radiation emitted by a first radiation source at a predetermined frequency that excites the atoms of said first isotope but does not substantially excite the atoms of said second isotope; (d) subjecting said vapor in its width direction to a second radiation emitted by a second radiation source at a predetermined frequency which ionizes the excited atoms of the first isotope but does not substantially ionize the atoms of the second isotope; and (e) separating said ionized atoms from the non-ionized atoms o 12o The method as defined in claim 11, wherein said material is uranium and said vapor is produced by maintaining said uranium in an evacuated environment at a temperature in excess of its melting point whereby said evaporation of said uranium is produced, said vapor comprising substantially only neutral atoms evaporated from a liquid surface of said material. 13. The method as defined in claim 12, wherein the isotopes of said material which are separated are U2^5 ana U238. H .a The method as defined in claim 13* wherein the isotope U235 i3 separated from said beam substantially in its said thickness direction. 15. The method as defined in claim lij., wherein said first radiation emitted by said first radiation source is in the visible spectrum, l6o The method as defined in claim 15, wherein the o wavelength of said first radiation is about 5027.3A and the wavelength of said second radiation is sufficient to ionize excited by absorption of a photon from said first radiation source and insufficient to ionize U2 8 from the ground state. 17. The method as defined in claim 1 , wherein the wavelength of said second radiation substantially corresponds to a transition between the excited state and an autoioniza-tion level of the excited atoms of said first isotope. 18. The method as defined in claim 11 wherein the separation of said ionized atoms comprises: (a) providing a magnetic field substantially normal to said width dimension; and (b) providing an electric current substantially parallel to said width dimension. 19. The method as defined in claim 18, wherein said electric current is pulsed, each pulse having a duration less than the mean free time for a charge exchange collision in said vapor. 20. The method as defined in claim 11, wherein said ionized atoms are given an energy of at least about one electron volt to cause them to expand at a speed greater than the thermal speed of unionized atoms in said vapor by about the square root of the ratio of the electron energy to V the mean energy in the atoms in said vapor whereby said ionized atoms separate from said unionized atoms. 210 The method as defined in claim 11 and additionally including the step of providing a magnetic field at the source of said vapor at an angle to the direction of flow of said vapor. 22o Apparatus for separating isotopes of a material containing a first isotope and a second isotope comprising: (a) scalable container means adapted to be connected to a vacuum pump, said container means having an opening concentric about an optical axis for admitting radiation to the interior of said container means; (b) means disposed within said container means for providing an atomic beam of said material normal to and passing through said optical axis, said beam having a dimension parallel to said optical axis that is long compared to its dimension normal to said optical axis; (c) first collector plate means disposed within said container means for receiving said atomic beam; (d) second collector plate means disposed within said container means adjacent said beam; (e) a first radiation source for producing a first beam of radiation at a predetermined frequency that excites the atoms of said first isotope but does not substantially excite the atoms of said second isotope; (f ) a second radiation source for producing a second beam of radiation at a predetermined frequency which ionizes the excited atoms of said first isotope but does not substantially ionize the atoms of said second isotope; (g) means for directing said beams of radiation into V said container along said optical axis; and (h) means for causing said ionized atoms to impinge on said second collector plate means β 23. The combination as defined in claim 22, wherein said means for providing said atomic beam includes oven means disposed below said optical axis for retaining said material at a temperature in excess of its melting point whereby said beam of atoms is produced from evaporation of said material. 2I.. The combination as defined in claim 23, wherein said oven means includes means for collimating said atomic beam. 25. The combination as defined in claim 2l .9 wherein said material is uranium, the isotopes which are separated are TJ235 and U238 and said vapor comprises substantially only neutral atoms evaporated from a liquid surface of said material. 26. The combination as defined in claim 2I., wherein said first radiation emitted by said first radiation source is in the visible spectrum. 27. The combination as defined in claim 25, wherein o the wavelength of said first radiation is about 027.3A and the wavelength of said second radiation is sufficient to ionize U235 excited by absorption of a photon from said first radiation source and insufficient to ionize U2-^ from the ground state. 28. The combination as defined in claim 26, wherein the wavelength of said second radiation substantially corresponds to a transition between the excited state and an autoionization level of the excited atoms of said first •t isotope* 29. The combination as defined in claim 22, wherein said means for causing said ionized atoms to impinge on said second collector plates comprises ∑ (a) means for providing a magnetic field at an angle to, extending along and passing through said optical axis; and (b) means for providing an electric current extending along and including said optical axis. 30. The combination as defined in claim 29, and additionally including means for pulsing said electric current, each pulse having duration less than the mean free time for a charge exchange collision in said beam. 31. The combination as defined in claim 22, wherein said means for causing said ionized atoms to impinge on said second collector plates comprises means for giving said ionized atoms an energy of at least about one electron volt to cause them to expand at a speed greater than the thermal speed of unionized atoms in said beam by about the square root of the ratio of the electron energy to the mean energy in the atoms in said beam whereby said ionized atoms impinge on said second collector plates. 32. The combination as defined in claim 29, and including additional means for providing a magnetic field at the source of said beam at an angle to the direction of flow of said beam. 33· A method of separating isotopes of a material containing a first isotope and a second isotope, substantially as hereinbefore described. 3l .» An apparatus for separating isotopes of a material containing a first isotope and a second isotope, substantially as hereinbefore described with reference to and as illustrated in the accompanying drawings. Pef the Applicants