Patent Application
20200368687
Several of the rare isotopes of light elements have exceedingly high monetary value in the chemical, nuclear, and medical fields. These elements including, but not limited to, 13C, 14C, 15N, 17O, and 18O are generally separated by the conventional processes of chemical exchange or distillation. For example, 15N and 14N have been separated by chemical exchange of NH3 in a mobile phase with NH4+ in the solid phase of a cation exchange resin. In this case, 15N is equilibrium shifted to the solid phase, by a factor of approximately 1.03, such that 15NH3 will elute slower from an adsorption column. Alternately, the isotopes 15N and 14N can be separated by cryogenic distillation of naturally occurring N2. The rarer 15N2 and 15N14N have slightly lower boiling points than the 14N2 such that they can be enriched over a number of column stages, as illustrated for example in U.S. Pat. No. 7,828,939. The primary drawback of each of these methods is the exceedingly low separation factor, between 1.01 and 1.1. These methods are, therefore, typically applied in a multi-stage configuration with tens or hundreds of stages needed to achieve a 99+% pure product.
Another method of light isotope separation is laser excitation and reaction, as illustrated for example in U.S. Pat. No. 4,387,010. Certain molecules have absorption bands, associated with vibrational, electronic, or vibronic excitations, which differ sufficiently between isotopologues such that one can be selectively targeted by a certain wavelength of light. As used here, the term “isotopologues” is defined as molecules which differs only in isotopic composition. Generally, the higher valued, or most desirable, isotopologue is targeted for excitation. The higher value isotopologue is then reacted with another molecule such that the product is isotopically enriched. A conventional means of separation is then used to isolate the product molecule such as adsorption, cryogenic distillation, etc. Methods of this nature have achieved enrichment factors of up to 30 in a single stage. For instance, a method was disclosed which utilized a gas filter of natural abundance NO gas to refine the UV light from an ArF laser cell such that the transmitted portion was selective towards 15NO excitation (U.S. Pat. No. 4,387,010).
The primary excitation sources utilized in the literature examples are tunable lasers. A narrow wavelength laser is tuned to a target wavelength in which the greatest selectivity advantage is gained for the higher-valued isotope over the lower-valued one. The laser light is then passed through a device, such as a flow-cell, which contains the reaction mixture. The drawback of laser excitation sources is a high capital cost for the laser and a low power efficiency. Additionally, tunable lasers are highly precise devices, which must be operated within a very narrow window of control.
In some special cases, a spectral emission source can be used. If a spectral line is known that lies close to the appropriate target wavelength, it can be used. In one instance an iodine-emission tube, emitting 206.2 nm light, was used to preferentially excite the 13C and 18O isotopes of CO and react them to form C3O2, CO2, and C(s) products. A Kr emission source, emitting 123.58 nm light, was also used to preferentially excite the 13C and 18O isotopes of CO and react them to yield C3O2 and CO2 products (U.S. Pat. No. 4,374,010). This design has a huge advantage in that gas discharge emission tubes are low cost and reasonably efficient. Unfortunately, very few emission lines are known which lie close enough to isotopically selective target wavelengths to be generally applied.
In another instance, the spectral emission source gas is actually composed of the enriched isotope component such that it radiates the target wavelength. An example of this was patented by Bergheaud et al, for 196Hg separation (U.S. Pat. No. 3,983,019). Unfortunately, this technique is significantly restricted to elemental or highly stable compounds for use as the emission source.
A number of reactors are described in the literature which utilize emission tubes to drive reactions. U.S. Pat. No. 7,695,675 describes the design of a photoreactor in which the reaction mixture is forced to flow in a helical path around the light source. This both increases the resonance time and maintains a high Reynolds number for the fluid. The patent KR101416067 describes a continuous flow reactor for water treatment in which a round mirror surface is used to reflect the transmitted light. This serves to increase the efficiency of the device. The patent US 2011/0237842A describes an arrangement where multiple emission tubes are arranged normal to the flow path. This ensures that the majority of the fluid stream receives a high intensity of light, even if the fluid is mostly opaque to the light. Some photoreactors, utilizing emission tubes, are modified with conventional filters such that the spectrum of light is modified in a preferential way for the reaction. For instance, EP 0803472A1 includes the use of quartz as an optical filter to preferentially transmit shorter wavelengths of light to a reactor. Finally, photoreactors have been designed in which Light Emitting Diodes (LEDs) are used as the light source (WO 2009/129993A1).
An isotope separation method, including introducing a first reactant stream (109), containing a natural abundance of at least one desired isotopologue molecule, a second reactant stream (110), and a recycle stream (112) into a photochemical reactor (101), thus producing a raw product stream (15), introducing the raw product stream (115) into a separation device (116), thus producing at least a product stream (117), a gas filter stream (113), and the recycle stream (112), and introducing at least a portion of the gas filter stream (113) into an unconventional (gas) filter (103), wherein the product stream (117) includes the at least one desired isotopologue molecule.
An isotope separation method, including introducing a first reactant stream (109), comprising a natural abundance of at least one desired isotopologue molecule, a second reactant stream (110), and a first recycle stream portion (112) into a first photochemical reactor (102), thus producing a raw product stream (115), introducing the raw product stream (115) into a first separation device (116), thus producing at least a crude product stream (117), a first gas filter stream (119), and a first recycle stream (112), introducing at least a portion of the crude product stream (123), and a second recycle stream portion (125) into a second photochemical reactor (127), thus producing an enhanced product stream (128), introducing the enhanced product stream (128) into a second separation device (129), thus producing at least a product stream (130), a second gas filter stream (133), and the second recycle stream (132), combining the first recycle steam (112) and the second recycle stream (132) and separating the combined stream into the first recycle stream portion (124) and the second recycle stream portion (125), wherein the product stream (130) comprises at least one desired isotopologue molecule.
For a further understanding of the nature and objects for the present invention, reference should be made to the following detailed description, taken in conjunction with the accompanying drawings, in which like elements are given the same or analogous reference numbers and wherein:
As schematically represented in the figures, a novel reactor 101 is disclosed for the separation of isotopes. Continuous spectrum light source 105 is passed through conventional filter, solid optical filter 104 to remove bulk undesirable wavelengths B. Light source 105 may be an incandescent lamp, or preferably, a light emitting diode (LED), or a gas discharge lamp. Light source 105 should be chosen for a peak power output near to the appropriate absorption band to drive the reaction. Light source 105 may include a mirrored back-plate 108 to increase the intensity of light directed towards the reactor. Conventional filter 104 may be chosen to target a range of light near to the appropriate absorption band and exclude light with energy of 40% less or 40% greater. For instance, if the target wavelength is in the ultraviolet region at 250 nm, conventional filter 104 would be chosen to exclude greater than 420 and less than 180 nm light; if the target light is in the infrared region at a wavenumber of 3,000 cm−1, conventional filter 104 would be chosen to exclude greater than 1,800 and less than 4,200 cm−1 light. Most crucially, light of higher energy should be excluded as it could result in undesirable side reactions or molecular dissociation.
The light is then passed through unconventional filter 103, herein referred to as a ‘gas filter’ or “optical gas filter”. Gas filter 103 consists of a chamber, with transparent walls, filled with the lower valued isotopologue. First transparent wall 106 defines the region proximate to, inside of, or outside of conventional (solid) filter 104. Second transparent wall 107 defines the inner boundary of unconventional (gas) filter 103, and defines the boundary of reaction chamber 102. With regard to the spectrum of light passing through the chamber D, the absorption spectrum of the low value isotopologue is eliminated and the transmittance spectrum, passed through, is appropriate for selective excitation of the higher valued isotopologue. Gas filter 103 may either be sealed or continuously purged to maintain its purity.
The light D transmitted by the gas filter 103 is passed to reaction chamber 102 where the mixture of isotopologues to be separated 109 plus a secondary reactant 110, forming at least part of combined reactant stream 111, are present. Filtered light selectively excites the higher valued isotopologue molecules present in first reactant stream 109, which then react with secondary reactant stream 110 to form raw product stream 115. Raw product stream 115 is isotopically enriched and can be separated by conventional means 116 from the reaction mixture.
Gas filter 103 function and design can be estimated starting from Beer's law equation 1.
where Aλi is the absorbance of light with wavelength λi, Ii is the intensity of transmitted light with wavelength i, Ii,0 is the intensity of light entering the gas filter, αi,j is the absorption coefficient of the gas species j to the wavelength i, Pj is the partial pressure of species j, and L is the length of the gas filter. Rearranging this equation to solve for the intensity of light transmitted gives the equation 2.
Ij=Ii,0*10−α
The absorption coefficients of each species can be measured experimentally in the region around the relevant wavelength. By equation 2, wavelengths of light which are absorbed by the primary species in the gas filter are absorbed exponentially faster than those absorbed by the scarcer species.
As can be seen in
The calculation shown in
The transmitted light is then passed on to reaction chamber 102. Reaction chamber 102 consists of a continuous flow reactor, having at least one wall 107 which is transparent to the target wavelength of light. Reactant mixture 111 is introduced at one end of reactor 102 and product mixture 115 is withdrawn from the other end of reactor 102. Reactant mixture 111 includes an isotopic mixture, including the target isotopologue and optionally, a secondary species which has some reactivity with that molecule. The transmitted light D selectively interacts with the higher valued isotopologue, promoting it to an excited state. The excited isotopologue then reacts to form a product species at an increased rate. Since the light excitation was selective towards the higher valued isotopologue, reaction stream 111 will be enriched in that product isotopologue and the un-reacted starting material will be depleted.
The enrichment achievable by the device is a function of both the selectivity of the light passed by the gas filter and the resolution of the relevant absorption peaks in the molecules' absorption spectra. The concentration of each excited species [xA] and [yA*] can be estimated by derivation from a steady state mass balance for each species:
where [jA*] is the concentration of excited state of species j, Ij is the intensity of light at wavelength i, h is Plank's constant, vi is the frequency of the light considered, kB is the Boltzmann constant, TSTP is the standard temperature of 273 K, jA0 is the concentration of ground state species j, and rd is the decay rate for the excited state, accounting for all relaxation mechanisms. The summation sign in Equation 3 indicates the necessity to account for all wavelengths of light which can cause the excitation. For instance, a single electronic transition in a diatomic molecule at room temperature is typically composed of more than five measurable vibrational bands, separated by energies on the order of 2,000 cm−1. Each vibrational band then contains a fine structure of over 20 measurable rotational bands, separated by energies on the order of 5 cm−1. The Equation 3 can be simplified by the assumption that only weak pumping occurs ([jA*]<<[jA0]), resulting in Equation 4.
Equation 4 allows for an estimation of the concentration of excited species in the reaction mixture, which can then be used to estimate the overall reaction rate. The Equation 4 can also be used to estimate the light intensity required for a given device productivity. The enrichment factor for two species that is enabled by the device under ideal conditions is then given by equation 5.
E=([yA*]/[xA*])/([yA0]/[xA0]) Equation 5
where yA is the lower concentration isotopologue and xA is the higher concentration one. If Equation 5 is solved by substitution with Equation 4, it can be referred to as the theoretical enrichment factor, which can readily be solved. For two isotopologues having well resolved absorption peaks, the theoretical enrichment enabled by the device as function of the gas filter length is shown in
The theoretical enrichment factor enabled by the device increases exponentially with the selectivity of the light passed by the gas filter, which is a direct function of the gas filter length. The case illustrated by
A number of different reaction systems can be employed in the reactor such that it is necessary to better clarify the reaction system. The photo-excitation reaction is given by reaction I.
n
A+hv→
y
A*+
x
A Reaction I
where nA is a chemical species containing a naturally occurring ratio of isotopologues, hv is a photon of light having the target wavelength, yA* is the excited state of the higher valued isotopologue, and xA represents one or more of the lower valued isotopologues. The excited state species then react to form products by one of several possible routes given by reactions II to V below.
y
A*+B→
y
C Reaction II
y
A*→
y
C+D Reaction III
y
A*+B→
y
C+D Reaction IV
y
A*+B→
y
C+D+
y
E Reaction V
where yC is the product with an enriched ratio of the higher valued isotope, D is one or more side-products, possibly having an altered isotope ratio, B is a secondary reactant species, and yE is one or more additional side-products with an enriched ratio of the higher valued isotope.
The preferred embodiment of the invention is one in which the fewest number of reaction products are formed such that the product stream can be easily separated. As such, the simplest case is given by Reaction II, the excited state molecule A* reacts with a secondary reactant species B to form a single product C. Since yA* was preferentially enriched in the higher valued isotope, the product yC will also be enriched. The Reaction III describes a different scenario; the excited state molecule spontaneously splits into two or more products, one of which is enriched in the higher valued isotope. Finally, more complex reaction chemistries are given by reactions IV and V. In these two systems, multiple reactants react to form multiple products, with one or more products enriched in the high valued isotope.
The reactor effluent stream is different in each case, containing three or more molecular species, the unreacted, value-depleted xA, the unreacted B, the value-enriched yC, one or more side-products D, possibly having an altered isotopic ratios, and one or more side-products yE having altered isotopic ratios. A conventional separation process can be employed to separate the product mixture, with the particular choice of technology based on the chemical properties of each species. One or more of the isolated products will be isotopically enriched. The unreacted starting material will be isotopically depleted.
Another possible side-reaction which must be considered in the reactor is Reaction VI, which accounts for excited state transfer between isotopologue molecules.
y
A*+
x
A→
y
A+
x
A* Reaction VI
Since the excitation energies of yA and xA are very nearly identical, the excited state energy can be readily transferred between the two species by a collision interaction. The Reaction VI provides a path by which the low valued isotopologue can react to form products such that it is termed ‘non-selective’. If the rate of Reaction VI is greater than the rate of product formation (for Reactions II, III, IV, or V), then the reactor products will have a low enrichment factor. The rate of Reaction VI can be decreased and the rate of product forming reactions increased by utilizing an excess of secondary reactant B such that A is dilute.
In order to maximize the use of light of the appropriate wavelength, a reactor with gas flow parallel to the direction of light can be envisioned such as shown in
The glass tube reactor arrangement can also be realized with extremely low diameter tubes, such as glass capillaries (hollow fibers), or so-called photonic crystals, which are optical fiber with an empty core surrounded by a plurality of empty channels which have the effect of forcing the light to propagate close to the core of the fiber. Such photonic crystals are manufactured for example by NKT Photonics or Photonics Bretagne.
Such a design as shown in
A preferable reactor geometry is given in
There are a few advantages of this radial reactor configuration. Since only a small fraction, perhaps less than 1%, of the emitted light is of the correct wavelength for selective excitation, a very high initial luminous output is needed. If a single source with sufficient power output is not available, multiple emission sources arrayed around the reactor can achieve the same effect. To allow for high throughput, higher space velocities of gas can be used. This is most efficiently contained in a small tube (or bundle of smaller tubes).
If gas filter 103 and reaction chamber 102 are designed around a selective set of absorption bands which have a strong adsorption coefficient, on the order of 5 cm−1, and a pressure in excess of 1 bara can be used, only a very short path length of less than 1 cm is necessary for the gas filter to achieve high selectivity. If these operating parameters are chosen, then a cylindrical reactor geometry is preferable. On the other hand, if the relevant absorption coefficients are weak, or a low operating pressure is chosen, then a cylindrical arrangement such as shown in
The specific method of using photochemical reactor 101 depends both on the specific type of reaction system being considered and on the optimal operating conditions: such as, but not limited to, pressure, temperature, and stoichiometry, for high isotopic selectivity.
In general, a process scheme similar to that shown in
n
A+hv→
y
A*+
x
A Reaction I
y
A*+B→
y
C+D Reaction IV
Stream 109, a mixture nA, consisting of the molecule A with a natural abundance of isotopologues, is mixed with stream 110, consisting of component B, and sent to the reactor chamber 102. In reactor chamber 102 the higher valued isotopologue yA is selectively excited by light from the gas filter by reaction I, then reacts with component B by reaction IV. The reactor generates the ‘product mixture’, raw product stream 115 composed of xA, B, yC, and D.
Raw product stream 115 may be separated by a variety of conventional separation processes, including one or more of the following: cryogenic distillation, pressure or vacuum swing adsorption, solvent absorption, or membrane diffusion. The particular type of separation is chosen giving priority to the higher valued products, typically, but not necessarily, the rare isotope enriched yC species. It may be the case that the rare isotope has a much higher value than the natural abundance starting material nA. In this case, the conventional separation should be designed for high recovery and/or high purity of the yC species. Alternately, it may be the case that either of the species, xA or B, have a moderate value and are present in significant excess with respect to the yC species. In this case, the conventional separation could be configured for high purity isolation of either species. Since the high valued isotopologue(s) are generally less than 1.5% in concentration versus the low valued isotopologue, a large volume of the low valued one will likely be present. To improve the economics of the process, stream 120, the isotopically depleted starting material xA, may be sold for conventional chemical use. Finally, if the species B is used in significant excess, recycle stream 112 may be returned to the reactor inlet. This may significantly reduce the required flowrate of stream 110, the make-up feed rate of B to the overall process.
Once raw product steam 115 has been separated, gas filter stream 113, comprising a portion of the low valued isotope stream, xA, may be recycled back to the reactor 101 to purge gas filter 103. Depending upon the light source and specific chemicals utilized, a side reaction is possible as given by reaction VII.
A*→D Reaction VII
where A is the excited state of either isotopologue and D is a degradation product with no isotopic enrichment. This non-selective reaction would quickly degrade the quality of the gas filter such that a continuous purge would be needed. The xA stream, which will be high volume, is ideally suited for this purpose. Separation device 116 may be configured to also remove first side product stream 118, consisting largely of degradation product D. Side-product(s), D, may have some value for conventional chemical use
Product stream 117, containing primarily the isotopically enriched product yC may be sold directly after purification. Alternatively, product stream 117 may be reacted by a variety of conventional processes to form other isotopically enriched species for sale (not shown).
In a preferred embodiment, the reactor 101 is operated at an elevated pressure, but not so high that collisional broadening reduces the selectivity of the device. The preferred pressure is on the order of 5 bara. At this pressure, the rotational fine structure of the absorbance spectrum is well resolved. Two different isotopologues necessarily have different rotational constants such that the spacing between rotational bands will result in high selectivity by the gas filter. At pressures exceeding 5 bara, collisional broadening results in overlap of the tails of each absorption peak, with a shape characteristically known as the Voigt profile.
Conversely, there is no preferable temperature at which the reactor is expected to operate better. Ideally, a temperature should be selected for the highest rate of the product forming Reaction II, III, IV, or V, but without significantly increased rate of the non-selective excitation transfer reaction VI or the degradation reaction VII. This should be decided on a case-by-case basis.
Accounting for some overlap in the absorbance spectra and the occurrence of the non-selective Reaction VI, an enrichment factor of between 2 and 30 is expected for the photochemical reactor. If the natural abundance of yA and xA are 0.4% and 99.6%, respectively, the product mix will contain an isotope composition of 0.8% to 12% of yC. It may be the case that the isotopically enriched product is sold directly. It is more likely the case that a higher level of enrichment is sought after such that it is preferable to reprocess the reaction product and send it to a second photochemical reactor stage. A multi-stage arrangement is proposed where the second, third, etc. stage reactor tubes receive light from the same source and through the same gas filter. An alternate reactor geometry is shown with regard to the radial design in
In the
Turning to
y
A+hv→
2y
A*+A Reaction I-2
2y
A*+B→
2y
C+D Reaction IV-2
In this multi-stage process, first stage reaction chamber 102 and first separation device 116 operate in the same way as that described above. In
Enhanced products mixture stream 128 is then sent to second separation device 129 which separates out product stream 130, including primarily the rare isotope enriched yC species, and side product stream including second degradation product D′. The use of a single gas filter simplifies the process piping and handling of the depleted xA stream.
The design of the reactor geometry can be modified in a number of other ways to improve its operation. For instance, the reactor tube or tubes can assume a helical shape similar to that disclosed in US2003049809A (not shown). The advantage of this arrangement is a higher Reynolds number of the fluid and therefore better mixing at the same space velocity. Alternately, it may be that a catalyst is known which accelerates the product forming reaction II, III, IV, or V. If the catalyst is transparent to the target wavelength, it could be coated 137 on the interior surface of reaction chamber 102. Alternately, the catalyst could be coated on a coiled wire and inserted into the reactor tube (not shown). Alternately, a multi-tube reactor could be used with the catalyst coated preferentially on one side 137 of each tube as shown in the multi-tube reactor as indicated in
An additional variation of the reactor design considers the use of two gas cell in series as shown in
First gas filter 138 may consist of an elemental or very stable compound which has a broad absorption region. For instance, CO2 has a broad, dissociative absorption band in the range of 140 to 165 nm. Water vapor has an absorption band in the range of 120 to 185 nm. Second gas filter 103 is then composed of the low valued isotopologue. Alternately, both the first gas filter 138 and second gas filter 103 may be swept with the same low valued isotopologue species. First filter 138 then acts as a sacrificial filter, from which a higher concentration of degradation products is produced. Primary gas filter 103 would then receive a lower intensity of light and produce fewer degradation products. The two filters could be operated at different pressures such that the first, at lower pressure, accumulates significant degradation impurities and is discarded afterwards. The second filter, at higher pressure, accumulates less impurities such that it can be sold to offset some operating expense.
As indicated in
A cooling system (not shown) may be required to regulate the device temperature. This may be different for various embodiments, depending on the intensity of the light sources, the heat transfer resistance of the various materials, and the overall geometry. A photoreactor with incorporated cooling and of similar geometry is disclosed in WO2016026576. Tubing with heat transfer media could be incorporated in the center of the reactor with a similar geometry as shown in
The operating methods for the photochemical reactor and process will be specific for each particular reaction system and are expected to conform to conventional chemical engineering design standards. Some operational aspects that are expected to be unique to the photochemical reactor are described herein.
Regarding the operation of the photochemical reactor process, isotopically depleted material xA may not be readily available in quantities sufficient to charge the gas filter for the initial start-up. In this case, the material A of natural isotopic abundance can be used until the separated products of the reactor effluent stream are available. Until the depleted xA is available, the gas filter will be less efficient at transmitting the target wavelength such that the reactor should be operated at a lower throughput.
Once the reactor is sized and built to a fixed dimension, the strongest lever by which the gas filter operation can be altered is the pressure. By increasing the gas filter pressure, the selectivity of the filter can be increased at the expense of the device productivity. The reaction selectivity can be easily monitored with conventional gas chromatography coupled with a mass spectrometer.
If the non-selective excited state transfer Reaction VI is known to be problematic, its rate can be decreased and the rate of product forming reactions increased by utilizing an excess of secondary reactant B such that A is dilute. Increasing the excess quantity of reactant B will lower the effective throughput of the device. The level of excess reactant B can be varied such that an optimal balance of high enrichment factor versus high device throughput is achieved.
Finally, the luminous intensity of the light source can also be altered to increase or decrease the reactor conversion at a given throughput. The light source should be operated at the greatest possible intensity for high throughput, but not so high that the photon flux approaches the decay rate of the target molecule, i.e. the assumption made in simplifying equation 3 to equation 4, that [jA*]<<[jA0]. This limit can easily be determined on a case by case basis.
Sentence 1. A photochemical reactor comprising:
Sentence 2. The reactor of sentence 1, wherein the at least one reaction chamber (102) is configured to react at least a first component (109) and a second component (110), wherein the first component (109) comprises a preferred isotopologue of a first compound.
Sentence 3. The reactor of sentence 1, wherein the light source (105) is selected from the group consisting of an incandescent lamp, a light emitting diode, or a gas discharge lamp.
Sentence 4. The reactor of sentence 1, wherein the solid optical filter (104) excludes light with energy either 40% greater than or 40% less than a predetermined adsorption band.
Sentence 5. The reactor of sentence 1, wherein the solid optical filter (104) consists of quartz, Pyrex, Vycor, or sodium chloride.
Sentence 6. The reactor of sentence 1, wherein the optical gas filter (103) comprises a gas media comprising an unwanted isotopologue of the first compound.
Sentence 7. The reactor of sentence 6, wherein the photochemical reactor is configured such that light emanating from the light source (105) first passes through the solid optical filter (104), then through the optical gas filter (103), resulting in light having a target wavelength bandwidth.
Sentence 8. The reactor of sentence 7, wherein the reaction chamber (102) comprises a curved surface at least a portion of which is essentially transparent to the target wavelength bandwidth.
Sentence 9. The reactor of sentence 7, wherein:
Sentence 10. The reactor of sentence 7, comprising multiple light sources, wherein:
Sentence 11. The reactor of sentence 1, wherein the reaction chamber (102) comprises a continuous flow reactor.
Sentence 12. The reactor of sentence 1, wherein the reaction chamber (102) comprises an interior surface that is at least partially covered with a catalyst (137),
Sentence 13. The reactor of sentence 1, wherein the reaction chamber (102) comprises an interior volume, wherein a coiled wire coated with catalyst (137) is inserted into the interior volume.
