Removal of technetium impurities from uranium hexafluoride

FIELD OF THE INVENTION
The present invention relates to methods of purifying uranium hexafluoride by removing technetium-99 impurities.
BACKGROUND OF THE INVENTION
Technetium-99 (.sup.99 Tc) is a contaminant that is typically present in an enriched UF.sub.6 product in low concentrations. This contaminant originates from the fission of .sup.235 uranium and is contained in reactor return uranium. Process equipment surfaces in gaseous diffusion plants which process UF.sub.6 hold the .sup.99 Tc as one or more volatile compounds which are slowly released over time. As a result, the enriched UF.sub.6 product which is withdrawn from diffusion plants invariably contains low concentrations of .sup.99 Tc, due to a slow leaching from equipment surfaces. The concentrations of the .sup.99 Tc impurity, although low, may easily exceed product specification limits, making the uranium hexafluoride product unacceptable to fuel fabricators. Presently, the specification limit for .sup.99 Tc in UF.sub.6 product is only 0.2 .mu.g/g .sup.235 U or, assuming 5% .sup.235 U enrichment, 0.010 .mu.g .sup.99 Tc/g U (0.010 ppm, U basis).
Methods for the removal of .sup.99 Tc from UF.sub.6 have typically involved gas phase operation (contacting gaseous UF.sub.6 with a metal fluoride, typically magnesium fluoride adsorbent). These methods are not effective at the low .sup.99 Tc concentrations which impact customer acceptance. For example, the lowest concentrations amenable to gas phase removal are above 0.1 .mu.g .sup.99 Tc/g U. Additionally, sufficient throughput must be maintained to provide adequate quantities of the purified UF.sub.6. The throughput for UF.sub.6 processing which can be obtained with gas phase operations is only about 100-500 lbs/ft.sup.2 /hour. If the gas velocity is increased to increase UF.sub.6 processing rates, .sup.99 Tc removal efficiency decreases sharply. Conversely, if gas velocity is decreased to maintain high .sup.99 Tc removal, processing rates suffer. In short, existing methods (gaseous UF.sub.6 with MgF.sub.2 adsorbent) are insufficient to combine both high removal efficiency of .sup.99 Tc for direct control at concentrations applicable to market acceptance, and high UF.sub.6 processing rates to produce an economic process without burdensome equipment size.
What is needed in the art are new methods of removing .sup.99 Tc from uranium hexafluoride which overcome the problems associated with existing methods. The present invention provides such processes.
SUMMARY OF THE INVENTION
The present invention provides processes for the removal of technetium from contaminated uranium hexafluoride containing technetium, typically technetium-99 which is present in several chemical forms or compounds. Some common volatile forms are pertechnetyl fluoride (TcO.sub.3 F), technetium hexafluoride (TcF.sub.6) and technetium oxytetrafluoride (TcOF.sub.4). Less volatile forms of technetium include TcO.sub.2 and TcF.sub.4 or TcF.sub.5. Volatility is a relative property and depends upon the temperature of the process. It is generally the .sup.99 Tc compounds which are volatile at UF.sub.6 handling temperatures that are present in the UF.sub.6 product.
The processes of the present invention involve:
(a) contacting contaminated uranium hexafluoride in liquid form with a solid metal fluoride, typically magnesium fluoride (MgF.sub.2) for a period of time sufficient for the technetium to become adsorbed onto the metal fluoride solid thereby producing a purified uranium hexafluoride liquid; and
(b) removing the purified uranium hexafluoride liquid from the solid metal fluoride, typically magnesium fluoride solid having adsorbed technetium.

BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 illustrates a UF.sub.6 cylinder "1S" which is used in the Examples.
FIG. 2 shows a liquid UF.sub.6 filtering apparatus containing 8-12 mesh MgF.sub.2 as an adsorbent and a 10 micron filter.
FIG. 3 illustrates a vacuum manifold system for filtering liquid UF.sub.6 and trapping the purified UF.sub.6 in a cooled collection vessel.

DETAILED DESCRIPTION OF THE INVENTION
This invention disclosure describes a process modification which leads to unexpectedly large improvement in the removal of .sup.99 Tc from UF.sub.6. The process reduces .sup.99 Tc in UF.sub.6 to below product specification levels for UF.sub.6 and further provides high UF.sub.6 throughputs. The methods described herein can be applied to control the removal of .sup.99 Tc at UF.sub.6 product withdrawal stations and/or UF.sub.6 liquid transfer facilities on a scale that is attractive for installation at uranium enrichment plants.
As noted in the Background, existing technology employs magnesium fluoride (MgF.sub.2) in the form of pellets which are placed in a reactor. Uranium hexafluoride is passed through the reactor in the gas phase at atmospheric pressures or less. Gas phase operation in itself is inherently limiting in UF.sub.6 throughputs or processing rates. Furthermore, this mode of operation fails to perform efficiently as .sup.99 Tc concentrations decrease. The lower concentration limit for useful removal and/or control of .sup.99 Tc varies with gas velocity and is estimated to be above 0.1 .mu.g/g U and probably much higher. (A typical result in gas phase operation at 2 psia is 2.2 .mu.g/g being reduced by 59% to 0.9 .mu.g/g). It is not possible to obtain UF.sub.6 processing rates above about 500 lbs/ft.sup.2 /hour at atmospheric pressure or below and still achieve acceptable reduction at concentration below 0.1 .mu.g/g U.
The present invention is based on the surprising discovery that when liquid UF.sub.6 (rather than gaseous UF.sub.6) is passed through a trap containing solid MgF.sub.2, a remarkable boost in .sup.99 Tc removal efficiency is realized at low .sup.99 Tc concentrations. As one example, at an initial concentration 0.018 .mu.g/g U, .sup.99 Tc is reduced to 0.0008 .mu.g/g U. In other demonstration tests, reductions to below the detection limit (0.0004 .mu.g/g U) are readily obtained. These high efficiencies were not expected considering the state of the art for MgF.sub.2 trapping technology. Furthermore, UF.sub.6 processing rates are well over 3000 lbs/ft.sup.2 /hour. Rates as high as 4600 lbs/fts.sup.2 /hour have been successfully demonstrated with no apparent decrease in efficiency. These rates have considerable economic impact as it is now feasible to install a reasonably sized .sup.99 Tc removal apparatus at any existing product withdrawal station or UF.sub.6 liquid transfer facility without significantly interfering with present operations. Accordingly, direct control of .sup.99 Tc in UF.sub.6 product is possible and results in further cost savings due to greater flexibility of diffusion operations. The highly purified UF.sub.6 product which is obtained using the methods herein, further assures customer acceptance of the UF.sub.6 product.
EMBODIMENT OF THE INVENTION
In view of the above surprising discovery, the present invention provides in one aspect a process for removal of technetium (.sup.99 Tc), from contaminated uranium hexafluoride containing the technetium. This process comprises:
(a) contacting the contaminated uranium hexafluoride in liquid form with a metal fluoride in solid form for a period of time sufficient for the technetium to become adsorbed onto the metal fluoride thereby producing a purified uranium hexafluoride liquid; and
(b) removing the purified uranium hexafluoride liquid from the solid metal fluoride having adsorbed technetium.
The present invention utilizes UF.sub.6 in liquid form to achieve the significant results provided in the Examples below. Uranium hexafluoride is a volatile white crystalline solid which melts at about 64.degree.-65.degree. C., but which sublimes at about 56.degree.-57.degree. C. under pressure of about 1 atmosphere. Accordingly, liquified uranium hexafluoride can be obtained under pressure of about 1.5 atmospheres and at working temperatures (typically provided in cylinders). Uranium hexafluoride also reacts vigorously with water and care should be taken to remove air and moisture from the trapping system.
Metal fluorides which are suitable for the removal of technetium from UF.sub.6 are generally high melting ionic solids with very low solubility in UF.sub.6. A particularly preferred metal fluoride is magnesium fluoride (MgF.sub.2) which is recognized for its technetium removal properties. However, other metal fluorides can also be used, such as AlF.sub.3, transition metal fluorides such as NiF.sub.2 and related high melting fluorides which are solids at temperatures at which UF.sub.6 begins to melt to a liquid. In other embodiments, the metal fluoride used in the invention can be a combination of different metal fluorides (e.g.,. a combination of MgF.sub.2 and AlF.sub.3).
The method of contacting the contaminated uranium hexafluoride with a metal fluoride typically involves placing a metal fluoride chemical trap in the UF.sub.6 flow line extending from the source of UF.sub.6 to a suitable receptacle for purified UF.sub.6. FIGS. 1-3 provide illustrations of the types of apparatus which can be used in the present invention. One of skill in the art will understand that other apparatuses could also be used to provide the necessary contact between liquid UF.sub.6 and the metal fluoride. By allowing the contaminated uranium hexafluoride to pass through, for example, a MgF.sub.2 chemical trap, contact is made between the MgF.sub.2 and the contaminated UF.sub.6, and technetium impurities are adsorbed onto the MgF.sub.2. Particular flow rates and contact times (between liquid UF.sub.6 and MgF.sub.2) will be dependent on the system requirements and capabilities. The examples below provide an indication of flow rates and contact times (or residence times) for a system comprising a "2S" or "1S" feed cylinder of UF.sub.6, a MgF.sub.2 chemical trap constructed from a "1S" cylinder body, and an attached "2S" cold trap. For this system, a flow rate of about 1.0.times.10.sup.-5 ft.sup.3 /sec to about 3.5.times.10.sup.-5 ft.sup.3 /sec is useful. These flow rates result in chemical trap residence (or contact) times of about 350 seconds to about 100 seconds. Other filters and chemical traps which operate on this same premise, including those versions which are of a larger or smaller scale and of greater or lesser trap voidage, will be known to those of skill in the art.
The amount and physical form (specific surface area and pellet size) of the metal fluoride used in the chemical traps will be dependent on a number of factors including the amount of UF.sub.6 to be purified, the size of the trap and the flow rate required. In one group of embodiments, the metal fluoride is MgF.sub.2 which is in the form of pellets, preferably about 1/8 inch to about 3/8 inch pellets, more preferably about 1/4 inch pellets. In other embodiments, the magnesium fluoride is in the form of 6-15 mesh, preferably 8-12 mesh. The amount of magnesium fluoride used will depend on the level of contamination of UF.sub.6 with .sup.99 Tc. Typically, magnesium fluoride will adsorb up to about 0.03 g .sup.99 Tc per g MgF.sub.2. Periodic monitoring of the chemical trap contents will determine whether the chemical trapping agent, MgF.sub.2, is in need of replacement.
Following contacting the contaminated UF.sub.6 with the metal fluoride chemical trap, the resultant purified liquid UF.sub.6 is removed from the trap. Typically, in a flow-through system, the removal occurs as a matter of course as the liquid UF.sub.6 (first contaminated, then purified) is pumped or transferred through the system. Accordingly, attached to the effluent port of the chemical trap is a cold trap for containing the purified UF.sub.6. The cold trap is typically a cylinder for longer term storage of the purified UF.sub.6 which is immersed in a cold temperature bath, for example a liquid nitrogen bath (-196.degree. C.). In a flow-through system the direction of liquid flow may be vertically upward or downward. Alternative modes of liquid-solid contact are also useful, including co-current, countercurrent and unicurrent flow of the liquid and solid phases. The unicurrent flow implies one of the phases is fixed, while the other is mobile. Both phases can be temporarily fixed (no flow) and either one of the two phases is then removed at a later time.
The following examples are offered solely for the purposes of illustration, and are intended neither to limit nor to define the invention.
EXAMPLES
A set of tests was designed to test the removal of .sup.99 Tc from liquid UF.sub.6 utilizing a MgF.sub.2 chemical trap. The tests consisted of transferring liquid UF.sub.6, containing a high concentration of .sup.99 Tc through a chemical trap and cold trapping the outlet of the trap. The trap contents were varied and a duplicate trap was operated for comparison. The outlet cold traps were subsampled and analyzed for .sup.99 Tc and the results evaluated for .sup.99 Tc removal efficiencies.
Trap Construction
Two identical MgF.sub.2 chemical traps were constructed using "1S" cylinder bodies (see FIG. 1). Both ends were drilled and a 10 micron filter (see FIG. 2) was silver soldered into the outlet of the traps. A cajon fitting was silver soldered to the inlet, which allowed access to the trap interior to facilitate trapping material changeouts. The 10 micron filter kept trapping material from contaminating the UF.sub.6. The traps were pressure checked at 200 psig, leak checked and passivated with fluorine. The traps were wrapped with a heat tape and connected to a Variac to control temperature at .about.100.degree. C.
Test Procedure
Testing was conducted using a vacuum manifold system typically used for liquid UF.sub.6 subsampling. The trap was connected to the cajon fitting used for subsampling UF.sub.6 cylinders. Tubing extended from the outlet end of the trap to a "2S" container immersed in a liquid nitrogen bath. The liquid UF.sub.6 was supplied in "2S" containers. The following steps were used in each test conducted.
1. Liquified "2S" feed cylinder.
2. Attached cold trap "2S" to manifold
3. Attached liquified "2S" feed cylinder to manifold
4. Pressure and leak checked apparatus
5. Applied liquid nitrogen to cold trap
6. Initiated liquid UF.sub.6 flow through filter
7. Maintained established pressure limits
8. Recorded transfer time and pressures
9. Subsampled filtered UF.sub.6
10. Analyzed filtered (purified) UF.sub.6 for .sup.99 Tc.
Control--Filter Test
The control test is a preliminary filter test in which no MgF.sub.2 trapping media is used. The test is used to determine the amounts of .sup.99 Tc which can be removed on a 10 micron filter in the absence of any trapping media. The test was conducted on traps 1 and 2 with the feed cylinder heel. The traps were washed and the liquid was sent for .sup.99 Tc analysis.
______________________________________PRELIMINARY TEST RESULTS______________________________________ Tc Inlet* Tc Outlet* % TcTrap No. Concentrations Concentrations Removal______________________________________Trap #1 0.488 0.429 12.1Trap #2 0.560 0.510 8.9______________________________________ *.mu.g Tc/g .sup.235 U
Wash SolutionsTrap No. Feed "2S"* Wash Filter* Wash______________________________________Trap #1 0.966 1.266Trap #2 1.56 0.741______________________________________ *.mu.g Tc Total
As can be seen from the above data, the 10 micron filters were not effective in trapping .sup.99 Tc from liquid UF.sub.6. The average percent of .sup.99 Tc removal from the liquid UF.sub.6 was about 10 percent with equal amounts of .sup.99 Tc found in the filters and the feed cylinders. This indicated no selective removal of .sup.99 Tc by the filter in the empty trap.
EXAMPLE 1
This example illustrates the effectiveness of a MgF.sub.2 trapping agent in removing .sup.99 Tc from liquid uranium hexafluoride.
Test Number 1
The chemical traps were cleaned and filled with 1/4" MgF.sub.2 pellets that were manufactured at the Portsmouth Gaseous Diffusion Plant in Piketon, Ohio. The filters were evacuated to a vacuum with heat applied, then treated with fluorine several times to remove excess water.
The inlet UF.sub.6 pressure was throttled at 30 psig.
______________________________________RESULTS - TEST NUMBER 1MgF.sub.2 FILTER TEST #1 TransferTrap Wt. Tc Inlet* Tc Outlet* .mu.g Tc Tc %No. Gms. Concentrations Concentrations Removed Removal______________________________________Trap 1494 0.429 0.0285 20.1 93.5#1Trap 1729 0.51 0.219 14.2 57.5#2______________________________________ *.mu.g Tc/g .sup.235 U
As the above data indicates, the traps removed .sup.99 Tc in much greater amounts than the controls. The amounts removed in Traps #1 and #2 differed significantly, however. The inequality may have been due to the difference in weight of MgF.sub.2 in the traps.
EXAMPLE 2
This example illustrates the impact of unequal weights of MgF.sub.2 on the ability of the traps to sequester .sup.99 Tc. In view of the results in Example 1, this test utilized equal amounts of MgF.sub.2 in the two traps.
Test Number 2
Two traps were loaded equally with 40 grams of 1/4" MgF.sub.2 pellets, then heated, evacuated, and fluorinated as described above. A pressure gauge was added to the outlet line of the traps. The pressure differentials were recorded to indicate any pressure drop occurring through the MgF.sub.2 trap. The flow rate of the liquid UF.sub.6 was estimated to be 1.62.times.10.sup.-5 ft.sup.3 /sec and the residence time was about 306 seconds.
______________________________________RESULTS - TEST NUMBER 2MgF.sub.2 FILTER TEST #2 TransferTrap Wt. Tc Inlet* Tc Outlet* .mu.g Tc Tc %No. Gms. Concentrations Concentrations Removed Removal______________________________________Trap 2036 0.437 0.078 20.5 82.3#1Trap 1741 0.440 0.198 11.7 55.2#2______________________________________ *.mu.g Tc/g .sup.235 U
The average flow rate of the liquid UF.sub.6 was estimated to be 3.32.times.10.sup.-5 ft.sup.3 /sec and the residence time was 150 seconds. The traps again behaved differently with resultant Tc removal very similar to the results of Test Number 1. The trap material was removed and the interior of the trap inspected by a video probe to verify that there was no difference in the trap construction. The inequality in Tc removal efficiency may be the result of the 1/4" pellets allowing channelling through the 11/2" diameter trap body. A subsequent test was established to determine if a smaller mesh MgF.sub.2 would provide more consistent results.
EXAMPLE 3
This example illustrates the use of 8-12 mesh MgF.sub.2 for the removal of .sup.99 Tc from liquid UF.sub.6.
Test Number 3
In this test the two traps were each loaded with 106 grams of 8-12 mesh MgF.sub.2. The traps were evacuated under heat and fluorinated as described above.
______________________________________RESULTS - TEST NUMBER 3MgF2 FILTER TEST #3Smaller Pellets 8-12 MeshTrap Weight Tc Inlet* Tc Outlet* .mu.g Tc Tc %No. (grams) Concentrations Concentrations Removed Removal______________________________________Trap 1816 0.379 0.0286 18.4 92.6#1Trap 1648 0.3906 0.0174 19.6 95.6#2______________________________________ *.mu.g Tc/g .sup.235 U
The increased surface area and bed density of the small mesh MgF.sub.2 produced consistent .sup.99 Tc trapping efficiency of greater than 90 percent. The flow rates of the liquid UF.sub.6 through the traps was estimated at 3.1.times.10.sup.-5 ft.sup.3 /sec and did not seem to be affected by the smaller mesh size of the MgF.sub.2.
EXAMPLE 4
This example illustrates the trapping consistency of the system in Example 3, through several cylinders of UF.sub.6.
Test Number 4
Testing was continued by running several "2S" containers of UF.sub.6 through trap #2 to determine consistency of .sup.99 Tc removal efficiency. It must be noted that the trap experienced a wet-air inleakage following cooldown after Test Number 3 (above). The gasket was changed and Test Number 4 was conducted.
______________________________________RESULTS - TEST NUMBER 4MgF.sub.2 FILTER TEST #4 Tc Inlet* Tc Outlets*Cylinder Transfer Concen- Concen- .mu.g Tc Tc %Number Wt. Gms. trations trations Removed Removal______________________________________2 1690 0.219 0.039 8.61 82.53 1687 0.198 0.029 8.1 85.54 1607 0.0174 0.035 -0.85 No Removal______________________________________ *.mu.g Tc/g .sup.235 U
More than 80 percent of the .sup.99 Tc was removed in two of the test cylinders. However, this amount is 10% less than the results seen in Test Number 3. The flow rate of the liquid UF.sub.6, was estimated at 3.65.times.10.sup.-5 ft.sup.3 /sec and the residence time was 136 seconds. The wet-air inleakage may have affected the results. More UF.sub.6 may have hydrolyzed on water sites of the MgF.sub.2 pellets taking up sites for .sup.99 Tc trapping to occur. Cylinder 4 was tested to see how the MgF.sub.2 would remove even low level .sup.99 Tc from liquid UF.sub.6. The .sup.99 Tc outlet concentrations were the same as previous tests which was apparently due to release of material from previous tests. However, the results could indicate sampling and detection errors at such low levels of technetium.
EXAMPLE 5
This example illustrates the removal of .sup.99 Tc from four cylinders of UF.sub.6 containing different concentrations of the contaminant.
Test Number 5
Test Number 5 consisted of running four more cylinders of UF.sub.6 through trap #2. The gaskets were changed as a precaution. The trap was evacuated after UF.sub.6 from each cylinder was passed through the trap.
Four cylinders containing several different concentrations of .sup.99 Tc were used for this test.
______________________________________RESULTS - TEST NUMBER 5FILTER 2 Tc Inlet* Tc Outlets*Cylinder Transfer Concen- Concen- .mu.g Tc Tc %Number Wt. Gms. trations trations Removed Removal______________________________________5 1568 0.465 <0.02 9.6 95.86 1469 0.408 0.016 9.7 96.17 1909 0.19 <0.008 11.8 95.88 301 1.40 0.021 5.7 98.6______________________________________ *.mu.g Tc/g .sup.235 U
Trap #2 removed 96% of the .sup.99 Tc from the four cylinders of UF.sub.6. The flow rate of the liquid UF.sub.6 was estimated at 3.38.times.10.sup.-5 ft.sup.3 /sec and the residence time was 146 seconds. The trapping efficiency discrepancies seen in earlier tests were probably a result of the wet-air inleakage experienced when the filter cooled and the aluminum gasket loosened. As a precaution the gaskets were changed and the trap was immediately evacuated and buffered with nitrogen for the remaining tests.
EXAMPLE 6
This example illustrates the efficiency of the present method using twelve cylinders of contaminated UF.sub.6.
Test Number 6
Twelve "2S" cylinders containing 1.303 .mu.g Tc/g .sup.235 U were trapped consecutively in this test. The trap was evacuated after each cylinder was transferred. Trap #2 was used with the same trapping media as used in the previous tests.
______________________________________RESULTS - TEST NUMBER 6Cylinder Transfer Inlet .mu.g Outlet .mu.g .mu.g .sup.99 Tc PercentNumber Wt. grams Tc/g .sup.235 U Tc/g .sup.235 U Removed Removal______________________________________ 9 1213 1.303 0.011 47.2 99.210 1449 1.303 0.011 56.3 99.211 1187 1.303 0.015 46.0 98.912 1248 1.303 0.013 48.4 99.013 1132 1.303 0.018 43.8 98.614 2181 1.303 0.009 85.0 99.315 1195 1.303 0.020 46.1 98.516 1352 1.303 0.009 52.7 99.317 1199 1.303 0.009 46.7 99.318 1160 1.303 0.025 44.6 98.119 1799 1.303 0.031 68.8 97.620 1542 1.303 0.016 59.7 98.8 16,657 645.3______________________________________
The 12 test cylinders used in this test had .apprxeq.99% of the .sup.99 Tc removed. The results were consistent at the 98-99% removal efficiency for the total 16,659 grams of UF.sub.6. The flow rate of the liquid UF.sub.6 was estimated to be 4.13.times.10.sup.-5 ft.sup.3 /sec and the residence time was 120 seconds.
The total .sup.99 Tc contained in chemical trap #2, after a total of 20 "2S" cylinders were filtered, was theoretically 718 .mu.g Tc. The high percentage removal of Tc indicates that saturation was not approached. These results further indicate that UF.sub.6 contact with MgF.sub.2 is more important than flow rates and residence time within the boundaries of our tests. This series of tests has demonstrated that with the 8-12 mesh MgF.sub.2 and a trap bed of 106 grams in the 7.times.11/2 inch chambers in excess of 28 Kg of UF.sub.6, were successfully stripped of .sup.99 Tc.
All publications, patents and patent applications mentioned in this specification are herein incorporated by reference into the specification to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated herein by reference.
Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be obvious that certain changes and modifications may be practiced within the scope of the appended claims.

Number	Name	Date
3806576	Carles et al.	Apr 1974
3848048	Moore	Nov 1974
5613186	Larson et al.	Mar 1997

Removal of technetium impurities from uranium hexafluoride

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

Government Interests

US Referenced Citations (3)

Non-Patent Literature Citations (2)

Entry
Milford, "Engineering Design of Oak Ridge Fluoride Volatility Pilot Plant" Ind. & Eng. Chem., vol. 50, No. 2, pp. 187-191, Feb. 1958.
Smiley et al, "Removal of Impurities from Uranium Hexafluoride by Selective Sorption Techniques", Trans. of Am. Nucl. Soc., vol. 10, No. 2, p. 507, Nov. 1967.