WO2001065570A2

WO2001065570A2 - Treatment for waste containing technetium

Info

Publication number: WO2001065570A2
Application number: PCT/GB2001/000845
Authority: WO
Inventors: Martin David Brownridge; Stephen A. Sharpe; Michael Edwin Dunbabin
Original assignee: British Nuclear Fuels PLC
Current assignee: Sellafield Ltd
Priority date: 2000-03-01
Filing date: 2001-02-28
Publication date: 2001-09-07
Anticipated expiration: 2002-09-01
Also published as: GB0004751D0; AU3579101A; WO2001065570A3

Abstract

There is described a process for the removal of soluble Tc from effluent which comprises: i reducing the soluble technetium by the addition of dithionite; and ii removing the insoluble technetium by a separation process.

Description

WASTE TREATMENT

The present invention relates to a novel process for the treatment of Tc".

The pre-treatment of technetium present in solution as the pertechnetate anion [Tc0 ]^" has been the subject of considerable studies.

The nuclear fuel cycle and, in particular, the processing of nuclear waste material is described in "The Nuclear Fuel Cycle" edited by P D Wilson, the content of which is incorporated herein by reference. In the processing of nuclear waste material Tc" is usually present in both the bulk effluent feed and medium active concentrate (MAC). The majority of the Tc is contained in the MAC and therefore attention has been focussed upon treating MAC. Due to its low specific activity (6.28 x 10^"4 TBq/g) Tc has a relatively long half life (2.13 x 10 years), and is therefore usually present in MAC in chemically significant concentrations.

Technetium is used widely in medicine as a radiopharmaceutical skeletal imaging agent, many papers have been published which discuss the chemical reduction of pertechnetate to other chemical species. However, the technologies described in such publications are of limited use due to the chemical composition of MAC, particularly the high nitrate concentrations, and the requirement not to affect the removal of other radionuclides, for example, the addition of chemical reagents to MAC could compromise the precipitation of the ferric floe upon neutralisation.

Other possible treatment methods for Tc" have been mentioned i.e. electrolytic or ion exchange methods, however development and implementation of such treatments would require significant effort and/or involve large additions to current processes. The pertechnetate anion [Tc0 ]^~ is a very stable species which is soluble under aqueous conditions. It has long been held that any treatment process must change the valency of the technetium to precipitate it from solution.

The International Atomic Energy Authority (IAEA) published in Technical Report Series No 337, 1992, 24; that the pertechnetate anion [Tc0 ]^" can be readily reduced in neutral or alkaline solution by the use of dithionite (S₂0₄ ^2"). The report describes that the resulting Tc0₂ is highly insoluble and can be readily removed by sedimentation or filtration. However, as the report states, re-oxidation to [Tc0 ]^" occurs rapidly in the presence of air or dissolved oxygen. Therefore, the IAEA disclosure is not considered to be enabling since, inter alia, the reduction by dithionite is found to be pH sensitive.

Our studies of dithionite reduction gave a high initial decontamination factor (DF) of > 10³ that decayed quickly over a period of 3 hours.

Concentration of Species in Feed Dr =

Concentration of Species in Product

It was suggested that the probable mechanism is reduction of the technetium VII to technetium IV which precipitates as Tc0 then gradually returns to solution as the technetium is reoxidised by dissolved oxygen:

3 S₂0₄ ^2" + 2Tc0₄ ^" + 40H^" -> 6S0₃ ^2" + 2Tc0₂ + 2H₂0

followed by 4 Tc0₂ + 30₂ + 40H^" → 4Tc0₄ ^" + 2H₂0

The initial high DF's achieved by dithionite suggest that if re-oxidation was the cause of re dissolution, then by excluding oxygen from the system, these initial high DF's would be maintained indefinitely, but completely excluding oxygen from the system is difficult to do. We have now found a novel process which allows for the satisfactory reduction of Tc " and avoids the problems of re-oxidation. This process is therefore suitable for removal of Tc from any Tc bearing stream, however, it is especially useful in removal of Tc from bulk effluent feed and medium active concentrate (MAC).

Moreover, we have found that DFs are dependent upon precise pH, whereas the IAEA disclosure makes no mention of pH. Thus, not only does the IAEA disclosure not touch towards the present invention, it can also be considered to be not an enabling disclosure.

Thus according to the invention we provide a process for the removal of soluble Tc from effluent which comprises;

(i) reducing the soluble technetium; and (ii) removing the insoluble separated reduced Tc.

The Tc is most likely to be present in effluent as Tc^vπ, e.g. [Tc0 ]^", and therefore it would be reduced substantially to Tc ^IV. As Tc0₂ is generally insoluble in aqueous solutions it can therefore be readily separated using conventionally known techniques, such as filtration, sedimentation, aggregation, etc.

The process of the invention is particularly advantageous in that it is able to minimise re-oxidation of Tc0₂ to e.g. Tc ^vπ. A number of methods have been found which mitigate re-oxidation and therefore any one or any combination of these steps may be incorporated in the process of the invention.

Thus according to a further feature of the invention we provide a process for the removal of Tc from effluent which comprises;

(i) reducing technetium whilst substantially preventing re-oxidation; and (ii) removing the separated Tc. The preferred method of preventing re-oxidation is the stepwise or continuous addition of dithionite over the course of the reaction. The frequency of the stepwise dithionite addition may vary and will depend upon, inter alia, the volume to surface area ratio of the vessel used. The volume to surface area ratio of the vessel used may vary, although a large ratio is preferred, requiring fewer repeat dosages of dithionite. However, we have found that, on the laboratory scale, a dithionite addition every 60 minutes is currently preferred. However, when scaled up, to operate on a plant the frequency of addition may be extended to as much as 60 hours. Alternatively, addition of dithionite can be effected in a continuous manner. The amount of dithionite added may also be varied depending upon, inter alia, the amount of Tc^vπ present. However, generally, we prefer dithionite to be added either stepwise or continuously, at a rate to maintain the concentration of dithionite within the range of 100 - 2000 ppm, preferably 100 - 1500 ppm. In a preferred aspect of the invention the dithionite may be added in large quantities initially, with smaller quantities being added at later additions. Thus the initial addition of dithionite may be from 500 to 1500 ppm, eg 1000 ppm. Later additions may be from 100 to 500 ppm, preferably 200 to 400 ppm, eg 300 ppm.

The reduction may be facilitated under a nitrogen blanket or a nitrogen sparge. Nitrogen sparge experiments were conducted which studied the reaction's progress over a period of 72 hours. With dithionite under air a high DF is achieved quickly but this decays away with time. Flowever, under a nitrogen sparge, the Tc removal is stabilised and even appears to increase over the time period, to a peak of 2 x 10 after 50 hours. Thus, the process of the invention may be carried out under a nitrogen sparge, a nitrogen blanket or a combination of both. However, it is a distinct advantage of the present invention that, e.g. stepwise addition of dithionite, can remove the need for a nitrogen blanket and/or a nitrogen sparge. This is especially advantageous in that on an industrial scale plant the use of a nitrogen blanket or sparge is impracticable. The process of the invention is also advantageous in that it generally does not have a detrimental effect on decontamination factors of other isotopes such as Ru¹⁰⁶, Sr⁹⁰ and Cs 137

Tests have been conducted which comprise the generation of ferric floe in situ by the neutralisation of MAC which indicate that with Ru¹⁰⁶ DF's of 50-200 are achieved. With the addition of dithionite, Ru¹⁰⁶ DF's of 200-1000 can consistently be recorded. Thus the dithionite also appears to have a slightly beneficial effect on Ru removal.

MAC is generally acidic in nature and we have found that the dithionite reduction of MAC is pH sensitive. The MAC may be substantially neutralised by the addition of a base. Conventionally known bases may be used, but the preferred bases are hydroxides, such as alkali metal hydroxides, eg potassium hydroxide and especially sodium hydroxide. Thus, in the process of the invention, MAC is substantially neutralised, e.g. to a pH of between 7 and 1 1, preferably between 9 and 1 1 , more preferably between 10 and 1 1 , eg pH 10.5. Lowering of the pH will occur upon addition of the dithionite.

The preferred pH for conducting dithionite reactions is pH 10.5. However, the addition of 1000 ppm dithionite will reduce the pH of the neutralised MAC to ~ 8. Although as the dithionite is consumed the pH will vary, it is within the scope of the present invention to adjust the pH after dithionite addition. The large change in pH on addition of dithionite can be adjusted by the addition of further sodium hydroxide to readjust the pH to 10.5. In the process currently employed to treat MAC the removal of some isotopes is pH dependent. In practice the DFs for Cs and Sr vary with pH but in opposite ways. Thus a balance is drawn in order to optimise both Cs and Sr DFs and the reaction is conducted within the range pFI6 to 14, preferably pH8 to 1 1 and more preferably at pH 10.9.

In an alternative embodiment the dithionite may be added as a solution, ie dithionite dissolved in an aqueous sodium hydroxide solution. The advantdage of the use of an aqueous dithionite/sodium hydroxide solution is that effects on the pH of the MAC are mitigated. Although the concentration of the aqueous sodium hydroxide solution may be varied, a preferred concentration is from 0.05 to 1.0 and preferably 0.1.

All treatments with dithionite, whether the pH is readjusted post dithionite addition or not, give very high Decontamination Factors for Cs¹³⁷; usually from 3 x 10³ to 10⁴.

Experiments have involved addition of solid sodium dithionite to the substantially neutralised MAC. This is because in solution the dithionite is rapidly oxidised. Clearly the addition of a solid reagent is undesirable on a plant scale. This problem may be overcome by preparing a dithionite solution under a nitrogen blanket, or by adding an excess of dithionite in order to counteract oxidation effects.

Iron present in the MAC may form ferric hydroxide (Fe(OH) ) as a floe. Advantageously, the Tc0₂ produced by the dithionite reduction may precipitate out of the reaction with the ferric hydroxide floe.

Earlier experimental work has involved a nitrogen sparge, bubbling N₂ slowly through the neutralised MAC in a dreschel bottle or using a nitrogen blanket. This has removed entrained oxygen before addition of the reducing agent and effectively maintained an inert atmosphere. On a plant scale, simple blanketing would be preferred to sparging for (a) simplicity of vessel internals and (b) prevention of entrained activity being carried into the ventilation system. More recent work has involved the use of repeated doses of dithionite without the use of a nitrogen blanket or sparge.

The dithionite process can, of course be applied to MAC permeate (filtrate) or similar streams which have been filtered to remove the ferric floe and the bulk of the radionuclides, i.e. streams containing Tc only. The absence of floe makes filtering the Tc0₂ more difficult and the DFs are lowered. Figure 1 shows the Tc DFs obtained in the presence of floe and in the absence of floe (like filtrate/permeate) at different pH after a single addition of 1000 ppm dithionite.

Thus according to a further feature of the invention we provide a process as hereinbefore described which includes a filtration step to remove other substances prior to the addition of dithionite and subsequent filtration of the precipitated Tc.

In an especially preferred embodiment the filtration step is conducted prior to the addition of dithionite.

In a preferred embodiment the process can be operated on a continuous basis rather than batch wise, and employ the use of a flooded reactor, e.g. a plug flow reactor, so that oxygen may then be excluded from the process altogether without the use of a nitrogen blanket or sparge.

In this case the Tc bearing stream is fed into the plug flow reactor concurrently with the sodium dithionite solution. The reactor is run flooded so the minimum amount of dithionite is added to effect reduction as oxygen can not absorb into the mixture. Sufficient dithionite is added to reduce the oxygen already absorbed, the chromium present and the Tc present. The mixture is passed down the reactor which is a sufficient length and at a rate sufficient to allow complete reduction of the Tc. The precipitated Tc0₂ is then filtered off through a 0.04 μm filter.

The invention will now be described with reference to the following Examples.

Table 1 Medium Active Feed (MAC Neutralised to 3x its volume) In all experiments where MAC liquor was used, the final reaction volume was adjusted after neutralisation to three times the original volume of MAC used. This table gives the activity levels in the reaction mixture. In some experiments a MAC simulant was used which contained 250 g/1 sodium nitrate, 30 ppm chromium, 100 ppm Tc in the form of pertechnate and various levels of dissolved iron.

Example 1 The Effects of Dithionite Reduction of MAC with a Nitrogen sparge

In each experiment a 40 ml aliquot of MAC was neutralised with 23% sodium hydroxide solution. Sodium nickel hexacyanoferrate II was added to a concentration of 600 ppm hexacyanoferrate to complex the caesium in the MAC. Some aliquots were then treated with dithionite to various concentrations. The pH was then adjusted back to various values e.g. pH 9 or pH 10.5. Some of the aliquots of liquor were then sparged with nitrogen for the duration of the experiment. Samples were removed periodically and filtered through a 0.05 μm filter. The sample filtrates were then acidified and submitted for analysis.

RESULTS

Table 2 The Effect of Dithionite With and Without a Nitrogen Sparge A series of three experiments were carried out to compare the effect of neutralisation of MAC to pH 10.5 alone, with neutralisation to pH 10.5 plus the addition of 1000 ppm dithionite, and neutralisation to pH 10.5 plus the addition of 1000 ppm dithionite under N₂ sparge. The pH was not adjusted after addition of dithionite.

Neutralisation by itself produced brown ferric floe and no DF for technetium. Neutralisation plus dithionite produced a black precipitate but after 50 hours the technetium DF was only 3.18 due to re-oxidation of the Tc precipitate to the soluble pertechnetate. With neutralisation, dithionite and N₂ sparging a black precipitate was formed and the DF after 50 hours was 1.86E4.

Table 3 The Effect of Dithionite at Different Concentrations

Dithionite was added at 1000, 500 and 200 ppm as a solid to MAC neutralised to pH 10.5 and with a nitrogen sparge. With 1000 ppm dithionite added the pH dropped to 7.8 and a black grainy ppt was formed. The presence of 500 ppm dithionite lowered the pFI to 9 and again a black grainy ppt was produced. In both cases Tc decontamination factors (DFs) were > 10³ after 72 hours, with 200 ppm dithionite added, however, the Tc DFs were never very high.

Table 4 Results of Dithionite Experiments with pH Readjusted to 10.5 after the Addition of 500 ppm Dithionite (with Nτ_Sparge). As expected a black grainy ppt was formed and the Tc DF after 72 hours was >10 .

The Cs, Sr and Ru DFs were unaffected.

Example 2

Reduction and Removal of Tc with Repeated Doses of 1000 ppm Dithionite 40 ml of MAC were neutralised with sodium hydroxide solution to a volume of 120 ml and pH 1 1.2. Dithionite was added (1000 ppm) as the mixture was stirred and then added again after every hour (1000 ppm). The pH was continually adjusted to maintain pH 1 1.2. A nitrogen sparge was not used in these experiments. A sample of the MAC solution was taken just before each dithionite addition and filtered through a 0.02 μm filter. The filtrate was then analysed for Tc, Sr and Cs.

The results are shown in Figure 2. The graph shows that high DFs are maintained for the duration of the experiment. The DFs appeared lower during the second half of the experiment but this was probably an analytical artefact; the redox potential confirms that dissolved oxygen was continually being used up and that the concentration did not rise enough to re-oxidise the Tc.

Example 3

Reduction and Removal of Tc with Repeated Doses of 300 ppm Dithionite 50 ml of MAC were neutralised with sodium hydroxide solution to a volume of 130 ml and pH 1 1.2. The mixture was stirred at a high rate. Dithionite was added (300 ppm) and then again at 300 ppm concentration after every hour. The pH was continually adjusted to maintain pH 11.2. A sample of the MAC solution was taken just before each dithionite addition and filtered through a 0.02 μm filter. The filtrate was then analysed for Tc, Sr and Cs.

The results are shown in Figure 3. The Tc DF is low initially but recovers to several thousand after a couple of additions. It would have been better to use an initial dose of 1000 ppm dithionite to get a high DF at the start of the experiment..

Example 4

The Effect of Surface Area to Volume Ratio of the Vessel Used in Tc Reduction and Removal with Dithionite.

1 litre of MAC simulant containing, inter alia, 100 ppm Tc, 500 ppm Fe and 30 ppm Cr was placed in each of several vessels, all with different volume to surface area ratios. Dithionite at 1000 ppm concentration was added to each. The pH of the solution was then adjusted back to 11.2. The mixtures in the vessels were then stirred at the same high rate. Samples of MAC were withdrawn periodically and filtered through a 0.02 μm filter. The filtrate was then analysed for Tc. The results in Figure 4 show that dithionite would need to be added less frequently to vessels with a low surface area to volume ratio.

Example 5

Some experiments were done in a different arrangement of apparatus. A sodium nitrate solution containing Tc (100 ppm) and Cr (30 ppm) was fed into a tube reactor at 175 ml/min together with a solution containing sodium dithionite at a concentration of 6 g/1 which was fed continuously at 25 ml/min. The mixture was allowed to pass down the tube reactor for a number of minutes and was then passed through the filter to remove the precipitated Tc0 .

DISCUSSION

Figure 1 illustrates a time dependence for Tc DF with good removal for a period of up to 3 hours and thereafter a gradual reduction in DF. This reduction is brought about by the re-oxidation of precipitated Tc0₂. Tables 2 to 4 show that the Tc DF can be maintained for long periods by the use of a nitrogen sparge or blanket. Alternatively, repeat doses of dithionite can be added to the mixture to maintain the reductive environment without the use of a nitrogen sparge or blanket.

Figure 4 shows that the bigger the volume to surface area ratio of the solution exposed to air, the longer the DF lasts because the concentration of oxygen in solution is reduced. That is the dithionite lasts longer. The significance of this is that the EARP plant itself is large with a large volume/surface area ratio and extrapolation from this data to the volume/surface area of the EARP plant (circled point) means that a dosing rate of only once every 60 hours would be required. There appears to be no adverse effect on the DFs for other radionuclides due to the use of dithionite. Table 1. Typical Activity Levels in the Reaction Mixture (MAC diluted 3 times)

Table 2. The Effect of Dithionite With and Without a Nitrogen Sparge pH Not Readjusted After Dithionite Addition

Table 3. The Effects of Dithiomte at Different Concentrations pH Not Readjusted After Dithionite Addition

Nitrogen Sparge Used

Table 4. Results of Dithionite Experiments with pH Readjusted to 10.5 After the Addition of 500 ppm Dithionite (with Nitrogen Sparge)

Claims

1. A process for the removal of soluble Tc from effluent which comprises;

(i) reducing the soluble technetium by the addition of dithionite; and (ii) removing the insoluble technetium by a separation process.

2. A process according to claim 1 in which the soluble Tc is present as Tc^V!I.

3. A process according to claim 1 in which Tc ^vπ is reduced to Tc ^IV .

4. A process according to claim 3 in which Tc ^IV is present substantially as Tc0₂.

5. A process according to claim 1 in which the dithionite is added as sodium dithionite.

6. A process according to claim 1 in which the reducing agent is added stepwise over the course of the reaction.

7. A process according to claim 1 in which the dithionite is added in a continuous manner.

8. A process according to claim 1 in which the pH of the effluent has been raised to between pH 7 to 11.

9. A process according to claim 1 in which the process is carried out without the use of a nitrogen sparge.

10. A process according to claim 1 which includes a filtration step prior to the stepwise addition of dithionite.

11. A process according to claim 1 in which the DF for Caesium and/or Strontium are substantially unaffected by addition of dithionite to the EARP process.