The raw materials of the nuclear industry (mainly uranium) are radioactive, its useful transmuted products, primarily plutonium but also specific radioisotopes, are more active, but the bulk of activity it has to handle is produced as an inevitable consequence of nuclear fission (i.e. fission products) and by the neutron activation of associated materials such as power reactor components. Table 7.2 indicates the main stages of the fuel cycle and the types of activity arising.
Much wet chemistry is involved in fuel reprocessing operations; the cleaning and decontamination of solids elsewhere in the fuel cycle ranging from reactor circuit components to protective clothing also use water as the primary medium, so it is readily appreciated that the treatment of active liquid waste is an important consideration in the nuclear industry.
Discharge Requirements for Effluents in Nuclear Industry:
ADVERTISEMENTS:
The activity in a liquid discharge is limited broadly by two criteria:
(i) Absolute discharge limits are fixed by the regulatory authority to ensure that no individual member of the public receives more than the maximum dose indicated by international recommendations, and that the average dose to the population at large does not exceed another, lower, figure.
(ii) Below these limits the activity is to be as low as reasonably achievable (Alara principle).
To derive discharge limits for- (a) can involve elaborate studies and calculations of the way radioactivity can reach the “critical group” of most exposed individuals – e.g. perhaps by concentration in aquatic organisms which are then consumed. To determine what is “reasonable” under (b) it may be necessary in addition to pursue a notional economic balance so as to ensure that money spent on more elaborate clean-up procedures is justified in terms of the radiation detriment saved. However, this kind of calculation does not apply to the absolute limits (a), which must be achieved irrespective of cost.
As elsewhere, dilution by the receiving water body (stream, river or sea) is of help in achieving acceptable concentrations but in many cases treatment will be needed. One objective of treatment is stated as being to concentrate the bulk of activity in as small a volume as possible and that indeed is most often the objective of the liquid treatment plant taken alone.
ADVERTISEMENTS:
However, the system requirement goes beyond that. Most of the environmentally harmful species which arise in nuclear effluents are not potentially useful for recovery and recycling, even were it economic to do so, and hence the ultimate objective is to isolate the activity from man’s environment.
To achieve this isolation from the biosphere for radioactive species of appreciable half-life requires that the activity so concentrated should be immobilised (i.e. solidified in an inert form, relative particularly to adventitious water) and then placed in a remote position, such as an underground repository.
The outline of this concept is shown in Fig. 7.9 and it will be noted that it involves possible routes of radiation to man other than via the plant liquid effluent. All the operations, including immobilisation, storage and disposal as well as treatment (and taking into account secondary arising’s such as used components) need to be carried out in such a way as to minimise the total return to the biosphere, according to criteria (a) and (b), and also to minimise the dose to the operators themselves.
For activity of reasonably short half-life (upto say 1 year or so) a delay period before discharge may be the best treatment – for economic reasons this also often requires pre-concentration of the activity.
Effluent Sources and General Methods of Management:
ADVERTISEMENTS:
Table 7.3 gives examples of process streams in which the activity arises at different types of site, and a broad indication of the volumes of effluent.
All sites practise a degree of segregation between streams, e.g. high, medium and low activity, detergent, oily and so on, to facilitate methods of treatment specific to ‘problem’ components. At the same time there is a requirement to make the best use of a few centralised effluent processes, rather than to have a proliferation of effluent treatments, each serving one process unit.
ADVERTISEMENTS:
Examples of management features are:
(i) A Reprocessing Plant:
The main process medium is nitric acid and the separations are performed by solvent extraction. Where salt additions can be avoided the activity can in principle be concentrated to join the first cycle raffinate (highly active waste) for further concentration and evaporation.
Segregation tends to be according to the main type of activity (α, β, γ long or short-lived) and to salt content. Some of the fission products most difficult to separate from U and Pu (viz. Ru, Zr/Nb) and which therefore appear in medium or low active wastes, have short half-lives. Hence the principle of delay and decay has a useful role with reprocessing wastes.
(ii) A Modern Power Station (AGR):
Liquid wastes are routed to a central treatment plant separately as oily, detergent, tritiated waste and also as the hydraulic water accompanying transferred solids (ion-exchangers, filter back-wash).
They are joined there by water from fuel element storage ponds already treated in a separate ion-exchange plant and are treated as appropriate by settling, various types of filtration and ion-exchange if required. Reception tanks allow for the batching of separate types through the appropriate treatment units while final delay tanks allow sentencing to discharge or recycle if necessary.
(iii) R and D Establishment:
Segregation is according to activity level, but low-level solutions containing complex agents pass to more intensive medium active treatment. In the laboratory appreciably active waste is segregated in carboys, dilute waste or washings pass via sink drains to external delay tanks where they are sentenced and routed via low active drains or by medium active tanker to the appropriate treatment plant. The different activity levels then receive appropriate intensity of treatment with the effluent from one level joining the feed to the next lower level (Fig. 7.10).
Chemical Treatment Methods:
1. General Methods:
The general methods available for treatment are:
(i) Concentration, e.g., by evaporation, ultimately to the point of solidification of the active fraction.
(ii) Removal from the bulk water in a solid form, for example, as a chemical precipitate.
Because the chemical concentration of radioactive species is often extremely low, direct chemical precipitation is seldom practical and the phenomena of co-precipitation, ion-exchange and adsorption need to be invoked. Considerable ingenuity has been involved in developing processes which are reasonably selective for the radioactive species in the presence of overwhelming quantities of inactive salts, since the latter can often be discharged.
Considering briefly the chemistry of the active species – the α-emitting actinides (Th, U, Pu, Am. Np) are multivalent and easily form insoluble (hydrated) oxides; they exist in appreciable concentration as metallic cations only at low pH. Pu and Np in particular have several possible valency states. As the pH approaches neutral, and in alkaline solutions, they form colloids or precipitates, but some at least can also exist as (anionic) complexes, e.g. with carbonate.
This variety of possible forms helps to explain the rather variable efficiency of their removal from mixed waste liquids by ion-exchange techniques, the more general method being (hydr-) oxide precipitation in the presence of a collector/absorber such as ferric or aluminium hydroxide. Where the solution composition is simpler they can be removed with high efficiency by ion-exchange.
The active fission products are mostly β and/or γ emitters and comprise a wide range of elements, among the most important being Cs (alkali metal) and Sr (alkaline earth) because of their high abundance, half-life and relative radiotoxicity and also Ru on account of the many complex forms it can take up.
Activation products of transition metals (e.g. Fe, Co, Mn) are also of low solubility in near neutral solutions. Like α species these are often present in the raw effluent in a suspended form, i.e., as radiocolloids, possibly absorbed on other particulates (e.g. Mg or Fe oxide).
Another by-product of nuclear reactions is tritium.
This is not generally removed from effluent streams provided it can be adequately diluted in the receiving source because it has a low biological effect and as an isotope of hydrogen in the water molecule it is very difficult to remove. On some inland sites abroad it has been suggested to evaporate tritiated effluent into the air, to achieve wider dispersion than would be possible into the local water courses.
2. Precipitation:
The most common inactive cations are Na and Fe from processing, Mg from fuel cladding (the only fuel element cladding which is readily soluble in nitric acid) and Ca. It is thus relatively easy to carry out a selective precipitation to remove nearly all the actinides using the carriers.
Other species are removed to some extent by ferric hydroxide but may also be precipitated by forming an insoluble compound of an analogous cation. For example, Sr is removed by precipitating calcium phosphate as hydroxyapatite at pH 10 or with BaSO4 at pH > 8.5. The latter method permits magnesium (inactive) to be retained in solution and discharged with the supernate.
MnO2 can also be used for Sr, as well for manganese and a number of amphoteric oxides or hydroxides, notably titanium hydroxide, have the ability to carry active cations down by physical absorption and by ion-exchange.
Divalent (Cu, Zn, Ni) ferrocyanides are specific absorbers for Cs and are widely used, while ruthenium is precipitated by reducing compounds such as ferrous hydroxide or cobalt sulphide. All these precipitates are generally formed in situ in a highly disperse state and are therefore very effective absorbers.
The French La Hague process, shown in Fig. 7.11, is an interesting illustration of the use of precipitation to treat the combined medium and low active effluents from a plant with “salty” wastes.
In general the equipment used for these floc precipitations mirrors that in the water industry, “flash” mixers are often used to obtain a highly disperse precipitate, followed by gentle agitation and gravity settling in large tanks or the use of a sludge blanket precipitator.
Decontamination factors (DF) of 1000 can be achieved for a species, 100 for β/γ species provided that appropriate precipitants are used but lower values (by a factor of 10) are common in practice with mixed effluents. One limitation of the DF is probably the carry-over of colloidal particles and fragments of floc.
The addition of extra compounds adds to the volume of floc for final disposal, and even with a simple process such as ferric precipitation the volume of floc before dewatering may be 1 % of the water treated for a floc solids addition of perhaps 50-100 ppm (10–2%) on the same basis (i.e. the settled floc is 99% water). Dewatering before immobilisation can be aided by freeze-thaw techniques to denature the floc, as well as by prolonged settling, but is then finally achieved by filtration or centrifugation.
3. Ion-Exchange:
Ion-exchange presents a more complicated flow-sheet than do most floc processes because of the steps involved—absorption, back-flush, regeneration and wash, plus periodic replacement of the bed. The alternative of “once through” operation avoids some complication at the expense of more frequent solid-handling operations, and hence it requires a high capacity ion-exchanger.
The regeneration liquids (when used) are generally concentrated by evaporation before immobilisation, so once-through operation can avoid that step also, although the final volume of solids is liable to be higher.
Space does not permit a comprehensive listing of types and applications of ion-exchangers – in principle an exchanger can be found to remove any of the soluble active species but the technique is at greatest advantage:
(i) In complete demineralisation of low salinity waters such as in the cooling circuits of water reactors,
(ii) Where the exchanger can selectively remove an active species in preference to the bulk species.
Striking examples of (ii) are found in the removal of Cs in the presence of macro-quantities of sodium ion in cooling pond or process waters. For instance, the CEGB employ both organic and synthetic inorganic resins for primary removal of Cs, with the regeneration liquors passing through a further zeolite molecular sieve which is then discharged for immobilisation. This sequence can give DFs of up to 500 for Cs.
In these cases the ion-exchanger used is in conventional particulate form in equipment similar to that for a deep-bed filter and is discharged hydraulically when exhausted, or when the regeneration efficiency falls below a set value. Elsewhere “Powdex”-type resins are used at power stations in a pre-coat filter mode and hence on a once-through basis. A related concept is that of vermiculite which also absorbs Cs.
Both organic bead resins and inorganic granules, natural or synthetic, have application- organic resins are suitable only for modest levels of activity, because of their degradation under radiation. Inorganic exchangers do not have this limitation and are of increasing interest because of their susceptibility to incorporation in a variety of matrices, including glass, ceramics and concrete.
Granular analogues of some precipitants such as hydrated titanium oxide and the ferrocyanides, have also been developed. Other types of ab- or adsorbent have also been used, notably activated charcoal for Pu removal and lately there has been some interest in inert substrates impregnated with a liquid extractant, a useful possibility where the trace quantities to be removed would not justify a liquid-liquid extraction operation.
A further novel technique which may find application is liquid membrane processing, which is also an analogue of solvent extraction. So far as is known, no work has yet been done on its use in the waste treatment field though it has been studied for application in the uranium processing industry.
Physical Treatment Methods:
1. Filtration:
A range of filters is employed, virtually all of them capable of being backwashed to recover the resultant activity. Secondary recovery is then by pressurised deep-bed sand filter, used for power station and other treatments, often as the first stage of treatment before ion-exchange.
In streams of pH above 3-4 these will generally remove much of the α load which is in suspended form. Indeed for suitably low active streams such as laundry waste, water filtration may be the only step necessary.
Pre-coat filters are also used for polishing duties; these have some potential for removal of colloids (or even solutes if the pre-coat has ion-exchange properties) but of course are less suitable for handling appreciable loads of suspended material. Ceramic candle and wedge wire filters are other types which can remove the finer range of solids but without adding to the discharge load. A modern power station treatment plant, can use filtration as the main treatment over a range of streams.
2. Evaporation:
Evaporation is perhaps the single most effective method of treatment, particularly in the absence of dissolved volatile activity such as iodine. Its decontamination is limited only by droplet carry-over and in normal practice with due attention to demisting, DF’s of 104 can be achieved in single stage equipment.
However, it is relatively very expensive in energy and equipment and is generally used as a final process before immobilisation:
(i) For high activity process liquids—the so-called highly active waste stream of a reprocessing plant and possibly for medium active raffinates of low salt content, which can then join the HAW for further evaporation and eventual vitrification.
(ii) For salty liquids of medium activity, particularly the used regenerating solutions from ion- exchange operations in a power station. (In this case the choice may be influenced by the relative cheapness of steam at a power station). Here the whole concentrate is then immobilised, salts and all, generally in concrete or bitumen.
The degree of concentration which can be reached is limited by the point at which precipitation of some salt starts, unless a crystallising evaporator is used. For low salt wastes the extent of evaporation can be very high, ultimately it may be limited by the increasing activity in the overheads as the evaporator bottoms become more concentrated (assuming the same equipment DF).
For medium active wastes conventional types of natural and forced circulation evaporators are used, sometimes in a two-effect system and at reduced pressure to reduce corrosion and incidentally to safeguard against outward loss of activity through any leak.
At the other extreme from “liquid only” evaporators are devices meant to produce a solid product for immobilisation, particularly the wiped film evaporator. Indeed this type of device has been adapted to handle bitumen as well, which is mixed with the concentrated liquid or slurry from which water evaporates as it passes along the heating surface, so that the final product is a molten bitumen with dispersed solids, which flows directly into the disposal drum to give an immobilised product.
3. Membrane Processes:
Reverse Osmosis – performs the same function as distillation but with a lower decontamination factor. The process consists of flowing water at high pressure past a semipermeable membrane, ions and large molecules are “filtered” out and a purified permeate emerges at the other side of the membrane, which at best will reject upto about 99% of salts.
Reverse osmosis is already displacing distillation from many large-scale industrial applications such as production of potable water. Although it is not suitable for the treatment of very salty or acid wastes on account of the membrane limitations to a mid-range of pH and to osmotic pressures of 20 or 30 bars, it can be used for the preliminary concentration of a bulk stream of modest activity, where an overall DF of 10 say, is adequate. (The degree of concentration per pass is usually less than ten-fold, but if the salt content is still fairly low a second stage (or batch recycle) may be used).
It has been used in the treatment of power station laundry waters, but perhaps the best example is the use in the Waste Treatment Centre at the Chalk River Nuclear Laboratories in Canada, to deal with reactor and laboratory wastes.
After a first concentration in spiral wound RO equipment, protected by ultrafiltration, the liquor passes to a tubular RO unit (commonly used in industry for handling process wastes) where it is further concentrated ready for feeding directly to a wiped film bitumeniser.
Ultrafiltration – uses more open membranes and lower pressures than RO to effect the removal of large molecules, colloids and any suspended matter. The process is sometimes used as a clean-up before RO to prevent fouling the membranes of the latter.
Its particular use in the nuclear industry would be for the concentration of colloidal particles:
(i) Which already exist in the waste stream, particularly colloidal particles of the actinides or activity present on other colloids.
(ii) Which are deliberately formed by addition of very low concentrations of precipitating agents such as ferrocyanide and copper ions.
In both of these modes the DF achievable is considerably greater than by sedimentation, even under centrifugal conditions, or than by normal filtration. The colloids can often be concentrated up to the point of coagulation – and this can considerably reduce the net addition of solids to the system compared with a normal floc process (even when the volume of used membranes is taken into account). For example, doses of 5-10 ppm of precipitant have been shown to give DFs of 100 for Cs and Sr. Fig. 7.12 illustrates in simplified form the flow schemes in such a process.
To date only mode (a) has been demonstrated at the industrial scale, to treat α effluents for the Mound Laboratory, USA, but the second mode has been studied there in some detail and with considerable success and it is hoped shortly to test it on an operational scale.
In Electrodialysis – the salts are removed from the liquid under an electric field and via an ion- exchange membrane which permits ion flow in one direction only. By a suitable choice of barriers a salt contained in the feed may be simply removed and concentrated into a parallel stream, or split into its component acid and base.
Because of the complicated plate and frame construction of equipment, entailing difficult membrane replacement, it is not favoured as a method of concentration, but has been used in special cases as a means of recycling valuable components (D3BO3 in a heavy water reactor, where D = deuterium and apparently for ordinary process recycle of acid/base in the USSR). The decontamination factor in the “clean” stream is limited by falling electrical conductivity at low salt concentration.
4. Electrical and Magnetic Processes:
A number of techniques are being explored for the use of electricity in waste treatment, on the broad principle that this is a method of promoting chemical or physicochemical changes without the addition of further materials.
A particular example of this is electrochemically controlled ion-exchange, in which H+ and OH– ions generated electrochemically in situ on the ion-exchanger are used instead of regenerate solutions. Other processes of interest include electro-osmotic dewatering of flocs, and electro-filtration for the concentration of colloidal suspensions, complementary methods of separating solids from water. These processes are at an early stage of exploration but may well play a useful part in future plant designs.
High gradient magnetic separation, by contrast, is already established in the non-nuclear field as a method for removing suspended particulates which have ferromagnetic or even paramagnetic properties. For the latter in particular it is necessary to use high fields and high field gradients; these are provided by a matrix collector of ferromagnetic material placed in the fluid between the poles of a powerful electromagnet.
Tests on the possible application to ferric flocs from nuclear waste plants have been described, though it has been concluded that a material of higher magnetic susceptibility would be desirable and that it may be possible to obtain such precipitates from the same iron solutions. One technique of interest here, namely the preparation of substitute ferrites of the formula MxFe3-xO4 where M is a divalent metal has also been suggested.
Treatment of Concentrated Activity:
For medium and low active solids and concentrates, such as form the product of liquid treatment processes, the following forms are used or available on an industrial scale:
(i) Concrete.
(ii) Bitumen.
(iii) Organic resins (mainly thermosetting).
In general the matching of the above matrices to the waste form is fairly straightforward:
Concrete can be used with inorganic sludges or ion-exchangers, but is not favoured for organic resins because of the differential swelling/contraction on water absorption. It may be used with neutralised concentrates, subject to chemical adjustments such as the use of sulphate resisting cements or the adjustment of the ionic composition of the concentrate.
Special compositions of concrete may also be preferred, to reduce the alkalinity of the concrete and the consequent displacement from its solid form of cationic activity in the associated waste.
Bitumen, which presents a relatively impermeable barrier, may be used for most types of residues but is sometimes suspect in the presence of oxidising species such as nitrates. It presents the possibility of simultaneous dewatering and encapsulation but requires a relatively complex plant.
Organic resins also form a relatively impermeable matrix and water tolerant resins are now available which can be used directly to encapsulate dam solids such as ion-exchangers, or even liquid concentrates when dispersed as an emulsion.
With low activity residues, provided that a satisfactory encapsulation process can be worked out and a sound solid product obtained, the main consideration will be the leach rates under storage, transport and disposal conditions.
The situation for storage or transport can, in principle, be predicted from product leach tests; but in the disposal situation we may be concerned with long-term behaviour (when appreciable quantities of α active species are present) and it is doubtful whether sufficient detail yet exists to determine a clear order of merit for the encapsulating media in that respect; i.e. to know the relative chemical stability after very long periods of burial.
The behaviour of waste in the final repository situation as well as of the other barriers between the activity and the biosphere are in fact the subject of substantial research programmes in a number of countries.
With more active solids an additional consideration is radiation stability of the solids and even heat release, which may put the organic media, bitumen and resins at a relative disadvantage.
Other possibilities under investigation include ceramic matrices – by careful selection these would provide better leach resistance and greater long-term chemical stability at the cost of a more complex (high temperature) process with higher costs and greater possibilities of radiation dose to operators. In the limit such processes tend to those favoured for highly active waste, particularly vitrification.
Thus there are no rigid recommendations which can be made to define which process will be suitable for any particular effluent; in most cases a careful study will be needed which includes the points mentioned here and involves some experiments with the actual waste. Selection may be made from a range of established processes with the proven performance and operational reliability to promote user confidence in most situations.
However, the continuing pressures to maintain or even reduce discharge levels with expanding process throughputs, together with the substantial costs expected for immobilisation and disposal should provide a strong incentive to introduce processes which promise improved decontamination and greater concentration factors.
The technology of waste management extends to the analysis into the remote future of the movement and possible effects of the radioactivity disposed of. This, as well as the range of techniques studied and practised for the more immediate treatment of wastes, is indicative of the very serious consideration which the nuclear industry gives to its effluents.