In order to assess the magnitude of the use of water in the non-ferrous metals industry and to determine the scope for economies, a survey was made of the water usage in five wrought aluminium plants and eleven copper and copper alloy manufacturing plants.
It was deduced from this survey that about 95 m3 of water is required to produce 1 ton of copper or copper alloy semi-manufactured product i.e. plate, sheet, strip, rod, wire tube etc.
ADVERTISEMENTS:
This is somewhat more than the amount of water used in the steel industry (50 m3/ton), rather less than is used in the paper industry (155 m3/ton) or the synthetic fibres industry (133 m3/ton) and comparable to the wallboard industry (98 m3/ton).
Analysis of the returns reveals that a large proportion of the water is used for cooling purposes. In fact, in the aluminium semi-fabricating industry this accounts for over 95% of the total requirement.
The returns from the copper industry showed that about 75% of the water is used for cooling, and although the proportion positively identified as for cooling was less in other countries, the use of a substantial proportion of the water was not identified and, might well be for cooling.
This cooling water is used for extracting heat from moulds in which metal is cast, for cooling furnaces, for cooling rolls, compressors and other ancillary equipment.
For the most part, this water is contaminated only with heat and it is therefore possible to recirculate the water for re-use after the heat has been removed. It is in this direction that there is the greatest scope for water economy.
Cooling water, treated if necessary to prevent scale formation and corrosion, may be re-circulated almost indefinitely if a suitable heat exchanger is included in the circuit.
ADVERTISEMENTS:
At the time at which this survey was made, the proportion of water re-circulated showed great variation from firm to firm, approaching 100% in a few cases down to virtually nil in many instances.
It is believed that increasing water costs together with the difficulty of abstracting more water from wells, lakes, rivers and canals to meet the demands of increasing production, have considerably changed the situation and that a much greater proportion of cooling water is now re-circulated, thus reducing the demand on water resources.
Contaminated Wastewater:
In the copper semi-fabricating industry, there is a substantial demand for water for rinsing work after it has been acid pickled. This demand is almost totally absent in the aluminium industry where the oxide film formed when the metal is heated for manipulation in various ways is protective and remains thin and scarcely visible.
Water used for pickle-rinsing is contaminated with the pickle liquor itself, in the case of copper and its alloys mainly sulphuric acid sometimes assisted by an oxidising agent such as sodium bichromate, and with metal salts.
ADVERTISEMENTS:
It may be seen that rinse water accounts on average for about 20% of the water consumption in the copper semi-fabricating industry. In individual firms it can rise to over 90% of the water demand depending on the nature of the product.
Reliable figures are difficult to obtain, but the water consumed for pickle-rinsing in the copper and copper alloy industry is probably of the order of 10 M m3/annum.
The demand for water for rinsing purposes in the metal finishing industry has been estimated as between 50 M m3 and 150 M m3 per annum. Because of its acidity and more particularly because of contamination with heavy metals such as copper, zinc, nickel, lead etc. this rinse water is unacceptable for discharge to public sewers or inland waters without costly preliminary treatment.
There is another and possibly growing source of contaminated wastewater brought about by the requirement to avoid atmospheric pollution. Both primary smelters and those dealing with secondary metals inevitably generate gaseous products which are contaminated with metal oxide fume and sometimes with chlorides and fluorides.
ADVERTISEMENTS:
Waste gases similarly contaminated also arise in re-melting operations. In most countries, it is now inadmissible to discharge such products to atmosphere without purification to reduce metal fume, chlorides, fluorides etc. to a level at which they are considered to be innocuous and, frequently, invisible.
This fume etc. is mainly sub-micron in particle size and it can only be effectively collected with filtration plant of the highest efficiency. Suitable plant includes high efficiency wet scrubbers of the venturi type and, though less common, irrigated electrostatic precipitators.
Although the water used for scrubbing is re-circulated, there is obviously need to bleed off contaminated water and to remove from time to time, the sludge of metallic oxides etc. which accumulate. Thus we have an example of an environmental problem in one sphere giving rise to a pollution and waste disposal problem elsewhere.
Treatment of Wastewater in Non-Ferrous Metal Industry:
Most plants which made returns seemed able to dispose of their effluent for little or no expenditure, presumably by making use of local rivers, lakes or canals. It is apparent that this state of affairs has changed dramatically over the past few years with pressure from the environmental lobby and the need to preserve supplies of potable water.
The simplest method for dealing with contaminated effluent is by dilution and this has possibly militated against the recirculation of cooling waters. They were a convenient way of diluting effluent to an extent such that the contamination was less obvious. Such measures can rarely lower the degree of contamination to an acceptable level. More heroic steps are needed.
The conventional treatment is neutralisation of the acid and precipitation of the heavy metals as hydroxides. Lime, added as a slurry, is usually found to be the cheapest reagent, but sodium hydroxide or even sodium carbonate have also been used.
The latter is possibly less desirable because the carbon dioxide generated tends to hinder the settling of the hydroxide precipitate. Where sodium bichromate is used as an oxidising agent in the pickle, the chromium has first to be reduced to the trivalent state, usually with sulphur dioxide.
Many plants, of greater or less complexity, have been installed for treating effluents from copper and copper alloy manufacture, flow sheets for a relatively simple plant and a more complex plant being given in Figs. 31.2 and 31.3.
One drawback to this system is the lengthy retention time needed for settling the precipitated metal hydroxides which results in large, space consuming settling tanks. The alternative flotation processes using air bubbles to carry the suspended solids to the surface are claimed greatly to reduce the space requirement.
Many plants for treating effluents arising from pickling processes were built for the Indian copper industry and elsewhere along the above lines, when the prime consideration was to improve the quality of the effluent to a level acceptable to public authority sewage systems or, in some cases, to a level at which discharge to inland waters was permitted. These plants have done much to lower the contamination of water courses, but various factors are combing to make it advisable to reconsider the situation.
Firstly in some countries, for example – the United States and Italy, authorities are proposing effluent discharge standards which are difficult, may be impossible, to meet by the conventional process.
Secondly, the precipitation process gives rise to quantities of hydroxide sludge which is becoming increasingly difficult to dispose of.
Thirdly, the high cost and increasing scarcity of water, together with the levy of charges for accepting treated industrial effluent suggest that the economics of the process now compare unfavourably with other more sophisticated treatments which allow the water to be recycled.
Effectiveness of Precipitation Treatment:
Data on the solubility of the heavy metal hydroxides are rather sparse and indicate that the solubility may be highly sensitive to pH and affected by the presence of other soluble salts. The factor of principal interest is the concentrations of heavy metals likely to be present in the overflow from a settling tank; part of the metal is in solution and the remainder in suspension. Experiments carried out on the lime treatment of synthetic pickle-rinse solutions under somewhat idealized conditions are illustrated in Fig. 31.4 to 31.6.
These indicate that the pH to which neutralisation is usually carried, around 8-8.5, the limiting concentration of copper in solution is in the region of 0.2-0.4 ppm; for zinc the figure is 1.0-3.0 ppm; while for nickel it is 0.5-1.0 ppm.
In the case of nickel and zinc, the concentration of metal in solution tends to fall as the pH rises to about pH 10 (though the reverse was found to apply with chromium). In practice, it is possible that somewhat higher figures would be obtained depending on the extent to which reaction with the lime was carried to completion. However, these concentrations are already higher than have been suggested as standards in some countries.
In practical effluent treatment, the separation of the precipitated metals is always less than perfect, so that to the metal in solution it is necessary to add metal in suspension. A not uncommon figure for the acceptable total of heavy metals in solution and in suspension is 30 ppm. This can be achieved readily. With good settling facilities about half this is reasonably practicable and the amount of metals in suspension could no doubt be reduced considerably by sand filtration.
One published account of the limits alleged to be under consideration by the U.S. Environmental Protection Agency for the metal finishing industry, summarised in Table 31.5 makes it clear that compliance would require a considerable improvement over the performance to be expected from a conventional lime precipitation treatment plant.
The necessity for these stringent requirements, which are in general more restrictive than those for drinking water, is open to debate, but it is doubtful whether the requirements could be met by a lime precipitation plant under industrial conditions.
Sludge Disposal in Non-Ferrous Metal Industry:
Precipitation treatments give rise to a metal hydroxide sludge with a very high proportion of water. It has been estimated that more than 100,000 tons per annum of such sludge is produced in the U.K. from the metal finishing industry and from the pickling of copper and copper alloys.
To make this easier to transport, the sludge is often passed through a filter-press or rotary vacuum filter, producing a filter cake still containing about 70% water, but sufficiently solid to handle. Not all of the sludge is filtered; some is taken away in road tankers and dumped.
It has been estimated that in India about 30,000 tons per annum of filter cake is produced from the copper and copper alloy industry together with the metal finishing industry. Although this filter cake contains valuable metal it has little or no commercial value and most of it is simply dumped.
The high moisture content is probably the main reason why the filter cake is unattractive to the secondary metal refiner, coupled with the fact that the cake usually contains a mixture of metals. Quantitative information is difficult to obtain, but the amount of metal wasted is probably of the order of 10000 tons/annum of copper and a comparable quantity of zinc plus considerable amounts of nickel.
The conservationist would no doubt deplore this loss of metal but of equal concern to the industrialist is the fact that sites for dumping such residues are becoming increasingly rare. Frequently the residues must be conveyed over long distances, incurring high transport charges, to an approved site where it is considered that leaching of the residues by rain water can cause no harm to water resources.
Much thought has been given to recovering metal from filter cake. Resolution of the metal hydroxides, followed by separation of the metals and recovery by solvent extraction procedures is one interesting possibility.
However, even at the high metal prices that were ruling a few months ago, the economics of the recovery process have not been sufficiently attractive to encourage commercial exploitation.
The problem of disposing of large quantities of hydroxide sludge seems to be an inherent drawback to the conventional effluent treatment process and one which is becoming increasingly urgent.
Economics of Precipitation Treatment:
Economically, the conventional precipitation treatment is becoming less and less attractive with each increase in the cost of water, since the treated effluent is not normally of sufficiently high quality to permit recycling.
Furthermore, it is common practice to make charges for receiving the treated effluent, the charge depending on both the volume of effluent and upon the degree of contamination. Thus industry has frequently to bear the double cost of purchasing water and of discharging the water after use. The costs vary considerably with the region.
It is unlikely that it has yet reached a plateau. It follows that there is now a strong and growing incentive to adopt practices which facilitate water recycling within the works. The cost of water for rinsing work after pickling or plating is a significant proportion of the total processing cost.
In fact the savings which could be made from recycling water are of greater significance than monies which might accrue from the recovery of metal from effluents, bearing in mind the probable cost of a recovery operation. Obviously a process which enables water to be recycled and metals to be recovered is doubly attractive.
Available Treatment Process in Non-Ferrous Metal Industry:
The methods of effluent treatment available to the non-ferrous metals industry include:
(i) Precipitation
(ii) Ion-exchange.
(iii) Electrolysis.
(iv) Crystallisation.
(v) Evaporation/distillation.
(vi) Cementation.
(vii) Reverse osmosis.
(viii) Electrodialysis.
(ix) Adsorption.
(x) Liquid/liquid extraction.
Precipitation, as outlined above is most widely used and it is applied mainly to the treatment of combined wastewaters arising from various pickling or finishing operations. However, it has been applied to the treatment of wastewater at source, i.e. at a single rinse tank, in the technique of so-called “integrated treatment”.
The metal hydroxide produced is then of simpler composition which may make recovery more attractive, but the main claim for the process is that the rinse water may be recycled. In practice, the buildup of chemicals in the rinse water requires a bleed-off and make-up with clean water which is stated to be about 20% of the volume circulating.
Electrolysis is widely used for regenerating pickling solutions used for treating copper, but dilution is too great for efficient electrolytic recovery of metals from rinse waters. While it is possible to obtain increased cathode efficiency by employing “fluidised bed” electrodes, it is doubtful whether metals could be removed down to a level acceptable for effluent discharge. Cementation suffers from the same drawback.
It is feasible to recover copper from dilute wastewaters by displacement using iron scrap, but the resulting effluent still contains several ppm of copper in solution, and the substitution of iron sulphate for copper sulphate does not assist in making the water suitable for recycling.
Crystalisation and evaporation or distillation are more suited to the recovery of metals from concentrated exhausted pickling solutions than from dilute rinse waters; similarly liquid-liquid extraction is more applicable to fairly concentrated solutions.
In fact in reviewing the available processes only ion-exchange and reverse osmosis appear to possess the required characteristics of enabling a high proportion of water to be recycled and preventing loss of metal with an effluent stream.
Ion-exchange is already being used in the electroplating industry and at least one plant is now in operation treating the effluent from a copper semi-fabricating works. Reverse osmosis as applied to effluent treatment is less well developed, but on first consideration it appears to have the attraction over ion-exchange in being basically simpler and continuous, with no regeneration cycle.
There are, however, doubts about the life of membranes under the conditions of low pH operating in effluent treatment plants dealing with wastewaters arising from pickling operations.
These doubts are unlikely to be resolved until long-term pilot scale tests have been made. Assuming a good life for the membranes, costs would appear to be similar to ion-exchange treatments. So far experiments seem to have been confined to the recycling of nickel plating rinse waters where conditions for the membranes are possibly less onerous.
Practical Application of Ion-Exchange:
A closed circuit ion-exchange treatment plant for pickling effluents has recently been described by Camenisch. It is illustrated diagrammatically in Fig. 31.7. Rinse waters are collected from nine pickling areas and suspended solids are removed in a settling tank which is monitored to ensure that concentrated pickle does not accidentally enter the regenerating system.
Before entering the ion-exchange units, the rinse water is filtered in a multi-bed filter containing layers of sand and activated carbon. This removes oil, grease, or surfactants which could poison the resins.
The filter is cleaned weekly by back washing. Both cation and anion exchange columns are used yielding demineralised water which is fed back via a storage tank to the pickling plants. Twin sets of ion-exchange columns are used and when one set is exhausted there is an automatic change over to the second set.
Regeneration takes place about every 12 hours and the regeneration cycle takes about 2 hours. The regeneration liquors are neutralised to precipitate the metal hydroxides, which, after settling, are filtered to produce a filter cake with about 30% solids. The filter water is processed through a small cation exchange column to remove the last traces of copper and zinc before discharge to the sewer.
The cost of water and of conventional effluent treatment is approaching that of the recycled water cost. If, as in some cases, a charge has also to be paid for discharging treated effluent, it may be seen that the ion-exchange plant is already offering an effluent treatment comparable in cost, if not cheaper than the conventional precipitation process.
Thus, although the high initial capital cost is a deterrent, long term economics appear to be changing in favour of ion-exchange because of the attraction of water recycling and the system would seem to be a strong contender for new installations.
New Pickling Formulations:
Changes in effluent treatment in the copper industry are being accompanied by modifications to pickling practice which simplify the treatment of rinse waters and also facilitate the electrolytic regeneration of the pickling solutions.
Although dilute sulphuric acid is the common pickling agent, for some purposes it is desirable to use an oxidising solution. Sulphuric acid-dichromate pickle is the traditional solution for brass and a range of copper alloys. The presence of the chromate ion complicates both electrolytic regeneration of the pickling acid and also the treatment of the effluent.
Over the past 5 year progress has been made in substituting hydrogen peroxide as the oxidising agent, which avoids the effluent treatment complication and does not interfere with the electrolytic recovery of copper from the pickle solution. Stabilised hydrogen peroxide is now available capable of operating in pickling baths and a chemical efficiency of 80% is stated to be achieved.
Recommended concentrations vary widely from as little as 2 g/l of 35% hydrogen peroxide to 30 g/l depending on the alloy being pickled, though there is a danger from the higher concentrations in that the oxygen bubbles generated can carry acid droplets into the atmosphere.
The appearance of work pickled in hydrogen peroxide containing solutions is different from that pickled with chromic acid as the oxidising agent and customer reaction may be a deterrent to the wider adoption of this newer pickling method.
On the other hand, it is claimed that the chromate conversion coating resulting from the dichromate pickles can lead to tool wear in drawing brass wire. Some operators have also reported difficulty in obtaining a satisfactorily clean finish on some copper alloys. This may be a question of optimising the pickling conditions and more work is obviously required.
The efficiency of the pickling solution decreases if the copper content builds up and there is an increasing tendency to cause staining of brass. Build-up of zinc in the pickle has little or no effect. By cycling the pickle through and electrolytic cell the copper can be plated out and sulphuric acid regenerated.
Drag-out of pickle solution and make up results in an equilibrium value being attained for the zinc concentration. Thus it is possible to keep the pickling solution working almost indefinitely with only fresh additions of acid to compensate for drag-out.
As a result of these developments significant progress is being made towards closed systems with very little waste for disposal. In a few plants, the possibility has already been demonstrated.
The cost of changes will undoubtedly delay wide application of the newer methods, especially in the present economic climate, but increasing charges for water, problems of effluent and waste disposal, and more stringent effluent purity standards are all tending to make the closed cycle concept more attractive.