FRESHWATER FROM THE SEA: DEVELOPMENTS IN BRITAIN

By R. S. SILVER, M.A., B.Sc., Ph.D., D.Sc., F.Inst.P., M.I.Mech.E., M.I.Mar.E.

Issued by British Information Services – The Certificated Engineer November 1965.

International interest in the conversion of seawater into freshwater has been shown at conferenc.es such as that arranged by the International Atomic Energy Agency at Geneva last year and the United Nations has published a survey of water desalination in developing countries. In October of this year conference and exhibition on desalination will be held in Washington, United States of America, and on show will be the facilities available in Britain for the construction of desalination plant anywhere in the world. 

In all this, there has been a tendency to overlook the fact that the existence of successful methods of desalination is due mainly to British research and engineering. More than two-thirds of the world's existing desalination capacity is of that country's design and manufacture. The achievement to date and the challenge for the future may usefully be considered in the hight of the basic science of the problem. 

When salt is dissolved in water, heat is given out (the heat of. solution) and the amount corresponding to the salts 10 seawater is about 0.67 kilocalorie per kilogram of water. Hence it is physically impossible to obtain. fresh wat r from the sea with an energy consumption of fewer than 0.67 kilocalories per kilogram (kcal/kg.). The ancient method of desalination by single distillation requires 540 kcal/kg, which is about 80o times the heat of solution. Thus superficially distillation does not appear attractive as an economic desalination process, but closer consideration reveals some more favourable characteristics. 

Second Law of Thermodynamics 

The objective of desalination is freshwater as a liquid. If the water is separated by evaporation, it is necessary subsequently to recondense the vapour. Hence in principle, the latent heat of evaporation can be recovered. If indeed only the conservation of energy (first law of thermodynamics) were involved then the energy requirement of the distillation process would only be the heat of solution. But, alas, we cannot overlook the second law of thermodynamics, which tells us that a finite temperature difference is needed to transfer heat through a surface of finite area. Heat can be supplied to the seawater only by exposing it to warmer surfaces. 

Suppose we have a process where seawater enters a plant, is distilled and the only freshwater emerges, the latent heat of evaporation having been utilised in some way. If the freshwater leaves at a temperature ten degrees C hotter than the seawater which enters, then we must supply 10.67 kcal/kg. If the process is so efficient that we only have to supply 1.67 kcal/kg, then the freshwater will emerge one degree C above the temperature of the seawater. A process was so efficient as to produce fresh water at the same temperature as the seawater is impossible in a plant of finite size. The larger the heat transfer surfaces the smaller the temperature difference between the feed and product-water and the lower the energy consumption. 

Thus we see that the size of the heat transfer surface required in a distillation plant must vary in some inverse way with energy consumption. The capital cost will vary similarly, and the cheapest product-water may not be that obtained with the lowest energy consumption. The essential task of research and development in this field is not simply to reduce energy consumption but to lower the capital cost of reducing it. 

'Multi-stage Flash' Process 

This fact is well illustrated by the development of the distillation process. Up to the late 1950s, the lowest feasible energy consumption was about 110 kcal/kg. This figure was obtained with the 'multiple effect pool boiling" method (Fig. 1). Then at one stroke, the introduction of the 'multi-stage flash' process (Fig. 2) made it possible to achieve 52 kcal/kg for rather less capital cost than pool boiling required at 110 kcal/kg. 

The first plants using the multi-stage flash process were British and were built and installed in the ordinary way of commercial business before the American plant at San Diego, which was Government sponsored as a demonstration plant-not to demonstrate the process as such, which had already been done by the British plants, but to obtain comparable data in a series of plants designed to demonstrate the different processes. 

Up to the present, the lowest energy consumption achieved by multi-stage flash distillation is 45 kcal/kg again in a British-built plant. In most cases investigated, with present costs and design, the cheapest product-water is obtained at somewhere between 50 and 70 kcal/kg. Some new technical advance is needed if both energy consumption and capital costs are substantial to be reduced. 

Reverse Osmosis 

This accounts for the current interest in other desalination processes. For example, reverse osmosis, a membrane diffusion process, is claimed to show possibilities of energy consumption as low as six kcal/kg. In this case, however, the energy is supplied as electricity so that allowing for the efficiency of generating the equivalent heat input, for comparison with distillation, must be about 18 kcal/kg. Moreover, distillation can use 'waste' heat at a relatively low temperature (for example, from a power station), the cost of which is only the cost of making it available. 

Using 'spent' steam bled from a power station, the energy necessary to make good the loss from the station is only 40 per cent of the energy made available for distillation. Hence a distillation plant with a heat consumption of 45 kcal/kg from such extraction steam uses effectively only 18 kcal/kg of prime fuel energy-the same as that estimated for reverse osmosis. So the crucial consideration is whether the capital cost of a reverse osmosis plant will be substantially lower. A similar situation affects the various freezing processes that are being studied and for which the estimated energy consumptions are of the same order as for reverse osmosis. 

For these reasons there is much interest in the savings which might be possible by the construction of very large plants-a capacity of 400 000 000 imperial gallons per day has been mentioned. Since the largest unit built so far (a British plant) is 1 400 000 imperial gallons per day, this is quite a step. The engineering problems are formidable, even for distillation about which most is known. Nuclear reactors as a heat source for distillation also tend to appear more favourable when considered in conjunction with large units, and this partly accounts for the interest in them. 

While exact costs are affected by local site conditions and financial arrangements, the present position is that multi-stage flash distillation can if associated with power production, give water costs at the site of approximately 4/- per 1 000 imperial gallons-a very reasonable figure for household and industrial water, but rather too high for irrigation. 

Government Co-ordination 

Practically all the research and development which led to the British lead in distillation, and particularly to multi-stage flash, was done by the manufacturing companies. Recently, however, the British Government decided to step up work on desalination and to improve its co-ordination. Under Clause 4 of the Science and Technology Act, the Minister of Technology has directed the United Kingdom Atomic Energy Authority to be responsible to the Government for the development of desalination methods. This was the first directive under the Act, and a substantial programme is now underway involving the U.K.A.E.A., industry, universities and Government establishments. 

Initially, the programme will amount to over ￡1 500 000 shared between Government and industry. All work under this programme is coordinated through the Desalination Research and Development Committee which has wide representation. 

The U.K.A.E.A., together with Weir Westgarth Ltd., has completed a design study of a 30 000 000 imperial gallons per day plant based on present technology. The U.K.A.E.A. is also studying the geometry of flash chambers, the basic mechanism of flashing, improved heat-exchangers, the reduction of scale formation and corrosion, together with optimisation studies of combined nuclear power and desalination plants. 

Work on other methods of desalination, such as reverse osmosis and electro-dialysis, is in process at Harwell, Berkshire, while the National Engineering Laboratory, East Kilbride, Glasgow, Scotland, is studying the properties of brine. Basic work on improved heat and mass transfer is being carried out at the Heriot-Watt College, Edinburgh.