It has been recognised in many quarters that, although it is possible by modern methods of sewage disposal to secure a high degree of purification from a chemical point of view, it may be necessary in certain cases to take steps to remove the large numbers of bacteria present in all such effluents. Experiments have been made which demonstrate that, when a pure culture of some specific organism is added to a sewage in a sufficiently large quantity, it may pass through the tanks and filters, and appear in the final effluent. This result is not necessarily a conclusive proof that the bacteria in sewage effluents are dangerous, as the experiments do not represent normal practical conditions. On the other hand, it is true that, in ordinary practice, sewage effluents contain large numbers of B. coli, which is admittedly of intestinal derivation, and although this bacillus is not a disease organism itself, its survival in an effluent is considered an indication of the presence of sewage matter, and consequently of the possibility of the survival of any pathogenic germs which may be present in the crude sewage. On this basis, scientists argue that sewage effluents are potentially dangerous—that there is a possibility of the pollution of drinking-water or shell-fish by the bacteria present in sewage effluents. This being so, it is evident that an additional process will, in some cases, be required to remove the bacteria, and in a few cases sand filters have been provided for this purpose. These, however, involve a comparatively high initial expense, and a considerable annual outlay for maintenance, and in some quarters it is considered that sterilisation may possibly be a means of securing the desired result at less cost, and with a higher percentage of removal of bacteria, and consequently with a higher degree of safety. It is true that a number of scientists, in replying to a question which was submitted to them on this subject by the Royal Commission on Sewage Disposal, stated that in their opinion sterilisation was impracticable, but this was in the year 1903, and there is good reason for assuming that if the same question were put to the same men to-day, the replies would in many cases be modified, if not entirely different. Practical experiments in the sterilisation of sewage effluents have been few and far between in this country, but in the United States of America a large number of reliable experiments under varying conditions have been made, and the results published. From these it is evident that sterilisation is not only possible, but economically practicable. Unfortunately, both in America and elsewhere, attempts were made to sterilise crude sewage and tank effluents, and the results of these experiments were so unsatisfactory, both in efficiency and cost, that they gave the impression that sterilisation was impracticable.

In the opinion of the author, sterilisation should be restricted to the destruction of living organisms, and should only be used in the case of liquids with a high degree of chemical purity, and a low content of matters in suspension. It is quite possible at the present time to produce sewage effluents which comply with these conditions, by means of properly designed, constructed and managed sewage disposal works, as long as these include suitable effluent settling tanks for the removal of the solids in suspension. Effluents which comply with the provisional standard suggested by the Royal Commission, and drinking-water supplies which are very slightly polluted, would be very suitable.

A number of different methods of sterilisation have been tried, but so far as the existing knowledge of the subject extends, the application of chlorine in one form or another is generally admitted to be the most efficient and economical process. The chlorine may be applied in the form of a solution of chloride of lime, or as a hypochlorite of sodium, or of magnesium. The solution of chloride of lime may be prepared by the sewage works manager, but this necessitates a considerable amount of care and knowledge in order to secure the correct strength at all times, and it is, of course, essential that the chloride of lime itself should always be of a known strength. The hypochlorite of sodium may be produced chemically, and this can be purchased of known strength from chemical manufacturers. For large volumes of sewage effluent, however, it will probably be found most economical to utilise an electrically produced hypochlorite of sodium, as this can be prepared on the works as and when required, of a uniform strength, and at a comparatively low cost.

In the case of the Digby process, briefly stated, this consists in passing an electric current through a solution of sodium chloride, with the result that the sodium chloride is broken up into its component parts, and chlorine is liberated at the positive pole, while sodium is deposited at the negative pole. The sodium and the chlorine are then allowed to recombine in the form of hypochlorite, and this solution is then ready for application to the liquid to be sterilised. Other processes are similar in principle but vary in detail.

This process naturally involves the use of an electric current and an apparatus for producing the hypochlorite. Of the latter there are several on the market, and an illustration of the Digby Meridioniser, which is manufactured and supplied by Messrs. Adams Hydraulics, Ltd., is given in Fig. 153. The particular feature of this apparatus consists in the manner in which the re-combination of the anode and cathode products are secured. Instead of the re-combination taking place in the main body of the electrolyte, it can only take place in the special porous compartment enclosing the electrodes. The re-combination may take place in either the anode or the cathode compartment, the products of the one compartment being conveyed to the other compartment. Thus the caustic hydrate from the cathode cells flows by gravity into the anode compartment, the two compartments being connected by a glass pipe. Fresh water or water containing an excess of alkali is run into the cathode cell. This process gives hypochlorite solution of low saline content, the only salt present in the resultant liquor being that due to diffusion, depending upon the porosity of the closely covering compartment walls, or upon such a reaction as that covered by the Blount hypothesis. A cross-section of this apparatus is shown at Fig. 153, in which A is the positive lead; BB negative leads, C outflow, D inflow, EE cathodes, F anode.

Another machine is that supplied by Messrs. Oxychlorides (1907), Limited. This machine consists essentially of a graphite anode of circular cross-section (except as to about 4 in. at the top, which is left open for the free escape of gases evolved during the electrolysis), and within it a metallic cathode of smaller circular cross-section. The annular space between the anode and cathode is filled with a solution of common salt, or with sea-water, through which a current is passed from a low-potential dynamo. In either case the ultimate result is to obtain some of the chlorine of the salt in an active or available form, the only difference being that in the case of using strong salt solution, a concentrated form of available chlorine may be obtained, while with sea-water a weaker solution results. Where sea-water is readily obtainable it is naturally more economical to make use of it, and to employ the larger volume of less concentration; while where sea-water is unobtainable and salt is expensive, or where chlorides in the effluent are objected to, it is more advantageous to prepare concentrated solutions, and to dilute them when required for use. This machine is made in various sizes to suit varying conditions. It was used by the Royal Commission on Sewage Disposal in connection with their experiments at Guildford, referred to in their Report of 1908, pages 198 to 201.

Electrolytic hypochlorite of magnesium is being produced daily, by the Borough of Poplar, for use as a disinfectant. The highly effective qualities of this solution, as well as the low cost of production, makes it a very valuable disinfectant, and for the results obtained every credit is due to Dr. F. W. Alexander, the Medical Officer of Health, who initiated the scheme, and to whose unbounded energy and enthusiasm the success of the work is due. This solution is, of course, equally suitable for the sterilisation of sewage effluents. The cost of production of this solution, of an average strength of from 4·5 to 5 grammes of available chlorine per litre (·45 to ·5 per cent. solution), is estimated at under one penny per gallon. The apparatus in use at Poplar has been supplied by the Farringdon Engineering Co., and an illustration of the plant is shown in Fig. 154.

It consists of four cells, each containing ten elements, consisting of one positive and two negative plates. The positive plates consist of thin platinum wire, wound upon slate slabs, and the negative electrodes are of zinc. The four cells are placed one above the other, so that the liquid passes through from one to the other by gravitation. The feed-tank at the top contains a solution of sodium chloride and magnesium chloride, and is fitted with a glass gauge to indicate the amount of solution in the tank. From this tank the solution passes through a small ball-valve cistern, so as to maintain a constant rate of flow. The feed-tank is also provided with a plate, operated by a chain carried over to the outside of the tank, by means of which the liquid to be electrolysed can be stirred from time to time, so as to secure a uniform strength of solution. The solution passes through the four electrolysers in series, being subjected to the action of a regulated current of 15 to 17 amperes at 230 to 250 volts, being 5·7 to 6·2 volts per cell. After the electrolysed solution leaves the last cell it runs into a small tank, where a fixed amount of hydroxide of magnesium is mixed with it, and it is claimed that by this means the solution is rendered stable, a quality which should be of much value where the solution has to be stored for any length of time.

The question as to what is the most suitable sterilising agent to use under certain conditions, and in what proportion it should be added to the sewage effluent, is a matter for the chemist and biologist. The method of application is, however, the duty of the engineer. As in the case of other chemicals, there are two ways in which it may be applied. The solution may be added to the sewage effluent in equal doses of varying strength, or in varying doses of equal strength. There is a third method, involving the variation of the dose and the strength of the solution, but while this is not impossible it is probably not practical. The chief difficulty to be overcome is the variation of both the rate of flow and the strength of the sewage, and the most practical solution is to prepare the sterilising agent of a uniform strength, and vary the doses in direct proportion to the flow of the effluent, the minimum dose being sufficient for the maximum strength of the sewage. This method was adopted by the author in the case of a small scheme of sewage disposal, which he prepared for a place where the only outlet for the final effluent was a discharge over an area of chalk subsoil, from which the water supply of a large town is drawn. In this case, he designed a simple apparatus which does not involve any special appliances, and which would be quite satisfactory for ordinary practical purposes. A new apparatus for the purpose in question has, however, recently been introduced by Messrs. Nixon and Mannock. As will be seen from the illustration, Fig. 155, it is based upon the application of the Venturi principle, and involves the use of a Venturi tube, as previously described under the heading of “Measuring Apparatus.” In fact, the same Venturi tube can be utilised to serve both for measuring the flow of the effluent, and for applying the sterilising agent in direct proportion to the flow of the effluent.

The apparatus consists of a cylinder C, the top of which is connected by means of a pipe fitted with a three-way cock to the “Upstream” end of the tube A. A similar connection is made from the bottom of the cylinder to the “throat” B. A piston of the type used in the Kent Standard Water Meters, and provided with a counterbalance weight, works in the cylinder by means of the difference of the pressure on the two sides of the Venturi tube. The chemical solution (e.g. a 5 per cent. solution of chlorine) is supplied to the underside of the piston, and the pressure on the upper side of the piston being greater than the pressure on the underside, the chemical is forced down by the piston and injected through the injection tube and regulating valve into the effluent at the “throat” of the Venturi tube. As the flow of the effluent through the Venturi tube produces a difference of pressure which varies as the square root of the velocity, the rate of injection will also vary in the same proportion. The injection is thus in exact proportion to the flow, and any variation of the flow will automatically cause a corresponding variation in the rate of injection.

When the chemical re-agent is exhausted, the piston will be at the bottom of the cylinder, and the pointer at zero. In order to recharge the cylinder with the chemical, the three-way cocks must be reversed by means of the hand lever, thereby cutting off pipes A and B, and simultaneously connecting the top of the cylinder to the waste pipe, and the bottom to the supply from the chemical storage tank, which is fixed at such a height that the head will rapidly force the piston up and re-fill the cylinder with the chemical. The three-way valves are then reversed, and the apparatus is again in full working order. The apparatus shown is applicable to the treatment of 1000 gallons per hour, and will only need recharging once per day of 24 hours.

A feature of this apparatus is that it is self-starting, and should the flow cease, the injection will also automatically stop, the static head on both sides of the piston being equal. There is absolute immunity from danger or over-injection of the chemical by this system, and this is a valuable factor in the treatment of potable water. Where absolutely necessary, the same firm can supply a de-chlorinating apparatus. By means of the indicator, the works manager is constantly informed of the exact amount of chemical injected, and the scale readings can be compared with those of a Venturi Meter operated by the same Venturi tube. This apparatus can be supplied of a larger size, and provided with automatic recharging gear for larger installations.

Whatever method may be adopted for applying the sterilising agent, it is essential in all cases to have a storage tank to receive the mixture in order to provide time for the chemical to have full effect. So far as can be ascertained at present, a storage capacity equal to one hour’s flow of the liquid to be sterilised will be sufficient under ordinary circumstances, but provision should be made for thoroughly mixing the chemical with the effluent and for drawing off any deposit which may occur in the tank without interfering with the normal working of the plant.

Although the present volume is devoted entirely to the disposal of sewage, it may be stated here that in the matter of sterilisation, the suggestions that have been made apply with equal force to drinking-water supplies. Where the water contains a considerable amount of matter in suspension, it would be advisable to provide means for ample storage and settlement before passing it through the sterilising plant.

Note.—An apparatus for the injection of chlorine solutions for the purpose of sterilising sewage effluents and drinking water has recently been brought out by the Candy Filter Co., Ltd., and has for some months been in practical operation, dealing with 200,000 gallons of river water per day for an important municipal waterworks in the country. In this case the installation includes a de-chlorinating process, and it is stated that the results of tests in actual work show that the sterilised water contains neither B. coli nor free chlorine.