The cell that I am now using has a resistance of two ohms, and has been entirely satisfactory up to the present time. To check the accuracy of the instrument all that is necessary is to have at your disposal a Daniells cell. Should you have any reason to suppose that your instrument is out of adjustment measure the electro-motive force of your Daniells cell by it, and should it differ from the known value of the electro-motive force of this cell, add or remove from the adjusting coil a few inches of wire until you get the desired reading of your voltmeter.

From my own experience with the instrument I think this adjustment will seldom be necessary.

MEETING 399.

Electrical Purification of Sewage.

BY MR. FRANK M. GILLEY.

The 399th meeting of the SOCIETY OF ARTS was held at the Institute on Thursday, February 13th, at 8 P.M., Prof. T. M. Drown in the chair.

After the reading of the records of the previous meeting, the chairman introduced Mr. Frank M. Gilley, of Chelsea, who read a paper on "Electrical Purification of Sewage."

Mr. GILLEY said: The treatment of sewage was attempted over a century ago, and the chemicals employed at that time were almost identical with those in use today, sulphate of iron and lime. England has lead in this work. Her cities are so populous and near together that the introduction of systems of drainage removed a nuisance farther down the stream. Rivers, or streams as they would be less pretentiously called, became filled with sewage deposit, the fish in them died, and the general poor health and discomfort of those who lived

on or near the banks called attention to the absolute necessity of removing or abating such nuisances. The removal consisted in the application to land or the discharge of the sewage into the sea or farther down the brook or river flowing through the town, perhaps near some other town. In many cases the nuisance was abated, never removed entirely, by the use of chemicals, and the precipitation of most of the suspended matter, and the destruction or oxydation of a little of the organic matter in solution. Suspended matter must either be deposited in settling tanks as sludge or in some river or harbor as mud until dredged out. The organic matter is the most difficult to destroy or remove, and is the part most dangerous to health. The oxydation of the organic matter can be accomplished by one method only, chemical action, whether that be produced by filtration, the addition of chemicals, or electrolytic action. A filter is sufficient when employed intermittently, for then the filtering material becomes aerated, i. e., charged with atmospheric oxygen during the periods of rest, and this oxygen destroys the organic matter of the sewage. Oxydation can also be accomplished more or less expensively by chemicals. The electrical method differs from the above methods in two points; first, the oxygen and chlorine that produce the burning up or oxydation of the organic matter is made by decomposing, by the aid of a current of electricity, the water and chlorides of the sewage itself; second, the oxygen and chlorine act much more powerfully at the moment of their formation, being set free in what is called a "nascent state.

Mr. Wm. Webster, F.C.S., engineer and contractor for the construction sewage works, including precipitation tanks, during a series of successful experiments in the purification of sewage by electrolysis, settled at last upon two metals suitable for electrodes, aluminum and iron, the latter, on account of its cheapness, practical on a commercial scale. He first tried large tanks in which the liquid was treated and allowed to settle.

Strips of sheet iron are placed in a jar of sewage and a current passed through the solution. Hydrogen is given off from the plate connected with the zinc pole of the battery, and from the positive pole, i. e., the strip where the current enters, chlorine and oxygen are set free in a "nascent" state, probably combining with the iron to form a hypochlorite of iron which is immediately reduced to ferrous car

bonate and oxide. If there be no dissolved O in the water the white

oxide is produced. But in any case the color is soon green, and finally red, ferric oxide (FeO3). On a small scale, or where there is no current or agitation, the precipitate buoyed up by the hydrogen is brought to the top together with the particles in suspension, and finally sinks. If overtreated, the fitrate or effluent, as the clear liquid is called, has a reddish tint from the presence of ferric oxide. The solid matter, or sludge, has little or no liability to decompose, but, of course, must be disposed of in some way, ploughed into land, pressed and sold or given away as a fertilizer, or taken out to sea.

The E. M. F. between the plates is at least .9 volt when of iron, 1.5 when of carbon. In practice London sewage requires one ampere per gallon for 10 minutes, or less than ampere an hour per gallon. More current would supercharge the liquid with iron salts. A slight greenish tinge and evident separation of the suspended matter are the signs of sufficient treatment. The color on leaving the sewer is nearly white and opaque. After treatment the liquid is clear and filled with the particles of sludge, which deposit quickly when allowed to rest for a few moments. Owing to the resistance of contacts and the liquid itself, about 2 volts is the difference of potential of the plates, and about 1 ampere per 5.5 square feet of iron surface exposed. In larger units, 23 h. p. in 24 hours treat 7,000,000 gals. of London sewage.

It may be advisable to pass the effluent through an electric filter composed of alternate layers of coke carbon and sand or porous earthenware, the coke being electrically connected in alternate sections to positive and negative terminals of a dynamo. No increased power is required, for when the electrical filter is used the treatment with the iron plates is not carried so far. The same device is also applied to household filters for drinking water, the contamination of which with sewage or vegetable organic matter is always to be feared, and water of purity as regards organic matter and living organisms produced easily. A few open circuit cells furnish sufficient current which, on account of the resistance of the water, is small and does not flow at all while the filter is not in use.

Where brackish or salt water can be obtained by the means of a porous diaphragm, a disinfecting fluid containing 18 grains chlorine per ampere per hour is made. The positive plate must be of carbon, the negative may be of iron. If in the porous jar containing the car

bon a piece of iron be connected also, a hypochlorite of that metal is made and may be used for the same purpose as the chlorine alone. One-third grain of chlorine is found to disinfect one gallon London sewage. By automatic attachments such an apparatus is used in the household, and the liquid supplied and drawn off at intervals when needed for use. The ordinary Leclanche cells, five or six in number, will last several months and produce two gallons of chlorine solution daily.

Mr. Webster's experimental station is located at Crossness about 13 miles from London. The sewage is pumped into a shoot 18 inches square, 400 feet long, and filled with wrought iron plates in groups of 15, a large number of which connected in multiple or parallel form sections in series with each other. A space of two or three feet in the shoot between sections is sufficient to prevent undue leakage. The 70 h. p. Mather and Platt dynamo gives 20 volts. The six sections take approximately 3 volts per section, current 320 amperes. It would be economical as regards the loss in the conductors to have more sections in series, but the number is limited to 25 or 30 on account of danger to the workmen, it being impossible to avoid grounds, as both ends of the shoot have liquid connection to the earth at the inlet and outlet. The consumption of the iron plates is from 1 to 2 grains per gallon of sewage treated if cast iron is used. Wrought iron scales badly and is more expensive, but from its lightness is well adapted to experimental work. The velocity in the shoot is about 10 to 25 feet a minute, or from 4,000 to 10,000 gallons per hour. The color shows no apparent change for 20 to 30 feet, then numerous bubbles come to the surface, and farther on these have a brown color, and near the end the liquid is dark. At points in the shoot the current runs over or under an adjustable board, giving a thorough mixing and a complete control of the level in each section of the shoot. It is desirable that the outlet should be below the level of the liquid in the precipitating tank that it may settle rapidly and to a compact form. As the churning action in the shoot has liberated all of the hydrogen, the precipitate settles at once, and in two hours the clear effluent may be drawn off and discharged directly or through an electric filter into any convenient stream; or without settling it may be run on to land which does not become clogged, and the precipitate or sludge rapidly drying on the surface is easily worked into the land. The sludge as

taken from the tanks forms .7 per cent of the total sewage, and, after remaining in a settling tank, is reduced to .4 per cent of the original amount treated. It then contains 90 to 95 per cent of water, and must be disposed of in some way. Deprived of 50 per cent of this water by presses (and the sludge formed by electrical treatment is better adapted than any other for the press), and 20 per cent more by drying, it is worth, by analysis, $10 per ton, but in any case should pay for carting.

Cast iron is thought to have an advantage over wrought iron in the larger per cent of carbon it contains, which results in the production of chlorine. The size is limited only by the possibilities of manufacture. Six feet by 3 feet by inch plates are said to be made for other purposes by French engineers. The metal dissolves evenly, any projecting portions offering less resistance and being more acted upon by the current. Though only inch apart the plates showed no signs of blocking, and the entire shoot can be emptied in 15 minutes for cleaning.

but

A test at the time of my visit was in progress by the chemists of the London County Council, and the consumption of iron determined by weighing the plates before and after a long run, and analyses made of the crude sewage and effluent. It must be remembered that such a reduction of the organic matter is desired as shall abate the nuisance that exists in the river. Perfect purification, nor anything approaching that, is not desired. 70,000,000 gallons at Crossness, and 90,000,000 at Barking is a large amount to treat by any system, the attempt is being made by the addition of 3.7 grains of lime and 1 grain of sulphate of iron per gallon, and the effluent disinfected by permanganate of soda manufactured at the works. The discharge is supposed to be at ebb tide, but takes place practically all the time that the water in the river is low enough. The manganate of soda and sulphuric acid are added just before discharge. Notwithstanding this treatment, complaints of the condition of the river are continually made. The amount of chemical used seems insufficient, for other cities used 10 to 20 grains instead of 5. But for all that between two and three million dollars is being spent on immense precipitating tanks at both the outfalls, with the view of extending the chemical treatment to the entire sewage discharge. These tanks, with some changes, can be made to serve for the electrolytic process of treatment when that is finally adopted.