THE PURIFICATION OF SEWAGE BY BACTERIAL OXIDATION IN ARTIFICIALLY AERATED FILTERS.

BY GEORGE E. WARING, JUNR., M.INST.C.E.

(FELLOW.)

The agents which effect the disintegration of dead organic tissues have but recently been identified. It was long thought that the destruction was accomplished by the direct chemical action of the oxygen of the air, but it is now known that minute living organisms (bacteria) are the means of combining organic wastes with oxygen and reducing them to the condition of plant food.

These bacteria are universally present. In many minds they are associated only with disease, but the proportion of germs which are inimical to man is infinitely small; the great majority of them are not only harmless but beneficent, in fact, they are absolutely indispensable, for without them plant food would be quickly exhausted, and animal life would be buried in its own filth.

The processes of disintegration may be broadly divided into two classes-decomposition and putrefaction (with many intermediate forms of change). Both accomplish the same ultimate effect-the reduction of complex organic structures into their original elements. This result is reached by a succession of changes. In decomposition these changes are inoffensive, and not injurious to health; in putrefaction the intermediate products are disgusting and dangerous. The essential difference between decomposition and putrefaction is, that in the former the oxidizing bacteria can work effectively only in the presence of abundant air, while the agents of putrefaction require little or no air for their operations. As organic structure gives way under decomposition, its constituents combine with atmospheric oxygen to form carbonic acid, water and mineral salts. As similar matter is disintegrated by putrefaction, the changing elements, in the absence of air, recombine and form offensive compounds, some of which are poisonous. It is the latter process which produces the disgusting changes and probably dangerous conditions in sewage which has been stored in mass, as in a cesspool.

If the same sewage be spread in thin sheets over the surface of the soil, so as to be freely exposed to the atmosphere, its impurities will be attacked by the bacteria of decomposition, the elements set free will be oxidized and inoffensively reduced to mineral form, and the purified water will sink into the ground, to reappear later in springs or wells. As the water subsides, the air will follow it into the pores of the surface soil, so that any decomposable matter lodged in them will be attacked and completely destroyed. After a short period of rest,

another application of sewage may be made, and this will be disposed of in the same way. If sufficient intervals be allowed between the applications, the tract will continue to receive and purify sewage indefinitely; indeed, as the bacteria in the soil multiply under the regular feeding, its capacity will increase rather than diminish. This is the substance of the method of disposal known as Broad Irrigation. The purification is accomplished by a purely natural process. All that we need to do is to bring the sewage, at regular intervals, into contact with a suitable area of soil.

The practice of Intermittent Downward Filtration is identical in theory with that of Broad Irrigation, and the efficiency of the two processes is practically equal. There is, however, a difference in capacity in favour of filtration, due to a more careful provision and maintenance of conditions which promote activity of the purifying organisms. The filter-beds are composed of natural soil, carefully selected for its porosity and under-drained, or are artificially constructed of suitable material. In this process the purification is accomplished, not only on the surface of the soil, but under the surface; for, in a properly constructed filter-bed, the material is so porous that, as the applied sewage sinks away, the air which follows it can penetrate the mass and make bacterial oxidation possible at a considerable depth. By thus extending the zone of bacterial activity downward, the capacity of a given area is much enlarged; for the purification depends upon the exposure of the sewage for a certain length of time to the action of the oxidizing organisms, and it is obvious that, without reducing this period of exposure, sewage can be passed more rapidly through a deep layer of purifying material than through the shallow layer upon which the process of irrigation depends.

In both irrigation and filtration, purification is secured, not by the mechanical straining which the sewage receives in its passage through the soil, but by the exposure of the liquid to bacterial action, in thin films, upon the surfaces of the particles of filtering material which are in contact with the air. Wherever air can penetrate the bacteria can live and act.

The thorough aeration of the soil, which, in both irrigation and filtration, depends upon gradual absorption from the atmosphere as the water of saturation sinks away, is, necessarily, a slow process. The amount of oxygen available for nitrification is, moreover, limited; for, when once the pores of the soil have filled with air, the underground atmospheric circulation is so slight that fresh oxygen is not supplied to take the place of that which has been used up, and the gaseous products of decomposition are not carried away, but remain to hinder, to constantly increasing degree, the purification which is taking place.

In 1891 it occurred to the writer that the capacity of a filter-bed might be increased by artificially supplying the air needed for the stimulation and sustenance of bacterial action. It seemed probable that the use of air under pressure would not only ensure the introduction of oxygen to every part of the filter but would make it

possible to change its gaseous contents as often as might be found desirable. To determine the value of this theory, an experimental plant, on a practical working scale, was erected and put in operation at Newport, R.I., in 1894. In outline, the process consisted of the mechanical deposition in filter beds of all solid matters carried in suspension in sewage, and their subsequent destruction by forced aeration, and the purification of the clarified sewage by bacterial oxidation of its dissolved impurities in an artificially aerated filter.

The details of the construction and operation of this plant have already been published.* It is sufficient for the purpose of this paper to say that the results accomplished exceeded the most sanguine expectations. The sewage used (pumped from the main outfall sewer of the city) contained not only the fresh wastes normally present but the putrid overflow of many old cesspools, yet the liquid leaving the tanks was clear, white, odorless and tasteless. It was collected in a large tank, where discoloration would have been at once apparent, and in this tank fish lived and thrived. Engineers and committee-men drank of it freely and pronounced it good, and frequent chemical analyses proved it acually clean-a good drinking water. An average of the figures representing the purification accomplished showed that 92.5 per cent. of the organic matter was removed. At one time a removal of 99-08 per cent. was effected.

This complete regeneration continued through five months until the experiment was concluded. The filtering material was never renewed, yet when the tanks were taken apart it was found to be as clean and sweet as beach-washed gravel. There was absolutely no suggestion of the hundreds of thousands of gallons of sewage which had passed through it. The filth had completely disappeared. The complex organic matters had been broken up into harmless mineral elements, some of which had escaped into the air, while the rest passed out with the effluent water.

Plans have recently been prepared for a sewage disposal plant, capable of purifying 1,000,000 gallons of sewage per day to a drinking water standard by filtration with forced aeration. The illustrations

facing page 80 represent a half plan of the filters with their distributing and collecting channels, and vertical sections showing the construction of the filters, means provided for drainage and aeration, &c.

In order that the essential features of the plant might be shown in the simplest and clearest form, free from complications incident to the overcoming of unfavorable natural conditions, it has been assumed that sewage can be delivered to the works by gravity, and that there is sufficient natural slope to admit of gravity removal of the effluent leaving the filters. The pumping of sewage or effluent is obviously a matter of secondary importance, easily arranged by any

* "The Purification of Sewage by forced Aeration," which can be obtained, in pamphlet form, on application to the writer at Newport, R.I., U.S.A.

competent engineer when once the theory of the process by which purification is accomplished is understood.

The following is a description of the plant :

SCREENS.

Before entering the filter-beds, the sewage passes through a screening chamber, which consists of two channels, side by side, with suitable gates for turning the flow into one or the other at will. Each of these channels is furnished with two removable vertical galvanized screens, the upper one of rods, say 14 inches apart (centre to centre), and the lower one of inch square mesh netting. These screens withhold the larger floating substances which are not capable of rapid disintegration. The objects retained by the upper and coarser one will be chiefly orange and lemon rinds, brushes, towels, rags and similar articles of an inoffensive nature, which can be dried and burned. The accumulations of the lower and finer screen will probably contain more putrescible matter. Much that is detained as the sewage flows through will gradually be broken up and will pass on to the filters. The remainder, if found objectionable, may be burned.

APRONS.

From this chamber the sewage passes to the distributing channel which lies between the shallow beds or 66 aprons" of broken stone, to which it escapes through openings which can be closed at will by sliding gates. The aprons are paved with brick set in cement mortar, and are divided by low brick walls. They are covered with coarse broken stone, 8 inches deep, save for a space 3 feet wide at the foot of each bed, which is left bare to facilitate the occasional forking over of the stone. When this is to be done, a strip of stone 3 feet wide, next to the foot of the bed, can be turned over on to the empty paved space. Another strip can then be forked on to the strip originally occupied by the stone just moved, and the operation can be repeated until the entire bed has been turned over and the space at the upper end, next to the inlets, is laid bare. When it is found desirable to fork over the stone again, the procedure is reversed and the bed is restored to its original position. This operation is only necessary at considerable intervals, but it is very useful in that it breaks up and tears apart the fibrous matters which it is the duty of the apron to remove, and which are often slow to decay. At other times, the aprons are cleansed merely by drainage and exposure to the air. It is, of course, important that the sewage be free from silt, for if this gets into the system in considerable quantity, it must be removed mechanically.

The aprons are to be used in alternation, and it is recommended that only so many be used at once as are necessary to accommodate the flow comfortably. When these are choked the sewage is to be turned to another set of similar capacity. The method of using a few beds at a time, allowing them to choke quickly, and then draining them, secures the maximum percentage of rest and aeration.

STRAINERS.

The flow from the aprons is collected at the foot of the beds by channels which convey it to the centre of the group of strainers and there distribute it through gates similar to those which control the inlets to the aprons.

There are twenty strainers in all, to be used in four sets of five each. Each strainer is a brick tank, 40 by 40 feet, with concrete floor (slightly sloping towards the central drainage tunnel), upon which is placed a false floor of hollow terra-cotta floor-brick, laid with open joints, which is intended to act as a collector and drain for the water when the tank is in use, or has just been thrown out of use, and as a distributor of air when the tank is recuperating. Each tank is divided into two equal parts by a vertical diaphragm of brick, which is built upon the hollow floor-brick so that there is free communication, under the diaphragm, between the two halves of the tank. The half which receives the sewage is filled to a depth of 5 feet 4 inches with fine broken stone (to inch), or with coarse gravel of fairly uniform size. The other half of the tank is filled with the same material, but only to a depth of 4 feet 4 inches. The sewage, freed from its coarser solids by its passage through the apron, flows down through the deeper side of this tank, passes under the diaphragm through the hollow flooring, and rises on the other side, overflowing into a collecting channel which carries it to the aerators. The rate of flow through the strainers must be slow enough to allow the deposition of the suspended matters of the sewage upon the filtering material. A million gallons per day, passing through a set of five of the strainers provided, would travel at the rate of half an inch per minute.

When the flow is first turned into a straining tank the sewage passes through freely, and the water level in the receiving and discharging compartments is practically the same, i.e., the level of the effluent overflow. As the sludge gathers, however, the service of the receiving compartment will become more and more choked, and the sewage will gradually rise, acquiring increased head as the resistance increases. When the accumulation threatens to overflow the walls of the receiving compartment, the flow of sewage is diverted to another tank, and the drainage valves in the bottom of the first tank are opened, allowing its contents to escape-slowly, so that the deposits upon the filtering material will not be washed out-into the drainage tunnel shown in the centre of section G-H, and thence through a trap into the pump-well. (This well receives nothing but the drainings of the strainers as they are thrown out of use. It is emptied, and its contents delivered to the aerators, for further treatment, by a 4-inch centrifugal pump.)

As soon as the tank is drained, so that the valves are no longer trapped by the escaping water, air, introduced into the tunnel under light pressure (say 4 oz.) by the blowers, passes into the tank and rises through the filtering material, being evenly distributed by the open-jointed hollow brick-floor. An abundance of oxygen is thus