July, 1916 THE JOURYAL OF INDUSTRIAL AND ENGINEERING

THE JOURYAL OF INDUSTRIAL tions of turbidity in connection with the methylene-blue putresci- bility test can establish a fairly definite working relat...
0 downloads 0 Views 323KB Size
July, 1916

T H E J O U R Y A L OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY

tions of turbidity in connection with the methylene-blue putrescibility test can establish a fairly definite working relation. Tabulation of a large number of results shows that the stability is IOO with turbidities of I O parts per million or less. With a turbidity of r j parts per million the relative stability varies usually between 50 and 100. With turbidities exceeding I j the quality of the effluent shows rapid deterioration. With turbidities ranging from 20 to 2 5 the stabilities are usually less than jo. To compare turbidities exceeding 25 parts per million with stabilities appears unsafe; such stabilities are, without exception, very low. Recently, with the advent of warmer weather, nitrite nitrogen is again increasing in the effluent and no doubt the ammonia index will again serve the purpose. The determination of other constituents, such as the organic nitrogen, albuminoid nitrogen, permanganate oxygen consumed, chlorine, and the bacterial content are merely of scientific interest. They are not essential to routine control. This holds good to a certain extent for the dissolved oxygen, but it is quite conceivable that the rate of deoxygenation a t a given temperature can be made to serve as an index of the degree of stabilization accomplished. The quantity af the settling suspended matter in the final effluent merely indicates the efficiency of the settling process and has nothing to do with the activated-sludge process proper. Packingtown sewage completely oxidized by the acfivatedsludge process is clear with a slightly yellowish tint barely noticeable in small bulk. 39TH

ST. PUMPIKG

STATSOX

CHICAGO,I L L I ~ O S S

DEVELOPMENT OF T H E PURIFICATION OF SEWAGE BY AERATION AND GROWTHS AT LAWRENCE MASSACHUSETTS By H. W. CLARK Chemist and Director, Water and Sewage Laboratories Massachusetts State Department of Health

\That is now known as treatment of sewage by the activated sludge process has been in operation a t the Lawrence Experiment Station of the State Department of Health of Massachusetts since early in 1912. During a series of experiments made at the station in 1911 i t was found that the presence of certain algal growths in bottles of weak sewage caused a purification of this sewage. This happened even in sealed bottles and with the liberation of oxygen. It was found also that these growths, aided by forced aeration, effected remarkable purification. Immediately following this early work, i. e . , in the spring of I 9 I 2 , extensive experiments were begun in regard to what could be accomplished in the actual purification of sewage by aeration aided by growths. For several months this work was carried on in gallon bottles and carboys and we found that by 24 hrs.’ aeration of sewage containing growths of various kinds we could obtain a n effluent which was stable with nitrates a t times equal to I j parts per million and containing only 33 per cent as much organic matter as the untreated sewage. A shorter period of aeration gave clarification and stability but nitrates were not formed. The sewage mas emptied from the bottles daily, only the growths and sewage slime being left.z This work was shown to Dr. Gilbert Fowler of Manchester, England, in the fall of 1912, and was his first view of sewage purified by what is now known as the activated sludge process. On his return to England he and his colleagues, Ardern and Lockett, began similar work and when Xrdern and Lockett’s paper was published3 a year and one-half after his visit to Lawrence, Fowler’s statement in regard to this English work and its results was as follows: “ I t is only right to admit that the work was really due to a visit to the Experiment Station a t Lawrence where he (Fowler) saw sewage which had been completely purified by 24 hrs.’ aeration.” M a s s . S2ate Bd. Health Rept., 1912, 344-45. ? I d e m . , p. 291. 3 J . SOC.Chem. I n d . , 33 (1914), 18. 1

653

Ardern and Lockett state in the paper just mentioned: “In November, 1912, Dr. Fowler visited the States * * * *. Shortly after his return he described to the authors, work in progress a t the Lawrence Experiment Station on the purification of sewage in the presence of organisms. Dr. Fowler suggested that work might be carried on on similar lines.” As a result of the first few months’ work a t Lawrence it appeared that offering a considerable surface to which growths would become attached might favor the purification of sewage by this method; hence, a tank containing a few pieces of slate separated from each other a t intervals of an inch or more was put into operation late in 1912. Spongy gelatinous growths, brown and gray in color, soon covered these slate layers and sides of this tank, and, of course, sewage slime was also prominent.’ These growths, aided by aeration and the circulation of sewage in the tank by the air currents, collected not only the suspended matter of the sewage but also a large percentage of the colloidal matter. Oxidation occurred and again, as in the bottle experiments, stable and well clarified effluents low in organic matter were obtained. This Lawrence work to the end of 1912 mas summarized by Dr. McLean Wilson of England, in his recent address a t Manchester as President of the Association of Sewage Works Managers, as follows: “Many investigators, including Dromm, Dupre and Dibdini Mason and Hine, Black and Phelps, Fowler and others, had sought to purify sewage by direct chemical oxidation by means of air currents and had failed. At Lawrence, however, the efficiency of growths in the purification of sewage by aeration was discovered” * * * * and this changed the current of investigation along this line of work. During the past three years the work a t Lawrence has been continued both along the aerated slate-tank method and also by that method known more commonly now as the activatedsludge process, and various articles have been published by us in the Engineering Record and elsewhere in regard to it. Fairly complete accounts have been given also in the report of the Lawrence Experiment Station during recent years. More than a year ago an article of mine appeared in the Engineering Record comparing Lawrence and hIanchestcr (England) work along this line. Up to that time we had shown, as stated in that article, that it was possible by j hrs.’ aeration of sewage in the Lawrence tank, with the use of 50,000 cu. ft. of air per hour per million gallons of sewage treated (l:4 cu. ft. per gal.), to render the tank effluent stable during more than 70 per cent of the time. I O hrs.’ aeration with the same volume of air per hour per million gallons of sewage rendered the tank effluent stable 90 per cent of the time. By j hrs.’ aeration, 80 per cent of the total suspended matters were collected and removed from the sewage, and by IO hrs.’ treatment, 90 per cent. Fire hours’ aeration removed the soluble organic nitrogen, including colloids, to the extent of 35 to 40 per cent, and this removal was increased t o 6 0 per cent by IO hrs.’ aeration. The albuminoid ammonia was reduced about 6 0 per cent by 5 hrs.’ and 8 0 per cent by IO his.’ aeration. Statements have been made in engineering journals in this country and abroad that the Lawrence aerating tank containing slate is a contact filter. This is not true, however. I n a contact filter a t least 6 j per cent of the total space is filled with filtering material allowing 35 per cent for air and sewage. I n our aerating tanks, of slate colloidors not more than 3 to 7 per cent is filled with slate or other material, leaving 94 or 9 j per cent of space free for sewage and air. It has been stated, also, t h a t a t Lawrence the object of our work was to prepare a sewage f o r filtration. This is untrue, of course, for as mentioned above we produced by aeration as early as 1912, an effluent practically 1

M a s s . Stale B d . Health Rept., 1912, 292.

T H E J O U R S A L O F I N D U S T R I A L A N D ENGIXEERILVG C H E M I S T R Y stable and containing nitrates equal t o 1:j parts per million. We have a t times studied filtration of these effluents, however. It is not m y purpose in the present paper to review all the Lawrence work along this line but to give some comparative results obtained during the past six or eight months by the two ways of working the process followed a t 1,awrence. The comparisons are made with two aerated slate tanks containing sludge and growths, Kos. 449 and 449B, and two activatedsludge tanks, so called, Kos. 46j and 482. These tanks hold only I I C to zoo gallons of sewage, each. The results appear in Table I. From Jan. I O t o bIarch 29 (3-cycle period with one-hour sedimentation a t the end of each period in Tank No. 4651, there were periods when a digestion of sludge occurred and the albuminoid ammonia in solution in the tank effluent averaged TABLE I-COMPARATIVE

h-0.

449 449B 465 482

(a) (b)

slightly greater than 40' F. and activated-sludge tanks were operated a t different temperatures, that of Tank 465 averaging O F. and of Tank 482 averaging j j F. Little difference in results T w s notqd a t first a t these temperatures, but a period of digestion of sludge occurred during a portion of the winter in activated-sludge Tank 465 and much organic matter wect into solution. On the whole, except for this period of several weeks when the organic matter of the sewage in Tank 4 6 j was going into solution, the effluents from these two activated-sludge tanks operated a t different temperatures were about equal, showing a general clearing ol the effluent, and 80 per cent of the samples stable. The work of the slate tanks during the winter period were much better, showing removals of j o to 60 per cent of the organic matter and practically every sample stable.

RESULTS:.kEKATED SLATE TANKS(NOS. 449

AKD

449B)

AXD

ACTIVATED-SLUDGE TAKKS ( N O S . 465

CUBIC FEETOF AIR USED Period of Per Per million Date of Treatment Aeration Hour aals. Treated July 1 to Oct. 6. 5 hrs. 50.000 250 000 Oct. 6 t o Feb. 15 6 hrs. 50,000 300:OOO Feh. l 5 to Mar. 29 3 hrs. 250,000 750,000 45OF. (a) 10 hrs. ( b ) 500,000 42' F. Oct. 6 to Jan. 10 (Period of settling, 1 hr.) 250,000 3-hr. Sample 750,000. 8-hr. Sample 2,000,000 55' F. Jan. 11 to Mar. 29 (Operated same as 465) Tank 449B receives effluent from Tank 449. Nitrification to 2.5 parts per 1,000,000. Total aeration period including aeration in Tank S o . 449.

TANK

AV.

Winter Temp. 41" F.

66 per cent greater than the albuminoid ammonia in solution in the sewage entering the tank and 1 2 per cent greater than the total albuminoid ammonia in the sewage entering the tank. This phenomenon was a t first a rather disquieting element in this investigation but its causes were easily contro!led. If these figures are compared it will be seen that a t Lawrence the results are about as follows: The slate process, as exemplified by Tanks 449 and 449B.gives considerably better purification in I O hrs., calculated upon removal of organic matter, than activated-sludge Tank 4 6 j in 3 hrs. and practically the same as Tank 465 in 8 hrs. By the slate tank method only 66 per cent as much air is used in I O hours as by the activated-sludge tank in 3 hrs. and only 37 per cent as much as is used by the activated-sludge tank in 8 hrs. Both methods give stable effluents, generally speaking, the slate-tank method requiring generally more time to accomplish this result but less air and the efRuent of the activated-sludge tank being the clearer in appearance. During the winter, as shown in the tables, the aerated slate tanks were operated a t temperatures

T-01. 8. NO. 7

AND

REDUCTION(IN PERCEKTAGES) ALBUMINOID .hlhlO~IA Oxygen susFree To- Dis- Con- pended "3 tal solved sumed Matter

.. ..

STABILITY OF

EFFLUEXTS

Percentof SamDles Stable

40

46

57

65

78

..

..

..

..

100 100

59

88

86

80

80

11

49 63

$4

22

sx68

;;

48

49

."

482)

..

74 63

,

.

75

.. (3 cycles) (1 cycle) (3 cycles)

86 100 100

At Lawrence the sludge from both methods has lost the offensive characteristics of sewage sludge; it is more dense, that is, more easily drained, and more or less granular when dried. It is of greater agricultural value, not only on account of the changes mentioned and increased nitrogenous contents, but also because a large percentage of the fatty matters present before treatment is destroyed. The governing factors in the success of this process of sewage treatment, as I have stated in previous articles, are: ( I ) The cost of power for supplying the large volume of air necessary; ( 2 ) a sewage that readily yields itself t o this method of treatment. It is not impossible to believe that certain selvages cannot be purified in this manner. In conclusion, I wish to state that we cannot a t Lawrence work out certain points in regard to such a method, these needing experiments upon a larger scale. It is probably true, however, that the method as first carried on by us a t Lawrence, without the use of slate, is the more practical. STATEDBPARTXENT O F HEALTH,BOSTON

CURRENT INDUSTRIAL NEWS HOT MECHANICAL PULP All attempts to make hot ground pulp with the aid of old or weakly constructed apparatus are futile, says the Paper-Maker. I n one mill experiments have been made in this direction but failed and were even dangerous t o life owing t o the bursting of the stones which, for want of axles thick enough, could not stand the increased pressure. Since, however, new machinery has been built for making hot pulp, the manufacture proceeds steadily and without trouble. The advantage of the hot grinding process compared with others may be recognized by the fact that four or five mills were hardly able to take care of the resultant coa.rse stuff, whereas nom- a single mill handles it easily. For tissue papers of less than 18 g. per sq. meter weight and which must contain 2 5 to 30 per cent of mechanical pulp, i t is advisable t o allow the pulp-wood logs t h a t are to be converted into hot mechanical pulp to soak for some time before grinding in hot mater, The mechanical pulp thus obtained is exceedingly

feltable and like wadding, so that on open, as well as on automatic machines, steady working and a very strong product will result. For the production of cheap kinds of imitation parchment, such a previously warmed stuff is to be highly recommended. The pulp combines more intimately, is not so plainly visible and the sheet of paper possesses remarkable toughness.-hIcMILLAx. BEST OIL FOR DIESEL ENGlNES The best results with Diesel engines are obtained, says the Setional Petroleum A-ews, with an oil of between 20 and 40' Baume gravity. Oils with a higher gravity have a flash point too low for cheap storage and also do not furnish as many heat units per gallon, while an oil with a gravity below this is usually not fluid enough to be piped in cold weather without heating. Oils as heavy as 14' can be used if precautions be taken in heating, so that the oil can be readily handled and the lighter dis-