I COMPLETE REPORT of J. J. de KINDER On the Vosmaer System of Producing Ozone for the Purification of Water and other purposes. Owned by The United Water Improvement Company PENNSYLVANIA BUILDING, PHILADELPHIA Philadelphia George H Buchanan Company 1905 (o?.%.\G Yvvi c i Digitized by the Internet Archive in 2018 with funding from University of Illinois Urbana-Champaign Alternates https://archive.org/details/completereportofOOdeki UY M < Cl. cd m n c cr — • — o ° " 3 3 ^ P — On ft> “I VI 3 O ft) “ a On —j o ro C kQ cr c o EL ! + CZ- 2 ! rt to »> on p rc ° _ 5 orq c ^ p o c 3 p ir. r~t- o o o fD p 7 T“ a> o p ft 03 00 O o 5 « P ft O c p^ pH t-n O *'■*4 * S 2 ^ ro * n O P ° pp — r .2 p • O H a rr ^ « p ' rt -c 03 y rs o ‘ ft hj l-H >-< -P CC 'O Oj MO ■- to 1—1 1—1 w -p -P CO tJ W on « M K) O -P On 01 I tO (O ON OJ to Oj to 2 CC N S' » 3 cr. ft o 3 C 3 X orq 5:2 -•Orq O o N o > . QTQ p z 3 P zr.^ c F 3'? = Per Cent, of Re¬ duction M x ft P-* • 3 r» 3 W O p ft & O c p 3 * ft ft C S ft ►n C/1 cT Ui P 3 r* 3 P— r-f 3 * JO O C crq 3 * ft >1 ft CL P r-t- CD *-! The mechanical operation of the plant was in charge of the regular engineer, while the bacteriological observations were in charge of Dr. Lebret, samples of water for bacterio¬ logical tests and all observations being taken in my presence. / “From a paper read by Mr. Vosmaer before the Society of Hydraulic Engineers of Holland, convened at ’s-Graven- hage on September 12 and 13, 1903, on the sterilization of water by the use of ozone, a translation of which follows: Translated from the Report of the Fourth Convention of the Society of Hydraulic Engineers of Holland, con¬ vened at ’s-Gravenhage, on September 12 and 13, 1902.” The Sterilization of Water by the Use of Ozone BY A. VOSMAER. Gentlemen —About two years ago I had an opportunity to discuss before you the question of ozone production, but at that time I was compelled to limit the discussion to the question of production only, not being able then to elaborate the subject of sterilizing water by ozone, inasmuch as there was not at our laboratory in Haarlem opportunity to apply it to the purpose of sterilizing water, and as a matter of fact the question of the manufacture of ozone is and remains the principal one, because all speculations in regard to water sterilization by ozone amount to nothing unless ozone can be manufactured on a sufficiently large scale to make it applicable. Two years ago I necessarily limited myself to the dis¬ cussion of other systems, saying little about our own. (See sterilization of drinking water by ozone in De Ingenieur, organ of the Society of Civil Engineers of Holland, 2d and 9th of September, 1899, Nos. 35 and 36.) But we are now able to consider our system, because it has been thoroughly tested. It is not my intention, therefore, to lecture on the question of ozone sterilization, inasmuch as I would have to weary you with repetitions. No doubt the publications of Siemens & Halske are known to you, and you are, no doubt, acquainted with their method of original operation with tubular apparatus. The first publication of Dr. Erlwein in 8 1901 covered apparatus with glass plates, while the installa¬ tion at Wiesbaden in 1902 again involved the original idea from which it is justifiable to conclude that the change from tubular apparatus to plates was not a lasting improvement. In the technical arrangement of our apparatus we have aimed, first of all, to secure absolute reliability and extreme simplicity, and in this we have completely succeeded, al¬ though at the expense of a slight economical loss in quantity of production. The apparatus of Siemens & Halske pro¬ duces a larger quantity per unit. The production in grams of ozone per kilowatt hour is greater than in our system, but whereas they attempt to secure reliability (in Wiesbaden) by the application of a number of controlling and danger signals we have secured absolute reliability through another system, viz: through the absence of all dielectrics. In our apparatus there is no mica or glass upon neither of which reliance can be placed, and our special electrical arrangement prevents absolutely the creation of the danger¬ ous arc discharges. It is, therefore, a matter of personal opinion whether it is more advantageous to have absolute reliability with a somewhat more expensive apparatus than non-reliability with a somewhat cheaper one. Both systems, therefore, have their advantages and disadvantages. Our ozone apparatus, which many of you have seen at work at Schiedam, has been much simplified and been reduced to one-tenth of its size for the same production. The system, however, has remained the same, only the technical arrange¬ ment as now in existence at Nieuwersluis is simplicity itself, but inasmuch as the apparatus has worked at Schiedam and at Nieuwersluis without any interruption of the slightest kind day in and day out with constant results, it is not necessary to discuss the question of reliability any further. We are mak¬ ing our apparatus for twenty-four hours’ work per day and guarantee its reliability. The reason why I emphasize the question of reliability so strongly is because it is just in this reliability that our strength lies and which justifies our claim 9 of superiority over any other apparatus for which a claim of higher production per unit may be made. But I desire to call your attention to the subject of steri¬ lization of water, and in particular to the results which we have obtained at Schiedam and Nieuwersluis. When the City government of Schiedam asked us for figures for an ozone installation and its cost per cubic meter of water treated, we requested permission to erect our plant at Schie¬ dam in order that we might have an opportunity to thor¬ oughly test it, inasmuch as this could not be readily done at Haarlem. This request was granted and in consequence we were able to operate the apparatus for two and one-half con¬ secutive years at Schiedam. Following in the footsteps of Tyndall we contented ourselves during the first half year and during part of the second half year by operating the sterilizer in accordance with Tyndall’s system, viz: entering' the ozone into the water in the direction of the latter’s flow. The results obtained were not satisfactory, and although we operated just as Tyndall did in Brussels, viz: with 20 cubic meters of water per hour and 40 liters of ozone of 3J/2 grams 0 3 per cubic meter the bacteriological results were not satisfactory. This induced us, inasmuch as our system of ozone making was the reverse of that employed by Tyn¬ dall, to reverse the operation of sterilization also and not to attach any value to Tyndall’s publications, and from the moment that we introduced a sterilizer on the principle of countercurrent our success was assured. The sterilizator is arranged as follows : (See accompanying figure.) The water enters at 1 and passes out at 2, while the ozone enters underneath at 3 and passes out at 4 (or rather the air passes out at 4) for at that point the ozone has com¬ pleted its work and is again changed to oxygen, or what was not ozone escapes into the atmosphere. The apparatus works continuously and entirely as countercurrent apparatus, which is naturally the most economical: peculiar construction of the lower part makes it impossible for the water to enter IO A 4 into the ozone supply, and prevents the ozone from passing out with the water. The first is obtained by the pres¬ ence of the perforated plate 5. So long as there is sufficient pres¬ sure water cannot pos¬ sibly pass, on the prin¬ ciple of the bottom of a Bessmer converter. The second is obtained by the presence of the perforated wall 6 and the screen 7. Our ster- ilizator sterilizer has a capacity of from 20 to 30 cubic meters per hour, a diameter of 30 centimeters (12") and a height of a little over 10 meters (34'). You are all acquainted with the fact that techni¬ cally counter-current apparatus are the most economical, and that is just the case here. A single example is suffi¬ cient to prove this. Ac¬ cording to the reports of Siemens & Halske, to whom further refer¬ ence will be made, they used for the purpose of ozonizing one cubic meter of water -f- two and one- half grains of ozone, whereas we hardly need one gram because of our perfect sterilizer. Ozone does not affect bac¬ teria by absorption. Absorption apparatus are, therefore, not advantageous, because there the intention is to dis¬ tribute a maximum quantity of gas through a minimum quantity of liquid, while in ozonization a maximum quantity of water must be treated with a minimum quantity of ozone. The economical results are there to prove this. I will now proceed to show what results we have ob¬ tained with our apparatus. In Schiedam they had water twice filtered; first, through coarse sand, and afterward through dune sand. We were asked to treat with ozone the water that had been filtered once. The purpose was, there¬ fore, to substitute ozonization for the second filtration. The first filtered water, which was very good, contained, accord¬ ing to Dr. ’t Hoff, from 200 to 1000 bacteria per cubic centi¬ meter, sometimes less than 200. After ozonization the fol¬ lowing figures were obtained. Results. Date 2 3 4 5 7 8 9 io 11 16 17 18 IO M.. . . f O 0 0 0 0 1 1 0 I I 2 5 I I 0 1 2 0 1 1 0 1 0 1 1 15 M.. . . { I O 2 0 0 0 0 1 0 0 5 0 1 0 ■j a l 2 I "> 0 I 2 1 5 2 20 M.. . . ( 2 2 0 1 2 0 0 1 0 4 2 2 t 2 0 0 0 1 4 4 0 0 4 1 1 In this table of two weeks’ run the figures given represent the number of bacteria left after ozonization. It represents three series, viz : The ozonization of io cubic meters (-(- 2700 gallons), 15 cubic meters (-)- 4050 gallons), and 20 cubic meters (-j- 5400 gallons) water per hour, with one and the same amount of energy expended for making the ozone, viz., 2 kw. This is a very good result compared with the table of Siemens & Halske, where it is true the number of bacteria was larger, but where instead 2f4 grams of ozone were used per cubic meter. We were working, however, under more favorable conditions, inasmuch as the water had been once filtered, while Siemens & Halske worked with rough filtered water. In our case, therefore, because the ozonization was supplementary to that of filtration we obtained better water, but at a greater expense. We desired, however, to apply the ozonization just as Siemens & Halske did—as a substitute for filtration—and decided, therefore, to transfer the installation to Nieuwersluis, because the water from the Vecht is of worse quality than that of the Maas. Besides, this Vecht water is not filtered (so called) but merely rough filtered by a plain mechanical filter (system, Kronke). Whereas in an ordinary filter the speed of filtration is some¬ thing like ioo millimeters per hour (4'') in this case, viz: with the Kronke rough filter the speed was 4000 (40") milli¬ meters per hour. When the filter has worked for one day it is stopped and the current reversed, while the drum is made to revolve, by which operation the sand in the drum is cleaned. There is, therefore, no question of any biological effect and it involves simply mechanical filtration or sifting of matter in suspension. That more ozone would be re¬ quired for the ozonization of this mechanically filtered Vecht water than for the Schiedam water was to be expected. We limited ourselves not to 10, 15 and 20, but to 5, 10 and 15 cubic meters of water per hour with the following results : Treatment of 15 Cubic Meters (- 4050 Gallons) per Hour Date Serial Number of Experiments (9 each day) Average Number of Bacteria in Raw Water Organic Matter in Solu¬ tion in Raw Water in M illigram Permanga¬ nate of. Potassium Per Cent, of Reduction in Organic Matter I 2 3 4 5 6 7 8 9 Aug. 4 3^ 35 49 11 13 13 18 12 21 2 3 6,860 I. I I 26 11 5 2 I 28 2 3 M 2 3 24 14 15 19 20 4,410 IO. I 2 5 a 7 2 3 29 16 12 10 11 15 14 12 l 6 4,770 IO.3 20 << 8 11 »5 6 ! 3 6 IO 11 14 12 I I 5 , 59 ° io -7 24 k < 9 5 . 4 5 5 5 4 4 n 5 5 4,050 10.8 20 Average 15 I Treatment of io Cubic Meters ( 2700 Gallons) per Hour Date Serial Number of Experiments (9 each day) Number of Bacteria in Raw Water Matter in Solu- Raw Water in am Permanga- Potassium Per Cent, of Reduction in Organic Matter 1 2 a 4 5 6 7 8 9 Average Organic tion in Milligr nate of July 7 4 10 7 5 19 7 8 9 9 2,300 11. 3 3 ° i 4 8 10 4 4 0 2 9 5 7 7 2,400 n -3 26 4 4 10 9 8 9 5 7 10 7 4 6 7 1,960 11.2 26 4 4 11 16 — — 8 10 7 *5 12 2 3 13 1,670 10.9 23 4 < 15 6 10 4 5 — 8 4 6 6 4,950 10.3 26 4 4 16 7 8 10 5 8 6 3 5 10 7 95 ° 10.1 3 1 4 4 17 9 7 10 1 ■1 10 2 3 11 6 920 9.9 28 4 4 18 10 8 4 6 4 2 — — — 6 800 9.8 23 4 4 19 2 2 7 2 2 7 -» a 3 880 9.2 1 7 4 4 21 0 1 1 6 0 1 — — — 1 55 o 8 39 4 4 25 M 16 9 6 11 9 3 9 7 Average 9 7 3,800 9 22 Treatment of 5 Cubic Meters (+1350 Gallons) per Hour Date Serial Number of Experiments (9 each day) a; hfi r of Bacteria in Water c Matter in Solu- in Raw Water in ?ram Permanga- of Potassium :nt. of Reduction ganic Matter 1 2 3 4 5 6 7 8 9 A vera: Numbe Raw '5 <-> 3 0 £L 0 S ts G O UC U Z 4 The raw water varies in number of bacteria from 700 to 20,000 per cubic centimeter. The Yecht is rather more of a ditch than a river. It receives a great deal of Polder water and after heavy rains the number of bacteria is raised to over 40,000 and often to over 70,000 per cubic centimeter, while in very dry weather the number is reduced to under 700. The original number of bacteria has, however, little effect on the result obtained by ozonization. To complete the table I have added the status of organic matter. When instead of 5 cubic meters we take 15 cubic meters we obtain a less favorable bacteriological result, as might be expected. In the meantime I became acquainted with what is said in a publication of Kaiserliche Gesnndheitsamt , wherein Dr. Ohlmuller and Dr. Prall furnished a report on the “Die Behandlung der Trinkwassers mit Ozon” (the treatment of water with ozone). This publication was made by authority of the government, and has, therefore, the unquestionable stamp of reliability and impartiality. In the article referred to, we find among others the following observations: [Translated from the German.] “It would be an unjustified requirement to ask that a water ozonizing plant should furnish water absolutely free from all bacteria. In no known arrangement of central water purification is this condition aimed at, nor is it necessary considered from a hygienic point of view, etc.”(See Arbeiten Gesundheitsambt f Band 8, p. 159.) A reduction of bacteria to 100 per cubic centimeter is considered sufficient. This result is readily obtained by ozonization by which the bac¬ teria (in Spree water), after the latter having been rough filtered through a Kronke filter to arrest the material in suspension, was reduced from 83,700-86,800 to 20-28. A verdict like this, considering the origin, allows us to confine ourselves to aiming not at complete sterilization, but at nearly complete sterilization with absolute safety; but the great question, of course, is this: Those 20 bacteria 1 5 which are left over—what are they? May they not be the most dangerous ones ? Although it was well known through experiments on a small scale that pathogenic bac¬ teria or disease germs offer feeble resistance and are easily killed, Drs. Ohlmuller and Prall experimented on a large scale, viz : with io cubic meters of water per hour after infect¬ ing the water with the bacilli of typhus and cholera, viz: the water contained + 17,000 cholera bacilli per cubic centi¬ meter, 35,000 typhus bacilli and 33,000 coli bacilli. Twenty- four tests were made and in all 24 sterile results were ob¬ tained—not entirely satisfied yet, they mixed infected water with ordinary Spree water 1 to 1. Before the ozonization the total was 43,800 to 45,107 per cubic centimeter, after the ozonization 5 and 6. “In dem nicht ozonisirten wasser konnten die cholerabachterien nachgewiesen werden, in dem ozonisirten waren sie vernichtet” (“in the water not ozonized the cholera bacteria could be detected; in the ozonized water they were destroyed”). The same tests were made on a larger scale, with the result that ozonization reduced the number of bacteria from 38,330-16,590 to 8-9, respectively, hence their verdict. These investigations proved that the treatment of water with ozone, as it was done in Martinikenfeld, destroyed the bacteria of cholera and typhus, and it is especially to be observed that the generally more resistant typhus bacilli, despite the higher oxidability of the water, were destroyed just as readily as the less resisting cholera bacilli. It is generally known that ozone first attacks the organic matter and then the bacteria, but we succeeded in limiting the effect on the organic matter and to completely destroy the bacteria. This is shown in the latest table of recent date, as of August it, where the reduction of organic matter is in¬ considerate, while the bacteriological result is very good. i6 Treatment of io Cubic Meters (-j- 2700 Gallons) per Hour Date Serial Number of Experiments (9 each day) Average Number of Bacteria in Raw Water Organic Matter in Solu¬ tion in Raw Water in Milligram Permanga¬ nate of Potassium Per Cent, of Reduction in Organic Matter I 2 3 4 5 6 7 8 9 Aug. I I 3 3 4 0 1 1 0 1 2 2 95 ° _ _ “ 12 4 0 4 0 1 0 1 2 3 2 1,000 .8 28% “ 13 2 2 2 4 2 0 1 6 1 0 2 1,030 8-4 33 % Average 2 We are able to produce water which is practically sterile. This does not mean, however, that in thousands of cubic meters per day not a single bacteria will be found. This is practically impossible. When, however, nine consecutive tests showed only two bacteria per centimeter, as the table showed, we are justified in calling this water sterile. We are able to leave it to the choice of others whether they desire practically sterile water with only 1 or 2 bacteria per cubic centimeter or a water low of bacteria ozonized, instead of filtered. The first, of course, is the most expensive. Drs. Ohlmuller and Prall further reached the following conclu¬ sions : First .—By the treatment of water with ozone a remark¬ able destruction of bacteria is produced, and in this regard ozonization is in general superior to separation of bacteria by sand filtration. Second .—The cholera and typhus bacilli are destroyed. Third .—Viewed from a chemical point of view the water is affected only insofar that the oxidability is reduced and free oxygen increased, both of which mean an improve¬ ment of the water. >7 Fourth .—The ozone which is introduced in the water is of no importance from a technical or health affecting con¬ sideration, as it readily changes into oxygen. Fifth .—The treatment of it improves the water by mak¬ ing it colorless. Sixth .—The taste is not affected, etc. Primary advantages are: First .—That it is a clean mechanical arrangement se¬ curing* a perfect regularity in operation without interruption from ice, conditions of weather, etc. Second .—That in times of threatening danger by epi¬ demics the quantity of ozone can be increased or the amount of water handled decreased. Third .—The capacity of the plant can be readily in¬ creased as it simply involves a question of multiplying the units of apparatus. Fourth .—It allows greater freedom in the choice of sources of supply, and, therefore, enters into the question of cost of transportation. These questions can, of course, not be exemplified in figures. The transportation of a water supply from long distances may involve millions. Again the cost of the installation depends on the quality of the water and size of the installation. Mr. Vosmaer claimed the plant required a minimum of attendance and that it is practically self-operative. During the entire tests the only attention required was the keeping up of steam to workable pressure, and exam¬ ining once a day the oil cups on the movable parts of the pumps, etc. In other words, the plant practically ran itself and the result obtained (see table) unquestionably sustained all that has been claimed for the Vosmaer system. i 8 Remarks by Dr. van’t Hoff, bacteriologist for the De¬ partment of Water Supply of the City of Rotterdam, at meeting referred to : I believe that the absence of complete sterilization in ozonized water, as discussed by Mr. Vosmaer, is attribu¬ table only to the presence of spore bacteria which are more resistant than the pathogenic. That being the case it is justifiable to say of this water that it is so barren of bacteria that it borders on sterility. OZONIZATION OF DRINKING WATER: What ozone is, and why it is claimed to be efficient and reliable as a positive remedy against the introduction and spread of pathogenic germs, such as typhoid fever, cholera, etc., through drinking water, will be better understood by reference to what has been said on the subject by well known scientists. M. M. Pattison Muir, M. A., Fellow and Professor in Chemistry of Gonville and Cains College, Cambridge, Eng¬ land, in “The Story of the Wanderings of Atoms," says: “When oxygen is submitted to the action of the silent electric discharge a portion of the ozygen is changed into a sub¬ stance which is a more energetic oxidizer than ozygen. This substance converts mercury, at the ordinary temperature, into black oxide of mercury, a reaction which is not accom¬ plished by oxygen; when brought into contact with a solu¬ tion of iodide of potassium it produces oxide of potassium, iodine, and oxygen, this also being a reaction which oxygen does not effect. When oxygen which has been submitted to the action of the silent electric discharge is cooled to about —i8o° C. (—292 0 F.) a blue liquid is formed; if this is allowed to evaporate oxygen passes off as a gas, and the modification of oxygen called ozone remains, presenting the appearance of a very dark blue liquid, and boiling at—106 0 C. (—214 0 F.). Liquid oxygen boils at about — 181 0 C. l 9 (—-294 0 F.). Ozone is completely changed into oxygen by heating to low redness; the weight of the oxygen obtained is equal to the weight of the ozone used; the volume of the oxygen, however, is one and one-half times as great as the volume of the ozone. Ozone is one and one-half times heavier than oxygen, bulk for bulk; and, as a contraction of volume attends the formation of ozone from oxygen, ozone may be called condensed oxygen. WHAT OZONE DOES: It purifies air and when introduced into smoky rooms all smell disappears. It can he used with advantage for the purpose of purifying air in public buildings, theatres, club rooms, and in fact wherever crowds congregate. It may also enter into therapeutic use for curing first stages of con¬ sumption, and in many ways in manufacturing industries, for instance: the treatment of linum usitatissimum before it can be combed into flax can be greatly simplified and cheap¬ ened and the quality of the flax improved by the use of ozone, since it takes the glue from the raw flax and at the same time imparts an improved color to it; but its great field lies in the direction of purifying drinking water. The following is a translation of the article by Drs. Ohlmuller and Prall, referred to by Mr. Vosmaer in his address before the Society of Hydraulic Engineers of Hol¬ land, convened at ’s-Gravenhage on September 12 and 13, 1902: The Treatment of Drinking Water with Ozone By Dr. Ohlmuller and Dr. Fr. Prall When in the year 1891 Froehlich and his cooperators had shown the way of obtaining ozone in any quantity and degree of concentration desired from the oxygen of the 20 atmospheric air, the Imperial Board of Health determined to demonstrate the action of this substance upon bacteria. These inquiries have confirmed previous observations bearing on this question, and have proved especially, “that ozone has a most powerful destructive influence on bacteria present in water, if the zmter is not too strongly contaminated with dead organic substances; the result is the same when the quantity of lifeless organic substance is oxidized to a certain degree." On the strength of this result it was not too bold to hope that after the production and application had been further de¬ veloped it would be possible to make the application of ozone practically useful, especially with a view to the destruction of bacteria in drinking water. The first attempts in this direction were made by Tyn¬ dall and his collaborator Schneller. These attempts induced Tyndall to build a large experimental station at Ouds- hoorn near Leyden. There water taken from the Old Rhine was treated with ozone. Van Ermengem tested the experiments and stated a satisfactory purification of the water. Tyndall then exhibited at the Paris hygiene exhibi¬ tion in 1895 an apparatus with a purifying capacity of 2 cubic meters per hour. Since 1895 Abraham and Marmier have occupied themselves with the process, particularly with a view to the practical application of ozone. Otto and Gosselin and also Vosmaer have done good work in the perfection of the apparatus. Otto’s apparatus acts well, as was shown by the tests applied by Loir and Fernbach. In 1897 the process was exhibited at the Brussels Ex¬ position and this led to the construction of an installation at Blankenberghe near Ostend, with a capacity of 2000 cubic meters per day; soon after, however, it was given up again, as it did not answer the expectations. Another installation was built in 1898 by Abraham and Marmier, by order of tbe municipal corporation of Lille; its capacity was 35 cubic meters per hour; the water was ob- 21 tainecl from wells in Emmerin. A scientific committee appointed by the corporation submitted the installation to a thorough bacteriological and chemical test and found it to answer every requirement. In 1900 the process, according to the Abraham-Marmier system, was on view at the Paris exhibition; the installation was described by Krull. There is also an installation at Schiedam near Rotter¬ dam; it has been built on Vosmaer’s system and has a capacity of 20 cubic meters per hour. Messrs. Siemens & Halske in Berlin, who, as has been stated, first made ozone practically applicable, have exerted themselves, since the possibility of purifying drinking water by ozone was pointed out, to perfect the process by the con¬ struction of apparatus both for the production and the appli¬ cation of ozone. Weyl had an opportunity of testing the apparatus while still in construction. Guided by the various results obtained they constructed a fairly large experimental station with a capacity of 10 cubic meters per hour, at Martinikenfeld, on the northeastern boundary of Berlin. The object of this station was not only to give those in¬ terested in the process an opportunity to get acquainted with it, but also to demonstrate its practicability. The installa¬ tion has been improved in various ways by hygienic and technical specialists and by members of scientific societies. The experiments made by the firm itself at the station, from a technical point of view as well as with regard to the condi¬ tion of the water treated with ozone, have been dealt with by Erlwein in his treatise, “Trinkwasserreinigung durch Ozon nach dem System von Siemens & Halske (A.-G.) Berlin." Owing to the great interest that had been taken in the subject of purification of drinking water by ozone, the Im¬ perial Board of Health decided to make a series of experi¬ ments at the above station. 22 The arrangement of the Martinikenfeld station is clearly shown by the drawing below. Ozone is prepared by pressing air by means of an air pump into the drying apparatus. It consists of a 23 refrigerator, the hose of which cools down the air and consequently reduces its moisture. The air dried in this way passes into the ozonizer, a box in which four pairs of plates made of glass and metal alternately have been ar¬ ranged. Between these “blue glow discharges” takes place, produced by an alternating current of 10,000 to 15,000 volts tension. When the air passes through the spaces between the plates in which the discharges take place, its oxygen is partly transformed into ozone. The ozonized air then passes upward through the sterilizing tower. In order to bring about the action of the ozonized air upon the water to be treated the following arrangement has been made: By means of a pump the water (taken from the Spree) is raised to a basin for raw water; from here it runs by gravity through a Kronke filter (of which there are two), where the visible suspended matter is separated, and col¬ lects in a second basin. This filtered water is led from there to the sterilizing tower. The tower is built of brick, 5 meters high and one meter in diameter; the inside is cemented ; it is closed all around with the exception of the inlets and outlets for ozonized air and water; the top is walled up; the bottom, which is provided on one side with a discharge pipe for the water, is placed in a water basin. This arrangement makes it impossible for the air entering at the bottom to escape; it is forced to go upwards through the tower and then, as far as it has not been used up, to escape at the top through a discharge pipe and to return to the ozoniz¬ ing* apparatus. The water, however, runs in the opposite direction; it enters at the top and runs out at the bottom. Consequently a countercurrent system is applied. In order to make the ozone act upon the water under the most favor¬ able circumstances, the water is spread over the greatest surface possible. The ozonizing space proper is therefore filled with pebbles as big as eggs, which are arranged upon a frame. By means of a douche and a sieve the water is 24 divided into numerous jets, falls on the pebbles and drips down again, while all the time it is exposed to the action of ozone; after this it runs into the basin in which the tower stands, and on this basin running over, it is collected in a reservoir. In order to test the efficiency of the installation a num¬ ber of experiments were made. They were executed on the following plan: First the refrigerator was worked for one hour, so as to make it reach its maximum capacity for drying- air; then the production of ozone began. In order that the sterilizing- tower might always be in the same condition at the beginning of the experiment ozonized air and water from the Charlottenburg waterworks were made to run in counter- currents through the tower for half an hour or an hour. Then the real test began; the water to be treated was con¬ ducted through the tower; after half an hour samples of water were taken at regular intervals from the raw water basin, from the basin for filtered water and from the sterili¬ zation tower before the water had reached the ‘‘overrunning” basin; for this purpose a metal tube bad been put in the wall of the tower, the mouth of which tube was sterilized before the beginning of the test by means of a flame. The samples for the bacteriological tests were taken with the usual pre¬ cautions and then dealt with at once with a view to the determination of the number of bacteria. Of the raw and the filtered water 0.2 cubic centimeter or corresponding dilu¬ tions with sterile water, of the ozonized water o. 1 cubic centimeter was used for the culture. Plates with ordinary gelatine were prepared, in order to obtain results that could lie compared with those arrived at elsewhere, and also plates with a culture substance consisting of gelatine, Agar & Heyden’s culture substance, as it had been proved that in the latter the bacteria occurring in water developed in greater numbers. These numbers are therefore nearer to the truth. The plates obtained from non-ozonized water were, in accord¬ ance with Neisser's method, “counted” after two days; those 25 prepared with ozonized water after five days. All the plates were kept in a culture case at 22 °. Moreover the raw, the filtered and the ozonized samples were submitted to a chemi¬ cal examination; its oxidability, the quantity of ammonia, nitrous and nitric acids were determined, and in the middle of each test the quantity of residue at 116 0 was fixed. Finally the appearance of the water as to color and clearness and its taste were noted. The production of ozone during these experiments was managed in such a way that a concentration of about 3 to 5 grams per cubic meter of air was obtained; during each experiment the quantity of ozone produced was only subject to insignificant, inevitable fluctuations. The percentage of ozone was during the experiment determined every half hour before and behind the sterilizing tower. As a rule 30 cubic meters of air went through the testing apparatus; in two tests 40 centimeters were employed with the same ozone con¬ centration, in order to determine the result of this increase. The quantity of water treated with ozone varied from 5 to 10 cubic meters per hour. Spree water and a mix¬ ture of such water and water from the Charlottenburg water- works in various proportions were employed to vary the oxidability and the number of bacteria within certain limits, as former experiments had shown that just these properties of the water are decisive for the action of ozone. A glance at the condition of Spree and of mixed water shows the mini¬ mum and the maximum values determined by the examina¬ tion of these two: Oxidability in Milligram of Oxygen N itrous Acid Nitric Acid Number of Bacteria Source of Water Left after Evaporation Ammonia Ordinary Gelatine Gelatine, Agar and Heyden Food I. Spree Water 172.4—233.6 7.52 — 10 88 O M 1 1 c O o-Spur 35,700—185,800 48,000—41 2,400 II. Spree Water after Filtrat’n I u d 5,100 5,010 4,980 30.100 31.100 31,000 2.91 3.49 3.72 2.24 3.07 3.04 17.2 21.5 23.2 3 55 3.63 3.66 7.12 6.16 6.40 4 56 4.64 4.96 2.56 1.52 1.44 43,900 48 000 40,800 22 24 28 85.700 83.700 86,800 20 24 28 30. III. 11 1 o O iD d D ^ Jr G 7,560 7,200 7,500 30,300 31,000 29.600 4.13 4.42 4.41 3.09 3.43 3.37 16.5 19 0 17.5 3.80 3.86 3.84 6 96 7 04 7.04 5.44 5.44 5.44 1.52 1.60 1.60 36,700 36.700 35.700 27 26 26 70,400 83.700 84.700 23 25 22 2 IV. 11 l 3 D r* U D D p.| 5 5,100 5,100 5,100 30,100 30,800 30,700 3.13 3.24 3.42 2.42 2.60 2.77 18.5 196 20.6 3.75 3.42 3.43 4.88 5.04 5.20 4.00 4.24 4.32 0.88 0.80 0.88 6.550 6,950 6,750 o i 13,050 14,500 19,350 o O 1 5 25. IV. 1130 1 3 D £ Jx D •- D *-• G G 7,480 7,260 7,440 30.100 80.100 30,000 3 20 3.12 3.23 2.50 2.32 2.37 12.6 12.9 13.0 2.53 8.03 3.16 5.36 5.68 5.68 4.56 4.96 4.96 0.80 0.72 0.72 22,250 26,150 25,000 13 zer flossen 64,700 69,300 63,900 16 32 14 6. V. 11 1 3 o C >-> o .0 o ci *2 O.— ctf C/2-^> r-H 7,560 7,320 7,340 41,100 40,900 40,500 4 66 3.71 3.38 4.02 3.08 2.78 25.3 20.7 18.6 2.95 3.09 2.93 4.24 4.24 4.24 3.44 3.44 3.52 0.80 0.80 0.72 7,550 7,950 7,950 2 3 2 18,350 19,800 19,000 1 2 1 9. V. 12 yso 3 V C *H D .G D *-• rt *-» Qh«t-H G C/. ^ > 7,500 7.560 7,440 31.100 30,000 30.100 3.17 3.04 3.35 2.41 2.27 2.56 13.1 12.1 13.5 2.87 3 79 2.88 4 32 4.32 4.40 3.52 3.52 3.60 0.80 0.80 0.80 16.950 17,700 26.950 3 3 5 32,350 35,000 62,900 3 3 6 15. V. 11 1 3 D £ *-• D .G D J; rt U r- 1 TJ< ^ 7,200 7,200 7,320 30.700 30.700 31,100 3.50 3 71 3.78 2.77 2.94 3.08 14.9 15.8 16.1 2.78 2.92 2.98 4.40 4.48 4.48 3.52 3.60 3.68 0.88 0.88 0.80 6,750 9,000 10,000 5 2 2 16,650 19,700 21,550 11 5 2 24. VI 11 12 1 D r- J_, D .23 d I-l cj Cu-2 G r—i r-^ 10,200 9,800 10,100 40,800 40,000 40,000 5.45 5.18 4.66 4.25 4.01 3.41 21.8 21.1 18.5 4.43 4.28 4 61 5.76 5.76 5.76 4.80 4.80 4.80 0.96 0.96 0.96 16 850 17,350 19,400 !) 3 7 32,150 50,000 42,350 *) 7 11 27. VI. 11 12 1 D C , D •— ^ u (4 V r—( r-l ^ 9.800 10,400 10,200 30,ii00 30.000 31,000 5.90 5.57 5.57 4.37 3.98 4.03 18.1 16.1 16.9 4.23 4.20 4.28 6.08 6.00 6.00 4.88 4.88 4.88 1.20 1.12 1.12 10,000 10.200 8,900 6 7 6 22,250 25,300 27.550 9 9 11 f o n 29 was used; the oxidability, however, was lower, 4.32-4.40 and consequently in spite of the greater number of bacteria (16,950-26,950) only 3-5 survived. That in the case of experiment 3 so much ozone was used up by the dead organic matter, is proved by the difference between the oxi¬ dation coefficients before and after the ozone treatment; in this case it amounted to from 1.60 to 1.68, in experiment 10, on the other hand, only 0.80 milligrams. The unfavorable result of experiment 8 is to be ac¬ counted for in the same way; here high oxidability (5.36- 5.68 milligrams) and a great number of bacteria (22,250- 26,150) coincided with the small quantity of ozone em¬ ployed (12.6-13.0 milligrams). That the result depends to a greater extent upon a low oxidation coefficient of the water than upon a great number of bacteria, was shown by previous experiments; when the oxidability was greater, the destruction of the bacteria be¬ came more difficult; in distilled water it took place very quickly. With these new experiments, too, the oxidation coefficient played, generally speaking, the same part. Still it must cause some surprise, that in some cases in which both the oxidability and the number of bacteria were low, the result did not answer the expectations. So, for instance, in the first samples of water in experiments 4 and 11, showing a consumption of oxygen of 4.72 respectively 4.40 milli¬ grams, 5 bacteria survived each time out of 6200, respec¬ tively 6750. This number seems too high as contrasted with the result of experiments 6 and 11, when with an oxidability of 4.96 milligrams 15,300 bacteria were all destroyed with the exception of four. The quantity of ozone employed does not sufficiently account for this result. To 1 I.iter of Air, Milligram of Ozone To 1 T iter of Water, Milligram of Ozone Of this Used Up Milligram Experiment 4.. . 4.07 25.8 2.30 “ n. 3 - 5 ° T 4-9 2.78 “ 6. 3-72 21.9 3-43 30 From this it appears probable that these results are to be ascribed to the presence of bacteria of greater resistance. The question arises if these cannot be destroyed by increasing the concentration of the ozone. This is not the case; the above table concerning these three experiments shows that the consumption is not proportionate to the concentration; when the latter was highest (experiment 4) the consumption was smallest. The condition of the water, especially its oxi¬ dation coefficient and the number of bacteria and the kinds of these, is decisive for the result. A concentration of about 3.0-5.5 grams ozone in 1 cubic meter of air, as applied in these experiments, is sufficient. Still though in the case of the experiments in the Lille installation concentrations of 5.8-9.5 g rams were applied, some germs of bacillus subtilis survived. Abraham does not think it advisable to go below a concentration of 4-5 grams. When the bacteriological results of these experiments are compared with those obtained at Lille and Oudshoorn, the Martinikenfeld experimenting station has shown a satis¬ factory capacity. That in the former fewer or no bacteria survived, is to be ascribed to the condition of the water with which the experiments were made there. In Oudshoorn the filtered water from the Old Rhine contained 385 bacteria, the oxygen consumption being 2.5 milligrams; the well water of Emmerin near Lille contained 988-2200 bacteria; the oxidability being as low as 0.8 milligrams. In the water used for the inquiries here the number of bacteria varied between 5700 and 48,000, the oxidability being from 4.24 to 7.12 milligrams. In Oudshoorn perfect sterility was obtained, in Lille specimens of bacillus subtilis occasionally escaped; in these experiments 1 to 28 bacteria survived (on gelatine plates). In spite of the water being of inferior condition, it has often been observed that one of the two plates that sup¬ plied the above averages was sterile ; this was the case with experiments 5, 7, 9 and 10. Even though the bacteria destroying action of ozone c3 O upon water is so considerable, still it is not to be avoided that some bacteria survive. This is owing to the greater resistance of certain bacteria; still the oxidation coefficient of the water is to be taken into account as well. It would be unreasonable to expect a water ozonizing installation always to supply water perfectly free from bacteria; this has not been obtained by any of the central water purifying methods now known. Nor is this necessary from a hygienic point of view; in the case of sand filtration, the method gen¬ erally applied for the purifying of surface water, a decrease of the number of bacteria to ioo per cubic centimeter is con¬ sidered sufficient. This object is reached to perfection with the ozone system, for it has been proved to he possible to reduce the number of 83,700-86,800 bacteria in Spree water roughly filtered through a Kroehnke filter in order to re¬ move suspended matter to 20-28. It may be emphasized that the water was taken from the Spree below Berlin, after it had been exposed to many forms of contamination on its course through this city. When the ozone method is employed—as well as in the case of sand filtration—the most favorable conditions of the “raw” water will of course be preferred, and even on esthetical grounds the pumping station will not be established below a city; here it was done only by way of experiment. By sand filtration the reduction of the number of bac¬ teria is brought about mechanically; the ozone method does this chemically. Experience shows that sand filtration is so efficacious that if the filter is properly attended to the danger of infection by purified water is reduced to an allowable mini¬ mum. Still C. Fraenkel and C. Piefke have proved that under certain circumstances pathogenic micro-organisms pass through the filter; all the same many years’ experience has shown that the use of water purified by sand filtration cannot be objected to. As regards the ozone methods, the behavior of patho¬ genic micro-organisms has still to be ascertained, as it has been proved that bacteria with greater resisting capacity 32 were not attacked under certain circumstances. Especially the cholera and the typhus bacillus are to be considered here as being causes of infection. In order to make preliminary researches experiments were made in the laboratory with a small apparatus of the same construction as those in the station. In order to render the capacity of this laboratory apparatus comparable with those at Martinikenfeld, the supply of air and water was arranged in such a way as to bring about the same reduction in the number of bacteria in a mixture of Spree and waterworks water in equal proportions as was ascertained in the experimenting station. With this apparatus water was treated containing cholera and typhus bacteria (for the sake of comparison coli bacteria were also employed), obtained from 1-3 days’ old agar cultures in ster¬ ilized Spree water at 30° or in a mixture of this water with equal parts of waterworks water; mixtures of water were employed to obtain various oxidation coefficients. The bac¬ teria in question were introduced in such great quantities that numbers of germs were obtained as never occur in ordinary impure water; this was done partly to force the experiment, partly with regard to the fact that the action of ozone is influenced by the number of germs and the oxidability. Be¬ fore and after the ozonization two gelatine plates were pre¬ pared with the water each time; the former were counted after two, the latter after ten days. The results were as follows: a. Bac. Cholera in Spree Water, 16,350 per Cubic Centimeter. Incuba¬ tion 1 day at 30°C. M ill igram of Ozone per Liter of Air Oxydisability Milligram of Oxygen Per Liter Reduction in Or¬ ganic Matter Number of Bac¬ teria after Ozo¬ nization Before Ozonization After Ozonization 5-49 7.92 6.88 I.04 O 4-53 7.92 6.96 O.96 O 3-45 7.92 7.04 0.88 O 2.88 7.92 7.12 0.80 O b. Bac. Cholera in Mixture of Spree and Main Water, i : i 17,850 per Cubic Centimeter. Incubation 1 day at 30°C. 5.19 5.72 5.°° 0.72 0 4.41 5-72 4.88 0.84 0 3-39 5-72 5-°4 0.68 0 2.61 5-72 5-04 0.68 0 c. Bac. Typhus in Spree Water, 39,050 per Cubic Centimeter. Incuba¬ tion 1 day at 30°C. 4 89 8.40 7 - 3 6 1.04 0 4.20 8.40 7-52 0.88 0 3-54 8.40 7.60 0.8S 0 2.73 S.40 7.76 0.64 0 d. Bac. Typhus in Mixture of Spree and Main Water 1 : 1 31,400 per Cubic Centimeter. Incubation 3 days at 30°C. 5 - 3 i 5.60 4.64 0.96 0 4 - 5 6 5.60 4.72 0.88 0 3-24 5.60 4.68 0.92 0 2-49 5.6° 4.76 0.84 0 e. Bac. Coli in Spree Water, 30,600 per Cubic Centimeter. 5.28 7.84 6.88 0.96 0 4 - 5 6 7.S4 6.96 0.88 0 3-42 7.84 6.88 0.96 0 2.49 7.84 6.86 0.88 0 f. Bac. Coli in Mixture of Spree and Main Water, 1 : 1 36,000 per Cubic Centimeter. 5 - 3 i 5-84 5 °4 0.80 0 4.44 5-84 5-°4 0.80 0 3-30 5-84 5.20 0.64 0 2.64 5-84 5-24 0.60 0 34 Against these experiments it may be objected that they do not irrefutably prove the destruction of the bacteria, as there is a possibility of their having their power of developing in gelatine through the action of ozone. Therefore two further experiments were made under equal circumstances as regards the passage of air and water, while cholera bacteria were added to the non-sterilized water, and their presence was ascertained by the “Enriching” method (by a peptone and salt solution). The results were as follows: Kind of Water V £< w <*_ u- c E $ Milligram of Oxygen Used to Oxidize i Liter of Water ion in Or- | Matter Total Number of Bacte¬ ria in One Cubic Cen¬ timeter OX) u ^ D ~ a Before Ozonization After Ozonization Reducti ganic Before Ozonization After Ozonization Spree and Main Water 3 48 6.16 5.28 0.88 45.>70 6 1 in 1. 3-54 6.22 5.28 O.96 43,890 5 These two experiments with the small apparatus show the same result as was observed in the great experimenting installation at Martinikenfeld, viz: a decrease of oxidability to the number of germs in the same degree. The capacity of the small apparatus was consequently properly adjusted in comparison with the large installation. “During the experiments six samples of 90 cubic centi¬ meters were taken from the ozonized water, and tw T o from the water not yet treated, and then submitted to the enriching method ; in the latter cholera bacteria could be detected, in the ozonized water they had been destroyed.” After these two preliminary experiments two definitive ones were made in the large ozonizing apparatus, in which cholera or typhus bacteria were added to a measured quantity of water. Measures were taken to prevent the spreading of 35 the pathogenic bacteria. The passage of the ozonized water from the “overrunning" basin to the reservoir was walled up. The connecting tube between the Kronke filter and the basin for filtered water was removed, and the joint screwed up; under the cock A' a basin containing a sublimate solution was placed to disinfect any water that might drip off. The cock, however, proved to be so perfectly tight, that no water dripped off. The quantity of water to be infected was made so great that it could be contained in the “overrunning’’ basin. This made it possible to sterilize all the water em¬ ployed ; this was done by making it boil with steam. For this purpose a pipe was laid from a boiler carrying a steam pres¬ sure of io atmospheres to the “overrunning” basin and the basin for filtered water. The arrangement of both experiments was as follows. First the basin for filtered water was filled with i cubic meter of water (a mixture of Spree water and water from the waterworks). For the cholera experiment it was used with¬ out any further preparation; for the typhus experiment it was heated to the boiling point by means of steam, in order to facilitate any difficult typhus diagnosis through the de¬ struction of the water bacteria. As appeared later on, the sterilization of the water was not perfect; perhaps the heat was not applied for a sufficient length of time; certain it is that during the cooling, which was brought about by leaving the water during the night and letting a piece of ice float in it, germs got into it. To the water in its natural condition cholera bacteria, to the boiled water typhus bacilli were added. Both cultures (agar) were i day old and had grown at 37 0 . Cholera bac¬ teria were added in such a quantity that accordingtoestimate there were 2000 in every cubic centimeter; as to the number of typhus bacilli the determination of the number of germs on the gelatine plates fixes it approximately. In order that 3 6 the pathogenic bacteria might be evenly distributed, the water was stirred vigorously. In the meantime the ozone apparatus was put in motion to make the refrigerator reach its maximum drying capacity and to obtain the desirable concentration for ozone. When the apparatus was in full action, the infected water was passed through the sterilizing tower. During the process samples of water were taken to- determine the oxidability and the number of germs, and others for the direction of patho¬ genic bacteria, viz: two samples before treatment, and ten of the ozonized water, which in the case of cholera amounted to 180 cubic centimeters, in the case of typhus to ioo cubic centimeters. When the experiment was finished, the basin for filtered water was filled with 1.75 cubic meters of water. Then steam from the boiler was conducted into this basin and into the “overrunning” basin; in the latter the steam tube was next pushed into the sterilizing tower, to make the water in it boil; then the water outside in the “overrunning” basin was treated in the same way. When the boiling heat had been kept up in both basins for half an hour, the water in the basin for filtered water, while kept at the same temperature, was allowed to run into the “overrunning” basin, and in order to sterilize the connecting tube to the sterilizing tower the latter was rinsed with 10 liters of alcohol. In the “over¬ running" basin the boiling heat was kept up for another one- quarter hour, and then the contents were allowed to run into a freshly dug sandhole iW meters deep, after which the hole was filled up. There are no wells in the vicinity of the hole. In order to sterilize the inside of the tower, ozonized air was passed through for one hour. By such precautions spreading of the bacteria was entirely excluded. It may be stated that during these two experiments the production of ozone was conducted by the Engineer Friberg; the infected and the ozonized water was not touched bv any one besides ourselves. 37 Of the results of the experiments we will first state those bearing upon the general inquiry: Quantity of Water per Hour in Cubic Meters Quantity of Air per Hour in Cubic Meters Quantity of Ozone Grams] per Cubic Meter Oxydisability of Water Expressed in Milligram of Oxygen Used Up per Liter of Water Oxydisability of Water Expressed in Milligram of Oxygen Used Up per Liter of Water After Ozonization Reduction in Oxydisa¬ bility Total Number of Bacte¬ ria per Cubic Centi¬ meter Before Ozoniza¬ tion Total No. of Bacteria per Cubic Centimeter After Ozonization Number of Bacteria, Cholera : 1 Part of Spree Water . 2 Parts of Main Water . 7 oc 3-76 4.64 4.00 O.64 38,330 8 Number of Bacteria, Typhus : 2 Parts of Spree Water . i Part of Main Water . 7 3 § 3-79 9 - 3 6 8.l6 1.20 16,590 9.2 Both experiments proceeded quite regularly; they had the same result as previous ones, viz: a decrease of oxida- bility and a considerable lessening: of the number of perms in the water. It is specially remarkable that in the typhus experiment there was such a high oxidation coefficient as had never before been observed. This makes the experiment all the more important, as the number of germs, too, though smaller than that in the cholera experiment, cannot be called small. The samples taken for the detection of pathogenic germs underwent a further treatment. Those obtained from the cholera experiment were, after the addition of pepton and salt, treated by the “enriching" method. Some of the “en¬ riched" water was spread on agar plates. From these the colonies suspected to consist of cholera bacilli were removed by oculation for further treatment by gelatine plates, cholera red reaction being made use of. The original cholera pure culture with which the water had been infected served as a basis of comparison. The detection of cholera bacteria in the infected, non-ozonized water was possible in the case of both samples through gelatine culture and nitrosoindol reaction. The ten samples of ozonized water were treated in the same way; the cholera red reaction, however, did not appear in any of the cultures that had been oculated as being suspected of containing cholera bacteria. The detection of typhus bacilli was also introduced by an “enrichment” process, viz: by the addition of bouillon in equal quantities of the water to be examined, and by leaving the samples for 18 hours in a tem¬ perature of 37 0 ; after this some was spread out on agar plates. Out of the typhus-like cultures agar pure cultures were prepared, and these were composed with the original typhus culture which had served to infect the water, as well as with a pure culture of bacteria coli commune by the usual methods of typhus diagnostics. From the infected, noil- ozonized water cultures were obtained, which in colored prep¬ arations, in suspended drops, by their growing on gelatine, agar and potatoes, by their behavior in sterile milk, grape- sugar solution, litmus and Maassen’s non-albuminous culture fluid, and especially through R. Pfeiffer's immunity reaction proved to be typhus bacilli. On the other hand, the 26 pure cultures (suspected to be typhus bacilli) obtained from agar plates out of ozonized water, proved on the above methods being applied, not to be typhus bacilli. “These two experiments, therefore, have proved that by the treatment of water with ozone in an installation like the one at Martinikenfeld the bacteria of cholera and typhus are destroyed. This result is confirmed particularly by the fact that generally speaking typhus bacilli with a greater power of resistance, in spite of the high oxidability of the water, were destroved as surely as the less resistant cholera bacteria.” The Result of the Chemical and the Physical Examination It must be required of any method that is to make any kind of water, subterraneous or surface water, fit for use, that the injurious ingredients of the water are removed or reduced to an allowable minimum, that the dissolved substances which 39 we consider an advantage in good drinking water, are not appreciably changed, and that no foreign substances which have an injurious influence upon water used in food or as a beverage, are added to the same. In what manner the ingredients that are not dissolved are influenced by the ozone process, has already been shown. The visible suspended matter is removed by the Kronke filter, the bacteria are with the exception of a few, destroyed by the action of ozone. It is true, it has not been proved that the cell substance of the dead bacteria is completely destroyed by oxidation; still this amount of organic matter even in water that is very rich in bacteria, is of no importance what¬ ever on account of its small quantity. As to the influence on dissolved substances the chemical analysis of the water gives the desired information; the values obtained varied between the following limits : I liter of water contained milligrams : Solid Matter 1 Organic Mattel (Oxygen) Ammonia Nitrous Acid Nitric Acid Filtered Spree Water . 181.4-204.2 6.16-7.12 0.2-0.3 O 0 to trace Mixed Water 232.8-330.0 4.24-6. 16 Trace to 0 24 O 0 to trace Ozoniz’d W’t’r 181.2-330.0 3 - 44 - 5-44 1 “ “ 0.24 O 0 to trace The residue figures show that the weight of the dis¬ solved, non-volatile ingredients of the water has not been altered by the action of ozone; they vary within the same limits in the ozonized water as in Spree and mixed water. On the other hand a considerable decrease of oxidable substances has taken place; this means an improvement of the water. Taking the great oxidizing power of ozone into considera¬ tion, it must cause surprise that the nitrogen compounds have been influenced so little; it might at least be expected that free ammonia would be oxidized into nitric acid. Special experi¬ ments throw light on this point; they were executed with the small laboratory apparatus; solutions of various concentra¬ tions in distilled water being used. 40 Froehlich had already shown on former occasions, that in air being ozonized small quantities of nitrous and nitric acid—depending on circumstances—are formed from the nitrogen. This is accounted for by the fact that on an electric spark passing through nitrogen and oxygen nitric acid is obtained. By preliminary experiments it was possible to ascertain that in the ozone apparatus a small percentage of the atmospherical nitrogen was changed into nitric acid; lower stages of oxidation, especially nitrous acid, did not come into play. The result obtained was as follows : 'ime taken for Experiments in Seconds Quantity of Air in Liters Quantity of Ozone in Milligram per Liter Total Quantity of Total Quantity Ozone IVi illigram of Nitric Acid 455 20 3.00 60.O 7.28 540 20 3-96 79.2 8.78 490 20 8.4 168.0 16.80 230 IO IO.44 IO4.4 12.19 In order to obtain a clear idea of the action of ozone upon the ammonia present in the water, the nitric acid given off by the air was extracted from the ozone by means of con¬ centrated sodium lye. The formation of nitric acid in water was so insignificant in these experiments, that a quantitative determination was not worth while; only the shades of blue of diphenylamine-sulphuric acid could be fixed. In the table below the presence of nitric acid is indicated as follows: By o + + H—I—b + + + no trace, hardly a trace, very slight trace, slight trace, a trace. 4 1 The result was as follows: Free Ammonia t/3 W v S .£ ES s a u 2.30 100 60 45 20 50 3 3*50 100 75 5° 3° 50 7 j 4-3° 100 80 55 35 50 7 d) 6.00 100 100 75 45 Combined to Sodium Per Cent. Oxidized to Nitric Acid 5o 3 2.80 100 95 85 5o 3 3-5° 100 100 94 5o 3 4-3° 100 100 100 5o 3 6.00 100 100 100 Here as opposed to the result obtained in the case of ammonia, it is to be observed, that the oxidation is greater when the dilution is stronger. Further it is particularly striking, that of the nitric acid in compound from a greater quantity is oxidized than of free nitric acid; this fact is ac¬ counted for by the alkaline reaction of the nitrous sodium. For it has been proved that the action of ozone upon oxidable substances is more powerful in the case of alkaline, than in the case of acid and of neutral reaction of water, j-. . O a . u ci o 5 G r± .2 c 1 < 50 4-3 3 50 3-o 7 0 5° 3-o 3 5o 3-o Milligram of Oxygen Required to Oxidize One Liter of Water Organic •g I ci N • —> N O N I u divide with them. These suburbs were en¬ tirely unable to secure a supply of spring water of their own. The next best thing available to them was to use the river water with filtration, and filters were established at several points by the company supplying them; and these filters were put in operation at about the same time that Paris secured sufficient spring water. “There were then these two bodies of population; Paris the inner city, supplied entirely with spring water, and the ring of suburbs outside supplied by filtered river water; and it was soon noticed that typhoid fever was much less preva¬ lent in the suburbs than it was in the city. Thus when the two kinds of supplies were put to the ultimate test the filtered river water proved better than the spring water. For the last few years a commission composed of many of the lead¬ ing scientists and engineers of Paris have been considering this question, and it has found pollutions entering the springs so numerous and so great as to amply explain the higher death rate among the users of this water. It has been pos¬ sible to shut off some of these sources of pollution, and some of the springs have been excluded and filters have been built to make up the deficiency with filtered river water, and the conditions are thus improved, but the matter of entirely correcting them is difficult. It has been proposed to filter the spring water supplies, and other and new sources are under consideration, but no decision has been reached as yet. “At the present time there seems to be a source of arbi¬ trary standard in the methods of water purification. That is to say, methods are extensively used which make the cost of purifying a million gallons of water, including interest and sinking fund charges, somewhere in the neighborhood of $10, and which are sufficient to remove 99 or 99F2 % of the bacteria of the applied water. This may fairly be called the best practice to-day. A purification like this serves to furnish a water from the Merrimac, or from the Hudson, in every respect as good, and perhaps better than is obtained from the best upland sources. It would be difficult to- produce suffi¬ cient reasons and to back them up, to induce a water board or City Council to authorize the necessary expenditure to produce a much more complete purification than this, but it is important to keep this matter in sight, and to note that, if 8 1 the pollution of our rivers increases, the purification can be made to keep pace with this increase, and it will be possible to obtain water of the best equality, even though conditions become far less favorable than they are to-day. “The triumphs in the past have been great. The typhoid fever death rates in a number of cities have been reduced by the installation of filters by 70% or 80% or more. General death rates have also been reduced by amounts which cor¬ respond to much more than the reduction in the typhoid fever rates. Perhaps the achievements of the future in this respect will not he more striking than those in the past, but this cer¬ tainly is true, that with the gradual raising of the standards of the quality of water supplies, and with the growth of a desire for cleaner and purer water, the art of water purifi¬ cation will advance and its application will be constantly ex¬ tended, with the greatest benefit to the people of our cities.” Some Lessons of the Butler Typhoid Epidemic. From Engineering Neivs, December 31, 1903. “Hard as it is upon fever stricken Butler, the typhoid epidemic in that community may be the means of saving thou¬ sands of lives elsewhere by directing public attention to the dangers of polluted water and inefficient sanitary administra¬ tion. “We say thousands of lives advisedly and with full knowledge that the total number of deaths in such epidemics as that of Butler or at Ithaca fall far short of any such num¬ ber. The fact is that the great mortality from typhoid fever results from the numerous cases which are occurring all the time in all parts of the country without the disease becoming sufficiently prevalent to be called epidemic. “In the past twenty years there have been only two typhoid epidemics comparable with that of Butler, viz : Plym¬ outh, Pa., in 1885, and Ithaca, N. Y., a few months ago. The total number of recorded typhoid cases in these three epidemics will probably be somewhere near 3500 to 4000 and 82 the recorded deaths three hundred or more. Meanwhile not a year has gone by when thousands upon thousands of per¬ sons have not been stricken and many thousands died on account of the pollution of the public water supplies of the United States. In other words, the great water-borne typhoid epidemics have been few and at long* intervals, but there are numerous cities with polluted water supplies in which typhoid is always prevalent. Thus it is not merely to avert the danger of epidemics that thousands of American communities need radically to reform their sanitary con¬ ditions. Such action is needed to eliminate the cases of typhoid which occur every year and often at all times of the year. What at first thought may seem to be a sweeping con¬ demnation of private ownership of water works is a fact that the three great typhoid epidemics at Butler, Ithaca and Plym¬ outh were all caused by water supplied by private companies. But on the other hand, what shall be said of tbe equally striking fact that municipal-owned waterworks head the list of cities with continuously high typhoid mortality? For the years 1898 to 1901, the City of Pittsburg, most of whose population is supplied with water from city works, had an average typhoid mortality of 113 per 100,000, standing at the top of the list of cities of thirty thousand and upwards. Charleston, S. C, under private ownership, came next (no), but was followed by three cities owning their works, namely: Youngstown, O.; Allegheny, Pa.; and Troy, N. Y. (109, 90 and 90 respectively). It would be interesting, and perhaps profitable, to go all through the list and see how the ownership stands as the typhoid mortality lowers, but our present space and purpose will not permit such an analysis. While there is reason to believe that a pure water supply can be more readily secured and maintained under municipal ownership, it is at the same time true that polluted water is now being furnished by hundreds of waterworks, includ¬ ing which are very many works owned by cities as well as those owned by private companies. “Even in cities where radical changes in the waterworks are not needed, there is frequently urgent need for a thorough sanitary study of the water supply, to be followed by constant sanitary supervision, all under the direction of technically trained men. “To inaugurate such sanitary reforms there is needed not only action by the towns and cities but by State authori¬ ties. It is not difficult to trace a direct connection between the epidemic at Butler and the fact that Butler is located in the State of Pennsylvania. That State, to put matters plainly, is well known among sanitarians to be notably backward in all matters concerning the protection of the public health. This is not a mere matter of opinion. It can be shown by statistics which represent human lives. “In order to show by concrete example what sanitary neglect in Pennsylvania really means, let us compare the records of typhoid fever mortality in its cities with those of cities in Massachusetts, a State much more densely populated, but well known the world over for its advanced position in matters of sanitation. Among the cities of the United States having 30,000 population and over for which typhoid mor¬ tality statistics for the four years 1898 to 1901 are available, Pennsylvania has 15 and Massachusetts has 17 cities. Nearly all the Pennsylvania cities are among those of high mortality and every one of the Massachusetts cities is among those having low typhoid death rates. The total range of the table, it may be added, is from the very high rate of 113 to the low rate of 6 deaths per hundred thousand of population per annum. “Pittsburg heads the table with 113 typhoid deaths per annum, and Allegheny comes fourth with 90. Johnstown stands six and York ninth, making four Pennsylvania cities among the first ten. The four Pennsylvania cities in the last half of the table stand number 67, 75, 94 and 95, the last two being Erie and Scranton, each with an average typhoid mor¬ tality of 25 per 100,000. 8 4 “In Massachusetts, New Bedford shows an average mortality of 30 per 100,000, Boston and Springfield show¬ ing an average mortality of 26 per 100,000. “To what shall the vastly superior showing of the Mas¬ sachusetts cities be ascribed? The records of the past show that it is due almost wholly to the relatively high character of the water supplies and to efficient sanitary administration. “Several years ago a number of Massachusetts cities were notorious for their high typhoid death rates; but the work of the State and local Boards of Health, backed by sound public opinion, has largely remedied former evils and has made the State foremost the world over for its efficient sanitation. The work of the Massachusetts State Board of Health in lessening water pollution and solving the problem of sewage and water purification has been recognized by the sanitary experts of all civilized nations as one of the most notable advances in public sanitation ever recorded. “Contrast with this record the record of Pennsylvania. Among sanitarians, foreign and American, the State is chiefly notable as an example of the results of sanitary neglect. “For decades the great city of Philadelphia has had to drink the foul waters of the Schuylkill and Delaware, and thousands of lives have been thereby sacrificed. Only now, after a generation of agitation and delay, has the city en¬ tered upon the construction of filtration plants to purify its water supplies. “Pittsburg, which, as we have stated above, has long- been notorious for its high typhoid rate and is now suffering to an extent which almost assumes the proportions of an epidemic, deserves credit for having a few years ago ex¬ pended a large sum of money in water purification experi¬ ments. It deserves nothing but discredit, however, for ignoring these experiments and delaying the work of water purification. Its citizens suffer and die by the hundred, while rival politicians and contractors struggle for party advantage 85 and personal gain in connection with the proposed construc¬ tion of filtration works. “Allegheny, so far as we can learn, also continues to pay a heavy penalty for impure water, notwithstanding the huge filter crib at its Allegheny River intake, which it con¬ structed at great expense a few years ago. ‘‘Reading, standing No. 33, or just below Philadel¬ phia, in the list of cities already reviewed, is making some progress towards the introduction of filters, but its past record is not one to be proud of. At Chester the water com¬ pany has installed filters, but it appears to have been driven to do it by litigation. “It would be interesting to inquire to what extent the backward sanitary condition of Pennsylvania is due to its local municipal government or to its State government. Un¬ doubtedly both are deserving of extreme censure. The State Legislature has not been liberal in its appropriation for the State Board of Plealth, nor has it clothed that body with the powers which it needs to carry on a fight against water pollution. Moreover, unless by quite recent legislation, needed powers for the conservation of the purity of public water supplies has not been granted to either local authorities or water companies. With such a record is it any wonder that Pennsylvania has to its debit two of the three great typhoid epidemics which have occurred in this country or that most of its large cities have for years reported heavy mortality from typhoid? “It may be said in defense of Pennsylvania, perhaps, that her sanitary condition is no worse than that of the States which have not been named and that a comparison with Mas¬ sachusetts, noted the world over for its sanitary achievement and its excellent State and local government, is unfair. Even if such a plea could be supported by facts, which is doubtful, is it a worthy defense? “Pennsylvania as a State is wealthy and prosperous; it can afford to better its sanitary condition. The cities which 86 have grown up within its borders entail a responsibility that may not be likely ignored. Sanitary conditions inherited from primitive days become unbearable as population grows more dense and as means for the spread of communicable disease are multiplied. “It is, therefore, no defense for Pennsylvania to cite the backward sanitary condition of States farther East or farther West. The State is rich enough to afford good sanitation and its densely populated cities demand such sanitation as the price of their citizens' lives. It is to be hoped that such object lessons as the calamity at Butler may have an influence towards bringing about reform." On page 166 of the Municipal Journal and Engineer , Vol. XVI, No. 4, April, 1904, appears an article entitled “Purification of Water by Ozone: A Chemical Process Requiring' no After Treatment—Frees Water from Bacteria and Iron—How it is Done." In this article Dr. Erlwein, of Berlin, speaking on this subject at the forty-third annual meeting of the German Union of Gas and Water Engineers, and describing the ozonizing apparatus in use for the purifi¬ cation of the water supply of Paderborn, in Germany, refer¬ ring to the ozonizing plant at Martinikenfeld, says: “It has been thoroughly tested with Spree water full of cholera, typhus and other disease germs artificially reared and added to the water. The result was to prove that the ozone treat¬ ment killed all the disease germs even when in greater num¬ bers than are present during epidemics. “As regards the chemical effect of the ozonization of water it diminishes the oxidization degree of the water by fifteen to twenty-five per cent, and increases its content of free oxygen. The ozone reaction vanishes a few seconds after the water leaves the tower, and no taste is given to the water * * *. Ozone is not only a bactericide but frees water so perfectly from iron that it may well be used for that purpose alone * * *. The ozonization process, too, should not cause forgetfulness of the immense service which the sand filtration of water has rendered to sanitation, in spite of the unquestionable efficacy of ozonization in killing all pathogenic bacteria, while sand filtration does nothing- more than diminish their number. The ozone process must always be taken seriously into consideration where difficulties arise in large towns as regards extending already existing filter beds or making new ones, when they may need the ac¬ quisition of land at heavy expense. The ozone plant is also excellent for places where it is impossible to rely too much on the waterworks staff. * * *. “* * * There are cases, too, in which iron and organisms giving taste and smell to the water can only be removed by ozone. In short, every case must be judged in¬ dependently for the determination of the most suitable sys¬ tem, or combination of systems, for purifying the water.” I think it is safe to say that the various authorities which the writer has quoted, and which authorities, both European and American, are conceded to rank among the highest, bear witness to the following facts: First .—'That filtration alone can in addition to clarify¬ ing water effect purification of the same, only to a degree varying with construction, maintenance, supervision, etc., of the plant, and Second .—That by ozonizing a previously clarified water absolute purity is obtained. Hence I cannot but express the belief that as an adjunct to any filtration system, sufficiently efficient to produce clear water, sterilization by ozone will prove to be one of the greatest blessings of this century. (Signed) J. J. de Kinder. II III Hill I II IIII III