Is Copper Sulphate Water Treatment Harmful?
Determining Amount of Copper in New York Distribution System After Copper Sulphate Treatment—Shows Not Enough to Affect Health
THE advantages in the use of copper sulphate to control the organisms in the water supplies of cities relying upon storage reservoirs has come to be recognized and the author of this paper has made many investigations as to the proper dosage and the method of treatment. His paper will be found of considerable value as it is based on his actual experience in the Catskill water supply of New York City, and the question as to whether enough of the copper goes into the distribution system to be harmful to health he answers positively in the negative.
During recent years it has been the practice of the water department of New York City to resort to continuous automatic treatment with copper sulphate of the Catskill water as it passes from Ashokan to Kensiro reservoir to prevent seeding the storage reservoir near the city with organisms prevailing in the upper watershed reservoir. The extent of this practice is shown as follows:
Synura has been present each fall in Ashokan reservoir in small quantity and to avoid any repetition of the experience with Synura in the winter of 1921 treatment of the water in the aqueduct has been continuous all winter until about the middle of February. After this date untreated water could not reach the effluent chamber of Kensico reservoir in several weeks or until ice would be off Kensico so that direct treatment of that reservoir could be made if necessary.
Method of Treatment at Kensico Reservoir
Svnura appeared as usual last fall in Ashokan reservoir and automatic treatment in the aqueduct with copper sulphate at the rate of one pound per million gallons. 0.12 ppm., was started on October 11, 1924, and continued until January 29, 1925. A separate growth of synura prevailed at the same time in Kensico reservoir so that it became necessary to treat the lower portion with the same dosage, which was carried out on October 23, 24, 25 and 27 from the dam to Bear Gutter point or about half way to the influent chamber. The distance between the influent and effluent chambers is about three miles. The depth ranges from ten feet at the influent chamber to one hundred feet at the lower end. The treated portion ranged from fifty to one hundred feet in depth and dosage was apportioned to volume. The water leaving Kensico reservoir passes through Hillview reservoir of 900 mg. capacity or about one and onehalf day’s storage at present rate of draft.
Searching for Copper in New York Distribution System
Several times during previous experience search had been made for copper in the distribution system of the city, evaporating one and one-half gallons of water for test, and not a trace had been found. This winter it was decided to make a more extended search. Of the regular daily samples collected at Shaft 23, Brooklyn, 200 cc. portions were taken and combined until 4 liters were obtained, representing about three weeks. This composite was analyzed for copper. The procedure began October 19, 1924, anti has been continued to date.
On November 24. 1924, a scries of samples was taken from the surface water of Kensico reservoir to ascertain the extent of residual synura following the treatment of the previous month. Three samples taken in the upper area from Bear Gutter point to the influent chamber were combined and tested for copper and likewise four samples taken between Bear Gutter point and the dam were combined and tested.
The following table shows the results of these examinations:
Determination of Copper in Water Supply
* Deposit dark and ammonia blue color slight.
† Average of weight and titration with exception of reservoir results.
+ Calculated to form in which it is applied.
Copper Present After Four Weeks
The astonishing feature isthat copper was continuously present until four weeks after the automatic treatment stopped, a period probably roughly equal to the time of flow from influent to effluent chamber. The amounts would indicate either an uneven admixture of the copper with the water or a certain striation of the precipitated copper. The dosage both of the reservoir and in the automatic treatment was 0.12 ppm. based on total volume. It seems probable that the copper combines with the organic matter of the microscopic organisms as well as with the alkalinity and may either float or sink or he continuously affected by the winter circulation of the water which takes place from top to bottom all winter. Apparently all the copper fed into the supply came through into the distribution system. The first composite from Shaft probably represents the reservoir treatment but all the others the aqueduct treatment. It will he interesting to note in future whether the same thing occurs in other periods than winter, since real stagnation occurs in summer.
The amounts of copper found were too small to he of sanitary significance. Whether they are of industrial importance, such as in photography or in dye mnaufacture, is a question for future decision, considering the experience of Baltimore with small amounts of manganese in the laundry industry.
Method of Determining Amount of Copper Present
The method employed in determining these minute amounts of copper was as follows:
The composite samples ranged from 2.3 to 4.2 liters and the weight of cupper determined from 0.1 milligram to 0.5 milligram. As a routine procedure 4 liters was adopted. As the ratio of copper to crystalline copper sulphate is as 1 :4. the amount of copper determined as Cu equals parts per million direct of copicr sulphate as applied.
The composite, acidified with 5 cc. of 1:1 nitric acid, is evaporated to dryness on a water hath in a platinum dish. The residue i> gent lx ignited by fanning with a gas dame until the small amount of organic matter remaining is burned off. ‘1 he evaporation with nitric acid destroys most of the organic matter and oxidizes any iron to ferric condition.
The residue is now transferred to a 3-inch porcelain evaporating dish by treating with 5 cc. of hot 1 :1 nitric acid and thoroughly scrubbing with a policeman, repeating the scrubbing with a second 5 cc. of acid and washing with two portions of hot water.
The contents in the porcelain dish is again taken to dryness to get rid of excess nitric acid and thoroughly dehydrate silica. The residue is moistened with 1 cc. of 1:1 nitric acid, diluted with !5cc. of hot water, filtered on a small filter and washed with successive small portions of hot water from the top down. The filtrate and washings are caught in a 60cc. non-sol beaker, volume about 3occ. This process gets rid of the large hulk of insoluble solids from the gallon of water taken.
The filtrate is now treated with 1 :1 ammonia in distinct excess to precipitate iron and hold copper in solution, boiled, settled, filtered on a small filter paper, and washed with hot water made slightly alkaline with ammonia. The filtrate is transferred to a 3-inch porcelain evaporating dish again, concentrated to about lOcc. to precipitate the last traces of iron, filtered through a small filter, washed with successive portions of hot water from the top down. The filtrate and washings amount to 20cc. and are caught in a 60-cc. non-sol beaker.
Then 5cc. of 1 :1 nitric acid and 5cc. of saturated solution of potassium sulphate are added. The latter assists copper to deposit completely in a bright clean manner. Electrolysis takes place oxer night (16 hours) using 110 volt I). C. lighting current through a 10 watt lamp, i. e., 0.10 ampere. The electrodes used are platinum spirals (of No. 16 B. & S. gauge wire) weighing about 13 grams each. About one inch is immersed in the liquid. Copper may he readily seen down to 0.1 mg. There should be no odor of ammonia in the liquid after electrolysis due to reduction of the nitric acid; if so the deposit will lie black and oxidized. The coils are washed with distilled water with the current on while lowering the beaker until free from the liquid. This is a necessary precaution as it doesn’t take nitric acid long to dissolve traces of copper with the current off. The coils are dipped in alcohol, heated over a gas Hame until the alcohol has evaporated without igniting, cooled in a desiccator and weighed.
Methods of Checking Results
As a check the deposit is dissolved by rolling the cathode in 3cc. of 1 :1 nitric acid in a test tube and washing in 3cc. of distilled water in another tube. The washing is added to the acid solution and nitrous oxides boiled off, then made alkaline with 1 :1 ammonia. Even traces of copper in small volume give a blue color and an estimate of amount may lie made by comparing with known amounts of copper solution similarly treated.
As a further check the ammoniacal solution is acidified with acetic acid in the test tube, treated with a crystal of potassium iodide, the size of a pea, and the resulting iodine set free titrated with N’/lOO thiosulphate, using a 2cc. pipette graduated to 0.02cc„ and starch solution as indicator. Each cc. N/100 thiosulphate equals 0.635 mg. Cu.
These two checks arc of advantage in denoting the purity of the copper deposit. The agreement between the electrolytic weight and the titration is shown in the table and is reasonably close.
(Excrrptfi from paper read before the annual convention of American Water Work Association at Louisville, Ky.)