Compliance monitoring of drinking water supplies - Environmental

---

“To assure a supply of drinking water which dependably complies with such maximum contaminant levels. . . .” With the signing of the Safe Drinking Water Act in 1974 (P.L. 93-523), it became the responsibility of the Environmental Protection Agency (EPA) and the states to ensure that safe water supplies are available to all users. As part of the program for meeting this goal, the EPA proposed a list of materials to be considered as possible health hazards when present in drinking water. After much public discussion from environmental and other public interest groups, the National Drinking Water Advisory Council, state agencies, and others, the EPA’s Interim Primary Regulations were issued on December 24, 1975. These regulations include: maximum allowable levels of ten inorganic chemicals and six organic pesticides maximum allowable levels of turbidity and coliform bacteria minimum monitoring frequencies for the contaminants how and when public notification must be given when any of the maximum allowable levels of contaminants cannot be met by the water system. In addition to the Interim Primary Regulations, the EPA issued regulations on the maximum permissible levels of radionucleotides in July 1976, and secondary regulations on March 31, 1977. Secondary regulations deal with aesthetic water qualities that affect consumer acceptance, such as levels of sulfate, iron, and manganese. They are not legally enforceable. Coliform and turbidity monitoring Compliance monitoring under the Interim Primary Regulations for turbidity and coliform bacteria began June 24, 1977 in approximately 40 000 community public water systems across the country. Community systems include those serving a resident population of at least 25 people with at least 15 service connections. Noncommunity systems, defined primarily as serving the traveling public in campgrounds, resorts, and the like, will have to begin testing for those two parameters by June 1979. For other contaminants, monitoring requirements are split by 870

EnvironmentalScience 8 Technology

type of water supply. In community surface-water systems, monitoring for inorganic chemicals and pesticides must begin by June 1978. June 1979 is the deadline for community groundwater systems. Nitrate is the only chemical that noncommunity systems are required to monitor, with samples to be taken by June 1979. Bulk of the work The most frequent testing required under the Safe Drinking Water Act is for turbidity and coliform. Turbidity, one of the few tests that can be performed by the plant operator, must be tested on-site at least once daily with a nephelometer. Maximum permissible monthly average is 1 NTU. Monitoring frequencies for coliform are dependent on community population. Water systems that serve up to 1000 persons must sample at least once a month. Systems with 12 000 users are to sample at least 13 times a month. The testing frequency increases with population, so a plant with over 25 000 users is requiredto take 30 samples/month; and those with over 5 million customers, 500 samples. EPA regulations specify that only state-certified laboratories can perform coliform analyses by either the multiple-tube fermentation technique or the membrane-filter method. In addition to coliform, all other inorganic chemical and pesticides analyses must be performed by certified laboratories. Recognizing the difficulties in collecting and transporting bacteria test samples, the EPA allows water plants to substitute free chlorine residual testing for up to 75% of their coliform monitoring requirement. (At press time, eight states-Ala., Ark., Conn., Ga., La., Miss., Neb., and 0kla.-had accepted primary enforcement responsibility for Safe Drinking Water Act compliance. Several of these have taken the option not to allow free chlorine substitution. Individual state officials should be checked.) Testing can be done by the plant operator. Samples must be taken at least once a day, at the rate of four times the bacterial testing requirement. A system required to send out 30 bacteria test samples a month to a certified laboratory could reduce that

number to eight, then test the free chlorine residual 3 times a day, or at least 88 times in the month, to make up the balance. DPD (N,Ndiethyl-p-phenylenediamine) has been specified as the method of choice for the free chlorine analyses. It is a fairly recent method, having only gained acceptance in the water industry within the past six or seven years. Since it is a relatively new reagent, it seems appropriate to discuss some of DPD’s characteristics and to offer a few suggestions for its best use in water testing.

laboratories analyzed an unknown of 980 pglL free chlorine with the DPD colorimetric method and obtained a relative standard deviation of 20.7% and a relative error of 15.6%. Analysis of the 800 pg/L unknown by the leuco crystal violet method by seventeen laboratories showed a relative standard deviation of 32.7%, with 7.1% relative error; while the methyl orange method in 26 laboratories yielded a 43.0% relative standard deviation and 22.0% relative error. It would appear the DPD method accuracy and precision is no worse than other available chlorine methods, and may be slightly better.

Available chlorine reagents Many methods have been reported for the analysis of chlorine. Standard Methods, 14th ed., lists at least eight different procedures for varying applications. When specifying a chlorine method, it is important to know what type of chlorine must be measured. Free chlorine is formed by the simple dissociation of chlorine in water, and exists as hypochlorous acid (HOCI) and the hypochlorite ion (OW). Free chlorine is considered the most effective disinfectant of the chlorine species. Combined chlorine forms by the reaction of free chlorine with nitrogenous and organic compounds. Combined species include monochloramine (NH2CI), dichloramine (NHCI2), and nitrogen trichloride (NC13). Most combined species require a longer contact time for disinfection and are not considered as efficient sanitizing agents. For free chlorine measurements, syringaldrazine, DPD, stabilized-neutral orthotolidine, and leuco crystal violet methods are available. Some of these methods can also differentiate between free and combined species. DPD, leuco crystal violet, orthotolidine arsenite, and methyl orange colorimetric methods, and an amperometic titration using phenylarsine oxide (PAO)are some of the common differentiating procedures. To measure both combined and free chlorine (or total chlorine), the iodometric titration and the unmodified orthotolidine method can be used.

Test options okayed by EPA For compliance monitoring,analysts can use titrimetric, visual, or spectrophotometric methods. Visual methods are not listed in APHA Standard Methods. In the titrimetric procedure, a 100-mL sample is buffered to about pH 6.5 with a phosphate buffer solution; 5 ml of indicator solution containing 0.1 % DPD with sulfuric acid and EDTA is added; and the sample titrated with 0.0282N ferrous ammonium sulfate. Endpoint is red to colorless. The indicator is also available 9s a single powdered reagent that is already combined with buffer, EDTA, and the like. Spectrophotometricanalyses generally require a 10- to 25-mL sample buffered to about pH 6.5 with 0.5 mL of DPD indicator solution added. Absorbance is read at 515 nm. Calibration curves are prepared with either chlorine solutions or potassium permanganate. DPD has another absorbance maximum around 550 nm, but this is rarely used. Test-kit methods offer the most convenience and simplicity. About 5 mL of sample is measured into a tube, reagent added, the sample mixed, and the color compared to a permanent standard, usually made of plastic or glass. Untreated water sample or demineralized water is used as a blank.

DPD for chlorine residual First suggested for chlorine use by Dr. A. T. Palin in 1956 (British Patent 813,493), DPD works as an oxidation-reduction indicator. DPD reacts with a variety of oxidants, iodine, bromine, some oxidized forms of manganese, as well as chlorine, and can be used to quantitatively determine any of these, provided the others are known to be absent. In its reaction with free chlorine, DPD is instantly and directly oxidized, losing one electron. The oxidized form is red. Combined chlorine species react more slowly than free chlorine, but can be catalyzed by adding small quantities of iodide to the indicator. The combined species react through a two-step mechanism where the chlorine species oxidize the iodide to iodine, which then reacts with the DPD. If the analyst wishes to differentiate between monochloramine, dichloramine, and nitrogen trichloride, certain reaction conditions can be met. Generally, sequential iodide additions to the reaction mixture will allow analysis of the different species. Most water treatment plants, however, are interested only in the ratio of free to combined chlorine. Though DPD is reportedto follow Beer’s Law up to 8 mg/L free chlorine, Standard Methods, 14th ed., recommends an upper limit for both the colorimetric and titrimetric procedures of 4 mg/L free chlorine or chloramine. Higher sample concentrations should be diluted with chlorine-free demineralized water. Lower detection limit for the colorimetric test is about 0.02 mg/L (1-cm cell), depending on the instrument design. Visual methods are slightly less sensitive, with 0.05 mg/L the lowest concentration of visible color. The endpoint in tritrimetric analyses is detectable to about 0.05 mg/L free chlorine. In a study published in Standard Methods, a synthetic unknown containing 800 pg/L free chlorine was analyzed in 19 laboratories by the DPD ferrous titrimetric method. Relative standard deviation was 39.8%, with a relative error of 19.8%. Twenty-six

Testing considerations Reagent stability: Great improvements have been made in DPD’s stability since the method was introduced. The indicator solution, as suggested by Palin, has a rather short shelf life, which is further reduced by increased temperature. Nicolson found the indicator in solution deterioratedto a significant degree within one month. With the introduction of solid reagents, the stability has been increased considerably, so, under normal conditions, shelf lives of a year are not unusual. For longest shelf lives, the reagent should be protected from light and heat and stored in a refrigerator. Interferences: When testing for free chlorine in the presence of high concentrations of combined chlorine, high free chlorine values may result from monochloramine breakthrough. This may show up in the spectrophotometric procedure as a continuous drift of the instrument readout to higher absorbance values. To avoid this problem, the free chlorine should be measured as soon as possible after reagent addition. When read at one minute, we have found 3.0 mg/L monochloramine will cause less than 0.1 mg/L interference in the free chlorine determination. Another advantage of this technique is it minimizes loss of chlorine caused by lengthy sample handling. Since DPD does react with other oxidants such as bromine, iodine, and ozone, these must be absent for accurate chlorine test results. The indicator is generally free from other interferences from materials commonly found in drinking water supplies, though high levels of dissolved solids may cause turbidity with some reagent formulations. Incorporation of EDTA into the indicator also helps to prevent interference from copper and dissolved oxygen. Interference from oxidized manganese in the titrimetric procedure may be treated with potassium iodide and sodium arsenite. pH control critical: If the buffer capacity is exceeded and the sample pH falls below about 6.2, interference from monochloramine breakthrough may occur. If sample pH is too high, dissolved oxygen may interfere. In these cases, sulfuric acid or sodium hydroxide should be added to bring the sample pH to about 6 or 7. Glassware cleanup: If sample cells and/or titration vessels Volume 11, Number 9, September 1977

871

I At last-

a disposable filter so good, it can sterilize airso economical, you won't mind replacing it!

Balston's new Microfibre@Disposable Filter Unit gives you truly high efficiencyfiltrationto the sterile level in gases, and 98% retention of 0.3 micron particlesin liquids-combined w convenience of complete disposability. It's ideal for ACTUAL sterilizing air and other gases-autoclavable SIZE filtration of samples to analyzers complete removal of oil, dirt and water from air and gases to 125 psig filtration of small quantities of liquids with minimum hold-up and no cross-contamination All this at a price of only about $2.75 each. For more detailed information, use the Reader Service Card. Fora FREE SAMPLE, write on your letterhead, describing your application. Address Dept. EST-17.

BALSTON, iNC. P 0 Box C, 703 MassachusettsAve Lexington, Massachusetts 02173 (617) 861-7240 * (617) 862-7455 Telex 92-3481 CIRCLE 2 ON READER SERVICE CARD

are not fullv cleaned after use with DPD.. the alass mav become stained wiih oxidized reagent. This can cause a high reagent blank in colorimetric procedures. Should this occur, we have found several rinses with methanol will remove all traces of color. Toxicity: Take normal care with DPD reagents, as with any chemical. Avoid ingestion and prolonged or repeated contact with the skin. Product options If titrimetric or spectrophotometric DPD methods are chosen, the analyst can purchase indicator powder from a number of manufacturers including Hach Chemical Co., Eastman Kodak Co., and Gallard-Schlesinger Chemical Manufacturing Co. Several distributors also handle the reagent. Indicator powder is produced as the free amine and as several water-soluble forms of DPD such as the oxalate and sulfate salts. Indicator premixed with buffer and chelants is available in prepackaged unit doses from Hach Chemical Co., Hellige, Inc., LaMotte Chemical Co., and Taylor Chemical Co. Various distributors offer the combined reagent in compressed tablets. If a test-kit method is chosen, several products are available. Test-kit methods are approved by the EPA for Safe Drinking Water Act Reportingand are recommended for on-site testing. Since low concentrations of free chlorine are very unstable in water, on-site testing is the best way to minimize error from chlorine loss during sample transport. The concentration range for most kits is between 0.1-4.0 mg/L chlorine, which is in the limits for compliance monitoring. Compact kits can be purchased from Hach Chemical Co., LaMotte Chemical Co., Hellige, Inc., Taylor Chemical Co., and distributors such as Fisher Scientific Co. A DPD chlorine analyzer for continuous monitoring is also available from Hach Chemical Co. This method is not accepted for reporting but can be used for control purposes, monitoring swimming pools, and the like. Wastewater analysis The DPD method of chlorine analysis is also accepted by the EPA for wastewater analysis, but only the spectrophotometric and titrimetric methods that are described in Standard Methods, 14th ed., are approved. The test-kit methods are not yet approved for NPDES (National Pollutant Discharge Elimination System) compliance monitoring. Additional reading Federal Register, 40,No. 248,p 59580. Bjorklund, J. G.,Rand, M. C., J. Am. Water Works Assoc., 60 (5), 608 (1968). APHA Standard Methods, 14th ed., 1975,pp 31 1-333. Nicolson, N. J., Water Research Association Technical Paper, T. P. 29 (1963),T. P. 47 (1965). Thomas Haukebo is a research chemist at Hach Chemical Co. and a member of the Standard Methods Joint Task Group on chlorine.

Jean Bernius is a technical writer at Hach Chemical Co.

CIRCLE 10

872

ON

READER SERVICE CARD

Environmental Science & Technology