Techniques of Water-Resources Investigations

of the United States Geological Survey

Chapter A3 METHODS FOR ANALYSIS OF ORGANIC SUBSTANCES IN WATER

By Donald F. Goerlitz and Eugene Brown

Book 5

LABORATORY ANALYSIS DEPARTMENT OF THE INTERIOR

WILLIAM P. CLARK, Secretary

U.S. GEOLOGICAL SURVEY

Dallas L. Peck, Director

First printing 1972 Second printing 1972 Third printing 1984

UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1984

For sale by the Distribution Branch, U.S. Geological Survey 604 South Pickert Street, Alexandria, VA 22304 PREFACE The series of manuals on techniques describes procedures for planning and executing specialized work in water-resources investigations. The material is grouped under major subject headings called books and further subdivided into sections and chapters. The unit of publication, the chapter, is limited to a narrow field of subject matter. This format permits flexibility in revision and publication as the need arises. Section A of Book 5 presents techniques used in water analysis. Provisional drafts of chapters are distributed to field offices of the U.S. Geological Survey for their use. These drafts are subject to revision because of experience in use or because of advancement in knowledge, techniques, or equipment. After the technique described in a chapter is sufficiently developed, the chapter is published and is for sale by the U.S. Geological Survey, 604 South Pickett Street, Alexandria, VA 22304. in

CONTENTS

Page Page Preface..,-.-----..- - ______III Part II, Ar Abstract. ______1 NitrogNitrogen, nitrate—Continued 1 RReferences. ______17 Nitrog 17 Part I, Sampling- ______2 DDiazotization method- 17 Sample preservation _,____.______.___.__._ 3 RReferences. ______18 References- ______-.______,-_-___--_-- 4 Nitrog 18 4 KKjeldahl method _ 18 Carbon, all forms. ____-______---____.____ 4 Reference- ______20 References..-,. __.______6 OxygeOxygen demand, chemical (COD). . 20 Chlorophylls- .__-__ __._„____,_____._____ 6 Dichromate oxidation method- 20 Extractive spectrophotometric method. __ 6 HReferences-, ______22 References. ______..„______8 Phenc 22 Color. ___---_-_._____. __---__-_-_-__-__. 8 EReferences. 24 Comparison method ______8 Pesticides—gasPestic chromatographic analysis. 24 9 References.Ii -__.____.__.______.___ 29 Extractable organic matter. _.-_._-__-____„_ 9 Insecl 30 Oils and waxes. ______9 CGas chromatographic method ______30 References _, ______.„_„____ _ .. _ 11 References.I ______32 Methylene blue active substances- ______11 ChlorChlorinated hydrocarbon insecticides in sus­ Synthetic anionic detergents ______11 penpended sediment and bottom material _____ 33 13 CGas chromatographic method ______33 Nitrogen, ammonia. ______13 Distillation method. ______13 ChlorChlorinated phenoxy acid herbicides in water. 35 References. _ . - ______15 (Gas chromatographic method ______35 IReference ______39 Nitrogen, nitrate, ______-___.___,___.,- 15 Brucine method. ______15 ChloiChlorinated phenoxy acid herbicides in sedi­ Reduction method. ______17 mement (tentative) ______39 (Gas chromatographic method ______39

TABLES

Page 1. Retention values for chlorinated hydrocarbon insecticides,.______25 2. Retention values for phosphorothioate insecticides.______26 3. Retention values for methyl esters of chlorinated phenoxy acid herbicides. 26 4. Insecticides in water: recovery of compounds added to surface-water samples. 32

METHODS FOR ANALYSIS OF ORGANIC SUBSTANCES IN WATER

By Donald F. Goerlirz and Eugene Brown

Abstract Because of the myriad of sources from both This manual contains methods used by the U.S. Geo­ pollution and natural processes, organic matter logical Survey for the determination of organic substances is present in almost all surface and ground waters in water. Procedures are included for the following cate­ and directly influences the water quality. gories of organic substances: Organic carbon, chlorophylls, In reviewing water-quality reports, one may color, detergents, nitrogen, oils and waxes, oxygen demand observe that although color, oxygen consumed (chemical), phenolic materials, herbicides, and insecti­ cides. Procedures are also given for the determination of (chemical oxygen demand), and other criteria for chlorinated hydrocarbon insecticides, as well as chlorinated the analysis of organic substances are described phenoxy acid herbicides, in sediment and bottom materials. in the introductory remarks, seldom are such data included in the tables of the report. These data Introduction are noticeably absent from past reports because such crude measurements did not appear to serve The yellow or brown color commonly associated any meaningful purpose in explaining most water with natural water is often the result of the de­ processes. The value of even the most primitive composition of naturally occurring organic matter. measurements of organic substances in water is The most abundant sources of this material in­ apparent, considering the general awareness of clude decaying vegetation, algae, and microscopic man's effect on the environment and the imple­ organisms. These substances are produced mostly mentation of programs to stem the tide of water on land and during runoff are flushed into the pollution, lake eutrophication, and similar prob­ water, where complex biological processes con­ lems. tinue. The role of natural organic substances in The purpose of this manual is to provide direc­ water processes is not very well understood. tions for collection and analysis of water samples These compounds are known to aid in transport­ containing organic substances as required by the ing and solubilizing many trace elements and are Geological Survey in making water-quality in­ important in weathering. Further, natural or­ vestigations. This manual is an updating and ganic substances interact with both organic and compositing of the organic analytical methods inorganic pollutants. contained in Geological Survey Water-Supply In addition to organic matter from natural Paper 1454, "Methods for Collection and Analysis sources, increasing amounts are entering water of Water Samples" (Rainwater and Thatcher, as a direct and (or) indirect result of man's 1960). The intense activity in the field of organic activities. Leading sources of this material are analysis is providing new and improved methods, industrial and domestic waste, agriculture, urban and as appropriate methods become available runoff, mining, and watercraft. Foaming deter­ they will be added to this manual. All methods as gents have been observed in water supplies. Fish written may be subjected to revision from time kills have been caused by toxic chemicals and to time. Methods labeled as tentative are as­ have been linked to nutrient-induced algae sumed to be adequate but need further testing blooms, which deplete the dissolved oxygen. before final acceptance. TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

Part I. Sampling Organic substances on the water surface or incorporated in the bed material present very Refer to Part I, Chapter Al of Book 5, for difficult sampling problems. Surface films may information on site selection, frequency, field appear chronically or intermittently and most safety, and first aid. Before a proper sampling often come from accidental spills or seepage. program can be initiated, the nature of the Samples taken from the surface of a body of organic compounds sought must be considered. water, especially moving water, are almost im­ As a general rule, organic compounds having less possible to evaluate. Quite often the samples than six carbon atoms are water soluble, whereas must be related to the total volume of water, compounds with more than six carbons are not. surface area, film area, and film thickness. At This generalization must be applied with care, present, no universally acceptable surface sam­ however, because the presence of hydroxyl or plers are available, and investigators are required other polar sites on a molecule increases its to fabricate their own. Floating hoops or rec­ solubility. Even very soluble compounds, how­ tangles covered with clean metal foil which ever, may prefer to reside on organic-layered delineate an area and allow skimming a surface sediment. Further, some insoluble compounds sample from calm water have been used with may be suspended on the water surface. In sum­ limited success. Other methods for collecting mary, organic matter in a body of water may be films and slicks from surface waters, such as distributed on the surface, in suspension, ad­ "dustpan skimmers," absorbent cloths, and floor sorbed on suspended sediment, in bed materials, mops, have been used, but none have been and (or) in solution. satisfactory for general application. The method Because of this wide and generally unpre­ most often used for describing a problem area dictable distribution of organic material in a when surface films are involved is identification body of water, the collection of a truly repre­ of the material, locating the source, and then sentative sample requires a great deal of care estimating the amount entering the water. and often requires the use of specialized sampling A rational approach to site selection for equipment. In most instances, samples for sampling stream bottoms is also difficult. Fortu­ organic analysis are best collected with the same nately, bed-material samplers are available. equipment and technique used for the collection Ideally, bottom material, including that which is of samples for suspended-sediment measure­ moving by saltation, rolling, or otherwise drifting ments. The technique and equipment used in along the streambed, should be included. It is such measurements have been extensively stud­ also desirable to collect the water from the 3 or ied, and they are discussed at length in one of a 4 inches of flow immediately above the stream series of reports by the Inter-Agency Committee bottom. This is commonly called the "unsampled on Water Resources (1963) and, more recently, zone/' because it is not sampled by suspended- in a report by Guy and Norman (1970). When sediment samplers or most other water-sampling more elaborate equipment is not available, a devices because of mechanical limitations. simple weighted bottle may often be used for Bed-material and suspended-sediment samplers sampling wells, lakes, and slow-moving streams may be obtained through the Federal Inter- with little suspended sediment. Other available Agency Project, U.S. Army Engineer District, samplers have rubber gaskets and valving ar­ St. Paul, Minn. One of these is the piston-type rangements that may either contaminate the bed-material hand sampler BMH-53 for shallow sample or remove organics from the water by water, and the other is the BM-54 or BM-60 adsorption. Such devices as the DH-48, D-49, bed-material sampler for deeper water. Both of or the P-46 samplers, which are specially de­ these are constructed to prevent sample washout signed to collect suspended-sediment samples when raised to the surface. Bed-material samplers from fast-moving streams at water velocity, to sample at the interface are now under develop­ must be modified for organic sample collection. ment and should be available soon. Most importantly, all the neoprene gaskets must Most investigations are not concerned with be replaced by inert plastic material, such as surface or bottom material, but rather with that teflon, and oil must be eliminated from valves. within the body of the water. Most of samples are ANALYSIS OF ORGANIC SUBSTANCES IN WATER collected from just beneath the surface of the hydrochloric acid. Following a rinse in distilled stream to as near the bottom as the sampling water, the glass containers are put into an oven apparatus will allow. Accurate samples are taken and heated at 300°C (Celsius) overnight. The either by depth integration or at a number of teflon cap liners and the metal closures are points in a transverse cross section. For small washed in detergent. After rinsing with distilled streams, a depth-integrated sample or a point water, the caps are set aside to air-dry. The liners sample taken at a single transverse position are rinsed in dilute hydrochloric acid and then located at the centroid of flow is usually adequate. soaked in redistilled acetone for several hours Larger streams require selection of several verti­ and heated at 200°C overnight. When the heat cals at centroids of equal flow after careful and treatments are completed, the bottles are removed often extensive measurement. Depth-integrated from the oven and capped with the closure and samples may also be taken by the ETR (equal teflon liner. The clean bottles are usually stored transit rate) method, at equally spaced points or shipped in a "duo-pak" container—a form- across the stream cross section. In collecting fitting expanded polystyrene case in a corrugated these samples, the sampler is lowered and raised cardboard carton. at each sampling point in the same length of time In actuality, the analysis begins with the (Guy and Norman, 1970). Lakes are sampled on collection of the sample. The sample must three-dimensional grids consistent with the shape represent the body of water from which it was and depth of the water body. Wells may be collected at the time it was taken. The importance sampled from the pump, provided the oils from of the sample to the final result cannot be over­ the mechanism do not contaminate the sample, emphasized. If possible, an analyst or a person otherwise a bottle sampler should be used. directly concerned with the particular study Compositing samples in the field is not recom­ should collect the samples. Inexperienced per­ mended. Individual samples for compositing sonnel should never be allowed to collect samples should be taken to the laboratory, where careful unless they are very closely supervised. If pos­ control can be maintained. Organic sediments sible, the principal investigator should give and soils tend to cling to sample containers, and personal instruction on sample collection. In all special precaution must be taken to avoid im­ instances, detailed printed instructions should proper handling. accompany each set of sample bottles. The sam­ Glass bottles are the most acceptable con­ ple containers should always be sent from the lab­ tainers for collecting, transporting, and storing oratory directly to the person taking the samples samples for organic analysis. Glass appears to be just before sampling. Extra or spare sample inert relative to organic materials and can with­ bottles should be actively inventoried and remain stand a rigorous cleaning procedure. Because under the control of the laboratory. organic materials are so plentiful in the environ­ ment, it is extremely difficult to collect samples free from extraneous contamination. Apparatus Sample preservation for sampling or processing samples must be scrupulously clean. Sample bottles, especially, Most water samples for organic analysis must must be free of contaminating materials. Boston be protected from degradation. Provisions for round glass bottles of 1-liter capacity with refrigerating or otherwise preserving the sample sloping shoulders and narrow mouths are satis­ should be available. Icing is the most ac­ factory for most applications. The closure should ceptable method of preserving a sample, but it be metal, preferably inert, and lined with teflon. is not always possible. Timing of collection This plastic is the only organic material that should be arranged so that the sample can reach should be allowed to contact the sample. the laboratory in a minimum of time. There is no AH sample bottles, whether new or used, must single preservative that may be added to a be cleaned before collecting organic samples, and . sample for all forms of organic analysis, but each the following procedure is recommended: After sample must be treated according to the analytical washing in hot detergent solution and rinsing in procedure to be performed. For the determina­ warm tapwater, the bottles are rinsed in dilute tions in this manual, the following general methods

488-959 O - 72 - 2 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS of sample preservation should be used: the chemical oxygen demand determination. This technique is not hampered by the presence of Oxygen demand, chemical: Add concentrated reducing substances or affected by the type of H2S04 (sp gr 1.84) at a rate of 2 milliliters (ml) organic material being oxidized. It cannot, how­ per liter of sample. ever, be used to replace the biological oxygen Carbon, inorganic: Refrigerate at 4°C. demand determination. Carbon, organic: Add concentrated H2SO4 at a rate of 2 ml per liter of sample and refrigerate 1. Summary of method at 4°C. Using a microsyringe, a fraction of a milliliter Chlorophylls: Refrigerate at 4°C. of sample is injected into a combustion tube Color: Refrigerate at 4°C. under a sweep of ox}^gen gas. All the carbon in Oils and ivaxes: Add concentrated H2S04 until the sample is converted to carbon dioxide and the pH of the sample is below 3.0. Generally, 5 measured by use of a nondispersive infrared ml per liter will be sufficient. analyzer (Van Hall and others, 1963). Surfactants: Refrigerate at 4°C.

Nitrogen: Add 40 milligrams (mg) of HgCl2 2. Application per liter of sample and refrigerate at 4°C. Water containing carbon from about 1.0 mg/1 Phenolic material: Acidify sample to pH 4.0 (milligram per liter) to 1,000 mg/1 may be with H3P04, add 1.0 gram (g) CuSO4 -5H20 per analyzed directly. For a more sensitive procedure, liter of sample, and refrigerate at 4°C. the analyst is referred to the work of Menzel and Herbicides: Acidify with concentrated H2SO4 Vaccaro (1964). Higher carbon concentrations at a rate of 2 ml per liter of sample and refrigerate must be diluted. Water containing organic at 4°C. Insecticides: None required for chlorinated particulate matter may be analyzed directly provided the largest particle diameter is f—| the compounds. syringe needle diameter. Otherwise, the particles must be reduced in size by homogenizing or References blending, and the resulting suspension must be Guy, H. P., and Norman, V. W., 1970, Field methods for fairly uniform and stable. the measurement of fluvial sediments: U.S. Geol. Depending on the sample treatment, results Survey Techniques Water-Resources Inv., book 3, should be reported as follows: chap. C2, 59 p. Rainwater, F. H., and Thatcher, L. L., 1960, Methods 2.1 Total carbon: Sample analyzed without of collection and analysis of water samples; U.S. Geol. filtration or acidification, Survey Water-Supply Paper 1454, 301 p. [U.S.] Inter-Agency Committee on Water Resources, 1963, 2.2 Total organic carbon: Sample analyzed A study of methods used in measurement and analysis without filtration, but acidified and purged with of sediment loads in streams, in Determination of inert gas to remove inorganic carbon. fluvial sediment discharge: Washington, U.S. Govt, 2.3 Dissolved carbon: Sample analyzed after Printing Office, Rept. 14, 151 p. filtration, but without acidification. 2.4 Dissolved organic carbon: Sample ana­ lyzed after filtration and acidification and purging to remove inorganic carbon. Part II. Analysis of samples In addition to these reporting forms, concen­ trations of inorganic carbon (dissolved and Carbon, all forms suspended) and suspended carbon (organic and The procedure given affords an accurate de­ inorganic) may be computed by differences from termination of carbon in water in solution as the above measurements. well as in suspended matter. Both organic and inorganic carbon can be selectively measured. 3. Interferences The organic carbon determination gives a much Strongly acidic solutions and some brines truer measure of the organic matter present in interfere with this technique by producing aqueous solution and (or) suspension than does infrared-absorbing fogs. Very volatile carbon ANALYSIS OF ORGANIC SUBSTANCES IN WATER compounds may be lost during carbonate elimina­ baseline between injections. Run duplicate de­ tion. Injections in excess of 0.08 ml result in terminations, at the lonst, for each standard and incomplete combustion owing to the large volume plot the milligrams per liter of carbon versus peak increase associated with the water vaporization. height measured from baseline on the recorder. 6.A.2 Dual-channel analyzer, Bockman Model

4. Apparatus 915, or equivalent. 6.A.2.1 Prepare a standard curve for the 4.1 Carbonaceous analyzer, Bookman Model organic carbon channel of the analyzer as in stop llf>-A, Model 915, or equivalent. 6.A.1 above. 4.2 Ulicroh'ter syringe, 100-//1 (inicroliter) 6.A.2.2 Prepare a standard curve for the capacity with large-bore needle, inorganic channel of the analyzer, using the same 4.3 Micvofiltration apparatus: Use only silver technique, by injection of standards prepared by metal filters having 0.45 /im (micrometer) maxi­ dilution of the carbonate-bicarbonate stock stand­ mum pore size, obtainable from Solas Flotronics. ard. Filters should be heated at 300°0 overnight to 6.B Analysis of samples remove organic matter. 6.B.I Single-channel analyzer 6.B.1.1 Depending on how the results are 5. Reagents to be reported, the sample may be filtered and All reagents must be chocked for organic con­ (or) acidified. If the sample is to be filtered tamination. through the metal membrane filter, the filtration 5.1 Carbonate-bicarbonate standard solution, must be done before acidification. Only enough 1.00 ml = 1.00 mg carbon: Dissolve 3.500 g filtrate for the determination need be collected. sodium bicarbonate and 4.418 g sodium car­ 6.B.1.2 Adjust the pH of the sample to 2 bonate, both dried at 1Q5°C for 1 hr (hour), in or below, using concentrated hydrochloric acid, carbon-free distilled water and dilute to 1,000 ml. and slowly bubble nitrogen gas through the solu­ 5.2 Hydrochloric acid, concentrated (sp gr tion for 3-5 min (minutes) to purge carbon 1.19). dioxide. 5.3 Xilrogen gas, free of carbon dioxide and G.B.I.3 Mix the sample well. Inject 20-80 organic impurities. jul of the sample into the combustion tube, using 5.4 Oj-ygen gas, free of carbon dioxide and the same technique as for the standards. Record organic impurities. the peak heights from at least two determinations 5.5 I'otassiwn hydrogen phthalate standard for each sample. solution, 1.00 ml —1.00 mg carbon: Dissolve 6.B.1.4 Prepare a blank from carbon-free 2.128 g potassium hydrogen phthalate, primary distilled water and all the reagents used, and standard grade, which has boon dried at 105°C analyze the blank as the sample. If more than for 1 hr, in carbon-free distilled water and dilute 1 percent of salts is present in the sample, the to l,00()ml. blank should be prepared similarly. 6.B.2 Dual-channel analyzer 6.B.2.1 Depending on whether results are 6. Procedure desired for total or dissolved carbon, the sample 6.A Instrument standardization may be injected untreated, or filtered through 6.A.I Single-channel analyzer, Bockman the 0.45-jum silver filter, but not acidified. Model 115-A, or equivalent. 6.B.2.2 Make duplicate 20-/-J injections of 6. A. 1.1 Prepare a series of standards each sample into each channel of the analyzer, containing 5, 10, 20, 40, 60, and 80 rng/1 of using the same technique as in the preparation of carbon by dilution of the potassium hydrogen the standard curves. phthalate standard stock solution. 6.A.1.2 Successively introduce 20-^1 sam­ 7. Calculations ples of the 20-mg/l standard into the carbona­ 7. A Single-channel analyzer ceous analyzer until the response is reproducible. 7.A.I Concentrations of carbon, in milli­ Allow the recorder to return to the original grams per liter, are obtained directly from ap- 6 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS propriate standard curves, depending on the Extractive spectrophotometric method sample treatment used. 7.B Dual-channel analyzer 1. Summary of method 7.B.I Concentrations of inorganic and in­ Individual chlorophylls are determined simul­ organic-plus-organic carbon, in milligrams per taneously without elaborate separation. The liter, in each sample are obtained from the sample is filtered, and the cells retained on the appropriate standard curves. filter are mechanically disrupted to facilitate 7.B.2 By difference, compute the concentra­ extraction of pigments into a solvent composed of tion of organic carbon, in milligrams per liter, 90 percent acetone and 10 percent water (volume When this difference is small and the determined per volume). The concentrations of chlorophylls values are large, results should be verified. This are calculated from measurements of absorbance may be done by injecting an acidified, nitrogen- of the extract at 4 wavelengths, corrected for the purged sample into the high-temperature furnace 90-percent acetone blank. to obtain a direct measure of organic carbon. 2. Application

8. Report The method is suitable for all natural waters. Carbon concentrations are reported as follows: The volume of water filtered should contain less Less than 10 mg/1, one significant figure; 10 than 10 jug (micrograms) chlorophyll a. (See mg/I and above, two significant figures. step 6.1.)

3. Interferences 9. Precision No precision data are available, but results Large amounts of inorganic sediment in the are believed reproducible within ±1 mg/1 at sample may clog the filter. Erroneously high the 100-mg/l level. values may result from the presence of fragments of tree leaves or other terrestrial plant material. References The presence of phaeopigments, the decom­ position products of chlorophyll, results in over­ Beckman Instruments, Inc., 1966, Laboratory carbona­ estimates of chlorophylls (Yentsch, 1965, 1969; ceous analyzer: Beckman Instructions 137879, Fuller- ton, Calif., 27 p. Lorenzen, 1965, 1967). ———— 1968, Total organic carbon analyzer: Beckman In­ structions 81706-B, Fullerton, Calif., 34 p. 4. Apparatus Menzel, D. W., and Vaccaro, R. F., 1964, The measure­ 4.1 Filtration equipment: Filter holder as­ ment of dissolved organic and particulate carbon in sembly, Millipore XX63 001 20, or equivalent, seawater: Limnology and Oceanography, v. 9, p. 138-142. and a source of vacuum. Van Hall, C. E., Safranko, John, and Stenger, V. A., 1963, 4.2 (flass pestle-type tissue homogenize Rapid combustion method for the determination of (grinder), 15-ml capacity, Corning 7725, or organic substances in aqueous solutions: Anal. equivalent. Motor drive with low-torque clutch Chemistry, v. 35, p. 315-319. to operate at about 500 rpm (revolutions per minute). 4.3 Membrane Jilter, Millipore HAWP 047 00 Chlorophylls type, HA 0.45-//m mean pore size, white, plain, The concentrations of photosynthetic pig­ 47-mm (millimeter) diameter, or equivalent. ments in natural waters vary with time and with 4.4 Stving-out centrifuge, 4,000-5,000 gravity, changing aquatic conditions. Chlorophyll a, b, with stoppered 15-ml graduated centrifuge tubes. and c concentrations are used to estimate the Saveguard centrifuge, Model CT-1140, or equiv­ biomass and the photosynthetic capacity of alent. phytoplankton. Ratios between the different 4.5 Spectrophotomeler (Beckman Model DU, forms of chlorophyll are thought to indicate the or equivalent), with a bandwidth of 3 nm (na­ taxonomic composition or the physiological state nometers) or less, allowing absorbance to be rend of the algal community. to ±0.001 units. Use cells with a light-path of from ANALYSIS OF ORGANIC SUBSTANCES itf WATER

1 to 10 cm (centimeters). If only chlorophyll a is greater than 0.8.) If the 750-nm reading is determined, instruments with interference filters greater than 0.005/cm light-path, reduce the with not more than 5-10 nm (Beckman Model turbidity as mentioned in step 6.8. B, or equivalent) half-bandwidth may be used. 4.6 Vacuum-pressure pump, Millipore 7. Calculations XX60 000 00, or equivalent. Subtract the absorbance at 750 nm from the absorbance at 663, 645, and 630 nm. Divide the 5, Reo gents difference by the light-path of the cells, in centi­ 5.1 Acetone solution: To 900 ml acetone, add meters. The concentrations of chlorophylls in the 100 ml distilled water. extract, as jug/ml (micrograms per milliliter), are 5.2 Magnesium carbonate, powdered. given by the following equations:

6. Procedure chlorophyll a, in M 6.1 Collect a sample of water which con­ = 11.646663-2.166645 + 0.106630 chlorophyll b, in Mg/ml tains less than 10 Mg, and preferably about 1 ^g, = -3.94e663+20.97e645-3.66e63o of chlorophyll a. A sample of 0.5-1.0 liter may be chlorophyll c, in adequate for fresh or estuarine waters; 4-5 liters of ocean water may be required. Samples must be refrigerated at 4°C until time of analysis. where: 6.2 Cover the surface of a 0,45-/im membrane absorbance at 663 nm filter with finely powdered MgC03, using about 10 mg/cm2 of filter area (about 170 mg for the ______—absorbance at 750 nm 47-mm-diameter filter). light-path, in cm 6.3 Filter the sample at no more than two- thirds atm (atmosphere). absorbance at 645 nm 6.4 Fold the filter with the plankton on the ______— absorbance at 750 nm inside and proceed immediately with the extrac­ light-path, in cm tion and measurement steps. If the sample must be stored, dry the filter in a silica gel desiccator absorbance at 630 nm in the dark at 1°C or less. Dry ice is recommended _____— absorbance at 750 nm 6630 = for storing samples in the field. light-path, in cm 6.5 Place the filter in a glass homogenizer. Add 2 to 3 ml acetone solution. Grind 1 min chlorophyll a, in Mg/1 (minute) at about 500 rpm. = chlorophyll a in Mg/ml 6.6 Transfer to a graduated centrifuge tube extract volume, in ml X and wash the pestle and homogenizer two or sample volume, in liters three times with acetone solution. Adjust the total volume to some convenient value, such as 5 or 10 ml dzO.l ml. Keep 10 min in the dark at 8. Report room temperature. Report chlorophyll concentrations as follows: 6.7 Centrifuge for 10 min at 4,000 to 5,000 Less than 1.0 Mg/1 (micrograms per liter), one gravity. decimal; 1.0 /ig/1 and above, two significant 6.8 Carefully pour or pipet the supernatant figures. into the spectrophotometer cell. Do not disturb the precipitate. If extract is turbid, try to clear 9. Precision by adding a little 100-percent acetone or distilled The precision of chlorophyll determinations is water, or by recentrifuging. influenced by the volume of water filtered, the 6.9 Read the absorbance at 750, 663, 645, range of chloroph}'!! values encountered, the and 630 nm against an acetone-solution blank. volume of extraction solvent, and the light-path (Dilute with acetone solution if the absorbance is of the spectrophotometer cells. 8 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

The precision for chlorophyll a determination to correspond to the platinum-cobalt scale of at the 5-//g level was reported to be ±5 percent Hazen (1892). The unit of color is that produced in one study (Strickland and Parsons, 1960). The by 1 mg of platinum per liter. A small amount of UNESCO (1966) report recommended that pre­ cobalt is added to aid in color matching. The cision be determined by each analyst. Hazen scale is satisfactory for most waters, but the hues and shades of some waters may not References easily be compared with standards. If the hue of the water does not compare with that of the Lorenzen, C. J., 1965, A note on the chlorophyll and phaeophytin content of the chlorophyll maximum: standard, there is very'little that can be done, Limnology and Oceanography, v. 10, p. 482-483. other than to visually compare the optical densi­ ———— 1967, Determination of chlorophyll and phaeo- ties of the sample and standard. Highly colored pigments: Spectrophotometric equations: Limnology waters should not be diluted more than necessary and Oceanography, v. 12, p. 343-346. Strickland, J. D. H., and Parsons, T. R., 1960, A manual because the color of the diluted sample often is of sea-water analysis: Canada Fisheries Research not proportional to the dilution. The colored Board Bull. 125, 185 p. glass disks should be recalibrated at frequent United Nations Educational, Scientific and Cultural Or­ intervals against platinum-cobalt standards, as ganization (UNESCO), 1966, Determination of photo- their color may fade with time. synthetic pigments in sea-water: Mon. Oceanog. Methodology 1, 69 p. Yentsch, C. S., 1965, Distribution of chlorophyll and 2. Application phaeophytin in the open ocean: Deep-Sea Research, This method may be used to measure the color v. 12, p. 653-666. ———— 1969, Methods for chemical analysis of fresh of samples whose colors reasonably match the waters, in Golterman, H. L., (ed.), International Hazen scale and which contain no excessive Biological Programs Chemical Methods: Oxford, amount of suspended matter. Blackwell Scientific Pubs., handb. 8, p. 114-119. 3. Interferences Color Turbidity generally causes the observed color to be higher than the true color, but there is some The color of water as considered herein is that disagreement as to the magnitude of the effect of due only to substances in solution. Color in water turbidity. The removal of turbidity is a recurrent may be of natural mineral, animal, or vegetable problem in the determination of color. Color is origin. It may be caused by metallic substances, removed by adsorption on suspended material. humus material, peat, algae, weeds, or protozoa. Filtration of samples to remove turbidity fre­ Industrial wastes may also color water. Color quently removes some of the color-imparting may range from zero to several hundred units. solutes, possibly by adsorption on the sediments In domestic water, any noticeable color is un­ or on the filter medium. Centrifuging is preferable desirable. Color-imparting solutes may dull clothes to filtration and is the only recommended method or stain food and fixtures. The U.S. Public Health for the removal of suspended matter. Service (1962) stated that the color shall not ex­ ceed 15 units in drinking and culinary water on 4. Apparatus carriers subject to Federal quarantine regula­ tions. Color is undesirable in water for many Color comparator, with standard color disks industries, particularly food processing, launder­ covering the range 0-500 color units. ing, ice manufacturing, bottled beverage, photo­ graphic, and textile (California State Water 5. Reagents Quality Control Board, 1963). None.

6. Procedure Comparison method 6.1 Remove turbidity by centrifuging the 1. Summary of method sample. The color of the water is compared with that 6.2 Fill one instrument tube with the sample of colored glass disks which have been calibrated of water, level the tube, insert the glass plug, ANALYSIS OF ORGANIC SUBSTANCES IN WATER 9 making sure that no air bubbles are trapped, and spills (Dept. of Interior, Office of Secretary, insert the tube into the comparator. 1968). Most oils are insoluble in water but may 6.3 Use distilled or demineralized water in be dispersed by natural and waste chemicals such the second tube as a blank. as soaps and detergents. Ester-type compounds 6.4 The color comparison is made by re­ can hydrolyze to become soluble, form soaps, and volving the disk until the colors of the two tubes further aid dispersion of the insoluble material. match. Oils, greases, fats, and waxes can severely damage water quality by (1) producing a visible film on 7. Calculations the surface, (2) imparting an odor to the water The color is read directly from the matching and causing a noxious taste, (3) coating the banks color standard, and the proper dilution factor is and bottoms of the water body by adsorbing on applied. sediment, and (4) destroying aquatic life (Federal Water Pollution Control Administration, 1968). 8. Report Keport color as follows: 1. Summary of method Oils and waxes are removed from the water by Color unit Record units to nearest extraction with organic solvents and are deter­ 1-50- ..... 1 mined gravimetrically. This method is similar in 51-100_ ..._. 5 principle to that found in "Standard Methods," 101-250. .... 10 12th edition (Am. Public Health Assoc., 1965). 251-500. ._.__ 20 For characterization, and for volatile substances, the analyst is referred to the papers by Kawahara 9. Precision (1969) and Maehler and Greenber (1968). Because of the many complicating factors in­ volved, the measurement of color is not a precise 2. Application determination. No statements on the repro- The extraction method is applicable to the ducibility of the tests can be made. analysis of waters containing oil, fat, grease, wax, and other solvent-soluble substances. References 3. Interferences California State Water Quality Control Board, 1963, Water quality criteria: Pub. 3-A, p. 168. Organic solvents vary considerably in their Hazen, Alien, 1802, A new color standard for natural ability to dissolve not only oily substances but waters: Am. Chem. Soc. Jour., v. 12, o. 427. other organic matter as well. The ester-type oils U.S. Public Health Service, MI62, Drinking water stand­ and waxes may be decomposed to alcohols and ards: Public Health Service Pub. 956, p. 6. acids or soaps. Glycerine, the alcohol from animal or vegetable fats and oils which are known as triglycerides, defies extraction by this technique. Extractable organic matter Soaps must be acidified to release the organic part of the molecule. Storage of the sample must Oils and waxes be avoided because many oils arc utilized by Oils and waxes in natural waters most likely micro-organisms. Volatile compounds (boiling come from vegetation and aquatic life. Oils or point <100°C) cannot be accurately determined fats and waxes from plants and animals are, for by this method. the most part, of the ester type—that is, the combination of an alcohol with an organic acid. 4. Apparatus Petroleum oils and tars, otherwise known as 4.1 Kuderna-Danish concentration apparatus, mineral oils, are almost exclusively hydrocarbon 250-ml capacity. Use an ungraduated receiver and in composition. Petroleum oils, on occasion, can a 1-ball Snyder column. enter the water from natural seeps but most often 4.2 Separatory funnel, 1-liter capacity, with result from industrial pollution and accidental unlubricated glass or teflon stopcock. 10 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

5. Reagents chloroform and add the washings to the separatory 5.1 Chloroform, boiling range 61°-62°C, dis­ funnel. Pour an additional 15 ml of chloroform tilled in glass. into the separatory funnel and shake 1-2 min. 5.2 Petroleum ether, boiling range 30°-60°C, Allow the layers to separate. Collect the chloro­ distilled in glass. form layer in the 125-ml erlenmeyer flask con­ 5.3 Sodium sulfate, anhydrous, ACS re­ taining the petroleum ether extract. agent, granular. 6.6 Filter the extract through a plug of ex­ 5.4 Sulfuric acid, 1:1 solution: Mix a volume tracted glass wool into a Kuderna-Danish ap­ of sulfuric acid (sp gr 1.84) with an equal volume paratus fitted with a tared (including a small of distilled water. boiling stone) receiver tube. Use liberal washings of petroleum ether to complete the transfer. 6.7 Remove the solvent by distillation on a 6. Procedure steam bath and take nearly to dryness. As the Surface samples must be taken with great care. last of the solvent is vaporizing, remove the (See p. 2.) The relationship of the affected sur­ apparatus from the heat and allow the vapor to face to the body of water should be well docu­ condense and wash down the sides of the flask. mented. A complete written description of the After cooling, remove the receiver and volatilize sampling technique should be made for future the last of the solvent with a gentle stream of air reference. Samples of the water beneath the or nitrogen gas. surface should be collected according to the 6.8 Wipe the sides of the receiver tube with recommended practice for organic samples. Usu­ a moist tissue and allow to dry in the room for ally 1-liter samples are sufficient for subsurface 30 min. Place the receiver in a desiccator for 30 sampling. min and then determine the weight of the sample. 6.1 Weigh the water and sample container with the cap removed. Pour the sample into the 7. Calculations separatory funnel. Weigh the drained sample bottle and record the weight of the sample to The concentration of material in the sample is three significant figures. (If an oil separation is calculated as follows: observed, the weight of water may be obtained Extractable organic matter (mg/1) after a preliminary petroleum ether extraction before acidification.) A-B X 1,000, 6.2 Rinse the sample bottle with 15 ml of petroleum ether and pour the washings into the separatory funnel. Rinse the sample bottle again where with another 15 ml pf petroleum ether and pour A = total weight (receiver+extract), in grams, the washings into the separatory funnel. B —receiver tare, in grams, and 6.3 Acidify the sample by the addition of C —water sample volume, in liters. 5.0 ml of 1:1 H2S04 per liter of sample. 6.4 Shake the separatory funnel vigorously The amount of material found in a surface sample for 1-2 min, stopping to vent the pressure after may be related to an affected area provided the the first few shakes. Allow the layers to separate film is uniform. The weight of material per for about 10 min. Draw off the water layer into square meter of surface may be calculated as the sample container and pour the solvent into follows: a 125-ml erlenmeyer flask containing about 0.5 g anhydrous sodium sulfate. Difficult emulsions Extractable organic matter (mg/m2) = — may be broken by shaking the funnel vigorously A. after most of the water has been withdrawn. where Beware of excessive pressure buildup in the funnel at this step. A =area sampled, in square meters (m2), and 6.5 Pour the sample back into the separatory B = weight of extracted material, in milli­ funnel. Rinse the sample bottle with 15 ml of grams. ANALYSIS OF ORGANIC SUBSTANCES IN WATER 11

8. Report on sediment, or floating on the surface. In addi­ Concentrations of extractable organic matter tion to acting synergistically to compound pollu­ are reported to two significant figures for values tion problems, some detergents are toxic to in excess of 10 mg/1, or to the nearest whole certain aquatic life. According to the report of milligram. the Federal Water Pollution Control Adminis­ tration (1968) levels of LAS (linear alkylate 9. Precision sulfonates) exceeding 0.2 mg/1 may be harmful The precision between samples depends upon to aquatic life. the homogeneity of the dispersion and the Three types of surface-active agents are in volatility of the material. Duplicate results vary­ general use. They are classified by their chemical ing less than 5 percent from the mean have been characteristics as anionic or cationic, indicating obtained on prepared samples. Results from the charge of the active ions, and as nonionic poorly dispersed material and surface film hydrophilic organic compounds. Fortunately, samples vary greatly. anionic detergents react with cationic detergents to precipitate. Production of anionic detergents References has greatly exceeded the other two, and, conse­ quently, greater amounts of anionic detergents American Public Health Association, 1965, Standard may be expected in waste water. Owing to this methods for the examination of water and waste- excess, cationic detergents are almost always water [12th ed.]: New York, Am. Public Health Assoc., Inc., 769 p. effectively precipitated and seldom appear in Kawahara, F, K., 1969, Laboratory guide for the identifica­ solution in sewage treatment facilities. Synthetic tion of petroleum products: Cincinnati, Federal Water nonionic detergents, alkyl or aryl polyols, have Pollution Control Admin., 41 p. not caused much concern because their use has Maehler, C. A., and Greenber, A. E., 1968, Identification been limited principally to industrial applications of petroleum in estuarine waters: Am. Soc. Civil Eng. Proc., Jour. Sanitary Eng. Div., v. 94, No. SA5, and as a minor additive in household detergents p. %9-978. for foam control. [U.S.] Department of the Interior, Office of the Secre­ Until 1965, ABS (alkyl benzene sulfonates) tary, 1968, Oil pollution; A report on pollution of anionic detergents were the most common the Nation's waters by oil and other hazardous sub­ surfactants on the market. Because ABS resist stances: Washington, U.S. Govt. Printing Office, 31 p. [U.S.] Federal Water Pollution Control Administration, biological degradation, they are not easily re­ 1968, Report of the committee on water-quality moved by waste treatment and were often found criteria: Washington, U.S. Govt. Printing Office, in water supplies, even drinking water. Since 234 p. then, LAS (linear alkyl sulfonates), for the most part, have replaced ABS in detergent formulation because LAS surfactants are biodegradable and Methylene blue active substances may be removed from waste water by ordinary treatment. Synthetic anionic detergents Both ABS and LAS are methylene blue active Methylene blue active substances (MBAS), substances but are not identical in color produc­ synthetic detergents or surfactants, occur in tion for spectrophotometric determination. Owing natural waters almost exclusively as a result of to this, methylene blue active substances are pollution. Detergents in natural water can measured in units relative to an ABS standard. drastically alter the normal regimen. As surface- active agents, they disrupt the stability at the 1. Summary of method interface between water and other substances. Methylene blue reacts with anionic surfactants, In the presence of surfactants, normally insoluble both ABS and LAS, to form a blue-colored dye organic compounds may be dispersed, and the complex. The complex is extracted in chloroform, function of vital membranes of aquatic organisms and the methylene blue active substances are de­ may be altered. Because they are water soluble, termined spectrophotometrically. This method is detergents can disperse toxic organic compounds similar in substance to "Standard Methods/' 12th normally incorporated in bottom muds, adsorbed edition (Am. Public Health Assoc., 1965).

488-959 O - 72 - 3 12 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

2. Application 5.3 Detergent standard solution II, 1.00 This method is applicable to the analysis of ml = 0.010 mg ABS: Dilute 10.0 ml detergent waters containing 0.025-100 mg/1 MBAS (meth- standard solution I to 1,000 ml. Prepare fresh ylene blue active substances relative to ABS weekly. standard). 5.4 Methylene blue reagent: Dissolve 0.35 g methylene blue in O.OLV sulfuric acid and dilute 3. Interferences to 1,000 ml with same. Phenols, proteins, and inorganic chlorides, 5.5 Sulfuric acid solution, 5A': Mix 245 ml cyanates, nitrates, and thiocyanates will complex concentrated H2S04 (sp gr 1.84) in 500 ml water methylene blue and give positive interference in and, after c'ooling, dilute to 1,000 ml. the determination. With ABS concentrations from 0.0 to 1.0 mg/1, tests have shown no inter­ 6. Procedure ference from the following individual constituents: Samples should be collected according to the 10 mg/1 N02, 25 mg/1 NO3, 5 mg/1 phenol, and recommended practice for organic analysis. Usu­ 1 mg/1 H_S. Organic compounds having amine ally a 1-liter sample is adequate. LAS detergents groups cause low results. are biodegradable; therefore samples should be analyzed as soon as possible after collection. All 4. Apparatus glassware should be rinsed with dilute hydro­ 4.1 Separatory funnels, 500-ml, unlubricated chloric acid after washing. As a further precau­ glass or teflon stopcock. tion, sample bottles, separatory funnels, and 4.2 Spedrophotometer, Beckman Model B, or beakers may be heat treated at 300°C overnight equivalent. to remove organic matter. With this instrument the following conditions 6.1 Measure a volume of sample containing have been used: less than 0.10 mg MBAS and not to exceed 100 ml into separatory funnel. Dilute to 100 ml if Wavelength._.._..______._-„___ 635 nm. necessary. Prepare a blank and detergent stand­ Cells_----_------_------10mm. ards in the same manner. Phototube.__.___...___-._____--___ Red-sensitive. Initial sensitivity setting ______2. 6.2 Adjust the pH of the sample to near Slit width.______0.2mm. neutral if necessary. To the sample, blank, and Blank___,______„______CHC1S. standard solution, add 1.0 ml 5A~ H2S04, mix, then add 5.0 ml methylene blue solution. Mix With these operating conditions, the following thoroughly. absorbances have been observed: 6.3 Add 25.0 ml chloroform and shake the contents of the separatory funnel for 1 min. MBAS (as ABS) Absorbance (mg) Allow the layers to separate. 0,01 _...... „...... __....„ 0.052 6.4 Drain the lower chloroform layer through .02____.____.__-__-__-______.150 a layer of absorbent cotton contained in a small .05...... -...... -..-...---. -438 filtering funnel into 1.0-cm cells and measure the .10...______.870 absorbance of the sample and standards against the blank at a wavelength of 655 nm. 5. Reagents 5.1 Chloroform, spectrophotometric grade. 7. Calculations 5.2 Detergent (ABS) standard solution I: Determine the amount of detergent contained Dissolve 1.030 g 97.1-percent-purity ABS powder in the sample, minus the blank, from the standard (obtained from Association American Soap and curve. Calculate the amount of MBAS in the Glycerine Producers, Inc., 295 Madison Ave., sample using the following equation: New York, N. Y. 10010, and dried over H2SO^ for 1 week) in 500 ml distilled water in a 1,000-ml MBAS (relative to ABS standard) mg/I volumetric flask with gentle swirling. Allow the mg MBAS foam to break and dilute to volume. Refrigerate X 1,000 during storage. ml sample ANALYSIS OF ORGANIC SUBSTANCES IN WATER 13

8. Report and an aliquot of the distillate is then Nesslerized. Report synthetic anionic detergents concentra­ Essentially, Nesslerization is the reaction between tions as follows: Less than 1 mg/1, two decimals; potassium mercuric iodide and ammonia which greater than 1 mg/1, two vsignificant figures. forms a red-brown complex of mercuric ammono- basic iodide:

9. Precision *Hg(NHa)I+2KI + HI. Deviation of ±10 percent may be expected in the range of 1-5 mg/1 in surface water. Concentrations of ammonia are then determined by standard spectrophotometric measurements. References Alternatively, the distillate may be titrated with American Public Health Association, 11)05, Standard standard sulfuric acid solution. methods for the examination of water and wastewater Additional information on the principle of the [12th ed.]: New York, Am. Public Health Assoc., determination is given by Kolthoff and Sandell Inc., 769 p. (1952, p. 633) and by Blaedel and Meloche (1963). [U.S.] Federal Water Pollution Control Administration, 1968, Report of the committee on water-quality criteria: Washington, U.S. Govt. Printing Office, 2. Application 234 p. This method is recommended for analysis of samples containing less than 2 mg of ammonia and ammonium ion per liter. Higher concentra­ Nitrogen, ammonia tions may be determined by the alternate titra- tion procedure provided. Ammonia nitrogen includes nitrogen in the forms of NH3 and NH4+1. As a component of the 3. Interferences nitrogen cycle, it is often present in water, but Calcium, magnesium, iron, and sulfide interfere usually in only small amounts. Ammonia is used with the Nesslerization, but the interference of in some water-treatment processes. More than the metals is eliminated by the distillation, and 0.1 mg/1 usually indicates organic pollution sulfide can be precipitated in the distillation flask (Rudolph, 1931). with a little lead carbonate. There is no evidence that ammonia nitrogen in Some organic compounds may distill over with water is physiologically significant to man or the ammonia and form colors with Nessler re­ livestock. Fish, however, cannot tolerate large agent which cannot be satisfactorily read with quantities. The toxicity to fish is directly related the spectrophotometer. Under such conditions, to the amount of free ammonia in solution; hence, the sample should be distilled into H3B03 and the toxicity is dependent on the pH of the water. titrated with standard H2SO4. Ammonia decreases the ability of hemoglobin to combine with oxygen, and the fish suffocate. Al­ 4. Apparatus though the tolerances of fish differ, 2.5 ml/1 of ammonia nitrogen is considered harmful in the 4.1 Kjeldahl distillation apparatus, 1,000-ml 7.4-8.5 pH range (Ellis and others, 1948). flasks. The low concentrations of ammonia in natural 4.2 Spectrophotometer, Beckman Model B, or waters are of little industrial significance, except equivalent. that ammonium salts are destructive to concrete. With this instrument, the following operating conditions have been used:

Distillation method Wavelength ______425 nm. Cells__, ______40 mm. 1. Summary of method Phototube. __.._--_._..._,_ Blue-sensitive. The sample is buffered to a pH of 9.5 to mini­ Initial sensitivity setting _ _ _ _ 1. Slit width. _..„__..__..._-. 0.3 mm. mize hydrolysis of organic nitrogen compounds. Blank. . _-__--_._-______,_ Ammonia-free water Ammonia is distilled from the buffered solution, plus reagents. 14 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

With these operating conditions, the following be neutralized with NaOH solution before pro­ absorbances have been observed: ceeding with the analysis. All glassware should be rinsed with ammonia-free water, prepared by \itrogen Absorbance passing distilled water through a mixed-bed ion- (mg} 0.02___. 0.24 exchange resin. .47 6.1 Free the distillation apparatus of am­ .06. .70 monia by boiling ammonia-free water until the .10. 1.16 distillate shows no trace using Nessler reagent (CAUTION: Deadly poison). 5. Reagents 6.2 Measure a volume of sample containing 5.1 Ammonium chloride standard solution I, less than 1.0 mg ammonia nitrogen (500 ml 1.00 ml-1.00 mg N: Dissolve 3.819 g NH4C1, maximum) into a 600-ml beaker, and adjust the dried overnight over sulfuric acid, in ammonia- volume to approximately 500 ml with ammonia- free water and dilute to 1,000 ml. free water. 5.2 Ammonium chloride standard solution II, 6.3 Add 25 ml borate buffer solution and 1.00 ml= 0.010 mg N: Dilute 10.00 ml ammonium adjust the pH to 9.5 with IM NaOH, if necessary. chloride standard solution I to 1,000 ml with 6.4 Immediately transfer the solution into ammonia-free water. Prepare fresh. the distillation flask and distill at a rate of 6-10 5.3 Borate buffer solution: Dissolve 9.54 g ml per minute; catch the distillate in a 500-ml Na2B407'10H20 in ammonia-free water. Adjust volumetric flask containing 50 ml boric acid the pH to 9.5 with IM NaOH (approx 15 ml) solution. The tip of the delivery tube must be and dilute to 1 liter with ammonia-free water. below the surface of the boric acid. 5.4 Boric acid solution: Dissolve 20 g H3BO3 6.5 Collect approximately 250 ml of dis­ in ammonia-free water and dilute to 1 liter. tillate, dilute to 500 ml with ammonia-free 5.5 Methyl red indicator solution: Dissolve water, and mix. 0.1 g methyl red indicator in 100 ml ethanol. 6.6.A Xesslerization procedure 5.6 Xessler reagent (CAUTION: HgI2 is a 6.6.A.I Pipet an aliquot of distillate contain­ deadly poison, and the reagent must be so ing less than 0.1 mg ammonia nitrogen (50.00 ml marked): Dissolve 100 g HgI2 and 70 g KI in a maximum) into a graduate, and adjust the volume small volume of ammonia-free water. Add this to 50.0 ml with ammonia-free water. mixture slowly, with stirring, to a cooled solution 6.6.A.2 Prepare a blank of ammonia-free of 160 g NaOH in 500 ml ammonia-free water and water and sufficient standards. Add 5 ml boric dilute to 1 liter. Allow the reagent to stand at acid solution to each, and adjust the volumes to least overnight and filter through a fritted-glass 50.0 ml. crucible. 6.6.A.3 Add 1.0 ml Nessler reagent (CAU­ 5.7 Sodium carbonate solution, 0.0357 A': Dis­ TION: Deadly poison), and mix. solve 1.892 g primary standard Na^COs in carbon 6.6.A.4 Allow the solutions to stand at least dioxide free water and dilute to 1,000 ml. 10 min, but not over 30 min. 5.8 Sodium hydroxide solution, LI/: Dissolve 6.6.A.5 Determine absorbance of test sample 40 g NaOH in ammonia-free water and dilute to and standards against the blank. 1 liter. 6.6.B Titration procedure 5.9 Sulfuric acid standard solution, 0.0357A', 6.6.B.I To the distillate, and an ammonia- 1.00 ml = 0.5 mg X: Mix 1.3 ml concentrated free water blank containing the same volume of H2SOi (sp gr 1.84) with demineralized water and H3B03, add 3 drops methyl red indicator solution, dilute to 950 ml before standardization. Stand­ :md titrate with sulfuric acid standard solution. ardize by titrating 25.0 ml 0.0357A' Na,CO3 to pH 4.5. 7. Calculations 6. Procedure 7.A Xesslerization procedure If acid has been added to the sample as a 7.A.I Determine mg N in aliquot from a plot preservative at the time of collection, this must of absorbances of standards. ANALYSIS OF ORGANIC SUBSTANCES IN WATER 15

7.A.2 Ammonia nitrogen as N, in mg/1 few tenths to several hundred milligrams per liter, but in unpolluted water they seldom exceed 1,000 500 = ————— X —~Tr~—— Xmg N in aliquot, 10 mg/1. Nitrate and chloride are major compo­ ml sample ml aliquot nents of human and animal wastes, and abnor­ 7.B Titration procedure mally high concentrations of both suggest 7.B.1 Ammonia nitrogen as N, in mg/1 pollution. Cyanosis due to methemoglobinemia may occur in infants whose drinking or formula water V. contains a high concentration of nitrates. The where nitrates, when ingested, are converted to nitrites in the digestive system of some infants. The ya = ml standard H2S()4 used in titration of nitrite ion oxidizes hemoglobin to methemoglobin sample minus ml used to titrate blank, and thereby causes cyanosis. It is widely recom­ Ntt = normality of standard H^SO-i, and mended that water containing more than 10-20 V8 = m\ of original sample used for distillation. mg/1 of nitrate nitrogen should not be used in 7.C Ammonia nitrogen as NH4+1, in mg/I = infant feeding (Comly, 1945). mg/1 as NX 1.288. Ammonia nitrogen as free Nitrates in large amounts are injurious to the NH3, in mg/1 = mg/1 as NX 1.216. dyeing of wool and silk and are undesirable in fermentation processes (California State Water 8. Report Quality Control Board, 1963). At least 2 mg/1 of Report ammonia nitrogen concentrations as nitrate prevents intercrystalline cracking of steel follows: Less than 1 mg/1, two decimals; 1 mg/1 in steam boilers. and above, two significant figures. Brucine method 9. Precision 1. Summary of method No precision data are available. The reaction between the alkaloid, brucine, and nitrate in acid medium produces a yellow References color that may be measured by standard spectro- Blaedel, W. J., and Meloche, V. W., 1963, Elementary photometric procedures. Close attention must be quantitative analysis—Theory and practice [2d ed.]: given to procedural technique if accuracy and New York, Harper & Row, 826 p. Ellis, M. M., Westfall, B. A.j and Ellis, M. D., 1948, precision are to be obtained. The procedure is Determination of water quality: U.S. Fish Wildlife similar to that of Jenkins and Medsker (1964). Service Research Kept. 9, 122 p. Kolthoff, I. M., and Sandell, E. B., 1952, Textbook of 2. Application quantitative inorganic analysis [3d ed.]: New York, This method may be applied to essentially Macmillan Co., 759 p. Rudolph, Z., 1931, Principles of the determination of the colorless water containing up to 5.0 mg of nitrate physical and chemical standards of water for drink­ per liter. Any significant amount of color should ing, industrial, and domestic purposes: Water Pollu­ be removed. Samples containing higher concen­ tion Abs, 4 [March], trations must be diluted.

3. Interferences Nitrogen, nitrate Organic color, nitrite ion, and all strong Nitrate is usually the most prevalent form of oxidizing and reducing agents interfere. The nitrogen in water because it is the end product interference by residual chlorine up to 5 mg/1 of the aerobic decomposition of organic nitrogen. may be eliminated by addition of sodium arsenite, Nitrate from natural sources is attributed to the and interference of up to 1 mg/1 of nitrite elimi­ oxidation of nitrogen of the air by bacteria and nated by use of sulfanilic acid. The interference to the decomposition of organic material in the by chloride is effectively masked by the addition soil. Fertilizers may add nitrate directly to water of a large amount of chloride ion to the reaction resources. Nitrate concentrations range from a mixture. 16 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

4. Apparatus 6.3 Add 2.0 ml sodium chloride solution, 4.1 Water bath, boiling. and mix well by swirling. 4.2 Spectrophotometer, Beckman Model B, or 6.4 Place the absorption tubes into a cold- equivalent. water bath (15°-20°C) and add 10.0 ml 29N With this instrument, the following operating HsSOa. Mix well by swirling, return to cold-water conditions have been used: bath, and allow the contents of the tubes to cool to water-bath temperature. Wavelength...____._.___--__.__ 410 nm. 6.5 Add 0.5 ml brucine-sulfanilic acid solu­ Cells..-._---____------_------23 mm. tion and mix thoroughly. Phototube-_._____-._-_._-._..-_ Blue-sensitive. NOTE.—If a deep-pink color forms immediately Initial sensitivity setting, ______2. Slit width (approximate)___-__.__ 0.10 mm. upon addition of the brucine-sulfanilic acid re­ agent, it is an indication of a high nitrate con­ With these operating conditions, the following centration—exceeding the range of the method. absorbances have been observed: Such sample aliquots must be discarded at this point and the samples reanalyzed, using a smaller NO3 . I bsorbn nee aliquot. (mg) 6.6 Remove the rack of tubes from the cold- 0.005. 0.115 water bath and place in a boiling-water bath for .220 .010. 20 min. The water bath must be sufficiently large .020, .440 .030. .640 so that boiling does not cease when the tubes are .040. .800 placed in it. This step is critical. All tubes must .050. .950 be heated uniformly. 6.7 Remove the tubes from the boiling-water 5. Reagents bath and immerse them in the cold-water bath. 5.1 Brucine-sulfanilic acid reagent: Dissolve Allow to cool before proceeding. This inhibits any 1 g brucine sulfate (CAUTION: Very poisonous) further color change. The cold-water bath must and 0.1 g sulfanilic acid in 70 ml hot demineralized be sufficiently large to cool all tubes uniformly. water. Add 3 ml concentrated HC1 (sp gr 1.19), Circulation of water in the bath is desirable. cool, and dilute to 100 ml. This solution is stable 6.8 Determine the absorbance of the sample for several months. The pink color that develops and standards against the blank within 1 hr. does not affect its usefulness. 5.2 Nitrate standard solution I, 1.00 ml = 1.00 7. Calculations mg N03 : Dissolve 1.631 g KNO3, dried overnight 7.1 Determine the mg NO3 in the sample over concentrated H^SO-i, in demineralized water from n plot of absorbances of standards. and dilute to 1,000 ml. 1 000 5.3 Nitrate standard solution II, 1.00 ml = 7.2 N03 in mg/I= —.-——— Xmg NO3 in 0.010 mg NO3 : Dilute 10.0 ml nitrate standard ml sample solution I to 1,000 ml with demineralized water. sample. 5.4 Sodium chloride solution: Dissolve 300 g 7.3 To convert NO3 to N, multiply by 0.2259. NaCl in 1 liter demineralized water. 5.5 Sulfuric acid, 29Ar : Add 500 ml concen­ 8. Report trated H2SC>4 (sp gr 1.84) to 125 ml demineralized water. Report NO3 concent rations us follows: Loss than 10 mg/1, one decimal; 10 mg/1 and above, 6. Procedure two significant figures. 6.1 Pipet a volume of sample containing less than 0.05 mg NO3 (10.0 ml maximum) into a 9. Precision 23-mm absorption cell and dilute to 10.0 ml. Single-laboratory analysis of two test samples 6.2 Prepare a demineralized-water blank resulted in mean values of 0.9 tind 2.9 mg/1, and and sufficient standards and adjust the volume standard deviations of 0.09 and 0.11 mg/1, of each to 10.0 ml. respectively. ANALYSIS OF ORGANIC SUBSTANCES IN WATER 17

Reduction method 2. Application Samples containing more than 30 mg/1 of This method may be applied to samples con­ NOs"1 may be analyzed by reduction using taining less than 4 mg nitrite per liter. Samples Devarda's alloy, distillation of the resulting containing higher concentrations must first be NHs, and titration with standard H2S04 solution. diluted. Details of the procedure are given by Blaedel 3. Interferences and Meloche (1963) and by Kolthoff and Sandell (1952). For high concentrations of nitrate, the None of the substances commonly occurring in method yields results which are comparable in natural water interferes with this method. accuracy to those obtained by the brucine method. 4. Apparatus References Spectropliotometer, Beckman Model B, or equivalent. Blaedel, W. J., arid Meloche, V. Wv 1963, Elementary With this instrument, the following operating quantitative analysis—Theory and practice [2d ed.]: New York, Harper & Row, 826 p. conditions have been used: California State Water Quality Control Board, 1963, Water quality criteria: Pub. 3-A, p. 226. Wavelength...... _..._._.._.... 535 nm. Comly, H. H., 1945, Cyanosis in infants caused by nitrates Cells-. „_-.-__-_..-___...--____ 10 mm. in well water: Am, Med. Assoc* Jour., v. 129. Phototube,._____.---____--.____ Blue-sensitive. Jenkins, D., and Medsker, L. L., 1964, Brucine method Initial sensitivity setting__..__--__ 2. for determination of nitrate in ocean, estuarine, and Slit width. _..____.____-__.___--.. 0.08 mm. fresh waters: Anal, Chemistry, v. 36, p, 610. Kolthoff, I. M., and Sandell, E. B., 1952, Textbook of With these operating conditions, the following quantitative inorganic analysis [3d ed.]: New York, Macmillan Co., 759 p. absorbances have been observed.

NO* Absorbance (mg) Nitrogen, nitrite 0.05, 0.53 .10. 1.06 Nitrite is unstable in the presence of oxygen .15. 1.59 and is, therefore, absent or present in only .20. 2.04 minute quantities in most natural waters under aerobic conditions. The presence of nitrite in 5. Reagents water is sometimes an indication of organic 5.1 Formic acirf, 87-90 percent. pollution. 5.2 1-naphthylethylenediamine dihydrochloride Recommended tolerances of nitrite in domestic solution: Dissolve 0.5 g 1-naphthylethylene­ water supplies differ widely. A generally accepted diamine dihydrochloride in 100 ml demineralized limit is 2 mg/1, but as little as 0.1 mg/1 has been water. Store in refrigerator. proposed (California State Water Quality Control 5.3 Xitriie standard solution I, 1.00 ml = 1.00 Board, 1963). Nitrite is undesirable in water used mg NO2 : Dissolve 1.850 KNO2 in demineralized in dyeing wool and silk and in brewing. water and dilute to 1,000 ml. 5.4 Nitrite standard solution II, LOO ml = Diazotization method 0.010 mg N02 : Dilute 10.0 ml nitrite standard solution I to 1,000 ml with demineralized water. 1. Summary of method 5.5 Sulfanilamide solution: Dissolve 0.5 g Nitrite is diazotized with sulfanilamide, and sulfanilamide in 100 ml demineralized water. the resulting diazo compound is coupled with 1-naphthylethylenediamine dihydrochloride to 6. Procedure form an intensely colored red dye (Rider and 6.1 Pipet a volume of sample containing less Mellon, 1945). The absorbance of the dye is than 0.20 mg N02 (50.0 ml maximum) into a proportional to the amount of nitrite present 100-ml volumetric flask and adjust the volume (Fishman and others, 1964). to 50 ml with demineralized water. 18 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

6.2 In a similar manner, prepare a blank and and waste from slaughter houses and chemical sufficient standards containing 0,00-0.20 mg plants often contain nitrogen in varying com­ N02, and adjust the volume of each to 50 ml binations. Organic nitrogen in unpolluted ground with demineralized water. Place in ice bath and water is usually very low. allow to cool for about 2 min. Organic nitrogen is not pathologically signifi­ 6.3 Add successively, while in the ice bath, cant but is sometimes an indication of pollution. and mixing thoroughly after each addition: 1.0 Organic nitrogen is important to considerations ml sulfanilamide solution, 4.0 ml formic acid, involving aquatic biology. and 1,0 ml 1-naphthylethylenediamine dihydro- chloride solution. KjeJdoh! method 6.4 Remove the flasks from the ice bath and allow at least 15 min for maximum color develop­ 1. Summary of method ment. Adjust each to exact volume with de- Organic nitrogen is degraded to the ammonium mineralized water. Mix thoroughly and measure ion by digestion with sulfuric acid in the presence the absorbance of samples and standards against of copper sulfate, which acts as a catalyst. The the blank. solution is made alkaline with sodium hydroxide, and the free ammonia is distilled off and Nessler- 7. Calculations ized, The color developed is proportional to the 7.1 Determine the mg N02 in the test organic nitrogen content. samples from a plot of absorbances of standards. Additional information on the principle of the 1,000 determination is given by Kolthoff and Sandell 7.2 N02 inmg/l = Xmg N02 in (1952, p. 537). ml aliquot sample. 2. Application 7.3 To convert N02 to N, multiply by 0.3043. This method may be applied to most natural water containing less than 2 mg of nitrogen per 8. Report liter. Higher concentrations must be reduced by Report NO2 concentrations as follows: Less dilution. than 1.0 mg/1, two decimals; LO mg/1 and above, two significant figures. 3. Interferences Nitrate and nitrite do not interfere. The effect 9. Precision of ammonium ions and ammonia is strictly Xo precision data are available. additive. Therefore, the organic nitrogen is normally determined on the residue of the am­ References monia nitrogen determination. California State Water Quality Control Board, 1963, Calcium, magnesium, iron, and sulfide interfere Water quality criteria: Pub. 3~A, p. 226. with the Nesslerization, but the interference of Fishman, M. J-, Skougstad, M. W., and Scarbro, G. F., the metals is eliminated by the distillation. 1964, Diaxotixation method for nitrate and nitrite: Sulfides interfere and must be precipitated in the Am. Water Works Assoc. Jour., v. 56, p. 633. distillation flask with a little lead carbonate be­ Rider, B. F., and Mellon, M. G., 1945, Colorimetric de­ termination of nitrites: Indus. Eng. Chemistry, Anal. fore addition of sodium hydroxide. Ed., v. 18, p. 76. Some organic compounds may distill over with the ammonia and form colors with Nessler re­ agent which cannot be satisfactorily read with Nitrogen, organic the spectrophotometer. Under such conditions, Organic nitrogen includes all nitrogenous or­ the sample should be distilled into H3BO3 and ganic compounds, such as amino acid, poly- titrated with standard H^SOj. peptides, and proteins. It is present naturally in all surface waters as the result of inflow of nitroge­ 4. Apparatus nous products from the watershed and the normal 4.1 Kjeldald distillation apparatus, 1,000-ml biological life of the stream. Effluents of sewage flasks. ANALYSIS OF ORGANIC SUBSTANCES IN WATER 19

4.2 Spectrophotometer, Beckman Model B, or 5.8 Sulfuric aczV/, concentrated (sp gr 1.84). equivalent. With this instrument, the following operating 6. Procedure conditions have been used: All glassware should be rinsed with ammonia- free water. Wavelength______,_„_„_ 425 nm. 6.1 Free the distillation apparatus of am­ Cells...-.._.._...._...._.. 40 mm. monia by boiling ammonia-free water until the Phototube^______--_..___,._ Blue-sensitive. distillate shows no trace using Nessler reagent Blank.._.__.___..._...._-_ Ammonia-free water plus reagents. (CAUTION: Deadly poison). Initial sensitivity setting. . . _ 1. 6.2 The residue from the ammonia nitrogen Slit width (approximate) ____ 0.3 mm. determination may be used for this determination. Alternatively, buffer a volume of sample contain­ With these operating conditions, the following ing less than 1.0 mg organic nitrogen (500.0 ml absorbances have been observed: maximum) to pH 9.5 with borate buffer solution and evaporate to approximately 20 percent of N Absorbance original volume to drive off ammonia. (mg) 6.3 Cool; add 10 ml concentrated H2S04 and 0.02_ 0.24 .04. .47 1 ml CuS04 solution. .06. .70 6.4 Digest under a hood until copious fumes .10- 1.16 are given off and the liquid becomes colorless or pale yellow. 5. Reagents 6.5 Cool and dilute to approximately 300 ml 5.1 Ammonium chloride standard solution I, with ammonia-free water. 1.00 ml-1.00 mg N: Dissolve 3.819 g NHiCl, 6.6 Add 50 ml 10.V NaOH cautiously down dried overnight over sulfuric acid, in ammonia- the side of the flask, free water and dilute to 1,000 ml. 6.7 Immediately connect the flask to the 5.2 Ammonium chloride standard solution II, distillation apparatus, and cautiously mix the LOO ml = 0.010 mg N: Dilute 10,00 ml NH4C1 contents by swirling gently. standard solution I to 1,000 ml with ammonia- 6.8 Distill at a rate of no more than 10 ml free water. Prepare fresh. nor less than 6 ml per minute; catch the distillate 5.3 Borate buffer solution: Dissolve 9.54 g in a 500-ml volumetric flask containing 50 ml l\a2B407'10H20 in ammonia-free water. Adjust boric acid solution. The tip of the delivery tube the pH to 9.5 with \M NaOH (approx 15 ml) and must be below the surface of the boric acid. dilute to 1 liter with ammonia-free water. 6.9 Collect approximately 250 ml distillate, 5.4 Boric acid solution: Dissolve 20 g H3B03 dilute to 500 ml with ammonia-free water, and in ammonia-free water and dilute to 1 liter. mix. 5.5 Copper sulfate solution: Dissolve 10 g 6.9.A Xesslerization procedure. Proceed as CuSO4 -5H20 in ammonia-free water and dilute directed in "Nitrogen, Ammonia/' steps 6.6.A.1- to 100 ml. 6.6.A.5. 5.6 Xessler reagent (CAUTION: HgI2 is a 6.9.B Titration procedure. Proceed as directed deadly poison, and the reagent must be so in "Nitrogen, Ammonia/' step 6.6.B.I.

marked): Dissolve 100 g HgI2 and 70 g KI in a 7. Calculation* small volume of ammonia-free water. Add this mixture slowly, with stirring, to a cooled solution 7.A Xesslerization procedure of 160 g NaOH in 500 ml ammonia-free water and 7.A.I Determine a reagent blank for each new dilute to 1 liter. Allow the reagent to stand at batch of H 2S04 by taking 300 ml ammonia-free least overnight and filter through a fritted-glass water through the entire procedure: crucible. mg reagent blank = mg N per 10 ml H2SO4 5.7 Sodium hydroxide solution, 10A': Dis­ ml aliquot solve 400 g NaOH in ammonia-free water and X dilute to 1 liter. ml distillate 20 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

7.A.2 Determine the mg N in the aliquot bonaceous organic pollution from sewage or in­ from a plot of absorbances of standards. dustrial wastes. 7.A.3 Normal, unpolluted river waters generally have Organic nitrogen as N, in mg/1 a COD from about 10 to 30 mg/1; mildly polluted river waters, 25 to 50 mg/1; and domestic sewage 1,000 500 about 250 mg/1 (It. C. Kroner, written commun., ml sample ml aliquot 1970). X[(mg N in aliquot) —(mg reagent blank)]. Tolerances for oxygen-demand values in feed 7.B Titmtion procedure water for low- and high-pressure boilers are 15 and 7.B.I Ammonia nitrogen as N, in mg/1 3 mg/1, respectively. Wash water containing more Va X X X 14,000 than 8 mg/1 has been reported to impart a bad odor to textiles; concentrations for water used in V. beverages and brewing range from 0.5 to 5.0 where mg/1 (California State Water Quality Control Board, 1963). Fn = ml standard H 2SOi used in titration of sample minus ml used to titrate blank, X« = normality of standard H2$O4, and Dichromate oxidation method

tr* = ml of original sample used for distillation. 1. Summary of method 7.C Ammonia nitrogen as NH l+l, in mg/l = Organic and other oxidizable material is oxi­ mg/1 as NX 1.288. dized by refiuxing with standard acid-dichromate Ammonia nitrogen as free NH;J, in mg/l = solution in the presence of silver sulfate catalyst. mg/1 as NX 1.216. The excess dichromate is titrated with standard ferrous ammonium sulfate, using orthophenan- 8. Report throline ferrous complex as indicator (American Report organic nitrogen concentrations as Society for Testing and Materials, 1968). follows; Less than LO mg/1, two decimals; 1 mg/1 and above, two significant figures. 2. Application This method can be used for analysis of natural 9. Precision waters and industrial wastes containing less than No precision data are available. 2,000 mg/I chloride ion and more than 50 mg/1 chemical oxygen demand (COD). Samples con­ Reference taining less than this amount should be analyzed Kolthoff, I. M., and Sandell, 10. B., 1952, Textbook of as directed in step 6.9. COD values for waters quantitative inorganic analysis |3d ed.j: New York, containing more than 2,000 mg/1 of chloride ion Macmillan Co., 759 p. should be corrected as indicated in step 6.10.

3. Interferences Oxygen demand, chemical (COD) Reducing substances such as ferrous iron and The oxygen-demand determination is a measure chlorides interfere since they are oxidized. of the readily oxidizable material in the water, and Chlorides constitute by far the largest and most it furnishes an approximation of the minimum common interference, being quantitatively oxi­ amount of organic and reducing material present. dized by dichromate in acid solution. One mg/1 In reality, the term "chemical oxygen demand" C 1!-1 is equivalent to 0.226 mg/1 COD. To elimi­ is defined by the method used for its determina­ nate chloride interference, mercuric sulfate is tion. In the method given below it is defined as added to the sample to form a soluble mercuric the amount of oxygen used by the sample when chloride complex. refluxed 2 hr with an excess of acid-potassium Care should be taken to prevent heating of the dichromate solution. The determined value may samph1 during addition of reagents to minimize correlate with natural-water color or with car­ loss of volatile constituents. ANALYSIS OF ORGANIC SUBSTANCES IN WATER 21

4. Apparatus 6.6 Allow flask to cool, and wash down con­ 4.1 Reflujr apparatus consisting of a 500-ml denser with 25 ml water. erlenmeyer flask and water-cooled condenser, 6.7 Dilute to 300 ml with demineralized with ground-glass joints and made of heat- water, cool to room temperature, and titrate the resistant glass. excess dichromate with 0.2500A' ferrous am­ 4.2 Hot plate or heating mantle. monium sulfate solution, using 8-10 drops Ferroin indicator solution. The end point is a sharp 5. Reagents change from blue green to reddish brown. 5.1 Ferrous ammonium sulfate standard solu­ 6.8 A demineralized-water blank is carried tion, 0.2500.Y: Dissolve 98.0 g FeSO^(KH4) 2S04 - through all steps of the procedure with each group 6H20 in demineralized water. Add 20 ml concen­ of samples. trated H2S04, cool, and dilute to 1 liter. To 6.9 Samples containing less than 50.0 mg/1 standardize, dilute 25.0 ml standard 0.2500.Y COD should be reanalyzed, using 0.025.Y solu­ K2Cr2O7 solution to 250 ml. Add 20 ml concen­ tions of potassium dichromate and ferrous trated H2S04 and cool. Titrate with the ferrous ammonium sulfate. A sample size should be ammonium sulfate solution, using 8-10 drops selected so that no more than half the dichromate Ferroin indicator. The solution must be stand­ is reduced. A further increase in sensitivity may ardized daily, or before use. be obtained by evaporating a larger sample to 5.2 Mercuric sulfate, powdered HgS04 . 150 ml in the presence of all reagents. A blank 5.3 Orthophenanthroline ferrous sulfate (Fer­ should be treated in a similar manner. roin) indicator solution: Dissolve 1.48 g 1,10- 6.10 To obtain more accurate COD values (ortho)-phenanthroline monohydrate and 0.70 g for samples containing more than 2,000 mg/1 of FeSO,-7H20 in 100 ml of water. The prepared chloride ion, the following procedure may be used indicator is available commercially. (Burns and Marshall, 1965). A series of chloride 5.4 Potassium dichromate standard solution, solutions are analyzed by the procedure indicated 0.2500A': Dissolve 12.259 g K2Cr,O7 primary above, except that 10 mg of HgS04 is added to standard, dried for 2 hr at 100°C, in demineralized each solution for each milligram of chloride ion water and dilute to 1,000 ml. present instead of a constant l~g quantity. The 5.5 Silver sulfate, powder, chloride concentrations should range from 2,000 5.6 Sulfuric acid, concentrated (sp gr 1.84). mg/1 to 20,000 mg/1, with the concentration interval not exceeding 4,000 mg/1. Plot the COD 6. Procedure values obtained versus milligrams per liter 6.1 Pipet 50,0 ml of sample or a smaller chloride. From this curve, COD values may be aliquot diluted to 50.0 ml into the reflux flask obtained for any desired chloride concentration. and add slowly, over a period of 2-3 min, 1 g This value is subtracted as a correction factor to HgSO4 ; allow to stand 5 min, swirling frequently. obtain the COD value of a sample. 6.2 Add 1 g Ag.:S04 and a few glass beads

that have been ignited at 600°C for 1 hr. 7. Calculations 6.3 Cool in ice water and add 75 ml concen­ trated H2SOj, slowly enough, with mixing, to Calculate the COD in each sample as follows: present appreciable solution heating with the 7.1 For samples not requiring chloride cor­ consequent loss of volatile constituents. rection: 6.4 Add 25.0 ml 0.2500A7 K2Cr2O7 solution and mix thoroughly by swirling. (a-6)cX8,000 COD, in mg/1 = 6.5 Attach flask to condenser, start water ml sample flow, and reflux for 2 hr. NOTE.—If contents are not well mixed, super­ 7.2 For samples requiring chloride correction : heating may result, and the contents of the flask may be blown out of the open end of the con­ denser. sample 22 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

\vhere: Mg/1 is harmful to many fish (Federal Water Pollution Control Administration, 1968). COD = chemical oxygen demand from di-

chromate, 1. Summary of method a = ml ferrous ammonium sulfate for The steam-distillable phenols react with 4-ami- blank, noantipyrine at pH lO.OrtO.2 in the presence of b = ml ferrous ammonium sulfate for potassium ferricyanide to form a colored anti- sample, pyrine dye. This dye is extracted from aqueous c = normality ferrous ammonium sulfate, solution with chloroform, and the absorbance is c/ = chloride correction value from graph measured at a wavelength of 460 nm. The con­ of chloride concentration versus COD, centration of phenolic compounds is expressed as and micrograms per liter of phenol (C6H5OH). This 1.20 = empirical compensation factor. method is similar in principle to, but different in 8. Report detail from, ASTM Method D 1783-62 (1969, p. 515-521). Report COD as follows: Less than 10 mg/1, whole numbers; 10 mg/1 and above, two signifi­ 2. Application cant figures. This method may be used to analyze waters

9. Precision containing from 0.0 to 1,000 Mg/1 of phenolic material. No precision data are available. The general precision of COD determinations has been re­ 3. Interferences viewed by the Analytical Reference Service of the Other phenolic compounds, as determined by U.S. Public Health Service (1965). this method, may produce less color than an equivalent amount of phenol itself. The introduc­ References tion of substituent groups to the benzene nucleus American Society for Testing and Materials, 1968, Water; of phenol lowers the sensitivity of the particular atmospheric analysis, pt. 23 of 1968 Book of stand­ compound to color formation. The composition of ards: Philadelphia, Am. Soc. Testing Materials, p. 244. Burns, E. R., and Marshall, C., 1965, Correction for various phenolic compounds which may be present chloride interference in the chemical oxygen demand in a given water sample is unpredictable. Phenol test: Water Pollution Control Federation Jour., v. itself, therefore, has been selected as the standard 37, p. 1716. for reference. Using this basis, the amount of California State Water Quality Control Board, 1963, phenol determined represents the minimum con­ Water quality criteria: Pub. 3-A, p. 233. U.S. Public Health Service, 1965, Water oxygen demand centration of phonolic compounds present in the no. 2: Public Health Service Study 21, Pub. sample. 999-WP-26. Certain bacteria, oxidizing and reducing sub­ stances, and highly alkaline waste waters may interfere with this method. Information for re­ Phenolic material moval of major interference may be found in Phenolic material in water resources is usually ASTM Method D 1783-62 (1969, p. 515-521). a result of pollution from oil refineries, coke plants, and from chemical manufacture. Mixed 4. Apparatus phenolic wastes at 0.02-0.15 mg/1 levels in water 4.1 Distillation apparatus, nil glass, consisting cause tainting of fish flesh. Low concentrations of of a 1-liter Pyrex distilling apparatus and a phenol impart a very disagreeable taste to drink­ water-cooled condenser. ing water. Reported thresholds of taste and odor 4.2 Funnels, Buchner type with fritted-glass range from 0.01 to 0.1 Mg/1- Chlorination produces disk (15-ml Corning 36060, or equivalent). an even more disagreeable taste and odor by re­ 4.3 Photometer, spectrophotometer or filter acting with the phenols to form chlorophenols. photometer suitable for use at a wavelength Concentrations of phonolic material up to 1,000 setting to 460 nm, and accommodating cells Mg/1 are not believed toxic to animals, but 5.0 having light paths of 1.0 and 10 cm. ANALYSIS OF ORGANIC SUBSTANCES JN WATER 23

4.4 pH meter, glass electrode. if necessary. (Add 5.0 ml of copper sulfate solu­ 4.5 Separator^ funnels, 1-liter capacity, tion if for any reason it was not added at sam­ Squibb type, with unlubrieated glass or teflon pling.) Transfer the solution to the distillation stopcock. apparatus, add boiling stones, and set for dis­ tillation. Collect 450 ml of distillate and stop. Add 5. Reagents 50 ml of distilled water to the residue and proceed All reagents must be prepared with phenol-free with the distillation until 500 ml of distillate is distilled water. De-ionized water is usually not collected.. satisfactory. 6.2 Prepare a 500-ml distilled-water blank. 5.1 Aminoantipyriue solution; Dissolve 2.0 g Also prepare 500-ml standards containing 5, 10, 4-aminoantipyrine in distilled water and dilute to 20, 30, 40, and 50 Mg of phenol, using the standard 100 ml. This solution is not stable for storage and phenol solution. should be prepared each day of use. 6.3 Treat the sample, blank, and standards 5.2 Ammonium chloride solution: Dissolve as follows: Add 10,00 ml ammonium chloride solu­ 20 g of reagent-grade ammonium chloride in tion and adjust the pH to 10.0±0.2 with concen­ water and dilute to 1 liter. trated ammonium hydroxide. Transfer the solu­ 5.3 Ammonium hydroxide, concentrated (sp tion to a 1-liter separatory funnel, add 3.00 ml of gr 0.90), ACS reagent grade. aminoantipyrine solution, mix, add 3.00 ml of 5.4 Chloroform, spectrophotometric grade. potassium ferricyanide solution, and again mix. 5.5 Copper sulfate solution: Dissolve 100 g of Allow the color to develop for 3 min, and a clear CuS04 '5H20 in water and dilute to 1 liter. to light-yellow solution should result. 5.6 Phenol standard solution, 1.00 ml = 1.00 6.4 Add 25.0 ml chloroform for 1- and 5-cm mg phenol: Dissolve 1.00 g analytical reagent cells and 50.0 ml for 10-cm cells. Shake the phenol in 1,000 ml freshly boiled and cooled dis­ separatory funnel vigorously for 1 min. Allow the tilled water. Solution may be used for up to 1 layers to separate and repeat the shaking. month. 6.5 After the layers have separated, draw off 5.7 Phosphoric acid solution: Dilute 10 ml the lower chloroform layer and filter through a 85-percent H3P04 to 100 ml with distilled water. 5-g layer of sodium sulfate, using the sintered 5.8 Potassium ferricyanide solution: Dissolve glass funnel, directly into the appropriate ab­ 8.0 g K3Fe(CN) 6 in water, dilute to 100 ml, and sorption cell. Avoid working in a draft to reduce filter. This solution is not stable and should be evaporation of the solvent. prepared each day of use. 6.6 Measure the absorbance of the sample 5.9 Sodium sulfate, anhydrous, ACS reagent and standards against the blank at a wavelength grade, granular. of 460 nm. Prepare a calibration curve plotting absorbance against micrograms of phenol. 6. Procedure Samples should be collected according to the 7. Calculations recommended practice for organic samples. Sam­ Phenol Oig/l)=-Xl,000, ples must be preserved with 10 ml of copper D sulfate and 2 ml of phosphoric acid solution. A where sealed glass ampoule of the preservative, with instructions, should accompany the sample con­ A = iug phenol measured, and tainer. A 1-liter sample should be collected for B = ml of the original sample used. each analysis. Samples should be protected from 8. Report light and analyzed as soon as possible. The Report phenolic material concentration for less analyst is referred to "Standard Methods/' 12th than 100 jug/1 to the nearest whole microgram, and edition (Am. Public Health Assoc., 1965), for greater than 100 pig/1 to two significant figures. the analysis of very alkaline or highly polluted water. 9. Precision 6.1 Measure 500 nil of the sample into a Precision at 5 mg/1 for phenol only is ±5 per­ beaker. Determine the pH and adjust below 4.0 cent but is variable for other phenolic materials. 24 TECHNIQUES OF WATER.RESOURCES INVESTIGATIONS

Because of interferences and because it is only a The technique of gas chromatography is most relative measure, the result should be considered useful for qualitative and quantitative analysis of as a minimum value. multicomponent mixtures. Small volumes of ex­ tract, as little as 1 jj, are injected into the gas References chromatograph, where the components are sepa­ rated and detected. The separation of vaporized American Public Health Association, 1965, Standard methods for the examination of water and waste- material takes place in the chromatographic water [12th ed.]: New York, Am. Public Health column as it is carried along by a fiow of inert gas. Assoc., Inc., 769 p. Actual separations occur as the component vapors American Society for Testing and Materials, 1969, Water; partition between the vapor phase and a non­ atmospheric analysis, pt. 23 of J969 Book of stand­ volatile stationary liquid incorporated in the ards: Philadelphia, 1032 p. [U.S.] Federal Water Pollution Control Administration, column. Each component, according to its physi­ 1968, Report of the committee on water-quality cal and chemical properties, enters and leaves the criteria: Washington, U.S. Govt. Printing Office, stationary liquid at a unique rate. Because this 234 p. partitioning is occurring between a moving vapor and a stationary liquid, components injected at one end of a column emerge from the other end Pesticides—Gas chromatographic at different times. analysis Several different devices are available for de­ tecting and measuring pesticides as they emerge The term "pesticide" encompasses a broad from the column. The electron-capture detector class of toxicants used to control insects, mites, is highly sensitive, responding to as little as 0.1 pg fungi, weeds, aquatic plants, and undesirable (picogram) of lindano, and is used extensively for animals. More specific designations include such detecting the presence of pesticides in water. terms as insecticides, miticides, fungicides, herbi­ Other somewhat less sensitive detectors, such as cides, and rodenticides. microcoulometric and flame photometric, are used Synthetic organic pesticides have introduced a in pesticide analysis because they respond only to far-reaching technological advance in the control specific elements incorporated -in the pesticide of pests. Although the compound DDT (dichloro- molecules, and thus aid identification. There arc diphenyltrichloroethane) was first synthesized in numerous books and papers on the analysis of 1874, its insecticidal properties were not dis­ pesticides by gas chromatography. One such covered until 60 years later. Since the introduc­ treatment is "Pesticide Residue Analysis Hand­ tion of synthetic chemical pesticides in the United book" (Bonelli, 1965). The book, "A Programmed States, annual production has reached 1 billion Introduction to Gas-liquid Chromatography" pounds. There are almost 60,000 pesticides (Pattison, 1969), is a good source for beginning formulations registered, and each contains at gas chromatographers. least one of approximately 800 different pesticide compounds (Simmons, 1969). 2. Application With the increased concern by noted world Organic insecticides and herbicides which are ecologists over the effects of toxic pesticides on volatile or can be made volatile for gas chromato- the environment, efforts are being made to substi­ graphic purposes may be analyzed by this tech­ tute more specific, fast-acting, and easily de- nique. gradable compounds for the chlorinated hydro­ carbon pesticides. These pesticides were developed 3. Interferences for general application and have proved to be The most .seriously interfering substances for very resistant to environmental degradation. chlorinated pesticide analysis are halogcnated organic compounds, such as "polychlorinated 1. Summary of method biphenyls" from industrial waste. Any compound Prepared extracts of water or sediment are or compounds having chemical properties similar analyzed for pesticides by gas chromatography. to the pesticides of interest may cause inter- ANALYSIS OF ORGANIC SUBSTANCES IN WATER 25 ference. The electron-capture detector is ex­ silicone oil (viscosity 12,500 centistokes) and 0.5 tremely sensitive but much less specific than the percent by weight Carbowax 20 M; and (2) with other detectors mentioned above and, as a result, 5 percent by weight QF-1 fluorinated silicone oil is much less reliable when interfering substances (also designated FS-1265) and 0.5 percent by are present. Special precautions are necessary to weight Carbowax 20 M. The support should be avoid contamination during sample handling and coated with the liquid phase by the "frontal to remove extraneous material from the sample analysis" technique (Smith, I960). The packing extract. materials are loaded in the glass columns using vibration and a vacuum to settle. The packing is 4. Apparatus held in place by small plugs of "silanized" glass 4.1 Electron-rapture yds cttromatoyraph: A gas wool. chromatograph having an electron-capture de­ The columns are installed in the gas chromato­ tector which, for an injection of 0.1 ng (nanogram) graph and are conditioned as follows: (1) Purge of aldrin, gives 100 mv-sec (millivolt-seconds) of the columns for 30 min with inert carrier gas. (2) response is adequate. A Varian-Aerograph Model Turn off carrier gas flow and heat the columns to 600-D, or equivalent, may be used. (A radioiso- 250°C for 2 hr. (3) Reduce the temperature to tope-byproduct-matcriiil license is required for 210°C and allow temperature to equilibrate for electron-capture detectors employing H3 or Ni63 30 min. (4) Turn on carrier gas flow to about 30 sources.) ml/min (milliliters per minute) and continue heat­ 4.2 Flame-photometric (/as chromatograph: A ing the column at 210°C for 12 hr. The column gas chromatograph equipped with a Melpar should not be connected to the detector during flame-photometric detector having filters for the column conditioning. specific detection of phosphorus or sulfur. Such After conditioning, the columns are ready for an instrument is the Micro-Tek Model MT-220 use. Performance and retention-time character­ flame-photometric gas chromatograph. A pro­ istics must be determined for each column by vision for venting solvent effluent between the use of standards. Retention data in tables 1, 2, column and the detector should be specified. 4.3 Microcoulometric-titratinij (/as chromato- yraph: A gas chromatograph connected to a Table 1.—Retention values for chlorinated hydrocarbon Dohrmann microcoulometer detection system. insecticides, relative to aldrin The system employs a Model S-200 sample-com­ Columns.-__.-__------_------1.8-m long X 1.8-mm ID bustion unit, a Model C-200 coulometer-amplifier, Carrier gas flow..„___..______.__ 30 ml/min nitrogen. Column temperature.. ._._._...... __.. 185°C. and a choice of titration cells, namely: for halides, Electron-capture detector. the T-300-S cell; for sulfur, the T-300-P cell; and for nitrogen, the T-400-H cell. This unit may Relative retention time be used with the Micro-Tek Model MT-200 gas Insecticides 5 percent 5 percent chromatograph, or equivalent. DC-200 QF-1 4.4 (ras chroinaloyraplric columns: The gas chromatographic columns are fabricated from Lindane. - _ _ . _ , _ _.. 0.46 0.80 1.5-m (meter) lengths of Pyrex glass tubing. For 79 .87 electron capture, 1.8-mm ID (inside diameter) Aldrin 1 ...-...... 1 00 1.00 1.18 1.34 tubing is used preferably, whereas for other Heptachlor epoxide. .__._. 1 . 28 2.06 modes of detection either 1.8-mm ID or 4-mm Dieldrin ____.__,__-___-. 1.92 3.31 ID glass tubing may be used. The smaller bore p,p'-DDE. ----.------. 1.98 2.30 columns accept injection volumes up to 10 M! and 2.15 3.93 the larger bore columns will accept volumes up p,p'-DDD_.-. ... ------2.53 4.24 o,p'-DDT._ ...... 2.69 3.01 to 80 /xl. p,p'-DDT... .------_. 3.41 4.69 Gas Chrom Q support, 60/80 mesh, is used for Methoxychlor, .._..--.-. 5.36 7.96 the preparation of two different column packings as follows: (1) With 5 percent by weight DC-200 i Aldrin retention times: DC-200, 4.18 min; QF-1, 2.00 min. 26 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

Table 2.—Retention values for phosphorothioate insecticides, purified grade or better, furnished in size 1A relative to parathion high-pressure cylinders. (CAUTION; Never use Columns...._.._..-..__..--..__..__.. 1.8-m long X 4.0-mm ID oxygen regulators for other gases.) Carrier gas flow. ______75 ml/min nitrogen. 4.8 Microbalance: A Cahn Gram Electro- Column temperature....___..,..._____. 185°C. Flame-photometric detector. balance, or equivalent. 4.9 Volumetric glassware: Class A volumetric Relative retention time flasks in 5, 10, and 25 ml sizes. The stoppers

Insecticides 5 percent 5 percent should fit well because volatile organic solvents DC-200 QF-1 are used for dilutions. Volumetric ware such as supplied by Kontes Glass Co., or equivalent, is Dioxathion ,______-__-__ 0.50 0.40 acceptable. Diazinon ______.55 .22 4.10 Integrating equipment: A compensating VC-13----.---- -_...-._. .71 .34 polar planimeter readable to the nearest 0,01 Methyl Parathion ______.72 1.04 Malathion______.93 .79 square inch is acceptable. Other instruments or Parathion 1 __..______1.00 1.00 methods of integration demonstrating greater Methyl Trithion ______2.2 2.7 accuracy may be used. Ethion__.______2.6 1.84 4.11 Recorder: A 1-mv (millivolt) full-scale Carbophenothion______2.9 2.48 response, 1-sec (second) pen speed, strip-chart recorder. Such a recorder having a fixed or select­ 1 Parathion retention times: DC-200, 3.82 min; QF-1, 4.55 mm. able chart speed of one-half inch per minute is acceptable. and 3 may be used as a guide for evaluating the columns. Column efficiency is measured by em­ 5. Reagents ploying the following equation: Solvents and reagents are specified for the particular isolation technique used. Recom­ mendations of the manufacturer should be followed for special reagents to be used with a particular gas chromatographic system. where 5.1 Benzene, pesticide-analysis quality: Na- n — number of theoretical plates, nograde, distilled in glass, or equivalent. Benzene fr = uncorrected retention time of peak, and is usually the solvent of choice for preparation of A/ = peak retention width (length of baseline concentrated standard solution because it is cut by the two tangents of the peak at relatively nonvolatile and the pesticide solution the half-height points). can be stored for long periods in a safety re­ frigerator. Using a ;>,//-DDT standard to test the column efficiency, a value of no less than 1,500 theoretical plates for a 1.8-m column is considered acceptable Table 3.—Retention values for methyl esters of chlorinated phenoxy acid herbicides, relative to 2,4-D for pesticide analysis. 4.5 Microliter capillary pipets: Volumetric Columns_-______.---.______-_ 1.8-m Ions X 1.8-mrn ID Carrier gas fiow______30 ml/min nitrogen. micropipets in 1, 5, 10, and 25 /_! sizes; the dis­ Column temperature. ______145°C. posable types are satisfactory. Klcctron-capturc detector. 4.6 Idicroliter syringes: Three microsyringes having capacities of 10, 50, and 100 p\, respec­ Relative retention time tively, are used. The syringe needle should be Herbicides 5 percent 5 percent DC-200 QF-1 about 2 inches long and have a point shaped to prevent punching out a core when penetrating the injection septum. 2,4-D l ...._...... 1. 00 1.00 Silvex-______.___ 1.42 1.22 4.7 Compressed gases: Use only the gases 2,4,5-T... _-__--_--. 1.91 1.80 recommended by the vendor for the particular instrument system being used. Also, select pre- 1 2,4-D retention times: DC-200, 3.U2 mm; QF-1, 3.05 min. ANALYSIS OF ORGANIC SUBSTANCES IN WATER 27

5.2 Pesticide standards: Reference or analyti­ the standardization should be written directly on cal-grade pesticide chemicals may be obtained the recorder chart. Calibration should be per­ from gas chromatography specialty suppliers and formed on both the DC-200 and the QF-1 often also by written request from the manu­ columns. facturer. It is desirable to obtain a particular 6.2 Sample analysis pesticide from at least two different suppliers. The sample extracts are analyzed in the same The pesticide standards should be refrigerated manner as the standards and under the same during prolonged storage, and appropriate hazard operating conditions. warnings should be posted on the refrigerator. 6.2.1 The first analysis is performed by electron-capture gas chromatography using the 6. Procedure DC-200 column. Concentration or dilution of the 6.1 Standardization extract may be required to allow a 5.0-Ml injec­ Each gas chromatographic system must be tion. Proceed with the analysis by injecting 5.0 calibrated to reference standards at the operating /il of the sample into the chromatograph, record­ conditions to be used for analysis. ing the extract volume and the volume injected. 6.1.1 Picogram standards: Weigh 1.00 mg of Do not make any subsequent injections until the pesticide on the microbalance and transfer into a last compound has eluted and the baseline has 10.00-ml volumetric flask., Dilute to volume with returned to normal. benzene and mix thoroughly. Prepare a series of 6.2.2 Run a calibration-retention-time stand­ required picogram standards from this solution. ard and a reagent blank as an analysis check. (Example: Take 1.00 jul of the above pesticide Should a pesticide be detected in the sample, a solution and dilute to 10.00 ml with the solvent standard containing the same pesticide at nearly to be used in the analysis. The concentration of the same concentration is also analyzed just after pesticide in the resulting solution is 10X10~12 the sample. g/Vl (grams per microliter), or 10 pg/^l (pico- 6.2.3 Pesticides detected in concentrations grams per microliter).) ranging from 0.01 jug/1 to 1.0 Mg/1 for water 6.1.2 Nanogram standards: Weigh 5.00 mg of samples, or from 0.10 Mg/kg (microgram per reference pesticide into a 5.00-ml volumetric kilogram) to 1.0 Mg/kg for sediment, must be flask and dilute to volume with benzene. Make a analyzed a second time by electron-capture gas series of appropriate nanogram standards from chromatography, on the QF-1 column, for con­ this solution. (Example: Take 10.0 /*! and dilute firmation. to 10.00 ml with the solvent to be used in the 6.2.4 The presence of pesticides at concen­ analysis. The concentration of pesticide in the trations greater than 1.0 Mg/1 in water or 1.0 resulting solution is 1.0X10"9 g//il, or 1.0 ng/Vl jug/kg in sediment samples must be confirmed by (nanogram per microliter).) microcoulometric or flame-photometric gas chro­ 6.1.3 Calibration: The picogram standards matography on both the DC-200 and QF-1 are used for electron-capture gas chromatography, columns. This requirement is not intended to and the nanogram standards are used for flame- restrict the use of specific detectors but rather to photometric and microcoulometric gas chroma­ indicate concentrations above which they must tography. A 5.0-jul volume of each of the ap­ be used. Specific detection should always be used propriate standard solutions is injected into the whenever practical. Volumes of extract up to 10 gas chromatograph. The concentration of pesticide Ml for the smaller diameter columns and up to 80 in the series of standard solutions should be such lA for the larger diameter columns may be in­ to calibrate either the full range of linear detector jected. In this instance, a check standard at nearly response or the range of anticipated pesticide the same concentration should also be run. concentration in the sample, whichever is less. The injection should be made so that the solution 7. Calculations enters the injection port in a single volume and Each gas chromatographic system must be in a reproducible manner. The volume injected calibrated with standards. The response of the should be measured by reading the syringe before gas chromatographic detector is usually the dis­ and after injection. All information pertinent to play of an analog signal on a strip-chart recorder. 28 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

The signal is recorded as a differential curve or by use of a planimeter or by any method of equal peak. The area inscribed beneath the peak is or greater accuracy is acceptable. If a planimeter proportional to the amount of material passing is used, the average of at least two measurements through the gas chromatographic detector. The is taken as the peak area. time elapsed from the introduction of the sample Interpretation of the chromatogram is very to the differential curve maximum is designated important to the precision of area measurement. as the retention time for a particular component, Reliable interpretation comes with experience and The retention time for a compound on a specified much can be gained by careful study of the elu- column is nearly unique and is used for quali­ tion patterns and peak shapes of individual mixed tative analysis. Also, the retention time relative standards. In general, peaks may appear in four to another selected compound is often used be­ different ways, which are: (1) A single peak, (2) cause this expression reduces variation usually two or more discrete peaks not completely sepa­ found in day to day operation. The response of rated, (3) a small peak or shoulder on the leading the chromatograph must be standardized at or trailing edge of a relatively large peak, and optimum conditions and enough determinations (4) two or more peaks perfectly overlapping one made so that the data may be treated by the another. The presentation of a single peak is method of least squares. During analysis, the ideal and allows precise area measurement. Peaks standard curve must be checked by running at not completely resolved are graphically separated least two standards at different concentrations so by drawing a line from the valley point between corrections can be made for day to day fluctua­ two adjacent peaks down to the baseline. It is tions. >very difficult to isolate a shoulder from the1 larger 7.1 Qualitative analysis peak in a reproducible manner. Also the retention Directly comparing the retention times of a time is biased toward the larger component. In sample component and a reference standard on this situation a line is drawn to conform with the both DC-200 and QF-1 columns is the method shape of the major peak. Although the area under used for qualitative identification. Additionally, the larger peak is usually quite reliable, that of specific detection is employed to further confirm the shoulder is not. Other steps should be taken, the presence of a particular component at levels such as gas chromatography using a different greater than 1.0 /*g/I for water samples and 1.0 column or techniques of column or thin-layer Mg/kg for sediment. Relative retention time, the chromatogrnphy, to isolate the shoulder com­ ratio of the retention time of an unknown to pound for quantitative determination. The same that of a selected standard, may be used to consideration must bo given to overlapping determine which reference standard to choose peaks. Components elating at nearly the same for comparison. The pesticides selected for this time to form a single peak are easily misinter­ purpose are: Aldrin for the chlorinated hydro­ preted. Correlation of retention times arid peak carbon insecticides, parathion for the phosphoro- areas on both the DC-200 and QF 1 columns is thioate insecticides, and the methyl ester of extremely important in this instance. To obtain 2,4-D for the chlorinated phenoxy acid herbi­ reliable qualitative and quantitative analysis, cides. The following equation is used in qualitative other isolation techniques may have* to be em­ identification: ployed whenever this occurs (Federal Water RTu Pollution Control Administration, 1969). RRT (relative retention time) = -77^.- 7.2.1 Standard curve rti r Using log-log graph paper, plot area of re­ whore sponse, in square inches fin2), against nnnogrnms of pesticide injected. If six or more values fall in = retention time of the unknown com­ the linear response region of the detector, the pound, and equation of the line may be found by the method RTr = rctcntion time of the reference com­ of least squares, as follows: pound. 7.2 Quantitative, analysis Measurement of gas chromatograrn peak ureas € ANALYSIS OF ORGANIC SUBSTANCES IN WATER 29 and substituting the weight of sample in kilograms for the sample volume (Vs) with the resulting concentration expressed as //g/kg. where 8. Report r = injection amounts (ng), Pesticides found in water samples are reported # = area response values (in2), as follows: At concentrations of less than 1.0 b — y intercept, /zg/1, two decimals and report less than 0.005 w — slope, and jitg/1 as 0,00 Mg/1; at concentrations of 1.0 Mg/1 n = number of points selected. and greater, two significant figures. Pesticides in For the equation of the straight line, sediment and soil samples are reported as follows: Less than 1.0 Mg/kg to one decimal; 1.0 jitg/kg and above, two significant figures. The identities the value for 6, the y intercept, is an indication of pesticides found in concentrations greater than of whether any experimental bias exists. It is 0.01 Mg/1 in water or 0.1 jug/kg in sediment must usually small enough to be insignificant so that be confirmed by two-column gas chromatography. the equation of the standard curve may be ex­ For concentrations greater than 1.0 jug/1 in water pressed as: and 10 Mg/kg in sediment, specific detection must y = wjr. be employed. Identities of compounds in concen­ The two or more daily response check standards trations greater than 10 Mg/1 in water and 100 are used to correct the slope of the standard Mg/kg in sediment must be confirmed by mass curve, as follows: spectromotry.

9. Precision Ac Precision of the gas chromatographic technique where is variable for multicomponent analysis. The re­ C = correction factor, sponse of one component may be considerably .4c = area of check standard obtained from greater or less than that of another. Peaks of the standard curve, and compounds having longer retention times are af­ As = area obtained from chromatogram of the fected more by instrumental noise and drift. check standard. Under ideal conditions repetitive analysis of a single component may be determined to a pre­ The slope of the standard curve is corrected by cision of ±3 percent. multiplying it by the correction factor. 7.2.2 Calculations for samples References The concentration of pesticides in water samples may be determined using the following Bonelli, E. J., 1965, Pesticide residue analysis handbook: equation: Walnut Creek, Calif., Varian Aerograph, 124 p. Pattison, J. B., 1969, A programmed introduction to gas- Concentration of pesticide (/ig/1) liquid chromatography: Philadelphia, Sadtler Re­ search Labs, Inc., 303 p. 1 Vert 1 _ A v ___ V ____ V __ Simmons, S. W., 1969, PHS pesticide program: Natl. Agr. — ^ Cm/^ ^ trVinj ' • ^ VsTT- J Chem. Assoc., News Pesticide Rev., v. 27, no. 5, p. 10-12. where Smith, E. D., 1960, Preparation of gas chromatographic A =area of component (in2), packings by frontal analysis: Anal. Chemistry, v. 32, Cm ~ corrected slope (in2/ng), p. 1049. [U.S.] Federal Water Pollution Control Administration, Vejrt = volume of extract (ml), 1968, Report of the committee on water quality Vinj= volume injected (ml), and criteria: Washington, U.S. Govt. Printing Office, Fs = volume of water sample (liters). 234 p. ———— 1969, FWPCA method for chlorinated hydrocarbon This equation may be used to calculate the con­ pesticides in water and wastewater: Cincinnati, centration of pesticides in sediment or soil by Federal Water Pollution Control Administration, 29 p. 30 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

Insecticides in water 4. Apparatus See step 4, "Gas Chromatographic Analysis of Gas chromatographic method Pesticides." 4.1 Concentrating apparatus: A Kudernii-

1. Summary of method Danish concentrator, 250-ml capacity with a 1-ball Snyder column, is used for .the initial con­ The insecticides are extracted directly from the centration step. Final concentration is performed water sample with n-hexane. After drying and in the receiver using .a 1-ball Snyder micro- removing the bulk of the solvent, the insecticides column. A' calibrated 4.00-ml receiver tube is are isolated from extraneous material by micro- used with the concentration apparatus. column adsorption chromatography. The insecti­ 4.2 Cleanup microcolumns: Disposable Pas­ cides are then analyzed by gas chromatography. teur pipets, 14-cm long and 5-mm ID, are used This method is a modification and extension of for the chromatographic cleanup columns. The the procedures developed by Lamar, Goerlitz, pipets are washed in warm detergent solution, and Law (1965, 1966). For the analysis of in­ thoroughly rinsed with dilute hydrochloric acid secticides in waters that are grossly polluted by and organic-free distilled water, then heated to organic compounds other than pesticides, the 300°C overnight to remove any traces of organic analyst is referred to the high-capacity cleanup matter. A column is prepared by plugging the procedure detailed in Federal Water Pollution pipet with a small amount of specially cleaned Control Administration "Method for Chlorinated glass wool, adding enough deactivated alumina Hydrocarbon Pesticides in Water and Waste- through a microfunnel to fill 3 cm of the column, water" (1969). followed by another1 0.5 cm of anhydrous sodium sulfate. 2 Application 4.3 Sandbath, ftuidizcd, Tecam, or equiva­ This method is usable for the analysis of water lent. only. The insecticides and associated chemicals 4.4 Separatory funnels, Squibb form, 1- or (aldrin, 7>,7>',DDD, p,pVDDE, o,?/-DDT, p,pf- 2-liter capacity. No lubricant is used on the DDT, dieldrin, endrin, hcptachlor, hcptachlor stopcocks. epoxide, isodrin, lindane (BHC), and methoxy- chlor) may bo determined to 0.005 jig/1 in 1-liter 5. Reagents water samples. The insecticides carbophonothion, 5.1 Alumina, neutral aluminum oxide, ac­ chlordan, dioxathion, diazinon, ethion, malathion, tivity grade I, Woelm. Weigh 19 g activated methyl parathion, Methyl Trithion, parathion, alumina into a 50-ml glass-stoppered orlenmoyor toxaphene, and VC-13 may be determined when flask and quickly add 1.0 ml distilled water. present to higher levels (method for organophos- Stopper the flask and mix the contents thoroughly phorus pesticides similar to that of Zweig and by tumbling. Allow 2 hr before uso. The do- Dovino, 1969). Also, the chemicals chlordeno, activated alumina may be used for 1 week. hoxachlorobicycloheptadiene, and hoxaehlorocy- 5.2 Benzene, distilled in glass, pesticide- clopentadieno, which are pesticide manufacturing analysis quality. precursors, may bo analyzed by this method. 5.3 ii-Hejrane, distilled in glass, pesticide- analysis quality. 3. Interferences 5.4 Sodium sulfate, anhydrous, granular. Any compound or compounds having chemical Prepare by heating at 300°O overnight and store and physical properties similar to the pesticide of at 130°C. interest may cause interference. The procedure 5.5 Water, distilled, obtained from a high- incorporates a column chromatographic tech­ purity tin-lined still. The feed water is passed nique which eliminates most extraneous material. through an activated carbon filter. Tho distillate1 Special precautions arc necessary to avoid con­ is collected in a tin-silvor-linod storage tank, and tamination during sampling arid analysis. the water is constnntlv irradiated with ultraviolet ANALYSIS OF ORGANIC SUBSTANCES IN WATER 31 light during storage. A gravity delivery system is pour the extract from the top of the separatory used, and no plastic material other than teflon is funnel into a 125-ml erlenmeyer flask containing allowed to contact the distilled water. about 0.5 g anhydrous sodium sulfate. 6.3 Repeat a second and third extraction of 6. Procedure the water sample in the same manner using 25 ml Samples should be collected according to the n-hexane each time, and collect the extracts in recommended practice for the collection of the 125-ml erlenmeyer flask containing the drying samples for organic analysis. A 1-liter bottle of agent. Cover the flask containing the extract with water should be collected for each sample. No foil and set aside for 30 min. preservative is used. Samples should be shipped 6.4 Filter the dried extract through glass promptly. Unless analyzed within a few days, wool into the Kuderna-Danish apparatus. Add a the water should be protected from light and re­ sand-sized boiling stone and remove most of the frigerated. If the sample contains sediment, then hexane by heating on a fluidized sandbath at the sediment must be analyzed separately. Re­ 100°C in a hood. When the ball in the Snyder move the sediment by centrifugation or nitration column just stops bouncing, remove the ap­ through a metal membrane filter. See step 6.1, paratus from the heat and allow to cool. Add "Chlorinated Hydrocarbon Insecticides in Sus­ another small boiling stone, fit the receiver with pended Sediment and Bottom Material." a Snyder microcolumn and reduce the volume to All glassware, except volumetric flasks, should between 0.4 and 0.5 ml on the sandbath. Set aside be washed in the usual manner, rinsed in dilute to cool. When changing columns, sand must be, hydrochloric acid and distilled water, and heat cleared from the glass joint before opening. treated at 300°C overnight. Instead of heat 6.5 Quantitatively transfer the contents of treating, the volumetric ware may be solvent the Kuderna-Danish receiver (0.4-0.5 ml) to the rinsed or steamed to remove organic matter. A top of a deactivated alumina cleanup micro- reagent and glassware blank should accompany column. Use a disposable pipet to transfer. Not each analysis. more than 0,1-0.2 ml hexane should be needed for 6.1 Water samples (800-900 ml) are ex­ washing. Using hexane, elute the extract from the tracted with n-hexane in such a manner that the column to a volume of 8.5 ml in a calibrated water and the container itself are exposed to the 10.00-ml receiver. Add only enough hexane so solvent. Weigh the uncapped bottle of water on that the solvent level enters the column packing a triple-beam balance and pour the sample into just as the 8.5-ml elution level is reached. Change a 1-liter separatory funnel. Allow the bottle to receivers and continue the elution using 1:1 drain for a few minutes, weigh again, and record benzene-hexane solvent. Collect 8.5 ml of eluate the weight of water to three significant figures. in a second receiver. The first fraction of eluate 6.2 Add 25 ml n-hexane to the empty sample should contain all the chlorinated hydrocarbon bottle and gently swirl to wash the sides of the insecticides, and carbophenthion, Methyl Tri- container with the solvent. Pour the contents of thion, and VC-13. The remaining phosphorus- the sample bottle into the separatory funnel containing pesticides are eluted in the benzene- containing the water. Stopper and shake the hexane fraction. Reduce the volume of each separatory funnel vigorously for 1 full min, eluate to 1.00 ml using a Kuderna-Danish micro- venting the pressure often. Allow the contents to apparatus on the sandbath. separate for 10 min and draw off the aqueous NOTE.—The insecticides are separated chromat- layer into the original sample bottle. If the hexane ographically in a predictable order on the micro- layer emulsifies, separate as much water as column, and this may be used to augment gas possible, then shake the contents of the funnel chromatographic analysis. Although alumina is very vigorously so that the liquids contact the the adsorbent of choice for the majority of water entire inside surface of the vessel. (CAUTION: and sediment samples, occasionally a second pass Vent often!) Allow the layers to separate and add through a different column is needed for more approximately 5 ml distilled water to aid the difficult samples. The analyst is referred to the separation, if necessary. Remove the water and work of Law and Goerlitz (1970) for a more 32 TECHNIQUES OK WATER-RESOURCES INVESTIGATIONS

Table 4.—Insecticides in water: recovery of compounds added to surface-water samples

TiisoctiouU1 und amount luMod (/_K/1)

Hrpta- Methyl Sample AMrin p,p'- P.P'- P,P'~ Dk'ldrin Kndrin Hcptn- chlor 1 jmhuir Mala- jmra- Para- No. nnn ni)K DOT rhlnr cpoxido ttuon thion tluon 0.019 0.080 o.oio 0.081 0.019 0.040 0 018 0.021 0 021 0 181 0 082 0.076

1. ______... 82 0 92 5 86.5 95.0 98.8 95 . 1 86 8 9. 0 90.7 92.9 75 . 1 99 0 2...... ______113 89.1 94.3 97.0 104 98.0 98 7 .M.9 101 106 91 6 96.0 3______... 90 . 1 96 0 93 . 5 103 99.6 86 0 95 2 99 2 99 0 120 89.8 110 4______... 92 . 1 95.5 92.1 101 104 81.9 96 3 98.1 107 89 :i 81.0 86 0 5 ... 07.0 95.0 93 2 9G.O 106 81.1 99

. . _ . 92 . 1 92 8 93.4 96.7 102 87.2 95 .I 96 7 103 103 91 98 3 Variance, ___ .... _ . _ 49 . 3 27.9 24.9 17.1 12.9 23.0 21 5 12 5 51 1 63.4 96 5 139 7 Std.dev. ____.__„...... 7.02 5.28 4.99 4 14 3 59 4.80 t 64 3 54 7.15 7 96 9 82 11.8 Moan error. _____ .... -7.9 -7.2 -6.6 -3.3 +2 0 -13 - 4 . 5 -H 3 430 +3.0 -9.0 -1 7 Total error »______..__ 22 18 13 12 9.2 22 14 10 17 19 29 25

1 McFarrpn, E. F., Liskn, 1.. ,!., and Parker, ,1. II . , 1970, ( 'ntonon foi .mining the• arooptubiilifv of analytical imethods Vn :il. Chemistry, v. 42, p. 355-358.

comprehensive treatment of the cleanup pro­ 9. Precision cedure. The results may vary as much as ±15 percent 6.6 Analyze the eluates by gas chromatog- for compounds in the 0.01- to O.lO-jug/l concen­ raphy under conditions optimized for the par­ tration range. Recovery and precision data are ticular gas chromatographic system being used. given in table 4. Run the first analysis on the electron-capture gas chromatograph using the DC-200 column. For References components in concentrations ranging from 0.01 Lamar, W. L., (.oerlit/, I). F., and Law, L. M. f 1065, Mg/1 to 1.0 /ug/1, a second analysis by electron Identification and measurement of chlorinated or­ capture on the QF 1 column is required. Pesti­ ganic pesticide* in water by electroi.-o.ipt.ure gas cides in concentrations greater than 1.0 ^.g/1 chromat ography: U.S. (Jcol. Survey Water-Supply must be analyzed by microcoulometric or flame- Paper 1817-H,*12p. Lamar, W. L., CloerliU, 1). F., and Law, L. M., 1966, photometric gas chromatography on both the 1 )etcM'mination of organic insertioides in water by DC-200 and the QF-1 columns. electron-capture gas ohroinatography, in Organic pesticides in the environ men t: Am. Chem. Soo., 7. Calculations Advances in Chemistry, ser. 60, p, 187-19*). See step 7, "(!us Chron.atographic Analysis/' Law, L. M., and (ioerlit./, I). F., 1970, Mieroc.olumn chrom.itngraphio cleanup for the analysis of pesticides in water: Assoc. Official Anal. Chemists Jour., v. 8. Report o:$, no. 0, p. 1276-1286. The pesticide concentrations in water samples [U.S.] Federal Water Pollution Control Administration, are reported as follows: Less than 1.0 jug/l, two 1969, FWPCA method for chlorinated hydrocarbon decimals and report less than 0.005 /jg/1 as 0.00 pesticides in water and wastewalei: Cincinnati, jug/1; 1.0 jug/I <-nd above, to two significant Federal Water Pollution Control Adm., 29 p. /weig, (iunter, and Devine, .1. M., 1969, Determination of figures. If more than one column or gas chroma! o- organophosphorus pesticides in water: Kesidue Kev., graphic system is used, report the lowest value. v. 26, p. 17-30. ANALYSIS OF ORGANIC SUBSTANCES IN WATER 33

Chlorinated hydrocarbon insecticides 4.1 Centrifuge: A medium-speed centrifuge with a head capable of accepting large-volume in suspended sediment and glass centrifuge bottles and tubes is adequate. bottom material 4.2 Concentrating apparatus: A Kuderna- Danish concentrator, 500-ml capacity with a Gas chromatographtc method 1-ball Snyder column. 4.3 Erlenmeyer flasks, 250-ml and 500-ml, 1. Summary of method having ground-glass stoppers, and having spring The insecticides are extracted from the sedi­ clips for securing the stoppers. ment or soil using acetone and n-hexane. The 4.4 Microfiltration apparatus: Use only silver vsolid is dispersed first in acetone, and then metal niters having 0.45 fjun maximum pore size, hexane is added to recover the acetone together obtainable from Solas Flotronics. Filters should with the desorbed insecticides. The extract is be rinsed with acetone and heated to 300°C over­ washed with distilled water and dried over night to reduce interfering substances, sodium sulfate. A preliminary gas chromato­ 4.5 Shaker table, or combination shaker table graphic analysis is performed before concentration and wrist-action shaker having a 12-container and cleanup. Following this, the volume is re­ capacity. duced and extraneous material is removed by microcolumn adsorption ehromatography. The 5. Reagents insecticides are determined by gas chromatog- 5.1 Acetone^ distilled in glass, pesticide- raphy. analysis quality,

2. Application 5.2 n-hejrane, distilled in glass, pesticide- analysis quality, Sediment and bed material may be analyzed 5.3 Sodium sulfate, anhydrous, granular. by this method. Water samples containing sus­ Prepare by heating at 300°C overnight and store pended sediment may also be analyzed by this at 130°C, technique. The insecticides aldrin, p,7/,DDD, 5.4 Water, distilled water obtained from a p,p'-DDE, o,p-DDT, p,p'-DDT, dioldrin, en- high-purity tin-lined still. The feed water is drin, heptachlor, heptachlor epoxide, isodrin, passed through an activated carbon filter. The lindane, and methoxychlor, may be determined distillate is collected in a tin-silver-lined storage down to 0.20 jug/kg for a 50.0-g sample, on a tank, and the water is constantly irradiated with dry-weight basis. The pesticide chemicals chlor- ultraviolet light during storage. A gravity system dan, chlordene, hexachlorobicydoheptadiene, hex- is used, and no plastic material other than teflon achlorocyclopentadiene, and toxaphene may also is allowed to contact the distilled water. be determined by this method.

3. Interferences 6. Procedure As in the analysis of water for pesticides, Samples should be collected according to the chlorinated hydrocarbon compounds similar to recommended practice for suspended sediment pesticide chemicals give the most interference. and bed materials. Special care should be taken Organic coextractivcs, occurring naturally in to avoid contaminating the sample with oil from sediments and soils, are usually adequately re­ the sampling device. Rubber gaskets should be moved from extracts by the cleanup micro- replaced with teflon. Two suspended-sediment column. Sulfur compounds present in some samples should be taken, one for insecticide bottom muds often hinder electron-capture chro- analysis and the other for determining the sedi­ matography but do not appear to interfere with ment concentration and particle-size distribution. microcoulometrie gas chromatography. A 1-liter suspended sediment sample is needed for the insecticide determination. 4. Apparatus At least 150 g of material should be collected See step 4, "Gas Chromatographic Analysis of for each sample when only solids are to be ana­ Pesticides." lyzed. All samples must be kept in watertight 34 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS glass containers to prevent water loss and con­ Na2S04, and continue the analysis as in the pro­ tamination. No preservative is added. Unless cedure, beginning step 6.4, "Insecticides in analyzed within a few days of collection, the Wator." samples should be refrigerated and protected 6.2 Analysis of sediment, soil, and bed-material from light. samples. A reagent blank must accompany the 6.1 Procedure JOT water samples having suspended analysis. sediment. A reagent blank must accompany the 6.2.1 Desiccated samples, such as bed mate­ analysis. rial from dry streams, should be moistened with 6.1.1 Allow the water-sediment sample to re­ distilled water (to about 15 percent by weight). main undisturbed until the sediment has settled. Samples to which water is added are first pul­ Weigh the uncapped bottles on a balance to three verized, then mixed with the water, and then significant figures and carefully decant the water kept in an airtight glass container. A minimum of into a soparatory funnol of appropriate size. (Sepa­ 2 hr should be allowed for equilibration. Start rate by contrifugation as in 0.2.2 below and (or) the analysis of homogeneous samples at step filtration through motal membrane* filters if 6.2.3, below. necessary.) 6.2.2 Excessive water in sediment and bot­ 6.1.2 Measure 10 ml acetone or a volume tom-mud samples must be separated from the approximately half the equivalent volume of solids in order to obtain a homogeneous fraction solid, whichever is greater, into the sample bottle of tho sample. A proportionate amount of this containing the sediment. Replace the cup and wator is used later so that any suspended material gently mix the contents of the bottle on a shaker is included in the analysis. This technique may table for 20 min. Add 25 ml n-hcxane and mix the also be used whenever wator and solids are to be contents for an additional 10 min. Decant the analyzed separately. Weigh the container and extract into the separatory funnel containing the contents and transfer the sample to centrifuge water from the sample. Repeat the extraction of bottles. Spin tho solids at a relative centrifuge the sediment in the same manner two more times, force of 500-1,000 times gravity. Use tho super­ using fresh acetone and hexane each time. natant water to complete the transfer and ropoat NOTK.—Additional hcxnno may be needed to the centrifugution as necessary. Decant tho recover the acetone extract from the sediment. separated water into the empty tared sample Also, the extract may have to be filtered through container, (See step 6.2.5.) Calculate the weight a plug of glass wool. Anhydrous sodium sulfate of the solid by difference. may bo added to uid in separating the solvent 6.2.3 Thoroughly mix tho moist solid until from the sediment. Add the sodium sulfate slowly homogeneous and then weigh 50.0 g into a 250-ml and mix to the desired consistency. A quantity orlonmoyor Mask having a ground-glass stopper. of sodium sulfate equal to the amount of sediment Also at this time, weigh an additional 10.00 g of may bo added if necessary. the solid into a tared 50-ml beakor to be heated 6.1.3 Shako tho combined extracts with the at 130°C overnight for moisture determination. water from the sample for 1 min. Rinse tho sodi- 6.2.4 Measure 40 ml acetone into the erlen- mont from iho sample bottle with distilled water meyer flask containing tho sample and clamp the and collect tho wator from tho sample in tho stopper in place. (If tho sediment is sandy, use sample bottlo. Tho sediment may bo discarded. 20 ml acetone instead.) Mix the contents of the Decant, tho extract from tho top of tho separatory flask for 20 rnin using a wrist-action shaker. Add funnol info a 250-ml orlonmoyor flask. 80 ml hexane and shako again for 10 min. Decant 6.1.4 Extract the wator from tho sample the extract into a separatory funnol containing with an additional 25 rnl hexano and discard tho 500 ml distilled water. Add 20 ml acetone to tho wator. Weigh tho sample bottlo to determine tho orlonmoyor flask and shako 20 min. Again add 80 weight of tho sample. ml hexane, shako 10 min, and decant, the extract 6.1.5 Combine tho extracts in tho separatory into the separatory funnol. Repeat as in tho second funnol and wash two times with 500 ml distilled extraction one more time, wator oach time. Collect tho extract in tho 250-ml NOTE. If the* sediment is not wet. enough to orlomnoyor flask containing approximately 0.5 g agglomerate when tho hexane is added, add water, ANALYSIS OF ORGANIC SUBSTANCES IN WATER 35 by drops, while swirling the flask and observe if 9. Precision this helps. Very sandy material may remain dis­ The recovery of pesticides from sediments and persed. Extremely wet, mucky sediments may be soils is mainly dependent on two factors: (1) the better handled by the addition of anhydrous ability of the solvent to remove the pesticide sodium sulfate. Add sodium sulfate, in small from the solid, and (2) the amount of solvent re­ quantities, until the desired consistency is at­ claimed at each extraction step. Comparative tained or until the amount added approximates studies of single and exhaustive extractions of the weight of the sediment. The extract volume soil samples taken from contaminated fields recovered should be measured at each extraction showed that the extraction technique described to insure that 75 percent or more is regained. If removed 90-95 percent of the chlorinated pesti­ not, additional extractions are necessary to ob­ cides. Dehydrated clay soils, however, proved tain quantitative removal of the insecticides. slow to yield the pesticides unless they were 6.2.5 If any water was separated in step premoistened. Apparently, the collapsed layers 6.2.2 above, mix thoroughly and weigh out an of certain dehydrated clays and the resulting aliquot equivalent to the fraction of solid taken agglomerates entrap the pesticides, and the for analysis. Transfer the aliquot of water into addition of water prior to analysis helps to open the separatory funnel containing the sample ex­ the layers and separate the aggregation. It is tract and distilled water. imperative that sufficient solvent be reclaimed 6.2.6 Gently mix the contents of the separa­ at each extraction to avoid low results. Removal tory funnel for about 1 min and allow the layers of 90-95 percent of the desorbed pesticides may to separate. Collect the water in a clean beaker be expected if at least 75 percent of the solvent is and decant the extract into a 500-ml erlenmeyer recovered at each extraction step. flask. Back-extract the water wash with 25 ml hexane. Combine the solvent layers and wash with fresh 500-ml quantities of distilled water two more times. Discard the water layers and Chlorinated phenoxy acid herbicides collect the washed extract in the 500-ml erlen­ in water meyer flask to which has been added about 0,5 g anhydrous sodium sulfate. Gas chromatographic method NOTE.—A preliminary gas chromatographic analysis at this point is helpful for determining 1. Summary of method the volume reduction necessary. Chlorinated phenoxy acids and their esters are Proceed with the analysis beginning at step extracted from the acidified water sample with 6.4, "Insecticides in Water/' ethyl ether. The extracts are hydrolyzed, and extraneous materials are removed with a solvent 7. Calculations wash. The acids are converted to their methyl See step 7, "Gas Chromatographic Analysis of esters and are further cleaned up on an adsorp­ Pesticides." tion microcolumn. The esters are determined by gas chromatography. This method is a modifica­ 8. Report tion and extension of the procedure developed by The pesticide concentrations in water-sediment Goerlitz and Lamar (1967). mixtures should be reported as in step 8, "In­ secticides in Water," and the sediment concen­ 2. Application tration should accompan}7 the report. The The method is usable for the analysis of esters concentration of pesticides in sediment, soil, and and salts of 2,4-D (2,4-dichlorophenoxyacetic bed material is reported on a dry-weight basis as acid), silvex [2-(2,4,5-trichlorophenoxy) propi- follows: Less than 1.0 Mg/kg to one decimal; 1.0 onic acid], 2,4,5-T (2,4,5-trichlorophenoxy- Atg/kg and greater to two significant figures. Be­ acetic acid), and similar herbicides found in cause negative bias exists in the extraction pro­ water. Concentrations as .low as 0.02 /zg/1 of cedure, insecticides found in sediments and solids 2,4-D and 0.005 Mg/1 of silvex and 2,4,5-T in are considered minimum amounts. 1 liter of water may be determined. 36 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

3. Interferences form in redistilled ether, making it hazardous for Halogenated organic acids and their salts and storage and subsequent use. esters cause interference when BF3-methanol 5.5 Florisil adsorbent, Florisil, PR grade, esterification is used, and both the acids and commercially activated at 650°C and stored at halogenated phenols interfere when diazomethane 130°C in a glass-stoppered bottle. is used for esterification. 5.6 Herbicides, chlorinated phenoxy acids, reference grade: 2,4-D mp (melting point) 138°- 4. Apparatus 139°C; silvex, mp 181°-182°C; and 2,4,5-T, mp See step 4, "Gas Chromatographic Analysis of 154°-155°C. The methyl esters of the herbicides Pesticides." may be obtained from commercial sources. The 4.1 Concentrating apparatus. A Kuderna- methyl esters may also be prepared by reacting Danish concentrator, 250-ml capacity, with a 0.5-1.0 g herbicide acid with 50 ml BF3-methanol l-ball Snyder column is used for the initial con­ reagent at reflux for 1 hr. The methyl ester is ex­ centration step. Final concentration is performed tracted in ether, washed with 5-percent Na2CO3 in the receiver using a l-ball Snyder micro- solution, and finally washed with distilled water. column. A 4.00-ml graduated receiver tube is The ether extract is dried over anhydrous used for the diazomethane esterification, and a Na.2C03, and the ester is isolated by removing 5.00-ml volumetric flask receiver is utilized for the ether under vacuum. the boron trifluoride-methanol esterification. 5.7 HI'ethanol, reagent grade, redistilled from 4.2 Erlenmeyer flasks, 250 ml and 500 ml, an all-glass packed-column still after reacting having ground-glass stoppers. with 5 g of magnesium lathe turnings per liter of 4.3 Pasteur pipets, disposable, 14-cm long solvent. and 5-mm inside diameter. 5.8 X-methyl-N-nitroso-p-toluenesulfonamide, 4.4 Sandbath, fluidized, Tecam, or equiva­ mp 60°-62°C. lent. 5.9 Potassium hydroxide reagent, 1M solu­ 4.5 Separatory funnels, Squibb form, some of tion: Prepare by dissolving 78 g KOH reagent- 1-, or 2-liter capacity and others of 60-ml ca­ grade pellets in 200 ml carbon-dioxide-free € pacity. No lubricant is used on the stopcocks. distilled water. Reflux for 8 hr to reduce inter­ fering substances. A calcium chloride tube filled 5. Reagents with Ascarite is used at the top of the reflux All reagents must be checked for purity as re­ condenser to exclude carbon dioxide. agent blanks using the gas chromatographic 5.10 Silicic acid, chromatographic grade, procedure. Effort is saved by selecting high- 100/200 mesh, heated at 300°C overnight and quality reagents that do not require further stored ut 130°C in a glass-stoppered bottle. preparation. However, some purification of re­ 5.11 Sodium sulfate, reagent grade, an­ agents may be necessary as outlined below. hydrous, granular; heat-treated at 300°C for 24 5.1 Boron trifluoride-methanol, esterification hr. The heat-treated material is divided, and one reagent: Dissolve 14,0 g BF3 gas in 86.0 g an­ part is labeled "neutral- sodium sulfate" and hydrous methanol. stored at 130°C in a glass-stoppered bottle. The 5.2 Benzene, distilled in glass, pesticide- other part is slurried with enough ether to cover residue quality, such as Nanograde, or equivalent. the crystals und acidified to pH 4 by adding a 5.3 2-(2-Ethoxyethoxy) ethanol, high purity, few drops of purified sulfuric jicid. (To determine AV6°C 1.4068. the pH, a srnnll quantity of the slurry is removed, 5.4 Ethyl ether, reagent grade, redistilled the ether evaporated, water added to cover the from an all-glass packed-column still, after re- crystals, ;ind the pH is measured on a pH meter.) fluxing over granulated sodium-lead alloy for The ether is removed by vacuum, and the treated 8 hr. Purity is checked by gas chromatography material is labeled "acidified sodium sulfate" and after a part is evaporated to one-tenth the stored at 130°C in a glass-stoppered bottle. original volume. The purified ether should be 5.12 Sodium sulfate solution, 0.35717: Prepare distilled as needed and should never be stored for by dissolving 50 g neutral sodium sulfntc in 1.0 more than 1 month. Explosive peroxides readily liter distilled water. ANALYSIS OF ORGANIC SUBSTANCES IN WATEU 37

5.13 Sulfuric acid, reagent grade (sp gr 1.84), the sample two more times, using 50 ml ether purified by distilling off water until a constant each time, and then combine the extracts in the boiling solution remains. The acid is refluxed for 250-ml erlenmeyer flask, about 4 hr. 6.3 Add 15 ml distilled water and a boiling chip to the extract and fit the flask with a 1-ball Snyder column. Remove the ether on a steam 6. Procedure bath in a hood and continue heating for a total Water samples for herbicide analysis should be of 90 min. collected according to the procedure described for 6.4 Allow to cool and transfer the water to a the collection of organic water samples. The 60-ml separatory funnel. Extract the basic solu­ samples may be preserved with sulfuric acid at tion once with 20 ml and two more times each the collection site only if the acid is supplied with with 10 ml of ether and discard the ether layers. the sampling package. Regardless of whether The herbicides remain in the aqueous phase. Add preserved or not, the samples must be iced or 2 ml cold (4°C) dilute sulfuric acid (1 part con­ refrigerated in the dark within 4 to 5 hr of collec­ centrated H2SO4 (sp gr 1.84) diluted to 4 parts tion. Samples must reach the laboratory within with distilled water) to the contents of the funnel 24 hr of collection if not acidified. A 1-liter sample to bring the pH to 2 or below, and extract the should be collected for each analysis. herbicides once with 20 ml and two times each 6.1 Immediately upon receipt in the labora­ with 10 ml of ether. Collect the extracts in a tory, the samples are acidified to pH 2 or lower 125-ml erlenmeyer flask containing about 0.5 g with the specially prepared sulfuric acid. If more acidified anhydrous sodium sulfatc. Cover the than 24 hr is required for shipment, then the flask with foil and allow the extract to remain in samples must be acidified at the collection site. contact with the sodium sulfate, preferably in an For this purpose, 5 ml of 1:1 diluted sulfuric acid, explosion-proof refrigerator, for at least 2 hr. sealed in a prescored glass ampoule must accom­ (Refer to steps 6.6 and 6.7, below, before con­ pany each empty sample container. Detailed tinuing.) instructions for proper addition should also be 6.5 Transfer the other solution into the included. After adding the acid the bottles Kuderna-Danish apparatus through glass wool in should be loosely capped for 5 min or so before a funnel, Use liberal washings of ether and break closing tightly. Refrigerate the sample until up the hardened sodium sulfatc to obtain quanti­ analysis. tative transfer. Concentrate the extract to about 6.2 Weigh the opened bottle containing the 0.5 ml on the fluidized sandbath heated to 60°- sample. Pour the sample into a 1- or 2-liter 70°C. Under no circumstances allow the extract separatory funnel. Allow the bottle to drain for to evaporate completely to dryness. Clear sand a few minutes and then weigh. Record the from the glass joints before opening. sample weight to three significant figures. Add 6.6 Esterijicatioji with diazouiethaiie. A 4.00-ml 150 ml ethyl ether to the sample bottle, rinse the graduated receiver tube is used with the Kuderna- sides thoroughly, and pour the solvent into the Danish apparatus when the sample is to be funnel. Shake the mixture vigorously for 1 min. esterified with diazomethano. Add a volume of Allow the contents to separate for at least 10 min. anhydrous methanol equal to 0.1 of the volume Occasionally, emulsions prevent adequate separa­ of the concentrated extract. Connect two 20- by tion. In this event, draw off the clear aqueous 150-mm test tubes in sorios with glass tubing layer, invert the separatory funnel and shake. through neoprene stoppers so that incoming (CAUTION: Vent the funnel frequently to pre­ nitrogen bubbles through the liquid in the tubes. vent forming excessive pressure.) Addition of At the outlet, position a piece of glass tubing small volumes of distilled water often aids re­ having a right-angle bend and a drawn-out tip moval of sediment from the ether layer. Small so that the gas can be bubbled through the amounts of water included in the extract do not sample. Add about 5 ml othor to the first test interfere with the analysis. Collect the extract in tube. To the second test tube, add 0.7 ml ether, a 250-ml ground-glass erlenmeyer flask containing 0,7 ml 2-(2-ethoxyethoxy) othanol, 1.0 ml 7M 2 ml 7I\f potassium hydroxide solution. Extract potassium hydroxide solution, and 0.1-0.2 g 38 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

N-methyl-N-nitroso-p-toluenesulfonamide. Imme­ for about 3 min for phase separation. Twirling diately position the second test tube and adjust the flask between the palms of the hands from the nitrogen flow through the apparatus to about time to time aids separation. The benzene layer 10 ml per minute. (CAUTION: Diazomethane is is pipetted from the receiver and passed through a toxic and explosive gas. The use of a good fume a small cleanup column prepared by plugging a hood is absolutely necessary.) Place the Kuderna- disposable pipet with glass wool and packing with Danish receiver so that the gas bubbles through 2,0 cm neutral anhydrous sodium sulfate over 1.5 the sample. Allow the reaction to proceed for cm Florisil adsorbent. The eluate is collected in a about 10 min, or less if the yellow color of diazo- graduated receiver. The transfer is completed by methane can be observed to persist in the sample repeating the extraction step with small quanti­ tube. Remove the tube containing the sample, ties of benzene until a final volume of 2.00 ml is stopper, and allow to stand in the hood for about attained. Add a few crystals of neutral anhydrous 30 min. Carefully discard all waste from the reac­ sodium sulfate to the benzene solution and tion. Add about 0.1-0.2 g silicic acid to the sample thoroughly mix for gas chromatographic analysis. solution to destroy excess diazomethane. After 6.8 Analyze the extract by gas chromatog­ evolution of nitrogen has subsided, pass the solu­ raphy under conditions optimized for the par­ tion through a disposable pipet plugged with ticular gas chromatographic system being used. glass wool and packed with 1.5 cm neutral Run the first analysis on the electron-capture anhydrous sodium sulfate over 1.5 cm Florisil chromatograph using the DC-200 column. For adsorbent. The eluate is collected in a graduated components in concentrations ranging from 0.01 receiver tube. The transfer is completed by //g/1 to 1.0 Mg/1, a second analysis by electron washing the receiver tube several times with small capture on the QF-1 column is required. Pesticides quantities of ether to a final volume of 2.00 ml. in concentrations greater than 1.0 jug/1 must be The tube is stoppered and the contents thoroughly analyzed using microcoulometric detection on mixed and analyzed by gas chromatography. both the DC-200 and QF-1 columns. 6.7 Esterijication ivith boron trifluoride-meth- atwl. A 5.00-ml volumetric receiver flask is used 7. Calculations with the Kuderna-Danish apparatus when the sample is to be esterified with boron trifluoride- See step 7, "Gas Chromatographic Analysis of methanol reagent. Prior to the initial concentra­ Pesticides." tion step described previously (see step 6.5, Each gas chromatographic system must be above), 0.5 ml benzene is added to the extract in calibrated with standards. Methyl ester stand­ the Kuderna-Danish apparatus. The extract is ards of the herbicides must be converted to the concentrated to less than 1 ml, and the walls of acid equivalent. During analysis, at least two the flask are washed down with a small amount standards should be run so that the standard of ether. Sand adhering to the joint is cleared off curve can be corrected for day-to-day instru­ with an air gun or brush, and the receiver is mental fluctuation. fitted with a 1-ball Snyder microcolumn. The liquid volume is further reduced to 0.5 ml in the 8. Report sandbath. After the benzene solution in the re­ The pesticide concentrations are reported as ceiver has cooled, 0.5 ml boron trifluoride- follows: Less than 1.0 Mg/1, two decimals, and methanol reagent is added. The Snyder column report less than 0.005 ptg/1 as 0.00 Mg/1; 1.0 /*g/l is used as an air-cooled condenser, and the con­ and above, two significant figures. If more than tents of the receiver are held at 50°C for 30 min one column or gas chromatographic system is in a sandbath. The reaction mixture is allowed used, report the lowest value found. to cool to room temperature. About 4.5 ml of the sodium sulfate solution is added to the reaction mixture so that the benzene-water interface is 9. Precision observed in the constricted neck of the receiver The results vary ±20 percent for 2,4-D at the flask. The flask is stoppered with a glass plug and 0.10-Mg/l level and ±10 percent at LO-jug/1 vigorously shaken for about 1 min. Allow to stand concentration. ANALYSIS OF ORGANIC SUBSTANCE^ IN WATER 39

Reference analysis quality, such as Nanograde, or equiva­ Goerlitz, D. F., and Lamar, W. L., 1967, Determination lent. of phenoxy acid herbicides in water by electron- 5.2 Hydrochloric acid, concentrated (sp gr capture and microcoulometric gas chromatography: 1.18), ACS reagent grade. U.S. Geol. Survey Water-Supply Paper 1817-C, 21 p. 5.3 Sodium sulfale solution, 0.35 Jl/: Prepare by dissolving 50 g neutral sodium sulfate in 1.0 liter distilled water.

Chlorinated phenoxy acid herbicides 6. Procedure in sediment (tentative) Samples should be collected according to the recommended practice for suspended sediment or Gas chromatographic method bed materials. The samples must be iced or re­ 1. Summary of method frigerated. The analysis should begin as soon as possible because the herbicides, particularly Chlorinated phenoxy acids and their salts and 2,4-D, may decompose significantly in a few esters are extracted from an acidified slurry of hours. Samples should be kept in tightly closed sediment and water with acetone and ether. The glass containers to prevent water loss and con­ extract is hydrolyzed, and extraneous materials tamination. A reagent blank must accompany the are removed with a solvent wash. The acids are analysis. converted to their methyl esters and are further 6.1 Excess water is separated from the sedi­ cleaned up on an adsorption microcolumn. The esters are determined by gas chromatography. ment as described in steps 6.1 and 6.2, "Chlo­ rinated Hydrocarbon Insecticides in Suspended

2. Application Sediment and Bottom Material." The water from suspended sediment samples is analyzed as in the This method may be used for the analysis of procedure, "Chlorinated Phenoxy Acid Herbi­ esters and salts of 2,4-D (2,4-dichlorophenoxy- cides in Water/7 using proportionally less ether acetic acid), silvex [2-(2 ,4,5-trichlorophenoxy) for smaller amounts of water. Water separated propionic acid], and 2,4,5-T (2,4,5-trichloro- from bed material is included in step 6.5, below. phenoxyacetic acid), and similar herbicides found 6.2 Thoroughly mix the moist solid until in sediments. homogeneous and weigh 50.0 g into a 250-ml eiienmeyer flask having a ground-glass stopper. 3. Interferences Also at this time, weigh an additional 10.00 g of Halogenated organic acids having properties sample into a tared 50-ml beaker to be heated at similar to the herbicides cause interference when 130°C overnight for moisture determination. BF3-methanol esterification is used, and both the 6.3 While stirring, slowly add water to the extraneous acids and halogenated phenols inter­ sample in the erlenmeyer flask until the mixture fere when diazomethane is used for esterification. has the consistency of paste, or until water begins to separate. Acidify the slurry to pH 2 or below 4. Apparatus by the addition, by drops, of concentrated hydro­ chloric acid. Allow to stand with occasional See steps 4, "Gas Chromatographic Analysis of stirring for 15 min, and insure that the pH re­ Pesticides'* and "Chlorinated Phenoxy Acid mains below 2. Add more acid if necessary and Herbicides in Water," until stabilized. 4.1 Shaker table, or combination shaker table 6.4 Measure 40 ml acetone into the erlen­ and wrist-action shaker having a 12-container meyer flask containing the acidified sample and capacity. clamp the stopper in place. Mix the contents of the flask for 20 min using the wrist-action shaker. 5. Reagents Add 80 ml ether and shake again for 10 min. De­ See step 5, "Chlorinated Phenoxy Acid Herbi­ cant the extract into an appropriate-sized separa- cides in Water." tory funnel containing 250 ml 0.35M sodium 5.1 Acetone, distilled in glass, pesticide- sulfate. 40 TECHNIQUES OF WATER-RESOURCES INVESTIGATIONS

NOTE.—If the sediment does not settle to allow droxide and 15 ml distilled water to the extract decanting the solvent, add anhydrous sodium in the 500-ml erlenmeyer flask. Add a boiling sulfate in small amounts until the mixture chip and fit the flask with a 1-ball Snyder column. separates. A quantity of sodium sulfate equal to Evaporate the ether on a steam bath in a hood the amount of sample may be added if necessary. and continue the heating for a total of 90 min. To ensure adequate recovery, measure the Continue the analysis, beginning at step 6.4, volume of extract at each decanting step. "Chlorinated Phenoxy Acid Herbicides in Water." Add 20 ml acetone to the erlenmeyer flask and shake 20 min. Again, add 80 ml ether, shake 10 7. Calculations min, and decant the extract into the separatory See step 7, "Gas Chromutographic Analysis of funnel. Repeat the process as in the second ex­ Pesticides." traction one more time, and collect the acetone- Each gas chromatographic system must be ether extract in the separatory funnel containing calibrated with standards. Methyl ester standards the 0.35M sodium sulfate solution. of the herbicides must be converted to the acid 6.5 If any water was separated in step 6.1 equivalent. During analysis, at least two stand­ above, mix thoroughly and weigh out an aliquot ards should be run so that the detector response equivalent to the fraction of solid taken for curve can be corrected for day-to-day instru­ analysis. Transfer the aliquot of water to the mental fluctuation. separatory funnel containing the sample extract and the sodium sulfate solution. 8. Report 6.6 Gently mix the contents of the separa­ Pesticide concentrations in sediment are re­ tory funnel for about 1 min and allow the layers ported as follows: Less than 1.0 jug/kg; two to separate. Collect the aqueous layer in a clean decimals; 1.0 Mg/kg and above, to two significant beaker and collect the extract in a 500-ml ground- figures. If more than one column or the gas glass erlenmeyer flask. Back-extract the water chromatographic system is used, report the wash with 25 ml ether. Separate the aqueous lowest value found. layer and discard. Pour the ether layer into the erlenmeyer flask containing the sample extract. 9. Precision 6.7 Add 5 ml 7M aqueous potassium hy­ No precision d;iUi are :ivail;ible ;it this time.