Organic substances in water

848-C Edited by Phillip E. Greeson

GEOLOGICAL SURVEY CIRCULAR 848-C

Library of Congress catalog-card No. 81-607886

Organic Substances in Water Why Study Organic Substances in Water?

Organic substances are those substances containing compounds of carbon and hydrogen; that is, hydrocarbons or compounds derived from them. At one time, it was believed that hydrocarbons could be formed only by living organisms- thus, the term "organic." Now, however, man has learned not only to produce many of these natural compounds synthetically but has created thoasands of new organic compounds in his quest for a better life.

The presence of organic materials in our enviro~ ment is so extensive that it could be said that we live in an organic world. The food we eat, the clothes we wear, the homes in which we live all contain large amounts of organic substances. Everyone is aware of the tremendous increase in the use of plastics within the past few years, and the more recent trend toward production of biodegradable organic consumer products. The degradation of such materials, along with the disposal of wastes from their manufacture, constitute an ever-increasing source of organic contamination to our environment. On the other hand, there are manufactured organic substances that do not degrade, and the persistence of some of these compounds has caused great concern among environmentalists.

Organic substances are important in water-quality studies for many reasons. The bacterial oxidation of organic carbon from natural sources and from municipal, industrial, and agricultural wastes is a major factor leading to depletion of oxygen in many streams. Nutrients in organic substances can impart color and odor to water, and can act as carriers of trace metals and other elements by combining with them chemically or inducing their sorption by suspended sediment.

Every day we are adding to the list of manmade organic compounds that are suspected of having detrimental effects on the environment-either aesthetic or toxic. A few years ago, industry initiated an expensive change in formulation of detergents to reduce the unwanted persistence of their foaming properties. An ever-increasing number of pesticides have been banned or severely limited in application because of their documented or suspected adverse effects on valuable living things- both plant and animal; these effects vary from reduction of the probability of survival of newly hatched fish to causing cancer or birth defects in humans. There has been a recent suggestion that some of the substances considered most dangerous to man may be produced from harmless organic compounds by the very processes which we use in our water treatment plants to make the water safe for human consumption.

Measurement of the unending variety of organic substances can be at many different levels of detail, ranging from the most gross measure, organic carbon, to determination of a specific organic compound or organometallic complex. These measurements can be either qualitative identifications or quantitative determinations. They may be for the purpose of establishing natural background levels of organic substances, or for regulatory detective work, such as identification of the source of polluting crude oil by its relative content of simple organic components.

Instrumentation for identification and quantification of organic substances is becoming more and more complex and expensive. Positive identifica- .provable to anyone who does not want to believe it. tion of a specific compound commonly requires a series of satisfactory comparisons with standards, using a different identification principle for each comparison, followed by a mass spectrogram for confirmation based on the way in which the compound breaks apart.

Questions about the effects on man or other forms of life of many of the organic substances present in our ground and surface waters will remain unanswered for many years. There is no good way to measure the chronic effect on man of longterm exposure to low concentrations of organic substances, except by waiting for that long term to pass. Even then, probably only a statistical inference can be drawn- one which will be un-

The more information that we have on the occurrence and behavior of the wide variety of organic substances in water, the better will be the basis for the inferences that we must draw. When we conclude that a particular organic compound or group represents a hazard (or a benefit), we will have at least a limited basis for deciding on an appropriate course of action.

Taste and Odor in Water

One may ask why briefing papers intended to describe organic sustances and their effects on the environment should include a discussion of taste and odor. Nature has endowed most persons with a sensory system much more responsive than many advanced instrumental devices. The senses of odor and taste are olfactory (smell) and gustatory (taste) responses.

Judgment about water supplies is foremost a sensory evaluation. Clarity, sparkle, freedom of color, and absence of objectionable taste and odor are desired properties. Except for certain sulfurous or highly mineralized waters, the cause of taste and odor of water is the organic substances content. The organic constituents that cause taste, odor, color, and toxicity are often at concentrations of micrograms per liter (p. g/L) or less. It is this ability to sense very low levels of contaminants that has afforded protection to man during evolutionary time. Surveillance is required, if water resources are to be protected and consumers are to be provided a high quality supply.

Sensory impairment from water has been chiefly a surface-water problem; ground water has been less affected. Adsorption processes in soils and rock usually remove dissolved and particulate organics. Humic components and other naturally occurring organics always have been present in surface water. Decomposition of vegetative matter is the principal source. However, man has complicated the situation. Not only has he released quantities of municipal and industrial wastewaters with concomitant organic pollution, but many of the organics are not biodegradable. Such biodegradably resistant organics are termed "refractory." They can originate from landfill leachates, agricultural runoff, and manufacturing wastes.

Waste-treatment technology has accelerated and not improved on natural processes for organic reduction. Even the best of treatment plants release refractory organics. As water reuse intensifies, because of population and industrial growth, so does the risk of exposure to refractory materials. In turn, refractory buildup requires more advanced wa.Ste-treatment processes.

Water-treatment plants generally can cope with low levels of organic wastes. They fail when slug discharges or organics are involved. Even the best of today's ·plants has limited analytical capability for providing early warning of slugs of organic constituents. By the time routine procedures identify a problem, there are already sources of consumer complaints in the distribution system. Seasonal and climatic factors also are involved. During low flow and decreased dilution, constituents normally present in water are more concentrated. Heavy rainfall or other runoff results in flushouts. Flushout dilution is offset by scour of decaying vegetative matter in benthic (bottom) layers. The associated organic materials often impose a severe taste and odor impact.

In the vernacular of the water technology, "taste and odor" are inseparable. Separation of the effect of odor from taste is impossible. There are four basic tastes: salty, sweet, sour, and bitter. Characterization of taste (flavor) by food technologists has been more successful than have attempts at characterization of odor. Flavor is the sum of taste and odor sensations. Odor characterization schemes have been proposed, but none has proven suitable for field use. Individuals engaged in related research often have problems describing odor character and communicating a precise description. that alter the trace organic composition. Chemical and physical processes also are involved. If sensory measurements are made on stored samples, the results should note this fact- including the duration and conditions of storage. Sample temperature at the time of collection should be recorded. Samples that are at a temperature exceeding that specified for the sensory test should be cooled to the proper level.

ODOR MEASUREMENTS

Measurements are made to determine intensity and to characterize (describe) odor. The principle of the measurement is relatively simple. The water sample is diluted with odor-free water until the least definitely perceptible odor is obtained. Increasing concentrations of odorant are tested beginning with levels below the threshold. Odor is reported as threshold odor number, TON, or as odor intensity index, On. These terms are related: TON =2°II.

The On is the number of times the concentration of the original sample is halved by addition of odor-free water to obtain the least definitely perceptible odor. The index is most convenient with industrial wastewaters. An On of 18 immediately imparts an impression of the strength based on the halving of the orginal sample 18 times. The equivalent TON is 262,000. The TON is the greatest dilution of the sample with odor-free water to yield the least definitely perceptible odor. A treatment-plant rawwater supply with a TON of 8 would have an on of

3. Note that when the TON= 1, the On=O. The aforementioned methods of testing for odor

describe precise controls for assuring odor-free glassware, odor-free dilution water, sample temperature, and environmental and operating conditions. It is imperative that the precautions and controls be observed, if meaningful results are to be obtained.

Research has established a basis for these controls. Ideally, the test area should be temperatureand humidity-controlled, and the air free of background odor. If the samples are colored, the testers will be biased in selecting the flask containing the sample from those containing only odor-free water. This problem can be eliminated by using opaque glassware for the samples or by monochromatic lighting that masks the sample color. Color is a factor in perception of taste. Food technologists have documented many instances when an individual attributed widely varying flavors to the same substance (such as ice cream) when taste-and-odor-free coloring was added. However, if the individual is blindfolded or monochromatic lighting hides the color change, consistent and objective judgments are made. Samples containing turbidity should be handled in the same manner as colored samples to preclude bias.

Odor tests are made in covered 500-mL, widemouth Erlenmeyer flasks containing 200 mL of solution. The solution may be entirely sample or some dilution of sample. The odor-free dilution water is prepared by passing the best available reagent-grade water, often freshly distilled, through a fresh activated-carbon column. The carbon is contained in an all-glass system. However, the carbon surface can serve as a host for algae or other organisms which produce odorous metabolites. Therefore, the condition of the carbon and the quality of the odor-free water must be monitored carefully.

The specification that sensory measurements be made in an odor-free environment is based on experimental experience. The senses are affected by background odor. Similar background odors inhibit threshold measurements. Dissimilar odors enhance threshold detection. This was demonstrated in studies using n-butural and m-cresol. The former has a pleasant, sweet odor at low concentration. The latter has a "fence post," noxious odor. Unfortunately, field tests are seldom made without some background odor. Most water treatment plants are not equipped with an odor-free laboratory. The odor of chlorine and related compounds prevails. Every effort should be made to conduct sensory tests in areas of minimum background odor.

Temperature is a factor in olfaction. The reason is volatility. As temperature increases, volatility is enhanced. Odor threshold measurements usually are made at 40°C (near body temperature) or 60°C. Research has shown that for selected stimuli in aqueous solution, the TON will not differ significantly over a range of 20° to 60°C. However, these were precisely controlled tests (Baker, 1963), under ideal laboratory conditions. The panelists in the controlled tests noted that at 60°C, a temperature preferred by water utilities because of hot-water domestic implications, the odorant was readily lost and repeat tests required fresh samples. A "steaming" effect on the olfactory system was described. At 20°C, the samples were "dead." The 40°C temperature was preferred. Details of the odor-threshold testing procedure will not be given. These are described in American Society for Testing and Materials (1979) and American Public Health Association and others (1976).

STANDARDS

Quantitative odor standards can be of two kinds. A limit can be set on odor of the water as a whole, or the concentrations of individual substances can be regulated. The concentrations of substances are specified in terms of a ratio between allowable concentration and the threshold odor or taste concentration. Such standards are difficult to administer. The great variation in sensory response by individuals raises the question of the choice of regulatory limit. However, there is another factor. Many times the sensory causing substances combine in a nonadditive manner.

Two odorants mixed in aqueous solution may create (1) additivity, the resulting odor is the sum of the individual odors, (2) synergism or intensification, the odor produced exceeds that predicted by simple addition, or (3) antagonism or suppression, the odor produced is less than that predicted by simple addition. These effects are described by Rosen and others (1962) and Baker (1963). Synergism is the effect experienced most often and it has practical implications. If an industrial or municipal waste effluent of acceptable threshold is discharged to a receiving body with no appreciable odor, synergism could produce a water of unacceptable odor or synergism could produce a water of unacceptable quality. Obviously, standards based on single components or even the waters treated as seperate entities would fail.

European communities have set recommended taste and odor numbers for surface water supplies at values of 3 to 20. This variation is caused by their definition of classes of water use, ranging from the cleanest to the most polluted class. The World Health Organization states in its 1971 International Standards for drinking water that there should not be objectionable taste and odor in water. The Commission of European Communities has issued a proposed directive stating that a TON of zero is recommended with a maximum TON of 2 at 12°C or 3 at 25°C for drinking water. It is interesting that Russian standards for organic substances in raw waters to be used as domestic (drinking) supplies are based solely on taste or odor threshold concentrations. Generally, sensory threshold values are well below toxicologically dangerous levels. For most of the many organic substances in water, there is no toxicological information, particularly for defining risk from chronic exposure to low concentrations.

Classification and Fractionation of Organic Solutes in Natural Waters

With the recent emphasis upon development of the Nation's fossil fuel resources to meet the growing need for energy, those scientists involved with water-quality studies are recognizing an increasing need for more definitive characterization of organic solutes in water so that changes in water quality due to organic residuals inputs can be determined. The need for organic solute characterization has long been recognized, but progress has been slow relative to inorganic solute characterization for the following reasons:

1. The large number of both known and unknown organic compounds which may be found in water make specific compound identification and quantification extremely difficult.
2. Organic solute concentrations in water are usually less than inorganic solute concentrations by two or more orders of magnitude.
3. Many organic solutes are unstable or volatile, which leads to water sampling and preservation problems.
4. Analytical instrumentation needs for organic solute characterization are usually complex, expensive, and often inadequate. The following brief discussion is intended to

present a simplified scheme of organic solute characterization which attempts to minimize these problems.

Although there is no single generally accepted classification scheme for organic compounds, separation techniques based on solubility and acidbase behavior have been in use for many years (figure 1) (Shriner and Fuson, 1940). These separation-classification schemes are useful, and indeed, are usually a necessary step in the identification of unknown organic compounds. However, they are time consuming, offer no means of concentrating the organic solutes, and do not relate the quantity of various subgroups to the total amount of organic solutes present. The development in recent years of instruments to rapidly determine dissolved organic carbon with precision and accuracy, as well as the development of synthetic resin adsorbents which concentrate and fractionate organic solutes by selective adsorption and desorption, permit greater utulization of these techniques in studies of organic solutes in water.

The present state of organic solute characterization is described graphically in a report by Christman and Hrutfiord (1973). They state that the development of organic water-quality standards have "* * * occurred on a piecemeal basis in response to current crises. At one time or another, scientific teams have deliberated the significance of and have established water-quality standards for specific compounds such as nitrilotriacetate (NTA) and phthalic acid; certain classes of compounds such as pesticides, phenols, polychlorinated biphenyls, phthalate esters, oil, and grease; and poorly defined groups of compounds known primarily by their properties or manner of isolation such as foaming agents, color, taste and odor, as well as the carbon adsorbable organics * * * "

s~nsol. ~L

sl s2

Characterization schemes based on these standards limit themselves for the following reasons:

1. They are not comprehensive schemes designed to cover all organic solutes in water. Therefore, they apply only to the specifics for which they were designed, and many organic solutes are not detected or even suspected.
2. A materials balance based on organic carbon is not used. Therefore, one has no idea how the specific compounds and compound classes relate to the total organic solute concentration.
3. Many of the parameters such as color, taste and odor, and oils and grease are outdated and nonspecific.
4. Most organic solute characterization schemes are based on methodology which incorporates the liquid-liquid extraction or carbon adsorption to extract and concentrate organic solutes, prior to analysis. Both of these analytical techniques have their limitations and drawbacks. Liquid-liquid extraction is an efficient method for

extracting certain classes of organic solutes from water; particulary those nonpolar solutes which have low-water solubility and high solubility in organic solvents immiscible with water (ASTM Method D-2778, 1975). However, liquid-liquid extraction is relatively ineffecient as a method for extraction of highly polar organic solutes which prefer the aqueous phase. Highly polar organic solutes comprise a significant proportion of organic compounds found in natural waters and include polybasic acids, hydroxy acids, polybasic alcohols, polysaccharides, and amino acids. An examination of the organic solute distribution coefficients listed in the International Critical Tables (Forbes and Anderson, 1928) shows that these types of highly polar compounds are not significantly extractable by organic solvents which are immiscible with water.

Adsorption of organic solutes from aqueous solution upon granular activated carbon currently is the most widely used method of concentrating and fractionating organic solutes (ASTM Method

an agent which removes organic solutes from water by adsorption; however, its use in quantitative analytical schemes is limited by the fact that many organic solutes are adsorbed irreversibly upon and altered by the activated carbon. The limitations are documented in the literature (Baker and Malo, 1967; Burnham and others, 1973.)

In a comparison of processes which concentrate organic solutes found in natural water, adsorption is inherently superior to liquid-liquid extraction, flash evaporation, freeze concentration, and freeze drying for the following reasons:

L It is a low-energy process which does not involve a phase transition. Adsorption does not remove the organic solute from the aqueous environment as does liquid-liquid extraction and freeze drying, and it takes less energy for a solute to be adsorbed on a surface than to cross a liquid-liquid interface. Extraction into a different solvent also often alters the chemical nature of the solute. The low temperature and high vapor pressure at which adsorption occurs prevents losses of volatile solutes which are lost during flash evaporation or freeze drying.

2. Adsorption is selective. Adsorbents can be selected, such that organic solutes are separated from inorganic solutes, and organic solutes can be fractionated into different groups depending upon which adsorption mechanism is used. Flash evaporation, freeze concentration, and freeze drying indiscriminately concentrate both organic and inorganic solutes, such that reactions between the solutes often occur during concentration.

The main disadvantage of adsorption is that frequently it is irreversible, as often occurs on activated carbon and certain exchange resins. Irreversible adsorption usually is due to the heterogenous nature of the adsorbent's surface which binds the adsorbed solute by multiple adsorption mechanisms. An adsorbent used for analytical concentration and separations should have a homogeneous surface, such that only one adsorption mechanism is operative so that solutes can be quantitatively desorbed for analysis.

A proposed classification scheme for organic solute characterization based on resin adsorbents is given in figure 2. This classification scheme has a logical progression from one level of specificity to the next, and it uses materials balance based on organic carbon to quantify organic solute concentrations and fractionations. The organic load parameter, dissolved organic carbon, is very specific for organic solutes. The meaning and significance of the dissolved organic carbon parameters is discussed in a report by Malcolm and Leenheer (1973).

,,,I I \

I I

• ::::=---....J

admittedly method dependent upon the analytical fractionation scheme given in figure 3. The mechanisms and methodology of hydrophobic organic solute adsorption is discussed by Junk and others (1974). Hydrophobic organic bases are separated by desorption from the resins with 0.1 M HCl and hydrophobic acids are separated by desorption using 0.1 M NH4 OH as the eluent. Hydrophobic organic neutral solutes remain on the column, and they are quantified by DOC differences. Most hydrophobic neutral solutes are soluble in organic solvents such as ethyl ether and menthanol, which are used as desorption solvents.

The DOC which remains in the eluent from the column packed with the XAD resins quantifies the hydrophilic organic solutes. Hydrophilic organic bases first are separated by passing the water sample through a glass column packed with a hydrogen-saturated cation-exchange resin which adsorbs organic bases which are protonated cations at sample pH 2. Hydrophilic organic acids next are separated in a similar manner by adsorption on a hydroxide-saturated anion-exchange resin. The DOC remaining in solution after the entire resin-adsorption procedure quantifies hydrophilic organic neutral solutes. Hydrophilic organic bases can be desorbed and eluted from the cation-exchange resin by 1 M NaOH and hydrophilic organic acids can be displaced from the anion-exchange resin by 1 M HCl. The many applications of ion-exchange resins for analytical separations are discussed by Khym (1974).

The analytical determination of selected organic compounds at the fourth level of classification can

derived from the DOC fractionation analysis. Fairly specific colorimetric methods exist for anionic detergents, phenols, polysaccharides, polyuronic acids, amino acids, and others. The DOC fractionation analysis greatly aids the compound class analyses because the resin adsorption precedure serves to concentrate trace organic solutes, and the resultant fractionation removes many organic and inorganic interferents. Only the hydrophilic organic neutral solute fraction is not concentrated by DOC fractionation analysis, and this fraction can be effectively freeze concentrated because most of the inorganic solutes which form precipitates when freeze concentrated, have been removed previously by ion exchange. The methodology and limitations of freeze concentration are discussed by Baker (1969).

The analytical determination of certain compound classes, specific organic solute compounds, and organic solute complexes usually demands additional fractionation by either liquid or gas chromatograph. Only about 10 percent of organic solute compounds are sufficiently volatile to be fractionated by gas chromatography; the remaining 90 percent must be fractionated by liquid chromatography. An excellent text on gas chromatography is by McNair and Bonelli (1969), and liquid chromatography is discussed by Hadden and others (1971). Most of the volatile organic solutes, such as various pesticides and hydrocarbons, are found in the hydrophobic neutral fraction. The volatility of organic solutes in the other fractions frequently can be enhanced by making nonpolar derivatives which have a greater volatility, thus making gas chromatography of these solutes possible. Gas chromatography still is preferred to liquid chromatography as a means of organic solute fractionation because of its greater versatility due to the absence of solvent during solute separation, and because of the greater number and sensitivity of solute detectors.

Final identification of specific organic solutes and solute complexes usually is accomplished by obtaining spectral information on the pure organic solute. Mass spectra, infrared spectra, nuclear magnetic resonance spectra, and ultraviolet-visible spectra are the most generally used spectra for identification purposes. Gas or liquid chromatography also can be used to identify an organic solute by comparison of its fractionation characteristics with the fractionation characteristics of various organic solute standards.

This paper is the first written communication about the proposed organic solute classification scheme; therefore, the data which follow are limited to a few samples which were processed by the DOC fractionation analysis to test levels 1-3 of classification scheme. Hopefully, these data will provide an indicator of how the organic solute classification scheme can be used, applied, and interpreted in water-quality studies.

The first natural water samples to be processed by part of the DOC fractionation analysis were two ground-water samples. The samples were collected to observe changes in organic solute composition which occurred when a coal seam about 90 m below land surface was ignited in a U.S. Bureau of Mines in situ coal gasification experiment. When the burn was extinguished, ground water flooded the burned zone and became contaminated with organic combustion products. The data are presented in table 1. Sample 1 is uncontaiminated ground water withdrawn from the coal seam outside the burned area. Sample 2 is contaminated ground water withdrawn from a well in the center of the burned area.

The data in table 1 could be interpreted in the following manner. The greater ratios of hydrophobic:hydrophilic DOC in sample 1 most likely indicate that the DOC of the uncontaminated sample is more hydrocarbon (hydrophobic) in nature, and is at a lower oxidation state as compared to the contaminated sample 2. Most likely, the combustion of coal during gasification gave rise to organic solutes with a higher oxidation state (more acids), and a greater polar functional group content (more hydrophilic) as compared to organic solutes in the uncontaminated ground water. The elemental carbon which resulted from the carbonization of coal during gasification also may have adsorbed some of the hydrophobic solutes in contaminated sample 2, such that the hydrophobic DOC concentration is reduced. In this case, DOC fractionation analysis showed large fraction ratio changes which can be interpreted in terms of organic solute changes.

The last sample processed by DOC fractionation analysis illustrates an application to studies involving organic-inorganic complexes. In a study of ground water of high selenium content near Golden, Colorado, it was noted that several samples contained high concentrations of DOC, as well as selenium. An experimental procedure based on DOC fractionation analysis was designed to observe also selenium fractionation both in the natural ground water and in a synthesized water sample which contained similar concentrations of selenium and other inorganic solutes with no DOC. Comparison of the data for the two samples should allow certain conclusions about the association between DOC and selenium. The data are presented in table 2.

The data clearly show that only minor amounts of selenium were associated with the hydrophobic organic soluted fractions which comprise 58 percent of the total DOC. Almost all of the selenium was adsorbed from both samples as an anion on the anion-exchange resin. Because the fractionation of selenium in sample 2, which contained only inorganic solutes, was identical to the selenium fractionation of sample 1 which was the ground water, it is virtually certain that selenium in the ground water is not directly associated with organic solutes. This is not a surprising conclusion because most organic solutes and selenium solutes have anionic natures which discourage complex formation.

This paper has shown how organic solutes can be classified, fractionated, and anlayzed so that meaningful data are obtained without resorting to highly complex organic analyses. DOC fractionation analysis should be useful for indicating relative abundances and changes in composition of organic subgroups, and a "finger-printing'' tool for various waters. DOC fractionation analysis also can be used as a preparative procedure for additional analyses ultimately leading to specific compound and compound-style identification.

Where this page came from

This page was imported from U.S. Geological Survey. Published by the U.S. Geological Survey and, as a work of the United States government, in the public domain.

Nobody has written it yet — it is the source material at a new address, which is why search engines are asked to skip it and why no one earns from it. It is up for grabs: take it on, and it is yours to rewrite and to earn from.