>EPA
United States
Environmental Protection
Agency •«
Office of
Drinking Walur
Washington CC 20460
July 1979
EPA-570/9-7(i-000
National Secondary
Drinking Water
Regulations
|U
LIBRARY
-------
-------
Preface
The National Secondary Drinking Water Regulations specify secondary
maximum contaminant levels (SMCLs) which, in the judgment of the
Administrator, are requisite to protect the public welfare. Contaminants
covered by these regulations are those which may adversely affect the
aesthetic quality of drinking water, such as taste, odor, color and appear-
ance, and which thereby may deter public acceptance of drinking water
provided by public water systems.
Secondary maximum contaminant levels (SMCLs) are established for
chloride, color, copper, corrosivity, foaming agents, iron, manganese, odor,
pH, sulfates, total dissolved solids and zinc. At considerably higher concen-
trations, these contaminants may also be associated with adverse health
implications. These secondary levels represent reasonable goals for drinking
water quality, but are not federally enforceable. Rather, they are intended
as guidelines for the States. The States may establish higher or lower levels
as appropriate to their particular circumstances dependent upon local
conditions, such as unavailability of alternate raw water sources or other
compelling factors, provided that public health and welfare are adequately
protected. However, odor, color and other aesthetic qualities are impor-
tant factors in the public's acceptance and confidence in the public water
system; thus, States are encouraged to implement these SMCLs so that
the public will not be driven to obtain drinking water from potentially
lower quality, higher risk sources.
Victor J. Kimm
Deputy Assistant Administrator for Drinking Water
U.S. Environmental Protection Agency
J. -, ENV.CBOMENIAL PROTECTION AGEtTCY
-------
Table of Contents
National Secondary Drinking Water Regulations
143.1 Purpose 1
143.2 Definitions 1
143.3 Secondary Maximum Contaminant Levels 2
143.4 Monitoring 2
Appendix A
Statement of Basis and Purpose for the National
Secondary Drinking Water Regulations 4
Chlorides 4
Color 8
Copper 10
Corrosivity 12
Foaming Agents 21
Iron 24
Manganese 26
Odor 28
pH 30
Sulfates 32
Total Dissolved Solids 34
Zinc 36
-------
National Secondary Drinking Water
Regulations
§ 143.1 Purpose
This part establishes National Secondary Drinking Water Regulations
pursuant to Section 1412 of the Safe Drinking Water Act, as amended (42
U.S.C. 300g-l). These regulations control contaminants in drinking water
that primarily affect the aesthetic qualities relating to the public accept-
ance of drinking water. At considerably higher concentrations of these
contaminants, health implications may also exist as well as aesthetic
degradation. The regulations are not Federally enforceable but are
intended as guidelines for the States.
§ 143.2 Definitions
(a) "Act" means the Safe Drinking Water Act as amended (42 U.S.C.
300f et seq.).
(b) "Contaminant" means any physical, chemical, biological, or radio-
logical substance or matter in water.
(c) "Public water system" means a system for the provision to the public
of piped water for human consumption, if such a system has at least fif-
teen service connections or regularly serves an average of at least twenty-
five individuals daily at least 60 days out of the year. Such term includes
(1) any collection, treatment, storage, and distribution facilities under
control of the operator of such system and used primarily in connection
with such system, and (2) any collection or pretreatment storage facilities
not under such control which are used primarily in connection with such
system. A public water system is either a "community water system" or a
"non-community water system."
(d) "State" means the agency of the State government which has juris-
diction over public water systems.
(e) "Supplier of water" means any person who owns or operates a public
water system.
(f) "Secondary Maximum Contaminant Levels" means SMCLs which apply
to public water systems and which, in the judgment of the Administrator,
are requisite to protect the public welfare. The SMCL means the maximum
permissible level of a contaminant in water which is delivered to the free
flowing outlet of the ultimate user of a public water system. Contaminants
added to the water under circumstances controlled by the user, except
those resulting from corrosion of piping and plumbing caused by water
quality, are excluded from this definition.
-------
§ 143.3 Secondary Maximum Contaminant Levels.
The Secondary Maximum Contaminant Levels for public water systems are
as follows:
Contaminant Level
Chloride 250 mg/1
Color 15 Color Units
Copper 1 mg/1
Corrosivity Non-Corrosive
Foaming Agents 0.5 mg/1
Iron 0.3 mg/1
Manganese 0.05 mg/1
Odor 3 Threshold Odor Number
pH 6.5-8.5
Sulfate 250 mg/1
Total Dissolved Solids (TDS) 500 mg/1
Zinc 5 mg/1
These levels represent reasonable goals for drinking water quality. The
States may establish higher or lower levels which may be appropriate
dependent upon local conditions such as unavailability of alternate source
waters or other compelling factors, provided that public health and welfare
are not adversely affected.
§ 143.4 Monitoring
(a) It is recommended that the parameters in these regulations should be
monitored at intervals no less frequent than the monitoring performed
for inorganic chemical contaminants listed in the National Interim Primary
Drinking Water Regulations as applicable to community water systems.
More frequent monitoring would be appropriate for specific parameters
such as pH, color, odor or others under certain circumstances as directed
by the State.
(b) Analyses conducted to determine compliance with § 143.3 should be
made in accordance with the following methods'.
(1) Chloride — Potentiometric Method, "Standard Methods for the Exami-
nation of Water and Wastewater," 14th Edition, p. 306.
(2) Color — Platinum-Cobalt Method, "Methods for Chemical Analysis of
Water and Wastes," pp. 36-38, EPA, Office of Technology Transfer, Wash-
ington, D.C. 20460, 1974, or "Standard Methods for the Examination of
Water and Wastewater," 13th Edition, pp. 160-162, 14th Edition,
pp. 64-66.
-------
(3) Copper — Atomic Absorption Method, "Methods for Chemical Analy-
sis of Water and Wastes," pp. 108-109, EPA, Office of Technology Trans-
fer, Washington, D.C 20460, 1974, or "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, pp. 210-215, 14th
Edition, pp. 144-147.
(4) Foaming Agents — Methylene Blue Method, "Methods for Chemical
Analysis of Water and Wastes," pp. 157-158, EPA, Office of Technology
Transfer, Washington, D.C. 20460, 1974, or "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, pp. 339-342, 14th
Edition, p. 600.
(5) Iron — Atomic Absorption Method, "Methods for Chemical Analysis
of Water and Wastes," pp. 110-111, EPA, Office of Technology Transfer,
Washington, D.C. 20460, 1974, or "Standard Methods for the Examina-
tion of Water and Wastewater," 13th Edition, pp. 210-215, 14th Edition,
pp. 144-147.
(6) Manganese — Atomic Absorption Method, "Methods for Chemical
Analysis of Water and Wastes," pp. 116-117, EPA, Office of Technology
Transfer, Washington, D.C. 20460, 1974, or "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, pp. 210-215, 14th
Edition, pp. 144-147.
(7) Odor - Consistent Series Method, "Methods for Chemical Analysis
of Water and Wastes," pp. 287-294, EPA, Office of Technology Transfer,
Washington, D.C. 20460, 1974, or "Standard Methods for the Examina-
tion of Water and Wastewater," 13th Edition, pp. 248-254, 14th Edition,
pp. 75-82.
(8) pH — Glass Electrode Method, "Methods for Chemical Analysis of
Water and Wastes," pp. 239-240, EPA, Office of Technology Transfer,
Washington, D.C. 20460, 1974, or "Standard Methods for the Examina-
tion of Water and Wastewater," 13th Edition, pp. 276-281, 14th Edition,
pp. 460-465.
(9) Sulfate - Turbidimetric Method, "Methods for Chemical Analysis
of Water and Wastes," pp. 277-278, EPA, Office of Technology Trans-
fer, Washington, D.C. 20460, 1974, or "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, pp. 334-335,
14th Edition, pp. 496-198.
(10) Total Dissolved Solids - Total Residue Methods, "Methods for
Chemical Analysis of Water and Wastes," pp. 270-271, EPA, Office
of Technology Transfer, Washington, D.C. 20460, 1974, or "Standard
Methods for the Examination of Water and Wastewater," 13th Edition,
pp. 288-290, 14th Edition, pp. 91-92.
-------
(11) Zinc — Atomic Absorption Method, "Methods for Chemical
Analysis of Water and Wastes," pp. 155-156, EPA, Office of Technology
Transfer, Washington, D.C. 20460, 1974, or "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, pp. 210-215, 14th
Edition, pp. 144-147.
Appendix A
Statement of Basis and Purpose for the National Secondary
Drinking Water Regulations
The following concepts and rationale were used in arriving at the specific
levels of the SMCLs and should enable those whose responsibility it is to
interpret or apply the regulations to do so with understanding, judgment
and discretion. The most recent reviews of aesthetic factors relating to
drinking water quality were prepared by Zoetemann in "Sensory Assess-
ment and the Chemical Composition of Drinking Water" and the National
Academy of Sciences (NAS) in its report, "Drinking Water and Health"
(1977). "Secondary Maximum Contaminant Levels" means SMCLs which
apply to public water systems and which, in the judgment of the Adminis-
trator, are requisite to protect the public welfare. The SMCL means the
maximum permissible level of a contaminant in water which is delivered to
the free flowing outlet of the ultimate user of a public water system.
Contaminants added to the water under circumstances controlled by the
user, except those resulting from corrosion of piping and plumbing caused
by water quality, are excluded from this definition.
Chlorides
The presence of too great a concentration of chloride ions in drinking
water can result in two undesirable effects. First, the consumer may
detect an objectionable taste in the water. Second, corrosion of the
pipes in hot water systems may occur. Existing evidence suggests that
consumers react to excessive amounts of chlorides by either treating the
water themselves, or by rejecting the water supply. Therefore, a limit has
been set for chloride ion concentration.
The literature contains a number of reports on the taste thresholds of
various salts. Whipple1, using a panel of 10 to 20 persons, found the
detectable range of concentration of various salts to be as shown in Table
1. Richter and MacLean2 studied the response of a larger panel to sodium
chloride in distilled water. Table 2 summarizes their results.
Lockhart, Tucker, and Merritt3 also studied the taste thresholds of the
ions in distilled water by studying their effects on the flavor of brewed
coffee. Using a triangular test with panels of 18 or more, they found
results that are summarized in Table 3. In a triangular taste test, the panel
members are asked to taste three samples. Two of the samples may contain
-------
either the salt being tested or distilled water, while the third is different
from the other two. The panel member is asked to identify the odd one.
Using this test procedure, the threshold concentration is defined as the
concentration at which the number of correct separations is 50% above
the chance of probability of one-third separations (i.e., when two-thirds
of the panel make the separations correctly).
Results shown in Table 1 and Table 3, which summarize the Whipple
and Lockhart et al. studies, are in agreement, even though different
methods were used for analysis. The Richter and MacLean study found
thresholds considerably below those of the other two studies. All three
studies support a secondary maximum contaminant level of 250 mg/1
for chloride.
Recently, Bruvold and his co-workers4' s' 6> 7 have studied the influ-
ence of mineralization on the taste of water. Much of their work has been
on the influence of total dissolved solids, but they noted that major taste
effects are produced by anions. Chloride produced an effect somewhere
between the milder sulfate and the stronger carbonate.6 ' 7
In addition to the influence of mineralization on the taste of water,
the deterioration of domestic plumbing and water heaters due to high
concentrations of Total Dissolved Solids (TDS may contain 50% chlorides)
should also be considered. According to Lawrence, the approximate life of
domestic water heaters for 200 mg/1 TDS water (100 mg/1 of chlorides)
ranged between 10 years and 13 years, but declining heater life as a
function of increasing TDS was fairly uniform: about 1 year shortened life
per 200 mg/1 additional TDS (100 mg/1 of chlorides)8 . Mineralization
can also affect other domestic plumbing and municipal waterworks
equipment.
Values for maximum chloride concentrations in drinking water set by
other agencies are as follows: World Health Organization's (WHO) Inter-
national Standards - 200 mg/19 ; WHO'S European Standards - 200
mg/110 ; USSR Standards - 500 mg/111 ; proposed European Commu-
nities (EC) Standards — 200 mg/112 ; and Food and Drug Administra-
tion's (FDA) Bottled Water Standards - 250 mg/113 .
Chloride cannot be readily removed from drinking water without the
use of distillation, reverse osmosis or electrodialysis which are effective
but relatively expensive; thus, the use of an alternate source or blending
is the most desirable approach.
Based upon the above experimental data and in agreement with the
standards presented above, a secondary maximum contaminant level of
250 mg/1 was determined to prevent most aesthetic effects. Consumers
may become accustomed to the taste of somewhat higher chloride
levels, but the economic effects of these higher levels ought to be
avoided.
-------
Table I — Range of Concentration of Various Salts Detected by Taste in Drinking
Water by Panel of 10 to 20 Persons
Concentration Detected — mg/1
Median
KC1 . . ,
NaCl
CaCl2
MgCl2
Sea Water 1 ...
Salt
. ... 525
300
250
500
Anion
250
182
160
372
300
Range
Salt
250-600
200^50
150-350
200-750
Anion
167-286
121-274
96-224
149-560
150-600
In terms of mg/1 chloride.
Source: Whipple, G.C., The Value of Pure Water. Wiley (1907).
Table 2 — Taste Threshold Concentrations of Panel of S3 A dulls for NaCl
Concentrations
NaCl
Difference from
distilled water
noted 160
Salt taste
identified 870
Mean
Cl
97
530
NaCl
100
650
Median
Cl
61
395
- mg/1
Range
NaCl Cl
70-600 42-364
200-2500 120-1215
Source: Richter, C.P. and MacLean, A., "Salt Taste Threshold of Humans," American
J. Physiol. Vol. 4, 126: 1-6, May 1939.
Table 3 — Taste Threshold Concentration of Salt and Ions in Water
Threshold Concentrations — mg/1
Salt Cation Anion
NaCl 345 135 210
KC1 650 340 310
CaCl2 347 125 222
Source: Lockhart, E.E., Tucker, C.L. and Merritt, M.C. The Effect of Water
Impurities on the Flavor of Brewed Coffee, Food Research, 20, 598-605 (1955).
-------
References
1, Whipple, G.C., "The Value of Pure Water," New York, New York,
John Wiley & Sons (1907).
2. Richter, G.O. and A. MacLean, "Salt Taste Threshold of Humans,"
Am. J. Physiol. 126: 1-6 (1939).
3. Lockhart, E.E., C.L. Tucker and M.C. Merritt, "The Effect of Water
Impurities on the Flavor of Brewed Coffee." Food Research, 20:
598-605(1955).
4. Bruvold, W.H. and H.J. Ongerth, "Taste Quality of Mineralized
Water," JAWWA, 61: 170-174 (April 1969).
5. Bruvold, W.H., H.J. Ongerth and R.C. Dillehay, Consumer Assess-
ment of Mineral Taste in Domestic Water, JAWWA 61: 575-580
(April 1969).
6. Bruvold, W.H. and W.R. Gaffey, "Evaluative Ratings of Mineral
Taste in Water," J. Perceptual and Motor Skills, 28: 179 (April
1969).
7. Bruvold, W.H. and W.R. Gaffey, "Rated Acceptability of Mineral
Taste in Water: II. Combinational Effects of Ions on Quality and
Action Tendency Ratings," J. Appl. Psychol. 53: 317(1969).
8. Lawrence, C.H., "I stimating Indirect Costs of Urban Water Use,"
J. of the Envir. Eng. Div., ASCE, pp. 517-533 (August 1975).
9. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Page 39.
10. "European Standards for Drinking Water," 2nd Edition, World
Health Organization. Geneva, 1970. Page 37.
11. Stuff en, D., "The Maximum Permissible Concentrations in the USSR
for Harmful Substances in Drinking Water," Toxicology.
Amsterdam, 1973. Page 190.
12. "Official Journal of the European Communities," Volume 18.
September 18, 1975. Page 9.
13. 21 CFR Part 103 - Quality Standards for Foods with no Identity
Standards - Bottled Water. Federal Register, Vol. 44, No. 45,
12172, March 6, 1979.
-------
Color
Experience has shown that if the water is highly colored, many people
will turn to alternative supplies that may be less safe, although the level
of this parameter is not known to be a measure of the safety of water.
However, color may be indicative of high organic chemical contamina-
tion, inadequate treatment, high disinfectant demand and the potential
for production of excess amounts of disinfectant by-products. At least
one State now maintains color among its primary standards.
Black and Christman1 analyzed natural color causing solids extracted
from a number of waters. They broadly classified these solids as aromatic,
polyhydroxy, methoxy and carboxylic acids. Natural color is also often
reported to be the result of fulvic and humic acid fractions in the water.
Although these acids are similar, Packham2 has collected data to show
that a true chemical difference exists between the fulvic and humic
fractions of color causing solids. Color in drinking water also results from
the presence of metals such as copper, iron and manganese. The entire
state-of-the-art and research needs in the area of color have been
extensively reviewed by the American Water Works Association's Research
Committee on Color Problems.3 ' 4
Color becomes objectionable to a considerable number of people at
levels over 15 color units (C.U.). At a level of 30 C.U., experts feel that
color will be detectable by all and aesthetically displeasing to most. How-
ever, a color level of 5 C.U. will be noted by many when viewed in a filled
bathtub, whereas a level of 3 C.U. will not be noticed.5 Rapid changes in
color levels may provoke more citizen complaints than a relatively high
constant color level and treatment plant operators should seek to prevent
or modify such changes in color levels where possible.
Color from industrial sources may present different problems in
measurement and handling than color from natural sources. Synthetic, as
well as natural colors, should be monitored following procedures given in
Standard Methods!"
Values for maximum color levels in drinking water set by other agencies
are as follows: WHO International Standards - 5 C.U.7; WHO European
Standards - 5 C.U.8 ; proposed EC Standards - 20 C.U.9 ; and FDA
Bottled Water Standards - 15 C.U.10.
Treatment for the removal of color from drinking water depends on
the nature of the colored material. Some color can be removed by con-
ventional treatment (oxidation, flocculation, filtration), but some types
of color may require activated carbon treatment.
Based upon the above data, a secondary maximum contaminant level
of 15 C.U. was determined to represent a color level which would prevent
the bulk of consumer complaints about colored water.
-------
References
1. Black, A.P. and R.F. Christman, "Characteristics of Colored Surface
Water," JAWWA 55: 753 (June 1963).
2. Packham, R.F., "Studies of Organic Color in Natural Water,"
Proc. Society for Water Treatment and Examination 13: 316
(1964).
3. Research Committee on Color Problems — Report for 1966,
JAWWA 59: 1023-1035 (August 1967).
4. Research Committee on Coagulation and Research Committee on
Color Problems, JAWWA 62: 311-314 (May 1970).
5. Quality Goals for Potable Water, A Statement Adopted by the
Board of Directors on January 28, 1968, JAWWA Yearbook
73-74: 60-63 (September 1973).
6. Standard Methods for the Examination of Water and Wastewater,
14th Edition: 64, American Public Health Association, New York,
New York (1975).
7. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Page 38.
8. "European Standards for Drinking Water," 2nd Edition, World
Health Organization. Geneva, 1970. Page 42.
9. "Official Journal of the European Communities," Volume 18.
September 18, 1975. Page 6.
10. 21 CFR Part 103 - Quality Standards for Foods with no Identity
Standards - Bottled Water. Federal Register, Vol. 44, No. 45,
12172, March 6, 1979.
-------
Copper
Copper imparts some taste to water and can cause staining of porcelain.
Physiologically, copper is important to human metabolism, and as such,
small amounts are generally regarded as non-toxic. However, large doses
may produce emesis, and prolonged oral administration may result in liver
damage. Therefore, for aesthetic and physiological reasons, it is desirable
to limit the concentration of copper in drinking water.
Copper is an essential and beneficial element in human metabolism, and
it is well known that a deficiency in copper results in nutritional anemia in
infants. The daily requirement for adults has been estimated to be 2.0
mg.4 Children of preschool age require about 0.1 mg of copper daily for
normal growth, and the average daily urinary excretion is in the order of
1.0 mg, with the remainder being eliminated in the feces. Because the
normal diet provides only a little more than is required, an additional
supplement from water would ensure an adequate intake. The distribution
of copper in the body is fairly uniform, except for the liver where it
appears to accumulate. The general health hazard from excess copper
intake at a level of a few milligrams per liter appears to be small, but a
few people are adversely affected by ingestion of even trace amounts of
copper. This disorder of copper metabolism is called Wilson's disease and
can be arrested by the use of chelating agents.2
The flurry of research during and following the acceptance of copper
and copper alloys as piping materials created a large amount of data on
the solubility and physiological effects of these metals, but relatively little
on the resulting taste of the water. Schneider quoted the Prussian Hygienic
Institute for Water, Soil, and Air in Berlin, which stated that it was
possible that a copper concentration of 3 to 5 mg/1 could affect the taste
of water adversely. Schneider suggested a few years later that the admis-
sable amount of copper in water be set at 5 mg/1, as a higher concentra-
tion would give the water a disagreeable taste. Spitta, critically reviewing
the literature on the effect of copper on human health, stated that 2 mg/1
of copper was a taste threshold concentration. Froboese, on the other
hand, reported that 1.5 mg/1 was the lowest concentration of copper
that normally could be tasted.1 Cohen found that individuals vary in the
acuity of their taste perception, and the threshold of taste for copper
varies from 1-5 mg/liter.3 Based on this research a copper concentration
of 1 mg/1 is reasonable to insure the absence of astringent taste effects.
Another reason for limiting the concentration of copper in water is
to avoid the staining of porcelain. Obrecht found that porcelain fixtures
can be stained blue or blue-green by low concentrations of copper in
water.5 The National Academy of Sciences2 (NAS) concluded that a
copper content of 0.5 mg/1 or less, in some soft waters, will cause stain-
ing of porcelain.
Green-stained drapes were also found in a Yale dormitory where the
water copper level was 4 mg/1. This same water was responsible for
imparting a green tint to the hair of silver blonde coeds. Determination
of copper levels in their hair were found to be 1,042 ppm as opposed to
normal values of 17-38 ppm.6
10
-------
Low copper concentrations can also cause black and white photo-
graphic prints to turn brown during processing. Furthermore, Obrecht
reported that copper concentrations greater than 1 mg/1 can produce
insoluble green "curds" by reacting with soap.5 In 1961, of 163 inter-
state supplies, 70 percent contained less than 0.2 mg/1. Of the 100
largest cities in the U.S., 94 contained less than 0.1 mg/1.4
Values for maximum copper concentrations in drinking water set
by other agencies are as follows: WHO International Standards —
0.05 mg/17 ; WHO European Standards - 0.05 mg/18 ; USSR Stand-
ards - 0.05 mg/19; proposed EC Standards - 0.05 mg/110; and FDA
Bottled Water Standards - 1 mg/1.11
Since copper in drinking water usually results from the action of
aggressive water on copper plumbing the copper content can be reduced
by stabilizing the water.
The SMCL for copper was set at 1.0 mg/1 because copper imparts a
detectable taste in water to some people at 1.0 mg/1. This is in agreement
with the Food and Drug Administration's value for copper at 1 mg/1.
References
1. Sollman, T.H., "A Manual of Pharmacology," Ed. 8, Philadelphia,
Pennsylvania, W.B. Saunders Co., pp. 1299-1302 (1957).
2. Drinking Water and Health, 1977. Report of the National Academy
of Sciences, Safe Drinking Water Committee. Pp. 308-309.
3. Cohen, J.M., L.J. Kamphake, E.K. Harris, and R.L. Woodward,
"Taste Threshold Concentration of Metals in Drinking Water,"
JAWWA 52: 660-670(1960).
4. Quality Goals for Potable Water, A Statement Adopted by the
Board of Directors on January 28, 1968, JAWWA Yearbook 73-74:
60-63 (September 1973).
5. Obrecht, M.F. and R.F. Myers, "Potable Water Systems in Build-
ings: A Treatsie on Corrosion and Deposit Control." Reinhold
Publishing Co., 1975, pp. 6-7.
6. Medical World News, p. 49, February 20, 1978.
7. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Page 40.
8. "European Standards for Drinking Water," 2nd Edition, World
Health Organization. Geneva, 1970. Page 37.
9. Stoffen, D., "The Maximum Permissible Concentrations in the
USSR for Harmful Substances in Drinking Water," Toxicology,
Amsterdam, 1973. Page 190.
10. "Official Journal of the European Communities," Volume 18.
September 18, 1975. Page 9.
11. 21 CFR Part 103 - Quality Standards for Foods with no Identity
Standards - Bottled Water. Federal Register, Vol. 44, No. 45,
12172, March 6, 1979.
11
-------
Corrosivity
Corrosion is a very significant concern, not only affecting the aesthetic
quality of the water, but also having serious economic impact and health
implications. Corrosion products containing materials such as lead and
cadmium have been associated with serious risks to the health of con-
sumers of drinking water. A number of indices (such as the Langelier
Index) are presently available for measuring the corrosivity of drinking
water but a single universal index applicable to all situations is not yet
generally available. For the present, the secondary regulations state that
the water should be "non-corrosive," as determined by the State.
It has been estimated by Hudson and Gilcreas that the annual loss from
water corrosiveness is about $375 million.1 In addition to the loss of
materials used to convey water, water leakage from deteriorating dis-
tribution systems can be substantial. In some instances, as much as 25
percent of the water leaving a treatment plant is lost before reaching
the consumer.
The concentration of trace metals in water collected in distribution
systems or at household taps is more relevant than raw water with respect
to water quality. Corrosion in a distribution system may add trace metals
to finished water before it reaches the consumer.
Corrosivity is a complex characteristic of water primarily related to
pH, alkalinity, dissolved oxygen, total dissolved solids, and hardness.2
Soft water tends to be more corrosive than hard water. As a result, certain
trace metals are found in higher concentrations in soft water. Because of
localized conditions, important variations in water quality will occur as
a result of the corrosion of pipe materials as water passes through the
distribution system. Water velocity can also play an important role in the
erosion and corrosion of pipe materials.2 ' 3 In addition, stationary water
in contact with pipes can have a significant effect on the dissolution of
distribution system materials. Furthermore, seasonal changes affecting
water quality will result in variations in the concentrations of dissolved
pipe materials at the tap.
Common piping materials used in distribution systems are iron,
steel, cement (reinforced concrete), asbestos-cement, and plastic. Lead,
copper, zinc, and alloys such as brass, bronze and stainless steel may also
be used in addition to ferrous metals in pumps, small pipes, valves and
other appurtenances. Trace metals and asbestos may be contributed to
the water by corrosion products or simply by solution of small amounts
of metals with which the water comes into contact.
Corrosion in distribution systems will result in the increase in levels of
toxic materials such as lead and cadmium. Products of corrosion present
in the water can also cause turbid waters and promote deposits under
stagnant conditions, thereby encouraging bacterial growth. The health
related aspects of corrosion are dealt with in the primary drinking water
regulations.4 '5 In addition to adverse health implications, corrosion will
also effect the taste, color and the aesthetic acceptability of the water.
The most commonly used materials in distribution systems whose
corrosion products adversely affect the aesthetic quality of waters are iron
12
-------
and copper. Ferrous materials are used for distribution mains as well as
for service lines and household plumbing. The use of copper for trans-
porting water is limited to service lines and household plumbing.
Corrosion of iron in distribution systems will result in the deposition
of ferric iron in the mains and pipes. Loosening of the deposits will cause
red water at the tap, which in turn will impart color and taste to the water
and will cause the staining of fixtures and laundry. A survey of 534
municipal water supply systems in 1969 revealed that 281 (52.6%) have
had red water problems.6
Another serious problem encountered with the corrosion of iron is
tuberculation. Tuberculation of the interior surface of pipes caused by
corrosion will result in a loss of the carrying capacity of the water through
the distribution system.2 Resistance to flow is caused by the roughening
of the interior surfaces of the pipes by the tubercules and by the reduction
of the inner diameter of the pipe due to the deposition of the corrosion
products. In order to overcome the loss of the carrying capacity, higher
pressures have to be generated at the pumping stations, which in turn
require additional energy. It has been shown by Hudson (1966) that the
cost of transporting the water increases as the 1.85th power of the loss
in efficiency of the pipe line, and the physical value of the installation
decreases almost directly with the loss of the carrying capacity.7
Deposits of corrosion products from iron may also serve as a breeding
ground for a number of life forms including bacteria, nematodes, small
crustaceans and protozoa. In addition to the health hazards associated
with these organisms, the presence of organic matter and the anaerobic
conditions prevailing in tubercules and deposits, some of the bacteria
present will inevitably reduce sulfates to hydrogen sulfide and other
odorous products rendering the water aesthetically unacceptable.8 In
addition, the sulfate reducing organisms will aid the acceleration of
corrosion. It has been shown that every milligram of sulfate reduced
will cause the corrosion of 2 milligrams of iron.9
The effect of water quality on the corrosivity of iron, steel and
galvanized steel has been categorized by Obrecht, et al., as shown in
Table 1.
13
-------
Table 1 — Potable water categorization is applicable to all metals for scaling but only to iron, steel
(not stainless), and galvanized steel for corrosivity1 (Obrecht, et al., 1975).'
Category
1A
IB
1C
2A
2B
2C
2D
3A
3B
3C
3D
3E
4A
4B
4C
Calcium
(Ca), ppm1
0-18
0-18
0-18
18-35
18-35
18-35
18-35
35-75
35-75
35-75
35-75
35-75
>75
>75
>75
Sulfate
(804), ppm
As found
0-25
0-60
> Ca but
not < 25
0-25
1% Ca
< P/2 Ca
As found
> 1% Ca
but < 3 Ca
<2Ca
>2Ca
<3Ca
Silica
(Si02), ppm
0-15
0-15
>15
0-15*
0-15
>15
As found
0-15
0-15
>15
As found
>30
0-30
0-30
>30
Dissolved
oxygen, ppm
1-10
0-1
1-5
1-10
1-10
1-8
0-1
1-10
1-10
1-10
0-1
1-10
1-10
1-10
1-10
Character
Extreme corrosion hot and cold.
Moderate corrosion hot and cold.
Extreme corrosion with
CO2 > 8 ppm.
Slight corrosion cold, considerable
hot. Aggressiveness reduced and
perhaps not troublesome due to
natural Si02 .
Considerable corrosion hot,
moderate cold. May be slightly
scale forming very hot.
Moderate to slight corrosion hot
and cold. May be scale forming
hot.
Corrosion unlikely. May be scale
forming hot.
Corrosion unlikely. May be scale
forming hot.
Moderate corrosion hot, slight
cold. Considerable scale
formation hot.
Considerable corrosion hot, slight
cold. Considerable scale
formation hot.
Considerable scale formation.
Slight corrosion hot.
Considerable scale formation.
Corrosion unlikely.
Corrosion unlikely hot and cold.
Excessive scale formation.
Excessive scale formation.
Corrosion unlikely to slight cold,
slight to moderate hot.
Excessive scale formation.
Galvanic corrosion considerable
hot and cold.
Excessive scale formation.
Corrosion unlikely.
•With SiO2 over 15 ppm, corrosion may be reduced in proportion to SiC>2 content.
NOTES: Presence of chlorides in concentrations greater than 100 ppm, with high sulfates, renders a
water more corrosive than indicated by category above. Presence of carbon dioxide in
concentrations exceeding 5 ppm accelerates corrosion processes where category groups
indicate corrosion. In concentrations exceeding 20 ppm, it may cause an indicated non-
corrosive water to be corrosive.
TERMS: Extreme or excessive — where effects necessitate immediate corrective action.
Considerable —where corrective action is desirable.
Moderate — where corrective action is questionable and depends on economy affected.
Slight — where effect is too slight to warrant corrective action.
Unlikely — where effects are possible but not probable.
Source: Obrecht, Malvern F. and Myers, James R. "Potable Water Systems in Buildings: A Treatsie on
Corrosion and Deposit Control, Materials Selection, System Design and Operation."
Reinhold Publishing Co., 1975.
14
-------
The presence of carbonate minerals will inhibit the corrosion of iron.
In soft waters containing no calcium, this inhibition is the greatest at pH
values ranging from 6.5 to 7.0 for mild steel, and 7.0 for cast iron provided
the alkalinity is five or ten times above the chloride and sulfate salt
concentration levels.2 Studies by Larson indicated that in soft waters
tuberculation of cast iron increased when the pH was increased within a
range of 7.5 to 9.O.2 The presence of calcium will also inhibit the corro-
sion of iron. It has been shown that in the presence of calcium com-
bined with a minimum alkalinity of 50 to 100 mg/1, the life of mild
steel will be extended considerably. In general, the greater the concentra-
tions of calcium, the greater the protective action of the water. Larson also
concluded that the combination of high alkalinity and calcium with the
inherent low pH will prevent corrosion more easily than low alkalinity
and calcium combined with high pH values.2
Corrosion products of copper also will adversely affect the aesthetic
quality of water. The presence of excessive amounts of copper in water
causes staining of fixtures,3 imparts taste, causes soap curds3 and in
extreme instances, imparts green color to the hair of individuals who
wash in such waters.1 °
The corrosion of copper relative to water quality has been well docu-
mented.2 ' 3 Results of a study by Obrecht, et al., relating to the corrosion
of copper pipes indicated that pitting of copper will occur when in contact
with cold water having a pH of less than 8.5, relatively low hardness, high
carbon dioxide and oxygen content and a saturation index of slightly
positive to strongly negative.3 High concentrations of sulfates, chlorides,
bicarbonates and total dissolved solids in the presence of dissolved oxygen
and free carbon dioxide concentrations in excess of 10 mg/1 will accelerate
the corrosion of copper.
In a study evaluating the effect of various California waters on copper
pipes, Cruse and Pomeroy11 reported that in the majority of instances
no pitting of copper occurred when copper was in contact with well
waters having carbon dioxide levels of 5-10 mg/1. Table 2 summarizes
the effect of various qualities of California waters on the corrosion of
copper.
In Washington County, Maryland, 669 copper determinations at the
tap were made in both public and private water systems.12 Efforts were
made to sample running water, rather than standing water in contact with
plumbing systems overnight, because the latter has been shown to have
greatly increased trace-metal concentrations in systems without corrosion
control. The correlation coefficients of copper concentration with pH,
with hardness, and with conductivity, were -0.369, -0.162-0.173, respec-
tively; all were significant (p=0.01).
Velocity of the water can also significantly affect the rate of corrosion
in copper pipes. Stagnant water conditions will promote concentration
cell corrosion of metals and alloys, including copper in water distribution
systems. Copper is also susceptible to erosion corrosion when exposed
to high velocity and high temperature waters. Erosion corrosion of
copper becomes a significant problem when the copper is in contact
with soft waters moving at a rate over 4 feet per second, having a
dissolved oxygen level exceeding 2 ppm and a carbon dioxide content
over 10 ppm.
15
-------
3|rS-
P,5 e
O ®
oS,
o
u
Ot-tNfNOvoOvOOXOCNON'-'
O in -H
VO
O\ -H
CS 01
,1
6
.3
OO ON
--. m
r- ts
CN cs
»-t ^" C4 O
Qi fA VO vo
O. _ _ ^
4)
M
00 "o "^ Tf
o
o
-H "O o\ vo OO
*—i % vo CO -*3-
B5
o
t-- ts r-
-H fS •-<
—I CN -H
" 5
"3
U
vo in
cs vo
O
r-'
fN CO
oo oo
E
o
•o
B
CO
O _
°. 1
"S "a
« .5
I «
« E
E C
° --B
^ S«
II 1
*• ? c
V 4> M
be feu *
04 « -H
53 S S
33£
s
a.
o 2
6 i S
.2 ? S
•C b* U
^ « fe
c .5 o.
'p o! o
,?|i
s!
11
o e
"05
'o ^ (2
bO £3 CA
-SI!
o
U
» S1 •*• ° J3 „
£ > C » ^ M 3
u 5 3
" « 5
T3 S
B
p ii
.OOT :
-------
Copper-zinc alloys are also subject to dezincification which consists
of the selective removal of zinc. Copper alloys such as muntz metal and
yellow brass are particularly subject to dezincification when expsoed to
high temperature soft waters containing free carbon dioxide and a
high chloride content. High pH levels (> 9.0 to 9.6) will also promote
dezincification.2 Lower zinc alloys usually are subjected to localized
(plug type) dezincification when exposed to neutral, alkaline or
slightly acidic solutions. On the other hand, higher content zinc alloys
will be subject to uniform dezincifications.3
Corrosion protection of water distribution systems, in general, may be
achieved by a number of methods including pH adjustment, addition of
lime, increasing alkalinity, and addition of phosphates or silicates.2
The extent and type of treatment selected is dependent upon the char-
acteristics of the water and the compatibility of treatment with regard
to the various materials used to convey the water through the distribu-
tion system. It should be noted that corrosion control aimed at the
reduction of contamination from one source may or may not reduce
contamination originating from other sources as well. Experience indi-
cates that among the previously mentioned treatment techniques corro-
sion control through the deposition of a protective calcium carbonate
film on the interior surfaces of the conveyors of the water may be the
most effective approach to non-selectively provide protection to a
number of materials such as asbestos cement, lead, iron, galvanized
steel, copper and various alloys that make up the distribution system.
This method is widely applicable and extensively used.1 s
The AWWA indicates that the maintenance of calcium carbonate
stability is the most effective method of preventing corrosive action on
iron mains. AWWA recommends 80 mg/1 hardness as CaCO$ is best for a
balance between scaling and corrosion characteristics, although a goal of
90 to 100 mg/1 for some supplies may be deemed desirable.1
Merrill and Shanks13 (1977) suggested that for best results for corro-
sion control, the alkalinity of the water should be maintained at least at
40 mg/1 as CaCC>3 while maintaining a slight oversaturation of the water
by 4-10 mg/1 CaCO3 within a pH range of 6.8 to 7.3.
Several investigators17'18) 19' 20 have reported that the addition of
polyphosphates or silicates will stabilize the iron and prevent staining and
red water problems. O'Connor9, however, points out that heat causes
polyphosphate to revert to orthophosphate, thereby destroying its
sequestering ability. Further, this treatment technique should not be
used when the original iron content exceeds 1 mg/1. If iron rust deposits
are present in the distribution system, these can be loosened and dispersed
in the water by sequestrants and create severe problems of rusty water.
Hatch21 and Moore .and Smith22 indicated that polyphosphates can also
introduce lead into stagnant water from lead pipes, particularly at pH
levels above 7 or 8. Larson19 notes that polyphosphate additions may
cause excessive bacterial growth in the distribution system, thereby
creating a need for increased disinfectant residuals.
There are a number of alternatives available for controlling the
corrosion of copper including the removal of dissolved oxygen by
17
-------
deaeration, decreasing the carbon dioxide by pH adjustment, addition
of inhibitors such as polyphosphates or silicates and maintenance of
flow rates (below 4 feet per second) within the distribution system.17
Table 3 presents a number of alternate treatment approaches for control
of corrosion.
The cost of controlling corrosion represents a small increment in cost
to a utility compared with significant benefit in (1) reduction of con-
taminants at the consumer's tap, (2) cost savings due to extending the
useful life of distribution system materials, (3) energy savings in trans-
porting the water due to the reduction of frictional losses, and (4)
reduction of water losses through leakage or breakage.
The additional costs incurred by the implementation of a corrosion
control program at a particular facility will be determined by the size of
the facility, the particular treatment selected, the quality of the water
distributed, increased monitoring requirements and additional workload
of the staff.
Hudson and Gilcreas1 estimated that the national annual costs of
stabilizing corrosive waters would be approximately $27 million which in
turn would result in an estimated economic savings of $375 million
annually. The following information was used in deriving this estimate: 1)
a total population of 180 million persons is served by community water
systems; 2) investment in community water-distribution systems comprises
about 60 percent of the capital cost of water utilities; 3) replacement cost
of community systems is estimated to be $125 billion; sixty percent of the
total is $75 billion; 4) even with stable waters, distribution capacity of
new piping declines with age at a rate of 1 percent/year and with unstable
waters, the decline commonly doubles, thus producing a further loss of 1
percent per year; and 5) half of the water supply systems have problems
with corrosive water.
A specific MCL for corrosivity is not established at this time. Instead,
the secondary regulations presently state that drinking water should be
"non-corrosive." A non-specific corrosivity standard is warranted under
the NSDWR because corrosive waters may adversely effect the aesthetic
quality of drinking water. However, the existence of corrosive waters is
left to be determined on a case-by-case basis through the exercise of
judgment by the States in implementing the secondary regulations.
References
1. Hudson, H.E., Jr. and F.W. Gilcreas, 1976. "Health and Economic
Aspects of Water Hardness and Corrosiveness." Journal of the
American Water Works Association (April): 201-204.
2. Larson, T.E. "Corrosion by Domestic Waters." Illinois State Water
Survey. Urbana. Bulletin, 59, 1975.
3. Obrecht, Malvern F. and Myers, James R. "Potable Water Systems in
Buildings: A Treatise on Corrosion and Deposit Control, Materials
Selection, System Design and Operation." Reinhold Publishing Co.,
1975.
18
-------
•osion Control
a
£
£
K
^
(j
•5-
|
^
I
**")
1
K
Comments/ Problems
Effectiveness
g
_o
A
_o
"3.
o.
<
+-*
c
(U
1
H
c
o
a
O
g
0
Best overall treatment approach. Over-
saturation may cause calcium deposits.
AWWA recommends 80 mg/1 hardness
as best.
Most effective in water
with low pH and hardness.
Excellent protection for
copper, lead and asbestos
cement pipe in stabilized
waters. Good protection
for galvanized and steel
pipe.
%%
la-
s 12 H »
J^ J
Should not be used to stabilize waters
without the presence of adequate
alkalinity and hardness. Will cause
dezincification of copper alloys at pH
9.0 to 9.6. May cause tuberculation in
iron pipes at pH 7.5 to 9.0.
Most effective in waters
with sufficient hardness
and alkalinity to stabilize
water. May provide ade-
quate protection against
lead corrosion in low
alkalinity soft waters.
ffi
&(
%
a
§
u
•a
«tH
H
o
•S
>•»
£
s
1
o
CO
Combination of high alkalinity and
hardness with low pH is more effective
than combination of high pH with low
hardness and alkalirity.
Most effective in water
with low pH and sufficient
hardness. Excellent pro-
tection for lead corrosion
in soft waters at pH 8.3.
>>
.«n
G
•si
"i ^
" 2
ox>
tl
^5
g E
II
o o
coco
S.fe*
May cause leaching of lead in stagnant
waters. May encourage the growth of alg
and microorganisms. May cause red wat<
if iron content is above 1 mg/1. May not
effective at low pH levels.
Effective at pH levels above
7.0. Good protection for
asbestos cement pipe.
Addition of lime may
increase effectiveness of
treatment for copper, steel,
lead and asbestos.
Cfl
_
ca o ft
S « •"
H *? ft
0 2 C
b ft o
6
*.,
d £
o ta
•a £
It
"d ^3
HH ft
May increase the potential of pitting in
copper and steel pipes. May not be com-
patible with some industrial processes.
Most effective in waters
having low hardness and
pH below 8. 4. Good
protection for copper,
galvanized and steel pipe.
0
£« -
o°S
« s a
OS'S
••Cl*-1 fe-
ed o a
c > w
C *2 o
o o O<
PH£ ft
5
'S
§„
•j3
-------
4. 40 CFR Part 141, National Interim Primary Drinking Water Regula-
tion Amendments, July 19, 1979.
5. "Statement of Basis and Purpose for Special Monitoring Require-
ments and Proposed Amendments to the National Interim Primary
Drinking Water Regulations," July 1979, Office of Drinking Water,
Criteria and Standards Division, EPA.
6. Schweitzer, G.W., "Zinc Glassy Phosphate Inhibitors in Potable
Waters." Presented by the Education Committee of AWWA and U.S.
Public Health Service, June 20, 1970, Washington, D.C.
7. Hudson, William D., "Studies of Distribution System Capacity in
Seven Cities." Journal of the AWWA, (February 1966): 157-163.
8. Victoreen, M.T., "Control of Water Quality in Transmission and
Distribution Mains." Journal of the AWWA (December 1974): 369-
370.
9. O'Connor, J.T., Hash, L. and Edwards, A.B., "Deterioration of
Water Quality in Distribution Systems." Journal of the AWWA
(March 1975): 113-116.
10. Medical World News, p. 49, February 20, 1978.
11. Cruse, H. and R.D. Pomeroy, "Corrosion of Copper Pipes." Journal
of the AWWA (August 1974): 479483.
12. Oliver, S., "Mood and Trace Metals in Drinking Water." 1974.
Master of Science Thesis, The Johns Hopkins University, School of
Hygiene and Public Health, Baltimore, Maryland.
13. Merrill, D.T. and Sanks, 1977. "Corrosion Control by Deposition of
CaCOs Films. Part 1, A Practical Approach for Plant Operators."
Journal of AWWA, November 1977.
14. AWWA Quality Goals for Potable Water, Adopted by the Board of
Directors on January 28, 1968.
15. Nelson, J.A. and Kingery, F.J., "One Utility's Approach to Solving
Copper Corrosion." Proceedings, AWWA Seminar, June 25, 1978.
16. Quality Goals for Potable Water, A Statement Adopted by the Board
of Directors on January 28, 1968.
17. Aulenbach, D.B., "Determining Phosphate Additive for Iron Control
in Water," JAWWA <5J: 197-198 (March 1971).
18. Dart, J.F. and P.D. Foley, "Preventing Iron Deposition with Sodium
Silicate," JAWWA 62: 663-668 (October 1970).
19. Larson, T.E., "Evaluation of the Use of Polyphosphates in the Water
Industry," JAWWA 49: 1581-1586 (December 1957).
20. O'Connor, J.T., "Iron and Manganese, Chapter 11: Water Quality
and Treatment," American Water Works Association, McGraw-Hill
Book Company, New York, New York (1971).
21. Hatch, G.B., "Inhibition of Lead Corrosion with Sodium Hexameta-
phosphate," JAWWA 33: 1179-1187 (July 1941).
22. Moore, E.W. and F.E. Smith, "Effect of Sodium Hexametaphos-
phate on the Solution of Lead," JAWWA 34 : 1415-1424 (September
1942).
20
-------
Foaming Agents
Many substances in water will cause foam when the water is agitated or air
is entrained as in a faucet. Foaming is an undesirable property of drinking
water because it is aesthetically displeasing in that such agents impart an
unpleasant taste, cause frothing and consumers often associate foaming
with contamination. Because no convenient standardized foamability test
exists and because surfactants are one major class of substances that cause
foaming, this property is determined indirectly by measuring the anionic
surfactant concentration in the water.
Surfactants are synthetic organic chemicals having high residual affinity
at one end of their molecule and low residual affinity at the other. Its
vigorous surface activity justifies not only its name but its use as a princi-
pal ingredient of modern household detergents. Surfactants may be di-
vided into two broad chemical classifications, ionic and non-ionic. Ionic
types may be either anionic (-) or catonic (+). Alkyl benzene sulfonate
(ABS) and linear alkyl benzene sulfonate (LAS) are typical anionic surfac-
tants.
Contamination of drinking water supplies by surfactants results from
their disposal, as household and industrial wastes, into sources of raw
water. Such contamination is appearing in both surface and ground water
sources. Other potential sources of human intake of surfactants are food
and inadequately rinsed cooking and household utensils and dinnerware.
More than 75 percent of the surfactants in household detergents are of
the anionic type. Before 1965, alkyl aryl sulfonates accounted for almost
three-quarters of these, the remainder being mostly alkyl sulfates. Next in
extent of such use were the non-ionics, the cationics making up only a
small percentage.1 In recent years, the requirement for biodegradability
led to widespread use of linear alkyl benzene sulfonate. Hence, the anionic
group of surfactants is most apt to be present in raw water supplies.2 It is
largely for this reason that the degree of detergent contamination is estab-
lished currently in terms of the concentration of anionic surfactants. While
the currently-employed method for determination of the concentration of
anionic surfactants (the methylene-blue method) measures more than
anionic surfactants (most of the interferences are positive, some of which
are foaming agents), this is not a problem at the levels usually encountered
in drinking water.
Concentrations of anionic surfactants found in drinking waters range
from 0 to 2.6 mg/1 in well water supplies and from 0 to 5 mg/1 in river
water supplies. In one instance, a municipal water supply containing 5
mg/1 during a period of drought necessitated use of an impounded, highly
purified sewage treatment plant effluent as a raw water supply.3
In a study of the taste effects of surfactants4 , ten percent of the indi-
viduals using water containing less than 1 mg/1 anionic sulfonate deter-
gents complained of an off-taste, whereas all those using water containing
1.5 mg/1 complained of an off-taste. Frothing was also a common com-
plaint, occurring most frequently at concentrations of 1 mg/1 and above.
The off-taste has been described as oily, fishy, or perfume-like.4 ABS and
21
-------
LAS themselves are essentially odorless. The odor and taste characteristics
are likely to arise from the degradation of waste products rather than the
detergents. The concentration of ABS or LAS in municipal sewage averages
10 mg/1. Each mg/1 of ABS or LAS present is an indication that 10 percent
of the water in which the surfactant is detected is of sewage origin. There-
fore, water containing the average concentration of 10 mg/1 surfactant
would likely be entirely of sewage origin.
From the basic toxicologic point of view, there are two reports which
are especially pertinent to the present consideration:
1. The Toxicologic Subcommittee of the Food Protection Committee of
the Food and Nutrition Board, National Research Council, published a
comprehensive report in 1956s bearing on the question of surfactants in
food. Reviewing extensively the acute and chronic toxic studies which
have been reported on these chemicals, they found that there appears to
be little specific relationship of toxicity to surface activity (reduction of
interfacial tension). In conclusion it was stated that:
a. There are no toxic effects common to all surfactants.
b. Surface activity per se is not a measure of toxicity.
c. The safety of each surfactant used in food must be determined
separately.
The report pointed out that surfactants may occur fortuitously in some
foods in amounts of a few parts per million and that: "It appears probable
that the interfacial tension existing in the digestive tract of a healthy
human is so low that it will not be further lowered by the small amounts
of synthetic surfactants which may be present in food."
2. In a report on an investigation dealing with the chronic and subacute
toxicity to rats of several surf ace-active agents, among which was sodium
alkyl aryl sulfonate, Fitzhugh and Nelson6 declared that: "The loxic
effects of the surface-active agents studied in the experiments were pro-
duced by irritation of the gastrointestinal tract (10,000 ppm or more in
the diet). To an extent which depended on the concentration of the
surface-active agents in the diet, this irritation prevented proper nutrition.
In severe cases of irritation, death resulted."
An ABS or LAS concentration of 0.5 mg/1 in drinking water, assuming
a daily adult human intake of 2 liters, would give a safety factor of
15,000, calculated on the results of subacute5 and 2-year7 tests on rats fed
diets containing ABS. In these rat studies, it was found that levels of ABS
in the diet of 0.5 percent and below produced no discernible physiological,
biochemical or pathological deviations from normal.
Human experience (6 subjects) with oral doses of purified ABS of 100
mg (equivalent to 2 liters of water containing 50 mg ABS/1) daily for 4
months led to no significant evidence of intolerance.8 LAS has foaming
characteristics similar to branched alkyl benzene sulfonate,9 and it can be
assumed that other properties are also similar.
Values for maximum anionic detergent concentrations in drinking
water set by other agencies are as follows: World Health Organization's
(WHO) International Standards - 1.0 mg/110; WHO's European Standards
22
-------
— 0.2 mg/111; proposed European Communities (EC) standards — 0.1
mg/112.
Foaming substances can be removed by conventional treatment or by
activated carbon. However, since the presence of foaming substances in
drinking water is frequently an indicator of contamination by sewage,
elimination of the source of contamination is the most appropriate treat-
ment approach.
The secondary maximum contaminant level of 0.5 mg/1 was based upon
levels that would prevent the occurrence of visible foam as discussed
above. Because of the association of other types of pollution with foaming
substances, the appearance of visible foam should be immediately investi-
gated and the source located and eliminated.
References
1. Schwartz, A.M., J.W. Perry and J. Berch, "Surface Active Agents and
Detergents," Vol. II, New York, New York, Interscience Publishers,
Inc. (1958).
2. Task Group Report, "Determination of Synthetic Detergent Content
of Raw Water Supplies," JAWWA 50: 1343-1352 (October 1958).
3. Metzler, D.F., R.L. Culp, H.A. Stollenburg, R.L. Woodward, G.
Walton, S.L. Chang, W.A. Clark, C.M. Palmec, and P.M. Middleton,
"Emergency Use of Reclaimed Water for Potable Supply at Chanute,
Kansas," JAWWA 50: 1021-1051 (August 1958).
4. Flynn, J.M., A. Andreoli, and A.A. Guerrera, "Study of Synthetic
Detergents in Ground Water," JAWWA 50: 1551-1562 (December
1958).
5. Food Protection Committee, "The Relation of Surface Activity to
the Safety of Surfactants in Foods." National Academy of Sciences,
National Research Council, Pub. 463, Washington, D.C. (1956).
6. Fitzhugh O .G. and A .A. Nelson, "Chronic Oral Toxicities of Surface
Active Agents," J. Am. Pharm. A. (Sc. Ed.) 37: 29-32 (1948).
7. Tusing, T.W., O.E. Painter and D.L. Opdyke, "Chronic Toxicity of
Sodium Alkylbenzene Sulfonate by Food and Water Administration
to Rats," Toxicol. Appl. Pharm. 2: 464473 (1960).
8. Freeman, S.( M.W. Burrill, T.W. Li, and A.C. Ivy, "The Enzymes In-
hibitory Action of an Alkyl Aryl Sulfonate and Studies on its Tox-
icity When Ingested by Rats, Dogs, and Humans," Gastroenterology
4: 332-343 (1945).
9. Rubinfield, J., E.M. Emerby and H.D. Cross, III, "Straight-chain
Alkylbenzenes: Structure and Performance Property Relations,"
The J. of the Am. Oil Chem. Soc., 41: 822-826 (December 1964).
10. "International Standards for Drinking Water," 3rd Edition, World
Health Organization, Geneva, 1971. Page 38.
11. "European Standards for Drinking Water," 2nd Edition, World
Health Organization, Geneva, 1970. Page 39.
12. "Official Journal of the European Communities," Volume 18.
September 18, 1975. Page 9.
13. Task Group Report, "Effects of Synthetic Detergents on Water
Supplies," JAWWA 49: 1355-1358 (October 1957).
23
-------
Iron
Iron and manganese have similar adverse environmental effects and fre-
quently occur together in natural waters. At one time, they were both
covered by a single recommended limit. However, the 1962 Public Health
Service (PHS) Standards1 recommended separate limits for iron and man-
ganese to reflect more accurately the levels at which adverse effects occur
for each.
Iron is a highly objectionable constituent in water supplies for either
domestic or industrial use. The domestic consumer complains of the
brownish color that iron imparts to laundered goods and plumbing fix-
tures. Iron also appreciably affects the taste of beverages.2 Taste thresh-
olds in drinking water are considerably higher than the levels which pro-
duce staining effects.3
When the concentration of iron exceeds 0.05 mg/1, some color may
develop, staining of fixtures may occur and precipitates may form.4 The
magnitude of such effects is directly proportional to the concentration of
iron in water. Experience shows that concentrations somewhat higher than
0.05 mg/1 may be generally tolerated. With certain exceptions, concentra-
tions up to 0.5 mg/1 may often be tolerated for industrial uses.2
The taste that iron imparts to water may be described as bitter or
astringent. Individuals vary in the acuity of taste perception, and it is dif-
ficult to establish a level that would not be detectable for the majority of
the population. In one study5 40% of a panel detected the taste of iron at
1.0 mg/1 in spring water but its authors mentioned that an earlier re-
searcher claimed 0.1 mg/1 as the taste threshold for ferrous iron.
The daily nutritional requirement is 1 to 2 mg, but intake of larger
quantities is required as a result of poor absorption. Diets contain 7 to 36
mg per day and average 16 mg.6 The amount of iron permitted in water to
minimize objectionable taste or staining effects (as much as 0.3 mg/1)
constitutes only a small fraction of the amount normally consumed and
does not have toxicologic significance.
Several investigators 3' 7'8'9 have reported that the addition of poly-
phosphates or silicates will stabilize the iron and prevent staining and red
water problems. O'Connor9, however, points out that heat causes poly-
phosphate to revert to orthophosphate, thereby destroying its sequestering
ability. Further, he cautions against use of this treatment technique when
the original iron content exceeds 1 mg/1. If iron rust deposits are present in
the distribution system, these can be loosened and dispersed in the water
by sequestrants and create severe problems of rusty water. Hatch10 and
Moore and Smith11 indicated that polyphosphates can also introduce lead
into stagnant water from lead pipes, particularly at pH levels above 7 or 8.
Larson8 notes that polyphosphate additions may cause excessive bacterial
growths in the distribution system, thereby creating a need for increased
disinfectant residuals.
Other maximum contaminant levels set for iron are as follows: WHO
International Standards — 0.1 mg/112 ; WHO European Standards — 0.1
mg/113 ; USSR Standards - 0.5 mg/114 ; proposed EC Standards - 0.3
mg/115; and FDA Bottled Water Standards - 0.3 mg/116 .
24
-------
Iron can be removed by oxidation and conventional treatment and by
an ion-exchange method specifically designed for iron removal. Utilization
of the "manganese greensand" filter using potassium permanganate to oxi-
dize the iron is also an effective method to remove iron. When the source
of iron is the corrosion of pipes and other water-contact surfaces, stabiliza-
tion or other appropriate water treatment processes can be employed to
minimize introduction of iron into the water.
The secondary maximum contaminant level of 0.3 mg/1 was determined
to represent a reasonable level at which taste effects would be minimized
in addition to minimizing the staining effects of iron. While the aesthetic
effect of levels somewhat higher than 0.3 mg/1 can be overcome by seques-
tration, there are drawbacks to the employment of this technique, as
noted above.
References
1. Public Health Service Drinking Water Standards, 1962, U.S. Depart-
ment of Health, Education and Welfare, PBS Pub. No. 956. Pages
42-43.
2. Riddick, R.M., H.L. Lindsay and A. Tomassi, "Iron and Manganese
in Water Supplies" JAWWA 50: 688-696 (May 1958).
3. Aulenbach, D .B., "Determining Phosphate Additive for Iron Control
in Water," JAWWA 63: 197-198 (March 1971).
4. Quality Goals for Potable Water, A Statement Adopted by the Board
of Directors on January 28, 1968, JAWWA Yearbook 73-74: 60-63
(September 1973).
5. Cohen, J.M., L.J. Kamphake, E.K. Harris and R.L. Woodward,
"Taste Threshold Concentrations of Metals in Drinking Water,"
JAWWA 52: 660-670 (May 1960).
6. Sollman, T.H., "A Manual of Pharmacology," Ed. 8, W.B. Saunders
Co.Philadelphia, Pennsylvania, pp. 1247-1267 (1957).
7. Dart, J.F. and P.O. Foley, "Preventing Iron Deposition with Sodium
Silicate," JAWWA 62: 663-668 (October 1970).
8. Larson, T.E., "Evaluation of the Use of Polyphosphates in the Water
Industry," JAWWA 49: 1581-1586 (December 1957).
9. O'Connor, J.T., "Iron and Manganese, Chapter 11: Water Quality
and Treatment," American Water Works Association, McGraw-Hill
Book Company, New York, New York (1 971).
10. Hatch, G.B., "Inhibition of Lead Corrosion with Sodium Hexameta-
phosphate," JAWWA 33: 1179-1187 (July 1941).
11. Moore, E.W. and F.E. Smith, "Effect of Sodium Hexametaphos-
phate on the Solution of Lead," JAWWA 54:1415-1424 (September
1942).
12. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Page 40.
13. "European Standards for Drinking Water," 2nd Edition, World
Health Organization. Geneva, 1970. Page 37.
14. Stoffen, D., "The Maximum Permissible Concentrations in the USSR
for Harmful Substances in Drinking Water," Toxicology. Amster-
dam, 1973. Page 190.
25
-------
15. "Official Journal of the European Communities," Volume 18.
September 18, 1975. Page 9.
16. 21 CFR Part 103 - Quality Standards for Foods with no Identity
Standards - Bottled Water. Federal Register, Vol. 44, No. 45,
12172, March 6, 1979.
Manganese
Manganese and iron have similar aesthetic effects and frequently occur to-
gether in natural waters. At one time they were both covered by a single
recommended limit, however, the 1962 PHS Standards1 recommended
separate limits for iron and manganese to reflect more accurately the levels
at which environmental effects occur for each. There are two reasons for
limiting the concentration of manganese in drinking water: (a) to prevent
aesthetic and economic damage, and (b) to avoid any possible physiologic
effects from excessive intake. The principal reason for limiting the concen-
tration of manganese is to reduce the aesthetic and economic prob-
lems.2'3> 4 The domestic consumer finds that it produces a brownish color
in laundered goods and impairs the taste of drinking water and other
beverages including coffee and tea.2'4 Taste thresholds in drinking water
are considerably higher than the levels which produce staining effects.5
In concentrations of only a few hundredths milligram per liter, manga-
nese may occasionally cause buildup of coatings in distribution piping. If
these coatings slough off, they can cause brown blotches in laundry items
and black precipitates. Griffin,3 as chairman of a task group on "Manga-
nese Deposition in Pipelines," stated that the concentration of manganese
which could be tolerated by the average consumer is 0.01 — 0.02 mg/1.
The World Health Organization (1970) suggests that such problems may
arise at concentrations of manganese greater than 0.05 mg/1, the same limit
recommended by the U.S. Public Health Service in 1962. In an unpub-
lished survey of 13 States reporting on levels of manganese giving rise to
water quality problems, only three States recommended levels as high as
0.2 mg/1, two permitted 0.15 mg/1 and four each permitted 0.1 mg/1 or
0.05 mg/1.1 Domestic complaints generally arise when the level of manga-
nese exceeds 0.15 mg/1 but aesthetic problems can occur at lower levels.
The application of chlorine increases the likelihood of precipitation of
manganese at low levels; unless the precipitate is removed it can cause
aesthetic problems.
From the health standpoint, data do not indicate at what level manga-
nese would be harmful when ingested.6' 7> 8l 9 The principal toxic effects
that have been reported are as a result of inhalation of manganese dust or
fumes. It has been estimated that the daily intake of manganese from a
normal diet is about 10 mg.10 In animals, at least, it has been shown to be
an essential nutrient, since diets deficient in manganese interfere with
growth, blood and bone formation, and reproduction. Hepatic cirrhosis
has been produced in rats when treated orally with very large doses. As far
as is known, the neurologic effects of manganese have not been reported
from oral ingestion in man or animal.11 Ingestion of manganese in moder-
26
-------
ate excess of the normal dietary level is not considered harmful. Concen-
trations of manganese found in water supplies are much less than those at
which adverse health effects have been observed.11
Values for maximum manganese concentrations in drinking water set
by other agencies are as follows: WHO International Standards — 0.05
mg/112 ; WHO European Standards — 0.05 mg/113 ; proposed European
Community (EC) Standards - 0.05 mg/114 ; and FDA Bottled Water
Standards - 0.05 mg/115.
Manganese in concentrations greater than 0.05 mg/1 may be sequestered
by the addition of poly phosphates or silicates.9'16 However, Hatch,17 and
Moore and Smith18 have indicated that polyphosphates can introduce lead
into stagnant water from lead pipes, particularly at pH levels above 7 or 8.
Larson19 also notes that the polyphosphate additions may lead to exces-
sive bacterial growths in the distribution system, thereby creating a need
for increased disinfectant residuals.
In summary, manganese is similar to iron in that it can be sequestered
to reduce its undesirable effects, but removal is preferable. Manganese can
be removed by oxidation followed by filtration. The iron removal filter
using "manganese greensand" is also effective for manganese removal.
The secondary maximum contaminant level of 0.05 mg/1 was deter-
mined from the above data, is in agreement with other agency standards,
and represents a level which prevents most aesthetic effects.
References
1. Public Health Service Drinking Water Standards, 1962, U.S. Depart-
ment of Health, Education and Welfare, PHS Pub. No. 956. Pages
4647.
2. Griffin, A.E., Manganese Removal with Chlorine and Chlorine Diox-
ide," J. New England Water Works Assn. 72: 321-327 (September
1958).
3. Griffin, A.E., "Significance and Removal of Manganese in Water
Supplies," JAWWA 52: 1326-1334 (October 1960).
4. Riddick, T.M., N.L. Lindsay and A. Tomassi, "Iron and Manganese
in Water Supplies," JAWWA 50: 688-696 (May 1958).
5. Cohen, J.M., L.J. Kamphake, E.K. Harris, and R.L. Woodward,
"Taste Threshold Concentrations of Metals in Drinking Water,"
JAWWA 52: 660-670 (May 1960).
6. Cotzias, G.C., "Manganese in Health and Disease," Physiol. Rev. 38:
503-532(1958).
7. Drill, V.A., "Pharmacology in Medicine," Ed. 2, New York, McGraw-
Hill, pp. 709, 787, 794 (1958).
8. Drinking Water and Health, 1977. Report of the National Academy
of Sciences, Safe Drinking Water Committee. Page 270.
9. Illig, G.L., Jr., "Use of Sodium Hexametaphosphate in Manganese
Stabilization," JAWWA 52: 867-874 (July 1960).
10. Sollman, T.H., "A Manual of Pharmacology," Ed. 8, Philadelphia,
Pennsylvania, W.B. Saunders Co., pp. 1278-1281 (1957).
11. von Oettingen, W.F., "Manganese: Its Distribution, Pharmacology
and Health Hazards," Physiol. Rev. 15: 175-201 (1935).
27
-------
12. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Page 40.
13. "European Standards for Drinking Water," 2nd Edition, World
Health Organization. Geneva, 1970. Page 37.
14. "Official Journal of the European Communities," Volume 18. Sep-
tember 18, 1975. Page 9.
15.21 CFR Part 103 - Quality Standards for Foods with no Identity
Standards - Bottled Water. Federal Register, Vol. 44, No. 45,
12172, March 6, 1979.
16. Dart, F J. and P.D. Foley, "Silicate as Fe, Mn Deposition Preventa-
tive in Distribution Systems," JAWWA 64: 244-249 (April 1972).
17. Hatch, G.B., "Inhibition of Lead Corrosion with Sodium Hexameta-
phosphate," JAWWA 33: 1179-1187 (July 1941).
18. Moore, E.W. and F.E. Smith, "Effects of Sodium Hexametaphos-
phate on the Solution of Lead," JAWWA 34: 1415-1424 (September
1942).
19. Larson, T.E., "Evaluation of the Use of Poly phosphates in the Water
Industry," JAWWA 49: 1581-1586 (December 1957).
20. Quality Goals for Potable Water, A Statement Adopted by the Board
of Directors on January 28, 1968, JAWWA 73-74 Yearbook, pp. 60-
63 (September 1973).
Odor
The absence of taste and odor of water helps to maintain the consumer's
confidence in the quality of their drinking water even though this does not
guarantee that the water is safe. Industrial processes such as food, beverage
and pharmaceutical manufacturers, require water essentially free of taste
and odor. Taste and odor tests are also useful as a check on the quality of
raw and finished water, for control of odor through the plant and the
determination of treatment dosages. Odor is also useful as a test of the
effectiveness of different kinds of treatment, and as a means of tracing the
source of contaminants.
According to psychologists, there are only four true taste sensations:
sour, sweet, salty and bitter. Dissolved inorganic salts of copper, iron,
manganese, potassium, sodium and zinc can be detected by taste. Concen-
trations producing taste range from a few tenths to several hundred milli-
grams per liter. As these tastes are not accompanied by odor, the taste test
must be used where they are involved.
All other sensations ascribed to the sense of taste are actually odors,
even though the sensation is not noticed until the material is taken into
the mouth. Odor tests are performed to arrive at qualitative descriptions
and approximate quantitative measurements of odor intensity. Odor tests
are less fatiguing than taste tests; hence, an operator can conduct odor
tests for a longer period. Higher temperatures can be used for odor evalua-
tions than for the taste test, with a resultant increase in sensitivity on some
samples.
28
-------
Some organic and inorganic chemicals contribute both taste and odor.
These may originate from municipal and industrial waste discharges,
natural sources (such as decomposition of vegetable matter), or from asso-
ciated microbiological activity. Because some odorous materials are detect-
able when present in only a few micrograms per liter and are often com-
plex, it is usually impractical and often impossible to isolate and identify
the odor-producing chemical.
Burttschell, Rosen, Middleton and Ettinger1 showed that chlorine,
when combined with phenol, produces particularly odorous chlorophen-
olic compounds. Further, they showed that about 20 percent of the
original phenol concentration was present as 2,4-dichlorophenol 18 hours
following chlorination, while 25 percent was present as 2,6-dichlorophenol
after the same period. Table 1 shows the odor threshold concentration for
phenol and its chlorinated products.
More recently, the AWWA Committee on Tastes and Odors reported on
odorous natural metabolites.2 Both geosmin and mucidone have been sug-
gested as causative agents of musty odors produced from algae and actino-
mycetes. Rosen, Mashni and Safferman3 isolated and identified geosmin
and 2-methylisoborneol from carbon-chloroform extracts collected from
two Ohio lakes exhibiting odors.
The Threshold Odor Number (TON) of water is the dilution factor re-
quired before the odor is minimally perceptible. A TON of 1 indicates that
the water has characteristics comparable to odor-free water, while a TON
of 4 indicates that a volume of the test water would have to be diluted to
4 times its volume before the odor became minimally perceptible. For pre-
cise work, a panel of five or more testers is required, and the TON is based
on the greatest amount of dilution which elicits a positive odor response
from one of the testers.
The following maximum odor levels have been established by other
agencies: Proposed European Community Standards — 2 TON at 12° C
and 3 TON at 25° C6 ; FDA Bottled Water Standards - 3 TON7 . While
other agencies recommend testing regularly for odor levels, these agencies
have not established limits.8
The control of odor in water requires knowledge of the nature of the
odorous material. Oxidative processes are effective when the odor is due
to sulfide, chlorophenols, and many other substances. Activated carbon is
effective against some odors of organic origin.
29
-------
Table 1 — Odor Threshold Concentrations
Compound Odor Threshold Concentration - ug/1
Burttschell,
etal.1'4 Baker4 de Grunt4 Zoeteman5
Phenol
2-chlorophenol
4-chlorophenol
2 ,4-dichlorophenol
2,6-dichlorophenol
2,4,6-trichlorophenol
2
2
2
3
1,000
5,900
1,240
210
0.2
0.5
7.5
5,900
0.2
210
8
The secondary maximum contaminant level of 3 TON was determined
to be the odor level that most consumers would find acceptable, and is in
agreement with other agency standards as noted above.
References
1. Burttschell, R.H., A.A. Rosen, P.M. Middleton and M.B. Ettinger,
"Chlorine Derivatives of Phenol Causing Taste and Odor/' JAWWA
51: 205-214 (February 1959).
2. AWWA Committee on Tastes and Odors, Research on Taste and
Odors, JAWWA 62: 59-62 (January 1970).
3. Rosen, A.A., C.I. Mashni and R.S. Safferman, "Recent Develop-
ments in the Chemistry of Odor in Water: The Cause of Earthy/
Musty Odor," Water Treatment and Examination, 19 Part 2: 106-
119(1970).
4. "Compilation of Odor Threshold Values in Air and Water," National
Institute for Water Supply, Voorbug, Netherlands, Central Institute
for Nutrition and Food Research TNO, Zeist, Netherlands, Pages 37-
50, June, 1977.
5. Zoeteman, B.C.J., Sensory Assessment and Chemical Composition of
Drinking Water. Gravenage, The Netherlands, Pages 148-151, (1978).
6. "Official Journal of the European Communities," Volume 18, Sep-
tember 18, 1975. Page 6.
7. 21 CFR Part 103 - Quality Standards for Foods with no Identity
Standards - Bottled Water. Federal Register, Vol. 44, No. 45,
12172, March 6, 1979.
8. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Pages 3841.
pH
Natural waters used as drinking water sources range from highly alkaline
ground waters to very acidic surface waters. Although from a health stand-
point a wide range of pH values for drinking water can be tolerated, a low
30
-------
pH may cause the need for heavier chlorination, whereas a high pH may
cause increased halogen reactions. Both excessively high and low pH's may
cause increased corrosivity and an unpleasant taste. These considerations
suggest employment of a narrower pH range.
At low and high pH's, water tends to be corrosive, in that it tends to
dissolve materials with which it is in contact. When metallic piping is in
contact with water of low pH, some of the metal is dissolved, imparting a
metallic taste to the water in some instances. If the metal is iron or copper,
oxides and carbonates will be deposited, leaving red or green stains, respec-
tively. If the metal is lead or galvanized (containing lead and cadmium as
impurities), there may be no apparent corrosive effect, but the water will
contain potentially hazardous amounts of these toxic metals.
Many factors may influence the corrosivity of water in addition to low
pH but it is a sufficiently common factor to merit recommending a mini-
num level. Water below a pH level of 6.5 will usually be corrosive.1
At high pH's, drinking water acquires a bitter taste. The high degree of
mineralization often associated with alkaline waters results in encrustation
of water pipes and water-using appliances.
In addition the disinfection activity of chlorine falls significantly as pH
levels rise.2 For example, several times as much chlorine is required for a
100 percent kill of Salmonella typhosa (currently called S. typhi) at a pH
of 9.8 as against a pH of 7.0.
High pH levels also appear to have an accelerating effect on the halo-
form reaction which produces chloroform and other trihalomethanes dur-
ing chlorination.3 Constraint of this reaction is certainly a desirable
objective.
While a number of factors, in addition to high pH, influence taste, en-
crustation, disinfection and the formation of trihalomethanes, it was deter-
mined that a maximum level of 8.5 is a reasonable goal. The WHO has set
the lower limit at a pH of 7.04, whereas, the European Communities have
proposed a range of 6.5 to 8.5.5
Adjustment of pH is accomplished by addition of alkaline or acidic
chemicals such as sodium hydroxide, lime, soda ash, carbon dioxide and
sulfuric acid.
The secondary maximum contaminant level for pH is therefore the
range 6.5 to 8.5. Corrosive properties are minimized by setting the lower
part of the pH range at 6.5 and the higher part of the pH range of 8.5
minimizes the other problems associated with higher pH values.
References
1. Water Quality and Treatment, Chapter 8, Corrosion Phenomena —
Causes and Cures, pp. 295-312, 3rd Edition, AWWA (1971).
2. Manual for Evaluating Public Drinking Water Supplies, Public Health
Services Publication No. 1820, pp. 3740 (1969).
3. Morris, J. Carrell, "Formation of Halogenated Organics by Chlorina-
tion of Water Supplies," EPA-600/1-75-002 (March 1975).
4. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Pages 31-41.
5. "Official Journal of the European Communities," Volume 18. Sep-
tember 18, 1975. Page 7.
31
-------
Sulphates
There are three reasons for limiting the concentration of sulfates in drink-
ing water: (a) waters containing appreciable amounts of sulfate tend to
form hard scales in boilers and heat exchangers; (b) sulfates cause taste
effects; and (c) sulfates can cause physiological effects (i.e., laxative ef-
fects) with excessive intake.1
Both sodium sulfate and magnesium sulfate are well known laxatives.
The laxative dose for both Glauber salts (Na2 864 10H2O) and Epsom
salts (MgSO4 7H2O) is about two grams. Two liters of water with about
300 mg/1 of sulfate derived from Glauber salt, or 390 mg/1 of sulfate from
Epsom salt, would provide this dose. Calcium sulfate is much less active in
this respect.
The laxative effect is commonly noted by newcomers and casual users
of water high in sulfates. One evidently becomes acclimated to use of these
waters in a relatively short time.
The North Dakota State Department of Health has collected informa-
tion on the laxative effects of water as related to mineral quality. This has
been obtained by having individuals submitting water samples for mineral
analysis complete a questionnaire that asks about the taste and odor of the
water, its laxative effect (particularly on those not accustomed to using it),
its effect on the flavor of coffee, and its effect on potatoes cooked in it.
Peterson2 and Moore3 have analyzed part of the data collected, particu-
larly with regard to the laxative effect of the water. Peterson found that,
in general, waters containing more than 750 mg/1 of sulfate showed a laxa-
tive effect while those with less than 600 mg/1 generally did not. Moore
showed that laxative effects were experienced by the most sensitive per-
sons, not accustomed to the water, when magnesium was about 200 mg/1
and by the average person when magnesium was 500-1,000 mg/1. Moore
analyzed the data as shown in Table 1. When sulfate plus magnesium ex-
ceed 1,000 mg/1, a majority of those who gave a definite reply indicated a
laxative effect. No other adverse health effects have been noted for con-
centrations less than about 500 mg/liter. The only observed physiological
effect at higher concentrations has been the induction of diarrhea.1
Table 2 presents some data collected by Lockhart, Tucker and Merritt4
and Whipple5 on the influence of sulfate on the taste of water and coffee.
Because of the milder taste of sulfate over chloride,6' 7 the taste of sulfate
would probably initially be detected in the 300^-00 mg/1 range, but some
people are able to detect taste at the 200 mg/1 level.1 The Peterson data2
and Table 1,3 however, indicated that from 600 to 1,000 mg/1 of sulfate
has a laxative effect on a majority of users. The data indicate that no sig-
nificant taste effects occur at 200-300 mg/1.
Values for maximum sulfate concentrations in drinking water set by
other agencies are as follows: WHO International Standards — 200 mg/18 ;
WHO European Standards - 250 mg/19 ; proposed EC Standards - 250
mg/110 ; and FDA Bottled Water Standards - 250 mg/1" .
Sulfate cannot be readily removed from drinking water without use of
distillation, reverse osmosis or electrodialysis which are effective but rela-
32
-------
tively expensive; thus, the use of an alternate source or blending are the
most desirable approaches.
The secondary maximum contaminant level of 250 mg/1 was deter-
mined as the level which would prevent most taste effects. This SMCL
will also prevent the undesirable laxative effects even in most casual
consumers.
Table I — Water Quality of Wells as Related to Presence or Absence
of Laxative Effects (Moore3)
Determination
Magnesium plus
sulfate
Sulfate
Range mg/1
0-200
200-500
500-1,000
1,000-1,500
1,500-2,000
2,000-3,000
Over 3, 000
0-200
200-500
500-1,000
1,000-1,500
1,500-2,000
2,000-3,000
Over 3, 000
Number Laxative
of Wells
in Range Yes No
51
45
56
36
14
21
14
56
47
56
34
16
20
8
9
7
11
18
6
13
5
10
9
13
16
9
9
3
34
27
38
10
4
3
1
36
28
26
10
4
3
0
Effects Percent
Present of Yes
Not Answers
Stated *
8
11
17
8
4
5
8
10
10
17
8
3
8
5
21
21
28
64
60
81
83
22
24
33
62
69
75
100
"This percentage is based only on the total of yes and no answers. It is probable that
a large proportion of the wells for which no statements were made were not regularly
used as water supplies.
Table 2 — Data on the Influence of Sulfate Salts on the Taste of Water and
Coffee (I ockhart. et al4; WhippleS)
Threshold Concentration - mg/1
Median
Salt
Na2SO4
CaSO4
MgS04
MgSO4
Salt
350
525
525
Average
500
Anion
327
370
419
400(5)
Range
Salt
250-550
250-900
400-600
Anion
169-372(4)
177-635 (4)
320-479 (4)
33
-------
References
1. Drinking Water and Health, 1977. Report of the National Academy
of Sciences, Safe Drinking Water Committee. Pages 425-428.
2. Peterson, N.L., "Sulfates in Drinking Water," Official Bulletin,
North Dakota Water and Sewage Works Conference, 18: 6-7, 11
(April-May 1951).
3. Moore, E.W., "Physiological Effects of the Consumption of Saline
Drinking Water," Bulletin of Subcommittee on Water Supply, Na-
tional Research Council, Appendix B, pp. 221-227 (January 10,
1952).
4. Lockhart, E.E., C.L. Tucker, and M.C. Merritt, "The Effect of Water
Impurities on the Flavor of Brewed Coffee," Food Research 20:
598-605(1955).
5. Whipple, G.C., "The Value of Pure Water," Wiley (1907).
6. Bruvold, W.H. and W.R. Gaffey, "Evaluation Ratings of Mineral
Taste in Water," J. Perceptual and Motor Skills 28: 179 (1969).
7. Bruvold, W.H. and W.R. Gaffey, "Rated Acceptability of Mineral
Taste in Water. II. Combinational Effects of Ions on Quality and Ac-
tion Tendency Ratings," J. Applied Psychol. 53: 317 (1969).
8. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Pages 3840.
9. "European Standards for Drinking Water," 2nd Edition, World
Health Organization. Geneva, 1970. Page 37.
10. "Official Journal of the European Communities," Volume 18. Sep-
tember 18, 1975. Page 9.
11.21 CFR Part 103 - Quality Standards for Foods with no Identity
Standards - Bottled Water. Federal Register, Vol. 44, No. 45,
12172, March 6, 1979.
Total Dissolved Solids
A SMCL for total dissolved solids (TDS) was established because high con-
centrations have adverse taste effects which force consumers to use other
water sources. Highly mineralized water also deteriorates distribution and
domestic plumbing and appliances.
It should be emphasized that there may be a great difference between a
detectable concentration and an objectionable concentration of the neu-
tral salts. The factor of acclimatization is particularly important. A num-
ber of public supplies in the United States provide water with more than
2,000 mg/1 of dissolved solids. Newcomers and casual visitors would cer-
tainly find these waters almost intolerable and although some of the resi-
dents use other supplies for drinking, many are able to tolerate (and pos-
sibly enjoy) these highly mineralized waters.
Recent data from Bruvold, et al.1 have demonstrated that more acute
taste effects occur as the total mineral content rises. Pangborn, et al.2 have
also shown that the temperature of the water influences the acceptability
of mineralized waters. One study is cited3 as showing that families using
34
-------
waters with TDS concentrations ranging from 500-1750 mg/1 expressed
dissatisfaction with their supply and turned to bottled water as an alter-
nate source. Waters containing TDS concentrations in the range of 10,000
mg/1 are considered unusable for drinking.
Mineralization has been shown to have an economic impact on water
distribution systems and household plumbing and appliances.4 According
to Lawrence5 , the life of home hot water heaters decreases one year for
each additional 200 mg/1 of TDS in water above a typical 200 mg/1 figure.
Mineralization can also cause precipitates to form in boilers or other
heating units, sludge in freezing processes, rings on utensils and precipi-
tates in foods being cooked.
The 1962 edition of the Public Health Service Drinking Water Stand-
ards3 included a limit for total dissolved solids (TDS) of 500 mg/1 because
of taste effects and because drinking water containing a high concentration
of TDS was likely to contain an excessive concentration of some specific
substance that would be aesthetically objectionable to the consumer. The
WHO International Standard and the FDA Bottled Water Standard are also
set at 500 mg/1.6 The AWWA recommended quality goal is 200 mg/1.7
The use of alternative sources and/or blending are recommended when
dissolved solids exceed the SMCL. The available treatment processes (dis-
tillation, reverse osmosis, electrodialysis) are effective but relatively
expensive.
The secondary maximum contaminant level of 500 mg/1 was deter-
mined to represent a reasonable goal which would prevent most aesthetic
effects of dissolved solids.
References
1. Bruvold, W.H., H.J. Ongerth and R.C. Dillehay, "Consumer Assess-
ment of Mineral Taste in Domestic Water," JAWWA 61: 575-580
(November 1969).
2. Pangborn, R.M. and L.L. Bertolero, "Influence of Temperature on
Taste Intensity and Degree of Liking of Drinking Water," JAWWA
64: 511-515 (August 1972).
3. Public Health Service Drinking Water Standards, 1962. U.S. Depart-
ment of Health, Education and Welfare, PHS Publication No. 956.
4. Patterson, W.L. and R.F. Banker, "Effects of Highly Mineralized
Water on Household Plumbing and Appliances," JAWWA 60: 1060-
1069 (September 1968).
5. Lawrence, H.H., "Estimating Indirect Cost of Urban Water Use,"
J. of the Envir. Eng. Div., ASCE, pp. 517-533 (August 1975).
6. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Pages 38-40.
7. Quality Goals for Potable Water, A Statement Adapted by the Board
of Directors on January 28, 1968, JAWWA Yearbook 73-74: 60-63
(September 1973).
35
-------
Zinc
Although zinc is an essential and beneficial element in human metabol-
ism1'2 , excessive amounts in drinking water may produce adverse physio-
logical and taste effects, a milky appearance, and may increase lead and
cadmium concentrations.
The daily requirement for preschool-age children is 0.3 mg Zn/kg. Total
zinc in an adult human body averages 2 grams. Zinc content of human
tissues ranges from 10-200 ppm wet weight; the retina of the eye and the
prostate contain the largest concentrations (500-1,000 ppm). Three per-
cent of all blood zinc is in the white cell. The daily adult human intake
averages 10-15 mg; excretion of zinc averages about 10 mg in the feces and
0.4 mg in the urine. Zinc deficiency in animals leads to growth retardation.
This can be overcome by adequate dietary zinc. The activity of several
body enzymes is dependent on zinc.
A group of individuals stationed at a depot used a drinking water sys-
tem containing zinc at 23.8 to 40.8 mg/1 and experienced no known harm-
ful effects. Communities have used waters containing from 11-27 mg/1
without harmful effects.3'4 Another report5 stated spring water contain-
ing 50 mg/1 was used for a protracted period without noticeable harm. On
the other hand, another supply containing approximately 30 mg/1 was
claimed to cause nausea and fainting.
Zinc salts act as gastrointestinal irritants. Although the illness is acute,
it is transitory. The emetic concentration range in water is 675-2,280 mg/1.
A wide margin of safety exists between normal intake from food and
water and those quantities likely to cause oral toxicity.
Zinc compounds also impart taste to drinking water. In tests performed
by a taste panel, 5 percent of the observers were able to distinguish be-
tween water containing 4 mg/1 (when present as zinc sulfide) and water
containing no zinc salts. The water was described as having a bitter or
astringent taste. Soluble zinc salts at 30 mg/1 may impart a milky appear-
ance to water, and at 40 mg/1, a metallic taste.6
Cadmium and lead are common contaminants of zinc used in galvaniz-
ing. Assuming that zinc is dissolved from galvanized water pipe no less
than cadmium, dissolution of zinc to produce 5 mg/1 would be accom-
panied by something less than the allowable 0.01 mg cadmium per liter
when cadmium contamination of the zinc is as high as 0.03 percent. Like-
wise, lead concentrations would likely be increased by something less than
the allowable 0.05 mg/1 when lead contamination of the zinc is as high as
0.6 percent. An AWWA statement cites two States as having a zinc limit of
1.0 mg/1 and recommends 1.0 mg/1 as a water quality goal.7
Values for maximum zinc concentrations in drinking water set by other
agencies are as follows: WHO International Standards - 5 mg/18 ; WHO
European Standards - 5 mg/19 ; FDA Bottled Water Standards - 5
mg/l10;USSR 16 hours standard - 1.0 mg/111 ; and the proposed EC
standards - 2.0 mg/112 (16 hours standard).
36
-------
The principal source of zinc in drinking water is the dissolution of gal-
vanizing by aggressive waters. Such dissolution can be minimized by stabil-
ization or neutralization. If zinc is a constituent of the source water, con-
ventional treatment may be effective.
The secondary maximum contaminant level of 5 mg/1 was determined
to be the level which would prevent most taste problems.
References
1. Drinking Water and Health, 1977. Report of the National Academy
of Sciences, Safe Drinking Water Committee.
2. Valle, B.L., "Zinc and Its Biologic Significance," Arch. Indus. Health
16: 147-154 (July 1957).
3. Anderson, E.A., C.E. Reinhard and W.D. Hammel, "The Corrosion
of Zinc in Various Waters," JAWWA 26: 49-60 (January 1934).
4. Bartow, E. and O.M. Weigle, "Zinc in Water Supplies," Indus. Eng.
Chem. 24: 463465 (1932).
5. Hinman, J.J., Jr., "Desirable Characteristics of a Municipal Water
Supply," JAWWA 30: 484494 (March 1938).
6. Kehoe, R.A., J. Cholak and E.J. Largent, "The Hygienic Significance
of the Contamination of Water of Certain Mineral Constituents,"
JAWWA 36: 645-657 (June 1944).
7. Quality Goals for Potable Water, A Statement Adopted by the Board
of Directors on January 28, 1968, JAWWA Yearbook 73-74: 60-63
(September 1973).
8. "International Standards for Drinking Water," 3rd Edition, World
Health Organization. Geneva, 1971. Page 3840.
9. "European Standards for Drinking Water," 2nd Edition, World
Health Organization. Geneva, 1970. Page.37.
10. 21 CFR Part 103 - Quality Standards for Food with no Identity
Standards - Bottled Water. Federal Register, Vol. 44, No. 45,
12172, March 6, 1979.
11. Stoffen, D., "The Maximum Permissible Concentrations in the USSR
for Harmful Substances in Drinking Water," Toxicology. Amster-
dam, 1973. Page 190.
12. "Official Journal of the European Communities," Volume 18.
September 18, 1975. Page 9.
13. Cohen, J.M., L.J. Kamphake, E.K. Harris, and R.L. Woodward,
"Taste Threshold Concentrations of Metals in Drinking Water,"
JAWWA 52: 660-670 (May 1960).
* U.S. GOVERNMENT PRINTING OFFICE 1980 O- 311-132'35
37
-------
------- |