f .. j *• ¦'t - — - ' ^
¦
" . -v "
v.- •' ' ... \ : ••
' ; ' - . '
v- -- " • ¦ - • -¦ ...
.f * • - . • , ' V ' : . .r
~ ««»-»•* * r ' ¦: ' * ¦ ' ' . *. ~ kVJ. .
, •- . ' • ' . ¦ • . . -i - .i +-
• • •' '¦ ----- - • " ¦ „"*¦-
,'ir^
- «- « ,| • . ' ;¦ . ' • ¦ • . • ^ '
'r* *'¦*
- r- -. -i
. •¦ ¦¦.-;... -
3?.V - •
•>'v-- ¦
v*.
«>sr ¦
V, M ••
IV
IV
' v C1
' .i ?;• •
' *
f
L
^ «. m
.. m. » '
>--V« •
Jr_' 1^-- -
^-sA,
_
¦4"%, „
V.,.
U, ' * : *
' &¦'[; "¦ uv"-v-vr- ^-
• f/; •• .• • "•
'¦ . - U,-
Project EPA 84-1.
DRAFT FINAL REPORT
REVIEW OF THE ANALYTICAL
PROCEDURES FOR THE STUDY OF
AQUATIC ORGANIC ACIDS AND
THEIR CONTRIBUTION TO" SURFACE
WATER ACIDITY
Submitted To:
United States Environmental
Protection Agency
Office of Monitoring Systems
and Quality Assurance
401 M Street, S.W.
Washington, D.C. 20460
% -
November 1984
-------
55SHSW
¦i ¦»?, '¦ ¦* fe
, •' #- «fe^. .>»•.'•
•su. ., •' ¦'- :. ¦ -: • - V
¦jpH
^ - '* J"
•• . .£T* •• -.^ ;'-':^i
§*• C? =..;^ ¦»- • i 4%^:, Va' ""
2. iv ¦«*, ¦ > * - >' •% <¦ •' :'>-- -.+ :: ^Eh .
--.*#.\v*. "* 1.-3;?.. a-dV4
.tf'i; -v. ." »¦?-.. - .„-. -...?,
"-- -- ;*«• .>-:•
.. ,3&* *t<«0w£. ^
¦'£. •• c"'
» n # sp
b^.3&
;.. * ¦¦ =?,» .
* -tta5
¦*41 a^„*v"
~J.;1
V- • '<*?&
.'•>? 1
®:' .
t& . .e
V J* ¦^v' .
' "** i fr ' I
Project EPA 84-1.1
DRAFT FINAL REPORT
REVIEW OF THE ANALYTICAL
PROCEDURES fFOR ,THE STUDY OF
AQUATIC ^ORGANIC ACIDS AND
THEIR .CONTRIBUTION TO" SURFACE
WATER ACIDITY
Submitted To:
United States Environmental
Protection Agency
Office of Monitoring Systems'
and'Quality Assurance
401 "M: Street, S.W.
Washington, D.C. 20460
November 1984
-------
Project EPA 84-1.1
DRAFT FINAL REPORT
REVIEW OF THE ANALYTICAL PROCEDURES FOR THE
STUDY OF AQUATIC ORGANIC ACIDS AND THEIR
CONTRIBUTION TO SURFACE WATER ACIDITY
Prepared By:
Anthony Janicki
Joseph Arlauskas
Kenneth Yetman
Harry Scott
Martin Marietta Environmental Systems
9200 Rumsey Road
Columbia, Maryland 21045
Submitted To:
United States Environmental Protection Agency
Office of Monitoring Systems and Quality Assurance
401 M Street, S.W.
Washington, D.C. 20460
November 1984
-------
Martin Marietta Environmental Systems
FOREWORD
This report was produced at the request of Dr. Courtney
Riordan, Director Office of Monitoring Systems and Quality
Assurance, United States Environmental Protection Agency,
under Interagency Agreement No. DW13931002-01-Q. The principal
coordinator for this project was Dr. Tibor Polgar, Director,
Technical and Scientific Operations, Martin Marietta Environmental
Systems
-------
Martin Marietta Environmental Syitemi
EXECUTIVE SUMMARY
The objective of this report is to provide a broad overview
of the currently available analytical methods for the isolation,
characterization, and quantification of naturally occurring
organic acids in surface waters, and to identify gaps in the
knowledge of organic acids which require further investigation.
The underlying motivation for this report is the need for an
analytical protocol for quantifying the contributions of
anthropogenically and naturally produced acids to the process of
surface water acidification.
The results of our review indicate that in order to determine
the contribution of organic acids to surface water acidity,
standardized analytical methods are needed for the following:
ISOLATION AND CONCENTRATION
A large number of methods for isolation and
concentration have been used. Recent reviews of
their advantages and disadvantages have been
conducted; however, no standardization of these
methods has been accomplished. Standardization of
the isolation and concentration methods is necessary
for the develpment of an analytical protocol.
• ACIDIC FUNCTIONAL GROUP ANALYSIS
The carboxyl group is the acidic functional
group of interest at pH conditions < 7.
Titration methods for determining the carboxyl
group content of the organic acids have been
developed but as yet are not standardized.
Standardization of the methods is necessary and
would require an operational definition of the
organic acids being analyzed.
• DISSOCIATION OF ACIDIC FUNCTIONAL GROUPS
The degree of dissociation of the acidic func-
tional groups must also be known in order to
quantify the contribution of the organic acids
to surface water acidity. Dissociation constants
have been estimated under laboratory conditions
using purified samples. It has not been shown
that these estimates can be validly applied to
quantify the amount of ionized organic acids under
v
-------
Martin Marietta Environmental Systems
in situ conditions in surface waters. The dis-
sociation behavior of hydrophilic organic acids
must also be addressed.
Besides standardization of the above methods, criteria
for determining the suitability of the approaches must be
defined. These criteria include:
• Applicability to waters within a wide range of chemical
conditions, including most notably low to high ionic
strength, low to circumneutral pH, and low to high
trace metal concentrations.
• Feasibility for use in large surveys, such as the National
Surface Water Survey, and long-term monitoring programs
which would require large numbers of samples to be
processed, some of which would have been obtained from
extremely remote areas.
• Provision of levels of precision, accuracy, and sensi-
tivity to satisfy the requirements of the particular
programs which may be served.
vi
-------
Martin Msriitta Environmental SyrtMM
TABLE OF CONTENTS
Chapter Page
FOREWORD iii
EXECUTIVE SUMMARY v
I. INTRODUCTION 1-1
II. NATURE AND SOURCES OF LAKE WATER ACIDITY.. II-l
III. ANALYTICAL METHODS III-l
IV. CONCLUSIONS IV-1
V. ANNOTATED BIBLIOGRAPHY V-l
VI. REFERENCES VI-1
RP-281
vii
-------
Martin Marietta Environmental Systami
LIST OF TABLES
No. Page
1 Operational carboxyl contents of fulvic acids
(PA) and humic acids (HA) from various loca-
tions in the United States and Canada II-8
2 Dissolved organic carbon (DOC) fractionation
data for selected freshwater systems of the
United States 11-11
3 Methods used to concentrate and isolate
organic constituents from water samples....... III-2
4 Advantages and disadvantages of silver mem-
brane and glass fiber filters for filtration
of water for the isolation of organic acids... III-4
5 Methods commonly used to isolate and concen-
trate aquatic humic substances III-6
6 Extraction scheme using XAD-8 to concentrate
aquatic humic substances as devised by
Thurman and Malcolm 111-12
viii
-------
Martin Marietta Environm«ntal Symmi
LIST OF FIGURES
Wo. Page
1 Spectrum of lakes whose acidity is derived
from anthropogenic and natural sources 1-2
2 Conceptual approach to determining the contri-
bution of naturally produced organic acids to
lake water acidity.*. 1-4
3 Gran plot depicting the effect of the presence
of weak acids on the estimation of the strong
acid fraction II-2
4 Functional groups of humic acids (HA) and fulvic
acids (FA).............................. II-5
5 Examples of the 4 types of elution patterns
found among 22 lakes in Minnesota (from Shapiro,
1967) II-6
6 Dissasocation of carboxylic acid and phenolic
groups in organic materials as a function of
pH........ 11-10
7 Analytical procedure for preparative dissolved
organic carbon fractionation* 111-14
8 Fractionation of organic solutes in water by
the method of Leenheer and Noyes. 111-16
9 Relationship bewteen pl( and pH for FA from two
aquatic systems................................. 111-19
ix
-------
Martin Marietta Environmental Systems
I. INTRODUCTION
Atmospheric deposition of man-made acidic materials has
been hypothesized to have resulted in the acidification of
surface waters. Alternate hypotheses propose that natural
sources of acidity are responsible for this acidification
(Patrick et al., 1981; Retzsch et al., 1982; Marcus et al.,
1983). Unquestionably, naturally acidic surface waters have
existed for thousands of years (Ruttner, 1963); however, this
does not preclude the more recent acidification from acid inputs
of anthropogenic origin. Therefore, in order to establish the
specific cause of lake acidification, the relative contributions
of natural and anthropogenic sources of acidity must be determined
Acid lakes fall within a spectrum ranging between the
extremes of natural or anthropogenic acidification (Fig. 1). A
number of striking distinctions have been found between the two
extreme lake types at the opposite ends of this spectrum (Havas
•t al., 1984). Typically, the waters of naturally acidic
lakes are highly colored (due to the presence of humic substances)
and have high dissolved organic carbon content and therefore
have low transparency (Wetzel, 1975). Conversely, lakes pre-
sumably acidified only by anthropogenic acids have high trans-
parency and low dissolved organic carbon concentrations.
Furthermore, according to Havas et al. (1984), the major acids
in naturally acidic lakes are humic and fulvic acids, while
sulfuric and nitric acids are of most importance in anthropo-
genically acidified lakes. Because of this major difference,
the solubility of trace metals in the two lake types also
differs. The solubility of most trace metals increases under
low pH conditions; however, the concentrations of the metals
can be effectively reduced by binding to the humic materials
present in naturally acidic lakes. Because of the above dis-
tinctions, it can be concluded that the two extreme lake types
can be discriminated easily. On the other hand, the acid lakes
that fall between these extremes may have Inputs from both
sources. The Important question is what portion of the acidity
of the lakes in the middle of this spectrum can be attributed
to natural or anthropogenic sources.
Previous attempts at estimating the relative contributions
of natural and anthropogenic acids can be placed into two
categories*
e Determination of strong and weak acid contributions to
lake water acidity via acid-base titration procedures
-------
ANTHRQPOGENICALLY
ACIDIFIED
Strong mineral acids
(h2so4, hno3)
Low DOC
High transparency
High metals concentrations
NATURALLY
ACIDIFIED
Weak organic acids
(hu»icr fulvic)
High DOC
Low transparency
Low metals concentrations
Figure 1. Spectrum of lakes whose acidity is derived from anthropogenic
and natural sources.
-------
Martin Marietta Environmental Syttwni
• Watershed studies that estimate the mass balance of
hydrogen ions in a water body and its associated
terrestrial environment.
The first approach is discussed in detail in Chapter II of this
report. The second approach, has provided gross estimates of
the natural acid contributions to total acidity that range
from 50-95% (Marcus et al., 1983). The higher estimates
have been derived by Rosenqvist et al. (1980). Rosenqvist
(1980) contends from these estimates that the acidification of
lakes in southern Norway is due primarily to changes in land
use which can increase the input of organic acids to surface
waters. Richter (1983) and Krug and Frink (1983) also advocate
the importance of the sources of natural acidity, most notably
the organic acids produced in the formation of natural soils.
This position has been strongly criticized (e.g., Henriksen,
1984; Johnson et al., 1984; Seip and Dillon, 1984; Wright,
1984). Despite their opposing views on the relative importance
of naturally acidic inputs, Richter (1983), Krug and Frink
(1983), and Henriksen (1980) all indicate that further research
is required to allow the quantification of the contributions
of acids from various sources. Oliver et al. (1983) approached
this problem by quantitatively estimating the contribution of
humic substances to lake water acidity. However, an important
exception is indicated, i.e., comparability of data is strongly
influenced by differences in analytical procedures. Therefore,
the need for a standardized analytical method for the quanti-
fication of organic acids is apparent.
It is the objective of this report to provide a broad
over review of the currently available analytical methods for
the isolation, characterization, and quantification of naturally
occurring organic acids (e.g., humic and fulvic) in surface
waters. The underlying motivation for this report is to identify
an analytical protocol which would serve to quantify the contri-
butions of anthropogenically and naturally produced acids to
the process of lake acidification. It is intended that this
report be viewed as a management document which identifies
gaps in the knowledge of organic acid determinations that
require further attention. Figure 2 illustrates the organization
of the approach taken in the remainder of this report.
1-3
-------
Mature and Sources of
Lake Hater Acidity
Organic
Inorganic
Anthropogenic
Natural
Natural
Anthropogenic
Analytical
Methods
Isolation
and
Concentration
Acid Functional
Group Analysis
Contribution to
Lake Water Acidity
Figure 2. Conceptual approach to determining the contribution of
naturally produced organic acids to lake water acidity
-------
Martin Marietta Environmental Syttwm
II. NATURE AND SOURCES OF LAKE WATER ACIDITY
Acidity ia defined as the quantitative capacity of a water
to react with hydroxyl ions (OH") (APHA, 1980). The components
of lake water acidity that contribute to this capacity include
uncombined carbon dioxide, organic acids, mineral acids, and
the salts of strong acids and weak bases (Wetzel, 1975).
Operationally, acidity is measured by titrating a sample with a
standard alkali (e.g., NaOH) to pH endpoints of 4.5 and 8.3.
A potentiometric pH titration technique called Gran plot
titration has been used to determine the strong and weak acid
components of lake water acidity (Askne and Brosset, 1972} Lee
and Brosset, 1978; Henricksen and Seip, 1980). In this technique,
the amount of hydrogen ions (H+) measured in a sample solution
is plotted against the amount of OH" added to the solution.
The linear part of the plot is extrapolated to the abscissa,
and this intercept is assumed to be the point of equivalence
for the strong acid content of the sample. The slope of the
linear part of the plot can vary from sample to sample (Lee
and Brosset, 1978). This variation is assumed to reflect the
presence of weak acids, since H+ is liberated by weak acids
upon addition of OH". Therefore, the plot (Fig. 3) is determined
by the kinds and concentrations of weak acids present in the
sample. Brosset (1976) has developed an equation that describes
the slope of the line:
m ¦ the change in the H+ concentration in the sample
with respect to the unit amount of OH" added
C^ ¦ the total concentration of the ith weak acid
¦ a function of H+ concentration and the dissociation
constants of the weak acids.
When strong acids are predominant, i.e., when ZCi<>i
approaches or equals 0 the slope of the line equals -1.
Values of m less than -1 indicate the presence of weak acids.
Molvaersmyr and Lund (1983) discussed problems that may
be encountered in using the Gran plot titration. First, the
titration should encompass a wide pH range to ensure that all
weak acids are undissociated at the start (at pH ~ 3.6) and
l+lCiOi
1
where
II-l
-------
Martin Marietta Environmental System*
1.5x10
1.0x10
¦
H
2
SB
Strong Acid Content,
ADDED OB"
(aoles/ml)
Figure 3. Gran plot depicting the effect of the presence of
weak acids on the estimation of the strong acid
fraction.
II-2
-------
Martin Marietta Environmental Syttams
fully dissociated at the end (at pR - 10.3). Interpretation
of the results may also present problems. A negative concen-
tration of strong acids leads to the conclusion that the sample
water contains bases. This conclusion, however, does not
preclude any influence from acid deposition. Atmospheric
inputs of strong acids may protonate these bases, resulting in
relatively high concentrations of weak acids.
Henrickson and Seip (1980) assumed that weak acidity was
due to aluminum, total organic carbon (TOO, and silica. They
regressed calculated concentrations of weak acid (WA) on the
observed concentrations of these three variables in 202 samples.
The form of the regression equation was:
[WA] ¦ A0 ~ AilAl] ~ A2CTOCI + A3[Si02l •
The regression showed that aluminum and organic carbon were the
more important contributors to the weak acid content.
Many investigators (e.g., Pisher, 1984; Lee and Brosset,
1978) have assumed that the linear part of the titration plot
represents only the strong acid concentrations in the sample.
The linear part of the titration curve could include some
fairly strong "weak acid,* e.g., phosphoric (K^ * 1.6 x 10~2,
K2 - 7 x 10"®), sulfurous (Kj " 1.72 x 10~2, Kj " ®*24
x 1Q"8), and organic^acids such as oxalic (Ki ¦ 5.9 x
10-2, K2 " 6«4 x 10"5).
INORGANIC ACIDS
A number of both strong and weak inorganic acids may
occur in lake water, but the strong ones such as H9SO4 or
HNO3 usually predominate. Benricksen and Seip (1980), have
found that aluminum and silica, and (to a lesser extent) am-
monium, are the most important inorganic species that contri-
bute to the weak acidity of lake waters. Inputs of inorganic
acids can be from either anthropogenic or natural sources. In
areas not receiving direct inputs of industrial effluents, the
primary input of anthropogenically produced inorganic acids is
via atmospheric deposition. Natural processes that generate
inorganic acids include oxidation, hydrolyais, cation exchange,
or nutrient uptake (Dickson, 1980). Also, atmospheric CO*
dissolves readily in water where, at a pB of 8 or lower, it is
slowly hydrated to form carbonic acid (Wetzel, 1975).
II-3
-------
Martin Marietta Environmtntai Syttams
ORGANIC ACIDS
Most (50-80%) of the dissolved, naturally derived organic
carbon in surface waters consists of humic substances (Liao,
1981; Christman and Gjessing, 1983; Thurman and Malcolm, 1981).
The humic matter of surface waters is primarily allochthonous,
i.e., derived from the catchment basin, and is a highly hetero-
geneous mixture of compounds representing plant and animal
remains in various degrees of decomposition. This heterogeneous
humic material is generally considered to be composed of three
operationally defined fractions (Gjessing, 1976):
e Humin, which is insoluble in both acidic and basic
solutions
e Fulvic acids (FA), which are soluble in both acidic and
basic solutions
e Humic acids (HA), which are soluble in basic solution
but insoluble in acidic solution*
These three fractions are structurally similar, but differ in
molecular weight and functional group content; compared with HA
or humin, FA generally are of lower molecular weight and contain
a higher proportion of oxygen-bearing functional groups (Schnitzer
and Khan, 1972). The major functional groups found in HA and
FA are shown in Figure 4. The inhibition of bacterial degrada-
tion by these functional groups and their associated acidity
ultimately can lead to accumulation of humic materials and the
formation of the typical dystrophic or bog lake described
earlier.
The study of aquatic, as well as terrestrial, humic materials
has been hampered by the heterogeneity of these materials, and
a great deal of research effort has been directed at elucidating
the physical characteristics and the elemental and functional
group composition of humics. The molecular weights of humic
material vary substantially. Shapiro (1967) fractionated the
humic materials from a number of Minnesota lakes and found
four basic distributional patterns of molecular weight (Fig.
5). Gjessing (1965) and Gjessing and Lee (1967) observed that
aquatic humic materials contained fractions of differing molecular
weight, from <10,000 to >100,000. Ghassemi and Christman
(1968) found that the molecular weights generally ranged between
700 and 10,000, but they cautioned that the pH of the eluants
used in the fractionation procedure can greatly affect the
molecular size distributions observed. Typically, increasing
II-4
-------
Martin Marietta Environmental Syttami
CARBOXYL
METHOXYL
-COOH
— OCH.
PHENOL
KETONE
9
>c=o
QUINONE
ALKENE
-c=c-
I I
4. Functional groups of humic acids (HA) and
fulvic acids (FA).
IX-5
-------
Martin Marietta Environmental Syttams
.6 -
.4-
.2 -
3.
e
o
«n
to
M
e
• o _
o
o
o
Figure 5. Examples of the 4 types of elution patterns found
among 22 lakes in Minnesota (from Shapiro, 1967) .
II-6
-------
Martin Marietta Environmental Syttamt
the pH increases both the molecular sizes observed and the
range of sizes observed; the humic polymers probably extend at
high pH. Ghassemi and Christian (1968) also concluded that
slightly colored waters obtain most of their color from the
low-molecular-weight fraction, whereas in moderately and highly
colored waters, more color was found in the higher-molecular-
weight fraction.
The elemental compositions of HA and FA generally differ
(Schnitzer and Khan, 1972). Carbon content ranges from 401
to 50% for FA and from 50% to 60% for HA, and oxygen content
ranges from 44% to 50% for FA and from 30% to 35% for HA.
Visser (1982, 1983) found that in elemental composition, aquatic
FA and HA were closer to each other than to terrestrial FA and
HA. Beck et al. (1974), however, found that river-water organic
matter closely resembled soil PA in chemical composition.
The relative proportions of their various functional group
components greatly influence the chemical reactivity of humic
materials and have also been used to determine their source.
It is generally believed that as humic materials mature, aromatic
components become more condensed, oxygen- and nitrogen-bearing
functional groups are split, aromaticity increases, and unsatu-
rated aliphatic chains may coalesce (Degens, 1969). These
maturational changes would explain the observations of Stuermer
et al. (1978) that HA from algal sources were more aliphatic
than those from vascular plants. Algal remains generally are
more easily degraded than vascular plant remains, which contain
relatively higher fractions of lignin and cellulose. The
changes postulated above also support the conclusion, based on
the high aromaticity of the HA carbon in a Welsh lake, that
the primary HA source for the lake was the surrounding peaty
watershed (Wilson et al., 1981).
The acidic functional groups of aquatic HA and FA have been
widely studied in an attempt to better understand their role in
metal complex formation and in trihalomethane formation as a
by-product of water chlorination. The carboxyl group (COOH)
is more labile than the phenolic OH"* group and, therefore,
contributes more to the ability of HA and FA to form metal
complexes. Visser (1982) observed that aquatic HA contained
more COOH groups and fewer phenolic OH" groups than soil
HA. He also observed a decrease in the number of acidic func-
tional groups with increasing molecular weight in both HA and
FA. Table 1 sets out the COOH content observed in HA and FA
from a number of aquatic systems (Oliver et al., 1983). In
general, the range in COOH content of FA from the various
sources is small. Also, COOH content is higher in FA than
in HA. Similar relative COOH contents for HA and FA were
found by Visser (1982).
II-7
-------
Table 1. Operational carboxvl contents of fulvic
acids (FA)
and
h untie
acids (HA)
from various
locations in the
United States
and Canada
(from Oliver
et al., 1983).
Location
DOC
Carboxvl
Content
(ma/A)
(ueg/mg C)
Rivera and Streams
Spencer Creek (Ont.)
8.0
PA + HA
11.5
Missouri R. (LA)
3.8
PA
11.0
HA
9.9
Ohio R. (OH)
3.5
FA
11.4
HA
7.4
Tampa R. (CO)
2.0
FA
11.3
HA
7.7
Ogeechee R. (CA)
7.0
FA
10.4
HA
5.1
Shawsheen R. (MA)
7.0
FA
10.4
Coes Creek (CO)
6.4
FA
10.1
Bear Creek (CO)
0.7
FA
11.3
——
Lakes
Pebbleleggitch (MS)
14
FA + HA
8.3
Brainard L. (CO)
2.8
FA
10.7
Island L. (MB)
30
FA
13.4
Castle L. (OR)
14S
FA
10.6
....
Wetlands
Spagrus Bog (MB)
30
FA
9.9
Swannee R. (CA/FL)
32
FA
11.0
HA
8.2
Hawaii Marsh (HI)
12
FA
10.3
Theresa's Bog (MA)
30
FA
10.1
Alpine Boa (CO)
3
FA + HA
9.2
—
Groundwaters
Biscayne (PL)
13
FA
11.4
Canal (A2)
3.2
FA
10.4
—
-
-------
Martin Marietta Environmental Syitamt
Because of their contribution to the acidity of surface
waters, the acid functional groups of aquatic HA and FA are of
major concern. Figure 6 shows that at pH values < 7 the
carboxylic acid groups are more dissociated than the phenolic
acid groups. Therefore, for the remainder of this report emphasis
will be placed on the methods for the determination of the
concentration of carboxylic acid groups. However, there are
organic acids other than FA and HA which can contribute to lake
water acidity. Table 2 shows that hydrophilic acids can be as
abundant as the hydrophobic acids, of which humic substances
are a major component (Aiken, in press a). The hydrophilic
acids are relatively strong acids and may be a major component
of the organic acidity in lake water. Therefore, methods for
the determination of the acidity of these compounds will also
be examined.
II-9
-------
M
H
I
a
CARBOXYLIC acio
GROUPS
PHENOLIC
CROUPS
3.0 4.0 5j0 6.0 7.0 8.0 9.0 10.0 IIJO 12.0 13D
pH
1
HI
2
2
I
Figure 6. Dissociation of carboxylie acid and phenolic groups in organic
material as a function of pH (from Wilson and Kinney, 1977).
-------
Table 2. Dissolved organic carbon (DOC) fractionation data for selected freshwater
systems of the United States.(a)
Carbon
HydrophobicsHydrophilics
Sample
Location
Date
DOC
Acids
Bases
Neutrals
Acids
Bases
Neutrals
Black Lake,
NC
Nov
1981
8.3
3.4
0.1
1.5
0.7
2.6
0.0
Ohio River at
Cincinnati,
OH
June
1981
3.7
1.2
0.0
0.9
1.3
0.3
0.0
Missouri River
at Sioux City,
IA
Aug
1981
3.4
0.7
0.8
0.0
1.7
0.3
0.0
Swannee River
at Fargo, GA
Dec
1982
38.2
16.0
0.2
1.0
19.2
1.4
0.6
South Platte^)
River,
Denver, CO
Nar
1980
8.6
1.4
0.1
3.1
1.7
1.9
0.4
(*) From Aiken (in press a) unless otherwise noted.
(b) From Leenheer (1981).
-------
Martin Marietta Environmental Syttaim
III.
ANALYTICAL METHODS
Most studies on the isolation and identification of organic
acids in freshwaters have centered on the humic and fulvic acid
(HA and FA, respectively) fractions. As discussed previously,
the chemical properties and structural features of HA and FA
have not been well defined, but both HA and FA are generally
considered to be complex polymers that contain both carboxyl
(COOH) and phenolic hydroxyl (OH-) acidic functional groups.
Additional chemical and physical features used to describe HA
and FA have been discussed in previous sections. Procedures
for separating (hydrophobic acids) HA and FA and hydrophilic
acids from natural waters will now be reviewed, along with
analytical methods for measuring the acidic carboxylic functional
groups in HA and FA.
ISOLATION TECHNIQUES
Aqueous solutions containing trace amounts of organic
constituents as complex mixtures must be concentrated or the
constituents isolated in order to obtain a sufficient quantity
of organics for separation and subsequent identification.
Some of the concentration and isolation techniques that have
been used for analysis and toxicity testing of organic consti-
tuents in water samples are listed in Table 3. The major
objective of any concentration and isolation scheme should be
to provide a final product that is free from chemical impurities
(e.g., inorganic species and other organic compounds) and that
has undergone minimal chemical alteration.
A critical review of the methods that have been commonly
used to isolate and concentrate aquatic humic substances has
recently been presented by Aiken (in press a). His findings,
including advantages and disadvantages of different techniques,
are summarized below.
Filtration is the slowest step in the separation process
because of the large volumes of water generally processed to
obtain sufficient quantities of humic substances. Although
use of a 0.45-v filter to separate the sample into dissolved
III-l
-------
Martin Marietta Environmental Systems
Table 3. Methods used to concentrate and isolate organic
constituents £rom water samples.
Concentration Techniques
Isolation Techniques
Freeze concentration
Lyophilization
Vacuum distillation
Reverse osmosis
Ultrafiltration
Solvent extraction
Activated carbon
Ion exchange
XAD resin
Precipitation
Centrifugation
Gas stripping
III-2
-------
Martin Marietta Environmental Syitami
and particulate fractions is generally acceptable, it most
likely represents a compromise between flow rate and rejection
of clay minerals.
Filter performance depends on chemical composition, flow
characteristics, and uniformity of pore size. Aiken evaluated
a variety of filter types for these criteria and their inertness
relative to the isolated compounds and recommended, based on
flow characteristics, glass-fiber filter tubes (greater surface
area), particularly when processing large volumes of water.
The two most commonly used filter types are the 0.3u
Balston's* Nicrofiber Filter Tubes (glass) and the 0.45w
Sela Flotronic Silver Membrane Filters. Their advantages and
disadvantages are presented in Table 4 (Aiken, in press a).
The glass-fiber filter tube has the additional advantage of
lower cost.
CONCENTRATION TECHNIQUES
Following the initial separation of the sample into its
dissolved and particulate fractions, the dissolved fraction
must be concentrated. Methods used to concentrate aquatic
humic substances have been reviewed by Aiken (in press a)(Table
5) and are discussed belowt
e Vacuum Distillation, Freeze-drying (lyophilization)
and Freeze Concentration
These methods are not well suited for concentrating
large sample volumes. In addition, dissolved inorganics
are concentrated along with the organics, requiring
further treatment to separate the humic materials.
Freeze-drying has been used most efficiently in con-
junction with other concentration methods as the
final step in aquatic humic isolation. Samples should
be desalted prior to freeze-drying.
e Coprecipitation
Inorganic compounds, such as FeClg# Fe(0H)3, and CaC03,
have been used to isolate humics from water by
coprecipitation. This method is not quantitative, as
shown by low recoveries, and is impractical for
processing large volumes of water.
III-3
-------
Martin Marietta Environmental Syitam*
Table 4. Advantages and disadvantages of silver membrane
and glass £iber filter for filtration of water
for the isolation of organic acids (from Aiken,
in press a).
Matrix Type
Advantages
Disadvantages
Silver Membrane
Glass Fiber
Uniform pore size
Does not alter
composition of
dissolved organic
carbon
Good flow
Economical
Slow flow
Expensive
Particles
larger than
nominal pore
size can pass
filter
Slight sorptior
of certain
organic com-
pounds
III-4
-------
Martin Marietta Environmental Systems
Ultrafiltration (UF) and Gel Permeation Chromatography
Ultrafiltration and S«phadex gel chromatography have
been used primarily to determine the molecular size
distribution of humic substances. Their disadvantages
include membrane-solute interactions (during the UF
concentration) and irreversible adsorption (on the
Sephadex gels).
Reverse Osmosis
Although large volumes of water can be processed, the
method concentrates most solutes which then need to be
further separated from humic substances. This is also
an expensive, equipment-intensive method.
Solvent Extraction
Although inorganic salts can be effectively separated
from organic matter by this method, poor extraction
efficiencies and slow extraction rates are major
disadvantages.
Sorption Methods
Liquid column chromatographic methods employing various
different sorbents have worked well in isolating
aquatic humic substances. Large volumes of water
have been processed by these methods, resulting in
high concentration factors for the isolated material.
Until the development of synthetic macroporous resins,
both in nonionic and anion-exchange forms, however,
the efficiency of isolating and concentrating aquatic
humic substances has been limited by irreversible
sorption and low sorption capacities.
- Alumina, Carbon, Nylon, and Polyamide
Althouth used for the isolation of humic materials,
these sorbents have not been as effective as the
macroporous resin sorbents. Major advantages and
disadvantages are shown in Table 5.
- Ion-Exchange Sorbents
Numerous exchange resins having a variety of matrix
polymers are available. lonizable functional groups
are chemically bonded to a hydrocarbon polymer matrix
and have mobile ions that can react with, or be
replaced by, other ions. Solute behavior on the ion-
exchange resin is governed by the nature of these
functional groups.
III-5
-------
Martin Marietta Environmental Syitami
Table 5. Methods commonly used to isolate and
concentrate aquatic huraic substances
Aiken, in press a).
(from
Method
Advantages
Disadvantages
Vacuum
Distillation
Low temperatures
All solutes
concentrated
Freeze-Drying
(Lyophilization)
Mild
High concentration
factors
Method is slow
All solutes,
except for
volatiles, are
concentrated
Freeze Concen-
tration
Mild
Inexpensive
Simple
Method is slow
All solutes
concentrated
Coprecipitation
Inexpensive
Effective for waters
high in DOC
Efficiency
dependent on
initial DOC
Inefficient
with large
volumes of
water
Isolated or-
ganic matter
must be
separated
from inorganic
salts
Ultrafiltration
Organic solutes
fractionated by
molecular size
Possible mem-
brane inter
actions
Large volumes can
be processed
Possible mem-
brane fouling
Reverse Osmosis
Ambient conditions,
mild
Large volumes can
be processed
All solutes
concentrated
Efficiency
concentration-
dependent
III-6
-------
Martin Mariana Environmental Systami
Table 5. Continued
Method
Advantages
Disadvantages
Solvent Extraction
Sorption
Inorganic salts
effectively excluded
Humic sub-
stances in-
soluble in
many solvents
Method is
slow
—Alumina
Organic acids
readily sorb to
basic adsorbent
Inefficient
desorption
Mild eluents
Structural
alterations
of organic
matter possible
—Nylon and
Polyamide
Powder
Efficient adsorption
Irreversible
sorption
possible
—Carbon
Inexpensive
Irreversible
sorption
possible
Simple procedure
Slow elution
rates
Large volumes of
water can be
readily processed
Slow sorption
rates with
high molecular-
weight species
Organic blanks
are low
Chemical al-
teration of
organic
solutes
possible
—Anion Exchange
Method is simple
Irreversible
sorption
probable
a. Strong-Base
Resins
Large volumes can
be processed
Fouling of
resins
possible
III-7
-------
Martin Marietta Environmental System*
Table 5. Continued
Method
Advantages
Disadvantages
High capacities
Cor macroporous
resins
Resin bleed
All anions
concentrated
b. Weak-Base
Resins on
Amphoteric
Matrix
Method is simple
All organic
anions con-
centrated;
humic sub-
stances must
be isolated
from hydro-
philic acids
Large volumes can
be processed
Resin needs
extensive
cleanup
High capacities
for macroporous
resins
Resin bleed
Efficient desorp-
tion
Desorption
with NaOH
Inorganic salts
removed
—Nonionic
Macroporous
Sorbents
Method is simple
Irreversible
sorption
possible on
styrenedivi-
nylbenzene
resins
Resins easily
regenerated
Desorption
with NaOH;
precautions
required to
prevent oxi-
dation of
humic sub-
stances
Large volumes can
be processed
Resin bleed
III-8
-------
Martin Marietta Environmental Syttemi
Table 5. Continued
Method
Advantages
Disadvantages
Large volumes can
be processed
High capacities
Resin bleed
Needs adjust-
ment to pH 2
prior to
absorption
III-9
-------
Martin Mariana Environmental Syftam*
Large volumes of water can be processed by these
methods and the resins can be easily regenerated;
however, extensive resin preparation is required
to minimize organic bleed. Also, since all organic
anions may be concentrated (on the anion exchange
resin) additional separation o£ huraic substances
may be required.
Cation exchange resins are not used for isolating
aquatic humic material because humic material is
strongly anionic and sorption, therefore, is poor.
These resins are used, however, to hydrogen-saturate
humic substances (convert them to free acid form)
as an intermediate step in other isolation schemes
(discussed later). This procedure removes trace
metals and other cations from the humic material.
Weak anion-exchange resins have been found to be
the most efficient ion-exchange sorbent for aquatic
humics. These contain weak-base functional groups,
macroporous structure, and a hydrophobic matrix.
Duolite A-7, a phenol-formaldehyde anion exchange
resin, has been shown to be effective not only for
isolation and concentration of aquatic humic sub-
stances, but for the hydrophilic acids as well
(Leenheer, 1981).
The sorption of organic anions on these resins is pH
dependent (by hydrogen bonding and ion exchange)
with maximum sorption occurring in the pH region
in which both resin and solute are uncharged (occurs
between pH 5.5 and 10 for the resin). By ionizing
both the anionic organic solutes and the anionic
resin, efficient desorption is attained (due to
charge exclusion). Further isolation of humic
materials, however, is required since all organic
anions are concentrated in weak-base exchange resins.
Diethylaminoethyl cellulose (DEAE), a weak anion ex-
changer, has also been used for concentrating
aquatic humic substances (Miles et al., 1983).
However, a much lower exchange capacity and poor
flow characteristics are major disadvantages.
- Nonionic Macroporous Sorbents
The nonionic macroporous Amberlite XAD resins
have been evaluated (Aiken et al., 1979) and
used most extensively for the isolation of
humic substances in freshwaters (Weber and
Wilson, 1975; Thurman and Malcolm, 1981; Liao,
111-10
-------
Martin Marietta Enviroiuiwritai Symnw
1981). These resins have been found to give
high recoveries of organic compounds from water
because of their high adsorption capacities and
ease of elution.
The sorption mechanism for these resins is the
hydrophobic effect where the nonpolar carbon
skeleton of the organic solutes adsorb to the
resin (nonpolar adsorption). Polar functional
groups (hydroxyl and carboxylic groups) are posi-
tioned in the aqueous phase and work to desorb
the molecule. The affinity for sorption of
solutes is determined by the solubility of the
solute and the solution pH (Thurman et al., 1978).
Adsorption and desorption can be controlled by
balancing the nonpolar and polar interaction.
XAD-8 (an acrylic ester polymer resin) was recom-
mended over four other XAO resins for concentrating
fulvic acids in freshwater (Aiken et al., 1979).
The acrylic ester polymer XAD resins (XAD-8 and
-7) were found to have higher adsorption capacities
for concentrating fulvic acids and were more
easily eluted than the styrene divinylbenzene
copolymer XAD resins (XAD-1, -2, and -4).
XAD-8 also had fewer bleed problems than XAD-7
during fulvic acid elution with NaOH.
It is apparent that there is no single technique that can
achieve quantitative isolation of similar groups of compounds
for testing and analyses. None of the methods discussed previously
can isolate humic substances (HA and FA) free from
inorganic salts and low molecular-weight organic acids. However,
correct combination of various techniques into a comprehensive
analytical procedure can result not only in quantitative isolution
of HA and PA (hydrophobic acids) but of the hydrophilic organic
acid fraction present in natural waters also. Such a scheme
has been devised by Thurman and Malcolm (1981) to concentrate
and isolate aquatic humic substances, while Leenheer (1981) has
devised a comprehensive fractionation procedure for all organic
solutes in water.
A comprehensive analytical procedure for isolating the
hydrophobic organic acids, i.e., HA and FA, using XAD-8 has
been proposed by Thurman and Malcolm (1981) and reviewed by
Aiken (in press b). The scheme is outlined in Table 6. The
authors have successfully applied this procedure to both surface
and groundwaters for the quantitative isolation of HA and FA
that had low ash content.
III-ll
-------
Martin Marietta Environmental Systems
Table 6.
Extraction scheme using XAD-8 to concentrate
aquatic humic substances as devised by
Thurman and Malcolm (1981) (from Aiken,
in press b).
Step
1.
Pilter sample through 0.45-v silver-membrane
filter and lower pH to 2.0 with HC1.
Step
2.
Pass acidified sample through XAD-8 column;
aquatic humic substances adsorb to resin.
Step
3.
Elute XAD-8 resin in reverse direction with 0.1 N
NaOH; acidify immediately to avoid oxidation of
humic material.
Step
4.
Reconcentrate on smaller XAD-8 column until DOC is
greater than 500.
Step
5.
Adjust pH to 1.0 with HC1 to precipitate humic
acid. Separate humic and fulvic acids by
centrifugation. Rinse humic acid fraction with
distilled water until AgNO^ test shows no Cl~.
Dissolve humic acid in 0.1 N NaOH and hydrogen
saturate by passing solution through cation-
exchange resin in H-form.
Step
6.
Reapply fulvic acid fraction at pH 2 to XAD-8
column. Desalt fulvic acid by rinsing column with
1-void volume of distilled water to remove HC1 and
inorganic salts; elute fulvic acid in reverse
direction with 0.1 N NaOH.
Step
7.
H-saturate fulvic acid fraction by immediately
passing 0.1 N NaOH eluate through cation-exchange
resin in H-form. Continue cation-exchange process
until final concentration of Na+ is less than 0.1
part per million.
Step
8.
Freeze-dry humic acid and fulvic acid fractions.
111-12
-------
Martin Marietta Environmental Syttwro
Before using this procedure to compare results from dif-
ferent sources of natural waters, some important considerations
must be made. The DOC content of the sample must be determined
at different points during the subsequent concentration steps
(Table 6). Efficient separation of the hydrophobic HA and FA
fractions from the hydrophilic fraction of organic solutes in
the sample is controlled by the degree of solute polarity and
by the ratio of the resin quantity to the volume of water
passed through the resin bed (Leenheer, 1981). For a given
volume of resin and a known column capacity factor, the volume
sample to be processed can be calculated to ensure 100% retention
of the solutes of interest (hydrophobic acids).
During the reconcentration step (Step 4), the resin-to-
sample volume ratio will change depending on the input DOC.
Although Thurman and Malcolm (1981) present XAD-8 column sizes
necessary to recover all the hydrophobic acids as a function of
DOC, the authors recommend monitoring the effluent during
reconcentration for DOC breakthrough.
Although a 0.45-u silver membrane filter is proposed in
Table 6, a glass-fiber filter is recommended (Aiken, in press
b).
Multiple reconcentration steps on XAD-8 are required for
samples with low DOC values, and concentration factors of 25,000
for both HA and FA are possible (Thurman and Malcolm, 1981).
Leenheer (1981) has devised an analytical protocol referred
to as DOC fractionation analysis that quantitatively classifies
organic solutes into hydrophobic-acid, -base and -neutral fractions,
and hydrophilic-acid, -base and -neutral fractions based upon
their adsorption on nonionic (XAD-8) and ion-exchange (Duolite
A-7 and BioRad A6 MP50) resins. This procedure was used to
separate the organic fractions from water and inorganic salts
and final recoveries of about 814 of the DOC were achieved.
The flow chart of the preparative scheme is shown in Figure 7.
This relatively complicated procedure combines rotary
evaporation, centrifugatlon, ion exchange, and adsorption
chromatography to separate each fraction. Extensive resin
cleanup procedures must be performed and additional parameters
(such as specific conductance, DOC, pH) must be measured during
several of the fractionation steps.
Zn addition to all the organic acids, all inorganic anions
are also concentrated on the A-7 resin and are coeluted from
the column with the organic acids. This eluate is fractionated
into hydrophobic and hydrophilic acid by adjusting the pH
and passing it through XAD-8, which retains the hydrophobic acid
111-13
-------
»I«D I
riltared iMfl* followed
by diet Ilied water rlnaa
Hydrophobic
l^eea
Ittol
1411 lad
faea ac
through col
by l.ll NCI
eanple
followed
taborlite Js^Heafa. thee
•oihlat attract kydrofkokte
•eitrili with aathaeol
Adjuat to pa 2
with act
|U|J
Iwahelut
Backflwah alwta with
•.I N NCI
I
«
I
I
I
I
Hydrophobic
acida
A
•
I
I
I
rllte uo-l tad*
»tag 4
Beokflueh elate with 1.1
NaON followed by dlatllled water
tea* H*(le through coltewi
M
M
I
!|SB_Z
forward elute with
1 N NN4M
¦ydrophillc baeea* - i
1
I
1
1
i
-„.L
¦to-Bad M NP M
hydrogen ioa aataratad
cation eachaage reala
Funp
Itap >
atapla ikroagk
coluan
Nydrophlllc actda plea* •
Inorganic anaonlun aalta
Step > -
Backflueh alwta with I N
NN«BM
•
•
•
i
•
«
•
•
•
I
_ i
Ouollte A-7 aeloa eachange
reala In freebaee (ora
Nydrophlllc neutrala In
detoolted water
Figure 7. Analytical procedure for preparative dissolved
organic carbon fractionation (fro* Leenheer, 1981).
-------
Martin Marietta Environmental Syttam*
fraction. The hydrophobic acids are desalted on XAD-8, hydrogen-
saturated, and freexe-dried according to Thurman and Malcolm
(1981). Desalting the hydrophilic acids is more difficult and
studies are currently being conducted to improve efficiency.
A modification of the fractionation procedure using
Duolite A-7 for organic acid concentration and isolation (Fig.
8) has been proposed by Leenheer and Noyes (cited in Aiken, in
press b). Comparisons of this method to the extraction scheme
for HA and FA using XAD-8 discussed earlier has been done in
actual field tests. The Duolite A-7 method was more efficient,
being able to process 8100 1 of water in 7 days, whereas 10,400
t of water required 60 days as a result of the slow filtration
rate through the silver membrane. Overall recoveries (based
on DOC) of 97% and 59% for the hydrophobic and hydrophilic
acid fractions, respectively, have been reported (Leenheer,
1981). Recovery data using representative hydrophilic and
hydrophobic organic acids as standards have not been tested.
Precision and accuracy data for that study were not generated.
The DOC measured in resin bleeds and reagent blanks did not
exceed 1.0 mg/l.
Future needs for continued development of these comprehensive
isolation procedures Include calibrating the fractionation
procedures with a wide variety of representative organic acid
standards and testing their applicability with a wide variety of
waters. In addition, Leenheer (1981) suggests that the variability
of manufactured resin adsorbents also be better assessed.
SAMPLE PRESERVATION
Concerns regarding sample preservation have not been
adequately addressed. Naturally occurring organic acids are
subject to both biological and chemical degradation. Aiken
(in press b) recommends filtering the sample immediately after
collection and processing it as soon as possible. The filtered
sample should be kept cold if processing delays are anticipated.
Chemical degradation of organic acids can also occur, e.g.,
oxidation at the elevated pB of the solutions that are used to
•lute the fractions from the resin. Therefore, the eluate
should be adjusted to pH 2 (and refrigerated) if sample processing
is interrupted (Aiken, in press b).
111-15
-------
Martin Marietta Environmental Syitam*
Water sample
Filtration
Suspended sediment-*
Hydrophobic basest
Weak hydrophobic acids*
Hydrophobic neutrals**
Amberlite XAD-8 resin
MSC-1 hydrogen-ion-saturated
cation-exchange resin
Hydrophilic bases-*
Duolite A-7 anion-exchange-
resin in free-base form
Strong hydrophobic and-*
hydrophilic acids
Hydrophilic neutrals
in deionized water
Pigure 8. Fractionation of organic solutes in water by the
method of Leenheer and Noyes (cited in Aiken, in
press b).
111-16
-------
Martin Marietta Environmental Syttami
ACIDIC FUNCTIONAL GROUP ANALYSIS
Most of the procedures used for identifying acidic functional
groups, i.e., measuring their concentration and their acid
dissociation properties in humic substances, have been performed
to evaluate metal-humic ligand equilibria (Schnitzer and Khan?
1972j Stevenson, 1982| Plaig et al., 1975; Reuter and Perdue,
1977i Perdue, 1979t Perdue, 1978> and Perdue et al., 1984).
However, these same methods can be used to assess the organic
acid contribution to aqueous acidity.
The main organic acid functional groups that are likely to
contribute to acidity of natural water are the carboxyl groups
(COOH), which have pK values less than 7 (Oliver et al., 1983).
Although a number of different methods have been used to determine
the concentration of carboxylic functional groups of humic
substances, an evaluation of these procedures as applied to
aquatic humus (Perdue, 1979; Perdue et al., 1980) concluded
that carboxyl determinations should be conducted in the absence
of polyvalent cations (e.g., Ca-acetate, Ca(N03)2)*
Oliver et al. (1983) have used a direct potentiometrie
titration (with 0.1 N NaOH) to determine the acidic carboxyl
group content of isolated and concentrated aquatic HA and FA.
Titrating the sample to pH 5 or 7 was assumed to provide a
reasonable estimate of the acidic carboxyl groups. From the
information provided by this analysis, they quantified the
acidic functional groups in aquatic humics that could potentially
dissociate in natural water at pH below 7. As shown earlier
(Table 1), the FA catboxyl content determined by this method
was similar among aquatic systems, ranging from 9.9 to
13.4 weq/mg DOC, with a mean of 10.7. The HA carboxyl content
was lower, ranging from 5.1 to 9.9 ueq/mg DOC (x • 7.7, std.
dev. ¦ 1.7). A weighted mean for all HA and FA was 10.5 ueq/mg
DOC (std. dev. - 1.0). This average carboxyl content differs
from the 20 weq/mg DOC suggested by Schnitzer (1978) for a
"model" soil FA, from the value of 5.5 ueq/mg DOC calculated
by Henricksen and Seip (1980), and from the values obtained by
Visser (1982), 13.2-17.4 ueq/mg DOC and 11.4-14.6 ueq/mg
DOC for FA and HA, respectfully. However, this study is very
important in terms of developing a generally applicable analytical
method to determine an operationally defined carboxyl content
for humic substances. As indicated by Perdue et al. (1980),
two criteria must be met for such a method to be considered
generally applicable:
e it must yield consistent results on a given
sample
111-17
-------
Martin Marietta Environmental Syttam*
• It must always measure the same quantity on different
samples, so that comparisons between samples are valid
and meaningful.
Purther work using the procedures employed by Oliver et al. is
needed to determine if these criteria are satisfied.
CONTRIBUTION TO LAKE
WATER ACIDITY
Quantitative evaluations of the contribution of aquatic
humic substances to acidification requires reliable information
on the identities, concentrations, and acid dissociation
properties of the acidic functional group. Isolation of a
chemically pure fraction of aquatic HA or PA has been discussed.
Methods for determining the concentrations of the acidic
functional groups have also been discussed. The final step in
estimating the contribution of organic acids to lake water
acidity is to determine their degree of dissociation. The
data derived from potentiometric titration curves have been
analyzed mathematically to determine dissociation constants
for HA and FA. Mass action quotients (K) have been derived
from potentiometric titrations to describe the dissociation of
fulvic acid (Gamble, 1970; Perdue et al., 1980j Oliver et al.,
1983). The term K is a weighted average of all the individual
dissociation constants (K^) for acid ionization in a heterogeneous
mixture of distinct non-identical functional groups (AjH)*
I IA1-HH+]
i-1 lA-JIH+1
K - - —
n ((CJ-IA-J)
I fAiH]
i-1
where
{Aj"! • acidic anion concentration
[A~l * concentration of ionized carboxyl groups
[CT] ¦ concentration of total carboxyl groups.
The values of K are^highly affected by pH. Pigure 9 shows the
change in pR (-log K) as a function of pH observed by Oliver et
al. (198JJ) for samples from two water bodies. The relationship
between pK and pH was nearly identical for the two samples.
Comparing their results with those of Gamble (1970) and Perdue
111-18
-------
Martin Marietta Environmental Symnw
m
PH
Figure 9. Relationship between pK and pH for PA from two
aquatic systems (from Oliver et al., 1983).
111-19
-------
Martin Marietta Environmantal SyitWM
*• a . (1980), Oliver et al. found very close agreement and
as a result propose that the "K of FA and HA can be estimated by
the equation:
pit - 0.96 + 0.90 pH - 0.039(pH)2.
They conclude that estimating the contribution of HA and FA
to the acidity of natural water, therefore, can be accomplished
by knowing their respective concentrations and the pH of the
sample. Since the carboxyl content of HA and FA can be inferred
from the DOC concentration ([COOHJ ~ 10*mgDOC/i), the
measurements required for estimating the acidic contribution
of HA and FA are pH and DOC.
111-20
-------
Martin Mariana Emrironntantal Syatam*
IV. CONCLUSIONS
It is quit* clear that the analysis of humic substances is
a non-trivial task. There are a large number of analytical
techniques that can be used to isolate and concentrate HA and FA
from aquatic samples. There also are a numerous approaches to
determining the acidic functional group content of humic sub-
stances. The International Humic Substances Society (Josephson,
1982) has proposed the following goals to aid in developing of
a routine protocol for the analysis of humic substances:
e Establish standards for the collection of HA and FA
samples
e Establish a standard isolation procedure for HA and FA
e Establish standard and reference samples of HA and FA.
Clearly, the work of Aiken in evaluating the isolation and
concentration techniques is a major step forward in the study of
aquatic humic substances. Continued work in the development of
a standardized procedure is planned. However, the need for an
analytical procedure for assessing the contribution of organic
acids to lake water acidity by the National Acid Precipitation
Assessment Program (NAPAP) is very pressing. Previous approaches
to estimating the contribution of organic acids to lake water
acidity, i.e.. Gran plot titration and the study of mass balance
in watersheds, do not fill this need. Data gathered via Gran
plot titration are not always comparable and there is the
potential for mistaking a strong organic acid, such as the
hydrophilic acids, for a strong mineral acid using this method.
Mass balance studies are clearly not suitable for monitoring
purposes due to the level of effort that must be employed for
even a single watershed. Therefore, alternative approaches
need to be considered.
Before doing so, however, criteria for determining the
suitability of these alternative approaches for the needs of
NAPAP must be defined. Some of these criteria includei
e Applicability to waters within a wide range of chemical
conditions, Including, most notably, low to high ionic
strength, low to circumneutral pH, and low to high
trace metal concentrations.
e Feasibility for use in large surveys, such as the National
Surface Hater Survey and long-term monitoring programs,
which would require large numbers of samples to be
iv-l
-------
Mutin Marietta Environmental Syttanu
processed# some of which would have been obtained from
extremely remote areas.
e Provision of levels of precision, accuracy, and sensitivity
to satisfy the requirements of the particular programs
that may be served.
Although the approach employed by Oliver et al. (1983)
appears promising, several aspects of their method require
further work* It is not clear whether the isolation and concen-
tration techniques employed are entirely suitable for waters
with a wide range of DOC. Isolation and concentration techniques
must provide quantitative i.e., measureable, recoveries, which
are nearly complete and reproducible. The use of 10 ueq/mg
DOC factor for estimating the acidic functional group (COOH)
content of the humic substances may not be fully warranted.
This factor did not fully agree with those derived by other
researchers and these differences may be due to inconsistencies
in the analytical methods employed or in the application of
data obtained from purified humic material as opposed to actual
water samples (Oliver et al., 1983). A detailed justification
of the use of a particular factor is warranted. Application
of the relationship observed between pH and the dissociation
constants obtained in the laboratory on purified samples may
also be dangerous. In situ lake-water conditions may likely
include high trace metal concentrations which may affect the
ionization of HA and PA.
There is one major gap in the knowledge needed to assess
the contribution of naturally occurring organic acids to lake
water acidity. The hydrophilic organic acids have been shown
to be present at concentrations very similar to those of the
hydrophobic acids (HA and PA) (Aiken, in press a). To date,
there appears to have been little or no consideration of the
relative contribution of these compounds to surface water
acidity. The utility of the isolation and concentration tech-
niques proposed by Leenheer (1981) for the hydrophilic acids
must be assessed. Also, the applicability of the potentiometric
titration techniques or other such techniques for determining
dissociation constants of the hydrophilic acids must be considered.
IV-2
-------
Martin Marietta Environmental Syitamt
V. ANNOTATED BIBLIOGRAPHY
A. GENERAL
Beck, K.C., J.H. Reuter, and E.M. Perdue. 1974. Organic and
inorganic geochemiatry of aoma Coaatal Plain rivara of the
aoutheaatern United Statea. Geochim. Cosmochim. Acta. 38;
341-364.
Objectives To perform a comprehenaive study of the distribution
of the organic and inorganic constituents of aeveral
rivers flowing through the Coastal Plain of southeast
Georgia.
Findings! Streams in the Coastal Plain region of Georgia,
especially the Satilla River, were characterized
by low suspended load, ionic strength, and pH, and
a predominance of organic constituents.
Sodium and chloride were dominant among the inorganic
ions, indicating that rainfall greatly influenced
the diatribution of major elements in these waters.
Low pH values in the Satilla River were the result of
a predominance of acidic organic matter in the
water.
The authors believe that, "The degree to which organic
matter will influence the overall chemiatry of a
surface water will depend not so much on the absolute
amount of organic matter as on the organic-inorganic
matter ratio."
Chemical analyses (which included elemental and
functional group analysis) and both infrared and
nuclear magnetic resonance spectroscopy indicate
that the river water organic matter is chemically
similar to aoil fulvic acida.
******
Da Baan, B. 1983. Use of ultraviolet spectroscopy, gel filtra-
tion, pyrolysis/mass spectrometry and numbers of benzoate-
metabolizing bacteria in a study of humification and
degradation of aquatic organic matter. Ins Aquatic and
Terrestrial Humic Materials. R.F. Christman and E.T.
Gjessing, Eds. Ann Arbor Sciences Ann Arbor, MI. pp
165-182.
V-l
-------
Martin Marietta Environmental Syitams
Objective* To study the humification and degradation of fulvic
acids by benzoate-metabolizing bacteria.
Findings* E2/E3 ratios obtained from Sephadex gel filtration
and UV spectroscopy is a simple and easy way to
determine the measure of the relative humification
of colored lake waters. Progressive humification
is predicted by relatively low E0/E3 ratios and
coincides with increasing molecular weight and
aromaticity of fulvic acids. Increasing E2/E3
ratios were believed to indicate fulvic acid degra-
dation and this was confirmed by incubation experi-
ments with bacteria enriched on benzoate.
The degradation of fulvic acids by benzoate-metabolizing
bacteria was believed to be the reason for increasing
E2/E3 ratios and decreasing mean molecular weight of
fulvic acids observed in Lake Tjerikemeer during the
summer*
Easily degradable substrates such as lactate and
benzoate appear to stimulate the degradation rate
of fulvic acids, suggesting that the rate of fulvic
acid degradation is related to the productivity of
colored lakes.
Ghassemi, N. and R.P. Christman. 1968. Properties of the
yellow organic acids of natural waters. Limnol. Oceanogr.
13*583-597.
Objective* To investigate the molecular size distribution of
the yellow organic acids of natural waters using
Sephadex gel filtration.
Findings* Results of the Sephadex data from six water samples
from Washington, Vancouver, British Columbia, and
Alaska indicate that, except for a fraction of the
sample having a molecular weight in excess of
50,000, color molecules are apparently in the
700-10,000 mol wt range. Interaction of color
acids with the gel, however, could not be determined
and results of this study cannot be considered as
a true representation of molecular size until they
are verified with another method. Furthermore, the
effect of sample concentration on molecular size
distribution is not yet known.
The effect of using eluents of different pH and
ionic composition was also examined and found to
affect the molecular size distributions observed.
V-2
-------
Martin Marietta Environmental Syttams
Iron association with color acids was found to be
highest in both low and high pH conditions. It
was lowest at a pH of 7-8. The data, however,
contradict the finding of Shapiro (1964) that the
iron-holding capacity of color increases with pH
to a maximum value at pH 10 and then decreases
rather sharply.
Strong chelating agents were used as eluents of
Sephadex and, with the exception of a snail amount
of iron bound to the large molecular weight fraction
of the sample, most of the iron could be pulled
away from the color molecules.
******
Perdue, E.M., J.H. Reuter, and R.S. Parrish. 1984. A statistical
model of proton binding by humus. Geochim. Cosmochim. Acta.
48 *1257-1263.
Objective* To develop and present a statistical model that
reasonably describes proton binding by aquatic humus.
Findings* A statistical model was presented and data from the
Satilla River was used to demonstrate its capabilities.
Nonlinear regression analysis of titration data was
used to estimate the concentration, mean log K value,
and variance of the log K distribution for each class
of functional groups in the model. The values
obtained for these parameters appear to be "chemically
reasonable," and an accurate description of the
extent of proton binding between pH 4.0-10.8 is
provided by the model.
******
Plechanov, N., B. Josefsson, 0. Dyrssen, and K. Lundquist.
1983. Investigations on humlc substances in natural waters.
Int Aquatic and Terrestrial Humic Material. R.P. Christman
and E.T. Gjessing, Bds. Ann Arbor Science* Ann Arbor,
Ml. pp. 387-405.
Objective* To isolate fulvic and humic acids from a polluted
and unpollutsd source using both XAD-7 adsorbent
and anion exchange resin, and to perform systematic
physiochemical analysis to study the structure and
properties of water-soluble humic substances.
V-3
-------
Martin Marietta Environmantai Syttama
Findings* Comparison of anion exchange resin and XAD-7 to
concentrate water-soluble humic substances appear
to result in quantitative rather than qualitative
differences in the composition of the materials
isolated.
Water-soluble humic substances isolated from two
Swedish lakes appear to originate at least in part
from lignin. In polluted waters, the lignin component
might be derived from lignosulfuric acids.
In addition to lignin-derived components, *H-NMR results
suggest that water-soluble humic substances contain
varying amounts of a second group of materials, charac-
terised by the presence of alkyl groups.
Humic and fulvic acids has similar molecular weight
distributions.
Reuter, J.H. and E.M. Perdue. 1977. Importance of heavy metal-
organic matter interactions in natural waters. Geochim.
Cosmochim. Acta. 41*325-334.
Objective:
Findings*
To outline and review available information on the
abundance and molecular nature of organic matter in
natural waters as well as the nature of the metal-
organic interactions.
Biopolymers (e.g., polypeptides, polysaccharides)
and geopolymers (humic substances) were the major
components of the dissolved organic carbon compounds
in natural waters. Of these, biopolymers make up
only a small fraction of the dissolved carbon in
natural waters, and humic substances were the main
contributors to dissolved organic carbon.
The majority of dissolved humic substances in river
systems resembled soil fulvic acids. A small fraction
(<1G%) consisted of higher-molecular-weight material
which more closely resembles soil humic acids.
Average total organic content of major rivers in the
U.S. ranges from 3 to 20 mg/ l -1.
The main contributor of dissolved humic substances to
rivers is believed to be soil organic matter. In some
local areas, however, man-made contribution from
secondary sewage treatment may be significant.
V-4
-------
fthrtin Marietta Environmental Syitams
The acidic character of aquatic humus allows it to
interact with heavy metal cations. The stability
of humic-metal complexes in natural waters is
higher than that of the corresponding inorganic-metal
complexes. The extent to which heavy metals are
complexed with aquatic humus will depend on the
concentration of humic substances and the competition
for available complexing sites between trace metals
and major cations.
Stuermer, D.H., K.E. Peters, and I.R. Kaplan. 1978. Source
indicators of humic substances and proto-kerogen. Stable
isotope ratios, elemental compositions and electron spin
resonance spectra. Geochim. Cosmochim. Acta. 42t989-997.
Objective: To examine a number of chemical properties of humic
and proto-kerogen substances isolated from very
different environments and to determine how the
chemical composition of the material is a reflection
of its source.
Findings* Humic acids and proto-kerogens from algal sources
were higher in nitrogen and more aliphatic than
those from vascular plants.
Oxygen content appears to be a measure of maturation
of organic matter, while sulfur content may reflect
the redox condition in which the material was
deposited.
Electronic spin resonance data indicated that the
transformation of humic substances to proto-kerogen
in sediments was accompanied by an increase in
aromaticity.
A combination of 4*3 and H/C ratios may be a simple
and reliable indicator which can be used to determine
if humic matter was formed from algal sources or
vascular plants.
Stable nitrogen isotopes appear to be a valuable
indicator of nutrient sources of the humic substances.
Stable hydrogen isotopes appear to be a crude
reflection of latitudinal and altitudinal differences
but are not a reliable source indicator.
V-5
-------
Martin Marietta Environmental Systama
*
Visser, S.A. 1982. Acid functional group content of aquatic
humic mattert Its dependence upon origin, molecular
weight and degree of humification of the material.
J. Environ. Scl. Health, A17j767-788.
Obiectivej To studv and measure acidic functional groups from
3
-------
Mutin M»ri«tt» Environmental Systwm
Carbon, hydrogen, and nitrogen content of Culvic and
humic acids all were inversely related to molecular
weight; oxygen content and C/H ratio showed a
direct relationship to molecular weight.
In temporal studies from May through September, carbon
content of aquatic humus was found to increase, while
oxygen content diminished.
In six-month laboratory humification experiments,
fulvic acids became oxidized to such an extent
that there were decreases in carbon, nitrogen, and
hydrogen contents. These findings contradict the
generally accepted belief that carbon and nitrogen
content of humic material increases during humification.
Visser, S.A. 1983. Application of Van Krevelen's graphical-
statistical method for the study of aquatic humic material.
Environ. Sci. Technol. 17t412-417.
Obiective: To use Van Krevelen's diagrams to study differences
in the structure and composition of humic matter
obtained from various natural aquatic environments
and from microbial cultures.
Findinas: Elemental analysis of different molecules weight
fractions of both fulvic and humic acids indicated
that carbon, hydrogen, and nitrogen content was
highest in the lower molecular weight fractions.
In contrast, oxygen content increased in higher
molecular weight fractions.
Humic and fulvic acids isolates from natural waters
were found to be more similar to each other than those
from terrestrial environments. Aquatic humus was also
less aromatic than soil humus, and in the case of fulvic
acids, richer in oxygen.
It is generally believed that humification normally
results in the formation of a more aromatic type of
humic matter. In laboratory culture experiments of
fulvic acids, however, aromaticity was found to
decrease with progressive humification.
Routine sampling indicated that aquatic humus
collected at the end of the summer was in a more
advanced state of humification than samples collected
at the beginning of the summer.
V-7
-------
Martin Marietta Environmantal Syttam*
Recently £ormed humic matter, especially fulvic acids,
were more sensitive to oxidation from ultraviolet
irradiation than more humified materials.
******
Wilson, M.A., P.P. Barron, and A.H. Gillam. 1981. The structure
of freshwater humic substances as revealed by 13C-NMR spectro-
scopy. Geochim. Cosmochim. Acta 45:1743-1750.
Objective!
Findinasi
To characterize humic substances from Lake
North Wales, using 13c-NMR spectroscopy in an attempt
to determine its origin.
Lake Celyn humic acid carbon was found to be 24%
csrboxyl and 40% aromatic. The high proportion of
aromatic carbon suggests the Lake Celyn humic
acids were largely formed from terrestrial humic
substances in the surrounding peaty watershed.
B. METHODS
Aiken, G.R., E.M Thurman, R.L. Malcolm, and H.P. Walton. 1979.
Comparison of XAD macroporous resin for the concentration
of fulvic acid from aqueous solution. Anal. Chem. 51:1799-
1803.
Objective* To examine and compare the effectiveness of five
macroreticular, nonionic Amberlite XAO resins used
to concentrate and isolate fulvic acids from aqueous
solutions.
Findings: Acrylic ester resins XAD-7 and XAD-8 were found
to be the best overall resins for concentrating and
isolating fulvic acids from natural waters. These
resins attain equilibrium more rapidly, have higher
absorption capacities, and are more efficiently
eluted than the three styrene divinylbenzene resins
(XAD—1, XAD-2, and XAD-4) that were tested. XAD-7,
however, did exhibit some bleeding problems, making
XAD-8 the resin of choice for isolating and concen-
trating fulvic acids from natural waters.
******
Alberts, J.J. 1982. The effect of metal ions on the ultraviolet
spectra of humic acid, tannic acid, and lignosulfonic acid.
Water Res. 16:1273-1276.
V-8
-------
Martin Marietta Environmental Syrtam*
Obiectlv». To study the effect of three metals (iron, aluminum,
and manganese) on the ultraviolet spectra of humic
acids, tannic acids, and lignosulfonic acids.
Finding*« The interaction of organic acids with metal ions,
especially iron, can change the spectral characteristics
of the organic material. If spectrophotometry tech-
niques are used to quantify these materials, the metal
organic interactions could result in false estimates.
Concentration techniques such as evaporation, which
concentrate inorganic salts as well as organic matter,
could promote additional interactions between metals
and organic matter daring the preparation of a sample
for analysis. This would further complicate attempts
to quantify organic matter concentrations using
ultraviolet spectroscopy.
Brun, G.L. and 0.0.L. Milburn, 1977. Automated fluorometric
determination of humic substances in natural water. Analyt.
Lett. 10*1209-1219.
Objective:
Findings:
To describe an automated procedure for fluorescence
spectrophotometrie analysis of humic substances in
natural fresh water samples.
An automated fluorescence spectrophotometry analysis
procedure using a Technicon "Auto Analyzer" II
System was described. The method could be used for
routine monitoring programs where high numbers of
samples need to be analyzed.
A tartrate-citrate complexing agent and a buffer line
were added to the system's manifold. The buffer line
increased the pH of the water sample to around 10,
which was found to increase sensitivity and prevent
precipitation of humic substances. Tartrate-citrate
complexing agents eliminated interference from high
levels of iron in the sample and prevented the pre-
cipitation of metal hydroxides at high pH.
Automated fluorometric analysis achieved 95« recovery
of humic acids standards even at dissolved iron
levels of 2 and 5 ag/ ft. These results are much
better than earlier UV spectrophotometry and
manual fluorometric procedures which have errors on
the order of 75% and 50%, respectively, when iron
level is at 5 mg/ 1.
V-9
-------
Martin Marietta Environmantal Systama
Buffle, J., P. Deladoey, and W. Haerdi. 1978. The use of ultra-
filtration for the separation and fractionation of organic
ligands in fresh waters. Anal. Chim. Acta. 101:339-357.
Objective:
Findings:
To study the use of membrane filtration for the
separation and fractionation of organic matter in
fresh waters under various experimental conditions.
The use of centrifugation, ultracentrifugation, gel
permeation chromatography, and membrane filtration
for the separation of organic matter from fresh water
were compared, and optimal conditions for ultrafiltra-
tion measurements were studied. Amicon membranes
UM05, UM02, and PN10 (or their equivalents) were
found to be the most useful membranes for ultrafil-
tration. Electrolyte concentrations, pressure, and
pH did not greatly affect results as long as extreme
conditions were avoided. Results, however, were
greatly dependent on interactions of the organic
matter with 1) other dissolved molecules or colloidal
particles; 2) the membrane, because of its absorbant
properties? and 3) other organic matter when
aggregations were formed.
Gjessing, E.T. 1965. Use of "Sephadex* gel for the estimation
of molecular weight of humic substances in natural waters.
Nature. 208:1091-1092.
Objective:
Findings:
To fractionate and characterize soluble humic
substances from Norwegian surface waters according
to their molecular size.
The moorland water consisted of at least two types
of humic substances that differ considerably in
molecular size: a large molecular weight fraction
with an estimated molecular weight of between
100,000—200,000, and a small molecular fraction
with a molecular weight below 10,000.
Both the large and small molecular weight fractions
were believed to consist of organic complexes of
ferrous iron, while the smaller molecules (molecular
weight of <50,000) also contained calcium.
V-10
-------
Martin Mariana Environmental Systams
Gjessing# E. and G.P. Lee. 1967. Fractionation of organic
natter in natural waters on Sephadex columns. Envir. Sci.
Technol. 1:631-638.
Objective! To fractionate organic natter in natural water into
apparent molecular weight groups using Sephadex columns
and to study the characteristics of the different
fractions.
Pindinosi Gel permeation chromatography was shown to be a
promising tool for fractionation of organic matter
in natural water. Moderately-colored creek water
was fractionated into 10 fractions. Results indicated
that natural waters are composed of several groups
of compounds# a significant number having molecular
weight above 50#000.
Various lakes# streams, and leachable organic matter
from lake sediments showed different organic carbon
elution patterns. Based on these results# it should be
possible to classify lakes according to their organic
matter elution patterns from different grades of Sephadex
and other gels.
The ratios of dichromatic-oxidizable organic matter#
color# and organic nitrogen vary between fractions.
These ranges may also aid in classification and study
of the composition of humic substances.
Low-colored waters appear to obtain most of their color
from the low-molecular-weight fraction# while in
moderately and highly colored waters# more color
is found in the high-molecular-weight fraction.
The authors question the average molecular weight
value# for natural water color (456)# previously
reported by Shapiro (1957).
******
Gjessing# E.T. 1970. Ultrafiltration of aquatic humus.
Environ. Sci. Technol. 4*437-438.
Objective:
Findings:
To examine the molecular size distribution of
aquatic humus using Diaflo ultrafiltration membranes.
Results suggest that
carbon and II of the
weights of less than 1#000.
V-ll
approximately 10% of the organic
colored matter have molecular
Most of the organic
-------
Mrtin Marietta Environm«ntai Systams
matter (90%) and approximately 50% of the colored
matter was collected on membranes with pore size
designed to retain molecules with molecular weight
larger than 20/000*
Lawrence, J. 1980. Semi-quantitative determination of fulvic
acid, tannin# and lignin in natural waters. Water Res. 14:
373-377.
Objective: To present an analytical procedure for the simultaneousr
semi-quantitative determination of fulvic, acids,
tannin# and lignin using ultraviolet spectrophotometry*
Findings: Methods for measuring fulvic, tannic, and lignosulfonic
acids using a multiple-wavelength ultraviolet
spectrophotometrie approach were presented.
Absorption spectra of test samples with known com-
positions of fulvic, tannic, and lignosulfonic
acids were examined and used to calculate their
concentrations. The deviations from known concentra-
tions were <15% for fulvic acid, <30% for tannic
acid, and <50% for lignosulfonic acid. These
accuracies were reported to be reasonable when
compared to other analytical procedures available
for natural organic compounds.
Natural water samples were analyzed for tannin,
and lignin by the proposed methods and by the
photophomolybdic acid method. Results indicated
reasonable agreement for the two methods in all
but two of the samples. The reason for the dis-
crepancy in two samples was not known.
The author believes that the analytical procedure
presented is a fast, reliable method for simultaneous
determination of fulvic acid, tannin, and lignin
concentrations with reasonable analytical accuracy.
******
Leenheer, J.A. 1981. Comprehensive approach to preparative
isolation and fractionation of dissolved organic carbon from
natural waters and wastewater. Envir. Sci. Technol.
15:578-587.
Objective: To present a comprehensive analytical procedure which
can be used to isolate most organic solutes from water.
V-12
-------
Martin Msriatta EnvJroimwntil Syanms
Findingst Methods for isolating complex mixtures of organic
solutes from water and organic salts are described*
The procedure involves the use of a series of
resin absorbents , and results in the fractionation
of the dissolved organic carbon in water into six
classes of compounds* 1) hydrophobic bases,
2) hydrophobic acids, 3) hydrophobic neutrals, 4)
hydrophilic bases, 5) hydrophilic acids, 6) hydrophilic
neutrals.
After secondary concentration and a desalting procedure,
81% of the dissolved organic carbon was recovered from
both oil shale retort wastewater and river water samples.
The techniques presented represent a comprehensive
analytical procedure for the isolation of most organic
solutes from water.
Mantoura, F.C. and J.P. Riley. 1975. The analytical concentration
of huntie substances from natural wat®rs» Anal* Chisi*
Acta. 76:97-106.
Objectives To examine the optimal conditions for the use of
Amberlite XAD-2 resin to remove huraic acid from water
and to study the physical chemistry of the absorption
process.
Pindinas: The thermodynamics of the absorption of humic and
fulvic acids on macroreticular polystyrene resin
Amberlite XAD-2 were examined. Under optimal
conditions, recoveries of 92% of the humic acids
and 75% of the fulvic acids were achieved.
Humic and fulvic acids could be fractionated on a
molecular weight basis during the desorption stage
by serial elutions at selected pH values.
Miles, C.J., J.R. Tuschall, Jr., and P.L. Brezonik. 1983.
Isolation of aquatic humus with Diethylaminoethylcellulose.
Anal. Chem. 55:410-411.
Objective: To examine the recovery of aquatic humus by DEAE-
cellulose using both batch and column methods.
V-13
-------
MMtin Miriam Environmental Systwnt
Findings; DEAE-cellulose, an anion exchange material, was
found to rapidly and efficiently concentrate and
isolate organic acids from natural waters at pH
levels from 4 to 8.
Inorganic anions, such as chloride and bicarbonate^
at levels typically found in freshwater were not
concentrated by the methods outlined. Thus, DEAE-
cellulose serves to both concentrate and separate
organic acids from other cations and neutral species
in natural waters.
Because of its simplicity, the batch technique may
be better suited for concentration of humic compounds
in the field than the resin absorption techniques
presently being used.
Perdue, E.M. 1978. Solution thermochemistry of humic substances*
I. Acid-base equilibria of humic acids. Geochim. Cosmochim.
Acta. 42:1351-1358.
Objective: To characterize and measure acid functional groups
of humic acids using titration calorimetry.
Findings: Titration thermochemistry was found to be a useful
technique for the characterization of acidic function
groups of humic substances and thermodynamic parameters
(pK) for this ionization. Results confirm the
generally accepted opinion that humic acids contain
both carboxy1 and phenolic hydroxy1 groups.
Calculation suggested that "at least one-third of the
phenolic hydroxy1 groups are not ortho to a carboxyl
group and thus cannot participate in chelation of
metal ions via salicylate—like functional groups.
The nature of the remaining two-thirds of the phenolic
hydroxyl groups is still unknown".
Perdue, E.M., J.H. Reuter, and M. Ghosal. 1980. The operational
nature of acidic functional group analyses and its impact
on mathematical description of acid-base equilibria in
humic substances. Geochim. Cosmochim. Acta. 44:1841-1851.
Objective: To evaluate several analytical methods used to
measure carboxyl content of aquatic humic substances,
including the calcium acetate exchange reaction,
other indirect and direct potentiometric titrations,
and therraometric titrations.
V-14
-------
Martin Marietta Environmental Syitam*
Findings: The carboxyl content of humic substances when
measured by titration methods is usually designated
as those functional groups which are sufficiently
acid: to protonate weak bases under a specified set
of experimental conditions. The method involves
at least two questionable assumptions: 1) that
all carboxyl groups in a sample are sufficiently
acidic to protonate a weak base, and 2) that no
other acidic functional groups are reacted. Therefore,
the present method can only yield an operationally
defined values of carboxyl content of humic substances.
Carboxyl content may be redefined as "those acidic
functional groups which are sufficiently acidic to
protonate ClUCOO" in the presence of a monovalent
counter-ion (e.g., Na ), with CH3COO" being in suf-
ficient excess."
Direct thermometric titration was used to place an
absolute lower limit on carboxyl content. Depending
on the other analytical methods used to measure
operational carboxyl content/ the values ranged from
5 to 43% greater than the absolute lower limit.
"The results suggest that the most appropriate analytical
method for determination of an operationally defined
carboxyl content would utilize a weak base that does
not contain a polyvalent cation in an indirect
titration on a reaction mixture from which all humic
reaction products have been removed prior to titration."
¦Because carboxyl contents can only be determined
0£erationally, any derived dissociation data (e.g.,
pK, a, etc.) are themselves operationally defined,
and thus not necessarily accurately representing
the actual dissociation equilibria of the acidic
functional groups."
Results of potentiometric and thermometric titration
studies found no evidence to justify the differention
of "carboxyl" groups into two or more distinct
sub-groups•
Reuter, J.H., N. Ghosal, E.S.K. Chian, and M. Giabbai. 1983.
Oxidative degradation studies on aquatic humic substances.
Ins Aquatic and Terrestrial Humic Material. R.F. Christman
and E.T. Gjessing, Eds. Ann Arbor Science: Ann Arbor,
MI. pp. 107-125.
V-15
-------
Martin Marietta Environmental Systems
Objective: To study the effect of varying conditions of oxidative
degradation of aquatic humus on the composition of
the degradation products, and to determine the bias
they may introduce into measurements of the relative
abundance of aromatic vs aliphatic components.
Findings: GC/MS spectra were both qualitatively and quantitatively
different depending on oxidation procedures used.
Under the mildest degradation conditions employing
potassium permanganate, oxalic acid was the major
reaction product. Prom the high oxalic acid/aryl
compound ratio, it was concluded that Satilla
River aquatic humus was highly aliphatic.
Increasingly severe oxidation conditions lead to pro-
gressive changes in the product mixture makeup and
introduced an unjustified bias favoring an interpre-
tation of a greater degree of aromaticity in the
sample.
Steelink, C., M.A. Mikita, and K.A. Thorn. 1983. Magnetic
resonance studies of humates and related model compounds
In: Aquatic and Terrestrial Humic Materials. R.F. Christman
and E.T. Gjessing, Eds. Ann Arbor Science: Ann Arbor, MI.
pp. 83-105.
Objective: To examine humic materials from a variety of well-
defined sources by proton nuclear magnetic resonance
(jH-NMR) and carbon-13 nuclear magnetic resonance
(13C-NMR) spectroscopy to identify structural
features common to fulvic and humic acids.
Findings: "NMR data of premethylated humic materials reveal
three major groups of hydoxyl functionality:
carboxy, phenolic, and carbohydrate moieties.
Within these groups, sharp peaks for aliphatic
carboxyl and 6-hydroxyl carbohydrates occurred in
humic and fulvic acid samples from a wide variety
of sources. Humic acids appear to have the highest
carbohydrate content."
Results of ifl-NMR indicate that fulvic acids have more
highly substituted aromatic rings than humic acids.
The addition of strong alkaline solution (i.e., sodium
hydroxide) to natural humic compounds was found to
distort the 13C-NMR spectra. The distortion is caused
by the formation of semiquinone radical anions which
cause the collapse or elimination of aromatic resonance.
V-16
-------
Martin Marietta Environmental Syctami
Stuermer, D.H. and J.R. Payne. 1976. Investigation of seawater
and terrestrial humic substances with carbon-13 and proton
nuclear magnetic resonance. Geochim. Cosinochiin. Acta.
40:1109-1114.
Objective*
Findings*
To investigate fulvic acids isolated from seawater
(Northwestern Sargosso Sea) using carbon-13 and proton
nuclear magnetic resonance, I.R. absorption spectroscopy,
and elemental analysis, and to compare findings to
existing terrestrial data.
The main structural differences between fulvic acids
isolated from seawater and terrestrial environments
is the low abundance of aromatic precursors (i.e.,
lignin) in seawater.
Thurman, E.M. and R.L. Malcolm. 1981. Preparative isolation
of aquatic humic substances. Environ. Sci. Technol. 15:463*
466.
Objective: To present a method of using macroporous resins to
obtain preparative quantities of low-ash, aquatic
humic substances.
Findinas: The procedure has been used successfully to concentrate
humic substances from various types of water samples,
including river water and ground water. Ash content is
typically less than 1%.
The method presented can be used to concentrate
humic substances and purify them from the bulk of
inorganic substances in water, even when the DOC
of the water is 0.7 mg C/i, and the humic concentration
is 50 ug/1 DOC.
Thurman, E.N. and R.L. Malcolm. 1983. Structural study of
humic substances: New approaches and methods. In:
Aquatic and Terrestrial Humic Materials. R.F. Christman
and E.T. Gjessing, Eds. Ann Arbor Science: Ann Arbor,
MI. pp. 1-23.
Objective: To fractionate and characterize fulvic acid from
the Swannee River, Georgia, using a threefold approach:
1) separation and purification of humic substances
V-17
-------
ftfertin Marietta Environmental Syitams
by liquid chromatography; 2) degradation o£ humic
substances by oxidation and methylation using 13c
reagents; and 3) identification of products and
structural components by capillary gas chromatography
- mass spectroscopy and 13c nuclear magnetic resonance
(13c-nmr).
Findings: Liquid chromatography was effective in separating
fulvic acids from inorganic substances, carbohydrates,
and fatty acids.
Results of l^C-NMR were compared against standard
methods of characterization (elemental analysis and
functional group titrations) and was shown to be a useful
technique for the structural analysis of humic substances*
Results suggest that fulvic acid in the Swannee River
consists of a mixture of plant degration products,
including lignins, terpenoids, flavonoids, and hydrolyz"
able tannins.
Weber, J.H. and S.A. Wilson. 1974. The isolation and characterize
ation of fulvic acid and humic acid from river water.
Water Res. 9:1079-1084.
Objective: To isolate, characterize, and compare fulvic acids
and humic acids from New Hampshire soil and two surface
water sites in New Hampshire.
Findings: New methods of isolating humic and fulvic acids
from water using anion exchange, cation exchange,
and molecule-absorbing resins were described.
Elemental and functional group analyses were performed.
The fulvic acid and humic acid samples isolated from
water were found to differ from each other but were
similar to analogous soil samples.
******
Wilson, M.A., R.P. Phelp, A.H. Gi1lam, T.D. Gilbert, and K.R.
Tate. 1983. Comparison of the structure of humic substances
from aquatic and terrestrial souces by pyrolysis gas chroma-
tography-mass spectrometry. Geochim. Cosmochim. Acta.
47:497-502.
Objective: To examine and compare the structure of humic
extracts isolated from terrestrial, marine, plankton,
and freshwater sources using pyrolysis gas chroma-
tography-mass spectrometry. (py-GC-MS).
V*18
-------
Martin Marietta Environmental System*
Pindinas: Thirty-seven compounds were identified in the four
samples and could be broken up into five main classes:
1) furans, pyrroles, and nitriles; 2) cyclopentenones;
3) phenols; 4) carboxylic acids, and alkanes; and
5) miscellaneous.
A wide range of biologically-related material charac-
teristic of proteins, carbohydrates, and peptidoglycans
were present in the samples and indicate their
geologically immature nature.
Results show that freshwater humic materials contain
aliphatic polycarboxylic acids derived from terrestrial
sources.
C. ACIDITY STUDIES
Glover, G.M. and A.H. Webb. 1979. Weak and strong acids in
the surface waters of the Tovdal region in S. Norway.
Water Res. 13:781-783.
Obiective: To study the concentration of weak and strong acids
in the surface waters of the Tovdal region of southern
Norway during the spring snow-melt period.
Findings: In the Tovdal River during spring snow-melt (i.e.,
late March and April), strong acid concentration
varied between 3 and 11 yeq t"1, while weak acid
concentrations varied between 67 and 106 peq t"1.
Weak acids contributed between 10 and 60% of the
hydrogen ion concentration.
The pH of the Tovdal River varied from 4.9 to 5.0.
The authors estimate that in the absence of excess
strong acids, the pH of the water would be 5.2-5.3
due to the presence of weak acids.
Weak acid concentrations and their contribution to
hydrogen ion concentrations were the least during
the period of most rapid thaw.
In studies of the weak acids, routine chemical analysis
suggest that aluminum and silicon accounted for 40-
60 ueq i~l while humic and fulvic acids accounted
for 20-50 yeq i"1.
Henriksen, A. ahd H.M. Seip. 1980. Strong and weak acids in
surface waters of southern Norway and southwestern Scotland.
Water Res. 14:809-813.
V-19
-------
Martin Marietta Environmental Systems
Obiectives: 1) To measure pH, strong and weak acid concentrations
(Gran's plot), and the concentrations of major ions
in lake water samples; and 2) to study the contri-
bution of strong and weak acids to lake acidity.
Findinast Variance in weak acid concentrations of lakes in
south Norway and Scotland was largely explained by
changes in the organic carbon and aluminum content
of the water. Aluminum was found to behave as a
tri- or tetrabasic acid. Other weak acids play
minor roles in lake acidification except at very
low aluminum and organic carbon concentrations.
It was estimated that each.mg of organic carbon
corresponds to about 5.5 veq of weak acid. This
is much lower than Schnitzer's (1978) estimates of
20 ueq mg~l C for "average" fulvic acids, and
the estimate of Oliver et al. (1983) of 10 ueq
mg'lc for fulvic and humic acid extracts.
These Differences may be related to differences
in the methods used to make estimates (see Molvaesmyr
and Lund, 1983).
Samples from Norway and Scotland show similar relation'
ships between strong acids and H+ concentrations
calculated from the pH. If the pH is higher than
about 5.5, the strong acids have negative values
(Gran plot, titration) corresponding to the presence
of bicarbonate or other bases. In the pH range of
4.8 to 5.5, the strong acid concentration is usually
positive, but less than the H+ concentration,
indicating contributions from weak acids, which
may have existed as bases before excess input of
strong acids started.
In areas where acid deposition from the atmosphere
has increased, there is also increased leaching of
aluminum from the soil, causing, increases in weak
acid concentrations as well.
Earlier studies by Glover and Webb (1979) hypothesized
that, in the absence of any excess strong acids, the
Tovdal River would have a pH of about 5.2-5.3 during
their study period. In this study however, the
authors pointed out that under "pristine" conditions,
bicarbonate instead of sulfate would have been
the major anion. The estimate of pH levels for
the Tovdal River under "pristine" conditions
would be >5.7.
V-20
-------
Martin Marittts Environmmtai Sytttma
Le», Y.H. and C. Brosaet. 1978. The alope of Gran's plot: A
U8eful function in the examination of precipitation, the
water-soluble part of airborne particles and lake water*
Water, Air and Soil Pollut. 10:457-469.
Objective: To refine and apply the Gran's plot acid-base
titration technique to the study of natural water samples.
Findinas: Titration procedures were refined and calibrated
using solutions of known acid compositions.
"Studies of rainwater samples and leaching solutions
of airborne particles showed that no unknown weak
acids that can affect pH in the range 4 to 5 were
present in these samples."
Preliminary investigations of several lake water sys-
tems gave a clear indication of the presence of
weak acids in titration results. All lake water
samples tested were shown to have one predominant
weak acid present, with a dissociation constant of
about 3 x 10-4. The concentration of weak acid
was determined and found to be linearly correlated
with the color (267.7 x absorbance at 400 nm) of
the sample.
Results "show that the slope of Gran's plot can be
used to identify and determine the concentration
of unknown weak acids." This is most applicable
when there is only one unknown weak acid present
in the system studied. In these cases, the acidity
constant and concentration can be exactly determined
through graphical methods. When more than one
weak acid is present, the problem is more difficult.
Lee, Y.H. 1980. The linear plot: A new way of interpreting
titration data, used in determination of weak acids in lake
water. Water, Air and Soil Pollut. 14:287-298.
Objective:
Findings:
1) To develop a method for determining the concen-
tration and dissociation constant (plT) of a monobasic
weak acid from the slope of Gran's plot; and 2) to
demonstrate the application of this method to an
investigation of weak acids (fulvic acids) in lake
water.
Methods were presented for determining the concentration
and dissociation constant of the monobasic weak
acid using a linear plot.
V-21
-------
Mwtifi Marietta Environmental Syttwm
Investigations of the weak acid content indicated the
presence of only one weak acid in the pH range
measured in four of the Swedish lakes studied. For this
weak acid, the dissociation constant (pK) appears to be
in the range of 4.5 to 4.7, and concentrations range
from 3 to 9 x 10""5 mole t~i. In two additional lakes,
the results indicated that at least one additional weak
acid with pK <4.4 must be present.
Results of this and an earlier study (Lee and Brosset,
1978) indicate a possible relationship between humus
content of the water and weak acid concentrations. The
acid in question is believed to correspond to the weak
carboxylic acid group (pK 4.5 to 5.5) which has been
identified in soil-extracted humus samples.
Little is known about the second weak acid present in
some lake samples. It was speculated, however, that it
may be another carboxyl group, either on an aromatic
ring ortho to a phenolic OH group, or ortho to carboxyl
groups in the humic acid. This carboxyl group would
correspond to a somewhat stronger acid with dissolution
constants (pK) of 2.5 to 3.4.
******
Molvaersmyr, K. and W. Lund. 1983. Acids and bases in fresh-waterss
Interpretation of results from Gran plots. Water Res. 17:303-
307.
Objective: 1) To discuss methodological problems which may be
encountered when the Gran plot procedure is applied
to freshwater samples, and 2) to analyze both
synthetic samples containing weak bases and buffers,
and natural water samples in order to illustrate
the different effects which may be observed.
Findings: To obtain correct results from the Gran titration,
the procedure must start at a sufficiently low pH
value and continue until a sufficiently high pH
value is reached. Normally a pH range of 3.6-10.3
should be covered. Failure to cover a large enough
range can result in erroneous results.
When bases are present in the sample, they will
appear in the titration as a negative value for
strong acids and a positive value for weak acids,
provided that the bases are not volatilized upon
acidification at the beginning of the Gran titration.
V-22
-------
Martin Marietta Environmental Syitams
In one lake in southern Norway (Lake Hovvatn),
the presence of strong acids was seen as an indicator
that the lake was influenced by acid precipitation.
In a second lake (Lake Langtjern), however, the
presence of bases could not be used as an indication
of absence of acid precipitation. The lake had a
relatively high concentration of weak acids, which
may have been formed as a result of acid precipitation,
through the protonization of bases which were
originally present in the lake.
******
Oliver, B.6., E.M. Thurman, and R.L. Malcolm. 1983. The contri-
bution of humic substances to the acidity of colored
natural waters. Geochim. Cosmochim. Acta. 47:2031-2035.
Objective: 1) To measure the carboxylic acid content and study
the dissociation behavior of humic and fulvic acids
extracted from natural waters, and 2) to calculate
the contribution these organic acids make to the
acidity of several highly colored, humic dominated
waters from Nova Scotia, Canada.
Findings: Historical data that measured the carboxylic acid
content of fulvic and humic acids from a variety of
environments (i.e., rivers and streams, lakes, wetlands,
and groundwater) were reviewed. In general, it appears
that the carboxyl content of aquatic humic substances
is approximately 10 ueq/mg C(+ lOt). The only
significant difference in carboxyl content between
environments was in humic substances from wetlands,
which contained 10% fewer carboxyl groups.
Dissociation behavior of combined fulvic and humic
acid extracts from Spencer Creek, Ontario, and
Pebbleloggitch Lake, Nova Scotia, were studied in
detail. Plots of the pK versus pH for the two
samples were similar, and almost identical to earlier
studies (Gamble, 1970} Perdue et al., 1980).
Based on data from this study and knowledge of the
pH of water and the mass action quotient of the fulvic
and humic acids, the following empirical equation
was proposed:
pf - 0.96 + 0.90 pH - 0.039(pH)2
Results of this study indicate that the acidic contri-
bution of the carboxylate anion (A~) of humic
substances in natural waters can be estimated by
measuring the humic substance concentration and
V-23
-------
Martin Marietta Environmental Systems
the pH of the water. The estimate is made in the
following manner. Dissolved organic carbon content
of a sample in mg/i~l is multiplied by 10 ueq/mg
DOC to calculate the organic acid concentration,
CT. The mass action quotient/ K, for a sample
organic acid is estimated from the sample pH using
the equation given above. Values are then substituted
into the following equation:
K[Ct]
[A-] -
K+lH+1
Using the above estimation procedure, the ionic
balances in three highly colored Nova Scotia rivers
were examined. Inclusion of the estimate of organic
anion (A") concentration resulted in a good
ionic balance. The results also indicated that
organic anions makes a significant contribution to
the calculated ionic balance.
Perdue, E.M. 1979. Solution thermochemistry of humic substances
II. Acid-base equilibria of river water humic substances.
In: Chemical Modeling of Aquatic Systems. E.A. Jenne, Ed.
ACS Symposium Series No. 93, American Chemical Society
pp. 99-114.
Objective: To use titration calorimetry to characterize the
acidic functional groups of humic substances.
Findings: In all samples, the carboxyl content determined using
titration calorimetry were less than values obtained
by the calcium acetate exchange reaction. Results
suggest that the calcium acetate exchange reaction
seriously overestimates the carboxyl content of
river water humic substances. Since the differences
between total acidity and carboxyl groups are used
to calculate phenolic content, most estimates of
phenolic content of river water are probably too
low.
Although the technique was used only on humic sub-
stances, it appears that past measurements of river
water humic acids and fulvic acids using calcium
acetate exchange reaction could have similar problems.
Previously reported high concentrations of carboxyl
groups in fulvic acids may be an artifact of the
calcium acetate exchange reaction.
v-24
-------
Martin Marietta Environmental Syttams
River water huraic substances appear to contain sig-
nificant concentrations of weakly acidic phenolic
hydroxyl groups. These groups are greater in
river water humic substances than in soil humic
acids. Results suggest that river water humic
substances have a greater capacity to form complexes
with metals than do soil humic acids.
******
Webb, A.H. 1982. weak acid concentrations and river chemistry
in the Tovdal River, Southern Norway. Water Res. 16:
641-648.
Objective: To study seasonal changes in both weak acid concen-
tration and the general chemistry of the Tovdal
River over a 15-month period.
Findings: Weekly water samples were analyzed for pHr sulphate,
chloride, nitrate, sodium, potassium, calcium,
magnesium, ammonium, aluminum, dissolved silica, and
weak and strong acids. Both atmosphere-derived
sulphate and ground-derived alkalis (calcium, mag-
nesium, and potassium) showed a steady increase during
the autumn and winter to a broad peak during the spring,
followed by a minimum concentration during the summer.
Nitrate and ammonium concentrations showed significantly
different seasonal patterns, primarily because they
are readily metabolized as nutrients.
In the spring of 1979 during the snow-melt period,
sulphate and nitrate peaks were matched by an equiva-
lent increase in the alkalis leached from the ground.
In the autumn of 1979, however, sulphate concentrations
increased without an equivalent increase in leached
alkalis, and water pH decreased.
Weak acid concentrations also exhibited a late
winter-spring maximum and a summer minimum. Seasonal
variation of weak acids could be largely explained
by variations in the concentration of inorganic weak
acid species of aluminum, ammonium, and dissolved
silica.
V-25
-------
Martin Marietta Environmental Systama
VI. REFERENCES
Aiken, G.R. In press a. Isolation and concentration techniques
for humic substances, Jto: Humic Substances in Soil,
Sediments and Waters Geochemistry, Isolation and
Characterization. New York: John Wiley and Sons, Inc.
Aiken, G.R. In press b. Isolation of organic acids from large
volumes of water by absorption on macroporous resins. In:
Concentration of organic solutes from water for biological
testing. Proceedings of the August 1984 American Chemical
Society Symposium.
Aiken, G.R., E.M. Thurman, R.L. Malcolm, and H.F. Walton. 1979.
Comparison of XAD macroporous resins for the concentration
of fulvic acid from aqueous solution. Anal. Chem. 51:1799-
1803.
American Public Health Assoc. (APHA). 1980. Standard methods
for examination of water and wastewater. 15th ed.
Washington, O.C.: American Public Health Assoc.
Askne, C. and C. Brosset. 1972. Determination of strong acid
in precipitation, lake water, and airborne matter. Atmos.
Environ. 6:695-696.
Beck, K.C., J.H. Reuter, and E.M. Perdue. 1974. Organic and
inorganic geochemistry of some coastal plain rivers of the
southeastern United States. Geochim. Cosmochim. Acta
38:341-364.
Brosset, C. 1976. A method of measuring airborne acidity:
Its application for the determination of acid content on
long-distance transported particles and in drainage water
from spruces. Proceedings of the First International
Symposium on Acid Precipitation and the Forest Ecosystem.
USDA Forest Service General Technical Report NE-23.
Christman, R.F. and E.T. Gjessing. 1983. Aquatic and terrestrial
humic material. Ann Arbor, Mich.: Ann Arbor Science Publishers.
Degens, E.T. 1969. Biogeochemistry of stable carbon isotopes.
In: Organic Geochemistry, G. Eglinton and M.T.J. Murphy, Eds.
New York, Heidelberg, and Berlin: Springer-Verlag, pp.304-329.
Dickson, W. 1980. Properties of acidified waters. In:
Ecological impacts of acid precipitation. Proceedings of
an Internation Conference. Sandefjord Norway, March 11-
14, 1980. D. Drablas and A. Tollon, Eds. SNSF Project NTIS
PC A17/MF A01. pp.84-92.
VI-1
-------
Martin Marietta Environmental Syitami
Fisher, J.E. 1984. Measurement of pH. Amer. Lab. 16:55-60.
Flaig, W., H. Beutelspacher, and E. Rietz. 1975. Chemical
composition and physical properties o£ humic substances.
In: Soil Components Vol. 1: Organic Components. J.E.
Gleseking, Ed. New York: Springer-Verlag.
Gamble, D.S. 1970. Titration curves of fulvic acidt The
analytical chemistry of a weak acid polyelectrolyte. Can.
J. Chera. 48:2662-2690.
Ghasserai, M. and R.F. Christman. 1968. Properties of the
yellow organic acids of natural waters. Limnol. Oceanogr.
13:583-597.
Gjessing, E.T. 1965. Use of "Sephadex" gel for the estimation
of molecular weight of humic substances in natural waters.
Nature. 208:1091-1092.
Gjessing, E.T. 1976. Physical and chemical characteristics of
aquatic humus. Ann Arbor, Mich.: Ann Arbor Science
Publisher Inc. 120p.
Gjessing, E. and G.F. Lees. 1967. Fractionation of organic
matter in natural waters on Sephadex columns. Environ. Sci.
Technol. 1:631-638.
Glover, G.M. and A.H. Webb. 1978. Weak and strong acids in
the surface waters of the Tovdal Region in S. Norway.
Water Res. 13:781-783.
Havas, M., T.C* Hutchinson, and G.E. Likens. 1984. Red herrings
in acid rain research. Environ. Sci. Technol. 18sl76A-186A.
Henrikson, A. 1980. Acidification of freshwaters: A large-
scale titration. Proceedings of an International Conference
Sandefjord, Norway. D. Drablas and A. Tollon, Eds. SNSF
Project NTIS PC A17/MF A01. pp.68-74.
Henriksen, A. 1984. Comment on acid precipitation in historical
perspective and effects on acid precipitation. Environ.
Sci. Technol. 18:631-632.
Henriksen, A., and H.M. Seip. 1980. Strong and weak acids in
surface waters of southern Norway and southwestern Scotland.
Water Res. 14:809.
Johnson, N.M., G.E. Likens, M.C. Foller, and C.T. Driscoll. 1984.
Science 225:1424-1425.
Josephson, J. 1982. Humic substances. Environ. Sci. Technol.
16:20A-24A
VI-2
-------
Martin Mariatta Environmental Syttami
Krug, E.C. and C.R. Prink. 1983. Acid rain on acid soil: A
new perspective. Science 221:520—525.
Lee, Y.H. and C. Brosset. 1978. The slope of Gran's plot: A
useful function in the examination of precipitation, the
water soluble part of airborne particles and lake water.
J. Water, Air, Soil Pollut. (Swed.) 10:457-469.
Leenheer, J.A. 1981. Comprehensive approach to preparative
isolation and fractionation of dissolved organic carbon
from natural waters and wastewaters. Environ. Sci. Technol.
15:578-S87.
Liao, W. 1981. Characterization of aquatic humic substances,
PhD disseration, Univ. of North Carolina at Chapel Hill.
Marcus, N.D., B.R. Parkhust, and F.E. Payne. 1983. An assessment
of the relationship among acidifying deposition, surface
water acidification, and fish popultions in North America.
Electric Power Research Institute Report No. EA-3127, Vol.
1.
Nolvearsmyr, K. and W. Lund. 1983. Acids and bases in fresh-
waters: Interpretation of results from Gran plots. Water
Res. 17:303-307.
Oliver, B.G., E.N. Thurman, and R.L. Malcolm. 1983. The
contribution of humic substances to the acidity of colored
natural waters. Geoch. Cosmochim. Acta 47:2031-2035.
Patrick, R., V.P. Binetti, and S.F. Halterman. 1981. Acid
lakes from natural and anthropogenic causes. Science
211:446-448.
Perdue, E.M. 1978. Solution thermochemistry of humic substances-
I. Acid-base equilibria of humic acid. Geochim. Cosmochim.
Acta 42:1351-1358.
Perdue, E.M. 1979. Solution thermochemistry of humic substances-
II. Acid-base equilibria of river water humic substances.
In: Chemical Modeling in Aqueous Systems. E.A. Jenne, Ed.
??-114. ACS Symp. Serv. No. 93., Am. Chem. Soc.
Perdue, E.M., J.H. Reuter, and M. Ghosal. 1980. The operational
nature of acidic functional group analysis and its impact
on mathematical descriptions of acid-base equilibria in
humic substances. Geochim. Cosmochim. Acta 44:1841-1851.
Perdue, E.M., J.H. Reuter, and R.S. Parrish. 1984. A statistical
model of proton binding by humus. Geochim. Cosmochim. Acta
48:1257-1263.
VI-3
-------
Mwtin Marietta Environmental Syst«m
Retzsch, W.C., A.G. Everett, P.F. Duhaime, and R. Nothwanger.
1982. Alternative explanations for aquatic ecosystem
effects attributed to acidic deposition. Prepared for The
Utility Air Regulatory Group by Everett and Assoc.
Rockville, MD. 77pp.
Router, J.H. and E.M. Perdue. 1977. Importance of heavy metal
organic matter interactions in natural waters. Geochim.
Cosmochira. Acta 41:325-334.
Richter, D.O. 1983* Comment on acid precipitation in historical
perspective and effects on acid precipitation. Environ.
Sci. Technol. 17:568-570.
Rosenqvist, I.T. 1980. Influence of forest vegetation and
agriculture on the acidity of fresh water. _In: Advances
in Environmental Science and Engineering Vol. 3. J.R.
Pfafflin and E.N. Ziegler, Eds. New York: Gordon and
Breach Science Publishers, pp.56-79.
Rosenqvist, I.T.r P. Jorgensen, and H. Rueslatten. 1980. The
improtance of natural H+ production for acidity in soil
and water. In: Proceedings of an International Conference
on the Ecological Impact of Acid Precipitation. D. Drablos
and A. Tollon, Eds. SNSF Project Report. Oslo, Norway.
NTIS, PC A17/MP A01.
Ruttner, F. 1963. Fundamentals of limnology (Translat. D.G.
Frey and F.E.J. Fry). Toronto, Canada: Univ. of Toronto
Press. 295pp.
Schnitzer, M. 1978. Humic substances: Chemistry and reactions.
In: Soil Organic Natter. Developments in Soil Science 8.
M. Schnitzer and S.U. Khan, Eds. New York: Elsevier
Scientific Publication Co. 319pp.
Schnitzer, M. and S.U. Khan. 1972. Humic substances in the
environment. New York: Marcel Dekker, Inc.
Seip, H.M., and P.J. Dillon. 1984. Science 225:1425-1426.
Shapiro, J. 1957. Chemical and biological studies on the
yellow organic acids of lake waters. Limnol. Oceanogr.
2:161-179. y
Shapiro, J. 1964. Effect of yellow organic acids on iron and
other metals in water. J. Am. Water Works Assoc. 56:1062-
1082.
Shapiro, J. 1967. Yellow organic acids in lake water:
Differences in the composition and behavior. In: Chemical
Environment in the Aquatic Habitat. Golterman, H.L. and
Clymo, R.S., Eds. Proc, IBP Symposium, Netherlands, pp.206-216.
IV-4
-------
Martin Marietta Environmental Symmt
Stevenson^ F.J. 1982. Humus Chemistry. Genesis, Composition,
Reactions. New York: John Wiley and Sons. 443pp.
Stuermer, D.H., K.E. Peters, and I.R. Kaplan. 1978. Source
indicators of huraic substances and proto-kerogen: Stable
isotope ratios, elemental compositions and electron spin
resonance. Geochim. Cosmochim. Acta 42:989-997.
Tanford, C. 1963. Physical chemistry of macromolecules. NY:
John Wiley and Sons, Inc. 551pp.
Thurman, E.M., R.L. Malcolm, and G.R. Aiken. 1978. Prediction
of capacity factors for aqueous organic solutes on a porous
acrylic resin. Anal. Chem. 50:775-779.
Thurman, E.M. and R.L. Malcom. 1981. Preparative isolation of
aquatic humic substances. Environ. Sci. Technol. 15:463-
466.
Visser, S.A. 1982. Acidic functional group content of aquatic
humic matter: Its dependence upon origin, molecular weight
and degree of humification of the material. J. Environ.
Sci. Health.
Visser, S.A. 1983. Comparative study on the elementary
composition of fulvic and humic acids of aquatic origin
and from soils and microbial substrates. Water Res.
17:1393.
Weber, J.H. and S.A. Wilson. 1975. The isolation and
characterization of fulvic and humic acid from riverwater.
Water Res. 9:1079-1084.
Wetzel, R.G. 1975. Limnology. Philadelphia: W.B. Saunders Co.
743pp.
Wilson, D.F. and P. Kinney. 1977. Effects of polymeric change
variations on the proton-metal ion equilibria of humic
material. Limnol. Oceanogr. 22:281-289.
Wilson, N.A., P.P. Barron, and A.H. Gillam. 1981. The structure
of fresh water humic substances as revealed by 13c-n.m.r.
Spectroscopy. Geochim. Cosmochim. Acta 45:1743-1750.
Wright, R.F. 1984. Science 225:1431-1432.
Vl-5
------- |