PB80-110323
Structure Reactivity Correlations for Environmental Reactions
SRI International, Menlo Park, CA
Prepared for
Environmental Protection Agency, Washington, D. C, Office of Toxic
Substances
September 1979
U.S. DEPARTMENT 0: COMMERCE
National Tech»:- jimation Service
NIIS
-------
EPA-560/11-79-012
STRUCTURE REACTIVITY
CORRELATIONS FOR
ENVIRONMENTAL REACTIONS
AUGUST 1979
FINAL REPORT
TASK FIVE
OFFICE OF TOXIC SUBSTANCES
ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
NOTICE
THIS DOCUMENT HAS BEEN REPRODUCED .
FROM THE BEST QOPY FURNISHED. US BY,
THE SPONSORING AGENCY; ALTHOUGH IT
IS RECOGNIZED THAT CERTAIN PORTIONS
ARE ILLEGIBLE, IT IS BEING RELEASED
IN THE INTEREST OF MAKING'A VAIL'ABLE
AS MUCH INFORMATION AS POSSIBL'E.
-------
. TECHNICAL REPORT DATA
{Plcasc-nad Iristnictioiis on the reverse before completing)
• REPORT NO.
• 2,
EPA-560/11-79-012
5. RECIPIENT'S ACCESSION-NO.'
'pr; ^o 11032?
i-. TITLE AND SUBTITLE
! Structure Reactivity Correlations for Environmental
Reactions
5.'REPORT DATE
September 1979
9. PERFORMING ORGANIZATION CODE
7. AUTHORlS)
Theodore Mill
a. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
SRI International
333 Raven%wood Avenue
Menlo Park, CA 94025
10. PROGRAM ELEMENT NO.
5722
11. CONTR ACT/OR ANTTNO.
68-01-41-09
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Technical Evaluation
Office of Toxic Substances
Environmental Protection Agency
Washington. D.C. 20460
,13. TYPE.OF REPORT AND PERIOD COVERED"
Final 4/1/79 to 8/15/79
14. SPONSORING AGENCY COPE
15. SUPPLEMENTARY NOTES
16. ABSTRACT • • • .• .•••.. • • -
Many of the key rate constants needed to predict rates of transformation and
transport of organic chemicals in water and air can be estimated from
structure-reactivity correlations (SRC) with reasonable accuracy. These rate ,
constants can be coupled with environmental parameters such.as pH or oxidant concent
trations to provide estimates of rates of these processes under a variety of environ-
mental conditions. SRC needed for zero-level testing are most abundant for hydroly-
sis oxidation and sorption, and relatively scarce for photolysis and volatilization.
Generalized SRC are readily used by non-expert, technically trained personnel whereas
application of detailed SRC or linear free energy relationships (LFER) require expert
knowledge in mechanistic chemical kinetics. An example is shown of the use of SRC
methodology to estimate rate and equilibrium constants for a specific chemical..
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS
c. COSATi Field/Group
Structure Reactivity Relationships,
Transformation, Hydrolysis, Photolysis,
Oxidation, Sorption, Volatilization
Environmental
Fate Assessment
Aquatic Systems
Atmospheric Systems
57A
99A
48G
68D
68E
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
21.
20. SECURITY CLASS (This pagef
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
EPA-560/11-79-OU
August 1979
STRUCTURE REACTIVITY CORRELATIONS
• FOR ENVIRONMENTAL REACTIONS
by
Theodore Mill
Physical Organic Chemistry'Department
Physical Sciences Division
Contract 68-01-41-r09
Project Officer
Asa Leifer
OFFICE OF TECHNICAL EVALUATION
OFFICE OF TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION .AGENCY
WASHINGTON, D. C. 20460
^International/
-------
DISCLAIMER
This report has been reviewed by the Office of Toxic Substances, U.S.
Environmental Protection Agency, and approved for publication. Approval does
not signify that the contents necessarily reflect the views and policies of
the U.S. Environmental Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
-------
PREFACE
This report was prepared under the general guidance of the Project
Officer, James Darr and the EPA Technical Monitor Asa Leifer. The report
was prepared at SRI by Theodore Mill with assistance from Kirtland
McCaleb (project leader). Valuable criticism was offered by Asa Leifer
(EPA). .
iii
-------
CONTENTS
Page
PREFACE ................... ill
FIGURE ,...,....,. vii
TABLES . . ... ........ ,.,;... . . . .%...'. ix
1. INTRODUCTION . . . . . .... . . . . . . ........ . .... 1
2. OBJECTIVES 3
3. RESULTS AND DISCUSSION . .... . . .:r.,.'. • • • •' .'.'. '. v 5
Fate Assessment Based on Test Protocols ......... 5
Environmental Processes and Kinetic Relations «'-.",.• .... 5
Prediction of Rate and Equilibrium Constants ....... 7
The Scope and Application of SRC ...... ... • • • .'• • • 7
The Scope and Application of LFER . , 8
Hammett Equation . . . . . . ... • • . . . . . . .8
Precision of LFER . . . . . . ... . . . . .,. . . • . . ..10
Bronsted Catalysis Equation ............. 12
Use of SRC and LFER-for Environmental Processes . . . 12
Chemical transformations . . . 13
Hydrolysis 14
Photochemistry ................. 15
Oxidation Processes . . . . . . . . . . ... . . • • 31
SRC for Oxidation .................. 33
Oxidation by R0a» ............... 38
Oxidation by Singlet Oxygen (I0a) ....... 42
Oxidation by H0» Radical .'. . . . . t. . . ... 42
Physical Transport. . . . . . . . . '.' '. . ..'.'. . . 43
Volatilization ..... . '.' .......... .,43
Sorption to Sediment and .Soil ..'..'•. , ... 45
4. METHODOLOGY FOR FATE ESTIMATION . .... i '....••...;.... ~..... 49,
5. CONCLUSIONS AND RECOMMENDATIONS . . . . .: :. . ". . ." V . • •: . . . 53'^
REFERENCES .'.... .'. . . .' . . \ . . 55
Preceding page blank
V
-------
FIGURE
Number Page
1 Soil or Sediment Partition Coefficient of Chemicals Versus
Solubility in Water ...................... 47
Preceding page blank
vii
-------
TABLES
Number . Page
1 Environmental Processes and Properties . .• . . 6
2 Selected Values'of Hammett o Values .............. 11
• , 3 Environmental Chemical Processes ............... 13
.. 4 .. Hydrolysis of Esters at pH 7 and 25*0 ... ..... . ,,...'.,' . ... 16.
5 Hydrolysis of Amides at pH 7 and 25°C . .'.'.., .-., ...... ..... 17.
6 Hydrolysis of Nitriles at 25°C . ... . . ./. . ,, . ,..'... 18
7 Hydrolysis of Acyl Chlorides at 25°C 18
8 Hydrplysis of Carbamates at pH 7 and 25'C ... '.. . .... .....,..,.,. . , ;: 19 >v.
9 Hydrolysis of Alkyl Halides at pH 7 and 25°C ......... 20
10 Hydrolysis of Phosphoric Acid Esters at pH 7 and 25°C ..... 21
11 Hydrolysis of Epoxides, Imides, and Cyclic Esters at pH 7 and
25°C ......................... J .:.. . . . ..' 22.,...
12 Summary of Chemicals. Persistent...to Hydrolysis: '.Half-rrlives > .
1 yr at 25°C and pH 7 . . . . . . , . . . . . . ... 23
13 Approximate Absorption Regions for Organic Molecules . . . . . 26
14 Quantum Yields for Photolysis of Ketones at 25'C: Structure and
Solvent Effects ......... 30
15 Quantum Yields of Selected Processes in the Presence and
Absence of Oxygen in Water ............ .'" ..... 32
16 Oxidant Concentrations in Water and Air , 33
17 Rates of Oxidation by R02» Radical .".... 34
18 Rates of Oxidation by Singlet Oxygen 35
19 Rates of Oxidation by H0» Radical 36
ix . ...
Preceding page blank
-------
Number : . ' ' . " ' Page
20 Rates of Oxidation by Os '.'.'•". •'•'•. . . . ..:. . . .... . . . .,• 37
21 Rates of Oxidation of Phenols by R0a» Radical . . 39
22 Rates of Oxidation of Aromatic-Amines by R02» Radical 41
-------
SECTION 1
INTRODUCTION
Of the several hundred billion pounds of synthetic chemicals manufactured
annually in this country, a significant amount enters the air,, water, and 50il,
often at very low concentrations. Although biotic and abiotic processes rapid-
•ly transform some or most of these chemicals to simpler and harmless forms, a
few chemicals persist, accumulate, and bioconcentrate, thereby affecting a wide
variety of plant and animal life. In some cases hv^man. health .is adversely
affected as well.
The Toxic Substances Control Act was enacted to provide the EPA's Office
of Toxic Substances (OTS) with the regulatory authority needed.to control and
minimize adverse effects of synthetic chemicals on the environment and on human
health.
One of the important activities of OTS is to evaluate the possible envi-
ronmental effects new chemicals may have if.marketed'and the-effects existing
chemicals may have because of their wide distribution, long persistence, high
toxicity, or large production volumes. To facilitate this task, OTS has pre-
pared a set of recommended test protocols for screening new chemicals for their
environmental persistence and effects (Federal Register, 1979). The recommended
protocols for fate are intended to provide data from simple laboratory kinetic
or equilibrium measurements that can be used to evaluate the probable lifetime
of a chemical in a specific environmental situation.
Despite the simplicity and low cost of the recommended test methods, their
widespread adoption and use by chemical manufacturers poses two major problems.
First, the total cost of performing most or all of the tests'may be particular-
ly burdensome for small manufacturing concerns that have limited staff and re-
sources. These cost requirements may stifle innovation by small companies,
particularly for potentially smallrvolume chemicals. Second, the large number
of new chemicals to be tested each year (estimated at between 200 and 500) will
generate a significant amount of kinetic and property data that must be evaluated
by knowledgeable staff in OTS; the process of evaluating such quantities of
data can lead to delays or, worse, inadequate assessments.
The test methods recommended by OTS reflect our present knowledge of the
important processes that control transport and transformation of chemicals in
the environment (Mill, 1979). Because some of the environmental processes
have been well-studied in the laboratory for many years, a considerable body
of empirical and theoretical knowledge is available from which to formulate
structure-reactivity correlations (SRC); these correlations can then be used
to relate molecular properties or structures to chemical or physical reactivity
-------
in a specific process. Hydrolysis of simple esters is a good example of how
the effect of changing the acid or alcohol structure can be systematized and
used to predict the rate constant for a new structure (Euranto, 1969).
. SRC area valuable tool•'"for predicting the probable range of reactivity
for a new chemical structure in specific environmental processes; they are
simple to use and can provide considerable savings in time and money for both
EPA and industry if they are used carefully as a pre-testing tool to select
only needed test protocols. This study was conducted to explore the scope of
SRC and'their potential value for predicting reactivity in environmental fate
processes (zero-level screening).
-------
SECTION 2
OBJECTIVES
The objectives of this study are to (1) evaluate the kinds and accuracy
of existing SRC for environmental fate processes, (2) use these SRC to formulate
some simple rules for including or excluding from testing specific types of
molecular structures, and (3) indicate the accuracy of quantitative predictions
of reactivity using SRC.
-------
SECTION 3
RESULTS AND DISCUSSION
FATE ASSESSMENT BASED ON TEST PROTOCOLS
The current methodology for fate assessment may be summarized as follows;
only the first two steps are now part of the proposed test program
• A chemical is screened in the laboratory to measure approximate rate
or equilibrium constants for all processes that might control the
transport or transformation of the chemical in the environment;.
• The rate constants (or half-lives) are compared under selected en-
vironmental conditions to evaluate the dominant processes that control
fate in specific environmental compartments.' •
• Additional detailed tests are performed for each dominant process
(usually one to three) to evaluate rate constants or equilibrium'
constants over a range of environmental conditions.
• The important rate or equilibrium.processes are integrated in a simple
multicompartment computer model'with environmental parameters and hy-
drologic or meterologic data. The model provides information on con-
centration as a function of both time and location in the environmental
location of interest. ' • .
A more detailed discussion and application of this methodology is found in
Smith et al. (1978). ....
ENVIRONMENTAL PROCESSES AND KINETIC RELATIONS
A detailed discussion of the kinetic or equilibrium processes thought to'
be important in the air, water, and soil is given in the recent paper by Mill...
(1979). Only a summary of that discussion is included here. Table 1 lists
the processes that should be considered in any comprehensive review of environ-
mental assessment. Each kinetic process listed in Table 1 can be formulated
as a reaction and kinetic relation
C + X —-*- products (1)
- Rate - -d[C]/dt - k [C][X ] (2)
where C is the chemical, 1^ is the specific rate constant for process n, and
Xjj is the chemical or biochemical species responsible for process n in the
environment. X may be H+, H0», soil organic content, or solar photon flux,
Preceding page blank
-------
Table 1 Environmental Processes and Properties
Process
Physical transport
Meteorological transport
Bio-uptake
Sorption
Volatilization
Run-off
Leaching
Fall out
Chemical transformation
Photolysis
Oxidation
Hydrolysis
. Reduction . ,
Biological tran6formation
Biotransformation
Key Environmental Property
a,b
Wind velocity
Biomass
Organic content of soil or sediments
Mass loading of aquatic systems
Turbulence, evaporation rate, re-
aeration coefficients, soil organic
content
Precipitation rate
Adsorption coefficient -
Particulate concentration,
Wind velocity
Solar irradiance, transmissivity .
of water or air , '
Concentrations of bxidants and
retarders
pH^ sediment or soil basicity or
acidity
Oxygen concentration* ferrous ibn
concentration and complexation state
Microorganism population and acclima-
tion level
At constant temperature.
bFrom Mill .(1979). <
-------
as just a few examples.
Equilibrium processes for sorptipn are treated in a similar fashion
state) (3)
(4)
where S is the sediment/soil mass and Koc is the measured value of K corrected
for the organic content (Mill, 1979; Kenaga, 1979; Smith and Bomberger, 1979).
Rate or equilibrium expressions such as (2) or (4) can be simplified to
pseudo-first-order forms if [XjJ is assumed to be constant during the/measure-
ment interval. . '' ' ' . ' .. : !. •.
(5)
Prediction of Rate and Equilibrium Constants.
A systematic approach to predicting values pf. 1% or KQC for new chemicals
rests on the well-known fact that despite almost limitless diversity in structure,
most organic chemicals share common reaction patterns and reactivites among
like-structured chemicals; SRC are a way of explicitly recognizing and
quantitating these similarities in reactivity for similar molecular structures.
Thus, the estimation of rate constants is actually a- tworpart procedure " '.. '.
• Selection of environmental processes (see Table 1) applicable to a
. specific chemical structure using simple, qualitative SRC
• Estimation of kinetic rate constants for potentially important processes
using quantitative estimation procedures such as linear free energy
relationships (LFER) (Shorter and Chapman, 1972) .
THE SCOPE AND APPLICATION OF SRC . "..••'
Chemists generally use screening SRC implicitly to estimate what kinds
of reactions a specific chemical will undergo, based on its .molecular structure.,
Thus, for example, all organic chemists know that alcohols cannot hydrolyze,
alkanes cannot absorb sunlight, but that acyl halides hydrolyze very rapidly;
fewer chemists could instintively classify thiadiazoles or dialkylaromatic
carbaraates as unreactive in hydrolysis or photolysis, and fewer still would
be able to estimate whether hindered aromatic phenols or amines will oxidize .
in the dark. Thus, screening SRC covers a range of easy- to-difficult ,>';
classifications of reactivities, but in each case only a yes or.no decision
. is required.
To predict how fast or significant a particular reaction 'will be for a
particular chemical structure chemists use quantitative SRC such as LFER.
For example, the sensitivity of ionization of phenylacetic. acids, in water at .. .
25°C to substituents is expressed as the op relation (Wells, 196'3)
-------
log K (XPhCHaCOOH) = log K(PhCH2COOH) + (0.562 ± 0.039)a (6)
Equation (6) is an example of LFER that link kinetic and thermodynamic parameters
to molecular structure in a quantitative framework. Both screening SRC and
LFER are useful for predicting environmental fate and persistence from
information about molecular structure and properties, but often only the
qualitative SRC is needed to select test protocols for a particular chemical.
This .is because of the several competing processes that might control the
persistence of a chemical in air, water or soil, only the one or two fastest
processes will control fate and the other slower ones will have little or no
influence on fate. Thus, if screening SRC predicts that processes A and B
will be faster than processes C, D, and E by a factor of 10 or more, it matters
little iif the rate factor is actually 10-. or 10,000 since the decision to exclude
tests for processes C, D, and E is a valid one for all rate factors larger
...than 10. Usually the confidence in these predictions increases with ah increase
in the rate factors. ..' ;•••'•' , .... . -.
For competing processes A and B, LFER may be usefully applied to predict
more quantitatively the values of k^ and kg. Unfortunately , application of
LFER to predicting a new rate constant from a set, of . measured ones is subject
to uncertainty and limitations, as discussed below.
THE SCOPE AND APPLICATION OF LFER
LFER relate rates contants and structural parameters through a general :
form, linear in log k (thus linear in AF^)' /;: "'-;'' •v:;''^!jJ"'-V ' ;-"-"' ' : ''"• ''-'';'
log kx = log kQ + op (7)
where kx is the unknown, rate constant k0 is known, and a and 3 are reaction,
.structure, or solvent parameters. The Hammett, Bronsted, Taft, Swain,
Grunwald-Winstein, and other LFER equations are of this form. T'he term free
energy implies a more fundamental thermodynamic foundation than is justified;
all these correlations are almost entirely empirical in origin and elaboration.
The range of reactivities that can be correlated with ' equation (7), is
indeed impressive i with some changes in kjj ranging Over 14 log units. These
are exceptional ranges, however, and most LFER cover 1 :or 2 log units in rate
constant (Wells, 1963; Shortef and Chapman, 1972).
Hammett .Equation . •
One of the best known and most widely used LFER is the Hammett equation
originally developed for correlating the acidity of substituted benzoic acids.
log k_< > log kn + op (8)
A . w ....
-------
In equation (8):
(1) k.£ and kg are rate or equilibrium constants for reaction of an
aromatic structure at ^ single temperature
(2) X is a m- or p-substituent in the ring; Y is the reaction center.
(3) k^ and kg refer to the X-substituent and H-substituent (parent),
respectively
(4) a is the substituent constant characteristic of the substituent alone
and independent of the type of reaction, but is a function of
temperature and solvent.
(5) p is the reaction constant characteristic of a specific reaction and
transition state. ' ''•'.-• • '• .• ; ' " '•'•'••• :-'\\-
Many modifications have been made to the Hammett equation to extend its
range of applicability; extensive reviews of LFER by Wells (1962) and Shorter
and Chapman (1972) are recommended for details. For this discussion we need
to focus on the two essential but limiting features of LFER: (I) the require-'
ment for a simple and constant reaction (similar transition states) for a valid
correlation, and (2) the largely empirical nature of LFER, which requires an
extensive experimental data set to develop the reaction constants, needed for
further prediction. These features restrict the usefulness of LFER for pre-
dieting new rate constants even though this objective was one of the mainsprings
for development of LFER.
Exner (1972) has discussed the range of validity of LFER, particularly
equation (8), in some detail. Many deviations from the LFER occur because the
reaction mechanism changes with the change in substituent. Examples are shown
in Exner's paper (see Figures 1.1-1.9). The application of LFER to a particular
reaction is often used as a sensitive, criterion for. detecting similarities and. '
differences in transition states for a similar group of chemical structures.
If LFER is used instead to predict a value for k£, then there must be some £
priori basis for knowing which particular value of p (or the corresponding
constant in another LFER) to choose for a particular reaction. For simple,
closely analogous reactions (e.g., ester hydrolysis, H-atom transfer from
aromatics), selection of correct values of p is generally straightforward; how-
ever, for some chemical structures and re'actions, selection of the correct
value of p may prove very difficult if not impossible. Two examples (from
Exner, 1972) will illustrate the point: (1) in a series of substituted cumyl
chlorides, solvolysis of p-nitrocumyl chloride has a measured rate constant :
almost one hundred times as fast as that predicted from the (good) correlation
equation based on m-substituents and nonconjugating p-substituents, and (2) .
acid-catalyzed addition of water to carbo'diimides shows excell'ent correlation
for 12 substituents; however, for 3-Me2N, the deviation from the line corresponds
to a factor of 10, and for 3-NOj, the deviation corresponds (approximately) to
a factor of 50 in rate. Failure, of the equation, apparently results from a
change in mechanism.
-------
Development of. a useful LFER does require a data base from which to derive
a best value of p.(or other reaction constants). A priori thermodynamic and
quantum mechanical procedures for calculating p from first principals ^-..,..
been tried with only minimal success. For.the immediate future, inves^'fccors
must continue- to rely on careful experimental studies to extend the usefulness
of LFER to environmental problems; where no data base exists, application of
LFER probably is imprudent.
Precision of LFER
Values of p, o, or other reaction constants are typically evaluated by
fitting data.on log k, log kx, or log K^ versus substituent constants, to a
regression equation using simple linear least squares methods. Error bounds
or confidence limits on p or a can then be calculated (Wells, 1963). Several
interesting points arise from examining the data of Wells. Most correlation
coefficients reported for r are > 0.98 [standard deviations typically are less
than 5%], but few reaction series cited contain more than eight substituehts.
Exher has proposed that a better measure of the usefulness of a LFER than r
is $:'•"•
<)> = [n(i-r2)/(n-2)]'s (9)
Applying equation (9) to a good data set where r » 0.998 for the Hammett equa-
tion and n =5 (Wells, 1963) gives .
= (5(1-0.998) 73)^ = 0.08
In this case, use of the Hammett equation for calculating log kjj. will, give a
mean deviation,only 8%, as large as would be obtained by^assuming' that '
substituents have no effect (p = 0). Howeverj for another example where r =
0.96 and n «*'5; <(> = 0.36; in this case use of the Hammett equation gives a
mean deviation which is 36% as large as would be obtained by assuming that p
= 0. If more data points were available such that n = 10, then 4> e 0.31; if
n = 25 i then $ = Oi29. •:•. •''•'' , ; .. . ...
Table 1> from Exner (1972),summarizes values of Hammett substituent
constants o for meta and para substituents. By definition H is zero'.. Other
substituents may have positive p or negative values of p ranging from 4-1.9 for
-Na+ to -0.66 for p-NH2. The range of values for k~ covered by these
substituents is almost'400 if p =1.00.
Values of a in Table 2 seem best suited for use in reactions in which the
transition states have no strong local charge development. Aromatic reactions
in which significant localized charges develop in the transition states are
better correlated by parallel substituent sets designated tf+ or a~. in general,
correlations of polar reactions with these substituents are riot as extensive
hbi'as .good as those with a.
Perhaps the main point to be emphasized is that statistical tests alone
cannot be used to judge whether a new chemical structure will fall on a
particular correlation line; some insight intd the probable reaction mechanism
is a necessary requirement tb using LFER as a reliable predictive tool.
-------
TABLE 2. SELECTED VALUES OF HAMMETT a VALUES3
, ..-•".
Substltuent
H
Me,Et,i-Pr
t-Bu
C=CH
Ph
CH,CN
CF,
CClj
CN
CHO
C(0).Me
C(0)NH2
CO,H
CO,R
NHj
N, . .. .
NO,
OH
CMe
OC(0)Me
SH
SOjNH,
F
Cl
Br
I
MMe*
N,+
CO,"
SO,"
' • . '• b
V
0.00
-0.04 to 0.
0 to -0.12
0.20
0.06 to 0.
..0.16
0.42 to 0.
. 0.40 '
0.56 to 0.
0.36
.0.31 to 0.
0.28
0.35 to 0.
0.36 to 0.
O.OOto -0.
. 0..27 ,; .,
0.70 to 0.
0.00 to 0.
0.12
0.26 to 0.
0.25 •"••
0.46 to 0.
0 . 34
0.37
0.39
0.35
0.88 to 0.
1.8
07
22
43
68
38
37
40
16
..
71*
12
39
53
99
-0.1 to 0.10
0.05 to 0.
31
0.00
rO.14 to 0.17
-0.18 to TO. 20
0.23
-0.01 to 0.02
0. 01 to 0.18
0.54 to 0.55
;:0.46
0.63 to 0.69
0.22 to 0.43
0.44 to 0.52
0.38
0.26 to 0.45
0.43 to 0.52
-0.57 to -0.66
... 0.15.. .',-..
6.78 'to 0.80
-0.36 to -0.37
-0.27
0.16 to 0.31
•0.15
0.57 to 0.62
0.06
0.23 .
0.23
0.18 to 0.28
0.82 to 0.96
1.9
-0.05 to 0.13
0.09 to 0.38
Exner (1972).
Range, where given, indicates extremes of
values listed.
11
-------
LFER for reactions of aliphatic structures have been developed by Taft
and reviewed in detail by Shorter (1972). In general, the Taft equation and
its elaborations do not correlate reactivity in aliphatic systems as well as
the simple Hammett equation does for aromatic systems and successful applica-
tion of LFER to aliphatic chemicals requires considerable insight into the
mechanisms of their reaction.
Another form of the Hammett equation is used to describe SRC for substi-
tution reactions in which the stability of the leaving group controls rate.
A~ + BX -T*~ ABX — *- AB + X~
: log kx i log kQ + plog Kj. (10)
In this form of the equation, log KX measures the stability of the anion
•relative to its; conjugate acid1 in: 'water or another basic solvent.
••..: Examples of the application of equation. .(11) are given in 'the section on
hydrolysis.,. ••:."/ ', . . ••'• /'' ••..••••••,••-•••• '•• • " • ••.'•••' '"-...
Bronsted Catalysis. Equation .
Structural changes in the reactant that effects a transformation at a
reaction center can be successfully correlated by the Bronsted equation; in
its simplest form, this equation relates acid or base, .strength :,tp catalytic ;
activity, 'in reactions involving' general' catalysis. ' ' ..... ' '""' ?
Because hydrolysis in natural waters usually involves only specific acid
(HsO"*") or base (OH-) catalysis,, the Bronsted equation is little used in en-
vironmental estimates* Correlation of pK^. or log k^ and rate constant, cited
above [equation (11) ]j is sometimes referred to as a Bronstedjcorrelation
(Williams, 1972).' ' : "; ' \' ' ;; : '
Us:e of SRC and. LF.ER for/; Environmental frrpcesses
fable 1 lists the processes believed to control transport and transforma-
tion in the environment. Some physical transport processes ^ such as
volatilization, can be related to therinbdynamic properties of the chemical in
dilute solution and are therefore amenable to correlation by LFER. Other
processes, such as sorptibh to sediment, particulate, or biomass, can be
correlated most readily with solubility or partitioning between organic and
aqueous or vapor phases, the relationship between molecular structure and
thermodynamic or kinetic parameters is less clear for these latter processes.
Biotransforinations are always mediated by enzymes through one or more
equilibrium processes usually coupled to ah irreversible bond-breaking or
bond-making step on the enzyme. In principle, the processes are susceptible
to treatment by LFERj in practice^ few such relatiohiships have been successful
" ,-•/: ''''•- . •""••''Vl- ' '" '" 12 .'•"•
-------
owing to the complexity of the overall process and the lack of detailed
understanding of the chemistry of individual steps. Some correlations of mi-
crobial and chemical rate constants are reported by Wolfe et fl. (1979).
A number of chemical transformations in the environment can be separated
into discrete elementary steps characterized by simple kinetic relationships,
many of which can be examined in detail. Therefore, many correlations can be
made between structural reactivity with reasonable assurance of their
applicability to environmental situations. For these reasons we have emphasized
the use of SRC and LFER for chemical transformations in this report.
Chemical Transformations , ... '. '.'.'.-'•••
Table 3 lists four basic types of chemical processes that occur in the
environment and the environmental agents that effect these processes.
TABLE 3. ENVIRONMENTAL CHEMICAL PROCESSES
Process
Agent, Property
Hydrolysis
Photolysis
Oxidation
Reduction
HaO, H30,, OH~, trace metals
sediment •'• • .. .'• • .
Solar irradiance, transmissivity
of water 'or air
H0«,
Fe2+
a02, 03, RO-
_j
For many of the processes listed in Table 3 we can formulate simple kinetic
relationships of the form
Rate - k [C][Pn]
(12)
where kjj is the rate constant for the n-th process, [C] is the concentration
of chemical and [Pn] is the concentration of environmental agent or property
of the n-th process. Prediction of the rates of many environmental processes
requires only prediction or measurement of the rate constant k^ because con-r
centrations of hydrolytic, oxidative and solar agents are now known for a
variety of environmental situations (Mill, 1979). The following sections will
examine the detailed processes and usefulness of SRC for predicting rate
constants.
13
-------
Hydrolysis • . • ' r...
Hydrolysis or organic compounds usually results in introduction of'a
hydroxyl function (-OH) into a-chemical, most commonly with the loss of a
leaving group • (-X). ' ..: ' '." ;' . ;
•: RX + H20 —*- ROH + HX (13)
il-C-X + HiC
R-C-OH + HX (14)
in water, the reaction is; catalyzed mainly by hydronium and hydroxyl ions,
hut in moist soil, loosely complexed metal ions such as copper or calcium may
also be .important catalysts for certain types of chemical structures. Sorption
of the chemical may also increase its reactivity toward H+ or H0~.
The general rate equation.for hydrolysis in water is
<-•:•;; /^'"^[cy-i/k^H*]^ '•" •••' (15)
where k^ is the measured first-order rate constant at a given pH. The last
term is the neutral reaction with water (second-order rate constant k.,') j and
in water it can be expressed as a pseudo-first-order rate constant k^.
This equation can be modified to account for the incursion of bound or
free metal-ion catalysis in soil or sediments, by including one or-more terms
for the form . : .
where k^ is the metal-^ion catalysis constant, [M]T is the total metal ion con-
centration, and k^ is the equilibrium constant for dissociation of the hydrated
ion complex. .Since a metal may be complexed in soil in several ways, the
descriptors heeded for the complete rate equation could be quite complex.
Equat.ion (15) shows that the total rate of hydrolysis in water is
pH^dependen't unless kA and k,, * 0. .
' - • - iMabey and Mill:(1978) have recently reviewed kinetic data for hydrolysis
of a variety of organic chemicals in aquatic systems and have reviewed the
chemical characteristics of most freshwater systems* These data have been
used in turn to calculate persistence (half-lives) of these same chemicals at
25° and at pH 7 in freshwater. Predictive test methods (screening and detailed)
for hydrolysis to develop the essential kinetic data: kjj, k^, kjj, kg, and their
temperature dependence (Arrhenius equation) have been prepared recently by SRI.
A variety of hydrolysis reactions have been observed on Soils and sediments.
In some cases, rates were markedly accelerated compared to, bulk solution, but
detailed understanding of mechanisms is limited and structure-reactivity re-
lationships appear to be available for only a few compounds.
-------
The review of Mabey and Mill (1978) is a useful point of departure for
summarizing the SRC for hydrolysis in pure or natural water where only H,OHO^
and H* are important catalytic agents. Organic chemicals can hydrolyze by a
variety of processes including both solvplytic and substitution mechanisms
catalyzed by H30+ and OH". We can narrow the scope of the enquiry by restricting
our interests to reactions proceeding at pH 7 and 25 ?C. General and detailed
effects of structure on hydrolytic reactivity, as measured by the halfrlife
at pH 7, are summarized in Tables 4 through 11. In some cases half-lives are
expressed, only as > or < 1 year or some multiple thereof.* By this
classification, one can quickly differentiate the reactive from the unreactive
chemical structures. For some classes, the information on LFER can be used
to make detailed estimates of reactivity based on structure.
Table 12 provides' an summary of hydrolyzable chemical structures that
will be expected to persist in water for significantly more than a year at pH
7 and 259C. These classes of chemicals probably tie ed not be screened in the
laboratory for hydrolysis using the current screening tests because these tests
are designed to estimate reliably only t:hpse half-lives of a few hours to a
few months; longer half -lives are not well-defined by the test method (Federal
Reg., 1979) and, indeed, current assessment procedures generally focus only
on chemicals that exhibit loss rates of days to weeks.
Photochemistry
Measurement Methods
The cutoff for the solar spectrum by the upper atmosphere is at about
290 nm; only absorption of photons by a chemical at this .or longer wavelengths.
can result in direct photochemical transformations. Direct absorption of sun-
light may result in cleavage of bonds, dimerization, oxidation, hydrolysis,
or rearrangement. No simple selection rules are available to predict the
specific chemical process that may occur, although some useful generalizations
have been found (Calvert and Pitts, 1967; Turro, 1978).
Quantitative aspects of direct photolysis in water, on soil surfaces,
or in the atmosphere have the same general kinetic relationships. The rate
of absorption of light, IA (rate constant ka) , by a chemical at one wavelength
is determined by e^, the molar absorbance; Ix, the intensity of the incident
light at wavelength X; and [C] , the concentration of chemical. At low concen-
trations of C where only a small percentage of the light is absorbed
The rate of direct photolysis of a chemical at wavelength X is obtained by
multiplying ka/,v by the quantum yield x, which is the efficiency for1
* • . . ,
This time limit is somewhat arbitrary but corresponds to the limits on
estimation .of rate constants using EPA screening test protocols.
15
-------
TABLE 4. • HYDROLYSIS OF ESTERS AT pH 7 AND 25*Ca'b
n C
Half-life (yr)
.Ri
Al
Al
H
Al
Ar. ' | "
Al
XCH2,XiCH
Ally!' ,. .
General SRC —-
...Ra....
Al .
Ar
Al
allyl
AI '. ••> ';;••• -•
Ar •- .', '
Al (X « Cl.^F)
•'AI•••••'••••'•'':'.', .• ./•'••''
> 1 (2-100)
« 1 .
< 1
» i
7
-•> 'i"'''.'. "'"'-''•••
-—'•''"''- ; Acid Catalysis
XArC(0)OEt
MeC(6)OAr . . , ,,„
:. '.'•.•',,.•. Base Catalysis
,.,v .^./,.,...p, ..
XArC(0)OAl
—^— Leaving Group pK,
p ='2.0
AlC(0)OArX: Wheti pKA of HOArX is less
than 8, 'tv- for hydrplysis: is < 1 .yr
'..'•'• , ' - , .': ' : '' • •.-..-. '....'..•
—-r —Steric• Effects '—:-^*- '"",'.'.
i MeC(6)OAl: As iiuik of A! increases,ffbo
Me to t-Bii; t^. increases by a factor of
•• '""' aitobs't 100T-'" -• " ••.•••••••••.••••;•.•••.• • -...••.
aFfom Tables 4.8 and 4.9 in Mabey and Mill (1978)
k« is the dominant term in k, .
CA1 * alkyl, air * aromatic.
dEurahtb (1969); ; '' ... • /-
16
-------
TABLE 5. HYDROLYSIS OF AMIDES AT pH 7 AND 25°C*
R1C(0)NR2R3b Half-life (yr)
Ri
Al
XA1
ClaCH
C13C
Al
XAr
XAr
ii
H
, H
H ;
H
Al
H
if _ . .. . . , . ; ,
H » 1
H > or » 1 for X = MeO, Cl, Br
H "' « i ! " •' ' '•'•"•''•_; i'"
H ''• •' • • - '•' "•' ':- ' "; .'•••• ••'::' ''-': •'•'•"••'"•''••'.•..., -.••.•;•.,'••••
-H • •' •'"» 1° " ••'••••''•'••• '"•'••.•:. ' ,'•'••
H » 1C, X - H to NO,
- H ••-•' •'•- • >» lc, ;X •- Me to'-Pr '';'''; '"'•
Table 4.10in Mabey and Mill (1978).
From footnote c, Table .4. ,
Estimated from H30+-catalyzed rate only at. 50"100°C.'
17
-------
TABLE 6. HYDROLYSIS OF NITRILES AT 25°Ca
••, -RCN :'
Al(Me, Et, Pr)
; v ;• ••/ Half-life (yr),
> 100C
•••> ioood
A.r2CH
At pH 9; slower at lower pH.
Footnote c, Table A.
cWidequist.
dZavaoiahu (1968) .' , , :
eBloch et al. (1973).
TABLE 7. HYDROLYSIS OF ACYL CHLORIDES AT 25°GC
fe
EtO.' .
RC(0)C1
General-
3- or in^XPh : . . . .
(X • * • NOi , Br , Med, ke)
LFER-^—
XPhC(0)Cl
Half-life
^10 sec
< AO sec
p « 1.57-2.0
c'd
^Kinivetx (1972).
Based on first-order rate constant for reaction with water*
^Kiniven , :
Hudson et al. (1970); ' rv '"'• •' /;>
18
-------
TABLE 8. HYDROLYSIS OF CARBAMATES AT pH .7 AND 25°C
' RiOC(0)NR2
• • - "
M Al
XI Ar
M Ar
M Al
4r v : . . Ar -1
(a- Al
\r • ' ... . A.r •
\r Al
*3N Al
XjN Al
XnCH3-n Ar
XArOC(0)N(Ar)H:
XArOC(0)N(Ar)Me:
XArOC(0)N..(Me)H:
XArOC(0)N(Me.)Al:
XArOC(0)N(Ar)H:
R,a " '" ' , Half-life, yr
R, ' '" :- ' '; ' ' •••,-••••
Al >> 1
Ar » 1
H » 1
H » 1
H •. • ; ..<•!••••'
.. H ... • '•' .••'••• ''• ':;..•'!••"
H . 1-150 . .
. '. X • Cl or F • .; . " ••. ' ..-'. ..'•.-;'
log kx » 15.2-1.34 PKaC
log kx - 13,6-1,15 pKad
log kx • 2.04 + 2.87ac ....
log kY - -1. 3-0. 26 pK d :
•A ' a
•'£..•< 5 yr if -pK •i'lO.Xat pH 8). ".,'.'.'. . •'"'
*5 ' ' ' 3 >• . ' . - • ' '••:••
c, > 1 yr for all values of. pK .(at. pH 8)
T . ••••••.••.•• \. • .. : '• 4 • • .. ;;
t, < 0.5 yr ;if pKg. < 12 (a,c pH 8) • -
Footnote c, Table 4.
bFrom Table 4.5 in Mabey and Mill (1978).
CWilliam3 (1972).
dWolfe ec al. (1978).
19
-------
TABLE 9. HYDROLYSIS OF ALKYL HALIDES AT pH 7 AND 25°Ca
RX .
Rb . .' .... X."; •; •':'•.. Half-life (yr)
Al . F • - , ' > 1
Al :/' . ' , ' Cl, Br, I,.. , . or j x , •.:.; . * ^
A10CH2 Cl " . < 1 hr
U_n _ ''.,•' Cl, Br n = 2-4 » 1;
From Table" A. l.Habey and Mill (1978)'.
Footnote c, Table 4.;
20
-------
TABLE 10. HYDROLYSIS OF PHOSPHORIC ACID ESTERS AND HALIDES AT pH 7 AND 25°Ce
RiP(0)R,R9
Ri Ra' '
Al AlO
Ar AlO ••„.
AlO AlO
,
i
ArO ArO
p-N02ArO p-NOjArO
Al Al
AlO MO
!A1 AlO
JA12N A1ZN
R3b Half-life (yr)
v ,/^ ,~v
.. Ph.os.pho nates RiP(0)(OR)s *" — »-«—•» .'. •" ..•'•• '...'-.'..'.."'
AlO : . » a •'•.•'•••
AlO (Al » Me or Et) ^ 1
ArO -v 1
p-rNOaArO « I
F ' « 1
F . < 1 .•
F ' ' ' .' • > 1
Cl • ' « 1
aFrom Tables 4.13-4.17, Mabey. and Mill, (1978).
'Footnote c, Table 4.
21
-------
TABLE 11. HYDROLYSIS OF EPOXIDES, IMIDES, AND CYCLIC ESTERS AT pH 7 AND 25°C
.. . ..-.•:• -
H
Me
Me
XCH2
XCH2
. ' ••; '• .'. -.•'-
Me
H -.,. . .-•,
LJ
. ,Ra/V^A
. H . . H .
H " _(H. ..
Me H
H H .
Me H
•'• . o.-X;= HO, 'Cl, "B'r "•••
H Me
, .- : ', H , . ,, •....i-H.-'.V-
S^'SSwJ
. p-Lactones ••''•
•Epokides and Itnides"
H
H
H
H
H
H
-Cyclic. Esters ;-
3*?
&-Sultones
Sulfates
Half-life (day)
..•••• 12
14
, 4 -
8-28
. v 16 , .'•
16
^ 1 yr(
« 1 yr
« 1 yr
< 1 yr
Table 4.7, Mabey and Mill (1978).
3From Table 4.18, Mabey and Mill (1978);
Ethyleneiraihei . ..,- ' . •'' ',
22
-------
TABLE 12. SUMMARY OF CHEMICALS PERSISTENT TO HYDROLYSIS: HALF-LIVES > 1 YR
AT 25°C AND pH 7
Category
Persistent Chemicals
Esters, RiC(0)OR?
Amides, RiC(0)NRaR3
Sitriles RCN
chlorides RC(Q)C1
Carbamates RiOC(0)NR3R3
Mkyl halides RX
Phosphorous acid esters and halides
RiP(0)R2R3
2poxides, lactones, sultones
AJ.1 Al esters of Al, Ar, or allylic.
' acids' '" •"'.'' • ..•;••:•• . ..;.• •• . • .• •
All amides where Ri -r R9 are Al .or Ar.;
only amides with halogenated alkyl Ri
hydrolyze ra;pidly.
All aliphatic ^or aromatic nitriles
'..''..
No acyl chlorides
All carbamates having only Al or Ar on
N and 0
'"Q\ Al*1 an(* polychloro- or polybromo"
methanes
All esters where Rj. is Al or Ar and Ra
and R3 are A10 or ArO (Phosphonates) ;
no esters where Ri 7 R3 are A10 or ArO
(phosphates); only esters where RI and
R.2 are AlaN and R3 is F (phosphono-r
halidates). : ••
Only hindered, bicyclic epoxides; no
simple lactones or sultones
23
-------
converting the adsorbed light into chemical reaction, measured as the ratio
of moles of substrate transformed to einsteins of photons absorbed. Thus at
a single wavelength, X . : , • -.,
where
kp(X) = ka(X)">X
The simplest and most direct method of using laboratory experiments
to estimate environmental photolysis rates in the field is to expose an aqueous
solution, a vapor phase sample, or a thin surface layer of a chemical to out-
door sunlight and monitor its rate 6f disappearance. At the same time photolysis
of another chemical having a well-characterized quantum yield and similar ab-
sorption spectrum should be carried out. this method will . take into account
variations, in sunlight intensity but avoids the heed for determining the detailed
spectrum or, the quantum yield for the chemical. ' • ' .. • .
Another method , for estimating environmental photolysis rates is based
on 'labbratbry measurements of .: at. : a. single wavelength ;; , .' ^ '...-; average suhlight
intensity (1^) data are available in the literature as a function of time of
day, season j and latitude (Mabey . et ; al.,, ... 1979 ; ,, Zepp and Cline, 1977) .
Since kp is equal to the product of ka and the quantum yield <}>,, and
since $ generally does not vary significantly with wavelength, the rate constant
in sunlight k ,^ is . ••• ; ' _' ..'.'.., •"'.';,!••'• ,-./; •-•!•-. •.••!•:•••.'•* '":: •;: '.;-; -• -'.:'-.''•'
and the half-life in sunlight is '• ''.
•• • -'-l:^:'...,' '. -.'(21)
Both computer and hand methods are available tb sum the products
of e^I^ over a wavelength range, and give a plot of '; the half-life of , the
chemical toward photolysis in water or, air as a function of the month of the
year and;, latitude (Mabey ;et al., 1979; Hendry et al.., 1979) ./ .. .
Comparisons were made at SRI between measured and calculated half-lives
for direct photolysis in sunlight of eight chemicals dissolved in water using
procedures described above; the comparisons gave excellent agreement, usually
within a factor of 2 (Smith et al., 1978).
Predictive. Methodblbgy
Equation (20) may be used ; to calculate an upper limit for kp by
assuming 4> - 1. This method is recommended as a screening tool since if the
rate constant is small compared to rate constants fbr other competing
.... ;" . ,..'. ..—••,- 24 ;.' . ./:•." • . •;:.
-------
environmental processes, no additional photolysis measurements are needed.
The foregoing discussion and equation (20) show that only the intensity
e and quantum yield at wavelengths in the solar spectrum are needed to predict
quite accurately the photochemical loss rate of a chemical in sunlight. For
the prediction, no information is needed on the actual chemical reaction oc-
curring. Application of SRC to photochemical reactions must then address two
separate problems: how does structure affect e, and <(>,?
There is no simple or succinct answer to this question. However,
some generalizations are possible because of a large body of empirical data :
organized on the basis of a slim theoretical framework (Calvert and Pitts (1967),
Baltrop and Coyle (1975), Turro (1978),
Spectral Properties
Equation (20) shows that the total rate; Iconstant for loss of a .
chemical by solar radiation is related to the sum of all absorption bands.
Thus, a chemical with a very weak tailing absorption spectrum,extending -into
the solar region for 50^100 nm could largely photolyze in less than a year if :
the quantum yield is reasonably high.* For this reason, .conventional.criteria
for weak or strong absorption spectra are not very useful. Table 13 lists
classes of chemicals that exhibit significant light absorption beyond 290 nm.
The list in Table 13 is by no means inclusive, but, as a general
rule, carbon singly bonded to carbon or to other more electro-negative elements
absorbs only weakly or not at all in the solar spectrum. Carbon multiply bonded
with electronegative elements does exhibit weak-tp-rstrong absorption between
290-350 nm, and conjugated structures absorb more"strongly and at higher wave^
lengths. Singly bonded heteroatoms such as OrO, S-0, and N-N also exhibit
weak-to-moderate absorption bands in the solar region. These familiar
generalizations provide the basis only for a crude separation of, chemical
structures into solar-active and solar-inactive categories. More detailed SRC
are difficult to formulate at this time because of the current limited knowledge
of the photophysics of excited states (Turro, 1978). ,
Energetics of Photoprocesses
The energy of light photons in the solar region decreases from 98.6
kcal/einstein at 290 nm to 35.7 kcal/einstein at 800 run.""" The kinds of
A hypothetical chemical with a constant absorbance e of 0.1 M"l cm"1 extending
from 290 to 350 nm will have a tu of 160 days if = 0.1, and a t^ of 16 days
fif 4> = 1.0 (Mabey et al., 1978). ' . " " ' *
The relation between energy in kcal/einstein and wavelength in nm is . :
photon energy = 28590/X . .
25
-------
TABLE 13. APPROXIMATE ABSORPTION REGIONS FOR
.',-.-' "•- , ORGANIC MOLECULES .:...'
Class
c ' • •
Aliphatics
hydrocarbons
fluorides, chlorides
bromides
•iodides
ethers, alcohols
aldehydes
ketories- . , , "
acids .
: esters
amides ;
'
' amines •
azines ''; •' -• " •'
azo ' ' ' •• •• ."• •'
nitro
nitroso ..•••'.
.'•': ; nitrite- "'."'" •' ' ' ':'•'-.'•'''''
nitrate • • ,. •: .
sulfide
....-«'••. , ..I".''1.,'
disulfide ' .' ..,;-,
Non-conjugated
Olefinics arid Acetylenes
.. hydrocarbons . .; ... . ,-.
-------
TABLE 13 (concluded)
Class
Aromatics
benzene
alkylbenzenes
halobenzenes
nitrobenzenes
aminobenzenes
azobenzene
phenols
• vinylbenzenes
acids
carbonyls •£
1
poly cyclic ''•
aromatics
heteroaromatics
Example
Benzene
Toluene
Cl- pr Br-benzene
Nitrobenzene
Trinitroluene
Aniline
Azobenzene
Phenol
*
Styrene
Stilbene '
4-nitrostilbene
Benzpic acid
Benzaldehyde
Benzophenone
Napthalene
Anthracene
Phenanthrene
Benzo[a]pyrene
Quinoline
Benzoquinoline
9H-Carbazole
Benzo [b ] thiophene
Region
.••'.' nm
" • '
200-280
200- > 280
200-280
200-330
200-450
200-3300
••'•-•'< 200- > 500
200-305 .
. .,200-300 ^
200-330
200-390
200-300
200-375
200-410
200-325
200- > 400
200-380
200-410
200-360
• • 200-380
200-390
200-312
Reference
a
a
a
a
d
• a - • • '/
' '•• a' • • • .'•' "
• : •?.-•• •
•;••;.-.•. •:.!•'. a •:.•'••• ,.
a • •
• , , '• a •••
" a- '
a
a
i
a
a
e •
, e
e
e ,P|I ,
e i
Spanggord et al (1979).
sSmith et al (1978).
27
-------
photoprocesses that can occur at specific wavelengths are therefore limited by
the energy available from absorption of one photon per molecule. Simple
homolysis reactions (Jt¥ -»• X» + Y») are limited by bond dissociation energies
of different bonds which vary from 112 kcal/mole for Ph-H to-. 38 kcal/mble for
RO-OR (Benson, 1976). Thus, a photbprocess that leads to. homolytic bond
cleavage can occur efficiently only below the wavelength limit imposed by the
bond dissociation energy. For example, photocleavage of the C-C1 bond (83
kcal/mole) in CH3C(0)C1 can occur only at wavelengths below 346 hin.
These comments also apply to concerted processes; however, the
energetics of the concerted process are very difficult to predict except that
the activation energy is always lower than for the nbriconcerted process in-
volving the same bond breaking and making steps. Thresholds for concerted
reactions will therefore be at Ibnger wavelengths.
'"'' • Quantum Yield Estimates
The simplest assumptions one can make about a photochemical quantum
yield is that it is zero, or one over the entire solar spectrum^ In the former
case, no-additional testing is heeded; in the latter case, k_, calculated from
.equation (20)-', is the upper limit for the rate constanti
More accurate estimates of ^ may be possible because SRC exist for .
the chemicals of interest or because a published value at one wavelength is
available. In either case application of equation (20) explicitly assumes
that ^ is constant over the range of wavelengths where light is absorbed.
This assumption is a reasonably good one providing the absorbed photon energy
is not b:elow the threshold energy needed ;to effect the reaction (see previous
section). In this case $ will quickly go to 2ero as X increases. Turro et
al. ~(1978) have discussed the exceptional cases where or products are
wavelength^dependent in solution. . ,
. Photochemists are particularly interested in studying efficient re-
act ions athat produce only a few produc:ts> usually in organic solvents or in
the" gas phase and in the absence of oxygen^ which is ah efficient quencher of
many photoprocesses. As a result, published quantum yield data on the bulk
of brgahi'C photoreactiohs provide, at best, only rough estimates of quantum
yields ifi aerated water or air; :. : ... . ;.
SRC for product quantum yields in'organic photochemistry are very
limited both by the number of measurements and by the complex photophysics
associated with even simple photolysis processes. The best studied processes
are those of ketones. A simplified photokinetic scheme for arylalkyl ketones
is shown below. The type II process involves cleavage to acetbphenone and
blefin via a biradical (BR). The competing pathways cbntrolling the fate of
the BR also determine the value of ty
28
-------
* '9 X
1-7T )9PhC(CH?)nX^
k. ^PhCO(CH2) X
' n
"< OK
PhC(CHa)n_1CHX (BR)
In. PhCO(CHa) X
. X
Ph§(CHa)n.lCHXvlc?
^ ^ PhGCH3 + olef in
PhX
Polar solvents such as alcohols (and presumably water) eliminate the back re-
action (kjj except where the substituent affects the relative
energy levels of the reactive (n-H*)3 and less reactive (H-fl*)3 states. Phase
effects on ketone photolysis may be significant if a-cleavage is the major
process because the solvent cage promotes recombination rather than dissociation.
hv. . . ''•'-' ' ''•"',,
RCOR' < *. [RCO + R'»] *- RCO -t- R'' ...
. cage , • ,. , . ..
Acetone exhibits a marked increase in cleavage rate on going from the liquid
to the vapor phase, but long-chain ketones, which can also cleave via a Type
II process, may exhibit smaller phase effects (Calvert and Pitts, 1967).
Table 14 summarizes quantum yield data for a- and B-cleavage of ketones and
indicates the magnitude of substituent, solvent, and phase effects on these
processes.
The complex photophysics may be additionally complicated by the
presence of oxygen. The effect of oxygen on photolytic processes is difficult
to predict, but some generalizations from other studies indicate that in the
presence of oxygen two processes can occur: (1) quenching by energy transfer
to give ground state species and either 30a or I0a and (2) chemical interaction
to give a peroxy-oxy radical
29
-------
Table 14; QUANTUM YIELDS FOR PHOTOLYSIS OF KETONES AT 25°C: STRUCTURE AND
SOLVENT EFFECTS •
Ketone
.Solvent?
-Ketone: aK;ieavage—-i-
Me2CO
Me2GO
MeCO-t-Bu
Cyclopentanone
2,2-Dimethylcy-
jclohexanone
PhCOt-Bu
PhCO'CH2Ph
4-MeOPhCOt-Bu
4-PhPhCOt-Bu
Or
V
Or
Or
Or
Or
Or
Or
Or
. 0.001
0.1
0.52
^ 0.2
> 0.4
A. 0.3 .
^ 0.4
^ 0.1
< 0.001
'':•'•'•'•''"''"''Ketone: B -Cleavage-
PhCO(CH2)aMe ... Or
PhCO(CH2)2Me : Al
|PhCOCHMe2 Or
PhCO(CH2)a-i-Pr . , Or
PhCp(CH2)CH(Me>OMe .Or
PhCO(CH2)3COsMe \ : . , Or
XPhc6(CHa)iMe';':;'-. .' '•';' Or
'••'•:,. X.= H ....••:' '•'•'" •..-'•'• .Or.
X = p-Meb Or
X = ffi-MeO • Or
X == p-Me Or
0.36
1.0
0.36
o.id
0;014
0.29
Reference
c
c
c
c
a
A
d
e
e
d
••d.
' b
Or *= nohpolar organic solvent^ ,V - vapor phase,
Al * alcohol>. ".'•;•'.'
Quantum yield in deaerated systeau
-------
*CR + 302 -«-*- PhCl
(1) PhCR + 3P2 -*-*-. PhCR + ^0, or 3P,
(2) PhCR + Pa —*r
Quenching by oxygen typically is diffusion controlled kq • (*V 10'-1010
M" l s~ x) and occurs with both singlet and triplet excited states. At one
atmosphere of air, quenching will limit the lifetime of triplets to < 1CT ' sec
(Turro , 1978). Unless some other fast intramolecular process intervenes,
oxygen quenching will significantly reduce the value of $. Kinetic analysis
indicates that quenching or reaction of keppne bir,adicals with oxygen will npt;
compete with cleavage, cyclization, or reversion $0 ketone,, Quenching pf the
precursor triplet species may be significant, however < Singlet oxygen, formed
in the quenching process, may react further with ground state molecules and
thereby lead to photosensitized oxygenation.
Reaction of oxygen to form bonds with excited species, is. also .
important and may serve either to quench the excited species or to form new
products. Acetone reacts with 02 to exchange oxygen, but cyclic ketones are
reported to cleave ;to new oxygenated products (Baltrop and Coyle, 1975).
Oxygen can also react with carbon radicals to trap them and thereby minimize
recombination and dramatically increase $.
R. 4- 0, -r-*~ R0a»
Aromatic rings exhibit a remarkable variety of photoprocesses in-
cluding coupling, rearrangement, contraction of expansion, 'and oxidation
(Barltrop and Coyle, 1975). Phptolyses of several aromatic chemicals in aerated
water show no specific pattern of quantum yields, dependence on oxygen concenr
tration, or product formation (Smith et al., 197.8). Indeed the frequent
citation of photooxidations of polyaromatics to form endoperoxides or quinones
may prove to be exceptions to the general reaction patterns of these chemicals.
Quantum yields for several aromatics phptplyzed in aerated pr.deaerated water
are listed in Table 15. In summary, detailed SRC for quantum yields are not
available at this time and a great deal of additional work 'is needed to provide"
the theoretical and empirical foundation needed for their development.
Oxidation Processes
Oxidation is a major loss process in the environment where significant
concentrations of oxidants are generated by photochemical processes involving
natural and anthropogenic light absorbers. In urban atmospheres, both H0»
radical and 03 are generated by a complex .photocycle involving N03, 02, .and
organic pollutants. In natural water, R02» radical and singlet Oa (J0a) are
generated by photolysis of humic materials dissolved or suspended in water.
Direct pho.tooxidation of many molecules also occurs by way of excited
31
-------
Table 15. QUANTUM YIELDS OF SELECTED PROCESSES IN THE PRESENCE AND ABSENCE
OF OXYGEN IN WATERa>b
Chemical; ...'
Behzanthracene
Benzpyrene<
Benz'quinbline
9-H-carbazole .. .....
Dibenzcarbazole
Oxygen
Present
Absent
Present
Absent
Present
Absent;
Present
Absent
Present
Absent"
aSolutions in oxygenated or atgbhated water.
bUnpublished data of Mill et al. (1979).
^Numbers in :parentheses ate powers of ten*
3.3(-3)
3(-3)
8.9(^4)
0
7.6(^3)
7. 6 (-3)
2.8(-3)
32
-------
state'interactions with triplet Oa, but the rates of'these processes are
controlled by the photophysics and photochemistry of excited states, not by
the thermal chemistry of oxidant"molecule interaction (see Photochemistry
section). ' . '
Recent reviews of oxidation in water and in air (Mill et al., 1979; Hendry
and Kenley, 1979) suggest that only a few selected oxidants are probably
important in either air or water. Table 16 lists these oxidants and their
average diurnal concentrations.
Table 16. OXIDANT CONCENTRATIONS IN WATER AND AIR
Oxidant Concentration, M
R03»
loa
H0«
03
1 x 10"
1 x 10-*2.
''."'• • , F ' ' ' '• ' ' ••' ','.'.
D
3.4 x 10- ,19
(8.2 x 10- " ppm)
1.7 x 10-'
(4.1 x ID"* ppm)
'From Mill et al. 0.978).
3From Hendry et al. (1979).
SRC For Oxidation
The concentrations and known reactivities of the major oxidants provide
a simple basis on which to separate reactive and nonreactive classes of chemicals
in oxidations in air or water. Those classes of chemicals that have half-lives
of more than one year may be considered as inert to these chemical oxidants
and thus can be excluded from screening tests for oxidation. Tables 17-20 .
list half-lives for a variety of organic structures in reactions -with R02>,
10a, K0«, and 03, based on the reasonable assumption that the rate of a specific
oxidation process follows the relation
-d(C]/dt
(22)
33
-------
TABLE 17 RATES OF OXIDATION BY R02» RADICAL
IN SOLVENTS AT 30°C
kpa
Class M~. s"1
Hydrocarbons 0.01
'Olefin 0.09
Benzyl i '.
Aldehyde 0.1
Alcohol . 0.01
Phenol 1 x 10*
Aromatic amine 1 x IP*
Hydroquinone 1 x id*
. .Hydroxylaiuine. : ,....;' " .',... •.•.'.!' x 10s '.:-'; '"•;•••':>;.,-
Hydroperoxide 1 x 10s
Polycyclic aromatic ,h,"i. x io3 , ; •';•:•,
Half -Life, t,b
; days
8 x 10s
9 x 10"'
8 x 10s
' 8 x 10"
8 x 10*
, 0.8
. '"'• ." 0.8
12 mih
-.-:.'.:•/; .'.•.^-;/120 miri •"•'"••' :; "•''" ' '
120 mih :
-.- '•;.:-:r;--B:V-... ...:...!'.••.:/•
^Per reactive X-H bond ; f rbm Hendry et al. (1974) and Howard (1972).
bfor I xlO~9 k R03; t, = in 2/8.64 x id"* kp in daysi - . ;
".•••'. : ..:: ••''•.'.'•••'.'..... - *i '••••'< •'•'• - - - ..•..••"•.••
34
-------
TABLE 18 RATES OF OXIDATION BY SINGLET OXYGEN
IN SOLVENTS AT 25 °C
^Per molecule from Foote (1975) and Mill et al. (1979).
bFor 1 x 10"ia M 10,; t, = Jin 2/klr. x lO"'* in sec.
*
Class
Alkanes
Alcohols and ethers
Ketones and Aldehydes.,
Aroraatics (including
amines and phenols.) ...
Alkyl halldes
Acids and esters
Unsubstituted olefin
Cyclic olefins
Substituted olefin.
Dialkyl sulfide
Diene
Imidazoles
Furans
Tr.ialkyleneamines
«-'°!' H
< 2 x 10*
< 2 x 10*
< 2 x 10*
< 2 x 10*
< ? x 10*
< 2 x 10*
3 x 103
2 x 10s !
1 x 10*
7 x 10*
1 x 107
4 x 107
1.4 x 10" '
8 x 10s
alf,Lifeb, t
> 100 yrs
• > 100 yrs
.,->.. 100 .y.rs. .
> 100 yrs .
> 100 yrs
> 100 yrs .
7.3 yrs . .
40 days,
8.0 days :
27 hrs .
19 hrs
4.8 hrs
1.0 hrs
14 mins ,-
.*. • .'. ' ' ' V" " '
• .•-.'• •
'"*'..
• - - -
35
-------
TABLE 19. RATE CONSTANTS FOR OXIDATION BY H0« RADICAL IN THE ATMOSPHERE AT 25"C.
.;•, : : ' Class- • ' ' .•"; '. .
n-Alkanes
(C* - C.)
iso-Alkanes
(G» - C,.)
Cycloalkanes
(C, - C«)
Halomethanes
(1-3 fluorines
or chlorines)
Haloethanes •
(1-3 chlorines
and 3-4 fluorines)
Butanone . .
pi- Alcohols
(C> - Cs)
sec-Alcohols •"'' .
(C, - CO . .
Ethers-
(C, - C.)
Terminal, olefins
(C, -:Ci) ••..•; :' .: '
internal olefins
(C, - C5)
Aroma Cics
benzene
toluene
xyienes
Triraethylbenzenes
Ethylbenzetie . ''
Propylbetiiene , .. : .. ':'
Cutoerie
o'-Cresol
Benzaldehyde
icrfkHo •;...
' .-M-,1 s-1 ••;•
1.3-5.0
0.6-2.9
0.7-4.1
1.2(^4) - 0,065
6(-3)-0.23
1.9
''"'2-
4.1
2.5-1.0
4.6-34 , •
29-90
0.82
3.5
. .. 5.9-^12 :
15-30
:4.«
•-'•-'3.5 •;'• • -. --•• •''•'
. • *-6'. ' ' ' '"" '.
20
7.6
.. .. • ; -b •
;-.; •:.•,. ,.L* ,,
. " Days
1.3-4.3
1.9-9.4
1.4-8
87-47,000
: 24-950
2.9
2.8
i.3-
0.6-2.2
"''•' '"- O'2-1.-2,.:.
: 0.0fr^0.2
'•'•:'.• 6.8
1.6
0;47-1.0
'-;•'. 0.2-0.4
. .... ' 1.3'
••''••• ' -1.6
.'•. . -;. i-2
0.3
; .' 0.74
aFrom Hendry and Kehley (1979).
bAssumes [HO»J is 3.4 x 1CT15 M and 10 hrs solar days: t,. * 0.69/k(3.4xlCT 19)x
(3600)x(10). . , . •••••:•
36
-------
TABLE 20 RATES OF OXIDATION BY OZONE IN
THE ATMOSPHERE AT 25°Ca
Class
-1 .1
M s
Alkanes . < 0.005
Terminal Olefina 5.8(3),
'Internal Olefins .. 110 (3)
Branched Internal .Olefins (300-900) (3)
Chloroethylenes < 60
Alkylaromatics < 60
Alkynes < 60
• Half^Life, t, days
' *5
> 2 (6)
0.2Tp.6
> 130
> 130
> 130 .
3From Hendry and Kenley (1979).
Per molecule.
CBased on 1 x 10~ M 09; t, = In/8.6 x 10" k- in days.
37
-------
If [OX].is constant equation (21) becomes
. ••'"." '' •'.. J. '".•."• ..'"'.' -dicl/dt ='-k^'ic] • : , ., (23)
and the half-life for chemical C is
t, = *n2/k0X (24)
Oxidation .by R02» ,
The range of measurable reactivities of organic compounds towards R02»
radicals covers about 1010 in kQy [Howard, 1972)]. Only chemicals that react
with rate constants.> 103 M"l s~will be oxidized in aquatic systems at
significant rates (Table 17). therefore, we.can exclude from testing any
compound that has only CH or aromatic structures and all of their simple
derivatives including carbonyls, esters, alcohols, halides, aliphatic amines,
and ethers.
Aromatic amines and phenols are two major classes of chemicals that are
reactive enough toward R02« radical to warrant detailed consideration regarding
SRC. Fortunately, because these compounds are useful as antioxidants, several
investigators have measured absolute rate constants for H-atom transfer to
R02» from many types of phenols and some amines :(Howard and Ingold, 1963;
Howard, 1973).
R02» '+ XPhOH' > R03H
Table • 21 summarizes the absolute rate constants (kinh) for selected classes
of phenols; . ;. values of p* for each subseiries of phenols are also listed. In,
all eases p+. is negative-,' indicating that the transition state has some positive
character stabilized by electron donor substituents. The reactivity of all
measured phenols ranges from a low of about 3 x 102 tC 1 s~ •.* '.,i or 4-CNPhOH to a
high of about 9 x lO4 M~ l s* l for k^jj for ,hydroquihohe extrapolated to the
&4me teinperaturei , Aromatic amines have been subject to less systematic study
than phenols; nonetheless1, the data listed in Table 22 shows that as a group,
amines are as reactive or moire so than phenols: values of kinh range from a
low of 9 x 102 K* s"1 for MeOiCPhNHMe to highs of 2 x, 10s M~ V &" 1 for diphenyl-
amines. Values of p are negative and are similar to those for
phenols (-0.8 to 0.6).
*
This assumption is usually referred to as a steady-state assumption and values
used for [OX] take into account the fact that oxidants depend on sunlight and
thus change in concentration diurnally, ,;
38
-------
TABLE 21. SRC FOR OXIDATION OF PHENOLS SY R0a« RADICAL IN SOLVENTS
Substituent
YT>U nn -it- f, R
.4-HO " '
4-Me
H • • . .
3-cr .
4-CN
p+ =* -1.49
+
p.;,- -3.7 i
+
. .P '. - -3.48
2/, / ,. » '\
. »4-!At-ou; 2
x io'4 ^nh ^i s^ ;: '
a
;.V : ...... •-. •?•??': '.-'-::-:;-^ '.'.V;.^.
• •! ; • ', _ •' 0,92".;. "' ' -;- '.";";. •';'
-.--... '• • "'• "•••• *)•,%$.•/• •••;'•'•'••;• "'•yv>.v:;;.,,.
• .' :'-; ."'' - °r?56 ••;:;.. - •• • -. ,^v^'
0.030
in styrene ' •:•-., . •
n chlprobenzene •. '_:'• ' • ' •
* 0.114 in.ipethyl methacrylate, ..
';•*•.•- " ' ' • ! . ,
-4— ArrnOH, at "^J/ ^ WiLn t— pUU* . • :
MeO
Me
H
Cl
CN
2.3 •
0.50'1
-p',-34.
\0.50 '
'0.10
p *» -1.00, r =? 0.94 in isopentane
-o-Alkyl PhOH at 65° C with polystyryl peroxy radical0^--
2>6-Me2-4-XPhOH
Me 2. '8 "''''."'"
Cl 1.2
CN • 0.73
H .0.89
p+ = -1.36 * -1.46 ±0.115
39
-------
TABLE 21 (Concluded)'
- .. Substltuent. X . .,. • . .. .
2-t-Bu-n-XPhOH :
H ' • ' •;.'.' .
4-MeO :--.. •••:... -
5-Me: .-. .:;:VY.'' ' '• : :' •
4-IN .. .,•...,...••••••: •• •-......
,..,••• '••'"',•.- •-•.• -..-f 'V • ^ . ,
"" •'•'•••• . ,, , .. .-. : .,:'• ' p = -l;-4o'
H • ..;••••'""-,
MeO ... ....'•-• •••'•• '•'
Me . ••!•, • ••
Cl . ...• ' • •• . •'-
CN
N02 . ., ..-.;. :.-.•
- • ' . ' i
p = -1.11
k. , M-1 s-1
inh ,
0.75
10.7
1.1 •
0.087
±A ATI
0.012 •
: I?
.0.80
•'"'" 6.32
0.077
^°-05;;^.V;
±0.026
^Howard and ingold, 1963a.
b"
Howard and Furimsky* 1973.
toward and Ingold, 1963b.
40
-------
TABLE 22 SRC FOR OXIDATION OF AMINES BY
R0,» RADICAL IN SOLVENTS
Amine 10~ k. ,M~ s~
inn
N-Methylanilines at 65pS'b-
4-H, .''"'' :- ' 0.4
4-Me ••• 1.2
3-Me 0.5
A-MeOC(O) 0.09
4-MeO X.9
P « - 1.6
-Oa,b
Diphenylamines at 65'
4,4'-Ha 4
4,4'-Me, ' 10
4,4/-(MeO)a 33
4,4/-(NOa), 0.16
4-MeO 20
3-C1 " . ' 1-8
4-NOa . . 0.6
P+= - 0.89
Napthylaminesat 40°Ca'
N-a-Napthyl 0.15
N-a-Naphthyl-N-Phenyl 1.25
N-0-Napthyl-N-Phenyl .10
N,N-Di-B-Napthyl 18
a In reaction with poly(peroxystyryl)peroxy radical.
b From Brownlie and Ingold 0-965).
41
-------
All these data were measured in nonpblar organic solvents. No useful
data are available to indicate the extent of solvent effects on the process,
for use in extrapolating to vater; however, Ingold and Howard (1963) have
speculated that p may increase with an increase in the solvent polarizability.
For the present we probably can use the relative values of k^ for phenols
and amines with few reservations, but the absolute values in water may prove
to be different from those.in organic solvents.
Oxidation by;. SingletjOxy^gen (1Qi) ' .
Zepp et al. (1978) demonstrated that sunlight irradiation of natural waters
led to production of 102 at,average concentrations of 1 x 1CT12 M, This^value
when combined with rate constants for oxidation a variety of organic structures
(Table 18) lead to predictions of very long half-lives for most simple structures.
Reactive molecules that warrant laboratory testing are electron-rich molecules
such as^branched olefihs, eneamines, polycyclic aromatics, arid sulfides.
• Estimation. Methods/ , ,. ..
. , , SRC , for X02 oxidations are very limited. Foote and Denny (1971)
measured the relative rates of oxidation of a series of substituted styrenes
and found that for allylic. oxidation correlation by the Hammett equation (a)
was satisfactory and p * MD.92. However, :epoxidation of the .styrene double
bond which accompanies allylic oxidation is :not correlated well, by d nor o+,
but is correlated by [a - 0.37 (o+-a) ]• to give-p = -0*87.
Oxidation by HO•Radical
' --HO*, radical' is'• tite jnost important oxidint''"ini'th'(i'''atiifespn'er6.iiv''ttv'reacts'
by H-atoin transfer and addition to double bonds arid aromatic rings. Table 19
lists irate constants for reaction of HO* radical with a variety of organic
structures. More detailed lists of rate constants are found in Hendry and
Kenley (1979) and Hampson and Gafvin (1977). Most organic molecules with -CH-
6r:-CH2- bonds or aromatic, react.rapidly .with half-olives of less than a day;
highly halogenated chemicals including, of^course, Freons are much less reactive
' of. unreactive; r•,.., .... v - .::-:^/'. '''v.•'.•'•. . •. .-•,'•••''••"''' v"'-: •' •;;/• ;;-''v "'-":--. ' . •-
;,; The.rate constant'kHQ fof Oxidation of, a specific structure can be
estimated using the additivity procedure Of Hendry and. Kenley (1979) which,
depending on the structure, requires estimation of the individual rate constants
for "H^atom transfer and addition. The expression for H-atom transfer is
} = £Vl*ikl (25)
in which a^ and p^ are substituent constants and k^ is the rate constant for
the i^th CH bond. Similar expressions have been developed for addition to
double bonds and to aromatic rings, the probable error.in values of kjj^ _
estimated:in this way is about a factbf of two (± 100%); A listing of
substituent constants is found in the report by Hehdfy and Kenley (1979).
42 :! '- . ' •••
-------
Oxidations by Ozone
Ozone is a selective oxidant that reacts only with electronrrich molecules
such as olefins, eneamines, some phenols and polycyclic aromatics. Rate
constants summarized in Table 20 show that of the structures listed only branched
alkenes react fast 'enough with ozone for this reaction to compete with oxidation
by H0« radical. A useful summary of specific rate constants for ozone oxidations
in the gas-phase is found in Hendry and Kenleys' report (1979).
Estimation Methods •
Huie and Herron (1974) were able to correlate rate constants, kg ,
for gas-phase oxidations with ionization potentials for olefins. As expected
the value of k0 increases with increasing ease of ionization. We are aware
of ho other SRC for ozone reactions in the atmosphere.
Physical Transport . '-.-.'-
Two major transport processes which can control environmental concentra-^
tions of chemicals in specific locations are sorbtion to sediment or soil and
volatilization from water, air. Movement in soil, run^-off, leaching to water
and bioconcentration in organisms are beyond the scope of this report.
Volatilization
Theory and Measurement. • ' . , " .-•''.
Volatilization of chemicals from water to air is now recognized as
an important transport process for a number of chemicals that have low
solubility and low polarity; vplatilizatipn from surfaces is also a major
transport process for many chemicals deliberately applied to fields. Despite
very low vapor pressure, many chemicals can volatilize at surprisingly rapid
rates owing to their very high activity coefficients in solution.
Mathematical expressions for the rate of volatilization from water,
have been developed by tiss and Slater (1974) and MacKay and Leinonen (1975).
The rate constant for volatilization from water (k N is given by the relation
vwj -
(25)
where
A =» Surface area (cm2)
V = Liquid volume (cm3)
Hc = Henry's law constant "(torr M~ l)
KL = Liquid film mass transfer coefficient
(cm hr~l)
KG = Gas film mass transfer coefficient
(cmhr~l)
43
-------
; R = Gas constant (torr °lT1 H"1)
T = Temperature (°K)
Mackay and Wolkoff '(1973) showed that an estimate of H can be
obtained from c
(27)
where Pgat is the .vapor pressure of pure chemical (of the hypothetical
super-cooled .liquid, if S is a solid) and [C]sat is the solubility of C in S
(mol liter"1). Equation (26) simplifies if H > 1000
where.mass transfer is liquid-phase limited; if H « 1000 then equation
becomes
(29>
and the process becomes limited by. gas phase mass transfer.
For high volatility, compounds a simple relative measurement for
volatility becomes possible because the rate constant for volatilization of
the chemical is proportional' to the rate constant for volatilization or re-
aeration of oxygen from .the same ^solution over 4 .range, of .turbulence ;- • '••
. ...;• ..... -=„ ? , (30,
; - . '• ..;..,. - vw ...... • ,. . . ..'.•• •'
If the value of k 2 in a real water body is known then
....,' •'' VW • '..-•'.". •••'..' ' ' '•':-1 " • ' ' •- •'
' •' . ' ""'..-•:' -p: ''•• - ...... ...... ^ • •. . • •'•••- j • . ' ••
. •, ,' k (water body) «nk * water body (31)
•'.' ....,:.' '!'•.. '•'•••' . vw .••.••:.•.. .• ., • '.'.••••• vw . . ; ... .• • • '":'••''.•••.'••' .•'•••
,;, .Spencer et;al. have 'reviewed volatilization processes from soil
surfaces . (197 3) * The .overall process .is .complicated by variable contributions
from volatilization of the chemical from the water at the surface, evaporation
of water itself and the wick effect that brings more water and dissolved chemical
to the surf ace ^ initially volatilization of the chemical from the surface
water- will -be rate controlling and equation (26) Can be used to estimate the
fate constant, this model will fail as a concentration gradient of chemical
is established through the soil column and at this time no simple laboratory
measurement will reliably measure the process in a way that can be
extrapolated to the field. "". ; : ^
• ;;:';••.'./' Predictive, Methods .- •'•.• '•/ .;•;•;"/". ; _ '. •..-•..;.'
-------
Smith et al. (1978) have shown that for chemicals with Hc > 1000 torr
M~ 4 a reasonably good estimate of the ratio k^/k^ can be made from the rela-r
tion between the ratio of volatilization rate constants and the ratio of
molecular diameters for Oa and the chemical
-F* (32)
k°a DC
vw
For chemicals having HC < 1000. torr M" l no satisfactory estimation
proved ure is available.
Values of HQ can be estimated from equation (27): if both solubility
and vapor pressure are known at temperatures close to 25°C. We know. of no re-
liable SRC that might be used to estimate Hr or solubility.
T
Sorption to Sediment and Soil ' ; ' •'.-••"
Measurement Procedures
Sediments and soils are complex mixtures of alumino-silicate minerals
(clays), metal oxides, water and humic materials. The proportions of these
components will vary widely from one source to another as will the particle
size distribution.
Many organic chemicals especially those that are nonpolar and -insoluble
in water, sorb strongly to sediments or soils. If the fraction of chemical
sorbed to sediment or soil is large, the overall loss rate of chemical by other
transformation processes will be slowed; in effect, sprption serves to buffer
the concentration of chemical present in the aqueous phase [see equation (12)].
In .some cases reversible sorption to sediments or soils may be followed by ir-
reversible transformation of the chemical in the sorbed state such as reduction
of carbon-halogen bonds (Williams and Bidleman, 1978). Possibly other
transformations may occur as well.
The equilibrium ratio of sorbed to non-sorbed chemical on a sediment
may be expressed as an equilibrium constant (constant temperature)
Kg => tC]g/[C]w (33)
The concentrations of C in sediment or soil are in yg ml" * ; for water 1 ml =
1 g, and Ks becomes dimensionless . Strongly sorbed chemicals such as
benzo [a]pyrene or mirex have Ks > 10s and weakly sorbed chemicals such as
nitroaromatics or quinoline have K < 10*.
s
Values of KS for a single chemical will vary with the composition
of the sediment. To place sediments in a more nearly equal basis the value
K can be expressed as
S
Kg = AKsc (.34)
45
-------
where A is the fraction of organic content expressed as mg C per mg sediment;
thus Ksc is a sorption constant corrected for the organic content. It follows
from, equation (34) that ifV.a''.* 1 then Ks is'equivalent, to'a partition co,effi-
cient such as the octariol-water coefficient"'(K :).. • ' v." •
•.••:•..•'' '•••'. , .... ' '•-.''• '.'••' , • OW ' • '
Predictive Methods '"'.'
Rece'nt studies and correlations by Karickoff et al. (1978) and by
Kanaga and Goring (1978) bring out the close direct relationship between Ksc
solubility, bioconcentration and the organic content of the sediment or soil.
Other studies notably by Chiou (1977) have shown a similar relationship between
K and solubility such that
log KQW = n log(solubility) + c (35)
Smith and Bomberger (19?9) have taken the data of Karickoff et al.,
Kanaga and Goring, and Smith et al. (1978), rescaled the data to one coordinate
set, and developed the following regression relation . .
•.,••••- .- •.•'•"'. ....': " .'... f log K = -0.782 16g[C] - 0.27, (36)
.'.'•' SC . , • • • •
; -.tpi is solubility in M '
Figure 1 shows the combined data plotted as log k0(;. (or Ksc) versus log
(solubility) and the regression line. Using equation (36) and the solubility
of the chemical in water the investigator can estimated Kgc within a power of
ten for most non-polar chemicals> -, an..accuracy Sufficient,;: ,in,most. cases, for
•screening purposes.••• • :p • • ••' • : •'••.:-•••••••->>••• ^ •••••• •<•*•*•• ••«•<,••:*•.?••'•••-. -v •...•..,-./•,.•,........•....
46
-------
108 =
i i.niiiij 11 [iniij i 11 niii| 11 mm) I-1mm] 11 mill) 11 mm) 11 mui
A Karickhoffet al. 41978)'
p Smith et al. (1978)
• Kenaga"and Goring {.1978),
11 i
10-11 10-10 1Q-9 1Q~8 10-7 lQ-« 10-5 10-4 10-3 1Q-2
SOLUBILITY — moles liter^
Figure 1. Soil or Sediment Partition Coefficient of 'Chemicals Versus
Solubility in Water (Smith and Bomberger, 1979).
47
-------
SECTION 4
METHODOLOGY FOR FATE ESTIMATION
The following calculational procedure illustrates how a. specific chemical
structure may be dissected into molecular fragments from which estimates of
rate or equilibrium constants for each environmental process may be made.
.Fictitious physical property data are-supplied to illustrate the use of the
methods, . ' - . • '. > ... ' . ''.''.••..•'' >'"""'•' :
Example; 2-n-Butylcarboxy-5-hydroxynapthalene, 5rHOCioH8T2-'C(0)0-n-Bu
Solubility: 6 x 10"* M at 259 (FICTITIOUS) f . :
Vapor Pressure: 8 x 10"fc torr at 25° (FICTITIOUS) \
uv Spectrum: 200-360 ran (FICTITIOUS)
Step 1. Calculation of Rate Constants for Chemical Transformation.
a. Hydrolysis - the ester funct ion,' nvbutylcarboxy, wil;l..hydro'lyze.
Table 4 shows that esters of the type ArC(0)OAl will hydrolyze at pH 7 and
25,C with half-lives > 1 yr. More detailed examination of data in Mabey and
Mill (1978) shows that these esters have a dominant base-catalyzed reaction,
down to pH 4.5, and that although no napthyl esters are listed, the lifetimes
for -simple aromatic esters is long enough at pH 7 to make lifetimes at pH 9
3-6 months or longer. ... . . . .....
b. Photolysis - the uv spectrum shows that the .chemical.will absorb
in the solar region. The detailed spectrum taken in 50% acetbnitrile/water
(v/v) (not shown) is tabulated using the intervals proposed by Mabey et al.
(1979) for solution photochemistry. .
\ '
UV Spectrum of Napthylester (FICTITIOUS)
Absorption
Interval Interval Range Coefficient
Center, nm nm If * cm" *
299 296.3-301.3 160
304 301.3-306.3 102
309 306.3-311.3 . • 57
314 311.3-316.3 32
319 316.3-321.3 13
323 321.3-325.0 3
340 325.0-355.0 0.7
370 . 355-385.0 . , . 0
Preceding page blank
49
-------
Spectral data are now combined with sunlight intensity data for summer season
over the same spectral range: in equation (20) to calculate, a maximum value of
kp. Intensity'data are taken from Table 3.2, Mabey et al. (1979). '
Solar Intensity, Summer
Interval .. Intensity, L^a
Center^ nm einsteins cm"2.day"1
299 A.4(-4)
304 . . 3.2(-3)
309 9.6<-3)
314 2.0(-2)
319 3.0(-2)
323 3.0(-2)
340 3.5(-l)
aRpunded off to. 2 places.
•Since ••'•', ..•••••.•• ' '. . ''•"•' ' .• •- '' •'••: .• . .. .••-.•-••:,• , ' '• .,.,-.• .. ^
. , . '..: •-•'"_' .:. .';. k.' • <(>Ze.L, (20)
-,', P..AA
we assign.a value of $ - 1 and £6inbiti£ the £* 'and L-* ' data to give
-.,..- A A •
k •- 2,3 day"1
A similar procedure could'be carried out for photolysis in the atmosphere bqt
because the result would be similar - intensity data and intervals are slightly
different (Hendry et al., 1979) - there is no need to calculate both results
in the zero, level screening procedure; ;!- • n ,....•.:. .
; c; Oxidation - oxidation "in water may b& an important fate for
this chemical-since it has a phenolic structure. Tables 17 and 21 show that
ci-naphthbi has k^0 of ^ 10s M*-1 ,'srli! the effect of a C(0)OR group in the
a-ppsition iii naptfialehe on the reactivity of a OH group in the 5*position is
to deactivate it but the magnitude 6f the eff^c.t is uncertain; If we assume
that reactivity in the: napthalehfe series foll6ws!that in benzene then a C(0)OR
giroup will slow the rate by * 3x (table 21) and k^Q2 £ 3 x I0fc MT l s~l. the
half-life of the chemical is then calculated by assuming that the average
concentration of R02» in water is 1 x 10"* M (Mill et al., 1979).
t, *• ln2/(3 x 10") x (1 x 1CT')
• . , *
•» 2.21 x 10" seconds
: . "=6.4 hrs.
50 :,
-------
Oxidation by J0a will not be important for this chemical (Table 18, tjj > 100
yrs) . Oxidation by H0« radical in the atmosphere will be important since
aromatics react with kjj0 > 1 x 10* M~ J s~ a (Table 19) and phenols are generally
more reactive. The half-life in the atmosphere can be calculated by assuming
that the average concentration of H0« in air is 3 x 10~xs M (Hendry et al.,
1979).
t^ - ln2/(> 1 x 10') x (3 x icr 13)
< 2.3 x 10* sec
< 64 hrs , . '•.-. '••'' • ••• . ' •,..••
< 6,4 days as 10-hr solar days
Oxidation by ozone will not be an important competing process. for this chemical'
(Table 20). . '.' ' •' -" ' ..'''."'•' :'-:-;;! . '"'''•
Step 2 - Calculation of Rate or Equilibrium Constants fpr Physical
Transformation
a. Volatility '- the .approximate method for estimating H from
equation (27) gives
•••'••• :••-,. •••••«• '8 x IQ"1 torr.- ..• .-:•..•'•••'•.•-
•' : ' .c; 6 x';10~6 M
f 130 torr JT1 at 25 °C . .
This value of Hc, lying between 10 and 1000 torr Tfl, cannot be ug.ed to clearly
exclude volatilization as an important; process. Testing is needed..
b. Sorption to sediment, r solubility data and the regression
equation (36) are used to calculate a value of K • . '
' SC . ' ., - .
log K > r.0.782 log[6 x 10"'] - 0-27
' • SC . '
^ 3.8
K = 6.5 x 103
sc
Step 3 - Summary of Estimated Rate Constants
a. Chemical transformations
Process 4
Hydrolysis
•-.. pH 5, 7 , ' .' •'• : . • •• ••>'!• yr ...;.-• "' ' "'' '
. .pH 9 •.".'..•-.•• '• .".... • 3-6 months,.
'51
-------
.Photolysis (=l) 0.3 day
(air or water)
.. • Oxidation . .
... water (R0a»). , , ' 6 hrs
air (H0») < 6 days
102 or 03 > 10 yrs
b. Physical transport • .
Volatilization He :•'.. 130
' ,'„ . Sorption K . . 6 x 103
sc . • .
the results of zero-level screening indicate that 2 n-butylcarboxy-5-hydroxy
•naphthalene may photolyze^ oxidize and hydrolyze in water rapidly enough to
warrant laboratory screening tests 'for sunlight photolysis in air and water,
•oxidation-by Kda» radical and H0» radical and hydrolysis at pH 9. No tests
•would appear to be needed for hydrolysis at pH 5 or 7 or oxidation by 102 or
03. Calculated values of HC and Ksc indicate a need for screening measurements
to evaluate volatility and sorption. .••••••' •
: A decision to carry but oxidation and photolysis both in air and water
would depend on whether volatilization measurements indicated the probable
importance of movement from air to water tb air. If the volatilization half-life
were more than ten times longer than half-lives for transformations in water,
probably ho tests in air wbuld be needed. .
The foregoing calculations illustrate the value of estimation methods for
calculating important kinetic and thermbdynamic constants prior to testing to
.eliminate some, processes;,from- the testing Scheme. The-example-•se'lected'and
its property data were deliberately chosen to inaxiMze the number of possible
processes the chemical might undergo;. Had we selected instead
2in-butylcarboxy-5-methoxynapthalene, reference to Table 17 would show that
•bxidatioh in water would be too slow tb warrant testing but the rates of Hy-
drolysis, photolysis and oxidation in. air would be relatively unaffected* .
Probably the sorption and volatility would.be mbte impbrtant since.solubility
would be-lower and the vapor pressure higher^ • v< . .., •';'-':'''•'•';'••••-..' •.; ::;
52
-------
SECTION 5
CONCLUSIONS AND RECOMMENDATIONS
From the foregoing discussions several conclusions and recommendations
for action emerge: . •
• Many of the important physical and cheioical processes in water and
air are well-enough understood to allow quantitative estimation'of
rate or equilibrium constants for specific chemical structures. :v. ..... .
feero-level screening).
• Two kinds of SRC are useful for application to environmental processes:
general SRC provide a sound and general basis for segregating many 10 <
1000). These should be developed.
• No reliable kinetic data base is available.for.microbiological :
transformations'from which SRC for these processes might be developed. •
Studies by Wolf et al. (1979) suggest ,a, relation between chemical and ,
53
-------
microbiblogical irate constants for hydrolysis. Additional studies of
this kind should be performed.
Transformations in soil and sediment are poorly understood. Only if
more detailed studies are performed to characterize the chemical species
involved in these processes will development of reliable SRC be possible.
Needed studies include hydrolysis, oxidation, photolysis and reduction.
. ' ' • ' V .
SRC now available can be used in a systematic and objective manner for
zero level screening; the methodology is an efficient and reliable
technique for selecting out those laboratory screening tests needed
for fate assessment thereby saving considerable time and money both
for EPA and chemical manufacturers. ,
Methodology for using SRC oil zero level screening needs to be optimized
and systemized in order that scientifically literate but inexperienced
personnel can use these methods reliably. Eventually computerized
decision-tree analysis programs should be developed but "hand" analysis
using tabulated SRC data is also efficient and easily used. ,
EPA should develop a comprehensive instruction manual for zero level
screening. The manual will include a logic-key decision sequence .to
characterize possible important processes for a specific molecular
.structure and a detailed tabular listing of equations, environmental
properties and rate constants for these processes. The decision
sequence will be keyed to the tables and equations. If properly
designed, non-expert personnel can use this manual for zero-level
screening with a high degree of confidence* ,,, ...', •.. , • '.,•• ••',.,. • : '
54
-------
REFERENCES
Baltrop, J. A., and J. D. Coyle. 1975. Excited States in Organic Chemistry.
John Wiley and Sons, Inc. New York.
Benson, S. W. 1976. Thermochemical Kinetics, Second Edition. John Wiley,
New York. •, • ' ' • •••"•• • . ": '."••• • ••'••;'... :".•/'"'
Bloch, S., F. Ravlin, A.Toupance, and R. Buyet. 1973. Etude de la de-
composition du malodinitriie en solution aqueuse.alkaline.. : Comp.:Rendu.,.,
Acad. Sci. Paris 276:915. •'-""•" " '" ' ' ' . , '':":':";':':
Brownlie, I. T., and K. U. Ingold. 1967. The Inhibited Autoxidation of Sty-
rene. Part VI. The Relative Efficiencies and .the'.Kinetics for .Inhibition
by N-aryl Anilines and N-alkyl Anilines, Can. J, Chem. 45:2419.
Calvert, J. G., and J. N. Pitts. 1967. Photochemistry. John Wiley, New York.
Chapman, N. B., and J. . Shorter. 1972. -Advances in •Linear Free Energy ,.
Relationships. Plenum Press, New York.
Chiou, G. T., B. H. Freed, D. W. Schmedding, and K. L. Kohnert. 1977. Partir
tion Coefficients and Bioaccumulation of Selected Organic Chemicals.
Environ. Sci. Technol. 11:475.
Euranto, E. K. 1969.-. Chemistry of Carboxylic Acid Esters. S.,,Patai,. Ed.. ""
Interscience Publishers, New York, p. 505-588. •" ' -••."''•'"''.
Exner, 0. 1972. The Hammett Equation -' the Present Position, in Advances in
Linear Free Energy Relationships. N. B. Chapman and J. Shorter, Eds. ..•: -
Plenum Press, New York. pp. 1-70.
Exner, 0. 1975. Range, Validity, and Precision of Correlation Equations.
Coll. Czech. Chem. Comm. 40:2781-9.
Federal Register. 1979. Toxic Substances Control: Discussion of Premanufacture
Testing Policy and Technical Issues; Requests for Comments. ^larch 16
Part IV, pp. 16240-16292. .
Foote, C. S. 1976. Free Radicals in Biology. W. A,. Pryor, ed. Academic ,
Press, Inc., New York. ,''
Foote, C. S., and R. W. Denny. 1971. Chemistry of Singlet .Oxygen. XIII.
Solvent Effects on the Reaction with Olefins. J. Am. Chem. Soc., 93:5168,
55
-------
Hendry, D. G., .T. Mill, L. Piszkiewicz, J. A. Howard, and H. K. Eigenmann.
1974. A Critical Review of H-Atom Transfer in the Liquid Phase. Chlorine
Atom, Alkyl, Trichloromethyl, Alkoxy and Alkylperoxy Radicals. J. Phys.
Chem. Ref. Data 3:937. . - ... •• - •
Hendry, D. G., T. Mill, and W. Mabey. 1979. Test Protocols for Environmental
Processes: Oxidation and Photolysis in Air. EPA Report (Draft). EPA
Contract 68-03-2227. .'-''.
Hendry, D. G. , and R. Kenley. 1979. Atmospheric Reaction products of Organic
Compounds. EPA Final Report. EPA Contract 560/12-79-001.
Howard, J. A., and K. U. Ingold. 1963. The Inhibited Autoxidation of Styrene.
Part II. Can. J. . Chem. 41:1744.
Howard, J. A. 1972. Absolute Rate Constants for Reactions of Oxyl Radicals.
•• Adv. Free 'Radical, Chem. 4:49. :
Howard, J. A, , and E. Furimsky. 1973. Arrheniiis Parameters for Reactions of
tert-Butylperoxy Radicals with Some.Hindered Phenols and Amines. Can. J.
Chem. 51:3738-3745; .
Hudsen, R. F. , and J. F. Wardill; 1950. Mechanism of Hydrolysis of Acid
Chlorides. Part I. J. Chem. Soc. , 1729-1733.
Kanaga, E. E., and C.A.I. Goring. 1978. Relationship Between Water Solubility,
Soil Sorption, Octanol-Water Partitioning and Bibconcentration of Chemicals
in Biota. ., Preprint, :Amer. Soc. .Testing Materials, '.Third "Aquatic •,;•.,,.,,•.
Tbxicoiogy Symposium, New brleans, Louisiana, October 17-18^
Karickoff, S. W., D. S. Brown, and T. A. Scott. 1978* Sorption of Hydrophobic
Pollutants on Natural Sediments. EPA Internal Report, Environmental Re-
search Laboratory, Athens, Georgia.
Kiniveri, A. 1972. Chemistry of Acyi Halides. S. Fatal / Ed. Interscience
'.'•,•'•.-., Publishers, New York.;. • ...:•. '.'....'.. • .. . •"/. ' ....
Liss, P'«' G;» and P: G, Slater; 1974* Flux of Gases Across the Air-Sea Inter-
' v' : Natute 247:181. .. ".. • : -•• '.'•''.'.';' .'..-. • ••.• .
. Mabey> W 1 ,Rv , T * Mill , and- D . G . Hendry . , 19 79 i ''.' Test Protbcbls f ot Environ*
'mental Pfbcesses: Photolysis in Water. EPA Report (Draft). EPA Contract
.68-03-2227.
Mabeyj W. R. , and T. Mill. 1978. Critical Review of Hydrolysis of Organic
Compounds in Water Under Environmental Conditions. J. Phys. Chem. Ref.
Data 7:383.
Mackay, D. , and A. Wolkoff> 1973. Rate of Evaporation bf Low-Solubility
Contaminants .from Water Bodies to Atmosphere. Environ. Sci. technol.
'•'•• 7:611. .' ' '••/.. ;, ;. • . ' ; . .
•..- '.'• ...•;--'•. ;/-: :. ', • ' •.'••'.-•'- '. • .' : 56' -v ' ' • • .
-------
Mill, T. 1979. Data Needed to Predict Environmental Fate of Organic Compounds.
Proceedings of Symposium on Environmental Fate held at meeting of Amer.
Chem. Soc., Miami, FL, Sept. 1978. Ann Arbor Publishers, in press.
Mill, T., W. R. Mabey, and D. G. Hendry. 1979. Test Protocols for Environ-
mental Processes: Oxidation in Water. EPA Report (Draft). EPA Contract
68-03-2227.
Perkampus, H. H., et al. 1971. UV Atlas of Organic Compounds. Vol. I-V.
Plenum Press, New York.
Shorter, J. 1972. Advances in Linear Free Energy Relationships. N. B.
Chapman and J. Shorter, Eds. Plenum-Press,.New York. p. 71.
Smith, J. H., W. R. Mabey, N. Bohonos, B, R. Holt, S. S. Lee, T.-W. Chou, D.
C. Bomberger, and T. Mill. 1978. Environmental Pathways of Selected
Chemicals in Freshwater Systems, Part I. EPA Report 600/7-77-113.
Smith, J. H., and D. Bomberger. 1979. Unpublished results.
Spanggord, R. J., T. Mill, T.-W. Chou, W. R. Mabey, J. H. Smith, and S. Lee.
1979. Environmental Fate Studies on Certain Munition Wastewater
Constituents. Phase I Literature Review, Final Report U.S. Army Medical
Bioengineering Research and Development Lab. Fort Detrick, MD.
Spencer, W.'F., W. Farmer, and M. M. Cliath? 1973.- Pesticide Volatilization.
Residue Rev. 49:1.
Turro, N. J., V. Ramamurthy, W. Cherry, and W. Farneth. 1978. Effect of Wave-
length on Organic Photoreactions in Solution. . Chem. Rev. 78:125-144.
Wagner, P. J., and A. E. Kemppainen. 1968. Is There Any Correlation Between
Quantum Yields and Triplet-State Reactivity in Type II Photoelimination?
J. Am. Chem. Soc. 90:5896-97.
Wells, P. R. 1963. Linear Free Energy Relationships. Chem. Rev. 63:171.
Widequist, S. 1956. The Alkaline Hydrolysis of Acetonitrile. Arkiv. for
Kemi. 10:256. :
Williams, A. 1972. Alkaline Hydrolysis of Substituted Phenyl n-Phenylcarba-
mates. J. Chem. Soc. Perkin II. p. 808.
Williams, R. R., and T. F. Bidleman. 1978. Toxaphene Degradation in Estuarine
Sediments. J. Agric. Food Chem. 26:280.
Wolfe, N. L., R. G. Zepp, and D. F. Paris. 1978. Use of Structure-Reactivity
Relationships to Estimate Hydrolytic Persistence of Carbamate Pesticides.
Water Res. 12:561.
Wolfe, N., et al. 1979. Unpublished results.
57
-------
Zavbianu, D. 1968. Structure and Reactivity of Nitriles. An. Univ. Bucuresti,
Ser. Stiint. Natur. Chim. 17:41-66.
Zepp, R. G., N. L. Wolfe, G. L. Baughman, and R. C. Hollis. 1978. Singlet
Oxygen in Natural Water. Nature 278:421.
Zepp, R. G., and D. M. Cline. 1977. Rates of Direct Photolysis in Aquatic
Environments. Environ. Sci. Technol. 11:359-366.
58
------- |