PB84-233287
New and Revised Chemical Fate Test Guidelines
October 1984
(U.S.) Environmental Protection Agency
Washington, DC
Oct 84
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
-------
PB84-233287
CG-1410
October, 1984
PARTITION COEFFICIENT (n-OCTANOL/WATER)
ESTIMATION BY LIQUID CHROMATOGRAPHY
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
• r’JrtI , ..n r Jnl 1W. 2.
PAGE
4. Tills and $ubtlt$s
3. Ricipienls Accs Ion No.
P 8 2332S7
. Itepon D.t.
Now it d ftc’vjscd Chcmic l 1t e Test uidc1ines
7. Aiatho.(s)
6.
8. Performing Organization Rept. No.
12. Spofisoring Organization N .m. sod Address
Same as box 9.
IL $uppl.msntsvy Notes
13. Type of Report & Period Covered
Annual Report
14.
/
S.. Instr uction, on R.v.rs.
9 P formIng Organization N.m. and Address 10. Proi.ct/Task/Worh Unit No.
Office of Pesticides and Toxic Substa ices
Off ice of Toxic Substances (TS—792) 11. ContractiC) or Grant(G) No.
United States Environmental Protection Agency
401 11 St.., S.W.
Washington,_D.C. 20460
Abstract (U,nlt 200 wfds)
These documents constitute a set of 2 new chemical fate test guidelines
and support documents that will be added to the chemical fate test
guidelines and support documents that had been published by NTIS in
October, 1982. There is also a major revision of one test guideline
and support document and minor revisions of 4 test guidelines and
support documents that had appeared in the October, 1982 publication.
These revisions were made in response to public comments.
17. Document Analysis a. 0 . .c,iptor. -___________ ______
b. Id .ntiflsrs/Qpen.Endsd Tarms
c. COSATI r..Id/s.t up
1$. Availability Statement 19. Security Class (Thit Reporll 21. No. oIPsgss
Unclassified
Re lease Unlimited 20. SScurit Ct$s (Tl sp; 1 l22P,ic.
Unclassified
.s ANS1-239. )
OPTIONAt. FOAM 212 (4—in
rerm.riy NTiS-B)
Qsparti’nsnf of Commerce
-------
CHEMICAL FAfE rEsT1N GUIDELINES
TABLA OF COt ITEr fS
(X •— I USt)
x;—I. RR)
CG—1150
CG—1 200
CG—1 250
CG—1 300
W— 150
(2G —1 400
CG— I A 10
u ;— [ 450
CG—1500
CG—1510
CG— lb O O
CX—i 700
CG—1710
CS—14 [ U
CS —I 500
CS—b LU
(2S— Ib O U
CS— 1700
CS—1710
I 9U 2
1982
1982
9
1982
1982
1982
1983
[ 984
(.982
1983
1983
1982
1983
1983
Support
Guideline I cument
No. No .
Most Recent
te
o Issue
Guideline/Support EkDcurr nt Title
Pth ICAL AND CHEMICAL P} JPERIIES
Absorption in Aqueous Solution:
uitravioleç/visible spectra
l )’ i I inj 1cn era1ure
t2nsity/I e1ative L ?nsity
Dissociation Constants in Water
I- nry’s Law Constant
Meltir Temperature
Particle Size DistriWtiorVFiber
Length and Di neter Distritutions
Partition Coefficient (n—Octanol,4 ater)
Partition Coefficient (n-Octanol/Water) —
Est itnat ion by Liquid Chr r toj raphy
pH ot dater Solution or Suspension
Water Solubility
Water Solubility (Generator Column Method)
Vapor Pressure
TRANSR)R OCESSES
Soil Thin-Layer Chranatography
Sedin - nt and Soil Adsorption Isotherm
-------
TNSR)RMP TI( PROCESSF S
Riodegrac1 t1on, P roblc Pquatic Ct —2OO() C —2Ofl(1 1 S2
Riocieqradation, Ready Cc—2010 1982
Biodegradation, Anaerobic (fl-2050 C —2O5O 1982
Riodegradation in Soil CG—2075 1982
Biodegradation, Sewage Treatment Sijnulations (r —21OO 1982
Cmplex Formation Ability in Water CG—4000 1982
Hydrolysis as a Thinction of at 25°C rn—5000 CS—SOOn 1992
Photolysis in Aqueous Solution in Sunlight CG—6000 CS-6000 1’ 84
Laboratory fleteni ination of the Direct.
Photolysis Reaction Ckianturn Viold in
the Laboratory and sunliqht- Photolysi.s (Y;—f OiO Cs— (fl() 1984
(‘ 1 as Phase Absorption Spectra and Photolysis CG—7000 CS—700fl 1983
-------
CC,—l4 fl (Ocfobt r, l 4)
Contents
Page
1. INTRODUCTION . 1
A. Rackqround and Purpose 1
R. Definitions
C. Principle of the Test Method 4
1). Applicability and Specificity
II. TEST PROCEDURES.
A. Test Conditions
l. Special Laboratory Fquipment
2. Purity of Solvents and Reagents
. Preparation of Reagents arid Solutions 7
1. Solvents 7
2. Calibration Mixture 7
3. Test Solution 9
C. Performance of the Test 9
III. DATA AND REPORTIN 10
A. TestReport 10
B. Specific Analytical Procedures 10
TV. REFERENCES ii
-------
C( — 4 fl (C ctobt-’r, 1 4
DISCLAIMER
Certain commercial equipment, instruments, and materials are
identified in this paper in order to specify adequately the
experimental procedure. In no case does such identification
imply recommendation or endorsement by the Environmental.
Protection qency, nor does it imply that the material,
instruments, or equipment identified are necessarily the best
available for the purpose.
1-i
-------
(4 ( U ‘)CtO() ’ r ,
I. INTRODUCTION
A. ackground and Purpose
Since the pioneering work of Hansch and Fulita (l9 4) in
the measurement and estimation of the octanol/water partition
coefficient (K 0 ), this property has become the cornerstone of a
myriad of structure—activity relationships (SAR). Hansch and Leo
(1979) have used the coefficient extensively for correlating
structural changes in drugs with changes observed in biological,
biochemical. or toxic effects. These correlations are then used
to predict the effect of a new drug for which a could he
measured.
In the study of the environmental fate of organic chemicals,
the coefficient has become a key parameter. It has been shown to
he correlated to water soluhility, soil/sediment adsorption
coefficient, and hioconcentration. The importance of this
property to SAR is indicated by its discussion in the First
chapter of Lyman, Reehl and Rosenhlatt’s (1952) comprehensive
compendium of methods for estimating the behavior of organic
compounds in the environment. These authors consider the
measurement or estimation of the coefficient to he the necessary
first step in assessing the fate of new chemicals.
Of the three properties that can he estimated from
water soluhility is the most important because it affect:s both
the fate and transport of chemicals. ‘or example, highly so i hI
chemicals become quickly distributed by the hydrologic cycle,
1
-------
CG—1410 (October, 9H4)
have low adsorption coefficients for soils and sediments, and
tend to he more easily degraded by microorganisms. In addition,
chemical transformation processes such as hydrolysis and
oxidation tend to occur more readily if a compound is soluble.
Direct correlations between and both the soil/sediment
adsorption coefficient and the hioconcentration factor are to he
expected. In these cases compounds that are more soluble in
octanol (more hydrophobic) would he expected to partition out of
the water and onto the organic portion of soils/sediments and
into lipophilic tissue. The relationship between and the
hioconcentration Factor, as developed by Neely et al. (T974), and
other similar relationships, are the principal means of
estimating hioconcentration factors. These factors are then used
to predict the potential for a chemical to accumulate in living
tissue. ks a rough estimate, chemicals with less than 10
will not accumulate in tissue while those with greater than
l0 will. Thus, although a chemical may he present in the
aqueous environment at subtoxic concentrations, if its is
greater than l0 it would accumulate to levels that may he toxic
nbt only to the organism hut also to the consumers of that
organism.
This test guideline describes a rapid, inexpensive method
based on reverse phase—high pressure liquid chromatoqraphv (RP-
RPLC) for estimating the octanol/water partition coefficient as
developed by Veith et al. (1q79). It is not intended, however,
to replace the standard shake—flask method described in Test
2
-------
( —14ttJ uctoher, L 4)
Guideline CG—1400, and should he used keeping in mind the
limitations described herein. The RP-HPLC method is intended to
give quick estimates of K 0 t particu1 arly for v ry hyclrophohic
substances and mixtures that cannot he analyzed usin i the sh ke
flake method.
. Definitions
(1) The octanol/water partition coefficient (KOW), as
estimated by this test method, is the ratio of the equilibrium
molar concentrations of a chemical in n—octanol and water, in
dilute solution; as such it is a dimensionless quantity. K 0 is
a constant for a given chemical at a qiven temperature. P ecauso
can assume such a wide range f values, from less than one to
greater than a million, depending on the structure of the
compound, is often reported log Ko .
•(2) The retention time, tR, is the time in minutes elapsed
between sample injection into the chromatograph and the peak
maximum (concentration) as recorded on a chromatogram. The
retention time is characteristic of the substance, the liquid
phase flow rate, and the stationary phase, at a given
temperature. With proper flow and temperature control, it c.in be
reproduced to within one percent and used to identify cnultiph
peaks. Although several substances can have nearly identical
retention times, each substance has only one retention time.
This retention time is not influenced by the presence of other
components. Retention times for this method vary between several
3
-------
CG—141fl (October, 19R4
minutes for substances with a lower to thirty minutes or
greater for substances with higher K 0 ’s.
C. Principle of the Test Method
This test method is based on a reverse—phase high pressure
chromatographic (HPLC) separation procedure developed by Veith et
al. (1979). The test substance (solute) is injected onto an HPLC
column containing a support onto which a long—chain hydrocarbon
has been permanently bonded. A methanol/water solvent system is
used to elute the solute which is subsequently analyzed using an
ultraviolet absorption detector, gas chromatograph, liquid
scintillator or other suitable detector. Ouring elution, the
solute moves alone the column by partitioning between the mobile
phase and the stationary hydrocarbon phase. The retention time
on the column is a function of the hydrophohicity of the
solute: A water soluble solute has a short retention time while
a hydrophobic solute has a long retention time. Once the
retention time is measured on the chrontatogram, the of the
substance is estimated from a previosly established linear
regression equation between log tR and log K 0 . The relationship
between these two variables is determined through a calibration
step that involves injecting into the chromatograph a mixture of
six .reference chemicals having a range of retention times ; nd
known octanol/water partition coefficients. The rotj n i fl tir ’
for each chemical is measured and a plot of log t vs. oq W
made. The data are also correlated using a linear regression an’l
the resulting equation is used to calculate log from the lorj
4
-------
j n lu
tR of test suhstances the correlation cnefficif nt: of the linear
regression gives a measure of the “goodness of fit” of the
calibration data to a straight line.
1). pplicahility and S cificity
The test method described in this guideline is designed to
calculate an estimated value of the octanol/water partition
coefficient using an empirically derived equation that relates
the of a substance to its experimentally determined retentjon
time on a HPLC column. It must he emphasized that the shake
flask method (CG—1400) remains the conventional method for
determining K 0 . The HPLC method described herein is a rapid
procedure for estimating log K 0 for a single substance or a
mixture of substances. Estimates of log K 0 should he limited to
within two log units of the minimum and maximum values of the
calibration substances, i.e. , the method is applicable to
substances with log K 0 between zero and eight. Tn the range of
two to six lcg units, estimates are within 22.5 2fl.l percent of
the values reported in the literature obtained using other
methods (Veith et al. 1979).
5
-------
CG— 4 0 (Octc h ’r,
II. TKS PROCEDI1R S
A. Test Conditions
1. Special Laboratory Equipment
(1) A liquid chromatograph equipped with a 61)00 psi pump, a
high—pressure stopflow injector, and appropriate recorder.
(2) A preparative scale reverse phase column (250 mm X 8 mm
1.0.), e.g., Varian Preparative Micropak C—H, consisting of a
stainless steel tube filled with 10 micron LiChrosorh to which
octadecylsilane is permanently bonded.
(1) For chemicals that absorb in the ultraviolet (i.e.,
aromatics), either 254—nm fixed wavelength detector or 190 to 600
run variable wavelength detector, can be used. For chemicals that
cannot be detected in the ultraviolet, a fraction collector can
he used to collect fractions at suitable intervals (0.51) to 1.0—
minutes near the retention time) for analysis by gas
chromatography, liquid scintillation, or other suitable,
sensitive, analytical detector.
2. Purity of Solv tsj ’ r
All solvents (water, methanol, acetone, and cyclohexane) rind
reagents used in this test procedure should he reagent or HPLC
grade and contain no impurities that could interfere with the
determination of the retention time of the test compound. Wat er
meeting ASTM Type II standards or an equivalent grade is
6
-------
CG—141() (October, 1984)
recommended to minimize the affects of dissolvo salts an otih r
impurities. STM Type II water is described in STM Dll 3—77,
“Standard Specification for Reagent Water”.
B. Preparation of Reagents and Solutions
1. Solvents
For column elution and preparation of buffers, mix
chroniatographic or reagent grade methanol and water in an 85:15
v/v ratio.
2. CalIbration Mixture
Prepare a standard calibration solution containing 200 mq/L
of each of the substances listed in Table 1 dissolved in acetone
and cyclohexane (3:1 v/v) other suitable solvent. Twenty
microliters of this solution injected into the chromatograph
should give an adequate recorder response (25% of scale) for
calibration purposes. However, both the concentration and amount
injected may he increased or decreased without afFecting the
retention times, since i_s independent of concentration in
dilute solutions.
7
-------
C(—14 1( (C rtb ’r, 1q. 4
T bl’ 1 • Mn Mured octanni /witer part i t- ion co IF i - I •ind
typical HPLC retention times for the chemicals used in
the calibration mixture.
Literature Measured Typical Retention Tim€
Chemical log K minutes log minutes
Renzene 2.13 4.12 (L 1
Rromohenzene 2.99 7.09 O.RS
Riphenyl 3.76 8.R5 0.95
Ribenzyl 4.81 15.87 1.20
p,p’—DDE 5.69 21.98 1.34
2,4,5,2’, S’—pCR 6.11 11.58
8
-------
1,—1’4L%, LLO fl- L, L C)*i
3. Test Solution
Solutions of the test substance(s) are prep red s i ni 1 ar to
the calibration mixture: by dissolving the substance(s) to he
tested in a 3:1 mixture of acetone and cyclohexane. The
concentration of the substance(s), as determined by trial and
error, should be sufficient to produce a chromatographic peak of
at least 25 percent of the recorder scale.
C. Performance of the Test
After Condi t oninq the column with 1’ m th nol—water or
buffered methanol—water, chromatoqraph the calibration mixture by
injecting 20 microliters of the mixture into the column. F1ute
the column using a solvent flow rate of about 2.0 mi/mm at a
pressure of approximately 1200 psi. Determine the retention time
for each substance in the mixture. The calibration mixture must
be chromatographed daily because the retention time is sensitive
to variations in the flow rate, temperature, solvent ratio, and
the retention properties of the column.
Chromatograph 20 microliters of the test solution(s)
directly following column calibration, using an identical flow
rate and pump pressure. Determine the retention time(s) for the
test substance(s).
9
-------
CG—141fl (October, 1984)
III. DATA AND RFPORTING
A. T st prt
using the measured retention times of the substances in the
calibration mixture and the log K for each substance (Table 1)
make a plot of log tR vs . log K 0 . From the data used in making
this plot compute a linear regression equation of the form:
log K 0 , = m log tR + h
i.e., y = mx + h. For each set of t-est conditions (flow rate,
pressure) report this equation, its correlation coefficient nd
the data used in its calculation.
Calculate an estimated log T for each test substance from
its retention time and corresponding regression equation. Report
the retention time, its logarithm and log along with the
above data.
B. Specific Anal ytical Procedures
(1) Provide a detailed description c i, or rt iorc’nco for,
the liquid chromatograph, separation column, and det ctor.
(2) Report the temperature at which the test(s) were
conducted.
(3) C,ive a description of any problems (and their
rectification) or changes in the test procedures.
10
-------
I ‘+ I U L LQ1)t L , I
IV. RH RENCFS
Hansch C, Fujita T. 1964. — — analysis. A method for the
correlation of biological activity and chemical structure. J Am
Chem Soc 86:1616.
Hansch C, Leo A. 1979. Substituent constants for correlation
analysis in chemistry and biology. 3. Wiley & Sons. New York.
Lyman WJ, Reehi WF, Rosenblatt DH. 1982. Handbook of Chemical
Property Estimation Methods: Environmental Behavior of Organic
Compounds. McGraw-Hill Book Company. New York.
Neely WR, Branson DR, Blau GE. 1974. Partition coefficient to
measure hioconcentration potential of organic chemicals in
fish. Environ Sd Technol 8:113.
Veith GD, Austin NM, Morris RT. l 79. A rapid ffli tih)C1 For
estimating log P for organic chemicals. Water Res 11:41.
ii
.1-i
-------
Cs— 4 10
Oct ober, 1984
PARTITION COEL ICIENT (n-OCTANOL/WATER)
ESTIMATION BY LIQUID CUROMATOGRAPt-fY
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AUEt 4CY
WASHINGTON, D.C. 20460
12
-------
L ’,- L’+.1tJ )e o) - r, L’fl)’*)
Tah e of
Piqo
I. NEED FOR THE TEST . 1
II. SCIENTIFIC ASPECTS 3
A. Rationale For the Test Method 3
B. Other Methods For Determining K 0 4
1. The Conventional Shake—Flask
Method 4
2. Estimation from Water Soluhility 5
3. Estimation Using the Fragment
Constant Method. 7
4. Estimation Using Thin—Layer
Chromatography
C. Rationale For the Test Conditions 9
1. Chromatoqraphic Column 9
2. Analytical Detector 9
3. Purity of Solvents and Reagents 9
4. Calibration Mixture TO
D. Test Data and Reporting 10
1. Test Report in
2. AnalyticaiProcedures. 11
III. REFERENCES 12
I
-------
CS—1410 (OctoLer, I98 .
I. NEED FOR THE TEST
The octanol/water partition coefticierit is one ot the
most frequently measured and most widely used parameters in
assessirv the environmental fate of organic chemicals. Its
importance is due primarily to the tact that is used to
predict the general lipophilicity or hydrophobicity ot a
chemical. Of greater significance on the quantitative level,
however, is the fact that has been tound to correlate with
the water solubility, the soil/sediment adsorption coetticient
and the bioconcentration factor.
Although all of the above mentioned properties are important
in predicting the fate and residence time ot a chemical in the
environment, the bioconcentration tactor (BCF) carries with it
the greatest consequences. This is because the BC is used tt
predict the potential tor a chemical to accumulate in living
tissue. When applied to the aquatic environment, the BC is
indicative of the potential for a chemical to accumulate or
bioconcentrate in the tissue ot fish and other lower aquatic
organisms. In terms of the K 0 , chemicals with a value less thar
10 will not bioconcentrate while those with a value greater trLan
will. Thus, although a chemical may be present in a stream
or lake at subtoxic concentrations, if its is greater than
iO it could accumulate to levels that may he toxic to not only
the organism itselt, but also to the consumers ui that tjaiii ;iii,
not the least important of which are Home sapiens.
14
1
-------
CS—1410 (October, l 4)
The direct determination of K 0 , using the con rentional
shake—flask method (CS— 1400) is suh ect to numerous experimental
difficulties. These include the formation of coUnidal.
dispersions (emulsions) during the shaking step, incomplete
separation of the octanol and water phases, difficulty in
analyzing very hydrophobic chemicals in the water phase, the
adsorption of the solute onto the surfaces of transfer vessels,
and the loss of volatile solute into the atmosphere. In addition
the method is time consuming and the chemicals used in the test
procedure, including the test substance, must he of high
purity. nme f the principal. advantages oF the HPL(’ method are:
1 ) the test substance need flot he pure and , i n Fact. , may
actually contain any number of impurities or other test
substances: 2) the method will give reliable estimates of log
for very hydrophobic chemicals, i.e, log K,. >5; and 3) the
method is easier to carry out and takes less time to perform than
the shake flask method.
In summary, the importance of in assessing the fate of
substances released into the environment and the con’ien ent.
trouble—free rapid procedures associated wit.h the IIPI,C method
make it ideally suited for use as a test method for n’ iitra1
organic substances, and especially for very hydrophobic
substances or multicomponent mixtures.
I
-L
2
-------
CS—1410 (October, .9
I I. SCIENTI ’IC_ASPECTS
A. Rationale tor the Test Method
The technique employed in this test method is reterred to as
reverse phase high pressure liquid chromatography (RP—l-IPLC). It
involves the use of a chrcxnatographic column that consists ot an
inert column support material onto which a non—polar C 18
hydrocarbon is permanently bonded. Chemicals are injected onto
this column and eluted with a polar solvent (in this case
methanol—water). As the chemicals move along the column, they
partition between the mobile phase (methanol—water) and the
stationary hydrocarbon phase. Those chemicals that arc more
soluble in the hydrocarbon phase, that is more hydrophobic, wil I
exhibit longer retention times (tR) or be eluted more slowly than
those that are more soluble in the polar methanol—water phase.
Thus, the retention time ot a chemical on the column is directly
proportional to its hydrocarbon/water partition coefficient.
This retention time can be correlated with the octanol/water
partition coefficient when a calibration curve is constructed
from the retention times of several reterence chemicals br wh’eh
has been established by using the conventional -;hake--i la ;k
method. The equation of the calibration curve ha ’; ttu i Olowiii’j
fo rm:
log K 0 = miog tR + b
3
IG
-------
C —14 [ U Octobec, L - )
McCall (1975) and Carlson et al. (1975) were the tirst to
recognize the value ot RP—UPLC for estimating K 0 . The technique
was later verified by Mirriess et al. (1978), Unger et a!.
(1978), and Veith et al. (1979) whose work serves is Lhe basis
for the method described in this juiduline.
B. Other Methods for_Deterraini y
1. The Conventional Shake lask Method
The conventional method for determining the octanol/water
coefficient involves shaking a solute with two immiscible
solvents and then measurirKj the solute concentration in the two
solvents after equilibration (Leo, et al. 1971). The
d istribuL ion of the SO lute between the two 50) IvetiLs is a ii rect
consequence at the thermodynamic requirements br eoju i 1 ibriuiri
that apply only to dilute solutions. The resulting ratio ot.the
two solute concentrations is the partition coetticient which is
constant at a given temperature. Numerous researchers have used
this method and published their results (Fujita et al.. 1964;
Hansch and Anderson 1967; Leo et al. 1971; Chiou et al. 1977).
Both EPA (1982) and OECD (1981) specify the use of this method in
their test protocols (see CG—1400) . And, although there is no
ASTM Standard Test Method for K 0 , the shake—i lask tuethoot i:;
“standard method” used in industry when i; to) I rwi oo1.
4
-------
CS—i • [ U (OCt t’F, L
As noted earlier, the shake task method is not wiLkiout
problems, the most important of which are the tormation ot
emulsions, surface adsorption of the solute, and the time
consuming nature of the test procedure. Flowever, with suitable
care and patience these problems are minimized anu basic
experimental data is obtained that can be correlated with other
important environmental fate parameters.
2. Estimation trom Water So tu
By definition, the partition coetticient expresses the
equilibrium ratio of an organic chemical partitioned between an
organic liquid (i.e., n—octanol) and water. This partitioning
is, in essence, equivalent to partitioning of the organic
chemical between itself and water. Thus, one would expect that a
correlation might exist between the partition coetticient (Kow)
and water solubility(s). Indeed, as shown by 4ackay ( 1 q77), th
of these properties are a function ot the aqueous phase act iv . iy
coefficient of the compound, and the correlation between the
and S is based partly on the relationship between this activity
coefficient and its reciprocal.
Chiou et al. (1977) studied the relationship between
the water solubility, 5, and found that, tor 34 orJanic
chemicals, an excellent linear correlation was observed between
log and log S that extended to more than eight orders ot
magnitude in water solubility (1O to IO ppm), arid six orders
L. P
-------
(28—1410 (October, 19d )
ot ma n i t ude iii It) to I 0 ) . V i ui t he r t t he I I
tog ross ion equal ion wa ; do r ived
log K 0 = 5.00 — 0.670 I j 8, r 2 = 0.970
where S is expressed imol/L, and r 2 is the coefficient of
determination.
To date a total of 18 different regression equations have
been derived that correlate water soLubility with the octanoi.
water partition coetticient. This large number ot equations
results troin the tact that solubility varies with 1t10 ITUflCt1OILII
group of the molecule. Lyman et al. (1982) have summarized these
equations along with the class and number ot chemicaLs that appLy
to each equation, the respective correlation coetticient, the
applicable range of S and values and the temperature at which
the solubility data were obtained. Recent developments in the
correlation between S and indicate that many ot the above
equations can be c bined into a single general relationship
(Lyman, 1984). This would greatLy simplity the esLim ition
procedure, but may introduce (J ruater error, rt icu I y I or
certain classes ot compounds.
This method has a definite advarit:aye in Lh it (:.in I’.
estimated from an experimentally aetermined par imeLer — w iL er
solubility — whose test procedure involves less experimental
6
-------
C S— L 4 Li) ( uc t. r ,
difficulties. Although this approach also has a cost advanta je
in that two parameters are determined tor the cost ot one, it
must be kept in mind that the is an estimated value subject
to the limitations usually associated with approximations. Thus,
while the estimation of Kow from water solubility is definitely a
valid method, the values obtained cannot be construed as beinj
equivalent to those obtained from the conventional shake—tlask
method.
•
Hansch and Leo (1979) have developed a method to estimate
from empirically derived atomic or yroup trajinent constants
(f) and structural factors (F). Iisiny these, the by K 0 is
calculated using the following equation:
by = Sum of fragments (t) + structuraL tactors (&)
Of course, the critical piece ot intormation requirt?d to d )pLy
this method is the structure of the chemical in 4uestion——whicll
is not always known. However, it the structure is known then
there are a total of over 200 fragments (t and F) available L at
take into account such structural factors as molecular
flexibility, unsaturation, multiple habogeriation, branching, and
H—polar fragment interactions. These and the inetriou ot
calculation, with examples, are reviewed extensively iy Lyman t(.
7
i d
-------
C — 1 410 (ct r,
al.. (1982). in addition, a large c I Lcct i on ( ibout 15,000) ol.
both measured and calculated values has been compilea by
Hansch and Leo (1979); the method is also available tor use on
computer (Chou and Jurs 1979).
Estimation by the fragment constant is probably the best
initial step in determining K 0 since it can be done without
experimentation and is quite reliable for a large number ot
C ( .MflfflOfl a r’jan ic cheini ca Is • However, I )r some I UliCt i ()IIa g
.jril (:()mI)Lex O h i jh I y SU1)St. i ( .Ut_e(j 1110 lecu Ie ; tile I lilleuL const ant
method (Jives erroneous, misleading results.
4. Estimation Using Thin-Laj r Chromtaj y
It has been reported that thin—layer chromatography can be
used to estimate K 0 (Mirriess et al. 19Th; i-lulst’totf and Perrin
1976). However, high-pressure liquid chromatography (HPLC) ič
far superior to thin—layer chromatography (TLC) because ot its
accuracy (i.e., definition of the peak, rcproducibility, ease at.
detect ion in many cases, and above al I Liie range ot aL.P I. caa I Ly
(HPLC is applicable over 5 orders at inag n i Lude ol j e 1 LC
is only applicable over 1.5 orders at magnitude ii k 0 ) (Mjr
et al. 1976).
8
-------
CS—l4ltJ (octobar, i9 -,
C. Rationale for the Test Conditions
1. cho col i
ye i th and Morris ( 1.9 7H ) desc r t1) ’ t h ’ use )t t’ i t hei . n
analytical or preparative scale reverse phase column containinj a
solid support to which octadecylsilane has been permanently
bonded. Both of these types columns are commercially
available. The preparative scale column has been recommended tor
use with this test guideline because the authors tound that this
column produced linear calibration curves with shorter retention
times using lower pressure than the analytical column.
2. Analytical Detector
Any suitable detector compatible with the HPLC and capable
of detecting the test substance(s) can he used. This includes a
u ” detector which is sensitive to the detection ot aromatic
compounds; gas chromatography or liquid scintillation may be usea
for chemicals that do not absorb in the uv.
3. Purity of Solvents and Reagents
To prevent any interference with the determination ot the
retention time of the test substances, all solvent ; and r( aJ( nI
must be chr natograph ic or reagent (3 rade • lb i i ti: lwJ ; he
of reagent grade water as described in ASiM UI 1tfl 7/: SI
Specification for Reagent Water.”
9
-------
4 L U ( ‘ t c H’ r , I
4. Calibration Mixture
This test method requires that a calibration or standaru
curve be developed that relates the log K 0 to log tR tar a group
of reference neutral organic chemicals. The for these
chemicals must be initially determined using the standard shake
flask method (CG-1400). The values should he well
established (i.e., have small standard deviation) and cover the
r ar je t ram about 2 to 6 1 ( j un its . Ih is range c in be ext ended
and/or 1. united with the use at other standard substances
[ Note: the calibration curve should have a high coet t icient at
determination, e.g., not less than r 2 =U.9751 . Ihe standaru
substances recommended for use in this test guideline meet these
criteria. The substances are dissolved in a suitable solvent aria
chromatographed prior to the test substances(s). because of
variations in the column, temperature, tlow rate and solvent
changes, the calibration curve must be determi tied ( Ia i ty
I). J L?Y ’ ) J.
1. Test ort
The data required in the test report are necessary to
estimate K 0 , and evaluate the accuracy and precision of each
step in the test procedure. These data include the tollowirig:
the retention time for each of the substances in the cat ibrat IO U
mixture, the and its logarithm, includirij the standard
_ ;,‘
Pd’
-------
CS—1410 (October, I* -.
(It’V i •i t. I r , ot t’ t(It I I lit’ ‘ t I II u t I tIt) I t IR’ I ; U i L
t he sta art.I shake t La ;k inettioti, t.Iie it it’ t.on e LhIt ion
coet t ic ient ot de t.e rm i nat. ion or cor r e tat. on coc I c t. e ut. ) lid
relates the logarithm ot tR and K 0 , Lhe LR anti lotj tR vaLues
along with the calculated log K 0 tor each ot the test
substances. The following intormation is also re uireQ regarulny
the chromatograph: tiow rate pressure and pH ot eluent and tbe
temperature at which the test was conducted.
2. Anailti cal Procedures
In order to assess the accuracy arid j is ton of t tie tilt i C
employed, its detector an(l column, a tletai ted tiescri t ion ot , or
reterence for these items must be provided. it any chdrtJos are
made or problems encountered in the test method, tliese stiould be
reported so that any necessary revisions can be incorporateci in
future editions.
24
11
-------
4 1 4 ( )c )n(’ 44
[ 11. REFERENCES
Carison RM, Carison RE, Kopperman EL. 1975. Determination ot
partition coefficients by liquid chromatography. J Chromato r
107:219.
Chou JT, Jurs PC. 1979. Computer assisted computation ot
partition coefficients trom molecular structure using tragment
constants. J Chem mt Comput Sci 19:172.
Chiou CT, Freed VU, Schmeddintj DW, Kohnert EL. 1977. PartitiOn
coetticient and bioaccuinuiation ot selected orj rnc chemica ls.
Environ Sd Technol 11:475.
EPA. 1982. Partition coet ticient ( n—octanol/wa ter) CL— !40U.
Test Guideline. USEPA. Ut t ice ot lox ic SubsLdnces. asU I ri ton
Fujita T, Iwasa J, Hansch C. 1964. A new substituerit constant,
derived from partition coefficients. J Am Chum Soc 86:5175.
Hansch C, Anderson SM. 1967. The ettect of intramolecular
hydrophobic bonding on partition coetticients. J Org Chem
23:2583.
Hansch C, Leo A. 1979. Substituent constants for correlation
analysis in chemistry and biology. New York: J. d1ley & Sons.
Uulshoft A, Perrin JH. 1976. A comparison or the determination
of partition coefficients ot 1,4—benzodiazepines by high—
performance liquid chromatography and thin—layer
chromatography. J Chromatogr l29:26 .
Leo A, Hansch C, Elkins U. 1971. Partition cecIl idlents arii
their uses. Chem Rev 71:525.
Lyman WJ. 1984. Personal communication.
Lyman WJ, Reehl WF, Rosenblatt DH. 1982. HandbooK ot Chemicu!
Property Estimation Methods. Environmental uctiavior ut Urgunic
Compounds. McGraw Hill Book Company. New York.
Mackay D. 1977. Environ Sci Technol 11:1219.
McCall JM. 1975. Liquid—liquid partition coetticients by hijh—
pressure liquid chromatography. J Med Chem 18:549.
Mirrless MS , Moulton SJ, Murphy Cl, Taylor P1. 1976. )lrect
measurement of octanol—water partition cecI l icieriLs by H i 0 h—
pressure liquid chromatography. J Med Chum 19:6L5.
12
-------
CS—I 4 I 0 ( Oct.ch r , I I 4
OECD. 19R 1. Organization for Economic Cooperation and
Development (OECD). Guidelines for Testinq Chemicals No. 1fl -
Partition Coefficient (n—flctanol/Water’ . flirec tor f
Information, OF.CD 2 Rue Andre—Pascal , 7 77S P R S CI 1) IX 16.
F rance.
Twitchett P1, Moffat AC. l 75. Hiqh pressure liquid
chromatography of drugs: An evaluation 01 an octadecylsi1an
stationary phase. J Chromatogr 111:149.
Unger SH, Cook JR, Rollenherg 1S. 197R. Simple procedures for
determining octanol—aqueous partition, distribution, and
ionization coefficients by reverse—phase high pressure liquid
chromatography. 3 Pharm Sd 7 l3 4
Veith GD, Austin NM, Morris RT. 1979. A rapid method for
estimating log P for organic chemicals. Water Res 13:43.
26
13
-------
CC—6 000
October, 1 R4
PHOTOLYSIS IN AOUEOUS SOLUTION IN SUMLIGHT
0f’1’ ICE OF TOX I C stIusrANch:s
OFFICE OF PESNCIDES AND TOXIC suHsrANCE
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, DC 20460
,.s
-------
CG—( ()U ) (( c t he r 1
1)1 C1 A 1 M I R
Certain commercial equipment, instruments, and mater aU. art’
Identi. fled in this Test Gutclol .fle I U )rder to Zt(1( ’(pI tel y spec i fy
the experimental procedure . In no case does such i. dent i fi cit t i on
imply recommendation or endorsement by the Environmental
Protection Agency, nor does it imply that the matertal,
instruments, or equipment identified are necessarily the best
available for the purpose.
4 )
I -,
—1—
-------
— h ) U (1 ( Oc t be r , 1 9 4 1
Contents
Page
I. INTRODUCTION • 1
A. Background and Purpose . 1
B. Definitions and Units 2
C. Principle of the Test Method 4
D. Applicability and Specificity 9
II. TEST PROCEDURES 12
A. Tier 1 Test: Ely—Visible Absorption Spectra —
Estimation of Aqueous Photolysis
Maximum Rate Constant and Minimum
Half—life in Sunlight 12
B. Tier 2 Test: Aqueous Photolysis in Sunlight . 14
1. Test Conditions 14
a. Special Laboratory Equipment 14
b . Pu r i t y of Wa t e r 1 4
C * Sterilization 15
d . pH Effects 15
e. Volatile Chemical Substances i S
f. Control Solution 16
g. Absorption Spectrum as a Criterion for
Performing the Aqueous Photolysis rest... 7
h. Sunlight Actinoinetor 17
i. Solar Irradiance Data 19
3. Geometry of the Reaction Vessel 19
k. Chemical Analysis of Solutions 2(1
2. Preparations 21
a. Preparation of Test Chemical Solution 21
b. Preparation of Buffer Solutions 22
c. Preparation of Actinometer Solution 22
3. Performance of the Tests 23
a. Phase 1 Experiments... 23
i. Procedure 1 24
ii. Procedure 2 25
iii. Procedure 3
iv. Analytical Methodology
b. Phase 2 Experiments
-ii- 1)
I
-------
CG—6000 (October, 1984)
Paqe
[ [ I. DATAANDREPORTTN(3 • 27
A. Tier 1 Test: tJV—Vis ihie Ahsorp(. LOfl Spe’t r i —
Estimation of Aqueous Phttolysis
Maximum Rate Constant and Minimum
Half—life inSunlight . 27
1. Treatment of Results... .. 27
2 • Test Data Report . . • . . . • . . • . * . . . 28
B. Tier 2 Test: Aqueous Photolysis in Sunlight 30
1. PhaselExperiments....... 30
a. Treatment of Results 30
h. Specific Analytical and Recovery
Procedures 32
c. Other Test Conditions
d. Test DataReport
2. Phase 2Experiments 34
a. Treatment of Results 34
b. Other Test Conditions..... 37
c • Test Data Report . 3.
IV. REFERENCES...... •1 •..•••••• •.......... ....•. ... 40
V • APPEt4DICIES . . . . . . . . . . . . • . . .. . . • . . . . . . 41
A. AppendixA. Tables 1—6. .... . 41
B. Appendix B. Illustrative Examples...... . . 52
1. Tier 1 Test: UV—Visihle Absorption
Spectra — Estimation of queous
Photolysis Maximum Rate
Constant and Minimum Half—life
in Sunlight 52
a. Illustrative Example 1.
h. Illustrative Example 2
2. Tier 2, Phase 1: Aqueous Photoly.sis
in Sunlight..... r) 7
a. Illustrative Example 3 57
3. Tier 2, Phase 2: Aqueous Photolysis
in Sunlight... .•.... • • 60
a. IllustrativeExample4
—iii—
-------
CG—6000 (October, 1984)
PHOTOLYSIS IN AQUEOUS SOLUTION IN SUNLIGHT
I. INTRODUCTION
A. Background and Purpose
Numerous chemicals enter natural aquatic systems from a
variety of sources. For example, chemical wastes are discharged
directly into natural water bodies, and chemicals leach into
natural water bodies from landfills. Pesticides are applied
directly into water bodies, and are applied to soils and
vegetation and subsequently leach into water bodies. Pollutants
present in aqueous media can undergo photochemical transformation
in the environment (i.e., in sunlight by direct photolysis or by
sensitized photolysis). As a result, there is considerable
interest in photolysis in solution, especially the photolysis of
pesticides. However, most of these studies have been qualita ivo
in nature and involved the identification of photolysis
products. Quantitative data in the form of rate constants and
half—lives are needed to determine the importance of
photochemical transformation of pollutants in aqueous media.
This Test Guideline describes a two—tiered screening level
approach for determining direct photolysis rate constants and
half—lives of chemicals in water in sunlight.
) I
‘I—’—
—1—
-------
cG—6000 (october, 984)
13. Definitions and Units
(1) “Radiant energy”, or radiation, is defined as the
energy traveling as a wave unaccompanied by transfer
of matter. Examples include x—rays, visible light,
ultraviolet light, radio waves, etc.
(2) “Absorbance (Ax)” is defined as the logarithm to the
base 10 of the ratio of the initial intensity (Is) of
a beam of radiant energy to the intensity (I) of the
same beam after passage through a sample at a fixed
wavelength A. Thus, A, = log 10 (TO/1).
(3) The “Beer—Lambert law” states that the absorhance of a
solution of a given chemical species, at a fixed
wavelength, is proportional to the thickness of the
solution (&), or the light pathlength, and the
concentration of the absorbing species (C)
(4) “Molar absorptivity (e’)” is defined as the
proportionality constant in the Beer—Lambert law when
the concentration is given in terms of moles per liter
(i.e., molar concentration). Thus, = cAC ., where
and c represent the absorhance and molar
absorptivity at wavelength A and t and C are
defined in (3). The units of are molar cm .
Numerical values of molar absorptivIty depend upon the
nature of the absorbing species.
—2—
-------
CG—6000 (October, 1984)
(5) A “first-order reaction” is defined as a reaction in
which the rate of disappearance of a chemical is
directly proportional to the concentration of th
chem icat and is not a funct IOn ü1 tue coricenLrat ion ot
any other chemical present in the reaction mixture.
(6) The ‘ half—life (tl/)” of a chemical is defined as the
time required for the concentration of the chemical
being tested to be reduced to one—half its initial
value.
(7) The “sunlight direct aqueous photolysis rate constant
(k )“ is the first—order rate constant in the units
pE
of day and is a measure of the rate of disappearance
of a chemical dissolved in a water body in sunlight.
(8) The “solar irradiance in water (LA)” is related to
the sunlight intensity in water and is proportional to
the average light flux (in the units of i( 3 einsteins
cm 2 day ) that is available to cause photoreaction
in a wavelength interval centered at A over a 24—
hour day at a specific latitude and season date.
(9) “The Grotthus—Draper law”, the first law of
photochemistry, states that only light which is
absorbed can be effective in producing a chemical
transformation.
(10) The “Stark—Einstein law”, the second 1 w ol
photochemistry, states that only one molecule is
‘It-)
—3—
-------
CG—6000 (October, 1q 4
activated to an excited state per photon or quantum of
light absorbed.
(11) The “reaction quantum yield (45j” for an excited
state process is defined as the traction of absorbed
light that results in photoreaction at. a fixed
wavelength A. It is the ratio of the number ol
molecules that photoreact to the number of quanta of
light absorbed or the ratio of the number of moles
that photoreact to the number of einsteins of light
absorbed at a fixed wavelength A.
(12) “Direct photolysis” is defined as the direct
absorption of light by a chemical followed by a
reaction which transforms the parent chemical into one
or more products.
As a convenient reference, a glossary of the important
symbols used in this Test Guideline is given in Section IV of the
Support Document [ CS—6000].
C. Principle of the Test Method
This Test Guideline is based on the principles developed by
Zepp and Cline (1977), Zepp (1978), the U.S. Environmental
Protection Agency, Office of Toxic Substances LUSEPA (1979)1,
Mill et al. (1981, l982a, 1982h), and Dul.in and MiIJ (I9 2).
Zepp and Cline (1977) published a paper on the rates of
direct photolysis in aquatic environments. The rates of aLl
4
-------
CG—6000 (October, 1984)
photochemical processes in a water body are affected by solar
spectral irradiance at the water surface, radiative transfer from
air to water, and the transmission of sLlnliqht in the water
body. It has been shown that for the phototysis of a chemical in
an optically thin aqueous solution, the kinetics of direct
photolysis can be described by the following equations:
ln(C 0 /Ct) = kpEt (1)
tl/ = O. 693 /kE (2)
kE = 4)Ek (3)
where is the reaction quantum yield of the chemical. in
dilute solution and is independent of the wavelength,
ka = the sum of kaA values of all wavelengths of
sunlight that are absorbed by the chemical (i.e., the light
absorption rate constant), t is the time, C 0 and C are the
concentrations of chemical at t = 0 and t, and tl/ represents
the half—life. The term kpE represents the first—order
photolysis rate constant for a water body in sunlight in the
units of reciprocal time.
Furthermore, under the same conditions CH0d dhove, t:ho
first—order direct photolysis rate constant, is j ivon hy
equation
—5—
-------
CG—6000 (October, 1984)
kpF = E CA1 A
where is the reaction quantum yield, c is the molar
absorptivity in the units molar cm’, L is the solar
irradiance in water in the units of iO einsteins cm 2 day’
[ Mill et al. (1982a)]., and the summation is taken over the range
A = 290 to 800 nm. L is the solar irradiance at shallow
A
depths for a water body under clear sky conditions and is a
function of latitude and season of the year.
The method of Zepp and dine (1977) and the method n MiU
et al. (19A2a) are applicable to sunlight incident on a water
surface such as natural water body. However, the method
developed in this guideline measures rate constants in tubes
(e.g., 13 x 100 mm) and the rate is faster in tubes. This is
discussed in more detail in Section II.B.J. Thus, equations 1
and 2 have to be modified to take this into account. For
simplicity, the following nomenclature is used. For water
bodies, the rate constant is designated as kpE with the subscript
E designating rates in the environment in water bodies. For
tubes, the rate constant is designated as k [ ). The correspond i nq
half—lives for water bodies and tubes are and
respectively. Thus, for tubes, equations 1 and 2 can be writterm
as
ln(d/C ) = kt (5)
0.693
ty 2 k
—6—
-------
CG—6000 (October, 1984)
A simple first—tier screening test has been developed using
equation 4. As an approximation, it is assumed that the reaction
quantum yield E is equal to one, the maximum value. As a
result, the upper limit for the direct photolysis sunlight rate
constant in aqueous solution is obtained and equation 4 becomes
(k ) = ) E: L (7)
pEmax. AX
Using equation 7 in equation 2, the lower limit for the half—life
is then given by
— 0.693
(tl/ f.) — (k ) (8)
rriin. pE max.
The molar absorptivity can be determined experimentally by the
method outlined in Section II.A and the values of LA are given
in Tables 3 to 6 as a function of latitude and season of the
year. These data can then he used in equation 7 to calculate
pE max . Finally, (kpE)max. can then he substituted in
equation 8 to calculate / min.
In a second—tier test method, an aqueous photolysis
screening test has been developed to determine rate constants and
half—lives in the presence of sunlight using equations 1, 2, 4,
5, and 6 EUSEPA (1979), Mill et al. (1981, 1982a, 1982h), and
Dulin and Mill (1982)). The second—tier test method is divided
into two phases. In phase one, the test chemical is photolyzc’d
n
—7—
-------
CG—6000 (October, 1984)
in suni ight In order to obtain an approx im Le rate eonst.ant
This method only gives an approximate rate constant since
it fails to measure sunlight intensities incident on the sample
during photolysis.
In phase two, a standard p—nitroacetophenone—pyridine
actiriometer (pNAP/PYR) is used to measure sunlight intensities
incident on the sample during photolysis [ Mill et al. (1982b) and
Dulin and Mill (1982)1. The rate constant for this actinorneter,
can he adjusted to match the approximate rate constant of
the test chemical by adjusting the concentration of pyridine.
Since the rate constant is a function of the reaction quantum
yield of the actinometer, the rate constant can he adjusted
according to the equation
= O.0169 [ PYR] (9)
where [ PYR] is the molar concentration of pyridine for a
p—nitroacetophenone (PNAP) concentration of 1.00 x M. The
reaction quantum yield for the test chemical, , is qiven by
= k c LA ( In)
k cALA
The reaction quantum yield of the test chemical, can be
determined in the following way. By measuring the concentration
—8—
-------
CG—6000 (October, 1984)
of test chemical and actinometer (PNAP) as a function ot time t.
in sunlight, the ratio of rate constants, (kd/ka), can be
determined using equation S (Section II.B.l.h). can he
determined from equation 9 at the molar concentration of pyridine
used in the standard actinometer. The term for the
actinometer has been tabulated as a function of latitude and
season of the year (Table 21. The term LA for the test
chemical can be obtained from the experimentally measured molar
absorptivities (Section II.A) and the values of listed in
Tables 3 to 6, as a function of latitude and season of the year.
With the values of , , and the appropriate J values,
kpE for the test chemical can be calculated as a function of
latitude and season of the year in the United States using
equation 4. The corresponding half—life can be calculated using
kpE in equation 2.
D. Applicability and Specificity
This Test Guideline is applicable to all chemicals which
have uv—visihle absorption maxima in the range 290-800 nm. Some
chemicals have absorption maxima significantly below 29() nm
consequently cannot undergo direct photolysis in sunlight (e.g.,
chemicals such as alkanes, alkenes, alkynes, saturated alcohols,
and saturated acids). This is a direct consequence of the
Grotthus—Draper law of photochemistry. Some chemicals have
absorption maxima significantly below 290 nm hut have measurable
—9— 1)
-------
CG—6000 (October, 1984)
absorption tails above the baseline in their absorption spectrum
at wavelengths greater than 290 run. Photolysis experiments
should be carried out for these chemicals.
These test methods are only applicable to pure chemicals and
not to the technical grade.
The first—tier screening test can be employed to estimate
(kpE)max and (t1/ )min. . If these data indicate that aqueous
photolysis is an important process relative to other
transformation processes (e.g., biodegradation, hydrolysis,
oxidation, etc.), then it is recommended that the second—tier
photolysis tests be carried out to determine environmentally
relevant rate constants and half—lives in sunlight. The data
obtained from this test can be used to determine kpF for the
test chemical as a function of latitude and season of the year
anywhere in the United States. These rate constants are in a
form suitable for preliminary mathematical modeling for
environmental fate of a test chemical.
The second—tier screening test is applicable to the direct
photolysis of chemicals in a homogeneous dilute solution with
absorbance less than 0.05 in the reaction cell at all wavelengths
greater than 290 nm and at shallow depths (less than 0.5 m).
These results are applicable to direct sunlight phot.olysis for
water bodies and clear sky conditions. In addition, these
experiments are limited to the direct photolysis of chemicals in
air—saturated pure water.
40
—10—
-------
CG—6000 (October, 1984)
Thi screening test has been designed to determine the molar
absorptivity of a test cheJcal, , and its reaction quantum
yield , . These parameters can bt’ Li 5 i tO iet e nii I
t flV tronmenta 1 y retevant aLu coii ;t ant ; at. low it;orhanc ’ ,wd
shallow depths in pure water as a function of latitude and season
of the year. Tables of solar lrradl3nce (Table 3—6) have been
included in this Test Guideline to carry out all the
calculations. However, the method is really very general and can
be extended to determine the rates of photolysis over a range of
other environmental conditions using a computer program. Zepp
and Cline (1977) have written a computer program to calculate the
rates of photolysis as a function of depth in water, as a
function of the attenuation coefficient of the water ( ) for
natural water bodies, the average ozone layer thickness that
pertains to the seasons and locatjon of interest, and as a
function of latitude and season of the year. This program has
been recently updated with the best available solar irradiance
data and is called the GC SOLAR program. The GC SOLAR computer
program is available on request. ER. Zepp, Environmental
Research Laboratory, U.S. Environmental Protection Agency,
College Station Road, Athens, Georgia 30601.1
41
—11—
-------
CG—600() (C ctober, 4)
II. TEST PROCEDURES
A. Tier 1 Test: UV-Visible Absorption Spectra —
Estimation of Aqueous Photolysis Maximum Rate
Constant and Minimum Half—Life in Sunlight
The uv—visible absorption spectra in aqueous solution can he
determined by the methods described in Test Guideline CG—1050.
It is recommended that the following additional procedures be
followed:
(1) F r chemicals which ionize or protonate (e.g.,
carboxylic acids, phenols, amines), carry out uv—visible
absorption studies at pT-Is at least two orders or magnitude above
the PKa and at least two orders of magnitude below the pK .
Prepare buffer solutions at 25°C using reagent grade chemicals
and distilled water as follows:
pHs in the range 3_6*: NaH 2 PO 4
HC 1
pHs in the range 6_8* KH 2 PO 4
NaOH
*Use the minimum concentration of hufh rs to itt In tht’ lwi ir4 II
pH.
—12—
-------
Cc—6000 (October, ‘I9 4)
pUs in the range ‘ : Prepa r&’ hut to rs
described in the
Uandbook ot Chemistry
and Physics.
Check the pH of all the buffer solutions with a pH meter at 25°C
and adjust to the proper pH, if necessary. These buffer -
solutions can then be added to the test chemical solution until
the desired pH is obtained. If these buffers are inadequate,
then adjust the pH of the test chemical sol Ut ion with 1 M HC or
NaOH at 25°C.
(2) Measure the absorbance, A l as a function of
wavelength in the range 290—800 nm in duplicate. If applicable,
measure A at each experimental pH. Record, in duplicate, the
baseline when both the sample and reference cells are filled with
blank solutions. These data will be used to calculate the molar
absorptivities for the appropriate wavelength intervals and
wavelength centers, Table 1 (Section V..), where th test
chern ica I absorbs light. The wavelength center is &Iet i nod t ho
midpoint of the interval range.
It must be emphasized that the molar ahsorptivities f
test chemical must be carefully determined, especially in the
tails of the absorption bands at A 290 nm. Large errors will
be encountered in calculating photolysis rate constants and half—
lives if these measurements are not carefully carried out.
., ‘)
I
—13—
-------
CG—6000 (October, 1984)
B. Tier 2 Test: P gueous PhotolysisinSunlt jht
1. Test Conditions
a. Special Laboratory Equipment
It is recommended that quartz tubes be used for the
photolysis of chemicals with appreciable absorption at
wavelengths below 340 nm. Chemicals that absorb appreciably at
wavelengths greater than 340 nm may he tested in borosilicate
tubes. Thin—walled borosilicate or quartz tubes are
recommended. Disposable culture tubes (13 x ‘LOO mm) with Tefl n—
lined screw caps or quartz tubes with quartz or boro.si 1 icato
stoppers, Teflon—lined, may be used as reaction vessels. Tubes
of 11 mm i.d. are recommended. For some chemicals, it may be
difficult to determine the concentration of the test chemical in
reaction tubes of small volume. For these chemicals, larger
volume reaction vessels are recommended provided that the cell
walls are thin and the pathlength of radiation through the vessel
is less than 0.5 meter.
b. Purity of Water
J .ip nt w.1 it , • , w.i I I t A i’M I’y I I A
standards, or an equivalent grade, is highly r mu enIed t. )
—14—
-------
CG—6000 ((‘)ctober, 1984)
minimize biodegradation. ASTM Type II A water is described in
ASTM D 1193—77, “Standard Specification for Reagent Water” . Air-
saturated water can be easily prepared by allowing the water to
equilibrate in a vessel plugged with sterile cotton.
c. Sterilization
It is extremely important to sterilize all qlassware and t()
use aseptic conditions in the preparation of all solutions and in
carrying out all photolysis experiments to eliminate or minimize
biodegradation. Glassware can be sterilized in an autoclave or
by any other suitable non—chemical method.
d. pH Effects
It is recommended that all photolysis experiinents be carried
out at pHs at least two orders of magnitude above the and it
least two orders of magnitude below the PKa for any chemical
which ionizes or protonates (e.g., carboxylic acids, phenols, and
amines). Buffers described in Section II.B.2.h. should he used.
e. Volatile Chemical Substances
Special care should he taken when testing a volatile
chemical so that the chemical substance is not lost due to
volatilization during the course of the phot:oly i ; experiu1 mI
Thus , it is important to e f feet: i ye I y ;ua I I h’ r ,? , I I HI V’ ; ;t’ I
1•
—15—
-------
(‘(;—(u)d() ( c t ‘)h F , 1 .1
Disposable culture tubes with Teflon—lined screw caps or quartz
tubes with quartz or borosilicate stoppers, Teflon—lined, are
recommended. Volatile compounds can be conveniently studied in
culture tubes equipped with Mininert® valves. Samples can be
introduced into or removed from the tubes through the septum in
these valves with no loss of substrate. Ps an alternative, the
tubes can he sealed with a torch. In addition, the reaction
vessels should he as completely filled as is poSSible to preVent
volatilization to any air space.
f. Control Solution
it is extremely important to take certain precautions to
prevent loss of chemical from the reaction vessels by processes
other than photolysis. For example, biodegradation and
volatilization can be eliminated or minimized by use of sterile
conditions and minimal air space in sealed vessels. Hydrolysis
is a process which cannot he minimized by such techniques. Thus,
control vessels containing test substance which are not: exposed
to sunlight are required. In this way, the loss of test cheinic l
for processes other than photolysis may be determined and
eliminated. For simplicity, if the loss of chemical in the
control is small (i.e., approximately 10% or less), one can
calculate a first—order loss, ki 05 , and subtract it from
(kp)obg to give the corrected direct photolysis rate const int
k . If hydrolysis is found to he significant (i. ., qreah’r tihin
U ’’
•‘)
—lh-
-------
CG—600() (Uctoher, 1984)
1 0 ) , hydrolys is stud es should be ei rt- i ed t u( t . I
Guide line CG—50001
g. Absorption Spectrum as a Criterion for
Performing the Aqueous Photolysis Test
This aqueous photolysis screening test is applicable to all
chemicals which have uv—visihle absorption maxima in the range
290—800 run. Some chemicals have absorption maxima significantly
below 290 run hut have measureahie absorption tails hove the
baseline in their absorption spectrum at wavelengths greater than
290 nm. Photolysis experiments should he carried out for these
chemicals. The absorption spectrum of the chemical in aqueous
solution can be measured by Test Guideline CG—1050.
h. Sunlight Actinometer
In order to quantify the rate of photolysis more precisely,
it is necessary to measure the sunlight intensity incident, on the
sample during photolysis. A standard p—nitroacetophenono—
pyridine actinometer (PNAP/PYR) has been developed IMill et d.
(1981, 1982b); Dulin and Mill (1982)) to measure the suniiqht
intensity incident on the sample during photolysis and this
actinometer has been incorporated in this Test Guideline.
According to equation 4, the rate constant is a function of the
reaction quantum yield. Furthermore, the reaction quantum yield
can be adjusted by varying the molar concentration of the
—17—
-------
CG—600C) (October, 1984)
pyridine according to equation 9. Hence, by varying the pyridine
concentration, the actinometer photolysis rate constant can be
adjusted so that the half—life can ranqe from several hours t
several weeks. The initial concentrat ton of PNAP i:; set: at 1.flt)
x lO M.
Using the test chemical photolysis rate constant,
determined in Tier 2, phase 1, and the variable =
listed in Table 2, the molar concentration needed to adjust the
rate of disappearance of PNAP in PNAP/PYR to match the rate of
disappearance of the test chemical is given by
EPYRI = 26.9 (k/k) (11)
Experiments are caned out by simultaneously photoly. ing the
test chemical and actinometer solutions. The concentrations of
test chemical and actinometer are measured periodically as a
function of time. These data are then used to determine the
ratio of the rate constants, k /k , using linear regression
analysis on the following equation
1n(C 0 ,C )c = (k /k ) ln(C 0 /C )a
with in ( a as the independ ni Vrl r I thli and n( (‘
I Iui •lt1l,. I$Il flt V.11. .lt) I . •rh’’ ,t I I Iii’ I) ! t ta iqht I i i
is the ratio of the rate constants, kd/ka.
pp
-18-
-------
CG—6000 (October, 1984)
i. Solar Irradiance Data
In order to calculate the reaction quantum yield of the test
chom i cal , a nti then ca 1 c u 1 ate k ,t ud t’ 1 ,, , it I ;
necessary to use the solar irrathance parameter L . L values
are proportional to the average light flux that is available to
cause photolysis in a wavelength interval centered at A over a
24—hour day at a specific latitude and season date. The
values are defined by the angle of declination of the sun at —20°
for winter, —10° for fall, +100 for spring and +20° for summer.
The actual dates for 1982 that correspond to these angles of
declination are January 21, April 16, July 24, and October 20,
for winter, spring, summer and fall, respectively FAA (1982)1.
The L values for these season dates are listed in Tables 3 to 6
(Section V.A) as a function of latitude and are applicable to
clear sky conditions, water bodies, shallow depths, and for
chemicals whose absorbance is less than 0.05 in pure water [ Mill
et al. (1984)].
j. Geometry of the Reaction Vessel
The method of Zepp and Cline (1977) and the method of Mill
et al. (l982a) are applicable to sunlight incident on a water
surface such as a natural water body while the method (lOve 1l)J)4 d
in th is Test ( u ide Ii ne rneisiiros rate ( ( )w; t. tnt s ( k ) I n I ul e ;
p
(e.g., 13 x lOf) mm) . however, rates in tubes are I ast r t han ju
:9
—19—
-------
(‘ ( —t- 1)fl(I ( t ’ h r , l R4 )
water bodies and it has been experimentally observed [ Mill et al.
(1982b)) that
k = 2.2k (13)
P
Because tubes are the simplest and easiest reaction vessels to
use, this Test Guideline recommends the use of tubes as reaction
vessels and the method has been modified to take into account. the
increased rate in tubes (equation 13).
k. Chemical Analysis of Solutions
In determining the concentration of the chemical in
solution, an analytical method should he selected which is most
applicable to the analysis of the specific chemical substance.
Chromatographic methods are generally recommended because of
their chemical specificity in analyzing the parent chemical
substance without interference from impurities. Whenever
practicable the chosen analytical method should have a precision
of ± 5 percent or better.
The p—nitroacetophenone in the chemical actinometer solution
is conveniently analyzed by high—pressure liquid
chromatography using a 30 cm C reverse—phase column and a uv
detector set at 280 nm. The mobile phase in volume. percent- i ;
ro
—20-
-------
(‘(;—t oOO (Oct (‘ 1 t , I
2.5% acetic acid, 50% acetonitrile, and 47.5% water which is
passed through the column at a flow rate of 2 mL/minute.
2. Preparations
a. Preparation of Test Chemical Solution
Prepare homogeneous solutions with the chemical at less I-han
one—half of its soluhiUty in water and at a concentration such
that the absorbance is less than 0.05 in the photolysis reaction
vesel at wavelengths greater than 290 nm. For very hydrophobic
chemicals, it is difficult and time consuming to prepare aqueous
solutions. To facilitate the preparation of aqueous solutions
containing very hydrophobic chemicals and to allow for easier
analytical measurement procedures, the following procedure may be
used to aid in the dissolution of the chemical. Dissolve the
pure chemical in reagent grade acetonitrile. Add pure water as
described under Test Conditions, Section [ I.B.l.b, or buffer
solution as described under Preparations, Section II.B.2.h., for
chemical substances which ionize or protonate, to an aliquot of
the acetonitrile solution. Do not exceed one volume—percent of
acetonitrile in the final solution. Place the reaction solution
in the appropriate photolysis reaction tubes as described in
Section II.B.]..a.
ti
—21—
-------
CG—6000 (October, 1Q84)
b. Preparation of Buffer Solutions
Prepare buffer solutions according to the procedures
outlined in Section II.A.l using reagent grade chemicals and pure
water as described under Test Conditions, Section II.B.l.b.
c. Preparation of Actinometer Solution
Using the test chemical photolysis rate constant, k ,
determined in Tier 2, Phase 1, and the variable k’ listed in
Table 2, the molar concentration of pyridine needed to adjust Lh?
rate of disappearance of p—nitroacetophenone (PNAP) to match the
rate of disappearance of the test chemical can he obtained from
equation 11. The variable (= is equal to the day—
average rate constant for sunlight absorption by PNAP which
changes with season and latitude. The value of k is selected
from Table 2 for the season nearest the mid-experiment date of
the Tier 2, Phase 1, studies and the decadic latitude nearesL the
latitude of the experimental site.
Once the molar concentration of pyridine IPYR1 has been
determined, an actinometer solution can he prepared as follows.
Dissolve 0.165 gms. of PNAP in 100 mL of acetonitrile (0.01 M)
Add 1 niL of this solution to a one liter volumetric flask. Add to
the volumetric flask the mass in grams, or the volume (v) of
pyridine at 20°C, obtained from the equations
—22--
-------
CG— OOO (October, HR4)
tnass(grams) 7()• I ( PYRI
(14)
V(mL) = 80.6IPYR1
Fill the volumetric flask with pure water (Section II.B.l.b.) to
give one liter of solution and shake vigorously to make sure that
the solution is homogeneous. The PNAP/PYR solution should be
wrapped with aluminum foil and kept from bright light.
3. Performance of the Tests
a. Phase 1 Experiments
For all experiments, prepare an aqueous solution of the
chemical substance, as described in Section II.B.2.a., and a
sufficient number of samples in quartz or borosilicate glass
tubes to perform all the required tests. Fill the tubes as
completely as possible and seal them. Prepare two control
samples in the absence of ultraviolet light and totally exclude
light by wrapping the tubes with aluminum foil or by other
suitable methods. These samples are analyzed for Ihi’ chem ieal
substance immediately after completion of the experiment to
measure the loss of chemical in the absence of light. Place the
samples, including the controls, outdoors in an area free of
shade and reflections of sunlight from windows and buildings.
Place the samples on a black, non—reflective background and
inclined at approximately 300 from the horizontal with the upper
—23—
-------
CG—6000 (October, 1984)
end pointing due north (in the northern hemisphere). Conduct the
photolysis experiments during a frost—free time of year (e.i.,
May, June, July, August, or September in the northern hemisphere
— — weather permitting) and start the experiments initially at
noon (1200 hours). Record the date and time the experiment was
begun, the date and time completed, the time of sunrise and
sunset on all days when photolysis experiments were performed,
the times exposure was stopped and restarted for intermittent
exposure, the weather conditions during the period, and the
latitude of the site. For chemical substances that ionize or
protonate, carry out photolysis experiments at the required pHs
as described under Test Conditions, Section IT.B.l.d.
If a significant loss of test chemical has occurred in the
control samples, determine the cause and eliminate or minimize
the loss. If hydrolysis is found to be significant, hydrolysis
studies should be carried out first (Section II.B.1.f.).
Use one of the following procedures, depending on how fast
the chemical substance photolyzes.
i. Procedure 1
If the chemical substance transforms 50—80% within 2R ddy ,
measure the concentration of the chemical substance, in
duplicate, at time t = 0 and periodically (at least four data
points at approximately equal time intervals) at noon (1200
hours) until at least 50% of the substance has been consumed. As
-------
CG—6000 (October, 1984)
a simplification, the sampling times can he estimated as the
photolysis experiments are in proqrcss. fleterinin ’ the
concentration of test chemical t rom two, F reshly t’n 1 , e ct ion
tubes for each time point. Determine the concent rat ion in each
of the two control solutions as soon as the photolysis
experiments are completed.
ii. Procedure 2
If the chemical substance transforms in the range of 20—50%
in 28 days, determine the concentration of the chemical
substance, in duplicate, at time t = 0. Determine the
concentration of the chemical in the two separate reaction tubes
and the two control tubes after 28 days of photolysis.
iii. Procedure 3
For chemical substances that transform in sunlight 50—80%
within one or two days, place the samples outside at noon (1200
hours) and analyze two samples for the concentration of the
chemical substance at t = 0, and in two, freshly opened, reaction
tubes at noon (1200 hours) the next day, and again, in two,
freshly opened, reaction tubes at noon (1200 hours) the second
day. Determine the concentration of the test chemica in ‘ irb 1
the two control solutions after the Ii rst: di y i i ph I. fl y i ; wd
as soon as the photolysis experiments haVe been C InJ)I ete(1 on I Ii
second day.
‘j’)
—25—
-------
CC,—600() (October, HS4
iv. Analytical Methodology
Select an analytical method which is most applicable t the
analysis of the specific chemical being tested (Section II.R. 1..
k.).
b. Phase 2 Experiments
Using the test chemical photolysis rate constant,
determined in Tier 2, Phase 1, prepare an actinometer solution,
as described in Section II.R.2.c., and a sufficient number of
samples in quartz tubes to perform all the required tests’. Fill
all the tubes as completely as possible, seal them, and cover
them with aluminum foil as soon as possible after preparation.
Prepare an aqueous solution of test chemical, as described in
Section It.R.2.a., and a sufficient numher of samples in quarts’.
or horosilicate tubes to perform all the required tests. FiU
these tubes as completely as possible, seal them, and cover them
with aluminum foil as soon as possible after preparation. Place
all the samples outdoors in an area free of shade and reflections
of sunlight from windows and buildings. Place the samples on a
black, nonreflective background and inclined at approximately 30°
from the horizontal with the upper end pointing due north (in the
northern hemisphere). Remove the foil from all samples excej
for the test chemical control solutions and the art nornrtor
control solutions at noon (1200 hours). Based on the r’siiH t
the Phase 1 experiments, determine the concentration of test
—26—
-------
CG—6000 (October, 1984)
chemical and actinometer (PN P), in triplicate, at. time t = 0 and
periodically (at least five data points at approximately equal
time intervals). Determine the concentration of PNP P in the
three actinometer control solutions and the concentration of test.
chemical in the three control solutions for each time point.
Select an analytical method which is most applicable to the
analysis of the specific chemical tested, Section IT.B.l.k., and
follow the procedure given in Section II.B.1.k. for the analysis
of PNAP.
III. DP TA AND REPORTING
A. Tier 1 Test: tJV—Visible Absorption Spectra —
Estimation of Aqueous Photolysis Maximum Rate_Constant
and Minimum Half—Life inSunlicjht
1. Treatment of Results
The molar absorptivity can be determined from the absorption
spectra using the expression
C
LA = AA/CR (15)
where AA is the absorbance at wavelength A, C is the rno ar
concentration of test chemical, and 9. is the c c i! p; th1en; h in
centimeters. The molar absorptivity of the chemical sh u1d he
determined for the wavelengths listed in Table 1 for a solution
-27-
-------
CG—60fl() ( )c tOhe r , I
of concentration C and In a cell with pathlenqth Q. It the
absorption curve is flat within the interval around the
wavelength Xcenter, may he determined from the absorbance
at Acenter using equation 15. If a large change in
absorbance occurs within this interval, obtain an average
absorbance at Acenter based on the absorbances at the two
boundaries of the interval. Calculate an average c using the
average value of in equation 15. Determine the molar
ahsorptivity for each replicate and calculate a mean value.
rising the molar absorptivities obtained from the spectra and
the values of the LA from Tables 3 to 6, the maximum rate
constant (k ,) can be calculated at a specific latitude and
p . max.
season of the year using equation 7. The minimum half—life,
can then be calculated using this (kpE)max in
equation 8.
Two hypothetical examples are presented in an P.ppenciix,
Section V.8.1, to illustrate how the test data obtained in the
first—tier screening test can be used.
2. Test Data Report
(1) Submit the original chart, or photocopy, containing (t
plot of absorbance of test chemical vs. wavelength
plus the baseline. Spectra should include a readable
wavelength scale, preferably marked at 10 nm
—28—
-------
CG—600() (( ctober, l’ A4)
intervals. Each spectrum should be clearly marked
with the test conditions.
(2) Report the concentration of the test chemical
solution, the type of absorption cell used (quartz or
borosilicate glass) and the pathlength.
(3) Report A and CA at Acenter for each repi icate
and the mean value.
(4) Report (kpE)max and (tlj )m for the summer and
winter solstices using the appropriate L values
from Tables 3—6 closest to the latitude of the
chemical manufacturing site.
(5) Report the identity and composition of the solvent
used in the spectral absorption study.
(6) For ionizable chemicals, report its PKa• Report the
type and concentration of the buffers employed for
each pH. Report the p1-is in which the photolysis
experiments were carried out.
(7) Describe the method employed in determining the test
chemical’ s concentration.
(8) Report the name, structure, and purity of the test
chemical.
(9) Submit a recent test spectrum on appropriate reference
chemicals for photometric and wavelength accuracy.
(10) Report the name and model o. the spectr()I)horolneter
used.
—29—
-------
(‘( —ht)()() (c1 (‘ h ’t , I 4
(11) Report the various control settings employed with the
spectrophotometer. These might include scan speed,
slit width, gain, etc.
B. Tier 2 Test: Aqueous Photolysis in Sunlight
1. Phase 1 Experiments
a. Treatment of Results
If a small loss of test substance in the control tubes has
occurred, use this data to make corrections to the measured
photolysis rate (Section [ I.13.l.f.). Note the site of photolysis
and its latitude and the weather conditions. For procedures 1
and 2 note the dates and times of actual exposure including times
of sunrise and sunset and, in case the cells are moved to prevent
freezing or for other reasons, make sure that these Limes are
recorded and that the cells are kept in a dark place when
exposure is not in progress.
(1.) For chemical substances which transform 50—80% within
28 days, use a concentration C , which corresponds to
less than 50% of the initial concentration of chemical
substance remaining, and the corresponding time t, in
days, along with the initial molar concentration
in equation 5 to calculate k 1 ) in days 1 . I r in the
-30-
-------
CG—6000 (October, 1984)
analysis of the two samples at time t = 0 and t,
calculate a mean value of C 0 and C , respectively, and
a value of k . If a slight loss of chemical has been
detected in the controls, then calculate a rate
constant as follows: Calculate an averajo
concentration C , based on the duplicate measurements
of concentration in the controls. Use this
concentration along with the average initial
concentration in equation 5 and calculate a rate
constant kioss Using this rate constant along with
the observed rate constant, the corrected rate
constant is then
(kp)obs. — k 105 5 (16)
Calculate the half—life, using the corrected
value in equation 6;
(2) for chemical substances which transform 20—50% in 28
days, use the mean concentration C remaining at t =
28 days along with the mean value of C 0 to calculate
k . Use the same procedure as described above to
calculate the value of k and ti , ,. If less than 20’ of
the chemical substance degrades in 28 days, report the
mean concentration of C and C 0 . In this case the
apparent half—life is reported as greater than 3
months; and
-31- Cl
1
-------
CG—6000 (October, l’ S4)
3) for chem tea 1 stiL -3t anees wh i el i t r ns t nu ‘ 0 or more i n
the first day, as described in procedure 3, ca ctilate
a full day k value using the mean concentration C 1 of
chemical substance remaining at noon (1200 hours)
after the first day along with the mean value of C 0
using equation 5. For chemical substances which
degrade less than 50% at noon (1200 hours) after the
first day but 50% or more at noon (1200 hours) the
second day, calculate k using the mean concentration
of chemical substances remaining at noon (1200 hours)
the second day. Calculate the half—life, ti 7 , 2 , using
the mean value of k in Equation 6. If a small 1o s
of test substance in the control tubes has occurred,
use this data to make corrections to the mesured
photolysis rate as described above. Note the dates of
photolysis, the latitude, and the site.
A hypothetical example is presented in an Appendix, Section
V.13.2, to illustrate how the test data obtained in the Tier 2,
Phase 1, test method can he used.
b. Specific Analytical and Recovery Procedures
(1) Provide a detailed description or reference for the
analytical procedures used, including thecalibration
data and precision; and
1”)
—32—
-------
CG—6000 (October , l9 4
(2) if extraction methods were used to separate the solute
from the aqueous solution, 1 rovide a description ot
the extraction method as well as the r1’ v ry dat a
c. Other Test Conditions
(1) Report the size, approximate cell wall thickness, and
type of glass used for the reaction tubes;
(2) report the initial pH of all test solutions, if
appropriate;
(3) for all procedures, report the dates of photolysis,
the time of sunrise and sunset on each photolysis day,
the site of photolysis and its latitude, and the
weather conditions. For procedures I and 2 submit the
dates and times of actual exposure, and the duration
of exposure, and, for intermittent exposure, the
fraction of each day during which photolysis occurred;
(4) if acetonitrile was used to solubilize the test
substance, report the percent, by volume; and
(5) if a significant loss of test chemical occurred in the
control solution, indicate the causes and how they
were eliminated or minimized.
d. Test Data Report
(1) For each photolysis experiment, report:
-33- G3
-------
C(;—60t)() ()c tohe r , q 4 ‘
i. The initial molar concentration ot test chemical
(C 0 ) of each replicate and the mean value.
ii. The molar concentration of test chemical. for
each replicate and the mean value for each time
point t.
iii. The molar concentration of each replicate
control sample and the mean value after
completion of the photolysis experiments.
(2) For procedure 1, 2, or 3, report the value of k . If
small losses of chemical are observed, report
(kp)obs k 1055 and k . Report the half—life (t i , , 2 )
calculated using the value of k .
2. Phase 2 Experiments
a. Treatment of Results
The objectives of this set of experiments is to determine
the sunlight reaction quantum yield, q , for a specific tiest
chemical. can he calculated usinq equation 10,
c k C LX a
a c (11))
k cALA
by the following steps:
—34—
-------
Cc 1 —60fl() ()ctober , lq 4
(1) Determine the ratio of the rate constants, k /kC,
described in Section 1:I.B.i.h using equation 12. If a
slight loss of test chemical or actinometer (PNAP) was
detected in the controls at any time t, then employ
the following procedure. Consider, as an example, the
loss of test chemical in the control at time t. Using
the average concentration of the test chemical in the
controls from the replicates at time t and the average
initial concentration, calculate ln(Co/Ct)Closs.
Using the average concentration of test chemical from
the replicates after photolysis time 1, calculate
ln(Co/Ct)Cobs . The corrected term is then
ln(C /C )c = ln(C Ic )C — ln(C /c )C . (17)
o t corr. o t obs. o t loss
The same procedure can be applied to obtain a
corrected term from the actinometer (PNA P).
Using the corrected terms for test chemical and/or
actinometer in equation 12, determine the ratio of the
rate constants (ka/kc) as described in Section
pp
It .B.l .h
(2) Determine the quantum yield of the actinorneter, ,
using equation 9 and the molar concentration of
pyridirie [ PYRI present in the actinometer.
(3) Determine the value of for the test chemical
as tollows: the molar absorptivities, , have
. )
-------
C(—600() 1c t I t’ r , 4)
been determined by the p1- edLlr ’ (J jVt fl in St c t i n
II.A. and the results have been tabulated aCcording tc
Section III.P.2. Choose the appropriate L valiie
(Tables 3 to 6) that correspond to the season closest
to the season in which the Phase 2 experiments were
performed and to the latitude nearest the latitude of
the experimental site. Calculate the product of
and L for each wavelength interval where has a
nonzero value. Sum the products of c L over all
wavelength intervals.
(4) fletermlne the value of for the actinometer,
as follows: These values have been calculated and are
given in Table 2. Choose the appropriate value that
corresponds to the season closest to the season in
which the Phase 2 experiments were performed and to
the latitude nearest the latitude of the experimental
site.
(5) Substitute the values of k /k , YC LA, anci
in equation 10 and calculate Lhe quantum
yield of the test chemical iii the env i rt nment ( i . •
in sunlight).
Once has been determined, equation 4 can he used to
calculate kpE at any season of the year and latitude using th
measured values of the molar absorptivities, c , and the
—36--
‘I I
-------
CG—6000 (October, lqS4
appropr tate “A values ( Tahi es I to h • The h i 1 t — Ii t’ c ui t h u
he cit 1 cul ated us t nj k in equ i t I on 2
- ph
A hypothetical example is presented in an Appendix, Section
V.B.3, to illustrate how the test data obtained in the Tier 2,
Phase 1, test method can be used.
b. Other Test Conditions
(1) Report the size, approximate cell wall thickness, and
type of glass used for tubes to hold the test chemical
and actinometer solutions.
(2) Report the initial pH of aH test. cheinic t1. solutions,
if appropriate, and the type and concentration of the
buffers employed for each pH.
(3) if acetonitrile was used to solubilize the test
chemical, report the percent, by volume, of the
acetonitrile, which was used.
(4) If significant loss of test chemical occured in the
control solution, indicate the causes and how they
were eliminated or minimized.
G7
—37—
-------
CG—6000 (October, 1984)
c. Test Data Report
(I) Report the initial, molar concentration ot hemtLal
(C 0 ) of each replicate and the mean value.
(2) Report the initial molar concentration of PNAP and the
molar concentration of pyridine used in the
act inometer.
(3) Report the time and date the sunlight photolysis
experiments were started, the time and date the
experiments were completed, and the elapsed photolysis
time in days.
(4) For each time point, report the three separate valuos
for the molar concentration of test chemical and PN\P
and the mean values.
(5) For each time point, report the three separate values
of the molar concentration of test chemical and PNT P
for the controls and the mean values.
(6) Tabulate and report the following data: t,
ln(Co/Ct)c, and 1n(C 0 /C )a. From the linear
regression analysis, report the ratio of the rate
constants, k /k , and the correlation co tficient
(7) If loss of test chemical and/or actin neti r w.i ;
observed during photolysis, then report the dald
lfl(c/C ), ln(Co/ct)ohs, ln(c 0 /ct)i for LIu?
test chemical and/or actinometer at each time t. F r un
the linear regression analyis of ln(Co/Ct)ccorr
-38- CS
-------
CG—6(l(lti (Oc tuber, 1 L 4
ln(co/ct)acorr, report the ratio of the rate
constants, k /k and the correlation coefficient.
(8) Report the reaction quantum yield of the actinometer
( ).
(9) Report the value of ka for the actinometer
corresponding to the season closest to the se isori in
which the photolysis experiments were carried out and
to the latitude nearest the latitude of the
experimental site.
(10) Tabulate the values of Acenter, E , and c L
for the test chemical corresponding to the season
closest to the season in which the photolysis
experiments were carried out and to the latitude
nearest the latitude of the experimental site.
(11) Report the value for the test chemical from
step 10.
(12) Report the reaction quantum yield of the test
chemical.
(13) Report kpE and for the summer and winter seasons
using the appropriate L values from Tables 3—6
closest to the latitude of the chemical manufacturing
site.
(14) For chemicals that ionize, report the data for sLep
1—13 for the experiments at the requ I red 1d1
-39- C9
-------
C( —hO( O ( c t l ’r , I 4 \
IV. REFERENCES
AA. 1982. AstronomiCal Almanac.
ASTM. 1978. Annual book of ASTM standards, M erican Society for
Testing and Materials, Philadelphia, PA., Part 31, Method D
1193—77.
Dulin D and Mill T. 1982. Development and application of solar
actinometers. Environ Sci and Technol 16:815.
Handbook of Chemistry and Physics. The Chemical Rubber Co.
Cleveland, Ohio.
Mill T, Davenport SI , E ilin DE, Mahey WR, and Bawol R. 1QR .
Evaluation and optimization of photolysis screeninj protocols.
EPA—560/5—Rl—00 3.
Mill T, Mahey WR, Bomberger DC, Chou T—W, Hendry DG, and Smith
JH. 1982a. Laboratory protocols for evaluating the fate of
organic chemicals in air and water. EPA—600/3--82--022.
Mill T, Mabey WR, Hendry DG, Winterle J, Davenport 3, Barich V,
Dulin D, and Tse D. l982b. Design and validation of screening
and detailed methods for environmental processes. EPA—
Mill T, Davenport JE, Winterle JS, Mabey WR, Drossman H, Tse D,
and Liu A. 1984. Toxic substances process data generation and
protocol development. EP Contract No. 68-03—2981 with EPA
Athens Research Laboratory, Office of Research and Development.
Draft final report.
TJSEPA. 1979. U.S. Environmental Protection Njency. Oftice ot
Toxic Substances. Toxic Substances Control: Discussion )t
premanufacture testing policy and technical issues. Requ’- st: ror
comment. Federal Register 44, 16240.
Zepp RG and Cline DM. 1977. Rates of direct photolvsis in
aquatic environment. Environ Sci and Technol 11:359.
Zepp RG. 1978. Quantum yields for reaction of pollutants in
dilute aqueous solution. Environ Sci and Technol 12:327.
—40—
-------
CG—6000 (October, 1984)
V. APPENDICES
A. Appendix A.
Tables 1—6
TABLE 1. WAVELENGTH CENTER AND INTERVALS ‘OR LA
Xcenter
(nm)
297.5
Interval
From (nm)
Range
To (nm)
298.7
(nm)
2.5
296.2
300.0
298.7
301.2
2.5
302.5
301.2
303.7
2.5
305.0
303.7
306.2
2.5
307.5
306.2
308.7
2.5
310.0
308.7
311.2
2.5
312.5
311.2
313.7
2.5
315.0
313.7
316.2
2.5
317.5
316.2
318.7
2.5
320.0
318.7
321.2
2.5
323.1
321.2
325.0
3.8
330.0
325.0
335.0
10.0
340.0
335.0
345.0
10.0
350
345.0
355.0
10.()
360
355.0
365.0
10.()
370
365.0
375.0
10.1)
380
375.0
385.0
10.0
390
385.0
395.0
10.0
400
395.0
405.0
10.0
—41—
-------
CG—6000 (October, 1984)
TARLE 1 (continued)
Acenter
(nm)
410
Interval
From (nm)
1 ange
To (nm)
415.0
X
(nm)
10.0
405.0
420
415.0
425.0
10.0
430
425.0
435.0
10.0
440
435.0
445.0
10.0
450
445.0
455.0
10.0
460
455.0
465.0
10.0
470
465.0
475.0
10.0
480
475.0
485.0
10.0
490
485.0
495.0
10.0
500
495.0
505.0
10.0
525
512.5
537.5
25
550
537.5
562.5
25
575
562.5
587.5
25
600
587.5
612.5
25
625
612.5
637.5
25
650
637.5
662.5
25
675
662.5
687.5
25
700
687.5
712.5
25
750
725.0
775.0
51)
800
775.0
825.0
50
—42—
-------
CG—6000 (October, 1984)
TABLE 2. DAY AVERAGED RkI’E coNsrAN’r (k ) * F)R S1INLIGUT
ABSORPTION BY PNAP AS A FUNCTION OF SEASON AN!) DECkDIC LATITUDE
Season
Latitude Spring Summer Fall Winter
20°N 515 551 409 327
30°N 483 551 333 233
40°N 431 532 245 139
5O°N 362 496 154 64
*ka = in d
—43—
-------
(‘( —ht OU ( c t t ’r , 1
ThBLE 3.
VPJAIES FOR LATITL1DE 2 OoNa,h,
3.67 (—1)
3.54 (—1)
Fa 11
7.77 (—5) 3.71
2.96 (—4) 1.62
8.21 (—4) 4.99
1.79 (—3) 1.17
3.24 (—3) 2.25
5.13 (—3) 3.72
7.33 (—3) 5.47
9.68 (—3) 7.40
1.21 (—2) 9.38
1.44 (—2) 1.13
2.55 (—2) 2.04
8.75 (—2) 7.08
1.10 (—1) 9.02
1.22 (—1) 1.01
1.35 (—1) 1.12
1.45 (—1) .2I
1.55 (—1) 1. 30
1.46 (—1) 1.22
2.09 (—1) 1.75
2.76 (—1) 2.31
2.84 (—1) 2.38
2.74 (—1) 2.30
(—5)
(—4)
(—4)
(—3)
(—3)
(—3)
(.—3)
(—3)
(—3)
(—2)
(—2)
(—2)
(—2)
(—1)
(-1)
(—1)
(—1)
(—1
(—1)
(—1)
(—1)
(—1)
Xcenter
(nm)
Winter
Spring
Summer
297.5
1.10
(—4)
1.52
(—4)
300.0
4.06
(—4)
5.26
(—4)
302.5
1.10
(—3)
1.35
(—3)
305.0
2.37
(—3)
2.79
(—3)
307.5
4.24
(—3)
4.86
(—3)
310.0
6.65
(—3)
7.45
(—3)
312.5
9.42
(—3)
1.04
(—2)
315.0
1.24
(—2)
1.35
(—2)
317.5
1.54
(—2)
1.66
(—2)
320.0
1.82
(—2)
1.96
(—2)
323.1
3.23
(—2)
3.45
(—2)
330.0
1.10
(—1)
1.17
(—1)
340.0
1.37
(—1)
1.45
(—1)
350.0
1.52
(—1)
1.60
(—1)
360.0
1.67
(—1)
1.76
(—1)
370.0
1.78
(—1)
1.88
(—1)
380.0
1.89
(—1)
2.00
(—1)
390.0
1.79
(—1)
1.89
(—1)
400.0
2.57
(—1)
2.71
(—1)
410.0
3.38
(—1)
3.57
(—1)
420.0
3.47
(—1)
430.0
3.35
(—1)
_44...
I
-------
CG—6000 (October, 1984)
TABLE 3 (continued)
a t 9 of LA are io— einsteins c 2 day . Multipi ical ion o
L by E in units of mo1ar cm’ gives rate const:ant:s in
units of day .
bThe second number in the columns in parenthesis is the power
ten by which the first number is multiplierl.
CBased on the GC SOLAR program.
I—
(—1)
(—1)
(—1)
(—1
(—1)
(—1)
(—1)
A center
J L
!
440.0
3.95
(—1)
4.18
(—1)
1.25
450.0
4.45
(—1)
4.70
(—1)
3.65
460.0
4.50
(—1)
4.75
(—1)
3.70
470.0
4.65
(—1)
4.91
(—1)
3.83
480.0
4.76
(—1)
5.03
(—1)
3.92
490.0
4.50
(—1)
4.76
(—1)
3.72
500.0
4.59
(—1)
4.85
(—1)
3.80
525.0
1.21
1.28
1.00
550.0
1.26
1.33
1.05
575.0
1.27
1.35
1.06
600.0
1.29
1.36
1.07
625.0
1.29
1.37
1.08
650.0
1.30
1.38
1.09
675.0
1.30
1.38
1.09
700.0
1.29
1.36
1.08
750.0
2.48
2.62
2.08
800.0
2.38
2.51
2.00
Wi uter
2.72 (—1)
3.07 (—1)
3.11 (—1)
3.22 (—1)
3.31 (—1)
3.13 (—1)
3.20 (—1)
8.48 (—1)
8.83 (—1)
8.92 (—1)
9.05 (—1)
9.15 (—1)
.24 (—1)
9.27 (—1)
9.21 (—1)
1.78
I .71
—45—
-------
CG—6000 (t)ctobet , -1 4
TABLE 4.
VALUES FOR LATITUDE 30 0Na,h,c
Acenter
(run)
Spring
Summer
Fall
Winter
297.5
5.73
(—5)
1.09
(—4)
3.18
(—5)
6.78
(—6)
300.0
2.50
(—4)
4.11
(—4)
1.46
(—4)
4.23
(—5)
302.5
7.65
(—4)
1.14
(—3)
4.64
(—4)
1.71
(—4)
305.0
1.79
(—3)
2.46
(—3)
1.12
(—3)
4.95
(—4)
307.5
3.43
(—3)
4.45
(—3)
2.19
(—3)
1.11
(—3)
310.0
5.64
(—3)
7.02
(—3)
3.67
(—3)
2.04
(—3)
312.5
8.27
(—3)
1.00
(—2)
5.46
(—3)
3.26
(—3)
315.0
1.12
(—2)
1.32
(—2)
7.43
(—3)
4.69
(—3)
317.5
1.41
(—2)
1.64
(—2)
9.48
(—3)
6.21
(—3)
320.0
1.70
(—2)
1.95
(—2)
1.15
(—2)
7.76
(—3)
323.1
3.04
(—2)
3.46
(—2)
2.07
(—2)
1.43
(—2)
330.0
1.05
(—1)
1.18
(—1)
7.23
(—2)
5.17
(—2)
340.0
1.33
(—1)
1.48
(—1)
9.23
(—2)
6.75
(—2)
350.0
1.47
(—1)
1.63
(—1)
1.03
(—1)
7.65
(—2)
360.0
1.62
(—1)
1.80
(—1)
1.15
(—1)
8.60
(—2)
370.0
1.73
(—1)
1.91
(—1)
1.24
(—1)
9.31
(—2)
380.0
1.84
(—1)
2.04
(—1)
1.33
(—1)
1.01
(—1)
390.0
1.74
(—1)
1.93
(—1)
1.25
(—1)
9.3Y
(—2)
400.0
2.50
(—1)
2.77
(—1)
1.79
(—1)
1.35
(—1)
410.0
3.29
(—1)
3.64
(—1)
2.36
(—1)
1.79
(—1)
420.0
3.38
(—1)
3.74
(—1)
2.43
(—1)
1.84
(—1)
430.0
3.26
(—1)
3.61
(—1)
2.35
(—1)
1.78
(—1)
—46—
-------
c(;—hfl () (c t ohe r , 1
TABLE 4 (continued)
A center
( nm) Spring Summer Fall Winter
440.0 3.86 (—1) 4.26 (—1) 2.79 (—1) 2.12 (—1)
450.0 4.34 (—1) 4.79 (—1) 3.14 (—1) 2.39 (—1)
460.0 4.39 (—1) 4.85 (—1) 3.19 (—1) 2.42 (—1)
470.0 4.54 (—1) 5.01 (—1) 3.30 (—1) 2.51 (—1)
480.0 4.65 (—1) 5.13 (—1) 3.38 (—1) 2.58 (—1)
490.0 4.40 (—1) 4.85 (—1) 3.20 (—1) 2.44 (—1)
500.0 4.49 (—1) 4.95 (—1) 3.27 (—1) 2.50 (—1.)
525.0 1.18 1.31 8.67 (—1) 6.61 (—1)
550.0 1.23 1.36 9.03 (—1) 6.87 (—1)
575.0 1.24 1.37 9.11 (—1) 6.93 (—1)
600.0 1.25 1.38 9.24 (—1) 7.04 (—1)
625.0 1.26 1.39 9.34 (—1) 7.15 (—1)
650.0 1.27 1.40 9.45 (—1) 7.27 (—1)
675.0 1.28 1.40 9.48 (—1) 7.32 (—U
700.0 1.27 1.39 9.42 (—U 7.31 (—i)
750.0 2.44 2.67 1.82 1.41
800.0 2.34 2.57 1.75 1.37
a t 5 of L are i0 einsteins cm 2 day . Multiplication of
L by c n units of mo1ar cm’ gives rate constants in
units of day .
bThe second number in the columns in parenthesis is the power of
ten by which the first number is multiplied.
CBased on the GC SOLAR program.
—47—
‘I
-------
CG—6000 (October, 1984)
TABLE 5.
1 A VPtLIfl* FOR lAl’ I 1(1 1Th 400 N’ , 1) , C
A c t n t e r
(nm)
Sprinq
Summer
Fall Winter
297.5
1.85
(—5)
6.17
(—5)
7.83
(—6)
5.49
(—7)
300.0
1.06
(—4)
2.70
(—4)
4.76
(—5)
5.13
(—6)
302.5
3.99
(—4)
8.30
(—4)
1.89
(—4)
3.02
(—5)
305.0
1.09
(—3)
1.95
(—3)
5.40
(—4)
1.19
(—4)
307.5
2.34
(—3)
3.74
(—3)
1.19
(—3)
3.38
(—4)
310.0
4.17
(—3)
6.17
(—3)
2.19
(—3)
7.53
(—4)
312.5
6.51
(—3)
9.07
(—3)
3.47
(—3)
1.39
(—3)
315.0
9.18
(—2)
1.22
(—2)
4.97
(—3)
2.22
(—3)
317.5
1.20
(—2)
1.55
(—2)
6.57
(—3)
3.19
(—3)
320.0
1.48
(—2)
1.87
(—2)
8.18
(—3)
4.23
(—3)
323.1
2.71
(—2)
3.35
(—2)
1.51
(—2)
8.25
(—3)
330.0
9.59
(—2)
1.16
(—1)
5.44
(—2)
3.16
(—2)
340.0
1.23
(—1)
1.46
(—1)
7.09
(—2)
4.31
(—2)
350.0
1.37
(—1)
1.62
(—1)
8.04
(—2)
4.98
(—2)
360.0
1.52
(—1)
1.79
(—1)
9.02
(—2)
5.68
(—2)
370.0
1.63
(—1)
1.91
(—1)
9.77
(—2)
6.22
(—2)
380.0
1.74
(—1)
2.04
(—1)
1.05
(—1)
6.78
(—2)
390.0
1.64
(—1)
1.93
(—1)
9.86
(—2)
6.33
(—2)
400.0
2.36
(—1)
2.76
(—1)
1.42
(—1)
4.H
(—2)
410.0
3.10
(—1)
3.64
(—1)
1.87
(—1)
1.21)
(—1)
420.0
3.19
(—1)
3.74
(—1)
1.93
(—1)
1.24
(—1)
430.0
3.08
(—1)
3.61
(—1)
1.87
(—1)
1.20
(—1)
—48—
-------
CG—6000 (October, l9 34)
ThBLE 5 (continued)
X center
(nm)
440.0
450.0
460.0
470.0
480.0
490.0
500.0
525.0
550.0
575.0
600.0
625.0
650.0
675.0
700.0
750.0
800.0
Spring
3.65 (—1)
4.11 (—1)
4.16 (—1)
4.30 (—1)
4.40 (—1)
4.16 (—1)
4.25 (—1)
1.12
1 • 16
1. 17
1.18
1 . 20
1. • 21
1.22
1.21
2.33
2.25
Summer
4.26 (—1)
4.80 (—1)
4.85 (—1)
5.02 (—1)
5.14 (—1)
4.86 (—1)
4.96 (—1)
1.31
1 • 36
1.37
1.38
1.40
1.41
1.41
1.40
2.69
2.59
Fall
2.22 (—1)
2.51 (—1)
2.54 (—1)
2.63 (—1)
2.70 (—1)
2.56 (—1)
2.62 (—1)
6.93 (—1)
7.21 (—1)
7.22 (—1)
7.39 (—1)
7.50 (—1)
7.62 (—1)
7.68 (—1)
7.66 (—1)
1.48
1.43
Winter
1.43 (—1)
1.61 (—1)
1.64 (—1)
1.69 (—1)
1.74 (—1)
1.65 (—1)
1.68 (—1)
4.45 (—1)
4.61 (—1)
4.61 (—1)
4.69 (—1)
4.82 (—1)
4.95 (-4)
5.03 (—1)
5.05 (—1)
9.84 (—1)
9.56 (—1)
a t 5 of L are io— einsteins çm 2 day. Multiplication of
Lx by c in the units of molar’ cm gives the rate const;ant
in units of day’.
hThe second number in the columns in parenthesis is th. power of
ten by which the first number is multiptiei.
cBased on the CC SOLAR program.
—49—
-------
CG—6000 ( ct bt r, 1’ 4
ThBLE 6.
VALtI1:S ‘OR I A till o K (l , h , c
Xcenter
(nm)
297.5
Spring
Summer
(—5)
Fall
9.58
(—7)
Winter
5.47
(—8)
3.61
(—6)
2.86
300.0
3.05
(—5)
1.50
(—4)
8.27
(—6)
4.17
(—7)
302.5
1.54
(—4)
5.33
(—4)
4.47
(—5)
2.62
(—6)
305.0
5.24
(—4)
1.39
(—3)
1.63
(—4)
1.34
(—5)
307.0
1.32
(—3)
2.89
(—3)
4.39
(—4)
5.14
(—s)
310.0
2.66
(—3)
5.05
(—3)
9.32
(—4)
.49
(—4)
312.5
4.53
(—3)
7.75
(—3)
1.66
(—3)
1.43
(—4)
315.0
6.82
(—3)
1.08
(—2)
2.58
(—3)
6.52
(—4)
317.5
9.34
(—3)
1.40
(—2)
3.64
(—3)
1.07
(—3)
320.0
1.19
(—2)
1.71
(—2)
4.76
(—3)
1.57
(—3)
323.1
2.25
(—2)
312
(—2)
9.19
(—3)
3.39
(—3)
330.0
8.26
(—2)
1.10
(—1)
3.48
(—2)
1.45
(—2)
340.0
1.09
(—1)
1.40
(—1)
4.71
(—2)
2.12
(—2)
350.0
1.22
(—2)
1.57
(—1)
5.43
(—2)
2.53
(—2)
360.0
1.36
(—1)
1.74
(—1)
6.18
(—2)
2.96
(-2)
370.0
1.47
(—1)
1.86
(—1)
6.76
(—2)
3.10
(—2)
380.0
1.57
(—1)
1.99
(—1)
7.37
(—2)
3.65
(—2)
390.0
1.48
(—1)
1.87
(—1)
6.89
(—2)
3.49
(—2)
400.0
2.12
(—1)
2.69
(—1)
9.90
(—2)
4.98
(—2)
410.0
2.80
(—1)
3.55
(—1)
1.31
(—1)
6.54
(—2)
420.0
2.89
(—1)
3.65
(—1)
1.35
(—1)
6.71
(—2)
430.0
440.0
2.79
3.31
(—1)
(—1)
3.52
4.17
(—1)
(—1)
1.31
1.55
(—1)
(—1)
6.47
7.66
(—2)
(—2)
—50—
-------
—6 i) () () ( c t r , 1 q 4
TABLE 6 (continued)
450.0
3.73
(—1)
4.69
(—1)
1.75
(—1)
8.62
(—2)
460.0
3.78
(—1)
4.75
(—1)
1.78
(—1)
8.74
(—2)
470.0
3.90
(—1)
4.91
(—1)
1.84
(—1)
8.95
(—2)
480.0
4.00
(—1)
5.03
(—1)
1.89
(—1)
9.15
(—2)
490.0
3.78
(—1)
4.76
(—1)
1.79
(—1)
8.62
(—2)
500.0
3.86
(—1)
4.85
(—1)
1.83
(—1)
8.77
(—2)
525.0
1.01
1.28
4.84
(—1)
2.28
(—1)
550.0
1.05
1.33
5.03
(—1)
2.32
(—1)
575.0
1.05
1.34
5.04
(—1)
2.28
(—1)
600.0
1.06
1.35
5.13
(—1)
2.32
(—1)
625.0
1.08
1.37
5.26
(—1)
2.42
(—1)
650.0
1.10
1.38
5.39
(—1)
2.53
(—1)
675.0
1.11
1.39
5.47
(—1)
2.61
(—1)
700.0
1.11
1.38
5.49
(—1)
2.66
(—1)
750.0
2.15
2.66
1.07
5.22
(—1)
800.0
2.08
2.57
1.04
5.11
(—1)
of
L\ by c
if\ unit
L are io— einsteins cm 2 day . Multiplication of
i the units of molar cm gives the rate constant
of day .
bThe second number in the columns in parenthesis is the power )f
ten by which the first number is multiplied.
CBased on the GC SOLAR program.
A center
(nm)
Spring
Summer
Fall
Winter
(_ j__4_.
—51—
-------
Cc,—6000 (October, l 84
B. Appendix B. Illustrative Examples
1. Tier 1 Test: IjV _ VisibleAbSOrI)tiOfl Spoctra —
Est imat ion of A eoUS Phot. is Rat ‘ Constant and
Minimum Flalf—lite in Sunlight
a. Illustrative Example 1.
A neutral organic chemical A was dissolved in water at a
concentration of 1.00 x l a — 3 M. UV—visible absorption spectra
were obtained in a 10.0 cm quartz absorption cell and no
absorbance was detected above the baseline in the region 290 nr’
and greater (i.e., A = 0 for A 290 nm). Since
= 0, then = C) (equation 15). using this result in
equation 7, it is found that (kpp)m 3 x = 0, indicating that no
direct photolysis can take place in sunlight at any latitude or
season of the year. This example illustrates the principle of
the Grotthus—Draper law, the first law of photocheruistry. That
is, in order for direct photolysis to take place in sunlight, the
chemical must absorb sunlight in the region A 290 nm.
b. Illustrative Example 2
Consider a plant located in Columbus, Georgia on bo
Chattahoochee River which produces an organic ChnmiCaI R WhiCh i ;
not an acid or a base. The waste effluent passes throiijh a
primary and secondary treatment plant and is then discharged
—52— c -’)
-------
(‘G— 6000 ( .‘)c t obt r , 1 ) t 4
directly into the river. The plant produces chemical B
continuously every day of the year. The plant is located at
32.5° north latitude. Estimate the maximum sunlight direct
photolysis rate constant and the corresponding minimum half—life
for this chemical in the river for the winter and summer seasons
under clear, skies.
Laboratory Experiments, Data, and Calculations
The water solubility of chemical B is 1.00 x iO M at
2°C. Chemical B was dissolved directly in water and a 1.00 x
io molar solution was prepared at 25°C. The uv-visible
absorption spectra were obtained according to the Tier 1
Procedure in a 10.0 cm quartz absorption cell in duplicate.
Using the wavelength interval range (from Table 1), the average
absorbance of the duplicate runs at Xcenter was obtained and
the results are summarized in Table 7.
—53—
-------
cG—6000 (C ctOher ,
TABLE 7. SUMMARY OF PHOTOLYSIS DATA FOR CHEMICAL B
A. SPECTRAL DATA
297.5
300.0
302.5
305.0
307 • 5
310.0
312.5
315.0
317.5
320.0
323.1
330.0
1.684
1.434
1.221
0.919
0.742
0.208
0.138
0.094
0.057
0.009
0.002
0.000
1684
1434
1221
919
742
208
138
94
57
9
2
0
Xcenter( rim)
c (M’ crn’)
— 4-.
-------
CC—6000 ( ctoher, 1984)
B. P}-IOTOLYSIS DATA
TABLE 7 (conLinued)
Acenter( nm)
Summer
Winter
(day )
C Lx (day )
0.18
L
6.78
E Lx
(—6)
0.01
297.5
1.09
(—4)
300.0
4.11
(—4)
0.59
4.23
(—5)
0.06
302.5
1.14
(—3)
1.39
1.71
(—4)
0.21
305.0
2.46
(—3)
2.26
4.95
(—4)
0.4’
307.5
4.45
(—3)
3.30
1.11
(—3)
0.82
310.0
7.02
(—3)
1.46
2.04
(—3)
0.42
312.5
1.00
(—2)
1.38
3.26
(—3)
0.45
315.0
1.32
(—2)
1.24
4.69
(—3)
0.44
317.5
1.64
(—2)
0.94
6.21
(—3)
0.35
320.0
1.95
(—2)
0.18
7.76
(—3)
0.07
323.1
3.46
(—2)
0.07
1.43
(—2)
0.03
330.0
1.18
(—1)
0.00
5.17
(—2)
0.0()
12.99
: 1.A
=
3. fl
*The units of L are in i0 3 einsteins cm 2 (lay 1 . rhe - cond
number in the c 1umns in parenthesis is the power of :en by whi:h
the first number is mu1tiplied.
—55—
qr-
14 )
-------
CG—6000 (October, 1984)
From the above data and equation 15, the average molar
absorptivity is
= 1000 (Ui)
From the average P value at A center, the average molar
absorptivity can be obtained from equation 18 and the results are
summarized in Table 7. Since the plant is located at 32.5° north
latitude, the closest L values are at 300 north latitude.
These values are obtained from Table 4 and are summarized in
Table 7 for the summer and winter seasons. Using the data from
Table 7 and equations 7 and 8, the following results are
obtained.
Summer Winter
(kpE)max = ) c L = 13.0 day (kpE)max = 3.3
(ta/ )min. = 0.053 day ( 1 1/ )min = 0.21 day
Since the chemical transforms rapidly for the summer and winter
seasons, it is necessary to carry out Tier 2 experiments to more
accurately define direct photolysis rates in aqueous media as a
function of the season of the year.
c “
—56—
-------
CG—6000 (October, H84
2.
a . Ti ustrat. ive Exampi e 3
Consider the same scenario as described in illustrative
example 2, Section V.B.l.h. Using the Tier 2, phase 1 Procedure,
carry out experiments to estimate the rate of direct photolysis
and half—life in aqueous solution in the spring for water bodies.
Photolysis Experiments and Calculations
Since chemical B absorbs appreciably below 340 run, 11 mm
i.d. quartz tubes were used (note: this tube has an approximate
pathlength of 1 cm). Chemical B was dissolved directly in pure
water and a 1.00 x i0 molar solution was prepared at 25°C.
Since the water solubility is 1.00 x io— M at 25°C, this sample
solution was well below one—half its water soluhility. The uv
spectrum of this solution in a one cm absorption cell indicated
that A was less than 0.05 at A 290 nm. Hence, under these
conditions, first—order kinetics are applicable.
A series of quartz tubes were filled with this aqueous
solution, sealed, and photolysis experiments were edrr ( (1 ( IO
sunlight according to the appropriate procedure descrih4 d in
Section II.B.3.a. The experiments were started at noon (1200
hours) on May 8, 1982. The weather conditions ar-c summarized for
jJ
-------
C( —6fl00 (Oct cr, 1 9 4
this period of time and the concent. rat ion data q i vcn represen
the mean of duplicate determinations.
(1) May 2, 1982: at t 0 (noon — 1200 hours)
C 0 = 1.00 x io— M.
(2) May 2, 1982: Noon to sunset — clear and sunny.
(3) May 3, 1982: Noon (1200 hours), C = 0.840 x lO 5 M.
(4) May 3, 1982: at 1400 hours the weather conditions
were cloudy with rain. The rain and cloudy weather
continued until 2200 hours.
(5) From sunrise, May 4, 1982 throuçjh 120(1 hours May 8,
1982, the weather was clear and sunny. At 1200 hours,
May 8, 1982, analysis of the samples gave an average
concentration of C = 0.400 x IO— 5 M. Since 60% of
chemical 13 transformed, the photolysis experiments
were terminated and the control samples were
analyzed. The average concentration of the control
samples was 0.9 7 x 10 M which was essential.ly the
same as c 0 . lence, no adventitious )roc s OS occurrt O
and the loss of chemical was only due to sunlight
photolysis.
Listed in Table 8 are the times of sunrise and sunset for
the dates sunlight photolysis experiments were carried out along
with the total number of hours of sunlight.
—58-
-------
CG—6000 (October, H84)
TABLE 8. SUMMARY OF’ TIMES FOR SUNRISE ANI) SUNSET POR TUE PERIOD
MAY 2—8, 1982
Date (1982) Sunrise (AM) Sunset (PM ) Total Sunlight Hours
5/2 0600 2010 14.2
5/3 0559 2011 14.2
5/4 0558 2012 14.2
5/5 0557 2013 14.3
5/6 0556 2014 14.3
5/7 0555 2015 14.3
5/8 0554 2016 14.4
The following data summarizes the dates photolyzed, . the
times exposed to sunlight, the total sunlight photolysis time for
each date in days, the total number of days of sunlight
photolysis, and the calculation of and t i , 7 .
Sunlight Photolysi
Time for Each Date
Date Times Photolyzed _______(d ys)
5/2 1200 hrs. to 2010 hrs. (8.2/14.2) 0.58
5/3 0559 hrs. to 1200 hrs. (6.0/14.2) 0.42
5/3 1200 hrs. to 1400 hrs. (2.0/14.2) 0.14
5/4 0558 hrs. to 2012 hrs. 1.00
5/5 0557 hrs. to 2013 hrs. 1.00
5/6 0556 hrs. to 2014 hrs. 1.00
5/7 0555 hrs. to 2015 hrs. 1.00
5/8 0554 hrs. to 1200 hrs. (6.1/14.4) 0.42
total 5.56 days
t 5.6 days; C 0 1.00 x 10 ; C 0.400 x
1n(C 0 /C ) = kCL
= (1/t) ln(C/C ) = (1/5.6)ln(1.00 x I0 /0.400 x
—59—
-------
CG—6 000 ( c tttht’ r , 4
= 0.16 days
0.693/0.16 days 1 = 4.3 days
Therefore, the rate constant for direct photolysis of chemical B
in tubes in pure water is 0.16 days 1 - and the corresponding half—
life is 4.3 days for the period of photolysis May 2—8, [ 982 at
32.5° north latitude. Using equation 13, the direct photolysi
rate constant (kpE) for water bodies is 0.073 days’ and the
corresponding half—life (t 4 , ) is 9.5 days.
3. Tier 2, Phase 2: Aqueous Photolysis in Sunlight
a. Illustrative Example 4
Consider the same scenario as described in illustrative
examples 2 and 3. rising the Tier 2, Phase 2, Procedure, carry
out experiments to determine the sunlight reaction quantum yie 1
and estimate the rate constant for direct photolysis in aqueous
solution and the half—life for water bodies and cleir ;ky
conditions for the summer and winter seasons.
—60—
-------
CG—6000 (October, 1984)
Photolysis Experiments and Ca culations
The sun 1 ight photol ys iS ex per i neut S were ca rr i ed eut in t he
heçjinninq of May, 1982 at. 32.5° north tat itude.
(1) Preparation of the Actinometer Solution
The results from the Tier 2, Phase 1, experiments indicated
that for the test chemical was 0.16 days . Since the
experiments were carried out in early May at 32.5° north
latitude, the value of was chosen from Table 2 which
corresponds to the spring season and at 300 north latitude; and
the value is 483 days . Using equation 11, the molar
concentration of pyridine required to adjust the actinometer rate
to match the rate of disappearance of the test chemical is
[ PYRJ = 26.9 (0.16/483) = 8.91 x iO— 3 molar.
Using this concentration of pyridine, an actinorneter solution was
prepared according to the procedure described in Section
II.B.h. The quantum yield for this actinometer is calculated
using equation 9.
= 0.0169 [ PYR ) = 0.0169(8.91 x 10 ) = .S1 x
( 1
—61—
-------
CG—6000 (October, 1984)
Procedures for Tier 2, Phase 2 experiments (Section
II.B.3.b) were followed and sunlight experiments were initiated
at 1200 hours on May 9, 1982. The mean initial concentration of
test chemical was 1.00 x l0 molar and the mean initial
concentration of PNAP was 1.00 x 10 molar. Samples of the
chemical and actinometer and the controls were analyzed in
triplicate periodically at 1200 hours on May 10, U, 13, 15, and
16. On May 16, the photolysis experiments were terminated. The
mean concentrations of all samples are summarized below.
Conc. of Conc. of Conc. of Conc. of
Date Chem. (M) Act. (M) Chem. Control (M) Act. Control
May 9 1.00 x l0 1.00 x l0 1.00 x 10 1.00 x 10
May 10 0.820 x l0 0.855 x 10 0.997 x 10 _s 1.00 X i0
May 11 0.654 x 10 0.710 x 10 1.00 x 10_S 0.997 x 10
May 13 0.440 x 10 0.515 x 10_s 0.996 x 10_s 0.999 x 10
May 15 0.299 x 10_S 0.383 x 10 0.999 x 10 0.998 x ur
May 16 0.233 x i0 0.304 x 10 0.997 x i0 0.996 x
Since no significant loss of PNAP or test chemical was obsecved
in the control samples, no adventitious processes occurred and
the loss of test chemical and PNAP was only due to sunlight
photolysis.
Using the above data, ln(C 0 /Ct) for the test chemical and
actinometer can be calculated and the results are summarized
below.
• ( T)
‘ lid
—62--
-------
(‘(—h(.)()() ( c t ‘ 1’)t’T• ,
The ratio of the rate constants, kC/k , is defined by
eq iation 12
ln(Co/Ct)c = (k /k ) 1n(C 0 /C )a (12)
Using all the data (including the time point t = 0) and
linear regression analysis, the slope is found to be 1.237 with a
correlation coefficient of 0.9998. Therefore
(kC/k&) = 1.24
pp
Using the molar absorptivities obtained in example 2,
Section V.B.l.h., and the L values for spring at 30° north
latitude, Table 4, the value of can he calcul.ated as
follows:
( 9
.
t (days)
Chemical
C x 10 (M) in (Co/Ct)c
Act inorneter
x xo (M) in (C 0 /C a
0
1.00
0.000
1.00
0.000
1
0.820
0.199
0.855
0.157
2
0.654
0.425
0.710
0.343
4
0.440
0.821
0.515
0.664
6
0.299
1.21
0.383
0.960
7
0.233
1.46
0.304
1.19
—63--
-------
CG —6000 (October, 1984)
* The units of
are i0 einsteins cm 2 day
= 9.96 days 1 .
For this experiment, k ( = c Lx) is 483 days (Table 2). AU
the pertinent data are summarized below:
= 1.24;
pp
ra
= 483 days ,
= 9.96 days ;
a
— 1.51 x
Substituting these results into equation 10 yields
+ = (1.2 4 )(483/9.96)(l.51x 10 )
= 9.08 x
Xcenter( rim)
(M cm ) I\ ( ys )
297.S
1684
5.73
(—5)
0.10
300.0
1434
2.50
(—4)
0.16
302.5
1221
7.65
(—4)
0.93
305.0
919
1.79
(—3)
1.65
307.5
742
3.43
(—3)
2.55
310.0
208
5.64
(—3)
1.17
312.5
138
8.27
(—3)
1.14
315.0
94
1.12
(—2)
1.05
317.5
57
1.41
(—2)
0.80
320.0
9
1.70
(—2)
0.15
323.1
2
3.04
(—2)
0.06
330.0
0
1.05
(—1)
0.00
—64—
-------
C—680() (O(’ t )t)t’t , I
The rate constants for direct photolysis ot t st chemical in
aqueous media and the half—life for water bodies and clear sky
conditions for the winter and summer seasons can he calcuL3ted as
follows: The values of have been calculated from example
1, Section V.B.1.h. For summer 13.0 days’; for winter
= 3.31 days 1 . The reaction quantum yield for the
chemical is 9.08 x i0 . Using these data in equation 4 yields
summer : kpE = 9.08 x l0 (13.0) = 0.118 days
winter: k = 9.08 x 10 (3.31) = 0.0301 days 1
— pE
These values can he substituted into equation 2 to obtain
the half—lives for these two seasons.
summer : ti/i = (0.693/0.118) = 5.9 days
winter : = (0.693/0.0301) = 23 days.
( r-
‘J
—65—
-------
C —tOOO
October, I’ H4
PHOTOLYSIS IN AOUEOUS SOLUTION IN SUNLIGHT
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXtC SUB TANC S
U • S • ENVI RONMENTAL PROTECT EON AGENCY
WASHINGTON, DC 20460
‘ I
-------
CS—6000 (October, 1984)
Contents
e
I. NEED FOR THE TEST
II. SCIENTIFIC ASPECTS ... 2
A. Rationale for the Selection of the Test Method 2
1. Historical Discussion 2
2. Selection of the Test Method 6
B. Development of the Test Method 9
1. Theoretical Aspects
2. Tier 1 Test: wi—Visible Absorption
Spectra—Estimation of Aqueous
photolysis Maximum Rate
Constant and Minimum Half—life
in sunlight 12
a. Determination of Molar Ahsorptivity 13
b. Solar Irradiance 15
c. Estimation of the Maximum Direct
photolysis Rate Constant and the
Corresponding Minimum Half-life 15
3. Tier 2 Test: Aqueous Photolysis in Sunlight... 16
a. Phase I Test 16
b. Phase 2 Test 16
i. Introduction 16
ii. Development of a Sunlight
Actinometer 18
iii. Geometry of the Reaction Vessel 22
iv. Phase 2 Test Method 23
4. Suimnary 27
C. Applicability and specificity 28
D. Rationale for the Selection of the Test:
Conditions
1. Special Laboratory Equipment
2. Purityof Water fl
3. Sterilization
4. Concentration of Solution
5. pH Effects
6. Control Solution 34
7. Absorption Spectrum as a Criterion for
Performing the Aqueous Photolysis Test 35
—1— ( ‘I . ,
-------
CS—6000 (October, 1984)
8. Sunlight Actinometer. . . . . • 35
9. Solar trradiance 1)ata . 38
10. GeometryoftheReactionVessel 43
11. ChemicalAnalysisofSolutions... ........ 43
12. Outdoor Experimental Conditions........ .. 44
III. REFERENCES.. •.... . . . . . ... . . . . . . . . . . . . . . . . . 46
IV. APPENDIX: GLOSSARY OF IMPORTANT SYMBOLS......... 49
1
—1.1—
-------
(-—6t):)() (October, 1 4
PHOTOLYSIS IN AOIJEOUS SOLIJrroN INS(JNLIGFIT
I. NEED FOR THE TEST
The majority of the earth’s surface is covered by water in
the form of oceans, seas, rivers, lakes, streams, or porids As a
result, chemicals are likely to enter aqueous media and could
then undergo transformation via direct aqueous photolysis,
Direct aqueous photolysis represents the transformation o a
chemical substance by direct absorption of radiant enerqy
(sunlight) into new chemicals different from their precursors.
Chemical substances which are present in aqueous media photolyze
at different rates depending upon the solar irradiance, the
chemical substance’s molar absorptivity at each wavelength of
solar radiation, and its reaction quantum yield at the
wavelengths of concern. Chemical substances which photolyze
rapidly under environmental conditions have relatively short
lifetimes in the environment. Consequently, the chemical fate
assessment may focus on the transformation products to a qreiter
extent than on the parent compound. On the other hand, if the
chemical substance is resistant to photolysis as well as to
the other possible transformation processes, the assessment
should focus on the parent chemical.
A cost—effective aqueous photolysis test is needed to assess
quantitatively the transformation of chemical substances in
sunlight. The importance of direct photolysis in sunliqht as a
—1—
-------
CS—600() (October, )‘ 84)
transformation process for chemical substances in aqueous media
in the environment can be determined quantitatively from data on
photolysis rate constants and half—lives.
This aqueous photolysis test represents a screening test to
allow one to determine how rapidly direct photolysis will, take
place in aqueous media under certain environmental conditions.
If the photolysis test data indicate that photolysis is a
relatively important transformation process and the initial risk
assessment indicates that there is a threat to the health of
humans and/or to the environment, then upper—tier tests may be
required to obtain more precise aqueous photolysis data over a
wide range of environmental conditions. These upper—tier tests
will also be concerned with determining the identity and fate of
the transformation products.
II. SCIENTIFIC P SPECTS
A. Rationale for the Selection of the Test Method
1. Historical Discussion
The scientific literature contains a number of publications
dealing with the photolysis in solution of various chemical
substances. Unfortunately, for one or more reasons, most of the
methods and data contained in the literature are of little or ne
use in determining aqueous photolysis rate constants and half.—
lives. Reasons for this include: (1) Many of the publications
-2-
-------
CS—6i)OO (October, VH4)
deal primarily with the inechan i sins nd products t rom direct
photolysis reactions rather than photolysis rates. (2) The
publications give no quantitative data on the rates of photolysis
under environmental conditions; i.e., many researchers used
sources of light which do not simulate sunlight or perforrded
experiments in solvents other than air—saturated water. (3) Some
studies are not based upon the fundamental laws of
photochemistry. (4) Many publications report the effects of
certain sensitizers which are environmentally irrelevant, or the
effect of other chemicals present in solution, on the chemical
substance being studied. (5) Many of the studies suffer from
poor experimental design, overlooking the controls necessary to
make sure that only photolysis and not biodegradation,
volatility, or other competing processes are taking place to
remove or transform the chemical substance. The following
paragraphs cite a few representative examples which illustrate
the points mentioned above, as well as the publication which
forms the basis for this Test Guideline.
A paper by Grunwell and Erickson (1q73) deals with the
photochemistry of parathion. This report is of: I it.I e rt lovnic
in the evaluation of photolysis rates of chemical substances for
several reasons. The main emphasis of this work is on the
identification of major products formed by the photolysis of
parathion rather than in the measurement of rate constants and
half—lives. Photolysis experiments were performed at three
wavelengths (254, 300, and 350 rim), of which only the last two
IC i
-------
(‘S—ht)Ot) (Oct er , ‘4 4 )
are environmentally relevant. The chemical was dissolved in
solutions of 20 percent water and either 80 percent ethanol or
tetrahydrofuran (by weight), neither of which is environmentally
relevant.
A paper by Langford et al. (1973) on the photolysis of
nitrilotriacetic acid claims environmental relevance by working
under environmental conditions. Radiation of 350 nm as well as
sunlight were used. Pure water and river water were used as
solvents. The authors presented no quantitative data on the rare
of photolysis. Detailed quantitative measurement of the
photolysis rate constant and half—life, along with adequate
controls, would he necessary to make this research useful for the
purposes of this Test Guideline.
Benson et al. (1971) photolyzed chiordane with both mercury
arc radiation and sunlight. Acetone was used as both a solvent
and photosensitizer. Sunlight photolysis was done only on the
pure compound applied as a thin layer under quartz glass. The
main emphasis of this work was to look at the chemical. structure
o the reaction products. Because rate constants were not
measured, acetone was used as both a solvent and suns i ti i zer, m 1
a mercury arc lamp was used as radiation source, their proceduru ;
have minimal applicability as a test method to screen for
environmental photolysis rates.
Su and Zabik (1972) studied the photochemistry of
arylamidine derivatives in distilled and natural water. A high
-4- 1C2
-------
CS—6000 (October, 1984)
pressure mercury arc filtered to remove radiation below 286 nm
was used . Prod uc t s of the react i on we re dole cm i nod l)U t no
kinetic studies were performed and no rite const nt or h i l t—
I ives wore report.e&i
‘Mancini (1978) presented a theoretical, framework for the
first—order photodecomposition of picloram in aqueous solutio 4 n
and tested the framework using some experimental data on the
photolysis of picloram in sunlight. In reality, the Mancini
framework is only an empirical approach describing the rate of
photolysis of picloram in sunlight.
Zepp and Cline (i977) published a paper on direct photolysis
in aqueous environments with equations for the determination of
direct photolysis rate constants in sunlight. This paper avuid
the problems illustrated above and serves as the Starting point
for the development of a test method for direct photolysis in
aqueous solution in sunlight. These equations translate readily
obtained laboratory data into rate constants and half—lives for
sunlight photolysis. photolysis half—lives can be calculated as
a function of season, latitude, time of day, depth in water
bodies, and thickness of the atmospheric ozone layer. A number
of published papers concerning the photolysis of chemicals in
sunlight have verified this method. (e.g., Wolfe ot al. I 376,
Smith et al. 1977, 1978, Zepp et al. 1975, 1976).
During the past few years mathematical models have become
available for predicting the behavior of chemicals in the
1C3
-------
CS—6t Ot) (Octoh ’r , ‘184
environment. A number of efforts have been made to validate the
models through studies of the fate of pollutants under various
field conditions or in outdoor ecosystems that have known inputs
of the chemical, controlled flow conditions, etc. Several of
these field studies confirmed that direct photochemical reaction
dominate the behavior of certain pollutants in natural waters and
the equations used in the model do a reasonably good job of
predicting rates in water bodies based on the work of Zepp and
Cline (1977). The following citations illustrate some of these
research efforts [ Broderius and Siiith (1980), Carey and RX
(1981), Crossland and Wolf (1984), Mabey et al. (1983), Sinimons
and Zepp (1984), Wolfe et al. (1982), Zepp (1983), and Zepp
(1982a)]
2. Selection of the Test Method
The methods in this Test Guideline were developed from e
thorough review of the research literature on the experimenti
determination of aqueous photolysis rate constants, the
principles outlined by Zepp and Cline (1977), Zepp (1978), the
method developed by the U.S. Environmental Protection Aqency,
Office of Toxic Substances [ USEPA (1979)], and by the research
carried out by SRI International [ Mill et al. (1981, 1982a,
1982b) and Dulin and Mill (1982)]
The proposed test method for the measurement of direct
sunlight photolysis of a chemical substance in aqueous sniut i
is based upon four fundamental Criteria. These criteria arc:
-6 ici
-------
CS—6000 (October, 1984)
(1) The test method should he based upon the fundamentals of.
photochem i stry . ( 2 ) The test met hod shun ii y i e 1 d u in t i t iv e
data on direct photolys is rates ot chem i cal subst in te us
media. (3) Sunlight should he used as the irradictLiurt source
because of its obvious environmental relevance as well as its ow
cost in comparison to artificial light sources (Zepp 1982). (4)
The test method should be designed to account for transformation
or chemical losses by mechanisms other than photolysis. For
example, the experiments should be designed to eliminate or
minimize biodegradation and volatilization and to account for
hydrolysis as a factor in the estimation of test substance
losses.
simple first—tier screening test is used to estimate the
direct aqueous photolysis rate constant and half—rife in sunliqht
as a function of latitude and season of the year. An upper limit
for this rate of photolysis can be obtained by assuming that
photolytic transformation occurs with a reaction quantum yietd of
unity and by equating the rate of reaction with the rate of
absorption of light. The latter rate, which represents the
maximum rate constant, can be estimated as a function of latitude
and season of the year by combining ultraviolet—visible
absorption spectral data with the appropriate solar irrr ii ince
data. The maximum rate constant can then he used to obt:a in tin
corresponding minimum half—life for photolysis in w ter in
sunlight. If these data indicate that aqueous photolysis is in
important transformation process relative to other transformation
-7-
‘-,
-------
CS—600() (OcLober, i’ 84
processes (e.g., biodegradation, hydrolysis, oxidation, etc.’
then additional tests (i.e., upper—tier tests) should be carried
out to determine environmentally relevant rate constants and
• half—lives.
second—tier aqueous photolysis screening test is used to
determine rate constants and half—lives in the presence of
sunlight. This method is divided into two phases. In phase one,
the test chemical is photolyzed in sunlight in order to obtain an
approximate rate constant [ USEPA (1979)1. However, this method
only gives an approximate rate constant since it fails to measure
sunlight intensities incident on the sample during photolysis.
In phase two, a standard actinometer is used to overcome this
deficiency [ Mill et al. (1981, 1982b) and Dulin and Mill
(1982fl. The rate constant of the standard actinometer can be
adjusted so that its half—life can range from a few hours to
several weeks. By using the data from phase one, the rate
constant of the standard actinometer can he adjusted to match Lhe
approximate rate constant of the test chemical. Then, by
simultaneously photolyzing the test chemical and the standard
actinometer in sunlight, the sunlight reaction quantum yield of
the test chemical can be determined. Combining this reaction
quantum yield with the test chemical molar ahsorptivity data fr1!m
the tier—one test and the appropriate solar irradiance (Iatd in
water, environmentally relevant rate constants (i •e•, ()r W [ It4 r
bodies under clear sky conditions) can be calculated as a
function of latitude and season of the year. These rate
—8—
_I_ ‘ )
-------
CS—6000 (October, 1984)
constants can then be used to determine the corresponding half—
lives for photolysis of a test chemical in sunlight in aqueous
media. These rate constants are in a form suitable for
preliminary mathematical modeling for environmental fate of a
test chemical.
B. Develo _ pment of the Test Method
1. Theoretical Aspects
The theory of the method of Zepp and dine (1977) is briefly
discussed to show that the proposed test method is based upon the
fundamental criteria given in Section II.A.2. These discussions
lay the foundation for the proposed method, show how the method
can be used to obtain direct sunlight photolysis rate constants
and half—lives, and indicate what test conditions must be
standardized in order to obtain meaningful aqueous photolysis
rate data.,.
Zepp and dine (1977) published a paper on the rates of
direct photolysis in aquatic environments. The rates of all
photochemical processes in a water body are affected by solar
spectral irradiance at the water surface, radiative transfer from
air to water, and the transmission of sunlight in the water
body. It has been shown that in dilute solution (i.e., the
absorbance of a chemical is less than 0.02 in the reaction cell.
at all wavelengths greater than 290 nm), the kinetic expression
for direct photolysis of a chemical at a molar concentration C is
107
-------
CS—6000 (October, 1984)
dt Ea — pE
kpE = Eka (2)
where is the reaction quantum yield of the chemical in dilute
solution and is independent of the wavelength, and ka= kaxsthe
sum of ka values of all wavelengths of sunlight that are absorbed
by the chemical (i.e., the light absorption rate constant). The
term kpE r epresents the photolysis rate constant for water bodies
in sunlight in the units of reciprocal time. Integrating
equation 1 yields
ln(C 0 /Ct) = kEt , (3)
where is the molar concentration of chemical at time t during
photolysis and C 0 is the initial molar concentration. By
measuring the concentration of chemical as a function of the time
t during photolysis in sunlight, kpE can be determined using
equation 3. In addition, equation (3) can be solved for the
condition C = C 0 /2 and the half—life of the chemical is given by
0.693
tl/ = (4)
Ir.Q
—10—
-------
CS—6000 (Oct Obt r , l 4
Furthermore, under the same conditions cited dh(we i.e.,
*
for homogeneous chemical solution with absorbance less than 0.02
in a reaction cell at all wavelengths greater than 290 nm and at
shallow depths (less than 0.5 m)], the first—order direct
photolysis rate constant, kpE is
= (5)
where is the reaction quantum yield which is independent f
the wavelength, €1 is the molar ahsorptivity in the units inl r 1
cm , and L is the solar irradiance in water in the unit.s 10 3
einsteins cm 2 day 1 [ Mill et al. (1981, 1982a, 1982b)]. is
the solar irradiance at shallow depths for a water body under
clear sky conditions and is a function of latitude and season of
the year.
The method of zepp and dine (1977) and the method of Mill
et al. (1982a) are applicable to sunlight incident on a water
surface such as a natural water body. However, the method
developed in this guideline measures rate enstants in t uht’ s
(e.g., 13 x 100 mm) and the rate is {aster in Lub s. This is
discussed in more detail in Section II.B.3.h.iii. Thus,
equations 3 and 4 have to be modified to take this into
*At an absorbance of 0.05, equation (5) is in error by only
11%. This is an acceptable error limit for a screening test.
and an absorbance of 0.05 has been adopted for use in this
Test Guideline.
.4 ,‘(
—11--
-------
CS—6000 (October, 1984)
account • For s tmpl icily, the to I lowi u i nienc I • t tOt’ I ;
For water bodies, the rate cot l.int i ; ‘ie ; i.qnat ( (i a ; wit Ii I lit’
subscript E designating rates in the environment in water
bodies. For tubes, the rate constant is designated as k . The
corresponding half—lives for water bodies and tubes are
and ti, respectively. Thus, for tubes, equations 3 and 4 can
be written as
ln(C 0 /Ct) k t (6)
0.693
t 1 = ____— (7)
p
2. Tier 1 Test: IN—Visible Absorption Spectra —
Estimation of Aqueous Photolysis Maximum Rate
Constant and Minimum Half—life in Sunlight
A simple first—tier screening test has been developed using
equation 5. As an approximation, it is assumed that the reaction
quantum yield +E is equal to one, the maximum value*. As a
esult, the upper limit for the direct photolysis sunlight rate
constant in aqueous solution is obtained and equation 5 ht t ’
*
t is possible that under certain circumstances the reaction
quantum yield can be greater than 1. For example, this
happens when a chain reaction occurs after the absorption of a
.quantum of light. However, this rarely occurs in dilute
aqueous solution.
—12—
-------
CS—6000 (f ctober , i 4
(k ) L ( fl
pEmax. XX
Substituting equation 8 in equation 4, the lower limit for the
half—life is then given by
0.693 0.693
(t 1 / ). = (k T = cxLx (9)
In order to estimate (kpE)max• it is necessary to obtain
data on the molar absorptivities of the test chemical and the
solar irradiance (equation 8). These two parameters can he
obtained in the following way.
a. Determination of Molar Absorptivity
The molar absorptivity of a solute (test chemical) can be
obtained from the Beer—Lambert law and the ultraviolet-visible
absorption spectrum. The Beer—Lambert law states that for an
absorbing solute present in water, the decrease in intensity of
light with thickness ( dI/dt) , at a fixed wavelength X, is
proportional to the intensity I and the concente rat ion of the
solute C in the medium iPrutton and Maron (1951)1. Thur tor ,
= c CI (10)
where is the proportionality constant. Equation 10 can he
easily integrated to give the standard Beer—Lambert equation
-13- Li
-------
( S—hOO(1 (.‘)ct r , 1
log 10 (t 0 /I) = (ii )
where Ł is the thickness of the solution or pathlength of the
cell in cm, i is the intensity of light at Ł. = 0, I is the
intensity of light at the thickness&, C is the molar
concentration, and is the molar absorptivity in the units
n Diar 1 cm 1 .
Since the absorbance, is equal to log 10 (t 0 /fl
equation 11 becomes
= c Ct (12)
The absorbance can be obtained from the ultraviolet—visible
absorption spectrum of the test chemical in aqueous solution by
the methods outlined in Test Guideline CG—l050. These data will
he used in equation 12 to calculate molar absorptivities tor th
appropriate wavelength intervals and wavelength centers where tht’
test chemical absorbs light. These wavelength intervals and
wavelength centers are listed in Table 1 of the Test ( uide1inc’
(CG—6000). The method outlined in the Test Guideline CG—1050 ‘;
the standard procedure used by all laboratories for determinin1j
the molar absorptivity of a chemical in solution.
A. .‘s
—14—
-------
CS—6000 (Octobt ’r, l 4
b. Solar Irradiance
The solar irradiance in water, is related to the
sunlight intensity in water and is proportional to the average
light flux that is available to cause photoreaction in the
wavelength interval centered at A over a 24—hour day. L is the
solar irradiance at shallow depths for a water body under clear
sky conditions and is a function of latitude and season of the
year. The solar irradiance is discussed in more detail in
Section II.C.9. and the values are listed in Tables 3—6 in the
Test Guideline (CG—6000) as a function of latitude and season of
the year.
C. Estimation of the Maximum Direct
Photolysis Rate Constant and the
Corresponding Minimum Half—Life
The maximum direct photolysis rate constant is defined by
equation 8. The L values are chosen from Tables 3—6 in the Test
Guideline (CG—6000) for the appropriate latitude and season of
the year. The molar absorptivity is determined according to
Section II.B.2.a. The product of and L is ca 1culated
at Aceriter for each wavelength interval where has a non—zero
value. Then the products of CALA are summed over all wavelength
intervals for Acenter > 299 nm. This sum is and is equal
to (kpE)max• (kpE)max is substituted into equation 9 U) ()ht1 iin
the minimum half—life.
—15—
-------
CS—6000 (October, 1984)
3. Tier 2 Test: Aqueous Photolysis in Sunlight
a. Phase 1 Test
The phase one aqueous photolysis test is based on the method
developed by the U.S. Environmental Protection Agency, Office of
Toxic Substance [ USEPA (1979)] using equations 6 and 7. By
ni asuring the concentration of chemical as a function of time t
during photolysis in sunlight, can he determined. This rate
constant is valid for the latitude and season of the year in
which the photolysis experiments were carried out. The half—life
can then he calculated using k in equation 7. This method only
gives an approximate rate constant since it fails to n easure
sunlight intensities incident on the sample during photolysis.
This data will be used in the phase two aqueous photolysis
screening test.
h. Phase 2 Test
The phase two aqueous photolysis sunlight photolysis test
method is based on the research carried out at SRI International
[ Mill et al. (1981, 1982b) and tXilin and Mill (1982)].
i. Introduction
The phase one test method has a limitation since it does r))t
measure sunlight intensities incident on the sample during
photolysis. SRI carried out a study on the use of i.nsolat ion
-16- 1 1
-------
CS—(i )0d ( ok.. I , I ) 04
methods for measuring solar intensity (e.g., radiometry,
photometry, and actinometry) . The most suitable method for
measuring sunlight intensity was found to be actinometry.
Radiometry is used to measure total solar flux, hut because
radiometry involves energy rather than photon flux, t is of
1. imited value for estimating photoselective photocheuuciI
reactions. Photometers, based on photocells and
photoinultipliers, have several advantages over radiometers, the
most important being that they are sensitive to photon flux
rather than energy flux. However, sensitive photometers, using
digital pulse—counting methods, are relatively expensive and
fragile. Chemical actinometry offers advantages over
instrumental methods. Actinornetry conforms to the geometry of
the reaction vessel and measures the actinic flux under known
sensitivity conditions. Thus, in general, actinoinetry, to
measure solar intensity, is simple to use in aqueous photolysis
experiments. The main complication in using actinometry in
outdoor experiments is that the sun is not a monochromatic
source, and the absorption of light by the actinoineter and
chemical rarely match. Therefore, a series of actinometecs would
be needed to match the light absorption of the test chemical in
the wavelength region A 290 nm. Fortunately, the most
important wavelength region for sunlight photolysis is at 290—400
run, where the high energy sunlight is concentrated, rind a I
number of chemicals absorb light in this region. Thereior , t
most useful actinometer is one which absorbs light in the regi ri
290—400 rim.
- 1—
i ’)
—17—
-------
CS 6000 C t h’ r 4
ii . F)eve 1 opment. ut a Sun light. i\c t i noinutc r
Listed below are several criteria for chemical and spectral
properties which should be given high priority in selecting
sunlight actinorneters. The ideal actinometer should:
(1) absorb light moderately and uniformly in the solar
region but be stable in room light;
(2) have a reaction quantum yield, , which is wavelcnqth
independent;
(3) have a reaction quantum yield that is indepu icut ut
dissolved oxygen and concentration of the actinometer;
(4) have an adjustable half—life ranging from several hours
to 90 days;
(5) have photolysis products that do not interfere with
photolysis, either chemically or spectrally;
(6) have simple, accurate, analytical methods available for
its analysis;
(7) he very stable in water in the dark;
(B) he moderately soluble in water and relatively
nonvolatile; and
(9) have little or no temperature dependence n
Based on a review of the photochemical literature, several
hundred examples of photochemical reactions were found, many of
which might fulfill some of the criteria established for the
ideal actinometer. To simplify the task, the review w i
—18—
-------
CS—b lU i c t ‘ r , 1 $
restricted to photoreactions that are bimolecular and are
therefore amenable to changes in the rate (i.e., to changes in
the value of ) with changes in concentration of one photoinert
reactant. Nucleophilic substitutions constituted one such class.
The most successful actinometer developed to date with a
variable reaction quantum yield is the system composed of
p—nitroacetophenone (PNAP) and pyridine (PYR) [ Mill et al. (l 8T,
1982b) and Dulin and Mill (1982H . Laboratory experiments at U
nrn and 25°C indicated that; the reaction quantum yield [ or
transformation of PNAP depended linearly on the molar
concentration of pyridine, at a fixed concentration of PNAP.
Regression analysis of the reaction quantum yield as a function
of the molar concentration of pyridine [ PYR] up to a
concentration of 0.2 M (at a fixed concentration of PNAP equal to
1.00 x l0 M) gave the relationship
= O.O 64(PYR] (1 l
with a correlation coefficient greater than 0.99. rt should be
noted that in each experiment, PNAP exhibited good iirst—ord& r
kinetics to 75% transformation. At pyridine concentrations
greater than 0.2 M [ Winterle (1984)], nonlinearity was
observed. Hence, equation 13 is only valid to a 0.2 M
concentration of pyridine. Further laboratory experiments were
carried out at 366 nm and 25°C. The data indicat: d that wa ;
essentially the same, within experimental error, at 311 urn arid
366 nm.
—19—
-------
CS—6000 (October, 1994)
To determine the temperature dependence of the PNAP/PYR
actinometer, photolysis rates were measured at 43°C and then
compared to the rates at 25°C. The rate of photolysis slowed
down by 12% at 43°C. Using the standard ArrhenIus equation, the
energy of activation was found to be —1.3 kcalories. Therefore,
in the range 1°C to 45°C, the photolysis rate constant of the
PNAP/PYR actinometer will be within ± 25% of its value at 25°C.
Thus, the rate constant for this actinometér, and hence its
reaction quantum yield, is not very sensitive to changes in
temperature.
The ultraviolet—visible absorption spectrum of PNAP was
oheained in an aqueous solution containing 1% acetonitrile (Mill
et al. (1982b)]. The molar absorptivi.ties were calculated using
the Beer—Lambert law as a function of wavelength (i.e. at the
wavelength centers corresponding to the interval ranges listed in
Table 1 of the Test Guideline) and the results are summarized in
Table 1. PNAP shows absorption of light in the solar region
A 290 nm to 410 nm with strong absorption in the region around
300 nm. As discussed in Section IIB.3.b.i., this actinometer is
extremely useful since a large number of chemicals absorb light
in this region.
An important variable for the PNAP/PYR actinometer is the
absorption rate constant k, in days , and is defined by the
equation
—20—
-------
CS—6000 (October, 1984)
TABLE 1. MOLAR ABSORPTIV [ TIES OF PNP P
A center (nm)
297.5 3790
300.0 3380
302.5 3070
305.0 2820
307.5 2590
310.0 2380
312.5 2180
315.0 1980
317.5 1790
320.0 1610
323.1 1380
330.0 959
340.0 561
350.0 357
360.0 230
370.0 140
380.0 81
390.0 45
400.0 23
410.0 (1
corresponds to the molar absorptivity at Acenter
aôeraged over the interval range listed in Table 1 of the
Test Guideline CG—6000.
—21—
-------
CS—hOt) l (Oct l)or , I ‘i 4
= (13
where is the molar absorptivity of PN I and is the solar
irradiance in water. The molar absorptivities of PMAP are listed
in Table 1. USing thiS data with the LA values listed in Tables
3—6 of the Test Guidelitie (CG—6000), has been calculated as a
function of latitude and Season of the year in the northern
hemisphere. The results are summarized in Table 2 of the Test
Guideline. This parameter is essential to carrying out the phase
two aqueous photolysis experiments.
iii. Geometry of the Reaction Vessel
The method of Zepp and dine (1977) and the method of Mill
et al. (1982a) are applicable to sunlight incident on a water
surface such as a natural water body. However, the method
developed in this Test Guideline measures rate constants (kg) fl
tubes (e.g., 13 x 100 rmn). Such a reaction vessel receives
radiation from all directions, due to scattering of the Iiqht )It
the walls, in contrast to a water hotly (e.g., a lake) whieh
receives sunlight only from above. IL is expected, t r 1’)r
that photoreactions would be faster in tubes. This w;is v’ril
experimentally [ Mill et al. (1982b)] . Laboratory experirnent
were carried out in tubes and dishes using the actinometers
p—nitroacetophenone/pyridine, p—nitroanisole/pyridine, and rose
bengal. The experiments using dishes represented a water body.
All the data indicated that the ratio of rate constants of tubes
to dishes was 2.2 0.3. Thus,
—22—
-------
CS—600() ( c t ot)& r , 1 ‘ •1 )
= y kpE (15)
with y = 2.2 ± 0.3. The factor y is catlecl the geometry factor.
For tubes, therefore, y = 2.2 while for water bodies y = 1.0.
iv. Phase 2 Test Method
In phase two, a standard PNAP/PYR actinometer is used to
measure sunlight intensities incident on the sample during
photolysis. The rate constant for this actinometer, k , can be
adjusted to match the approximate rate constant of the test
chemical by adjusting the concentration of pyridirte. Since the
rate constant is a function of the reaction quantum yield of the
actinorneter, the rate constant can be adjusted according to
equation 13.
Using equations S and 15, the rate constant in tubes is
given by
= 1 E CALX (16)
Therefore, the rate constants for the actinometer, k , and
the test chemical, k , in tubes are
k M7)
-23- 1 i
-------
CS—60t)() ( ) t ( bt r , 4
= ( 18)
Since the phase two test method involves the simultaneous
photolysis of test chemical and actinometer in identical tubes,
equation 18 can be divided into equation 17. Carrying out this
operation and solving for q yields
c a
k cL
k ) €xLx
The reaction quantum yield of the test. chemical, , can he
determined in the loUnwing way.
(1) Determination of the ratio (k /k )
Using equation 6, the equations for the test chemical and
actinometer in tubes are
1n(C 0 /C )c= k t (20)
ln(C 0 /C )a = (21)
Since the phase two test method involves the simultaneous
photolysis of the test chemical and actinorneter in identical
tubes, equation 20 can be divided by equation 21. Carrying out
this operation and rearranging yields
4 “•‘•)
Ls it..’
-24—
-------
(‘S—hL)OO (t’ ct t’ lt’t , I l 4
lrl(Co/Ct)c = (k /ka) ln(C 0 /CtY (22)
ln(Co/Ct)c is a linear function of 1n(C 0 /C )a with a slope given
by
slope = (k/k ) (23)
The most precise and easiest method of determining the slope is
by linear regression analysis on the experimental data.
Therefore, by measuring the concentration ot test chemical and
actinometer at the same time as a function of time, these data
can be fitted to equation 22 by linear regression analysis with
ln(C 0 /C )a as the independent variable and ln(Co/Ct)c as the
dependent variable to give the best straight line. The slope,
the ratio of the rate constants, is given by equation 23.
(2) The term for the actinometer has been tabulated as a
function of latitude and season of the yeat [ Table 2 in the
Test Guideline].
(3) The term for the Test Guideline can he obtained from
the experimentally measured molar absorptivities (Section
II.B.2) and the values of [ listed in Tables 3 to 6 in
the Test Guideline as a function of latitude and season of
the year]
1 ’
I ’
—25—
-------
CS—iS000 (October, l9 4)
4 ) can he determined f rorn eq 1it ion 1 it t h ’ in 1 . r
concentration of pyrid inc used in the standard act i neter
With the value of 4, , c , and the appropriate L values,k
for the test chemical can be calculated as a function of latitude
and season of the year using equation 5. The corresponding half—
life can be calculated using kpE in equation 4. These rate
constants and half—lives correspond to sunlight photolysis in
water bodies under clear sky conditions and at shallow depths.
Photolysis experiments were carried out in the ahoratory
using a photochemical “merry—go—round” reactor (PMGRR) at 31 1 nni
and in sunlight using the test chemicals dihenzothiophene,
nitrobenzene, and p—methoxyacetophenone [ Mill et al. (1982h)1.
These data were used to calculate the laboratory reaction quantum
yield at 313 nm [ (313 nm)], the sunlight reaction quantum yield
the half—life calculated from the laboratory data
1t 11 (calc.)1, and the half—life in sunlight [ tI/ (obs.)]. These
data are summarized in Table 2. Comparison of the corresponding
quantum yield data and half—life data for the laboratory
experiments with those in sunlight indicated thai the resti It ; •t re
in good agreement confirming the phase two lest method.
—26—
-------
C —t’WtI ( c t c bt’ r , 1 t 4
TABLE 2. CALCULATED AND OBSERVED QUANTUM YIELDS ANt)
ENVIRONMENTAL PHOTOLYSIS HALF—LIVES FOR SOME TEST CHEMICALS
ty (calc.) tj (obs.)
Chemical E days days
Dibenzothiophene 5.0 x io— 7.3 x i0 9.1 6.4
Njtrobenzene 1.8 x i0 1.9 x iO— 4 19.8 19.3
P-Nethoxyaceto—
phenone 9.5 x 1O 13 x io— 11.9 12.0
4. Summary
This screening test has been designed to determine the molar
absorptivity of a test chemical, and its reaction quantum
yield, 4 . These parameters can be used to determine
environmentally relevant rate constants at low absorbance and
shallow depths in pure water as a function of latitude and season
of the year. Tables of solar irradiance (Tables 3—6) have been
included in this Test Guideline to carry out all the
calculations. However, the method is really very general and can
be extended to determine the rates of photolysis over a range of
other environmental conditions using a computer program. Zepp
and Cline (1977) have written a computer program to calculate the
rates of photolysis as a function of depth in water, in natural
waters as a function of the attenuation coefficient of the
water (cz ) the average ozone layer thickness that pertains to
the season and location of interest, and as a function of
latitude and season of the year. This program ha’; beeii r cent.I y
-27-
-------
CS—t-t10(l ( ( C t ot ’ r , 1
updated with the best available solar irradiance data and is
called the GC SOLAR computer program. The GC SOLAR computer
program is available on request. ER. Zepp, Environmental
Research Laboratory, U.S. Environmental Protection Agency,
College Station Road, Athens, Georgia 30601.1
This Test Guideline represents a screening test to obtain
the reaction quantum yield of a chemical. Since it does not
require sophisticated (and expensive) photochemical equipment,
the reaction quantum yield can be obtained easily and at a modest
cost. However, the reaction quantum yield can he obtained more
precisely in the laboratory using special photochemical equipment
and monochromatic light by procedures described by Zepp (1978)
and Mill et al. (1982a). A reaction quantum yield obtained by
one of these procedures is acceptable as a replacement for the
determination of the reaction quantum yield in sunlight... An
upper—tier test has been published as a Test Guideline in NTIS
(CG—6010) which gives detailed procedures for determining the
reaction quantum yield in the laboratory using a monochromatic
light source.
C. Applicability and Specificity
This test method is applicable to all chemicals which have
uv—visible absorptions in the range 290—800 nm. Solar radiation
reaching the earth’s surface has a sharp cutoff at.. a wavelength
of approximately 290 nm due to the absorption by ozone
1 ‘‘4”
i .ss.)
-------
CS—6000 (October, 1984)
(1961), Peterson (1976), Zepp and Cline (1977), nemerjian
(1980)] . The long wavelength limit is set by thermochemistry
since light of wavelength greater than 800 nm is not of
sufficient energy to break chemical bonds of ground state
molecules [ Calvert and Pitts (1966), Benson (1976)]. Photolysis
does not occur unless there is absorption of radiant energy.
This is the direct consequence of the Grotthus—traper law, the
first law of photochemistry. If a chemical in aqueous solution
only absorbs light at wavelengths below 290 nm it will not
undergo direct photolysis in sunlight. A few examples of
chemicals that only absorb light below 290 run and therefore
should not be tested by this Test Guideline are alkanes, alkenes,
alkynes, saturated alcohols, and saturated acids. It is possible
that some chemicals will absorb radiation mainly at wavelengths
below 290 nm but may have an absorption tail that extends above
290 nm. Photolysis experiments should he carried out for these
chemicals.
This Test Guideline is only applicable to pure chemicals and
not to the technical grade. Overestimates of molar
absorptivities usually occur when technical substances are tested
because the impurities frequently absorb in the same spectral
region as the pure chemical.
4
I A
—29—
-------
CS O0O (October, l984
D. Rat ionale for__the Selecr ion ti;t
1. Special Laboratory Equipment
All the laboratory equipment needed to determine the uv—
visible absorption spectrwn of a test chemical in aqueous
solution is listed in the references in Test Guideline CG—1050.
The equipment required to carry out this test is standard
equipment which is commonly used in most laboratories.
In order to carry out the photolysis experiments, special.
tubes are necesary to contain the reaction solutions durinq
photolysis. Reaction tubes with 11 mm inside diameter are
recommended as they are inexpensive, easy to seal to prevent
volatilization of the chemical, and can be easily mounted on a
rack in sunlight. Disposable culture tubes (13 x 100 mm) with
Tef lon-line screw caps are readily available from commercial
labpratory supply companies. For some chemicals, it may be
difficult to determine their concentrations in the small volumes
present in 11 nun i.d. tubes. For such chemicals, the use o
large reaction vessels is permissible as lonq as the pathlenqtli
is less than 0.5 meters. This is necessary in order that. Ii rst
order kinetics are obeyed (Section II.B.l) . Furthermore, the
cell walls should be relatively thin to minimize the loss of
sunlight through the cell walls. Reaction tubes of either quartz
or thin—walled borosilicate glass may be used. If a chemical
absorbs strongly at A 340 run, then quartz tubes should he
used. iartz will transmit 100% of the sunlight in the region
—30-
-------
CS—6000 (Octobec, 1984)
290—340 nm while boros iii cate q 1 a ;s w i II bsorb ome ot t hi -
light • The ahsorpt ion spec I run ot the cheni i i n
sol. Ut ion , as measured by Test Gu ide Ii ne (‘G— It) ‘ () , in be used t o
determine which type of glass to use.
It is extremely important that all reaction tubes be capable
of being sealed (without the use of grease) and be filled as
cc npletely as possible to prevent volatilization of test chemical
or water (see Section II.C.6). Screw caps lined with Teflon
inserts are recommended. Grease should be avoided since
hydrophobic chemicals might adsorb it. Volatile compounds can be
conveniently studied in culture tubes equipped with gas—tight
Mininert valves. Samples can he introduced into or removed from
the tubes through the septum in these valves with no loss of
substrate [ Zepp [ 1984)1. As an alternative, the tubes can be
sealed with a torch.
2. Purity of Water
Pure water is used because dissolved impurities could
sensitize or otherwise affect the rate of photolysis. In
addition, the water needs to he sterile because hact:eri may
consume or alter the chemical substance durirvj Ihe J)rolon(Jed
periods of testing which may occur in the coursu of a raI’
determination. Thus, pure water [ e.g., ASTM Type If A (ASTM
1979)], or an equivalent grade, is recommended in this Test
Guideline. Furthermore, it is important that the water he
1 9
-------
CS—600() (October, 1984)
saturated with air prior to preparation of the te t • u 1
solutions to simulate env ironenent al .i edi t ions. Ai r—sat ue t ‘ i
water can he easily prepared by a I low i n j he w ei t o ii i t’r
in a vessel plugged with sterile cotLon.
3. Sterilization
Sterilization is necessary to kill the bacteria and
therefore eliminate or minimize biodegradation which could
interfere with the photolysis rate determination. The presence
of bacteria in either the test solutions or controls may cause
biodegradation of the test substance, especially when the
photolysis experiments are carried out over a long period ol
time. This will intrcxiuce an error in the concentration of the
test chemical. Thus, it is extremely important to use aseptic
con ditions in carrying out all photolysis experiments to minimize
biodegradation. Glassware can be sterilized easily in an
autoclave or by use of any other suitable non—chemical method.
4. Concentration of Solution
It is extremely important that solutions of chemical
substances used in this Test Guideline he prepared .et low
concentrations in order to approximate environmental condit:iorc;
and to allow first—order kinetics assumptions to apply (see
Section 11.8.1.). Furthermore, an absorhance of 0.05 has been
adopted in this Test Guideline since solutions with this
—32—
-------
CS—6000 (C ctoher, 1984)
absorbance can ho easily ach it ved in t h’ I l r i( ry . Tb i
absorhance 1 imlt yields an l err r , an tbpt able’ ‘tFt t t ‘t
screening test ($ oction It.} .I).
If the chemical substance is too difficult to dissolve in
pure form to permit reasonable handling and analytical
procedures, then a test solution may be prepared by first
dissolving a chemical in reagent grade acetonitrile. The final
acetonitrile concentration in the test solution should be no more
than one volume percent in order to avoid solvent effects (Smith
et al. 1977, 1978). Acetonitrile was chosen as a solvent as it
is soluble in water, is non—polar and thus effective in
dissolving many substances which are insoluble in water, it does
not absorb radiation over the wavelength range of 2’ O to 800 nm,
and it causes minimal solvent effects (i.e., minimum shifts in
bands and changes in absorbance) for test substances.
5. pH Effects
The molecular structure of a chemical substance which
ionizes or protonates is a function of the pH. As a result, the
absorption spectrum and consequently the rate of photolysis may
change with pH. The recommended procedure is t determi n” I 1i•
uv-visible spectra under conditions in which only one
strongly predominates in solution. This can he accompIi ihud by
preparing buffered solutions containing the test chemical at pHi;
at least 2 units above the PKa and at least 2 units below the
1’i-
-------
:; —t t t t t , I ‘ 1
PKa. For example, the 1 a for phenol is 9.8g. i\t p i below
7.89, the protonated form of the phenol strongly predomirtates t
a pH of greater than 11.89, the phenoxide form of the phenol
strongly predominates. The reaction quantum yield for ionizable
compounds should also be determined in buffered solutions, where
only one species strongly predominates. Thus, photolysis
experiments should be carried out at the same pt-Is used to obtain
the spectrum of each species.
Since buffers coul(I influence the rate of photiolysis, the
recommended buffers for use in this Test Guideline were carefully
chosen to be transparent to radiation between 290 and 800 nm and
should be kept at very low concentrations to avoid buffer effects
which may cause transformation of the substance by, for example,
catalysis.
6. Control Solution
Undetected loss of a test substance Lb rough vu lut. i Ii za t i <)tI
hydrolysis, or other processes during the course of the
photolysis experiment will result in the determination of
erroneously large rate constants for aqueous photolysis.
Therefore, for volatile chemical substances, it is important that:
the reaction vessels and control vessels be filled as completely
as possible and sealed in order to avoid evaporative losses. To
correct for possible losses, control solutions of test substance,
in darkened vessels, are placed side—by--side wi Lb the J)hot ul y ; i
-34- 4.”)
I S
-------
-- & c ( I 4
vessels and the contents of the control vessels are analyzed at
prescribed times during the experiment and at the end of the
experiment. In this way, the loss of test chemical for processes
other than photolysis may be detemined and eliminated or
accounted for in determining k . For simplicity, if the loss of
chemical in the control is small (i.e., approximately 10% or
less), one can calculate a first—order loss, k 1055 , and subtract
It from (kp)ohq. to give the corrected di rect: photolysis rate
constant k . If hydrolysis is found to he signilicant (i.e.,
greater than 10%), hydrolysis studies should be carried out first
to define precisely the kinetics of this process [ Test Guideline
CG—50001
7. Absorption Spectrum as a Criterion for
Performing the Aqueous Photolysis Test
The test method is applicable to all chemicals that have uv-
visible absorptions in the range 290—800 nm (section ET.C).
Thus, the uv—visible absorption spectrum ofa chemical in aqueous
solution will give a good indication of whether it would be
useful to carry out sunlight aqueous photolysis tests.
8. sunlight Actinometer
In order to quantify the rate of photolysis more precisely,
it is necessary to measure the sunlight intensity incident on the
sample during photolysis. A standard p—nitroacetophenonu—
-35-
-------
(‘S—6OOtl C Oct c’t ’ r , I 1 I .1
pyridine act inotneter (PNAP/PYR) has been deve ecI in i t-he
details have been discussed in Section II.B.3.h.ii. Accordini t
equation 5, the rate constant is a function of the reaction
quantum yield. Furthermore, the reaction quantum yield can be
adjusted by varying the molar concentration of pyridine according
to equation 13. Hence, by varying the pyridine concentration,
the PNAP photolysis rate constant can be adjusted so that the
half—life can range from several hours to several weeks. Tn
order to make the actinoineter operable, it is necessary that.
[ PYR] 10 EPNAP] and < 0.2 M. Therefore, the concentration ot
PNAP must be approximately 1 x 10 M. In the recommended
actirtometer formulation, the concentration of PNAP is set at
1.00 x M. The concentration required to adjust the rate of
the actinometer to equal the rate of the test chemical can be
derived as follows. Since it is necessary to adjust the rate of
photolysis of the chemical to that of the test chemical, then
k = ka (fl)
According to equation 15, the rate constant for the actinometer
in tubes is
= 2.2k F (24)
Hdwevet, k E for the actinometer can be obtained from equation
5,
—36— 1:4.1
-------
CS—6O () (Uctober, H84)
= 4 c Lx (5)
and since for the actinometer is
4 k (25)
Substitution of equation 25 into equation 24 yields
= 2.2$ k (26)
According to equation 13, = O.0169(PYR). Substituting this
result in equation 26 yields
ka = O.0372 [ pYR] (27)
Finally, substituting equation 27 into equation 23 and so1 vinq
for EPYR1 yields
[ PYR] = 26.9 (k /k ) . (2R)
The term is the sunlight rate constant for the test chemical
in tubes and the variable k(= c Lx) for the actinometer can he
obtained from Table 2 of the Test Guideline.
For 1 liter of solution, the mass of pyridine required to
adjust the rate of the actinometer is
4 — .r---
37
-------
CS—6000 ( ctoher, 1984)
tn.a s(qrains) t.W. ( L YR
where F.W. = 79.10 g/mole, the formula weight of pyridine. The
density of pyridine is 0.9819 g/mL at 20°C [ Weast (1973—1974)1.
Thus, the volume of pyridine at 20°C required to equal this mass
is
V(mL) = 80.6 EPYR] (30)
9. Solar Irradiance Data
In order to calculate the reaction quantum yield of the test
chemical, • , and then calculate k E and tj , it is necessary
to use solar irradiance data L . The theoretical basis for
the data is discussed in the following paragraphs.
In a water body exposed to the sun, the intensity of light
passing into a water body is composed of light directly from the
SUfl (I ) and the sunlight from the sky (I ). Leighton (1961)
discussed the calculation and contribution of direct and sky
light to the total intensity of sunlight fitterinij tiuwn lhr ujh
the troposphere at a point on the earth’s surface as a funct.ton
of the zenith angle Z, Figure 1. Zepp and Cline (1977) exlend ’d
the treatment to the behavior of and I that enter a waler
body, Figure 1. Part of the theoretical treatment has been
summarized in Section II.B.l and the following discussion extends
this treatment. According to equation 2
4 ‘‘4’
)
—38—
-------
CS—6000 (October, 19 14
kpE = Eka = (2
Furthermore, Zepp and Cline showed that
kx 2.303 cxj’Zx (31)
where 2 A = IdASeCO + 1.2 l (32)
In equation 31, the term is the molar absorptivity of a
chemical in aqueous solution, j is a constant which converts the
intensity units into units that are compatible with the
concentration units of C for the chemical (i.e., j = 6.02 x io20
when C is the molar concentration of the chemical). In equation
32, ‘dx and I are the direct and sky sunlight intensity in
photons cm 2 sec. , and U is defined by the refractive index of
water, Figure 1. is the actinic irradiance in water at
shallow depths in the units of photons cm 2 sec. . Zepp and
Cline (1977) tabulated values for 40°N latitude as a function
of midseason dates for spring, summer, winter, and fall.
Substitution of equation 31 into equation 2 yields
kE = 2.303f’BE cAz x (33)
—39—
-------
cS—( OOO (act o ’t r, 1 -’ 3
(o 4r)
Z Angle of incidence (solar zenith angle in case of direct sunlight)
0 — Angle of refraction
n — Refractive index of water
FIGURE 1 REFLECTION AND REFRACTION OF A L;GHT BEAM PASSING
FROM ATMOSPHERE INTO A WATER aoov
(From Zepp and Cline. 1971)
/
/
/
I
1’
I
sin Z
sIn 0
I
—4 n—
-------
CS—6000 (October, 1984)
Based on their theoretical work, 7.epp ctn&1 (‘1 t 1k’ tLtV ’ Wr t t t t’I
a computer program for 39 speciNc wavelength intervals from
296.3 to 825nm which calculates instantaneous rate constants at
specific times of day for a given latitude and midseason date and
summed them up to obtain kpE. This program has been recently
updated with the best available solar irradiance date and is
called the GG SOLAR computer program [ Zepp (1984)1. The term
midseason date used by Zepp and dine does not refer to a day
midway in the season, but rather to a day during the season when
the angle of declination of the sun is at specific values (listed
in a subsequent paragraph) . The program further calculates a
photolysis rate constant kpEI which is averaged over a 24—hour
calendar day in the units of days . Required data inputs are:
(1) molar absorptivities of the chemical at wavelengths greater
than 297.5 nm; (2) the refractive index of the medium (e.g., for
water n = 1.34); (3) the attenuation coefficient, a , of the
medium (e.g., aX for distilled water or for natural water
bodies); (4) the reaction quantum yield, of the chemical;
(5) the solar declination, solar right ascension, and siderial
time for the date of interest obtained from the Astronomical
Almanac fAA (1982)1; (6) the latitude and longitude; dfld (7) tlw
average ozone layer thickness that pertains to the season and
location of interest.
For routine computations, the program calculates rate
constants and half—lives for midseason dates corresponding
to solar declinations of +10° for spring, +200 for summer, —10°
—41—
-------
CS—&000 (October, 1984)
for fall, and 2O° for winter, as a functAon ot lcc il time c t
day, long itude, and latitude for 00 t- qO°N in 1L)’ increment
To obtain the L x data, Mill et al.. (l982a), used the Zepp
and’ dine computer program with = 1, = 1, and the absorption
coe fficients for pure water. The Lx values were obtained
for the midseasons of winter, spring, summer, and fall for the
latitudes 200 to 50°N and these data are listed in Tables 3 to 6
of the Test Guideline. The L 1 values correspond to the 1982
mid eason dates of January 21, April 16, July 24, and October 20
[ AA: (1982)1. The values of Lx are in the units of
einsteins cm 2 day . The values are applicable to clear sky
conditions, water bodies, shallow depths, and for chemicals with
very weak absorbance in pure water. The Zepp and dine computer
program has been recently updated and is called the GC SOLAR
program [ Zepp (1984)]. Miii et al. (1984) used this program
with = 1., • = 1, and the absorption coefficient ax for pure
water to obtain updated L values as a function of latitude and
season of the year. These Lx values are summarized in Tables 3—
6 of the Test Guideline.
The solar irradiance data pertain to midday, midse. son,
values in the units of photons cm 2 sec 1 while the solar
irradiance L data represent day—averaged, rnidseason, values i
the units of i0 einsteins cnr 2 day . The L data are used in
equation 5 to calculate day—averaged, midseason, direct
photolysis rate constants. The data are used in equation 33
to calculate midday, midseason, direct photolysis rate constants.
—42—
-------
CS—6000 (October, 1984)
10. Geometry of the Reaction Vessel
The method of Zepp and dine (1977) and the method of MiU
et al. (1982a) are applicable to sunlight incident on a water
surface such as a natural water body. However, the method
developed in this Test Guideline measures rate constants in tubes
(e.g., 13 x 100 mm). Such a reaction vessel receives radiation
from all directions, due to scattering of the light off the
walls, in contrast to a water body such as a lake which receives
sunlight only from above. It is expected, therefore, that
photoreactions would be faster in tubes. This was verified
experimentally [ Mill et al. (1982b)] and this has been discussed
in detail in Sections II.B.3.b.iii. For these experiments, tubes
are the simplest and easiest reaction vessels to use since they
can be sealed easily to minimize or eliminate volatilization and
can be handled easily. Hence, this Test Guideline recommends the
use of tubes as reaction vessels and the method has been modified
to take into account the increased rate in tubes (equation 15)
11. Chemical Analysis of Solutions
The analytical techniques employed in the determination of
the concentration of the test substances are left to selection hy
the tester. This is in recognition of the many difft rent
techniques available and the practical advantage of being ; hIe to
make particular use of one of the properties of the subs :ance;
e.g., the NMR or uv spectrum of the substance, or its
—43—
-------
CS—6)O (C ctoher ,
chromatographic behavior. P nalytica1. tt chniqu& s th t pt’rm it ttit’
detthrmination of the test compound to the exclusion of impurities
or photolysis reaction products are recommended to the extent
practicable. Therefore, chromatographic techniques are
particularly desirable. Whenever practicable, an analytical
procedure should he used which has a precision of ±5% or
better. The specific technique which is utilized should he
adequately described.
The p—nitroacetophenone in the chemical actinometer solution
is conveniently analyzed by high—pressure liquid chromatography
using a C 18 reverse—phase column and a uv detector set at 280
nm. The mobile phase in volume percent is 2.5% acetic acid, 50%
acetonitrile, and 47.5% water which is passed through the column
at a flow rate of 2 mL/minute. This analytical procedure was
specifically developed to analyze for PNAP, especially at high
pyridine concentrations (i.e., for pyridine concentrations close
to 0.2 M) [ Mill et al. (1982b), Winterle (1984)1.
12. Outdoor Experimental Conditions
It is important that the photolysis reaction vessels
containing the chemical substance be placed in an area free from
shade and reflections and on a black, non—reflecting, hackgr und
to ensure that they only receive direct and sky r I tiion tro,ri
the sun. The reaction vessels should he tilted .i1 30° lriin
horizontal with the upper end pointing due north in the northi in
_A4_
Ł
-------
CS—6000 (October, 1984)
hemisphere so that they present a larqe surface are i and minimum
pathlength to the sun and create minimal internal reflections.
It is recommended that the photolysis experiments be carried
out during a frost—free time of the year (i.e., May, June, July,
August, or September in the northern hemisphere-—temperature
permitting). This period of time was chosen because the solar
intensity is a maximum and consequently the rate of photolysis
will be a maximum. Thus, the kinetics of photolysis will be
easier to follow, especially for chemicals that photolyze
slowly. Purthermore, in many parts of the United States, t.he
temperature falls below 0°C during the winter months. Therefore,
if the photolysis experiments are carried out in the winter, the
dilute aqueous reaction solution would freeze, the tubes would
break, and the samples would be destroyed. It may be possible to
avoid this problem by placing the reaction tubes in a housing
transparent to the appropriate wavelengths of radiation and
thermostated to control the temperature at 25 ±5°C. Thus, with
this device it may be possible to carry out photolysis experi-
ments at any time of the year. However, transmission of sunU ht.
through the housing could present a (jifficull problem to solve.
113
—45—
-------
CS— 6O()O (October, 984
III. REFERENCES
AA. 1982. Astronomical Almanac.
ASTM 1978. Annual book of ASTM standards, merican Society for
Testing and Materials, Philadelphia, PA., Part 31, Method D.
Benson WR, Lombardo P, Egry IJ, Ross, RD, Barron RP, Masthrook
DL Hansen EA. 1971. Chiordane photoalteration products: Their
preparation and identification. 3 Agr Food Chem 19:857—862.
Benson SW. 1976. Thermochemical kinetics. Second edition.
Wiley—Interscience, N.W., N.Y.
Broderius SJ and &nith LC. 1980. Direct photolysis of
hexacyanoferrate complexes. EPA—600/3—80—003.
Calvert JG and Pitts JN, Jr. 1966. Photochemistry. Wiley,
N.Y., N.Y.
Carey 311 and Fox ME. 1981. Photodegradation of the lampricide
3—trifluoromethyl—4—nitrophenol (TEM). 1. Pathway of the direct
photolysis in solution. 3. Great Lakes RCS. 7:234.
Crossland NO and Wolff CJM. 1984. The fate and biological
effects of pentachiorophenol in outdoor ponds. I. E’ate:
Predicted and observed. Submitted for publication to Aquatic
Tox.
Denterjian KL, Schere KL, and Peterson JT. 1980. Theoretical
estimates of actinic (spherically integrated) flux and photolysis
rate constants of atmospheric species in the lower troposphere.
Mv Sci and Tech 10:389.
Thilin D and Mill T. 1982. Development and application of solar
actinometers. Environ Sci and Technol 16:815.
Grunwell JR, Erickson RH. 1973. Photolysis of parathion. Now
products. 3 Agr Food Chem 21:929—931.
Langford CH, Wingham M, Sastrin vs. 1973. Ligand phntoox 1Li1 i r1
in copper (II) complexes of nitrilotriacetic acid. Environ Sci
and Technol 7:820—822.
Leighton PA. 1961. Photochemistry of air pollution. New York:
Academic Press, Inc.
Mabey WR, Tse D, Baraze A and Mill T. 1983. Photolysis of
nitroaromatics in aquatic systems. I. 2,4,6—trinitrotoluene.
Chemosphere 12: 3.
-46-
-------
CS—600() (October, 1984)
Mancini JL. 1978. Arialyis framework for photodecomposition in
water. Environ Sci and Technol 12:1274—1276.
Mill T, Davenport JE, Dulin DE, Mabey WR and Bawoi R. 1 S1.
Evaluation and optimization of photolysis screenin protocols.
U.S. Environmental Protection Agency. EPA 56O/S— l—0O1.
Mill T, Mahey WR, Homt)erqer !X, Chou T—W, Hendry l x ;, nd n it h
311. 1982a . Laboratory pro1oco s 1 or eva uatinq the tat e t
organic chemicals in air and water. EPA—600/3—82—022.
Mill T, Mabey WR, Hendry DG, Winterle 3, Davenport 1, Barich V,
Dulin D, and Tse D. 1982b. Design and validation of screening
and detailed methods for environmental processes. EPA—
Mill T, Davenport JE, Winterle JS, Mabey WR, Drossman H, Tse D,
and Liu A. 1984. Toxic substances process data generation and
protocol development. EPA Contract No. 68-03—2981 with EPA
Athens Research Laboratory, Office of Research and Development.
Draft final report.
Peterson JT. 1976. Calculated actinic fluxes (290—700nm) for
air pollution photochemistry applications.
Prutton CF and Maron SlI. 1951. Fundamental principles of
physical chemistry. Chapter Xxii. The Macmillan Co. New York,
New York.
Simmons MS and Zepp RG. 1984. The influence of humic substances
on photolysis of nitroaromatics in aqueous systems. Submitted
for publication to Water Res.
Smith JH, Mabey WR, Bohonos N, Holt BR, Lee SS, Chou T—W,
Bomberger DC, Mill T. 1977. Environmental pathways of selected
chemicals in freshwater systems. Part I. Background and
experimental procedures. U.S. Environmental Protection Agency.
EPA 600/7—77—113.
nith JH, Mabey WR, Bohonos N, Holt BR, Lee SS, Chou T—W,
Bomberger DC, Mill T. 1977. Environmental pathways of selected
chemicals in freshwater systems. Part IT. Laboratory studies.
Athens, GA: U.S. Environmental Protection Agency. EPA 600/7—7fl—
074.
Su CC and Zabik MJ. 1972. photochemistry of hioactive
compounds. Photolysis of arylamidine derivatives in w t; ,r. ,J
Agr Fbod Chem 20:320—323.
USEPA. 1979. U.S. Environmental Protection Agency. Oil ice UI
Toxic Substances. Toxic Substances Control: DiscLmsion ef
premanufacture testing policy and technical issues. Reque ;L f r
Comment. Federal Register 44, 16240.
-47-
1 )
-------
CS—6000 (October, 1984)
Winterle J. 1984. private, communication.
Weast RC. 1973—1974. Handbook of eheinistry and physics.
Che nical Rubber Company, Cleveland, Ohio.
Wolfe, NL, Zepp RG, Schlotzhauer PF, and Sink M. 1982.
Transformation pathways of hexachiorocyclopentadiene in the
aquatic environment. Chemosphere 11:91.
Wolfe, NL, ZeppRG, Baughman GL, Fincher RC, Gordon JA. 1976.
Chemical and photochemical transformation of selected pesticides
in aquatic systems. Athens, GA: U.S. Environmental Protection
Agency. EPA 600/3—767—067.
Zepp RG, Wolfe NE ., Gordon JA, Baughman GL. 1975. Dynamics of
2,4rD esters in surface water. Hydrolysis, photolysis, and
vaporization. Environ Sci and Technol 9:1144—1150.
Zepp RG, Wolfe NL, Gordon JA, Fincher RC. 1976. Light—induced
transformations of methoxychlor in aquatic systems. 3 Agr Food
Chern 24:727—733.
Zepp RG, Wolfe NL, Azarraga LV, Cox RH, Pape LW. 1977.
photochemical transformation of the DDT and methoxychlor
degradation products, DDE and DMDE, by sunlight. Arch Environ
Contain Toxicol 6:305—314.
Zepp RG and dine DM. 1977. Rates of direct photolysis in
aquatic environment. Environ Sci Technol 11:359—366.
Zepp RG. 1978. Ouantum yields for reaction of pollutants in
dilute solution. Environ Sci and Technol 12:327.
Zepp RG. 1982. Experimental approaches to environmental
photochemistry. The handbook of environmental chemistry.
0. Hutzinger, Editor. Springer—Verlag.
Zepp RG. 1983. Mathematical modeling of DDE in a quarry.
Kinetics of photolysis of DDE. Unpublished results, U.S. EPA,
Athens, GA.
Zepp RG. 1984. Private communication.
Zepp RG. 1982a. Photochemical transformations in(lIJce(1 by .s 1 ir
ultraviolet radiation in marine ecosystems. In: The role ot
solar ultraviolet radiation in marine ecosystems. Calkins .J,
Editor. Plenum Publishing Corporation, N.Y.
-48-
-------
CS— OOO (October, H$4
IV. APPENDIX: GLOSSARY OF IMPORTANT SYMBOLS
PYR Pyridine
PNAP p—Nitroacetophenone
A Wavelength A
Absorbance at wavelength A.
a Actinometer (composed of PNAP/PYR).
C
Ł Molar absorptivity of a chemical. C.
Molar absorptivity of the actinometer.
Ł light pathlength; the distance traveled by a beam of
light passing through the system.
Sunlight reaction quantum yield of chemical c in
water.
Sunlight reaction quantum yield of the actinometer in
water. Since the r acti n quantum yield is
independent of A, $. . = • (i.e., the reaction quantum
yield of the act inoTWeter measured in the laboratory).
[ C] Molar concentration of chemical c.
[ PYR) Molar concentration of pyricline.
- dECl/dt Direct photolysis rate of chemical. c.
k E Direct photolysis sunlight rate constant in water
p bodies in the environment.
(k E) Maximum direct photolysis sunlight rate constant in
max. water bodies in the environment.
Direct photolysis sunlight rate constant of chemical
P c in water in tubes.
ka Direct photolysis sunlight rate constant of the
p actinometer in water in tubes.
kaA Specific light absorption of a photoreactive chemir.tI
at a low concentration and at wavelonqt:h A.
AQ
_I_ .
-------
CS—E t OO (Octcb ’r ,
k Specific light absorption rate constant inLe ratt d
a. over all wavelengths absorbed by the chemical.
ka Specific light absorption rate constant integrated
a over all wavelengths absorbed by the actinometer.
Sunlight half—life of a chemical in water in tubes.
(t 1 ) The minimum sunlight half—life of a chemical in water
nUn, bodies in the environment.
I The numbers of ph 9 tons of light of wavelength A in
the system per cm per second.
L So1 r irr diance in water in the units l( 3 einsteins
A cnC day’.
y The geometry factor which represents the ratio of the
rate constants in tubes (k ) to the rate constant in
water bodies in the enviroRment (kpE).
i:s
—50—
-------
CG—60 [ U
October, 1984
LABORATORY DETERMINATION OF THE DIRECT PHOTOLYSIS REACTION
QUANTUM YIELD IN AQUEOUS SOLUTION AND SUNLIGHT PHOTOLYSIS
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, DC 20460
1:9
-------
h Lilt) ( : • t ‘b ’ , I t I
D1 CLAI MKR
Certain commercial equipment, instruments, and materials ate
identified in this Test Guideline in order to adequately specify
the experimental procedure. In no case does such identification
imp] y recommendation or endorsement by the Environmental
Protection Agency, nor does it imply that the material,
insbruments, or equipment identified are necessat ily the best
available for. the putpose.
—a-.
-------
(‘— c ) 0 ( t ‘ ‘ t, L ) 4
(‘ )flt t’ rtt
I . INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . • 1
A. Background and Purpose ....... .. 1
B. Definitions and Units ......................•• 3
C. princip1eoftheTeStMeth0d...... *.*.. ” 6
D. Applicabtl1tYafldSPeC1f1c1tY.. 15
II. TEST PROCEDURES •.... •e•s••ss•••••*•••• 18
A • Test ConditionS .. . . . .• . •. ... .... 18
1. IJV—Visible Absorption
Specirophot_otneter . . . . . • . . . . . 18
2. Special Photochernical Lahoratory
Equipment • 18
a. Designoft_heAPParatus... ..... 19
i. Photochemical “Merry—Go—
Round” Reactor • . . . . . • • • . . . . • . . . . 19
ii. Photochemical Optical Bench ..... 21
b . Light Sources • . . • . . . • . . . . • .. . . . . 22
c. Light Filtering Systems •.• ......e... 22
d . Re a C t ion Ce 1 1 S . . . . . . . . . . . , . . . . . • • . . . • 2 3
3. Cell pathlength . . . . . . . . . . . . . . 25
4 • Solvents . . . • .... 25
5. Sterilization 26
6. p8 1 ffects 26
7. Volatile Chemical Suttanc’ S 27
H. Contro’ SolUtiOn 27
9 . Absorpt ion SpeC t. ruin s a Cr 1 r 1 On
for Perforininythe Reaction
Ouantum Yield Experiments . 28
10. Actinometers 28
a. Low Optical Density Actinometers ..... 29
i. p_NitrOaCetOPheflofle—
Pyridine Actinorueter
PNAP/PYR) . . . . • . . . . . . . • . . . 29
ii. p_NitroafliSOlePyridine
ActinOmeter(PNA/PYR) 31
b. 1-ugh Optical Density ActiflOtneter
(Ferrioxalate Actinometer) 33
—ii—
-------
(‘ —t () 111 ( ‘t ,
COfltti t’ tS ( con’ t
11. Chemical Analysis of Solutions . 34
a. Chemical Analysis of Test
Chemical Solutions 34
b. Chemical Analysis of
p—Nitroacetophenone (PNAP) 34
c. Chemical Analysis of
p—Nitroanisole(PNA) 34
d. Chemical Analysis of
Ferrous Ion in the
Ferrioxalate Actinomete . 35
B. Procedure One: Determination of Reaction
Ouantum Yield by the Low Opt;ical Den it:y
Test Chemical and Actino;neLor Method
1. Phase 1 Experiments: IJV—Visihle
Absorption Spectra 35
2. Phase 2 Experiments: Trial
Photolysis Experiments 37
a. Determination of the Approximate
Rate Constant of the Test Chemical.... 37
i. Preparation of Buffer
Solutions .. 37
ii. Preparation of Test
Chemical Solution 38
iii. Performance of the Test 38
h. Actinometry ExperirnerltTs 4()
i . Preparation of Act i noinel ‘r
Solutions 41)
ii. Performance of the Test 42
3. Phase 3 Experiments: Determination
of the Reaction Ouanturn Yield of the
Test Chemical 43
C. Procedure Two: Determination of the
Reaction Quantum Yield by the Low Optical
Density Test Chemical and High Optical
Density Actinometer Method 44
1. Phase 1 Experiments: UV—Visihle
Absorption Spectra 44
r-’)
—iii — 1 luS
-------
(‘(—b( l) ( ‘ ct 1 ’r ,
(ontent s ( (‘t’m’ t
Pa&j e
2. Phase 2 Experiments: Determination
of the Reaction Quantum Yield of the
Test Chemical . . 45
a. Preparation of Buffer Solutions . 45
b. Preparation of Test Chemical
Solution . • 45
c. Preparation of Ferrioxalate
Actinometer Solution 45
d. Determination of the Quantum
Yield of the Test Chemical 46
III. DP%TAA1 D REPORTING 47
A. Procedure One: Determination of the Reaction
Quantum Yield by the Low Optical Density Test
Chemical and Actinometer Method 47
1. Phase 1 Experiments: LW—Visible
Absorption Spectra 47
a. Treatmentof Results 47
b. Test Data Report . 49
2. Phase 2 Experiments: Trial Photolysis
Experiments . . . . . . Si
a. Determination of the Approximate
Rate Constant of the Test
Chemical Si
j. Treatment of Results Si
ii. Specific Analytical and
Recovery Procedures 52
iii. Other Test Conditions 52
iv. Test Data Report
b. Actinometry Experiments 54
i. Treatment of Results . 54
ii. Test Data Report. 54
3. Phase 3 Experiments: Determination of the
Reaction Quantum Yield of the Test
Chemical . . . 55
a. Treatment of Results
b. Other Test Conditions
C. Test Data Report
— lv—
-------
CG— 60 1 0 ( )c t r , T 4
Contents ( con t
Paçje
B. Procedure Two: Determination of the Reaction
Ouantum Yield by the Low Optical Density Test
Chemical and High Optical Density Actinometer
Method . . . 60
1. Phase 1 Experiments: UV—Visible
Absorption Spectra
2. Phase 2 Experiments: Determination of
the Reaction cX antum Yield of the Test
Chemical. 60
a. Treatment of Results 60
h. Specific Analytical and Recovery
Procedures 62
c. Other Test Conditions 62
d. Test Data Report 63
IV. REFERENCES . . . . . . . . . . . . . . . . . . 65
V. APPENDIX .•.... ...... . . .
Hypothetical Illustrative Example: Determination
of the Reaction Ouantum Yield by the Low Optical
Density Test Chemical and Actinometer Method
and Sunlight Photolysis . 67
-v-
-------
CG—60 10 (October, 1984)
LABORATORY DETERMINATION 01’ THE DIRECT pHoroLYsIS
REACTION QUANTUM YIELD IN AQUEOUS SOLUTION AND
SUNLIGHT PUOTOLYSIS
I. ENTRODIICT ION
A. Background and Purpose
Numerous chemicals enter natural aquatic water bodies
from a variety of sources. Some pollutants present in aqueous
media can undergo photochemical transformation in sunlight by
direct photolysis. Therefore, quantitative data in the form of
rate constants and half—lives are needed to determine the
importance of direct photochemical transformation of pollutants
in aqueous media.
Test Guideline CG—6000, the first in a series ot
aqueous photolysis test methods, was designed to determine the
molar absorptivity and reaction quantum yield of a test chemical
in aqueous solution. These parameters can be combined with solar
irradiance data to determine environmentally relevant rate
constants and half—lives in aqueous solution as a function of
latitude and season of the year anywhere in the United States.
Test Guideline CG—6000 was developed as a screening
test to obtain the direct photolysis reaction quantum yield oi u
Chemical in aqueous solution by carrying out photoLysi
experiments in sunlight. This method does not require
sophisticated and expensive photochernical equiprhent and therefore
the reaction quantum yield can be easily determined and at a
modest cost. However, there are circumstances when this method
1
-------
CG—6010 (October, 1984)
may not be app i. i cable . 1’or example , t. h i s p )cedu re i not
applicable for determining the reaction quantum yield of a test
chemical when the temperature outdoors falls below zero degrees
centrigrade. Furthermore, depending upon the status of a risk
assessment for a specific chemical, a more precise value of the
reaction quantum yield may be required. Thus, a more
comprehensive procedure is needed to determine the direct
photolysis reaction quantum yield in the laboratory using
specialized photochemical equipment and monochromatic (or narrow
hand) light. This Test Guideline (CG—6010) describes laboratory
procedures for determining the direct photolysis reaction quantum
yield in aqueous solution.
The reaction quantum yield obtained in this Test
Guideline can be combined with molar absorptivity data and solar
irradiance data to determine environmentally relevant rate
constants and half—lives in aqueous solution as a function of
latitude and season of the year anywhere in the United States.
The procedures described in this Test Guideline are
very detailed and the theory of photolysis in aqueous solution is
relatively complicated. In order to follow these procedures, it
is recommended that the Technical Support Document should he
first studied carefully.
4 r4
_L, )
2
-------
CG—601() (Oclober, 484)
B. Definitions and Units
(1) “Radiant energy”, or radiation, is defined as the
energy traveling as a wave unaccompanied by transter
of matter. Examples include x—rays, visible light,
ultraviolet light, radio waves, etc.
(2) “Absorbance (A. ) “ is defined as the logarithm to the
base 10 of the ratio of the initial intensity (Is) of
a beam of radiant energy to the intensity (I) of the
same beam after passage through a sample at a fixed
wavelength A . Thus, A log 10 (I /1).
(3) The “Beer—Lambert law” states that the absorbance of a
solution of a given chemical species, at a fixed
wavelength, is proportional to the thickness of the
solution (t) , or the light pathlength, and the
concentration of the absorbing species (C).
(4) “Molar absorptivity ( ) “ is defined as the
proportionality constant in the Beer—Lambert law when
the concentration is given in terms of moles per litter
(i.e., molar concentration). Thus, AA ‘A - wh rt,
and c represent the absorhance and molar
absorptivity at wavelength x and and C are
defined in (3). The units of are mo1ar cm’.
Numerical values of molar absorptivity depend upon the
nature of the absorbing species.
3
-------
CG—6010 (October, 1984)
(5) A “first—order reaction” is defined as a reaction in
which the rate of disappearance of a chemical is
directly proportional to the concentration of the
chemical and is not a function of the concentration of
any other chemical present in the reaction mixture.
(6) A “zero—order reaction” is defined as a reaction in
which the rate of disappearance of a chemical is
independent of the concentration of the chemical or
the concentration of any other chemical present in the
reaction mixture.
(7) The “first—order half—life (ti 1 2 )” of a chemical is
defined as the time required for the concentration of
the chemical to be reduced to one—half its initial
value.
(8) The “sunlight direct aqueous photolysis rate constant
(kpE)” is the first—order direct photolysis rate
constant in the units of day’ and is a measure of the
rate of disappearance of a chemical dissolved in a
water body in sunlight.
(9) The “solar irradiance in water (Lx) “ is related to
the sunlight intensity in water at shallow depths and
is proportional to the average light flux (in the
units of lO einsteins cm 2 day ) that is available
to cause photoreaction in the wavelength int rvaI
centered at A over a 24—hour day at a specific
latitude and season date.
I ?ç)
4
-------
CG—6010 (October, 1984)
(10) “The Grotthus—Draper law”, the first law of
photochernistry, states that c nIy li iht which i
absorbed can tR t tect: i V’ ’ Iii j)rI)du ’ i t q . ctu’ n I c.i I
transformation.
(11) The “Stark—Einstein law”, the second law of
photochemistry, states that only one molecule is
activated to an excited state per photon or quantum of
light absorbed.
(12) The “reaction quantum yield ( ) “ for an excited
state process is defined as the fraction of absorbed
light that results in photoreaction at a fixed
wavelength A . It is the ratio of the number o
molecules that photoreact to the number of quanta of
light absorbed or the ratio of the number of moles
that photoreact to the number of einsteins of light
absorbed at a fixed wavelength x.
(13) “Direct photolysis” is defined as the direct
absorption of light by a chemical followed by a
reaction which transforms the parent chemical into one
or more products.
As a convenient reference, a glossary of Lhe important.
symbols used in this Test Guideline is given in Section IV of th
Support Document ICS—6010J
4
5
-------
CG—6010 (October, 1984)
C. Principle of the Test Method
This Test Guideline is based on the principles
developed by Zepp (1978), the use of low optical density
actinometers developed by Mill et al. (1981, 1982) and Dulin and
Mill (1982), and the high optical density ferrioxalate
actinometer developed by Parker (1953) and Hatchard and Parker
(1956).
Zepp (1978) published a paper on the determination o
the reaction quantum yield for the reaction of pollutants in
dilute aqueous solution in the laboratory. Based on this work,
two procedures are described to determine the reaction quantum
yield of a test chemical at low optical density in the
lab®ratory. These procedures involve the use of: (1) low
optical density test chemical and actinometer; and (2) low
optical density test chemical and high optical density
actinometer.
(1) Procedure One: Determination of the Reaction Quantum
Yield by the Low Optical Density Test (‘ht mi u •uid
Actinometer Method
For a low optical density test chemical and actinorneter
in which the absorbance of aqueous solutions is less than 0.02,
the reaction quantum yield of a test chemical at wavelength
A is given by the equation
6
-------
CG—6010 (October, 1984)
(kpc/kpa)(Cxa/ xc) a (1)
where ŁAa and are the molar absorptivities of the
actiriometer and test chemical, respectively, at wavelenjth
A a is the reaction quantum yield of the act-inometer at
wavelength A and kpa and are the first—order direct
photolysis rate constants for actinometer and test chemical,
respectively. These rate constants are defined by the equations
Lfl(C/Ct) = kpat (2)
Lfl(c/C ) = k t , (3)
where (Co)a and (Ct)a are the molar concentrations of actinometer
at time t 0 and t and (C 0 ) and (Ct) are the molar
concentrations of test chemical at time t = 0 and t. Since kpa
and kpc are first—order rate constants, the half—lives of
actinometer and test chemical are:
(tl/ 2 )a = O•6 93 /kpa (4)
(t1i) = 0.693/k (5)
If both the actinometer and test chernica so1t t.ior ; r’
photolyzed in identical cells, equation 3 can he divided into
equation 2. Carrying out this operation and rearranginq the
resultant equation yields
I P1
-I - - ’ - .
7
-------
CG—6010 (October, 1984)
= (kpc/kpa)2 fl(Co/Ct)a . (6)
Procedure One involves the simultaneous phot:oly ;is ot
test chemical and actinorneter in an Ace--type photocheiniciL
“merry—go—round” reactor (PMGRR) using monochromatic light of
wavelength A . Two low optical density actinometers have been
developed by SRI International for the Office of Toxic
Substances/U.S. Environmental Protection Agency to measure the
light intensity incident on the sample during photolysis (Mill et
al. (1981, 1982) and Dulin and Mill (1982)]. These actinometers
are: (1) p—nitroacetophenone--pyridine actiriometer (PNAP/PYR);
and (2) p—nitroanisole—pyridine actinometer (P A/PYR). The rate
constant of each of these actinometers can he adjusLed to match
the rate constant of the test chemical by adjusting the
concentration of pyridine. Since the rate constant is a function
of the reaction quantum yield of the actinometer, the rate
constant can be adjusted according to the following equations:
PNAP/PYR actinometer a = 0.0 169 [ PYRJ (7)
PNA/PYR actinometer 4 ’a = O.437 [ PYR] + 0.000282 , (8)
where [ PYRJ is the molar concentration of pyridine for a PNAP or
PNA concentration of approximately 1 x M.
8 1 ( c)
_a_ j,
-------
CG—6010 (OcLobet -, 1984)
The laboratory procedure [ or determining the reaction
quantum yield of a test chemical in aqueous solution has been
divided into three phases using a uv—visihle absorption
spectrophotometer, an Ace—type PMGRR, and a 450 watt medium
pressure mercury lamp with appropriate filters to isolate the
monochromatic wavelength A . In Phase 1, the molar
absorptivities of a test chemical, 3l3c and C366c are
determined with a uv—visihle absorption spectrophotometer using
procedures outlined in Test Guidelines CG—1050 and CG—6000.
Based on these results, photolysis experiments are carried out at
313 or 366 nm corresponding to the higher value of C313c or
C366c
The Phase 2 procedure is composed of trial photolysis
experiments at the chosen wavelength A (313 or 366 nm) in an
Ace-type PMGRR to determine the approximate rate constant and
half-life of the test chemical and to choose the appropriate
actinometer which has a rate constant approximately the same as
the rate constant of the test chemical. Pirst, an aqueous
solution of test chemical at low optical density is photolyzed in
the PMGRR at the chosen wavelength A to determine kpc and (ty 2 c
using equations 3 and 5, respectively. If (t1 c is less than 12
hours, use the PNA/PYR actinometer. If (t]/ 2 )c is greater than 12
hours, then use the PNAP/PYR actinorneter. Trial photolysis
experiments are then carried out at wavelength A with the
Chosen low optical density actinometer to determine the molar
Concentration of pyridine needed to make the rate cori ;L nt ot tjii,
actinometer approximately match the rate constant () [ Ih( L( Mt
Chemical
9
-------
CG—6010 (October, 1984)
In the Phase 3 procedure, low pt ical density aqueous
solutions of test chemical and actinometot- (at a t ixud molar
concentration of pyridine [ PYRI) are photolyzed in identical
tubes in the PMGRR at wavelength A . Concentrations of test
chemical and actinometer are measured as a function of time.
These data are used in equation 6 to determine the ratio of the
rate constants (kpc/kpa)• The reaction quantum yield of the
actinometer a employed in these experiments, can be determined
at the molar concentration of pyridine [ PYR] using equation 7 or
8. These data along with the molar absorptivity of the test
chemical (c or ) and the act inometer
313c 366c
3l3a or C366a) are substituted in equation I to determine the
reaction quantum yield of the test chemical
(2) Procedure Two: Determination of the Reaction Quantum
Yield by the Low Optical Density Test Chemical and High
Optical Density P ctinometer Method
For a low optical density test chemical
(absorhance < 0.02) and for a high optical density actinometer
(absorbance >2), such as the ferrioxalate actinomotur, thu
reaction quantum yield of the test chemical at wavolenyth
A is given by the equation
,c =a pcpa) (2303Cxct ’ (9)
10
-------
CG—6010 (October, 1984)
where is the reaction quantum yieli f the ferrioxalatt
actinometer at wavelength A ‘ is the molar absorptivity of
the test chemical at wavelength A, t is the cell pathlength,
is the first—order direct photolysis rate constant for the
test chemical, and kpa is the zero—order direct photolysis rate
constant for the ferrioxalate actinometer. These rate constants
are defined by the equations
Lfl(Co/Ct)c = k t (10)
(Ct)a = kpat (11)
where (Co)c and (Ct) are the molar concentrations of the test
chemical at time t = 0 and t and (Ct)a is the molar concentration
of the ferrous ion formed at time t.
The laboratory procedure for determining the reaction
quantum yield of the test chemical is divided into two phases
using a uv—visible absorption spectrophotometer and a
photochemical optical bench (POB) or a PMGRR containing a 450
watt medium pressure lamp with appropriate filters to isolate the
frionochromatic wavelength x • In Phase 1, the molar
absorptivities of test chemical (C313c and C366c) are
determined using spectroscopic procedures outlined in Test
Guidelines CG—l050 and CG—6000. Based on these resuiLs,
photolysis experiments are carried out at 313 or 366 nm
corresponding to the higher value of or C366c
.4 1
_1_ ‘..)
-------
CG—6010 (October, 1984)
In the Phase 2 procedure, aqueous low optical density
test chemical solution (absorbance < 0.02) and aqueous hidh
optical density solution of ferrioxalate actinotneter (absorhance
>2) are photolyzed sequentially in identical cells in PUB or
PMGRR at wavelength x . In the first and third set of
experiMents, the ferrioxalate actinometer is photolyzed for a few
minutes and the molar concentration of ferrous ion formed
is measured as a function of time t. These data are used in
equation 11 to determine an average value of the actinometer rate
constant (kpa)ave . The second series of experiments involves
the photolysis of the aqueous solution of test chemical in the
POB or PMGRR in identical cells to those used in the actinometer
experiments and the molar concentration of test chemical (Ct) is
measured as a function of the time t. These data are used in
equation 10 to determine kpc• Using and (kpa)ave the molar
absorptivity of the test chemical 3l3c or e 366 ) , the
pathlength of the cell t , and the reaction quantum yield of the
ferrioxalate actinometer a at wavelength A in equation 9,
the reaction quantum yield of the test chemical can he
determined.
As described in detail in Test Guideline CG—6000 and
the associated Technical Support Document CS—6000, and briefly in
Technical Support Document CS—6010, the direct sunlight
photolysis of a chemical in an optically thin aqueous solution
can be described by the following equations:
12
-------
CG—6010 (October, 1984)
tn(C 0 /Ct) = kpEt (12)
= O.693/k 1 .. (13)
kpE = Eka (14)
where is the reaction quantum yield of the chemical in
dilute solution and is independent of the wavelength, ka =
E kaA the sum of ka values for all wavelengths of sunlight
that are absorbed by the chemical(i.e., the light absorption rate
constant), t is the time, C 0 and C are the molar concentrations
of chemical at t = 0 and t, and t1/ represents the half—life.
The term kpE represents the first—order direct photolysis rate
constant for a chemical in a water body in sunlight in the units
of reciprocal time. In general, the reaction quantum yield 4 ’E
is equivalent to the reaction quantum yield c determined in
the laboratory.
Furthermore, under the same conditions cited above, the
first—order direct photolysis rate constant, kpEl is given by the
eguat ion
kpE = cLx (ii))
where is the reaction quantum yield, is tht rnoJ ir
absorptivity in the units molar—’ cm 1 , L is the ol. r
irradiance in water in the units ,O einsteins cm 2 day , and
the summation is taken over the range A = 290 to 800 nm. is
13
-------
CG—60L() (Oclober, 1984)
the solar irradiance at shallow depths for a water body under
clear sky conditions and is a function of latitude and season of
the year. Solar irradiance data are tabulated in Test Guideline
CG—6000, Tables 3—6, as a function of latitude and season of the
year.
A simple screening test has been developed in Test
Guideline CG—6000 using equation is. As an approximation, it has
been assumed that the reaction quantum yield •E is equal to
one, the maximum value. As a result, the upper limit for the
direct photolysis sunlight rate constant in aqueous solution is
obtained and equation 15 becomes
(kpE)max = ECxcL . (16)
Using equation 16 in equation 13, the lower limit for the half—
life is then given by
(t1i )min = Q•69 3 /(kpE)max. (17)
The molar absorptivity can be determined experimentally by the
method outlined in Section II.B,l and the solar irradiarice data
are tabulated in Tables 3—6 of Test Guidelines CG—6000. These
data can then be used in equation 16 to calculate (kpE)mflax..
Finally, (kpE)max can be substituted in equation 17. t()
Based on these data, a decision can he m id wh Lher
to determine the reaction quantum yield of the test chemical in
the laboratory.
14
-------
CG—60I0 (Octou ’r, l S4)
Once the reaction quantum yield has been determined in
the laboratory by Procedures One or Two of this Test Guideline
(CG—6010), it can be combined with the molar absorptivtty data
a nd the appropriate LA values to calculate
function of latitude and season of the year anywhere in the
United States using equation 15. The corresponding half—life can
be calculated using kpE in equation 13.
D. Applicability and Specificity
F’or environmental photochemistry, the general
procedures outlined in this Test Guideline are applicable to all
chemicals which have uv—visible absorption maxima in the range
290—800 nm. Some chemicals have absorption maxima significantly
below 290 nm and consequently cannot undergo direct photolysis in
sunlight (e.g., chemicals such as alkanes, alkenes, alkynes,
saturated alcohols, and saturated acids). This is a direct
consequence of the Grotthus—Draper law of photocheinistry. Some
chemicals have absorption maxima significantly below 290 n;n but
have measurable absorption tails above the baseline in their
absorption spectrum at wavelengths greater than 290 nm.
Photolysis experiments should be carried out for these
chemicals.
These test methods are applicable to pure chemicals and
not to technical grade chemicals.
The molar absorptivity data (cu,) obtained in S cti n
II.B.1 can be combined with the appropriate L data to estiinat:e
15
1’)
-------
C(—6010 (Oct ocr, 984)
( .it sh. I ow d ’pt Ii,; in wot ‘i t; •i
funct ion of latitude and season of the ycir in the u. S. It tho se
data indicate that aqueous photolysis is a important process
relative to other transformation processes (e.g., biodegradation,
hydrolysis, oxidation, etc.), then it is recommended that the
reaction quantum yield • be determined in the laboratory by
Procedures One or Two outlined in this guideline. Once has
been determined, it can be combined with the molar absorptivity
data and the appropriate L data to calculate kpE at
shallow depths in water bodies as a function of latitude and
season of the year in the United States. The corresponding h iI-
life can then he calculated from kpE.
Procedure One is only applicable to solutions of test
chemicals and actinometers which have low optical densities
(i.e., absorbance < 0.02). Procedure Two is only applicable to
solutions of test chemicals with low optical densities and
actinomčters with high optical densities (i.e., absorbance >2).
Procedure one, as described in detail in this Test
Gui (Ic line, is limited to the spectral rog ion 290—40() urn becnise
the recommended low optical density actinoineters are only
sensitive to light in this region. However, Procedure Two using
the ferrioxalate actinorneter is useful in the spectral region
290—500 nm. This procedure can be extended up to approximately
750nm using the Reinicke’s salt actinometer [ de Mayo and Shizuka
(1976)]
16
-------
CG—6010 (October, 1984)
There is a third procedure which could be used to determine
in the laboratory. This procedure involves the
determination of • under the conditions when the test chemical
and actinometer solutions both have high optical density.
However, in general, this method is I united to only very SOlUt)le
or strongly absorbing organic compounds. Thus, this method is
not described in detail as a procedure in this Test Guideline.
However, this method is briefly described in the associated
Technical Support Document CS-6010 (Section II.E.3).
This Test Guideline has been designed to determine the
molar absorptivity of a test chemical, and its quantum
yield, c • These parameters can be used to determine environ-
mental rate constants at low absorbance and shallow depths in pure
water as a function of latitude and season of the year. Tables o [
solar irradiance (Table 3—6) have been included in Test Guideline
CG-6000 to carry out all the calculations. However, the method is
really very general and can be extended to determine the rates of
photolysis over a range of other environmental conditions using a
Computer program. Zepp has written a GC SOLAR computer program to
Calculate the rates of photolysis as a function of depth in water,
as a function of the attenuation (or absorption) coefficient of the
water (c ) for natural water bodies, the average.ozone layer
thickness that pertains to the seasons and location ot inter’,;t,
and as a function of latitude and season of the year. Thu compu r
program is available on request. [ R. Zepp, Environmental Re.se. rch
t.aboratory, U.S. Environmental Protection Agency, College Station
Road, Athens, Georgia 30601.1
4 1
17 -
-------
CG—601() (Octo3er, l 84)
II. TEST PROCEDURES
A. Test Conditions
‘•
The recommended uv—visih [ e absorption spectrophotoineter
is described in Test Guideline CG—1050.
2. Special Photochernical Laboratory Equipment
There are a number of different designs of photochemica].
equipment which can be used to measure the reaction quantum yield
in the laboratory and the one chosen will depend on Lhe equipment
available and to a certain extent on the light source and filter
system chosen for the measurement. The apparatus chosen must
contain a light source, appropriate Niters, sample holders, and
cells which allow solutions of test chemical and actirioneter to be
reproducibly irradiated with a uniform and constant amount of light
at a discrete or narrow band wavelength x . The temperature of
the reaction cells must be at a reasonably constant temperature t 1
± 2°C in the range 20—30°C. The apparatus should be housed in a
separate part of a laboratory and properly shielded so that
laboratory personnel are not exposed to uv light and to exclude
extraneous light that could contribute to the photorea t ion ol:
chemicals. If the ferrioxalate actinometer is USed, i n
actinornetry experiments must he carried Out in a d. rk reom with
photographic “safe lights”. Excellent reviews and d s ripLions 01
a wide variety of photochernical equipment used to measure the
18
‘S
-------
‘(;—( () It) ( () t l)’t , 1 ‘4 4
reaction quantum yield in the laboratory are given by Calvert and
pitts (1966), de Mayo and Shizuka (1976), Murov (1973), and Mill et
al., (1982a) and these references are highly recommended.
a. Design of the Apparatus
i. Photochemical “Merry—Go—Round” Reactor
The design of the photochemical “merry—go--round” reactor
(PMGRR) has been described in the literature [ Moses et al. ( 969),
Murov (1973), de Mayo and Shizuka (1976)]. In the design of a
PMGRR, the light source is in the center of the apparatus with
reaction tubes arranged in a ring around the light source. The
ring rotates around the light source to give a uniform irradiation
of the reaction tubes, Figure 1A. Glass filters (f) may be
inserted between the light source and the reaction tubes. Filter
solutions (f) may also be contained in the immersion well holding
the light source or in a glass donut that surrounds the light
source. In order to dissipate the heat generated by the light
source, the PMGRR may he immersed in a water bath or a stream ot
air may he passed through the space between the light source and
the filters. The filter solutions can be circulated to an external
cooling source. There are two types of PMGRR equipment; the Ace—
type and the Moses—type and both are very similar. The Ace—typo
PMGRR is designed so that the entire reaction cell is irradiated
while the Moses—type PMGRR [ Moses et al. (1969)] is designed with
Windows (which act as slits with a fixed aperture) so that only i
narrow portion of the reaction cell is irradiated. ;iric ttu Am’ —
type PMGRR is the only one commercially available, this Test
Guideline has been specifically designed to use this PMGRR.
19
-------
i ct ,
I t
I =a —
0
S
e.’ I
L
R
(b)
Figure 1A.
(a ) e M - Go - Round” Apparatus
(b) Cylindrical cell ( R). l incident light;
a transmitted light p sed through cell;
S a slit
F 1 1F 2 F 3 F 4 J
Figure 18. A typicel opucel train, where E: point source arc;
L: lens; F: solunori and gIa s filters; R 1 : reaction cell;
R 2 : actinometer cell; V a: thermostat vessel; W: quartz window.
(a)
20
-------
CG—6010 (October, 1984)
However, if the Moses—type PMGRR is available, this Test Guideline
has to be slightly modified as descri.bed in Sect ion .1 .1 .c ot
the Technical support Document. (‘S—6() 10.
Small cylindrical tubes are used in the PMGRR to hold small
volumes of test chemical arid actinometer solutions. One tube is
used for each datum point measured.
ii. Photochemical Optical Bench
The photochemical optical bench (POB) is composed of a
light source, a lens, glass or chemical solution filters, and a
reaction vessel. The component parts are mounted on a rail in the
optical bench. A typical P013 apparatus is depicted in } ‘igure 11 .
The light source is located at the end of the bench and contains a
housing around a light source and a lens to collimate the beam.
The light passes through glass filters and/or through cells
containing filter solutions to transmit wavelength . Filter
solutions in the cell may be circulated through a cooling system
and the glass filters should be cooled with a stream of air to
prevent heat buildup. The photochemical reaction vessel containing
the test chemical solution or actinorneter solution is nount:od
coaxially with the lamp on the bench so that: the i It rid I igh I
enters the window of the reaction vessel. The react: ion vessel
Should be temperature controlled by circulating COflStaflt t:ernperdtur
water through side walls. The light flux is usuaJly measured
before and after the photolysis of the test chemical solution.
Commercially available optical benches are highly recommended hut
21 1 5
-------
(‘C—6 lC (Oct er ,
simple “home—made” benches can he used [ A ndre (1977H . If i
“home—made” bench is used, it must he coTnpletely descr hed.
h. Light Sources
There are a number of light sources which ar available
for use in photochemical studies: for example, low, medium, and
high pressure mercury lamps; xenon lamps; and lasers. The
characteristics and application of these lamps are described in
detail by Calvert and Pitts (1966), Murov (1973), and (I C Mayo and
Shizuka (1976). The use of a POB or PMGRR will often narrow the
choice of lamp with regard to use of a point source or a tube—
type lamp since the former can he focused to give a collimated
beam while the latter cannot. The light source should emit light
at a constant and high intensity. The 450 watt medium pressure
mercury lamp is highly recommended for use in this Test
Guideline.
c. Light Filtering Systems
Monochromatic or narrow band wavelength lijht is
essential for the accurate determination of the reaction quantum
yield of a test chemical. Various systems for isolating
monochromatic or narrow band wavelengths are described by (‘advert
and Pitts (1966), Murov (1973), and cle Mayo and Shizuka (1976)
Two filter systems commonly used to isolate the 3 1 md 66 in1
from a 450 watt medium pressure mercury lamp are highly
recommended and have been incorporated in this Test (iiideHne.
These two filter systems are described as Follows:
22
-------
CG—6010 (October, 1984)
(1) 313 nm filter ing system: (‘orni nq ( —7 ‘4
f ii ter* with 0 . 005 M pot .i ; i tim ch roin t ‘ ;o 1 tit i on
containing 39 sodium carbonate.
(2) 366 nm filtering system: Corning Glass Cs 0—52 and
Cs 7—60 filters.
Reagent grade chemicals should he used to prepare
the chemical filter solution. Since this filter sO1.Ut )fl (and in
ijeneral any f i I tier solut. ion) (1eqra(1es. 1ow1 y over pr 1onqe(1
periods of photolysis, the solution should he careful.Iy
monitored. Even when tap water is used to cool the lamp, the
buildup of solid material or algae may reduce the light
intensity; and this must be checked repeatedly.
d. Reaction Cells
In general, reaction cells of large volume are
appropriate for POR equipment while small reaction cells are used
for PMGRR equipment . These cel is should bO cOflSt. rtict ‘d of
horosilicate glass or quartz. ctinorneLer and test. chemical
solutions must he photolyzed in identical cells and ;tiou1 1
contain the same volumes of actinometer and test chemical
solution.
1 1f the test chemical does not absorb light at wavelengths
greater than 400 nm, this filter is not needed. The 754 filter
is designed to block out visible light.
23
-------
CG—6d10 (Octouer, S4)
For the PMGRR equipment, disposable culture tubes
(13 x 100 mm) with Teflon—lined screw caps or quartz t:uh s (13 x
100 mm) with quartz ground glass s oppors or boro i 1 i cut
caps , Te f Ion—li ned , are reco nme nded I or us as r ’ac t I on t nbc
Grease should be avoided since hydrophobic chemicals might adsorb
to it. In carrying out the photolysis experiments, one tube is
used for each datum point measured. The pathlength of these
cells is discussed in Section II.A.3.
For the POB equipment, the most common and
functional design for reaction cells is a cylindrical shape with
optically flat circular windows fused to each end of the cylinder
and at right angles to its axis. The windows should he made of
material that will transmit 100% of the light at the desired
wavelengths. Optically flat quartz windows are recommended. The
size of the cylindrical reaction cell and windows should be
consistent with the dimensions of the light beam used in the
equipment. The reaction cell should be of sufficient volume to
permit removal of samples for analysis without significantly
altering the volume of the reaction solution in the cell. The
cell pathlength can be measured directly or it can be determined
by the procedure discussed in Section tI.A.3. Details for tht?
construction of these reaction cells may he found in (‘alverL anti
Pitts (1966) and de Mayo and Shizuka (1976).
24
-------
CG—6010 (October, 1984)
3. Cell Pathlength
Zepp (1978) described an experimental method using an
isolated wavelength band to determine the effective pathlength of
any cell. This procedure is described in the associated
Technical Support Document CS—6010, Section rI.D). This
procedure has been used to measure the effective pathlength of
Corning Glass culture tubes 13 x 100 mm and it was Found that
was 11.2 mm (Mill et al. (1982a)I. rn a similar manner, reaction
tubes made from borosilicate glass stock of o.d. 12 mm had an
effective pathlength of 10.0 mm [ Mill et al. (1982a)J. This
procedure can be used to measure the pathlength of cylindrical
cells designed for the POB apparatus. However, it is recommended
that the pathlength of the cylindrical cells be measured directly
with a precise centimeter ruler or an equivalent measuring
device.
4. Solvents
If the half—life of an aqueous solution of test
chemical in the photochemical equipment is less than 24 hours,
then distilled water meeting ASTM Type II standards, or an
equivalent grade, is recommended for use in this Test
Guideline. If the half—life of an aqueous solution of test
chemical in the photochemical equipment is greater than 24 hours,
then water meeting ASTM type hA standards, or an iuiValOnt
grade, is highly recommended for use in this Test Guidel in.’ I c)
minimize biodegradation. ASTM Type II and lEA waler
25
-------
C;—6010 ( ct n r, i’ 4)
described in ASTM 1) 1193—77. Air saturated water i re iuired t
photolysis of test chemical solutions. Air saturated water can
be easily prepared by allowing the ASTM Type II water ta
equilibrate in a vessel plugged with cotton or ASTM Type hA
water to equilibrate in a vessel plugged with sterile cotton.
Reagent grade acetonitrile is recommended as the
organic cosolvent with water in photochernical studies.
Spectrograde acetonitrile or methanol is recommended t or
sp. ctro.scopic studies to determine the molar absorptivi Ly’ L t the
test chemical.
5. Sterilization
If the half—life of an aqueous solution of test
chemical in the photochemical equipment is greater than 24 hours,
then it is important to sterilize all glassware and to use
aseptic conditions in the preparation of all solutions for
photolysis studies in order to eliminate or minimize
biodegradation. Glassware can he sterilized in an autoclave or
by any other suitable non—chemical method.
6. pH Effects
It is recommended that all photolysis and molar
absorptivity experiments be carried out at pHs at least two pH
units above the and at least two pH units below the pK for
any chemical that ionizes or protonates (e.g., carboxylic acids,
phenols, and amines). Buffers described in Section II.H.l should
be used.
26
-------
CG—6010 (October, 1984)
7. Volatile Chemical Substances
Spec i a 1 ca re s hiu 1(1 be t: a ken wh.’ u t es t d a va I at i I e
chemical. so that t.ho chomi cal ;uhst,in ’t’ i ; ii t I t ‘iii’
volatilization durinq the course of the photolysis experiment.
Thus, it is important to effectively seal the reaction vessels.
Disposable culture tubes with Teflon—lined screw caps or quartz
tubes with quartz or borosilicate screw caps, Teflon—lined, are
recommended. Grease should not be used. Volatile compounds can
be conveniently studied in culture tubes equipped with gas—tight
Mininert® valves. Samples can be introduced into or removed from
the tubes through the septum in these valves with no loss of
substrate. As an alternative, the tubes can he sealed with a
torch. In addition, the reaction vessels should h as completely
filled as is possible to minimize volatilization to any air
space.
8. Control Solution
It is extremely important to take certain precautions
to prevent loss of chemical from the reaction vessels by
processes other than photolysis. For example, biodegradation and
volatilization can be eliminated or minimizer] by iis of st.eril(
Conditions and minimal air space in sealed ves3els. IIydr oIy ; 1’;
is a process which cannot he minimized by such t:echni pi s. Ihw;,
Control vessels containing test substance which are not eXposed
to light are required. In this way, the loss of test. chemical. in
processes other than photolysis may be determined and eliminated
27
i 1
L
-------
CG—6010 (October, 1984)
or minimized. For simplicity, if the loss of chemical in tne
control is small (i.e., approximately 10% or less), one can
calculate a first—order loss, klossl and subtract it froTt
(kp) bs. to give the corrected direct phorolysis r-it ’ n ti .t
k . If hydrolysis is found to he sitjni t icuut (qreil bin 1O’
hydrolysis studies should he carried out first. fTest (uide ine
CG—5000J
9. Absorption Spectrum as a Criterion for Performing
the Reaction Quantum Yield Experiments
This aqueous photolysis test is applicable to all
chemicals rhich have uv—visible absorption maxima in the range
290—800 nm. Some chemicals have absorption maxima significantly
below 290 nm hut have measurable absorption tails b ve the
baseline in their absorption spectrum at avelenjtJis greater than
290 nm. Photolysis experiments should he carried out for these
chemicals to determine the reaction quantum yield. The
absorption spectrum of the chemical in aqueous soluf ion can be
measured by the procedures given in Test Guideline CG—iU ).
10. A ctinometers
Chemical actinometers are used in reaction quantum
yield experiments to measure the integrated light intensity
incident on the sample during photolysis. Chemica! atHnintfter
have photochemical reactions which have well—defined r ’u(:LfmnI
quantum yields *a at wavelength x
i ’T)
Ł0 IS
-------
CG—6010 (October, 1984)
a. Low Optical Density P ctinometers
Two low optical. density actinometers (absorbance
<0.02) are described which can he used to determine the re ct-.i n
quantum yield of a test chemical at low optical density
(absorbance < 0.02 ) in Procedure One of this Test Guideline.
These actinometers are: (i) p—nitroacetophenone—pyridirie
actinometer (PNAP/PYR); and (ii) p—nitroanisole—pyridine
actinometer (PNA/PYR). The rate constants and half—1.ives of
these actinometers can he adjusted to match the rate constant and
half—life of the test chemical by adjusting the concentration ot
pyridine. With a 450 watt medium pressure mercury lamp in an
Ace-type PMGRR, the PNAP/PYR actinometer can be adjusted with
pyridine to have half—lives that range from greater than 12 hours
to several weeks while the PNA/PYR actinometer can be adjusted
with pyridine to have half-lives that range from approximately 15
minutes to approximately 12 hours.*
i. p—rjitroacetophenone—Pyridine
Actinometer (PNAP/PYR)
The reaction quantum yield of the PNAI /I YR
actinometer is a function of the molar concentration oh pyri(hn
EPYR} and this relationship is given by equation 7. Thi’;
equation is valid up to 0.2 M pyridine. The initial
Th a Moses—type PMGRR is used, then new criteria have to he
defined to determine which of these two actinometers should he
used. For details see Section II.J.11.c of the Technical
Support Document CS—6010.
29
-------
CG—6010 (October, i 4)
concentration of PNAP (C 0 ) is precisely set in the range of
approximately 1 x 10 M. The molar absorptivities of PNAP at
313 arid 366 rim are: 3l3a = 2,056 ‘1 cm ’; and
366a = 160 M’ cm . The chemical analysis (it PNA v n
Section II.A.1l.h.
Trial photolysis experiments are required to
determine the concentration of pyridine needed to adjust the rate
constant of the actinometer to approximateLy equal the rate
constant of the test chemical (within ± 50%). In the Phase 2
experiments of Procedure One, the rate constant of the test
chemical kpc is determined in the Ace—type PMGRR with a 450 watt
medium pressure mercury lamp. As an approximation, for trie
PNAP/PYR actinometer in the PMGRR with a 450 watt medium pressure
mercury lamp, the concentration of pyridine needed to adjust: the
rate constant of the actinorneter to approximately equal the rate
constant of the test chemical at 313 and 366 rim is given by the
following equations:
at 313 nm [ PYR] = 1.16 kpc (18)
at 366 nm [ PYR) = 5.95 , (19)
where kpc is in the units of (hours) ’ and [ PYRI is the molar
concentration of pyridine.
A trial experiment is then carried out in tti Phase
2 procedure by preparing the actinometer with the molar:
30
-------
CG—6010 (October, 1984)
concentration of pyridine EPYR] estimated from equations lB or 19
and photolyzing the actinometer solution at a fixed molar
concentration of PNAP (C 0 ) in the PMGRR at 313 or 366 nm. Alter
approximately 50% transformation, the concentrat ion of PNAP (Ce)
is recorded at time t and an approximate rate constant kpa is
calculated using these data in equation 2. This trial experiment
is repeated by adjusting the molar concentration of pyridine
[ PYR] until the rate constant of the test chemical and
actinometer are approximately equal.
ii. p—Nitroanisole—Pyridine
Actinometer (Prs A/PYR)
The reaction quantum yield of the PNA/PYR
actinometer is a function of the molar concentration of pyridine
IPYRI and this relationship is given by equation B. This
equation is valid up to 0.02 M pyridine. The initial
concentration of PNA (C 0 ) is precisely set in the range of
approximately 1 x l0 M at 366 nm and at approximately 0.4 x l0 M
at 313 nm. The molar absorptivities of PNA at 313 arid 366 nm
are: 313a = 10,300 M ’ cm 1 ; and Ł366a_ 1,990 r4 1 cm’. The
chemical analysis for PNA is given in Section lI.A.ll.c.
Trial photolysis experiments are required to
determine the concentration of pyridine needed to adjust: the rat
Constant of the actinometer to approximately equal th’ rate
Constant of the test chemical (within ± 50%). In the Phase 2
experiments of Procedure One, the rate constant of the test
Chemical is determined in an Ace—type PMGRR with a 450 watt
r
31
-------
medium pressure mercury lamp. As an approximation, for the
PMA/PYR actinometer in the PMGRR with a 450 watt medium pressure
mercury lamp, the concentration of pyridine needed to adjust the
rate constant of the actinometer to equal the rate constant of
the test chemical at 313 and 366 nm is given by the following
equations:
; t 313 nm IPYRJ = 8.93 x 10 (k 1 ) . - 0.0722) (20)
at 366 nm [ PYRI = 1.85 x iO— 3 (k — 0.0349) , (21)
where is in the units of (hours) and [ PYR] is the moiar
concentration of pyridine.
A trial experiment is then carried out in the Phase
2 procedure by preparing the actinometer with the molar
concentration of pyridine [ PYR] estimated from equation 20 or 21
and photolyzing the actinometer solution at a fixed molar
concentration of PNA (C 0 ) in the Ace—type PMGRR at 313 or 366
nm. After approximately 50% transformation, the concentration o
PNA (Ct) is recorded at time t and an approximate rate consL nt
kpa is calculated using these data in equation 2. This trial
experiment is repeated by adjusting the molar concentration of
pyridine [ PYR] until the rate constant of the test chemical and
actinometer are approximately equal.
-ç r (‘
32 1 )
-------
CG—6010 (October, 1984)
b. High Optical Density Actinometer
(Ferrioxalate Act inometer)
The ferrioxalate actinometer at high optical
density (absorbance > 2) is a widely used actinometer by
photochemists and has been recommended for use in Procedure Two
[ Murov (1973), de Mayo and Shizuka (1976), Calvert and Pitts
(1966)1. This actinometer is applicable over the environmentally
relevant range 290—500 nm. The net photochemical reaction is
e 3 + (C 2 0 4 2 )/2 — ---> Fe 2 + CO 2 (22)
At high optical density, the reaction kinetics are zero—order and
the kinetics are followed by measuring the molar concentration of
Fe 2 formed as a function of time t using equation 11. The Fe 2
formed is measured by procedures outlined in Section II.A.ll.d.
Murov (1973) gives a detailed procedure for using this
actinometer and it is highly recommended. This procedure should
be modified slightly to use the ferrioxalale at 0.15 M. The
irradiated solution is diluted lOOfold prior to anaiysu for
ferrous ion, which must not be allowed to exceed 0.005 M. Table
1 of the associated Technical Support Document CS—6010 lists the
reaction quantum yield as a function of wavelength. At 0.15 M,
the quantum yield at 313 and 366 rim is 1.20 and 1.18,
respectively. All ferrioxalate actinometry experiments must he
Carried out in a dark room with photographic “safe lights.”
Ferrioxylate is available from Alfa Inorganics. If the test
33
1,l
-------
CG—6010 (October, 1984)
cheinica I absorbs 1. ight at. wave leng t.h qreat r than O ) inn, t hen
Reinicke’s salt can he used as the acLin meter. The u e et
Reinicice’s salt actinorneter is described in Technic i1 Suppert
Document CG—6010, Section I1.C.2.c.
11. Chemical Analysis of Solutions
a. Chemical Analysis of Test Chemical Solutions
In determining the concentration of the chemical in
solution, an analytical method should be selected which is most
appi icahie to the analysis of the specific chemic il st nce
Chroma tograph ic methods are jene ra I ly recommended t)ec, usO 0
their chemical specificity in analyzing the parent chonicil
substance without interference from impurities. Whenever
practicable the chosen analytical method should have a precision
of ± 5 percent or better.
b. Chemical Analysis of p—t’Jitroacetophenone (PNAP)
The p—nitroacetophenone (PNAP) in the chemical
actinometer solution is conveniently analyzed by hi;h—pressure
liquid chromatography using a 30 cm C 18 reverne—ph se column in
a uv detector set at 280 nm. The mobile phase in V) I in ’
is 2.5% acetic acid, 50% acetonitrile, ant -] 47.5% wuter which ii
passed through the column at a flow rate of 2 mL/minut:e. EMili
et al. (1982) and Mill and Dulin (1982)1.
c. Chemical Analysis of p—Nitroanisole (PNA)
The p—nitroanisole (PNA) in the chemical
actinometer solution is conveniently analyzed by high—pressure
I t-: -
-------
CG—E O U) ()ct. h ’r, 19B4
liquid chromatography using a 30 cm C 18 reverse—phase column and
a uv detector set at 280 nm. The mobile phase in volume percent
is 50% acetonitrile and 50% water which is passed through the
column at a flow rate of 2 mL/minute [ Dulin and Mill. (1982)1.
d. Chemical Analysis of Ferrous Ion in
the Ferrioxalate Actinometer
The concentration of Fe 2 formed in the photolysis
of the ferrioxalate actinometer is measured spectrophotometrically
via the formation of a red phenanthroline complex and determining
the absorbance of the complex at 510 nm. Murov (1973) describes
a detailed procedure for measuring the molar concentration of
ferrous ion formed in the photolysis reaction and this procedure
is highly recommended. This procedure has been modified slightly
to use the ferrioxalate at 0.15 M. As a result, the irradiated
solution has to be diluted 100—fold prior to analysis for ferrous
ion, which must not he allowed to exceed 0.005 M.
B. Procedure One: Determination of the Reaction fluantum
Yield by the Low Optical Density Test Chemical. and
Actiriorneter Method
1. Phase 1 Experiments: fly—Visible Absorption Spectra
The uv—visible absorption spectra in aqueous
Solution can be determined by the methods described •in Thst
Guidelines CG—1050 and CG—6000. It is recommended that t h
following additional procedures be followed:
35
-------
ct;— t ) I ü ( ) t , j Ľ9 4
(a) For chemicals which 1t)fl i ze or protoihi tc ( e . .
carboxylic acids, phenols, amines), carry out uv—visible
absorption studies at pHs at least two pH units above the pKa and
at least two pH units below the PKa Prepare buffer solutions at
25°C using reagent grade chemicals and disti’led water as
follows:
pFIS in the range 3_6*: NaH 2 PO 4 , HC 1
pus in the range 6_8*: KU 2 PO 4 , NaOH
pus in the range >8: Prepare buffers as described in the
Handbook of Chemistry and Physics.
Check the Pu of all the buffer solutions with a pH meter at 25°C
and adjust to the proper pH, if necessary. These buffer
solutions can then be added to the test chemical solution until
the desired pH is obtained. tf these buffers are inadequate,
then adjust the pH of the test chemical solution with I M FIC1 or
NaOI-1 at 25°C.
(5) Measure the absorbance, , as a function of
wavelength in the range 290—800 nm in duplicate. If applicable,
measure at each experimental pH [ Section II.A.6, p.2 6 1.
*Use the minimum concentration of buffers to attain th* desired
pH.
36
-------
CG— 60 1 0 ( Jc t oh ’ r / I 4 4
Record, in duplicate, the baseline where both the sample and
reference cells are filled with blank solutions. These data will
be used to calculate the molar absorptivities for the appropriate
intervals and wavelength centers, Table 1, Test Guideline CG—
6000, where the test chemical absorbs light. The wavelength
center is defined as the midpoint of the interval range listed in
Table 1.
Measure at 313.0 nm and 366.0 nm in
duplicate. These data will he used to calculate thc molar
absorptivities of the test chemical, and 366c
Photolysis experiments should be carried out at the wavelength
corresponding to the higher value of the molar absorptivity.
It must be emphasized that the molar
absorptivities of the test chemical (c),c) must he carefully
determined, especially in the tails of the absorption bands at
X . 290 nm. Large errors will occur when calculating
photolysis rate constants and half—lives if these measurements
are not carefully carried out.
2. Phase 2 Experiments: V Tria PhoLoI sisExVeriment
a. Determination of the Approximate Rat
Constant of the Test Chemical
i. Preparation of Buffer Solutions
Prepare buffer solutions according to the
procedures outlined in Section II.B.]. (p. 35) using reagent grade
chemicals and pure water as described under Test Conditions,
Section II.A.4 (p. 2 5), for chemicals that reversibly ionize or
protonate. 37
Vt
-------
CG—6Ci( (ctch r ,
ii. Prilirat i n t 1st (‘hem i cal el ut i en
Prepare a homogeneous se I ut ion nt t est chem i c,i I
below its water soluhility and at an absorhance less than (1.02 in
the photolysis reaction vessel at 313 or 366 nm. For very
hydrophobic chemicals, it is difficult and time consuming to
prepare aqueous solutions. To facilitate the preparation of
aqueous solutions containing very hydrophobic chemicals and to
allow for easier analytical procedures, the following procedure
may he used as an aid in the dissolution of the test chemical.
Dissolve the pure test chemical in reagent grade aceten it r i ie
Add pure water as described under Test Conditions, Section II. \.4
(p. 25), or buffer solution, as described in Section TI.k.l (p.
35), for chemicals which ionize or protonate, to an aliquot of
the acetoriitrile solution. Do not exceed one—volume percent of
acetonitrile in the final solution.
iii. Performance of the Test
Prepare an aqueous solution of test chemical as
described in Section II.B.2.a.ii (p. 35) and a sufticierit number
of samples in quartz or horosilicate glass tubes (Si ’t ion
II .A. 2.d , p. 23) to perform all the required testy;. M•.p;1r
initial concentration of test chemical (C 0 ) as described in
Section II.A.1l.a (p. 34) in duplicate. Fill the tubes is
completely as possible and seal them. Do not use grease.
Prepare two control samples in the absence of ultraviolet light
and totally exclude light by wrapping the tubes with aluminum
foil or by any other suitable method [ Section II.A.R, p.27]
1 ‘ )
_I’ —‘
-------
CG—6010 (October, 1984)
Place the samples, including the controls, in the Ace—type PMGRR
with a 450 watt medium pressure mercury 1.clinp and appropriatt
filters as described in Sect ion It . A. 2 .. . i (p. 1 ) . 1 ii . .\.
c (p. 22) . The re i c t ion tubes shou 1(1 he con t: roE ted I
temperature of t 1 ± 2°C within the range 20 — 30°C. Photolyze
the samples at 313 or 366 nm corresponding to the higher value of
the molar absorptivity of the test chemical (c 313 or
determined in the Phase 1 experiments, Section II.B.1, p. 35)
until approximately 50% of the test chemical has transformed.
Measure the molar concentration of the test chemical (Ct) in
duplicate at time t. For test chemicals that ionize or
protonate, carry out the photolysis experiments at the required
pHs as described under Test Conditions, Section II.A.6 (p. 26).
After the photolysis experiments are completed,
determine the concentration of test chemical in the controls in
duplicate. If a significant loss of test chemical has occurred
in the controls, determine the cause arid eliminate, or minimize,
the loss. If hydrolysis is found to be significant, hydrolysis
experiments should be carried out first [ Section II.A.8, p. 2711
The data obtained in this sectidn will he used to
determine an approximate kpc and (tV 2 c using equations 3 arid 5,
respectively.
39
-------
CG—601() (October, 1984:
h. P ctinoinetry Experiments
These experunent.s are ties iqneti t cho s’ the
appropriate low optical density actinometer [ i.e., PN AP/PYR,
Section II.P .l0.a.i (p. 29) 0rPNA/PYR, Section ll. .lfl.a.it, p.
311 and to adjust the rate of the chosen actinometer so that the
rate constant of the actinometer (kpa) is approximately equal to
the rate constant of the test chemical (k ). Based on the
photolysis experiments in Section II.B.2 (p. 37), if (tly 2 )c is
less than 12 hours, use the PNA/PYR actinometer; f ti/ is
qreat:er than 12 hours, use the PN P/PYR act inometer.
i. Preparation of ctinotneter Solutions
PNAP/PYR Actinometer : Using the test chemical
photolysis rate constant determined in Section lI.B.2.a (p.
37) and equations 18 or 19, determine the molar concentration of
pyridine needed to adjust the rate constant of the actinorneter to
approximately equal the rate constant of the test chemical at the
appropriate wavelength 313 or 36 nm (chosen based on the results
of the Phase 1 experiments) . Once the molar coneent rut ion t
pyridine [ PYR] has been estimated, the uct inoinet ( ru 1 it l ol l
he prepared as follows. Dissolve 0.0l 5 grams of PNAP iii i() nt.
of acetonitrile (0.01 M). Md 1 mL of this solJti’)n to u
liter volumetric flask. Add to the volumetric flask the muss 1
grams, or the volume V of pyridine at 20°C, obtained from the
equat ions
11
40
-------
CG—6010 (October, 1984)
mass (grams) = 79.1 IPYR1 (23)
V(mL) = S0.6 [ PYR1 (24)
Fill the volumetric flask with pure water (Section TT.A.4, p. 2 )
to give one liter of solution and shake vigorously to make sure
that the solution is homogeneous. The resultant concentration of
PNAP is 1.00 x l0 M. The PNAP/PYR solution should he wrapped
with aluminum foil and kept from bright light.
PNA/PYR Actinometer : Using the test chemical
photolysis rate constant determined in Section II.B.2.a (p.
37) and equations 20 or 21, determine the concentration of
pyridine needed to adjust the rate constant of the actinoineter to
approximately equal the rate constant of the test chemical at the
appropriate wavelength 313 or 366 nm (chosen based on the results
of the Phase 1 experiments) . Once the molar concentration of
pyridine has been determined, an actinometer solution can be
prepared as follows. If photolysis experiments are to he carried
out at 366 om, dissolve 0.0153 grams of PNA in 10 mL of
acetonitrile (0.01 M) . Add 1 mL of this solution to a one liter
flask. Add to the volumetric flask the mass in grams, or t:he
volume of pyridine at 20°C as defined by equations 23 mi1 24.
Fill the volumetric flask with pure water (Section H.A.4, p. 2’)
to give one liter of solution and shake vigorously tO mfl 1k( sur(
41
-------
C( h ) 1 u c t “he , I 4
that the solut ion is homogeneous. The resii 1 taut c’ent r. t i o i
1.00 x 10 M. The PN /PYR solution should he wrapped with
aluminum foil and kept from bright light.
If photolysis experiments are to he carried out a
313 nm, dissolve 0.00612 grams of PNA in 11) mL of acetonitrile
(0.004 M) . Then follow the procedure described above. The
resultant PWA concentration is 0.400 x 1O M. The P A/PYR
solution should he wrapped with aluminum foi and kept from
hr i’;ht light
ii. Performance of the Test
Prepare a solution of the appropriate actinometer
as described in Section II.B.2.b.i (p. 40). Measure the initial
molar concentration of actinometer C 0 [ PNAP or PNA as described in
Sections II.A.11.b, c (p. 34)1. Prepare a sufficient number of
samples in borosilicate or quartz tubes [ Section II.’ .2.d (p.
23)1 to perform all the required tests. The tubes should he
ident ical to the tubes used to deterini ne (Sect ion
)
It • B .2 .a . iii , p. 38) . Fill the tubes as completely as poss i ble
and seal them. It is important that the act i noineter tubes
contain the same volume as that used in the test chemical tubec
(Section II.B.2.a.iii, p. 38). Prepare two control samples in
the absence of ultraviolet light and totally exclude light by
wrapping the tubes with alurninimum foil or by any other suitable
method [ Section II.A.8, p. 27] . Place the samples, including t:he
controls, in the Ace—type PMGRR with the 450 watt rn ’- diuin resume
mercury lamp and the appropriate filters. The tubec ;hould be
42 1 ‘
-------
CG—6010 (October, 1984)
controlled to a temperature t ± 2°C. Photolyze the samples at
the chosen wavelenqth A (313 or 366 nm) until approximately
of the act inometer has transformed. Determine the concentrat ion
of the actinometer (Ct) in duplicate by the appropriate procedure
described in Section II.A.ll.h, c (p. 34) at time t. Calculate a
rate constant using equation 2. If the rate constant kpa IS not
approximately the same as the rate constant of the test chemical
then repeat the experiment with an appropriate concentration
of pyridine until kpa is approximately equal to kpc (within ±
50%).
3. Phase 3 Fxperiments: Determination of the Reaction
Ouantucn Yield of the Test. Chemical
Based on the results of the Phase I and 2 experiments,
use the appropriate actinometer with the required concentration
of pyridine to make the rate constant of the actinometer
aproximately match the rate constant of test chemical. Follow
the procedure outlined in Section II.B.2.h.i (p. 40) to prepare
the actinometer and a sufficient number of samples in
horosilJcate qlass or quartz tubes [ Section II .A.2.d (p. 23)1 to
perform all the required tests. Determine the initi.il
concentration of the actinometer (Co)a in duplicat:e acc()r(lin’; to
the appropriate procedure in Section [ I.A.ll (p. 34). Fill all.
the tubes as completely as possible, seal them, and cover half ol
the tubes with aluminum foil as soon as possible after
preparation. Prepare an aqueous solution of test chemical as
described in Section II.B.2.a.ii (p. 38) and determine the
43
-------
CG—6010 (October, ln 4
initial concentration of test chemical (C 0 ) in duplicate
according to the procedure described in Section II.A .ll.a (p.
34). Prepare a sufficient number of samples in borosilicate or
quartz tubes to perform all the required tests. Fil’ all the
tubes as completely as possible, seal them, and cover half of the
tubes with aluminum foil as soon as possible after preparation.
The reaction tubes for actiriometer and test chemical must he
identical and contain the same volume of solution. Place the
tubes to he photolyzed in the Ace—type PMC’,RR containinq a 45(1
watt medium pressure mercury lamp and the appropriate filters,
Sections II.A.2.b,c (p. 22), and all the control tubes close to
the PMGRR. The tubes should he temperature controlled to a
temperature t ± 2°C. Based on the results of the Phase 2
experiments, determine the concentration of test chemical and
actinometer periodically between 10 and 80% transformation in
duplicate [ at least 6 data points at approximately equal.
times) . Determine the concentration of act inometer and t:est
chemical in the controls in duplicate at each time point.
C. Procedure Two: Determination of the Reaction Ouantum
Yield by the r w Optical Density Test Chemical and Hi h
Optical Density Actinometer Method
1. Phase 1 Experiments:
Follow the procedure outi [ ned in Sect ion TI •K. I (p.
35) and measure the absorhance of the test chemical, A , a
Ac
function of wavelength in the range 290—800 nm in duplicate.
Measure at 313.0 and 366.0 run in duplicate. The ahsorbanck
1: ‘
44
-------
CG—6010 (October, 1984)
data at 313.0 and 366.0 nm will. he used to calculate the molar
ahsorptivities of the test chemical, c and , and
3Hc 66c
photolysis experiments should he carried nut at the wavelenqth
corresponding to the highest value ot the molar ahsorp iv i ty.
2. Phase 2 Experiments: Determination of the Reaction
Quantum Yield of the Test Chemical
a. Preparation of Buffer Solutions
Prepare buffer solutions according to the
procedures outlined in Section II.B.2.a.i (p. 37) for a test
chemical which ionizes or protonates.
h. Preparation of Test Chemical Solution
Prepare a homogeneous solution of test chemical
according to the procedures outlined in Section Il.B.2.a.ii (p.
38).
c. Preparation of Ferrioxalate Actiriometer
Solution
Prepare a ferrioxalate actinometer solution at 0.15
M as described in Section iI.A.10.b (p. 33) followinq the
procedure outlined by Murov (1973) which has been nodit ted
S1i ht1y. Since the ferrioxalate actinorneter solition i ; v’ ry
sensitive to visible light, all these actinometer experiments
must be carried out in a dark room with photographic “safe
lights.”
i )
45
-------
CG—6010 (Dctober, 1984)
d. Determination of the Ouantum Yield of
Test Chemical
Prepare a so lut ion of the ferrioxa late act i u()mete r
at high optical density (absorbance > 2) as described in Section
II.C.2.c (p. 45) and completely fill the reaction cell described
in Section rr.A.2.d (p. 23). Place the reaction cell in the
photochemical optical bench (POB) as described in Section
tI.A.2.a.ii (p. 21) with a 450 watt medium pressure mercury lamp
(Section rI.A.2.h, p. 22) and appropriate filters (Section
11 • P . 2 • c , p. 22 ) . The react ion cell shou Li he cou t r ) I 1 el t 0 a
temperature of ± 2°C in the range 20 — 30°C. The litter
system used should isolate the wavelength 313 or 366 urn
corresponding to the larger value of or C 366 c determined
in the Phase 1 experiments. Photolyze the actino neter solution
in the reaction cell for 1, 3, 6, 9, 12, and 15 minutes and
measure the concentration of Fe 2 + formed spectrophotometrically
by the procedure outlined in Section II..ll.d (p. 35) in
duplicate at each of the time points. No time points should be
tdken when the optical (lens ty of the act i noenet:e r t a I s be 1 OW
2. Only withdraw small volumes for analysis so that tI total
volume of the actinometer solution in the reaction c I {j()e ; not
change appreciably. This procedure is repeated in a third set
experiments soon after the test chemical photolysis experiinent:s
are performed.
In the second set of experiments, a solution of
test chemical is prepared according to the procedure outlined in
46
-------
CG—601() (October, 1984)
Section [ I.C.2.h (p. 45) at low optical density (ah rhancc
<0.02) . Measure the in it ia c mcent rit I n ( in tupi I cit c hv
the procedure outlined in Section IT.A.I .a (p. 34). 1i11 tiwc,
reaction cells with the same volume as that used in the
actinometer experiments. One of these cells is wrapped with
aluminum foil, placed close to the POB and temperature controlled
to tj ± 2°C. The second reaction cell (which is the same one
used for the actinometry experiments) is placed in the POB and
photolyzed at t ± 2°C at the same wavelength used in the
actinometry experiments. Determine the concentration of test
chemical periodically between 10 and 80% transformation in
duplicate Eat least data points at approximately equal
times 1 . Only withdraw small volumes for analysis so that the
total volume of the test chemical solution in the reaction cell
does not change appreciably. Determine the concentration of the
control in duplicate at each time point.
As an alternative procedure, the same series of
experiments described above can be carried out in a PMGRR using
individual reaction tubes for each datum point.
III. DATA AND REPORTING
A. Procedure One: Determination of the Re Li.r I)u.uulii,ri
Yield by the L/Dw Optical Density Test Chernilind
Actinorneter Method
1. Phase 1 ExperimentS: ¶JV—Visihle Ahsorption j ectrd
a. Treatment of Results
47
-------
CG—60U) (Octeber, 984)
The molar absorptivity (e ,) can be determined
from the absorption spectra using the expression
= Axc/C (25)
where is the absorbance at wavelength A, C is the molar
concentration of test chemical, t is the cell pathlength in
centimeters. The molar absorptivity of the chemical should be
determined for the wavelengths listed in Table I, rest Guideline
CG—6000, for solution of concentration C and in i c lJ with
pathlength t . If the absorption curve is flat within the
interval around the wavelength center, may be determined
from the absorbance at x center using equation 25. If a
large change in absorbance occurs within this interval, obtain an
average at A center based on the two absorbances at the
boundary of the interval. Calculate an average c c using
the average value of in equation 25. Determine the molar
ahsorptivity for each replicate and calculate a mean value.
Deterrni ne the molar absorpt iv i Ly ot: the Lest
chemical at 313.0 and 366.0 nm for a so hit ion ot L s L diem i c I i
a molar concentration C and in a cell of pathlencjtih Q USiIlj
equation 25. Determine and the molar ahsorptivity at 3I .()
and 366.0 nrn for each replicate and calculate a mean value.
Using the molar absorptivities obtained from the
spectra and the values of from Table 3 to 6 of Test
Guideline CG—6000, calculate the maximum direct photolysis rate
constant (kpE)max at a specific latitude (corresponding to the
48
‘,/_‘ ‘_)
-------
CG—60 10 (C ctc)ber , lq 4
manufacturing site) and season of the year using equation 16.
The corresponding minimum half—life, (t )mjn can then he
calculated using this (kpE)max in equation 17.
h. Test Data Report
(1) Report the name, structure, and purity of the test
chemical.
(2) Submit a recent test spectrum on appropriate
reference chemicals for photometric and wavelength
accuracy.
(3) Submit the original chart, or photocopy, containing
a plot of absorbance of test chemical vs.
wavelength plus the baseline. Spectra should
include a readable wavelength scale, preferably
marked at 10 nm intervals. Each spectrum should he
clearly marked with the test conditions.
(4) Report the concentration of the test chemical
solution, the type of absorption cell used (quartz
or horosilicate glass) and the pathlength.
(5) Report and at A center for each replicate
and the mean value.
(6) Report the identity and composition of the solvent
used in the spectral absorption study.
49
-------
CG—6010 (October, 1984)
(7) Report C 313 and C366c along with A 3 i 3 and A 366
for each replicate and the mean value.
8) Report (kpp)mjx ani ( ti,. ..) 111 tc r th ’ ;u nmer iu i
winter si)lst & ’ 4 ii. in I lit’ I i r( pr1iI ‘ V.I lu ’ ;
from Tables 3—6, Test u i&1t 1 ne ( — ()1) ) , Cl ) - !St I C
the latitude of the chemical manufacturing site.
(9) For chemicals that ionize or protonate, report the
data for steps 1—8 at the required p1-Is [ Section
II.A.6 (p. 26)1.
(10) For a chemical that ionizes or protonates, report
its PKa* Report the type and concentration of the
buffers employed for each pH.
(11) Describe the method employed in determining the
test chemical’s concentration.
12) Report the name and model of the spectrophotometer
used.
(13) Report the various control settings employed with
the spectrophotometer. These might include scan
speed, slit width, gain, etc.
(14) If a Moses—type PMGRR was used in these
experiments, describe it completely and report the
aperture in cm 2 .
2 ,-I
50
-------
(—bOlU (oct her, l 4
2. Phase 2Exj eri’nent:s: rn il Pot ‘l I r i men t
a. Determination of the Approximate Rate ‘onst int
of the Test Chemical
i. Treatment of Results
From the photolysis experiments carried out at 313
or 366 run (photolysis wavelength chosen based on the results of
the Phase 1 experiments), use the concentration (Ct)
corresponding to approximately 50% of the initial molar
concentration of chemical remaining at time t along with the
initial molar concentration and the time t in hours in
equation 3 to calculate in hours . From the analysis of the
two samples at time t = 0 and t, calculate a mean value of (C 0 )
and (Ct)c, and a value of If a slight loss of chemical has
been detected in the controls, then calculate a rate constant as
follows. Calculate an average concentration (Ct)c based on
duplicate measurements of concentration in the controls at the
end of the experiment (time t) . Use this concentration along
with the average initial concentration (C 0 ) anti t in i’quat ion
and calculate a rate constant k 1055 . Using this rate constant
along with the observed rate constant in the photo] y.s is
experiments, the corrected rate constant is then
= (kpc)obs — kloss (:26)
Calculate the half—life, (t >,i using the corrected value of k 1 )(
in equation 5.
)f r—
I ’ ; ,
c i
-------
CG—6010 (October, 1984
ii. Spec it 1C TLfl yt I .1U 1 1’Ľ V( 1\
(1) Provide a tiet: ii led descr ion or r’t o cuco tor t o’
analyt ical procedures used, inc 1 ud I n i t ho
calibration data and precision; and
(2) if extraction methods were used to separate the
solute from the aqueous solution, provide a
description of the extraction method as well as the
recovery data.
iii. Other Test Conditions
(I) Report the size, approximate ceU wiM thickness
pathlength 2. , and type of qiass ust’d for the to t
chemical reaction tubes. If the ceU patThlonqth
was measured by the procedure described in the
Technical Support Document, Section 11.1), report
all the data obtained in these experiments.
(2) Report the initial pH of all test solutions, if
appropriate.
(3) If acetonitrile was used to soluhilize the test
substance, report the percent, by volume.
(4) If a significant loss of t.f? -t cto ni ’if mirt •i iu
the control solution, md icate th r;’ m l how
they were eliminated or minimized.
‘S
‘ .
52
-------
CC—6( lO (C ct her ,
iv. Test Data Report
(1) Report the wavelength used to photolyze the test
chemical.
(2) Report the initial molar concentration of test
chemical (C 0 ) of each replicate and the mean
value.
(3) Report the molar concentration of test chemical
(Ct)c for each replicate and the mean value.
Report the time t.
(4) Report the molar concentration of each replicate
control sample and the mean value after completion
of the experiment. Report the time t.
(5) Report the value of and (t] c• If small losses
of chemical were observed, report (kpc)ohc ,
and kpc Report the half—Life (tI,, 2 calculated
using the value of kpc .
(6) For chemicals that ionize or protonate, report the
data for steps 1—5 for the experiments at the
required pHS. Report the initial pH of all test
chemical solutions and the type and concentration
of the buffers used for each pH.
-------
C—b ) 1 C) ( ‘ )C t oho r , 1 4
h . ! c t i ne I ry Kx p er i men t
i . Tretiment. of Re ;ul t-
Follow the same cli scuss U)n of the t re. t me nt of
results, Section Itl.A.2.a.i (p. 51), to determine the rate
constant of the actinometer kpa using equation 2. Repeat these
calculations for the trial actinometry experiments to determine
the molar concentration of pyridine EPYR1 needed to adjust the
rate constant of the actinometer (kpa) to be approximately equal
to the rate constant of the test chemical (k ).
ii. Test Data Report
(1) Report the wavelength used to photolyze I he
actinorneter.
(2) Report the size, approximate cell wall thickness,
pathlength Ł , and the type of glass used for the
actinometer reaction tubes.
(3) For each trial experiment:
(a) report the actinorneter used;
(h) report the initial concentration of
actinometer (C 0 ) 1 for each rep] ic t-e and the
mean value;
(c) report the molar concent rat: ion of I
actinometer (Ct)a for each rep] icat :e and I
mean value. Report the time 1;
(Cl) report the molar concentration of each
replicate control sample nd the mean value
after the completion of the experiment.
Report the time t; and
-------
CG—60l0 (October, 1984)
(e) report the value of kpa. [ f small losses
were observed for the act inoineter, report
k 1 ot)S , k and k 1 ) 1
(4) II the Moses—type PMGRR was used, list the criteria
and equations used as described in Section
II.1.11.c of the Technical Support Documeuit CS—
6010.
3. Phase 3 Experiments: Determination of the Reaction
Quantum Yield of the Test Chemical
a. Treatment of Results
objective of this set of experiments is to
react ion quantum yield for a sped lie test cheinica l
optical density with a low optical density
The reaction quantum yield can he calculated
= (kpc/kpa)(Cxa/Cxc) a (1)
by the following steps.
(1) Photolysis experiments are carried out: by
Simultaneously photolyzing the test chemical and actinLnnet r In
the PMGRR at the chosen wavelength x . The corlcentratil)rl ot
test chemical and actinometer are measured periodically as a
fu ctjon of time. These data are then used to determine the
The
determine the
at low
act i name te r.
using equation 1,
55
2Y)
-------
CG—6010 Dctc ber,
ratio of the rate constants (kpc/kpa) employing linear regression
analyses of the data on equation 6,
fl (Co/Ct)c = (kpc/kpa)Lfl(Co/Ct)a (6)
with fl(Co/Ct)a as the independent variable and n (C 0 /C )
as the dependent variable. The slope of the best straight Iine
is the ratio of the rate constants (kpc/kpa)* If a stight loss
of test chemical or actinoineter was detected in the controls at:
any time t, then employ the following procedure. Consider, as an
example, the loss of test chemical in the control at time t.
Using the average concentration of the test chemical fri the
controls from the replicates at time t and the average initial
concentration, calculate n (Co/Ct)closs. Using the average
concentration of test chemical from the replicates after
photolysis time t, calculate n(C 0 /C 1 ) S . The corrected
term is then
= tn(c/Ct)0h — . (27)
The same procedure can be applied to obtain a corrected term for
the actinometer. Using the corrected terms for test chemical
and/or actinometer in equation 6, determine the ratio of the rate
constants (kpc/kpa) as described above.
56 2 1
-------
cc ;—6O 10 (Oct c be r , I 4 4
(2) Determine the quantum yield of the actinometer
using equation 7 or 8 and the molar concentration of pyrUiine
[ PYRI present in the actinometer.
(3) Use the molar absorptivities of test chemical
and actinometer xa at the wavelength the photolysis
experiments were carried out (i.e., 313 or 366 rim).
(4) Substitute the values of (kpa/kpc) CXa and
in equation 1 and calculate the reaction quantum yield of
the test chemical
C
A hypothetical example is presented in the
Appendix, Section V., to illustrate how all the data can be used
in Procedure One to determine the reaction quantum yield of the
test chemical and to determine (kpE) and (tj 71 ) as a function of
latitude and season of the year.
b. Other Test Conditions
(1) Report the size, approximate ceU wait
thickness, the pathlerigth z , and type ot glass
used for tubes used to hold the Lest chemical
and actirionieter solutions.
(2) Report the initial pH oF all test chemical
solutions, if appropriate, and the type and
concentration of the buffers employed for each
pH.
21
57
-------
CG—60l.U (Oc oer, i 4)
3) If ace toni I r i le was u -eti t so I ub i I i i. ’ h ’ t ;
chemical., report the percent, by vo um ’ , ot t h’
acetonitrile which was used.
(4) If significant loss of test chemical occurred
in the control solution, indicate the causes
and how they were eliminated or minimized.
c. Test Data Report
( I) Report the wavelength used in the photolysis
experiments.
(2) Report the actinometer used.
(3) Report the initial molar concentration of test
chemical (C 0 ) of each replicate and the mean
value.
(4) Report the initial molar concentration of
actinometer (Co)a and the molar concentration
of pyridine used.
(5) For each time point, report the two separat: .?
values for the molar encentration t)t ti 5
chemical (Ct) and actinometor nd 1i
mean values.
(6) For each time point, report the two sep r. Le
values of the molar concentration of test.
chemical and actinometer controls and the mean
values.
58 22
-------
CG—6010 (Octoh r, 1984)
(7) Tabulate and report the tot towing data:
t, n(C 0 /C 1 ) , and Qn(C 0 /Ct) . ‘r ui tHte
linear req ress ion analysis, rt-’j)ort t ti ’ rat i
the rato constant.s ( , and I ht’
correlation coefficient.
(8) If loss of test chemical and/or actiriometer was
observed during photolysis, then report the
data Łn(Co/Ct)co •, n(C 0 /Ct)0bs.,
and tn(Cc/Ct) 1 05S for the test chemical
and/or actinometer for each time point. From
the linear regression analysis of test chemical
and/or actinometer
Łn(Co/Ct)a)r• , report the ratio of the rate
constants (kpc/kpa) and the corre lit nn
coefficient.
(9) Report the reaction quantum yield of the
actiriometer
(10) Report the reaction quantum yield of the test
chemical
(11) Report (kpE)max.I (kpE)l (tl/ )miri .1 and (t )
for the summer and winter soistices using the
appropriate L values from Tables 3—(, ‘l e st
Guideline CG—6000, closest to th I ; t I wi” t
the chemical manufacturing site.
(12) For chemicals that ionize or protrniate, repert
the data for steps 1—11 for the experini nts at
the required pHs [ Section 1I.A..6 (p. 26)1.
2 ’
-------
Cc—6C) C ( c : e r , I - -
B. Procedure Two: Deterrnina ion of the Re t i o u u t
Yield
Opt ical Densi tinomerMeth d
1. Phase 1 Experiments:
The treatment of results and data reporting are e actY y
the same as described in Section Itl.P .l (p. 47)
2. Phase 2 Experiments: T)eterminationof the Reaction
Ouanturn Yield of the Test Chemical
a. Treatment of Results
The objective of this set of experiments is to
determine the reaction quantum yield of a tt st t’hemjcil ( p) at
low optical density with the high optical density ferrioxilat:e
actinometer. can he calculated using equation 9,
= a (kpc/kpa)(2 303CAc ) ’ (9)
by the following steps.
(1) In the first set of experiments, th. - irrioxite
actinometer is photolyzed in a POh at h
wavelength A (313 or 366 nm) ani t ie welir
concentration of ferrous ion (C ) H n ;iir d i: a
function of the time t. These data are fitted t
equation 11 using linear regression analysis. TFi
slope is equal to kpa. The data obtained from the
60 2 1
-------
(‘( —60 0 (Oct r , I 44
third set of experiments are used in the same
manner as described above to obtain another
actinometer rate constant. These two rate
constants are used to obtain an average actinometor
rate constant (kpa)ave.•
(2) Tn the second series of experiments, the Lest
chemical is photolyzed in the same cell in the POB
and the concentration of test chemical (Ct) is
measured as a function of the time t. These data
are fitted to equation 3 using linear regression
analysis. The slope is equal to If a slight
loss of test chemical is detected in the controls,
then follow the procedure outlined in Section
II.A.2.a.i (p. 19) to calculate a corrected test
chemical rate constant.
3) Use the appropriate quantum yield of the
ferrioxylate actinorneter (at 0.15 M) at the chosen
wavelength: at 313 nm, a l.20 at 366 nrn,
= 1.1 5.
a
(4) Use the molar ahsorptivity of the Lest chemical
(c ) at the appropriate photolysis wavelength
(313 or 366 rim) determined in Section IIl.B.l (p.
60)
(5) Use the known pathlength of the re.ic ion 1 Q..
Section ii.A .3 (p. 25).
)
-------
CS—6010 (October, 1984)
on the actinometer ) , and the molar absorptivity oF the
act inometer x a • The pert inent equaL ions for the ow opt: i c I
dens I ty act i nometo r are oqua t. ions 4 1 and 20 . I or Pr cedut e ne
it is necessary to derive equations under the condit ion that: the
rate constant of the actinometér equals the rate constant of the
test ch mica1; that is,
k = k (57)
pa PC
Substituting equation 41 in equation 57 yields
= 2303 ’a x xa• . (‘5 3 )
Substituting equation 41 in equation 20 yields
y’ a= 3•01Xl0 1 L aI tC aI’
These equations are perfeätly general and are applicable for use
in a POB or PMGRR using any light source filtered to yield a
monochromatic wavelength band x . Equations will be derived for
a typical 450 watt medium pressure mercury lamp in an Ace—type oi
a Moses—type PMGRR. The pertinent information has been taken
from the paper by Dulin and Mill (1982), the data from tJori,
u.C.l .a and b, and from Mill (1984).
2T..’
62
-------
CS—6010 (October, 1984)
a. P_rs4itroacetophenone_pyridi,le Act.inomet:et
(PNAP/PYR) in an Ace—type PMGRR.
For the PNAP/PYR actinometer, the relationship o the
reaction quantum yield as a function of the molar
concentration of pyridine [ PYR] is given by equation 30
= 0.0l69 [ PyR] . (30)
For the 450 watt medium pressure mercury lamp in an Ace—type
PMGRR used by Mill and Dulin (1982) the approximate average light
intensity at 313 nm and 366 nm was: 1(313) = 3.0 x io6
einsteins sec L 1 ; 1(366) = 7.5 x l0 eins tejns sec’ L’.
Conversion of these intensities to hours yields 1(313) 1.08 x
10—2 einsteins hr. L 1 and 1(366) = 2.70 x 10—2 einstejns hr.
L . For the cells used by Dulin and Mill (1982), =01.0
[ Mill (1984)1. The molar absorptivitjes of PNAP at 313 and 366
1 _1 1 1
nm are: = 2056 M cm and = 160 M cm (Section
1I.C. 1.a)
At 313 nm, the following relationships CU)
calculated. Substituting equation 30 in equation 5U m d u inj
the above pertinent data:
—2 -
= 2.303(1.69 x 10 [ PYR])(1.08 x 10 )(1.0)(2.056 x 10
kpc = O.864 [ PYRJ (60)
63
4
-------
C(—bt) 10 tIct h’r , 1 Ľ l 4
(3) For the test chemical photolysis experiment;
(i) report the initial molar concentration of
chemical (C 0 ) 0 of each replicate and the mea:
value;
(ii) report the two separate values of the moLir
concentration of test chemical (Ct)(.. and the
mean value at time t. Report the time L;
(iii) report the two separate values of
the molar concentration of the test chemical
controls and the mean value for each time
point. Report t for each time point; and
(iv) report from the linear regression analyses of
the data, the rate constant of the test
chemical (k 0 )l and the correlation
coefficient.
(4) Report the reaction quantum yield of the act nometer
and
(5) Report the react ion quantum yield of the test
chemical ( ).
(6) Report (kpF)maxi (kpE)l (t1,/ )min.i nnd (tl/ ) for
the summer and winter solstices using the
appropriate Lx values from Tables 3—6, Test
Guideline CG—6000, closest to the latitude of tht’
chemical manufacturing site.
(7) For chemicals that ionize or proton it , r(p()r
data for steps 1—6 for the experunen n l he
required pHs (Section II.A.6 (p. 2 )I
64
-------
C —6UlO ( )ct t ?r, L ) 4
l v. REE RKNCI S
Andre JC , Nictause M, Joussal—I)ubion .1, and I)e l sc 1 - 7 7.
Photodegradation of pyridine in aqueous soition. J Chem l d
54:387.
ASTM. 1983. Annual Book of ASTM Standards, American Society tor
Testing and Materials, Philadelphia, PA. Part 31, Method D 1193-
77.
Calvert JG and Pitts JN. 1966. Photochemistry. John Wiley and
Sons, Inc., New York, New York.
de Mayo P and Sh izuka P. 1976. Measurement of react ion duantum
yields. “Creation and Detect ion of the Excited State,” Vol. 4,
A.R. Ware, Ed. Marcel Dekker, Inc., New York, New York.
Dul.in D and Mill T. 1982. Development and application of solar
act.inometers. Environ Sci and Tech 16:815.
Handbook of Chemistry and Physics. 1983. The Chemical Rubber
Co., Cleveland, Ohio.
Hatchard CG and Parker CA. 1956. A new sensitive chemical
actinometer II. Potassium ferrioxylate as a standard chemical
actinometer. Proc Royal Soc of London A 235:518.
Mill T, Davenport JE, Dtj tin J)E, Mabey WR, and Rawol R. I I
Eva luat ion and opt iini zat ion of photolys is screen i n j proL)COU;
EPA—560/5—81—003.
Mill T, Mahey WR, 1-Jendry DG, Winterle J, Davenport JE, Karicli V 1
Duliri D, and Tse D. 1982. Design and vat idit ion ot cro ’ni n; and
detailed methods for environmental processes. EPA-
Mill T, Mabey WR, Boinberger DC, Chou T—W, }-lendry D(, and Srnit:h
JH. l982a. Laboratory protocols for evaluatinj the fate of
organic chemicals in air and water. EPA—600/3--82--022.
Murov SL. 1973. Handbook of photochemistry. Marcel Dekker,
Inc., New York, New York.
65
-------
CG—6tJ10 )ctoeec,
Moses ‘G , Li u RSII , and Mon roe I M • q 69 • The “tier ry —qo-— round
quantum yield apparatus. Mol Photochein 1:245.
Parker CA. 1953 • A new sens i I i ye chein i c I ic ti i noinet or I . Se ine
details with potassium ferrioxylale. i’rec Ray Soc of Lendm
A220:104.
Zepp RG. 1978. Quantum yields for reaction of pouutants in
dilute aqueous solution. Environ Sci and Technol 12:327.
2’ ’)
66
-------
CG—6010 (October, 1984)
V. APPENDIX
Hypothetical Illustrative Exainple: Determination af th
Reaction Quantum Yield by the Low Optical Dens t Test
Chemical and Actinometet- Method and Sunlight Phot:olysis
Consider a chemical plant just south of PeOria, Illinois on
the Illinois River which produces an organic chemical A which is
not an acid or a base. The waste effluent passes through a
primary and secondary treatment plant and the waste, Which still
contains some chemical A, is then discharged into the river. The
plant is located at 40.7 degrees north latitude. Information is
needed on photolysis rates and half—lives of chemical A in
aqueous media in the summer and Winter seasons.
The company reseach laboratory is located in Peoria,
Illinois and the required photolysis data was needed in
January. Since the temperature outdoors was well below freezing
during this month, the outdoor sunlight photolysis experiments,
described in Test Guideline CG—6000, could not be carried out.
Thus, it was necessary to carry out photolysis experiments in the
laboratory. The research laboratory was equipped to carry out.
photolysis experiments with an Ace—type PMGRR arid hus I’r c dt4r*
One was used.
9’ 1
d. t..
67
-------
(‘(;—tS(l 1 ’ (t ct - ,
• , C 1 .‘ u 1 .
1. Phase 1 Experiments
Chemical Pk had a saturated water soluhility of 3.9 x
M at 25°C. In the Phase I procedure 1 the uv—visihle
absorption spectrum was obtained for chemical at a
concentration of 1.00 x 10 M in a 10.0 cm quartz cell in
duplicate. flsing the wavelength interval range from Table I o1
TeSt- (;u ide line CG—6000 the average a orhance t iu ’ 1 c te run-;
at A center was obtained and the re u I t.s are nuininuri ,‘.‘&i in TaHb’
1. In addition, the average absorhance was rneanireti at lL) un
and at 366.0 rim and these results are summarized in Table 1.
From the above data and equation 25, the average molar
absorptivity is
= 1O (28)
Xc Xc
must he emphasized that has to he averaged over th
wavelength intervals that correspond to the same intervals for
the values centered at A . This has already been taken
care of in this Test Guideline since the wavelength interval
ranges listed in Table 1 coincide with the same wavelength
intervals for L centered at A in Tables 3—6 of Test
Guideline CG—6000.
‘)- ‘_)
68
-------
CG—6010 (October, 1984)
297.5
300.0
302.5
305.0
307.5
310.0
312.5
315.0
317.5
320.0
323.1
330.1
340.0
350.0
360.0
370.0
380.0
390.0
400.0
410.0
313.0
366.0
0.5904
0.5414
0.4924
0.4434
0.3810
0.3124
0.2714
0 . 2224
0.2004
0.1514
0.1310
0.1030
0.0645
0. 0473
0.0287
0. 0 I. 70
0.0085
0. 0030
0.0010
0.0000
0.2610
0.0185
5904
5414
4924
4434
3810
3124
2714
2224
2004
1514
1310
1030
645
473
287
E •7()
85
3(3
1(3
U
TABLE 1.
A center (nm)
SUMMARY OF SPECTRAL AND PUOTOLYSIS DATA FOR CHEMICAL A
A. SPECTRAL DATA
A (M cm
Ac X c
2610
185
69
-------
C —6U 10 (Oct be
TABLE 2. Continued
297.5
300.0
302.5
305.0
307.5
310.0
312.5
315.0
317.5
320.0
323.1
330.0
340.0
350.0
360.0
370.0
380.0
390.0
400.0
410.0
B. PHOTOLYSIS DATA
6.17(—5)
2.70(—4)
8.30(—4)
1 .95(—3)
3.74(—3)
6.17(—3)
9.07(—3)
1.22(—2)
1.55(—2)
1 .87(—2)
3.35(—2)
1.16(—1)
1.46(—1)
1 .62(—1)
1.79(—1)
1.91(—1)
2.04(—1)
1.93(—1)
2.76(—1)
3.64(—1)
(day )
0.4
1.5
4.1
8.7
14.3
19.3
24.6
27.1
31. 1
28.3
43.9
119.5
94.2
76.6
51.4
32.5
17.3
5.8
2.8
0.0
1*
5.49(—7)
5.13(—6)
3.02(—5)
L19(—4)
3. 3R(—4)
7.53(—4)
1. 39(—3)
2.22(—3)
3 .] 9(—3)
4.23(—3)
8.25(—3)
3.l6(—2)
4.3 1(—2)
4.98(—2)
5.68(—2)
6.22(—2)
6. 78 ( —2)
6.33(—2)
9.1 1(—2)
l.20(—1)
0.0
0.()
0.2
• L)
1.
2.4
3.8
4.
6.4
6.4
10.8
32.6
27.8
23.6
16.3
10.6
5.8
9
0.9
0 . 0
603.4
Vc •1 =156.2
L
*The units of
nunber in the
is multiplied.
r. , are in einsteins ai 2 day .
co’lunns in parenthesis is the power of
The second
ten by which the first nurnber
2 d1
A center (nm)
Suiinei
Win t r
LA cl I)
70
-------
CG—6010 (October:, 1984)
From the average value of at x center, the average molar
absorptivity can be obtained from equation 28 and these results
are summarized in Table l.A. In addition, using at 313.0
and 366.0 nm in equation 28, the molar absorptivity
C313c and C366c can be obtained and these values aie also given
in Table l.A.
Since the plant is located at 40.7 degrees north latitude,
the closest values are at 40 degrees north latitude. These
values are obtained from Table 5 of Test Guideline CG—6000 and
are summarized in Table l.B. for the summer and winter
soistices. Using the data from Table l.B. and equations 16 and
17, the following results are obtained.
Summet Solstice Winter Solstice
(kp )max. = )cA.cL = 603 days (kpE)max = = 156 days’
(t1, / )mjn = 1.1 x 10 days (t1ii )min x 10 days
Since the chemical transforms rapidly on the summer and winter
solstices, it is necessary to determine the reaction quantum
yield of chemical A in the laboratory using Procedure One and to
obtain direct sunlight photolysis rates and half—lives in aqueous
media during the summer and winter seasons.
d )
71
-------
CG—hU 10 (Dct cL , 1’ 4
2. Phase 2 Experiments
a. Determination of the Apptoximate Rate Constant
of the Test Chemical
Chemical A was dissolved diiectly in pure water and a
homogeneous solution was prepared. Analysis of duplicate sanip es
indicated that the average concentration was 5.00 x io6 M.
Using the uv spectral data obtained, the absorhance at 313 nm in
a one cm absorption cell containing a solution at a concentration
of s.oo x io 6 M was less than 0.02 [ i.e., A 3 i 3 2610 (5.00 X
1 0 6 )U.00) = 0.011. Hence, the test chemical solut ion in
ri’ ox imate 1 y cm path length tubes was at low oj t a I d-sns i ty
A series of tubes (13 x 100 mm), with an ettective
pathlength of 1.12 cm, were filled with the aqueous solution Ut
test chemical and sealed. Since ‘> Ł (Table 1),
313c 366L
photolysis experiments were carried out at 313 nm in a PMGRR
using the procedure described in Section II.B.2.a (p. 37).
Duplicate photolyzed and control samples were removed
periodically and analyzed for the concentration or test
chemical. At t = 39.4 hours, the average concentration at
photolyzed sample (Ct) was 2.25 x io6 M and tha .IV raJO
concentration of control sample was 4.99 x 10 M. ltew ,
approximately 55% of the test chemical transforin d. nce no
loss of chemical was observed in the control sample, no
adventitious processes occurred and the loss of chemical was only
due to photolysis. Using equation 3 and the above data:
72
-------
CG—6010 (Octohet, 1984)
n(5.00 x 10-6/2.25 x 106) = k(39.4)
k = 0.0203 hours’ (29)
PC
b. Actinometry Experiments
Utilizing equation 5 and the value of kj)c obtained in
equation 29, the half—life of the test chemical is then
(t˝)c = 0.693/0.0203 34.1 houts. (30)
Since (tl > 12 hours, the PNAP/PYR actinometer is requited for
the photolysis experiments.
Using equation 18 and the approximate rate
constant from equation 29, the Concentration of pyridine
needed to make the rate constant of the actinometet approximately
equal to the rate constant of the test chemical is given below:
IPYRJ = 1.16 = 1.16(0.0203)
fPYRJ = 2.36 x 1ir 2 M (31)
An actinometer solution was then prepared accotding to the
procedure given in Section II.B.2.b.i (p. 40) and the
concentration of pyridine was 2.36 x io2 M. The concenttation
of PNAP was measured in duplicate and the average concentrat:ion
(C 0 ) was 0.900 X i0 M. Using uv spectr at (1dt. , ; r hdri
at 313 nm in a one cm absorption cell containing in :t r )u1t’t( r
solution at a concentration of 0.900 x 10 M was less Lh in 0.02
2 2
-------
C —t5d id ) t er:, l’ 4
i . e A 3 1 3a (2056) (0 .900 X 1 0 (1 . 00) 0 . 1)18 ) . d Onc e,
actinometer solution in approximately 1 cm pathien jth tubes wa
at low optical density.
A series of tubes, identical to those used in trie test
chemical photolysis experiments, were filled with the actinornete
solution and sealed. The photolysis experiments were carried out
in the same PMGRR at 313 nm using the procedure described in
Section II.B.2.h.ii (p. 42). Duplicate photolyzed and control
samples were per iodically removed and analyzed for the
concentrat. ion of PNJAP. At 50.6 hours, the aver i ’ rrcentt at er
of photolyzed sample (Ct)a was 0.360 x 10 M and the ver a e
concentration of the control sample was 0.900 x l0 M. Thus,
approximately 60% of PNAP transformed. Since no loss of chemical
was observed in the control sample, no adventitious processes
occured and the loss of chemical was only due to photolysis.
Using equation 2 and the above data:
Łn(0.900 x 1o /O.36O x 10 ) = k (50.6)
k = 0.0181 hours ( 2)
a
Thus, at the molar concentration of pyt i(ilne ji yen by ec ijal i ir
31, kpa is approximately the same as (within ± 50%).
3. Phase 3 Experiments
A solution of test chemical was prepared according tO
the procedure described in Section II.B.2.a.ii (p. 38) and the
concentration (C 0 ) was measured in duplicate arid the aver .ige
74
-------
CG—6010 (October, 1984)
concentration was found to be 5.00 x io6 M. An actinometer
sol Ut ion ( PNAP/PYR) was pr epat (3d aCu)t. ti i FRJ to L ht t &iut
descr ihed in Section I1.R.2.h. i (p. 40) with a py idin ’
concenttation of 2.36 x i0 2 M (equation 31). The a etage
concentration of Pt’JAP (Co)a from duplicate samples was measured
and found to be 9.00 x iO— 6 M. These solutions were placed in
identical tubes (13 x 100 mm), sealed, and photolyzed at 313 nm
in the PMGRR according to the procedure given in Section II.B.3
(p. 43). The average concentration of duplicate samples of test
chemical, test chemical control, actinometer (PNAP), and
actinometet control, obtained in this photolysis exp riment, is
summatized in Table 2.
Table 2. Photolysis Data for Test Chemical A and (PNAP/PYR)
Act inometer
Average l st themical Concentration x10 6 Average PNAP Concentration x i0 6
t(hours) Photolyzed Samples Controls Photolyzed Samples Controls
0 5.00 5.00 9.00 9.00
12 3.92 4.98 7.24
24 3.07 5.00 3
36 2.41 4.99 4.69 9.09
48 1.89 4.98 J.78 0.90
60 1.48 4.99 3.04 3.99
72 1.16 5.00 2.44
75
-------
CG—60U 1)ctoher, 1 4
Since no significant loss of PNAP or test chemicai
was observed in the control samples, no adventitious processes
occurred and the loss of test chemical and P P was only due t:
photolysis. Utilizing the data in Table 2, Łn(C 0 ./C ) for thc
test chemical and actinometer solution can he calculated and the
results are summarized in Table 3.
Table 3.
0
12
24
36
48
60
72
Photolysis Data for Test Chemical and
ctinometer (PNAP/PYR)
5.00
3.92
3.07
2.41
1.89
1.48
1.16
0.000
0.243
0.488
0.730
0.973
1.22
1.46
9.00
7.24
5.83
4.69
3.78
3.04
2.44
1) .
O . 434
0.652
0.868
1.09
1.31
by equation 6,
The ratio of the rate constants (kpc/k [ )l) is (1 flII(d
tn(C/Ct)
= (k /k ) R.n(C /C
PC pa o ta
(6)
t (hours)
i st th ical
t c’ 10 6 (M)
I ctinoTleter (PNAP)
x l0 6 (M) s n(C 0 /C ).
76
-------
CG—6010 (October, 1984)
lit iii z i ng a 1. 1 the data in Table inC wi i t tTh ! t i Iflt’ i nt t = 0
and lineat regression analysis, the slope is t ound t ) be 1.12
with a correlation coefficient of 1.000.
The quantum yield of the PWP P/PYR actinometet is given
by equation 7,
= 0.0169 [ PYR). (7)
Since the pyLidine Concentration in the actinomete was 2.36 x
i02 M (equation 3!), the quantum yie hi ot the act i ninet:et (
i.s 3.99 x ion.
The reaction quantum yield of the test chemical is
given by equation 1,
= a (1)
The pettinent data ate summarized as follows:
3.99 x 10 , (k/k) = 1.12, xc 2610, and
= 2056. Substituting these data in equation I yields
(l.12)(2056/2610)(3.99 x lOg)
= 3.52 x 10 (33)
The rate constants for direct photolysis of test
chemical in aqueous media and the half—life fot water bodies ui d
77 2 .
-------
CG—6010 (October, 1984)
clear sky cond it ions for the wi n1 r irid summer ses ns c m
calculated as follows: The values o YCALX have been
calculated for the summer and winter soistices, Table 1.H. F’oi
sumrnet = 603 days ; foi wintet = 156 ciays .
The reaction quantum yield for chemical A is 3.52 x 10
(equation 33). Using these data in equation 16 yields
summer : kpE = 3.52 x l0 (603) = 0.213 days (34)
winter : kpE 3.52 x 10 (l56) = 0.0549 days’. (35)
These values can he substituted into equation 18 to ohLa n the
half—lives for these two seasons.
summer (t1j )c = 0.693/0.213 = 3.3 days (36)
winter (ty ) = 0.693/0.0549 = 13 days (37)
These results are valid for clear—sky conditions and at shallow
depths in the Illinois River.
e, ‘‘)
‘- I S
78
-------
CS— 60 10
October, 1984
LABORATORY DETERMINATION OF THE DIRECT PHOTOLYSIS REACTION
QUANTUM YIELD IN AQUEOUS SOLUTION AND SUNLIGHT PFIOTOLYSIS
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
2 3
-------
— I I C t i , 1 ) I
Contents
I. NEED FOR THE TEST..... 1
II. SCIENTIFICASPECTS 4
A. HistoricalDiscussion 4
B. TheoreticalDjscussion 9
C. Actinometry . 14
1. Low Optical Density Actinometers 16
a. p—Nitroacetophenone—Pyridine Actinotneter
(PNAP/PYR) 1
h. p—Nitroanisole—Pyridine Act inoTneter
(PNA/PYR) 17
2. High Optical Density Actinometers t8
a. Potassium Ferrioxalate Actinometer 19
b. o—Nitrobenzaldehyde Actinometer 23
C. Reinecke’s Salt Actinometer 25
D. Determination of the Cell Pathlength 26
E. Development of the Test Method 30
1. Procedure One: Determination of the Reaction
Ouantum Yield by the Low Optical Density Test
Chemical and Actinorneter Method
a. Theoretical. Discussion
h. Procedure One: Determination of th
Reaction Ouantum Yield by the Simidtan& ous
Photolysis of Test Chemical and Actin net:or
at Low Optical Density
i. Phase 1 Experiments
ii. Phase 2 Experiments
iii. Phase 3 Experiments
2. Procedure Two: Determination of the Reaction
Ouantum Yield by the Low Optical Density Test
Chemical and High Optical Density Actinometer
Method 38
a. Theoretical Discussion 38
b. Procedure Two: Determination of the
Reaction Quantum Yield by the inw
Optical Density Test Chemical. and IJHh
Optical Density Actinometer Method
1
-------
CS—6010 (October, 1984)
Conts. (Continued)
Pa e
i. Phase 1 Experiments 40
ii. Phase2Experiments 40
3. Procedure Three: Determination of the
Reaction Quantum Yield by the High Optical
Density Test Chemical and Actinorneter Method 42
F. Quantum Yield as a Function of Wavelength 46
G. Photolysis and Molar Absorptivity in Water 46
H. Solar Irradiance Data 47
1. Applicability and Specificity of the Test Method 48
J. Rationale for the Selection of the Test Conditions 50
1. UV—Visihle Absorption Spectrophotometer 50
2. Special Photochemical Labotato y Equipment 50
a. Design of Apparatus 50
h. Light Sources 51
C. Light Filtering Systems 52
d. Reaction Cells. 54
e. Temperature Control 55
3. Cell Pathlength
4. Solvents 56
5. Sterilization 57
6. Preparation of Test Chemical Solution 58
7. PH Effects . 58
8. Volatile Chemical Substances 59
9. Control Solution 60
10. Absorption Spectrum as a Criterion
for Performing the Reaction Quantum
Yield Experiments 61
11. Actinometry 61
a. p—Nitroacetophenone—Pyr id i n’ Act. i nonu t r
(PNAP/PYR) in an Ace—Type PM;kk
b. p—Nitroanisole—Pyridine Actinomet’r
(PNA/PYR) in an Ace-Type PMGRR
c. The PNAP/PYR and PNA/PYR Actinorneters
in a Moses—Type PMGRR 67
12. Chemical Analysis of Solutions 7!
a. Chemical Analysis of Test Chemical Solutions.... 71
b. Chemical Analysis of p—Nitroacetophenone (PNAP) 72
c. Chemical Analysis of p—Nitroanisole (PNA) 72
d. Chemical Analysis of Ferrous Ion in the
FerrioxalateActinometer 73
—ii—
-------
CS—6010 (October, 1 4)
Conts. (Continued)
III . REFERENCES . 74
IV. APPENDIX: GLOSSARY OF IMPORTANT SYMBOLS 77
“, I)
iij
-------
(‘f;— - U U ( ( ) t )I) ‘r , I 4 )
LABORATORY DETERM I NATION 01 TEl F D I RFCT PHO’I’OLYS IS REAC1’ 1 ON
QUANTUM YIELD IN AQUEOUS SOLUTION AND SUNLIGHT PHOTOLYSIS
I. NEED FOR TE € TEST
The majority of the earth s surface is covered by water in
the form of oceans, seas, rivers, lakes, streams, or ponds. As a
result, numerous chemicals enter natural. aquatic systems from a
variety of sources. For example, chemical. wastes are discharqed
directly into natural water bodies, and chemicals reach into
natural water bodies from landfills. Pesticides are applied
directly into water bodies, and are applied to soils and
vegetation and subsequently migrate into water bodies. Some
pollutants present in aqueous media can undergo photochemical
transformation in the environment (i.e., in sunlight by direct
photolysis or by sensitized photolysis). As a result, there is
considerable interest in photolysis in solution, especially the
photolysis of pesticides. However, most of these studies have
been qualitative in nature and involved the identification of
photolysis products. Quantitative data in the form ot raLe
constants and half—lives are needed to determine the importanuo
of photochemical transformation of pollutants in aqueous media.
Direct aqueous photolysis represents the transformation of
a chemical substance by direct absorption of radiant energy
(sunlight) into products different from their precursors.
Chemical substances which are present in aqueous media photolyze
-------
CS—6 J1O COctober, 1984)
at different rates depending upon the solar irradiance, the
chemical substance’s molar absorptivity at each wave1en th of
solar radiation, and its photolysis reaction quantum yield at th ’
wavelengths of concern. Chemical substances which photolyze
rapidly under environmental conditions have relatively short:
lifetimes in the environment. Consequently, the chemical fate
assessment may focus on the transformation products to a greater
exte t than on the parent compound. On the other hand, if the
chemical substance is resistant to photolysis as well as to all
the other possible transformation processes, the assessment
should focus on the parent chemical.
Test Guideline CG—6000, the first in a series of aqueous
photolysis tests, was designed to determine the molar
absorptivity and the reaction quantum yield of a test chemical in
aqueous solution in sunlight. These parameters are then combint”i
with solar irradiance data to determine environmental rate
constants in aqueous solution as a function of latitude and
season of the year anywhere in the United States.
Test Guideline CG—6000 was developed as a screening test
to obtain the direct photolysis reaction quantum yield of a
chemical in aqueous solution by carrying out photolysis
experiments in sunlight. This method does not require
sophisticated and expensive photochemical equipment dn(1 tberet r
the reaction quantum yield can he easily determined and al
moderate cost. However, there are circumstances when this rfle I )i
may not be applicable. For example, this procedure is not
2 2
-------
CS—601() (Octoher, U384
applicable when the teinperat Lirt’ out door’; I at Is hi’ low ..eio Ľ1eS t ‘e. ’;
cent igrade. l’urthermoro, dej)efld tug ipou h e St t ti-; ot i rt sk
assessment for a specific chemical , a more proc i so va I tie t I he
reaction quantum yield may he required. Thus, a more
comprehensive procedure is needed to determine the reaction
quantum yield in the laboratory using specialized photochemical
equipment and monochromatic light.
Test Guideline CG—6010, the second guideline in a series
of photolysis procedures, represents an upper—tier test to
(10 to rrn i no the react: ion quantum y ie 1(1 1 n the I a bora I ory .
data w i I I he w-;eti to do to rrni no how rap 1 ly di rect: phot o lys i .S w t. I
take place in aqueous media under certain environmental
conditions. If the photolysis data indicate that photolysis is a
relatively important transformation process and that the initial
risk assessment indicates that there is a threat to the health of
humans and/or the environment, then other upper—tier photolysis
tests may be required to obtain more precise photolysis data over
a wide range of environmental conditions. These upper—tier tests
will also he concerned with determining the identity 111(1 f ite 01
the trans’orrnation products.
-------
Cs—6010 (dctcher, I’3 34)
Il. SCIEr’JTIPIC ASPECTS
A. Historical Discussion
An historical discussion of the theory and methods of
determining direct photolysis rate c nstant and ha I I —I i v ; ii
the environment (i.e., in sunlight) is given iii (let:ai I lfl SupporL
Document CS—6000, Photolysis in Aqueous Solution in Sunlight. In
addition, Zepp (1982) gives a detailed discussion of experimenta’
approaches to environmental photochemistry. The following
paragraphs give a brief summary of some of the relevant
principles of environmental photochemistry.
Zepp and dine (1977) published a paper describing the
theory of calculating rates of direct photolysis in aquatic
environments. The rates of all photochemical processes in a
water body are a f fected by solar spect ra 1 i rrad i a nee at the wa t e r
surface, radiative transfer from air to water, and the
transmission of sunlight in the water body. ‘It has been shown
that in an optically thin aqueous solution (i.e., the absorhance
of a chemical is less than 0.02 in the reaction cell at all
wavelengths greater than 290 nm), the kinetic expression for
direct photolysis of a chemical at a molar concentration C is
given by the first—order differential equation
— (dC/dt) = +EkaC = k PEC (I)
kpE= $Eka (2)
-------
CS—60li) (October, 984
where is the react ion quantum y ie Id of the chom i I in di 1 ut e
solution and is independent of the wavelength (Section h.P) 1 and
ka , the sum of ka values for all wavelengths of
sunlight that are absorbed by the chemical (i.e., the light
absorption rate constant). The term kpE represents the first—
order direct photolysis rate constant for a chemical in a water
body in sunlight: in the units of reciprocal time. Tlite1r Itintj
equat iOfl I yields
n(C 0 /C ) = kpEt (3)
where is the molar concentration of chemical at time t during
photolysis and C 0 is the initial molar concentration. By
measuring the concentration of chemical as a function of the time
t during photolysis in sunlight, can be determined using
equation 3. Tn addition, equation (3) can he solved for the
condition C = C 0 /2 and the half—life of the chemical is given by
= O. 693 /kpE
Furthermore, under the same conditions cited above
[ i.e., for a homogeneous chemical solution with absorhance loss
than 0.02 in a reaction cell at all wavelengths greater than
290 nm and at shallow depths (less than 0.5 m)}, the first—order
direct photolysis rate constant, kpF is
5
‘) •1
-------
Cs— 6 ) 1 ) ( )c i ’r , I ‘-) 4
kpE $E AcLA
where • F’ is the react ion quant urn y le Id wh I ch s ndt’ &’nde u I
the wavelength, is the molar absorptiv ty in the units
molar cm , and L is the solar irradiance in water in the
units 1O einsteins cm 2 day 1 [ Mill et al. (1981, 1982a,
1982bfl. is the day averaged solar irradiance at shallow
depths for a water body under clear sky conditions and is a
function of latitude and season of the year.
A simple screening test has been developed in Test
Guideline CG—6000 using equation 5. As an approximation, it has
been assumed that the reaction quantum yield E is equal to one,
the maximum value.* As a result, the upper limit for the direct
photolysis sunlight rate constant in aqueous solution is obtained
and equation S becomes
(kE) Ac X (6)
Substituting equation 6 in equation 4, the lower limit for the
half—life is then given by
jt is possible that under certain circumstances the re et i On
quantum yield can be greater than 1. For example, this h pperis
when a chain reaction occurs after the absorption o i quimtuin ot
light. However, this rarely occurs in dilute aqueous solution.
6 2’
-------
CS—6010 (October, 1984)
= O.693/(k ) (7)
pE max.
In order to estimate (kpE)max using equation 6, it is necessary
to obtain the molar absorptivities (CA of the test chemical
and the solar irradiance data L . Details for determining
are given in Test Guideline CG—6000 along with a listing of
solar irradiance data L (Tables 3—6). A procedure is
described how to use these data to estimate (kpE)max and
/ min. using equations 6 and 7.
Test (;uideline CG—6000 also (lesCrit)eS a procedure tor
determining the reaction quantum yield of a test chemical in
aqueous solution in sunlight. The reaction quantum yield can
then he combined with the and L data in equation to
determine the direct photolysis rate constant kpE and the
corresponding half—life t using equation 4. Furthermore, it
should be noted that the method is very general and can be
extended to determine rates of photolysis over a range of
environmental conditions using a computer program. Zepp and
Cline (1977) have written a computer program to caJcuUite rates
of phololysis as a function of depth in water, in n itural waters
as a function of the water absorption or aLtenu d: ori rovf I
(a ) , the average ozone layer that pertains to the ieason and
location of interest, and as a function of latitude and season ol
the year. This program has been recently updated with the best
available solar irradiance data and is called the CC SOLAR
Computer program. The GC SOLAR computer program is available on
7
2I’
-------
CS—6010 (QctobeL, 1984)
request ER. Zepp, Environmental Research Laboratory, U.S.
Env it onmenta I Protect ion Agency, Co I loge Stat. ion Road, Athens,
Georgia 30601]
Test Guideline CG—6000 was developed as a screening
test to determine the reaction quantum yield of a test chemical
in aqueous solution by carrying out photolysis experiments in
sunlight. This method does not require sophisticated and
expensive photochemical equipment and thus the reaction quantum
yield can be determined easily and at a modest cost. However,
the method does have some limitations. For example, this Test
;uirlel me can only be cart ied out (tnt i iij the w it in 5 ISOfl 5 &)t t he
year in the United States. Unfortunately, in many parts ot this
country, the temperature falls below 0°C during the winter
months. Therefore, if photolysis experiments are carried out in
the winter in certain parts of the country, the dilute aqueous
solutions would freeze, the tubes would break, and the samples
would be destroyed. Furthermore, under certain circumstances, a
more precise value of the reaction quantum yield may be
required. For example, if the photolysis test data indicate that
photolysis is a relatively impot tant t.r insf( nat i on pr cess and
the initial risk assessment indicates that. there H a threat I
the health of humans and/or the environment, then an uf)J)er—t 00
test may be required to obtain a more precise value of the
reaction quantum yield for a specific test chemical. Finally,
some laboratories have photochernical equipment using
monochromatic light already set up and may prefer to use it.
8 21
-------
CS—6010 (October, 1984)
Thus , a more comprehensive procedure i s needed to de t e rin i ne t he
react ion quantum yield of a test chum i c dl in the Llhordt ory us i ii
specialized photochernical equipment.
B. Theoretical Discussion
Methods for determining the reaction quantum yield* of
pollutants in dilute aqueous solution have been described by Zepp
(1978) and some theoretical aspects of photochemistry will be
discussed in this section.
Based on the first and second laws of
hotOchemistry, the general rate equation for aqueous photTolysu-
of a chemical in a cell containing volume V of solution and
exposed area A is
Rate = — (dC/dt) = • I (A/V)Fs F’cx (8)
where c is the molar concentration of the chemical, t is the
time, is the reaction quantum yield of the chemical, I is
the incident light intensity at wavelength x , is the
fraction of the light absorbed by the system, and is the
fraction of light absorbed by the chemical IRaizani dud ( 1rIssi I
(1970J. For complex organic molecules in solution, is a ;s’ m’H
t Since the term quantum yield is applied to other photochernical
processes including fluorescence, phosphorescence, and
sensitizations, the term “reaction quantum yield” should always
be explicitly stated.
9
-------
CS—6010 (October, 984;
to bto 1iit1t p no1t r’ t t)f t h waVe It I1(Jt h, I .
CA dt I I nt’(i I )y t O’( OI. I t ( n
= 1 — io ax + (9)
F c C/(cz +c C) (10)
where is the absorption (or attenuation) coefficient of
water, is the molar ahsorptivity of the chemical, and . is
tho 1)athlPfl(jLh of I iijht
I y I Ct I I fl I A OX A/V ) , U.i i Ofl I ho ‘C( )Ifl ’ ‘
Rate = _(dC/dt) = •I 1 F Fx . (11)
For an aqueous solution of a chemical with low optical
density or absorbance, the incident light on the solution is
weakly absorbed. That is, only a fraction of the incident light
is absorbed by the system. PSA , defined by equation 9, takes
the form
SA 1 — 10 X = 1 — CXp 1—2. 31)3 (z A (II o)
= 1 — exp(—x) (12)
where x = 2.303 x4txc)t (13)
10
)
-------
CS—6010 (October, 1’)84)
and the expansion of exp(—x) in a series [ Pierce (1929)] yields
exp(—x) = 1 — x + x 2 /2! — x 3 /3! + . (14)
For weak absorbance, x < 0.02; and thus the terms x 2 /2!
x 3 /3!, . . . are small relative to x and can be neglected.
Thus, equation 14 becomes
x = 1 — exp(—x) (15)
Substituting equation 15 into equation 12 and utilizing equation
13 yields
= x = 2 . 3 O 3 (cz +c C)L (16)
Substituting equations 16 and 10 in equation 11 yields the first—
order differential equation
=
- (dC/dt) = kC (17)
where k is the first—order rate constant and is given by the
expression
k = 2.303 I c 2. (IH)
p x x
11
-------
CS—6010 (October, 1984)
Integrating equation 17 under boundary condit ions
(t = 0, C 0 ) and (t, Ct) yields
tn(C 0 /Ct) = kpt (19)
Equation 19 can be solved for the condition C C 0 /2 and the
half—life of the chemical (ti, 2 ) is given by
/2 O.693/k (20)
For an aqueous solution of a chemical (e.g., an
actinometer or test chemical) with high optical density or
absorhance, essentially all the light is absorbed. Under these
conditions the absorbance is >2. Since the absorhance is equal
to (rz +c C)t and is >2, the term
10 A AC
and F 5 defined by equation 9 is then given by the relationship
= 1 (21)
Furthermore, since essentially all the light is absorbed by the
chemical, c C >> and defined by equation 10 is then
given by the relationship
12
-------
CS—6OlC (Oct eber , 4
= 1 (22)
Substituting the results of equations 21 and 22 in equation 11
yields the zero—order differential equation
— (dC/dt) = “A (23)
Tnte(J rat I ng equat inn 23 under the boundary COt d it l flS ( tT =0,
and (t,Ct) yields
(C 0 — Ct) = k t (24)
where k is the zero—order rate constant and is given by the
express ion
k (25)
A
Equations 23, 24, and 25 are applicahl.e to the dis-
appearance of chemical (e.g., test chemical and actinometer) and
the reaction kinetics are studied by measuring the disappearance
of chemical as a function of time. However, there are instances
when the reaction kinetics are studied by following the formation
of one of the products (e.g., the ferrous ion in the ferrioxalat:c
actjnometer discussed in Section II.C.2.a). For a condition in
which the rate of disappearance of chemical equals the rate ot
formation of one of the products,
2
-------
C.S—6010 (October, 1q84)
— ((IC/lit) A = dC /clt A , ( •h
where Cr is the molar concentration of reactant and is the
molar concentration of one of the products. Using equation (26)
in equation 11 and proceeding in the same manner as above, the
following equations are obtained:
(dC /dt) = (27)
(C ) = k t (28)
tp p
= ‘A , (29)
where (Ct) now represents the molar concentration of the product
formed at time t.
C. Actiriometry
Chemical actinometers are used in reaction quantum
yield experiments to measure the integrated light intensity
incident on the sample during photolysis. Chernic i tinoinet’r
are photochemical reactions which have been calibrited diri t ly
or indirectly with light sources of known light flux ind hdve
well defined reaction quantum yields a at specific wavo]enqtJi ;
A.
2 T’)
14
-------
CS— E t) 1 0 ( Oc ( In’ r ,
In the environment, photolysis takes place in aqueous
media. Thus, the reaction quantum yield of a test chemical is
required in aqueous solution. Complications are encountered when
reaction quantum yield experiments are carried out in the
recommended cylindrical cells (Section 11.0) with water as the
solvent for the test chemical and the actinometer is dissolved in
an organic solvent. These complications arise because the liqht
flux measured is a function of the refractive index of the water
and the organic solvent relative to the glass used in the
fabrication of the cylindrical cells. As a result, the reaction
quantum yield measured can be in error. This effect will he
discussed in more detail in Section 11.0. Therefore, in order to
measure precise reaction quantum yields of test chemicals in
cylindrical cells, actinorneters in aqueous solution are highly
desirable. Four aqueous actinorneters will he described in the
following sections which can he used to obtain environmentally
relevant reaction quantum yields in the spectral region 290—
750 nm. In cases where aqueous solutions are not practical, the
recommended actinometer solvent is acetonitrile. Acetonitrile
20°
has an index of refraction ri = 1.344 which is close to the
0°
refractive index of water (n = 1.332). Thus, the o—
nitrobenzaldehyde actinometer dissolved in acetoriitrile is also
described (Section II.C.2.b) and can he used in these reaction
quantum yield experiments. There are other actinometers that
could be used in these experiments. However, if actiriorneter ;
other than the ones described in this section are used, they must
2. 1
-------
Cs— 60 10 ( c t r , I ‘4 4
be completely described. For detailed information on
actinometers, the descriptions by Calvert and Pitts (1966), de
Mayo and Shizuka (1976), and Murov (1973) are highly recommended.
1. Low Optical Density Actinometers
Two low optical density actiriometers will be described
which can be used to determine the reaction quantum yield of a
test chemical at low optical density ( i .e. , ahsorb inc < 0.02)
These actinometers are ideal for environmental actinometry
experiments since they are in aqueous solution and have variable
reaction quantum yields which are well defined. These two low
optical density actinorneters, composed of the systems p-
nitroacetophenone—pyridine and p—nitroanisole—pyridine, have been
recently developed for the Office of Toxic Substances, U.S.
Environmental Protection Agency, by SRI International [ Mill et
al. (1981, 1982) and Dulin and Mill (1982)]. The
p—nitroacetophenone--pyridine actinometer has already been
incorporated in Test Guideline CG—6000 to measure the reaction
quantum yield of a test chemical in aqueous solution outdoors in
sunlight.
a. p—Nitroacetophenorie—Pyridine Actinometer (PNAP/PYR)
The p—nitroacetophenone—pyridirie actinometer (PNAP/PYR)
has been thoroughly described by Mill et al. (1982), Dulin and
Mill (1982), and in Test Guideline CG—6000. Laboratory
experiments indicated that the reaction quantum yield for Ll
16
‘T ’ )
‘i-, 1,.#
-------
CS—60 10 (October, 1984)
photolytic transformation of Pt’ AP depended linearly on the molar
concentration of pyridine at a fixed molar concentration of
PNAP. Regression analysis of the data on the reaction quantum
yield as a function of the molar concentration of pyridine [ PYR]
up to a concentration of 0.2 M [ Winterle (1984)1 (at a fixed
concentration of PNAP of approximately I x —5 M) gave the
relationship
= 0.0169 [ PYR] (30)
with a correlation coefficient of 0.9986. Additional laboratory
experiments indicated that a was independent of the wavelength.
The ultraviolet—visible absorption of PNAP was obtained
in an aqueous solution containing 1% acetonitrile [ Mill et al.
(19821 and the molar absorplivities at 313 and 366 nm
313a and C366a) were 2,056 and 160 M’ cm’, respectively.
With a 450 watt medium pressure mercury lamp in an Ace—type
photochemical “merry—go—round” reactor, this actinoineter: can be
adjusted with pyridine to have half—lives that range from greater
than 12 hours to several weeks (Section II.J .1l.a).
b. p—NitroanisolePyridine Actinometer (PNA/PYR)
The p—nitroanisole—pyridine actinometer (pNA/PYR) has
been thoroughly described by Mill et al. (1982) and J)ulin nd
Mill (1982). Laboratory photolysis experiments ir Jic, t:od tl I
the reaction quantum yield for the transformation o PNA 1pend I
linearly on the molar concentration of pyridine at a fixed
17
-------
CS—6010 (October, 1984)
concentration of PNA. Regression analysis of the data on th ?
reaction quantum yield as a function of the molar concentration
of pyridine [ PYRJ up to a concentration of 0.02 M (at a fixed
concentration of PNA of approximately 1 x 10 M) ve the
relationship
= 0.4371PYR1 + 0.000282 (31)
with a correlation coefficient of 0.9998. Additional laboratory
work indicated that was independent of the wavelength.
The ultraviolet—visible absorption spectrum of PNA was
obtained in an aqueous solution containing 1% acetonitrile [ MU1
et al. (1982)1 and the molar absorptivities at 313 and 366 nrn
313a and c 366a were 10,300 and 1,990 M 1 cm 1 ,
respectively. With a 450 watt medium pressure mercuty lamp in an
Ace—type photochernical “merry—go—round” reactor, this actirtometer
can be adjusted with pyridine to have half—lives that range from
approximately 15 minutes to more than 12 hours (Section
II.J.ll.b)
2. High Opt ica I ens Act In one tiers
There are three excellent high opt ic;iI (1l n’ ity
ctinometers (absorbance > 2) which can he used to d ’, t’ rmir
reaction quantum yield of a test chemical. These ir’ the
ferrioxalate actinometer, the Reinecke’s salt actinoineter, and
the o—nitrobenzaldehyde actinorneter. The ferrioxalate
18
-------
CS—6010 (October, 1984)
actinometer is an ideal actiriometer for environmental actinometry
since it is in aqueous solution and is most suitable tor use in
the spectral range 290—500 nm. This actinometer is highly
recommended and has been adopted for use in this Test
Guideline. The one disadvantage of this actinometer is that
actinornetry experiments must be carried out in a dark room with
photographic “sate lights” since it is sensitive to visible
light. The o—nitrobenzaldehyde actinometer is another excellent
high optical density actiriometer, but it must be used in
acetonitrile. It is easy to use but is only sensitive in the
range 280—400 nm. This actinometer is described in this Support
Document and can be used as an optional actinometer in place of
the ferrioxalate actinometer, provided the test chemical absorbs
in the spectral region 290—400 nm. This actinometer follows
zero—order kinetics and is applicable up to 25% transformation
(Mill. (1984)]. Reinecke’s salt is another excellent aqueous
solution actinometer which is recommended for use in the spectral
region 500—750 nm. Because this actiriometer is sensitive to
visible light., actinometry experiments must be carried out in a
dark room with photographic “safe lights”. All these
actinometers are high efficiency actinometers which have well
defined quantum yields at fixed wavelengths.
a. Potassium Ferrioxalate Actinome1 r
The potassium ferrioxalate actinometer (JeV lop d by
Parker (1953) and Hatchard and Parker (1956) is a widely u d
19
i )
-------
CS—6010 (October, 1984)
solution phase actinometer by many photochemists. It. 1S USCLI at
high optical density (absorhance > 2) and is applicable ovet a
wide range of wavelengths (254—577 nm). However, it is most
useful in the range 254—500 nm. Since it is an excellent
actinometer and has a large wavelength range of sensitivity in
the environmentally relevant region 290—500 nm, it has been
adopted for use in this Test Guideline.
Quantum yields for the ferrioxalate actinometer are
given in Table 1 as a function of wavelength ide Mayo and Shizuka
(1976)). At 0.15 M ferrioxalate, the quantum yield at 313 nm is
1.20 and the average quantum yield at 366 nm is 1.18. The net
photochemical reaction is
Fe 3 + (C 2 0 4 )272 hv > Fe + CO 2 (32)
Prom the stoichiometry of equation 32
—d [ Fe 3 )/dt =d [ Fe 2 )/dt (33)
At high optical density, equations 27, 28, and 29 ire app1icibI
and the reaction kinetics are zero—order. ‘I’he kineti’, of liii’;
actinometer are followed by measuring the molar concentration Of
Fe 2 + formed as a function of time. The molar concentration of
Fe 2 + is measured spectrophotometrically via the formation of Lh
red phenanthroline complex and the determination of the
20
-------
CS—6010 (October, 1984)
absorbance of the complex at 510 nm. Ferric ion forms oniy a
weak complex with pherianthrolirie which is transparent at 510
nm. Murov (1973) gives a detailed procedure for using the
ferrioxalate actinometer and this procedure is highly
recommended. Zepp (1984) has found that it is most useful to use
the ferrioxalate actinometer at a concentration of 0.15 M. At
0.15 M, the ferrioxalate actinometer can accept a much greater
dose of light than at 0.006 M. The irradiated solut ion is
diluted 100—fold prior to analysis of Fe 2 +, which must not he
allowed to exceed 0.005 M. Thus, it is recommended that the
Murov procedure by modified as described by Zepp.
-------
CS—601() (Cctc bt r, Q64\
1’,lblEb I
()u. nt urn Vi t 1 t1 ; a)f
( .it
Wavelength
nm
(K Fe(C O
M
577/9
546
509
480
468
436
416
405
392
365/6
358
334
313
297 /30 2
254
0.15
0.15
0.15
0 • 15
0.15
0.15
0.006
0.006
0.006
0.006
0.006
0.15
0.006
0. 006
0.006
0.15
0 .006
0.006
0.013
0.15
0 .86
0.93
0.92
1.01
•
1.11
1.14
1 . i
1.21,
1.15,
1 . 25 h
1.23
1.24
I
1 . 24
1.25
I . 26
1. 20 a
a. Lee .1 and .Selinqer Fill. 1964. .J (‘horn hy:; 3 ): 1 .
b. Wegner EE and Adam .son AW. 1966. .1 Am (‘h4’n S c l: 94.
c. Zepp (1984).
-------
CS—6010 (October, 1984)
With a 450 watt medium pressure mercury lamp filtered
at 313 or 366 nm in a photochemical “merry—go—round” reactor,
ferrioxalate decomposes rapidly and samples should he taken for
analysis in the range of 1 to 15 minutes.
h. o — itrohenzaldehyde ctinometer
The o—nitrohenzaldehyde actirtometer was oriqinally
developed by Pitts et al. (1964) for use in a KBr matrix. This
actinometer was then used in polymethacrylate film and procedures
for use in this matrix are fully described by de Mayo and Sht uk
(1976)
Smith et al. (1977, 1978) described the use of the p—
nitrobenzaldehyde actinorneter in solution with ace onitri1e as
the solvent. Dultn nd Mill (l982 reported the use of this
ctinometer in acetonitrile 15 .1 hiqh optical density
actinometer. The photochemical reaction is
p l o 1 d.4 J 1O
)
and the reaction quantum yield was reported to he O.5( i und i
independent of the wavelength. It is sensitive in the spect:r 1
region 280—400 rim. The photochemical reaction was studied by
-------
CS—6010 (October, 1984)
measuring the concentration of o—nitrobenzaldehycle (c —N .)
remaining at time t using reverse—phase high pressure li ui i
chromatography using 50% acetonitrile in water . : the eluent .
the calibration experiments by Dulin and Mill ( 1 q 2 ), a
of approximately i x io 2 M o—NR in acetonitrile was photolyzed
in quartz or borosilicate tubes for periods up to 16 minutes at
313 and 366 nm in an Ace—type photochemical “merry—go—round”
reactor with a 450 watt medium pressure mercury lamp. They
reported that the light intensity I at 313 nm varied from
2.5 x io—6 to 3.5 x io6 einsteins sec. 1 liter’ and at 366 nm
I. varied from 6.7 x io6 to 8.2 x io6 einsteins sec.’
A
liter . Mill (1984) reported that the zero—order kinetics were
obeyed up to 25% transformation and the molar ahsorptivities at
313 and 366 nm (C313a and C366a) were 1500 and HI N1 cm ,
respectively. At 313 nm all the light is absorbed by o—NB in a 1
cm cell. At 366 nm, essentially all the light is absorbed by o—
NB in a 1 cm cell (98.5% of the light is absorbed at 366 nm) . If
this actjnometer is used at A > 366 nm to 400 nm, the
concentration of 0—NB must be adjusted in the reaction cell so
that the optical density is > 2 in the reaction cell throughout
irradiation.
The o—nitrohenzaldehyde actinometer in t()Il I riI.
solution is an excellent actinometer for use in laboratory
photolysis experiments. it is easy to use hut has the limitation
that it is only sensitive in the spectral region 280-400 nm. H
is highly recommended for use with test chemicals that absorb in
2 )
-------
CS—6010 ( ct ber, l9 4)
the req ion 290—40(1 nm. It is described in detai 1 in t hi sect i n
hut not in the Test (‘,uideljne. If this actinometzer is used n
the Test Guideline, then the Test Data Report of Test Guideline
CG—60l0 should state it. It should he noted that since the
photochemical reaction is followed by measuring the concentration
of o—nitrobenzaldehyde as a function of time, equations 23, 24,
and 25 are applicable.
c. Reinecke’s Salt Actinometer
The ferrioxalate actinometer is a very useful
actinorneter hut has the limitation that it is inr,.st: useful in the
environmentally relevant spectral range 290—500 nm. The
Reinecke’s salt actinometer [ KCr(NU 3 ) 2 (MCS) 4 ] , developed by
wegner and Adarnson (1966), can be used in the spectral range 500—
750 nm. Thus, it is recommended for use to determine the
reaction quantum yield of test chemicals that absorb in the range
500—750 nm.
Upon photolysis of this actinometer, the Reinecke’s
salt gives the hydrate Cr(NH 3 ) 2 (NCS) 3 (I-1 2 0) and NCS. The
formation of thiocyanate ion can he determined with terric I( n
spectrophotometrically. The actinometer is used in lp eow;
solution with an optical density a 2.2. De Mayo and Shizuk i
(1976) give a detailed procedure for using this actinomet:nr nd
this procedure is highly recommended.
2
-------
CS—6010 (October, 1984)
D. Determination of the Cell Pathlength
From the theoretical discussion given in Section 11.8,
it is evident that the pathlength of the light is an important
variable, especially when low optical density actinonieters are
used (equation 18). For a 1—cm square cell the pathlenqth is
easily defined. However, for a cylindrical cell where the light
falls on rounded surfaces, the pathlength is not easily defined
because of refraction and reflection in the cell. Zepp (1978)
developed an experimental method for determining the effective
pathlength of any cell, and in particular for cylindrical, cells
which are used in environmental studies.
If the molar concentration (C) of the light absorbing
substance is high, then xC >> and to a good approximation
( + C) = c C . Using this result in equations 9 and 10
yields
F’ —
(35)
F
CA
Under these conditions, equation 11 becomes
(Rate) = —(dC/dt) — lo cxCL) (36)
9,”)
s
26
-------
Cs—t 0 10 (Oct , I ‘)$ 4)
For the special case when essentially all the light is
absorbed by the chemical, > 2 and 1O x — 0.
Equation 36 becomes
(Rate)max “A
Dividing equation 37 into 36 yields
( a t., ) R i te) mu x — x X (38)
Rearranging equation 38 yields
1 — X =
and —1og 10 (1 — X) = c Ct (39)
A plot 01 —10(310(1 — X) vs. cxC yields a straight line witi a
slope equal to the path length of the c cl 1 . Zepp (1978)
measured the effective cell pathlength of small cylindrical cells
used in his environmental photochemical studies by employing the
benzophenone—serisitized isomerization of cis—1,3—pentadiene.
This procedure has been used to measure the effective pathienyth
of Corning Glass culture tubes 13 x 100 nm and it was found that
L was 11.2 mm. [ Mill et al. (1982a)] . In a similar manner,
reaction tubes made from borosilicate stock of 12 mm o.1. had an
effective pathlength of 10.0 mm. [ Mill et al. (1982a)1.
27
-------
CS—6010 (October, 1984)
The following procedure can he used to determine the ceN
pathlength of cy ii ndr ica 1 eel Is us i n i t he e—n t i en I
actinometer (0—NB) which is descr ibed in Sect ion ii •C . .I.
Prepate o —NB scilut ions with an o t i’a I di’n;i y “ with
known concentration
2. Prepare, by sequential dilutions of the solution above,
0—NB solutions with optical densities 1 to 0.1 (e.g.,
00’s approximately 1, 0.8, 0.6, 0.4, 0.2, arid 0.1) with
known concentrations C 1 , C 2 , C 3 , C 4 , C 5 , and C 6 .
3. Photolyze the seven solutions (Cmt C 1 , C 2 , C 3 , C 4 1 C 5
C 6 ) for identical times in a PMGRR and analyze for o—N 3
by the pr ocedur e outlined in Sect; ion I [ . C .2. b.
Reaction should correspond to less than 25
transformation of 0—NB.
4. Determine the ratios etc.... These
ratios are designated X 1 , X 2 , etc...
5. Plot of the quantity [ —log 10 (1—X 1 )] versus
(—1og 10 (l—X 2 )1 versus c> C 2 etc. for each set of data
points; c is the absorption coefficient for 0—NB at
the photolysis wavelength x
6. Calculate the slope of the best strai.jht line which i
equal to the effective cell Lh enqth t pi. t i ri
39). A more accurate and precise value of rnu/ hI-
obtained by a fit of the data to equation 39 w;inq
linear regression analysis.
‘ I
28
-------
(‘S—60 1() (Oct bt t , 1q84)
These photolysis experiments can he carried out at
wavelengths 313 or 366 nm. The molar absorptivities of a-NB at
these wavelengths are: C313a = 1500 M 1 cm 1 and 366a 181
cm [ Mill (1984)].
When reaction quantum yield experiments are carried Out in
cylindrical cells (i.e., tubes) using different solvents for
actinometer and test chemical, the light flux measured, and thus
the reaction quantum yield of the test chemical in an Ace-type
photochemical “merry—go—round” reactor, wiLt he solvent dependent
because of the different refractive indices of the solvents
used. Vesley (1971) measured the reaction quantum yield for the
photoisomerization of cis—1,3—pentadiene in berizene at 365 nm in
an Ace—type photochemical “merry—go—round” reactor using the
standard actinorneters ferrioxalate in water and benzophenorie—
benzhydrol in benzerie. The reaction quantum yield of the
photoisornerizatiori reaction was found to be 0.64 when the
ferrioxalate actinometer (in water) was used while the reaction
quantum yield was found to he 0.55 when the henzophenone—
benzhydrol actinometer (in benzene) was USed. Vesley attributed
the differences in the values of the reaction quantum yield to
the differences in refractive indices of water and henz ne
relative to the refractive index of glass. Benzene has a
refractive index ri 200 =l.50l which is similar to the refractive
index of pyrex n 20 ° =l.52 and thus all the incident light passed
into the solution whereas with water, with a refractive index
200 1.332, some of the light is refracted and reflected at the
29
-------
CS—6010 (October, 1984)
inner cell walls. As a result, the samples in which the henzene
was used as the solvent “see” more photons than the aqueous
actinometer solutions. Thus, the ferrioxalate actinometer
underestimated the light that actually entered the solution in
the cell when benzerie was the solvent for the photoisomerization
reaction and thus overestimated the true reaction quantum yield
for the cis—l,3—pentadiene by 15%.
Since in environmental photochemistry water is the medium
in which photolysis takes place,it is important that actinometry
experiments he performed in aqueous solution whenever
practical. In those cases where aqueous actinometers are not
practical, the actinometer solvent actonitrile (n 200 =l.344) may
be used since its refractive index is close to that of pure
water.
T . Development of the Test Method
In general, the reaction quantum yield is deterni ned
under the experimental conditions in which all the incii nt lidht
is absorbed by the chemical (absorhance > 2) . The concentrations
of chemicals used in these experiments ranje from icr M tO
greater than 0.1 M. However, most organic pollutants are
hydrophobic so that the water solubility is generally below
l0 M. Hence, for these chemicals in aqueous solution, only a
small fraction of the incident light is absorbed (absorhance <
0.02). Zepp (1978) described two procedures for deterinininq the
reaction quantum yield of organic pollutants in very dilute
2’.’;
-------
CS—6010 (October, 1984)
aqueous solution with an absorbance < 0.02. One procedure
measures the light intensity incident on the sample u inq a hi h
optical density act i nometet ( e .q . , t- he fe i xa at e
actinometer). This procedure is used extensively by
photochemists and therefore it has been incorporated in this Test
Guideline.
In addition, Zepp (1978) described a second procedure
for determining the reaction quantum yield by comparing the
photolysis rate of the pollutant and actinorneter under conditions
in wh 1 ch the ii j h t is only weakly a hsor bed by bot:h COIflI)Ot1 ntis
SUch CX t iments ate conveniently made in a photochetni ca I “ine r
go—round” reactor. This elegant method has not been used because
the standard chemical actinorneters cannot he analyzed at the low
concentrations required for this technique. However, two
excellent low optical density variable reaction quantum
yield actinometers have been recently developed for the Office of
Toxic Substances/U.S. Environmental Protection Agency by SRI
International [ Mill et al. (1981) and Dulin and Miii (1982)1 and
these actinometers are discussed in Section ir.C.). One o these
actinometers (PNAP/PYR) has already been incoLpot aI:ed in Test.
Guideline CG—6000. Thus, this second procedute hat; ut been
incorporated in this Test Guideline (CG—6010).
There is another procedure which can he USed to
determine the reaction quantum yield of a test chemical at hijh
Optical density by measuring the light incident on the test
chemical with a high optical density actinometer (e.g., the
31
-------
CS—6010 (October, l 84
ferr loxalate act i nometer) • This exce 1 out procodu ro , howov ’r
has only limited applicability to those test c i nica1’ which ar
very soluble in water and have high optical density
absorhance 2 ) . Thus , th is procedure has not bo’n i ncorpora t o i
in this Test Guideline. However, it is discussed in this Support
T cument and may be used for certain organic pollutants. If this
procedure is used for these pollutants, then it must be described
in detail.
1. Procedure One: Determination of the Reaction Quantum
Yield by the Low Optical Density Test Chemical and
P ctinometer Method
a. Theoretical Discussion
Since aqueous solutions of the test chemical and
actinometer will he photolyzed at low optical density (absorhance
< 0.02), equation 18 is applicable. Using this equation, the
equations for test chemical (c) and actinometer (a) can he
written as follows:
= (40)
kpa = 23 O 3 aIACAa (41)
If the aqueous solutions of test chemical and actinometer are
photolyzed simultaneously in identical cells with the same
pathlength and at a fixed wavelength A (or in a narrow
2
-------
CS—601() (Octobet. , 1984)
wavelength band) in a photochemical “merry—go—round” reactor
(PMGRR), then equation 41 can be divided into 40. Carrying out
this operation and rearranging the result yields
c (kpc/kpa)kxa/cxc)$a (42)
using equation 19, the equations for test chemical (c)
an(1 actinometer (a) can he written as follows:
= k t (43)
tn(Co/Ct)a = kpat (44)
the half—life of the test chemical (C) and actinometer (a) can be
written as (t , 2 ) and (t]y 2 )a and can be obtained from equation
20. Again, since both the actinometer and the test chemical
solutions are photolyzed simultaneously in identical cells in a
PMGRR, then equation 44 can be divided into equation 43.
Carrying out this operation and rearranging the result yields
= (kpc/kpa) Ln(Co/Ct)a
During photolysis at wavelength x , the concentrations of test
chemical and actinometer are measured periodically as a function
of the time t. These data are then used to determine the ratio
4,
( 1.
33
-------
CS— 6iJ I 0 ( )c t The r , 954
of the rate constants (kpc/kpa) using linear regression analysis
on equation 45. The slope of the best straight line is the ratio
of the rate constants.
Using spectroscopic procedures, the molar absorptivities oi
test chemical and actinometer and ) can be determined •-i
the absorbing wavelength ) . Cnmhininq these data il n with
the ratio of the rate constants (kpc/kk) ) anJ t ht r act i an
quantum yield of the actinometer the reaction quantum
yield of the test chemical can be determined using equation
42.
b. Procedure One: Determination of the Reaction
Quantum Yield by the Simultaneous Photolysis
of Test Chemical and Actinometer at Low
Optical Density
ProcetlU re One involves the s imu I taneous )ht)t a tys is at
aqueous solutions of test chemical and actinometer at low optical
density (absorbance < 0.02) in an Ace—type PMGRR* to determine
the reaction quantum yield of the test chemical. The excitation
source is a 450 watt medium pressure mercury lamp with
appropriate filters to isolate the 313 or 366 nm hands. The
*The Ace—type PMGRR is designed so tnat the entire reaction c Il
is irradiated. This type of PMGRR is commercially available.
There is a Moses—types PMGRR [ Moses et al. (1969)1 whi:h 1 tli,r;
slightly from the Ace—type PMGRR in that it contains irithw;
(i.e., slits) so that only a narrow portion of the solit Ion in
the reaction cell is irradiated. Por details see Section
II.J.2.a and II.J.11.c.
34
-------
CS—6010 (October, 1984)
procedure has been divj(led i nt-o three t ses . In h i t’ I , t t ’
molar absorptivities of test: chernic i I (&‘) r ’ del t’rmj ned it 1 nd
366 nm 3l3 and c 366 ) using uv—visible absorption
spectroscopy. Based on these results, photolysis experiments are
carried out at 313 or 366 nm corresponding to the higher value of
C313c or Ł366c• The Phase 2 procedure is composed of trial
experiments in a PMGRR at 313 or 366 nm to determine the approxi—
mate rate constant and half—life of the test chemical and to
choose the appropriate actinometer which has a rate constant
approximately the same as the rate constant of the test
chemical. In the Phase 3 procedure, aqueous low optic al density
solutions of test chemical and actinometer (at a fixe(i molar
concentration of pyridine [ PYRI ) are photolyzed simultaneously in
a PMGRR at 313 or 366 nm. The data obtained from the Phase 1, 2,
and 3 experiments are then used to determine the reaction quantum
yield of the test chemical.
i. Phase 1 Experiments
In the Phase 1 experiments, the molar absorptivities ot
the test chemical (c) are determined at 313 and 36 nrn
313c and Ł 366c using procedures outlined in l’o t (uidel in e
CG—1050 and CG—6000. The method outlined in these two jUidi I irn ;
is the standard procedure used by all laboratories Uor
determining the molar absorptivity of a chemical in solution.
Photolysis experiments will then be carried out in the Phases 2
and 3 experiments at 313 or 366 nm corresponding to the higher
35
-------
CS—6010 (October, 1984)
value of C313c or Ł366c . By photolyzing the test chemical at
the wavelength corresponding to the higher value of molar
absorptivity, the rate of photolysis is maximized.
ii. Phase 2 Experiments
First, a dilute aqueous solution of test chemical (c)
at a low optical density (absorbance < 0.02) and at an initial
concentration (C 0 ) is photolyzed in an Ace-type PMGRR at
wavelength 313 or 366 nm (irradiation wavelength chosen based on
the Phase 1 experiments) using a 450 watt medium pressure mercury
lamp and appropriate filters to isolate the monochromatic
wavelength A . After approximately 50% transformation, the
concentration of test chemical (C ) is measured at time t.
These data are used in equation 43 to tiotermi no in approx i mato
rate constant kpc• This result is used in equation 2() to obtain
the half—life of the test chemical (tY 2 )c• If (tY 2 )c is less than
12 hours, use the actiriomneter PNA/PYR (Sections II.C.l.h and
II.J.11.b). If (tl,, 2 )c is greater than 12 hours, then use the
actinometer PNAP/PYR (Sections u.C.l.a and II.J.ll.a).*
Utilizing equation 44, trial experiments are carried
out with the appropriate actinometer, at an absorhance < 0.02, to
determine the molar concentration of pyridine IPYRI no4 14?d to
make the rate constant of the act inometer approx :nal I y qtia I I >
*jf a Moses—type PMGRR is used, then new criteria have to be
definea to determine which of these two actinometers have to be
used. For details see Section II.7.11.c.
36
-------
CS—6010 (October, 1984)
th( rat_*’ ony tant of the t e -t cheinjc, I (wit ft in j i’ . V t lU
details, see Section Il.J.EI.
According to Sections U.C.i.a, h, the re 5 iction uantuin
yield of these actinometers is a function of the mol 5 ir
concentration of pyridine; and these relationships are given by
equations 30 and 31:
PNAP/PYR actinometer 4 ’a = 0.0 169 [ PYRJ (30)
PNA/PYR actinorneter a = 0.437 [ PYR] + 0.000282, (31)
where IPYRJ is the molar concentration of pyridine for a fixed
concentration of PNAP or PNA of approximately 1 x 10 M.
iii. Phase 3 Experiments
An aqueous solution of test chemical at low optical
density and at a concentration (Co)c is prepared and placed in
small cylindrical tubes. Based on the results of the Phase 2
experiments, the appropriate actinometer is chosen and the
concentration of pyridine is adjusted so that: the rate constant
of the act inometer is approximately equal to the r.ite )It t-1nI oh
the test chemical (i.e., within ± 50%) . An aqueous art I n orn ter
solution is prepared at concentration (C 0 ) with a molar
Concentration of pyridine [ PYR] . The actinometer solution is
placed in identical tubes used for the test chemical solution.
The volumes of actinometer and test chemical solution should he
37
-------
CS—60l0 (October, 1984)
the same. Aqueous solutions of test chemical and actinometer are
simultaneously photolyzed in the PMGRR at the chosen wavelength
313 or 366 nm using the 450 watt medium pressure lamp and
appropriate filters. The concentrations of test chemical (C )
and actinometer (Ct)a are measured as a function of the time t.
Using these data in equation 45 and linear regression analysis,
the ratio of the rate constants (kpc/kpa) is obtained. The
reaction quantum yield of the actinometer Can he determined
at the molar concentration of pyridino employed in this
experiment (i.e., substituting [ PYR] in ejuation 30 or 31).
Using these data along with the molar absorptivities of test
chemical 3l3c or c 366 ) and the actinometer
313a or c1 3 6 6 a) , given in Sections II.C.1.a, b, in equation
42, the reaction quantum yield of the test chemical can be
determined.
2. Procedure Two: Determination of the Reaction Ouantun
Yield by the Low Optical Density Test Chemical adH
Optical Density Actinometer Method
a. Theoretical Discussion
Since the test chemical Cc) is at low optical density,
equations 18 and 19 are applicable:
kpc 2 * 3 O 3 cIxCxcZ (18)
Ln(C 0 /C ) = kpct ( 9)
38 2 1
-------
CS—6010 (October, 1984)
Furthermore, since the actinorneter is at high optical density and
the ferrioxalate actinometer (Section II.C.2.a) is widely used by
photochemists, equations 28 and 29 are applicable:
(Ct)a = kpat (28)
kpa = (29)
where (Ct)a now represents the product formed which is [ Pe 2 ]
Solving for in equation 1.8 and I in equation 29 yields the
following two equations
k /E2.3O 3 I c ] (46)
= kp /
and substituting equation 47 in equation 46 yields
c apcpa2 3 Cxc (4 )
b. Procedure Two: Determination of the Reaction
Quantum Yield by the Low Optical Density Test
Chemical and High Optical Density Actiriolneter
Method
39
)
-------
CS—6010 (October, 1984)
Procedure Two involves the photolysis of an aqueous
solution of test chemical at low optical density and d l i
solution of an actinometer at high optical density in i
photochemical optical bench (POB) or in a PMGRR to determine the
reaction quantum yield of a test chemical. The excitation sourc
is a 4S0 watt medium pressure mercury lamp with appropriate
filters to isolate the 313 or 366 nm bands. The procedure has
been divided into two phases. In the Phase 1 experiments, the
molar absorptivities of test chemical (c 3 i 3 and c366c) are
determined at 313 and 366 nm using uv—visible absorption
spectroscopy. Based on these results, photolysis experiments are
carried out at 313 or 366 nm corresponding to the higher value ot
313c or c 366c . In the Phase 2 experiments, aqueous low
optical density solutions of test chemical and aqueous high
optical density solutions of the ferrioxalate actinometer are
photolyzed sequentially in a P08 or in a PMGRR at 313 or 366
nm. These data, along with the Phase 1 data, are used to
determine the quantum yield of the test chemical.
i. Phase 1 Experiments
The phase I experiments are identical to those
described in Section II.E.l.b.i.
ii. Phase 2 Experiments
In the first series of experiments, a dilute aqueous
solution of the ferrioxalate actinometer (a) at high optical
40
-------
CS—6010 (October, l $4)
den tty (I.e., t absorhance 2) is placed in a c I I ( r cel i )
and photolyzed in a POB (or a PMGRR) at wavelength 313 or 366 rim
(based on the results of the Phase 1 experiments) using a 450
watt medium pressure mercury lamp and appropriate filters. The
molar concentration of ferrous ion formed (Ct)a is measured as a
function of the time t. These data are fitted to equation 28
using linear regression analysis to obtain kpa• No measurements
should be used when the absorbance of the actinometer falls below
2. In a third set of experiments at the end of this procedure,
kpa is determined again in the same manner as described above.
The results of these two experiments are averaged to obtain
(kpa)ave.
In a second set of experiments, a dilute aqueous
solution of test chemical (C) at low optical density (absorbance
< 0.02) and at an initial concentration (Co)c is then photolyzed
in the same POB or PMGRR. The reaction cell or cells must be
identical to those used in the actinometry experiments and
contain the same volume of solution. The concentration of test
chemical (Ct) is measured as a function of time and these data
are fitted to equation 19 using linear regression analysis to
obtain kpc• Using kp and (kpa)ave.i the molar absorpt vit.y ol
the test chemical (c 313 or 366c , and the pathlength of:
the reaction cell or cells (which must he known or determined by
the procedure described in Section 11.0) in equation 48, the
reaction quantum yield of the test chemical ( ) can he
determined.
41
-------
CS—6010 (OctoLer, 1984
3. Proct dure Thrt t : f)et rini nut ‘ui )t thc. i t)uunt urn
Vie t(1 th :\ :
Method
Since the test chemical (c) and act inomet r (a) soIuti n -
will be photolyzed at high optical density (absorhance > 2),
equation 25 is applicable. Using this equation, the equations
for test chemical (C) and actinometer (a) can be written as
follows:
= CIx (49)
kpa a 1 x
If aqueous solutions of the test chemical and act.inometer ace
photolyzed simultaneously in identical cells at a fixed
wavelength x in a PMGRR, then equation 50 can be divided into
equation 49. Carrying out this operation and rearranging the
result yields
a pc kpa) (51)
P ssuming that the concentration of test chemical ar act
are measured as a function of time during photo ysis, th ri
equation 24 is applicable. Thus, the equations for test chwnicui
(c) and actinorneter (a) can be written as follows:
42
-------
CS—6010 (October, ‘3 4)
(C 0 — Ct) = k t (52)
(C 0 Ct)a kpat
Since both the test chemical and actinometer solutions are
photolyzed at wavelength x in a PMGRR in identical cells, then
equation 55 can be divided into equation 54. Carrying out this
operation and rearranging the result yields
(C 0 — Ct)c = (kpc/kpa)(Co — Ct)a (54)
During the photolysis, the molar concentrations of test chemical
and actinometer are measured periodically as a function of time
t. These data are then used to determine the ratio of the rate
constants (kpc/kpa) using linear regression analysis on equation
54. The slope of the best straight line is the ratio of the rate
constants.
If it is now assumed that the high optical density
actinometer is the ferrioxalate actinometer, then equation 28 is
applicable. The equations for test chemical (C) and actinometer
(a) can be written as follows:
(C 0 — Ct)c = kpct (! 2)
k t (tn)
t 1 a — pa
43
-------
CS—6010 (Oc t )er, 4)
and dividing equation 55 into equation 52 and rearranqin tht
result yields
(C 0 — Ct) (kpc/kpa)(Ct)a (56)
If now the molar concentration of ferrous ion (Ct)a is measured
along with the molar concentration of test chemical (Ct) as
function of the time t, then these data can he substituted int:’
equation 56 and the ratio of the rate constants (kpc/kpa) can bt’
obtained using linear regression analysis.
Finally, using the ratio of the rate constants
(kpc/kpa) as described above and the reaction quantum yield of
the actinometer at wavelength x in equation 51, tne
reaction quantum yield of the test chemical can be
obtained.
Another procedure can be used to determine the reaction
quantum yield of a test chemical ( ) by sequentially
photolyzing actinometer and test chemical solutions. In this
method, aqueous solutions of test chemical and actinometer at
high optical density (absorbance > 2) are photolyzed sequenti.il y
at a wavelength A in a POB (or PMGRR). In the first series ot
experiments, the actinometer solution is photolyzed in a cell (or
cells) in a POB (or PMCRR) and the molar concentration of
actinometer (a) is measured as a function of the time t. IJsing
these data in equation 53 and linear regression analysis, the
44
-------
CS—6010 (October, I 84)
actinometer rate constant is obtained since the slope equals
kpa• tf the ferrioxalate actinometer is used, then equ:ition 55
is applicable. By neasuri ng the concent rat ion ot rrou ; i fl
formed (C a as a funet ion of the time t , in t h ’ii I ’
determined using these data in equatjon 55 and linear regression
analysis. In the third series of experiments, the above
procedure is repeated and another actinometer rate constant is
obtained. The results of these two experiments are averaged to
yield (kpa)ave. . Since the reaction quantum yield of the
actinometer is known at wavelength x , and the actiriometer
rate constant (kpa)ave. has been obtained by the procedure
described above, then I can be determined using equation 50.
In the second series of experiments, an aqueous
solution of test chemical at high optical density is photolyzei
at wavelength A in a cell (or cells) in a POB (or PMGRR). The
cell (or cells) must be identical to those used in the
actinometer experiments and contain the same volume of
solution. The molar concentration of test chemical is measured
as a function of the time t and these data are used in equation
52 to obtain using linear regression analysis. Using the
values of and I , as described above, in equation 49, the
reaction quantum yield of test chemical can be determinei.
45
-------
( - —hU I L) ( )CI )i’ , ‘J: 4 )
F. Ouantum Yield as a Function of Wavelength
In Section II.A, it has been shown that the rate
constant kpE in the environment is a function of the reaction
quantum yield and it has been assumed that the reaction quantum
yield is independent of the wavelength. This assumption is valid
for most complex molecules in solution because the photoreaction
from second or higher electronic states usually cannot compete
with the rapid radiationless decay of excited molecules to their
first excited state [ Zepp (1978)). Thus, this Test Guideline
only recommends measuring the reaction quantum yield ot a Lest
chemical at one wavelength. However, there are exceptions to
this rule (e.g., organornetallics) and this must be considered
when carrying out these experiments. Determination of the
reaction quantum yield at more than one wavelength may be
discussed in a Test Rule for specific chemicals or classes of
chemicals.
G. Photolysis and Molar Absmptivit 1 inWator
Since water is the environment i] ly relev: nt. ineHuin in
which photolysis takes place, these reaction quantum yieLd
experiments should be carried out in pure water. Up to 1%
acetonitrile, however, is acceptable. Furthermore, whenever
possible, the molar absorptivity of the test chemical should he
determined in pure water. However, determination of the molar
absorptivity is difficult for hydrophobic chemicals and is
essentiaUy impossible for very hydrophobic chemicals ( .q., 1) 1)1,
46 ‘ ‘
I. I
-------
CS—6010 (October, 19R4)
(‘l s, an(i po1yarotn t Ic I1ydroc Irl)()n.s) . i\ ; an a It ‘‘mat v ’’
Zepp(1978) recommended the determination of the molar
absorptivity in mixtures of water and acetonitrile or methanol.
Since these solvents are effective in dissolving many very
hydrophobic organic chemicals, have refractive indices very close
to that of water and thus cause minimal solvent effects (i.e.,
minimum shifts and changes in the absorbance of strong absorption
bands of non—polar compounds), and do not absorb radiation at
wavelengths greater than 290 nm, they have been recommended t r
use in this Test Guideline for determining the molar absorplivity
of test chemicals. However, a minimum amount of the solvent
should be used to measure the molar absorptivity of a test
chemical.
H. Solar Irradiarice Data
In order to calculate the sunlight direct photolysis
ratio constant and the corresponding half—life ti 1 for a test
chemical as a function of latitude and season of the year
anywhere in the United States, it is necessary to use the solar
irradiance parameter Lx • Lx values are proportional to the
average light flux that is available to cause phot ilysis In tJit ’
wavelength interval A over a 24—hour day at a s citic latitim(1( ’
and season date. The L values are defined by the angle of
declination of the sun at _200 for winter, _lOO for fall, +10°
for spring and +20° for summer. The actual dates for 1982 that:
correspond to these angles of declination are January 21,
April 16, July 24, and October 20, for winter, spring, sumtn r and
V
47 I
-------
CS—60U) (Oct b ’r, V 84)
fall, respectively [ AA ( N82) I . The values art’ I i’ t ‘ i ii
Tables 3 to 6 of Test Guideline CG—6000 as a function ( t lititiude
and season of the year and are applicable to clear sky coridi—
tions, shallow depths in water bodies, and t or chemicals whose
absorbance is less than 0.02 in pure water [ Mill et al.
(1982a)J. The theoretical basis for the data is discussed in
Section 11.0.9 of Technical Support Document CS-6000.
1. Applicability and Specificity of the_Test
Method
Because of the theoretical princij)les outlined in
Section II.E, Procedure One is only applicable to test chemicals
and actinometers which have low optical density (i.e.,
absorbance < 0.02) while Procedure Two is only applicable to a
test chemical with low optical density and with actinometers at
high optical density (i.e., absorbance > 2).
For environmental photochernistry, the general
procedures outlined in this Test Guideline are ap licabU to all
chemicals which have uv—v is ib I c abserpt iflns in the rt I n 2 H)—d0t)
nm. Solar radiation reaching the earth’s surtace ha- 1
cutoff at a wavelength of approximately 290 nm dut U
absorption by ozone [ Leighton (1961), Peterson (1976), Zepp and
dine (1977), Demerjian (1980)1. The long wavelength limit is
set by thermochemistry since light of wavelength greater than 800
nm is not of sufficient energy to break chemical bonds of ground
state molecules [ Calvert and Pitts (1966), Benson (1976)1.
Photolysis does not occur unless there is absorption of radiant
48
-------
CS—6010 (Oct ther , 1984)
energy. This is the direct consequence of the Grotthus-Draper
law, the first law of photochemistry. If a chemical in aqueous
solution only absorbs light at wavelengths below 290 nm it will
not undergo direct photolysis in sunlight. A few ox irnples o
chemicals that only absorb light below 290 nm and tht ret ore
should not be tested by this Test Guideline are alkanes, alkenw;,
alkynes, saturated alcohols, and saturated acids. It is possible
that some chemicals will absorb radiation mainly at wavelengths
below 290 nm but may have an absorption tail that extends above
290 nm. Photolysis experiments should he carried out for these
chemicals.
Procedure One, as described in this Test Guideline, is
limited to the wavelength region 290—400 nm because the only welt
developed low optical density actinometer.s ate sensitive in this
region (Sections II.C.l.a, h). Fortunately, this disadvantage is
minimal since a large number of chemicals absorb light in this
spectral region.
Procedure Two, as described in this Test Guideline, is
applicable to the wavelength region 290—750 nm. The ferrioxalate
actinorfleteL is the most useful in the region 290—500 nm (Section
II.C.2.a) while the Reinicke’S salt actinometer is useful in the
region 500—750 nm (Section II.C.2.c).
This Test Guideline is only applicable to pute
chemicals and not to the technical gtade. Ovetosti:nat’ Of 1fl0l %Y
absorptivities usually occur when technical grade ;ut:;t. nci s aY(
tested because the impurities frequently absorb in tJ ;arn
spectral region as the pure chemical.
49 4)- —
-------
CS —6010 (October, 1984)
J. Rationale for the Section of the r t
Condi t tons
I
All the laboratory equipment needed to determine tńe
uv—visible absorption spectrum of a test chemical in aqueous
solution is listed in the references in Test Guideline CG—1050.
The equipment required to carry out this test is .stand zd
equipment which is commonly used in most laboratories.
2. Special Photochemical Laboratory Equipment
a. Design of Apparatus
The apparatus used to measure photochemical kinetics in
the laboratory is usually the “merry—go—round” reactor (PMGRR) or
the optical bench (POB). There are two types of PMGRR equipment;
the Ace—type and the Moses—type. The Ace—type PM(;RR is designed
so that the entire reaction cell is irradiated while the Moses—
type PMGRR EMoses et al. (1969)J is designed with windows (which
act as slits with a fixed aperture) so that only a narrow portion
of the reaction cell is irradiated. Only the Ace—type is
commercially available from Ace Glass Company, Vineland, New
Jersey; the Moses-type PMGRR would have to be specially built in
a machine shop. Since the Ace—type PMGRR is the onIy ont
commercially available, this Test Guideline has been )e t : 1 /
designed to use this PMGRR. However, if the Moses-type ‘M(P k ;
available, this Test Guideline has to he modified .sL ght1y i ;
described in Section lI..J ,ll.c,
50
-------
CS—6010 (October, 1984)
POB equipment is commercially available from a number ot
sources. One type of POB which can he used is a Schoeitel
Reaction Chemistry System [ Schoeffel Industries, 1w ’ .j which
contains a 1000 watt xenon lamp, a h igh i ntens it y im n R’h r mit r
and a sample compartment.
Both PMGRR and POB equipment provide for constant and
uniform irradiation of a solution at a fixed distance from the
light source and allow for the use of a light—filter system
between the light sources and the reaction solution. These two
pieces of equipment are standard equipment used by photochemists
to study photochemical reaction kinetics in the laboratory and
are described in numerous sources in the literature fe.g.,
Calvert and Pitts (1966), de Mayo and Shizuka (1976), Murov
(1973), Zepp (1982), Moses et al. (1969), and MiL’ et a1.
1982a) I
b. Light Sources
There are a number of light sources which are available
for use in photochemical studies (e.g., low, medium, and high
pressure mercury lamps; xenon lamps; and lasers). The selection
of the specific light sources should he determined by the
following criteria ide Mayo and Shizuka (1976)1:
(1) The wavelength of the light absorbed by cheJnic jls(A)
should be of sufficient intensity to induce p It()ly J
in a reasonable amount of time. The intensity ot th
light source should be greater than iol4
quanta/sec ./cm 2 .
51
1 ,
-------
CS—6010 (Octo er,
(2) The light source should be stable for prolonged periods
of irradiation.
3 ) If possible , the 1 igh t emana t i ng t torn t h ’ i it ou tc
should be a point source for ease of collimation.
(4) A monochromatic light source is desirable since the
excitation wavelength may he a narrow band without the
loss of intensity.
The 450 watt medium pressure mercury lamp has most of
these characteristics and is widely used by many photochernists.
The PMGRR is designed to accornodate this lamp. Therefore, it has
been recommended for use in this Test Guideline. Numerous other
light sources are available and may he used in this Test
Guideline. These light sources are described in detail by
Calvert and Pitts (1966), de Mayo and Shizuka (1976), and Murov
(1973). If a light source other than the 450 watt medium
pressure mercury lamp is used, it must be completely described in
the Test Data Report Section of this Test Guideline.
C. Light ‘iltering Systems
Monochromatic or narrow band wavelength light is
essential for the accurate determination of the react i r ju.irituw
yield of a test chemical by the procedures outlined in t his fi,;t
Guideline [ see the theoretical discussions in Sections N. .
and El. A serious error can arise with an incorrectly filtered
light source that allows light (i.e., leak light) at other thdn
the irradiation wavelength . . If is dramatically larger
at the leak wavelength, a small fractional leak can produce a
52
I ’ ,
-------
CS—60 [ U ( )ct Dt’t ,
large error hecau.st the it act ion ol the 1 ight b d
linear function of
Because the reaction quantum yields of pol’utants ate
usually low,it is important to have high intensity monochromatic
light so that faster photolysis rates are obtained in the
laboratory equipment. High intensity grating monochrornators can
he used to isolate narrow—band radiation from an intense uv
source such as mercury or xenon lamps; however, this equipment is
expensive. Alternatively, a narrow wavelength band can atsO
isolated by the use of chemical solutions and/or glass filters;
and a number of these combinations have been described by Calvert.
and Pitts (1966), Murov (1973), de Mayo and Shizuka (1976), and
Mill et al. (1982a). These filter systems are inexpensive.
G1 ss interference filters are also very efficient in isolating
narrow wavelength bands of light.
Two filter systems are described iii this Test Guideline
to isolate the 313 and 366 rim bands from a 450 watt medium
pressure mercury lamp. These filter systems are inexpensive ;in i
are commonly used by many photochemists. Hence, t:hey are hiLjh [ y
recommended and have been incorporated for: use in this Test
Guidelir e. For the 313 nm filter system, it is recommended that
a 0.005 M solution of potassium chrocnate he used to tilter out
the 302 nm mercury band. If the test chemical does not absorb
light at wavelengths greater than 400 rim, the glass filter
Corning CS—754 is not needed. The 754 filter is designed to
block out visible light. If other light filtering systems are
used other than those recommended in the Test (;uid tir , h(O
53
-------
CS—6010 (Oct ht’t , I. ) 4)
they must he completely described in the Test Data Rep r t ‘t i n
of this Test Guideline. For these light filtering systems, t is
important to check the light emanating from them with i
monochromator to make sure that the desitecl wavelen ith b. n 1 h i’-;
been isolated.
d. Reaction Cells
In general, reaction cells of large volume ate
appropriate for POB equipment while small reaction cells ate used
for the PMGRR equipment. Small volume reaction cells (e.j.,
rectangular cuvettes of approximately 4 rnL) can be used with the
Schoeffel Reaction Chemistry System [ Zepp (1984)1. Test chemical
and actinometer reaction solutions must be photolyzed in
identical cells [ see the theoretical discussions in Sections
II.B.and E].
For the PMGRR equipment, small tubes are required
because of the nature of the design of this equipment. These
tubes can be constructed from quartz or borosilicate glass.
React ion tubes with an ins ide diameter t 1 1 mm •i r r
as they are inexpensive anti .ue e u;y U) St i U) j)tiVi flt viUd I i—
zation of the test chemical. Grease should he v it1t 1 ;iric’
hydrophobic chemicals might adsorb to it. Disposable cultute
tubes (13 x 100 mm) with Teflon—lined screw caps ate recommended
since they are inexpensive, easy to seal, and are readily
available from commercial laboratory supply companies. Volatile
compounds can be conveniently studied in culture tubes equipped
with gas—tight Mininert valves. Sampling can he introduced iflt()
54
-------
CS—6OlO (October, 1984)
or removed from the tubes through the septum in these valves with
no loss of substrate [ Zepp (1984)]
For the POB equipment , the most common nd t unc t i ona I
design for reaction cells is a cylindrical shape with 1 tic liy
flat circular windows fused to each end of the cylinder and at
right angles to its axis. Optically flat quartz windows are
recommended because they transmit 100% of the light at 313 and
366 nm. The reaction vessel should be of sufficient volume to
permit removal of samples for analysis without significantly
altering the volume of the reaction solution in the cell.
Details for the construction of these reaction cells may be found
in Calvert and Pitts (1966) and de Mayo arid Shizuka (1976).
e. Temperature Control
The effect of temperature on photochemical Processes jii
solution is usually considered to he small since these reactions
occur from the electronic excited state and this process usually
has small or essentially zero activation energy. However, a
number of photochemical reactions have activation energies that
are in the range 3—7 kcal./mole fBarltrop and Coyle (1975)1.
Thus, the reaction rate may vary with temperature to a small
extent. Therefore, the reaction cell should he controlled to i
temperature t ± 2°C in the range 20 — 30°C.
3. Cell Pathienyth
Zepp (1978) described an experimental method 1.or u -;i wj
an isolated wavelength band to determine the effective pathlenrjth
55
21.
-------
CS—6010 (Octonet, i S4)
t of a y Ce Il , and in pat t I t’II I at tot sma 1 1 cy 1 it lt I
used in environmental studies. This p ocedute is desut ibed in
Section II.D. This procedure has been used to measure the
effective pathlength of the recommended Corning Glass cuLtute
tubes 13 x 100 mm; and the effective pathlength was found to be
11.2 mitt [ Mill et al. (1982a)]. These researchers also measured
reaction tubes made from borosilicate glass stock of 12 mm o.d.
and the effective pathlength was found to he 10.0 mm. If the
cell pathlength was measured by this procedure, then report all
the data obtained in these experiments in the Test Data Report
Section of the Test Guideline including a complete c1CSCtI )tiOfl ot
the actinometer used. This procedure can be used to measure the
effective pathlength of cylindrical cells designed for the POB
apparatus; however, it is easier to measure the pathlength of
these cylindrical cells directly with a precise centimeter ruler
or an equivalent measuring device.
4. Solvents
Pure distilled water is used because dissolved
impurities could sensitize or othetwise affect the tate cd
photolysis. If the half—life of an aqueous o [ ut ion of tist
chemical in the photochemical equipment is less than 24 hours,
then distilled water meeting ASTM Type II standards (or an
equivalent grade) is adequate and is recommended for use.
In addition, if the half—life of an aqueous solution of
test chemical in the photochemical equipment is greater than 24
hours, then the water needs to be sterile because bacteria may
56
‘ - ‘_)
d
-------
CS—6010 (October, 1984)
consume or alter the to t. chemical dun ug proton ed p ’t
testing which may occur in the course of the rate
determination. Thus, water meeting ASTM Type [ TA standards, or
an equivalent grade, is highly recommend to minimize
biodegradation.
Furthermore, it is important that the water be
saturated with air prior to preparation of the test chemical
solutions to simulate environmental conditions. Air—saturated
Type II water can be easily prepared by allowing the water to he
equilibrated in a vessel plugged with cotton while air-saturated
ASTM Type hA water can he prepared by allowing the water to
equilibrate in a vessel plugged with sterile cotton.
Reagant grade acetonitrile is recommended as the
organic cosolvent with water in photochemical studies to avoid
the potential effects of impurities on the rate of photolyis.
Spectrograde acetonitrile or methanol is recommended for
spect.roscopic studies to determine the molar absorptivity of the
test chemical. Overestimates of molar absorptivities may occur
when spectroscopic grade solvents are not used because the
impurities may absorb in the same spect rai reg ion a the I
chemical. All scientists use .spectro.sr:r)I)ic (Jr.I(ie )Iv flI ti
molar absorptivity and other spectroscopic .stuthes.
5. Sterilization
Sterilization is necessary to kill the bacteria and
therefore eliminate or minimize biodegradation which could
interfere with the photolysis rate determination. The presence
57
-------
CS—6010 (Octoher, 19S4)
of bacteria in either the test sohit ions or coi’itroI in y c iu’; ’
biodegradation of the test substance when the photeIysi
experiments are ca rr ied on t OVC r a 1 onq period ot t i m ’ . 1’h i
will introduce an error in the concentration of the test
chemical. Thus, it is extremely important to use aseptic
conditions in carrying out long—term photol.ysis experiments to
minimize biodegradation. Glassware can be sterilized easily in
an autoclave or by use of any other suitable non—chemical method.
6. Preparation of Test Chemical Solution
If the chemical substance is too dir t icu ft t. di ye
in pure form to permit reasonable handling and analytic l
procedures, than a test solution may be prepared by tirst
dissolving a chemical in reagent grade acetonitrile. The final
acetonitrile concentration in the test solution should be no more
than one volume percent in order to avoid solvent effects (Smith
et al. 1977, 1978). Acetonitrile was chosen as a solvent as it
is soluble in water, is non—polar and thus effective in
dissolving many substances which are insoluble in water, it does
not absorb radiation over the wavelength range of 290 to 0() urn,
and it causes minimal solvent effects (i.e., miniin’iui ;hif I’; iii
bands and changes in absorbance) for test substances.
7. pH Effects
The molecular structure of a chemical substance which
ionizes or protonates is a function of the p1-I. As a result, the
absorption spectrum and consequently the rate of photolysis may
58
2T1
-------
Cs— 60 10 (Oct Oho r , L,) 4
change with pH. The recommended procedure is to detcrinine the
uv—visible spectra under conditions in which only one species
strongly predominates in solution. This can be accomplished by
preparing buffered solutions containing the test chemical at pHs
at least 2 p 1- I units above the PKa and at least 2 pH units below
the pKa* For example, the P 1
-------
CS—6010 (October, 1984)
recommended. Grease should not be used since hydr pholic
chtnntca is may acis()r t) to it . V() lat i li’ cnouut1 ; c.in l
conveniently studied in culture tubes oqu pped with +l’;—t i ht
Mi n i net t va 1 yes . Sarnp I c s can be jut. r oduce(i n t o o W( V( ‘d f r
the tubes through the septum in these valves with no loss of
substrate [ Zepp (1984)]. As an alternative, the tubes can be
sealed with a torch, In addition, the reaction vessels should he
as completely filled as possible to prevent volatiUzation to any
air space.
9. Control Solution
Undetected loss of a test substance throu jh
volatilization, hydrolysis, or other rocesses dur i n j the urst
of the photolysis exper iment will result in Lhe determinut: i n ot
erroneously large rate constants for aqueous photolysis.
Therefore, for volatile chemical substances, it is important that
the reaction vessels and control vessels be filled as completely
as possible and sealed in order to avoid evaporative losses. To
correct for possible losses, control solutions of test substance,
in darkened vessels, are placed in the apparatus or: alongside the
apparatus and the contents of the control vessels are analyzed at
the same time as the photolyzed samples (i.e., at ton ). In
this way, the loss of test chemical for roc ;ses ol h’;r t h u
photolysis may be determined and eliminated or acourit’ d for iii
determining kpc For simplicity, if the loss of chemical in
the control is small (i.e., approximately 10% or less), one Cam)
calculate a first—order loss, k 10 5 , and subtract it from
60
-------
CS—6010 ( )ctaber, 1984)
(kpc))bs to give the c irecled direct pheto [ y ; H Fit it ;t n t
. If hydrolysis is found to he significant (greater than
10%), hydrolysis studies should be carried out first to define
precisely the kinetics of this process [ Test Guideline CG—5000J
10. Absorption Spectrum as a Criterion for
Performing the Reaction Quantum Yield
Experiments
The test method is applicable to all chemicals that
have uv—visible absorptions in the range 290—800 nm. Thus, the
uv—visible absorption spectrum of a chemical in aqueous solution
will give a good indication of whether it would be useful to
carry out these photolysis experiments and determine the reaction
quantum yield of the test chemical. For more details, see
Section 11.1.
ii. Actinometry
A detailed discussion of actinometry is jiven in
Section iI.C. However, some further details must he considered
for the two low optical density actinometers PNAP/PYR and PNA/PYR
discussed in Sections u.C.l.a and b, respectively. In order to
carry out Procedure One, it is necessary to choose the
appropriate actinometer and to adjust the concentration of
pyrizlirie so that the rate constant of the actinometer
approximately equals the rate constant of the test h tnical. l}i
rate constant of the actinometer is a function of th irii,f,ir
concentration of pyridine [ PYRI, the intensity of light - indent
61
-------
CS—6010 (October, 1984)
on the actinometer , and the molar absorptivity t the
actinometer . The pettinent eqUationS fot the It w M t icii
density act i nometer are equat i ns 4 1 and 2). ‘ r 1 r c Itu ‘
it is necessary to cler ive equat ions under the cond it. i n I hat: th
rate constant of the actinometer equals the rate constant of the
test chemical; that is,
k = k (57)
pa pc
Substituting equation 41 in equation 57 yields
= 2303 ’ a 1 x xa ( 8)
Substituting equation 41 in equation 20 yields
3.0lXlO’1 IxL xJ’ ( )
These equations are perfectly general and are applicable fat use
in a P08 or PMGRR using any light source filtered to yield a
monochromatic wavelength band A . Equations will be derived foi
a typical 450 watt medium pressure mercury lamp in an Ace—type or
a Moses—type PMGRR. The pertinent information has beer, tak r ,
from the paper by Dulin and Mi ii (1982), the data from .SectHon ;
u.C.l.a and b, and from Mill (1984).
2 ‘.
62
-------
CS—6010 (0ctebe , 1484)
a . p—Nit roace tophenone— Pyr id i no 1 \c t i noiuot: o
(PNAP/PYR) in an Ace—type PMt;RR.
Por the PNAP/PYR actinometer, the re1ation hip et iho
reaction quantum yield as a function of the molar
concentration of pyridine [ PYR] is given by equation 30
0.0 169 [ PYR) (30)
Pot the 450 watt medium pressure mercury lamp in an Ace—type
PMGRR used by Mill and Dulin (1982) the approximate aveLage light
intensity at 313 nm and 366 nm was: 1(313) 3.0 x
einsteins sec 1 [ [ 1; 1(366) 7.5 x l0 einsteins sec L’.
Conversion of these intensities to hours yields 1(313) 1.08 x
i c r 2 einsteins hr. L 1 and 1(366) 2.70 x io2 einsteins hr.
L’. For the cells used by Dulin and Miii (1982), g. =1.0
[ Mill (1984)J. The molar absorptivities of PNAP at 313 and 366
_1 _1
nm are: C313a = 2056 M cm and C366a 160 M’ cm (Section
rr.C.1.a).
At 313 nm, the following relationships L:.lfl L)t
calculated. Substituting equation 30 in equation ‘ .tnd w;
the above pertinent data:
-2 -
k = 2.303(1.69 x 10 [ PYRJ)(1.08 x 10 )(1.0)(2.056 x 10
PC
k = 0.864 [ PYR) (6(J)
PC
63
-------
CS—6010 (Octobet, 1984)
EPYR] = 1.16 (61)
where is in the units (hours) and EPYR] is in the units ot
molar concentration. Equation 61 corresponds to equation 18
listed in the Test Guideline. Furthermore, since kpa
(equation 57), equation 60 becomes
kpa = 0.864 [ PYR) (62)
and substituting this result in equation 20 yields
(ti..) = O.802/ [ PYR] (63)
a
where (t1/ )a is the half—life of the actinometer in hours. The
half—life of the actinometer can range from 16 hours at
IPYRJ = 0.05 M to 33 days at [ PYR] = M. Thus, this
actinorneter is suitable frn use with test chemicals havin j half—
lives > 12 houts to sevetal weeks.
The above calculation can be epeated at 366nm and the
results are:
[ PYRI = 5.95 kp (64)
where is in the units of (hoursY 1 and [ PYRI is in the units
of molar concentration. Equation 64 corresponds to e uat:ion 19
listed in the Test Guideline. The half—life of the iCt j,u,rnet .•t
in hours is given by the equation
64
-------
CS—6010 (October, 984)
(t1, a 4.13/EPYR)
Using the molar absorptivity data for PNA1 at 313 and
366 nm, the absorbance A313a and A366a is less than 0.02 in a
1.00 cm absorption cell containing a PNAP solution of
concentration 1 X 10 M. The recommended tubes used for
these photolysis experiments are approximately 1 cm. Hence, PNAP
-5
solutions of concentration < 1 x 10 M have absorbance less
than 0.02 in the recommended tubes and thus cot respond to law
optIca’ (iensity Thus, PNAP solutions of CoflCentrat(on
-5
I x 10 M have been recommended for use in this Test
Guideline.
b. p—Nitroanisole—Pytidine Actinorneter (PNA/PYR)
in an Ace—Type PMGRR
The above calculations can be repeated as described
above using equation 31 for the PNA/PYR actinometer.
0.437 [ PYRI + 0.000282 ( H)
a
1(313) = 1.08 x io2 einsteinS hr. L’ and 1(366) 2.70 x
einsteins hr. L ; t 1.0; C• 3 13a = 1.03 x i0 4 M’ cm’ r 1
c = 1.99 x M 1 cm (Section II.C.1.h).
366a
Thus, at 313 nm:
[ PYRI = 8.93 X 10 3 (kpc 0.0722), (66)
65
-------
CS—6010 (October, 1984)
where is in the units (hoursY and EPYRI is iii the units ot
molar concenttat Ion. Equation 66 cm tesponds to t- qLI;1t ion 2 1
listed in the Test Guideline. The half—life of the actinoineteF
in hours is given by the equation
(tl/ 2 )a = (162EPYRI + O.104) (67)
The half—life of the actinometer can range from 15 minutes at
(PYR) = O.02M to 8.3 hours at EPYR] 10 4 M. Thus, this
actinometet is suitable for use with test chemicals havin j half—
lives of > 15 minutes to apptoximately 12 houts:
At 366 nm:
[ PYRJ = 1.85 X l0 2 (k c — 0.0349) (68)
where is in the units of (hourY and [ PYR) is in the units
of molar concentration. Equation 68 corresponds to equation 21
listed in the Test Guideline. The half—life of the actinometet
is given by the equation
(t]/ )a = (78h [ RJ ÷ 0.0504 ’ (69)
Using the molar absorptivity of P JA at 366 nm, the
absorbance A366a is less than 0.02 in a 1.00 cm absotption cell
containing a PNA solution 1 x M . The recommended tubes
used for these photolysis experiments are approximately 1 cm.
66
-d
-------
CS—6010 (October, 1984)
Hence, PNA solutions of concentration 1 x l0 M have
absorhance less than 0.02 in the recommended tubes and this
corresponds to low optical density at 366 run. Howove ., at. 313
nm, the molar absorptivity is large and A313a is less than 0.02
in a 1.00 cm absorption cell containing a PNA solution of
concentration < 0.2 x l0 M. PNA is difficult to analyze at a
concentration significantly below 0.2 x io M. As a compromise,
the allowable absorbance has been raised to 0.04 and the
concentration of PNA in a 1.00 cm cell is 0.4 X M. Thu.s,
the concentration of PNA has been set at 0.4 x l0 M to
photolysis experiments at 313 nm in the recommended tubes. PNA
can be analyzed in this concentration range and only a small
error (< 10%) has been introduced in the method.
c. The PNAP/PYR and PNA/PYR Actinometers in a Moses-Type
PMGRR
Test Guideline CG—6010 has been designed for use with
an Ace—type PMGRR which is commercially available. However, iF a
Moses—type PMGRR (Section II.J.2.a) is available, the discussions
given in Sections II.3.11.a and b have to he modiFied. As
indicated in Section II.J.2.a, the Ace—type PMGRR is desiyned SI)
that the entire reaction tubes are irradiated while the Moses—
type PMGRR is designed with windows, which act as slits, and as a
result irradiation only occurs in a narrow portion of the
reaction cells. Thus, the light intensity is reduced in a floses—
type PMGRR thereby reducing the reaction rate of a chemical.
Since the reaction rate of the chemical is reduced, the
67
-------
CS—6010 tOctober, 1 )84)
corresponding half—life is increased. Mill (1984) has indicated
that his Ace—type PMGRR had an effective aperture of 5.35 cm 2
while his Moses—type PMGRR (which he had specially built) had an
aperture of 1.21 cm 2 . Since is proportional to the reciprocal
of the rate constant and is proportional to the li ht
intensity, /2 is proportional to the reciptocal of the light
intensity. Furthermore, since the light intensity is
proportional to the apettute (A in cm 2 ), ty 2 is ptopottiona1 t
the reciprocal of the apertute.
Using the subscript notation Ace to represent the Ace-
type PMGRR and the subscript notation Test to represent the
Moses—type PMGRR, then for a test chemical (c), one can write
Ace El/(Ap)Ace]
Test El/(Ap)Testl
assuming that the same reaction cells are placed in both types of
PMGRR (and thus have the same pathlength).
c,Ace/’ t4 c,TestJ = (Ap)Test/(Ap)Ace
/ c,Test c,Ace1p)Ace” p)TestI (69)
68
-------
CS—6010 (October, L9 4)
From the previous discussions, the criteria for the choice
of actinometer were: if (t is less than 12 hours, use the
PNA/PYR actinometer; if (t],t )c is greater than 12 hours, then use
the PNAP/PYR actinometer. Thus, (t}/ c Ace 12 houts. Since
(Ap)Ace = 5.35 cm 2 , then equation 69 becomes
/ c,Test 641 p Test (70)
where (t1 1 / 2 )c,Test is in hours and A is the aperture in cm 2 .
Therefore, if a Moses—type PMGRR is used in this Test Guideline
(CG—6010), the aperture (An) must be measured and substituted in
equation 70 and the test criteria will be: if (t1/ c,TeSt <
64 /(Ap)Test then use the PNA/PYR actinometer; and if
> 64 /(Ap)Test then use the PNAP/PYR actinometer.
Using the same approach as described above, equations 6 and
64 (for the Pt’4AP/PYR actinometer) and equations 66 and 68 (for
the Pr’ A/PYR actinometer) can be modified as follows. Based on
the above discussions, it can be readily shown that
(kp)c,Test (Ap)Test
(kp)c,Ace (Ap)Ace
(kp)c,Ace = p c,Test Ap)Ace ’ p )ie.sLJ (7U
For the Ace—type PMGRR, the aperture is 5.35 cm 2 [ M i i i ( 1984)1.
Thus, equation 71 becomes
69
0)’
-------
CS—6010 (October, 1984)
(kp)c,Ace 5 . 35 (kp)c,Test/(Ap)Test (72)
where the rate constants are in the units of hours’ and (Ap)Test
is the aperture of the Moses—type PMGRR in cm 2 .
For the PNAP/PYR actinometer, equations 61 arid 64
define the pyridine concentration needed to adjust the rate
constant of the actinometer to equal the rate constant of the
test chemical in the Ace—type PMGRR at 313 and 366 rim,
respectively. The corresponding equations for the Moses-type
PMGRR can he obtained by substituting equation 72 in equations 61
and 64. The tesults are:
at 313 nm
(PYR] G . 2 l(kp)c,Test/(Ap)Test
at 366 nm
[ PYR) = 3 1. 8 (kp)cTest/(Ap)Test (74)
where (kp)c,Test is in the units of bout , (A [ ))l e.s i the
aperture of the Moses—type PMGRR in the units of c i a 2 , and jJ YRJ
is in the units of molar concentration.
Following the same procedure as above fot the PNA/PYR
actinometer, equations 66 and 68 can be modified usinq equation
72. The results are:
70
-------
CS-6010 (October, 1984)
at 313 rim
[ PYR] = 8.93 x io— E5•35(kp)cTest/(Ap)Test1 — 0.0722} ; (75)
at 366 nm
[ PYRJ 1.85 X 10_2{ [ 5.35(kp)c,Test/(Ap)Test} _ 0.03493 (76)
where (kp)c Test is in the units of hour , (Ap)Test is the
aperture of the Moses—type PMGRR in the units of cm 2 and [ PYR] is
in the units of molar concentration.
12. Chemical Analysis of Solutions
a. Chemical Analysis of Test. Chemical Solutions
The analytical techniques employed in the determination
of the concentration of the test substances are left to selection
by the tester. This is in recognition of the many different
techniques available and the practical advantage of being able to
make particular use of one of the properties of the substance;
e.g., the NMR or (JV spectrum of the substance, or its
chromatographic behavior. Analytical techniques that permit the
determination of the test compound to the exclusion of imnputities
or photolysis reaction products are recommended to th xt nt:
practicable. Therefore, chromatOgraPhic techniques;
particularly desirable. Whenever practicable, an anaIyti.c
71
-------
CS—6010 (Octoae , H 4)
‘U ot (1(jUt e qhou id h( i ; ’d which has a ec is i Ofl ol i ‘c k !
better. The specific technique which is utilized should he
completely described.
b. Chemical Analysis of p—Nitroacetophenone (PNAPI
The p—nitroacetophenone (PNAP) in the chemical
actinometer solution is conveniently analyzed by high—pressure
liquid chromatography using a 30 cm C 18 reverse—phase column arid
a uv detector set at 280 rim. The mobile phase in volume percent
is 2.5% acetic acid, 50% acetonitrile, and 47.5% water. which is
passed through the column at a flow tate of 2 mL/minute. This
analytical procedure was specifically developed to analyze for
PNAP, especially at high pyridine concentrations (i.e., for
pyridine concentrations up to approximately 0.2 M) [ Winterle
(1984)1.
c. Chemical Analysis of p—Nitroanisole (P A)
The p—nitroanisole (PNA) in the chemical actinometer
solution is conveniently analyzed by high—pressute li ui t i
chromatography using a 30 cm C 18 reverse—phase column and a uv
detector set at 280 rim. The mobile phase in volume petc nt. is
50% acetonitrile and 50% water which is passed through the column
at a flow rate of 2 mL/minute. This analytical procedura was
developed by Dulin and Mill (1982).
4) ’
72
-------
(2S—bUIU Wctooer,
1. Chemical Analysis of Ferrous Ion in the
Ferrioxalate Actinometer
The molar concentration of Fe 2 + formed in the
photolysis of the ferrioxalate actinometer is measured
spectrophotornetrically via the formation of a red phenanthroline
complex and determining the absorbance at 510 nm. This is the
standard procedure for determining the molar concentration of
Fe 2 formed in the photolysis of the ferrioxalate actinometer.
The procedure described by Murov (1973) is excellent and has been
recommended. The procedure has been modified slightly by using
the concentration of ferrioxalate at 0.15 M. As a result, the
irradiated solution has to be diluted 100—fold prior to analysis
for Fe 2 , which must not he allowed to exceed 0.005 M Izepp
(1984)).
73
-------
CS—6010 (Octcber, 1984)
III. REFERENCES
AA. 1982. Astronomical Almanac.
Barltrop JA and Coyle JD. 1975. Excited states in organic
chemistry. John Wiley and Sons, Inc., New York, New York.
Ralzani V and Carassiti V. 1970. Photochernistry of coordination
compounds. Academic Press, New York, New York.
Benson SW. 1976. Thermochemical kinetics. Second Edition.
Wiley-Interscience, New York, New York.
Calvert JG and Pitts JN, Jr. 1966. Photochemistry. John Wiley
and Sons, Inc.,, New York, New York.
de Mayo P and Shizuka p. 1976. Measurement of reaction quantum
yields. ucreation and Detection of Excited States,” Vol. 4, WR
Ware, Ed. Marcel Dekker, Inc., New York, New York.
Demerjian KL, Schere KL, and Peterson JT. 1980. Theorct.ical.
estimates of actinic (spherically integrated) flux and photolysis
rite constants of atmospheric species in the lower tr posphcre.
Adv Sci and Tech 10:389.
Dulin D and Mill T. 1982. Development and application ot solar
actinometers. Environ Sci and Tech 16:815.
Hatchard CG and Parker CA. 1956. A new sensitive chemical
actinometer II. Potassium ferrioxylate as a standard chemical
actinometer. Proc Royal Soc of London A 235:518.
Leighton PA. 1961. Photochernistry of air pollution. A ii rnic
Press, New York, New York.
Mill T, Davenport JE, Dulin DE, Mabey WR, and F3awol k. 19H1.
Evaluation and optimization of photolysis screening protcu !s.
U.S. Environmental Protection Agency. EPA—560/5—81-0O3.
74
U -.--
-------
CS—6t) It) C üc t r , ‘
Mill T, Mabey WR, Flendry DC, Winterle J, Davenport J, Barich V,
Dulin D, and Tse D. 1982. Design and validation of screening
and detailed methods for environmental process. EPA—
Mill T, frlabey WR, Bomberger DC, Chou T—W, Hendry DG, and Smith
JH. 1982a. Laboratory protocols for evaluating the fate of
organic chemicals in air and water. EPA—600/3—82—022.
Mill T. 1984. Private communication. SRI International, Menlo
Park, California.
Moses FG, Liu RSH, and Monroe BM. 1969. The ‘ merry—jo—cound”
quantum yield apparatus. Mol Photochem l(2):245.
Murov SL. 1973. Handbook of photochemistry. Marcel Dekker,
Inc., New York, New York.
Parker CA. 1953. A new sensitive chemical actinometer I. Some
details with potassium ferrioxylate. Proc Royal Soc of London A
220:104.
Peterson JT. 1976. Calculated actinic fluxes (290—700nm) for
air pollution photochemistry applications. EPA—600/4—76—0025.
Pierce BO. 1929. A short table of integrals. Ginn and Company,
New York, New York.
Pitts JN, Wan JKS.. and Schuck EA. 1964. An o—nitrobenzaldehyde
actinometer and its application to a kinetic study of the
photoreduction of benzophenone by benzhydrol in a pressed
potassium bromide disk. J Am Chem Soc 86:3606.
Smith JH, Mabey WR, Bohonos N, Bolt BR, Lee SS, Chou T—W,
Bomberger DC, Mill T. 1977. Environmental pathways of selected
chemicals in freshwater systems. Part I. Background and
experimental procedures. U.S. Environmental Protection Agency,
Athens, GA. EPA 600/7—77—113.
Smith JH, Mabey WR, Bohonos N, Bolts BR, Lee SS, Chon T—W,
Bomberger DC, Mill T. 1978. Environmental pathway ol
chemicals in freshwater systems. Part I I. Laboratory stidi”s.
U.S. Environmental Protection Agency, Athens, GA. PA
600/7—78—074.
75 ‘) I
-------
CS—6010 (October, 1984)
Vesley CF. 1971. Complications in measuring quantum yielis
using cylindrical cells. Mol Photochern 3(2):193.
Wegner EE and Adamson AW. 1966. Photochemistry of complex ions
III . Absolute quantum yields for the photolysis of some aqueous
chromium (III) complexes. Chemical actinornetry in the lon j
wavelength visible region. J Am Chern Soc 88:394(1966).
Winterle J. 1984. Private communication. SRI International,
Menlo Park, California.
Zepp RG and Cline DM. 1977. Rates of direct photolysis in
aquatic environment. Environ Sci and Technol 11:359—366.
Zepp kG. 1978. Quantum yields for reaction of pollutants in
dilute solution. Environ Sd and Technol 12:327.
Zepp RG. 1982. Experimental approaches to environmental
photochemistry. The handbook of environmental chemistry.
). Hutzinger, Editor. Springer-Verlag, New York, New York.
Zepp RG. 1984. Private communication. Environmental Research
Laboratory, U.S. Environmental Protection Agency, College Station
Road, Athens, Georgia 30601.
•) ‘)
76
-------
CS—6010 (October, 19 4)
IV. APPENDIX: GLOSSARY OP IMPORTANT SYMBOLS
A Wavelength A
Absorhance at wavelength A.
Fraction of light absorbed by the system t
wavelength x
Fraction of light absorbed by a chemical (c)
at wavelength A
Molar absorptivity of a chemical (c) at
wavelength A
Molar absorptivity of an actinometer (a) at
wavelength A
Absorption (or attenuot. ion) coet icieti o
at wave length A
Ł The light pathlength; the distance Lrave eO by a
beam of light passing through the system.
Reaction quantum yield of an aqueous solution of
actinometer (a).
Reaction quantum yield of an aqueous solution of
chemical (C).
Sunlight reaction quantum yield of a chemical (c)
in a water body in the environment.
—d [ C}/dt. Direct photolysis rate of chemical (C).
k F’ Direct photolys is sun 1 ight. rate cc ns LarltT in i w: t ‘ r
p body in t:. he env i ronme n
(k ,) Maximum direct photolysis rate consL n in .i w il e
pF mCx.
body in the environment.
k Direct photolysis rate constant of chemicaL Cc) in
p water measured in the laboratory.
k Direct photolysis rate constant of actinometer (a)
pa in water measured in the laboratory.
Specific light absorption of a photoreactive
chemical at low concentration and at
wavelength A
k Specific light absorption rate constant integrated
a over all. wavelengths absorbed by the chemical.
77
-------
CS—6010 (October, 1984)
(ti 1 ) Half—life o a chemical (c) in water.
C
t , ) Hal — I i fe ot an act i noinett ( a ) in wa t ‘
a
tl Half—life of a chemical in a water body in sunIi ht
in the environment.
)/ min. The minimum sunlight half—life of a chemical in a
water body in sunlight in the environment.
The number of photons (or einstei 9 ) of light of
wavelength in the system per cm per second.
The number of einsteins of light of wavelength
in the system per liter per hour (or second).
Solar irradiance in a water body at shallow depths
in the units milli einsteins per cm 2 per second.
PYR Pyridine.
EPYR) Molar concentration of pyridine.
PNAP p-Nitroacetophenone.
PNA p—Nitroanisole.
0—NB o—Nitrobenzaldehyde
PNAP/PYR p—Nitroacetophenone—pyridine actinometer.
PNA/PYR p—Nitroanisole—pyridine actinometer.
POB Photochemical optical bench.
PMGRR Photochemical “Merry—Go—Round” Reactor.
Aperture in the PMGRR.
78
-------
C —6OlO (Oc ode , 19b4)
(t ,,)C Half—life of a chemical (c) in water.
Half—life of an actinometer (a) in water.
Ualf—life of a chemical in a water body in sunliqht
in the environment.
(t1/ )min The minimum sunlight halL—iife of i chemic ii in i
water body in sunlight in the environm flt.
‘OX The number of photons (or einstein ) o ijt t o
wavelength A in the system per cm pet second.
The number of einsteins of light of wavelertcjt5 X
A in the system per liter per hour (at second).
Solar irradiance in a water body at shaiJow doj c s
fri the units milli einsteins per cm 2 per second.
PYR Pyridir ie.
IPYRI Molar concentration of pyridine.
PNAP p—Nitr oa cetopheflOfle.
PNA p—NittOafliSOle.
0—Nh o_ itrohenzaldehyde
PNAP/PYR p_NitroaCetOpheflOne—PYtidl fle act inometar.
7N \/PYR p —NitroaniSOle—pyridifle actinometet.
P0 5 Photochemical optical bench.
PMGRR Photochemical “Merry—Go--ROun& Reactor.
Aperture in the P ’1GRR.
78
------- |