Acetone-photosensitized reduction of carbon tetrachloride by 2-propanol in aqueous solution

Article abstract of DOI:10.1021/es9902120

The kinetics and mechanism of the reductive dehalogenation of CCl₄ to chloroform were studied in the presence of aqueous acetone and 2-propanol, which was added as a hydrogen donor. The rate of reduction was very fast, stoichiometrically converted 3 mM CCl₄ to chloroform in ~ 2 min. The zero-order reaction occurred at 2.7 x 10 ^-6 M/sec rate under typical conditions (0.69 M acetone, 5.7 M 2-propanol, 3 mM CCl₄) using a 75 W Xe lamp. The mechanism was first order in CCl₄ near the end of the reaction. The zero-order reaction rate increased with acetone, 2-propanol, and with absorbed light intensity. Other combinations of carbonyls, alcohols, and halogenated organics such as methyl ethyl ketone, methanol, ethanol, 1-propanol, and 1-butanol are also effective in dehalogenation via this process. Relative rates of dehydrogenation for a series of chlorinated methanes and ethanes such as trichloroethylene and perchloroethylene are presented.

Full text of DOI:10.1021/es9902120

Environ. Sci. Technol. 2000, 34, 1229-1233
The electron distribution in the excited states plays a
crucial role in determining chemical reactivity (1-4). Both
and T states are n, π* in character, meaning that the
Acetone-Photosensitized Reduction
of Carbon Tetrachloride by
S
1
1
unpaired n-orbital electron is localized at the O-atom, while
the lone π*-orbital electron is delocalized over the functional
group. The result is that the O-atom is radical-like, behaving
much like an alkoxy radical in H-abstraction reactions.
In the presence of molecular oxygen, ketones can be used
to sensitize the photooxidation of alcohols (4, 5). Oxygen
insertion into the alcohol results in the formation of a
hydroxyhydroperoxide that readily hydrolyzes in aqueous
solution to generate a new carbonyl (as distinct from the
original photosensitizer) and hydrogen peroxide. Although
the reaction is normally viewed as photooxidation of the
alcohol substrate, it can just as easily be viewed as photo-
2
-Propanol in Aqueous Solution
E R I C A . B E T T E R T O N *
Department of Atmospheric Sciences, University of Arizona,
Tucson, Arizona 85721-0081
N A D I A H O L L A N , R O B E R T G . A R N O L D ,
S T E F A N G O G O S H A , K R I S T A M C K I M , A N D
Z H I J I E L I U
Department of Chemical and Environmental Engineering,
University of Arizona, Tucson, Arizona 85721
reduction of O
2
. In principle, this photochemical system
should be capable of reducing many target species besides
O
2
, depending on the redox potential of the target.
Here we apply some of the fundamental knowledge gained
Simple carbonyl chromophores can be used to facilitate
the solar-promoted reductive dehalogenation of halo-organics.
Here we describe the kinetics and mechanism of the
reductive dehalogenation of carbon tetrachloride (CT) to
chloroform (CF) in the presence of aqueous acetone and
on ketone photochemistry over the past four decades to the
problem of remediating solvent-contaminated water. Specif-
ically, we describe, apparently for the first time, the kinetics
and mechanism of the photosensitized reductive dehalo-
genation of carbon tetrachloride to chloroform (CF) by
2
-propanol (iP) in the presence of acetone (A) in aqueous
2-propanol, which is added as a hydrogen donor. The reaction
solution. The rate of reduction is extremely rapid (e.g., 3 mM
CT can be destroyed in 2 min by exposure to sunlight), and
quantum yields are high. The overall reaction is
is rapid in sunlight, stoichiometrically converting 3 mM
CT to CF in approximately 2 min. Under typical conditions
(0.69 M acetone, 5.7 M 2-propanol, 3 mM CT) using a
7
5 W Xe lamp, the observed zero-order reaction occurs
CCl + (CH ) CHOH f CHCl + (CH ) CO + HCl
4
3 2
3
3 2
-
6
-1
at a rate of 2.7 × 10 M s . The mechanism appears to
become first order in CT near the end of the reaction.
The zero-order reaction rate increased with both acetone-
and 2-propanol concentration and also with absorbed
light intensity. Other combinations of carbonyls, alcohols,
and halogenated organics are also effective.
-
1
∆G° ) -115 kJ mol (4)
Although the liquid-phase photolysis of 1,4-dichlorobu-
tane (DCB) is sensitized by singlet (n, π*) acetone, the process
reportedly occurs by collisional deactivation of excited
acetone by DCB and not via H-atom abstraction (8, 9).
It is too soon to determine if carbonyl-photosensitized
treatment systems will be of any practical use, but ongoing
work (to be reported elsewhere) shows that the system is
effective for a wide range of halogenated alkanes and alkenes.
Combinations of other carbonyls (aromatic and aliphatic
aldehydes and ketones) and alcohols are also useful.
Introduction
Carbonyl photochemistry, both in the gas and liquid phases,
has been studied in remarkable detail over the course of
several decades and has been the subject of numerous articles
and reviews (e.g., see refs 1-7, and references therein). As
a result, ketones are well-known to undergo three types of
photochemical reactions: (i) Type I cleavage of a bond R to
the carbonyl chromophore (eq 1), (ii) intermolecular and
intramolecular hydrogen abstraction (eq 2); and (iii) addition
of the excited-state carbonyl to olefins (eq 3).
Experimental Section
Materials. Deionized water was further purified to >18MΩcm
by passage through a Millipore Milli-Q system. Acetone
(
Fisher, HPLC grade), 2-propanol (Mallinckrodt, AR), and
carbon tetrachloride (ACROS, 99.8%) were all used as
received. Other ketones and alcohols tested in this study
were all >99% purity (Aldrich), except for 2,3-butanedione
(
biacetyl) which was 97% (Aldrich). GC/ FID analysis showed
that the 2-propanol contained traces of acetone (0.3mM)
which was taken into account in analyzing the results of the
subsequent photolysis experiments.
Methods. Polar organics (acetone and 2-propanol) were
determined in the headspace by gas chromatography (GC)
(
Hewlett-Packard 5880; DB-WAXETR 30m × 1.5 mm; He;
isothermal 60 °C; FID; HP5880A integrator), while haloge-
nated organics were determined in pentane extracts (2.5 µL:
1
.27 mL) by GC/ ECD (Hewlett-Packard 7673 auto injector;
HP 5790; DB 624 30m × 1.5 mm; He, isothermal 40-100 °C,
depending on analyte; HP 3392A integrator). Unidentified
peaks were analyzed by GC/ MS (Varian 3400; Finnigan
MAT700 ion trap detector; Megabore DB-5 60 m). Chloride,
the only anion detected, was determined by ion chroma-
Here we focus solely on the H-abstraction reaction and
how it may be employed to reductively dehalogenate carbon
tetrachloride (CT) in a photocatalytic system.
*
Corresponding author phone: (520)621-2050; fax: (520)621-6833;
e-mail: better@atmo.arizona.edu.
2
tography (IC) (Dionex DX-100; AG4A, AS4A; 0.85 mM Na -
1
0.1021/es9902120 CCC: $19.00

2000 Am erican Chem ical Society
VOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 2 2 9
Published on Web 02/29/2000

3
CO eluant). Electronic absorption spectra of ketones and
solvents were recorded on a Hitachi U-2000 spectropho-
tometer using 1 cm rectangular quartz cuvettes. Hydrogen
peroxide was determined spectrophotometrically after ad-
dition of titanium sulfate (10).
Solar photolysis experiments were performed by exposing
the contents of a sealed 17-mL borosilicate glass serum vial
(
Wheaton) to sunlight. The vial was placed over a darkened
surface and supported at an angle perpendicular to the solar
beam. The diameter of the cylindrical vial was 2.5 cm. Acetone
has a low extinction coefficient where the n-π* band overlaps
with the solar spectrum (see later) resulting in relatively high
transmission (>10%T) for the solar experiments, which were
therefore probably not light-limited, in contrast to the
experiments conducted with the Xe arc lamp source. The
actinic flux (≈295-320 nm) reaching the contents of the vial
1
5
-1
was estimated to be ≈6 × 10 photon s , after accounting
for the transmittance of the glass wall (determined by UV-
visible spectrophotometry) but ignoring possible effects of
reflection and refraction.
Most experiments were conducted using an Oriel 75W Xe
arc lamp assembly consisting of a fused silica collimating
lens, a 10-cm IR water filter, a 400-nm cut off filter to remove
visible light, an adjustable iris diaphragm, and a fiber input
adapter which focused light onto a ≈1 m × 1 mm core optical
fiber (3M Corp.). The fiber passed through a septum into a
FIGURE 1. (a) Results of a typical experiment showing mainly
zero-order acetone-photopromoted reduction of carbon tetrachloride
14
-1
to chloroform (75 W Xe lamp, I
.69 M; [iP] ) 5.7 M; optical depth ) 2.2 M cm; anoxic). Linear
regression parameters for CT loss rate: slope ) -(1.43 ( 0.03) ×
A
) 6.6 × 10 photon s ; [A] )
0
1
7-mL glass vial reactor. Some cladding (2-4 mm) was
removed from the fiber tip that was positioned 3.0 or 3.7 cm
above the bottom of the vial, depending on the liquid volume
-
1
2
1
0 ; intercept ) 2.92 ( 0.03; r ) 0.998; n ) 9. (b) Results of a
similar experiment showing reduction of 10-fold higher initial
chloroform concentration that begins as a zero-order decay but
apparently becomes first-order after approximately 100 minutes.
Linear regression parameters for CF loss rate: slope ) -(1.8 (
(
13.7 or 17.0 mL). The contents were stirred with a glass-
coated magnetic stir bar. Reactions were studied only at room
temperature (≈25 °C).
In a typical photolysis experiment, 13.7 mL of an aqueous
mixture of 2-propanol (5.7 M) and acetone (0.69 M) was added
-1
1
2
0
.06) × 10 ; intercept ) (2.83 ( 0.03) × 10 ; r ) 0.996; n ) 6.
Exponential fit parameters for CF loss rate: y ) (1.1 ( 0.2)) ×
to the vial, purged with N
2
or He for 10-20 min, and sealed
2
-2
2
1
0 × exp[-(2.3 ( 0.1) × 10 ]x; r ) 0.983; n ) 5.
with a Teflon-coated rubber septum. No significant loss of
acetone due to volatilization could be detected by GC/ FID
over this time period. The vial was wrapped in aluminum
foil to protect the contents from light prior to injecting the
desired amount of CT. The reaction was initiated by exposure
to light (solar or Xe arc lamp). Liquid-phase samples were
withdrawn at known intervals and analyzed by GC or IC. The
high solubility of CT in the alcoholic solution obviated the
problem of significant headspace losses.
0.1 µM s-1) and CF (3.0 ( 0.1 µM s-1) were similar under
these low-intensity light conditions. It should be noted that
this was not the case during solar photolysis when the zero-
order rates were dependent on the nature of the target
halocarbon (Table 1). The quantum yield for CT loss (ΦCT
)
was ≈30, indicating the existence of an efficient chain
A
reaction. Here we define ΦCT ) (-d[CT]/ dt)/ I , during the
zero-order period.
The photon flux into the solution from the end of the
optical fiber, determined by ferrioxalate actinometry, was
The generally accepted mechanism (4) for the photo-
sensitized reduction of O forms the basis for our proposed
mechanism for the CT-containing system.
2
1
5
-1
5
.1 × 10 photon s , assuming an average quantum yield
of 1.22 over the photoactive wavelength range of 200-400
nm (12). However, we were primarily interested in the flux
absorbed by the n-π* electronic band (=210-310 nm), which
was estimated in the following way. The water in the 10-cm
IR filter was replaced with 0.22 M aqueous acetone (optical
depth ) [acetone] × path length ) 2.2 M cm), and the
ferrioxalate actinometry was repeated. The flux absorbed by
A + hν f A* IA
(R1)
A* + iP f 2iP k ) 9.7 × 10 M s-1 (15, 20) (R2)
•
5
-1
2
•
•
iP + CT f A + CT + HCl
k3 ) 1.8 × 10 M s-1 (18, 19) (R3)
8
-1
1
4
-1
the acetone n-π* band was found to be 6.6 × 10 photon s
,
i.e., 13% of the original incident flux. By arranging all
subsequent Xe lamp experiments to maintain a reactor optical
depth of 2.2 M cm (e.g., 0.69 M acetone × 3.2 cm), the
CT + iP f CF + iP k ≈ 1.4 × 10 M s-1 (17) (R4)
•
•
2
-1
4
photolytically active flux, I
A
-8
, was always known to be 6.6 ×
iP f A + iP k ≈ 4 × 10 M s-1 (14)
•
8
-1
2
(R5)
5
1
4
-1
-1
1
0
photon s (7.5 × 10 M s ).
NET: iP + CT f A + CF + HCl
Results and Discussion
Using the Xe lamp, the disappearance of CT was typically
accompanied by stoichiometric production of CF (Figure
Excitation occurs in the first step, where ground-state
acetone (A) absorbs a photon, undergoes an n-π* transition
-
+
1
a), Cl , and H (data not shown). The reaction was zero-
to the short-lived excited singlet state (S
1
), and rapidly
undergoes intersystem crossing (10 s (1, 16, 17)) to triplet
state (T ) A* with a quantum efficiency close to unity. Although
9
-1
order in CT until about 90% completion after which the
reaction order appeared to change, possibly becoming first-
order near the end. This is more easily seen during the
photoreduction of a more recalcitrant target such as CF
1
the overlap of the acetone n-π* electronic absorption band
(=210-310 nm) with the solar spectrum at the earth’s surface
is not great (Figure 2), it is sufficient to promote rapid
(
Figure 1b). As expected, the reduction rates for CT (2.7 (
1
2 3 0
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 7, 2000

TABLE 1. Comparison of Photopromoted Zero-Order Loss Rates
for Some Chloroalkanes and -alkenes under Comparable
a
Conditions in Sunlight
target species
relative rate
CCl4 (CT)
CHCl3 (CF)
CCl3CHCl2 (PCA)
CHCl2CHCl2 (1,1,2,2-TECA)
CCl3CHCl2 (1,1,1,2-TECA)
CHCl2CH2Cl (1,1,2-TCA)
CCl3CH3 (1,1,1-TCA)
CHCl2CH3 (1,1-DCA)
CHCldCCl2 (TCE)
1.0
(1.4 ( 0.1) × 10
-
1
-
1
(5.1 ( 0.41) × 10
-
1
2
1
(1.48 ( 0.03) × 10
-
(5.1 ( 0.41) × 10
-
3
(3.0 ( 0.5) × 10
-
(2.12 ( 0.04) × 10
-
-
1
2
(3.9 ( 0.4) × 10
(4.2 ( 0.7) × 10
(6 ( 3) × 10-
2
CCl2dCCl2 (PCE)
a
0
.69 M acetone; 5.7 M 2-propanol; 8:30 a.m . local tim e, 8/16/97.
Under these conditions the absolute rate for CT reduction was 1.0 ×
-5
-1
1
0
M s .
FIGURE 3. The exponential (Beer’s law) decrease in carbon
tetrachloride loss rate and chloroform production rate as the
concentration of acetone in the optical filter was increased
demonstrates that the n-π* wavelength range is of prime importance
1
4
-1
in this system (75 W Xe lamp, I
A
) 6.6 × 10 photon s ; [A] ) 0.69
M; [iP] ) 5.7 M; optical depth ) 2.2 M cm; anoxic). Exponential fit
parameters for CT loss rate: y ) (1.6 ( 0.2)) × 10 × exp[-(3.3 (
1
-
1
2
0
.3) × 10 ]x; r ) 0.984; n ) 6.
FIGURE 2. Despite the poor overlap of the acetone n-π* band with
the calculated solar spectrum at the earth’s surface, the rate of
carbon tetrachloride reduction is notably rapid in sunlight.
dehalogenation of CT when coupled with the long photo-
chemical chain lengths.
Triplet A* behaves much like an alkoxy radical and so
readily abstracts hydrogen from suitable donors (2). The rate
of H-abstraction from 2-propanol (iP) is much faster than
from other potential donors such as water or ground-state
acetone (1, 16). Consequently, we propose that the chain
FIGURE 4. Logarithmic increase of reaction rate with acetone
concentration when the system is not necessarily light-limited is
explained through Beer’s law dependence of absorbed light intensity,
reaction is initiated only in R2 where two identical 1-hydroxyl-
•
A
I , in the reactor.
1
-methylethyl ketyl (iP ) radicals are generated sone from
the alcohol and one from the ketone. If different ketones
and/ or alcohols are chosen, then two different radicals will
be produced.
detected (<400 nM), we assume that R6, R7, and R8 are
unimportant under our experimental conditions.
The following experiment showed that it is the n-π*
A
wavelengths that are effective. I was varied by filling the IR
water filter with a series of aqueous acetone solutions of
known concentration to selectively filter out the useful n-π*
wavelengths. The resulting plot of reaction rate vs filter
acetone concentration is exponential at fixed reactor acetone
concentration (Figure 3) as expected from a consideration
of Beer’s law.
•
The iP radical is a powerful reductant ((CH
3
)
2
)CO/
•
(
CH
3
)
2
C OH ) -1.81 V vs NHE) (13) that propagates the chain
reaction in R3 where it is proposed that electron donation
to CT is accompanied by Cl elimination. This is followed by
H-atom abstraction by the trichloromethyl radical (CT ) and
-
•
production of chloroform (CF) in R4.
The only termination step in the proposed mechanism is
iP radical disproportionation (R5). Other possible termina-
•
By contrast, the Beer’s law dependence of the absorbed
light intensity (I ) in the reactor results in a logarithmic
A
tion steps include the radical-radical recombination reac-
tions R6, R7, and R8.
dependence of the zero-order reaction rate on [A] over a
five-order of magnitude concentration range (Figure 4). At
sufficiently high [A] the rate should eventually become
independent of [A] when every photon is absorbed. However,
this could not be demonstrated because it was not possible
to prepare a solution that contained more than 6.9 M acetone
and yet maintain the required 5.7 M iP (pure acetone has a
concentration of 13.6 M). This [A] corresponds to absorption
of 99% of the actinic flux.
CT f C Cl k ) 3.7 × 10 M s-1 (22) (R6)
•
8
-1
2
2
6
6
•
•
CT + iP f CCl C(CH ) COH k7
(R7)
3
3 2
•
2
iP f (CH ) C(OH)C(OH)(CH ) k ) k / 3.4 (14) (R8)
3 2 3 2 8 5
Since the expected chlorinated products, C
CCl C(CH OH, were not detected (<10 nM) by GC/ ECD or
3 2
GC/ MS, respectively, and nor was (CH C(OH)C(OH)(CH )
2
Cl
6
or
Using the Xe lamp, the standard reaction is characterized
by two phases: an initial period that is zero-order in CT
(>90% of the reaction) followed by a second period that is
3
3 2
)
3
)
2
VOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 2 3 1

2
the original O contained in the reactor, including the
-
6
headspace, i.e., 2.6 × 10 mol.
The predicted net 1:1 stoichiometric ratio of carbonyl
production (R3) compared to CF production (R4) was
confirmed by switching to ethanol, which yields acetaldehyde
that can more easily be detected against the high background
concentration of acetone (instead of 2-propanol which yields
acetone). The acetaldehyde concentration was found to
increase from zero to 1.2 mM as CF increased to 1.4 mM
under our standard conditions.
Other chlorocarbons besides CT can also be dehaloge-
nated via this process. Relative rates of dehalogenation for
a series of chlorinated methanes and ethanes, obtained under
identical conditions in sunlight, are summarized in Table 1.
The substituted methanes were apparently dehalogenated
at a rate that was inversely related to the C-Cl bond energy
(
CT > CF). Among the ethanes, reduction occurred more
readily at the most heavily substituted carbon atom. For
example, 1,1,1,2-TECA is reduced 30 times faster than 1,1,2,2-
TECA, while 1,1,1-TCA reduction was 70 times faster than
FIGURE 5. Zero-order reaction rates for both carbon tetrachloride
loss and chloroform production increase linearly with 2-propanol
14
-1
concentration (75 W Xe lamp, I
A
) 6.6 × 10 photon s ; [A] )
1
,1,2-TCA. Although PCE and TCE were degraded, no reaction
0
.69 M; [CT] ) 3 mM; optical depth ) 2.2 M cm; anoxic). Linear
0
-
products other than Cl could be identified. The expected
hydrogenolysis products, TCE and DCE, respectively, were
not observed.
-
2
regression parameters for CT loss rate: slope ) (2.3 ( 0.1) × 10 ;
-
2
2
intercept ) (1.7 ( 0.7) × 10 ; r ) 0.997; n ) 12.
Combinations of other carbonyls, including methyl ethyl
ketone, 2-pentanone, 4-methyl-2-pentanone, biacetyl, ben-
zophenone, acetophenone, methanol, ethanol, 1-propanol,
and 1-butanol, are also effective and are currently under
investigation.
apparently first-order in CT (Figure 1). This suggests the
existence of two different rate-limiting steps, depending on
[CT]. We consider each phase in turn. During the zero-order
period, when [CT] is still high and the propagation steps R3
and R4, in particular, are fast, the overall loss of CT is limited
by an earlier step in the mechanism. This is postulated to be
R2 because of the observed first-order dependence on [iP]
under these conditions (Figure 5). The slope of this plot is
Acknowledgments
This publication was made possible by grant number P42
ES004940 from the National Institute of Environmental
Health Sciences, NIH with funding provided by EPA. We thank
Brian Barbaris and Kristen Taylor for their assistance in the
laboratory and Jennifer Slack and Cynthia Malbrough for
help in preparing the manuscript.
2
therefore equal to k A*. Since water and ground-state acetone
are both poor H-donors (1), initiation reactions equivalent
to R2 do not occur in the absence of iP, and the intercept is
within 2-3 SDs of zero. No further kinetic information
concerning subsequent steps in the proposed reaction
mechanism can be extracted from the kinetic data during
the zero-order period since R2 is the rate-limiting step.
If we next focus on the second phase (last 10% of the
reaction), when [CT] is low due to reduction to CF, the
reaction becomes first-order in [CT]. Here it is postulated
that a later step in the proposed mechanism, i.e., R4, becomes
Literature Cited
(1) Scaiano, J. C. J. Photochem. 1973/74, 2, 81-118.
(2) Cowan, D. O.; Drisko, R. L. Elements of Organic Chemistry;
Plenum Press: New York, 1976.
(
3) Churio, M. S.; Greal, M. A. J. Chem. Educ. 1997, 74, 436-438.
rate-limiting. This is reasonable since k is 3-5 orders of
4
magnitude smaller than the other second-order rate con-
stants. During this first-order CT period a kinetic analysis
becomes more useful. The rate law can be written as
(4) Coxon, J. M.; Halton, B. Organic Photochemistry, 2nd ed.;
Cambridge University Press: London, 1987.
(
5) Rubin, M. B. CRC Handbook of Organic Photochemistry and
Photobiology; Harspool, W. M.; Song, P.-S., Eds.; CRC Press:
Boca Raton, 1995; pp 430-436.
•
ν ) -d[CT]/ dt ) d[CF]/ dt ) k [CT][iP ]
(7)
(6) Calvert, J. G.; Pitts, J. N., Jr. Photochemistry; Wiley: New York,
966.
3
1
(
7) Wagner, P.; Park, B.-S. Organic Photochemistry; Padwa, A., Ed.;
Marcel Dekker: New York, 1991; Vol. 2, pp 227-366.
8) Golub, M. A. J. Am. Chem. Soc. 1970, 92, 2615-2619.
9) Golub, M. A. J. Am. Chem. Soc. 1969, 91, 4925-4926.
Under conditions of steady illumination, making pseudo-
steady-state assumptions for [A*], [iP ], and [CT ] results in
the following rate law.
•
•
(
(
(
(
(
10) Schumb, W. C. Hydrogen Peroxide; Reinhold: New York, 1955;
pp 548-549.
11) Seinfeld, J. H. Atmospheric Chemistry and Physics of Air Pollution;
1
/ 2
ν ) k (2I / k ) [CT]
(8)
3
A
5
Wiley-Interscience: New York, 1986.
Thus, under pseudo-steady-state conditions, the reaction
becomes first-order in [CT], as observed (Figure 1).
A 120-min time lag was observed before CT reduction
12) Rabek, J. F. Experimental Methods in Photochemistry and
Photophysics; Wiley-Interscience: Chichester, 1982; Part 2, p
9
44.
began when the reactor was purged with air instead of He.
(
(
13) Wardman, P. J. Phys. Chem. Ref. Data 1989, 18, 1637-1755.
14) Blank, B.; Henne, A., Laroff, G. P.; Fischer, H. Pure Appl. Chem.
•
This arises because O
). After the O
of CT proceeded, but at a higher rate than expected based
on the results of O -free experiments. A rate increase could
occur if the O reaction produced more carbonyl chromo-
phore. Hydrogen peroxide should be produced stoichio-
metrically from the dissolved O . The observed H of
3.5 × 10 mol was reasonably close to that estimated for
2
is a better iP scavenger than CT (1-
4
2
was consumed by iP oxidation, the degradation
1
975, 41, 475-494.
(
(
15) Porter, G.; Dogra, R. O.; Loutfy, S. E.; Sagumori, S. E.; Yip, R. W.
Faraday Trans. 1973, 69, 1462-1474.
16) Turro, N. J. Modern Molecular Photochemistry; Benjamin
2
2
Cummings: 1978; p 91.
2
2
O
2
(17) Alfassi, Z. B. Chemical Kinetics of Small Organic Radicals; CRC
Press: Boca Raton, 1988; Vol. 3, pp 41-69.
-
6
≈
1
2 3 2
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 7, 2000

(
(
18) Stefan, M. L.; Hoy, R. H.; Bolton, J. R. Environ. Sci. Technol.
996, 30, 2382-2390.
19) Steacie, E. W. R. Atomic and Free Radical Reactions, 2nd. ed.;
Reinhold: New York, 1954; Vol. 1.
(22) Ross, A. B.; Neta, P. Rate constants for reactions of aliphatic
carbon-centered radicals in aqueous solution. Gov. Doc. #
NSRDS-NBS 70; Department of Commerce, National Bureau of
Standards: 1982.
1
(
(
20) Backstrom, H. L. J.; Sandros, K. J. Chem. Phys. 1955, 23, 2197.
21) Farhataziz, Ross, A. B. Selected specific rates of reactions of
transients from water in aqueous solution: III. Hydroxyl radical
and perhydroxyl radical and their radical ions; Gov. Doc. #
NSRDS-NBS 59; Department of Commerce, National Bureau of
Standards: 1977.
Received for review February 22, 1999. Revised manuscript
received November 8, 1999. Accepted January 3, 2000.
ES9902120
VOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1 2 3 3