Mechanistic Investigations Concerning the Aqueous Ozonolysis of Bromacil

Article abstract of DOI:10.1021/jf9600420

Bromacil ozonolysis was examined to determine the mechanism of product formation in an effort to optimize a chemical-microbial remediation strategy for contaminated waters. Two debrominated products, 3-sec-butyl-5-acetyl-5-hydroxyhydantoin (II) (24%) and 3-sec-butylparabanic acid (III) (56%), and a dibromohydrin, 3-sec-butyl-5,5-dibromo-6-methyl-6-hydroxyuracil (IV) (20%), were formed. The latter compound, arising from HOBr addition to bromacil, reverted back to starting material, causing the treated solution to remain somewhat phytotoxic. Mass balance studies provided evidence for parallel reaction pathways as opposed to a series pathway where II gives rise to III. Addition of hydrogen peroxide slightly decreased the rate of bromacil degradation while the addition of tert-butyl alcohol (t-BuOH), a hydroxy radical scavenger, increased the degradation rate, strongly suggesting that the mechanism does not involve hydroxy radicals but direct ozone attack at the double bond. A much lower yield of IV, 6%, relative to the control was observed with H₂O₂, whereas a slightly higher yield, 23%, was found with t-BuOH.

Full text of DOI:10.1021/jf9600420

1006
J. Agric. Food Chem. 1997, 45, 1006−1011
Mech a n istic In vestiga tion s Con cer n in g th e Aqu eou s Ozon olysis of
Br om a cil
,
†
†,‡
‡
†,§
Cathleen J . Hapeman,* Brent G. Anderson, Alba Torrents, and Aurel J . Acher
Natural Resources Institute, Agricultural Research Service, U.S. Department of Agriculture,
Beltsville, Maryland 20705, and Environmental Engineering Program, Department of Civil Engineering,
University of Maryland, College Park, Maryland 20742
Bromacil ozonolysis was examined to determine the mechanism of product formation in an effort to
optimize a chemical-microbial remediation strategy for contaminated waters. Two debrominated
products, 3-sec-butyl-5-acetyl-5-hydroxyhydantoin (II) (24%) and 3-sec-butylparabanic acid (III)
(56%), and a dibromohydrin, 3-sec-butyl-5,5-dibromo-6-methyl-6-hydroxyuracil (IV) (20%), were
formed. The latter compound, arising from HOBr addition to bromacil, reverted back to starting
material, causing the treated solution to remain somewhat phytotoxic. Mass balance studies
provided evidence for parallel reaction pathways as opposed to a series pathway where II gives rise
to III. Addition of hydrogen peroxide slightly decreased the rate of bromacil degradation while the
addition of tert-butyl alcohol (t-BuOH), a hydroxy radical scavenger, increased the degradation rate,
strongly suggesting that the mechanism does not involve hydroxy radicals but direct ozone attack
at the double bond. A much lower yield of IV, 6%, relative to the control was observed with H2O2,
whereas a slightly higher yield, 23%, was found with t-BuOH.
Keyw or d s: Bromacil; remediation; bromohydrin; ozonolysis
INTRODUCTION
1979). A comparison of the various oxidation methods,
direct and indirect photolysis and ozonation, of bromacil
was recently conducted (Acher et al., 1994). Treated
solutions were subsequently submitted to biomineral-
ization using soil, activated sludge, and a previously
isolated microorganism, Klebsiella terragena (strain
DRS-I) (Hapeman et al., 1995). All oxidized waste
solutions showed enhanced biodegradation relative to
the parent compound, and in general the phytotoxicity
was removed. However, some phytotoxic effects were
observed from the ozonated bromacil solutions. These
data would suggest that ozonation be precluded in
oxidation/biomineralization treatment strategies for
bromacil; however, inherent problems also exist with
photolytic oxidation. Waste systems that are opaque,
for example, will not transmit light, which will impede
degradation; furthermore, the photosensitizer, methyl-
ene blue, may be regulated in some areas, presenting
an additional remediation issue.
Efforts have been underway to develop strategies to
remediate pesticide-contaminated matrices, including
ground and surface waters, irrigation and greenhouse
effluents, and waste waters. Early observations showed
that pesticide-containing solutions pretreated with ul-
traviolet light or ozone were more readily degraded by
soil microflora and were significantly less phytotoxic
(Saltzman et al., 1982; Somich et al., 1990); this led to
development of binary (chemical-microbial) remedia-
tion processes (Somich et al., 1990; Hapeman-Somich,
1
1
991; Acher et al., 1994; Hapeman et al., 1995; Massey,
995). Bromacil (5-bromo-3-sec-butyl-6-methyluracil, I),
the pesticide of interest in this study, is a nonselective
and relatively stable herbicide inhibiting photosynthesis
and is used for general weed control on noncrop lands
and citrus orchards (Worthing and Hance, 1991).
Photodecomposition of bromacil has been investigated
in aqueous solutions (Kearney et al., 1969) (λ ) 254 nm)
and in thin solid films (λ ) 254 nm) (J ordan et al., 1965)
although the products were not identified. Experiments
simulating natural conditions by exposure of aqueous
solutions to direct solar irradiation (summer) for 4
months yielded only 2.2% of a single dealkylated pho-
toproduct, indicating that bromacil is very stable toward
sunlight (Moilanen and Crosby, 1974). Addition of
methylene blue, a photosensitizer and generator of
singlet oxygen, to an aerated aqueous solution (Acher
and Saltzman, 1980) substantially increased the bro-
macil degradation rate. The two principal products
were identified as 3-sec-butyl-5-acetyl-5-hydroxyhydan-
toin and a dimer linked at C5 (Acher and Dunkelblum,
The remaining phytotoxicity in the ozonated bromacil
solutions was attributed a priori to the formation of the
somewhat unstable dibromohydrin (Acher et al., 1994).
Formation of the dibromohydrin is novel in that bro-
mohydrins formed during ozonation typically arise from
bromide present in the initial system (Cavanaugh et al.,
1
992; Glaze et al., 1993). In this case, however, bromide
is an ozonolysis byproduct that is subsequently involved
in reactions with starting material. The focus of the
current study is to elucidate the mechanism of bromacil
oxidation when treated with ozone in an effort to
minimize the formation of the phytotoxic oxidation
product, i.e., the dibromohydrin. In addition, by deter-
mining whether degradation proceeds via direct ozo-
•
nolysis or hydroxy radical ( OH), the conditions can be
altered so as to optimize the oxidation rate.
EXPERIMENTAL PROCEDURES
*
Corresponding author [e-mail: chapeman@asrr.
arsusda.gov; fax: (301) 504-5048].
†
U.S. Department of Agriculture.
University of Maryland.
On sabbatical leave from Volcani Center, Institute
‡
Rea gen ts. Bromacil was obtained gratis from Agan Ltd.,
Israel. Recrystallization from 2-propanol yielded chromato-
graphically pure compound (needle crystals, mp 158-159 °C).
§
of Soils and Water, Bet Dagan 50250, Israel.
S0021-8561(96)00042-8 CCC: $14.00
© 1997 American Chemical Society

Mechanism of Bromacil Ozonolysis
J. Agric. Food Chem., Vol. 45, No. 3, 1997 1007
F igu r e 2. Products of bromacil ozonolysis.
runs to determine the concentration of II, III, and IV. The
concentration of bromacil was obtained using a standard curve
calculated from known concentrations of bromacil.
1
4
Liqu id Ch r om a togr a p h y/Ma ss Sp ectr om etr y. LC/MS
electron ionization (EI) spectra (70 eV) were obtained on a
Hewlett-Packard Model 5988A mass spectrometer with 3.0
Pascal software equipped with a Hewlett-Packard Model
F igu r e 1. Total counts in HPLC eluent of [2- C]bromacil
ozonolysis reaction mixture. Peaks at t
6
4
R
) 1.1, 3.8, 4.4, and
.4 min correspond to II, III, bromacil, and IV, and areas of
8, 275, 235, and 102 dpm‚min, respectively.
5
9980A particle beam LC/MS interface (desolvation chamber
temperature ) 50 °C; source temperature ) 200 °C). LC
separations were achieved employing a Zymark (Hopkinton,
MA) Encore HPLC system equipped with the previously
described Beckman C-18 analytical column using a 5 min
linear gradient of 20-40% acetonitrile in acetic acid buffer (pH
Stock solutions of bromacil were prepared with ultrapure
water (18 MΩ/cm, Modulab, type I HPLC, Continental Water
System Corp., San Antonio, TX).
Ozon olysis of Br om a cil. Ozonolysis experiments were
carried out at room temperature in a previously described 550
mL reactor (Somich et al., 1988). Ozone was generated using
a PCI Model GL-1B ozone generator (PCI Ozone Corp., West
Caldwell, NJ ) with oxygen feed. Ozone (in oxygen) was fed
into the reactor at a constant rate of 1 L/min; the stream was
maintained by use of mass flow controllers. Ozone concentra-
tions in the feed line (ca. 1.3% w/w ozone/oxygen) and the off
gas line were determined using an ozone monitor (Model HC-
4
) at a rate of 0.4 mL/min.
Ozon olysis in th e P r esen ce of Hyd r ogen P er oxid e.
Solutions of 382 µM bromacil and 0, 0.069, 3.4, or 6.9 mM H
O
2 2
were subjected to ozonolysis in a manner similar to that above
on the same day under exactly the same conditions, i.e., the
generator was stabilized prior to the first run and was not shut
down between runs. Several sets of experiments were con-
ducted, giving rise to similar results. Rates were determined
1
2, PCI Ozone Corp.). Ozone concentrations in solution (0-11
2 2
relative to the control (no H O ) run on the same day.
µM) were measured using the indigo method (Bader and
Hoign e´ , 1981). The amount of ozone consumed in the reaction
was calculated by subtraction of the off gas and aqueous
concentrations from the feed line concentration. Typical
consumption ranged from 3 to 5 mol of ozone consumed per
mole of bromacil degraded. Samples were removed at various
intervals, purged with nitrogen, and immediately analyzed by
HPLC as described below.
Ozon olysis in th e P r esen ce of ter t-Bu tyl Alcoh ol.
Solutions of 382 µM bromacil and 0, 13.3, or 52.9 mM t-BuOH
were subjected to ozonolysis on the same day under exactly
the same conditions in a manner similar to that above. Several
sets of experiments were conducted, giving rise to similar
results. Rates were determined relative to the control (no
t-BuOH) run on the same day.
HP LC An a lyses. Samples were analyzed directly by HPLC
employing two Gilson (Middleton, WI) Model 303 HPLC pumps
equipped with Gilson systems controller software and a Gilson
Model 116 UV detector (210 and 222 nm monitored). Separa-
tions were achieved using 40% acetonitrile/phosphoric acid
buffer (pH 2) at a flow rate of 1.0 mL/min on an Ultrasphere
C-18 (ODS), 5 µm, end-capped, 4.6 mm × 25 cm steel-jacketed
column (Beckman Instruments, Inc., Fullerton, CA).
RESULTS AND DISCUSSION
Ozon olysis of Br om a cil. Three products, II-IV,
and several very minor products were detected by HPLC
when solutions of bromacil were treated with ozone
(Figure 2). The structures of II (3-sec-butyl-5-acetyl-5-
hydroxyhydantoin), III (3-sec-butylparabanic acid), and
IV (3-sec-butyl-5,5-dibromo-6-methyl-6-hydroxyuracil)
were determined previously (Acher et al., 1994). The
principal fragments from product mass spectra obtained
by LC/MS (EI) of the reaction solution (Figure 3) were
as follows. II mass (relative abundance): 185 (3), 172
Qu a n tita tion of Br om a cil a n d Its P r od u cts. Standard
curves for the purposes of quantitation were obtained by
treating with ozone a 400 mL solution of 400 ppm of bromacil
1
4
(
1.53 mM) spiked with 30 µCi of [2- C]bromacil. Samples
were removed periodically during the reaction and assayed
directly by HPLC; effluent fractions were collected at 15 s
intervals. Fractions were analyzed on a Beckman Model
LS6000IC liquid scintillation counter (Columbia, MD) using
ScintiVerse II cocktail (Fisher Chemical, Fisher Scientific, Fair
Lawn, NJ ). Chromatograms were reconstructed at each
reaction time interval from the amount of radioactivity
detected in each fraction. In addition to bromacil, only the
(
24), 171 (43), 142 (9), 116 (44), 115 (100), 70 (28). III:
155 (2), 141 (75), 115 (29), 84 (11), 70 (100), 56 (39). IV:
327, 329, and 331 (15); 301, 303, and 305 (28); 260 and
262 (9); 231 and 233 (13); 205 and 207 (100); 142 (55);
1
27 (18); 70 (41). The dibromohydrin, IV, was relatively
unstable and after a day at room temperature would
lose HOBr and re-form bromacil. Isolated yields of II,
III, and IV obtained from flash column chromatography
followed by semipreparative HPLC separation were ca.
1
4
three major products were observed in the C chromatogram;
several very minor products that were at the level of detection
were not quantified (Figure 1). The area of each peak (in
dpm‚min) corresponded to a known mass as determined from
the total counts present in each run. The calculated mass was
plotted versus the area of the peak obtained using the UV
detector for each run; these curves were used in subsequent
5
%, 20%, and 5% of II, III, and IV, respectively; the
minor products were not isolated (Acher et al., 1994).
The actual product yields were determined using [2- C]-
1
4

1008 J. Agric. Food Chem., Vol. 45, No. 3, 1997
Hapeman et al.
F igu r e 3. LC/MS total ion chromatogram of bromacil ozo-
nolysis reaction mixture. Peaks at t ) 7.9, 18.2, 20.6, and
4.8 min correspond to II, III, bromacil, and IV, respectively.
R
2
F igu r e 5. Proposed parallel mechanism for bromacil ozonoly-
sis.
F igu r e 4. Proposed in-series mechanism for bromacil ozo-
nolysis.
Ta ble 1. Com p a r ison of Oxid a tion Ra tes a n d F in a l
P r od u ct Yield s
yield (%)
reactn conditions
k′ (min-1)
II
III
IV
O3
0.29 ( 0.01
0.23 ( 0.02
0.97 ( 0.01
24
18
21
56
72
55
20
6
23
O3 + 6.9 mM H2O2
O3 + 52.9 mM t-BuOH
bromacil; treatment of aqueous bromacil with ozone
afforded 24, 56, and 20% of II, III, and IV, respectively
F igu r e 6. Summary of degradation pathways: in-series
versus parallel mechanisms.
(
Table 1; Figure 1).
Degr a d a t ion P a t h w a y: In -Ser ies a n d P a r a llel
would correspond to a disappearance in II. Further-
more, the series mechanism cannot be operational to
any extent because II does not decrease nor does III
increase after all the bromacil is depleted. Therefore,
the ozonolysis of bromacil proceeds via a parallel mech-
anism exclusively.
Mech a n ism s. The simplest degradation pathway for
bromacil ozonolysis, assuming direct attack by ozone
rather than a hydroxy radical mechanism, might involve
the formation of an ozonide, V, which upon collapse
would give an unstable epoxide, VI (Figure 4). Loss of
a proton and bromide followed by rearrangement would
afford II. Further oxidation of II to a peroxide would
subsequently yield III. Hydrolysis of the epoxide, VI,
in the presence of hydrobromic acid would give rise to
IV. This in-series mechanism contrasts with a parallel
mechanism, where reaction of bromacil with ozone
forms the ozonide, V, which can collapse to form II or
III directly (Figure 5). The formation of IV in the
parallel mechanism would arise from hypobromous acid,
HOBr, attack on bromacil itself. These mechanisms are
summarized in Figure 6.
Rea ction Kin etics. The kinetics of bromacil ozo-
nolysis can be expressed as a function of the concentra-
tions of bromacil, the oxidants, and HOBr (eq 1), where,
k2, k3, and k4 are the rate constants for the formation
of II, III, and IV, respectively; [I] is the concentration
of bromacil, [oxidant] is the concentration of ozone or
•
OH, and [HOBr] is the concentration of hypobromous
acid. Ozone was continuously applied at a constant
rate, and concentrations in the reactor increased rapidly
to ca. <11 µM. Furthermore, the concentration of ozone
is small relative to the concentration of bromacil as the
initial concentrations of bromacil were approximately
2 orders of magnitude larger than those of ozone. The
concentration of HOBr can also be assumed to be rather
low because its in-situ production is slow (Haag and
The degradation profile of bromacil ozonolysis (Figure
) clearly shows that the formation of III is independent
7
of II. If III were a product of II, a delay would be
expected in the formation of III and its appearance

Mechanism of Bromacil Ozonolysis
J. Agric. Food Chem., Vol. 45, No. 3, 1997 1009
F igu r e 7. Bromacil ozonolysis product profile.
F igu r e 8. Bromacil ozonolysis in the presence of 0, 3.4, or
-
1
2
6
.9 mM hydrogen peroxide: k′(no H
2
1
O
2
) ) 0.29 min (r
2
)
-
0
H
.97), k′(3.4 mM H
2
O
2
) ) 0.27 min (r ) 0.97), k′(6.9 mM
Hoign e´ , 1983) and it is consumed fairly rapidly. The
kinetic equation can then be reduced to the pseudo-first-
order equation (eq 2) where k′ is defined by eq 3.
-
1
2
2 2
O ) ) 0.23 min (r ) 0.99).
d[I]/dt ) -k [I][oxidant] - k [I][oxidant] -
2
3
k4[I][HOBr] (1)
d[I]/dt ) - k′[I]
(2)
k′ ) k [oxidant] + k [oxidant] + k [HOBr] (3)
2
3
4
•
Dir ect Ozon olysis ver su s a OH Mech a n ism . The
oxidation of bromacil in the presence of ozone can
proceed via direct attack of ozone on the carbon-carbon
double bond or via a hydroxy radical mechanism, which
is very typical of aqueous ozonation (Masten and Davies,
1
994). The addition of hydrogen peroxide is expected
to increase hydroxy radical processes in ozonation
Peyton and Glaze, 1987). In this study, however,
(
addition of hydrogen peroxide did not effect a profound
change in the rate of bromacil degradation, although
the reaction was somewhat slower: k′(no H2O2) ) 0.29
-
-
1
1
-1
min
( 0.01 min
versus k′(6.9 mM H2O2) ) 0.23
F igu r e 9. Product profile of bromacil ozonolysis in the
presence of 6.9 mM hydrogen peroxide.
-1
min ( 0.02 min (Figure 8). A very slight decrease
was observed in k′ when 3.4 mM H2O2 was added, and
no change was observed when 0.069 mM H2O2 was used
increased the rate of degradation by nearly an order of
(
data not shown). This result was somewhat unex-
-1
-1
-1
magnitude: k′(no t-BuOH) ) 0.29 min ( 0.01 min
pected because, in previous studies, hydrogen peroxide
increased the degradation rate of pesticides during
ozonation (Somich et al., 1988; Kearney et al., 1988;
Somich et al., 1990). However, the product ratio showed
an increase in the formation of III (72%) and a sub-
stantial decrease in the formation of IV to 6% (Table 1;
Figure 9).
-1
versus k′(52.9 mM t-BuOH) ) 0.97 min ( 0.01 min
(
Figure 10). Interestingly, the yield of IV actually
increased slightly to 23% (Table 1; Figure 11).
These experiments strongly suggest that the mecha-
nism of bromacil ozonolysis proceeds via direct ozone
attack on the double bond and not via a hydroxy radical
mechanism. Furthermore, none of the major products
of bromacil degradation involve oxidation or removal of
the sec-butyl moiety on the imide nitrogen, as might be
expected in a hydroxy radical system. If this process
occurs, it is a very minor pathway at best. Apparently,
the electron-withdrawing nature of the imide function-
•
In contrast, the addition of t-BuOH, a known OH
scavenger, will cause a decrease in the rate of degrada-
tion if the mechanism involves hydroxy radical oxida-
tion, because t-BuOH will compete with the substrate
•
for OH. Furthermore, hydroxy radicals can attack
ozone and remove a portion of the ozone from the system
Peyton and Glaze, 1987). Alternatively, if the mech-
•
ality is sufficient to deter attack by OH on the sec-butyl.
(
anism proceeds via direct ozone attack, addition of
t-BuOH could effect an increase in the degradation rate
by allowing more ozone to react directly with the
substrate rather than OH. In this study, treatment of
bromacil with ozone in the presence of 52.9 mM t-BuOH
Mech a n ism for Br om a cil Degr a d a tion . Addi-
tional proof that the ozonolysis of bromacil does not
involve hydroxy radicals was obtained by irradiating an
aqueous solution of bromacil containing nitrate. (Ni-
trate has been shown to generate hydroxy radicals upon
•

1010 J. Agric. Food Chem., Vol. 45, No. 3, 1997
Hapeman et al.
of bromacil yields IV (March, 1977; Haag and Hoign e´ ,
1
983; Cavanaugh et al., 1992; Glaze et al., 1993).
CONCLUSIONS
The mechanism for the ozonolysis of bromacil was
shown to proceed via direct ozone attack as opposed to
a hydroxy radical process. Furthermore, the reaction
proceeds via parallel pathways as opposed to an in-
series pathway. The dibromohydrin, which decomposes
to the phytotoxic starting material, bromacil, is believed
to be the reason for the residual phytotoxic effect of
ozonated bromacil solutions (Acher et al., 1994). The
results presented here indicate that under conditions
•
in which OH is scavenged, the rate of bromacil ozo-
nolysis is significantly enhanced; however, the formation
of the phytotoxic precursor, IV, remains constant or
slightly increases. The addition of hydrogen peroxide,
although a slower reaction, apparently diminishes the
formation of IV substantially. The results of this work
suggest that, in situations where photolytic remediation
of bromacil is precluded and elimination of phytotoxicity
is desirable, conditions which favor a hydroxy radical
process (higher pH and H2O2 addition) would be pre-
ferred. This is generally the more common method of
ozonation for most waste streams as decomposition of
many organic species proceeds via a hydroxy radical
process (Masten and Davies, 1994).
F igu r e 10. Bromacil ozonolysis in the presence of 0, 13.3 or
-
1
2
5
2.9 mM t-BuOH: k′(no t-BuOH) ) 0.29 min (r ) 0.97), k′-
-1 2
(
13.3 mM t-BuOH) ) 0.62 min
(r ) 0.90), k′(52.9 mM
-
1
2
t-BuOH) ) 0.97 min (r ) 0.92).
ACKNOWLEDGMENT
We thank J ames Oliver (USDA/ARS, Beltsville, MD)
for discussions concerning the proposed mechanisms of
bromacil ozonolysis.
LITERATURE CITED
Acher, A. J .; Dunkelblum, E. Identification of sensitized
photooxidation products of bromacil in water. J . Agric. Food
Chem. 1979, 27, 1164-1167.
Acher, A. J .; Saltzman, S. Dye-sensitized photooxidation of
bromacil in water. J . Environ. Qual. 1980, 9, 190-194.
Acher, A. J .; Hapeman, C. J .; Shelton, D. R.; Muldoon, M. T.;
Lusby, W. R.; Avni, A.; Waters, R. Comparison of formation
of biodegradation of bromacil oxidation products in aqueous
solutions. J . Agric. Food Chem. 1994, 42, 2040-2047.
Anderson, B. G. M.S. Thesis, University of Maryland, 1996.
Bader, H.; Hoign e´ , J . Determination of ozone in water by the
indigo method. Water Res. 1981, 15, 449-456.
Cavanaugh, J . E.; Weinberg, H. S.; Gold, A.; Sangalah, R.;
Marbury, D.; Glaze, W. H.; Collette, T. W.; Richardson, S.
D.; Thurston, A. D., J r. Ozonation byproducts: identification
of bromohydrins from the ozonation of natural waters with
enhanced bromide levels. Environ. Sci. Technol. 1992, 26,
1658-1662.
F igu r e 11. Product profile of bromacil ozonolysis in the
presence of 52.9 mM t-BuOH.
irradiation (Zepp et al., 1987).) The products generated
during this process were examined by HPLC equipped
with a photodiode array detector. None of the products
observed had retention times or UV spectra similar to
those of the products observed during ozonolysis (data
not shown) (Anderson, 1996). A small amount of II was
observed in an aqueous bromacil solution that contained
no nitrate, which was not entirely unexpected (Acher
and Dunkelblum, 1979; Acher et al., 1994).
Glaze, W. H.; Weinberg, H. S.; Cavanaugh, J . E. Evaluating
the formation of brominated DBPs during ozonation. J . Am.
Water Works Assoc. 1993, 85, 96-103.
These data, then, are consistent with the following
mechanism (Figure 5). Ozone attack on the bromacil
carbon-carbon double bond gives rise to the ozonide,
V, which can collapse to form the epoxide, VI. Opening
Haag, W. R.; Hoign e´ , J . Ozonation of bromide-containing
waters: kinetics of formation of hypobromous acid and
bromate. Environ. Sci. Technol. 1983, 17, 261-267.
Hapeman, C. J .; Karns, J . S.; Shelton, D. R. Total mineraliza-
tion of aqueous atrazine in the presence of ammonium
nitrate using ozone and Klebsiella terragena (strain DRS-
I): mechanistic considerations for pilot scale disposal. J .
Agric. Food Chem. 1995, 43, 1383-1391.
Hapeman-Somich, C. J . Mineralization of pesticide degrada-
tion products. In Pesticide Transformation Products. Fate
and Significance in the Environment; Somasundaram, L.,
Coats, J . R., Eds.; ACS Symposium Series 459; American
Chemical Society: Washington, DC, 1991; pp 133-147.
-
of the epoxide and loss of Br followed by intramolecular
rearrangement affords II. Alternatively, the ozonide
can open directly, cleaving the C5-C6 bond. Bromide
release, ring closure, and subsequent loss of the acyl-
-
type moiety gives rise to III. Finally, Br can react
-
1
-1
directly with O3 (k ) 160 M
s
) (Haag and Hoign e´ ,
-
1
983) to give OBr. The electrophilic addition of HOBr
HOBr h OBr, pKa ) 8.8) to the C5-C6 double bond
-
(

Mechanism of Bromacil Ozonolysis
J. Agric. Food Chem., Vol. 45, No. 3, 1997 1011
Hoign e´ , J .; Bader, H. The role of hydroxy radical reactions in
ozonation processes in aqueous solutions. Water Res. 1976,
Somich, C. J .; Kearney, P. C.; Muldoon, M. T.; Elsasser, S.
Enhanced soil degradation of alachlor by treatment with
ultraviolet light and ozone. J . Agric. Food Chem. 1988, 36,
1322-1326.
Somich, C. J .; Muldoon, M. T.; Kearney, P. C. On-Site
treatment of pesticide waste and rinsate using ozone and
biologically active soil. Environ. Sci. Technol. 1990, 24, 745-
749.
1
0, 377-386.
J ordan, L. S.; Mann, J . D.; Day, B. E. Effects of ultraviolet
light on herbicides. Weeds 1965, 13, 43-46.
Kearney, P. C.; Woolson, E. A.; Plimmer, J . R.; Isensee, A. R.
Decontamination of pesticides in soils. Residue Rev. 1969,
2
9, 137-144.
Kearney, P. C.; Muldoon, M. T.; Somich, C. J .; Ruth, J . M.;
Voaden, D. R. Biodegradation of ozonated atrazine as a
wastewater disposal system. J . Agric. Food Chem. 1988, 36,
Worthing, C. R., Hance, R. J ., Eds. The Pesticide Manual, A
World Compendium, 9th ed.; British Crop Protection Coun-
cil: Old Woking, Surrey, Great Britain, 1991; pp 1240-
1
301-1306.
1
241.
March, J . Advanced Organic Chemistry. Reactions, Mecha-
nisms, and Structure, 2nd ed.; McGraw-Hill: New York,
Zepp, R. G.; Hoign e´ , J .; Bader, H. Nitrate-induced photooxi-
dation of trace organic chemicals in water. Environ. Sci.
Technol. 1987, 21, 443-450.
1
977; pp 741-742.
Massey, J . H. Ph.D. Thesis, University of Arkansas, 1995.
Masten, S. J .; Davies, S. H. R. The use of ozonation to degrade
organic contaminants in wastewaters. Environ. Sci. Technol.
Received for review J anuary 23, 1996. Revised manuscript
1
994, 28, 180A-185A.
X
received September 30, 1996. Accepted December 13, 1996.
Moilanen, F. W.; Crosby, D. G. The photodecomposition of
bromacil. Arch. Environ. Contam. Toxicol. 1974, 2, 3-8.
Peyton, G. R.; Glaze, W. H. Mechanism of photolytic ozonation.
In Photochemistry of Environmental Aquatic Systems; Zika,
R. G., Cooper, W. J ., Eds.; ACS Symposium Series 327;
American Chemical Society: Washington, DC, 1987; pp 76-
Mention of specific products or suppliers is for identification
and does not imply endorsement by U.S. Department of
Agriculture to the exclusion of other suitable products or
suppliers.
J F9600420
8
8.
Saltzman, S.; Acher, A. J .; Brates, N.; Horowitz, M.; Gevelberg,
A. Removal of phytotoxicity of uracil herbicides in water by
photodecomposition. Pestic. Sci. 1982, 13, 211-217.
X
Abstract published in Advance ACS Abstracts, Feb-
ruary 1, 1997.