Oxidation of aminodinitrotoluenes with ozone: Products and pathways

Article abstract of DOI:10.1021/es990190h

An investigation of the products from the reaction of ozone with aminodinitrotoluenes (ADNTs) provides information about the oxidation pathway. Studies conducted at low conversions of 2- and 4-ADNT show 2:1 ozone/ADNT stoichiometries, prompt formation of glyoxylic and pyruvic acids, and NO₂^- and NO₃^- (NO_x) ions. Reaction schemes to account for these results involve a 1,3-dipolar cycloaddition of ozone to selected double bonds of the aromatic ring, leading to ring cleavage. ¹⁵N-Labeling experiments indicate that the amino function is not involved in the initial ozone oxidation and eventually is incorporated into pyruvamide (2-ADNT) and oxamic acid (4-ADNT) before being oxidized to nitrate.

Full text of DOI:10.1021/es990190h

Environ. Sci. Technol. 2000, 34, 497-504
Oxidation of Aminodinitrotoluenes
with Ozone: Products and
Pathways
R O N A L D J . S P A N G G O R D , *
C . D A V I D YA O , A N D T H E O D O R E M I L L
SRI International, 333 Ravenswood Avenue,
Menlo Park, California 94025
accelerated with electron-donating groups and retarded with
electron-withdrawing groups. Reaction pathways and prod-
An investigation of the products from the reaction of
ozone with aminodinitrotoluenes (ADNTs) provides information
about the oxidation pathway. Studies conducted at low
conversions of 2- and 4-ADNT show 2:1 ozone/ADNT
stoichiometries, prompt formation of glyoxylic and pyruvic
acids, and NO2- and NO3 (NOx) ions. Reaction schemes
to account for these results involve a 1,3-dipolar cycloaddition
of ozone to selected double bonds of the aromatic ring,
leading to ring cleavage. N-Labeling experiments indicate
that the amino function is not involved in the initial
ozone oxidation and eventually is incorporated into
pyruvamide (2-ADNT) and oxamic acid (4-ADNT) before
being oxidized to nitrate.
-
15
ucts can become complex, depending on the nature and
number of the substituents (3). Both electron-donating
(
amino, methyl) and electron-withdrawing (nitro) groups in
ADNTs create reactive centers for ozonide decomposition in
water and lead to a multiplicity of products. The ADNTs
serve as interesting examples of ozone oxidations in aromatic
systems with multiple reaction sites and the products provide
pathway information in electron-deficient systems. With the
exception of a few simple aromatics and polycyclic aromatics,
detailed characterization of polysubstituted aromatic ozone
reactions has not been done.
Introduction
Aminodinitrotoluenes [2-amino-4,6-dinitrotoluene (2-ADNT)
and 4-amino-2,6-dinitrotoluene (4-ADNT)] are environmen-
tal pollutants that arise from the microbial biotransformation
of 2,4,6-trinitrotoluene (TNT). These chemicals are unique
to soils, lagoons, and groundwaters near TNT production
and handling facilities. In the remediation of such environ-
ments, consideration must be given to the behavior of these
pollutants toward the remediation technology. In the previous
paper (1), we investigated the kinetics of the ozone-hydrogen
peroxide (peroxone) oxidation of 2- and 4-ADNT and found
that, under most peroxone usage conditions, the oxidation
of ADNTs is controlled by ozone. In using an oxidation
remediation process, it becomes important to know the final
end products formed under the treatment conditions. In this
investigation, we determined the products resulting from
ozone oxidation of 2- and 4-ADNT and evaluated products
Experimental Section
Chem icals. 2- and 4-ADNT were prepared in our laboratory
at 99% purity using the method of Zbarskii (4). 15N-Labeled
aminodinitrotoluenes were prepared by the method of
Spanggord and Clizbe (5). The remaining chemicals were
obtained from Aldrich Chemical Co. (Milwaukee, WI) or
synthesized where indicated. An authentic standard of oxamic
acid aldehyde was prepared using the method of Tits and
Bruylant (6); pyruvamide was prepared using the method of
Claison and Shadwell (7). Both standards yielded chromato-
graphic retention volumes and UV spectral data identical to
those for ozonation products from ADNTs.
Ozone Oxidations. Ozone oxidations were conducted in
aqueous solutions adjusted to pH 5.0 with phosphate buffer.
Ozone was generated from a Welsbach ozone generator and
bubbled into water. Ozone concentrations were measured
using the indigo colorimetric method (8).
An ozone stock solution was added in 20 µM increments
to a rapidly stirred stock solution of ADNT (150 µM). After
each addition, the ADNT was analyzed by reverse-phase high-
performance liquid chromatography (HPLC), using a gradient
program developed for aldehydes and ketones as their 2,4-
dinitrophenylhydrazones (DNPHs, 9) and using an Alltech
Altima C18 column (5 µm, 4.6 × 250 mm) and UV detection
at 350 nm. Under these conditions, the ADNTs eluted at 11.4
min. Carboxylic acids were analyzed by ion chromatography
using a Supelcogel 610H column and UV detection at 210
1
5
from oxidation of N-amino-labeled ADNTs, from which
multiple potential oxidation pathways can be advanced.
Many of the mechanistic approaches to the oxidation of
organic compounds by ozone can be described by the Criegee
mechanism (2, 3). The initial steps for ozonation of an
aromatic compound appear to involve sequential formation
of a π-complex (I), an addition product (II), a zwitterion (III),
and an isoozonide (IV) as shown here for benzene.
In aqueous systems, water can play a major role in the
decomposition of intermediates I-IV to form aldehydes or
ketones and hydroxyhydroperoxides (V). If the aromatic ring
has substituents, reaction with ozone will be significantly
*
Corresponding author phone: (650)859-3822; fax: (650)859-4321;
e-mail: ronald.spanggord@sri.com.
1
0.1021/es990190h CCC: $19.00

2000 Am erican Chem ical Society
VOL. 34, NO. 3, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4 9 7
Published on Web 12/29/1999

FIGURE 1. HPLC profile of carbonyl-containing products resulting from the ozonation of 2-ADNT. B, glyoxylic acid; C, 2,4-dinitrophenylhydrazine;
H, pyruvamide; D, pyruvic acid; I, 2-ADNT, J, glyoxal; F, pyruvaldehyde.
nm or a Dionex PAX-500 column with conductivity detection.
Structural identifications were made by mass spectrometry
and/ or chromatographic retention times coupled with UV
spectral data obtained from a Waters photodiode array
detector (PDA) as compared to authentic standards.
monitoring the formation of carbonyl-containing compounds
(as their 2,4-dinitrophenylhydrazones), carboxylic acids, and
nitrite and nitrate using incremental additions of ozone to
oxidize up to 95% of the ADNT. Typical HPLC profiles of
carbonyl-containing compounds formed at 95% conversion
of 2- and 4-ADNT are shown in Figures 1 and 2. Products
identified from 2-ADNT include glyoxylic acid, pyruvic acid,
pyruvamide, glyoxal, and pyruvaldehyde. From 4-ADNT, the
products include pyruvic acid, glyoxylic acid, oxamic acid,
oxamic acid aldehyde, pyruvaldehyde, and a compound
tentatively identified as a tricarbonyl-containing amide. The
formation and decline of these products as a function of
ozonation time are shown in Figure 3 for 2- and 4-ADNT. It
appears that, for both compounds, pyruvic acid is generated
as a stable end product.
x
NO Ions. Nitrite and nitrate were analyzed by ion-
exchange liquid chromatography with UV detection at 210
nm according to the method of Thayer and Huffaker (10).
1
5
N-Labeled nitrite and nitrate were analyzed by mass
spectrometry using the method of Glover and Hoffsommer
11) after conversion of each anion to nitrobenzene.
(
Results
Products at High (95%) ADNT Conversions. The oxidations
of 2- and 4-ADNT by ozone were followed by HPLC by
4
9 8
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 3, 2000

FIGURE 2. HPLC profile of carbonyl-containing products resulting from the ozonation of 4-ADNT. A, oxamic aldehyde; B, gloxylic acid;
C, 2,4-dinitrophenylhydrazine; D, pyruvic acid; G, 4-ADNT; E, product 1; F, pyruvic aldehyde.
HPLC profiles, showing a similar set of short-chain
carboxylic acids for each ADNT, are found in Figure 4. These
acids include acetate, glycolate, formate, glyoxylate, pyruvate,
malonate, and oxalate. Most of these acids, with the exception
of glyoxylic acid, are terminal end products from the
ozonation of ADNT intermediates.
account for only 50% of the available nitro groups at low
conversions of 2-ADNT and 25-35% of oxidized 4-ADNT.
They account for >90% of the nitro groups at high conversions
(Tables 1 and 2).
Ozonation of 15N-labeled amino-2- and 4-ADNT produced
NO
x
ions that were converted to nitrobenzene and analyzed
Nitrogen Balances. Nitrogen balances in oxidized ADNTs
for 15N content by mass spectrometry by comparing the ratio
1
4
15
are largely composed of NO
69-99%) conversions of ADNTs, with the exception of
pyruvamide, which appears in significant amounts in the
high conversions of 2-ADNT (Table 1). NO ions together
x
ions at low (6-45%) and high
of m/ z 123/ 124 ( N-labeled nitrobenzene/ N-labeled ni-
trobenzene). No 15N-labeled nitrobenzene was detected,
indicating that amine nitrogen is not involved in the initial
ozonation of ADNTs. 15N-Labeled oxamic acid was identified
(
x
VOL. 34, NO. 3, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4 9 9

FIGURE 3. Quantitative determination of 2,4-dinitrophenylhydrazone derivatives of products from (A) 2-ADNT and (B) 4-ADNT.
a
TABLE 1. Nitrogen Balance on Products from the Ozonation of
2-ADNT (µmol)
%
N
pyruv-
total
% N
based
ADNT ∆ADNT nitrate nitrite amide product N balance on NOx
1
60
0
109
140
150
158
0
200
390
410
425
0
20
30
22
18
0
0
240
447
458
465
5
2
1
0.8
0.4
0.3
18.9
27.2
25.9
21.8
73
106
102
98
67
100
96
Carbon Balances. Carbon balances were determined by
quantifying known products as a function of the amount of
ozone added to the ADNT solutions. Data for 2-ADNT, in
Table 3, show that recovery of carbon in products is nearly
constant at about 48% over the conversion range of 67-99%.
Data for 4-ADNT, in Table 4, show a lower recovery of carbon
between 32 and 40%. Over the course of the oxidations, total
organic carbon (TOC) was found not to change significantly,
2
.0
93
a
Sam ple calculation: At the first point, 109 µm ol of ADNT consum ed
corresponding to 327 µm ol of nitrogen produced (3 × 109). Total
m icrom oles of nitrogen from products ) 240 µm ol. Percent nitrogen
balance ) 240/327 × 100 ) 73%. Percent N balance from NO
200 + 20)/327 ) 67%.
x
)
(
in the reaction mixture from ozonation of 4-ADNT by mass
spectrometry after converting it to the methyl ester. Its mass
spectrum showed a molecular ion one mass unit higher than
reported for methyl oxamate by Woodburn et al. (12). After
extended ozonation, >95% of the nitrogen in both ADNTs
2
indicating that little carbon was oxidized to CO . We believe
several four- and five-carbon carbonyl products from ring
cleavage account for the missing carbon; these products
appear in HPLC profiles of product dinitrophenylhydrazone
(DNPH) derivatives but could not be identified or quantified
due to a lack of available standards.
x
appear as NO ions, showing that pyruvamide and oxamide
intermediates are eventually oxidized further.
Equations 1 and 2 illustrate the pathways for conversion
of ADNT nitrogens to products at low conversions.
Products at Low ADNT Conversions. Studies were
conducted with ozone as the limiting reagent to determine
5
0 0
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 3, 2000

TABLE 2. Nitrogen Balance on Products from the Ozonation of 4-ADNT (µmol)
oxamic acid
aldehyde
oxamic
acid
total
product N
% N
balance
% N based
on NOx
ADNT
67
∆ADNT
nitrate
nitrite
1
0
103
125
148
157
0
121
182
320
415
0
42
20
15
14
0
0
12
13
13
15
0
178
219
352
449
6
4
1
1
4
2
9
0
3.5
3.7
3.8
4.8
58
58
79
95
53
54
75
91
a
TABLE 3. Carbon Balance on Ozonation Products of 2-ADNT (µmol)
glyoxylic pyruvic pyruv-
pyruv-
C balance,
%
2-ADNT ∆ADNT glycolate acetate formate oxalate
acid
acid
amide glyoxal aldehyde product C
1
60
0
109
140
150
158
0
0
0
0
0
0
70.0
87.9
109
125
0
0
1.0
3.0
8.8
0
127
147.6
126
96.2
0
0
0
0
7.2
13.2
13.8
11.4
5
2
1
0.8
0.4
0.3
54.2
70.4
86
37.5
53.4
62.4
80.1
56.7
81.6
77.7
65.4
16.6
23.4
21.8
17.4
369
478
500
505
48
49
48
46
2
.0
8.2
100.8
a
Sam ple calculation: At the first point, 109 µm ol of ADNT consum ed corresponding to 763 µm ol of carbon (7 × 109). The m icrom oles of carbon
from products found ) 369. Carbon balance ) 396/763 × 100 ) 48%.
TABLE 4. Carbon Balance on Ozonation Products of 4-ADNT (µmol)
glyoxylic pyruvic oxamic oxamic unknown
pyruv-
product C
% C
a
4-ADNT ∆ADNT glycolate acetate formate oxalate
acid
acid
aldhyde
acid
1
aldehyde balance balance
1
67
0
103
125
148
157
0
2.4
5.8
20
39.2
0
0
0
0
0
0
0
0
7.2
28.6
20.8
0
0
60.6
90
121.8
143
0
0
24
26
26
30
0
0
6
4
1
1
3.9
104.3
132.8
185
26.2
31.4
30.6
18.6
7.0
7.4
7.6
9.6
13.6
18.4
14.4
8.8
4.5
7.5
6.9
3.9
243
327
441
469
34
37
43
43
1.6
9.3
0.3
194.7
a
Assum es four carbon product.
initial products formed in the oxidation and their initial mole
ratios. Table 5 shows the result of low conversion oxidations
of 2-ADNT, starting with 100 mM 2-ADNT, and evaluating
products after 6%, 19%, 28%, and 45% 2-ADNT oxidation by
ozone. The ozone/ 2-ADNT stoichiometry is close to 2.
Glyoxylic and pyruvic acids were produced in a 2:1 mole
ratio with no evidence for production of pyruvamide. This
result suggests that ring cleavages in the vicinity of the amine
group are not involved in the initial oxidative steps. The
finding that both glyoxylic and pyruvic acids are produced
at the same time suggests that multiple sites are involved in
the initial cleavages and that some selectivity exists for
particular sites. Nitrate and nitrite are also produced in a 2:1
ratio, which is controlled at least in part by the rapid oxidation
x
of nitrite to nitrate by ozone; however, the NO / ADNT ratio
is less than 2. Carbon and nitrogen balances in these
experiments were low: only 13% of the carbon was accounted
for while 50% of nitrogen (based on nitro groups) was found
-
as NO
x
ions
The results for 100 µM 4-ADNT appear in Table 6 and
represent product distributions after 10%, 18%, 24%, and
3
9% 4-ADNT oxidation by ozone. In this case, pyruvic and
glyoxylic acids were the first small acids to result from 4-ADNT
cleavages, and they were formed in a 2:1 ratio. A small amount
of oxamic acid aldehyde was also observed, again indicating
that multiple cleavage sites occur during the initial oxidation.
However, the relative mole ratios indicate some selectivity
for initial ring rupture. Ozone/ 4-ADNT ratios began to
approach 2. Interestingly, nitrite/ nitrate ratios were 2:1 while
the opposite was observed with 2-ADNT. Also, the NO
ADNT ratio approaches 0.6 instead of 2 as observed with
-ADNT. Nitrogen balances are 31-34%, based on nitro
groups in oxidized ADNT; however, carbon balances are only
% based on oxidized 4-ADNT.
x
/ 4-
2
FIGURE 4. HPLC profile of carboxylic acids generated from the
ozonation of 2-ADNT (A) and 4-ADNT (B).
3
VOL. 34, NO. 3, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5 0 1

TABLE 5. Products from Ozonation of 2-ADNT at Low Conversions (µmol)
glyoxylic
acid
pyruvic
acid
% N in
products
% C in
products
2-ADNT
∆ADNT
∆O3
nitrate
nitrite
O3/ADNT
NOx/ADNT
1
00
0
6
19
28
45
0
17.2
34
51
85
0
6.4
10.5
28.1
29.5
0
3.5
5.6
15.3
16.2
0
0
0
0
1.6
0.85
1.55
1.02
9
8
7
5
4
1
2
5
2.5
5.2
7.3
10.3
1.1
2.5
3.6
5.4
2.8
1.8
1.8
1.9
80
48
43
51
20
13
13
12
TABLE 6. Products from Ozonation of 4-ADNT at Low Conversions (µmol)
glyoxylic
acid
oxamic acid
aldehyde
pyruvic
acid
% N in
products
% C in
products
4-ADNT
∆ADNT
∆O3
nitrate
nitrite
O3/ADNT
NOx/ADNT
1
00
0
10
18
24
39
0
14
28
44
75
0
0
4.1
5.6
11.6
12.2
0
0
0
0
0
9
8
7
6
0
2
6
1
2.6
3.4
4.4
11.7
0.2
0.4
0.6
0.9
0.10
0.14
0.23
0.37
0.4
0.9
1.0
1.6
1.4
1.6
1.8
1.9
0.67
0.50
0.67
0.61
34
34
33
31
2.6
3.0
2.8
2.7
SCHEME 1. Pathways for Ozonation of 2-ADNT
Ozonation very likely involves 1,3-dipolar cycloaddition
Discussion
of ozone to selected double bonds of the aromatic ring to
form one or more isomeric ozonides via the Criegee mech-
anism (13, 14). Early formation of pyruvic and glyoxylic acids,
but not pyruvamide, from ozonation of 2-ADNT points to
selective attack at C5 to form C6-C5 and C5-C4 ozonides
Mechanistic Pathway. Scheme 1 illustrates a plausible
pathway for ozonation of 2-ADNT, accounting for the results
of product studies at low conversions: the ∼2:1 ozone:ADNT
stoichiometry, prompt formation of selected short-chain
carbonyl compounds, and NO
x
ions. A similar scheme (not
(
Scheme 1). Since the estimated lifetimes of aliphatic ozonides
shown) describes the pathways for ozonation of 4-ADNT.
are less than a millisecond (14), the initial ADNT ozonides
probably cleave to ring-opened diene dicarbonyls and
nitrohydroperoxides before reacting with additional ozone
1
5
The N-labeling experiments show that amino nitrogen is
not involved in ozone-ADNT interactions.
5
0 2
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 3, 2000

FIGURE 5. Ultraviolet spectra of the 2,4-dinitrophenylhydrazones of (A) pyruvic acid, (B) pyruvaldehyde, and (C) multicarbonyl intermediate
from the ozonation of 4-ADNT.
to give secondary ozonides. Cleavage of a second ozonide or
diozonide would give detectable two- or three-carbon
aldehydes and acids plus four- or five-carbon chain frag-
ments, which hydrolyze by multiple routes to unidentified
carbonyl intermediates. Continued ozonation results in
further oxidative degradation of intermediate carbonyls and
amides. Eventually, almost all of the organic nitrogen
converges to the oxidatively stable nitrate ion.
Material Balance. Element balances on products account
for 30-50% of the reacted 2-ADNT nitrogen as NO ions at
x
1
0-40% conversion (Table 5). However, only 3-20% of
carbon in ozonized 2- or 4-ADNT is identified in glyoxylic
15
and pyruvic acids. The N-labeled experiments with the 2-
x
and 4-ADNT show that essentially all NO ions initially arise
from the nitro groups, not the amino group, probably by
hydrolysis of the intermediate nitrohydroperoxides (5 and 6
in Scheme 1). The significant shortfalls in carbon at low
conversions means that intermediates 5 and 6 hydrolyze to
form aminonitrodienoic aldehydes or acids, which remain
undetected or unidentified by HPLC analysis (see above).
HPLC traces of product mixtures are devoid of the longer
retained peaks, with UV absorption characterizing oxidized
aromatic rings such as hydroxyanilines or quinomethides
that might form by electrophilic or radical ozonation (13,
A similar evaluation of the ozonation of 4-ADNT also leads
to conclusion that ozone attacks the aromatic ring of 4-ADNT
selectively and at sites remote from the amine carbon.
Products of ring cleavage of 4-ADNT show spectral properties
of polycarbonyl DNPHs. For example, Figure 5 shows the
spectral profiles of the DNPHs of pyruvic acid, pyruvaldehyde,
and an unknown product from the ozonation of 4-ADNT.
The unknown DNPH shows the characteristic spectrum of
a monocarbonyl and a conjugated dicarbonyl compound,
suggesting that at least three carbonyl groups are present in
the unknown product. Some possible intermediates that
possess these features are shown below.
1
5).
Carbon and nitrogen balances on oxidized ADNTs steadily
improve with conversion as the intermediate carbonyls are
oxidatively cleaved to shorter, more stable oxalic or pyruvic
acids, and amide nitrogen in oxamic acid and pyruvamide
is oxidized to nitrate (Tables 1-4). Similar results are reported
for ozonation of aniline and aminobenzoic acid, where most
nitrogen is accounted for as ammonia, with very little nitrate
formation and glyoxal, the major organic intermediate from
VOL. 34, NO. 3, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5 0 3

SCHEME 2. Proposed Reaction Pathways for Nitro Ketones
NH, with SERDP funds from the Waterways Experiment
Station, Vicksburg, MS, under Contract DACA-39-95-K-0046.
Literature Cited
(
1) Spanggord, R. J.; Yao, C. C. D.; Mill, T. Environ. Sci. Technol.,
in press.
(
(
(
2) Criegee, R.; Wenner, G. Ann. 1949, 546, 9.
3) Bailey, P. S. Chem. Rev. 1958, 58, 925-1010.
4) Zbarskii, V. L.; Sonis, M. A.; Orlova, E. Y. Zh. Prikladnoi Khim.
1
971, 44, 2578-2579.
5) Spanggord, R. J.; Clizbe, L. A. J. Labelled Compd. Radiopharm.
998, 41, 615-621.
6) Tits, M.; Bruylant, A. Bull. Soc. Chim. Bel. 1948, 57, 50-64.
7) Claison, H.; Shadwell, K. In Berichte Organische Chemie; 1910;
Band III, p 620.
8) Bader, H.; DN Hoign e´ , J. Water Res. 1981, 15, 448-456.
(
aniline, is oxidized to oxalic acid on further ozonation (15-
1
1
7).
(
(
Nitrohydroperoxide intermediates 5 and 6 in Scheme 1
almost certainly hydrolyze promptly to nitro ketones and
and then undergo further hydrolytic or redox cleavage,
2 2
H O
(
(
producing aldehydes and nitrate or carboxylic acids and
nitrite. The nitro ketone structure has not been reported to
our knowledge, and its existence must be transient in water.
An alternative intramolecular pathway from 5 or 6 to
carboxylic acids and nitrate cannot be ruled out. Nitrite ion
rapidly oxidizes to nitrate with ozone (18). Scheme 2 sum-
marizes these reactions.
The nitro ketone hydrolyzes to the carboxylic acid and
nitrite by pathway A. Nitrite would be oxidized to nitrate by
ozone using pathway C. Pathway B is a reduction leading to
an aldehyde and nitrate. Since nitrite is observed to form
with aldehydes in the initial oxidation (glyoxal, pyruvalde-
hyde, and oxamic acid aldehyde), both pathways A and B
appear to be possible for the transformation of a nitro ketone.
The treatment of 2- or 4-ADNT with ozone produces a
variety of small molecules with no observable mineralization.
The products, however, are readily assimilated into biologic
pathways such as the tricarboxylic acid cycle (pyruvic acid)
and glyoxylate cycle (glyoxylic acid) (19). Ozonation offers a
rapid and effective technology to reduce the hazards of
ADNTs in the environment. One application could be the
finishing step after bioslurry treatment of 2,4,6-trinitrotoluene
9) Kieber, R. J.; Mopper, K. Environ. Sci. Technol. 1990, 24, 1477-
1
481.
(
(
(
10) Thayer, J. R.; Huffaker, R. C. Anal. Biochem. 1980, 27, 110-119.
11) Glover, D. J.; Hoffsommer, J. C. J. Chromatogr. 1974, 94, 334-7.
12) Woodburn, K. B.; Fontaine, D. D.; Bjerke, E. L. Environ. Toxicol.
Chem. 1989, 8, 769-775.
(13) Bailey, P. Reactions of ozone with various organic functional
groups. In Water and Wastewater Treatment; Rice, R. G., Cotruvo,
J. A., Eds.; International Ozone Institute: Cleveland OH, 1975;
Vol. 1, pp 101-119.
(
14) Dowideit, P.; von Sonntag, C. Environ. Sci. Technol. 1998, 32,
112-1119.
1
(
(
(
(
15) Caprio, V.; Insola, H. Ozone Sci. Eng. 1985, 7, 169-17.
16) Caprio, V.; Insola, H. Inquinamento 1984, 26, 33-36.
17) Dore, M.; Legube, B. J. Fr. Hydrol. 1983, 14, 11-30.
18) Neta, P.; Huie, R. E.; Ross, A. B. J. Phys. Chem. Ref. Data 1988,
1
7, 1083.
(
(
19) Brock, T. D. Biology of Microorganisms, 7th ed.; Prentice Hall:
Englewood Cliffs, NJ, 1994; pp 107-108, 612-613.
20) Griest, W. H.; Vas, A. A.; Stewart, A. J.; Ho, C.-H. Chemical and
Toxicological Characterization of Slurry Reactor Biotreatment
of Explosives-Contaminated Soils; Report SFIM-AEC-ET-CR-
9
1
6186; Oak Ridge National Laboratory: Oak Ridge, TN, August
998.
(
TNT; 20).
Received for review February 17, 1999. Revised manuscript
received August 2, 1999. Accepted October 24, 1999.
Acknowledgments
This study was funded by the U.S. Army Corps of Engineers
Cold Regions Research and Engineering Laboratory, Hanover,
ES990190H
5
0 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 3, 2000