Oxidation of dichloroanilines and related anilides catalyzed by iron(III) tetrasulfonatophthalocyanine

Article abstract of DOI:10.1002/(sici)1099-0682(199912)1999:12<2319::aid-ejic2319>3.0.co;2-1

We investigated the degradation of polychlorinated pollutants, such as dichloroanilines and related anilides, catalyzed by iron(III) tetrasulfonatophthalocyanine (FePcS) with potassium monopersulfate or hydrogen peroxide as oxidant. The reaction is influenced by the positions of the two chloro-substituents and by the nature of the oxidant. The FePcS- catalyzed oxidation of 3,5-dichloroaniline with potassium monopersulfate leads to the formation of more biodegradable products (carboxylic acids) and to potentially toxic dimers (azo and azoxy compounds). The oxidation of 3,4- dichloroaniline by FePcS/H₂O₂ converts this pollutant into coupling products. The formation of dimers in the catalytic oxidation of dichloroanilines can be avoided by acylation of the amine function.

Full text of DOI:10.1002/(sici)1099-0682(199912)1999:12<2319::aid-ejic2319>3.0.co;2-1

FULL PAPER
Oxidation of Dichloroanilines and Related Anilides Catalyzed by Iron(III)
Tetrasulfonatophthalocyanine
Anke Hadasch[a] and Bernard Meunier*[a]
Keywords: Dichloroanilines / Propanil / Iron(III) tetrasulfonatophthalocyanine / Oxidations / Catalysts
We investigated the degradation of polychlorinated
pollutants, such as dichloroanilines and related anilides,
catalyzed by iron(III) tetrasulfonatophthalocyanine (FePcS)
with potassium monopersulfate or hydrogen peroxide as
potassium monopersulfate leads to the formation of more
biodegradable products (carboxylic acids) and to potentially
toxic dimers (azo and azoxy compounds). The oxidation of
3,4-dichloroaniline by FePcS/H2O2 converts this pollutant
oxidant. The reaction is influenced by the positions of the into coupling products. The formation of dimers in the
two chloro-substituents and by the nature of the oxidant. The
FePcS-catalyzed oxidation of 3,5-dichloroaniline with
catalytic oxidation of dichloroanilines can be avoided by
acylation of the amine function.
the following two isomers: 3,4-dichloroaniline (3,4-DCA)
and 3,5-dichloroaniline (3,5-DCA). 3,4-DCA is formed
during the biodegradation of the herbicide Propanil (3,4-
dichloropropionanilide).[2] From 1970 to 1987, on average,
3Ϫ7 kg of Propanil/ha were applied each year to about 70
to 100% of rice culture in the United States.[7] 3,5-DCA,
which is the most toxic DCA isomer, is a relevant precursor
in the industrial synthesis of azo dyes.
Introduction
Dichloroanilines (DCAs) are used as precursors on a
large scale in the industrial synthesis of pesticides, plastics
and dyes.[1] The contamination of aquatic environments by
DCAs can occur directly through their accidental release
during industrial processing or indirectly by microbial
degradation of DCA-derived pesticides in soil. The herbi-
cide Propanil (3,4-dichloropropionanilide, 3,4-DCPA) is
used extensively around the world for control of weeds in
rice cultures.[2] Although Propanil itself has a low toxicity,
its conversion into 3,4-DCA by microorganisms in soil is a
problem, since the acute exposure to this DCA isomer re-
sults in renal and hepatic toxicity in vivo.[3] In order to
solve this problem several investigations have been carried
out to find efficient systems for DCA degradation by re-
ductive or oxidative dehalogenation. Using the latter
method, Pieper et al. reported the degradation of 3,4-DCA
by lignin peroxidase from Phanerochaete chrysosporium.[4]
Whereas 60% of the pollutant was degraded to a quin-
oneimine, 10Ϫ15% of 3,4-DCA was transformed into the
more toxic tetrachloroazobenzene.
We recently developed a new catalytic system for the oxi-
dative degradation of recalcitrant pollutants, such as po-
lychlorinated phenols[5] and polycyclic aromatic hydro-
carbons,[6] using iron(III) tetrasulfonatophthalocyanine
(FePcS) as a catalyst. We found that the FePcS-catalyzed
oxidation of 2,4,6-trichlorophenol (TCP) with hydrogen
peroxide leads to the formation of chloromaleic acid as the
main product and also to carbon dioxide.[5aϪ5d] Because of
its high catalytic activity in the degradation of TCP, we de-
cided to investigate the FePcS-catalyzed oxidation of other
poorly biodegradable compounds such as dichloroanilines.
In order to investigate the chemical degradation of di-
chloroanilines (DCA) we studied the catalytic oxidation of
Results
We first investigated the oxidative degradation of the two
different dichloroanilines 3,5-DCA and 3,4-DCA using
FePcS as the catalyst and potassium monopersulfate or hy-
drogen peroxide as the oxidant. The final concentrations of
DCA and oxidant (H2O2 or KHSO5) in the reaction mix-
ture were 1 m and 5 m, respectively. The solution con-
tained 0.03 m of FePcS, which corresponds to a molar ra-
tio of catalyst to substrate of 3%. The reactions were carried
out at room temperature at pH 7 in an aqueous mixture of
acetone/phosphate buffer (v/v ϭ 1:9) with a final volume
of 4 mL. The results are reported in Table 1.
Table 1. Oxidation of 3,4-DCA and 3,5-DCA
Run Substrate
Oxidant Conversion [%] after ClϪ/DCA[a]
10 min
30 min
[b]
[c]
[b]
[c]
1
2
3
4
5
6
7
8
3,4-DCA
3,4-DCA
3,4-DCA
3,4-DCA
3,5-DCA
3,5-DCA
3,5-DCA
3,5-DCA
KHSO5
100
41
90
0
100
38
80
0
100
100
100
0
0.4
0.5
0
[b]
H2O2
KHSO5
[c]
H2O2
Ϫ
KHSO5
100
78
0.6
0.4
0.1
Ϫ
[b]
H2O2
KHSO5
100
0
[c]
H2O2
[a]
[a]
Laboratoire de Chimie de Coordination du CNRS, 205 route de
Number of released chloride ions per converted DCA molecule
[b]
Narbonne,
(determined after 60 min). Ϫ Reactions carried out with FePcS,
[c]
F-31077 Toulouse cedex 4, France
E-mail: bmeunier@lcc-toulouse.fr
molar ratio FePcS/DCA ϭ 3%. Ϫ
Reactions carried out wi-
thout FePcS.
Eur. J. Inorg. Chem. 1999, 2319Ϫ2325
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1434Ϫ1948/99/1212Ϫ2319 $ 17.50ϩ.50/0
2319

A. Hadasch, B. Meunier
FULL PAPER
Under these conditions 3,4-DCA and 3,5-DCA were
In order to identify the reaction products we extracted
completely converted by the FePcS/KHSO5 system within the aqueous mixture after the oxidation of 3,5-DCA by
10 min (runs 1 and 5). When hydrogen peroxide was used FePcS/KHSO5 and analyzed the organic phase by 1H NMR
instead of potassium monopersulfate, the reaction rates de- spectroscopy and GC-MS. Three groups of oxidation prod-
creased and the conversions of 3,4-DCA and 3,5-DCA were ucts could be distinguished: (i) ring cleavage products, (ii)
41% and 38% after 10 min and increased to 100% and 78% products resulting from oxidation of the amine function
within 30 min, respectively (runs 2 and 6). The average and (iii) oxidative coupling products (Figure 1). The forma-
dechlorination values of these reactions corresponded to 0.5 tion of ring cleavage products probably occurred via the
released chloride ions per converted substrate molecule. In intermediates 2,6-dichloro-1,4-quinoneimine (1) and 2,6-di-
other words, one covalent chlorine atom of 50% of DCA chloro-1,4-benzoquinone (2). Their respective reduced
molecules was transformed into a chloride ion.
forms, 2,6-dichloro-4-aminophenol and 2,6-dichloro-1,4-
In control reactions we tested the behavior of these DCA hydroquinone, were identified by GC-MS. The ring cleavage
molecules without FePcS catalyst. Surprisingly, we ob- of intermediates 1 and 2 can lead to the formation of
served fast reactions of DCAs with KHSO5 and the pol- chloromaleic acid (3) and formic acid (4) as occurs in the
lutants were completely converted within 30 min (runs 3 degradation of 2,4,6-trichlorophenol by FePcS/H2O2.[5]
and 7). Despite full substrate conversions, the dechlori- Carboxylic acids 3 and 4 were identified by 1H NMR spec-
nation values were negligible, suggesting the formation of troscopy after extraction of the reaction mixture in the
oxidation products different to those obtained in the cata- manner previously described for the TCP oxidation prod-
lyzed reaction with FePcS/KHSO5. A similar oxidation re- ucts.[5c] 3,5-Dichloronitrosobenzene (5) and 3,5-dichloro-
action was also observed on replacing acetone by ethanol, nitrobenzene (6) resulted from the oxidation of the amine
suggesting that a dioxirane was not entirely responsible for function. Both compounds are minor products of the cata-
this noncatalyzed oxidation of DCAs (see ref.[8] for a review lytic oxidation reaction, since only traces of these com-
article on the formation of dioxiranes from ketones and pounds were detected by GC-MS and HPLC. Most of the
monopersulfate). The absence of chloride release suggested 3,5-DCA was transformed into coupling products such as
that the amine function was the only oxidation site in these 3,5,3Ј,5Ј-tetrachloroazobenzene (7) and 3,5,3Ј,5Ј-tetrachlo-
noncatalyzed oxidations with monopersulfate. The control roazoxybenzene (8). A third dimer was also detected by
reactions with hydrogen peroxide gave no substrate conver- GC-MS. The MS data for this compound ([Mϩ] ϭ 316,
sions without catalyst (runs 4 and 8).
isotopic peaks indicating four chlorine atoms) are in agree-
These results indicate that the catalytic activity of FePcS ment with 1,3,6,8-tetrachlorophenazine (9).
is higher with potassium monopersulfate than with hydro-
In order to compare the distribution and the structure of
gen peroxide. Nevertheless, full substrate conversion of 3,4- the degradation products obtained with FePcS/KHSO5 and
DCA could also be reached using the clean oxidant H2O2 the oxidation compounds of the noncatalyzed reaction, the
(water being the only by-product after reaction). The oxidation of 3,5-DCA was carried out with KHSO5 only.
dechlorination values in these catalyzed oxidations of DCA In this case, 38% of 3,5-DCA was transformed into 3,5-
derivatives with KHSO5 or H2O2 were similar and sug- dichloronitrobenzene (6), which is the major reaction prod-
gested the same level of pollutant degradation regardless of uct quantified by HPLC. In accordance with the negligible
the nature of the oxidant.
dechlorination value (0.1 ClϪ/3,5-DCA), the other products
Figure 1. Products formed in the oxidation of 3,5-DCA by FePcS/KHSO5
2320
Eur. J. Inorg. Chem. 1999, 2319Ϫ2325

Oxidation of Dichloroanilines and Related Anilides Catalyzed by Iron(III) Tetrasulfonatophthalocyanine
FULL PAPER
identified by GC-MS were 3,5-dichloronitrosobenzene (5) Table 2. Oxidation of 3,4-DCPA and 3,5-DCAA
and small amounts of dimers 7 and 8. It should be noted
Run Substrate
Oxidant Conversion [%] after ClϪ/DCA[a]
that no ring cleavage products could be identified by GC-
MS in this noncatalyzed oxidation of 3,5-DCA.
10 min
30 min
We subsequently characterized the reaction products of
the FePcS-catalyzed oxidation of 3,4-DCA with hydrogen
peroxide as the oxidant (Figure 2). After a reaction time
of 1 h, 69% of the pollutant was transformed into more
hydrophobic products, which precipitated from the aqueous
mixture. The separation of the precipitated components by
column chromatography allowed the isolation and quantifi-
cation of products 10Ϫ13. As shown in Figure 2, this di-
chloroaniline was mainly transformed into tetrachloroazo-
benzene 10 (22%), whereas the tetrachloroazoxybenzene 11
(7%) and the two tetrachlorophenazine isomers 12 and 13
(3%) were isolated in lower yields. We also analyzed the
aqueous phase after filtration, but no ring cleavage products
could be identified.
[b]
[c]
[b]
[c]
1
2
3
4
5
6
7
8
3,4-DCPA
3,4-DCPA
3,4-DCPA
3,4-DCPA
KHSO5
31
15
1
34
34
2
0.3
1.0
0
[b]
H2O2
KHSO5
[c]
H2O2
0
0
0
3,5-DCAA KHSO5
22
1
4
0
22
3
8
0
0
[b]
3,5-DCAA H2O2
0
3,5-DCAA KHSO5
0
[c]
3,5-DCAA H2O2
0
[a]
Number of released chloride ions per converted 3,4-DCPA or
[b]
3,5-DCAA molecule (determined after 60 min). Ϫ
Reactions
carried out with FePcS, molar ratio FePcS/3,4-DCPA and FePcS/
[c]
3,5-DCAA ϭ 3%. Ϫ Reactions carried out without FePcS.
gible conversions (runs 3 and 7) and no reaction occurred
The oxidation of chlorinated aromatic amines leads to with H2O2 (runs 4 and 8).
the formation of chlorinated azo and azoxy compounds,
The dechlorination value of the oxidation of 3,4-DCPA
which are cytotoxic and teratogenic molecules.[1] In order by FePcS/H2O2 corresponded to one released chloride ion
to avoid the formation of these dimers during the oxidation per converted substrate molecule and indicated an advanced
of DCA we decided to protect the amine function with an degradation of the anilide. In order to identify the degra-
acetyl or a propionyl group and we synthesized the 3,4- dation products, the reaction mixture was extracted after
dichloropropionanilide 3,4-DCPA (Propanil) and the 3,5- the oxidation of 3,4-DCPA by FePcS/H2O2 and the organic
dichloroacetanilide 3,5-DCAA. To compare the reaction phase was analyzed by 1H NMR spectroscopy and GC-MS.
behavior of these anilides with that of the corresponding Four oxidation products were distinguishable and the pro-
anilines, we checked the degradation of 3,4-DCPA and 3,5- posed structures have to be considered as current working
DCAA with hydrogen peroxide or potassium monopersulf- hypotheses (Figure 3).
ate using FePcS as a catalyst. The reaction conditions were
identical to those of DCA oxidations and the results are onyl-1,4-quinoneimine (14), which is the most oxidized
shown in Table 2. product that was identified by GC-MS. The additional oxy-
3,4-DCPA was degraded to 2-chloro-hydroxy-N-propi-
The FePcS-catalyzed oxidations of 3,4-DCPA and 3,5- gen atom is probably introduced into the ring as the mass
DCAA with potassium monopersulfate led to conversions fragment of the propionyl group was unchanged after the
of 34% and 22% within 30 min, respectively (runs 1 and 5). oxidation. The GC-MS data of another product indicated
The catalytic system using hydrogen peroxide as the oxidant the oxidation of the aromatic ring of 3,4-DCPA. In accord-
was able to convert 34% of 3,4-DCPA, while only a small ance with a molecular peak [(M3,4-DCPA ϩ 16)ϩ] and a mass
amount of 3,5-DCAA was oxidized (runs 2 and 6). We also fragment suggesting an unchanged amide function we pro-
checked the reaction behavior of the anilides in the absence pose the formation of 3,4-dichlorohydroxypropionanilide
of a catalyst: the control reactions with KHSO5 led to negli- (15). We detected a second product having a molecular peak
Figure 2. Products formed in the oxidation of 3,4-DCA by FePcS/H2O2; indicated percentages are isolated yields after a 69% conversion
of the substrate
Eur. J. Inorg. Chem. 1999, 2319Ϫ2325
2321

A. Hadasch, B. Meunier
FULL PAPER
dation potential for the isomer with chloro-substituents in
positions 3 and 4.
The oxidation of both DCA derivatives catalyzed by
FePcS is faster with potassium monopersulfate than with
hydrogen peroxide. Sorokin et al. reported that the oxi-
dation of 2,4,6-trichlorophenol (TCP) catalyzed by FePcS
was also remarkably influenced by the oxidant and oc-
curred much faster with KHSO5.[5] This characteristic
could be explained by the formation of different active spec-
ies of the catalyst with potassium monopersulfate or hydro-
gen peroxide. The authors proposed that the activation of
KHSO5 by FePcS leads to the formation of an iron(IV)-oxo
species (FeIVϭO), which is able to abstract electrons from
aromatic substrates such as DCA or TCP, but unable to
transfer the oxygen atom of the iron(IV)-oxo species to an
olefin, styrene or cyclohexene.[5c] Furthermore, recent data
Figure 3. Proposed structures for the products of the oxidation of regarding the oxidation of polycyclic aromatic hydro-
3,4-DCPA by FePcS/H2O2
carbons by FePcS/H2O2 indicate that the activation of H2O2
by FePcS probably generates two different kinds of active
species: an iron(IV)-oxo entity (FeIVϭO) and an iron(III)-
hydroperoxo species (FeIIIϪOOH).[6] The oxidation poten-
tial of the iron(IV)-oxo species is expected to be higher than
that of the lower valent iron(III)-hydroperoxo entity. When
potassium monopersulfate is used as the oxidant, the con-
centration of iron(IV)-oxo species is higher than with hy-
drogen peroxide and, consequently, we obtained faster sub-
strate conversions.
[(M3,4-DCPA ϩ 16)ϩ]. The corresponding mass spectrum
indicated that oxidation on the aromatic ring or on the pro-
pionyl group had not taken place. These characteristics are
in agreement with the proposed formula of 3,4-dichloro-
N-hydroxypropionanilide (16). Furthermore, we detected a
product with a mass spectrum suggesting the incorporation
of an oxygen atom into the propionyl substituent. The fol-
lowing structural hypotheses have to be considered: (i) for-
mation of an ester by insertion of an oxygen atom, (ii) oxi-
dation of the methylene group or (iii) oxidation of the
methyl group. A carbonyl group can be oxidized to the cor-
responding ester by an oxygen atom donor such as m-
chloroperbenzoic acid or KHSO5 through a BayerϪVilliger
reaction.[9] This kind of reaction is rather unexpected in the
The formation of the oxidation products from 3,5-DCA
can be explained in terms of the mechanism depicted in
Figure 4. Two successive one-electron oxidations of 3,5-
DCA and a deprotonation lead to the formation of the cat-
ion A, with a positive charge delocalized on the nitrogen
atom and on the ortho or para positions of the ring. The
nucleophilic addition of a water molecule to the nitrogen
atom and further deprotonation gave the 3,5-dichloro-hy-
droxylaniline B (pathway a). As proposed by Pothuluri et
al.,[2] compound B can react with another substrate mol-
ecule to form the 3,5,3Ј,5Ј-tetrachlorohydrazobenzene C,
which is then oxidized to the 3,5,3Ј,5Ј-tetrachloroazoben-
zene (7) and to the 3,5,3Ј,5Ј-tetrachloroazoxybenzene (8).
The addition of a water molecule at the para position of the
cation A (pathway b in Figure 4) generates intermediate D,
which is transformed into 2,6-dichloro-1,4-quinoneimine
(1) by deprotonation and removal of two electrons. The pro-
posed reaction mechanism suggests that the degradation of
dichloroanilines and their transformation into dimers pro-
ceeds via the same intermediate A. In the case of 3,5-DCA,
two successive one-electron oxidations are necessary to gen-
presence of hydrogen peroxide even if a nucleophilic FeIII
-
peroxo species is formed with FePcS.[5c] Therefore, we pro-
pose the oxidation of the acyl methylene group, since sec-
ondary CϪH bonds of a methylene group in the α-position
with respect to a carbonyl group should be weaker than
tertiary CϪH bonds of a methyl group. The proposed prod-
uct is the 3,4-dichloro-2-hydroxy-propionanilide 17. Coup-
ling products, such as the azobenzene 10 and the azoxyben-
zene 11, were not detected in the degradation of 3,4-DCPA
by FePcS/H2O2.
Discussion
In the first part of this work we investigated the behavior erate the 3,5-dichloro-hydroxylaniline B, whereas four elec-
of two dichloroanilines, 3,4-DCA and 3,5-DCA, in oxi- trons have to be abstracted from 3,5-DCA to produce the
dations catalyzed by FePcS. In order to check the influence 2,6-dichloro-1,4-quinoneimine 1. This electrophilic com-
of the chloro-substituents on the reaction rate we compared pound can then be further oxidized to ring-cleavage prod-
the conversions obtained for these two dichloroaniline iso- ucts by addition of peroxidic oxidants, either activated or
mers. The reactions catalyzed by FePcS with potassium not activated by FePcS, as shown in quinone intermediates
monopersulfate as oxidant were too fast to detect differ- observed in the oxidative degradation of 2,4,6-trichloro-
ences in the reaction rates by HPLC analyses. However, in phenol.[5c]
the catalytic oxidations with hydrogen peroxide, 3,4-DCA
In the second part of this work we investigated the cata-
is converted faster than 3,5-DCA, suggesting a lower oxi- lytic oxidation of the dichloroanilides 3,4-DCPA and 3,5-
2322
Eur. J. Inorg. Chem. 1999, 2319Ϫ2325

Oxidation of Dichloroanilines and Related Anilides Catalyzed by Iron(III) Tetrasulfonatophthalocyanine
FULL PAPER
were always produced in the catalytic oxidations of DCA
derivatives under similar conditions.
Conclusion
The present study indicates that the catalytic systems
FePcS/KHSO5 and FePcS/H2O2 are able to oxidize recalci-
trant pollutants such as dichloroanilines and related anil-
ides. One part of the pollutant 3,5-DCA was partially trans-
formed by FePcS/KHSO5 into more biodegradable prod-
ucts such as chloromaleic acid (3) and formic acid (4),
whereas the other part of 3,5-DCA was oxidized to toxic
dimers such as the azobenzene 7 and the azoxybenzene 8.
The FePcS-catalyzed oxidation of 3,4-DCA with hydrogen
peroxide gave mainly coupling products without ring cleav-
age products.
The noncatalyzed oxidations of DCA isomers with
KHSO5 led to fast conversions of the pollutants without
chlorine release, indicating that the FePcS catalyst is neces-
sary for the oxidative ring cleavage of the dichloroaniline
derivatives.
The oxidation of DCA by these bio-inspired systems is,
like the degradation of 3,4-DCA by lignin peroxidase from
Phanerochaete chrysosporium,[4] a double-edged sword,
since the formation of more biodegradable compounds and
toxic dimers occurs simultaneously. The formation of oligo-
mers could be avoided when these amines were converted
into amides before the catalytic oxidation. Unfortunately,
the electron-withdrawing acyl substituents are responsible
for slower substrate conversions, but the optimization of the
reaction conditions might be able to produce full substrate
conversions without the formation of toxic coupling prod-
ucts in these catalytic degradation reactions of dichloroani-
line derivatives.
Figure 4. Proposed mechanism for the oxidation of 3,5-DCA by
FePcS/KHSO5; compounds indicated by numbers have been identi-
fied whereas derivatives assigned by capital letters are only puta-
tive intermediates
DCAA, prepared by reacting the respective DCA isomers
with propionic or acetic anhydride. The results of the cata-
lytic oxidations indicate that the introduction of an elec-
tron-withdrawing substituent such as a propionyl or an ace-
tyl group reduced the conversion rates of these anilides
compared with the corresponding amines. In addition, the
anilide 3,5-DCAA, with two chloro-substituents on posi-
tions 3 and 5 of the aromatic ring, is more difficult to oxid-
ize than the 3,4-isomer 3,4-DCPA. It should be noted that
we observed the same tendency in the oxidations of 3,5-
DCA and 3,4-DCA because only the latter was fully con-
verted by FePcS/H2O2 (Table 1, runs 2 and 6).
Experimental Section
Materials: Potassium monopersulfate (the triple salt [KHSO5]2
[KHSO4][K2SO4], Curox ) was a gift from Peroxid Chemie GmbH.
Hydrogen peroxide was obtained from Acros as a 35 wt.% solution.
3,4-DCA, 3,5-DCA, 3,5-dichloronitrobenzene and 2,6-dichloro-
1,4-benzoquinone were purchased from Aldrich. Iron tetrasulfona-
tophthalocyanine (FePcS) was prepared according to the modified
method of Weber and Busch.[11][12]
The oxidation of 3,4-DCPA by FePcS/H2O2 gave the hy-
droxylated N-propionyl-quinoneimine 14, which is formed
after release of one chloride. The analytical data suggest
Analytical procedures: The conversions of substrates were moni-
tored by HPLC (Waters 510 pump, Waters 486 detector) using a
that the formation of products 15, 16 and 17 resulted from µ-Bondapak C18 column. The eluent was a mixture of acetonitrile/
phosphate buffer 0.05  (pH 7.0) in proportions 55:45 (v/v) for
oxidations of 3,5-DCA and 3,4-DCPA and 40:60 (v/v) for oxi-
dations of 3,4-DCA and 3,5-DCAA. The flow rate was 1 mL/min
and the detection was performed at 220 nm. Gas chromatography-
mass spectrometry data (GC-MS), except for compounds 1 and 2,
were recorded on a Hewlett-Packard 5890 instrument by electron-
impact ionization at 70 eV. The carrier gas for GC-MS was helium
oxidations of the aromatic ring, the amide function or the
acyl chain of 3,4-DCPA, respectively. The dechlorination
value (1.0 ClϪ/converted 3,4-DCPA) indicates that other
unidentified dechlorinated products should be formed,
since 14 is the only mono-chlorinated compound observed.
Coupling products such as azo or azoxy compounds were
not detected in the catalytic oxidation of 3,4-DCPA by
FePcS/H2O2. The presence of the propionyl substituent on
the aniline derivative blocked the formation of toxic dimers
and
a non polar capillary column (12 m ϫ 0.2 mm HL-1,
crosslinked methylsilicone) was used. The injector temperature was
250°C, and analyses were performed at 80°C for 2 min, then up
such as the azobenzene 10 and the azoxybenzene 11, which to 250°C (10°C/min) for a further 15 min. The GC-MS data for
Eur. J. Inorg. Chem. 1999, 2319Ϫ2325 2323

A. Hadasch, B. Meunier
compounds 1 and 2 were obtained using a Hewlett-Packard 5890 verified the GC-MS data by co-injection of a solution containing
FULL PAPER
spectrometer with a nonpolar capillary column (50 m ϫ 0.22 mm
BPX5). The injector temperature was 250°C, and analyses were
3,5-DCA and chloromaleic acid, and the same adduct (retention
time and mass spectrum) was formed during the analysis. Chloro-
performed at 100°C for 0.5 min, then up to 325°C (6°C/min) for a maleic acid was prepared according to the method of Gruzdev
further 30 min. 1H NMR spectra were recorded with a Bruker WM
250 spectrometer. The concentration of chloride ions was deter-
mined by the mercuric thiocyanate method.[10]
and Gruzdev.[13]
1
Formic acid (4): H NMR ([D6] DMSO): δ ϭ 8.24 (s, CH). Formic
acid formed an adduct with traces of unchanged 3,5-DCA under
the conditions of GC-MS analysis. GC-MS data of the adduct: m/
z ϭ 191 [(M ϩ 2)ϩ], 189 [Mϩ], 163 [(M ϩ 2 Ϫ CO)ϩ], 161 [(M Ϫ
CO)ϩ]. We verified the data by co-injection of a solution containing
3,5-DCA and formic acid, and the same adduct (retention time and
mass spectrum) was formed during the analysis.
General Procedure for Catalytic Oxidations: The reaction mixture,
with a final volume of 4 mL, contained 4 µmol of substrate, 0.12
µmol of FePcS and 5 µmol of KHSO5 or 6 µmol of H2O2. The
reaction mixtures were prepared as follows: 80 µL of a substrate
solution (50 m) in acetone was mixed with 80 µL of an internal
standard (4-fluoronitrobenzene, 50 m) in acetone. 1.5 mL of an
aqueous FePcS solution (80 µ) was added and the reaction mix-
ture was adjusted to a final volume of 4 mL with 400 µL of phos-
phate buffer (500 m, pH 7), 240 µL of acetone and 1.7 mL of
water for 3,5-DCA, 3,4-DCA and 3,4-DCPA (acetone/water ϭ 1:9,
v/v) or with 400 µL of phosphate buffer (500 m, pH 7), 640 µL
of acetone and 1.3 mL of water for 3,5-DCAA (acetone/water ϭ
2:8, v/v). Finally, 175 µL of H2O2 (0.12 ) or 50 µL of KHSO5 (0.4
) was added and the reaction mixture was stirred at room tem-
perature.
3,5-Dichloronitrosobenzene (5): GC-MS: m/z ϭ 177 [(M ϩ 2)ϩ],
175 [Mϩ], 147 [(M ϩ 2 Ϫ NO)ϩ], 145 [(M Ϫ NO)ϩ], 111 [(M ϩ 2
Ϫ HNOCl)ϩ], 109 [(M Ϫ HNOCl)ϩ].
3,5-Dichloronitrobenzene (6): GC-MS: m/z ϭ 193 [(M ϩ 2)ϩ], 191
[Mϩ], 163 [(M ϩ 2 Ϫ NO)ϩ], 161 [M Ϫ NO)ϩ], 147 [(M ϩ 2 Ϫ
NO2)ϩ], 145 [(M Ϫ NO2)ϩ]. Ϫ HPLC: product 6 was also charac-
terized by co-injection of an authentic commercial sample which
had the same retention time. The quantification of compound 6 in
the oxidation of 3,5-DCA with KHSO5, corresponding to run 7 in
Table 1, was recorded by HPLC with 4-fluoronitrobenzene as an
internal standard.
Characterization of the Oxidation Products of 3,5-DCA: The reac-
tion was carried out under the conditions described for run 5 in
Table 1 with a larger (25 times) reaction volume (100 mL). The
reaction mixture was stirred for 1 h at room temperature. The ace-
tone and water were then evaporated under vacuum at 50°C and
the residue was dried for 1 h under vacuum at room temperature.
The residue was dissolved in 10 mL of 1  HCl saturated with
NaCl (resulting pH ϭ 2). The products were extracted with diethyl
ether (4 ϫ 60 mL). After evaporation of the ether, the dark brown
residue was dried under vacuum for 2 h at room temperature and
dissolved in 1 mL of diethyl ether. The resulting solution was ana-
lyzed by GC-MS. In order to analyze the product mixture by 1H
NMR spectroscopy the diethyl ether was evaporated and the resi-
due was dissolved in 1 mL of deuterated dimethyl sulfoxide ([D6]
DMSO).
3,5,3Ј,5Ј-Tetrachloroazobenzene (7): GC-MS: m/z ϭ 322 [(M ϩ
4)ϩ, 12], 320 [(M ϩ 2)ϩ, 20], 318 [Mϩ, 14], 175 [(M ϩ 2 Ϫ
C6H3Cl2)ϩ, 18], 173 [(M Ϫ C6H3Cl2)ϩ, 30], 147 [(M ϩ 2 Ϫ
C6H3N2Cl2)ϩ, 55], 145 [(M Ϫ C6H3N2Cl2)ϩ, 82].
3,5,3Ј,5Ј-Tetrachloroazoxybenzene (8): GC-MS: m/z ϭ 338 [(M ϩ
4)ϩ, 3], 336 [(M ϩ 2)ϩ, 7], 334 [Mϩ, 6], 322 [(M ϩ 4 Ϫ O)ϩ, 13],
320 [(M ϩ 2 Ϫ O)ϩ, 24], 318 [(M Ϫ O)ϩ, 20], 175 [(M ϩ 2 Ϫ
C6H3Cl2O)ϩ, 20], 173 [(M Ϫ C6H3Cl2O)ϩ, 33], 147 [(M ϩ 2 Ϫ
C6H3Cl2N2O)ϩ, 64], 145 [(M Ϫ C6H3Cl2N2O)ϩ, 100].
1,3,6,8-tetrachlorophenazine (9): GC-MS: m/z ϭ 320 [(M ϩ 4)ϩ,
48], 318 [(M ϩ 2)ϩ, 100], 316 [Mϩ, 76], 283 [(M ϩ 2 Ϫ Cl)ϩ, 17],
281 [(M Ϫ Cl)ϩ, 20], 248 [(M ϩ 2 Ϫ Cl2)ϩ, 6], 246 [(M Ϫ Cl2)ϩ, 9].
Characterization of the Oxidation Products of 3,4-DCA. ؊ Pro-
cedure for the catalytic oxidation: The reaction mixture, with a final
volume of 200 mL, contained 2 mmol of 3,4-DCA, 80 µmol of
FePcS (catalyst/substrate ratio: 4%) and 10 mmol of H2O2. The
reaction mixture was prepared as follows: 324 mg of 3,4-DCA and
82 mg of FePcS (M ϭ 1022) were dissolved in a previously pre-
pared solvent mixture containing 40 mL of acetone and 110 mL of
water. Phosphate buffer (50 mL, pH 7, 500 m) was added. Finally,
1 mL of H2O2 (12 , 35 wt.%) was added in five aliquots (5 ϫ 200
µL every 5 min) and the reaction mixture was stirred at room tem-
perature.
In order to identify the intermediate products, we modified the
work-up described above and acidified the reaction mixture after
5 min of reaction time without prior evaporation of water and ace-
tone. After extraction with 2 ϫ 60 mL of diethyl ether and evapor-
ation of the ether, the residue was contaminated with buffer salts
because of their solubility in acetone. The residue was dissolved
in 20 mL of diethyl ether and filtered to separate the salts from
oxidation products.
2,6-Dichloro-1,4-quinoneimine (1): The mass spectrum corre-
sponded to the 2,6-dichloro-4-aminophenol, which is the reduced
form of compound 1. GC-MS: m/z ϭ 180 [(M ϩ 2)ϩ], 178 [Mϩ],
144 [(M ϩ 2 Ϫ HCl)ϩ], 142 [(M Ϫ HCl)ϩ], 116 [(M ϩ 2 Ϫ
HCOCl)ϩ], 114 [(M Ϫ HCOCl)ϩ].
Isolation of the Oxidation Products: After a reaction time of 60 min,
the mixture was filtered and the remaining precipitate was dried
for 3 h under vacuum. Yield: 223 mg (69%). In order to separate
the hydrophobic products from paramagnetic iron salts, the precipi-
tate was dissolved in diethyl ether and filtered. The filtrate was
evaporated under vacuum and 155 mg (48%) of organic products
were isolated. Afterwards, the products were separated by liquid
chromatography (column on silica, eluent: dichloromethane/hex-
ane ϭ 65:35) and analyzed by GC-MS and 1H NMR spectroscopy.
2,6-Dichloro-1,4-benzoquinone (2): The mass spectrum corre-
sponded to the 2,6-dichloro-1,4-hydroquinone, which is the reduced
form of compound 2. GC-MS: m/z ϭ 179 [(M ϩ 2)ϩ], 177 [Mϩ],
143 [(M ϩ 2 Ϫ HCl)ϩ], 141 [(M Ϫ HCl)ϩ], 115 [(M ϩ 2 Ϫ
HCOCl)ϩ], 113 [(M Ϫ HCOCl)ϩ].
1
Chloromaleic acid (3): H NMR ([D6] DMSO): δ ϭ 6.67 (s, CH).
Chloromaleic acid formed an adduct with traces of unchanged 3,5-
3,4,3Ј,4Ј-Tetrachloroazobenzene (10): Yield: 71 mg (22%). Ϫ GC-
DCA under the conditions of GC-MS analysis. GC-MS data of MS: m/z ϭ 322 [(M ϩ 4)ϩ, 10], 320 [(M ϩ 2)ϩ, 19], 318 [Mϩ, 14],
the adduct: m/z ϭ 277 [(M ϩ 2)ϩ], 275 [Mϩ], 242 [(M ϩ 2 Ϫ Cl)ϩ], 175 [(M ϩ 2 Ϫ C6H3Cl2)ϩ, 18], 173 [(M Ϫ C6H3Cl2)ϩ, 28], 147
240 [(M Ϫ Cl)ϩ], 198 [(M ϩ 2 Ϫ CO2Cl)ϩ], 196 [(M Ϫ CO2Cl)ϩ], [(M ϩ 2 Ϫ C6H3N2Cl2)ϩ, 53], 145 [(M Ϫ C6H3N2Cl2)ϩ, 82]. Ϫ 1H
161 [(M ϩ 2 Ϫ CHCCl(CO)2)ϩ], 159 [(M Ϫ CHCCl(CO)2)ϩ]. We NMR (CDCl3): δ ϭ 7.61 (d, 1 H, J ϭ 7.5 Hz, C5-H, C5Ј-H), 7.79
2324
Eur. J. Inorg. Chem. 1999, 2319Ϫ2325

Oxidation of Dichloroanilines and Related Anilides Catalyzed by Iron(III) Tetrasulfonatophthalocyanine
FULL PAPER
(dd, 1 H, J ϭ 7.5 Hz, J ϭ 2.5 Hz, C6-H, C6Ј-H), 8.00 (d, 1 H, J ϭ
2.5 Hz, C2-H, C2Ј-H).
Characterization of the Oxidation Products of 3,4-DCPA: The reac-
tion was carried out under the conditions as described for run 2 in
Table 2 with a larger (25 times) reaction volume (100 mL). The
workup was identical to that used for the characterization of the
oxidation products of 3,5-DCA.
3,4,3Ј,4Ј-Tetrachloroazoxybenzene (11): Yield: 22 mg (7%). Ϫ GC-
MS: m/z ϭ 338 [(M ϩ 4)ϩ, 3], 336 [(M ϩ 2)ϩ, 7], 334 [Mϩ, 6], 322
[(M ϩ 4 Ϫ O)ϩ, 13], 320 [(M ϩ 2 Ϫ O)ϩ, 25], 318 [(M Ϫ O)ϩ, 20],
175 [(M ϩ 2 Ϫ C6H3Cl2O)ϩ, 21], 173 [(M Ϫ C6H3Cl2O)ϩ, 32], 147
[(M ϩ 2 Ϫ C6H3Cl2N2O)ϩ, 64], 145 [(M Ϫ C6H3Cl2N2O)ϩ, 100].
2-Chloro-hydroxy-N-propionyl-1,4-quinoneimine (14): GC-MS: m/
z ϭ 215 [(M ϩ 2)ϩ], 213 [Mϩ], 185 [(M Ϫ C2H4)ϩ], 161 [(M ϩ 2
Ϫ CNC2H4)ϩ], 159 [(M Ϫ CNC2H4)ϩ].
1
Ϫ H NMR (CDCl3): δ ϭ 7.52 (d, 1 H, J ϭ 7.5 Hz, C2Ј-H), 7.57
(d, 1 H, J ϭ 7.5 Hz, C2-H), 7.98 (dd, 1 H, J ϭ 10 Hz, J ϭ 2.5 Hz,
C6Ј-H), 8.12 (dd, 1 H, J ϭ 10 Hz, J ϭ 2.5 Hz, C6-H), 8.38 (d, 1 H,
J ϭ 2.5 Hz, C2Ј-H), 8.39 (d, 1 H, J ϭ 2.5 Hz, C2-H).
3,4-Dichlorohydroxypropionanilide (15): GC-MS: m/z ϭ 235 [(M ϩ
2)ϩ], 233[Mϩ], 179 [(M ϩ 2 Ϫ COC2H4)ϩ], 177 [(M Ϫ COC2H4)ϩ].
3,4-Dichloro-N-hydroxypropionanilide (16): GC-MS: m/z ϭ 235
[(M ϩ 2)ϩ], 233[Mϩ], 217 [(M ϩ 2 Ϫ H2O)ϩ], 215 [(M Ϫ H2O)ϩ],
179 [(M ϩ 2 Ϫ COC2H4)ϩ], 177 [(M Ϫ COC2H4)ϩ].
Tetrachlorophenazines 12 and 13: Yield: 9.5 mg (3%). Ϫ GC-MS:
m/z ϭ 320 [(M ϩ 4)ϩ, 48], 318 [(M ϩ 2)ϩ, 100], 316 [Mϩ, 76], 283
[(M ϩ 2 Ϫ Cl)ϩ, 17], 281 [(M Ϫ Cl)ϩ, 20], 248 [(M ϩ 2 Ϫ Cl2)ϩ, 6],
1
246 [(M Ϫ Cl2)ϩ, 9]. Ϫ 3,4,6,7-tetrachlorophenazine 12: H NMR
3,4-Dichloro-2-hydroxypropionanilide (17): GC-MS: m/z ϭ 235 [(M
ϩ 2)ϩ], 233 [Mϩ], 191 [(M ϩ 2 Ϫ CO2)ϩ], 189 [(M Ϫ CO2)ϩ], 163
[(M ϩ 2 Ϫ CO2C2H4)ϩ], 161 [(M Ϫ CO2C2H4)ϩ].
(CDCl3): δ ϭ 7.87 (d, 1 H, J ϭ 9 Hz, C2-H, C8-H), 8.09 (d, 1 H,
J ϭ 9 Hz, C1-H, C9-H). Ϫ 1,2,6,7-tetrachlorophenazine 13: 1H
NMR (CDCl3): δ ϭ 7.93 (d, 1 H, J ϭ 9 Hz, C3-H, C8-H), 8.25 (d,
1 H, J ϭ 9 Hz, C4-H, C9-H).
Preparation of 3,4-Dichloropropionanilide (3,4-DCPA): Propionic
anhydride (20 mL, 156 mmol) was added to 3,4-dichloroaniline
(600 mg, 3.7 mmol). The mixture was stirred at 60°C for 4 h and
then at 100°C for 2 h. The propionic acid and unreacted propionic
anhydride were removed by distillation (bath temperature: 70°C)
under vaccum. The remaining orange oil was dissolved in 5 mL of
dichloromethane and the product was precipitated by addition of
100 mL of hexane. The white powder was filtered off and dried
under vacuum for 4 h at 25°C. Ϫ Yield: 550 mg (68%). Ϫ 1H NMR
([D6] DMSO): δ ϭ 1.19 (t, 3 H, CH3), 2.44 (q, 2 H, CH2), 7.59
(dd, 1 H, C6-H), 7.67 (d, 1 H, C5-H), 8.12 (d, 1 H, C2-H), 10.29 (s,
1 H, NH). Ϫ GC-MS: m/z ϭ 219 [(M ϩ 2)ϩ], 217 [Mϩ], 163 [(M
ϩ 2 Ϫ COCH2CH2)ϩ], 161 [(M Ϫ COCH2CH2)ϩ]. Ϫ C9H9Cl2NO
(218): calcd. C 49.54, H 4.13, N 6.42; found C 49.62, H 3.79, N
6.38.
Acknowledgments
A. H. thanks the European Community for a PhD fellowship
(TMR grant No ERBFMBICT950030). The authors are grateful
to Alexander Sorokin (LCC, Toulouse), Alain Rabion and Laurent
Fraisse (ELF-Atochem, Lacq) for fruitful discussions.
[1]
J. Struijs, J. E. Rogers, Appl. Environ. Microbiol. 1989, 55, 2527.
[2]
J. V. Pothuluri, J. A. Hinson, C. E. Cerniglia, J. Environ. Qual.
1991, 20, 330.
[3]
M. A. Valentovic, J. G. Ball, D. K. Anestis, G. O. Rankin, Toxic,
In Vitro 1995, 9, 75.
[4]
D. H. Pieper, R. Winkler, H. Sandermann, Angew. Chem. Int.
Ed. Engl. 1992, 31, 68.
[5] [5a]
´
A. Sorokin, J.-L. Seris, B. Meunier, Science 1995, 268, 1163.
[5b]
Ϫ
A. Sorokin, S. De Suzzoni-Dezard, D. Poullain, J.-P.
[5c]
Noe¨l, B. Meunier J. Am. Chem. Soc. 1996, 118, 7410. Ϫ
A.
B.
[5d]
Preparation of 3,5-Dichloroacetanilide (3,5-DCAA): Acetic anhy-
dride (10 mL, 102 mmol) was added to a solution of 3,5-dichloro-
aniline (300 mg, 1.85 mmol) in 10 mL of acetonitrile. The reaction
mixture was stirred for 2 h at 100°C. The acetic acid and unreacted
acetic anhydride were removed by distillation (bath temperature:
100°C) under vaccum. The remaining beige oil was dissolved in
5 mL of dichloromethane and the product was precipitated by the
addition of 100 mL of hexane. The white powder was filtered off
and dried under vacuum for 3 h at 25°C. Ϫ Yield: 310 mg (82%).
Sorokin, B. Meunier, Chem. Eur. J. 1996, 2, 1308. Ϫ
Meunier, A. Sorokin, Acc. Chem. Res. 1997, 30, 470.
A. Sorokin, B. Meunier, Eur. J. Inorg. Chem. 1998, 1269.
[6]
[7]
U.S. Environmental Protection Agency, Guidance for the regis-
tration of pesticide products containing propanil as the active in-
gredient USEPA Case 0226, U.S. Gov. Print. Office, 12-23-87,
Washington, 1987.
[8]
[9]
W. Adam, L. Hadjiarapoglou, Topics Curr. Chem. 1993, 164, 45.
M. Renz, B. Meunier, Eur. J. Org. Chem. 1999, 737.
T. M. Florence, Y. J. Farrar, Anal. Chim. Acta 1971, 54, 373.
J. H. Weber, D. H. Busch, Inorg. Chem. 1965, 4, 469.
A. Hadasch, A. Sorokin, A. Rabion, B. Meunier, New J. Chem.
1998, 45.
[10]
[11]
[12]
1
Ϫ H NMR ([D6] DMSO): δ ϭ 2.18 (s, 3 H, CH3), 7.36 (s, 1 H,
C4-H), 7.76 (s, 2 H, C2-H, C6-H), 10.41 (s, 1 H, NH). Ϫ GC-MS:
m/z ϭ 205 [(M ϩ 2)ϩ], 203 [Mϩ], 163 [(M ϩ 2 Ϫ COCH2)ϩ], 161
[(M Ϫ COCH2)ϩ]. Ϫ C8H7Cl2NO (204): calcd. C 47.06, H 3.43, N
6.86; found C 47.17, H 3.15, N 6.91.
[13]
V. G. Gruzdev, G. V. Gruzdev, J. Gen. Chem. USSR 1986, 56,
1872.
Received April 13, 1999
[I99130]
Eur. J. Inorg. Chem. 1999, 2319Ϫ2325
2325