Catalytic oxidation of 2,4,6-tribromophenol using iron(III) complexes with imidazole, pyrazole, triazine and pyridine ligands

Article abstract of DOI:10.1016/j.molcata.2015.12.017

Five types of non-heme iron complexes, coordinated with imidazole, pyrazole, triazine and pyridine ligands, which had been previously synthesized, were used in the following studies. Among these complexes, the mer-[FeCl₃(terpy)] complex showed the highest catalytic activity for the oxidative degradation of 2,4,6-tribromophenol (TrBP) using KHSO₅ as an oxygen donor. The turnover numbers for the degradation and debromination of TrBP in the mer-[FeCl₃(terpy)]/KHSO₅ catalytic system were estimated to be 1890 ± 1 and 4020 ± 216, respectively. The catalytic activity was significantly inhibited at pH 4-7 in the presence of a humic acid, a major component of landfill leachates. However, the percent of TrBP degradation and debromination increased at pH 8. GC/MS analyses showed that a major oxidation product was 2,6-dibromo-p-benoquinone (DBQ) and its level decreased with increasing reaction time, suggesting that organic acids (identified by LC/TOF-MS) are formed via the ring-cleavage of DBQ. Mineralization to CO₂ was observed to be 15% as a result of the oxidation for a 3 h period, where TOC values before and after the reaction were measured. Absorption spectra of mer-[FeCl₃(terpy)] with m-chloroperoxybenzoic acids as an oxygen donor in acetonitrile showed that a center metal, Fe, formed a peroxide complex with the oxygen donor.

Full text of DOI:10.1016/j.molcata.2015.12.017

Journal of Molecular Catalysis A: Chemical 413 (2016) 100–106
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Catalytic oxidation of 2,4,6-tribromophenol using iron(III) complexes
with imidazole, pyrazole, triazine and pyridine ligands
Mami Igarashia, Qianqian Zhua, Masahide Sasakib, Ritsu Kodamaa, Kohki Odaa,
Masami Fukushimaa,∗
a Laboratory of Chemical Resources, Hokkaido University, Sapporo 060-6828, Japan
b National Institute of Advanced Industrial Science and Technology (AIST), 2-17-2-1 Tsukisamu-Higashi, Toyohira-ku, Sapporo 062-8517, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Five types of non-heme iron complexes, coordinated with imidazole, pyrazole, triazine and pyridine
ligands, which had been previously synthesized, were used in the following studies. Among these com-
plexes, the mer-[FeCl3(terpy)] complex showed the highest catalytic activity for the oxidative degradation
of 2,4,6-tribromophenol (TrBP) using KHSO5 as an oxygen donor. The turnover numbers for the degrada-
tion and debromination of TrBP in the mer-[FeCl3(terpy)]/KHSO5 catalytic system were estimated to be
1890 1 and 4020 216, respectively. The catalytic activity was significantly inhibited at pH 4–7 in the
presence of a humic acid, a major component of landfill leachates. However, the percent of TrBP degrada-
tion and debromination increased at pH 8. GC/MS analyses showed that a major oxidation product was
2,6-dibromo-p-benoquinone (DBQ) and its level decreased with increasing reaction time, suggesting that
organic acids (identified by LC/TOF-MS) are formed via the ring-cleavage of DBQ. Mineralization to CO2
was observed to be 15% as a result of the oxidation for a 3 h period, where TOC values before and after
the reaction were measured. Absorption spectra of mer-[FeCl3(terpy)] with m-chloroperoxybenzoic acids
as an oxygen donor in acetonitrile showed that a center metal, Fe, formed a peroxide complex with the
oxygen donor.
Received 4 November 2015
Received in revised form
19 December 2015
Accepted 21 December 2015
Available online 28 December 2015
Keywords:
Non-heme
Terpyridine
Catalytic oxidation
2,4,6-Tribromophenol
Humic acid
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
been reported in the case of a heterogeneous Fe(III)-porphyrin cat-
alyst, which was introduced into an ionic liquid supported Fe3O4
2,4,6-Tribromophenol (TrBP) is used in the production of fungi-
cides, wood preservatives [1] and an intermediate in the production
of brominated flame retardants (BFRs) [2]. TrBP is eluted from land-
fills where wastes derived from electrical devices including BFRs
are disposed of, and this leads to water pollution [3]. In particu-
lar, humic acids (HAs) are present in landfill leachates in the form
of dissolved organic matter and interactions with HA can enhance
water solubility and inhibit the oxidative degradation of organic
pollutants [4,5]. Because TrBP has been reported to have endocrine
disrupting effects, the TrBP in landfill leachates must be reduced.
Previous studies suggest that Fe(III)-porphyrins, a model of
heme enzymes, can be effective for the oxidative degradation of
bromophenols and chlorophenols [6–18]. Among the halogenated
phenols, bromophenols are relatively resistant to oxidative degra-
dation and dehalogenation [8]. However, mineralization to CO2 has
[19]. However, in a homogeneous catalytic system, mineralization
is minimal, because of catalyst self-degradation.
On the other hand, non-heme enzymes, such as methane
monooxygenase, riske dioxygenase, lipoxygenase and catechol
dioxygenase, involve iron as a center metal. These enzymes can
be activated by an oxygen donor, and organic substrates are oxi-
dized by active species such as L•+-Fe(IV) = O, L-Fe(IV) = O and HO•
that are produced in such systems [18]. Stable non-heme iron
complexes with tetradentate ligands had been synthesized and
applied as catalysts for the selective oxidation of olefins [19–24].
However, there are no reports, regarding the catalytic oxidation
of halogenated phenols using non-heme iron complexes. Catechol
dioxygenases can oxidize catechol, a phenol derivative, via ring-
cleavage [20]. In particulars, Dhanalkshmi et al [25]. reported that
iron complexes, when coordinated with tridentate ligands, such
as pyridine-2,6-dicarboxylic acid, 2,2ꢀ:6,2ꢀꢀ-terpyridine (terpy) and
2,6-bis(benzimidazol-2ꢀ-yl) pyridine, showed higher activities for
the oxidation of catechol. This finding suggests that a non-heme
∗
Corresponding author. Fax: +81 11 706 6304.
E-mail address: m-fukush@eng.hokudai.ac.jp (M. Fukushima).
http://dx.doi.org/10.1016/j.molcata.2015.12.017
1381-1169/© 2015 Elsevier B.V. All rights reserved.

M. Igarashi et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 100–106
101
trum, ꢀ(C H) 3047 and 767 cm−1, ꢀ(C C, C N) 1610, 1468 and
1453 cm−1
iron complex with a tridentate ligand might be effective for the
oxidation of phenol derivatives.
.
Five types of non-heme iron complexes with monodentate imi-
dazole, pyrazole and tridentate nitrogen ligands (Fig. 1), which had
been previously synthesized and their structures determined by X-
ray crystallography [26], were prepared. In the present study, the
catalytic activities of the non-heme iron complexes for the oxida-
tive degradation of TrBP and the influence of HA on the oxidation of
TrBP were investigated. In addition, to characterize the oxidation
products of TrBP produced by the non-heme catalyst, the reaction
mixtures and CH2Cl2 extracts derived from them were analyzed
by LC/TOF-MS and GC/MS, respectively. The detection of the acti-
vated species of the non-heme complex by the oxygen donor was
examined by observing the visible absorption spectrum.
2.3. Assay for TrBP degradation
A 10-mL aliquot of 0.02 M NaH2PO4/Na2HPO4 buffer (pH 4–8)
was placed in a 50-mL Erlenmeyer flask and a 50 ␮L aliquot of
0.01 M TrBP in acetonitrile was then added to the solution. Aqueous
catalysts were added to be produced concentrations of 0.1–10 ␮M,
and a 0.1 mL aliquot of 0.1 M KHSO5 or H2O2 was then added to
the mixture to start the reaction. After shaking for 30 min, a 1 mL
aliquot of the reaction mixture was placed in a glass vial, and 0.5 mL
2-propanol was then added. A 20 ␮L aliquot of the mixture was
injected into a PU980-type HPLC pumping system (JASCO) to ana-
lyze the remaining TrBP in the reaction mixtures. The mobile phase
was a mixture of methanol and water (78:22 in volume), which was
acidified with aqueous 0.08% H3PO4, and the stationary phase was
a COSMOSIL 5C18-AR-II column (4.6 × 250 nm). The flow rate of
the eluent, the column temperature and the detection wavelength
were set at 1.0 mL min−1, 50 ◦C and 290 nm, respectively.
2. Materials and methods
2.1. Materials
Imidazole (im), 1-methylimidazole (meim) (Wako Pure Chem-
ical Industries), 3-methylpyrazole (mepy), 2,4,6-tris(2-pyridil)-
1,3,5-triazine (tptz), terpy and TrBP (Tokyo Chemical Industry)
were used without further purification. An iron(III) chloride anhy-
drate was obtained from Nacalai Tesque. KHSO5 was obtained as
a triple salt, 2KHSO5·KHSO4·K2SO4 (Merck). Other reagents were
obtained from Wako Pure Chemical Industries and were used with-
out further purification. An HA used in this study was extracted
from Shinshinotsu peat soil, as described in a previous report [27].
The elemental compositions of the HA were as follows: C 54.5%, H
5.35%, N 2.17%, S 0.66%, O 35.1%, ash 2.22% [28].
The Br− that is released as a result of the oxidation was
analyzed by an ion chromatography (Dionex IC-120 type, Thermo-
Fishers) with conductivity detection. The mobile phase was
an aqueous solution of 2.7 mM Na2CO3/0.3 mM NaHCO3, and
the separation column was an IonPacAS12A analytical column
(4 × 200 mm, Thermo-Fishers) with an IonPacAG12A guard column
(4 × 50 mm). The flow rate and column oven temperature were set
at 1.5 mL min−1 and 35 ◦ C, respectively. To evaluate the percent
mineralization to CO2, the concentration of total organic carbon
(TOC) in the reaction mixture was analyzed before and after the
reaction using a TOC-V CSH type analyzer (Shimadzu). In this test,
a stock solution of 0.01 M TrBP was prepared using aqueous 0.01 M
NaOH.
2.2. Synthesis and characterization of non-heme iron complexes
2.4. Identification of oxidation products by GC/MS
Five types of non-heme iron complexes (Fig. 1) were prepared
based on the methods previously reported by Cotton et al [26]. The
detailed procedures for their preparation can be found in Supple-
mentary Material (Text SM-1). The FT-IR spectra of the synthesized
complexes were measured using an FT/IR 600-type spectropho-
tometer (Japan Spectroscopic Co., Ltd.). Elemental analyses (C,
H, N and Cl) of the complexes were carried out by the Instru-
mental Analysis Division, Equipment Management Center Creative
Research Institution, Hokkaido University (Sapporo, Japan). The Fe
contents of the complexes were determined using an inductively
coupled plasma-atomic emission spectrophotometer (ICPE9000
type, Shimadzu) after wet digestion with a mixture of HNO3/HCl
and the appropriate dilution with ultra-pure water. Analytical
data, product yields, elemental composition and FT-IR spectra for
each complex were as follows: mer-[FeCl3(meim)3], yield 13.4%;
elemental analysis, observed (calculated), %C 35.03 (35.27), %H
4.27 (4.44), %N 20.44 (20.57), %Cl 26.08 (25.43), %Fe 13.7 (13.38);
A 30 mL aliquot of a 0.02 M phosphate buffer (pH 7) containing
200 ␮M TrBP and 2.7 ␮M catalyst was placed in a 100-mL Erlen-
meyer flask. A 300 ␮L aliquot of 0.1 M KHSO5 was added and the
flask subjected to shaking at room temperature for 1 min, 30 min
and 24 h. After the reaction period, a 2 mL aliquot of 1 M ascorbic
acid and a 0.5 mL aliquot of 1 mM anthracene in hexane and acetone
(1:1) as an internal standard (ISTD) were then added. To adjust the
pH of the solution to 11–11.5, a 15 mL aliquot of 600 g L−1 K2CO3
was added, and 5 mL of acetic anhydride was then added dropwise
into the mixture to acetylate the phenolic hydroxyl groups in the
TrBP and the oxidation products. This solution was then extracted 3
times with 20 mL of CH2Cl2. After dehydration of the organic phase
with Na2SO4 anhydride, the CH2Cl2 was removed by flushing with
a stream of dry N2 gas at 35 ◦C and the resulting residue was re-
dissolved in 300 ␮L of CH2Cl2. A 1 ␮L aliquot of the solution was
injected into a GC-17A/QP5050 type GC/MS (Shimadzu). Separation
was accomplished with a 100% methylsiloxane capillary column
(0.25 ␮m thickness, 0.25 mm i.d. × 25m). Temperature gradient:
65 ◦C (1.5 min); 65–120 ◦C (35 ◦C min−1); 120–130 ◦C (4 ◦C min−1);
300 ◦C (10 min).
IR spectrum, ꢀ(C H) 2952 and 1420 cm−1, ꢀ(C N) 1108 cm−1
;
[FeCl2(im)4]Cl, yield 53.8%; elemental analysis, observed (calcu-
lated), %C 32.04 (35.45), %H 3.66 (3.96), %N 24.80 (20.67), %Cl 24.9
(35.45), %Fe 13.0 (11.80); IR spectrum, ꢀ(C H) 3050 cm−1, ꢀ(C N)
1261 and 1055 cm−1; [FeCl2(mepy)4]Cl, yield 7.48%; elemental anal-
ysis, observed (calculated), %C 38.90 (39.16), %H 4.78 (4.93), %N
22.79 (22.84), %Cl 21.84 (21.72), %Fe 11.40 (11.45); IR spectrum,
ꢀ(C H) 2964 and 1442 cm−1, ꢀ(C N) 1103 and 1049 cm−1; mer-
[FeCl3(tptz)], yield 75.7%; elemental analysis, observed (calculated),
%C 43.51 (43.10), %H 2.97 (2.79), %N 16.94 (16.51), %Cl 21.04 (21.03),
%Fe 11.00 (11.45); IR spectrum, ꢀ(C H) 1523 and 1468 cm−1, ꢀ(C C,
2.5. Identification of organic acids using LC/TOF-MS
A 0.1 mL aliquot of aqueous 0.1 M KHSO5 was added to an ammo-
nium formate buffer (pH 8), which contained 50 ␮M TrBP and
10 ␮M catalyst. After a 30 min reaction period, a 1 mL aliquot of the
reaction mixture was pipetted into a glass vial and 0.5 mL methanol
was then added. The organic acids in the reaction mixture were
identified and using a 1200 type LC/TOF-MS (Agilent). The sta-
tionary phase was a ZORBAX Extend-C18 column (2.1 × 100 mm,
C
N) 1523 and 1468 cm−1; mer-[FeCl3(terpy)], yield 71.8%; elemen-
tal analysis, observed (calculated), %C 45.50 (45.6), %H 2.77 (2.81),
%N 10.61 (10.63), %Cl 26.78 (26.93), %Fe 13.70 (13.40); IR spec-

102
M. Igarashi et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 100–106
Fig. 1. Structures of the synthesized complexes: mer-[FeCl3(N-methylimidazole)3] (1), [FeCl3(imidazole)4] (2), [FeCl3(methylpyrazole)4] (3), mer-[FeCl3(2,4,6-tris(2-pyridyl)-
1,3,5-triazine)] (4), mer-[FeCl3(terpyridine)](5).
1.8 ␮m particle size, Agilent). The mobile phase was a mixture
of acetonitrile/aqueous ammonium formate (10 mM), which was
changed from; 1/99 to 80/20 for 35 min and kept 80/20 for 2 min.
degradation was observed in the presence of H2O2 or KHSO5
alone. For the case of complexes with monodentate ligands mer-
[FeCl3(meim)3], [FeCl2(mepy)3]Cl and [FeCl2(im)4]Cl, the percent
TrBP degradation was less than 10%, while the iron complex with
a tridentate ligand, mer-[FeCl3(tptz)], indicated 7–8% of degrada-
tion. However, approximately 100% of the TrBP was degraded by
the mer-[FeCl3(terpy)] complex in the presence of KHSO5, although
only 5% was degraded when H2O2 was used as the oxygen donor.
Thus, only the mer-[FeCl3(terpy)]/KHSO5 system showed catalytic
activity for oxidative degradation of TrBP. The lower activity of
the complexes with monodentate ligands, mer-[FeCl3(meim)3],
[FeCl2(mepy)3]Cl and [FeCl2(im)4]Cl, can be attributed to their
instability, compared to the complexes with multidentate ligands.
Because the triazine ligand in mer-[FeCl3(tptz)] has lower elec-
tron donating characteristics, the electron density of the center
metal, Fe, would be lower than that in mer-[FeCl3(terpy)]. Thus,
KHSO5 that can act as an electrophile easily attack to the mer-
[FeCl3(terpy)]. Zucca et al. [29]. reported a Mn complex with
pyridine that had a higher catalytic activity, because of the higher
electron density in the center metal. Thus, the catalytic activity of
mer-[FeCl3(terpy)] was the highest of all the catalysts tested.
The degradation and debromination kinetics of TrBP by the mer-
[FeCl3(terpy)]/KHSO5 system were investigated (Fig. SM-1). The
percent degradation of TrBP reached 99% within 1 min, and the
percent debromination was reached a plateau at approximately
55% when the reaction period was 5 min. Thus, the oxidation
of TrBP was too rapid to permit the reaction kinetics in the
mer-[FeCl3(terpy)]/KHSO5 system to be interpreted. When TrBP
and KHSO5 were again added to the reaction mixture after a
30 min period, no degradation of TrBP and/or debromination were
observed, suggesting that the mer-[FeCl3(terpy)] catalyst had been
deactivated. Reiff and Baker, Jr. [30]. reported that terpy ligands
might be oxidized in the course of the reaction, e.g., a complex
Fe(terpyO3) with a 2,2ꢀ,2ꢀꢀ-terpyridine-l, lꢀ,lꢀꢀ-trioxide (terpyO3) lig-
and. Thus, the deactivation of the mer-[FeCl3(terpy)] catalyst after
2.6. UV–vis absorption spectra
To observe the active species of a non-heme iron complex
that was activated by an oxygen donor, 10 ␮L of 0.1 M m-
chloroperoxybenzoic acid (mCPBA) was added to a 2.5 mL of the
acetonitrile solution of 200 ␮M mer-[FeCl3(terpy)] in a quartz cell
(1 cm × 1 cm), and UV–vis absorption spectra were recorded at
160 s intervals using a Jasco V-650 iRM type spectrophotometer.
3. Results and discussion
3.1. Catalytic activities for the synthesized complexes
Table 1 shows the percent TrBP degradation for the synthe-
sized complexes using H2O2 or KHSO5 as the oxygen donor at
pH 7 after a 30 min reaction period. In control cases, no TrBP
Table 1
Influence of catalyst and oxygen donor type on the TrBP degradation. The reaction
conditions were as follows: [catalyst] 0.5 ␮M, [KHSO5] 1 mM, [TrBP] 50 ␮M, reaction
time 30 min and pH 7.
No
1
Catalysts
Oxygen donor
TrBP degradation (%)
mer-
[FeCl3(meim)3]
H2O2
KHSO5
H2O2
KHSO5
H2O2
KHSO5
H2O2
KHSO5
H2O2
KHSO5
6.61 3.26
5.51 5.52
7.67 2.68
9.18 2.30
7.07 1.34
4.40 3.88
8.42 8.42
7.21 0.30
5.29 0.12
99.5 0.2
2
3
4
5
[FeCl2(im)4]Cl
[FeCl2(mepy)4]Cl
mer-
[FeCl3(tptz)]
mer-
[FeCl3(terpy)]

M. Igarashi et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 100–106
103
Fig. 2. Influence of KHSO5 concentration on TrBP degradation and debromination.
The reaction conditions were as follows: pH 7, [TrBP] 50 ␮M, [mer-[FeCl3(terpy)]]
10 ␮M, reaction time 30 min.
Fig. 4. Influence of pH on TrBP degradation (without HA ᭿, with HAᮀ) and debromi-
nation (without HA ᭹, with HA ꢁ) in the absence and presence of HA. The reaction
conditions were as follows: [TrBP] 50 ␮M, [mer-[FeCl3(terpy)]] 0.5 ␮M, [KHSO5]
1 mM, reaction time 30 min. [HA] 50 mg L−1
.
the first reaction can be attributed to the self-oxidation of the terpy
ligand.
at pH 6–8 and the percent debromination reached 50%. How-
ever, the degradation and debromination of TrBP were remarkably
inhibited at pH 4–7, while at pH 8 the percent degradation and
debromination increased to 80 and 25%, respectively. In the Fe(III)-
ligand (Fe(III)-L)/KHSO5 system, the reactive species produced from
Fe(III)-L and KHSO5 ferryl-oxo species (Fe(IV) O) and/or sulfate
radicals (SO4•−) [31–34], that can serve as electrophiles. The pKa
values for TrBP and HA were reported to be 6.31 [35] and 3.5–5.8
[36,37], respectively. Deprotonated species such as phenolate and
carboxylate anions for TrBP and HAs can increase the reactivity
for electrophiles [38,39]. At pH 4–7, the HAs are largely present as
deprotonated species, compared to those for TrBP, because of their
pKa values. Thus, the significant inhibition of the degradation and
debromination of TrBP at pH 4–7 in the presence of HA may be
attributed to the fact that levels of electron-rich phenolate and car-
boxylate anions, which is susceptible to attach by an electrophile,
are more dominant in HA than in TrBP. The pH values of a typical
landfill have been reported to be around 8 [5]. The catalytic system
using mer-[FeCl3(terpy)]/KHSO5 showed higher degradation and
debromination efficiencies at pH 8, and this can be an advantage
for degrading bromophenols in landfill leachates.
Fig. 2 shows the influence of KHSO5 concentration on the
percent degradation and debromination of TrBP. The percent
degradation and debromination increased with increasing KHSO5
concentration and reached a plateau at concentrations above
500 ␮M. These results indicate that the mer-[FeCl3(terpy)] complex
can catalyze the oxidation of TrBP by KHSO5.
Fig.
3 shows the influence of the concentration of mer-
[FeCl3(terpy)] on the degradation and debromination of TrBP
in the presence of 1 mM KHSO5. The percent degradation of
TrBP increased linearly with increasing concentration of mer-
[FeCl3(terpy)]. However, the debromination increased from 0.01 to
0.03 ␮M of mer-[FeCl3(terpy)], while a further addition of the cata-
lyst resulted had no further effect on debromination. The Turnover
number (TON) represents one of the indices of catalytic activity and
is calculated by dividing the concentration of the degraded TrBP
or the Br− released by the catalyst. At catalyst concentrations of
0.01 ␮M, the TON values for the degradation and debromination of
TrBP were estimated to be 1890 1.4 and 4020 216, respectively.
3.2. Effect of HA on degradation and debromination of TrBP
HAs are a major component of landfill leachates, which con-
tain TrBP. Because HAs can inhibit the oxidation of TrBP in landfill
leachates [6,7,19], their influence on the degradation and debromi-
nation of TrBP was examined for a variety of pH values (Fig. 4). In
the absence of HA, approximately 100% of the TrBP was degraded
3.3. Oxidation products
To analyze the byproducts produced as the result of TrBP oxi-
dation by the mer-[FeCl3(terpy)]/KHSO5 catalytic system, GC/MS
analyses of CH2Cl2 extracts of reaction mixtures were conducted
after acetylation with acetic anhydride. GC/MS chromatograms
after reaction periods of 1 min, 30 min and 24 h are shown in Fig. 5,
and the assignments of the reaction products from mass spectra
are shown in Fig. 6 (compounds 1–4). The substrate, TrBP (Peak
1 in Fig. 5), was detected in the case of a 1 min reaction period,
while 2,6-dibromo-p-benzoquinone (DBQ) was detected as a major
product (Peak 2 in Fig. 5), and levels of DBQ isomers were also
detected (Peaks 3 and 4 in Fig. 5). The levels of DBQ decreased
with increasing reaction time, and disappeared after 24 h. However,
the formation of dimers, such as phenoxy phenol derivatives, was
not detected during the reaction, as reported for the oxidation of
2,4,6-trichlorophenol by an Fe(III)-porphyrin/KHSO5 catalytic sys-
tem [10]. Thus, it can be expected that the decrease in DBQ in
Fig. 5 is due to aromatic ring cleavage, resulting in the formation
of organic acids. Thus, an analysis by LC/TOF-MS was conducted
to identify the organic acids in the reaction mixture. As shown in
Fig. 6, some brominated organic acids are produced. These results
suggest that mineralization to CO2 occurs via ring cleavage. To elu-
cidate this, the TOC in the reaction mixture after 3 h of the reaction
Fig. 3. Influence of mer-[FeCl3(terpy)] concentration on the percent TrBP degra-
dation and debromination. The reaction conditions were as follows: pH 7, [TrBP]
50 ␮M, [KHSO5] 1 mM, reaction time 30 min.

104
M. Igarashi et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 100–106
3.4. Reactive species from mer-[FeCl3(terpy)]
Radical species, such as HO• or SO4•− might be generated by the
activation of mer-[FeCl3(terpy)] with KHSO5, and these species can
contribute to the oxidation of organic substrates [41]. Ethanol is
generally used as a scavenger of such radical species, as described
in the following reactions [43]:
HO• + CH3CH2OH → H2O + •CH(CH3)OH
−
SO4− + CH3CH2OH → HSO4 + •CH(CH3)OH
If the above reactions were to be dominant in the presence of
ethanol, the percent degradation and debromination of TrBP could
be expected to decrease. As shown in Fig. SM-2, the levels of degra-
dation and debromination of TrBP were not changed in the presence
of various concentrations of ethanol. These results show that HO•
•−
and SO4 are not involved in the oxidation of TrBP. In the Fe(III)-
porphyrin catalysts, ferryl-oxo species (Fe(IV) = O) are the major
reactive species that reproduced as the result of activation with an
oxygen donor [31–34,44,45]. Fe(IV) = O species are also detected in
non-heme enzymes [46], suggesting that mer-[FeCl3(terpy)] may
be activated to Fe(IV) = O when KHSO5 is present. In the formation of
Fe(IV) = O species, the oxygen donor (X-O-O−) can coordinate with
the center metal of the complex (L-Fe(III)) to form a peroxide inter-
mediate (L-Fe(III)-O-O-X), and the OO bond of the L-Fe(III)-O-O-X
can be cleaved, as in the pathways below [42]:
Fig. 5. GC/MS chromatograms of CH2Cl2 extracts of reaction mixtures for reaction
times of 1 min, 30 min and 24 h. The reaction conditions were as follows: [TrBP]
200 ␮M, [mer-[FeCl3(terpy)]] 2.7 ␮M, [KHSO5] 1 mM.
period was compared with that before the reaction, and the per-
cent conversion of CO2 to the degraded TrBP was estimated to
be 15%. In the FeTPPS/KHSO5 system, 20% of the degraded pen-
tachlorophenol was mineralized when cyclodextrin was added to
the reaction mixture to form the supramolecular catalyst [40],
while this is not effective in the case of bromophenol [41]. However,
for the case of mer-[FeCl3(terpy)], bromophenol was mineralized
to CO2 without the need for any additives, indicating that the
mer-[FeCl3(terpy)]/KHSO5 catalytic system is effective in the degra-
dation of TrBP.
Homolysis : L-Fe(III)-O-O-X → L-Fe(IV) = O + XO•
Heterolysis : L-Fe(III)-O-O-X → L•+-Fe(IV) = O + XO−
In the mer-[FeCl3(terpy)]/KHSO5 system, XO• (i.e., SO4•−) was
found to not participate in the reaction, because the efficiency of
degradation of TrBP was not affected in the presence of ethanol,
a radical scavenger (Fig. S2). Thus, if L-Fe(III)-O-O-X were formed,
the OO bond could be cleaved via heterolysis to form L+•-Fe(IV) = O.
Fig. 6. Organic acids detected by LC/TOF-MS and possible degradation pathways of TrBP by the mer-[FeCl3(terpy)]/KHSO5 catalytic system.

M. Igarashi et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 100–106
105
Acknowledgment
This work was supported by JSPS KAKENHI Grant Number
25241017.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molcata.2015.12.
017.
References
[1] N. Nichkova, M. Germani, M.P. Marco, J. Agric. Food Chem. 56 (2008) 29–34.
[2] C. Thomsen, E. Lundanes, G. Becher, Environ. Sci. Technol. 36 (2002)
1414–1418.
[3] C. Thomsen, E. Lundanes, G.J. Becher, J. Environ. Monit. 3 (2001) 366–370.
[4] K.I. Choi, S.H. Lee, M. Osako, Chemosphere 74 (2009) 460–466.
[5] M. Osako, Y.J. Kim, S. Sakai, Chemosphere 57 (2004) 1571–1579.
[6] Q. Zhu, Y. Mizutani, S. Maeno, M. Fukushima, Molecules 18 (2012) 5360–5372.
[7] Q. Zhu, Y. Mizutani, S. Maeno, R. Nishimoto, T. Miyamoto, M. Fukushima, J.
Environ. Sci. Health A 48 (2013) 1593–1601.
[8] M. Fukushima, Y. Mizutani, S. Maeno, Q. Zhu, H. Kuramitz, S. Nagao, Molecules
17 (2012) 48–60.
[9] Q. Zhu, S. Maeno, R. Nishimoto, T. Miyamoto, M. Fukushima, J. Mol. Catal.
A-Chem. 385 (2014) 31–37.
Fig. 7. UV–vis absorption spectra of mer-[FeCl3(terpy)] (black line) in CH3CN and
variations in spectra in the presence of mCPBA as an oxygen donor. Spectra were
recorded at intervals of 160s after adding mCPBA. The reaction conditions were as
follows: [mer-[FeCl3(terpy)]] 200 ␮M, [mCPBA] 400 ␮M. Insert figure represents the
expansion of the spectra from 600 to 500 nm.
[10] M. Fukushima, S. Shigematsu, J. Mol. Catal. A-Chem. 293 (2008) 103–109.
[11] M. Fukushima, J. Mol. Catal. A-Chem. 286 (2008) 47–54.
[12] M. Fukushima, Y. Tanabe, K. Morimoto, K. Tatsumi, Biomacromolecules 8
(2007) 386–391.
[13] M. Fukushima, K. Tatsumi, J. Mol. Catal. A-Chem. 245 (2006) 178–184.
[14] S. Rismayani, M. Fukushima, A. Sawada, H. Ichikawa, K. Tatsumi, J. Mol. Catal.
A-Chem. 217 (2004) 13–19.
[15] M. Fukushima, A. Sawada, M. Kawasaki, H. Ichikawa, K. Morimoto, K. Tatsumi,
M. Aoyama, Environ. Sci. Technol. 37 (2003) 1031–1036.
[16] M. Fukushima, H. Ichikawa, M. Kawasaki, A. Sawada, K. Morimoto, K. Tatsumi,
Environ. Sci. Technol. 37 (2003) 386–394.
[17] T. Miyamoto, Q. Zhu, M. Igarashi, R. Kodama, S. Maeno, M. Fukushima, J. Mol.
Catal. B-Enzym. 119 (2015) 64–70.
[18] T. Miyamoto, R. Nishimoto, S. Maeno, Q. Zhu, M. Fukushima, J. Mol. Catal.
B-Enzym. 99 (2014) 150–155.
[19] Q. Zhu, S. Maeno, M. Sasaki, T. Miyamoto, M. Fukushima, Appl. Catal.
B.-Environ. 163 (2015) 459–466.
[20] M. Abu-Omar, A. Loaiza, N. Hontzeas, Chem. Rev. 105 (2005) 2227–2252.
[21] G. Bilis, K.C. Christoforidis, Y. Deligiannakis, M. Louloudi, Catal. Today 157
(2010) 101–106.
[22] M. Klopstra, G. Roelfes, R. Hage, R.M. Kellog, B.L. Feringa, Eur. J. Inorg. Chem. 4
(2004) 846–856.
[23] Q. Zhang, J.D. Gorden, C.R. Goldsmith, Inorg. Chem. 52 (2013) 13546–13554.
[24] G. Bilis, P. Stathi, A. Mavrogiorgou, Y. Deligiannakis, M. Louloudi, Appl. Catal.
A-Gen. 470 (2014) 376–389.
[25] T. Dhanalakshmi, M. Bhuvaneshwari, M. Palaniandavar, J. Inorg. Biochem. 100
(2006) 1527–1534.
[26] S.A. Cotton, V. Franckevicius, J. Fawcett, Polyhedron 21 (2002) 2055–2061.
[27] R.S. Swift, Methods of Soil Analysis, part 3, Soil Science Society of America,
Madison, 1996, pp. p1018.
[28] M. Fukushima, S. Tanaka, K. Nakayasu, K. Sasaki, K. Tatsumi, Anal. Sci. 15
(1999) 185–188.
Although the direct observation of Fe(IV) = O species is difficult,
intermediates derived from L-Fe(III)-O-O-X can be detected in
absorption spectra, since they show an absorption maximum
at 450–550 nm [47–49]. The mer-[FeCl3(terpy)] complex did not
absorb in the visible-light region above 460 nm. When absorp-
tion spectra were measured with KHSO5 as an oxygen donor, no
peaks were observed in the range 450–550 nm. Because KHSO5 is
a strong oxidant, the self-degradation of mer-[FeCl3(terpy)] can be
rapid. The heterolytic cleavage of L-Fe(III)-O-O-X is a polar reac-
tion, with anionic and cationic species being formed. A highly polar
solvent such as water can enhance the rate of the reaction, and
the intermediates disappear immediately via the self-degradation.
Thus, acetonitrile, which a lower polarity than water, and mCPBA as
the X-O-O− were employed in our attempts to observe the L-Fe(III)-
O-O-X intermediate in absorption spectra. Fig. 7 shows absorption
spectra of mer-[FeCl3(terpy)] in the presence of mCPBA. After
adding mCPBA, the absorption maximum at 551 nm, which can
be assigned as (terpy)Cl2FeIII-O-O-SO3−, increased with increasin−g
reaction time, indicating the formation of (terpy)Cl2FeIII-O-O-SO3
[34,35]. These results suggest that the OO bond in the L-FeIII-O-O-
SO3− cleaves via heterolysis and L•+-Fe(IV) = O can act as the active
species. The FeIV = O species for non-heme complexes have been
detected using FT-IR spectrometry at −35 ◦C [50].
[29] P. Zucca, G. Cocco, S. Maneca, D. Steri, E. Sanjust, J. Mol. Catal. A-Chem. 394
(2014) 129–136.
6. Conclusion
[30] W.M. Reiff, W.A. Baer Jr., Inorg. Chem. 9 (1970) 570–576.
[31] C. Konstantinos, M. Louloudi, E.R. Milaeva, Y. Deligiannakis, J. Catal. 270
(2010) 153–162.
[32] E.R. Milaeva, O.A. Gerasimova, A.L. Maximov, E.A. Ivanova, E.A. Karachanov, N.
Hadjiliadis, M. Louloudi, Catal. Commun. 8 (2007) 2069–2073.
[33] W. Nam, H.J. Han, S.Y. Oh, Y.J. Lee, M.H. Choi, S.Y. Han, C. Kim, S.K. Woo, W.
Shin, J. Am. Chem. Soc. 122 (2000) 8677–8684.
[34] C. Perollier, C. Pergrale-Mejean, A.B. Sorokin, New J. Chem. 29 (2005)
1400–1403.
[35] J. Eriksson, S. Rahm, N. Green, Å. Bergman, E. Jakobsson, Chemosphere 54
(2004) 117–126.
[36] M. Fukushima, S. Tanaka, K. Hasebe, M. Taga, H. Nakamura, Anal. Chim. Acta
302 (1995) 365–373.
[37] M. Fukushima, S. Tanaka, H. Nakamura, S. Ito, Talanta 43 (1996) 383–390.
[38] A.T. Stone, Environ. Sci. Technol. 21 (1987) 979–988.
[39] D. Hatzipanayioti, G. Tzeferakos, P. Petropouleas, Chem. Phys. 345 (2008)
119–129.
[40] M. Fukushima, K. Tatsumi, Environ. Sci. Technol. 39 (2005) 9337–9342.
[41] T. Miyamoto, S. Maeno, Q. Zhu, M. Fukushima, J. Mol. Catal. B-Enzym. 110
(2014) 147–153.
[42] S. Maeno, Q. Zhu, M. Sasaki, T. Miyamoto, M. Fukushima, J. Mol. Catal.
A-Chem. 400 (2015) 56–63.
Catalytic activities of five non-heme complexes (mer-
[FeCl3(meim)3], [FeCl2(im)4]Cl, [FeCl2(mepy)]Cl, mer-[FeCl3(tptz)]
and mer-[FeCl3(terpy)]), which had been previously synthesized,
were evaluated, in terms of the degradation of TrBP in the presence
of H2O2 or KHSO5 as oxygen donors. The mer-[FeCl3(terpy)]/KHSO5
catalytic system showed a higher activity than other catalytic sys-
tems. TON values for the degradation and debromination of TrBP
by the mer-[FeCl3(terpy)]/KHSO5 catalytic system were estimated
to be 1890 and 4020, respectively. The major oxidation prod-
uct was DBQ at the initial stage of the reaction, while its level
decreased with increasing reaction time. This can be attributed to
the formation of organic acids via the cleavage of aromatic rings,
with 15% of the degraded TrBP being mineralized to CO2. In the
oxidation, KHSO5 activates iron(III) to form the peroxy-complex
of L-FeIII-O-O-X, suggesting that the reactive species for the
degradation of TrBP is L•+-FeIV = O.

106
M. Igarashi et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 100–106
[43] G.D. Fang, D.D. Dionysiou, S.R. Al-Abed, D.M. Zhou, Appl. Catal. B-Environ. 129
(2013) 325–332.
[48] K.B. Jensen, C.J. McKenzie, L.P. Nielsen, J.Z. Pedersen, H.M. Sevendsen, Chem.
Commun. (1999) 1313–1314.
[44] K.C. Christoforidis, M. Louloudi, E.R. Milaeva, Y. Deligiannakis, J. Catal. 270
(2010) 153–162.
[49] F.G. Gelalcha, G. Anilkumar, M.K. Tse, A. Brückner, M. Beller, Chem. Eur. J. 14
(2008) 7687–7698.
[45] K. Yamaguchi, Y. Watanabe, I. Morishima, J. Am. Chem. Soc. 115 (1993)
4058–4065.
[50] J.P. Bigi, W.H. Harman, B. Lassall-Kaiser, D.M. Robles, T.A. Stich, J. Yano, R.D.
Britt, C.J. Chang, J. Am. Chem. Soc. 134 (2012) 1536–1542.
[46] R.A. Leising, B.A. Brennan, L. Que, J. Am. Chem. Soc. 113 (1991) 3988–3990.
[47] G. Bilis, P. Stathi, A. Mavrogiorgou, Y. Deligiannakis, L. Louloudi, Appl. Catal.
A-Gen. 470 (2014) 376–389.