Iron(II, III)-catalyzed oxidative N-dealkylation of amines with dioxygen

Article abstract of DOI:10.1016/j.molcata.2003.11.018

A study on the oxidation of N,N-dimethylaniline and other alkylated anilines by dioxygen, air, and hydrogen peroxide catalyzed by several labile Fe complexes in acetonitrile was carried out. N,N-dimethylaniline (DMA) was oxidized to N-methylaniline and formaldehyde as well as to N-methylanilide. Formaldehyde and the excess of substrate underwent condensation reaction to produce 4,4′-methylenebis(N,N-dimethylaniline). Small quantities of N-methylformanilide and formanilide were also formed during oxidation of N,N-dimethylaniline and N-methylaniline, respectively. The system was very labile even in the presence of bpy as a ligand and it was difficult to formulate the nature of the reactive species. However, the results confirmed that Fe(III) was reduced by the substrate to Fe(II), which interacted with substrate and activated dioxygen for demethylation process.

Full text of DOI:10.1016/j.molcata.2003.11.018

Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
Iron(II, III)-catalyzed oxidative N-dealkylation of amines with dioxygen
∗
Dorota Naróg, Urszula Lechowicz, Tadeusz Pietryga, Andrzej Sobkowiak
Rzeszów University of Technology, P.O. Box 85, 35-959 Rzeszów, Poland
Received 12 November 2002; accepted 13 November 2003
Abstract
III
3+
II
2+
III
3+
II
2+
Labile iron complexes (e.g., [Fe (bpy)
2
]
, [Fe (bpy)
2
]
, [Fe (H
2
O)
6
]
, and [Fe (H
2
O)
6
]
) in base-free acetonitrile ac-
MeCN
MeCN
MeCN
MeCN
tivate dioxygen for the direct oxygenation of N,N- and N-alkylated amines to form N-dealkylated products and corresponding aldehydes.
N,N-dimethylaniline (DMA) is oxidized to N-methylaniline (MA) and formaldehyde as well as to N-methylanilide. Formaldehyde and
ꢀ
the excess of substrate undergo condensation reaction to produce 4,4 -methylenebis(N,N-dimethylaniline) (MBDME). Small amounts of
N-methylformanilide (MF) and formanilide are also formed during oxidation of N,N-dimethylaniline and N-methylaniline respectively.
Iron(III) catalysts are reduced by the substrate to iron(II), which activates dioxygen. Dioxygen activation step is preceded by the equilibrium
reaction between iron(II) catalyst and substrate.
©
2003 Elsevier B.V. All rights reserved.
Keywords: N-demethylation; N-dealkylation; Dioxygen activation; Alkylated amines; Iron complexes
1
. Introduction
dealkylation processes, dioxygen as a direct oxidant has
been little studied [5b,j,k,n].
Oxidations of amines are of considerable biological sig-
In spite of the large number of studies important uncer-
tainties over the mechanism of oxidative N-dealkylation
by cytochrome P450 still persist [8]. One mechanism sug-
gests that N-dealkylation involves a classical hydrogen
atom transfer (HAT) - hydroxyl recombination sequence.
The other one assumes that the N-dealkylation process is
initiated by abstracting an electron from the amine (single
electron transfer (SET)). Recent reports [3z,9] postulate the
direct oxygen atom transfer from metal-dioxygen adduct
to organic molecule in the concerted C–H insertion mech-
anism. It includes the involvement of dioxygen adducts
either (peroxo)diiron(III) or di(oxo)diiron(IV) as reactive
intermediates that first hydroxylate the C–H bonds ␣ to
the nitrogen affording the N-demethylated product and
corresponding aldehyde or ketone as a final product.
In our previous studies [5k] we have found that in ace-
nificance. Amines occur as amino acids, neurotransmitters
and as components of the diet. Amine moieties are common
in drugs and other xenobiotics ingested into the body [1].
N-dealkylation is a convenient method of degrading rich
molecules, often with the result that they are transformed
into more polar entities that are excreted from the body
due to enhanced solubility. Moreover, an unprecedented
mechanism of DNA alkylation damage repair by oxida-
tive demethylation process has been recently described
[
2]. Therefore, the reaction attracts a lot of attention and
several heme-proteins such as P-450 and peroxidases [3]
and non-heme iron proteins [4] like lipoxygenases have
been used as catalysts for the reaction. A number of model
systems of N-dealkylation by means of chemical [3h,j,5]
(
including ligand dealkylation in a catalyst complex [5t–z]),
II
electrochemical [6], and photochemical [7] methods have
also been studied. These model reactions are often per-
formed with various alternative oxygen sources including
iodosobenzene and hydroperoxides in the presence of ap-
propriate iron or other metal complexes. Except for ligand
tonitrile [Co (bpy)2](ClO4)2 activates dioxygen via the
reversible formation of a µ-dioxygen complex. The com-
plex dehydrogenates (oxidizes) N-methylanilines to give
formaldehyde and the demethylated aniline. The reaction
is first-order with respect to the concentrations of catalyst,
dioxygen, and substrate. When hydrogen peroxide is used
as the oxidant, N-methylamides are also present among
∗
Corresponding author. Tel.: +48-17-865-1656;
II
2+
the products. The [Co (bpy)2] /O2 complex does not re-
act however, with hydrocarbons. Even 1,4-cyclohexadiene
fax: +48-17-854-9830.
E-mail address: asobkow@prz.rzeszow.pl (A. Sobkowiak).
1
381-1169/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2003.11.018

2
6
D. Nar o´ g et al. / Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
−
in an aqueous tetramethylammonium chloride solution that
was adjusted to give a potential of 0.00 V versus SCE.
The latter was contained in a Pyrex tube with a cracked
soft-glass tip, which was placed inside a Luggin capillary
(
C–H bond energy, ꢀHDBE = 73 kcal mol 1) remains
unchanged in the reaction mixture after 72 h. The ab-
sence of any activation by [Co (bpy)2] for oxidation of
,4-cyclohexadiene by dioxygen indicates that the reactive
species formed in the system are unable to abstract even
weakly bounded hydrogen atoms. In contrast, the oxidation
of N-methylanilines (ꢀHDBE for the C–H and N–H bonds is
about 90–95 kcal mol ) appears to occur by the removal of
two hydrogen atoms in a single step. The first-order depen-
dence of the reaction rate with respect to the catalyst, dioxy-
gen, and substrate concentrations is consistent with substrate
binding to [Co (bpy)2] in an initial reversible step, which
is followed by the formation of the µ-dioxygen reaction
complex. The later facilitates the almost concerted removal
of two hydrogen atoms from each bound substrate to form
demethylated aniline, formaldehyde and water molecules.
The lack of reactivity for aniline and N-phenylaniline
is consistent with this proposition. The presence of sec-
ondary product - 4,4 -methylenebis(N,N-dimethylaniline)
MBDMA) - formed in the consecutive reaction between
N,N-dimethylaniline and formaldehyde confirms the sug-
gested dehydrogenation pathway.
Our recent studies have indicated that redox properties of
iron catalyst have crucial role in the oxidation of hydrocar-
bons by dioxygen and hydrogen peroxide. Labile iron(III)
complexes in base-free acetonitrile are reduced by cyclo-
hexene even in the presence of dioxygen [10], and also by
hydrogen peroxide [11]. Therefore, we have undertaken
the investigations on oxidation of N,N-dimethylaniline
and other alkylated anilines by dioxygen, air, and hydro-
gen peroxide catalyzed by several labile iron complexes
II
2+
1
[12]. The UV-vis spectrophotometric measurements were
made with a Hewlett-Packard Model HP-8453 diode array
rapid scan spectrophotometer.
−
1
2.2. Chemicals and regents
The reagents for the investigations and syntheses were
the highest purity commercially available and were used
without further purification. The solvent for all of the exper-
iments was Burdick and Jackson “distilled in glass” grade
acetonitrile (MeCN, 0.002% H2O). High-purity argon gas
was used to deaerate the solutions. Tetraethylammonium
perchlorate (TEAP, GFS Chemicals) was dried in vacuo over
CaSO4 for 24 h prior to use. Iron(II) and (III) perchlorates,
II
2+
II
III
ꢀ
Fe (ClO ) ·6H O (99+%) and Fe (ClO ) ·9H O (99+%)
4 2
2
4 3
2
were obtained from GFS Chemicals. The organic substances
(
ꢀ
included: 2,2 -bipyridine (bpy, 99 + %), aniline [C H NH2,
6
5
9
9
9.5 + %] N,N-dimethylaniline [DMA, C H N(CH3)2,
6
5
9.5 + %], N-methylaniline [MA, C H NH(CH3), 99 + %],
6
5
N,N-diethylaniline [DEA, C H N(C2H )2, 99 + %],
6
5
5
N-ethylaniline [EA, C H NH(C2H ), 98%], N-methylfor-
6
5
5
manilide [MF, HCON(CH3)C H , 99%], formanilide
6
5
ꢀ
HCONHC H , 99%], 4,4 -methylenebis(N,N-dimethylani-
6 5
[
line) (CH2[C H4N(CH3)2]2 98%), biphenyl (PhPh, 99+%),
6
and diphenyl diselenide [PhSeSePh 98%] were obtained
from Aldrich.
II
II
Synthesis of [Fe (MeCN)4](ClO4)2: The [Fe (MeCN)4]
III
3+
II
2+
III
3+
(ClO ) complex was prepared by multiple recrystallizations
([Fe (bpy)2]
, [Fe (bpy)2]
, [Fe (H2O) ]
,
4 2
6
MeCN
MeCN
MeCN
II
II
2+
of [Fe (H2O)6](ClO4)2 from MeCN [13,14].
[
Fe (H2O) ]
) in acetonitrile.
6
MeCN
ꢀ
II
2+
Iron(II) bis(2,2 -bipyridine): The [Fe (bpy)2]
com-
MeCN
II
plex was prepared in-situ by mixing [Fe (MeCN)4](ClO4)2
in MeCN with a stoichiometric ratio of bipyridine. The other
metal complexes were prepared in situ by mixing the cor-
responding salt with stoichiometric ratios of ligand in ace-
tonitrile.
2
. Experimental
2
.1. Equipment
The reaction products were separated and identified with
2.3. Methods
a Hewlett–Packard 4890A Series gas chromatograph (GC)
equipped with an HP-1 capillary column (cross-linked
methyl-silicone-gum phase, 12 m × 0.2 mm i.d.) and by
gas chromatography–mass spectrometry (MS) (Agilent
Technologies 6890N gas chromatograph with 5973N
mass-selective detector). A three-electrode potentiostat
The appropriate amounts of metal salt and ligand were
combined in acetonitrile followed by the addition of sub-
strate (1 M) (total volume = 5 ml). The solution was sat-
urated with dioxygen (O2, 1 atm) or air (O2, 0.2 atm) or
3
high-purity argon gas (O2, 0 atm). The reaction cell (25 cm
(
Bioanalytical Systems Model CV-50W) was used to record
vial with cut-out cap and Teflon-faced septum) had a 20-ml
headspace, which provided a reservoir to maintain a con-
stant solution concentration of dioxygen. The reactions were
allowed to proceed for 24 h with constant stirring at room
the cyclic voltammograms (CV). The experiments were
conducted in a 15 mL electrochemical cell with provision
to control the presence of dioxygen with an argon-purge
system. The working electrode was a Bioanalytical Systems
◦
temperature (23± 1 C). After the experiment samples of the
2
glassy-carbon inlay (area, 0.09 cm ), the auxiliary electrode
reaction solution were injected into a capillary-column gas
chromatograph for analysis. In some cases the reaction was
quenched with water, and the product solution was extracted
with diethyl ether. Product species were characterized by
a platinum wire (contained in a glass tube with a medium
porosity glass-frit and filled with a solution of supporting
electrolyte), and the reference electrode a Ag/AgCl wire

D. Nar o´ g et al. / Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
25
27
GC–MS. Authentic samples were always used to confirm
product identifications and to produce standard curves for
quantitative assays of the product species. Biphenyl (10 mM)
was used as an internal standard. All experiments were done
in triplicate. The presented values of concentration are the
mean values of three independent experiments.
MA
2
0
MBDMA
15
1
0
5
0
3
. Results
In acetonitrile simple iron(II) and iron(III) complexes
MF
activate dioxygen to oxygenate N,N-dimethylaniline to
yield N-methylaniline (MA), N-methylformanilide (MF),
and 4,4 -methylenebis(N,N-dimethylaniline) (MBDMA)
ꢀ
0
5
10
15
20
25
(Table 1). Under argon atmosphere only traces of prod-
(A)
Time, h
ucts were detected. The product profiles after 24 h reaction
II
2+
25
20
time indicate that the 1 mM [Fe (bpy)2]
/O2(1 atm)
MeCN
MA
combination is the most reactive - 48.8 products/catalyst
turnovers; with air (0.21 atm O2), however the reaction
rate is reduced only to 34.1 product/catalyst turnovers.
MBDMF
[
4
To calculate product/catalyst turnover, the amounts of
,4 -methylenebis(N,N-dimethylaniline), as a product of
15
ꢀ
consecutive reaction between N,N-dimethylaniline and
formaldehyde, was not taken into account]. The use of
Fe(III) instead of Fe(II) causes the reaction yield slightly to
decrease. Fig. 1A and B show typical dependences of prod-
MF
10
5
II
2+
ucts concentrations versus time for [Fe (H2O) ]
and
6
MeCN
II
2+
[Fe (bpy)2]
as catalysts, respectively. The presented
MeCN
0
plot indicates that the reaction is completed after approx-
0
5
10
15
20
25
(B)
Time, h
Table 1
Fig. 1. Dependence of products concentration on time for oxidation of
Iron-induced oxidation of 1 M N,N-dimethylaniline by dioxygen and
air to produce N-methylaniline, N-methylformanilide, and 4,4 -methyl-
enebis(N,N-dimethylaniline)
1 M N,N-dimethylaniline in acetonitrile by air catalyzed by: (A) 1 mM
ꢀ
II
2+
II
2+
[Fe (H O) ]
2
6 MeCN
2 MeCN
and (B) 1 mM [Fe (bpy) ] .
Catalyst
Oxidant
MA
MF
MBDMA
imately 10 h, the concentration profiles of N-methylaniline
and 4,4 -methylenebis(N,N-dimethylaniline) are similar,
(mM) (mM) (mM)
ꢀ
II
2+
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
mM [Fe (H2O)6]
O2
21.9
19.6
22.9
15.6
31.4
5.8
2.3
2.5
18.0
18.3
18.7
13.3
22.1
7.8
MeCN
and the final concentration of the later is slightly smaller.
The concentration of N-methylformanilide is generally
smaller than the other products. The presence of a radical
trap diphenyl diselenide (PhSeSePh, 0.1 M) in the reaction
system causes the decrease of the reaction yield, however
no coupling products with –SePh were detected.
II
2+
mM [Fe (H2O)6]
mM [FeIII(H2O)6]
mM [FeIII(H2O)6]
Air
O2
MeCN
3+
2.4
MeCN
3+
Air
2.0
MeCN
II
2+
mM [Fe (bpy)2]
O2
17.4
12.6
11.1
5.3
MeCN
II
2+
O2, Ph2Se2a
mM [Fe (bpy)2]
MeCN
II
2+
The values of initial reaction rates indicate that: (a)
mM [Fe (bpy)2]
mM [FeIII(bpy)2]
mM [FeIII(bpy)2]
Air
O2
Air
O2
Air
O2
O2
Air
O2
23.0
23.3
14.8
30.6
22.8
37.1
38.4
21.7
32.1
19.3
16.6
12.3
22.1
5.7
MeCN
II
2+
3+
the use of 1 mM [Fe (bpy)2]
instead of 1 mM
MeCN
MeCN
II
2+
3+
[Fe (H2O)6]
as catalyst causes the initial reaction
2.1
MeCN
MeCN
rate to increase (15.0 versus 9.2 and 12.3 versus 8.6 for
dioxygen and air, respectively), the effect is not so visible
when the concentration of the catalyst is equal to 16 mM;
(b) the use of air instead dioxygen causes only slight de-
crease in the initial reaction rates. The last observation is
also supported by the results presented in Table 1. The de-
crease of dioxygen concentration 4.8 times by the use of
air (1/0.21) causes the decrease of the amounts of products
after 24 h approximately 1.5 times, which also indicates that
the reaction order with respect to dioxygen is less then 1.
II
2+
6 mM [Fe (H2O)6]
2.7
MeCN
II
2+
6 mM [Fe (H2O)6]
6 mM FeIII(H2O)6]
2.4
MeCN
3+
3.2
10.8
29.4
9.2
MeCN
II
2+
6 mM [Fe (bpy)2]
3.6
MeCN
II
2+
6 mM [Fe (bpy)2]
6 mM [FeIII(bpy)2]
3.9
MeCN
3+
3.8
16.4
MeCN
Reaction time: 24 h.
a
Concentration of PhSeSePh equal to 0.1 M, no coupling products
SePh were detected.
–

2
8
D. Nar o´ g et al. / Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
Table 2
0
Iron-induced oxidation of 1 M N-methylaniline and N-ethylaniline by
I,
A
dioxygen and air to produce aniline
1
-20
Catalyst
Oxidant Aniline from
Aniline from
4
3
N-methylaniline N-ethylaniline
-40
-60
mM
5.6
mM
3.5
3.6
1.7
6.5
3.6
6.9
6.4
6.3
8.1
4.7
8.1
5.9
II
2+
1
1
1
1
1
1
1
1
1
1
1
1
mM [Fe (H2O)6]
O2
Air
O2
O2
Air
O2
O2
Air
O2
O2
Air
O2
2
MeCN
II
2+
mM [Fe (H2O)6]
7.4
MeCN
-80
mM [FeIII(H2O)6]
3+
1.4
MeCN
II
2+
-100
-120
mM [Fe (bpy)2]
8.2
MeCN
II
2+
mM [Fe (bpy)2]
11.0
13.5
15.2
12.1
8.8
MeCN
mM [FeIII(bpy)2]
3+
E vs SCE, V
0,5 0,0
MeCN
-
140
II
2+
6 mM [Fe (H2O)6]
2,0
I,
1,5
1,0
MeCN
II
2+
(a)
6 mM [Fe (H2O)6]
MeCN
6 mM [FeIII(H2O)6]
3+
MeCN
II
2+
6 mM [Fe (bpy)2]
11.5
12.0
3.5
0
A
II
2+
6 mM [Fe (bpy)2]
-
-
-
-
20
40
60
80
1
6 mM [FeIII(bpy)2]
3+
MeCN
4
Reaction time: 24 h.
2
The described iron catalyst/dioxygen systems convert
also N-methylaniline to aniline and traces of formanilide,
as well as N-ethylaniline to aniline. The yields of these
reactions however, are much smaller then in the case of
N,N-dimetylaniline (Table 2). When N,N-diethylaniline was
used as a substrate, only traces of aniline were detected
after 24 h reaction time.
3
-
-
-
100
120
140
E vs SCE, V
0,5 0,0
2,0
1,5
1,0
(
b)
Generally, in all reactions (Tables 1 and 2) the presence
of 2,2 -bipyridine as a ligand causes an increase in the prod-
ꢀ
Fig. 2. Voltammograms in acetonitrile of: (a) (1) 5 mM [FeIII(H2O)6]
3+
,
MeCN
III
ucts yield, Fe(III) is a less effective catalyst then Fe(II),
the decrease of catalyst concentration increases the prod-
uct/catalyst turnover ratio, and the reaction order with re-
spect to dioxygen is <1.
Fig. 2 illustrates the oxidation/reduction character of the
catalyst/substrate system. Cyclic-voltammetric measure-
ments indicates that the 2:1 (molar ratio) combination of
(2) 2.5 mM N,N-dimethylaniline, (3) combination of 5 mM [Fe
+
(
H2O)6]3
and 2.5 mM N,N-dimethylaniline after 2 h in argon atmo-
MeCN
II
2+
; (b) 5 mM [FeIII(bpy)2]
3+
sphere, and (4) 5 mM [Fe (H2O)6]
.5 mM N,N-dimethylaniline, (3) combination of 5 mM [FeIII(bpy)2]
and 2.5 mM N,N-dimethylaniline after 2 h in argon atmosphere, and (4)
mM [Fe (bpy)2]
, (2)
MeCN
MeCN
3+
2
MeCN
II
2+
5
.
MeCN
III
3+
[
Fe (H2O) ]
with N,N-dimethylaniline causes the ap-
6
MeCN
peaks however, are shifted towards more negative poten-
tials) and the disappearance of DMA oxidation peak. The
complete disappearance of substrate occurs in the range of
Fe(III) to DMA molar ratios from 2:1 to 1:2. Analogous be-
havior was observed when other substrates were used. The
results presented in Fig. 2 indicate that Fe(III) is reduced to
Fe(II) by methylated anilines and that there are interactions
between Fe(II) and methylated anilines or products of their
oxidation.
pearance of Fe(II) oxidation peak-curve 3 in Fig. 2a (for the
clarity only the anodic scans of the cyclic voltammograms
were shown). The peak, however is shifted towards more
negative potentials (from +1.6 to +1.4 V) in comparison to
the Fe(II) oxidation peak in the absence of DMA, which is
presented by curve 4. At the same time the disappearance of
the substrate (DMA) oxidation peak (curve 2) is observed.
III
3+
Similar effects exist when [Fe (bpy)2]
is used in
MeCN
III
3+
place of [Fe (H2O) ]
(Fig. 2b). The cyclic voltam-
mogram of the acetonitrile solution of 1:2 combination
6
To confirm the last statement the cyclic-voltammetry
of Fe(II)/DMA system was performed. After mixing the
reagents the oxidation peak of DMA is lowered (Fig. 3a,
MeCN
of Fe(II) and bpy (curve 4) shows three oxidation peaks,
II
2+
2+
II
which can be attributed to: [Fe (bpy)3]
(+1.15 V)
MeCN
II
curve 1) and oxidation peak of [Fe (H2O)6]
is slightly
MeCN
II
2+
2+
[
(
15], [Fe (bpy)2]
(+1.35 V), and [Fe (H2O) ]
shifted towards more positive potentials (Fig. 3a, curve 2).
After 2 h under Ar atmosphere the DMA oxidation peak
almost completely disappeared and Fe(II) oxidation peak is
moved towards more negative potentials (Fig. 3a, curve 3)
as observed during interaction between Fe(III) and DMA.
6
MeCN
MeCN
+1,6 V) whereas the combination of Fe(III) with bpy does
not show any anodic peaks (curve 1). The addition of DMA
III
3+
to the solution of [Fe (bpy)2]
causes the appearance
MeCN
of Fe(II) oxidations peaks (curve 3 - the second and third

D. Nar o´ g et al. / Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
29
1
,0
0
20
40
60
80
I,
A
Absorbance
0
0
,9
,8
Absorbance, = 465 nm
-
-
-
-
3
0,8
0,7
0
0
0
,6
,5
,4
2
0,3
,2
0,6
0,4
0,2
0,0
Time, h
15 20
0
0
5
10
4
-
-
-
100
120
140
1
E vs SCE, V
0,5 0,0
2
,0
1,5
1,0
,
nm
(
a)
3
00
400
500
600
700
0
I,
A
Fig. 4. The UV-Vis spectrum for the combination of 0.1 mM
N,N-dimethylaniline and 0.1 mM [Fe (H2O)6]
gon atmosphere. Inset: the dependence of absorbance at λ = 465 nm on
III
3+
-
-
-
-
20
40
60
80
4
in acetonitrile in ar-
MeCN
time.
3
8
0
0
2
3
I,
A
-
-
-
100
120
140
6
2
1
40
E vs SCE, V
0,5 0,0
20
2
,0
1,5
1,0
0
.
(
b)
-
20
Fig. 3. Voltammograms in acetonitrile of: (a) (1) 2.5 mM N,N-dimethyl-
II
2+
-40
aniline, (2) combination 5 mM [Fe (H2O)6]
and 2.5 mM N,N-
MeCN
dimethylaniline after mixing, (3) the same as (2) after 2 h in ar-
-
60
II
2+
gon atmosphere, and (4) 5 mM [Fe (H2O)6]
; (b) (1) 2.5 mM
2+
MeCN
II
N,N-dimethylaniline, (2) combination 5 mM [Fe (bpy)2]
and 2.5 mM
-80
MeCN
E vs SCE, V
N,N-dimethylaniline after mixing, (3) the same as (2) after 2 h in argon
-
100
II
2+
atmosphere, and (4) 5 mM [Fe (bpy)2]
.
MeCN
2,0
1,5
1,0
0,5
0,0
-0,5
(
A)
Again, in Fig. 3 only anodic scans were shown for clarity.
When labile Fe(II)-bpy complex is used the existence of in-
teractions between Fe(II) and DMA is not so clearly visible.
However, the decrease of the height of DMA oxidation peak
in time is observed and the height of the peak correspond-
ing to coordinately saturated Fe(II) complex (at +1.15 V, by
80
I,
A
60
40
20
0
.
II
2+
analogy to [Fe (bpy)3]
) is increased (Fig. 3b, curve 3).
MeCN
As previously, the analogous situation occurs in the range
of Fe(II) to DMA molar ratios from 2:1 to 1:2, and simi-
lar effects were observed when other methylated anilines
were used instead of DMA. These results also indicate that
interactions between Fe(II) and methylated amines occur.
-20
-
-
-
40
60
80
III
3+
E vs SCE, V
The interactions between [Fe (H2O) ]
and DMA
6
MeCN
-
100
were also investigated spectrometrically. When the com-
pounds are combined in acetonitrile in argon atmosphere
the absorption band of DMA disappears and a new band
appears at λmax = 465 nm (Fig. 4). This behavior was
observed from 2:1 up to 1:3 Fe(III)/DMA molar ratios. If
2
,0
1,5
1,0
0,5
0,0
-0,5
(
B)
Fig. 5. Voltammograms in acetonitrile of: (A) combination of 5 mM
III
3
+
[
Fe (H2O)6]
and 2.5 mM N,N-dimethylaniline; (B) combination of
MeCN
5 mM [FeIII(bpy)2]
3+
and 2.5 mM N,N-dimethylaniline; both after 2 h
MeCN
+
however, the one mole equivalent of H was added to DMA
in argon atmosphere.

3
0
D. Nar o´ g et al. / Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
the effect occurred also for higher Fe(III)/DMA ratios. The
height of the band increases initially and then either reaches
a plateau (for 2:1 Fe(III)/DMA molar ratios, Fig. 4, inset)
or decreases (for higher Fe(III)/DMA ratios). When Fe(II)
catalyst is used instead of Fe(III) there are no new peaks
present on the UV spectra. The described effect is less visi-
ble (lower absorption peak height) in the presence of dioxy-
with N,N-dimethylaniline to produce N-methylaniline and
formaldehyde in a stoichiometric reaction:
2
(bpy)2 FeIII − OH + PhN(CH3)2
→
+
2
PhNHCH3 + H2C(O) + H2O + 2(bpy)2 FeII
2+
(1)
The process occurs according to 2:1 (iron to substrate)
stoichiometry. However, the condensation reaction between
N,N-dimethylaniline (DMA) and formaldehyde to produce
III
3+
gen in the solution. The replacement of [Fe (H2O) ]
6
MeCN
by Fe(III)-bpy complex also attenuates the observed ef-
ꢀ
fect. In addition, the described absorption band slightly
4,4 -methylenebis(N,N-dimethylaniline) occurs at the com-
II
2+
overlaps with [Fe (bpy)3]
adsorption peak, which
parable rate to demethylation reaction, which follows from
the concentration profiles (Fig. 1). The occurrence of the
condensation process (Eq. (2))
MeCN
presence indicates that Fe(II) is formed in the reaction.
When the described UV-Vis absorption band is present, on
(2)
cyclic-voltammograms the reduction peak at +0.72 V (ver-
sus, SCE) - or at +0.60 V in the presence of bpy - appears
regardless of the presence of bpy in the system (Fig. 5). It is
characteristic that the height of the peak increases directly
from the rest potential, without an existence of a residual
current. If anodic scan is performed first the small, broad
anodic peak also arising from the rest potential is observed.
All applied substrates: N,N-dimethylaniline, N-methylani-
line, N,N-diethylaniline, and N-ethylaniline are electrochem-
ically oxidizable, the shapes of their CV-peaks are the same
and their oxidation potentials are very similar and equal to:
causes that the overall stoichiometry of the reaction between
Fe(III) and N,N-dimethylaniline is equal to 2:3.
Cyclic-voltammetric measurements show that for the mo-
lar combinations of Fe(III) to the substrate from 2:1 to 1:2
the oxidation peak of DMA is absent. However, the product
of the reaction - N-methylaniline - should be visible, it is also
oxidizable at +0.90 V like DMA. The absence of substrate
or products oxidation peaks after reaction indicates that the
Fe(II) formed interacts with the MA (it was confirmed by in-
dependent CV-measurements of Fe(II)/MA system), which
also causes the Fe(II) oxidation peak to be shifted towards
negative potentials. These possible interactions explain the
existence of described effects up to 1:2 molar ratio of iron
to the substrate.
+
0.90, +0.90, +0.95, and +0.80 V, respectively.
4
. Discussion
In acetonintile in the presence of a base Fe(III) can ab-
stract hydrogen atom from alkyl C–H bond [8d,10,17,18].
Therefore, in the initial step of the reaction (1) the C–H
bond of methylene group is broken with formation of water
molecule. The total bond breakage energy for reaction (1)
is the sum of C–H bond breakage in N,N-dimethylaniline
The results presented indicate that the substrate - N,N-
III
3+
III
dimethylaniline–reduces [Fe (H2O) ]
and [Fe
6
MeCN
+
(
bpy)2]3
to corresponding Fe(II) complexes. The re-
MeCN
duction of iron(III) complexes by alkylated amines was pre-
viously observed [8d] and demethylation of tertiary amines
to secondary amines by potassium hexacyanoferrate(III) in
the presence of a base is long-time known reaction [16].
−1
(H–CH2(CH3)NPh, (∆HDBE = 352 kJ mol , by anal-
ogy to H–CH2N(CH3)3 [20]) and Fe–OH bond in
(bpy)2 Fe –OH (∆HDBE = 222 kJ mol
+
III
−1
[21] to form
2
III
3+
complex and
Moreover, [Fe (1,10-phenanthroline)3]
water molecule. Because the total bond-breakage en-
−1
Fe(ClO4)3 in aprotic solvents are known as a one-electron
oxidants, which generates a series of cation-radicals form
hydrocarbons [17]. In a recent paper we have reported
that Fe(III) catalysts are reduced by cyclohexene during
its iron-induced oxygenation by dioxygen in acetonirtile
ergy (∆HDBE = 352 + 222 = 574 kJ mol ) is greater
than the free-energy of bond-formation for the HO–H
−
1
bond (-∆GBF = 465 kJ mol
[21a]), for the observed
process (Eq. (1)) to occur at reasonable rates requires
formation of an intermediate with an iron-carbon bond
2
+
III
−1
[10]. These facts indicate that in the presence of the large
{(bpy)2 Fe CH2N(CH3)Ph}; −ꢀGBF ∼ 110 kJ mol
excess of N,N-dimethylaniline (like in the reaction con-
ditions applied in the paper), iron catalysts remains in
the reduced state, even in the presence of dioxygen. On
the basis of present results and those from related stud-
ies [8c,10,11,18], and taking into account properties and
iron(III) aqua-complexes [19] (we cannot avoid traces of
water in acetonitrile) we can postulate that in the absence
[21b]). The latter will react at diffusion-controlled rates with
a second (bpy)2 Fe OH to give the observed demethyla-
+
III
2
tion products (Scheme 1).
The results obtained indicate that Fe(III) complexes
are reduced by the substrate and Fe(II) is responsible for
dioxygen activation. The CV-measurements show that the
peak of Fe(II) oxidation in the presence of DMA is shifted
towards negative potentials, which gives evidence that an
of dioxygen only (bpy)2 Fe –OH species is reactive
+
III
2

D. Nar o´ g et al. / Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
31
H2
C
bpy) 2+FeIII-OH
+
{(bpy)2
2+FeIII-CH (CH )NPh}
2 3
(
H
2
CH3
N
H O
2
(
bpy)22+FeIII-OH
H
CH3
N
O
II
2+
+
+
2Fe (bpy)
2
H
H
Scheme 1. Proposed mechanism for N,N-dimethylaniline oxidation by iron(III) in the absence of dioxygen.
interaction between the iron catalyst and substrate occurs.
The described in [15b] process of ligand exchange in the
Fe/bpy system in acetonitrile supports the assumption that
Fe(II) interacts with the substrate even in the presence of
bpy. In experimental part the evidences were shown that the
reaction order with respect to dioxygen is less then 1. This
situation occurs when an equilibrium step between substrate
and catalyst precedes the addition of dioxygen [22].
The proposed intermediate can: (a) reacts with another
substrate to form demethylated product, formaldehyde, and
Fe(II) (path A); (b) undergoes decomposition to formanilide
derivative, water, and the Fe(II) catalyst (path B); (c) re-
acts with iron(II) catalyst to produce demethylated product,
formaldehyde, and ␣-oxo diiron(III) species, which often is
the final products of Fe(II) oxidation (path C).
The mechanism is consistent with the results obtained. It
is also coherent with the recently introduced proposition of
the direct oxygen atom transfer from metal-dioxygen adduct
to the ␣ C–H bond in amine molecule in the concerted in-
sertion mechanism [5z]. The mechanism elucidates the ob-
served fact that the amounts of N-methylformanilide formed
is always much smaller than N-methylaniline - the large con-
centration of substrate favors the path A. The effect however,
Based on these facts and previously published results
[10,11,18] the proposed putative mechanism, which includes
the activation of dioxygen and the substrate, is outlined in
2
+
IV
Scheme 2. The reactive species {L2 Fe (OOH)(CH2(CH3)
NPh} is a hypothetical one and was proposed on the ba-
sis of the results obtained for metal-induced oxygenation of
organic substrates in basic, nonaqueous solutions [10,18].
Similar reactive intermediate was postulated in the models of
hydrogen peroxide activation [23]. Moreover, several anal-
ogous hydroperoxo complexes of iron have been recently
isolated and characterized [24].
II
2+
is less visible for the 1 mM [Fe (bpy)2]
/dioxygen or
MeCN
air systems. Clearly the presence of bpy as a ligand at low
catalyst concentrations causes the increase of the efficiency
of path B. At the higher catalyst concentration the decrease
H2
C
K
L22+FeII
+
H
{L22+Fe (CH ) NPh}
II
CH3
3 2
N
O2
OOH
L22+FeIV
CH (CH )NPh
2
3
(Path A)
(Path B)
(Path C)
PhN(CH3)2
L22+FeII
2
PhNH(CH ) + 2H C(O)
PhN(CH )CH(O)
2PhNH(CH ) + 2H C(O)
3 2
3
2
3
+
L22+FeII
+ H O
+ L22+FeIIIOFeIIIL22+
2
+
L22+FeII
Scheme 2. Proposed mechanism for iron(II)-activation of dioxygen for oxygenation of N,N-dimethylaniline.

3
2
D. Nar o´ g et al. / Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
of the reaction yield is observed - in this case path C is more
efficient and it leads to catalyst decomposition.
References
The interesting fact described in the paper is the existence
a chromophore at λmax = 465 nm, which is formed only
with Fe(III) [not Fe(II)] and DMA, when the concentration
the last one not exceed 3 molar equivalents with respect to
Fe(III). The species cannot be attributed to a Fe(II)/DMA
complex because the oxidation peak of Fe(II)/DMA formed
after mixing Fe(III) with DMA (Fig. 2a, curve 3) is present
also when the chromophore is not visible. The observed ab-
sorption band can be attributed to a µ-oxo dimer having
[1] F.P. Guengerich, O. Okazaki, Y. Seto, T.L. MacDonald, Xenobiotica
25 (1995) 689.
[2] S.C. Trewick, T.F. Henshaw, R.P. Hausinger, T. Lindhal, B. Sedgwick,
Nature 419 (2002) 174.
[3] (a) M. Sono, M.P. Roach, E.D. Coulter, J.H. Dawson, Chem. Rev.
96 (1996) 2841;
(b) G.T. Miwa, J.S. Walsh, G.L. Kedderis, P.F. Hollenberg, J. Biol.
Chem. 258 (1983) 14445;
(c) G.L. Kedderis, P.F. Hollenberg, J. Biol. Chem. 258 (1983) 8129;
(d) G.L. Kedderis, P.F. Hollenberg, J. Biol. Chem. 259 (1984) 3663;
III
III
4+
(e) G.L. Kedderis, P.F. Hollenberg, Biochemistry 24 (1985) 6158;
a structure [LxFe OFe Lx] , where L represents DMA
and/or bpy and a solvent (H2O or MeCN). It is known that
Fe(III) in the presence of traces of water shows the tendency
to hydrolyze and/or to form complexes [19,25]. One of the
major species formed is the µ-oxo dimer. It can also activate
dioxygen, however its catalytic activity is lower then that
for monomeric species [25c]. The excess of DMA as a base
and also complexing agent can prevent the formation of the
products of hydrolysis, therefore the chromophore is not vis-
ible in the condition. However, in the presence of equimolar
(
(
f) F.P. Guengerich, C.H. Yun, T.L. Macdonald, J. Biol. Chem. 271
1996) 27321;
(g) J.I. Manchester, J.P. Dinnocenzo, L.A. Higgins, J.P. Jones, J. Am.
Chem. Soc. 119 (1997) 5069;
(
h) Y. Goto, Y. Watanabe, S. Fukuzumi, J.P. Jones, J.P. Dinnocenzo,
J. Am. Chem. Soc. 120 (1998) 10762;
i) E. Baciocchi, M.F. Gerini, O. Lanzalunga, A. Lapi, S. Mancinelli,
(
P. Mencarelli, J. Chem. Soc., Chem. Commun. (2000) 393;
(j) E. Baciocchi, M.F. Gerini, O. Lanzalunga, A. Lapi, M.G. Lo
Piparo, S. Mancinelli, Eur. J. Org. Chem. (2001) 2305.;
(
k) C.L. Shaffer, M.D. Morton, R.P. Hanzlik, J. Am. Chem. Soc.
123 (2001) 8502;
l) M.N. Bhakta, K. Wimalasena, J. Am. Chem. Soc. 124 (2002)
844;
+
amounts of H with respect to DMA the same chromophore
(
1
is visible for higher Fe(III)/DMA molar ratios. This is prob-
ably due to the fact that the presence of protons in the solu-
tion can promote ligands exchange, as described in [15b] and
favor the formation of the species. The species can undergo
additional reactions, which includes further hydrolysis [26],
ligand exchange, and reduction by DMA, which leads to its
disappearance. The cathodic peak observed on the CV-curve
at +0.72 V (or at +0.60 V in the presence of bpy) is visi-
ble only when the discussed absorption band is present can
be attributed to the reduction of the µ-oxo dimer. The ir-
(m) C.L. Shaffer, S. Harriman, Y.M. Koen, R.P. Hanzlik, J. Am.
Chem. Soc. 124 (2002) 8268.
[4] (a) L. Que Jr., R.Y.N. Ho, Chem. Rev. 96 (1996) 2607;
b) G.G. Hover, A.P. Kulkarni, Chem. Biol. Interact. 124 (2000) 191.
5] (a) P. Shannon, T. Bruice, J. Am. Chem. Soc. 103 (1981) 4580;
b) F. Benedini, G. Galliani, M. Mali, B. Rindone, S. Tollari, J.
Chem. Soc. Perkin. Trans. II (1985) 1963;
c) C.M. Dicken, F.L. Lu, M.W. Nee, T.C. Bruice, J. Am. Chem.
Soc. 107 (1985) 5776;
d) J.R. Lindsay Smith, D.N. Mortimer, J. Chem. Soc., Chem. Com-
mun. (1985) 64;
e) K. Fujimori, S. Fujiwara, T. Takata, S. Oae, Tetrahedron Lett.
7 (1986) 581;
f) J.R. Lindsay Smith, D.N. Mortimer, J. Chem. Soc., Perkin Trans.
II (1986) 1743;
g) E.I. Karasevich, A.M. Khenkin, A.E. Shilov, J. Chem. Soc.,
Chem. Comm. (1987) 731;
(
[
(
(
(
III,III
4+
reversible electrochemical reduction of [Fe2
OL4S2]
(
2
(
dimer [L = bpy or (-)4,5-pinene-bpy; S = H2O or MeCN] is
reported [15b]. It is interesting that the reduction peak starts
directly from the rest potential, and also the small anodic
peak arises directly from the rest potential. Probably in the
experimental conditions the formation of a small amounts of
(
(
(
(
h) D. Ostrovic, C.B. Knobler, T.C. Bruice, J. Am. Chem. Soc. 109
1987) 3444;
i) S. Murahashi, T. Maota, K. Yonemura, J. Am. Chem. Soc. 110
II
III
3+
mixed valence µ-oxo complex [LxFe OFe Lx] , is pos-
sible as well, and its oxidation is observed. The possibil-
ity of formation of such complexes was also reported in
(1988) 8256;
j) S. Murata, M. Miura, M.J. Nomura, J. Chem. Soc., Chem. Com-
mun. (1989) 116.;
(
[15b].
The described results show that the system is very labile
(
(
k) A. Sobkowiak, D.T. Sawyer, J. Am. Chem. Soc. 113 (1991) 9520;
l) S. Murata, K. Suzuki, A. Tamatani, M. Miura, M. Nomura, J.
even in the presence of bpy as a ligand and it is difficult to
formulate the nature of the reactive species. However, the
results confirm that Fe(III) is reduced by the substrate to
Fe(II), which interacts with substrate and activates dioxygen
for demethylation process.
Chem. Soc., Perkin Trans. I (1992) 1387;
(m) S. Murahashi, T. Naota, N. Miyaguchi, T. Nakato, Tetrahedron
Lett. 33 (1992) 6991;
(
(
(
n) J.P. Hage, J.R. Schnelten, D.T. Sawyer, Bioorg. Med. Chem. 4
1996) 821;
o) E. Kotani, T. Takeya, H. Shimizu, H. Egawa, T. Yamamoto, S.
Tobinaga, Chem. Pharm. Bull. 45 (1997) 2089;
(p) E. Baciocchi, O. Lanzalunga, A. Lapi, L. Manduchi, J. Am.
Chem. Soc. 120 (1998) 5783;
Acknowledgements
(
(
(
(
q) E. Baciocchi, A. Lapi, Tetrahedron Lett. 40 (1999) 5425;
r) K. Krohn, J. Küpke, J. Prakt. Chem. 341 (1999) 509;
s) S. Fukuzumi, J. Porphyrins Phthalocyanines 4 (2000) 398;
t) S. Mahapatra, J.A. Halfen, W.B. Tolman, J. Am. Chem. Soc. 118
This work was supported by the Rzeszów University of
Technology internal grant U-5905/BW and by the State
Committee for Scientific Research (Poland) within COST
Action D21.
(1996) 11575;

D. Nar o´ g et al. / Journal of Molecular Catalysis A: Chemical 212 (2004) 25–33
33
(
u) V. Mahadevan, Z. Hou, A.P. Cole, D.E. Root, T.K. Lal, E.I.
[18] D.T. Sawyer, A. Sobkowiak, T. Matsushita, Acc. Chem. Res. 29
(1996) 409.
Solomon, T.D.P. Stack, J. Am. Chem. Soc. 119 (1997) 11996;
(
(
v) D. Lee, S.J. Lippard, J. Am. Chem. Soc. 120 (1998) 12153;
w) D. Lee, j. Du Bois, D. Petasis, M.P. Hendrich, C. Krebs, B.H.
[19] (a) F.A. Cotton, G. Wilkinson, Advanced Inorganic Chemistry, Wiley,
New York, fifth ed., 1988, p. 717;
Huynh, S.J. Lippard, J. Am. Chem. Soc. 121 (1999) 9893;
(b) D.T. Richens, The Chemistry of Aqua Ions, Wiley, New York,
1997, pp. 369–385.
(
(
(
x) D. Lee, S.J. Lippard, J. Am. Chem. Soc. 123 (2001) 4611;
y) D. Lee, S.J. Lippard, Inorg. Chem. 41 (2002) 827;
z) M. Taki, S. Teramae, S. Nagatomo, Y. Tachi, T. Kitagawa, S.
[20] D.R. Lide (Ed.), CRC Handbook of Chemistry and Physics, 73rd
ed., CRC Press, Boca Raton, FL, 1992, pp. 9–139.
[21] (a) D.T. Sawyer, Inorg. Chim. Acta 226 (1994) 99;
(b) A. Qiu, D.T. Sawyer, J. Porphyrins Phthalocyanines 1 (1997)
125.
Itoh, S. Fukuzumi, J. Am. Chem. Soc. 124 (2002) 6367.
6] L.R. Hall, R.T. Iwamoto, R.P. Hanzlik, J. Org. Chem. 54 (1989)
[
2
446.
[
7] (a) J. Santamaria, R. Ouchabane, J. Rigaudy, Tetrahedron Lett. 30
[22] (a) A.L. Feig, M. Becker, S. Schindler, R. van Eldik, S.J. Lippard,
Inorg. Chem. 35 (1996) 2590;
(
(
(
1989) 3977;
b) H. Gan, D.G. Whitten, J. Am. Chem. Soc. 113 (1991) 9409;
c) T. Zhang, T. Oyama, A. Aoshima, H. Hidaka, J. Zhao, N. Serpone,
(b) A.L. Feig, A. Masschelein, A. Bakac, S.J. Lippard, J. Am. Chem.
Soc. 119 (1997) 334;
J. Phtochem. Photobiol. 140 (2001) 163.
8] (a) L.R. Hall, R.P. Hanzlik, J. Biol. Chem. 265 (1990) 12349;
(c) L.I. Simándi, T.L. Simándi, J. Mol. Catal. A Chem. 117 (1997)
299;
[
(
(
b) O. Okazaki, F.P. Guengerich, J. Biol. Chem. 268 (1993) 1546;
c) S.B. Karki, J.P. Dinnocenzo, J.P. Jones, K.R. Korzekwa, J. Am.
(d) S.V. Kryatov, E.V. Rybak-Akimova, V.L. MacMurdo, L. Que Jr.,
Inorg. Chem. 40 (2001) 2220.
Chem. Soc. 117 (1995) 3657;
d) H. Chen, M.J. de Groot, N.P.E. Vermeulen, R.P. Hanzlik, J. Org.
Chem. 62 (1997) 8227.
[23] (a) S. Nashino, H. Hosomi, S. Ohba, H. Matsushima, T. Tokii, Y.
Nishida, J. Chem. Soc., Dalton Trans. (1999) 1509;
(b) K. Chen, L. Que Jr., Angew. Chem. Int. Ed. 38 (1999) 2227;
(c) S.J. Lange, H. Miyake, L. Que Jr., J. Am. Chem. Soc. 121
(1999) 6330;
(
[
9] M. Taki, S. Itoh, S. Fukuzumi, J. Am. Chem. Soc. 124 (2002) 998.
10] A. Sobkowiak, D. Naróg, D.T. Sawyer, J. Mol. Catal. A Chem. 159
[
(
2000) 247.
(d) H. Miyake, K. Chen, S.J. Lange, L. Que Jr., Inorg. Chem. 40
(2001) 3534;
(e) K. Chen, L. Que, J. Am. Chem. Soc. 123 (2001) 6327.
[24] (a) C. Kim, K. Chen, J. Kim, L. Que Jr., J. Am. Chem. Soc. 119
(1997) 5964;
[
[
11] T. Paczesniak, A. Sobkowiak, J. Mol. Catal. A Chem. 194 (2003) 1.
12] D.T. Sawyer, A. Sobkowiak, J.L. Roberts, Jr., Electrochemistry for
Chemists, second ed., Wiley, New York, 1995, p. 189.
13] H. Sugimoto, D.T. Sawyer, J Am. Chem. Soc. 107 (1985) 5712.
14] C. Sheu, S.A. Richert, P. Cofré, B. Ross, A. Sobkowiak, D.T. Sawyer,
J.R. Kanofsky, J. Am. Chem. Soc. 112 (1990) 1936.
[
[
(b) R.Y.N. Ho, G. Roelfes, B.L. Feringa, L. Que Jr., J. Am. Chem.
Soc. 121 (1999) 264;
[
15] (a) S.A. Richert, P.K.S. Tsang, D.T. Sawyer, Inorg. Chem. 28 (1989)
(c) K. Chen, L. Que, J. Chem. Soc., Chem. Commum. (1999)
1375;
2
471;
(
b) M.-N. Collomb Dunand-Santhier, A. Deronzier, C. Duboc Toia,
(d) G. Roelfes, M. Lubben, K. Chen, R.Y.N. Ho, A. Meetsma, S.
Genseberger, R.M. Hermant, R. Hage, S.K. Mandal, V.G. Young Jr.,
Y. Zang, H. Kooijman, A.L. Spek, L. Que Jr., B.L. Feringa, Inorg.
Chem. 38 (1999) 1929.
M. Fontecave, K. Gorgy, J.-C. Lepr eˆ te, S. Ménage, J. Electroanal.
Chem. 469 (1999) 53.
[
16] M. Hudlicky, Oxidations in Organic Chemistry, ACS Monograph
1
86, American Chemical Society, Washington, DC, 1990, p. 37.
[25] (a) D.M. Kurtz Jr., Chem. Rev. 90 (1990) 585;
(b) V.S. Kulikowa, O.N. Gritsenko, A.A. Shteinman, Mendeleev
Commun. (1996) 119;
(c) E.I. Karasevich, V.S. Kulikowa, A.E. Shilov, A.A. Shteinman,
Russ. Chem. Rev. 67 (1998) 335.
[26] (a) R.K. Murmann, E.A. Healy, J. Am. Chem. Soc. 83 (1961) 2092;
(b) W.M. Reiff, W.A. Baker Jr., N.E. Erickson, J. Am. Chem. Soc.
90 (1968) 4794;
[
17] (a) S. Fukuzumi, C.L. Wong, J.K. Kochi, J. Am. Chem. Soc. 102
(
(
(
(
(
1980) 2928;
b) C.J. Schlesener, J.K. Kochi, J. Org. Chem. 49 (1984) 3142;
c) C.J. Schlesener, C. Amatore, J.K. Kochi, J. Phys. Chem. 90
1986) 3141;
d) D.D.M. Wayner, D.J. McPhee, D. Griller, J. Am. Chem. Soc.
1
(
8
(
10 (1988) 132;
e) S. Fukuzumi, Y. Tokuda, M. Fujita, J. Phys. Chem. 96 (1992)
413;
f) S. Fukuzumi, Y. Tokuda, T. Etano, T. Okamoto, J. Otera, J. Am.
Chem. Soc. 115 (1993) 8960.
(c) D.L. Ehman, D.T. Sawyer, Inorg. Chem. 8 (1969) 900;
(d) P.G. David, P.C. de Mello, Inorg. Chem. 12 (1973) 2188;
(e) P.S. Braterman, J-I. Song, R.D. Peacock, Inorg. Chem. 31 (1992)
555.