C-H benzylic oxidation promoted by dinuclear iron DBDOC iminopyridine complexes

Article abstract of DOI:10.1016/j.ica.2014.12.016

Several benzofurobenzofuran and methanodibenzodioxocine iminopyridine derivatives have been used as ligands to form mononuclear and dinuclear iron complexes. Complexes 6, 7, 8, 9 and 10 were able to promote the catalytic oxidation of benzylic C-H bonds to ketones in moderate to high yields. The effects of backbone scaffold, nuclearity (mononuclear versus dinuclear) and nitrogen hybridization (iminopyridine versus aminopyridine) were studied. A strong effect on the yields of the nature and position of the substituents in the substrates was observed.

Full text of DOI:10.1016/j.ica.2014.12.016

Inorganica Chimica Acta 431 (2015) 156–160
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
C–H benzylic oxidation promoted by dinuclear iron DBDOC
iminopyridine complexes
b
Oriol Martínez-Ferraté a,b, George J.P. Britovsek c, Carmen Claver a, , Piet W.N.M. van Leeuwen
⇑
a Departament de Química Física i Inorgànica, Universitat Rovira i Virgili, C/Marcel.li Domingo s/n, 43007 Tarragona, Spain
b Institute of Chemical Research of Catalonia (ICIQ), Avgda. Països Catalans 16, Tarragona 4307, Spain
c Department of Chemistry, Imperial College London, Exhibition Road, South Kensington, SW7 2AY London, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Several benzofurobenzofuran and methanodibenzodioxocine iminopyridine derivatives have been used
as ligands to form mononuclear and dinuclear iron complexes. Complexes 6, 7, 8, 9 and 10 were able
to promote the catalytic oxidation of benzylic C–H bonds to ketones in moderate to high yields. The
effects of backbone scaffold, nuclearity (mononuclear versus dinuclear) and nitrogen hybridization
(iminopyridine versus aminopyridine) were studied. A strong effect on the yields of the nature and
position of the substituents in the substrates was observed.
Received 15 December 2014
Accepted 22 December 2014
Available online 7 January 2015
Keywords:
Dinuclear complexes
Fe(II) complex
C–H oxidation
DBDOC
Ó 2015 Elsevier B.V. All rights reserved.
Iminopyridines
1. Introduction
The first example of a stereospecific alkane hydroxylation pro-
moted by an iron catalyst was reported by Que and co-workers
The selective oxidation of saturated alkanes is a challenging
reaction due to the high strength of C–H bond. This transformation
can be promoted by homogeneous catalysts [1]. The reaction is of
great interest for industry as a means to directly functionalize
saturated hydrocarbons [2].
[12]. The reaction was catalyzed by a tris(pyridin-2-ylmethyl)
amine (TPA) iron complex. Also in 1997, Nishida and co-workers
reported
N1,N2-dimethyl-N1,N2-bis(pyridin-2-ylmethyl)ethane-
1,2-diamine (BPMEN) as a convenient ligand system for iron pro-
moted oxidations [13]. Modification of BPMEN resulted in interest-
ing ligands for iron catalyzed C–H bond oxidations [14–16]. The
two main structures of tetradentate ligands in iron catalyzed oxi-
dation of alkanes are derived from TPA and BPMEN, having tripodal
and linear structures, respectively [17]. 1,4-Dimethyl-7-(pyridin-2-
ylmethyl)-1,4,7-triazonane (TACN) resulted in an interesting tetra-
dentate N-donor ligand for the oxidation of alkanes [18,19]. These
non-heme ligands are summarized in Fig. 1.
Recently, Bauer and co-workers, reported the use of iminopyri-
dine iron complexes in the oxidation of activated C–H bonds to
ketones under mild conditions [23,24]. With this precedent in
mind, we considered of interest the application of Fe complexes
with large bite angle N-ligands able to form dinuclear complexes
as catalyst in the oxidation of C–H aiming at a possible cooperative
effect of the two metals.
At the end of nineteenth century, Fenton reported an oxidising
system based on Fe(II) complexes and hydrogen peroxide [3]. The
application of this system allows the oxidation of C–H bonds via
the generation of hydroxyl radicals and hence the reaction pro-
ceeds with low selectivity and functional group tolerance [4,5].
On the other hand it is well-known that biological systems per-
form selective oxidation of alkanes under mild conditions. For
instance Methane Monooxygenase (MMO) oxidizes methane to
methanol [6]. The aliphatic oxidation is catalyzed by two types of
enzymes named heme and non-heme proteins [6,7]. These pro-
teins were used as models to develop new iron-based catalytic sys-
tems that promoted the oxidation of aliphatic compounds [8–10].
The main drawback in the development of biomimetic ligands
is the formation of radicals. Thus the ligand design focused on sys-
tems that should be capable to oxidize alkanes selectively avoiding
the Fenton type oxidations. Non-heme ligands are more attractive
ligands due to their tunability compared to heme ligands [8,9,11].
2. Results and discussion
Our group previously synthesized several imino- and aminopyr-
idine DBDOC and BFBF compounds [25] which were used as
ligands in the synthesis of iron mononuclear and dinuclear
complexes. Reaction is summarized in Fig. 2.
⇑
Corresponding author.
E-mail address: carmen.claver@urv.cat (C. Claver).
http://dx.doi.org/10.1016/j.ica.2014.12.016
0020-1693/Ó 2015 Elsevier B.V. All rights reserved.

O. Martínez-Ferraté et al. / Inorganica Chimica Acta 431 (2015) 156–160
157
Table 1
formed when the present bidentate imine ligands coordinate to
iron centres [23]. Complex 6 present the pyridine donors trans to
one another.
Selected angles and distances for complex 6.
Bond distance (Å)
Fe1–N1
Fe1–N2
Fe1–N3
1.965(3)
1.956(3)
1.961(3)
Fe1–N4
Fe1–N5
Fe1–N6
1.965(2)
1.948(3)
1.946(3)
The different complexes were used as catalysts in the oxidation
of diphenylmethane 11. The reaction was optimized using catalyst
6. Several parameters were shown to influence the reaction and the
results of the optimization are summarized in Table 2. First the
influence of oxidant loadings was explored. Large excesses of
oxidant can contribute to faster catalyst decomposition; it was
recently found that amide was formed by oxidation of the imino-
pyridine, but the amide complex was still active in catalysis [24].
A positive linear relationship between yield and quantity of
oxidant was observed (entries 1–3). The usage of an internal stan-
dard led to depleted yields and thus an external standard was
found to be required. It is suggested that changes in reaction media
such as polarity or solubility causing inclusion of the internal
standard were responsible for the decreased yields. Due to the
decomposition of the metal centres by peroxides slow addition of
the peroxide to the system was required [4,5,27,28]. In our case,
when the total amount of peroxide was added at the beginning
of the reaction (entry 3 versus 5) the conversion decreased and
gas evolution was observed. Only pyridine and acetonitrile were
tested as solvents (entries 3 and 6). Pyridine proved to be a more
suitable solvent but since pyridine can readily coordinate to Fe
and act as a good ligand, we decided to use acetonitrile as
solvent. From the literature it is known that benzophenone can
be prepared with low yields (8%) via a non-metal catalyzed
reaction [24].
Angle (°)
N6–Fe1–N5
N5–Fe1–N2
N6–Fe1–N3
N2–Fe1–N3
N6–Fe1–N1
N5–Fe1–N1
85.94(14)
86.87(13)
90.65(13)
96.91(13)
93.99(13)
90.88(13)
N2–Fe1–N1
N3–Fe1–N1
N6–Fe1–N4
N5–Fe1–N4
N2–Fe1–N4
N3–Fe1–N4
81.67(13)
94.55(13)
91.21(12)
93.57(12)
93.67(11)
81.28(11)
Crystals of complex 6 were grown from CH3CN:Et2O (Fig. 3).
Complex 6 is composed by two Fe nuclei and two ligands 1 and
is a dinuclear complex and present a long distance between both
Fe nuclei (over 9 Å).
The iron centres showed an octahedral geometry with angles
around 90° and short N–Fe distances, shorter than 2 Å (see Table 1),
indicative of low spin complexes [26]. Several isomers could be
N
N
N
N
N
N
N
N
The reaction evolved quickly in the first 2 h and a chemical yield
of 52% was determined. Later the reaction rate decreased and at 4
and 6 h only 68% and 73% yield were obtained. Despite this
decrease in activity, the catalyst was still active, as after 20 h yield
the reaction had progressed further to 82%. The concentration of
tBuOOH decreased after 4 h due to decomposition, lowering the
conversion and the reaction had almost stopped.
Finally different oxidants were tested in the iron catalyzed oxi-
dation of diphenylmethane to benzophenone (Fig. 4). The best oxi-
dant for this reaction was found to be tert-butyl hydroperoxide 13
which gave 73% yield after 6 h. Cumene hydroperoxide 14 was
found to be a less suitable oxidant as only 15% yield was obtained.
Surprisingly, hydrogen peroxide 15 gave the lowest yield of all
tested oxidants, only 10%, which is most likely due to fast decom-
position of hydrogen peroxide by the metal catalyst [4,5,27,28].
meta-Chloroperbenzoic acid (m-CPBA, 16) gave yields of up to
TPA
BPMEN
N
N
N
N
N
N
N
N
PDP
TACN
Fig. 1. Non-heme ligands used in C–H iron oxidations. A revolution in the iron
catalyzed alkane oxidation was the work reported in 2007 by Chen and White [20].
An iron complex modified with (2S,20S)-1,10-bis(pyridin-2-ylmethyl)-2,20-bipyrr-
olidine (PDP) gave rise to a selective catalyst for the hydroxylation of secondary and
tertiary C–H bonds. Substrates can be transformed with predictable selectivity
based on steric and electronic factors [20–22].
[Fe2L2(CH3CN)4](BF4)4
or
n=1
n=2
Fe(BF4)2.6H2O
+
n L
THF, r.t, 3 h
[FeL2(CH3CN)2](BF4)2
n=1, 2
L=
( )m
N
N
N
N
N
N
N
O
O
O
O
O
O
H
H
N
N
N
N
N
4
5
3
1 m=1
2 m=0
Dinuclear
[Fe2L2(CH3CN)4](BF4)4
Mononuclear
[FeL2(CH3CN)2](BF4)2
6
7
8
L=1
9
L=4
2
10
5
L=
L=
L=3
Fig. 2. Synthesis of mononuclear and dinuclear complexes with ligands 1–5 and catalysts 6–10.

158
O. Martínez-Ferraté et al. / Inorganica Chimica Acta 431 (2015) 156–160
Oxidant scope
80
70
60
50
40
30
20
10
0
OOH
13
OOH
14
H2O2
15
OOH
Cl
O
13
14
15
16
Oxidant
16
Fig. 4. Oxidant scope in the catalytic oxidation of 11. Reaction conditions: 0.3 mmol
of 11, 0.015 eq. of 6, in 0.5 mL of ACN, 6 h and r.t. Slow addition of oxidant, over
30 min, every 5 min. Chemical yield determined by GC using o-dichlorobenzene as
external standard.
O
11
12
[Fe]
+
O
HOO
13
6h, r.t.
ACN
R
R
Fig. 3. ORTEP-plots (thermal ellipsoids shown at 50% probability levels) of complex
6. Non-relevant hydrogen atoms and tetrafluoroborate counterions have been
omitted for the sake of clarity.
17
21
R=H
[Fe]
6, 7, 8, 9, 10
R=H
4-CF3 18
4-OMe 19
4-CF3 22
4-OMe 23
Table 2
Diphenylmethane oxidation: optimization of reaction conditions.
20
24
2-OMe
2-OMe
O
Scheme 1. Substrate and catalyst scope in iron catalyzed oxidation of benzylic C–H
bonds.
6
+
oxidant
S, r.t.
11
12
Table 3
Diphenylmethane oxidation with iron complexes.
Entry
Solvent
Oxidant eq.
Yielda (%)
Entry
Catalyst
Chemical yielda (%)
1
2
3
ACN
ACN
ACN
ACN
ACN
Py
4
6
8
8
8
8
30
49
73
57
65
80
1
2
3
4
5
6
7
8
9
10
73
50
20
41
49
4b
5c
6
0.3 mmol of 11, 0.015 eq. of 6, oxidant tBuOOH in 0.5 mL of solvent, 6 h and r.t. Slow
0.3 mmol of 11, 0.015 eq. of 6, oxidant tBuOOH in 0.5 mL of ACN, 6 h and r.t. Slow
addition of oxidant over 30 min, every 5 min.
addition of oxidant over 30 min, every 5 min.
a
a
Chemical yield determined by GC using o-dichlorobenzene as external
Chemical yield determined by GC using o-dichlorobenzene as external
standard.
standard.
b
Internal standard.
Fast addition of oxidant.
c
lower conversions (entry 2). The complex with aminopyridine 8
gave lower yields, 20%, probably due to the nature of the ligand.
Monomeric species (entries 4 and 5) were found to perform
similarly.
The conversion was affected by the electronic properties of the
N-donor. For example, complex 8 containing the amino derivative
was the worst catalyst for this transformation with only 20% yield
(entry 3). It is known from literature that secondary amines can be
oxidized in the presence of iron and oxidant [29,30], thus they are
less stable than imine complexes and the conversion decreased for
these ligands.
34% and this moderate conversion was attributed to the low solu-
bility of the oxidant in the reaction media (410 mg in 0.5 mL of
solvent).
Several substrates were screened, along with different Fe com-
plexes. A schematic representation of the reaction is shown in
Scheme 1.
The results obtained in the oxidation of diphenylmethane 11
with different iron catalyst are summarized in Table 3. Low to
moderate conversions were obtained. The best yields were
achieved with complex 6, up to 73% (entry 1). The nature of the
backbone was found to influence the activity, with catalyst 7 giving
The catalytic oxidation was extended to ethylbenzene deriva-
tives, the results of which are summarized in Fig. 5. Ethylbenzene
is a more challenging substrate than diphenylmethane as it is less

O. Martínez-Ferraté et al. / Inorganica Chimica Acta 431 (2015) 156–160
159
4.2. Synthesis of Fe complexes
Effect of ethylbenzene substituent
80
70
60
50
40
30
20
10
0
A
Schlenk tube was charged with Fe(BF4)2ꢀ6H2O (34 mg,
0.1 mmol). Then 3 mL of dry THF and 0.1 mL of TEOF (0.6 mmol)
were added. The solution was stirred for 30 min and it became
light purple. Ligand was added (0.1 mmol) to the stirring solution.
The resulting solution was allowed to stir during 3 h. Then 10 mL
of dry diethyl ether were added and the solution was filtered and
dried. The resulting solid was clean 3 times with diethyl ether
and dried under vacuum.
6
7
8
9
[Fe212(CH3CN)4](BF4)4, 6. Yield: 74% 1H NMR (400 MHz, CD3CN)
d = 10.50 (s), 9.68 (s), 7.31 (s), 6.35 (s), 6.06 (s), 5.15 (s), 3.67 (s),
3.45 (s), 3.25 (s), 1.30 (s), 1.15 (s), 0.91 (s). C66H60B4F16Fe2N12O4ꢀ4H2-
O (1616.41): Calc. C, 49.05; H, 4.24. Found C, 48.81; H, 4.24%.
[Fe222(CH3CN)4](BF4)4, 7. Yield: 65%. 1H NMR (500 MHz, Acetoni-
trile-d3) d = 9.35–8.66 (m), 8.60–7.86 (m), 7.85–7.19 (m, 0H), 7.11–
6.63 (m), 5.12 (m), 3.37–2.07 (m), 1.27 (s). C64H56B4F16Fe2N12O4ꢀ
4H2O (1516.41): Calc. C, 48.40; H, 4.06. Found C, 48.56; H, 4.24%.
[Fe232(CH3CN)4](BF4)4, 8. Yield: 71%. 1H NMR (500 MHz, CD3CN)
d = 8.54 (s), 8.01 (s), 7.99 (s), 7.92 (s), 6.57 (s), 6.32 (s), 6.07 (s), 4.76
10
17
18
19
20
Substrate
Fig. 5. Effect of substituents in ethylbenzene in iron catalyzed oxidation of benzylic
C–H bonds. Reaction conditions: 0.3 mmol of substrate, 0.015 eq. of 6, oxidant
tBuOOH in 0.5 mL of ACN, 6 h and r.t. Slow addition of oxidant, over 30 min, every
5 min. Chemical yield determined by GC using o-dichlorobenzene as external
standard.
(bs), 3.92 (s), 2.26 (s), 2.22 (s).
C
66H68B4F16Fe2N12O4ꢀ4H2O
(1624.29): Calc. C, 48.80; H, 4.72. Found C, 48.52; H, 4.62%.
[Fe42(CH3CN)2](BF4)2, 9. Yield: 71% 1H NMR (500 MHz, Acetoni-
trile-d3) d = 8.73 (d, J = 17.8 Hz), 8.35 (s), 7.78 (s), 7.15 (d,
J = 30.8 Hz), 6.97–6.77 (m), 4.90 (s), 3.78 (d, J = 114.6 Hz), 2.70–
2.07 (m), 1.34 (s). C50H46B2F8FeN6O4ꢀ3H2O (1616.41): Calc. C,
55.69; H, 4.86. Found C, 55.35; H, 4.89%.
activated towards oxidation. A similar trend was observed, imino-
pyridine complexes were more active than aminopyridine com-
plexes. Likewise for 11, no cooperative effect was observed as
mononuclear and dinuclear complexes oxidize ethylbenzene to
acetophenone 21 with comparable yields. Electronic effects were
observed for substituents in the para position, substrates 18 and
19. Thus EDG activated the substrate towards oxidation due to
an increase of electron density of the ring and 19 was converted
with up to 75% yield, which is very close to 11 which contains
two aryl groups and is doubly activated. On the contrary, EWD
groups in the substrates gave only 32% yields. The most hindered
substrate, 20 gave lower yields and the best catalytic system gave
only 40% (Fig. 4). Two different effects could explain this
behaviour: (a) the steric hindrance could block the approach of
the substrate to the iron centre and (b) the intermediate alcohol
may coordinate as a chelate to the Fe catalyst, slowing down the
reaction.
4.3. Catalytic tests
A Schlenk tube was charged with the substrate (0.3 mmol) and
the catalyst (4.5 lmol) dissolved in ACN (0.5 mL). The oxidant
tBuOOH (0.32 mL, 70 wt% in H2O, 2.4 mmol) was slowly added
within 30 min, and then solution was shaken for 6 h at room tem-
perature. 0.2 mL of the solution were dissolved in 1 mL of DCM and
the yield was determined by GC. After that 1 lL of solution was
injected to a GC (Pressure = 111.7 KPa of He, T oven-initial = 50 °C
during 0 min, ramp temperature 15 °C/min, Final oven tempera-
ture = 200 °C during 10 min, T inj = Tdet = 250 °C). Conditions were
different for diphenylmethane (Pressure = 111.7 KPa of He, T oven-
initial = 100 °C during 0 min, ramp temperature 25 °C/min, T oven-
final = 320 °C during 10 min, T-inj = T-det = 250 °C).
3. Conclusions
4.4. X-ray crystallography
In summary, several iminopyridine and aminopyridine iron
complexes were applied in the catalytic oxidation of benzylic
C–H bonds. This reaction offered moderate to high yields which
were highly affected by electronic and steric properties of the sub-
strate; EWG containing and sterically demanding substrates were
the most difficult to oxidize. As general trend the best catalyst
was the dinuclear complex 6 containing ligand 1. No cooperative
effect between the iron nuclei was observed. The results presented
in this work are comparable to those previously reported in litera-
ture [23,24].
Measurements were made on a Bruker-Nonius diffractometer
equipped with a APPEX 2 4 K CCD area detector, a FR591 rotating
anode with Mo K
a radiation, Montel mirrors as monochromator
and a Kryoflex low temperature device (T = ꢁ173 °C). Full-sphere
data collection was used with w and j scans. Programs used: Data
collection APEX2 V. 1.0-22 (Bruker-Nonius 2004), data reduction
SAINT+Version 6.22 (Bruker-Nonius 2001) and absorption correction
SADABS V. 2.10 (2003).
Empirical formula
Formula weight
Space group
Unit cell dimensions
a (Å)
C72H69B4F16Fe2N15O4
1667.36
P4(2)/n
4. Experimental
4.1. General considerations
23.7118(19)
90.00
23.7118(19)
90.00
Solvents were purchased from Sigma–Aldrich as HPLC grade
and dried with an SPS system of ITC-inc. Reagents were used as
commercially available. Gas chromatographic analyses were run
a
(°)
b (Å)
b (°)
c (Å)
on
a Hewlett-Packard HP 5890A instrument (split/splitless
3.5686(14)
injector, J&W Scientific, IA, 25 m column, internal diameter
0.25 mm, film thickness 0.33 mm, carrier gas: He, F.I.D. detector)
equipped with a Hewlett-Packard HP 3396 series II integrator.
(continued on next page)

160
O. Martínez-Ferraté et al. / Inorganica Chimica Acta 431 (2015) 156–160
[12] C. Kim, K. Chen, J. Kim, L. Que, J. Am. Chem. Soc. 7863 (1997) 5964.
c
(°)
90.00
[13] T. Okuno, S. Ito, S. Ohba, Y. Nishida, Dalton Trans. (1997) 3547
[14] L. Gómez, I. Garcia-Bosch, A. Company, J. Benet-Buchholz, A. Polo, X. Sala, X.
Ribas, M. Costas, Angew. Chem., Int. Ed. 48 (2009) 5720.
[15] L. Gómez, M. Canta, D. Font, I. Prat, X. Ribas, M. Costas, J. Org. Chem. 78 (2013)
1421.
[16] M. Canta, D. Font, L. Gómez, X. Ribas, M. Costas, Adv. Synth. Catal. 356 (2014)
818.
[17] K. Chen, L. Que, Chem. Commun. (1999) 1375
[18] A. Company, L. Go, M. Gu, X. Ribas, J.M. Luis, J. Am. Chem. Soc. 129 (2007)
15766.
[19] G. Roelfes, M. Lubben, R. Hage, L. Que, B.L. Feringa, Chem. Eur. J. 6 (2000) 2152.
[20] M.S. Chen, M.C. White, Science 318 (2007) 783.
[21] M.C. White, Science 335 (2012) 807.
[22] M.S. Chen, M.C. White, Science 327 (2010) 566.
[23] P. Shejwalkar, N.P. Rath, E.B. Bauer, Dalton Trans. 40 (2011) 7617.
[24] M. Lenze, E.T. Martin, N.P. Rath, E.B. Bauer, ChemPlusChem 78 (2013) 101.
[25] O. Martínez-Ferraté, Synthesis of Dinuclear Complexes. From Ligand Design to
Catalysis, Universitat Rovira i Virgili, 2013. http://www.tesisenred.net/handle/
10803/126531.
U (Å3)
7628.9(12)
4, 1.452 mg/m3
0.478
3416
44832
9123 [Rint = 0.1428]
721
0.0965, 0.2319
0.1660, 0.2641
0.777 and ꢁ0.638
Z, Dcalc
M (mmꢁ1
F(000)
Reflections collected
Independent reflections
Parameters
)
R1, wR2 [I > 2
r(I)]
R1, wR2 (all data)
Largest peak/hole (e Åꢁ3
)
Acknowledgements
[26] S.C. Bart, E.B. Lobkovsky, E. Bill, K. Wieghardt, P.J. Chirik, Inorg. Chem. 46
(2007) 7055.
[27] M.D. Godbole, M.P. Puig, S. Tanase, H. Kooijman, A.L. Spek, E. Bouwman, Inorg.
Chim. Acta 360 (2007) 1954.
[28] N. Carvalho, A. Horn, O. Antunes, Appl. Catal., A Gen. 305 (2006) 140.
[29] A. Machkour, D. Mandon, M. Lachkar, R. Welter, Eur. J. Inorg. Chem. 2005
(2005) 158.
The authors are grateful to the Spanish Ministerio de Educación
y Ciencia (CTQ2010-14938/BQU, Consolider Ingenio 2010) and the
Generalitat de Catalunya – Spain (for financial support 2009SGR116)
and the Universitat Rovira I Virgili for awarding a research grant to
Oriol Martínez (URV 2010PFR-URV-B2-01).
[30] J. Larsen, K.A. Jsrgensen, J. Chem. Soc., Perkin Trans. 2 (1992) 1213.
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.ica.2014.12.016.
References
[1] A.E. Shilov, G.B. Shul’pin, Chem. Rev. 97 (1997) 2879.
[2] J.F. Brazdil, Top. Catal. 38 (2006) 289.
[3] H.J.H. Fenton, J. Chem. Soc., Trans. 65 (1894) 899.
[4] J. Prousek, Pure Appl. Chem. 79 (2007) 2325.
[5] C. Pavan, J. Legros, C. Bolm, Adv. Synth. Catal. 347 (2005) 703.
[6] S. Friedle, E. Reisner, S.J. Lippard, Chem. Soc. Rev. 39 (2010) 2768.
[7] F.P. Guengerich, Chem. Res. Toxicol. 14 (2001) 611.
[8] L. Que, W.B. Tolman, Nature 455 (2008) 333.
[9] M. Costas, K. Chen, L. Que, Coord. Chem. Rev. 202 (2000) 517.
[10] M. Costas, Coord. Chem. Rev. 255 (2011) 2912.
[11] A. Correa, O. García, Chem. Soc. Rev. 37 (2008) 1108.