Combined experimental and theoretical study on aromatic hydroxylation by mononuclear nonheme iron(IV)-oxo complexes

Article abstract of DOI:10.1021/ic700462h

The hydroxylation of aromatic compounds by mononuclear nonheme iron(IV)-oxo complexes, [Fe^IV(Bn-tpen)(O)]²⁺ (Bn-tpen = N-benzyl-N,N′,N′-tris(2-pyridylmethyl)ethane-1,2-diamine) and [Fe^IV(N4Py)(O)]²⁺ (N4Py = N,N-bis(2-pyridylmethyl)-N- bis(2-pyridyl)methylamine), has been investigated by a combined experimental and theoretical approach. In the experimental work, we have performed kinetic studies of the oxidation of anthracene with nonheme iron(IV)-oxo complexes generated in situ, thereby determining kinetic and thermodynamic parameters, a Hammett ρ value, and a kinetic isotope effect (KIE) value. A large negative Hammett ρ value of -3.9 and an inverse KIE value of 0.9 indicate that the iron-oxo group attacks the aromatic ring via an electrophilic pathway. By carrying out isotope labeling experiments, the oxygen in oxygenated products was found to derive from the nonheme iron(IV)-oxo species. In the theoretical work, we have conducted density functional theory (DFT) calculations on the hydroxylation of benzene by [Fe^IV(N4Py)(O)]²⁺. The calculations show that the reaction proceeds via two-state reactivity patterns on competing triplet and quintet spin states via an initial rate determining electrophilic substitution step. In analogy to heme iron(IV)-oxo catalysts, the ligand is noninnocent and actively participates in the reaction mechanism by reshuttling a proton from the ipso position to the oxo group. Calculated kinetic isotope effects of C₆H₆ versus C₆D₆ confirm an inverse isotope effect for the electrophilic substitution pathway. Based on the experimental and theoretical results, we have concluded that the aromatic ring oxidation by mononuclear nonheme iron(IV)-oxo complexes does not occur via a hydrogen atom abstraction mechanism but involves an initial electrophilic attack on the π-system of the aromatic ring to produce a tetrahedral radical or cationic σ-complex.

Full text of DOI:10.1021/ic700462h

Inorg. Chem. 2007, 46, 4632−4641
Combined Experimental and Theoretical Study on Aromatic
Hydroxylation by Mononuclear Nonheme Iron(IV) Oxo Complexes
−
Sam P. de Visser,*,‡ Kyungeun Oh,† Ah-Rim Han,† and Wonwoo Nam*,†
Department of Chemistry, DiVision of Nano Sciences, and Center for Biomimetic Systems,
Ewha Womans UniVersity, Seoul 120-750, Korea, and The Manchester Interdisciplinary Biocenter
and the School of Chemical Engineering and Analytical Science, The UniVersity of Manchester,
131 Princess Street, Manchester M1 7DN, United Kingdom
Received March 9, 2007
IV
2+
The hydroxylation of aromatic compounds by mononuclear nonheme iron(IV)
−oxo complexes, [Fe (Bn-tpen)(O)]
IV
2+
(
Bn-tpen N-benzyl-N,N ,N -tris(2-pyridylmethyl)ethane-1,2-diamine) and [Fe (N4Py)(O)] (N4Py ) N,N-bis(2-
)
′
′
pyridylmethyl)-N-bis(2-pyridyl)methylamine), has been investigated by a combined experimental and theoretical
approach. In the experimental work, we have performed kinetic studies of the oxidation of anthracene with nonheme
iron(IV)
value, and a kinetic isotope effect (KIE) value. A large negative Hammett
value of 0.9 indicate that the iron oxo group attacks the aromatic ring via an electrophilic pathway. By carrying out
isotope labeling experiments, the oxygen in oxygenated products was found to derive from the nonheme iron-
−oxo complexes generated in situ, thereby determining kinetic and thermodynamic parameters, a Hammett
F
F
value of 3.9 and an inverse KIE
−
−
(
IV)
−
oxo species. In the theoretical work, we have conducted density functional theory (DFT) calculations on the
hydroxylation of benzene by [Fe (N4Py)(O)]2+. The calculations show that the reaction proceeds via two-state
IV
reactivity patterns on competing triplet and quintet spin states via an initial rate determining electrophilic substitution
step. In analogy to heme iron(IV)
mechanism by reshuttling a proton from the ipso position to the oxo group. Calculated kinetic isotope effects of
versus C confirm an inverse isotope effect for the electrophilic substitution pathway. Based on the
experimental and theoretical results, we have concluded that the aromatic ring oxidation by mononuclear nonheme
iron(IV) oxo complexes does not occur via a hydrogen atom abstraction mechanism but involves an initial electrophilic
attack on the -system of the aromatic ring to produce a tetrahedral radical or cationic -complex.
−oxo catalysts, the ligand is noninnocent and actively participates in the reaction
C
H
6 6
6 6
D
−
π
σ
Introduction
have been identified in both enzymatic and biomimetic
2
-5
reactions.
A high-valent iron(IV)-oxo intermediate has
Oxygen-activating enzymes with mononuclear nonheme
iron active sites are involved in many metabolically important
oxidative transformations in nature. In the catalytic cycles
been characterized as an active oxidizing species in the
catalytic cycle of Escherichia coli taurine, namely an
1
2
R-ketoglutarate-dependent dioxygenase (TauD). In biomi-
of dioxygen activation by these enzymes, high-valent iron-
metic studies, mononuclear nonheme iron(IV)-oxo com-
plexes bearing tetradentate N4 and pentadentate N5 and N4S
(IV)-oxo species are frequently invoked as the key inter-
1
mediates responsible for the oxidation of organic substrates.
Very recently, such nonheme iron(IV)-oxo intermediates
(
2) (a) Price, J. C.; Barr, E. W.; Tirupati, B.; Bollinger, J. M., Jr.; Krebs,
C. Biochemistry 2003, 42, 7497-7508. (b) Price, J. C.; Barr, E. W.;
Glass, T. E.; Krebs, C.; Bollinger, J. M., Jr. J. Am. Chem. Soc. 2003,
125, 13008-13009. (c) Riggs-Gelasco, P. J.; Price, J. C.; Guyer, R.
B.; Brehm, J. H.; Barr, E. W.; Bollinger, J. M., Jr.; Krebs, C. J. Am.
Chem. Soc. 2004, 126, 8108-8109. (d) Bollinger, J. M., Jr.; Price, J.
C.; Hoffart, L. M.; Barr, E. W.; Krebs, C. Eur. J. Inorg. Chem. 2005,
4245-4254. (e) Bollinger, J. M., Jr.; Krebs, C. J. Inorg. Biochem.
2006, 100, 586-605.
*
To whom correspondence should be addressed. E-mail: wwnam@
ewha.ac.kr (W.N.), sam.devisser@manchester.ac.uk (S.P.d.V.).
†
Ewha Womans University.
The University of Manchester.
‡
(
1) (a) Neidig, M. L.; Solomon, E. I. Chem. Commun. 2005, 5843-5863.
(
b) Abu-Omar, M. M.; Loaiza, A.; Hontzeas, N. Chem. ReV. 2005,
105, 2227-2252. (c) Decker, A.; Solomon, E. I. Curr. Opin. Chem.
Biol. 2005, 9, 152-163. (d) Kryatov, S. V.; Rybak-Akimova, E. V.;
Schindler, S. Chem. ReV. 2005, 105, 2175-2226. (e) Borovik, A. S.
Acc. Chem. Res. 2005, 38, 54-61.
(3) (a) Shan, X.; Que, L., Jr. J. Inorg. Biochem. 2006, 100, 421-433. (b)
Decker, A.; Clay, M. D.; Solomon, E. I. J. Inorg. Biochem. 2006,
100, 697-706. (c) Neese, F. J. Inorg. Biochem. 2006, 100, 716-726.
4632 Inorganic Chemistry, Vol. 46, No. 11, 2007
10.1021/ic700462h CCC: $37.00
© 2007 American Chemical Society
Published on Web 04/20/2007

Mononuclear Nonheme Iron(IV)-Oxo Complexes
ligands have been synthesized and characterized with various
spectroscopic techniques including X-ray crystallography.
capable of oxygenating a pendent aromatic ring on the
ligand. The authors have provided mechanistic evidence that
3-5
9
The synthetic nonheme iron(IV)-oxo complexes have shown
reactivities in a variety of oxidation reactions, including
metal-centered electrophilic oxidants, presumably mono-
nuclear nonheme iron(IV)-oxo intermediates, are involved
in the arene hydroxylations, but direct evidence for the
participation of such intermediates in the catalytic cycles has
yet to be reported. More recently, Rybak-Akimova, Nam,
and their co-workers demonstrated independently that inter-
mediates generated in the reactions of mononuclear nonheme
alkane hydroxylation, olefin epoxidation, and the oxidation
of PPh
3
, sulfides, and alcohols.4
,5
Aromatic hydroxylation is an important chemical process
mediated by many metalloenzymes, including cytochromes
6
,7
P450 and nonheme iron mono- and dioxygenases. In
mononuclear nonheme iron enzymes, aromatic amino acid
hydroxylases, such as phenylalanine, tyrosine, and tryptophan
hydroxylases, catalyze the aromatic hydroxylation of the
namesake amino acids, concomitant with the two-electron
oxidation of the requisite organic cofactor tetrahydropterin
iron(II) complexes and oxidants such as H
2 2
O and m-
chloroperbenzoic acid hydroxylate benzoic acids to the
1
1,12
corresponding salicylic acids efficiently.
However, the
nature of active oxidizing intermediates and mechanisms for
the aromatic ring hydroxylations were not clearly understood
in the studies. Therefore, we deemed it timely to do a
combined experimental and theoretical study into the mech-
anisms of aromatic hydroxylation by nonheme iron-oxo
biomimetics. Since the aromatic hydroxylation reactions
reported so far have been investigated mainly under catalytic
conditions, we have performed mechanistic studies of the
aromatic ring hydroxylation with in situ-generated mono-
nuclear nonheme iron(IV)-oxo complexes. To support the
experimentally obtained conclusions we conducted density
functional theory (DFT) calculations on the hydroxylation
of benzene by a nonheme iron(IV)-oxo intermediate for the
first time. These experimental and theoretical results have
been discussed in the light of elucidating mechanisms of the
aromatic ring hydroxylation by mononuclear nonheme iron-
(IV)-oxo complexes.
8
to its quinonoid dihydropterin form. A mechanism was
proposed in which iron(IV)-oxo intermediates are involved
as active oxidants in the hydroxylation of aromatic rings,
although such iron(IV)-oxo species have not yet been
experimentally observed in the catalytic cycles of the
enzymes. In biomimetic studies, the catalytic hydroxylation
of aromatic compounds by nonheme iron(II) complexes has
been demonstrated in a number of cases.9 Notably, Que
and co-workers have shown that mononuclear nonheme iron-
,10
(II) complexes in combination with alkyl hydroperoxides are
(
4) (a) Rohde, J.-U.; In, J.-H.; Lim, M. H.; Brennessel, W. W.; Bukowski,
M. R.; Stubna, A.; M u¨ nck, E.; Nam, W.; Que, L., Jr. Science 2003,
2
99, 1037-1039. (b) Lim, M. H.; Rohde, J.-U.; Stubna, A.; Bukowski,
M. R.; Costas, M.; Ho, R. Y. N.; M u¨ nck, E.; Nam, W.; Que, L., Jr.
Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 3665-3670. (c) Kaizer, J.;
Klinker, E. J.; Oh, N. Y.; Rohde, J.-U.; Song, W. J.; Stubna, A.; Kim,
J.; M u¨ nck, E.; Nam, W.; Que, L., Jr. J. Am. Chem. Soc. 2004, 126,
4
72-473. (d) Kim, S. O.; Sastri, C. V.; Seo, M. S.; Kim, J.; Nam, W.
Results and Discussion
J. Am. Chem. Soc. 2005, 127, 4178-4179. (e) Oh, N. Y.; Suh, Y.;
Park, M. J.; Seo, M. S.; Kim, J.; Nam, W. Angew. Chem., Int. Ed.
Reactivities of Nonheme Iron(IV)-Oxo Complexes in
Aromatic Hydroxylation. The mononuclear nonheme iron-
(IV)-oxo complexes, [Fe (Bn-tpen)(O)] (1) (Bn-tpen )
N-benzyl-N,N′,N′-tris(2-pyridylmethyl)ethane-1,2-diamine)
2
005, 44, 4235-4239. (f) Sastri, C. V.; Park, M. J.; Ohta, T.; Jackson,
T. A.; Stubna, A.; Seo, M. S.; Lee, J.; Kim, J.; Kitagawa, T.; M u¨ nck,
IV
2+
E.; Que, L., Jr.; Nam, W. J. Am. Chem. Soc. 2005, 127, 12494-12495.
(g) Bukowski, M. R.; Koehntop, K. D.; Stubna, A.; Bominaar, E. L.;
Halfen, J. A.; M u¨ nck, E.; Nam, W.; Que, L., Jr. Science 2005, 310,
000-1002. (h) Klinker, E. J.; Kaizer, J.; Brennessel, W. W.;
Woodrum, N. L.; Cramer, C. J.; Que, L., Jr. Angew. Chem., Int. Ed.
005, 44, 3690-3694. (i) Sastri, C. V.; Oh, K.; Lee, Y. J.; Seo, M.
S.; Shin, W.; Nam, W. Angew. Chem., Int. Ed. 2006, 45, 3992-3995.
j) Park, M. J.; Lee, J.; Suh, Y.; Kim, J.; Nam, W. J. Am. Chem. Soc.
006, 128, 2630-2634.
IV
2+
1
and [Fe (N4Py)(O)] (2) (N4Py ) N,N-bis(2-pyridyl-
methyl)-N-bis(2-pyridyl)methylamine), used in this work are
2
4
c
shown in Figure 1. The reactive species were prepared by
treating their corresponding iron(II) complexes, Fe(Bn-tpen)-
(
2
(
CF
zene (PhIO) in a solvent mixture of CH
) at 25 °C. Upon addition of anthracene to the solutions,
3
SO
3
)
2
and Fe(N4Py)(CF
3
SO
3
)
2
, with solid iodosylben-
(
5) (a) Grapperhaus, C. A.; Mienert, B.; Bill, E.; Weyherm u¨ ller, T.;
Wieghardt, K. Inorg. Chem. 2000, 39, 5306-5317. (b) Balland, V.;
Charlot, M.-F.; Banse, F.; Girerd, J.-J.; Mattioli, T. A.; Bill, E.; Bartoli,
J.-F.; Battioni, P.; Mansuy, D. Eur. J. Inorg. Chem. 2004, 301-308.
3
CN and CH Cl (1:
2
2
13
1
the intermediates reverted to the starting iron(II) complexes
with a clear isosbestic point at 490 nm for 1 and 538 nm for
(
c) Martinho, M.; Banse, F.; Bartoli, J.-F.; Mattioli, T. A.; Battioni,
P.; Horner, O.; Bourcier, S.; Girerd, J.-J. Inorg. Chem. 2005, 44, 9592-
596.
9
2
and showed pseudo-first-order decay as monitored by a
(
6) (a) Sono, M.; Roach, M. P.; Coulter, E. D.; Dawson, J. H. Chem.
ReV. 1996, 96, 2841-2887. (b) Meunier, B. In Biomimetic Oxidations
Catalyzed by Transition Metal Complexes; Meunier, B., Ed.; Imperial
College Press: London, 2000; pp 171-214. (c) Meunier, B.; de Visser,
S. P.; Shaik, S. Chem. ReV. 2004, 104, 3947-3980. (d) Cytochrome
P450: Structure, Mechanism, and Biochemistry, 3rd ed.; Ortiz de
Montellano, P. R., Ed.; Kluwer Academic/Plenum Publishers: New
York, 2005. (e) Shaik, S.; Kumar, D.; de Visser, S. P.; Altun, A.;
Thiel, W. Chem. ReV. 2005, 105, 2279-2328.
(10) (a) Kitajima, N.; Ito, M.; Fukui, H.; Moro-oka, Y. J. Am. Chem. Soc.
1993, 115, 9335-9336. (b) Funabiki, T.; Yokomizo, T.; Suzuki, S.;
Yoshida, S. Chem. Commun. 1997, 151-152. (c) Menage, S.; Galey,
J.-B.; Dumats, J.; Hussler, G.; Seite, M.; Luneau, I. G.; Chottard, G.;
Fontecave, M. J. Am. Chem. Soc. 1998, 120, 13370-13382. (d)
Mekmouche, Y.; Menage, S.; Toia-Duboc, C.; Fontecave, M.; Galey,
J.-B.; Lebrun, C.; Pecaut, J. Angew. Chem., Int. Ed. 2001, 40, 949-
952. (e) Mehn, M. P.; Fujisawa, K.; Hegg, E. L.; Que, L., Jr. J. Am.
Chem. Soc. 2003, 125, 7828-7842. (f) Avenier, F.; Dubois, L.; Latour,
J.-M. New. J. Chem. 2004, 28, 782-784.
(
7) (a) Solomon, E. I.; Brunold, T. C.; Davis, M. I.; Kemsley, J. N.; Lee,
S.-K.; Lehnert, N.; Neese, F.; Skulan, A. J.; Yang, Y.-S.; Zhou, J.
Chem. ReV. 2000, 100, 235-349. (b) Costas, M.; Mehn, M. P.; Jensen,
M. P.; Que, L., Jr. Chem. ReV. 2004, 104, 939-986.
(11) Taktak, S.; Flook, M.; Foxman, B. M.; Que, L., Jr.; Rybak-Akimova,
E. V. Chem. Commun. 2005, 5301-5303.
(
8) (a) Flatmark, T.; Stevens, R. C. Chem. ReV. 1999, 99, 2137-2160.
(
b) Klinman, J. P. J. Biol. Inorg. Chem. 2001, 6, 1-13. (c) Fitzpatrick,
(12) Oh, N. Y.; Seo, M. S.; Lim, M. H.; Consugar, M. B.; Park, M. J.;
Rohde, J.-U.; Han, J.; Kim, K. M.; Kim, J.; Que, L., Jr.; Nam, W.
Chem. Commun. 2005, 5644-5646.
P. F. Biochemistry 2003, 42, 14083-14091.
(
9) (a) Lange, S. J.; Miyake, H.; Que, L., Jr. J. Am. Chem. Soc. 1999,
1
21, 6330-6331. (b) Jensen, M. P.; Lange, S. J.; Mehn, M. P.; Que,
(13) The intermediates, 1 and 2, do not react with benzene and naphthalene
due to the low reactivity of the nonheme iron(IV)-oxo species.
E. L.; Que, L., Jr. J. Am. Chem. Soc. 2003, 125, 2113-2128.
Inorganic Chemistry, Vol. 46, No. 11, 2007 4633

de Visser et al.
Figure 1. Structure of ligands and iron(IV)-oxo complexes used in this
study. The hydrogen atoms have been omitted for clarity.
UV-vis spectrophotometer (Figure 2a for 1 and Supporting
Information, Figure S1 for 2). Pseudo-first-order fitting of
the kinetic data allowed us to determine the kobs values to
-
2
-1
-3 -1
be 5.9(4)•10
s
and 2.5(3)•10
s
for the reactions of 1
and 2 at 25 °C (Figure 2a and Figure S1, insets), indicating
that 1 is more reactive than 2 toward aromatic hydroxylation.
The greater reactivity of 1 than 2 has also been observed in
alkane hydroxylations.4 Product analysis of the reaction
mixtures revealed that anthraquinone was produced in high
yields in the reactions (∼40% based on intermediates
generated). The formation of quinone products has been well
c
documented in aromatic hydroxylation reactions catalyzed
by metalloporphyrins,1
4-16
and a mechanism has been
proposed in which two metal-oxo complexes are needed
for the formation of one quinone product.1 The pseudo-
first-order rate constants increased proportionally with
substrate concentration, thus leading us to determine the
4,15
-
1
-1 -1
second-order rate constant to be 6.8(5)•10 M •s in the
anthracene oxidation by 1 (Figure 2b). Activation parameters
for the anthracene oxidation were determined by plotting
first-order rate constants measured at different temperatures
Figure 2. (a) UV-vis spectral changes of 1 (1 mM) upon addition of 10
equiv of anthracene (10 mM) at 25 °C. Inset shows absorbance traces
monitored at 737 nm. (b) Plot of kobs against anthracene concentration to
determine the second-order rate constant at 0 °C. (c) Plot of first-order-rate
constants against 1/T to determine activation parameters.
(
273-298 K) against 1/T (Figure 2c); the activation enthalpy
q
and entropy values were calculated to be ∆H ) 13(1)
-
1
q
-1
-1
kcal•mol and ∆S ) -22(2) cal•mol •K .
We then investigated the electronic effect of substrate on
the aromatic hydroxylation by nonheme iron(IV)-oxo spe-
cies. The first-order rate constants determined in the oxidation
of various para-X-substituted anthracenes by 1 indicate that
the electron-donating ability of the para-substituents influ-
ences the reaction rate significantly (Supporting Information,
p
Table S1). By plotting the rates as a function of σ of the
para-substituents, a good linear correlation was obtained with
a large negative Hammett F value of -3.9(2) (Figure 3),
indicating a high charge separation in the transition state of
Figure 3. Hammett plot of log krel against σp of para-X-substituted
anthracenes. First-order rate constants for para-substituted anthracenes were
determined at 0 °C (Supporting Information, Table S1), and the krel values
were calculated by dividing kobs of para-X-anthracene by kobs of anthracene.
(
14) (a) Song, R.; Sorokin, A.; Bernadou, J.; Meunier, B. J. Org. Chem.
1
997, 62, 673-678. (b) Sorokin, A.; Meunier, B. Eur. J. Inorg. Chem.
1
998, 1269-1281.
(
15) Higuchi, T.; Satake, C.; Hirobe, M. J. Am. Chem. Soc. 1995, 117,
8
879-8880.
16) Khavasi, H. R.; Davarani, S. S. H.; Safari, N. J. Mol. Catal. A: Chem.
002, 188, 115-122.
the aromatic ring oxidation. Such a large negative F value
implies that the iron-oxo group attacks the aromatic ring
(
2
4634 Inorganic Chemistry, Vol. 46, No. 11, 2007

Mononuclear Nonheme Iron(IV)-Oxo Complexes
Scheme 1. Incorporation of Isotopically Labeled 18O into Product
Table 1. Rates Determined in the Hydroxylation of Anthracene and
Anthracene-d10 by 1 and 2a,b
1
2
rate constant (10- s-1
3
)
rate constant (10-3 s-1
)
C14H10
C14D10
kH/kD
C14H10
C14D10
kH/kD
7.1
7.7
0.92
2.4
2.8
0.88
a
Rates were determined by adding 10 equiv of anthracene (10 mM) to
a reaction containing 1 or 2 (1 mM each) in a solvent mixture of CH3CN
b
and CH2Cl2 (1:1) at 0 °C for 1 or at 25 °C for 2. Rate constants are
averaged by three determinations.
via an electrophilic pathway. Moreover, it is worth noting
that a very small Hammett F value was observed in the
oxidation of alcohols by nonheme iron(IV)-oxo complexes
Scheme 2. Possible Reaction Mechanisms of Anthracene
Hydroxylation by Nonheme Iron(IV)-Oxo Intermediates
(
i.e., F value of ∼ -0.1), in which the alcohol oxidation
4e
occurs via a hydrogen atom abstraction mechanism.
Kinetic isotope effects (KIE) were also investigated
kinetically in the hydroxylation of undeuterated and deuter-
ated anthracenes by nonheme iron(IV)-oxo complexes
(
Table 1). The calculated k
H
/k
D
values of ∼0.9 in the
reactions of 1 and 2 indicate the inverse KIE in the aromatic
ring oxidation reactions; the observation of the inverse KIE
2
3
is consistent with the sp -to-sp hybridization change during
2
the addition of an electrophilic iron-oxo group to the sp
center of aromatic ring to form a σ adduct.17-19 Further, the
KIE value of ∼0.9 rules out a hydrogen atom abstraction
mechanism; the oxidation of ethylbenzene and benzyl alcohol
by 2 occurs via a hydrogen atom abstraction mechanism,
and large KIE values of ∼50 were observed in these
oxygen of anthraquinone does not exchange with H 1
under the reaction conditions.
8O
2
In summary, we have shown that nonheme iron(IV)-oxo
complexes are capable of conducting aromatic hydroxylation
reactions; the kinetic and thermodynamic parameters were
determined in the oxidation of anthracene. Based on the large
negative Hammett F value and inverse KIE value, we have
proposed that the aromatic oxidation does not occur via a
hydrogen atom abstraction mechanism (Scheme 2, pathway
ii) but involves an initial electrophilic attack on the π-system
of the aromatic ring to produce a tetrahedral radical or
cationic σ-complex (Scheme 2, pathway i). By carrying out
4
c,4e
reactions.
Finally, the source of oxygen incorporated into oxygenated
products was investigated by carrying out isotope labeling
IV
studies in the hydroxylation of anthracene by [Fe (Bn-tpen)-
1
8
2+
18
18
(
(
O)] (1- O). First, 1- O was prepared by reacting Fe-
1
8
18
18
2
Bn-tpen)(OTf) with PhI O, and the O percentage in 1- O
was determined to be 75% by taking an Electrospray
Ionization Mass Spectrum of the intermediate (Supporting
Information, Figure S2).20 Then, 1- O was reacted with
anthracene. To increase the accuracy of product analysis, a
18
18
O-labeled water experiments, the oxygen in oxygenated
product was found to derive from iron-oxo intermediates.
DFT Calculations on the Hydroxylation of Benzene by
Mononuclear Nonheme Iron(IV)-Oxo Species. To support
the experimental findings we have performed DFT calcula-
II
2+
catalytic hydroxylation of anthracene by [Fe (Bn-tpen)]
1
8O at
and PhIO was also carried out in the presence of H
2
2
5 °C under inert atmosphere. By analyzing the reaction
IV
solutions with GC and GC-MS, we found that anthraquino-
ne was produced as a major product (35% based on PhIO
added) and that the percentages of M, M + 2, and M + 4
products were 13(3), 36(4), and 51(5)%, respectively (Scheme
tions on a model reaction of aromatic hydroxylation by [Fe -
2
+
(N4Py)(O)] (2) using benzene as a substrate. Earlier DFT
2
2
calculations showed that 2 reacts with alkanes via two-
state reactivity patterns on competing triplet and quintet spin-
state surfaces; therefore, here we also calculated the reaction
mechanisms on these two spin-state surfaces. The two spin
states only differ in the occupation of the metal 3d set of
2
1
1
). The observation that the anthraquinone product con-
18
tained significant amounts of O derived from the iron-
oxo group demonstrates unambiguously that the source of
oxygen incorporated into the product is the iron(IV)-oxo
3
2
2
1
1
5
orbitals and 2 has occupation π*
2
π* π* , while 2
x -y xz yz
2
1
complex. In control experiments, we checked that the
(
21) (a) Nam, W.; Valentine, J. S. J. Am. Chem. Soc. 1993, 115, 1772-
(
17) (a) Griffin, G. W.; Horn, K. A. J. Am. Chem. Soc. 1987, 109, 4919-
1778. (b) Bernadou, J.; Fabiano, A.-S.; Robert, A.; Meunier, B. J.
Am. Chem. Soc. 1994, 116, 9375-9376. (c) Lee, K. A.; Nam, W. J.
Am. Chem. Soc. 1997, 119, 1916-1922. (d) Groves, J. T.; Lee, J.;
Marla, S. S. J. Am. Chem. Soc. 1997, 119, 6269-6273. (e) Bernadou,
J.; Meunier, B. Chem. Commun. 1998, 2167-2173. (f) Meunier, B.;
Bernadou, J. Struct. Bonding 2000, 97, 1-35.
4
9
1
926. (b) Hanzlik, R. P.; Ling, K.-H. J. J. Am. Chem. Soc. 1993, 115,
363-9370. (c) Vannelli, T.; Hooper, A. B. Biochemistry 1995, 34,
1743-11749.
(18) Shiota, Y.; Suzuki, K.; Yoshizawa, K. Organometallics 2005, 24,
3532-3538.
(
(
19) de Visser, S. P.; Shaik, S. J. Am. Chem. Soc. 2003, 125, 7413-7424.
20) Seo, M. S.; In, J.-H.; Kim, S. O.; Oh, N. Y.; Hong, J.; Kim, J.; Que,
L., Jr.; Nam, W. Angew. Chem., Int. Ed. 2004, 43, 2417-2420.
(22) (a) Kumar, D.; Hirao, H.; Que, L., Jr.; Shaik, S. J. Am. Chem. Soc.
2005, 127, 8026-8027. (b) Hirao, H.; Kumar, D.; Que, L., Jr.; Shaik,
S. J. Am. Chem. Soc. 2006, 128, 8590-8606.
Inorganic Chemistry, Vol. 46, No. 11, 2007 4635

de Visser et al.
Scheme 3. +Raeaction Mechanism and Nomenclature of Structures of Intermediates and Transition States for the Benzene Hydroxylation Reaction by
2
FedO(N4Py)
a
We also calculated a long-range complex (5,3RC) between 5,3FedO(N4Py)2+ and benzene (not shown in this scheme).
1
1
xz
1
yz
1
22
has occupation π*
2
2
π* π* σ* . The triplet spin
of an intermediate species with a cation on the benzene rest-
x -y
xy
5,3
5,3
state is the ground state with the quintet spin state higher in
energy by 13.1 (LACVP) and 8.0 (LACV3P+*) kcal mol ,
while zero-point corrections at the LACVP level reduce the
group ( A) or a radical on the benzene restgroup ( B). In the
case of benzene and ethylbenzene hydroxylation by oxoiron
heme models, the mechanisms passing the cationic and radi-
cal intermediates were close-lying, and the ordering of the rate
determining steps was dependent on small external perturba-
tions such as a dielectric constant or hydrogen bonding.1
We tested two different mechanisms for the relay of the ipso
proton to the oxygen atom. First, a direct hydrogen atom
-1
-1
triplet-quintet energy gap further to 11.9 and 6.7 kcal mol ,
respectively, for the two basis sets. Therefore, the triplet-
quintet energy gap is sensitive to the quality of the basis set
and ZPE corrections, as observed before by Shaik and co-
workers who found the same trends in the energetics. By
contrast, other nonheme iron(IV)-oxo species such as the
active species of TauD has a quintet spin ground state well
separated from the triplet.23 Therefore, the nature of the
ligands bound to iron regulates the triplet-quintet energy
splitting and consequently the reactivity patterns. Our
optimized geometries of 3 2 are shown in Figure S7 of the
Supporting Information, and the results are in perfect
agreement with earlier work.2
Subsequently, we calculated the benzene hydroxylation by
using several possible reaction mechanisms (Scheme 3).
The top two mechanisms in Scheme 3 are analogous to the
ones studied for oxoiron heme catalysts19,25 and proceed via
an electrophilic substitution of benzene with attack on the
oxygen atom of the oxoiron group, resulting in the formation
9,25b
2
2
5
,3
5,3
(HA) transfer from the ipso position of A and B to the
5
,3
oxygen atom to give phenol products via barriers TSHA,A
or TSHA,B, respectively, was tested. Second, a proton shuttle
5,3
mechanism via one of the nitrogen atoms of N4Py was tested,
5
,3
whereby first a proton-transfer intermediate ( C) is formed
,5
5,3
5,3
5,3
5,3
from A and B via barriers TSPT,A and TSPT,B
,
respectively. Subsequently, the proton is reshuttled to the
2,24
5,3
oxygen atom via a barrier TSreb to finally give the phenol
5
,3
products ( D). In the case of benzene hydroxylation by
oxoiron heme catalysts, the latter mechanism was found to
be much lower in energy than a direct hydrogen atom
5,32
19
transfer. In addition, we tested a direct hydrogen abstraction
2
+
5,3
from benzene by FedO(N4Py) reactant ( 2) leading to a
5
,3
hydroxoiron complex ( E) with a nearby benzyl radical,
followed by OH rebound to the benzyl radical to form
phenol. Alkane hydroxylations by oxoiron heme and non-
heme catalysts typically react via an initial hydrogen ab-
straction followed by rebound of the hydroxo group.4
(
23) (a) de Visser, S. P. Angew. Chem., Int. Ed. 2006, 45, 1790-1793. (b)
de Visser, S. P. J. Am. Chem. Soc. 2006, 128, 9813-9824.
24) (a) de Visser, S. P.; Shaik, S.; Sharma, P. K.; Kumar, D.; Thiel, W.
J. Am. Chem. Soc. 2003, 125, 15779-15788. (b) de Visser, S. P.;
Kumar, D.; Cohen, S.; Shacham, R.; Shaik, S. J. Am. Chem. Soc. 2004,
(
c,24b,26
1
26, 8362-8363.
(
25) (a) Bathelt, C. M.; Ridder, L.; Mulholland, A. J.; Harvey, J. N. Org.
Biomol. Chem. 2004, 2, 2998-3005. (b) de Visser, S. P. Chem. Eur.
J. 2006, 12, 8168-8177.
(26) (a) de Visser, S. P.; Ogliaro, F.; Sharma, P. K.; Shaik, S. Angew. Chem.,
Int. Ed. 2002, 41, 1947-1951. (b) de Visser, S. P.; Ogliaro, F.; Sharma,
P. K.; Shaik, S. J. Am. Chem. Soc. 2002, 124, 11809-11826.
4636 Inorganic Chemistry, Vol. 46, No. 11, 2007

Mononuclear Nonheme Iron(IV)-Oxo Complexes
Figure 4. Potential energy surface for the hydroxylation of benzene by 5,3FedO(N4Py)2+ via cationic intermediates. The proton shuttle mechanism in the
triplet spin state is in blue, and the one in the quintet spin state is in red. All energies are obtained at the UB3LYP/LACV3P+* level with ZPE corrections
-
1
at UB3LYP/LACVP. Energies are in kcal mol relative to the reactant complex in the triplet spin state. Values in parentheses are ∆E+ZPE energies with
5
,3
solvent corrections. Also shown are optimized geometries of A with bond distances in Å.
5
Optimized geometries of all optimized structures are shown
in the Supporting Information, while we will focus here on
the major trends and reaction mechanisms.
Although the barrier leading to B is the lowest, we expect
5
5
fast equilibration between A and B as the cationic
intermediates ( A) are substantially lower in energy than
the radical intermediates ( B) by 5.4 (triplet) and 6.9
5
,3
5
,3
Figure 4 shows the potential energy surface for the reaction
mechanisms of benzene hydroxylation via cationic interme-
-1
(quintet) kcal mol . The reason for the fact that the cationic
intermediates are more stable than the radical ones is because
the FedO(N4Py) model has an overall charge of +2, and
the charge repulsion within the model will favor the cationic
intermediates. By contrast, the arene hydroxylation by heme
models used an overall neutral system. There are some
critical geometric differences between A and B, which may
have an impact on the triplet-quintet spin-state crossing.
Due to double occupation of the π*x2-y2 orbital in A, the
5,3
diates ( A), whereas Figure 5 shows the one that passes
5,3
radical intermediates ( B). Thus, initially the reaction
5
,3 5,3
bifurcates into radical and cationic intermediates ( A, B),
5
,3
but later the surfaces converge again to form structures
C
and 5 D. The cationic mechanisms require a relatively high
) of 30.6 (quintet) and 35.7
triplet) kcal mol , while the barriers leading to radical
intermediates are significantly lower. In particular, the barrier
,3
5,3
5
5
C-O formation barrier ( TS
(
A
-1
5
5
5
(
TS
B
) leading to B is barely higher in energy than the
average Fe-N distance of the four Fe-N bonds in the xy-
5
5
reactant complex and even lower in energy than isolated
plane of symmetry is considerably longer in A than in B
(i.e., 2.231 Å versus 2.175 Å), while the axial Fe-N distance
is virtually the same (compare the structures in Figures 4
5
reactants. This implies that the approach of benzene to 2
a
will rapidly lead to the formation of a radical intermediate
without much activation barrier. This contrasts the energetics
calculated for benzene hydroxylation by FedO(Por)SH,
whereby comparable barriers leading to radical and cationic
II
5,3
and 5). Since the metal is in oxidation state Fe in A but
III
5,3
in Fe in B, the iron-oxygen bond is longer in A (i.e.,
5
5
1.963 Å in A and 1.792 Å in B). Note as well the short
-
1
19
5
5
intermediates of around 18 kcal mol were obtained.
C-O distance in A as compared to B (i.e., 1.363 Å versus
Inorganic Chemistry, Vol. 46, No. 11, 2007 4637

de Visser et al.
Figure 5. Potential energy surface for the hydroxylation of benzene by 5,3FedO(N4Py)2+ via radical intermediates. The proton shuttle mechanism in the
triplet spin state is in blue, and the one in the quintet spin state is in red. All energies are obtained at the UB3LYP/LACV3P+* level with ZPE corrections
-
1
at UB3LYP/LACVP. Energies are in kcal mol relative to the reactant complex in the triplet spin state. Values in parentheses are ∆E+ZPE energies with
5
,3
solvent corrections. Also shown are optimized geometries of B with bond lengths in Å.
1
.530 Å), which is almost a double bond in A but a single
proton. Geometrically, the proton is at a similar distance to
the oxygen atom as to the C carbon atom: the O-H distance
in C is of the same order of magnitude as the C-H
5
,3
bond in B. Another structural difference between A and
1
5,3
5,3
B is the fact that the arene group is aligned with the Fe-O
bond in the cationic intermediates, whereas it is tilted in the
radical intermediates. Although the triplet spin state is the
ground state in the reactants, the quintet spin-state surface
crosses the triplet spin state before the first transition state,
and in the rest of the reaction mechanism the quintet spin
state stays the ground state. Thus, since the barriers on the
quintet spin-state surface are well lower than those on the
triplet spin-state surface, this implies that the population of
the quintet spin state in the reactants and the spin-state
crossing from the triplet to the quintet spin-state surface
during the reaction determine the speed of the reaction and
the amount of products formed.
distance, so that in principle the proton can be reshuttled to
either the oxygen to form product or back to the C
atom to either intermediate A or B. However, structure
1
carbon
5
,3
5,3
5
,3
5,3
C is well lower in energy than A (by at least 16.3 kcal
-
1
5,3
-1
mol ) and B (by at least 21.7 kcal mol ), whereas the
formation of product is exothermic. Since, the charge on the
oxygen atom is much more negative than the one on the
carbon atom, the process leading to product formation will
be favorable. Moreover, structures A and B have less
aromaticity in the benzene group than C. Reshuttling the
proton to the oxygen atom reinstates the broken Fe-N bond
to its original geometry. As follows from the geometry scans
5
,3
From the intermediate complexes a direct proton shuttle
to one of the nitrogen atoms of the N4Py unit can take place
in Figure S5, the proton reshuttle barriers ( TSreb) on the
triplet and quintet spin-state surfaces are negligible, which
5
,3
5,3
to form structure C which thereafter leads to the phenol
would imply that C are first-order saddle points; however,
5
,3
product D via a negligible barrier. Figure 6 shows the
frequency calculations confirm them as stable intermediates.
optimized geometries of 5 C and D. Essentially, C has
a phenolate anion bound to iron, and the N4Py ligand has
undergone a dramatic change, namely one of the pyrimidine
rings has lost its Fe-N bond to make space for the incoming
,3
5,3
5,3
Since the reshuttle barrier ( TSreb) is small, the lifetime of
5,3
5
,3
C will be limited, and it will be converted into 5,3D very
quickly. Generally, in the electrophilic substitution reactions
5
,3
the proton shuttle barriers (via TSPT) are lower in energy
4638 Inorganic Chemistry, Vol. 46, No. 11, 2007

Mononuclear Nonheme Iron(IV)-Oxo Complexes
Figure 6. Optimized geometries of 5,3C and 5,3D with bond lengths in Å.
5
,3
5,3
5
than the direct hydrogen transfer barriers ( TSHA), which
confirms the proposed proton shuttle mechanism as the most
likely reaction mechanism. In particular, the proton shuttle
intermediates ( TS
B E
and TS ) already have significant
radical character developed on the substrate moiety, whereas
the FC6H6 value is low in TS , which means that the
A
transition states occur late. Note as well the large spin density
on the N4Py unit as a result of occupation of the σ*xy
orbital with one electron in the radical complexes. Due to
5
,3
5
-1
5
barrier TSPT,B is just 0.8 kcal mol above B at the
LACV3P+* level of theory (Table S3), while ZPE correc-
tions bring it below B by 0.6 kcal mol (Figure 5). The
latter artifact should be seen as an almost barrierless proton
transfer from the ipso position in B to form the proton-
5
-1
5
,3
A
the creation of a cationic complex in TS the charge
5
distributions are quite different from those in 5,3TS
B
5
: QC6H6
5
5
5
transfer intermediate C. Moreover, the quintet pathway in
) 0.52 in TS
A
and 0.23 in TS
B
, QFe ) 0.50 in TS
A
5
and
5
5
Figure 5 shows barely a significant barrier for benzene
hydroxylation, implying that the reaction should proceed
extremely fast without accumulation of intermediates.
Finally, we ran calculations for a direct hydrogen abstrac-
tion from benzene by 5,32 reactants to form a hydroxo-iron
complex (bottom reaction in Scheme 3 above), followed by
rebound of the hydroxo group to the C H • radical to form
6 5
the phenol product. The two hydroxo-iron complexes ( E)
are local minima and are higher in energy than RC by 13.9
quintet) and 20.3 (triplet) kcal mol-1 at the UB3LYP/
0.74 in TS
B
, and Q
O
) -0.25 in TS
A
and -0.44 in TS
B
.
Thus, the FeO group has a smaller positive charge in the
cationic complexes as compared to the radical ones.
Subsequently, we calculated the reactants and the rate
5,3
5,3
5
determining transition states ( TS
A
,
B E
TS , TS ) in a
dielectric constant of ꢀ ) 5.7 (Figures 5-7). Addition of a
dielectric constant to the system increases all barriers and
in particular the ones passing radical intermediates. Some
barriers increase by more than 5 kcal mol , but TS
A
increases by only 2.4 kcal mol . Nevertheless, even while
5,3
3
-1
5
-1
(
5
LACV3P+* level of theory with the ZPE correction
A
TS is less destabilized than the other transition states, the
included. On the triplet spin-state surface we did not manage
ordering of the various transition states stays the same. The
high barriers in an environment with a large dielectric
constant may explain why experimentally the reaction
3
to locate a hydrogen abstraction barrier ( TS
E
), which implies
3
that E is a shallow minimum, while on the quintet spin-
5
-1
2+
state surface a barrier ( TS
E
) of 15.5 kcal mol is obtained.
between FedO(N4Py) and benzene did not occur. Nev-
5,3
5
The subsequent radical rebound process from E is highly
ertheless, our estimated free energy of activation of TS
B
of
-1
endothermic especially on the triplet spin-state surface
16.1 kcal mol obtained at UB3LYP/LACV3P+* with ZPE,
thermal, and solvent (ꢀ ) 5.7) corrections at the LACVP
level of theory is in very good agreement with the experi-
(Figure S21). Moreover, the direct hydrogen abstraction
barriers are well higher in energy than the electrophilic
5
-1
substitution barrier TS
B
; therefore, the hydrogen abstraction
mental free energy of activation of 19.6 kcal mol at 298 K
mechanism will not be able to compete with the electrophilic
substitution mechanism.
(vide supra) although a different substrate was used. This is
further evidence on the reliability and accuracy of the DFT/
UB3LYP methodology for calculating these types of difficult
reaction mechanisms.
So what is it that determines the relative energies of the
reaction barriers, and how do external perturbations or
hydrogen by deuterium substitution influence the energetics?
To answer these questions, consider first in Figure 7 the
Also shown in Figure 7 are the kinetic isotope effects (k
, KIE) for the reaction with C versus C as calculated
with the Eyring (KIE ) and Wigner (KIE ) models (see
H
/
k
D
H
6 6
6 6
D
5
,3
geometries of the rate determining transition states ( TS
A
,
E
W
5
,3
5
TS
B
, and TS
E
) with bond lengths in Å and group spin
Methods section). Thus, the electrophilic substitution transi-
tion states give inverse kinetic isotope effects, while the
densities identified. The transition states leading to the radical
Inorganic Chemistry, Vol. 46, No. 11, 2007 4639

de Visser et al.
Figure 7. Optimized geometries of 5,3TSA, 5,3TSB, and TSE with bond lengths in Å. The imaginary frequency of the transition state is given in cm-1. Also
5
-
1
3
shown are group spin densities (F) as taken from the UB3LYP/LACV3P+* data and relative free energies (∆G in kcal mol relative to RC) in the gas
phase and under the influence of a dielectric constant of ꢀ ) 5.7. Finally, the kinetic isotope effects of replacing C6H6 by C6D6 using the Eyring (KIEE) and
Wigner (KIEW) models are given.
hydrogen abstraction barrier gives typical values of KIE
E
)
oxoiron complexes have been performed. Experimental
evidence (e.g., KIE value of <1) demonstrates unambigu-
ously that the reaction does not proceed via an initial
hydrogen abstraction mechanism. Instead, the aromatic
hydroxylation proceeds via an electrophilic substitution
reaction, whereby either a radical or cationic intermediate
is formed. Detailed theoretical calculations on various
possible reaction mechanisms confirm the electrophilic
substitution mechanism via radical intermediates as the most
likely pathway. Theory has provided further insights and
shown that the ligand is noninnocent and able to reshuttle
the proton from the substrate via the ligand to the oxo group.
1
0.5 and KIE ) 13.2. Our calculated KIE values of the
W
electrophilic substitution reactions match the experimentally
obtained KIEs perfectly (compare Table 1 and Figure 7) and
point to an electrophilic substitution pathway as the rate
determining step in the reaction in agreement with the
thermodynamics shown in Figures 4 and 5.
In summary, theory predicts that the reaction mechanism
for arene hydroxylation by FedO(N4Py)2+ proceeds via an
electrophilic substitution reaction leading to radical interme-
diates and a subsequent proton-shuttle mechanism similarly
to the one predicted for oxoiron heme catalysts will convert
IV
the radical intermediates into phenol. Theory predicts [Fe -
N4Py)(O)]2 to react with benzene via two-state reactivity
+
Methods
(
on competing triplet and quintet surfaces, whereby the quintet
spin-state surface seems more reactive. However, since the
reactant state is most likely to be in a dominant triplet spin
state, the rate determining step in the reaction will be the
spin-crossing from the triplet to the quintet spin-state surface.
Thus, the spin-state crossing will involve the transfer of a
â-spin electron in π*x2-y2 into an R-spin electron in σ*xy,
and since these orbitals are located on the same center, it is
expected that it will occur fast. Once the spin-state crossing
has occurred and either a cationic or radical intermediate
Materials and Instrumentation. Acetonitrile (anhydrous) and
dichloromethane (anhydrous) were obtained from Aldrich Chemical
Co. and purified by distillation over CaH prior to use. All reagents
2
purchased from Aldrich were the best available purity and used
without further purification unless otherwise indicated. Iodosyl-
benzene was prepared by a literature method.27
H
18
O (95% O-
enriched) was purchased from ICON Services Inc. (Summit, NJ).
Fe(Bn-tpen)(CF SO and Fe(N4Py)(CF SO were synthesized
according to published procedures.
18
2
3
3
)
2
3
3 2
)
28
UV-vis spectra were recorded on a Hewlett-Packard 8453
DN
spectrophotometer equipped with Optostat variable-temperature
5
5
(
A or B) is formed, the reaction will proceed rapidly via a
liquid-nitrogen cryostat (Oxford Instruments). Product Analysis for
the anthracene hydroxylation was performed on Agilent Technolo-
gies 6890N gas chromatography equipped with a FID detector (GC)
and Thermo Finnigan Focus DSQ mass spectrometer interfaced with
proton shuttle mechanism via one of the nitrogen atoms of
the N4Py ligand that is used as a springboard to transfer the
proton toward the oxygen atom. This mechanism is found
to be more efficient than a direct hydrogen atom transfer.
(
27) Organic Syntheses; Saltzman, H., Sharefkin, J. G., Eds.; Wiley: New
York, 1973; Collect. Vol. V, p 658.
28) (a) Duelund, L.; Hazell, R.; McKenzie, C. J.; Nielsen, L. P.; Toftlund,
H. J. Chem. Soc., Dalton Trans. 2001, 152-156. (b) Lubben, M.;
Meetsma, A.; Wilkinson, E. C.; Feringa, B.; Que, L., Jr. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1512-1514.
Summary
(
Combined experimental and theoretical (DFT) studies into
the reaction mechanism of arene hydroxylation by nonheme
4640 Inorganic Chemistry, Vol. 46, No. 11, 2007

Mononuclear Nonheme Iron(IV)-Oxo Complexes
Focus gas chromatography (GC-MS). Electrospray ionization mass
spectra (ESI MS) of intermediates were collected on a Thermo
Finnigan (San Jose, CA) LCQ Advantage MAX quadrupole ion
trap instrument, by infusing samples directly into the source at 25
µL/min using a syringe pump. The spray voltage was set at 4.7 kV
and the capillary temperature at 220 °C.
atoms.30 An analytical frequency calculation in Gaussian-03
confirmed the nature of the stationary points: all transition states
are characterized by a single imaginary frequency for the correct
mode, and all intermediates, reactants, and products have real
frequencies only.32 In the past we showed that frequencies calculated
with Gaussian and Jaguar give identical results; therefore, we
exported the optimized geometries from Jaguar to Gaussian for an
Kinetic Studies and Product Analysis. In general, [Fe(IV)-
Bn-tpen)(O)]2 (1 mM) and [Fe(IV)(N4Py)(O)] (1 mM) were
+
2+
analytical frequency analysis. All energies reported in this work
33
(
prepared by stirring Fe(Bn-tpen)(CF
respectively, with excess solid PhIO in a solvent mixture of CH
CN and CH Cl (1:1) at 25 °C. The reaction solutions were filtered
through a 0.45-µM filter to remove solid PhIO remained. Then,
2
appropriate amounts of anthracene (diluted in 150 µL of CH Cl )
were added to the resulting solution, and spectral changes of the
intermediates were directly monitored by a UV-vis spectropho-
tometer. Rate constants, kobs, were determined by pseudo-first-order
fitting of the decrease of absorption bands at 737 nm for 1 and
3
SO
3
)
2
and Fe(N4Py)(CF
3
SO
3
)
2
,
-
are energies obtained with the LACV3P+* basis set corrected with
zero-point energies at LACVP level unless otherwise stated. To
explore the potential energy surface, we initially ran extensive
geometry scans (see the Supporting Information) between the
various local minima whereby one degree of freedom was fixed.
The maxima of these geometry scans were used as a starting point
for the transition state calculations. To estimate the effect of the
environment on the relative energies of the transition states we ran
single point calculations in Jaguar using the Self-Consistent
Reaction Field (SCRF) model on the rate determining transition
states.30 We used a typical dielectric constant of ꢀ ) 5.7 and a
probe radius r ) 2.67 Å mimicking chlorobenzene since this
dielectric constant falls midway in the range of experimentally used
solvents. Kinetic isotope effect studies for the replacement of all
hydrogen atoms in benzene by deuterium atoms were studied using
3
2
2
2
695 nm for 2. All the rate constants are averages of at least three
determinations.
Product analysis was performed by injecting reaction solutions
directly into GC and GC-MS. Product yields were determined by
comparison with standard curves of known authentic samples and
decane as an internal standard. The source of oxygen in oxygenated
18
q
q
products was determined by preparing O-labeled [Fe(IV)(Bn-
the semiclassical Eyring model (KIE
and further tunneling corrections due to Wigner (KIE
applied by multiplying KIE with the tunneling ratio (Qt,H/Qt,D)
E
) exp((∆G
D
- ∆G
H
)/RT))
) were
1
8
2+
18
tpen)( O)] (1- O) and using the intermediate directly in the
W
hydroxylation of anthracene. First, 18O-labeled PhIO was prepared
E
18
2
q
by treating PhIO (5 mM, diluted in 50 µL of CH
10 µL, 0.52 mmol, 95% 18O enriched). Then, PhI O was added
to a solution of Fe(Bn-tpen)(OTf) (1 mM) in a solvent mixture (2
mL) of CH CN and CH Cl (1:1), followed by the addition of
3
OH) with H
2
O
t
whereby Q ) 1 + (hν/kT) /24. In these equations ∆G is the free
18
(
energy of activation, R is the gas constant, T is the temperature
(298 K), h is Planck’s constant, k is Boltzmann’s constant, and ν
is the imaginary frequency in the transition state.34
2
3
2
2
anthracene (15 mM). The resulting solution was directly analyzed
by GC-MS. The 18O percentage in 1-18O was determined by taking
Acknowledgment. This research was supported by the
Creative Research Initiative Program of MOST/KOSEF
(W.N.), the BK 21 Program (K.O. and A.-R.H.), and the
National Service of Computational Chemistry Software
(NSCCS) for providing valuable CPU time (S.P.dV.). We
thank Dr. Mi Sook Seo at Ewha Womans University for
taking ESI MS of intermediates.
an Electrospray Ionization Mass Spectrum (Supporting Information,
Figure S2).20 To increase the accuracy of the product analysis, the
18
hydroxylation of anthracene was performed by adding PhI O (5
mM) into a reaction solution containing Fe(Bn-tpen)(CF SO (1
3
3 2
)
mM) and anthracene (15 mM) under catalytic conditions in a
drybox. Then, the reaction mixture was stirred for 0.5 h at room
temperature. 16O and O compositions in anthraquinone were
analyzed by the relative abundances of the following mass peaks:
m/z ) 208 (M), 210 (M + 2), and 212 (M + 4) for anthraquinone.
Theory. We have studied the reaction of FedO(N4Py)2 with
benzene, by utilizing the unrestricted hybrid density functional
method UB3LYP throughout.29 Initial geometry optimizations and
geometry scans to explore the potential energy surface were done
18
Supporting Information Available: Experimentally obtained
UV-vis spectral changes of 2 (Figure S1), the mass spectrum of
+
18
1- O (Figure S2), tables of first-order rate constants of various
para-X-substituted anthracenes by 1 (Table S1), absolute and
relative energies, group spin densities, charges of all optimized
geometries using either the LACVP or LACV3P+* basis set, and
kinetic isotope effects, five figures with geometry scans and 15
figures with optimized geometries and Cartesian coordinates of all
species mentioned in this work, and ref 32 in full.
30
with the Jaguar 5.5 program package. We used a modest Los
Alamos type double-ú quality LACVP basis set on iron that contains
a core potential, while all other atoms (C, N, O, H) are described
with a 6-31G basis set.31 Subsequent, single point calculations on
the optimized geometries were done with the LACV3P+* basis
set that contains a triple-ú quality basis set with diffuse functions
on iron (LACV3P+) and a 6-311+G* basis set on the rest of the
IC700462H
(32) Frisch, M. J. et al. Gaussian-03: ReVision C.01; see the Supporting
Information for the complete reference citation.
(
33) Ogliaro, F.; Harris, N.; Cohen, S.; Filatov, M.; de Visser, S. P.; Shaik,
S. J. Am. Chem. Soc. 2000, 122, 8977-8989.
(
29) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652. (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
(34) (a) Kumar, D.; de Visser, S. P.; Sharma, P. K.; Cohen, S.; Shaik, S.
J. Am. Chem. Soc. 2004, 126, 1907-1920. (b) de Visser, S. P. J. Biol.
Inorg. Chem. 2006, 11, 168-178.
(
30) Jaguar 5.5; Schr o¨ dinger, LLC: Portland, OR, 2003.
31) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299-310.
(
Inorganic Chemistry, Vol. 46, No. 11, 2007 4641