Iron catalyzed competitive olefin oxidation and ipso-hydroxylation of benzoic acids: Further evidence for an FeV=O oxidant

Article abstract of DOI:10.1021/ic101144s

The iron complex [Fe^II(TPA)(CH₃CN) ₂](OTf)₂ (1) [TPA = tris(2-pyridylmethyl)amine] with H₂O₂ as an oxidant performs ipso-hydroxylation of electron-withdrawing benzoic acids at room temperature, leading to multiple turnovers of corresponding phenols. ipso-Hydroxylation competes with olefin epoxidation and cis-dihydroxylation in the presence of olefins, with the product ratios being modulated by the relative amounts of benzoic acid, olefin, and water. It is proposed that benzoic acid and water compete for the available sixth site on the [(TPA)Fe^III(OOH)] intermediate, which undergoes O-O bond heterolysis to form, respectively, the Fe^V(O)(O₂CAr) and Fe^V(O)(OH) oxidants that determine the product outcome. The putative Fe^V(O)(O₂CAr) oxidant decays either by undergoing oxidative decarboxylation and subsequent ipso-hydroxylation to form the observed phenol product or by oxo-transfer to olefins to form epoxide. The observed higher yield of phenol over epoxide or cis-diol in all cases studied, where an electronwithdrawing benzoic acid is present in the reaction mixture, suggests that intramolecular decay of the putative Fe^V(O)-(O ₂CAr) oxidant is favored over intermolecular olefin oxidation. These results support the mechanistic framework postulated for Fe(TPA) oxidative catalysis and further strengthens the notion that oxoiron(V) species are the key oxidants in these reactions.

Full text of DOI:10.1021/ic101144s

Inorg. Chem. 2010, 49, 9479–9485 9479
DOI: 10.1021/ic101144s
Iron Catalyzed Competitive Olefin Oxidation and ipso-Hydroxylation
of Benzoic Acids: Further Evidence for an FeVdO Oxidant
Parthapratim Das and Lawrence Que, Jr.*
Department of Chemistry and Center for Metals in Biocatalysis, University of Minnesota,
207 Pleasant Street SE, Minneapolis, Minnesota 55455
Received June 7, 2010
The iron complex [FeII(TPA)(CH3CN)2](OTf)2 (1) [TPA = tris(2-pyridylmethyl)amine] with H2O2 as an oxidant
performs ipso-hydroxylation of electron-withdrawing benzoic acids at room temperature, leading to multiple turnovers
of corresponding phenols. ipso-Hydroxylation competes with olefin epoxidation and cis-dihydroxylation in the presence
of olefins, with the product ratios being modulated by the relative amounts of benzoic acid, olefin, and water. It is
proposed that benzoic acid and water compete for the available sixth site on the [(TPA)FeIII(OOH)] intermediate,
which undergoes O-O bond heterolysis to form, respectively, the FeV(O)(O2CAr) and FeV(O)(OH) oxidants that
determine the product outcome. The putative FeV(O)(O2CAr) oxidant decays either by undergoing oxidative
decarboxylation and subsequent ipso-hydroxylation to form the observed phenol product or by oxo-transfer to olefins
to form epoxide. The observed higher yield of phenol over epoxide or cis-diol in all cases studied, where an electron-
withdrawing benzoic acid is present in the reaction mixture, suggests that intramolecular decay of the putative FeV(O)-
(O2CAr) oxidant is favored over intermolecular olefin oxidation. These results support the mechanistic framework
postulated for Fe(TPA) oxidative catalysis and further strengthens the notion that oxoiron(V) species are the key
oxidants in these reactions.
Introduction
by a nonbiological iron center. These catalysts used H2O2 as
oxidant in place of the combination of O2, electrons, and protons
required in the enzyme reactions. Complex 1 and other non-
heme iron complexes have also been shown to use H2O2 as
oxidant for a variety of other reactions8 including the hydroxyl-
ation of alkanes,9-15 the epoxidation of olefins,12,16-20 and
Nonheme iron oxygenases have emerged as a very important
group of enzymes that perform a variety of interesting oxidative
transformations.1-3 Of particular interest is the family of Rieske
dioxygenases4,5 that catalyze the cis-dihydroxylation of arene
double bonds in the biodegradation of aromatic molecules
by soil bacteria.1,6,7 Such a transformation was unique to a
biological iron center, until quite recently. We initiated an
effort more than ten years ago to develop bioinspired non-
heme iron complexes that can carry out the cis-dihydroxyla-
tion of CdC double bonds and found a family of complexes
exemplified by [FeII(TPA)(MeCN)2]2þ(1) that represented
the first examples of catalysts for olefin cis-dihydroxylation
(8) Chen, K.; Costas, M.; Que, L., Jr. J. Chem. Soc., Dalton Trans. 2002,
672–679.
(9) Chen, K.; Que, L., Jr. J. Am. Chem. Soc. 2001, 123, 6327–6337.
(10) Chen, K.; Que, L., Jr. Chem. Commun. 1999, 1375–1376.
(11) Company, A.; Gomez, L.; Guell, M.; Ribas, X.; Luis, J. M.; Que, L.,
Jr.; Costas, M. J. Am. Chem. Soc. 2007, 129, 15766–15767.
ꢀ
(12) Company, A.; Gomez, L.; Fontrodona, X.; Ribas, X.; Costas, M.
Chem. Eur. J. 2008, 14, 5727–5731.
;
(13) Britovsek, G. J. P.; England, J.; White, A. J. P. Inorg. Chem. 2005, 44,
8125–8134.
*To whom correspondence should be addressed. E-mail: larryque@umn.
edu.
(1) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr. Chem. Rev. 2004,
104, 939–986.
(2) Abu-Omar, M. M.; Loaiza, A.; Hontzeas, N. Chem. Rev. 2005, 105,
2227–2252.
(3) Bruijnincx, P. C. A.; Koten, G. v.; Gebbink, R. J. M. K. Chem. Soc.
Rev. 2008, 37, 2716–2744.
(4) Kauppi, B.; Lee, K.; Carredano, E.; Parales, R. E.; Gibson, D. T.;
Eklund, H.; Ramaswamy, S. Structure 1998, 6, 571–586.
(5) Karlsson, A.; Parales, J. V.; Parales, R. E.; Gibson, D. T.; Eklund, H.;
Ramaswamy, S. Science 2003, 299, 1039–1042.
(6) Gibson, D. T., Ed. Microbial Degradation of Organic Compounds;
Marcel Dekker: New York, 1984.
(7) Wolfe, M. D.; Lipscomb, J. D. J. Biol. Chem. 2003, 278, 829–835.
(14) Britovsek, G. J. P.; England, J.; White, A. J. P. Dalton Trans. 2006,
1399–1408.
(15) Britovsek, G. J. P.; England, J.; Spitzmesser, S. K.; White, A. J. P.;
Williams, D. J. Dalton Trans. 2005, 945–955.
ꢀ
(16) Mas-Balleste, R.; Costas, M.; Berg, T. v. d.; Que, L. J. Chem. Eur.
;
J. 2006, 12, 7489–7500.
(17) Chen, K.; Costas, M.; Kim, J.; Tipton, A. K.; Que, L., Jr. J. Am.
Chem. Soc. 2002, 124, 3026–3035.
(18) Chen, K.; Que, L., Jr. Angew. Chem., Int. Ed. 1999, 38, 2227–2229.
€
(19) Company, A.; Feng, Y.; Guell, M.; Ribas, X.; Luis, J. M.; Que, L.,
Jr.; Costas, M. Chem. Eur. J. 2009, 15, 3359–3362.
;
(20) Bruijnincx, P. C. A.; Buurmans, I. L. C.; Gosiewska, S.; Moelands,
M. A. H.; Lutz, M.; Spek, A. L.; Koten, G. v.; Klein Gebbink, R. J. M.
Chem. Eur. J. 2008, 14, 1228–1237.
;
r
2010 American Chemical Society
Published on Web 09/24/2010
pubs.acs.org/IC

9480 Inorganic Chemistry, Vol. 49, No. 20, 2010
Das and Que
the hydroxylation of arenes.21-25 Very recently the 1/H2O2
combination was shown to carry out the cis-dihydroxylation
of naphthalene to cis-1,2-dihydro-1,2-naphthalenediol, mak-
ing it the first functional model of the enzyme naphthalene
1,2-dioxygenase.21
In this paper, we focus on the ipso-hydroxylation of benzoic
acids catalyzed by 1, its relationship with olefin epoxidation
and cis-dihydroxylation, and mechanistic insights derived
therefrom. Complex 1 has been shown to hydroxylate benzoic
acids regioselectively with H2O2 as an oxidant. Depending on
the position of the substituent in the arene ring, the benzoic
acid can undergo either ortho-hydroxylation (forming salic-
ylates) or oxidative decarboxylation followed by hydroxyla-
tion at the ipso-carbon (making phenols).22 The salicylates
obtained from ortho-hydroxylation are excellent bidentate
ligands and form stable iron(III) complexes that exclude the
possibility of catalytic turnover. However, as reported in this
paper, the phenolates obtained from ipso-hydroxylation are
more susceptible to displacement by substrate, and multiple
turnovers of phenol formation are observed for benzoic acids
with electron withdrawing substituents. Furthermore, experi-
ments in the presence of 1-octene or tert-butyl acrylate show
that ipso-hydroxylation and olefin oxidation are competitive
reactions and provide further support for the FeVdO oxidant
proposed for 1-catalyzed oxidations.
Figure 1. Phenol formation catalyzed 1 (1 mM) with 10 equivalents of
H2O2 as oxidant in the presence of 10 (lavender-blue), 25 (maroon), or 50
(yellow) equivalents of benzoic acids at room temperature. TON (turn
over number) was calculated as moles of product/mol of the catalyst.
Results and Discussion
[FeII(TPA)(CH3CN)2](OTf)2 (1) has been shown to be an
effective olefin oxidation catalyst to produce cis-diol and epoxide
with H2O2 as the oxidant.17 The 1/H2O2 combination has also
been shown to attack a variety of benzoic acids and effect
either ortho-hydroxylation (producing salicylates) or ipso-
hydroxylation (producing phenolates),22 depending on the
position of the substituent in the aromatic ring. 3-Substituted
benzoic acids are converted to salicylates, while 2- and
4-substituted ones produce phenolates as a result of oxidative
decarboxylation and ipso-hydroxylation. Further explora-
tions of this chemistry, presented in this paper with benzoic
acids having electron-withdrawing substituents, reveal that
ipso-hydroxylation can be catalytic.
Figure 1 shows the results of experiments with 1 as catalyst,
10 equiv of H2O2, and varying amounts of benzoic acid. We
chose to study the oxidations of perfluoro-, 2-nitro-, and
2-chlorobenzoic acids, which have respective pKa’s of 1.88,
2.19, and 2.92. The yields of the corresponding ipso-hydroxyl-
ated products increased with a larger excess of the benzoic
acid. As much as 75% of the H2O2 was converted to phenol
product. The phenol products were readily identified by their
characteristic GC retention times and their mass spectral
patterns from GC-MS analysis. This is the first example of
the catalytic production of phenol from benzoic acids from
an iron catalyst and the environmentally friendly oxidant
H2O2 at ambient temperature and pressure. However, much
lower phenol yields were observed for benzoic acids with
electron donating substituents due to overoxidation of the
resulting phenols. For example, in the 1-catalyzed oxidation
of 2,6-dimethylbenzoic acid, the products found were 0.5
equiv of 2,6-dimethylphenol and 1 equiv of 2,6-dimethylbenzo-
quinone. As a consequence, our subsequent experiments
focused only on the benzoic acids with electron withdrawing
substituents. The observed increasedyield of phenol products
in Figure 1 with higher concentrations of benzoic acid reflects
pre-equilibrium binding of the benzoic acid to iron(III)-
hydroperoxo species A prior to O-O bond cleavage that
leads to the formation of C, as proposed in Scheme 1.
To gain further insight into the relative reactivities of the
electron withdrawing benzoic acids with respect to 1/H2O2,
reactions were performed involving pairwise combinations of
perfluoro-, 2-nitro-, and 2-chlorobenzoic acids. These com-
petition experiments (Figure 2, Table 1) showed that 2-chloro-
benzoic acid was the easiest to oxidize and perfluorobenzoic
Experimental Section
Materials and Methods. All chemicals and solvents were pur-
chased from Aldrich and were used without additional puri-
fication unless otherwise noted. CH3CN solvent was dried over
CaH2 before use. H218O (97% 18O-enriched) was obtained from
Cambridge Isotope Laboratories Inc. (Andover, MA). H218O2
(90% 18O-enriched) was obtained from Isotec (Sigma-Aldrich)
Inc. The complex [FeII(TPA)(CH3CN)2](OTf)2 (1) was prepared
in an anaerobic glovebox according to the published procedures.26
Product analyses were performed on a Perkin-Elmer Sigma 3 gas
chromatograph (AT-1701 column, 30 m) with a flame ionization
detector. GC-MS experiments were carried using an HP 6890
gas chromatograph (HP-5 column, 30 m) with an Agilent 5973
mass detector.
Catalytic Studies. Catalytic studies were performed using
0.2 mL of a 10 mM solution of 1 along with 0.287 mL of a 70 mM
solution of H2O2 (10 equiv) at room temperature in acetonitrile
where the final concentration of 1 in the solution was 1 mM.
H2O2 was added using a syringe pump over 25 min with an addi-
tional 5 min of stirring. Ten, 25, or 50 equivalents (relative to 1)
of perfluorobenzoic, 2-nitrobenzoic, or 2-chlorobenzoic acid
were added in various experiments. For competition experi-
ments with olefins, various amounts of olefin were added prior
to the introduction of H2O2. Each catalytic result reported repre-
sented the average of at least three experiments.
(21) Feng, Y.; Ke, C.-y.; Xue, G.; Que, L., Jr. Chem. Commun. 2009,
50–52.
ꢀ
(22) Makhlynets, O. V.; Das, P.; Taktak, S.; Flook, M.; Mas-Balleste, R.;
Rybak-Akimova, E. V.; Que, J., L Chem. Eur. J. 2009, 15, 13171–13180.
;
ꢀ
(23) Alonso, D. A.; Najera, C.; Pastor, I. M.; Yus, M. Chem. Eur. J.
;
2010, 16, 5274–5284.
(24) 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.
(25) Taktak, S.; Flook, M.; Foxman, B. M.; Que, L., Jr.; Rybak-Akimova,
E. V. Chem. Commun. 2005, 5301–5303.
(26) Lim, M. H.; Rohde, J.-U.; Stubna, A.; Bukowski, M. R.; Costas, M.;
€
Ho, R. Y. N.; Munck, E.; Nam, W.; Que, L., Jr. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 3665–3670.

Article
Inorganic Chemistry, Vol. 49, No. 20, 2010 9481
Figure 2. Competition reactions between any two of perfluoro- (A),
2-nitro- (B), and 2-chlorobenzoic acid (C) (25 equiv each with respectto1)
and 1/H2O2 (1/10) at room temperature. Data for the reactions of 1/H2O2
and the individual benzoic acids (25 equiv with respect to 1) are also shown
for comparison. Black, gray, and red blocks depict amounts of per-
fluorophenol, 2-nitrophenol, and 2-chlorophenol, respectively.
Scheme 1. Proposed Pre-Equilibrium Binding of the Benzoic Acid to
the FeIII-OOH Intermediate
Table 1. 1-Catalyzed Competitive ipso-Hydroxylation of Benzoic Acidsa
Figure 3. 1-Octene (A) and tert-butyl acrylate (B) oxidation by H2O2
with 1 mM 1 as catalyst at room temperature (1/H2O2/1-octene = 1/10/100)
in the presence of different amounts of substituted benzoic acids. Green,
blue, and yellow blocks represent phenol, epoxide, and cis-diol products,
respectively.
benzoic acid
perfluorophenol
2-nitrophenol
2-chlorophenol
A
4.4 (1)
2.8 (2)
A þ B
3.3 (2)
5.1 (1)
2.9 (1)
B
B þ C
C
A þ C
3.1 (3)
6.7 (4)
3.6 (3)
1.8 (1)
effect of aromatic acids (e.g., benzoic acids) on olefin
oxidation catalysis has not been investigated. It would be
interesting to know how the addition of benzoic acids might
affect the epoxide/cis-diol ratio and whether the oxidation of
the arene ring was competitive with olefin oxidation.
a Reaction conditions: 1 (1 mM), 10 equiv H2O2, 25 equiv of each
benzoic acid designated in acetonitrile at room temperature. A, B, and C
represent perfluoro-, 2-nitro-, and 2-chlorobenzoic acid, respectively.
acid was the hardest, consistent with the cumulative electron-
withdrawing abilities of the substituents, as reflected by the
pKa’s of the three acids. This trend may be rationalized by
either (a) a lower affinity of the stronger acid for FeIII-OOH
(Scheme 1) and/or (b) a greater difficulty in oxidizing the
benzoate with more electron withdrawing substituents.
Competition between Olefin Oxidation and ipso-Hydroxyl-
ation. Complex 1 has been previously shown to be an effec-
tive catalyst for olefin epoxidation and cis-dihydroxylation
using H2O2 as an oxidant.17,27,28 Furthermore, the addition
of acetic acid to the aforementioned catalytic system led to
significantly increased formation of epoxide at the expense
of cis-diol.27,29 It was proposed that the presence of acetic
acid changed the nature of the active oxidant in the catalytic
reaction from FeV(O)(OH), where one or both oxygen atoms
could be transferred to the olefin substrate, to FeV(O)(OAc),
where only oxo-atom transfer could occur.29 However, the
Figure 3 presents the results of catalytic olefin oxidation
experiments by 1/H2O2 carried out with different amounts
of added benzoic acid in the presence of a 100-fold excess
of 1-octene or tert-butyl acrylate (relative to 1); these
results are also tabulated in Table 2. Olefin oxidation
products (cis-diol and epoxide for 1-octene and cis-diol
for tert-butyl acrylate) were observed in the reactions
along with multiple turnovers of phenol, clearly suggest-
ing that ipso-hydroxylation of the benzoic acids competes
with olefin oxidation. Adding more benzoic acid in-
creased the yield of phenol at the expense of the olefin
oxidation products. The results of the competition experi-
ments are similar for both 1-octene and tert-butyl acrylate
(Figure 3), despite the fact that two olefins are quite differ-
ent in terms of their electronic properties. Indeed, in a
previously reported competitive oxidation of 1-octene
and tert-butyl acrylate in the absence of added benzoic
acid, 1-octene oxidation products were favored by 4:1
over those of tert-butyl acrylate, indicating that 1-octene
is the olefin more preferably oxidized by the 1/H2O2 com-
bination.28 However, in all experiments presented in
Figure 3, phenols were the dominant products. These results
demonstrate that ipso-hydroxylation of the benzoic acids
ꢀ
(27) Mas-Balleste, R.; Fujita, M.; Hemmila, C.; Que, L., Jr. J. Mol. Catal.
A: Chem. 2006, 251, 49–53.
(28) Fujita, M.; Costas, M.; Que, L., Jr. J. Am. Chem. Soc. 2003, 125,
9912–9913.
ꢀ
(29) Mas-Balleste, R.; Que, L., Jr. J. Am. Chem. Soc. 2007, 129, 15964–
15972.

9482 Inorganic Chemistry, Vol. 49, No. 20, 2010
Das and Que
Table 2. Competitive Olefin Oxidation and ipso-Hydroxylation Experiments Catalyzed by 1a
10 equiv acid
25 equiv acid
epoxide
50 equiv acid
benzoic acid
perfluoro
1-octene
phenol
epoxide
0.5 (1)
diol
phenol
diol
phenol
epoxide
0.4 (1)
diol
0
100
250
500
1000
0
100
250
500
1000
0
2.7 (2)
4.0 (2)
4.4 (1)
4.2 (3)
3.4 (3)
2.6 (2)
1.8 (2)
5.1 (1)
4.2 (3)
3.8 (4)
3.3 (3)
2.9 (4)
6.7 (4)
6.8 (2)
7.1 (5)
7.2 (3)
6.8 (4)
7.4 (3)
4.3 (3)
1.4 (1)
0.4 (1)
1.2 (1)
1.8 (4)
2.0 (3)
0.9 (1)
2.1 (2)
1.7 (2)
1.5 (5)
1.0 (4)
4.4 (2)
3.6 (2)
5.8 (1)
4.6 (2)
0.3 (2)
1.1 (4)
3.2 (4)
0.2 (1)
0.6 (2)
1.2 (3)
1.5 (5)
1.1 (2)
1.0 (3)
1.0 (3)
0.9 (3)
0
0
0.4 (2)
0.5 (1)
2-nitro
3.6 (2)
4.8 (2)
7.8 (1)
8.0 (3)
100
250
500
1000
0
0
0
0
0
1.0 (2)
1.0 (1)
1.3 (2)
1.5 (3)
2-chloro
10 equiv acid
25 equiv acid
50 equiv acid
epoxide
tert-butyl acrylate
phenol
epoxide
diol
phenol
4.4 (1)
5.4 (3)
5.1 (1)
4.1 (2)
6.7 (4)
6.4 (4)
epoxide
diol
phenol
diol
0
100
0
100
0
2.7 (2)
4.1 (1)
4.4 (2)
3.3 (1)
3.6 (2)
3.7 (4)
7.4 (3)
6.2 (2)
5.8 (1)
4.8 (1)
7.8 (1)
7.8 (2)
perfluoro
2-nitro
0
0
0
3.1 (4)
1.9 (4)
2.3 (5)
0
0
0
2.1 (1)
1.2 (5)
1.0 (5)
0
0
0
1.5 (1)
0.5 (2)
0.6 (3)
2-chloro
100
a Competitive oxidations carried out in acetonitrile at room temperature with 1 mM 1, 10 equiv H2O2, and 100 equiv olefin in the presence of various
amounts of the listed benzoic acid. H2O2 was delivered by a syringe pump over 25 min with an additional 5 min of stirring before the reaction was worked
up. Values listed reflect the average of at least three experiments. A TON value of 0 is listed for experiments where the actual TON is <0.1.
is favored over olefin oxidation, despite the lower concen-
tration of the benzoic acid. This advantage may derive from
the likely intramolecular nature of the ipso-hydroxylation
versus the intermolecular nature of the olefin oxidation.
Figure 4 and Table 2 present results of experiments
where the amount of olefin added was varied from 100 to
1000 equivalents in the presence of 25 equiv benzoic acid,
resulting in an increase in the yields of olefin oxidation
products. For 2-chlorobenzoic acid (Figure 4C), the amount
of 2-chlorophenol formed remained constant at about 6.9
TON, while the amount of cis-diol product increased from
1 to 1.5 TON; however, less than 0.1 TON epoxide was
observed. For 2-nitrobenzoic acid (Figure 4B), the amount
of cis-diol product remained approximately constant at
1 TON, but the yield of epoxide increased at the expense
of phenol, with the combined amount of phenol and epoxide
holding constant at about 4.4 TON. In this series of experi-
ments, the diol-to-epoxide ratio of 7 observed in the absence
of added acid decreased to 0.6 upon addition of 1000 equiv
1-octene (Table 2). A similar behavior was observed for
perfluorobenzoic acid (Figure 4A), with epoxide yield
increasing at the expense of phenol and the combined
amount of phenol and epoxide remaining at about 4 TON.
Interestingly for this acid, the amount of cis-diol formed
increased from 0.9 TON at 100 equiv 1-octene to 2.1 TON
at 250 equiv, which then decreased with more added
olefin. Similar to that observed for 2-nitrobenzoic acid,
the diol-to-epoxide ratio in the presence of 1000 equiv of
1-octene decreased from 7 in the absence of acid to 0.8 in
its presence. These results demonstrate the likely partici-
pation of multiple equilibria and oxidation pathways that
determine the outcome of the catalysis experiments.
The preference of 1-catalyzed oxidation for ipso-hydrox-
ylation could be counteracted by the addition of water.
Experiments carried out by introducing varying amounts
of water at a fixed concentration of acid and olefin
(Figure 5) showed that water competed with the benzoic
acid in determining the nature of the products. The results
indicated that higher amounts of water in the medium
increased the yield of cis-diol at the expense of phenol
formation, while epoxide yield remained relatively low in
all experiments, unlike the results presented in Figure 4.
There would appear to be multiple equilibria that affect
the outcome of these catalytic oxidation experiments.
Labeling Studies. 18O labeling experiments have previ-
ously provided important insights into the mechanism of
olefin oxidation by the FeIITPA catalyst family17 and are
also useful in the present case for discerning the reaction
pathways available. We showed previously that 1-cata-
lyzed olefin oxidations afforded epoxide products, where
about 10% of the incorporated oxygen atom was derived
from H2O and the other 90% was from H2O2, and cis-diol
products where one O atom was derived from H2O and
the other was from H2O2.17 The incorporation of label
from H2O implicated a water-assisted mechanism for
O-O bond cleavage. By comparing the 18O labeling results
in the presence or absence of perfluorobenzoic acid, it should
be possible to discern what effect the added benzoic acid
has on the mechanism of 1-catalyzed oxidation.
The 18O labeling results are listed in Table 3. As previ-
ously reported, the addition of increasing amounts of
H218O resulted in a greater incorporation of 18O into the
cis-diol and epoxide products in the absence of added
benzoic acid. In the presence of 25 equiv of C6F5CO2H,
the labeling outcome for the cis-diol was similar; however,
much less 18O from H218O was incorporated into the epoxide
product, as previously reported for 1-catalyzed epoxida-
tions in the presence of acetic acid.29 Interestingly, in the

Article
Inorganic Chemistry, Vol. 49, No. 20, 2010 9483
by H218O2 was essentially quantitative, implicating a
common oxidant that effects these two oxidative trans-
formations but is different from the oxidant that carries
out the cis-dihydroxylation.
Mechanistic Insights. Scheme 2 depicts our current
understanding of the mechanistic landscape for 1-cata-
lyzed olefin oxidation and ipso-hydroxylation, which
takes into account the results from this paper and from
our previous work on olefin oxidation17,29 and arene ring
oxidation.22 The iron(III)-hydroperoxo species (P1), gen-
erated from the reaction of 1 and H2O2, has been trapped
and characterized in detail by various spectroscopic tech-
niques, namely UV-vis,30 electrospray ionization (ESI)-
MS,30 EPR,30,31 and resonance Raman spectroscopy.31,32
We propose that intermediate P1 is the branching point in
the reaction of 1 and H2O2 leading to the formation of
various high-valent iron oxidants. The eventual fate of P1
is dependent on what is bound at the labile sixth site and
can be tracked by monitoring the percent 18O incorpora-
tion into the products. The addition of a large excess of
H2O results in the displacement of the CH3CN ligand on
P1 to form the aquated intermediate P2. Decay of P2 via
a water-assisted heterolytic O-O bond cleavage step
affords the FeV(O)(OH) (O2) oxidant where the oxo atom
derives from H2O2 and the hydroxo O atom derives from
H2O. These notions are corroborated by 18O-labeling
experiments (Table 3). Olefin oxidation in the presence
of 1000 equiv H218O results in the incorporation of
18O label into the oxidation products with 9% into the
epoxide and 77% into one oxygen atom of the cis-diol.
The complementary labeling experiment carried out with
a 2% solution of H218O2 (containing 100 equiv H216O per
equiv H218O2 added) confirms the H218O result, where the
epoxide product is 82% labeled and 92% of the cis-diol is
singly labeled.
Figure 4. Oxidation of perfluorobenzoic acid (A), 2-nitrobenzoic acid
(B), and 2-chlorobenzoic acid (C) catalyzed by 1 (1 mM) with H2O2 as
oxidant in the presence of variable amounts of 1-octene at room temper-
ature (1/H2O2/substituted benzoic acid = 1/10/25). Yellow, blue, and green
blocks represent cis-diol, epoxide, and phenol, respectively.
As the amount of excess H2O is decreased, the extent of
water incorporation into the olefin oxidation products
diminishes, as expected. For example, with 20 equiv added
H218O (per H2O2 used), there is a significant fraction
(∼45%) of the cis-diol that is unlabeled (Table 3), imply-
ing that both oxygen atoms of H2O2 must be incorporated
into the cis-diol for this subpopulation. Thus, P1 itself
must be involved in cis-dihydroxylation via a nonwater-
assisted mechanism that cleaves the O-O bond of the
iron(III)-hydroperoxo intermediate and delivers both the
O atoms to the cis-diol product. In fact, such a nonwater-
assisted O-O bond cleavage to form a FeV(O)(OH)
species (O1) has been proposed on the basis of density-
functional theory (DFT) calculations for the related
FeBPMEN [BPMEN=N,N0-dimethyl-N,N0-bis(2-pyridyl-
methyl)ethane-1,2-diamine] complex and found to be
energetically viable.33 Thus, both P1 and P2 participate in
this mechanistic landscape, with their respective involve-
ment being determined by the extent of solvent exchange
between MeCN and H2O.
Figure 5. 1-Octene oxidation catalyzed by 1 (1 mM) with H2O2 as
oxidant at room temperature (1/1-octene/substituted benzoic acid/H2O2 =
1/100/25/10) in the presence of variable amounts of water. The left side
(A) represents the results for perfluorobenzoic acid, while the right side
(B) depicts the study with 2-nitrobenzoic acid. Blue, yellow, and green
blocks represent epoxide, cis-diol, and phenol, respectively.
complementary labeling experiment performed with 10
equiv of H218O2 in the presence of 1000 equiv from H216O,
there was almost no change in the extent of label incor-
poration for cis-diol in the presence and absence of added
perfluorobenzoic acid, but the fraction of 18O-labeled
epoxide increased from 82% in the absence of added
acid to 92% in the presence of 25 equiv of C6F5CO2H.
The same extent of 18O label incorporation was also
observed for perfluorophenol. Given that the H2O2 used
was 90% 18O labeled, the labeling of epoxide and phenol
(30) Kim, C.; Chen, K.; Kim, J.; Que, L., Jr. J. Am. Chem. Soc. 1997, 119,
5964–5965.
(31) Mairata i Payeras, A.; Ho, R. Y. N.; Fujita, M.; Que, L., Jr. Chem.
Eur. J. 2004, 10, 4944–4953.
;
(32) Ho, R. Y. N.; Roelfes, G.; Feringa, B. L.; Que, L., Jr. J. Am. Chem.
Soc. 1999, 121, 264–265.
~
(33) Quinonero, D.; Morokuma, K.; Musaev, D. G.; Mas-Balleste, R.;
Que, L., Jr. J. Am. Chem. Soc. 2005, 127, 6548–6549.

9484 Inorganic Chemistry, Vol. 49, No. 20, 2010
Das and Que
a
Table 3. Effect of Added C6F5CO2H on the 18O-Label Distributions in the Products of 1-Catalyzed Oxidation of 1-Octene by H2O2
no acid
25 equiv C6F5CO2H
1,2-octanediol
1,2-epoxyoctane
1,2-octanediol
1,2-epoxyoctane
phenol
equiv of added H218O
or H218O2
16O
18O
16O16
O
16O18
O
18O18
O
16O
18O
16O16
O
16O18
O
18O18
O
16O
18O
200 H218O
500 H218O
1000 H218O
96
93
91
18
4
7
9
82
43
32
23
5
57
68
77
92
100
98
98
8
46
38
31
5
54
62
69
93
100
100
100
9
2
2
92
b
10 H218O2
3
2
91
a Reaction conditions: 1(1 mM)/H2O2/1-octene = 1/10/1000 in acetonitrile at room temperature. b The commercially available 2% aqueous solution
of H218O2 contains a 1:100 molar ratio of H218O2 and H216O.
Scheme 2. Mechanistic Landscape for Oxidations by the Nonheme Iron Catalyst 1 with H2O2 as Oxidanta
a P and O represent the various proposed iron-hydroperoxo and iron-oxo intermediates, respectively.
The addition of carboxylic acids complicates this mecha-
nistic landscape by introducing yet another potential
ligand that can displace the CH3CN solvate from P1. In
the case of added acetic acid, olefin oxidation yields
predominantly epoxide, with the epoxide O-atom derived
exclusively from H2O2.29 These results can be rationalized
by the formation of an FeV(O)(O2CR) (O3) oxidant from
P3 in a carboxylic-acid-assisted pathway. This oxidant
acts mainly as an oxo transfer agent, converting the olefin
substrate into epoxide. Lacking a bound hydroxo group,
O3 cannot effect olefin cis-dihydroxylation, although a
small amount of a novel byproduct has been identified
that derives from the transfer of both the oxo and acetato
groups to the substrate, resulting in the cis-hydroxy-
acetoxylation of the olefin.34
an additional wrinkle to the mechanism is introduced,
as the benzoate can undergo oxidative decarboxylation
and ipso-hydroxylation to afford phenol as a product. In
fact, electron withdrawing benzoic acids are catalytically
converted to phenols. The catalytic results presented in
this paper demonstrate the existence of an equilibrium
among P1, P2, and P3, as indicated by an increase in the
yields of phenol products as more benzoic acid is added
to the catalytic oxidation system (Figure 1) and a corre-
sponding decrease when water is added (Figure 5). When
olefin is present in the reaction mixture, there is a compe-
tition between ipso-hydroxylation and olefin epoxidation
that is affected by the relative amounts of benzoic acid
and olefin (Figures 3 and 4) and the phenol and epoxide
products exhibit the same extent of label incorporation
from H218O2 (Table 3), suggesting that they share a common
oxidant.
This same carboxylic-acid-assisted pathway can also be
invoked to rationalize the results presented in this paper
on the effect of adding benzoic acids with electron with-
drawing substituents. With benzoate replacing acetate in O3,
The observed oxidative decarboxylation and ipso-hydroxyl-
ation of electron withdrawing benzoic acids is postulated
to occur via intermediate O3. Internal electron transfer
from the bound benzoate to the FeVdO center generates
a carboxyl radical and an FeIV=O species, which then
ꢀ
(34) Mas-Balleste, R.; Fujita, M.; Que, L., Jr. Dalton Trans. 2008, 1828–
1830.

Article
Inorganic Chemistry, Vol. 49, No. 20, 2010 9485
recombine after loss of CO2 to form the ipso-hydroxylated
product. However, it is also plausible for the decay of P3
to bypass O3 entirely and proceed directly by O-O bond
homolysis to generate the carboxyl radical and the FeIV=O
species. Figure 4 shows data that strongly disfavor the
homolysis alternative. With the amount of perfluoro-
benzoic acid held constant at 25 equiv, the introduction
of 1-octene resulted in the formation of the corresponding
epoxide and diol products. Interestingly, as the 1-octene
concentration increased, the yield of cis-diol changed by a
factor of 2, but the amount of epoxide formed grew 5- to
7-fold. Thus, the diol-to-epoxide ratio decreased from 7:1
in the absence of the acid to 0.8:1 in its presence. Further-
more, the epoxide appeared to form at the expense of the
phenol product. In our interpretation of the data, the
amount of diol formed reflects the extent to which the
water-assisted pathway is operating, consistent with the
amount of label incorporation into the diol from H218O
(Table 3). However, in the presence of perfluorobenzoic
acid, label incorporation into the epoxide decreased from
H218O and increased from H218O2. Both observations are
consistent with a switch in the oxidant that forms epoxide
from O2 in the absence of added perfluorobenzoic acid to
O3 in its presence (Table 3). Thus, there is a competition
between ipso-hydroxylation of the bound perfluoro-
benzoate and the epoxidation of added 1-octene, an out-
come that can only be rationalized by the involvement of
O3, which is derived from the carboxylic-acid-assisted
heterolysis of the O-O bond of P3. A similar reactivity
pattern was observed for 2-nitrobenzoic acid (Figure 4B).
In the case of 2-chlorobenzoic acid (Figure 4C), less than
0.1 equiv of epoxide could be detected under all condi-
tions studied; the very low yield of epoxide may be attri-
buted to the greater ease of oxidizing 2-chlorobenzoate
versus perfluoro- or 2-nitrobenzoate. A similar competi-
tion between olefin epoxidation and oxidative decarboxyl-
ation has been noted previously for the catalysis of olefin
epoxidation by 1/H2O2 in the presence of phenylacetic
acid, where the yield of epoxide was correlated inversely
with the yield of benzaldehyde, which was derived from
oxidative decarboxylation of the acid.29 In the case of
acetic acid, O3 appears directed only toward the olefin
epoxidation.
Lastly, the absence of any label incorporation from
H218O into C6F5OH (Table 3) excludes the participation
of the FeVdO oxidant O2 in ipso-hydroxylation. Were
the water-assisted pathway also involved in ipso-hydroxyl-
ation, the phenol would have been partially 18O labeled,
resulting from the formation of O30 from O2 by ligand
exchange after the oxo/hydroxo tautomerization of O2
had occurred (Scheme 2). That the water-assisted path-
way is not involved is supported by the essentially quan-
titative incorporation of label from H218O2 into the
C6F5OH product (91% from 90% labeled peroxide).
Also, the almost identical 18O label distribution in cis-
diol with H218O2 in the presence or absence of added
perfluorobenzoic acid further emphasizes that cis-dihy-
droxylation is derived purely from the water-assisted
pathway (via P2 and O2). The clear difference in labeling
outcomes for the diol and the phenol products demon-
strates that olefin cis-dihydroxylation and ipso-hydroxyl-
ation must result from two related but distinct reaction
pathways that involve two different FeVdO oxidants.
These mechanistic speculations are supported by recent
EPR evidence reported by Talsi and co-workers that
suggest the trapping of an FeVdO oxidant in the reac-
tion of 1 and peracids at -70 °C.35 This transient S=1/2
species was shown to decay at a rate dependent on the
concentration of the added olefin and found to yield
epoxide, demonstrating its involvement in olefin epoxi-
dation. However, direct proof of the iron oxidation state
of the epoxidizing species was not obtained; such proof
will be essential to place the mechanistic landscape of
Scheme 2 on even firmer ground.
Acknowledgment. This work was supported by a grant
from the U.S. Department of Energy (FFG02-03ER15455
to L.Q.).
(35) Lyakin, O. Y.; Bryliakov, K. P.; Britovsek, G. J. P.; Talsi, E. P. J. Am.
Chem. Soc. 2009, 131, 10798–10799.