Cyclohexene as a substrate probe for the nature of the high-valent iron-oxo oxidant in Fe(TPA)-catalyzed oxidations

Article abstract of DOI:10.1039/c3nj00524k

Cyclohexene is shown to be a versatile substrate probe in shedding light on the nature of the high-valent oxidant generated by bio-inspired nonheme iron catalysts. Cyclohexene provides the oxidant a choice to attack an allylic C-H bond or a CC bond, and t

Full text of DOI:10.1039/c3nj00524k

NJC
View Article Online
View Journal
LETTER
Cyclohexene as a substrate probe for the
nature of the high-valent iron-oxo oxidant in
Fe(TPA)-catalyzed oxidations
Cite this: DOI: 10.1039/c3nj00524k
Received (in Porto Alegre, Brazil)
15th May 2013,
Williamson N. Oloo, Yan Feng, Shyam Iyer, Sean Parmelee, Genqiang Xue and
Lawrence Que, Jr.*
Accepted 28th June 2013
DOI: 10.1039/c3nj00524k
www.rsc.org/njc
Cyclohexene is shown to be a versatile substrate probe in shedding
light on the nature of the high-valent oxidant generated by bio-
inspired nonheme iron catalysts. Cyclohexene provides the oxidant a
Q
choice to attack an allylic C–H bond or a C C bond, and the
Scheme 1 Possible high valent iron intermediates proposed for olefin oxidation
selectivity observed should depend on the electronic properties of
the oxidant. Allylic oxidation products are predominantly observed in
the reaction with [FeIV(O)(TPA)(NCCH3)]2+ (TPA = tris(2-pyridylmethyl)-
amine), while epoxide and cis-diol are the major products found in the
[FeII(TPA)(NCCH3)2]2+-catalyzed reaction with H2O2. This difference
suggests that the oxidant generated in the latter case must be distinct
from [FeIV(O)(TPA)(NCCH3)]2+ and supports an oxoiron(V)-hydroxo
species that has been proposed as the oxidant.
by bio-inspired iron catalysts and H2O2.
epoxidation and cis-dihydroxylation.7 In this paper, we report
the use of cyclohexene as a substrate to probe the nature of the
high-valent oxidant.
Cyclohexene is a versatile substrate probe, first used by
Groves to probe the oxidative reactivity of cytochrome P450
and a related model complex.8 Cyclohexene provides the oxidant
with two possible channels of reactivity, namely abstraction of
an allylic C–H bond and oxygen-atom transfer to the CQC bond.
The ratio of products from these two channels should reflect
the relative preferences of the oxidant to undergo H-atom
abstraction and O-atom transfer and thus can shed light on
the nature of the oxidant, in particular its electronic properties.
Our experiments first focused on [FeIV(O)(TPA)(NCCH3)]2+ (2),
which was generated using ArIO as oxidant (ArIO = 1-(tert-
butylsulfonyl)-2-iodosylbenzene, a more soluble, non-polymeric
analog of PhIO) in CH3CN. Fig. 1 shows that 2 was formed at
À40 1C in >90% yield upon addition of a slight stoichiometric
excess of ArIO to a CH3CN solution of 1. Complex 2 exhibits a
near IR band at 711 nm that arises from ligand field transitions
of the S = 1 FeIVQO unit and is characteristic of this class of
compounds.9 The presence of 2 was confirmed by ESI-MS
where a dominant ion at m/z = 511.0348 that is formulated as
[Fe(TPA)(O)](OTf)+ on the basis of the mass (calcd = 511.0350)
and isotope distribution pattern observed (Fig. 1, inset). The
lmax of 2 obtained from ArIO is slightly blue-shifted relative
to that of 2 obtained previously with AcOOH as oxidant
(lmax 722 nm).5 This spectral difference may reflect differences
in the nature of the ligand coordinated to the available site
High-valent intermediates are implicated in the catalytic cycles
of dioxygen activating nonheme iron enzymes,1 such as
a-ketoglutarate-dependent oxygenases2 and Rieske dioxygenases.3
These enzymes carry out a variety of biologically important
and potentially synthetically useful transformations, such as
the hydroxylation and halogenation of C–H bonds and the
epoxidation and cis-dihydroxylation of CQC bonds. Pursuant
to an understanding of biological dioxygen activation, much
effort has focused on synthesizing iron complexes whose
reactivities can serve to model these enzymes.4 Scheme 1 shows
the various high-valent species that have been postulated as
possible oxidants for various nonheme iron-catalyzed reac-
tions, but only one of these, namely A, has been directly observed
and spectroscopically characterized.5 Among the polydentate
ligands used to support iron centers in these biomimetic efforts
is the tripodal tetradentate TPA ligand (TPA = tris(2-pyridylmethyl)-
amine). The corresponding iron(II) complex [FeII(TPA)(NCCH3)2]-
(OTf)2 (1) has been found to react with H2O2 to generate oxidants
that can carry out hydroxylation of C–H bonds,6 as well as olefin
Department of Chemistry and Center for Metals in Biocatalysis, University of
Minnesota, Minneapolis, Minnesota 55455, USA. E-mail: larryque@umn.edu;
Fax: +1 612 6250389; Tel: +1 612 6247029
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Letter
NJC
Subsequent work-up and GC analysis of the resulting reaction
mixture revealed formation of cyclohex-2-enol, cyclohex-2-enone,
and cyclohexene oxide (Table 1), but no cis-cyclohexane-1,2-diol
was detected. The allylic alcohol and ketone products were
generated in yields of 23% and 28%, respectively, relative to 2,
while cyclohexene oxide was generated in 7% yield (Table 1,
entry 5). When 2 was generated using AcOOH, only allylic
oxidation products were observed (Table 1, entry 6). These
results show that 2 strongly favours H-atom transfer (HAT) over
O-atom transfer (OAT) and confirm an earlier observation of
Mas-Balleste and Que, who found that the reaction between 2
and cyclooctene afforded o10% cyclooctene oxide.12
The next set of experiments explored the oxidation of
cyclohexene with 1 in combination with H2O2 at 25 1C. In a
typical experiment, 10 eq. H2O2 (30% solution in H2O) were
added to a CH3CN solution of 1 (1.0 mM) and cyclohexene
(250 eq.) at room temperature under inert atmosphere,
followed by work up and product analysis by gas chromato-
graphy (GC). In this case, all four possible products were
obtained (Table 1, entry 1), with the cis-diol and epoxide
representing the major products of the reaction. These results
correspond to an HAT/OAT ratio of 1 : 4, a reactivity preference
that is opposite to the >7 : 1 ratio found for 2, thus excluding 2
as the oxidant responsible for the OAT products in 1-catalyzed
cyclohexene oxidation with excess H2O2. Interestingly however,
the use of near stoichiometric amounts of H2O2 afforded a
HAT/OAT ratio of B1 : 1 (Table 1, entries 3 and 4). These results
suggest that the reaction of 1 and H2O2 during the first turnover
generates an oxidant that favours C–H bond cleavage, such as 2,
and evolves in subsequent turnovers to an oxidant that favours
attack of the CQC bond. This intriguing observation, though
requiring further investigation, emphasizes the utility of cyclo-
hexene as a mechanistic probe.
To ensure that cyclohexene oxidation followed the pattern
established for other substrates in 1-catalyzed oxidations,6,7
isotopic labeling experiments were carried out in the presence
of H218O (1000 eq.). Consistent with previous results, both
cyclohexenol and cyclohexene oxide show partial incorporation
of an 18O from water, while one of the diol O-atom essentially
derives from water. These results can readily be rationalized by
the participation of the previously proposed FeV(O)(OH) oxi-
dant,6,7 where both the hydroxo group derived from water and
the oxo group derived from H2O2 are transferred to the substrate
to make the cis-diol product, while oxo/hydroxo tautomerization
prior to substrate attack by the high-valent FeQO unit leads to
partial labeling of the epoxide and enol products (Scheme 2).
Kinetic evidence supporting this ‘water-assisted’ mechanism has
recently been reported by Oloo et al. in the 1-catalyzed oxidation
of olefins by H2O2.13 By monitoring the (TPA)FeIII–OOH inter-
mediate observed at À40 1C, the decay rate of this species
showed saturation behaviour with increasing water concen-
tration and a H2O/D2O KIE of 2.5. This water-assisted decay of
the FeIII–OOH intermediate affords the FeV(O)(OH) oxidant 3,
which then reacts rapidly with the substrate (Scheme 2).
Fig. 1 Conversion of 1 (1 mM in CH3CN, black trace) to 2 (red trace) upon
treatment with 1.2 eq. ArIO at À40 1C under N2. Inset: Observed isotope
distribution pattern for the dominant ion in the ESI-MS spectrum of 2 with
m/z = 511.0348.
cis to the oxo group, as a previous study on a series of
[FeIV(O)(TPA)(X)]2+ complexes (X = Cl, Br, O2CCF3) has shown
that the maximum of the broad near IR band associated with the
FeIV(O)(TPA) unit is sensitive to the nature of the cis ligand.10 In
the present case, this ligand is likely to be CH3CN, as there are
no other obvious ligands in the solution besides the triflate
counterions and solvent. On the other hand, the AcOOH used as
the oxidant in the earlier experiment is an equilibrium mixture
of AcOOH, H2O2, and HOAc, so the FeQO complex formed with
AcOOH may have other moieties bound to the FeQO unit.
Introduction of 250 eq. cyclohexene upon maximum formation
of 2 resulted in its rapid decay. The decay of 2 could be fit well
with a single exponential function, and the resulting first order
rate constants showed a linear dependence on cyclohexene
concentration, showing that 2 directly oxidizes cyclohexene
(Fig. 2). The corresponding second order rate constant of
0.019 MÀ1 sÀ1 compares well with the value of 0.070 MÀ1 sÀ1
obtained at À40 1C for the oxidation of cyclohexadiene by the
closely related [FeIV(O)(N4Py)]2+ (N4Py = bis(2-pyridylmethyl)-
bis(2-pyridyl)methylamine) complex.11
Fig. 2 Observed rate of decay of (TPA)FeIV(O) complex
2 as a function
The nature of the high-valent oxidant clearly changes when
the 1-catalyzed oxidation is carried out in the presence of acetic
of cyclohexene concentration at À40 1C.
A
second order decay rate of
0.019 MÀ1 sÀ1 was obtained.
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Letter
Table 1 Productsa of cyclohexene oxidation under various reaction conditions
Catalyst
Oxidant
A
K
E
D
Ref.
1
1
10 HOOH
10 HOOH
200 HOAc
2 HOOH
1 HOOH
1.2 ArIO
1.2 AcOOH
1 PhIO
10 HOOH
1.4b (46%)
1.4
0.5
0.2
2.3b (36%)
6.8
5.0b (81%)
0.6
This work
This work
1
1
1
1
0.6
0.4
0.23
0.31
0.15
0.5
0.2
—
0.28
0.31
—
0.2
0.1
0.07
—
0.55
—
0.6
0.2
—
—
—
This work
This work
This work
This work
8
Fe(TPP)Cl
FeII(L)
0.4
5.6
14
a
A = cyclohexenol; K = cyclohexenone; E = cyclohexene oxide; D = cis-cyclohexane-1,2-diol; TPP = meso-tetraphenylporphin; L = PhDPAH =
(di-(2-pyridyl)methyl)benzamide. Entries in parentheses represent % 18O incorporation into products for the reaction run in the presence of
b
1000 eq. H218O.
peracids at À60 1C.17 They associated the EPR signal with the
proposed FeV(O)(O2CR) species, but this conclusion has recently
been challenged by results from DFT calculations of Wang, Shaik
and co-workers.18
The last entry in Table 1 is FeII(PhDPAH) PhDPAH = (di-(2-
pyridyl)methyl)benzamide,
a nonheme iron catalyst that
catalyses olefin cis-dihydroxylation using H2O2 as oxidant.14
Even less is known about the mechanism of this reaction as not
a single intermediate has yet been detected. However, based on
reactivity and mechanistic studies, an FeIV(OH)2 oxidant (B in
Scheme 1) has been proposed to form upon reaction of the
iron(II) catalyst with H2O2. The incorporation of some 18O label
from H218O into the diol product requires O–O bond cleavage to
occur prior to olefin oxidation to allow for the possibility of
Scheme 2 Proposed water-assisted mechanism for the oxidation of hydro-
carbons by 1 and H2O2.
acid, as evidenced by the product yields in Table 1, entry 2. label exchange. The notion of an FeIV(OH)2 oxidant is supported
Under these reaction conditions, formation of epoxide is by DFT calculations by Comba on a related Fe(N4) catalyst.19
significantly increased at the expense of cis-diol, while main- Interestingly, oxidation of cyclohexene by this catalyst/H2O2
taining allylic oxidation at about the same level. Thus the HAT/ combination affords mainly the cis-diol product as expected,
OAT ratio remains more or less the same at about 1 : 4. The with an HAT/OAT ratio of 1 : 6. The HAT/OAT preference is
change in olefin oxidation products has been proposed to arise opposite to that for 2, suggesting that the electronic properties
from the modification of the water-assisted mechanism to a of the oxidant formed in the Fe(PhDPAH)/H2O2 reaction must
carboxylic-acid-assisted version, where the acetic acid replaces differ from those of 2, despite sharing the same formal iron
the water ligand in the FeIII–OOH intermediate, to form the oxidation state. One possible difference may be the spin state of
corresponding FeV(O)(OAc) oxidant D (Scheme 1).12 With the the iron(IV) center. Given the combination of N,N,O-donor atoms
loss of the hydroxo ligand from the oxidant, cis-dihydroxylation in the PhDPAH ligand, it seems likely that the iron(IV) center
is no longer possible.
would in this case be high-spin, compared to the S = 1 state that
Interestingly, the HAT/OAT ratio found for the catalytic has been established for 2.5 Alternatively, the similarity of the
Fe(TPA)/H2O2 reactions resembles that reported by Groves for HAT/OAT ratio of Fe(PhDPAH)/H2O2 with those of Fe(TPA)/H2O2
the stoichiometric Fe(TPP)Cl/PhIO reaction (Table 1, entry 7),8 and Fe(TPP)Cl/PhIO may suggest involvement of a formally
implicating oxidants that favour OAT significantly over HAT. iron(V) oxidant. More work is needed to gain insight into the
The Fe(TPP)Cl/PhIO (TPP = tetraphenylporphinate dianion) nature of the oxidant in the Fe(PhDPAH)–H2O2 system.
combination has been shown to generate an FeIV(O)(porphyrin
In summary, we have employed cyclohexene as a substrate
radical cation) oxidant15 and can be described formally as an probe to show that the FeIV(O) species 2 cannot be the oxidant
FeV(O) complex. The oxidants for the Fe(TPA)–H2O2 systems are responsible for the epoxide and cis-diol products formed in the
also proposed to be formally FeV(O) species, but they have not 1-catalysed oxidation of cyclohexene. This conclusion stems
been stable enough to be characterized as thoroughly as the from experiments showing that the reaction of cyclohexene
porphyrin-based oxidant. Costas and Cronin have provided with 2 afforded mainly, if not exclusively, the HAT-derived products
convincing mass spectrometric evidence for the FeV(O)(OH) cyclohex-2-enol and cyclohex-2-enone, while corresponding
species generated from the reaction of a related Fe(N4) catalyst catalytic reactions with Fe(TPA)/H2O2 yielded mainly OAT-derived
with H2O2 at À40 1C,16 while Talsi and co-workers have observed products, namely cyclohexene oxide and cis-cyclohexane-1,2-diol.
a transient S = 1/2 EPR signal in the reaction of various Fe(N4) Comparison of the cyclohexene oxidation results listed in Table 1
complexes with H2O2 in the presence of carboxylic acids or with suggests that this divergence in HAT/OAT selectivity for the
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Letter
NJC
Fe(TPA)-based high-valent species may reflect differences in the formation of 2, cyclohexene was added and the NIR band of 2
nature of the high-valent oxidants and supports the involve- was monitored in order to obtain the decay rate of 2.
ment of formally oxoiron(V) oxidants C and D (Scheme 1) that
Catalytic oxidations and isotopic labeling experiments
have been proposed as the active species in the Fe(TPA)/H2O2
reactions.6,7,12,13 Whether the preference for HAT over OAT is a
In a typical reaction carried out at 25 1C, 3 mL of a CH3CN
solution of 1.0 mM 1 and 0.25 M cyclohexene was prepared
general reactivity pattern for nonheme FeIV(O) complexes
remains to be seen, although the dearth of examples of olefin
epoxidation by such complexes supports this notion. However
an exception to this trend is the FeIV(O) complex of a pyridine
containing macrocyclic ligand, which was recently reported by
Rybak-Akimova to effect epoxidation of cyclooctene.20 This
result suggests that the nature of the supporting ligand can
affect the olefin epoxidation ability of an FeIV(O) unit. Such
complexities in reactivity may extend the utility of cyclohexene
as a probe substrate for the mechanisms of other nonheme iron
oxidation catalysts.
under nitrogen. The reaction was initiated by the addition of
10 eq. 30% H2O2. In reactions conducted in the presence
of AcOH or H218O, the additive was added prior to the addition
of H2O2. After the reaction mixture was stirred for 30 minutes,
the reaction mixture was worked up and analyzed by gas
chromatography according to published procedures.7 Isotopic
labeling experiments were also carried out according to pub-
lished procedures.7
Acknowledgements
This work was supported by the US Department of Energy,
Office of Basic Energy Sciences (grant DE-FG02-03ER15455).
We very much appreciate the suggestion of a reviewer to do the
Methodology
Materials and general procedures
All reagents were purchased from Aldrich and used as received low-equivalent H2O2 experiments listed in Table 1.
unless otherwise indicated. Preparation and handling of air-
sensitive materials were carried out under an inert atmosphere
by using either standard Schlenk and vacuum line techniques
Notes and references
or a glove box. Labeled water (95% 18O) was purchased from
ICON Isotopes (Summit, NJ). Cyclohexene was purified by
vacuum distillation before use. The Fe(TPA) complex5 and the
1-(tert-butylsulfonyl)-2-iodosylbenzene oxidant (ArIO)21 were
prepared according to published literature procedures.
1 M. Costas, M. P. Mehn, M. P. Jensen and L. Que, Jr., Chem.
Rev., 2004, 104, 939.
2 C. Krebs, D. G. Fujimori, C. T. Walsh and J. M. Bollinger,
Acc. Chem. Res., 2007, 40, 484.
3 S. Chakrabarty, R. N. Austin, D. Y. Deng, J. T. Groves and
J. D. Lipscomb, J. Am. Chem. Soc., 2007, 129, 3514.
4 L. Que, Jr. and W. B. Tolman, Nature, 2008, 455, 333.
5 M. H. Lim, J. U. Rohde, A. Stubna, M. R. Bukowski,
M. Costas, R. Y. N. Ho, E. Munck, W. Nam and L. Que, Jr.,
Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 3665.
6 K. Chen and L. Que, Jr., J. Am. Chem. Soc., 2001,
123, 6327.
7 K. Chen, M. Costas, J. Kim, A. K. Tipton and L. Que, Jr.,
J. Am. Chem. Soc., 2002, 124, 3026.
8 J. T. Groves and D. V. Subramanian, J. Am. Chem. Soc., 1984,
106, 2177.
9 A. Decker, J. U. Rohde, E. J. Klinker, S. D. Wong, L. Que, Jr.
and E. I. Solomon, J. Am. Chem. Soc., 2007, 129, 15983.
Physical methods
UV-visible spectra were recorded on a Hewlett-Packard 8453A
diode array spectrometer with samples maintained at low
temperature by using a cryostat from Unisoku Scientific Instru-
ments, Osaka. Electrospray mass spectrometry was performed
on a Finnigan-MAT (San Jose, CA) LCQ ion trap instrument.
The spectrum of the (TPA)FeIV(O) species 2 was obtained
by direct introduction of the solution of the intermediate
generated at À40 1C into the injector port of the spectrometer.
The capillary heater on the instrument was turned off and the
flow of the atomizing gas was increased to minimize thermal
decomposition of the unstable intermediate. Product analyses
were performed on a Perkin-Elmer Sigma 3 gas chromatograph 10 J. U. Rohde, A. Stubna, E. L. Bominaar, E. Munck, W. Nam
(AT-1701 column, 30 m) with a flame ionization detector. GC
and L. Que, Jr., Inorg. Chem., 2006, 45, 6435.
mass spectral analyses were performed on a HP 5898 GC (DB-5 11 D. Wang, K. Ray, M. J. Collins, E. R. Farquhar, J. R. Frisch,
column, 60 m) with a Finnigan MAT 95 mass detector or a HP
6890 GC (HP-5 column, 30 m) with an Agilent 5973 mass
L. Gomez, T. A. Jackson, M. Kerscher, A. Waleska, P. Comba,
M. Costas and L. Que, Jr., Chem. Sci., 2013, 4, 282.
detector. A 4% NH3–CH4 mix was used as the ionization gas 12 R. Mas-Balleste and L. Que, Jr., J. Am. Chem. Soc., 2007,
for chemical ionization analyses.
129, 15964.
13 W. N. Oloo, A. J. Fielding and L. Que, Jr., J. Am. Chem. Soc.,
2013, 135, 6438.
were 14 P. D. Oldenburg, Y. Feng, I. Pryjomska-Ray, D. Ness and
Reactivity studies
Reactivity studies for the (TPA)FeIV(O) complex
2
conducted under nitrogen atmosphere and were monitored
L. Que, Jr., J. Am. Chem. Soc., 2010, 132, 17713.
via UV-visible spectroscopy at À40 1C. In a typical experiment, 15 (a) J. T. Groves, R. C. Haushalter, M. Nakamura, T. E. Nemo
a 1.0 mM solution of 1 in CH3CN was treated with 1.2 eq. 1-(tert-
butylsulfonyl)-2-iodosylbenzene or AcOOH. Upon maximum
and B. J. Evans, J. Am. Chem. Soc., 1981, 103, 2884;
(b) J. E. Penner-Hahn, K. S. Eble, T. J. McMurry,
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Letter
M. Renner, A. L. Balch, J. T. Groves, J. H. Dawson and 18 Y. Wang, D. Janardanan, D. Usharani, K. Han, L. Que, Jr.
K. O. Hodgson, J. Am. Chem. Soc., 1986, 108, 7819;
(c) J. T. Groves, J. Inorg. Biochem., 2006, 100, 434.
16 I. Prat, J. S. Mathieson, M. Guell, X. Ribas, J. M. Luis,
L. Cronin and M. Costas, Nat. Chem., 2011, 3, 788.
and S. Shaik, ACS Catal., 2013, 3, 1334.
19 P. Comba, G. Rajaraman and H. Rohwer, Inorg. Chem., 2007,
46, 3826.
20 W. Ye, D. M. Ho, S. Friedle, T. D. Palluccio and E. V. Rybak-
Akimova, Inorg. Chem., 2012, 51, 5006.
17 (a) O. Y. Lyakin, K. P. Bryliakov, G. J. P. Britovsek and
E. P. Talsi, J. Am. Chem. Soc., 2009, 131, 10798; 21 D. Macikenas, E. Skrzypczak-Jankun and J. D. Protasiewicz,
(b) O. Y. Lyakin, R. V. Ottenbacher, K. P. Bryliakov and
E. P. Talsi, ACS Catal., 2012, 2, 1196.
J. Am. Chem. Soc., 1999, 121, 7164 (see additional safety note
in J. Am. Chem. Soc. 2011, 133, 4151).
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.