On the reactivity of mononuclear iron(V)oxo complexes

Article abstract of DOI:10.1021/ja208007w

Ferric tetraamido macrocyclic ligand (TAML)-based catalysts [Fe{C ₆H₄-1,2-(NCOCMe₂NCO)₂CR ₂}-(OH₂)]PPh₄ [1; R = Me (a), Et (b)] are oxidized by m-chloroperoxybenzoic acid at-40 °C in acetonitrile containing trace water in two steps to form Fe(V)oxo complexes (2a,b). These uniquely authenticated Fe^V (O) species comproportionate with the Fe ^III starting materials 1a,b to give μ-oxo-(Fe^IV) ₂ dimers. The comproportionation of 1a-2a is faster and that of 1b-2b is slower than the oxidation by 2a,b of sulfides (p-XC₆H ₄SMe) to sulfoxides, highlighting a remarkable steric control of the dynamics. Sulfide oxidation follows saturation kinetics in [p-XC ₆H₄SMe] with electronrich substrates (X = Me, H), but changes to linear kinetics with electron-poor substrates (X = Cl, CN) as the sulfide affinity for iron decreases. As the sulfide becomes less basic, the Fe^IV/Fe^III ratio at the end of reaction for 2b suggests a decreasing contribution of concerted oxygen-atom transfer (Fe^V → Fe^III) concomitant with increasing electron transfer oxidation (Fe^V → Fe^IV). Fe^V is more reactive toward PhSMe than Fe^IV by 4 orders of magnitude, a gap even larger than that known for peroxidase Compounds I and II. The findings reinforce prior work typecasting TAML activators as faithful peroxidase mimics.

Full text of DOI:10.1021/ja208007w

COMMUNICATION
pubs.acs.org/JACS
On the Reactivity of Mononuclear Iron(V)oxo Complexes
Soumen Kundu, Jasper Van Kirk Thompson, Alexander D. Ryabov,* and Terrence J. Collins*
Department of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213, United States
S Supporting Information
b
Scheme 1. Starting Ferric Complexes (1) and the Intercon-
versions of FeIII, FeIV, and FeV Species at ꢀ40 °C in MeCN
ABSTRACT: Ferric tetraamido macrocyclic ligand (TAML)-
Containing 0.2% (v/v) H2O
based catalysts [Fe{C6H4-1,2-(NCOCMe2NCO)2CR2}-
jj
j
(OH2)]PPh4 [1; R = Me (a), Et (b)] are oxidized by
m-chloroperoxybenzoic acid at ꢀ40 °C in acetonitrile con-
taining trace water in two steps to form Fe(V)oxo com-
plexes (2a,b). These uniquely authenticated FeV(O) species
comproportionate with the FeIII starting materials 1a,b to
give μ-oxo-(FeIV)2 dimers. The comproportionation of 1aꢀ
2a is faster and that of 1bꢀ2b is slower than the oxidation by
2a,b of sulfides (p-XC6H4SMe) to sulfoxides, highlighting a
remarkable steric control of the dynamics. Sulfide oxidation
follows saturation kinetics in [p-XC6H4SMe] with electron-
rich substrates (X = Me, H), but changes to linear kinetics
with electron-poor substrates (X = Cl, CN) as the sulfide
affinity for iron decreases. As the sulfide becomes less basic,
the FeIV/FeIII ratio at the end of reaction for 2b suggests a
decreasing contribution of concerted oxygen-atom transfer
(FeV f FeIII) concomitant with increasing electron transfer
oxidation (FeV f FeIV). FeV is more reactive toward PhSMe
than FeIV by 4 orders of magnitude, a gap even larger than
that known for peroxidase Compounds I and II. The find-
ings reinforce prior work typecasting TAML activators as
faithful peroxidase mimics.
processes might modulate the catalysis. Herein we describe these
reactivity features for 2a,b, the FeV(O) complexes4 by studying
the oxidations of aryl methyl sulfides (ArSMe) to sulfoxides. We
show that 2a,b can serve as kinetically competent reactive
intermediates in peroxidase-like catalysis and the reactivity
depends strongly upon subtle steric effects in the macrocyclic
ligand.
Complex 2a was prepared at ꢀ40 °C from 1a in MeCN
containing 0.2% (v/v) H2O by adding m-chloroperoxybenzoic
acid (mCPBA).4 The starting 1a (2.0 ꢁ 10ꢀ4 M) reacted with 0.5
equiv of mCPBA (1.0 ꢁ 10ꢀ4 M) to give the μ-oxo-(FeIV)2
dimer (Figure 1A). A second 0.5 equiv of mCPBA (1.0 ꢁ 10ꢀ4
M) converted the μ-oxo-(FeIV)2 dimer to 2a (Figure 1B). The
FeIII f FeIV conversion (k3/4) is ca. 10 times faster than the
FeIV f FeV transformation (k4/5) (Table 1), where k3/4 is the
effective second-order rate constant at low [mCPBA] [Figure 1S
in the Supporting Information (SI)] and k4/5 was measured
directly at [FeIV] = [mCPBA] (Figure 2S). On the basis of the
extinction coefficient reported (and supported by M€ossbauer
quantification),4 the amount of 2a formed was not less than 95%
of the total iron. Prior to measurements of the rates of oxidation
by 2 of the sulfide series p-XC6H4SMe, potentially interfering
interactions among FeIII, FeIV, and FeV (Scheme 1) were
investigated kinetically.
igh-valent ironꢀoxo complexes are key reactive intermedi-
H
ates in numerous biochemical oxidations. The most oxi-
dized are commonly isoelectronic with Fe(V) having one
oxidation equivalent residing on porphyrin and the second on
Fe(IV). While there is a substantial literature of well-character-
ized FeIVoxo species,1 FeV(O) reactive intermediates have rarely
been proposed. Examples include the Rieske dioxygenase enzyme2
and certain non-heme iron biomimetic systems.3 In the tetraamido
macrocyclic ligand (TAML) environment, however, FeV(O) has
been trapped and authenticated via spectroscopic, structural, and
First, FeV(O) converts to FeIV in MeCN. This process is much
slower than the reduction of FeV(O) by ArSMe. Therefore, it is
not an interfering reaction. The decay is first-order in FeV (Figure
3S), and the first-order rate constants k5/4 are given in Table 1.
Significantly, the decay of FeV to FeIV (k5/4) is 10 times faster for 2b
than for 2a (Table 1), in agreement with the macrocyclic ligand
CꢀH cleavage mechanism proposed elsewhere.9
5,6
theoretical studies.4 Iron(IV)oxo5 and μ-oxo-(FeIV)2 TAML
species also have been well-characterized. For multiple reasons, it
is important to examine mechanistically the reactivity of these
authentic FeV(O) complexes. First, as Collman and colleagues
have pointed out, “Only by associating particular complexes with
the kinetic behavior of the catalytic reaction can one be certain
that this complex is contributing to the catalytic process,”7 a
challenge that has been encountered and discussed previously in
biomimetic oxidation chemistry.8 Second, FeV(O) reactivities
must be compared with those of the corresponding FeIV species
to assess the relative and collaborative roles in catalysis. Third,
the rates of FeV(O) reactions with other iron complexes in the
catalytic medium should be measured to learn how such
Second, both the formation of FeV(O) and the compropor-
tionation with FeIII (k5+3/4) have been postulated to occur in
water above 0 °C, where FeV(O) is thought to have a fleeting
existence.10 Since comproportionation between 2 and 1 formed
upon substrate oxidation would tame the behavior of FeV(O)
Received: August 24, 2011
Published: October 10, 2011
r
2011 American Chemical Society
18546
dx.doi.org/10.1021/ja208007w J. Am. Chem. Soc. 2011, 133, 18546–18549
|

Journal of the American Chemical Society
COMMUNICATION
Figure 2. (A) UVꢀvis changes for 2b (5 ꢁ 10ꢀ5 M) reduction by
PhSMe (5 ꢁ 10ꢀ4 M). (B) Absorbance at 375 nm (O) and the amount
of PhMeSO formed (b) measured by HPLC; see the SI) vs time under
Figure 1. Spectral changes for the (A) FeIII f FeIV and (B) FeIV f FeV
conversions. Conditions: [1a] = 2 ꢁ 10ꢀ4 M; [mCPBA] = (1 + 1) ꢁ
10ꢀ4 M [added in two steps for the sequential conversions in (A) and
(B)]; [H2O] = 0.2% (v/v); MeCN; ꢀ40 °C. Figure 6S shows similar
spectral changes with 1b.
the same conditions. The solid line was calculated using kobs = 0.084 sꢀ1
Conditions: ꢀ40 °C, 0.2% (v/v) H2O/MeCN.
.
points held even in excess ArSMe (Figure 7S). On addition of at
least a 10-fold excess of PhSMe (chosen for the most extensive
studies), the UVꢀvis absorbances at 370 nm (2a) or 375 nm
(2b) rapidly increased. The full spectra indicated that 2a was
converted to FeIV quantitatively (Figure 8S) and that 2b pro-
duced a mixture of FeIII and FeIV (Figure 2A). This difference
results from the divergent comproportionation rates (see below).
Figure 2B shows the matching kinetics of the reduction of 2b and
the formation of PhMeSO (HPLC analysis; see the SI). At the
end of the reactions, PhMeSO was produced quantitatively
(HPLC and GCꢀMS analysis for both the 2a and 2b processes),
even though considerable FeIV remained at the end of the process
in each case. The reaction of FeIV with PhSMe is slow under
the experimental conditions (see below). Analyses of PhSMe/
PhMeSO were performed at room temperature after quenching
with the more reactive p-MeOC6H4SMe. Two-electron reduction
of FeV to FeIII by PhSMe should produce equivalent amounts
of FeIII and sulfoxide. One-electron reduction of FeV to FeIV
should first produce the corresponding intermediate cation radi-
cal ArMeS•+.13 In water, ArMeS•+ is known to react with O2 to
form sulfoxide, where the oxygen in sulfoxide originates from
water and not from O2.14 Thus, a consistent explanation for the
observed stoichiometry is that ArMeS•+, produced in an FeV to
FeIV step, is swept on rapidly by O2 to ArMeSO, leaving residual
FeIV while giving one sulfoxide for each FeV f FeIV event.
Table 1. Rate Constants (k5/4 in sꢀ1, All Others in Mꢀ1 sꢀ1
)
for the Interconversions between FeIII, FeIV, and FeV Species
(Scheme 1) in the Absence of Sulfide at ꢀ40 °C in MeCN
Containing 0.2% (v/v) H2O
a
b
complex
k3/4
k4/5
k5/4
k5+3/4
1a (R = Me) 280 ( 20 18 ( 1 (1.0 ( 0.1) ꢁ 10ꢀ5 (4 ( 1) ꢁ 104
1b (R = Et) 800 ( 100 47 ( 3 (1.11 ( 0.02) ꢁ 10ꢀ4 35 ( 1
a k3/4 = kK, as calculated from the established rate law: v = kK[FeIII]t-
[mCPBA]/(1 + K[mCPBA]). b Measured directly at [FeIV] = [mCPBA]
(see the SI for details).
toward the remaining substrate, it was important to measure k5+3/4
as we did here in MeCN to evaluate potential moderating
effects on the catalysis. When an equimolar amount of FeIII
was added to a solution of freshly prepared FeV(O) at ꢀ40 °C,
comproportionation led quantitatively to μ-oxo-(FeIV)2 (Figures
4S and 5S). Studies of 2a,b revealed that this rate is under ligand
control. The comproportionation of 2b and 1b is 103 times
slower than that of 2a and 1a (Table 1), yet 2a and 2b are similar
electronically. Thus, we attribute the large difference in k5+3/4
to the bulkier ethyl (2b) versus methyl (2a) groups on each face
of the macrocycle that hinder the formation of μ-oxo-(FeIV)2.
We conclude that steric effects should be explored through
TAML activator design to control the selectivity. In porphyrin
chemistry, it is well-known that steric effects can influence
comproportionations.11
Having observed and quantified the rates of processes
relating FeV, FeIV, and FeIII in the absence of a substrate, we
next studied the oxidations of ArSMe by 2. At [2] = 2 ꢁ 10ꢀ4 M
(used to produce Figure 1), the ArSMe oxidations were too
fast for conventional UVꢀvis techniques. Therefore, 2 was used
at a lower concentration (5 ꢁ 10ꢀ5 M), where excess mCPBA
(1 ꢁ 10ꢀ4 M, 2 equiv) was required for quantitative conversion
of 1 to 2. On the time scales of the sulfide oxidations under the
conditions employed, this excess mCPBA did not react with either
ArSMe or 1 fast enough to affect the kinetic behavior (see the SI).
Oxidations of organic sulfides to sulfoxides by biological and
biomimetic catalysts are known to proceed by both oxygen-atom
transfer (OT) and electron transfer (ET) mechanisms, suggest-
ing that either process or both could be found here.12 The
FeV(O) oxidations of sulfides were assayed at the isosbestic
points for FeIII and FeIV interconversions (370 and 375 nm for
2a,b, respectively; Figure 1 and Figures 6SA,B). The isosbestic
Attempts to observe ArMeS•+ or the peroxyl radical ArMeSO2
•+
by EPR spectroscopy were unsuccessful (see the SI). This does
not rule out these putative intermediates but mandates that
further reactions to sulfoxide be fast if these are indeed formed.
Oxidation of PhSMe by FeVO in the presence of H218O gave 25%
incorporation of 18O in the product PhMeSO.4 Both the OT (via
fast oxo ligand exchange) and ET14 pathways can lead to labeled
PhMeSO.
The absorbance-versus-time traces for reduction of 2a,b are
exponential, indicating first-order kinetics in 2a,b, which holds
for at least four half-lives. Thus, presumptive conversion of
ArMeS•+ to ArMeSO by O2 does not intrude into the kinetic
analysis of FeV(O) reactions. The reactions of FeV(O) with p-
XC6H4SMe [X = Me (2b); H, Cl, CN (2a,b)] were measured under
pseudo-first-order conditions in excess of sulfide. With X = MeO
(2a,b) and Me (2a), the processes were too fast for pseudo-first-
order treatment, so another approach was used for the kinetic
analysis (see below). The values of kobs for the reduction of FeV(O)
by ArSMe were collected in excess sulfide at concentrations of
(0.5ꢀ4.2) ꢁ 10ꢀ3 M (Figure 3A for 2b and Figure 10SA for 2a).
18547
dx.doi.org/10.1021/ja208007w |J. Am. Chem. Soc. 2011, 133, 18546–18549

Journal of the American Chemical Society
COMMUNICATION
Table 2. Rate and Equilibrium Constants for the Oxidations
of p-XC6H4SMe by FeV(O) and Relative Contributions of the
OT and ET Pathways, kOobTs/kEobTs, for 2ba
2b (R = Et)
2a (R = Me)
K1/Mꢀ1 kII/Mꢀ1 sꢀ1
k
/k
K1/Mꢀ1 kII/Mꢀ1 sꢀ1
OT ETb
obs obs
X
MeO
Me
H
5000 ( 400
9000 ( 700
3900 ( 250
800 ( 100 600 ( 80
1.9 ( 0.1
210 ( 30
190 ( 30 0.88 ( 0.07 500 ( 50 450 ( 50
Figure 3. (A) Pseudo-first-order rate constants kobs plotted against the
[p-XC6H4SMe]. Conditions: [2b] = 5 ꢁ 10ꢀ5 M; [H2O] = 0.2% (v/v);
ꢀ40 °C in MeCN. (B) Hammett plot for kII vs σ+ (see the text for
details). Figure 10S shows similar plots for 2a.
Cl
80 ( 2
0.57 ( 0.09
165 ( 5
13 ( 5c
CN
4.4 ( 0.5 0.20 ( 0.01 23 ( 4c
11.5 ( 0.5
a See the text for details. Conditions: ꢀ40 °C in 0.2% (v/v) H2O/
MeCN. b Calculated at the time of 95% conversion of FeV(O) at
[ArSMe] = 5 ꢁ 10ꢀ4 M. c Obtained using 0.005ꢀ0.05 M p-
NCC6H4SMe.
Scheme 2. Proposed Reaction Mechanism
0.05[Fe]tεV + [FeIV]εIV, so [FeIII]/[FeIV] = (εIV[Fe]t ꢀ
A722)/(A722 ꢀ 0.05[Fe]tεIV). The time at which A722 should
be read was calculated from kobs for each sulfide at [p-
XC6H4SMe] = 5 ꢁ 10ꢀ4 M. The [FeIII]/[FeIV] ratios (=k
/
OT
For 2a,b, the kobs values show similar trends, but 2a is 2ꢀ6 times
more reactive than 2b, suggesting that Me-versus-Et steric effects are
again significant.
Over the concentration range in Figure 3, the kobs dependen-
cies on [ArSMe] are hyperbolic for electron-rich and linear for
electron-poor sulfides for both 2a,b. Nevertheless, this behavioral
obs
kEobTs) are included in Table 2. These ratios indicate that (i) the ET
pathway is favored for the electron-poor sulfides, and (ii) the OT
pathway is favored for the electron-rich sulfides. Hammett plots
were made for both 2a,b. The fit for 2b (Figure 3B) improved
o
when log(kII/kII ) values were plotted against σ+ constants17
diversity is consistent with a common rate law: kobs
=
(F+ = ꢀ2.15 ( 0.09, r2 = 0.994) instead of σ (F = ꢀ3.0 ( 0.4, r2 =
0.939), suggesting that resonance effects are significant. For 2a,
kII[ArSMe]/(1 + K1[ArSMe]) eq 1 provided that K1[ArSMe] , 1
for sulfides with X = Cl, CN. This was confirmed by measuring
kobs for 2a with p-NCC6H4SMe at higher [ArSMe] [(0.5ꢀ5) ꢁ
10ꢀ2 M], where saturation kinetics emerged. The rate law
requires the reversible formation of an FeV(O)ꢀsulfide adduct,
such as I (Scheme 2). PhMeSO (5 ꢁ 10ꢀ5 M) did not inhibit the
oxidation of PhSMe (5 ꢁ 10ꢀ4 and 2 ꢁ 10ꢀ3 M) by 2a or 2b
(5 ꢁ 10ꢀ5 M), showing that the sulfoxide product does not bind
to FeVO under the reaction conditions.
the plot of log(kII/kII ) versus σ+ (Figure 10SB) had a similar
o
slope of ꢀ2.1 ( 0.3 (F+). Thus, the electrophilicities of 2a,b are
similar, supporting the idea that steric effects result in different
comproportionation rates. The absolute F values are larger than
those for the oxidation of organic sulfides to sulfoxides by
reconstituted cytochrome P450 (F+ = ꢀ0.16, σ+),18 (salen+•)-
Fe(IV)oxo complexes (F = ꢀ0.65 to ꢀ1.54, σ),19 or a mono-
nuclear non-heme Fe(IV)oxo complex (F = ꢀ1.0, σp).20 This
suggests that the sulfide X substituent is in strongest electronic
communication with the reactive center in the 2a,b case. To obey
the rate law, I cannot react with another sulfide and be involved in
the bimolecular OT pathway, as this would require a higher kinetic
order in sulfide. The geometric constraints appear to forbid
intramolecular OT through I (Scheme 2). Also to obey the rate law,
the ET pathway could proceed either through a bimolecular step
involving uncoordinated FeVO and sulfide (Scheme 2), with I
forming in an unproductive equilibrium, or via homolysis of the
FeꢀS bond in I. Both processes can deliver the observed saturation
kinetics of kobs on [ArSMe]. The kinetic data presented cannot
distinguish between the two cases. It has been proposed that
(salen+•)Fe(IV)oxo complexes bind sulfides reversibly at the oxo
ligand,19 which would represent another way to produce saturation
kinetics. However, we find it hard to imagine that this process could
be reversible: once the sulfur has approached the oxo ligand at the
van der Waals distance or less, the electronic reorganization required
to make the sulfoxide would occur instantaneously, at which point
the reaction would be over.
Because reactions of p-XC6H4SMe [X = OMe (2a,b), Me
(2a)] were too fast for pseudo-first-order treatment, kII was
determined at [FeV(O)] = [ArSMe] = 5 ꢁ 10ꢀ5 M assuming
second-order kinetics (2b; Figure 9S). It is realistic because for
the most electronically similar sulfide, p-MeC6H4SMe, at
[ArSMe] = 5 ꢁ 10ꢀ5 M with K1(2b) ≈ 103 Mꢀ1, eq 1 becomes
kobs/[ArSMe] ≈ kII. The kII values were obtained by fitting the
2
data to the equation At = A0 + (a0 kIIΔεt)/(1 + a0kIIt), where Δε
is the extinction coefficient difference of the FeIII and FeV species
and A0 and At are the absorbances at times 0 and t, respectively
(see the SI). kII for X = OMe (Table 2) is comparable to that for
oxidation of SMe2 by (meso-Mes4porphyrin+•)FeIVoxo under
similar conditions.15
In the case of 2b, where comproportionation is slow, the UVꢀ
vis spectra show that both FeIII and FeIV are generated by sulfide.
This implies that kobs reflects a summation of the rates for the
parallel reductions of FeV(O) to FeIII (OT pathway) and/or of
FeV(O) to FeIV (ET pathway),16 i.e., kobs = k + k and k
k
/
OT
ET
OT
obs
obs
obs
= [FeIII]/[FeIV]. k /k could be determined for 2b but
ET
obs
OT ET
obs obs
not for 2a (fast comproportionation). The relative contributions
of the OT and ET pathways for 2b were calculated from the
absorbance at 722 nm, where the molar absorptivities of FeV and
FeIV are the same but FeIII does not absorb (Figure 6SB). When
the FeV conversion is 95%, the absorbance at 722 nm is given by
Having studied how FeVO interacts with sulfides, we con-
cluded by examining the comparative behavior of FeIV with
PhSMe. The reactivity of the μ-oxo-(FeIV)2 dimer was quantified
by measuring at 750 nm the initial rates of its reaction with
PhSMe (Figure 4) to form FeIII and PhMeSO (Figure 7S). The
18548
dx.doi.org/10.1021/ja208007w |J. Am. Chem. Soc. 2011, 133, 18546–18549

Journal of the American Chemical Society
COMMUNICATION
Gupta for EPR assistance. S.K. thanks the R. K. Mellon Founda-
tion for a Presidential Fellowship in the Life Sciences. J.V.K.T.
thanks CMU for a Small Undergraduate Research Grant.
’ REFERENCES
(1) (a) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr. Chem.
Rev. 2004, 104, 939. (b) Kryatov, S. V.; Rybak-Akimova, E. V.; Schindler,
S. Chem. Rev. 2005, 105, 2175. (c) Pestovsky, O.; Stoian, S.; Bominaar,
E. L.; Shan, X.; M€unck, E.; Que, L., Jr.; Bakac, A. Angew. Chem., Int. Ed.
2005, 44, 6871. (d) Meunier, B. Models of Heme Peroxidases and
Catalases; Imperial College Press: London, 2000.
(2) Shan, X.; Que, L., Jr. J. Inorg. Biochem. 2006, 100, 421.
(3) (a) Das, P.; Que, L., Jr. Inorg. Chem. 2010, 49, 9479. (b) Prat, I.;
Mathieson, J. S.; G€uell, M.; Ribas, X.; Luis, J. M.; Cronin, L.; Costas, M.
Nat. Chem. 2011, 3, 788. (c) Company, A.; Gꢀomez, L.; G€uell, M.; Ribas,
X.; Luis, J. M.; Que, L., Jr.; Costas, M. J. Am. Chem. Soc. 2007, 129, 15766.
(d) Harischandra, D. N.; Zhang, R.; Newcomb, M. J. Am. Chem. Soc.
2005, 127, 13776.
(4) Tiago de Oliveira, F.; Chanda, A.; Banerjee, D.; Shan, X.;
Mondal, S.; Que, L., Jr.; Bominaar, E. L.; M€unck, E.; Collins, T. J. Science
2007, 315, 835.
(5) Chanda, A.; Shan, X.; Chakrabarti, M.; Ellis, W. C.; Popescu,
D. L.; Tiago de Oliveira, F.; Wang, D.; Que, L., Jr.; Collins, T. J.; M€unck,
E.; Bominaar, E. L. Inorg. Chem. 2008, 47, 3669.
(6) Ghosh, A.; Tiago de Oliveira, F.; Yano, T.; Nishioka, T.; Beach,
E. S.; Kinoshita, I.; M€unck, E.; Ryabov, A. D.; Horwitz, C. P.; Collins,
T. J. J. Am. Chem. Soc. 2005, 127, 2505.
(7) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry; Univer-
sity Science Books: Mill Valley, CA, 1987.
(8) (a) Lyakin, O. Y.; Bryliakov, K. P.; Britovsek, G. J. P.; Talsi, E. P.
J. Am. Chem. Soc. 2009, 131, 10798. (b) Chen, K.; Costas, M.; Kim, J.;
Tipton, A. K.; Que, L., Jr. J. Am. Chem. Soc. 2002, 124, 3026.
(9) Bartos, M. J.; Gordon-Wylie, S. W.; Fox, B. G.; Wright, L. J.; Weintraub,
S. T.; Kauffmann, K. E.; M€unck, E.; Kostka, K. L.; Uffelman, E. S.; Rickard,
C. E. F.; Noon, K. R.; Collins, T. J. Coord. Chem. Rev. 1998, 174, 361.
(10) (a) Ghosh, A.; Mitchell, D. A.; Chanda, A.; Ryabov, A. D.;
Popescu, D. L.; Upham, E.; Collins, G. J.; Collins, T. J. J. Am. Chem. Soc.
2008, 130, 15116. (b) Popescu, D.-L.; Vrabel, M.; Brausam, A.; Madsen,
P.; Lente, G.; Fabian, I.; Ryabov, A. D.; van Eldik, R.; Collins, T. J. Inorg.
Chem. 2010, 49, 11439.
Figure 4. Kinetics of reduction of FeIV to FeIII by PhSMe. (A) Initial
rate of reduction of FeIV as a function of [FeIV]. Conditions: [PhSMe] =
2.1 ꢁ 10ꢀ3 M, 0.2% (v/v) H2O/MeCN, ꢀ40 °C. (B) Initial rate of
reduction of FeIV as a function of [PhSMe]. Conditions: [FeIV] = 1 ꢁ
10ꢀ4 M, 0.2% (v/v) H2O/MeCN, ꢀ40 °C.
initial rate varied linearly with [FeIV] and exhibited saturation
kinetics with [PhSMe] (Figure 4), suggesting the reversible
formation of a sulfide adduct. The data were fitted to the rate
equation v = k4/3[FeIV][PhSMe]/(1 + K[PhSMe]) to obtain the
values K = (120 ( 25) Mꢀ1 and k4/3 = (6 ( 1) ꢁ 10ꢀ2 Mꢀ1 sꢀ1
,
where k4/3 corresponds to the rate constant characterizing the
reactivity of FeIV. The oxidant FeV(O) is more reactive than this
FeIV species by 4 orders of magnitude, a gap even larger than that
known for Compounds I and II of horseradish peroxidase.21
In conclusion, the first detailed reactivity studies of authentic
Fe(V)oxo complexes have revealed rapid sulfide oxidations. The
FeV(O) reagents form from FeIII via FeIV, where comproportio-
nation might be involved. FeV(O) is 4 orders of magnitude more
reactive toward sulfides than FeIV. This suggests that the FeV(O)
species have the high reactivity needed for catalytic oxidations.22
Among the potentially intruding degradation and compropor-
tionation reactions of the FeV(O) reagent, only the latter is active
under the conditions employed. The FeV(O) engages in oxygen-
atom transfer with organic sulfides, establishing an FeV f FeIII
redox process. Nevertheless, despite high rates of formation of
sulfoxides from sulfides at low temperature, the OT dominates
only slightly over ET and then only for the electron-rich sulfides
in the studied ArSMe series. Electron-withdrawing groups induce
the sulfides to react predominantly via ET. The comproportiona-
tion between FeV and FeIII is very sensitive to steric effects and is
strongly hindered by the relatively minor exchange of a methyl
for an ethyl substituent. Steric effects also slow the reactions of
sulfides with 2b relative to 2a. This points to a large scope for
designing steric effects into TAML systems to control the
selectivity in oxidation catalysis.
(11) Collman, J. P.; Gagne, R. R.; Reed, C. A.; Halbert, T. R.; Lang,
G.; Robinson, W. T. J. Am. Chem. Soc. 1975, 97, 1427.
(12) Goto, Y.; Matsui, T.; Ozaki, S.-i.; Watanabe, Y.; Fukuzumi, S.
J. Am. Chem. Soc. 1999, 121, 9497.
(13) Kobayashi, S.; Nakano, M.; Kimura, T.; Schaap, A. P. Biochem-
istry 1987, 26, 5019.
(14) (a) Sch€oneich, C.; Aced, A.; Asmus, K.-D. J. Am. Chem. Soc.
1993, 115, 11376. (b) Bonifaꢁciꢀc, M.; Hug, G. L.; Sch€oneich, C. J. Phys.
Chem. A 2000, 104, 1240.
(15) Franke, A.; Fertinger, C.; van Eldik, R. Angew. Chem., Int. Ed.
2008, 47, 5238.
’ ASSOCIATED CONTENT
(16) Espenson, J. H. Chemical Kinetics and Reaction Mechanisms, 2nd
ed.; McGraw-Hill: New York, 1995.
(17) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
(18) Watanabe, Y.; Iyanagi, T.; Oae, S. Tetrahedron Lett. 1980,
21, 3685.
S
Supporting Information. Experimental details and kinetic
b
measurements. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
(19) Sivasubramanian, V. K.; Ganesan, M.; Rajagopal, S.; Ramaraj, R.
J. Org. Chem. 2002, 67, 1506.
(20) Sastri, C. V.; Seo, M. S.; Park, M. J.; Kim, K. M.; Nam, W. Chem.
Commun. 2005, 1405.
Corresponding Author
tc1u@andrew.cmu.edu; ryabov@andrew.cmu.edu.
(21) Dunford, H. B. Heme Peroxidases; Wiley-VCH: New York, 1999.
(22) Collins, T. J.; Khetan, S. K.; Ryabov, A. D. In Handbook of Green
Chemistry; Anastas, P. T., Crabtree, R. H., Eds.; Wiley-VCH: Weinheim,
Germany, 2009; pp 39ꢀ77.
’ ACKNOWLEDGMENT
Support from the Heinz Endowments, the Institute for Green
Science and CMU is acknowledged (T.J.C). We thank Dr.
Deboshri Banerjee for preliminary measurements and Rupal
18549
dx.doi.org/10.1021/ja208007w |J. Am. Chem. Soc. 2011, 133, 18546–18549