Mechanism of Alcohol Oxidation by FeV(O) at Room Temperature

Article abstract of DOI:10.1021/acs.inorgchem.5b01937

Selective oxidation of alcohol to its corresponding carbonyl compound is an important chemical process in biological as well as industrial reactions. The heme containing enzyme CytP450 has been known to selectively oxidize alcohols to their corresponding carbonyl compounds. The mechanism of this reaction, which involves high-valent Fe^IV(O)-porphyrin^?+ intermediate with alcohol, has been well-studied extensively both with the native enzyme and with model complexes. In this paper, we report for the first time the mechanistic insight of alcohol oxidation with Fe^V(O) complex of biuret TAML (bTAML), which is isoelectronic with Fe^IV(O)-porphyrin^?+ intermediate form in CytP450. The oxidations displayed saturation kinetics, which allowed us to determine both the binding constants and first-order rate constants for the reaction. The K and k values observed for the oxidation of benzyl alcohol by Fe^V(O) at room temperature (K = 300 M^-1, k = 0.35 s^-1) is very similar to that obtained by CytP450 compound I at -50 °C (K = 214 M^-1, k = 0.48 s^-1). Thermodynamic parameters determined from van't Hoff's plot (ΔH～ -4 kcal/mol) suggest hydrogen bonding interaction between substrate and bTAML ligand framework of the Fe^V(O) complex. Analysis of H/D KIE (k_H/k_D ～ 19 at 303 K), Hammett correlation and linearity in Bell-Evans-Polyanski plot points to the C-H abstraction as the rate determination step. Finally, experiments using Fe^V(O¹⁸) for benzyl alcohol oxidation and use of the "radical clock" cyclobutanol as a substrate shows the absence of a rebound mechanism as is observed for CytP450. Instead, an ET/PT process is proposed after C-H abstraction leading to formation of the aldehyde, similar to what has been proposed for the heme and nonheme model compounds.

Full text of DOI:10.1021/acs.inorgchem.5b01937

Article
pubs.acs.org/IC
Mechanism of Alcohol Oxidation by FeV(O) at Room Temperature
Munmun Ghosh,†,§ Y. L. K. Nikhil,§ ,† Basab B. Dhar,‡ and Sayam Sen Gupta*,†
†Chemical Engineering Division, CSIR-National Chemical Laboratory, Pune 411008, India
‡Department of Chemistry, Shiv Nadar University, Goutam Buddha Nagar, UP 201314, India
S
* Supporting Information
ABSTRACT: Selective oxidation of alcohol to its corresponding carbonyl compound
is an important chemical process in biological as well as industrial reactions. The heme
containing enzyme CytP450 has been known to selectively oxidize alcohols to their
corresponding carbonyl compounds. The mechanism of this reaction, which involves
high-valent FeIV(O)−porphyrin•+ intermediate with alcohol, has been well-studied
extensively both with the native enzyme and with model complexes. In this paper, we
report for the first time the mechanistic insight of alcohol oxidation with FeV(O)
complex of biuret TAML (bTAML), which is isoelectronic with FeIV(O)−porphyrin•+
intermediate form in CytP450. The oxidations displayed saturation kinetics, which
allowed us to determine both the binding constants and first-order rate constants for
the reaction. The K and k values observed for the oxidation of benzyl alcohol by
FeV(O) at room temperature (K = 300 M−1, k = 0.35 s−1) is very similar to that
obtained by CytP450 compound I at −50 °C (K = 214 M−1, k = 0.48 s−1).
Thermodynamic parameters determined from van’t Hoff’s plot (ΔH∼ −4 kcal/mol)
suggest hydrogen bonding interaction between substrate and bTAML ligand framework of the FeV(O) complex. Analysis of H/D
KIE (kH/kD ∼ 19 at 303 K), Hammett correlation and linearity in Bell-Evans-Polyanski plot points to the C−H abstraction as the
rate determination step. Finally, experiments using FeV(O18) for benzyl alcohol oxidation and use of the “radical clock”
cyclobutanol as a substrate shows the absence of a rebound mechanism as is observed for CytP450. Instead, an ET/PT process is
proposed after C−H abstraction leading to formation of the aldehyde, similar to what has been proposed for the heme and
nonheme model compounds.
understand biological oxidation but also to develop bioinspired
catalysts for environmentally benign catalysis. Mechanism of
alcohol oxidation with CytP450 has been evaluated under both
catalytic and single turnover conditions.28,29 The proposed
mechanism consisted of the following steps: (i) binding of
benzyl alcohol to the enzyme, (ii) α-CH abstraction from
benzyl alcohol, (iii) rebound of the −OH from FeIV(OH)
heme, and (iv) dehydration of the gem-diol to form an
aldehyde. Single turnover experiments performed with
quantitative generation of CytP450 compound I at −50 °C
afforded a binding constant of 214 M−1 and a first-order rate
constant of 0.48 s−1 corresponding to the C−H abstraction.28
In contrast, alcohol oxidation with model compounds
containing the FeIV(O)−porphyrin•+ and nonheme FeIV(O)
displayed no binding of the substrate to the Fe(O)
intermediates. In addition, instead of a “rebound” mechanism,
the oxidation was shown to proceed by a two-electron
oxidation process.30 Herein, we provide the first report for
the mechanistic insights into alcohol oxidation by an in situ
generated bTAML FeV(O) at room temperature. We show that
similar to compound I studies, binding of the benzyl alcohol to
the bTAML FeV(O) occurs with binding constants of ∼300
M−1. However, the mechanism of oxidation involves a two-
INTRODUCTION
■
High valent iron oxo species of heme and nonheme containing
monooxygenase are invoked as reactive intermediates for the
catalytic oxidation of organic substrates in numerous biological
processes.1−5 For heme containing enzymes such as cyto-
chrome P450 (CytP450), the FeIV(O)−porphyrin•+ is an active
oxidant, which catalyzes alkane hydroxylation, olefin epox-
idation, and alcohol oxidation.6−9 In the nonheme family of
enzymes, both FeIV(O) and FeV(O) have been frequently
proposed as the reactive intermediates for the oxidation of
organic substrates. The reactivity landscape of both FeIV(O)−
and nonheme FeIV(O)−porphyrin•+ toward oxidation of
several organic substrates like alkanes, alkenes, and alcohols
have been explored in detail.10−15 In contrast, reports on the
reactivity of FeV(O) toward organic substrates, an intermediate
proposed for Rieske dioxygenase family, are scarce.16−20 We
have recently reported the formation of bTAML-FeV(O) of
FeIII-bTAML at room temperature that has been characterized
by several spectroscopic techniques.21 The reactivity of FeV(O)
using bTAML and TAML ligands toward oxidation of C−H
bonds in alkanes, CC bonds in alkenes, and sulphoxidation
has been recently reported by us and others.18,19,21−24
Oxidation of alcohol to the corresponding carbonyl compound
is an important chemical process in biological as well as
industrial reactions.25−27 Hence understanding the mechanism
of this process is very important in our quest not only to
Received: August 25, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b01937
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Schematic Representation for Benzyl Alcohol Oxidation by FeV(O)
(bTAML)] m/z 431.0763 (calculated m/z 431.0773) as previously
electron process similar to what has been observed for the
heme and nonheme model compounds.30
reported.21 Twenty microliters of a benzyl alcohol solution of 10−1
M
benzyl alcohol was added to the labeled FeV(O18) solution and
immediately checked the GC-MS.
EXPERIMENTAL SECTION
■
Materials. Fe−bTAML (1) was synthesized by our previously
reported method.31 Meta-chloroperbenzoicacid (mCPBA, Aldrich
77%) was purified by an established method. Acetonitrile (HPLC
grade, Aldrich) was used passing through an activated neutral alumina
column and then dried as described elsewhere. 18O-enriched water
(98%) was procured from the Shanghai Research Institute of Chemical
Industry (China). Benzyl alcohol (Aldrich, 99.8%) was passed through
activated neutral alumina and distilled prior to use. All reactions were
carried out without any special precautions under atmospheric
conditions unless otherwise specified.
RESULTS AND DISCUSSIONS
■
Kinetics of Benzyl Alcohol Oxidation. FeV(O) was
generated at 25 °C from the parent bTAML activator,
(Et4N)2[FeIII(Cl)-(bTAML] (1), in CH3CN by adding
equimolar (1.2 equiv) amounts of mcpba, as reported before.21
The unprecedented stability21 of 2 at RT allowed us to perform
extensive kinetic studies under single-turnover conditions to
ascertain the mechanism of benzyl alcohol oxidation by FeV(O)
using UV−vis spectroscopy. Upon addition of benzyl alcohol
(20 equiv. 10−3 M) to a solution of 2 (5 × 10−5 M), the green
color of the reaction mixture immediately changed to violet.
This is ascribed to the formation of the μ-oxo−FeIV dimer
which has been characterized earlier by UV−vis spectroscopy
and NMR.21 Although the two electron oxidation of alcohol to
aldehyde is expected to reduce FeV species to parent FeIII
complex 1, the fast comproportionation reaction between 1 and
2 (eq 2, 1.0 × 105 M−1 s−1; much faster than the rate of alcohol
oxidation)21 leads to the formation of the μ-oxo−FeIV dimer.
This FeIV species slowly regenerates back to the parent FeIII
complex quantitatively over time (Figure 1).
General Instrumentation. UV−vis spectral studies were carried
out using an Agilent diode array 8453 spectrophotometer with an
attached electrically controlled thermostat. Gas chromatography (GC)
was performed on a PerkinElmer Arnel Clarus 500 instrument
equipped with a hydrogen flame ionization detector; HP5 columns
(polar) (12 m × 0.32 mm × 1.0 μm) were used with helium as the
carrier gas at a flow rate of 1 mL min−1. GC-MS was performed on an
Agilent 5977A mass selective detector interface with an Agilent 7890B
gas chromatograph using a HP-5 ms capillary column (30 m × 0.32
mm × 0.25 μm, J&W Scientific). HR-MS was performed in a Thermo
Scientific Q-Exactive Orbitrapanalyzer using an electrospray ionization
source connected with a C18 column (150 m × 4.6 mm × 8.0 μm).
Reactivity of FeV(O) (2) toward Benzyl Alcohol Oxidation.
Triplicate kinetic runs were carried out for each experiment, and mean
values are reported here. The kinetics were monitored in either the
kinetic mode of the spectrophotometer or the scanning spectral
kinetics mode using a 1.0 cm quartz cell at 399 nm (Isosbestic points
of FeIV species and FeIII) at 25 °C. All kinetic experiments were carried
out in CH3CN. During kinetic measurements, FeV(O) was generated
by adding 1.2 equiv of mCPBA and then substrate was added to it
under pseudo-first order conditions. (Concentration of 2 was from 2 ×
10−5 to 1 × 10−4 M, depending upon reactivity of the substrate, and
different benzyl alcohol concentrations were chosen according to the
experiment, which were from 2 × 10−4 to 1.8 × 10−1 M.) Pseudo-first-
order rate constant (kobs) was calculated at the isosbestic wavelength
FeV(O) + C6H5CH2OH → C6H5CHO + FeIII
(1)
by using a nonlinear curve fitting [At = Aα − (Aα − Ao)e(−k t)] and
obs
had good agreement in rate constant value within 10% error.
Single-Turnover Reaction. Excess substrate (20 equiv. 1 × 10−3 M)
was added to a freshly prepared 500 μL solution of 2 [generated by
reaction of 1 (5 × 10−5 M) and mCPBA (6 × 10−5 M)] at room
temperature (RT)]. Reaction mixture was stirred for 2−3 min. After
completion of the reaction (determined by UV−vis spectroscopy), the
products (yield %) were quantified by GC using a calibration curve
obtained with authentic benzaldehyde of respective benzyl alcohol.
The products were identified by GC-MS.
18O Incorporation Experiment. To 200 μL of an acetonitrile
solution of 1 (5 × 10−5 M), 1.2 equiv of mCPBA was added, which
leads to the formation of FeV(O) (2). H2O18 (20 μL) was introduced
into the solvent media of species 2, and the solution was allowed to
stand at −22 °C for approximately 3 h. The HRMS showed 70%
incorporation of O18 leading to the formation of [FeV(O18)
Figure 1. UV−vis spectral changes upon reaction of 2 (5 × 10−5 M)
with 1 mM of benzyl alcohol. (Inset) Absorbance vs time plot at 399
nm. Red line is the first-order fit according to the equation At = Aα −
(Aα − A0)e−kobst
.
B
DOI: 10.1021/acs.inorgchem.5b01937
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
The pseudo-first-order rate constant (kobs), measured at the
isosbestic point of FeIII and FeIV, correlated linearly with the
substrate concentration when the substrate concentration was
varied between 0.25 to 1.0 mM. The slope of the straight line
provided us the second-order rate constant k2 of 75 10 M−1
s−1 (Figure SI 1). After completion of the reaction,
benzaldehyde was identified as the sole product (∼72%) and
no further oxidation to benzoic acid was observed. On the basis
of the fast comproportionation between 1 and 2, only 50%
yield of the benzaldehyde is expected upon reaction of 2 with
alcohol (Figure SI 2). The higher yields of benzaldehyde
(∼72%) observed are due to the subsequent reaction of alcohol
with the μ-oxo-FeIV dimer (Figure 8; reaction discussed later;
Figure SI 3). Similarly, second-order rate constants of different-
substituted benzyl alcohols were calculated (Figure SI 4−6).
When the single turnover experiments were performed at
higher concentration of alcohol (>1 mM, 30 equiv), saturation
behavior of the pseudo-first-order rate constant (kobs) was
observed for benzyl alcohol and its substituents (Figure 2 and
very similar to those obtained for the oxidation of benzyl
alcohol by CytP450 compound I at −50 °C (K = 214 M−1, k =
0.48 s−1).28 Likewise, equilibrium constants (K) and rate
constants (k) of different substituted benzyl alcohols were also
calculated (SI Table). For model compounds, binding of
alcohol to metal-oxo has been recently reported for the
oxidation of alcohols by RuIV = O complex in water where an
equilibrium constant of 92 6 M−1 was obtained at 280 K.33
To determine the thermodynamic parameters for the pre-
equilibrium process and the subsequent oxidation reaction, the
oxidation of benzyl alcohol by 2 were carried out at different
temperatures, and all the values of K and k are summarized in
Table 1. The plots of the equilibrium constants K and the rate
Table 1. Equilibrium Constants and Rate Constants at
Different Temperatures Determined in the Oxidation of
Benzyl Alcohol by bTAML-FeV(O) Complex from Equation
4
temperature (K)
equilibrium constant K (M−1
)
rate constant k (s−1
)
300
293
288
283
300 25
350 25
405 15
454 28
0.35
0.24
0.15
0.12
constants k relative to the inverse of the reaction temperatures
T−1 (Eyring plots and van’t Hoff plots, respectively; Figure 3,
panels A and B) allowed us to obtain the thermodynamic
parameters for the pre-equilibrium processes and the activation
parameters for the alcohol oxidation, respectively. As indicated
by the thermodynamic parameters from the van’t Hoff’s plot,
the formation of the precursor complex is exothermic and the
value of ∼4 kcal/mol suggests hydrogen-bonding interaction
between substrates and bTAML ligand framework of FeV(O).
This is quite reasonable since the bTAML ligand framework
has four carbonyl oxygen close to the FeV(O) in the ligand
moiety, and these are capable of forming H-bonds with the
alcohol (Figure 9). The activity parameters obtained from the
Eyring plot (ΔH‡ = 10 kcal mol−1; ΔS‡ = −26 cal mol−1 K−1)
are indicative of a well-organized transition state (Table 2).
These are also extremely similar to the activation parameters for
CytP450 compound I (ΔH‡ = 13.2 kcal mol−1) and FeIV(O)
radical cation of electron-rich porphyrins (ΔH‡ = 9.6 kcal
mol−1; ΔS‡ = −26 cal mol−1 K−1) described earlier.28 This is
expected since the FeIV(O) radical cation is isoelectronic with
the FeV(O) of our complex. However, for model heme-systems,
no binding of the substrate to the complex is reported. In
addition, no saturation behavior was observed for toluene
oxidation (0.1 M, 1000 equiv) by FeV(O) (second-order rate
constant k2 = 0.136 M−1 s−1) that we reported earlier.21 This
leads us to speculate that the amide oxygens allow H-bonding
to form an adduct of bTAML-FeV(O) with benzyl alcohol.
To probe the generality of these observations, the oxidation
of cyclohexanol (>100 equiv. 5 mM) with FeV(O) 2 (5 × 10−5
M) was also studied under single turnover conditions. As with
benzyl alcohol, saturation behavior of the pseudo-first-order
rate constant kobs was observed (Figure SI 8). Fitting kobs to
concentration of the substrate using saturation rate law (3) gave
the equilibrium constants of the pre-equilibrium processes (K =
77 3 M−1) and the rate constants of the subsequent oxidation
Figure 2. Pseudo-first-order kinetic analysis for the oxidation of benzyl
alcohol at 300 K (red), 293 K (black), 288 K (pink), and 283 K (blue).
Figure SI 7). This points to the existence of a pre-equilibrium
process before the oxidation reaction.32 It is suggesting a
noncovalent interaction between FeV(O) and benzyl alcohol
leading to an intermediate formation. Taking into account the
pre-equilibrium process (eq 3), the kobs obtained was plotted
against different concentration of benzyl alcohol. Fitting kobs to
concentration of the substrate using saturation rate law (4) gave
the equilibrium constants (K) of the pre-equilibrium processes
and the rate constants (k) of the subsequent oxidation
reactions.
K
k
FeV(O) + benzyl alcohol ⇌ [FeV(O)·benzylalcohol] → product
(3)
kK[benzyl alcohol]
1 + K[benzyl alcohol]
kobs
=
(4)
The equilibrium constant K300, the rate constant k300 at 300
K, was determined to be 300 25 M−1 at 300 K and 0.35
0.05 s−1, respectively. Consistent with this analysis, the product,
K300 × k300 (104.7 M−1 s−1) is equal within error to the value of
k2 measured at lower benzyl alcohol concentrations (80 10
M−1 s−1). The large equilibrium constant values indicate strong
binding interaction between 2 and benzyl alcohol during the
course of the reaction. The K and k values we have observed are
reactions (k = 0.19
0.023 s−1). The rate constant k for
cyclohexanol was determined to be pertinently slower (0.19 s−1
at 298 K) than benzyl alcohol. The equilibrium constant of the
C
DOI: 10.1021/acs.inorgchem.5b01937
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. (A) Eyring plot. (B) vant Hoff plots for oxidation of benzyl alcohol with a FeV(O) complex.
Table 2. Thermodynamic Parameters Obtained from Eyring
and van’t Hoff Plot
Table 3. Second-Order Rate Constant of C6H5CH2OH (kH)
and C6H5CD2OH (kD) at Different Temperatures and
Corresponding KIE
ΔH [kcal mol−1
ΔS [cal K−1 mol−1
]
−4.13 0.3 ΔH‡ [kcal mol−1
−2.45 0.5 ΔS‡ [cal K−1 mol−1
]
+10.25 0.8
]
]
−26.46 1.2
temperature (K)
kH (M−1 s−1
)
kD (M−1 s−1
)
KIE
19
303
283
273
263
75 10
4
0.54
0.4
2
3
4
7
62
33
14
7
7
3
2
25
33
54
pre-equilibrium binding was less than that observed for benzyl
alcohol. This is expected since benzyl alcohol has a lower pKa
than cyclohexanol (16 compared to 15.4) and hence is expected
to form stronger H bonds. Also, the conformational flexibility
of the −OH group in cyclohexanol is lower than benzyl alcohol.
We anticipated that the −OH in the benzyl alcohol or in
cyclohexanol is responsible for the pre-equilibrium process
resulting in an ordered transition state during the oxidation
reaction.
1
0.02
0.25
Mechanistic Study for the Oxidation of Benzyl
Alcohol. In order to obtain a mechanistic picture for the
oxidation process, several experiments were performed. This
includes determination of kinetic isotope effect (KIE) of C−H
abstraction, linear free energy relationship (LFRE) with
substituted benzyl alcohols, experiments using FeV(O18), and
oxidation of cyclobutanol as a radical clock.
Kinetic Isotope Effect. The kinetics of benzyl alcohol with
high valent FeV(O) (5 × 10−5 M) species were studied using
(d2)-benzyl alcohol (C6H5−CD2OH) as the substrate (1.5−3
mM) to determine the kinetic isotope effect (kH/kD). A very
Figure 4. Plot of ln k vs 1000/T (Arrhenious plot) of C6H5CH2OH
(brown line) and C6H5CD2OH (blue line).
large KIE value of 19
2 was determined at 30 °C. Such
benzyl alcohol, these two values (ln AD and Ea) were
determined to be 19.2 and 10.7 kcal/mol, respectively. This
corresponds to AH/AD = 0.025 and ΔEa = 4.1 kcal/mol. Both
these values suggest the presence of a tunnelling effect.
All the above data indicates that the activation of alcohols
occurs exclusively by hydrogen atom abstraction from the α-
CH of benzyl alcohol, and that C−H bond cleavage is the rate-
determining step in the alcohol oxidation. The large negative
enthalpy value and well-organized transition state observed
from the activation parameters discussed above agree well with
the high nonclassical KIE.
nonclassical large H/D KIE values (>7), observed for FeIV
intermediates in enzymes such as methane monooxygenase34 or
TaUD oxoiron intermediates35 has been ascribed to a hydrogen
atom tunnelling reaction. For oxidation of benzyl alcohol by
CytP450 compound I, kH/kD has been determined 28 2 at 23
°C.36
In order to determine if tunnelling was taking place, the
temperature-dependent second-order rate constants of benzyl
alcohol and (d2)-benzyl alcohol were performed between 30 °C
and −10 °C. There are three key criteria for determining
tunneling effects in terms of the Arrhenius equation as has been
described by Bell.37 These include (i) the KIE kH/kD > 7 and is
temperature dependent; (ii) the ratio AH/AD < 0.7, where A is
the pre-exponential term in the Arrhenius equation, and (iii) a
difference in activation energy (ΔEa) between C−H and C−D
substrates greater than the ZPE difference of 1.13 kcal mol−1.
We observe that for benzyl alcohol oxidation, the KIE increases
with decreasing temperature (Table 3). The Arrhenius pre-
exponential factor (ln AH) and activation energy Ea were
determined to be 15 and 6.6 kcal/mol (Figure 4). For the (d2)-
Linear Free Energy Relationship. Oxidation of various p-
substituted benzyl alcohol (0.25−1.50 mM) by FeV(O) (5 ×
10−5 M) was also studied to validate if hydrogen atom
abstraction from the α-CH of benzyl alcohol represented the
rate-determining step. From eq 4, the equilibrium constant
values (K) for various substituted alcohols were determined to
be 300
60 M−1 (SI Table). This connotes that para-
substituents on benzyl alcohol have very little or no effect on
the pre-equilibrium process. The krel (krel = kX/kH) value for
benzyl alcohol and its analogue was plotted against the
D
DOI: 10.1021/acs.inorgchem.5b01937
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Hammett parameter, and ρ value −0.67 was obtained from the
slope (Figure 5). The relatively small value of ρ indicates that
(>85%), thus indicating that the oxidation by FeV(O) is a
two-electron process.
Oxidation Using 18O labeled FeV(O). Oxidation of benzyl
alcohol was finally carried out with an 18O-labeled FeV(O)
complex to further prove the two-electron oxidation process.
Upon adding 25 equiv of benzyl alcohol (2.5 mM) to the
solution of 18O-labeled catalyst, only <5% of 18O incorporation
was observed in benzaldehyde as determined by GC-MS
(overall yield of benzaldehyde >70%, Figure SI 9−10).41 If the
oxidation had occurred through a two-step rebound process, a
1:1 mixture of 16O- and 18O-labeled benzaldehyde should have
been observed. The small amount of 18O incorporated product
was likely the result of oxygen atom exchange between 16O
benzaldehyde and H218O leading to <5% incorporation of 18O
in the benzaldehyde product (Figure 7). In all likelihood, the
Figure 5. Hammett plot of log krel against substituent constants of
benzyl alcohols.
consistent with the mechanism proposed on the basis of KIE,
the reaction rates are not greatly influenced by the electron-
withdrawing and electron-donating ability of the para
substituents. The Hammett value for the FeV(O) is slightly
higher than that reported for synthetic FeIV(O)−porphyrin•+
(ρ = −0.39)30 but lower than that obtained for the CytP450
enzymes (ρ = −1.19).28
Further, to confirm C−H abstraction as the rate-determining
step, we have plotted k2 (80 M−1 s−1) of benzyl alcohol (BDE =
79 kcal mol−1)38 in the BEP graph, which was previously
determined for the mechanism of alkane hydroxylation (SI 11).
This plot showed a competent linear fit confirming C−H
abstraction as the rate-determining step.
Figure 7. Benzyl alcohol oxidation with FeV(O18), < 5% incorporation
of O18 into benzaldehyde.
oxygen of benzyl alcohol is retained in the product
benzaldehyde. This is in contrast to CytP450, where 30% 18O
incorporation in benzaldehyde is observed upon using 18O2 as
the oxidant, indicating a two-step rebound mechanism.28 A 2-
electron oxidation process similar to that observed by us has
been reported for the oxidation of benzyl alcohol by both heme
FeIV(O) radical cation and nonheme FeIV(O) model
complexes.
Oxidation with Cyclobutanol. Two possibilities exist for the
carbon radical of benzyl alcohol formed after α-CH abstraction
to undergo further oxidation leading to aldehyde formation.
The first involves the classical “rebound mechanism” in which
the FeIV−OH formed upon C−H abstraction rebounds the
−OH group to form a gem-diol which subsequently dehydrates
to form the aldehyde. In an alternative scenario, carbon radical
of benzyl alcohol can be oxidized which then transfers a H+ to
the solvent cage leading to aldehyde formation. To distinguish
between the one electron and two electron process, cyclo-
butanol (100 equiv. 10−2 M) has been taken as a substrate
(Figure 6).39,40 In this experiment upon oxidation, if ketone is
Mechanistic Interpretation. To summarize the observations,
the first step of alcohol oxidation by bTAML FeV(O) involves
substrate binding with the FeV(O) complex (evident from the
saturation kinetics). This is followed by H atom abstraction as
is evidenced by very large KIE values and linearity obtained in
the BEP graph (SI 11). The relatively small value of ρ obtained
from the Hammett plot also supports this hypothesis.
Following the H atom abstraction leading to the formation of
bTAML FeIV(OH) and the cyclobutanol radical cage, the
following possibilities exist leading to product formation: (i)
diffusion of cyclobutanol radical from the cage followed by
further oxidation, (ii) rapid trapping of the cyclobutanol radical
via rebound to form the gem-diol, and (iii) ET/PT from the
carbon radical to the FeIV−OH leading to aldehyde formation.
The possibility (i) is negated by the radical clock experiment
using cyclobutanol, where cyclobutanone is obtained as the
major product (only minor ring-opened product was observed).
The 18O labeling experiments which show <5% incorporation
of 18O in the benzaldehyde product negate the possibility of
rebound (ii). We believe that FeIV−OH is a strong enough
oxidant (FeIV/III redox potential is ∼0.6 V42) to oxidize the
benzyl alcohol radical via ET/PT to form the product aldehyde.
Hence the likely mechanism for alcohol oxidation by bTAML
FeV(O) involves a 2-electron process leading to a “net hydride
transfer” (Figure 9).
Figure 6. Oxidation of cyclobutanol by FeV(O) showing cyclo-
butanone as a major product.
the sole product then the reaction occurs by hydrogen atom
abstraction followed by electron transfer. On the other hand, if
a carbon radical formed by H atom abstraction of cyclobutanol
and then escapes from the solvent cage, a ring open product, 4-
hydroxybutyraldehyde, can be formed. Our experiment with
FeV(O) show predominant formation of cyclobutanone
Reactivity of μ-oxo-FeIV Dimer. The >50% yields of
benzaldehyde formation observed in the single turnover
experiments is unlikely because the fast comproportionation
between 1 and 2 would only allow 50% of 2 to react with
E
DOI: 10.1021/acs.inorgchem.5b01937
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
benzyl alcohol and yield a maximum of only 50% benzaldehyde
(with respect to 2). The higher yields of benzaldehyde
observed are due to the slow reaction of alkanes with the μ-
oxo-FeIV dimer, similar to what has been observed for alkene
oxidation by FeV(O).22 Hence the reactivity of μ-Oxo-FeIV
dimer with benzyl alcohol was studied at RT. The μ-oxo-FeIV
dimer was generated by addition of 0.5 equiv of m-CPBA in the
FeIII complex. The synthesized dimer had a characteristic UV−
vis spectra similar to the previously reported μ-oxo-FeIV dimer
(formed in equimolar amounts of FeIII and FeV(O);
21
1
characterized by H NMR and EPR). The resulting dimer
species reacted with benzyl alcohol to form the corresponding
benzaldehyde with rates that are ∼200 fold slower than FeV(O)
(Figure SI 3). Evidently, when the concentration of benzyl
alcohol was increased to 1000 equiv, no saturation was
observed for the kobs showing that substrate-mediated
disproportionation was not occurring. On the basis of prior
work from our group where the epoxidation of alkenes with μ-
oxo-FeIV dimer was studied,22 we hypothesized that the μ-oxo-
FeIV dimer exists in an equilibrium with FeV(O) and FeIII
species and the FeV(O) was the species responsible for
oxidation of benzyl alcohol. The reaction of benzyl alcohol
with FeV(O) leads to more disproportionation of the dimer,
leading to the formation of more FeV(O), which reacts with
benzyl alcohol until all of it is converted back to the starting
FeIII complex (Figure SI 3). To prove this hypothesis, the
reactions of μ-oxo-FeIV with benzyl alcohol were carried out
with varying concentrations of the parent FeIII complex.
Reduction of reaction rates was observed with increasing
concentrations of the parent FeIII complex. The plot of reaction
rates versus increasing concentration of FeIII in the reaction
mixture shows a competent fit into the kinetic model.
Figure 8. Plot of initial rate vs concentration of FeIII, complex 1, added
for reaction of μ-oxo-FeIV dimer (2 × 10−5 M) with benzyl alcohol
(0.015 M). k1 (0.00596) and k−1 (21833) were obtained according to
the nonlinear curve fit (eq 7). Reaction was performed at RT in
acetonitrile solvent under air.
k2
FeV(O) + benzyl alcohol → products
(6)
k1k2[μ − oxo − FeIVdimer][benzyl alcohol]
Figure 9. Proposed mechanism for alcohol oxidation with FeV(O).
rate =
k−1[FeIII] + k2[benzyl alcohol]
(7)
With the use of the second-order rate constant (k2) that was
determined earlier for benzyl alcohol oxidation using FeV(O),
k1 and k−1 were calculated under single turnover conditions
using eq 7 as 0.00596 and 21833, respectively (see Figure 8).
The comproportionation k−1 is in the error range of the value
reported elsewhere independently.21
oxidation reaction were generated. Thermodynamic parameters
determined from van’t Hoff’s plot (ΔH ∼ −4 kcal/mol)
suggest a hydrogen-bonding interaction between substrate and
bTAML ligand framework in the FeV(O) complex. The first-
order rate constant obtained for biuret FeV(O) is much smaller
compared to the FeIV(O) radical cation of CytP450 but is
comparable to the electron-rich heme model of CytP450.
Analysis of H/D KIE for benzyl alcohol and its benzylic
dideuterio isotopomer (kH/kD ∼ 19 at 303 K), Hammett
correlation with substituted benzyl alcohols and linearity in
Bell-Evans-Polyanski plot points to the C−H abstraction as the
rate determination step (Figure SI 11). Finally, experiments
using FeV(O18) for benzyl alcohol oxidation and use of the
“radical clock” cyclobutanol as a substrate shows the absence of
a rebound mechanism as is observed for CtyP450. Instead, an
ET/PT process is proposed after C−H abstraction, leading to
formation of the aldehyde, similar to what has been proposed
for the heme and nonheme model compounds. In summary,
the reactivity of FeV(O) toward alcohol is reported for the first
time. We have shown binding of substrate −OH to the bTAML
ligand framework of FeV(O). Such an interaction had not been
observed earlier for model complexes. These secondary
CONCLUSION
■
Oxidation of benzyl alcohol by bTAML FeV(O) was studied at
room temperature in CH3CN. The oxidations displayed
saturation kinetics, which allowed us to determine both the
binding constants and first-order rate constants for the reaction.
The saturation kinetics observed are similar to the oxidation of
benzyl alcohol by CytP450 compound I at −50 °C under single
turnover conditions.28 The K and k values observed for the
oxidation of benzyl alcohol by FeV(O) at RT (K = 300 M−1, k =
0.35 s−1) is very similar to that obtained by CytP450 compound
I at −50 °C (K = 214 M−1, k = 0.48 s−1). Such saturation
kinetics have not been observed for model heme compound I
and nonheme FeIV(O) reported earlier.30 For oxidations of the
benzyl alcohol conducted between 10 and 27 °C, a van’t Hoff
function for binding and an Eyring function for the first-order
F
DOI: 10.1021/acs.inorgchem.5b01937
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(18) de Oliveira, F. T.; Chanda, A.; Banerjee, D.; Shan, X.; Mondal,
interactions can be used to design substrates for selective
oxidation of complex natural products. Work in such direction
is being pursued in our laboratory.
S.; Que, L.; Bominaar, E. L.; Munck, E.; Collins, T. J. Science 2007, 315
̈
(5813), 835−838.
(19) Kundu, S.; Thompson, J. V. K.; Ryabov, A. D.; Collins, T. J. J.
Am. Chem. Soc. 2011, 133 (46), 18546−18549.
(20) Chen, K.; Que, L. J. Am. Chem. Soc. 2001, 123 (26), 6327−6337.
(21) Ghosh, M.; Singh, K. K.; Panda, C.; Weitz, A.; Hendrich, M. P.;
Collins, T. J.; Dhar, B. B.; Sen Gupta, S. J. Am. Chem. Soc. 2014, 136
(27), 9524−9527.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b01937.
(22) Singh, K. K.; Tiwari, M. k.; Dhar, B. B.; Vanka, K.; Sen Gupta, S.
Inorg. Chem. 2015, 54 (13), 6112−6121.
(23) Singh, K. K.; Tiwari, M. k.; Ghosh, M.; Panda, C.; Weitz, A.;
Hendrich, M. P.; Dhar, B. B.; Vanka, K.; Sen Gupta, S. Inorg. Chem.
2015, 54 (4), 1535−1542.
Plot of kobs against concentration; GC chromatogram;
UV-vis spectral changes; absorbance vs time plot; plots of
kobs against concentration; plots of absorbance vs time;
plot of kobs vs substrate; table of equilibrium constant,
rate constants, and second-order rate constant of
substituted benzyl alcohols; GC/MS data; and log k′2
vs BDEC−H (PDF)
(24) Kundu, S.; Thompson, J. V. K.; Shen, L. Q.; Mills, M. R.;
Bominaar, E. L.; Ryabov, A. D.; Collins, T. J. Chem.−Eur. J. 2014, 21
(4), 1803−1810.
(25) Sheldon, R. A.; Arends, I. W. C. E.; ten Brink, G.-J.; Dijksman,
A. Acc. Chem. Res. 2002, 35 (9), 774−781.
(26) Stahl, S. S. Angew. Chem., Int. Ed. 2004, 43 (26), 3400−3420.
(27) Whittaker, J. W. Chem. Rev. 2003, 103 (6), 2347−2364.
(28) Wang, Q.; Sheng, X.; Horner, J. H.; Newcomb, M. J. Am. Chem.
Soc. 2009, 131 (30), 10629−10636.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: ss.sengupta@ncl.res.in.
(29) Vaz, A. D. N.; Coon, M. J. Biochemistry 1994, 33 (21), 6442−
6449.
Author Contributions
M.G. and Y.L.K.N. contributed equally.
(30) Oh, N. Y.; Suh, Y.; Park, M. J.; Seo, M. S.; Kim, J.; Nam, W.
Angew. Chem., Int. Ed. 2005, 44 (27), 4235−4239.
(31) Panda, C.; Ghosh, M.; Panda, T.; Banerjee, R.; Sen Gupta, S.
Chem. Commun. 2011, 47 (28), 8016−8018.
(32) Mader, E. A.; Davidson, E. R.; Mayer, J. M. J. Am. Chem. Soc.
2007, 129 (16), 5153−5166.
(33) Ohzu, S.; Ishizuka, T.; Hirai, Y.; Jiang, H.; Sakaguchi, M.; Ogura,
T.; Fukuzumi, S.; Kojima, T. Chemical Science 2012, 3 (12), 3421−
3431.
§
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
S.S.G. acknowledges DST, New Delhi (SERB, EMR/2014/
000106) for funding. M.G. acknowledge CSIR (Delhi) for
fellowship. B.B.Dhar acknowledges DST, New Delhi (SERB/F/
3945/2013-14) for funding.
(34) Nesheim, J. C.; Lipscomb, J. D. Biochemistry 1996, 35 (31),
10240−10247.
(35) Price, J. C.; Barr, E. W.; Glass, T. E.; Krebs, C.; Bollinger, J. M. J.
Am. Chem. Soc. 2003, 125 (43), 13008−13009.
(36) Pan, Z.; Horner, J. H.; Newcomb, M. J. Am. Chem. Soc. 2008,
130 (25), 7776−7777.
REFERENCES
■
(1) Denisov, I. G.; Makris, T. M.; Sligar, S. G.; Schlichting, I. Chem.
Rev. 2005, 105 (6), 2253−2278.
(37) Bell, R. P. Trans. Faraday Soc. 1959, 55 (0), 1−4.
(38) Blanksby, S. J.; Ellison, G. B. Acc. Chem. Res. 2003, 36 (4), 255−
263.
(2) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L. Chem. Rev. 2004,
104 (2), 939−986.
(39) Pestovsky, O.; Bakac, A. J. Am. Chem. Soc. 2004, 126 (42),
(3) Groves, J. T. Proc. Natl. Acad. Sci. U. S. A. 2003, 100 (7), 3569−
13757−13764.
3574.
(40) Rocek, J.; Radkowsky, A. E. J. Am. Chem. Soc. 1968, 90 (11),
2986−2988.
(4) Que, L.; Tolman, W. B. Nature 2008, 455 (7211), 333−340.
(5) Rittle, J.; Green, M. T. Science 2010, 330 (6006), 933−937.
(6) Ortiz de Montellano, P. R. Chem. Rev. 2010, 110 (2), 932−948.
(7) Shaik, S.; de Visser, S. l.; Kumar, D. JBIC, J. Biol. Inorg. Chem.
2004, 9 (6), 661−668.
(41) Seo, M. S.; In, J.-H.; Kim, S. O.; Oh, N. Y.; Hong, J.; Kim, J.;
Que, L.; Nam, W. Angew. Chem., Int. Ed. 2004, 43 (18), 2417−2420.
(42) Panda, C.; Debgupta, J.; Díaz Díaz, D.; Singh, K. K.; Dhar, B. B.;
Sen Gupta, S. J. Am. Chem. Soc. 2014, 136 (35), 12273.
(8) Pan, Z.; Newcomb, M. Inorg. Chem. 2007, 46 (16), 6767−6774.
(9) Krest, C. M.; Onderko, E. L.; Yosca, T. H.; Calixto, J. C.; Karp, R.
F.; Livada, J.; Rittle, J.; Green, M. T. J. Biol. Chem. 2013, 288 (24),
17074−17081.
(10) Kaizer, J.; Klinker, E. J.; Oh, N. Y.; Rohde, J.-U.; Song, W. J.;
Stubna, A.; Kim, J.; Munck, E.; Nam, W.; Que, L. J. Am. Chem. Soc.
̈
2004, 126 (2), 472−473.
(11) Bell, S. R.; Groves, J. T. J. Am. Chem. Soc. 2009, 131 (28),
9640−9641.
(12) Han, J. H.; Yoo, S.-K.; Seo, J. S.; Hong, S. J.; Kim, S. K.; Kim, C.
Dalton Transactions 2005, 2, 402−406.
(13) Que, L. Acc. Chem. Res. 2007, 40 (7), 493−500.
̀
(14) Krebs, C.; GalonicI Fujimori, D.; Walsh, C. T.; Bollinger, J. M.
Acc. Chem. Res. 2007, 40 (7), 484−492.
(15) Nam, W. Acc. Chem. Res. 2007, 40 (7), 522−531.
(16) Chow, T. W.-S.; Wong, E. L.-M.; Guo, Z.; Liu, Y.; Huang, J.-S.;
Che, C.-M. J. Am. Chem. Soc. 2010, 132 (38), 13229−13239.
(17) Prat, I.; Mathieson, J. S.; Guell, M.; Ribas, X.; Luis, J. M.;
̈
Cronin, L.; Costas, M. Nat. Chem. 2011, 3 (10), 788−793.
G
DOI: 10.1021/acs.inorgchem.5b01937
Inorg. Chem. XXXX, XXX, XXX−XXX