Trapping a Highly Reactive Nonheme Iron Intermediate That Oxygenates Strong C-H Bonds with Stereoretention

Article abstract of DOI:10.1021/jacs.5b09904

An unprecedentedly reactive iron species (2) has been generated by reaction of excess peracetic acid with a mononuclear iron complex [Fe^II(CF₃SO₃)₂(PyNMe₃)] (1) at cryogenic temperatures, and characterized spectroscopically. Compound 2 is kinetically competent for breaking strong C-H bonds of alkanes (BDE ≈ 100 kcal·mol^-1) through a hydrogen-atom transfer mechanism, and the transformations proceed with stereoretention and regioselectively, responding to bond strength, as well as to steric and polar effects. Bimolecular reaction rates are at least an order of magnitude faster than those of the most reactive synthetic high-valent nonheme oxoiron species described to date. EPR studies in tandem with kinetic analysis show that the 490 nm chromophore of 2 is associated with two S = 1/2 species in rapid equilibrium. The minor component 2a (～5% iron) has g-values at 2.20, 2.19, and 1.99 characteristic of a low-spin iron(III) center, and it is assigned as [Fe^III(OOAc)(PyNMe₃)]²⁺, also by comparison with the EPR parameters of the structurally characterized hydroxamate analogue [Fe^III(tBuCON(H)O)(PyNMe₃)]²⁺ (4). The major component 2b (～40% iron, g-values = 2.07, 2.01, 1.95) has unusual EPR parameters, and it is proposed to be [Fe^V(O)(OAc)(PyNMe₃)]²⁺, where the O-O bond in 2a has been broken. Consistent with this assignment, 2b undergoes exchange of its acetate ligand with CD₃CO₂D and very rapidly reacts with olefins to produce the corresponding cis-1,2-hydroxoacetate product. Therefore, this work constitutes the first example where a synthetic nonheme iron species responsible for stereospecific and site selective C-H hydroxylation is spectroscopically trapped, and its catalytic reactivity against C-H bonds can be directly interrogated by kinetic methods. The accumulated evidence indicates that 2 consists mainly of an extraordinarily reactive [Fe^V(O)(OAc)(PyNMe₃)]²⁺ (2b) species capable of hydroxylating unactivated alkyl C-H bonds with stereoretention in a rapid and site-selective manner, and that exists in fast equilibrium with its [Fe^III(OOAc)(PyNMe₃)]²⁺ precursor.

Full text of DOI:10.1021/jacs.5b09904

Subscriber access provided by NATIONAL TAIWAN UNIV
Article
Trapping a Highly Reactive Non-heme Iron Intermediate
That Oxygenates Strong C-H Bonds with Stereoretention
Joan Serrano-Plana, Williamson N. Oloo, Laura Acosta-Rueda, Katlyn K. Meier, Begoña Verdejo, Enrique
Garcia-España, Manuel G. Basallote, Eckard Münck, Lawrence Que, Anna Company, and Miquel Costas
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b09904 • Publication Date (Web): 24 Nov 2015
Downloaded from http://pubs.acs.org on November 25, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 25
Journal of the American Chemical Society
1
2
3
4
5
Trapping a Highly Reactive Non-heme Iron Intermediate That Oxygenates
Strong C-H Bonds with Stereoretention
6
7
8
9
Joan SerranoꢀPlana,a Williamson N. Oloo,b Laura AcostaꢀRueda,c Katlyn K. Meier,d Begoña Verdejo,e
Enrique GarcíaꢀEspaña,e Manuel G. Basallote,c,* Eckard Münck,d,* Lawrence Que Jr.,b,* Anna Company,a,*
Miquel Costas a,*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aGrup de Química Bioinspirada, Supramolecular i Catàlisi (QBISꢀCAT), Institut de Química Compuꢀ
tacional i Catàlisi (IQCC), Departament de Química, Universitat de Girona. Campus de Montilivi,
E17071, Girona (Catalonia, Spain). Eꢀmail: anna.company@udg.edu, miquel.costas@udg.edu
bDepartment of Chemistry and Center for Metals in Biocatalysis, University of Minnesota, Minneapolis,
MN 55455 (United States). Eꢀmail: larryque@umn.edu
cUniversidad de Cádiz, Facultad de Ciencias, Departamento de Ciencia de los Materiales e Ingeniería
Metalúrgica y Química Inorgánica, Apdo. 40, 11510 Puerto Real, Cádiz (Spain). Eꢀmail: maꢀ
nuel.basallote@uca.es
dDepartment of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, PA 15213 (Unitꢀ
ed States). Eꢀmail: emunck@cmu.edu
eInstituto de Ciencia Molecular (ICMol). Universidad de Valencia. C/ Catedrático José Beltrán, 2
46980, Paterna (Valencia, Spain)
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 25
1
2
3
Abstract
4
5
6
7
8
9
An unprecedentedly reactive iron species (2) has been generated by reaction of excess peracetic acid with
a mononuclear iron complex [FeII(CF3SO3)2(PyNMe3)] (1) at cryogenic temperatures, and characterized
spectroscopically. 2 is kinetically competent for breaking strong CꢀH bonds of alkanes (BDE ~ 100
kcal·molꢀ1) through a hydrogenꢀatom transfer mechanism, and the transformations proceed with stereoreꢀ
tention and regioselectively, responding to bond strength, as well as to steric and polar effects. Bimolecuꢀ
lar reaction rates are at least an order of magnitude faster than those of the most reactive synthetic highꢀ
valent nonheme oxoiron species described to date. EPR studies in tandem with kinetic analysis show that
the 490ꢀnm chromophore of 2 is associated with two S = ½ species in rapid equilibrium. The minor comꢀ
ponent 2a (∼5% iron) has gꢀvalues at 2.20, 2.19 and 1.99 characteristic of a lowꢀspin iron(III) center and
it is assigned as [FeIII(OOAc)(PyNMe3)]2+, also by comparison with the EPR parameters of the structuralꢀ
ly characterized hydroxamate analogue [FeIII(tBuCON(H)O)(PyNMe3)]2+ (4). The major component 2b
(∼40% iron, gꢀvalues = 2.07, 2.01, 1.95) has unusual EPR parameters and it is proposed to be
[FeV(O)(OAc)(PyNMe3)]2+, where the O–O bond in 2a has been broken. Consistent with this assignment,
2b undergoes exchange of its acetate ligand with CD3CO2H and very rapidly reacts with olefins to proꢀ
duce the corresponding synꢀ1,2ꢀhydroxoacetate product. Therefore, this work constitutes the first example
where synthetic nonꢀheme iron species responsible for stereospecific and site selective C–H hydroxylaꢀ
tion are spectroscopically trapped, and its catalytic reactivity against C–H bonds can be directly interroꢀ
gated by kinetic methods. Therefore, the accumulated evidence indicates that 2 consists mainly of an exꢀ
traordinarily reactive [FeV(O)(OAc)(PyNMe3)]2+ (2b) species capable of hydroxylating unactivated alkyl
C–H bonds with stereoretention in a rapid and siteꢀselective manner, and that exists in fast equilibrium
with its [FeIII(OOAc)(PyNMe3)]2+ precursor.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 25
Journal of the American Chemical Society
1
2
3
Introduction
4
5
6
7
8
9
The oxygenation of nonactivated alkyl CꢀH bonds is a very challenging reaction owing to their notorious
chemically inert character. Consequently, highly reactive species are necessary to break these bonds.1
However, because of their high reactivity, these species can only rarely be detected. In nature, CꢀH hyꢀ
droxylation is mainly performed by ironꢀdependent enzymes, which create highly electrophilic species via
finely controlled partial reduction of the O2 molecule. Although it is now well established that the oxidaꢀ
tion of C–H bonds can be carried out by oxoiron(IV) species generated in the catalytic cycles of both
heme (e.g. cytochrome P450)2,3 and nonꢀheme enzymes (e.g. αꢀketoglutarateꢀdependent enzymes),4,5 an
FeIII–OOH species has been observed and implicated as the C–H oxidizing species in the catalytic cycle
of the antibiotic anticancer drug bleomycin, and in naphthalene dioxygenase, an enzyme that belongs to
the Rieske oxygenase family.6,7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In parallel, biomimetic synthetic chemistry has produced a number of FeIII–OOR (R = H, alkyl, acyl)
compounds,4,8 but not one has been shown capable of directly attacking strong alkyl C–H bonds.9,10,11,12
Taking inspiration from nature, selected nonꢀheme iron complexes have been developed that elicit siteꢀ
selective and stereoretentive CꢀH bond oxidation upon reaction with H2O2. Selectivity and stereospecificiꢀ
ty exhibited by these catalysts provide strong support in favor of the intermediacy of metalꢀbased oxidants
akin to those involved with ironꢀdependent oxygenases. Presumably because of their high reactivity, these
oxidants do not accumulate in solution, and it is challenging to establish their nature.
Herein we report the trapping and spectroscopic characterization of a metastable iron species that is kinetꢀ
ically and catalytically competent to attack strong alkane CꢀH bonds and can carry out this transformation
at unprecedented fast reaction rates and with stereospecificity. This reagent also proves to be sensitive to
the strength of the C–H bond, polar effects and steric constraints, and therefore it bears the characteristics
of a selective C–H hydroxylating agent.
Results and Discussion
Ferrous complex [Fe(CF3SO3)2(PyNMe3)] (1) (Figure 1) was prepared and characterized in the solid state
1
by Xꢀray crystallography and in solution by HꢀNMR spectroscopy (see SI for details). Xꢀray analysis
shows an octahedral ferrous center coordinated to the four nitrogen atoms of the macrocyclic ligand
(PyNMe3) and to two triflates disposed cis to each other. The average FeꢀN distance is 2.2 Å, indicating
1
the presence of a highꢀspin iron(II) center.13,14 The HꢀNMR spectrum of 1 in acetonitrileꢀd3 at 298 K
reveals broad bands that span from 80 to ꢀ30 ppm, demonstrating that the iron center remains in the highꢀ
spin configuration (S = 2) in solution. This is further confirmed by a magnetic moment of 4.94 ꢀB measꢀ
ured by the Evans method. The basic chemical features of 1, namely the presence of a tetradentate aminoꢀ
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 25
1
2
3
4
5
6
7
8
9
pyridine ligand that leaves two labile coordination sites in cis relative position, is the same as those of
iron complexes with general formula [Fe(N4L)(X)2]n+ (representative cases are N4L = BPMEN, BPMCN,
TPA, PDP, PyTACN;15 X = CH3CN or CF3SO3) that mediate selective and stereoretentive CꢀH hydroxꢀ
ylation. While mechanistic studies on these catalysts have implicated oxoiron(V) species as the oxidants
responsible for their catalytic activity, the characterization of such species has remained elusive because
they do not accumulate in solution.16ꢀ19
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Treatment of 1 with 10 equiv peracetic acid (AcOOH) in the presence of excess cyclohexane at 25 °C
under air resulted in the formation of cyclohexanol and cyclohexanone in a 5:1 ratio with a total turnover
number of 9.5. Under analogous conditions, adamantane oxidation occurs with a high selectivity for the
tertiary C–H site (normalized 3º/2º ratio = 23), while hydroxylation of the tertiary C–H bonds of cisꢀ1,2ꢀ
dimethylcyclohexane occurs with a high degree of stereoretention (92%). The sum of these observations
suggests that the combination of 1 and peracetic acid generates a metal centered oxidant capable of atꢀ
tacking strong C–H bonds. Furthermore the regioꢀ and stereospecificity of the reactions excludes the inꢀ
volvement of unselective HO• or related species and longꢀlived carbonꢀcentered radicals.20 To gain insight
into the nature of the oxidant, we monitored the reaction of 1 with 4 equiv peracetic acid at ꢀ41 ºC in aceꢀ
tonitrile by UVꢀVis spectroscopy and observed the rapid formation of a metastable brown species 2 (t1/2
=
70 s) with visible absorption features at 490 and 660 nm having a 7:1 relative absorbance ratio (Figure 1).
Figure 1. UVꢀVis spectra of 1 (1 mM, dashed line) and 2 (violet line). Solid black lines show the progresꢀ
sive formation of 2 upon addition of 4 equiv peracetic acid to 1 at ꢀ41 oC in acetonitrile over the course of
1 minute. Inset: the thermal ellipsoid plot (50% probability) of [FeII(CF3SO3)2(PyNMe3)] (1). Hydrogen
atoms have been omitted for clarity.
4
ACS Paragon Plus Environment

Page 5 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Kinetic analysis of the formation of 2 by stopped-flow UV-vis spectroscopy. To get more information
on the formation of intermediate 2, detailed cryoꢀstoppedꢀflow experiments were carried out using a diꢀ
odeꢀarray detector. The reaction of 1 with peracetic acid in acetonitrile solution at ꢀ35 ºC shows spectral
changes that clearly reveal multiꢀphasic kinetics in which intermediate 2 forms and then decays (Figure
S6ꢀS7). Global fitting analysis requires the involvement of another intermediate (B) preceding formation
of 2 in order to obtain a satisfactory fit of the spectral changes, but B has no visible absorption (Figure
S8). The kinetic model consists of three consecutive steps (A→B →C→D), where A = 1, C = 2 and D is
the species formed upon decay of 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Under pseudoꢀfirst order conditions of excess oxidant, the rate constant for the first resolved step (A→B)
shows a linear dependence on the concentration of AcOOH (Figure S9), which yields a second order rate
constant kA→B = 54 ± 1 Mꢀ1 sꢀ1 at ꢀ35 ºC. During this phase, the appearance of a band at 330 nm and the
absence of spectral features in the visible region suggest that this step consists of the oxidation of the
starting complex to form an FeIII species that corresponds to intermediate B. B then converts to 2 with a
rate constant that does not show a clear dependence on the concentration of the oxidant (kB→C = 0.12 ±
0.06 sꢀ1). The large uncertainty associated with kB→C value is probably due to the existence of some paralꢀ
lel reaction of intermediate B, as revealed by the small deviation of the experimental kinetic traces at the
absorption maximum of 2 from those expected for the 3ꢀstep kinetic model. Nevertheless, conversion of
B to 2 requires the presence of more oxidant, as 2 does not form unless supraꢀstoichiometric amounts of
AcOOH are added. The rate constant for the third step, which leads to decomposition of intermediate 2, is
also independent of the concentration of oxidant and shows a value of kC→D= 0.007 ± 0.001 sꢀ1.
Because peracetic acid solutions contain additional species (H2O2, acetic acid, water), it cannot be ruled
out that some of the observed kinetic features may be caused by those additional species. To clarify this
potential ambiguity, some kinetic experiments in acetonitrile solution were also carried out with pernonaꢀ
noic acid, which can be obtained as a pure solid. Reaction of 1 with 10 equiv. of pernonanoic acid in aceꢀ
tonitrile at ꢀ40 ºC produced a metastable species 2’ with analogous spectroscopic features as 2 (see supꢀ
porting information for details). Again, an adequate fit of the spectral changes observed during the reacꢀ
tion of the starting complex 1 with the oxidant requires a model with three consecutive kinetic steps, the
first one showing a second order rate constant of kA→B = 11.9 ± 0.6 Mꢀ1sꢀ1 (Figure S13a) and the third one
being independent of the concentration of pernonanoic acid, kC→D = (2.3±0.3)×10ꢀ3 sꢀ1. Importantly, unꢀ
like with peracetic acid, the rate of the second step in this case does show a dependence on the concentraꢀ
tion of the oxidant (kB→C = 1.1±0.1 Mꢀ1sꢀ1, Figure S13b), supporting our suspicion that the second step in
the peracetic acid reaction is complicated by one or more of the additional species present in solution.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 25
1
2
3
4
Importantly, the second order rate constants for the reactions of alkanes with 2 generated from either
peracetic or pernonanoic acid are congruent.
5
6
7
8
9
Kinetic analysis of the reaction of 2 with alkanes by stopped-flow UV-vis spectroscopy. The decay of
2 is significantly accelerated by the addition of cyclohexane (Figure S14), requiring the use of stoppedꢀ
flow UVꢀVis methods to follow the disappearance of 2 (490 nm) as a function of cyclohexane concentraꢀ
tion (Figure 2a). The kinetic traces show pseudoꢀfirst order behavior and can be fitted with single expoꢀ
nentials. The corresponding kobs values are in turn found to be linearly dependent on substrate concentraꢀ
tion (Figure 2a), affording a second order rate constant (k2) of 2.8±0.1 Mꢀ1sꢀ1 at ꢀ41 ºC. Analysis of the
reaction rates of 2 with cyclohexane as a function of temperature (Figure S15) reveal an activation barrier
with a relatively small activation enthalpy (ꢁH≠ = 37 ± 3 KJ·molꢀ1) and a negative activation entropy (ꢁS≠
= ꢀ76 ± 8 J·Kꢀ1·molꢀ1), consistent with a bimolecular reaction. Comparison of the k2 values for substrates
with weaker C–H bonds show an inverse correlation between the logarithms of second order kinetic conꢀ
stants (normalized by the number of equivalent C–H bonds that can be cleaved) versus BDECꢀH (bond
dissociation energy) of the weakest C–H bond of the hydrocarbon (Figure S18). These results demon-
strate that C–H bond cleavage by 2 is an important component of the rate determining step and that even
unactivated C-H bonds can be cleaved by 2 at fast reaction rates.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2 a) Kinetic data for the reaction of 2 with cyclohexane at ꢀ35 ºC (cyclohexane, filled circles; cyꢀ
clohexaneꢀd12, open circles). b) Reactivity of 2 with various substrates demonstrating its selectivity.
Figure 2b shows that 2 exhibits significant selectivity in its reactions with various substrates. For the oxiꢀ
dation of cyclohexane, cyclohexanol (A) and cyclohexanone (K) products were obtained with an A/K
ratio of 4.5. In the case of adamantane, 1ꢀadamantanol and 2ꢀadamantanol/one were produced with a
6
ACS Paragon Plus Environment

Page 7 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
normalized 3º/2º selectivity of 29, while the hydroxylation of cisꢀ1,2ꢀdimethylcyclohexane to transꢀ1,2ꢀ
dimethylcyclohexanol occurred with >95% retention of configuration. In addition, the oxidation of 1,1ꢀ
dimethylcyclohexane yielded a mixture of C2, C3 and C4 alcohols and ketones in a relative ratio of
10:50:40, which showed that the sterically more congested C2 site was less prone to attack. Lastly, the
hydroxylation of 1ꢀbromoꢀ3,7ꢀdimethyloctane resulted in the preferential attack of the more electronꢀrich
C7ꢀH bond over the C3ꢀH bond by a factor of 8. Taken together, these observations demonstrate a metalꢀ
based oxidant that discriminates among multiple CꢀH bonds based not only on the C–H bond strength but
also on steric and polar factors.21ꢀ26
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The significance of the ability of 2 to react with substrates having strong C–H bonds becomes more obviꢀ
ous when its reaction rates are compared with those determined for other ironꢀbased oxidants. Thus far,
hydroperoxoiron(III) species have only been shown capable of attacking weak CꢀH bonds such as those
of xanthene.11,27 Cleavage of much stronger CꢀH bonds such as those of cyclohexane has been described
only for a few highꢀvalent nonheme oxoiron complexes. Surprisingly, as shown in Figure 3 and Table 1, 2
is by far the most reactive of such species. The reaction of 2 with cyclohexane at ꢀ41 °C is at least 4 orꢀ
ders of magnitude faster than several S = 1 oxoiron(IV) complexes.28,29 Also, when compared to the two
previously reported oxoiron(V) complexes supported by tetraamido macrocyclic ligands (TAML), the
reaction rate for 2 is respectively 4 and 2 orders of magnitude faster,30,31 even without correcting the 65ꢀ
degree difference in temperature in the second case. Furthermore, 2 is more than 200 times more reactive
than an iron(III)ꢀiodosylarene adduct, which can be construed as a masked oxoiron(V) species.32 The
complexes that most closely approach 2 in C–H bond cleavage reactivity are the S = 1 complex
[FeIV(O)(Me3NTB)]2+ (which is proposed to operate through a highly reactive high spin (S = 2) excited
state) and the recently described S = 2 complex [FeIV(O)(TQA)(NCMe)]2+; both react with cyclohexane
an order of magnitude more slowly than 2 (Table 1).33,34 Thus 2 is the most effective species observed to
date in solution that can oxidize cyclohexane.
Intermediate 2 stands out from the other complexes in Figure 3 in yet another respect. As shown in Figure
2a, 2 oxidizes cyclohexane and cyclohexaneꢀd12 at different rates, exhibiting a classical kinetic isotope
effect (KIE) of 5. This value is obtained as well from the product analysis of competitive oxidations of
cyclohexane and cyclohexaneꢀd12, corroborating that 2 is in fact the actual C–H bond cleaving agent. In
contrast, the other complexes in Figure 3 carry out C–H bond cleavage with large nonclassical KIE valꢀ
ues. The subset consisting of the oxoiron(IV) complexes exhibits KIE values of 25ꢀ50 that have attributed
to Hꢀatom tunneling,28,35 while the other subset comprising the two oxoiron(V) complexes and the
iron(III)ꢀiodosylarene adduct exhibit lower values of 9ꢀ26 (Table 1). As KIE values can depend on temꢀ
perature and the nature of the substrate, the best comparison is for the oxidation of cyclohexane at ꢀ40 °C
between 2 and [FeIV(O)(TQA)(NCMe)]2+ where the KIE values differ by a factor of 5. What this differꢀ
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 25
1
2
3
4
5
6
7
8
9
ence tells us about the distinct natures of the oxidants and how they cleave C–H bonds remain to be unꢀ
covered, but it is clear that there is a difference. On the other hand, the classical KIE value of 5 found for
2 matches those found for nonhemeꢀironꢀcatalyzed hydroxylation of cyclohexane by H2O2. For these cataꢀ
lysts, there is strong indirect evidence for an FeV(O) oxidant, but it has not yet been spectroscopically
identified.16,36 In at least three cases, evidence for the FeV(O) oxidant has been obtained in the gas phase
by electrospray ionization mass spectrometry.16,18,37 In a few cases, hydroperoxyꢀ or acylperoxyiron(III)
intermediates have been observed and spectroscopically characterized;38ꢀ41 unlike 2, these intermediates
persist in a steadyꢀstate phase at ꢀ40 °C, during which oxidation products are formed catalytically. The
decay occurs upon depletion of H2O2 at a rate independent of the nature of the substrate and its concentraꢀ
tion. This kinetic behavior makes them precursors to the actual oxidant. In contrast, 2 is unique as the
only catalytic intermediate shown to effect C–H bond cleavage directly.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Alkane oxidation reactions by selected nonꢀheme iron highꢀvalent species.
k2 (cyclohexane)
A/Kb
(cyclohexane)
Highꢀvalent species
2
KIEc
ref
(Mꢀ1sꢀ1) (T)a
2.8(1) (ꢀ40 °C)
0.37 (ꢀ40 °C)
1/0.2
5 (cyclohexane)
25 (cyclohexane)
26 (PhEt)
this work
[FeIV(O)(TQA)(NCMe)]2+
[FeIV(O)(Me3NTB)]2+
FeIII(13ꢀTMC)/OIAr adduct
[FeV(O)(TAML)]–
[FeV(O)(bTAML)]–
[FeIV(O)(PyTACN)]2+
[FeIV(O)(BnTPEN)]2+
[FeIV(O)(N4Py)]2+
0/0.35
34
33
32
30
31
29
28
28
0.25 (ꢀ40 °C)
ꢀ
0.011 (ꢀ40 °C)
0.00026 (ꢀ40 °C)
0.023 (25 °C)
0.0004 (25 °C)
0.00039 (25 °C)
0.11/0.04
10 (cumene)
26 (PhEt)
ꢀ
0.4/0.02
ꢀ
9 (PhCH3)
27 (DHA)
50 (PhEt)
0.03/0.13
0.000055 (25 °C) 0.03/0.04
27 (PhEt)
b
aSecondꢀorder rate constant determined in cyclohexane oxidation. Equivalents of cyclohexanol (A) and
c
cyclohexanone (K) with respect to Fe obtained in the oxidation of cyclohexane. KIE = kinetic isotope
effect. Ligand abbreviations: TQA = tris(2ꢀquinolylmethyl)amine; Me3NTB = tris((Nꢀmethylꢀ
b
benzimidazolꢀ2ꢀyl)methyl)amine; TAML = tetraamido macrocyclic ligand; TAML = biuret tetraamidate
macrocyclic ligand; 13ꢀTMC = 1,4,7,10ꢀtetramethylꢀ1,4,7,10ꢀtetraazacyclotridecane; PyTACN = 1,4ꢀ
dimethylꢀ7ꢀ(2ꢀpyridylmethyl)ꢀ1,4,7ꢀtriazacyclononane;
BnTPEN
=
NꢀbenzylꢀN,N′,N′ꢀtris(2ꢀ
pyridylmethyl)ꢀ1,2ꢀdiaminoethane; N4Py = N,Nꢀbis(2ꢀpyridylmethyl)ꢀNꢀ(bisꢀ2ꢀpyridylmethyl)amine.
8
ACS Paragon Plus Environment

Page 9 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Cyclohexane oxidation rates (k2) obtained at ꢀ40 or 25 °C for various nonheme iron complexes.
Dark blue bars represent k2 values at ꢀ40 °C, while light blue bars correspond to k2 values at 25 °C.
EPR insights into the nature of 2. EPR spectroscopy turned out to be an excellent tool to characterize
the nature of the iron center in 2. We have studied a series of EPR samples prepared at ꢀ50 °C using a 1:3
(v/v) acetonitrile:acetone glassing solvent mixture.42 This procedure yielded well resolved EPR features in
the g = 2 region that could be monitored concurrent with the growth and decay of the 490ꢀnm chromoꢀ
phore (Figure 4a and Figures S28ꢀS32). At maximum accumulation of this visible chromophore, three
major EPRꢀactive species can be observed. There is a highꢀspin iron(III) species with signals in the g = 9
and 4ꢀ6 regions that account for ~40% of total Fe; this species exhibits a time course that does not correꢀ
late at all with the kinetic behavior of 2, suggesting that it is associated with side reactions often observed
in this kind of chemistry. This highꢀspin iron(III) species can also be observed in the Mössbauer spectrum
of the same sample (Figure S33) and represents ~40% of the total Fe in the sample as well.
There are also two S = ½ species. The minor component, representing 5% of total Fe, exhibits g values at
2.20. 2.19, and 1.99 (2a, gmax = 2.20 species), and its simulated spectrum is shown in blue in Fig. 4a. The
major component, representing ≈ 40% of total Fe, has gꢀvalues at 2.07, 2.01, and 1.95 (2b, gmax = 2.07
species); its simulation is shown in red in Fig. 4a. Significantly, the growth and decay of both EPR speꢀ
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 25
1
2
3
4
5
6
7
8
9
cies closely track the time course of the 490ꢀnm chromophore (Figure 4b). More importantly, addition of
50 equiv. cyclohexane at ꢀ50 °C accelerated the decay of the 490ꢀnm chromophore, in tandem with the
decay of both 2a and 2b. Upon complete decay of the 490ꢀnm chromophore, a new lowꢀspin iron(III)
center at g values at 2.21, 2.14, and 1.97 (labeled the gmid = 2.14 species) was observed, which represents
28% of total Fe. However its formation is not directly connected with the decay of the 490ꢀnm chromoꢀ
phore either in the absence or the presence of cyclohexane (Table S6). The most important observation
from this study is that both 2a and 2b are kinetically connected with the 490ꢀnm chromophore and are
therefore likely to be species that are in equilibrium with each other. (For more detailed discussion, see
text associated with Table S6 and Figures S28ꢀS32.)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. A) Xꢀband EPR spectrum of a 1 mM solution of 1 in 1:3 (v/v) acetonitrile:acetone reacted with
5 equiv. AcOOH at ꢀ50 °C and frozen at t = 120 s. EPR conditions: T = 20 K, 0.2 mW microwave power,
1 mT modulation. Shown is the g = 2 region of the experimental spectrum (black). The theoretical curves
(red and blue) are S = ½ SpinCount simulations. Blue: gmax = 2.20 species (2a) simulated with g = 2.20,
2.19, 1.99 (5% of Fe). Red: gmax = 2.07 species (2b) simulated with g = 2.07, 2.01, 1.95 (40% of Fe). The
sharp feature at g = 2.00 belongs to an unknown minor species (< 2%). Spectra along the time course that
also include the low field region are shown in SI. B) Time course of 2 generated by adding 5 equiv.
AcOOH to a 1 mM solution of 1 in 1:3 (v/v) acetonitrile:acetone at ꢀ50 ºC. Black squares: formation and
10
ACS Paragon Plus Environment

Page 11 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
selfꢀdecay of the 490ꢀnm chromophore. Grey line shows the decay of the chromophore after the addition
of cyclohexane (50 equiv) upon maximum formation of 2 (near 120 s). Red open and filled circles repreꢀ
sent the respective amounts of the gmax = 2.20 and gmax = 2.07 species (2a and 2b) during the formation
and selfꢀdecay of the 490 chromophore. Blue open and filled circles respectively mark the time course of
the gmax = 2.20 and 2.07 species (2a and 2b) during the decay of 2 after the addition of 50 equivalents of
cyclohexane.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
At near maximal development of the 490ꢀnm chromophore, the two isomeric S = ½ species in equilibrium
with each other account for ≈ 45% of the Fe. If only the gmax = 2.20 species (2a) contributes to the 490ꢀ
nm band, it would have an unreasonably large ε490 of 36,000 Mꢀ1cmꢀ1, whereas a more reasonable ε490
≈
4500 Mꢀ1cmꢀ1 results if only the 2.07 species (2b) contributes to the absorption band. Clearly, 2b must be
associated with the 490 nm band, if not alone then in concert with 2a. It is interesting to note that an ε490
≈ 4500 Mꢀ1cmꢀ1 (assuming that both 2a and 2b contribute to the chromophore) would compare well with
that of the 460ꢀnm chromophore (ε460 ≈ 4000 Mꢀ1cmꢀ1) recently found in the reaction of [FeII(TPA*)]2+
with H2O2/AcOH or AcOOH at ꢀ40ºC, which is associated with an S = ½ acylperoxoiron(III) intermediate
[FeIII(OOAc)(TPA*)]2+ (3, TPA* = tris(3,5ꢀdimethylꢀ4ꢀmethoxyꢀ2ꢀpyridylmethyl)amine), even though the
latter has a much greater EPR anisotropy (g = 2.58, 2.38, 1.72).40 It should also be noted that Talsi et al.
have recently found a similar transient species in epoxidation reactions with the [Fe(TPA*)] catalyst perꢀ
formed at ꢀ65 ºC with gꢀvalues like those of the gmax = 2.07 species.43 However this species accumulates
to only ∼1% of total Fe in the [Fe(TPA*)] sample.
By analogy to the 460ꢀnm chromophore 3 derived from the Fe(TPA*) catalyst, we first considered the
assignment of the 490ꢀnm intermediate 2 to a lowꢀspin [FeIII(OOAc)(PyNMe3)]2+ complex. In previous
work, we reported that the Mössbauer and EPR data of 3, having g = (2.58, 2.38, 1.72), fit well to the
GriffithꢀTaylor model, which generally describes the gꢀvalues of lowꢀspin FeIII complexes quite well.44,45
This model considers spinꢀorbit coupling within a T2g set that is energetically well separated from the Eg
set and lowꢀlying charge transfer states. The assignment of 3 as a lowꢀspin S = ½ acylperoxoiron(III) was
corroborated by resonance Raman and ESIꢀMS experiments.40 To serve as an analog for the putative
[FeIII( 2ꢀOOAc)(PyNMe3)]2+ species, the [FeIII(tBuCON(H)O)(PyNMe3)]2+ complex (4) has been syntheꢀ
κ
sized and structurally characterized, where a hydroxamate replaces the bidentate peracetate ligand (see SI
for details). Complex 4 exhibits an axial EPR spectrum with g = 2.21 and 1.94 (Figure S25) that is similar
κ
to that of 2a, suggesting that the latter arises from the [FeIII( 2ꢀOOAc)(PyNMe3)]2+ complex.
In stark contrast, the gꢀvalues of 2b do not fit to the GriffithꢀTaylor model. To obtain the observed g valꢀ
ues would require T2g splittings in excess of 10000 cmꢀ1, and the resonance at g = 2.01 would be especially
troublesome to fit.44,45 While this argument does not necessarily rule out the possibility of 2b as a lowꢀ
spin FeIII complex, the small spread of its gꢀvalues (2.07, 2.01, 1.95) would indicate a very unusual nonꢀ
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 25
1
2
3
4
heme lowꢀspin iron(III) center that has not yet been observed.35 So, an alternative description needs to be
considered.
5
6
7
8
Comparison of the gꢀvalues of 2b with those reported for a handful of FeV(O)(L)31,43,46,50 and
FeIV(O)(L•+)52ꢀ54 species (where L•+ stands for a ligand radical cation) suggests a possible solution (Table
2). In particular, the gꢀvalues of 2b and [FeV(O)(NC(O)Me)(TMC)]+ (5, g = 2.053, 2.010, 1.971) closely
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
57
resemble each other.46 Furthermore, both exhibit highly anisotropic Fe Aꢀtensors. Examination of the
EPR spectrum of an 57Feꢀenriched sample of 2 revealed a large magnetic hyperfine interaction of 47 MHz
at gmid = gy = 2.01 and two much smaller components at gmax and gmin (see Figure S34). For comparison,
57
the EPR spectrum of an Feꢀenriched sample of 5 also showed a large component of 47 MHz, but this
was observed along the direction of gmax = gx = 2.05 (the unpaired electron of the FeV is in dyz),47 suggestꢀ
ing some (possibly significant) differences between 2b and 5. Nevertheless, these two complexes are
clearly different from the S = ½ FeIV(O)(L•+) species such as Compounds I of horseradish peroxidase,
chloroperoxidase, and cytochrome P450 (Table 2),48,49 which all have axial 57Fe Aꢀtensors with |Ax| ≈ |Ay|
>> Az. Both the dyz and dxz orbitals of the S = 1 FeIV(O) complexes are singly occupied, in contrast to the
S = ½ FeV(O) centers, which have only one unpaired electron in either the dxz or dyz orbital (with dxy douꢀ
bly occupied). An x/y anisotropy is also observed for [FeV(O)(TAML)]–, the first characterized example
of a bona fide oxoiron(V) complex (Table 2).50 We can thus safely conclude that 2b cannot arise from an
57
FeIV(O)(L•+) complex. However, the presence of a highly anisotropic Fe Aꢀtensor does not allow us to
distinguish between an S = ½ peroxoiron(III) and an S = ½ oxoiron(V) center, as similar anisotropies of
the 57Fe Aꢀtensor have been observed for some lowꢀspin iron(III) complexes having gꢀvalues close to g =
2.0, such as the S = ½ [FeIII(η1ꢀOOH)(N4Py)]2+ complex (Table 2).51
At this point, our usual strategy would have been to carry out a complementary Mössbauer analysis in
order to distinguish between an S = ½ peroxoiron(III) and an S = ½ oxoiron(V) center. However our atꢀ
tempts to identify unique Mössbauer features associated with the gmax= 2.07 species (2b) were stymied by
the spectral complexity of samples (Figure S33) that have at least four EPRꢀactive species, all of which
would be expected to exhibit Mössbauer spectra with magnetic hyperfine features even without an apꢀ
plied magnetic field (See SI for further comments). What emerges from our EPR studies is that 2b must
have an unusual electronic structure, either in an S = ½ iron(III) description with a small spread of gꢀ
values that has yet to be observed for a nonꢀheme lowꢀspin iron(III) center35 or an S = ½ oxoiron(V) deꢀ
scription for which there are only three well characterized examples for comparison, namely the TMC
complex and the two TAML complexes listed in Table 2.
Table 2. EPR properties of established and proposed FeV(O)(L) and FeIV(O)(L•+) complexes and related
peroxoiron(III) species.
12
ACS Paragon Plus Environment

Page 13 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
57Fe Ax, Ay
EPR g values
Complex a
ref
50
(MHz)
[FeV(O)(TAML)]–
[FeV(O)(bTAML)]–
1.99, 1.97. 1.74
1.98, 1.94, 1.73
2.05, 2.01, 1.97
2.07, 2.01, 1.96
1.97, 1.93, 1.91
2.09, 2.05, 2.02
2.1, 2.1, 2
ꢀ49.3, ꢀ1.5
31
46
[FeV(O)(NC(O)Me)(TMC)]+ (5)
[FeV(O)(TPA*)]3+ b
[FeIV(O)(Clꢀacac•+)(Me3TACN)]2+ c
[FeIV(O)(TBP8Cz•+)]
[FeIV(NTs)(TBP8Cz•+)]
HRPꢀCpd I
ꢀ47, ꢀ17
43
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
52
53
54
48
49
1.99 (broad)
ꢀ26, ꢀ26
CPOꢀCpd I
2.00, 1.72, 1.61 ꢀ31 T, ꢀ30 T
49
40
CYP119ꢀCpd I
[FeIII(κ2ꢀOOAc)(TPA*)]2+ (3)
[FeIII(η1ꢀOOH)(N4Py)]2+
2.00, 1.96, 1.86 ꢀ28 T, ꢀ32 T
2.58, 2.38, 1.72 ꢀ45 T, +19 T
51
2.16, 2.11, 1.98
ꢀ9, ꢀ53
2.07, 2.01, 1.95 Ax = 47 >> Ay
2a
2b
this work
2.20, 2.19, 1.99 not determined
b
aAbbreviations used: TAML = tetraamidate macrocyclic ligand, TAML = biuret tetraamidate macrocyꢀ
clic ligand, TMC = tetramethylcyclam, TPA* = tris(3,5ꢀdimethylꢀ4ꢀmethoxyꢀpyridylꢀ2ꢀmethyl)amine,
Me3TACN = 1,4,7ꢀtrimethylꢀ1,4,7ꢀtriazacyclononane, Clꢀacac = 3ꢀchloroꢀacetylacetonate, TBP8Cz =
octakis(4ꢀtertꢀbutylphenyl)corrolazine, HRP = horseradish peroxidase; CPO = chloroperoxidase, CYP119
= cytochrome Pꢀ450 from Sulfolobus acidocaldarius, N4Py =
bis(2ꢀpyridylmethyl)ꢀbis(2ꢀ
pyridyl)methylamine. bThis species has the same gꢀvalues as 2b but represents only ca. 1% of the iron in
the sample. The iron(V) oxidation state proposed for this species has not established.
cThe proposed formulation for this intermediate needs further spectroscopic corroboration.
The insights derived from the EPR analysis must be interpreted within the context of the kinetic results,
which show that 2a and 2b both rise and fall in concert with the 490ꢀnm chromophore and are thus in
equilibrium with each other. If 2b is assigned to a lowꢀspin iron(III) center with an unusual electronic
structure, it would most likely be a geometric isomer of the acylperoxoiron(III) center associated with the
2a. Such a geometric isomer would presumably have the bidentate acylperoxo ligand bound trans to difꢀ
ferent N atoms of the supporting PyNMe3 macrocycle. However, the subsequent rate determining cleavꢀ
age of the strong substrate C–H bond would have to be effected by an acylperoxoiron(III) center, which is
thus far unprecedented. On the other hand, assignment of 2b to an S = ½ FeV(O) center would make it an
electromer of the acylperoxoiron(III) center associated with 2a that would form by a reversible O–O bond
cleavage step (Scheme 1). There is experimental precedent for such equilibria as well as DFT calculations
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 25
1
2
3
4
5
6
7
showing that higherꢀvalent electromers of acylperoxoiron(III) species are energetically accessible (see
subsequent section for discussion of these topics). The proposed oxoiron(V) species that we favor for the
assignment of the gmax = 2.07 species (2b) would certainly have the oxidative power required to cleave
strong C–H bonds directly, as demonstrated in this paper for the 490ꢀnm chromophore.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Reversible OꢀO bond cleavage of acylperoxoiron(III) species 2a to oxoiron(V) species 2b.
Mass spectroscopic insights into the nature of 2. Additional experimental evidence has been obtained
in solution at ꢀ40 °C corroborating the formation of a FeV(O)(O2CR) derivative that is in equilibꢀ
rium with the acylperoxoiron(III) intermediate. Highꢀresolution cryospray mass spectrometry
(CSIꢀMS) experiments at ꢀ40 ºC showed 2 to exhibit a major peak at m/z 528.0893 with an assoꢀ
ciated isotopic pattern fully consistent with its formulation as {Fe(OOAc)(PyNMe3)](CF3SO3)}+
(Fig. 5a) or either of its higherꢀvalent electromers. CD3COOD was added into the 1/AcOOH reacꢀ
tion mixture during the formation of 2 at ꢀ40 °C. This experiment resulted in the appearance of a
new feature at m/z 531 (Fig. 5b), indicating the substitution of the acetate (CH3CO2) by d3ꢀacetate
(CD3CO2). Control experiments showed that CD3CO2D and AcOOH did not exchange under the
reaction conditions (see SI), so there must be a pathway for the incorporation of labeled acetate
into the primary mass peak associated with 2. We propose carboxylate exchange with the FeV(O)
electromer as representing the most facile pathway that rationalizes the observed deuterium inꢀ
corporation into 2 from CD3COOD.
14
ACS Paragon Plus Environment

Page 15 of 25
Journal of the American Chemical Society
1
2
3
4
a
528.0893
5
6
7
Experimental
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
523
528
533
528.0947
Calculated
for 2
523
528
m/z
533
b
528
531
+3
523
528
533
Figure 5 Cryospray mass spectral (CSIꢀMS) experiments. a) Experimental and calculated patterns of 2
generated by reaction of 1 with 4 equiv peracetic acid in acetonitrile at ꢀ40 ºC. b) Spectrum of 2 generated
with 4 equiv peracetic acid in the presence of 20 equiv acetic acidꢀd4 at ꢀ40 ºC. An M+3 ion was observed
at m/z = 528, indicating that the CD3CO2 can be incorporated into this species. See Figures S35ꢀS36 for
full spectra.
Corroboration for carboxylate exchange with 2 was obtained by investigating the reactions of 1/AcOOH
with cyclooctene in the presence of CD3COOD. In room temperature reactions with 20 equiv. AcOOH
and no added CD3COOD, cyclooctene was converted to cyclooctene oxide as the major product (17
TON) and syn-2ꢀacetoxycyclooctanol as the minor product (2 TON). The latter has been reported in cyꢀ
clooctene oxidations catalyzed by Fe(TPA) and Fe(BPMCN) with H2O2 in the presence of AcOH15,55,56
and was proposed to arise from a [3 + 2] cycloaddition of a fleeting FeV(O)(OAc) oxidant to the C=C
bond of the olefin substrate. It should be noted that the exclusive synꢀstereochemistry observed for the
product is consistent with this scenario but not with a twoꢀstep mechanism involving initial epoxidation
followed by epoxide ring opening. A control experiment in the absence of 1 afforded ∼ 5 times less epoxꢀ
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 25
1
2
3
4
5
6
7
8
9
ide product but, importantly, there was no trace of synꢀ2ꢀacetoxycyclooctanol. The syn-2ꢀ
acetoxycyclooctanol product observed was 10% trideuterated in reactions of 1 and AcOOH with 200
equiv. cyclooctene in the presence of 200 equiv. CD3COOD, both at ꢀ40 °C and at 25 °C, and also upon
addition of 200 equiv. CD3COOD and 200 equiv. cyclooctene after maximal formation of 2 at ꢀ40 °C,
indicating that the acetate exchange in 2 occurs very fast (see Section VII of SI). Up to 29% of labeled
product was obtained after increasing the CD3COOD amount to 500 equiv at 25 ºC. These results clearly
demonstrate that there must be a mechanism for the incorporation of deuterated acetate into the oxidant
generated from 1 and AcOOH.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Comparison with literature precedents for acylperoxoiron(III) complexes. Intermediate 2 resembles
in several respects the closely related lowꢀspin acylperoxoiron(III) intermediate 3 recently characterized
at ꢀ40 °C in the reaction of FeII(tpa*) with H2O2/AcOH40 but differs in others. Both 2 and 3 exhibit simiꢀ
larly intense visible chromophores in the 450ꢀ500 nm region that are associated with S = ½ EPR signals
and undergo selfꢀdecay in MeCN at ꢀ40 °C with kobs values of 0.02 sꢀ1. However, the reactivity of 2 signifꢀ
icantly differs from that of 3. Upon its generation, 3 persists in a steady state phase and undergoes rateꢀ
determining O–O bond cleavage that is substrateꢀindependent. In contrast, 2 decays at a rate that depends
on the nature of the substrate and its concentration, indicating that, unlike 3, 2 directly reacts with alꢀ
kanes. Moreover, the substrateꢀdependent decay of 2 is associated with two S = ½ species with gmax = 2.20
and 2.07 (2a and 2b), respectively. These two species maintain a constant intensity ratio as 2 decays at
different rates in the course of oxidizing various substrates. To accommodate these observations, we proꢀ
pose the possibility of a reversible O–O bond cleavage equilibrium between the [FeIII(OOAc)(PyNMe3)]2+
(2a, gmax = 2.20) and the highly oxidizing highꢀvalent ironꢀoxo species [FeV(O)(OAc)(PyNMe3)]2+ (2b,
gmax =2.07) which acts as the actual oxidant (Scheme 2). Reversible cleavage of the O–O bond has in fact
been demonstrated for a peroxocarbonatoiron(III) complex by Suzuki and coꢀworkers, resulting in the
incorporation of labeled water into the bound peroxocarbonate moiety and label scrambling.57 It is highly
unlikely that the observed incorporation of deuterated acetate into 2 occurs by direct exchange with the
acylperoxoiron(III) species 2a, but it could easily come about by ligand exchange of CD3COOD with the
FeV(O)(OAc) electromer 2b. Our experimental results thus support the notion of energetically accessible
higherꢀvalent electromers in equilibrium with the acylperoxoiron(III) species, a notion first proposed by
Shaik and coꢀworkers based on DFT calculations on related systems.17,40 The existence of this equilibrium
allows 2 to attack strong C–H bonds directly and rapidly, a kinetic behavior that makes 2 unique among
the few characterized firstꢀrow transition metal acylperoxo complexes reported so far including iron,58,59
copper,60ꢀ62 and nickel,63 which are generally much more stable and sluggish in cleaving C–H bonds.
16
ACS Paragon Plus Environment

Page 17 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Proposed mechanism for the reaction of 1/AcOOH with substrates.
Comparison of the C–H bond cleavage reactivities of 2 and Cpd I. In Figure 3, 2 was shown to be at
least an order of magnitude faster with respect to HAT from cyclohexane than the most reactive nonheme
ironꢀoxo complex described to date. A comparison can be also made between 2 and Cpd I, the highly
reactive oxoiron(IV) oxidant in heme chemistry. The second order rate constant (k2) of 2.8 Mꢀ1sꢀ1 for the
reaction of 2 with cyclohexane at 232 K is 25ꢀfold faster than that measured for the Cpd I model
[FeIV(O)(TDCPP•)]+) (TDCPP = tetrakis(2,6ꢀdichlorophenyl)porphyrin) at the same temperature.33 On the
other hand, the k2 value for the reaction of 2 with toluene at 298 K of 6×103 Mꢀ1 sꢀ1, as estimated from the
Eyring parameters obtained at lower temperature (ꢁH≠ = 27.3(0.1) kJ molꢀ1, ꢁS≠ = ꢀ82(5) J Kꢀ1 molꢀ1, Figꢀ
ure S19), is only two orders of magnitude smaller than the k2 value for pꢀtoluic acid oxidation recently
determined by Groves for the Cpd I intermediate of a hemeꢀthiolate peroxygenase.64 When the toluene
oxidation rate of 2 is mapped onto a plot of log k2 values of different oxidants towards toluene vs. the
strength of the O–H bond formed,65 an estimate of ~ 101 kcal·molꢀ1 can be made for the FeO–H bond that
forms in the HAT reaction by 2 (Figure S20), which is substantially larger than those determined for
FeIIIO(R) (~ 85 kcal·molꢀ1) and heme FeIV(O) oxidants (~90 kcal·molꢀ1)66,67 and just 2 kcal·molꢀ1 smaller
than the 103 kcal·molꢀ1 value associated with the hemeꢀthiolate peroxygenase Cpd I.64
Summary and Perspectives
We have thus collected a set of spectroscopic and reactivity data on 2 that shed light on a remarkable oxiꢀ
dant 2 that is generated at ꢀ40 °C by the reaction of peracetic acid with the nonheme iron catalyst 1. Oxiꢀ
dant 2 is an order of magnitude faster at cleaving the C–H bonds of cyclohexane than the most reactive
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 25
1
2
3
4
5
6
7
8
nonheme oxoiron(IV) complexes reported thus far33,34 and four orders of magnitude faster than oxꢀ
oiron(V) TAML complexes at the same temperature.30 Oxidant 2 is also capable of catalytic turnover,
unlike the latter complexes, which at best can only carry out stoichiometric substrate oxidation. Furtherꢀ
more, despite its high reactivity, 2 effects siteꢀselective C–H bond functionalization that is responsive to
electronic and steric parameters. Most significantly, in stark contrast to the reactivity of several nonꢀheme
oxoiron(IV) and oxoiron (V) complexes that can cleave strong C–H bonds but generate longꢀlived carꢀ
bonꢀcentered radicals,28,29,31,68 2 carries out C–H hydroxylations with high stereoretention. This observaꢀ
tion is especially remarkable because structurally related iron complexes catalyze this challenging reacꢀ
tion with reactivity patterns analogous to those exhibited by 2.24,25,69 Selectivity in C–H bond reactivity
converts them into very interesting tools for synthetic organic chemistry, yet the active species that acꢀ
count for this selectivity have not been directly observed. Therefore, this work constitutes the first case
where the species responsible for stereospecific and site selective CꢀH hydroxylation with this class of
catalysts is spectroscopically trapped, and its catalytic reactivity against CꢀH bonds could be directly
demonstrated. Results presented here identify the major component of 2, namely 2b, as a species having
unusual electronic structure. The g values of this S = ½ species exhibit a small g anisotropy that is not
observed thus far for a nonꢀheme lowꢀspin iron(III) center but resembles that of a recently described oxꢀ
oiron(V) complex 5. We thus favor the latter formulation as a rationale for the unprecedentedly C–H bond
cleavage reactivity found in the reaction of 1 with peracetic acid.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
Materials. Reagents and solvents used were of commercially available reagent quality unless otherwise
stated. Solvents were purchased from Scharlab, Acros or SigmaꢀAldrich and used without further purifiꢀ
cation. Peracetic acid was purchased from Aldrich as a 32 wt% solution in acetic acid containing less than
6% H2O2. Preparation and handling of airꢀsensitive materials were carried out in a N2 drybox (Jacomex)
with O2 and H2O concentrations < 1 ppm. Synthesis of compound 1 is detailed in the Supporting Inforꢀ
mation.
Physical methods. High resolution mass spectra (HRꢀMS) were recorded on a Bruker MicrOTOFꢀQ
IITM instrument using ESI or Cryospray ionization sources at Serveis Tècnics of the University of Giꢀ
rona. Samples were introduced into the mass spectrometer ion source by direct infusion using a syringe
pump and were externally calibrated using sodium formate. A cryospray attachment was used for CSIꢀMS
(cryospray mass spectrometry). Temperature of the nebulizing and drying gases was set at ꢀ40ºC. The
instrument was operated in positive ion mode. NMR experiments were performed on a Bruker Ultrashield
Avance III400 and Ultrashield DPX300 spectrometers. UV/Vis spectroscopy was performed with an Agꢀ
18
ACS Paragon Plus Environment

Page 19 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
ilent 50 Scan (Varian) UV/Vis spectrophotometer with 1 cm quartz cells. Low temperature control was
achieved with a cryostat from Unisoku Scientific Instruments, Japan. GC product analyses were perꢀ
formed on an Agilent 7820A gas chromatograph equipped with a HPꢀ5 capillary column
30mx0.32mmx0.25ꢂm and a flame ionization detector. Stoppedꢀflow experiments were carried out using
an SFM4000 Bioꢀlogic instrument provided with a cryoꢀstoppedꢀflow accessory fitted to a Huber CCꢀ905
bath.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mössbauer spectra were recorded with two spectrometers, using Janis Research (Wilmington, MA) Suꢀ
perVaritemp dewars that allow studies in applied magnetic fields up to 8.0 T in the temperature range
from 1.5 to 200 K. Mössbauer spectral simulations were performed using the WMOSS software package
(WEB Research, Minneapolis). Perpendicular (9.63 GHz) mode Xꢀband EPR spectra were recorded on a
Bruker EPP 300 spectrometer equipped with an oxford ESR 910 liquid helium cryostat and an Oxford
temperature controller. The quantification of the signals was relative to a CuꢀEDTA spin standard. Softꢀ
ware for EPR analysis was provided by Dr Michael P. Hendrich of Carnegie Mellon University.
Generation of 2. In a typical experiment, a 1 mM solution of 1 in dry acetonitrle was prepared inside the
glovebox. 2 mL of this solution were placed in a UVꢀVis cuvette (2 ꢀmols of 1). The quartz cell was
capped with a septum and taken out of the box, placed in the Unisoku cryostat of the UVꢀVis spectrophoꢀ
tometer and cooled down to 238 K. After reaching thermal equilibrium an UVꢀVis spectrum of the startꢀ
ing complex was recorded. Then, 50 ꢀL of a solution containing 66 ꢀL of 32% peracetic acid in 2 mL of
dry acetonitrile were added (8 ꢀmols). The formation of a band at λmax = 490 nm and a shoulder at λmax
=
660 nm was observed. 2 was fully formed after 70 seconds. In the reactions where the intermediate speꢀ
cies is generated in acetone, a 4 mM solution of the complex in acetonitrile was prepared and then further
diluted with the corresponding amount of dry acetone.
Kinetic analysis of reactions against alkanes. The experiments were carried out by mixing thermostated
acetonitrile solutions of 1 and peracetic acid (1:4 ratio) in a delay line. The resulting solution was aged for
the time required to achieve the maximum concentration of 2 and then it was mixed with a solution of the
substrate. The concentration of the substrate in the observation cell was changed by using different ratios
of the volumes in the second mixing. In any case, the substrate concentration was always in pseudoꢀfirst
order excess with respect to 2, and the kinetics of reaction was monitored by measuring the spectral
changes with time using a diodeꢀarray detector. The data acquired after the second mixing were analyzed
using the standard software of the stoppedꢀflow instrument. In all cases a satisfactory fit was obtained for
the disappearance of 2 using a single exponential.
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 25
1
2
3
4
5
6
7
8
9
Analysis of the oxidation products. Once 2 was fully formed, 100ꢀL of a solution containing the correꢀ
sponding equivalents of the desired substrate were added in the cuvette. The decay of the band at λ = 490
nm was monitored and after complete decay the cuvette was let to attain room temperature. Biphenyl was
added as internal standard and the iron complex was removed by passing the solution through a short path
of silica. The products were then eluted with ethyl acetate and then analyzed by GC.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Catalytic experiments at room temperature with 1/AcOOH
In a typical experiment, 2 mL of a 1 mM solution of 1 in aceonitrile were placed in a 20 mL vial together
with the appropriate amount of the desired substrate. Then, the corresponding amount of peracetic acid
solution diluted in acetonitrile was added over 30 minutes. The solution was stirred for extra 15 minutes.
For alkane oxidation, an internal standard (biphenyl) was added after the reaction was completed. The
iron complex was removed by passing the solution through a short path of silica followed by elution with
ethyl acetate (2 mL). Finally, the solution was subjected to GC analysis. For the oxidation of cyclooctene,
the sample was treated with acetic anhydride (1 mL) together with 1ꢀmethylimidazole (0.1 mL), stirred
for 15 minutes at r.t. to esterify the alcohol products. Then 3 mL of ice were added and stirred for 10
minutes. Biphenyl was added as an internal standard at that point and then the organic products were exꢀ
tracted with CHCl3 (2 mL). The organic layer was extracted and washed with H2SO4 (2 mL, 1M), saturatꢀ
ed aq. NaHSO3 (2 mL) and water (2 mL). The organic layer was dried over MgSO4, filtered through a
short path of Celite® and subjected to GC or GCꢀMS analysis.
Supporting information
Experimental details for the preparation and characterization of ligand PyNMe3, and compounds 1, 2’ and
4; Xꢀray data for 1 and 4 (CIF); UV/vis stoppedꢀflow experiments for the generation and reactivity of 2;
details on EPR and Mössbauer experiments; carboxylate exchange experiments of 2 with acetic acidꢀd4.
This material is available free of charge via the Internet at http://pubs.acs.org.
Acknowledgements
Financial support for this work was provided by the European Commission (2011ꢀCIGꢀ303522 to A.C.
and ERCꢀ2009ꢀStGꢀ239910 to M.C.), the Spanish Ministry of Science (CTQ2012−37420ꢀC02ꢀ01/BQU
to M.C., CTQ2012ꢀ37821ꢀC02ꢀ02 to M.G.B., CTQ2013ꢀ43012ꢀP to A.C. and CSD2010ꢀ00065 to M.C.,
M.G.B. and E.G.E) and Generalitat de Catalunya (ICREA Academia Award to M.C.). The Spanish Minisꢀ
try of Science is acknowledged for a Juan de la Cierva contract to B.V. (JCIꢀ2011ꢀ09302) and for a Raꢀ
món y Cajal contract to A.C. (RYCꢀ2011ꢀ08683). We thank financial support from INNPLANTA project
INPꢀ2011ꢀ0059ꢀPCTꢀ420000ꢀACT1 to Dr. Xavi Ribas. We also thank Dr. Laura Gómez (Serveis Tècnics
de Recerca, Universitat de Girona) for helpful advice in setting up the HRꢀMS experiments and helpful
20
ACS Paragon Plus Environment

Page 21 of 25
Journal of the American Chemical Society
1
2
3
4
5
discussions. The work at the University of Minnesota and Carnegie Mellon University was respectively
supported by grants from the US Department of Energy, Office of Basic Energy Sciences (grant DEꢀ
FG02ꢀ03ER15455 to L.Q.) and the US National Science Foundation (CHEꢀ1305111 to E.M.).
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
References
(1)
(2)
(3)
Dyker, G. Handbook of C-H Transformations; WileyꢀVCH: Weinheim, 2005.
Meunier, B.; de Visser, S. P.; Shaik, S. Chem. Rev. 2004, 104, 3947ꢀ3980.
Montellano, P. R. O. d. Cytochrome P450: Structure, Mechanism, and Biochemistry;
Springer: New York, 2005.
(4)
(5)
(6)
(7)
(8)
Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L. Chem. Rev. 2004, 104, 939ꢀ986.
AbuꢀOmar, M. M.; Loaiza, A.; Hontzeas, N. Chem. Rev. 2005, 105, 2227ꢀ2252.
Kovaleva, E. G.; Lipscomb, J. D. Nat. Chem. Biol. 2008, 4, 186ꢀ193.
Barry, S. M.; Challis, G. L. ACS Catal. 2013, 3, 2362ꢀ2370.
Company, A.; LloretꢀFillol, J.; Costas, M. In Comprehensive Inorganic Chemistry II;
Reedijk, J., Poeppelmeier, K., Eds.; Elsevier: Oxford, 2013; Vol. 3, p 487ꢀ564.
(9)
Park, M. J.; Lee, J.; Suh, Y.; Kim, J.; Nam, W. J. Am. Chem. Soc. 2006, 128, 2630ꢀ2634.
(10)
Seo, M. S.; Kamachi, T.; Kouno, T.; Murata, K.; Park, M. J.; Yoshizawa, K.; Nam, W.
Angew. Chem.-Int. Edit. 2007, 46, 2291ꢀ2294.
(11)
Cho, J.; Jeon, S.; Wilson, S. A.; Liu, L. V.; Kang, E. A.; Braymer, J. J.; Lim, M. H.;
Hedman, B.; Hodgson, K. O.; Valentine, J. S.; Solomon, E. I.; Nam, W. Nature 2011, 478, 502ꢀ505.
(12)
Thibon, A.; Jollet, V.; Ribal, C.; SénéchalꢀDavid, K.; Billon, L.; Sorokin, A. B.; Banse, F.
Chem.-Eur. J. 2012, 18, 2715ꢀ2724.
(13)
52, 9229ꢀ9244.
Prat, I.; Company, A.; Corona, T.; Parella, T.; Ribas, X.; Costas, M. Inorg. Chem. 2013,
(14)
Britovsek, G. J. P.; England, J.; White, A. J. P. Inorg. Chem. 2005, 44, 8125ꢀ8134.
BPMEN = N,N´ꢀdimethylꢀN,N´ꢀbis(2ꢀpyridylmethyl)ethaneꢀ1,2ꢀdiamine; BPMCN =
(15)
N,N'ꢀdimethylꢀN,N'ꢀbis(2ꢀpyridylmethyl)ꢀtransꢀ1,2ꢀdiaminocyclohexane;
pyridylmethyl)amine; PDP = N,N´ꢀBis(2ꢀpyridylmethyl)ꢀ2,2´ꢀbipyrrolidine; PyTACN = 1,4ꢀDimethylꢀ7ꢀ
(2ꢀpyridylmethyl)ꢀ1,4,7ꢀtriazacyclononane.
TPA
=
trisꢀ2ꢀ
(16)
Chen, K.; Que, L. J. Am. Chem. Soc. 2001, 123, 6327ꢀ6337.
(17)
3, 1334ꢀ1341.
Wang, Y.; Janardanan, D.; Usharani, D.; Han, K.; Que, L., Jr.; Shaik, S. ACS Catal. 2013,
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 25
1
2
3
4
(18)
Prat, I.; Mathieson, J. S.; Güell, M.; Ribas, X.; Luis, J. M.; Cronin, L.; Costas, M. Nat.
Chem. 2011, 3, 788ꢀ793.
5
6
7
8
(19)
(20)
(21)
Oloo, W. N.; Que, L. Acc. Chem. Res. 2015, 48, 2612ꢀ2621.
Costas, M.; Chen, K.; Que, L. Coord. Chem. Rev. 2000, 200, 517ꢀ544.
9
Even though the BDECꢀH of CH3CN (93 kcal·molꢀ1) is lower than that of some of the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
hydrocarbon substrates used in this work (e.g. BDEC6H12 = 99.3 kcal·molꢀ1), acetonitrile oxidation does
not take place due to polar effects. See: B. P. Roberts Chem. Soc. Rev. 1999, 28, 25.
(22)
1421ꢀ1433.
Gomez, L.; Canta, M.; Font, D.; Prat, I.; Ribas, X.; Costas, M. J. Org. Chem. 2013, 78,
Canta, M.; Font, D.; Gomez, L.; Ribas, X.; Costas, M. Adv. Synth. Catal. 2014, 356, 818ꢀ
(23)
830.
(24)
(25)
(26)
Chen, M. S.; White, M. C. Science 2007, 318, 783ꢀ787.
Chen, M. S.; White, M. C. Science 2010, 327, 566ꢀ571.
Oloo, W. N.; Que, L. In Comprehensive Inorganic Chemistry II; Reedijk, J.,
Poeppelmeier, K., Eds.; Elsevier: Oxford, 2013; Vol. 6, p 763ꢀ778.
(27)
Liu, L. V.; Hong, S.; Cho, J.; Nam, W.; Solomon, E. I. J. Am. Chem. Soc. 2013, 135,
3286ꢀ3299.
(28)
Kaizer, J.; Klinker, E. J.; Oh, N. Y.; Rohde, J.ꢀU.; Song, W. J.; Stubna, A.; Kim, J.;
Münck, E.; Nam, W.; Que, L. J. Am. Chem. Soc. 2004, 126, 472ꢀ473.
(29)
Company, A.; Prat, I.; Frisch, J. R.; MasꢀBalleste, R.; Guell, M.; Juhasz, G.; Ribas, X.;
Munck, E.; Luis, J. M.; Que, L.; Costas, M. Chem.-Eur. J. 2011, 17, 1622ꢀ1634.
(30)
Kundu, S.; Thompson, J. V. K.; Shen, L. Q.; Mills, M. R.; Bominaar, E. L.; Ryabov, A.
D.; Collins, T. J. Chem.-Eur. J. 2015, 21, 1803ꢀ1810.
(31)
Ghosh, M.; Singh, K. K.; Panda, C.; Weitz, A.; Hendrich, M. P.; Collins, T. J.; Dhar, B.
B.; Gupta, S. S. J. Am. Chem. Soc. 2014, 136, 9524ꢀ9527.
(32)
Hong, S.; Wang, B.; Seo, M. S.; Lee, Y.ꢀM.; Kim, M. J.; Kim, H. R.; Ogura, T.; Garciaꢀ
Serres, R.; Clémancey, M.; Latour, J.ꢀM.; Nam, W. Angew. Chem.-Int. Edit. 2014, 53, 6388ꢀ6392.
(33)
Seo, M. S.; Kim, N. H.; Cho, K.ꢀB.; So, J. E.; Park, S. K.; Clemancey, M.; GarciaꢀSerres,
R.; Latour, J.ꢀM.; Shaik, S.; Nam, W. Chem. Sci. 2011, 2, 1039ꢀ1045.
(34)
Biswas, A. N.; Puri, M.; Meier, K. K.; Oloo, W. N.; Rohde, G. T.; Bominaar, E. L.;
Münck, E.; Que, L. J. Am. Chem. Soc. 2015, 137, 2428ꢀ2431.
(35)
Sastri, C. V.; Lee, J.; Oh, K.; Lee, Y. J.; Lee, J.; Jackson, T. A.; Ray, K.; Hirao, H.; Shin,
W.; Halfen, J. A.; Kim, J.; Que, L.; Shaik, S.; Nam, W. Proc. Natl. Acad. Sci. USA 2007, 104, 19181ꢀ
19186.
22
ACS Paragon Plus Environment

Page 23 of 25
Journal of the American Chemical Society
1
2
3
4
(36)
Company, A.; Gómez, L.; Güell, M.; Ribas, X.; Luis, J. M.; Que, L.; Costas, M. J. Am.
Chem. Soc. 2007, 129, 15766ꢀ15767.
5
6
7
(37)
3448ꢀ3452.
Hitomi, Y.; Arakawa, K.; Funabiki, T.; Kodera, M. Angew. Chem.-Int. Edit. 2012, 51,
8
9
(38)
Makhlynets, O. V.; Oloo, W. N.; Moroz, Y. S.; Belaya, I. G.; Palluccio, T. D.; Filatov, A.
S.; Muller, P.; Cranswick, M. A.; Que, L.; RybakꢀAkimova, E. V. Chem. Commun. 2014, 50, 645ꢀ648.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(39)
Oloo, W. N.; Fielding, A. J.; Que, L. J. Am. Chem. Soc. 2013, 135, 6438ꢀ6441.
(40)
Oloo, W. N.; Meier, K. K.; Wang, Y.; Shaik, S.; Munck, E.; Que, L. Nat. Commun. 2014,
5, 4046/4041ꢀ4046/4049.
(41)
(42)
Makhlynets, O. V.; RybakꢀAkimova, E. V. Chem.-Eur. J. 2010, 16, 13995ꢀ14006.
The EPR of 2 in MeCN shows broad signals and it is difficult to integrate. Thus, 2 was
generated in the glassy 1:3 MeCN:acetone solvent mixture to afford dramatically sharpened EPR signals.
(43)
Lyakin, O. Y.; Zima, A. M.; Samsonenko, D. G.; Bryliakov, K. P.; Talsi, E. P. ACS Catal.
2015, 5, 2702ꢀ2707.
(44)
(45)
(46)
McGarvey, B. R. Coord Chem. Rev. 1998, 170, 75ꢀ92.
Taylor, C. P. S. BBA Protein Struct. 1977, 491, 137ꢀ148.
Van Heuvelen, K. M.; Fiedler, A. T.; Shan, X.; De Hont , R. F.; Meier, K. K.; Bominaar,
E. L.; Münck, E.; Que, J. L. Proc. Natl. Acad. Sci. USA 2012, 109, 11933ꢀ11938.
(47)
For the extreme octahedral case, for which the tetragonal and rhombic distortions are
small compared to the spinꢀorbit coupling constant (λ ≈ 380 cmꢀ1), the gꢀvalues are close to g = 2.00; see
for instance Figure 3 in Telser, J. J. Braz. Chem., 17, 1501ꢀ1515 (2006) for A2 ≈ 1/3. This, perhaps
unlikely, case would be recognizable by Mössbauer spectroscopy, as the 57Fe magnetic hyperfine tensor
would differ considerably from that observed for the other region for which the gꢀvalues are close to g =
2, namely when the tetragonal distortion is large compared to λ.
(48)
23, 4743ꢀ4754.
Schulz, C. E.; Rutter, R.; Sage, J. T.; Debrunner, P. G.; Hager, L. P. Biochemistry 1984,
(49)
Rittle, J.; Green, M. T. Science 2010, 330, 933ꢀ937.
(50)
de Oliveira, F. T.; Chanda, A.; Banerjee, D.; Shan, X.; Mondal, S.; Que, L.; Bominaar, E.
L.; Münck, E.; Collins, T. J. Science 2007, 315, 835ꢀ838.
(51)
Roelfes, G.; Vrajmasu, V.; Chen, K.; Ho, R. Y. N.; Rohde, J.ꢀU.; Zondervan, C.; la Crois,
R. M.; Schudde, E. P.; Lutz, M.; Spek, A. L.; Hage, R.; Feringa, B. L.; Münck, E.; Que, L. Inorg. Chem.
2003, 42, 2639ꢀ2653.
(52)
Tse, C.ꢀW.; Chow, T. W.ꢀS.; Guo, Z.; Lee, H. K.; Huang, J.ꢀS.; Che, C.ꢀM. Angew. Chem.-
Int. Edit. 2014, 53, 798ꢀ803.
(53)
8040ꢀ8048.
McGown, A. J.; Kerber, W. D.; Fujii, H.; Goldberg, D. P. J. Am. Chem. Soc. 2009, 131,
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 25
1
2
3
4
(54)
Leeladee, P.; Jameson, G. N. L.; Siegler, M. A.; Kumar, D.; de Visser, S. P.; Goldberg, D.
P. Inorg. Chem. 2013, 52, 4668ꢀ4682.
5
6
7
8
(55)
(56)
MasꢀBalleste, R.; Fujita, M.; Que, L. Dalton Trans. 2008, 1828ꢀ1830.
Iyer, S. R.; Javadi, M. M.; Feng, Y.; Hyun, M. Y.; Oloo, W. N.; Kim, C.; Que, L. Chem.
Commun. 2014, 50, 13777ꢀ13780.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(57)
Furutachi, H.; Hashimoto, K.; Nagatomo, S.; Endo, T.; Fujinami, S.; Watanabe, Y.;
Kitagawa, T.; Suzuki, M. J. Am. Chem. Soc. 2005, 127, 4550ꢀ4551.
(58)
Zhang, X.; Furutachi, H.; Tojo, T.; Tsugawa, T.; Fujinami, S.; Sakurai, T.; Suzuki, M.
Chem. Lett. 2011, 40, 515ꢀ517.
(59)
(60)
Khusnutdinova, J. R.; Luo, J.; Rath, N. P.; Mirica, L. M. Inorg. Chem. 2013, 52, 3920ꢀ
3932.
Ghosh, P.; Tyeklar, Z.; Karlin, K. D.; Jacobson, R. R.; Zubieta, J. J. Am. Chem. Soc.
1987, 109, 6889ꢀ6891.
(61)
(62)
(63)
Sanyal, I.; Ghosh, P.; Karlin, K. D. Inorg. Chem. 1995, 34, 3050ꢀ3056.
Kitajima, N.; Fujisawa, K.; Moroꢀoka, Y. Inorg. Chem. 1990, 29, 357ꢀ358.
Nakazawa, J.; Terada, S.; Yamada, M.; Hikichi, S. J. Am. Chem. Soc. 2013, 135, 6010ꢀ
6013.
(64)
Wang, X.; Peter, S.; Kinne, M.; Hofrichter, M.; Groves, J. T. J. Am. Chem. Soc. 2012,
134, 12897ꢀ12900.
(65)
(66)
(67)
(68)
(69)
Mayer, J. M. Acc. Chem. Res. 1998, 3, 441ꢀ450.
Wolak, M.; van Eldik, R. Chem.-Eur. J. 2007, 13, 4873ꢀ4883.
Groves, J. T.; Gross, Z.; Stern, M. K. Inorg. Chem. 1994, 33, 5065ꢀ5072.
Kwon, E.; Cho, K.ꢀB.; Hong, S.; Nam, W. Chem. Commun. 2014, 50, 5572ꢀ5575.
Gómez, L.; GarciaꢀBosch, I.; Company, A.; BenetꢀBuchholz, J.; Polo, A.; Sala, X.; Ribas,
X.; Costas, M. Angew. Chem.-Int. Edit. 2009, 48, 5720ꢀ5723.
24
ACS Paragon Plus Environment

Page 25 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
Table of contents
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25
ACS Paragon Plus Environment