Reductive Electrochemical Activation of Molecular Oxygen Catalyzed by an Iron-Tungstate Oxide Capsule: Reactivity Studies Consistent with Compound i Type Oxidants

Article abstract of DOI:10.1021/acscatal.0c00897

The reductive activation of molecular oxygen catalyzed by iron-based enzymes toward its use as an oxygen donor is paradigmatic for oxygen transfer reactions in nature. Mechanistic studies on these enzymes and related biomimetic coordination compounds designed to form reactive intermediates, almost invariably using various "shunt" pathways, have shown that high-valent Fe(V)=O and the formally isoelectronic Fe(IV) =O porphyrin cation radical intermediates are often thought to be the active species in alkane and arene hydroxylation and alkene epoxidation reactions. Although this four decade long research effort has yielded a massive amount of spectroscopic data, reactivity studies, and a detailed, but still incomplete, mechanistic understanding, the actual reductive activation of molecular oxygen coupled with efficient catalytic transformations has rarely been experimentally studied. Recently, we found that a completely inorganic iron-tungsten oxide capsule with a keplerate structure, noted as {Fe₃₀W₇₂}, is an effective electrocatalyst for the cathodic activation of molecular oxygen in water leading to the oxidation of light alkanes and alkenes. The present report deals with extensive reactivity studies of these {Fe₃₀W₇₂} electrocatalytic reactions showing (1) arene hydroxylation including kinetic isotope effects and migration of the ipso substituent to the adjacent carbon atom ("NIH shift"); (2) a high kinetic isotope effect for alkyl C - H bond activation; (3) dealkylation of alkylamines and alkylsulfides; (4) desaturation reactions; (5) retention of stereochemistry in cis-alkene epoxidation; and (6) unusual regioselectivity in the oxidation of cyclic and acyclic ketones, alcohols, and carboxylic acids where reactivity is not correlated to the bond disassociation energy; the regioselectivity obtained is attributable to polar effects and/or entropic contributions. Collectively these results also support the conclusion that the active intermediate species formed in the catalytic cycle is consistent with a compound I type oxidant. The activity of {Fe₃₀W₇₂} in cathodic aerobic oxidation reactions shows it to be an inorganic functional analogue of iron-based monooxygenases.

Full text of DOI:10.1021/acscatal.0c00897

Subscriber access provided by EDINBURGH UNIVERSITY LIBRARY | @ http://www.lib.ed.ac.uk
Article
Reductive Electrochemical Activation of Molecular Oxygen
Catalyzed by an Iron-Tungstate Oxide Capsule: Reactivity
Studies Consistent with Compound I Type Oxidants
Marco Bugnola, Kaiji Shen, Eynat Haviv, and Ronny Neumann
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.0c00897 • Publication Date (Web): 11 Mar 2020
Downloaded from pubs.acs.org on March 12, 2020
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 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 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.
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 11
ACS Catalysis
1
2
3
4
5
6
Reductive Electrochemical Activation of Molecular Oxygen Catalyzed
by an Iron-Tungstate Oxide Capsule: Reactivity Studies Consistent with
Compound I Type Oxidants
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
‡
,†
‡
Marco Bugnola, Kaiji Shen, Eynat Haviv, and Ronny Neumann*
Department of Organic Chemistry, Weizmann Institute of Science, Rehovot, Israel, 76100
ABSTRACT: The reductive activation of molecular oxygen catalyzed by iron-based enzymes towards its use as an oxygen donor is paradig-
matic for oxygen transfer reactions in nature. Mechanistic studies on these enzymes and related biomimetic coordination compounds de-
signed to form reactive intermediates, almost invariably using various “shunt” pathways, have shown that high valent Fe(V)=O and the for-
mally isoelectronic Fe(IV)=O porphyrin cation radical intermediates are often thought to be the active species in alkane and arene hydroxyla-
tion and alkene epoxidation reactions. Although this four-decade long research effort has yielded a massive amount of spectroscopic data,
reactivity studies and detailed, but still incomplete, mechanistic understanding, the actual reductive activation of molecular oxygen coupled
with efficient catalytic transformations has rarely been experimentally studied. Recently, we found that a completely inorganic iron-tungsten
oxide capsule with a Keplerate structure, noted as {Fe30
W
72}, is an effective electrocatalyst for the cathodic activation of molecular oxygen in
water leading to the oxidation of light alkanes and alkenes. The present report deals with extensive reactivity studies of these {Fe30
W72} elec-
trocatalytic reactions showing (1) arene hydroxylation including kinetic isotope effects and migration of the ipso substituent to the adjacent
carbon atom (“NIH shift”); (2) a high kinetic isotope effect for alkyl C-H bond activation; (3) dealkylation of alkylamines and alkylsulfides;
(
4) desaturation reactions; (5) retention of stereochemistry in cis-alkene epoxidation and (6) unusual regioselectivity in the oxidation of
cyclic and acyclic ketones, alcohols and carboxylic acids where reactivity is not correlated to the bond disassociation energy; the regioselectivi-
ty obtained is attributable to polar effects and/or entropic contributions. Collectively these results also support the conclusion that the active
intermediate specie formed in the catalytic cycle in consistent with a Compound I type oxidant. The activity of {Fe30
oxidation reactions shows it to be an inorganic functional analog of iron based monoxygenases.
W72} in cathodic aerobic
INTRODUCTION Collectively, iron-based monooxygenase
enzymes have diverse structures and active site configurations, and
also different biological roles and therefore different target sub-
Despite the extensive research efforts described above there is a
very noticeable paucity of reports on the actual reductive activation
of O in the context of both heme and non-heme iron chemistry
2
1
strates. Despite this diversity in structure and biological function,
where a highly catalytically active intermediate is formed that re-
sults in alkane oxidation. In an early example, ascorbate was used as
they have many common mechanistic similarities. In fact, especially
for hydroxylation reactions, it can be generalized that molecular
reductant to activate O
2
over a diiron complex, but the reaction was
sub-stoichiometric. Cathodic electrocatalytic reactions present
intriguing possibilities for the activation of O . In this context, early
8
oxygen activation takes place via the bonding of O to a Fe(II) cen-
2
ter, followed by reductive formation of (hydro)peroxo iron inter-
mediates. In the case of heme proteins and related biomimetic
porphyrin-based models, lysis of the O-O yields a Fe(IV)=O por-
phyrin cation radical intermediate, also known as Compound I,
2
9
use of manganese(III) porphyrins showed low activity, and an
electrocatalytic alkane oxidation with cytochrome P-450 failed
allowing only the more facile sulfide oxidation. In the past few
years there have been 3 reports of cathodic activation of O towards
1
0
2
responsible for the observed reactivity profiles. For the non-heme
2
diiron soluble methane monooxygenase enzyme, a closed bis(µ-
C-H bond hydroxylation and alkene epoxidation based on iron-
catalysts. (1) A combination of Alkane Hydroxylase (AlkB) and
Rubredoxin-2 (AlkG) enzymes were successfully used for hydrox-
3
oxo)diiron(IV) active species has been the consensus, until recent-
ly where new evidence points to a more open bisFe(IV) structure
4
11
with both terminal and bridging oxo moieties. Monoiron non-
ylation of higher alkanes. (2) Picket-fence iron porphyrins with an
heme Fe(IV) oxo biomimetic species have also been extensively
studied and in some cases quite reactive, but their activity appears
axial thiol ligand were assembled on electrodes and used for alkane
12
hydroxylation and alkene epoxidation. (3) An inorganic
5
III
VI
to be low compared to enzyme catalysts. The first non-heme
{Fe 30
W
72} molecular capsule was used by us as a homogeneous
iron(V)-oxo species based on the tetraanionic macrocyclic
tetramide ligand (TAML) showed low activity relative to the isoe-
electrocatalyst in water for alkane oxygenation under ambient con-
ditions, notably ethane to acetic acid by a cascade of oxidation reac-
6
lectronic heme compound I. On the other hand, recent research
tions CH
3
CH
3
® CH
3
CH
2
OH ® CH
3
CHO® CH COOH. Oxi-
3
on Fe(V) oxo species suggest that they are more reactive than their
dation of alkenes was also observed, for example, CH =CH ®
2
2
7
analogous Fe(IV) oxo species, although the published reactivity
13
CH
2
(O)CH
2
® HOCH
2
CH
2
OH ® HCHO ® HCOOH. As
profiles for such intermediates is much more limited than for the
isoelectronic Fe(IV)=O porphyrin cation radical species.
determined from the X-ray diffraction derived structure,
III
VI
{
Fe 30
W
72} is icosahedral with 20 pores and is assembled by link-
ACS Paragon Plus Environment

ACS Catalysis
} oxide units with 30 Fe(III) atoms,
Page 2 of 11
ing 12 pentagonal {(W)W
5
species responsible for reactivity. Based on a typical iron-porphyrin
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 1. The inner core of the capsule contains water and ~25
sulfate/bisulfate anions, yielding overall a water soluble polyan-
catalytic cycle, a similar working hypothesis for a mono-iron cata-
lytic cycle is presented in Figure 2.
1
4
ion. Each Fe atom, 3 lining each pore, is hexa-coordinated to 4
bridging oxygen atoms in the equatorial plane (oxo, Fe-O = 1.9787
Thus, the present report deals with extensive reactivity studies of
e
the {Fe30
W
72}/O cathodic reactions. Reactions investigated in-
2
± 0.002 Å); these oxygen atoms are bound to distorted octahedral
neighboring W(VI) atoms. The coordination sphere around
Fe(III) is completed in the axial plane by terminal oxygen atoms on
the outer surface, Oa,o, and inner surface, Oa,i, of the capsule with
cluded (1) arene hydroxylation reactions encompassing kinetic
isotope effects and migration of the ipso substituent to the adjacent
carbon atom (“NIH shift”) or elimination of the ipso substituent
(dehalogenation); (2) oxidation of 1,2-ethanediol towards the
measurement of the kinetic isotope effect for alkyl C-H bonds (3)
oxidation dealkylation of alkylamines and to some degree of alkyl-
sulfides; (4) desaturation reactions; (5) cis-alkene epoxidation with
retention of stereochemistry and (6) oxidation of various substrates
with functional groups, where unusual regioselectivity was ob-
served where reactivity is not correlated to the bond disassociation
energy. The unusual regioselectivity observed is attributed to polar
effects (water solvent and polyanionic catalyst) and/or entropic
contributions. Collectively these results also support a conclusion
that a Compound I type active intermediate species is formed in
bond distances of Fe-O
coordination sphere around each Fe atom can be defined as Fe-
®W) (O with ~180 ° bond angles between opposite oxygen
a
= 2.041 Å and 2.045 Å respectively. The
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(O
e
4
)
a 2
atoms, Figure 1. The Fe-Fe distances are ~6.4 Å.
the catalytic cycle that explains the activity of {Fe30
aerobic oxidation reactions.
W72} in cathodic
RESULTS AND DISCUSSION
Some information on the reactivity of the reduced capsule with
III
VI
72
O
2
was available by UV-vis spectroscopy. Thus, after {Fe 30
W
}
II
III
VI
was reduced by 20 electrons, brown {Fe 20Fe 10
W
72} was then
exposed to O
2
in the absence of further reducing electrons. It rather
slowly transformed through two isosbestic points at 242 and 390
nm in about 50 min at 4 °C, Figure 3.
Figure 1. The iron-tungsten capsule, {FeIII30WVI72} from single crystal
14
XRD. Left—ball and stick model; Fe--green, W-black, O-red, S-
yellow. Not shown are the water molecules and NH cations that are
4
mostly on the periphery of the outer surface. Right – Coordination
around the Fe atoms.
Beyond the alkane and alkene oxidations noted above, the previ-
III
VI
ous research on electrocatalytic oxidation with {Fe 30
W
72} re-
III
VI
vealed by cyclic voltammetry that (a) the yellow {Fe 30
W
72
}
compound shows a fast, reversible, diffusion limited reduction of
Fe(III) to Fe(II) at 0.47 V vs. Ag/AgCl; (b) at 0 V vs. Ag/AgCl the
capsule IIIis reduced by 10 electrons, yielding
a
brown
II
VI
{Fe 10Fe 30
W 72} species, commensurate with one reduced iron
II
III
VI
atom, Fe(II), per pore. (c) Upon exposure of {Fe 10Fe 30
W 72} to
O the initial formation of an intermediate was deduced by EPR but
2
not directly observed; (d) no further intermediate species were
identified, but the formation of hydroxy radicals as reactive species
during catalysis was ruled out.
Figure 3. Difference UV-vis spectra showing the evolution of proposed
III
III
VI
II
III
VI
{
(Fe OOH)10Fe 20
W
72} from {Fe 20Fe 10
W
72} at 4°C in double
Sub-O
O
W
O
W
W
W
III
e–
distilled water. A 1.7 µM solution of {Fe 30WVI72} capsule was reduced
in situ by addition of 20 reducing equivalents of Na
tra were recorded by opening the cuvette to air.
O
W
O
O
IIIO
O
O
O
IIO
W
Fe
O
Fe
O
O
O
O
2
S
2
4
O and the spec-
FeIV
O
O
W
W
O
W
W
W
W
W
O
Sub
O2
The presence of the isosbestic points indicates that the reaction
stoichiometry is constant and no secondary reactions occurred
during this time period, but not necessarily is there a quantitative
H+
O
W
O
W
O
W
W
W
O
O
VO
–H O
2
O
O
O
Fe
O
O
O
e–, H+
O
FeIII
FeIII
15
O
O
conversion one unique species to another. One possible explana-
O
W
O
O
O
W
W
W
O
W
tion is that the transformation that is being observed is as shown in
equation 1:
W
O
OH
O
Figure 2. A possible catalytic cycle or hypothesis for cathodic activa-
tion of O and oxygenation. The electron distribution for the active
II
–
+
III
Fe + O
2
+ 2 e + H ® Fe –O–OH
(1)
2
Support for this explanation is that similar transformations were
observed for other iron substituted polyoxometalates where the
reduced end-on water stabilized hydroperoxo species were isolated
species is unknown. The presentation of Fe(V)=O versus Fe(IV)-O•+
is a formalism. Other reactive intermediates can also be considered.
III
VI
The high catalytic activity of {Fe 30
W 72} under cathodic aero-
bic conditions is an impetus to gain insight into the possible active
ACS Paragon Plus Environment

Page 3 of 11
ACS Catalysis
EPR spectrum associable to a combination of the two conformers
1
6
and identified by x-ray crystallography. An alternative explanation
is that the reaction being observed is as shown in equation 2:
2
0
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
of the BMPO/OOH• adduct, was obtained Figure 5. The manner
in which the BMPO/OOH• adduct is formed and then observed
under these conditions is typically indicative of the trapping of a
superoxide anion. In addition, the experiments exclude the for-
mation of “free” hydroxyl radicals and homolytic cleavage in the
proposed iron(III)-hydroperoxo intermediate species.
II
+
III
4
Fe + 4H + O
2
® 4 Fe + 2 H
2
O
(2)
III
where Fe –O–OH would only be an initial shorter lived inter-
mediate. Support for this explanation is that the spectrum obtained
after 50 min is practically identical to the spectrum of {Fe 30
III
VI
W 72}.
A further resonance Raman experiment was carried out where
III
VI
72
{
{
Fe 30
W
}
was reduced by 20 electrons to yield
72} either electrochemically, Figure 4 ,or by addition
II
III
VI
Fe 20Fe 10
W
of Na
2
S
2
O
4
, Figure S1 and then exposed to O . With both reduction
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
–1
methods the spectra showed the appearance of a peak at 893 cm
1
6
after addition of O
signed to
2
(air) Figure 4. This peak is tentatively as-
a
high spin iron(III)-hydroperoxo species,
III
III
VI –1 17
{
(Fe OOH)10Fe 20
W
72}, typically observed at > 850 cm . The
18
–1
expected isomer shift for an experiment with
which unfortunately coincides with strong M-O vibrational modes
of the capsule and therefore was not observed.
O
2
is ~50 cm ,
Figure 5. Formation of the BMPO/OOH• spin adduct. A 14 µM solu-
III
VI
tion of Na
x
{Fe 30
W
72} was reduced in situ at 4°C in double distilled
water with 10 % DMSO by addition of 10 reducing equivalents electro-
chemically; 10 mg BMPO were added and then the solution was
opened to air.
The further spectroscopic investigation of this catalytic cycle af-
ter the proposed formation of the hydroperoxo species was stymied
by (a) the aqueous reaction medium that made low temperature
capture of reactive intermediates difficult, (b) the strong absorp-
5
7
tion by tungsten of Co gamma rays that led to a very noisy and
5
7
uninformative Fe Mössbauer spectrum even of the pristine start-
III
VI
ing compound {Fe 30
W 72}, (c) the strong and broad Raman
III VI
peaks of W-O and Fe-O in {Fe 30
W 72}, Figure 4, that prevented
Figure 4. Normalized Raman spectra showing the formation of the
proposed {(Fe OOH)10Fe 20
µM solution of Na
water in situ by addition of 20 reducing equivalents electrochemically
III
III
VI
II
III
VI
observation of new peaks that may be associable to additional Fe-
based intermediates and (d) the magnetic exchange interactions
between Fe atoms that prevented useful EPR measurements.
W
72} from {Fe 20Fe 10
W 72}. A 14
{Fe 30WVI72} was reduced at 4°C in double distilled
III
x
III
III
VI
(
blue). The spectrum of {(Fe OOH)10Fe 20
W
72} was recorded by
In lieu of the lack of spectroscopic data on isolatable reactive in-
termediates, reactivity studies with different types of substrates
could be excellent indicators of a possible active intermediate that
would explain the overall reactivity observed. The cathodic oxida-
18
opening the cuvette to air (red) or
O
2
(green).
III
VI
The Raman spectrum of {Fe 30
upon exposure to O
iron(III)-superoxo species, {(Fe OO•)10Fe 20
W
72} reduced by 10 electrons
2
showed no evidence for the formation of an
III
VI
III III VI
tion of ethane to acetic acid was efficient (1.9 µmol {Fe 30
W
72},
W
72}. That is no
–
1
2.5 mL H O, 1 bar air, 2 bar ethane, 20 °C, 24 h, platinum mesh
2
peak expected at ~1100 cm was observed. Further along the reac-
tion pathway, the suggested high iron(III)-hydroperoxo species,
cathode, platinum wire anode, cell potential 1.8 V) the reaction of
methane was two-orders of magnitude slower or less efficient. We,
therefore, initially focused on the challenging oxygenation of ben-
zene since the hydroxylation of arenes is also paradigmatic for
compound 1. Benzene reacted to initially form phenol, which fur-
ther reacted to yield hydroquinone and catechol, Scheme 1.
III
III
VI
{
(Fe OOH)10Fe 20W 72} could undergo heterolytic bond cleav-
age to yield an Fe(V)-oxo species as hypothesized in Figure 2. Al-
ternatively, homolytic bond cleavage would yield a hydroxyl radi-
cal. The formation of “free” hydroxyl radical can in principle be
identified by trapping it with a nitrone spin trap such as BMPO (5-
tert-butoxycarbonyl 5-methyl-1-pyrroline N-oxide) that would
yield a BMPO/OH• spin adduct with its associated EPR spectrum.
Indeed, such a spectrum was initially observed, Figure S2. Howev-
er, this experiment has an important caveat in that BMPO/OH•
spin adducts are typically also observed as a result of the decay of
originally formed BMPO/OOH• spin adduct that may also be
metal catalyzed. In order to differentiate between this decay pro-
cess and “free” OH• radicals directly reacting with BMPO, DMSO
should be added to the solution as a scavenger (reacts faster than
Scheme 1. Cathodic Aerobic Hydroxylation of Benzene
OH
OH
OH
OH
+
HO
7
.6
5.5
2.5
Reaction conditions: 1.9 μmol {FeIII30WVI72}, 2.5 mL D
2
O, air 1 bar, 100
µmol benzene, 20 °C, 24 h, platinum mesh cathode, platinum wire anode, po-
tential 1.8 V. Faradaic efficiency 24%. Products are reported in μmol and were
1
characterized and quantified by H NMR (see Figure S3). There was no reac-
III
VI
tion without {Fe 30
W
72} excluding the possibility of anodic activation of H O
2
BMPO) of free OH• radicals. The formation of BMPO/CH • spin
adducts is expected when free OH• radicals are present.
was not observed. Instead, under these experimental conditions an
3
1
8-20
at the Pt anode.
This
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 11
Since a strong oxidant was surmised, Figure 2, the absence of The results for the oxidation of other halobenzenes summarized
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
quinones in the reaction mixture was initially puzzling. It was, how-
ever observed that benzoquinone is easily reduced to hydroqui-
none under the reaction conditions used. This along with the
prevalence of the uncoupled oxidation reaction (reduction of the
reactive Fe=O species before its reaction with the substrate) can
explain the Faradaic efficiency observed.
Scheme 2. Cathodic Aerobic Oxidation of 1,4-Difluorobenzene
in Table 1. The most notable observations are that (1) the overall
reactivity was BrPh > ClPh > FPh which is proportional to Ham-
mett s values and commensurate with a positively charged transi-
+
tion state in the rate determining step and an electrophilic oxidant.
This is what would be expected from the mechanism presented in
Scheme 2 and the results of the kinetic isotope effect experiment
shown below. In addition, as may be also expected, nitrobenzene
was less reactive (2.6 µmol products) and yielded mostly 3-
nitrophenol (96%) and some hydroquinone (4 %). (2) The reac-
tion at the ipso position was predominant for fluorobenzene but
less so for bromobenzene and chlorobenzene leading to the per-
haps counterintuitive observation that defluorination is more facile
2
1
F
F
F
F
F
F
–
Fe(III)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
Fe(V)=O
H
+
HO
F
O
F
1.5%
O
Fe(III)
H , e–
OH
OH
OH
despite the stronger C-F bond (C
6
H
5
–F = 125.6 Kcal/mol, C
6
H –
5
F
+ F–
OH
25
Fe(V)=O
Cl = 95.5 Kcal/mol; C –Br = 80.4 Kcal/mol); C-halogen bond
6
H
5
HO
cleavage is not rate determining. It should be noted that the prefer-
ence for defluorination over dichlorination has been observed in
the past, for example in the cytochrome P-450 catalyzed oxidation
78.2%
20.3%
F
Reaction conditions: 1.9 μmol {FeIII30WVI72}, 2.5 mL D
O, air 1 bar, 100
2
µmol 1,4 difluorobenzene, 20 °C, 24 h, platinum mesh cathode, platinum wire
anode, potential 1.8 V. Faradaic efficiency 5.3%. A total of 2.94 μmol products
were formed, which were characterized and quantified versus dimethylsulfone
23a
of 1,3,5-trichloro-2,4,6-trifluorobenzene. Subsequent DFT calcu-
lations point towards the probability that the initial p attack of the
1
19
arene ring before formation of the sigma-bond intermediate is rate
as standard by H NMR and F-NMR(see Figures S4-S7).
26
determining and has a lower barrier for C-F versus C-Cl.
The hydroxylation of benzene also presented the opportunity to
study the mechanism of this arene hydroxylation via investigation
of the “NIH shift” reaction. A typical mechanism proposes the for-
mation of a s-complex followed hydride migration and formation
of a ketone. Keto to enol tautomerization, thermodynamically driv-
en by the formation of the arene, leads to products with partial
The mechanism presented in Scheme 2, implies that the rate de-
termining step of the reaction does not involve C-H bond cleavage,
therefore, there should be no primary kinetic isotope effect, KIE,
for the hydroxylation of arenes. Since the benzene has limited solu-
bility in water and is also quite volatile, the KIE was measured for
phenol, using ~1:1 C
6
H
5
OH/C
6
5
D OH in a competition experi-
2
2
retention of the migrated hydrogen atom. Loss of the ipso hydro-
gen has also been observed and an analogous dehalogenation is a
major pathway for haloarene oxidation that is catalyzed by bifacial
ment. KIE ~1.0 was measured (see Figures S11-S13 for the data)
supporting a mechanism that does not involve C-H bond cleavage
in the rate determining step. Another experiment that was carried
out was to test for the possible exchange of the oxo moiety of the
2
3-24
iron phthalocyanine complexes, presumably because halides are
better leaving groups compared to hydride. We used 1,4-
difluorbenzene as a substrate, where product formation through the
2
7
proposed Fe(V)=O active species with water. This was done by
1
8
carrying out a cathodic aerobic oxidation of reaction using O
2
and
1
shift reaction and dehalogenation can be easily followed by H and
1
6
III
VI
19
H
2
O. Thus, 1.9 µmol {Fe 30
W
72} in 2.5 mL DDW (natural
abundance H O), was deaerated by 5 pump/thaw cycles using
99.999% N and then 30 mL 96.2% O was added to the sealed
F NMR, Scheme 2. The results clearly show the predominance of
1
6
2
the dehalogenation reaction, but also the NIH shift reaction was
observed, both hallmarks of an iron-oxo reactive intermediate. The
prevalence of attack at the ipso position is very noticeable even in
the presence of an activating hydroxyl moiety as evidenced by the
final product distribution via the initial formation of 4-fluorophenol
as intermediate.
18
2
2
electrochemical cell. Electrolysis was carried out at 20 °C for 2 h
using a platinum mesh cathode and a platinum wire anode at a
potential of 1.8 V. Phenol was analyzed by GC-MS and showed
1
8
16
16
7
9% Ph OH and 21% Ph OH. Considering that O is continu-
2
ously formed at the anode, the experiment shows that the reactive
Table 1. Cathodic Aerobic of Halobenzenes
18
oxidant remained mostly O labeled. This allows one to conclude
1
8
that the proposed Fe(V)= O active species reacts faster with the
Substrate
Total yield, µmol Products, mol %
benzene or is reduced in an uncoupled transformation (not shown)
1
6
16
fluorobenzene
4.7
2-fluorohydroquinone 13
compared to an exchange reaction with H O to yield Fe(V)= O,
2
hydroquinone/catechol 64
formic acid 23
Scheme 3. Note that the absence of a phenyl radical, even if it could
exist, would require a C-H bond activation in the rate determining
chlorobenzene
bromobenzene
5.9
2-chlorohydroquinone 46
step which was not observed. Thus, in this type of arene hydroxyla-
18
4-chlorophenol 5
tion reaction one can rule the reaction of such a radical with O
2
to
18
hydroquinone/catechol 20
formic acid 23
2-bromohydroquinone 36
yield Ph OH.
Scheme 3. Cathodic Hydroxylation of Benzene with
18
O
2
7.5
FeIII
4- bromohydrocatechol 8
18O
+
–
hydroquinone/catechol 40
formic acid 16
2 (H , e )
OH
2
H216O
slow
FeV
FeV
O
O
Reaction conditions: 1.9 μmol {FeIII30WVI72}, 2.5 mL D
00 µmol substrate, 20 °C, 24 h, platinum mesh cathode, platinum wire
anode, potential 1.8 V. Products were characterized and quantified by
O, air 1 bar,
2
fast
Reaction conditions: 1.9 μmol {FeIII30WVI72}, 2.5 mL H
O, 1.3 mmol 96.2%
1
2
18O
, 100 µmol benzene, 20 °C, 2 h, platinum mesh cathode, platinum wire
2
1
19
H NMR and F-NMR (see Figures S8-S10).
anode, potential 1.8 V.
ACS Paragon Plus Environment

Page 5 of 11
ACS Catalysis
Alkene epoxidation is a common reaction catalyzed also by vari-
droxylation regioselectivity will be presented below for reactions
involving C-H bond activation by HAT of functionalized alkanes
including those substituted with electron donating hydroxyl
groups, that is alcohols.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ous biomimetic models of heme and non-heme iron compounds.
{Fe 30
epoxide formed, for example ethene oxide from ethene quickly
reacts via C-C bond cleavage to form the corresponding formalde-
hyde and then formic acid. It was shown that cytochrome P-450
III
VI
W 72} also catalyzes the oxidation of alkenes, but the initial
Scheme 5. Cathodic Aerobic Oxidation of Thioanisole
2
8
epoxidized alkenes by stereoselective cis-addition, and as ex-
pected the more reactive the oxidant, e.g. Mn(V)-oxo versus
Mn(IV)-oxo porphyrins, the higher retention of cis-geometry. A
water soluble cis-stilbene was chosen as a suitable substrate. Thus,
the oxidation of 75 µmol cis-4,4'-(ethene-1,2-diyl)dibenzoic acid
basified with 1.8 eq of LiOH was carried out in the presence of 1.9
{Fe30W }
72
PhS(O)Me + PhS(O) Me + Ph-Ph + PhSSPh
2
PhSMe
94.1%
2.2%
2.3%
1.3%
2
9
Reaction conditions: 1.3 μmol {FeIII30WVI72}, 2.0 mL H
O, 1.3 mmol, air 1
2
bar, 100 µmol thioanisloe, 20 °C, 24 h, platinum mesh cathode, platinum wire
anode, potential 1.8 V, total conversion of thioanisole - 9.2 mol %.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The oxidation of thioanisole was also examined in order to com-
pare S-dealkylation versus S-oxidation, Scheme 5. Here very clearly
the S-oxidation is the preferred reaction with only formation of a
small amount of S-dealkylation products.
III
VI
µmol {Fe 30
W
72} in 3 mL D O. Electrolysis was at 20 °C for 15
2
min using a platinum mesh cathode and a platinum wire anode at a
potential of 1.8 V. H NMR spectra from 0.5 mL aliquots, after
precipitation of {Fe 30
H NMR showed a peak associable to the cis-epoxide at 4.24 ppm,
Figure S14, but no peak at <4.0 ppm that could be attributed to the
trans-epoxide.
1
III
VI
W 72} with CsCl, after 1, 5 and 15 min, the
1
As shown previously the aerobic reductive electrocatalytic oxy-
gen of light alkanes in water was rather extraordinary with the oxi-
dation of ethane to acetic acid being the preferred reaction upon
30,31
After 15 min the further C-C cleavage reaction
1
3
consideration of both activity and selectivity. In order to gain
additional data, the KIE of the oxidation of ethylene glycol was
studied. The observable products of this reaction were glycolalde-
hyde and formic acid, Figure S16. The reaction was first order in
ethylene glycol and the KIE = 12.2, Figures S17-S19. Thus, the KIE
supports a hydrogen abstraction-oxygen rebound mechanism for
hydroxylation of aliphatic C-H bonds and is different than the KIE
for phenol. As proposed the arene and alkane hydroxylation occur
through different pathways.
yielded 4-carboxybenzaldehyde as the predominant product.
N-dealkylation and to a much lesser extent, S-dealkylation reac-
tions are also well established for high valent Compound I. The
2
competing transformation is oxygen transfer to the heteroatom,
which is much more common for sulfides. The oxidation of tri-
III
VI
ethylamine catalyzed by {Fe 30
W 72} reacted only via a N-
dealkylation pathway, Scheme 4.
Scheme 4. Cathodic Aerobic Oxidation of Triethylamine
Carbon-hydrogen bond activation and oxygenation was studied
for various additional functionalized compounds. In the case of
cyclic ketones, Scheme 6, one can observe that for cyclopentanone
the major product (>98%) is cyclopent-2-en-1-one. On the other
hand, for cyclohexanone and cycloheptanone, no significant oxy-
genation at the carbon atom adjacent to the ketone was observed,
even though the bond dissociation energy of the a-C-H bond to
(
a) (Et)3N
(Et)2NCH(OH)CH3
(Et)2NCH2CH2OH
(CH3CH2)2NH + CH3CHO
3
CH COOH
6.1 µmol
(
b)(Et)3N
(CH3CH2)2NH + 2 HCHO
HCOOH
1.6 µmol
20.6 µmol
26.2 µmol
Reaction conditions: 1.9 μmol {FeIII30WVI72}, 2.5 mL D
O, air 1 bar, 50 µmol
2
triethylamine, 20 °C, 24 h, platinum mesh cathode, platinum wire anode, poten-
tial 1.8 V. Faradaic efficiency 5.3%. A total of 2.94 μmol products were formed,
which were characterized and quantified versus dimethylsulfone as standard by
33
1
the ketone is significantly weaker. The very high selectivity ob-
H NMR, Figure S15.
served for the formation of 1,4-cyclohexanedione from cyclohexa-
none is especially compelling.
Such N-dealkylation reactions are thought to take place via for-
mation of a a-hydroxylation intermediate that decomposes to yield
Scheme 6. Cathodic Aerobic Oxidation of Cyclic Ketones
3
2
diethylamine and acetaldehyde, Scheme 4a. The mechanism by
which the a-hydroxylation intermediate is formed is under debate,
with evidence for pathways involving either hydrogen atom transfer
(HAT) or formation of a nitrogen centered radical cation by elec-
tron transfer. Either way in the reaction presented in Scheme 4a,
one indeed could expect the formation of acetaldehyde as the initial
product that then would be oxidized quickly to acetic acid. Howev-
er, in the past acetaldehyde was always an observable by-product in
the oxidation of ethane. Now we also observed separately that the
oxidation of trifluoroethanol under the conditions described in
Scheme 4 yielded trifluoroacetaldehyde/trifluoroacetic acid in a
16.4 mol % conv.
O
O
O
TON 31.5
TOF 780 min–1
>98%
O
O
O
21.8 mol % conv.
TON 41.9
TOF 1280 min–1
O
~97%
1.5%
1.5%
O
O
O
O
18.2 mol % conv.
TON 35.0
TOF 1200 min–1
4
2%
O
O
58%
O, air 1 bar, 250
>
5:1 ratio after 24 h, bringing up the question - Why is acetalde-
Reaction conditions: 1.3 μmol {FeIII30WVI72}, 2.0 mL H
2
hyde absent in the analysis of the triethylamine oxidation reaction?
In addition, what is not explicable for such a reaction sequence and
not observed in the oxidation of acetaldehyde, is the formation of
significant amounts of formaldehyde. Note also in this context the
good mass balance between the amount of diethylamine and other
products formed. This raises the possibility of a competing or dom-
µmol substrate, 20 °C, 48 h, platinum mesh cathode, platinum wire anode,
1
potential 1.8 V. Conversions were determined by H NMR. Turnover number
III
VI
(TON) is mol products/mol {Fe 30
W
72}. Turnover frequency (TOF) was
III
VI
34
determined as an estimate of the coverage of {Fe 30
Quantification of the products and their identification was by H NMR and
NMR; see Figures S20-S25.
W 72}on the anode.
1 13
C
The oxidation of the analogous cyclic alcohols might have been
expected to yield the corresponding ketones, considering the BDE
inant HAT initiated reaction at the b-position, to form a b-
hydroxylation intermediate, Scheme 4b. Such an intermediate
could yield formaldehyde, acetic acid and diethylamine by any
number of conceivable pathways. Support for such unusual hy-
of the a-C-H bond in a linear primary alcohol is about 5 kcal/mol
35
weaker than the corresponding b-C-H bond, and the observation
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 11
that the reactivity of cyclohexanol with a tert-butylperoxy radical
favors the reaction at the a-C-H bond versus b-C-H bond by three
orders of magnitude. Despite this expectation, very surprisingly,
dized exclusively at the b-C-H bond yielding the 3-oxoadipate Fig-
ure S36.
It is typically observed that aliphatic hydrocarbons react at the C-
H bonds with the lowest BDE and therefore according to a
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
6
cyclic alcohols were oxidized at the a-C-H bond only with a 50-60
37,38
%
selectivity. In addition, oxidation occurred in significant amounts
Brønsted-Evans-Polanyi correlation.
Although the examples
(40-50%) also at b-C-H bond to yield very selectively the a, w-
shown above in Schemes 4, 6-8 show unusual regioselectivity, there
are some examples where it was observed that oxidation of (func-
tionalized) aliphatic hydrocarbons are in contrast to reactivity cor-
related to BDE’s and the Brønsted-Evans-Polanyi formalism. In
general for aliphatic C-H bond activation polar effects, stereo-
electronic effects, solvents and strain can influence regioselectivi-
dialdehyde as shown in Scheme 7.
Scheme 7. Cathodic Aerobic Oxidation of Cyclic Alcohols
1
4.3 mol % conv.
OH
O
O
O
O
TON 25.1
TOF 850 min
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
–1
59%
39%
41%
3
9
OH 9.6 mol % conv.
O
O
ty. This is the case also for metalloporphyrin catalyzed oxyhalo-
genation reactions where weak a-C-H bonds adjacent to electron
withdrawing moieties such as ketones and esters were shown to be
O
O
TON 17.8
TOF 570 min–1
HO
50%
8.7%
2.3%
O
O
OH
O
O
O
40
deactivated. Furthermore, using a non-heme iron catalyst, oxida-
12.9 mol % conv.
TON 22.0
tion reactions at distal methylene positions was observed where
both electronic, that is electron withdrawing groups, and steric
effects related to the structure of the substrate and those emanating
from the catalyst structure have been utilized to attain non-
TOF 790 min–
1
51%
41%
3.6%
O
O
4.4%
Reaction conditions: 1.3 μmol {FeIII30WVI72}, 2.0 mL H
2
O, air 1 bar, 250
µmol substrate, 20 °C, 48 h, platinum mesh cathode, platinum wire anode,
1
41
potential 1.8 V. Conversions were determined by H NMR. Turnover number
conventional chemo/regioselectivity.
III
VI
(
TON) is mol products/mol {Fe 30
W
72}. Turnover frequency (TOF) was
III
VI
Since {Fe 30
W 72} has a spherical capsular structure the site at
III
VI
32
determined as an estimate of the coverage of {Fe 30
Quantification of the products and their identification was by H NMR and
NMR; see Figures S26-S31
W 72}on the anode.
1 13
which catalysis occurs is important in terms of steric and confine-
ment effects. Thus, one can consider three different environments
for the active oxidizing intermediate. (1) on the inner surface of the
C
Scheme 8. Oxidation of Primary Linear Alcohols
III
VI
{
Fe 30
W
72} capsule where the active oxo unit is in the axial posi-
tion, Oa,i; (2) on the outer surface where the active oxo unit is in the
axial position, Oa,o; or (3) in the pore where the active oxo unit is in
1
3.0 mol % conv.
OH
O
OH
TON 30.0
TOF 790 min
OH
13%
–1
32%
O
55%
O
the equatorial position, O . In the past there have been measure-
e
ments on analogous capsules with slightly larger pores, {Mo132},
that demonstrated by NMR measurements the encapsulation of
alkanes and arenes when the inner surface was lined with acetate or
propionate ligands as opposed to sulfate in the presently used
O
O
OH
2
2%
OH 14.3 mol % conv.
9
%
O
TON 32.2
OH
TOF 880 min–1
OH
O 30%
2
4%
III
VI
42,43
{
Fe 30
W
72} capsule. The likelihood of an active oxo unit being
Reaction conditions: 1.3 μmol {FeIII30WVI72}, 2.0 mL H
O, air 1 bar, 300
2
at the Oa,i position is low considering the fast reaction of bulky sub-
strates such as cis-4,4'-(ethene-1,2-diyl)dibenzoate. To verify this
point, {Mo132} with sulfate ligands was prepared as an analog of
µmol substrate, 20 °C, 48 h, platinum mesh cathode, platinum wire anode,
1
potential 1.8 V. Conversions were determined by H NMR. Turnover number
III
VI
(TON) is mol products/mol {Fe 30
W
72}. Turnover frequency (TOF) was
III
VI
32
III
VI
determined as an estimate of the coverage of {Fe 30
Quantification of the products and their identification was by H NMR and
W 72}on the anode.
{
Fe 30
W
72} since the presence of paramagnetic Fe(III) prevented
1
1
3
C
44
direct NMR measurments. Encapsulation studies even with small
substrates such as ethane, using also low concentration sensitive
chemical exchange saturation transfer H NMR, showed no indi-
cation that ethane was encapsulated. This finding suggests that the
NMR; see Figures S32-S35. Some products (≤15%) were not identified.
1
45
The observation that the a, w-dialdehydes did not react within a
relatively long reaction time to yield observable quantities of car-
boxylic acids was also quite unexpected and in line with the large
amounts of formaldehyde remaining in the oxidation triethylamine,
Scheme 4 and the slow formation of trifluoroacetic acid from tri-
fluoroacetaldehyde. In this context, the oxidation of 1,4-butanediol
O
a,i position is not the active site.
Another aspect to consider is that the initial catalyst,
III
VI
{Fe 30
W
72}, is coordinatively saturated around the iron(III) cen-
ter. In the past we have computationally shown for the Keggin type
yielded only 4-hydroxybutanal that in D O appeared as a mixture of
2
H
5
PV Mo10O40 polyoxometalate that water and the surrounding
2
the acyclic product and the intramolecular hemiacetal, lactol.
Therefore, it is conceivable that the formation of hydrates and
hemiacetals stabilizes aldehydes towards further oxidation in these
high valent metal centers can stabilize coordinatively unsaturated
IV
species through the breaking of bridging (equatorial) V -O
bonds. In that case the terminal V=O is significantly shorter than
4
6
III
VI
{
Fe 30
W
72} catalyzed reactions. On the other hand, the reaction
the bridging V-O bond. In this case, however, the axial Fe-O
is slightly longer (2.041 Å) than the bridging Fe-O bond (1.979 Å)
although the former is too short to be considered a labile aqua lig-
and. At this point the binding site for O activation is still undeter-
a
bond
of 1,4-butanediol only at the a-C-H bond was in contrast to what
was observed with the cyclic alcohols. Thus, the reactivity of linear
primary alcohols was also surveyed, Scheme 8. As may be observed
e
2
reactions occur at the CH
2
OH moiety and at all the secondary
mined but our tentative proposal is activation at the equatorial
position, also from statistical considerations (4 positions versus 1
position), Figure 2.
methylene groups, but not at the primary methyl groups yielding a
non-selective mixture of products. Based on the reactivity profiles
that have been observed it can be noted that certain substrates
could be very selectively oxidized. For example, adipic acid, solubil-
ized in water by the addition of one equivalent of LiOH was oxi-
CONCLUSIONS
Oxidation of alkanes, alkenes, arenes and functionalized aliphatic
and aromatic compounds catalyzed by iron-heme and iron non-
heme biomimetic compounds have been extensively investigated
ACS Paragon Plus Environment

Page 7 of 11
ACS Catalysis
2
,3
13
for about 40 years. Very significant mechanistic insight has been
gained during this period. Synthetic applications are less developed,
previously postulated to explain the ethane to acetic acid oxida-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tion via formation of ethanol and acetaldehyde. (6) Dealkylation of
triethyl amine and to some extent of thioanisole also supports
compound I type species as active species.
A further survey of the oxidation of water-soluble substrates with
aliphatic C-H bonds revealed reactivity that was not commensurate
with the Brønsted-Evans-Polanyi formalism where reactivity is a
function of the bond disassociation energy (BDE). Such regioselec-
tivity has been observed in the past with electron-withdrawing sub-
stituents such as ketones and carboxylic acids. The regioselectivity
attained has been attributed in the past to polar effects. Consider-
especially as regards the reductive activation of O and probably
2
also because of the inherent instability of the organic ligands used.
At the basis of this research project were the following questions:
(1) Could an inherently stable inorganic metal oxide containing
iron function to activate O under electrochemical reducing condi-
2
tions? (2) If this is the case, is the reaction mechanism related to
heme and non-heme-iron oxidation chemistry? Phenomenological-
ly, we had already previously observed that an iron-tungsten cap-
III
VI
sule, {Fe 30
W
72}, was indeed able to cathodically activate O lead-
ing to interesting transformations such as the oxidation of ethane to
acetic acid at ambient conditions, low potentials and in water.
This paper reports on some mechanistic details concerning the
catalytic cycle of this reaction. A generic catalytic cycle for O acti-
vation, Figure 2, involves reduction of Fe(III) to Fe(II) which was
shown to occur at 0.67 V versus NHE, for one Fe atom in each of
the 20 identical pores of {Fe 30
Fe(II) center would yield by the Weiss model the superoxide,
Fe(III)-OO•–, as the first plausible intermediate in the catalytic
cycle. We were unable to observe such an intermediate by Raman
spectroscopy (expected n -1100-1200 cm ), but the formation of
such an intermediate was deduced using a BMPO spin trap. Since
electrons are often delocalized in polyoxometalates (rates of intra-
molecular electron transfer are ∼10 to 10 s for polyoxotung-
states at room temperature), the further intramolecular reduction
of a superoxo species to a hydroperoxo intermediate is reasonably
fast. In fact, reduction of {Fe 30
reduced species, that is two Fe(II) atoms per pore of {Fe 30
yielded an observable species by Raman spectroscopy that was
assigned as a hydroperoxo species, Fe(III)-OOH upon addition of
. Such hydroperoxo species have been shown to be stabilized by
hydrogen bonding to water as determined by X-ray diffraction for a
different iron containing polyoxometalates. Here a Raman spec-
trum supported such a scenario - the formation of a hydroperoxo
intermediate, nominally {(Fe OOH)10Fe 20
isotope labeling experiments with O
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
III
VI
ing the anionic outer surface of the {Fe 30
W 72} capsule, where the
1
3
reaction likely occurs according to the NMR measurements, and
the water solvent the explanation of polar effects on the reactivity
seems reasonable. More uniquely here, primary and cyclic alcohols,
and likely amines also showed unusual regioselectivity and did not
react according to the Brønsted-Evans-Polanyi formalism. Especial-
ly notable here are the C-C bond cleavage reaction for cyclic alco-
hols and triethylamine and the unexpected low reactivity of alde-
hydes with weak aldehydic C-H bonds. On may speculate that this
may be due to substrate protonation or binding to a metal. Finally,
it should be noted that entropic effects may also play an important
role in the selectivity that has been observed. In the absence of an
2
III
VI
13
W
72}. Addition of O
2
to such a
47
–1
‡
‡
isolatable intermediate, the activation energies, DH and DS , can-
not be directly evaluated. It is however worth noting that the reac-
tion of an isolated Mn(V)=O polyoxometalate-based species with
1
1
12 -1
4
8
1
-octene to yield 1-octene oxide was under entropic control with a
III
VI
W
72} to form a twenty electron
‡
–1
‡
–1 –1 49
DH < 4 kcal mol and DS < –50 cal mol K . It is reasonable in
this context that the higher charge of the transition state due to
charge transfer from the substrate to the anionic polyoxometalate
during the rebound reaction versus the initial polyoxometalate
leads to increased solvation of transition state by the water solvent
molecules. This may lead to the electrostriction of the solvent, a
loss of degrees of freedom, thereby leading to a high negative en-
tropy of activation. Such electrostriction of the solvent may be es-
III
VI
72
W
}
O
2
1
5
III
III
VI
W
72}, although the
were foiled by interfering
W 72} capsule, rendering the experiment less
III
VI
pecially significant in or at the pores of {Fe 30
W 72}.
1
8
2
III
VI
In summary, we presented data and explanations for the unique
peaks of the {Fe 30
III
VI
reactivity of {Fe 30
W
72} for the cathodic activation of O and
2
conclusive.
subsequent oxygenation. The formation of a Fe(III)-OOH species
is followed by the formation of a compound I type species active
species that shows reactivity commensurate with many reactivity
studies carried out in the past. Unique regioselectivity obtained in
the oxidation of several substrate types is atypical and explained by
polar and/or entropic effects. Finally, although reactivity is clearly
commensurate with compound I type active intermediate, the spec-
tral identification of this species remains elusive. The simplest ex-
planation as presented in Figure 2 is that the active species is an
Fe(V)=O intermediate. Another reasonable possibility is the for-
mation of a bis Fe(IV)=O species by fast intramolecular electron
transfer within a pore of the capsule with a bridging O-W-O unit as
shown in Scheme 9.
As noted above, further Mössbauer, Raman and EPR spectro-
scopic measurements to identify the active oxidizing moiety were
not feasible because of the strong absorption of g-rays by W, over-
lapping vibrational modes of the capsule and magnetic exchange
interactions. In lieu of spectroscopic studies, extensive reactivity
studies were carried out on various substrate types whose activity,
especially towards Fe(IV)=O-porphyrin•+ (compound I) and
isoelectronic Fe(V)=O species has been thoroughly investigated.
(1) Oxidation of 1,4-difluorobenzene showed both a NIH shift
reaction and dehalogenation reactions quite unique for compound
I. (2) A further KIE experiment on phenol (KIE = ~ 1) also sup-
ported an arene oxidation mechanism where an addition reaction
of a Fe(V)=O species at the arene ring is rate determining. (3)
Scheme 9. Representation of the Formation of a bis Fe(IV)=O
Species from an Initial Fe(V)=O Species.
1
8
18
Oxidation of benzene using O
2
showed high incorporation of O
and (4) epoxidation of cis-4,4'-(ethene-1,2-diyl)dibenzoate
showed conservation of stereochemistry. Both experiments suggest
that the active species formed is a very potent oxidant so that ex-
change with 55 M water is slower than oxygen transfer to benzene
and a planar intermediate in the epoxidation reaction is not formed
before epoxide ring formation. (5) A KIE experiment carried out
on ethylene glycol (KIE = 12.2) is supportive of an oxygen rebound
mechanism associated with a compound I type species that was
O
O
O
O
Fev
FeIII
FeIV
W
FeIV
W
O
O
O
O
EXPERIMENTAL SECTION
III
VI
Synthesis of the {Fe 30
W
72
} capsule. The synthesis was
performed according to the reported procedure and yielded the
III
VI
14
ammonium salt, (NH
4
)
x
{Fe 30
W
72} with a typical yield of 10-15%.
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 11
For most of the experiments including the reactivity studies
Temperature was controlled by a Ventecon low temperature cell. Thus,
{Fe 30WVI72
III
}
was used. For Raman and CEST NMR
of
72} was dissolved in 20 mL of 18.2 MW water and the
200 mg (Na)
{Fe 30
III
W
VI
72} was dissolved in 2.5 mL DDW and placed
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(NH
4
)
x
x
experiments that required higher solubility 1.5
g
in an undivided electrolysis cell. The solution was purged 10 times with
III
VI
III
VI
(NH
4
x
) {Fe 30
W
Ar and placed in a glove box. (Na)
x
{Fe 30
W
72} was reduced by 20
solution was transferred to a cellulose dialysis membrane tube with a
molecular weight cut-off of 12000. The cellulose dialysis membrane
electrons per capsule at -0.6 V versus Ag/AgCl (4 mA*h). The solution
was sampled into a 1 cm quartz cuvette, cooled in the Raman
1
8
spectrometer to 4 °C and opened to air or
Raman spectrum.
2
O for measurement of the
tube was immersed in a solution of 14 g Na
water for 15 h. The cellulose dialysis membrane tube then immersed in
0 mL of 8.2 MW water for 1 h to remove excess sulfate (6 times). This
2
SO
4
in 60 mL of 8.2 MW
EPR spectroscopy. EPR spectra were measured at RT by a X-band
CW EPR using a Bruker ELEXSYS 500 spectrometer equipped with a
Bruker ER4102ST resonator at 20 K with microwave power of 20 mW,
0.1 mT modulation amplitude and 100 kHz modulation frequency.
5
entire procedure was repeated twice and after that the solution inside
III
VI
the dialysis tube was evaporated to dryness to give (Na)
x
{Fe 30
W
72}.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Electrochemical Oxidation Reactions. Typically, the required
III
VI
III
VI
Thus, 50 mg (NH
4
x
) {Fe 30
W
72} was dissolved in 2.5 mL DDW and
amounts of (NH
4
)
x
{Fe 30
W
72} were dissolved in of D
2
O or H O was
2
placed in an undivided electrolysis cell. The solution was purged 10
times with Ar and placed in a glove box. (NH
placed in an undivided electrolysis cell. The required amount of
substrate (100-300 µmol) of substrate was added while stirring. The
electrocatalysis was performed for the given time period by applying a
DE= -1.8 V between cathode and anode. The working electrode
(cathode) was an 80 mesh platinum gauze and the counter electrode
III
VI
4
)
x
{Fe 30
W
72} was
reduced by 20 electrons per capsule at -0.6 V versus Ag/AgCl (4
mA*h). BMPO (10 mg) was added and the solution was sampled into
a flat cell while opened to air.
Gas chromatography. GC-MS measurements were carried out on
an Agilent 6890 gas chromatograph equipped with a 5973 mass
selective detector. A HP-5MS column (30ꢀmꢀ×ꢀ0.32ꢀmm i.d.; 0.25ꢀμm
film thickness) was used with He as carrier gas. GC-FID measurements
were also carried out on Agilent 6890 gas chromatograph. A HP-5MS
column (30ꢀmꢀ×ꢀ0.32ꢀmm i.d.; 0.25ꢀμm film thickness) was used with
He as carrier gas.
(
anode) a platinum wire. Reactions carried out in D O were solution
2
19
1
13
were typically analyzed by H NMR, C NMR, and F-NMR (11.7 T
Bruker Advance III HD Spectrometer) as required after precipitating
III
VI
(NH
4
)
x
{Fe 30
W
72} by addition of CsCl. In these reactions, products
1
were quantified by H NMR after addition of dimethylsulfone as
standard. Product quantification was performed by recording 100
scans at the appropriate time intervals. For reactions carried out in
H O, similar procedures were used, but analysis was carried out after
extraction of substrates and products into CDCl
directly assessed from the
efficiences were 20-35%.
2
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS
Publications website.
Raman and NMR spectra, and kientic profiles (PDF)
3
. Product yields were
1
H
NMR spectra. Typical Faradaic
Kinetic Isotope Effect - Phenol. A solution of (NH
120 mg in 6 mL of D O) was placed in an undivided electrolysis cell.
3.8 μmol of a mixture phenol/ deuterated phenol (0.99: 1) was
4
)
x
{FeIII30WVI72
}
(
1
2
AUTHOR INFORMATION
Corresponding Author
added. The electrocatalytic oxidation was performed for 3 hours
applying DE= -1.8 V between cathode and anode. The working
electrode (cathode) was a 80 mesh platinum gauze and the counter
electrode (anode) a platinum wire. A sample was taken every 30 min.
*
Email: ronny.neumann@weizmann.ac.il
1
The solution was analyzed by H-NMR after precipitating the capsule
ORCID
with CsCl and adding dimethylsulfone as standard. The quantification
was performed recording 100 scans with the appropriate time intervals.
The solutions were then analyzed by GC-MS, after extracting the
products with ethyl acetate.
Ronny Neumann: 0000-0002-5530-1287
Author Contributions
The manuscript was written through contributions of all authors. All
authors have given approval to the final version of the manuscript. ‡
These authors contributed equally.
Kinetic Isotope Effect
-
Ethylene Glycol
. A solution of
III
VI
(
NH
4
x
) {Fe 30
W
72} (120 mg in 6 mL of 10%D O + 90%H2O) was
2
placed in an undivided electrolysis cell. 22.2 μmol of a mixture ethylene
glycol/ deuterated ethylene glycol (1: 1) was added. The
electrocatalytic oxidation was performed for 5 hours applying DE = -
Notes
The authors note no competing financial interest
1
.8V between cathode and anode. The working electrode (cathode)
was a 80 mesh platinum gauze and the counter electrode (anode) a
platinum wire. A sample was taken and analyzed every hour, for 5 h.
ACKNOWLEDGMENT
This research was supported by the Israel Science Foundation grant
1
2
The solutions were analyzed by H-NMR and H-NMR, after
precipitating the capsule with CsCl. A standardized solution containing
1
237/18 and the Divadol Foundation. John T. Groves (Princeton
University) is thanked for the attempt to measure Mössbauer spectra.
R.N. is the Rebecca and Israel Sieff Professor of Organic Chemistry.
dimethylsulfone or methanol-d was added to quantify the substrates.
4
1
The quantification was performed recording 100 scans for H-NMR
2
and 250 scans for H-NMR at the appropriate time intervals.
REFERENCES
UV-vis spectroscopy. UV/Vis spectra at 4°C were recorded using a
diode array spectrometer Agilent 8453 and a Unisoku Cryostat for
1. Wang, V. C.-C.; Maji, S.; Chen, P. P.-Y.; Lee, H. K.; Yu, S. S.-F.;
Chan, S. I. Alkane oxidation: Methane monooxygenase, related
enzymes, and their biomimetics. Chem. Rev. 2017, 117, 8574-8621.
2. Huang, X.; Groves, J. T. Oxygen activation and radical
transformations in heme proteins and metalloporphyrins. Chem. Rev.
2018, 118, 2491-2553.
3. (a) Janiewski, A. J.; Que, L. Dioxygen activation by nonheme
diiron enzymes: Diverse dioxygen adducts, high valent intermediates,
and related model compounds. Chem. Rev. 2018, 118, 2454-2592; (b)
temperature control. Thus,
3
mL of 1.7 µM solution of
III
VI
(
NH
4
)
x
{Fe 30
W
72} in a 1 cm quartz cuvette was purged 10 times with
Ar and placed in the UV-vis spectrometer at 4 °C for 10 min. Sodium
dithionite (2.4 mg, 20 electron equivalents) was added yielding the
II
III
VI
reduced dark brown capsule, nominally {Fe 20Fe 10
W 72}. The
cuvette was opened to air and spectra were recorded every 30 s.
Raman spectroscopy. Raman spectra were measured on a Thermo
Scientific DRX2 spectrometer using a 455 nm solid state laser.
ACS Paragon Plus Environment

Page 9 of 11
ACS Catalysis
11. Tsai, Y.-F.; Luo, W.-I.; Chang, J.-L.; Chang, C.-W.; Chuang, H.-
Hong, S.; Lee, Y. M.; Ray, K.; Nam, W. Dioxygen activation chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
by synthetic mononuclear nonheme iron, copper and chromium
complexes. Coord. Chem. Rev. 2017, 334, 25-42; Engelmann, X.;
Monte-Perez, I.; Ray, K. Oxidation reactions with bioinspired
mononuclear non-heme metal-oxo complexes. Angew. Chem. Int. Ed.
C.; Ramu, R.; Wei, G.-T.; Zen, J.-M.; Yu, S. S.-F. Electrochemical
hydroxylation of C3–C12 n-alkanes by recombinant alkane
hydroxylase (AlkB) and rubredoxin-2 (AlkG) from pseudomonas
putida GPo1. Sci. Rep. 2017, 7, 8369.
12. Mukherjee, M.; Dey, A. Electron transfer control of reductase
versus monooxygenase: Catalytic C−H bond hydroxylation and alkene
epoxidation by molecular oxygen. ACS Cent. Sci. 2019, 5, 671−682.
13. Bugnola, M.; Carmieli, R.; Neumann, R. Aerobic
2
016, 55, 7632-7649.
. Cutsail, G. E.; Banerjee, R.; Zhou, A.; Que, L.; Lipscomb, J. D.;
4
DeBeer, S. High-resolution extended X-ray absorption fine structure
analysis provides evidence for a longer Fe···Fe distance in the Q
intermediate of methane monooxygenase. J. Am. Chem. Soc. 2018
,
electrochemical oxygenation of light hydrocarbons. ACS Catal. 2018
8, 3232-3236.
,
1
40, 16807-16820.
5. (a) Puri, M.; Que, L. Toward the synthesis of more reactive S = 2
non-heme oxoiron(IV) complexes. Acc. Chem. Res. 2015, 48,
443−2452; (b) Nam, W. Synthetic mononuclear nonheme iron–
oxygen intermediates. Acc. Chem. Res. 2015, 48, 2415−2423; (c) Seo,
M. S.; Kim, N. H.; Cho, K.-B.; So, J. E.; Park, S. K.; Clemancey, M.;
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14. Todea, A. M.; Merca, A.; Bögge, H.; Glaser, T.; Pigga, J. M.;
Langston, M. L.; Liu, K. T.; Prozorov, R.; Luban, M.; Schröder, C.;
III
VI
2
Casey, W. H.; Müller, A. Porous capsules {(M)M }12Fe 30 (M=Mo ,
5
VI
W ): Sphere surface supramolecular chemistry with 20 ammonium
ions, related solution properties, and tuning of magnetic exchange
interactions. Angew. Chem. Int. Ed. 2010, 49, 514–519.
́
Garcia-Serres, R.; Latour, J.-M.; Shaik, S.; Nam, W. A mononuclear
nonheme iron(iv)-oxo complex which is more reactive than
cytochrome P450 model compound I. Chem. Sci. 2011, 2, 1039−1045;
rd
15. Braslavsky, S. E. Glossary of terms used in photochemistry 3
edition. Pure Appl. Chem. 2007, 79, 293-465.
(
d) Biswas, A. N.; Puri, M.; Meier, K. K.; Oloo, W. N.; Rohde, G. T.;
16. Barats, D.; Leitus, G.; Popovitz-Biro, R.; Shimon, L. J. W.;
Neumann, R. A stable “end-on” iron(III)-peroxo complex in water
Bominaar, E. L.; Munck, E.; Que, L. Modeling TauD-J: A high-spin
̈
derived from
molecular oxygen. Angew. Chem. Int. Ed. 2008, 47, 9908–9912.
7. Chen, K.; Que, L. Stereospecific alkane hydroxylation by non-
a multi iron(II) substituted polyoxometalate and
nonheme oxoiron(IV) complex with high reactivity toward C–H
bonds. J. Am. Chem. Soc. 2015, 137, 2428−2431; (e) McDonald, A.
R.; Que, L. High-valent nonheme iron-oxo complexes: Synthesis,
structure, and spectroscopy. Coord. Chem. Rev. 2013, 257, 414– 428;
1
heme iron catalysts: Mechanistic evidence for an Fe=O active species.
J. Am. Chem. Soc. 2001, 123, 6327-63339.
6. (a) de Oliveira, F. T.; Chanda, A.; Banerjee, D.; Shan, X.; Mondal,
S.; Que, L.; Bominaar, E. L.; Münck, E.; Collins, T. J. Chemical and
18. Zhao, H.; Joseph, J.; Zhang, H.; Karoui, H.; Kayanaraman, B.
Synthesis and biochemical applications of a solid cyclic nitrone spin
trap: a relatively superior trap for detecting superoxide anions and
glutathiyl radicals. Free Radical Biol. Med. 2001, 31, 599-606.
19. Keszler, A.; Kayanaraman, B.; Hogg, N. Comparative
investigation of superoxide trapping by cyclic nitrone spin traps: the
use of singular value decomposition and multiple linear regression
analysis. Free Radical Biol. Med. 2003, 35, 1149-1157.
V
Spectroscopic Evidence for an Fe -Oxo Complex. Science 2007, 315,
835– 838; (b) Kwon, E.; Cho, K.-B.; Hong, S.; Nam, W. Mechanistic
insight into the hydroxylation of alkanes by a nonheme iron(V)–oxo
complex. Chem. Commun. 2014, 50, 5572– 5575; (c) Ghosh, M.;
Singh, K. K.; Panda, C.; Weitz, A.; Hendrich, M. P.; Collins, T. J.; Dhar,
V
B. B.; Sen Gupta, S. Formation of a Room Temperature Stable Fe (O)
Complex: Reactivity Toward Unactivated C–H Bonds. J. Am. Chem.
Soc. 2014, 136, 9524– 9527; (d) Collins, T. J.; Ryabov, A. D. Targeting
of High-Valent Iron-TAML Activators at Hydrocarbons and Beyond.
Chem. Rev. 2017, 117, 9140– 9162
20.
https://www.bruker.com/fileadmin/user_upload/8-PDF-
Docs/MagneticResonance/EPR_brochures/superoxide.pdf
21. Holtmann, D.; Hollmann, F. The oxygen dilemma: A severe
challenge for the application of monooxygenases? ChemBioChem,
2016, 17, 1391-1398.
7. (a) Dantignana, V.; Serrano-Plana, J.; Draksharapu, A.; Magallón,
C.; Banerjee, S.; Fan, R.; Gamba, I. Guo, Y. Que, L. Jr.; Costas, M.;
Company, A. Spectroscopic and Reactivity Comparisons between
Nonheme Oxoiron(IV) and Oxoiron(V) Species Bearing the Same
Ancillary Ligand. J. Am. Chem. Soc. 2019, 141, 15078-15091; (b) Fan,
R.; Serrano-Plana, J.; Oloo, W. N.; Draksharapu, A.; Delgado-Pinar, E.;
Company, A.; Martin-Diaconescu, V.; Borrell, M.; Lloret-Fillol, J.;
García-España, E.; Guo, Y.; Bominaar, E. L.; Que, L.; Costas, M.;
Münck, E. Spectroscopic and DFT characterization of a highly reactive
22. Sono, M.; Roach, M. P.; Coulter, E. D.; Dawson, J. H. Heme
containing oxygenases. Chem. Rev. 1996, 96, 2841-2888.
23. (a) Rietjens, I. M. C. M.; Vervoort, J. A new hypthoses for the
mechanism of cytochrome-P450 dependent aerobic conversion of
hexahalogenated benzenes to pentaholgenated phenols. Chem. Res.
Toxicol. 1992, 5, 10–19. (b)Koerts, J.; Soffers, A. E. M. F.; Vervoort, J.;
De Jager, A.; Rietjens, I. M. C. M. Occurrence of the NIH Shift upon
the Cytochrome P450-Catalyzed in Vivo and in Vitro Aromatic Ring
Hydroxylation of Fluorobenzenes. Chem. Res. Toxicol. 1998, 11, 503–
512.
24. (a) Afanasiev, P.; Sorokin, A. B. μ-Nitrido Diiron Macrocyclic
Platform: Particular Structure for Particular Catalysis. Acc. Chem. Res.
2016, 49, 583–593; (b) Colomban, C.; Kudrik, E. V.; Afanasiev, P.;
Sorokin, A. B. Catalytic Defluorination of Perfluorinated Aromatics
under Oxidative Conditions Using N-Bridged Diiron Phthalocyanine.
J. Am. Chem. Soc. 2014, 136, 11321–11330.
25. Luo, Y.-R. Comprehensive Handbook of Chemical Bond
Energies, CRC Press, 2007, pp 1688.
26. Hackett, J. C.; Sanan, T. T.; Hadad, C. M. Oxidative
dehgaolgenation of perhalogenated benzenes by cycchrome P-450
compound. Biochemistry, 2007, 46, 5924-5940.
27. Seo, M. S.; In, J. H.; Kim, S. O.; Oh, N. Y.; Hong, J.; Kim, J.; Que,
L.; Nam, W. Direct evidence for oxygen-atom exchange between
nonheme oxoiron(IV) complexes and isotopocally lableled water.
Angew. Chem. Int. Ed. 2004, 43, 2417-2420.
V
nonheme Fe −oxo intermediate. J. Am. Chem. Soc. 2018, 140,
3916−3928; (c) Serrano-Plana, J.; Oloo, W. N.; Acosta-Rueda, L.;
Meier, K. K.; Verdejo, B.; García-España, E.; Basallote, M. G.; Münck,
E.; Que, L.; Company, A.; Costas, M. Trapping a Highly Reactive
Nonheme Iron Intermediate That Oxygenates Strong C-H Bonds with
Stereoretention. J. Am. Chem. Soc. 2015, 137, 15833– 15842.
8. Menage, S.; Galey, J.-B.; Hussler, G.; Seit, M.; Fontecave, M.
Aromatic Hydroxylation by H
Diiron(III) Complex. Angew. Chem. Int. Ed. 1996, 35, 2353-2355.
. (a) Creager, S. E.; Raybuck, S. A.; Murray, R. W. An efficient
2
O
2
and O
2
Catalyzed by a μ-Oxo
9
electrocatalytic model cytochrome P-450 epoxidation cycle. J. Am.
Chem. Soc. 1986, 108, 4225-4227; (b) Horwitz, C. P.; Creager, S. E.;
Murray, R. W. Electrocatalytic olefin epoxidation using manganese
Schiff-base complexes and dioxygen. Inorg. Chem. 1990, 29, 1006–
1
011.
1
0. Udit, A. K.; Hill, M. G.; Gray, H. B. Electrochemical generation
of a high-valent state of cytochrome P450. J. Inorg. Biochem. 2006
100, 519–523.
,
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 11
2
8. Ortiz de Montellano, P. R.; Mangold, B. L. K.; Wheeler, C.; 37. Panov, G. I.; Parfenov, M. V.; Parmon, V. N. The Brønsted-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Kunze, K. L.; Reich, N. O. Stereochemistry of cytochrome P-450-
catalyzed epoxidation and prosthetic heme alkylation. J. Biol. Chem.
Evans-Polanyi correlations in oxidation catalysis. Catal. Rev.: Sci. Eng.
2015, 57, 436-477.
38. Salamone, M.; Bietti, M. Tuning reactivity and selectivity in
hydrogen atom transfer from aliphatic C-H bonds to alkoxyl radicals:
Role of structural and medium effects. Acc. Chem. Res. 2015, 48,
2895-2903.
1
983, 258, 4208-4213.
9. Groves, J. T.; Stern, M. K. Olefin epoxidation by manganese(IV)
porphyrins: Evidence for two reaction pathways. J. Am. Chem. Soc.
987, 109, 3812-3814.
0. Mark, V. Nucleophilic reactions of trivalent phosphorus
2
1
3
39. Liu, W.; Groves, J. T. Manganese catalyzed C-H halogenation.
Acc. Chem. Res. 2015, 48, 1727-1735.
compounds: A new synthesis of epoxides. J. Am. Chem. Soc. 1963, 85,
1
884-1885.
40. Chen, M. S.; White, M. C.; Combined effects on selectivity in
Fe-catalyzed methylene oxidation. Science, 2010, 327, 566-571.
41. Kopilevich, S.; Gottlieb, H.; Keinan Adamsky, K.; Müller, A.;
Weinstock, I. A. The uptake and assembly of alkanes within a porous
nanocapsule in water: New information about hydrophobic
confinement. Angew. Chem. Int. Ed. 2016, 55, 4476-4481.
42. Sarma, B. B.; Avram-Biton, L.; Neumann, R. Encapsulation of
arenes within a porous molybdenum oxide nanocapsule. Chem.–Eur. J.
2016, 22, 15231-15236.
31. Stereoselective cis epoxidation would also be observed if the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
hydroperoxo species, Fe(III)-OOH were the reactive intermediate and
this possibility cannot be excluded. Since the experiments show fast C-
C bond cleavage of this epoxide, such a reaction not generally being
associated with such hydroperoxides, we prefer to assign just one
reactive intermediate to explain reactivty at this time.
31. Shaik, S.; Cohen, S.; Wang, Y.; Chen, H.; Kumar, D.; Theil, W.
P=450 enzymes: Their structure, reactivity and selectivity modeled by
QM/MM calculations. Chem. Rev. 2010, 110, 949-1017.
43. Müller, A.; Krickemeyer, E.; Bögge, H.; Schmidtmann, M.;
Botar, B.; Talismanova, M. O. Drawing small cations into highly
charged porous nanocontainers reveals “water” assembly and related
interaction problems. Angew. Chem. Int. Ed. 2003, 42, 2085 – 2090.
44. Lokesh, N.; Seegerer, A.; Hioe, J.; Gschwind, R. Chemical
exchange saturation transfer in chemical reactions: A mechanistic tool
for NMR detection and characterization of transient intermediates. J.
Am. Chem. Soc. 2018, 140, 1855-1862.
45. Efremenko, I.; Neumann, R. Computational insight into the
initial steps of the Mars-van Krevelen mechanism: Electron transfer
and surface defects in the reduction of polyoxometalates. J. Am. Chem.
Soc. 2012, 134, 20669–20680.
32. Hudzik, J. M.; Bozzelli, J. W. Thermochemistry and bond
dissociation energies of ketones. J. Phys. Chem. A, 2012, 116, 5707-
5722.5
33. Since such electrochemcial reactions occur at the electrotrode
surface, the turnover frequency (TOF) was calculated as follows: TOF
=
molproducts/(molcat on electrode * time) where molproducts is the total of all
2
14
2
products. The surface of the Pt cathode was 2.5 cm = 2.5 *10 nm ;
III
VI
the diameter of {Fe 30
W
72} was taken as 3.0 nm which can be
2
translated to a coverage of 7.065 nm per molecule. At a somewhat
14
arbitray 50% coverage molcat on electrode
.77*10 molecules = 2.94*10 mol.
= 2.5*10 *0.5/7.065 =
13
-11
1
34. Oyeyemi, V. B.; Keith, J. A.; Carter, E. A. Trends in bond
46. Weiss, J. J. Nature of the iron-oxygen bond in oxyhaemoglobin,
Nature, 1964, 202, 83-84.
47. Weinstock, I. A. Homogeneous-phase electron-transfer reactions
of polyoxometalates. Chem. Rev. 1998, 98, 113-170.
48. Khenkin, A. M.; Kumar, D.; Shaik, S.; Neumann, R.
Characterization of Manganese(V) Oxo Polyoxometalate
Intermediates and their Properties in Oxygen Transfer Reactions. J.
Am. Chem. Soc. 2006, 128, 15451-15460.
dissociation energies of alcohols and aldehydes computed with
multireference averaged coupled-pair functional theory. J. Phys. Chem.
A, 2014, 118, 3039-3050.
35. Puchkov, S. V.; Buneeva, E. I.; Perkel’, A. L. Reactivity of
cyclohexanol CH bonds toward the tert-butylperoxy radical. Kinet.
Catal. 2002, 43, 813-820.
3
6. Mayer, J. M. Understanding hydrogen atom transfer: From bond
strengths to Marcus theory. Acc. Chem. Res. 2011, 44, 36-46.
ACS Paragon Plus Environment

Page 11 of 11
ACS Catalysis
F
OH
OH
F
OH
F
1
2
3
4
5
6
7
8
9
OH
F
F
OH
Arene hydroxylation, dehalogenation, NIH shift
Compound I Equivalent
O
O
OH
O
O
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Aliphatic C-H hydroxylation, polar/entropic effects
ACS Paragon Plus Environment
1
1