Hangman effect on hydrogen peroxide dismutation by Fe(iii) corroles

Article abstract of DOI:10.1039/c2cc30580a

Hangman Fe(iii) corroles catalyse H₂O₂ disproportionation at a faster rate and display a more pronounced hangman effect than their one electron oxidized analogues owing to their ability to bypass high energy intermediates by redox-leveling derived from the use of the corrole as a non-innocent ligand.

Full text of DOI:10.1039/c2cc30580a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 4175–4177
www.rsc.org/chemcomm
COMMUNICATION
Hangman effect on hydrogen peroxide dismutation by Fe(III) corrolesw
Daniel J. Graham, Dilek K. Dogutan, Matthias Schwalbe and Daniel G. Nocera*
Received 26th January 2012, Accepted 7th March 2012
DOI: 10.1039/c2cc30580a
Hangman Fe(III) corroles catalyse H2O2 disproportionation at a
faster rate and display a more pronounced hangman effect than
their one electron oxidized analogues owing to their ability to
bypass high energy intermediates by redox-leveling derived from
the use of the corrole as a non-innocent ligand.
as compared to its [(HCX)FeIIICl–corrꢀ+] and non-hangman
congeners.
The systematic tuning of the steric and electronic properties
of corrole metal complexes may be achieved through macro-
cycle meso-substituent variation.21–24 Given that the electron
rich and redox active corrole p-system is prone to reaction
with O2 resulting in macrocycle decomposition, corroles bearing
electron-withdrawing pentafluorophenyl meso-substituents
were chosen for this study, as they display superior turnover
numbers (TONs) for O2-producing reactions relative to corroles
with electron-releasing meso-substituents.17 Beginning with the
free base hangman corrole,6 the previously reported method of
refluxing in DMF or pyridine with excess FeCl2, followed by
purification with diethyl ether, delivered the pyridine or diethyl
ether adducts 2-py and 3-Et2O.25 Isolation of 3 as the diethyl
ether adduct was desired so as to avoid any deleterious acid–base
chemistry of the hanging carboxylic acid with pyridine; indeed,
the product may be obtained in high yield (96.0%). 2-py was
isolated in high yield (94.6%) as the pyridine adduct owing to its
enhanced resistance to oxidation and ease of handling. With
[FeIII–corr] in hand, their one-electron oxidized congeners are easily
obtained by the methods of Gross et al.25 Washing dichloro-
The strategic placement of an acidic or basic functionality in the
secondary coordination sphere of transition metal complexes is
known to facilitate catalysis and stabilize energetic intermediates
during O–O bond formation or cleavage.1–3 With the installation
of acid–base groups on functionalized xanthene moieties featuring
porphyrin, corrole, or salen ligands,4,5 the so-called ‘‘hangman
effect’’ has proven beneficial in the catalytic reduction of O2,6
production of H2,7 oxidation of H2O,8 epoxidation of olefins,9
and disproportionation of H2O2.10–13 In every case, the hangman
effect manifests itself in a lowered overpotential and/or significant
increase in catalytic rate.
Metallo-corroles can promote a variety of transformations,
including atom efficient aziridination and cyclopropanations.14
Accounts of [FeIII–corr] catalyzed decomposition of reactive
nitrogen and oxygen containing species in biology and the
pertinence of such a reaction chemistry to disease14–16 motivated
us to investigate the role of the hangman effect in corrole H2O2
dismutation activity; but quizzically, the presence of the hanging
group decreased both the lifetime and activity of the catalyst.17
Mahammed and Gross had previously shown that Fe(III) corrole
([FeIII–corr]) complexes are active H2O2 dismutation catalysts.18
In light of these insights, we tentatively ascribed the deteriora-
tion in performance of the hangman system to the oxidation
state of the pre-catalyst,17 which is most accurately described
as an FeIIICl centre complexed by a corrole radical cation,
[FeIIICl–corrꢀ+].19,20 In this case, H2O2 dismutation would
necessarily pass through a [FeV–corrꢀ+] intermediate, which is
thermodynamically challenging to access and prone to decom-
position, thus obscuring any hangman effect. If this conten-
tion is correct, a pronounced hangman effect should be
observed for H2O2 dismutation from a hangman corrole
resting state residing one redox level lower, i.e., [FeIII–corr]. We
now report the superior rates of catalytic H2O2 disproportiona-
tion by xanthene modified hangman corrole [(HCX)FeIII–corr]
methane solutions of 1-py, 2-py, and 3-Et2O (Scheme 1) with
+
dilute HCl in H2O delivered the corresponding [FeIIICl–corrꢀ
]
complexes 1–Cl, 2–Cl, and 3–Cl quantitatively.
The coordination environment of 1, 2, and 3 differs from
1–Cl, 2–Cl, and 3–Cl only in the nature of the axial ligand. The
iron centres in 1–Cl25 and [(HCX)FeIIICl–corrꢀ+ 17 are known
]
to be five-coordinate, with one chloride bound axially, though
the coordination environments of 2-py and 3-Et2O are not
certain without crystallographic evidence. Nonetheless,
1H NMR spectra are consistent with the presence of only
one axial ligand. The xanthene scaffold imposes an asymmetry
and axial ligands can reside on the syn and anti face of the
corrole macrocycle and consequently two distinct sets of axial
ligand NMR shifts would be observed for a six-coordinate
hangman complex. However, the 1H NMR spectra of 2-py and
Department of Chemistry, 6-335, Massachusetts Institute of Technology,
Cambridge, MA 02139, USA. E-mail: nocera@mit.edu
w Electronic supplementary information (ESI) available: Synthetic
procedures, NMR, Mass, and UV-vis spectra. See DOI: 10.1039/
c2cc30580a
Scheme 1
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 4175–4177 4175

View Article Online
Fig. 1 Titration of pyridine, 0 (—), 4.1, 8.3, 12.4, 16.6, 20.7 and
41 (- - -) mM, into a 15 mM solution of 2 in THF, monitored by UV-vis.
Fig. 2 Turnover number for catalytic H2O2 disproportionation by
iron corroles: 1-py (’), 2 ( ), 3 ( ), 1–Cl ( ), 2–Cl ( ) and 3–Cl ( ).
TON measured over the first 2 min of the reaction. TON determined
from volumetric measurement of O2 produced.
3-Et2O display only one unique set of axial ligand NMR shifts
(Fig. S1 and S3, ESIw).
Table 1 TON and TOF of catalytic H2O2 dismutationa
Although pre-catalysts 1-py, 2-py, and 3-Et2O differ in the
identity of their axial ligands, experimental conditions were
such that the iron coordination environments were identical
immediately preceding catalysis. The labile nature of the
pyridine ligand on Fe corroles has been established by
1H NMR experiments.26 We provide further support for
pyridine exchange upon dissolution of 2-py in THF with the
UV-vis titration data provided in Fig. 1. An absorption
spectrum of 2-py in pentane displayed the expected Soret band
at 407 nm and a prominent Q-band at 556 nm (Fig. S5, ESIw).
Upon changing the solvent to THF, a shoulder at 333 nm
increased in intensity, a shoulder at 368 nm was replaced by a
shoulder at 385 nm, and a Q-band at 758 nm was blue-shifted
to 737 nm. We ascribe these changes to displacement of the
pyridine by THF to produce 2. The spectra shown in Fig. 1 for
the titration of 2 with pyridine in THF confirm this contention.
The final absorption spectrum of 2 in the presence of excess
pyridine is identical to that of 2-py in pentane. The similarities
observed in the absorption spectra obtained upon dissolving
2-py and 3-Et2O in THF suggest that the coordinated Et2O
molecule in 3-Et2O is also displaced promptly upon dissolution
to form 3, and furthermore that the coordination environments
of 2 and 3 in THF are similar (Fig. S6, ESIw).
The dismutation chemistry of 1, 2, and 3 was compared to
their one-electron oxidized congeners 1–Cl, 2–Cl, and 3–Cl.
The hydrophobicity of 1–3 together with their tendency to be
instantaneously oxidized in chlorinated solvents led us to a
3 : 1 THF : MeOH solvent system for dismutation experiments.
Fig. 2 plots the TON for each of the compounds employed in
this study. In each experiment, a 0.50 mM solution of catalyst in
2 mL 3 : 1 THF : MeOH was exposed at once to an excess
(1000 equiv.) of H2O2 and the amount of generated O2 was
quantified using an apparatus described in the ESI.w In order to
avoid initial O2-induced catalyst decomposition, all solvents
and reagents were thoroughly purged with argon prior to each
experiment. As shown in Table 1, the [FeIII–corr] systems 1–3
clearly exhibit enhanced activity over their one electron-oxidized
analogues in the given solvent system. The total amount of O2
generated by 2 and 3 (1.9 mL and 2.0 mL, respectively) was less
than the amount generated by 1 (2.3 mL); this is reflected in the
reduced TONs. Moreover, the total amount of O2 generated by
Compound
TONb
TOFc/minÀ1
1
2
3
1–Cl
2–Cl
3–Cl
95.7
79.0
83.2
54.1
59.6
55.7
71
75
170
53
58
77
a
Extent of reaction determined from volumetric measurement of
b
c
O2 produced. Calculated after 2 min. Determined from first 20 s
of the reaction.
2 and 3 was approximately equal, suggesting that they decompose
at similar rates with respect to O2 concentrations, and faster than
that observed for 1. We attribute this observation to the greater
stability of 1 as compared to 2 and 3. The presence of a third
pentafluorophenyl meso-substituent on the macrocycle of 1 as
compared to 2 and 3 serves to decrease electron density in the
corrole p-system, thereby increasing catalyst lifetime by retarding
oxidation by O2 and subsequent m-O formation. Introduction of
the electron releasing xanthene backbone shortens the catalyst
lifetime and TON. Notwithstanding, the clear benefit of the
hangman effect is evident in the initial rate of O2 production
for 3, which far surpasses the initial rates displayed by both
1 and 2.
1–Cl, 2–Cl and 3–Cl display inferior catalytic activity
compared to 1, 2, and 3. The similarity between TONs of
1–Cl, 2–Cl and 3–Cl may be explained by the facile formation
of catalytically inactive m-O complexes, known to result from
the reaction of [FeIIICl–corrꢀ+] and OHÀ.25 The diminished
activity of the [FeIIICl–corrꢀ+] may be understood within the
context of the prevailing mechanistic model for H2O2 dismu-
tation at the metal cofactors, which include the Fe porphyrin
centres of catalase enzymes.27–30 Dismutation is believed to
initiate with the coordination of peroxide to the Fe{n} centre to
form a Fe{n}–OOH adduct. Subsequent addition of H+ and loss
of water raise the formal oxidation state by two levels to furnish
[Fe{n+2}(QO)] oxo. Reaction with another molecule of H2O2
results in the release of O2, H2O, and regenerates the Fe{n+2}
.
The presence of an acid–base moiety in the secondary coordi-
nation sphere of the metal, as provided in a hangman motif,
c
4176 Chem. Commun., 2012, 48, 4175–4177
This journal is The Royal Society of Chemistry 2012

View Article Online
acknowledges the MIT Energy Initiative and the National
Science Foundation for predoctoral fellowships.
Notes and references
1 R. L. Shook and A. S. Borovik, Inorg. Chem., 2010, 49, 3646.
2 C. K. Chang, Y. Liang, G. Avilds and S.-M. Peng, J. Am. Chem.
Soc., 1995, 117, 4191.
3 Y. Gao, J. Liu, M. Wang, Y. Na, B. Akermark and L. Sun,
Tetrahedron, 2007, 63, 1987.
4 J. Rosenthal and D. G. Nocera, Acc. Chem. Res., 2007, 40, 543.
5 J. Rosenthal and D. G. Nocera, Prog. Inorg. Chem., 2007, 55, 483.
6 D. K. Dogutan, S. A. Stoian, R. McGuire Jr., M. Schwalbe,
T. S. Teets and D. G. Nocera, J. Am. Chem. Soc., 2011, 133, 131.
7 C. H. Lee, D. K. Dogutan and D. G. Nocera, J. Am. Chem. Soc.,
2011, 133, 8775.
8 R. McGuire Jr., D. K. Dogutan, Y. Shao-Horn and D. G. Nocera,
J. Am. Chem. Soc., 2011, 133, 9178.
9 J. Y. Yang, J. Bachmann and D. G. Nocera, J. Org. Chem., 2006,
71, 8706.
Scheme 2
10 C. J. Chang, L. L. Chng and D. G. Nocera, J. Am. Chem. Soc.,
2003, 125, 1866.
11 J. Rosenthal, L. L. Chng, S. D. Fried and D. G. Nocera, Chem.
Commun., 2007, 2642.
12 S.-Y. Liu and D. G. Nocera, J. Am. Chem. Soc., 2005, 127, 5278.
13 L. L. Chng, C. J. Chang and D. G. Nocera, Org. Lett., 2003,
5, 2421.
14 I. Aviv-Harel and Z. Gross, Chem.–Eur. J., 2009, 15, 8382.
15 Z. Gershman, I. Goldberg and Z. Gross, Angew. Chem., Int. Ed.,
2007, 46, 4320.
16 L. Kupershmidt, Z. Okun, T. Amit, S. Mandel, I. Saltsman,
A. Mahammed, O. Bar-Am, Z. Gross and M. B. H. Youdim,
J. Neurochem., 2010, 113, 363.
17 M. Schwalbe, D. K. Dogutan, S. A. Stoian, T. S. Teets and
D. G. Nocera, Inorg. Chem., 2011, 50, 1368.
18 A. Mahammed and Z. Gross, Chem. Commun., 2010, 46, 7040.
19 S. Ye, T. Tuttle, E. Bill, L. Simkhovich, Z. Gross, W. Thiel and
F. Neese, Chem.–Eur. J., 2008, 14, 10839.
20 E. Steene, T. Wondimagegn and A. J. Ghosh, J. Phys. Chem. B,
2001, 105, 11406.
21 Z. Gross, N. Galili, L. Simkhovich, I. Saltsman, M. Botoshansky,
facilitates the protonation of the peroxy species and formation of
the oxo as has been observed by stopped-flow spectroscopy.31,32
For the studies described here, the redox non-innocence of the
corrole ligand provides a mechanism for redox levelling of
the potential required for dismutation by avoiding a high
energy [FeV(QO)] intermediate (Scheme 2). In the protonation
of the initial peroxy to form the oxo, only FeIV is needed with
the participation of the corrole ligand. Conversely, with a
[FeIIICl–corrꢀ+] pre-catalytic state, oxo formation must proceed
through the FeV formal oxidation state even with the partici-
pation of the corrole. Considering the dramatically different
potentials for [FeIV(QO)] as compared to [FeV(QO)], driving
dismutation from a [FeIII–corr] resting state permits the
thermodynamically costly [FeV(QO)] to be circumvented.
In summary, we have shown that the oxidation state of the
hangman pre-catalyst plays a substantial role in determining
the catalytic activity of Fe corrole catalyzed H2O2 dismutation
reactions, consistent with observations made for other [Fe–corr]
catalyzed reactions.33 [FeIII–corr] is nearly twice as active towards
H2O2 dismutation as their related [FeIIICl–corrꢀ+] complexes.
Presumably, this is due to their ability to avoid high energy
iron–oxo intermediates by relying on the redox chemistry
of the corrole to redox level the metal by contributing an
oxidizing equivalent to the overall dismutation transfor-
mation. Whereas the hangman effect is obscured from a
[FeIIICl–corrꢀ+] resting state, it is prevalent for a [FeIII–corr]
resting state. Recent measurements show that proton transfer
from the hanging group can be extremely fast (>106 sÀ1).34
Such a facile proton transfer to a peroxy intermediate is
greatly beneficial to promoting dismutation, and hence is
undoubtedly at the origin of the enhanced catalytic activity
of the hangman system.
D. Blaser, R. Boese and I. Goldberg, Org. Lett., 1999, 1, 599.
¨
22 Z. Gross, N. Galili and I. Saltsman, Angew. Chem., Int. Ed., 1999,
38, 1427.
23 Z. Gross, G. Golubkov and L. Simkhovich, Angew. Chem., Int.
Ed., 2000, 39, 4045.
24 D. T. Gryko and K. J. Jadach, J. Org. Chem., 2001, 66, 4267.
25 L. Simkhovich, I. Goldberg and Z. Gross, Inorg. Chem., 2002,
41, 5433.
26 L. Simkhovich, A. Mahammed, I. Goldberg and Z. Gross,
Chem.–Eur. J., 2001, 7, 1041.
27 S.-I. Ozaki, M. P. Roach, T. Matsui and Y. Watanabe, Acc. Chem.
Res., 2001, 34, 818.
28 T. L. Poulos and J. Kraut, J. Biol. Chem., 1980, 255, 8199.
29 T. Matsui, S.-I. Ozaki, E. Liong, G. N. Phillips and Y. Watanabe,
J. Biol. Chem., 1999, 274, 2838.
30 E. Derat and S. Shaik, J. Phys. Chem. B, 2006, 110, 10526.
31 J. D. Soper, S. V. Kryatov, E. V. Rybak-Akimova and D. G.
Nocera, J. Am. Chem. Soc., 2007, 129, 5069.
32 S.-Y. Liu, J. D. Soper, J. Y. Yang, E. V. Rybak-Akimova and
D. G. Nocera, Inorg. Chem., 2006, 45, 7572.
33 L. Simkhovich and Z. Gross, Tetrahedron Lett., 2001, 42, 8089.
34 M. M. Roubelakis, D. K. Bediako, D. K. Dogutan and D. G. Nocera,
Energy Environ. Sci., 2012, 3, DOI: 10.1039/c2ee21123h.
We thank Robert McGuire Jr. for insightful discussions along
with funding from DOE DE-FG02-05ER15745. D.J.G. gratefully
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 4175–4177 4177