Mechanism of catalytic aziridination with manganese corrole: The often postulated high-valent Mn(V) imido is not the group transfer reagent

Article abstract of DOI:10.1021/ja0665489

The reaction of Arl=NTs (Ar = 2-(tert-butylsulfonyl)benzene and Ts = p-toluenesulfonyl) and (tpfc)Mn (tpfc = 5,10,15-tris(pentafluorophenyl)corrole), 1, affords the high-valent (tpfc)Mn^V=NTs, 2, on stopped-flow time scale. The reaction proceeds

Full text of DOI:10.1021/ja0665489

Published on Web 12/20/2006
Mechanism of Catalytic Aziridination with Manganese
Corrole: The Often Postulated High-Valent Mn(V) Imido Is Not
the Group Transfer Reagent
Michael J. Zdilla and Mahdi M. Abu-Omar*
Contribution from the Brown Laboratory, Department of Chemistry, Purdue UniVersity,
560 OVal DriVe,West Lafayette, Indiana 47907
Received September 11, 2006; E-mail: mabuomar@purdue.edu
Abstract: The reaction of ArIdNTs (Ar ) 2-(tert-butylsulfonyl)benzene and Ts ) p-toluenesulfonyl) and
(tpfc)Mn (tpfc ) 5,10,15-tris(pentafluorophenyl)corrole), 1, affords the high-valent (tpfc)MnVdNTs, 2, on
stopped-flow time scale. The reaction proceeds via the adduct [(tpfc)MnIII(ArINTs)], 3, with formation constant
K3 ) (10 ( 2) × 103 L mol-1. Subsequently, 3 undergoes unimolecular group transfer to give complex 2
with the rate constant k4 ) 0.26 ( 0.07 s-1 at 24.0 °C. The complex (tpfc)Mn catalyzes [NTs] group transfer
from ArINTs to styrene substrates with low catalyst loading and without requirement of excess olefin. The
catalytic aziridination reaction is most efficient in benzene because solvents such as toluene undergo a
competing hydrogen atom transfer (HAT) reaction resulting in H2NTs and lowered aziridine yields. The
high-valent manganese imido complex (tpfc)MndNTs does not transfer its [NTs] group to styrene. Double-
labeling experiments with ArINTs and ArINTstBu (TstBu ) (p-tert-butylphenyl)sulfonyl) establish the source
of [NR] transfer as a “third oxidant”, which is an adduct of Mn(V) imido, [(tpfc)Mn(NTstBu)(ArINTs)] (4).
Formation of this oxidant is rate limiting in catalysis.
Introduction
years ago, examples of stable metal imides were extremely rare
for the mid-to-late metals of the first-row (3d series).6,7 Besides
Imido complexes of high-valent manganese and iron have
been postulated as the active reagents responsible for group
transfer in aziridination reactions.1 For example, Groves and
Carreira and co-workers have elegantly implicated MnVd
N(C(O)CF3) in porphyrin and Schiff base ligand environments,
respectively, as group transfer reagent to carbon-carbon double
bonds.2 Even though Mn(V) nitrides have been stabilized by
various ancillary ligands, terminal Mn(V) imides are rare.3 We
have been able to obtain isolable Mn(V) imides stabilized by
5,10,15-tris(pentafluorophenyl)corrole (tpfc) via the action of
organic aryl azides on Mn(III).4 More recently Goldberg and
co-workers carried out successfully the same reaction with a
corrolazine auxiliary ligand.5 Additionally and only until a few
their prevalence in catalysis, multiply bonded species, especially
for porphyrinoid ligands, play a part in biomimetic and bio-
inspired chemistry of metalloenzymes.8
For the reasons cited above, the study and characterization
of reactive intermediates constitute a major research goal in
mechanistic chemistry. The consensus mechanism for atom and
group transfer has pointed to LnM(V)dE (where M ) Mn or
Fe and E ) O or NR) as the oxidizing species.9,1a,2 Investigation
into ruthenium bis(tosylimido porphyrin) complexes have sup-
ported this notion.10 Besides this traditional mechanism, evi-
dence has been reported recently for involvement of a porphyrin
metal-oxidant intermediate (eq 1) as a competing oxidant with
(7) Recent examples of well-defined multiply bonded terminal ligands have
been reported in low coordinate geometries for Fe: (a) Brown, S. D.; Peters,
J. C. J. Am. Chem. Soc. 2005, 127, 1913. (b) Kaizer, J.; Klinker, E. J.; Oh,
N. Y.; Rohde, J. U.; Song, W. J.; Stubna, A.; Kim, J.; Mu¨nck, E.; Nam,
W.; Que, L., Jr. J. Am. Chem. Soc. 2004, 126, 472. (c) Bart, S. C.;
Lobkovsky, E.; Bill, E.; Chirik, P. J. J. Am. Chem. Soc. 2006, 128, 5302.
For Co: (d) Shay, D. T.; Yap, G. P. A.; Zakharov, L. N.; Rheingold,
A. L.; Theopold, K. H. Angew. Chem. Int. Ed. 2005, 44, 1508. (e) Dai, X.
L.; Kapoor, P.; Warren, T. H. J. Am. Chem. Soc. 2004, 126, 4798. (f)
Jenkins, D. M.; Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2002, 124,
11238. For Ni: (g) Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc.
2002, 124, 9976. (h) Kogut, E.; Wiencko, H. L.; Zhang, L. B.; Cordeau,
D. E.; Warren, T. H. J. Am. Chem. Soc. 2005, 127, 11248.
(8) (a) Groves, J. T. In Cytochrome P450: Structure, Mechanism, and
Biochemistry, 3rd ed.; Ortiz de Montellano, P. R.; Ed.; Kluwer Academic/
Plenum Publishers: New York, 2004; pp 1-44. (b) Collman, J. P.;
Boulatov, R.; Sunderland, C. J.; Fu, L. Chem. ReV. 2004, 104, 561. (c)
Denisov, I. G.; Makris, T. M.; Sligar, S. G.; Schlichting, I. Chem. ReV.
2005, 105, 2253. (d) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr.
Chem. ReV. 2004, 104, 939. (e) Kryatov, S. V.; Rybak-Akimova, E. V.;
Schindler, S. Chem. ReV. 2005, 105, 2175. (f) Abu-Omar, M. M.; Loaiza,
A.; Hontzeas, N. Chem. ReV. 2005, 105, 2227.
(1) (a) Mu¨ller, P.; Fruit, C. Chem. ReV. 2003, 103, 2905. (b) Eikey, R. A.;
Abu-Omar, M. M. Coord. Chem. ReV. 2003, 243, 83.
(2) (a) Groves, J. T.; Takahashi, T. J. Am. Chem. Soc. 1983, 105, 2073. (b)
Du Bois, J.; Tomooka, C. S.; Hong, J.; Carreira, E. M. Acc. Chem. Res.
1997, 30, 364.
(3) Mn(V) nitrides are know for porphyrin and other ancillary ligands. See for
example: (a) Buchler, J. W.; Dreher, C.; Lay, K. L.; Lee, Y. J. A.; Scheidt,
W. R. Inorg. Chem. 1983, 22, 888. (b) Hill, C. L.; Hollander, F. J. J. Am.
Chem. Soc. 1982, 104, 7318. (c) Meyer, K.; Bendix, J.; Metzler-Nolte, N.;
Weyhermu¨ller, T.; Wieghardt, K. J. Am. Chem. Soc. 1998, 120, 7260. (d)
Golubkov, G.; Bendix, J.; Gray, H. B.; Mahammed, A.; Goldberg, I.;
DiBillio, A. J.; Gross, Z. Angew. Chem. Int. Ed. 2001, 40, 2132. (e) Chang,
C. J.; Connick, W. B.; Low, D. W.; Day, M. W.; Gray, H. B. Inorg. Chem.
1998, 37, 3107.
(4) Eikey, R. A.; Khan, S. I.; Abu-Omar, M. M. Angew. Chem. Int. Ed. 2002,
41, 3592.
(5) Lansky, D. E.; Kosack, J. R.; Narducci-Sarjeant, A. A.; Goldberg, D. P.
Inorg. Chem. 2006, 45, 8477.
(6) (a) Glueck, D. S.; Wu, J. X.; Hollander, F. J.; Bergman, R. G. J. Am. Chem.
Soc. 1991, 113, 2041. (b) Verma, A. K.; Nazif, T. N.; Achim, C.; Lee, S.
C. J. Am. Chem. Soc. 2000, 122, 11013.
9
10.1021/ja0665489 CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 16971-16979
16971

A R T I C L E S
Zdilla and Abu-Omar
the multiply bonded oxo complex.11 Through convincing isotope
labeling experiments, Goldberg and co-workers presented yet
another possible “third oxidant” in which the Mn(V) oxo
corrolazine forms an adduct with the oxidant, and the oxygen
atom of the adduct is the one that transfers to substrate (eq 2).12
where por ) porphyrin; M ) Fe or Mn; XO ) H2O2, ROOH,
or PhIdO; and Y ) substrate such as organic sulfides or
alkenes.
Figure 1. Comparative UV-vis spectra of 2 and (tpfc)MnNMes in
acetonitrile.
(ArIdNTs) adduct of (tpfc)MnVdNTs, analogous to the “third
oxidant” proposed by Goldberg and co-workers.12
where Cz ) corrolazine and Y ) PhSMe.
Results and Discussion
Our stable and isolable Mn(V) imido corrole complexes are
not oxidizing and do not effect [NR] group transfer to organic
substrates.4,13 The synthetic methodology we employed in
preparing (tpfc)MnVdNR was limited to aryl groups with alkyl
or halide substituents in the 2,6-(ortho) positions.14 Therefore,
we reasoned that more electron-withdrawing substituents are
needed on the imido ligand such as acyl or tosyl to induce
efficient group transfer. These complexes were not possible to
synthesize using organic azides because tosyl azide, for example,
cannot be easily activated by thermolysis or photolysis.14 In this
work, we report on the reaction of soluble iodonium ylides15
(ArIdNTs, Ar ) 2-(tert-butylsulfonyl)benzene and Ts )
p-toluenesulfonyl) and (tpfc)MnIII, 1, to afford characterizable
MnVdNTs, 2 (it should be noted that while we refer formally
to MnIII and MnV in complexes 1 and 2, it is possible that a
more electronically accurate description for either is the one-
electron reduced metal with a corrole radical cation16). The
manganese corrole complex (tpfc)MnIII is an effective catalyst
for group transfer to styrene substrates in benzene. The reaction’s
main highlights are low catalyst loading and no requirement
for excess olefin. When the reaction is conducted in toluene,
aziridine yields are lowered because of intervening hydrogen
atom transfer (HAT) from the solvent. While the terminal Mn-
(V) tosyl imide is reactive toward HAT reactions, it does NOT
transfer the terminal tosyl imide ligand to styrene substrates.
Furthermore, we present unambiguous evidence through labeling
experiments that the oxidant in this system is an iodoimine
Reaction of (tpfc)MnIII(EtOAc) with ArIdNTs. The man-
ganese(III) complex of tpfc is often written in the literature as
[(tpfc)Mn] without specifying axial ligation because crystal-
lization proved difficult without addition of a ligand additive
such as OPPh3.17 We have developed a successful purification/
crystallization protocol that affords 5-coordinate manganese with
a single axial ethyl acetate ligand (see Supporting Information
for structural data). Addition of ArIdNTs or PhIdNTs to (tpfc)-
Mn(EtOAc), 1, at room temperature in a variety of dry organic
solvents (CH2Cl2, CH3CN, benzene, or toluene) results in a color
change from green to red-brown over several seconds with ArI/
PhI as the organic byproduct. The resulting manganese complex
can be isolated by scraping microcrystalline residue from the
sides of the vial after fast drying of the reaction mixture. The
compound has been isolated in this way only in sparing amounts
(the majority of the material remains in a mixture at the bottom
of the vial) but decomposes in solution back to [(tpfc)MnIII]
over a period of 10 h at ambient temperature.
The resulting red-brown manganese complex can be assigned
as the tosylimide of Mn(V), (tpfc)MnVdNTs (2) by spectro-
scopic comparison to the structurally characterized arylimido
analogues.4 The electronic spectrum of 2 shows distinctly
different band structure from the Mn(III) starting material and
compares well with the UV-vis of (tpfc)MnVdNMes (Mes )
2,4,6-trimethylphenyl), exhibiting two soret bands near 350 and
400 nm. The Q-band near 520 nm is less distinct and appears
as a shoulder (Figure 1). Furthermore, the ESI mass spectrum
of 2 exhibits a parent m/z peak of 861.9, which corresponds to
(tpfc)Mn(N) as a result of N-S bond cleavage in the gas phase.
(9) (a) McLain, J. L.; Lee, J.; Groves, J. T. In Biomimetic Oxidations Catalyzed
by Transition Metal Complexes; Meunier, B., Ed.; Imperial College Press:
London, 2000; pp 91-169. (b) Meunier, B.; Bernadou, J. Top. Catal. 2002,
21, 47. (c) Watanabe, Y. In The Porphyrin Handbook; Kadish, K. M., Smith,
K. M., Guilard, R., Eds.; Academic Press: New York, 2000; Vol. 4, pp
97-117.
(10) (a) Au, S.-M.; Huang, J.-S.; Yu, W.-Y.; Fung, W.-H.; Che, C.-M. J. Am.
Chem. Soc. 1999, 121, 9120. (b) Leung, S. K.-Y.; Tsui, W.-M.; Huang,
J.-S.; Che, C.-M.; Liang, J.-L.; Zhu, N. J. Am. Chem. Soc. 2005, 127, 16629.
(c) Liang, J.-L.; Yuan, S.-X.; Huang, J.-S.; Che, C.-M. J. Org. Chem. 2004,
69, 3610. (d) Liang, J.-L.; Huang, J.-S.; Yu, X.-Q.; Zhu, N. Y.; Che,
C.-M. Chem. Eur. J. 2002, 8, 1563. (e) Au, S.-M.; Fung, W.-H.; Cheng,
M.-C.; Che, C.-M.; Peng, S.-M. Chem. Commun. 1997, 1655.
(11) (a) Collman, J. P.; Chien, A. S.; Eberspacher, T. A.; Brauman, J. I. J. Am.
Chem. Soc. 2000, 122, 11098. (b) Nam, W.; Choi, S. K.; Lim, M. H.; Rohde,
J.-U.; Kim, I.; Kim, J.; Kim, C.; Que, L. Jr. Angew. Chem. Int. Ed. 2003,
42, 109. (c) Nam, W.; Lim, M. H.; Lee, H. J.; Kim, C. J. Am. Chem. Soc.
2000, 122, 6641.
(12) Wang, S. H.; Mandimutsira, B. S.; Todd, R.; Ramdhanie, B.; Fox, J. P.;
Goldberg, D. P. J. Am. Chem. Soc. 2004, 126, 18.
(13) Edwards, N. Y.; Eikey, R. A.; Loring, M. I.; Abu-Omar, M. M. Inorg.
Chem. 2005, 44, 3700.
(14) Abu-Omar, M. M.; Shields, C. E.; Edwards, N. Y.; Eikey, R. A. Angew.
Chem. Int. Ed. 2005, 44, 6203.
(15) Macikenas, D.; Skrzypczak-Jankun, E.; Protasiewicz, J. D. J. Am. Chem.
Soc. 1999, 121, 7164.
(16) (a) Turner, P; Gunter, M. J. Inorg. Chem. 1994, 33, 1406. (b) Groves,
J. T.; Haushalter, R. C.; Nakamura, M.; Nemo, T. E.; Evans, B. J. J. Am.
Chem. Soc. 1981, 103, 2884.
(17) Bendix, J.; Gray, H. B.; Golubkov, G.; Gross, Z. Chem. Commun. 2000,
1957.
9
16972 J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006

Aziridination with Managanese Corrole
A R T I C L E S
of the second π-bond would lower the energy of the corre-
sponding d orbital, resulting in the observed triplet state (Figure
2). This trend has been observed by Odom and co-workers in
a series of bis-imido metal complexes of the chromium group.
Using NMR shielding and DFT arguments, this study concluded
that a bend at nitrogen coincides with a more shielded (hence
more nitrene-like) imide ligand, which is an overall poorer
electron donor than a linear imide.19 The S ) 1 spin state has
been observed for porphyrinato manganese(V) oxo by magnetic
measurements.20 It should be noted that the observed value of
µeff (3.24 BM) is slightly higher than the theoretical value of
2.83 BM due to some inevitable decomposition of 2 back to
the higher magnetic moment 1. When 2 is allowed to decompose
back to green 1, the expected spin state (S ) 2, µeff ) 4.72
BM) of MnIII(tpfc) is observed by magnetic measurement (Evans
method). S ) 2 is observed for 1 regardless of axial ligation.21
In addition to water, solvents such as toluene and acetonitrile
result in decomposition of 2, though at much slower rate, and
via HAT instead of protolysis. The organic byproduct from de-
composition of 2 is H2NTs. Even though complex 2 lasts hours
in dry benzene, it still decomposes back to MnIII, 1, through
pathways that we cannot fully account for at this time. Small
amounts of adventitious moisture that cannot be avoided could
be responsible for the slow decomposition in benzene. However,
we did not observe formation of (tpfc)MnV(N)3d or disubstituted
diazene upon decomposition of 2 (NMR, UV-vis, and GC).
The absence of the latter suggests that this system is not
amenable to bimolecular coupling of the imido group. Formation
and decomposition of 2 is summarized in Scheme 1.
The EPR spectrum of 2 generated in situ exhibits a number
of assignable signals, suggesting some degree of speciation in
solution.22 The concentration of these extra species is most likely
very minor, given the comparability of the absorption spectrum
of 2 to that of (tpfc)MnVdNMes, which suggests that the
majority of manganese in solution is in the form of terminal
imido. Among the EPR signals observed is a broad feature at g
) 4, which we assign to 2. While no comparable signal has
been reported for a high-spin manganese(V) porphyrin or corrole
complex to our knowledge, this signal is consistent with a spin
forbidden transition ∆ms ) (2 for a triplet-state complex and
thus a weak signal.22d,e Upon addition of H2O to form
diamagnetic (tpfc)MndO, this feature disappears. In summary,
characterization of complex 2 is consistent with high-spin (tpfc)-
Figure 2. Geometric effects on electronic structure of d2 metals with linear
terminal mesitylimido (left) and bent terminal tosylimido (right) under Cs
symmetry. The dxz and dyz orbitals are lowered in energy in the tosylimido
complex due to weaker overall π-donation by the tosylimido ligand in
comparison to mesitylimido. The bend at nitrogen in the tosylimido complex
further lowers the energy of the dyz orbital by disrupting the π-overlap
between the nitrogen py and metal dyz orbitals, resulting in a high-spin
complex.
Even though the manganese tosylimide was problematic in
affording large enough single crystals for X-ray analysis, the
structure of (tpfc)CrVdNTs, prepared from the reaction of ArId
NTs and (tpfc)CrIII(py)2, was solved (see Supporting Informa-
tion).
Unlike the diamagnetic (tpfc)MnVdNMes, 2 is paramagnetic,
exhibiting broad 19F signals, and no visible resonances in the
1
1H NMR spectrum are observed. Instead, the diamagnetic H
and 19F NMR signals belonging to (tpfc)MnVdO (5) are
observed in small amounts as a result of protolysis by H2O.18
Magnetic susceptibility measurements (Evans method) demon-
strate a µeff of 3.24 BM, most consistent with an S ) 1 complex.
While this spin state differs from the S ) 0 (tpfc)MnVdNMes,
it is consistent with a terminal imido with poor π-donation,
which lowers the energy of the dxz and dyz orbitals in comparison
to NMes. Additional electronic structure information can be
inferred from the structure of (tpfc)CrVdNTs in the Supporting
Information. While the CrsN terminal imido bond length (1.65
Å) is comparable to that of the known (tpfc)CrdNMes and
(tpfc)MndNMes, the nitrogen atom of the tosylimido exhibits
a bond angle of 150.7°, while the mesitylimidos are linear at
nitrogen. This bend at nitrogen suggests at most a single strong
N-Cr π-bond, perpendicular to the Cr-N-S plane. Assuming
structural similarity for the manganese complex, the absence
(19) Ciszewski, J. T.; Harrison, J. F.; Odom, A. L. Inorg. Chem. 2004, 43, 3605.
(20) Groves, J. T.; Kruper, W. J.; Haushalter, R. C. J. Am. Chem. Soc. 1980,
102, 6375.
(21) The S ) 2 spin state is observed when there is no axial ligation in solution,
as well as when the complex exists as a pyridine adduct based on magnetic
measurement by Evans’ method (µeff ) 5.28 BM, see Supporting Informa-
tion).
(18) (a) Gross, Z.; Golubkov, G.; Simkhovich, L. Angew. Chem. Int. Ed. 2000,
39 (22), 4045. (b) Addition of H2O to a red-brown solution of 2 results in
formation of a red solution with diamagnetic â-pyrrole 1H NMR signals
belonging to (tpfc)MnVdO, concurrently with TsNH2, the two compounds
appearing in a 1:1 ratio. We have confirmed the identity of this hydrolysis
product by comparison of the electronic absorption and 1H NMR spectra
to the manganese-oxo complex generated by reaction of 1 with PhIO. The
electronic absorption and 1H NMR spectra are identical whether generated
directly or by hydrolysis of 2, although the behavior of this manganese-
oxo is different in our hands than reported by Gross et al,18a which indicates
that the manganese oxo lasts minutes at millimolar concentrations (hours
in our hands), and hours at micromolar concentrations (days in our hands).
Additionally, this report claims that cold temperatures (-75 °C) are
necessary to observe broad b-pyrrole signatures, while we are able to obtain
a well-resolved spectrum with sharp signatures at room temperature. Lastly,
our electronic absorption spectrum differs substantially from Gross’s
reported spectrum for isolated (tpfc)MnVdO, but matches the spectrum he
reports for (tpfc)MnVdO generated in situ (spectra are provided in the
Supporting Information).
(22) (a) A broad signature near g ) 8 is consistent with residual Mn(III).22b As
manganese(III) porphyrin complexes are normally silent by X-band EPR,17
it is unclear whether this signal results from 1 or from trace Mn3+ in another
ligand environment. A hyperfine feature present at 4.3 is comparable to
those for Mn2+ ion in a high zero-field-splitting environment, such as in
the manganese superoxide dismutases.22c This feature remains after addition
of H2O. A broad signal at g ) 4 is assigned to 2.22d,e An intense 16-line
feature at g ) 2.0 is consistent with a S ) 1/2 manganese dimer.22e This
signal appears as a result of hydrolysis of 2 by H2O and is believed to be
a minor byproduct accompanying the formation of the major diamagnetic
(tpfc)MnVdO product observed by electronic absorption spectroscopy and
1H and 19F NMR. See Supporting Information for spectra. (b) Talsi, E. P.;
Bryliakov, K. P. MendeleeV Commun. 2004, 3, 111. (c) Hsieh, Y.; Guan,
Y.; Tu, C.; Bratt, P. J.; Angerhofer, A.; Lepock, J. R.; Hickey, M. J.; Tainer,
J. A.; Nick, H. S.; Silverman, D. N. Biochemistry. 1998, 37, 4731. (d)
Eisch, J. J.; Gitua, J. N.; Doetschman, D. C. Eur J. Inorg. Chem. 1994, 33,
1406. (e) Van der Waals, J. H.; de Groot, M. S. J Chim. Phys. 1964, 1643.
(f) Chen, H.; Tagore, R.; Das, S.; Incarvito, C.; Faller, J. W.; Crabtree,
R. H; Brudvig, G. W. Inorg. Chem. 2005, 44, 7661.
9
J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006 16973

A R T I C L E S
Zdilla and Abu-Omar
Scheme 1. Formation and Decomposition of MnV Tosylimido 2
MnVdNTs. Detailed spectroscopic characterization of 2 by
parallel-mode EPR and X-ray absorption spectroscopy will be
the subject of another report.
catalyst. The results were again negative. In conclusion,
irrespective of the starting form of the catalyst (MnIII vs MnV)
the kinetics profiles for aziridine formation were identical.
Next we considered the manganese-oxo complex resulting
from hydrolysis of 2 (vide supra) as the active catalyst. When
the aziridination of styrene is carried out in solvents that have
not been dried to remove water, the reaction mixture is bright
red (from (tpfc)MnO) instead of red-brown (from (tpfc)MnNTs),
and no aziridine is formed. Furthermore, when (tpfc)MnO is
generated directly from 1 and PhIO in dry solvent and subse-
quently added to a mixture of styrene and PhINTs, minimal
aziridination is observed. While (tpfc)MnO has been reported
to catalyze epoxidation of styrene,18a we did not observe stoich-
iometric oxidation of styrene by (tpfc)MnO to form styrene
oxide. From these observations, we conclude that the manganese
oxo complex 5 is not the catalyst responsible for group transfer.
Finally we considered the “third oxidant” suggested by
Goldberg and co-workers.12 Here the “third oxidant” could be
an iodoimine adduct of 2, (tpfc)MnV(NTs)(ArIdNTs) (4), or a
transient Mn(VII) bisimido, (tpfc)MnVII(NTs)2 (6). To address
the validity of either pathway we turned to labeling experiments.
We prepared the tert-butyl analogue of ArIdNTs with a p-tert-
butyl substituent instead of p-methyl on the tosylimine (Scheme
2). We first established that this reagent (ArIdNTstBu) indeed
forms aziridine with styrene in the presence of catalyst 1 at a
comparable rate to that observed for ArIdNTs, isolated the
aziridine, and obtained its GC retention time (15.2 min), which
is clearly distinguishable from the aziridine with the parent
p-tolylsulfonylimine (11.6 min). The expectations as detailed
in Scheme 2 are as follows: If the “third oxidant” is an adduct
Identity of the Active Manganese Catalyst Responsible
for Group Transfer. To our surprise the manganese(V)
tosylimide 2 does not transfer [NTs] to styrene to give aziridine.
With excess styrene, the red-brown solution of 2 turns green
and H2NTs along with ArI are detected by GC; however, no
aziridine is produced. Under these stoichiometric conditions we
have not been able to identify any organic products from styrene.
Purification of styrene via distillation to remove potentially
reactive impurities did not change the outcome. The one
affirmative conclusion we can ascertain from this reaction is
that (tpfc)MnVdNTs does not transfer [NTs] to styrene to give
aziridine. This finding may not be so surprising if we consider
the recent studies of Newcomb and co-workers on the reaction
kinetics of metal-oxo species containing porphyrin and corrole
ancillary ligand.23 In these studies, photochemically generated
(tpfc)MnVdO was found to be ∼105 less reactive as an oxygen
atom transfer species toward cyclooctene than its porphyrin
analogue (tpfp)MnVdO (tpfp ) 5,10,15,20-tetrakis(pentafluo-
rophenyl)porphyrin).
Further confirmation that complex 2 is not the active form
of the catalyst came from steady-state kinetics. Oxidative for-
mation of 2 (vide infra) is fast as compared to reduction of 2
back to 1. Nevertheless, under steady-state conditions, formation
of aziridine (followed by 1H NMR) was zero-order in [styrene]
(Figure 3). Also the rate of aziridine formation (initial rate
method) exhibited first-order dependence on the initial concen-
tration of ArINTs. Thus, formation of the oxidizing species is
rate-determining under catalytic steady-state conditions.
We turned first to the mechanism proposed by Nam and
Que11b for oxidation with iron porphyrin and PhIdO, namely,
a metal-oxidant adduct as the active species (eq 1). One
experimental finding we already had against this mechanism is
that the reaction between ArIdNTs and 1 to give 2 is not
reversible. To probe this further, we monitored the catalytic
reaction starting with a solution of (tpfc)MnIII (10 mol %) and
styrene and adding ArIdNTs last. We looked in the initial stages
to see if there is a burst of product that would result from the
adduct (tpfc)MnIII(ArIdNTs) prior to formation of complex 2.
The result was negative. We also looked for an induction period
if we started with the manganese tosylimide 2 as the starting
Figure 3. Steady-state kinetics for catalytic aziridination. 2 equiv of solid
ArINTs were stirred in a 24 mM solution of styrene with 1.5 mol % 1 as
catalyst (0.39 mM). Experimental data are represented by points. The solid
line represents the theoretical linear fit to the linear region: slope )
(-4.0 ( 0.1) × 10-5 M min-1; R ) 0.998.
(23) (a) Zhang, R.; Newcomb, M. J. Am. Chem. Soc. 2003, 125, 12418. (b)
Zhang, R.; Horner, J. H.; Newcomb, M. J. Am. Chem. Soc. 2005, 127,
6573. (c) Zhang, R.; Chandrasena, R. E. P.; Martinez, E., II; Horner, J. H.;
Newcomb, M. Org. Lett. 2005, 7, 1193. (d) Harischandra, D. N.; Zhang,
R.; Newcomb, M. J. Am. Chem. Soc. 2005, 127, 13776.
9
16974 J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006

Aziridination with Managanese Corrole
A R T I C L E S
Scheme 2. Potential Outcomes of the Double-Labeling Experiment for Different Aziridination Mechanisms
(4) of Mn(V) imide, only one aziridine would be formed. If the
“third oxidant” is a transient Mn(VII) bisimido (6), 50:50
mixture of two possible aziridines will be obtained.
smaller amount of TsNH2 is likely due to some decomposition
of ArIdNTs under these dilute (mM) stoichiometric conditions.
Nevertheless, more than 70% of the initial styrene was converted
to azirdine. As for the difference of a factor of 10 between the
two double-labeling experiments, we attribute this to the limited
precision of the experiment, precisely, in measuring an exact
equivalent of iodoimine relative to the starting Mn(III) complex
(1) at this scale. Furthermore, the reactivity of ArIdNTstBu is
quite comparable to that of ArIdNTs under steady-state
conditions and thus would not account for the observed
difference. In summary, a difference of 10 between the yields
of the two possible aziridines is sufficient to decisively
discriminate between the two possible mechanisms.
Probing Electronic Effects on Group Transfer to Alkenes.
Since the rate-determining step under steady-state kinetics does
not involve the alkene, styrene, information on the group transfer
step cannot be obtained from direct kinetic measurements.
However, measurement of product ratios resulting from two
styrene substrates that compete for the active intermediate would
be directly proportional to the ratio of the substrates’ rate
constants as long as the oxidant is limiting and the styrene
substrates are used in excess (Scheme 3 and eq 3). Three styrene
We performed this double labeling experiment twice. In the
first we generated complex 2a, (tpfc)MnVdNTstBu and added
1 equiv of ArIdNTs concurrently with 1 equiv of styrene. The
reaction product was analyzed by GC. The ratio of the two
aziridines NTs:NTstBu ) 100:1 (Figure 4). In the second
experiment, we generated complex 2 and added 1 equiv of ArId
NTstBu concurrently with 1 equiv of styrene. The ratio of the
two aziridines NTs:NTstBu ) 1:10. These findings provide
unambiguous evidence that the group transfer catalyst is an
adduct analogous to Goldberg’s “third oxidant” as illustrated
in Scheme 2, which also accounts for zero-order dependence
on [styrene] and lack of aziridine formation from Mn(V) imido
2. The presence of TstBuNH2 in the GC trace is because the
reaction was allowed to stand until the solution turned green,
that is all the Mn(V) imido reverted to manganese(III). The
Scheme 3. Ratio of Rate Constants from Ratio of Products under
Competition Kinetics
Figure 4. Gas chromatogram from the labeling competition experiment
between transfer of NTstBu vs NTs to olefin. In this experiment, to (tpfc)-
MnNTstBu was added 1 equiv of ArINTs and styrene concurrently. The
predominant signal at 11.6 is indicative of aziridination by NTs, while the
small signal at 15.2 is indicative of the minor aziridination by NTstBu. Other
signals appearing in the trace areTsNH2 at 6.7 min, TstBuNH2 at 8.2 min,
and ArI at 8.7 min.
[X-Ph-Az]∞ kX[X-PhCHdCH2]0
)
(3)
[Ph-Az]∞
kH[PhCHdCH2]0
9
J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006 16975

A R T I C L E S
Zdilla and Abu-Omar
Figure 5. Hammett plot from competition between substituted sty-
renes. 0.1 equiv of ArINTs was stirred with a 1:1 mixture of substituted
olefin and styrene (20 mM for each) and 1.5% catalyst 1 (0.29 mM).
Ratio of rate constants was calculated based on the ratio of aziridine
products. Points represent experimental data except for the origin point set
for styrene. The solid line represents the theoretical linear fit: slope ) -0.06
( 0.2.
substrates (4-Cl, 3-Cl, and 4-CF3-styrene) were selected for
competition with styrene. The aziridine product ratios for each
competition experiment were determined by GC and were found
to be quite close to unity in each instant. A Hammett plot of
log(kx/kH), calculated from the aziridine product ratio per eq 3,
versus the Hammett substituent constant σ is shown in Figure
5. There is no significant electronic effect for the group transfer
step. The reaction constant F is essentially zero (-0.06 ( 0.2).
This observation is consistent with a nonpolar transition state
that is unaffected by electronic substitutents on the aryl ring.24
The most likely scenario is a radical-type mechanism in which
the olefin approaches from the unhindered â-carbon to form an
intermediate nitrogen and carbon radical cage, followed by rapid
recombination to afford the product (eq 4). This mechanism
has been invoked recently for aziridination in one report by
ruthenium and in another for manganese and iron porphyrins
based upon secondary isotope effects and product stereochemical
arguments, respectively.10a,25 The mechanism in eq 4 is also
consistent with the observation that 1 is an effective catalyst
for R-disubstituted styrenes but not for â-substituted styrene:
Figure 6. (A) Stack UV-vis plot showing isosbestic conversion of 1 to 2
by titration with ArINTs. (B) A typical kinetic trace for the formation of 2
from 1 and excess ArINTs monitored using stopped-flow spectroscopy at
600 nm (data designated by points, fit designated by solid line). Conditions:
T ) 24 °C, [1] ) 2.53 × 10-5 M, and [ArINTs] ) 6.3 × 10-4 M.
Figure 7. Plot of pseudo-first-order rate constant vs [ArINTs] exhibiting
saturation kinetics. Points represent experimental data. Solid line represents
theoretical fit. k4 ) 0.26 ( 0.07 s-1 and K3 ) (10 ( 2) × 103 L mol-1 at
24.0 ( 0.2 °C, R ) 0.98.
Kinetics and Mechanism of Manganese(V) Imido Forma-
tion. While PhIdNTs is the common nitrene source in aziri-
dination reactions,1a its use in quantitative kinetics is hampered
by its lack of solubility and polymeric nature. The development
of molecular analogues such as ArIdNTs (Ar ) 2-(tert-
butylsulfonyl)benzene)15 that are soluble makes kinetic studies
feasible. We studied the kinetics of the reaction of ArIdNTs
with (tpfc)Mn(EtOAc) using stopped-flow spectroscopy. Under
pseudo-first-order conditions (limiting manganese corrole and
excess ArIdNTs) disappearance of manganese(III) complex 1
monitored at λ ) 600 nm is first order in 1 (Figure 6) and the
plot of kψ (pseudo-first-order rate constants) versus [ArIdNTs]
(Figure 7) features kinetic saturation in oxidant. Two kinetically
indistinguishable mechanisms can account for the observed rate
law (Scheme 4 and eqs 5 and 6). The rate law for mechanism
A in Scheme 4 based on the steady-state approximation is given
in eq 5. This pathway (Scheme 4A) is discounted on the basis
that negligible inhibition is observed by added ethyl acetate.
At ethyl acetate concentrations comparable to those in our
kinetic runs, inhibition is not observed. A 2000-fold excess of
EtOAc is required to decrease the reaction rate by a meager
10%. Additionally, we have UV-vis evidence to suggest that
the predominant manganese species in solution is the 4-coor-
dinate free metal corrole, not the 5-coordinate EtOAc adduct.
In benzene, titration of 1 with EtOAc does lead to a change in
the UV-vis spectrum, but the equilibrium constant of 76
calculated by the absorbance change with respect to ethyl acetate
concentration suggests that, in a typical 5 × 10-5 M solution
(24) (a) Dinc¸tu¨rk, S.; Jackson, R. A. J. Chem. Soc. Perkin Trans. 2. 1981, 1127.
(b) Walling, C.; Briggs, E. R.; Wolfstirn, K. B.; Mayo, F. R. J. Am. Chem.
Soc. 1948, 70, 1537. (c) Rybtchinski, B.; Vigalok, A.; Ben-David, Y.;
Milstein, D. J. Am. Chem. Soc. 1996, 118, 12406.
(25) Simonato, J.-P.; Pe´caut, J.; Scheidt, R.; Marchon, J.-C. Chem. Commun.
1999, 989.
9
16976 J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006

Aziridination with Managanese Corrole
A R T I C L E S
Scheme 4. Possible Mechanisms for Formation of 2
Table 1. Catalytic Aziridination of Olefinsa
d[(tpfc)MndNTs]
) kψ[(tpfc)Mn(EtOAc)] )
dt
k1k2[(tpfc)Mn(EtOAc)][ArIdNTs]
k-1[EtOAc] + k2[ArIdNTs]
(5)
(6)
K3k4[Mn]T[ArIdNTs]
1 + K3[ArIdNTs]
d[(tpfc)MndNTs]
dt
) kψ[Mn]T )
of 1‚EtOAc used in catalysis, 1 is 99.6% unligated. Additionally,
the UV-vis spectrum of crystalline 1‚EtOAc dissolved in
benzene is identical to the spectrum of 1 in which EtOAc has
been removed from the solid by sublimation. These results
provide strong evidence that 1 does not exist as a 5-coordinate
ethyl acetate adduct in benzene solution.
Given the above results, reversible formation of iodoimine
adduct (tpfc)MnIII(ArIdNTs), 3, followed by unimolecular
oxidation of Mn(III) to Mn(V) (Scheme 4B and eq 6) is the
favored mechanism. Fitting the experimental data in Figure 7
to the rate law based on the prior equilibrium approximation
(eq 6) yields k4 ) 0.26 ( 0.07 s-1 and an equilibrium constant
for formation of 3 K3 ) (10 ( 2) × 103 L mol-1 at 24.0 (
0.2 °C.
a Conditions: 2 equiv of solid ArINTs was stirred with styrene (20 mM)
and 2 mol % 1 (0.4 mM) in benzene for 48 h. b Yields were determined by
integration of methine and methylene aziridine 1H NMR signals against an
internal standard ((SiEt3)2O). c Values in parentheses represent isolated
yields. Isolated yields were lower due to overlapping of the aziridine and
ArI bands from the silica column.
It is worth noting that the reaction of 1 with ArIdNTs to
give complex 2 is not reversible. The addition of ArI (100 equiv)
to a solution of 2 does not result in reversion to 1 according to
Le Chatelier’s principle. This is in contrast to a reported
evidence that the reaction of iron porphyrin with PhIdO is
reversible.11b
Figure 8. 1H NMR spectral region showing aziridine methine and
methylene signals as well as TsNH2 N-H signals from in situ reactions.
Conversion to aziridine is optimized in benzene (bottom), while formation
of TsNH2 from HAT dominates in toluene (top).
yields were determined by 1H NMR spectroscopy. Even though
complex 1 is an effective catalyst for aryl olefins, it does not
catalyze aziridination of aliphatic alkenes such as 1-hexene and
cyclohexene or olefins with â-subtitution such as cis- and trans-
â-methyl styrene or cis- and trans-stilbene.
Catalytic Group Transfer to Olefins. The classical aziri-
dination reagent PhIdNTs was introduced as a nitrene source
for group transfer reactions to CsH and CdC bonds catalyzed
by heme enzymes and biomimetic metalloporphyrins.26 It has
also been employed with chiral transition metal catalysts in
asymmetric aziridination.1a,27 A major drawback for the over-
whelming majority of known catalysts is the requirement of
excess olefin (10- to 100-fold relative to PhIdNTs). For
example, iron corroles in the oxidation states +3 and +4
catalyze styrene aziridination with catalyst:PhIdNTs:styrene )
1:100:10 000.28a Recently, Zhang and co-workers have reported
first an iron and then a cobalt porphyrin complex, which have
achieved aziridination catalysis on a wide array of olefins with
a limiting olefin (olefin:nitrene ratio of 1:2); moderate (5-10
mol %) catalyst loadings are necessary to achieve reasonable
to good yields.28b,c 1 catalyzes the aziridination of styrene
substrates effectively in benzene with low catalyst loading (1-2
mol %) and without requiring excess olefin (Table 1). Aziridine
The reaction between ArIdNTs and 1 is much faster than
the reduction of 2 back to 1. Thus, under steady-state conditions
the major form of the catalyst is (tpfc)MnVdNTs, 2, and the
color of the reaction solution is red-brown throughout the
reaction until all substrates are consumed at which time the
catalyst reverts to the green Mn(III) (given that ArIdNTs is
not used in large excess). When toluene was used as solvent,
we noted lowered aziridine yields accompanied by substantial
TsNH2 byproduct formation. A comparison of the catalytic
reaction with styrene in the two solvents (benzene versus
toluene) is shown in Figure 8. We attribute this to hydrogen
atom transfer (HAT) from toluene. The ability of 2 to effect
hydrogen abstraction from benzylic hydrogens is demonstrated
by the HAT reaction of complex 2 with 9,10-dihydroanthracene
(DHA) to give anthracene. The reaction displays a primary
kinetic isotope effect kH/kD ) 3.5 for DHA versus deutero-DHA.
The details of HAT reaction with complex 2 will be taken up
in a future paper. However, we cannot exclude the possibility
that another form of manganese corrole that is present in small
concentration under catalytic conditions is more effective than
(26) (a) Breslow, R.; Gellman, S. H. J. Chem. Soc. Chem. Commun. 1982, 1400.
(b) Mansuy, D.; Mahy, J.-P.; Dureault, A.; Bedi, G.; Battoni, P. J. Chem.
Soc. Chem. Commun. 1984, 1161.
(27) Llewellyn, D. B.; Adamson, D; Arndtsen, B. A. Org. Lett. 2000, 2, 4165.
(28) (a) Simkhavich, L.; Gross, Z. Tetrahedron Lett. 2001, 42, 8089. (b) Vyas,
R; Gao, G.-Y.; Harden, J. D.; Zhang, P. Org. Lett. 2004, 6, 1907. (c) Gao,
G.-Y.; Harden, J. D.; Zhang, P. Org. Lett. 2005, 7, 3191.
9
J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006 16977

A R T I C L E S
Zdilla and Abu-Omar
by 14.7 g (75 mmol) of pentafluorobenzaldehyde. The reaction mixture
was heated at 60 °C open to atmosphere to afford a gradual color change
to brown. Once all solvent had evaporated, the mixture was heated for
an additional 4 h and then cooled at room temperature for 12 h. The
solid was dissolved in 100 mL of CH2Cl2 and filtered to remove
alumina; 8.52 g (37.5 mmol) of solid 2,3-dichloro-5,6-dicyano-1,4-
benzoquinone was added slowly (to avoid overboiling) with stirring to
afford a color change to black. This solution was stirred for 8 h and
separated on a silica column with 1:1 CH2Cl2:hexanes as eluant. The
location of product corrole on the column is apparent by its red
fluorescence under long-wave UV light. The solvent was removed from
the product-containing fraction in vacuo, and the product was separated
on a second column of alumina with CH2Cl2:hexanes as eluant. A 1:2
mixture was used to elute the initial fractions; a 1:1 mixture was then
substituted until the product began eluting, and a 2:1 mixture was used
to encourage elution of the remainder of the product. The solvent was
removed in vacuo, and the solid was recrystallized from CH2Cl2 and
hexanes to give 0.59 g of H3(tpfc) (0.72 mmol, 4.0%) as purple-gray
crystals.
complex 2 at HAT reactions, such as the “third oxidant” 4
suggested above for the aziridination reaction. Recently the
groups of Borovik and Holland have reported on HAT from
putative Fe(IV) imido intermediates.29
Conclusion
The manganese(V) tosylimide with a corrole auxiliary ligand
has been synthesized and spectroscopically characterized. The
reaction of (tpfc)MnIII(EtOAc) and ArIdNTs proceeds rapidly
on stopped-flow time scale, and the experimental rate law is in
agreement with a mechanism that involves iodoimine adduct
formation, (tpfc)MnIII(ArIdNTs), in route to the manganese-
(V) imido product, (tpfc)MnVdNTs. While (tpfc)Mn serves as
an effective catalyst for aziridination of styrene substrates
without requiring excess olefin, the imido complex (tpfc)MnVd
NTs is not the active intermediate. Labeling experiments
demonstrate unambiguously that the active group transfer
catalyst is an adduct of Mn(V), (tpfc)MnV(NTs)(ArIdNTs).
Mn(tpfc)EtOAc‚EtOAc (1) was prepared according to a modified
procedure of Gross et al.18a Before chromatographic separation, the dried
reaction mixture was heated gently under vacuum for 30 min to remove
all traces of DMF. The crystalline material obtained after chromatog-
raphy is suitable for use as a reagent or catalyst but can be further
purified by recrystallization from ethyl acetate/heptane (82% recovery).
UV-vis (C6H6): λmax [nm] (log ꢀ) ) 400 (4.64), 409 (4.67), 422 (4.60),
450 (4.28), 478 (4.21), 504 (4.02), 593 (4.16), 611 (4.16). Anal. Calcd
for C41H16F15MnN4O2‚C4H8O2: C, 52.75; H, 2.26; N, 5.47. Found: C,
52.71; H, 2.26; N, 5.53. MS (MALDI) m/z: 847.73 (theory ) 848.426
for Mn(tpfc)).
Experimental Section
General. All operations were carried out under inert atmosphere
using standard glovebox and Schlenk line techniques30 except where
otherwise noted. Benzene, toluene, and n-pentane were distilled from
sodium benzophenone ketyl; acetonitrile and pyrrole were distilled from
CaH2; CH2Cl2 was obtained from an Anhydrous Engineering solvent
purification system. Solvents were degassed and stored over activated
4-Å molecular sieves in a glovebox for 24 h prior to use. NMR solvents
were purchased from Cambridge Isotope Laboratories. Those used for
aziridination kinetics and for observation of 2 were stored over activated
4-Å molecular sieves for 24 h prior to use. TstBuNH2 was purchased
from Oakwood Products, Inc. Olefins were purchased from Aldrich
(X-styrene (XdH, 4-CF3, 4-Cl, 3-Cl), R-methylstyrene, cis- and trans-
stilbene, and trans-â-methylstyrene) and TCI (cis-â-methyl styrene).
All purchased chemicals were used as obtained from the manufacturers
unless specified otherwise.
2-(tert-Butylsulfonyl)(p-(tert-butylbenzene)sulfonyliminoiodo)-
benzene (ArINTstBu) was prepared according to a modified procedure
for the preparation of ArINTs.15 To a stirred yellow solution of ArI-
15
(OAc)2 (3.00 g, 6.74 mmol) in MeOH (50 mL), a solution of KOH
(1.5 g, 27 mmol) in MeOH (50 mL) was added, followed immediately
by a solution of TstBuNH2 (1.45 g, 6.80 mmol) in MeOH (50 mL) under
atmospheric conditions. The resulting cream-colored suspension was
stirred for 4 h, poured into 1 L of ice, and chilled at 4 °C for 10 h. The
mixture was filtered to give 0.922 g of cream-colored solid, and the
filtrate was saved. The residue was washed with 8 mL of C6H6 for 30
min and filtered to give 0.421 g of ArINTstBu. The initial filtrate was
evaporated in vacuo by 20 mL and chilled at 4 °C for 10 h to give
more cream-colored solid, which was filtered and washed with benzene
to give 0.559 g of more ArINTstBu. Total yield: 0.980 g, 28%. 1H NMR
UV-vis spectra were recorded on a Shimadzu UV-2501PC scanning
spectrophotometer. NMR spectra were obtained on Inova/Varian 300
MHz spectrometers. Gas chromatography was carried out on an Agilent
Technologies 6890N Network GC System with a J&W Scientific DB-5
capilary column. Oven temperature was ramped from 70 to 150 °C at
25 °C/min, then from 150 to 230 at 15 °Ì/min in constant flow mode
at 2.5 mL/min. Kinetics were performed using an Applied Photophysics
SX.18MV stopped-flow analyzer. Data were fit to theoretical curves
using KaleidaGraph.
(CD3CN) δ: 1.23 (s, 9H, Ts-tBu), 1.38 (s, 9H, SO2 Bu), 7.37 (dd, 2H,
t
Ts-m-H), 7.67 (m, 2H, Ar-H 3- and 4-position), 7.11 (dd, 2H, Ts-o-
H), 7.80 (ddd, 1H, Ar-H ortho to I), 7.89 (ddd, 1H, Ar-H ortho to
S). MS (ESI) m/z: 535.73 (theory ) 535.454 for [ArINHTstBu]+).
Synthesis: 2-(tert-Butylsulfonyl)iodobenzene (ArI) and 2-(tert-
butylsulfonyl)(p-toluenesulfonyliminoiodo)benzene (ArINTs) were
prepared according to the method of Macikenas et al.15
p-Toluenesulfonyliminoiodobenzene (PhINTs) was prepared ac-
cording to the method of Heuss et al.31
General Procedure for Catalytic Aziridination. To a stirred
suspension of ArINTs (580 mg, 1.18 mmol) in benzene (24 mL), a
green solution of 1 (10 mg, 0.0098 mmol) in benzene (1 mL) was added
to afford a color change to red-brown. Neat styrene (67.5 µL, 0.59
mmol) was added, and the reaction was stirred for 24 h. The reaction
mixture was opened to air, and the solvent was removed in vacuo. The
residue was separated on a silica column to isolate aziridine.
5,10,15-Tris(pentafluorophenyl)corrole ((tpfc)H3). A facile syn-
thesis of H3(tpfc) has been described by Gross et al.32 We have been
unable to reproduce the yields described and have therefore adopted
the following revised protocol to maximize product yield: to 15 g of
Brockmann I alumina with enough dry CH2Cl2 to cover the alumina,
5.04 g (75 mmol) of pyrrole was added with slow stirring, followed
N-(p-Tolylsulfonyl)-2-phenylaziridine. Aziridine was prepared as
describe above. Separation was achieved using 1:6 diethyl ether:benzene
as eluant. The fractions were collected in 1 mL increments and spotted
onto TLC plates. TLC analysis shows aziridine in a spot at Rf ) 0.69.
Quantitative isolation of aziridine was unsuccessful due to overlapping
of the aziridine and ArI (Rf ) 0.56) bands from the silica column. The
resulting azirdine solid was dissolved in a few drops of diethyl ether
and chilled to -20 °C. After 24 h, the supernatant was decanted to
give 103 mg of tan crystals (67%). 1H NMR (CDCl3) δ: 2.40 (d, 1H,
(29) HAT from a putative [FeIVdNR]: (a) Lucas, R. L.; Powell, D. R.; Borovik,
A. S. J. Am. Chem. Soc. 2005, 127, 11596. (b) Eckert, N. A.; Vaddadi, S.;
Stoian, S.; Lachicotte, R. J.; Cundari, T. R.; Holland, P. L. Angew. Chem.
Int. Ed. 2006, 45, 6868.
(30) Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-SensitiVe
Compounds, 2nd ed.; Wiley: New York, 1986; 326 pp.
(31) Heuss, B. D.; Mayer, M. F.; Dennis, S.; Hossain, M. Inorg. Chim. Acta
2003, 342, 301.
(32) Gross, Z.; Galili, N.; Simkhovich, L.; Saltsman, I; Botoshansky, M.; Bla¨ser,
D.; Boese, R.; Goldberg, I. Org. Lett. 1999, 1 (4), 599.
9
16978 J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006

Aziridination with Managanese Corrole
A R T I C L E S
CH2), 2.42 (s, 3H, CH3), 2.9 (d, 1H, CH2), 3.79 (dd, 1H, CH), 7.2-
7.36 (m, 7H, Ph, Ts-m-H), 7.88 (d, 2H, Ts-o-H). mp. 87-88.5 °C (lit.
88-89 °C).33
determined by gas chromatography: N-(p-tolylsulfonyl)-2-(p-trifluo-
romethylbenzene)aziridine ) 14.47 min, N-(p-tolylsulfonyl)-2-(p-
chlorobenzene)aziridine ) 10.95 min, N-(p-tolylsulfonyl)-2-(m-
chlorobenzene)aziridine ) 14.93 min. Competition kinetics were carried
out by stirring 1 mL of a 0.3 mM solution of 1 (0.3 µmol) with a 1:1
mixture of substituted olefin and styrene (20 µmol of each) and limiting
ArINTs (1.0 mg, 2 µmol). These were stirred for 24 h and analyzed by
gas chromatography. The integral of each substituted aziridine was
compared with that of phenyl aziridine (retention time ) 11.61 min).
All product ratios were approximately 1.
N-(p-(tert-Butylbenzenesulfonyl)-2-phenylaziridine. Aziridine was
prepared as described above. Separation was achieved using 1:3 ethyl
acetate:hexanes as eluant. The fractions were collected in 1 mL
increments and spotted onto TLC plates. TLC analysis shows aziridine
in a spot at Rf ) 0.55. Quantitative isolation of aziridine was
unsuccessful due to overlapping of the aziridine and ArI (Rf ) 0.41)
bands from the silica column. The resulting azirdine solid was
1
recrystallized from hot hexanes to give tan crystals (43%). H NMR
Double-Labeling Experiment To Determine the Nature of the
Oxidizing Species in Aziridination. A vial containing a mixture of 1
(2.1 mg, 2.0 µmol) and ArINTstBu (1.1 mg, 2.0 µmol) in benzene (1
mL) was stirred until the solution became red-brown (ca. 10 min). This
solution was added to a vial containing ArINTs (1.0 mg, 2.0 µmol)
and 1.0 mL of a 2.0 mM solution of styrene (2.0 µmol) in benzene. A
second reaction was set up in the same way, except that the order of
ArINTstBu and ArINTs was reversed. Both of these were stirred for 8
h and analyzed by GC. N-(p-(t-Butylbenzenesulfonyl)-2-phenylaziridine
and N-(p-tolylsulfonyl)-2-phenylaziridine were identified in the chro-
matogram by comparison to the known retention times, and the ratio
of products was determined by comparison of their respective peak
integrals.
t
(C6D6) δ: 0.97 (s, 9H, Bu), 1.81 (d, 1H, CH2), 2.73 (d, 1H, CH2),
3.77 (dd, 1H, CH), 6.96-7.21 (m, 7H, Ph, Ts-m-H), 7.96 (d, 2H, Ts-
o-H) mp. 103-106 °C. MS (CI) m/z: 316 ([PhCHCH2NHTstBu]+).
Kinetic Measurements: Kinetics of Formation of 2. Pseudo-first-
order kinetic measurements on the formation of 2 were carried out using
a stopped-flow analyzer. Equal volumes of solutions of 1 and ArINTs
were injected into the chamber and monitored at 600 nm. Reaction
profiles were fit to theoretical single-exponential functions to determine
the pseudo-first-order rate constants (kψ) in accordance with the follow-
ing relationship: Abst ) Abs∞ + (Abs0 - Abs∞) exp(-kψt). Stock
solutions were prepared by dissolving solid 1 and ArINTs in dry ben-
zene (dissolution of poorly soluble ArINTs was encouraged by heating).
The limiting solution of 1 (2.54 × 10-5 M) was tested with a range of
ArINTs concentrations (3.38 × 10-4-1.69 × 10-3 M). Because of the
limited solubility of ArINTs in benzene, this range could not be expand-
ed to larger concentrations of ArINTs. To widen the range, a further
dilution of ArINTs (1.69 × 10-4 M) was examined. For this measure-
ment, 1 was diluted to 1.27 × 10-5 M in order to maintain the 10-fold
ArINTs concentration excess necessary for pseudo-first-order kinetics.
Catalytic Aziridination of Olefins: Steady-State Kinetics. To a
stirred suspension of ArINTs (58 mg, 120 µmol) in benzene (2 mL), 1
(1.0 mg, 0.98 µmol) in benzene (0.5 mL) was added to afford a color
change to red-brown. Neat styrene (6.9 µL, 59 µmol) was added,
followed by (Et3Si)2O (5 µL) as internal standard. At regular intervals,
aliquots of 0.2 mL were removed from the reaction mixture and diluted
into 0.5 mL of benzene-d6 for NMR analysis. Styrene concentrations
were determined by comparison between the olefinic 1H peak integrals
and those of the internal standard.
Kinetics of Aziridination by Initial Rates Method. Three reactions
were prepared for kinetic examination by mixing solutions of ArINTs
and 1 with styrene in benzene. Benzophenone (11.4 mM) was used as
an internal standard for GC analysis. Initial concentration of ArINTs
was varied while all other species were held constant: [ArINTs] )
1.4-12 × 10-4, [styrene] ) 0.025, and [1] ) 2.0 × 10-5 M. The
catalytic reaction was initiated by addition of 1. After 5 min, 1 µL of
solution was drawn and analyzed by GC for the determination of
aziridine concentration.
Catalytic Aziridination of Olefins: Competition Kinetics. Aryl-
substituted aziridines were prepared in situ according to the general
protocol described above. Conversion to aziridine was confirmed by
1H NMR analysis, and retention time for each compound was
Spectroscopy on 2: Absorption spectroscopy on 2 was achieved
by stirring a solution of 1 mL of 1 (ca 10-3 M) with ArINTs or PhINTs
(ca. 1 mg) for 10 min or until the solution was red. The solution was
decanted from the solid and analyzed by absorption spectroscopy in a
0.10 mm path length cell. UV-vis (C6H6): λmax [nm] (log ꢀ) ) 360
(sh, 4.48), 404 (4.58), 530 (sh, 3.70).
Mass spectrometry was achieved by evaporation of a toluene
solution of 2 (generated from 1 mg of 1 and 2 mg of PhINTs). The
microcrystalline solid which formed on the side of the vial was scraped
away from the excess PhINTs at the bottom of the vial. The
microcrystalline solid was analyzed by ESI mass spectrometry. MS
(ESI) m/z: 861.89 (theory ) 862.433 for (tpfc)MnN).
Acknowledgment. The NSF is greatly acknowledged for
support of this work (CHE-0443147). We thank Dr. Phillip E.
Fanwick for his help with the X-ray structural determinations.
We also thank Dr. Jyotishman Dasgupta for his assistance in
interpretation of EPR spectra.
Supporting Information Available: Experimental procedures
for removal of EtOAc from 1 and corresponding IR spectra;
details of magnetic measurements; titration of 1 with EtOAc;
19F NMR and EPR spectroscopy on 2; 1H NMR and electronic
absorption spectroscopy on (tpfc)MnVdO; plot of initial rate
versus [ArINTs]; synthesis and EPR of (tpfc)CrVdNTs; and
X-ray structural data for (tpfc)Mn(EtOAc) (1) and (tpfc)Cr-
(NTs). This material is available free of charge via the Internet
at http://pubs.acs.org.
(33) Seden, T. P.; Turner, T. W. J. Chem. Soc. C 1968, 876.
JA0665489
9
J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006 16979