On the mechanism of the reaction of organic azides with transition metals: Evidence for triplet nitrene capture

Article abstract of DOI:10.1002/anie.200462647

(Chemical Equation Presented) High-valent imido complexes are obtained by reaction of the manganese-(III) corrole [Mn(tpfc)] with aryl azides by a novel reaction mechanism in which singlet nitrene undergoes intersystem crossing to the triplet state, which

Full text of DOI:10.1002/anie.200462647

Angewandte
Chemie
Imido Complexes
DOI: 10.1002/anie.200462647
On the Mechanism of the Reaction of Organic
Azides with Transition Metals: Evidence for
Triplet Nitrene Capture**
Mahdi M. Abu-Omar,* Catherine E. Shields,
Nicola Y. Edwards, and Rebecca A. Eikey
Organic azides (RN3) have received extensive attention
recently because of their central role in several “spring-
loaded” reactions that have been dubbed “click chemis-
try”.[1–4] In organometallic chemistry, azides have been used
[*] Prof. Dr. M. M. Abu-Omar, C. E. Shields
Department of Chemistry
Purdue University
560 Oval Drive
West Lafayette, IN 47907 (USA)
Fax: (+1)765-494-0238
E-mail: mabuomar@purdue.edu
N. Y. Edwards, Dr. R. A. Eikey
Department of Chemistry and Biochemistry
University of California
Los Angeles, CA 90095 (USA)
[**] This work was supported by the Department of Chemistry at Purdue
University, the National Science Foundation, and the Beckman
Foundation. We are grateful to Professors Matthew Platz and
Kendall Houk for helpful discussions.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2005, 44, 6203 –6207
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6203

Communications
extensively for the preparation of imido
=
(M NR) complexes, in which dinitro-
gen is released as byproduct.[5,6] As part
of our interest in developing nitrene
(NR) transfer to organic substrates by
transition-metal catalysis, we have used
aryl azides, which are quite stable and
easy to make,[7,8] to prepare mononu-
clear terminal imido complexes of man-
ganese(v) and chromium(v) cor-
roles.[9,10] Despite many known reac-
tions of organic azides with transition
metals, little is known about the reac-
tion mechanism. Organoazido metal
complexes have been proposed to ini-
tially form for coordinatively unsaturated metal com-
plexes.[11–14] However, only in two instances have such
complexes been characterized and shown to proceed to the
imido product upon heating or photolysis, presumably via a
four-membered metallacycle.[15–17] Herein we report a novel
mechanism for the reaction of aryl azides with manganese
corroles in which the metal does not form an organoazido
complex.
Encouraged by the successful synthesis of terminal imido
manganese(v) complexes from mesityl azide and trichloro-
phenyl azide,[9] we explored the utility and generality of this
reaction (Scheme 1). We noted that ortho substitution is
mandatory, as is thermal or photochemical activation. Phenyl
azide with substituents in the para position as well as the
electron-withdrawing sulfonyl azide did not produce imido
metal complexes.
Results presented herein and the scope of the reaction
(Scheme 1) gave valuable insight into the mechanism of the
reaction of organic azides with transition-metal corroles. Our
mechanistic proposal is presented in Scheme 2. On thermal or
photochemical decomposition of aryl azide, singlet nitrene is
formed and dinitrogen extruded.[18,19] The singlet nitrene then
decomposes along two pathways. The first is cyclization to
azepine B via azirine A as a steady-state intermediate.[20] The
Scheme 2. Mechanism of the reaction of organic azides with [MnIII(tpfc)].
ultimate fate of the azepine is polymerization.[19] However, in
highly dilute solutions, polymerization of B is suppressed and
it reverts to singlet nitrene, which relaxes by intersystem
crossing (isc) to the lower-energy triplet state.[21] The final
thermodynamic product is azo compound C, assumed to form
by dimerization of triplet nitrene.[22] This second pathway, isc
of singlet nitrene to triplet, is responsible for the formation of
the imido manganese complex in our system and is most
consistent with the observed scope of the reaction (Scheme 1)
and kinetic data, as detailed below.
Ortho substituents are mandatory for the formation of
imido metal complexes in this system. This is in agreement
with results showing that singly ortho-substituted singlet aryl
[23]
nitrenes cyclize away fromthe substituent,
and ortho
methylation increases the rate of intersystem crossing.[21]
Furthermore, the lack of reaction with fluorinated phenyl
azides and sulfonyl azide is consistent with these substituents
destabilizing the open-shell singlet nitrene.[24] Indeed, under
our reaction conditions only starting material was recovered
with these substrates (Scheme 1).
The kinetics of imido metal formation were studied in
detail for the reaction of mesityl azide (MesN3) with [MnIII-
(tpfc)]. Changes in the UV/Vis spectra for a typical reaction
are displayed in Figure 1. The disappearance of the bands at
Scheme 1. Scope of the reaction of aryl azides with transition-metal corroles.
6204 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6203 –6207

Angewandte
Chemie
Figure 1. UV/Vis spectral changes for the reaction of [MnIII(tpfc)]
(15 mm) with mesityl azide (2.1 mm) in toluene at 808C recorded every
6 min.
479 and 609 nmand appearance of the band at 536 nmare
III
indicative of a change in the oxidation state fromMn
to
MnV. The tight isosbestic points at 510 and 565 nmdemon-
strate clean conversion without significant accumulation of an
intermediate.
The progress of the reaction was monitored at 536 nm
with [MnIII(tpfc)] limiting and MesN3 in excess. The exper-
imental reaction rate was found to be first-order in mesityl
azide (Figure 2) and exhibited a complex dependence on the
concentration of the manganese complex. A simplified rate
law that includes the reaction steps described in Scheme 2 is
given in Equation (1).
Figure 2. A) Time profiles for the reaction of [Mn(tpfc)] (15 mm) and
mesityl azide (3, 5, and 12 mm) in toluene at (80.0ꢁ0.2)8C. The linear
fit is over the initial 30% of the reaction to illustrate zeroth-order kinet-
ics that deviate from linearity in the later stages of the reaction. B) A
plot of the initial rate Vi versus the mesityl azide concentration show-
ing first-order dependence on the azide concentration.
dc
½MnVimideŠ
k1kPc
cMesN
3
¼
ð1Þ
dt
kPc
½Mnðtpfcފ
½MnðtþpfcÞkŠ dc3NMes
Under conditions where kPc[Mn(tpfc)] @ kdc3NMes, the rate law
simplifies to dc /dt = k1cMesN and the rate is zeroth-order in
(k1) was determined experimentally herein (Figure 2B); the
rate constant for intersystemcrossing ( kisc) was taken from
published data,[27] and the rate constants for triplet nitrene
dimerization to give the azo compound (kd) and for polymer-
ization of azepine (k2) were set to the diffusion limit (8.0 ”
109 mꢀ1 sꢀ1). The remaining three rate constants were floated
to give the best fit to the experimental profiles. The agree-
ment between experimental and simulated kinetics is illus-
trated in Figure 3B and the Supporting Information. It is
worth noting that formation of the azo compound sets in after
the initial 30% of reaction, which agrees with the observed
kinetic behavior (see above). Nevertheless, GC-MS analyses
of reaction mixtures at the end of the reaction show only
azide, and only at longer decomposition times is the azo
compound detected (see Supporting Information). This
apparent contradiction is most likely due to low concentra-
tions of the azo compound (below the GC-MS detection
threshold) on the reaction timescale (2 h).
=
[MnV NR]
3
the concentration of the manganese corrole. However, as the
reaction proceeds and [Mn(tpfc)] is consumed, the above
assumption is no longer valid and the kinetics divert from
zeroth order. This expectation is consistent with the time
profiles shown in Figure 2A, which are linear over the initial
30% of the reaction (ca. 0.02DOD) and deviate fromzeroth
order in the later stages. The experimental time profiles do
not fit an exponential pseudo-first-order equation (see
Supporting Information). Furthermore, under no circum-
stance would the reaction rate be independent of the metal
concentration if an organoazido adduct was formed initially
or the metal corrole reacted directly with the organic azide.
Similar kinetic behavior was observed when reaction progress
was followed at 479 nm, which corresponds to the disappear-
ance of the MnIII reactant (see Supporting Information).
The observed kinetics were successfully modeled with the
kinetic simulation program KINSIM.[25,26] Figure 3A shows
the simulated time profiles for [MnIII(tpfc)], [MnV(tpfc)], azo
compound, azepine, and polymer arising from the azepine.
The rate constants used in association with Scheme 2 as the
reaction mechanism are given in the caption of Figure 3. The
rate constant for the thermal decomposition of mesityl azide
To date, the reaction of organic azides with transition-
metal complexes to give imido metal compounds has been
accepted to proceed via an organoazido metal intermediate.
While this mechanism holds true in many instances and has
been demonstrated in a couple of reaction systems, it is not
exclusive.[15–17] We have described herein a new mechanism
Angew. Chem. Int. Ed. 2005, 44, 6203 –6207
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6205

Communications
molecular sieves and distilled over CaH2 prior to use. When required,
solvents were deoxygenated by freeze/pump/thaw cycles. Deuterated
solvents were purchased fromCambridge Isotopes and used without
further purification. All other solvents and reagents were of reagent
grade and used as received. The syntheses of H3(tpfc),[28,29] [Mn-
(tpfc)],[30] [Cr(tpfc)(py2)],[31] [Mn(tpfc)(NMes)][9] (Mes = 2,4,6-
(CH3)3C6H2), [Mn(tpfc)(NAr)][9] (Ar= 2,4,6-Cl3C6H2), [Mn(tpfc)-
(NAr’)][10] (Ar’ = 2,6-Cl2C6H3), [Cr(tpfc)(NMes)],[10] and [Cr(tpfc)-
(NAr)][10] followed previously described methods. All azides were
prepared by using the Sandmeyer reaction or slight variations
thereof.[7,8] Azo compounds were compared to authentic samples.
Photolysis experiments were carried out with a Hanovia Model 673A-
0360 550 W medium-pressure mercury arc lamp. Rate measurements
were carried out by conventional UV/Vis spectroscopy on a Shimadzu
UV-2501 spectrophotometer equipped with a temperature-controlled
cell holder. Formation of [MnV(NMes)(tpfc)] was monitored at
536 nm, c[Mn(tpfc)] = 15 or 30 mm, and cMesN = 2.0–12.0 mm at (80 ꢁ
3
0.2)8C.
Received: November 17, 2004
Revised: June 13, 2005
Published online: August 29, 2005
Keywords: azides · imido ligands · kinetics ·
.
macrocyclic ligands · manganese
[1] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056; Angew. Chem. Int. Ed. 2001, 40, 2004.
[2] R. Huisgen, J. Sauer, H. J. Sturm, J. H. Markgraf, Chem. Ber.
1960, 93, 2106.
[3] Z. P. Demko, K. B. Sharpless, Angew. Chem. 2002, 114, 2117;
Angew. Chem. Int. Ed. 2002, 41, 2113
[4] Z. P. Demko, K. B. Sharpless, Angew. Chem. 2002, 114, 2114;
Angew. Chem. Int. Ed. 2002, 41, 2110.
Figure 3. A) Time profiles simulated by the program KINSIM for the
reaction of [MnIII(tpfc)] with mesityl azide according to Scheme 2. Sim-
ulation conditions: c[MnIII(tpfc)] =15 mm, cMesN =5.3 mm,
3
k1 =8.3ꢀ10ꢀ7 sꢀ1, kisc =2.0ꢀ107 sꢀ1 [27] kr =1.0ꢀ1010 sꢀ1
,
,
[5] R. A. Eikey, M. M. Abu-Omar, Coord. Chem. Rev. 2003, 243, 83.
[6] D. E. Wigley, Prog. Inorg. Chem. 1994, 42, 239.
[7] S. Murata, S. Abe, H. Tomioka, J. Org. Chem. 1997, 62, 3055.
[8] E. Leyva, D. Munoz, M. S. Platz, J. Org. Chem. 1989, 54, 5983.
[9] R. A. Eikey, S. I. Khan, M. M. Abu-Omar, Angew. Chem. 2002,
114, 3743; Angew. Chem. Int. Ed. 2002, 41, 3591.
[10] N. Y. Edwards, R. A. Eikey, M. I. Loring, M. M. Abu-Omar,
Inorg. Chem. 2005, 44, 3700 – 3708.
[11] G. LaMonica, S. Cenini, J. Chem. Soc. Dalton Trans. 1980, 1145.
[12] G. L. Hillhouse, J. E. Bercaw, Organometallics 1982, 1, 1025.
[13] J. H. Osborne, A. L. Rheingold, W. C. Trogler, J. Am. Chem. Soc.
1985, 107, 7945.
[14] D. M. Antonelli, W. P. Scaefer, G. Parkin, J. E. Bercaw, J.
Organomet. Chem. 1993, 462, 213.
[15] G. Proulx, R. G. Bergman, J. Am. Chem. Soc. 1995, 117, 6382.
[16] G. Proulx, R. G. Bergman, Organometallics 1996, 15, 684.
[17] M. G. Fickes, W. M. Davis, C. C. Cummins, J. Am. Chem. Soc.
1995, 117, 6384.
kꢀr =1.0ꢀ107 sꢀ1, k2 =kd =8.0ꢀ109 mꢀ1 sꢀ1, kp =2.0ꢀ106 mꢀ1 sꢀ1
.
B) Simulated (gray) and experimental (black) data for the formation of
[MnV(NMes)(tpfc)] from the reaction of 15 mm [Mn(tpfc)] and 5 mm
mesityl azide.
for the reaction of a manganese corrole with aryl azides to
give imido metal complexes. In this mechanism, the tran-
sition-metal complex captures a triplet nitrene formed by
thermal or photochemical activation of the organic azide. The
compelling evidence presented for this mechanism is 1) com-
plex dependence of the kinetics on metal concentration with
zeroth order over 30% of the reaction), 2) the kinetics have
been modeled successfully for the reaction mechanism
described in Scheme 2, and 3) substituents in both ortho
positions (Me or Cl, but not F) are required. Contrary to
conventional wisdom that highly reducing metal complexes
are needed for a reaction with organic azides, we have
demonstrated that effective imido metal formation via
nitrene capture does not require a highly reducing center.
Investigations into harnessing this mechanistic paradigm for
selective nitrene transfer under thermal or photochemical
catalysis are in progress.
[18] S. Patai, Z. Rappoport in The Chemistry of Azides (Eds.: S. Patai,
Z. Rappoport), Wiley-Interscience, New York, 1995.
[19] W. T. Borden, N. P. Gristan, C. M. Hadad, W. L. Karney, C. R.
Keminitz, M. S. Platz, Acc. Chem. Res. 2000, 33, 765.
[20] N. P. Gritsan, Z. Zhu, C. M. Hadad, M. S. Platz, J. Am. Chem.
Soc. 1999, 121, 1202.
[21] N. P. Gristan, D. Tigelaar, M. S. Platz, J. Phys. Chem. A 1999, 103,
4465.
[22] E. Leyva, M. S. Platz, G. Persy, J. Wirz, J. Am. Chem. Soc. 1986,
108, 3783.
Experimental Section
Compounds were prepared and handled by standard vacuum-line and
glove-box techniques. Toluene and acetonitrile were predried over
[23] R. J. Sundberg, S. R. Sutar, M. Brenner, J. Am. Chem. Soc. 1972,
94, 513.
[24] W. L. Karney, W. T. Borden, J. Am. Chem. Soc. 1997, 119, 3347.
6206 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6203 –6207

Angewandte
Chemie
[25] B. A. Barshop, R. F. Wrenn, C. Frieden, Anal. Biochem. 1983,
130, 134.
[26] C. T. Zimmerle, C. Frieden, Biochem. J. 1989, 258, 381.
[27] N. P. Gritsan, A. D. Gudmundsdottir, D. Tigelaar, M. S. Platz, J.
Phys. Chem. A 1999, 103, 3458.
[28] Z. Gross, N. Galili, I. Saltsman, Angew. Chem. 1999, 111, 1530;
Angew. Chem. Int. Ed. 1999, 38, 1427.
[29] Z. Gross, N. Galili, L. Simkhovich, I. Saltsman, M. Botoshansky,
D. Blaser, R. Boese, I. Goldberg, Org. Lett. 1999, 1, 599.
[30] Z. Gross, G. Golubkov, L. Simkhovich, Angew. Chem. 2000, 112,
4211; Angew. Chem. Int. Ed. 2000, 39, 4045.
[31] A. E. Meirer-Callahan, A. J. DiBilio, L. Simkhovich, A. Maham-
med, I. Goldberg, H. B. Gray, Z. Gross, Inorg. Chem. 2001, 40,
6788.
Angew. Chem. Int. Ed. 2005, 44, 6203 –6207
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6207