Probing the role of an FeIV tetrazene in catalytic aziridination

Article abstract of DOI:10.1039/c4cc05124f

An iron(IV) tetrazene complex has been synthesized that is an important addition to a previously proposed catalytic aziridination cycle. It provides two key insights about the aziridination: an iron(IV) imide is formed and this imide can bind an additional ligand in the cis position.

Full text of DOI:10.1039/c4cc05124f

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Probing the role of an FeIV tetrazene in catalytic
aziridination†
Cite this: Chem. Commun., 2014,
50, 13967
S. Alan Cramer,a Raul Hernandez Sanchez,b Desirae F. Brakhagea and
´
´
´
David M. Jenkins*a
Received 3rd July 2014,
Accepted 9th September 2014
DOI: 10.1039/c4cc05124f
www.rsc.org/chemcomm
An iron(IV) tetrazene complex has been synthesized that is an important
Tetrazenes are famous examples of non-innocent ligands
addition to a previously proposed catalytic aziridination cycle. It pro- that are usually formed by addition of an organic azide to a
vides two key insights about the aziridination: an iron(IV) imide is formed metal imide in a [2+3] cycloaddition.14–16 While Trogler noted
and this imide can bind an additional ligand in the cis position.
early examples ‘‘defy a simple description’’,17 they have since
been classified into three groups by Holland18 that allow for the
The synthesis of non-heme iron(IV) complexes has been of con- determination of bonding mode to the metal center. Well-
siderable interest due to their roles in biomimetic chemistry as characterized iron tetrazene complexes have recently been
well as catalytic group transfer reactions.1,2 Iron(IV) oxo complexes, prepared by Holland and Riordan,18,19 but the reactivity of iron
prepared by Que,3 Borovik,4 and Nam,5 have been shown to be tetrazene complexes has not been discussed previously. This
effective at O-atom transfer. Likewise, iron(IV) nitrides have been communication presents the first example of an iron(IV) tetrazene
considered as intermediates in the formation of ammonia from complex, which is also a component of a catalytic aziridination
dinitrogen.6,7 Finally, several iron(IV) imide complexes have been cycle. This novel complex is both fully characterized structurally
prepared and implemented for a variety of nitrene transfer and spectroscopically, and its reactivity provides insight into the
reactions.8,9 In all cases, even when the iron(IV) complex cannot aziridination catalytic cycle.
be isolated, these intermediates are often considered reactive
species in catalytic cycles.10,11
Addition of tolyl azide to 1 at 40 1C resulted in formation of
[(Me,EtTCPh)Fe(( p-tolyl)N4( p-tolyl))](PF6)2 (2), in a 74% isolated
When we previously reported an atom economical C2 + N1 crystalline yield. As previously discussed by Gade and Holland,18,20
aziridination with [(Me,EtTCPh)Fe(NCCH3)2](PF6)2 (1) as the catalyst, metallotetrazenes are generally formed by a [2+3] cycloaddition
we detected a transient imide species, [(Me,EtTCPh)FeQNAr]2+ of an additional equivalent of organic azide to the metal imide
(Ar = aryl) by electrospray ionization mass spectrometry (Scheme 1). Unlike the six-coordinate [(Me,EtTCH)Fe(O)(NCCH3)](OTf)2
(ESI-MS).12 Based on this result, and the later synthesis of (OTf = triflate),13 where S = 1, complex 2 was found to be diamagnetic
[(Me,EtTCH)Fe(O)(NCCH3)]2+, a terminal FeIV oxo, by Meyer13 (S = 0) at room temperature. 1H NMR exhibits diastereotopism at the
we believed that it would be possible to isolate a stable FeIV methylene and ethylene protons of the N-heterocyclic carbene (NHC)
imido to study the aziridination reaction in a step-wise manner, linkers that is consistent with several other rigid tetracarbene
which may lead to greater elucidation of the mechanism. complexes that we have synthesized.21–23 Notably, the 13C NMR
Despite considerable effort, we have been unable to isolate showed a resonance for the carbene carbon at 169.79 ppm that
[(Me,EtTCPh)FeQNAr]2+. Instead, even at relatively low tempera- is shifted significantly upfield (B26 ppm) from 1.12
tures, two equivalents of organic azide react with 1 to form the
The X-ray crystal structure of 2 confirms the formation of an
corresponding iron tetrazene, [(Me,EtTCPh)Fe(( p-tolyl)N4( p-tolyl))]- iron tetrazene complex (Fig. 1A). Notably, 2 adopts a trigonal
(PF6)2 (2) (Scheme 1).
prismatic geometry with a slight twisting distortion (fAve = 10.41)
as opposed to the octahedral geometry seen in 1 (trigonal
prismatic, f = 01; octahedral, f = 601).24 While there are several
examples of trigonal prismatic FeII complexes,25,26 there are only
a few FeIII examples,27 and to date no examples with FeIV have
been synthesized. The trigonal prismatic geometry is favorable
since this reduces steric interactions between the aryl rings on
the tetrazene and the aryl rings on the tetracarbene macrocycle.
The average Fe–NHC bond length is 1.98 Å, which is slightly
a Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996,
USA. E-mail: jenkins@ion.chem.utk.edu
b Department of Chemistry and Chemical Biology, Harvard University, Cambridge,
Massachusetts, 02138, USA
† Electronic supplementary information (ESI) available: Complete experimental
details, additional NMR figures, and X-ray crystallographic files. CCDC 1012071
and 1012072. For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c4cc05124f
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13967--13970 | 13967

View Article Online
Communication
ChemComm
Scheme 1 Reaction cycle of aziridination showing purported imide intermediate (center) and isolated tetrazene complex 2 (right).
ligand in a cis position which is well illustrated by the for-
mation of 2 (Fig. 1B). Second, the X-ray structure can help
deduce the oxidation state of the iron center via careful
evaluation of the N–N bond lengths of the tetrazene ligand.
The N10–N11 bond length is 1.292(3) Å which is shorter than the
N9–N10 or N11–N12 bond lengths (1.345(3) and 1.343(3), respectively)
suggesting that the central nitrogens have a double bond. This
double bond is indicative of a dianionic tetrazene ligand, which is
suggestive of an FeIV metal center. In contrast, a previously reported
FeII-tetrazene by Holland adopts a radical anion binding mode
where the radical is based at the tetrazene ligand, but the non-
innocent tetrazene ligand does adopt a dianionic binding mode
upon chemical reduction.18
In order to confirm the nature of 2 as a low spin (S = 0) FeIV
57
¨
complex, zero field Fe Mossbauer spectroscopy was collected
¨
on this species. Iron(IV) species investigated by Mossbauer
spectroscopy range in isomer shifts from as negative as ꢀ0.34
to as positive as 0.18 mm s .
ꢀ1 3,11,28 In particular, two examples
of FeIV complexes have been isolated and investigated with
polydentate carbene ligands, a nitride by Smith and an oxo by
Meyer.13,29 The spectrum of 2 (Fig. 1C) shows an isomer shift (d)
Fig. 1 (A) X-ray crystal structure of 2 from a top down view. (B) X-ray
crystal structure of 2 from a side view with selected parts of the macro-
cycle omitted for clarity. Orange, blue and grey ellipsoids (50% probability)
represent Fe, N and C, respectively. H atoms and solvent molecules have
been omitted for clarity. Selected bond lengths (Å): Fe–C1, 1.952(3); Fe–C2,
1.987(3); Fe–C3, 1.954(3); Fe–C4, 2.011(2); Fe–N9, 1.896(2); Fe–N12, 1.921(2),
of ꢀ0.01 mm sꢀ1 and quadrupole splitting (|DEQ|) of 0.62 mm sꢀ1
,
which is consistent with a low spin FeIV metal center. The
combination of diamagnetism observed in NMR (with 13C
¨
carbene shift), short N10–N11 bond distance, and Mossbauer
N9–N10, 1.345(3); N10–N11, 1.292(3); N11–N12, 1.343(3). (C) Mossbauer spectrum
¨
spectroscopy, demonstrates that the tetrazene ligand in 2 is a
dianionic ligand and the iron is in a tetravalent low spin state.
Based on both structural and spectroscopic evidence collected,
of 2: d, |DEQ| (mm sꢀ1): ꢀ0.01, 0.62.
shorter than the average bond length for 1, but is in line with we conclude that 2 is the first example of a trigonal prismatic
Meyer’s octahedral FeIV–oxo tetracarbene complex.13
FeIV complex.
A close examination of the X-ray crystal structure demon-
Because tetrazene ligands are often non-innocent, we inter-
strates two notable points. First, the crystal structure provides rogated whether the redox behavior of 2 was metal or ligand based.
a hint as to a possible mechanism for aziridination from the Cyclic voltammetry shows a single reversible reduction wave
Fe-imide intermediate. Since the alkene must be present near centered at ꢀ1.05 V versus ferrocene in an acetonitrile solution.
the metal imide prior to group transfer, the tetracarbene ligand Complex 2 was reduced with one equivalent of cobaltocene yield-
then must be flexible enough to accommodate an additional ing [(Me,EtTCPh)Fe((p-tolyl)N4(p-tolyl))](PF6) (3). The X-ray structure
13968 | Chem. Commun., 2014, 50, 13967--13970
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
complexes,30 not the subsequent nitrene coupling. Hillhouse
and Cundari implicated a metallotetrazene as the intermediary
step (based on DFT calculations) in the conversion of a Ni(II)
imido to form diazene derivatives, although they did not report
any experimental evidence for the tetrazene intermediate.32
Although several catalysts have been successful in the conver-
sion of organic azides to diazene, including a recent example by
Peters, none of these examples have been proposed to proceed
through a tetrazene intermediate.33–35 Heating 2 at 85 1C in
CD3CN formed 1 and 4,40-dimethylazobenzene after 24 hours
(Scheme 1). This result is significant because it shows that 1
can be regenerated below the temperature at which we perform
aziridination catalysis. In addition, we believe that this is the
first well-characterized step-wise formation of an azobenzene
from a metallotetrazene.
Since we could reform 1, we believed that 2 would also be a
competent catalyst for aziridination. Addition of cyclooctene and p-
tolyl azide in a 20 : 1 ratio to 2 (at 0.2% catalyst loading) gave a 77%
yield of 9-(p-tolyl)-9-azabicyclo[6.1.0]nonane by NMR integration
with an internal standard (less than 5% 4,40-dimethylazobenzene
was formed during the catalysis). These results encouraged us to
inquire whether direct group transfer could occur from the tetra-
zene ligand on 2. Heating 2 in neat cyclooctene at 90 1C formed
a mixture of both 4,40-dimethylazobenzene and 9-( p-tolyl)-9-
azabicyclo[6.1.0]nonane in a 2.3 : 1 ratio respectively, based on
NMR integration (Fig. S2, ESI†). Since multiple products are formed
from 2, it appears that multiple reaction mechanisms occur
simultaneously at this temperature. To our knowledge, this is the
first example of formation of an aziridine from a metallotetrazene.
In conclusion, we have synthesized the first iron(IV) tetra-
Fig. 2 (A) X-ray crystal structure of 3 from a top down view. Orange, blue
and grey ellipsoids (50% probability) represent Fe, N and C, respectively.
H atoms, counteranions, and solvent molecules have been removed for
clarity. Selected bond lengths (Å): Fe–C1, 1.963(3); Fe–C2, 1.963(3); Fe–C3,
1.976(3); Fe–C4, 1.958(3); Fe–N9, 1.947(2); Fe–N12, 1.980(2), N9–N10
,
1.368(3); N10–N11, 1.272(3); N11–N12, 1.377(3). (B) Mossbauer spectrum of
¨
3: d, |DEQ| (mm sꢀ1): 0.10, 1.13. (C) Frozen solution EPR of 3 at 77 K: gJ
2.094(1), g> = 1.997(1).
=
of 3 reveals the same general geometry about the metal center
(Fig. 2A) except the dihedral angle (f) has increased to 22.11. zene complex, [(Me,EtTCPh)Fe(( p-tolyl)N4( p-tolyl))](PF6)2, which
The central N–N bond distance in 3 is significantly shorter (about has been found to be an important component of our catalytic
0.10 Å) than the peripheral N–N bond distances suggesting that cycle for aziridination. The complex is low spin and in a
the tetrazene ligand remained dianionic.
trigonal prismatic geometry and based on spectroscopic and
The spectroscopic data for 3 also shows that the tetrazene is structural data the tetrazene binds in the dianionic mode. The
in the dianionic binding mode in the reduced state. The X-ray structure demonstrates a key point that is salient to this
¨
Mossbauer spectrum of 3 is shifted to a more positive isomer unique catalytic cycle, namely that the macrocyclic tetracarbene
shift relative to its FeIV congener, consistent with an FeIII state is flexible enough to allow an alkene to bind in a cis position
(Fig. 2B).13,30 More tellingly, the EPR spectra of 3 shows an axial prior to loss of aziridine. The iron(IV) tetrazene complex will
pattern with gJ = 2.094(1) and g> = 1.997(1) consistent with a release 4,40-dimethylazobenzene allowing for the regeneration
low spin (S = 1/2) FeIII (Fig. 2C).31 This EPR spectrum is distinct of the original catalyst and is itself a competent starting point
from Wagner’s trigonal prismatic FeIII complexes,27 which are for catalytic aziridination. Perhaps more remarkably, the tetra-
high spin, as well as the other iron tetrazenes of Riordan and zene will yield 9-( p-tolyl)-9-azabicyclo[6.1.0]nonane (as well as
Holland.18,19 Unlike their results with iron tetrazenes, we have 4,40-dimethylazobenzene) in the presence of excess cyclooctene
found the ligand behaves in a fully ‘‘innocent’’ manner regard- which has not been previously demonstrated. These combined
less of Fe oxidation state.
results suggest that an iron(IV) imide is formed during catalysis.
S.A.C. and D.M.J. are grateful to the NSF (CAREER grant
Due to 1 being an excellent aziridination catalyst and since 2
is formed by addition of excess organic azide to 1, the question CHE-1254536) for financial support. R.H.S. is grateful to CON-
´
arose as to whether 2 was a competent catalyst for aziridination ACYT (Consejo Nacional de Ciencia y Tecnologıa) and Funda-
´
´
or whether it was a poison to the catalytic cycle. We performed cion Mexico for a doctoral fellowship.
both stoichiometric and catalytic tests with 2 to probe its role in
the aziridination catalytic cycle. Our initial reaction was to try to
reform 1 by removing the tetrazene ligand by thermal decom-
position (Fig. S1, ESI†). Previous thermolysis reactions with
metallotetrazenes have noted the extrusion of N2,17 yet the only
noted by-product of such reactions were the respective diimide
Notes and references
1 M. M. Abu-Omar, A. Loaiza and N. Hontzeas, Chem. Rev., 2005, 105,
2227–2252.
2 M. Costas, M. P. Mehn, M. P. Jensen and L. Que, Jr., Chem. Rev.,
2004, 104, 939–986.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13967--13970 | 13969

View Article Online
Communication
ChemComm
3 J.-U. Rohde, J.-H. In, M. H. Lim, W. W. Brennessel, M. R. Bukowski, 19 M. T. Mock, C. V. Popescu, G. P. A. Yap, W. G. Dougherty and C. G.
A. Stubna, E. Muenck, W. Nam and L. Que, Jr., Science, 2003, 299,
1037–1039.
4 D. C. Lacy, R. Gupta, K. L. Stone, J. Greaves, J. W. Ziller, M. P.
Riordan, Inorg. Chem., 2008, 47, 1889–1891.
20 T. Gehrmann, J. Lloret Fillol, H. Wadepohl and L. H. Gade,
Organometallics, 2012, 31, 4504–4515.
Hendrich and A. S. Borovik, J. Am. Chem. Soc., 2010, 132, 21 Z. Lu, S. A. Cramer and D. M. Jenkins, Chem. Sci., 2012, 3, 3081–3087.
12188–12190.
22 H. M. Bass, S. A. Cramer, A. S. McCullough, K. J. Bernstein, C. R.
Murdock and D. M. Jenkins, Organometallics, 2013, 32, 2160–2167.
23 H. M. Bass, S. A. Cramer, J. L. Price and D. M. Jenkins, Organometallics,
2010, 29, 3235–3238.
24 S. Banerjee, A. Ghosh, B. Wu, P.-G. Lassahn and C. Janiak, Polyhedron,
2005, 24, 593–599.
5 S. Fukuzumi, H. Kotani, Y.-M. Lee and W. Nam, J. Am. Chem. Soc.,
2008, 130, 15134–15142.
6 J. J. Scepaniak, C. S. Vogel, M. M. Khusniyarov, F. W. Heinemann,
K. Meyer and J. M. Smith, Science, 2011, 331, 1049–1052.
7 T. A. Betley and J. C. Peters, J. Am. Chem. Soc., 2004, 126, 6252–6254.
8 I. Nieto, F. Ding, R. P. Bontchev, H. Wang and J. M. Smith, J. Am. 25 R. S. Srivastava, M. A. Khan and K. M. Nicholas, J. Am. Chem. Soc.,
Chem. Soc., 2008, 130, 2716–2717. 1996, 118, 3311–3312.
9 E. R. King, E. T. Hennessy and T. A. Betley, J. Am. Chem. Soc., 2011, 26 T. Owen, F. Grandjean, G. J. Long, K. V. Domasevitch and
133, 4917–4923. N. Gerasimchuk, Inorg. Chem., 2008, 47, 8704–8713.
10 J. England, Y. Guo, K. M. Van Heuvelen, M. A. Cranswick, G. T. 27 J. Wagner, H. Goerls and H. Keutel, Inorg. Chim. Acta, 2005, 358,
Rohde, E. L. Bominaar, E. Munck and L. Que, J. Am. Chem. Soc.,
2011, 133, 11880–11883.
11 J. Kaizer, E. J. Klinker, N. Y. Oh, J.-U. Rohde, W. J. Song, A. Stubna,
J. Kim, E. Muenck, W. Nam and L. Que, Jr., J. Am. Chem. Soc., 2004,
126, 472–473.
12 S. A. Cramer and D. M. Jenkins, J. Am. Chem. Soc., 2011, 133,
19342–19345.
13 S. Meyer, I. Klawitter, S. Demeshko, E. Bill and F. Meyer, Angew.
Chem., Int. Ed., 2013, 52, 901–905.
14 K. E. Meyer, P. J. Walsh and R. G. Bergman, J. Am. Chem. Soc., 1995,
117, 974–985.
15 A. F. Heyduk, K. J. Blackmore, N. A. Ketterer and J. W. Ziller, Inorg. Chem.,
2005, 44, 468–470.
16 R. I. Michelman, R. G. Bergman and R. A. Andersen, Organometallics,
1993, 12, 2741–2751.
17 W. C. Trogler, Acc. Chem. Res., 1990, 23, 426–431.
808–813.
28 J. England, M. Martinho, E. R. Farquhar, J. R. Frisch, E. L. Bominaar,
E. Munck and L. Que, Jr., Angew. Chem., Int. Ed., 2009, 48,
3622–3626.
29 C. Vogel, F. W. Heinemann, J. Sutter, C. Anthon and K. Meyer,
Angew. Chem., Int. Ed., 2008, 47, 2681–2684.
30 K.-M. Lin, P.-Y. Wang, Y.-J. Shieh, H.-Z. Chen, T.-S. Kuo and Y.-C.
Tsai, New J. Chem., 2010, 34, 1737–1745.
31 K. Meyer, E. Bill, B. Mienert, T. Weyhermueller and K. Wieghardt,
J. Am. Chem. Soc., 1999, 121, 4859–4876.
32 N. D. Harrold, R. Waterman, G. L. Hillhouse and T. R. Cundari,
J. Am. Chem. Soc., 2009, 131, 12872–12873.
33 A. Takaoka, M.-E. Moret and J. C. Peters, J. Am. Chem. Soc., 2012,
134, 6695–6706.
34 N. P. Mankad, P. Mueller and J. C. Peters, J. Am. Chem. Soc., 2010,
132, 4083–4085.
18 R. E. Cowley, E. Bill, F. Neese, W. W. Brennessel and P. L. Holland, 35 R. A. Zarkesh, J. W. Ziller and A. F. Heyduk, Angew. Chem., Int. Ed.,
Inorg. Chem., 2009, 48, 4828–4836.
2008, 47, 4715–4718.
13970 | Chem. Commun., 2014, 50, 13967--13970
This journal is ©The Royal Society of Chemistry 2014