Oxidative group transfer to a triiron complex to form a nucleophilic μ3-nitride, [Fe3(μ3-N)]-

Article abstract of DOI:10.1021/ja2003445

Utilizing a hexadentate ligand platform, a high-spin trinuclear iron complex of the type (^tbsL)Fe₃(thf) was synthesized and characterized ([^tbsL]^6- = [1,3,5-C₆H ₉(NPh-o-NSi^tBuMe<

Full text of DOI:10.1021/ja2003445

COMMUNICATION
pubs.acs.org/JACS
Oxidative Group Transfer to a Triiron Complex to Form a Nucleophilic
μ3-Nitride, [Fe3(μ3-N)]-
Tamara M. Powers, Alison R. Fout, Shao-Liang Zheng, and Theodore A. Betley*
Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street, Cambridge, Massachusetts 02138, United States
S Supporting Information
b
Scheme 1
ABSTRACT: Utilizing a hexadentate ligand platform, a
high-spin trinuclear iron complex of the type (tbsL)Fe3(thf)
was synthesized and characterized ([tbsL]6- = [1,3,5-C6H9-
(NPh-o-NSitBuMe2)3]6-). The silyl-amide groups only
permit ligation of one solvent molecule to the tri-iron core,
resulting in an asymmetric core wherein each iron ion
exhibits a distinct local coordination environment. The
triiron complex (tbsL)Fe3(thf) rapidly consumes inorganic
azide ([N3]NBu4) to afford an anionic, trinuclear nitride
complex [(tbsL)Fe3(μ3-N)]NBu4. The nearly C3-symmetric
complex exhibits a highly pyramidalized nitride ligand that
resides 1.205(3) Å above the mean triiron plane with short
Fe-N (1.871(3) Å) distances and Fe-Fe separation
(2.480(1) Å). The nucleophilic nitride can be readily
alkylated via reaction with methyl iodide to afford the
neutral, trinuclear methylimide complex (tbsL)Fe3(μ3-NCH3).
Alkylation of the nitride maintains the approximate C3-
symmetry in the imide complex, where the imide ligand
resides 1.265(9) Å above the mean triiron plane featuring
lengthened Fe-Nimide bond distances (1.892(3) Å) with
nearly equal Fe-Fe separation (2.483(1) Å).
reaction chemistry of a triiron scaffold, we targeted a ligand plat-
form that would restrict exogenous ligands from binding to the
n our lab we are currently investigating the synthesis and
metal ions. Following a similar protocol for the preparation of
I
reactivity of polynuclear metal complexes. Recent interest in
tame-based scaffolds (tame = 1,1,1-tris(aminomethyl)ethane),4
the tris-amine R-R-R-1,3,5-tris-aminocyclohexane hydrobromic
such systems stems from their potential to cooperatively bind
and activate small molecules, utilizing the expanded redox
reservoir afforded by the polynuclear core.1 This design strategy
may prove effective as functional surrogates for polynuclear
reaction sites found in heterogeneous catalysts (e.g., Fe(111)
face in Haber-Bosch dinitrogen reduction2) or in metalloen-
zyme cofactors (e.g., FeMo-cofactor in nitrogenase3). The poly-
nuclear reaction sites in the cited examples likely facilitate bind-
ing, activation, and chemical breakdown of substrate. New
methods for generating polynuclear reaction sites featuring
coordinatively unsaturated metal ions are required before the
cooperative reactivity can be assessed. Herein we report the
synthesis of a reactive triiron complex that exhibits the ability to
mediate multielectron reaction chemistry within the trinuclear
core, as demonstrated by activation of inorganic azide to produce
a nucleophilic triiron μ3-nitride complex.
acid (tach 3HBr)5 was derivatized with o-fluoronitrobenzene via
3
nucleophilic aromatic substitution (6 equiv Cs2CO3, ACN 120 °C
for 72 h). Subsequent reduction of the bright-orange trinitro
species using Zn dust (15 equiv) and saturated NH4Cl in
tetrahydrofuran afforded the hexamine platform 1,3,5-C6H9-
(NHPh-o-NH2)3 in good overall yield (80%). Deprotonation
of 1,3,5-C6H9(NHPh-o-NH2)3 with 3 equiv of n-butyllithium
followed by reaction with tert-butyldimethylsilylchloride af-
t
forded the target ligand 1,3,5-C6H9(NHPh-o-NHSiMe2 Bu)3
(tbsLH6) in good isolated yield (45%) on a multigram scale.
Metalation of the ligand platform was effected by heating
tbsLH6 with Fe2(mes)4 (1.5 equiv) to 75 °C for 12 h in tet-
rahydrofuran. Evolution of mesitylene and consumption of ligand
was determined by 1H NMR, though the reaction product is 1H
NMR silent. Storing the brown oil in cold hexanes (-33 °C)
precipitated the triiron complex (tbsL)Fe3(thf) (1) in good
Previously, we have shown the utility of using hexaamide-
based ligand platforms to proximally coordinate three divalent
metal ions, supporting trinuclear cores with a wide span of metal
ion separations (2.274(1)-2.847(1) Å).4 To probe the cooperative
Received: January 13, 2011
Published: February 18, 2011
r
2011 American Chemical Society
3336
dx.doi.org/10.1021/ja2003445 J. Am. Chem. Soc. 2011, 133, 3336–3338
|

Journal of the American Chemical Society
COMMUNICATION
Figure 1. Solid-state structures for (a) (tbsL)Fe3(thf) (1), (b) [(tbsL)Fe3(μ3-N)][NBu4] (2) (side view in d), and (c) (tbsL)Fe3(μ3-NMe) (3) (side
view in e) with the thermal ellipsoids set at the 50% probability level (hydrogen atoms, solvent molecules, and Bu4N cation in 2 omitted for clarity; Fe
orange, C black, H white, N blue, O red, Si pink). Bond lengths (Å) for 1: Fe1-Fe2, 2.6129(5); Fe1-Fe3, 2.5061(5); Fe2-Fe3, 2.6118(5); Fe-Nbase
,
2.047(2); Fe-Ntbs, 1.950(2); Fe-O, 2.1162(18); for 2: Fe1-Fe2, 2.4212(7); Fe1-Fe3, 2.5444(7); Fe2-Fe3, 2.4737(7); Fe-N7avg, 1.871(3); Fe-
N
base, 2.030(3); Fe-Ntbs, 1.950(3); for 3: Fe1-Fe2, 2.449(3); Fe1-Fe3, 2.487(2); Fe2-Fe3, 2.513(2); Fe-N7avg, 1.892(3); N7-C43, 1.457(14);
Fe-Nbase, 1.996(10); Fe-Ntbs, 1.904(10).
overall yield (80%, Scheme 1). Crystallographic analysis of single
crystals of 1 provided the composition consisting of three iron
ions and an asymmetrically bound hexaamide ligand (Figure 1a).
Unlike the trinuclear complexes we have previously reported, the
large silyl substituents on the ligand platform prevent two of the
three apical amide groups from bridging adjacent metal ions.
Only a single silylamide bridges Fe1 and Fe2, while the other two
silylamides are terminally bound to Fe2 and Fe3, giving each iron
ion a distinct coordination environment. The average Fe-Fe
separation is 2.577(6) Å, which is comparable to that for
previously reported complexes.4 The zero-field 57Fe M€ossbauer
spectrum of 1 reflects its asymmetry as the 110 K spectrum can
only be modeled using three quadrupole doublets (δ, |ΔEQ|
The ligand reorganization accommodates binding of the monatomic
nitride ligand. The nitride is heavily pyramidalized (Σ(Fe-N7-Fe)
= 248.94(14)°, NH3 is 319.8°), sitting 1.205(3) Å above the triiron
basal plane. The solution magnetic moment for paramagnetic 2 is
7.3(2) μB, while the zero-field 57Fe M€ossbauer spectrum of 2
obtained at 120 K shows two quadrupole doublets (δ, |ΔEQ|
(
mm/s)): component 1 0.37, 1.78 (48%); component 2 0.39, 1.23
(51%)). The lower isomer shift of 2 compared to 1 reflects the
increased oxidation of the trinuclear core.4
In contrast to the electrophilic nature of terminal Fe-nitride
complexes,6 the nucleophilicity of the nitride in complex 2 can be
demonstrated by its rapid reaction with methyl iodide to afford a
hexane-soluble methyl imido complex (tbsL)Fe3(μ3-NCH3) (3)
with generation of Bu4NI. Previous examples of triiron imido
complexes were synthesized via reaction of iron carbonyl pre-
cursors with silylazide, nitroarene, or alkyldiazene reagents.10
Storing complex 3 in hexanes at -33 °C deposited crystals
suitable for X-ray diffraction analysis. The solid-state molecular
structure for 3 is shown in Figure 1c. Structurally similar to 2,
complex 3 features a central μ3-imide ligand with average Fe-N
and Fe-Fe bond lengths of 1.892(3) and 2.483(3) Å, respec-
tively. The imide ligand sits 1.265(9) Å above the triiron basal
plane, slightly extended from the nitride (see core highlights
shown in Figure 1d and e). Like 2, the zero field 57Fe M€ossbauer
spectrum of paramagnetic 3 (5.3(2) μB) obtained at 110 K shows
two quadrupole doublets with identical isomer shifts (δ, |ΔEQ|
(
mm/s)): component 1 0.89, 1.68; component 2 0.49, 1.57; com-
ponent 3 0.50, 1.92). The 1H NMR silent material is consistent
with the large solution magnetic moment for paramagnetic 1
(12.0(2) μB) determined by the Evans’ method.
Reaction of complex 1 with tetrabutylammonium azide at
room temperature results in the dissociation of thf from 1 and
consumption of the azide as judged by the absence of the azide
stretch (νN3) in the IR spectrum. Storing the reaction product in
diethyl ether at -33 °C deposited crystals of the reaction product
suitable for X-ray diffraction analysis. The solid-state molecular
structure for the product, shown in Figure 1b, confirmed the
breakdown of the azide to produce the nearly C3-symmetric
nitride product [(tbsL)Fe3(μ3-N)]NBu4 (2) (Scheme 1).
While formation of iron-nitride complexes proceeding via
thermal or photolytic decomposition of iron azides embedded in
tetraazamacrocylic ligand environments is well precedented,7
most polynuclear (nuclearity exceeding two) iron-nitride spe-
cies form via reduction of nitrosyl ligands8 or via metathetical
routes using N(SnMe3)3.9 However, these routes give rise to
unpredictable nuclearity and cluster geometries. Formation of
the anionic nitride 2 proceeds via the two-electron oxidation of 1
where the overall complex geometry is dictated by the trinucleat-
ing ligand (tbsL).
(
mm/s)): component 1 0.36, 1.67 (65%); component 2 0.37, 0.94
(35%)). Obtaining a spectrum at 180 K causes the two quadru-
pole doublets to coalesce (δ, |ΔEQ| (mm/s)): 0.34, 1.44), sug-
gesting the Fe nuclei relax faster than the M€ossbauer time scale
(10-7 s).
In summary, the silyl-substituted ligand platform [tbsL]6-
supports triiron complex formation, maintaining a degree of
coordinative unsaturation at the iron ions. Despite the asymme-
try of the triiron thf-bound complex, the compound rapidly reacts
with a single equivalent of inorganic azide to produce a nearly C3-
symmetric, μ3-nitride complex. The observed nucleophilic re-
activity of the anionic nitride demonstrates the utility of embed-
ding a polynuclear reaction site within a single ligand manifold.
This strategy permits control of the nuclearity of the resultant
complex and elaboration of a cooperatively bound substrate, as
demonstrated by alkylation of the nitride to afford a bridging
Complex 2 features three iron centers that bind the central μ3-
nitride ligand with average Fe-N and Fe-Fe bond lengths of 1.871
(3) and 2.480(1) Å, respectively. Each of the iron ions sit in dis-
torted tetrahedral, nearly cis-divacant octahedral, sites, bridged by two
ligand internal amide residues (average Fe-Nbase 2.030(3) Å) and
capped by one terminal silyl-amide (average Fe-Ntbs 1.950(3) Å).
3337
dx.doi.org/10.1021/ja2003445 |J. Am. Chem. Soc. 2011, 133, 3336–3338

Journal of the American Chemical Society
COMMUNICATION
imido complex. Work is currently underway to understand the
electronic structure of these complexes and canvass the scope of
reactivity of this and related trinuclear platforms with small-
molecule substrates.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures and
b
spectral data for 1, 2, and 3; M€ossbauer spectrum and parameters
for 1, 2, and 3; selected crystallographic data and bond lengths for
1, 2, and 3; CIF file for 1, 2, and 3. This material is available free of
charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
betley@chemistry.harvard.edu
’ ACKNOWLEDGMENT
We thank Harvard University for financial support, Prof. R. H.
Holm for the generous use of his M€ossbauer spectrometer,
Novartis for a predoctoral fellowship for T.M.P., and the NIH
for an NIH Ruth L. Kirschstein NRSA fellowship for A.R.F.
’ REFERENCES
(1) (a) Adams, R. D. J. Organomet. Chem. 2000, 600, 1–6. (b) Suzuki,
H. Eur. J. Inorg. Chem. 2002, 1009–1023. (c) Dyson, P. J. Coord. Chem.
Rev. 2004, 248, 2443–2458. (d) Pap, J. S.; DeBeer George, S.; Berry, J.
F. Angew. Chem., Int. Ed. 2008, 47, 10102–10105.
(2) (a) Bozso, F.; Ertl, G.; Grunze, M.; Weiss, M. J. Cat. 1977,
49, 18–41.(b) Ertl, G. In Catalytic Ammonia Synthesis; Jennings, J.
R., Ed.; Plenum Press: New York, 1991; pp 109-131.
(3) (a) Burgess, B. K.; Lowe, D. J. Chem. Rev. 1996, 96, 2983–3012.
(b) Dos Santos, P. C; Igarashi, R. Y.; Lee, H.-I.; Hoffman, B. M.; Seefeldt,
L. C.; Dean, D. R. Acc. Chem. Res. 2005, 38, 208–214. (c) Hoffman, B.
M; Dean, D. R.; Seefeldt, L. C. Acc. Chem. Res. 2009, 42, 609–619.
(4) Zhao, Q.; Betley, T. A. Angew. Chem., Int. Ed. 2011, 50, 709–712.
(5) Bowen, T.; Planalp, R. P.; Brechbiel, M. W. Bioorg. Med. Chem.
Lett. 1996, 6, 807–810.
(6) (a) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004,
126, 6252–6254. (b) Vogel, C.; Heinemann, F. W.; Sutter, J.; Anthon,
C.; Meyer, K. Angew. Chem., Int. Ed. 2008, 47, 2681–2683. (c) Scepaniak,
J. J.; Fulton, M. D.; Bontchev, R. P.; Duesler, E. N.; Kirk, M. L.; Smith,
J. M. J. Am. Chem. Soc. 2008, 130, 10515–10517.
(7) (a) Summerville, D. A.; Cohen, I. A. J. Am. Chem. Soc. 1976, 98,
1747–1752. (b) Wagner, W.-D.; Nakamoto, K. J. Am. Chem. Soc. 1988,
110, 4044–4045. (c) Meyer, K.; Bill, E.; Mienert, B.; Weyherm€uller, T.;
Wieghardt, K. J. Am. Chem. Soc. 1999, 121, 4859–4876.
(8) (a) Fjare, D. E.; Gladfelter, W. L. J. Am. Chem. Soc. 1981,
103, 1572–1574. (b) Hourihane, R.; Spalding, T. R.; Ferguson, G.;
Deeney, T.; Zanello, P. J. Chem. Soc., Dalton Trans. 1993, 43–46. (c)
Della Pergola, R.; Bandini, C.; Demartin, F.; Diana, E.; Garlaschelli, L.;
Stanghellini, P. L.; Zanello, P. J. Chem. Soc., Dalton Trans. 1996,
747–754.
(9) Bennett, M. V.; Stoian, S.; Bominaar, E. L.; Munck, E.; Holm,
R. H. J. Am. Chem. Soc. 2005, 127, 12378–12386.
(10) Silylazide: (a) Barnett, B. L.; Kruger, C. Angew. Chem., Int. Ed.
1971, 10, 910–911. Nitroarene:(b) Landesberg, J. M.; Katz, L.; Olsen, C.
J. Org. Chem. 1972, 37, 930–936. (c) Ragaini, F.; Song, J.-S.; Ramage, D. L.;
Geoffroy, G. L.; Yap, G. A. P.; Rheingold, A. L. Organometallics 1995,
14, 387–400. Alkyldiazene:(d) Wucherer, E. J.; Tasi, M.; Hansert,
B.; Powell, A. K.; Garland, M.-T.; Halet, J.-F.; Saillard, J.-Y.; Vahrenkamp,
H. Inorg. Chem. 1989, 28, 3564–3572.
3338
dx.doi.org/10.1021/ja2003445 |J. Am. Chem. Soc. 2011, 133, 3336–3338