Catalytic C-H bond amination from high-spin iron imido complexes

Article abstract of DOI:10.1021/ja110066j

Dipyrromethene ligand scaffolds were synthesized bearing large aryl (2,4,6-Ph₃C₆H₂, abbreviated Ar) or alkyl ( ^tBu, adamantyl) flanking groups to afford three new disubstituted ligands (^RL, 1,9-R₂-5-mesityldipyrromethene, R = aryl, alkyl). While high-spin (S = 2), four-coordinate iron complexes of the type (^RL)FeCl(solv) were obtained with the alkyl-substituted ligand varieties (for R = ^tBu, Ad and solv = THF, OEt₂), use of the sterically encumbered aryl-substituted ligand precluded binding of solvent and cleanly afforded a high-spin (S = 2), three-coordinate complex of the type (^ArL)FeCl. Reaction of (^AdL)FeCl(OEt₂) with alkyl azides resulted in the catalytic amination of C-H bonds or olefin aziridination at room temperature. Using a 5% catalyst loading, 12 turnovers were obtained for the amination of toluene as a substrate, while greater than 85% of alkyl azide was converted to the corresponding aziridine employing styrene as a substrate. A primary kinetic isotope effect of 12.8(5) was obtained for the reaction of (^AdL)FeCl(OEt₂) with adamantyl azide in an equimolar toluene/toluene-d₈ mixture, consistent with the amination proceeding through a hydrogen atom abstraction, radical rebound type mechanism. Reaction of p-^tBuC₆H₄N₃ with (^ArL)FeCl permitted isolation of a high-spin (S = 2) iron complex featuring a terminal imido ligand, (^ArL)FeCl(N(p- ^tBuC₆H₄)), as determined by ¹H NMR, X-ray crystallography, and ⁵⁷Fe Moessbauer spectroscopy. The measured Fe-N_imide bond distance (1.768(2) A) is the longest reported for Fe(imido) complexes in any geometry or spin state, and the disruption of the bond metrics within the imido aryl substituent suggests delocalization of a radical throughout the aryl ring. Zero-field ⁵⁷Fe Moessbauer parameters obtained for (^ArL)FeCl(N(p- ^tBuC₆H₄)) suggest a Fe^III formulation and are nearly identical with those observed for a structurally similar, high-spin Fe^III complex bearing the same dipyrromethene framework. Theoretical analyses of (^ArL)FeCl(N(p-^tBuC ₆H₄)) suggest a formulation for this reactive species to be a high-spin Fe^III center antiferromagnetically coupled to an imido-based radical (J = -673 cm^-1). The terminal imido complex was effective for delivering the nitrene moiety to both C-H bond substrates (42% yield) as well as styrene (76% yield). Furthermore, a primary kinetic isotope effect of 24(3) was obtained for the reaction of (^ArL)FeCl(N(p- ^tBuC₆H₄)) with an equimolar toluene/toluene-d₈ mixture, consistent with the values obtained in the catalytic reaction. This commonality suggests the isolated high-spin Fe ^III imido radical is a viable intermediate in the catalytic reaction pathway. Given the breadth of iron imido complexes spanning several oxidation states (Fe^II-Fe^V) and several spin states (S = 0 → ³/₂), we propose the unusual electronic structure of the described high-spin iron imido complexes contributes to the observed catalytic reactivity.

Full text of DOI:10.1021/ja110066j

ARTICLE
pubs.acs.org/JACS
Catalytic C-H Bond Amination from High-Spin Iron Imido Complexes
Evan R. King, Elisabeth T. Hennessy, and Theodore A. Betley*
Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street, Cambridge, Massachusetts 02138, United States
S Supporting Information
b
ABSTRACT: Dipyrromethene ligand scaffolds were synthesized bearing large
aryl (2,4,6-Ph3C6H2, abbreviated Ar) or alkyl (tBu, adamantyl) flanking groups to
afford three new disubstituted ligands (RL, 1,9-R2-5-mesityldipyrromethene, R =
aryl, alkyl). While high-spin (S = 2), four-coordinate iron complexes of the type
(RL)FeCl(solv) were obtained with the alkyl-substituted ligand varieties (for R =
tBu, Ad and solv = THF, OEt2), use of the sterically encumbered aryl-substituted
ligand precluded binding of solvent and cleanly afforded a high-spin (S = 2),
three-coordinate complex of the type (ArL)FeCl. Reaction of (AdL)FeCl(OEt2)
with alkyl azides resulted in the catalytic amination of C-H bonds or olefin
aziridination at room temperature. Using a 5% catalyst loading, 12 turnovers were
obtained for the amination of toluene as a substrate, while greater than 85% of alkyl azide was converted to the corresponding
aziridine employing styrene as a substrate. A primary kinetic isotope effect of 12.8(5) was obtained for the reaction of
(AdL)FeCl(OEt2) with adamantyl azide in an equimolar toluene/toluene-d8 mixture, consistent with the amination proceeding
through a hydrogen atom abstraction, radical rebound type mechanism. Reaction of p-tBuC6H4N3 with (ArL)FeCl permitted
isolation of a high-spin (S = 2) iron complex featuring a terminal imido ligand, (ArL)FeCl(N(p-tBuC6H4)), as determined by 1H
NMR, X-ray crystallography, and 57Fe M€ossbauer spectroscopy. The measured Fe-Nimide bond distance (1.768(2) Å) is the
longest reported for Fe(imido) complexes in any geometry or spin state, and the disruption of the bond metrics within the imido aryl
substituent suggests delocalization of a radical throughout the aryl ring. Zero-field 57Fe M€ossbauer parameters obtained for
(ArL)FeCl(N(p-tBuC6H4)) suggest a FeIII formulation and are nearly identical with those observed for a structurally similar, high-
spin FeIII complex bearing the same dipyrromethene framework. Theoretical analyses of (ArL)FeCl(N(p-tBuC6H4)) suggest a
formulation for this reactive species to be a high-spin FeIII center antiferromagnetically coupled to an imido-based radical (J = -
673 cm-1). The terminal imido complex was effective for delivering the nitrene moiety to both C-H bond substrates (42% yield) as
well as styrene (76% yield). Furthermore, a primary kinetic isotope effect of 24(3) was obtained for the reaction of
(ArL)FeCl(N(p-tBuC6H4)) with an equimolar toluene/toluene-d8 mixture, consistent with the values obtained in the catalytic
reaction. This commonality suggests the isolated high-spin FeIII imido radical is a viable intermediate in the catalytic reaction
pathway. Given the breadth of iron imido complexes spanning several oxidation states (FeII-FeV) and several spin states (S = 0 f
3/2), we propose the unusual electronic structure of the described high-spin iron imido complexes contributes to the observed
catalytic reactivity.
been reported that mimics the catalytic transfer of the metal-
ligand multiply bonded functionality found in the biological
systems.8-13
I. INTRODUCTION
Introducing functionality into unactivated C-H bonds re-
mains a significant challenge both in the realm of complex
molecule synthesis as well as in the elaboration of simple
hydrocarbon feedstocks into value-added commodity
chemicals.1,2 Biological C-H bond functionalization is primarily
performed by iron-containing enzymes that utilize dioxygen as
the terminal oxidant. A key structural element of the putative
hydroxylation catalyst in both heme (where iron is embedded in
a porphyrin) and non-heme systems is a transiently formed
terminal iron oxo species, typically thought to involve multiple-
bond character.3,4 Furthermore, the reactivity of this intermedi-
ate is believed to be dictated by its electronic structure.5-7 In
non-heme enzymes four such FeIV(oxo) complexes have been
characterized, and their reactivity has been linked to a common
electronic feature: namely a high-spin ground state (S = 2).4
However, an isolable, high-spin synthetic analogue has not
Parallel to the work targeted at iron-mediated hydroxylation
chemistry, C-H bond amination14-20 and olefin
aziridination14,16,21-23 have been the focus of much recent
synthetic work, although many mechanistic details and their
interplay in effecting chemo- and regioselectivity remain poorly
understood. The synthesis and characterization of Fe(imido)
complexes as isoelectronic surrogates to Fe(oxo) functionalities
have been targeted in the pursuit of effecting viable catalytic
delivery of the nitrene functional unit to a C-H bond or olefinic
substrates. Iron imido complexes have now been characterized in
four oxidation states spanning a range of spin states (FeII, S = 0;24
Received: November 18, 2010
Published: March 15, 2011
r
2011 American Chemical Society
4917
dx.doi.org/10.1021/ja110066j J. Am. Chem. Soc. 2011, 133, 4917–4923
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Ligand and Metal Complex Syntheses
Table 1. Spectral and Magnetic Properties of Complexes 1-4
complex
tbuL)FeCl(thf)
AdL)FeCl(OEt2) (1)
(ArL)FeCl (2)
[(ArL)FeCl]2(μ-N(Ph)(C6H5)N) (3)
(ArL)FeCl(NC6H4-p-tBu) (4)
μeff (μB)
S
λ/nm (ε/M-1 cm-1
)
δ (mm/s)b
|ΔEQ| (mm/s)b
(
(
5.2(2)
5.1(1)
5.1(2)
7.8(2)
5.3(1)
2
2
2
497 (40 000)a
494 (59 000)a
554 (26 000)
551 (27 000)
553 (26 000)
0.98
0.68
0.33
0.29
3.70
0.68
2.15
2.29
5/2, 5/2
2
a UV/vis reported for pyridine adduct. b Recorded at 105 K.
FeIII, S = 1/2, 1, 3/2;25-30 FeIV, S = 1;31-33 FeV, S = 1/234) and
have been shown to engage in group transfer to carbon monoxide
to produce isocyanates25,35 and to isocyanides to produce
carbodiimides,35 undergo hydrogenation,27 and perform H atom
abstraction from C-H bonds.36,37 Herein we report room-
temperature, catalytic C-H bond and olefin functionalization
from a transiently formed, high-spin (S = 2) iron imido complex.
h, R = adamantyl 88%, R = Bu 65%).41,42 Disubstituted
dipyrromethene ligands (XL, X = 1,9-substituent) were prepared
using literature procedures to produce the ligands in good overall
yields ((ArL)H, 49%; (tBuL)H, 73%; (AdL)H, 71%; Scheme 1b).37
Dipyrromethene deprotonation with phenyllithium in thawing
benzene afforded the lithium salts (ArL)Li, (tBuL)Li, and (AdL)Li
as brightly colored powders in nearly quantitative yields (88-
92%) for subsequent transmetalation to iron (Scheme 1b).
Formation of the iron dipyrromethene complexes proceeds
cleanly from reaction of the lithio dipyrromethene species with a
thawing slurry of FeCl2 in an ethereal solvent. For example,
reaction of (AdL)Li with FeCl2 in thawing diethyl ether cleanly
affords the solvated complex (AdL)FeCl(OEt2) (1) as a lumines-
cent green-brown solid following precipitation (yield 58%;
Scheme 1c). Utilizing the aryl-substituted ligand (ArL)Li under
similar reaction conditions affords the three-coordinate species
(ArL)FeCl (2) as a bright purple solid following crystallization
(46%). The compositions and purity of 1 and 2 were established
t
II. RESULTS AND DISCUSSION
During our investigations of iron dipyrromethene complexes
as heme surrogates, we observed that reaction of an FeII complex
with organic azides led to facile intramolecular delivery of the
nitrene functional group into a ligand C-H bond.37 The reaction
was postulated to proceed via a high-valent FeIV(NR) species,
akin to the hydroxylation pathway of cytrochrome P450 and its
functional analogues.38,39 Extending this reactivity to an inter-
molecular reaction required removal of reactive C-H bonds
from the ligand platform. Thus, dipyrromethene platforms were
targeted featuring sterically encumbered aryl or alkyl substituents
that lack weak C-H bonds to circumvent intramolecular C-H
bond activation pathways. Pyrroles substituted in the 2-position
were afforded using modified Negishi coupling for 2-arylpyrroles
and directed Friedel-Crafts alkylations for 2-alkylpyrroles. For
example, 2,4,6-Ph3-C6H2Br was cleanly coupled to sodium
pyrrolide to afford 2-Ar(pyrrole) using conditions outlined by
Sadighi and co-workers (3 equiv of sodium pyrrolide, 3 equiv of
ZnCl2, Pd2(dba)3 (0.67%)/S-Phos (1.33%); THF, 90 ꢀC, 24 h,
84%; Ar = 2,4,6-Ph3C6H2; see Scheme 1a).40 Alkyl-substituted
pyrroles were synthesized in good yields by reaction of the
corresponding alkyl chloride (R-Cl), AlCl3, and ethyl pyrrole-2-
carboxylate following decarboxylation (KOH, glycol, 200 ꢀC, 12
by H NMR, UV-visible spectroscopy, 57Fe M€ossbauer spec-
1
troscopy, and combustion analysis, the data of which are
compiled in Table 1. The respective geometries of four-coordi-
nate 1 and three-coordinate 2 were verified by X-ray diffraction
studies on a single crystal of each (Figure 1; see the Supporting
Information). The solid-state molecular structure of 1 shows a
trigonal-pyramidal geometry with a diethyl ether molecule cap-
ping the pyramid. Complex 2 has a trigonal-planar geometry
about iron, wherein the large 2,4,6-Ph3C6H2 ligand substituents
flank iron above and below the [N2FeCl] plane. The average
bond lengths of the four-coordinate complex (Fe-NL
=
2.035(4) Å, Fe-Cl = 2.249(1) Å) are expanded relative to
three-coordinate species (Fe-NL = 1.966(7) Å, Fe-Cl =
2.154(2) Å). The shorter bond lengths in the three-coordinate
4918
dx.doi.org/10.1021/ja110066j |J. Am. Chem. Soc. 2011, 133, 4917–4923

Journal of the American Chemical Society
ARTICLE
Figure 1. Solid-state molecular structures for (a) (tBuL)FeCl(thf), (b) (AdL)FeCl(OEt2) (1), and (ArL)FeCl (2) with thermal ellipsoids at the 50%
probability level. Color scheme: Fe, orange; Cl, green; C, black; N, blue. Hydrogens, solvent molecules, and aryl ring disorder in 2 are omitted for clarity.
Bond lengths (Å) are as follows. (tBuL)FeCl(thf): Fe-N1, 2.026(2); Fe-N2, 2.028(2); Fe-Cl, 2.255(1); Fe-O, 2.077(2). (AdL)FeCl(OEt2) (1): Fe-
N1, 2.028(3); Fe-N2, 2.042(3); Fe-Cl, 2.250(2); Fe-O, 2.090(2). (ArL)FeCl (2): Fe-N1, 1.966(5); Fe-N2, 1.966(5); Fe-Cl, 2.154(2).
•
1
Table 2. LC/MS (1H NMR) Yields at Various Temperatures
of Products Generated in Catalytic Amination Reaction with 1
coupling two PhCH2 ) is also detectable by H NMR. While
elevated temperatures may facilitate amine dissociation from the
iron catalyst, it does so at the expense of the thermal stability
of the iron catalyst. Complex 1 is also effective for nitrene
delivery to olefinic substrates. Near-quantitative nitrene transfer
was observed when 1 is reacted with N3Ad in styrene at
room temperature, giving 85% of the corresponding aziridine
(17 TON based on 20 equiv of azide).
yield/10-2 mmol
temp/ꢀC BnAdNH PhC(H)NAd AdNH2 PhCH2CH2Ph TON
25
60
6.5 (4.8)
9.8 (11)
8.1 (8.1)
5.7 (5.6)
0.19 (0.20)
0.19 (0.10)
0.58 (0.72)
0.29 (0.13)
0.12
0.46
2.1
6.7 (5.0)
10 (11)
(0.17)
(0.81)
(1.7)
Catalysis screens for the amination of toluene showed com-
90
8.7 (8.8)
6.0 (5.7)
plex 1 to be the most active, whereas the tert-butyl analogue
120
1.6
(
tBuL)FeCl(thf) only showed evidence for 5.6 turnovers at 60 ꢀC,
and complex 2 showed only trace amounts of aminated product
by mass spectrometry. While THF found in the precatalyst (e.g.,
(AdL)FeCl(thf)) is tolerated during catalysis, adding additional
THF to the reaction suppresses amination. Addition of 5 equiv of
THF to an amination reaction (20 equiv of N3Ad/[1]) yields
only 50% benzyladamantylamine with respect to the iron catalyst
1. Addition of 50 equivalents of THF completely suppresses the
reaction, as THF presumably outcompetes the azide from
binding to the catalyst. Addition of a Lewis acid promoter (i.e.,
1 equiv of B(C6F5)3) to scavenge THF from the precatalyst did
not increase the turnover observed for any of the precatalysts
screened. Furthermore, the reaction is impeded by product
inhibition. Addition of 10 equiv of adamantylamine or 15 equiv
of benzyladamantylamine to 1 under typical catalytic conditions
(5% 1, toluene 60 ꢀC) suppresses the catalytic amination to an
undetectable amount.
The detection of 1,2-diphenylethane during the formation of
benzyladamantylamine strongly suggests that an H atom abstrac-
tion pathway is operative. As a probe for direct H atom transfer, a
competition experiment employing a 1:1 ratio of toluene to its
perdeuterio analogue provides a kinetic isotope effect, kH/kD, of
12.8(5) for precatalyst 1. The kH/kD ratio is consistent with a
C-H bond-breaking event contributing to the rate-determining step
and the reaction proceeding by a hydrogen atom abstraction/
rebound mechanism, as illustrated in Scheme 2. The observed
KIE exceeds the values Che and co-workers reported for
stoichiometric C-H amination from isolated RuVI bis-imido
complexes (KIE: 4.8-11)45,46 and those reported by Warren and
co-workers for nitrene delivery from a putative [Cu]2(imido)
complex (KIE: 5-6),47 though the isotope effect reported for
ethylbenzene amination by a Cu(amido) species is substantially
higher (KIE: 70).48 The large KIE value exceeds the classical
value for hydrogen atom transfer (6.5)49 and is thus suggestive of
species are likely a result of both decreased steric repulsion and
the increased electrophilicity of iron in the absence of a fourth
donor. The room-temperature solution magnetic moments
determined using the method of Evans are 5.2(2) μB for 1 and
5.1(2) μB for 2, both consistent with high-spin (S = 2) iron(II).43
Zero-field 57Fe Mossbauer analysis of 1 (δ, |ΔEQ| (mm/s) 0.98,
3.70) and 2 (δ, |ΔEQ| (mm/s) 0.67, 0.68) corroborate this
assignment (see Table 1). The three-coordinate iron chloride 2
features an isomer shift lower than for the other four-coordinate
dipyrromethene FeII complexes and a particularly small quadru-
pole splitting, though these unusual parameters are similar to
those reported for structurally similar trigonal-planar, three-
coordinate β-diketiminate complexes.44
As the reaction of organic azides with (MesL)FeCl(THF) gives
rise to intramolecular amination of a ligand C-H bond,37 we
canvassed the reactivity of complex 1 with organic azides. When
alkyl azides (e.g., (H3C)3CN3, 1-azidoadamantane) are added to
1 in toluene, rapid azide consumption is evident and the product
of intermolecular nitrene insertion into a benzylic C-H bond of
toluene (PhCH2NHR) is observed. Catalytic turnover is ob-
served at room temperature when multiple equivalents of azide
are used. Reactions of 1-azidoadamantane (N3Ad) with 1 in
toluene at room temperature yielded a mixture of benzylada-
mantylamine (95%), benzyladamantylimine (2.8%), and ada-
mantylamine (1.8%) for a total of 6.7 turnovers (TON) (see
Table 2). The turnover is maximized by running the reaction at
60 ꢀC (TON = 10-12) but decreases substantially at elevated
temperatures (TON: 8.7 at 90 ꢀC; 6 at 120 ꢀC). The ratios of the
amination products change at elevated temperatures as well (at
120 ꢀC: benzyladamantylamine (75%), benzyladamantylimine
(4%), and adamantylamine (21%)). At elevated temperatures
(60-120 ꢀC) the presence of 1,2-diphenylethane (product of
4919
dx.doi.org/10.1021/ja110066j |J. Am. Chem. Soc. 2011, 133, 4917–4923

Journal of the American Chemical Society
ARTICLE
Scheme 2. Proposed Catalytic Cycle for the Amination of
C-H Bonds by Reaction of 1 with Organic Azides
1H NMR as a C2-symmetric species distinct from the starting
material. The room-temperature magnetic moment for this
species is 5.3(1) μB, consistent with an S = 2 complex. Zero-
field 57Fe M€ossbauer analysis of the crude reaction product at
100 K reveals a single iron-containing species (δ, |ΔEQ| (mm/s)
0.29, 2.29) that is nearly superimposable with the spectrum
obtained for 3, suggesting the new product also contains FeIII
(Figure 3a).
An X-ray diffraction study on single crystals grown from the
reaction of 2 with p-tBuC6H4N3 revealed the product to be a
monomeric species bearing a terminally bound imido ligand,
(ArL)FeCl(NC6H4-p-tBu) (4) (Figure 2b). The Fe-Cl
(2.210(1) Å) and Fe-NL (2.005(2) Å) bond lengths are
consistent with those found in the bimolecularly coupled FeIII
dimer 3. The Fe-NAr bond length in 4 (1.768(2) Å) is
elongated relative to those in previously reported terminal imido
complexes (e.g., [PhBP3]Fe(N-tol) 1.658(2) Å, S = 1/2;24
3
(
Menacnac)Fe(NAd) 1.662(2) Å, S = /2;30 (iPrPDI)Fe(NAr)
1.705-1.717 Å, S = 1;27 [(N4Py)Fe(NTs)]2þ 1.73 Å, S = 154),
suggesting any iron-imido multiple-bond character is severely
attenuated. To account for the anomalously long Fe-NAr bond
length in 4, the similar M€ossbauer parameters for 3 and 4, and the
observed magnetic moment of monomeric 4, we propose 4 is
comprised of a high-spin FeIII (d5, S = 5/2) center antiferromag-
H atom tunneling akin to that observed in TauD monooxy-
genases (KIE: 37)4 and several FeIV-oxo model complexes
reported by Que and Nam (KIE: 18-50).11,50-52
1
netically coupled to an imido-based radical (S = /2). The
presence of an aryl-delocalized radical can be gleaned from the
C-C bond distances within the nitrene aryl ring (N3-C1,
1.331(2) Å; C1-C2, 1.423(3) Å; C2-C3, 1.372(3) Å; C3-C4,
1.406(3) Å; C4-C5, 1.405(3) Å; C5-C6, 1.378(3) Å; C6-C1,
1.414(3) Å).55 In comparison , for a para-substituted FeIII imido
species that does not feature radical character, the average bond
distances within the imide aryl moiety are C-N = 1.383 Å,
While our proposed mechanism suggests the intermediacy of
an FeIV imido complex prior to the rate-determining step of H
atom abstraction, we sought to validate this hypothesis via
isolation or characterization of the putative intermediate. React-
ing a thawing solution of 2 in benzene with a stoichiometric
amount of phenyl azide quantitatively produces a 1H NMR silent
complex following consumption of the azide, as ascertained by
the disappearance of the azide stretch (νN ) by infrared spec-
troscopy. An X-ray diffraction study on crys3tals grown from the
reaction mixture revealed the product to be the bimolecularly
coupled species [(ArL)FeCl]2(μ-N(Ph)(C6H5)N) (3), shown in
Scheme 3, the solid-state molecular structure of which is shown
in Figure 2a. The coupled product 3 presumably arises from
radical coupling of two monomeric (ArL)FeCl(NPh) moieties to
intermolecularly form a new N-C bond. Complex 3 features
chemically distinct amide N(Ph)R ligation to one iron center
(Fe2) and ketimide ligation to the second (Fe1). The (NPh)
ketimide formulation is supported by the dearomatization of the
C1-C6 ring, featuring localized double bonds (C2-C3 =
1.328(3) Å, C5-C6 = 1.331(3) Å, C1-N6 = 1.276(3) Å).
The room-temperature magnetic moment for 3 is 7.8(2) μB,
slightly lower than the calculated value of 8.3 μB for two
noninteracting high-spin FeIII centers (S = 5/2). The FeIII
formulation is corroborated by zero-field 57Fe M€ossbauer anal-
ysis of 3 at 100 K, which reveals that both iron centers in 3 have
isomer shifts and quadrupole splitting parameters consistent with
high-spin FeIII (δ, |ΔEQ| (mm/s) 0.33, 2.15). The parameters
are significantly distinct from those of both the FeII precursor 2
(δ, |ΔEQ| (mm/s) 0.68, 0.68) and the four-coordinate FeII
species 1 (δ, |ΔEQ| (mm/s) 0.98, 3.70) (Table 1).
Cipso-Cortho = 1.401 Å, Cortho-Cmeta = 1.378 Å, and Cmeta
-
Cpara = 1.387 Å.25 Probing the electronic structure by DFT
corroborates the proposed electronic structure. Broken-symme-
try calculations estimate the antiferromagnetic magnetic ex-
change coupling (J)56 to be -673 cm-1 and the calculated
M€ossbauer parameters match well with those observed for 4
(calculated δ, |ΔEQ| (mm/s) 0.34, -2.00).57,58 The calculated
spin density plot (R - β) for 4, shown in Figure 3b, illustrates
this exchange interaction.
Gratifyingly, monomeric 4 is a reactive source of arylnitrene.
Complex 4 reacts rapidly with 1,4-cyclohexadiene to produce
H2N(C6H4-p-tBu) and benzene, in addition to reacting with
PMe2Ph to produce the phosphinimide Me2PhPdN(C6H4-p-tBu)
and 2. Stirring 4 in toluene at room temperature produces the
benzylic C-H amination product PhCH2NH(C6H4-p-tBu)
(42% yield by 1H NMR), where the remainder of the aryl azide
is converted into the free aniline. The kinetic isotope effect for
the putative imido species 4 was determined via a competition
experiment employing a 1:1 ratio of toluene to its perdeuterio
analogue, providing a kinetic isotope effect, kH/kD, of 24(3).
While this value exceeds the kinetic isotope effect determined for
precatalyst 1, it is of the same order of magnitude and suggests
complex 4 is representative of the transient group transfer
reagent formed in the catalytic runs employing 1 as a precatalyst.
Reacting 4 with styrene (200 equiv) at room temperature
produces the aziridine Ph(CHCH2)N(C6H4-p-tBu) in good
In an effort to obtain a monomeric imido complex, 2 was
reacted with p-tBuC6H4N3, where the aryl para substitution was
selected to sterically prevent the radical coupling pathway
observed in the formation of 3.53 In contrast to the dimerization
observed in the reaction of 2 with phenyl azide, a new product
from the reaction of 2 with p-tBuC6H4N3 is easily discernible by
1
yield (76% by H NMR). One competitive reaction pathway
observed in the aziridination reaction is the radical polymeriza-
tion of styrene initiated by 4, though this could be minimized by
dilution of the styrene substrate. In both the amination and
4920
dx.doi.org/10.1021/ja110066j |J. Am. Chem. Soc. 2011, 133, 4917–4923

Journal of the American Chemical Society
ARTICLE
Scheme 3. Synthesis of the Bimolecularly Coupled FeIII Imido Precursor 3 and the Terminal Imido Complex 4 with a Delocalized
Imido-Based Radical, FeIII(•NAr), and Their Subsequent Reactivity with C-H Bonds
Figure 2. Solid-state core molecular structures for (a) 3 and (b) 4. Ligand aryl substituents, hydrogen atoms, and solvent molecules are omitted for
clarity. Bond lengths (Å) are as follows. For 3: Fe1-N3, 1.810(2); Fe1-Cl1, 2.202(1); C1-N3, 1.276(3); C1-C2, 1.462(4); C2-C3, 1.328(3); C3-
C4, 1.490(4); C4-C5, 1.489(4); C5-C6, 1.331(3); C6-C1, 1.462(4); N4-C4, 1.484(3); Fe2-Cl2, 2.228(1); Fe2-N4, 1.886(2). For 4: Fe-Cl,
2.210(1); Fe-N3, 1.768(2); N3-Cl, 1.331(2); C1-C2, 1.423(3); C2-C3, 1.372(3); C3-C4, 1.406(3); C4-C5, 1.405(3); C5-C6, 1.378(3); C6-
C1, 1.414(3).
aziridination reactions free amine, H2NC6H4-p-tBu, was ob-
served by HR-MS, and 2 was formed as the predominant iron-
containing product, as ascertained by 1H NMR. Heating 4 in the
presence of substrate leads to decreased yields of the correspond-
ing amine and aziridine products. Traces of diazene (RNdNR)
are also observable in the HR-MS of pure 4, though how it forms
remains unclear. Furthermore, dimeric 3 is also a reactive source
of phenylnitrene, affording PhCH2NHPh upon heating 3 in
toluene to 80 ꢀC (observed by HR-MS). Thus, dimeric 3 must be
in equilibrium with its monomeric precursor, (ArL)FeCl(NPh),
which reacts analogously to 4 to effect intermolecular C-H bond
amination (Scheme 3).
Given the reactivity observed for 4, we propose that an imido
radical strongly coupled to a high-spin FeIII ion, (L)FeIIICl(•NR),
is the putative group-transfer reagent in both the amination and
aziridination catalytic processes. The reactive species reported
herein is a departure from the typical FeIV assignment invoked in
Fe-mediated group transfer catalysis.3,38,39 The isolation of the
stable terminal imido radical 4 and the imido precursor 3 lend
support to this assignment. Reaction with alkyl azides in the
catalytic reaction should localize the radical character on the
imido N, giving rise to the observed enhanced reactivity toward
C-H bond or olefinic substrates. We attribute the observed
group transfer catalysis to the electronic structure, more specifi-
cally the localized radical character on the imido N, as the
reactivity reported here is distinct from that of other Fe(imido)
complexes.24-34,54,59 Metal imido complexes which bear radical
character on the imido fragment have been invoked in other
transition-metal complexes to explain the observed reactivity.47,60,61
Warren and co-workers invoke a terminal CuIII(imido) species47
and have demonstrated that a terminal CuII(amido) com-
pound,48 both of which are proposed to have radical character
at N, are key intermediates in their C-H bond amination
chemistry. A cobalt(III) imido species, reported by Theopold,
which aminates an intramolecular C-H bond is also believed to
have nitrogen radical character at room temperature despite a
diamagnetic ground state.62 Thus, while the terminal imido
radical bound to a ferric center (4) is the first example of an
4921
dx.doi.org/10.1021/ja110066j |J. Am. Chem. Soc. 2011, 133, 4917–4923

Journal of the American Chemical Society
ARTICLE
Figure 3. Terminal imido radical complex 4 exhibiting M€ossbauer metrical parameters nearly identical with those of the bimolecularly coupled diferric 3,
illustrated in (a) by the zero-field 100 K 57Fe M€ossbauer spectral data, presented as black dots with spectral fits as solid lines (δ, |ΔEQ| (mm/s)): FeII 2
(red) 0.68, 0.68; (FeIII)2 3 (green) 0.33, 2.15; FeIII(•NAr) 4 (blue) 0.29, 2.29) and antiferromagnetic coupling between the high-spin FeIII ion with the
terminalimido radical in4, illustratedin(b) bythe calculated spin density population(R- β) for4 (S = 2) byDFT (B3LYP/TZVP, SV(P); ORCA 2.758).
isolated complex featuring this type of high-spin electronic
configuration, the electronic structure may arise in other syn-
thetic and biochemical catalytic cycles.4,7,63-66
4, selected bond lengths for 1-4, solid-state structures for 1-4,
crystallographic data for 1-4, and a description of computational
methods and calculated molecular orbitals for 4. This material is
available free of charge via the Internet at http://pubs.acs.org.
III. CONCLUSIONS
’ AUTHOR INFORMATION
Catalytic C-H bond amination and olefin aziridination have
been observed from the reaction of organic azides with a simple
iron(II) coordination complex supported by dipyrromethene
ligands. Kinetic isotope analysis of the amination reaction
suggests the C-H bond-breaking event contributes to the
rate-limiting step of the reaction, followed by a radical rebound.
Isolation of a reactive intermediate reveals the putative nitrene-
delivery precursor to be a high-spin (S = 2) iron complex
featuring a terminal imido ligand. Crystallographic, spectro-
scopic, and theoretical analyses suggest a formulation for this
reactive species to be a high-spin iron(III) center antiferromag-
netically coupled to an imido-based radical. The terminal imido
complex was effective for delivering the nitrene moiety to both
the C-H bond and olefinic substrates. The similarities observed
in the kinetic isotope effects observed during catalytic runs and
those using the isolated imido suggest the high-spin FeIII(imido
radical) is representative of the group transfer reagent in the
catalytic sequence. Given the breadth of iron imido complexes
spanning several oxidation states (FeII-FeV) and several spin
Corresponding Author
*E-mail: betley@chemistry.harvard.edu.
’ ACKNOWLEDGMENT
We thank Harvard University, the ACS PRF Fund (Type G),
and the NSF (CHE-0955885) for financial support and Prof. R.
H. Holm for the generous use of his M€ossbauer spectrometer. E.
T.H. thanks the DOE SCGF for a predoctoral fellowship. We
thank Dr. Andrew Tyler for help with GC and LCMS studies.
’ REFERENCES
(1) Bergman, R. G. Nature 2007, 446, 391.
(2) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507.
(3) Ortiz de Montellano, P. R.; Ed. Cytochrome P450: Structure,
Mechanism, and Biochemistry, 4th ed.; Kluwer Academic/Plenum Pub-
lishers: New York, 2005.
(4) Krebs, C.; Fujimori, D. G.; Walsh, C. T.; Bollinger, M. J., Jr. Acc.
Chem. Res. 2007, 40, 484.
3
states (S = 0 f /2), we propose that the unusual electronic
structure of the described high-spin iron imido complexes
contributes to the observed catalytic reactivity. The high-spin
nature and radical nitrene character serve to destabilize the Fe-
imido bond, which presumably leads to the observed reactivity.
Work is currently underway to determine the generality and
scope of the nitrene-transfer reaction and to find whether this
group transfer reaction pathway is amenable for the delivery of
other functional groups.
(5) Decker, A.; Rhode, J.-U.; Klinker, E. J.; Wong, S. D.; Que, L., Jr.;
Solomon, E. I. J. Am. Chem. Soc. 2007, 129, 15938.
(6) Bernasconi, L.; Louwerse, M. J.; Baerends, E. J. Eur. J. Inorg.
Chem. 2007, 3023.
(7) Ye, S.; Neese, F. Curr. Opin. Chem. Biol. 2009, 13, 89.
(8) Pestovsky, O.; Stoian, S.; Bominaar, E. M.; Shan, X.; M€unck, E.;
Que, L., Jr.; Bakac, A. Angew. Chem., Int. Ed. 2005, 44, 6871.
(9) Kaizer, J.; Klinker, E. J.; Oh, N. Y.; Rhode, J.-U.; Song, W. J.;
Stubna, A.; Kim, J.; M€unck, E.; Nam, W.; Que, L., Jr. J. Am. Chem. Soc.
2004, 126, 472.
’ ASSOCIATED CONTENT
(10) Kumar, D.; Hirao, H.; Que, L., Jr.; Shaik, S. J. Am. Chem. Soc.
2005, 127, 8026.
S
Supporting Information. Text, figures, tables, and CIF
b
(11) England, J.; Martinho, M.; Farquhar, E. R.; Frisch, J. R.;
Bominaar, E. L.; M€unck, E.; Que, L., Jr. Angew. Chem., Int. Ed. 2009,
48, 3622.
files giving details of the syntheses of 1-4, reactivity and catalysis
experiments for 1-4, M€ossbauer spectra and parameters for 1-
4922
dx.doi.org/10.1021/ja110066j |J. Am. Chem. Soc. 2011, 133, 4917–4923

Journal of the American Chemical Society
ARTICLE
(12) England, J.; Guo, Y.; Farquhar, E. R.; Young, V. G., Jr.; M€unck,
E.; Que, L., Jr. J. Am. Chem. Soc. 2010, 132, 8635.
(13) Lacy, D. C.; Gupta, R.; Stone, K. L.; Greaves, J.; Ziller, J. W.;
Hendrich, M.; Borovik, A. S. J. Am. Chem. Soc. 2010, 132, 12188.
(14) M€uller, P.; Fruit, C. Chem. Rev. 2003, 103, 2905.
(15) Davies, H. M. L.; Long, M. S. Angew. Chem., Int. Ed. 2005,
44, 3518.
(16) Halfen, J. A. Curr. Org. Chem. 2005, 9, 657.
(17) Cenini, S.; Gallo, E.; Caselli, A.; Ragaini, R.; Fantauzzi, S.;
Piangiolino, C. Coord. Chem. Rev. 2006, 250, 1234.
(18) Davies, H. M. L.; Manning, J. R. Nature 2008, 451, 417.
(19) Collet, F.; Dodd, R. H.; Dauban, P. Chem. Commun.
2009, 5061.
(51) Nam, W. Acc. Chem. Res. 2007, 40, 522.
(52) Klinker, E. J.; Shaik, S.; Hirao, H.; Que, L., Jr. Angew. Chem., Int.
Ed. 2009, 48, 1291.
(53) Peters and co-workers reported a Cu di-p-tolyl aminyl radical,
in which the tolyl methyl groups prevented radical coupling to form
aryl-aryl-linked dimers, which were seen if methyl was replaced with
hydrogen by using a diphenyl amido unit. See: Mankad, N. P.; Antholine,
W. E.; Szilagyi, R. K.; Peters, J. C. J. Am. Chem. Soc. 2009, 131, 3878.
(54) Klinker, E. J.; Jackson, T. A.; Jensen, M. P.; Stubna, A.; Juhꢀasz,
G.; Bominaar, E. L.; M€unck, E.; Que, L., Jr. Angew. Chem., Int. Ed. 2006,
45, 7394.
(55) Chaudhuri, P.; Verani, C. N.; Bill, E.; Bothe, E.; Weyherm€uller,
T.; Wieghardt, K. J. Am. Chem. Soc. 2001, 123, 2213.
(20) Zalatan, D. N.; Du Bois, J. Top. Curr. Chem. 2010, 292, 347.
(21) Tanner, D. Angew. Chem., Int. Ed. 1994, 33, 599.
(22) Osborn, H. M. I.; Sweeney, J. B. Tetrahedron: Asymmetry 1997,
8, 1693.
(56) With spin Hamiltonian H = -2JSFe
SNR and coupling constant
3
J = -(EHS - EBS)/(ÆS2æHS - ÆS2æBS).
(57) Ye, S.; Tuttle, T.; Bill, E.; Simkhovich, L.; Gross, Z.; Thiel, W.;
Neese, F. Chem. Eur. J. 2008, 14, 10839.
(23) Sweeney, J. B. Chem. Soc. Rev. 2002, 31, 247.
(24) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 1913.
(25) Brown, S. D.; Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003,
125, 322.
(26) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003, 125, 10782.
(27) Bart, S. C.; Lobkovsky, E.; Bill, E.; Chirik, P J. J. Am. Chem. Soc.
2006, 128, 5302.
(58) Neese, F. ORCA-An ab initio, Density Functional and Semi-
empirical Electronic Structure Package, Version 2.7-00 ed.; Universit€at
Bonn, Bonn, Germany, 2009.
(59) Mankad, N. P.; M€uller, P.; Peters, J. C. J. Am. Chem. Soc. 2010,
132, 4083.
(60) Lu, C. C.; DeBeer George, S.; Weyherm€uller, T.; Bill, E.; Bothe,
E.; Wieghardt, K. Angew. Chem., Int. Ed. 2008, 47, 6384.
(61) Kogut, E.; Wiencko, H. L.; Zhang, L.; Cordeau, D. E.; Warren,
T. H. J. Am. Chem. Soc. 2005, 127, 11248.
(28) Lu, C. C.; Saouma, C. T.; Day, M. W.; Peters, J. C. J. Am. Chem.
Soc. 2007, 129, 4.
(29) Scepaniak, J. J.; Young, J. A.; Bontchev, R. P.; Smith, J. M.
Angew. Chem., Int. Ed. 2009, 48, 3158.
(30) Cowley, R. E.; DeYonker, N. J.; Eckert, N. A.; Cundari, T. R.;
DeBeer, S.; Bill, E.; Ottenwaelder, X.; Flaschenriem, C.; Holland, P. L.
Inorg. Chem. 2010, 49, 6172.
(62) Shay, D. T.; Yap, G. P. A.; Zakharov, L. N.; Rheingold, A. L.;
Theopold, K. H. Angew. Chem., Int. Ed. 2005, 44, 1508.
(63) Groves, J. T.; Van Der Puy, M. J. Am. Chem. Soc. 1974, 96, 5274.
(64) Groves, J. T.; McClusky, G. A. J. Am. Chem. Soc. 1976, 98, 859.
(65) Chen, M. S.; White, M. C. Science 2007, 318, 783.
(66) Chen, M. S.; White, M. C. Science 2010, 327, 566.
(31) Verma, A. K.; Nazif, T. N.; Achim, C.; Lee, S. C. J. Am. Chem.
Soc. 2000, 122, 11013.
(32) Thomas, C. M.; Mankad, N. P.; Peters., J. C. J. Am. Chem. Soc.
2006, 128, 4956.
(33) Nieto, I.; Ding, R.; Bontchev, R. P.; Wang, H.; Smith, J. M.
J. Am. Chem. Soc. 2008, 130, 2716.
(34) Ni, C.; Fettinger, J. C.; Long, G. J.; Brynda, M.; Power, P. P.
Chem. Commun. 2008, 6045.
(35) Cowley, R. E.; Eckert, N. A.; Elhaik, J. E.; Holland, P. L. Chem.
Commun. 2009, 1760.
(36) Cowley, R. E.; Holland, P. L. Inorg. Chim. Acta 2011, in press.
(37) King, E. R.; Betley, T. A. Inorg. Chem. 2009, 48, 2361.
(38) Groves, J. T. J. Chem. Educ. 1985, 62, 928.
(39) Sono, M.; Roach, M. P.; Coulter, E. D.; Dawson, J. H. Chem.
Rev. 1996, 96, 2841.
(40) Rieth, R. D.; Mankad, N. P.; Calimano, E.; Sadighi, J. P. Org.
Lett. 2004, 6, 3981.
(41) Bailey, D. M.; Johnson, R. E.; Albertson, N. F. Organic Syntheses;
Wiley: New York, 1988; Collect Vol. 6, 618.
(42) Harman, W. H.; Harris, T. D.; Freedman, D. E.; Fong, H.;
Chang, A.; Rinehart, J. D.; Ozarowski, A.; Sougrati, M. T.; Grandjean, F.;
Long, G. L.; Long, J. R.; Chang, C. J. J. Am. Chem. Soc. 2010, 132, 18115.
(43) Evans, D. F. J. Chem. Soc. 1959, 2003.
(44) Andres, H.; Bominaar, E. L.; Smith, J. M.; Eckert, N. A.;
Holland, P. L.; M€unck, E. J. Am. Chem. Soc. 2002, 124, 3012..
(45) Au, S. M.; Huang, J. S.; Yu, W. Y.; Fung, W. H.; Che, C. M.
J. Am. Chem. Soc. 1999, 121, 9120.
(46) Leung, S. K. Y.; Tsui, W. M.; Huang, J. S.; Che, C. M.; Liang,
J. L.; Zhu, N. Y. J. Am. Chem. Soc. 2005, 127, 16629.
(47) Badiei, Y. M.; Dinescu, A.; Dai, X.; Palomino, R. M.; Heinemann,
F. W.; Cundari, T. R.; Warren, T. H. Angew. Chem., Int. Ed. 2008, 47, 9961.
(48) Wiese, S; Badiei, Y. M.; Gephart, R. T.; Mossin, S.; Varonka,
M. S.; Melzer, M. M.; Meyer, K.; Cundari, T. R.; Warren, T. H. Angew.
Chem., Int. Ed. 2010, 49, 8850.
(49) Anslyn, E. V.; Dougherty, D. A. In Modern Physical Organic
Chemistry; University Science Books: Sausalito, CA, 2006.
(50) Que, L., Jr. Acc. Chem. Res. 2007, 40, 493.
4923
dx.doi.org/10.1021/ja110066j |J. Am. Chem. Soc. 2011, 133, 4917–4923