Selective extraction of N2 from air by diarylimine iron complexes

Article abstract of DOI:10.1021/ja311021u

Treatment of cis-(Me₃P)₄FeMe₂ with ortho-substituted diarylimines afforded 2 equiv of MeH, PMe₃, and {mer-κC,N,C′-(Ar-2-yl)CH₂N=CH(Ar′-2-yl)}Fe(PMe ₃)₃ (Ar = 3,4,6-(F)₃-C₆H, Ar′ = 3,5-(CF₃)₂-C₆H₂, 1a; Ar = 3,4,6-(F)₃-C₆H, Ar′ = 3,4,5-(F)₃-C ₆H, 1b; Ar = 4,5,6-(F)₃-C₆H, Ar′ = 3,5-(CF₃)₂-C₆H₂, 1c; Ar = C ₆H₄, Ar′ = 3-(OMe)-C₆H₃, 1d; Ar = 4,5,6-(F)₃-C₆H, Ar′ = 3,6-Me₂-C ₆H₃, 1e; Ar = C₆H₄, Ar′ = 3,6-Me₂-C₆H₂, 1f). Exposure of 1a-f to O ₂ caused rapid degradation, but substitution of the unique PMe ₃ with N₂ occurred when 1a-f were exposed to air or N ₂ (1 atm), yielding {mer-κC,N,C′-(Ar-2-yl)CH ₂N=CH(Ar′-2-yl)}Fe(PMe₃)₂L (L = N ₂, 2a-f); CO, CNMe, and N₂CPh₂ derivatives (L = CO, 3a-d,f; L = CNMe, 8b; L = N₂CPh₂, 9b) were prepared. Dihydrogen or NH₃ binding to {mer-κC,N,C′-(3,4,6-(F) ₃-C₆H-2-yl)CH₂N=CH-(3,4,5-(F) ₃-C₆H-2-yl)}Fe(PMe₃)₂ (1b′, S = 1 (calc)) to provide 5b (L = H₂) or 6b (L = NH₃) was found comparable to that of N₂, while PMe₃ (1b) and pyridine (L = py, 7b) adducts were unfavorable. Protolytic conditions were modeled using HCCR as weak acids, and trans-{κC,N-(3,4,5-(F) ₃-C₆H₂)CH₂N=CH(3,4,6-(F) ₃-C₆H-2-yl)}Fe(PMe₃)₃(CCR) (R = Me, 4b-Me; R = Ph, 4b-Ph) were generated from 1b. Exposure of 1b to N₂O or N₃SO₂tol generated 2b and Me₃PO or Me ₃P=N(SO₂)tol, respectively. Calculations revealed 2b to be thermodynamically and kinetically favored over the calculated Fe(III) superoxide complex, ³[FeO₂], relative to 1b′ + N₂ + O₂. The correlation of 1b′ + ³O ₂ to ³[FeO₂] is likely to have a relatively high intersystem crossing point (ICP) relative to 1b′ + N₂ to 2b, thereby explaining the dinitrogen selectivity.

Full text of DOI:10.1021/ja311021u

Article
pubs.acs.org/JACS
Selective Extraction of N2 from Air by Diarylimine Iron Complexes
Erika R. Bartholomew,† Emily C. Volpe,† Peter T. Wolczanski,*,† Emil B. Lobkovsky,†
and Thomas R. Cundari‡
†Department of Chemistry & Chemical Biology, Baker Laboratory, Cornell University, Ithaca, New York 14853, United States
‡Department of Chemistry, Center for Advanced Scientific Computing and Modeling (CASCaM), University of North Texas, Box
305070, Denton, Texas 76203-5070, United States
S
* Supporting Information
ABSTRACT: Treatment of cis-(Me3P)4FeMe2 with ortho-
substituted diarylimines afforded 2 equiv of MeH, PMe3, and
{mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar′-2-yl)}Fe(PMe3)3 (Ar
= 3,4,6-(F)3-C6H, Ar′ = 3,5-(CF3)2-C6H2, 1a; Ar = 3,4,6-(F)3-
C6H, Ar′ = 3,4,5-(F)3-C6H, 1b; Ar = 4,5,6-(F)3-C6H, Ar′ =
3,5-(CF3)2-C6H2, 1c; Ar = C6H4, Ar′ = 3-(OMe)-C6H3, 1d; Ar
= 4,5,6-(F)3-C6H, Ar′ = 3,6-Me2-C6H3, 1e; Ar = C6H4, Ar′ =
3,6-Me2-C6H2, 1f). Exposure of 1a−f to O2 caused rapid degradation, but substitution of the unique PMe3 with N2 occurred
when 1a−f were exposed to air or N2 (1 atm), yielding {mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar′-2-yl)}Fe(PMe3)2L (L = N2, 2a−
f); CO, CNMe, and N2CPh2 derivatives (L = CO, 3a−d,f; L = CNMe, 8b; L = N2CPh2, 9b) were prepared. Dihydrogen or NH3
binding to {mer-κC,N,C′-(3,4,6-(F)3-C6H-2-yl)CH2NCH-(3,4,5-(F)3-C6H-2-yl)}Fe(PMe3)2 (1b′, S = 1 (calc)) to provide 5b
(L = H2) or 6b (L = NH3) was found comparable to that of N2, while PMe3 (1b) and pyridine (L = py, 7b) adducts were
unfavorable. Protolytic conditions were modeled using HCCR as weak acids, and trans-{κC,N-(3,4,5-(F)3-C6H2)CH2N
CH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)3(CCR) (R = Me, 4b-Me; R = Ph, 4b-Ph) were generated from 1b. Exposure of 1b to N2O
or N3SO2tol generated 2b and Me3PO or Me3PN(SO2)tol, respectively. Calculations revealed 2b to be thermodynamically
and kinetically favored over the calculated Fe(III) superoxide complex, 3[FeO2], relative to 1b′ + N2 + O2. The correlation of 1b′
+ 3O2 to 3[FeO2] is likely to have a relatively high intersystem crossing point (ICP) relative to 1b′ + N2 to 2b, thereby explaining
the dinitrogen selectivity.
tions53−57 that operate in cycles proposed by Chatt, Leigh, and
I. INTRODUCTION
Hidai,22−28 and related functionalizations.35−38,58−60 The
The binding of dinitrogen to transition metals, most notably
degree of dinitrogen activation, or the extent of reduction of
iron, is of historical and fundamental interest due to its
N2 upon binding, has often provided key examples of
complexation and subsequent reduction.61,62 As for modeling
the Haber−Bosch process, the chemistry of nitrides,63−65 in
particular iron66−74 and ruthenium nitrides,75,76 and the
conversion of this functionality to ammonia69,70,76 has been a
central focus.
Dinitrogen complexes of iron have been under intense
scrutiny30−32,40,62,77−88 due to the presence of iron in the
nitrogenase cofactor responsible for the reduction of N2 to two
NH3 and H2 via 8 electrons and 8 protons (6 each for the N2; 2
each for the H2), which requires 16 equiv of Mg-ATP.1−13 The
majority of studies concerning iron-bound dinitrogen com-
plexes utilize reducing equivalentsprovided by the metal or
by exogenous agentsand protons to effect the conversion to
ammonia or hydrazine, i.e., Chatt, Leigh, etc. cycles.22−28
Recently, dinitrogen reduction cycles utilizing iron have been
developed in aqueous solution,83,84 and model systems that
feature two iron centers have been explored.86−88
activation and conversion to ammonia by nitrogenase
enzymes1−13 and in the heterogeneous iron- and ruthenium-
catalyzed Haber−Bosch process (BASF, 1913).14−18 The latter
hydrogenation of dinitrogen to ammonia is arguably the most
important industrial operation conducted on earth,14 as it is the
source of roughly 5 × 1011 kg of fertilizer as anhydrous
ammonia, urea, ammonium nitrate (the nitrate is produced
from oxidation of NH3 via the Ostwald process), ammonium
phosphates, and ammonium sulfate.18 It has been estimated
that >1% of the planet’s energy supply is appropriated for
Haber−Bosch operations and that >50% of the nitrogen in our
body is derived from the fertilizer it generates.
Roughly 50 years after the Haber−Bosch process was
initialized, evidence of the complexation of dinitrogen to Ru2+
proved to be no less astonishing, as Allen and Senoff’s
communication19 “was rejected for publication by J. Am. Chem.
Soc. on the grounds that it was impossible”.18 Since the
discovery and characterization of (H3N)5Ru(N2)2+ salts,19−21 a
remarkable number of dinitrogen complexes have been
synthesized,22−32 and various bonding modes of N2 have
been realized,33−39 some even leading to dinitrogen
scission,40−48 hydrogenation to ammonia,49−52 catalytic reduc-
Received: November 8, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1
Somewhat lost in the excitement to ascertain the mechanism
of biological dinitrogen reduction or find insights or improve-
ments to the Haber−Bosch process are the fundamental aspects
of selective dinitrogen binding, a factor critical in the processing
of natural gas. Most specifications for natural gas require <4%
N2 in the feed stream, and dinitrogen removal processing
techniques that rely on cryoscopic distillation, adsorption/
desorption approaches, and even the latest membrane
technologies are expensive.89−91 In addition, methane can be
a competitive binder in sieves and membranes, rendering these
technologies more difficult to implement.
RESULTS
■
Diarylimine Iron(II) Dinitrogen Complexes. 1. Discovery
of N2 Extraction from Air. Diarylimine iron(III) complexes
(i.e., [{mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar-2-yl-X)}Fe-
(PMe3)3]OTf) were previously examined as precursors to
azaallyl ligands via deprotonation, but all attempts, even with
bases such as LiN(TMS)2, regenerated the corresponding
Fe(II) derivatives, {mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar-2-yl-
X)}Fe(PMe3)3. If deprotonation occurred, the resulting high-
energy CNCnb orbital of the putative azaallyl complex {mer-
κC,N,C′-(Ar-2-yl)CHNCH(Ar-2-yl-X)}Fe(PMe3)3 was likely
to internally transfer an electron to the Fe(III) center,
generating an azaallyl radical that elicited hydrogen atom
transfer to afford the Fe(II) product.98
One way of circumventing the internal oxidation would be to
lower the CNCnb azaallyl orbital below the “t2g” set of the
pseudo-octahedral Fe(III) species. As a consequence, fluori-
nated diarylimines were considered as a means to render the
ligand more electron-withdrawing, and imine Im-b (Scheme 1)
was readily synthesized via condensation of the inexpensive
constituents 2,4,5-trifluorobenzaldehyde and 3,4,5-trifluoroben-
Dioxygen is normally an anathema to organometallic
systems, and one would hardly expect to conduct dinitrogen
binding reactions in its presence, unless an excess of exogenous
reducing agent is employed. Consider the biological system
pertaining to nitrogenase. In vivo, the NifEN protein delivers
the FeMo-cofactor to its operational nitrogenase protein in a
reducing environment,8−11 while in vitro studies are typically
done with dioxygen scrubbed from the system. However, there
are some exceptions. Allen and Senoff’s (H3N)5RuN22+ salt was
shown to form in the presence of dioxygen,92 and [HFe-
99
zylamine. Treatment of cis-(Me3P)4FeMe2 with Im-b
appeared to provide the diarylimine100−103 tris-phosphine
complex {mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH-
(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)3 (1b), but upon examination
after transfer to a dinitrogen-filled glovebox, only the dinitrogen
complex trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2N
CH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2(N2) (2b) was observed.
Moreover, the red color of 2b persisted when the glassware
used to prepare the complex was exposed to the air.
(dppe)2(THF)]+ and H2(H2)Fe(PEtPh2)4 both form their
93
respective dinitrogen complexes [HFe(dppe)2N2]+
H2(N2)Fe(PEtPh2)4 upon exposure to air.
and
94
Reported herein is the extraction of dinitrogen from the air
by true iron organometallic complexes, which contain two
iron−carbon bonds derived from diarylimine ligands that
sterically and electronically tune the metal center for N2
binding,95−97 while rendering oxidative destruction by dioxygen
kinetically unfavorable. Calculations on the system provide a
rationale for the observed selectivity, and an assessment of how
the compounds might be utilized for dinitrogen removal is
provided.
The apparent stability of the dinitrogen complex in the
presence of air suggested that selective binding of dinitrogen
was feasible; hence, cis-(Me3P)4FeMe2 and Im-b were
combined in a J. Young NMR tube into which benzene-d6
was vacuum transferred. NMR spectroscopy confirmed the
B
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Dinitrogen and Carbonyl Stretching Frequencies for the Complexes trans-{mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar-2-
yl)}Fe(PMe3)2L (L = N2, 2a−f; L = CO, 3a−d,f) and Stability t1/2 Values for 2a−f under Various Conditions (Wet Air,
Atmospheric Conditions; Dry Air, Atmospheric Conditions with the Water Removed; p(H2O) ≈ 22-30 Torr)
preparation of 1b after 1−6 h. Air was admitted to the tube, and
it was vigorously shaken; a second NMR spectrum indicated a
mixture of 1b and 2b, and further admissions of air afforded a
complete conversion to the dinitrogen complex 2b.
trans-{mer-κC,N,C′-(6-(OCH3)-C6H2-2-yl)CH2NCH(C6H4-
2-yl)}Fe(PMe3)2(N2) (2d) in 80% yield. Scheme 1 shows that
a group of diarylimines, ranging from the most electron
withdrawing (3,5-(CF3)2-C6H3)CH2NCH(2,4,5-(F)3-C6H2)
(Im-a) to the modestly electron donating (2,5-Me2-C6H3)-
CH2NCH(C6H5) (Im-f), afford the dinitrogen complexes
trans-{mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar-2-yl)}Fe-
(PMe3)2(N2) (2a−f). All of the dinitrogen complexes
selectively extract N2 from air, and they all have substituents
ortho to the activated aryl positions; only one ortho substituent
appears necessary to cause PMe3 substitution. IR spectra of
2a−f revealed the expected inverse correlation of dinitrogen
stretching frequencies vs electron-withdrawing capacity of each
diarylimine (Table 1). The order suggests that ortho
substituents are more important and that the inductive
influence of the o-methoxy group of 2d outweighs its π-
donor capacity.
3. Diarylimine Tris-PMe3 Iron Intermediates. In order to
observe the putative tris-PMe3 precursors to the dinitrogen
complexes, cis-(Me3P)4FeMe2 was treated with Im-a−Im-f in
NMR tube scale reactions where dinitrogen was kept from the
system. NMR spectra corresponding to diamagnetic {mer-
κC,N,C′-(Ar-2-yl)CH2NCH(Ar-2-yl)}Fe(PMe3)3 were ob-
tained for 1b,d, whereas chemical shifts and line widths
consistent with paramagnetic complexes were observed for
1a,c,e,f. Evans method104 studies on the crude solutions of
1a,c,e,f afforded μeff values ranging from 3.0 to 3.3 μB, which are
consistent with plausible S = 1 ground states possessing
significant contributions from spin−orbit coupling.105 The
spectra for 1b,d were similar to that of 1g and those previously
2. Scope of Dinitrogen Complexes. Inductively, the
electron-withdrawing fluorines on Im-b should attenuate π-
back-bonding to N2. In contradiction of this logic, the six
fluorine substituents were initially considered as an electronic
feature that somehow elicited selective dinitrogen extraction
over dioxygen from air. Those considerations were rapidly
dashed as cis-(Me3P)4FeMe2 was treated with (2,3,4-(F)3-
C6H2)CH2NCH(2,3,4-(F)3-C6H2) (Im-g), a diarylimine
considered to have an electron-withdrawing capacity similar
to that of Im-b, but without ortho substituents. Formation of
the triphosphine complex {mer-κC,N,C′-(4,5,6-(F)3-C6H-2-
yl)CH2NCH(4,5,6-(F)3-C6H-2-yl)}Fe(PMe3)3 (1g) was
noted, but the complex did not react with dinitrogen. Clearly
the electron-withdrawing capacity of the diarylimine did not
appear to matter significantly. The result corroborated previous
findings that {mer-κC,N,C′-(C6H4-2-yl)CH2NCH(C6H4-2-
yl)}Fe(PMe3)3 (1h) and {mer-κC,N,C′-(5-(CF3)-C6H2-2-yl)-
CH2NCH(C6H4-2-yl)}Fe(PMe3)3 (1i) were also impervious
to substitution by dinitrogen.
Another previously explored derivative, {mer-κC,N,C′-(6-
(OCH3)-C6H2-2-yl)CH2NCH(C6H4-2-yl)}Fe(PMe3)3
(1d),98 whose methoxy-substituted diarylimine ligand was
considered to be significantly more electron donating than
Im-b, had only been prepared on an NMR tube scale and had
not been exposed to N2. An attempted isolation of 1d under a
nitrogen atmosphere instead yielded the dinitrogen derivative
C
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. (a) Molecular view (50% probability ellipsoids) of trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe-
(PMe3)2(N2) (2b) and (b) a view down the P−Fe−P axis with the phosphine methyl groups removed. Selected bond distances (Å) and angles
(deg): Fe−C1, 1.996(2); Fe−C14, 2.014(2); Fe−N1, 1.935(2); Fe−N2, 1.811(2); Fe−P1, 2.2292(7); Fe−P2, 2.2389(7); N2−N3, 1.104(3); N1−
C7, 1.308(3); N1−C8, 1.459(3); C6−C7, 1.443(3); C8−C9, 1.501(4); N1−Fe−N2, 178.17(10); C1−Fe−C14, 163.06(9); P1−Fe−P2, 176.04(3);
N1−Fe−C1, 81.25(9); N1−Fe−C14, 81.82(9); N1−Fe−P1, 91.47(6); N1−Fe−P2, 91.32(6); N2−Fe−P1, 89.54(7); N2−Fe−P2, 87.74(7); C1−
Fe−P1, 91.62(7); C14−Fe−P1, 89.13(7); C1−Fe−P2, 91.59(7); C14−Fe−P2, 88.49(6); C1−Fe−N2, 97.20(9); C14−Fe−N2, 99.73(9); Fe−N2−
N3, 178.9(2); Fe−C1−C2, 134.17(19); Fe−C1−C6, 111.81(15); Fe−C14−C13, 131.84(16); Fe−C14−C9, 113.79(18).
recorded for 1h,i, including the diagnostic A2B pattern in the
31P{1H} NMR that was used to verify the meridional
conformation, including JPP values of 58 (1b), 62 (1d), and
61 Hz (1g).98
Table 2. Selected Crystallographic and Refinement Data for
trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-
(F)3-C6H-2-yl)}Fe(PMe3)2(N2) (2b) and trans-{mer-
κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-
2-yl)}Fe(PMe3)2Cl (12b)
It is plausible that significant dissociation of PMe3 to the five-
coordinate {mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar-2-yl)}Fe-
(PMe3)2 (1′a,c,e,f) complexes, which are likely to be triplets
according to DFT calculations, occurs for these complexes. The
spectrum of trans-{mer-κC,N,C′-(3,5-(CF3)2-C6H3)CH2N
CH(2,4,5-(F)3-C6H2)}Fe(PMe3)31 (1a) contradicts this possi-
bility, as paramagnetically shifted H NMR spectral resonances
for the unique phosphine (δ −11.46) and trans-PMe3 (δ
−4.86) groups are observed in addition to that of free PMe3.
Rapid exchange with free PMe3 via a dissociative process is
thereby ruled out, and it appears unlikely that five-coordinate
1′a impacts the spectrum of 1a.
2b
12b
formula
formula wt
space group
Z
C20H23F6N3P2Fe
537.20
C20H23F6NP2ClFe
544.63
Cc
P21/c
4
4
a, Å
19.0597(7)
9.1733(4)
15.7972(10)
90
10.6916(9)
8.1275(7)
26.737(2)
90
b, Å
c, Å
α, deg
β, deg
121.8450(10)
90
100.653(4)
90
γ, deg
V, Å3
ρcalcd, g cm−3
μ, mm−1
temp, K
λ (Å)
The 1H NMR spectra of 1c,e,f are broader and fewer features
are revealed, including overlapping resonances for the distinct
PMe3 sites. The 1H NMR spectrum of 1c included a resonance
for free PMe3; thus, it is likely that its unique PMe3 is bound
strongly enough not to exchange with the free ligand on the
time scale of the NMR observation. DFT calculations indicated
that S = 0 and S = 1 states were close in energy for the tris-
phosphine complexes (vide infra).
2346.2(2)
1.521
2283.3(3)
1.584
0.839
0.974
173(2)
173(2)
0.71073
0.71073
ab
,
R indices (I > 2σ(I))
R1 = 0.0364
wR2 = 0.0815
R1 = 0.0434
wR2 = 0.0858
1.024
R1 = 0.0370
wR2 = 0.0903
R1 = 0.0528
wR2 = 0.1004
ab
,
R indices (all data)
4. Structure of trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)-
CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2(N2) (2b). The mo-
lecular structure of trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)-
CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2(N2) (2b) was
determined by single-crystal X-ray diffraction techniques, and
views of the complex are given in Figure 1. A limited amount of
crystallographic and refinement data are given in Table 2. A
relatively short d(N−N)30 value of 1.104(3) Å is observed in
concert with the ν(NN) of 2107 cm−1 observed in its IR
spectrum, and the d(Fe−NN) value of 1.811(2) is significantly
shorter than the iron−imine distance of 1.935(2) Å, but both
are within normal ranges. The iron−aryl bond distances are
1.996(2) and 2.014(2) Å, which are again short but are in range
of those for other arene-activated species derived from cis-
(Me3P)4FeMe2.100−103 The most significant metric parameter is
the C1−Fe−C14 bite angle of 163.06(9)° and its comple-
mentary C1−Fe−N2 and C14−Fe−N2 angles of 97.20(9) and
c
GOF
1.023
a
b
2
R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 = [∑w(|Fo| − |Fc|)2/∑wFo ]1/2
.
c
GOF (all data) = [∑w(|Fo| − |Fc|)2/(n − p)]1/2; n = number of
independent reflections, p = number of parameters.
99.73(9)°, respectively. The bite angle of the diarylimine is
critical, because it is consequential to σ*/dxz mixing within the
diarylimine plane that permits greater overlap and a better
energy match with the N(π*) orbital, thereby enhancing back-
bonding. The remaining core angles among adjacent sites
average 90.1(16)°, and the N1−Fe−N2 and P1−Fe−P2 angles
are 178.17(10) and 176.04(3)°, respectively.
5. Molecular Orbital View of π-Back-Bonding to N2. Figure
2 illustrates a truncated molecular orbital diagram of trans-{mer-
κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-2-
yl)}Fe(PMe3)2(N2) (2b), highlighting the “t2g”, π-back-
D
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. Truncated molecular orbital diagram of trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2(N2) (2b)
showing the greater π back-bonding from the dxz orbital relative to dyz as a consquence of σ*/dxz mixing.
bonding set of orbitals. Despite the electron-withdrawing
capability of the hexafluorinated diarylimine ligand, efficient
back-bonding to dinitrogen occurs because of the aforemen-
tioned mixing of diaryl to σ* character with dxz, rendering the
orbital better at back-bonding. If the bite angle is assumed to be
similar for the remaining imines, the stability of all the
dinitrogen complexes is well justified. Note that the diagram
also reveals a low-lying diarylimine π* orbital. It is likely that
most of the color variations of the dinitrogen and related
complexes stem from MLCT bands in the visible region that
are derived from occupation of this orbital. The σ* orbitals
to a >1000 rate of degradation difference (a rough first-order
decay was observed) that cannot be simply accounted for by
the approximate factor of 5 difference in [O2] in solution and
above. Only for 2c was the degradation rate as fast in dry air as
it was in 1 atm of O2 once the difference in [O2] was taken
account. Assuming dinitrogen dissociation is the initial step,
these experiments suggest there is a kinetic preference for
dinitrogen binding; otherwise, the dry air and 1 atm dioxygen
conditions would yield degradation t1/2 values that are near a
factor of 4. The selective, reversible binding of N2 in dry air is a
plausible explanation for the difference vs the fast degradation
in 1 atm of O2. The dinitrogen ligand, at least on 2b-15N2,
cannot be rapidly and reversibly lost, nor does Nα (δ 369.55
2
2
2
dx −y and dz are buried amidst a number of diarylimine and
iron−dinitrogen π* orbitals. The energies are given in eV, but
the values of virtual vs filled orbitals are likely to be inaccurate,
although relatively accurate within each set.106−108
relative to NH3(l)) rapidly interconvert with Nβ (δ 334.15),
1
since a J
value of 4 Hz is resolved in the 15N NMR
15N15
N
6. Stability of Dinitrogen Complexes in Air and Dioxygen.
Table 1 gives degradation t1/2 values for the disappearance of
dinitrogen complexes 2a−f under normal air (wet), dry air, and
pure dioxygen conditions, all at 1 atm. The experiments were
conducted in J. Young NMR tubes, which were exposed to the
particular atmospheric conditions, vigorously shaken, moni-
spectrum.109
It is tempting to conclude that the t1/2 values in 1 atm of
dioxygen reflect the rates of irreversible dinitrogen loss.
Without independent studies, associative or associative
interchange paths cannot be dismissed under these pseudo-
first-order conditions. A purely dissociative path might correlate
with the ν(NN) of 2a−f, as the lower stretching frequencies
could be construed as indicative of stronger binding; no such
correlation is evident, given that 2e (2058 cm−1) is one of the
swiftest to degrade under 1 atm of O2 and 2c (2102 cm−1) is
one of the slowest. It is interesting to note that {mer-κC,N,C′-
(4,5,6-(F)3-C6H-2-yl)CH2NCH(4,5,6-(F)3-C6H-2-yl)}Fe-
(PMe3)3 (1g), a tris-PMe3 derivative with no ortho
substituents, was stable to 1 atm of O2 for >2 weeks.
The degradation path in 1 atm of O2, aside from the
disappearance of 2a−f, is not obvious. No organic materials are
observed in the 1H NMR spectra of the orange solutions other
than 2a−f, and Evans method104 measurements on samples
monitored to greater than 95% completion afford μeff values
ranging from 3.1 to 3.5 μB. While 1 atm of O2 is known to
broaden signals, it typically does not make them disappear. The
possibility of paramagnetic {mer-κC,N,C′-(Ar-2-yl)CH2N
CH(Ar-2-yl)}Fe(PMe3)2(O2) complexes was considered in
1
tored by H NMR spectroscopy, and periodically refreshed.
The longest-lived samples had t1/2 values of over 2 weeks in dry
air, although a few samples persisted for only a few hours. In
normal “wet” air, samples had much shorter degradation t1/2
values of 1−2 h, suggesting that the most rapid decomposition
mode involved protonation. Free diarylimine was typically
observed to grow in as the presumed protolytic degradation
proceeded. In samples of pure O2 (1 atm), the t1/2 values were
the shortest, ranging from <10 to 30 min. Virtually no
correlation with substituents was noted, as among the most
long-lived were both electron-withdrawing and electron-
donating diarylimine chelates. As a consequence, it does not
appear that outer-sphere electron transfer (eT) to dioxygen is a
particularly credible degradation path.
Dioxygen (1.0 atm) degradation t1/2 values were in most
cases substantially shorter than those in dry air (∼0.2 atm of
O2). For example, the <20 min vs >2 weeks for 2b corresponds
E
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
view of calculations indicative of an Fe(III) superoxide
possessing a triplet GS. IR studies failed to elicit bands in the
regions expected, and crystalline material obtained from the J.
Young tubes was identified as Me3PO, consistent with oxidative
destruction. An “NMR silent” C6D6 solution of 2b was
subjected to an aqueous quench, and 1H NMR spectral analysis
of this mixture revealed Im-b and Me3PO along with a solid. It
is conceivable that the O2 degradations generate paramagnetic
LnFeOx aggregates that have a significant effect on the
observation of NMR spectra. In support of this proposal,
treatment of {mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH-
(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)3 (1b) with 1 atm of O2
generated a black solid, but no observable Me3PO or free
Im-b was observed by NMR spectroscopy.
4a-Ph, ν(CC) 2044 cm−1, eq 2) by HCCR; no attempts at
isolation were made.
8. Alternate Route to N2 Complexes via N2O. While
exploring the synthesis of diarylimine diphosphine Fe(II)
adducts, {mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH-
(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)3 (1b) and trans-{mer-
κC,N,C′-(3-OMe)-C6H3-2-yl)CH2NCH(C6H4-2-yl)}Fe-
(PMe3)3 (1d) were treated with N2O110 on NMR tube scales,
but the corresponding dinitrogen derivatives trans-{mer-
κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-2-
yl)}Fe(PMe3)2(N2) (2b) and trans-{mer-κC,N,C′-(3-OMe)-
C6H3-2-yl)CH2NCH(C6H4-2-yl)}Fe(PMe3)2(N2) (2d), re-
spectively, were formed instead according to eqs 3 and 4
7. Terminal Alkynes as Proton Donors. Attempts to view
hydrolysis intermediates via the addition of 1 equiv of H2O to
trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-
(F)3-C6H-2-yl)}Fe(PMe3)3 (1b) invariably led to complete
degradation of a portion of the starting dinitrogen complex;
hence, terminal alkynes were considered as weakly acidic
substrates that would allow observation of intermediates
derived from proton transfer. Treatment of 1b with HCCR
(R = Me, Ph) afforded trans-{κC,N-(3,4,5-(F)3-C6H2)CH2N
CH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)3(CCR) (R = Me, 4b-Me,
69%, ν(CC) 2081 cm−1; R = Ph, 4b-Ph, 79%, ν(CC) 2050
cm−1) as dark red solids. As eq 1 indicates, protonation of
(>95%). Compound 2b was subsequently found to be a very
modest catalyst for the conversion of PMe3 to Me3PO with
N2O with a pseudo-first-order rate constant of ∼4 × 10−4 s−1 at
23 °C (eq 5). This is roughly a factor of 500 faster than the
uncatalyzed rate constant of ∼8 × 10−7 s−1 under identical
conditions.
Since 2b failed to catalyze oxygen atom transfer from N2O to
any useful substrates (i.e., olefins etc.), the reactivity was not
examined further. A related nitrene transfer64,111,112 from tosyl
azide to PMe3 was noted (eq 6), and the reaction proved to be
somewhat sporadic, but useful. The alternative nitrous oxide
method failed to generate the dinitrogen complex from trans-
{mer-κC,N,C′-(4,5,6-(F)3-C6H-2-yl)CH2NCH(4,5,6-(F)3-
C6H-2-yl)}Fe(PMe3)3 (1g), but the nitrene transfer reagent
cleanly affected the transformation on an NMR tube scale (eq
7). The generation of trans-{mer-κC,N,C′-(4,5,6-(F)3-C6H-2-
yl)CH2NCH(4,5,6-(F)3-C6H-2-yl)}Fe(PMe3)2(N2) (2g)
permitted measurement of its ν(NN) value of 2070 cm−1,
which revealed that having two o-F substituents (cf. 2b, 2107
cm−1) rather than one meta and one para renders the
diarylimine of 2b substantially more inductively withdrawing.
the(3,4,5-(F)3-C6H-2-yl)CH2− group has occurred to generate
a dangling benzyl group attached to the κC,N-phenylimine
residue. The three phosphines remain in a meridional
configuration, while the acetylides are likely to occupy the
sterically hindered site opposite the aryl and between the imine
and unique PMe3, although the NMR spectra do not
distinguish this structure from one in which the acetylide and
PMe3 positions are reversed. Note that the slightly longer Fe−
Ar bond (2.014(2) Å relative to 1.996(2) Å) has been
protonated, but whether this is a kinetic or thermodynamic
product was not ascertained. Exposure to excess HCCR
removed the diarylimine entirely and afforded trans-(Me3P)4Fe-
(CCR)2 as one of the products. Protonation of the Fe−Ar bond
on the benzylic side of the diarylimine seems general, as trans-
{mer-κC,N,C′-(3,5-(CF3)2-C6H2-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)3 (1a) was also converted to trans-{κ-C,N-
(3,5-(CF3)2-C6H3)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe-
(PMe3)3(CCR) (R = Me, 4a-Me, ν(CC) 2077 cm−1; R = Ph,
F
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
H) value of 0.77 Å,115−117 assuming rapid rotation of the
dihydrogen unit. The dihydrogen complex could not be
isolated, due to a slow process that leads to hydrogenation of
the Fe−Ar bonds and production of Im-b along with the
release of PMe3.
Ammonia adduct 6b was isolated in 98% yield as a purple
solid after repeated exposure of 2b to NH3 in benzene. Pyridine
derivatives could not be obtained from the addition of RC6H4N
to 2b, but 1b was treated with pyridine N-oxide to provide 7b
(L = py); this was subsequently shown to be unstable under N2,
and conditions could not be found to measure a Kc. Pyridine
adduct 7b was isolated in 70% from 1b generated in situ from
(Me3P)4FeMe2 and Im-b, and was obtained as red micro-
crystals.
3. Related Adducts. Attempts to utilize the diarylimine bis-
phosphine Fe(II) platform to generate double bonds was
successful in regard to π-back-bonding, as dinitrogen, carbonyl,
and isocyanide complexes have revealed. The red-brown methyl
isocyanide complex 8b (ν(CN) 2073 cm−1), shown in eq 8, was
Related Diarylimine Fe(II) Adducts. 1. Carbonyl De-
rivatives. As shown in Scheme 1, a select number of related
carbonyl derivatives were generated, some via isolation and
others on NMR tube scales, as a check on how the substituents
affected the ν(CO) values in the infrared spectra. Table 1
indicates that the ν(CO) of {mer-κC,N,C′-(Ar-2-yl)CH2N
CH(Ar-2-yl)}Fe(PMe3)2(CO) (3a−d,f) reveal a trend identical
with that of the dinitrogen derivatives, with roughly the same
spread in values. The values range from 1950 to 1882 cm−1,
which are fairly low for Fe(II) species, and provide another
indication of the unique back-bonding situation illustrated in
Figure 2.
2. NH3, PMe3, and H2 Binding Studies. Scheme 2 shows
rough equilibrium constants (Kc; all components relative to 1
M) measured for the formation of trans-{mer-κC,N,C′-(3,4,5-
(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe-
(PMe3)2(N2) (2b) from trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-
yl)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)3 (1b) and the
generation of ammonia and dihydrogen complexes trans-{mer-
κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-2-
yl)}Fe(PMe3)2L (L = H2, 5b; L = NH3, 6b) via 2b. The
dinitrogen, ammonia, and dihydrogen113−115 complexes all had
similar relative free energies, with the PMe3 complex about 1.6
kcal/mol above the dinitrogen complex.
Dihydrogen complex 5b was characterized by a broad triplet
1
(JPH = 12 Hz) at δ −13.88 in its H NMR spectrum. Variable-
temperature T1 measurements were conducted in toluene-d8,
and the T1(min) value was found at 228 K, affording an r(H−
Scheme 2
G
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 3
Figure 3. (a) Molecular view (50% probability ellipsoids) of trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe-
(PMe3)2Cl (12b) and (b) a view down the P−Fe−P axis with the phosphine methyl groups removed. Selected bond distances (Å) and angles (deg):
Fe−C1, 2.031(2); Fe−C8, 2.028(2); Fe−N1, 1.9389(17); Fe−Cl, 2.2306(6); Fe−P1, 2.2602(6); Fe−P2, 2.2463(6); N1−C7, 1.299(3); N1−C14,
1.460(3); C6−C7, 1.434(3); C13−C14, 1.497(3); N1−Fe−Cl, 178.04(5); C1−Fe−C8, 160.70(9); P1−Fe−P2, 174.29(2); N1−Fe−C1, 79.80(8);
N1−Fe−C8, 80.96(8); N1−Fe−P1, 91.90(5); N1−Fe−P2, 93.74(5); Cl−Fe−P1, 86.64(2); Cl−Fe−P2, 87.75(2); C1−Fe−P1, 89.88(6); C8−Fe−
P1, 89.23(6); C1−Fe−P2, 92.02(6); C8−Fe−P2, 90.74(6); C1−Fe−Cl, 98.88(7); C8−Fe−Cl, 100.31(6); Fe−C1−C2, 133.60(18); Fe−C1−C6,
111.90(16); Fe−C8−C9, 131.47(16); Fe−C8−C13, 114.41(15).
prepared in 69% yield. Unfortunately, the generation of Fe(IV)
species where the sixth ligand is a π donor has not met with
much success. Azide and diazo reagents64 did not react with
dinitrogen complexes under mild conditions, and higher
temperatures resulted in degradation. Only in the case of
diphenyldiazomethane was a dark purple product isolated, and
its spectral characteristics were consistent with an adduct,118
trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-
(F)3-C6H-2-yl)}Fe(PMe3)2(NN-CPh2) (9b, 76%, eq 9).
Oxidations to Fe(III) Complexes. 1. Outer-Sphere eT.
Fe(III) species containing metal−carbon bonds are still
relatively uncommon,98 and no Fe(III) derivative has been
shown to bind dinitrogen; hence, 1e oxidations of the
dinitrogen complexes were explored, as shown in Scheme 3.
Oxidation of trans-{κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2N
CH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2(N2) (2b) with [Cp2Fe]-
BF4 in the presence of LiI caused the release of N2, and the
ferric iodide complex trans-{κC,N,C′-(3,4,5-(F)3-C6H-2-yl)-
CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2I (11b) was iso-
lated as a dark green solid in 42% yield. When the oxidation
was conducted in the absence of LiI, the red solution darkened
but there was no precipitate and no indication of ν(NN) in the
IR spectrum of the solution. A subsequent addition of LiI
generated 11b; thus, it is likely that the THF adduct 10b was
the initial product in this sequence.
2. Inner-Sphere eT. While the iodide complex 11b was
generated via outer-sphere electron transfer (eT), the plausible
lability of the dinitrogen in the ferrous N2 complexes suggested
that inner-sphere paths were also available. Inner-sphere metal-
based reagents capable of reacting in nonpolar media are
H
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 3. Calculated (M06/6-311+G(d)) Binding Energies for trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2 (1′b) + L (L = N2, NH3, PMe3, H2) and Related Reactions (P′ = PMe3)
uncommon; therefore, Ph3CCl was used to convert 2b and
trans-{κC,N,C′-(3,5-(CF3)2-C6H2-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2(N2) (2a) to their corresponding Fe(III)
chloride derivatives 12b and 12a, respectively, in very good
yields. Each halide complex had a tractable 1H NMR spectrum,
and Evans method104 measurements showed each to have an S
= 1/2 ground state with a significant spin−orbit contribution:105
11b, μeff = 2.0(1) μB; 12b, μeff = 2.0(1) μB; 12a, μeff = 2.1(1) μB.
3. Structure of trans-{κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2N
CH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2Cl (12b). Selected crystallo-
graphic and refinement data pertaining to the X-ray structure of
trans-{κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2Cl (12b) are given in Table 2, and views
of the molecule are provided in Figure 3. The iron−imine
distance of 1.9389(17) Å is essentially the same as in 2b, while
the iron−carbon bonds are slightly longer (0.02−0.04 Å) at
2.031(2) and 2.028(2) Å, as are the d(Fe−P) values of
2.2602(6) and 2.2463(6) Å. It is plausible that the contraction
of the 3d orbitals upon oxidation, while typically causing
shorter bond distances in molecules with greater ionic
character, renders covalent interactions longer due to decreased
orbital overlap.98 As the Fe−Ar distances increase, the bite
angle of the diarylimine ligand decreases to 160.70(9)°, and the
related N1−Fe−C(Ar) angles are 79.80(8) and 80.96(8)°,
while their complementary C1−Fe−Cl and C8−Fe−Cl angles
are 98.88(7) and 100.31(6)°, respectively. Remaining core
angles among adjacent ligands average 90.2(23)°, and the N1−
Fe−Cl and P1−Fe−P2 angles are 178.04(5) and 174.29(2)°,
respectively; hence, the core is very similar to that of 2b.
similarly, while binding of PMe3 and H2 is significantly less
favorable. With respect to the experimental relative binding
studies, dihydrogen is the outlier, as it is calculated to be 9.9
kcal/mol less favorable than N2 binding, while experimentally it
is found to be roughly equal to N2 and NH3. Dinitrogen and
dihydrogen are often found to be comparable ligands.113−115
PMe3 is unfavorable, and the calculation again appears to
overestimate its relative binding energy. A closer look at free
energies of binding pertaining to N2 vs PMe3 reveals the
difference to be entropic in origin, as the relative binding of
PMe3 manifests a ΔΔS° value of 16.5, which can be construed
as reflecting the heavy steric penalty as the phosphine abuts the
o-fluorine substituents.
The binding assessments continued with a calculation of
dinitrogen binding to trans-{mer-κC,N,C′-(4,5,6-(F)3-C6H-2-
yl)CH2NCH(4,5,6-(F)3-C6H-2-yl)}Fe(PMe3)2 (1′g), which
does not contain o-fluorines. The binding of N2 to this five-
coordinate fragment (ΔG°calc = −19.4 kcal/mol) was calculated
to be enthalpically (−2.9 kcal/mol) and entropically (5.7 eu)
more favorable than to 1′b (ΔG°calc = −14.8 kcal/mol), a clear
indication of how the o-fluorines inductively destabilize
dinitrogen binding and have a deleterious entropic effect
(d(o-F···Nbnd) = 2.9 Å; d(o-F···Nterm) = 3.1 Å) on even a linear
ligand such as N2. In order to make sure these arguments are
not compromised by the unique dispositions of the fluorine
substituents in 2b,g, the hypothetical unsubstituted diarylimine
iron dinitrogen complex was compared to one containing just
two o-fluorines, and the results are quite similar.
Spin State of (diarylimine)Fe(PMe3)3 (1a−g). The
precursors to the dinitrogen complexes, tris-phosphine species
{mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar-2-yl)}Fe(PMe3)3 (1a−
f), were shown to be diamagnetic when the diarylimine ortho
substituents were two fluorines (1b) or one methoxide (1d).
Paramagnetic properties (S = 1) were observed for diarylimine
ortho substituents that were one fluorine and one CF3 (1a),
one CF3 (1c), and one methyl group (1e,f). Since a methoxy
group can be considered smaller than a methyl due to its
orientation, a correlation between paramagnetism and larger
substituents exists, highlighting the influence of steric features
on electronic structure.
DISCUSSION
■
Relative Binding Energies. In order to assess the steric
and electronic factors present in diarylimine Im-b, calculations
on the binding energies of L (L = N2, NH3, PMe3, H2) to the
putative five-coordinate trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-
yl)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2 (1′b) were
conducted, and the results are given in Table 3. Initial efforts
utilizing B3LYP ran counter to the experimental observations,
and a switch to the M06119/6-311+G(d)120 functional121 was
rationalized on the basis of the need for better calculations of
van der Waals interactions: i.e., the sterics of binding. The
calculations are reasonable, showing that N2 and NH3 bind
Calculations on {mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)-
CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)3 (1b) were con-
I
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 4. Calculated Metric Parameters for the Singlet, Triplet, and Quintet States of {mer-κC,N,C′-(3,4,5-(F)3-C6H-2-
yl)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)3 (1b)
spin state
S = 0
2.07, 2.08
S = 1
2.04, 2.06
S = 2
2.20, 2.21
d(Fe−C) (Å)
d(Fe−Ptr) (Å)
d(Fe−Pcis) (Å)
d(Fe−N) (Å)
2.26, 2.29
2.36
2.26, 2.26
3.18
2.57, 2.59
2.69
1.97
2.15
2.20
∠C−Fe−C (deg)
∠C−Fe−N (deg)
∠Ptr−Fe−Pcis (deg)
∠Ptr−Fe−Ptr (deg)
∠Ptr−Fe−N (deg)
∠Pcis−Fe−N (deg)
∠Ptr−Fe−C (deg)
∠Pcis−Fe−C (deg)
159.7
155.8
150.8
80.1, 80.1
89.6, 93.6
173.8
77.9, 77.9
87.0, 87.0
172.8
75.2, 75.7
89.5, 91.0
176.9
86.6, 90.7
173.5
93.1, 93.4
175.8
88.7, 91.1
174.9
85.9, 88.2, 90.5, 94.5
95.1, 105.1
89.1, 89.4, 92.2, 92.2
97.9, 106.2
87.9, 89.1, 91.3, 91.6
99.7, 109.5
ΔErel (kcal/mol)
ΔHrel (kcal/mol)
ΔGrel (kcal/mol)
−7.0
−4.7
3.9
−3.0
−2.3
2.3
0.0
0.0
0.0
ducted for S = 0, 1, 2, and the metric parameters and energies
of these spin states are given in Table 4. There is a general
lengthening of bond distances from singlet to triplet to quintet
and a corresponding change in bond angles. Although the
observed GS of 1b is S = 0, the singlet state is calculated to be
(1′b) appears to be kinetic in origin, since prolonged exposure
of the N2 complexes to O2 causes degradation, and no clean
dioxygen complexes were identified. In addition, exposure of 1b
to O2 produced Me3PO and a black solid, consistent with
irreversible O2 binding and subsequent decomplexation. Recall
that stability in air is significant, even in the worst cases,
implicating reversible trapping by dinitrogen and irreversible
destruction by dioxygen.
The results of calculations on the binding of N2 vs O2 to
trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-
(F)3-C6H-2-yl)}Fe(PMe3)2 (1′b) are provided in Figure 4, and
some insights are illustrated, albeit with certain assumptions.
First, note that the N2 complex, in short form denoted as
1[FeN2], is calculated to be 2.7 kcal/mol more stable than a
dioxygen complex that is best construed as a ferric superoxide
1
1
2
1.8 kcal/mol above the triplet state ((dxy) (dz ) ) in free energy
a n d 3 . 9 k c a l / m o l a b o v e t h e q u i n t e t s t a t e
2
1
1
1
1
2
2
2
((dyz) (dxz) (dxy) (dz ) (dx −y ) ). While the calculated free
energies obviously do not conform to experiment, the ΔErel
and ΔHrel values parallel the experimental observations, at least
in terms of predicting diamagnetic behavior. Relative energy
calculations of comparative closed-shell vs open-shell systems
can present problems for density functional methods;106−108
hence, the important outcome of the calculations is the
relatively similar values of the three states. Moreover, the
calculated entropic corrections, which the calculations suggest
are responsible for tipping the free energy balance toward the
high-spin state, may be exaggerated in a gas-phase calculation.
The Fe−P bond distances of the phosphine opposite the
imine-N change dramatically from S = 0 (2.36 Å) to the S = 1
3
3
−
species, [FeO2] or [FeIII(O2 )]. In order to produce either
3
species, the five-coordinate precursor [Fe] (i.e., 1′b) must
intersystem cross, since it possesses a triplet ground state (GS)
that is (dxy)1(dz2)1, which can be thought of in orbital symmetry
terms as a π1σ1. The product [FeN2] has an abbreviated
1
2
(3.18 Å) state as dz becomes half-occupied, while the d(Fe−
electron configuration of (dxz)2(Nσ)2 which is essentially π2σ2;
hence, the conversion of 1′b + N2 (π1σ1 + σ2) to 2b is not
orbital symmetry allowed, nor is it spin allowed. The GS of
Ptr) values remain the same. Since the calculated GS of the
putative five-coordinate intermediate, 1b′, is a triplet,
intersystem crossing from 1b to 1b′ probably occurs smoothly
as the iron−phosphine bond is elongated, and the barrier to
PMe3 loss is likely to be close to the calculated binding free
energy of 8.0 kcal/mol. A similar observation has been made
regarding spin state changes in dinitrogen bonding to related
Fe(II) compounds.95
1
3[Fe] + N2 correlates to an excited state (ES) on the bound
substrate side (left) of Figure 4 that is 20.9 kcal/mol above
1
1[FeN2]. The [Fe] or (dxz)2 configuration of 1′b is only 12.6
kcal/mol above the GS, and the intersystem crossing point
1
(ICP) to afford 1b (1[FeN2]) from 1′b (3[Fe] + N2) is the
lowest for either substrate, requiring only the conversion of HS
Fe(II) to LS Fe(II) as the main electronic obstacle to
surmount. Cases of N2 bonding that incur an intersystem
crossing event have been similarly rationalized.95−97
Origin of the Selectivity of N2 Extraction over O2 from
Air. The selectivity for binding dinitrogen over dioxygen to a
five-ccordinate fragment such as trans-{mer-κC,N,C′-(3,4,5-
(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2
J
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. Correlation diagram for the binding of dinitrogen vs dioxygen to five-coordinate trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2N
CH(3,4,6-(F)3-C6H-2-yl)}Fe(PMe3)2 (1′b), showing the lower intersystem crossing point (ICP) for the former. Since the quintet state of the bound
O2 complex was not located, the ICP for dioxygen binding is somewhat arbitrary; see the text for a discussion of its placement.
1
1
The binding of dioxyen to GS 1′b (3[Fe] + 3O2) is
complicated, in part due to the inability to locate a geometry
pertaining to an O2-bound quintet state to which it would
A simple pairing of the Fe((dz ) ) and O(σ ) spins affords the
calculated ground-state Fe(III) superoxide species 3[FeO2]
((dxz)1(Oσ)2(Oπ)1). The ICP that arises from this correlation
is significantly higher than that for N2, which accounts for the
N2 vs O2 selectivity, but its value is again dependent on the
nature of the quintet surface.
2
1
1
2
correlate. The GS may thus be construed as (dxy) (dz ) +
π*1π*1, and its conversion to the Fe(III) superoxide species
3[FeO2] ((dxy)1(Oσ)2(Oπ)1) is slightly endergonic (+2.7 kcal/
mol). No steric interference to dioxygen binding was noted in
the calculations; hence, the additional thermodynamic prefer-
ence for dinitrogen binding stems purely from electronic
factors. It is plausible that the strong field imparted by the
phosphines and diaryl ligands renders the ferrous centers less
susceptible to oxidation and thus endergonic with respect to the
calculated Fe(III) superoxide. While the lowest spin state
1
In summary, while the [FeN2] complex (e.g., 2b) is 4.9
kcal/mol below the calculated first-formed dioxygen adduct,
there is a significantly greater kinetic preference for N2 binding.
The kinetic preference likely stems from both spin and orbital
symmetry constraints. A quintet surface that would correlate
3
3
directly with the reactant [Fe] + O2 surface has not been
3
3
located and may indicate that O2 cannot bind to [Fe]; such
an adduct would not optimize. This is not surprising, because
an octahedral Fe(II) center that is S = 1 would have an electron
in a σ* orbital and would likely be unstable relative to an
1
3
correlation occurs from [Fe] + O2 at 12.6 kcal/mol, the
orbital symmetry of this state, which is (dxz)2(π*π*) or
(π2)(π*π*), does not correlate with the orbital symmetry of the
bound-O2 GS. It also seems unlikely that the ICP derived from
this “spin correlation” would serve to significantly differentiate
between N2 and O2 because it is likely to be only slightly higher
than the ICP for dinitrogen binding, although it is highly
dependent on where and if a suitable quintet surface exists. This
lowest O2-bound state best correlates with a high -energy 3[Fe]
+ 1O2 state in which the singlet oxygen 1(π*1π*) configuration
−
electron transfer to afford a ferric state and O2 . It is precisely
such an event that would lead to the calculated superoxide
product, but finding this ICP without a ready correlation is
difficult. The orbital symmetry of the calculated superoxide
3
complex denoted as [FeO2] correlated with a high-energy
reagent surface; hence, it is quite likely that the ICP is high in
energy. Since a dioxygen adduct could not be isolated or
observed spectroscopically due to rapid degradation, this
(1Σ+ ) can be considered (π*1σ1) along the reaction coordinate.
g
K
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
1
Figure 5. Key for H, 13C{1H}, 31P, and 19F NMR assignments.
Chemical shifts are reported relative to benzene-d6 (1H, δ 7.16;
13C{1H}, δ 128.39), THF-d8 (1H, δ 3.58; 13C{1H}, δ 67.57), and
CD2Cl2 (1H, δ 5.32; 13C{1H}, δ 54.00). Infrared spectra were recorded
on a Nicolet Avatar 370 DTGX spectrophotometer interfaced to an
IBM PC (OMNIC software). UV−vis spectra were obtained on an
Ocean Optics USB2000 spectrometer. Solution magnetic measure-
ments were conducted via the Evans method in benzene-d6.104
Elemental analyses were performed by Robertson Microlit Labo-
ratories, Madison, NJ, or Complete Analysis Laboratories, Inc.,
Parsippany, NJ.
analysis is predicated on the calculated initially formed
dioxygen species, the Fe(III) superoxide [FeO2].
3
CONCLUSIONS
■
Due to steric interactions, ortho-substituted diarylimine ligands
cause ready dissociation of PMe3 from {mer-κC,N,C′-(Ar-2-
yl)CH2NCH(Ar-2-yl)}Fe(PMe3)3 (1a−f) complexes, and
the resulting five-coordinate species, {mer-κC,N,C′-(Ar-2-yl)-
CH2NCH(Ar-2-yl)}Fe(PMe3)2 (1′a−f), selectively extract
dinitrogen from air. The compounds eventually degrade,
rapidly in wet air due to irreversible protonation reactions,
but far less swiftly in dry air, where the selective recombination
of 1′a−f and dinitrogen renders the systems significantly stable
in the presence of dioxygen. Calculations show the dinitrogen
complexes {mer-κC,N,C′-(Ar-2-yl)CH2NCH(Ar-2-yl)}Fe-
(PMe3)2N2 (2a−f) are kinetically and thermodynamically
preferred over the calculated dioxygen adduct, a ferric
superoxide, while experiments are consistent with rapid
degradation of the iron organometallic species subsequent to
dioxygen binding. Dinitrogen, ammonia, and dihydrogen bind
similarly to the five-coordinate species 1′b, while PMe3 and
pyridine are disfavored, principally due to the aforementioned
steric features. Ligands with π-accepting capability bind
strongly, but formation of iron multiple bonds could not be
effected. Ferric derivatives generated via inner (2a,b) and outer-
sphere oxidations of (2b) do not bind N2. It is conceivable that
a chemical means to remove dinitrogen from hydrocarbon feed
streams can be designed on the basis of the principles
discovered, even with trace dioxygen present. Furthermore,
perhaps the μ6-carbide present in the operational cluster of
nitrogenase, i.e., [(Cys)Fe(μ3-S)3Fe3(μ-S)3(μ6-C)Fe3(μ3-
S)3Mo(His)(Homocitrate)]n,122 helps impart a strong field
and an electronic influence on dinitrogen bonding, akin to the
species herein.
Procedures. 1. General Procedure for Synthesis of Imines. To a
suspension of MgSO4 (5−8 equiv) in CH2Cl2 were added 1.5 mmol of
aldehyde and 1.5 mmol of amine. After it was stirred for 12 h, the
mixture was filtered and concentrated to yield a colorless to pale
1
yellow oil in >98% purity (by H NMR). All aryl positions are CH
unless noted. For Im-a and Im-c−Im-g, see the Supporting
Information.
Im-b (b = d = e = k = l = m = CF). 1H NMR (C6D6, mult, J (Hz);
assignt): δ 7.71 (td, 9, 7; c), 6.24 (td, 10, 6; f), 7.98 (s, g), 3.91 (s, h),
6.51 (dd, 8, 7; j, n). 13C{1H} NMR (C6D6, mult, JCF (Hz); assignt): δ
135.89 (a), 157.63 (d, 250; b), 105.89 (d, 250; c), 151.63 (d, 250; d),
152.37 (d, 250; e), 115.25 (f), 153.35 (g), 62.91 (h), 120.47 (i), 111.64
(j, n), 147.70 (d, 250; k, m), 139.02 (d, 250; l). 19F NMR (C6D6, mult,
J (Hz); assignt): δ −123.84 (ddd, 16, 11, 6; b), −128.48 (dtd, 22, 9, 5;
d), −134.76 (dd, 21, 8; m,k), −141.26 (tdd, 20, 14, 7; e), −163.20 (tt,
21, 7; l).
2. General Procedure for Tris-PMe3 Complexes. Since 1a−f cannot
be exposed to dinitrogen, spectral assays of the (diarylimine)Fe-
(PMe3)3 complexes were conducted on NMR tube scale reactions. In a
typical reaction, ∼8 mg (0.020 mmol) of (Me3P)4FeMe2 and 0.020
mmol of imine were loaded into a J. Young NMR tube into which
∼0.6 mL of C6D6 was vacuum-transferred. The reactions went to
completion in 1−6 h, and the contents were never exposed to
dinitrogen. Evans method measurements were conducted on the crude
paramagnetic compounds generated in situ. For 1c−f, see the
Supporting Information.
1
1a (b = d = e = CF, kp = mq = CCF3). H NMR (C6D6): δ 66.13
(2H), 34.01 (2H), 32.57 (2H), −4.86 (18H), −11.46 (9H). Free
PMe3 was noted at δ 0.80. 31P{1H} NMR (C6D6): δ −60.0 (br s, free
PMe3); no other resonances were noted. μeff(296 K) = 3.0, 3.3 μB.
1b (b = d = e = k = l = m = CF). 1H NMR (C6D6, mult, assignt): δ
6.39 (s, c, j), 8.27 (s, g), 4.24 (s, h), 0.46 (s, r) 1.36 (s, s). 13C{1H}
NMR (C6D6, mult, JCF (Hz); assignt): δ 117.39 (a), 161.49 (d, 230;
b), 103.11 (c), 157.94 (d, 230; d), 159.10 (d, 250; e), 111.40 (f),
163.13 (g), 65.59 (h), 111.77 (i), 96.50 (j), 146.44 (d, 250; k), 144.97
(d, 250; l), 157.54 (d, 250; m), 111.26 (n), 16.97 (t, 10; r), 16.08 (d,
13; s). 19F NMR: (C6D6, mult, JCF (Hz); assignt): δ −106.35 (d, 32;
m), −147.15 (dd, 30, 8; d), −120.42 (dd, 21, 7; k) −166.60 (dd, 33,
20; p), −116.08 (dd, 25, 10; b), −133.28 (dd, 32, 9; e). 31P{1H} NMR
(C6D6, mult, JCF (Hz); assignt): 14.14 (d, 58; o), 9.60 (t, 58; p).
3. General Procedure for Dinitrogen Complexes. In a 100 mL
bomb reactor charged with Fe(PMe3)4Me2 (0.200 g, 1.02 mmol) and
imine (1 equiv) was transferred 15 mL of benzene. The mixture was
placed under an atmosphere of N2 at 23 °C. The solution was stirred
for 6 h. Upon removal of solvent, the crude mixture was dissolved in
Et2O, the solution was filtered, and the residue was washed (4 × 10
mL of Et2O). Crystallization from hexanes at −78 °C afforded
product. For 2a,c−g, see the Supporting Information.
EXPERIMENTAL SECTION
■
General Considerations. Since a number of the preparations
pertaining to different diarylimine complexes are repetitive, this section
focuses on those pertaining to Im-b. For the remaining complexes and
spectral data, see the Supporting Information.
All manipulations were performed using either glovebox or high-
vacuum-line techniques. Hydrocarbon solvents containing 1−2 mL of
added tetraglyme and ethereal solvents were distilled under nitrogen
from purple sodium benzophenone ketyl and vacuum-transferred from
the same prior to use. Benzene-d6 and toluene-d8 were dried over
sodium, vacuum-transferred, and stored under N2. THF-d8 was dried
over sodium benzophenone ketyl. Methylene chloride-d2 was dried
over CaH2, vacuum-transferred and stored over activated 4 Å
molecular sieves. Fe(PMe3)4Me2 was prepared according to a literature
procedure.99 Compounds 1d,h,i were previously reported.98 All other
chemicals were commercially available and used as received. All
glassware was oven-dried.
NMR spectra (see Figure 5 for assignment key) were obtained using
Mercury-300 and INOVA 400, 500, and 600 MHz spectrometers.
L
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
2b (b = d = e = k = l = m = CF). Dark red crystalline 2b (0.184 g)
was obtained in 67% yield. H NMR (C6D6, mult, assignt): δ 7.79 (s,
(d, 260; l), 16.65 (t, 12; r), 23.39 (d, 20; s), 120.89 (q), 121.30 (r),
130.66 (s), 129.97 (Co), 128.33 (Cm), 123.35 (Cp). 19F NMR (C6D6,
mult, JCF (Hz); assignt): δ −119.07 (dd, 33, 10; d), −160.97 (tt, 21, 7;
l), −132.10 (dd, 33, 7; e), −133.68 (dd, 22, 8; k, m), −118.30 (dd, 20,
8; b). 31P{1H} NMR (C6D6, mult, JCF (Hz); assignt): δ 15.68 (“d”, 61;
r), 13.46 (td, 62, 12; s). Anal. Calcd for C31H38F6P3NFe: C, 54.17; H,
5.57; N, 2.04. Found: C, 54.36; H, 5.41, N, 2.28. IR: ν(CC) 2020
cm−1.
1
g), 6.36 (m, c, j), 3.95 (s, h), 0.44 (s, r). 13C{1H} NMR (C6D6, mult,
JCF (Hz); assignt): δ 133.26 (a), 160.4 (d, 230; b), 97.34 (c), 157.19
(d, 230; d), 157.93 (d, 230; e), 112.47 (f), 161.98 (g), 64.04 (h),
142.75 (i), 104.45 (j), 148.17 (d, 240; k), 138.46 (d, 240; l), 150.80 (d,
240; m), 111.07 (n), 12.90 (t, 12; r). 19F NMR (C6D6, mult, JCF (Hz);
assignt): δ −119.36 (d, 31; m), −145.94 (dd, 19, 11; k), −118.92 (dd,
24, 11; b), −166.13 (dd, 31, 20; l), −128.81 (dd, 30, 25; e), −132.18
(ddd, 30, 10, 3; d). 31P{1H} NMR (C6D6, mult, assignt): δ 19.95 (s, r).
15N NMR (1b-15N2, 60.8 MHz, referenced to NH3(l)): δ 334.15 (br d,
7. trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2(H2) (5b). a. Observation of 5b. Into a J. Young
tube (2.1 mL volume) was added 2b (20 mg, 0.037 mmol) in C6D6 or
toluene-d8. The tube was degassed by multiple freeze−pump−thaw
cycles, and dihydrogen (660 Torr) was added at 23 °C. The reaction
3
1
2
15
1J N = 4 Hz, J NP < 1.5 Hz, Nβ), 369.55 (q, J N = 5 Hz, J
=
15
15N15
15N15
NP
5 Hz, Na).109 IR (C6D6): ν(N2) 2107 cm−1. Anal. Calcd for
C20H23N3F6FeP2: C, 44.72; H, 4.32; N, 7.82. Found: C, 44.85; H,
4.22; N, 7.88.
1
1
was monitored by H and 19F NMR spectroscopy. H NMR (C6D6,
mult, JCF (Hz); assignt): δ 8.08 (s, g), 6.48 (m, c, j), 4.24 (s, h), 0.22 (t,
3; r), −13.88 (t, 12; s). 19F NMR (C6D6, mult, JCF (Hz); assignt): δ
−107.83 (d, 35; e), −118.85 (m, b, m), −132.15 (dd, 31, 9; d),
−146.57 (dd, 20, 4; k), −166.00 (ddd, 35, 19, 7; l). 31P NMR (C6D6):
δ 20.54.
4. General Procedure for Carbonyl Complexes. In a 100 mL bomb
reactor charged with (PMe3)4FeMe2 (0.100 g, 0.513 mmol) and imine
(1 equiv) was transferred 15 mL of benzene at −78 °C. The solution
was warmed to 23 °C and stirred for 4 h. Upon removal of solvent and
excess PMe3, the crude mixture was redissolved in benzene, placed
under an atmosphere of dry CO, and stirred for 12 h. Solvent and
excess CO was removed from the bomb reactor via vacuum transfer,
and the crude solid was dissolved in Et2O, filtered, and washed (3 × 10
mL) with Et2O. Crystallization from hexanes at −78 °C afforded
product. For 3a,c,d,f, see the Supporting Information.
3b (b = d = e = k = l = m = CF). Dark red microcrystals were (0.110
g) isolated in 80% yield. 1H NMR (C6D6, mult, assignt): δ 6.36 (m, c,
j), 7.96 (s, g), 4.18 (s, h), 0.44 (s, r). 13C{1H} NMR (C6D6, mult, JCF
(Hz); assignt): δ 149.13 (a), 158.93 (d, 240; b), 97.81 (c), 158.08 (d,
240; d), 154.57 (d, 240; e), 142.67 (f), 160.95 (g), 63.40 (h), 147.26
(i), 104.59 (j), 157.15 (d, 240; k), 151.83 (d, 240; l), 156.24 (d, 240;
m), 132.88 (n), 14.72 (t, 13; r), 190.97 (s). 19F NMR (C6D6, mult, JCF
(Hz); assignt): δ −110.14 (d, 30; m), −145.40 (dd, 28, 8; k), −118.70
(ddd, 22, 9, 4; b), −120.92 (ddd, 30, 20, 5; e), −165.24 (dd, 31, 22;, l),
−131.28 (ddd, 30, 10, 4; d). 31P{1H} NMR (C6D6, mult, assignt): δ
20.54 (s, r). IR (C6D6): ν(CO) 1936 cm−1.
5. trans-{κC,N-(3,4,5-(F)3-C6H2)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe-
(PMe3)3(CCMe) (4b-Me). In a 25 mL round-bottom flask charged with
Fe(PMe3)4Me2 (0.050 g, 0.128 mmol) and Im-b (0.038g, 0.128
mmol) was transferred an 8 mL amount of benzene. The reaction
mixture was stirred for 4 h at 23 °C. Propyne was added to the
reaction mixture via gas bulb (0.128 mmol), and it was stirred for 10 h.
Upon removal of solvent, the crude solid was filtered and washed with
Et2O (3 × 5 mL). Dark red 4b-Me was isolated (0.055 g) in 69% yield.
1H NMR (C6D6, mult, assignt): δ 6.30 (m, c), 8.23 (s, g), 4.79 (s, h),
b. T1(min) Measurement. 5b was prepared in toluene-d8 and
allowed to equilibrate for 48 h. 1H NMR spectra were recorded at 500
MHz, and temperature calibration was performed for each measure-
ment (T (K), T1 (ms)): 298, 21; 288, 18; 278; 16; 268, 15; 258, 13;
238, 11; 218, 11; 198, 15. The T1(min) value of 10.7 ms (226 K) was
obtained by plotting ln T1 (ms) vs 1/T (K−1) and fitting with linear
regression. The d(H−H) value of 0.77 Å was calculated by assuming
rapid rotation of H2 and using the following equations: dipolar
relaxation, 1/T1 = 0.3γH4(h/2π)2(J(ω) + 4J(2ω))/rHH6; spectral
density function, J(ω) = Aτ/(1 + ω2τ2), where A = 0.25 for rapid
rotation. The temperature dependence of the correlation time is τ = τ0
exp[Ea/RT], and at T1, τ = 0.62/(2πν); simplifying, rHH = 4.611(T1
(min)/ν)1/6
.
c. Kc Measurement. 5b was prepared from 2b in C6D6 as above and
allowed to equilibrate for 48 h. Kc was calculated by direct integration
of 2b, 5b, and H2, and the amount of N2 in solution was estimated
from the Henry’s law constant of N2 in benzene and assuming the total
amount of N2 (gas and solution) was equal to that of 5b.
8. trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2NH3 (6b). In a 100 mL bomb charged with 2b
(0.100 g) was transferred 15 mL of benzene at −78 °C. An excess of
ammonia dried over sodium was transferred to the bomb at −78 °C.
The bomb was slowly warmed to 23 °C, and the contents were stirred
for 0.5 h. The solution turned from yellow-red to bright red and
eventually bright purple. The excess ammonia and benzene were
removed in vacuo. The addition of benzene and excess ammonia was
repeated three times. Crude product was assayed by transferring
benzene-d6 to an NMR tube in the absence of N2. Bright purple 6b
was isolated (0.096 g) in 98% yield. 1H NMR (C6D6, mult, assignt): δ
6.44 (s, c, j), 8.14 (s, g), 3.89 (s, h), 0.37 (s, s), 0.48 (s, r). 13C{1H}
NMR (C6D6, mult, JCF (Hz); assignt): δ 150.70 (a), 161.64 (d, 240;
b), 95.29 (c), 161.19 (d, 240; d), 160.65 (d, 240; e), 145.89 (f), 163.67
(g), 64.93 (h), 147.80 (i), 103.64 (j), 159.87 (d, 230; k), 149.31 (d,
230; l), 152.57 (d, 230; m), 135.51 (n), 12.51 (t, 12; r). 19F NMR
(C6D6, mult, JCF (Hz); assignt): δ −128.99 (d, 35; m), −168.31 (dd,
30, 20; l), −119.72 (dd, 24, 7; b), −148.39 (dd, 32, 9; e), −136.60 (dd,
20, 10; k), −135.51 (dd, 34, 7; d). 31P{1H} NMR (C6D6, mult,
assignt): δ 21.42 (s, r). Anal. Calcd for C20H26F6N2P2Fe: C, 45.65; H,
4.98; N, 5.32. Found: C, 45.62; H, 5.09, N, 5.35.
9. trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2py (7b). In a 50 mL bomb charged with
(Me3P)4FeMe2 (0.050 g, 0.128 mmol) and Im-b (0.039g, 0.128
mmol) was transferred 8 mL of Et2O, and the reaction mixture was
stirred for >4 h at 23 °C. The Et2O and PMe3 were removed in vacuo,
and the residue was triturated with Et2O to remove excess PMe3.
Another 8 mL of Et2O was transferred to the flask, and a solution of
pyridine N-oxide in Et2O (0.126 M) was added dropwise; this mixture
was stirred for 12 h at 60 °C. Product 7b was filtered and washed (3 ×
5 mL) with Et2O and crystallized at −78 °C (0.053 g, 70%). The red
microcrystalline solid was assayed by transferring C6D6 to an NMR
tube containing the solid, in the absence of N2. 1H NMR (C6D6, mult,
6.96 (t, 7, j, n), 0.73 (s, r), 1.28 (d, 6, s), 2.30 (s, CH3). 13C{1H} NMR
(C6D6, mult, JCF (Hz); assignt): δ 133.26 (a), 157.66 (d, 260; b),
96.10 (c), 158.32 (d, 260; d), 151.54 (d, 260; e), 110.75 (f), 165.29
(g), 62.03 (h), 131.79 (i), 117.45 (j, n), 152.16 (d, 260; k, m), 139.69
(d, 260; l), 16.79 (t, 10; r), 16.06 (d, 10; s), 62.51 (Cα), 103.34 (Cβ),
3.43 (s, CH3). 19F NMR (C6D6, mult, JCF (Hz); assignt): δ −119.52
(dd, 22, 7; b), −161.31 (t, 21; l), −132.76 (dd, 33, 7; d), −133.99 (dd,
22, 8; k, m), −118.35 (dd, 34, 18; e). 31P{1H} NMR (C6D6, mult, JCF
(Hz); assignt): δ 16.08 (“d”, 62; r), 14.12 (td, 62, 11; s). Anal. Calcd
for C26H36F6P3NFe: C, 49.94; H, 5.80; N, 2.24. Found: C, 50.04; H,
5.42, N, 2.45. IR: ν(CC) 2081 cm−1.
6. trans-{κC,N-(3,4,5-(F)3-C6H2)CH2NCH(3,4,6-(F)3-C6H-2-yl)}Fe-
(PMe3)3(CCPh) (4b-Ph). In a 25 mL round-bottom flask charged with
Fe(PMe3)4Me2 (0.050g, 0.128 mmol) and Im-b (0.038g, 0.128 mmol)
was transferred an 8 mL amount of benzene, and the reaction mixture
was stirred for 4 h at 23 °C. Phenylacetylene was added to the reaction
via syringe (14 uL, 0.128 mmol), and the reaction mixture was stirred
for 10 h at 23 °C. Upon removal of solvent, the crude solid was filtered
and washed with Et2O (3 × 5 mL). Dark red 4b-Ph was isolated
1
(0.070 g) in 79% yield. H NMR (C6D6, mult, JCF (Hz); assignt): δ
6.31 (m, c), 8.18 (s, g), 4.82 (s, h), 6.85 (t, 6; j, n), 0.74 (s, r), 1.27 (d,
5; s), 7.54 (d, 7; CoH), 7.25 (t, 7; CmH), 7.00 (t, 7; CpH). 13C{1H}
NMR (C6D6, mult, JCF (Hz); assignt): δ 133.46 (a), 160.19 (d, 260;
b), 96.55 (c), 157.76 (d, 260; d), 153.53 (d, 260; e), 125.97 (f), 165.29
(g), 62.32 (h), 150.11 (i), 114.69 (j, n), 151.62 (d, 250; k, m), 149.41
M
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
JCF (Hz); assignt): δ 6.53 (s, c), 8.81 (s, g), 4.24 (s, h), 8.01 (s, j), 0.54
(s, r), 6.89 (d, 6; CoH), 7.10 (m, CmH, CpH). 13C{1H} NMR (C6D6,
mult, JCF (Hz); assignt): δ 135.03 (a), 161.38 (d, 250; b), 94.93 (c),
158.56 (d, 250; d), 153.06 (d, 250; e), 129.96 (f), 163.67 (g), 62.96
(h), 138.79 (i), 115.11 (j), 144.10 (d, 250; k), 134.47 (d, 250; l),
151.19 (d, 250; m), 124.47 (n), 12.03 (t, 10; r), 141.19 (Co), 124.96
(CmH), 135.27 (Cp). 19F NMR (C6D6, mult, JCF (Hz); assignt): δ
−134.48−134.77 (m, d, k, m), −119.96 (dd, 22, 7; b), −162.23 (td, 22,
4; l), −127.45 (dd, 34, 23; e). 31P{1H} NMR (C6D6, mult, assignt):
20.15 (s, r). Anal. Calcd for C25H28F6N2P2Fe: C, 51.04; H, 4.80; N,
4.76. Found: C, 51.02; H, 4.90, N, 4.76.
concentrated solution in Et2O at −40 °C. 1H NMR (C6D6): δ 12.45 (1
H), −4.61 (2 H), −9.86 (2 H), −16.52 (18 H). μeff(296 K) = 1.9, 2.0
μB.
Single-Crystal X-ray Diffraction Studies. Upon isolation, the
crystals were covered in polyisobutenes and placed under a 173 K N2
stream on the goniometer head of a Siemens P4 SMART CCD area
detector (graphite-monochromated Mo Kα radiation, λ = 0.71073 Å).
The structures were solved by direct methods (SHELXS). All non-
hydrogen atoms were refined anisotropically unless stated, and
hydrogen atoms were treated as idealized contributions (Riding
model).
1. trans-{κC,N,C′-(2,4,5-trifluorophen-2-yl)CHNCH2(3,4,5-tri-
fluorophen-2-yl)}Fe(PMe3)2 (N2) (2b). A red block (0.25 × 0.20 ×
0.15 mm) was obtained from benzene. A total of 9138 reflections were
10. trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2(CNMe) (8b). In a 50 mL round-bottom flask
charged with (Me3P)4FeMe2 (0.200 g, 0.512 mmol) and Im-b (0.155
g, 0.512 mmol) was transferred 15 mL of benzene. The reaction
mixture was stirred for 4 h at 23 °C. Methyl isocyanide (27 uL, 0.512
mmol) was added to the flask with continuous stirring for 6 h at 23 °C.
The solvent was removed in vacuo, and the crude solid was filtered
and washed in Et2O (3 × 10 mL). Red-brown solid 8b was isolated
collected with 5119 determined to be symmetry independent (Rint
=
0.0298), and 4493 were greater than 2σ(I). A semiempirical absorption
correction from equivalents was applied, and the refinement utilized
2
2
w−1 = σ2(Fo ) + (0.0437p)2 + 0.1193p, where p = (Fo + 2Fc2)/3.
2. trans-{mer-κ-C,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2Cl (12b). A red block (0.40 × 0.15 × 0.10 mm)
was obtained from diethyl ether. A total of 23077 reflections were
1
(0.200 g) in 69% yield. H NMR (C6D6, mult, JCF (Hz); assignt): δ
6.40 (m, c, j), 8.18 (s, g), 4.32 (s, h), 0.53 (“t”, 4; r), 2.85 (t, 2; CH3).
13C{1H} NMR (C6D6, mult, JCF (Hz); assignt): δ 139.19 (a), 156.94
collected with 6176 determined to be symmetry independent (Rint
=
0.0321), and 4841 were greater than 2σ(I). A semiempirical absorption
correction from equivalents was applied, and the refinement utilized
(d, 250; b), 103.73 (c), 156.57 (d, 250; d), 150.38 (d, 250; e), 143.55
(f), 159.30 (g), 63.78 (h), 137.22 (i), 133.25 (j), 147.68 (d, 250; k),
138.53 (d, 250; l), 150.00 (d, 250; m), 157.33 (n), 14.62 (t, 12; r),
96.13 (s), 29.86 (CH3). 19F NMR (C6D6, mult, JCF (Hz); assignt): δ
−114.16 (d, 31; m), −147.14 (ddd, 19, 10, 4; k), −119.94 (ddd, 23, 8,
2; b), −166.70 (ddd, 32, 19, 5; l), −124.57 (ddd, 33, 21, 3; e), −133.85
(ddd, 31, 10, 2; d). 31P{1H} NMR (C6D6, mult, assignt): δ 24.04 (s, r).
Anal. Calcd for C22H26F6P2N2Fe: C, 48.02; H, 4.76; N, 5.09. Found:
C, 48.09; H, 4.62, N, 5.03. IR: ν(CN) 2073 cm−1.
2
2
w−1 = σ2(Fo ) + (0.0483p)2 + 1.1000p, where p = (Fo + 2Fc2)/3.
Computations. Calculations were carried out at the M06119/6-
311+G(d)120 level of theory. An ultrafine grid was used for integration
in all calculations. Simulations were performed with the Gaussian 09
program.121 All structures were optimized with restraint of neither
symmetry nor geometry. Open-shell complexes were modeled within
the framework of the unrestricted Kohn−Sham formalism; spin
contamination was deemed to be minimal via calculation of the
⟨S ⟩UDFT expectation value. Systems were judged to be minima via
11. trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2(N2CPh2) (9b). In a 10 mL round-bottom flask
charged with 2b (0.045 g, 0.084 mmol) and diphenyldiazomethane
(0.017 g, 0.088 mmol) was transferred 5 mL of Et2O at −78 °C. The
reaction mixture was warmed to 23 °C and stirred for 16 h. The brown
solution was filtered, and solvent was removed in vacuo, leaving a dark
2
̂
calculation of the energy Hessian. Quoted energetics are free energies
(kcal/mol), unless specified otherwise, and were determined with
unscaled vibrational frequencies assuming standard temperature and
pressure.
1
purple microcrystalline solid (0.045 g, 0.064 mmol) in 76% yield. H
ASSOCIATED CONTENT
* Supporting Information
NMR (C6D6, mult, JCF (Hz); assignt): δ 6.37 (m, c, j), 7.78 (s, g), 3.98
(s, h), 0.34 (t, 3; r), 7.61 (d, 8; CoH), 7.30 (t, 8; CmH), 6.98 (t, 8;
CpH). 13C{1H} NMR (C6D6, mult, assignt): δ 136.38 (a), 160.93 (b),
97.60 (c), 159.40 (d), 158.11 (e), 120.27 (f), 162.21 (g), 64.31 (h),
131.82 (i), 104.71 (j), 156.99 (k), 141.98 (l), 156.08 (m), 124.94 (n),
13.15 (r), 131.83 (p), 131.09 (q), 132.12 (Co), 126.02 (Cm), 125.45
(Cp). 19F NMR (C6D6, mult, JCF (Hz); assignt): δ −114.53 (d, 31; m),
−145.55 (dd, 20, 8; k), −119.22 (dd, 22, 9; b), −132.42 (dd, 31, 9; d),
−124.31 (ddd, 30, 23, 5; e), −165.49 (dd, 31, 20; l). 31P NMR (C6D6,
mult, assignt): δ 17.52 (s, r). Anal. Calcd for C33H33F6P2N3Fe: C,
56.35; H, 4.73; N, 5.97. Found: C, 56.31; H, 4.83, N, 5.91.
■
S
CIF files giving crystallographic data for 2b and 12b and text
and figures giving full experimental details, including all spectral
details. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
E-mail: ptw2@cornell.edu
■
12. trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2I (11b). In a 25 mL round-bottom flask charged
with 2b (0.100 g, 0.186 mmol), [Cp2Fe]BF4 (0.061g, 0.186 mmol),
and excess lithium iodide (0.075 g, 0.560 mmol) was transferred 8 mL
of THF at −78 °C. The reaction mixture was warmed to 23 °C and
stirred for 16 h. Solvent was removed from the olive green solution,
and the crude brown solid was filtered and washed (3 × 5 mL) with
toluene. The brown solid was then filtered and washed (3 × 5 mL)
with pentane, leaving the green-brown paramagnetic 11b (0.050 g,
0.079 mmol) in 42% yield. 1H NMR (C6D6): δ 19.95 (2 H), 12.35 (1
H), −10.25 (1 H), −13.49 (1 H), −16.62 (18 H). Anal. Calcd for
C20H23F6NP2IFe: C, 37.76; H, 3.64; N, 2.20. Found: C, 37.62; H, 3.70,
N, 2.26. μeff(296 K) = 1.9, 2.0 μB.
13. trans-{mer-κC,N,C′-(3,4,5-(F)3-C6H-2-yl)CH2NCH(3,4,6-(F)3-
C6H-2-yl)}Fe(PMe3)2Cl (12b). In a 25 mL round-bottom flask charged
with 2b (0.085 g, 0.158 mmol) and triphenylmethyl chloride (0.066 g,
0.237 mmol) was transferred 8 mL of benzene at −78 °C. The
reaction mixture was warmed to 23 °C and stirred for 16 h. The
solvent was removed in vacuo, and the crude red solid was filtered and
washed (3 × 5 mL) with hexane, leaving microcrystalline red 12b
(0.060 g, 0.110 mmol) in 70% yield. Single crystals were grown from a
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National Science Foundation (CHE-1055505,
P.T.W.), the U.S. Department of Energy (BES DE-FG02-
03ER15387, T.R.C.), and Cornell University for financial
support.
REFERENCES
■
(1) Howard, J. B.; Rees, D. C. Proc. Natl. Acad. Sci. U.S.A. 2006, 103,
17088−17093.
(2) Rees, D. C.; Tezcan, F. A.; Haynes, C. A.; Walton, M. Y.;
Andrade, S.; Einsle, O.; Howard, J. B. Philos. Trans. R. Soc. A 2005,
363, 971−984.
(3) Einsle, O.; Tezcan, F. A.; Andrade, S. L. A.; Schmid, B.; Yoshida,
M.; Howard, J. B.; Rees, D. C. Science 2002, 297, 1696−1700.
(4) Hoffman, B. M.; Dean, D. R.; Seefeldt, L. C. Acc. Chem. Res.
2009, 42, 609−619.
N
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(5) Eady, R. R. Chem. Rev. 1996, 96, 3013−3030.
(6) Rubio, L. M.; Ludden, P. W. J. Bacteriol. 2005, 187, 405−414.
(7)
(8) Hu, Y. L.; Ribbe, M. W. Microbiol. Mol. Biol. Rev. 2011, 75, 664−
677.
(9) Hu, Y. L.; Lee, C. C.; Ribbe, M. W. Dalton Trans. 2012, 41,
1118−1127.
(10) (a) Hu, Y. L.; Ribbe, M. W. Coord. Chem. Rev. 2011, 255, 1218−
1224. (b) Hu, Y. L.; Ribbe, M. W. Acc. Chem. Res. 2010, 43, 475−484.
(11) Wiig, J. A.; Hu, Y. L.; Lee, C. C.; Ribbe, M. W. Science 2012,
337, 1672−1675.
(43) Nikiforov, G. B.; Vidyaratne, I.; Gambarotta, S.; Korobkov, I.
Angew. Chem., Int. Ed. 2009, 48, 7415−7419.
(44) Caselli, A.; Solari, E.; Scopelliti, R.; Floriani, C.; Re, N.; Rizzoli,
C.; Chiesi-Villa, A. J. Am. Chem. Soc. 2000, 122, 3652−3670.
(45) Kawaguchi, H.; Matsuo, T. Angew. Chem., Int. Ed. 2002, 41,
2792−2793.
(46) Clentsmith, G. K. B.; Bates, V. M. E.; Hitchcock, P. B.; Cloke, F.
G. N. J. Am. Chem. Soc. 1999, 121, 10444−10445.
(47) Hebden, T. J.; Schrock, R. R.; Takase, M. K.; Muller, P. Chem.
Commun. 2012, 48, 1851−1853.
(48) Semproni, S. P.; Milsmann, C.; Chirik, P. J. Angew. Chem., Int.
Ed. 2012, 51, 5213−5216.
(12) Lancaster, K. M.; Boemelt, M.; Ettenhuber, P.; Hu, Y.; Ribbe,
M. W.; Neese, F.; Bergmann, U.; DeBeer, S. Science 2011, 334, 974−
976.
(13) Spatzal, T.; Aksoyoglu, M.; Zhang, L.; Andrade, S. L. A.;
Schleicher, E.; Weber, S.; Rees, D. C.; Einsle, O. Science 2011, 334,
940.
(14) Erisman, J. W.; Sutton, M. A.; Galloway, J.; Klimont, Z.;
Winiwarter, W. Nat. Geosci. 2008, 1, 636−639.
(15) Ertl, G. Z. Anorg. Allg. Chem. 2012, 487−489.
(16) Schlogl, R. Angew. Chem., Int. Ed. 2003, 42, 2004−2008.
(17) Ertl, G. Angew. Chem., Int. Ed. 2008, 47, 3524−3535.
(18) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements, 2nd
ed.; Butterworth-Heinemann (Elsevier Science): New York, 1997;
Haber−Bosch, pp 421−422; Ostwald, pp 466−467. For Allen and
Senoff, see ref 28.
(19) Allen, A. D.; Senoff, C. V. Chem. Commun. 1965, 621−622.
(20) Allen, A. D.; Bottomley, F.; Harris, R. O.; Reinsalu, V. P.; Senoff,
C. V. J. Am. Chem. Soc. 1967, 89, 5595−5599.
(21) (a) Leigh, G. J. Can. J. Chem. 2005, 83, 277−278. (b) Taube, H.
Coord. Chem. Rev. 1978, 26, 1−5.
(22) Chatt, J.; Dilworth, J. R.; Richards, R. L. Chem. Rev. 1978, 78,
589−625.
(23) Richards, R. L. Coord. Chem. Rev. 1996, 154, 83−97.
(24) Leigh, G. J. J. Organomet. Chem. 2004, 689, 3999−4005.
(25) Leigh, G. J. Acc. Chem. Res. 1992, 25, 177−184.
(26) Hidai, M. Coord. Chem. Rev. 1999, 185−186, 99−108.
(27) Hidai, M.; Mizobe, Y. In Metal Ions in Biological Systems; Marcel
Dekker: New York, 2002; Vol. 39, pp 121−161.
(28) Weiss, C. J.; Groves, A. N.; Mock, M. T.; Dougherty, W. G.;
Kassel, W. S.; Helm, M. L.; DuBois, D. L.; Bullock, R. M. Dalton Trans.
2012, 41, 4517−4529.
(49) Pun, D.; Bradley, C. A.; Lobkovsky, E.; Keresztes, I.; Chirik, P. J.
J. Am. Chem. Soc. 2008, 130, 14046−14047.
(50) Hanna, T. E.; Keresztes, I.; Lobkovsky, E.; Chirik, P. J. Inorg.
Chem. 2007, 46, 1675−1683.
(51) (a) Bernskoetter, W. H.; Olmos, A. V.; Lobkovsky, E.; Chirik, P.
J. Organometallics 2006, 25, 1021−1027. (b) Bernskoetter, W. H.;
Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2005, 127, 14051−14061.
(52) (a) Bernskoetter, W. H.; Pool, J. A.; Lobkovsky, E.; Chirik, P. J.
J. Am. Chem. Soc. 2005, 127, 7901−7911. (b) Pool, J. A.; Bernskoetter,
W. H.; Chirik, P. J. J. Am. Chem. Soc. 2004, 126, 14326−14327.
(c) Pool, J. A.; Lobkovsky, E.; Chirik, P. J. Nature 2004, 427, 527−530.
(53) Yandulov, D. V.; Schrock, R. R. Science 2003, 301, 76−78.
(54) (a) Schrock, R. R. Acc. Chem. Res. 2005, 38, 955−962.
(b) Schrock, R. R. Angew. Chem., Int. Ed. 2008, 47, 5512−5522.
(55) Nishibayashi, Y. Dalton Trans. 2012, 41, 7447−7453.
(56) Arashiba, K.; Miyake, Y.; Nishibayashi, Y. Nat. Chem. 2011, 3,
120−125.
(57) Tanaka, H.; Sasada, A.; Kouno, T.; Yuki, M.; Miyake, Y.;
Nakanishi, H.; Nishibiyashi, Y.; Yoshizawa, K. J. Am. Chem. Soc. 2011,
41, 3498.
(58) (a) Knobloch, D. J.; Semproni, S. P.; Lobkovsky, E.; Chirik, P. J.
J. Am. Chem. Soc. 2012, 134, 3377−3386. (b) Semproni, S. P.;
Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2011, 133, 10406−10409.
(c) Knobloch, D. J.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc.
2010, 132, 15340−15350. (d) Knobloch, D. J.; Lobkovsky, E.; Chirik,
P. J. J. Am. Chem. Soc. 2010, 132, 10553−10564. (e) Knobloch, D. J.;
Lobkovsky, E.; Chirik, P. J. Nat. Chem. 2010, 2, 30−35.
(59) Bernskoetter, W. H.; Lobkovsky, E.; Chirik, P. J. Angew. Chem.,
Int. Ed. 2007, 46, 2858−2861.
(60) Mori, M.; Hori, K.; Akashi, M.; Hori, M.; Sato, Y.; Nishida, M.
Angew. Chem., Int. Ed. 1998, 37, 636−637.
(61) (a) Manriquez, J. M.; Bercaw, J. E. J. Am. Chem. Soc. 1974, 96,
6229−6230. (b) Manriquez, J. M.; Sanner, R. D.; Marsh, R. E.; Bercaw,
J. E. J. Am. Chem. Soc. 1976, 98, 3042−3044.
(62) Smith, J. M.; Sadique, A. R.; Cundari, T. R.; Rodgers, K. R.;
Lukat-Rodgers, G.; Lachicotte, R. J.; Flaschenriem, C. J.; Vela, J.;
Holland, P. L. J. Am. Chem. Soc. 2006, 128, 756−769.
(63) Smith, J. M.; Subedi, D. Dalton Trans. 2012, 41, 1423−1429.
(64) Saouma, C. T.; Peters, J. C. Coord. Chem. Rev. 2011, 255, 920−
937.
(29) Bazhenova, T. A.; Shilov, A. E. Coord. Chem. Rev. 1995, 144,
69−145.
(30) Crossland, J. L.; Tyler, D. R. Coord. Chem. Rev. 2010, 254,
1883−1894.
(31) Hazari, N. Chem. Soc. Rev. 2010, 39, 4044−4056.
(32) Holland, P. L. Can. J. Chem. 2005, 83, 296−301.
(33) Ballmann, J.; Munha, R. F.; Fryzuk, M. D. Chem. Commun.
2010, 46, 1013−1025.
(34) Gambarotta, S.; Scott, J. Angew. Chem., Int. Ed. 2004, 43, 5298−
5308.
(65) Hohenberger, J.; Ray, K.; Meyer, K. Nat. Commun. 2012, 3, 1−
13.
(35) Fryzuk, M. D.; Johnson, S. A. Coord. Chem. Rev. 2000, 200−202,
379−409.
(66) (a) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126,
6252−6254. (b) Rohde, J.-U.; Betley, T. A.; Jackson, T. A.; Saouma, C.
T.; Peters, J. C.; Que, L. Inorg. Chem. 2007, 46, 5720−5726.
(67) Vogel, C.; Heinemann, F. W.; Sutter, J.; Anthon, C.; Meyer, K.
Angew. Chem., Int. Ed. 2008, 47, 2681−2684.
(68) Scepaniak, J. J.; Young, J. A.; Bontchev, R. P.; Smith, J. M.
Angew. Chem., Int. Ed. 2009, 48, 3158−3160.
(69) Scepaniak, J. J.; Bontchev, R. P.; Johnson, D. L.; Smith, J. M.
Angew. Chem., Int. Ed. 2011, 50, 6630−6633.
(70) Scepaniak, J. J.; Vogel, C. S.; Khusniyarov, M. M.; Heinemann,
F. W.; Meyer, K.; Smith, J. M. Science 2011, 331, 1049−1052.
(71) Berry, J. F.; Bill, E.; Bothe, E.; George, S. D.; Bienert, B.; Neese,
F.; Wieghardt, K. Science 2006, 312, 1937−1941.
(72) Aliaga-Alcalde, N.; George, S. D.; Mienert, B.; Bill, E.;
Wieghardt, K.; Neese, F. Angew. Chem., Int. Ed. 2005, 44, 2908−2912.
(36) Shaver, M. P.; Fryzuk, M. D. Adv. Synth. Catal. 2003, 345,
1061−1076.
(37) Fryzuk, M. D. Acc. Chem. Res. 2009, 42, 127−133.
(38) Ballmann, J.; Yeo, A.; Patrick, B. O.; Fryzuk, M. D. Angew.
Chem., Int. Ed. 2011, 50, 507−510.
(39) Chirik, P. J. Dalton Trans. 2007, 16−25.
(40) Rodriguez, M. M.; Bill, E.; Brennessel, W. W.; Holland, P. L.
Science 2011, 334, 780−783.
(41) (a) Laplaza, C. E.; Johnson, M. J. A.; Peters, J. C.; Odom, A. L.;
Kim, E.; Cummins, C. C.; George, G. N.; Pickering, I. J. J. Am. Chem.
Soc. 1996, 118, 8623−8638. (b) Figueroa, J. S.; Piro, N. A.; Clough, C.
R.; Cummins, C. C. J. Am. Chem. Soc. 2006, 128, 940−950.
(42) Korobkov, I.; Gambarrotta, S.; Yap, G. P. A. Angew. Chem., Int.
Ed. 2002, 41, 3433−3436.
O
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(73) Meyer, K.; Bill, E.; Mienert, B.; Weyhermueller, T.; Wieghardt,
K. J. Am. Chem. Soc. 1999, 121, 4859−4876.
(100) (a) Camadanli, S.; Beck, R.; Florke, U.; Klein, H. F.
̈
Organometallics 2009, 28, 2300−2310. (b) Beck, R.; Zheng, T.; Sun,
(74) Wagner, W. D.; Nakamoto, K. J. Am. Chem. Soc. 1989, 111,
1590−1598.
H.; Li, X.; Florke, U.; Klein, H.-F. J. Organomet. Chem. 2008, 693,
̈
3471−3478. (c) Beck, R.; Sun, H.; Li, X.; Camadanli, S.; Klein, H.-F.
Eur. J. Inorg. Chem. 2008, 3253−3257. (d) Klein, H.-F.; Camadanli, S.;
(75) Walstrom, A.; Pink, M.; Yang, X.; Tomaszewski, J.; Baik, M.-H.;
Caulton, K. G. J. Am. Chem. Soc. 2005, 127, 5330−5331.
(76) Askevold, B.; Nieto, J. T.; Tussupbayev, S.; Diefenbach, M.;
Herdtweck, E.; Holthausen, M. C.; Schneider, S. Nat. Chem. 2011, 3,
532−537.
Beck, R.; Leukel, D.; Florke, U. Angew. Chem., Int. Ed. 2005, 44, 975−
̈
977. (e) Klein, H.-F.; Camadanli, S.; Beck, R.; Florke, U. Chem.
̈
Commun. 2005, 381−382.
(101) (a) Liu, N.; Li, X.; Sun, H. J. Organomet. Chem. 2011, 696,
2537−2542. (b) Shi, Y.; Li, M.; Hu, Q.; Li, X.; Sun, H. Organometallics
2009, 28, 2206−2210.
(77) (a) Moret, M.-E.; Peters, J. C. J. Am. Chem. Soc. 2011, 133,
18118−18121. (b) Moret, M.-E.; Peters, J. C. Angew. Chem., Int. Ed.
2011, 50, 2063−2067.
(102) Bhattacharya, P.; Krause, J. A.; Guan, H. R. Organometallics
2011, 30, 4720−4729.
(78) Saouma, C. T.; Moore, C. E.; Rheingold, A. L.; Peters, J. C.
Inorg. Chem. 2011, 50, 11285−11287.
(103) (a) Martin, A. Chem. Rev. 2010, 110, 576−623. (b) Kulkarni,
A. A.; Daugulis, O. Synthesis 2009, 4087−4109.
(79) Takaoka, A.; Mankad, N. P.; Peters, J. C. J. Am. Chem. Soc. 2011,
133, 8440−8443.
(104) (a) Evans, D. F. J. Chem. Soc. 1959, 2003−2005. (b) Schubert,
E. M. J. Chem. Educ. 1992, 69, 62.
(80) Lee, Y.; Mankad, N. P.; Peters, J. C. Nat. Chem. 2010, 2, 558−
565.
(105) (a) Figgis, B. N.; Hitchman, M. A. Ligand Field Theory and Its
Applications; Wiley-VCH: New York, 2000. (b) Carlin, R. L.
Magnetochemistry; Springer-Verlag: New York, 1986.
(106) Zhang, G.; Musgrave, C. B. J. Phys. Chem. A 2007, 111, 1554−
1561.
(107) (a) Poli, R.; Cacelli, I. Eur. J. Inorg. Chem. 2005, 12, 2324−
2331. (b) Petit, A.; Cacelli, I.; Poli, R. Chem. Eur. J. 2006, 12, 813−
823.
(81) Whited, M. T.; Mankad, N. P.; Lee, Y. H.; Oblad, P. F.; Peters, J.
C. Inorg. Chem. 2009, 48, 2507−2517.
(82) Hendrich, M. P.; Gunderson, W.; Behan, R. K.; Green, M. T.;
Mehn, M. P.; Betley, T. A.; Lu, C. C.; Peters, J. C. Proc. Natl. Acad. Sci.
U.S.A. 2006, 103, 17107−17112.
(83) Gilbertson, J. D.; Szymczak, N. K.; Tyler, D. R. J. Am. Chem. Soc.
2005, 127, 10184−10185.
(108) (a) Sydora, O. L.; Wolczanski, P. T.; Lobkovsky, E. B.; Buda,
C.; Cundari, T. R. Inorg. Chem. 2005, 44, 2606−2618. (b) Kuiper, D.
S.; Douthwaite, R. E.; Mayol, A.-R.; Wolczanski, P. T.; Lobkovsky, E.
B.; Cundari, T. R.; Lam, O. P.; Meyer, K. Inorg. Chem. 2008, 47,
7139−7153. (c) Kuiper, D. S.; Wolczanski, P. T.; Lobkovsky, E. B.;
Cundari, T. R. J. Am. Chem. Soc. 2008, 130, 12931−12943. (d) Kuiper,
D. S.; Wolczanski, P. T.; Lobkovsky, E. B.; Cundari, T. R. Inorg. Chem.
2008, 47, 10542−10553.
(84) (a) Crossland, J. L.; Balesdent, C. G.; Tyler, D. R. Inorg. Chem.
2012, 51, 439−445. (b) Crossland, J. L.; Young, D. M.; Zakharov, L.
N.; Tyler, D. R. Dalton Trans. 2009, 9253−9259. (c) Crossland, J. L.;
Balesdent, C. G.; Tyler, D. R. Dalton Trans. 2009, 4420−4422.
(d) Crossland, J. L.; Zakharov, L. N.; Tyler, D. R. Inorg. Chem. 2007,
46, 10476−10478. (e) Gilbertson, J. D.; Szymczak, N. K.; Crossland, J.
L.; Miller, W. K.; Lyon, D. K.; Foxman, B. M.; Davis, J.; Tyler, D. R.
Inorg. Chem. 2007, 46, 1205−1214.
(109) Field, L. D.; Hazari, N.; Li, H. L.; Luck, I. J. Magn. Reson. Chem.
2003, 41, 709−713.
(85) (a) Sellmann, D.; Hille, A.; Rosler, A.; Heinemann, F. W.; Moll,
M.; Brehm, G.; Schneider, S.; Reiher, M.; Hess, B. A.; Bauer, W. Chem.
Eur. J. 2004, 10, 820−830. (b) Sellmann, D.; Hille, A.; Heinemann, F.
W.; Moll, M.; Reiher, M.; Hess, B. A.; Bauer, W. Chem. Eur. J. 2004,
10, 4214−4224.
(110) (a) Chang, C. J.; Harman, W. H. J. Am. Chem. Soc. 2007, 129,
15128−15129. (b) Piro, N. A.; Lichterman, M. F.; Harman, W. H.;
Chang, C. J. J. Am. Chem. Soc. 2011, 133, 2108−2111.
(111) (a) Brown, S. D.; Betley, T. A.; Peters, J. C. J. Am. Chem. Soc.
2003, 125, 322−323. (b) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc.
2005, 127, 1913−1923. (c) Thomas, C. M.; Mankad, N. P.; Peters, J.
C. J. Am. Chem. Soc. 2006, 128, 4956−4957.
(86) Saouma, C. T.; Muller, P.; Peters, J. C. J. Am. Chem. Soc. 2009,
131, 10358−10359.
(87) Saouma, C. T.; Kinnery, R. A.; Hoffman, B. M.; Peters, J. C.
Angew. Chem., Int. Ed. 2011, 50, 3446−3449.
(112) Cowley, R. E.; Eckert, N. A.; Elhaïk, J.; Holland, P. L. Chem.
Commun. 2009, 1760−1762.
(88) Yuki, M.; Miyake, Y.; Nishibayashi, Y. Organometallics 2012, 31,
2953−2956.
(113) Kubas, G. J. Metal-Dihydrogen and Sigma-Bond Complexes:
Structure, Theory, and Reactivity; Kluwer Academic: New York, 2001.
(114) (a) Heinekey, D. M.; Oldham, W. J., Jr. Chem. Rev. 1993, 93,
913−926. (b) Jessop, P. G.; Morris, R. H. Coord. Chem. Rev. 1992,
121, 155−284. (c) Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993,
32, 789−805.
(89) Haring, H. W., Ed. Industrial Gases Processing; Wiley-VCH:
Weinheim, Germany, 2008.
(90) Kerry, F. G. Industrial Gas Handbook: Gas Separation and
Purification; CRC Press, Taylor and Francis Group: Boca Raton, FL,
2007.
(91) Kidnay, A. J.; Parrish, W. R.; McCartney, D. G. Fundamentals of
Natural Gas Processing, 2nd ed.; CRC Press, Taylor and Francis
Group: Boca Raton, FL, 2011.
(115) Szymczak, N. K.; Tyler, D. R. Coord. Chem. Rev. 2008, 252,
212−230.
(116) (a) Bautista, M. T.; Earl, K. A.; Maltby, P. A.; Morris, R. H. J.
Am. Chem. Soc. 1988, 110, 7031−7036. (b) Earl, K. A.; Jia, G.; Maltby,
P. A.; Morris, R. H. J. Am. Chem. Soc. 1991, 113, 3027−3039.
(117) Desrosiers, P. J.; Cai, L.; Lin, Z.; Richards, R.; Halpern, J. J. Am.
Chem. Soc. 1991, 113, 4173−4184.
(118) Sydora, O. L.; Kuiper, D. S.; Wolczanski, P. T.; Lobkovsky, E.
B.; Dinescu, A.; Cundari, T. R. Inorg. Chem. 2006, 45, 2008−2021.
(119) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157−167.
(120) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem.
Phys. 1980, 72, 650−654.
(121) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
(92) Allen, A. D.; Bottomley, F. Can. J. Chem. 1968, 46, 469.
(93) Aresta, M.; Giannoccaro, P.; Rossi, M.; Sacco, A. Inorg. Chim.
Acta 1971, 5, 203−206.
(94) Giannoccaro, P.; Rossi, M.; Sacco, A. Coord. Chem. Rev. 1972, 8,
77−79.
(95) Franke, O.; Wiesler, B. E.; Lehnert, N.; Nather, C.;
̈
Ksenofontov, V.; Neuhausen, J.; Tuczek, F. Inorg. Chem. 2002, 41,
3491−3499.
(96) (a) Franke, O.; Wiesler, B. E.; Lehnert, N.; Peters, G.; Burger,
P.; Tuczek, F. Z. Anorg. Allg. Chem. 2006, 632, 1247−1256. (b) Studt,
F.; Tuczek, F. J. Comput. Chem. 2006, 27, 1278−1291.
(97) de la Jara Leal, A.; Tenorio, M. J.; Puerta, M. C.; Valerga, P.
Organometallics 1995, 14, 3839−3847.
(98) Volpe, E. C.; Wolczanski, P. T.; Lobkovsky, E. B. Organo-
metallics 2010, 29, 364−377.
(99) Karsch, H. H. Chem. Ber. 1977, 110, 2699−2711.
P
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.1; Gaussian, Inc., Wallingford, CT, 2009.
(122) Harris, T. V.; Szilagyi, R. K. Inorg. Chem. 2011, 50, 4811−4824.
Q
dx.doi.org/10.1021/ja311021u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX